preliminary economic assessment of the maracás vanadium project, 1.4 million tonnes per year...

8/19/2019 Preliminary Economic Assessment of the Maracás Vanadium Project, 1.4 Million Tonnes Per Year Processing Plant

http://slidepdf.com/reader/full/preliminary-economic-assessment-of-the-maracas-vanadium-project-14-million 1/400

NI 43-101F1 Technical Report

Preliminary Economic Assessment of the Maracás Vanadium Project,1.4 Million Tonnes per Year Processing Plant

Municipality of Maracás,Bahia State, Brazil

Prepared for:

Largo Resources Ltd.

Prepared by:

Donald Arsenault, P.Eng.B. Terrence Hennessey, P.Geo.Hebert Lopes Oliveira, BSc, MAIGJane Spooner, MSc, P.Geo.Kevin Tanas, P.Eng.Scott Weston, MSc, P.Geo.

Report No: ADV-TO-00007

Date: March 4, 2013



ADV-TO-00007 / March 4, 2013 Page iiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer

clauses contained in the body of the report

DATE AND SIGNATURE PAGE

The effective date of this Technical Report, entitled “Preliminary Economic Assessment of the Maracás

Vanadium Project, 1.4 Million Tonnes per Year Processing Plant” is March 4, 2013. The undersignedhave prepared the Technical Report in accordance with National Instrument 43-101 guidelines forTechnical Reports.

Donald Arsenault {Signed and Sealed} March 4, 2013

Donald Arsenault, P. Eng. Dated at Toronto, Ontario

Vice President Mining – Canada

RungePincockMinarco (Canada) Limited

B. T. Hennessey {Signed and Sealed} March 4, 2013

B. Terrence Hennessey, P.Geo. Dated at Toronto, Ontario

Vice President

Micon International Limited

Hebert Lopes Oliveira {Signed and Sealed} March 4, 2013

Hebert Lopes Oliveira, BSc, MAIG Dated at Belo Horizonte, MG Brazil

Senior Mineral Resource Geologist

Coffey Mining Pty Ltd

Jane Spooner {Signed and Sealed} March 4, 2013

Jane Spooner, MSc, P.Geo. Dated at Toronto, Ontario

Vice President


Kevin Tanas {Signed and Sealed} March 4, 2013

Kevin Tanas, P.Eng. Dated at Toronto, Ontario

Principal Mining Consultant

RungePincockMinarco (Canada) Limited

Scott Weston {Signed and Sealed} March 4, 2013

Scott Weston, MSc, P.Geo. Dated at Vancouver, British Columbia

Sector Leader, Mining

Hemmera Envirochem Inc.



ADV-TO-00007 / March 4, 2013 Page iiiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


CERTIFICATE OF QUALIFICATIONS



SUITE 900 - 390 BAY STREET, TORONTO ONTARIO, CANADA M5H 2Y2Telephone (1) (416) 362-5135 Fax (1) (416) 362 5763

CERTIFICATE

B. TERRENCE HENNESSEY

As the author of portions of the report titled “Preliminary Economic Assessment of the MaracásVanadium Project, 1.4 Million Tonnes per Year Processing Plant” dated March 4, 2013 on

certain mineral properties of Largo Resources Ltd. in Bahia state, Brazil, I, B. TerrenceHennessey, P.Geo., do hereby certify that:

1. I am employed by, and carried out this assignment for:


Suite 900, 390 Bay StreetToronto, Ontario

M5H 2Y2

tel. (416) 362-5135

fax (416) 362-5763

e-mail: [email protected]

2. I hold the following academic qualifications:

B.Sc. (Geology) McMaster University 1978

3. I am a registered Professional Geoscientist with the Association of Professional

Geoscientists of Ontario (membership # 0038); as well, I am a member in good standing

of several other technical associations and societies, including:

The Australasian Institute of Mining and Metallurgy (Member)

The Canadian Institute of Mining, Metallurgy and Petroleum (Member).

4. I have worked as a geologist in the minerals industry for over 30 years.

5. I do, by reason of education, experience and professional registration, fulfill the

requirements of a Qualified Person as defined in NI 43-101. My work experience

includes 7 years as an exploration geologist looking for iron ore, gold, base metal and tindeposits, more than 11 years as a mine geologist in both open pit and underground mines

and over 15 years as a consulting geologist working in precious, ferrous and base metals

as well as industrial minerals.

6. I visited Largo’s Maracás project on several occasions during the periods June 26 to 29,

2006, April 13 to 17, 2007 and September 26 to 30, 2011 to review exploration results

and exposures of mineralization. The property had not previously been visited by me.



2

7. I am responsible for the preparation of Sections 3.1, 7.1, 7.2, 7.3A, 7.4A, 8, 9, 10.1A to

10.5A, 11.1A, 11.2A, 12.1A to 12.4A, 14.1A to 14.11A, and summaries thereof in

Section 1, of the Technical Report

8. I am independent of the parties involved in the transaction for which this TechnicalReport is required, as defined in Section 1.5 of NI 43-101.

9. I have had no prior involvement with the mineral properties in question.

10. I have read NI 43-101 and the portions of this Technical Report for which I amresponsible have been prepared in compliance with the instrument.

11. As of the date of this certificate, to the best of my knowledge, information and belief, the portion of the Technical Report for which I am responsible contains all scientific and

technical information that is required to be disclosed to make this Technical Report not

misleading.

Effective date: September 17, 2012

Dated this 4th day of March, 2013

“B. Terrence Hennessey” {signed and sealed}

B. Terrence Hennessey, P.Geo.




20 Meteor Drive Etobicoke Ontario M9W 1A4 CanadaT (+1) (416) 213 1255 F (+1) (416) 213 1260coffey.com

Certificate of Qualified Person – Hebert Lopes Oliveira

I, Hebert Lopes Oliveira, Geologist Engineer , as an author of this report entitled “PreliminaryEconomic Assessment of the Maracás Vanadium Project, 1.4 Million Tonnes per Year

Processing Plant” prepared for Largo Resources Ltd. (“Largo”), and effective date 4th March

2013 (the “Technical Report”), do hereby certify that:

1. I am a consultant with Coffey Consultoria e Serviços Ltda, of Av Afonso Pena, 4001, 12°

andar, Bairro Serra- CEP 30.130-008, Belo Horizonte city, Minas Gerais State, Brazil;

2. I am a graduate of Federal University of Ouro Preto, Minas Gerais State, Brazil and hold

a Bachelor of Science Degree in Geologist Engineer (2003) and I have practiced my

profession continuously since 2003.

3. I am a professional geologist with 10 years of relevant experience in Resource and

Reserve estimation, involving mining properties in Brazil, including iron, gold, tungstein,

vanadium, copper, etc.

4. I am a member of the Australian Institute of Geoscientists (“MAIG”) #5170.

5. I have read the definition of "Qualified Person" set out in National Instrument 43-101 (“NI

43-101”) and certify that by reason of my education, affiliation with a professional

association (as defined in NI 43-101) and past relevant work experience, I fulfill the

requirements to be a "Qualified Person" for the purposes of NI 43-101.

6. I visited Largo’s Maracás Vanadium Project on September 10

th

to 13

th

, 2012 to reviewexploration results and exposures of mineralization. The property had not previously

been visited by me.

7. I am responsible for Sections: 7, items, 7.3B, 7.4B; 10, items 10.1B,10.2B; 11, items

11.1B, 11.2B, 11.3B, 11.4B, 11.5B, 11.6B; 12, items 12.1B, 12.2B, 12,3B, 12.4B and

12.5B; 14, items 14.1B and 14.2B of this Technical Report.

8. I am independent of Largo, pursuant to section 1.5 in NI 43-101.

9. I have had no prior involvement with the property that is the subject of the Technical

Report.

10. I have read NI 43-101, and portions of this Technical Report for which I am responsible

have been prepared in compliance with NI 43-101 and Form 43-101F1.

11. At the effective date of the Technical Report, to the best of my knowledge, information,

and belief, the portion of the Technical Report for which I am responsible contain all

scientific and technical information that is required to be disclosed to make this

Technical Report not misleading.




20 Meteor Drive Etobicoke Ontario M9W 1A4 CanadaT (+1) (416) 213 1255 F (+1) (416) 213 1260coffey.com

12. I do not have nor do I expect to receive a direct or indirect interest in the Maracás

Vanadium Project of Largo and I do not beneficially own, directly or indirectly, any

securities of Largo or any associate or affiliate of such company.

Belo Horizonte, Brazil, 4th March, 2013.

Hebert Lopes OliveiraBSc (Geologist Engineer), MAIG #5170.



SUITE 900 - 390 BAY STREET, TORONTO ONTARIO, CANADA M5H 2Y2Telephone (1) (416) 362-5135 Fax (1) (416) 362 5763

CERTIFICATE OF QUALIFIED PERSON

JANE SPOONER, M.Sc., P.Geo.

As a co-author of this report entitled “Preliminary Economic Assessment of the Maracás Vanadium Project, 1.4 Million Tonnes

per Year Processing Plant”, prepared by RungePincockMinarco (Canada) Limited and dated March 04, 2013, I, Jane Spooner,P.Geo., do hereby certify that:

1. I am employed by, and carried out this assignment for

Micon International LimitedSuite 900, 390 Bay StreetToronto, OntarioM5H 2Y2

tel. (416) 362-5135 fax (416) 362-5763e-mail: [email protected]

2. I hold the following academic qualifications:

B.Sc. (Hons) Geology, University of Manchester, U.K. 1972M.Sc. Environmental Resources, University of Salford, U.K. 1973

3. I am a member of the Association of Professional Geoscientists of Ontario (membership number 0990); as well, I am amember in good standing of the Canadian Institute of Mining, Metallurgy and Petroleum and the Prospectors andDevelopers Association of Canada.

4. I have worked as a specialist in mineral market analysis for over 35 years.

5. I do, by reason of education, experience and professional registration, fulfill the requirements of a Qualified Person asdefined in NI 43-101. My work experience includes the analysis of markets for base and precious metals, industrial

and specialty minerals (including vanadium), coal and uranium.

6. I have not visited the project site.

7. I am responsible for the preparation of Section 19 of this report entitled “Preliminary Economic Assessment of theMaracás Vanadium Project, 1.4 Million Tonnes per Year Processing Plant”, dated March 04, 2013.

8. I am independent of Largo Resources Ltd., as described in Section 1.5 of NI 43-101.

9. I have had no prior involvement with the mineral property in question other than providing consulting services.

10. I have read NI 43-101 and the portions of this report for which I am responsible have been prepared in compliance with

the instrument.

11. As of the date of this certificate, to the best of my knowledge, information and belief, the sections of this TechnicalReport for which I am responsible contain all scientific and technical information that is required to be disclosed to

make this report not misleading.

Signing Date: March 4, 2013

“Jane Spooner” {signed and sealed}

Jane Spooner, M.Sc., P.Geo.



ADV-TO-00007 / March 4, 2013 Page ivThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Table of Contents1 Summary ............................................................................................................................................ 1-1

1.1 Introduction ................................................................................................................................. 1-1

1.2 Property Description and Location ............................................................................................. 1-1

1.3 Exploration and Drilling .............................................................................................................. 1-3

1.4 Mineral Resource Estimates ...................................................................................................... 1-4

1.5 Mining Methods .......................................................................................................................... 1-6

1.6 Recovery Methods ................................................................................................................... 1-12

1.7 Environmental, Permitting and Tailings Management ............................................................. 1-16

1.8 Market Studies and Contracts .................................................................................................. 1-21

1.9 Capital and Operating Costs .................................................................................................... 1-21

1.10 Economic Analysis ................................................................................................................... 1-23

1.11 Recommendations ................................................................................................................... 1-28

2 Introduction ......................................................................................................................................... 2-1

2.1 Use of Report ............................................................................................................................. 2-1

2.2 Terms of Reference ................................................................................................................... 2-2

2.3 Sources of Information ............................................................................................................... 2-2

2.4 Site Visits .................................................................................................................................... 2-2

2.5 Terms and Units ......................................................................................................................... 2-3

3 Reliance on Other Experts ................................................................................................................. 3-1

3.1 Geology, Mineralogy, Resource Estimates (Micon) ................................................................... 3-1

3.2 Mineral Processing and Metallurgical Testing ........................................................................... 3-2

3.3 History ........................................................................................................................................ 3-2

3.4 Market Studies and Contracts .................................................................................................... 3-2

3.5 Environmental Studies, Permitting and Social or Community Impact........................................ 3-2

3.6 Infrastructure References ........................................................................................................... 3-3

3.7 Cost Estimate ............................................................................................................................. 3-3

3.8 Geology, Mineralogy, Resource Estimates (Coffey Mining) ...................................................... 3-3

4 Property Description and Location ..................................................................................................... 4-1

4.1 Location ...................................................................................................................................... 4-1

4.2 Property Status........................................................................................................................... 4-2

5 Accessibility, Climate, Local Resources, Infrastructure and Physiography ....................................... 5-1

5.1 Access ........................................................................................................................................ 5-1

5.2 Infrastructure .............................................................................................................................. 5-1

5.3 Climate and Physiography ......................................................................................................... 5-1



ADV-TO-00007 / March 4, 2013 Page vThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


6 History ................................................................................................................................................ 6-1

6.1 Summary .................................................................................................................................... 6-1

6.2 Exploration History ..................................................................................................................... 6-1

6.3 Historical Drilling ........................................................................................................................ 6-2

6.4 Historical Resource Estimates ................................................................................................... 6-3 6.5 Historical Technical and Environmental Studies ........................................................................ 6-4

7 Geological Setting and Mineralization ................................................................................................ 7-1

7.1 Regional Geology ....................................................................................................................... 7-1

7.2 Rio Jacaré Intrusion ................................................................................................................... 7-2

7.3.A Property Geology (Gulçari A Deposit) .................................................................................... 7-3

7.4.A Mineralization (Gulçari A Deposit) .......................................................................................... 7-7

7.3.B Property Geology (Satellite Deposits) .................................................................................... 7-9

7.3.1.B Novo Amparo North Deposit ............................................................................................ 7-10

7.3.2.B Novo Amparo Deposit ...................................................................................................... 7-12

7.3.3.B São José Deposit ............................................................................................................. 7-14

7.3.4.B Gulçari B Deposit ............................................................................................................. 7-16

7.3.5.B Gulçari A North Deposit ................................................................................................... 7-18

7.4.B Mineralization ....................................................................................................................... 7-20

8 Deposit Types .................................................................................................................................... 8-1

9 Exploration ......................................................................................................................................... 9-1

9.1 2006 Exploration Program ......................................................................................................... 9-1


9.2.1 Previous Geophysical Surveys .......................................................................................... 9-2

9.2.2 Discussion of Present Geophysical Techniques ................................................................ 9-2

9.2.3 Geophysical Survey Results .............................................................................................. 9-3


9.3.1 2011-2012 Exploration Program ........................................................................................ 9-3

10 Drilling .......................................................................................................................................... 10-1

10.1.A Drilling By Previous Operators ............................................................................................. 10-1

10.2.A 2007 Largo Drill Program ..................................................................................................... 10-1 10.3.A 2008 Largo Drill Program ..................................................................................................... 10-7

10.4.A 2011-2012 Largo Drill Program ............................................................................................ 10-9

10.5.A Logging ............................................................................................................................... 10-19

10.1.B Drilling By Previous Operators ........................................................................................... 10-19

10.2.B Maracás Drilling Program 2011/2012 ................................................................................ 10-23



ADV-TO-00007 / March 4, 2013 Page viThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


10.2.1.B Novo Amparo North Deposit .................................................................................... 10-25

10.2.2.B Novo Amparo Deposit .............................................................................................. 10-27

10.2.3.B São José Deposit ..................................................................................................... 10-29

10.2.4.B Gulçari B Deposit ....................................................................................................... 10-32

10.2.5.B Gulçari A Norte Deposit ............................................................................................. 10-33 11 Sample Preparation, Analysis and Security ................................................................................. 11-1

11.1.A Sampling Method and Approach .......................................................................................... 11-1

11.1.1.A Previous Operators ................................................................................................... 11-1

11.1.2.A Largo Core Sampling ................................................................................................ 11-2

11.2.A Sample Preparation, Analyses and Security ........................................................................ 11-3

11.2.1.A Pre-2006 Analytical Work ........................................................................................... 11-3

11.2.2.A Largo Analytical Work ................................................................................................. 11-3

11.1.B Sample Preparation and Analysis ........................................................................................ 11-4

11.2.B Core Logging ........................................................................................................................ 11-5

11.3.B Core Sampling...................................................................................................................... 11-8

11.4.B Density Determination ........................................................................................................ 11-10

11.5.B Sample Preparation and Analysis ...................................................................................... 11-12

11.6.B Adequacy of Procedures .................................................................................................... 11-13

12 Data Verification ........................................................................................................................... 12-1

12.1.A Pre-2006 ............................................................................................................................... 12-1

12.2.A Largo .................................................................................................................................... 12-1

12.2.1.A 2006 ........................................................................................................................... 12-1

12.2.2.A Early 2007 .................................................................................................................. 12-2

12.2.3.A 2007 Exploration Drilling Program ............................................................................. 12-4

12.3.A Micon .................................................................................................................................. 12-12

12.4.A Conclusions ........................................................................................................................ 12-12

12.1.B Geological Database .......................................................................................................... 12-13

12.2.B Quality Control.................................................................................................................... 12-13

12.3.B Standard and Blank Samples ............................................................................................. 12-14

12.4.B Comparative Data Analysis ................................................................................................ 12-15 12.5.B Data Quality Summary ....................................................................................................... 12-18

13 Mineral Processing and Metallurgical Testing ............................................................................. 13-1

13.1 Introduction ............................................................................................................................... 13-1

13.2 Process Technical and Economical References...................................................................... 13-1

13.3 Material Characterization, Mineralogy and Metallurgy ............................................................. 13-1



ADV-TO-00007 / March 4, 2013 Page viiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


13.3.1 Mineralogy and Chemical Analysis .................................................................................. 13-2

13.3.2 Comminution .................................................................................................................... 13-5

13.3.3 Concentration (Magnetic Separation) .............................................................................. 13-6

13.3.4 Roasting, Leaching and Precipitation............................................................................... 13-6

13.3.5 AMV Calcination ............................................................................................................. 13-16 13.3.6 Ferrovanadium Test ....................................................................................................... 13-17

13.4 PGM Testwork........................................................................................................................ 13-17

13.5 Pilot Plant Testing (by Largo) ................................................................................................. 13-18

13.5.1 Beneficiation of Ore Samples at Fundacao Gorceix ...................................................... 13-19

13.5.2 Davis Tube Testing of Low and High Grade Core Samples .......................................... 13-19

13.5.3 Kiln Roasting .................................................................................................................. 13-19

13.5.4 Leaching ......................................................................................................................... 13-20

14 Mineral Resource Estimates ........................................................................................................ 14-1

14.1.A Database .............................................................................................................................. 14-1

14.2.A Geological Solids and Domain Interpretation ....................................................................... 14-2

14.3.A Block Model Geometry ......................................................................................................... 14-4

14.4.A Specific Gravity .................................................................................................................... 14-5

14.5.A Population Statistics ............................................................................................................. 14-6

14.5.1.A Grade Capping ......................................................................................................... 14-14

14.5.2.A Compositing ............................................................................................................. 14-14

14.6.A Experimental Variograms ................................................................................................... 14-23

14.7.A Kriging and Resource Estimation ....................................................................................... 14-37

14.8.A Mineral Resources ............................................................................................................. 14-39

14.9.A Sensitivity Studies .............................................................................................................. 14-48

14.10.A Confirmation of Estimation ............................................................................................ 14-49

14.11.A Responsibility for Estimation ......................................................................................... 14-49

14.1.B Mineral Resources Estimates ............................................................................................ 14-50

14.1.1.B Introduction .............................................................................................................. 14-50

14.1.2.B Database ................................................................................................................. 14-50

14.1.3.B Geological Model...................................................................................................... 14-50 14.1.4.B Block Model .............................................................................................................. 14-56

14.1.4.1.B Density .................................................................................................................. 14-59

14.1.4.2.B Volumetric Block Model Validation ....................................................................... 14-59

14.1.4.3.B Compositing .......................................................................................................... 14-60

14.1.4.4.B Statistical Analysis ................................................................................................ 14-60



ADV-TO-00007 / March 4, 2013 Page viiiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


14.1.4.5.B Variography .......................................................................................................... 14-61

14.1.4.6.B Grade Estimation .................................................................................................. 14-65

14.1.5.B Grade Estimation Strategy ....................................................................................... 14-66

14.1.5.1.B Estimation Strategy and Search Neighbourhood ........................................................ 14-66

14.1.5.2.B NN-Check - Resource Validation ................................................................................. 14-72 14.2.B Resource Reporting ........................................................................................................... 14-76

15 Mineral Reserve Estimates .......................................................................................................... 15-1

16 Mining Methods ............................................................................................................................ 16-1

16.1 Introduction ............................................................................................................................... 16-1

16.2 Pit Optimization ........................................................................................................................ 16-1

16.3 Mine Design ............................................................................................................................. 16-5

16.4 Equipment .............................................................................................................................. 16-11

16.4.1 Equipment Strategy ........................................................................................................ 16-11

16.4.2 Productivity and Fleet Size ............................................................................................. 16-12

16.5 Mine Schedule.......................................................................................................................... 16-1

16.5.1 Mine Development Sequence .......................................................................................... 16-1

16.6 Drill and Blast ........................................................................................................................... 16-6

16.7 Mine Safety and Rescue .......................................................................................................... 16-7

17 Recovery Methods ....................................................................................................................... 17-8

17.1 Process Description ................................................................................................................. 17-8

17.2 Crushing and Grinding ............................................................................................................. 17-8

17.2.1 Magnetic Separation ........................................................................................................ 17-8

17.2.2 Concentrate Dewatering and Tailings Disposal ............................................................... 17-8

17.2.3 Roasting ........................................................................................................................... 17-3

17.2.4 Leaching ........................................................................................................................... 17-3

17.2.5 Precipitation...................................................................................................................... 17-3

17.2.6 AMV Drying ...................................................................................................................... 17-4

17.2.7 Ferrovanadium Production ............................................................................................... 17-4

17.2.8 Sodium Sulphate Production ............................................................................................ 17-4

18 Infrastructure ................................................................................................................................ 18-1 18.1 Process Water .......................................................................................................................... 18-1

18.2 Potable Water........................................................................................................................... 18-1

18.3 Sewage Treatment ................................................................................................................... 18-1

18.4 Fuel and Lubricant Storage and Distribution ............................................................................ 18-1

18.5 Air ............................................................................................................................................. 18-2



ADV-TO-00007 / March 4, 2013 Page ixThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


18.5.1 Air Emissions and Air Quality Monitoring .............................................................................. 18-2

18.6 Heating ..................................................................................................................................... 18-2

18.7 Power Supply ........................................................................................................................... 18-2

18.8 Buildings ................................................................................................................................... 18-3

18.9 Assay Laboratory ..................................................................................................................... 18-5 18.10 Miscellaneous Buildings ....................................................................................................... 18-5

18.11 Explosives Magazine ........................................................................................................... 18-5

18.12 Communications................................................................................................................... 18-5

18.13 Roads ................................................................................................................................... 18-5

18.14 Tailings Facility ..................................................................................................................... 18-6

18.14.1 Tailings Disposal Ponds ............................................................................................... 18-6

18.15 Waste Management ............................................................................................................. 18-7

19 Market Studies and Contracts ...................................................................................................... 19-1

19.1 Marketing and Market Research .............................................................................................. 19-1

19.2 The Market for Vanadium ......................................................................................................... 19-1

19.2.1 Demand ............................................................................................................................ 19-2

19.2.2 International Trade ........................................................................................................... 19-3

19.2.3 Vanadium Prices .............................................................................................................. 19-3

19.2.4 Price Outlook .................................................................................................................... 19-4

19.3 Contracts .................................................................................................................................. 19-5

20 Environmental Studies, Permitting and Social or Community Impact.......................................... 20-1

20.1 Regulatory Framework Overview ............................................................................................. 20-1

20.2 Environmental Permitting Status .............................................................................................. 20-2

20.3 Environmental Baseline Conditions ......................................................................................... 20-3

20.3.1 Climate and Physiography ............................................................................................... 20-3

20.3.2 Water Resources.............................................................................................................. 20-4

20.3.3 Flora Characterization ...................................................................................................... 20-5

20.3.4 Fauna Characterization .................................................................................................... 20-8

20.3.5 Aquatic Biota .................................................................................................................. 20-10

20.4 Social and Economic Baseline ............................................................................................... 20-12 20.4.1 Populations Dynamics .................................................................................................... 20-13

20.4.2 Employment Structure and Unemployment Rate ........................................................... 20-13

20.4.3 Economic Aspects .......................................................................................................... 20-14

20.4.4 Land Use and Occupation .............................................................................................. 20-15

20.4.5 Villages around the Project ............................................................................................ 20-15



ADV-TO-00007 / March 4, 2013 Page xThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


20.4.6 Historical and Cultural Heritage ..................................................................................... 20-20

20.4.7 Living Standards............................................................................................................. 20-21

20.4.8 Education ....................................................................................................................... 20-22

20.4.9 Health ............................................................................................................................. 20-22

20.4.10 Housing Conditions and Infrastructure ....................................................................... 20-23 20.4.11 Leisure, Tourism and Culture ..................................................................................... 20-23

20.4.12 Public Safety .............................................................................................................. 20-24

20.4.13 Property Disputes and Rural ...................................................................................... 20-24

20.4.14 Water Supply .............................................................................................................. 20-24

20.5 Environmental Impact Assessment, Mitigation and Compensation ....................................... 20-25

20.5.1 Physical Environment ..................................................................................................... 20-25

20.5.2 Biotic Environment ......................................................................................................... 20-36

20.5.3 Environmental Mitigation ................................................................................................ 20-38

20.6 Social and Economic Environment ........................................................................................ 20-40

20.6.1 Job and Income Generation ........................................................................................... 20-40

20.6.2 Boosting the Local and Regional Economy ................................................................... 20-41

20.6.3 Improvement of Access and Roads ............................................................................... 20-41

20.6.4 Pressure on the Educational System ............................................................................. 20-42

20.6.5 Pressure on the Public Health System .......................................................................... 20-42

20.6.6 Pressure on the Water Supply System .......................................................................... 20-42

20.6.7 Inhibition of Immigration and Attraction of Population.................................................... 20-43

20.6.8 Expectations and Population’s Opinion about The Project ............................................ 20-44

20.7 Geotechnics and Hydrology ................................................................................................... 20-45

20.7.1 Hydrological Studies ...................................................................................................... 20-45

20.7.2 Geological / Geotechnical Characterization of the Overall Project Area ....................... 20-49

20.7.3 Geological / Geotechnical Characterization of the Overall Project Area ....................... 20-49

20.7.4 Overall Geological / Geotechnical Characteristics ......................................................... 20-50

20.7.5 Geotechnical Studies ..................................................................................................... 20-52

20.8 Current Activities and Plans ................................................................................................... 20-59

20.8.1 Project Organization and Sustainability team ................................................................ 20-59 20.8.2 Equator Principles Audit Review .................................................................................... 20-60

20.8.3 Socio-Environmental Action Plan and Environmental Management System ................ 20-60

21 Capital and Operating Costs ........................................................................................................ 21-1

21.1 Project Capital Cost ................................................................................................................. 21-1

21.2 Operating Cost Summary ......................................................................................................... 21-2



ADV-TO-00007 / March 4, 2013 Page xiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


22 Economic Analysis ....................................................................................................................... 22-1

22.1 Project Evaluation .................................................................................................................... 22-1

22.1.1 Estimated Product Price and Revenue ............................................................................ 22-1

22.1.2 After-Tax Cash Flow......................................................................................................... 22-1

22.1.3 Cash Flow Analysis .......................................................................................................... 22-1 22.1.4 Sensitivity Analysis ........................................................................................................... 22-5

23 Adjacent Properties ...................................................................................................................... 23-1

24 Other Relevant Data and Information .......................................................................................... 24-1

25 Interpretations and Conclusions................................................................................................... 25-1

26 Recommendations ....................................................................................................................... 26-1

26.1 Expansion and Definition Drilling Program and Budget ........................................................... 26-1

26.2 Mineral Processing and Metallurgical Testing ......................................................................... 26-2

26.3 Open Pit Mining ........................................................................................................................ 26-3

26.4 Hydrology ................................................................................................................................. 26-3

26.5 Environmental Studies ............................................................................................................. 26-4

27 References ................................................................................................................................... 27-1



ADV-TO-00007 / March 4, 2013 Page xiiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


List of Tables Table 1-1 Maracás Property Exploration Permits ............................................................................... 1-3

Table 1-2 Total Maracás Drilling ......................................................................................................... 1-4

Table 1-3 Gulçari A Deposit Mineral Resource as at December 2007(1) ............................................ 1-5

Table 1-4 Gulçari A and Satellite Deposits Mineral Resources .......................................................... 1-6 Table 1-5 Pit Optimization Parameters ............................................................................................... 1-7

Table 1-6 Optimum Final Pit Results .................................................................................................. 1-8

Table 1-7 Approximate Pit Geometries ............................................................................................... 1-8

Table 1-8 Mine Design Parameters .................................................................................................... 1-8

Table 1-9 Summary of Pit Resources ............................................................................................... 1-10

Table 1-10 Mine Design and Production Target Criteria .................................................................... 1-10

Table 1-11 Summary of Key Process Design Criteria ........................................................................ 1-15

Table 1-12 Summary of Processing Recoveries ................................................................................ 1-16

Table 1-13 Main Characteristics of the Non-magnetic Dry Stack Tailings Pond ............................... 1-19

Table 1-14 Environmental Authorization and Permits ........................................................................ 1-21

Table 1-15 Ferrovanadium and Vanadium Pentoxide Pricing ............................................................ 1-21

Table 1-16 Capital Cost Summary – USD (000) ................................................................................ 1-22

Table 1-17 Average Annual Operating Cost Summary ...................................................................... 1-23

Table 1-18 Project Cash Flow ............................................................................................................ 1-24

Table 1-19 Sensitivity Analysis ........................................................................................................... 1-26

Table 3-1 Key Project Personnel ........................................................................................................ 3-1

Table 4-1 Maracás Property Exploration Permits ............................................................................... 4-3

Table 6-1 Summary of Diamond drilling, Maracás Property ............................................................... 6-3

Table 6-2 Historical Diamond Drilling, Gulçari A Deposit (1981 – 1987) ............................................ 6-3

Table 6-3 Historical “Reserve” Estimate (1986) - Gulçari A ............................................................... 6-4

Table 10-1 Largo 2007 Maracás Drill Program .................................................................................. 10-1

Table 10-2 Drill-Hole Summary for the 2007 Gulçari A Drill Program ................................................ 10-3

Table 10-3 2007 Gulçari A Drill Results ............................................................................................. 10-4 Table 10-4 2007 Late Drill Results ..................................................................................................... 10-7

Table 10-5 2008 Drill Program Summary ........................................................................................... 10-7

Table 10-6 2008 Drill Program Information ........................................................................................ 10-8

Table 10-7 2008 Drill Program Summary of Significant Results ........................................................ 10-9

Table 10-8 Largo 2011 - 2012 Drill Program .................................................................................... 10-10

Table 10-9 Gulçari A Zone Drilling ................................................................................................... 10-10

Table 10-10 Gulçari A Zone Norte Drilling ......................................................................................... 10-11

Table 10-11 Gulçari B Zone Drilling ................................................................................................... 10-11

Table 10-12 Gulçari B Sul Zone Drilling ............................................................................................. 10-12

Table 10-13 São José Zone Drilling ................................................................................................... 10-12

Table 10-14 Novo Amparo Zone Drilling ............................................................................................ 10-12

Table 10-15 Novo Amparo Norte Zone Drilling .................................................................................. 10-13 Table 10-16 2011-2012 Drill Program Summary of Significant Results ............................................. 10-13

Table 10-17 Total Maracás Drilling .................................................................................................... 10-18

Table 10-18 Drillhole Information ....................................................................................................... 10-20

Table 10-19 Drillhole Information GA ................................................................................................. 10-20

Table 10-20 Assay Result GA Deposit ............................................................................................... 10-22

Table 10-21 Maracás Drill Program 2011/2012 ................................................................................. 10-24

Table 10-22 Summary NAN Deposit .................................................................................................. 10-25



ADV-TO-00007 / March 4, 2013 Page xiiiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Table 10-23 Assay Result NAN Deposit ............................................................................................ 10-26

Table 10-24 Summary NA Deposit ..................................................................................................... 10-28

Table 10-25 Assay Result NA Deposit ............................................................................................... 10-28

Table 10-26 Summary SJ Deposit ..................................................................................................... 10-29

Table 10-27 Assay Result SJ Deposit ................................................................................................ 10-30

Table 10-28 Summary GB Deposit .................................................................................................... 10-32

Table 10-29 Assay Result GB Deposit ............................................................................................... 10-33

Table 10-30 Summary GAN Deposit .................................................................................................. 10-34

Table 10-31 Assay Result GAN Deposit ............................................................................................ 10-35

Table 11-1 Density Summary ........................................................................................................... 11-12

Table 12-1 Micon Check Samples ................................................................................................... 12-12

Table 12-2 Standards and Blanks QAQC Summary Results ........................................................... 12-13

Table 12-3 Quality Control Sample Summary .................................................................................. 12-14

Table 12-4 Internal Standard Detection limits .................................................................................. 12-14

Table 12-5 Standards and Blanks QAQC Summary Results ........................................................... 12-14

Table 12-6 QAQC Program Summary ............................................................................................. 12-16

Table 13-1 SGS Test Sample Analyses - 2007.................................................................................. 13-2

Table 13-2 SGS Test Sample Analyses – 2012 ................................................................................. 13-3 Table 13-3 Maracás Ore Mineralogical Composition ......................................................................... 13-3

Table 13-4 Bond Rod Mill Grindability Test Summary ....................................................................... 13-5

Table 13-5 Ball Mill Grindability Test Summary ................................................................................. 13-5

Table 13-6 Davis Tube Summary on Composites – Concentrate – Gulçari A ................................... 13-7

Table 13-7 Davis Tube Summary on Composites – Concentrate – Gulçari A Norte ......................... 13-8

Table 13-8 Davis Tube Summary on Composites – Concentrate – Gulçari B ................................... 13-8

Table 13-9 Davis Tube Summary on Composites – Concentrate – São José ................................... 13-9

Table 13-10 Davis Tube Summary on Composites – Concentrate – Novo Amparo .......................... 13-10

Table 13-11 Davis Tube Summary on Composites – Concentrate – Novo Amparo Norte ................ 13-10

Table 13-12 Davis Tube Summary on Composites – Tailings – Gulçari A ........................................ 13-11

Table 13-13 Davis Tube Summary on Composites – Tailings – Gulçari A Norte .............................. 13-12

Table 13-14 Davis Tube Summary on Composites – Tailings – Gulçari B ........................................ 13-12

Table 13-15 Davis Tube Summary on Composites – Tailings – São José ........................................ 13-13

Table 13-16 Davis Tube Summary on Composites – Tailings – Novo Amparo ................................. 13-14

Table 13-17 Davis Tube Summary on Composites – Tailings – Novo Amparo Norte ....................... 13-14

Table 13-18 Salt Roast Leaching with Sodium Carbonate - Overview Test Conditions and Results 13-15

Table 13-19 Bulk Leach and Desilication Test Results ...................................................................... 13-15

Table 13-20 AMV Precipitation Recovery and Analysis (wt%) ........................................................... 13-16

Table 13-21 AMV Calcination Results ................................................................................................ 13-16

Table 13-22 Analysis of V2O5 ............................................................................................................. 13-17

Table 13-23 Davis Tube Testing Results ........................................................................................... 13-19

Table 13-24 Leaching Tests Results .................................................................................................. 13-20

Table 14-1 Rock Type and Mineral Domain Codes at Gulçari A ....................................................... 14-3 Table 14-2 Average Specific Gravity for the Gulçari A Deposit, CBPM Data .................................... 14-5

Table 14-3 Average Specific Gravity for the Gulçari A Deposit, Largo Data ..................................... 14-6

Table 14-4 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw V2O5 Data ..... 14-12

Table 14-5 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw V2O5 Data ..... 14-13

Table 14-6 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw Pt Data ......... 14-13

Table 14-7 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw Pd Data ........ 14-14

Table 14-8 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, V2O5 Composites .. 14-22



ADV-TO-00007 / March 4, 2013 Page xivThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Table 14-9 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, V2O5 Composites .. 14-22

Table 14-10 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Pt Composites ...... 14-23

Table 14-11 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Pd Composites ..... 14-23

Table 14-12 2007 Ordinary Kriging Interpolation Parameters, Measured Resources ....................... 14-37

Table 14-13 2007 Ordinary Kriging Interpolation Parameters, Indicated Resources......................... 14-38

Table 14-14 2012 Ordinary Kriging Interpolation Parameters, Measured Resources ....................... 14-38

Table 14-15 2012 Ordinary Kriging Interpolation Parameters, Indicated Resources......................... 14-38

Table 14-16 2012 Ordinary Kriging Interpolation Parameters, Inferred Resources ........................... 14-38

Table 14-17 Solid and Domain Precedence for Block Modelling of the Gulçari A Deposit ................ 14-39

Table 14-18 Gulçari A Deposit Mineral Resource as at December, 20071 ........................................ 14-40

Table 14-19 Gulçari A Deposit Mineral Resource as at September, 20121 ....................................... 14-41

Table 14-20 Sensitivity Table, Measured and Indicated Resource, 2007 Gulçari-A Deposit ............ 14-48

Table 14-21 Sensitivity Table, M & I Resource, Gulçari A Deposit within Micon 2012 Pit Shell ........ 14-49

Table 14-22 Drillhole Database Summary ......................................................................................... 14-50

Table 14-23 Gulçari A North - Block Model Summary ....................................................................... 14-56

Table 14-24 Gulçari B - Block Model Summary ................................................................................. 14-56

Table 14-25 Novo Amparo - Block Model Summary .......................................................................... 14-56

Table 14-26 Novo Amparo North - Block Model Summary ................................................................ 14-56 Table 14-27 São José - Block Model Summary ................................................................................. 14-57

Table 14-28 Volumetric Validation - Geological Model and Block Model........................................... 14-60

Table 14-29 Basic Statistical Analysis Summary ............................................................................... 14-61

Table 14-30 Variogram Models Summary .......................................................................................... 14-62

Table 14-31 Ordinary Kriging Strategy - Satellite deposits ................................................................ 14-67

Table 14-32 Confidence Level of Key Criteria.................................................................................... 14-76

Table 14-33 Grade Tonnage Table – 17th September 2012 ............................................................. 14-77

Table 14-34 Cut-off grade definition based on economic parameters ............................................... 14-77

Table 16-1 Pit Optimization Parameters ............................................................................................ 16-2

Table 16-2 Pit Optimization - Gulçari A .............................................................................................. 16-3

Table 16-3 Pit Optimization - Gulçari A Norte .................................................................................... 16-3

Table 16-4 Pit Optimization - Gulçari B .............................................................................................. 16-4

Table 16-5 Pit Optimization - São José .............................................................................................. 16-4

Table 16-6 Pit Optimization - Novo Amparo ....................................................................................... 16-4

Table 16-7 Pit Optimization - Novo Amparo Norte ............................................................................. 16-5

Table 16-8 Optimum Final Pit Results ................................................................................................ 16-5

Table 16-9 Approximate Pit Geometries ............................................................................................ 16-6

Table 16-10 Mine Design Parameters .................................................................................................. 16-6

Table 16-11 Waste Pile Geometries .................................................................................................... 16-6

Table 16-12 Waste Pile Design Parameters ........................................................................................ 16-6

Table 16-13 Summary of Pit Resources ............................................................................................ 16-11

Table 16-14 Primary Fleet Productivity Estimates ............................................................................. 16-13

Table 16-15 Annual Primary Equipment List ...................................................................................... 16-14 Table 16-16 Annual Ancillary Equipment List ..................................................................................... 16-15

Table 16-17 Mine Design and Production Target Criteria .................................................................... 16-1

Table 17-1 Summary of Key Process Design Criteria ........................................................................ 17-9

Table 17-2 Summary of Processing Recoveries ................................................................................ 17-9

Table 19-1 World Vanadium Primary and Co-product Output ........................................................... 19-2

Table 19-2 World Crude Steel Production .......................................................................................... 19-3

Table 19-3 Three-year Average Prices for Vanadium Pentoxide and Ferrovanadium ...................... 19-5



ADV-TO-00007 / March 4, 2013 Page xvThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Table 20-1 Environmental Authorization and Permits ........................................................................ 20-3

Table 20-2 Vulnerable Species .......................................................................................................... 20-7

Table 20-3 Rare Species .................................................................................................................... 20-8

Table 20-4 Testwork results - ABA-M – Waste Rock- Gulçari A ...................................................... 20-27

Table 20-5 Testwork results – ABA-M- Waste Rock- Gulçari A ....................................................... 20-28

Table 20-6 Testwork Results – ABA-M –Magnetite Rocks- Gulçari A ............................................. 20-30

Table 20-7 Testwork Results – ABA-M –Magnetite rocks- Gulçari A .............................................. 20-31

Table 20-8 Maracás Vanadium Plant Atmospheric Emissions Estimate ......................................... 20-34

Table 20-9 Max. short-term pollutant conc. from plant emissions and related emissions stds ........ 20-34

Table 20-10 Max. long term pollutant conc. from plant emissions and related emissions stds ......... 20-35

Table 20-11 Mitigation Measures – Construction Phase ................................................................... 20-39

Table 20-12 Mitigation Measures – Operations Phase ...................................................................... 20-40

Table 20-13 Rainfall Height Rates (mm) ............................................................................................ 20-48

Table 20-14 Design Flow Volumes for the Hydraulic Structures ........................................................ 20-49

Table 20-15 Characterization of Soils and Outcrops in the Project`s Area ........................................ 20-51

Table 20-16 Main Characteristics of the Non-Magnetic Dry Stack Tailings Dump ............................ 20-54

Table 20-17 Main Characteristics Catchment Dike ............................................................................ 20-55

Table 20-18 Main Geometric Characteristics of the Leached Calcine Tailings Pond ........................ 20-57 Table 20-19 Main Geometric Characteristics of the Chloride Control Purge Pond ............................ 20-57

Table 20-20 Main Characteristics of the Ridges ................................................................................ 20-58

Table 20-21 Action Plan – Environmental Programs ......................................................................... 20-61



Table 22-1 Project Cash Flow ............................................................................................................ 22-3


Table 25-1 Largo Updated Mineral Resource Estimate ..................................................................... 25-1

Table 25-2 Action Plan – Environmental Programs ........................................................................... 25-3




Table 26-1 Proposed Resource Upgrade Drilling Budget .................................................................. 26-2



ADV-TO-00007 / March 4, 2013 Page xviThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


List of Figures Figure 1-1 Maracás Project Property ................................................................................................... 1-2

Figure 1-2 General Mine Layout .......................................................................................................... 1-9

Figure 1-3 LOM Total Material Movement by Pit ............................................................................... 1-11

Figure 1-4 LOM Saleable Vanadium Product .................................................................................... 1-12 Figure 1-5 Conceptual Process Flow sheet – Vanadium Pentoxide ................................................. 1-13

Figure 1-6 Conceptual Process Flow sheet – Ferrovanadium .......................................................... 1-14

Figure 1-7 Water Balance Flow Chart ............................................................................................... 1-17

Figure 1-8 Layout of Dry Stack Tailings Ponds ................................................................................. 1-18

Figure 1-9 Project NPV ...................................................................................................................... 1-27

Figure 1-10 Project IRR ....................................................................................................................... 1-27

Figure 4-1 Maracás Project Location Map ........................................................................................... 4-1

Figure 4-2 Maracás Project Property ................................................................................................... 4-2

Figure 5-1 Maracás Project With Gulçari A Hill in Background ........................................................... 5-2

Figure 7-1 Maracás Area Simplified Regional Geology Map .............................................................. 7-2

Figure 7-2 Maracás Property Geology Map ........................................................................................ 7-4

Figure 7-3 Gulçari A Deposit, Detailed Surface Geology Map ............................................................ 7-5

Figure 7-4 Gulçari A Deposit - Drill Section 6175N ............................................................................. 7-6

Figure 7-5 Gulçari A Deposit - Drill Section 6225N ............................................................................. 7-7

Figure 7-6 Property Geological Map With the Deposits .................................................................... 7-10

Figure 7-7 Integrate Map of NAN Deposit ......................................................................................... 7-11

Figure 7-8 Cross Section of NAN Deposit ......................................................................................... 7-12

Figure 7-9 Integrated Map of NA Deposit .......................................................................................... 7-13

Figure 7-10 Cross Section of NA Deposit ............................................................................................ 7-14

Figure 7-11 Integrated Map of SJ Deposit ........................................................................................... 7-15

Figure 7-12 Cross Section of SJ Deposit ............................................................................................ 7-16

Figure 7-13 Integrated Map of GB Deposit .......................................................................................... 7-17

Figure 7-14 Cross Section of GB Deposit ........................................................................................... 7-18 Figure 7-15 Integrated Map of GAN Deposit ....................................................................................... 7-19

Figure 7-16 Cross Section of GAN Deposit ......................................................................................... 7-20

Figure 10-1 Gulçari A Deposit Drill-Hole Plan ..................................................................................... 10-2

Figure 10-2 Zone Location Map......................................................................................................... 10-17

Figure 10-3 Headquarters of Maracás ............................................................................................... 10-24

Figure 10-4 Project Drillhole Storage Facility .................................................................................... 10-25

Figure 10-5 Drillhole of NAN Deposit ................................................................................................ 10-27

Figure 10-6 Mineralized zone - FNA15 .............................................................................................. 10-29

Figure 10-7 Location of Drillhole FSJ15 ............................................................................................ 10-32

Figure 10-8 Location of Drillhole FGB14 ........................................................................................... 10-33

Figure 10-9 Zone Mineralized Drillhole FGAN05............................................................................... 10-37

Figure 11-1 Drillhole Core with log intervals delimited ........................................................................ 11-5 Figure 11-2 Drillhole Core with tagged intervals delimited .................................................................. 11-6

Figure 11-3 Drillhole Core Log Example ............................................................................................. 11-7

Figure 11-4 Drillhole Core with tagged intervals delimited .................................................................. 11-8

Figure 11-5 Drillhole Core electric diamond bladed core .................................................................... 11-9

Figure 11-6 Batch Sample Storage Facility ....................................................................................... 11-10

Figure 11-7 Density Determination by Archimedes Principle ............................................................ 11-11

Figure 12-1 Gulçari A Core Duplicate Sampling.................................................................................. 12-3



ADV-TO-00007 / March 4, 2013 Page xviiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Figure 12-2 Gulçari A Core Duplicate Sampling With Hole FGA-41 Removed ............................... 12-4

Figure 12-3 High Standard V2O5 Assay Results .............................................................................. 12-5

Figure 12-4 Low Standard V2O5 Assay Results ............................................................................... 12-6

Figure 12-5 High Standard Platinum Assay Results ........................................................................ 12-6

Figure 12-6 High Standard Palladium Assay Results ...................................................................... 12-7

Figure 12-7 Field Blank Assay Results ............................................................................................ 12-8

Figure 12-8 Field Duplicate Assay Results ...................................................................................... 12-9

Figure 12-9 Sample Duplicate Assay Results .................................................................................. 12-9

Figure 12-10 Secondary Laboratory Check Assays – V2O5 ............................................................ 12-10

Figure 12-11 Secondary Laboratory Check Assays – Pt ................................................................ 12-11

Figure 12-12 Secondary Laboratory Check Assays – Pd ............................................................... 12-11

Figure 12-13 Standard to V2O5 produced by Largo and Internationally certified ............................ 12-15

Figure 12-14 Duplicates Summarized Quality Control .................................................................... 12-17

Figure 13-1 Ti-magnetite Associations by Size – Massive .............................................................. 13-4

Figure 13-2 Free Ti-Magnetite and Ilmenite Middles shown at -212 microns .................................. 13-4

Figure 14-1 Gulçari A Deposit Geological Domain Solids - Looking North ..................................... 14-3

Figure 14-2 Magnetite Domain, Lognormal Histogram, Raw Assays - V2O5 ................................... 14-6

Figure 14-3 Magnetite Domain, Log Probability Plot, Raw Assays - V2O5 ...................................... 14-7 Figure 14-4 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays - V2O5................. 14-7

Figure 14-5 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays - V2O5 .................... 14-8

Figure 14-6 Magnetite Domain, Lognormal Histogram, Raw Assays – Pt ...................................... 14-8

Figure 14-7 Magnetite Domain, Log Probability Plot, Raw Assays – Pt .......................................... 14-9

Figure 14-8 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays – Pt .................... 14-9

Figure 14-9 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays – Pt ...................... 14-10

Figure 14-10 Magnetite Domain, Lognormal Histogram, Raw Assays – Pd ................................... 14-10

Figure 14-11 Magnetite Domain, Log Probability Plot, Raw Assays – Pd ...................................... 14-11

Figure 14-12 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays – Pd ................. 14-11

Figure 14-13 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays – Pd .................... 14-12

Figure 14-14 Magnetite Domain, Lognormal Histogram, Composite Data Set - V2O5 .................... 14-16

Figure 14-15 Magnetite Domain, Log Probability Plot, Composite Data Set - V2O5 ....................... 14-16

Figure 14-16 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set - V2O5 .. 14-17

Figure 14-17 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set - V2O5 ..... 14-17

Figure 14-18 Magnetite Domain, Lognormal Histogram, Composite Data Set – Pt ........................ 14-18

Figure 14-19 Magnetite Domain, Log Probability Plot, Composite Data Set – Pt ........................... 14-18

Figure 14-20 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set – Pt ..... 14-19

Figure 14-21 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set – Pt ......... 14-19

Figure 14-22 Magnetite Domain, Lognormal Histogram, Composite Data Set – Pd ....................... 14-20

Figure 14-23 Magnetite Domain, Log Probability Plot, Composite Data Set – Pd .......................... 14-20

Figure 14-24 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set – Pd .... 14-21

Figure 14-25 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set – Pd ........ 14-21

Figure 14-26 Down-hole Variogram - V2O5 in Magnetite Domain ................................................... 14-24 Figure 14-27 Variogram - Direction of Intermediate Continuity - V2O5 in Magnetite Domain .......... 14-25

Figure 14-28 Variogram - Direction of Maximum Continuity - V2O5 in Magnetite Domain .............. 14-25

Figure 14-29 Variogram - Direction of Minimum Continuity - V2O5 in Magnetite Domain ............... 14-26

Figure 14-30 Down-hole Variograms - V2O5 in Magnetite-Pyroxenite Domain ............................... 14-26

Figure 14-31 Variogram - Direction of Int. Continuity - V2O5 in Magnetite-Pyroxenite Domain ...... 14-27

Figure 14-32 Variogram - Direction of Max. Continuity - V2O5 in Magnetite-Pyroxenite Domain .... 14-27

Figure 14-33 Variogram - Direction of Min. Continuity - V2O5 in Magnetite-Pyroxenite Domain ..... 14-28



ADV-TO-00007 / March 4, 2013 Page xviiiThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Figure 14-34 Down-hole Variogram - Pt in Magnetite Domain ....................................................... 14-28

Figure 14-35 Variogram - Direction of Intermediate Continuity - Pt in Magnetite Domain .............. 14-29

Figure 14-36 Variogram - Direction of Maximum Continuity - Pt in Magnetite Domain................... 14-29

Figure 14-37 Variogram - Direction of Minimum Continuity - Pt in Magnetite Domain .................... 14-30

Figure 14-38 Down-hole Variograms - Pt in Magnetite-Pyroxenite Domain ................................... 14-30

Figure 14-39 Variogram - Direction of Int. Continuity - Pt in Magnetite-Pyroxenite Domain ........... 14-31

Figure 14-40 Variogram - Direction of Maximum Continuity - Pt in Magnetite-Pyroxenite Domain 14-31

Figure 14-41 Variogram - Direction of Minimum Continuity - Pt in Magnetite-Pyroxenite Domain . 14-32

Figure 14-42 Down-hole Variogram - Pd in Magnetite Domain ....................................................... 14-32

Figure 14-43 Variogram - Direction of Intermediate Continuity - Pd in Magnetite Domain ............. 14-33

Figure 14-44 Variogram - Direction of Maximum Continuity - Pd in Magnetite Domain.................. 14-33

Figure 14-45 Variogram - Direction of Minimum of Continuity - Pd in Magnetite Domain ............... 14-34

Figure 14-46 Down-hole Variograms - Pd in Magnetite-Pyroxenite Domain .................................. 14-34

Figure 14-47 Variogram - Direction of Int. Continuity - Pd in Magnetite-Pyroxenite Domain .......... 14-35

Figure 14-48 Variogram - Direction of Maximum Continuity - Pd in Magnetite-Pyroxenite Domain14-35

Figure 14-49 Variogram - Direction of Minimum Continuity - Pd in Magnetite-Pyroxenite Domain 14-36

Figure 14-50 Gulçari A Deposit, Section 6200N - Domain and V2O5 Grade Distributions .............. 14-42

Figure 14-51 Gulçari A Deposit, Section 6150N - Domain and V2O5 Grade Distributions .............. 14-42 Figure 14-52 Gulçari A Deposit, Section 6200 N - Isoshell Grade Contours .................................. 14-43

Figure 14-53 Gulçari A Deposit, Plan View 280 m Level, Domain and V2O5 Grade Distributions .. 14-43

Figure 14-54 Gulçari A Deposit, Plan View 170 m Level, Domain and V2O5 Grade Distributions .. 14-44

Figure 14-55 Gulçari A Deposit, Plan View 240 m Level - Isoshell Grade Contours ...................... 14-44

Figure 14-56 Gulçari A Deposit, Section 6075 Showing Resource Confidence Categories ........... 14-45

Figure 14-57 Gulçari A Deposit, Section 6175 Showing Resource Confidence Categories ........... 14-46

Figure 14-58 Gulçari A Deposit, Section 6075 Showing V2O5 Grade Distributions ........................ 14-46

Figure 14-59 Gulçari A Deposit, Section 6175 Showing V2O5 Grade Distributions ........................ 14-47

Figure 14-60 Gulçari A Deposit, Plan View 210 m Level, Showing V2O5 Grade Distributions ........ 14-47

Figure 14-61 Vertical Sections Location .......................................................................................... 14-51

Figure 14-62 Geological Model - Gulçari A North - 3D View ........................................................... 14-52

Figure 14-63 Geological Model - Gulçari B - 3D View ..................................................................... 14-52

Figure 14-64 Geological Model - Novo Amparo - 3D View ............................................................. 14-53

Figure 14-65 Geological Model - Novo Amparo North - 3D View.................................................... 14-53

Figure 14-66 Geological Model - São José - 3D View .................................................................... 14-54

Figure 14-67 Typical Vertical Sections - Gulçari A North ................................................................ 14-55

Figure 14-68 Block Model - Gulçari A North Deposit....................................................................... 14-57

Figure 14-69 Block Model - Gulçari B Deposit ................................................................................ 14-57

Figure 14-70 Block Model - Novo Amparo Deposit ......................................................................... 14-58

Figure 14-71 Block Model - Novo Amparo North Deposit .............................................................. 14-58

Figure 14-72 Block Model - São José Deposit ................................................................................ 14-59

Figure 14-73 Variographical Analysis - Example Gulçari A North deposit (orebody2) ................... 14-63

Figure 14-74 Variographical Analysis - Example Novo Amparo North deposit (orebody1) ............ 14-64 Figure 14-75 Anisotropic Ellipse to Estimate Gulçari A North ......................................................... 14-68

Figure 14-76 Anisotropic Ellipse to Estimate Gulçari B ................................................................... 14-69

Figure 14-77 Anisotropic Ellipse to Estimate Novo Amparo ............................................................ 14-69

Figure 14-78 Anisotropic Ellipse to Estimate Novo Amparo North (Orebody 1) ............................. 14-70

Figure 14-79 Anisotropic Ellipse to Estimate Novo Amparo North (Orebody 2) ............................. 14-70

Figure 14-80 Anisotropic Ellipse to Estimate São José (Orebody 1) .............................................. 14-71

Figure 14-81 Anisotropic Ellipse to Estimate São José (Orebody 2) .............................................. 14-71



ADV-TO-00007 / March 4, 2013 Page xixThis report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


Figure 14-82 V2O5 % kriged by Coffey Mining – NN-Check Comparative Statistic ......................... 14-73

Figure 14-83 Swath Plot – East Coordinate (X) – V2O5 (%) ............................................................ 14-74

Figure 14-84 Swath Plot – North Coordinate (Y) – V2O5 (%) .......................................................... 14-75

Figure 14-85 Swath Plot – Elevation (Z) – V2O5 (%) ....................................................................... 14-75

Figure 16-1 General Mine Layout .................................................................................................... 16-7

Figure 16-2 Mine Layout - Gulçari A Phase 1 .................................................................................. 16-8

Figure 16-3 Mine Layout - Gulçari A Phase 2 .................................................................................. 16-8

Figure 16-4 Mine Layout - Gulçari A Norte ...................................................................................... 16-9

Figure 16-5 Mine Layout - Gulçari B ................................................................................................ 16-9

Figure 16-6 Mine Layout - São José .............................................................................................. 16-10

Figure 16-7 Mine Layout - Novo Amparo ....................................................................................... 16-10

Figure 16-8 Mine Layout - Novo Amparo Norte ............................................................................. 16-11

Figure 16-9 Mining Progression - Gulçari A ..................................................................................... 16-2

Figure 16-10 Mining Progression - Gulçari A Norte........................................................................... 16-2

Figure 16-11 Mining Progression - Gulçari B .................................................................................... 16-3

Figure 16-12 Mining Progression - São José .................................................................................... 16-3

Figure 16-13 Mining Progression - Novo Amparo ............................................................................. 16-4

Figure 16-14 Mining Progression - Novo Amparo Norte ................................................................... 16-4 Figure 16-15 LOM Total Material Movement ..................................................................................... 16-5

Figure 16-16 LOM Total Material Movement by Pit ........................................................................... 16-5

Figure 16-17 LOM Saleable Vanadium Product ................................................................................ 16-6

Figure 17-1 Conceptual Process Flow sheet – Vanadium Pentoxide ............................................. 17-1

Figure 17-2 Conceptual Process Flow sheet – Ferrovanadium ....................................................... 17-2

Figure 18-1 Plant Site Layout........................................................................................................... 18-4

Figure 19-1 Past Ferrovanadium Price Trend .................................................................................. 19-4

Figure 20-1 Jacaré River (Dry period) ............................................................................................. 20-4

Figure 20-2 Pedras Dam reservoir at Porto Alegre........................................................................ 20-12

Figure 20-3 Average Monthly Rainfall for the Pluviometric Station at Fazenda Alagadiço ........... 20-47

Figure 20-4 Intensity, Duration and Frequency Curves ................................................................. 20-48

Figure 20-5 Flow Chart of the Water Balance for the Project ........................................................ 20-50

Figure 20-6 Typical section of the Non-Magnetic Dry Stack Tailings Dump contour structure ..... 20-53

Figure 20-7 Layout of Non-Magnetic Dry Stack Tailing Dump ...................................................... 20-53

Figure 20-8 Map showing the location of the Sediment Catchment Dyke ..................................... 20-55

Figure 20-9 Section of the Sediments Catchment Dyke ................................................................ 20-55

Figure 20-10 Typical Cross-Section of the Leached Calcine Tailings Pond ................................... 20-56

Figure 20-11 Typical Cross-Section of the Chloride Control Purge Pond ....................................... 20-57

Figure 22-1 Project Net Present Value ............................................................................................ 22-6

Figure 22-2 Project Internal Rate of Return ..................................................................................... 22-6



ADV-TO-00007 / March 4, 2013 Page 1-1This report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer


1 Summary

1.1 Introduction

Largo Resources Ltd. (Largo) is currently constructing its Maracás Vanadium Project (the “Project”)

located in Bahia, Brazil. This Preliminary Economic Assessment reports on an expanded productionscenario for the Project. The Company commissioned the Report to re-scope the Project by incorporatinga new production stream of both vanadium pentoxide (V2O5) and ferrovanadium (FeV) as opposed to anFeV only scenario (set out in the September 2008 Definitive Feasibility Study as amended in May 2009(see www.sedar.com) prepared by Aker Solutions Canada Inc. (the "Feasibility Study")) as well as anexpanded production scenario which considered the mineral resources available to Largo set out in its

Amended Technical Report filed on SEDAR on December 21, 2012 (the "December Report").

1.2 Property Description and Location

The Maracás Project property consists of twenty (20) exploration permits totaling 28,587 hectares (seeFigure 1-1). These exploration permits are granted and listed in Table 1-1. Three of these exploration

permits include the original two, (DNPM 870134/82 and DNPM 870135/82), and are owned by Vanádiode Maracás Ltda (VML) which is controlled 99.9% directly and indirectly by Largo. The remaining 17exploration permits are owned by Largo Mineração Ltda, a wholly owned subsidiary of Largo.

Largo reports that all but two exploration permits are currently registered as exploration licenses and arein good standing. An application to convert two of the exploration licenses to exploitation licenses (mininglicenses) has been made. Grant of the mining license is pending, however, the Localization License (LL)and the Installation License (LI) for the Project have been granted. There are no fees due on theexploration permits until the exploitation license has been granted.

Largo entered into an agreement with VALE and Odebrecht, dated October 16, 2006 giving it an option toacquire a 90% interest in the Maracás property from the two Brazilian companies for a purchase price ofUSD10.0 million. Under the agreement, Largo maintained the exploration permits in good standing. OnDecember 22, 2012 Largo bought out VALE and Odebrecht giving Largo 100% ownership of Companhiade Maracás.

Largo reports that, to its knowledge, there are no existing environmental liabilities with the property.

The Maracás Project is located within the greater municipality of Maracás in eastern Bahia State, Brazil.Maracás lies about 250 km southwest of the City of Salvador, the capital of Bahia. The distance by road

from Salvador to the Project is 405 kilometers via a paved secondary road from the main coastal highwayin Bahia, with the project being accessed by about 50 km of secondary highway and gravel road west ofthe town of Maracás. Access to water, the electric power grid and a railroad is within reasonable

distance, and a trained workforce familiar with the mining and mineral exploration industries exists in thestate of Bahia and also within in the country generally.

The Project has been the subject of extensive prior exploration and feasibility work by Companhia Baianade Pesquisa Mineral (CBPM), CAEMI Mineração e Metálurgia S.A. (CAEMI) and Odebrecht, since thediscovery of vanadium there in 1980. According to reports by CBPM, CAEMI, Odebrecht and theirconsultants, the previous work had outlined the existence of four vanadium deposits, including a large





Figure 1-1 Maracás Project Property





Table 1-1 Maracás Property Exploration Permits

deposit of vanadium-rich titaniferous magnetite mineralization at Gulçari A and other smaller depositsknown as Gulçari B, Nova Amparo and Sāo José.

1.3 Exploration and Drilling

At the time of acquisition of Maracás in the fall of 2006, Largo had performed no exploration on theproperty, but had completed a due diligence review program which checked the logging of, andresampling of certain original diamond drill core from the 1980s, and checked the surveyed locations ofvarious trenches and drill-hole collars. This work confirmed the previous sampling and interpretation atthe Project and Largo used the data to complete an “in-house” mineral resource estimate for the Gulçari

A deposit at Maracás.

Since the acquisition of Maracás, Largo has completed three drill programs in 2007, 2008 and 2011-12.These programs have confirmed previous interpretations, increased the size of the deposit and createdenough data to allow for an update of the previous mineral resource and an upgrade of their confidencecategory. The total drilling completed on the property has tested 7 zones with 209 holes totaling35,286.59 m (see Table 1-2) of which Largo has drilled 140 holes totaling 29,371.29 m between 2007 and2012.

Requested

AreaCity

DNPM no Date (há) U.F.:BA

VML 870.134/1982 9/03/1982 Vanádio 2,000.00 Maracás

VML 870.135/1982 9/03/1982 Vanádio 2,000.00 MaracásVML 872.346/2010 14/10/2010 Vanádio 977.29 Maracás

Largo 871.550/2007 24/05/2007 Vanádio 1,877.30 Iramaia / Maracás

Largo 871.551/2007 24/05/2007 Vanádio 1,300.00 Man. Vitorino / Iramaia

Largo 871.552/2007 24/05/2007 Vanádio 925.65 Maracás

Largo 871.961/2008 8/04/2008 Vanádio 1,273.41 Maracás / Iramaia


Largo 870.498/2007 14/03/2007 Vanádio 1,000.00 Maracás


Largo 871.246/2010 28/06/2010 Vanádio 1,000.00 Manoel Vitorino



Largo 871.944/2010 6/09/2010 Vanádio 2,000.00 M. Vitor./Iramaia/ Marac.

Largo 871.943/2010 6/09/2010 Vanádio 2,000.00 Iramaia, Maracás

Largo 871.941/2010 6/09/2010 Vanádio 927 Maracás




Largo 873.878/2011 27/09/2011 Vanádio 1,709.53 Maracás/Iramaia

20 26,301.90

Holder DNPM Protocol

Substance

Total





Table 1-2 Total Maracás Drilling

There has been sufficient drilling in this area to demonstrate the continuity of the magnetite-rich horizonswhich is also supported by the ground magnetic survey that traces the known zones on surface. Theground magnetic survey also has identified a number of deposits that had not been previously tested.

1.4 Mineral Resource Estimates

As noted above, the property is host to a number of mineralized zones, namely Gulçari A, Gulçari ANorte, Gulçari B, São José, Novo Amparo and Novo Amparo Norte. In this technical reportRungePincockMinarco (RPM) have used the Gulçari A, Gulçari A Norte, São José, Novo Amparo and

Novo Amparo Norte Mineral Resources as at 2012 to determine Mineable Resources that support thedesign of an open pit and processing plant having a production rate of 1,400,000 t/a of vanadium ore.

In 2012 Largo retained Micon International Limited (Micon) to audit and accept responsibility for a MineralResource estimate for Gulçari A using all available data up to 2012. This estimate is presented in thisreport.

Area Type No. of Holes Total Metres

Gulçari A 1981-87 53 5,153

2007 45 11,196

2011-12 11 3,118 Total 109 19,466

Gulçari B 1981-83 7 270

2011-12 10 1,368

Total 17 1,638

Gulçari A Norte 2007 3 566

2008 1 211

2011-12 12 1,767

Total 16 2,544

Gulçari B Sul 2011-12 6 1,150

Total 6 1,150

Sāo Jose 1983 2 115 2008 9 2,210

2011-12 14 2,390

Total 25 4,714

Novo Amparo 1983 7 377

2007 9 1,502

2008 1 285

2011-12 2 358

Total 19 2,522

Novo Amparo Nor 2011-12 17 3,252

Total 17 3,252

Grand Total 209 35,287





In addition, Largo retained Coffey Mining Pty Ltd (Coffey Mining) to prepare Mineral Resource estimatesfor the Gulçari A North, Gulçari B, São José, Novo Amparo and Novo Amparo North deposits. Theseestimates are also presented in this report.

The Gulçari A deposit, as outlined from the drill programs, now extends 400 m along strike, and to avertical depth of over 350 m with true widths ranging from 11 to 100 m and with an average width of about

40 m. This deposit is part of a mineralizing system that extends the length of the property. All the assaysfrom this drill program are completed and results received.

A summary of the mineral resources for Gulçari A in 2007, as accepted by Micon, is set out in Table 1-3. They are reported at a 0.66% V2O5 cut-off grade. Only the magnetite and magnetite-pyroxenite domainshave been reported. The vanadium grade is presented as vanadium pentoxide as is the custom in theindustry.

Table 1-3 Gulçari A Deposit Mineral Resource as at December 2007(1)

(1) Resources within a USD 3.50 Whittle pit shell using USD 42.23/t operating costs and reported at a 0.66% V 2 O5

cut-off grade.

A three dimensional block model was generated to enable grade estimation. The selected block size wasbased on the geometry of the domain interpretation and the data configuration. The block size of 5-m Eby 5-m N by 5-m RL was selected. The “percent” block modeling technique was used to represent the

volume of the interpreted wireframe models. Sufficient variables were included in the block modelconstruction to enable grade estimation and reporting.

Resource estimation for the Maracás vanadium deposit was undertaken using ordinary kriging (OK) asthe principal estimation methodology for V2O5. The OK estimates were completed using Gemcom miningsoftware.

The following corresponds to the list of variables contained in the received block model data:

Contained Contained

V2O5 Pt Pd PGM V2O5 V2O5

Category/Zone Tonnes (%) (g/t) (g/t) (g/t) (t) (lb)

Measured

Magnetite Zone 2,450,000 2.17 0.39 0.15 0.54 53,500 117,300,000

Indicated

Magnetite Zone 5,920,000 1.90 0.23 0.11 0.34 112,700 247,900,000

TOTAL 8,370,000 2.00 0.28 0.12 0.40 166,000 365,200,000

Measured

Magnetite-Pyroxenite 3,410,000 0.92 0.20 0.06 0.26 31,400 69,100,000

Zone

Indicated

Magnetite-Pyroxenite 5,480,000 0.93 0.12 0.07 0.19 50,900 112,000,000 Zone

TOTAL 8,890,000 0.93 0.15 0.07 0.22 82,300 181,100,000

GRAND TOTAL 17,260,000 1.44 0.21 0.09 0.30 248,300 546,300,000





Lithology code Bulk density assigned by lithology-type constant values (t/m3) Vanadium pentoxide in percent (%V2O5) Titanium dioxide in parts per million (TiO2 ppm) Palladium grade in parts per million (Pd ppm) Platinum grade in parts per million (Pt ppm) A class code to distinguish Measured, Indicated, and Inferred resource blocks.

As noted above, the company has completed a revised block model and updated mineral resourceestimate for the existing Gulçari A deposit and five new satellite deposits incorporating the drilling fromthe 2011 program including 72 holes totaling 13,401m. The five satellite deposits which extend north fromGulçari A for eight kilometers include from south the north: Gulçari A Norte, Gulçari B, São José, Novo

Amparo and Novo Amparo Norte. All are hosted in the Jacaré River Intrusion.

The details for the deposit are presented in Table 1-4.

Table 1-4 Gulçari A and Satellite Deposits Mineral Resources

* Resource within an open pit model using USD 34.20/t all in operating costs and reported at a 0.45% V 2 O5 cut-off,

reviewed and confirmed by B. Terrence Hennessey (Micon)

** Resource within a pit shell using USD 34.20/t all in operating cost and reported at a 0.45% V 2 O5 cut-off, reviewed

and confirmed by Hebert Oliveira (Coffey Mining)

1.5 Mining Methods

The Maracás mining project has been planned as an open pit mine using a combination of hydraulicexcavators, large front end loaders and 40 tonne haul trucks as the primary mining equipment.

There are two hard rock types of mineralization which are referred to as Magnetite-Pyroxenite (MG) andMagnetite (HG) and are characterized by their contained vanadium expressed as the V2O5% content.

Design of the ultimate pit was based on the results of the Whittle Lerchs-Grossmann shell analysis.Whittle is a software package that uses the Lerchs-Grossman algorithm to determine the approximateshape of a near optimal pit shell based on applied cut-off grade (COG) criteria and pit slopes. Theseshells are generated from the geologic grade (block) models, economic and physical criteria.

V2O5

(%)

Measured 8,870,000 1.37 121,500

Indicated 15,770,000 0.96 151,400

M &I 24,640,000 1.11 272,900

Inferred 2,610,000 0.76 19,800

Gulçari A Norte** Inferred 9,730,000 0.84 81,388

Gulçari B** Inferred 2,910,000 0.70 20,312

Novo Amparo** Inferred 1,560,000 0.72 11,255

Novo Amparo Norte** Inferred 9,720,000 0.87 84,453

Sao Jose** Inferred 3,900,000 0.89 34,706

Satellite Deposits (5)** Inferred 27,820,000 0.83 232,114

Deposits Category Tonnes Contained V2O5

(tonnes)

Gulçari A*





For the Whittle analysis the grade models for vanadium, rock density and resource category (measuredand indicated) models were exported from Gemcom, and imported into Whittle. Values for each of theblocks in the block model were set by Whittle based on the recovery, economic and physical parametersshown in Table 1-5.

Table 1-5 Pit Optimization Parameters

RPM developed a range of pit shells for a range of vanadium prices and cut-off grades to evaluate theeconomic viability of MG and HG material. A range of shells were developed to determine the project’s

sensitivity and the basis for the designed ultimate pit. As the cut-off grades decreased, the tonnes

increased and resultant grade decreased, since the lower cut-off grade increases the revenue per tonneand thus the tonnage available for mining.

There are several different methods for selecting a pit shell for use as the basis for the ultimate pit design.In the case of the Maracás Project, the decision was made to use pit shells which corresponded to a USD26.88 per kilogram of V in FeV sale price, the 3-yr average price at the time of the Whittle Study. The pitoptimization analyses were completed assuming a 45 degree slope as per the 2009 Feasibility Study.

With the addition of the satellite pits to the Maracás project, the original Gulçari A ultimate pit designed inthe 2012 Technical Report will be expanded to include Phase 2 which exhibits a lower average grade andhigher strip ratio than Phase 1. The low strip ratio satellite pits will be used to blend against Gulçari APhase 2.

The optimum final pit results are shown in Table 1-6.

Description Units GA GAN GB SJ NA NAN

Product Price

V in FeV $/kg 26.88 26.88 26.88 26.88 26.88 26.88

Fe Bi -Product $/t 70.00 70.00 70.00 70.00 70.00 70.00

Operating Costs

Ore Mining $/t 3.40 3.29 3.35 3.73 3.96 5.01

Waste Mining $/t 3.40 3.29 3.29 3.29 3.29 3.29

Processing $/t 54.87 48.77 56.50 50.63 56.72 51.47

Cutoff Grades

Vanadium & Iron Ore % 0.35% 0.43% 0.32% 0.44% 0.32% 0.43%

Recovery

Mining % 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%

Processing to V2O5 % 75.38% 64.60% 73.88% 64.68% 71.75% 65.93%

Processing from V2O5 to V in FeV % 52.94% 52.94% 52.94% 52.94% 52.94% 52.94%

Physical Parameters

Magnetite Density t/m3 4.45 4.28 4.42 4.30 4.36 4.26

Manetite-Pyroxenite Density t/m3 3.57 3.32 3.35 3.33 3.38 3.33

Waste Density t/m3 3.09 3.03 2.90 3.05 3.00 3.06

Fe Bi -Product Percentage % 36.00% 23.36% 40.72% 28.12% 41.89% 29.95%

Pit Slope

Section 0-360 degrees 45 45 45 45 45 45





Table 1-6 Optimum Final Pit Results

The open pits will be accessed by a 15 m wide haul road at a 10 percent gradient. All roads and rampswill have two lanes sloped at the side to provide drainage, sub-grade support and a wearing surface.

The final pits are designed to allow sufficient blending of HG and MG ore to produce a ROM plant feed of1.10 percent V2O5, on average, over the life of the project.

After analysis of V2O5 percent cut-off grades, blending targets and the screening method for upgradingthe vanadium ore, it was determined that the economically feasible regions of the pit based on the currentparameters were between the maximum and minimum bench elevations as laid out in Table 1-7.

Table 1-7 Approximate Pit Geometries

.

Table 1-8 outlines the primary mine design parameters.

Table 1-8 Mine Design Parameters

Figure 1-2 shows the general mine layout.

Description Units GA\PH1 GA\PH2 GAN GB SJ NA NAN Total

Ore Tonnes t 12,486,222 13,329,384 3,908,970 1,970,140 2,397,409 1,830,930 6,936,829 42,859,884

Ore V2O5 Grade % 1.42% 1.01% 0.86% 0.70% 0.91% 0.79% 0.87% 1.06%

Waste Tonnes t 29,224,128 147,468,572 23,406,536 10,162,339 14,963,906 8,209,116 35,235,360 268,669,957

Total Tonnes t 41,710,350 160,797,956 27,315,506 12,132,479 17,361,315 10,040,046 42,172,189 311,529,841

Strip Ratio wton:oton 2.34 11.06 5.99 5.16 6.24 4.48 5.08 6.27


Maximum Bench RL m 310 320 320 330 340 350

Minimum Bench RL m -50 210 210 230 240 190

Ramp Access RL m 299 320 313 320 337 354

Width m 875 275 250 250 225 340

Length m 950 800 475 625 400 740

Description Units ValueTwo Lane Ramp Width m 15

Ramp Grade % 10

Bench Face Angle degrees 70

Pit Slope degrees 45

Bench Height m 20

Bench Width m 8





Figure 1-2 General Mine Layout

Novo Amparo Norte

Novo Amparo

Gulçari A Norte

Gulçari A

Gulçari B

São José





Table 1-9 summarizes the pit resources based on the design criteria used.

Table 1-9 Summary of Pit Resources

Ore and waste mining will be carried out using 91 t Hydraulic Excavators equipped with 6.0 m3 bucket,having a digging depth of 14.1 m to excavate and load 24.0 m3 (40 tonne) capacity trucks. The diggingdepth allows for top loading of haul trucks and the ability to maintain wall angles at the prescribed benchheights. In terms of operational flexibility, the excavator enables controlled excavation for mid-bencheswhere the Gabbro, HG and MG zones intersect.

For the drilling activities, a 5.5 inch diameter drill will be used in waste material and 3.5 inch diameter drillin ore. Preliminary evaluation considered a Sandvik 1500i drill that is equipped with both 3.5 inch and 5.5inch drill diameters capabilities to ensure required productivity rates and minimal operational time losses.

The support equipment required for this fleet includes a grader with a 4.3 m 3 blade width and a 310horsepower Track Dozer comparable to a Caterpillar D8T class.

A small hydraulic excavator equipped with a 2.6 m3 bucket is required to support the primary digging fleetand mine development work.

The mine schedule incorporates an ore blending mining method of both HG and MG ore bodies. Themine plan targets an annual concentrate production rate of 509kt per year with a 3-year ramp up period ofkiln production. Mining is scheduled to operate for 360 days per year. The main mining operations includeoverburden stripping and storage for reclamation, waste rock removal to piles, and ore excavation andtransportation to the primary crusher at the processing plant.

A mining recovery rate of 100 percent has been adopted for both HG and MG ore bodies. The main minedesign and production criteria are outlined in Table 1-10.

Table 1-10 Mine Design and Production Target Criteria

The excavation of the open pits will be mechanized and advance in successive benches. Mine productionbegins in Gulçari A. Phase 1 provides the majority of the ore to the plant for the first 12 years ofproduction. The second pushback of Gulçari A (Phase 2) starts in production Year 8 and continues to theend of the project life (Year 29) providing the majority of production from Year 23 onward.


Ore Tonnes t 13,748,627 11,861,228 3,160,466 1,813,670 2,415,008 1,293,369 7,444,814 41,737,182

Ore V2O5 Grade % 1.34% 1.24% 0.86% 0.69% 0.91% 0.82% 0.87% 1.12%

Waste Tonnes t 27,547,646 153,114,265 21,232,370 11,410,836 17,798,173 9,712,012 41,993,697 282,809,000Total Tonnes t 41,296,273 164,975,494 24,392,836 13,224,505 20,213,182 11,005,381 49,438,511 324,546,182


Operation Units Stripping ROM Stripping ROM

Average Production Rate kt/annum 5,171 1,412 12,269 1,512

Kiln Rate kt/annum na 509 na 509

V2O5 % % na 1.32 na 0.99

Bench Face Angle degrees 70 70 70 70

Final Berm Width m 8 8 8 8 Maximum Bench Height m 20 20 20 20

Years 3 - 10 Years 11 - EOM





Between production years 12 and 22 the majority of the ore production will be provided by the satellitepits while Gulçari A Phase 2 stripping occurs. Primary ore production moves from Gulçari A to Gulçari ANorte, Gulçari B, São José, Novo Amparo and then to Novo Amparo Norte. Sustained ore production inYear 22 becomes available from Gulçari A Phase 2 while Novo Amparo Norte is active. At this pointproduction is balanced between Novo Amparo Norte and Gulçari A Phase 2 to provide higher grade feedto the mill.

Figure 1-3 shows the total material movement by pit for the operation. Figure 1-4 outlines the LOMsaleable vanadium product schedule.

Figure 1-3 LOM Total Material Movement by Pit





Figure 1-4 LOM Saleable Vanadium Product

1.6 Recovery Methods

The flow sheet selected for this study has been based on metallurgical testwork compiled by SGS

Lakefield Research Group Limited (SGS) in 2007, a study undertaken by IMS Process Plant in 1990, afeasibility study completed by Lurgi in 1986 and a detailed Technical study produced by Engenharia eConsultoria Mineral S.A. (ECM) in 1990.

The testwork compiled by SGS included mineral processing investigations using magnetic separation torecover vanadium contained in magnetite and hydrometallurgical extraction using roasting, leaching,precipitation and calcining to produce an intermediate vanadium product.

In 2010 pilot scale testing was undertaken by Largo’s consultant, Les Ford, to test bulk samples of high -grade and low-grade ore with respect to recovery and leaching performance. This work simulated theproposed process route from ROM ore to a final V2O5 product.

The process flow sheet selected for the Maracás process plant comprises three stages of crushing, onestage of grinding, two stages of magnetic separation, magnetic concentrate roasting, vanadium leaching,ammonium metavanadate (AMV) precipitation, AMV filtration, AMV calcining to V2O5 flake as finalproduct. From Year 3, part (up to 65%) of the V2O5 produced will be smelted to produce ferrovanadium(FeV).See Figure 1-5 and Figure 1-6.

The plant (originally sized to process 960,000 t/a ROM) will be capable, after due modification, to process1,400,000 t/a of feed ore with an average grade of 1.10% V2O5. The plant has an operating regime of



ADV-TO-00007 / March 4, 2013 Page 1-13This report has been prepared for Largo Resources Ltd. and must be read in its entirety and subject to the third party disclaimer clauses contained in the body of the report

Figure 1-5 Conceptual Process Flow Sheet – Vanadium Pentoxide




Figure 1-6 Conceptual Process Flow Sheet – Ferrovanadium





365 d/a, 7 d/wk, 24 h/d and plant utilization of 89%, resulting in an average nominal hourly throughput of180 t/h. The plant will produce an equivalent of 11,400 t/y V2O5, (LOM average), and from Year 3 willpartly convert to production of 3,830 t/a (LOM average) of vanadium as ferrovanadium equivalent to 65%of total V2O5 from Year 5.

Recovery for V2O5 is estimated to be 72.5% while an overall average recovery of 68.4% V is expected for

ferrovanadium production over the life of mine.

Table 1-11 provides a summary of key process design criteria used for the process design. Processingrecoveries vary between the different ore deposits and are outlined in Table 1-12. Since blending ofmagnetite and magnetite-pyroxenite bearing material is planned to occur at the crusher, the magneticrecovery factors calculated are based on blended HG an MG grades.

Table 1-11 Summary of Key Process Design Criteria

(1) Includes Kiln Recovery of 87.5% in line with actual operations at the Rhovan and Vantech plants and 97%

leaching efficiency. Values chosen in consultation with Les Ford from actual similar operations, rather than

the pilot results that displayed lower recoveries.

Criterion Units Value

Nominal ore processing rate t/a 1,400,000

Average production of vanadium oxide) t/a 11,400

Average V2O5 head grade % 1.10

Ore specific gravity 4.3

Plant availability % 89

Plant operating hours h/yr 7 800

Average plant daily ore throughput t/d 4,320

Number of crushing stages 3

Crusher product size (80% passing) mm 7.5

Number of grinding stages 1

Grind product size microns 150

Bond ball mill work index (metric) kWh/t 12.9

Magnetic product solids yield % 34

V2O5 recovery to magnetic concentrate % 88.4

Average magnetic concentrate V2O5 content % 2.88

Roasting reaction zone residence time h 1

Leach retention time h 1

Average roasting/leach V2O5 conversion (1) % 84.8

Estimated AMV Precipitation V2O5 recovery % 97.5

Estimated conversion to Ferro vanadium –

vanadium recovery% 94.5-95.7

Total average estimated recovery to V 2O5 % 72.5

Total estimated vanadium recovery to FeV % 68.4 Average ferrovanadium production (LOM), as V t/a 3,830





Table 1-12 Summary of Processing Recoveries

* Recoveries calculated on a blended HG and MG material basis.

The final magnetic concentrate will be thickened and filtered. The filter cake will be fed to the roastingsection of the plant. The nonmagnetic tailings fraction from the beneficiation plant will be thickened,filtered and conveyed to the tailings storage area for dry stacking.

An off-gas control system will collect any dust entrained in the gas from the roaster. To meet localenvironmental regulation, an electrostatic precipitator will be installed to remove such particulates. The

quantity of sodium sulphate added to the kiln will be controlled and reduced in order to ensure compliancewith the emission limits for SO2. Since the sodium sulphate dosage was already maximized for the basecase, additional sodium carbonate will be added to the kiln in the expanded case to ensure efficientextraction of vanadium and the excess sodium sulphate produced in the evaporator will be stockpiled ona sealed area.

1.7 Environmental, Permitting and Tailings Management

The overall water balance of the project was determined in order to quantify water availability and identifyrequirements for tailings disposal and storage. Figure 1-7 shows a water balance flow chart involving thestructures under consideration and their corresponding flow rates.

The process plant make-up water during operations is 81.6 m3/hr where 75.0 m3/hr is provided from theRio de Contas and 5.6 m3/hr being provided from the water content in the mined ore. The licenseprovided by the federal water agency, ANA, (Agência Nacional De Aguas) allows a water-take of 300m3/hr from the Rio de Contas. The pumping system from the Rio de Contas is sized at 200 m3/hr. Thereis a circulating water load within the plant with the net make-up being the aforestated 75 m3/hr.

Originally, as seen in previous technical reports, the plant design incorporated a conventional slurriedtailings system with a tailings pond resulting in a greater demand for water. This water demand has beengreatly reduced with the introduction of “dry-stacked” tailings reflected in this report.

A number of geological and geotechnical characterization activities were carried out to provide input datafor the engineering design work associated with the open pit, the process plant installations, waste andore stock piles, tailings disposal and flood control systems.

The process plant is planned to be constructed northwest of the Gulçari A pit where natural elevationsrange from 300 to 325 m. The extraction process results in the need to provide several tailings storagefacilities and waste piles as noted below. Tailings generated by the process include leached calcine fromthe processing kiln discharge, filter cake from the desilication process, chloride control purge from theevaporation circuit and primary inert tailings originating from magnetic separation.


Magnetic Recovery * % 91.00% 77.85% 89.17% 77.94% 86.58% 79.47%

Scavenger Recovery % 1.00% 1.00% 1.00% 1.00% 1.00% 1.00%

Concentrate to V2O5 Yield % 81.93% 81.93% 81.93% 81.93% 81.93% 81.93%

V2O5 recovery % 75.38% 64.60% 73.88% 64.68% 71.75% 65.93%V2O5 to V in FeV Recovery % 94.50% 94.50% 94.50% 94.50% 94.50% 94.50%

Stoichiometry Recovery % 56.00% 56.00% 56.00% 56.00% 56.00% 56.00%

V in FeV Recovery % 39.89% 34.19% 39.10% 34.23% 37.97% 34.89%





Figure 1-7 Water Balance Flow Chart

The area destined for the construction of the Nonmagnetic Tailings Dry Stacking Area is locatednorthwest of the Gulçari A pit outside the João Creek Basin. The first waste pile is located southwest ofthe Gulçari A deposit. The Chloride Purge Tailings Pond is to be located south of the process plant andthe Leached Calcine Tailings Dump to the north.

The Non-Magnetic Dry Stack Tailings Dump consists of a series of ponds formed by rock-fill structuressealed by compacted clayey/saprolitic material, as illustrated in Figure 1-8. The proposed arrangementwill suffice to contain around 6.25M m3 of tailings that will be produced during the mine’s first 10 years of

vanadium production from the Gulçari A Phase 1 pit. Further study is required to expand this tailingsfacility to adequately meet the storage requirements of the remainder of the project life including theGulçari A Phase 2 pit and satellite pits.

The construction sequence will advance through stages. Two ponds will be constructed during pre-

production year 2 and production year 1 for dry stacking of nonmagnetic tailings close to the plant, andwill suffice for the first two years of plant operation. Tailings from Year 3 and beyond will be transported toan expanded version of the pond system shown in Figure 1-8.





Figure 1-8 Layout of Dry Stack Tailings Ponds

* Dry stack tailings cell system design to be expanded in the future to accommodate tailings storage requirements

beyond production Year 10.

The current design of the pond structure will be (considering a working life of 10 years) 9m high, 820mlong, 900m wide and will have a volume of 624 000m3. Table 1-13 presents its most relevantcharacteristics.

The leached calcine tailings and the desilication process tailings will all be discharged into the samestructure (pond). This structure has been designated herein as the Leached Calcine Tailings Pond. Thechloride control purge tailings are to be deposited in a separate pond. The envisaged constructionlocation for these ponds is northwest of the open pit, close to the industrial plant.

The dikes will be built using compacted earth and their base areas will be leak-proofed using double-layergeomembrane liner featuring a leak detection system.

To prevent silting of João River with the fines washed from ore and waste stockpiles and major structuresfrom the mine site, the construction of a dike, named Sediments Catchment Dike has been envisaged.

This dike as well as the pit-production ridges will be comprised of rock fill originating from the pit, togetherwith compacted earth and transition material placed between the foregoing.





Table 1-13 Main Characteristics of the Non-magnetic Dry Stack Tailings Pond

* Dry stack tailings pond characteristics to be updated with revised design to accommodate tailings storage

requirements beyond production Year 10.

The dike will be constructed at João Creek and its design is such that it will capture the entire run offdrainages around the site assuring enough residence time to allow fines sedimentation during intenserains.

The area designed for the open pit intercepts of the João Creek and three direct tributaries.Consequently, this warrants the installation of a protection system to impede the influx of surface waterrunoff into the pit to enable mining activities to proceed.

For the initial phase, VOGBR, a Brazilian geotechnical consultant, envisaged the pit protection system toconsist of dikes and channels. The reason for the selection of this alternative provided a good balancebetween the volume of the required cuts and landfills. However, following ensuing Project developmentVanádio de Maracás decided to use a system of protection ridges, because of the proximity and usage ofmaterial from within the open pit, which was considered to be more economic.

This system consists of a pair of ridges located around the open pit, denominated as the Northern Ridgeand the Southern Ridge. The objective is to redirect the waters to points downstream from the open pit.The protection ridges form a barrier around the open pit to intercept the watercourses and to raise theirwater levels, so they can naturally surpass the topographical elevations variations and flow by gravitybearing the surface runoff downstream from the open pit.

The protection ridges will be installed pursuant to the mining activity plan defined for the open pit. TheNorthern Ridge should be completed by the close of Year 2 while the Southern Ridge will be required formining activity development as of Year 3.

The use of overburden material from the open pit was considered as a premise when dimensioning the

protection ridges. Accordingly, the ridge structures were conceived consisting of rock fill, transitionmaterial and compacted earth (residual soil/saprolitic material).

The borrow materials that will be used for the ridge construction are to be obtained from the open pit, theactual ridge location and the industrial plant area. Rocky overburden (processed material) is envisagedfor use as transition material within the ridges.

Maximum height (m) 9

Length of stack structure (m) 820

Width of stack structure (m) 900

Maximum crest elevation (m) 309Minimum downstream elevation (m) 295

Center road width (m) 8

Height of slopes between berms (m) 4

Tailings capacity - (m3) 6500 000

Maximum area occupied (m2) 740 000

NON-MAGNETIC TAILINGS DAM





The material originating from the open pit, which will be placed on the waste pile or the ore stockpile, ispredominantly rock consisting of boulders of varying sizes. The area destined for the waste pile coversapproximately 47 ha.

In addition, Largo have retained Mineral Engenharia em Meio Ambiente Ltda (Mineral) to complete anenvironmental audit incorporating the requirements of Equator Principle n 04. This audit resulted in an

Action Plan that incorporates the programs necessary for compliance with Brazilian laws and regulationsand applicable environmental performance standards and Environment, Health and Safety (EHS)guidelines.

The results of this audit are presented in Section 20.8 Current Activities and Plans.

A review of the Brazilian permitting process is presented as follows:

When a Class II mineral extraction project (as defined by the Mining Code) is presented for developmenta multidisciplinary technical review team is appointed by the State Council for Environmental Matters(CEPRAM) to review the project. The team sets the Terms of Reference for the Environmental Impact

Assessment (EIA) and the Relatório de Impacto Ambiental (RIMA). The RIMA is a document thatsummarizes the full impact assessment for review by the public.

The Terms of Reference for the EIM/RIMA include a social impact, alternatives and archaeologicalassessment for the Project, in addition to the basic physical and biological environmental impactassessment.

Environmental permitting in the State of Bahia is the responsibility of the Instituto do Meio Ambiente eRecursos Hídricos (INEMA), which is the institution that regulates, approves and issues theenvironmental permits or licenses. INEMA replaced the Instituto de Meio Ambiente (IMA) by state decreeon May 4, 2011.

The following types of environmental licenses are necessary for the Largo Maracás Vanadium Project inBahia:

Location License (LL) – approves the location and conceptual design of the project, attesting itsenvironmental feasibility and determines the basic requirements and conditions to be observed inthe subsequent permitting stages;

Installation License (LI) – is granted so that the Project can be installed or constructed inaccordance with the specifications presented in the plans, programs and projects proposed by theenvironmental studies that were approved, including the environmental control measures and otherconditions;

Operational License (OL) – is granted for the project to commence the operational phase after thefulfillment of all the requirements of the previous license have been confirmed and the conditions

and procedures to be observed during the operation are defined; Water Rights Grant – is obligatory for the lawfulness and legitimacy of any usage of water

resources; Vegetation Suppression Authorization (ASV) – is necessary to alter the land use for the installation

or expansion of mining operations.

At this time, the Maracás Vanadium Project is fully licensed and well advanced as shown in Table 1-14.





Table 1-14 Environmental Authorization and Permits

A complete study of the environmental conditions and social aspects of the area impacted by the Projectwas provided by Brandt Meio Ambiente Ltda dated May 2008 and completed by Fontes & Handel

Ambiental Ltd dated June 2011. A summary of their work is presented in Item 20 Environmental Studies,Permitting and Social or Community Impact.

1.8 Market Studies and Contracts

RPM used the three year average pricing from Metal bulletin (www.MetalBulletin.com) as their base casescenario for metal pricing for both vanadium pentoxide and ferrovanadium, as outlined in Table 1-15. Metal Bulletin is a widely accepted source for non-LME listed minor metal pricing and industry

information, and will be the source for sale pricing as is stipulated in the Company’s off -take contract withGlencore International Plc. (“Glencore”).

Table 1-15 Ferrovanadium and Vanadium Pentoxide Pricing

* Vanadium pricing determined based pricing as reported on Metal Bulletin.

1.9 Capital and Operating Costs

Project Capital Costs, as of March 2013, are estimated to be USD 389.1 million including an allowancefor contingencies of USD 22.0 million, equivalent to 6% of total capital expenditure. The capital costsummary as presented in Table 1-16 are for actual expenditures through December 2012, USD 33.4million, the total pre-production capital through to December 2013 of USD 230.4 million and remaining

Permits Status

Localization License - LLAwarded by CRA-CEPRAM( Environmental State

Council ) nº3941, May 2009/2014

Water Rights GrantAwarded by ANA (National Water Agency)

nº684, September 2009/2019

Installation License - LIAwarded by INEMA(Bahia State Environmental

Institute) nº1286, October 2011/2016

Vegetation Suppression Authorization - ASVAwarded by INEMA (Bahia State Environmental

Institute) nº0245 , October2011/2013

Vegetation Suppression Authorization- ASVAwarded by INEMA(Bahia State Environmental

Institute) nº0246 , October 2011/2013

Intervention in Area Permanent Preservation Authorization - IAPAwarded by INEMA(Bahia State Environmental


Legal Reserve Approval -ARLAwarded by INEMA(Bahia State Environmental


Vegetation Suppression Authorization- ASVAwarded by INEMA(Bahia State Environmental

Institute) nº049 , June 2012/2014

USD/lb V2O5 USD/kg V2O5

(at Jan 17, 2013) (at Jan 17, 2013)

Vanadium Pentoxide 6.37 14.04 6.47 14.28

USD/kg V

(at Jan 17, 2013)

Ferrovanadium - 28.01 32.00 -


US$/kg V

http://c/Users/mbarton/AppData/Local/Temp/www.MetalBulletin.com

http://c/Users/mbarton/AppData/Local/Temp/www.MetalBulletin.com





Table 1-16 Capital Cost Summary – USD (000)

other capital and sustaining capital costs of USD 125.4 million from 2014 to 2042. The total capitalexpenditure during the operation is USD 355.8 million, including acquisition to increase or replace minefleet equipment, plant and infrastructure.

A 2.00 BRL/USD exchange rate was applied to the project economics.

The operating expenditure is based on contractor mining during the preproduction years to Year 3 of theproject. From Year 4 onward, mining will be conducted by Largo. The strategy was determined as themost cost effective for the operation and ensures sustainability of a skilled labor force.

The average unit cost for operational activities is USD 61.50/t of ore. The lifetime annual average of alloperating costs included from years 1 to 29 amounts to USD 88.1 million. The breakdown of mining,

processing, general and administration costs, insurance and royalties as presented in Table 1-17

The average unit production cost for V2O5 from Year 1 to 29 is USD 3.10 per pound. Given the plannedcompletion of construction and commissioning of the FeV plant for Year 3, the average production costfrom Year 4 to 29 for FeV production is USD 15.26 per kilogram.

Capital ExpenditureSunk Cost

(2011-2012)

Remaining

20122013

2014

- 2042Total

Project Development and Studies 12,098 - - - 12,098

Enivronment 1,411 1,819 1,287 - 4,517 Management of Works 377 3,184 3,529 - 7,090

Project Management - 7,106 1,905 - 9,011

Project Costs (Basic & Executive) 6,657 6,704 357 - 13,718

Civil Works 37 33,829 12,580 - 46,446

Equipment 2,012 47,959 5,633 - 55,604

Installation and Commissioning - 21,236 19,707 - 40,943

Administration Costs 1,292 64 56 - 1,411

Consulting 10 64 56 - 129

Acquisition of Land 3,625 2,730 1,910 - 8,265

Acquisition of Mineral Rights 4,900 - - - 4,900

Imported Equipment 946 32,725 3,342 - 37,013

Seguros - 304 321 - 625

Contingency - 13,028 8,914 - 21,941

Plant Expansion 35,000 35,000

FeV Plant Construction - - - 15,000 15,000

Mining Equipment - - - 52,772 52,772

Sustaining Earthworks - - - 14,215 14,215

Closure Cost - - - 8,418 8,418

Total Capital Expenditures 33,365 170,752 59,594 125,405 389,117





Table 1-17 Average Annual Operating Cost Summary

Mining and Process Plant operating costs are largely variable per tonne of product while the General and Administrative costs are fixed per year. RPM has reviewed the basis of the operating cost estimate andconsiders the costs to be appropriate for evaluating economic viability of the project.

1.10 Economic Analysis

The economic analysis was completed for a 1.4 million tonne per year processing plant capacity and isbased on the pit resources outlined Table 1-18.

The market prices of vanadium product projected in the cash flow analysis are based on USD 14.04 perkilogram V2O5 and USD 28.01 per kg FeV. A USD 70.00 per tonne of iron ore sold FOB mine site isincluded as a secondary product.

Total annual revenue for the project is based on 1.4 Mt/y production to be reached in 2016 and continuingfor the life of the project averaging USD 192.0 million per year (gross revenue). The current planestimates 9.7 kt of V2O5 production and sales and 377 kt iron ore during 2015.

Net cash flow totals USD 1,845.3 million over the 29 year designed mine life. Economic results of the

Project cash flow model indicate an Internal Rate of Return (IRR) of 26.3 percent and a Net PresentValue of USD 554.0 million at an 8 percent discount rate. The 8 percent discount rate is consideredappropriate for this evaluation as the overall project risks are considered to be relatively low in terms oftotal capital committed, geological risk and market risk.

Operating CostReais/ Tonne

ROM

USD/Tonne

ROM

Mining 28.59 14.29

Plant 75.30 37.65

G&A 4.07 2.03Insurance 1.38 0.69

Royalty 13.66 6.83

Production Cost FOB 123.00 61.50




Table 1-18 Project Cash Flow

Project Period Units Value PP1 PP2 YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 YR10 YR11 YR12 YR13 YR14

Mining Physicals - - 647.85 6,881.22 9,493.08 9,553.20 9,430.78 8,933.66 8,945.42 9,332.56 9,094.36 7,592.76 5,858.53 4,151.78

Ore Mined to plant kt 40,585 628 850 570 1,326 1,413 1,413 1,413 1,413 1,411 1,413 1,473 2,082 1,644 1,254

Ore mined to stockpile kt 1,152 556 596 - - - - - - - - - - - - -

Waste Mined kt 282,020 1,700 3,249 2,601 3,385 4,238 4,238 2,846 2,087 4,238 8,476 11,858 12,654 20,000 18,366 15,097

Total Rock Mined kt 323,757 2,256 4,473 3,451 3,955 5,564 5,651 4,259 3,500 5,651 9,887 13,270 14,127 22,082 20,010 16,351

Stripping Ratio (Waste:Ore) 6.76 3.06 2.66 3.06 5.93 3.20 3.00 2.01 1.48 3.00 6.01 8.39 8.59 9.61 11.17 12.04

Ore from stockpile to plant kt 1,152 - - 223 842 86 - - - - - - - - - -

Total Material Movement kt 324,909 - 2,256 4,473 3,674 4,797 5,651 5,651 4,259 3,500 5,651 9,887 13,270 14,127 22,082 20,010 16,351

Total Plant Feed kt 41,737 - - 628 1,074 1,413 1,413 1,413 1,413 1,413 1,413 1,411 1,413 1,473 2,082 1,644 1,254

Grade of Vanadiam %V2O5 1.10% 0.00 1.16 1.20 1.22 1.29 1.37 1.38 1.36 1.29 1.29 1.35 1.28 0.86 0.80 0.69

Conta ined Vanad iam tonnes 455,205 - - 7,311 12,854 17,190 18,258 19,376 19,499 19,249 18,234 18,258 19,048 18,931 17,948 13,122 8,629

Concentrate recovered kt 14,147 226 387 509 509 509 509 509 509 508 509 492 502 509 506

Grade of concentrate recovered %V2O5 2.98 3.06 3.11 3.30 3.51 3.53 3.48 3.30 3.31 3.45 3.47 2.84 2.16 1.54

Total V2O5 Produced Tonnes 331,064 5,509 9,685 12,952 13,757 14,599 14,692 14,504 13,739 13,757 14,353 13,986 11,677 9,010 6,385

Recovered V2O5 tonnes 135,633 5,511 9,689 12,309 6,881 5,112 5,144 5,078 4,810 4,817 5,025 4,897 4,088 3,155 2,236

Recovered V in FeV tonnes 103,523 - - 343 3,643 5,026 5,057 4,993 4,729 4,736 4,941 4,814 4,020 3,101 2,198

Recovered FeV: 80.00% tonnes 129,404 - - 429 4,554 6,282 6,322 6,241 5,912 5,920 6,176 6,018 5,024 3,877 2,747

Iron Ore ByProduct kt 14,011 - - 220 377 496 502 503 503 503 504 503 504 488 498 505 504

Revenue

Revenue V2O5 0 00 U S$ 1 ,7 61 ,4 70 71,569 125,830 159,859 89,366 66,385 66,806 65,949 62,473 62,555 65,263 63,597 53,096 40,969 29,033

Revenue V in FeV 000 US$ 2,682,208 - - 8,886 94,383 130,208 131,032 129,353 122,535 122,696 128,006 124,739 104,143 80,356 56,946

Revenue of Iron Ore 000 US$ 980,794 15,430 26,377 34,738 35,118 35,241 35,239 35,244 35,262 35,210 35,247 34,129 34,866 35,378 35,263

Total Revenue 000 US$ 5,424,472 - - 86,999 152,207 203,482 218,867 231,834 233,077 230,546 220,270 220,461 228,516 222,465 192,105 156,703 121,242

Operating Costs 1.84

Mining unit operating cost US$/t 4.29 - 4.23 4.29 4.34 4.28 2.01 2.15 2.58 2.77 2.08 1.63 1.52 1.55 1.33 1.50 1.70

Drilling US$/t 0.83 0.83 0.83 0.83 0.83

Blasting US$/t 0.73 0.77 0.70 0.74 0.69

Load and Haul US$/t 2.23 2.13 2.27 2.27 2.27

Development US$/t 0.50 0.50 0.50 0.50 0.50

55,000

Mining annual cost 000 US$ 596,558 9,545 19,193 15,939 20,537 11,365 12,151 10,975 9,681 11,750 16,083 20,144 21,863 29,434 30,035 27,716

Process to V2O5 0 00 U S$ 1 ,4 11 ,0 13 31,372 42,765 50,168 50,885 51,649 51,721 51,575 50,981 50,964 51,457 51,008 52,404 48,412 44,432

Process V2O5 to FeV 000 US$ 245,350 - - - 813 8,634 11,911 11,986 11,832 11,209 11,223 11,709 11,410 9,526 7,350 5,209

Insurance on OPEX 000 US$ 28,861 1708 691 766 1,178 1,135 1,051 995 972 957 1,006 1,024 1,079 871 1,024 926 809

Total Operating Costs - FOB 000 US$ 2,281,783 1,708 10,236 51,331 59,882 72,653 71,936 76,705 75,654 74,046 74,945 79,295 84,390 85,152 92,389 86,724 78,167

Total Royalties 000 US$ 285,139 4,769 7,997 10,648 11,234 11,889 11,928 11,779 11,311 11,402 11,877 11,617 10,316 8,539 6,708

Tota l Ope rat ing C os t ( including R oy al ti e s) 0 00 U S$ 2 ,5 66 ,9 22 1,708 10,236 56,101 67,879 83,301 83,169 88,595 87,583 85,825 86,256 90,697 96,267 96,769 102,705 95,263 84,874

V 2O 5 P or ti on 55, 338 66,576 76,704 36,234 26,069 25,688 25,128 25,505 27,055 28,828 29,131 31,881 30,059 27,203

Unit Operating Costs FeV portion - - 4,881 45,201 60,785 60,154 58,956 59,009 61,902 65,698 65,952 69,101 63,457 55,929

Cost per Processed Tonne US$ / t 61.50 89.39 63.22 58.97 58.87 62.71 62.00 60.75 61.06 64.30 68.15 65.68 49.33 57.96 67.69

Cost per kg (Recovered V2O5) US$ / kg 7.01 10.18 7.01 6.37 5.42 5.25 5.15 5.10 5.46 5.78 5.89 6.10 7.98 9.76 12.48

Cost per kg V in FeV US$ / kg 15.6 21.6 15.6 14.4 12.6 12.3 12.1 12.0 12.7 13.3 13.5 13.9 17.4 20.8 25.9

Total Capital Expenditure 000 US$ 355,752 170,752 59,594 5,082 38,133 21,000 11,587 1,057 35 - 3,561 2,554 3,018 539 8,245 3,946 269

Operating Income Before Interest and Taxes

Before Tax Cashflow (Annual) 000 US$ 2,501,798 172,460- 69,830- 25,816 46,195 99,181 124,110 142,183 145,460 144,722 130,453 127,211 129,231 125,158 81,156 57,494 36,098

Before Tax Cashflow (Cumulated) 000 US$ 172,460- 242,290- 216,474- 170,279- 71,097- 53,013 195,196 340,656 485,377 615,831 743,041 872,272 997,430 1,078,586 1,136,080 1,172,178

Total Depreciation 000 US$ 277,475 20,967 20,967 20,967 22,655 22,709 22,524 22,637 23,099 21,621 21,574 2,234 2,832 2,882 3,274

Total Taxes 000 US$ 670,865 - - 740 3,847 11,928 15,472 18,220 18,748 18,618 16,372 16,102 16,418 41,794 26,630 18,568 11,160

Net After Tax Profit

After Tax Cashflow (Annual) 000 US$ 1,845,338 172,460- 69,830- 25,077 42,348 87,254 108,638 123,963 126,712 126,104 114,082 111,108 112,813 83,364 54,526 38,926 24,938

Net CashFlow

Real Discount Rate 8.0%

Discounted Cashflow 000 US$ 554,008 172,460- 64,658- 21,499 33,617 64,134 73,937 78,118 73,935 68,130 57,069 51,465 48,384 33,105 20,049 13,253 7,861

Cumulative Discounted CashFlow 000 US$ 172,460- 237,118- 215,618- 182,001- 117,867- 43,930- 34,188 108,123 176,253 233,322 284,787 333,171 366,276 386,325 399,577 407,439

Net Present Value 000 US$ 554,008

Internal Rate of Return 26.3%

Periods to discounted payback Years 4.0 - 1 2 3 4 5




Table 1-18 Project Cash Flow (Cont.)

Project Period Units Value YR15 YR16 YR17 YR18 YR19 YR20 YR21 YR22 YR23 YR24 YR25 YR26 YR27 YR28 YR29 Total

Mining Physicals 6,798.26 5,290.77 5,280.13 6,501.14 6,398.21 6,217.00 6,074.98 6,381.40 6,504.17 6,540.72 7,137.55 8,195.18 9,372.43 10,334.89 8,610.17 195,552.20

Ore Mined to plant kt 40,585 1,779 1,441 1,437 1,683 1,657 1,654 1,526 1,555 1,473 1,413 1,413 1,413 1,413 1,413 1,017 40,585

Ore mined to stockpile kt 1,152 - - - - - - - - - - - - - - - 1,152

Waste Mined kt 282,020 20,000 20,000 22,291 20,293 18,991 16,081 12,572 8,814 7,067 6,244 5,208 4,168 3,301 1,568 391 282,020

Total Rock Mined kt 323,757 21,779 21,441 23,728 21,975 20,648 17,735 14,097 10,370 8,540 7,657 6,620 5,581 4,714 2,980 1,408 323,757

Stripping Ratio (Waste:Ore) 6.76 11.24 13.88 15.51 12.06 11.46 9.72 8.24 5.67 4.80 4.42 3.69 2.95 2.34 1.11 0.38 6.76

Ore from stockpile to plant kt 1,152 - - - - - - - - - - - - - - - 1,152

Total Material Movement kt 324,909 21,779 21,441 23,728 21,975 20,648 17,735 14,097 10,370 8,540 7,657 6,620 5,581 4,714 2,980 1,408 324,909

Total Plant Feed kt 41,737 1,779 1,441 1,437 1,683 1,657 1,654 1,526 1,555 1,473 1,413 1,413 1,413 1,413 1,413 1,017 41,737


C on ta in ed Van ad ia m ton ne s 4 55 ,2 05 16,138 11,895 11,851 15,054 14,743 14,231 13,332 13,936 13,674 13,350 14,568 16,727 19,130 21,094 17,574 455,205

Concentrate recovered kt 14,147 502 509 505 509 504 509 495 509 509 509 509 509 509 509 366 14,147

Grade of concentrate recovered %V2O5 2.54 1.95 1.97 2.40 2.38 2.30 2.30 2.36 2.40 2.42 2.64 3.03 3.46 3.82 4.42

Total V2O5 Produced Tonnes 331,064 10,455 8,137 8,120 9,998 9,840 9,561 9,343 9,814 10,003 10,059 10,977 12,603 14,414 15,894 13,242 331,064

Recovered V2O5 tonnes 135,633 3,661 2,849 2,843 3,501 3,445 3,348 3,271 3,436 3,502 3,522 3,843 4,413 5,047 5,565 4,636 135,633

Recovered V in FeV tonnes 103,523 3,599 2,801 2,795 3,442 3,387 3,291 3,216 3,378 3,443 3,463 3,779 4,338 4,962 5,471 4,558 103,523

Recovered FeV: 80.00% tonnes 129,404 4,499 3,501 3,494 4,302 4,234 4,114 4,020 4,223 4,304 4,328 4,723 5,423 6,202 6,839 5,698 129,404

Iron Ore ByProduct kt 14,011 498 506 502 505 501 505 492 505 505 505 505 504 504 503 361 14,011

Revenue

Revenue V2O5 0 00 U S$ 1 ,7 61 ,4 70 47,540 36,998 36,924 45,462 44,743 43,475 42,482 44,625 45,484 45,739 49,913 57,309 65,541 72,272 60,211 1,761,470

Revenue V in FeV 000 US$ 2,682,208 93,245 72,569 72,423 89,170 87,758 85,273 83,325 87,528 89,212 89,713 97,899 112,406 128,553 141,754 118,098 2,682,208

Revenue of Iron Ore 000 US$ 980,794 34,882 35,400 35,118 35,354 35,054 35,365 34,438 35,359 35,354 35,353 35,330 35,290 35,246 35,210 25,298 980,794

Total Revenue 000 US$ 5,424,472 175,668 144,967 144,465 169,987 167,555 164,113 160,246 167,512 170,049 170,805 183,142 205,005 229,340 249,236 203,606 5,424,472


Mining unit operating cost US$/t 4.29 1.45 1.48 1.46 1.54 1.61 1.68 1.79 1.96 1.94 2.02 2.26 2.54 2.58 3.12 5.11

Drilling US$/t 0.83

Blasting US$/t 0.73

Load and Haul US$/t 2.23

Development US$/t 0.50

Mining annual cost 000 US$ 596,558 31,608 31,774 34,579 33,757 33,289 29,767 25,248 20,336 16,550 15,456 14,956 14,186 12,149 9,305 7,188 596,558

Process to V2O5 0 00 U S$ 1 ,4 11 ,0 13 49,998 46,761 46,617 49,366 48,998 48,890 47,729 48,614 48,364 48,121 48,834 50,098 51,505 52,655 44,668 1,411,013

Process V2O5 to FeV 000 US$ 245,350 8,529 6,638 6,625 8,157 8,028 7,800 7,622 8,006 8,160 8,206 8,955 10,282 11,759 12,967 10,803 245,350

Insurance on OPEX 000 US$ 28,861 956 934 1,101 953 938 907 831 794 751 791 754 766 759 761 674 28,861

Total Operating Costs - FOB 000 US$ 2,281,783 91,091 86,107 88,922 92,233 91,252 87,364 81,429 77,751 73,825 72,574 73,499 75,332 76,172 75,687 63,332

Total Royalties 000 US$ 285,139 9,518 7,974 8,002 9,269 9,138 8,901 8,612 8,880 8,926 8,937 9,537 10,603 11,766 12,695 10,365

Total Opera ting Cos t ( including Royal ties) 000 US$ 2 ,566 ,922 100,609 94,080 96,924 101,503 100,391 96,265 90,041 86,631 82,751 81,511 83,036 85,935 87,938 88,382 73,697

31,510 29,903 30,908 31,950 31,613 30,246 28,149 26,799 25,385 24,935 25,197 25,729 25,894 25,611 21,429

Unit Operating Costs 67,376 62,428 64,280 67,806 67,046 64,272 60,192 58,085 55,620 54,830 56,094 58,462 60,303 61,032 51,018

Cost per Processed Tonne US$ / t 61.50 56.55 65.31 67.45 60.33 60.59 58.20 59.02 55.70 56.20 57.70 58.78 60.83 62.25 62.56 72.48

Cost per kg (Recovered V2O5) US$ / kg 7.01 8.81 10.75 11.12 9.34 9.39 9.25 8.82 8.01 7.46 7.29 6.75 6.00 5.29 4.74 4.75

Cost per kg V in FeV US$ / kg 15.6 19.0 22.7 23.4 20.0 20.1 19.8 19.0 17.5 16.5 16.1 15.1 13.7 12.4 11.3 11.3

Total Capital Expenditure 000 US$ 355,752 2,057 4,681 4,401 1,035 851 788 - 551 - 5,368 1,595 1,635 - - 3,418


Before Tax Cashflow (Annual) 000 US$ 2,501,798 73,002 46,205 43,141 67,449 66,314 67,060 70,204 80,329 87,298 83,925 98,510 117,435 141,402 160,853 126,492

Before Tax Cashflow (Cumulated) 000 US$ 1,245,180 1,291,385 1,334,526 1,401,975 1,468,289 1,535,349 1,605,553 1,685,882 1,773,180 1,857,106 1,955,616 2,073,050 2,214,453 2,375,306 2,501,798

Total Depreciation 000 US$ 277,475 3,375 3,583 17,865 3,003 2,611 3,440 2,471 1,889 2,005 1,951 1,015 445 438 1,141 1,303

Total Taxes 000 US$ 670,865 23,673 14,491 8,594 21,912 21,659 21,631 23,029 26,670 29,000 27,871 33,149 39,777 47,928 54,302 42,564


After Tax Cashflow (Annual) 000 US$ 1,845,338 49,329 31,714 34,547 45,538 44,655 45,429 47,175 53,659 58,298 56,054 65,362 77,658 93,474 106,551 98,333

Net CashFlow


Discounted Cashflow 000 US$ 554,008 14,399 8,571 8,645 10,552 9,581 9,025 8,677 9,139 9,194 8,185 8,837 9,722 10,835 11,436 9,772

Cumulative Discounted CashFlow 000 US$ 421,838 430,409 439,054 449,606 459,186 468,211 476,888 486,028 495,221 503,406 512,243 521,965 532,800 544,236 554,008



Periods to discounted payback Years 4.0





RPM developed a sensitivity analysis for the cash flow model based on variations in key project elementsof metal price, operating and capital costs. The sensitivity of the Project’s IRR and NPV to +/ - 15 percentchanges to key assumptions is shown in Table 1-19.

Table 1-19 Sensitivity Analysis

Spider charts are shown in Figure 1-9 and Figure 1-10 below for the Project’s sensitivity to price, capital

expenditure, production costs and foreign exchange rate (FOREX), with key assumptions varying plusand minus 5 percent. The element with the most impact on the project cash flow is metal price, followedby FOREX, production cost and capital expenditures (+/- 15 percent).

ItemIRR

(%)

NPV

(USD Million)

Base Case 26.3% 554.0

Capex +15% 23.3% 510.7

Capex -15% 30.1% 597.3

Sale Price (V205) +15% 29.2% 638.1

Sale Price (V205) -15% 23.4% 470.0

Sale Price (FeV) +15% 28.1% 646.9

Sale Price (FeV) -15% 24.4% 461.1

Sale Price (Fe) +15% 27.2% 591.0

Sale Price (Fe) -15% 25.4% 517.0

Production Cost +15% 23.9% 462.6

Production Cost -15% 28.7% 645.4

Exchange Rate 2.3 31.2% 652.6






Figure 1-9 Project NPV

Figure 1-10 Project IRR





1.11 Recommendations

In the opinion of RungePincockMinarco, the 1.4 Million Tonnes per Year Processing Plant, hereindescribed as the expansion case, has merit. As the project continues to advance from the feasibility stageinto the design and construction phases there are areas of the project that should be given additionalconsideration beyond what is required for this level of study. Below is a summary of recommendations toconsider as the project advances.

Expansion and Definition Drilling Program and Budget

Largo completed a recommended diamond drilling program designed to expand the mineral resources onits Maracás property. This was a follow up of favorable results of an earlier exploration program ofground geophysics (magnetic survey), detailed geological mapping and sampling. Part of this programfocused on investigating the area between the Gulçari A deposit and the other five known prospects(satellite deposits); Gulçari A Norte, Gulçari B, São José, Novo Amparo, and Novo Amparo Norte to thenorth along the 8 km strike length of the Rio Jacaré intrusion.

This area became the focus of the drill program to explore for additional mineralization along strike fromthe known resources at Gulçari A. The plan was to do enough drilling on the satellite deposits in order tobring them into the inferred mineral resource category at this time. A program comprised of 13,400 m ofdrilling in 72 holes has been completed and the initial modeling of the new satellite deposits has beencarried out as is described in this report.

It is recommended that any Inferred resources identified in the new satellite deposits be upgraded to theMeasured and Indicated category through an additional definition drilling program. A suitable programwould comprise of 21,200 m of drilling in 194 holes.

Micon and Coffey Mining have reviewed this proposed resource upgrade drilling program and budgetand, in light of the observations made during their site visit and noted in this report, find it reasonable and

warranted. It is recommended that Largo proceed with the program.

Mineral Processing and Metallurgical Testing

The following recommendations for the Mineral Processing and Metallurgical Testing should beconsidered for this expanded case and include the ore from the satellite deposits:

The amount of water required to wash the leach residue to achieve the stated PLS concentrationand minimize vanadium losses should be confirmed;

The ability to wash the AMV cake minimizing re-dissolution of vanadium should be confirmed; The new fuel requirements for roasting the concentrate and heat losses should be confirmed by

modeling with vendor’s pr oprietary design using the new increased feed and production data;

The calciner energy requirements and off-gas composition should be quantified by modeling withvendor’s proprietary design and specific gas emission testing;

The calciner off-gas from the calcining tests should be analyzed to provide a basis for treatmentrequirements;

Identification of corrosive gases from the calcination process in the testwork to date suggests futuretestwork is required to quantify the content and volume of gas flows for materials selection incalcining equipment and gas-handling ductwork;





Critical equipment capacities and size should be confirmed with vendors in view of the newproduction scheme, keeping a contingency factor as per industry practice;

Mass and water balance should be re-run to verify usage and design/operation criteria used for thiseconomic evaluation.

In addition, the metallurgical parameters through specific pilot-scale tests should be confirmed for the

expansion case:

The required wash water on the leach residue to achieve the stated vanadium losses and thevanadium concentration in the PLS should be confirmed;

Confirmation that the target vanadium concentration in the PLS is still economically achievable withlower grade concentrates;

The vanadium losses through dissolution during washing of the AMV filter cake should beconfirmed;

The compliance of calcine-leach vanadium recovery with the values stated in this report and usedas basis for the economic analysis should be confirmed;

Gas volume and composition during roasting should be verified;

Available technology and associated costs for producing ferrovanadium should be confirmed.

Open Pit Mining

The following recommendations for the Mine Operation should be considered:

Geotechnical investigation to refine pit slope stability, including the satellite deposits, should beintroduced as mining progresses;

During Year 1 a drill and blast study should be introduced to optimize drilling and blastingparameters for all rock types;

Due to the strong demand for mining equipment it is recommended that Largo advance to thepurchasing phase of securing mining equipment to avoid an impact to the project schedule due to

potential long lead times; Further study to determine the viability of expanding the mineable ore resource through underground

mining methods should be carried out; Further refinement of the mine plan to smooth the material handling requirements, hence, the

equipment fleet size requirements over the life of the project should be carried out; The waste placement strategy should be further studied in order to minimize waste haulage distance

and cost; The economic impact of the use of remote crushing facilities for ore transportation from the satellite

pits to the main plant site should be analyzed; Additional topographic data in the vicinities of the satellite pits and their associated development

areas should be acquired to improve on the final designs of open pits, waste piles, drainage systemsetc.

Hydrology

The hydrological characteristics and water balance model should be reconciled with seasonal rainfall andvalidated by means of a stochastic model within the first two years of operation. The estimated cost tocomplete this work is USD 122,000.





Environmental Studies

The following recommendations for the Environmental Studies should be considered:

All tailings storage facilities have been designed for the original 15-year mine plan outlined in theDecember Report. It is recommended that all storage facilities including the Leached Calcine

Tailings Dump, the Chloride Tailings Pond and the Non-Magnetic Tailings Dump be redesigned forthe storage requirements of this expansion case, including their characteristics, placement anddevelopment schedule;

Testing should be carried out on the waste material of the satellite deposits for acid generatingproperties, and the waste placement strategy should be further refined in order to minimize the riskof any potential acid rock drainage;

The existing water management plan should be further developed through the use of dykes andpossibly ditches for Gulçari A and all satellite deposits in order to mitigate against risk of pit f loodingand acid rock drainage;

Condemnation drilling for satellite pits needs to be completed; Baseline studies for each pit and the overall site to identify the impact of the satellite pit footprint to

develop the EIA for the future mining activities.





2 Introduction

This Preliminary Economic Assessment (PEA) was produced as a replacement to the previous TechnicalReport produced in 2012 (Arsenault, et al). The main differences between the mine plan presented atthis time (known as the expanded case) as compared to the previous plan (known as the base case) areas follows:

The Gulçari A Pit has been expanded to include a second phase pushback; Five additional satellite deposits have been added to the mine plan, comprising Gulçari A Norte,

Gulçari B, São José, Novo Amparo and Novo Amparo Norte; The plant output has increased from 960,000 t/a with an average feed grade of 1.35% V2O5 to

1,400,000 t/a with an average feed grade of 1.10 V2O5; Open pit mining increases from 2.8M t/a to 11.1M t/a, on average, over the life of the project; The mine life increases from 15 years to 29 years; The production of ferrovanadium employs an aluminothermic reaction as opposed to a DC furnace; Results from metallurgical testing indicate that it is more economical to produce vanadium

pentoxide (V2O5) for life of the project, converting 65% of it to ferrovanadium (FeV) from Year 5onward.

In 2012 Largo completed a revised block model and updated mineral resource estimate for the existingGulçari A deposit and five new satellite deposits. These have been discussed in Item 1 Summary andpresented in detail in Item 14 Mineral Resource Estimates. However, it should be noted that the mineralresource estimate used to design the mine and the processing plant and used in the capital and operatingcost estimates are the same as those used in the 2009 Technical Report.

This Report is intended to be used by Largo to further the development of the Maracás Project byproviding an estimate of resources and reserves, classification of resources and reserves (in accordancewith the Canadian Institute of Mining, Metallurgy and Petroleum (CIM) Classification system) and an

economic evaluation of the Maracás Vanadium Project located in the municipality of Maracás, BahiaState, Brazil.

Largo may use the Report for any lawful purposes to which it is suited. The Report has been prepared ingeneral accordance with the guidelines provided in NI 43-101 Standards for Mineral Projects.

The economic evaluation contained in this Report specifically pertains to the mineral resources (as atDecember 2012) associated with the Gulçari A, Gulçari A Norte, Gulçari B, São José, Novo Amparo andNovo Amparo Norte deposits.

2.1 Use of Report

This Report is intended to be used by Largo subject to the terms and conditions of its contract with RPM.That contract permits Largo to file this Report as a Technical Report with Canadian Securities Regulatory

Authorities pursuant to provincial securities legislation. Except for the purposes legislated underprovincial securities laws, any other use of this Report by any third party is at that party’s sole risk.





2.2 Terms of Reference

In preparing this report, RPM was responsible for those items listed below:

Open pit mine design; Review of Capital and Operating Costs prepared by others; Cash Flows using data supplied by others; Economic Evaluation; Preparation of Report according to NI 43-101 format.

It should be noted that Items 7-12 and Item 14 were prepared under separate contract to Largo by MiconInternational Limited (Micon) and Coffey Mining Pty Ltd (Coffey Mining). Item 19 was prepared exclusivelyby Micon.

Capital cost estimates are expressed in first quarter 2013 United States dollars (USD) with no allowancefor escalation, interest costs or financing during construction. Cost estimates and factors were preparedby Largo. The capital costs are based on designs prepared by Promon and have an average level of

accuracy of +/- 15%.

Operating costs were prepared by Largo with support from Promon for each phase of the operation andinclude operating labour, fuel, replacement parts, operating supplies, maintenance labour and supplies,plant consumables, power and shipping for plant and infrastructure.

Capital and operating cost estimates including mine services and consumables for the mine werereviewed by RPM. All project cost estimates including mine, plant and infrastructure were reviewed byRPM.

Actual operating costs, production and economic returns may differ materially from those anticipated bythe technical report, and depend on a variety of factors, some of which are outside the control of the

Company.

2.3 Sources of Information

Information on resources and reserves determination used in the preparation of this Report was obtainedfrom Largo officials and technical staff for the major part. The information was taken from companydocuments or obtained from Largo staff in personal communications.

A complete list of the consultants and contributors who have provided supporting reports and data for theReport is provided in Item 3 Reliance on Other Experts.

2.4 Site Visits

Site visits were completed by the following personnel associated with the preparation of this Report:

RungePincockMinarco; Kevin Tanas, P.Eng.; Mining, Processing, Infrastructure, Cost Estimates Rod Ogilvie, P.Geo., P.Eng.; Geology Micon; B. Terrence Hennessey, P. Geo.; Geology and Resource Estimates RungePincockMinarco; David De Lemos Pires, B.Sc., M. Sc; Open Pit Mining, Infrastructure, Cost

Estimates, Reserve Estimates





Coffey Mining Pty Ltd; Hebert Lopes Oliveira, BSc Geologist Eng, MAIG; Geology and ResourceEstimates

2.5 Terms and Units

The following terms and definitions are used in this report.

Largo refers to Largo Resources Ltd. RPM refers to RungePincockMinarco and its representatives. Gulçari A Project refers to the Gulçari A Vanadium Project located in the, Maracás Municipality,

Bahia State, Brazil, including the proposed mine area, process plant location, and other relatedfacilities.

DNPM refers to the National Department for Mineral Production (Departemento Nacional deProdução Mineral.

SEMA/BA refers to the State Secretary of Environment (Secretaria de Estado do Meio Ambiente)for Bahia State.

RPM has based all measurements in the metric system, and has identified exceptions to this, notablywhen listing both English and Metric standards. Currencies are generally based on January 2013 USDollar, converting to the Brazilian Real at 2.0 Real per US Dollar for the long-term exchange rate. A shortterm exchange rate of 2.0 Real per US Dollar was used in the capital and operational expenditureestimates to reflect current exchange rates. Unless otherwise stated, Dollars are United States Dollars,and weights are in metric tonnes of 1,000 kilograms (2,204.62 pounds). The following abbreviations areused in this report:

Abbreviation Acronym

AA Atomic Absorption

AC Air conditioning

ADIS Automated Digital Imaging System

ADR Adsorption, Desorption, Recovery

Ag Silver

AGP Acid Generation Potential

AMV Ammonium Metavanadate

ANA Angencia Nacional de Aguas

ANFO Ammonium Nitrate Fuel Oil

ARD Acid Rock Drainage

As Arsenic

ASC Aluminum Standard

ASV Vegetation Suppression Authorization

Au Gold

AuEq Gold equivalent

B.O.O. Build/own/operate

BFA Bench face angle

CAPEX Capital Expenditure

CBPM Companhia Baiana de Pesquisa






CCA Capital cost allowances

CEA Cumulative Expenditure Account

CEMA Conveyor Equipment Manufacturers Association

CEPRAM State Council for Environmental Matters

CIC Carbon in columnCIM Canadian Institute of Mining, Metallurgy and Petroleum

CO Carbon monoxide

CO2 Carbon dioxide

COG Cut-off grade

CSMAT Controlled Source Audio-Magnetotelluric Tensor

Cu Copper

CuEq Copper equivalent

DCS Distributed Control Systems

DDIP Dipole Dipole Induced Polarisation

DFS Definitive Feasibility StudyDIA Declaración de Impacto Ambiental

EIA Environmental Impact Assessment

EM (VLF) Electromagnetic, very low frequency

EMPs Environmental Management Plans

EPC Engineer, procure, construct

EPCM Engineer, procure, construction management

ESIA Environmental and Social Impact Assessment

EW Electrowinning

FCA Free carrier allowed

FeV FerrovanadiumFOB Free on board

FOREX Foreign Exchange

GA General Arrangement

GDP Gross Domestic Product

HDPE High Density Polyethylene

HG Magnetite

HSEC Health, Safety, Environment & Community

HV High voltage

HVAC Heating, ventilation and air conditioning

ICP Inductively Coupled Plasma

ID2 Inverse Distance Squared

HDI Human Development Index

IFC International Finance Committee

IFS Initial Feasibility Study

IIA Informe de Impacto Ambiental

IMO International Maritime Organisation






IP Induced polarization

IRR Internal Rate of Return

kWh Kilowatt hour

LAM Lithology; Alteration and Mineralization

LI Installation LicenseLL Localization License

LME London Metal Exchange

LOM Life of mine

LV Level voltage

MCC Motor control centres

MCE Maximum credible earthquake

MDE Maximum design earthquake

MG Magnetite Pyroxenite

MII Measured, Indicated and Inferred resources

MISC MiscellaneousMLI McClelland Laboratories International

Mo Molybdenum

MoS2 Molybdenum disulphide

MTO Material take off

MP Particulate Matter

MV Medium voltage

MW Megawatt

NAG Net Acid Generation

NaHS Sodium hydro sulphide

Nox Nitrous oxidesNPC Net Present Cost

NPV Net Present Value

OK Ordinary Kriging

OL Operational License

OPEX Operating Expenditure

P Au Gold price

Pb Lead

PCu Copper price

PEA Preliminary Economic Assessment

PFS Prefeasibility StudyPGM Platinum group minerals

PRAD Plan Rehabilitation of Degraded Areas

PSA Pit slope angle

PTSV Technical Project for Vegetation Suppression

QA/QC Quality assurance/Quality control

R Au Projected gold recovery






RCu Projected copper recovery

RIMA Relatório de Impacto Ambiental

RL Relative Level

RMR Rock mass rating

ROM Run of mineRQD Rock Quality Designation

SECA Sistema de Estudo Climaticos e Ambientais

SFC Secretaria de Conservação de Florestas

SG Supergene

SG Specific Gravity

SGS Lakefield Research Group

SOx Sulphur oxides

SXEW Solvent extraction electrowinning

TBD To be determined

TR Reference TermsheetUPS Uninterruptible power supplies

UTM Universal Transverse Mercator (coordinate system)

V2O5 Vanadium Pentoxide

VALE Companhia Vale de Rico Doce

VFD Variable frequency drive

VML Vanádio de Maracás Ltda

VSD Variable speed drive

XRD Mineralogical characterization

XRF X-ray fusion

The following Units of Measure are used in this report:

Unit Abbreviation

American Dollar USD

Bond Ball Mill Work Index (metric) kWh/t

Brazil Real $R

Canadian Dollar CAD

Centigrade

Centimeter Cm

Cubic metre M3

Cubic metres per second M3/s

Day D

Dead weight ton (imperial ton – long ton) Dwt

Dry metric tonne Dmt

Foot/feet Ft

Gram G





Unit Abbreviation

Gram/litre g/L

Gram/tonne g/t

Hour H

Hours per Year h/yr

Kilogram KgKilogram per tonne Kg/t

Kilometer Km

Kilopascal kPa

Kilovolt kV

Kilovolt amp kVA

Kilowatt kW

Kilowatt hour kWh

Litre L

Litre per second L/s

Megawatt MWMetre M

Metre per hour m/h

Metre per second m/s

Metric tonne T

Metric tonne per hour t/h

Metric tonner per day t/d

Micron Mm

Milligram Mg

Milligram per litre Mg/L

Millimeter MmMillion M

Million tonnes Mt

Million tonnes per annum Mt/a

Parts per billion ppb

Parts per million ppm

Percent %

Second S

Short ton T

Square metres M2

Tonnes per Annum t/a

Tonnes per Day t/d

Troy ounce Oz

Wet metric tonne Wmt

Work index WI

Year yr





The following Chemical Symbols are used in this report:

Common Chemical Symbols

Aluminum AlVanadium VCalcium Ca

Chlorine ClCobalt CoCopper CuGold AuIron FeLead PbMagnesium MgManganese MnMolybdenum MoNitrous Oxides NOx Oxygen O2

Potassium KSilver AgSulfur STitanium TiPlatinum PtLead PbVanadium Pentoxide V2O5

Vanadium Trioxide V2O3





3 Reliance on Other Experts

The authors of this Report state that they are the Qualified Persons as identified in Table 3-1.

Table 3-1 Key Project Personnel

During the preparation of this Report, RPM has relied on the contributions of a variety of specialistconsultants who have provided reports and studies for the Technical Report. RPM has not audited thesereports.

The financial analysis that RPM have prepared are based on commodity prices and owner’s cost

estimates provided by Largo, which have not been audited by RPM.

Permitting status and the present status of mining rights and areas cited in this document have beenprovided by Largo and have not been audited by RPM.

Assistance and information was obtained from the following sources:

3.1 Geology, Mineralogy, Resource Estimates (Micon)

Micon has reviewed and analyzed exploration data provided by Largo and its consultants, and has drawnits own conclusions therefrom, augmented by its direct field examination. Micon has not carried out anyindependent exploration work, drilled any holes or carried out any significant program of sampling andassaying. However, the presence of vanadium-bearing mineralization in the area is substantiated by theprior exploration history.

Grab samples of mineralization exposed in exploration trenches on the property, or in the historical drillcore, and collected by Micon, independently confirm the presence of vanadium mineralization at the

project.

While exercising all reasonable diligence in checking, confirming and testing it, Micon has relied upon thedata presented by Largo, and previous operators of the project, in formulating its opinion.

The various agreements under which Largo holds title to the mineral lands for this project have not beenthoroughly investigated or confirmed by Micon and Micon offers no opinion as to the validity of the mineraltitle claimed. The descriptions were provided by Largo.

Don Arsenault, P.Eng. RungePincockMinarco

B. Terrence Hennessey, P.Geo. Micon International Limited

Hebert Lopes Oliveira, BSc, MAIG Coffey Mining Pty Ltd

Jane Spooner, MSc, P.Geo. Micon International Limited

Kevin Tanas, P.Eng. RungePincockMinarco

Scott Weston, MSc, P.Geo. Hemmera Envirochem Inc.





The description of the property is presented here for general information purposes only, as required by NI43-101. Micon is not qualified to provide professional opinion on issues related to mining and explorationtitle or land tenure, royalties, permitting and legal and environmental matters. Accordingly, the author hasrelied upon the representations of the issuer, Largo, for Section 4 of this report, and has not verified theinformation presented therein.

3.2 Mineral Processing and Metallurgical Testing

The following reports were used as a basis for the design and the development of associated plant capitaland operating cost estimates:

Lurgi, “Feasibility Study, Maracás Vanadium Project, prepared for Pedeiras Valeria Ltda.,

Salvador/Bahia, Brazil”, may 1986. Rautaruukki Oy Tutkimuskeskus Research Centre, “Laboratory Research of the Suitability of the

Otankäki Process for Extracting Vanadium from Maracás Ore”, December 1989. Engenharia e Consultoria Mineral S.A., “Projeto Vanádio de Maracás Projecto Conceitual e

Estimativa de Investimento, Produção: 4,500 t/a de V2O5”, September 1990.

IMS Process Plant, “Vanádio de Maracás Ltda., Vanadium Pentoxide Production Plant”, 1990. SGS Minerals Services, “The Beneficiation Characteristics of Samples from The Vanádio De

Maracás Deposit” November 2007 SGS Minerals Services, “Recovery of Vanadium from the Maracás Ore Deposit”. April 2008. SGS Minerals Services, “The Solid-Liquid Separation of the Maracás Ore Deposit”, July 2008 Vendors’ budgetary quotes Largo Resources Ltd. (Les Ford), “Pilot Plant Testing of Maracás Magnetite Ore”, Oct 2010. Ausenco Minerals and Metals, “Conceptual design of alternatives for non-magnetic tailings

deposition”, Sep 2010.

3.3 History

The history of the Maracás property has been previously described by Menezes (2005). Information forthis Item 6 is taken from Brio et al (1981), J. Menezes (2005) and other unpublished documents fromprevious operators.

3.4 Market Studies and Contracts

The product pricing information used in Item 19 Market Studies and Contracts was provided to RPM byLargo, and originates from the Metal Bulletin. Micon used this information along with market outlook datafrom Roskill Information Services to develop Item 19 of this report.

3.5 Environmental Studies, Permitting and Social or Community Impact

Baseline study by Brandt Meio Ambiente Ltda, dated May 2008 and completed by Fontes &Handall Ambiental Ltda dated June 2011

Equator Principles Compliance Audit carried out by Mineral Engenharia e Meio Ambiente Ltdadated December 2011

Archaeological studies were carried out in 2007 by the company Arqueologia Brasil – Projetos,Pesuisas e Planejamento Cultural e Arqueologico Ltda.





In order to verify the impacts of the emission of pollutants a mathematical simulation was carriedout by SECA – Sistema de Estudos Climaticos e Ambientais, an independent company based inSão Paulo

Hydrological and geological characterization studies were completed in 2008 by VOGBR andrevised in 2011

A review of the basic design of the Non-Magnetic Dry Stack Tailings Dump was completed by

Ausenco Minerals in 2010 on the basis of the geotechnical parameters provided by VOGBR The design and location of the waste pile and ore stockpiles for Gulçari A were determined by NCL

do Brasil based on the parameters provided by VOGBR.

All of the above studies were summarized by Largo and presented to RPM for review.

3.6 Infrastructure References

The infrastructure design was completed by Promon Engenharia (Promon) from their São Paulo officesand presented to RPM for review.

3.7 Cost EstimateThe CAPEX and OPEX estimates were developed by Largo with support by Promon and verified againstofficial quotes and cross checked against Brazilian industry resources by RPM.

3.8 Geology, Mineralogy, Resource Estimates (Coffey Mining)

This report was prepared as a National Instrument 43-101 Technical Report, in accordance with Form 43-101F1, for Largo by Coffey Mining Pty Ltd (Coffey Mining). The quality of information and conclusionscontained herein are consistent with the level of effort involved in Coffey Mining’s services and based on:

information available at the time of preparation by Largo, third party technical reports prepared by Government agencies and previous tenements holders,

along with other relevant published and unpublished third party information, and the assumptions, conditions ,and qualifications set forth in this report. This report is intended to be

used by Largo, subject to the terms and conditions of its contract with Coffey Mining. This contractpermits Largo to file this report as a Technical Report with Canadian Securities Regulatory

Authorities pursuant to National Instrument 43-101, Standards of Disclosure for Mineral Projects. Any other use of this report by any third party is at that party’s sole risk.

A final draft of this report was provided to Largo, along with a written request to identify any materialerrors or omissions, prior to lodgment.

The authors of this report are not qualified to provide extensive comment on legal and environmentalissues associated with the Largo exploration permits in Brazil included in Section 4 of this report.

Assessment of these aspects has relied on information provided by Largo and has not beenindependently verified by Coffey Mining.

During the diligence in reviewing the satellite deposits and geological database used to estimate themineral resource, Coffey Mining confirmed and tested it.





The geological database, mineralization, exploration descriptions used in this document were provided byLargo, its contracted consultants or previous owners or project operators of the property. Coffey Mininghas review this work and confirmed it with its diligence during the site visit.

Similarly, neither Coffey Mining nor the author of this report is qualified to provide comment onenvironmental issues associated with Largo.

No warranty or guarantee, be it express or implied, is made by Coffey Mining with respect to thecompleteness or accuracy of the legal or environmental aspects of this document. Coffey Mining does notundertake or accept any responsibility or liability in any way whatsoever to any person or entity in respectof these parts of this document, or any errors in or omissions from it, whether arising from negligence orany other basis in law whatsoever.





4 Property Description and Location

4.1 Location

The Maracás Project is located within the greater municipality of Maracás in eastern Bahia State, Brazil.Maracás lies about 250 km southwest of the City of Salvador, the capital of Bahia. The distance by roadfrom Salvador to the Project is 405 kilometers via a paved secondary road from the main coastal highwayin Bahia, with the project being accessed by about 50 km of secondary highway and gravel road west ofthe town of Maracás. Access to water, the electric power grid and a railroad is within reasonabledistance, and a trained workforce familiar with the mining and mineral exploration industries exists in thestate of Bahia and also within in the country generally.

Maracás is a farming and agricultural support community and its inhabitants are reported to welcome therenewal of mineral exploration. The exploration permits are isolated and there are no adjacentexploration, or mining properties.

Figure 4-1 Maracás Project Location Map





4.2 Property Status

The Maracás property consists of twenty (20) exploration permits totaling 28,587.06 hectares (see Figure4-2).

Figure 4-2 Maracás Project Property

These exploration permits are granted and listed in the following table:





Table 4-1 Maracás Property Exploration Permits

Three of these exploration permits highlighted in pink on the map including the original two, DNPM870134/82 and DNPM 870135/82, are owned by Vanádio de Maracás Ltda. (VML) which is controlled99.9% directly and indirectly by Largo. The remaining 17 exploration permits, highlighted with a hatchpattern, are owned by Largo (see Figure 4-2). These exploration permits run contiguously north-southwith each exploration permit forming rectangle of various dimensions and cover the prospective horizonfor over 40 kilometres. The UTM Geographic zone is 24L and the Datum used for the area is Corrego

Alegre. The centre of the original two exploration permits is at Latitude 13º 41' S, Longitude 40º 40' W.

Largo reports that all but two exploration permits are currently registered as exploration licenses and arein good standing. An application to convert two of the exploration licenses to exploitation licenses (mininglicenses) has been made. Grant of the mining license is pending, however, the Localization License (LL)and the Instillation License (LI) for the Project have been granted. There are no fees due on theexploration permits until the exploitation license has been granted.

Largo entered into an agreement with VALE and Odebrecht, dated October 16, 2006 giving it an option toacquire a 90% interest in the Maracás property from the two Brazilian companies for a purchase price ofUSD10.0 million. Under the agreement, Largo maintained the exploration permits in good standing. On

Requested

AreaCity

DNPM no Date (há) U.F.:BA



VML 872.346/2010 14/10/2010 Vanádio 977.29 Maracás

Largo 871.550/2007 24/05/2007 Vanádio 1,877.30 Iramaia / Maracás

Largo 871.551/2007 24/05/2007 Vanádio 1,300.00 Man. Vitorino / Iramaia


Largo 871.961/2008 8/04/2008 Vanádio 1,273.41 Maracás / Iramaia




Largo 871.246/2010 28/06/2010 Vanádio 1,000.00 Manoel VitorinoLargo 871.085/2010 28/06/2010 Vanádio 2,000.00 Manoel Vitorino


Largo 871.944/2010 6/09/2010 Vanádio 2,000.00 M. V itor./Iramaia/ M arac.

Largo 871.943/2010 6/09/2010 Vanádio 2,000.00 Iramaia, Maracás

Largo 871.941/2010 6/09/2010 Vanádio 927 Maracás




Largo 873.878/2011 27/09/2011 Vanádio 1,709.53 Maracás/Iramaia

20 26,301.90

Holder DNPM Protocol

Substance

Total





December 22, 2012 Largo bought out VALE and Odebrecht giving Largo 100% ownership of Companhiade Maracás.

Largo reports that, to its knowledge, there are no existing environmental liabilities with the property.





5 Accessibility, Climate, Local Resources, Infrastructure and Physiography

5.1 Access

The town of Maracás is accessible by paved secondary highway from the main Brazilian coastal highwaythrough Bahia state. It is approximately 405 road kilometres from Salvador. The Project is accessed bypaved secondary highway west from Maracás for a distance of 29 km followed by 20 km of gravel road toa ranch gate. The Project is located on the ranch and a 2.5-km sand and gravel trail leads to the smallhill where the Gulçari A deposit outcrops.

Maracás has a small general aviation airstrip but no commercial air service. Salvador, being a statecapital and one of the larger cities in Brazil, is served by an international airport with many flights a dayfrom São Paulo and Brasilia.

5.2 Infrastructure

Domestic power and telephone service are available both at the property and in the town of Maracás,which is linked to the power grid. Maracás has a population of approximately 20,000. Water is availablefrom a number of rivers and creeks which drain the general area.

Brazil has a large and very active mining industry. Infrastructure for mining equipment, services andpersonnel are available in a number of centres including São Paulo, Belo Horizonte and Cuiaba. TheJacobina gold mine is located 275 km to the north of Maracás. There are several other small activemines in the general area and thus some local mining services are available in Salvador. There is a railline close to the property and deep water port facilities are available at Salvador. The village of Porto

Alegre is located south of the project.

5.3 Climate and Physiography

The local climate has two distinct seasons, one is typically hot and humid and the other a dry seasonwhich occurs in the winter. This climate is characteristic of much of the interior of the state of Bahia. Theaverage day time temperature in the interior is near 30ºC. The temperatures drop to a low in May andJune when minimum day time temperatures stay above 10ºC. Daytime temperatures rise to 40ºC inJanuary and February. Rainfall is about 1000 mm/a.

The rainy season runs from November to March. During that time, the rains are intense, and thetemperature is high. Some low-lying areas can experience flooding. The dry season is from July toSeptember. The climate does not create any problem for exploration with diamond drilling or othergeological / geochemical work. Tropical weathering can create specific issues for geochemistry and

mapping. Exploration can be carried out at any time without difficulty.

Approximately 23% of the State of Bahia lies at less than 300-m asl, 70% is between 300- to 900-m asland 7% is above 900 m. There are three types of relief observed, high plateau, coastal, and areasextending between the coast and the high plateau. The Maracás property is located in the regionbetween the coast and the high plateau in an area of moderate to low-lying relief.





At the Project site itself, the maximum relief is about 30 m. The surrounding terrain is typical ranch / farmland with low trees and shrubs and consists of a number of relatively flat plateaus adjacent to a series ofcreeks and ponds. The property is bounded to the east by a steep cliff that rises 300 m to an area ofhigher land where the town of Maracás is located. Figure 5-1 is a photograph of the Maracás propertywith the hill on which Gulçari A outcrops visible in the background.

Figure 5-1 Maracás Project With Gulçari A Hill in Background

Occasional outcrops of pegmatite dike and gabbro are present on the property. The local overburden,which consists of alluvial soils, ranges from 3 to 10 m in thickness.

The local land is used primarily for agriculture with ranching being the primary activity on the land at theMaracás Project exploration permits.





6 History

The history of the Maracás property has been previously described by Menezes (2005). Information inthis section is taken from Brito et al (1981), Galvão et al (1984), Menezes (2005) and other unpublishedinternal documents from previous operators.

6.1 Summary

Exploration of the Rio Jacaré mafic to ultramafic intrusion by the geologists of CBPM started in 1980during a regional geological survey. This work led to the discovery of the vanadium-rich titaniferousmagnetite occurrence on what now is part of the Maracás property. In 1981, CBPM conducted anexploration program which included geological mapping, ground geophysical surveys (magnetic and VLFelectromagnetic surveys), test pitting and trenching, and diamond drilling of two holes totalling 147 m. In1983, CBPM continued work and focused on the Gulçari A deposit when it completed an additional 12holes totalling 985 m.

Over the past 27 years, the Maracás Project has undergone several additional phases of exploration andeconomic evaluation, including geophysical surveys, prospecting, trenching, diamond drilling programs,geological studies, resource estimates, petrographic studies, metallurgical studies, mining studies andeconomic analyses. These studies have progressively enhanced the knowledge base for the Gulçari Adeposit. The following is an historical summary of work that has taken place since the CBPMinvolvement, taken, in part, from the reports by CBPM (1981 and 1984), a report by Menezes (2005), andpresent information from existing reports.

6.2 Exploration History

In 1984, CBPM formed a joint venture with the Odebrecht Group, after the conclusion of a prefeasibilitystudy completed by CEPED, the State of Bahia Research and Engineering Centre. The joint ventureformed a new company, Vanádio de Maracás Ltda., in order to explore and develop a mine andmetallurgical plant to exploit the vanadium-bearing titaniferous magnetite deposit.

During that year, Odebrecht carried out systematic exploration work including 1492 m of diamond drillingin 18 holes on the Gulçari A deposit. Over the next 3 years (1985 to 1987), Odebrecht completed threemore drill programs to further define the resource at Gulçari A including nine holes totalling 971 m, eightholes totalling 1136 m and four holes for 421 m, respectively. An additional 24 vertical holes totalling 648m were drilled for geotechnical information regarding open-pit boundaries and overburden. These holeswere not analyzed or included in the database used for the resource estimate presented in this Report.Odebrecht also carried out an exploration program testing the three other prospects on the property(Gulçari B, São José and Nova Amparo) which included geological mapping, ground geophysical surveys

(magnetics and VLF), trenching, and diamond drilling of 13 holes totalling 661 m.

In 1986, CBPM and Odebrecht completed a “reserve” estimate for the Gulçari A deposit, which is

discussed in Item 8.3 below.

Odebrecht conducted a number of petrographic studies, metallurgical tests and feasibility studiesintermittently from 1984 to 1988. These were performed with recognized engineering companies such asCEPED from Brazil, Lurgi GMBH and Gesellschaft für Electrometallurgie (GFE) from Germany, Mintek





from South Africa, Rautaruukki Oy from Finland, Jaakko Poyry Engineering and Engenharia e ConsultoriaMineral S.A. (ECM), both from Brazil. The conclusion of these studies in 1990 resulted in what wasdetermined to be a feasible project for the production of 4000 t/a of vanadium pentoxide (V 2O5) with partconverted to ferrovanadium as a final, value-added product. As a consequence of the evaluation, the

joint venture decided to contract the Finnish company, Rautaruukki Oy, for the implementation of finalpilot plant tests and basic engineering design for the Project.

Early in the 1990s, Odebrecht, owning 93% of Vanádio de Maracás’ shares, decided to reform the joint

venture on a 50 / 50 basis with CAEMI with the intention of bringing to the Project expertise in mining,metallurgy and marketing.

In 1990, a sampling program of the drill core from Maracás was conducted. The samples were alsoanalyzed for platinum and palladium. A total of 167 samples from 10 drillholes were selected for analysis.Ninety-six samples were from magnetite and 71 samples from pyroxenite. The results indicatedpotentially significant platinum and palladium values associated with high-grade V2O5 values.

The following is a list of studies completed on the property by the above listed companies:

CBPM Geological Study Lurgi GMBH Feasibility Study Rautaruukki Oy Feasibility Study Jaakko Poyry Feasibility Study Natron Environmental Impact Study ECM Feasibility Study CAEMI / MBR 1996 Revision of Feasibility Study CRU Market Study CRA Market Study IMS Process Plant Study MINTEK Test Reports

Paulo Abib Geo-Statistical Evaluation & Mining Plan VML - DNPM Economic Development Plan 1996 CAEMI Feasibility Study 1999 Economic Update Report of 1996 Study.

In 2006, Largo signed an agreement with Odebrecht, and CAEMI for the Maracás property. Largo carriedout a resampling program to analyze a portion (approximately 10%) of the old drill core (CBPM andOdebrecht) to verify the past analytical database on the property (see Item 11). Analyses were done atSGS Minerals (SGS) laboratories, both in Belo Horizonte, Brazil and Lakefield, Ontario. The sampleswere analyzed for FeO, Fe2O3, SiO2, TiO2 and V2O5 by the XRF method and for platinum and palladiumby a 50-g fire assay technique. Based on the verification of the database, Largo completed a revisedblock model and NI 43-101-compliant mineral resource estimate which was the subject of an earlier

report (Hennessey, 2006).

6.3 Historical Drilling

Between 1981 and 1987, previous Owners of the Project drilled 66 holes totalling 5814 m, testing fourdeposits on the Maracás property. These consist, from south to north, of the Gulçari A, Gulçari B, SãoJosé and Nova Amparo deposits. A summary of the drilling completed, by deposit, is set out in Table 6-1.





Table 6-1 Summary of Diamond drilling, Maracás Property

The vast majority of work has been done on the Gulçari A deposit which is the focus of this Report.Previous diamond drilling on the Gulçari A deposit, completed by CBPM and Odebrecht is summarized inTable 6-2. The analytical results from this diamond drilling are the basis for the subsequent, historical“reserve” estimates and metallurgical and feasibility studies on the deposit.

Table 6-2 Historical Diamond Drilling, Gulçari A Deposit (1981 – 1987)

6.4 Historical Resource Estimates

As discussed above, a mineral “reserve” was estimated for the Project in 1986. This historical mineral

reserve was based on detailed geological mapping, geological sections incorporating structural geologyand mineralogical and analytical results, sampling of 46 trenches totalling 1950 m and 53 diamonddrillholes totalling 5153 m, density tests and ore microscopy. The work was prepared by staff geologistsfrom both CBPM and Odebrecht.

The Odebrecht “geological reserve” was a kriged estimation done on a block model with block

dimensions of 5 m by 5 m by 15 m. At the time, the mineral inventory was classified as measuredreser ves above 150 m of vertical depth. This historical mineral “reserve” estimate, as summarized inMenezes (2005), is set out in Table 6-3. The historical estimate was reported at various cut-off grades asshown in the table and is not compliant with 43-101.

No. of Length

Deposit Holes (m)

Gulçari A 53 5,153

Gulçari B 4 169

Sao Jose 2 115 Nova Amparo 7 377

TOTAL 66 5,814

Period

No. Of

Holes Length (m)

1981 2 147

1983 12 985

1984 18 1,492

1985 9 971

1986 8 1,136

1987 4 421

53 5,153 TOTAL

CBPM/Odebrecht

CBPM/Odebrecht

CBPM/Odebrecht

CBPM

CBPM

CBPM/Odebrecht

Company





Table 6-3 Historical “Reserve” Estimate (1986) - Gulçari A

Largo embarked on a program to verify the database and provide an NI 43-101-compliant mineral

resource estimate for the deposit (Hennessey, 2006). That estimate was then updated in 2007(Hennessey, 2007).

From the information provided above, Paulo Abib Engenharia S.A. built up a complete geological andanalytical database in the 1990s. This database was used in a geostatistical study of the deposit wheregrades were interpolated into a block model by ordinary kriging. The geostatistics were also used togenerate a variographic analysis of the deposit.

First, the deposit was interpreted and modeled on section into one titaniferous magnetite lens and twobroad envelopes of lower-grade mineralization. The estimate was prepared using a block model withblock dimensions of 5 m by 5 m by 15 m. The comparison between the kriging results on cross-sections

and the geological cross-sections showed remarkable similarity.

6.5 Historical Technical and Environmental Studies

In 2008 Aker Solutions were retained by Largo to complete a Definitive Feasibility Study (“DFS”) for theMaracás Vanadium Project. The results of that study were presented in a NI 43-101 Technical Reporttitled “Technical Report of the Feasibility Study for the Maracás Vanadium Project, Brazil”, Amended

version dated May 2009.

Since the completion of the DFS, Largo has continued to advance the project. Additional studies andtechnical effort have been completed in the following areas;

Pilot scale metallurgical testing at Fundação Gorceix, Ouro Preto, August- September 2010 Conceptual design of alternatives for disposal of non-magnetic tailings, Ausenco Minerals, May-

June 2010 Dry stacking feasibility study, Ausenco Minerals, September 2010 Land Agreement with Banco Economico S/A, , to secure land rights for mining and project

development at São Conrado and São Pedro da Goiania land proprieties, July 2011 Environmental permitting with the issuance of:

o Localization License - CEPRAM Resolution nº 3941, 10-10-2008

Cut-off Grade Average Grade Tonnes

(%V2O5) (%V2O5) (millions)

- 1.13 13.2

0.1 1.13 13.1

0.2 1.14 13.00.3 1.19 12.4

0.4 1.28 11.4

0.5 1.37 10.4

0.6 1.45 9.6

0.7 1.53 8.8

0.8 1.62 7.9

0.9 1.72 7.1

1.0 1.82 6.4





o Installation (Construction) License - INEMA Resolution nº 1286, 10/20/2011o Grant of Water Rights- ANA resolution nº 684, 09-16-2011

Air pollutant emissions modeling and simulation by SECA, February 2012 13,401 meters of additional resource drilling in Gulçari A and B trends (not yet incorporated into the

geological model) Promon Engenharia - basic engineering work for hydrometallurgical plant and infrastructure,

process flowsheet re-evaluation and design, infrastructure engineering, capital and operating costupdates, production throughput review, March - November 2011.

HYDROS Engineering- basic engineering work for water pipeline and, capital and operating costsestimates, May - November 2011

VOGBR, basic engineering of main geotechnical structures, waste dump, tailings facility, sitedrainage design, hydrogeological studies and costs estimates, May - November 2011

RPM, basic mine design, with Capex and Opex estimates, May - November 2011 Project financing led by Bank ITAÚ BBA

This Technical Report incorporates the results of the updated engineering and environmental studies.





7 Geological Setting and Mineralization

In Item 7 both Micon and Coffey Mining have provided individual sections. These have been designatedwith the letter “A” for Micon and “B” for Coffey Mining.

Items 7.1 to 7.2 are common for both Micon and Coffey Mining.

Items 7.3.A and 7.4.A have been prepared by Micon and deal with the Gulçari A deposit.

Items 7.3.B and 7.4.B have been prepared by Coffey Mining and deal with the satellite deposits of Gulçari A North, Gulçari B, São José, Novo Amparo and Novo Amparo North.

7.1 Regional Geology

Brito (2000), Sá et al. (2005) and Teixeira et al. (2000) have described the regional geological setting forthe Maracás property. These references give a detailed description of the geotectonic evolution of the

São Francisco craton. The following is a brief summary of their work.

The Rio Jacaré intrusion, which hosts the Maracás Project vanadium mineralization, is located in thesouth-central part of Bahia state in northeastern Brazil. It lies within the Archean São Francisco craton,which in this area is composed of the Contendas-Mirante Complex and the Gavião and Jequié blocks(see Figure 7-1).

The intrusion is located on the eastern edge of the Contendas-Mirante supracrustal sequence, whichforms a large anticlinorium trending approximately north-south. The supracrustal rocks are locatedbetween the early Archean Gavião block to the west, which is composed predominantly of tonalite-trondhjemite granodiorite, and the Archean Jequié block to the east, which is composed predominantly ofcharnockite and enderbite intrusive rocks with strong calc-alkaline affinities and granulite facies

metamorphic rocks (Teixeira et al., 2000).

The Contendas-Mirante sequence is thought to be younger than the adjacent Gavião and Jequié blocksand consists of an Archean basal volcanic unit overlain by a Paleoproterozoic member containing flyschand metavolcanic rocks that are overlain by a clastic member. A Rb/Sr age of 2.0 Ga for the granite,derived from melting of the Contendas-Mirante metapelites, corresponds to the timing of the Tran

Amazonian orogeny (2.14 to 1.94 Ga; Teixeira et al., 2000). The Contendas-Mirante sequence wasdeformed by the collision of the Gavião and Jequié blocks during the Tran Amazonian orogeny and isnow located along part of the major Contendas-Jacobina lineament (Teixeira et al., 2000).





Figure 7-1 Maracás Area Simplified Regional Geology Map

7.2 Rio Jacaré IntrusionThe Rio Jacaré mafic-ultramafic intrusion is composed mainly of gabbro. It is a linear sheet-like structurethat strikes almost north-south, with a length of approximately 70 km, an average width of 1.2 km, and adip of 70° E. The intrusion has been described previously as a sill intruded into the volcanic rocks of thelower unit of the Contendas-Mirante gneissic complex (Brito, 1984; Galvão et al., 1986). However, theRio Jacaré intrusion is fault bounded to the east and west, and, therefore, its contacts with both theContendas-Mirante sequence and Jequié block are tectonic.

The age of the intrusion is poorly known. Whole rock dating of rocks from the intrusion itself includes aPb/Pb age of 2.47 Ga ±72 Ma, a Sm/Nd age of 2.8 Ga ± 68 Ma, and a zircon age of 2.64 Ga ± 5 Ma (Britoet al., 2001). The intrusion is cut by granitic pegmatite veins that are closely related to a granite intrusion

that has an age of 1.94 Ga ±54 Ma (Brito et al., 2001).

Metamorphism and deformation have modified many of the igneous textures and minerals of theintrusion. Relict minerals are rare, but some igneous textures are still preserved such as olivine cumulatetextures and layering between pyroxenite and gabbro. The pyroxene in these rock types is now largelyaltered to hornblende, which is in turn replaced by actinolite, tremolite and chlorite in many samples. Thepresence of amphibole and garnet in the gabbro and magnetitite (an igneous rock composed largely ofmagnetite) in the Rio Jacaré intrusion indicates amphibolite grade metamorphism.





The intrusion has been divided into an Upper and a Lower zone. The Lower zone is approximately 400-mthick and consists of gabbro with some diorite and minor anorthosite. Anorthosite also occurs as a layernear the bottom of the Lower zone, and has a mean thickness of 15 m. It is composed of plagioclase withminor quartz and chlorite that replaces pyroxene.

The gabbro is massive, coarse grained, and slightly foliated, whereas the diorite is massive and mainly

fine grained. The primary igneous mineralogy of the gabbro consisted of plagioclase and orthopyroxeneas cumulate phases, with interstitial clinopyroxene. The orthopyroxene, clinopyroxene and olivinemineralogy has been examined in detail by Brito (2000). Quartz and biotite are present as minor phases,and apatite and titanite are common accessory phases.

Within the Lower zone, there are lenses of magnetite-rich rocks. The outer margins of the lenses consistof magnetite-bearing pyroxenite with 30% to 70% opaque minerals. The centres of the lenses consist ofmassive magnetite (magnetitite). These bodies were previously described as forming pipes and plugsintruded into the gabbro of the Lower zone (Brito, 1984). However, the contact relationships with thegabbro are not clear, because the bodies are usually bounded by faults and are poorly exposed. Theyare described in greater detail below.

The Upper zone has an average thickness of 600 m and is formed mainly of layered gabbro varying fromleucogabbro to melagabbro with some cyclic units of gabbro, pyroxenite, magnetite-bearing pyroxenite,and magnetitite. The pyroxenite consists of thin layers, typically a few centimetres to less than 1 m inthickness, and which are in many cases associated with the magnetite bodies.

7.3.A Property Geology (Gulçari A Deposit)

The north-south trending Rio Jacaré intrusion underlying the Maracás exploration permits can be tracedfor the full 8 km of strike length which occurs on the property. Along this trend, hosted within the RioJacaré intrusion, occur the main vanadium-rich magnetite bodies in four known locations, Gulçari A,Gulçari B, São José and Novo Amparo, from south to north (see Figure 7-2). The intrusion trends N 20°

E and dips 70° southeast.

The Gulçari A deposit is hosted in the gabbro of the Lower zone, whereas the other three occur in theUpper zone.





Figure 7-2 Maracás Property Geology Map

The Lower zone consists of four members including; medium to coarse-grained mesocratic gabbros,melano-gabbro consisting of banded pyroxenite and gabbro, two magnetite horizons within gabbros andpyroxenite and a tonalite member. The vanadium-rich magnetite layer is associated with the mesocraticgabbro. All the units, including the deposit, are cut by later pegmatite dikes.





The Gulçari A body contains the largest concentrations of vanadium-rich magnetite known on theproperty (Brito, 1984). This deposit crops out over an area of approximately 400-m along strike, up to150-m width and is known to extend to approximately 350-m vertical depth, where it remains open. Thedeposit has been disrupted by northwest-southeast faulting.

It is composed of magnetitite grading into magnetite-rich pyroxenite, pyroxenite, and then gabbro which

contains layers or lenses of magnetite-bearing pyroxenite that are sometimes sheared (see Figure 7-3).The main magnetitite body is on average about 25-m thick and thins to the south.

Figure 7-3 Gulçari A Deposit, Detailed Surface Geology Map

It has been suggested that, although the magnetite-pyroxenite body of Gulçari A is disrupted by faulting, itis a part of the main igneous sequence of the Rio Jacaré intrusion rather than a separate pipe-likeintrusion crosscutting the gabbro (Sá, 1992).

Figure 7-4 and Figure 7-5 are example cross-sections through the Gulçari A deposit, looking north and

showing drilling and interpreted geology.

As in Gulçari A, the Gulçari B, São José and Novo Amparo bodies consist of magnetitite closelyassociated with pyroxenite layers and hosted in gabbro. The magnetitite layers have a width between 8and 13 m and lengths of up to 250 m, with the layers being truncated by faulting. The magnetite layers inthe Upper zone have sharp magmatic contacts with gabbro below and gradational contacts with thegabbro above.





Figure 7-4 Gulçari A Deposit - Drill Section 6175N





Figure 7-5 Gulçari A Deposit - Drill Section 6225N

7.4.A Mineralization (Gulçari A Deposit)The vanadium at Maracás is associated with the titaniferous magnetite. Within the deposit, thetitaniferous magnetite is the major oxide phase, followed by ilmenite. Magnetite occurs as primary grainsthat may be partly martitized. There is also fine-grained magnetite as inclusions in the silicate grains.This magnetite occurs as a secondary alteration after uralitization of pyroxene and serpentinization ofolivine.

The magnetite normally occurs as anhedral grains, with grain sizes of between 0.3 and 2.0 mm, andforms a polygonal mosaic together with ilmenite. Ilmenite also occurs as inclusions in the magnetite,commonly displaying exsolution textures. Silicate phases associated with the magnetite include uralite,augite, plagioclase, hornblende, and rare grains of clinopyroxene, olivine and spinel. Magnetite from the

magnetitite in the Lower zone (Gulçari A) has higher V2O5 concentrations (2.2% to 4.5%) than magnetitefrom the Upper zone (Gulçari B and Novo Amparo, 0.3% to 2.5% V2O5; Brito, 2000).

Rare olivine and pyroxene grains are observed within the magnetitite, but most are altered to serpentineor chlorite. The Rio Jacaré intrusion has been intensely metamorphosed, so the pyroxene compositionsobserved probably reflect metamorphic re-equilibration rather than original magmatic compositions. Inaddition, Brito (2000) also documented the presence of orthopyroxene. Garnet and biotite are present inthe Gulçari B and Novo Amparo deposits.





Sulphides account for up to 1% of the rock in the magnetitite. The major phases are chalcopyrite andpentlandite with only very minor pyrite and pyrrhotite. Chalcopyrite is much more abundant than the othersulphides and is most common in the rock types containing 50% magnetite or less. It commonly occursinterstitially in magnetite or ilmenite enclosed by amphibole and plagioclase. Pentlandite is much lessabundant and occurs in the magnetitite. The pentlandite tends to occur interstitially to the magnetite andilmenite in silicates but, locally, composite grains of pyrrhotite, pentlandite, and chalcopyrite are enclosed

in magnetite. Minor sphalerite and galena grains are found together in the silicates, associated with theother sulphides especially in the magnetite-poor rock types. However, the dominant trace minerals arenickel and cobalt sulpharsenides and arsenides and cobalt-rich pentlandite. In many cases the arsenidesare associated with the sulphides and appear to be alteration products of the sulphides.

High platinum and palladium values have been found in the magnetite zones in the Rio Jacaré intrusion.They are much richer in platinum-group metals than the surrounding silicate rocks, and there aresignificant correlations among all of the PGMs and between PGM and copper.

In the magnetite zones, palladium-rich minerals, especially bismuthides and antimonides, are the mostabundant PGM minerals. In most cases, these occur with interstitial silicates or within silicate inclusions

in magnetite and ilmenite grains, and are associated with pentlandite and, in a few cases, with arsenides.Sperrylite is the most abundant platinum mineral and is associated with silicates interstitial to magnetiteand ilmenite grains. At sites where the igneous mafic minerals have been altered to amphiboles,sperrylite may be altered to platinum-iron alloys.

It is suggested that copper, nickel and PGM were concentrated in the magnetite layers by thecoprecipitation of a small quantity of sulphide with the magnetite. These PGM-bearing base metalsulphides subsequently exsolved the platinum minerals. The association of palladium minerals with basemetal sulphides and the small variation in the Pt/Pd ratio (4:1) suggests that the PGMs have not beenextensively remobilized in the magnetite.

The association of PGM enrichment with magnetite layers in the Rio Jacaré intrusion has similarities with

that of the Rincón del Tigre, Skaergaard and Stella Complexes. This enrichment is rarely associated withvisible sulphides, but suggests a possible new target for PGM exploration.

The Rio Jacaré intrusion hosts massive magnetite pod-like bodies confined to a layered sequence ofmafic and ultramafic cumulates. They are named the lower and upper magnetite seams. The lowermagnetite seam is represented by the Gulçari pod which occurs within the lower Transition zone. Thiszone is a 400-m-long, 150-m-thick pod tested to a vertical depth of about 350 m. It is a sequence ofmagnetite, pyroxenite and gabbro layers carrying vanadiferous iron ore with a mean grade of 2% V 2O5 that displays PGM values up to 5 ppm Pt, 1.7 ppm Pd and an average grade of 400 ppb total PGMs.

Magnetite layers are interbedded with ultramafic and mafic cumulates. The ultramafic cumulates aretransgressive towards the contact with the Lower zone gabbros and consist of olivine-magnetite

cumulates, and clinopyroxene-magnetite heteradcumulates. The massive magnetite layers are made upof ilmenite-magnetite heteradcumulates that form 2 cm to 3 m thick layers containing variable amounts ofclinopyroxene. Associated mafic cumulates are rhythmically micro-layered gabbro, magnetite, andmagnetite-pyroxenite bands.

The outer contacts of the magnetite pods exhibit hornblende-rich rocks and dunite in places. Thesefeatures are suggestive of a zoned pattern that Brito (1984) interpreted as similar to the magnetite pipe-like bodies of the Bushveld Complex. The upper Transition zone pod-like magnetite bodies are groupings





of magnetite seams and pyroxenites that form 150 m long, 20 m thick masses of vanadiferous iron orewith mean grade of 0.5% V2O5 and maximum total PGM contents of 1.3 ppm and mean grade of 380 ppb.The upper Transition zone also contains a low-grade copper sulphide mineralization which is confined tothe lowermost layers of the upper magnetite seams.

7.3.B Property Geology (Satellite Deposits)

The N20°E-oriented and 70° SE-dipping Paleoproterozoic Rio Jacaré Layered Sill (RJLS) occursthroughout the 40 km long Maracás Project exploration permits. Several vanadium-rich titanomagnetiteoccurrences have been identified within the sill. The occurrences are the following, from south to north:Gulçari A, Gulçari A North, Gulçari B, São José, Novo Amparo and Novo Amparo North (Figure 7-6).

The RJLS was subjected to detailed geological mapping and magnetometric ground geophysical surveythat led to the location of the abovementioned vanadium-rich titanomagnetite occurrences, which aresimilar to the known Gulçari A vanadium deposit. This fact shows that the RJLS has a great anduntouched potential for the discovery of new blind vanadium deposits like the Novo Amparo NorthDeposit.

The RJLS is divided in two magmatic stratified zones, as follows: The Lower Zone that consists of fourmembers including: a medium to coarse-grained mesocratic gabbros; a melano-gabbro consisting ofbanded pyroxenite and gabbro; two magnetite-rich horizons within gabbros and pyroxenites; and, atonalite intrusive. The Upper zone has an average thickness of 600 m and is formed mainly of layeredgabbro varying from leucogabbro to melano-gabbro with some cyclic units of gabbro, pyroxenite,magnetite-bearing pyroxenite, and magnetitite. The pyroxenite consists of thin layers, typically a fewcentimeters to less than 1 m in thickness, and which are in many cases associated with the magnetitebodies.

The Gulçari A, Gulçari A North and São José deposits are found within the gabbro unit of the Lower Zoneof the RJLS, while the Gulçari B, Novo Amparo and Novo Amparo North occur in its Upper Zone.





Figure 7-6 Property Geological Map With the Deposits

The vanadium-rich magnetite layer is associated with the mesocratic gabbro. All the rock units of theRJLS are cut by a later (1.9 Ga) Paleoproterozoic pegmatite dike swarm.

The Gulçari A body contains the largest concentrations of vanadium-rich magnetite known on theproperty (Brito, 1984). This deposit crops out over an area of approximately 400 m along strike, up to 150m width and is known to extend to approximately 350 m vertical depth, where it remains open. The

deposit has been disrupted by NW-trending faults.

7.3.1.B Novo Amparo North Deposit

The Novo Amparo North Deposit is the northernmost deposit and it corresponds to two vanadium-bearingmassive magnetite layers with 620 meters in strike length. Figure 7-7 shows the integrated geologic andmagnetic map of the Novo Amparo Norte deposit.





The geology of this deposit is represented by gabbro green to gray in color, fine to medium grain, mainlycomposed of plagioclase, pyroxene, amphibole and biotite. Magnetite gabbro showing dark gray in color,strong foliation, very magnetic (25% of magnetic oxides).

Figure 7-7 Integrate Map of NAN Deposit

This deposit is overlain by Tertiary cover, and it was detected by the ground magnetometry survey carriedout in 2011. At that time it was executed a drilling program aimed to test this magnetic anomalies and thefirst drillhole crossed a massive magnetite layer of 20 meters thick. All drillholes executed along theanomaly intercepted the mineralized zones. Figure 7-8 shows a representative cross section of this

deposit.





Figure 7-8 Cross Section of NAN Deposit

The layer has an average width of 18.6 meters, with an average grade of 0,87 % of V2O5.The massivemagnetite is hosted by magnetite gabbro, and sometimes in the gabbro. This rock are dark gray to blackin color, fine to medium grained, with more than 60% of magnetic oxides, pyroxene, amphibole, garnet

and disseminated sulphide (pyrite). In some places exhibit small interbedded layer of magnetitepyroxenite, magnetite gabbro and anorthosite.

The entire set has direction N20° E with a dip of 70° SE and is cut by transverse faults. At the north of thearea, the mineralized zone is interrupted by a fault of NW-SE direction.

Diamond drilling at the Novo Amparo North Deposit amounts to 17 diamond drillholes, totalling 3.281meters, located in a 50 meters spaced lines grid.

7.3.2.B Novo Amparo Deposit

The Novo Amparo Deposit is located in the Upper Zone of the RJLS. The mineralized body presents itself

as a tabular body with 285meters in length and widths ranging from 11 to21 meters, and the averagegrade of 0,72 % of V2O5.It has its north and south ends truncated by faulting. All the rocks deposit areoriented in the direction NNE/SSW, with dip of 70°SE. Figure 7-9 shows the integrated geologic andmagnetic map.





Figure 7-9 Integrated Map of NA Deposit

The deposit consists of a variety of amphibolitized gabbros, leucocratic, fine to medium grain, stronglyfoliated and it has a composition: plagioclase, biotite and amphibole and garnet, similar to the Gulçari Bdeposit.

The massive ore corresponds to the lithotype over 60% of magnetic oxides. It has dark gray to black colorand massive structure. The foliation, when present, is mainly observed in the oriented gangue minerals.

The main oxide is titanomagnetite and gangue consists of amphibole, biotite, chlorite and garnet. Interlayered with the massive ore are common the small levels of magnetite pyroxenite, magnetite gabbro andpyroxenite.

The magnetite gabbro occurs in the western part of the deposit and are characterized by fine grainsometimes coarse, composed of plagioclase, amphibole, titanomagnetite, and disseminated sulphide(pyrite and chalcopyrite).The deposit is cut longitudinally by powerful bodies of quartz-feldspar pegmatite.

These pegmatites cross the mineralized zone. Structurally have direction N 30° and dips 60-70°to theNW. Figure 7-10 is a typical cross section of the NA deposit.





Figure 7-10 Cross Section of NA Deposit

In Novo Amparo deposit were executed 21drillholes totalling1632 meters, located in a 25 meters spacedlines grid.

7.3.3.B São José Deposit

The São José deposit like Novo Amparo and Gulçari B Deposit is situated in the Upper Zone of the RJLS.It is characterized by two blind mineralized bodies: The first one is 524 meters long and 11 meters thick,and the second one is 404 meters in length and 5 meters thick. The average grade of the deposit is0,89% of V2O5. Figure 7-11 shows the integrated map of São José Deposit.





Figure 7-11 Integrated Map of SJ Deposit

The two mineralized zones have a NNE/SSW strike and a strong foliation with a dip of 65° to SE. This

deposit is hosted by magnetite gabbro on the East side, characterized by a coarse grained, foliated,composed of plagioclase, and amphibole, with garnet and magnetite. The west side the ore body is incontact with fine grain gabbro strongly foliated with narrow bands of pyroxenite. The Magnetitite are darkgray to black, fine to medium grained and massive structure. Figure 7-12 shows the representative crosssection of the São José deposit.

The São José deposit is divided in two parts. In the east part 11 drillholes were driven, 2 of them made byCBPM and 9 made by Largo. In the west part 15 drillholes were driven by Largo. The total drilled lengthwas 4.713 meters. This report considers only the west part, with sections every 50 meters.





Figure 7-12 Cross Section of SJ Deposit

7.3.4.B Gulçari B Deposit

The Gulçari B Deposit is situated within the Upper Zone of the RJLS, and it is 200 meters in strike length,and 4 to 13 meters thick, and 100 meters depth, with an average grade of 0.70% V 2O5. Geological

contacts are all fault-bounded. It displays a tabular geometry made up mainly by a massive magnetitehorizon hosted by gabbros and subordinate pyroxenite lenses. Late pegmatite dikes cut the entirevanadium deposit at an attitude of N20°E, in strike length, and a dip of 65° to NW.

The main rock-type in this deposit is a fine to coarse-grained isotropic and leucocratic amphibolitizedgabbro, oriented to NNE-SSW, with dips of 65° to 75° to SE. Banded structures are locally shown.Discordant pegmatite dikes are also present. Figure 7-13 shows the integrated map of this deposit.





Figure 7-13 Integrated Map of GB Deposit

A common feature of these gabbros is the presence of garnet, sulfides (pyrite and chalcopyrite) anddisseminated magnetite. This unit is referred as the Magnetite Gabbro and it contains some 0.3% V2O5.They consist mainly of amphibole and plagioclase in varying proportions, and sometimes occuranorthositic terms. Figure 7-14 shows the typical cross section of Gulçari B deposit.





Figure 7-14 Cross Section of GB Deposit

The magnetitite bodies are gray to black in color, mainly fine grain, but locally coarse-grained, with oxides(mainly titanomagnetita) above 70%. The other oxides present in decreasing order of participation are:

Ilmenite and ulvospinel. The gangue is composed of pyroxene, hornblende, biotite and garnet.

Inter layered with the magnetic ore bodies occur subordinate strata of gabbros and pyroxenites. The firstare the leucocratic to mesocratic, fine to medium grained, composed mainly ofamphibole and plagioclase,with the presence of sulphides (pyrite and chalcopyrite). Sometimes they present a anorthositiccomposition. The pyroxenites are foliated, amphibolitized, with fine disseminations of titanomagnetite.

The Gulçari B deposit has been investigated by 7 diamond drillholes executed by CBPM and 10 diamonddrillholes by Largo, totaling 1622 meters, with sections spaced of 25 meters.

7.3.5.B Gulçari A North Deposit

The Gulçari A North deposit is located north of the Gulçari A deposit, hosted by the Lower Zone of theRJLS. It is characterized by two massive magnetite horizons. The main body is 1 km long and 7.5 metersin average thickness, and the second body with 350meters long with a width of 4 meters, located east ofthe first. The average grade of the depositis0,84% V2O5. Figure 7-15 shows the geological map of Gulçari

A North deposit.





Figure 7-15 Integrated Map of GAN Deposit

The geology of the deposit is represented by the following rock types: magnetite-gabbro with medium tocoarse grain with magnetic oxides around 25%. It has disseminated sulfide and remobilized into fractures.Locally, small layers of massive magnetite occur.

Fine to medium grained gabbros with a composition of plagioclase, pyroxene, amphibole and biotite, with

narrow layers of magnetite-gabbro, pyroxenite, and anorthosites. The massive magnetite has black colorwith 60-70% of magnetic oxides with disseminated sulphide (pyrite and chalcopyrite) and small layers ofmagnetite- pyroxenite and anorthosite.

The two mineralized zones have direction N20° E with dip ranging from 60-65° to the SE. In the Southpart of the deposit the main ore body is cut by fault with NW / SE direction. Figure 7-16 shows therepresentative cross section of GAN deposit.





Figure 7-16 Cross Section of GAN Deposit

Due to the presence of significant layers of magnetite pyroxenite and pyroxenite in all drillholes thisdeposit has many similarities with the Gulçari A deposit, maybe a north ward continuation of this deposit.

Were executed on the Gulçari A North deposit a total of 17drillholes with sections every 100 meters.

7.4.B Mineralization

The vanadium at Maracás is associated with the titaniferous magnetite. Within the deposit the titaniferousmagnetite is the major oxide phase, followed by ilmenite. Magnetite occurs as primary grains that may bepartly martitized. There is also fine grained magnetite as inclusions in the silicate grains. This magnetiteoccurs as a secondary alteration after uralitization of pyroxene and serpentinization of olivine.

The magnetite normally occurs as anhedral grains, with grain sizes of between 0.3 and 2.0 mm and formsa polygonal mosaic together with ilmenite. Ilmenite also occurs as inclusions in the magnetite, commonlydisplaying exsolution textures. Silicate phases associated with the magnetite include uralite, augite,plagioclase, hornblende, and rare grains of clinopyroxene, olivine and spinel. Magnetite from themagnetitite in the Lower Zone (Gulçari A, Gulçari A North, and São José) has higher V2O5 concentrations(2.2 to 4.5%) than magnetite from the Upper Zone (Gulçari B, Novo Amparo, Novo Amparo North, 0.3 to2.5%; Brito, 2000).





Rare olivine and pyroxene grains are observed within the magnetitite but most are altered to serpentineor chlorite. The RJLS has been intensely metamorphosed so the pyroxene compositions observedprobably reflect metamorphic re-equilibration rather than original magmatic compositions. In addition,Brito (2000) also documented the presence of orthopyroxene. Garnet and biotite are present in theGulçari A North, Gulçari B, São José, Novo Amparo, and Novo Amparo North deposits.

Sulphides account for up to 1% of the rock in the magnetitite. The major phases are chalcopyrite andpentlandite with only very minor pyrite and pyrrhotite. Chalcopyrite is much more abundant than the othersulphides and is most common in the rock types containing 50% magnetite or less. It commonly occursinterstitially in magnetite or ilmenite enclosed by amphibole and plagioclase. Pentlandite is much lessabundant and occurs in the magnetitite. The pentlandite tends to occur interstitially to the magnetite andilmenite in silicates but, locally, composite grains of pyrrhotite, pentlandite, and chalcopyrite are enclosedin magnetite. Minor sphalerite and galena grains are found together in the silicates, associated with theother sulphides especially in the magnetite-poor rock types. However, the dominant trace minerals arenickel and cobalt sulpharsenides and arsenides and cobalt-rich pentlandite. In many cases the arsenidesare associated with the sulphides and appear to be alteration products of the sulphides.

High platinum and palladium values have been found in the magnetite zones in the RJLS. They are muchricher in platinum-group metals than the surrounding silicate rocks, and there are significant correlationsamong all of the PGM and between PGM and copper.

In the magnetite zones palladium-rich minerals, especially bismuthides and antimonides, are the mostabundant platinum minerals. In most cases these occur with interstitial silicates or within silicateinclusions in magnetite and ilmenite grains and are associated with pentlandite and, in a few cases, witharsenides. Sperrylite is the most abundant platinum mineral and is associated with silicates interstitial tomagnetite and ilmenite grains. At sites where the igneous mafic minerals have been altered toamphiboles, sperrylite may be altered to platinum-iron alloys. It is suggested that copper, nickel and PGMwere concentrated in the magnetite layers by the coprecipitation of a small quantity of sulphide with themagnetite. These PGM-bearing base metal sulphides subsequently exsolved the platinum minerals. The

association of palladium minerals with base metal sulphides and the small variation in the Pt/Pd ratio (4:1)suggests that the PGM have not been extensively remobilized in the magnetite.

The association of PGM enrichment with magnetite layers in the RJLS has similarities with that of theRincón del Tigre, Skaergaard and Stella Complexes. This enrichment is rarely associated with visiblesulphides but suggests a possible new target for PGM exploration.

The RJLS hosts massive magnetite pod-like bodies confined to a layered sequence of mafic andultramafic cumulates. They are named the lower and upper magnetite seams. The lower magnetite seamis represented by the Gulçari pod which occurs within the lower Transition zone. This zone is a 400-mlong, 150-m thick pod tested to a vertical depth of about 350 m. It is a sequence of magnetite, pyroxeniteand gabbro layers carrying vanadiferous iron ore with a mean grade of 2% V2O5 that displays PGM

values up to 5 ppm Pt, 1.7 ppm Pd and an average grade of 400 ppb total PGM.

Magnetite layers are interbedded with ultramafic and mafic cumulates. The ultramafic cumulates aretransgressive towards the contact with the Lower zone gabbros and consist of olivine-magnetitecumulates, and clinopyroxene-magnetite heteradcumulates. The massive magnetite layers are made upof ilmenite-magnetite heteradcumulates that form 2 cm to 3 m thick layers containing variable amounts ofclinopyroxene. Associated mafic cumulates are rhythmically micro-layered gabbro, magnetite, andmagnetite-pyroxenite bands.





The outer contacts of the magnetite pods exhibit hornblende-rich rocks and dunite in places. Thesefeatures are suggestive of a zoned pattern that Brito (1984) interpreted as similar to the magnetite pipe-like bodies of the Bushveld Complex (Willemse, 1979). The upper Transition zone pod-like magnetitebodies are groupings of magnetite seams and pyroxenites that form 150 m long, 20 m thick masses ofvanadiferous iron ore with mean grade of 0.5% of V2O5, and the total PGM contents of 1.3 ppm and meangrade of 380 ppb. The upper Transition zone also contains a low grade copper sulphide mineralization

which is confined to the lowermost layers of the upper magnetite seam.





8 Deposit Types

Work carried out to date indicates that the Maracás vanadium-rich titaniferous magnetite deposit issituated in a geologic environment similar to other magmatic vanadium-rich magnetite deposits. The RioJacaré intrusion that hosts the deposit is a mafic to ultramafic layered intrusion characterized by bothrhythmic and cryptic layering that includes a pyroxenite/gabbro layer in the upper part. This gabbro layerhosts the Gulçari A vanadium deposit. The deposit has been assigned this way by the manner in which itwas generated as an early magmatic ore type deposit formed by concentration through liquidimmiscibility.

The primary minerals of the gabbro layer are plagioclase, augite and magnetite. Secondary alterationhas altered most of the augite to uralite. Plagioclase occurs as elongated laths, the only idiomorphicmineral present, with the allotriomorphic minerals of magnetite and augite occurring as grains betweenthe plagioclase laths. Magnetite was the last mineral to crystallize and, because of the amount ofvanadium incorporated in it, the magnetite gabbro layer has the potential to have economic value. Thecrystal form of the magnetite does not indicate the concentration of the magnetite by an accumulation of

separate magnetite crystals; on the contrary, the textural features suggest the formation of a separateoxide liquid.

The enrichment of PGMs, in association with magnetite, has been described in other mafic-ultramaficcomplexes. In association with the Main Magnetite Layer in the Upper zone of the Bushveld Complex,PGMs are concentrated in anorthosite just below the layer. However, the magnetite seams themselvescontain very little PGMs.

In the Skaergaard intrusion in Greenland, palladium is concentrated with copper and gold in anorthositiclayers associated with magnetite at the top of the middle zone, having values of up to 3 ppm Pd and 200ppb Pt. In the Rincón del Tigre Complex in Bolivia, platinum and palladium are concentrated with copperat the base of the magnetite-bearing gabbro in the upper part of the intrusion with maximum precious

metals concentrations of 1.8 ppm Pd and 0.68 ppm Pt in separate 1-m samples. In all of these intrusions,magnetite formed after extensive fractionation of the magma.

Samples of oxide layers from the Birch Lake area of the Duluth Complex exhibit PGM-depleted, -undepleted, and -enriched mantle-normalized patterns similar to Rio Jacaré samples. The PGM patternsfor Birch Lake and Rio Jacaré are similar for platinum and palladium, but the Birch Lake patterns aregenerally enriched in osmium to rhodium compared to the Rio Jacaré samples. The Longear, WymanCreek and Boulder Lake magnetite units of the Duluth Complex also closely resemble the Rio Jacarédepleted and undepleted samples. This comparison suggests that the Rio Jacaré magnetitite is mostsimilar to those found in layered intrusions associated with continental tholeiites. The depleted RioJacaré PGM patterns are also similar to those obtained from the northern limb of the Upper zone of theBushveld.

The sulphide content of the Rio Jacaré rocks is very low and the rocks are metamorphosed. Sulphides inthe magnetite and magnetite-bearing pyroxenite of the Lower zone show enrichment in platinum,palladium, rhodium, iridium, gold and nickel, but not in copper, compared to the Upper zone.Recalculated sulphide concentrations for samples from Gulçari A are similar to those of PGM-enrichedsulphides from Birch Lake and Medvezhy Creek in the Norilsk Complex. In contrast, recalculated





sulphide concentrations of the Gulçari B and Novo Amparo samples resemble the PGM-depletedMinnimax sulphides and Dunka Road sulphides of the Duluth Complex.





9 Exploration

9.1 2006 Exploration Program

Largo signed a letter of intent to acquire the Maracás Project in October, 2006. As a result, there hadbeen no time to conduct any exploration on the property before the completion of the previous mineralresource estimate (Hennessey, 2006). During the late spring and summer of 2006, Largo conducted aprogram of technical due diligence prior to completion of the transaction. This work added to thedatabase of information for the property and may, therefore, be regarded as exploration.

As of the end of 2006, Largo had re-established the exploration grid on the Gulçari A deposit, surveyed it,and also check surveyed the drill-hole collars, most of which were marked with casing or plastic pipe.

Most of the CBPM and Odebrecht exploration drill core was intact and stored in a rentedoffice/warehouse facility in the town of Maracás. Largo staff re-logged much of this core and conducted asignificant program of check sampling. This program involved quarter-sawing of the same sample

intervals selected previously and check assaying for V2O5 and TiO2, as well as analyzing for PGMs. Theresults of this program are discussed in Item 12, Data Verification, below.


The 2007 exploration program conducted at Maracás by Largo consisted of the following:

175 line-km of line cutting 175 line-km of ground magnetic surveying 136 line-km of induced polarization (IP) surveying geological mapping of the property at a scale of 1:2,500 resampling old drillholes from 1981 through 1986 for PGMs surveying thin section and lithogeochemical studies diamond drilling, 61 holes totalling 13 876 m.

The surveying program was completed in order to put all of the previous work and future work into thesame grid system. It was decided to use UTM (Universal Transverse Mercator / Corrego Alegre)coordinates for the grid system. All drillholes and trenches completed since 1981 have been converted toUTM.

The entire property has been covered by 175 line-km of line cutting. The grid lines are 2.5 km long andoriented east-west with 100-m line spacing and 25-m stations along the lines. This line cutting work has

been done, in order to conduct geological mapping, sampling and ground geophysical surveys (magneticand IP). Geological mapping was done at a scale of 1:2,500 over the entire property concentrating onfavourable areas that have a limited amount of information. These include Gulçari A, Gulçari B, Novo

Amparo and São José. This work was completed in order to get a better understanding of the area’s

potential prior to conducting further drill testing.

Ground magnetic surveying was completed over the entire property. It was hoped that the magneticsurvey would help in understanding the geology that underlies the property and trace the magnetite-rich





horizons associated with the mineralization along strike and at depth. The results were reasonablyencouraging given that the magnetite horizons are good magnetic anomalies that respond reasonablystrongly.

A total of 136 line-km of IP surveying have been completed on the property. IP responds well to themagnetite and disseminated sulphide mineralization found at Maracás. Geophysical surveys are

considered important in this phase of work and their use will be discussed in the next section.

Data compilation, re-logging and additional resampling of previously drilled holes (1981 to 1986) wereundertaken. This work was done to correlate the lithologies between holes and from section to section,and to test the platinum and palladium potential of the deposit, in order to better understand thegeological setting and help in future work plans.

Petrographic analysis was carried out on 56 polished thin sections from drillholes representing the variousrock types, including highly mineralized samples from Gulçari A and Novo Amparo. They were used tocharacterize the rock types and mineralization in the immediate area around the deposits. Themineralized samples were also analyzed with an inductively coupled argon plasma (ICP) multi-elementpackage. Fresh, relatively unaltered samples were also chosen for whole rock analysis, in order tocharacterize the intrusive rocks in the belt.

A diamond-drill program of 61 holes totalling 13 876 m was completed in 2007 on the property. Theprogram began in the middle of February and ended in November. A more complete description of theprogram follows in Item 10 of this Report.

9.2.1 Previous Geophysical Surveys

A limited number of geophysical surveys were conducted by previous operators (CBPM, Odebrecht andCAEMI,) over the Maracás property during the period 1980 to 1986. These include magnetic and verylow frequency EM (VLF) surveys. Survey coverage was total grid for both. A review of this coverage is

beyond the scope of this Report. Any new drill targets will be generated by the new geophysical surveys.

9.2.2 Discussion of Present Geophysical Techniques

Systems such as magnetic and IP surveys are the optimum methods for detecting both massivemagnetite and disseminated sulphides in the Rio Jacaré belt. Both techniques generally give goodresponses to this style of mineralization. The advantage of spectral IP over traditional IP is in its ability todistinguish between strictly massive magnetite and a mixture of massive magnetite and disseminatedsulphides.

The total field magnetic responses reflect major changes in the magnetite content of the underlying rockunits. The amplitude of the magnetic responses relative to the regional background assists in identifying

specific magnetic and nonmagnetic units related to, for example, gabbro, pyroxenite, magnetite units,felsic intrusions and sedimentary rocks. Alteration and fault zones often have distinctive nonmagneticbelow background responses.

Spectral IP surveys involve measurement of the magnitude and relative phase of the polarization voltagethat results from the injection of an alternating current into the ground. Polarization voltages primarilyresult from electrochemical action within the pores and pore fluids of the material being energized.Measurements of the relative phase shift between the transmitted current and the measured signal and





the magnitude of the polarization voltage are taken over a range of different frequencies, typicallybetween 0.125 and 1000 Hz. This results in a distinct IP response spectrum or ‘dispersion’ at each

measurement position that can be characterized with Cole-Cole theoretical spectral parameters calledtau, M-IP and c, such as relaxation time and chargeability which are influenced by chargeable grain sizeand the type of chargeable source.

These spectral parameters complement chargeability and apparent resistivity data and have provensuccessful for detecting favourable gold and PGM mineralization.

A combination of magnetic and spectral IP techniques appears to be more diagnostic in detecting themassive magnetite and disseminated sulphide mineralization on the property, and was thusrecommended as the only geophysical tool remaining that can provide diagnostic survey coverage of theproperty. With the new advances in IP, the spectral IP method is a useful ground geophysical tool indetecting the disseminated sulphide mineralization along strike and at depth.

9.2.3 Geophysical Survey Results

Three relatively continuous, parallel to sub-parallel, magnetic trends can be traced north-south using theresults of the new ground magnetic survey data. These are associated, from west to east, with Gulçari A,Novo Amparo-São José-Gulçari B and a third unknown trend. The two western trends are associatedwith the Rio Jacaré layered mafic intrusion, whereas the eastern, most magnetic trend, is of unknownorigin. The few available outcrops suggest that the trend is underlain by felsic intrusive rocks. There is astrong possibility, however, that there are northeast-trending listric normal faults and that a magnetichorizon hosted in the mafic layered intrusion at depth may be responsible for the trend.

The IP conductors discovered are coincident with the trends of strong magnetic highs. In 2008 work toexamine the spectral IP parameters was ongoing and nearing completion. At that time Largo reportedthat initial indications were the spectral data suggested areas along the magnetic trend where there ispotential for disseminated sulphides. Largo also reported that these would become targets for future drill

programs at Maracás.


In 2008 Largo conducted a 5,000 m drill program at Maracás, largely designed to test IP deposits otherthan Gulçari A. This program is described in Section 10 below.

9.3.1 2011-2012 Exploration Program

In 2011 and 2012 Largo completed an additional diamond drilling campaign on the property consisting ofapproximately 13,400 m. While some of this drilling was on the Gulçari A deposit the majority was againtargeting other known anomalies on the property. The results of the program are described in Section 10

below.





10 Drilling

In Item 10, Drilling, both Micon and Coffey Mining have provided individual sections Although there issome duplication of information it was felt by RPM that the commentary was sufficiently different to justifypresenting both Micon and Coffey Mining data.

The Micon items have been designated with a letter “A” and the Coffey Mining items have beendesignated with the letter “B”.

10.1.A Drilling By Previous Operators

The drilling completed by previous operators, CBPM and Odebrecht, has been described earlier in thisReport in Item 6, above. Figure 10-1 is a plan view of the historic drilling that was available at the Gulçari

A deposit for the previous resource estimate (Hennessey, 2006).

10.2.A 2007 Largo Drill Program

During 2007, Largo completed a drill campaign consisting of 61 holes totalling 13 876 m as set out in theTable 10-1. The location of the 2007 drilling at Gulçari A is shown in red on Figure 7-3 above. Thissection of the Report is concerned solely with the updated mineral resource estimate at Gulçari A, andthe drilling on other deposits will not be discussed in detail.

Table 10-1 Largo 2007 Maracás Drill Program

Boart Longyear (Geoserv Pesquisas Geológicas S/A) began the program with one drill rig on February15, 2007, and added a second drill rig on March 5, 2007. Two rigs continued on the property until August19, 2007, at which time drilling was completed on the Gulçari A deposit. One rig was released and thesecond drill went to Novo Amparo where 11 holes totalling 1852 m were completed. The drill then coredfive regional holes testing geophysical deposits totalling 828 m. Drilling was completed on October 29,2007.

Boart Longyear drilled with NQ-sized core and averaged 1000 m per rig per month. Core recovery was

good with a reported average of 90%. Detailed drill-hole information for Gulçari A is listed in Table 10-2.

Areas Type/Purpose No. of Holes Total Metres

Gulçari A Resource 42 10 896

Gulçari A Metallurgical and Geotechnical 3 300

Novo Amparo Exploration 11 1 852

Regional Exploration 5 828 TOTAL 61 13 876





Figure 10-1 Gulçari A Deposit Drill-Hole Plan

The principal focus of the diamond-drill program was to upgrade the confidence of the previous Inferredresource at Gulçari A to the Measured and Indicated categories, and to expand upon it sufficiently to

demonstrate its potential economic viability. A secondary objective was to evaluate the PGM potential inthe Gulçari A deposit.





Table 10-2 Drill-Hole Summary for the 2007 Gulçari A Drill Program

* Holes drilled for metallurgical testing and geotechnical information.

The drilling at Novo Amparo was designed to test and characterize the mineralization 4 km to the northalong strike of Gulçari A and, in particular, the sulphide content and PGM potential of the mineralization.Finally, from the ground magnetic survey completed, five deposits were selected that had not been

UTM Coordinates

(Corrego Alegre) Elevation Azimuth Dip Depth

Drill Hole N E (m) (º) (º) (m)

FGA 56 8,486,090 318,346 294.04 290 -55 229.10

FGA 57 8,486,090 318,346 295.04 290 -70 265.10

FGA 58 8,486,125 318,353 294.96 290 -70 238.40 FGA 59 8,486,151 318,358 294.76 290 -45 138.60

FGA 60 8,486,032 318,360 297.52 290 -50 254.00

FGA 61 8,486,162 318,364 296.29 290 -60 139.60

FGA 62 8,486,074 318,332 296.09 290 -50 214.00

FGA 63 8,486,038 318,334 293.09 290 -50 229.30

FGA 64 8,486,074 318,332 295.09 290 -65 267.00

FGA 65 8,486,020 318,402 299.25 290 -50 197.50

FGA 66 8,486,073 318,409 299.18 290 -65 247.15

FGA 67 8,486,090 318,296 301.09 290 -45 206.60

FGA 68 8,486,111 318,405 297.73 290 -65 270.60

FGA 69 8,486,133 318,319 302.71 290 -45 151.00

FGA 70 8,486,053 318,213 302.74 290 -45 94.70

FGA 71 8,486,097 318,454 300.84 290 -65 236.20

FGA 72 8,486,020 318,336 296.00 290 -45 285.60

FGA 73 8,486,130 318,435 299.00 290 -50 224.55

FGA 74 8,486,130 318,435 299.00 290 -80 292.80

FGA 75 8,486,033 318,288 293.99 290 -45 267.60

FGA 76 8,486,011 318,281 293.68 290 -55 151.10

FGA 77 8,486,145 318,429 297.82 290 -60 159.70

FGA 78 8,486,000 318,320 295.64 290 -65 190.00

FGA 79 8,486,133 318,478 300.03 290 -60 229.00

FGA 80 8,485,941 318,301 294.70 290 -60 222.95

FGA 81* 8,486,101 318,250 302.40 290 -50 205.00

FGA 82 8,486,201 318,926 298.99 290 -60 169.40

FGA 83 8,486,065 318,380 298.05 290 -65 188.60

FGA 84* 8,486,110 318,226 302.40 290 -50 50.50

FGA 85* 8,486,120 318,205 313.30 290 -50 44.50

FGA 86 8,486,048 318,432 302.00 290 -65 214.10

FGA 87 8,486,079 318,523 305.00 290 -65 352.00

FGA 88 8,486,055 318,479 309.00 290 -65 336.64

FGA 89 8,486,051 318,620 304.00 290 -65 436.00

FGA 90 8,485,993 318,498 309.00 290 -65 404.25

FGA 91 8,486,106 318,594 300.00 290 -60 382.20

FGA 92 8,485,991 318,430 301.00 290 -60 351.80

FGA 93 8,486,007 318,530 304.00 290 -65 358.10

FGA 94 8,485,968 318,412 301.00 290 -65 336.00

FGA 95 8,486,068 318,567 301.41 290 -65 413.95

FGA 96 8,486,129 318,531 304.10 290 -60 369.65

FGA 97 8,486,130 318,435 298.00 290 -65 280.30

FGA 98 8,486,020 318,402 296.00 290 -73 325.00

FGA 99 8,485,955 318,465 302.00 290 -65 394.30 FGA100 8,485,962 318,348 296.00 290 -65 181.50

TOTAL 45 11,196





previously tested. These showings occur along magnetic trends that can be trace across the property for4 km from Novo Amparo in the north to Gulçari A in the south.

At the time the Gulçari A deposit, as outlined from the drill programs, extended 400 m along strike, to avertical depth of over 320 m with true widths ranging from 11 to 100 m with an average width of about 40m. This deposit is part of a mineralizing system that extends for 8 km across the property. All the results

from the drill program up to hole FGA-99 (the 2007 drill program) were completed and incorporated in theblock model at that time.

The results from the Gulçari A drill program are summarized in Table 10-3. The results for Novo Amparoand the five regional holes are not part of the updated Gulçari A mineral resource.

Of the 45 holes drilled at the Gulçari A deposit, 39 intersected wide well-mineralized zones. Table 10-3below is a summary of all significant Largo assay results from the drilling at Gulçari A up to the end of2007.

Table 10-3 2007 Gulçari A Drill Results

From To V2O5 Pt Pd PGM IntervalTrue

Thickness

Hole Number (m) (m) (%) (g/t) (g/t) (g/t) (m) (m)

FGA56 82.00 119.00 1.68 0.28 0.21 0.49 37.00 32.00

including 102.00 118.00 2.17 0.40 0.29 0.69 16.00 15.00

FGA57 104.00 140.10 1.85 0.27 0.21 0.48 36.10 30.00

including 121.00 138.10 2.31 0.36 0.28 0.66 19.10 17.00

FGA58 61.30 120.00 2.22 0.45 0.11 0.56 58.70 54.00

including 79.00 103.00 2.55 0.44 0.11 0.55 24.00 21.00

and 136.00 158.00 2.12 0.41 0.20 0.61 22.00 22.00

including 136.00 141.00 1.79 1.07 0.33 1.40 5.00 5.00

FGA59 50.00 122.00 1.88 0.52 0.08 0.60 72.00 72.00

including 65.00 73.00 2.65 0.91 0.10 1.01 8.00 8.00

FGA60 75.00 91.00 0.82 0.12 0.13 0.25 16.00 16.00

FGA61 76.00 121.00 1.97 0.44 0.11 0.55 45.00 44.00

including 95.00 106.00 2.64 0.57 0.21 0.78 11.00 10.00

FGA62 62.00 104.00 1.77 0.24 0.22 0.46 42.00 42.00

including 83.00 103.00 2.13 0.36 0.34 0.70 20.00 20.00

and 160.00 191.00 1.13 0.08 0.11 0.19 31.00 31.00






From To V2O5 Pt Pd PGM Interval

True

Thickness


FGA63 54.47 70.00 0.91 0.18 0.11 0.29 15.53 15.00

FGA64 72.00 116.00 1.41 0.24 0.21 0.45 44.00 42.00 including 96.87 114.34 2.03 0.37 0.33 0.70 17.47 16.00

and 238.05 262.40 1.27 0.44 0.28 0.72 24.35 23.00

FGA65 129.08 155.52 1.09 0.18 0.08 0.26 26.44 25.00

FGA66 136.91 182.80 1.46 0.22 0.18 0.40 45.89 43.00

including 156.94 182.00 2.03 0.39 0.31 0.70 25.06 23.00

FGA67 30.00 76.00 2.10 0.44 0.17 0.61 46.00 46.00

including 32.00 55.00 2.47 0.42 0.14 0.56 23.00 23.00

including 62.00 73.00 2.21 0.54 0.26 0.80 11.00 11.00

and 137.00 154.00 1.42 0.13 0.10 0.23 17.00 17.00

FGA68 105.00 147.00 1.80 0.36 0.14 0.50 42.00 40.00 including 124.00 147.00 2.25 0.40 0.19 0.59 23.00 21.00

and 168.00 217.00 1.64 0.28 0.22 0.50 49.00 47.00

FGA69 44.26 112.73 2.42 0.53 0.12 0.65 68.00 68.00

FGA70 23.00 36.00 2.15 0.21 0.09 0.30 13.00 13.00

FGA71 151.22 193.20 2.07 0.24 0.08 0.32 41.98 40.00

FGA72 211.55 224.60 1.30 0.12 0.08 0.20 13.05 12.00

and 229.90 239.15 1.43 0.13 0.11 0.24 9.25 9.00

and 249.15 268.15 1.06 0.04 0.05 0.09 19.00 18.00

FGA73 128.60 148.00 2.81 0.54 0.05 0.59 19.40 17.50

and 151.46 195.46 1.62 0.24 0.10 0.34 44.00 40.00

FGA74 136.38 175.38 1.86 0.24 0.03 0.27 39.00 39.00

including 148.38 173.38 2.15 0.21 0.04 0.25 25.00 25.00

FGA75 181.00 216.89 1.19 0.05 0.04 0.09 35.00 35.00

FGA76 89.00 118.00 1.43 0.08 0.06 0.14 29.00 29.00

including 105.00 118.00 1.93 0.09 0.08 0.17 13.00 13.00

and 128.00 135.00 1.32 0.06 0.08 0.14 7.00 7.00

FGA77 127.50 145.42 2.37 0.22 0.03 0.25 17.92 17.92

FGA78 106.95 117.40 0.68 0.08 0.07 0.15 10.45 9.00

FGA79 148.75 221.00 1.60 0.26 0.06 0.32 72.25 70.00

including 150.00 191.00 2.32 0.35 0.08 0.43 41.00 38.00

FGA80 83.70 90.70 0.69 0.07 0.04 0.11 7.00 7.00 and 94.70 98.70 1.02 0.09 0.11 0.20 4.00 4.00

and 114.10 118.10 1.22 0.09 0.08 0.17 4.00 4.00






Since the completion of the 2007 resource estimate report the results for hole FGA100 have becomeavailable. They are set out in Table 10-4.

From To V2O5 Pt Pd PGM Interval

True

Thickness


FGA82 No significant results

FGA83 134.00 152.93 1.77 0.23 0.21 0.44 18.93 18.93

including 139.00 152.93 1.99 0.26 0.25 0.51 13.93 13.93

and 159.72 172.72 1.18 0.05 0.06 0.11 13.00 13.00

FGA86 178.50 191.16 1.43 0.01 0.02 0.03 12.66 10.00

and 241.00 247.45 1.45 0.19 0.14 0.33 6.45 5.00

and 259.00 265.00 1.28 0.06 0.02 0.08 6.00 5.00

and 276.00 298.00 1.54 0.33 0.19 0.52 22.00 20.00

FGA87 244.50 259.83 1.93 0.16 0.21 0.37 15.33 15.33

and 267.00 289.00 1.82 0.19 0.22 0.41 22.00 22.00

and 306.00 323.00 0.86 0.05 0.06 0.11 17.00 17.00

FGA88 177.10 183.10 1.02 0.13 0.10 0.23 6.00 6.00 and 303.10 308.10 1.07 0.25 0.16 0.41 5.00 5.00

FGA89 297.00 356.00 2.11 0.19 0.07 0.26 59.00 59.00

including 302.00 338.00 2.40 0.20 0.07 0.27 36.00 36.00

FGA90 279.00 293.00 1.05 0.09 0.07 0.16 14.00 14.00

FGA91 285.60 294.60 1.32 0.21 0.10 0.31 9.00 9.00

FGA92 249.00 270.00 0.99 0.07 0.05 0.12 21.00 21.00

FGA93 264.00 278.00 1.94 0.24 0.18 0.42 14.00 14.00

and 299.00 309.00 1.31 0.10 0.07 0.17 10.00 10.00

FGA94 172.00 242.12 1.35 0.09 0.05 0.14 70.12 65.00

and 269.00 281.00 1.12 0.08 0.05 0.13 12.00 10.00

and 292.47 298.00 1.28 0.07 0.06 0.13 5.53 5.00

FGA95 268.80 336.00 1.46 0.11 0.10 0.21 67.20 65.00

including 268.80 284.60 2.24 0.09 0.09 0.18 15.80 15.80

FGA96 283.84 290.00 0.81 0.08 0.04 0.12 6.16 6.16

FGA97 142.73 193.00 1.83 0.28 0.19 0.47 50.27 50.27

including 143.73 173.00 2.26 0.34 0.16 0.50 29.27 29.27

and 218.00 228.00 1.40 0.30 0.30 0.60 10.00 10.00

FGA98 175.00 187.00 0.80 0.20 0.06 0.26 12.00 12.00

FGA99 224.00 233.00 1.38 0.10 0.11 0.21 9.00 9.00

and 237.00 241.00 1.55 0.06 0.04 0.10 4.00 4.00





Table 10-4 2007 Late Drill Results

10.3.A 2008 Largo Drill Program

In May 2008, Largo began a 5000 m drill campaign to test high priority IP deposits for PGMmineralization. Boart Longyear (Geoserv Pesquisas Geológicas S/A) began the program with one drill rigon May 28, 2008 and continued until September 19, 2008, at which time the drill program was terminateddue to the capital market collapse. It was decided that it was more prudent to discontinue drilling andsave the resources. At the time the program was stopped, Largo had completed 16 holes totalling 3,843m. The program is summarized in Table 10-5 below:

Table 10-5 2008 Drill Program Summary

Ten of the sixteen holes were analyzed in 2008. The remaining six holes were analyzed in 2011, duringthe 2011-2012 drill program. Boart Longyear drilled with NQ-sized rods and averaged 1000 m per month,per rig. Core recovery was good with a reported average of 90%. Detailed drillhole information is listedin Table 10-6 below:

V2O5Pd Pt

(%) (g) (g)

FGA100 83.00 85.00 2.00 2.00 1.62 0.14 0.13 Gulçari A

and 93.00 95.00 2.00 2.00 1.24 0.10 0.10and 119.00 122.00 3.00 3.00 1.45 0.05 0.10

Hole

Number From To

Interval

(m)

True

thickness

(m) Zones

Areas TypeNo of

Holes

Total

metres

Gulçari A Exploration 1 211.00

Gulçari A Norte Exploration 5 1,137.20

São Jose Exploration 9 2,209.50

Novo Amparo Exploration 1 285.00

Total 16 3,842.70





Table 10-6 2008 Drill Program Information

The focus of the diamond drill program was on testing a number of high priority PGM deposits on theproperty. These holes were targeted based on the magnetic and IP geophysical surveys completed in2007, lithogeochemical results from drill core sampling done across the property and some modeling,lithological and petrographic studies done by Dr. Keays.

Dr. Keays considers that there is a strong possibility for a PGM-rich horizon to occur at a stratigraphicinterval higher than that in which the Gulçari A deposit is located. The field work complete to dateincluding geological mapping and sampling, magnetic and Spectral IP ground surveys has identified anumber of high priority deposits to be tested, in particular by the sulphide content and PGM potential ofthe mineralization.

Significant assay results for all the 2008 drilling are reported in Table 10-7 below.

Drill

Hole # N E

FGA 101 8,486,386 318,630 300 270 -45 211.00

FGAN 01 8,487,100 318,775 307 270 -45 259.00FGAN 02 8,487,000 318,785 310 270 -45 260.80

FGAN 03 8,487,000 318,925 312 270 -45 225.40

FGAN 04 8,486,800 318,700 308 270 -45 196.00

FGAN 05 8,486,800 318,875 330 270 -45 196.00

FSJ 03 8,488,300 319,200 325 290 -50 244.30

FSJ 04 8,488,200 319,200 330 290 -50 250.00

FSJ 05 8,488,350 319,250 323 290 -50 232.60

FSJ 06 8,488,100 319,225 332 290 -45 229.30

FSJ 07 8,488,000 319,243 330 290 -45 280.00

FSJ 08 8,487,303 318,806 321 290 -45 253.00FSJ 09 8,488,398 319,104 320 290 -50 178.00

FSJ 10 8,487,800 319,200 332 290 -50 262.30

FSJ 11 8,487,298 318,863 320 270 -50 280.00

FNA 19 8,489,599 319,679 307 270 -45 285.00

TOTAL 16 3,842.70

UTM Coordinates

(Corrego Alegre) Elevation

(metres)

Azimuth

(°)

Inclination

(°)

Depth

(metres)





Table 10-7 2008 Drill Program Summary of Significant Results

10.4.A 2011-2012 Largo Drill Program

Between May 16, 2011 and February 16, 2012, Largo completed a drill campaign consisting of 72 holestotaling 13 401 m as set out in Table 10-8 below.

Layne Christensen (Layne do Brasil Sondagens Ltda) began the program with one drill rig on May 16,2011 and added a second drill rig on June 1, 2011. Two rigs continued on the property until December20, 2011 at which time one was released and the second drill went to Gulçari A Norte where 8 holestotaling 1006.55 m were completed. Drilling was completed on February 5, 2012. Layne Christensendrilled with NQ-sized rods and averaged 900 m per month, per rig. Core recovery was good with areported average of about 90%. Detailed drillhole information for the seven zones tested is set out inTable 10-9 to Table 10-15 below.

From To V2O5 Pt Pd PGM

Hole Number (m) (m) (%) (g/t) (g/t) (g/t)

FSJ 03 75.00 90.00 0.55 15.00 15.00

FSJ 04 22.00 35.00 0.56 13.00 13.00 and 128.00 138.00 0.50 10.00 10.00

FSJ 05 2.00 21.00 0.55 0.32 0.12 0.44 19.00 19.00

FSJ 06 68.00 88.00 0.52 0.34 0.14 0.48 20.00 20.00

including 83.00 88.00 0.63 0.37 0.25 0.63 5.00 5.00

FSJ 07 20.00 30.00 0.74 0.40 0.12 0.52 10.00 10.00

and 113.00 129.00 0.65 0.40 0.22 0.62 16.00 16.00

FSJ 08 No significant results

FGA101 93.00 98.27 1.00 0.10 0.10 0.20 5.27 5.00

and 156.00 163.00 1.08 0.32 0.31 0.63 7.00 7.00

including 157.22 160.00 1.17 0.73 0.69 1.42 2.78 2.78

FNA 19 No significant resultsFGAN01 211.80 220.00 1.00 0.1 0.1 0.2 8.20 8.00

including 216.00 220.00 1.16 0.1 0.1 0.2 4.00 4.00

FGAN02 192.00 195.45 1.00 0.1 0.1 0.2 3.45 3.00

FGAN03 No significant results

FGAN04 105.00 107.82 0.95 0.1 0.1 0.2 2.82 2.82

and 153.78 162.00 1.02 0.22 0.25 0.47 8.22 8.22

including 157.00 160.63 1.21 0.39 0.44 0.83 3.63 3.50

FGAN05 48.94 62.70 0.75 0.2 0.1 0.3 13.76 13.50

True

Thickness

(m)

Interval

(m)





Table 10-8 Largo 2011 - 2012 Drill Program

Table 10-9 Gulçari A Zone Drilling

Areas Type

No. of

Holes Total Metres

Gulçari A Exploration 11 3,117.61

Gulçari A Norte Exploration 12 1,766.73

Gulçari B Exploration 10 1,367.81Gulçari B Sul Exploration 6 1,150.00

São Jose Exploration 14 2,389.75

Novo Amparo Exploration 2 357.95

Novo Amparo Norte Exploration 17 3,251.50

Total 72 13,401.35

Elevation Azimuth Dip Depth

(m) (º) (º) (m)FGA 102 8,486,018 318,696 304 290 -70° 506.00

FGA 103 8,486,086 318,554 301 290 -70° 364.05

FGA 104 8,486,081 318,058 303 290 -45° 196.85

FGA 105 8,486,029 318,554 310 290 -70° 404.90

FGA 106 8,486,120 318,082 304 290 -60° 195.04

FGA 107 8,486,169 318,089 306 290 -60° 154.07

FGA 108 8,486,074 318,602 303 290 -75° 446.60

FGA 109 8,486,103 318,119 338 290 -60° 194.00

FGA 110 8,486,125 318,194 339 290 -60º 104.00

FGA 111 8,485,902 318,422 295 290 -60º 300.50FGA 112 8,486,176 318,390 301 290 -50º 251.60

Hole ID Northing Easting





Table 10-10 Gulçari A Zone Norte Drilling

Table 10-11 Gulçari B Zone Drilling


(m) (°) (º) (m)

FGAN 06 8,486,501 318,658 305 290 -45° 189.30

FGAN 07 8,486,644 318,695 307 290 -45° 198.10FGAN 08 8,486,896 318,728 307 290 -45° 216.65

FGAN 09 8,486,382 318,562 301 290 -45° 130.75

FGAN 10 8,486,458 318,581 305 290 -45° 130.45

FGAN 11 8,486,520 318,612 312 290 -45° 122.48

FGAN 12 8,486,605 318,620 304 290 -45° 131.40

FGAN 13 8,486,676 318,620 310 290 -45° 117.60

FGAN 14 8,486,814 318,629 312 270 -45° 118.30

FGAN 15 8,486,926 318,652 319 290 -45° 121.85

FGAN 16 8,487,009 318,710 312 290 -45° 128.50

FGAN 17 8,487,116 318,711 316 290 -45° 161.35



(m) (°) (º) (m)

FGB 08 8,487,166 318,945 319 290 -65° 99.15

FGB 09 8,487,127 318,982 320 290 -65° 102.62

FGB 10 8,487,091 318,928 316 290 -65° 99.33

FGB 11 8,487,068 318,984 320 290 -65° 171.06

FGB 12 8,487,090 318,991 322 290 -70° 136.90FGB 13 8,487,044 318,976 324 290 -70° 173.40

FGB 14 8,487,013 318,957 319 290 -70° 164.70

FGB 15 8,486,992 318,902 319 290 -60° 103.80

FGB 16 8,486,978 318,937 322 290 -70° 155.70

FGB 17 8,486,937 318,933 325 290 -70° 161.15






Table 10-12 Gulçari B Sul Zone Drilling

Table 10-13 São José Zone Drilling

Table 10-14 Novo Amparo Zone Drilling


(m) (°) (º) (m)

FGBS 01 8,486,749 318,992 317 290 -45° 212.15

FGBS 02 8,486,840 319,041 325 290 -45° 167.25FGBS 03 8,486,606 318,996 309 290 -45° 174.80

FGBS 04 8,486,496 318,988 312 290 -45° 219.75

FGBS 05 8,486,393 318,983 314 290 -45° 169.70

FGBS 06 8,486,293 318,940 325 290 -45° 206.35



(m) (°) (º) (m)

FSJ 12 8,488,280 318,907 319 290 -45° 368.80FSJ 13 8,488,230 318,884 320 290 -45° 187.70

FSJ 14 8,488,230 318,884 320 290 -70° 169.15

FSJ 15 8,488,364 318,940 325 290 -65° 152.30

FSJ 16 8,488,171 318,863 321 290 -45° 161.20

FSJ 17 8,488,564 319,041 323 290 -45° 124.60

FSJ 18 8,488,656 319,059 338 290 -45° 156.40

FSJ 19 8,488,492 318,989 331 290 -45° 142.40

FSJ 20 8,488,425 318,957 327 290 -45° 132.20

FSJ 21 8,488,344 318,997 322 290 -45° 180.65

FSJ 22 8,488,346 318,899 319 290 -45° 153.85FSJ 23 8,488,313 318,811 320 290 -45° 150.70

FSJ 24 8,488,260 318,812 321 290 -45° 153.80

FSJ 25 8,488,196 318,800 319 290 -45° 156.00



(m) (°) (º) (m)

FNA 20 8,489,491 319,632 349 290 -45° 137.70

FNA 21 8,489,690 319,623 343 290 -45° 220.25






Table 10-15 Novo Amparo Norte Zone Drilling

The focus of the diamond drill program was to further delineate additional resources on the Maracásproperty. The area encompassed by the drilling includes a 6.5 km strike length from, south to north,Gulçari A North to Novo Amparo Norte and a 1.5 km strike length on the east side from São José toGulçari B Sul (See Figure 10-2). The results from the 2011-2012 program are summarized in Table

10-16 below.

Table 10-16 2011-2012 Drill Program Summary of Significant Results


(m) (°) (º) (m)

FNAN 01 8,492,565 319,989 354 290 -45° 213.65

FNAN 02 8,492,479 319,953 362 290 -45° 169.95FNAN 03 8,492,414 319,923 354 290 -45° 167.00

FNAN 04 8,491,855 319,967 341 290 -45° 239.40

FNAN 05 8,491,780 319,895 335 290 -45° 150.50

FNAN 06 8,492,657 320,053 351 290 -45° 184.15

FNAN 07 8,491,679 319,865 343 290 -45° 240.75

FNAN 08 8,492,732 320,115 355 290 -45° 186.00

FNAN 09 8,492,303 319,981 339 290 -45° 170.80

FNAN 10 8,492,400 319,978 339 290 -45° 175.80

FNAN 11 8,492,885 320,117 352 290 -45° 223.10

FNAN 12 8,492,756 320,048 354 290 -45° 196.95FNAN 13 8,492,458 320,015 355 290 -45° 187.50

FNAN 14 8,492,634 320,120 354 290 -45° 231.50

FNAN 15 8,492,536 320,051 365 290 -60° 286.90

FNAN 16 8,492,361 319,900 350 290 -45° 113.20

FNAN 17 8,492,448 319,934 351 290 -45° 114.35


PD

(g/t)

PT

(g/t)

FGA101 93.00 98.27 5.27 5.00 1.00 0.10 0.10 Gulçari A Norte

and 156.00 163.00 7.00 7.00 1.08 0.31 0.32

including 157.22 160.00 2.78 2.78 1.17 0.69 0.73

FGA102 385.95 399.40 13.45 10.00 1.05 0.16 0.16 Gulçari A

FGA103 254.68 322.80 78.14 62.00 1.64 0.15 0.10 Gulçari A

including 260.00 290.00 30.00 25.00 2.09 0.16 0.11FGA104 31.00 35.90 4.90 4.90 0.70 0.21 0.22 Gulçari A

ZonesHole Number From To

Interval

(m)

True

Thickness

(m)

V2O5

(%)






PD

(g/t)

PT

(g/t)

FGA105 99.00 103.39 4.39 4.30 1.00 - - Gulçari A

and 274.60 293.57 18.97 15.00 1.34 0.15 0.12

and 308.18 327.18 19.00 15.10 1.00 0.12 0.14

FGA106 37.50 52.44 14.94 14.00 1.03 0.31 0.18 Gulçari A

including 47.34 51.62 4.28 4.00 1.32 0.61 0.41

FGA107 29.00 34.20 5.20 5.00 1.00 0.10 0.10 Gulçari A

FGA108 154.00 159.40 5.40 5.00 1.00 Gulçari A

and 287.00 318.00 41.00 35.00 1.68 0.10 0.14

including 294.00 305.00 11.00 8.00 1.87 0.04 0.09

including 307.00 317.00 10.00 7.00 2.09 0.24 0.27

FGA109 - 37.00 37.00 37.00 1.00 0.06 0.09 Gulçari A

including - 3.00 3.00 3.00 1.87 0.10 0.19and 93.00 99.00 6.00 5.00 1.01 0.03 0.03

and 136.00 141.00 5.00 4.50 1.28 0.33 0.46

FGA110 14.00 35.00 21.00 15.00 1.00 0.12 0.14 Gulçari A

FGA111 240.00 243.00 3.00 3.00 1.00 0.31 0.18 Gulçari A

FGA 112 47.00 61.65 14.65 0.90 0.01 0.33 Gulçari A

and 91.20 94.00 2.80 1.82 0.13 0.52

and 96.60 107.00 10.40 1.19 0.22 0.30

FGAN06 169.00 176.00 7.00 6.00 1.09 Gulçari A North

including 172.00 175.00 3.00 3.00 1.18 0.46 0.46

FGAN07 177.00 184.00 7.00 6.00 1.05 0.41 0.44 Gulçari A North

FGAN08 138.00 143.00 5.00 5.00 0.88 Gulçari A Northand 204.00 207.93 3.93 3.50 0.95



and 86.00 89.00 3.00 3.00 0.52 0.41 0.28


and 48.00 52.00 4.00 4.00 1.00

and 110.00 112.00 2.00 2.00 0.56 0.56 0.34




and 90.00 95.10 5.10 5.00 1.00 0.23 0.42


and 104.75 116.25 12.50 12.00 1.03 0.14 0.26



and 104.25 106.40 2.15 2.00 1.00

and 137.00 146.00 9.00 8.00 1.10 0.09 0.09

Hole Number From To

Interval

(m)

True

Thickness

(m)

V2O5

(%) Zones






PD

(g/t)

PT

(g/t)

FGB08 Gulçari B

FGB09 Gulçari BFGB10 13.00 29.00 16.00 16.00 0.83 0.08 0.20 Gulçari B

FGB11 92.90 101.30 8.40 8.40 0.89 0.15 0.32 Gulçari B

FGB12 Gulçari B

FGB13 98.00 108.60 10.60 10.00 0.80 0.09 0.22 Gulçari B

FGB14 88.80 99.40 10.60 10.00 0.76 0.09 0.20 Gulçari B

FGB15 15.00 37.00 22.00 22.00 0.66 0.05 0.11 Gulçari B

FGB16 72.00 100.00 28.00 26.00 0.62 0.04 0.08 Gulçari B

FGB17 85.00 105.00 20.00 20.00 0.67 0.06 0.14 Gulçari B

FGBS01 9.00 17.00 8.00 8.00 0.50 Gulçari B Sul



FGBS04 46.00 58.00 12.00 12.00 0.43 0.06 0.18 Gulçari B Sul

FGBS05 54.00 70.00 16.00 16.00 0.42 0.09 0.20 Gulçari B Sul


and 55.00 65.00 10.00 10.00 0.36 0.06 0.20

FSJ12 50.00 54.00 4.00 4.00 0.88 São Jose West

and 80.00 83.00 3.00 3.00 0.80 São Jose West

FSJ13 71.00 75.00 4.00 4.00 0.86 São Jose West

and 91.00 93.00 2.00 2.00 0.96 São Jose West

FSJ14 96.00 100.20 4.20 4.20 0.90 São Jose West

and 117.00 118.75 1.75 1.75 1.20 São Jose West

FSJ15 35.00 46.00 11.00 11.00 0.98 São Jose West

including 40.00 45.00 5.00 5.00 1.16 São Jose WestFSJ16 90.00 94.00 4.00 4.00 0.90 São Jose West

FSJ17 57.00 66.00 9.00 9.00 1.08 São Jose West

including 60.00 66.00 6.00 6.00 1.25

and 83.47 91.60 8.13 8.00 0.77

FSJ18 44.00 50.00 6.00 6.00 1.00 São Jose West

FSJ19 62.00 67.00 5.00 5.00 0.78 São Jose West

FSJ20 21.00 29.00 8.00 8.00 0.93 São Jose West

and 53.00 57.40 4.40 4.40 0.82

FSJ21 77.00 89.90 12.90 12.50 1.05 São Jose West

including 83.00 89.90 6.90 6.50 1.23

and 112.00 115.20 3.20 3.00 0.71

FSJ22 12.00 23.00 11.00 11.00 0.67 São Jose West

FSJ23 7.70 9.00 1.30 1.30 1.14 São Jose West

and 26.65 29.85 3.20 3.20 0.73

V2O5

(%) ZonesHole Number From To

Interval

(m)

True

Thickness

(m)

No significant results

No signifi cant results

No signifi cant results






PD

(g/t)

PT

(g/t)

FSJ24 9.00 12.00 3.00 3.00 1.00 São Jose West

FSJ25 31.00 34.00 3.00 3.00 0.94 São Jose WestFNA21 45.15 52.00 6.85 6.85 0.80 Novo Amparo

FNAN01 79.00 100.00 21.00 20.00 1.00 Novo Amparo Norte

including 83.00 99.00 16.00 15.00 1.11

and 136.00 141.90 5.90 5.00 0.82 0.10 0.10


and 72.00 78.15 6.15 6.00 0.90

and 91.00 93.00 2.00 2.00 1.10 1.46 0.53


including 55.52 66.00 10.48 10.00 1.08

and 81.00 90.00 9.00 8.50 0.88 0.29 0.16

FNAN04 Novo Amparo Norte

FNAN05 Novo Amparo NorteFNAN06 109.00 122.00 13.00 13.00 0.98 Novo Amparo Norte

including 115.00 122.00 7.00 7.00 1.25

and 152.00 155.00 3.00 3.00 0.89

and 161.00 163.10 2.10 2.00 1.09 0.76 0.33

FNAN07 Novo Amparo Norte


including 164.00 170.00 6.00 5.00 1.10 Novo Amparo Norte



including 112.45 134.45 22.00 21.00 1.18

and 144.45 152.00 7.55 7.00 0.83

including 149.45 152.00 2.55 2.00 0.93 0.80 0.48


including 197.00 200.00 3.00 2.50 0.93 1.00 0.42


and 123.00 135.00 12.00 11.00 0.90

including 133.00 135.00 2.00 2.00 1.03 1.38 0.69


including 143.00 149.00 6.00 5.00 1.24



and 260.65 273.40 12.75 11.50 0.84

including 268.00 273.40 5.40 5.00 0.94 0.81 0.47FNAN16 78.50 94.70 15.50 15.00 0.90 Novo Amparo Norte

including 86.00 94.70 8.70 8.50 1.09

including 88.00 94.70 6.70 6.50 1.15 1.56 0.76

FNAN17 86.00 88.00 2.00 2.00 1.17 1.46 0.72 Novo Amparo Norte

Hole Number From To

Interval

(m)

True

Thickness

(m)

V2O5

(%) Zones








Figure 10-2 Zone Location Map

The total drilling completed on the property has tested 7 zones with 209 holes totalling 35 286.59 m (seeTable 10-17) of which Largo has drilled 140 holes totalling 29 371.29 m between 2007 and 2012.





Table 10-17 Total Maracás Drilling

There has been sufficient drilling in this area to demonstrate the continuity of the magnetite-rich horizonswhich is also supported by the ground magnetic survey that traces the known zones on surface. Theground magnetic survey also has identified a number of deposits that had not been previously tested.

The Gulçari A deposit, as outlined from the drill programs, now extends 400 m along strike, and to avertical depth of over 350 m with true widths ranging from 11 to 100 m and with an average width of about

Area Type No of Holes Total Metres

Gulçari A 1981-87 53 5 152.57

2007 45 11 195.94

2011-12 11 3 117.61Total 109 19 466.12

Gulçari B 1981-83 7 270.28

2011-12 10 1 367.81

Total 17 1 638.09

Gulçari A Norte 2007 3 566.40

2008 1 211.00

2011-12 12 1 766.73

Total 16 2 544.13

Gulçari B Sul 2011-12 6 1 150.00

Total 6 1 150.00

São Jose 1983 2 115.15

2008 9 2 209.50

2011-12 14 2 389.75

Total 25 4 714.40

Novo Amparo 1983 7 377.30

2007 9 1 502.10

2008 1 285.00

2011-12 2 357.95

Total 19 2 522.35

Novo Amparo Nort 2011-12 17 3 251.50

Total 17 3 251.50

Grand Total 209 35 286.59





40 m. This deposit is part of a mineralizing system that extends the length of the property. All the assaysfrom this drill program are completed and results received.

Results from the 2008 and 2011-2012 drilling were not fully available for use in the technical updatepresented in this report. The 2008 to 2012 results were later used to update the mineral resource,presented in Section 14 so as to make the report current.

10.5.A Logging

Largo had rented a farmhouse immediately adjacent to the Maracás property and about 2 km fromGulçari A. This house was used as an office, bunkhouse for the geologists and a core logging andstorage facility. Covered and shaded logging racks have been built for the geologists to lay out andexamine core.

The core boxes had nailed-on lids and were delivered to this location daily, where they were sorted byhole and stacked. Later, the lids were removed and they were placed on the logging racks, where boxmarkings and footage blocks are checked for accuracy.

Holes are logged in a conventional manner with lithologies and mineralization marked up with a lumbercrayon and described, as well as the recording of basic geotechnical observations (rock qualitydesignation, RQD). Particular attention was placed on the degree of magnetism in the core. Loggingwas performed using a computer and the “Logger” front-end data collector program written for Gemcom®.

At the time of Micon’s site visit in April 2007, drilling had just commenced and core had not yet beenphotographed. Micon was informed that this was to be done, and a digital camera had just beenpurchased. At the time of Micon’s 2011 visit, core was being photographed. Other than this change thecore logging procedures used in later drill programs have remained the same.

10.1.B Drilling By Previous Operators

The drilling completed by previous operators, CBPM and Odebrecht, has been described earlier in thisreport in Section 6, History, above.

During 2007 Largo completed a drill campaign consisting of 61 holes totalling 13,876 m.

Boart Longyear (Geoserv Pesquisas Geológicas S/A) began the program with one drill rig on February15, 2007 and added a second drill rig on March 5, 2007. Two rigs continued on the property until August19, 2007 at which time drilling was completed on Gulçari A deposit. One rig was released and the seconddrill went to Novo Amparo where 11 holes totalling 1,852 m were completed. The drill then scored fiveregional holes testing geophysical deposits totalling 828 m. Drilling was completed on October 29, 2007.

Boart Longyear drilled with NQ-sized core and averaged 1,000 m per rig per month. Core recovery wasgood with a reported average of 90%. Detailed drillhole information is listed in Table 10-18.





Table 10-18 Drillhole Information

The principal focus of the diamond drill program was to upgrade the confidence of the existing inferredresource at Gulçari A to the measured and indicated categories and to expand upon it sufficiently todemonstrate its potential economic viability. A secondary objective was to evaluate the PGM potential in

the Gulçari A deposit. Table 10-19 shows the drillhole summary for Gulçari A program.

Table 10-19 Drillhole Information GA

Areas Type/Purpose Number of Holes Total (m)

Gulçari A Resource 42 10,896

Gulçari A Metallurgical & Geotech 3 300

São José Exploration 9 2,324

Novo Amparo Exploration 11 1,852

Regional Exploration 5 828

Total - 70 16,200

N E

FGA 56 8,486,090 318,346 294.04 290 -55 229.10

FGA 57 8,486,090 318,346 295.04 290 -70 265.10

FGA 58 8,486,125 318,353 294.96 290 -70 238.40

FGA 59 8,486,151 318,358 294.76 290 -45 138.60

FGA 60 8,486,032 318,360 297.52 290 -50 254.00

FGA 61 8,486,162 318,364 296.29 290 -60 139.60 FGA 62 8,486,074 318,332 296.09 290 -50 214.00

FGA 63 8,486,038 318,334 293.09 290 -50 229.30

Drillhole

UTM Coordinates

(Corrego Alegre)Elevation

(m)

Azimuth

(°)

Dip

(°)

Depth

(m)





Table 10-19 Drill Information GA

N E

FGA 64 8,486,074 318,332 295.09 290 -65 267.00

FGA 65 8,486,020 318,402 299.25 290 -50 197.50 FGA 66 8,486,073 318,409 299.18 290 -65 247.15

FGA 67 8,486,090 318,296 301.09 290 -45 206.60

FGA 68 8,486,111 318,405 297.73 290 -65 270.60

FGA 69 8,486,133 318,319 302.71 290 -45 151.00

FGA 70 8,486,053 318,213 302.74 290 -45 94.70

FGA 71 8,486,097 318,454 300.84 290 -65 236.20

FGA 72 8,486,020 318,336 296.00 290 -45 285.60

FGA 73 8,486,130 318,435 299.00 290 -50 224.55

FGA 74 8,486,130 318,435 299.00 290 -80 292.80

FGA 75 8,486,033 318,288 293.99 290 -45 267.60

FGA 76 8,486,011 318,281 293.68 290 -55 151.10 FGA 77 8,486,145 318,429 297.82 290 -60 159.70

FGA 78 8,486,000 318,320 295.64 290 -65 190.00

FGA 79 8,486,133 318,478 300.03 290 -60 229.00

FGA 80 8,485,941 318,301 294.70 290 -60 222.95

FGA 81* 8,486,101 318,250 302.40 290 -50 205.00

FGA 82 8,486,201 318,926 298.99 290 -60 169.40

FGA 83 8,486,065 318,380 298.05 290 -65 188.60

FGA 84* 8,486,110 318,226 302.40 290 -50 50.50

FGA 85* 8,486,120 318,205 313.30 290 -50 44.50

FGA 86 8,486,048 318,432 302.00 290 -65 214.10

FGA 87 8,486,079 318,523 305.00 290 -65 352.00

FGA 88 8,486,055 318,479 309.00 290 -65 336.64

FGA 89 8,486,051 318,620 304.00 290 -65 436.00

FGA 90 8,485,993 318,498 309.00 290 -65 404.25

FGA 91 8,486,106 318,594 300.00 290 -60 382.20

FGA 92 8,485,991 318,430 301.00 290 -60 351.80

FGA 93 8,486,007 318,530 304.00 290 -65 358.10

FGA 94 8,485,968 318,412 301.00 290 -65 336.00

FGA 95 8,486,068 318,567 301.41 290 -65 413.95

FGA 96 8,486,129 318,531 304.10 290 -60 369.65

FGA 97 8,486,130 318,435 298.00 290 -65 280.30

FGA 98 8,486,020 318,402 296.00 290 -73 325.00

FGA 99 8,485,955 318,465 302.00 290 -65 394.30

FGA 100 8,485,962 318,348 296.00 290 -65 181.50

Total 45 11,196.00

Drillhole

UTM Coordinates

(Corrego Alegre)Elevation

(m)

Azimuth

(°)

Dip

(°)

Depth

(m)





The drilling at Novo Amparo and São José was designed to test and characterize the mineralization 2 Kmand 4 km to the north along strike of Gulçari A and, in particular, the sulphide content and PGM potentialof the mineralization.

Finally from the ground magnetic survey completed, five deposits were selected that had not beenpreviously tested. These showings occur along magnetic trends that can be trace across the property for

4 km from Novo Amparo in the north to Gulçari A in the south.

The Gulçari A deposit as outlined from drill program extends 400 m along strike, to a vertical depth ofover 320 m with true widths ranging from 11 to 100 m with an average width of about 40 m. This depositis part of a mineralizing system that extends for 8 km across the property. All the results from the drillprogram are completed and incorporated in the block model. The results from the Gulçari A drill programare summarized below. Results for Novo Amparo do not impact on this report as they are not part of thisupdated mineral resource. Of the 45 holes drilled at the Gulçari A deposit, 39 intersected wide wellmineralized zones. Table 10-20 is a summary of all significant assay results from the drilling at Gulçari A.

Table 10-20 Assay Result GA Deposit

FGA56 82.00 119.00 1.68 0.28 0.21 0.49 37.00 32.00

including 102.00 118.00 2.17 0.40 0.29 0.69 16.00 15.00

FGA57 104.00 140.10 1.85 0.27 0.21 0.48 36.10 30.00

including 121.00 138.10 2.31 0.36 0.28 0.66 17.10 17.00

FGA58 61.30 120.00 2.22 0.45 0.11 0.56 58.70 54.00

including 79.00 103.00 2.55 0.44 0.11 0.55 24.00 21.00

and 136.00 158.00 2.12 0.41 0.20 0.61 22.00 22.00

including 136.00 141.00 1.79 1.07 0.33 1.40 5.00 5.00

FGA59 50.00 122.00 1.88 0.52 0.08 0.60 72.00 72.00including 65.00 73.00 2.65 0.91 0.10 1.01 8.00 8.00

FGA60 75.00 91.00 0.82 0.12 0.13 0.25 16.00 16.00

FGA61 76.00 121.00 1.97 0.44 0.11 0.55 45.00 44.00

including 95.00 106.00 2.64 0.57 0.21 0.78 11.00 10.00

FGA62 62.00 104.00 1.77 0.24 0.22 0.46 42.00 42.00

including 83.00 103.00 2.13 0.36 0.34 0.70 20.00 20.00

and 160.00 191.00 1.13 0.08 0.11 0.19 31.00 31.00

FGA63 54.47 70.00 0.91 0.18 0.11 0.29 15.53 15.00

PGM(g/t)

Interval(m)

TrueThickness (m)

HoleNumber

From (m) To (m) V2O5 (%) Pt (g/t) Pd (g/t)





Table 10-20 Assay Result GA Deposit

10.2.B Maracás Drilling Program 2011/2012During 2011/2012 Largo completed a drill campaign consisting of 73 holes totalling 13,401 m distributedin 7 deposits as set out in Table 10-21.

This report is concerned solely with the mineral inferred resources estimate at Novo Amparo Norte, Novo Amparo, São José, Gulçari B and Gulçari A Norte. Layne do Brasil Sondagem Ltda began the programwith two drill rigs on May 15, 2011.These rigs continued on the property until February 5, 2012 at whichtime drilling was completed on all areas. Layne drilled with NQ-sized core and averaged 1,000 m per rigper month. Core recovery was good with a reported average of 90%. The drillhole testimonials are storedin the facility at the headquarters of Maracás (Figure 10-3 and Figure 10-4).

FGA64 72.00 116.00 1.41 0.24 0.21 0.45 44.00 42.00

including 96.87 114.34 2.03 0.37 0.33 0.70 17.47 16.00

and 238.05 262.40 1.27 0.44 0.28 0.72 24.35 23.00FGA65 129.08 155.52 1.09 0.18 0.08 0.26 26.44 25.00

FGA66 136.91 182.80 1.46 0.22 0.18 0.40 45.89 43.00

including 156.94 182.00 2.03 0.39 0.31 0.70 25.06 23.00

FGA67 30.00 76.00 2.10 0.44 0.17 0.61 46.00 46.00

including 32.00 55.00 2.47 0.42 0.14 0.56 23.00 23.00

including 62.00 73.00 2.21 0.54 0.26 0.80 11.00 11.00

and 137.00 154.00 1.42 0.13 0.10 0.23 17.00 17.00

FGA68 105.00 147.00 1.80 0.36 0.14 0.50 42.00 40.00

including 124.00 147.00 2.25 0.40 0.19 0.59 23.00 21.00

and 168.00 217.00 1.64 0.28 0.22 0.50 49.00 47.00FGA69 44.26 112.73 2.42 0.53 0.12 0.65 68.47 68.00

FGA70 23.00 36.00 2.15 0.21 0.09 0.30 13.00 13.00

FGA71 151.22 193.20 2.07 0.24 0.08 0.32 41.98 40.00

FGA72 211.55 224.60 1.30 0.12 0.08 0.20 13.05 12.00

and 229.90 239.15 1.43 0.13 0.11 0.24 9.25 9.00

and 249.15 268.15 1.06 0.04 0.05 0.09 19.00 18.00

FGA73 128.60 148.00 2.81 0.54 0.05 0.59 19.40 17.50

and 151.46 195.46 1.62 0.24 0.10 0.34 44.00 40.00

Interval

(m)

True

Thickness (m)

Hole

NumberFrom (m) To (m) V2O5 (%) Pt (g/t) Pd (g/t)

PGM

(g/t)





Table 10-21 Maracás Drill Program 2011/2012

Figure 10-3 Headquarters of Maracás

AREAS TYPE PURPOSE N° OF HOLES TOTAL METERS

GULÇARI A RESOURCE 11 3,117.61

GULÇARI A NORTE RESOURCE 12 1,766.73

GULÇARI B RESOURCE 10 1,367.81

GULÇARI B SUL EXPLORATION 6 1,150.00

SÃO JOSÉ RESOURCE 14 2,389.75

NOVO AMPARO RESOURCE 2 357.95

NOVO AMPARO NORTE RESOURCE 17 3,251.50

TOTAL 72 13,401.35





Figure 10-4 Project Drillhole Storage Facility

10.2.1.B Novo Amparo North Deposit

Table 10-22 shows the drillhole summary of Novo Amparo Norte deposit.

Table 10-22 Summary NAN Deposit

Hole ID Northing Easting Elevation(m)

Azimuth(°)

Dip(°)

Depth(m)

Year

FNAN01 8,492,565.00 319,989.00 354.00 290 -45 213.65 2011

FNAN02 8,492,479.00 319,953.00 362.00 290 -45 169.95 2011

FNAN03 8,492,414.00 319,923.00 354.00 290 -45 167.00 2011

FNAN04 8,491,855.00 319,967.00 341.00 290 -45 239.40 2011

FNAN05 8,491,780.00 319,895.00 335.00 290 -45 150.50 2011

FNAN06 8,492,657.00 320,053.00 351.00 290 -45 184.15 2011

FNAN07 8,491,679.00 319,865.00 343.00 290 -45 240.75 2011

FNAN08 8,492,732.00 320,115.00 355.00 290 -45 186.00 2011

FNAN09 8,492,303.00 319,981.00 339.00 290 -45 170.80 2011

FNAN10 8,492,400.00 319,978.00 339.00 290 -45 175.80 2011





Table 10-22 Summary NAN Deposit

The Novo Amparo North deposit as outlined from drill program extends 620 metres along strike withaverage width of 18 meters and average grade of 0.87% of V 2O5. Around of 17 holes drilled at thisdeposit, 39 intersected wide well mineralized zones.

Table 10-23 Assay Result NAN Deposit

Hole ID Northing EastingElevation

(m)

Azimuth

(°)

Dip

(°)

Depth

(m)Year

FNAN11 8,492,885.00 320,117.00 352.00 290 -45 223.10 2011

FNAN12 8,492,756.00 320,048.00 354.00 290 -45 196.95 2011

FNAN13 8,492,458.00 320,015.00 355.00 290 -45 187.50 2011FNAN14 8,492,634.00 320,120.00 354.00 290 -45 231.50 2011

FNAN15 8,492,536.00 320,051.00 365.00 290 -60 286.90 2011

FNAN16 8,492,361.00 319,900.00 350.00 290 -45 113.20 2012

FNAN17 8,492,448.00 319,934.00 351.00 290 -45 114.35 2012

3,251.50TOTAL

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FNAN 01 80.00 100.00 1.025 41.45 11.92 16.68 20.00 17.5

FNAN 01 136.00 141.90 0.822 25.52 7.10 31.11 5.90 5.2

FNAN 02 61.00 67.00 1.047 23.37 11.85 17.59 4.00 5.2

FNAN 02 72.00 78.15 0.898 15.00 8.22 26.21 6.15 5.4

FNAN 02 91.00 94.10 0.873 22.81 5.86 34.03 3.10 2.7

FNAN 03 55.52 73.70 0.924 40.89 12.19 17.03 18.18 17.9

FNAN 03 81.00 90.00 0.883 25.36 6.87 31.68 9.00 7.9 FNAN 06 108.00 122.00 0.949 42.97 13.30 13.89 14.00 12.2

FNAN 06 152.00 163.10 0.648 22.23 8.99 32.39 3.10 2.6

FNAN 08 159.00 173.35 0.831 38.39 11.60 19.41 1.35 12.5

FNAN 09 81.90 84.00 0.498 22.21 5.37 36.47 2.10 1.8

FNAN 09 102.00 106.80 0.836 25.99 7.19 30.10 4.80 4.2

FNAN 10 106.45 135.45 1.050 41.19 12.00 16.49 29.00 25.3

FNAN 10 143.45 152.00 0.786 24.09 6.54 32.16 8.55 7.5

FNAN 11 158.00 159.00 1.030 37.52 11.05 21.74 1.00 0.9

FNAN 11 188.00 200.00 0.757 24.43 7.06 32.90 12.00 10.5

FNAN 12 70.00 84.00 0.939 40.14 12.29 17.79 14.00 12.2





Table 10-23 Assay Result NAN Deposit

Figure 10-5 shows an example of the holes in the Novo Amparo North. Table 10-23 is a summary of allsignificant assay results from the drilling at Novo Amparo North.

Figure 10-5 Drillhole of NAN Deposit

10.2.2.B Novo Amparo Deposit

Table 10-24 below shows the drillhole summary of Novo Amparo Deposit.

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FNAN 12 121.00 135.56 0.820 25.64 7.22 30.74 14.56 11.0

FNAN 13 129.00 150.00 0.792 40.84 12.84 16.43 21.00 18.3 FNAN 14 190.00 200.40 0.857 41.32 12.77 15.73 10.40 9.1

FNAN 15 203.00 225.00 1.107 42.74 12.43 14.15 22.00 15.8

FNAN 15 258.00 273.40 0.780 24.05 6.87 32.91 13.95 11.1

FNAN 16 47.00 63.50 0.698 41.37 13.78 16.08 16.50 14.4

FNAN 16 78.50 89.70 0.821 26.07 7.25 29.39 11.20 9.8

FNAN 16 92.20 94.70 1.083 21.21 5.70 40.90 2.50 2.2

FNAN 17 57.00 73.00 0.865 34.71 8.63 23.77 16.00 14.9

FNAN 17 77.00 88.00 0.954 26.46 6.88 30.27 11.00 10.6





Table 10-24 Summary NA Deposit

The Novo Amparo deposit as outlined from drill program is 285 meters along strike and range from 11 to21 meters widths, with average grade of 0.72% of V2O5. Only drillhole FNA20 cross the mineralized zone.Table 10-25 below is a summary of all significant assay results from the drilling at Novo Amparo deposit.

Table 10-25 Assay Result NA Deposit

Figure 10-6 shows the mineralized zone of drillhole FNA15.


(m)

Azimuth

(°)

Dip

(°)

Depth

(m)Year

FNA 20 8,489,491.00 319,632.00 349.00 290 -45 137.70 2011

FNA 21 8,489,690.00 319,623.00 343.00 290 -45 220.25 2011

357.95TOTAL

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FNA 08 41.00 42.00 0.640 41.21 13.58 15.37 1.00 0.9

FNA 09 57.17 62.73 0.652 36.01 10.31 22.52 5.56 4.8 FNA 09 70.20 72.87 0.644 37.61 12.26 19.69 2.67 2.3

FNA 09 76.55 80.29 0.848 48.98 16.48 7.11 3.74 3.2

FNA 10 76.00 84.00 0.600 33.24 9.45 26.95 8.00 6.6

FNA 10 89.00 106.23 0.697 45.73 15.65 10.50 17.23 14.1

FNA 11 97.00 100.00 0.597 34.63 10.34 24.36 3.00 2.6

FNA 12 95.00 103.30 0.566 34.73 10.56 23.69 8.30 7.2

FNA 13 72.00 97.00 0.669 42.15 14.43 14.55 25.00 17.7

FNA 14 54.00 73.00 0.652 40.73 13.53 16.91 7.00 16.5

FNA 15 83.00 107.00 0.773 49.01 17.27 7.76 24.00 20.8

FNA 16 84.48 90.54 0.728 45.99 15.47 10.69 6.06 5.2

FNA 16 97.00 100.00 0.570 46.99 16.32 9.46 3.00 2.6 FNA 18 141.30 166.00 0.650 43.81 14.70 13.03 24.70 14.2

FNA 01 19.75 32.40 0.791 45.44 16.01 9.94 12.65 7.3

FNA 01 35.35 46.93 0.788 46.77 16.93 8.09 11.58 6.6

FNA 02 22.27 54.55 0.854 48.16 17.32 7.17 32.28 16.1

FNA 04 29.55 44.50 0.610 37.28 12.39 21.55 14.95 8.6

FNA 05 33.25 45.72 0.866 48.00 16.26 7.89 12.47 9.6

FNA 07 26.20 40.60 0.764 43.45 15.29 12.35 14.40 7.2

FNA 07 44.90 47.80 0.755 37.05 11.98 21.41 2.90 1.5





Figure 10-6 Mineralized zone - FNA15

10.2.3.B São José Deposit

Table 10-26 shows the drillhole summary of São José Deposit.

Table 10-26 Summary SJ Deposit


Azimuth(°)

Dip(°)

Depth(m)

Year

FSJ 12 8,488,280.00 318,907.00 319.00 290 -45 368.80 2011

FSJ 13 8,488,230.00 318,884.00 320.00 290 -45 187.70 2011

FSJ 14 8,488,230.00 318,884.00 320.00 290 -70 169.15 2011

FSJ 15 8,488,364.00 318,940.00 325.00 290 -65 152.30 2011

FSJ 16 8,488,171.00 318,863.00 321.00 290 -45 161.20 2011

FSJ 17 8,488,564.00 319,041.00 323.00 290 -45 124.60 2011

FSJ 18 8,488,656.00 319,059.00 338.00 290 -45 156.40 2011

FSJ 19 8,488,492.00 318,989.00 331.00 290 -45 142.40 2011

FSJ 20 8,488,425.00 318,957.00 327.00 290 -45 132.20 2011





Table 10-26 Summary SJ Deposit

The São José deposit is characterized by two blind mineralized bodies. The first is 524 meters long and11 meters thick. The second one is 404 meters in length and 5 meters thick. The global average grade is0.89% V2O5. Table 10-27 is a summary of all significant assay results from the drilling at São Josédeposit.

Figure 10-7 shows the location of drillhole FSJ15 in the São José Deposit.

Table 10-27 Assay Result SJ Deposit


(m)

Azimuth

(°)

Dip

(°)

Depth

(m)Year

FSJ 21 8,488,344.00 318,997.00 322.00 290 -45 180.65 2011

FSJ 22 8,488,346.00 318,899.00 319.00 290 -45 153.85 2011

FSJ 23 8,488,313.00 318,811.00 320.00 290 -45 150.70 2011FSJ 24 8,488,260.00 318,812.00 321.00 290 -45 153.80 2011

FSJ 25 8,488,196.00 318,800.00 319.00 290 -45 156.00 2011

2389.75TOTAL

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FSJ 09 125,10 126,60 1,500 na na na 1,50 1,4

FSJ 09 126,60 127,60 0,780 na na na 1,00 1,0

FSJ 09 142,00 148,80 0,743 na na na 6,80 6,5

FSJ 12 50,00 54,00 0,880 37,23 10,96 21,00 4,00 3,9

FSJ 12 59,00 60,00 1,110 40,70 11,10 17,10 1,00 1,0

FSJ 12 80,00 83,00 0,797 37,37 11,07 20,60 3,00 2,9FSJ 12 94,00 96,00 0,790 29,73 7,29 27,50 2,00 2,0

FSJ 13 71,00 75,00 0,860 40,33 12,35 15,60 4,00 3,9

FSJ 13 90,75 93,00 0,991 33,52 8,59 24,82 2,25 2,2

FSJ 14 96,00 100,20 0,904 41,27 12,11 17,13 4,20 3,4

FSJ 14 117,00 118,75 1,241 39,89 10,50 17,89 1,75 1,4

FSJ 15 35,00 46,00 0,934 40,72 12,20 17,15 11,00 10,8

FSJ 15 52,00 53,00 1,320 40,90 11,10 15,90 1,00 1,0

FSJ 15 71,00 72,00 0,900 23,27 6,10 32,97 1,00 1,0

FSJ 16 90,00 93,00 0,897 43,40 13,37 14,93 3,00 2,9

FSJ 16 109,00 110,00 0,860 30,80 7,62 28,30 1,00 1,0

FSJ 17 57,00 66,00 1,080 43,50 12,76 14,59 9,00 8,8





Figure 10-7 Location of Drillhole FSJ15

10.2.4.B Gulçari B Deposit

Table 10-28 shows the drillhole summary of Gulçari B Deposit.

Table 10-28 Summary GB Deposit

The Gulçari B deposit as outlined from drill program is 200 meters along strike and range from 4 to 13metres widths, with average grade of 0.70% of V2O5. Table 10-29 is a summary of all significant assayresults from the drilling at Gulçari B deposit. The Figure 10-8 shows the drillhole FGB14.


Azimuth(°)

Dip(°)

Depth(m)

Year

FGB 08 8,487,166.00 318,945.00 319.00 290 -65 99.15 2011

FGB 09 8,487,127.00 318,982.00 320.00 290 -65 102.62 2011

FGB 10 8,487,091.00 318,928.00 316.00 290 -65 99.33 2011

FGB 11 8,487,068.00 318,984.00 320.00 290 -65 171.06 2011

FGB 12 8,487,090.00 318,991.00 322.00 290 -70 136.90 2011

FGB 13 8,487,044.00 318,976.00 324.00 290 -70 173.40 2011

FGB 14 8,487,013.00 318,957.00 319.00 290 -70 164.70 2011

FGB 15 8,486,992.00 318,902.00 319.00 290 -60 103.80 2011

FGB 16 8,486,978.00 318,937.00 322.00 290 -70 155.70 2011

FGB 17 8,486,937.00 318,933.00 325.00 290 -70 161.15 2011

1,367.81TOTAL





Table 10-29 Assay Result GB Deposit

Figure 10-8 Location of Drillhole FGB14

10.2.5.B Gulçari A Norte Deposit

The Table 10-30 shows the drillhole summary of Gulçari A Norte Deposit.

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FGB 01 18,95 27,22 0,75 42,48 15,05 13,76 8,27 5,3

FGB 01 29,07 35,40 0,84 43,79 15,27 12,78 6,33 4,1FGB 02 18,50 29,05 0,77 44,46 15,41 13,24 10,55 6,8

FGB 06 10,60 36,44 0,75 47,30 17,07 8,98 25,84 8,8

FGB 10 13,00 29,00 0,88 44,93 14,99 12,37 16,00 11,3

FGB 11 87,22 90,35 0,72 41,90 13,89 15,35 3,13 2,2

FGB 11 95,00 101,30 0,84 42,56 13,60 15,35 6,30 4,5

FGB 12 91,30 92,25 0,66 31,90 10,30 25,10 0,95 0,6

FGB 13 98,00 108,60 0,80 45,31 15,08 11,88 8,90 6,8

FGB 14 88,80 99,40 0,76 43,65 14,78 13,49 10,60 6,8

FGB 15 17,00 39,25 0,68 43,44 15,04 13,38 22,25 14,3

FGB 16 73,00 100,00 0,60 41,55 14,43 16,58 21,00 17,3FGB 17 85,00 99,60 0,67 41,71 14,18 15,92 14,60 9,4

FGAN 05 48,94 62,70 0,60 34,88 10,75 23,95 13,76 12,0





Table 10-30 Summary GAN Deposit

Table 10-30 Summary GAN Deposit

The Gulçari A North deposit is characterized by two blind mineralized bodies. The main ore body is 1 Kmlong and 7.5 metres in average thickness. The second one is 350 metres long and 4 metres and width of4 metres. The global average grade is 0.84% V2O5.

Table 10-31 is a summary of all significant assay results from the drilling at Gulçari A north deposit.

Figure 10-9 shows the mineralized zone of Drillhole FGAN05.


(m)

Azimuth

(°)

Dip

(°)

Depth

(m)Year

FGAN 06 8,486,501 318,658 305 290 -45 189.3 2011

FGAN 07 8,486,644 318,695 307 290 -45 198.1 2011

FGAN 08 8,486,896 318,728 307 290 -45 216.65 2011FGAN 09 8,486,382 318,562 301 290 -45 130.75 2011

FGAN 10 8,486,458 318,581 305 290 -45 130.45 2011

FGAN 11 8,486,520 318,612 312 290 -45 122.48 2011

FGAN 12 8,486,605 318,620 304 290 -45 131.,40 2012

FGAN 13 8,486,676 318,620 310 290 -45 117.6 2012

FGAN 14 8,486,814 318,629 312 270 -45 118.3 2012

FGAN 15 8,486,926 318,652 319 290 -45 121.85 2012

FGAN 16 8,487,009 318,710 312 290 -45 128.5 2012


(m)

Azimuth

(°)

Dip

(°)

Depth

(m)Year

FGAN 17 8,487,116 318,711 316 290 -45 161.35 2012

1,766.73TOTAL





Table 10-31 Assay Result GAN Deposit

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FGAN 01 143,10 145,00 0,737 na na na 1,90 1,7

FGAN 01 148,67 150,00 0,860 na na na 1,33 1,2FGAN 01 181,00 186,00 0,476 na na na 5,00 4,4

FGAN 01 211,80 220,00 0,781 na na na 8,20 7,2

FGAN 02 192,00 195,45 0,957 na na na 3,45 3,0

FGAN 02 199,47 200,52 1,300 na na na 1,05 0,9

FGAN 04 105,00 107,82 0,943 41,97 12,30 17,06 2,82 2,5

FGAN 04 140,00 143,00 0,527 20,53 4,29 38,40 3,00 2,6

FGAN 04 153,78 160,63 1,118 29,19 7,35 27,72 6,85 6,0

FGAN 06 86,00 89,50 0,939 39,93 11,53 18,69 3,50 3,2

FGAN 06 91,53 92,70 1,300 41,10 10,70 16,90 1,17 1,1

FGAN 06 124,14 125,05 1,140 37,30 9,76 20,20 0,91 0,8FGAN 06 151,00 154,00 0,520 19,37 3,96 39,70 3,00 2,8

FGAN 06 169,00 176,00 1,084 29,44 7,32 27,19 5,00 6,4

FGAN 07 160,00 161,00 0,540 20,30 4,10 39,90 1,00 0,9

FGAN 07 177,00 184,00 0,984 26,69 6,60 30,74 3,00 6,4

FGAN 08 138,00 143,00 0,878 37,86 11,13 20,56 5,00 4,6

FGAN 08 145,00 147,00 0,760 27,20 6,62 31,90 2,00 1,8

FGAN 08 204,10 207,93 0,945 27,74 6,30 30,26 3,83 3,5

FGAN 09 28,00 30,10 0,905 39,20 11,62 19,62 2,10 1,9

FGAN 09 32,30 33,50 0,870 29,40 7,52 29,02 1,20 1,1

FGAN 09 60,00 61,00 0,920 32,41 7,79 25,75 1,00 0,9





Table 10-31 Assay Result GAN Deposit

Hole NumberFrom

(m)

To

(m)

V2O5

(%)

Fe

(%)

TiO2

(%)

SiO2

(%)

Interval

(m)

True

Thickness

(m)

FGAN 09 89,30 97,50 0,891 24,94 5,63 33,36 8,20 7,2

FGAN 10 26,00 28,15 0,945 40,15 11,96 18,06 2,15 2,0FGAN 10 30,00 31,00 0,688 24,91 6,03 34,30 1,00 0,9

FGAN 10 58,00 59,00 0,581 22,93 5,51 36,10 1,00 0,9

FGAN 10 75,00 76,00 0,484 28,45 6,56 31,60 1,00 0,9

FGAN 10 86,00 89,00 0,525 18,04 3,73 41,60 3,00 2,8

FGAN 11 34,00 38,00 1,037 41,67 12,35 16,68 4,00 3,7

FGAN 11 48,00 52,00 1,000 41,04 12,20 17,81 4,00 3,7

FGAN 11 82,00 84,00 0,631 24,53 5,63 35,00 2,00 1,8

FGAN 11 110,00 111,00 0,676 21,29 4,70 37,00 1,00 0,9

FGAN 12 29,80 31,35 1,070 38,60 11,57 16,67 1,55 1,4

FGAN 12 85,00 86,00 0,450 18,59 3,55 40,12 1,00 0,9

FGAN 12 103,00 104,00 0,930 26,97 6,59 28,83 1,00 0,9FGAN 13 74,50 75,50 0,520 33,80 7,70 24,90 1,00 0,9

FGAN 13 82,50 86,50 0,520 19,46 3,86 40,32 4,00 3,7

FGAN 14 31,00 34,30 1,033 41,67 12,40 16,70 3,30 3,0

FGAN 14 75,00 78,00 0,515 19,15 4,06 40,10 3,00 2,8

FGAN 14 90,00 95,10 0,954 27,00 6,33 31,19 5,10 4,7

FGAN 15 60,00 63,00 1,001 40,77 12,30 17,85 3,00 2,8

FGAN 15 65,00 66,40 0,979 35,60 8,53 25,20 1,40 1,3

FGAN 15 104,75 116,25 1,014 27,65 6,27 30,20 11,50 10,6

FGAN 16 109,00 112,00 1,081 43,61 13,05 15,03 3,00 2,8

FGAN 16 115,00 117,00 0,732 28,35 6,96 31,50 2,00 1,8

FGAN 17 75,00 79,20 0,860 38,09 12,51 18,84 4,20 3,9FGAN 17 81,15 82,00 1,185 36,33 9,62 21,80 0,85 0,8

FGAN 17 104,25 107,40 0,898 30,94 7,65 28,06 3,15 2,9

FGAN 17 139,00 146,00 1,087 31,64 7,70 26,17 7,00 6,4

FGA 101 93,00 95,90 0,857 na na na 2,90 2,6

FGA 101 97,63 99,36 0,626 na na na 1,73 1,5

FGA 101 128,00 129,00 0,620 na na na 1,00 0,9

FGA 101 157,22 162,00 1,201 na na na 4,78 4,2

FSJ 08 126,00 128,92 0,973 na na na 2,92 2,7

FSJ 08 133,00 133,75 0,680 na na na 0,75 0,7

FSJ 08 145,93 151,00 0,457 na na na 5,07 4,7

FSJ 08 178,34 190,00 0,816 na na na 11,66 10,7FSJ 11 184,00 187,96 0,870 na na na 3,96 3,5

FSJ 11 191,00 192,00 0,680 na na na 1,00 0,9

FSJ 11 208,00 209,00 1,020 na na na 1,00 0,9

FSJ 11 238,55 250,25 0,558 na na na 11,70 10,3





Figure 10-9 Zone Mineralized Drillhole FGAN05

Largo has rented a farmhouse immediately adjacent to the Maracás property and about 2 km fromGulçari A. This house is used as an office, bunkhouse for the geologists and a core logging and storage

facility. Covered and shaded logging racks have been built for the geologists to lay out and examine core.

The core boxes had nailed on lids and were delivered to this location daily where they were sorted byhole and stacked. Later the lids were removed and they were placed on the logging racks where boxmarkings and footage blocks are checked for accuracy.

Holes are logged in a conventional manner with lithologies and mineralization marked up with a lumbercrayon and described, as well as the recording of basic geotechnical observations (RQD). Particularattention was placed on the degree of magnetism in the core. Logging was performed using a computerand the “Logger” front end data collector program written by Gemcom.





11 Sample Preparation, Analysis and Security

In Item 11, Sample Preparation, Analysis and Security, both Micon and Coffey Mining have providedindividual sections. Although there is some duplication of information, it was felt by RPM that thecommentary and emphasis was sufficiently different to justify presenting both Micon and Coffey Miningdata.

The Micon items have been designated with the letter “A” and the Coffey Mining items have beendesignated with the letter “B”.

11.1.A Sampling Method and Approach

11.1.1.A Previous Operators

Sampling of mineralization within the Study area by previous operators has been conducted by bothdiamond-drilling and trench-sampling methods.

The actual sampling method and approach carried out by CBPM (1981 and 1983) is not known by Largo.However, during Largo’s visit to the core facility, it was observed that the drill core had been carefullysawn in half with all of the drillholes at Maracás available for review in the core shed. The remaining halfshowed the core to be very competent. It is believed by Largo, given the competent nature of the rock,together with clearly-marked sample intervals, that careful sampling procedures were carried out and thatthe sampling method had been conducted in a professional manner. Micon visited the core shed andspent several days reviewing core. Nothing viewed during the visit would cause Micon to come to adifferent conclusion. Sampled intervals were easy to identify and the core was in good condition.

Personal communication between R. A. Campbell and Marcos Nunes, then Project Geologist forOdebrecht, described the drill-core sampling procedures for 1984 through 1987, as set out below.

“Drill core was split using a diamond blade tile saw. The core pans were cleaned between each split

sample. The remaining half of the core sample was replaced back in the core box. Half the core wasthen bagged along with its corresponding sample tag and bagged for shipment. Commercial truckingshipped the core samples to GEOSOL (1983 to 1987) Laboratory, Paulo Abib Engenharia S.A. laboratory(1985 to 1987) both in Belo Horizonte. Core trays with the remaining half core sample were placed incore racks at the exploration office in Maracás and remain intact for future reference.”

“During the Odebrecht drill programs samples were crushed, ground completely to pulp passing -150mesh and then split at the GEOSOL and Paulo Abib Engenharia S.A. preparation facilities in BeloHorizonte, Brazil. The split pulps were then analyzed by XRF method also in GEOSOL and Paulo AbibEngenharia S.A. laboratory in Belo Horizonte.”

“The sample core length was 2.0 m in all cases in past sampling programs. The layered nature of thedeposit and thicknesses of the mineralized zone of from 4 to 100 m justified this interval. The potentialmining method of large tonnage open pit was also considered when selecting sample intervals. Smallerintervals would only be taken if there was a particular geological reason to do so.”





“Channel samples were also cut on 2.0 m intervals and usually no greater, due to the large amount of

material generated. Channel intervals were also governed by topography and geology so their lengthsvaried on occasion.”

11.1.2.A Largo Core Sampling

11.1.1.1 2006 and Early 2007 Re-logging

Until April 2007, Largo had drilled no core of its own. It had only resampled old drillholes from earlierprograms.

In collecting its samples, Largo split all core using a diamond-blade tile saw. Half of the core was placedin a numbered plastic sample bag with the sample tag. The remaining half-core was placed back in thecore box. A brick was briefly sawn between each sample cut, in order to clean the blade and prevent anycontamination between samples.

The half-core was then sealed in the bags along with its corresponding sample tag. The sample bagswere placed into larger “rice” bags, in groups of 15 samples, for shipment. The samples were transportedin a company-operated vehicle from the office in Maracás to Salvador, where they were handed over to acommercial transport company for delivery by truck to SGS Geosol Laboratórios Ltda. (SGS) in BeloHorizonte. The core trays with the remaining quarter-core were placed in core racks at the core storagefacility/office in Maracás, so as to be available for future reference.

The sampled core length was 2.0 m in all cases, in order to duplicate past sampling programs. Largoagreed that the layered nature of the deposit and thicknesses of the mineralized zone of from 2 to 100 m

justified this interval.

11.1.1.2 2007 Exploration Drill Program

Sample boundaries were marked up by the geologists during the logging process. Generally, core wasselected for sampling based on magnetite content or nearby strong alteration. Intervals to be sampledwere marked in red lumber crayon. The beginning of each sample interval was marked on the edge ofthe core box with a felt tip marker and with a sample tag, affixed to the box with a staple, at the end of theinterval. Overall about 45% of the core drilled was sampled.

Sampling commenced several metres prior to the beginning of mineralization and proceeded down-hole,usually at 1-m intervals, until a major lithologic contact. Sampling did not cross these contacts. Sampleintervals could be shortened or lengthened depending on these observations. Magnetite and magnetite-pyroxenite were generally sampled separately, if the magnetite bands were approximately 1 m in size orlarger.

Samples were collected by sawing the core in half with a diamond-blade tile saw at the logging facility.Once sawn, half-core was placed in a numbered plastic sample bag with the corresponding sample tag.The remaining half-core was placed back in the core box. A brick was briefly sawn between each samplecut, in order to clean the blade and prevent any contamination between samples. The sample bags wereplaced into larger plastic containers, in groups of 15 samples, for shipment. The samples weretransported in a company-operated vehicle from the office in Maracás to Jequié where they were handedover to a commercial transport company for delivery by truck to SGS in Belo Horizonte.





Logged and sampled core boxes were stored in a roofed, fenced-in enclosure with a concrete floor andknee wall, within the fenced yard of the farm house.

These procedures continued to be used in the 2008 and 2011-2012 drill programs.

11.2.A Sample Preparation, Analyses and Security

11.2.1.A Pre-2006 Analytical Work

Personal communication between R. A. Campbell and Mr. Marco Nunes also described the samplepreparation and analytical protocols in use prior to Largo’s involvement in the Project, but after 1983.

According to Nunes, CBPM and Odebrecht used GEOSOL and Paulo Abib Engenharia S.A. as theiranalytical laboratories during the 1981 to 1983 exploration programs and again between 1984 and 1987.Their procedures were summarized as follows.

“A total of 1,675 core samples were prepared at GEOSOL and Paulo Abib Engenharia S.A. laboratoriesin Belo Horizonte, Brazil. These core samples were packaged in batches of 40 samples which included

two replicates, one reference standard and one blank, inserted randomly. All samples underwentstandard crushing and pulverizing techniques. The entire drill sample was passed through a primarycrusher to yield a fine crushed product, with better than 75% of the sample passing 2 mm. When thecrushed sample yielded approximately 2 kg the entire sample was pulverized.”

“A crushed 2 kg sample was ground using a ring and puck mill pulverizer. The pulverizer uses a chrome

steel ring and puck set. All samples were pulverized to greater than 95% of the ground material passingthrough a -150 mesh screen. Grinding with chrome steel may impart trace amounts of iron and chromiuminto the sample.”

“Core samples were then analyzed at GEOSOL and Paulo Abib Engenharia S.A.’s laboratories in Belo

Horizonte, Brazil. All samples were analyzed for FeO, Fe2O3, SiO2, TiO2 and V2O5. Routinely a sample

weight of 0.66 grams was fused with 7.2 g of flux to prepare each bead. However there were variationsto this ratio for some matrices, giving a lower limit of detection of 0.01% and an upper limit of detection of5.0% for V2O5. Samples were fused into a glass disc using a Lithium Borate flux much as described fornormal fused glass beads. For “ore grade” materials flux composition and sample/flux ratios were varied

to ensure all of the sample dissolves and that recrystallization does not occur as the melt is cooled.”

The CBPM and Odebrecht sample pulps are still available to Largo. They have been placed in storage inthe office in Maracás. While not climate-controlled storage, the sample pulps are protected from directexposure to the elements and sunlight in a secure location.

11.2.2.A Largo Analytical Work

All sample preparation and primary analyses of drill core from the 2006/2007 resampling program and the2007, 2008 and 2011-2012 drill programs was performed by SGS in Belo Horizonte, Brazil and Lakefield,Ontario. The samples were analyzed for FeO, Fe2O3, SiO2, TiO2 and V2O5 by the XRF method and forplatinum and palladium by a 50 g fire assay technique. This was changed to a 20 g fire assay for the2007 and later drill programs, as a result of some initial problems with flux and the amount of magnetite insome of the samples. The XRF method gives a lower limit of detection of 0.01% V2O5.





SGS claims that the quality assurance system in place at its laboratories “complies with the requirements

of the international standards ISO 9001:2000 and ISO 14001:2004 to chemical analysis and geochem ofsoils, rocks and ores” (SGS Minerals, 2006).

Core samples were prepared at SGS in Belo Horizonte using the following protocol:

weigh, dry and reweigh sample; primary crush to -2 mm (70% passing); pulverize split fraction to -150 mesh in chrome steel ring mill pulverizer.

The fire assay procedure employed for platinum and palladium used a 50 g aliquot (later changed to 20 gas described above) with aqua regia digestion, followed by an atomic absorption (AA) spectroscopyfinish. Since this was a check sampling program, there were no field duplicates and field blanks insertedby Largo for this resampling program.

SGS prepared and analyzed its own laboratory duplicates and inserted its own internal referencestandards and blanks. A complete set of the SGS quality control data files was obtained and verified bythe Largo staff member responsible. There were no abnormalities detected in either the procedures orthe results. SGS has ISO/IEC 17025 accreditation for its mineral analytical services.

It is Largo’s opinion that the sample preparation, security and analytical procedures were satisfactory.

Micon concurs with this view.

11.1.B Sample Preparation and Analysis

The drillhole core intersections from Largo’s drilling programmes are stored under an open sided roofed

structure at the exploration camp.

The drillhole core is stored in wooden core boxes with a nominal capacity of approximately 4 m for NQ or

NQ2 sized drill core and 3 m for HQ or HW sized core. The drillhole number, Project name, box number,and downhole depths are stamped onto an aluminum tag and affixed to the edge of the box. Woodendownhole core depth markers are placed in the core box by the driller and affixed with an aluminum tagstamped with the depth, the length of the interval, and the length of the recovered sample.





Figure 11-1 Drillhole Core with log intervals delimited

The staff geologists are responsible for logging drillhole cores. Drillhole core logging and sampling isdone at Largo´s exploration team.

11.2.B Core Logging

Before sampling, the geologist completes a graphic log and logs the core in detail for lithology, structure,mineralization and alteration. (Figure 11-2 and Figure 11-3)

Sample intervals and sample numbers are also recorded on the report drillhole log. This information isstored physically and digitalized, Largo´s team has been kept the information on site, and another placesdue security policy.

Core sample recovery is not recorded by the geologist, although a record of the drillhole recovery on arun by run basis is recorded manually by the driller and it is checked by sampler always drilled core is

delivered to logging and sampling in the facility area. It was illustrated in Figure 11-2.

This information is typed by a geological technician into a digital file for each hole. The recovery in themineralized zones is generally very good, on average better than 95%.





Figure 11-2 Drillhole Core with tagged intervals delimited




Figure 11-3 Drillhole Core Log Example





11.3.B Core Sampling

After drillhole core is received at the sampling shed, the entire length of the drillhole is checked andmarked for lithological contacts. Samples are marked down the entire length of the hole at different lengthof intervals, always respecting at lithological contacts. The sample length depend on the width of the corezone considered interesting to be sampled. A double arrow on the side of the channel marked on the boxwith a pen indicates the start and end of the sample interval, after the paper sample number tags areplasticized and stapled to the core box next to the corresponding sample. It is illustrated in Figure 11-4.

Figure 11-4 Drillhole Core with tagged intervals delimited

Samples are selected down the entire length of the drillhole core, sawn in half with an electric diamondbladed core saw, and sampled prior to logging. Half core samples are selected by a Largo´s geologytechnician or by a trained sampler. Figure 11-5 illustrates the bladed core procedure.

The samples are then placed in a numbered plastic bag along with a paper sample tag, and tied closedwith a piece of string. Sample weight is around 3 kg.





Figure 11-5 Drillhole Core electric diamond bladed core

Batch samples are placed in a larger plastic bag, loaded onto a truck owned and driven by a locally basedtransport company, and driven to the SGS laboratory, or Intertek Lab, which ever is available. In the caseof SGS, it is forwarded to a sample preparation facility in Belo Horizonte, State of Minas Gerais, or in caseof Intertek to São Paulo, State of São Paulo.

Largo then receives the sample back from the lab for storage at the project site. It is stored and available

for viewing inside the drilling facility seen in Figure 11-6.

Coffey Mining considers the core sampling security to be above current industry best practice.

Coffey Mining considers the sampling preparation, analysis and security to be in line with current industrybest practice and adequate for the purpose of mineral resource estimation. The QAQC procedure ofinsertion of standard, blank and duplicate samples is discussed in detail in Section 12 as Coffey Mining’s

data verification program.





Figure 11-6 Batch Sample Storage Facility

11.4.B Density Determination

The mass density or density of a material is defined as its mass per unit volume. Mathematically, density(ρ) is defined as mass (m) divided by volume (V). From this equation, mass density must have units of a

unit of mass per unit of volume (eg. g/cm3 , kg/m³, etc.).

The methodology used to determine the density was Archimedes principle further reasoned that if theliquid in this volume were removed and replaced by an object of exactly the same size and shape as thisliquid portion, none of the liquid pressure forces acting on its surface would change. Because the object isexactly the same shape and volume as the fluid removed, it would fit exactly into the previous volumewithout compressing the surrounding fluid. Therefore, Archimedes (287 - 212 B.C) had concluded that thenet buoyant for B upward on any object immersed in a fluid is equal to the weight of the fluid displaced.

All Largo´s Density samples were analyzed in Federal University of Bahia State, using the Archimedesprinciple. Some of this procedure is illustrated in Figure 11-7.





Figure 11-7 Density Determination by Archimedes Principle





Table 11-1 Density Summary

11.5.B Sample Preparation and Analysis

Sample preparation and analysis of core samples taken by Largo were performed by SGS Lakefield-Geosol Ltda. (“Geosol”), an ISO 9000-2001 certified laboratory. Sample preparation procedurescompleted by the Geosol preparation laboratories based in Belo Horizonte, Minas Gerais State were:

Volumetric with Dicromate of Potash;

Fire Assay - ICP; LOI: Loss on ignition - Sample Calcination between 405ºC / 1000ºC; Melting with tetraborate of lithium - X-Ray fluorescence;

Sample pulps were analyzed for V2O5 using a fire assay technique with an atomic absorption finish.Selected samples were subsequently sent for elements analysis by ICP spectrometry as describedbefore.

Target Lithology Number of Samples Density (g/cm3)

MAGNETITITO 26 4.36

MAGNETITA GABRO 11 3.38

GABRO 11 3.00

ANORTOSITO 3 2.88

PEGMATITO 2 2.61

MAGNETITITO 30 4.26


GABRO 11 3.06

ANORTOSITO 5 2.86

PEGMATITO 3 2.6

MAGNETITITO 19 4.3


GABRO 9 3.05ANORTOSITO 3 2.84

PEGMATITO 3 2.58

MAGNETITITO 14 4.28


GABRO 8 3.03

ANORTOSITO 5 2.84

PEGMATITO 2 2.62

MAGNETITITO 15 4.42


GABRO 3 2.90ANORTOSITO 1 2.83

PEGMATITO 2 2.63

GULÇARI A NORTH

GULÇARI B

NOVO AMPARO

NOVO AMPARO

NORTH

SÃO JOSE





11.6.B Adequacy of Procedures

Quality control data from other exploration companies prior to Largo’s involvement has not been checked.In addition, no meaningful legacy analysis data has been considered in this document.

The current analytical method is appropriate. Sufficient quality control data exists to allow thorough review

of the analytical performance of samples taken by Largo.

The sampling methods, chain of custody procedures, and analytical techniques are all consideredappropriate and are compatible with accepted industry standards although the sample preparation ofvanadium should be reviewed in light of the QAQC analysis in the following section.





12 Data Verification

In Item 12, Data Verification, both Micon and Coffey Mining have provided individual sections. Althoughthere is some duplication of information it was felt by RPM that the commentary and emphasis wassufficiently different to justify presenting both Micon and Coffey Mining data.

The Micon items have been designated with the letter “A” and the Coffey Mining items have beendesignated with the letter “B”.

12.1.A Pre-2006

It is reported that CBPM and Odebrecht made use of replicate and blank samples along with referencestandard materials during their sampling and assaying programs. No detailed results are available for theresults of these programs. Therefore, before using the data in a resource estimate, Largo decided toconduct a check sampling program on approximately 8% of the mineralized core. That program isdescribed in Item 12.2 below.

12.2.A Largo

12.2.1.A 2006

Largo’s first drill core resampling program was conducted during May, 2006, in order to verify the

precision and accuracy of the V2O5 grades reported during the 1981 through 1987 drill campaigns. It wasalso used to provide additional data on the PGM content of the mineralization. A total of 123 quarter-coresamples from eight drillholes (7.3% of samples) were analyzed. Check analyses were systematicallycompleted at a second laboratory during the resampling program.

The original analyses were done at GEOSOL and Paulo Abib Engenharia S.A. laboratories in Belo

Horizonte, Brazil between 1981 and 1987. The 2006 duplicate sample analyses were conducted at SGS,both in Belo Horizonte, Brazil and Lakefield, Ontario. Every effort was made to use similar techniquesand sample sizes, in order to compare results. Check analyses on the 2006 duplicate samples wereanalyzed by Ultra Trace Analytical Laboratories in Perth, Australia (Ultra Trace). A total of 25 pulpsamples (20%) from the resampling program were sent to Ultra Trace for analysis to compare against theduplicate results. Again, every effort was made to use similar techniques.

The duplicate samples sent to SGS were analyzed for the major oxides (FeO, Fe2O3, SiO2, TiO2 andV2O5) using borate fusion XRF. The lower detection limit for V2O5, for this method, was 0.01%, while theupper limit was 5%. Elements were reported as oxides. The samples were also analyzed by SGS forplatinum and palladium by fire assay with an AA finish on a 50-g sample. Internal quality control

procedures included duplicate and blank sample and certified reference material analysis. These datawere used to check the analytical reproducibility and precision of the assays.

Ultra Trace’s XRF method used a fusion technique, with a high -energy X-ray instrument. The resultingdetection limits are reported to often be better than those obtained by pressed powder methods on olderinstruments. The lower detection limit for V2O5 was 0.01%, while the upper limit was 5%. A sampleweight of 0.66 g was fused with 7.2 g of flux to prepare each bead. However, there are variations to thisratio for some matrices. Samples were fused into a glass disc using a lithium borate flux much as





described for normal fused glass beads. For ore grade materials, flux composition and sample/flux ratiosare varied to ensure the entire sample dissolves and that recrystallization does not occur as the melt iscooled.

Two comparisons of the results were carried out:

1. Original sample results versus SGS duplicate sample (quartered core) results2. SGS duplicate sample (quartered core) results versus Ultra Trace check pulp sample results.

In both cases, the agreement is generally good. Overall, there is little evidence of any systematic orconditional bias. The correlation coefficient between the original samples and the duplicate samples is0.84, a number considered reasonably good for quarter-core field duplicate samples. Any variability inthe sample results can be attributed to a number of conditions including differences in sample mass orhalf-core versus quartered core. This is partly confirmed by the comparison of the SGS duplicatesamples and the check sample results of the pulps from Ultra Trace where the correlation coefficient is0.89.

It is concluded, therefore, that the analytical reproducibility is satisfactory, and that the analytical accuracy

is equally acceptable. Consequently, Largo chose to use the original assay data for the geostatisticalanalysis.

Micon concurs with Largo’s decision and concludes that the data are suitable for use in the resource

estimate presented herein.

12.2.2.A Early 2007

In accordance with the recommendations made in Micon’s December, 2006 Technical Report

(Hennessey, 2006), Largo has continued with a program of resampling of old drill core from the Gulçari Adeposit.

The results, which are graphed on Figure 12-1, generally show good agreement clustered about the 45°black reference line, with the exception of a clustered group of data (see red ellipse) appearing to fall on aflatter line of about 30° dip. Further investigation revealed that all of these data points were from DrillholeFGA-41.





Figure 12-2 Gulçari A Core Duplicate Sampling With Hole FGA-41 Removed

12.2.3.A 2007 Exploration Drilling Program

Largo’s drill-core sampling program was conducted from April 3 to October 30, 2007. A qualityassurance/quality control (QA/QC) program was conducted, in order to verify the precision of the V2O5,platinum and palladium grades reported during the drill campaign. It was also used to provide additionaldata on the PGM content of the mineralization.

The QA/QC procedures consisted of the insertion of one of two certified reference standards, two fieldduplicate samples, and one field blank sample with each batch of samples sent to the laboratory. Thelaboratory batch size is 40 samples, of which 5 were Largo QA/QC samples. There were also laboratory-inserted blanks and duplicates used by SGS in accordance with its own QA/QC policy.

Certified Reference Standards

Largo had two certified reference standards made using pulps from an earlier drill program that sampledmaterial from the Gulçari A deposit. The standards consisted of both a high-grade (magnetitite) andlower-grade (magnetite-pyroxenite) material. The high and low standards were inserted at a rate ofapproximately one each per batch (40 samples).

These certified reference standards were prepared and packaged by CDN Resource Laboratories ofDelta, B.C. Each sample was pulverized in a large rod mill, screened through a 200 mesh screen usingan electric sieve and homogenized in a large rotating mixer. Each standard was sealed in plastic toprevent gravity separation and oxidation.

Each of the standards underwent blind round robin assaying for V2O5 by five laboratories and the datawere reviewed and certified by Barry W. Smee, Ph.D., P.Geo. (Smee & Associates Consulting Ltd., see

y = 0.9701x

R 2 = 0.9221

0.00

1.00

2.00

3.00

4.00

5.00

0.00 1.00 2.00 3.00 4.00 5.00

Original Assays (% V2O5)

S G S C h e c k A s s a y s ( % V 2 O 5 )





Appendix 2 of Hennessey, 2007). Both standards were also analyzed for PGM and the high-grademagnetitite was found to have high enough values to potentially be useful as a precious metals standard.Experience with it has found that while it repeated well as a platinum standard, it did not perform well as apalladium standard.

All of the analytical results for the high and low standards were tracked on control charts on a continuous

basis from April 3 to October 30, 2007. Each of these charts tracks the results of assaying of a singlestandard over time and plots it against the accepted value (the mean from the round robin assayprogram) and ±2 standard deviations (SD) from the mean. Staying within the ±2 SD lines is acceptableperformance for precision and accuracy at a laboratory.

The performance of the in-house V2O5 standards used by Largo is judged to be acceptable. Figure 12-3to Figure 12-6 show the V2O5 analytical results for the high and low standard, as well as the platinum andpalladium results for the high standard.

Figure 12-3 High Standard V2O5 Assay Results

Red line = the accepted value, blue lines = ± 2 SD

Maracas Project - V2O

5 High Standard

2.30

2.40

2.50

2.60

2.70

2.80

2.90

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Lab Analyses

V 2 O 5 %





Figure 12-4 Low Standard V2O5 Assay Results


Figure 12-5 High Standard Platinum Assay Results


Maracas Project - V2O5 Low Standard

0.80

0.85

0.90

0.95

1.00

1.05

1.10

1.15

1.20

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

Lab Analyses

V 2 O 5 %

Maracas Project - High Standard Pt Results

450

500

550

600

650

700

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Lab Analyses

P t p p b





Figure 12-6 High Standard Palladium Assay Results


Field Blanks

Field blanks of a known barren rock were randomly inserted at least once in every 40 samples, usuallyresulting in one sample per batch. This was done to check for cross contamination at any point in the

sample preparation or assaying. Micon has reviewed the analytical results for the field blanks (see Figure12-7) and found them to be acceptable.

Maracas Project - High Standard Pd Results

110

120

130

140

150

160

170

180

190

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Lab Analyses

P d p p b





Figure 12-7 Field Blank Assay Results

Duplicate Samples

Two field duplicate samples were randomly inserted in each batch. These samples are used todetermine the precision of the assay laboratory and the degree of nugget effect introduced in sampling.Detailed records were kept at the core shed for all field duplicate sample locations.

SGS also introduced sample duplicates prepared at the laboratory into the stream. These samples areuseful in determining the analytical precision of the laboratory.

The results of the field duplicate sampling are presented in Figure 12-8 and the results of the sample

duplicate assaying are presented in Figure 12-9. Micon has reviewed these results and finds them to beacceptable and better behaved than most.

Field Blank Results

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171

V 2 O 5 ( % )





Figure 12-8 Field Duplicate Assay Results

Figure 12-9 Sample Duplicate Assay Results

Secondary Laboratory Checks

Check analyses were systematically completed at a second laboratory during the drill program, in order totest the precision and relative bias of the primary laboratory. A total of 500 pulp samples from 40drillholes (9.1% of samples) were analyzed.

The original analyses were done at the SGS laboratory in Belo Horizonte, Brazil. The 2007 duplicatesample analyses were conducted at ALS Chemex laboratory in Vancouver, B.C. Every effort was made

Maracas Project - Original vs Duplicate Analyses

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

Original Samples V2O5%

D u p l i c a t e S a m p l e s V 2 O 5 %

Original vs Duplicate V2O5 Laboratory Analyses

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

Original Values V2O5%

D u p l i c a t e V a l u e s

V 2 O 5 %





to use similar techniques and sample sizes, in order to compare results. The longest lapse of timebetween the original assays at SGS and the secondary checks at ALS Chemex was 4 months, and theshortest period of time was 2 months.

The samples sent to SGS were analyzed for the major oxides (FeO, Fe2O3, SiO2, TiO2 and V2O5) usingborate fusion XRF. The lower detection limit for V2O5, for this method, was 0.01%, while the upper limit

was 5%. Elements were reported as oxides. The samples were also analyzed by SGS for platinum andpalladium by fire assay with an AA finish on a 20-g sample. Internal quality control procedures includedduplicate and blank sample and certified reference material analysis. These data were used to check theanalytical reproducibility and precision of the assays.

ALS Chemex’s XRF method used a fusion technique, with a high-energy X-ray instrument. The lowerdetection limit for V2O5 was 0.01%, while the upper limit was 5%. A sample weight of 0.66 g was fusedwith 7.2 g of flux to prepare each bead. Samples were fused into a glass disc using a lithium borate fluxmuch as described for normal fused glass beads. The samples were also analyzed for platinum andpalladium by fire assay with an AA finish on a 20-g sample.

Comparisons of the results for V2O5, platinum and palladium were carried out (see Figure 12-10, Figure12-11 and Figure 12-12) in an analysis of original sample results versus ALS Chemex duplicate sample(pulps) results. In all cases, the agreement is generally good. Overall, there is a little evidence for a slightsystematic high bias for the ALS Chemex V2O5 results. The correlation coefficient between the originalsamples and the duplicate samples for V2O5, platinum and palladium were all 1.02.

Largo, therefore, concluded that the analytical reproducibility was satisfactory, and that the analyticalaccuracy is equally acceptable. Consequently, Largo chose to use the original 2007 assay data for thegeostatistical analysis. Micon supports the decision.

Figure 12-10 Secondary Laboratory Check Assays – V2O5

V2O5 (%)

y = 1.0159x

R2 = 0.9963

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00

Original V2O5 values SGS

D u p l i c a t e V 2 O 5 v a l u e s C h e m e x





Figure 12-11 Secondary Laboratory Check Assays – Pt

Figure 12-12 Secondary Laboratory Check Assays – Pd

Summary

The purpose of adding a QC program to any drill program is to verify the accuracy and precision of thelaboratory’s results and to react immediately to any deviation at the laboratory demonstrated by the

standards. Use of the certified reference materials in this case has meant that the laboratory results forthe metals can be said to be reasonably accurate and precise.

Pt (ppb)

y = 1.0152x

R2 = 0.9911

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Original Pt (ppb) SGS

D u p l i c a t e P t ( p p b ) C h e m e x

Pd (ppb)

y = 1.0246x

R2 = 0.9804

0

100

200

300

400

500

600

700

800

900

1000

0 100 200 300 400 500 600 700 800 900 1000

Original Pd (ppb) SGS

D u p l i c a t e P d ( p p b )

C h e m e x





The SGS pulp duplicates from the 2007 drilling generally show close agreement and little bias betweenfirst and second assay. The correlation coefficients confirm the close agreements between the originaland the laboratory duplicates. Largo’s field duplicates also showed good reproducibility.

Largo’s 180 field blanks routinely returned very low values for all elements with three excepti ons. Twosamples, LML6272 and VML1682, assayed above background precious metal levels and V2O5 and one

sample, LML6056, showed elevated V2O5 levels.

The performance of the certified reference standards were generally good for the high V2O5, low V2O5 and platinum determinations with over and under limit palladium assays returned. The high V2O5 resultswere stable and consistent and all clustered between the ±2 SD control limits. The results for low V2O5 standard were good with only one sample that fell outside the ± 2SD control limits. The high V2O5 standard is not really a precious metal standard. However, the values for platinum were very consistent,except for one sample and within acceptable ±2 SD limits. The palladium results on the other hand werevariable showing no consistent drift.

12.3.A Micon

During the first site visit in June 2006, Micon visited the Maracás Project site and the surface exposure ofthe Gulçari A deposit. Magnetite and magnetite-pyroxenite mineralization exposed on surface in outcropand trenches on Gulçari A hill were seen. Significant amounts of magnetite mineralization, consistentwith the claimed widths and grades were seen. Micon also visited the office/warehouse facility inMaracás to look at old core stored there.

During the visit, Micon collected two samples of mineralization from exposures at the Project site and fourmore samples of quarter-sawn core from the core warehouse. The samples were collected and assayedto confirm the presence of vanadium, titanium and PGMs. The results of the sample program are set outin Table 12-1.

Table 12-1 Micon Check Samples

The results obtained by Micon have confirmed the presence of vanadium, titanium and PGMmineralization at Maracás at grades which were expected and which broadly agree with the historicalresults.

12.4.A Conclusions

The data verification work completed by Largo and Micon has led to confidence in the database compiledby the original operators of the property. Largo’s ongoing QA/QC program has also led to confidence in

the newly-generated data.

Sample

Location From To V2O5 TiO2 Au Pt Pd From To V2O5 TiO2

(m) (m) (%) (%) (g/t) (g/t) (g/t) (m) (m) (%) (%)

Maracás 1 Outcrop 2.67 17.03 0.007 0.490 0.155 N/A N/A

Maracás 2Surface

trench 2.00 12.52 0.011 0.713 0.265 N/A N/A

Maracás 3 FGA-40 112 113 3.11 18.81 0.005 0.653 0.031 112 113 6.3 17.4

Maracás 4 FGA-41 63 64 0.70 5.04 0.041 0.063 0.060 63 64 1.3 5.3

Maracás 5 FGA-38 92 93 2.04 15.86 0.007 0.224 0.171 92 94 2.2 16.1

Maracás 5 FGA-54 74 75 2.40 19.20 0.003 0.360 0.040 73 75 2.8 18.1

magnetitite

mag-pyroxenite

Sample

Micon Results Original Results





Micon concludes that the Maracás database is suitable for use in the mineral resource estimate reportedherein. In Hennessey (2006), Micon recommended that Largo proceed with planned check drilling and tocontinue re-assaying of historical drill core for PGMs. Micon also recommended that Hole FGA-41 be re-drilled. This work has now been completed and its results included in the updated mineral resourceestimate (Hennessey, 2007) that has been utilized for the present report.

Only the data verification results for the drillholes employed in the Gulçari A mineral resource estimateused in this report have been presented herein.

12.1.B Geological Database

Coffey Mining has validated the Largo database for Maracás Vanadium Project using the GemcomSurpac Software System Database Audit tool with no inconsistencies noted.

A comparison of hardcopy assay and geological logging versus the digital database was performed on atotal of 10% of the Largo - Maracás Vanadium Project drillholes. No errors were identified with theoriginal log and the digital database.

12.2.B Quality Control

Coffey Mining has not been able to verify the oldest drill sample QAQC data before 2006, in spite of it hadobserved in geological database.

A quality control and quality assurance program was implemented by Largo on the complete drillingprogram samples analyzed at Intertek laboratory and SGS laboratory (Table 12-2), including:

Table 12-2 Standards and Blanks QAQC Summary Results

In total of blanks, standards and duplicates for quality control Largo had respect the rule as indicated, 2standards (Low Grade and High Grade), 1 duplicate + 1 blank for a batch of 32 samples, totalizing 36 perbatch.

The quality control data has been assessed statistically by Coffey Mining using a number of comparative

analyses for available datasets. The objectives of these analyses were to determine relative precision andaccuracy levels between various sets of assay pairs and the quantum of relative error. This assessment isthe basis of the data verification program undertaken by Coffey Mining. No duplicate samples have beencollected or analyzed by Coffey Mining.

Sample Type Objective Procedure

Blank

Verification if there is contamination in the

samples during the physical preparation

process and chemical analysis.

Always respecting an interval of 32 samples

StandardVerification of the Laboratory analysis

precisionInserted after each 16 samples

DuplicatesVerification of the Laboratory analysis

AccuracyAlways respecting an interval of 32 samples





12.3.B Standard and Blank Samples

Coffey Mining performed an analysis on blanks data provided by Largo. The blank material was sourcedby Largo from unmineralized building material acquired at one specific supplier in Maracás city, close siteat the project and submitted at a frequency of about five percent.

Overall the blank and standard data is within acceptable limits, and this results was presented in Table12-3, Table 12-4 and Table 12-5 below.

Table 12-3 Quality Control Sample Summary

Largo had produced two types of standards for introduce on its process of sample analysis. Eachstandard was analyzed in six different Labs. These Labs were certified on the best practices of the marketanalysis.

Figure 12-3 is presenting how the standards are stored and it is packaged to be used in the samplingroutine analysis.

Table 12-4 Internal Standard Detection limits

Table 12-5 Standards and Blanks QAQC Summary Results

Chemical Element Sample Type Number of Samples Analyzed

Standard 380

Blanks 372V2O5

Standard Mean Mean + 2SD Mean - 2SD Mean + 3SD Mean - 3SD

V2O5 High grade 2.60 2.76 2.54 2.84 2.46

V2O5 Low grade 0.988 1.054 0.922 1.087 0.889

Blank / V2O5 0.01 0.005 0.015 199 0.001 0.18 0.008 84.422 15.578

High Grade/ V2O5 2.60 2.52 2.68 187 2.38 2.53 2.45 3.21 96.79

Low grade/ V2O5 0.988 0.955 1.021 193 0.95 1.06 1.015 63.212 36.788

Blank / V2O5 0.01 0.01 0.01 173 0.005 0.05 0.007 95.954 4.046

XRF_SGS

%

inside

precision

limits

Min

(%)

Max

(%)

Mean

(%)

%

outisde

precision

limits

Site Project

Standard / VariableOrigin

(%)

Min

(%)

Max

(%)

Sample

N°

Reference values Analyzed Results Results





Figure 12-13 Standard to V2O5 produced by Largo and Internationally certified

The Annex B - QAQC Analysis presents all detailing about the standard and blanks presented on thissection.

12.4.B Comparative Data Analysis

The comparative data was statistically analyzed using the Coffey Mining QC Assure software. Theobjective of this is to determine the relative precision among some pairs of results and to quantify therelative error.

The duplicate samples are sorted from a half of drill core sample. The analyzed duplicates and replicatesare summarized in Table 12-6.

Largo maintained the routine to duplicate and replicate in order to improve the confidence in the labresults.





Table 12-6 QAQC Program Summary

From 196 existing field duplicate sample pairs in the database for the V2O5 % variable more than 88.78%of the data pairs are in the acceptable 10% precision limits for this type of duplicate. Another approachcan be observed from 275 existing lab duplicate sample pairs in the database for the V2O5 % variablewith more than 93.82% of the data pairs 10% precision limits for this type of duplicate. Commonly, thisresult is over the acceptable lower limit of 90%. Also in sequence is possible to present Replicate (LabSGS) with 93.6%, Check Lab SGS vs Intertek 96.94% and Check Lab SGS vs ALS is 94.43%. The fieldduplicate and was presented in Figure 12-14, more graphs were presented in Annex B - QAQC Analysis

Coffey Mining concludes that the results of the sample duplicates of Largo are normal for the style of

mineralization and acceptable for mineral resource estimation purposes.

Chemical Element Sample Type Number of Samples Analyzed

Field Duplicate 196

Duplicate (Lab SGS) 275

Replicate (Lab SGS) 296

Check Lab SGS vs Intertek 359Check Lab SGS vs ALS 305

V2O5V2O5





Figure 12-14 Duplicates Summarized Quality Control

V2O5_DUP V2O5 Units ResultNo. Pairs: 196 196 Pearson CC: 0.992Minimum: 0.009 0.010 % Spearman CC: 0.986

Maximum: 1.810 1.650 % Mean HARD: 4.191Mean: 0.248 0.248 % Median HARD: 1.563Median 0.170 0.170 %Std. Deviation: 0.286 0.278 % Mean HRD: -1.041Coefficient ofVariation: 1.152 1.117 Median HRD 0.000

0

20

40

60

80

100

0.001 0.01 0.1 1 10

H A R D ( % )

Mean of Data Pair (%)

Mean vs. HARD Plot(All Data)

Mea n HAR D: 4.191 Med ia n HAR D: 1.563

Precision: 10%

0

20

40

60

80

100

0 10 20 30 40 50 60 70 80 90 100

H A R D ( % )

Rank (%)

Rank HARD Plot(All Data)

Precision: 10%

88.776% of data are withinPrecision limits

0

20

40

60

-1.0 0.0 1.0

F r e q u e n c y ( % )

HRD (/100)

HRD Histogram(All Data)

Mean HR D: -1.041 Med ia n HRD: 0.000Precision: +/-10%

-100

-50

0

50

0.001 0.01 0.1 1 10

H R D ( % )


Mean vs. HRD Plot(All Data)

Mea n HRD: -1.041 Media n HR D: 0.000Precision: +/-10%

0.0001

0.001

0.01

0.1

1

0.001 0.01 0.1 1 10 A b s o l u t e D i f f e r e n

c e ( % )


T & H Precision Plot (Assay Pairs)(All Data)

10% 20% 30% 50%

0.001

0.01

0.1

1

0.01 0.1 1 10

M e d i a n A D ( % )

Grouped Mean of Pair (%)

T & H Precision Plot (Grouped Pairs)(All Data)

10% 20% 30% 50%

0.0

0.5

1.0

1.5

2.0

0 .0 0 .2 0 .4 0 .6 0 .8 1 .0 1 .2 1 .4 1. 6 1 . 8 2 .0

V 2 O 5 ( % )

V2O5_DUP (%)

Correlation Plot(All Data)

P.CC = 0.992 S.CC= 0.986 Ref. Li ney = 0.964x + 0.009

0.0

0.5

1.0

1.5

2.0

0. 0 0. 2 0 . 4 0 .6 0 .8 1 .0 1 .2 1 .4 1 .6 1 .8 2 .0

V 2 O 5 ( % )

V2O5_DUP (%)

QQ Plot(All Data)

Ref. Line y = 0.970x + 0.008

Duplicate - Largo Resources - V2O5(All Data)

Printed: 24-Sep-2012 04:37:13 Data Imported: 24-Sep-2012 04:31:56 Page 1





12.5.B Data Quality Summary

The standards data has shown a moderate to high accuracy as returned by the SGS Geosol laboratory,Intertek and it is noted that SGS supplied the standards to BGC.

Coffey Mining, after analyzing all procedures and results gathered under the QA/QC program undertaken

by Largo, conclude that the data is of sufficient quality to support a resource estimate.





13 Mineral Processing and Metallurgical Testing

13.1 Introduction

The process design is based primarily on the metallurgical testwork performed by SGS in 2007, a studyundertaken by IMS Process Plant in 1990, a feasibility study completed by Lurgi in 1986, a metallurgicalstudy performed by Rautaruukki Oy Research Centre between 1987 and 1989, and the detailed technicalstudy produced by Engenharia e Consultoria Mineral S.A. (ECM) in 1990. A list of metallurgical andprocess technical and economic references can be found in Item 13.2.

Testwork was undertaken by SGS between April and November 2007 to investigate the recovery ofvanadium from Maracás mineralization. This program included mineral processing investigations usingmagnetic separation to recover vanadium contained in magnetite and hydrometallurgical extraction usingroasting, leaching, precipitation and calcining to produce an intermediate vanadium oxide product.

Additional SGS testwork was undertaken in 2012 to investigate beneficiation recoveries and concentrateanalyses for the additional ore-bodies included in the expanded plan presented in this report.

Pilot scale testing was undertaken by Largo in 2010 to test bulk samples of high grade and low grade orewith respect to recovery and leaching performance.

13.2 Process Technical and Economical References

The following reports were used as a basis for the design and the development of associated plant capitaland operating cost estimates.

Lurgi, “Feasibility Study, Maracás Vanadium Project, prepared for Pedeiras Valeria Ltda.,

Salvador/Bahia, Brazil”, May 1986. Rautaruukki Oy Tutkimuskeskus Research Centre, “Laboratory Research of the Suitability of

the Otankäki Process for Extracting Vanadium from Maracás Ore”, December 1989. Engenharia e Consultoria Mineral S.A., “Projeto Vanádio de Maracás Projecto Conceitual e

Estimativa de Investimento, Produção: 4,500 t/a de V2O5”, September 1990. IMS Process Plant, “Vanádio de Maracás Ltda., Vanadium Pentoxide Production Plant”,

1990. SGS Minerals Services, “The Beneficiation Characteristics of Samples from The Vanádio De

Maracás Deposit” November 2007. SGS Minerals Services, “Recovery of Vanadium from the Maracás Ore Deposit”, April 2008.

SGS Minerals Services, “The Solid-Liquid Separation of the Maracás Ore Deposit”, July

2008 Vendors’ budgetary quotes

Largo Resources Ltd. (Les Ford), “Pilot Plant Testing of Maracás Magnetite Ore”, Oct 2010. Ausenco Minerals and Metals, "Conceptual design of alternatives for non-magnetic tailings

deposition", Sep 2010.

13.3 Material Characterization, Mineralogy and Metallurgy

The main basis for the process design is the testwork results by SGS, input f rom external consultants whohave extensive experience on vanadium plant design and operation, as well as Largo’s experience. The





metallurgical design parameters discussed below are mainly based on the data included in the referenceslist above.

The objective of the SGS testwork conducted on samples of vanadium-magnetite mineralization was tolook into the suitability of the magnetic separation, salt roasting and hydrometallurgical process forextracting vanadium from Maracás mineralization. Additionally SGS investigated the magnetic separation

efficiency on lower grade mineralization.

13.3.1 Mineralogy and Chemical Analysis

The Maracás mineralization consists of vanadium-containing titanium magnetite, which is the major oxidephase within the deposit. Magnetite occurs as primary grains that may be partly martitized. There is alsofine-grained magnetite as inclusions in the silicate grains. This included magnetite, which containsnegligible vanadium, as a secondary alteration formed after uralitization of pyroxene and serpentinizationof olivine.

The chemical analyses of the metallurgical composite samples reported by SGS in its November 2007metallurgical testwork are presented in Table 13-1. Additional analyses carried out in 2012 are presentedin Table 13-2.

Table 13-1 SGS Test Sample Analyses - 2007

Source: SGS / Lakefield





Table 13-2 SGS Test Sample Analyses – 2012

Table 13-3 presents the mineralogical composition for both massive and oxide ore samples. Magnetiterepresents about 60.9% of the massive ore and ilmenite represents about 19.9%.

Table 13-3 Maracás Ore Mineralogical Composition

In Figure 13-1, the mineralogical associations of Ti-magnetite have been illustrated.

In Figure 13-2, free Ti – magnetite and ilmenit middles shown at -212 microns.

Deposit V2O5 (%) Fe (%) TiO2 (%) FeO (%) Fe2O3 (%) LOI (%) SiO2 (%) Al2O3 (%) CaO (%) MgO (%) K2O (%) Na2O (%)

Gulçari A N 0.85 30.57 8.02 17.43 24.34 -1.01 27.52 9.87 5.09 2.91 0.60 1.28

Gulçari B 0.74 42.65 14.71 22.17 36.41 -2.29 14.07 6.20 1.79 2.00 0.38 0.34

São José 0.84 33.35 9.51 19.03 26.52 -1.08 24.82 9.77 4.25 2.02 0.58 1.21

Novo Amparo 0.71 42.60 14.35 24.31 33.89 -2.11 13.91 6.02 2.26 2.51 0.32 0.40

Novo Amparo Norte 0.90 32.94 9.89 20.64 24.76 -1.01 23.82 10.54 4.54 1.57 0.54 1.39

Mineral Massive Sample (%) Oxide Sample (%)

Ti-magnetite 60.90 67.30

Ilmenite 19.90 23.50

Goethite 0.32 0.99

Cummingtonite / Actin 6.51 2.08

Hornblende 4.43 1.42

Clinopyroxene 0.82 0.24

Orthopyroxene 0.06 0.03 Chlorites 2.54 2.71

Biotite+Phlogopite 1.56 0.03

Mica/Clay-Mg 1.51 0.70

Talc 0.34 0.19

Quartz 0.40 0.31

Feldspars 0.51 0.06

Other silicates 0.21 0.20

Apatite 0.02 -

Carbonates 0.08 0.16

Sulphides 0.14 0.07

Others - 0.01

TOTAL 100.00 100.00





Figure 13-1 Ti-magnetite Associations by Size – Massive


Figure 13-2 Free Ti-Magnetite and Ilmenite Middles shown at -212 microns





13.3.2 Comminution

Laboratory crushing tests, overseen by Rautaruukki Oy, were undertaken in the crushing laboratory ofLokomo factory of Rauma-Repola Oy, Finland in December 1989. The average crushing work index was6.9 kWh/t and the maximum 11.3 kWh/t.

The average abrasion index of the Maracás test samples was 0.42. Massive rock abrasion index was0.243.

No recent crushing testwork was performed.

Grinding testwork was undertaken by SGS in 2007. The rod mill and ball mill work indices for differentcomposites are shown in Table 13-4 and Table 13-5.

Table 13-4 Bond Rod Mill Grindability Test Summary


Table 13-5 Ball Mill Grindability Test Summary






The massive and oxide composites were both soft with respect to rod mill grindability, which characterizescoarse (primary) grinding. The rod mill work indices for the massive and oxidized composites measured10.4 and 10.0 kWh/t, respectively. The ball mill work indices, which represent fine secondary grinding,were medium. They measured 14.2 and 12.9 kWh/h, respectively, for the massive and oxidizedcomposites.

13.3.3 Concentration (Magnetic Separation)

Davis tube tests were performed by SGS in 2007, in order to determine the effect of the followingparameters on vanadium recovery:

Grind; Davis tube intensity; Ore variability.

The results on the composite samples are summarized in Table 13-6 and Table 13-12. Maximumseparation efficiency is obtained at 100% passing 50 microns. With feed grades composites of 2-2.4%

V2O5 the vanadium recovery was, above 94% for the massive bulk composite and above 91% for theoxidized bulk composite, and the silica content was low, 1.18% maximum, for samples with less than 106microns.

Additional Davis Tube tests were performed in 2012 in order to determine concentrate recoveries andgrades on the additional ore bodies included in the expanded mine plan. The additional work, whoseresults are listed in Table 13-7 through Table 13-11 indicate vanadium recoveries of 78-89% at headgrades of 0.7-0.9% V2O5 with average concentrate grades ranging from 1.6% to 2.9% V2O5.

Davis Tube Recovery test results for tailings composites can be seen in Table 13-13 through Table13-17.

13.3.4 Roasting, Leaching and Precipitation

The SGS testwork program also investigated conditions of salt roasting for extracting vanadium from themagnetite concentrate.

Vanadium was successfully recovered from Maracás magnetic concentrates by salt roast leaching withsodium carbonate. Massive and oxidized magnetite samples of various compositions were treated in abench program, resulting in vanadium extraction exceeding 90%. Using the bulk magnetic concentratesproduced at SGS, an optimum vanadium leach extraction of 93.5% (Test L15, massive magnetite) and95.0% (Test L18, oxidized magnetite) were obtained.

Table 13-18 shows conditions of tests performed in 2007 on different samples.

A second bench program was carried out using bulk roasted samples. The bulk roasting was performedby FEECO, a kiln manufacturing company in Green Bay, WI, U.S.A., in a batch operating rotary kiln. Thevanadium extraction was generally lower than what was observed in the preliminary bench program.Vanadium extractions were 85.6% for the massive magnetite and 92.3% for the oxidized sample.

It is noted that since bulk roasting’s purpose was to generate enough material for leaching, the testingwas not performed under optimal recovery conditions.




Table 13-6 Davis Tube Summary on Composites – Concentrate – Gulçari A





Table 13-7 Davis Tube Summary on Composites – Concentrate – Gulçari A Norte


Table 13-8 Davis Tube Summary on Composites – Concentrate – Gulçari B


Mass Rec (%) V2O5 Rec (%) V2O5 (%) Fe (%) TiO2 (%) FeO (%) Fe2O3 (%) LOI (%) SiO2 (%) Al2O3 (%) CaO (%) MgO (%)

FGAN 01 23.65 83.15 2.74 63.62 2.64 31.59 55.88 2.47- 3.38 2.01 0.67 0.54

FGAN 02 42.04 88.69 2.89 63.50 3.63 31.96 55.30 3.26- 2.95 2.32 0.43 0.36

FGAN 04 19.80 62.55 3.24 61.69 2.51 30.50 54.33 4.45 2.35 1.01 0.58

FGAN 06 22.06 67.05 3.29 64.29 2.76 32.32 55.99 3.03 1.85 0.69 0.30

FGAN 07 17.52 68.26 3.65 65.43 1.94 32.62 57.25 2.02 1.56 0.49 0.26

FGAN 08 24.62 77.23 2.99 65.80 2.97 32.82 57.62 3.42- 0.93 1.50 0.17 0.05

FGAN 09 25.28 79.11 2.88 57.69 4.16 28.70 50.56 2.78- 7.44 3.68 1.34 0.59

FGAN 10 15.30 79.02 32.42 FGAN 11 29.90 80.92 2.58 67.10 2.90 33.25 58.97 0.71 1.32 0.12 0.17

FGAN 12 23.45 73.41 32.11

FGAN 13 9.85 45.60 2.67 67.34 0.75 32.83 59.78 1.83 1.00 0.41 0.27

FGAN 14 19.61 93.49 31.96

FGAN 15 21.60 77.43 3.79 67.10 1.92 33.10 59.15 0.61 1.00 0.15 0.19

FGAN 16 30.48 86.13 2.70 66.50 3.31 33.15 58.22 0.82 1.47 0.14 0.18

FGAN 17 27.84 88.28 2.98 66.35 3.08 32.59 58.61 0.68 1.58 0.15 0.20

FGAN 05 27.89 81.48 1.82 65.59 5.25 31.05 59.27 3.60- 0.92 1.69 0.14 0.22


FGB 10 43.22 86.22 1.74 63.11 6.58 24.38 63.18 2.59- 0.60 2.09 0.04 0.22

FGB 11 38.42 89.51 1.97 65.50 4.13 32.18 57.93 3.41- 0.92 2.05 0.12 0.27

FGB 12 23.47 75.04 2.11 65.80 2.91 33.15 57.30 3.45- 1.01 1.09 0.18 0.10 FGB 13 44.15 91.84 1.72 65.15 4.60 32.92 56.53 3.41- 0.83 2.18 0.10 0.30

FGB 14 41.13 90.46 1.72 64.69 4.97 32.89 55.99 3.62- 0.74 2.20 0.08 0.27

FGB 15 41.14 89.51 1.49 63.98 6.05 29.11 59.11 3.11- 0.82 2.38 0.07 0.28

FGB 16 40.47 90.52 1.45 64.56 5.77 31.73 57.00 3.44- 0.84 2.28 0.09 0.27

FGB 17 37.62 87.19 1.66 65.05 5.38 30.65 58.96 3.43- 0.99 1.80 0.15 0.21




Table 13-9 Davis Tube Summary on Composites – Concentrate – São José



FSJ 09 20.43 73.43 3.39 66.98 1.86 33.98 57.94 -3.49 0.87 1.41 0.16 0.05

FSJ 12 34.25 75.98 2.06 59.36 5.04 30.55 50.92 -2.90 6.43 3.40 0.97 0.51

FSJ 13 35.38 86.87 2.13 62.24 4.38 31.91 53.54 -2.90 3.61 2.64 0.68 0.38

FSJ 14 36.14 72.65 2.14 62.74 3.98 31.97 54.21 -3.27 3.46 2.54 0.53 0.30

FSJ 15 35.16 90.66 2.32 64.40 3.31 32.02 56.46 -3.46 2.69 2.17 0.42 0.24

FSJ 16 35.62 95.39 2.24 64.90 3.04 31.71 57.58 -3.33 2.76 1.83 0.38 0.28

FSJ 17 28.68 73.81 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

FSJ 18 36.46 87.84 2.36 64.69 3.92 32.41 56.50 -3.23 1.15 1.59 0.17 0.14FSJ 19 13.45 58.13 3.37 65.30 1.45 34.15 55.40 -3.23 2.29 1.77 0.43 0.19

FSJ 20 29.35 75.94 2.69 63.78 3.41 32.11 55.52 -3.28 2.78 2.18 0.45 0.24

FSJ 21 33.74 81.34 2.63 64.32 3.30 2.23 2.00 0.38 0.22

FSJ 22 23.47 71.67 2.40 64.86 2.65 27.39 62.33 2.52 1.75 0.43 0.11

FSJ 23 29.22 74.02 2.16 65.99 2.58 25.09 66.48 -2.08 1.62 1.28 0.25 0.13

FSJ 24 29.65 84.88 2.29 64.30 5.00 19.22 70.60 -1.73 0.59 1.12 0.05 <0.1

FSJ 25 25.32 88.45 2.49 67.05 2.63 32.59 59.60 -3.30 0.73 1.37 0.10 <0.1

FSJ 08 17.82 71.56 2.79 65.99 2.21 34.06 56.50 1.11 1.10 0.25 0.18

FSJ 11 25.28 79.70 2.67 64.65 3.06 32.61 56.16 -3.15 2.65 2.03 0.48 0.27




Table 13-10 Davis Tube Summary on Composites – Concentrate – Novo Amparo


Table 13-11 Davis Tube Summary on Composites – Concentrate – Novo Amparo Norte



FNA 08 34.19 84.26 1.58 66.60 2.06 31.34 60.50 3.30- 1.70 1.49 0.28 0.27

FNA 09 37.34 84.13 1.72 64.73 4.70 31.19 57.91 3.22- 1.76 2.07 0.26 0.42

FNA 10 41.96 88.04 1.56 64.51 5.15 32.12 56.54 3.24- 1.44 2.18 0.22 0.45

FNA 11 26.13 78.23 1.87 67.20 2.00 31.55 61.00 3.23- 1.21 1.18 0.21 0.20

FNA 12 27.28 73.45 1.70 63.10 4.76 30.67 56.15 3.02- 2.93 2.54 0.47 0.57

FNA 13 41.88 84.38 1.53 63.54 5.09 31.34 55.98 3.00- 1.77 2.50 0.23 0.53 FNA 14 37.01 86.57 1.73 65.64 4.03 32.21 58.07 3.33- 0.87 2.04 0.12 0.30

FNA 15 52.41 92.57 1.37 63.78 6.15 31.53 56.17 3.11- 1.34 2.78 0.12 0.59

FNA 16 48.67 88.83 1.32 63.93 5.25 30.47 57.56 3.22- 1.63 2.64 0.19 0.52

FNA 18 42.19 87.37 1.44 64.43 4.72 31.18 57.48 3.18- 1.66 2.52 0.25 0.48


FNAN 01 34.83 83.07 2.60 64.15 3.72 32.47 55.61 3.31- 2.34 2.09 0.40 0.26

FNAN 02 28.80 82.19 2.89 64.93 2.61 31.69 57.66 3.33- 2.30 1.89 0.39 0.21

FNAN 03 32.91 81.87 2.58 62.81 3.73 32.06 54.17 3.10- 3.26 2.35 0.55 0.29

FNAN 06 26.99 57.85 -

FNAN 08 34.21 80.17 2.10 61.87 4.06 31.27 53.72 2.95- 3.74 2.61 0.65 0.37 FNAN 09 17.30 65.34 2.84 59.91 2.29 30.48 51.77 2.75- 6.40 2.91 1.19 0.51

FNAN 10 32.42 78.18 2.63 63.58 3.44 31.56 55.82 2.20 2.31 0.41 0.25

FNAN 11 19.34 64.20 2.94 59.53 3.68 27.64 54.32 1.31- 5.56 3.06 1.14 0.40

FNAN 12 35.22 113.33 4.66 98.34 3.38 48.92 86.19 4.96- 2.73 2.64 0.52 0.16

FNAN 13 35.43 86.56 1.95 66.22 3.13 30.16 61.20 3.16- 1.09 1.71 0.16 0.11

FNAN 14 36.97 87.22 2.08 65.16 3.94 31.62 57.96 3.21- 1.59 1.92 0.29 0.18

FNAN 15 30.06 77.30 2.88 64.73 3.04 31.44 57.61 3.23- 1.90 1.85 0.35 0.21

FNAN 16 24.32 72.33 2.71 64.96 2.57 31.43 57.95 3.10- 2.13 1.82 0.52 0.10

FNAN 17 22.56 71.99 3.16 65.91 2.79 32.48 58.14 3.49- 1.15 1.65 0.27 0.12




Table 13-12 Davis Tube Summary on Composites – Tailings – Gulçari A





Table 13-13 Davis Tube Summary on Composites – Tailings – Gulçari A Norte


Table 13-14 Davis Tube Summary on Composites – Tailings – Gulçari B



FGAN 01 76.35 16.85 0.27 21.08 8.27 13.21 15.46 0.49- 37.11 9.69 6.49 6.39

FGAN 02 57.96 11.31 0.29 28.80 16.80 22.55 16.10 1.74- 25.40 10.10 3.83 2.73

FGAN 04 80.20 37.45 0.37 20.15 8.71 15.37 11.73 0.40- 34.70 13.12 7.21 3.62

FGAN 06 77.94 32.95 0.37 19.11 9.13 14.73 10.91 0.00- 35.60 14.21 7.34 3.37

FGAN 07 82.48 31.74 0.36 17.15 7.25 11.18 12.10 0.08- 38.65 14.40 7.77 3.57

FGAN 08 75.38 22.77 0.25 20.84 11.09 13.88 14.33 0.49- 33.99 13.24 5.64 2.79

FGAN 09 74.72 20.89 0.30 18.66 8.61 13.03 12.20 0.14 37.73 14.02 6.31 2.78

FGAN 10 84.70 0.28 20.26 9.26 14.04 13.35 0.81 36.45 12.23 6.16 2.90 FGAN 11 70.10 19.08 0.21 23.67 14.07 19.43 12.27 0.96- 30.90 11.73 5.11 2.42

FGAN 12 76.55 0.39 21.93 9.97 19.40 9.82 1.08 30.83 12.89 5.81 2.07

FGAN 13 90.15 54.40 0.30 17.38 5.35 12.42 11.10 0.11 40.82 12.37 7.81 3.84

FGAN 14 80.39 0.30 20.39 9.27 9.68 18.38 0.11 35.50 12.45 6.70 3.55

FGAN 15 78.40 22.57 0.35 21.52 9.60 11.27 18.26 0.17- 33.85 11.59 6.27 4.58

FGAN 16 69.52 13.87 0.21 25.42 14.65 14.45 20.26 0.94- 29.22 10.98 4.72 2.56

FGAN 17 72.16 11.72 0.24 20.58 11.62 12.01 16.12 0.46- 32.86 13.13 6.30 3.00

FGAN 05 72.11 18.52 0.14 23.55 13.66 11.81 20.52 1.48- 32.28 10.25 5.60 4.06


FGB 10 56.78 13.78 0.18 29.59 22.23 11.41 29.63 0.23- 21.09 7.92 2.57 3.22

FGB 11 61.58 10.49 0.16 26.90 19.53 17.84 18.67 1.49- 24.30 9.05 3.65 3.93

FGB 12 76.53 24.96 0.20 20.30 12.50 16.36 10.90 0.57- 32.80 14.00 6.01 2.21

FGB 13 55.85 8.16 0.13 28.98 23.52 18.48 20.85 2.12- 20.95 8.28 3.06 3.64

FGB 14 58.87 9.54 0.13 28.37 21.77 17.43 21.24 2.03- 22.17 8.90 3.37 3.24

FGB 15 58.86 10.49 0.14 28.86 22.11 16.18 23.29 1.74- 21.25 8.86 2.93 3.00

FGB 16 59.53 9.48 0.11 27.62 22.29 15.01 22.82 1.96- 23.37 9.08 2.89 2.90

FGB 17 62.38 12.81 0.12 27.25 19.86 15.05 22.27 1.41- 24.97 9.33 2.40 2.88




Table 13-15 Davis Tube Summary on Composites – Tailings – São José



FSJ 09 79.57 26.57 0.26 17.11 8.70 9.58 13.78 -0.07 38.25 15.84 7.14 2.39

FSJ 12 65.75 24.02 0.24 24.38 13.51 18.32 14.50 -0.60 30.51 12.11 4.89 2.01

FSJ 13 64.62 13.13 0.24 25.16 15.09 17.99 15.98 0.46 27.71 11.80 5.80 2.02

FSJ 14 63.86 27.35 0.21 26.35 16.36 20.51 14.89 -1.52 26.96 11.72 4.47 1.93

FSJ 15 64.84 9.41 0.20 26.49 17.04 16.44 19.54 -1.76 26.33 11.95 3.94 1.96

FSJ 16 64.38 4.61 0.20 26.43 16.85 16.28 19.70 -1.70 26.88 10.73 4.15 2.15

FSJ 17 81.77 26.87 0.28 17.89 8.52 8.19 16.46 -0.68 37.82 15.49 7.21 2.51

FSJ 18 63.54 12.16 0.19 24.28 16.73 16.98 15.82 -0.75 28.14 12.01 4.06 2.11

FSJ 19 86.55 41.87 0.38 17.30 6.75 11.42 12.00 0.44 38.70 15.40 6.56 2.34

FSJ 20 70.65 24.06 0.24 23.51 14.17 16.29 15.48 -1.08 30.02 12.97 5.03 1.95

FSJ 21 66.26 18.66 0.26 25.89 16.38 19.27 15.60 -1.07 26.99 11.84 4.44 2.19

FSJ 22 76.53 28.33 0.28 25.62 16.23 16.07 18.79 0.09 27.86 11.67 3.98 1.49

FSJ 23 70.78 25.98 0.33 27.42 17.25 18.17 18.97 1.99 22.84 10.14 4.57 1.91

FSJ 24 70.35 15.12 0.26 27.20 14.30 8.38 29.60 1.23 28.10 11.30 3.14 1.45

FSJ 25 74.68 11.55 0.18 20.30 11.66 13.28 14.20 -0.55 35.35 12.90 5.72 2.43

FSJ 08 82.18 28.44 0.26 17.87 7.73 11.04 13.30 -0.21 39.66 11.97 7.41 3.95

FSJ 11 82.25 13.31 0.26 18.00 7.59 9.94 14.67 -0.23 39.52 12.78 7.27 4.17




Table 13-16 Davis Tube Summary on Composites – Tailings – Novo Amparo


Table 13-17 Davis Tube Summary on Composites – Tailings – Novo Amparo Norte



FNA 08 65.81 15.74 0.16 27.40 20.40 17.38 19.80 1.23- 22.20 10.30 3.63 2.69

FNA 09 43.86 6.66 0.14 31.00 29.70 27.15 14.20 1.14- 14.00 6.55 1.19 3.38

FNA 10 58.04 11.96 0.15 26.20 21.64 21.16 13.96 0.89- 22.90 8.38 3.87 4.16

FNA 11 73.87 21.77 0.17 22.40 14.20 11.49 19.20 0.63- 32.80 10.60 4.80 3.79

FNA 12 72.72 26.55 0.19 22.34 12.78 12.64 17.84 0.84- 32.47 10.13 5.75 4.05

FNA 13 58.12 15.62 0.17 27.16 22.64 21.12 15.36 1.27- 21.54 8.34 3.68 3.89 FNA 14 62.99 13.43 0.15 26.16 20.63 19.45 15.77 1.25- 24.29 8.95 3.72 3.95

FNA 15 47.59 7.43 0.12 31.52 30.52 25.69 16.53 1.88- 13.70 6.92 1.53 3.33

FNA 16 52.19 11.62 0.08 30.30 26.10 20.79 20.20 1.77- 17.70 8.26 2.08 3.68

FNA 18 57.81 12.63 0.12 28.69 23.24 19.20 19.66 1.62- 20.96 8.81 3.48 3.50


FNAN 01 65.17 16.93 0.26 24.46 15.31 19.07 13.79 0.54- 28.19 12.22 4.90 2.23

FNAN 02 28.80 82.19 2.89 64.93 2.61 31.69 57.66 3.33- 2.30 1.89 0.39 0.21

FNAN 03 67.09 18.13 0.25 23.80 14.33 17.99 14.04 0.76- 29.57 13.22 4.83 1.85

FNAN 06 73.01 42.15 0.38 25.04 15.73 19.92 13.69 0.08- 26.51 11.17 5.11 2.33

FNAN 08 65.79 19.83 0.26 25.79 15.84 19.13 15.63 1.25- 26.68 11.75 4.86 2.08

FNAN 09 82.70 34.66 0.29 17.16 7.93 11.91 11.33 0.79 37.20 15.75 6.68 1.99 FNAN 10 67.58 21.82 0.27 24.04 14.96 16.70 15.79 0.96- 28.78 12.56 5.52 2.27

FNAN 11 84.09 37.14 0.34 17.28 7.89 8.07 15.75 0.39 38.54 15.83 6.90 2.03

FNAN 12 113.80 35.69 0.41 29.97 16.72 19.17 21.50 0.88- 50.70 22.75 9.69 2.53

FNAN 13 64.57 13.44 0.76 36.39 15.40 24.83 24.43 1.70- 18.75 9.27 3.07 1.34

FNAN 14 63.03 12.78 0.21 27.52 18.52 17.63 19.76 1.01- 23.93 11.03 4.12 2.15

FNAN 15 69.94 22.70 0.29 23.02 14.10 16.94 14.08 0.42- 29.39 12.83 5.69 2.28

FNAN 16 75.68 27.67 0.29 22.90 13.69 16.04 14.89 0.04 30.29 13.48 5.42 1.67

FNAN 17 77.44 28.01 0.27 20.60 10.03 10.87 17.36 0.42- 34.30 14.24 6.71 2.47





Table 13-18 Salt Roast Leaching with Sodium Carbonate - Overview Test Conditions and Results


Bulk leach tests were performed using bulk roasted samples to produce feed liquor for ammoniummetavanadate (AMV) precipitation. Vanadium leach extractions were 78% for the massive bulk and91.0% for the oxidized sample. Table 13-19 shows the assays on bulk concentrate, bulk leach residueand pregnant leach solution (PLS).

Table 13-19 Bulk Leach and Desilication Test Results


Roast Stages Leach

Test Sample Temp Time Na2CO3 Alumina Temp Time Solids Recovery Final Res Final PLS, mg/L

ºC h dosage, % dosage, % ºC h wt% %V %V V SiL1A 60 2 19.6 95.5 0.10 4500 3.0

L1B 95 2 19.4 95.7 0.09 4200 7.2

L2A 60 3 20.6 95.0 0.10 4200 2.7

L2B 95 3 20.7 95.5 0.09 4100 3.5L3A 25 3 22.1 93.7 0.13 4800 2.2

L3B 60 3 22.1 93.9 0.13 4900 4.4

L4 massive PGM 7-Aug 1200 1 4.0 0.7 60 2 49.0 95.2 0.10 15000 4.9

L5 massive PGM 8-Aug 1250 1 5.4 Na2SO4 1.3 60 2 46.1 95.0 0.11 15000 5.7

L6 massive PGM 9-Aug 1200 2 4.0 1.3 60 2 46.7 94.0 0.12 13000 6.3L7 massive PGM 20-Aug 1200 2 4.0 0 60 2 45.8 96.0 0.09 14000 10.0

L8 massive confirmatory 13-Sep 1200 2 4.0 1.3 60 2 47.0 92.8 0.14 12100 6.6


L10 massive confirmatory 18-Sep 1200 2 4.0 1.3 60 2 52.2 92.4 0.15 16500 5.8L11 oxidized confirmatory 19-Sep 1200 2 4.0 1.3 60 2 47.6 93.3 0.14 15000 6.1


L13 massive confirmatory 2-Oct 1200 1 4.5 1.3 60 1 49.5 93.2 0.13 13000 7.6

L14 oxidized confirmatory 11-Oct 1200 2 4.5 1.3 60 1 50.5 94.8 0.10 14500 2.7L15 massive bulk 10-Oct 1250 1 4.5 1.3 60 1 51.5 93.5 0.12 13800 7.7

L16 massive bulk 15-Oct 1150 1 4.5 1.3 60 1 50.4 91.8 0.15 12600 5.2

L17 oxidized confirmatory 16-Oct 1150 1 4.5 1.3 60 1 48.7 93.7 0.13 14700 7.2L18 oxidized confirmatory 17-Oct 1250 1 4.5 1.3 60 1 51.2 95.0 0.10 15700 9.4

2.0

2

1250 2 4.025-Jul

massive PGM 26-Jul 1250

Date2007

31-Julmassive PGM 1.34.011200

4.0 1.3

massive PGM

Units V Si Al Fe Mg Ca Na K Ti P Mn Cr

Bulk Massive Magnetite

Concentrate% 1.76 1.51 1.13 60.3 1.11 0.29 0.02 0.02 4.41 <0.00 0.15 0.01

Bulk Oxidized MagnetiteConcentrate

% 1.95 0.55 0.98 60.6 0.43 0.07 0.01 <0.01 4.89 <0.00 0.13 0.01

Bulk Massive LeachResidue

% 0.40 0.73 1.50 64.0 0.41 0.09 0.58 0.02 3.20 0.00 0.11 0.00

Bulk Oxidized Leach

Residue% 0.21 0.72 1.40 72.0 0.28 0.04 0.55 0.02 3.80 0.00 0.11 0.00

Bulk Massive PLS mg/L 12500 199 1.00 4.5 5.68 113 8790 15.00 3.40 5.00 12Bulk Massive PLS after deSi

mg/L 13300 45.7

Bulk Oxidized PLS mg/L 27500 59.9 5.70 10.7 15.10 372 17800 16.00 15.10 5.00 14.30Bulk Oxidized PLS afterde-Si

mg/L 28000 26.8

Bulk Massive LeachExtraction % 78.6 (Fe Tie)

Bulk Oxidized LeachExtraction

% 91.0 (Fe Tie)





A desilication step was applied after bulk leaching by the addition of aluminum sulphate. The filtrate fromthe massive sample leach and desilication was used in the oxide sample leach for the purpose of highvanadium concentration in the solution.

The final filtrate from the desilication step, after the oxide leach, contains 28 g/L V and 26.8 mg/L Si.

The following contributing factors are presented as possible explanation(s) for the lower than expectedvanadium extraction:

Different furnaces were used for the bulk roasting; Temperature variation differences from the testing done by SGS and FEECO; Indirect (SGS) versus direct flame heating in bulk roasting; The feed to the bulk leach was a blend of various batch roasting samples, produced under various

conditions.

Vanadium precipitation efficiency from the combined bulk leach liquor was calculated to be 99.9%.However, during washing of the AMV solids, some AMV redissolved. A final product, assaying 41.6% Vand only 0.09% Si, was produced. Table 13-20 shows precipitation recovery and AMV assay. Furtherwork is required to investigate methods to minimize soluble vanadium losses.

Table 13-20 AMV Precipitation Recovery and Analysis (wt%)


13.3.5 AMV Calcination

The AMV calcination tests were performed by SGS. Four temperature conditions were tested. At the twohigher temperatures of 450ºC and 500ºC, conversion to V2O5 was completed by the time the first samplewas withdrawn at 2 h. There was no significant change from the 2 h results in the 4-, 6-, 12- or 24-hsamples.

For the 400ºC calcination test, nitrogen concentration had decreased below detection limits at 2 h, but thevanadium concentration was only 93% (V2O5), while the weight loss was approximately 18%. While therewas some variability from sample to sample in this test, there did appear to be some further weight lossafter 2 h; this was corroborated by a further increase in percent V2O5. The results are presented in Table13-21.

Table 13-21 AMV Calcination Results

¹ Average of remaining samples, 4 to 24 h.

V

Recovery V Na Si Al Ca Fe Mg Ti

69 41.6 0.17 0.09 0.04 0.007 0.04 0.03 0.02

Bulk AMV Product Analysis

Calcining Temperature, ºC 400 400 450 500

Time 2 h Average¹ Average Average

Weight Loss, wt% 18 20 21 22

V2O5 in Calcine, wt% 93 95 97 99





AMV calcination tests showed that it was possible to calcine AMV to V2O5 in less than 2 h attemperatures above 450°C, but longer calcination times are generally used in industry depending on the

AMV bed thickness.

Testwork of converting AMV to V2O5 was also performed by Mintek, South Africa. The assay of V2O5 produced is shown in Table 13-22.

Table 13-22 Analysis of V2O5

Based on the data presented in Table 13-22, if this V2O5 material is used as feed to ferrovanadiumproduction, the ASTM and ISO specifications for an 80% V ferroalloy may be met, provided that the levelsof impurity elements in the other raw materials are sufficiently low.

The testwork to date concentrated on producing V2O5 which is the conventional route for producingferrovanadium. No testwork was performed to simulate V2O3 production.

The feasibility study, however, was based on producing V2O3 as an intermediate product which wouldhave been the feed for the production of ferrovanadium. The V2O3 route can save significantly onconsumption of aluminum and other materials that are used in the ferrovanadium production, whichmeans lower operating costs. The production of V2O3 from AMV is achieved through the use of relativelynew technology, and no detailed information was available at the time of testwork. Technology andequipment required for producing V2O3 from AMV could be available from Drytech in South Africa.However, subsequently, Largo has decided to produce V2O5 and ferrovanadium as primary products.

Accordingly, there are no current plans to produce V2O3 as an intermediate product.

13.3.6 Ferrovanadium Test

No testwork has been performed.

13.4 PGM Testwork

A Platinum Group Metals (“PGM”) scoping study program was undertaken by SGS on assay reject

samples from the beneficiation testwork program in order to investigate the potential for PGM recovery. Ahead sample was taken which assayed 0.74 g/t platinum and 0.19 g/t palladium. A series of tests werecarried out which included, flotation, magnetic separation, gravity, hot cyanide leach and platsol.

A series of six flotation tests were performed in order to examine rougher recovery. Several parameterswere investigated which included residence time, reagent additions and grind size. From these testsoptimum conditions were selected and two cleaning tests were performed. Flotation rougher recoveryranged from 51% to 70% platinum recovery and 51% to 65% palladium recovery. As a result of two-stage cleaning the platinum recovery achieved was 26% with a concentrate grade of 52 g/t platinum anda palladium recovery of 17% recovery resulting in a concentrate grade of 9 g/t.

Temp. Duration

Test by °C hrs V2O5 V Si Al2O3 Fe CaO K Cr NA P K2O + Na2O

SGS 450 4 98.40 55.12 <0.03 1.75 0.04 <0.20 <0.17 0.01 0.17 <0.01 0.42

SGS 500 4 99.50 55.73 <0.03 1.77 0.04 <0.20 <0.17 0.01 0.17 <0.01 0.43

Mintek 400 16 97.30 54.50 0.04 0.02 0.00 0.00 0.00 0.00

Assay, wt%





It must be noted that one of the major issues with the cleaning test wok was in the size of the sample.Due to the low amount of sulphides in the ore a rough mass pull of only 3% was achieved, insufficient tosustain proper froth stability. The results obtained are therefore unreliable.

In order to properly evaluate cleaning kinetics, a much larger sample is required. This can be achieved byeither a pilot trial run or to a lesser degree, six or eight cycle, lock cycle tests using a 10 kg charge for

each cycle. This would have a twofold effect, one it would enable a buildup of material in each cleaningstage and two it would become a closed circuit test with re-circulation of many tailing streams. Bothfactors would lead to an improvement in PGM recovery.

Other tests examined included magnetic separation, which was carried out on the whole ore. It wasdetermined that over 80% of the PGM’s were associated with the non -magnetics, thus leading to lowrecoveries. Also gravity separation was performed on finely ground non-magnetic material. Recoveryrates for both platinum and palladium were very low, with a platinum recovery at 2.5% resulting in aconcentrate grade of 40 g/t, while palladium recovery was 0.3% with a concentrate grade of 1.7 g/t.

A series of hot cyanidation tests were performed on the whole ore. Recoveries for both platinum andpalladium ranged from 8% and 78 % respectively. As a result of the low platinum recoveries it wasdecided to investigate Platsol. Whole ore samples were once again used as not enough flotationconcentrate could be generated. Platinum and palladium recoveries in excess of 80% were achieved.

Due to the limited amount of testwork that was performed on the ore and the early encouraging resultsfrom Platsol process testing it is recommended that further development be done for PGM recovery withspecial attention to reagent conditions, grind size and flotation times. The results indicate that flotation,followed by Platsol, needs to be re-examined in greater detail.

13.5 Pilot Plant Testing (by Largo)

After completion of the Definitive Feasibility Study (“DFS”) in 2010, at the request of the financing bank’s

technical consultant, a pilot scale program was initiated to prove the viability of producing vanadium fromMaracás ore and to confirm the process data reported in the feasibility study. The testwork was done atFundação Gorceix and involved obtaining a sample of the Maracás ore, beneficiating the ore to produce avanadium concentrate and then roasting the concentrate in a k iln to convert vanadium into a soluble form.

The roasted concentrate was then leached in water to produce a vanadium solution that was furtherprocessed through desilication and AMV precipitation steps. The AMV thus produced was then analyzedand calcined at SGS to produce V2O5. The complete process route has been described in the DFS.

It was not possible with available facilities to pilot the production of V2O3 and Ferrovanadium from AMV.Since these are state of the art technologies utilized by major FeV producers their exclusion from the pilotprogram was considered acceptable as long as the AMV produced is of acceptable quality.

Largo prepared three composite samples representing:

massive high-grade mineralization – 800 kg from drill core; massive high-grade mineralization – 20 tonnes from outcrop; disseminated low-grade mineralization.





Past testwork determined that the surface ore grind should be relatively coarse to minimize magneticlosses so it was decided to adopt a grind mesh of 150 microns instead of the 100 microns that was usedin the DFS design criteria.

13.5.1 Beneficiation of Ore Samples at Fundacao Gorceix

Approximately half of the total sample was fed into the beneficiation equipment which consisted of a ballmill, in closed circuit with a classifier, in which the fine fraction produced was fed onto a 1500 Gauss wetmagnetic separator. The magnetic product from this separator was then fed onto an 800 Gauss magneticseparator as a cleaning stage.

The concentrate produced was 86% -150 micron in size, indicating that in the actual plant operation wewould have lower energy consumption and smaller equipment as compared with the DFS design.

The results indicated that the quality of the 1500G magnetic product was sufficient without the need of asecond stage 800G separator. Both reject and magnetic product from the 800G separator were thereforepartially dried to around 3% moisture, and kept separate for re-combination before roasting. The materialswere drummed and transported to Phoster in Belo Horizonte.

13.5.2 Davis Tube Testing of Low and High Grade Core Samples

Representative samples of high grade ore (800 kg) and lower grade disseminated ore (600 kg) werecollected from cores at Maracás and shipped to Fundacao Gorceix, for sizing to minus150 microns.Samples of the milled products, plus a blend of the two products in the ratio 57% high grade and 43% lowgrade, were prepared and sent to SGS for Davis Tube magnetic separation.

The concentrate produced during pilot plant testing indicated comparable quality and V yield to theconcentrate produced at SGS from the high grade/low grade blend. Low grade ore on its own producesan acceptable concentrate quality, but the V yield is significantly lower than that obtained on high grade or

blended ores.

Table 13-23 Davis Tube Testing Results

(*) [The V yield of the blend is just a mathematical balance of the two yield from individual composites when taking

their ratios into account ]

13.5.3 Kiln Roasting

The 800G concentrate and the 800G “reject” were carefully mixed along with 29 kg of Sodium carbonate

which was added to each 500 kg batch of the mixed concentrate, equivalent to 5.48% Na2CO3 and thenwell mixed with the concentrate. The kiln was heated up over a period of 6 hours to give a refractorytemperature at the discharge end of 1000º C.

V in Ore V in Concentrate SiO2 in Concentrate V yield to Concentrate

% % % %

Pilot plant ore sample 1.51 1.92 0.8 90.9

Low grade core sample 0.41 1.81 2.23 68.3(*)

High grade core samples 1.14 2.01 0.96 96.5

Blend (low grade+high grade) 0.82 1.98 1.09 90.3(*)





The lower temperatures (1150C) utilized during the trial run (below design conditions) affected vanadiumrecoveries that were lower than anticipated for a commercial operation. The calcined material exiting thekiln contained lumps that disintegrated very easily but no hard “fused” material was evident. Unlike some

other operations the kiln runs did not experience any surface build up possibly due to the low levels ofSiO2.

13.5.4 Leaching

The results of the leaching tests are shown as follows:

Table 13-24 Leaching Tests Results

Other testwork was completed on the de-silication, precipitation and calcination aspects of the process.

The general conclusion reported by Largo was that recoveries achieved on the pilot plant are lower thanthe ones reported in the DFS. In particular the kiln recovery in the DFS is reported at 95% whereas in thepilot plant the recovery was 85%. The small kiln used in the pilot tests was not optimized for achievinghigh vanadium recoveries indicating that greater than 85% values could be obtained with actualcommercial size equipment.

De-silication and precipitation recoveries achieved should also be expected in actual plant operation.Reagent consumptions experienced in most areas were lower than the ones adopted for the DFS.Flocculent and sodium hydroxide may not be required.

It is noted that the pilot plant report used as basis for this subchapter 13.5 includes summary results ofthe unit operations as analyzed, interpreted (some calculated) and reported by Largo. No independentanalysis was made of the “raw” data produced from the program.

Sample Mass Ratio

ID No to concentrate V % V2O5 % Na2O %

Concentrate to kiln Average 1.000 1.94 3.46 -

Kiln Feed inc. Na2CO3 calculated 1.050 1.93 3.44 3.04

Kiln Discharge due to loss of CO2 calculated 1.030 1.97 3.52 3.11 Leached Residue Sample1 Washed Res1 0.975 0.36 0.65 1.34

Leached Residue Sample 2 Washed Res2 0.975 0.32 0.58 1.39

Leached Residue Sample 3 LMAR00289 0.975 0.35 0.62 n/a

Leached Residue Sample 4 290-298 0.975 0.32 0.57 n/a

Average Leached residue 0.975 0.34 0.61 1.37

% into Solution 3.210 83.00 83.00 55.00

Kiln recovery analyzed on site 85.04 85.04

Leach Recovery (calculated) 97.60 97.60

Overall Kiln leach recovery 83.00 83.00





14 Mineral Resource Estimates

In Item 14, Mineral Resource Estimates, both Micon and Coffey Mining have provided individual sections.

The Micon data has been designated with a letter “A” and is primarily concerned with the preparation ofthe Gulçari A deposit mineral resources.

The Coffey Mining data has been designated with the letter “B” and is primarily concerned with the

preparation of the satellite deposits mineral resources.

All block modeling, three-dimensional (3D) solids interpretation and grade interpolation for the resourceestimates for Gulçari A presented herein were carried out by Farshid Ghazanfari using Gemcom® software (Version 6.0 in 2007 and 6.2 in 2012). The statistical and geostatistical analyses wereperformed using Snowden Supervisor software (Version 7). Scanning software and AutoCAD were usedfor recovery and digitizing of old section and plan view geological maps from previous operators. Theprocedures used and the resulting mineral resource estimate were reviewed by Micon.

The first mineral resource estimate for Gulçari A presented in this report was estimated in 2007. Resultsfrom the 2008 and 2011-2012 drilling described in Section 11 were not available for that estimate. The2008 to 2012 drill results were used to update the mineral resource estimate in September, 2012,however, the 2007 mineral resource estimate was used for the Technical Update, also presented in thisreport (which was prepared over the spring and summer of 2012). For this reason both estimates arediscussed and presented herein.

14.1.A Database

For the original 2006 resource estimate, a Gemcom® database for the Gulçari A vanadium deposit hadbeen created from hard copies of the old drill logs and assay certificates (see Hennessey, 2006). All the

original coordinates from previous operator’s programs were changed from local grid to UTM grid. Thecollars for all drillholes and trenches were also converted to UTM coordinates.

The 2007 database contains complete assays for V2O5 and TiO2 as well as more extensive results forPGMs than were available for the 2006 estimate. The TiO2 assay results are significant, often greatlyexceeding 10%. However, it was not clear at the time of estimation that the titanium would berecoverable from the process proposed by earlier operators and, therefore, potentially economic. For thisreason, a titanium resource was not estimated.

A total of 97 drillholes and 21 trenches were manually entered into Excel spreadsheets then imported intothe Gemcom® database. The database was validated with Gemcom’s internal validation tools and double

checked with hard copies of original data. The drill-hole workspace in the Gemcom® database includes

different tables for the drill-hole and trench data. In early 2012, 13 additional drillholes (FGA-100 to FGA-112) were drilled and the Gemcom database updated.

Down-hole survey data were available only for longer holes (more than 300 m in length). This is notconsidered to be a serious problem, as the other holes are generally quite short. Drillhole collar surveyswere available and several of their surveys had recently been checked by Largo. More complete surveydata were available, in intervals, for the trenches.





All but 10 of the drillholes and one of the trenches from the Gulçari A property have been used in the tworesource estimates. DrillHoles FGA-82 and FGA-101 and Trench TGA-34 were not used as they arelocated outside of the mineralized area and the geological domain model. Drillholes FGA-41 and FGA-55were excluded from database, due to a discrepancy in historical assay results (see Section 12 above).Drillholes FGA-45, FGA-46 and FGA-47 were not used since there is no complete set of assay dataavailable. Drillholes FGA-81, FGA-84 and FGA-85 were drilled for metallurgical testwork and have not

been sampled.

14.2.A Geological Solids and Domain Interpretation

Original paper maps, including several cross-sections and one geological map from CAEMI (ca. 1980s),were scanned and digitized, had their coordinates corrected and were imported into the AutoCADsoftware package. The local mine grid was transformed to UTM grid coordinates. This local resourcegrid was used as a base to determine the orientation and boundaries of the deposit for the new blockmodel estimation. This local resource grid was rotated 20° clockwise, in order to be perpendicular to thestrike of deposit.

A new set of cross-sections was constructed in Gemcom every 25 m, as appropriate for the drill-holespacing. These cross-sections are perpendicular to the strike of the deposit and parallel to the rotatedlocal resource grid to avoid any rotation of the block model. A new set of plan views was also constructedevery 10 m between the 100- and 320-m elevations (elevations used are true, in metres above sea level).

The geological interpretation details shown on the scanned maps were checked and polylines weredrawn in AutoCAD following them. All of the drawings in AutoCAD were rotated 20° clockwise to matchthe sections in Gemcom. A midpoint in the 20º-rotated cross-sections in Gemcom was considered asreference anchor point and all of the lines in AutoCAD were rotated to match this point.

Diamond drillholes which were on, or very close to, a section line have been resurveyed by recent GPSsurveys. The surface map elevations were all adjusted by adding 7.00 m. The original elevations were

all surveyed, based on a known survey control point which was entered as being 7.00 m lower than theGPS-surveyed elevations.

The AutoCAD drawings were imported into Gemcom and used as a reference for geological domaininterpretation and construction. Different lithological domains were selected, based on the “Litho” table in

the Gemcom database, the scanned maps and sections, and their imported cross-sections into Gemcom.

Eight rock type codes were needed for the Maracás project (see Table 14-1).

From these, five major geological domains have been identified for modelling of the Gulçari A deposit.These five domains can be identified easily in drill logs and core boxes and also on the sections. Therock types of the layered mafic intrusive control the mineralization at Gulçari A. The geological rock types

were therefore used to control the mineralized domain solids.





Table 14-1 Rock Type and Mineral Domain Codes at Gulçari A

Geological and mineralized domains were constructed on vertical sections and then checked forconsistency in plan view.

The geological domains for magnetite, magnetite-pyroxenite and pegmatite were entirely contained withinthe modelled deposit on all sections. The gabbro and pyroxenite were only partially modelled, as theselithologies extend well beyond the deposit boundaries.

The magnetite and magnetite-pyroxenite mineral domains compose more than 98% of the volume of thedeposit. The remaining contained pegmatite and gabbro are internal waste materials. Since the Gulçari

A deposit is cut by volumetrically significant pegmatitic dikes, modeling of them was considered essential(seeFigure 14-1 Figure 14-1). Due to their size, they are considered to be separable during mining andare not included as internal dilution in the mineral resource.

Figure 14-1 Gulçari A Deposit Geological Domain Solids - Looking North

Rock Type Rock Code Used as Domain?

Magnetite Code 1 Yes

Magnetite-pyroxenite Code 2 Yes

Melano-gabbro Code 3

Pegmatite Code 4 YesGabbro Code 5 Yes

Pyroxenite Code 7 No

Soil and weathered rocks Code 8





Table 14-3 Average Specific Gravity for the Gulçari A Deposit, Largo Data

14.5.A Population Statistics

The univariate statistical analysis was performed using the Snowden software, V2O5 values (in weight %)and PGM values (in ppb) which were extracted from the Gemcom database for each domain. Bothunivariate statistics and variogram analysis were completed for the drill-hole and trench data combinedtogether as one dataset. The lognormal histograms and log probability plots for V2O5 and PGM assays inthe magnetite and magnetite-pyroxenite domains are shown in Figure 14-2 through Figure 14-13, respectively. (Figures shown are for the 2007 data. Similar results were returned in 2012.). The

univariate statistical data for the raw assays and the log histograms are shown in Table 14-4, throughTable 14-7. Table 14-4 shows the 2007 V2O5 results and Table 14-5 shows the 2012 results.

The histogram of V2O5 results for the magnetite domain shows that 50% of the vanadium assay valuesare above 2.06% (2.04% in 2012). The geometric mean for the magnetite-pyroxenite domain is about0.6% and 50% of the assays are above this value (see Table 14-5 to Table 14-7). These values arerepresentative of the mean for the uncut and non-composited assay data in the magnetite and magnetite-pyroxenite domains.

Figure 14-2 Magnetite Domain, Lognormal Histogram, Raw Assays - V2O5

Rock Type Rock Code Average SG

Massive and banded magnetite 1 4.33

Magnetite-pyroxenite 2 3.63

Pyroxenite 7 3.50

Gabbro 5 2.95Pegmatite 4 2.53





Figure 14-3 Magnetite Domain, Log Probability Plot, Raw Assays - V2O5

Figure 14-4 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays - V2O5





Figure 14-5 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays - V2O5

Figure 14-6 Magnetite Domain, Lognormal Histogram, Raw Assays – Pt





Figure 14-7 Magnetite Domain, Log Probability Plot, Raw Assays – Pt

Figure 14-8 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays – Pt





Figure 14-9 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays – Pt

Figure 14-10 Magnetite Domain, Lognormal Histogram, Raw Assays – Pd





Figure 14-11 Magnetite Domain, Log Probability Plot, Raw Assays – Pd

Figure 14-12 Magnetite-Pyroxenite Domain, Lognormal Histogram, Raw Assays – Pd





Figure 14-13 Magnetite-Pyroxenite Domain, Log Probability Plot, Raw Assays – Pd

Table 14-4 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw V2O5 Data

2007 Data (%)

Lognormal Histogram DataMagnetite-

Pyroxenite Domain

Magnetite

Domain

Total Number of Samples 2,365 1757

Minimum Histogram Value 0.01 0.01Maximum Histogram Value 4.40 6.78

Minimum Population Data 0.01 0.01

Maximum Population Data 4.40 6.78

75% Population 1.01 2.48



Mean 0.784 2.007

Geometric Mean 0.598 1.821

LOG-Est. Mean 0.869 2.082

Standard Deviation 0.516 0.759Variance 0.266 0.576

Skewness 1.208 0.383

Coefficient of Variation 0.658 0.378





Table 14-5 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw V2O5 Data

2012 Data (%)

Table 14-6 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw Pt Data

2007 Data (ppb)


Pyroxenite Domain

Magnetite

Domain

Total Number of Samples 3,581 1,830

Minimum Histogram Value 0.01 0.01Maximum Histogram Value 4.40 6.78


Maximum Population Data 4.40 6.78




Mean 0.737 1.969




Skewness 1.467 0.25



Pyroxenite DomainMagnetite Domain


Minimum Histogram Value 5 5

Maximum Histogram Value 4998 5026Minimum Population Data 5 5

Maximum Population Data 4998 5026

75% Population 161 450



Mean 137.2 339


LOG-Est. Mean 155.4 398

Standard Deviation 233.902 328.109

Variance 54331.82 107655.33Skewness 7.727 4.376






Table 14-7 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Raw Pd Data

2007 Data (ppb)

14.5.1.A Grade Capping

Log probability plots can be used to determine grade capping (or top cutting) values for assays in eachdomain. The graphs in Figure 14-2 through Figure 14-13 were analysed for consistent lognormallydistributed populations and the point at which those populations broke down. Lognormal populationsform straight lines on probability plots and the point at which data can be considered to be outliers canusually be readily determined from them.

Top cuts were selected by examining the histograms and probability plots for the grade at which outliersbegin to occur. The 1,757 (1,830 in 2012) assays available for the magnetite domain and 2,635 (3,581 in2012) assays for the magnetite-pyroxenite domain are an adequate number for top cut analysis. In 2007a top cut of 3.80% V2O5 was chosen for the magnetite domain based on the log probability plot. Thisvalue was used to cut all assay data greater than 3.80% V2O5 to 3.80% in the drillhole and trench assaytables. Top cuts for PGMs in the magnetite domain were 1,000 ppb and 550 ppb for platinum andpalladium, respectively, based on log probability plots (see Figure 14-7 and Figure 14-9). The samenumbers were used in 2012.

A top cut of 2.0% V2O5 was chosen for the magnetite-pyroxenite domain based its log probability plot.

Top cuts for PGMs in magnetite-pyroxenite domain were 620 ppb and 350 ppb for platinum and palladiumrespectively (see Figure 14-11 and Figure 14-13).

14.5.2.A Compositing

Grade interpolation methods such as kriging or inverse distance weighting can weight average assays bydistance away from, and clustering around, a block being interpolated; but they cannot weight by sample



Total Number of Samples 1 643 1873

Minimum Histogram Value 5 5Maximum Histogram Value 705 1106

Minimum Population Data 5 5





Mean 70.6 143.2




Variance 8 356.74 15 803.34







length. All informing sample values are assumed to be point values in space. For this reason, it isnecessary to composite assay results to a common length before grade interpolation.

Before compositing of the assays in the drillholes and trenches, a true thickness was calculated for eachhole and trench based on the average inclination of magnetite and magnetite-pyroxenite layers in eachsection. True thickness was then entered in the assay table to be used in compositing as a weighting

field. This an option in the compositing profile of the Gemcom software package. (This option wasemployed in 2007 but not in 2012.)

An assay composite table was created starting at the drill-hole collar using all cut assay data from thegrade capping. All drillhole and trench assay data were composited in 5-m intervals. The updateddomain intersections and results for % or percent V2O5 were saved in a different table with their rockcodes. No lower cut-off grade was used. These data were then extracted in point format withcoordinates for each domain and used for geostatistical interpretation, as well as for resource estimation.

A composite length of 5 m was chosen so that the internal dilution and variance would be the same asthat for the selected SMU and block size.

If there were no assays for a sample interval, or missing assay intervals in drillholes, the backgroundvalue of 0.15% V2O5, which is the median value for gabbro domain, was used. In the case of trenches,due to the frequent intervals of no sampling and resulting large intervals with few assays, no backgroundvalue was selected for compositing. Therefore, all missing samples there were treated as null (not zero)values. The same approach was applied to the PGM data, since there are many missing intervals forPGMs within the model.

After compositing, all intervals less than 1 m in length (0.5 m in 2012) were deleted from the database,except those intervals in the boundary of magnetite and magnetite-pyroxenite. This was done to preventshort sample bias. The magnetite and magnetite-pyroxenite boundary is a transitional boundary. Theshort intervals were kept, because there are many medium- to high-grade values associated with this

boundary.

Figure 14-14 through Figure 14-25 show lognormal histograms and log probability plots for the V2O5 andPGM results in the magnetite and magnetite-pyroxenite domains from the drillhole and trench compositedassay dataset. Table 14-8 through Table 14-11 show the univariate statistical data for the V2O5 and PGMassay composites of all drillhole and trench data in the magnetite and magnetite-pyroxenite domains. Thedata shown in these figures are from 2007. Similar results were obtained in 2012. V2O5 data in Table14-8 and Table 14-9 are from 2007 and 2012 respectively.





Figure 14-14 Magnetite Domain, Lognormal Histogram, Composite Data Set - V2O5

Figure 14-15 Magnetite Domain, Log Probability Plot, Composite Data Set - V2O5





Figure 14-16 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set - V2O5

Figure 14-17 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set - V2O5





Figure 14-18 Magnetite Domain, Lognormal Histogram, Composite Data Set – Pt

Figure 14-19 Magnetite Domain, Log Probability Plot, Composite Data Set – Pt





Figure 14-20 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set – Pt

Figure 14-21 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set – Pt





Figure 14-22 Magnetite Domain, Lognormal Histogram, Composite Data Set – Pd

Figure 14-23 Magnetite Domain, Log Probability Plot, Composite Data Set – Pd





Figure 14-24 Magnetite-Pyroxenite Domain, Lognormal Histogram, Composite Data Set – Pd

Figure 14-25 Magnetite-Pyroxenite Domain, Log Probability Plot, Composite Data Set – Pd





Table 14-8 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, V2O5 Composites

2007 Data (%)

Table 14-9 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, V2O5 Composites

2012 Data (%)



Total Number of Samples 799 484

Minimum Histogram Value 0.02 0.23Maximum Histogram Value 2 3.08


Maximum Population Data 2 3.08




Mean 0.824 2.005




Variance 0.161 0.43






Minimum Histogram Value 0.02 0.01

Maximum Histogram Value 2 3.8Minimum Population Data 0.02 0.01

Maximum Population Data 2 3.8




Mean 0.763 1.978




Variance 0.158 0.462

Skewness 0.949 -0.044






Table 14-10 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Pt Composites

2007 Data (ppb)

Table 14-11 Statistical Data for Magnetite and Magnetite-Pyroxenite Domains, Pd Composites

2007 Data (ppb)

14.6.A Experimental Variograms

Experimental semivariograms (hereafter called variograms) were prepared for the magnetite andmagnetite-pyroxenite domains using the composited assay dataset for V2O5 (see above), platinum and




Minimum Histogram Value 1 5Maximum Histogram Value 620 1000

Minimum Population Data 1 5





Mean 129.3 324.9









Minimum Histogram Value 4.8 5

Maximum Histogram Value 350 453

Minimum Population Data 4.8 5





Mean 67.3 138.4




Variance 4845.37 10077.17Skewness 1.906 0.787






palladium. Variograms were constructed by applying an average strike (N 20° E) of the deposit and thetwo other mutually orthogonal directions. Single- and double-structure spherical models were used to fitexperimental directional variograms that are illustrated in Figure 14-26 through Figure 14-49. Thevariograms are traditional and are log transformed. A lag distance of 10 m and a maximum separation of200 m between samples were selected to create the variograms.

Down-hole variograms were used instead of omnidirectional variograms to model the nugget (y-intercept)for each domain (see Figure 14-26 and Figure 14-30 for V2O5 and Figure 14-34, Figure 14-38, Figure14-42 and Figure 14-46 for PGM). The angular tolerance for down-hole variograms was 90° to allow allcomposite samples to fall on the variograms. The angular tolerance used for the experimental variogramcalculations was ±20º (±10º in 2012).

Based on these variograms, three directions of continuity (major, intermediate and minor) weredetermined for the magnetite and magnetite-pyroxenite domains (see Figure 14-27 to Figure 14-29, Figure 14-31 to Figure 14-33, Figure 14-35 to Figure 14-37, Figure 14-39 to Figure 14-41, Figure 14-43 toFigure 14-45 and Figure 14-47 to Figure 14-49). The figures shown are for the 2007 data. Similar resultswere obtained in 2012.

Figure 14-26 Down-hole Variogram - V2O5 in Magnetite Domain





Figure 14-27 Variogram - Direction of Intermediate Continuity - V2O5 in Magnetite Domain

Figure 14-28 Variogram - Direction of Maximum Continuity - V2O5 in Magnetite Domain





Figure 14-29 Variogram - Direction of Minimum Continuity - V2O5 in Magnetite Domain

Figure 14-30 Down-hole Variograms - V2O5 in Magnetite-Pyroxenite Domain





Figure 14-31 Variogram - Direction of Intermediate Continuity - V2O5 in Magnetite-Pyroxenite Domain

Figure 14-32 Variogram - Direction of Maximum Continuity - V2O5 in Magnetite-Pyroxenite Domain





Figure 14-33 Variogram - Direction of Minimum Continuity - V2O5 in Magnetite-Pyroxenite Domain

Figure 14-34 Down-hole Variogram - Pt in Magnetite Domain





Figure 14-35 Variogram - Direction of Intermediate Continuity - Pt in Magnetite Domain

Figure 14-36 Variogram - Direction of Maximum Continuity - Pt in Magnetite Domain





Figure 14-37 Variogram - Direction of Minimum Continuity - Pt in Magnetite Domain

Figure 14-38 Down-hole Variograms - Pt in Magnetite-Pyroxenite Domain





Figure 14-39 Variogram - Direction of Intermediate Continuity - Pt in Magnetite-Pyroxenite Domain

Figure 14-40 Variogram - Direction of Maximum Continuity - Pt in Magnetite-Pyroxenite Domain





Figure 14-41 Variogram - Direction of Minimum Continuity - Pt in Magnetite-Pyroxenite Domain

Figure 14-42 Down-hole Variogram - Pd in Magnetite Domain





Figure 14-43 Variogram - Direction of Intermediate Continuity - Pd in Magnetite Domain

Figure 14-44 Variogram - Direction of Maximum Continuity - Pd in Magnetite Domain





Figure 14-45 Variogram - Direction of Minimum of Continuity - Pd in Magnetite Domain

Figure 14-46 Down-hole Variograms - Pd in Magnetite-Pyroxenite Domain





Figure 14-47 Variogram - Direction of Intermediate Continuity - Pd in Magnetite-Pyroxenite Domain

Figure 14-48 Variogram - Direction of Maximum Continuity - Pd in Magnetite-Pyroxenite Domain





Figure 14-49 Variogram - Direction of Minimum Continuity - Pd in Magnetite-Pyroxenite Domain

In the Supervisor software, naming convention Directions 1, 2 and 3 are defined as follows:

1, is the direction of intermediate continuity along the dip of the magnetite and magnetite-pyroxenite layers (Y);

2, is the direction of minimum continuity across the dip of the magnetite and magnetite-pyroxenitelayers (X);

3, is the direction of maximum continuity along the strike of the magnetite and magnetite-pyroxenitelayers (Z).

Gemcom ZYZ angles were extracted from the positive axis rotation (around Y) to apply to the searchellipsoid for interpolating grade. The Gemcom ZYZ angles were 195º, -14º, and 252º for the principalazimuth, principal dip and intermediate azimuth, respectively, in the magnetite domain. The GemcomZYZ angles were 200º, 0º, and 290º for the principal azimuth, principal dip and intermediate azimuth,respectively, in the magnetite-pyroxenite domain. For 2012 the ZYZ angles were 160º, -55º, and -20º inboth magnetite and magnetite pyroxenite domains.

For V2O5, the search volume limits along the anisotropy X, Y and Z directions were set at ranges of 100,100 and 60 m, respectively, for the magnetite domain (115, 120 and 30 m in 2012) and 100, 100 and 55

m for the magnetite-pyroxenite domain (130, 160 and 30 m in 2012). These search radii were determinedby the maximum range of the variograms (in order to interpolate the blocks as Indicated resources), butcan be extended empirically if needed for Inferred resources. For the 2012 resource update, where thereare inferred resources, this was done. Given the 25 m-spaced drill sections at Gulçari A, it was felt thatthe search radius should be more than large enough that maximum numbers of samples to krig are foundfor most blocks. All blocks in the 2007 3D model for the magnetite and magnetite-pyroxenite domainswere filled using the maximum range of variograms. As a result, no Inferred resources were estimated.For the 2012 update the remaining blocks were filled with a search of two times the variogram range and





classified as inferred. In order to estimate Measured resources, the range at half of the gamma value(variance) of the variogram was selected as the search distance in 2007. This was relaxed to two thirdsin 2012.

14.7.A Kriging and Resource Estimation

The resource estimates for the magnetite and magnetite-pyroxenite domains were prepared usingordinary kriging. Ellipsoid search anisotropy was applied by using the ZYZ method described above andassuming that there is no special deposit orientation. This method relies upon all three axes beingorthogonal. This method also yields a unique orientation for the resulting anisotropy search ellipsoidwhich is independent from block model orientation. No rotation is applied to the block model.

No specific filtering and no cross validation was performed during interpolation. Anisotropic inversedistance interpolation (ID2) was also used, with the same search ellipsoid, to compare the difference ingrade distribution between the two methods of interpolation. Ordinary kriging returned an acceptabledistribution of grade and tonnage and was chosen as the primary method for interpolation because of thewell-modelled variograms which were obtained.

For purposes of interpolation, some constraints were applied. The minimum and maximum number ofsample composites used to estimate the grade of a block were set at 3 and 10, respectively, for theIndicated category, and 5 and 10 for the Measured category, based on the drilling density and requireddegree of confidence. For the 2012 update 2 and 10 were used for the inferred category. If there werenot at least this minimum number of samples within the defined search volume then a default value (0.00)was assigned to that block. If more than the maximum number of samples lay within the defined searchvolume then only the closest ones were selected for use.

The maximum number of samples to be selected per hole was set to two. If more than two samples fromone drillhole were found within the search ellipse, the program looks further for other points (samples),even though the next closest may still be from same drillhole. This was done to prevent too much

smearing of samples from up and down hole, since Gulçari A is a well layered deposit. The constraintcriteria selected for use in 2007 are summarized in Table 14-12 and Table 14-13. Those used in the 2012update are shown in Table 14-14 to Table 14-16.

Table 14-12 2007 Ordinary Kriging Interpolation Parameters, Measured Resources

Principal Intermediat Dip Strike Across Dip Max. No. Min. No. Max. No.

Azimuth Dip Azimuth Range Range Range Per Hole Samples Samples

(º) (º) (º) (m, Y) (m, X) (m, Z) # # #

Magnetite (V2O5) 195 -14 252 20 35 20 2 5 10

Magnetite-

Pyroxenite (V2O5)200 0 290 33 33 16 2 5 10

Magnetite (Pt) 195 -14 252 16 19 10 2 5 10

Magnetite-

Pyroxenite (Pt)200 0 290 19 31 11 2 5 10

Magnetite (Pd) 195 -14 252 14 24 13 2 5 10

Magnetite-

Pyroxenite (Pd)200 0 290 18 30 14 2 5 10

Domain





Table 14-13 2007 Ordinary Kriging Interpolation Parameters, Indicated Resources

Table 14-14 2012 Ordinary Kriging Interpolation Parameters, Measured Resources

Table 14-15 2012 Ordinary Kriging Interpolation Parameters, Indicated Resources

Table 14-16 2012 Ordinary Kriging Interpolation Parameters, Inferred Resources

The grade interpolation of the block model was performed within the magnetite and magnetite-pyroxenitedomains without consideration of the post-mineralization gabbro and pegmatite intrusives. These wereaccounted for afterwards by calling up the waste solid models, "nulling" the overlapping blocks in theblock model and reassigning them to other rock codes. This was done using Gemcom's needling routine.Needling was performed for all domains coded by a rock code, by block model level, and on a regular gridwith a 5 by 5 integration level.

Principal Intermediat Dip Strike Across Dip Max. No. Min. No. Max. No.

Azimuth Dip Azimuth Range Range Range Per Hole Samples Samples

(º) (º) (º) (m, Y) (m, X) (m, Z) # # #

Magnetite (V2O5) 195 -14 252 100 100 60 2 3 10

Magnetite-

Pyroxenite (V2O5) 200 0 290 100 100 55 2 3 10

Magnetite (Pt) 195 -14 252 75 100 40 2 3 10

Magnetite-

Pyroxenite (Pt)200 0 290 60 100 40 2 3 10

Magnetite (Pd) 195 -14 252 50 90 45 2 3 10

Magnetite-

Pyroxenite (Pd)200 0 290 60 100 45 2 3 10

Domain

Dip Strike Across Dip Max. Min. Max.

Range Range Range No. Per No. No.

(m, X) (m, Y) (m, Z) Hole Samples Samples

Magnetite (V2O5) 160 -55 -20 40 40 15 2 5 10

Magnetite-

Pyroxenite (V2O5)160 -55 -20 40 40 15 2 5 10

Domain Z Y Z




Magnetite (V2O5) 160 -55 -20 115 120 30 2 3 10

Magnetite-

Pyroxenite (V2O5) 160 -55 -20 130 160 30 2 3 10

Domain Z Y Z




Magnetite (V2O5) 160 -55 -20 230 240 60 2 -2 10

Magnetite-

Pyroxenite (V2O5)160 -55 -20 260 320 90 2 2 10

Domain Z Y Z





A solid precedence was applied to resolve the overlap among the different mineralized domains and"waste" solids that occurred when the gabbro and pegmatite dike solids were applied. Where solidsoverlap in a block, Gemcom assigned the data to the first possible solid, in order of precedence with Solid1 being needled before Solid 2 and so on. Once named by a precedence solid, a block will not berenamed by a lower precedent. This procedure was also used to assign blocks on the boundary of themagnetite and magnetite-pyroxenite domains to one domain or the other.

Gemcom's needling utility was used to develop a "partial percentage model" with precise portions of eachblock being assigned to the different domains. Afterwards each portion was weighted by assigning thegrade. By applying a percent model, each domain that occupies a portion of each block selected will beassigned with the appropriate rock code from its order of precedence.

A volumetrics report using a percent model, will calculate the volumes by multiplying the block size by theblock percent. This method is an effective way to accurately estimate volumes of different mineralizationtypes such as in the vicinity of the pegmatite dikes and mineralized solids at Gulçari A. The magnetiteand magnetite-pyroxenite domains were locally diked-out by late post-mineralization intrusive rocks tobelow 150 m (500 m in 2012). The use of a percent model is arguably a better choice to deal with the

volumes in these different domains.

The details of the domain and solid precedence for needling purposes can be seen in Table 14-17.

Table 14-17 Solid and Domain Precedence for Block Modelling of the Gulçari A Deposit

14.8.A Mineral Resources

The 2007 mineral resources in this report were estimated using the Canadian Institute of Mining,Metallurgy and Petroleum (CIM), CIM Standards on Mineral Resources and Reserves, Definitions andGuidelines prepared by the CIM Standing Committee on Reserve Definitions and adopted by CIM CouncilDecember 11, 2005. The 2012 resource update is compliant with the definitions adopted by CIM Councilon November 27, 2010.

Under the CIM definitions, a mineral resource must be potentially economic in that it must be "in suchform and quantity and of such a grade or quality that it has reasonable prospects for economicextraction." Micon has used a cut-off grade of 0.66% V2O5 for the reporting of the 2007 Measured andIndicated resource. This grade was chosen based on initial USD 3.50/lb V2O5 pit shell around theinterpolated blocks. Whittle software was used to define three pit shells based on relevant technical andeconomic parameters taken from the Maracás scoping study (Jacobs et. al., 2007). The resources were

Precedence Order Rock Type Rock Code

1 Air

2 Dike 4

3 Gabbro 5

4Magnetite-Pyroxenite

enclaves within magnetite2

5 Magnetite 1

6 Magnetite-Pyroxenite 2





reported from a block model with Gemcom software and a pit shell optimized with Whittle 4X softwareusing an average USD 42.23/t operating cost (mining, processing, and G&A costs).

For the 2007 estimate three pit shells were designed based on USD 3.50/lb, USD 5.00/lb and USD7.50/lb V2O5 were run. The reported resource here lies entirely within the USD 3.50/lb pit shell andrepresents a potential open pit mineral resource. The 2012 estimate was also pit shell-constrained using

an all-in operating cost of $34.20/t and was reported at a 0.45% V2O5 cut-off grade reflecting higher V2O5 prices. The pit shell was optimized by Jonathon Steedman of Micon under the direction of B. TerrenceHennessey, P.Geo.

The mineral resources for the Gulçari A deposit, as determined in 2007 by the methodology describedabove, and reported at a 0.66% V2O5 cut-off grade, are set out in Table 14-18. Only the magnetite andmagnetite-pyroxenite domains have been reported. The vanadium grade is presented as vanadiumpentoxide as is the custom in the industry. The mineral resources reported in Table 14-18 were current asof December 21, 2007 and were used in the 2008 feasibility study as well as the Technical update to thefeasibility presented in this report. For this reason the table is included in this report. The mineralresources were updated in September, 2012 to reflect recent drilling and a reinterpretation of the dykes

cutting the deposit (see Figure 14-56 to Figure 14-60 in Section 14.8.A Mineral Resources).Table 14-18 Gulçari A Deposit Mineral Resource as at December, 2007

1

1 Resources within a USD 3.50 Whittle pit shell using USD 42.23/t operating costs and reported at a 0.66% V 2 O5 cut-

off grade.

For the purposes of this Report, all mineral resources were classified in the Measured and Indicated

categories. No Inferred mineral resources were determined in 2007.

Consistent with the parameters developed in the previous preliminary assessment (Jacobs et al., 2007),the Measured and Indicated resources presented here have an average cut-off grade of 0.66% at a V2O5 price of USD 3.50/lb. A price for V2O5 at USD 5.00/lb is consistent with a cut-off grade of 0.459% and atUSD 7.50/lb a cut-off grade of 0.306%. The results of the 2012 resource update are presented in Table14-19. PGM grades were not determined for the 2012 estimate.

Contained Contained

V2O5 Pt Pd PGM V2O5 V2O5

Category / Zone Tonnes (%) (g/t) (g/t) (g/t) (t) (lb)

Measured

Magnet ite Zone 2,450,000 2.17 0.39 0.15 0.54 53,300 117,300,000

Indicated

Magnet ite Zone 5,920,000 1.90 0.23 0.11 0.34 112,700 247,900,000

TOTAL 8,370,000 2.00 0.28 0.12 0.40 166,000 365,200,000

Measured

Magnetite-

Pyroxenite Zone3,410,000 0.92 0.20 0.06 0.26 31,400 69,100,000

Indicated

Magnetite-

Pyroxenite Zone5,480,000 0.93 0.12 0.07 0.19 50,900 112,000,000

TOTAL 8,890,000 0.93 0.15 0.07 0.22 82,300 181,100,000

GRAND TOTAL 17,260,000 1.44 0.21 0.09 0.30 248,300 546,300,000





Table 14-19 Gulçari A Deposit Mineral Resource as at September, 20121

Although the mineral deposit at Gulçari A is not a recent discovery, Micon is unaware of any previousattempt to permit a mine at the site prior to Largo’s involvement. Micon is not aware of anyenvironmental, permitting, legal, title, taxation, socioeconomic, marketing or political issues which wouldadversely affect the mineral resources estimated herein. Brazil and Bahia State are mining-friendly

jurisdictions and mining has taken place in the general area for many years.

Figure 14-50 through Figure 14-55 are example plan and sectional views of the 2007 Gulçari A modelillustrating various features including the initial pit model based on USD 3.50/lb V2O5.

Figure 14-50 and Figure 14-51 are vertical sectional views (6200 N and 6150 N) of the block modelshowing the distribution of the magnetite (red polyline boundaries) and magnetite-pyroxenite (greenpolyline boundaries) domains and coded blocks. The interior text on each block is colour-coded by graderange for V2O5. Two sets of pegmatite dikes are seen where blocks have been nulled as waste materials.

Figure 14-52 is an example section (6200 N) showing isoshell grade contour of block grades for themagnetite and magnetite-pyroxenite domains. Yellow contours define the boundary of magnetite-pyroxenite domain with waste rocks and dikes and also the lower grade pyroxenite unit.

Figure 14-53 and Figure 14-54 are example plan views of the deposit (at the 280- and 170-m elevation)showing distribution of the magnetite (red boundaries) and magnetite-pyroxenite (green boundaries)coded blocks. The interior text on each block is colour-coded by grade range for V2O5. Waste materialssuch as gabbro (blue boundaries) and dike (gray boundaries) are excluded and shown as a whitebackground.

Figure 14-55 is an example plan view (at the 240-m elevation) showing isoshell grade contours of blockgrades for the magnetite and magnetite-pyroxenite domains. Yellow contours define the boundary of themagnetite-pyroxenite domain with waste rocks, dikes and the lower grade pyroxenite unit.

Contained

Category / Zone Tonnes V2O5 V2O5

(t) (%) (t)

Measured 8,870,000 1.37 121,500Indicated 15,770,000 0.96 151,400

Measured + Indicated 24,640,000 1.11 272,900

Inferred 2,610,000 0.76 19,800





Figure 14-50 Gulçari A Deposit, Section 6200N - Domain and V2O5 Grade Distributions

Figure 14-51 Gulçari A Deposit, Section 6150N - Domain and V2O5 Grade Distributions





Figure 14-52 Gulçari A Deposit, Section 6200 N - Isoshell Grade Contours

Figure 14-53 Gulçari A Deposit, Plan View 280 m Level, Domain and V2O5 Grade Distributions





Figure 14-54 Gulçari A Deposit, Plan View 170 m Level, Domain and V2O5 Grade Distributions

Figure 14-55 Gulçari A Deposit, Plan View 240 m Level - Isoshell Grade Contours

Figure 14-59 through Figure 14-60 are example plan and section views of the 2012 Gulçari A modelillustrating various features including the updated pit model based on V2O5 as well as the new pegmatitedike interpretation.





Figure 14-56 and Figure 14-57 are vertical sectional views (6075 N and 6175 N) of the block modelshowing the distribution of the measured, indicated and inferred resources. The pegmatite dikes areseen where blocks have been nulled as waste materials.

Figure 14-58 and Figure 14-59 are example section views (6075 N and 6175 N) of the deposit colour-coded by grade range for V2O5. Waste materials such as gabbro and pegmatite dike are excluded and

shown as a white background.

Figure 14-60 is a plan view of the deposit at the 210 elevation colour-coded by grade range for V 2O5.Waste materials such as gabbro and pegmatite dike are excluded and shown as a white background.

Figure 14-56 Gulçari A Deposit, Section 6075 Showing Resource Confidence Categories





Figure 14-57 Gulçari A Deposit, Section 6175 Showing Resource Confidence Categories

Figure 14-58 Gulçari A Deposit, Section 6075 Showing V2O5 Grade Distributions





Figure 14-59 Gulçari A Deposit, Section 6175 Showing V2O5 Grade Distributions

Figure 14-60 Gulçari A Deposit, Plan View 210 m Level, Showing V2O5 Grade Distributions





14.9.A Sensitivity Studies

The mineral resources for the magnetite and magnetite-pyroxenite domains were estimated incrementallyfor consecutive grade groups by kriging. This allows for the results to be reported incrementally fordifferent cut-off grade categories and presented in a sensitivity analysis. Table 14-20 presents the resultsof a sensitivity analysis of the Measured and Indicated mineral resources to cut-off grade. This analysiswas performed using the USD 5.00 V2O5 pit shell rather than the USD 3.50 shell from the resourceestimate.

Table 14-20 Sensitivity Table, Measured and Indicated Resource, 2007 Gulçari-A Deposit

(Using USD 5.00/lb V2O5 Pit Shell)

Table 14-21 shows a similar analysis completed on the 2012 resource model.

V2O5 Contained Contained

Cutoff Grade Tonnes Grade V2O5 V2O5 Pt Pd

(%V2O5) (‘000) (%) (t) (‘000 lb) (g/t) (g/t)

0.36 22,998 1.25 287,000 632,700 0.20 0.08

0.46 22,534 1.27 286,000 630,500 0.20 0.08

0.56 21,172 1.32 279,000 615,100 0.21 0.09

0.66 18,327 1.43 262,000 577,600 0.22 0.09

0.76 15,289 1.57 240,000 529,100 0.25 0.10

0.86 13,329 1.69 225,000 496,000 0.27 0.11

0.96 11,576 1.82 211,000 465,200 0.29 0.12

1.06 10,547 1.90 200,000 440,900 0.30 0.12

1.50 7,604 2.14 163,000 359,400 0.33 0.14

2.00 4,477 2.35 105,000 231,500 0.36 0.13

2.50 1,077 2.74 29,500 65,040 0.44 0.10

3.00 130 3.19 4,100 9,030 0.48 0.08





Table 14-21 Sensitivity Table, Measured and Indicated Resource, Gulçari A Deposit within Micon 2012Pit Shell

(Using USD 6.75/lb V2O5 Pit Shell)

* Table 14-21 was reported from the 2012 block model by Farshid Ghazanfari from the 2012 Whittle pit using

Gemcom software. The 2012 mineral resources in Table 14-19 were reported from the Whittle pit optimized on the

block model using Surpac software. Because of the different reporting algorithms used in the two software packages

minor discrepancies exist in the reported quantities.

14.10.A Confirmation of Estimation

After grade interpolation was completed, Micon reviewed all sections in the Gulçari A model in Gemcom.Block grades on each section were compared to nearby drillholes and the geological domain model.Checks were made for consistency of block grades with nearby informing drill-hole samples, nulling ofblocks in dikes and consistency of grade interpolation with the hard boundaries of the geological domainmodel.

14.11.A Responsibility for Estimation

The 2007 mineral resource estimate presented in this report was prepared by Farshid Ghazanfari, withtechnical input from Robert Anderson Campbell, P.Geo. and B. Terrence Hennessey, P.Geo. Themineral resources and methodology employed to estimate them have been reviewed, and overallresponsibility for them has been accepted by B. Terrence Hennessey, P.Geo.

The mineral resource estimate for the Gulçari A deposit was updated in September, 2012 to include the

results of 13 recent drillholes and a reinterpretation of the unmineralized pegmatite dykes cutting thedeposit. Other than the inclusion of the new drill results, the amended geological model and the minorchanges noted throughout Section 14, no major changes to the 2007 methodology have been made. The2012 mineral resource results have been reviewed, and overall responsibility for them accepted, by B.Terrence Hennessey, P.Geo.

V2O5 Contained

Cutoff Grade Tonnes Grade V2O5

(%V2O5) (‘000) (%) (t)

0.45* 24,628 1.11 273,000

0.55 21,490 1.19 257,000

0.65 17,556 1.33 233,000

0.75 14,353 1.47 211,000

0.85 11,702 1.62 190,000

0.95 9,896 1.75 173,000

1.05 8,733 1.85 162,000

1.50 6,198 2.10 123,000

2.00 3,445 2.35 81,0003.00 95 3.19 3,000





14.1.B Mineral Resources Estimates

14.1.1.B Introduction

The purpose of this Technical Report is to report the mineral resource of the Satellite orebodies,incorporating the most current economic parameters and reflecting the geological information as of theend of August 2012. To accomplish the update, Coffey Mining had to review and define the mineralresource estimating parameters and methods, economic limits of the pit, etc.

Coffey Mining carried out the Maracás Vanadium Project material mineral resources estimation based onthe current available information that has been obtained from the data from the diamond drilling program.

The grade estimates have been classified as Inferred Mineral Resources in accordance with CIMDefinition Standards (2010) and NI 43-101. A brief discussion on the classification criteria is provided.

14.1.2.B Database

To produce a consistent drilling database of the Satellite deposits, the drilling data from MaracásVanadium Project, originally stored in Access MS-Access software. This database contains informationabout location and orientation of drillholes, descriptive lithology, analytical and density results.

Coffey Mining carried out an electronic validation of the database in Gemcom Surpac software. No errorswere found, but gaps or overlapping data were corrected, and used after validation. Table 14-22summarizes the drillhole database used for the mineral resource estimations.

Table 14-22 Drillhole Database Summary

14.1.3.B Geological Model

The geological modeling of the Maracás Vanadium Project was performed by Largo´s Staff team throughthe interpretation of vertical sections based on the drillholes (Figure 14-61), using lines and polygons togenerate solids (wireframes). In accordance with lithological domains definition, mineralized zones were

modeled (Figure 14-67). The value of 0.3% of V2O5 was applied as cut-off grade and lithological types(Magnetites) guideline on drillhole mineralization definition due to the geological and grade continuity.

The topography survey used in the models was developed from topographical points collected in the fieldwith a total station campaign. This work is undertaken routinely by the mine, and monthly the surveyorexecutes this procedure to update the database.

AREAS TYPE PURPOSE N° OF HOLES TOTAL METERS

GULÇARI A NORTE Mineral Resource 18 3226.53

GULÇARI B Mineral Resource 17 1862.92

SÃO JOSÉ Mineral Resource 15 2567.75

NOVO AMPARO Mineral Resource 16 2135.05

NOVO AMPARO NORTE Mineral Resource 14 2620.85

80 12413.1TOTAL





Figure 14-61 Vertical Sections Location

The limits of mineral resources were determined using the drillhole information. The lateral continuity ofthe model is as delineated by the geological model that was sent by Maracás Vanadium Project team.The shape of the geological model is presented below; the orebodies model was treated individuallyduring the resource estimation (Figure 14-62 to Figure 14-66).





Figure 14-62 Geological Model - Gulçari A North - 3D View

Figure 14-63 Geological Model - Gulçari B - 3D View





Figure 14-64 Geological Model - Novo Amparo - 3D View

Figure 14-65 Geological Model - Novo Amparo North - 3D View





Figure 14-66 Geological Model - São José - 3D View

All geological model solids were validated before the start of the estimation process. After that, thegeological model attribute was included into Block Model data to be used during the statistical estimationprocedures.

Coffey Mining has reviewed the geological model for all Satellites orebodies, and its opinion this work hasbeen completed in a professional manner and that the outcomes can be relied upon for the purposes of

resource estimation.




Figure 14-67 Typical Vertical Sections - Gulçari A North





14.1.4.B Block Model

Coffey Mining used Gemcom Surpac and applied exactly the parameters (origin, limits, orientation andsizes) of Largo in its estimation.

A Block model was constructed using 5 metres x 5 metres x 5 metres parent cell size within the project

field limits (Table 14-23 to Table 14-27) and it was illustrated in Figure 14-68 to Figure 14-72.

Table 14-23 Gulçari A North - Block Model Summary

Table 14-24 Gulçari B - Block Model Summary

Table 14-25 Novo Amparo - Block Model Summary

Table 14-26 Novo Amparo North - Block Model Summary

Type Y X Z

Minimum Coordinates 8486365 318180 50

Maximum Coordinates 8487555 318670 360

User Block Size 5 5 5

Min. Block Size 2.5 2.5 2.5

Rotation Bearing: 20 Dip: 0 Rotation: 0

Type Y X Z






Type Y X Z

Minimum Coordinates 8489535 319175 80Maximum Coordinates 8490095 319435 390




Type Y X Z



User Block Size 5 5 5Min. Block Size 2.5 2.5 2.5






Table 14-27 São José - Block Model Summary

Vertical sections were interpreted using cut-off grade 0.3% V2O5 inside Magnetites lithotypes, which wasbased on the estimation plan.

Also, the estimate were done for separated dataset of weathered rocks, lithological composition for fiveblock models estimated for the deposit site project.

Figure 14-68 Block Model - Gulçari A North Deposit

Figure 14-69 Block Model - Gulçari B Deposit

Type Y X Z




Min. Block Size 2.5 2.5 2.5Rotation Bearing: 20 Dip: 0 Rotation: 0





Figure 14-70 Block Model - Novo Amparo Deposit

Figure 14-71 Block Model - Novo Amparo North Deposit





Figure 14-72 Block Model - São José Deposit

14.1.4.1.B Density

The mass density or density of a material is defined as its mass per unit volume. Mathematically, density(ρ) is defined as mass (m) divided by volume (V). From this equation, mass density must have units of aunit of mass per unit of volume (eg. g/cm3 , kg/m³, etc.).

The material used in the tests performed by Largo was explained in detail in Section 11.4.B.

An average density of each type of lithology were estimated and tagged in the block model reference.The values adopted were based on procedures practiced by Largo in accordance with best practices.

14.1.4.2.B Volumetric Block Model Validation

According to the block model definitions, all blocks in the model have been associated with themineralized envelope defined as orebodies, always considering the cut-off grade of 0.3% of V2O5 definitions.

The visual comparison between the geological model (solids or wireframes) and the block model shows agood fit for the five satellite deposits.

Table 14-28 shows the visual and numeric comparison between wireframes and blocks. Coffey Miningconsider high adherence all values over 95%, also it can be observed for all satellite deposits.





Table 14-28 Volumetric Validation - Geological Model and Block Model

14.1.4.3.B Compositing

In compositing to an appropriate regular downhole length, the aim is:

To achieve uniform sample support; To reduce the impact of random variability; and

To minimize the effect of averaging samples of a skewed distribution.

After examining the raw sample lengths of sampled intervals, and in consideration of the local geology,composites were generated using appropriate nominal lengths with 75% of range at end of intervals (1m).Compositing was applied to the mineralized intervals inside the geological model.

Compositing procedure was applied for all of satellites samples validated to be used on the estimateroutine made by Coffey Mining

14.1.4.4.B Statistical Analysis

Descriptive and distribution statistics have been compiled based upon the 1m composite grade data for

the mineralized area and domains. The statistical examination and the grade characteristics of themineralized intervals of each target area were summarized in the Table 14-29.

Target Orebody Solid Volume Block Model Volume Adherence (ratio)

ore1 2,583,170 2,553,672 0.99

ore2 767,071 756,578 0.99

Gulçari B - 872,083 865,406 0.99

Novo Amparo - 838,124 808,281 0.96

ore1 2,228,089 2,145,938 0.96

ore2 1,805,296 1,635,938 0.91

ore1 1,101,622 1,072,906 0.97

ore2 508,653 501,875 0.99Sao Jose

Gulçari A North

Novo Amparo North





Table 14-29 Basic Statistical Analysis Summary

14.1.4.5.B Variography

Variography is used to describe the spatial variability or correlation of an attribute recognized as aregionalized variable. The spatial variability is traditionally measured by means of a variogram, which isgenerated by determining the averaged squared difference of data points at a nominated distance or lag(h). The averaged squared difference (Variogram or (h)) for each lag distance is plotted on a bivariateplot where the X-axis is the lag distance and Y-axis represents the average squared differences (h) forthe nominated lag distance.

In this report, the term “variogram” is used as a generic word to desi gnate the function characterizing the

variability of variables versus the distance between two samples. The traditional measures have beenapplied for the estimation studies completed for the mineralized intervals of Maracás Vanadium Project.

Gemcom Surpac software has been employed to generate and model the Variography. The rotation isreported as input for grade estimation, with Y (rotation around Y axis), X (rotation around X) and Z(rotation around Z axis) also being referred to as the major, semi major and minor axes.

Variography Discussion

The downhole experimental variograms were calculated to establish the structures for composite grades.

Variogram maps and variogram models were constructed using Gemcom Surpac software. The

variography results are summarized in Table 14-30. Figure 14-73 and Figure 14-74 are examples ofvariographic analysis for one satellite deposit in Maracás Vanadium Project.

During the validation work, each domain was checked and the final result was interesting, because alldata domains presented particular behavior in relation of the spacial geostatistical distribution (Table14-30).

Target Lithology Variable Mean VarianceStd.

Deviation

Coeff

Variation

N°

SamplesMinimum Maximum

Orebody 1 V2O5 0.93 0.06 0.24 26% 25 0.53 1.25

Orebody 1_1 V2O5 0.96 0.05 0.22 23% 46 0.44 1.53

Orebody 1_2 V2O5 0.84 0.1 0.31 37% 81 0.07 1.45

Orebody 1_3 V2O5 1.08 0.02 0.13 12% 7 0.85 1.22

Orebody 2 V2O5 0.71 0.05 0.23 33% 25 0.45 1.3

Orebody 2_1 V2O5 0.62 0.08 0.28 46% 13 0.15 1.26

GB Orebody 1 V2O5 0.71 0.03 0.17 23% 174 0.02 1.18

NA Orebody 1 V2O5 0.72 0.04 0.19 27% 242 0.13 1.24

Orebody 1 V2O5 0.93 0.09 0.3 33% 209 0.22 1.52

Orebody 2 V2O5 0.82 0.06 0.25 30% 110 0.29 1.49

Orebody 1 V2O5 0.92 0.07 0.27 30% 103 0.46 1.5

Orebody 2 V2O5 0.79 0.03 0.19 24% 32 0.44 1.2SJ

GAN

NAN




Table 14-30 Variogram Models Summary

C0 C1 A1 C2 A2 Bearing Plunge Dip

% % m % m ° ° °

Orebody 1 V2O5 0.02 0.0238 70 0.0275 100 197 0 70 1.38 14.10

Orebody 2 V2O5 0.02 0.0238 70 0.0275 100 197 0 70 1.38 14.10

Gulçari B Orebody 1 V2O5 0.004 0.007 30 0.015 100 197 0 60 1.75 10.07

Novo Amparo Orebody 1 V2O5 0.005 0.0111 24 0.0213 100 192 0 80 1.00 13.25

Orebody 1 V2O5 0.01 0.084 100 - - 207 0 80 1.27 10.17

Orebody 2 V2O5 0.015 0.0557 100 - - 197 0 70 1.22 10.00

Orebody 2 V2O5 0.006 0.029 100 - - 212 16 50 1.41 15.25

Orebody 1 V2O5 0.01 0.061 100 - - 203 8 60 1.30 10.07

Gulçari A North

Novo Amparo North

Sao Jose

Target Lithology Variable

Ratio

SemiMajor/

Major

Ratio

Minor/

Major





Figure 14-73 Variographical Analysis - Example Gulçari A North deposit (orebody2)





Figure 14-74 Variographical Analysis - Example Novo Amparo North deposit (orebody1)





14.1.4.6.B Grade Estimation

Coffey Mining had chosen Ordinary Kriging (OK) method to be used to estimate V2O5% on the all ofsatellites orebodies of Maracás Vanadium Project.

Ordinary Kriging (OK) is one of the most common geostatistical methods for grade estimation of the

block. In this interpolation technique, the contributing composited samples are identified through a searchapplied from the centre of each block. The weights are determined to minimize the variance error,considering the spatial localization of the selected composites and the modelled variogram. The grade ofthe weighted composited sample is combined to generate the estimation of the block grade and thevariance.

Typical ordinary kriging assumptions

The typical assumptions for the practical application of ordinary kriging are:

Intrinsic stationarity or wide sense stationarity of the field Enough observations to estimate the variogram.

The mathematical condition for applicability of ordinary kriging is:

The mean E[Z(x)] = μ is unknown but constant The variogram γ(x,y) = E[(Z(x) − Z(y))2] of Z(x) is known.

Ordinary Kriging equation

The kriging weights of ordinary kriging fulfill the unbiasedness condition

and are given by the ordinary kriging equation system:

The additional parameter μ is a Lagrange multiplier used in the minimization of the kriging errorto honour the unbiasedness condition.

Ordinary kriging interpolation

The interpolation by ordinary kriging is given by:





Ordinary kriging error

The kriging error is given by:

Properties of kriging

(Cressie 1993, Chiles&Delfiner 1999, Wackernagel 1995)

The kriging estimation is unbiased:

The kriging estimation honours the actually observed value:

The kriging estimation is the best linear unbiased estimator of Z(x) if the assumptionshold. However (e.g. Cressie 1993)As with any method:

o If the assumptions do not hold, kriging might be bad.o There might be better nonlinear and/or biased methods.o No properties are guaranteed, when the wrong variogram is used. However typically still a

'good' interpolation is achieved.o Best is not necessarily good: e.g. in case of no spatial dependence the kriging interpolation

is only as good as the arithmetic mean.

o Kriging provides as a measure of precision. However this measure relies on thecorrectness of the variogram.

14.1.5.B Grade Estimation Strategy

14.1.5.1.B Estimation Strategy and Search Neighbourhood

The established Kriging plan, for all attributes, considered three estimation steps as presented in theTable 14-31. It was used to honour the geostatistical rule established with the Exploratory Data analysis.





Table 14-31 Ordinary Kriging Strategy - Satellite deposits

Maximum Number of Samples

per Drillhole

1 70 3 12 22 100 3 12 2

3 180 3 12 2

4 500 1 12 2

1 70 3 12 2

2 100 3 12 2

3 180 3 12 2

4 500 1 12 2

1 70 3 12 2

2 100 3 12 2

3 180 3 12 2

4 500 1 12 2

1 70 3 12 2

2 100 3 12 2

3 180 3 12 2

4 500 1 12 2

1 70 3 12 2

2 100 3 12 2

3 180 3 12 2

4 500 1 12 2

Searching Parameters: Bearing=192; Plunge=0; Dip=-80; Major/Semi-Major Ratio= 1; Major/Minor Ratio:13.25;

Novo Amparo North

Variables: V2O5% (orebody 1)

Searching Parameters: Bearing=207; Plunge=0; Dip=-80; Major/Semi-Major Ratio= 1.27; Major/Minor Ratio:10.17;




Gulçari B

Variables: V2O5%


Novo Amparo

Variables: V2O5%

Step Search RadiusMinimum Number

of Samples

Maximum Number

of Samples

Gulçari A North

Variables: V2O5%





Table 14-31 Ordinary Kriging Strategy - Satellite deposits

Figure 14-75 Figure 14-81 to are illustrating all ellipses used to be the representative geometry of theanisotropy

Figure 14-75 Anisotropic Ellipse to Estimate Gulçari A North

1 70 3 12 2

2 100 3 12 2

3 180 3 12 24 500 1 12 2

1 70 3 12 2

2 100 3 12 2

3 180 3 12 2

4 500 1 12 2


Searching Parameters: Bearing=211; Plunge=16.27; Dip=-50; Major/Semi-Major Ratio= 1.4; Major/Minor Ratio:15.24;

São José


Searching Parameters: Bearing=203; Plunge=8.18; Dip=-60; Major/Semi-Major Ratio= 1.29; Major/Minor Ratio:10.07;





Figure 14-76 Anisotropic Ellipse to Estimate Gulçari B

Figure 14-77 Anisotropic Ellipse to Estimate Novo Amparo





Figure 14-78 Anisotropic Ellipse to Estimate Novo Amparo North (Orebody 1)

Figure 14-79 Anisotropic Ellipse to Estimate Novo Amparo North (Orebody 2)





Figure 14-80 Anisotropic Ellipse to Estimate São José (Orebody 1)

Figure 14-81 Anisotropic Ellipse to Estimate São José (Orebody 2)





14.1.5.2.B NN-Check - Resource Validation

The NN-Check, a comparative analysis of OK with the Nearest Neighbour results for V2O5%, shows basicdifferences in grade distributions. There is relative smoothing in the OK results that are compatible withthis estimation technique.

Validation of estimated grades was carried out with a comparative Nearest Neighbor estimation (NN).This validation consists of a comparative statistical analysis of global results for V2O5 (%) to themineralized intervals (Figure 14-82).

Coffey Mining has been considered NN-Check one of method to validate the interpolate method toestimate because it enables an important comparison in the neighbourhood of the sample used tocalculate the mean of the block model.





Figure 14-82 V2O5 % kriged by Coffey Mining – NN-Check Comparative Statistic

14.1.5.3.B Swath Plot - Validation Resource

The Swath Plot is a common validation from local comparative graphs for estimated grades by OK andNN-check considering x, y, or z coordinates.





The local validation by the Swath Plot method shows the smoothing effect in kriging estimation and localbias in depth at Maracás Vanadium Project (Figure 14-83 to Figure 14-85). The smoothing effectpresented in the figures is acceptable. The results of Swath Plot validation are in accordance with a highconfidence level for the resource estimate.

Figure 14-83 Swath Plot – East Coordinate (X) – V2O5 (%)





Figure 14-84 Swath Plot – North Coordinate (Y) – V2O5 (%)

Figure 14-85 Swath Plot – Elevation (Z) – V2O5 (%)





14.2.B Resource Reporting

The Vanadium deposits of Maracás Project, including all satellite deposits had presented in totally on thisreport, have been classified as Inferred Resources based on the assessment of the among input data,geological interpretation, quality of grade estimation and the results of the pilot tests.

The key criteria assessed as part of the resource classification and the resource quantity is presented inTable 14-32.

Table 14-32 Confidence Level of Key Criteria

Items Discussion Confidence

Drilling TechniquesAll drill holes were completed by rotative diamond drilling methods

and are industry standard approach.High

Logging Standard nomenclature and apparent good quality High

Drill Sample RecoveryThe database presents a good average recovery of 95% but there was a

moderate recovery variation.Moderate to High

Sub-sampling Techniques and

Sample Preparation

Sampling was planned on a nominal 1m interval following the drilling

progress. The fie ld preparation and the laboratory preparation were

industry standard. The blank samples show one low contaminationtendency that needs to be analyzed.

High

Quality of Assay Data

In the general context, the parameters from the quality control

analysis of the reference samples from exploration are inside the

acceptance limits.

Moderate to High

Drill hole SurveyingAll drill holes need to be surveyed (instead of 15m per interval)

deviation measurements.Low

Location of Sampling Poi ntsThe field samples and the drill holes collars were collected using

precise topographic survey.High

Data Density and Distribution

The drilling spacing is not regular covering the entire target area

(considering all targets). The drilling spacing is not enough to enable

confident interpretation of the spatial distribution of the orebodies,

also on deep.

Low to Moderate

Database Integrity The drill hole database was presented without errors. High

Geological InterpretationThere is a very good understanding of the stratigraphy and structural

controls to construct a robust geological model .Moderate to High

Density

There is a sufficient number of density tests available to resource

measurements, but tests with more samples can improve the spatial

variability analysis of density values.

Moderate

Estimation and Modeling

Techniques

The modeling was performed in 3D platform with maximum

confidence with sampling intervals (snap to point to drillhole

information).Ordinary Kriging (OK) method was used to V 2O5%

estimation. The OK estimation techniques are accepted industry

considered best practice.

High





The results of the mineral resource estimate presented here demonstrate a Mineral Resource inferred of27.82 millions of tonnes of vanadium at 0.83% of the mineralized material.

The total Inferred Mineral Resource is presented in Table 14-33, the Maracás Vanadium Project MineralResources was based on the economic scenario of USD 34.20/ tonne and cut-off (0.45%).

Table 14-33 Grade Tonnage Table – 17th September 2012

Table 14-34 presents a summarized scenario applied on cut-off grade definition for Satellite deposits. Thesource data for cut-off definition was based on Gulçari A deposit parameters not detailed on thisdocument. All of the information was provided by Largo´s team.

Table 14-34 Cut-off grade definition based on economic parameters

The Mineral Inferred resource was constrained by mathematics open pit design that was interpretedbased on the cut-off grade definition explained on Table 14-34.

V2O5

(%)

Gulcari A Norte Inferred 9 730 000 0.84 81 400

Gulcari B Inferred 2 910 000 0.7 20 300

Novo Amparo Inferred 1 560 000 0.72 11 300

Novo Amparo Norte Inferred 9 720 000 0.87 84 400

São Jose Inferred 3 900 000 0.89 34 700

Satellite Deposits (5) Inferred 27 820 000 0.83 232 100

Resource within a pit shell using US$34.20/t operating cost and reported at a 0.45% V 2O5 cutoff, run and reviewed by Hebert

Oliveira (Coffey Mining)

Mineral Resources - Exclusive

Satellite Target Resource Classification Tonnes Contained V2O5 (tonnes)

Variable Value Unit

Vanadium Price 10.50 US$/t

Process Costs 32.40 US$/t

Mining Costs 1.42 US$/t

G&A 2.12 US$/t

Metallurgical Recovery 71% %

Cut Off Grade 0.45 %





15 Mineral Reserve Estimates

The guidelines set out in CIM Standards of Mineral Resources and Reserves Definitions and Guidelinesadopted by the CIM council, December 13, 2005 require that all material classified as Measured,Indicated, or Inferred be reported as a Mineral Resource. Therefore, all in-pit mineral resources will bepresented in Section 16 of this Preliminary Economic Assessment.





16 Mining Methods

16.1 Introduction

The Maracás project is planned to be an open pit vanadium mining project consisting of six open pits,which will utilize a combination of hydraulic excavators, large front end loaders and 40 tonne haul trucksas the primary mining equipment. There are two hard rock types, which are referred to as Magnetite-Pyroxenite (MG) and Magnetite (HG) and are characterized by their contained vanadium expressed asthe V2O5 % content.

Based on the mine optimization analysis, several ultimate pits were designed in the North-East region ofthe deposit. The mine schedule delivers an average annual production of 1.1 Mt of run-of-mine (ROM)hard-rock ore with an average V2O5 grade of 1.30% during the first 10 years of production, and an annualaverage of 1.5Mt of ore at 1.10% for the remainder of the 29-year project. Material that is comprised of0.58% V2O5 of blended HG and MG material is planned to be stockpiled separately at a rate of 40 kt perannum. On average, 3.7Mt of waste material is handled annually during the first 10 years of production,

and 12.3Mt annually thereafter.

Using this production schedule, capital and operating cost estimates were developed for the projectincluding mine and processing plant expenditures. These estimates were then incorporated into aneconomic model to determine the viability of the project based on a delivered product price of USD 14.04per kilogram V2O5 and USD 28.01 per kilogram ferrovanadium FOB.

16.2 Pit Optimization

Design of the ultimate pit was based on the results of the Whittle Lerchs-Grossmann shell analysis.

Whittle is a software package that uses the Lerchs-Grossmann algorithm to determine the approximate

shape of a near-optimal pit shell based on applied cut-off grade criteria and pit slopes. These shells aregenerated from the geologic grade models, economic and physical criteria.

For the Whittle analysis the grade models for vanadium and rock density were exported from Vulcan, andimported into Whittle. Values for each of the blocks in the model were set by Whittle based on recovery,economic and physical parameters shown in Table 16-1.

The sale price for ferrovanadium (FeV) used for the pit optimization exercises was based on the 3-yearaverage price period from mid-December 2009 to mid-December 2012, which is consistent with industrybest practice for mineral reserve estimation. The sale price of the Fe byproduct is also based on the 3-yrmarket history.

The unit mining operating cost used for the Gulçari A pit was based on detailed study carried outpreviously for the 2012 Technical Report. Unit mining costs for the satellite pits were estimated usingincremental haulage analysis, based on haul distance and grade profiles.

The planned open pit mining recovery rate is 100 percent, based on the assumption made in the 2009Feasibility Study. Davis Tube Recovery test results were used to develop unit processing cost estimatesfor each of the deposits based on their respective ore qualities.





Table 16-1 Pit Optimization Parameters

Based on analysis of the 2009 Feasibility Study and 2012 SGS magnetic recovery testwork, RPM wasinstructed by Largo to apply average magnetic recoveries for both HG and MG material, a yield fromconcentrate to V2O5 of 81.93%, a V2O5 to V in FeV recovery of 94.50% and a stoichiometry recovery of56.0% for HG and MG.

RPM developed a range of pit shells for a range of vanadium prices and cut-off grades to evaluate theeconomic viability of MG and HG material. A range of shells were developed to determine the project’ssensitivity and the basis for the designed ultimate pit. As the cut-off grades decreased, the tonnes

increased and resultant grade decreased, since the lower cut-off grade increases the revenue per tonneand thus the tonnage available for mining. Table 16-2 to Table 16-7 details the pit optimization results.

There are several different methods for selecting a pit shell for use as the basis for the ultimate pit design.In the case of the Maracás Project, the decision was made to use pit shells which corresponded to a USD26.88 per kilogram of V in FeV sale price, the 3-yr average price at the time of the Whittle Study. The pitoptimization analyses were completed assuming a 45 degree slope as per the 2009 Feasibility Study.

With the addition of the satellite pits to the Maracás project, the original Gulçari A ultimate pit designed inthe 2012 Technical Report will be expanded to include Phase 2 which exhibits a lower average grade andhigher strip ratio than Phase 1. The low strip ratio satellite pits will be used to blend against Gulçari APhase 2, as outlined in Section 16.5.1, in order to smooth the material handling requirements, hence,

smooth the equipment fleet requirements.

The optimum final pit results are shown in Table 16-8.


Product Price

V in FeV $/kg 26.88 26.88 26.88 26.88 26.88 26.88

Fe Bi -Product $/t 70.00 70.00 70.00 70.00 70.00 70.00

Operating Costs

Ore Mining $/t 3.40 3.29 3.35 3.73 3.96 5.01 Waste Mining $/t 3.40 3.29 3.29 3.29 3.29 3.29

Processing $/t 54.87 48.77 56.50 50.63 56.72 51.47

Cutoff Grades

Vanadium & Iron Ore % 0.35% 0.43% 0.32% 0.44% 0.32% 0.43%

Recovery

Mining % 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%

Processing to V2O5 % 75.38% 64.60% 73.88% 64.68% 71.75% 65.93%

Processing from V2O5 to V in FeV % 52.94% 52.94% 52.94% 52.94% 52.94% 52.94%

Physical Parameters

Magnetite Density t/m3 4.45 4.28 4.42 4.30 4.36 4.26

Manetite-Pyroxenite Density t/m3 3.57 3.32 3.35 3.33 3.38 3.33

Waste Density t/m3 3.09 3.03 2.90 3.05 3.00 3.06

Fe Bi -Product Percentage % 36.00% 23.36% 40.72% 28.12% 41.89% 29.95%Pit Slope

Section 0-360 degrees 45 45 45 45 45 45





Table 16-2 Pit Optimization - Gulçari A

Table 16-3 Pit Optimization - Gulçari A Norte

Minimum Maximum Rock Ore Strip Max Min V2O5 V2O5 CON CON

Pit Rev Ftr Rev Ftr Tonnes Tonnes Ratio Bench Bench Units Grade Units Grade

1.00 0.10 0.20 - - - - - - - -

2.00 0.30 0.30 130,170 81,794 0.59 105 99 213,431 2.61 2,944,567 36.00

3.00 0.40 0.40 12,763,769 3,533,000 2.61 105 73 7,573,897 2.14 127,187,986 36.00

4.00 0.50 0.50 27,881,405 6,609,453 3.22 105 64 12,303,684 1.86 237,940,307 36.00

5.00 0.60 0.60 37,256,448 9,930,405 2.75 105 62 15,745,023 1.59 357,494,589 36.00

6.00 0.70 0.70 41,710,350 12,486,222 2.34 105 60 17,674,671 1.42 449,503,989 36.00

7.00 0.80 0.80 148,297,026 21,553,907 5.88 105 34 27,373,240 1.27 775,940,682 36.00

8.00 0.90 0.90 181,061,433 24,167,027 6.49 105 29 29,811,110 1.23 870,013,004 36.00

9.00 1.00 1.00 202,508,306 25,815,606 6.84 105 27 31,171,156 1.21 929,361,860 36.00

10.00 1.10 1.10 215,183,525 26,716,279 7.05 105 25 31,850,009 1.19 961,786,101 36.00

11.00 1.20 1.20 221,806,402 27,316,432 7.12 105 25 32,192,057 1.18 983,391,595 36.00

12.00 1.30 1.30 224,215,953 27,747,694 7.08 105 25 32,344,600 1.17 998,917,030 36.00

13.00 1.40 1.40 227,874,674 28,149,737 7.10 105 25 32,501,184 1.15 1,013,390,585 36.00

14.00 1.50 1.50 228,616,286 28,425,612 7.04 105 25 32,556,154 1.15 1,023,322,061 36.00

15.00 1.60 1.60 230,286,053 28,724,174 7.02 105 25 32,625,989 1.14 1,034,070,310 36.00

16.00 1.70 1.70 232,806,468 29,050,361 7.01 105 25 32,705,660 1.13 1,045,813,036 36.00

17.00 1.80 1.80 234,519,192 29,320,518 7.00 105 25 32,756,453 1.12 1,055,538,680 36.00

18.00 1.90 1.90 235,849,017 29,573,629 6.97 105 24 32,793,563 1.11 1,064,650,663 36.00

19.00 2.00 2.00 236,330,273 29,819,191 6.93 105 24 32,811,033 1.10 1,073,490,889 36.00



1.00 0.50 0.50 223 122 0.82 25 24 160 1.30 2,858 23.36

2.00 0.60 0.60 84,698 37,507 1.26 55 24 38,528 1.03 876,088 23.36

3.00 0.70 0.70 940,259 245,934 2.82 55 23 244,596 0.99 5,744,596 23.36

4.00 0.80 0.80 8,532,993 1,735,467 3.92 57 23 1,524,711 0.88 40,537,355 23.36

5.00 0.90 0.90 17,336,542 2,948,042 4.88 57 22 2,559,832 0.87 68,860,862 23.36

6.00 1.00 1.00 27,315,506 3,908,970 5.99 57 22 3,378,584 0.86 91,306,371 23.36

7.00 1.10 1.10 39,149,770 4,834,516 7.10 57 20 4,153,410 0.86 112,925,410 23.36

8.00 1.20 1.20 52,555,712 5,702,123 8.22 57 18 4,872,804 0.85 133,191,095 23.36

9.00 1.30 1.30 68,593,277 6,596,567 9.40 57 17 5,605,418 0.85 154,083,666 23.36

10.00 1.40 1.40 90,759,312 7,661,408 10.85 57 15 6,478,504 0.85 178,956,386 23.36

11.00 1.50 1.50 108,287,524 8,375,734 11.93 57 13 7,080,131 0.85 195,641,733 23.36

12.00 1.60 1.60 121,131,252 8,851,936 12.68 57 13 7,476,536 0.84 206,764,932 23.36

13.00 1.70 1.70 131,352,634 9,217,030 13.25 57 13 7,768,564 0.84 215,292,836 23.36

14.00 1.80 1.80 135,266,897 9,345,264 13.47 57 13 7,871,069 0.84 218,288,143 23.36

15.00 1.90 1.90 142,038,968 9,539,143 13.89 57 13 8,031,192 0.84 222,816,815 23.36

16.00 2.00 2.00 148,590,378 9,718,934 14.29 57 13 8,176,971 0.84 227,016,388 23.36





Table 16-4 Pit Optimization - Gulçari B

Table 16-5 Pit Optimization - São José

Table 16-6 Pit Optimization - Novo Amparo



1.00 0.60 0.60 903 345 1.62 42 40 326 0.94 14,061 40.72

2.00 0.70 0.70 625,956 278,701 1.25 42 34 213,680 0.77 10,880,485 39.04

3.00 0.80 0.80 2,364,439 750,911 2.15 42 30 550,156 0.73 29,179,666 38.86

4.00 0.90 0.90 4,977,894 1,186,216 3.20 43 26 840,550 0.71 46,607,055 39.29

5.00 1.00 1.00 12,132,479 1,970,140 5.16 43 16 1,384,916 0.70 76,380,134 38.77

6.00 1.10 1.10 19,018,376 2,497,452 6.62 43 11 1,758,990 0.70 97,078,146 38.87

7.00 1.20 1.20 25,911,533 2,922,474 7.87 43 8 2,060,399 0.71 113,872,784 38.96

8.00 1.30 1.30 33,931,741 3,337,355 9.17 43 4 2,354,185 0.71 130,512,748 39.11

9.00 1.40 1.40 38,717,876 3,554,413 9.89 43 3 2,510,793 0.71 139,067,552 39.13

10.00 1.50 1.50 42,305,484 3,688,222 10.47 43 3 2,609,093 0.71 144,208,308 39.10

11.00 1.60 1.60 43,389,581 3,723,487 10.65 43 3 2,636,119 0.71 145,524,432 39.08

12.00 1.70 1.70 45,378,810 3,783,971 10.99 43 3 2,681,637 0.71 147,723,596 39.04

13.00 1.80 1.80 46,284,463 3,808,276 11.15 43 3 2,699,636 0.71 148,702,252 39.05

14.00 1.90 1.90 48,169,975 3,855,294 11.49 43 3 2,735,126 0.71 150,412,088 39.01

15.00 2.00 2.00 48,244,753 3,857,061 11.51 43 3 2,736,893 0.71 150,412,088 39.00

Minimum Maximum Rock Ore Strip Max Min V2O5 V2O5 CON CONPit Rev Ftr Rev Ftr Tonnes Tonnes Ratio Bench Bench Units Grade Units Grade

1.00 0.60 0.60 196,786 71,624 1.75 47 41 80,054 1.12 2,013,999 28.12

2.00 0.70 0.70 1,476,388 408,996 2.61 47 37 406,745 0.99 11,500,536 28.12

3.00 0.80 0.80 6,524,127 1,275,880 4.11 47 30 1,187,725 0.93 35,876,984 28.12

4.00 0.90 0.90 11,293,538 1,831,763 5.17 47 27 1,688,940 0.92 51,508,185 28.12

5.00 1.00 1.00 17,361,315 2,397,409 6.24 48 22 2,179,200 0.91 67,413,854 28.12

6.00 1.10 1.10 23,669,689 2,849,442 7.31 48 20 2,567,663 0.90 80,124,784 28.12

7.00 1.20 1.20 29,889,073 3,230,289 8.25 48 17 2,890,154 0.89 90,833,971 28.12

8.00 1.30 1.30 38,488,224 3,690,647 9.43 48 13 3,267,592 0.89 103,778,946 28.12

9.00 1.40 1.40 46,760,130 4,063,717 10.51 48 11 3,583,685 0.88 114,269,457 28.12

10.00 1.50 1.50 57,331,809 4,482,284 11.79 49 8 3,934,269 0.88 126,039,307 28.12

11.00 1.60 1.60 64,599,153 4,746,717 12.61 49 5 4,152,923 0.87 133,475,002 28.12

12.00 1.70 1.70 75,408,541 5,099,386 13.79 49 3 4,447,123 0.87 143,391,825 28.12

13.00 1.80 1.80 84,146,358 5,365,784 14.68 49 1 4,662,936 0.87 150,882,750 28.12

14.00 1.90 1.90 92,278,331 5,583,963 15.53 49 1 4,847,841 0.87 157,017,813 28.12 15.00 2.00 2.00 95,530,848 5,667,151 15.86 49 1 4,916,419 0.87 159,356,986 28.12



1.00 0.60 0.60 2,649 1,567 0.69 54 52 1,435 0.92 65,636 41.89

2.00 0.70 0.70 612,817 258,240 1.37 54 46 209,071 0.81 10,414,647 40.33

3.00 0.80 0.80 2,592,807 824,203 2.15 54 40 651,255 0.79 28,925,639 35.10

4.00 0.90 0.90 6,718,366 1,479,935 3.54 54 33 1,171,556 0.79 47,651,753 32.20

5.00 1.00 1.00 10,040,046 1,830,930 4.48 54 29 1,447,875 0.79 58,075,585 31.72

6.00 1.10 1.10 12,856,413 2,050,847 5.27 54 27 1,623,886 0.79 64,343,106 31.37

7.00 1.20 1.20 15,504,284 2,224,441 5.97 54 26 1,761,545 0.79 69,260,628 31.14

8.00 1.30 1.30 19,584,125 2,442,283 7.02 54 23 1,933,195 0.79 75,751,708 31.02

9.00 1.40 1.40 25,191,420 2,695,091 8.35 54 19 2,128,382 0.79 83,776,424 31.08

10.00 1.50 1.50 30,184,030 2,893,560 9.43 54 17 2,283,041 0.79 90,025,281 31.11

11.00 1.60 1.60 34,546,182 3,051,404 10.32 54 15 2,405,281 0.79 95,013,408 31.14

12.00 1.70 1.70 42,299,084 3,303,388 11.80 54 12 2,594,989 0.79 103,340,933 31.28

13.00 1.80 1.80 48,372,747 3,484,209 12.88 54 10 2,732,979 0.78 109,090,229 31.31

14.00 1.90 1.90 58,004,496 3,749,121 14.47 54 7 2,935,381 0.78 117,355,469 31.30

15.00 2.00 2.00 61,489,737 3,836,066 15.03 54 6 3,002,384 0.78 120,058,121 31.30





Table 16-7 Pit Optimization - Novo Amparo Norte

Table 16-8 Optimum Final Pit Results

16.3 Mine Design

The mining method planned for the operation is conventional open pit excavation. The pit will beaccessed by a 15 m wide haul road incorporating a 10 percent gradient and two lane ramps. The rampswill have protective side berms whose height equals half that of the truck’s tire diameter. A minimum

width between haul trucks travelling on same haul road is 3.5 times the truck width, as per standardglobal mining operations. All required roads and ramps will have two lanes sloped 2.0 cm/m at the sides,drainage ditches placed along roadsides, sub-grade support preparation and wearing surface. Thegrades of the haul roads are within the operating range of the mining equipment.

The two ore bodies are characterized by High Grade and Medium Grade vanadium that will be mined inflitches of 5m and 10m benches, targeting a 70 degree face slope. The incorporation of 5m and 10m flitchmining allows for selective mining to increase ore body knowledge, control of grade and ability to blendHG and MG ore. The final pit design will incorporate 20 m high benches, a 45 degree overall and an 8mberm width. The benches will have a slight decline from crest to the toe of the upper bench face slope, inthe direction of the open pit to provide rainfall water drainage and to maintain design slope angles

Approximate final pit geometries are outlined in Table 16-9. Table 16-10 outlines the primary mine designparameters.



1.00 0.60 0.60 672,369 299,794 1.24 57 49 317,847 1.06 8,979,161 29.95

2.00 0.70 0.70 5,433,394 1,706,519 2.18 57 42 1,658,879 0.97 51,112,120 29.95

3.00 0.80 0.80 16,100,517 3,816,443 3.22 57 35 3,497,331 0.92 114,306,528 29.95

4.00 0.90 0.90 27,127,313 5,327,722 4.09 57 28 4,769,762 0.90 159,570,873 29.95

5.00 1.00 1.00 42,172,189 6,936,829 5.08 58 22 6,060,017 0.87 207,765,260 29.95

6.00 1.10 1.10 60,238,117 8,280,203 6.27 58 15 7,205,376 0.87 248,000,689 29.95

7.00 1.20 1.20 75,623,489 9,255,507 7.17 58 10 8,013,538 0.87 277,212,049 29.95

8.00 1.30 1.30 96,000,666 10,429,051 8.21 58 5 8,923,030 0.86 312,360,904 29.95

9.00 1.40 1.40 1 19,786, 598 11,618,408 9.31 58 1 9,832,271 0.85 347,983,397 29.95

10.00 1.50 1.50 137,483,109 12,417,489 10.07 58 1 10,436,303 0.84 371,916,663 29.95

11.00 1.60 1.60 150,590,879 12,942,986 10.63 59 1 10,832,550 0.84 387,655,845 29.95

12.00 1.70 1.70 159,371,998 13,271,533 11.01 59 1 11,070,125 0.83 397,496,180 29.95

13.00 1.80 1.80 165,504,979 13,468,618 11.29 59 1 11,220,996 0.83 403,399,056 29.95

14.00 1.90 1.90 170,369,359 13,611,052 11.52 59 1 11,330,681 0.83 407,665,114 29.95

15.00 2.00 2.00 175,558,791 13,758,336 11.76 59 1 11,439,126 0.83 412,076,421 29.95

Description Units GA\PH1 GA\PH2 GAN GB SJ NA NAN TotalOre Tonnes t 12,486,222 13,329,384 3,908,970 1,970,140 2,397,409 1,830,930 6,936,829 42,859,884

Ore V2O5 Grade % 1.42% 1.01% 0.86% 0.70% 0.91% 0.79% 0.87% 1.06%

Waste Tonnes t 29,224,128 147,468,572 23,406,536 10,162,339 14,963,906 8,209,116 35,235,360 268,669,957

Total Tonnes t 41,710,350 160,797,956 27,315,506 12,132,479 17,361,315 10,040,046 42,172,189 311,529,841






Table 16-9 Approximate Pit Geometries

Table 16-10 Mine Design Parameters

A stockpile will need to be maintained to store ore mined during the preproduction period. Stockpiles andwaste piles were developed using 10 m benches, 8 m berm width, 15 m width haul roads incorporating a10 percent gradient, and a 37 degree face slope. The waste pile capacities and approximate geometriesare outlined in Table 16-11. Table 16-12 details the waste pile design parameters.

Table 16-11 Waste Pile Geometries

Table 16-12 Waste Pile Design Parameters

Figure 16-1 shows the general mine layout and Figure 16-2 through Figure 16-8 display the individual pitand waste pile designs.


Maximum Bench RL m 310 320 320 330 340 350

Minimum Bench RL m -50 210 210 230 240 190

Ramp Access RL m 299 320 313 320 337 354

Width m 875 275 250 250 225 340

Length m 950 800 475 625 400 740

Description Units Value

Two Lane Ramp Width m 15

Ramp Grade % 10

Bench Face Angle degrees 70

Pit Slope degrees 45

Bench Height m 20

Berm Width m 8

Description Units GA PH1 GA PH2 GAN GB SJ NA NAN

Capacity Mm3 15.2 55.8 8.5 5.2 8.1 4.2 19.0

Maximum Height m 50 50 50 50 50 50 50

Footprint Area sq km 0.47 1.19 0.31 0.18 0.28 0.16 0.55

Width m 690 1160 375 360 350 295 575

Length m 1060 1300 950 525 850 590 975

Distance to Pit Exit m 1000 1000 1000 1000 1000 1000 1000

Description Units Value

Ramp Width m 15

Ramp Grade % 10

Lift Face Angle degrees 37

Dump Slope degrees 25

Lift Height m 10

Berm Width m 8





Figure 16-1 General Mine Layout

Novo Amparo Norte

Novo Amparo

Gulçari A Norte

Gulçari A

Gulçari B

São José





Figure 16-2 Mine Layout - Gulçari A Phase 1

Figure 16-3 Mine Layout - Gulçari A Phase 2





Figure 16-4 Mine Layout - Gulçari A Norte

Figure 16-5 Mine Layout - Gulçari B





Figure 16-6 Mine Layout - São José

Figure 16-7 Mine Layout - Novo Amparo





Figure 16-8 Mine Layout - Novo Amparo Norte

Table 16-13 summarizes the pit resources based on the design criteria used.

Table 16-13 Summary of Pit Resources

16.4 Equipment

16.4.1 Equipment Strategy

Ore and waste mining will be carried out using 91 t Hydraulic Excavators equipped with 6.0 m3 bucket,having a digging depth of 14.1 m to excavate and load 24.0 m3 (40 tonne) capacity trucks. The diggingdepth allows for top loading of haul trucks and the ability to maintain wall angles at the prescribed benchheights. In terms of operational flexibility, the excavator enables controlled excavation for mid-bencheswhere the Gabbro, HG and MG zones intersect.

A wheel loader with a 6.0 m3 rock bucket will be used for each mining front. A wheel loader is required formobility within the mining areas due to the requirement to maintain two mining faces containing HG andMG at any one time. The additional front end loader will assist with low grade stockpiles to trickle feed


Ore Tonnes t 13,748,627 11,861,228 3,160,466 1,813,670 2,415,008 1,293,369 7,444,814 41,737,182 Ore V2O5 Grade % 1.34% 1.24% 0.86% 0.69% 0.91% 0.82% 0.87% 1.12%

Waste Tonnes t 27,547,646 153,114,265 21,232,370 11,410,836 17,798,173 9,712,012 41,993,697 282,809,000

Total Tonnes t 41,296,273 164,975,494 24,392,836 13,224,505 20,213,182 11,005,381 49,438,511 324,546,182






the crusher. In periods where stockpile of low grade material is not required, the loader will also work asa clean-up machine to maintain haul roads and piles.

Between 7 and 31 haul trucks with 24.0 m3 (40 tonne) capacity will be required to transport the estimateddaily overburden stripping waste, ROM ore and stockpile material over the life of the project.

Financial analysis determined that this is the most economical and safe method for the operation giventhe susceptibility to long periods of rainfall and high moisture content that will be inherently containedwithin the Saprolite material.

For the drilling activities, a 5.5 inch diameter drill will be used in waste material and 3.5 inch diameter drillin ore. Preliminary evaluation considered a Sandvik 1500i drill that is equipped with both 3.5 inch and 5.5inch drill diameters capabilities to ensure required productivity rates and minimal operational time losses.

The support equipment required for this fleet includes a grader with a 4.3 m 3 blade width and a 310horsepower Track Dozer comparable to a Caterpillar D8T class.

A small hydraulic excavator equipped with a 2.6 m3 bucket is required to support the primary digging fleet

and mine development work.

16.4.2 Productivity and Fleet Size

The performance of the loading units and trucks has been calculated upon the basis of current industrymining equipment performance for the prescribed equipment classes. A detailed estimate of the primaryfleet productivity is outlined in Table 16-14.

Table 16-15 and Table 16-16 detail the annual primary and ancillary fleet requirements.





Table 16-14 Primary Fleet Productivity Estimates

Item Description Unit Waste LOW GRADE HIGH GRADE

Digger Configuration Type

Digger Make/Model

Truck Configuration Type

Truck Type

Mining Style Type Bulk Bulk BulkFlitch Height (m) 10 10 10

Weathering Type Fresh Fresh Fresh

Dry Density (t/bcm) 3.0 3.5 4.5

Moisture Content (%) 3% 3% 3%

Swell Factor (%) 25% 25% 25%

Wet Density Loose t/m3 2.5 2.9 3.7

Wet Bank Density t/m3 3.1 3.6 4.6

Maximum susp. Load (t) 18.0 18.0 18.0

Bucket weight (t) 12.4 14.4 18.5

Bucket Heaped Cap. (m3) 5.0 5.0 5.0

Fill Factor (%) 100% 93% 87%

Bkt Cap. Volume (bcm) 5.0 4.7 4.4

Bkt Cap. Weight (t) 7.5 7.5 7.5

Bkt Payload (t) 12.4 13.4 16.1

Suspended Load (t) 19.9 20.9 23.6

Suspended Load Factor (%) 110% 116% 131%

Bkt Cap. Weight (bcm) 3.0 2.6 2.0

Bkt Cap. Adopted (bcm) 3.0 2.6 2.0

Tray Capacity (SAE) (m3) 25.1 25.1 25.1

Trk Fill Factor (%) 87% 87% 87%

Body payload Capacity (t) 38.2 38.2 38.2

Volume Limit (KAC) (bcm) 17.5 17.5 17.5

Rated Payload (t) 43.2 50.4 64.8

Assumed Overload (%) 5% 5% 5%

Adjusted Payload (t) 45.3 52.9 68.0

Weight Limit (bcm) 14.0 14.0 14.0

Adopted Capacity (bcm) 14.0 14.0 14.0Min. Bucket Fill (%) 0.3 0.3 0.3

Calc Passes Per load Qty 4.6 5.4 6.9

Calc Passes Per load (rounded) 5.0 6.0 7.0

Actual Trk Load (bcm) 12.1 12.5 11.3

Actual Trk Load (t) 37.5 45.0 52.5

Dump Time (min) 1.5 1.5 1.5

Cycle Time (sec) 30 30 30

1st pass (sec) 12 12 12

Truck exchange (sec) 45 45 45

Additional Passes (sec) 120 150 180

Loading Time (min) 3.0 3.5 4.0

Max. Productivity (bcm/OH) 246.8 217.1 172.1

Productivity (t/DOH) 610 626 638

Annual Operating days (days) 360 360 360

Annual Calendar hours (hrs) 8640 8640 8640

Effective Utilisation (%) 60% 60% 60%

Productivity (Mtpa) 3.2 3.2 3.3

Shovel Details

Truck Details

Digger

Productivity

Load and Haul

Description

Excavator

CAT385C L

Rear Dump Haultruck

CAT 770

Material Details




Table 16-15 Annual Primary Equipment List

Item Units PP1 PP2 YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 YR10 YR11 YR12 YR13 YR14

HAULING

Cycle Time min 16.0 17.3 18.0 19.3 15.7 15.2 14.6 15.8 15.7 16.4 17.2

Haul Distance (one way) km 2.1 2.3 2.4 2.5 2.1 2.1 2.0 2.1 2.2 2.3 2.4

Truck Productivity t/hrs 152.6 141.1 133.3 125.5 147.1 154.4 160.2 150.5 153.0 148.1 144.5

Truck Operating Hours hrs 37,021 40,056 31,948 27,882 38,416 64,032 82,832 93,863 144,284 135,154 113,143

F'cast Truck Avail. % Contractor Fleet 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85%

F'cast Truck Util. % 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77%

F'cast Truck Eff. Util. % 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65%

Trucks Onsite units 7 8 8 8 8 12 15 17 26 26 26

Truck Productivity kt/unit 807 706 532 437 706 824 885 831 849 770 629

DIGGING

F'cast Digger Eff. Util. % 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60%

Excavator units 2 2 2 2 2 3 3 3 5 5 5

FEL units Contractor Fleet 2 2 2 2 2 2 3 3 5 5 5

Diggers Onsite units 4 4 4 4 4 5 6 6 10 10 10

Digging Productivity kt/unit 1,413 1,413 1,065 875 1,413 1,977 2,212 2,354 2,208 2,001 1,635

DRILLING

F'cast Drilling Eff. Util. % 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40%

Drills Onsite units Contractor Fleet 3 3 3 3 3 3 3 3 3 3 3

Drilling Productivity kt/unit 1,884 1,884 1,420 1,167 1,884 3,296 4,423 4,709 7,361 6,670 5,450

Item Units YR15 YR16 YR17 YR18 YR19 YR20 YR21 YR22 YR23 YR24 YR25 YR26 YR27 YR28 YR29

HAULING

Cycle Time min 16.8 17.0 17.6 19.1 20.9 22.8 24.7 26.2 27.1 27.6 29.3 30.7 32.7 34.5 36.3

Haul Distance (one way) km 2.4 2.5 2.6 2.9 3.1 3.4 3.6 3.8 3.8 3.7 3.9 4.1 4.3 4.5 4.6

Truck Productivity t/hrs 144.3 142.4 139.1 129.3 118.5 110.8 103.3 98.4 93.4 94.9 86.7 82.9 79.3 71.4 63.9

Truck Operating Hours hrs 150,893 150,579 170,539 170,020 174,286 160,069 136,532 105,419 91,429 80,687 76,350 67,302 59,435 41,730 22,039

F'cast Truck Avail. % 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85% 85%

F'cast Truck Util. % 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77% 77%

F'cast Truck Eff. Util. % 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65% 65%

Trucks Onsite units 27 27 31 31 31 29 25 19 17 15 14 12 11 8 4

Truck Productivity kt/unit 807 794 765 709 666 612 564 546 502 510 473 465 429 373 352

DIGGING

F'cast Digger Eff. Util. % 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60% 60%

Excavator units 5 5 5 5 5 4 3 3 2 2 2 2 1 1 1

FEL units 5 5 5 5 4 4 3 2 2 2 2 2 1 1 1

Diggers Onsite units 10 10 10 10 9 8 6 5 4 4 4 4 2 2 2

Digging Productivity kt/unit 2,178 2,144 2,373 2,198 2,294 2,217 2,350 2,074 2,135 1,914 1,655 1,395 2,357 1,490 704

DRILLING

F'cast Drilling Eff. Util. % 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40% 40%

Drills Onsite units 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

Drilling Productivity kt/unit 7,260 7,147 7,909 7,325 6,883 5,912 4,699 3,457 2,847 2,552 2,207 1,860 1,571 993 469




Table 16-16 Annual Ancillary Equipment List

Item Model PP1 PP2 YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 YR10 YR11 YR12 YR13 YR14

Graders 160K 2 2 2 2 2 3 4 4 7 7 7

Water Truck 20kL 1 1 1 1 1 1 2 2 3 3 3

Track Dozers D8T 4 4 4 4 4 5 6 6 10 10 10

Wheel Loader WA200 2 2 2 2 2 2 2 2 2 2 2

Fuel Truck 20kL 1 1 1 1 1 1 1 1 1 1 1

Roller 612H Contractor Fleet 1 1 1 1 1 1 1 1 1 1 1

Small Excavator 766 1 1 1 1 1 1 1 1 1 1 1

Crane Small 1 1 1 1 1 1 1 1 1 1 1

Service Truck 20kL 1 1 1 1 1 1 1 1 1 1 1

Lighting Plants 3 Burner 4 4 4 4 4 4 4 4 4 4 4Light Vehicles Landcrusier 10 10 10 10 10 10 10 10 10 10 10

Item Model YR15 YR16 YR17 YR18 YR19 YR20 YR21 YR22 YR23 YR24 YR25 YR26 YR27 YR28 YR29

Graders 160K 7 7 8 8 8 7 6 5 4 4 4 3 3 2 1

Water Truck 20kL 3 3 3 3 3 3 2 2 2 2 2 1 1 1 1

Track Dozers D8T 10 10 10 10 9 8 6 5 4 4 4 4 2 2 2

Wheel Loader WA200 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

Fuel Truck 20kL 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Roller 612H 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Small Excavator 766 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Crane Small 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Service Truck 20kL 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Lighting Plants 3 Burner 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

Light Vehicles Landcrusier 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10





16.5 Mine Schedule

The mine schedule incorporates an ore blending mining method of both HG and MG ore bodies. Themine plan targets an annual concentrate production rate of 509kt per year with a 3-year ramp up period ofkiln production. Mining is scheduled to operate for 360 days per year. The main mining operations includeoverburden stripping and storage for reclamation, waste rock removal to piles, and ore excavation andtransportation to the primary crusher at the processing plant.

A mining recovery rate of 100 percent has been adopted for both ore bodies. The main mine design andproduction criteria are outlined in Table 16-17. A planned stockpiling rate of 36,000 t/annum wasincorporated to account for plant outages.

Table 16-17 Mine Design and Production Target Criteria

16.5.1 Mine Development Sequence

The excavation of the open pits will be mechanized and advance in successive benches. Mine productionbegins in Gulçari A. Phase 1 provides the majority of the ore to the plant for the first 12 years ofproduction. The second pushback of Gulçari A (Phase 2) starts in production Year 8 and continues to theend of the project life (Year 29) providing the majority of production from Year 23 onward.

Between production years 12 and 22 the majority of the ore production will be provided by the satellitepits while Gulçari A Phase 2 stripping occurs. Primary ore production moves from Gulçari A to Gulçari ANorte, Gulçari B, São José, Novo Amparo and then to Novo Amparo Norte. Sustained ore production inYear 22 becomes available from Gulçari A Phase 2 while Novo Amparo Norte is active. At this pointproduction is balanced between Novo Amparo Norte and Gulçari A Phase 2 to provide higher grade feedto the mill.

Waste production rates fluctuate throughout the schedule as the satellite pits become ore bound limitingwaste mining. A strip ratio of 3:1 is targeted during Gulçari A Phase 1 production. Stripping is increasedas Gulçari A Phase 1 nears completion. The increased stripping is required to ensure that Gulçari APhase 2 stripping is completed to sustain ore production going forward once the ore from the satellite pitsis finished. This waste strategy ensures sustained mill feed throughout the project life.

Figure 16-9 through Figure 16-14 outline the mining progression of each pit. Figure 16-15 depicts thetotal material movement for the operation. Figure 16-16 shows the total material movement by pit for theoperation. Figure 16-17 outlines the LOM saleable vanadium product schedule.

Operation Units Stripping ROM Stripping ROM

Average Production Rate kt/annum 5,171 1,412 12,269 1,512

Kiln Rate kt/annum na 509 na 509

V2O5 % % na 1.32 na 0.99

Bench Face Angle degrees 70 70 70 70

Final Berm Width m 8 8 8 8

Maximum Bench Height m 20 20 20 20

Years 3 - 10 Years 11 - EOM





Figure 16-9 Mining Progression - Gulçari A

Figure 16-10 Mining Progression - Gulçari A Norte





Figure 16-11 Mining Progression - Gulçari B

Figure 16-12 Mining Progression - São José





Figure 16-13 Mining Progression - Novo Amparo

Figure 16-14 Mining Progression - Novo Amparo Norte





Figure 16-15 LOM Total Material Movement

Figure 16-16 LOM Total Material Movement by Pit





Figure 16-17 LOM Saleable Vanadium Product

16.6 Drill and Blast

The drill and blast requirements were defined by RPM based on global standards for drilling and blastingin Magnetite and Magnetite-Pyroxenite deposits. A summary of findings are outlined as follows:

The waste material Gabbro will be drilled and blasted with ANFO material and sequentiallyexcavated by hydraulic excavators and trucks;

The HG and MG ore materials, characterized by Magnetite and Magnetite-Pyroxenite, require ablend of ANFO and emulsion explosives to ensure safe and efficient mining;

Target production, given a blast size of 25 kt per blast, ranges from 3 to 6 blasts per week peractive pit;

The waste material will be drilled using a drill pattern of 2.5 m burden and 2.9 m spacing sufficientto support a safe blast of 5 m bench height;

Explosive powder factor is an industry economic measurement of the amount of explosive required

to fracture a cubic metre of rock. The powder factor for waste was based on 0.45 kg per m3 forwaste material using ANFO product. For ore, a powder factor of 0.59 kg per m3 was applied,incorporating a 70% ANFO and 30% Emulsion blend. These assumptions are congruent withsimilar deposits around the globe;

The ore will be drilled using a drill pattern of 1.8 m burden and 2.1 m spacing sufficient to support asafe blast of 5 m bench height;

Hole length 5.5 m including 0.5 m subdrill;





Total hole length of 5.5 m consists of a stemming height 1.7 m, and a column charge 3.9 m will beused for both ore and waste;

Three track-mounted drilling machines, equipped with pneumatic drills capable of drilling 3.5 inchdiameter holes in ore and 5.5 inch diameter holes in waste, will be required;

During Year 1, a drill and blast study to ascertain the optimal drilling and blasting parameters willbe focused on increased drill bit size, delay timing, and explosive powder factors for all material

types.

16.7 Mine Safety and Rescue

The Maracás Project will use international standards of “Loss Control Management” for personnel training

and control all activities related to environmental health and safety according to Brazilian labour laws, asoutlined below.

SESMT, a service specialized in safety and occupational medicine, will provide a full time safetyassociate and a part time physician. They will be legally responsible for the health and safety ofthe workers in the entire Project site;

The operation will be fenced and supported by a guard house to control and deter the circulation ofpersonnel unrelated to the operation. Internally, signs will warn about possible hazards;

Operational equipment will be inspected periodically. Each piece of equipment will have a loggingcard to record maintenance and other services performed. Supplier instructions will be required tobe followed. Portable fire extinguishers will be strategically distributed around the mine site;

Special attention will be given to the storage, transportation, and handling of explosives. Allregulatory codes and legal requirements will be followed, including the required licensing by the

Army Bureau and the Police. The handling of explosives will always be done and supervised by aregistered professional blaster;

A dust management system will incorporated within the operation; The safety of the mining activities will be the responsibility of a registered licensed mining engineer; Given that the mine site is located a long distance from municipal medical services, SESMET will

have an ambulance and one safety attendant to assist in emergencies. A system will be in place toensure there is a minimum of two employees trained in first aid and immobilization procedures forevery shift.





17 Recovery Methods

The process flow sheet selected for the Maracás expanded process plant comprises three stages ofcrushing, one stage of grinding, two stages of magnetic separation, magnetic concentrate roasting,vanadium leaching, ammonium metavanadate (AMV) precipitation, AMV filtration, AMV calcining to V 2O5

flake as final product. From Year 3, part (up to 65%) of the V 2O5 produced will be smelted to produceferrovanadium (FeV).

The plant (originally sized to process 960,000 t/a ROM) will be capable, after due modification, to process1,400,000 t/a of feed ore with an average grade of 1.10% V2O5. The plant has an operating regime of365 d/a, 7 d/wk, 24 h/d and plant utilization of 89%, resulting in an average nominal hourly throughput of180 t/h. The plant will produce an equivalent of 11,400 t/y V2O5, (LOM average), and from Year 3 willpartly convert to production of 3,830 t/a (LOM average) of vanadium as ferrovanadium equivalent to 65%of total V2O5 from Year 5.

Recovery for V2O5 is estimated to be 72.5% while an overall average recovery of 68.4% V is expected forferrovanadium production over the life of mine.

Table 17-1 provides a summary of key process design parameters used for the process design.Processing recoveries vary between the different ore deposits and are outlined in Table 17-2. Sinceblending of magnetite and magnetite-pyroxenite bearing material is planned to occur at the crusher, themagnetic recovery factors calculated are based on blended HG an MG grades.

17.1 Process Description

A simplified process flow diagram for the production of vanadium pentoxide is presented in Figure 17-1. Figure 17-2 outlines the process flow for the production of ferrovanadium.

17.2 Crushing and Grinding

The ore is crushed via a three-stage crushing circuit comprising a primary jaw crusher, a secondary conecrusher and a vibrating sizing screen. The fine crushed product is fed to a 4,200-m³-capacity stockpilefrom which it is withdrawn at a controlled rate to feed the grinding circuit. The grinding mill grinds to aproduct size of minus 150 microns.

17.2.1 Magnetic Separation

The vanadium is contained within the magnetite fraction of the resource. Magnetite is recovered by usinglow intensity magnetic separators. The Maracás beneficiation circuit comprises crushing, screening andgrinding followed by magnetic separation.

17.2.2 Concentrate Dewatering and Tailings Disposal

The final magnetic concentrate will be thickened and filtered. The filter cake will be fed to the roastingsection of the plant. The nonmagnetic tailings fraction from the beneficiation plant will be thickened,filtered and conveyed to the tailings storage area for dry stacking.





Table 17-1 Summary of Key Process Design Criteria

(1) Includes Kiln Recovery of 87.5% in line with actual operations at the Rhovan and Vantech plants and 97%

leaching efficiency. Values chosen in consultation with Les Ford from actual similar operations, rather than

the pilot results that displayed lower recoveries.

Table 17-2 Summary of Processing Recoveries

* Recoveries calculated on a blended HG and MG material basis.

Criterion Units Value

Nominal ore processing rate t/a 1,400,000

Average production of vanadium oxide) t/a 11,400

Average V2O5 head grade % 1.10

Ore specific gravity 4.3

Plant availability % 89

Plant operating hours h/yr 7 800

Average p lant dai ly ore throughput t/d 4,320

Number of crushing stages 3

Crusher product size (80% passing) mm 7.5

Number of grinding stages 1

Grind product size microns 150

Bond ball mill work index (metric) kWh/t 12.9

Magnetic product solids yield % 34

V2O5 recovery to magnetic concentrate % 88.4

Average m agnetic concentrate V2O5 content % 2.88

Roasting reaction zone residence time h 1

Leach retention time h 1

Average roasting/leach V2O5 conversion(1) % 84.8

Estimated AMV Precipitation V2O5 recovery % 97.5

Estimated conversion to Ferro vanadium –

vanadium recovery% 94.5-95.7

Total average estimated recovery to V2O5 % 72.5

Total estimated vanadium recovery to FeV % 68.4

Average ferrovanadium production (LOM), as V t/a 3,830


Magnetic Recovery * % 91.00% 77.85% 89.17% 77.94% 86.58% 79.47%

Scavenger Recovery % 1.00% 1.00% 1.00% 1.00% 1.00% 1.00%

Concentrate to V2O5 Yield % 81.93% 81.93% 81.93% 81.93% 81.93% 81.93%

V2O5 recovery % 75.38% 64.60% 73.88% 64.68% 71.75% 65.93%

V2O5 to V in FeV Recovery % 94.50% 94.50% 94.50% 94.50% 94.50% 94.50%

Stoichiometry Recovery % 56.00% 56.00% 56.00% 56.00% 56.00% 56.00%

V in FeV Recovery % 39.89% 34.19% 39.10% 34.23% 37.97% 34.89%




Figure 17-1 Conceptual Process Flow sheet – Vanadium Pentoxide




Figure 17-2 Conceptual Process Flow sheet – Ferrovanadium





17.2.3 Roasting

The wet magnetic concentrate filter cake, containing approximately 2.88% V2O5, will be roasted at hightemperature (+1,150 °C) in a rotary kiln. In order to extract the vanadium from the magnetite particles,

this roasting process needs to be undertaken in the presence of a sodium flux (sodium carbonate and/orsodium sulphate) to form a water soluble salt, (sodium vanadate). The wet magnetic concentrate will beblended with the sodium salts prior to roasting.

An off-gas control system will collect any dust entrained in the gas from the roaster. To meet localenvironmental regulation, an electrostatic precipitator will be installed to remove such particulates. Thequantity of sodium sulphate added to the kiln will be controlled and reduced in order to ensure compliancewith the emission limits for SO2. Since the sodium sulphate dosage was already maximized for the basecase, additional sodium carbonate will be added to the kiln in the expanded case to ensure efficientextraction of vanadium and the excess sodium sulphate produced in the evaporator will be stockpiled ona sealed area.

In the testwork completed by SGS, both sodium carbonate and sodium sulphate were used as the saltreagent. No obvious recovery differences were noticed in the results. Consequently, in the current plantdesign, recycled sodium sulphate is being used, together with sodium carbonate as sodium salt makeup.The ratio of these two salts will be carefully controlled in order to remain below the allowable SO2 limits.

Following roasting and cooling, the calcine will be fed to the leaching process.

17.2.4 Leaching

The hot calcine, at about 400°C, will be conveyed, by means of a metal conveyor, to a regrinding mill,where it will be further cooled by mixing with process water. The large lumps that form during calcining

will be broken down in this mill. Vanadium salt contained in the roasted calcine will be leached in twoagitated tanks installed in series. The total leaching time will be one hour.

The leach discharge will be then filtered and washed using a vacuum belt filter. The pregnant solutioncontaining approximately 90 g/L V2O5 is pumped to a desilication stage, which is designed to removesoluble silica in the solution. The leach residue will be removed from the leaching system and conveyedto a dedicated storage area. The residue contains approximately 60% Fe and 5% TiO2 and will be soldas an iron ore byproduct.

The desilication is achieved in two agitated tanks in series with the addition of aluminum sulphate andsulphuric acid. The silica in the solution will be reduced from about 200 mg/L to less than 30 mg/L. Afterdesilication, the solution will be pumped to a filter where the solids will be removed and sent to disposal

along with the calcined leached tailings. The filtrate, pregnant leach solution (PLS), will be pumped to theprecipitation stage.

17.2.5 Precipitation

The clear leach liquor will be pumped to another series of agitated tanks, where the vanadium isprecipitated as ammonium metavanadate (AMV), with the addition of ammonium sulphate. Theprecipitate is filtered and fed to a dryer prior to being calcined to produce V2O3 powder. The barren





solution, which contains sodium and ammonium sulphate salts, will be treated to recover sodium sulphatepart of which is recycled to the roasting stage.

17.2.6 AMV Drying

Wet AMV solids will be dried in a flash dryer. The dried AMV will be calcined under oxidizing conditions to

produce V2O5 and then melted and cast into flakes for sale with a portion of it being converted toferrovanadium from Year 3.

17.2.7 Ferrovanadium Production

Ferrovanadium (FeV), which is an alloy of vanadium metal and iron, is typically produced by theconversion (aluminothermic reduction) of V2O5 as is proposed at Maracás. The following simplifiedreaction present the aluminothermic process used to produce ferrovanadium from V2O5.

(a) 3 V2O5 + 10 Al = 6V + 5 Al2O3.

Lime, iron scrap and aluminum prills are also used in the production of FeV. V2O5 flake as well as the

other ingredients are mixed in stoichiometric amounts and discharged into a refractory lined pot. Thecontents of the pot are ignited under an extraction hood with a fuse of V2O5 and aluminum powder. Thereaction takes place autogenously and the FeV settles to bottom of the pot and the liquid slag above it.The pot is allowed to cool and inverted to allow the ferrovanadium ingot and the slag to separate. Theslag is discarded and the ferrovanadium is crushed and screened, and packed to customer’s

requirements.

Product storage bins will be provided, from where products are packaged into steel drums and transferredto the product warehouse.

17.2.8 Sodium Sulphate Production

A crystallization circuit has been designed to recover sodium sulphate salt from the barren leach liquor.

The barren liquor, which contains ammonium sulphate, sodium sulphate plus small amounts of dissolved AMV and impurities, is first concentrated by evaporation to a predetermined ammonium sulphateconcentration. This allows anhydrous sodium sulphate to be crystallized from the concentrated solution.The resulting crystalline product is then centrifuged, dried and recycled to the roasting stage as part of thesodium flux.

The concentrated mother liquor from this process contains approximately 24% ammonium sulphate and21% sodium sulphate. The mother liquor is cooled and recycled back to precipitation as a source ofammonium sulphate. Approximately 2.9 tph of this slurry is purged to a sealed purged dam as a mean of

controlling chlorides build up in the circuit.





18 Infrastructure

The infrastructure requirements for the project are summarized in the following sections and areincorporated in the capital cost estimate for the project.

18.1 Process Water

A water main, about 29 km long, will be installed in order to meet the beneficiation plant and themetallurgical plant process water requirements. This pipeline connects the water intake at Rio de Contasto the plant raw water tank using two centrifugal pumps (one operational and one stand by).

The water pumped from the Rio de Contas will be used primarily by the process plant. Once the plant isin operation, water recovered from the water reclamation circuit will also be used. The thickeners will besized to contain the reclaimed water. A water demineralizing unit and a cooling tower will be installed totreat the water for equipment cooling.

The water taken from Rio de Contas will be stored in a concrete tank with a total volume of 2800 m³. Theraw water tank will contain enough water for 12 hours of plant operation. This tank will also contain apermanent water reserve of 240 m³ for firefighting purposes, which is in accordance with the laws ofBahia State and the National Fire Protection Association. The water reserve is enough for 2 hours offirefighting.

A concrete process water tank, with a designed volume of 260 m³, will be built at the process plant forwater storage. This tank will store recovered water from the thickeners. In case make-up water isneeded, it will be supplied from a centrifugal pump installed at the raw water tank.

18.2 Potable Water

A water treatment plant, with a capacity of 8 m3/h, will be installed for potable water and gland waterdistribution.

A concrete tank for potable water, with a volume of 220 m³, sufficient for 24 hours of consumption, will beprovided.

18.3 Sewage Treatment

Two sewage treatment plants will be used to process sewage from the industrial areas. These plants arecompact in size and the treated effluent will be used for wetting gardens and dust control on roads. Bothplants will be built during the construction phase.

18.4 Fuel and Lubricant Storage and Distribution

Diesel fuel will be delivered to the site by road tankers and will be offloaded into the fuel farm from whereit will be pumped to various areas for use in the mine. Diesel fuel distribution will be limited to loading andunloading facilities and metering equipment at the diesel fuel tank.





Lubricants will be delivered to the site in drums. The drums will be stored in a secure area in accordancewith relevant state regulations. The lubricants will be distributed to hose reels in the truck shop servicebay with barrel pumps.

18.5 Air

A screw compressor will supply high-pressure air for instruments, plant general use and tanks. Arefrigerant air drier and filters will be supplied in order to ensure that instrument air will be of good quality.The compressor will be appropriately stored in a compressor building.

18.5.1 Air Emissions and Air Quality Monitoring

The air monitoring system will provide information on which to base the strategy for managing airemissions. The project anticipates the installation of three air monitoring stations, one within theplant, one upstream, and one downstream of the industrial area. These stations will make continuousmonitoring of SO2, NO2, and MP totals. Along with these stations there will be continuous analyzers forSO2 and NO2 in the chimney stack, which will ensure that emissions remain within the allowable limits

permitted by environmental legislation.

18.6 Heating

A complete system, consisting of one fire tube boiler, with a capacity of 2600 kg/h, at an operationalgauge of 10 kgf/cm² and temperature of 180ºC (saturated steam), burners operating with fuel oil B-type(low sulphur grade), fuel oil storage tank, day tank, heat exchangers for heating the fuel oil and all theother required devices, will be required.

18.7 Power Supply

The electrical power requirements for the plant, including the beneficiation, hydrometallurgy and installed

utilities were estimated to be approximately 11 MW. In order to fulfill this demand, the power supply willbe provided at 138 kV, 60 Hz, through a transmission line 85 km long from Coelba’s Ibicoara Substation.

A step-down substation of 13.8 kV will be installed at the plant. Initially, a power transformer of 13.8 kV15 / 20 MVA will be installed. There is sufficient storage space for a second power transformer to beinstalled in the future.

The 13.8-kV power distribution system for the plant will be supplied by means of isolated cables orconventional aerial cable system. Substations will be designed for meeting the requirements of theconcentrator, hydrometallurgy plant, crushing and supporting areas.

The power required at the water pumping intake at Rio de Contas will be supplied by a substation to befed by the local supplier company at 13.8 kV.

All electrical distribution will be in cable trays using armor interlocked PVC coated cables. The processand plant site ancillary facilities switchgear and electrical equipment will be installed in modular electricalrooms adjacent to, or within, their respective buildings.





18.8 Buildings

The administration building will be of a single-storey prefabricated panel construction and will includegeneral areas for engineering, geology, administration personnel, offices for the general manager, minemanager, plant manager, administration staff, chief engineer, chief geologist, EH&S and medical careroom.

The architectural and constructability design of these buildings will consider the climate characteristics,environmental comfort, ergonomics, durability, standards and codes suitable for a Project of this size.The construction system consists of reinforced concrete pillars, beams and slabs with concrete blocksmasonry walls finished with mortar layer and painting, tiles or vinyl floor and metallic roofing.

A general layout of the of the plant facility and office buildings etc. is shown in Figure 18-1.




Figure 18-1 Plant Site Layout





18.9 Assay Laboratory

A fully equipped assay laboratory will be included on the plant site. The laboratory will perform dailyanalysis of mining and process samples. The laboratory will be a single-storey structure located close tothe warehouse.

18.10 Miscellaneous Buildings

A main gatehouse will be located at the entrance to the plant site. This building will be a single-story blockwork structure. An additional security post will be located at the entrance to the process plant. A first aidpost, equipped for “first response”, will be located in the vicinity of the main gatehouse.

18.11 Explosives Magazine

The bulk explosives magazine will be located away from the mining facilities, at a location to be decidedlater. Raw materials, such as ammonium nitrate, fuel oil and primary explosives used in the explosivemanufacturing process will be brought to the site by road, with the explosives being stored in explosive

magazines until required.

A detonator magazine and the industrial explosives magazine will be required as separate units situatedwithin close proximity to the mine or industrial area, and enclosed within a security fence. Location andconstruction of the magazines will follow the requirements of the Brazilian Army.

18.12 Communications

The Maracás project will be connected to the public communication system through telephone andinternet services. An allowance has been made in the plant and mining costs for an adequate supply ofhand held and vehicle mounted radio sets.

RPM recommends the fitting of in-vehicle radios to all of the hydraulic excavators, drill rigs, graders andwheel loaders, to the majority of the haul trucks and to the service vehicles and supervisors transport toassist in safe operations at the project site.

An automated emergency call system will be installed to be able to call directly to the hospital, clinic, orpolice.

A Wi-Fi system that will be utilized with the mine mobile dispatch system will be installed at the mine toprovide a messaging service between the dispatch center and the fleet of mobile equipment.

18.13 Roads

The on-site roads, from the main reception to the primary crushing plant at the beneficiation plant areconsidered to be industrial roads. Some studies were developed for improvements to the existing publiccounty road, which has a length of about 42 km, between the BA-026 crossing the Maracás Project areaand the village of Porto Alegre. Largo has been engaged in discussions with the Bahia State trafficagency who are considering a proposal by Largo that the road be paved by the Bahian State governmentin partnership with the company. Negotiations on this issue have not been concluded. However, the





current road is being upgraded by the minor improvements needed to ensure timely and efficienttransportation of goods, supplies and workers.

18.14 Tailings Facility

Three types of tailings will be produced and stored in the following facilities:

Leached Calcine Tailings Dump (to be sold as iron ore byproduct) Chloride Purge Tailings Pond Non-Magnetic Tailings Dump

The leached calcine tailings and non-magnetic tailings dumps will be constructed using a “dry stacking”

method.

All tailings facility characteristics and designs included in this report relate to the previous productionscenario as laid out in the Technical Report “Technical Report for the Largo Maracás Vanadium Project, 1

Million Tonnes per Year Processing Plant, Brazil”, issued Dec 21, 2012.

The same tailings management strategies will be employed for this expanded production scenario toaccommodate required tailings storage for life of the project, including production from the Gulçari APhase 2 pit and all satellite pits. Further study is required to determine the new design characteristics ofthese expanded facilities.

18.14.1 Tailings Disposal Ponds

The tailings generated by the beneficiation process and the vanadium ore processing plant originate fromthe following process:

leached calcine from the processing kiln discharge.

filter cake from the desilication process. ferrovanadium slag. chloride control purge from the evaporation circuit. primary inert tailings from magnetic separation.

The leached calcine tailings, desilication process filter cake and ferrovanadium slag will all be dischargedinto the Leached Calcine Tailings Dump. The Leached Calcine Tailings Dump will be constructed with alife of 2 years and capacity of 220,000 m3. The tailing from the Leached Calcine Tailings Dump has thepotential for commercialization of an iron ore concentrate byproduct, as has been assumed in this report.

The chloride control purge tailings from the evaporation circuit are to be deposited in the Chloride PurgeTailing Pond. The Chloride Purge Tailings Pond will be built at the outset to accommodate 3 years of

tailings production. The pond will later be expanded, possibly in a different location, if required, toaccommodate LOM tailings.

The Non-Magnetic Tailings Dump is designed to receive the primary inert tailings originating from themagnetic separation after thickening and filtering. The proposed tailings containment system consists of aseries of reservoirs formed by rock-fill structure and sealed by compacted clayey / liner on the upstreamside. This dump will be constructed with a capacity of 630,000 m3 and will be monitored for leachates in





the first three years of the operation. If no leachates are considered to be hazardous, the tailings will bestored in an unlined facility for the balance of the life of the mine.

The dikes to form the ponds will be built using compacted earth (residual soil / saprolite) and their baseareas will be lined using double-layer geomembrane liner featuring a leak detection system. Theconstruction sequence consists of clearing vegetation from the areas to be occupied by the ponds,

removal of organic material and excavation of material inappropriate for foundations. The entire perimeterof each pond will be protected by rock-fill channels.

18.15 Waste Management

Solid waste generated from the mine plant site, including ancillary buildings, will primarily be domesticand industrial non-hazardous waste. A comprehensive Waste Management Plan will be developed for theproject. Solid waste will include:

refuse from construction (scrap wood, metal, concrete, etc.); refuse from the mine (empty drums, packing materials, etc.); and

general domestic garbage from the offices and ancillary buildings (paper, refuse, food, etc.)

Construction debris, inert waste and used tires will be placed in designated cells and proper use will bedefined. Solid domestic and industrial waste from the mine plant facilities will be recycled and re-used inan approved manner, where feasible. Other solid waste will be placed in waste receptacles.





19 Market Studies and Contracts

19.1 Marketing and Market Research

CPM Group (“CPM”) of New York, U.S.A. was commissioned by Largo to prepare a confidential reportentitled “Vanadium Industry Outlook” dated July 30, 2008. Since this, no reports have been specifically commissioned on Largo’s behalf. Micon understands, however, that Largo subscribes to severalassociations, publications and industry related websites, as listed below.

Associate Member of Vanitec (www.vanitec.org)o A global organization which encourages and assists technical research and education about

vanadium and its world-wide application in the steel, titanium and chemical industries.Vanitec convenes international representatives of companies involved in the mining,processing, manufacturing, research and use of vanadium and vanadium-containingproducts.

Subscription to Metal Bulletin (www.metalbulletin.com)o An industry accepted source for news, pricing information, expert market commentary and

statistics. Metal Bulletin publishes pricing information for over 900 metals, includingvanadium.

Subscription to Metal Pages (www.metalpages.com)o An industry accepted source for metal prices, news, conferences and information for non-

ferrous metals, rare earths and ferro alloys. Roskill Information Services - “Vanadium: Global Industry Markets & Outlook”

o Report includes data and information relating to: Vanadium sources and resources, globalproduction and consumption, supply demand outlook, historical and price forecasts, a reviewof production by country, uses of vanadium and an overview of the international market

o Roskill is considered a leader in independent, international metals and minerals research,producing 75 market reports, databooks and newsletters designed for the purposes offormulating company strategies, following industry trends, competitor analysis, and gaining acomplete overview of a single industry.

Attendance at various industry related conferences and eventso Byron Capital Markets Electric Metals conferenceso American Metals Market, Steel Success Strategies Conferenceso Metal Pages Titanium Alloys Conferences

19.2 The Market for Vanadium

Vanadium is recovered principally from magnetite and titano-magnetite ores, either as the primaryproduct or as a co-product with iron. It is also recovered as a secondary product from fly ash, petroleum

residues, alumina slag, and from the recycling of spent catalysts used for some crude oil refining andwhich have accumulated vanadium.

Roskill estimates that 80% of vanadium comes from mined ores and the balance from secondary sources.(Roskill, 2010).

Vanadium pentoxide is the principal intermediate product from treatment of magnetite ores, vanadiferousslags and secondary materials. It is used directly in non-metallurgical applications and in the production





of a range of vanadium chemicals. It is also the starting material for production of ferrovanadium andmaster alloys. Most vanadium is used in the form of ferrovanadium as a steel additive.

Production and demand figures may be reported in terms of contained vanadium metal or the pentoxide(V2O5) equivalent. Trade statistics are reported in terms of gross weight.

World production since 2007 reported by the U.S. Geological Survey (USGS) is summarized in Table19-1.

Table 19-1 World Vanadium Primary and Co-product Output(Tonnes contained vanadium in ores, concentrates and slag)

1 Estimated.

U.S. Geological Survey, 2010 and 2011 Minerals Yearbooks and Mineral Commodity Summaries, January, 2013.

Comparable statistics for ferrovanadium output by country are not readily available.

19.2.1 Demand

It is estimated that over 90% of vanadium production is used in the steel industry in a wide range of steelformulations to meet a variety of end-use applications. It is also used in titanium-aluminum and other

non-ferrous alloys, in catalysts for the production of maleic anhydride and sulphuric acid and petroleumcracking, in batteries and in a number of chemical applications.

Vanadium consumption trends reflect the general trend of steel making and production of high strengthsteel, in particular. In turn, conditions in the steel industry are affected by global economic conditions.Table 19-2 shows crude steel output in the five largest producing countries, and the world total, from2006. The increasingly dominant position of China in the steel industry is shown clearly and, in 2012, itaccounted for approximately 47% of world crude steel production. Although production faltered in 2008,and dropped sharply in 2009 as a result of the global economic crisis, recovery was rapid in 2010 andcontinued through 2012.

2006 2007 2008 2009 2010 2011 2012(1)

China 17,500 19,000 20,000 21,000 22,000 23,000 23,000

Kazakhstan 1,000 1,000 1,000 1,000 1,000 1,000 1,000

Russia 15,100 14,500 14,500 14,500 15,000 15,200 16,000

South Africa 23,780 23,486 20,295 14,353 22,606 22,000 22,000

United States - - 520 230 1,060 590 270

Total 57,400 58,000 56,300 51,100 61,700 61,800 62,300





Table 19-2 World Crude Steel Production(Thousand tonnes)

World Steel Association, www.worldsteel.org

Vanadium increases the strength of a variety of steels by forming carbides and nitrides. High strengthlow-alloy (HSLA) steels are estimated to account for about 40% of use of vanadium in steels which are

used in oil and gas pipelines, in a range of automotive components, for pressure vessels and reinforcingbars (rebar) for concrete. In these applications, HSLA steels provide increased strength and weldabilityand reduced weight compared with other steels. Full alloy steels are the second largest market forvanadium in the steel industry, followed by tool and carbon steels.

As noted above, most vanadium is used in the form of ferrovanadium as a steel additive. The threeprincipal ferrovanadium grades contain 40% V, 60% V and 80% V. The 60% V and 80% V grades areproduced by reduction of the pentoxide (V2O5) or trioxide (V2O3), generally using the aluminothermicprocess. Lower grade ferrovanadium is generally produced by reduction of slag or other vanadium-containing feedstocks by the silicothermic process.

Titanium alloys are the principal non-ferrous alloys using vanadium. These have high strength to weightratios and are used in aircraft components, including structural elements, hydraulic systems and jetengine parts. The vanadium used in the form of vanadium-aluminum master alloys.

19.2.2 International Trade

In 2011, South Africa and China were the leading exporters of ferrovanadium at approximately 10,600 t(gross weight) and 6,800 t, respectively. Other significant exporters were Australia and the CzechRepublic at over 4,500 t each. The principal importers in 2011 were Germany and Japan, importingapproximately 5,300 t and 4,800 t, respectively. (United Nations COMTRADE database, DESA/UNSD).

19.2.3 Vanadium Prices

Ferrovanadium and vanadium pentoxide are the principal commercially-traded vanadium products.Neither these, nor any other vanadium products are traded by means of an exchange or terminal marketsuch as the London Metal Exchange or COMEX Division of the New York Mercantile Exchange(NYMEX). Prices for ferrovanadium and vanadium pentoxide are quoted in publications including MetalBulletin, Ryan’s Notes and Metal Pages (www.metalpages.com).

2006 2007 2008 2009 2010 2011 2012

China 421,024 489,712 512,339 577,070 637,400 683,265 708,784

India 49,450 53,468 57,791 63,527 68,321 72,200 76,715

Japan 116,226 120,203 118,739 87,534 109,599 107,595 107,235

Russia 70,830 72,387 68,510 60,011 66,942 68,743 70,608

United States 98,557 98,102 91,350 58,196 80,495 86,247 88,598

Others 492,904 513,130 492,476 389,503 465,954 472,010 458,283

Total 1,248,991 1,347,002 1,341,205 1,235,841 1,428,711 1,490,060 1,510,223

http://www.worldsteel.org/








Transactions are usually negotiated under six- or 12-month contracts between producers and consumersor trading houses. Prices are generally quoted in terms of US dollars per pound or per kilogram grossweight for vanadium pentoxide containing 98% V2O5 while ferrovanadium prices are quoted in terms ofUS dollars per kilogram of contained vanadium.

Figure 19-1 illustrates the trend in ferrovanadium prices over the past five years from January, 2008.

Prices reached all-time highs in the second quarter of 2005 and, as shown in Figure 19-1, peaked againthrough the early part of 2008. Between mid-2010 to end-2012, ferrovanadium prices declined fromaround USD 30/kg V to around USD 25/kg V. As of the end of December, 2012, prices quoted by MetalBulletin were USD 27.50 to USD 28.40/kg V.

Figure 19-1 Past Ferrovanadium Price Trend(USD/kg V, high in red, low in blue)

Metal Bulletin data provided by Largo.

The price peak in 2005 reflected conditions of strong demand coupled with rapid drawdown of inventorythat had been built up in the early-2000s. In the early part of 2008, high prices reflected concerns overpower generation capability affecting the South African ferrovanadium industry and the effects of anearthquake in China on ferrovanadium production. Roskill notes, however, that the vanadium industry issignificantly larger than in it was 10 years ago and that a higher proportion of transactions are directly

between producers and consumers based on long term supply agreements, both of which reduce theimpact on prices of short term changes in the balance between supply and demand (Roskill, 2010).

19.2.4 Price Outlook

Over the short term, world economic growth is expected to be moderate but supported by relatively strongconditions in Asia compared with most high income countries. The World Steel Association reported thatglobal apparent steel use will increase by 2.1% in 2012 compared with 6.2% achieved in 2011 largely as

0

10

20

30

40

50

60

70

80

90

100

11/1/08 20/3/09 28/5/10 5/8/11 10/10/12





a result of weaker conditions in the second half of the year. The association projects that demand willimprove by 3.2% in 2013 to reach a record 1.455 Mt. (World Steel Association, October, 2012).Vanadium demand may be expected to follow this trend and then to grow somewhat more rapidly asglobal economic conditions improve.

Given the relative volatility of vanadium prices over the decade to the end of 2009, it was decided that the

three-year average from mid-January, 2010 to mid-January, 2013 reflect relatively stable marketconditions that represent a reasonable basis for projected prices used in this study.

Analysis of price data published by Metal Bulletin and provided by Largo yielded the averages shown inTable 19-3.

Table 19-3 Three-year Average Prices for Vanadium Pentoxide and Ferrovanadium

Metal Bulletin data provided by Largo.

19.3 Contracts

On 14 May, 2008, Largo announced that it had entered into an off-take agreement with GlencoreInternational AG (Glencore) for all vanadium products from the Maracás project. The agreement remainsin effect for a six-year period following the start of commercial production and may be renewed for afurther six-year term. The agreement includes the provision for a discount from market vanadium prices

in consideration of marketing and logistical services to be provided by Glencore. For the term of theagreement, Largo will not need to undertake its own marketing efforts.


(at Jan 17, 2013) (at Jan 17, 2013)

Vanadium Pentoxide 6.37 14.04 6.47 14.28

USD/kg V

(at Jan 17, 2013)

Ferrovanadium - 28.01 32.00 -


US$/kg V





20 Environmental Studies, Permitting and Social or Community Impact

20.1 Regulatory Framework Overview

Environmental permitting in the State of Bahia is the responsibility of INEMA - Instituto do Meio Ambiente,which is the institution that regulates, approves and issues environmental permits or licenses.

The permitting process in Bahia takes into consideration the nature and size of the projects and activitiesunder consideration, the characteristics of the affected ecosystem and the supporting capacity of the areabeing impacted.

The following types of environmental licenses are necessary for the Project:

Location License (LL): The LL is granted in the preliminary planning phase of the project or operation,and it approves the location and the conceptual design of the project, attesting its environmentalfeasibility and determining the basic requirements and conditions to be observed in the subsequent

permitting stages;

Installation License (LI): The LI is granted so that the project or operation can be installed (orconstructed), in accordance with the specifications presented in the plans, programs and projectspecifications proposed by the environmental studies that were approved, including the environmentalcontrol measures and other conditions;

Operational License (LO): The LO is granted for the project to commence the operational phase, afterthe fulfillment of all the requirements of the previous licenses has been confirmed and the conditions andprocedures to be observed during the operation are defined.

The licenses and authorizations are granted based on an analysis of the environmental studies that have

been completed. This analysis considers the objectives, criteria and norms for the conservation,preservation, protection and improvement of the environment, the possible cumulative impacts and theplanning and land use guidelines of the State. For mining projects that are exceptionally large, as it is thecase of the Largo Project, the preparation of the EIA and RIMA must comply with the TR (ReferenceTerm Sheet) issued specifically for the project.

The environmental licenses have a definite validity or term, which can be renewed or extended, based onthe nature of the project and activities. The validity is defined for each license and is stated on theenvironmental certificate issued; the validity period starts on the day the license is published in the OfficialNewspaper of the State of Bahia. If the renewal of any license is requested 90 days or more in advanceof its expiration date, the validity of that license is automatically extended until INEMA issues a formalresponse, positive or negative.

For permitting purposes, the classification of mining operations in Bahia is divided in five categories:micro, small, medium, large and exceptionally large. These categories are determined based on threecriteria: built area, total investment (capital investment + cash flow, in Brazilian Currency, $R), andnumber of employees. The project is classified based on the highest ranking of the three criteria. TheMaracás Vanadium Project is classified as exceptionally large.





The grant or license for water resources usage in the State of Bahia is governed by the State Decree N.6,296, from March 21st, 1997; State Law N. 10,431, from December 20th, 2006; and Federal Decree N.24,643, from July 10th, 1934 (Código de Águas).

The water grant is obligatory for the lawfulness and legitimacy of any usage of water resources whoseobjective is the construction, expansion or alteration of any project that requires surface or groundwater,

as well as for any work that alter the water regime, quantity or quality.

The grant, as a concession, (as it is the case of the Largo Project) has a maximum validity of 30 yearsand it is renewable. The state water grants are issued by the SRH (Secretaria Recursos Hidricos),through specific publications in the Official Newspaper of the State of Bahia. The request for water grantsin mining involves the execution of hydrologic, hydrogeological and hydro chemical studies.

In the case of a long drought or water shortage, the water grants can be altered, in order to assure thathuman supply has priority.

In the State of Bahia, the SFC (Secretaria de Conservação de Florestas) has the duty of issuingauthorizations for native vegetal suppression, which are necessary to alter the land use for the installationor expansion of mining operations. The authorizations are only conceded once the environmental,technical and economic feasibility of the project is proved. The administrative process involving theauthorization for vegetal suppression must be conducted by the SFC, based on a specific TR (ReferenceTerm sheet). The documents to be submitted include the PTSV (Technical Project for VegetationSuppression) and the Forest Inventory, as required by the SEMARH Norm 29/05, as well as the PRAD(Plan Rehabilitation of Degraded Areas).

The CONAMA Norm 369/2006, from March 28th, 2006, defines extraordinary cases, in which thecompetent authority may authorize intervention or vegetal suppression in APP (Area of PermanentPreservation), for implementation of projects, plans and activities that result in significant public benefits.

The clause I, letter C, from article 2 of the Norm 369/06 explicitly recognizes the public benefits ofminerals exploration and extraction granted by the competent authority, except sand, clay, silt and gravel.

The section II, article 7 of the norm deals specifically with the activities of mineral exploration andextraction in order to obtain environmental licenses. It is understood, therefore, that once the guidelinesand obligations defined by the CONAMA Normative 369/2006 are observed, the public benefits of theMaracás Project are automatically recognized for environmental permitting purposes and for purposes ofobtaining authorization for vegetal suppression and intervention inside APP’s.

20.2 Environmental Permitting Status

The Project is fully licensed and well advanced as follows:





Table 20-1 Environmental Authorization and Permits

All of the permit approvals required to initiate construction of the project are in place. In addition topreviously awarded licenses and permits, an Operational License – LO will still be required prior to thecommencement of mining operations. Such license shall be award by INEMA at the end of constructionphase, during the testing and commissioning period.

20.3 Environmental Baseline Conditions

The environmental baseline study was carried out as part of the preparation of the EIA. The baselinestudy was completed by Brandt Meio Ambiente Ltda in June 2011. The study also included supplementalinformation related to an Equator Principles Compliance Audit carried out by Mineral Engenharia e Meio

Ambiente Ltda (Mineral) dated December 2011 requested by Brazilian finance institutions Itaú BBA,Bradesco and Votorantin who are involved in the financing of the Project. Mineral are, in fact, acting asthe financing bank’s environmental auditor.

20.3.1 Climate and Physiography

The local climate has two distinct seasons, the rainy season (hot and humid) from October to March, andthe dry season from April to September. The average daytime temperature in the Project area is 22.3ºC.

The months of May to August are the most representative ones for winter conditions, at which time theaverage temperature is 18ºC.

Rainfall in the Project area ranges from 480 mm at the Porto Alegre gauge to 630 mm at the Alagadiçogauge both of which are located in the local area adjacent to the Project. There is rainfall each month ofthe year, with the driest period being from May to September, when the monthly precipitation is below 20mm.

Permits Status

Localization License - LLAwarded by CRA-CEPRAM( Environmental State

Council) nº3941, May 2009/2014

Water Rights Grant

Awarded by ANA (National Water Agency) nº684,

September 2009/2019

Installation License - LIAwarded by INEMA(Bahia State Environmental

Institute) nº1286, October 2011/2016

Vegetation Suppression

Authorization - ASV

Awarded by INEMA (Bahia State Environmental

Institute) nº0245 , October2011/2013


Authorization- ASV

Awarded by INEMA(Bahia State Environmental


Intervention in Area

Permanent Preservation

Authorization - IAP



Legal Reserve Approval -ARL

Awarded by INEMA(Bahia State EnvironmentalInstitute) nº0182 , October 2011/2016


Authorization- ASV


Institute) nº049 , June 2012/2014





The Project is located in the middle branch of the Jacaré River, about 3-km west of the western border ofthe Maracás plateau, in a region of very flat terrain with maximum differences of 25 to 30 m. The altitudein the area averages between 310 and 340m and seldom exceeds 400m. The Jacaré River valleyconstitutes a long north-south depression with an average width of 10 km and a length of approximately70 km.

The surrounding terrain is typical ranch/farm land with low trees and shrubs, and relatively-flat platformadjacent to a series of creeks and ponds. The property is bounded to the east by a steep cliff that rises300 m to an area of higher land, the Maracás plateau.

20.3.2 Water Resources

The project is located in the Rio de Contas basin, which has an elongated shape with length of 400 kmand average width of 185 km. The Project is located in the municipality of Maracás located in the Sincorásub-basin.

The Sincorá River is a perennial river, although most of its tributaries are intermittent during the dryseason. At the river flow gauging station of Jequié, the 26-yr average is 25 m3/s, while at the Ubaitabastation (closer to the delta), the flow is 99 m3/s. The main left side tributaries are the Ourives, Sincorá deSantana, Jacaré and Caldeiras Rivers.

The Jacaré River is located 4-km downstream from the Gulçari -A deposit where it receives the flow fromthe João River. The Jacaré River and its tributaries are intermittent watercourses.

Figure 20-1 Jacaré River (Dry period)

Surface water quality monitoring was divided into two groups. One group involves two monitoringstations at the João River (upstream and downstream from the mine site area) and two monitoringstations at the Jacaré River (upstream and downstream from the confluence with the João River). The





20.3.3.2 Arborous Steppe Savannah

This subtype of vegetation shows almost the exact same floristic characteristics as the previous type ofvegetation, however, the individuals of this type are shorter, so there is more ground-level insolation. Itcorresponds to the Arborous-bushy Caatinga, according to Carvalho e Junior (2005).

The shorter individuals that occur in this type of vegetation can be categorized as bushy-arborous andbushy. There is a change in the prevailing species, with the smaller ones being more frequent, includingMimosa spp e de Spondias Tuberosa, Maytenus rigid, Capparis yco and Jatropha ribifolia, besidesSideroxylon obtusifolium and various individuals from the malvaceae family, resulting in a denserherbaceous strata than before. Nevertheless, there are also arborous individuals, more spread out, aswell as individuals from the Bromeliaceae and Cactaceae families, reaching 10% to 60% cover in someareas.

This type of vegetation is found in areas further away from the rivers, with more compact soils and lowhumidity.

20.3.3.3

Recovering Steppe SavannahIts characteristics derive from the absence of a variety of species, with the plants being more spaced out.

Arborous or bushy individuals from the two previous formations can occur (2- to 7-m tall, in general). Dueto past exploration, their branches are not so numerous or dense.

The management and use of the soil is the main factor that shapes this type of formation, since man-made fires are frequently seen, with the purpose of “cleaning” the land for pastures and agriculture and

originating a new vegetation cover, at a secondary stage of ecological evolution.

20.3.3.4 Anthropized Vegetation

It is characterized by various types of clean pastures or bushy-arborous vegetation cover. In this case,the denomination is dirty pasture and if left without management, they can evolve to recoveringvegetation.

In any of these pastures, exotic grass is introduced, normally: Aristida setifolia (capim-panasco),Bracchiaria decumbens (braquiária) and Cenchrus ciliaris (capim-bufel). The annual herb Estilosantes(Styloranthes humiles) and the algodãozinho de seda (Calotropis procera) can be cited as intrusivevegetation more associated with disturbance of the natural vegetation than with the management ofpastures.

The management of these pastures is mainly done for cattle and to a lesser extent for caprines. Thelatter is more related to the presence of communities. The cattle raising activity requires larger areas and

an abrupt alteration of the natural environment, with continuous suppression of natural vegetation and itssubstitution by exotic grass species. Goat raising is less frequent and does not require the continuousremoval of natural vegetation, although it is as impacting as cattle raising.

The local residents use the land to plant and to raise cattle and goats. The plant they use to re-vegetatethe area is the Cactaceae, called palmatória or, Palma (Opuntia palmadora, Cactaceae), which is eatenas a water source for cattle during the dry season. The poor soil and the lack of regular rainfall keepagriculture from developing satisfactorily. At Porto Alegre, the original Caatinga vegetation was replaced





to a large extent by small plantations, where the population uses the water from the Rio de Contas forirrigation.

20.3.3.5 Forest Inventory

The forest inventory carried out by Brandt in 2008 and audited by Mineral Engenharia em Meio Ambiente

Ltda (Mineral) in 2011 showed 141 vegetation species, with 111 being identified at the species level(78.7%), 13 at the genus level (9.2%), 10 species remained to conferatum (cf.) and seven species werenot identified. The reasons were largely due to the absence of reproductive material, available literatureor available herbarium material for comparison.

The most representative family was Fabaceae (20.57%), followed by Cactaceae (8.51%), Malvaceae andEuphorbiaceae (6.4%). These four families encompass roughly 41.9% of the species. The remainingspecies are distributed in 39 other families, with a total of 43 botanical families.

The utilization of region’s native species as a source of natural remedies, food and water source s whenthe dry season becomes too intense is widely known among the local residents. The skin and leaves oftrees like the pau-ferro (Caesalpinia férrea) and the quixabeira (Bumelia sartorum) are used to producetea against rheumatism and diabetes. The fruit from the icó (Capparis yco) and from the umbuzeiro(Spondias tuberosa) are part of the diet of the local population, with the latter forming undergroundtubercles that are used to quench the thirst during harsh dry periods.

Based on the official list of endangered flora, from the IBAMA Norm 06/2008, and from the list of IUCNthe followings tables presents the species found and considered vulnerable and rare.

Table 20-2 Vulnerable Species

Name Popular Name

Astronium fraxinifolium Gonçalo – Alves

Myracroduon urundeuva Aroeira do sertão

Pereskia cf. aculeata ***

Pilocereus piauhyensis Facheiro

Caryocar brasiliensis Pequi

Anadenanthera macrocarpa Angico

Chloroleucon tortum Jurema

Mimosa caesalpinifolia Sabiá

Amburana cearensis Umburana

Psidium rufum Araça

Manilkara elata Maçaranduba

Sideroxylum obtusifolium Quixabeira

Schinopsis brasiliensis Braúna





Table 20-3 Rare Species

The species categorized as rare, are those with a density smaller than one individual per hectare,according to the methodology proposed by Kageyama e Gandara (1993). These species should also bethe targets of environmental management actions for conservation.

20.3.4 Fauna Characterization

The fauna characterization was based on a qualitative survey of the various vertebrate groups throughoutthe Project’s area of direct and indirect influence. The survey was carried out in the beginning of 2008and audited in 2011 by Mineral, through field surveys, interviews with local residents and specializedliterature.

During the field survey, the entire area of direct influence (ADI) and area of indirect influence (AII) werecovered aiming at identifying wild fauna species, which are present in the region. For the mastofaunasurvey, direct observations were made through the method of linear transects, walking along existingtrails in search of animals or traces such as animal tracks, fur, feces or dead animals. The animals were

identified according to the references Cabrera, 1961; Silva, 1994; Emmons, 1997.

The same method (linear transects) was used for the avifauna survey, but with the help of binoculars.The AID and AII were searched, along the existing trails, with the purpose of finding individuals, nests,dead animals or vocalizations. The guide for field surveys by Souza, 2004 was used, as well asspecialized literature (Sic, 1997).

The survey of herpetofauna, reptiles and amphibian was conducted directly at the ponds, water bodies,fallen logs, hollow trees and shadows. The methodology used for identification was recommended byPeters & Orejas Miranda, 1971; Marques et al., 2001; Campbell & Lamar, 2004; and Kwet & Dibernardo,1999.

20.3.4.1 Mammals

The region is home to a diversified mammal fauna, including species ranging from small mammals(rodents) to large-sized animals. The mammals are capable of occupying a large variety of habitats.There are not a significant number of big animals in the region of the project, but there are numeroussmall-sized species, such as bats and rodents. The majority of the mammals are singular and nocturnal.They also hide at the slightest sign of danger, and consequently are seldom observed.

Scientific Name Popular name Family

Astronium fraxinifolium Sete cascas Anacardiaceae

Cnidoscolus pubescens Cansanção Euphorbiaceae

Jacaranda cuspidifolia Pau de colher Bignoniaceae

Luehea paniculata Açoita cavalo Malvaceae

Mimosa tenuiflora Buranhém Fabaceae

No identified Borracha Not identified

No identified Pau de curral Not identified

No identified Pinheiro Roxo Not identified

Patagonula bahiensis Casca fina Fabaceae





Many mammals have cinegetic value, hence being threatened by hunters. Footprints of a Puma Concolor(sussuarana) were observed in the region. This is a large-sized mammal that has its largest habitat in theProject’s region. 47 species were identified, which belong to 19 families.

From the mammals that have economic value, due to their consumption as food by the local residents,the following stand out: armadillos (Dasypodidae), preás (Caviidae), mocó (Caviidae), agoutis

(Dasyproctidae) and tapetis (Leporidae).

According to the Official List of Brazilian Fauna Threatened by Extinction (MMA - Instrução Normativa, no3, de 27/05/03), the Brazilian three-banded armadillo (Tolypeutes tricinctus) is categorized as threatened.The following species are classified as vulnerable: giant anteater (Myrmecophaga tridactyla), ocelot(Leopardus pardalis), little spotted cat (Leopardus tigrinus) e a cougar (Puma concolor).

20.3.4.2 Avifauna

The avifauna was the most representative group during the field survey and it was connected to variousaspects of the area’s vegetation. Birds are considered imperative in any ecosystem, due to the fact that

they combat plagues, contribute with flower pollination, seed spreading, in the control of rodent andvenomous animals, collecting and recycling biological wastes and as a bio indicator of environmentalconditions.

Limiting the scope of analysis only to the caatinga of the State of Bahia, Fiúza (1999) listed 283 speciesand Lima (2004) found roughly 280 species. The studies carried out in the region of Porto Alegre, in thecounty of Maracás (in the Project’s area of influence), considered the species that live in the region all

year-round and the ones that visit the area seasonally or temporarily. Thus, the number of bird speciesfound is 247; they are distributed through 51 zoological families (according to CBRO, 2005).

There is a large diversity of species in the area, since it has 88.2% of all species found in the entire Stateof Bahia. However, this diversity is not uncommon, because the region shows good conservation of

natural resources. The aquatic birds, which use the region’s ponds, lakes and rivers for variouspurposes, help explain this diversity.

The families with the largest number of individuals (above 4% species / family) were: Tyranidae (20species), Emberezidae (19 species), Thraupidae (16), Furnairdae (15), Thamnophilidae (14) andTrochilidae (11). These six families encompass 38.3% of all species observed.

The consumption of wild avifauna as food and its raising as a pet are common in many regions of Bahia.This situation applies to the county of Maracás, as well. This tradition threatens numerous species,including xerimbabos and cinegetic species. The tinamidae family (quails, partridges and tinamous) andthe columbidae family (doves and pigeons) are frequently sought as food in the region. The mainornamental birds are the melro (Icberidae), galo-de-campina (Fringillidae), canário-da-terra (Fringillidae),

caatinga parakeet (Psittacidae), maritaca (Pittaicidae), among others.

Based on the Official List of Brazilian Fauna Threatened by Extinction (MMA - Instrução Normativa, No 3,de 27/05/03), none of the species is categorized as threatened.





20.3.4.3 Herpetofauna

Though the caatinga reptiles are considered well surveyed (Vanzolini et al., 1980), unexpected findingssuggest that little is known about the patterns, which govern their evolution and differentiation (Rodrigues,2005). This group is evenly distributed throughout the caatinga, with the only absentees being thecrocodilians.

The reptile fauna of the region was very rich. During the field survey, the following groups wereregistered (seen or mentioned during interviews): chelonian, serpents and amphisbeanidae.

The venomous fauna of the region shows ample distribution in every biome, except for the jararaca-da-caatinga (Bothrops erythromelas), which is restricted to the caatinga. Among the species that stand out,there is the tropical rattlesnake (Crotalus durissus), jararaca (Bothrops jararaca), Jaracussu (Bothropserythromelas), cobra-patrona (Bothrops sp.) and coral snake (Micrurus sp.).

In this group, the species that are more susceptible to human activities are the snakes and lizards thatlive in forested environments, mainly the small ones and the ones adapted to the microclimate. Thesespecies are usually incapable of enduring the high temperatures of the open fields. Hence, themaintenance of the remaining forested environments is an essential condition for the survival of thesecommunities. A total of 19 species, distributed in 9 families, were registered.

The lizards are known as frugivorous reptiles. They play an important role in the spreading of seeds.

With respect to the amphibians, 12 species were found in the following families: Bufonidae,Leptodactylidae, Hylidae e Microhylidae (Quadro B-listanfíbios). These are mainly found in ponds, wellsand streams; but larger amphibians can be found farther away from water resources and drier areas.

Among the latter, the following can be cited: toads (Bufo crucifer) and treefrogs (Hyla spp.). In morehumid areas, the rãs (Leptodactylus spp.) and the black frogs (Ololygoe spp.) can be found.

The red-tailed boa (Boa constrictor), rainbow boa (Epicrates cenchria), Argentine tegu (Tupinambisteguixin e Tupinambis meriane) and the tuberculate toad-headed turtle (Phrynoes tuberculatus eGeochelone carbonaria) are considered the main cinegetic species in the region. However, they do nothave the same cinegetic value as the mammals, thus they are only occasionally hunted or captured.

The snakes of epidemiological importance, which occur in the region, include: tropical rattlesnake(Crotalus durissus), jararaca-vermelha (Bothrops erythromelas), jararaca (Bothrops sp) and the coralsnake (Micrurus sp). All of them are capable of causing accidents, which, if not properly treated, can leadto death or irreversible damage. The occurrence of accidents involving snakes is a common aspect ofrural communities, where agriculture and pasture are the main labor activities.

20.3.5 Aquatic Biota

For aquatic biota, according to the EIA (2008), samples were collected along the Rio de Contas River tothe end of the reservoir at the Pedras Dam.

Samples from zooplankton, phytoplankton, ichthyoplankton, macrophyte and benthic communities werecollected.





20.3.5.1 Zooplankton community

The qualitative analyzes of the samples identified components of zooplankton community, distributed inthe groups Protozoa, Rotifera, Crustacea, Copepoda, Cladocera, Crustacea and Insecta.

The samples showed a high concentration of rotifers, cladocerans and copepods, which may be related

to anthropogenic eutrophication of the system, due to excess organic matter.

20.3.5.2 Phytoplankton community

The qualitative analysis of the samples identified components of the phytoplankton community distributedin the following divisions: Chlorophyta, Cyanophyta, Bacillariophyta and Euglenophyta.

The results showed a high concentration of the class Chlorophyta probably associated with eutrophicationby inputs of inorganic nutrients from agricultural activities for livelihood.

The high concentrations of this class are associated to the change of hydrodynamic regime of systemsdue to damming of rivers, BICUDO (2005), causing deep and abrupt changes in the conditions ofphytoplankton communities, which reflect the transformation of an open system for transportation to amore closed.

20.3.5.3 Icthyoplankton community

According to the EIA (2008) the identification of representatives of this community came from observationof Pedras Dam and also the collection of eggs and larval fish states.

The qualitative analysis identified larvae distributed in the following families: Erythrinidae, Cichlidae andCharacidae, Poeciliidae, and Engraulidae Ariidae.

In the observation carried out upstream of the Pedras Dam were identified adult families: Characidae(piaba) Erythrnidae (betrayed) Anostomidae (piau) Loricariidae (charitable or catfish), cichlids andScianidae (croaker). Some of the observed species, such as gender Hoplias sp (Traíra), Astyanax splarvae (Piaba), genus Poecilia sp (Barrigudinho), is abundant in most of the watersheds of the Northeastand Bahia.

One of the representatives of the Family Cichlidae, urolepis Oreochromis (Tilapia), is considered exoticand introduced in Brazil through fishing projects in watersheds. The occurrence of peacock bass (Cichlasp) was confirmed by in situ verification of fishing carried out upstream of the Pedras Dam.

Some species Aspistor sp (Catfish), observed not belong to the basins of Bahia and were introducedthrough fishing projects.

In addition, representatives of the Family Engraulidae, Anchoa spinifer (sardines) or Anchoviella vaillanti(anchovy), described species into rivers of the Northeast, are fed to coastal communities.

20.3.5.4 Benthic community

During sampling were identified representatives belonging to the class Gastropoda, with a predominanceof Melanoides tuberculatus (Muller, 1774) (Grastropoda; Thiaridae). This species is native to East Africa,





Southeast Asia, China and Indo-Pacific Islands, and its introduction in Brazil is probably related to trade inornamental fish and plants (France, 2007). The Melanoides tuberculatus occurs in disturbed areas andusually associated with inputs of organic matter.

20.3.5.5 Aquatic Macrophytes

Aquatic macrophytes are important components of aquatic ecosystems because they contribute toimprove the structure and diversity of habitats, interfere with nutrient cycling and participate in the base offood webs (Esteves, 1998).

A hydroelectric dam on the Rio de Contas river has created the Pedras Dam Lake. According to thesampling program, only 2 species (Salvinus Chara sp and sp) were identified in this lake. However,according to residents there are other species of aquatic macrophytes.

Figure 20-2 Pedras Dam reservoir at Porto Alegre

20.4 Social and Economic Baseline

The objective of this section is to present and characterize the various aspects of the social and economicstructure of the project area since such aspects tend to be modified and reorganized during theinstallation and operation phases of the Project.

Although the county of Maracás will directly benefit with regards to the Project, the indirect benefitsextend beyond that county. The changes in the municipality of Maracás related to economic structure, jobopportunities, taxes, income of families and companies and distribution of work force, among others, willresult in new relationships between the municipality and its micro regional neighbors.

The assessment conducted for the environmental study was based on a systemic approach involving theintegration of 25 counties which make up the micro region that will be impacted by the Project. The





present assessment will emphasize the Maracás municipality and the area of direct influence of theProject.

The social and economic assessment was based on interviews with local residents through March 2008(primary data) and on secondary data obtained from an audit review report carried up by MineralEngenharia Ambiental Ltda in 2011.

20.4.1 Populations Dynamics

The evolution study of the population of the counties included in the Maracás micro region encompassedthe last four decades with the purpose of portraying a broad view of the demographic processesexperienced by the 25 municipalities. Generally, the urban population showed larger growth rates;changing the distribution of the population, but still maintaining a rural profile in the majority of thecounties.

From 1991 to 2000, the average annual population growth rate of the 27 counties remained positive(1.62%), resulting in a population growth from 509,378 to 532,409. The urban population of all countiesgrew from 276,868 to 325,905, with Maracás showing the largest annual urban population growth rate(4.75%). Only nine counties (33.3% of the counties) had predominantly urban populations, with theremainder 18 counties staying rural.

In the same period, the average annual population growth rate in Maracás was 1.73%. This county, whichin 1991 still showed a predominantly rural population (44.91% urbanization rate), in 2000, was consideredan urban county, with 58.44% of its population in cities.

In 2007, the population of Maracás was the third largest of all the counties studied, being smaller thanJequié and Jaguaquara. Its 2000 to 2007 average annual population growth rate was 1.11%. It isnoteworthy that in the same period, the average annual population growth rate of the region’s main

county, Jequié, fell 0.12%.

20.4.2 Employment Structure and Unemployment Rate

In Brazil, the service sector employs the largest number of economically-active people. In Bahia, therelative importance of the different economic sectors is similar to the national situation: 29.4% ofemployment in agriculture and grazing; 56.2% in services; and 14.4% in industry. The 27 counties,subject to the study, showed similar distribution of economic activities: service sector, followed by theagriculture and grazing sector and the industrial sector.

In the Maracás county, the largest share of workers are found in the agriculture and grazing sector(45.7%), while the industrial sector provides 11.6% of the county’s employment.

In general, the unemployment rate of the Maracás micro region shows a considerable variability (3.92%to 24.94%). Maracás is among the counties with the largest unemployment rates in the region - 19.41% in2000 - its rate is above the Bahia’s and Brazil’s average.

In the county of Maracás, the share of the population that is younger than 29 years is 62.56%, beingslightly above the region’s average; and the older than 60 years portion is smaller than the region’s

average, 8.92% as opposed to 10%. The county is faced with the challenge of providing education,





leisure and work opportunities to its predominantly young population and the demand for theseopportunities tends to grow in the short range.

The comparison of the literate population in the micro regional area of influence of the Project from 1991to 2000 reveals a significant increase in literacy rates. In 1991, the literacy rate of the population between5 and 9 years was 16.9%. In 2000, this figure grew to 44.4%. Between 10 and 14 years, 54.8% of the

people were literate in 1991 and this rate increased to 91.6% in 2000. The growth in literacy rates wasobserved in every age portion in Maracás, even among the oldest people.

20.4.3 Economic Aspects

In 1991, the average monthly income at Maracás was R$62.10 (eighth place in the region). From 1991 to2000, the county showed the smallest income growth rate in the region (18.4%). In 2000, its average percapita income was only R$73.50.

The Gini index is a measure of the income concentration and varies from 0 to 1. The closer it is to 1, theworse the income distribution is (income is more concentrated). If the value is closer to 0, the income ismore evenly distributed through the population.

In 1991, the best income distribution of all the counties in the micro region was measured in Maracás(0.44). From 1991 to 2000, the income distribution in the counties followed different paths, but in most ofthem, there was more income concentration. In 2000, Maracás still showed one of the best incomedistributions of the micro region, with 0.5.

The economic activities of the counties of the micro region are highly concentrated in the county ofJequié. Of all the goods produced in the region, Jequié alone is responsible for 45.24%. That meansalmost half of all the economic activity of the entire region is limited to one county, so, this one countytends to exert a strong attractive force over the others.

The service sector is the main source of income for the counties in the Project’s area of influence.However, the economical results of agriculture and grazing activities are also significant. Maracásrepresents 3.95% of all income generated in the region.

The per capita gross internal product is led by Jequié which remains in front of all other counties with avalue of R$7,091.48 per year. It is important to emphasize that this value is greater than the state valueof R$6,582.00. Maracás per capita gross internal product (R$2,318.45) is the ninth largest of the microregion. The smallest one of all the 27 counties is the Iramaia’s, with R$1,785.93.

20.4.3.1 Production Structure on the Maracás County

From 1996 to 2006, there was a reduction in the number of businesses that produced cow milk in

Maracás. In 1996, there were 400 milk-producing businesses and in 2006, only 282 businesses were stillactive. In the same period, the cow milk production in the municipality increased by 51%, while goat milkproduction declined 67% and egg production decreased 96%.

The seasonal production in Maracás, especially sugar cane, beans, and tomato, decreased from 1997 to2006. The only exception was mamona, whose production increased 172%. The area used to growproduce expanded 8.5% in that period.





The total area with permanent crops has been subject to significant changes in the last 10 years. From1996 to 2006, this area in the county of Maracás shrunk 20%. Coffee production increased from 2000 to2006, but the 2006 production was smaller than that of 1996. The orange and lemon productions did notchanged significantly, while the “passion fruit” production fell 89.45% from 1996 to 2000.

20.4.4 Land Use and Occupation

The study of the land use and occupation focused on the communities around the Project site and on thecapital of the Maracás County. The study examined public services infrastructure, land management andthe dependence of local residents on natural resources (i.e., water resources), being that this is asemiarid region with water scarcity.

The goal was to provide an adequate view of the area where the effects of the Project will be greatest.This mapping covered the Project’s directly influenced area (DIA), which includes all pits, waste rock/ore

piles, tailings dry stack, processing units and all other operational and administrative support units.

20.4.4.1 Rural Properties

In the DIA there are four large rural properties, averaging more than 2400 ha, where extensive and semi-extensive cattle raising activities are carried out, which require large pasture areas and small bushes. Alllandowners of these properties visit them regularly and employ a number of employees to run theoperation.

There are also three smaller properties in the DIA with only one proprietor residing onsite and taking careof the land. Secondary activities include household agriculture and traditional cheese production.Temporary employment is common in the region.

The facilities that exist in the largest properties include barns and warehouses and all show goodconstruction standards including brick walls. Most of these facilities are sized adequately to the property’s

production.

The construction standard for the houses of the workers is not ideal, but most of them have electric powersupply. Water is supplied through rainwater collection and water trucks. There are only a few housesand, due to the size of the properties, they are relatively far apart from one other

The temporary rural workers live in the nearby villages. Most of these villages are along the main roadthat connects the highway BA-030 (at the community of Pé de Serra) to the village of Porto Alegre.

20.4.5 Villages around the Project

20.4.5.1 Pé de Serra

This village lies on BA-330 highway in the foothills of the mountains that divide the east-central (upper)and west (bottom) of the municipality of Maracás. The community is characterized by a cluster that wasestablished and developed on the edge of the highway.

Among the communities surrounding the Project, Pé de Serra is the one with the best infrastructure,including power, telephone network, water supply through underground wells, some paved roads and





garbage collection. There are inns, small shops and bars. Many homes are vacant waiting for newresidents.

There is no sewage collection infrastructure therefore many houses have poor sanitary facilities erectedoutside the house.

The supply of drinking water is provided through wells with high levels of salts, necessitating the use ofdesalination plants operated by the city to make the water potable for human consumption. The watersupply is at its limit, and the residences are supplied on alternate days to ensure all have access to water.

There is a Family Health Center and a school that offers elementary school grades. Young people arerequired to move to Maracás to continue their studies.

20.4.5.2 Água Branca

This village is located in an area adjacent to the municipal road that connects Pé de Serra and Porto Alegre, east-southeast of the project.

Água Branca is the closest town directly affected area by the Project. It is characterized as a cluster oftypical rural buildings, with houses located on the edge of the road, amid agricultural lands and pastures.

The community has just over 15 homes of farm workers who survive basically from the land work in thelarge surrounding properties. The houses are a very simple pattern, some using adobe-type construction.Most do not have toilets and water supply is provided from wells which require the use of desalination.The main source of income is farm work and some people have already performed services for theMaracás Project with a formal contract.

Public transport is limited to a line connecting the center of Porto Alegre to Maracás, a couple of timesduring the day. School transport is served by buses and vans provided by the municipality of Maracás.

Among the nine communities visited, Água Branca is the one that has the highest expectations regardingthe Project. Specifically residents are viewing the proposed infrastructure (which was promised at apublic hearing) as extremely positive. The infrastructure includes a better road and network connection ofa water supply.

20.4.5.3 Antonio Caetano

The village of Antonio Caetano is comprised of 10 houses located near the Santo Antonio Farm, ownedby Mr. Antonio Caetano Neto. This village is located near the present Project headquarters along themunicipal road that connects Pé de Serra to Porto Alegre.

Households are made up of rural workers who survive basically from the land in large surroundingproperties. It is common to see the cultivation of palm and other crops for subsistence farming, such asbeans, corn and watermelon within the properties. The survival of such subsistence crops is threatenedby low water availability in the region.

The houses are generally very simple in pattern, some of adobe construction. Most do not have toilets,and the water supply is inadequate with the use of wells, rainwater collection systems, and water trucksprovided by the city.





Public transport is limited to a route connecting the center of Porto Alegre to Maracás, a couple timesduring the day. School transport is served by buses and vans provided by the municipality of Maracás.

20.4.5.4 Braga

This village has very similar characteristics to the Antonio Caetano. It is located along a neighboring

municipal road that connects Pé de Serra to Porto Alegre, east-southeast of the project.

20.4.5.5 Caldeirãozinho

This village is located at the margins of the municipal road, near the district of Porto Alegre, and followsthe pattern of occupancy and structure observed in the other vil lages in the region.

The buildings are simple, with small areas used to grow subsistence crops. They are served by theelectric power grid and have a phone network available. The majority of homes do not have toilets andsewage is held in common pits. The existing trade focuses on Pindobeiras, since the acquisition ofproducts not found in the region are available in Maracás. It has a public school that provides elementaryeducation. The streets are not paved and public transportation is limited to the route connecting Porto

Alegre to Maracás.

20.4.5.6 Jacaré

This village is approximately four kilometers south-southwest from the future plant site situated on thebanks of the Jacaré River. It borders the municipalities of Maracás and Iramaia and consists ofapproximately 20 homes of low constructive pattern, some of adobe construction. There is no electricityor telephone network and the water supply is limited to individual tanks to capture rainwater or tankertrucks supplied by the City of Maracás.

It is common to have small yards where usually palm is grown for human and animal consumption. There

is a school that offers grade 4 elementary education. To continue their education, students would need togo to Porto Alegre.

20.4.5.7 Lagoa Comprida

This village is located east-southeast of the project along a municipal road that connects the neighboringvillages of Pé de Serra and Porto Alegre. The village has a small number of rural and agricultural workers'households who survive from the land deals in the large surrounding properties. It is common to cultivatepalm and other crops such as beans, corn and watermelon for subsistence. This village has strongrelations with the communities of Água Branca and Antonio Caetano based on their proximity.

The houses follow the same regional pattern with most not having toilets, and the water supplied with the

use of wells, rainwater collection systems, and water trucks provided by the city.

Public transport is limited to the same line connecting the center of Porto Alegre to Maracás, a few timesduring the day.





20.4.5.8 Pindobeiras

Pindobeiras is established at the margins of the municipal road, near the district of Porto Alegre. Thevillage of Pindobeiras is comprised of 30 houses with a population of 122. Five houses are currentlyvacant. Houses are simple, but are served by the electric power grid and telephone network. The majorityof homes do not have toilets.

According to the local leader, since the population does not receive government assistance for thedevelopment of agriculture in the settlement, it develops activities in large farms in Maracás. Publictransportation is limited to the service connecting Porto Alegre to Maracás.

20.4.5.9 Porto Alegre

Situated close to the Pedras Dam, Porto Alegre is an urban center predominantly residential of low tomedium constructive pattern. Community employment is predominantly from fishing and the cultivation offruits and vegetables at lake banks, including corn, watermelon, and mango. In some areas cattleranching was identified.

Porto Alegre is the region’s most populous village with approximately 1020 residences. Residents

reported that the increase in the population began with the construction of the railway on the oppositeside of the lake.

Infrastructure includes a power grid, public water supply, health services, education and locations forsocial interaction and leisure. Sanitation is inadequate, with the use of mass pits which have highpotential for soil contamination.

The main resource of the Porto Alegre residents is the lake, which provides the water for development ofagriculture, fishery and leisure. According to local information, approximately 80 men are engaged inagriculture and fisheries with the most common fish caught being tilapia and piranha. The shrimp fishery

is largely staffed by a group of approximately 25 women.

20.4.5.10 Quilombola and Indian Communities

Throughout the Project’s DIA (made up of the 27 counties), only the county o f Jequié has a Quilombolacommunity that is officially recognized. Quilombola are the descendents of slaves who escaped fromslave plantations that existed in Brazil until abolition in 1888. The most famous Quilombola was Zumbiand the most famous Quilombo was Palmares.

The people of Maracás consider the communities of Cuscus, Pindobeiras, Caldeirão dos Miranda andJacaré as being quilombolas, in spite of not being legally recognized by the Palmares Foundation andINCRA (National Institute for Colonization and Land Reconstruction).

There are no Indian communities in the county of Maracás.

The majority of the nine villages within the Project area have less than 24 houses. The exceptions are Péde Serra and Porto Alegre, which have approximately 220 houses. Most of the houses show humbleconstruction standards, with brick walls and clay. Several houses were built in the 1950s. Many houseshave backyards with small plantations and animals, which are for their own consumption. Villages are





accessed by a main road, so houses are relatively close to one another. The public services are modestwith some villages having schools, health clinics and small commercial centers.

Out of the nine villages, only Jacaré does not have electric power supply. Porto Alegre has some pavedroads. Water is obtained from wells, rainwater collected from rooftops and water trucks supplied by thecity. In the case of Porto Alegre, the water is drawn from the Rio de Contas, which is dammed 85 km

downstream from the community.

Porto Alegre is the village that shows the best sanitary conditions.

The main economic activity for many residents of these villages is working with cattle on the farms. InPorto Alegre, there is also irrigated production of fruits and vegetables.

20.4.5.11 Municipality of Maracás

Maracás is an urban conglomerate located in the central portion of the county on a plateau that is at ahigher elevation than the surrounding valleys. It is a typical town in the interior of Bahia, with a smallpopulation that cultivates the habits and customs of their ancestors. Currently known as the “city of

flowers”, Maracás had a population of 21,832 in 2010.

Downtown Maracás has two main avenues, Brasília and João Durval, and has good urban infrastructurewith paved streets, a water supply system (the water is withdrawn from the Boca do Mato reservoir, 9 kmaway), an electric power supply and telephone lines. There is regular waste collection and streetsweeping services making the public avenues much more aesthetic than many villages. There ishowever no wastewater collection system and the domestic sewage is disposed of in pits, usually in thebackyard of the houses.

Maracás has legal regulations regarding urbanization, such as those established in the city’s

development plan.

The tertiary sector, represented by activities of trade and services are the main source of municipalincome and in 2008 it participated with 65% of Gross Domestic Product (GDP).

The agricultural sector plays an important role in the economy of the city with the GDP accounting for23% of its composition. Labor absorption accounts for 45.7% of registered jobs in 2008.

The city's economy is not diversified, resulting in low employment. The unemployment rate was 19.41% in2000, exceeding state and national levels. This generates a high number of people living below thepoverty line. In 2003, according to IBGE data, 54.63% of the population of the city was below the povertyline.

Commerce and services are distributed along the main avenues and streets. The local commerce isrelatively diversified and in harmony with the size of the city. It includes clothing stores, cellular phoneshops, furniture stores, house appliances stores, bookstores, construction shops, grocery stores,drugstores, and restaurants. Among the institutional services are: a branch of the Banco do Brasil,lotteries, post offices, schools, social centers, health clinic, a Catholic Church and an Evangelical Church.

The educational system is made up of eight public schools, which offer elementary and high schoolcourses. Five of these schools are municipal and three are run by the State. There is a private





elementary school and private kindergarten schools. The municipal education system is supplementedby kindergartens available for the children that live downtown and in the surrounding neighborhoods.

The local college is called Faculdade de Tecnologia e Ciências (FTC), and has on-line graduationcourses including business, biology, mathematics, languages as well as some specialization courses.

Water supply is insufficient for the city. For human supply, the municipality makes use of surface waters(rivers and springs) and underground (wells). The county is supplied by a small dam known as Boca doMato.

Violence and drug use are the most common social problems. According to a representative of theJudiciary, violence involving young people is increasing, with narcotics most often cited as a generator.This scenario may reflect different situations, such as lack of perspective and employment opportunities,poor choice of leisure involvement with people from different places, among many others.

20.4.6 Historical and Cultural Heritage

Maracás is a typical city of the Bahia’s countryside, with a small population that perpetuates the habits

and customs of its ancestors. The name of the city comes from Maracás, which is an Indian tool used bythe Cariris tribe that lived in the Paraguaçu region.

According to studies by Prof. Carlos Ott, the Portuguese arrived at the Paraguaçu valley in search of goldand diamonds and found the Indians. Many bloody battles were fought, resulting in the disappearance ofthe Indians. These Indians are still remembered today in the history of Maracás as being brave andaggressive warriors.

The influence of the gold cycle on the county can be seen today by the city’s architecture, showing

houses with styles that are typical of that period. The houses are narrow, with high doors and windows,similar to the ones found in Chapada Diamantina and Ouro Preto.

The Portuguese occupation of the region is evident through the main houses of the farms, in colonialstyle. One example is the Santa Rita farm, which still keeps the big main house with its XVIII centuryfurniture and chapel.

Since the occupation of the region started in the period in which Brazil was still a colony of Portugal, andduring which slavery still existed in Brazil, the presence of black people is very pronounced.

According to Prof. Marina Silva, Maracás was one of the five Brazilian towns that hosted Germans duringthe Second World War. The German presence in the county is clearly seen by the main church’s

architecture, in German gothic style and by other houses in the same style.

Besides the areas of historical and architectural value, there are some areas that are part of the county’snatural heritage and should be protected. For instance, the Jequiriça River headwaters park rebuilt bythe municipal government; the Eucalyptus Park; the water spring of Jequiriça River; and mountains of theregion.





20.4.6.1 Archaeological Heritage

The report “Programa de Diagnóstico e Prospecção para o Projeto Vanádio de Maracás, Maracás, Bahia”

submitted to INEMA for Installation Permitting, presents the archaeological studies carried out in 2007 bythe company Arqueologia Brasil - Projetos, Pesquisas e Planejamento Cultural e Arqueológico Ltda.,whose principal office is in Espírito Santo do Pinhal - São Paulo. The archaeological diagnostic and

survey was approved by Acervo - Centro de Referência em Patrimônio e Pesquisa, based on PortoSeguro - Bahia. The leader of the technical team was Prof. Dr. Walter Fagundes Morales (archaeologiste sociologist), and the other archaeologists of the team were: Luiz Augusto Vivas, Flávia Prado Moi,Daniel Bertrand e Diego Palma Rocha.

The archaeological studies were authorized through the IPHAN publication N. 162, from July 30, 2007,which deals with the permission to carry out archaeological survey and analysis for the Project, in thecounty of Maracás, State of Bahia.

The archaeological survey was concentrated in the areas of the mining rights DNPM 870.134/82 andDNPM 870.135/82, where the mineral deposits of the Project are located.

In these areas, 20 archaeological sites and 41 occurrences of archaeological materials were identified.Every archaeological site and occurrence is identified by its geographical coordinates and described inthe report, which provides many photographs as well.

The recovery and monitoring to protect the archaeological traces and structures at the surface orsubsurface is under way. During the recovery and monitoring stage, every archaeological site andoccurrence must be revisited and minimum characterization interferences must be done. This last stageof the archaeological study must include the continuation of the heritage education activities. Theprograms of archaeological recovery and heritage education will be included in the PCA (EnvironmentalControl Plan), which is part of the environmental permitting process for the Project.

20.4.7 Living Standards

The human development index - HDI was created in 1989 to represent the level of development andliving standards of a community. The intention of the people who created this index was to representdevelopment based on three criteria: life expectancy, education and GDP per capita.

The criterion life expectancy - life expectancy at birth - aims to represent the health condition of a society.In 1991, the Maracás life expectancy index (0.541) was categorized as medium, so the county was thesixteenth in the micro region. In 2000, the county improved its life expectancy index by 6.7%, reaching0.577 and, thus, occupying the 12th position.

In 1991, generally speaking, the educational indices of the counties in the Project’s area of influence were

worse than the life expectancy ones. However, from 1991 to 2000, this index increased and overcamethe life expectancy values. The Maracás’ educational index in 1991 was 0.490 (low). In 2000, this valuehad increased to 0.714, showing a 59.4% increase and, hence, was classified as medium.

The GDP per capita index is the one that has the smallest contribution to the HDI of the region’s counties.

In 1991, Maracás had a GDP per capita index of 0.462, placing it in the seventh position, out the 27counties of the micro region. In the period from 1991 to 2000, the growth of that index in Maracás wasonly 6.1%, so the county moved down to the fifteenth position, with a GDP per capita index of 0.490.





Out of all the 27 counties located in the Project’s area of influence, in 1991, Maracás was in the

fourteenth position, with an HDI of 0.498 (low). From 1991 to 2000, the increase in the HDI index of thatcounty was the sixteenth best (22.3%) and its HDI reached 0.609 (medium). So the county was in thesixteenth position, losing two positions. The most important of the three criteria, in the case of Maracás,was the education index (0.759).

20.4.8 Education

In 1991, Jequié had the largest average school years of all the counties in the Project’s area of influence.

That county had the same value as the State of Bahia, 3.3 years. Maracás’ average was only 1.5 years

and that placed the county in the ninth position among the counties in the micro region. In 2000, Maracásmoved to fifth place with a 73.3% increase in its average school years, reaching 2.6 (still low).

Maracás’s position, with respect to adult illiteracy (older than 25 years), was 14th in 1991, and the value

was 56.7%. In 2000, Maracás reduced its adult illiteracy rate to 38.6%.

In general, data showed an improvement in the number of Maracás residents that have educationalservices. However, additional efforts are needed to reach the educational level of the Bahia State(average of 4.5 school years and illiteracy rate of 28.5% in 2000).

At the municipal and state system, Maracás provides education at three levels: kindergarten, elementaryand high school. The rural population has free transportation to the schools, provided by the countyadministration. The population of the villages located near the area of the Project has access to municipalschools.

20.4.9 Health

The hospital beds available in the Project’s area of influence are predom inantly private owned (52.5% or741 beds). The public hospital beds total 693 or 47.5%.

Maracás has 64 hospital beds, with 40 being municipal owned and 24 privately owned. This represents aratio of 1.9 beds for a thousand inhabitants; a value that is smaller than the OMS standard. With respectto hospital beds, Maracás occupies the 13th out of the 27 counties of the micro region.

Maracás shows one of the worst (26th) healthcare coverage of all the 27 studied counties, with only67.4% of the population assisted with such programs.

Of the 27 counties that make up the micro region, Jequié has the largest number and the greatest varietyof medical equipment. Even so, the medical services are still very precarious. But, due to the evenworse situation of all other counties, Jequié is the region’s center for medical care.

Maracás has only one piece of X-ray equipment (100-500 mA). Thus, the Maracás’ population needs togo to another county if a more complex medical examination is necessary.

In 2008, the healthcare system had six public health clinics; four family health units; one healthcarecenter; one clinic specialized in birth surgery and one hospital. There is a shortage of doctors for theurban and rural communities. Due to this shortage, the public health clinics in the rural zone operate withnurses and assistants, while doctors are available usually once a month.





From 1999 to 2005, the main cause of death in Maracás was caused by brain / vascular diseases. Heartstrokes were the second and diabetes mellitus and transit accidents were third.

20.4.10 Housing Conditions and Infrastructure

In the last census (2000), Maracás had 7,430 families living in private houses (31,678 people) and 7,430

people declared they provided money for their family; 5,288 were spouses; 16,330 were sons anddaughters; 152 were parents or mother / father in law; 1,035 had another type of relationship, and 220had no family relationship with the owner of the house.

According to the same census, there were 6,832 houses with adequate sanitary installations in Maracás.Out of these 6,832, 28% had treated water supply, 0.9% wastewater collection, and 76.7% had theirwastes properly disposed of.

The data from the last census shows that, though the population of Maracás does not have easy accessto basic consumption goods (refrigerators, etc.) and to sanitation services, the housing conditions are notas bad as that seen on large urban centers, mainly with respect to the number of people per room.

The Maracás’ water supply system has 2,914 active water connections, supplying water to 18,533

inhabitants, through a 58-km-long network. The county does not have a sewage collection network, sowaste water is disposed of by the individual households.

The public cleaning services are limited to the county’s capital and includes trees’ trimming, streetsweeping and waste collection. The waste is disposed of at a simple landfill that started in November2005.

The city does not have a rainwater drainage system, but the reconstruction of the BA-026 (connectsMaracás to Contendas do Sincorá) includes storm water drainage adjacent to the road.

The following roads are used to get to Maracás: BA-026 (Maracás /Contendas do Sincorá), BA-250(Maracás /Lajedo do Tabocal) and BR-330 (Maracás /Jequié). There is a bus station, where theintermunicipal routes connect to Salvador, Vitoria da Conquista, Iramaia, Jaguaquara, Jequié, Ilheus andPorto Seguro, and interstate routes from Rio de Janeiro and São Paulo arrive. The county has a landingstrip called Luís Eduardo Magalhães.

Also it has a community radio station, loud-speaker services, telephone lines, cellular coverage, postoffices and four small newspapers. The TV broadcasts are TV - Sudoeste, Aratú, TVE Bahia andBandeirantes.

There is a municipal market on Saturdays when products from Maracás and Jaguaquara are soldincluding live animals (i.e. pigs, goats, chicken, ducks), cereals, grains, vegetables and meat. Almost all

the products and meat consumed in the county are produced locally. The cereals are brought from othercities in the southeastern region.

20.4.11 Leisure, Tourism and Culture

The cultural aspects of Maracás are intimately related to the religion of the population. The mostimportant cultural events, both in the rural and urban zones, show traces of popular Catholicism mixedwith enjoyment. The main events of the county are: Ternos de Reis, Festejos Juninos (Trezenas de





Santo Antônio e São João), Festa de São Roque, Festa de Nossa Senhora da Graça and Cosme eDamião.

The city has a few public leisure areas, such as the Eucalyptus Park, where environmental institutions arelocated (ADAB, EBDA, Production Secretary and Flower Project of Maracás). The park is also used forvarious sports. Besides this park, there is the park of the springs of the Jequiriçá River, where there are

ecological tracks, sports court and municipal squares.

There are also very few leisure areas in the communities surrounding the future Project area. Most of thevillages have only small soccer fields. The exception is Porto Alegre that has a sports court built by themunicipal government.

20.4.12 Public Safety

Maracás’ public safety is ensured by the 19º Batalhão da Polícia Militar (Millitary Police) by the 4ª CIA dePolícia Comunitária (Communitarian Police), by Delegacia de Polícia Civil (Civil Police Delegacy) and bythe Guarda Municipal de Maracás (County Police).

The Military Police is composed of 14 police officers and 1 vehicle. This structure is enough to ensure thepublic safety of the county and the services provided include: rural surveillance, school surveillance, roadblocking at night and drug traffic combat, among others.

The Civil Police has five officers and five public agents, as well as one vehicle. There is no firedepartment in the municipality.

20.4.13 Property Disputes and Rural

In the Project’s DIA or DII, there are no property or land disputes. There is a program for rural settling(Pakhaeta) in the California Farm, village of Pindobeiras. The area available for settling is of 2035 ha and

is large enough to settle 63 families. The registration process is underway. All the people to be settledwill be rural workers in the region.

20.4.14 Water Supply

The water sources for human consumption include surface water bodies (rivers and springs) andgroundwater (wells). The public raw water comes from a small dam called Boca do Mato. The villages ofPorto Alegre and Pindobeira obtain their water from the reservoir of the Pedras Dam, Rio de ContasRiver. The other villages get their water from wells and water trucks.

Due to the fact that it is located in the margin of the Pedras Dam, the community of Porto Alegre can usethe water from the Rio de Contas for various purposes, including irrigation, leisure and fishing. The

multiple water uses occur in no other village in the region.

With irrigation and fertile soils, the rural properties of Porto Alegre deliver fruits for export (mango, papayaand cashew) and vegetables for the regional market. Nevertheless, irrigation is not a threat to the supplyof water for human consumption, because the lake volume is greater than 80,000,000 m3.





Although the fruit production is very large at Porto Alegre, fishing is still the most important economicactivity in the district of Porto Alegre. Besides fishing, the people also produce fresh water shrimp, whichis sold in Jequié.

Navigation is done only in the Pedras Dam, but it is restricted to small fishing boats and a special boatthat transports goods and people among Porto Alegre, Jequié and Iramaia.

The reservoir is also used for leisure purposes such as carnivals and other events, when tents and publicshows are set up.

20.5 Environmental Impact Assessment, Mitigation and Compensation

This section was founded on a technical report prepared by Mineral Engenharia em Meio Ambiente Ltda,an independent consulting company retained by Largo in 2011 to carry out an Equator PrinciplesCompliance Audit in the Maracás Vanadium Project.

The Audit report focused on the aspects recognized by Principle 2 – Environmental Assessment, and asadvocated, it identified and discussed the impacts and relevant social and environmental risks of theproject, during the phases of installation, operation and decommissioning. Furthermore, it examined theproposed controls, monitoring programs and mitigation measures and appropriate management programsfor enforcement of the principles.

20.5.1 Physical Environment

20.5.1.1 Erosion and Silting of water bodies

The excavation, removal and storage of soil create points susceptible to erosion resulting in laminar flowof rainwater, which can generate points of silting. The deforested areas, the excavated slope, openingsof access roads, water catchment systems are all susceptible to erosion albeit low. This is due to such

reasons as low annual precipitation and the smooth topography in the DIA. During periods of short,intense rainfall, the occurrence of solids removal and silting of the João River and Jacaré River can occur.

Environmental, Health and Safety Guidelines for Mining establish some premises to be taken for theprevention of erosion processes and settlings in industrial and mining activities. To decrease theincidence of erosion processes in the area and settling of water bodies, the project will develop severalmitigating measures, including the implementation of the various measures proposed in the PRAD (Planof Rehabilitation of Degraded Areas). An example is the erection of protection barriers for the North andSouth ridges around the Gulçari A pit to avoid sediment being washed inside the pit.

The access routes will be constructed with drainage channels to drain off rain water for a containmentbasin, where the sediments will be cached. The channels are designed for the rainy season, when the

maximum rainfall in the region occurs.

The existing water bodies in the DIA of the project, specifically the João and Jacaré Rivers may have theirquality affected by solid and substances being washed from the installations. The groundwatercontamination can occur from the percolation of substances in the soil. In summary, the risks ofcontamination are centered on the following sources:





Waste piles; Non-magnetic dry stacked tailing ponds; Leached calcine tailings dumps; Chloride purge tailings ponds; Storm water drainage system; Effluent of industrial plant; and, Oily effluent.

20.5.1.2 Waste piles

The project will feature external waste dumps or piles generated from the mining of various open pits. Thewaste material originating from the open pits, which will be placed in a waste pile or catchment dyke, ispredominantly rock consisting of boulders of varying sizes. The area destined for the Gulçari A Phase 1waste pile covers approximately 47 ha and the area destined for the Gulçari A Phase 2 waste pile coversapproximately 119 ha. The waste piles for the satellite pits vary in size ranging from 16 ha to 55 ha.

To check the risk of groundwater contamination by seepage of substances in soil or surface water,leaching and solubilization testwork of waste material was conducted at SGS laboratories, followingprocedures regulated by the Brazilian Association of Technical Norms-ABNT, according to NBR10.004/2004.

As to investigate the likelihood of acid rock drainage, the three types of rocks were submitted to ABA-Mtests for prediction of acid drainage. The Table 20-4 and Table 20-5 show the results of the analyses.




Table 20-4 Testwork results - ABA-M – Waste Rock- Gulçari A




Table 20-5 Testwork results – ABA-M- Waste Rock- Gulçari A





According to the testwork, the gabbro, which constitutes the greater part of the mine waste, has a smallpotential for acid generation. The remaining rocks (pyroxenite and pegmatite) do not have the potential togenerate acids.

20.5.1.3 Non- Magnetic Dry Stack Ponds

The non-magnetic dry stack ponds are formed from the deposition of weakly or non-magnetic mineralsobtained from the magnetic separation of titano-magnetite ore after grinding and filtering. This tailings drystack structure was designed and incorporated into the operational layout as an alternative solution toconventional tailings deposition in the Jacaré River valley, in order to protect and preserve an arborouscaatinga area near the mine site.

The magnetite rock was subjected to acid drainage prediction tests, solubilization and leaching of heavymetals, in accordance with the norms of ABNT. The results of such testing are shown in the Table 20-6and Table 20-7 below:




Table 20-6 Testwork Results – ABA-M –Magnetite Rocks- Gulçari A





Table 20-7 Testwork Results – ABA-M –Magnetite rocks- Gulçari A

As shown in Table 20-6 only the magnetite originating from the 100 meters level horizon, has a lowpotential of acid generation. The upper and middle magnetite has no acidic drainage generation potential.

According to the Table 20-7 this residue was classified as inert (class IIB) according to ABNT NBR10.004/2004. Therefore, the risk of leaching and percolation of substances in the soil from the reaction inthe stack with rainwater it is not expected.

According to the design of the tailings facility the first cell will be sealed and monitored as a precaution.The design of the final drainage system for the system will be done after completion of all the "ponds".

20.5.1.4 Calcine Residue Stack

The tailings identified as calcine residue, consist essentially of synthetic hematite (calcined magnetite)that will contain, regardless of the effectiveness of the leaching and filtering processes, some residues ofvanadium and sodium salts soluble in the form of sodium vanadate.

Such material, after filtering and washing, will be stacked in a lined and impermeable structure. Theresidue stack will be wetted by sprinkling water on top of it, washing progressively for removal of solublesalts. The solution will be collected in tanks and returned by pumping to the metallurgical plant.

The residue remaining on the stack is considered a Class I (dangerous), and the leaching pad is fullywaterproof with high density dual layer polyethylene.

It should be noted that Largo will be selling this material as iron ore; therefore, an excessive buildup ofthis material will not occur. Over the life of the project, approximately 500 kt of this iron ore byproduct willbe sold on an annual basis and a temporary storage facility will be allocated accordingly.

20.5.1.5 Chloride Salts Residue Pond

The chloride salts pond contains a solution rich in chlorine, present in the purging of the evaporation

system. The effluent flow is in order of 2.9 m³/h, arranged in a dam type structure ("pond") lined with highdensity polyethylene, dual layer.

20.5.1.6 Storm Water Drainage System

Each area of the plant has a containment reservoir in which rainwater and washing water from processwill be continually collected and re-circulated.





20.5.1.7 Oily Effluent

The maintenance of machinery, vehicles and equipment, is a potential source of effluents containing oilsand greases as is contamination from other chemicals. The generation of this effluent or contaminationcan occur in places of storage in all stages of project development.

To mitigate the possible impacts on the quality of surface and groundwater resources, the project willdevelop the following measures:

• Both the solid material from the tailings dry stack as the solution percolates will be systematicallyand periodically sampled and tested to determine the levels of soluble salts of vanadium present.

As soon as it is confirmed the tailings neutralization as class IIB (non-hazardous and inert) on theconformity of ABNT NBR 10,004/2004, the stack will be covered with sterile rock and soil, andvegetated.

• The plant is designed to not generate effluents. The only effluent produced is the chlorine saltsolution, which is sent to a dike designed for evaporation. All other solutions are recycled back inthe process.

• For the sanitary effluent treatment, during the installation phase of the project, a provisional ETS(Effluent Treatment Station) will be used, which will be operated until the end of commissioning andinstallation phases. After this period it will be decommissioned and turned off. The effluent from theETS will be recirculated to the tank and used to wet the calcine leach pad. The effluent from theETS will be submitted to the Effluent Monitoring Program, with quarterly analysis of physical-chemical parameters, these being at a minimum: pH, BOD (Biochemical Oxygen Demand),suspended solids, dissolved solids, total coliforms, color, turbidity, nitrates, nitrites, total nitrogen,total phosphorus, sulphates and sulfides.

• Each plant area has a containment tank where rainwater and washing water will be continually

collected and recycled to the process and all contention structures (non-magnetic tailings stack,waste and ore piles) must have containment barriers that prevent the contamination/sedimentationof natural drainage.

• The Plan for Monitoring and Quality Control of Surface Water, Groundwater and Sediment (anEnvironmental Management Plan) predicts biannual sampling of 14 points for surface water andsediment collecting and 10 wells for monitoring ground water.

• For control of oily wastewater, the areas of mechanical repairs will equipped with waterproof floorwith collection systems and water/oil separators, constructed in accordance with legal standardsand compatible with the flows. The effluent from oil and water separators will be sampled quarterlyfor pH, BTEX (Benzene, Toluene, Ethylbenzene, and Xilene) levels, TPH (total Petroleum

Hydrocarbons) and oils and greases.

20.5.1.8 Soil Contamination

The Solid Waste Management Program presents guidelines for the packaging and disposal of wastegenerated in the project. The plan identifies the waste types, volume of generation, packaging anddisposal of all waste generated in all the project phases. It also presents a list of recycling companies thatshould be intended for recyclable waste.





According to the company plan, a landfill project nearby the industrial area was recently submitted toINEMA for disposal of non-toxic wastes. The sludge of ETE will be wrapped in a drying bed to separatesolid and liquid phases, and prepared for final destination as fertilizer in vegetation of degraded areas andlandscaping of the enterprise areas.

The sludge generated in the ETA will be wrapped in a drying bed to separate solid and liquid phases, and

prepared to be donated to the nearby potteries.

20.5.1.9 Change in Air Quality

Fugitive dust from moving vehicles on unpaved access roads, ore and waste stockpiles, especially in thedry period, generates particulate material. The operation of machinery and equipment will generateatmosphere gases resulting from the combustion of diesel in combustion engines and dust, potentiallyalso leading to a change in the quality of the air.

In the comminution process, primary crushing will also generate particulate material. In the calcinationstage, kiln emissions will contain SO2, CO2, and H2O. All the sulfur contained in the feed (reagents andfuel) is converted to SO2 and, in smaller proportions, in SO3. The chlorine present is converted intogaseous HCl and released into the furnace.

In the drying process of the MVA (Ammonium Metavanadate), a scrubbing system will be installed forMVA dust recovery.

The ammonia removal system will comprise a scrubbing system to remove any residue of ammonia.Sulphuric acid reacts with ammonia to produce ammonium sulfate, which will be pumped into the reagentrecovery system.

The future production of ferrovanadium will be by the Aluminothermic process which does not use anelectric furnace. The fumes generated in the process of casting will be extracted by a hood and passed

by bag filters for removal of particulates. Dust collected will be re-routed to the furnace.

In order to verify the impacts of the emission of pollutants of the project on air quality in the regions closeto the project, the atmospheric dispersion of pollutants, considering the base case scenarios of FeV andV2O5 production, was modeled. A mathematical simulation was carried out by SECA – Sistema deEstudos Climáticos e Ambientais, an independent consulting company, based in São Paulo, applying thelast generation EPA-AERMOD model, using four years of weather data, terrain topography, pollutantemission data generated by the project and the exhaust parameters on a grid of the domain area of 900km2, in the municipality of Maracás, State of Bahia.

Table 20-8 shows an estimate of atmospheric emissions for the base case production scenarios,according to the study of Atmospheric Dispersion carried out by SECA – Sistema de Estudos Climáticos e

Ambientais.





Table 20-8 Maracás Vanadium Plant Atmospheric Emissions Estimate

The results of the modeling showed that independent of the regulated pollutant, there is no violation of airquality standards in the short and long term of CONAMA resolution No. 3/90 neither IFC standardregulations.

Maximum results obtained in the study of dispersion concentrations and air quality limits established inlegislation are presented in Table 20-9 and Table 20-10 below.

Table 20-9 Maximum short-term pollutant concentrations from plant emissions and related emissions

standards

NOX SO2 MP V NH3

Kiln off gas 23,1 120 0,59 0,005 0

AMV Flash Dryer 1,2 0,28 0,1 0,045 0

AMV Reduction Kilns 0 0 0,04 0,023 9,1

FeV Furnace Baghouse 0 0 0,02 0,001 0

Quench Scrubber 0 0 0,14 0,026 0

Screen Dust Collector 0 0 0,56 0,002 0

Crushing Dust collector 0 0 0,83 0,002 0

Kiln off gas 23,1 120 0,59 0,005 0

AMV Flash Dryer 1,2 0,28 0,1 0,045 0

AMV De-ammoniator Kiln 0 0 0,07 0,029 19,7

Quench Scrubber 0 0 0,14 0,026 0

Screen Dust Collector 0 0 0,56 0,002 0

Crushing Dust collector 0 0 0,83 0,002 0

Scenary SourcesEmission rates (g/s)

Ferro Vanádium FeV

Vanadium Pentóxide V2O5

MP10 SO2 NOx Vanádio,V. NH3

(24h) (24h) (1h) (24h) (24h)

13,5 89 174 0,296 59,8

13,5 89 174 0,346 124,4

150

(24h)

365

(24h)

320

(1h)- -

75

(24h)

125

(24h)

200

(1h)- -

Pollutant

Primary Standard

CONAMA 3/90

Scenario 1 FeV Production

Scenario 2 V2O5 Production

Maximum Concentrations (µg/m3) - 24 hours

Equator Principles





Table 20-10 Maximum long term pollutant concentrations from plant emissions and related emissionsstandards

Maximum daily mean concentration of SO2 was 89 µg/m3 for both scenarios. This value represents 24%of the daily CONAMA standard for SO2 of 365 µg/m3 and 71.2% of the daily IFC standard -125 µg/m3.The maximum point is found in the Southeast and 4 km away from the chimneys of the mining company.

Maximum annual average concentration of SO2 was 7.3 µg/m

3

for both scenarios. This value represents8.8% annual standard of CONAMA 3/90 for SO2, 80 µg/m3, and 14.6% of the annual IFC standard of 125µg/m3. The maximum point is found to the East and 7 km away from the project.

Maximum hourly mean concentration of NOX was 174 µg/m3, for both scenarios. This value is 1.8 timessmaller than the default zone of CONAMA 3/90 for the NO2 of 320 µg/m3 and 1.2 times smaller than thedefault zone of the IFC for the NO2 of 200 µg/m3. The maximum point is found in the Northeast and the4.0 miles away from the project. Maximum annual average concentration of NOX was 1.0 µg/m3, for bothscenarios. This value represents 1% of the annual CONAMA standard of NO2 100g/m3 and 2.5% of theannual IFC standard of NO2 40 µg/m3. The maximum point is found to the East and 8.0 miles away fromthe project.

It will be necessary to develop a new dispersion model for the expansion scenario. The quantities ofpollutants (apart from SO2) are expected to increase as is the volume of gas, increasing the stack exitvelocities. This expected increase will result in different dispersion levels; however, the levels areexpected to be within the approved ambient air guidelines. A new dispersion model will be developed toconfirm this expectation.

The project will include atmospheric emissions control equipment such as an electrostatic precipitator inthe rotary kiln exhaust system, gas washers, scrubbing systems and filters in other 6 stacks. Theemissions will be quarterly monitored to the pollutants: MP, NOx, SOx, CO, V and NH3.

The Air Quality Monitoring program proposes the installation of three monitoring stations in the area. Thestations will have continuous meters of SOx, NOx and MP and sampling and monitoring of V and NH3 at

the air quality monitoring points will be conducted twice a year.

For combustion gas emissions on mobile sources, the project will require the adoption of preventivemaintenance and control of emissions from vehicles, equipment and machinery in order to ensure thatoperational conditions are always as per normal standards. The frequency of maintenance of the fleet tocontrol emission of pollutants will be set by the equipment standards.

MP10 SO2 NOx Vanádio,V. NH3

(24h) (24h) (1h) (24h) (24h)

2,3 7,3 1,0 - -2,3 7,3 1,0 - -

50

(Year)

80

(Year)

100

(Year)- -

35

(Year)

50

(Year)

40

(Year)- -

Pollutant

Maximum Concentrations (µg/m3) - 24 hours

Scenario 1 FeV ProductionScenario 2 V2O5 Production

Primary Standard

CONAMA 3/90

Equator Principles





After kiln commissioning, the study of Atmospheric Dispersion will be repeated to confirm the parametersfound in the model.

20.5.1.10 Change in Noise Levels

The sources of noise in the study area will be linked to the movement of small vehicles, trucks, machinery

and equipment used for the opening of access roads, preparation of mine site and infrastructure.

During the operation phase there are activities with intense movement of machinery and equipment thatcan change the sound pressure level, mainly in the areas of mining and processing. The use ofexplosives for blasting and the activities carried out in day-to-day life (workshops, offices, cafeterias andother) are also sources of noise.

The community closest to the location of the mine is the village of Água Branca, located about 4 km fromthe industrial plant. Considering the distance of the communities to the installations, the sound levels inthese communities will be impacted by moving vehicles, machines and equipment near the villages of

Água Branca and Pindobeiras.

A Noise Monitoring Program will be established with points and periodicity defined. To mitigate theimpacts of increasing sound levels in the vicinity of the installations, the project proposes the followingmitigating measures:

• The purchase of machinery and equipment with low noise potential;

• Establish routine procedures for motor inspection and preventive maintenance;

• To focus the activities of greater intensity of noise during the day, preferably between 8:00am and5:00 pm;

• The Noise Monitoring Program will establish a biannual monitoring during implementation andoperation phases, at the same locations used to determine the baseline data

In addition to the measures proposed by the project, the audit report suggested that every six months thenoise monitoring points are reviewed, in order to check for possible new developments and impacts in thevicinity of the project.

20.5.2 Biotic Environment

20.5.2.1 Fauna

The fauna environmental impacts will occur from vegetal suppression, earth-moving, soil exposition, solid

waste generation including organic food, atmospheric emissions, fugitive dust and gases, noise, thetransit of machines and vehicles on the roads and the changes to be introduced in ecological corridors.

The vegetal suppression causes the loss of fauna unintentionally by the invasion and change of the landuse. In the areas with the occurrence of amphibians and reptiles, these will suffer directly the effects ofsuppression of vegetation, with reduction of number of individuals. Bird impacts due to removal ofvegetation will be more intense during the deforestation, when dispersion of several species of birds insearch of refuge, food and safest places in the surrounding areas occurs.





According to the data, the identified endangered species are not restricted to a single habitat, but on thecontrary, occupy various habitats and have sufficient living area. The endangered fauna, mainlycharacterized by mammals (cats) are not restricted to the area of the project and move around easily.

Another aspect that can mitigate this impact is that the region offers a good amount of naturalenvironment where fauna can roam.

The presence of extensive vegetal refuges can support the mammals in dispersion, with goodenvironmental quality and connectivity, and may encourage the displacement of mammals and birds insearch of refuge, food and security.

The conservation of the João River valley, where, originally, the tailings dam was proposed to be built,also will mitigate this impact, since it is closer to the area to be suppressed giving shelter to fauna. Theherpetofauna will have some of their representatives dispersed to adjacent areas, but it can alsoexperience loss of local populations. Another reason for the loss of fauna will result from negativeecological interaction between species, competing for food and space.

The deforestation eliminates habitats where wildlife gets food, shelter and places to roam. To reduce thisimpact it is important to preserve areas with pristine vegetation in the surrounding areas of the mine siteand, in accordance with the standard 8 item 6 of IFC (International Finance Corporation), a net loss ofbiodiversity should be avoided. The requirement is a compensation of losses by creating ecologicallycomparable areas for the preservation of biodiversity.

According to the impacts described and to minimize the loss of wild fauna due to suppression, the projectenvironmental management plan proposes the following plans and programs:

• The Plan for Deforestation Actions. The main objective of this plan is to minimize the loss of localfauna by ensuring that deforestation actions are performed in a progressive manner and oriented ina single direction. This is intended to create opportunity for the spontaneous movement of animalsinto new areas and provide mechanisms and actions to prevent unauthorized intervention. The

plan also supports the presence of professional experts in wildlife management before and duringthe deforestation actions for the rescue and salvage of fauna.

• The Fauna Rescue and Relocation Plan. The objective of this plan is to rescue and relocate faunaindividuals unable to escape through the passages that would be created by the previous Plan. Theplan also provides for a Fauna Provisional Rehabilitation Center with infrastructure and equipmentrequired for such activity in order to facilitate the management of individuals saved in the area ofthe project.

• Acquisition and legal establishment of a conservation area with 2178 hectares of pristinearboreous caatinga, San Conrado Farm, Municipality of Iramaia, well beyond the legal requirement(Law nº 4771, September 15, 1965, Código Florestal) .

20.5.2.2 Flora

The vegetal suppression will lead to a series of impacts that will impact directly and indirectly the flora ofthe area of influence. The vegetation affected by the suppression is: bushy Caatinga (37.07 ha),bushy/dense Caatinga (167.91 ha), bushy/arborous Caatinga (1.53 ha) and dense/arborous Caatinga(13.14 ha). The vegetation is at various levels of conservation and diversity.





According to the studies, the potential impact of the project on the flora is negative. The potential impactwill have direct and local coverage, because it acts on the area directly affected (ADA) and the area ofdirect influence (AID), interfering negatively in the dynamics of surrounding populations.

To minimize those potential impacts, it is proposed that ethnobotanical programs be established (throughtargeted management) to encourage the continuation and multiplication of the species identified as

endangered and which have historic-cultural value and are rare. Such a program requires the collectionof seeds and seedlings of species (forms of germplasm), for their germination/development in the nurseryand the eventual reintroduction to the natural environment. For the success of this program, it is importantto identify an appropriate location for the maintenance of seedlings/seedling and training of staff.

20.5.3 Environmental Mitigation

The following plans and programs were proposed as indicated in the Table 20-11 and Table 20-12 as partof the mitigation measures to reduce the overall environmental impacts of the Project and are currentlybeen applied to the preliminary stages of installation phase.

These are intended to preserve (in the most feasible way) the current natural conditions of the site and toreduce the risks that could compromise worker health and safety.

No conditions were observed by Mineral during their project audit that would compromise theenvironment and worker safety. The Project has a very strong commitment to preserve natural resourcesand to improve the current social conditions at the Maracás micro region.





Table 20-11 Mitigation Measures – Construction Phase

Subject Mitigation Measure

Ditching and silt catchment

Geomembrane coating

Environmental Compensation Program

Remediation PlanPRAD - Plan of Rehabilitation of Degraded Areas

Plan of Deforestation

Fauna Rescue and Relocation Plan

Environmental Training and awareness Campaign

Environmental Compensation Program – Legal Reserve

Erosion Control Program

Flora Rescue and Relocation Plan

Ethno botanical Programs


Remediation ProgramEnvironmental Compensation Program – Legal Reserve

Discharge Management program

Water Monitoring Program

Environmental Compensation Program

Remediation Program

Storm Water Natural Drainage Modification Plan

Remediation Plan


Remediation Plan

Containment Ponds in Waste and Tailings StructuresSoil Remediation Plan

Noise Monitoring Program

Maintenance Program

Dust Control

Maintenance Program

Waste Material Disposal Waste Management Program

Accident prevention Plan

Emergency Plan

Fire And Explosion Risk Accident Prevention Plan

Workers Safety

Morphology

Surface/underground water

Storm water drainage

Soil

Noise

Air Quality

Erosion process

Land use

Fauna

Flora





Table 20-12 Mitigation Measures – Operations Phase

20.6 Social and Economic Environment

The Vanadium Maracás Project has expended substantial effort characterizing and understanding thesocio-economic context of the project as has been reported in the EIA studies prepared by Integratio,independent consultants responsible for social- communication aspects of the project.

This section summarizes major social and economic impacts arising from different stages of the projectas portrayed in the EIA and Audit report prepared by Mineral Engenharia e Meio Ambiente Ltda, in 2011.

20.6.1 Job and Income Generation

In accordance with the impact study, the duration of the installation period is estimated to be 22 months.During this period approximately 1,200 jobs will be created. The functional framework will be comprisedof machine operators, assemblers of metal structures (welders, boilermakers) and construction workers(bricklayers, carpenters and helpers). During this period, the estimated salaries and expenses will beUSD 15,000,000.

The offer of employment will constitute the main benefit attributed to the project by the population.

Subject Mitigation Measure

Ditching and silt catchment

PRAD - Plan of Rehabilitation of Degraded Areas.

Land use Erosion Control Program

Environmental Training and awareness campaignEnvironmental Compensation Program – Legal Reserve



Discharge Management program

Water Monitoring Program

Remediation Program

Storm Water Natural Drainage Modification Plan

Remediation Plan


Soil Remediation PlanNoise Monitoring Program

Maintenance Program

Dust Control

Monitoring Program (MP, NOx, SOx, CO, V and NH3).

Maintenance Program

Waste Material Disposal Waste Management Program

Accident Prevention Plan

Emergency Plan

Fire and Explosion risk Accident Prevention Plan

Erosion

Fauna

Flora

Surface/underground water

Noise

Air Quality

Workers Safety

Soil

Storm water drainage





During operations the project will be responsible for generating approximately 400 direct jobs with themajority related to the operation of equipment (crushing, grinding and concentration systems), as well asadministrative, managerial and operational positions.

Wage expenditures are estimated to be USD 5,000,000 per year. A significant portion will be spent in themarket of Maracás which will be stimulated. The direct spending for goods, services and materials

purchased on the local market and additional spending of Maracás County due to increased taxesrevenues will also add economic stimulus.

In order to increase the positive effects on the municipality, management has promised to give priority tothe hiring of local workers and to provide training to people to acquire the necessary skills required for the

jobs.

The work force training program is associated with the Environmental Management system, in order tocompensate the population directly affected by the project, as stated in the performance standard 1 ofIFC, "avoid, minimize or offset the negative impacts on workers, communities and the environment."Training or educating manpower can be an efficient way to benefit the local population with thegeneration of jobs, since the majority of the population does not have the skills or education required forthe venture.

20.6.2 Boosting the Local and Regional Economy

The municipality will notice an increase in economic activities which will result in the creation of new jobs,income generation and increased public revenue.

This stimulus will be motivated by workers’ wages, company spending and the municipal public

administration expenses.

New project investment is expected to provide the opportunity for new startup companies, as well as

established companies expanding. The increase in public revenue will not only engage additionalmanpower, but also increase the implementation of public policies for health, education, infrastructure,security, social and economic development.

The distribution of the expenses pertaining to investment (USD 250 million) will be as follows: 10% inMaracás, purchase of land, labor, taxes, transport and rents; 30% in the State of Bahia, purchase ofcement, supplies, taxes and services; 20% in the State of Minas Gerais, purchase of engineering,services, equipment and steel; 20% in the State of São Paulo, purchase of equipment and; finally, 20%overseas for purchase of equipment from South Africa and China.

In order to maximize this positive impact, the company will develop a training program for local supplierswith the objective of ensuring that local communities be included appropriately in the businesses that

have potential to affect them.

20.6.3 Improvement of Access and Roads

The Project will improve approximately 42 km of existing dirt road between the villages of Pé de Serraand Porto Alegre during the construction process. This road will potentially be further upgraded incollaboration with the Bahia State Government in the future. It is proposed that sections of the road will bereconstructed and the entire corridor will be paved with an asphaltic cover, improving the access between





villages. The timing of this proposed improvement has not yet been determined and no agreementbetween the Project and the State of Bahia is currently in place, although discussions are ongoing.

20.6.4 Pressure on the Educational System

Pressure on the educational system will occur due to the attraction of people from other localities. During

the estimated two year construction period the existing infrastructure is not estimated to be sufficient forthe absorption of additional families.

According to information obtained from the Education and Culture department of Maracás, theeducational system is overdesigned for actual educational duties. However since it is not possible toestimate accurately the amount of migrant families in the construction phase, it is necessary to start theproposed labor training program in order to minimize the need for recruitment from other municipalities.

Project management will also establish agreements with private schools of the region or will engage withthe public administration, to support if necessary, the expansion of schools. This planning shall adhere tothe Socio-Environmental Plan.

20.6.5 Pressure on the Public Health System

Maracás relies on Jequié as well as Planaltino, Iramaia and Lajedo do Tabocal for health support. Thismeans that part of the medical consultations, exams and emergency care of the population is conductedin Jequié and more complex cases are completed in Vitória da Conquista and Salvador.

Among the micro regional municipalities, Maracás has the better health structure and meets the minimumhealth demands of municipalities of this type. This health system stands out in the sphere of primaryhealth care, but with regard to clinical and hospital specialties care, the city presents weaknesses – anaspect that puts additional demands on a significant negative impact.

To mitigate such impacts, the project proposes a set of integrated initiatives including:

• Installation of a medical clinic in the camp with full time medical team where most clinical care andemergency demands will be met;

• Establishment of agreements with hospitals enrolled in the local and regional context for healthcare of workers and their families;

• Preferential hiring of local manpower to minimize the population growth and the consequentadditional pressure on health; and,

• Donation of medical equipment such as X-ray, ambulances, etc.

20.6.6 Pressure on the Water Supply System

Water supply, water capacity and water availability are heterogeneous and diverse in the municipality. InMaracás and in the district of Porto Alegre, communities that will receive the majority of migrants, thecurrently installed capacity will be subject to additional demand.





The villages of Caldeirãozinho, Pindobeiras, Jacare, Água Branca and Lagoa Comprida that rely on watersupply through trucks and tanks will require additional demand, and this will result in an additional stressto the current problems of supply.

The village of Pé de Serra is supplied by an artesian well with reasonable water availability.

The project will be erecting a 10 inch diameter, 27 km long, water pipeline that will bring raw water fromRio de Contas` dam to a water treatment plant in the site area for project consumption. The systemcapacity is designed for 250 cubic meters per hour and its construction is under way, as it will benecessary for supplying water during the construction phase.

As a socioeconomic compensation, the project will build a Water Treatment Station (ETA) at the closestvillage of Água Branca that will be handed over to the municipality for water distribution by themunicipality.

The surrounding communities of Jacaré, Lagoa Comprida and Caldeirãozinho are also expected tobenefit from improvements in water supply from an anticipated effective integration of the communitiessurrounding the project.

20.6.7 Inhibition of Immigration and Attraction of Population

Migratory flows generated by a large project are usually thought in terms of additional pressure that willfocus on public infrastructure, with an emphasis on health services, supply, educational, and publicsafety. Although this evaluation is relevant and plausible to be applied within the framework of socioenvironmental forecasts, in some cases, it disregards other important aspects.

The project maximizes the employability conditions in the Maracás area, by not only contributing to theareas in which it is proposed, but also, to the social and economic conditions of the large urban centersthat receive the migration flows. Maracás and other micro regional municipalities will compete for that

portion of the local population which has the opportunity to stay in their own environment (therefore in thefamily context), and achieve some degree of social and domestic success One of the major problemsexperienced in the country rises from cultural differences promoted by migratory processes includingfinancial losses.

The attractiveness of the project to the population are a function of its magnitude of investment, job-generating potential, and favorable socio-economic conditions within the area of influence especially thedemographic attributes and the general framework of employment.

Given the socio-economic conditions of the micro region, strongly marked by the lack of employmentopportunities, the installation and operation phases will attract a contingent of workers. Specifically,Maracás will be a "locus" of convergence of people from other localities; a fact that will change the daily

life of its inhabitants.

On the basis of not having social ties established within the communities and, therefore subject to minorsocial coercion of local cultural codes, relocated people tend to generate annoyances to the localpopulation, as well as infringing on their benchmarks of conduct and interaction. Therefore, it is necessaryto not only establish mechanisms of control over the population that will join the city, but also prepare thelocal community for contextual and cultural transformations. The County representative of the Social





Welfare group explained that one of the biggest concerns that emerge through the project is the increaseof crime, prostitution and drug use.

The proposed measures for solutions of this impact constitute a complex system of integrated initiatives.Information related to the Project and recruitment locations will not be concentrated in rural areascharacterized by poor infrastructure. Only major centers like Maracás, Pé de Serra and the Porto Alegre

will have information and recruitment workplaces.

With regards to hired employees, inside the framework of training activities, they will be oriented inrelation to the social relationship, with emphasis on the establishment of harmonious relations with thelocal community.

Regarding the community, the company will support educational and communication activities to becarried out by local institutions. Instead of proposing educational actions (centralizing and inspiring socioenvironmental concerns of its interest), the company programs will combine the various communicationgroups and social institutions to guide the local population in the effects resulting from the additionalpopulation.

These initiatives are intended to recognize and help find solutions to the anticipated problems ofprostitution, unwanted pregnancy, increased consumption of drugs, and at the same time strengtheninstitutions through reflection and engagement in matters relating to the society.

With regards to public security, the company (via the Social Communication Program), will be inpermanent interaction with the authorities supporting monitoring programs as well as the support for theactivities in line with the variability of demands.

Finally, it should be noted that the preferential hiring of local labor constitutes a measure which willminimize the number of people outside the municipal environment, both during the installation andoperational phase. It is the goal to recruit a significant percentage of people from the local community,

who already form the local context.

20.6.8 Expectations and Population’s Opinion about The Project

The expectations of the population in relation to the project are based on field surveys conducted by theteam that prepared the EIA with the communities surrounding the project, the villages of Pé de Serra andPorto Alegre and the small communities of Água Branca, Lagoa Comprida, Pindobeiras, Caldeirãozinhoand Jacaré. At Maracás, interviews were conducted with institutional representatives of the Departmentsof Urban Planning, Public Works and Services, Industry, Commerce and Agriculture, Education andCulture, Health and Welfare. Also included were Mineral team impressions, which developed exploratoryinterviews with some residents and local leaders in 2011.

The perception of various social segments enrolled in the Maracás project has significant convergence,and the expectations are high, mainly due to the county having few employment opportunities and theProject is seen as a possible source of jobs and income. Fundamentally, there is social awareness of thepositive effects of the economic plan, which includes the creation of jobs and their effects on the generalquality of life. Faced with a socioeconomic context characterized by unemployment, (where the municipalgovernment is characterized as a major employer), the decrease of agricultural activities and themigration from the rural areas to the city, the project emerges as a landmark for the establishment ofsocial and economic conditions in the municipality.





During a public hearing, people received information about the project, and the construction of the tailingsdam in the Jacaré river. Concerns were raised about the tailing dam since the water supply is always amajor concern. However, the conventional tailings dam was subsequently replaced by a “dry -stacked”

method, which mitigated the concern. Another factor that causes discomfort among residents is the hiringof local labor. Overall, residents of Maracás communities do not have the formal education orqualifications to meet specific job skills, however the outlook is positive with the opportunity for

professional preparation since the region is poor on technical and vocational schools.

With regards to social issues, the main concerns were related to increased flow of people and relatedeffects such as an increase in prostitution, theft, drug trafficking/consumption and violence. In order tominimize or resolve these issues, the project management proposes to implement a comprehensiveinformative program of Social Communication (PCS), which contemplates the democratization ofinformation about the various impacts of the project and also the initiatives of mitigation and enhancementthat will be adopted.

The PCS implementation is under way, and five actions are being developed: (I) Participation of thecommunity, (ii) Disclosure, (III) The consultation process, (IV) Mechanisms for complaints and (V)

Monitoring by a specialized person. These actions are part of the Environmental Management System ofthe project. The environmental concerns of the population of the county’s capital are rare, but they are

more frequent in the villages close to the Project, because the focus is on job and income issues.

Though the expectations are high towards the Project, a majority of people do not know much about thevanadium ore to be mined, its uses, or even how it will be processed. To some, the Project will onlybenefit the local population through jobs.

In summary, it is clear that the population is excited about the possibility of creating jobs and increasingincome, but it is also clear that the population knows very little about the Project or the type of activity thatwill take place in the region. Thus, it is necessary to carry out information events to inform the populationabout the Project as it is planned through the PCS.

20.7 Geotechnics and Hydrology

Hydrological and geological characterization studies were completed in 2008 by VOGBR and revised in2011 for Basic Engineering. Below a summary of these studies is presented.

20.7.1 Hydrological Studies

The pluviometric and fluviometric data that were adopted for this current study were obtained from theBrazilian National Waters Agency – ANA – Agencia Nacional de Águas. The selection of the stations wasbased on their location. The selected stations are:

Pluviometric Station at Fazenda Alagadiço, ANA code - 01340019, and;

Fluviometric Station at Roçados, ANA code - 52265000.

The climatological characterization has been based on average monthly figures from the Ituaçu Station(code 83,292), obtained from the INMET Publication entitled “Normais Climatológicas”, 1992.

(a) Climatological Characterization





The Project area is located in the southwestern region of the State of Bahia (Brazil), in the municipality ofMaracás. The climate, pursuant to the Koppen classification (Climatologia do Brasil, Edmon Nimer,1979), is determined as being hot and semi-arid tropical, with 6 dry months. The average annualtemperature is 24°C, with December, January and February being the hottest months and July and

August those with the lowest temperatures.

The pluviometric regimen is characterized by one rainy period during the summer during which thewettest quarter is from November to January and by one dry period in the winter during the quarterranging from June to August. The average annual rainfall is approximately 600 mm. Figure 20-3 showsthe average monthly rainfall for the Project region.

The region’s annual evaporation rate is high, close to 1600 mm, with approximately 180 hours average

monthly sunlight and average relative humidity between 50% to 73%, pursuant to the average monthlydata obtained from the Ituaçú weather station.

(b) Hydrographical Characterization

The Project area is inserted within the hydrographical basin of the creek that is locally known as “João”,

which is an affluent to the right bank of the Jacaré River, which in turn is a tributary to the Rio de ContasRiver. The basin extends over approximately 57.4 km2 and the length of its main course is 18 km, with550 m difference of elevation; water flows are intermittent.





Figure 20-3 Average Monthly Rainfall for the Pluviometric Station at Fazenda Alagadiço (ANA Code -01340019)

(c) Intense Rains

The main characteristics of the intense rains include the total amount of rainfall, their spatial and temporaldistribution and their frequency of occurrence. The interest given to the study of the intense rains isrelated to the use rainfall volume data in the hydraulic dimensioning of the various project structures.

The figures relative to various return periods for rainfall height and intensity ratio, and for duration andfrequency which are presented in Table 20-13 and Figure 20-4, respectively, were derived on the basis ofstatistical treatment of daily rainfall volume figures obtained from the weather station at Fazenda

Alagadiço.

0

10

20

30

40

50

60

70

80

90

100

110

120

Jan Fev Mar Abr Mai Jun Jul Ago Set Out Nov Dez

Month

R a i n f a l l ( m m )

Annual Total = 631,7





Table 20-13 Rainfall Height Rates (mm)

Figure 20-4 Intensity, Duration and Frequency Curves

2 5 10 20 25 50 100 200 500 1 10

5 min 5.76 7.29 8.3 9.27 9.58 10.5 11.5 12.4 13.6 14.6 17.7

10 min 10.3 13.1 14.9 16.6 17.2 18.9 20.6 22.2 24.5 26.1 31.7

15 min 14.1 17.8 20.3 22.6 23.4 25.7 28 30.3 33.3 35.6 43.1

20 min 17.2 21.7 24.7 27.6 28.6 31.4 34.2 37 40.7 43.4 52.725 min 19.8 25.1 28.6 31.9 33 36.2 39.5 42.7 46.9 50.1 60.8

30 min 22.1 28 31.9 35.6 36.8 40.4 44 47.6 52.4 55.9 67.8

1hr 31.4 39.7 45.2 50.4 52.1 57.3 62.4 67.5 74.2 79.3 96.2

2hr 40.4 51.1 58.2 65 67.2 73.8 80.4 87 95.7 102 124

4hr 48.5 61.3 69.8 78 80.6 88.6 96.5 104 115 123 149

6hr 52.7 66.7 76 84.8 87.7 96.3 105 114 125 133 162

8hr 55.6 70.3 80.1 89.4 92.4 102 111 120 132 141 171

10hr 57.7 73 83.1 92.9 95.9 105 115 124 137 146 177

12hr 59.4 75.2 85.6 95.6 98.8 109 118 128 141 150 182

14hr 60.9 77 87.7 97.9 101 111 121 131 144 154 187

24hr 65.8 83.2 94.7 106 109 120 131 142 156 166 202

DurationReturn Period - TR (years)

1

10

100

1.000

1 10 100 1.000 10.000

Duration (minutes)

I n t e n s i t y ( m m / h )

T = 2 years

T = 5 years

T = 10 years

T = 20 years

T = 25 years

T = 50 years

T = 100 years

T = 200 yearsT = 500 years

T = 1.000 years

T = 10.000 years





(d) Design Output Volumes

The studies to calculate the design output volumes were designed to provide input data for use in thehydraulic dimensioning of the structures envisaged for the Project.

Used as a criterion, the design output volumes correspond to 1,000-yr return period high water events.

Table 20-14 presents the design flow volumes for the envisaged hydraulic structures.

Table 20-14 Design Flow Volumes for the Hydraulic Structures

(e) Site Water Balance

The process plant make-up water during operations is 81.6 m3/hr where 75.0 m3/hr is provided from theRio de Contas and 5.6 m3/hr being provided from the water content in the mined ore. The licenseprovided by the federal water agency, ANA, (Agência Nacional De Aguas) allows a water-take of 300m3/hr from the Rio de Contas. The pumping system from the Rio de Contas is sized at 200 m3/hr. Thereis a circulating water load within the plant with the net make-up being the aforestated 75 m3/hr.

Originally, the plan for a conventional slurried tailings system with a tailings pond resulted in a greaterdemand for water. This water demand has been greatly reduced with the introduction of “dry -stacked”

tailings.

20.7.2 Geological / Geotechnical Characterization of the Overall Project Area

The geological/geotechnical characterization of the project area is intended to provide input data forengineering design work, namely: the pit, process plant installations, waste and stockpiles, tailingsdisposal system and flood control system.

Figure 20-5 shows a water balance flow chart, involving the structures under consideration and theircorresponding flow rates.

20.7.3 Geological / Geotechnical Characterization of the Overall Project Area

The geological/geotechnical characterization of the project area is intended to provide input data forengineering design work, namely: the pit, process plant installations, waste and stockpiles, tailings

disposal system and flood control system.

Structure Elements Flow Volume (m³/s)

Northern Ridge 20.6

Southern Ridge 202.4

Sediments Catchment Dyke

Flood Control System





Figure 20-5 Flow Chart of the Water Balance for the Project

The main activities performed to achieve the proposed objectives are as follows:

overall geological / geotechnical mapping of the entire installation area, envisaging the recognition,characterization, distribution and use of the overburden and outcropping substrate rock materialsas cut and fill and foundation material;

mapping of areas subject to flooding and waterlogged areas, expressive erosion features,landslides and areas with rock boulders;

analysis of field investigations and monitoring, consisting of inspection test pits, geotechnicaldrilling campaigns to allow direct viewing of the different types of soils / saprolite "in situ";

elaboration of geological / geotechnical vertical sections for the main units within the development,and diagnosis of the foundation conditions of the main component units of the Master Plan, and

diagnosis of the foundations conditions of the main components units of the Master Plan.

20.7.4 Overall Geological / Geotechnical Characteristics

The geological / geotechnical mapping for the Project area, in conjunction with the geotechnicalinvestigations, allowed the identification, characterization and distribution of the different types ofmaterials throughout the area as illustrated in Table 20-15.





Table 20-15 Characterization of Soils and Outcrops in the Project`s Area

(a) Geological / Geotechnical Characteristics of the Unit’s Foundations

Process Plant

The industrial area is planned to be installed northwest of the future pit, where natural elevations rangefrom 300 to 325 m. NSPT hammer rating generally varies from 20 to 35 strokes. The top of the rock layeris in general very shallow, between 1.0 to 4.0 m. However, the top of the rock layer can be deeper inlocalized areas (down to 10 m). The band of weathered rock is in general 2.0 to 3.0-m thick, below which

bedrock is encountered. Ground water level varies from 0.5 to 12.0 m.

Non-Magnetic Dry Stack Tailings Dump

The area destined for the installation of the Non-magnetic Tailings Dry Stacking area is located northwestof the Gulçari A pit within the João Creek basin. The residual soils encountered throughout the area areconstituted by clayey-silty material, which presents sandy portions of very firm to hard consistency withinits matrix. The contact between the soil / rock occurs not deeper than 4.0 m within the area. Groundwater level varies from 0.5 to 12.0 m.

Flood Control System

Brownish silty colluvium (northern ridge) or red silty-clayey colluvium (part of the southern ridge) arepresent but both with thin layers. Underneath, residual soil / saprolite is found, with SPT figures normallyhigher than 30. This material can present minimum thickness of 1 to 2 m, close to drainage channels, andranging up to approximately 11 m in higher regions. The top of the rock layer, in these places, occurs atdepths of around 12 m. Groundwater is encountered close to the soil / rock contact depth.

Waste Pile

Type of Material Tactile-Visual Classification Average Thickness (m)

Alluvium

Yellowish brown, fine to medium-sized sand, with boulders and

pebbles of quartz, and other more rarely encountered types of rounded rock (quartzites, andesites). 0.2 to 0.5

Red ColluviumClayey si lt and / or sandy si lt matrix, red colored with smal l

amounts of pebbles and granules of quartz or other rock.0.5 to 3.0 (Max. 8)

Brown ColluviumBrown colored matrix, normally silty, variable quantities of

granules and pebbles of weathered rock and / or quartz.0.5 to 2.0

Residual Soil /

Saprolite

Soils consisting of clayey-silty material with amounts of sand and

fragments of weathered rock, which may contain fol iation

structures and veins of quartz.

1.5 to 3.0 (max. 10)

Outcrops

Slightly or very weathered and / or fractured undifferentiated

rocks, consisting of granitoids, gabbros, pegmatites, andesites,

etc.





The first waste pile is located southwest of the Gulçari A pit. The region is overlain with brown colluvialsoil less than 1-m thick. The residual soil / saprolite, which is firm to very firm, is no more than 4-m thick inplaces. The NSPT for this material is close to 50 strokes. The top of the rock layer is encountered at thevery shallow depth of around 3 m. A second waste pile is planned for placement to the east of the Gulçari

A pit to accommodate the Phase 2 waste material. In addition, a waste pile will be constructed for eachsatellite pit.

Tailings Disposal Ponds (Leached Calcine and Chloride Purge Ponds)

Both the leached calcine and the chloride purge ponds, which are located south of the industrial plant, arealmost entirely located over silty brown colluvium that contains varying amounts of pebbles. The layer ofresidual soil / saprolite in these areas is encountered at a depth of 4 or 5 m. The top of the rock layer canoccur in places at down to 10-m depth. The water level was estimated to occur, in more topographicallyelevated portions, at approximately 6-m depth.

20.7.5 Geotechnical Studies

a) Non-Magnetic Dry Stack Tailings

This section presents a review of the basic design of the Non-Magnetic Dry Stack Tailings Dumpelaborated by Ausenco Minerals, 2010 on the basis of the geotechnical parameters provided by VOGBR.

The dam`s original design completed by VOGBR in 2007 envisaged a conventional valley type damsituated on the João Creek at the latitude 13° 42’ 28”S and longitude 40° 42’ 11”W, in a pristine forested

environmental. It was projected to receive the primary tailings, to retain some of the sedimentsoriginating from the creek basin and to assure a permanent water flow from João River. A study of analternative location for this type of structure was avoided with adoption of the Dry Stacking concept,eliminating a major concern of the nearby populations - a dam break after intense rains.

The new tailings facility consists of a series of ponds formed by rock-fill structures sealed by compactedclayey / saprolitic material, as illustrated in Figure 20-7. The proposed arrangement will suffice to containaround 6.25M m3 of tailings that will be produced during the mine’s first 10 years of vanadium productionfrom the Gulçari A Phase 1 pit. Further study is required to expand this tailings facility to adequately meetthe storage requirements of the remainder of the project life including the Gulçari A Phase 2 pit andsatellite pits.

The construction sequence will advance through stages. Two ponds will be constructed during pre-production year 2 and production year 1 for dry stacking of nonmagnetic tailings close to the plant, andwill suffice for the first two years of plant operation. Tailings from Year 3 and beyond will be transported toan expanded version of the pond system shown in Figure 20-7.

The pond construction sequence consists of stages of vegetation clearing, removal of organic topsoil andother material until the depth appropriate for the foundations of rock fill structures.

The current design of the pond structure will be (considering a working life of 10 years) 9m high, 820mlong, 900m wide and will have a volume of 624 000m3. Figure 20-6 and Figure 20-7 presents a sectionand a layout of the Non-Magnetic Dry Stacking Tailings Dump.





Table 20-16 presents its most relevant characteristics.

Figure 20-6 Typical section of the Non-Magnetic Dry Stack Tailings Dump contour structure

Figure 20-7 Layout of Non-Magnetic Dry Stack Tailing Dump

* Dry stack tailings cell system design to be expanded in the future to accommodate tailings storage requirements

beyond production Year 10.





Table 20-16 Main Characteristics of the Non-Magnetic Dry Stack Tailings Dump

* Dry stack tailings pond characteristics to be updated with revised design to accommodate tailings storage

requirements beyond production Year 10.

(a) Sediments Catchment Dike (Waste pile)

To prevent the silting of João River with the fines washed from ore and waste stockpiles and majorstructures from mine site, the construction of a dike, named Sediments Catchment Dike, has beenenvisaged.

This dike, as well as the pit-protection ridges will be comprised of rock fill originating from the pit togetherwith compacted earth and transition material placed between the foregoing.

The dike will be constructed at João Creek at latitude 13º 41 46`` and longitude 46º 41`32`` and its designwas conceived to catch and capture the entire run off drainages around the site assuring enough

residence time for fines sedimentation during intense rains. The barrier and crest were designed withfiltering properties allowing the water to percolate and/or surpass the crest. The open drain systemconsists of a channel installed in the central region of the dike, with trapezoidal geometry and rockprotection and a dissipation basin against erosion potential.

The main characteristics of the sediment Dike are presented in Figure 20-8, Figure 20-9, and Table 20-17show the location and a cross section of the barrier.


Length of stack structure (m) 820

Width of stack structure (m) 900

Maximum crest elevation (m) 309Minimum downstream elevation (m) 295

Center road width (m) 8

Height of slopes between berms (m) 4

Tailings capacity - (m3) 6500 000

Maximum area occupied (m2) 740 000

NON-MAGNETIC TAILINGS DAM





Figure 20-8 Map showing the location of the Sediment Catchment Dyke

Figure 20-9 Section of the Sediments Catchment Dyke

Table 20-17 Main Characteristics Catchment Dike

Maximum height (m) 9.5

Length of structure (m) 455Maximum crest elevation (m) 288.5

Minimum downstream-side elevation (m) 287

Slant of embankment 1V:2,5H

Crest width (m) 6





(b) Tailings Disposal Ponds

The tailings generated by the process are of three types: leached calcine from the processing kilndischarge, filter cake from the desilication process and chloride control purge from the evaporation circuit.

The leached calcine tailings and the desilication process tailings will all be discharged into the same

structure (pond). This structure has been designated herein as the Leached Calcine Tailings Pond. Thechloride control purge tailings are to be deposited in a separate pond. The envisaged constructionlocation for these ponds is northwest of the open pit, close to the industrial plant.

The dikes will be built using compacted earth and their base areas will be leak-proofed using double-layergeomembrane liner featuring a leak detection system. The construction consists of clearing vegetationfrom the areas to be occupied by the ponds, removal of organic material and excavation of materialinappropriate for foundations. The perimeter of each pond will be protected by rock-fill channels.

The leached calcine tailings pond will be built initially as a first stage to handle the first two years ofproduction. The pond will later be expanded, if required, possibly in a different location, to accommodateLOM tailings production for the expanded production scenario.

The Chloride Control Purge Pond will be built at the outset to accommodate 3 years of tailings production.The pond will later be expanded, if required, possibly in a different location, to accommodate LOM tailings

Typical sections of the ponds are presented in Figure 20-10 and Figure 20-11, respectively.

Figure 20-10 Typical Cross-Section of the Leached Calcine Tailings Pond





Figure 20-11 Typical Cross-Section of the Chloride Control Purge Pond

The main geometric characteristics of the Leached Calcine Tailings Pond and the Chloride Control PurgePond are presented in Table 20-18 and Table 20-19.

Table 20-18 Main Geometric Characteristics of the Leached Calcine Tailings Pond

* Leached calcine tailings pond sized according to 10 years of operation at the expanded production rate. Design to

be updated to accommodate tailings storage requirements beyond production Year 10.

Table 20-19 Main Geometric Characteristics of the Chloride Control Purge Pond

* Chloride control purge pond sized according to 10 years of operation at the expanded production rate. Design to be

updated to accommodate tailings storage requirements beyond production Year 10.

Maximum dike height (m) 12Area occupied (m2) 120,000

Maximum capacity - volumetric (m3) 2 500 000

Slant of pond slope 1V:2H

Slant of pile slope 1V:3H

Maximum height of the slopes between

the pile berms (m)10

Pile berm width (m) 5

Pond crest width (m) 7

Leached Calcine Tailings Pond


Area occupied (m2) 10,000

Maximum capacity volumetric (m3) 17500

Slant of pond slope 1V:2H

Pond crest width (m) 7

Chloride Control Purge Pond





(c) Flood Control System

The area destined for the Gulçari A Phase 1 open pit intercepts the João Creek and three directtributaries. Consequently, this justifies the installation of a protection system to impede the influx ofsurface water runoff into the pit to allow mining activities to proceed.

For the initial phase, VOGBR, proposed that the pit protection system consist of dikes and channels. Theselection of this design allows for a good balance between the volume of earth cuts and landfills.However following ensuing project development, it was decided to use a system of protection ridgesbecause of the proximity and the economical usage of waste material from within the open pit.

The said system was designed with a pair of ridges located around the open pit, identified as the NorthernRidge and the Southern Ridge. The objective is to redirect the water downstream from the open pit. Theprotection ridges form a barrier around the open pit to intercept the watercourses and to raise their waterlevels, so they can naturally pass the topographical elevation variations and flow by gravity bearing thesurface runoff downstream from the open pit.

The definition of the structures was in keeping with the adoption of a 100 m minimum distance from thefinal pit limit. Furthermore, issues relating to the volume of landfill, feasibility of execution and hydraulicworkings were also considered in efforts to optimize the adopted system.

The protection ridges will be installed pursuant to the mining activity plan defined for the Gulçari A Phase1 pit. The Northern Ridge should be completed by the close of Year 2, while the Southern Ridge will berequired for mining activity development as of Year 3. It should be noted that vegetation clearing, removalof organic materials and excavation of no appropriate materials for foundation must precede theforegoing.

The use of overburden material from the Gulçari A Phase 1 pit was considered as a premise whendimensioning the protection ridges. Accordingly, the ridge structures were designed to consist of rock fill,

transition material and compacted earth (residual soil / saprolitic material). Table 20-20 presents asummary of the main characteristics of the ridges.

Table 20-20 Main Characteristics of the Ridges

This item presents and analyses the arrangement of the waste pile and ore stockpile elaborated by NCLdo Brasil on the basis of geometric parameters provided by VOGBR.

The material from the open pit will be placed on the waste pile or the ore stockpile, as predominantly rockconsisting of varying sizes of boulders. The area of the Gulçari A Phase 1 waste pile coversapproximately 48 ha. The updated engineering and production planning carried out from 2010 onwards

Characteristics

Northern Southern

Maximum height (m) 14 13

Length of structure (m) 915 875

Maximum crest elevation (m) 312 313

Minimum downstream elevation (m) 298 300

Slant of downstream-side embankment 1V:1.5H 1V:1.5H

Crest width (m) 6 6

Ridges





does not consider the stockpiling of low grade magnetite-pyroxenite ore, but rather considers feeding theplant with a blended V2O5 grade material.

NCL elaborated the arrangements for both the waste pile and the ore stockpile based on the geometricparameters provided by VOGBR. The elevation of the waste pile crest is 360 m.

The geometry of the proposed waste pile stipulated the following parameters:

Bench face angle = 36° Overall angle = 26° Berm width = 8 m Height of banks = 10 m Width of ramps = 15 m Maximum pile height = 60 m Area occupied = 48 ha Capacity = 30 Mt.

RPM has further refined the Gulçari A Phase 1 waste pile design to occupy an area of 47 ha, maintainingall other design criteria listed above. With the addition of Gulçari A Phase 2 pit, the North and Southridges originally designed for redirecting water around the Gulçari A Phase 1 pit may need to be relocatedor rehandled at a later date. Further study will be carried out on these ridges in the future, as well as anyadditional ridges required for the satellite pits to the north.

20.8 Current Activities and Plans

Maracás project construction is under way and the key socio-environmental activities that are beingimplemented are vegetal suppression, fauna and flora rescue and relocation, and public communication.

The Action Plan containing the integrated framework of environmental management plan components

during operations is required to be developed and applied at the project by an adequate environmentalorganization and structure in a suitable time frame.

20.8.1 Project Organization and Sustainability team

In order to fulfill Largo’s commitments to complying with the Equator Principals, a ‘Sustainability Team’

was created in February 2011. It is comprised the following skills and positions:

a senior Safety/Health/Environmental/Community manager responsible for the project; a safety engineer; an environmental engineer; a psychologist; biologists; and, several technicians responsible for the monitoring, communication, re-vegetation, and reclamation

activities

The Safety/Health/Environmental/Community manager and the safety engineer report directly to theProject Construction Manager. The participation of senior corporate management relating to theenvironmental aspects of the project is required for the implementation of the EMPs described in the

Actions Plan.





20.8.2 Equator Principles Audit Review

As mentioned previously, Largo, and its parent company Vanádio de Maracás S.A, have committed tocomply with best management practices accepted by the International Mining Industry, all Brazilianenvironmental laws and regulations, and the conditions of all environmental permits issued for the project.

A finance consortium composed of Itaú-BBA, Bradesco and Banco Vororantin, have provided a portion ofthe finance for the project in the form of a loan. A requirement of such financial Institutions/IFCGuidelines related to participation in projects such as the Maracás project is the performance of asocio/environmental audit by a qualified, independent third party. In discussion with Bank representatives,it was agreed that the audit could be performed by Mineral Engenharia em Meio Ambiente Ltda (Mineral),a well-known environmental engineering consulting firm, familiar with project permitting in Brazil andhaving recognized expertise in the performance of environmental audits.

The scope of the audit followed generally accepted auditing principles, reviewing the Ten EquatorPrinciples and respective Performance Standards.

As part of the background review, an evaluation of the permitting and regulatory environmentsurrounding the project was completed. This involved a review of the currently issued permits andpermits that are pending or in the process of being obtained. During the EIA process, a number ofcommitments were made by the project in the form of preparing and adopting a series of EnvironmentalManagement Plans. The content, status and implementation of these plans was reviewed and assessed.The EMP review and the implementation of the environmental management process required for theproject was also assessed in terms of the status of the project during the audit. Included in this aspect ofthe audit was a review of the monitoring of the project by Largo’s corporate personnel, and the project’s

compliance with its permits and internationally accepted best management practices.

20.8.3 Socio-Environmental Action Plan and Environmental Management System

The Action Plan is the accomplishment of Equator Principle nº 04. It states that an action plan should bedrawn up with a description of the actions required for management of the mitigating measures, correctiveactions and supplementing measures identified through the Environmental Assessment.

The Action Plan was drawn up by Largo corporate staff jointly with Mineral, and it incorporates theprograms necessary for compliance with the Brazilian laws and regulations and applicable environmentalperformance standards and EHS guidelines.

The list of programs is presented in the next section, Table 20-21 which describes each program.





9) Plan Rehabilitation of Degraded

Areas Plan to manage the rehabilitation and

reclamation activities of the mine siteConstruction and Operation

10) Plan of Social Communication

Plan to manage the public and internal

communication to assure transparency and

democratization of information. Construction and Operation

11) Plan of Hiring Local People Plan to hire 60% of people from Municipality of

Maracás Construction

12) Plan of Labor Training Plan to manage the training of skills and

competences of local people for project

opportunities

Construction and Operation

13) Plan of Guidance of Local

Suppliers Plan to guide local suppliers on project

opportunities for new business Construction

Risk Management Plan- PGR Environmental Risks Prevention Plan – PPRA Health Control Plan- PCMSO

8) Environmental Monitoring Plan Plan to monitor specific aspects throughout the

life of the mine. Establishment of an “ObserverConstruction and Operation

16) Closure and Reclamation PlanPlan to identify the concurrent and ultimate

reclamation and closure of the mine area andDecommissioning

14) Health and Safety Plans Construction and Operation

15) Contingency Plan Plan to identify e enforce actions in the event of

unforeseen events or an “upset” condition andConstruction and Operation

Table 20-21 Action Plan – Environmental Programs

The Maracás Vanadium Project must meet the standards required with respect to the legal aspects of the

environmental licensing process, as the License of Implementation and Environmental Control Programsdemand and be consistent with the Equator Principles as required by the International FinanceCorporation (IFC) and the banks (Itaú, Votorantim and Bradesco) financing a portion of the Project.

Integratio has been involved with the Maracás Vanadium Project since November 2012 providingguidance and support for the physical, biotic and anthropic/socio-economic characteristics of the project.The objective of the integration of the three characteristics is to meet the requirements of the PD and theIFC, while observing the conditions and restrictions of all licenses.

Program Description Phase

1) Environment, Health and Safety

Organization Plan

Plan to manage, apply and supervise the Safety,

Health and Environment plans


2) Fauna Management Plan Plan to manage and protect wild fauna. Construction and Operation

3) Forest Management Plan Plan to manage and protect forest. Construction and Operation

4) Aquatic Biota Management Plan Plan to manage and protect forest. Construction and Operation

5) Water Management PlanPlan to manage industrial and domestic waste

waters and their treatment prior to releases toConstruction and Operation

6) Surface Water Management PlanPlan to manage the meteoric water and the

mitigation measures to storm events andConstruction and Operation

7) Waste Management PlanPlan to manage the solid waste, inert industrial

wastes and oily contaminated soils at theConstruction and Operation





The audit company for the PD is Mineral Engineering and Environment (MEE). To date, they haveperformed two audits; the first being in December 2011 and the second being in August 2012. MEE andthrough the course of their audits have identified certain non-conformities. These non-conformities arebeing addressed. Existing response mechanisms are being amended and where no responsemechanism exists, new ones are being developed.

The general planning consists of the following:

Situational Diagnosis including;o Analysis of the Social Environmental of the projecto Stakeholders Mapping and Analysis

Assessment of the Requirements and Referential of Development Performance Standards of the IFC Conditionings of the Environmental Control Programs Existent activities and commitments

o Impacts and expectations for the role of the Company by the Stakeholders Geographic scope (Areas of influence)

Relevant environmental questions Operationalizationo Action Plan

Macro Plan Detailing Monitoring of implementation

o Evaluation Internal pre-audits External audit (prepping and monitoring) Periodic Follow up with financial agents and external audit

Internal Governanceo Constitution of the Environment Evaluation and Management System (Standard 1 of

the IFC) Definition of the form the full monitoring of the physical, biotic and social

economic dimensions of the Projecto Empowerment of the Sustainability Coordination on the Management Practiceso Implementation of the Social Environment Internal Commiteeo Addressing and response to the internal demandso Definition of the monitoring and evaluation perspectives (performance indexes)

Development of the Social Investment and Responsibility Guidelines Design of Projects and Programs of local development and mitigating the environmental impact.

During MEE’s audit, Performance Standards 1 (PD-1) was adhered to which focuses on the structure ofthe Environmental Evaluation and Management System. Integratio’s priority is for the consolidation of

these systems to enable the company to answer the social, environmental, and economic demands of theMaracás Vanadium Project. The next audit was scheduled for February 2013.

It is important to emphasize the social and environmental investment requirement of the Banco Nacionalde Desenvolvimento (BNDES), which is to strengthen the social and economic development of theregions that will receive investments.





The requirement of addressing the social and economic values is a new approach for BNDES which nowplaces an emphasis on the initiatives of sustainable development so as to not cause social, economic andenvironmental repercussion to the surrounding regions of the project.

The Main Equator Principles Framework for VMSA Social Investments is as follows:

Principle 2: Social and Environmental AssessmentThis assessment proposes mitigation and management relevant and appropriate to the natureand scale of the proposed project.

o Social assessment (Main Impacts from Environmental Impact Study) Population expectations Provide employment and income Stimulation of local and regional economy - investment

Principle 3: Applicable Social and Environmental Standards Assessment will refer to the applicable IFC Performance Standards

o Performance Standard 1: Management System and Environmental Assessment Stakeholders engagement

Principle 5: Consultation and DisclosureFor projects with significant adverse impacts on affected communities, the process will ensurethere is informed consultation and opportunity for participation as a means to address any issues.

The Main IFC approach for VMSA Social Investments is as follows:

Addressing the Social Dimensions of Private Sector Projects - Good Practice Note (IFC)o Traditionally, the “do-no-harm” approach of the World Bank Group's social safeguard

policies has made social mitigation plans the primary entry point for distributing benefitsto local communities impacted by IFC investments.

o Unlike mitigation and compensation which have important but limited objectives ofprotecting affected persons from adverse impacts, sustainable development actions

enable the wider population in a project's area of influence to gain access to and takebetter advantage of the range of opportunities brought about by private sectordevelopment.





21 Capital and Operating Costs

21.1 Project Capital Cost

Project Capital Costs, as of March 2013, are estimated to be USD 389.1 million including an allowancefor contingencies of USD 22.0 million, equivalent to 6% of total capital expenditure. The capital costsummary as presented in Table 21-1 are for actual expenditures through December 2012, USD 33.4million, the total pre-production capital through to December 2013 of USD 230.4 million and remainingother capital and sustaining capital costs of USD 125.4 million from 2014 to 2042. The total capitalexpenditure during the operation is USD 355.8 million, including acquisition to increase or replace minefleet equipment, plant and infrastructure.

A 2.00 BRL/USD exchange rate was applied to the project economics.


Capital ExpenditureSunk Cost

(2011-2012)

Remaining

2012 20132014

- 2042 Total


Enivronment 1,411 1,819 1,287 - 4,517

Management of Works 377 3,184 3,529 - 7,090



Civil Works 37 33,829 12,580 - 46,446

Equipment 2,012 47,959 5,633 - 55,604


Administration Costs 1,292 64 56 - 1,411

Consulting 10 64 56 - 129

Acquisit ion of Land 3,625 2,730 1,910 - 8,265

Acquisit ion of Mineral Rights 4,900 - - - 4,900


Seguros - 304 321 - 625

Contingency - 13,028 8,914 - 21,941





Closure Cost - - - 8,418 8,418 Total Capital Expenditures 33,365 170,752 59,594 125,405 389,117





21.2 Operating Cost Summary

The operating expenditure is based on contractor mining during the preproduction years to Year 3 of theproject. From Year 4 onward, mining will be conducted by Largo. The strategy was determined as themost cost effective for the operation and ensures sustainability of a skilled labor force.

The average unit cost for operational activities is USD 61.50/t of ore. The lifetime annual average of alloperating costs included from years 1 to 29 amounts to USD 88.1 million. The breakdown of mining,processing, general and administration costs, insurance and royalties as presented in Table 21-2.


The average unit production cost for V2O5 from Year 1 to 29 is USD 3.10 per pound. Given the plannedcompletion of construction and commissioning of the FeV plant for Year 3, the average production costfrom Year 4 to 29 for FeV production is USD 15.26 per kilogram.

Mining and Process Plant operating costs are largely variable per tonne of product while the General and Administrative costs are fixed per year. RPM has reviewed the basis of the operating cost estimate andconsiders the costs to be appropriate for evaluating economic viability of the project.


ROM

USD/Tonne

ROM

Mining 28.59 14.29

Plant 75.30 37.65

G&A 4.07 2.03

Insurance 1.38 0.69

Royalty 13.66 6.83






22 Economic Analysis

The economic analysis was completed for a 1.4 million tonne per year processing plant capacity and isbased on the pit resources outlined Table 16-13.

The financial evaluation incorporates the methodology of capitalization of waste within pre-productionyears as per industry standards. The capital expenditure is classified as additional civil works and resultsin a reduced strip ratio of 6.27 as the basis for the project cash flow.

RPM’s study indicates that the economic analysis show positive results with a net cash flow of USD 554.0million at an 8 percent discount rate, and with an internal rate of return of 26.3 percent. The followingsection is a detailed discussion on the cash flow and economic parameters for the project.

22.1 Project Evaluation

22.1.1 Estimated Product Price and Revenue

The market prices of vanadium product projected in the cash flow analysis are based on USD 14.04 perkilogram V2O5, USD 28.01 per kg FeV. A USD 70.00 per tonne of iron ore sold FOB mine site is includedas a secondary product.

Total annual revenue for the project is based on 1.4 Mt/y production to be reached in 2016 and continuingfor the life of the project averaging USD 192.0 million per year (gross revenue). The current planestimates 9.7 kt of V2O5 production and sales and 377 kt iron ore during 2015.

22.1.2 After-Tax Cash Flow

The Project’s annual net cash flows were modeled based on the projected revenue, operating costs and

capital expenditures summarized in Section 21 of this report. Table 22-1 details the project cash flow.

The royalty indicated is a contractual obligation to the state of Bahia calculated at 3 percent of the FOBsales revenue. The Brazilian CFEM (Mining Rights Tax) is also included based on 2 percent of the FOBsales revenue. The maximum Brazilian income tax rate of 34 percent on taxable income for miningproducers (25% plus 9% “Social Contribution”) has been assumed. The Sudene regional tax incentive iscalculated 25% of the gross profit subject to tax at a rate of 25%. All taxes and tax incentives have beenconsidered in the cash flow analysis. Potential reductions in income tax will be pursued during Projectimplementation.

22.1.3 Cash Flow Analysis

Based on the projections discussed previously, the Project is expected to generate a net cash flow ofUSD 25.0 million during the first full year of full production (2014); including USD 230.3 million in initialcapital expenditures during the preproduction years (2012-2013). Annual sustaining capital expenditure isestimated at USD 4.3 million per year from production Year 1 onwards.

Net cash flow totals USD 1,845.3 million over the 29 year designed mine life. Economic results of theProject cash flow model indicate an Internal Rate of Return (IRR) of 26.3 percent and a Net Present





Value of USD 554.0 million at an 8 percent discount rate. The 8 percent discount rate is consideredappropriate for this evaluation as the overall project risks are considered to be relatively low in terms oftotal capital committed, geological risk and market risk.




Table 22-1 Project Cash Flow

Project Period Units Value PP1 PP2 YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 YR10 YR11 YR12 YR13 YR14

Mining Physicals - - 647.85 6,881.22 9,493.08 9,553.20 9,430.78 8,933.66 8,945.42 9,332.56 9,094.36 7,592.76 5,858.53 4,151.78

Ore Mined to plant kt 40,585 628 850 570 1,326 1,413 1,413 1,413 1,413 1,411 1,413 1,473 2,082 1,644 1,254

Ore mined to stockpile kt 1,152 556 596 - - - - - - - - - - - - -

Waste Mined kt 282,020 1,700 3,249 2,601 3,385 4,238 4,238 2,846 2,087 4,238 8,476 11,858 12,654 20,000 18,366 15,097

Total Rock Mined kt 323,757 2,256 4,473 3,451 3,955 5,564 5,651 4,259 3,500 5,651 9,887 13,270 14,127 22,082 20,010 16,351

Stripping Ratio (Waste:Ore) 6.76 3.06 2.66 3.06 5.93 3.20 3.00 2.01 1.48 3.00 6.01 8.39 8.59 9.61 11.17 12.04

Ore from stockpile to plant kt 1,152 - - 223 842 86 - - - - - - - - - -

Total Material Movement kt 324,909 - 2,256 4,473 3,674 4,797 5,651 5,651 4,259 3,500 5,651 9,887 13,270 14,127 22,082 20,010 16,351

Total Plant Feed kt 41,737 - - 628 1,074 1,413 1,413 1,413 1,413 1,413 1,413 1,411 1,413 1,473 2,082 1,644 1,254


Conta ined Vanad iam tonnes 455,205 - - 7,311 12,854 17,190 18,258 19,376 19,499 19,249 18,234 18,258 19,048 18,931 17,948 13,122 8,629

Concentrate recovered kt 14,147 226 387 509 509 509 509 509 509 508 509 492 502 509 506

Grade of concentrate recovered %V2O5 2.98 3.06 3.11 3.30 3.51 3.53 3.48 3.30 3.31 3.45 3.47 2.84 2.16 1.54

Total V2O5 Produced Tonnes 331,064 5,509 9,685 12,952 13,757 14,599 14,692 14,504 13,739 13,757 14,353 13,986 11,677 9,010 6,385

Recovered V2O5 tonnes 135,633 5,511 9,689 12,309 6,881 5,112 5,144 5,078 4,810 4,817 5,025 4,897 4,088 3,155 2,236

Recovered V in FeV tonnes 103,523 - - 343 3,643 5,026 5,057 4,993 4,729 4,736 4,941 4,814 4,020 3,101 2,198

Recovered FeV: 80.00% tonnes 129,404 - - 429 4,554 6,282 6,322 6,241 5,912 5,920 6,176 6,018 5,024 3,877 2,747

Iron Ore ByProduct kt 14,011 - - 220 377 496 502 503 503 503 504 503 504 488 498 505 504

Revenue

Revenue V2O5 0 00 US $ 1 ,7 61 ,4 70 71,569 125,830 159,859 89,366 66,385 66,806 65,949 62,473 62,555 65,263 63,597 53,096 40,969 29,033

Revenue V in FeV 000 US$ 2,682,208 - - 8,886 94,383 130,208 131,032 129,353 122,535 122,696 128,006 124,739 104,143 80,356 56,946

Revenue of Iron Ore 000 US$ 980,794 15,430 26,377 34,738 35,118 35,241 35,239 35,244 35,262 35,210 35,247 34,129 34,866 35,378 35,263

Total Revenue 000 US$ 5,424,472 - - 86,999 152,207 203,482 218,867 231,834 233,077 230,546 220,270 220,461 228,516 222,465 192,105 156,703 121,242


Mining unit operating cost US$/t 4.29 - 4.23 4.29 4.34 4.28 2.01 2.15 2.58 2.77 2.08 1.63 1.52 1.55 1.33 1.50 1.70

Drilling US$/t 0.83 0.83 0.83 0.83 0.83

Blasting US$/t 0.73 0.77 0.70 0.74 0.69

Load and Haul US$/t 2.23 2.13 2.27 2.27 2.27

Development US$/t 0.50 0.50 0.50 0.50 0.50

55,000

Mining annual cost 000 US$ 596,558 9,545 19,193 15,939 20,537 11,365 12,151 10,975 9,681 11,750 16,083 20,144 21,863 29,434 30,035 27,716

Process to V2O5 0 00 US $ 1 ,4 11 ,0 13 31,372 42,765 50,168 50,885 51,649 51,721 51,575 50,981 50,964 51,457 51,008 52,404 48,412 44,432

Process V2O5 to FeV 000 US$ 245,350 - - - 813 8,634 11,911 11,986 11,832 11,209 11,223 11,709 11,410 9,526 7,350 5,209

Insurance on OPEX 000 US$ 28,861 1708 691 766 1,178 1,135 1,051 995 972 957 1,006 1,024 1,079 871 1,024 926 809

Total Operating Costs - FOB 000 US$ 2,281,783 1,708 10,236 51,331 59,882 72,653 71,936 76,705 75,654 74,046 74,945 79,295 84,390 85,152 92,389 86,724 78,167

Total Royalties 000 US$ 285,139 4,769 7,997 10,648 11,234 11,889 11,928 11,779 11,311 11,402 11,877 11,617 10,316 8,539 6,708

Tota l Ope rat ing C ost ( including R oy al ti e s) 0 00 U S$ 2 ,5 66 ,9 22 1,708 10,236 56,101 67,879 83,301 83,169 88,595 87,583 85,825 86,256 90,697 96,267 96,769 102,705 95,263 84,874

V 2O5 P or ti on 55, 338 66,576 76,704 36,234 26,069 25,688 25,128 25,505 27,055 28,828 29,131 31,881 30,059 27,203

Unit Operating Costs FeV portion - - 4,881 45,201 60,785 60,154 58,956 59,009 61,902 65,698 65,952 69,101 63,457 55,929

Cost per Processed Tonne US$ / t 61.50 89.39 63.22 58.97 58.87 62.71 62.00 60.75 61.06 64.30 68.15 65.68 49.33 57.96 67.69

Cost per kg (Recovered V2O5) US$ / kg 7.01 10.18 7.01 6.37 5.42 5.25 5.15 5.10 5.46 5.78 5.89 6.10 7.98 9.76 12.48

Cost per kg V in FeV US$ / kg 15.6 21.6 15.6 14.4 12.6 12.3 12.1 12.0 12.7 13.3 13.5 13.9 17.4 20.8 25.9

Total Capital Expenditure 000 US$ 355,752 170,752 59,594 5,082 38,133 21,000 11,587 1,057 35 - 3,561 2,554 3,018 539 8,245 3,946 269


Before Tax Cashflow (Annual) 000 US$ 2,501,798 172,460- 69,830- 25,816 46,195 99,181 124,110 142,183 145,460 144,722 130,453 127,211 129,231 125,158 81,156 57,494 36,098

Before Tax Cashflow (Cumulated) 000 US$ 172,460- 242,290- 216,474- 170,279- 71,097- 53,013 195,196 340,656 485,377 615,831 743,041 872,272 997,430 1,078,586 1,136,080 1,172,178

Total Depreciation 000 US$ 277,475 20,967 20,967 20,967 22,655 22,709 22,524 22,637 23,099 21,621 21,574 2,234 2,832 2,882 3,274

Total Taxes 000 US$ 670,865 - - 740 3,847 11,928 15,472 18,220 18,748 18,618 16,372 16,102 16,418 41,794 26,630 18,568 11,160


After Tax Cashflow (Annual) 000 US$ 1,845,338 172,460- 69,830- 25,077 42,348 87,254 108,638 123,963 126,712 126,104 114,082 111,108 112,813 83,364 54,526 38,926 24,938

Net CashFlow


Discounted Cashflow 000 US$ 554,008 172,460- 64,658- 21,499 33,617 64,134 73,937 78,118 73,935 68,130 57,069 51,465 48,384 33,105 20,049 13,253 7,861

Cumulative Discounted CashFlow 000 US$ 172,460- 237,118- 215,618- 182,001- 117,867- 43,930- 34,188 108,123 176,253 233,322 284,787 333,171 366,276 386,325 399,577 407,439



Periods to discounted payback Years 4.0 - 1 2 3 4 5




Table 22-1 Project Cash Flow (cont’d)

Project Period Units Value YR15 YR16 YR17 YR18 YR19 YR20 YR21 YR22 YR23 YR24 YR25 YR26 YR27 YR28 YR29 Total

Mining Physicals 6,798.26 5,290.77 5,280.13 6,501.14 6,398.21 6,217.00 6,074.98 6,381.40 6,504.17 6,540.72 7,137.55 8,195.18 9,372.43 10,334.89 8,610.17 195,552.20

Ore Mined to plant kt 40,585 1,779 1,441 1,437 1,683 1,657 1,654 1,526 1,555 1,473 1,413 1,413 1,413 1,413 1,413 1,017 40,585

Ore mined to stockpile kt 1,152 - - - - - - - - - - - - - - - 1,152

Waste Mined kt 282,020 20,000 20,000 22,291 20,293 18,991 16,081 12,572 8,814 7,067 6,244 5,208 4,168 3,301 1,568 391 282,020

Total Rock Mined kt 323,757 21,779 21,441 23,728 21,975 20,648 17,735 14,097 10,370 8,540 7,657 6,620 5,581 4,714 2,980 1,408 323,757

Stripping Ratio (Waste:Ore) 6.76 11.24 13.88 15.51 12.06 11.46 9.72 8.24 5.67 4.80 4.42 3.69 2.95 2.34 1.11 0.38 6.76

Ore from stockpile to plant kt 1,152 - - - - - - - - - - - - - - - 1,152

Total Material Movement kt 324,909 21,779 21,441 23,728 21,975 20,648 17,735 14,097 10,370 8,540 7,657 6,620 5,581 4,714 2,980 1,408 324,909

Total Plant Feed kt 41,737 1,779 1,441 1,437 1,683 1,657 1,654 1,526 1,555 1,473 1,413 1,413 1,413 1,413 1,413 1,017 41,737


C on ta in ed Van ad ia m ton ne s 4 55 ,2 05 16,138 11,895 11,851 15,054 14,743 14,231 13,332 13,936 13,674 13,350 14,568 16,727 19,130 21,094 17,574 455,205

Concentrate recovered kt 14,147 502 509 505 509 504 509 495 509 509 509 509 509 509 509 366 14,147

Grade of concentrate recovered %V2O5 2.54 1.95 1.97 2.40 2.38 2.30 2.30 2.36 2.40 2.42 2.64 3.03 3.46 3.82 4.42

Total V2O5 Produced Tonnes 331,064 10,455 8,137 8,120 9,998 9,840 9,561 9,343 9,814 10,003 10,059 10,977 12,603 14,414 15,894 13,242 331,064

Recovered V2O5 tonnes 135,633 3,661 2,849 2,843 3,501 3,445 3,348 3,271 3,436 3,502 3,522 3,843 4,413 5,047 5,565 4,636 135,633

Recovered V in FeV tonnes 103,523 3,599 2,801 2,795 3,442 3,387 3,291 3,216 3,378 3,443 3,463 3,779 4,338 4,962 5,471 4,558 103,523

Recovered FeV: 80.00% tonnes 129,404 4,499 3,501 3,494 4,302 4,234 4,114 4,020 4,223 4,304 4,328 4,723 5,423 6,202 6,839 5,698 129,404

Iron Ore ByProduct kt 14,011 498 506 502 505 501 505 492 505 505 505 505 504 504 503 361 14,011

Revenue

Revenue V2O5 0 00 US $ 1 ,7 61 ,4 70 47,540 36,998 36,924 45,462 44,743 43,475 42,482 44,625 45,484 45,739 49,913 57,309 65,541 72,272 60,211 1,761,470

Revenue V in FeV 000 US$ 2,682,208 93,245 72,569 72,423 89,170 87,758 85,273 83,325 87,528 89,212 89,713 97,899 112,406 128,553 141,754 118,098 2,682,208

Revenue of Iron Ore 000 US$ 980,794 34,882 35,400 35,118 35,354 35,054 35,365 34,438 35,359 35,354 35,353 35,330 35,290 35,246 35,210 25,298 980,794

Total Revenue 000 US$ 5,424,472 175,668 144,967 144,465 169,987 167,555 164,113 160,246 167,512 170,049 170,805 183,142 205,005 229,340 249,236 203,606 5,424,472


Mining unit operating cost US$/t 4.29 1.45 1.48 1.46 1.54 1.61 1.68 1.79 1.96 1.94 2.02 2.26 2.54 2.58 3.12 5.11

Drilling US$/t 0.83

Blasting US$/t 0.73

Load and Haul US$/t 2.23

Development US$/t 0.50

Mining annual cost 000 US$ 596,558 31,608 31,774 34,579 33,757 33,289 29,767 25,248 20,336 16,550 15,456 14,956 14,186 12,149 9,305 7,188 596,558

Process to V2O5 0 00 US $ 1 ,4 11 ,0 13 49,998 46,761 46,617 49,366 48,998 48,890 47,729 48,614 48,364 48,121 48,834 50,098 51,505 52,655 44,668 1,411,013

Process V2O5 to FeV 000 US$ 245,350 8,529 6,638 6,625 8,157 8,028 7,800 7,622 8,006 8,160 8,206 8,955 10,282 11,759 12,967 10,803 245,350

Insurance on OPEX 000 US$ 28,861 956 934 1,101 953 938 907 831 794 751 791 754 766 759 761 674 28,861

Total Operating Costs - FOB 000 US$ 2,281,783 91,091 86,107 88,922 92,233 91,252 87,364 81,429 77,751 73,825 72,574 73,499 75,332 76,172 75,687 63,332

Total Royalties 000 US$ 285,139 9,518 7,974 8,002 9,269 9,138 8,901 8,612 8,880 8,926 8,937 9,537 10,603 11,766 12,695 10,365

Tota l Opera ting Cost ( including Royal ties) 000 US$ 2 ,566 ,922 100,609 94,080 96,924 101,503 100,391 96,265 90,041 86,631 82,751 81,511 83,036 85,935 87,938 88,382 73,697

31,510 29,903 30,908 31,950 31,613 30,246 28,149 26,799 25,385 24,935 25,197 25,729 25,894 25,611 21,429

Unit Operating Costs 67,376 62,428 64,280 67,806 67,046 64,272 60,192 58,085 55,620 54,830 56,094 58,462 60,303 61,032 51,018

Cost per Processed Tonne US$ / t 61.50 56.55 65.31 67.45 60.33 60.59 58.20 59.02 55.70 56.20 57.70 58.78 60.83 62.25 62.56 72.48

Cost per kg (Recovered V2O5) US$ / kg 7.01 8.81 10.75 11.12 9.34 9.39 9.25 8.82 8.01 7.46 7.29 6.75 6.00 5.29 4.74 4.75

Cost per kg V in FeV US$ / kg 15.6 19.0 22.7 23.4 20.0 20.1 19.8 19.0 17.5 16.5 16.1 15.1 13.7 12.4 11.3 11.3

Total Capital Expenditure 000 US$ 355,752 2,057 4,681 4,401 1,035 851 788 - 551 - 5,368 1,595 1,635 - - 3,418


Before Tax Cashflow (Annual) 000 US$ 2,501,798 73,002 46,205 43,141 67,449 66,314 67,060 70,204 80,329 87,298 83,925 98,510 117,435 141,402 160,853 126,492

Before Tax Cashflow (Cumulated) 000 US$ 1,245,180 1,291,385 1,334,526 1,401,975 1,468,289 1,535,349 1,605,553 1,685,882 1,773,180 1,857,106 1,955,616 2,073,050 2,214,453 2,375,306 2,501,798

Total Depreciation 000 US$ 277,475 3,375 3,583 17,865 3,003 2,611 3,440 2,471 1,889 2,005 1,951 1,015 445 438 1,141 1,303

Total Taxes 000 US$ 670,865 23,673 14,491 8,594 21,912 21,659 21,631 23,029 26,670 29,000 27,871 33,149 39,777 47,928 54,302 42,564


After Tax Cashflow (Annual) 000 US$ 1,845,338 49,329 31,714 34,547 45,538 44,655 45,429 47,175 53,659 58,298 56,054 65,362 77,658 93,474 106,551 98,333

Net CashFlow


Discounted Cashflow 000 US$ 554,008 14,399 8,571 8,645 10,552 9,581 9,025 8,677 9,139 9,194 8,185 8,837 9,722 10,835 11,436 9,772

Cumulative Discounted CashFlow 000 US$ 421,838 430,409 439,054 449,606 459,186 468,211 476,888 486,028 495,221 503,406 512,243 521,965 532,800 544,236 554,008



Periods to discounted payback Years 4.0





22.1.4 Sensitivity Analysis

RPM developed a sensitivity analysis for the cash flow model based on variations in key project elementsof metal price, operating and capital costs. The sensitivity of the Project’s IRR, and NPV sensitivity to +/ -15 percent changes to key assumptions is shown in Table 22-2.


Spider charts are shown in Figure 22-1 and Figure 22-2 below for the Project’s sensitivity to price, capital

expenditure, production costs and foreign exchange rate (FOREX), with key assumptions varying plusand minus 5 percent. The element with the most impact on the project cash flow is metal price, followedby FOREX, production cost and capital expenditures (+/- 15 percent).

ItemIRR

(%)

NPV

(USD Million)

Base Case 26.3% 554.0

Capex +15% 23.3% 510.7

Capex -15% 30.1% 597.3

Sale Price (V205) +15% 29.2% 638.1

Sale Price (V205) -15% 23.4% 470.0

Sale Price (FeV) +15% 28.1% 646.9

Sale Price (FeV) -15% 24.4% 461.1

Sale Price (Fe) +15% 27.2% 591.0Sale Price (Fe) -15% 25.4% 517.0









Figure 22-1 Project Net Present Value

Figure 22-2 Project Internal Rate of Return





23 Adjacent Properties

There are no adjacent properties.





24 Other Relevant Data and Information

There is no other relevant data and information.





25 Interpretations and Conclusions

At the Maracás property in central Brazil, Largo has a vanadium deposit, Gulçari A, which is anadvanced-stage development project. Several programs of surface mapping and diamond drilling and a

significant amount of metallurgical and engineering studies have been completed there since the 1980's. A vanadium-titanium-PGM-mineralized mafic to ultra-mafic intrusion has been outlined at Gulçari A andrecent exploration has resulted in the discovery and initial delineation of other nearby zones of similarmineralization located at or near surface.

As a result of the recently completed exploration program the mineral resources at Maracás haveexpanded. The new updated mineral resource estimate is presented in Table 25-1.

Table 25-1 Largo Updated Mineral Resource Estimate

* - Resource within an open pit model using USD 34.20/t all in operating costs and reported at a 0.45% V 2 O5 cut-off,

reviewed and confirmed by B. Terrence Hennessey (Micon International Limited).

** - Resource within a pit shell using USD 34.20/t all in operating cost and reported at a 0.45% V 2 O5 cut-off, reviewed

and confirmed by Hebert Oliveira (Coffey Mining).

The Maracás deposit is to be mined as an open pit utilizing a combination of hydraulic excavators, largefront end loaders and 40 tonne capacity haulage trucks. The mine production schedule resulted in anaverage annual production rate of 1.4 Mt of ROM ore at a 1.10 percent grade (V2O5) ex-pit of blended HGand MG material and AMT of waste over the 29 year life of mine project.

The process flow sheet selected for the Maracás process plant comprises three states of crushing, orestage of grinding, two stages of magnetic separation, magnetic concentrate roasting, vanadium leaching,ammonium metavanadate (AMV), precipitation, AMV filtration and AMV calcining to V2O5 flakes as thefinal product. From Year 3, part (up to 65%) of the V2O5 produced will be smelted to produceferrovanadium (FeV).

The plant (originally sized to process 960,000 t/a ROM) will be capable, after due modification, to process1,400,000 t/a of feed ore with an average grade of 1.10% V2O5. The plant has an operating regime of365 d/a, 7 d/wk, 24 h/d and plant utilization of 89%, resulting in an average nominal hourly throughput of180 t/h. The plant will produce an equivalent of 11,400 t/y V2O5, (LOM average), and from Year 3 will

V2O5

(%)

Gulçari A* Measured 8,870,000 1.37 121,500Indicated 15,770,000 0.96 151,400

M &I 24,640,000 1.11 272,900

Inferred 2,610,000 0.76 19,800

Gulçari A Norte** Inferred 9,730,000 0.84 81,388

Gulçari B** Inferred 2,910,000 0.70 20,312

Novo Amparo** Inferred 1,560,000 0.72 11,255

Novo Amparo Norte** Inferred 9,720,000 0.87 84,453

Sao Jose** Inferred 3,900,000 0.89 34,706

Satellite Deposits (5)** Inferred 27,820,000 0.83 232,114

Deposits Category TonnesContained V2O5

(tonnes)





partly convert to production of 3,830 t/a (LOM average) of vanadium as ferrovanadium equivalent to 65%of total V2O5 from Year 5.

Recovery for V2O5 is estimated to be 72.5% while an overall average recovery of 68.4% V is expected forferrovanadium production over the life of mine.

The environmental baseline study was carried out by Brandt Meio Ambiente Ltda in June 2011 as part ofthe preparation of the EIA.

In addition, Mineral Engenharia e Meio Ambiente Ltda completed an environmental impact assessmentmitigation and compensation audit in December 2011. This audit followed the dictates of an EquatorPrinciples Compliance Audit and resulted in the preparation of an Action Plan. This plan incorporates theprograms necessary for compliance with the Brazilian laws and applicable environmental performancestandards and EHS guidelines.

The list of programs are presented in Table 25-2.





Table 25-2 Action Plan – Environmental Programs

Project Capital Costs, as of March 2013, are estimated to be USD 389.1 million including an allowancefor contingencies of USD 22.0 million, equivalent to 6% of total capital expenditure. The capital costsummary as presented in Table 25-3 are for actual expenditures through December 2012, USD 33.4million, the total pre-production capital through to December 2013 of USD 230.4 million and remaining

Program Description Phase

1) Environment, Health and Safety

Organization Plan

Plan to manage, apply and supervise the

Safety, Health and Environment plansConstruction and Operation

2) Fauna Management Plan Plan to manage and protect wild fauna. Construction and Operation

3) Forest Management Plan Plan to manage and protect forest. Construction and Operation

4) Aquatic Biota Management Plan Plan to manage and protect forest. Construction and Operation

5) Water Management Plan

Plan to manage industrial and domestic

waste waters and their treatment prior to

releases to the environment.


6) Surface Water Management Plan

Plan to manage the meteoric water and the

mitigation measures to storm events and

potential impact on environment.


7) Waste Management Plan

Plan to manage the solid waste, inert

industrial wastes and oily contaminated

soils at the project.


8) Environmental Monitoring Plan

Plan to monitor specific aspects throughout

the life of the mine. Establishment of an

“Observer Commission” consisting of localgovernment, ONGs, County leaders, which

regularly assess the implementation of the

project.


10) Plan of Social Communication

Plan to manage the public and internal

communication to assure transparency and

democratization of information.


11) Plan of Hiring Local PeoplePlan to hire 60% of people from

Municipality of MaracásConstruction

12) Plan of Labor Training

Plan to manage the training of skills and

competences of local people for project

opportunities


13) Plan of Guidance of Local SuppliersPlan to guide local suppliers on project

opportunities for new businessConstruction

Risk Management Plan- PGR

Environmental Risks Prevention Plan –

PPRA

Health Control Plan- PCMSO

15) Contingency Plan

Plan to identify e enforce actions in the

event of unforeseen events or an “upset”

condition and to simulate emergency

response.


16) Closure and Reclamation Plan

Plan to identify the concurrent and ultimate

reclamation and closure of the mine areaand any offsite impacts or disturbances.

Decommissioning

9) Plan Rehabilitation of Degraded

Areas

Plan to manage the rehabilitation and

reclamation activities of the mine siteConstruction and Operation

14) Health and Safety Plans Construction and Operation





other capital and sustaining capital costs of USD 125.4 million from 2014 to 2042. The total capitalexpenditure during the operation is USD 355.8 million, including acquisition to increase or replace minefleet equipment, plant and infrastructure.


The average unit cost for operational activities is USD 61.50/t of ore. The lifetime annual average of alloperating costs included from years 1 to 29 amounts to USD 88.1 million. The breakdown of mining,processing, general and administration costs, insurance and royalties as presented in Table 25-4.

The average unit production cost for V2O5 from Year 1 to 29 is USD 3.10 per pound. Given the plannedcompletion of construction and commissioning of the FeV plant for Year 3, the average production cost

from Year 4 to 29 for FeV production is USD 15.26 per kilogram.

Capital Expenditure Sunk Cost(2011-2012)

Remaining2012

2013 2014- 2042

Total


Enivronment 1,411 1,819 1,287 - 4,517

Management of Works 377 3,184 3,529 - 7,090



Civil Works 37 33,829 12,580 - 46,446

Equipment 2,012 47,959 5,633 - 55,604


Administration Costs 1,292 64 56 - 1,411 Consulting 10 64 56 - 129

Acquisit ion of Land 3,625 2,730 1,910 - 8,265

Acquisit ion of Mineral Rights 4,900 - - - 4,900


Seguros - 304 321 - 625

Contingency - 13,028 8,914 - 21,941





Closure Cost - - - 8,418 8,418

Total Capital Expenditures 33,365 170,752 59,594 125,405 389,117






The project will be subjected to the following royalty and taxes.

3 per cent of the FOB sales revenue, to the State of Bahia 2 per cent of the FOB sales revenue, to the Brazilian CFEM (Mining Rights Tax) Maximum income tax rate of 34 percent on taxable income for mining producers (25% plus 9%

“Social Contribution”) All taxes and tax incentives have been considered in the cash flow analysis. Potential reductions will be

pursued during project implementation.

Net cash flow totals USD 1,845.3 million over the 29 year mine life. Economic results of the cash flowmodel indicate an Internal Rate of Return of 26.3 per cent and a Net Present Value of USD 554.0 millionat an 8 per cent discount rate.

RPM have developed a sensitivity analysis for the cash flow mode based on variations in key projectelements of metal price, operating and capital costs. These have been presented in Table 25-5.



ROM

USD/Tonne

ROM

Mining 28.59 14.29

Plant 75.30 37.65

G&A 4.07 2.03Insurance 1.38 0.69

Royalty 13.66 6.83


ItemIRR

(%)

NPV

(USD Million)Base Case 26.3% 554.0

Capex +15% 23.3% 510.7

Capex -15% 30.1% 597.3

Sale Price (V205) +15% 29.2% 638.1

Sale Price (V205) -15% 23.4% 470.0

Sale Price (FeV) +15% 28.1% 646.9

Sale Price (FeV) -15% 24.4% 461.1

Sale Price (Fe) +15% 27.2% 591.0

Sale Price (Fe) -15% 25.4% 517.0









26 Recommendations

In the opinion of RungePincockMinarco, the 1.4 Million Tonnes per Year Processing Plant, hereindescribed as the expansion case, has merit. As the project continues to advance from the feasibility stageinto the design and construction phases there are areas of the project that should be given additionalconsideration beyond what is required for this level of study. Below is a summary of recommendations toconsider as the project advances.

26.1 Expansion and Definition Drilling Program and Budget

Largo completed a recommended diamond drilling program designed to expand the mineral resources onits Maracás property. This was a follow up of favorable results of an earlier exploration program ofground geophysics (magnetic survey), detailed geological mapping and sampling. Part of this programfocused on investigating the area between the Gulçari A deposit and the other five known prospects(satellite deposits); Gulçari A Norte, Gulçari B, São José, Novo Amparo, and Novo Amparo Norte to thenorth along the 8 km strike length of the Rio Jacaré intrusion.

This area became the focus of the drill program to explore for additional mineralization along strike fromthe known resources at Gulçari A. The plan was to do enough drilling on the satellite deposits in order tobring them into the inferred mineral resource category at the time. A program comprised of 13,400 m ofdrilling in 72 holes has been completed and the initial modeling of the new satellite deposits has beencarried out as is described in this report.

It is recommended that the Inferred resources identified in the new satellite deposits be upgraded to theMeasured and Indicated category through an additional definition drilling program. A suitable programwould comprise 21,200 m of drilling in 194 holes and is summarized in Table 26-1 below.

Micon and Coffey Mining have reviewed this proposed resource upgrade drilling program and budget

and, in light of the observations made during their site visit and noted in this report, find it reasonable andwarranted. It is recommended that Largo proceed with the program.





Table 26-1 Proposed Resource Upgrade Drilling Budget

26.2 Mineral Processing and Metallurgical Testing

The following recommendations for the Mineral Processing and Metallurgical Testing should beconsidered for this expanded case and include the ore from the satellite deposits:

The amount of water required to wash the leach residue to achieve the stated PLS concentrationand minimize vanadium losses should be confirmed;

The ability to wash the AMV cake minimizing re-dissolution of vanadium should be confirmed; The new fuel requirements for roasting the concentrate and heat losses should be confirmed by

modeling with vendor’s proprietary design using the new increased feed and production data; The calciner energy requirements and off-gas composition should be quantified by modeling with

vendor’s proprietary design and specific gas emission testing; The calciner off-gas from the calcining tests should be analyzed to provide a basis for treatment

requirements; Identification of corrosive gases from the calcination process in the testwork to date suggests future

testwork is required to quantify the content and volume of gas flows for materials selection incalcining equipment and gas-handling ductwork;

Critical equipment capacities and size should be confirmed with vendors in view of the newproduction scheme, keeping a contingency factor as per industry practice;

Mass and water balance should be re-run to verify usage and design/operation criteria used for thiseconomic evaluation.

Unit Cost Cost

(US$) (US$)

Contract Diamond Dri ll ing 21,200 m

(including mob/demob, camp

costs and related items)NQ-size drilling

194 holes

Certified reference material 4,000

Assay costs: 9,733 samples 60.00/sample 584,000

Freight 80,000

Field Expenses 140,000398,000

125,000

115,000

90,000

4,936,000

Vehicle Rental

Meals

Total Drilling Budget

Program Activity Amount

Proposed Expansion Drilling Budget Diamond

Drilling Program (Step-out drilling from Gulçari A)

160.38/m 3,400,000

Staff

Office Expenses





In addition, the metallurgical parameters through specific pilot-scale tests should be confirmed for theexpansion case:

The required wash water on the leach residue to achieve the stated vanadium losses and thevanadium concentration in the PLS should be confirmed;

Confirmation that the target vanadium concentration in the PLS is still economically achievable withlower grade concentrates;

The vanadium losses through dissolution during washing of the AMV filter cake should beconfirmed;

The compliance of calcine-leach vanadium recovery with the values stated in this report and usedas basis for the economic analysis should be confirmed;

Gas volume and composition during roasting should be verified; Available technology and associated costs for producing ferrovanadium should be confirmed.

26.3 Open Pit Mining

The following recommendations for the Mine Operation should be considered:

Geotechnical investigation to refine pit slope stability, including the satellite deposits, should beintroduced as mining progresses;

During Year 1 a drill and blast study should be introduced to optimize drilling and blastingparameters for all rock types;

Due to the strong demand for mining equipment it is recommended that Largo advance to thepurchasing phase of securing mining equipment to avoid an impact to the project schedule due topotential long lead times;

Further study to determine the viability of expanding the mineable ore resource through undergroundmining methods should be carried out;

Further refinement of the mine plan to smooth the material handling requirements, hence, the

equipment fleet size requirements over the life of the project should be carried out; The waste placement strategy should be further studied in order to minimize waste haulage distance

and cost; The economic impact of the use of remote crushing facilities for ore transportation from the satellite

pits to the main plant site should be analyzed; Additional topographic data in the vicinities of the satellite pits and their associated development

areas should be acquired to improve on the final designs of open pits, waste piles, drainage systemsetc.

26.4 Hydrology

The hydrological characteristics and water balance model should be reconciled with seasonal rainfall and

validated by means of a stochastic model within the first two years of operation. The estimated cost tocomplete this work is USD 122,000.





26.5 Environmental Studies

The following recommendations for the Environmental Studies should be considered:

All tailings storage facilities have been designed for the original 15-year mine plan outlined in theDecember Report. It is recommended that all storage facilities including the Leached CalcineTailings Dump, the Chloride Tailings Pond and the Non-Magnetic Tailings Dump be redesigned forthe storage requirements of this expansion case, including their characteristics, placement anddevelopment schedule;

Testing should be carried out on the waste material of the satellite deposits for acid generatingproperties, and the waste placement strategy should be further refined in order to minimize the riskof any potential acid rock drainage;

The existing water management plan should be further developed through the use of dykes andpossibly ditches for Gulçari A and all satellite deposits in order to mitigate against risk of pit floodingand acid rock drainage;

Condemnation drilling for satellite pits should to be completed; Baseline studies for each pit and the overall site to identify the impact of the satellite pit footprint to

develop the EIA for the future mining activities.





27 References

Aurox Resources Limited, 2006; Press Release, January 11, 2006, Balla Balla - Largest Vanadium OreReserve In Australia, 4 p.

Brandt Meio Ambiente Ltda; Environment Study, Project Vanádio de Maracás, Largo Mineraçáo Ltda,Maio 2008.

Brito, R. S. C., 1984; Geologia do Sill Estratificado do Rio Jacaré-Maracás, Bahia: Congresso Brasileirode Geologia, Sociedade Brasileira Geociências, 9th, Rio de Janeiro, Anais do 33, p. 4316 –4334.

Brito, R. S. C., 2000; Geologia e Petrologia do Sill Máfico Ultramáfico do Rio Jacaré, Bahia Estudio dasMineralizacoes de Fe-Ti-V e Platinoides Associados: Unpublished Ph.D. thesis, Brasilia, Brazil,Universidade do Brasilia, no. 39 325 p.

Brito, R. S. C., et al, 1981; Projeto Fazenda Gulçari, Companhia Baiana de Pesquisa Mineral (CBPM),Salvador, Convenio SME-CBPM, 102 p.

CPM Group; Vanadium Industry Outlook, Largo Resources Ltd., July 2008.

ECM S.A. Projetos Industrias; Processing and Infrastructuré Study, Projeto Vanádio de Maracás, LargoResources Ltd., July 2008.

Engenharia e Consultoria Mineral S.A., “Projeto Vanádio de Maracás Projecto Conceitual e Estimativa deInvestimento, Produção: 4,500 t/a de V2O5”, September 1990.

Galvão, C. F., 1984; Projeto Fazenda Gulçari – Alvo A, National Department of Mineral Production

(DNPM), Salvador, v. 1, 102 p.

Galvão, C. F., and Nonato, I. F. B. P., 1984; Projeto Fazenda Gulçari III, Companhia Baiana de PesquisaMineral (CBPM), Salvador, Convenio SME-CBPM, 51 p.

Galvão, C. F., Vianna, I. A., Nonato, I. F. B. P., and Brito, R. S. C., 1986; Depósito de MagnetitaVanadífera de Fazenda Gulçari, Maracás, Bahia, in Schobbenhaus, C. and Coelho, C.E.S. eds.,Principais Depósitos Minerais do Brasil: National Department of Mineral Production (DNPM), Brasília, v. 2(XL), p. 493 –501.

Hennessey, B. T., 2006; A Review of, and A Mineral Resource Estimate For, The Maracás VanadiumProject, Municipality Of Maracás, Bahia State, Brazil. An NI 43-101 Technical Report prepared for Largo

Resources Ltd. by Micon International Limited, filed on SEDAR, www.sedar.com.

Hennessey, B. T., 2007; An Updated Mineral Resource Estimate for The Maracás Vanadium Project,Municipality Of Maracás, Bahia State, Brazil. An NI 43-101 Technical Report prepared for LargoResources Ltd. by Micon International Limited, filed on SEDAR, www.sedar.com.



Hennessey, B. T. et al, 2008; An Updated Preliminary Assessment of the Maracás Vanadium Project,Municipality of Maracás, Bahia State, Brazil. An NI 43-101 Technical Report prepared for LargoResources Ltd. by Micon International Limited, filed on SEDAR, www.sedar.com.

IMS Pr ocess Plant, “Vanádio de Maracás Ltda., Vanadium Pentoxide Production Plant”, 1990.

Jacobs, C. A. et al., 2007; A Preliminary Assessment of the Maracás Vanadium Project, Municipality ofMaracás, Bahia State, Brazil. An NI 43-101 Technical Report prepared for Largo Resources Ltd. byMicon International Limited, filed on SEDAR, www.sedar.com.

Lopes, J. J. V., 1991; Composicão Mineralogica e Tenor de Oxidos Estimados: Gulçari A: Furos deSondagem FGA01 - FGA55, Unpublished internal report, Vanádio de Maracás Ltda, Salvador, 85 p.

Lurgi Staff, 1986; Feasibility Study: Maracás Vanadium Project, Lurgi Unpublished internal report, LurgiGMBH, Frankfurt, Germany, 18 p.

Marques, R., 2006; Draft Projecto Vanádio de Maracás, Estudo de Impacto Ambiental.

Menezes, J., 2005; Feasibility Study Summary, Unpublished internal report, Vanádio de Maracás Ltda,Salvador, 56 p.

NCL Brasil Ltda; Mine Feasibility Study, Executive Summary, Maracás Vanadium Project, Bahia, Brazil,for Largo Resources Ltd., August 2008.

PMA, Precious Metals Australia Limited, 2005; Press Release, December 20, 2005, JORC Compliant OreReserve at Windimurra - Largest in the World, 5 p.

Poylio, E., and Hiltunen, A., 1987; DDT - Test and Mineralogical Characterizations of Maracás Samples AA - 29 -90, Unpublished Research Report, Rautaruukki Oy Tutkimuskeskus Research Centre, Finland, 5p.

Rautaruukki Oy Tutkimuskeskus Research Centre, “Laboratory Research of the Suitability of the Otankäki

Process for Extracting Vanadium from Maracás Ore”, December 1989.

Roskill Information Services Ltd., 2010, Vanadium Global Industry Markets and Outlook, Twelfth Edition.

Sá, J. H. S., 1985; Pesquisa de Platinoides no Complexo do Rio Jacaré: Internal report, Bahia, Brazil,Companhia Baiana de Pesquisa Mineral, Salvador, 9 p.

Sá, J. H. S., 1992; Feições Petroquímicas Relacionadas a Estrutura do Corpo Gulçari A no ComplexoMáfico - Ultramáfico do Rio Jacaré, Maracás, Bahia [abs.]: 37o Congresso Brasileiro do Geologia, São

preliminary economic assessment of the maracás vanadium project, 1.4 million tonnes per year...

Documents