dissertation sansha nel final sn

Bibliography

Catalytic steam gasification of large coal particles 99

Bibliography

ALLEN, T. 1997. Particle Size Measurement: Volume 2. 5th ed. Taylor and Francis Group.

pp.528.

ALLEN, T. and A. A. KHAN. 1970. Critical evaluation of powder sampling procedures. The

Chemical Engineer. 238, pp.108-112.

BEND, S. L. 1992. The origin, formation and petrographic composition of coal. Fuel. 71,

pp.851-870.

CHAPMAN, S. K. 1986. Working with a Scanning Electron Microscope. Kent: Lodgemark

Press Ltd. pp.113.

CLOKE, M. and E. LESTER. 1994. Characterization of coals for combustion using

petrographic analysis: a review. Fuel. 73(3), pp.315-319.

COLLOT, A.-G. 2006. Matching gasification technologies to coal properties. Coal Geology.

65, pp.191-212.

DAWES, S. B. and P. MAZUMBER. 2006. Method for preparing catalysts. US: 7,097,880.

DOMAZETIS, G., J. LIESEGANG, and B. D. JAMES. 2005. Studies of inorganics added to

low-rank coals for catalytic gasification. Fuel Processing Technology. 86, pp.463-486.

ENGLAND, T., P. E. HAND, and D. C. MICHAEL et al. 2002. Coal Preparation in South

Africa. Pietermaritzburg: Natal Witness Commercial Printers. pp.298.

EVERSON, R. C., H. W. J. P. NEOMAGUS, H. KASAINI, and D. NJAPHA. 2006. Reaction

kinetics of pulverized coal-chars derived from inertinite-rich coal discards: Gasification with

carbon dioxide and steam. Fuel. 85, pp.1076-1082.

FALCON, R. 1986. A brief review of the origin, formation, and distribution of coal in Southern

Africa. Mineral Deposits of Southern Africa. 2, pp.1879-1898.

FALCON, R. M. S. and C. P. SNYMAN. 1986. An Introduction to Coal Petrography: Atlas of

Petrographic Constituents in the Bituminous Coals of Southern Africa. Johannesburg: The

Geological Society of South Africa. pp.27.

FRAISSARD, J. P. 1997. Physical adsorption: experiments, theory, and applications.

Dordrecht: Kluwer Academic Publishers. pp.619.

FRERIKS, I. L. C., H. M. H. VAN WECHEM, J. C. M. STUIVER, and R. BOUWMAN. 1981.

Potassium-catalysed gasification of carbon with steam: a temperature-programmed

desorption and Fourier Transform infrared study. Fuel. 60, pp.463-470.

Bibliography


GALLAGHER JR., J. E. and C. A. EUKER JR. 1979. Catalytic coal gasification for SNG

manufacture. In: 6th Annual International Conference on Coal Gasification, Liquefaction and

Conversion to Electricity, 1979. Pittsburgh.

GASIFICATION TECHNOLOGIES COUNCIL. 2008. Gasification: Redifining Clean Energy.

Gasification. 1(1), pp.1-25.

GREGG, S. J. and K. S. W. SING. 1982. Adsorption, Surface Area and Porosity. 2nd ed.

London: Academic Press Inc. pp.303.

GUZMAN, G. L. and E. E. WOLF. 1982. Kinetics of the K2CO3-Catalyzed Steam Gasification

of Carbon and Coal. Ind. Eng. Chem. Process Des. Dev. 21, pp.25-29.

HABASHI, F. 1970. Principles of Extractive Metallurgy: Hydrometallurgy. New York: Gordon

and Breach. pp.457.

HANAOKA, Y., E. BYAMBAJAV, and T. KIKUCHI et al. 2010. Steam Gasification of Low

Rank Coals with Ion-Exchanged Sodium Catalysts Prepared Using Natural Soda Ash.

(Paper presented at Twenty Seveth Annual International Pittsburgh Coal Conference on 14

October 2010.) Instanbul, Turkey. 9p. (Unpublished.).

HANSON, S., J. W. PATRICK, and A. WALKER. 2002. Effect of coal particle size on

pyrolysis and steam gasification. Fuel. 81, pp.531-537.

HARVEY, R. D. and R. R. RUCH. 1986. Mineral matter in Illinois and other U.S. coals. In:

Volume 301 of ACS symposium series, pp.10-40. Pennsylvania.

HATTINGH, B. B. 2009. The determination of the reaction mechanism involved in the CO2

gasification of inertinite-rich, high ash coal. Potchefstroom.

HAYASHI, J. and K. MIURA. 2004. Pyrolysis of Victorian Brown Coal. In: C. LI, (ed).

Advances in Science of Victorian Brown Coal, Oxford: Elsevier, pp.134-217.

HIGMAN, C. and M. VAN DER BURGT. 2008. Gasification. 2nd ed. Gulf Professional

Publishing. pp.435.

HIPPO, E. J. and D. TANDON. 1996. Low temperature steam-coal gasification catalysts.

Preprints Fuel Division., pp.216-220.

HUANG, J. and A. P. WATKINSON. 1996. Coal gasification in a stirred bed reactor. Fuel.

75, pp.1617-1624.

HUTTINGER, K. J. and C. NATTERMAN. 1994. Correlation between coal reactivity and

inorganic matter content for pressure gasification with steam and carbon dioxide. Fuel. 73,

pp.1682-1684.

Bibliography


ISHIDA, M. and C. Y. WEN. 1968. Comparison of kinetic and diffusional models for solid-gas

reactions. American Institute for Chemical Engineering Journal. 14, p.311.

JAFFRI, G. and J. ZHANG. 2008. Catalytic gasification characteristics of mixed black liquor

and calcium catalyst in mixing (air/steam) atmosphere. Journal of Fuel Chemistry and

Technology. 4(36), pp.406-414.

JEFFREY, L. S. 2005. Characterization of the coal resources of South Africa. The Journal of

The South African Institute of Mining and Metallurgy, pp.95-102.

JORDAAN, J. 1986. Highveld Coalfield. Mineral Deposits of Southern Africa. 2, pp.1985-

1994.

KARIMI, A., N. SEMAGINA, and M. R. GRAY. 2011. Kinetics of catalytic steam gasification

of bitumen coke. Fuel. 90, pp.1285-1291.

KRUGER, D. J. 2008. The preparation of activated carbon from South African coal for use in

PGM extraction. Potchefstroom.

LANG, R. J. 1986. Anion effects in alkali-catayzed steam gasification. Fuel. 65, pp.1324-

1330.

LANG, R. L. and R. C. NEAVEL. 1982. Behaviour of calcium as steam gasification catalyst.

Fuel. 61, pp.620-626.

LEE, W. J. and S. D. KIM. 1995. Catalytic activity of alkali and transition metal salt mixtures

for steam char gasification. Fuel. 74(9), pp.1387-1393.

LESSARD, R. R. and R. A. REITZ. 1981. Catalytic Coal Gasification: An emerging

technology for SNG. Texas.

LIU, Z. and H. ZHU. 1986. Steam gasification of coal char using alkali and alkaline-earth

metal catalysts. Fuel. 65, pp.1334-1338.

LU, G. Q. and D. D. DO. 1992. A kinetic study of coal reject-derived char activation with

CO2, H2O, and air. Carbon. 30, pp.21-29.

MAGEE, J. S. and G. E. DOLBEAR. 1998. Petroleum Catalysis in Nontechnical Language.

Tulsa: PennWell Publishing Company. pp.215.

MATSUKATA, M., E. KIKUCHI, and Y. MORITA. 1992. Migration of alkali and alkaline earth

metal elements into carbon black. Fuel. 71, pp.705-707.

MCCORMICK, R. L. and M. C. JHA. 1995. Effect of Catalyst Impregnation Conditions and

Coal Cleaning on Caking and Gasification of Illinois No. 6 Coal. Energy & Fuel. 9, pp.1043-

1050.

Bibliography


MCKEE, D. W. and D. CHATTERJI. 1978. The catalysed reaction of graphite with water

vapour. Carbon. 16, pp.53-57.

MERCK CHEMICALS. 2011. 104928 Potassium carbonate. [online]. [Accessed 15

September 2011]. Available form World Wide Web: <www.merck-chemicals.com>

MINING TECHNOLOGY. 1997. Twistdraai Coal Mine, South Africa. [online]. [Accessed 20

October 2010]. Available form World Wide Web: <http://www.mining-technology.com>

MOLINA, A. and F. MONDRAGON. 1998. Reactivity of coal gasification with steam and

CO2. Fuel. 77, pp.1831-1839.

MOULIJN, J. A., M. B. CERFONTAIN, and F. KAPTEIJN. 1984. Mechanism of the

potassium catalysed gasification of carbon in CO2. Fuel. 63, pp.1043-1047.

NISHIYAMA, Y. 1991. Catalytic gasification of coals-Features and possibilities. Fuel

Processing Technology. 29, pp.31-42.

OBERHOLZER, A. L. 2009. Characterisation and steam gasification of large, low grade coal

particles using a specially designed thermo gravimetric analyser. Potchefstroom.

O'NEILL, F. ([email protected]) 15 Aug. 2011. Discussion on error percentage on

XRF analyses. E-mail to: Nel, S. ([email protected]).

OSBORNE, D. G. 1988. Coal Preparation Technology. London: Graham & Trotman.

pp.1175.

PEREGO, C. and P. VILLA. 1997. Catalyst preparation methods. Catalysis Today. 34,

pp.281-305.

PETERSEN, E. E. 1957. Reaction of Porous Solids. AIChE. 3(4), pp.443-448.

PINHEIRO, H. J. 1999. A techno-economic and historical review of the South African coal

industry in the 19th and 20th centuries AND analyses of coal product samples of South

African collieries 1998-1999 (In Bulletin 13. SABS:Pretoria. 97p.).

RANA, M. S., E. M. CAPITAINE, C. LEYVA, and J. ANCHEYTA. 2007. Effect of catalyst

preparation and support composition on hydrodesulfurization of dibenzothiophene and Maya

crude oil. Fuel. 86, pp.1254-1262.

SCHMAL, M., J. L. F. MONTEIRO, and J. L. CASTELLAN. 1982. Kinetics of coal

gasification. Ind. Eng. Chem. Process. Des. Dev. 21, pp.256-266.

SCHOEMAN, L. ([email protected]) 6 Dec. 2010. Discussion of XRF procedure used by

UIS Analytical Services. E-mail to: Nel, S. ([email protected]).

Bibliography


SCHOPF, J. M. 1966. Definitions of Peat and Coal and of Graphite that Terminates the Coal

Series (Graphocite). The Journal of Geology. 74(5), pp.584-592.

SHADMAN, F. and D. A. SAMS. 1985. Interaction between Carbon and Alkali Metals during

Catalytic Carbon Gasification. In: Proc. 17th Biennial Conf. Am. Carbon Soc., 1985.

Lexington:, pp.182-183.

SHADMAN, F., D. A. SAMS, and W. A. PUNJAK. 1987. Significance of the reduction of

alkali carbonates in catalytic carbon gasification. Fuel. 66, pp.1658-1663.

SHARMA, A., T. TAKANOHASHI, and K. MORISHITA et al. 2008. Low temperature catalytic

steam gasification of HyperCoal to produce H2 and synthesis gas. Fuel. 87, pp.491-497.

SHETH, A., Y. D. YEBOAH, and A. GODAVARTY et al. 2003. Catalytic gasification of coal

using eutectic salts: reaction kinetics with binary and ternary eutectic catalysts. Fuel. 82,

pp.305-317.

SHUFEN, L. and S. RUIZHENG. 1992. Kinetic studies of lignite char pressurized gasification

of CO2, H2 and steam. Fuel. 73, p.413.

SMITH, D. and R. WHITTAKER. 1986. The Coalfields of Southern Africa: An Introduction.

Mineral Deposits of Southern Africa. 2, pp.1875-1879.

SOMMERFELD, D. A., J. JATURAPITPORNSAKUL, L. L. ANDERSON, and E. M. EYRING.

1993. Microscopic studies on the dispersion of iron/molebdenum bimetallic catalysts in

hydrotreated Blind Canyon, Wyodak and Pittsburgh #8 coals. Fuel Processing Technology.

34, pp.197-206.

STIEGEL, G. J. and R. C. MAXWELL. 2001. Gasification technologies: the path to clean,

affordable energy in the 21st century. Fuel Processing Technology. 71, pp.79-97.

SUN, Q., W. LI, H. CHEN, and B. LI. 2004. The CO2-gasification and kinetics of Shenmu

maceral chars with and without catalyst. Fuel. 83, pp.1787-1793.

SU, J. L. and D. D. PERLMUTTER. 1985. Porosity effects on Catalytic Char Oxidation Part I:

A Catalyst Deposition Model. AlChE Journal. 31(6), pp.957-964.

SUZUKI, T., M. MISHIMA, and J. KITAGUCHI et al. 1984. The catalytic steam gasification of

one Australian and three Japanese coals using potassium and sodium carbonates. Fuel

Processing Technology. 8, pp.205-212.

SZEKELY, J. 1977. Reactions between porous solids and gases. In: L. LAPIDUS and N. R.

AMUNDSON, (eds). Chemical Reactor Theory: A Review, New Jersey: Prentice-Hall, Inc.,

pp.269-313.

Bibliography


TAKARADA, T., Y. TAMAI, and A. TOMITA. 1986. Effectiveness of K2CO3 and Ni as

catalysts in steam gasification. Fuel. 65, pp.679-683.

TIEDT, L. R. 2011. Discussion regarding SEM photo's of raw and impregnated samples.

(Potchefstroom).

TIMPE, R. C., R. E. SEARS, and G. G. MONTGOMERY. 1987. Characterization of low-rank

coal char used in the production of hydrogen. University of North Dakota Energy Research

Center, Grand Forks. pp.1-9.

VAN DYK, J. C., M. J. KEYSER, and M. COERTZER. 2006. Syngas production from South

African coal sources using Sasol-Lurgi gasifiers. International Journal of Coal Geology. 65,

pp.243-253.

VAN NIEKERK, D., R. J. PUGMIRE, and M. S. SOLUM et al. 2008. Structural

characterization of vitrinite-rich and inertinite-rich Permian-aged South African bituminous

coals. International Journal of Coal Geology. 76, pp.290-300.

VERAA, M. J. and A. T. BELL. 1987. Effect of alkali metal catalysts on gasification of coal

char. Fuel. 57, pp.194-200.

WAGNER, N. J. and B. HLATSHWAYO. 2005. The occurence of potensially hazardous

trace elements in five Highveld coals, South Africa. International Journal of Coal Geology.

63, pp.228-246.

WANG, J., M. JIANG, Y. YAO, and J. CAO. 2009. Steam gasification of coal char catalyzed

by K2CO3 for enhanced production of hydrogen without formation of methane. Fuel (2009),

doi:10.1016/j.fuel.2008.12.017.

WANG, J., M. JIANG, and Y. YAO et al. 2009. Steam gasification of coal char catalyzed by

K2CO3 for enhanced production of hydrogen without formation of methane. Fuel. 88,

pp.1572-1579.

WANG, J., Y. YAO, J. CAO, and M. JIANG. 2010. Enhanced catalysis of K2CO3 for steam

gasification of coal char by using Ca(OH)2 in char preparation. Fuel. 89, pp.310-317.

WARD, C. R. 2002. Analysis and significance of mineral matter in coal seams. Coal

Geology. 50, pp.135-168.

WCI. 2009. The Coal Resource: A Comprehensive Overview of Coal. [online]. [Accessed 5

March 2010]. Available form World Wide Web: <www.worldcoal.org>

WCI. 2009. The Coal Resource: A Comprehensive Overview of Coal. London.

Bibliography


WU, S., J. GU, and L. LI et al. 2006. The reactivity and kinetics of Yanzhou coal chars from

elevated pyrolysis temperatures during gasification in steam at 900-1200 C. Process Safety

and Environmental Protection. 84, pp.420-428.

YE, D. P., J. B. AGNEW, and D. K. ZHANG. 1998. Gasification of a South African low-rank

coal with carbon dioxide and steam: kinetics and reactivity studies. Fuel. 77, pp.1209-1219.

YEBOAH, Y. D., Y. XU, and A. SHETH et al. 2003. Catalytic gasification of coal using

eutectic salts: identification of eutectics. Carbon. 41, pp.203-214.

YUH, S. Y. and E. E. WOLF. 1983. K2CO3-catalysed steam gasificationof supercritical

extracted chars. Fuel. 62, pp.738-741.

YUH, S. J. and E. E. WOLF. 1984. Kinetic and FT-i.r. studies of the sodium-catalysed steam

gasification of coal chars. Fuel. 63, pp.1604-1609.

ZARZYCKI, R. and A. CHACUK. 1993. Absorption: Fundamentals and applications.

Michigan: Pergamon Press. pp.638.

ZHANG, Y., S. HARA, S. KAJITANI, and M. ASHIZAWA. 2009. Modeling of catalytic

gasification kinetics of coal char and carbon. Fuel (2009), doi:10.1016/j.fuel.2009.06.004.

ZHU, W., W. SONG, and W. LIN. 2008. Catalytic gasification of char from co-pyrolysis of

coal and biomass. Fuel Processing Technology., pp.890-896.

Appendix A


Appendix A

Appendix A contains all relevant calculations regarding impregnation. This includes the

procedure and calibration for ISE measurements, calculations done to determine the catalyst

loading, and additional CT scans.

A.1. Procedure and calibration for ISE measurements

Procedure

The potassium ion (K+) concentration of the 0.5 M solution was calculated to be 39 098 ppm,

which is outside the concentration range specified for this potassium ion specific electrode.

Therefore, the solution was diluted to ensure accurate measurements. For each

concentration measurement, a 1 mL sample was taken from the original impregnation

solutions, containing the various particle size coal samples. The 1 mL impregnation solution

was diluted to 100 mL using deionised water, after which 2 mL potassium ion strength

adjuster (ISA) was added. For every 50 mL solution, 1 mL ISA is required. The ISA is

required for accurate ISE measurements, since it provides a constant background ionic

strength for samples and standards. The ISA allows accurate measurements of potassium

ion (K+) concentrations as low as 0.04 ppm. Glass beakers containing the diluted

impregnation solution sample were covered and put in a water bath, set at 23 °C. This was

done to ensure that the temperatures for the various diluted samples were uniform during

ISE measurements. Once the temperatures of the diluted samples stabilised, the ISE

measurements were taken with the potassium electrode. The diluted samples were stored

in the water bath and multiple measurements were taken randomly. Therefore, an average

catalyst loading was calculated for the ISE method, and the 95 % confidence interval was

determined.

Calibration

The potassium electrode was calibrated before each ISE measurement to ensure accurate

readings. During calibration of the ISE probe, it was found that the most accurate

measurements were taken when the probe setting was adjusted to measure millivolts (mV).

In order to determine the concentration of the K2CO3 impregnation solution, a calibration

graph had to be constructed. Three calibration solutions were prepared with K2CO3

concentrations of 0.4, 0.45 and 0.5 M. It was assumed that the decrease in solution

concentration would be in the range of 0.4 M to 0.5 M. The calibration curve is presented in

Figure A.1.

Appendix A


Figure A.1: ISE calibration curve

As can be seen from Figure A.1, the calibration curve is constructed by plotting the ISE

measurements (mV) against concentration. Numerous calibration measurements were

performed, and a 95 % confidence interval was calculated. The 95 % confidence interval is

illustrated by the error bars in Figure A.1. A trendline was fitted through the calibration data,

in order to obtain a calibration equation. The following calibration equation was obtained,

from which the concentration of K2CO3 in the impregnation solution was calculated:

(Equation A.1)

A.2. Calculation of catalyst loading from ISE results

The calibration curves presented in Appendix A.1, were used to determine the final

potassium concentration of the impregnation solutions. Once the final K2CO3 concentration

was determined, the difference in initial and final concentrations was determined:

[ ]2 3( ) 52.75 ( ( )) 93.71ISEmeasurement mV K CO M= × −

Appendix A


2 3 2 3 2 3[ ] [ ] [ ]decrease i fK CO K CO K CO= − (mol/L) (Equation A.2)

It is assumed that the amount with which the potassium concentration in the impregnation

solution decreases, is equal to the amount of potassium taken up by the coal samples.

Therefore, the potassium ion concentration is calculated from equation A.2:

2 3 2 3[ ] [ ]decrease adsorbedK CO K CO= (mol/L) (Equation A.3)

2 3[ ] 2 ([ ] )adsorbed adsorbed

K K CO+= × (mol/L) (Equation A.4)

From the concentration of the adsorbed potassium, [ ]adsorbed

K + , the weight (g) of potassium

adsorbed per weight unit (g) of coal can be calculated. Firstly, the [ ]adsorbed

K + is multiplied

by the total volume of the final impregnation solution to calculate the total moles of

potassium adsorbed:

[ ]adsorbed adsorbed solution

Total mole K K Volume+= + × (mol) (Equation A.5)

The weight of potassium can be calculated by multiplying the total mol adsorbedK +

with the

molecular weight of potassium:

(g) (Equation A.6)

In order to determine the weight percentage of potassium loaded onto the coal sample, on a

coal basis, adsorbed

Mass K is divided by the total mass of the coal sample:

.% 100adsorbedloaded

Mass KWt K

Mass coal sample= × (%) (Equation A.7)

A.3. Calculating K from K2O (XRF results)

The results obtained for the XRF analysis reported the potassium content in oxide form,

K2O. In order to determine the potassium (K) loading obtained through impregnation, the

elemental K had to be calculated from K2O, using the following calculation:

adsorbed adsorbed KMass K Total mole K MW+

= ×

Appendix A


(%) (Equation A.8)

XRF analysis of the raw, un-impregnated coal indicated a potassium (K) content of 0.44

wt.%. This value was subtracted from the XRF results of the impregnated coal samples, in

order to determine the amount of potassium added during impregnation.

The results presented on an ash basis are the results obtained from XRF analysis, as XRF

analysis is an ash analysis. The XRF results (ash basis) were used to calculate the catalyst

loading on a coal basis.

A.4. XRF results

Two different impregnated samples of each particle size was ashed and sent for XRF

analysis, for repeatability. An average of the results obtained for each set of samples was

used to calculate the potassium loading of the large coal particles. The repeatability values

obtained from XRF are presented in Table A.1. It should be noted that these values are the

original XRF results, and are presented on an ash basis.

Table A.1: XRF results in wt.% K2O

Particle size (µm) Sample A (wt.% K2O) Sample B (wt.% K2O) Average (wt.% K2O)

-212 µm 15.31 17.32 16.31

5 mm 7.78 8.08 7.93

10 mm 7.22 7.29 7.25

20 mm 8.08 6.22 7.15

30 mm 8.56 4.36 6.47

The average results shown in Table A.1 were used to calculate the potassium loading of the

coal particles (wt.%, on a coal basis).

A.5. Catalyst loading calculations

The potassium loading was calculated with the wt.% K, calculated from the wt.% K2O

results. In order to determine the amount of potassium added during impregnation, the initial

2

22 ( .%)( .%) K

K O

K O wt MWK wt

MW

× ×=

Appendix A


potassium content of the raw coal was subtracted from the total potassium content,

determined from the results in Table A.1.

(wt.% K2O/g ash) (Equation A.9)

The value for the initial potassium oxide (K2O) content of the raw coal is 0.53 wt.%, as

presented in Section 4.5.2. From the previously mentioned value, the potassium (K) content

is calculated to be 0.44 wt.%.

The potassium content, on a mass basis, can be calculated from loaded

K .

%( )( ) ( )

100 100

loadedmass

KAsh air dry basisK = × (g/g coal) (Equation A.10)

The total wt.% potassium added during impregnation, on a coal basis, can be calculated as

follows:

100%loaded massK K= × (wt.% K/g (Equation A.11)

A.6. Procedure and data processing for CT scans

Sample Preparation

The coal samples were placed in florist foam blocks to secure the sample position. Separate

CT scans were obtained for the different particle sizes, before and after impregnation.

Figure A.2 and Figure A.3 illustrate the positioning of the coal particles for CT scans.

2 2 2loaded total initialK O K O K O= −

Appendix A


Figure A.2: Positioning of 5 mm particles for CT scans

Figure A.3: Positioning of 20 mm particle for CT scans

As can be seen from Figure A.2 and Figure A.3, the coal particles are securely positioned in

florist foam to ensure accurate scans. The advantage of using florist foam is that the shape

of the specific coal particle is indented into the foam block. The indents of the particles are

used as guidelines to ensure that the same particle is positioned in the same manner for

scans taken before and after impregnation. The process of comparing CT scans obtained

before and after impregnation is simplified by maintaining a constant sample position. As

observed from the figures above, the sample size of the different particle sizes vary.

Accurate CT scans can be obtained for the 5 mm and 10 mm particle sizes, using 2 to 4

particles per sample. However, in order to ensure the accuracy of the CT scans for the 20

mm and 30 mm particles, only one particle was used per sample. CT scans were obtained

for all four particle sizes of the raw coal. After the scans were acquired the raw coal particles

were impregnated, as described in section 5.3.3.1. After impregnation, CT scans were

obtained for the impregnated coal particles.

Obtaining data for the 3D images involves 4 steps. During the first step, the coal sample is

rotated 360 degrees and high resolution digital radiographs are acquired every 0.5 degrees.

Appendix A


The second step involves geometric and shading correction. During data correction, spatial

and intensity irregularities are removed. Correction algorithms are also used to remove any

distortions and to correct for beam hardening. The third step is known as reconstruction,

where the individually corrected scans are combined using Cone Beam Back Projection to

obtain a CT image. The CT image is made up of millions of pixels known as voxels. The

final step is viewing the image results. X-Tek Graphical User Interface (XGUI) is used to

acquire, reconstruct and view the images.

Data processing

The CT scans obtained with the HMXST CT system, were processed with VGStudio Max 2.1

software. This software is used for visualisation and processing of voxel data, obtained from

CT scans. Although this software is very powerful and comprises of numerous functions,

only the volume rendering and volume analyser functions were used in this study.

The volume rendering tool is used to manipulate the volume object in different ways. This is

done by using the grey value histogram in the opacity manipulation area. The visibility of a

certain object in a CT scan is dependent on the opacity/intensity of the voxels and the grey

values of the elements which the object is comprised of. By using the volume rendering tool,

the opacity of certain grey value ranges can be varied to enhance or eliminate their

appearance. The interface of the volume rendering tool is illustrated in Figure A.4.

Figure A.4: Volume rendering tool interface

Appendix A


As seen from Figure A.4 , the histogram in the opacity manipulation area has two

distinguishable peaks. These peaks indicate that the CT scan has two discernable grey

value ranges (densities). The first peak represents the air around the object, in this case the

coal particle. The first step in processing the CT scans is to manipulate the opacity of the

voxels which are included in the grey value range of air so that they are not visible. This

removes the dullness and gives a brighter image. The second peak represents the second

largest amount of voxels which fall into a certain grey value range. In this case the second

peak symbolises voxels of the carbon element, since the majority of the coal particle is made

up of carbon. The fact that only two peaks are visible on the histogram, does not indicate

that the CT scan only comprises of air and carbon. This just implies that the amount of

voxels depicting the two peaks is significantly more than the voxels falling into other grey

value ranges. The first two peaks/intervals are disabled so that only the higher density

materials, i.e. the minerals, are visible on the CT scans.

It is possible to see voxels in all grey value ranges, by selecting a different colour for each

range. This is done by first dividing the histogram into various sections, known as intervals.

As seen from Figure A.4, the histogram is divided into three intervals. The first interval is

disabled, which means that the opacity of this grey value range is zero, and the air voxels

are not visible. The second interval represents the grey value range in which the carbon

element is obtained; while the third interval represents the grey value range in which all other

elements are obtained, such as minerals. In order to distinguish between the various

elements, a different colour can now be selected for each interval. As seen from Figure A.4,

the second interval is white (displays grey), while the third interval is red. The following CT

image was obtained when using the above-mentioned volume rendering settings:

Appendix A


Figure A.5: CT scan obtained from volume rendering analysis

As observed from Figure A.5, the carbon in the coal particles has a grey colour, while the

minerals in the coal particles are coloured red. The mineral bands in the coal particles are

clearly distinguishable from the carbon. This shows that the voxels of carbon and minerals

fall into different grey value ranges due to variation in intensity.

The second interval can also be disabled in order to eliminate the carbon from the image so

that only the minerals in the coal particles are visible, as shown in Figure A.6. A grey colour

was selected to enhance the visibility of the minerals.

Appendix A


Figure A.6: CT scan illustrating minerals in coal particles

As seen from Figure A.6, only the minerals are visible in the CT scan. During volume

rendering it is possible that the grey values of different voxels representing carbon and

minerals overlap, and that some of the carbon is still included in scans showing only the

minerals. The histogram is based on intensities of various elements, which cannot be

directly related to the density of the elements. Therefore, in order to find the grey value

range for a specific element, such as potassium, the image needs to be calibrated for water

(with a known density), as well as the densities of the desired elements.

The volume analyser tool was used to determine if the amount of potassium added to the

coal during impregnation, is substantial enough to increase the mineral volume of the coal

sample. The volume of the coal sample and the minerals in the coal sample are analysed

using the data histogram, as illustrated in Figure A.7.

Appendix A


Figure A.7: Volume analyser data histogram

The data histogram is used to analyse the properties of the voxel data within a selected

range. The range can either be specified by moving the red slider from left to right, or by

manually specifying the cursor range. The same grey value ranges used for the volume

rendering analysis are used for the volume analysis. Firstly, the volume of the entire coal

sample is determined by manually specifying the grey value range. This includes the carbon

elements in the coal sample as well as the mineralogical elements. The range between the

two red sliders is used to determine the volume, and the volume is given in mm3. Secondly,

only the volume of the minerals in the coal sample is determined. The mineral volume is

then used to calculate the volume percentage (vol.%) of minerals in the coal sample. This

analysis is done for all particle sizes, before and after impregnation. The vol.% of minerals

in the raw coal sample is compared to the vol.% of minerals in the impregnated coal sample,

to determine if impregnation increases the mineral volume.

A.7. CT scans

A CT scan of four 5 mm raw particles is shown in Figure A.8. Only the minerals present in

the particles are illustrated in the scans.

Appendix A


Figure A.8: CT scan of 5 mm raw coal particles

As seen from Figure A.8, the mineral distribution throughout the coal particles is irregular.

This clearly confirms the heterogeneity of coal, as well as the difference in composition

between two particles of the same size. The CT scan of the impregnated 5 mm coal

particles is presented in Figure A.9.

Figure A.9: CT scan of 5 mm impregnated coal particles

When the CT scan of the impregnated 5 mm particles (Figure A.9) is compared to the scan

of the raw coal particles (Figure A.8), a slight increase in minerals is observed as a result of

impregnation.

Appendix A


The following images are CT scans of the 10 mm particles, before and after impregnation.

Figure A.10: CT scan of 10 mm raw coal particles

Figure A.11: CT scan of 10 mm impregnated coal particles

As can be seen from the CT scans presented in Figure A.10 and Figure A.11, a slight

increase in mineral matter is observed for the particle on the left. However, no discernible

difference exists between the mineral matter of the raw particle and that of the impregnated

particle on the right.

The CT scans for the 10 mm and 20 mm coal particles are shown in this section.

Appendix A


The subsequent images are slice view images of the 10 mm particles, before and after

impregnation. Figure A.12 shows the clipping plane of the slice views for the 10 mm

particles.

Figure A.12: Clipping plane of 10 mm particles

The slice views of the 10 mm particles are presented in Figure A.13 and Figure A.14.

Figure A.13: Slice view of raw 10 mm particles

Appendix A


Figure A.14: Slice view of impregnated 10 mm particles

As seen from the slice views of the 10 mm particles, as shown in Figure A.13 and Figure

A.14, an increase in minerals due to impregnation is not apparent. These results are similar

to those found for the 5 mm particles presented in Section 5.4.4.2.

The following CT images presented are of the 20 mm coal particles, before and after

impregnation.

Figure A.15: CT scan of raw 20 mm coal particle

Appendix A


Figure A.16: CT scan of impregnated 20 mm coal particle

The CT scans of the 20 mm coal particles, as presented in Figure A.15 and Figure A.16,

clearly show that the coal particle is comprised of numerous mineral bands. There is no

discernible difference between the raw and impregnated 20 mm particles, as presented

above. The clipping plane for the slice views of the 20 mm particles is shown in Figure A.17:

Figure A.17: Clipping plane of 20 mm particles

The slice view of the raw and impregnated coal particles, as shown in Figure A.18 and

Figure A.19, also illustrates the mineral band included in the coal particle.

Appendix A


Figure A.18: Slice view of raw 20 mm coal particle

Figure A.19: Slice view of impregnated 20 mm coal particle

Appendix A


A.8. Volume predictions

The volume analyser tool was used to measure the particle and mineral volume in Section

5.4.4.2. The accuracy of the volume analyser was validated by calculating the radius of the

particles from the measured volume obtained from the volume analyser. The radius was

calculated with the equation used to determine the volume of a sphere, since coal particles

were selected to be spherically shaped:

(Equation A.12)

The volume prediction data is presented in Table A.1. The results presented in Table A.1

include the measured volumes obtained from the volume analyser, the radii calculated from

the measured volumes, and the actual radii of the particles.

Table A.2: Validation of volume analyser measurements

Particle size Measured volume

(mm3)

Calculated radius

(mm)

5 mm 98.56 2.87

10 mm 318.47 4.24

20 mm 5129 10.7

30 mm 13462.82 14.76

As seen from the results presented in Table A.2, the radii calculated from the measured

volume is a good prediction when compared to the intended radii of the particles. The

particles were hand selected and sieved to obtain an average particle radius measurement.

A.9. ISE error predictions

As seen from the results in section 5.4.3.2, the 95 % confidence interval estimates a

relatively large error. This can be attributed to the subtraction of two values, the initial and

final impregnation solution concentration, each having an error. A summary of the initial and

final concentrations, along with the respective errors calculated, is presented in Table A.3:

34( )3

sphereV r= ×Π ×

Appendix A


Table A.3: Initial and final impregnation solution concentrations

Particle size Initial concentration (M) Final concentration (M) Difference (M)

5 mm 0.504 ± 0.007 0.44 ± 0.005 0.06 ± 0.01

10 mm 0.504 ± 0.007 0.46 ± 0.006 0.04 ± 0.01

20 mm 0.504 ± 0.007 0.46 ± 0.007 0.04 ± 0.01

30 mm 0.504 ± 0.007 0.49 ± 0.005 0.01 ± 0.01

As seen from the results, the errors for the difference in initial and final concentration is

higher than for the individual values alone. This error increases since two values, with

respective errors, are subtracted from each other.

Appendix B


Appendix B

B.1. Conversion-time graphs

Figure B.1: Conversion-time graphs for 5 mm particles

Appendix B


Figure B.2: Conversion-time graphs for 10 mm particles

Appendix B


B.2. Experimental values for IM and VM and ash content

Table B.1: Experimental values for inherent moisture and volatile matter and ash contents

Temperature (°C) Raw

(IM and VM)*

Catalysed

(IM and VM)*

Raw

(Ash)

Catalysed

(Ash)

5 mm particles

800 0.29 0.25 0.16 0.17

0.29 0.29 0.14 0.09

825 0.30 0.27 0.14 0.14

0.32 0.27 0.13 0.13

850 0.30 0.27 0.14 0.13

0.30 0.25 0.15 0.17

875 0.29 0.30 0.14 0.12

0.29 0.26 0.13 0.15

10 mm particles

800 0.29 0.25 0.12 0.17

0.33 0.26 0.12 0.20

825 0.32 0.23 0.14 0.17

0.26 0.29 0.15 0.11

850 0.26 0.26 0.15 0.15

0.32 0.26 0.08 0.15

875 0.31 0.27 0.13 0.12

0.26 0.28 0.14 0.17

* (IM – Inherent moisture, VM – Volatile matter)

B.3. Experimental errors for average conversion runs

The errors for the average conversion plots were calculated using the following equation:

(Equation B.1)

( )

( )

=

−

−

=

∑N

1,i 2,i

i 1 1,i 2,i

X X

X X

2error %

N

Appendix B


Where X1,i and X2,i are the ith carbon conversion of run 1 and run 2, and N is the amount

of data point used for error determination.

Table B.2: Error % for average conversion runs

Sample 800 °C 825 °C 850 °C 875 °C

5 mm Raw 4 1 1 1

5 mm Cat 4 4 3 3

10 mm Raw 2 3 4 6

10 mm Cat 6 4 1 2

B.4. Particle size influence for 825 °C, 850 °C and 875 °C

Figure B.3: Influence of particle size on reactivity at 825 °C

Appendix B




Appendix B


B.5. Determination of reactivity, k

The slope from the –ln(1-X) vs. t plots were determined by fitting a linear line through the

data, as presented in Figure B.7 and Figure B.7.

Figure B.6: Determination of reactivity, k, from slope (5 mm)

Appendix B


Figure B.7: Determination of reactivity, k, from slope (10 mm)

B.6. Temperature profile during gasification

The following graph illustrates the temperature measurements acquired during a gasification

experiment conducted at 875 °C.

Appendix B


Figure B.8: Temperature profile during gasification experiment at 875 °C

As can be seen from Figure B.8, the temperature does not vary considerably during

gasification. A 95 % confidence interval of ± 1 °C was calculated from the experimental

logged data.

dissertation sansha nel final sn

Documents