Membrane Pilot Plant TrialsMembrane Pilot Plant TrialsAlburyAlbury Desalination Project Peer Review WorkshopDesalination Project Peer Review Workshop

2

Membrane Pilot Plant Trials

Lasse Blom, Liz TaylorAlbury Desalination Project Peer Review Workshop

20th November 2008

Future with WaterFuture with Water

3

Outline1. Pilot plant capability

– The 3-stage membrane pilot (the hardware)

– Simplified P&ID– Membrane element piloting

• Pilot performance• Hydraulic loads• Analytical results

2. Locations trial results– Paper Machine– Recycled Fibre plant– Tertiary Clarifier

3. Tertiary Clarifier– Membrane selections– Membrane comparison tables– Chemical usage

4. Proposed membrane arrangement– Final Configuration

– Pre-filtration– Membranes

– Optimum operation (cleaning frequency)

– Overall conclusions

4

The 3-stage membrane pilot

5

Pilot Plant TrialsIntroduction – Picture from test site

6

Pilot Plant Trials The 3-stage membrane pilot

▪ Pilot stages

Stage 3:

Stage 2:

Stage 1:PT

FF

P

MF/UF

P

PT

F

FF

P

NF/RO

VSD

F

P

PT

F

FF

P

NF/RO

VSD

F

7

Pilot Plant Trials The 3-stage membrane pilot

P

TP

F

FF

F F = Flow

= Pressure

= Temperature

P

T

▪ Simplified pilot P&ID of the three stages

Feed

Permeate

Retentate

VSD

F

P

Data logging:- Feed pressure- Loop temperature- Pump speed (VSD)- Permeate flow- Retentate flow- Operating modus

Acid/caustic

pHAnti-scalent

8

Membrane element piloting

9

Pilot Plant Trials Membrane element piloting

▪ An illustrating example of hydrodynamic simulation & analysis simulation vs a fullscale installation

1.

Permeate

Retentate2. 3. 4. 5. 6.

Feed

Retentate

Permeate

Retentate

Permeate

Worse location! Then a good conservative

location for testing the membranes!

Pilot

Fullscale simulation

10

Pilot Plant Trials Membrane element piloting - hydrodynamic

▪ The pilot crossflow profile – Customised design

Feed Retentate

600 L/h

350+600+50=1000 L/h

1st element:180/((1000+820)/2)=10%2nd element:120/((820+700)/2)=8%

This recovery is in theupper range of what is recommended (7%).4040 element has normallya crossflow between 450 to1450 L/h.

1000

50 L/h

820

700

~120 L/h~180 L/h

350 L/h

Membrane settings:TMP control, gives an averagetotal permeate flow of 300 L/h.VCF = 8

Estimate that the permeatingis slightly higher in the first element than the second.

11

Pilot Plant Trials Membrane element piloting - Energy

▪ Two calculation methods for energy consumption– Process to power grid - Theoretical calculations– VSD to power grid - Actual pump calculations

▪ These are mainly used for troubleshooting.

PQ

VSD

ηp

ηm

12

Pilot Plant TrialsMembrane element piloting - Analysis

▪ Analysis performed at the Albury lab– Total suspended solids (SS onto 0.45 um paper), only feed– Total dissolved inorganic solids (0.45 um paper, 180°C

evaporated)– Turbidity– Colour– UV254 (Lignin)– Chemical Oxygen Demand (on total)– Conductivity– pH– Total Alkalinity (HCO3

-)– Cations: Na+, Mg2

+, Ca2+, SiO2, Mn3

+, Fe3+

– Anions: Cl-, SO42-

13

Pilot Plant TrialsMembrane element piloting – Membranes

▪ Tested membranes & locations

nanaXPolyamide~30KOCH XR4KOCH XR4Spiral WoundBWRO

naXXPES/PVP200 kDCapfil - UFC M5X-Flow, NORITHollow FiberMF

Test locations

Membrane type: Manufacturer: Name: MWCO: Material: TC PM RCF

UF

Tubular polymeric X-Flow, NORIT Comp.-F5385 - ID 8mm 30 kD PVDF X X X

Tubular polymeric X-Flow, NORIT Comp.-F5385 - ID 5mm 30 kD PVDF X na na

Tubular ceramic Tami Industries InsideCeram 8 kD TiO2 X na na

NF

Spiral Wound Filmtec NF270-400 300 D Polyamide X X X

Spiral Wound Osmonics Duracon NF4040F 150-300 D Polyamide X X X

Spiral Wound Hydranautics Hydracore 50-4040 1000 D SPS X na na

Spiral Wound Hydranautics ESNA1-LF/LF2 200 D Polyamide X na na

Spiral Wound KOCH TFC-SR2 300-400 D na X na na

SWRO

Spiral Wound Hydranautics ESPA2 ~30 Polyamide X na na

Spiral Wound Hydranautics SWC1-4040 ~30 Polyamide X X X

14

Pilot Plant TrialsMembrane element piloting

▪ Initial start-up tests to find the optimum operating parameters– It is only the RO that has a linear flux as the pressure

increases, the other membrane vary with changing pressure and these test gives us an indication of where the operating parameters should be:

Temperature adjusted permeate flux with process water (PW)(Various NF membranes, feed water from UF Tubular 30 kD, circulation flow of 500 L/h)

0

20

40

60

80

100

120

0 250 500 750 1000 1250 1500

Pressure (kPa)

Tem

pera

ture

adj

uste

d pe

rmea

te fl

ux (l

mh@

25oC

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautic ESNA1

Process water flux at various Volume Concentration Factors (VCF-Test)(Feed water from UF Tubular 30 kD, circulation flow of 500 L/h)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0

Volume Concentration Factor - VCF

Perm

eabi

lity

flux

(lmh

@ 2

5°C

/bar

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautics ESNA1

Pressure test

Permeate Flux Test on Freshwater (FW) and Process Water (PW)(NF Osmonics membrane, feed water from UF Tubular 30 kD, VCF=6, circulation flow is 500 L/h)

0

20

40

60

80

100

120

0 250 500 750 1000 1250 1500 1750Pressure (kPa)

Perm

eate

flux

(lm

h)

Temp.adj. FW Flux

Temp.adj. PW Flux

Critical Flux

Optimum TMP

Concentration testCritical flux test

15

Locations trial results

16

Pilot Plant TrialsLocations trial results

▪ Based on initial analysis screening of 21 locations in the mill, three test locations was selected for piloting

RCF

TC

PM

Parameters RCF PM TC

Temp. (°C) 54 52 31

COD–GFC (mgO2/L) 2192 1850 434

Turbidity (NTU) 39 29 3

Cat.dem. (meq/L) 1652 881 662

Colour (Hazen) 229 85 185

Cond. (µS/cm) 2632 2001 2133

Sodium 580 318 348

pH 7.4 6.6 7.5

Total Solids (mg/L) 3594 2956 1900

TDIS (mg/L) 2623 1860 1583

Iron (mg/L) 0.44 0.82 0.42

Manganese (mg/L) 1.0 1.6 1.0

Calcium (mg/L) 44 96 70

CaCO3 (mg/L) 427 146 554

Magnesium (mg/L) 6.8 10.1 8.7

Chloride (mg/L) 98 30 8

Silica (mg/L) 150 50 84

Sulphate (mg/L) 277 327 259

17

Pilot Plant TrialsTest location 1 – Paper machine

▪ PM test conclusions– Three different white water qualities were tested

• Rich Clear-, Clear,- and Super Clear filtrate– The ceramic tubular membrane plugged quickly.– The HF operated well, but needed very good pre-filtration.– All the three water qualities gave high fouling of the UF

Tubular membrane but it operated the best out of all the options

• We learned that too high circulation flow (cross flow velocity) generated too much foam in the system.

– The VCF could at most be 4 for the NF stage.– Only two runs performed with the RO due to problems

with the UF stage which did not operate long enough.– Operation summary:

• UF Tubular (VCF=2, TMP Control) – 3 days• NF Osmonics (VCF=4, TMP Control) – 2 days• RO Hydranautics (VCF=5, TMP Control) – 2 days

18

Pilot Plant TrialsTest location 1 – Paper machine

▪ PM test conclusions - pictures

Too much foaming for all the 3-stages!

Plugged ceramic membrane!

Spin Klin plugged

Test site!

Filtrated water qualities!

Plugged UF inline filter!

Bow screen

19

Pilot Plant TrialsTest location 2 – Recycled Paper Furnish

▪ RCF test conclusions– Too high circulation flow generates too much foam in the

system and was not sustainable for the ceramic and tubular membrane.

– Backwashing on the UF Tubular membrane improved the operating time drastically, from only 1 day to 7 days.

– The MF HF operated for 10-13 days, with backwashing.– The NF operation is around 5 days when the upstream pre-

filtration is a UF. The NF was the bottle neck in this location.– The RO operated relatively well in this position (VCF=3).– Operation summary:

• UF Tubular (VCF=5, TMP Control) – 7 days• NF Osmonics (VCF=4, TMP Control) – 5 days• RO Hydranautics (VCF=3, TMP Control) – 8 days

20

Pilot Plant TrialsTest location 2 – Recycled Paper Furnish

▪ RCF test conclusions - pictures

Test site!

Some foaming issues!

Filtrated water qualities!

Spin Klin operated well!

The HF operated well!

21

Test location 3: Tertiary Clarifier

22

Pilot Plant TrialsTest location 3 – MF/UF stage

1st stage treatment - Normalised permeate flux - Ultrafiltration and Microfiltration(TMP Control, various pre-filtration)

0

20

40

60

80

100

120

140

160

180

200

10/08

/0811

/08/08

12/08

/0813

/08/08

14/08

/0815

/08/08

16/08

/0817

/08/08

18/08

/0819

/08/08

20/08

/0821

/08/08

22/08

/0823

/08/08

24/08

/0825

/08/08

26/08

/0827

/08/08

28/08

/0829

/08/08

30/08

/0831

/08/08

1/09/0

82/0

9/08

3/09/0

84/0

9/08

Time of operation - fictive dates

Nor

mal

ised

per

mea

te fl

ux (L

MH

@25

°C) 200 kD - Hollow Fiber Straw (250 kPa) - 3.6 m2

30 kD - Tubular polymeric (180 kPa) - 15.3 m2

8 kD - Tubular ceramic (300 kPa) - 1.5 m2

23

Pilot Plant TrialsTest location 3 – MF/UF stage

1767558590775.3306Tubular

(30 kD, 50% circ) (3

13445530.8250779.7240Hollow Fibre (200 kD)

Membrane:PermeateProduced

(l/h)

Specific energy:

(kWh/m3) (1

Average flux:(lmh

@25°C)

TMP:(kPa)

Specific Flux:

(lmh/bar)VCF:

Spec. invest. cost

(A$/m2)

Operating time:(days)

Tubular Ceramic (8 kD)

50 42.7 41 300 13.7 3 1123 2 (2

Tubular (30 kD,

84% circ) (4

530 3.5 38 180 21 5 675 22

1) Compare only as relative numbers, and not absolute.2) Could be operated longer, but due to low flow had to be cancelled because of time limits.3) One element (5.1 m2).4) Three elements in series (15.3 m2).

▪ Comparison of the Ultrafiltration membranes trialled

24

Pilot Plant TrialsTest location 3 – MF/UF stage

▪ Comparison of the MF/UF membranes trialledMF/UF Cleansing Efficiency Comparison

0

10

20

30

40

50

60

70

80

90

100

Turbidity Colour COD Fe Ca Mn Mg Na TDIS 180 SiO2

% C

lean

sing

Effi

cien

cy

Ceramic - 8 kD

Hollow Fibre - 200 kD - Feed diluted 15% - Backwash

Tubular - 30 kD

255.25

Pilot Plant TrialsTest location 3 – MF/UF conclusions

▪ The Tubular membrane was selected due to its operation duration of up to 14 days and the need for less pre-filtration

▪ The Hollow Fibre (Tube) operated for a similar time to the Tubular polymeric membrane, but it needs better pre-filtration and backwashing water and makes the arrangement more complex.

▪ The specific energy consumption is quite different. The Tubular membrane uses half of the energy used for the Hollow Fibre. The main difference is that the HF uses the energy for pressure.

▪ It should be noted that the Tubular membrane has a lower cut-off (30 kD) which is much denser than the Hollow Fibre membrane of 200 kD

▪ One negative issue with the Tubular membrane installation compared to the Hollow Fibre, is that it uses more space, but this will not be an issue for Albury.

26

Pilot Plant TrialsTest location 3 – Nanofiltration

Nanofiltration membranes - Temperature adjusted permeability(VCF=8, feed water UF Tubular 30 kD, cir.flow of 600 l/h)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

26/01

/0827

/01/08

28/01

/0829

/01/08

30/01

/0831

/01/08

1/02/0

82/0

2/08

3/02/0

84/0

2/08

5/02/0

86/0

2/08

7/02/0

88/0

2/08

9/02/0

810

/02/08

11/02

/0812

/02/08

13/02

/0814

/02/08

Operation time (days) - fictive dates

Nor

mal

ised

per

mea

bilit

y (lm

h@25

°C/b

ar) Filmtec 270 (300 kPa)

KOCH SR2 (300 kPa)

Osmonics DK (500 kPa)

Hydranautics ESNA1 (300 kPa)

27

Pilot Plant Trials Test location 3 – SAR optimisation

Minor improvement of the soilstructure!

Mg2+

Ca2+

Na+

NFMultivalent ions Monovalent ions

[ ][ ] [ ]

2

22 ++

+

+=

MgCaNaSAR

Comparing NF Membrane Removal Efficiencies

0

10

20

30

40

50

60

70

80

90

100

Sodium Calcium Magnesium

Rem

oval

Effi

cien

ciy

(%)

Filmtec 270

KOCH SR2

Osmonics DKHydranautics ESNA1

“Tailor-made-water”

28

Pilot Plant TrialsTest location 3 – Nanofiltration

Sodium - Removal EfficiencyTubular Ultrafiltration - Misc Nanofiltration - Hydranautics Reverse Osmosis

29

21

28

83

25

29

0

10

20

30

40

50

60

70

80

90

100

UF Permeate NF Permeate RO Permeate

Tubular VCF 5 Misc. Nanofiltration VCF 8 Reverse Osmosis VCF 5

Rem

oval

Effi

cien

cy (%

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautics ESNA1

Chemical Oxygen Demand - Removal EfficiencyTubular Ultrafiltration - Misc Nanofiltration - Hydranautics Reverse Osmosis

95

36

94

71

93

98

0

10

20

30

40

50

60

70

80

90

100

UF Permeate NF Permeate RO Permeate

Tubular VCF 5 Misc. Nanofiltration VCF 8 Reverse Osmosis VCF 5

Rem

oval

Effi

cien

cy (%

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautics ESNA1

Calcium - Removal EfficiencyTubular Ultrafiltration - Misc Nanofiltration - Hydranautics Reverse Osmosis

72

21

54

92

5653

0

10

20

30

40

50

60

70

80

90

100

UF Permeate NF Permeate RO Permeate

Tubular VCF 5 Misc. Nanofiltration VCF 8 Reverse Osmosis VCF 5

Rem

oval

Effi

cien

cy (%

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautics ESNA1

Total Dissolved Inorganic Solids - Removal EfficiencyTubular Ultrafiltration - Misc Nanofiltration - Hydranautics Reverse Osmosis

43

22

44

82

45

51

0

10

20

30

40

50

60

70

80

90

100

UF Permeate NF Permeate RO Permeate

Tubular VCF 5 Misc. Nanofiltration VCF 8 Reverse Osmosis VCF 5

Rem

oval

Effi

cien

cy (%

)

Filmtec 270

KOCH SR2

Osmonics DK

Hydranautics ESNA1

29

Pilot Plant TrialsTest location 3 – Nanofiltration

NF Filmtec 270 - Normalised permeability - No acid vs pH<6.8 vs pH<6.3 vs pH<4.5(TMP Control (300 kPa), VCF=8, feed water UF Tubular 30 kD, cir.flow of 550 L/h)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

27/09

/0829

/09/08

1/10/0

8

3/10/0

8

5/10/0

8

7/10/0

8

9/10/0

811

/10/08

13/10

/0815

/10/08

17/10

/0819

/10/08

21/10

/0823

/10/08

25/10

/0827

/10/08

29/10

/0831

/10/08

2/11/0

8

4/11/0

8

6/11/0

8

Operation time (days)

Nor

mal

ised

per

mea

bilit

y (lm

h/ba

r@25

°C)

No acid (300 kPa)

pH < 6.8 (300 kPa)

pH < 6.3 (300 kPa)

pH < 4.5 (300 - 500 kPa)

29 days16 days

>11 days

35 days

30

Pilot Plant TrialsTest location 3 – Nanofiltration

pH = ~8.0 – Two week of operation

pH < 6.8- Four weeks of operation- Similar to the Osmonicswithout any chemical dosing

▪ Comparison of the Filmtec 270 permeate tank with different pH dosing levels

pH < 4.5- At least 11 days

31

Pilot Plant TrialsTest location 3 – Nanofiltration

▪ Summary of the NF membranes tested (average of 10 first days of operation)

3514586.430019.11.6190Filmtec 270

300 D, pH<6.3

1614587.030020.81.6250Filmtec 270

300 D

2914587.630022.71.4220Filmtec 270

300 D, pH<6.8

134586.730020.01.5152Hydra. ESNSA1LF/LF2 200 D

Membrane:PermeateProduced

(l/h)

Specific energy:

(kWh/m3)

Average flux:

(lmh@25°C)

TMP:(kPa)

Specific Flux:

(lmh@25°C/bar)

VCF:

Spec. invest. cost:

(A$/m2)

Operating time:(days)

KOCH SR2300-400 D

235 1.6 21.0 300 7.0 8 55 14

Osmonics DK150-300 D

185 2.0 21.0 500 4.2 8 125 14

Filmtec 270300 D, pH<4.5

185 2.5 16.3 450 3.6 8 145 >11

32

Pilot Plant TrialsTest location 3 – Nanofiltration conclusions

▪ In general, most of the NF membranes trialled are quite similar in respect to concentration levels, while the flow differs

▪ When selecting membranes based on best permeate quality, the Osmonics DK membrane performed the best

▪ When selecting based on best permeate flux performance, the Filmtec 270 membranes performed best

▪ The ESNA1 had a relative high flux and a good removal efficiency, so it is placed between the Osmonics DK and Filmtec 270

▪ The KOCH SR2 membrane looked promising, but it looks as if the performance deteriorated during the trials due to membrane degradation

▪ We selected the Filmtec 270 for pre-filtration before the RO stage during our longevity tests.

33

Pilot Plant TrialsTest location 3 – Reverse Osmosis - Flux

SWRO (Hydranautics SWC1) vs BWRO (Hydranautics ESPA2) - Normalised pressure(Flux Control, VCF=5, feed water from NF Filmtec 270, Circ.flow of 400 l/h,

no H2SO4 vs pH<6.8 vs pH<6.3 vs pH<4.5, legend figures are 90% percentile)

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

12/09/08 13/09/08 14/09/08 15/09/08 16/09/08 17/09/08 18/09/08 19/09/08 20/09/08 21/09/08 22/09/08 23/09/08 24/09/08 25/09/08 26/09/08 27/09/08 28/09/08

Operation days

Nor

mal

ised

pre

ssur

e (k

Pa @

25°C

)

SWRO - no acid (2460 kPa @25°C)

BWRO - no acid (1660 kPa @25°C)

BWRO - pH<6.8 (790 kPa @25°C)

BWRO - pH<6.3 (2900 kPa @25°C)

BWRO - pH<4.5 (340 kPa @25°C)

6 days 7 days

>11 days

4 days 15 days

34

Pilot Plant TrialsTest location 3 – Reverse Osmosis - Flux

SWRO (Hydranautics SWC1) vs BWRO (Hydranautics ESPA2) - Normalised permeate flux(Flux Control, VCF=5, feed water from NF Filmtec 270, Circ.flow of 400 l/h,

no H2SO4 vs pH<6.8 vs pH<6.3 vs pH<4.5, legend figures are 90% percentile)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

12/09/08 13/09/08 14/09/08 15/09/08 16/09/08 17/09/08 18/09/08 19/09/08 20/09/08 21/09/08 22/09/08 23/09/08 24/09/08 25/09/08 26/09/08 27/09/08 28/09/08

Operation days

Nor

mal

ised

per

mea

te fl

ux (l

mh

@25

°C)

SWRO - no acid (8.6 LMH @25°C)

BWRO - no acid (7.8 LMH @25°C)

BWRO - pH<6.8 (7.3 LMH @25°C)

BWRO - pH<6.3 (5.8 LMH @25°C)

BWRO - pH<4.5 (6.4 LMH @25°C)

6 days

7 days

>11 days

4 days

15 days

35

Pilot Plant TrialsTest location 3 – Reverse Osmosis

▪ Comparison of the Reverse Osmosis membranes trialled

152751.24855.67.4100BWRO

HydranaticsESPA2, pH<6.3

6.02750.95356.63.7100BWRO

HydranaticsESPA2

7.02751.45006.83.7100BWRO

HydranaticsESPA2, pH<6.8

5.04250.96205.35.2120BWROKOCHXR4

Membrane:PermeateProduced

(l/h)

Specific energy:

(kWh/m3)

Average flux:

(lmh@25°C)

TMP:(kPa)

Permeability:(lmh/bar @25°C)

VCF:Spec. invest.

cost:(A$/m2)

Operating time:(days)

SWROHydranautics

SWC1120 6.6 6.9 1460 0.5 5 78 4.0

BWROHydranatics

ESPA2, pH<4.5

100 6.8 6.0 310 2.0 5 27 >11

36

Pilot Plant TrialsTest location 3 – Reverse Osmosis conclusions

▪ Using BWRO versus SWRO reduced the energy consumption by 25% at the same flux.

▪ The BWRO removal efficiency was slightly better than the SWRO.

▪ The KOCH XR4 membrane degraded during our tests.▪ There does not seem to be much variation between the

RO membranes as between the NF and UF membranes. We selected the Hydranautics ESPA2 membrane for our longevity test.

Removal Efficiency Summary for the SWRO and the BWRO

0

10

20

30

40

50

60

70

80

90

100

COD Chloride Alkalinity TDIS Silica

Rem

oval

Effi

cien

cy (%

)

SWRO

BWRO

37

Pilot Plant TrialsTest location 3 – Chemical usage

▪ Cleaning of membranes– Ultrasil 110 (Caustic)– Ultrasil 75 (Acid)– Hypochlorite– Ultrasil 02 (Surfactant)

▪ Anti-scaling– CleanTreat 4500 (Ecolab)– Hypersperse (GE Betz)– Flocon (GE Betz)

▪ pH scale control– Acid (Citric, acetic, hydrochloric, sulphuric)– Caustic (NaOH)

▪ Preservation– Ultrasil 73 (Ecolab)

▪ Sanitiser– Oxysan – H2O2 based (Ecolab)

38

Pilot Plant TrialsTest location 3 – CIP

Fresh Water Flux (LMH @25°C) after each chemical cleaningNanofiltration Filmtec 270 - 109 days of effective operation

0

10

20

30

40

50

60

70

80

90

100

25/09/2007 14/11/2007 3/01/2008 22/02/2008 12/04/2008 1/06/2008 21/07/2008 9/09/2008 29/10/2008 18/12/2008

Date

Tem

p. a

dj F

lux

(lmh

@25

°C)

▪ The UF Tubular, 295 operating days, it is dropping the most in the beginning, but flattens as the CIP procedure is improved by soaking time and acid clean also during the maintenance CIP’s.

▪ The Filmtec 270, 109 operating days, stable.

▪ The Hydranautics ESPA2, 64 operating days. relatively stable.

Fresh Water Flux (LMH @25°C) after each chemical cleaningUF Tubular - 295 days of effective operation

0

10

20

30

40

50

60

70

80

90

25/09/2007 14/11/2007 3/01/2008 22/02/2008 12/04/2008 1/06/2008 21/07/2008 9/09/2008 29/10/2008 18/12/2008

Date

Tem

p. a

dj F

lux

(lmh

@25

°C)

Fresh Water Flux (LMH @25°C) after each chemical cleaningNanofiltration BWRO ESPA2 - 64 days of effective operation

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

1/06/2008 21/06/2008 11/07/2008 31/07/2008 20/08/2008 9/09/2008 29/09/2008 19/10/2008 8/11/2008 28/11/2008 18/12/2008

Days

Tem

p. a

dj F

lux

(lmh

@25

°C)

39

Proposed membrane arrangement

40

Pilot Plant TrialsProposed membrane arrangement

▪ Overall process layout

UF NF RO

H2SO4, pH<6.8 H2SO4, pH<6.3

Anti-scaling

Case 2a

UF NF

H2SO4, pH<6.8Case 3

4.5 ML/d

2.2 ML/d

41

Pilot Plant TrialsProposed membrane arrangement

▪ Summary of permeate fluxes

152751.24855.67.4100BWRO

HydranauticsESPA2, pH<6.3

22675521180383.5530UF Tubular

(30 kD, 84% circ)

2914587.630022.71.4220NF Filmtec 270300 D, pH<6.8

Membrane:PermeateProduced

(l/h)

Specific energy:

(kWh/m3)

Average flux:

(lmh@25°C)

TMP:(kPa)

Permeability:(lmh/bar @25°C)

VCF:Spec. invest.

cost:(A$/m2)

Operating time:(days)

42

Pilot Plant TrialsProposed membrane arrangement

▪ Overall layout

UF Tubular NF Filmtec 270 BWRO Hydra. ESPA2Pre-filter

8 week 1 week 4 week 2 week

pH<6.8, H2SO4Anti-scaling for Silica control

Acid CIP

pH<6.3, H2SO4

43

Pilot Plant TrialsProposed membrane arrangement

▪ Ultrafiltration Tubular membrane

The pilot used 3 m long elements A full scale plant should use 1.5 m long elements

44

Pilot Plant TrialsProposed membrane arrangement

▪ The NF & RO crossflow profile based on the pilot trials

4" Crossflow profile calculation - based on pilot trials(NF Filmtec 270 & RO Hydranatuics ESPA2)

2,130

1,770

1,450

1,170

930

730

570450

1,530

1,330

1,150

990850

730630

550 490 450

0

500

1,000

1,500

2,000

2,500

10 9 8 7 6 5 4 3 2 1Number of elements

4" C

ross

flow

pro

files

(L/h

)

NF 4" Crossflow

RO 4" Crossflow

45

Pilot Plant TrialsProposed membrane arrangement

▪ A proposed NF configuration– Conservative and flexible

7301170 (1140)

5701170 (1140)

5701450

240 200

240 200 160

280 240 200 160

Recovery% = 87.4% (VCF=8)Configuration: 3 stages in a 4:3:2 array8” Qp = 15,000 L/h 8” Qf = 17,162 L/h8” Qr = 2,162 L/h

1st Stage 2nd Stage 3rd Stage

Future Case 2a: will need 12 trainsFuture Case 3: will need 6 trains

46

Pilot Plant TrialsProposed membrane arrangement

▪ A proposed RO configuration– Conservative and flexible

7301330 (1260)

6301330

180 160 140 120

180 160 140 120 100

Recovery% = 72.5% (VCF=5)Configuration: 2 stages in a 5:4 array8” Qp = 7,250 L/h 8” Qf = 10,000 L/h8” Qr = 2,750 L/h

1st Stage 2nd Stage

Future Case 2a: 25 trains of this

47

Pilot Plant TrialsOverall operating conclusions

▪ Operating the pilot plant ourselves has given us the information we find relevant for this study

▪ UF Tubular membrane was chosen because of its robustness

▪ For our chosen Tubular polymeric membrane, a simple pre-filtration technique such as sand filters or a bow-screen will be good enough

▪ We need to operate in TMP control on the UF and NF, but Flux control was best on the RO.

▪ We need to adjust the pH for both the NF and RO to control the scaling issues

▪ SAR could not be improved much by using NF membranes

48

Thank you for your time!Questions?

For more details, please contact: lasse.blom@yahoo.com

- The End -