nanofiltration membrane characterization using mass transfer data

Nanofiltration Membrane Characterization

using Mass Transfer Data with Emphasis

on Temperature Effects

Ramesh R. SharmaTrussell Technologies, Inc., Pasadena, CA

Shankar ChellamDepartment of Civil and Environmental Engineering,

University of Houston, Houston, TX

Importance of Nanofiltration

• Capable of achieving high removals of

natural organic matter, disinfection by-

products, pesticides, arsenic, hardness,

etc.

• Lower cost than reverse osmosis

technology

• However, transport mechanisms are not

yet fully understood

Thin Film Composite Nanofilter

Porous support~ 40 µm

Reinforcing fabric~ 120 µm

Thin skin ~ 1 µm

EFFECTIVE skin layer parameters: pore size distribution,

porosity, tortuosity, thickness, and charge density

Temperature – Important Variable

NF/RO feed water temperature changes over time

• 1 - 26 °C – River Oise, France

37 MGD NF Plant, (Ventresque et al., 1997)

• 1 - 26 °C – Occoquan Reservoir, Virginia, USA

NF Pilot scale Study (Chellam et al., 1997)

• 10 - 25 °C – Ehime, Japan (Sea Water)

3.7 MGD RO plant (Taniguchi and Kimura, 2000)

• 10 - 35 °C – Kuwait (Sea Water)

32 MGD RO plant (Abdel-Jawad et al., 2001)

Motivation – Temperature Effects

With increase in temperature

• Salt passage↑↑↑↑ (Mehdizadeh, et al. 1989)• Natural organic matter passage↑↑↑↑ (Her, et al. 2000)• Arsenic passage↑↑↑↑ (Waypa, et al. 1997)

•Water permeability↑↑↑↑ (Sharma, et al., 2003)





IDEALIZED NF PORE

STRUCTURE


High TemperatureLow Temperature





Seasonal Variation

Water Quality Parameters

• Composition

• Nature

• Concentration

• Temperature

Membrane Parameters

• Pore size distribution

• Porosity

• Thickness

• Fixed charge density

NF Performance

Fouling Selectivity

Objectives

Investigate temperature (5 – 41°°°°C) effects

–Water Permeation

– Convective transport

• Pore size distribution and sieving

– Diffusive transport

• Solute permeability

– Develop insights into pore structure,morphology from selectivity measurements

Membranes Employed

Manufacturer Model Composition MWCO

Osmonics,

Minnetonka, MN

DL Polyamide ~200

Koch Fluid Systems,

San Diego, CA

TFCS Polyamide ~ 300

Manufacturer specifications

Water Permeation

Schematic of Nanofiltration Apparatus

Pressure

transducer

Temperature Controlled Room

DAQ

Computer

Feed tank

Compressed air

Gear pump

Permeate

Valve

P

Flow meter

Flow sensor

Temperature sensor

Concentrate Recycle

Validity of Darcy’s Law – Nanofiltration

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

0

1

2

3

4

5

6

7

8

9

10

DL membrane 41 °C

23 °C

5 °C

15 °C

35 °C

Flux (µµ µµm/s)

Pressure differential, ∆∆∆∆P (MPa)

PR

J

LR

PLJ

m

V

p

m

PV

∆=

=

∆=

1

1

µ

µ

Validity of Darcy’s Law – Microfiltration

0 40 80 120 160 200

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

Flux x viscosity (N cm/m

2)

Pressure (kPa)

4 ºC

23 ºC

35 ºC

46 ºC

R2 = 0.998

0.2 µm membrane

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 20 40 60 80 100 120 140 160 180 200

Pressure (kPa)

Flux (cm/s)

4 °C

22 °C

35 °C

46 °C

0.2 µm membrane

Increasing

Temp.

Temperature Effects on Water Permeation

0.0 0.2 0.4 0.6 0.8

0

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Flux (µm/s)

∆P (MPa)

0.0 0.2 0.4 0.6 0.80

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Jv×µ (MN/m

)

∆P (MPa)

PR

J

PLJ

m

V

PV

∆=

∆=

1µ


0.0 0.2 0.4 0.6 0.8

0

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Flux (µm/s)

∆P (MPa)

0.0 0.2 0.4 0.6 0.80

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Jv×µ (MN/m

)

∆P (MPa)

p

m

PV

LR

PLJ

µ=

∆=

1


0.0 0.2 0.4 0.6 0.8

0

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Flux (µm/s)

∆P (MPa)

0.0 0.2 0.4 0.6 0.80

2

4

6

8

10

DL membrane

41 °C

23 °C

5 °C

15 °C

35 °C

Jv×µ (MN/m

)

∆P (MPa)

p

m

PV

LR

PLJ

µ=

∆=

1

+

−=T273

1

296

1 TCFexp

J

(T)J

Cv,23

v

o

Temperature Effect on Transport

Arrhenius Equation

−−=

° 296

11ln

23TR

E

L

L

CP

p

where

Lp = Water or solute permeability

E = Activation energy

T = Temperature (K)

R = Universal gas constant

3.2 3.3 3.4 3.5 3.6-0.2

-0.1

0.0

0.1

0.2

3.2 3.3 3.4 3.5 3.6-0.2

-0.1

0.0

0.1

0.2

Ln (Lp)/(L

p) 23 °C

TFCS membrane

Ew = 21.3 ± 0.5 kJ/mol

1000/T (T in K)

Ln (Lp)/(L

p) 23 °C

DL membrane

Ew = 23.8 ± 0.9 kJ/mol

Pure Water Permeability

Temperature Correction Equations

+

−=T273

1

298

1 2617exp

J

(T)J

Cv,25

v

o

TFCS Membrane

+

−=T273

1

298

1 2918exp

J

(T)J

Cv,25

v

o

DL Membrane

+

−=T273

1

298

1 2100exp

J

J

C25,v

v

o

MF/UF Membrane

+

−=T273

1

298

1

Eexp

J

(T)J act

Cv,25

v

Ro

Need for Membrane-Specific TCFs

5 10 15 20 25 30 35 40

0.6

0.8

1.0

1.2

1.4

1.6

Permeate flux normalized to 25 °C

Temperature (°C)

TFCS

DL

Generic

1.03T-25

5 10 15 20 25 30 35 40-4

-2

0

2

4

6

8

10

12

Relative error (%)

Temperature (°C)

Perfect fit

DL

TFCS

Neutral Solute Transport and

Changes in Membrane Morphology

Solutes Employed

Solute

Molecular

weight

(g/g-mol)

Molar

volume

(cm3/mol)

Stokes

radius

(nm)

Mean

molecular

radius (nm)

PEG (100K) 100,000 5,180,014.8 8.26 NA

PEG (35K) 35,000 1,813,014.8 4.87 NA

PEG (20K) 20,000 1,036,014.8 3.68 NA

αααα-cyclodextrin 973 886.8 0.700 0.866

Raffinose 594 480 0.583 0.573

Sucrose 342 325 0.471 0.508

Dextrose 180 166 0.366 0.374

Xylose 150 148 0.300 0.326

t-butyl alcohol 74 103.6 0.278 0.303

Glycerol 92 85.1 0.258 0.302

Ethylene glycol 62 63.5 0.211 0.270

Ethanol 46 62.6 0.198 0.239

Methanol 32 37 0.148 0.211

Water 18 14.8 0.150 0.140

Choice of Solute Size Parameter

1.5 3.0 4.5 6.0 7.5 9.0

0

200

400

600

800

1000

1.5 3.0 4.5 6.0 7.5 9.0

1.5

3.0

4.5

6.0

7.5

1.5 3.0 4.5 6.0 7.5 9.00

200

400

600

800

1000

Y = 150.97x - 345

R2 = 0.98

Molecular weight (gm/mol)

Mean molecular radius (A °)

Y = 0.87x + 0.04

R2 = 0.95

Stokes radius (A °)


Y = 132.27x - 290

R2 = 0.98

Molar volume (cm

3/mol)


MMR as a representative

solute size parameter

Experimental Conditions

Flux: 1 to 50 L/m2 h

Constant and low recovery (< 1%)

Constant cross flow velocity: 9 and 19 cm/s

Feed water TOC: 20 mg/L

Temperature: 5 , 15, 23, 35 and 41 °°°°C

Irreversible Thermodynamics Model

MembraneQf, Cf Qp, Cp

Qc, Cc

)1(

)1(1Rejection

F

F

C

C

m

p

σσ−−

=−=

−−= vJ

P

σ1expF where

cJdx

dcxPJ VS )1( σ−+

∆−= (Spiegler and Kedem 1966)

σRejection and 0FJ∆P

0Rejection and 1F0J0∆P

v

v

→→⇒∞→⇒∞→

→→⇒→⇒→

Rejection – Transmembrane Pressure

0.0 0.2 0.4 0.6 0.8

0

20

40

60

80

100

Temperature = 23 °C

Recovery < 1%

Raffinose

Sucrose

Dextrose

Xylose

Glycerol

t-butyl alcohol

Ethylene glycolEthanol

Rejection (%)

∆∆∆∆P (MPa)



Qc, Cc

)1(

)1(1Rejection

F

F

C

C

m

p

σσ−−

=−=

−−= vJ

P

σ1expF where

cJdx

dcxPJ VMS )1( σ−+


)(intercept 0J

1J :Note


v

v

v

→⇒∞→

→→⇒∞→⇒∞→

Rejection – Transmembrane Pressure

0.0 0.2 0.4 0.6 0.8

0

20

40

60

80

100


Recovery < 1%

Raffinose

Sucrose

Dextrose

Xylose

Glycerol

t-butyl alcohol

Ethylene glycolEthanol

Rejection (%)

∆∆∆∆P (MPa)

Rejection – Inverse Flux

0.0 0.2 0.4 0.6 0.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

DL membrane


Recovery < 1%

Xylose

Ethanol

Ethylene glycol Glycerol

t-butyl alcohol

Dextrose

SucroseRaffinose

Inverse permeate flux, 1/Jv (µµµµm/s)

-1

Solute rejection (-)


0.0 0.2 0.4 0.6 0.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

DL membrane


Recovery < 1%

Xylose

Ethanol


t-butyl alcohol

Dextrose

SucroseRaffinose


-1


Reflection coefficient


0.0 0.2 0.4 0.6 0.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

DL membrane


Recovery < 1%

Xylose

Ethanol


t-butyl alcohol

Dextrose

SucroseRaffinose


-1


Reflection coefficient

Permeability

Lognormal Model

• Assumption of lognormal distribution of pore

sizes

• No hydrodynamic lag involved

( )dr

S2

)rln()rln(exp

r

1

2S

1*)r(

*r

02p

2

p∫

−−

π=σ

Rejection at Jv→∞

∞→vJ

Lognormal Distribution

0.0 0.2 0.4 0.60.0

0.2

0.4

0.6

0.8

1.0

0.0 0.2 0.4 0.60.0

0.2

0.4

0.6

0.8

1.0

Reflection coefficient (-)

TFCS membrane

r1 = 0.24 nm

SP1 = 0.1 nm

23 °C

Recovery < 1%

Solute Stokes radius (nm)

DL membrane

r1 = 0.3 nm

SP1 = 0.07nm

23 °C

Recovery < 1%


Mean molecular radius (nm)

Purely Sieving Bimodal Distribution

0.1 1 10

0.0

0.2

0.4

0.6

0.8

1.0TFCS

R2 > 0.99

23°C



PEGs


Purely Sieving Bimodal Distribution

0.1 1 10

0.0

0.2

0.4

0.6

0.8

1.0TFCS

R2 > 0.99

23°C



0.1 1 100.01

0.1

1

10

Type 2: r2 = 8 nm

SP2 = 0.1 nm

h2 = 0.1

Type 1: r1 = 0.24 nm

SP1 = 0.2 nm

h1 = 0.9

Pore density function (nm

-1)

Stokes radius (nm)

PEGs


Purely sieving bimodal distribution

0.1 1 10

0.0

0.2

0.4

0.6

0.8

1.0TFCS

R2 > 0.99

23°C



0.1 1 10

0.01

0.1

1

10

Type 2: r2 = 8 nm

SP2 = 0.1 nm

h2 = 0.1

Type 1: r1 = 0.24 nm

SP1 = 0.2 nm

h1 = 0.9

Pore density function (nm

-1)

Stokes radius (nm)

PEGs


0.1 0.2 0.3 0.4 0.5 0.6

0.0

0.2

0.4

0.6

0.8

1.0

DL membrane

5°C

15°C

23°C

35°C

41°C


Solute Stokes radius (nm)Mean molecular radius (nm)

Temperature Effect on PSD

T

Increasing Mean Pore Radii

0 5 10 15 20 25 30 35 40 45

0.27

0.30

0.33

0.35

Mean pore radius (nm)

Temperature (°C)

DL

TFCS



Qc, Cc

)1(

)1(1Rejection

F

F

C

C

m

p

σσ−−

=−=

−−= vJ

P

σ1expF where

cJdx

dcxPJ VS )1( σ−+


0Rejection and 1F0J0∆P


v

v

→→⇒→⇒→

→→⇒∞→⇒∞→

Arrhenius Relationship

3.15 3.24 3.33 3.42 3.51 3.60

0.0

0.5

1.0

1.5

2.0

2.5

3.15 3.24 3.33 3.42 3.51 3.60

0.0

0.5

1.0

1.5

2.0

2.5

3.0

1000/T (T in K)

Ln(P/P

5°C)

TFCS

dextro

se

xyloseethylene glycol

t-butyl alcoholglycerol

ethanol

t-butyl alcohol

dextro

se

xylose

glycerol

ethylene glycol

DL

−=RT

EexpPP

p

o

Activation Energy – Mean Molecular Radius

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0

10

20

30

40

50

60

70

0.0 0.1 0.2 0.3 0.4 0.50

10

20

30

40

50

60

70

80

50th percentile of ratio of molecular size to

membrane pore radius (λλλλ0.5)

Dextrose

Xylose

t-butyl alcohol

Glycerol

Ethylene glycol

Ethanol

Activation energy of permeation (kJ/mol)

Activation energy (kJ/mol)


Conclusion

Low Temperature High Temperature




Lessons Learned!

• NF membranes are NOT like RO membranes

• Because pore sizes are lognormally

distributed, even a few large pores may

determine membrane selectivity

• Need for a membrane specific temperature

correction factor

• Rejection of larger size contaminant is more

sensitive to temperature

Peer Reviewed Articles• Temperature Effects on Sieving Characteristics of

Nanofiltration Membranes: Pore Size Distributions and

Transport Parameters. JMS (2003) 223, 69-87

• Temperature Effects on Morphology of Porous Nanofiltration

Membranes. ES&T, (2005) 39, 5022-5030.

• Temperature and Concentration Effects on Electrolyte

Transport across Porous Nanofiltration Membranes. JCIS,

(2006) 298, 327-340.

• Frictional interpretation of thermodynamic transport

parameters for porous nanofiltration membranes. Journal of

Water Supply: Research and Technology—AQUA Vol 55 No 7-

8 pp 571–587

Acknowledgement

Funding agency National Science

Foundation

Collaborators Dr. Advincula and Prasad for

molecular mechanics simulations

Questions!

nanofiltration membrane characterization using mass transfer data

Documents