Tailoring of water swollen PVA membrane for hosting carriers in CO2 facilitated transport membranes

Tailoring of water swollen PVA membrane for hosting carriers in CO2 facilitated transport membranes

Accepted Manuscript Tailoring of Water swollen PVA membrane for hosting carriers in CO2 facilitated transport membranes Muhammad Saeed, Sikandar Rafiq...

1MB Sizes 0 Downloads 9 Views

Accepted Manuscript Tailoring of Water swollen PVA membrane for hosting carriers in CO2 facilitated transport membranes Muhammad Saeed, Sikandar Rafiq, Linda Hildegard Bergersen, Liyuan Deng PII: DOI: Reference:

S1383-5866(16)32638-7 http://dx.doi.org/10.1016/j.seppur.2017.02.022 SEPPUR 13552

To appear in:

Separation and Purification Technology

Received Date: Revised Date: Accepted Date:

7 December 2016 9 February 2017 9 February 2017

Please cite this article as: M. Saeed, S. Rafiq, L. Hildegard Bergersen, L. Deng, Tailoring of Water swollen PVA membrane for hosting carriers in CO2 facilitated transport membranes, Separation and Purification Technology (2017), doi: http://dx.doi.org/10.1016/j.seppur.2017.02.022

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Tailoring of Water swollen PVA membrane for hosting carriers in CO2

2

facilitated transport membranes

3

Muhammad Saeed1,3, Sikandar Rafiq1,2, Linda Hildegard Bergersen3,

4

Liyuan Deng1

5

[email protected]

6

[email protected]

7

Department of Chemical Engineering, Norwegian University of Science and Technology

8

(NTNU), Trondheim, 7491, Norway

9

Department of Chemical Engineering, COMSATS Institute of Information Technology,

10

1.5km Defence Road, Off Raiwind Road, Lahore 54000, Pakistan

11

Electron Microscopy Laboratory at Department of Oral Biology, University of Oslo (UiO),

12

Oslo, 0316 Norway

13 14 15

Abstract:

16

Composite membranes were synthesized by incorporation surface modified multi-walled

17

carbon nanotubes (CNTs) in polyvinyl alcohol (PVA) for investigating the facilitated

18

transport of CO2 under humidified conditions. The effect of pH and nano filter concentration

19

on swelling index and CO2/N2 separation performance were observed for membrane by gas

20

permeation tests. A progressive increasing trend in the swelling behavior of was observed

21

with respect to time for all developed membranes and 1% CNT at pH12 showed the highest

22

equilibrium swelling% up to 256±30%. Furthermore, the experimental membrane swelling

23

data was in agreement with the theoretical model as a function of time. Gas permeation test

24

conducted on premixed CO2/N2 gas with 10% CO2 showed CO2 permeance increased from

25

0.18±0.02 to 0.40±0.03m3 (STP)/m2 bar-h at 1.22bara for pure PVA membrane as pH shifted

26

from acidic to the basic regime. This increased permeance resulted from the higher uptake of

27

water by the membrane as water acted as a carrier and facilitate CO2 transport across the

28

membrane. Improved permeance was observed with CNT at 1% at pH 0.44±0.02 under 100%

29

humidified conditions with a consistent selectivity of 60±1.

1

1 2 3

Keywords:

4

CO2 capture, membrane technology, carbon nanotubes, composite membrane, empirical

5

modelling, mixed matrix membrane, electron microscopy.

6

Highlights:

7

Water swollen composite membrane for post combustion CO2 capture

8

Modeling of membrane swelling degree

9

Multi-walled carbon nanotube enhanced membrane for CO2 capture

10

Effect of humidity on membrane separation performance

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 2

Introduction

1

1.

2

Post-combustion CO2 production from fossil fuel combustion is the largest point source of

3

global CO2 emissions. The CO2 concentration in flue gas from natural gas-fired power plants

4

varies between 4–10%, which increases for the coal-based power plant (up to 13.5 % CO2).

5

Among the several technologies developed for CO2 capture and emission reduction, the

6

membrane is one of the most promising techniques. In recent years, membrane technology

7

has been intensively investigated for CO2 separation since it offers several advantages over

8

other techniques. Membrane separation of CO2 is an environmentally friendly as compared

9

to conventional absorption process, it has lower capital cost and higher specific surface area,

10

it has a modular design and greater suitability for applications in remote and compact

11

locations[1-4]. Development of highly efficient membranes is the most important parameter

12

in determining the efficiency and separation performance of a membrane system. During

13

membrane development, permeability and selectivity parameters are of great importance and

14

acquiring high permeability and selectivity are key issues for efficient CO2 separation[5, 6].

15

Conventional porous membrane offers high gas permeance but low CO2/N2 selectivity

16

whereas highly selective self-supported dense membranes offers high selectivity but low CO2

17

permeance and low mechanical strength. The concept of the composite membrane with a thin

18

dense layer of selective material over porous support has revolutionized the membrane gas

19

separation technology. Furthermore, carrier-mediated (CM) transport of CO2 has resulted in

20

membranes with high CO2/N2 separation performance. These membranes have shown great

21

potential to reach a performance beyond the Robeson upper bound region[7, 8]. The transport

22

mechanism in a carrier-mediated transport membrane involves reversible reactions between

23

the reactive carriers incorporated in the membrane material with the target gas molecule and

24

this interaction governs the gas transportation across the membrane[9-11]. In various

25

industrial processes which involve CO2 capture especially at post-combustion conditions

26

from flue gas, the feed streams are saturated with water vapor due to which CO2 permeability

27

may reduce as a result of competitive sorption of water in the membranes[12]. However, in

28

the case of CM membranes, in particular, amine-promoted and mimic enzyme membrane, the

29

presence of moisture causes an enhancement in carrier activity. This in turn greatly enhances

30

CO2 transport by involving in the reversible reaction of CO2 with the carrier and hence these

31

membranes are highly efficient for CO2 separation[13-17]. These membranes possess high

32

selectivity and

33

membranes especially at low CO2 concentrations and thus driving forces[18].

permeability as compared to

3

the dry gas operation and conventional

1

Research work on fixed site carrier membrane by Kim et al.[11] was carried out using a

2

water swollen crosslinked polyvinyl amine (PVAm)/polyvinyl alcohol (PVA) that acted as

3

fixed carrier.

4

separation performance of the membrane. The CO2 gas molecules in presence of moisture

5

formed HCO3- ions which had higher diffusivity in the membrane and hence

6

transported quickly and selectively through the membrane. The membrane exhibited a high

7

selectivity of CO2 over N2 and CH4. Similarly, Ansaloni et al. [19] reported the effect of water

8

on CO2/H2/CH4 separation performance for CM membrane containing amine functionalized

9

carbon nanotubes dispersed in PVA film to produce a water swollen composite membrane.

10

Park and Lee[20] reported a water swollen PVA membrane fixated with carbonate salt to

11

facilitate CO2 transport in humid conditions and enhance the CO2/N2 separation performance.

12

More recently, a novel concept of the carbonic anhydrase-like mimic enzyme in PVA

13

composite membranes reported by Saeed and Deng [17] showed high CO2/N2 separation

14

performance in humid conditions. Some fixed site carrier membranes were prepared by Zang

15

et al. applying amine containing polymers like poly(N-vinyl-γ-sodium aminobutyrateco-co-

16

sodium acrylate), pentaerythrityl tetraethylenediamine (PETEDA) dendrimer, polyvinyl

17

amine (PVAm). The resulting membranes were shown to have good stability and were quite

18

effective for CO2 transport across the membranes[21-24].

19

In the case of the mixed feed gas, membrane swelling by sorption of a component leads to

20

higher permeance. Hydrophilic membrane materials like PVA swells in presence of humidity

21

and results in higher CO2 permeance. FTMs with a carrier like an amine,

22

carbonate/bicarbonate salt or mimic enzyme require the aqueous environment to efficiently

23

separate CO2 from a mixture of gasses. A hydrophilic polymer matrix with high tensile

24

strength and water uptake, excellent film forming ability and low cost is an ideal candidate to

25

develop a membrane that can host these carriers for mediated transport of CO2. PVA is

26

extensively used to cast a defect free membrane onto a porous support. The swelling behavior

27

of PVA can be tailored by physical or chemical crosslinking to adjust its swelling degree and

28

maintain membrane integrity in aqueous environments. Taking advantage of its high swelling

29

behavior, good film forming ability and high tensile strength, PVA has been used to host

30

carriers and nanoparticles that mediate facilitated transport of CO2[17, 19, 20, 25].

31

The transport of CO2 through the water swollen membrane is carried out as carbonate and

32

bicarbonate and this is induced by the presence of water in the humidified flue gas streams as

33

mentioned earlier. The use of aminated polymers in CO2 facilitated transport membranes

34

have been studied extensively in past years and several of the investigators found that these

The humidity in gas stream played an important role in improving the

4

were

1

aminated polymeric membranes depicted high performance as water swollen conditions

2

rather than in a dry state[26-31]. This is because CO2 transformed into HCO3− mobile ions

3

due to the interaction between CO2 water and carrier groups which increased wet membrane

4

performance[30-32]. The water in a hydrophilic membrane swells the selective later by

5

producing aqueous domains in tightly packed polymer chains. This swollen domain not only

6

helps the carrier like amines and mimic enzyme to hydrolyze CO2 and produce HCO3− ions

7

but due to higher diffusivity of gasses in the aqueous phase, increases the permeance of

8

diffusing gasses. This was further verified by Park et al. [20] as they controlled the swelling

9

degree of PVA membrane containing potassium carbonate by chemical crosslinking with

10

glutaraldehyde and observed a decline in gas permeance with an increase in the degree of

11

crosslinking. Thus the presence of water promotes CO2 transportation across using facilitated

12

transport membrane. Similarly in Mimic enzyme promoted membrane reported by Saeed and

13

Deng [17], water played a key role in facilitating the transport of CO2 through the membrane.

14

Figure 1 shows the CO2 transportation mechanism by the carrier attached to the polymer

15

backbone, facilitated by H2O forming bicarbonate on the feed side of the membrane. CO2 is

16

released on the permeate side as a result of protonation of bicarbonate as a result of reversible

17

reaction i.e equation 1 (a,b)

18

CO2 + H2O → H2CO3

(1a)

19

H2CO3 → CO2 + H2O

(1b)

20 21

Figure 1 A mechanism of facilitated transport in a facilitated transport membrane. Adapted from Park et al. [20]

22

5

1

Just like chemical crosslinking of polymer ,physical crosslinking has also been reported in

2

the literature to alter the membrane swelling behavior [17, 33]. Besides post casting

3

membrane treatment like thermal treatment and surface chemical crosslinking, it is also

4

possible to treat membrane casting solution to modify the separation performance of a

5

facilitated transport membrane. Kim et al. [2] adjusted the pH of membrane casting solution

6

to adjust the swelling of polymer and improve the CO2/N2 separation of FSC membrane.

7

Saeed and Deng[17] also studied the effect of pH on the separation performance of mimic

8

enzyme membrane. Deng and Hagg [34] maintained the swelling degree of PVAm membrane

9

at high pressure by embedding the polymer layer with carbon nano spacers to maintain high

10

separation performance.

11

In this work, the PVA-based nanocomposite membrane has been investigated for facilitated

12

transport of CO2 under humidified conditions. To altering the membrane swelling to tailor

13

membrane performance, the effect of pH, selective layer thickness and nanoparticles contents

14

as surface modified hydrophilic carbon nanotubes are explored over the performance of the

15

membrane.

16

Experimental

17

2.

18

2.1. Materials

19

Polyvinyl alcohol Mwt 89000-99000 (89% hydrolyzed) was supplied by Sigma-Aldrich,

20

NaOH (99% pure) was supplied by VWR and PSf ultrafiltration membrane (MWCO 50,000)

21

was acquired from Alfa Laval10% CO2 in N2 and pure CH4 was supplied by yara.The multi-

22

walled CNTs (VGCF-XTM, diameter 15nm, length 1–3µm) were kindly supplied by Showa

23

Denko K.K. (SDK, Japan). A SEM and TEM images of the VGCF-XTM CNT can be found

24

in Figure 3 (a,b)[35].

25

26

2.2. Membrane development

27

A composite membrane with a defect-free thin dense layer of PVA (0.4 ~0.85µm thickness)

28

over porous polysulfone support was developed using dip coating technique. A 10% bt

29

weight solution of PVA in DI water was produced gravimetrically. The stock solution was

30

heated to 90 oC in a heating cabinet to get a clear solution. The stock solution was mixed by 6

1

rolling overnight at room temperature. The pH of PVA solution was measured to be around 5

2

using pH papers. This PVA solution was diluted with DI water and 0.1M NaOH solution to

3

produce a membrane casting solution with 2% or 1% PVA at different pH. The pH of casting

4

solutions was adjusted from 5 to 7, 9, 10 and 12.

5

Membranes containing CNTs were prepared by diluting the 10% PVA stock solution with

6

CNTs, 0.1M NaOH and distilled water to produce casting solution of 2% and 1% PVA, 0.5, 1

7

and 1.5 % CNTs with respect to PVA at pH of 5, 9, 10 and 12. In this work membranes at

8

different pH and CNTs loadings were prepared with a uniform dense layer of polymer over

9

porous PSf. The thickness of the dense selective layer was altered by adjusting the

10

concentration of PVA in membrane casting solution from 2.0% to 1.0% wt PVA to study the

11

effect of membrane thickness on overall membrane performance.

12

A flat sheet Psf ultrafiltration membrane washed with DI water was masked on a glass slab to

13

prevent the solution from wetting the back side of the membrane. This support membrane

14

was then dipped into the casting solution for 30 seconds. After lifting the membrane above

15

the coating bath, it was held nearly vertical to allow extra solution flowing off evenly and

16

form a homogeneous and continuous film. The coated membrane was recoated after drying it

17

for 3 hours in a dust free environment at room temperature to ensure a defect free surface.

18

The twice coated membrane was left in the dust-free space overnight, and then to dry in a

19

convection oven at 45oC for 3 hours. Dried membranes were heat treated at 110oC for 1 hour

20

to physically crosslink the membrane. The membrane preparation procedure is schematically

21

illustrated in figure 2. DI Water

NaOH solution

CNTs in water

PVA 10% PVA stock solution

2% PVA casting solution

DI Water Heat at 90oC for 6 h and overnight rolling

Rolling

Crosslinking Oven

Sonnication for CNTs PSf membrane Porous support DI Water

22 23

Dip coating Dipped rwice for 30 S

Masking of support on glass slab

Washed with running DI water for 30 min

Figure 2: Schematic diagram of membrane casting procedure.

7

Dip coated membrane dried at room temperature

110oC for 1 hr

1

2.3.Membrane morphology

2

Electron microscopy was used to examine the surface and cross-section of the membrane to

3

ensure a defect free dense film of the membrane over porous support and determine the

4

thickness of the selective layer. A scanning electron microscope (LVSEM Hitachi S3400N)

5

was used to analyze the thickness of each layer.Cross-sectional samples were prepared by

6

freeze fracturing the membranes in liquid nitrogen. Samples were sputtered with gold to

7

produce a conductive layer of 20nm thickness before observation. A SEM image of the cross-

8

section of a PVA membrane is shown in Figure3 (c).

(a) 9

(b)

(c)

Figure 3. SEM (a) and TEM (b) images of VGCF-XTM CNTs, reprinted from [35], (c) Thin layer of PVA over

10

PSf.

11

2.4.Swelling test

12

The swelling behavior of the self-supported membrane under humid conditions at 20 oC was

13

measured by gravimetric analysis. 5 ml of 2% PVA solution with and without CNTs at

14

different pH (5,7,9,12) was powered on Teflon dishes having the same diameter. The

15

solutions were dried overnight at room temperature to produce self-supported membranes of

16

same dimensions. The produced self-supported

17

convective oven and then physically crosslinked at 110oC for 1 hour.

18

The samples (polymer films) were vacuum dried at 20oc for 3 hours prior to the tests. The

19

dried membranes were weighed and then placed in a closed container saturated with water

20

vapors. Samples were weighted after regular intervals to determine the uptake of water by a

21

membrane. The degree of swelling was calculated based on gravimetric analysis of samples.

22

Equation (2) for calculation of swelling degree is presented below.

23

% swell =

Ws − Wd × 100 Wd 8

membranes were dried at 45 oC in a

(2)

1

Where Ws and Wd are the masses of the swollen and dry membranes, respectively.

2

The kinetic of the swelling in the PVA and PVA/CNT samples at pH 5, 9, 10 and 12 were

3

studied by using a simple analysis based on the second order equation: the equilibrium

4

swelling degrees of the samples with the swelling–time curve can be evaluated, as expressed

5

in equation 3a [36]. The integration of equation 3a with the limits S = S0 at t = t0 and S = S at

6

t = t gives equation 3b:

7

dS = kS (Seq − S )2 dt

(3a)

8

1 1 t = + ( )t 2 S k S Seq Seq

(3b)

9

Where S is the swelling degree at time t, kS is the swelling rate constant and Seq denotes the

10

theoretical equilibrium swelling degree.

11

2.5.Permeation test

12

The developed composite membranes were tested for CO2 permeance and CO2/N2 selectivity

13

by using a premixed gas containing 10% CO2 in N2 as feed gas at various feed pressures

14

under humidified conditions. Pure methane was used as a sweep gas to maintain a constant

15

driving force across the membrane. The composition of the permeate gas was analyzed

16

online by a gas chromatograph equipped with a thermal conductivity detector

17

(MicroGC3000). The permeance of CO2 and N2 were calculated using equation 4.

PA J = A l ∆ pA

18

(4)

19

Here PA is the permeability of transporting gasses through the selective film of thickness l. J A

20

is the flux of gas at a partial pressure difference of ∆pA across the membrane. The selectivity

21

( α ) was calculated from permeance of CO2, PCO2, and N2, PN2 , as expressed in Eq. (5) [37]

22

α=

PCO 2 PN 2

(5)

23

Experiments were conducted at feed pressure between 1.2 bara and 5 bara at a temperature of

24

25oC and constant sweep flow at atmospheric pressure. The experimental setup is further

25

explained by Kim et al. [11].

9

1

The permeance of the gas was reported in units of [m3 (STP)/ (m2 h bar)]. For a convenient

2

comparison of the results, the unit conversion table of some commonly used gas permeance

3

units is given in Table 1.

4

Table 1 Permeance unit conversion table

GPU

m3(STP)/m2.bar.h

mol/m2.Pa.s

m3(STP)/m2.Pa.s

GPU

1

0.00274

3.38E-10

7.60E-12

m3(STP)/m2.bar.h

365.5

1

1.24E-07

2. 78E-09

mol/m2.Pa.s

2.96E+09

8.09E+06

1

0.0225

m3(STP)/m2.Pa.s

1.32E+11

3.60E+08

44.5

1

5

6

3.

7

Self-supported, physically crosslinked PVA membrane with and without CNTs at various pH

8

levels were bone dried in vacuum at 20oC. These dried membranes were weighed and placed

9

in a closed container saturated with water vapors. Gravimetric analysis was periodically

10

conducted to calculate the swelling degree of these samples. The weight of all the self-

11

supported membranes was periodically registered to calculate the swelling degree of the

12

membrane by using equation 10. Multiples samples were analyzed to calculate an average

13

swelling degree. The percentage swelling of membranes with respect to time is presented in

14

figure 4 (a) showing that with membrane water uptake increases with increase in pH of

15

casting solution. Drozdov et al. [38] also studied the effect of pH on swelling of hydrophilic

16

polymers and found that by increasing pH of hydrogel the swelling degree increases.

17

Similarly, the addition of multi-walled carbon nanotubes (CNTs) also results in an increase of

18

membrane swelling. In figure 4(b) a significant increase in swelling degree is observed for

19

membrane containing 0.5 and 1% CNTs with respect to PVA membrane. However, increase

20

in membrane swelling is not very significant as the CNT loading increases from 1% to 1.5%.

21

The carbon nanotubes are added to the membrane as nano spacers in a polymeric matrix.

22

Carbon nanotubes a generally hydrophobic in nature and forms clusters. Although the CNTs

23

used in this work are surface modified to be hydrophilic but their homogeneous dispersion in

24

aqueous polymeric solution is limited to lower concentration.

25

concentration of CNT (2% and above) were also prepared. However, due to the unstable

26

suspension of CNTs at this concentration membranes were not cast. The results presented in

Results and Discussions

10

Solutions with a higher

1

figure 4(b) indicate that there is no significant increase in membrane swelling with respect to

2

increasing in CNT loading from 1% to 1.5%.

3 4

(a)

5

6

(b)

7 8

Figure 4(a) Membrane swelling with respect to time for PVA membrane at pH 5, 9, 10, 12 at 20o C. (b)

9

Membrane swelling with respect to time for PVA membrane at pH 5 containing a different concentration of

10

CNTs at 20oC.

11

PVA is a hydrophilic membrane material and it swells by absorbing moisture from the humid

12

environment. By adjusting the pH from 5 to 12, its swelling degree can be increased.

13

Similarly, by adding CNT as nano spacers, its swelling degree can be altered. Membranes 11

1

were cast with 1% CNTs with respect to PVA and pH of casting solution was adjusted to pH

2

12. These casted membranes were also tested for swelling degree at 20oC and humidified

3

environment. The gravimetric analysis was used to calculate the swelling degree. The

4

percentage swelling of PVA membrane at pH 5 and pH 12 with and without 1% CNTs are

5

presented in figure 5. For a facilitated transport membrane like mimic enzyme promoted

6

membrane and polyvinyl amine based fixed site carrier membrane, swelling degree can be

7

adjusted by varying pH and/or addition of nanoparticles to tailor its separation performance.

8 9

Figure 5: Swelling degree of PVA membrane with respect to time at pH 5 and 12 with and without 1% CNTs at

10

20oC.

11

In figure 4 and 5, an increase in membrane swelling with respect to time is evident, however;

12

none of the membranes reaches equilibrium swelling after 4 days of testing. To calculate the

13

equilibrium swelling degree (Seq) and kinetic rate constant of swelling (ks), modeling of

14

experimental data was needed.

15

A numerical regression using an in-house built program named “Modfit” was conducted in

16

Matlab to model the swelling of membranes with respect to time. The correlation presented

17

in equation 6b was used with an objective function (t/S) and two variable parameters “ks” and

18

“Seq”. The numerical regression was conducted to fit the model with experimental data based

19

on residual decay to reach a global minimum. The experimental and modeled membrane

20

swelling at a different time for PVA membrane at pH 5,9,10 and 12 is presented in 7(a).

21

PVA membrane containing CNTs (0.5 to 1.5% ) at pH 5 is presented in figure 6(b) and PVA

22

membrane with and without 1% CNTs at pH 5 and 12 are compared in figure 6 (c).

12

1 2

(a)

3 4

(b)

13

1

(C)

2 3

Figure 6.(a) Time/swelling degree (t/S) as a function of time (t) for PVA with pH 5, 9, 10 and 12 at 20oC and

4

water saturated environment. (b) Time/swelling degree (t/S) as a function of time (t) for PVA with pH 5

5

containing 0.5%, 1% and 1.5% CNTs at 20oC and water saturated environment. (c) Time/swelling degree (t/S)

6

as a function of time (t) for PVA with pH 5, and 12 with and without 1% CNTs at 20oC and water saturated

7

environment.

8

The experimental data and modeled“ t/S” plotted in figure 5 indicates that modeling agrees

9

with experimental data. The calculated kinetic rate constant for swelling (ks) and equilibrium

10

swelling (Seq) constant are presented in table 2. A parity plot of modeled results and

11

experimental is presented in . This parity plot indicates that variation of model results with

12

respect to experimental data is less than 10% hence, confirming a good parameter fitting and

13

reliable model for prediction of equilibrium swelling degree.

14

Table 2 Equilibrium swelling and kinetic rate constant Membrane

S(eq)

ks

PVA pH 5

154±20

4,10E-05

PVA PH 9 PVA pH 10 PVA pH 12 0.5% CNT, PVA pH 5

165±22 192±25 208±16 188±14

2,24E-05 2,94E-05 6,79E-05 1,11E-04

1% CNT, PVA pH 5 1.5% CNT, PVA pH 5 1% CNT pH 12

194±19 190±12 256±30

1,93E-04 1,28E-04 2,03E-04

15 14

1 2

Figure 7 Parity plot of experimental “t/S” and modeled “t/S”.

3 4

The CO2/N2 separation performance was evaluated using a premixed gas contain 10% CO2 in

5

N2 at room temperature to simulate flue gas. Experiments were conducted with various feed

6

pressures ranging from 1-5bar. Methane was used as a sweep gas with constant flowrate at

7

atmospheric pressure. Both feed and sweep gasses were humidified before entering the

8

membrane cell. The membranes were humidified overnight to swell the PVA selective layer

9

by water. The CO2 permeance and CO2/N2 selectivity were calculated based on an online GC

10

analysis of permeate stream. The effect of membrane pH on CO2/N2 separation performance

11

at 1.2 bara feed pressure is presented in figure 9. This figure indicates that by increasing the

12

pH of membrane casting solution the CO2 permeance of membrane also increases. For a

13

composite membrane of PVA with a film thickness of 0.80~0.85µm, the CO2 permeance has

14

increased from 0.18 to 0.40±0.03 m3 (STP)/m2 bar-h. The CO2 permeance of PVA membrane

15

has increased to become more than doubled as the pH of the membrane is shifted from acidic

16

to the basic regime. The CO2/N2 selectivity, on the other hand, varies between 58 and 60

17

which could be considered as constant. The CO2/N2 selectivity of water based on Henry’s law

18

constant is also in the same range. This indicates that water is the carrier for CO2 facilitated

19

transport in the membrane. By adjusting the pH of the membrane, its swelling degree is

20

increased which resulted in higher uptake of water by a membrane. This high water uptake

21

facilitated the transport of CO2 in PVA film and resulted in higher CO2 permeance at constant

22

CO2/N2 selectivity.

15

100

0.6

80

0.5 60 0.4 40 0.3

Selectivity (CO2/N2)

3

2

CO2Permeance (m (STP)/m .bar.h)

0.7

20

0.2

0.1

0 4

6

8

10

12

pH

1 2

Figure 8 Effect of membrane pH on the CO2 permeance and CO2/N2 selectivity of the membrane. (Humidified

3

10% CO2 in N2 at 1.22bara feed gas and CH4 sweep at atmospheric pressure and operating temperature of 20o C)

4

The effect of membrane pH on CO2 separation is seldom reported in the literature. However,

5

Kim et al. [2] investigated the effect of pH on CO2 permeance and CO2/N2 selectivity of

6

PVAm/PVA membrane. They also observed a significant increase in CO2/N2 separation

7

performance.

8

Mixed matrix membranes containing functionalized and non-functionalized nanoparticles

9

dispersed in the polymeric matrix are often reported in the literature to enhance the separation

10

performance of membranes. Rafiq et al.[39] used non-functionalized silica particles to

11

improve CO2/CH4 separation of polysulfone/polyimide membrane. Similarly, Deng and

12

Hagg[34] used multi-walled carbon nanotubes to improve the CO2/CH4 separation

13

performance of PVAm. Surface modified multi-walled carbon nanotubes were dispersed in

14

PVA. The concentration of CNTs was varied between 0 to 2% with respect to PVA.

15

Generally, the carbon nanotubes are hydrophobic. Although the CNTs in this work were

16

surface treated to be hydrophilic but still their stable suspension in aqueous polymeric

17

solution was difficult to maintain at a concentration of 2% and higher CNTs with respect to

18

PVA. Self-supported films of PVA containing CNTs are presented in figure 9. In this figure,

19

it can be seen that the color of membrane turns from clear white to gray and then dark black

20

as the concentration of CNTs is increased from 0 to 1.5% with respect to PVA.

16

1

(a)

2

(b)

(c)

(d)

3

Figure 9 (a) Self supported membranes (a) PVA membrane without CNTs, (b) PVA membrane containing 0.5%

4

CNTs, (c) PVA membrane containing 1% CNTs, (d) PVA membrane containing 1.5% CNTs.

5 6

Nanocomposite membranes with a thin dense layer of PVA containing CNTs without pH

7

adjustment were tested for CO2/N2 separation performance by using premixed feed gas

8

containing 10% CO2 in N2 at 1.2 bara feed pressure and methane as sweep under humidified

9

conditions. The CO2 permeance and CO2/N2 selectivity of membrane containing different

10

loadings of CNTs are compared in figure 10. 100

0.6

80

2

CO2Permeance (m (STP)/m .bar.h)

0.7

3

60 0.4 40 0.3 20

0.2

0.1

0 0.0

11

Selectivity (CO2/N2)

0.5

0.5

1.0

1.5

CNT content (%)

12

Figure 10 Effect of CNT loading on the CO2 permeance and CO2/N2 selectivity of the membrane. (Humidified

13

10% CO2 in N2 at 1.22bara feed gas and CH4 sweep at atmospheric pressure and operating temperature of 20o C)

14

The CO2 permeance of membrane significantly increases by adding carbon nanotubes to the

15

polymeric matrix. The swelling tests of membranes with water also showed similar results. 17

1

This indicates that carbon nanotubes added to the polymeric matrix help increase the swelling

2

of the selective layer which then results in higher gas permeance.An increase in CO2

3

permeance from 0.18 to 0.34 m3 (STP)/m2 bar-h was observed when CNT loading was

4

increased from 0 to 1% with respect to PVA. However, at higher loading both CO2

5

permeance and CO2/N2 selectivity decreased. The carbon nanotubes were dispersed in PVA

6

by sonication. The sonicated membrane solution was then filtered to ensure that aggregated

7

particles are removed to get a stable suspension. The carbon nanotubes were intended to act

8

as nano spacers between the polymeric chains to produce gaps and increase water uptake and

9

gas permeance. However, at 1.5% CNT loading the nanoparticles started aggregating and

10

resulted in a decrease in CO2 permeance. The CNTs also affected the selectivity of the

11

membrane and resulted in a decreased CO2/N2 selectivity of 45 at 1.5% CNT loading. An

12

optimal CNT loading of 1% with respect to PVA is suggested based on highest CO2

13

permeance.

14

The separation performance of PVA membrane can be tailored by adding CNTs or by

15

adjusting the pH. The membrane containing 1% CNTs at different pH were tested to suggest

16

an optimal set of conditions based on highest possible separation performance. A thin film of

17

PVA (~0.8µm) containing 1% CNTs at pH 5, 10 and 12 was cast over porous PSf support.

18

The prepared membranes were tested for CO2/N2 separation performance under humidified

19

conditions and feed pressure of 1.2 bara with methane as a sweep at atmospheric pressure.

20

CO2 permeance and CO2/N2 selectivity are presented in figure 11.

21 100

0.6

80

2

CO2Permeance (m (STP)/m .bar.h)

0.7

3

60 0.4 40 0.3 20

0.2

0.1

0 4

22

6

8

10

pH

18

12

Selectivity (CO2/N2)

0.5

1

Figure 11 Effect of membrane pH on the CO2 permeance and CO2/N2 selectivity of membrane containing 1%

2

CNTs. (humidified 10% CO2 in N2 at 1.22bara feed gas and CH4 sweep at atmospheric pressure and operating

3

temperature of 20oC)

4

The experimental results presented in figure 10 shows that by adjusting the pH of PVA

5

membrane containing 1% CNTs from 5 to 12, the permeance of CO2 is increased from 0.34

6

to 0.44±0.02 m3 (STP)/m2 bar-h. This is a significant increase in permeance as compared to

7

PVA membrane without CNTs at pH 5. The CO2 permeance of membrane has increased by

8

144% just by adding a small quantity of CNTs and pH adjustment. The CO2/N2 selectivity

9

which decreased from 59 to 51 by adding CNTs is also now recovered to 60. This indicated

10

that a PVA membrane can be tailored by adding CNTs and pH adjustment to increase its gas

11

permeance while maintaining its selectivity.

12

Table 3. CO2 permeance and CO2/N2 selectivity of prepared PVA membranes.

Membrane

CO2 permeance [m3(stp)/m2-bar h]

CO2/N2 Selectivity

PVA pH 5

0.18±0.02

58±3

PVA pH 7

0.26±0.01

56±4

PVA pH 9

0.34±0.02

59±1

PVA pH 10

0.35±0.02

58±2

PVA pH 12

0.40±0.03

60±1

0.5% CNT pH 5

0.27±0.03

58±2

1% CNT pH 5

0.34±0.02

51±3

1.5% CNT pH 5

0.24±0.01

45±5

1% CNT pH 10

0.40±0.03

56±2

1% CNT pH 12

0.44±0.02

60±1

13 14

The effect of feed pressure on the CO2/N2 separation performance was also studied by

15

varying the feed pressure from 1.2 to 4.5 bara. The experimental results as presented in figure

16

11 compares the CO2 permeance of PVA membrane with and without 1% CNTs at pH 5 and

17

pH 12. These figures show that by increasing the pH the permeance of CO2 has increased

18

significantly but the CO2 permeance of PVA membrane at pH 12 declines rapidly as the

19

pressure is increased from 1.2 bara. This is a typical behavior observed in facilitated transport

20

membranes. By the addition of CNTs to the polymeric film this decline in permeance is

21

greatly reduced. The carbon nanotubes no only help improve membrane swelling and CO2

22

permeance but also maintain its performance at higher pressure. Similar effects were

23

observed by Deng and Hagg [34] for CO2/CH4 separation by CNT reinforced PVAm 19

1

membrane. A CNT reinforced PVA membrane at pH 12 is 200% more permeable as

2

compared to PVA membrane even at higher pressure. The CO2/N2 selectivity varied between

3

50 and 60 for all the tested pressures, thus showing no significant effect with respect to pH or

4

CNT loading.

% PVA 2% % PVA 2%+CNT0.5% % PVA 2%+CNT1.0% % PVA 2%+CNT1.5% % PVA 2%+CNT1.0% pH10 % PVA 2%+CNT1.0% pH12 % PVA 1%

0.6

2

CO2Permeance (m (STP)/m .bar.h)

0.7

3

0.5

0.4

0.3

0.2

0.1 1.0

5

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Pressure (bar)

6

Figure 12 Effect of feed pressure on CO2 permeance for membrane with and without CNTs at pH 5 and pH 12.

7

(Humidified 10% CO2 in N2 and CH4 sweep at atmospheric pressure and operating temperature of 20oC)

8 9

The effect of membrane thickness on CO2/ N2 separation performance was also studied by

10

casting 1% PVA membrane at pH 12 to produce 0.4~0.45µm thick membranes on PSF

11

support. An increase in CO2 permeance from 0.4 to 0.48 m3 (STP)/m2 bar-h was observed at

12

constant CO2/N2 selectivity around 60.

13

20

100

0.5

80

0.4 60 0.3 40 0.2 20

0.1

0.0

0 0

1

Selectivity (CO2/N2)

3

2

CO2Permeance (m (STP)/m .bar.h)

0.6

5

10

15

20

25

30

Time (Days)

2

Figure 13 CO 2 permeance of PVA membrane at pH 12 with 1-month operation (Humidified 10% CO2 in N2 at

3

1.22bara feed gas and CH4 sweep at atmospheric pressure and operating temperature of 20oC)

4

The casted membrane with PVA solution was tested for CO2 permeance for over a period of

5

1 month to study the stability of the membrane. The membrane with a film thickness of

6

0.4~0.45µm of PVA at pH 12 was tested at 1.2 bar feed pressure with 10% CO2 in N2

7

humidified conditions. The CO2/N2 separation performance of membrane over time is

8

presented in figure 13. The CO2 permeance was found to be 0.48 m3 (STP)/m2 bar-h. This

9

membrane was kept in humidified conditions and tested periodically over a period of 1 month

10

to monitor any change in CO2 permeance. After 1 month of testing the CO2 permeance was

11

found to be 0.46 m3 (STP)/m2 bar-h and CO2/N2 selectivity was around 60. This shows that

12

the membrane can be used for stable operation over a long period of time.

13

The developed membrane shows appreciable CO2/N2 separation performance under the

14

humidified conditions. To investigate the effect of humidity on the membrane separation

15

performance, experiments were conducted with mixed feed gas containing 10% CO2 in N2

16

at 1.2bara feed pressure and pure CH4 as a sweep with constant feed/sweep flow rates and

17

various humidity levels at 25oC. The effect of humidity on CO2 permeance and CO2/N2

18

selectivity of the membrane are presented in figure 14. For a PVA membrane without CNTs

19

or pH adjustment, a sharp increase in both CO2 permeance and CO2/N2 selectivity was

20

observed as the humidity level of feed gas increased from 80% relative humidity to 100%

21

humidity. This indicates that water plays a crucial role in the separation performance of PVA

22

membrane. The presence of water not only improves CO2 permeance of membrane but also

23

improves the CO2/N2 selectivity. This indicates that CO2 does not only transport as a gas

21

1

through the water swollen membrane but it also interacts with moisture in the membrane to

2

form complexes which help in selective transport of CO2 within the membrane.

0.45

100 90

PVA 2%

0.40

80

0.35

70

0.30

60

0.25

50

0.20

40

0.15

30

0.10

20

0.05

10

0.00

0 30

40

60

70

80

90

100

RH (%)

3 4

50

Selectivity (CO2/N2)

3

2

CO2Permeance (m (STP)/m .bar.h)

0.50

Figure 14: Effect of Relative humidity (RH%) on CO2 permeance and CO2/N2 selectivity of the membrane

5

without pH adjustment. (Feed gas containing 10% CO2 in N2 at 1.2 bara pressure and CH4 as sweep with

6

constant flow rate at atmospheric pressure and 25oC).

7 8

The swelling test indicated that water uptake of the membrane increases with increase in pH.

9

Similarly, the permeance results showed an increase in CO2 permeance and CO2/N2

10

selectivity with respect to pH. The PVA membrane with pH 12 was also tested for CO2/N2

11

separation performance at various humidity levels. The experimental results presented in

12

figure 15 indicate that both CO2 permeance and CO2/N2 selectivity increases significantly as

13

the relative humidity is increased over 70%. As evident from swelling tests that PVA

14

membrane with pH 12 is more hydrophilic as compared to PVA at pH 5, the increase in

15

hydrophilicity and water uptake is also evident in figure 15. At 70% RH the CO2 permeance

16

and CO2/N2 selectivity of PVA membrane at pH 12 are similar to the CO2 permeance and

17

CO2/N2 selectivity of PVA membrane at pH 5 at 85% RH. This indicates that by tailoring the

18

hydrophilicity and water uptake of membrane its separation performance can be improved

19

even at low humidity levels.

22

100

0.45

90

PVA 2% pH12

0.40

80

0.35

70

0.30

60

0.25

50

0.20

40

0.15

30

0.10

20

0.05

10

0.00

0 60

70

80

90

Selectivity (CO2/N2)

3

2

CO2Permeance (m (STP)/m .bar.h)

0.50

100

RH (%)

1 2

Figure 15: Effect of Relative humidity (RH%) on CO2 permeance and CO2/N2 selectivity of the membrane at

3

pH 12. (Feed gas containing 10% CO2 in N2 at 1.2 bara pressure and CH4 as a sweep with constant flow rate at

4

atmospheric pressure and 25o C).

5

The carbon nanotubes were added to PVA membrane to reduce membrane compaction and

6

increase water uptake. The membrane containing 1% CNTs at pH 12 indicated highest CO2

7

permeance and CO2/N2 selectivity at humidified conditions. The membrane was tested with a

8

mixed feed gas containing 10% CO2 in N2 as feed and CH4 as a sweep at a constant feed

9

pressure of 1.2bara and constant feed/sweep flow rates at 25oC. Relative humidity was varied

10

from 50 to 100% humidified feed gas. The CO2 permeance and CO2/N2 selectivity were

11

calculated and presented in figure 16. The experimental results indicate that at 60% RH CO2

12

permeance and CO2/N2 selectivity of the membrane is similar to the separation performance

13

of PVA membrane (pH12) at 70% RH and PVA membrane (pH 5) at 85% RH. This indicates

14

that by adjusting pH and adding CNTs as nano spacers the swelling behavior of membrane

15

can be tailored to improve both CO2 permeance and CO2/N2 selectivity. A membrane with

16

CNTs and pH 12 is more reliable for operation in conditions where the relative humidity may

17

change significantly.

18 19 23

100

PVA 2%+ CNT1% pH12

0.45

90

0.40

80

0.35

70

0.30

60

0.25

50

0.20

40

0.15

30

0.10

20

0.05

10

0.00

0 50

60

70

80

90

Selectivity (CO2/N2)

3

2

CO2Permeance (m (STP)/m .bar.h)

0.50

100

RH (%)

1 2

Figure 16: Effect of Relative humidity (RH%) on CO2 permeance and CO2/N2 selectivity of the membrane at

3

pH 12 and 1% CNTs. (Feed gas containing 10% CO2 in N2 at 1.2 bara pressure and CH4 as a sweep with

4

constant flow rate at atmospheric pressure and 25oC).

5

6

4.

7

A hydrophilic thin film composite membrane was tested and optimized for hosting carrier in

8

a facilitated transport membrane for CO2 capture at post-combustion humidified conditions.

9

A defect-free thin film composite membrane with a selective layer of PVA was developed

Conclusions

10

over porous support. The swelling degree of the physically crosslinked membrane was

11

tailored by adjusting the pH of casting solution from acidic to basic and addition of surface

12

modified multiwalled carbon nanotubes. Furthermore, empirical modeling of experimental

13

data using in-house built Modfit was conducted to determine equilibrium swelling degree.

14

The modeled results show that by adjusting pH and optimizing CNT concentration the

15

swelling degree of the polymer was increased from 154 to 256%.

16

The developed membranes were tested for CO2/N2 separation performance with simulated

17

flue gas containing 10% CO2 in N2. The experimental results indicate that CO2 permeance

18

has increased from 0.18 to 0.44 m3/m2-bar.h. By tailoring the membrane pH and optimizing 24

1

the CNT concentration, CO2 permeance has increased by 144% with average CO2/N2

2

selectivity around 60.

3

The thickness of the selective layer was reduced from 0.8µm to 0.4µm to improve the CO2

4

permeance from 0.44 to 0.48 m3/m2-bar.h at CO2/N2 selectivity of 60. The effect of humidity

5

in gas stream on CO2/N2 separation performance was also studied. Experimental results

6

indicate that by increasing the equilibrium swelling degree of the membrane, its

7

hydrophilicity has increased. Although the tailored membrane gives higher separation

8

performance at 100% relative humidity, it also shows better CO2/N2 separation performance

9

at lower humidity levels as compared to PVA membrane without CNTs and pH adjustments.

10 11

Acknowledgements

12

The Authors will like to acknowledge Faculty of Natural Science at Norwegian University of

13

Science and Technology (NTNU) and Electron Microscopy Lab at Institute of Oral Biology

14

in University of Oslo for their funding to conduct this work.

15

References

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

1. 2. 3. 4. 5. 6.

7. 8. 9. 10.

Wee, S.-L., C.-T. Tye, and S. Bhatia, Membrane separation process—Pervaporation through zeolite membrane. Separation and Purification Technology, 2008. 63 (3): p. 500-516. Kim, T.-J., et al., Separation performance of PVAm composite membrane for CO2capture at various pH levels. Journal of Membrane Science, 2013. 428(0): p. 218224. Wind, J.D., D.R. Paul, and W.J. Koros, Natural gas permeation in polyimide membranes. Journal of Membrane Science, 2004. 228(2): p. 227-236. Li, B., et al., Advances in CO2 capture technology: A patent review. Applied Energy, 2013. 102(0): p. 1439-1447. Roussanaly, S., et al., Membrane properties required for post-combustion CO2 capture at coal-fired power plants. Journal of Membrane Science, 2016. 511: p. 250264. Scholes, C.A., S.E. Kentish, and G.W. Stevens, Carbon dioxide separation through polymeric membrane systems for flue gas applications. Recent Patents on Chemical Engineering, 2008. 1 (1): p. 52-66. Robeson, L.M., Correlation of separation factor versus permeability for polymeric membranes. Journal of Membrane Science, 1991. 62 (2): p. 165-185. Rafiq, S., L. Deng, and M.-B. Hägg, Role of Facilitated Transport Membranes and Composite Membranes for Efficient CO2 Capture – A Review. ChemBioEng Reviews, 2016. 3(2): p. 68-85. Robeson, L.M., Polymer membranes for gas separation. Current Opinion in Solid State and Materials Science, 1999. 4(6): p. 549-552. Liao, J., et al., Fabrication of high-performance facilitated transport membranes for CO2 separation. Chemical Science, 2014. 5 (7): p. 2843-2849. 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

11.

12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

27. 28. 29. 30.

Kim, T.-J., B. Li, and M.-B. Hägg, Novel fixed-site–carrier polyvinylamine membrane for carbon dioxide capture. Journal of Polymer Science Part B: Polymer Physics, 2004. 42(23): p. 4326-4336. Freeman, B. and Y. Yampolskii, Membrane Gas Separation. 2011: Wiley. Grainger, D. and M.-B. Hägg, Techno-economic evaluation of a PVAm CO2-selective membrane in an IGCC power plant with CO2 capture. Fuel, 2008. 87(1): p. 14-24. Hussain, A. and M.-B. Hägg, A feasibility study of CO2 capture from flue gas by a facilitated transport membrane. Journal of Membrane Science, 2010. 359(1–2): p. 140-148. Brunetti, A., et al., Membrane technologies for CO2 separation. Journal of Membrane Science, 2010. 359(1–2): p. 115-125. Lie, J.A., et al., Optimization of a membrane process for CO2 capture in the steelmaking industry. International Journal of Greenhouse Gas Control, 2007. 1 (3): p. 309-317. Saeed, M. and L. Deng, CO2 facilitated transport membrane promoted by mimic enzyme. Journal of Membrane Science, 2015. Accepted. Kang, Y.S., et al., Solid-State Facilitated Transport Membranes for Separation of Olefins/Paraffins and Oxygen/Nitrogen, in Materials Science of Membranes for Gas and Vapor Separation. 2006, John Wiley & Sons, Ltd. p. 391-410. Ansaloni, L., et al., Facilitated transport membranes containing amino-functionalized multi-walled carbon nanotubes for High-pressure CO2 separations. Journal of Membrane Science, 2015. 490(0). Park, Y.-I. and K.-H. Lee, Preparation of water-swollen hydrogel membranes for gas separation. Journal of Applied Polymer Science, 2001. 80(10): p. 1785-1791. Zhang, Y., Z. Wang, and S. Wang, Novel fixed-carrier membranes for CO2 separation. Journal of Applied Polymer Science, 2002. 86(9): p. 2222-2226. Zhang, Y., Z. Wang, and S.C. Wang, Selective permeation of CO2 through new facilitated transport membranes. Desalination, 2002. 145(1–3): p. 385-388. Wang, Z., et al., CO2-facilitated transport through poly(N-vinyl-γ-sodium aminobutyrate-co-sodium acrylate)/polysulfone composite membranes. Journal of Applied Polymer Science, 2006. 100 (1): p. 275-282. Wang, Z., et al., Novel CO2 selectively permeating membranes containing PETEDA dendrimer. Journal of Membrane Science, 2007. 290 (1–2): p. 250-258. Ansaloni, L., et al., Influence of water vapor on the gas permeability of polymerized ionic liquids membranes. Journal of Membrane Science, 2015. 487 : p. 199-208. Matsuyama, H., M. Teramoto, and K. Iwai, Development of a new functional cationexchange membrane and its application to facilitated transport of CO2. Journal of Membrane Science, 1994. 93 (3): p. 237-244. Matsuyama, H., et al., Effects of membrane thickness and membrane preparation condition on facilitated transport of CO2 through ionomer membrane. Separation and Purification Technology, 1999. 17 (3): p. 235-241. Matsuyama, H., et al., Facilitated transport of CO2 through various ion exchange membranes prepared by plasma graft polymerization. Journal of Membrane Science, 1996. 117(1–2): p. 251-260. Yoshikawa, M., et al., Selective permeation of carbon dioxide through synthetic polymer membranes having pyridine moiety as a fixed carrier. Journal of Applied Polymer Science, 1988. 35 (1): p. 145-154. Matsuyama, H., K. Hirai, and M. Teramoto, Selective permeation of carbon dioxide through plasma polymerized membrane from diisopropylamine. Journal of Membrane Science, 1994. 92 (3): p. 257-265.

26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

31.

32. 33.

34. 35. 36. 37. 38. 39.

Matsuyama, H., M. Teramoto, and H. Sakakura, Selective permeation of CO2 through poly 2-(N,N-dimethyl)aminoethyl methacrylate membrane prepared by plasma-graft polymerization technique. Journal of Membrane Science, 1996. 114(2): p. 193-200. Yamaguchi, T., et al., Transport Properties of Carbon Dioxide through Amine Functionalized Carrier Membranes. Industrial & Engineering Chemistry Research, 1995. 34(11): p. 4071-4077. Deng, L. and M.-B. Hägg, Swelling behavior and gas permeation performance of PVAm/PVA blend FSC membrane. Journal of Membrane Science, 2010. 363 (1–2): p. 295-301. Deng, L. and M.-B. Hägg, Carbon nanotube reinforced PVAm/PVA blend FSC nanocomposite membrane for CO2/CH4 separation. International Journal of Greenhouse Gas Control, 2014. 26(0): p. 127-134. Takahiro, M. and O. Toshiya, Optical resonance in far-infrared spectra of multiwalled carbon nanotubes. Applied Physics Express, 2015. 8(5): p. 055101. Mohan, Y.M., et al., Swelling behavior of semi-interpenetrating polymer network hydrogels composed of poly(vinyl alcohol) and poly(acrylamide-co-Sodium methacrylate). Journal of Applied Polymer Science, 2005. 98(1): p. 302-314. Mulder, M., Basic Principles of Membrane Technology. 2nd ed. 2003, Netherland: Kluwer Academic Publishers inc. Drozdov, A.D., C.G. Sanporean, and J.d. Christiansen, Modeling the effects of temperature and pH on swelling of stimuli-responsive gels. European Polymer Journal, 2015. 73: p. 278-296. Rafiq, S., et al., Separation of CO2 from CH4 using polysulfone/polyimide silica nanocomposite membranes. Separation and Purification Technology, 2012. 90 : p. 162-172.

26

27