CO2 capture by polymeric membranes composed of

Open Phys. 2017; 15:662–670

Reaseach Article

Open Access

Ikuo Taniguchi*, Norihisa Wada, Kae Kinugasa, and Mitsuru Higa*

CO2 capture by polymeric membranes composed of hyper-branched polymers with dense poly(oxyethylene) comb and poly(amidoamine) https://doi.org/10.1515/phys-2017-0077 Received July 26, 2017; accepted September 2, 2017

1 Introduction

Abstract: Due to CO2 -philic nature of polyoxyethylene (POE), a dense POE comb structure was tethered onto PMMA backbone to develop CO2 separation membranes over N2 . The resulting hyper-branched polymers displayed preferential CO2 permeation. When the polymer thin layer was formed on a high gas permeable polydimethylsiloxane (PDMS) support by a spray-coating manner, the resulting thin film composite (TFC) membranes displayed very high CO2 permeability. However, the CO2 selectivity, which was the permeability ratio of CO2 over N2 , was moderate and lower than 50. To enhance the selectivity, poly(amidoamine) (PAMAM) was introduced to the hyper-branched polymers in the CO2 -selective layer of the TFC membranes. The CO2 selectivity increased from 47 to 90 with increasing PAMAM content to 40 wt%, and it was drastically enhanced to 350 with PAMAM content of 50 wt%. Differential scanning calorimetry (DSC) and laser microscope revealed formation of PAMAM-rich domain at the higher amine content, where CO2 could readily migrate in comparison to the other polymeric fractions.

Membrane separation is one of the most effective separation technologies in terms of chemical engineering. Transmembrane difference inconcentration or chemical potential drives the separation, and this method basically does not require additional energies, such as heating/cooling. As represented by reverse osmosis membranes for water purification, various membrane separation technologies have been established and commercialized. In gas separation by membranes, no phase change in the separation results in lower energy consumption, which can be the biggest advantage in comparison to other technologies, such as cryogenic separation processes. In addition, smaller foot print, instant operation, and lower maintenance cost due to no moving parts are beneficial for industrial applications. Membranes for air separation to enrich N2 , H2 removal from NH3 or hydrocarbons, vapor recovery, and acid gas removal from natural gas have been currently industrialized. Especially, cellulose acetate membranes has been used for three decades for the natural gas sweetening, in which CO2 is separated over CH4 [1]. CO2 separation membranes are current topic to capture CO2 at the mass emission sources such as coalfired plants and steel works in CO2 Capture and Storage (CCS) [2]. CCS was approved by one of the effective options to mitigate carbon emission by the Paris Agreement in 2015. Cost reduction is the key for implementation of CCS, and the capture dominates major part of the cost. Liquid amine scrubbing has been studied to capture CO2 [3] and used in the CCS demonstration projects. After CO2 in the source gas is separated by aqueous amine in an absorber, it is regenerated by heating in a desorber. The heating process results in large energy consumption even at thermal power stations. As described above, membrane separation could reduce the required energy to capture CO2 and hold potential to be an effective alternate capture technology. A number of CO2 separation membranes have been developed [4, 5]. Among them, polymeric membranes would be suitable for the CO2 capture due to synthetic

Keywords: CO2 capture, gas permeability, macrophase separation, membrane separation, polymeric membrane PACS: 66.30.je, 68.60.Wm

*Corresponding Author: Ikuo Taniguchi: International Institute for Carbon-Neutral Energy Research (WPI-I2 CNER), Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan, E-mail: [email protected] Norihisa Wada: Graduate School of Science and Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube 755-8611, Japan Kae Kinugasa: International Institute for Carbon-Neutral Energy Research (WPI-I2 CNER), Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan *Corresponding Author: Mitsuru Higa: Graduate School of Science and Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube 755-8611, Japan, E-mail: [email protected]

Open Access. © 2017 Ikuo Taniguchi et al., published by De Gruyter Open. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 License. Unauthenticated

Download Date | 11/13/17 1:14 AM

CO2 capture by polymeric membranes composed of hyper-branched polymers

and module preparation feasibility, large-scale production, and much lower cost in comparison to inorganic membranes. For membrane separation, both of high flux/permeability and selectivity are required, although they show a trade-off relation [6]. For example, thermallyrearranged polymers were recently developed and exhibited excellent CO2 separation performance over N2 and CH4 [7]. The CO2 permeability exceeded over 1,000 barrer (1 barrer: 10−10 cm3 (STP)cm/(cm2 ·s·cmHg)). However, the CO2 selectivity was moderate or lower than 50. According to the report, the selectivity should be greater than 50 for post-combustion CO2 capture by membranes [8]. In this research group, hyper-branched polymers with dense POE comb graft have been developed, and the polymeric membranes showed high CO2 separation properties over N2 [9, 10]. While the polymers formed a self-standing membrane with thickness above 80 µm by a solution casting method, they required a support when reducing the thickness below 80 µm to increase gas transport properties. The polymer thin film layer was thus formed on a commercial PDMS support (thickness: 60 µm) by spraycoating, the resulting TFC membranes showed CO2 permeability coefficient of 1,470 barrer with the selectivity of 25, where the thickness of the CO2 -selective polymer layer was 15 µm [10]. As well as the other CO2 separation membranes, the low selectivity limits the use for CO2 capture. Thus, improving the CO2 selectivity has been a challenge as well as elevating CO2 permeability. For polymeric or dense membranes, gas permeability can be explained by multiplication of solubility and diffusivity, while the selectivity is given by the ratio of permeabilities. To improve CO2 solubility to the membranes, amines have been used [11–15]. Here, amines act as a CO2 carrier to facilitate the transportation through the membrane. In this research, poly(amidoamine) (PAMAM) was introduced to improve CO2 selectivity of the polymeric membranes from hyper-branched polymers. Incorporation of the amine to the polymeric membranes has been extensively investigated, and the resulting membranes displayed excellent CO2 separation performance especially over H2 [16]. The preferential CO2 permeation was elucidated and explained by a CO2 -selective molecular gate mechanism [17]. Dissolved CO2 in the membranes interacted to the primary amines of PAMAM with the formation of quasi-crosslinks of PAMAM, which suppressed the permeation of other gaseous species, and as a result, the PAMAM containing membranes displayed quite high CO2 selectivity. Effect of PAMAM introduction on the CO2 separation performance was studied, and interplay between the membrane morphology and the gas transport properties were also investigated in this paper.

| 663

2 Experimental 2.1 Materials Monomers, methyl methacrylate (MMA), 4-chloromethyl styrene (Cl-St), POE methacrylate (POEM, M n : 500, average oxyethylene unit: 9), ethyl 2-bromoisobutyrate (EBI), and PAMAM (20 wt% solution in methanol) were obtained from Sigma-Aldrich (MO, USA). A radical scavenger, dibutylhydroxytoluene, in the monomers was removed by passing through a basic alumina column (MP Biomedicals, Eschwege, Germany). Copper(I) chloride (CuCl), 2,2’-azobisisobutyronitrile (AIBN), and 4,4’-dimethyl-2,2’dipyridyl (bpy) were purchased from Wako (Tokyo, Japan). AIBN was purified by recrystallization from methanol to obtain needle-like crystals before use. Other organic and inorganic chemicals were reagent grade and used without further purifications. A PDMS sheet with 60 µm thickness was available from As One (Cat.# 3-345-095, Osaka, Japan) and used as a high gas permeable support for TFC membranes.

2.2 Synthesis of macroinitiator The hyper-branched polymers were developed using the same procedures as previously reported [9, 10]. PMMA backbone as a macroinitiator was first prepared by free radical polymerization of MMA and Cl-St. In brief, 40 mL of MMA (380 mmol) and 0.54 mL of Cl-St (3.8 mmol) were dissolved in 200 mL of toluene, and the solution was applied onto a basic alumina column to remove the radical inhibitor. A 342 mg of AIBN (2.1 mmol) was added to the solution, where the monomer to initiator ratio was 90. After N2 bubbling for 30 min at room temperature, the reaction vessel was tightly sealed and kept at 70 °C for 1 h under stirring. The obtained macroinitiator was recovered by reprecipitation in methanol. The chemical structure and Cl-St content were determined by 1 H NMR in CDCl3 on a JEOL JNM-EX270 FT (Tokyo, Japan). The yield and Cl-St content were 74.7 % and 0.78 mol%, respectively. Cl-St was statistically distributed along the polymer chain, when the monomer inclusion was less than 2 mol% [9]. Molecular weight and the polydispersity of the macroinitiators were measured by size exclusion chromatography (SEC). The SEC system consisted of a Hitachi L-6000 pump (Hitachi, Tokyo, Japan) and a RI-2031 plus intelligent RI detector (JASCO, Tokyo, Japan). AKD-805 (Shodex,Tokyo, Japan) was used in a 860-CO column oven (JASCO). Tetrahydrofuran was used as an eluent at a flow rate of 0.8 mL/min and

Unauthenticated Download Date | 11/13/17 1:14 AM

664 | Ikuo Taniguchi et al. at 40°C, and the sample concentration was ca. 10 mg/mL. PMMA standards (Sigma-Aldrich) were used for molecular weight calibration. With the polymerization conditions, the macroinitiators had the weight-average molecular weight (M w ) of >50 kDa with polydispersity of 1.7-2.3 [9]. δH (270 MHz; CDCl3 ): 7.4-6.6 (Ar), 4.54 (ClCH2 ), 3.60 (OCH3 ), 2.1-0.8 ppm (CCH3 and polymer backbone)

sulting polymer solution, where the PAMAM fraction was 44 wt% in the solutes. The obtained solution was filtered with a membrane filter (nominal pore size: 0.2 µm) to remove dust particles and sprayed several times with an air gun-spray on the support. The obtained TFC membranes were dried under vacuum. The TFC membranes with various PAMAM contents were obtained, and the thickness of the active layer was 10-60 µm measured on a Mitutoyo digimatic micrometer (Tokyo, Japan).

2.3 Synthesis of hyper-branched polymer Dense POE comb was propagated from Cl moiety of the polymer backbone by atom transfer radical polymerization (ATRP) of POE methacrylate (POEM). ATRP is one of facile living polymerization techniques of various allyl monomers to obtain polymers with low molecular weight dispersity and can control chain length of polymers or degree of polymerization of monomers [9]. Hyper-branched polymers with well-defined chemical structure were thus obtained. For example, POEM (12.3 mL, 27.1 mmol) was dissolved in 30 mL of N-methylpyrrolidone (NMP) and the solution was passed through the basic alumina column. The macroinitiator above synthesized (3.0 g, 0.31 meq. of Cl-moiety), CuCl (0.31 mmol), and bpy (0.62 mmol) were added to the POEM solution. The obtained reaction suspension was kept at 90°C under stirring after N2 bubbling for 30 min at room temperature to remove O2 . After 48 h, polymeric fraction was collected by reprecipitation in petroleum ether/ethanol solution (5/1 by vol.) with trace HCl, and then re-dissolved in toluene. The transition metal on the active chain end was removed by silica gel chromatography. Finally, the hyper-branched polymer, PMMA-g-PPOEM, was obtained. The chemical structure and POEM introduction were confirmed by 1 H NMR in CDCl3 . The yield and POEM fraction were 61.9 % and 69.9 wt%, respectively. δH (600 MHz; CDCl3 ): 7.3-6.9 (Ar), 4.08 (COOCH2 CH2 ), 3.83.5 (COOCH3 and CH2 CH2 O), 3.38 (CH2 OCH3 ), 2.1-0.6 ppm (Ar-CH2 CH2 , CCH3 and polymer backbone)

2.4 Preparation of TFC membrane TFC membranes were fabricated by spray-coating of hyperbranched polymer and PAMAM solution onto a commercial PDMS support to form a CO2 -selective layer by the previously reported manner [10]. A 0.3 g of hyperbranched polymer (POEM: 69.9 wt%) was dissolved in acetone (7.0 mL) under ultrasonic irradiation in a water bath at 100 W and 37 Hz for 10 min (Sharp UT106, Osaka, Japan). A 1.5 mL of PAMAM solution (240 mg) was added to the re-

2.5 Gas permeability measurement Gas permeation properties of the composite membranes were examined by a vacuum time-lag method on a Tsukuba Rika Seiki K-315N-02 (Tokyo, Japan) at 1 atm and 35 °C as shown in Figure 1, where the effective membrane area A was 18.86 cm2 , the volume of reservoirs V 0 (upper) and V 1 (lower) were 22.87 and 22.93 cm3 , respectively. After evacuating the membranes in a cell at 35°C for at least 25 min to remove residual gaseous species, the target gas was charged in the upper reservoir at 1 atm and 35°C, and pressure increase dp2 /dt (cmHg/s) in the lower reservoir was then monitored as a function of time. The gas permeability P (cm3 (STP)cm/(cm2 s cmHg) is calculated from the following equation (Equation 1), and expressed in barrer. Here, T 1 and T 2 were 35°C, the thickness of the TFC membranes are l cm (l = lcom ), and the dp2 (blank)/dt was obtained from a leak test prior to gas permeation experiments. (︂ )︂ V1 V0 1 l + P = 273 A 273 + T1 273 + T2 76p1 (︂ )︂ dp2 dp2 (blank) (1) − dt dt A linear relationship between p2 and t can be found at a steady state, and the gas permeability coefficient P is determined from the slope. The permeability coefficient of the support alone was experimentally determined and 2,870 barrer for CO2 and 284 barrer for N2 under the operation conditions. The permeability coefficient of the selective layer was then calculated by using a series resistance model as follows (Equation 2) [18, 19]. l comp l l = PDMS + sel P comp P PDMS P sel

(2)

where P comp , P PDMS , and P sel , are permeability coefficients of the composite membranes, the PDMS support, and the CO2 -selective layer, respectively. Accordingly, lcomp , lPDMS , and lsel are the thicknesses of the composite membrane, PDMS, and selective layer, respectively.



In this vacuum time-lag method, the apparent diffusion coefficient Dapp will be determined from the time-lag θ, which is the period of time to reach the steady state and the intercept in x-axis of the plot p vs. t, with the following equation (Equation 3) [20]. However, the difference in time-lag θ could not be determined precisely due to the thinner active layer thickness on the gas permeation system used. l2 D app = (3) 6θ Gas permeance Q is defined as P divided by l (Equation 4) and expressed in GPU (1 GPU = 10−6 cm3 (STP)/(cm2 ·s·cmHg). The ideal separation factor of a membrane for gas A over gas B α(A/B) is given by the ratio of the permeability coefficients (or permeances) as shown in Equation 5. Q= α A/B =

P l

(4)

PA QA = PB QB

(5)

| 665

from −100 to 100°C under N2 atmosphere. The surface observation of polymeric films was carried out on an Olympus OLS4000 3D laser measuring microscope.

3 Results and discussion 3.1 Preparation of hyper-branched polymers Hyper-branched polymers of PMMA-g-PPOEM (PPOEM: polyPOEM) were prepared by two-step polymerization as shown in Figure2. PMMA macroinitiators were developed by free radical polymerization of MMA and Cl-St. Cl-St was statistically introduced into PMMA when the content was less than 2 mol% [9]. This means that average distance between Cl moieties is controlled, and the results were listed in Table 1. The average Cl moiety introduction was between 1.02 to 1.42 % of the monomer unit, and the yield was 6175 % after twice purification process by reprecipitation. O O

O

O O

MMA

O

O

Feed gas

V0

O

O Cl

POEM

O O

Cl

Reservoir

p1

Cl-St

Macroinitiator

Hyper-branched polymer PMMA-g-PPOEM

Membrane

O

Figure 2: Synthetic scheme of hyper-branched polymer

T1: 35°C Permeate gas

Thermostatic chamber 1

p2 Reservoir

V1

T2: 35°C

Vacuum Pump Thermostatic chamber 2

Figure 1: Experimental set up for gas permeation system

2.6 Membrane characterizations Thermal properties of the polymeric materials were studied by differential scanning calorimetry (DSC) on a Netzsch DSC 204 F1 Phoenix (Kanagawa, Japan). The samples were first cooled down to −100°C and equilibrated, and then the DSC spectra were recorded with a heating rate of 10°C/min

Then, the POEM propagation was initiated from the Cl moiety on the PMMA macroinitiator to form dense POE comb structure. POEs with molecular weight greater than 2,000 show crystallinity, which often suppress gas permeation [21]. Herein, the average oxyethylene unit of POEM was 9, and the POE is not considered crystalline. The degree of polymerization of POEM or the POE comb length was also readily controlled by ATRP. These synthetic procedures allowed preparing hyper-branched polymers with well-defined chemical structures. The results of PMMAg-PPOEM synthesis were summarized in Table 2. PMMAg-PPOEMs with three different POEM contents were successfully synthesized. The POEM fraction was from 34.4 to 69.9, where the average degree of polymerization DP of POEM was 8.1 to 45.6. The obtained hyper-branched polymers were collected by reprecipitation in petroleum ether/ethanol solution (5/1 by vol.). Although the mixture was the best poor solvent among examined for the hyperbranched polymers, POEM was not completely precipitated in the solution. Thus, the yield decreased from 78


666 | Ikuo Taniguchi et al. Table 1: Synthetic conditions of PMMA macroinitiator preparation by free radical polymerization and the results of polymerization

Run MMA 1 2 3

0.38 0.38 0.38

Feed/mol Cl-St 4.9 × 10−3 4.9 × 10−3 4.9 × 10−3

AIBN

Cl-St in feed /mol%

Cl-St in polymer* /mol%

Yield /%

2.1 × 10−3 2.1 × 10−3 2.1 × 10−3

1.29 1.29 1.29

1.02 1.42 1.29

74.7 61.2 66.6

Monomers were dissolved in 20 mL toluene. Reaction time was for 1 h at 70∘ C. * Determined by 1 H NMR to 62 % with increasing POEM fraction introduced in the product.

at -25 °C may result from phase separation, although it has not been clarified.

The hyper-branched polymers bearing dense POE comb show preferential CO2 permeation over N2 [9, 10], and thus would use for post-combustion CO2 capture. However, the CO2 permeability has to be improved for the use. One of the plausible approaches is reducing the membrane thickness. Although a self-standing membrane of the hyperbranched polymers with the thickness of >80 µm was readily available by casting the polymer solution, they require a support to be handled when reducing the membrane below the thickness [10]. The polymer solutions were spraycoated on a PDMS support to form a thin CO2 -selective layer, and the thickness of the active layer of the resulting TFC membranes was 10-60 µm. With PAMAM introduced membranes, incompatibility between POE and PAMAM resulted in macrophase separation [22, 23], and such membrane morphology would be a factor to characterize the gas transport properties. Herein, the same amount of the hyper-branched polymer was used at each sample. The phase separation was investigated by DSC and laser microscope. Figure 3 displays the DSC thermograms of PMMA-g-PPOEM (POE: 60 wt%) and mixture of the polymer and PAMAM. PAMAMg-PPOEM(PAMAM: 0 wt%, a) showed slightthermal transition at -60 °C, which was corresponded to the glass transition of POE. On the other hand, PAMAM containing polymers (PAMAM: 50 wt%, c) exhibited much larger transition at around -45 °C, which would indicate melting of the amine. This result suggested the presence of PAMAM-rich phase upon macrophase separation between the polymer matrix and PAMAM. On the contrary, the polymeric membrane with lower PAMAM inclusion (PAMAM: 40 wt%, b) did not show such apparent thermal transition, which implied that some portion of PAMAM became miscible to the polymer at the lower PAMAM inclusion. A faint transition

Exo./a.u.

3.2 Membrane formulation and morphology

a b c Temperature/°C

Figure 3: DSC thermograms of PAMAM introduced polymeric membranes from hyper-branched polymer (POEM: 60 wt%): (a) pristine PMMA-g-PPOEM: (b) PMMA-g-PPOEM with PAMAM (40 wt%): (c) PMMA-g-PPOEM with PAMAM (50 wt%):

Surface morphology of the polymeric membranes was studied by laser microscope to see the phase separationon the membrane surface as depicted in Figure 4. The surface of the lower PAMAM introduction (40 wt%) membrane was smoother than that of the higher amine loading membrane (50 wt%). With the higher amine loading membrane, bright and dark regions were more clearly confirmed on the membrane surface. The image in Figure 4b indicated the presence of PAMAM-rich domain upon macrophase separation, which was induced by the 10 wt% difference in the amine fraction. This was well consistent with that determined by DSC.

3.3 CO2 separation properties The gas permeation performance of the TFC membranes was determined by a vacuum time-lag method. This strategy is often used to examine gas diffusivity from the timelag until gas permeation reaches a steady state, in which



| 667

Table 2: Synthetic conditions of hyper-branched polymer preparation by ATRP and the results of polymerization

MI

*

MI-1 MI-2 MI-3

a

MI /g

Cl-St in polymer /mol%

CuCl /10−4 mol

bpy /10−4 mol

POEM /10−4 mol

Yield /%

DP of POEM**

POEM fraction** /wt%

61.9 3.00 3.5 6.4 271 45.6 1.02 1.42 3.00 3.6 7.3 66.5 10.7 286 1.29 3.00 5.2 132 78.1 10.1 8.1 Macroinitiator was dissolved in 30 mL NMP. Reaction time was for 48 h at 90°C. * Macroinitiator, ** Determined by 1 H NMR b

Figure 4: Surface images of PAMAM introduced polymeric membranes of hyper-branched polymer (POEM: 60 wt%) by laser microscope (bar: 500 µm). a: PMMA-g-PPOEM with PAMAM (40 wt%); b: PMMA-g-PPOEM with PAMAM (50 wt%)

pressure increase per unit time is constant [20]. However, in this research, the thickness of the CO2 -selective layer of the TFC membranes was too small to detect the time-lag precisely, and apparent diffusivity Dapp could not be determined by Equation 3. The gas transport properties were thus expressed only with permeability coefficient P. The gas permeation properties of various TFC membranes were summarized in Table 3, and effect of membrane thickness on the CO2 separation performance was drawn in Figure 5. The CO2 permeability of both the TFC membranes and the active layers was depended on the corresponding thicknesses, and thinner membranes seemed to give higher gas transport properties as shown in Figure 5. With higher POEM fraction (43.0 and 69.9 wt%), the CO2 permeability coefficient was greater than that of the membrane with lower POEM fraction (34.4 wt%). POEM is CO2 philic, and a discontinuous increase in CO2 permeability was confirmed at the POEM fraction of 40 wt% in our previous work [9, 10]. Above the POEM fraction, POE-rich phase was formed upon microphase separation between POE and the rest of the polymeric fraction, which would enhance the CO2 solubility to provide the higher gas permeability. In the case of the hyper-branched polymers with POEM fraction of 69.9 wt%, the TFC membrane with the thickness lcomp of 86 µm and PAMAM content of 20.0 wt%

69.9 43.0 34.4

exhibited the highest CO2 permeability P comp (CO2 ) of 454 barrer, and the CO2 selectivity over N2 α comp (CO2 /N2 ) was 69.1. On the other hand, the highest CO2 permeability coefficient of the active layer P sel (CO2 ) was 176 barrer (thickness: 37 µm and PAMAM content: 13.3 wt%) with 80.4 in the selectivity α sel (CO2 /N2 ). The obtained results were higher than those of the polymeric membranes without PAMAM incorporation. Effect of PAMAM inclusion on the gas transport properties was thendisplayed in Figure 6. With both of the TFC membranes and the CO2 -selective layer, PAMAM introduction successfully contributed to enhance the CO2 selectivity over N2 , in which the amine worked as a CO2 carrier in facilitated transportation manner [24, 25]. However, the amine loading most likely suppressed the gas permeation in these experimental conditions. The increased CO2 selectivity was resulted from slight decrease in the CO2 permeability relative to large drop in the N2 permeability with increase of the amine content in the CO2 -selective layer. Similar results of PAMAM containing membranes have been reported in the mixed gas experiment for CO2 separation over H2 [16, 17]. When the amine was immobilized in a crosslinked PEG network, the CO2 permeability of the resulting polymeric membranes did not change much with increase of the amine loading. On the contrary, the H2 permeability significantly dropped at higher amine loading (> 40 wt%) to give quite high CO2 selectivity over H2 greater than 1,000 at 5 kPa of CO2 partial pressure and 80 % relative humidity. As well as H2 , N2 is non-polar and would suffer from salting out effect by the polar amine at the higher amine loading, and as a result the solubility of N2 would decrease sharply in comparison to that of CO2 with increase of PAMAM content. Mechanism of preferential CO2 permeation was elucidated in the CO2 separation. PAMAM containing membranes got swollen under humidity, and a part of CO2 turned to bicarbonate in the membrane with drastic enhancement of the CO2 permeability. Bicarbonate formed under humidity was thus the major migrating species through the membrane [17].


668 | Ikuo Taniguchi et al. Table 3: Gas permeation properties of TFC membranes with different POEM and PAMAM fractions at 35 °C Run

POEM /wt%

PAMAM /wt%

lcomp /µm

P comp (CO2 ) /barrer

P comp (N2 ) /barrer

α comp

lsel /µm

P sel (CO2 ) /barrer

P sel (N2 ) /barrer

α sel

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 PDMS

34.4 34.4 34.4 43.0 43.0 69.9 69.9 69.9 69.9 69.9 69.9 69.9 69.9 69.9 69.9 -

0 33.3 33.3 50.0 50.0 13.3 13.3 20.0 20.0 20.0 40.0 50.0 50.0 50.0 50.0 -

87 70 87 118 121 90 97 86 88 110 92 100 110 115 118 60

363 178 127 161 135 369 419 464 387 297 289 269 224 249 161 2,870

8.33 3.86 3.68 1.32 0.38 4.22 5.66 6.57 8.47 3.33 3.34 2.08 2.08 0.70 1.32 284

43.6 46.1 34.4 122 355 87.4 74.0 69.1 45.7 89.2 86.5 129.3 108 356 122 10.1

27 10 27 58 61 30 37 26 28 50 32 40 50 55 58 -

123 26.9 40.7 81.5 69.7 135 176 154 136 143 108 114 106 125 81.5 -

2.64 0.56 1.16 0.65 0.19 1.42 2.19 2.02 2.75 1.52 1.17 0.84 0.95 0.34 0.65 -

46.8 48.1 35.2 125 364 94.7 80.4 76.4 49.3 93.9 91.9 136 112 372 125.3 -

1 barrer = 10−10 cm3 (STP)cm/(cm2 ·s·cmHg)

b

Thickness/µm

Permeability/barrer

α(CO2/N2)

α(CO2/N2)

Permeability/barrer

a

Thickness/µm

Figure 5: Effect of thickness of (a) TFC membranes and (b) CO2 -selective layer on gas transport properties with various POEM fractions: 34.4 (diamond), 43.0 (square), and 69.9 wt% (circle): filled, open, and red symbols denote CO2 permeability, N2 permeability, and selectivity, respectively.

However, the gas permeation was studied under dry conditions in this report, and CO2 would move as a carbamate form from feed to permeate side in the dry conditions. Although the CO2 solubility was improved by the addition of PAMAM, migration of carbamate through PAMAM network was much lower than that of bicarbonate formed under humidity, and thus the CO2 permeability was not increased much as expected. However, the CO2 selectivity over N2 could be enhanced by the PAMAM loading, and some membranes exhibited the selectivity over 350 with higher CO2 permeability than those without the amine. In gas transportation in dense polymers, permeability coefficient P is given by diffusivity coefficient D times solubility coefficient S as P=D×S [26]. Hence Equation 6: α(CO2 /N2 ) =

D(CO2 ) S(CO2 ) D(N2 ) S(N2 )

(6)

Here, kinetic diameters of CO2 and N2 are 3.3 and 3.6 Å, respectively, thus the diffusion selectivity D(CO2 )/D(N2 ) becomes greater than 1. In addition, POE of the hyperbranched polymers gives rise to CO2 solubility, which contributes the solution selectivity S(CO2 )/S(N2 ) >1. As a result, the TFC membranes showed preferential CO2 permeation over N2 . To enhance the CO2 permeation selectivity, PAMAM was incorporated into the CO2 -selective layer. As a result, the selectivity was successfully elevated due to further increase of the solution selectivity S(CO2 )/S(N2 ).

4 Conclusions A polymeric membrane from PMMA-g-PPOEM hyperbranched polymers showed preferential CO2 permeation over N2 , and the CO2 permeability was increased by ele-



| 669

b Permeability/barrer

PAMAM/wt%

α(CO2/N2)

α(CO2/N2)

Permeability/barrer

a

PAMAM/wt%

Figure 6: Gas transport properties as a function of PAMAM dendrimer inclusion: (a) TFC membranes and (b) CO2 -selective layer with various POEM fractions: 34.4 (diamond), 43.0 (square), and 69.9 wt% (circle): filled, open, and red symbols denote CO2 permeability, N2 permeability, and selectivity, respectively.

vating the POE fraction and reducing the membrane thickness. In comparison to the high CO2 permeability, the CO2 selectivity over N2 was moderate (< 50) and not high enough for effective CO2 capture, such as at thermal power stations. The CO2 selectivity was improved by the addition of PAMAM. The amine loading contributed to increase the CO2 solubility to the polymeric membranes. The improved CO2 separation performance of the TFC membranes and/or the CO2 -selective layer was strongly depended on the amine content. When the content exceeded 50 wt%, the significant increase of the CO2 selectivity was found and reached 350 at the operation conditions. At the higher loading, low compatibility between the amine and the hyper-branched polymers induced phase separation to provide PAMAM-rich phase, in which CO2 migrated preferentially, determined by DSC and laser microscope. In other words, formation of the amine-rich phase would be a key for elevating the CO2 selectivity. The CO2 permeability was not enhanced by the amine introduction, but would be done by reducing the thickness of the active layer as reported [10]. The obtained results would be beneficial to develop CO2 separation materials. Acknowledgement: The authors acknowledge Prof. Anne M. Mayes of Massachusetts Institute of Technology for invaluable support to develop the hyper-branched polymers. This research was partially supported by Grant-inAid for Scientific Research (C) Grant Number JP17899334 and the Advanced Carbon Technology Research and Development Program from Japan Science and Technology Agency. The International Institute for Carbon-Neutral Energy Research (WPI-I2 CNER) is supported by the World Premier International Research Center Initiative (WPI), MEXT, Japan.

References [1]

Schell W.J., Wensley C.G., Chen M.S.K., Venugopal K.G., Miller B.D., Stuart J.A., Recent advances in cellulosic membranes for gas separation and pervaporation, Gas Sep. Purif., 1989, 3, 162169 [2] Powell C.E., Qiao G.G., Polymeric CO2 /N2 gas separation membranes for the capture of carbon dioxide from power plant flue gases, J. Membr. Sci., 2006, 279, 1-49 [3] Sharma S.D., Azzi M., A critical review of existing strategies for emission control in the monoethanolamine-based carbon capture process and some recommendations for improved strategies, Fuel, 2014, 121, 178-188 [4] Yang H., Xu Z., Fan M., Gupta R., Slimane R.B., Bland A.E.,Wright I., Progress in carbon dioxide separation and capture: A review, J. Environ. Sci., 2008, 20, 14-27 [5] Wang S., Li X., Wu H., Tian Z., Xin Q., He G., Peng D., Chen S., Yin Y., Jiang Z., Guiver M.D., Advances in high permeability polymerbased membrane materials for CO2 separations, Energy Environ. Sci., 2016, 9, 1863-1890 [6] Robeson L.M., The upper bound revisited, J. Membr. Sci., 2008, 320, 390-400 [7] Park H.B., Han S.H., Jung C.H., Lee Y.M., Hill A.J., J. Membr. Sci., 2010, 359, 11-24. [8] Merkel T.C., Lin H., Wei X., Baker R., J. Membr. Sci., 2010, 359, 126-139 [9] Taniguchi I., Kinugasa K., Egashira S., Higa M., Preparation of well-defined hyper-branched polymers and the CO2 separation performance, J. Membr. Sci., 2016, 502, 124-132 [10] Taniguchi I., Wada N., Kinugasa K., Higa M., A strategy to enhance CO2 permeability of well-defined hyper-branched polymers with dense polyoxyethylene comb graft, J. Membr. Sci., 2017, 535, 239-247 [11] Matsuyama H., Terada A., Nakagawara T., Kitamura Y., Teramoto M., Facilitated transport of CO2 through polyethylenimine/ poly(vinylalcohol) blend membrane, J. Membr. Sci., 1999, 163, 221-227 [12] Kim T.-J., Li B., Hägg M.-B., Novel fixed-site-carrier polyvinylamine membrane for carbon dioxide capture, J. Polym. Sci., Part B: Polym. Phys., 2004, 42, 4326-4336


670 | Ikuo Taniguchi et al. [13] Kovvali A.S., Chen H., Sirkar K.K., Dendrimer membranes: a CO2 -selective molecular gate, J. Am. Chem. Soc., 2000, 122, 7594-7595 [14] Cai Y., Wang Z., Yi C., Bai Y., Wang J., Wang S., Gas transport property of polyallylamine-poly(vinylalcohol)/polysulfone composite membranes, J. Membr. Sci., 2008, 310, 184-196 [15] Zhao Y., Ho W.S.W., Steric hindrance effect on amine demonstrated in solid polymer membranes for CO2 transport, J. Membr. Sci., 2012, 415-416, 132–138 [16] Taniguchi I., Duan S., Kazama, S., Fujioka Y,Facile Fabrication of a Novel High Performance CO2 Separation Membrane: Immobilization of Poly(amidoamine) Dendrimers in Poly(ethylene glycol) Networks, J. Membr. Sci., 2008, 322, 277-280 [17] Taniguchi I., Urai, H., Kai, T., Duan, S., Kazama, S., A CO2 - selective molecular gate of Poly(amidoamine) dendrimer immobilized in a poly(ethylene glycol) network, J. Membr. Sci., 2013, 444, 96-100 [18] Paul D.R., Gas transport in homogeneous multicomponent polymers, J. Membr. Sci., 1984, 18, 75-86 [19] Rowe B.W., Freeman B.D., Paul D.R., Physical aging of ultrathin glassy polymer films tracked by gas permeability, Polymer, 2009, 50, 5565-5575

[20] Paul D.R., Effect of immobilizing adsorption on the diffusion time lag, J. Polym. Sci. Part A-2, Polym. Phys., 1969, 7, 181118188 [21] Liu S.L., Shao L., Chua M.L., Lau C.H., Wang H., Quan S., Recent progress in the design of advanced PEO-containing membranes for CO2 removal, Prog. Polym. Sci., 2013, 38, 1089-1120 [22] Taniguchi I., Kazama S., Jinnai H., Structural analysis of poly(amidoamine) dendrimer immobilized in cross-linked poly(ethylene glycol), J. Polym. Sci. B: Polym. Phys., 2012, 50, 1156-1164 [23] Taniguchi I., Duan S., Kai T.,Kazama S.,Jinnai H., Effect of phase-separated structure on CO2 separation performance of poly(amidoamine) dendrimer immobilized in a poly(ethylene glycol) network, J. Mater. Chem. A, 2013, 1, 14514-14523 [24] Cussler E.L., Aris R., Bhown A., On the limits of facilitated diffusion, J. Membr. Sci., 1989, 43, 149-164 [25] Zhang Y., Sunarso J., Liu S., Wang R., Current status and development of membranes for CO2 /CH4 separation: A review, Int. J. Greenh. Gas Control, 2013, 12, 84-107 [26] Matteucci S., Yamploskii Y., Freeman B.D., Pinnau I., Transport of Gases and Vapors in Glassy and Rubbery Polymers, In: Yamploskii Y., Pinnau I., Freeman B.D. (Eds.), Materials Science of Membranes for Gas and Vapor Separation,Wiley, Hoboken, NJ, 2006
