Effect of Carbonic Anhydrase on CO2 Separation Performance of Thin Poly(amidoamine) Dendrimer/Poly(ethylene glycol) Hybrid Membranes

The effect of carbonic anhydrase (CA) on the separation performance of thin poly(amidoamine) (PAMAM) dendrimer/poly(ethylene glycol) (PEG) hybrid membranes was investigated. CA, a type of enzyme, was used to promote CO2 hydration and dehydration reactions and to assess whether these reactions were the rate-limiting step in CO2 permeation through the membrane. The relationship between the membrane thickness and the CO2 permeance was evaluated in CO2/H2 or CO2/He separation using PAMAM/PEG hybrid membranes (thickness: 10–100 μm) with and without CA. Without CA, the CO2 permeance of PAMAM/PEG hybrid membranes was not inversely proportional to the membrane thickness. On the other hand, with CA, the CO2 permeance was inversely proportional to the membrane thickness. It was implied that, without CA, the rate-limiting step of CO2 transport was either the CO2 hydration reaction at the feed side or the CO2 dehydration reaction at the permeate side. On the other hand, with CA addition, the rate-limiting step of CO2 transport was diffusion, and CO2 permeance could be increased without sacrificing the selectivity by reducing membrane thickness. The effect of the position of CA (i.e., on the surface and/or reverse surface) on CO2 separation performance was investigated to evaluate which reaction was the rate-limiting step of CO2 permeation through the membrane. It was suggested that the rate-limiting step of CO2 permeation was CO2 dehydration reaction at the permeate side.


Introduction
CO 2 capture and storage (CCS) is widely accepted as an important option for mitigating climate change, and has been attracting worldwide attention [1]. CO 2 capture includes CO 2 separation from flue gas (post-combustion, CO 2 /N 2 ), CO 2 separation from natural gas (CO 2 /CH 4 ), and CO 2 separation from integrated gasification combined cycle (IGCC) processes (pre-combustion, CO 2 /H 2 ) [2,3]. For practical application of the CCS technology, cost-effective methods for CO 2 capture are required. Many studies have focused on the development of effective CO 2 capture and separation technologies. Membrane separation would be one of the most promising approaches among them in terms of technological and economic perspectives. Polymeric membranes [4,5], inorganic membranes [6,7], ionic liquid membranes [8], and facilitated transport membranes [9,10] were studied for CO 2 separation. Most of the CO 2 selective membranes were developed for post-combustion. If CO 2 selective membranes are used for pre-combustion (high pressure gas), H 2 can be kept at high pressure in the retentate side and directly fed to gas turbine without compression, and CO 2 can transport though the membrane without using vacuum pump in the permeate side, so it is energy and cost saving to apply a CO 2 selective Membranes 2019, 9, 167 2 of 7 membrane for pre-combustion. However, it is very difficult to separate CO 2 over H 2 , because the molecular size of CO 2 is greater than that of H 2 , and only limited numbers of CO 2 selective membranes were reported for this purpose [11][12][13].
It was reported that poly(amidoamine) (PAMAM) dendrimer showed excellent CO 2 /N 2 separation performance in a liquid immobilized membrane (ILM) under 200 kPa low pressure [14]. In our group, PAMAM dendrimer/crosslinked-polymer hybrid membranes, such as PAMAM dendrimer/poly(ethylene glycol) (PEG) hybrid membranes [11][12][13] have been developed by immobilizing PAMAM dendrimer into the cross-linked polymer matrix for use in high pressure CO 2 separation. We found that these membranes show high CO 2 /H 2 separation performance at pressurized conditions, and have the potential to be applied for pre-combustion. In our previous paper, we developed hybrid membranes composed of PAMAM dendrimer and polyethylene glycol dimethacrylate (PEGDMA) and a compatible cross-linker 4GMAP that enabled thickness less than 100 µm [13]. However, we found that CO 2 permeance (Q CO 2 ) was not inversely proportional to thickness, and CO 2 /H 2 separation performance was reduced from ca. 30 µm to ca. 10 µm by reducing the membrane thickness. The experimental results were explained by the facilitated transport theory [15]. The permeances of both CO 2 and H 2 increased with decreasing thickness. However, since the membrane thickness dependence of Q CO 2 was lower than that of H 2 , selectivity of CO 2 over H 2 decreased with decreasing the membrane thickness.
In this paper, we investigated the effect of enzyme on the CO 2 separation performance, in order to find the solution to obtain both high Q CO 2 and selectivity for the thin PAMAM/PEG hybrid membranes. Carbonic anhydrase (CA) is a well-known enzyme to promote CO 2 hydration and dehydration reaction (CO 2 + H 2 O ⇔ H 2 CO 3 ) [16,17]. CA is an efficient catalyst for CO 2 hydration and dehydration with a turn over number of 106 mol-CO 2 /(mol-CA s) at the maximum. The carbonic anhydrase active region is shown in Figure 1 [16]. The CO 2 hydration mechanism of carbonic anhydrase is shown in Figure 2 [16,17]. A carbon dioxide molecule is attached to the Zn 2+ active site to form a meta-stable complex. The complex is then attacked by a Lewis base (OH − ) to produce bicarbonates (HCO 3 − ).
In this two-step process, CO 2 is converted to HCO 3 − and the active site in the CA is left un-reacted.
Therefore, in this paper, in order to overcome the limitation of the CO 2 separation properties, the effect of CA on separation properties was investigated. As far as the authors know, this paper is the first to report the effects of CA on CO 2 separation performance of the facilitated transport membranes in detail, such as the relationship between Q CO 2 and membrane thickness, rate-limiting step of CO 2 permeation, etc.
Membranes 2019, 9, x 2 of 7 saving to apply a CO2 selective membrane for pre-combustion. However, it is very difficult to separate CO2 over H2, because the molecular size of CO2 is greater than that of H2, and only limited numbers of CO2 selective membranes were reported for this purpose [11][12][13]. It was reported that poly(amidoamine) (PAMAM) dendrimer showed excellent CO2/N2 separation performance in a liquid immobilized membrane (ILM) under 200 kPa low pressure [14]. In our group, PAMAM dendrimer/crosslinked-polymer hybrid membranes, such as PAMAM dendrimer/poly(ethylene glycol) (PEG) hybrid membranes [11][12][13] have been developed by immobilizing PAMAM dendrimer into the cross-linked polymer matrix for use in high pressure CO2 separation. We found that these membranes show high CO2/H2 separation performance at pressurized conditions, and have the potential to be applied for pre-combustion. In our previous paper, we developed hybrid membranes composed of PAMAM dendrimer and polyethylene glycol dimethacrylate (PEGDMA) and a compatible cross-linker 4GMAP that enabled thickness less than 100 μm [13]. However, we found that CO2 permeance (QCO₂) was not inversely proportional to thickness, and CO2/H2 separation performance was reduced from ca. 30 μm to ca. 10 μm by reducing the membrane thickness. The experimental results were explained by the facilitated transport theory [15]. The permeances of both CO2 and H2 increased with decreasing thickness. However, since the membrane thickness dependence of QCO₂ was lower than that of H2, selectivity of CO2 over H2 decreased with decreasing the membrane thickness.
In this paper, we investigated the effect of enzyme on the CO2 separation performance, in order to find the solution to obtain both high QCO₂ and selectivity for the thin PAMAM/PEG hybrid membranes. Carbonic anhydrase (CA) is a well-known enzyme to promote CO2 hydration and dehydration reaction (CO2 + H2O ⇔ H2CO3) [16,17]. CA is an efficient catalyst for CO2 hydration and dehydration with a turn over number of 106 mol-CO2/(mol-CA s) at the maximum. The carbonic anhydrase active region is shown in Figure 1 [16]. The CO2 hydration mechanism of carbonic anhydrase is shown in Figure 2 [16,17]. A carbon dioxide molecule is attached to the Zn 2+ active site to form a meta-stable complex. The complex is then attacked by a Lewis base (OH − ) to produce bicarbonates (HCO3 − ). In this two-step process, CO2 is converted to HCO3 -and the active site in the CA is left un-reacted. Therefore, in this paper, in order to overcome the limitation of the CO2 separation properties, the effect of CA on separation properties was investigated. As far as the authors know, this paper is the first to report the effects of CA on CO2 separation performance of the facilitated transport membranes in detail, such as the relationship between QCO₂ and membrane thickness, rate-limiting step of CO2 permeation, etc.

Membrane Preparation
A polymeric membrane was fabricated by photopolymerization of PEGDMA in the presence of PAMAM dendrimer in water. The composition of precursor solution is PAMAM (50 wt %), PEGDMA (42.5 wt %) and 4GMAP (7.5 wt %) in water.
A schematic diagram of membrane fabrication of thin PAMAM/PEGDMA/4GMAP with or without CA hybrid membranes is shown in Figure 3. The hybrid membrane was prepared by casting precursor solution on a quartz plate, followed by the UV curing at 312 nm UV for 1.5 min and transferred onto PES support membrane. The membrane thickness was controlled by sandwiching the reaction mixture between quartz plates with stainless steel spacers (10-100 μm in thick).

Membrane Preparation
A polymeric membrane was fabricated by photopolymerization of PEGDMA in the presence of PAMAM dendrimer in water. The composition of precursor solution is PAMAM (50 wt %), PEGDMA (42.5 wt %) and 4GMAP (7.5 wt %) in water.
A schematic diagram of membrane fabrication of thin PAMAM/PEGDMA/4GMAP with or without CA hybrid membranes is shown in Figure 3. The hybrid membrane was prepared by casting precursor solution on a quartz plate, followed by the UV curing at 312 nm UV for 1.5 min and transferred onto PES support membrane. The membrane thickness was controlled by sandwiching the reaction mixture between quartz plates with stainless steel spacers (10-100 µm in thick).

Membrane Preparation
A polymeric membrane was fabricated by photopolymerization of PEGDMA in the presence of PAMAM dendrimer in water. The composition of precursor solution is PAMAM (50 wt %), PEGDMA (42.5 wt %) and 4GMAP (7.5 wt %) in water.
A schematic diagram of membrane fabrication of thin PAMAM/PEGDMA/4GMAP with or without CA hybrid membranes is shown in Figure 3. The hybrid membrane was prepared by casting precursor solution on a quartz plate, followed by the UV curing at 312 nm UV for 1.5 min and transferred onto PES support membrane. The membrane thickness was controlled by sandwiching the reaction mixture between quartz plates with stainless steel spacers (10-100 μm in thick).  To investigate the effect of CA additive, composite membranes with the selectivity layer PAMAM/PEGDMA/4GMAP shown were prepared with the reverse surface, with CA coated on the surface of the PES support membrane by spraying in advance (Figure 3 (3)). A schematic diagram of CA addition by spray method of CA onto the membranes is shown in Figure 4. The membrane thickness is determined by a KeyenceVHX-1000 digital microscope (Tokyo, Japan) [14]. To investigate the effect of CA additive, composite membranes with the selectivity layer PAMAM/PEGDMA/4GMAP shown were prepared with the reverse surface, with CA coated on the surface of the PES support membrane by spraying in advance (Figure 3 (2)). A schematic diagram of CA addition by spray method of CA onto the membranes is shown in Figure 4. The membrane thickness is determined by a KeyenceVHX-1000 digital microscope (Tokyo, Japan) [14].

Gas Separation Experimental
A schematic diagram of the gas separation experiment setup is shown in [18]. A CO2/H2 or CO2/He (80/20 by vol.) gas mixture was humidified at 80% relative humidity and then fed to a flatsheet membrane cell at a flow rate of 100 mL/min. As we reported in our previous papers, our membrane needs relative humidity as high as 80% RH to show high separation performance [18]. The CO2 partial pressures of the feed side was 560 kPa (total feed pressure 700 kPa). Dry Ar was supplied at a flow rate of 10 mL/min to the permeate side of the cell as a sweep gas. The test operating temperatures were at 40 °C. The CO2 and He concentrations in both feed and permeate gas were measured by gas chromatography. Permeance, Q, and selectivity, CO2/H2 or CO2/He were calculated as expressed in [18]. A CO2/H2 (80/20 by volume) gas mixture was used for the gas separation experimental of Section 3.1. A CO2/He (80/20 by volume) gas mixture was used for the gas separation experimental of Sections 3.2 and 3.3. He was used instead of H2 for the safety reason. The relationship between QHe (permeance of He) and QH₂ (permeance of H2) was as follows: QHe ≈ 0.8 QH₂. On the other hand, QCO₂ was almost the same for both CO2/He and CO2/H2 separation.

Effect of Membrane Thickness on CO2 Permeance and CO2/H2 Selectivity
The effect of membrane thickness on CO2 permeance and CO2/H2 selectivity was studied at 560 kPa of CO2 partial pressure and 80% RH at 40 °C with hybrid membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 by wt %; thickness = 10-100 μm. It was found that thinner membranes gave higher CO2 permeation properties, as shown in Figure 5. However, CO2 permeance was not inversely proportional to membrane thickness (QCO₂ ∝ L −0.62 ). On the other hand, H2 permeance was inversely proportional to membrane thickness (QH₂ ∝ L −0.95 ). As a result, selectivity of CO2 over H2 decreased as the membrane thickness decreased.
The amino group contributes to transport of CO2 though membrane as a bicarbonate ion (HCO3 − ) in the wet membrane, while H2 is only transported by a solution-diffusion mechanism [16]. The ratelimiting step of CO2 permeation was the reaction from CO2 to a bicarbonate ion (HCO3 − ) at the feed side, or the reaction from HCO3 − to CO2 at the permeate side. Since the membrane thickness dependence of QCO₂ is lower than that of H2, selectivity of CO2 over H2 decreased as the membrane thickness decreased. The experimental results were explained by the facilitated transport theory

Gas Separation Experimental
A schematic diagram of the gas separation experiment setup is shown in [18]. A CO 2 /H 2 or CO 2 /He (80/20 by vol.) gas mixture was humidified at 80% relative humidity and then fed to a flat-sheet membrane cell at a flow rate of 100 mL/min. As we reported in our previous papers, our membrane needs relative humidity as high as 80% RH to show high separation performance [18]. The CO 2 partial pressures of the feed side was 560 kPa (total feed pressure 700 kPa). Dry Ar was supplied at a flow rate of 10 mL/min to the permeate side of the cell as a sweep gas. The test operating temperatures were at 40 • C. The CO 2 and He concentrations in both feed and permeate gas were measured by gas chromatography. Permeance, Q, and selectivity, CO 2 /H 2 or CO 2 /He were calculated as expressed in [18]. A CO 2 /H 2 (80/20 by volume) gas mixture was used for the gas separation experimental of Section 3.1. A CO 2 /He (80/20 by volume) gas mixture was used for the gas separation experimental of Sections 3.2 and 3.3. He was used instead of H 2 for the safety reason. The relationship between Q He (permeance of He) and Q H 2 (permeance of H 2 ) was as follows: Q He ≈ 0.8 Q H 2 . On the other hand, Q CO 2 was almost the same for both CO 2 /He and CO 2 /H 2 separation.

Effect of Membrane Thickness on CO 2 Permeance and CO 2 /H 2 Selectivity
The effect of membrane thickness on CO 2 permeance and CO 2 /H 2 selectivity was studied at 560 kPa of CO 2 partial pressure and 80% RH at 40 • C with hybrid membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 by wt %; thickness = 10-100 µm. It was found that thinner membranes gave higher CO 2 permeation properties, as shown in Figure 5. However, CO 2 permeance was not inversely proportional to membrane thickness (Q CO 2 ∝ L −0.62 ). On the other hand, H 2 permeance was inversely proportional to membrane thickness (Q H 2 ∝ L −0.95 ). As a result, selectivity of CO 2 over H 2 decreased as the membrane thickness decreased.
The amino group contributes to transport of CO 2 though membrane as a bicarbonate ion (HCO 3 − ) in the wet membrane, while H 2 is only transported by a solution-diffusion mechanism [16].
The rate-limiting step of CO 2 permeation was the reaction from CO 2 to a bicarbonate ion (HCO 3 − ) at the feed side, or the reaction from HCO 3 − to CO 2 at the permeate side. Since the membrane thickness dependence of Q CO 2 is lower than that of H 2 , selectivity of CO 2 over H 2 decreased as the membrane thickness decreased. The experimental results were explained by the facilitated transport theory [15,16]. To obtain high Q CO 2 and selectivity, CA was added into CO 2 carrier (PAMAM) to obtain much higher reactivity with CO 2 in next section.

Effect of CA Addition on the CO2 Separation Properties
The effect of CA addition and membrane thickness on CO2 permeance and CO2/H2 selectivity was studied at 560 kPa of CO2 partial pressure and 80% RH at 40 °C with hybrid membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 by weight ratio with addition 1 wt % CA (membrane thickness = 10-60 μm). As can be seen from Figure 6, CO2 permeance was inversely proportional to membrane thickness (QCO₂ ∝ L −0.98 ). He permeance was also inversely proportional to membrane thickness (QHe ∝ L −0.94 ). As a result, CO2/He selectivity kept constant. It was suggested that CA addition enhanced the reaction rate of CO2 hydration at the feed side and the dehydration at the permeate side, and that the rate-limiting step of CO2 transport rate became diffusion.
CO2 permeace of as high as 2.47 × 10 −11 m 3 (STP)/(m 2 s Pa) accompanied with a CO2/He selectivity of 26.8 was achieved by the membrane with CA addition, ca. 55 μm thick. The enhancement in CO2 permeace is 270% compared with the membrane without CA addition (QCO₂: 6.71 × 10 −12 m 3 (STP)/(m 2 s Pa), ca. 55 μm thick). CO2 permeace of as high as 1.08 × 10 −10 m 3 (STP)/(m 2 s Pa) accompanied with a CO2/He selectivity of 28.7 was achieved by the membrane with CA addition, ca. 15 μm thick. The enhancements in CO2 permeace is 490% compared with the membrane without CA addition (QCO₂: 1.84 × 10 −11 m 3 (STP)/(m 2 s Pa), ca. 15 μm thick). It was indicated that CA addition was effective to break though the limitation of CO2 permeation of the thin facilitated transport membrane.

Effect of CA Addition on the CO 2 Separation Properties
The effect of CA addition and membrane thickness on CO 2 permeance and CO 2 /H 2 selectivity was studied at 560 kPa of CO 2 partial pressure and 80% RH at 40 • C with hybrid membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 by weight ratio with addition 1 wt % CA (membrane thickness = 10-60 µm). As can be seen from Figure 6, CO 2 permeance was inversely proportional to membrane thickness (Q CO 2 ∝ L −0.98 ). He permeance was also inversely proportional to membrane thickness (Q He ∝ L −0.94 ). As a result, CO 2 /He selectivity kept constant. It was suggested that CA addition enhanced the reaction rate of CO 2 hydration at the feed side and the dehydration at the permeate side, and that the rate-limiting step of CO 2 transport rate became diffusion. CO 2 permeace of as high as 2.47 × 10 −11 m 3 (STP)/(m 2 s Pa) accompanied with a CO 2 /He selectivity of 26.8 was achieved by the membrane with CA addition, ca. 55 µm thick. The enhancement in CO 2 permeace is 270% compared with the membrane without CA addition (Q CO 2 : 6.71 × 10 −12 m 3 (STP)/(m 2 s Pa), ca. 55 µm thick). CO 2 permeace of as high as 1.08 × 10 −10 m 3 (STP)/(m 2 s Pa) accompanied with a CO 2 /He selectivity of 28.7 was achieved by the membrane with CA addition, ca. 15 µm thick. The enhancements in CO 2 permeace is 490% compared with the membrane without CA addition (Q CO 2 : 1.84 × 10 −11 m 3 (STP)/(m 2 s Pa), ca. 15 µm thick). It was indicated that CA addition was effective to break though the limitation of CO 2 permeation of the thin facilitated transport membrane.

Effect of CA Addition on the CO2 Separation Properties
The effect of CA addition and membrane thickness on CO2 permeance and CO2/H2 selectivity was studied at 560 kPa of CO2 partial pressure and 80% RH at 40 °C with hybrid membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 by weight ratio with addition 1 wt % CA (membrane thickness = 10-60 μm). As can be seen from Figure 6, CO2 permeance was inversely proportional to membrane thickness (QCO₂ ∝ L −0.98 ). He permeance was also inversely proportional to membrane thickness (QHe ∝ L −0.94 ). As a result, CO2/He selectivity kept constant. It was suggested that CA addition enhanced the reaction rate of CO2 hydration at the feed side and the dehydration at the permeate side, and that the rate-limiting step of CO2 transport rate became diffusion.
CO2 permeace of as high as 2.47 × 10 −11 m 3 (STP)/(m 2 s Pa) accompanied with a CO2/He selectivity of 26.8 was achieved by the membrane with CA addition, ca. 55 μm thick. The enhancement in CO2 permeace is 270% compared with the membrane without CA addition (QCO₂: 6.71 × 10 −12 m 3 (STP)/(m 2 s Pa), ca. 55 μm thick). CO2 permeace of as high as 1.08 × 10 −10 m 3 (STP)/(m 2 s Pa) accompanied with a CO2/He selectivity of 28.7 was achieved by the membrane with CA addition, ca. 15 μm thick. The enhancements in CO2 permeace is 490% compared with the membrane without CA addition (QCO₂: 1.84 × 10 −11 m 3 (STP)/(m 2 s Pa), ca. 15 μm thick). It was indicated that CA addition was effective to break though the limitation of CO2 permeation of the thin facilitated transport membrane. Figure 6. Effect of CA addition on the CO2 separation properties of PAMAM/PEG hybrid membranes. Figure 6. Effect of CA addition on the CO 2 separation properties of PAMAM/PEG hybrid membranes.

Effect of Position of CA on CO 2 Separation Performance
The effect of position of CA on CO 2 separation performance was studied at 560kPa of CO 2 partial pressure and 80% RH at 40 • C, using hybrid composite membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 (weight %) with CA at different positions: (1) without CA; (2) S: surface; (3) RS: reverse surface; (4) S/RS. The thicknesses of these membranes were ca. 20 µm. The results are shown in Figure 7. CO 2 permeance of the membrane with CA at S position was not higher than that of membrane without CA. On the other hand, the membranes with CA at RS and S/RS positions showed CO 2 permeance and selectivity around twice as high as that of the membrane without CA or the membrane with CA at S position. From these results, it could be seen that the existence of CA at the RS position was more important than the S position. RS position is the place where CO 2 dehydration occurs at the permeate side, and the S position is the place where CO 2 hydration occurs at the feed side. Therefore, it was suggested that the rate-limiting step of CO 2 permeation was CO 2 dehydration reaction at the permeate side. These findings will be useful in the development of the facilitated transport membrane with high CO 2 separation properties. Detailed research on the CO 2 permeation mechanism is ongoing.

Effect of Position of CA on CO2 Separation Performance
The effect of position of CA on CO2 separation performance was studied at 560kPa of CO2 partial pressure and 80% RH at 40 °C, using hybrid composite membranes of PAMAM/PEGDMA/4GMAP = 50/42.5/7.5 (weight %) with CA at different positions: (1) without CA; (2) S: surface; (3) RS: reverse surface; (4) S/RS. The thicknesses of these membranes were ca. 20 μm. The results are shown in Figure  7. CO2 permeance of the membrane with CA at S position was not higher than that of membrane without CA. On the other hand, the membranes with CA at RS and S/RS positions showed CO2 permeance and selectivity around twice as high as that of the membrane without CA or the membrane with CA at S position. From these results, it could be seen that the existence of CA at the RS position was more important than the S position. RS position is the place where CO2 dehydration occurs at the permeate side, and the S position is the place where CO2 hydration occurs at the feed side. Therefore, it was suggested that the rate-limiting step of CO2 permeation was CO2 dehydration reaction at the permeate side. These findings will be useful in the development of the facilitated transport membrane with high CO2 separation properties. Detailed research on the CO2 permeation mechanism is ongoing.

Conclusions
Effect of CA enzyme on CO2 separation performance of thin poly(amidoamine) dendrimer/poly(ethylene glycol) hybrid membranes was investigated by examining the relationship between membrane thickness and CO2 permeance using the membranes with or without CA. CO2 permeance was inversely proportional to membrane thickness with CA, and not proportional to membrane thickness without CA. The addition of CA exhibited significantly enhanced CO2 separation performances. The membrane with CA at the permeate side showed higher CO2 separation performance than that of the membrane with CA at the feed side. Therefore, it was indicated that the rate-limiting step of CO2 permeation was CO2 dehydration reaction at the permeate side. These findings will be useful in the development of the CO2 facilitated transport membrane with high CO2 separation properties.
Author Contributions: S.D. participated in its design, methodology, validation and formal analysis. S.D. carried out writing-original draft preparation and writing-review. T.K. participated in its design and helped to draft and writing-review the manuscript. S.N. conceptualized and supervised project administration.
Funding: This article is based on results obtained from a project commissioned by the Ministry of Economy, Trade and Industry (METI), Japan.

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
Effect of CA enzyme on CO 2 separation performance of thin poly(amidoamine) dendrimer/ poly(ethylene glycol) hybrid membranes was investigated by examining the relationship between membrane thickness and CO 2 permeance using the membranes with or without CA. CO 2 permeance was inversely proportional to membrane thickness with CA, and not proportional to membrane thickness without CA. The addition of CA exhibited significantly enhanced CO 2 separation performances. The membrane with CA at the permeate side showed higher CO 2 separation performance than that of the membrane with CA at the feed side. Therefore, it was indicated that the rate-limiting step of CO 2 permeation was CO 2 dehydration reaction at the permeate side. These findings will be useful in the development of the CO 2 facilitated transport membrane with high CO 2 separation properties.
Author Contributions: S.D. participated in its design, methodology, validation and formal analysis. S.D. carried out writing-original draft preparation and writing-review. T.K. participated in its design and helped to draft and writing-review the manuscript. S.-i.N. conceptualized and supervised project administration.
Funding: This article is based on results obtained from a project commissioned by the Ministry of Economy, Trade and Industry (METI), Japan.