Scholarly article on topic 'CO2/CH4 Mixed Gas Separation Using Carbon Hollow Fiber Membranes'

CO2/CH4 Mixed Gas Separation Using Carbon Hollow Fiber Membranes Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
OECD Field of science
Keywords
{"carbon membrane" / "hollow fiber" / "mixed gas separation" / "poly(phenylene oxide)"}

Abstract of research paper on Materials engineering, author of scientific article — Miki Yoshimune, Kenji Haraya

Abstract In this study, the permeation properties of single and binary CO2/CH4 mixture were investigated using a module of carbon hollow fiber membranes derived from sulfonated poly(phenylene oxide) (SPPO). SPPO carbon membrane module had a sharp pore-size distribution in the range of 0.35-0.4nm, and showed high CO2/CH4 ideal selectivity for both single and binary gas separation. The effects of permeation temperature, total feed pressure and CO2 concentration in the feed on separation performances of the carbon membrane module were also investigated. It was found that the CO2/CH4 ideal selectivity decreased slightly with the increasing permeation temperature, total feed pressure, and CO2 concentration in the feed.

Academic research paper on topic "CO2/CH4 Mixed Gas Separation Using Carbon Hollow Fiber Membranes"

Available online at www.sciencedirect.com

SciVerse ScienceDirect

Energy Procedia 37 (2013) 1109 - 1116

GHGT-11

CO2/CH4 mixed gas separation using carbon hollow fiber

membranes

Miki Yoshimune*, Kenji Haraya

National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, 305-8565, Japan

Abstract

In this study, the permeation properties of single and binary CO2/CH4 mixture were investigated using a module of carbon hollow fiber membranes derived from sulfonated poly(phenylene oxide) (SPPO). SPPO carbon membrane module had a sharp pore-size distribution in the range of 0.35-0.4 nm, and showed high CO2/CH4 ideal selectivity for both single and binary gas separation. The effects of permeation temperature, total feed pressure and CO2 concentration in the feed on separation performances of the carbon membrane module were also investigated. It was found that the CO2/CH4 ideal selectivity decreased slightly with the increasing permeation temperature, total feed pressure, and CO2 concentration in the feed.

© 2013 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of GHGT

Keywords: carbon membrane ; hollow fiber ; mixed gas separation ; poly(phenylene oxide)

1. Introduction

In recent times, biogas has become an important renewable energy source and is considered a strategy for greenhouse gas reduction. Typical biogas mainly contains about 60 vol.% CH4 and 40 vol.% CO2; thus, effective separation of CO2/CH4 is necessary for biogas processing. Many membrane separation techniques using CO2 selective polymeric membranes, inorganic membranes, facilitated transport membranes, ionic liquid membranes and mixed matrix membranes have already been investigated. Carbon membranes have also exhibited attractive characteristics such as excellent shape selectivity and high chemical stability, required for CO2/CH4 separation [1].

Carbon membranes are mainly prepared by the pyrolysis of aromatic polymers such as polyfurfuryl alcohol, polyimide, phenolic resin, etc. Several factors including precursor selection, membrane preparation and pyrolysis process are well recognized as important steps to enhance the separation performance of carbon membranes. These factors immensely affect the pore structure of the resulting carbon membrane [2]. Till date, several researchers have studied the CO2/CH4 mixed gas separation by

1876-6102 © 2013 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of GHGT doi: 10.1016/j.egypro.2013.05.208

using various carbon membranes. For example, Sedigh et al. [3] used polyfurfuryl alcohol as a precursor for the preparation of supported carbon films. The membranes were tested by using equimolar binary mixtures of CO2 and CH4, and a quaternary mixture of CO2, CO, H2, and CH4. The separation factors for CO2/CH4, ranging between 34 and 37, were obtained for the binary and quaternary mixtures. Ogawa et al. [4] used polyamic acid as a precursor for the preparation of carbon hollow fiber membranes. The membranes showed as high as 310 CO2/CH4 selectivity when 65 mol% of CO2/CH4 gas was fed at 25 °C. Tin et al. [5] reported 50 mol% of CO2/CH4 gas separation, by using carbon molecular sieve membranes from P84 polyimide. This carbon membrane, pyrolyzed at 650 °C, showed CO2 permeability of 500 Barrer and selectivity of 89 at 35 °C. Anderson et al. [6] also made supported nanoporous carbon membranes from polyfurfuryl alcohol (PFA) for the separation of CO2/CH4. The permeation of mixed CO2/CH4 gas (10/90 vol.%) was evaluated across a range of operating temperatures from 35 to 200 °C. The CO2 selectivity of the membranes, pyrolyzed at 550 °C, decreased from 7.5 to 2.8 with increasing operating temperature.

We investigated the carbon hollow fiber membranes, derived from poly(phenylene oxide) (PPO) and its derivatives with excellent performance for gas separation such as O2/N2 and CO2/CH4 [7-9]. Recently, the development of flexible carbon hollow fiber membranes was successfully achieved using sulfonated PPO (SPPO) as a precursor polymer [10]. SPPO carbon hollow fiber membrane, pyrolyzed at 600 °C showed not only good CO2/CH4 selectivity, but also better mechanical stability for preparation of a membrane module. The aim of the present study was to examine the CO2/CH4 mixed gas separation performance using a small module of carbon hollow fibers. The effects of operation temperature, feed pressure and CO2 concentration in the feed on the CO2/CH4 separation performance of the membrane module were examined.

2. Experimental

2.1. Membrane preparation

The carbon hollow fiber membranes were prepared by the pyrolysis of SPPO polymeric hollow fibers as a precursor membrane. The hollow fiber precursor membranes were prepared by dry/wet spinning process using SPPO as a precursor polymer. Then the precursor membrane was pyrolyzed at 600 °C for one hour under vacuum at a heating rate of 10 °C min 1 to obtain the carbon membrane. The details of membrane preparation and characterization have already been reported in our previous work [10]. The membrane module was prepared by sealing a bunch of carbon hollow fibers with epoxy resin in a polyethersulfone tube holder. The holder size was 18 cm in length and 2 cm in diameter. The obtained membrane module contained 195 carbon hollow fibers and its effective membrane area was 259 cm2. The module was fixed into the pressure tight stainless case with Viton O-rings at the point of use. The picture of the membrane module is shown in Fig. 1.

2.2. Gas permeation test

A conventional time-lag method was used to measure the single and binary gas permeation properties. The experimental setup of the time-lag system is illustrated in Fig. 2. Pure and mixed gas at predetermined concentrations in gas cylinders was used. The total feed pressure was controlled by the back pressure regulator (BPR). Prior to each measurement, both the feed and permeate sides of the membrane module were evacuated to eliminate residual gases and impurities. In the single gas system, H2 (0.289 nm), CO2 (0.33 nm), O2 (0.346 nm), N2 (0.364 nm), and CH4 (0.38 nm) were tested as penetrants at 90, 50 and 25 °C under a pressure difference of 0.1 MPa. For CO2 and CH4 gas permeation test, the feed pressure was varied from 0.1 MPa to 0.99 MPa while the permeate side of the membrane was

evacuated. The gas permeance was expressed in SI units as mol m 2 s 1 Pa where 1 mol m 2 s 1 Pa 1 = 2.99 x 103 cm3 (STP) cm 2 s 1 cm Hg 1. The ideal selectivity was determined by considering the ratio of permeances for selected gas pairs.

The mixed gas separation experiments were conducted by using the binary gas mixtures (20, 40, 60 and 80 vol.% CO2/CH4) at 90, 50 and 25 °C. Gas compositions of feed, permeate and retentate stream were determined on line through the sampling rotary valve (SRV) by gas chromatography, equipped with a thermal conductivity detector (TCD). The membrane module was placed in the constant temperature bath (CTB1) as the counter-current configuration. The feed gas was admitted into the bore-side of the module, and the permeate gas was evacuated by a vacuum pump (VP) from the shell-side of the membranes.

Fig. 1. A picture of the membrane module of SPPO carbon hollow fibers.

AV1-5: Air actuate valve, BPR: Back pressure regulator, CTB1-2: Constant temperature bath, CVC: Constant volume cylinder, GC: Gas chromatography, MFC: Mass flow controller, P1-3: Pressure sensor, SC: Sample cylinder, SL: Sampling loop, SRV: Sampling rotary valve, TWV1-3: Three way valve, V1-5: Stop valve, VP: Vacuum pump

Fig. 2. Schematic diagram of experimental set-up for the gas permeation test.

3. Results and discussion

3.1. Single gas permeation

Fig. 3 shows the plots of permeances against the kinetic diameters of gas molecules for the SPPO carbon membrane module at 90, 50 and 25 °C under a pressure difference of 0.1 MPa. The gas permeances increased with higher operation temperature, and sharply decreased with the increasing size of penetrate molecule. This indicated that the obtained membrane module had a sharp pore-size distribution in the range of 0.35-0.4 nm, and the gas transport mechanism was mainly based on the molecular sieving. The ideal selectivities of selected gas pairs are summarized in Table 1. Higher ideal selectivities were obtained for smaller gases such as H2 and CO2 over larger molecules due to the sieving effect, while increased temperature lowered the selectivities of all gas pairs. The ideal selectivities of CO2/CH4 were 47, 93 and 197 at 90, 50 and 25 °C, respectively, a very promising result for the separation of CO2 and CH4. The CO2 and CH4 activation energies of the membrane module were evaluated to be 8.6 and 27.4 kJ mol 1, respectively, which are close to the values obtained in our previous report [11].

^ 10-8

| 10-9

I 10-10

<¡3 Q.

0.25 0.3 0.35 0.4

Molecular diameter (nm)

Fig. 3. Plots of single gas permeances against the kinetic diameters of gas molecules measured at 90, 50 and 25 °C. Table 1. Ideal selectivities of selected gas pairs in the single gas system measured at 90, 50 and 25 °C.

Temperature (°C) H2/N2 CO2/N2 Ideal selectivity, a O2/N2 H2/CH4 CO2/CH4

90 108 20 6.9 249 47

50 231 40 10.0 537 93

25 400 58 12.1 1354 197

Fig. 4 shows the effect of total feed pressure on CO2 and CH4 permeances in the single gas system, measured at 90, 50 and 25 °C. The results are summarized in Table 2. The CO2 permeances slightly decreased with increasing total feed pressure, while the CH4 permeances remained nearly-constant. As a result, CO2/CH4 ideal selectivities also lowered at high total feed pressure. This degree of reduction tends to be larger at lower permeation temperature. This phenomenon can be attributed to the Langmuir-type

adsorption effect of CO2, in which adsorption coefficients decrease with increasing relative pressure, in addition to the molecular sieving effect.

£ 10"8

E 10-9

a> 10-10 c TO

i 10-11

a) lu Q.

0 0.2 0.4 0.6 0.8 1 1.2 Total feed pressure (MPa)

Single gas co 2 r

Î T I 11]

ch -o—o—o-o— 4

-O-O--0—o-

• ,O:900C ♦^:500C ■,□ :25°C

Fig. 4. Effect of total feed pressure on CO2 and CH4 permeances in the single gas system measured at 90, 50 and 25 °C. Table 2. CO2, CH4 permeances and CO2/CH4 ideal selectivities at different total feed pressure in the single gas system.

Temperature Total feed pressure (MPa)

(°C) 0.1 0.3 0.5 0.8 0.99

CO2 permeance (mol m -2 s-1 Pa-1)

90 8.76x10-9 8.64x10-9 8.60x10-9 8.51x10-9 8.49x10-9

50 6.73x10-9 6.49x10"9 6.30x10"' 5.92x10-9 5.72x10-9

25 4.65x10-9 4.39x10"9 4.24x10"9 4.05x10-9 3.96x10-9

CH4 permeance (mol m -2 s-1 Pa-1)

90 1.88x10-10 1.90x10"10 1.90x10"10 1.94x10-10 1.96x10-10

50 7.23x10-11 7.06x10"n 6.92x10"1] 6.64x10-11 6.71x10-11

25 2.36x10-11 2.41x10-11 2.45x10-11 2.45x10-11 2.47x10-11

CO2/CH, ideal selectivity (-)

90 46.6 45.4 45.2 43.8 43.4

50 93.1 92.0 91.0 89.2 85.2

25 197 182 173 165 161

3.2. Binary gas separation performance

Fig. 5 shows the effect of CO2 concentration in the feed on CO2 and CH4 permeances in the binary gas system, measured at 90, 50 and 25 °C. The results are summarized in Table 3. The total feed pressure was controlled at 0.1 MPa, and the permeate side was vacuumed. At the permeation temperature of 90 °C, the CO2 and CH4 permeances were almost constant, thus indicating that the permeation through the SPPO carbon membrane module was hardly dependent on the feed concentration. However, the CO2

permeances were slightly decreased and CH4 permeances were slightly increased with increasing CO2 concentration in the feed gas at lower temperature. This resulted in decreasing CO2/CH4 ideal selectivities at high CO2 concentration. Compared to the single gas system, lower gas permeances and CO2/CH4 ideal selectivities are observed at higher CO2 concentration in the mixed gas system. This trend was opposite to the previous works, which may be due to the differences in pore structure caused by using different polymer precursor in this study [2-6]. According to Fig. 2, the pore-size of the SPPO carbon membrane module was ca. 0.35-0.4 nm, which was a little smaller than that of other carbon membranes. Kusakabe et al. [12] discussed the separation of CO2/N2 mixture by polyimide-derived carbon membranes and suggested that when the pore-size is narrower than the sum of CO2 and N2 molecular sizes, one is barely able to pass the other in the pore. In this way, the CO2 permeation decreased by the permeation of CH4, and the CH4 permeances increased with the adsorbed CO2 in the mixed gas system.

10-7 1 co2

? 10-8 CI......1......1......B

'e 10-9

a) 10-10

i= 10-11

• ,O:900C ♦,O:500C ■,□ :25°C

0 20 40 60 80 100 CO2 concentration in the feed (vol.%)

co2 1-•-1 1-< 1-<

0-o-o-—o-o-n

......'

1-□-□-□-

• ,0:90°C ^,0:50°C ■ ,□ :25°C

Fig. 5. Effect of CO2 concentration in the feed on CO2 and CH4 permeances in the binary gas system measured at 90, 50 and 25 °C. Table 3. CO2, CH4 permeances and CO2/CH4 ideal selectivities at different feed CO2 concentration in the binary gas system.

Temperature Binary gas Single gas

(°C) CO2 concentration in the feed (vol.%)

CO2 permeance (mol m-2 s- 1 Pa-1)

90 9.27: >10-9 8.81>10-9 8.99>10-9 8.84>10-9 8.76>10-9

50 6.81 >10-9 6.70>10-9 6.67>10-9 6.37>10-9 6.73>10-9

25 5.08: >10-9 4.84>10-9 4.68>10-9 4.65>10-9 4.65>10-9

CH4 permeance (mol m-2 s- 1 Pa-1)

90 1.85> :10-10 1.78>10-10 1.88>10-10 2.01>10-10 1.88>10-10

50 6.45> :10-11 6.61>10-11 6.97>10-11 6.97>10-11 7.23>10-11

25 2.55> :10-11 2.79>10-11 2.93>10-11 3.15>10-11 2.36>10-11

CO2/CH4 ideal selectivity (-)

90 50.2 49.6 47.9 43.9 46.6

50 106 101 95.7 91.4 93.1

25 199 173 160 148 197

Fig. 6 shows the effect of total feed pressure on CO2 and CH4 permeances in the binary gas system, measured at 90, 50 and 25 °C. The results are summarized in Table 4. The feed CO2 concentration was 40 vol.%, which is the common biogas composition. As observed in the single gas system (Fig. 4), the CO2 permeances decreased with increasing total feed pressure. The amount of change was a little larger in the mixed gas system. On the contrary, CH4 permeances increased with increasing total feed pressure, which was more apparent at lower permeation temperature. As a result, CO2/CH4 ideal selectivities also lowered at higher total feed pressures. This degree of reduction in ideal selectivities was larger than that in the single gas system. This trend can be explained by the combined effects of CO2 adsorption and pore-size limitation as mentioned above.

ra 10-8

E 10-9 ö ¡m

® 10-10

Binary gas (40 vol.%CO2)

ril........1.....IT

• ,0:90°C ♦,O:500C ■ ,□ :25°C

0 0.2 0.4 0.6 0.8 1 Total feed pressure (MPa)

Fig. 6. Effect of total feed pressure on CO2 and CH4 permeances in the binary gas system (feed gas: 40 vol.% CO2) measured at 90, 50 and 25 °C.

Table 4. CO2, CH4 permeances and CO2/CH4 ideal selectivities at different total feed pressure in the binary gas system (feed gas: 40 vol.% CO2).

Temperature Total feed pressure (MPa)

(°C) 0.1 0.3 0.5 0.8 0.99

CO2 permeance (mol m "2 s"1 Pa"1)

90 8.81x10"' 8.42x10-9 8.13x10-9 7.98x10"9 7.79x10"9

50 6.70x10"9 5.86x10-9 5.46x10-9 5.22x10"9 5.00x10"9

25 4.84x10"' 4.31x10-9 4.22x10-9 3.95x10"9 3.80x10"9

CH4 permeance (mol m "2 s"1 Pa"1)

90 1.78x10"10 1.77x10-10 1.75x10-10 1.87x10"10 1.89x10"10

50 6.61x10"" 6.48x10-11 6.42x10-11 6.51x10"11 6.45x10"11

25 2.79X10"11 3.06x10-11 3.36x10-11 3.57x10"11 3.69x10"11

CO2/CH ideal selectivity (-)

90 49.6 47.5 46.4 42.8 41.2

50 101 90.4 85.0 80.1 77.6

25 173 141 126 111 103

4. Conclusions

Single as well as mixed gas permeations of CO2 and CH4 through the SPPO carbon membrane module were investigated in this study. The module was successfully prepared using 195 carbon hollow fiber membranes with a sharp pore-size of ca. 0.35-0.4 nm. The SPPO carbon membrane module achieved excellent gas separation performance, and the CO2/CH4 ideal selectivity was 197 at 25 °C in the single gas system. The CO2 permeances were slightly affected by the total feed pressure due to the adsorption effect of CO2 in tandem with the molecular sieving mechanism. In the mixed gas system, the CO2 and CH4 permeances were approximately equal to the single gas system at 90 °C. However, the permeation through the membrane module was dependent on the feed concentration at lower temperatures. When 40 vol.% CO2/CH4 mixture was provided, the CO2 and CH4 permeances were influenced by the total feed pressure. CO2/CH4 ideal selectivity decreased from 173 to 103 when the total feed pressure was changed from 0.1 to 0.99 MPa at 25 °C. The results indicated that the pore-size of the membrane has an important role for the high CO2/CH4 separation performance under various conditions.

References

[1] Ismail AF, David LIB. A review on the latest development of carbon membranes for gas separation, J. Membr. Sci. 2001; 193: 1-18.

[2] Saufi SM, Ismail AF. Fabrication of carbon membranes for gas separation - a review, Carbon 2004; 42: 241-259.

[3] Sedigh MG, Onstot WJ, Xu L, Peng WL, Tsotsis TT, Sahimi M. Experiments and simulation of transport and separation of gas mixtures in carbon molecular sieves membranes, J. Phys. Chem. 1998; A102: 8580-8589.

[4] Ogawa M, Nakano Y. Separation of CO2/CH4 mixture through carbonized membrane prepared by gel modification, J. Membr. Sci. 2000; 173: 123-132.

[5] Tin PS, Chung TS, Liu Y,Wang R. Separation of CO2/CH4 through carbon molecular sieve membranes derived from P84 polyimide, Carbon 2004; 42: 3123-3131.

[6] Anderson CJ, Pas SJ, Arora G, Kentish SE, Hill AJ, Sandler SI, Stevens GW. Effect of pyrolysis temperature and operating temperature on the performance of nanoporous carbon membranes, J. Membr. Sci. 2008; 322: 19-27.

[7] Yoshimune M, Fujiwara I, Suda H, Haraya K. Novel carbon molecular sieve membranes derived from poly(phenylene oxide) and its derivatives for gas separation, Chem. Lett. 2005; 958-959.

[8] Yoshimune M, Fujiwara I, Suda H, Haraya K. Gas transport properties of carbon molecular sieve membranes derived from metal containing sulfonated poly(phenylene oxide), Desalination 2006; 193: 66-72.

[9] Yoshimune M, Fujiwara I, Haraya K. Carbon molecular sieve membranes derived from trimethylsilyl substituted poly(phenylene oxide) for gas separation, Carbon 2007; 45: 553-560.

[10] Yoshimune M, Haraya K. Flexible carbon hollow fiber membranes derived from sulfonated poly(phenylene oxide), Sep. Purif Technol. 2010; 75: 193-197.

[11] Yoshimune M, Mizoguchi K, Haraya K. Alcohol dehydraion by pervaporation using a carbon hollow fiber membrane derived from sulfonated poly(phenylene oxide), J. Membr. Sci. in press.

[12] Kusakabe K, Yamamoto M, Morooka A. Gas permeation and micropore structure of carbon molecular sieving membranes modified by oxidation, J. Membr. Sci. 1998; 149: 59-67.