- Email: [email protected]
Matrimid-based carbon tubular membrane: Effect of carbonization environment
G Model
JIEC-2618; No. of Pages 5 Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Matrimid-based carbon tubular membrane: Effect of carbonization environment N. Sazali a,b, W.N.W. Salleh a,b,*, N.A.H.M. Nordin a,b, A.F. Ismail a,b a b
Advance Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia Faculty of Petroleum and Renewable Energy Engineering (FPREE), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia
A R T I C L E I N F O
Article history: Received 27 May 2015 Received in revised form 6 August 2015 Accepted 11 August 2015 Available online xxx Keywords: Heat treatment Carbonization environment Carbon membrane Argon Nitrogen
A B S T R A C T
Among gas separation materials, carbon membrane exhibits the most interesting performance in terms of selectivity, stability, and gas permeance. By controlling and optimizing carbonization environment, excellent gas separation performances can be achieved. In this study, tubular supported carbon membrane was prepared using Matrimid as polymeric precursor. In order to produce high performance carbon membrane, the effect of carbonization conditions on the gas permeation properties was investigated. The polymer solution was coated on the surface of the tubular support by using dip-coating method. Carbon membranes were fabricated by heat treatment process under controlled carbonization environments; Ar or N2. Pure gas permeation tests were performed using CO2, CH4, and N2 at room temperature with pressure 8 bar. Based on the results, the highest CO2/CH4 and selectivity of 87.34 and CO2/N2 selectivity of 79.60 were obtained by carbon membrane carbonized under Ar gas. Despite the higher carbonization temperature, the carbonization under Ar created more permeable pores as compared to N2 environment. ß 2015 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Nowadays, membranes processes are very important in the industry; for example in medical applications, separation of petrochemicals, and water and waste water treatment. Conventional separation methods used for purification of chemical products, extraction, crystallization, and distillation are energy and cost intensive [1]. Over 50% of the energy costs in the chemical industry are used for the separation of gaseous and liquid mixtures. The costs for difficult separations can be reduced significantly with these membrane technologies [2,3]. It is well known that the membrane performance appears to be a trade-off between selectivity and permeability, for example a highly selective membrane tends to have a low permeability [4]. When permeability is higher, the cost of the system can be lowered and the membrane needed is also small. In order to surpass Robeson’s upper bound, some methods can be applied to suitably prepare the
* Corresponding author at: Faculty of Petroleum and Renewable Energy Engineering (FPREE), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia. Tel.: +60 75535388; fax: +60 75535625. E-mail address: [email protected] (W.N.W. Salleh).
membranes. Previous study mentioned that membranes which have the potential to exceed such upper bound are inorganic membranes [5]. In recent years, special attention has been concentrated on the relationship between the membrane polymer structure and gas separation performances [6]. Separation using polymeric membrane produces high performance separation method and provides lower operating cost compared to other separation techniques [7]. It is widely used for wastewater treatment, gas separation, seawater desalination, distillation, and dialysis [8]. Polymeric membrane is suitable to replace the conventional separation method due to modest energy and modular equipment requirement. Although polymeric membrane based gas separations offer low energy requirement and low capital cost, Robeson trade-off limit between permeability and selectivity is the main obstacle that hinders the polymeric membrane process application [9]. The limitations suffered by polymeric membranes have encouraged researchers to develop new class of membrane. Yet, in harsh environment, these membranes are not suitable to be used, for instance, those prone towards erosion besides high temperatures. Among various classes of membrane, carbon membrane produced from carbonization of polymeric precursor has superior gas separation properties [10].
http://dx.doi.org/10.1016/j.jiec.2015.08.014 1226-086X/ß 2015 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
2
Carbon membrane separation development has become a great tool, and has undergone a rapid improvement since the past few years [11]. Because of their unique characteristics, carbon membranes are used in place of polymer membranes; owing to their high thermal and chemical resistance, long life, higher pore volume, and the simultaneous diffusion and chemical reaction [12]. It has been reported that the permeability obtained from carbon membranes is higher than those typically found in polymeric membranes, and these selectivity is achieved without sacrificing their productivity [13]. The carbonization environments are crucial to provide ideal carbonization conditions for Matrimidbased carbon membrane fabrication. Hence, the experiment was carried out by carbonizing Matrimid-based polymeric membrane under either N2 or Ar atmosphere. The results support the potentiality of this simple and relatively fast procedure, which offers new ways of designing and directly characterizing supported carbon membranes for the gas separation. Matrimid 5218 carbonized under vacuum was reported to produce less permeable but more selective carbon membranes compared to an inert gas carbonization system [14]. The carbonization using Argon gas produce better selectivity compared with other inert gas [15]. However, their potential use is often limited by the inability to economically manufacture large or complexshaped components displaying reliable performance. Furthermore, since carbon membrane is brittle, it requires a strong support or substrate to sustain its structure. When a supporting substrate is used for development of carbon membrane, the substrate must be chemically and physically stable and possess a diffusion resistance, which is lower than that of the carbon membrane [16]. While majority of the reported carbon membrane used flat substrate, tubular support hardly draws similar attraction even though it is mechanically stronger against a compressing pressure and higher membrane area per unit module volume. In this study, the carbon membrane fabrication was different to others study due to the supported used made from TiO2 (4.5–5.5 mm) with a coating of ZrO2 (2–3 nm) on the inner surface which can stand high temperature till 1200 8C. With that, it is important to focus the factors that make carbon membranes very attractive and useful as separation tools [17]. Since the development of the tubular supported carbon membrane is still in research stage, it is important to identify the most ideal carbonization environment that would result in high performance of carbon tubular membrane. Hence, this study aims to develop tubular supported carbon membrane for CO2/CH4 and CO2/N2 separations using different carbonization environments, which are Argon (Ar) and Nitrogen (N2) atmosphere.
solution was prepared by dissolving 15 wt%. of Matrimid with NMP for 7 h with mechanical stirring. The mixture was sonicated to remove all bubbles from the solution. Supported polymeric membranes were prepared by dip-coating the tubular support into the polymeric solution for 15 min. During the dip-coating process, the substrate was slowly dipped into and withdrawn from a tank containing the solution, with a uniform velocity, in order to obtain a uniform coating. The dip-coating cycle was repeated for three times to eliminate pinholes presences on the prepared carbon membrane. The carbonization profile used in this study is adapted from the studies by Salleh and groups [18]. The supported polymeric membranes were undergone ageing at 80 8C for 24 h. The membranes were then immersed in methanol for 2 h before being placed at 100 8C for 24 h inside oven to allow slow removal of the solvent. Subsequently, the supported polymeric membranes were placed in the centre of the Carbolite (Model CTF 12/65/550) wire wound tube furnace to undergo heat treatment process. The length of the furnace tube is approximately 75 cm (heating zone 60 cm) and the diameter is 12 cm. The membranes were subjected to stabilization step at 300 8C at a heating rate of 2 8C/min. During this step, the membranes were held for 30 min at 300 8C. After that, the temperature was increased to the final carbonization temperature of 850 8C at a heating rate of 2 8C/min. During heat treatment process, the atmosphere in the furnace is controlled by flowing the gas using mass flow controller. The types of gas flow environments used throughout the heat treatment process were Ar and N2 atmosphere with the flow rate of 200 ml/min. Finally, the membranes were cooled down naturally to room temperature. The detailed heat treatment profile used in this study is shown in Fig. 1. Membrane characterization The analysis on the existence of the functional group in the prepared membranes was analyzed using a Universal ATR (UATR, Single Reflection Diamond for the Spectrum Two) (PerkinElmer, L1600107). The cross section of the prepared carbon membrane was observed using Scanning Electron Microscopy (SEM). Gas permeation measurement A simple bubble flow metre was used to obtain the permeation properties of the prepared membranes. The apparatus used for this measurement can be found elsewhere [18,19]. The performance of
Experimental Materials Matrimid 5218 was purchased from Merck while N-methyl-2pyrrolidone (NMP) was purchased from Sigma–Aldrich. In the fabrication of carbon tubular membrane, tubular support with 8 cm in length, 1.3 cm in diameter and 0.3 cm in thickness was used. The tubular support was purchased from Shanghai Gongtao Ceramics Co., Ltd with a nominal cut-off of 1 kDa (TAMI). The support are made from TiO2 (4.5–5.5 mm) with a coating of ZrO2 (2–3 nm) on the inner surface. The porosity of the support was 40– 50% with an average pore size of 0.2 mm. Carbon membrane preparation In order to eliminate water vapour or moisture from the polymer and equipments, they were dried inside the oven for one day before the dope solution preparation. Polymer precursor
Fig. 1. Heat treatment profile.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
the membrane was dictated by two important parameters which were permeance and selectivity. A carbon tubular membrane was fitted with rubber O-rings to allow the membrane to be housed and placed inside a stainless steel tubular module with length of 14 cm without leakages. Pure gas CO2 (0.33 nm), CH4 (0.38 nm) and N2 (0.36 nm) were fed into the module at a trans membrane pressure of 8 bar. The permeance, P/l (GPU), and selectivity, a, of the membranes were calculated using equations below: Permeance, (P/l)i: ðP=lÞi ¼
Qi
Dp A
¼
Qi
(1)
D p pDl
1 GPU ¼ 1 106 cm3 ðSTPÞ cm2 s1 cmHg1 where P/l is the permeance of the membrane (GPU), Qi is the volumetric flow rate of gas i at standard temperature which measured by the time taken gas to travel at fix volume and pressure (cm3 (STP/s)), p is the pressure difference between the feed side and the permeation side of the membrane (cmHg), A is the membrane surface area (cm2), D is the an outer diameter of the membrane (cm) and l is the effective length of the membrane (cm). Selectivity, a:
ai= j ¼
ðP=lÞi ðP=lÞ j
(2)
where ai/j is the selectivity of gas penetrant i over gas penetrant j, (P/l)i and (P/l)j are the permeance of gas penetrant i and j, respectively [18]. Each measurement value is the result of several different membranes and the precision in gas permeance for each membrane was found to be within the error less than 10%. The error that occurred was probably due to the fluctuation occurred during the data collection by simple soap film flow metre. Results and discussion Carbon membrane properties In this study, the molecular orientation of the prepared membrane is measured via Frontier Transform Infrared spectroscopy (FTIR) as shown in Fig. 2. The reduction of peaks can be clearly seen at both carbonization environments. It is proves that the
3
polymeric precursor membranes have transformed into carbon membranes under both carbonization environments [7]. There is no apparent different on functional group of carbon membrane under different carbonization environments were observed. It is reported that the changing trends of the peaks slightly similar even different carbonizations environments were applied during the fabrication of the carbon membranes [20]. Fig. 3(a) and (b) shows the SEM images of the cross section of the Matrimid-based carbon membranes prepared under N2 and Ar gas environment, respectively. Both membranes showed a dense structure with no apparent different in the induced by carbonization environment. These results suggested that the dense carbon membrane can be obtained regardless the type of gas environment used. The different between both results was on the thickness of the resultant carbon membrane. The thickness of the carbon membrane prepared under Ar and N2 environment of 65.9 and 66.4 mm was obtained, respectively. It can be stated that, the carbon membrane prepared under Ar environment are more stable than N2 environment. This is because, Ar having an octet configuration and very stable compared to the other inert gas. It will not easily react with other compounds and it will depend on the heat treatment conditions. Besides, N2 only has 5 electrons at outer shell and will more easily react with other compounds. Therefore, Ar provides a better structural arrangement compared to N2 gas [20]. Previous study reported that, although these both gases generally produce satisfactory reproducible results, all of them require high temperature treatment [21]. Gas permeation measurements The carbonization process was performed under two types of gases which are Ar and N2. The measurement of pure gas permeation test was done at 8 bar in the room temperature. It has been reported that the carbonization environment play an important role in the resultant carbon membrane performance [20]. Figs. 4 and 5 show the gas permeation data of the carbon membrane prepared under different carbonization environments. The results revealed that the resultant membranes carbonized under Ar gas flow could produce higher selectivity and permeability compared to those carbonized under N2 gas flow. It is because, carbon membrane produced under high density environment (Ar r = 1.784 g/l) would provide better skin integrity than carbon membrane produced under low density environment (N2
Fig. 2. FTIR analysis of Matrimid-based carbon membranes prepared under N2 and Ar environments.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 4
N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
Fig. 3. SEM images of the Matrimid-based carbon membranes prepared under (a) N2 environment, (b) Ar environment.
r = 1.251 g/l). According to Su and Lua [14] the carbonization process under Ar atmosphere can produce the highest CO2/CH4 selectivity of 476.74 for Kapton-based carbon membrane prepared at 1073 K.
Despite the higher carbonization temperatures, the heat treatment under Ar gas flow created more permeable pores as compared to N2 atmosphere. Based on literature, the membranes carbonized under inert atmospheres could produce ideal gas permeance and selectivity except for Helium gas [14]. In addition, different gas environment would result in different carbonization reaction rate. It is claimed that, the carbonization process can be accelerated to form more porous matrix by using inert gas flow [15]. During the carbonization process, the inert gas flow could sweep away by products that may block pores formed on the skin layer of the carbon membrane. Based on this study, it can be concluded that the effective carbonization environment for the preparation of the Matrimid-based carbon tubular membrane was under Ar gas flow. Conclusion
Fig. 4. Gas permeation results of the carbon membrane prepared under different carbonization environments.
Two different modes of carbonization process on Matrimidbased carbon tubular membrane had been carried out. The carbonization processes were conducted under Ar and N2 gas flow. Permeation results showed that the utilization of Ar environment instead of N2 environment during carbonization could increase gas permeance of CO2, N2, and CH4. The highest CO2/ CH4 and CO2/N2 separation of 87.34 0.04 and 79.60 0.03 were obtained for carbon membrane prepared under Ar environment, respectively. This might be due to the capability of Ar gas that favoured the volatile compounds release during carbonization, which could sweep away by-products that might block pores formed on the skin layer of the carbon membrane. Acknowledgement The authors acknowledge the financial support by the Research University Grant, Universiti Teknologi Malaysia (UTM) for the research activities undertaken in Advanced Membrane Technology Research Centre (AMTEC). References
Fig. 5. CO2/CH4 and CO2/N2 selectivity of the carbon membrane prepared under different carbonization environments.
[1] E. Favre, in: E. Drioli, L Giorno (Eds.), Comprehensive Membrane Science and Engineering, Elsevier, Oxford, 2010, p. 155. [2] R.W. Baker, K. Lokhandwala, Ind. Eng. Chem. Res. 47 (7) (2008) 2109. [3] Y. Xiao, et al. Prog. Polym. Sci. 34 (6) (2009) 561. [4] M.L. Chua, Y.C. Xiao, T.-S. Chung, Chem. Eng. Sci. 104 (0) (2013) 1056. [5] M. Askari, T.-S. Chung, J. Membr. Sci. 444 (0) (2013) 173. [6] R.W. Baker, Ind. Eng. Chem. Res. 41 (6) (2002) 1393. [7] W.N.W. Salleh, et al. Sep. Purif. Rev. 40 (4) (2011) 261. [8] M.G. Buonomenna, J. Bae, Renew. Sust. Energy Rev. 43 (0) (2015) 1343. [9] L.M. Robeson, J. Membr. Sci. 62 (2) (1991) 165. [10] K. Bricen˜o, et al. Int. J. Hydrog. Energy 37 (18) (2012) 13536. [11] A.F. Ismail, in: E. Drioli, L Giorno (Eds.), Comprehensive Membrane Science and Engineering, Elsevier, Oxford, 2010, p. 275. [12] E.K. Chatzidaki, et al. Eur. Polym. J. 43 (12) (2007) 5010.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx [13] [14] [15] [16] [17]
W.N.W. Salleh, A.F. Ismail, Sep. Purif. Technol. 88 (0) (2012) 174. J. Su, A.C. Lua, J. Membr. Sci. 305 (1–2) (2007) 263. M. Kiyono, P.J. Williams, W.J. Koros, J. Membr. Sci. 359 (1–2) (2010) 2. L. Li, et al. J. Membr. Sci. 450 (0) (2014) 469. S.S. Hosseini, et al. Sep. Purif. Technol. 122 (0) (2014) 278.
[18] [19] [20] [21]
W.N.W. Salleh, A.F. Ismail, Sep. Purif. Technol. 80 (3) (2011) 541. W.N.W. Salleh, A.F. Ismail, AIChE J. 58 (10) (2012) 3167. M. Wang, et al. J. Alloys Comp. 617 (0) (2014) 219. T.A. Centeno, J.L. Vilas, A.B. Fuertes, J. Membr. Sci. 228 (1) (2004) 45.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
5
JIEC-2618; No. of Pages 5 Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Matrimid-based carbon tubular membrane: Effect of carbonization environment N. Sazali a,b, W.N.W. Salleh a,b,*, N.A.H.M. Nordin a,b, A.F. Ismail a,b a b
Advance Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia Faculty of Petroleum and Renewable Energy Engineering (FPREE), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia
A R T I C L E I N F O
Article history: Received 27 May 2015 Received in revised form 6 August 2015 Accepted 11 August 2015 Available online xxx Keywords: Heat treatment Carbonization environment Carbon membrane Argon Nitrogen
A B S T R A C T
Among gas separation materials, carbon membrane exhibits the most interesting performance in terms of selectivity, stability, and gas permeance. By controlling and optimizing carbonization environment, excellent gas separation performances can be achieved. In this study, tubular supported carbon membrane was prepared using Matrimid as polymeric precursor. In order to produce high performance carbon membrane, the effect of carbonization conditions on the gas permeation properties was investigated. The polymer solution was coated on the surface of the tubular support by using dip-coating method. Carbon membranes were fabricated by heat treatment process under controlled carbonization environments; Ar or N2. Pure gas permeation tests were performed using CO2, CH4, and N2 at room temperature with pressure 8 bar. Based on the results, the highest CO2/CH4 and selectivity of 87.34 and CO2/N2 selectivity of 79.60 were obtained by carbon membrane carbonized under Ar gas. Despite the higher carbonization temperature, the carbonization under Ar created more permeable pores as compared to N2 environment. ß 2015 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Nowadays, membranes processes are very important in the industry; for example in medical applications, separation of petrochemicals, and water and waste water treatment. Conventional separation methods used for purification of chemical products, extraction, crystallization, and distillation are energy and cost intensive [1]. Over 50% of the energy costs in the chemical industry are used for the separation of gaseous and liquid mixtures. The costs for difficult separations can be reduced significantly with these membrane technologies [2,3]. It is well known that the membrane performance appears to be a trade-off between selectivity and permeability, for example a highly selective membrane tends to have a low permeability [4]. When permeability is higher, the cost of the system can be lowered and the membrane needed is also small. In order to surpass Robeson’s upper bound, some methods can be applied to suitably prepare the
* Corresponding author at: Faculty of Petroleum and Renewable Energy Engineering (FPREE), Universiti Teknologi Malaysia, 81310 Skudai, Johor Darul Takzim, Malaysia. Tel.: +60 75535388; fax: +60 75535625. E-mail address: [email protected] (W.N.W. Salleh).
membranes. Previous study mentioned that membranes which have the potential to exceed such upper bound are inorganic membranes [5]. In recent years, special attention has been concentrated on the relationship between the membrane polymer structure and gas separation performances [6]. Separation using polymeric membrane produces high performance separation method and provides lower operating cost compared to other separation techniques [7]. It is widely used for wastewater treatment, gas separation, seawater desalination, distillation, and dialysis [8]. Polymeric membrane is suitable to replace the conventional separation method due to modest energy and modular equipment requirement. Although polymeric membrane based gas separations offer low energy requirement and low capital cost, Robeson trade-off limit between permeability and selectivity is the main obstacle that hinders the polymeric membrane process application [9]. The limitations suffered by polymeric membranes have encouraged researchers to develop new class of membrane. Yet, in harsh environment, these membranes are not suitable to be used, for instance, those prone towards erosion besides high temperatures. Among various classes of membrane, carbon membrane produced from carbonization of polymeric precursor has superior gas separation properties [10].
http://dx.doi.org/10.1016/j.jiec.2015.08.014 1226-086X/ß 2015 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
2
Carbon membrane separation development has become a great tool, and has undergone a rapid improvement since the past few years [11]. Because of their unique characteristics, carbon membranes are used in place of polymer membranes; owing to their high thermal and chemical resistance, long life, higher pore volume, and the simultaneous diffusion and chemical reaction [12]. It has been reported that the permeability obtained from carbon membranes is higher than those typically found in polymeric membranes, and these selectivity is achieved without sacrificing their productivity [13]. The carbonization environments are crucial to provide ideal carbonization conditions for Matrimidbased carbon membrane fabrication. Hence, the experiment was carried out by carbonizing Matrimid-based polymeric membrane under either N2 or Ar atmosphere. The results support the potentiality of this simple and relatively fast procedure, which offers new ways of designing and directly characterizing supported carbon membranes for the gas separation. Matrimid 5218 carbonized under vacuum was reported to produce less permeable but more selective carbon membranes compared to an inert gas carbonization system [14]. The carbonization using Argon gas produce better selectivity compared with other inert gas [15]. However, their potential use is often limited by the inability to economically manufacture large or complexshaped components displaying reliable performance. Furthermore, since carbon membrane is brittle, it requires a strong support or substrate to sustain its structure. When a supporting substrate is used for development of carbon membrane, the substrate must be chemically and physically stable and possess a diffusion resistance, which is lower than that of the carbon membrane [16]. While majority of the reported carbon membrane used flat substrate, tubular support hardly draws similar attraction even though it is mechanically stronger against a compressing pressure and higher membrane area per unit module volume. In this study, the carbon membrane fabrication was different to others study due to the supported used made from TiO2 (4.5–5.5 mm) with a coating of ZrO2 (2–3 nm) on the inner surface which can stand high temperature till 1200 8C. With that, it is important to focus the factors that make carbon membranes very attractive and useful as separation tools [17]. Since the development of the tubular supported carbon membrane is still in research stage, it is important to identify the most ideal carbonization environment that would result in high performance of carbon tubular membrane. Hence, this study aims to develop tubular supported carbon membrane for CO2/CH4 and CO2/N2 separations using different carbonization environments, which are Argon (Ar) and Nitrogen (N2) atmosphere.
solution was prepared by dissolving 15 wt%. of Matrimid with NMP for 7 h with mechanical stirring. The mixture was sonicated to remove all bubbles from the solution. Supported polymeric membranes were prepared by dip-coating the tubular support into the polymeric solution for 15 min. During the dip-coating process, the substrate was slowly dipped into and withdrawn from a tank containing the solution, with a uniform velocity, in order to obtain a uniform coating. The dip-coating cycle was repeated for three times to eliminate pinholes presences on the prepared carbon membrane. The carbonization profile used in this study is adapted from the studies by Salleh and groups [18]. The supported polymeric membranes were undergone ageing at 80 8C for 24 h. The membranes were then immersed in methanol for 2 h before being placed at 100 8C for 24 h inside oven to allow slow removal of the solvent. Subsequently, the supported polymeric membranes were placed in the centre of the Carbolite (Model CTF 12/65/550) wire wound tube furnace to undergo heat treatment process. The length of the furnace tube is approximately 75 cm (heating zone 60 cm) and the diameter is 12 cm. The membranes were subjected to stabilization step at 300 8C at a heating rate of 2 8C/min. During this step, the membranes were held for 30 min at 300 8C. After that, the temperature was increased to the final carbonization temperature of 850 8C at a heating rate of 2 8C/min. During heat treatment process, the atmosphere in the furnace is controlled by flowing the gas using mass flow controller. The types of gas flow environments used throughout the heat treatment process were Ar and N2 atmosphere with the flow rate of 200 ml/min. Finally, the membranes were cooled down naturally to room temperature. The detailed heat treatment profile used in this study is shown in Fig. 1. Membrane characterization The analysis on the existence of the functional group in the prepared membranes was analyzed using a Universal ATR (UATR, Single Reflection Diamond for the Spectrum Two) (PerkinElmer, L1600107). The cross section of the prepared carbon membrane was observed using Scanning Electron Microscopy (SEM). Gas permeation measurement A simple bubble flow metre was used to obtain the permeation properties of the prepared membranes. The apparatus used for this measurement can be found elsewhere [18,19]. The performance of
Experimental Materials Matrimid 5218 was purchased from Merck while N-methyl-2pyrrolidone (NMP) was purchased from Sigma–Aldrich. In the fabrication of carbon tubular membrane, tubular support with 8 cm in length, 1.3 cm in diameter and 0.3 cm in thickness was used. The tubular support was purchased from Shanghai Gongtao Ceramics Co., Ltd with a nominal cut-off of 1 kDa (TAMI). The support are made from TiO2 (4.5–5.5 mm) with a coating of ZrO2 (2–3 nm) on the inner surface. The porosity of the support was 40– 50% with an average pore size of 0.2 mm. Carbon membrane preparation In order to eliminate water vapour or moisture from the polymer and equipments, they were dried inside the oven for one day before the dope solution preparation. Polymer precursor
Fig. 1. Heat treatment profile.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
the membrane was dictated by two important parameters which were permeance and selectivity. A carbon tubular membrane was fitted with rubber O-rings to allow the membrane to be housed and placed inside a stainless steel tubular module with length of 14 cm without leakages. Pure gas CO2 (0.33 nm), CH4 (0.38 nm) and N2 (0.36 nm) were fed into the module at a trans membrane pressure of 8 bar. The permeance, P/l (GPU), and selectivity, a, of the membranes were calculated using equations below: Permeance, (P/l)i: ðP=lÞi ¼
Qi
Dp A
¼
Qi
(1)
D p pDl
1 GPU ¼ 1 106 cm3 ðSTPÞ cm2 s1 cmHg1 where P/l is the permeance of the membrane (GPU), Qi is the volumetric flow rate of gas i at standard temperature which measured by the time taken gas to travel at fix volume and pressure (cm3 (STP/s)), p is the pressure difference between the feed side and the permeation side of the membrane (cmHg), A is the membrane surface area (cm2), D is the an outer diameter of the membrane (cm) and l is the effective length of the membrane (cm). Selectivity, a:
ai= j ¼
ðP=lÞi ðP=lÞ j
(2)
where ai/j is the selectivity of gas penetrant i over gas penetrant j, (P/l)i and (P/l)j are the permeance of gas penetrant i and j, respectively [18]. Each measurement value is the result of several different membranes and the precision in gas permeance for each membrane was found to be within the error less than 10%. The error that occurred was probably due to the fluctuation occurred during the data collection by simple soap film flow metre. Results and discussion Carbon membrane properties In this study, the molecular orientation of the prepared membrane is measured via Frontier Transform Infrared spectroscopy (FTIR) as shown in Fig. 2. The reduction of peaks can be clearly seen at both carbonization environments. It is proves that the
3
polymeric precursor membranes have transformed into carbon membranes under both carbonization environments [7]. There is no apparent different on functional group of carbon membrane under different carbonization environments were observed. It is reported that the changing trends of the peaks slightly similar even different carbonizations environments were applied during the fabrication of the carbon membranes [20]. Fig. 3(a) and (b) shows the SEM images of the cross section of the Matrimid-based carbon membranes prepared under N2 and Ar gas environment, respectively. Both membranes showed a dense structure with no apparent different in the induced by carbonization environment. These results suggested that the dense carbon membrane can be obtained regardless the type of gas environment used. The different between both results was on the thickness of the resultant carbon membrane. The thickness of the carbon membrane prepared under Ar and N2 environment of 65.9 and 66.4 mm was obtained, respectively. It can be stated that, the carbon membrane prepared under Ar environment are more stable than N2 environment. This is because, Ar having an octet configuration and very stable compared to the other inert gas. It will not easily react with other compounds and it will depend on the heat treatment conditions. Besides, N2 only has 5 electrons at outer shell and will more easily react with other compounds. Therefore, Ar provides a better structural arrangement compared to N2 gas [20]. Previous study reported that, although these both gases generally produce satisfactory reproducible results, all of them require high temperature treatment [21]. Gas permeation measurements The carbonization process was performed under two types of gases which are Ar and N2. The measurement of pure gas permeation test was done at 8 bar in the room temperature. It has been reported that the carbonization environment play an important role in the resultant carbon membrane performance [20]. Figs. 4 and 5 show the gas permeation data of the carbon membrane prepared under different carbonization environments. The results revealed that the resultant membranes carbonized under Ar gas flow could produce higher selectivity and permeability compared to those carbonized under N2 gas flow. It is because, carbon membrane produced under high density environment (Ar r = 1.784 g/l) would provide better skin integrity than carbon membrane produced under low density environment (N2
Fig. 2. FTIR analysis of Matrimid-based carbon membranes prepared under N2 and Ar environments.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 4
N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx
Fig. 3. SEM images of the Matrimid-based carbon membranes prepared under (a) N2 environment, (b) Ar environment.
r = 1.251 g/l). According to Su and Lua [14] the carbonization process under Ar atmosphere can produce the highest CO2/CH4 selectivity of 476.74 for Kapton-based carbon membrane prepared at 1073 K.
Despite the higher carbonization temperatures, the heat treatment under Ar gas flow created more permeable pores as compared to N2 atmosphere. Based on literature, the membranes carbonized under inert atmospheres could produce ideal gas permeance and selectivity except for Helium gas [14]. In addition, different gas environment would result in different carbonization reaction rate. It is claimed that, the carbonization process can be accelerated to form more porous matrix by using inert gas flow [15]. During the carbonization process, the inert gas flow could sweep away by products that may block pores formed on the skin layer of the carbon membrane. Based on this study, it can be concluded that the effective carbonization environment for the preparation of the Matrimid-based carbon tubular membrane was under Ar gas flow. Conclusion
Fig. 4. Gas permeation results of the carbon membrane prepared under different carbonization environments.
Two different modes of carbonization process on Matrimidbased carbon tubular membrane had been carried out. The carbonization processes were conducted under Ar and N2 gas flow. Permeation results showed that the utilization of Ar environment instead of N2 environment during carbonization could increase gas permeance of CO2, N2, and CH4. The highest CO2/ CH4 and CO2/N2 separation of 87.34 0.04 and 79.60 0.03 were obtained for carbon membrane prepared under Ar environment, respectively. This might be due to the capability of Ar gas that favoured the volatile compounds release during carbonization, which could sweep away by-products that might block pores formed on the skin layer of the carbon membrane. Acknowledgement The authors acknowledge the financial support by the Research University Grant, Universiti Teknologi Malaysia (UTM) for the research activities undertaken in Advanced Membrane Technology Research Centre (AMTEC). References
Fig. 5. CO2/CH4 and CO2/N2 selectivity of the carbon membrane prepared under different carbonization environments.
[1] E. Favre, in: E. Drioli, L Giorno (Eds.), Comprehensive Membrane Science and Engineering, Elsevier, Oxford, 2010, p. 155. [2] R.W. Baker, K. Lokhandwala, Ind. Eng. Chem. Res. 47 (7) (2008) 2109. [3] Y. Xiao, et al. Prog. Polym. Sci. 34 (6) (2009) 561. [4] M.L. Chua, Y.C. Xiao, T.-S. Chung, Chem. Eng. Sci. 104 (0) (2013) 1056. [5] M. Askari, T.-S. Chung, J. Membr. Sci. 444 (0) (2013) 173. [6] R.W. Baker, Ind. Eng. Chem. Res. 41 (6) (2002) 1393. [7] W.N.W. Salleh, et al. Sep. Purif. Rev. 40 (4) (2011) 261. [8] M.G. Buonomenna, J. Bae, Renew. Sust. Energy Rev. 43 (0) (2015) 1343. [9] L.M. Robeson, J. Membr. Sci. 62 (2) (1991) 165. [10] K. Bricen˜o, et al. Int. J. Hydrog. Energy 37 (18) (2012) 13536. [11] A.F. Ismail, in: E. Drioli, L Giorno (Eds.), Comprehensive Membrane Science and Engineering, Elsevier, Oxford, 2010, p. 275. [12] E.K. Chatzidaki, et al. Eur. Polym. J. 43 (12) (2007) 5010.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
G Model
JIEC-2618; No. of Pages 5 N. Sazali et al. / Journal of Industrial and Engineering Chemistry xxx (2015) xxx–xxx [13] [14] [15] [16] [17]
W.N.W. Salleh, A.F. Ismail, Sep. Purif. Technol. 88 (0) (2012) 174. J. Su, A.C. Lua, J. Membr. Sci. 305 (1–2) (2007) 263. M. Kiyono, P.J. Williams, W.J. Koros, J. Membr. Sci. 359 (1–2) (2010) 2. L. Li, et al. J. Membr. Sci. 450 (0) (2014) 469. S.S. Hosseini, et al. Sep. Purif. Technol. 122 (0) (2014) 278.
[18] [19] [20] [21]
W.N.W. Salleh, A.F. Ismail, Sep. Purif. Technol. 80 (3) (2011) 541. W.N.W. Salleh, A.F. Ismail, AIChE J. 58 (10) (2012) 3167. M. Wang, et al. J. Alloys Comp. 617 (0) (2014) 219. T.A. Centeno, J.L. Vilas, A.B. Fuertes, J. Membr. Sci. 228 (1) (2004) 45.
Please cite this article in press as: N. Sazali, et al., J. Ind. Eng. Chem. (2015), http://dx.doi.org/10.1016/j.jiec.2015.08.014
5