Gas separation properties of functionalized carbon nanotubes mixed matrix membranes

Gas separation properties of functionalized carbon nanotubes mixed matrix membranes

Separation and Purification Technology 78 (2011) 208–213 Contents lists available at ScienceDirect Separation and Purification Technology journal home...

977KB Sizes 0 Downloads 28 Views

Separation and Purification Technology 78 (2011) 208–213

Contents lists available at ScienceDirect

Separation and Purification Technology journal homepage:

Gas separation properties of functionalized carbon nanotubes mixed matrix membranes S.M. Sanip a , A.F. Ismail a,b,∗ , P.S. Goh a , T. Soga c , M. Tanemura c , H. Yasuhiko c a b c

Advanced Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia Faculty of Petroleum and Renewable Energy Engineering, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia Department of Frontier Materials, Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Aichi 466-8555, Japan

a r t i c l e

i n f o

Article history: Received 29 October 2010 Received in revised form 31 January 2011 Accepted 3 February 2011 Keywords: Multiwalled carbon nanotubes Mixed matrix membrane Gas separation

a b s t r a c t The synergistic combinations of organic polymers for separation applications with inorganic substances such as multi-walled carbon nanotubes (MWNTs), have resulted in a new class of membrane material called mixed matrix membrane (MMM) for the separation of CO2 /CH4 gases. Mixed matrix membrane incorporated functionalized MWNTs (f-MWNTs) were fabricated by the solution casting method, in which the f-MWNTs were embedded into the polyimide membrane and the resulting membranes were characterized. The effect of nominal MWNTs content between 0.5 and 1.0 wt% on the gas separation properties was investigated. The mixed matrix membranes showed 100% enhancement for the selectivity of CO2 /CH4 compared to the corresponding neat polymer membrane. This new class of mixed matrix membrane has the ability to separate gases at the molecular level and has the potential to ultimately reduce the energy consumed in present-day separation operations. This study has shown that addition of CNTs to polymeric membranes has improved separation properties of the membranes to a certain extent. © 2011 Elsevier B.V. All rights reserved.

1. Introduction With the rapidly increasing interest in carbon dioxide capture to mitigate global warming, gas separation polymeric membranes is currently an active and vibrant research area to address this issue [1,2]. The energy efficiency and simplicity of membrane gas separation make it extremely attractive for carbon dioxide capture. Membrane has the ability to selectively pass one component in a mixture while rejecting others and possesses a number of advantages over conventional processes as it is more compact, energy efficient, green technology and economical [3]. One of the best methods to improve membrane separation properties would be to modify existing polymers with the incorporation of fillers. Mixed matrix membranes (MMMs) combine useful molecular sieving properties of inorganic fillers with the desirable mechanical and processing properties of polymers [4]. The current trend in polymeric membranes is the incorporation of fillerlike nanoparticles to improve the separation performance. Most mixed matrix membranes have shown higher gas permeabilities, improved or similar gas selectivities compared to the corresponding pure polymer membranes [5–7].

∗ Corresponding author at: Advanced Membrane Technology Research Centre (AMTEC), Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia. Tel.: +60 7 5535807; fax: +60 7 5535925. E-mail addresses: [email protected], [email protected] (A.F. Ismail). 1383-5866/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.seppur.2011.02.003

Membrane separation in MMM follows the solution-diffusion model whereby the permeation of molecules through membranes is controlled by two major parameters; diffusivity coefficient (D) and solubility coefficient (S). The diffusivity is a measure of the mobility of individual molecule passing through the voids between the polymeric chains in a membrane material. However, separation is not just diffusion dependent but is also reliant on the physical–chemical interaction between the various gas species and the polymer, which determines the amount of gas that can accumulate in the membrane polymeric matrix. For ideal gases, the permeability is related to the gas permeation rate through the membrane (Q), the surface area of the membrane (A), the thickness of the membrane (l) and the driving force for separation, the pressure difference across the membrane (p) [8,9]: P Q = l A p


The ideal selectivity (˛) of one gas, a, over another gas, b, is defined as: ˛=

Pa Pb


Membrane permeability is inversely proportional to the membrane area required for separation. However, for most membranes, there is a trade-off between selectivity and permeability. A highly permeable membrane tends to have low selectivity, and vice versa as described by Robeson [10]. The incorporation of organic fillers

S.M. Sanip et al. / Separation and Purification Technology 78 (2011) 208–213

has resulted in new membrane materials with properties lying far beyond the Robeson upper-bound limit for the organic polymers. Hence this has provided novel routes towards the preparation of MMM with enhanced gas separation properties. MMM based on carbon nanotubes explore the possibilities of overcoming the trade-off between selectivity and permeability through the reengineering and modification of materials. The rapid mass transport behavior of carbon nanotubes which is orders of magnitude higher than any other porous filler has made it the ideal candidate for gas separation membranes [11]. Hence it is desirable to construct highly permeable and selective membranes containing carbon nanotubes inside a polymer matrix that could easily be scaled up to large area membranes. However the preparation of satisfactory CNT based MMM is still a great challenge that needs to be overcome so that the full potential and applications of MMM can be realized. Several researchers have embarked upon this challenge and showed promising results as presented in Table 1 [12–19]. Effective use of CNTs in MMM depends greatly on the ability to disperse the CNTs uniformly through the matrix. As-produced CNTs tend to assemble into crystalline ropes or bundles due to the strong van der Waals attraction among the tubes. Ropes are typically made up of 100–500 tubes and formed highly entangled networks [20]. The dispersion problem for CNTs is rather different from other conventional fillers, such as spherical particles and carbon fibers, because of the difficulties associated with dispersion of the entangled CNTs during processing and poor interfacial interaction between CNTs and polymer matrix. This is due to the characteristics of CNTs having a small diameter in nanometer scale with high aspect ratio (>1000) and thus extremely large surface area [21]. The functionalization of CNTs is an effective way to prevent nanotube aggregation, which helps to better disperse and stabilize the CNTs within a polymer matrix. There are several approaches for functionalization of CNTs including defect functionalization, covalent functionalization and noncovalent functionalization [22]. Noncovalent functionalization of CNTs with cyclodextrins (CDs) appears to be a promising way to further develop environmentfriendly hybrid materials with a combination of properties. The interaction of CNTs with CDs has been previously studied [23–27]. CDs are cyclic oligosaccharide of 6–8 glucopyranoside units which can be represented as toroids with an inner cavity of several angstroms in diameters [27]. Chen in his study, discovered that SWNTs can be efficiently cut simply by grinding in CDs. This solid state process successfully avoids treatment in strong acids and oxidants which can severely damage the small-diameter nanotubes, and the laborious method of sonication in any solvent which could make scaling-up difficult [23]. In this work, as-grown MWNTs were first functionalized with beta-cyclodextrin (beta-CD). The beta-CD functionalized MWNTs were characterized using Raman spectroscopy to show that the nanotubes were not damaged upon treatment as the smooth nanotubes walls will result in a fast transport of the gas molecules studied. MMM was then prepared with the addition of f-MWNTs in different ratios. The asymmetric membranes were then prepared using solution blending of polyimide (PI) in N-methyl-2-pyrolidone (NMP) and f-MWNTs in NMP. The mixture was then cast into thin sheets of MMM. The as-prepared membranes were characterized for their morphology using field emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM). The effects of the CO2 /CH4 gas separation properties upon the nominal addition of 0.5–1.0 wt% of f-MWNTs were investigated. To date, few researchers have prepared MMM based on CNTs and further work on this area is critical in order to develop MMM with higher selectivity and productivity.


Fig. 1. As-grown MWNTs (50,000×).

2. Experimental 2.1. MWNTs synthesis The as-grown MWNTs samples used were synthesized by catalytic decomposition of acetylene (C2 H2 ) over supported metal catalyst; iron and cobalt (Fe/Co) which was obtained from Sigma and Acros, respectively [28]. The support used was alumina hydrate (Al2 O3 ) powder (Merck). The resultant MWNTs produced having diameters in the range of 10–40 nm and more than 500 nm in length, is depicted as in Fig. 1. The MWNTs were used as prepared without further purification. 2.2. Functionalization of MWNTs The MWNTs were then dispersed in beta-cylcodextrin (beta-CD) which was obtained from Sigma. The mixtures were ground with a known amount of ethanol until a homogeneous powder is achieved [23,25]. They were then further mixed through a ball-milling process. This procedure resulted in a fine homogeneous cyclodextrin functionalized MWNTs (f-MWNTs) powder. 2.3. Mixed matrix membrane preparation 0.5–1 wt% f-MWNT were solution-blended with the polyimide (PI) polymer (Alfa-Aesar) in N-methyl-2-pyrrolidone (NMP) solvent with a polymer weight of 20–23%. The MMM were prepared by phase inversion process of the f-MWNT/PI solution using water as a coagulant. The prepared MMM were left in water for 24 h to encourage further solvent exchange. The MMM were then dried in atmosphere for 24 h. 2.4. Characterization techniques JASCO Raman Spectroscopy System was used to detect the degree of defects of f-MWNTs by green laser, with excitation wavelength of 532 nm and power of 14.1 mW. The samples were scanned three times at room temperature with acquisition time of 1 min. Zeiss Supra 35VP FESEM, equipped with wave (energy) dispersive X-ray (EDS) instrument was used to study the morphology of MMM. Transition Electron Microscopy (TEM) was performed by JEOL, JEM-4000, EX11. The samples were prepared by sonication of the f-MWNTs in ethanol and a few drops of the resultant suspension was put onto a holey carbon grid and dried. The TEM morphology of the cross section of MMM was analysed using JEOL JEM-1011. The samples were wrapped in epoxy resin and sliced thinly using a microtome at room temperature.


S.M. Sanip et al. / Separation and Purification Technology 78 (2011) 208–213

Table 1 Selectivities of some CNTs MMM. References


Predictions/simulation selectivity data

Experimental selectivity data


Chen et al. [11]


Not available

Zhang et al. [12]


303 K, 138 kPa for feed pressure drop 303 K, 20 bar

Razavi et al. [13] Huang et al. [14] Kim et al. [15] Cong et al. [16]


CH4 /H2 3–15 (mixture gases) CO2 /CH4 4.4 (mixture gases) CO2 /N2 4–11 CO2 /CH4 9–10 Not available Not available

Tseng et al. [17]


Not available

Weng et al. [18]


Not available

Kim et al. [19]


Not available

Not available Not available Not available CO2 /CH4 1–1.5 CO2 /N2 30–40 CO2 /N2 30–40 CO2 /N2 4.1 O2 /N2 3.3 CO2 /O2 1.3 H2 /CH4 6.5–8 CO2 /CH4 2–3.5 CO2 /CH4 16–18.8 O2 /N2 5–5.1 CH4 /N2 1.17–1.27

300 K, 0.15 MPa 303 K, 10 bar 308 K, 50 psi 298 K, 10 psi 299 K, 2 atm

298 K, 2 kg/cm2 308 K, 4 atm

2.5. Gas permeability measurement The gas permeation properties of the mixed matrix membrane were measured using variable-pressure constant-volume method with a pre-calibrated permeation cell described elsewhere [29]. The gas permeability measurement for all the pure gases was performed at 35◦ C with pressures up to 10 bar and measured three times for each membrane sample. Pure CO2 , CH4 , N2 and O2 were used as the test gases. 3. Results and discussions 3.1. Raman spectroscopy The Raman spectra analysis can provide qualitative information on the status of sidewall functionalization, which corresponds to the change of properties of the treated MWNTs. The D band (1350 cm−1 ) represents a disorder induced effect which is attributed to lattice distortions due to disordered sp2 bonded carbon atoms, or to the presence of structural defects (such as the presence of chemical groups), or both [30,31]. The G-band is due to tangential vibration of carbon (1500–1600 cm−1 ) which is related to the vibration of sp2 -bonded carbon atoms in a twodimensional hexagonal lattice, such as in a graphitic layer. The presence of functional groups on the surface of the MWNTs will


Fig. 3. TEM of f-MWNTs (250,000×).

see an increase in the defect peak for the hexagonal graphitic layers. As such, an increase in the intensity of D-band is observed when MWNTs are functionalized as depicted in Fig. 2 [32]. Both D and G band peaks are still present for the functionalized MWNTs samples. This observation has shown that the graphitic and nanotube structure remained stable upon functionalization with beta-CD. This treatment did not damage the small diameter tubes as evidenced from their TEM micrograph. This is important because the smooth walls of the MWNTs have the advantage of creating nano channels

f-MWNT with beta-CD G peak

D peak

Raman Intensity

0.06 0.04 0.02 0.00


-0.02 -0.04 1100

as-grown CNTs 1216






Raman Shift cm-1 Fig. 2. Raman spectra of as-grown MWNTs and f-MWNTs.

Fig. 4. TEM of cross-section of MMM.

S.M. Sanip et al. / Separation and Purification Technology 78 (2011) 208–213


Fig. 5. Cross section morphologies of (a) CNT-MMM, (b) f-MWNTs-MMM 0.5%, (c) f-MWNTs-MMM 0.7% and (d) f-MWNTs-MMM 1% (200,000×).

for the fast transport of gases. As a result, the gas separation performance of the MMM prepared showed an improvement with the addition of the f-MWNTs. 3.2. Transition electron microscopy The TEM micrograph shown in Fig. 3 depicts the f-MWNTs. The multiwalled carbon nanotubes graphitic walls are still visible indicating that the structure of the MWNTs did not undergone any destructive process during the functionalization. This observation was supported by their Raman spectra as seen earlier. The

nanotubes hollow structure facilitate the fast transport of gases through the smooth cavity and hence increases the performance of the MMM for gas separation. The cross section of the MMM in Fig. 4, also showed that the fMWNTs were well dispersed in the polymer matrix with the lengths of the CNTs being cut to between 100 and 300 nm in length. This is also indicative that functionalization with beta-CD have successfully opened, cut and disentangled the MWNTs therefore enhancing the dispersion of MWNTs. This treatment has also improved the interfacial interactions between the CNTs and the polymer matrix which is an important parameter for preparing a good MMM.

Permeance (GPU) 0.7% CD MMM

Selecvity for CO2/CH4 for MMM














Permeance (GPU)


0.7%CD-MMM 1.0%CD-MMM






0 1








Pressure (bar) Fig. 6. Permeance (GPU) of pure gases for MMM with 0.7% f-MWNTs at different pressures.

0.00 5






Pressure (bar) Fig. 7. Selectivity for CO2 /CH4 gases for MMM with different CNTs loadings.


S.M. Sanip et al. / Separation and Purification Technology 78 (2011) 208–213

Polymer matrix f-MWNTs



Polymer matrix




Polymer matrix

Fig. 8. Schematic of gas permeation through MMM with different ratios of f-MWNTs: (a) 0.5%, (b) 0.7% and (c) 1.0%.

3.3. Field emission scanning electron microscopy The cross section of SEM micrographs in Fig. 5(a)–(d) showed different morphologies for MMM with different CNTs ratio. Fig. 5(b) and (c) depicts the MMM with f-MWNTs of ratios 0.5 and 0.7%. The f-MWNTs with beta-CD have resulted in the formation of uniform finger-like structures which can be attributed to an enhanced compatibility between the inorganic and the organic phases. While the as-grown MWNT-MMM in Fig. 5(a) showed non-uniform fingerlike structures and large pore holes. This can be attributed to the incomplete dispersion of the MWNTs within the polymer matrix due to stronger interactions between the MWNTs bundles [33] and slower solvent exchange process during coagulation hence the formation of larger finger-like structures resembling pore holes [28]. However, at 1% ratio of f-MWNTs, a more viscous solution has resulted in a dense-like structure as observed in Fig. 5(d). The threshold limit for the addition of CNTs to the polymer matrix to prevent viscous, thixotrophic networks agglomeration is typically around 1% [34]. Hence, from these observations, the optimum ratio for the addition of f-MWNTs is 0.7% which have corresponded to the significant improvement in the separation properties of the MMM as observed in Figs. 6 and 7.

The selectivity of the neat PI and different ratios of f-MWNTs is summarized in Fig. 7. The addition of nominal amounts of f-MWNTs has resulted in more than 100% increase in the selectivity of the MMM. The optimum ratio for enhanced selectivity was at 0.7% loading of the f-MWNTs. At 0.5% and 1.0% loading the selectivity was found to be almost the same or lower than the neat PI polymer. The increase in pressure did not exert much change on the selectivity of the MMM indicating that the addition of the f-MWNTs have given additional mechanical strength to the MMM and that the produced membranes are stable up to a pressure of 10 bar [35]. However, for MMM with 0.5% loading of as-grown CNTs, no selectivity was observed after a pressure of 7 bar. This is due to the formation of large nanogaps and poor interfacial interaction between the CNTs and polymer matrix as a result of untreated CNTs. The enhanced selectivity observed at 0.7% ratio of f-MWNTs MMM is also a result of a fast diffusion of gas through the welldispersed CNTs particles in the MMM as represented schematically in Fig. 8. The oriented CNTs channels have provided for a smooth flow and penetration of gas through the polymer matrix. Above this ratio, agglomeration and saturation of CNTs particles in the polymer matrix hinders the fast transport of gases due to increased resistance through the nanotubes cavity.

3.4. Gas separation properties of MMM 4. Conclusion The transport of gases in a porous membrane is usually governed by the solution-diffusion mechanism presence in all polymeric membranes. The permeance for 0.7% ratio of f-MWNTs MMM for different pure gases, CO2 , CH4 , N2 and O2 is reported as in Fig. 6. The permeation rate is in the order of CO2 > CH4 > O2 > N2 and favours that of CO2 . The MMM with 0.7% f-MWNTs showed a high permeance for CO2 as compared to the other gases. The high permeation rate observed for CO2 , is a result of the more strongly adsorbing properties of CO2 through the channels of CNTs when compared to O2 and N2 . The presence of functional groups such as the hydroxyl groups on the beta-CD has also contributed towards the enhancement of the permeance of the MMM. The increase in the solubility of CO2 in the MMM was the result of the strong interaction between the CO2 molecules and the functional groups on the f-MWNTs [19]. An increased in free volume of the polymer matrix as a result of the disruption of the polymer chains packing due to the interaction of the CNTs and the polymer segmental chains have also contributed towards the increment of the permeance by enhancing the gas diffusion [16]. The addition of open-ended and shortened CNTs has also resulted in favorable permeability increase for CO2 as compared to the other gases. On the other hand, the slower diffusing CO2 molecules reduced the mobility of the adsorbed CH4 which has resulted in an increase in the selectivity for CO2 as evidenced from Fig. 7 [19].

These results have shown that improvement of the solubility and homogeneous dispersion of CNTs in the MMM is a result of the functionalization treatment of MWNTs with beta-CD. Raman spectra analysis has confirmed that functionalization did not destroy the nanotubes structure. The morphologies of the MMM also indicated that at 0.7% loading of f-MWNTs, the structures of the MMM showed uniform finger-like structures which have facilitated the fast gas transport through the polymer matrix. It may also be concluded that addition of open ended and shortened CNTs to the polymer matrix can improve its permeance by increasing diffusivity through the CNTs smooth cavity. The f-MWNTs also offered favorable effects towards increasing gas permeability by enhancing the diffusion and adsorption of CO2 in the polymer matrix. Increased in free volume and the presence of nanotubes channels have resulted in more than 100% improved selectivity for the MMM. Hence the addition of f-MWNTs can be seen to enhance the permeability without compromising on the selectivity of MMM.

Acknowledgements The authors wish to thank Universiti Teknologi Malaysia (UTM), JSPS under the RONPAKU Fellowship Program (for Ms. S.M. Sanip) and special thanks to Prof. Tanemura from NIT and JEOL of Japan for the TEM micrographs.

S.M. Sanip et al. / Separation and Purification Technology 78 (2011) 208–213

References [1] S. Bertelle, T. Gupta, D. Roizard, C. Vallières, E. Favre, Study of polymer–carbon mixed matrix membranes, Desalination 199 (2006) 401–402. [2] A. Brunetti, F. Scura, G. Barbieri, E. Drioli, Membrane technologies for CO2 separation, J. Membr. Sci. 359 (2010) 115–125. [3] C.A. Scholes, E.K. Sandra, G.W. Stevens, Carbon dioxide separation through polymeric membrane systems for flue gas applications, Recent Patents on Chemical Engineering, 2008, 1, 52–66, 1874-4788/08, Bentham Science Publishers Ltd. [4] T.S. Chung, L.Y. Jiang, Y. Li, S. Kulprathipanj, Mixed matrix membranes (MMMs) comprising organic polymers with dispersed inorganic fillers for gas separation, Prog. Polym. Sci. 32 (2007) 483–507. [5] S. Kim, T.W. Pechar, E. Marand, Poly(imide siloxane) and carbon nanotube mixed matrix membranes for gas separation, Desalination 192 (2006) 330–339. [6] M. Moaddeb, W.J. Koros, Gas transport properties of thin polymeric membranes in the presence of silicon dioxide particles, J. Membr. Sci. 125 (1997) 143–163. [7] J.H. Kim, Y.M. Lee, Gas permeation properties of poly(amide-6-b-ethylene oxide)–silica hybrid membranes, J. Membr. Sci. 193 (2001) 209–225. [8] W. Richard, Baker, Membrane Technology and Applications, 2nd ed., John Wiley and Sons Ltd., 2004. [9] S.H. Han, G.W. Kim, C.H. Jung, Y.M. Lee, Control of pore characteristics in carbon molecular sieve membranes (CMSM) using organic/inorganic hybrid materials, Desalination 233 (2008) 88–95. [10] L.M. Robeson, The upper bound revisited, J. Membr. Sci. 320 (2008) 390–400. [11] H. Chen, D.S. Sholl, Predictions of selectivity and flux for CH4 /H2 separations using single walled carbon nanotubes as membranes, J. Membr. Sci. 269 (2006) 152–160. [12] X. Zhang, X.H. Shao, W.H. Wang, D. Cao, Molecular modeling of selectivity of single-walled carbon nanotube and MCM-41 for separation of methane and carbon dioxide, Sep. Purif. Technol. 74 (2010) 280–287. [13] S.S. Razavi, S.M. Hashemianzadeh, H. Karimi, Modeling the adsorptive selectivity of carbon nanotubes for effective separation of CO2 /N2 mixtures, J. Mol. Model, DOI 10 1007/s00894-010-0810-9, Publish online 8 August 2010. [14] L.L. Huang, L.H. Zhang, Q. Shao, L.H. Lu, X.H. Lu, S. Jiang, W.F. Shen, Simulations of binary mixture adsorption of carbon dioxide and methane in carbon nanotubes: temperature, pressure, and pore size effects, J. Phys. Chem. C 111 (2007) 11912–11920. [15] S. Kim, L. Chen, J.K. Johnson, E. Marand, Polysulfone and functionalized carbon nanotube mixed matrix membranes for gas separation: theory and experiment, J. Membr. Sci. 294 (2007) 147–158. [16] H. Cong, J. Zhang, M. Radosz, Y. Shen, Carbon nanotube composite membranes of brominated poly(2 6-diphenyl-1,4-phenylene oxide) for gas separation, J. Membr. Sci. 274 (2007) 177–185. [17] H.H. Tseng, I.A. Kumar, T.H. Weng, C.Y. Lu, M.Y. Wey, Preparation and characterization of carbon molecular sieve membranes for gas separation—the effect of incorporated multi-wall carbon nanotubes, Desalination 240 (2009) 40–45.


[18] T.H. Weng, H.H. Tseng, M.Y. Wey, Preparation and characterization of multi-walled carbon nanotube/PBNPI nanocomposites membrane for H2 /CH4 separation, Int. J. Hydrogen Energy 34 (2009) 8707–8715. [19] S. Kim, J.R. Jinschek, H. Chen, D.S. Sholl, E. Marand, Scalable fabrication of carbon nanotube/polymer nanocomposite membranes for high flux gas transport, Nano Lett. 7 (2007) 2806–2811. [20] P.C. Ma, N.A. Siddiqui, G. Marom, J.K. Kim, Dispersion and functionalization of carbon nanotubes for polymer-based nanocomposites: a review, Composites Part A 41 (2010) 1345–1367. [21] K. Sears, L. Dumee, J. Schutz, M. She, C. Huynh, S. Hawkins, M. Duke, S. Gray, Recent developments in carbon nanotube membranes for water purification and gas separation, Materials 3 (2010) 127–149. [22] N.G. Sahoo, S. Rana, J.W. Cho, L. Li, S.H. Chan, Polymer nanocomposites based on functionalized carbon nanotubes, Prog. Polym. Sci. 35 (2010) 837–867. [23] J. Chen, M.J. Dyer, M.F. Yu, Cyclodextrin-mediated soft cutting of single-walled carbon nanotubes, J. Am. Chem. Soc. 123 (2001) 6201–6202. [24] S. Samal, K.E. Geckeler, Cylcodextrin-fullerenes: a new classs of water-soluble fullerenes, Chem. Comm. 13 (2000) 1101–1102. [25] F. Peng, C. Hu, Z. Jiang, Novel ploy(vinyl alcohol)/carbon nanotube hybrid membranes for pervaporation separation of benzene/cyclohexane mixtures, J. Membr. Sci. 297 (2007) 236–242. [26] K. Liu, H. Fu, Y. Xie, L. Zhang, K. Pan, W. Zhou, Assembly of ␤-cyclodextrins acting as molecular bricks onto multiwall carbon nanotubes, J. Phys. Chem. C 112 (2008) 951–957. [27] L.Y. Jianga, T.S. Chung, ␤-Cyclodextrin containing Matrimid® subnanocomposite membranes for pervaporation application, J. Membr. Sci. 327 (2009) 216–225. [28] S.M. Sanip, A.F. Ismail, M. Aziz, T. Soga, Functionalized carbon nanotubes for mixed matrix membrane, IEICE Trans. Electron. E92-C (December (12)) (2009) 1427–1431. [29] T.D. Kusworo, A.F. Ismail, A. Mustafa, T. Matsuura, Dependence of membrane morphology and performance on preparation conditions: the shear rate effect in membrane casting, Sep. Purif. Technol. 61 (2008) 249–257. [30] T. Belin, F. Epron, Characterization methods of carbon nanotubes: a review, Mater. Sci. Eng. B 119 (2005) 105–118. [31] N. Larouche, B.L. Stansfield, Classifying nanostructured carbons using graphitic indices derived from Raman spectra, Carbon 48 (2010) 620–629. [32] J. Qui, G. Wang, C. Zhou, Preparation and characterization of amphiphilic multiwalled carbon nanotubes, J. Nanopart. Res 10 (2008) 659–663. [33] E.V. Perez, K.J. Balkus Jr., J.P. Ferraris, I.H. Musselman, Mixed-matrix membranes containing MOF-5 for gas separations, J. Membr. Sci. 328 (2009) 165–173. [34] H. Qian, E.S. Greenhalgh, M.S.P. Shaffer, A. Bismarck, Carbon nanotube-based hierarchical composites: a review, J. Mater. Chem. 20 (2010) 4751–4762. [35] T. Liu, Y. Tong, W.D. Zhang, Preparation and characterization of carbon nanotube/polyetherimide nanocomposite films, Compos. Sci. Technol. 67 (2007) 406–412.