Incorporation of Carbon Nanotubes Into Lyotropic Crystal Phase

3562 DOI: 10.1021/la902960a Langmuir 2010, 26(5), 3562–3568Published on Web 09/03/2009

pubs.acs.org/Langmuir

© 2009 American Chemical Society

Incorporation of CarbonNanotubes into aLyotropic LiquidCrystal byPhase

Separation in the Presence of a Hydrophilic Polymer

Xia Xin,*,† Hongguang Li,† Stefan A. Wieczorek,† Tomasz Szymborski,† Ewelina Kalwarczyk,†

Natalia Ziebacz,† Ewa Gorecka,‡ Damian Pociecha,‡ and Robert Hozyst*,†,§

†Department III, Institute of Physical Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224Warsaw, Poland, ‡Department of Chemistry, University ofWarsaw, Al. Zwirki i Wigury 101, Warsaw, Poland,

and §Department of Mathematics and Natural Sciences, College of Science, Cardinal Stefan WyszynskiUniversity, Dewajtis 5, 01-815 Warsaw, Poland

Received August 10, 2009

Single-walled carbon nanotubes (SWNTs) were incorporated into a lyotropic liquid crystal (LLC) matrix formed byn-dodecyl octaoxyethene monoether (C12E6) at room temperature through spontaneous phase separation induced bynonionic hydrophilic polymer poly(ethylene glycol) (PEG). The quality of SWNTs/LLC composite was evaluated bypolarized microscopy observations and small-angle X-ray scattering (SAXS) measurements. The results obtainedclearly indicated that SWNTs have been successfully incorporated into the LLCmatrix up to a considerable high contentwithout destroying the LLC matrix, although interesting changes of the LLC matrix were also induced by SWNTsincorporation. By varying the ratio of PEG to C12E6, the type of LLCmatrix can be controlled from hexagonal phase tolamellar phase. Temperature was found to have a significant influence on the quality of SWNTs/LLC composite, andtube aggregation can be induced at higher temperature. When SWNTs were changed to multiwalled carbon nanotubes(MWNTs), they became difficult to be incorporated into LLC matrix because of an increase in the average tubediameter.

Introduction

Because of a combination of their unusual structural, mechan-ical, and electronic properties, carbon nanotubes (CNTs) haveshown great potential applications in physics, chemistry and lifescience.1-3 The freshly produced rawCNTs usually exist as ropesrather than single tubes because of the van der Waals attractionbetween adjacent tubes. This is especially evident for single-walledcarbon nanotubes (SWNTs) with high length to width ratio. Theaggregation of SWNTs is a disadvantage in their applications,because the aggregates are difficult to process into useful nano-material structures. In recent years, it has been shown thatsurfactants and amphiphilic macromolecules can prevent nano-tube aggregation in aqueous solution under ultrasonication. Inthis way dispersions containing randomly oriented single tubes in

water have been obtained.4-19 As SWNTs are anisotropic nano-particles, they exhibit most of their remarkable properties in asingle direction: along the tube axis.20 The achievement of uni-form alignment is therefore a crucial condition in most of theapplications proposed to exploit the unique features of CNTs.Several methods have been developed for aligning dispersednanotubes,21-28 among which nanotube alignment by usinglyotropic liquid crystal (LLC) phases is a promising one.29-39

*To whom correspondence should be addressed. E-mail: [email protected] (R.H.), [email protected] (X.X.).(1) Hirsch, A. Angew. Chem., Int. Ed. 2002, 41, 1853.(2) Song, W. H.; Kinloch, I. A.; Windle, A; H. Science 2003, 302, 1363.(3) Moulton, S. E.; Maugey, M.; Poulin, P.; Wallace, G. G. J. Am. Chem. Soc.

2007, 129, 9452.(4) Bonard, J. M.; Stora, T.; Salvetat, J. P.; Maier, F.; St€ochli, T.; Dischl, C.;

Forr�o, L.; Heer, W.; Chatelain, A. Adv. Mater. 1997, 9, 827.(5) Vigolo, B.; P�enicaud, A.; Coulon, C.; Sauder, C.; Pailler, R.; Kpirmet, C.;

Bermoer, P. Science 2000, 290, 1331.(6) O’Connell, M. J.; Bachilo, S. M.; Huffman, C. B.; Moore, V. C.; Strano,

M. S.; Haroz, E. H.; Rialon, K. L.; Boul, P. J.; Noon, W. H.; Kittrell, C.; Ma, J.;Hauge, R. H.; Weisman, R. B.; Smalley, R. E. Science 2002, 297, 593.(7) Bachilo, S. M.; Strano, M. S.; Kittrell, C.; Hauge, R. H.; Smalley, R. E.;

Weisman, R. B. Science 2002, 298, 2361.(8) Strano, M. S.; Dyke, C. A.; Usrey, M. L.; Barone, P. W.; Allen, M. J.; Shan,

H.; Kittrell, C.; Hauge, R. H.; Tour, J. M.; Smalley, R. E. Science 2003, 301, 1519.(9) Moore, V. C.; Strano, M. S.; Haroz, E. H.; Hauge, R. H.; Smalley, R. E.

Nano Lett. 2003, 3, 1379.(10) Nap, R.; Szleifer, I. Langmuir 2005, 21, 12072.(11) Shvartzman-Cohen, R.; Nativ-Roth, E.; Baskaran, E.; Levi-Kalisman, Y.;

Szleifer, I.; Yerushalmi-Rozen, R. J. Am. Chem. Soc. 2004, 126, 14850.(12) Shvartzman-Cohen, R.; Levi-Kalisman, Y.; Nativ-Roth, E.; Yerushalmi-

Rozen, R. Langmuir 2004, 20, 6085.(13) Grunlan, J. C.; Liu, L.; Kim, Y. S. Nano Lett. 2006, 6, 911.

(14) Li, H.; Zhou, B.; Lin, Y.; Gu, L.;Wang,W.; Fernando, K. A. S.; Kumar, S.;Allard, L. F.; Sun, Y. P. J. Am. Chem. Soc. 2004, 126, 1014.

(15) Kang, Y.; Taton, T. A. J. Am. Chem. Soc. 2003, 125, 5650.(16) Chen, J.; Liu, H.; Weimer, W. A.; Halls, M. D.; Waldeck, D. H.; Walker,

G. C. J. Am. Chem. Soc. 2002, 124, 9034.(17) Zheng, M.; Jagota, A.; Smeke, E. D.; Diner, B. A.; Mclean, R. S.; Lustig,

S. R.; Richardson, R. E.; Tassi, N. G. Nat. Mater. 2003, 2, 338.(18) Star, A.; Steuerman, D.W.; Heath, J. R.; Stoddart, J. F.Angew. Chem., Int.

Ed. 2002, 41, 2508.(19) Bandyopadhyaya, R.; Nativ-Roth, E.; Regev, O.; Yerushalmi-Rozen, R.

Nano Lett. 2002, 2, 25.(20) Lagerwall, J.; Scalia, G.; Haluska, M.; Dettlaff-Weglikowska, U.; Roth, S.;

Giesselmann, F. Adv. Mater. 2007, 19, 359–364.(21) Nakashima, N.; Tomonari, Y.; Murakami, H. Chem. Lett. 2002, 31, 638.(22) Nakashima, N.; Tanaka, Y.; Tomonari, Y.; Murakami, H.; Kataura, H.;

Sakaue, T.; Yoshikawa, K. J. Phys. Chem. B 2005, 109, 13076.(23) Wang, H.; Christopherson, G.; Xu, Z.; Porcar, L.; Ho, D.; Fry, D.; Hobbie,

E. Chem. Phys. Lett. 2005, 416, 182.(24) Casavant, M.; Walters, D.; Schmidt, J.; Smalley, R. J. Appl. Phys. 2003, 93,

2153.(25) Fischer, J. E.; Zhou,W.; Vavro, J.; Llaguno, C.; Guthy, C.; Haggenmueller,

R.; Casavant, M. J.; Walters, D. E.; Smalley, R. E. J. Appl. Phys. 2003, 93, 2157.(26) Kamat, P. V.; Thomas, K. G.; Barazzouk, S.; Girishkumar, G.; Vinodgopal,

K.; Meisel, D. J. Am. Chem. Soc. 2004, 126, 10757.(27) Nativ-Roth, E.; Regev, O.; Yerushalmi-Rozen, R. Chem.Commun. 2008,

17, 2037.(28) Nativ-Roth, E.; Yerushalmi-Rozen, R.; Regev, O. Small 2008, 4, 1459.(29) Stewart, G. T. Liq. Cryst. 2004, 31, 443.(30) Badaire, S.; Zakri, C.; Maugey, M.; Derre, A.; Barisci, J. N.; Wallace,

G. G.; Poulin, P. Adv. Mater. 2005, 17, 1673.(31) Song, W.; Kinloch, I. A.; Windle, A. H. Science 2003, 302, 1363.(32) Davis, V. A.; Ericson, L. M.; Parra-Vasquez, A. N. G.; Fan, H.; Wang, Y.;

Prieto, V.; Longoria, J. A.; Ramesh, S.; Saini, R. K.; Kittrell, C.; Billups, W. E.;Adams, W. W.; Hauge, R. H.; Smalley, R. E.; Pasquali, M.Macromolecules 2004,37, 154.

DOI: 10.1021/la902960a 3563Langmuir 2010, 26(5), 3562–3568

Xin et al. Article

Because of the high viscosity of the LLC phase, however, theincorporation of SWNTs is usually difficult, and rigorous sonica-tion combined with a heating process was usually performed forbetter introduction of SWNTs. This method is inconvenient asexcess heating can destroy the LLC matrix. Particularly in thecase of a nonionic surfactant the LLC phase undergoes a transi-tion to the micellar phase at high temperatures. Also the heatingprocess causes desorption of surfactant molecules from SWNTsand subsequent tube aggregation. Therefore, approaches tocreate a spontaneous phase transition from the isotropic SWNTsdispersions to LLC phases are highly desired.40

In this work, we report a newmethod of SWNTs incorporationinto LLC phase formed spontaneously by a nonionic surfactant(C12E6, n-dodecyl hexaoxyethylene glycol monoether) duringphase separation in the presence of a hydrophilic polymer(PEG, poly(ethylene glycol), molecular weight: 20000). To thebest of our knowledge, this is the first example of spontaneousphase separation from isotropic CNTs dispersions to well-or-dered CNTs/surfactant hybrids in the presence of polymers. Thetype of LLC phase (hexagonal or lamellar) can be convenientlycontrolled by varying the ratio of PEG to C12E6. The quality ofSWNTs/LLC composite as a function of SWNTs content wasevaluated by polarized microscope observations and small-angleX-ray scattering (SAXS) measurements. The influences of tem-perature and tube diameter on the composite were also investi-gated. In the latter case, SWNTswere replaced byMWNTswhichhave a much larger tube diameter.

Experimental Section

Chemicals and Materials. SWNTs, with a diameter <2 nmand a length of 2-15 μm, were purchased from ShenzhenNanotech Port Co., Ltd., which were prepared by chemical vapordeposition (CVD). MWNTs with a purity of ∼95% are called“unbundledMWNTs” and were obtained fromAhwahnee Tech-nology Inc. Each of tubes has 3-5 layers with average tubediameter ranging from 20 to 50 nanometers and a length from0.5 to 200 μm. Nonionic surfactant n-dodecyl hexaoxyethenemonoether (C12E6) was purchased from Fluka, of purity betterthan 98%. The melting temperature given by Fluka ChemicalsCo. is 27-28 �C. The molecular weight is 450.66. PEG with amolecular weight of 20000 was purchased fromFluka. Deionizedwater (18.2 MΩ cm) was used in the preparation of all thesamples.

Sample Preparation Method. Stock solution of 0.1 wt %C12E6 was prepared by dissolving the desired amount of C12E6 inwater at room temperature. Dispersions of SWNTs in 0.1 wt %C12E6 aqueous solution were obtained by sonicating the mixtureat 43-45 kHz and 60 W. After 2 h, a homogeneous blackdispersion was obtained. The as-prepared SWNTs dispersionscontain well-dispersed SWNTs as well as tube bundles. Smallamount of black precipitates were noticed, after deposition atroom temperature for 2 weeks. The upper phase, which can bestable at room temperature for months, was collected for furtheruse.

For preparation of SWNTs-LLC hybrids, desired amounts ofPEG and C12E6 were added to aqueous dispersions of SWNTs in0.1 wt % C12E6. The mixtures were homogenized by stirring forabout 10 minutes. Then the mixture was allowed to equilibrate at

room temperature, during which simultaneous phase separationwas noticed. Finally two phases were formed: an upper LLCcontaining black SWNTs and a bottom transparent isotropicphase.

Techniques. Polarized microscopy observations were carriedout on a Nikon Eclipse E400 microscope equipped with aLINKAM THMS 600 heating/cooling stage. The computerprogram that controlled the stagewasLinkSys 2.36. The precisionof temperature control was within ( 0.01 �C. UV-vis spectros-copy was performed on a MultiSpec-1501, Shimadzu (Kyoto,Japan) apparatus. SAXS measurements were carried out using aCuKR radiation of 0.15418 nmwith the Bruker Nanostar systemand aVantec 2000 area detector. Samples were prepared by fillingthin glass capillaries (1.5 mm in diameter) which were flame-sealed immediately. The signal intensities were obtained throughintegration of the two-dimensional patterns over the azimuthalangle. All the experiments were carried out at room temperatureunless other stated.

Results and Discussion

Incorporation of SWNTs into LLC Matrix by Phase

Separation. The ordering of C12E6 surfactant in dilute aqueoussolutions induced by PEG has been well documented.41-43WhenPEG of suitable molecular weight was added to isotropic micellarphase of C12E6, a slow phase separation process was noticed andfinally, a C12E6 hexagonal phase formed at the top of the vial. Incurrent work, this interesting process was applied for the pre-paration of the SWNTs/LLC hybrids. In principle it was notobvious that carbon nanotubes would be incorporated into thesurfactant matrix. Normally at high surfactant or polymer con-centration the dilute solution of carbon nanotubes undergoes aprecipitation process. To obtain a high concentration of carbonnanotubes in an ordered surfactant matrix we applied thefollowing method. Dispersion of SWNTs in 0.1 wt % C12E6

aqueous solution was prepared by sonication (see ExperimentalSection), to which C12E6 and PEG with a molecular weight of20000 were added. After being homogenized by stirring for about10 minutes, the mixture was allowed to equilibrate at roomtemperature, during which spontaneous phase separation wasnoticed. Finally an upper LLC phase containing black SWNTsand a bottom transparent isotropic phase formed.A typical phaseseparation process is schematically shown in Figure 1 wherethe weight fractions of C12E6 and PEG to water are 10 wt %and 20 wt%, respectively. The phase separation may take severalhours or days, depending on the exact composition of themixture.It was found to be reversible. When the sample after equilibriumwas homogenized again by hand-shaking, it can rebound to thetwo phases spontaneously.

FromFigure 1 we observe that all SWNTs enter into the upperLLC phase without any precipitation. This indicates the highefficiency of this method to incorporate SWNTs into LLC phase.Here SWNTs were incorporated into LLC phase at roomtemperature via spontaneous phase separation. This is a bigadvantage over existing methods where vigorous sonicationcombined with a heating process is necessary to incorporateSWNTs into LLC matrix. The previous method operates withhigh viscosity solution (with high concentration of surfactantsforming ordered phases) while our method is performed at lowviscosity micellar solutions. In some control experiments, we(33) Dierking, I.; Scalia, G.; Morales, P.; LeClere, D. Adv. Mater. 2004, 16, 865.

(34) Dierking, I.; Scalia, G.; Morales, P. J. Appl. Phys. 2005, 97, 044309.(35) Song, W. H.; Windle, A. H. Macromolecules 2005, 38, 6181.(36) Scalia, G.; Lagerwall, J.; Schymura, S.; Haluska, M.; Giesselmann, F.;

Roth, S. Phys. Status Solidi B 2007, 244, 4212.(37) Patrick, M.; Lynch, D. Nano Lett. 2002, 2, 1197.(38) Weiss, V.; Thiruvengadathan, R.; Regev, O. Langmuir 2006, 22, 854.(39) Jiang, W. Q.; Yu, B.; Liu, W. M.; Hao, J. C. Langmuir 2007, 23, 8549.(40) Zhang, S. J.; Kumar, S. Small 2008, 4, 1270.

(41) Holyst, R.; Staniszewski, K.; Demyanchuk, I. J. Phys. Chem. B 2005, 109,4881.

(42) Makulska, S.; Chudy, E.; Urbaniak, K.; Wieczorek, S. A.; Zywocinski, A.;Holyst, R. J. Phys. Chem. B 2007, 111, 7948.

(43) _Zywoci�nski, A.; Korda, A.; Gosk, J.;Wieczorek, S. A.;Wilk, A.; Hozyst, R.J. Am. Chem. Soc. 2007, 129, 13398.

3564 DOI: 10.1021/la902960a Langmuir 2010, 26(5), 3562–3568

Article Xin et al.

observed that it was much more difficult to directly incorporateSWNTs into C12E6 hexagonal phase without the preceding phaseseparation. Also, we found the heating process has a highpotential to destabilize the SWNTs/LLC composites (see below).

Since the volume of the SWNTs dispersion shrunk significantlyduring phase separation, it can be well concluded that theconcentration of SWNTs in the upper LLC phase has beengreatly increased. From the volume fraction of the upper phasewe estimated the increase to be 4-5 fold. The concentration ofSWNTs in the upper LLC phase can be also controlled by simplychanging the concentration of SWNTs dispersed in 0.1 wt %C12E6 aqueous solutions.Unlike ionic surfactants such as SDS4-8

which prevent nanotube aggregation mainly by electrostaticrepulsion, C12E6 impede nanotube aggregation mainly by thewell-known steric repulsion. When mixed with SWNTs, thehydrophobic chains of C12E6 interact with the sidewall of theSWNTs (this kind of interaction was often called “wrapping” or“decorating”) while the EO groups would extend into water andhence create the steric repulsion.18 This wrapping phenomenonhas been used by others to explain the dissolution of SWNTs insolutions containing large polymer molecules.44,45

It was found that samples containing more solid SWNTsgained heavier black color after sonication and subsequentdeposition, as proved by visual inspection and turbidity measure-ments (Figure 2). This is because more tubes were available to be

dispersed into the bulk aqueous solutions as the amount of addedSWNTs increased. This observation is consistent with the phe-nomena known from the solutions of PEO-PPO-PEO triblockcopolymer.46 Different SWNTs dispersions were used as thestarting solutions. Also SWNTs/LLC composite with differentSWNTs content was obtained after phase separation (Figure 3).Polarized Microscopy Observations and SAXS Results.

We used a polarized microscope and SAXS to characterize thequality of SWNTs/LLC composites. Typical samples investigatedhad a composition of 10 wt% C12E6 and 20 wt% PEG to water.At this composition, C12E6 was highly condensed into an upperphase after phase separation, forming anLLCphase of hexagonaltype. In the absence of SWNTs, the upper C12E6 hexagonal phaseexhibited a typical fan-like texture under a polarized microscope(Figure 4, A). When SWNTs were incorporated, the upper C12E6

Figure 1. Schematic illustration of the phase separation process. (A) SWNTs dispersion in 0.1 wt%C12E6 aqueous solution; (B) mixture ofSWNTs dispersion, C12E6 (10 wt%) and PEG (20 wt%) after being homogenized; (C) intermediate stage during phase separation; (D) finalstate after equilibrium.

Figure 2. Left: Photos of typical samples of SWNTs dispersed in 0.1 wt%C12E6 aqueous solutions. The weight percents of added SWNTsare 0.01 wt % (A), 0.03 wt% (B), 0.05 wt% (C), 0.1 wt% (D), and 0.2 wt% (E), respectively. Photos were taken 2 weeks after sonication.Right: Transmittance of the SWNTs dispersions for samples A-E. These datawere obtained at a wavelength of 500 nm, and 0.1 wt%C12E6

in water was used as the baseline.

Figure 3. SWNTsembedded in theupper hexagonalphase formedby C12E6 (10 wt%) after phase separation induced by PEG 20000(20 wt %). The weight percent of SWNTs in the upper phasewas calculated to be 0 wt % - reference sample(A), 0.05 wt % (B),0.15 wt % (C), 0.25 wt % (D), 1 wt % (E). The images were takenmore than 1 month after preparation.

(44) Coleman, J. N.; Dalton, A. B.; Curran, S.; Rubio, A.; Davey, A. P.; Drury,A.; McCarthy, B.; Lahr, B.; Ajayan, P. M.; Roth, S.; Barklie, R. C.; Blau, W.J. Adv. Mater. 2000, 12, 213.(45) Curran, S. A.; Ajayan, P. M.; Blau, W. J.; Carroll, D. L.; Coleman, J. N.;

Dalton, A. B.; Davey, A. P.; Drury, A.; McCarthy, B.; Maier, S.; Strevens, A.Adv.Mater. 1998, 10, 1091.

(46) Xin, X.; Xu, G. Y.; Zhao, T. T.; Zhu, Y. Y.; Shi, X. F.; Gong, H. J.; Zhang,Z. Q. J. Phys. Chem. C 2008, 112, 16377.

DOI: 10.1021/la902960a 3565Langmuir 2010, 26(5), 3562–3568

Xin et al. Article

condensed phase retains the texture of hexagonal phase up to aconsiderable high concentration of incorporated SWNTs, that is,∼1 wt % in the current study (Figure 4D). This observationindicated that the incorporation of SWNTs did not lead tocollapse of the host hexagonal matrix. On the other hand, thecontrast of the micrograph decreased with increased concentra-tion of SWNTs, as could be seen in A, B, and D in Figure 4. Thisobservation could be partly ascribed to a decrease of transparencyof the sample with increased SWNTs concentration. However,the incorporation of SWNTs might also cause some changes ofthe host hexagonal matrix.

From images obtained without crossed polarizers, some differ-ences couldbe also seenwith increasing SWNTs content.When theconcentration of incorporated SWNTs in the upper LLC phasewas below 1 wt %, microscopically homogeneous samples can be

obtained. We did not observe large SWNTs aggregates under themicroscope. Such an example is given in Figure 4C. When theconcentration of incorporated SWNTs reached 1 wt % someSWNTs aggregates were noticed (Figure 4E). This observationwas a sign of the microscopic phase separation between SWNTsand the host hexagonal matrix.Macroscopically the SWNTs/LLCcomposite appeared as a homogeneous upper phase. Probably atsuch high content of SWNTs, the capability of the host hexagonalmatrix to incorporate SWNTs was exceeded.

To gain further details about SWNTs/LLC composites, SAXSmeasurements were carried out as a function of SWNTs content,as shown in Figure 5. Interestingly, with increasing amount ofSWNTs, the d-spacing of the hexagonal lattice, calculated fromthe q value of the first peak (d=2π/qmax), increased continuously.For example, the d-spacing of the hexagonal phase without

Figure 4. Textures of the upper C12E6 hexagonal phases without (A) and with (B-E) incorporated SWNTs. The weight fractions of C12E6

and PEG to water before phase separation are 10 wt% and 20 wt%, respectively. The concentrations of incorporated SWNTs in the upperLLC phases are 0.25 wt % (B, C) and 1 wt % (D, E), respectively. B, D: with crossed polarizers; C, E: without crossed polarizers.

3566 DOI: 10.1021/la902960a Langmuir 2010, 26(5), 3562–3568

Article Xin et al.

SWNTs is 5.16 nm, which shifts to 5.35 nm when the content ofSWNTs reaches 0.25wt%.This observation clearly indicates thatthe incorporation of SWNTs does cause some changes of the hosthexagonal matrix, which is also consistent with polarized micro-scopy observations. The increase in d-spacing indicates thatswelling of the host hexagonal matrix was induced by theincorporation of SWNTs.Controlling LLC Phase type by Varying Ratio of PEG to

C12E6. The upper C12E6 LLC phase is formed because of thecondensation of the micellar phase by PEG during phase separa-tion. The type of LLC phases and their lattice parameters are thusdominated by the concentration of bothC12E6 andPEGaswell astheir mixing ratio. Normally a highmixing ratio of PEG to C12E6

induces not only changes of lattice parameters but also differenttypes of LLC phases after the phase separation. These observa-tions show how, in a convenient way, to control the host of LLCphase for SWNTs incorporation.

We first fixed the concentration of C12E6 at 10 wt % whileincreasing the amount of added PEG. The weight fraction ofSWNTs in 0.1 wt % C12E6 aqueous dispersion was fixed at0.05wt%.At a composition of 10wt%C12E6 and 20wt%PEG,the final content of SWNTs in the upper LLC hexagonal phasewas 0.25 wt % since the volume shrunk ∼5 fold. When theamount of addedPEGwas changed tobe 25wt%, the upperLLCphase was still hexagonal; however, a decrease in d-spacing wasobserved as compared to the case of 20 wt % PEG (Figure 6,curve b). Thus the upper LLC phase was more condensed athigher amount of addedPEGand the concentration of SWNTs inthe upper LLC phase was also increased. It also accounted for thesmaller d-spacing value. The amount of added PEG was furtherincreased to 36 wt % leading to the formation of the lamellarphase (Figure 6, curve c). Swelling was also observed for thelamellar phase induced by SWNTs incorporation, as shown oncurves c and curves d in Figure 6.

The type of the upper LLC phase could also be changed byfixing PEG concentration while varying the concentration ofC12E6. For this purpose PEG concentration was fixed at 20 wt%

while the concentration of C12E6 was varied between 5 wt % and15 wt % and the photographs of the samples were shown inFigure 7A. It was found when the concentration of C12E6 waslowered from 10 wt % to 5 wt %, the d-spacing of the upperhexagonal phase decreased from 5.35 to 4.91 nm (Figure 7 B,curve a and b). In this case the upper LLC phase was morecondensed at lower C12E6 concentration which corresponded to ahigher mixing ratio of PEG to C12E6. Increasing concentration ofC12E6 led to swelling of the LLC phase. In fact, at a C12E6

concentration of 15 wt %, the upper phase had already lost theordered structure (Figure 7B, curve c).Temperature Effects. The properties of SWNTs/LLC com-

posites could be remarkably influenced by temperature. C12E6 isthe nonionic surfactant, whose hydrophilic part undergoes con-tinuous dehydration upon temperature increase. The hydropho-bic chain ofC12E6 above a critical temperature influences the non-covalent interactionwith the sidewall of SWNTs. In the sample of10 wt % C12E6/20 wt % PEG/0.05 wt % SWNTs for example,when the upper SWNTs/LLC composite was heated to 70 �C,above the cloud point of C12E6,

41-43 the fan-like texture disap-peared and nanotube aggregation occurred (Figure 8 A). Whenthe sample was cooled back to 25 �C, the fan-like texture wasrecovered but the aggregated tubes could not be reincorporatedinto the LLC matrix (Figure 8 B). From the SAXS results(Figure 8 C), the d-spacing before temperature rise was 5.35nm. After the temperature rise and subsequent cooling thed-spacing decreased to 5.04 nm. It indicated that SWNTs wereextruded out of the LLC matrix. This experiment gave usadditional proof that before temperature rising, SWNTs hadbeen dispersed and incorporated into the LLC matrix. It alsoimplied that the heating process used in previous method toincorporate SWNTs into LLC phases might significantly reducethe quality of final SWNTs/LLC composites, highlighting theadvantage of the method reported in the current work. HereSWNTs were incoporated into LLC matrix at room temperature

Figure 5. Results of SAXS measurements of SWNTs/LLC com-posites as a function of the concentration of incorporated SWNTsin the upper LLC phase (shown in the insets). The d-spacing of thehexagonal phases are also given. Theweight fractions ofC12E6 andPEG to water before phase separation was 10 wt % and 20 wt %,respectively. The dashed lines are guides to the eyes.

Figure 6. SAXSresults for SWNTs/LLCcompositeswith increas-ing PEG concentration. The weight fraction of C12E6 to water wasfixed at 10wt%while that of SWNTs in 0.1wt%C12E6 dispersionwas fixed at 0.05 wt %, respectively. The weight percent of PEGwas (a) 20 wt %, (b) 25 wt %, and (c) 36 wt %, respectively. Theresult of the sample with 36 wt % PEG without SWNTs was alsoshown (curve d) for comparison. The dashed line is a guide to theeyes.

DOI: 10.1021/la902960a 3567Langmuir 2010, 26(5), 3562–3568

Xin et al. Article

by spontaneous phase separation, and in this way the heatingprocess was avoided.Influence of Tube Diameter. We also investigated the influ-

ence of the tube diameter by changing SWNTs to multiwalledcarbon nanotubes (MWNTs). With a decrease of the van derWaals interactions between adjacent tubesMWNTs were easier tobe dispersed by 0.1 wt % C12E6 aqueous solution (SupportingInformation). However, the subsequent incorporation ofMWNTsinto LLC matrix was more difficult than SWNTs, as proved bypolarized microscopy observations and SAXS measurements.While considerable amount of SWNTs could be incorporated intothe LLC matrix with a satisfactory quality, tube aggregation wasalready observed even at lowMWNTs content of 0.05 wt%.Withincreasing concentration of incorporated MWNTs in theMWNTs/LLC composites, tube aggregation becamemore evident(Supporting Information). At the same time, the textures of LLCphase became weaker and finally disappeared at 0.25 wt %MWNTs (Supporting Information).

Figure 9 summarizes the changes of the d-spacing of the LLCmatrix induced by MWNTs incorporation. Similarly to the caseof SWNTs/LLC composites (see Figure 5), the incorporation ofMWNTs first caused a continuous increase of the d-spacing of thehost LLCmatrix. However, when the content of the incorporated

MWNTs exceeded 0.1 wt%, a decrease in d-spacing was noticed.It was probably due to the microscopic phase separation ofMWNTs within the LLCmatrix. This microscopic phase separa-tion made the LLC more condensed, accounting for an evensmaller d-spacing compared with that of the LLCmatrix withoutMWNTs incorporation. The difficulty ofMWNTs incorporationinto the LLC matrix at high content could be ascribed to a muchlarger tube diameter compared with SWNTs.Driving Forces of SWNTs/LLC Composite Formation.

Recently, Poulin and co-workers3 reported an observation ofmacroscopic phase separation in the hyaluronic acid-SWNT-water system, and attributed the effect to excluded volume inter-actions.We suggested, in the present manuscript, that the physicalorigin of the observed phase behavior is similar, and is related toentropic interactions leading to osmotic depletion forces.47 Severalstudies of depletion-induced phase separations were reported incolloidal systems.48 In the case ofmicellar/polymer solutions it was

Figure 7. Photograhs (A) and SAXS results (B) of SWNTs/LLC composites at varying C12E6 concentration. The weight fraction of PEG towaterwas fixed at 20wt%while thatofSWNTs in0.1wt%C12E6dispersionwas fixed at0.05wt%, respectively.Theweight percentofC12E6

is (a) 5 wt %, (b) 10 wt %, and (c) 15 wt %, respectively. The dashed line is a guide to the eyes.

Figure 8. Typical polarizedmicrograph of the SWNTs/LLC com-posites heated to 70 �C (A) and then cooledback to 25 �C (B). Tubeaggregation was noticed. SAXS results before and after tempera-ture rising are also shown (C) which were measured (a) at 25 �Cbefore temperature rising, (b) at 70 �C, and (c) after cooleddown to25 �C. The weight fractions of C12E6 and PEG to water beforephase separation are 10wt%and 20wt%,while that of SWNTs in0.1 wt % C12E6 dispersion was fixed at 0.05 wt %.

Figure 9. SAXSresults ofMWNTs/LLCcomposites as a functionof the concentration of incorporated MWNTs in the upper LLCphase. as shown in the insets. The weight fractions of C12E6 andPEG to water before phase separation are 10 wt % and 20 wt %,respectively. Two dashed lines are guides to the eyes.

(47) Richard, C.; Balavoine, F.; Schultz, P.; Ebbesen, T. W.; Mioskowski, C.Science 2003, 300, 75.

(48) Gang, H.; Krall, A. H.; Weitz, D. A. Phys. Rev. Lett. 1994, 73, 3435.

3568 DOI: 10.1021/la902960a Langmuir 2010, 26(5), 3562–3568

Article Xin et al.

demonstrated that the hexagonal ordering of surfactants wasinduced only when the polymer radius of gyration, RG, was largerthan the size of thewater channels,Lwater, in thehexagonal phase.

42

In the present manuscript the radius of gyration for PEG 20000calculated according to the equation RG = 0.02M 0.58 nm is6.3 nm, and the size of the water channels of the hexagonal phasefor C12E6 is 1.5 nm, which means the radius of gyration for PEG20000 is larger than the water channels in the hexagonal phase ofC12E6 (depletion interactions), so the phase separation occurredandorderedC12E6-rich phase formed.The cartoon in the Scheme 1shows the main idea of the phase separation and further orderingin the carbon nanotubes/surfactant/polymer/water system.

Conclusions

We described a new method for incorporation of SWNTs intoLLCphase formed by the nonionic surfactantC12E6. Themethodis based on the phase separation in the presence of polymer PEG.Since the LLC phase was formed spontaneously through phaseseparation, therefore, long-lasting sonication and heating pro-cesses were avoided. In our method SWNTs/LLC compositeswere obtained by the phase separation induced by the polymeraddition. The final concentration of SWNTs in the upper C12E6

liquid crystal phase was 4-5 folds higher compared with that ofpreformed CNTs aqueous dispersions. This observation can bevery helpful in practical applications where high-concentrated

SWNTs dispersions are needed. There is a vast combination ofsurfactant and polymer which can be selected for this method.Our preliminary experimental results also show that the LLCordering induced by poly(styrenesulfonate) (PSS) in sodiumdodecyl sulfate (SDS) aqueous solution or poly(diallydimethyl-ammonium chloride) (PDADMAC) in cetyltrimethylammoniumbromide (CTAB) aqueous solution can also lead to the incorpora-tion of SWNTs into the ordered phases of SDS and CTAB.Efforts in this direction and the investigation of the alignment ofCNTs in these LLC phases are currently underway and will bereported in the near future.

Acknowledgment. The work was supported by the Projectoperated within the Foundation for Polish Science Team Pro-gramme co-financed by the EU “European Regional Develop-ment Fund” TEAM/2008-2/2 and also by Research Grant fromthe Human Frontier Science Program Organization. N.Z. ac-knowledges a Ph.D. scholarship from the President of the PolishAcademy of Sciences.

Supporting Information Available: The results ofMWNTsdispersed in 0.1 wt%C12E6 aqueous solution and the typicalpolarized micrographs of C12E6 hexagonal phases contain-ingMWNTs. This material is available free of charge via theInternet at http://pubs.acs.org.

Scheme 1. Schematic Representation of the Phase Separation Process in the Four Component Mixture of the Surfactant C12E6, PEG 20000,

SWNTs, and Water