C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1

.sc ienced i rec t .com

Avai lab le a t www

ScienceDirect

journal homepage: www.elsevier .com/ locate /carbon

Rheological properties of graphene oxide liquidcrystal

http://dx.doi.org/10.1016/j.carbon.2014.08.0850008-6223/� 2014 Elsevier Ltd. All rights reserved.

* Corresponding author.E-mail address: sangouk.kim@kaist.ac.kr (S.O. Kim).

Pradip Kumar, Uday Narayan Maiti, Kyung Eun Lee, Sang Ouk Kim *

Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Department of Materials Science & Engineering KAIST,

Daejeon 305-701, Republic of Korea

A R T I C L E I N F O

Article history:

Received 3 June 2014

Accepted 22 August 2014

Available online 29 August 2014

A B S T R A C T

We report the rheological properties of liquid crystalline graphene oxide (GO) aqueous dis-

persion. GO dispersions exhibit typical shear thinning behaviors of liquid crystals, which is

described by power law or simple Curreau model. Irrespective of the shear rate, shear vis-

cosity exhibits sudden decrease with the increase of GO composition around a critical vol-

ume fraction, /c = 0.33%, demonstrating typical colloidal isotropic–nematic phase

transition. Dynamic measurements reveal the liquid-like (isotropic phase, G 0 > G00) behavior

at a low GO composition (/ � 0.08%) and solid-like (liquid crystalline) behavior at higher

compositions (/ � 0.45%), where G 0 exceeds over G00. Nematic gel-like phase is confirmed

at a higher GO composition over / > 0.83%, where both G 0 and G00 moduli are nearly inde-

pendent of frequency (x). Simple power law scaling arguments are introduced to model

the dependence of yield stress and viscoelastic moduli on the GO composition. We also

observed the yield stress and rigidity percolation transition above phase transition compo-

sition /c > 0.33% with a percolation exponent of 1.3 ± 0.1. These rheological insights provide

valuable information for the liquid crystalline processing of GO based materials including

fibers, sheets and other complex structures for electronic/optoelectronic and energy stor-

age/conversion applications.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Mesoscopic ordering of anisotropic molecules or particles in

fluid states, known as liquid crystalline (LC) phase, has

attracted a great deal of research attention for the signifi-

cance in display devices and highly ordered materials pro-

cessing [1–4]. Among various liquid crystals, carbonaceous

mesophases are of particular research interest, as they offer

valuable routes to functional carbon based materials [5–10].

Recently, graphene oxide (GO), an oxygenated form of graph-

ene with spontaneous solution dispersibility, has also been

reported to form discotic colloidal liquid crystalline phase in

aqueous dispersion as well as in several organic solvents

[11–19]. Liquid crystalline GO dispersions have already been

utilized in a number of areas including carbon based fiber

spinning, alignment of one-dimensional nanomaterials, and

multifunctional composites, liquid crystal displays, etc [13–

19]. Manipulation of atomic thick platelet ordering by external

stimuli or dynamic perturbations may provide unprecedented

routes to highly ordered carbon based materials assembled

from genuine two-dimensional building blocks.

Processing of liquid crystalline materials frequently

encounters technological issues involved with complex mul-

tiscale structures and dynamics. For many applications,

454 C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1

liquid crystalline phase formation and corresponding visco-

elastic properties play significant roles to determine the

material structures and properties [20–22]. In the particular

case of GO liquid crystals, Gao and Xu reported the shear vis-

cosity reduction at isotropic to nematic phase transition as an

evidence for LC phase formation [23]. Yong and co-workers

reported typical shear thinning and relevant shear alignment

behavior of GO liquid crystals. [24]. Nonetheless, detailed rhe-

ological properties of GO liquid crystals have not been inves-

tigated thus far, despite their crucial significance for

optimized processing of high-performance graphene-based

materials [23–25]. The aim of this work is to investigate on

the various aspects of the viscoelastic properties of GO liquid

crystals. Thorough investigation over a broad GO composi-

tional range up to A = 1.8% reveals the typical rheological

behaviors involved with the isotropic-to-nematic phase tran-

sition and nematic gel formation.

2. Experimental

GO dispersion was prepared by the oxidative exfoliation of

graphite powder (Bay Carbon, SP1 Graphite powder; lot #

060706), following modified Hummers method, as described

elsewhere [11]. Briefly, 1 g of graphite powder was oxygenated

in 40 mL H2SO4, while stirring for 15 min, which was followed

by slow adding of 3.5 g potassium permanganate under an ice

bath. The mixture was then continuously stirred for 2 h at

35 �C. After completion oxygenation, the excess amount of

deionized water and 35% H2O2 were added to the mixture.

The obtained yellow mixture was thoroughly filter washed

with 1 M HCl solution and deionized water and re-dispersed

in 1 L of deionized water. The monolayer exfoliation was

achieved by the sonication of the washed GO dispersion in

water bath ultrasonicator. Subsequent purification was per-

formed by dialysis and centrifugation to remove ionic impuri-

ties and unexfoliated graphite oxides. A predetermined

amount of the concentrated dispersion was diluted in deion-

ized water to prepare the GO dispersions with a desired

composition.

To characterize the GO sheet size and thickness, extremely

diluted GO dispersion was dried at Si substrate surface.

Atomic force microscopy (SPA400) image was recorded in

the non-tapping mode under ambient conditions. Orienta-

tional arrangement of GO sheets was characterized by FE-

SEM (Hitachi S-4800 SEM). Optical birefringence of GO liquid

crystal dispersion was observed with light microscopy

between crossed polarizers. Rheological characterization of

GO dispersions was performed on a stress controlled

Advanced Rheometer (Bohlin Instrument) in both steady

and dynamic mode. All measurements were performed with

cone-plate configuration (radius: 44 mm, cone angle: 4�) with

truncated gap of 150 lm. In the steady-state measurements,

shear viscosity and shear stress were monitored as function

of shear rate (from 0.01 to 100 s�1). The shear rate was

increased continuously with the 30 s integration time for each

shear rate. Dynamic frequency sweep tests were performed

from 0.1 to 100 rad/s. All samples are compressed slowly to

the set gap and allowed to equilibrate for 10 min prior to each

measurement. All measurements were performed at room

temperature (20 �C).

3. Results and discussion

3.1. Characterization of GO liquid crystal dispersion

Fig. 1a shows the SEM image of as-synthesized GO platelets

with the mean lateral size of �700 nm. AFM image of GO

sheets with a height profile is shown in Fig. 1b, which con-

firms the monolayer thickness of �1 nm. The structure and

properties of GO dispersions are strongly dependent on the

interactions between the platelets. The steric hindrance from

overlapping of GO platelets arises at high composition, results

in configurational entropy driven excluded-volume effects

[12]. Well-known Onsager’s model accounts for the formation

of nematic liquid crystals based on purely entropy-driven

mechanism [26]. The thermodynamic equilibrium liquid crys-

tal phase originates from the competition between orienta-

tional and positional entropies. In dilute dispersion, the

orientational entropy is maximized due to random orienta-

tion of GO platelets. By contrast, above critical GO concentra-

tion, the GO platelets orientate parallel to each other in order

to minimize the excluded volume of the approaching plate-

lets. Thus, the parallel arrangement of anisotropic GO plate-

lets maximizes the packing or positional entropy that

governs the formation of nematic liquid crystalline GO disper-

sions. The liquid crystalline phase can be easily identified

from polarized optical microscopy. GO dispersions does not

show any optical birefringence at low volume fraction, /

� 0.08%. Biphasic dispersion is observed between 0.15% and

0.45%. Entire nematic phase is confirmed above a volume

fraction of 0.45%. Typical optical birefringent texture of single

nematic LC phase for / � 0.45% is shown in Fig. 1c. This bire-

fringence texture indicates the characteristic of anisotropic

phases originates from the alignment of anisotropic platelets.

The liquid crystalline GO dispersions are further confirmed

from freeze-dried sample. Aqueous dispersion of / = 0.45%

quenched in liquid N2 and subsequently freeze-dried. The

freeze-dried sample was subjected to SEM observation.

Fig. 1d clearly demonstrates the ordered GO platelets in the

nematic phase.

3.2. Rheological study

3.2.1. Steady shear flowMolecular ordering is particularly significant in the systems of

anisotropic colloidal particles, polymers or worm-like

micelles, where external shear can readily induce alignment

and orientational order. To understand the orientational

behavior of GO dispersions under shear, steady state shear

flow curves were measured. Fig. 2a shows the evolution of

shear viscosity against shear rate for various GO composi-

tions. GO dispersion exhibits Newtonian and non-Newtonian

(shear thinning) behavior strongly depend on the /. At low /

= 0.08%, shear viscosity shows Newtonian behavior at low

and high shear rate with a shear thinning region at interme-

diate range. At a high GO composition, GO dispersion exhibits

typical shear thinning behaviors. Moreover, the viscosity is

found to show non-monotonic behavior with composition,

which will be discussed later. Observed shear thinning is

associated to the orientation of GO platelets in nematic dis-

persion. The flow behavior can be approximated by power

(a) (b)

(c) (d)

Fig. 1 – (a) SEM image of synthesized GO flakes deposited on Si substrate. (b) AFM image of GO sheets with height profile

showing the typical monolayer thickness of �1 nm. (c) Polarized light microscopy of GO aqueous dispersion showing

optically birefringent texture. (d) Freeze dried SEM image of GO liquid crystalline phase. (A colour version of this figure can be

viewed online.)

C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1 455

law and Curreau shear thinning models. In power law shear

thinning behavior (g � cx�1) with g the viscosity, k a constant,

the shear rate and x the power law exponent [27]. The expo-

nent x = 1 is retrieved for Newtonian flow and x < 1 is for

shear thinning fluids. In Fig. 2a, we have fitted the power

law in the low shear rate regime (0.01–1 s�1). The exponents,

x decrease from 0.9 to 0.29 with volume fraction (see Table 1).

The exponent, x = 0.9 matches well with the one for isotropic

dispersion at / = 0.08% with Newtonian behavior, while

x = 0.29 for / = 1.8% corresponds the typical values of weak

gels or concentrated dispersions. The shear thinning over

the entire shear rate range can be also described by the sim-

ple Curreau equation [28], gðcÞ � gðc! 0Þ=½1þ ðsrcÞ2�m; where

m, and sr define the shear thinning exponent and orienta-

tional relaxation time in the system, respectively. We find that

shear thinning exponent of m = 0.38 for / = 0.08%, is close to

liquid-like (isotropic) behavior. By contrast, exponent,

m = 0.61 for / = 1.8 close to highly entangled polymer melts

and concentrated solutions [29]. Moreover, relaxation time,

sr increase from 4.5 to 45 s when / increase from 0.08% to

1.8%. This increase in relaxation time of GO platelets should

manifest the maximization of packing and enhanced orienta-

tion of GO platelets along the shear direction.

Viscosity behavior of liquid crystalline dispersions

depends not only on the GO composition but also on molecu-

lar arrangements in the dispersion. Fig. 2b shows the evolu-

tion of shear viscosity with volume fraction, / for shear

rates of 0.01, 8.5, and 92 s�1. Interestingly, viscosity exhibits

non-monotonic behavior with increasing /. At a low fraction

of A = 0.08, GO dispersion is in isotropic phase. Afterwards,

viscosity increased up to a critical fraction (/c � 0.33) and

get maxima. With the further increase in /c, viscosity goes

down and exhibits minima before increasing again. This vis-

cosity reduction above a critical volume fraction, /c � 0.33, is

caused by the formation of low-viscosity nematic liquid crys-

talline phase. The viscosity minima approximately corre-

spond to the GO composition for entire nematic phase

formation. The transition from isotropic to nematic phase

can be observed for the GO composition range between the

maxima and minima in shear viscosity [30]. In this work,

/c = 0.33% is considered as the critical GO composition for iso-

tropic–nematic phase transition, which is close to theoretical

predicated value (0.25%) for polydispersed infinitely thin

platelets [31]. We note that this GO liquid crystal phase tran-

sition behavior agrees well with our previous optical charac-

terization with polarized optical microscopy [11].

The shear effect on GO liquid crystal dispersion is further

described by shear stress-shear rate flow curve (Fig. 3). The

flow curves were measured for a wide range of GO composi-

tion from 0.08% to 1.8%. The representative shear stress-shear

rate plots for / = 0.08% and 0.83% are presented. At / = 0.08%,

shear stress varies almost linearly up to the shear rate of 1 s�1

with a constant viscosity (see Fig. 2a), which confirm the typ-

ical Newtonian behavior in isotropic phase. As shear rate

increases, horizontal stress plateau are observed for both

samples, where the constant shear stress values are

Fig. 2 – (a) Evolution of shear viscosity with shear rate at different GO composition. (b) Non-monotonic constitutive relation

between shear viscosity vs GO composition demonstrating isotropic to nematic phase transition.

Table 1 – Power law (g � cx�1), and Carreau equationðgðcÞ � gðc! 0Þ=½1þ ðsrcÞ2�

mÞ fitted parameters.

/ (vol%) x sr/s m

0.08 0.9 4.4 0.380.16 0.77 15 0.300.33 0.53 24 0.520.45 0.68 22 0.330.83 0.41 33 0.531.20 0.28 45 0.561.80 0.29 42 0.61

456 C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1

measured from 1 to 18 s�1 for / = 0.08% and from 0.1 to 5.9 s�1

for / = 0.83%, respectively. For / = 0.83% we performed mea-

surements decreasing the shear rate, as shown in inset of

Fig. 3. As arrows indicate sheer rate change direction, hyster-

esis is clearly observed. The observed stress plateaux and

hysteresis loops in steady state flow curves are often associ-

ated to the typical bistability of various soft materials, includ-

ing aqueous surfactant solutions, emulsions, and liquid

crystals to concentrated colloidal suspensions [32–36]. In

liquid crystalline materials, such bistability can be linked to

the phenomenon of shear banding; separation of the flow in

high and low shear regions. Under a shear flow, many com-

plex fluids have been reported to form macroscopic bands,

perpendicular to the flow direction at different shear rate lev-

els. In liquid crystalline systems, an ordered phase co-exits

with a disordered one in the shear-banding regime. Our obser-

vation of such a bistability can be associated to the different

orientational states of GO platelets in the direction of shear

flow as well as banded structure formation. In addition to

shear banding GO liquid crystal also exhibits yield stress (ry)

behavior. This steady shear yield stress can be estimated from

the observed stress plateau (Fig. 3), where shear stress starts

again increase as shear rate further increases. The observed

yield stress is plotted as a function of GO composition along

with dynamic yield stress in the inset of Fig. 4b.

Fig. 3 – Shear stress plotted against shear rate. The inset shows hysteresis observed when the shear stress was measured

with increasing and decreasing the shear rate for / = 0.83%.

C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1 457

3.2.2. Dynamic characteristicsTo establish the limits of the linear viscoelastic regime (LVR),

where dynamic moduli are independent on the applied stress,

oscillatory stress sweep measurements are performed. We

note that non-Newtonian systems such as polymer solution,

concentrated colloidal suspensions, and liquid crystalline dis-

persion do not show true LVR. For practical measurement, lin-

ear viscoelastic regime can be defined as the region in which

both G 0 and G00 are nearly constant with stress. Fig. 4a shows

the typical stress sweep measurements for liquid crystalline

GO dispersion of / = 1.2%, where both G 0 (solid) and G00 (open)

moduli are nearly constant up to a critical stress of 1.27 Pa.

Above the critical value, both moduli become stress depen-

dent, G 0 begin to decrease due to rupture of the GO network

and eventually cross G00. The crossover of G 0 and G00 can be

employed to estimate the yield stress (ry) of GO dispersions

[37]. Fig. 4b shows the strong dependency of yield stress on

GO composition. Along with the increase of /, ry first moves

to high values (/c = 0.33) and then decreases at single nematic

phase fraction (/N = 0.45) and finally increases again to higher

values. This intermittent decrease in yield stress at single

nematic phase composition is caused by the alignment or ori-

entation of GO platelets that reduces the entanglements

among platelets. The yield stress values are plotted against

GO composition up to / P 0.45 in the inset of Fig. 4b. Simple

scaling argument; ry � am is used to quantitatively link the /

dependence of the yield stress with m = 2.86 and 3.14 for

dynamic and steady yield stresses, respectively. Similar scal-

ing exponents have been reported for discotic laponite gels

and concentrated yield stress fluids [38].

Frequency-dependent measurements for the storage G 0(x)

and loss G00(x) moduli, exhibit typical liquid- or solid-like

behavior for volume fraction ranged from / = 0.08% to 1.8%

in the linear viscoelastic regime at constant stress of 0.1 Pa.

Fig. 5a shows the typical frequency sweep measurements

for the GO dispersion with / = 0.08, 0.45, and 1.8%. For /

= 0.08%, both moduli, G 0 (solid symbol) and G00 (open symbol)

have very low amplitudes and fluctuate continuously with

frequency. By contrast, for / > 0.08%, G 0 and G00 increase

monotonically up to a critical frequency and exhibits a sharp

minima before fluidization. This observation of minima

(dynamic instability) will be discussed later. We further

observed that G 0 and G00 become independent of frequency

as / P 0.83%. Low-frequency G 0 and G00 moduli can be

described by the power-law relationship, G(x) � kxn, where

the exponent n is the magnitude of slope in the double loga-

rithmic plot and k is the intercept which defines the degree of

network connectivity [39,40]. The linear viscoelasticity model

proposed by Winter [41] predicts that the stress relaxation fol-

lows the power-law frequency dependence G 0 � xn with

0 < n < 1. Stoichiometric balanced and imbalanced cross-

linked networks showed n < 0.5 in the presence of excess

cross-linker, and n > 0.5 in the short of cross-linker, respec-

tively. This description strictly applies to chemically cross-

linked gels. It has been known that n > 0 for a physical (weak)

gel, whereas n = 0 for a chemical (strong) gel [42,43]. The expo-

nent n can therefore be used as a quantitative measure of the

resemblance of an elastic network system. In viscoelastic

measurements, the liquid-like behavior is governed by G 0 / x2

and G00 / x. The data in Fig. 5a are fitted with the aforemen-

tioned power law and the obtained exponent values are plot-

ted as a function of / (Fig. 5b). At low /, both G 0 and G00 are of

the same order of magnitude and exhibits frequency depen-

dence, G 0 � x1.65 and G00 � x0.79, confirms the isotropic phase.

At / P 0.33, G 0 exceeds over G00 across the low frequency

range and indicates that GO sheets relaxation time is longer

than experimentally accessible value, exhibiting nematic

phase (solid-like) behavior, whereas n = 0.28 for / = 0.45%.

For even higher volume fraction of / = 1.8%, G 0 and G00 are

independent of frequency with very low exponent of

n = 0.05, confirming the gel-like phase formation. It is note-

worthy that our finding is qualitatively similar to previous

Fig. 4 – (a) Dynamic oscillatory stress sweep measurements for / = 0.16% (sphere), and / = 0.83% (triangle) GO dispersions.

Solid symbol for storage (G 0) and open symbol for loss (G00) moduli. (b) Double logarithmic plot of estimated yield stress as a

function of GO composition. Inset shows the double logarithmic plot of dynamic yield stress (solid sphere) and steady shear

yield stress (open sphere) for fully nematic GO dispersions.

458 C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1

observations of aqueous CNT dispersions [44–46]. Taken

together, we may conclude that the elasticity arises from

GO platelet network and nematic ordering of GO platelets.

Our results are consistent with the recent observations of

gel-like phase formation of aqueous GO dispersion at high

composition [47,48]. The formation of GO liquid crystalline

gel or three-dimensional network is attributed to the large

and flexible two-dimensional structure of GO platelets and

the force balance between their static repulsion and bonding

interaction. Increasing the bonding force or decreasing the

repulsion force between GO sheets in dispersion will reinforce

the gel network. In GO liquid crystalline dispersions, two sorts

of inter-sheets forces work together; repulsive forces due to

the negatively charged functional groups (oxygenated func-

tional group) and attractive forces arising from p–p interaction

among the graphitic patches and vander Walls interaction. At

a high GO composition, the strength of attractive forces may

dominant due to the small inter-sheets separation. This

resulted in strong sheets interlinked GO gel-like phase

formation.

We have above discussed that GO dispersion exhibits

liquid crystalline and thixotropic fluid like features. This

thixotropic behavior may arise from the breakup of intercon-

nected network and the enhanced orientational ordering of

Fig. 5 – (a) Frequency sweep measurements for / = 0.08% (circle), 0.33% (triangle) and 1.2% (square). Solid and open symbol

represents for storage and loss modulus, respectively. (b) Variation of scaling exponents (n) as a function of GO composition.

(c) Storage modulus (G 0) vs. GO composition. Inset shows the double logarithmic plot of G00 vs. reduced volume fraction (/ � /*)

obtained at angular frequency (x) of 0.01 rad/s. (d) GO composition dependent frequency minima where instability occur in

storage (G 0, open sphere), loss (G00, solid sphere) moduli.

C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1 459

GO platelets. In frequency sweep measurements elastic stor-

age modulus at low frequencies defined as plateau modulus

G00 is plotted as a function of / as shown in Fig. 5c. The notice-

able increase of G00 more than five orders of magnitude

between / = 0.16% and 0.33% is due to rigidity percolation.

Near the percolation threshold, G00 can be correlated to the

volume fraction / through a power law relationship as

G00 / u� u0ð Þp; where /* is the percolation volume fraction,

and p is the percolation exponent. We obtain percolation vol-

ume fraction of /* = 0.30 and exponent of p = 1.32 ± 0.08 by

varying /* to minimize the error in the slope of a double log-

arithmic plot of G00 vs (/ � /*) (see inset of Fig. 5c). This power

law increase in the elastic modulus is possible due to the

nematic ordering of GO platelets. The percolation exponent

of 1.32 obtained in current work is comparable to single-

walled, multi-walled nanotube dispersions and their polymer

nanocomposites [46,49–51]. A comparison of our measured

value of p = 1.32 ± 0.08, with previously speculated three-

dimensional elastic percolated systems, shows that the

inter-particle interaction for GO platelets is weak. This obser-

vation of rigidity percolation transition in GO liquid crystal-

line dispersions should be useful for the analysis of GO

fillers in a polymer matrix to estimates the percolation

threshold.

Lastly, we found the interesting dynamic instabilities in

GO liquid crystalline dispersion at a critical frequency, which

strongly depends on the GO composition. The dynamic insta-

bilities in a fluid mechanics are responsible for a variety of

important common phenomena, such as waves on the sea

surface or Taylor vortices in Couette flow [52]. Observation

of distinct minima in the frequency sweet test (G 0, G00) dis-

played in Fig. 5a signifies existence of dynamic instability.

As obvious from the Fig. 5a, the critical instability frequency

shifts to the higher values with the increase of GO composi-

tion (Fig 5d). The mechanism for such kind of instability has

not been clearly understood, however, there are few reports

on the oscillatory flow induced instabilities for other colloidal

systems. Torralba et al. [53] have reported that oscillatory flow

of the Maxwell fluid is unstable at high driving frequencies. In

other report, the binary mixture under horizontal vibrations

have revealed the existence of a new dynamical instability

due to size segregation under oscillatory shear and its con-

nections to pattern formation in granular flows [52]. Our find-

ing of dynamic instability should be significantly considered

in the processing of GO based materials.

4. Conclusions

We have reported that aqueous GO dispersion exhibit the

well-defined LC phases and gel-like behavior based on

rheological measurements. The GO dispersion exhibit shear

460 C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1

thinning behavior, which is well described by power law and

Carreau models. Concentration dependent shear viscosity

confirms the typical colloidal isotropic–nematic phase transi-

tion at /c = 0.33%. Viscoelastic measurements revealed that

liquid crystals and gel-like phase exhibit well-defined elastic

shear modulus and yield stress, which increase with GO com-

position following a power law. Moreover, GO liquid crystal-

line dispersion shows the rigidity percolation transition

with an onset of solid-like elasticity at /* = 0.3% and the per-

colation exponent of 1.32 ± 0.08. Our observations of strongly

nonlinear rheology under flow (shear thinning, banding and

dynamic instabilities) demonstrate that precise rheological

characterization of GO liquid crystalline dispersion is

required for their processing into desired structures and

properties.

Acknowledgment

This work is financially supported by Institute for Basic Sci-

ence (IBS) [IBS-R004-G1-2014-a00], and Asian Office of Aero-

space Research and Development (AOARD) [AOARD-14-1403].

R E F E R E N C E S

[1] Bisoyi HK, Kumar S. Liquid-crystal nanoscience: an emergingavenue of soft self-assembly. Chem Soc Rev 2011;40:306–19.

[2] Lagerwall JPF, Scalia G. A new era for liquid crystal research:applications of liquid crystals in soft matter nano-, bio- andmicrotechnology. Curr Appl Phys 2012;12:1387–412.

[3] Kaafarani BR. Discotic liquid crystals for opto-electronicapplications. Chem Mater 2011;23:378–96.

[4] Yan J, Rao L, Jiao M, Li Y, Cheng H-C, Wu S-T. Polymer-stabilized optically isotropic liquid crystals for next-generation display and photonics applications. J Mater Chem2011;21:7870–7.

[5] Zhang S, Kumar S. Carbon nanotubes as liquid crystals. Small2008;4:1270–83.

[6] Ko H, Tsukruk VV. Liquid-crystalline processing of highlyoriented carbon nanotube arrays for thin-film transistors.Nano Lett 2006;6:1443–8.

[7] Zakr C, Blance C, Grelet E, Zamora-Ledezma C, Puech N,Anglaret E, et al. Liquid crystals of carbon nanotubes andgraphene. Philos Trans R Soc Lond B Biol Sci 2013;371.20120499-16.

[8] Zakri C. Carbon nanotubes and liquid crystalline phases. LiqCryst Today 2007;16:1–11.

[9] Lee SH, Lee DH, Lee WJ, Kim SO. Tailored assembly of carbonnanotubes and graphene. Adv Funct Mater 2011;21:1338–54.

[10] Maiti UN, Lee WJ, Lee JM, Oh Y, Kim JY, Kim JE, et al.Chemically modified/doped carbon nanotubes & graphenefor optimized nanostructures & nanodevices. Adv Mater2014;26:40–67.

[11] Kim JE, Han TH, Lee SH, Kim JY, Ahn CW, Yun JM, et al.Graphene oxide liquid crystals. Angew Chem Int Ed2011;50:3043–7.

[12] Aboutalebi SH, Gudarzi MM, Zheng QB, Kim JK. Spontaneousformation of liquid crystals in ultralarge graphene oxidedispersions. Adv Funct Mater 2011;21:2978–88.

[13] Jalili R, Aboutalebi SH, Esrafilzadeh D, Konstantinov K,Moulton SE, Razal JM, et al. Organic solvent-based grapheneoxide liquid crystals: a facile route toward the next

generation of self-assembled layer-by-layer multifunctional3D architectures. ACS Nano 2013;7:3981–90.

[14] Xu Z, Gao C. Graphene chiral liquid crystals and macroscopicassembled fibres. Nat Commun 2011;2:571–9.

[15] Jalili R, Aboutalebi SH, Lzadeh DE, Shepherd RL, Chen J,Aminorroaya-Yamini S, et al. Scalable one-step wet-spinningof graphene fibers and yarns from liquid crystallinedispersions of graphene oxide: towards multifunctionaltextiles. Adv Funct Mater 2013;23:5345–54.

[16] Li Y, Wu Y. Coassembly of graphene oxide and nanowires forlarge-area nanowire alignment. J Am Chem Soc2009;131:5851–7.

[17] Maiti UN, Lim J, Lee KE, Lee WJ, Kim SO. Three-dimensionalshape engineered, interfacial gelation of reduced grapheneoxide for high rate, large capacity supercapacitors. Adv Mater2014;26:615–9.

[18] Kim JY. Kim SO liquid crystals: electric fields line up grapheneoxide. Nat Mater 2014;13:325–6.

[19] Shim J, Yun JM, Yun T, Kim P, Lee KE, Lee WJ, et al. Two-minute assembly of pristine large-area graphene based films.Nano Lett 2014;14:1388–93.

[20] Kharchenko SB, Douglas JF, Obrzuti J, Grulke EA, Migler K.Flow-induced properties of nanotube-filled polymermaterials. Nat Mater 2004;3:564–8.

[21] Theron SA, Zussman E, Yarina AL. Experimentalinvestigation of the governing parameters in theelectrospinning of polymer solutions. Polymer2004;45:2017–30.

[22] Tan Y, Song YY, Zheng Q. Hydrogen bonding-drivenrheological modulation of chemically reduced grapheneoxide/poly(vinyl alcohol) suspensions and its application inelectrospinning. Nanoscale 2012;4:6997–7005.

[23] Xu Z, Gao C. Aqueous liquid crystals of graphene oxide. ACSNano 2011;5:2908–15.

[24] Yang X, Guo C, Ji L, Li Y, Tu Y. Liquid crystalline and shear-induced properties of an aqueous solution of graphene oxidesheets. Langmuir 2013;29:8103–7.

[25] Korkut S, Roy-Mayhew JD, Dabbes DM, Milius DL, Aksay IA.High surface area tapes produced with functionalizedgraphene. ACS Nano 2011;5:5214–22.

[26] Onsager L. The effects of shape on the interaction of colloidalparticles. Ann N Y Acad Sci 1949;51:627–59.

[27] Bird RB, Armstrong RC, Hassager O. Dynamics ofpolymeric liquids, fluid mechanics, vol. 1. NewYork: Wiley; 1987.

[28] Tanner RI. Engineering rheology. New York: OxfordUniversity Press; 2000.

[29] Douglas JF. Shift in polymer blend phase-separationtemperature in shear flow. Macromolecules1992;25:1468–74.

[30] Kiss GD. Rheology and rheo-optics of concentrated solutionsof helical polypeptides [Ph.D. thesis]. University ofMassachusetts; 1979, p 308.

[31] Bates MA, Frenkel D. Nematic–isotropic transition inpolydisperse systems of infinitely thin hard platelets. J ChemPhys 1999;10:6553–9.

[32] Bonn D, Meunier J. Bistability in non-Newtonian flow:rheology of lyotropic liquid crystals. Phys Rev E1998;58:2115–8.

[33] Yuan X-F, Jupp L. Interplay of flow-induced phase separationsand rheological behavior of complex fluids in shear bandingflow. Europhys Lett 2002;60:691–7.

[34] Wunenburger AS, Colin A, Leng J, Arneodo A, Roux D.Oscillating viscosity on a lyotropic lamellar phase undershear flow. Phys Rev Lett 2001;86:1374–7.

[35] Eiser E, Molino F, Porte G. Nonhomogeneous textures andbanded flow in a soft cubic phase under shear. Phys Rev E2000;61:6759–64.

C A R B O N 8 0 ( 2 0 1 4 ) 4 5 3 – 4 6 1 461

[36] Spenley NA, Cates ME, McLeish TC. Nonlinear rheology ofwormlike micelles. Phys Rev Lett 1993;71:939–42.

[37] Shih WY, Shih W-H, Aksay IA. Elastic and yield behavior ofstrongly flocculated colloids. J Am Ceram Soc 1999;82:616–24.

[38] King Jr HE, Milner ST, Lin MY, Singh JP, Mason TG. Structureand rheology of organoclay suspensions. Phys Rev E2007;75:021403–20.

[39] Jampala S, Manolache S, Gunasekaran S, Denes F. Plasma-enhanced modification of xanthan gum and its effect onrheological properties. J Agric Food Chem 2005;53:3618–25.

[40] Kim C, Yoo B. Rheological properties of rice starch–xanthangum mixtures. J Food Eng 2006;75:120–8.

[41] Winter HH. Can the gel point of a cross-linking polymer bedetected by the G 0 – G00 crossover? Polym Eng Sci1987;27:1698–702.

[42] Ross-Murphy SB, Shatwell KP. Polysaccharide strong andweak gels. Biorheology 1993;30:217–27.

[43] Ross-Murphy SB. Structure-properties relationships in foodbiopolymer gels and solutions. J Rheol 1995;39:1451–63.

[44] Davis VA, Ericson LM, Parra-Vasquez NAG, Fan H, Wang Y,Prieto V, et al. Phase behavior and rheology of SWNTs insuperacids. Macromolecules 2004;37:154–60.

[45] Hobbie EK, Fry DJ. Rheology of concentrated carbon nanotubesuspensions. J Chem Phys 2007;126. 124907-7.

[46] Hough LA, Islam MF, Janmey PA, Yodh AG. Viscoelasticity ofsingle wall carbon nanotube suspensions. Phys Rev Lett2004;93:168102–4.

[47] Bai H, Li C, Wang X, Shi G. On the gelation of graphene oxide.J Phys Chem C 2011;115:5545–51.

[48] Dan B, Behabtu N, Martinez NA, Evans JS, Kosynkin DV, TourJM, et al. Liquid crystals of aqueous, giant graphene oxideflakes. Soft Matter 2011;7:11154–9.

[49] Du F, Scongna RC, Zhou W, Brand S, Fischer JE, Winey KI.Nanotube networks in polymer nanocomposites: rheologyand electrical conductivity. Macromolecules 2004;37:9048–55.

[50] Sabzi M, Jiang L, Liu LF, Ghasemi I, Atai M. Graphenenanoplatelets as poly(lactic acid) modifier: linear rheologicalbehavior and electrical conductivity. J Mater Chem A2013;1:8253–61.

[51] Sahimi M, Arbabi S. Mechanics of disordered solids. II.Percolation on elastic networks with bond-bending forces.Phys Rev B 1993;47:703–11.

[52] Ciamarra MP, Coniglio A, Nicodemi M. Shear instabilities ingranular mixtures. Phys Rev Lett 2005;94:188001–4.

[53] Torralba M, Castrejon-Pita AA, Hernandez G, Huelsz G, del RıoJA, Ortın J. Instabilities in the oscillatory flow of a complexfluid. Phys Rev E 2007;75:056307–56309.