4800 Langmuir 2008, 24, 4800-4805

Conducting Film from Graphite Oxide Nanoplatelets and

Poly(acrylic acid) by Layer-by-Layer Self-Assembly
Jihuai Wu,* Qunwei Tang, Hui Sun, Jianming Lin, Haiyong Ao, Miaoliang Huang, and
Yunfang Huang
The Key Laboratory for Functional Materials of Fujian Higher Education, Institute of Material Physical
Chemistry, Huaqiao UniVersity, Quanzhou 362021, China

ReceiVed January 10, 2008. In Final Form: February 8, 2008

A [poly(acrylic acid)/graphite oxide]n [(PAA/GO)n] film with a conductivity of 60 S‚cm-1 was grown layer-by-layer
(LbL) using Langmuir-Blodgett self-assembly techniques. GO nanoplatelets were prepared from natural graphite by
oxidizing, ball milling, exfoliating, and modifying with cationic surfactant cetyltrimethylammonium bromide (CTAB).
The X-ray diffraction pattern reveals that PAA and GO stack orderly LbL and repeatedly in the (PAA/GO)n films,
and about three carbon molecular layers are superposed on each GO sheet. Fourier transform infrared spectra offer
evidence for the interaction between the carboxylic groups on PAA and the CTAB on the surface of the GO nanoplatelets.
Electrochemistry measurements show that the conductivity of the (PAA/GO)n film depends on the carbon-carbon
interlayer height of the GO sheet, and the (PAA/GO)n film has a typical positive temperature coefficient effect above
the PAA melting temperature. The atomic force microscopy images reveal that CTAB molecules stack in a well-
ordered head-to-head structure on both surfaces of the GO nanoplatelets and the GO nanoplatelets are embeded
between PAA layers.

Introduction surface layer of hemicylindrical micelles13-15 (shown in Figure

The thin film composites prepared by self-assembly have been
investigated widely because of their unique properties1,2 in
comparison with those of conventional composites. Typically,
the thin film based on graphite oxide (GO) nanoplatelets is
expected to be an excellent conducting material, which is obtained
by the wet chemical layer-by-layer (LbL) self-assembly of
oppositely charged polyelectrolytes and GO nanoplatelets.2-9
However, it is believed that the carboxyl group in poly(acrylic
acid) (PAA) cannot fix a cationic polyelectrolyte10,11 and is located
only at the edge of GO carbon layers.12
With the aim of enhancing the electrostatic forces and
improving the stability of poly(acrylic acid)/graphite oxide
composite (PAA/GO)n film, we introduced cationic surfactant
cetyltrimethyl-ammonium bromide (CTAB) onto the surface of
GO nanoplatelets and expected to obtain a highly conducting
(PAA/GO)n film. PAA shows negative charge and hydrophilicity,
and GO shows a weak negative charge and hydrophobicity in
solution. When the GO nanoplatelets are modified with cationic
surfactant CTAB, both surfaces of hydrophobic GO nanoplatelets
are covered with alkyl chains and undergo the replacement of
a large number of hydrophilic ammonium ions situated on the
* Corresponding author. E-mail: jhwu@hqu.edu.cn. Tel: +86-595-
22693899. Fax: +86-595-22693999.
(1) Li, H.; Park, S.; Reif, J.; LaBean, T.; Yan, H. J. Am. Chem. Soc. 2004, 126,
418-419.
(2) Kovtyukhova, N.; Ollivier, P.; Martin, B.; Mallouk, T. Chem. Mater. 1999,
11, 771-778.
(3) Du, X.; Xiao, M.; Meng, Y.; Hay, A. Carbon 2005, 43, 195-213.
(4) Szabo, T.; Szeri, A.; Dekany, I. Carbon 2005, 43, 87-94.
(5) Liu, Z.; Wang, Z.; Yang, X.; Ooi, K. Langmuir 2002, 18, 4926-4932.
(6) Xu, J.; Hu, Y.; Song, L.; Wang, Q.; Fan, W. Carbon 2002, 40, 2961-2973.
(7) Sun, J.; Wu, T.; Liu, F.; Wang, Z.; Zhang, X.; Shen, J. Langmuir 2000,
16, 4620-4624.
(8) Fery, A.; Scholer, B.; Cassagneau, T.; Caruso, F. Langmuir 2001, 17,
3779-3783.
(9) Huang, S.; Artyukhin, A.; Wang, Y.; Ju, J. J. Am. Chem. Soc. 2005, 127,
14176-14177.
(10) Mermoux, M.; Chabre, Y.; Rousseau, A. Carbon 1991, 29, 469-474.
(11) Lerf, A.; He, H.; Riedl, T.; Forster, M.; Klinowsk, I. J. Solid State Ionics
1997, 101-103, 857-862.
(12) Szabo, T.; Tombacz, E.; Illes, E.; Dekany, I. Carbon 2006, 44, 537-545.
1). The surface of modified GO nanoplatelets is changed to be
electronegative and hydrophilic and can interact with PAA by
electrostatic forces and a polar effect; therefore, LbL self-assembly
from PAA and modified GO nanoplatelets can be successfully
realized.
Experimental Section
Materials. CTAB (C16H33N+(CH3)3Br-) (purity >99%, cation
exchange capacity (CEC) ) 0.9 mM16) was purchased from Aldrich
and used without further purification. The graphite used to prepare
the oxide graphite was natural flake graphite ultrafine powder with
an average size of <2 µm as proven by the SKN-1000 particle sizer:
its ash content was <0.7 wt % as measured by a thermoanalytical
measurement (supplied by Shandong Qingdao Graphite Company,
China). The natural flake graphite powder was vacuum dried at
80 °C for 24 h before use. Concentrated sulfuric acid and fuming
nitric acid were used as oxidizing and intercalating agent to prepare
expanded GO nanoplatelets.17 Acrylic acid monomer (AA) was
distilled under reduced pressure prior to use. Ammonium persulfate
(APS) purified by recrystallization from a 66 wt % ethanol/water
solution was a radical initiator for the synthesis reaction of PAA.
N,N′-Methylene bisacrylamide (NMBA) was used as a cross-linker
for the preparation of PAA. The above materials were all purchased
from Shanghai Chemical Reagents Co., China.
Exfoliation of GO and Modification by CTAB. Expanded GO
was prepared according to the method in refs 18 and 19. Briefly,
under appropriate cooling and stirring, a mixed solution of
concentrated sulfuric acid and fuming nitric acid (4:1 v/v) was slowly
added to a three-necked flask containing natural flake graphite. After
(13) Xu, S.; Wang, C.; Zeng, Q.; Wu, P.; Wang, Z.; Yan, H. Langmuir 2002,
18, 657-660.
(14) Bandyopadhyay, S.; Shelley, J.; Tarek, M.; Moore, P. J. Phys. Chem. B
1998, 102, 6318-6322.
(15) Srinivas, G.; Nislsen, S.; Moore, P.; Klein, M. J. Am. Chem. Soc. 2006,
128, 848-853.
(16) Manne, S.; Cleveland, J.; Gaub, H.; Stucky, G.; Hansma, K. Langmuir
1994, 10, 4409-4413.
(17) Chen, G.; Wu, D.; Weng, W.; Wu, C. Carbon 2003, 41, 619-621.
(18) Nakajima, T.; Matsuo, Y. Carbon 1994, 32, 469-475.
(19) Ramesh, P.; Sampath, S. Anal. Chem. 2003, 75, 6949-6957.

10.1021/la800095z CCC: $40.75 © 2008 American Chemical Society

Published on Web 03/22/2008
Conducting (PAA/GO)n Film
Figure 1. Structure of modified graphite oxide and poly(acrylic acid).
Figure 2. Schematic representation of the LbL self-assembly of the (PAA/GO)n film.
reacting for 16 h, the mixture was carefully diluted with distilled
water, filtered, and washed until the pH of rinsewater became neutral.
The filtered product was dried at 100 °C overnight, and the resulting
preoxidized graphite was subjected to a thermal shock at 1050 °C
for 20 s in a muffle furnace to form expanded GO.
The expanded GO was exfoliated by using a ball milling method
as follows: 2 g of expanded GO was saturated with 300 mL of an
aqueous alcohol solution (water/alcohol ) 1:1 v/v) for 12 h. Then
the suspension was milled for 8 h in a high-energy planetary ball
mill at 600 r‚min-1, which resulted in the exfoliation of the expanded
GO and the formation of GO nanoplatelets (with about three carbon
molecular layers in each nanoplatelet). The slurry was centrifugated
and vacuum dried at 70 °C, and the resulting GO nanoplatelets were
kept in a desiccator before use.
The GO nanoplatelets were immersed in a CTAB aqueous solution
(concentration <1 mM) with a weak agitation for 24 h, and the
superfluous CTAB solution was filtered. After that, modified GO
nanoplatelets were obtained.
Preparation of PAA Solution. The PAA solution was prepared
using an aqueous polymerization method.20-22 Distilled AA monomer
(15 g) with a predetermined degree of neutralization was dissolved
in 15 mL of deionized water to achieve a concentration of 50 wt %
at ambient temperature. 0.006 g of cross-linker NMBA was dissolved
in the monomer solution. The mixed solution was then vacuum
degassed for 30 min. Radical initiator APS (0.15 g) was added to
(20) Wu, J.; Lin, J.; Zhou, M. Macromol. Rapid Commun. 2000, 21, 1032-
1034.
(21) Wu, J.; Wei, Y.; Lin, J.; Lin, S. Polymer 2003, 44, 6513-6520.
(22) Lin, J.; Wu, J.; Yang, Z.; Pu, M. Macromol. Rapid Commun. 2001, 22,
422-424.
Langmuir, Vol. 24, No. 9, 2008 4801
the mixed solution consisting of monomer and cross-linker. Under
a nitrogen atmosphere, the reaction mixture was stirred and heated
to 80 °C in a water bath for 30 min. The resulting polymer with a
molecular weight of about 30 000 was dissolved in deionized water
to form a colloid solution with a concentration of 50 wt % PAA.
Self-Assembly of the (PAA/GO)n Film. The (PAA/GO)n film
was grown on a planar glass substrate. The glass surface was cleaned
2 h prior to use with a concentrated sulfuric acid solution containing
potassium dichromate. Then the cleaned glass substrate was
immediately dipped into the PAA colloid solution for 15 min. After
being vacuum dried at 80 °C for 30 min, the glass substrate was
immersed in a suspension of GO nanoplatelets (about three carbon
molecular layers in each nanoplatelet) for 30 s to allow GO adsorption
take place.4 The weakly adsorbed GO nanoplatelets and some
impurities were removed from the surface by rinsing with deionized
water, thus a (PAA/GO)1 film was obtained. A (PAA/GO)n multilayer
film was grown by repeating the PAA and GO adsorption cycle
(where subscript n is designated as the number of times that the
adsorption cycle is repeated and the bilayer number of PAA/GO).
The deposition steps and the schematic structure of the assembled
film are sketched in Figure 2.
Characterization. GO nanoplatelets and (PAA/GO)n films were
identified by powder X-ray diffraction (XRD) on a Bruker D8-
ADVANCE X-ray diffractometer at 40 kV and 40 mA for
monochromatized Cu KR (λ ) 0.154 nm) radiation. The XRD data
for indexing and cell parameter calculations were collected in
scanning mode with a scanning speed of 5 deg·min-1 in the 2θ range
from 5 to 70°. Basal distances of (002) planes were calculated
according to the Bragg equation. The crystallite sizes were obtained
from the X-ray line width using Scherrer’s equation4,23
4802 Langmuir, Vol. 24, No. 9, 2008 Wu et al.

Kλ
Lc ) (1)
βhkl cos θhkl

where λ, θ, and β are the wavelength of X-rays, the Bragg angle,

and the pure diffraction line broadening (in radians), respectively.
K corresponds to the shape factor, with a value of 0.89.24 β is defined
here as the observed full diffraction peak breadth at half-maximum
intensity of the (hkl)) diffraction peak. Lc is the average dimension
of the crystallite perpendicular to the diffracting (hkl) plane.
Differential scanning calorimetry (DSC) curves of (PAA/GO)n
films were measured with a DSC2910 derivatograph (USA TA
Instruments) in the temperature range of 100-160 °C at a heating
rate of 10 °C‚min-1 for the study of thermal properties of the
samples.25,26
For the measurement of conductivity, each sample was milled
and pressed into pellet form (diameter, 13 mm; thickness, 1 mm)
at a pressure of 14 MPa using a Carver model C press. The electrical Figure 3. FTIR spectra of PAA, the GO platelet, and the (PAA/
conductivity of sample was measured with a four-probe conductivity GO)3 film.
test meter (RTS-9, China, Guangzhou) at ambient temperature.
Fourier transform infrared (FTIR) spectra of the samples were
identified using a Nicolet Impact 410 FTIR spectrophotometer. The
samples were prepared by adding the PAA powder or GO
nanoplatelets to solid KBr, grinding, and pressing into a pellet. The
spectrum of the (PAA/GO)n film was obtained with the ambient
atmosphere as the background.
Atomic force microscopy (AFM) images of the GO nanoplatelets
modified with CTAB and the (PAA/GO)n film were obtained with
an NS3A-02 Nanoscope III. A scanning probe microscope (Veeco
Co.) operates in tapping mode. A 1 × 1 cm2 piece of the film grown
on a glass substrate was carefully cut and fixed to the sample holder.

Results and Discussion

FTIR Spectra. The FTIR spectra of GO nanoplatelets, PAA,
and (PAA/GO)3 film samples are shown in Figure 3. The FTIR
spectrum of GO nanoplatelets displays a consistency with that
reported by Peckett et al.27 The broad, strong band at ∼3600-
3300 cm-1 is due to O-H stretching in H-O-H, and the peaks
at ∼2925 and ∼2863 cm-1 correspond to the presence of a C-H
Figure 4. XRD patterns of (PAA/GO)n films. (The degree of
bond; these results are consistent with the report by Rabah for neutralization of PAA is 60%.)
the anodic oxidation of graphite in H2SO4 in the presence of
Na2SO4.28 The absorption peak at ∼1636 cm-1 is due to -Cd Table 1. XRD Parameters of the GO Composite Films
C- stretching as part of the ring breathing mode in the GO sample 2θ (deg) d002 (Å) Lc (Å) N
skeleton, whereas the band at ∼1708 cm-1 is attributed to -Cd
graphite powder 26.5 3.28 macroscopic
O stretching. The band at ∼1422 cm-1 is due to -CHn GO 23.1 3.88 107 28
deformation, and the band at ∼1079 cm-1 is due to -C-O- (PAA/GO)3 16.35 5.42 16.88 3.11
(i.e., hydroxyl, aether) stretching in PAA.29 The absorption peaks (PAA/GO)8 15.86 5.58 16.52 2.96
at ∼1560 and 1460 cm-1 are attributed to the vibration of the (PAA/GO)13 15.21 5.82 16.17 2.78
-COO- group. The absorption bands below ∼1000 cm-1 are
moved to ∼1720 and ∼1442 cm-1, respectively. The shift of the
due to the presence of trace sulfate groups intercalated between
-COO- bands reveals the interactions between the carboxylic
graphite planes.27
groups on PAA and CTAB adsorbed on the surfaces of GO
For the spectrum of the (PAA/GO)3 film, the broad bands at
nanoplatelets.
∼3600-2900 cm-1 belong to O-H stretching from adsorbed
X-ray Diffraction. XRD is a powerful tool for the structural
water, and the peak at ∼2674 cm-1 is due to an asymmetric
analysis of layered materials such as GO. Figure 4 shows the
-C-H stretching vibration. The peak at ∼1725 cm-1 is due to
XRD patterns of the graphite powder, GO, and the (PAA/GO)n
-CdO stretching of the carboxylic group, which is overlaped
films with different adsorption cycles, and Table 1 summarizes
by the absorption band of GO. Furthermore, the absorption bands
the XRD parameters. The 2θ diffraction peaks at 26.5 and 53.2°
of the -COO- group at ∼1560 cm-1 (CdO stretching mode for
correspond to the (002) and (004) planes of natural graphite (d002
-COOH) and ∼1460 cm-1 (O-H bending for -COOH) have
) 3.37 Å).30,31 In the course of strong oxidation, the interlayer
(23) Bartram, F. Handbook of X-rays; McGraw-Hill: New York, 1967; pp
gallery of GO expands as oxygen-containing groups are
17.1-17.17. incorporated into the carbon layers: the interlayer distance of
(24) Fukuda, K.; Kikuya, K.; Isono, K.; Yoshio, M. J. Power Sources 1997, carbon-carbon molecular layers is expanded from 3.37 to 3.86
69, 165-168.
(25) Messe, L.; Perdigon, A.; Clarke, S. Langmuir 2005, 21, 5085-5093. Å (where the 2θ diffraction angle shifted from 26.5 to 23.1°).
(26) Wang, G.; Yang, Z.; Li, X.; Li, Z. Carbon 2005, 43, 2564-2570.
(27) Peckett, J.; Trens, P.; Gougeon, R.; Poppl, A.; Harris, R. Carbon 2000, (30) Zhang, B.; Wang, T.; Zhang, S.; Qiu, J.; Jian, X. Carbon 2006, 44, 2764-
38, 345-353. 2769.
(28) Rabah, M.; Abdul, A.; Ismail, A. J. Appl. Electrochem. 1981, 11, 41-47. (31) Pardo, H.; Faccio, R.; Araujo-Moreira, F.; Lima, O.; Mombru, A. Carbon
(29) Yang, D.; Sacher, E. Chem. Mater. 2006, 18, 1811-1816. 2006, 44, 565-569.
Conducting (PAA/GO)n Film
Figure 5. DSC curves of (PAA/GO)n films with bilayer numbers
(n) of 3, 8, and 13.
The average carbon molecular layer number (N) of GO
nanoplatelets stacking along the c axis (N) in each GO sheet can
be calculated in terms of the crystallite size and the basal
spacing: N ) Lc/d002. Line broadening in GO powder diffrac-
tometry shows that Lc is about 107 Å, which indicates that parallel
to layer expansion partial exfoliation of the macrocrystalline
graphite particles also occurs (N ) 28).4 The Lc for GO
nanoplatelets in (GO/PAA)n is about 16 to 17 Å, indicating a
two- to four-carbon molecular layer.
XRD patterns of the (PAA/GO)n films reveal that after
exploiting the coating on the films of about three carbon molecular
layers (N ≈ 3), the average carbon-carbon (C-C) interlayer
height of GO nanoplatelets is swollen (5.42-5.82 Å) with
concomitant line broadening (where the position and breadth
remained constant). This can be explained by the fact that the
interlayer interaction between carbon molecular layers is
weakened by sandwiching in PAA layers. However, Table 1
indicates that with the increase in the repeating adsorption cycle
times (n) in the preparation or the bilayer number (n) in (PAA/
GO)n films from 3 to 13, the average carbon layer number (N)
in each GO sheet decreases from 3.11 to 2.78 and the C-C
interlayer height in each GO sheet increases significantly from
5.42 to 5.82 Å, which means that partial disaggregation takes
place in the suspension and the self-assembly selects these thicker
platelets.12
Thermal Analysis. DSC was employed to search for the
thermal behavior of the (PAA/GO)n films. Figure 5 shows the
DSC curves of (PAA/GO)n films with different bilayers. It can
be seen that the film composites display an obvious endothermic
peak at about 135 °C, which is similar to that of the other
composite as reported by Xiong et al.32 The endothermic peak
is seen near the melting temperature of PAA. A small increase
in the melting temperature is observed with an increase in bilayer
number (n) for (PAA/GO)n films. This may be due to the fact
that with the increased bilayer number (n) in (PAA/GO)n films
the average carbon layer number (N) in each GO sheet (Table
1) decreases, the C-C interlayer height in each GO sheet (Table
1) increases, and the interaction between the carbon layer and
PAA layer is reinforced.
Electrical Conductivity. To reveal the influence of bilayer
number (n) and the PAA degree of neutralization on the electrical
conductivity of (PAA/GO)n films, the conductivities of (PAA/
(32) Xiong, C.; Zhou, Z.; Xu, W.; Hu, H.; Zhang, Y.; Dong, L. Carbon 2005,
43, 1788-1792.
Langmuir, Vol. 24, No. 9, 2008 4803
Figure 6. Conductivities of (PAA/GO)n films as a function of the
bilayer number (n) and the PAA degree of neutralization.
GO)n films were measured, and the results are depicted in Figure
6. It can be seen that the conductivities of the films are very low
at n < 13 for films with a PAA degree of neutralization of 60%.
In contrast, the conductivity increases dramatically from 1.06 to
47.62 S‚cm-1 when n increases from 13 to 18 for the (PAA/GO)n
film with a degree of PAA neutralization of 60%, and the
conductivity increases more obviously for the sample with a
PAA degree of neutralization of 100%. In particular, the
conductivity of the (PAA/GO)24 film with a PAA degree of
neutralization of 100% reaches 60 S‚cm-1.
The influence of the PAA degree of neutralization on the
conductivities of the films can be explained by the interaction
between PAA and GO layers in (PAA/GO)n films. In the
preparation, PAA is neutralized with potassium hydroxide, and
with the increase in the degree of neutralization, the content of
polyacrylate (-COO- group) increases and the content of PAA
(-COOH group) decreases in the solution. Under a higher degree
of PAA neutralization, more negatively charged -COO- groups
are assembled in the (PAA/GO)n films, and the electrons in the
carbon layer of GO sandwiched in PAA transfer more easily,
owing to the electrostatic repulsion of negatively charge -COO-
groups. Therefore, the (PAA/GO)n film with a higher degree of
PAA neutralization possesses a higher conductivity.
It is very interesting that the conductivity of the (PAA/GO)n
film depends on its bilayer number (n), which has not been
reported and explained so far. The charge percolation theory33-36
is a very effective tool for explaining the conducting phenomenon
for various composites and films. According to the charge
percolation theory, when the number of conducting components
in a composite or film exceeds a percolation threshold value (wt
% or vol %), the conductivity of the composite or film will
abruptly increase as a result of the connection of conducting
channels. It is known that a 1-D conducting channel or a 2-D
conducting plane has been constructed in the (PAA/GO)n film
and (PAA/GO)n films with different bilayer numbers (n) have
an approximate percentage of conducting component GO.
According to the results in Figure 6, the abrupt increase in the
conductivity of the (PAA/GO)n film is related to its bilayer number
(n), instead of the percentage of conducting component GO. In
other words, the phenomenon in which the conductivity of the
(PAA/GO)n film depends on the corresponding bilayer number
(n) is difficulty to explain by the charge percolation theory.
(33) Ramirez-Garcia, S.; Alegret, S.; Cespedes, F.; Forster, R. Anal. Chem.
2004, 76, 503-512.
(34) Watts, P.; Hsu, W.; Kroto, H.; Walton, D. Nano Lett. 2003, 3, 549-553.
(35) Bernasconi, L.; Madden, P. J. Phys. Chem. B 2002, 106, 12916-12928.
(36) Lu, J.; Weng, W.; Chen, X.; Wu, D.; Wu, C.; Chen, G. AdV. Funct. Mater.
2005, 15, 1-6.
4804 Langmuir, Vol. 24, No. 9, 2008
Figure 7. Conductivity of (PAA/GO)n (n ) 3, 13, 21) films vs
temperature.
For the (PAA/GO)n film, its conductivity depends on electron
movement in the carbon interlayer of the GO nanoplatelets. It
can be seen from Table 1 that with the increase in the bilayer
number (n) in (PAA/GO)n films the average carbon layer number
(N) in each GO sheet decreases and the C-C interlayer distance
between carbon layers in each GO sheet increases. This implies
that the interaction between carbon layers in each GO sheet is
weakens, the astriction of the carbon layer to electrons in the
interlayers is loosened, and electron movement is improved. When
the C-C interlayer height reaches a threshold value (or n reaches
a threshold value), the electrons in the carbon interlayer become
free, and the conductivity of the film increases sharply.
It is understandable that the interlayer threshold height allowing
free interlayer electrons is different for different films and
preparation conditions. For instance, under our experimental
conditions, the C-C interlayer threshold height for the (PAA/
GO)n film with a PAA degree of neutralization of 60% is >5.82
Å or the bilayer number (n) is >13. Therefore, it is anticipated
that an expanded graphite/polymer film with higher conductivity
has a C-C interlayer height >5.82 Å.
Influence of Temperature on the Electrical Conductivity
of Films. Figure 7 presents the relationship between the
conductivities of (PAA/GO)n (n ) 3, 13, 21) films and
temperature. It can be seen that with the increase in temperature
the conductivities of the (PAA/GO)n films decrease slowly to
<120 °C, decrease sharply from 120 to 140 °C, and then
appreciably increase to >140 °C (melting temperature). In other
words, (PAA/GO)n (n ) 3, 13, 21) films show a representative
positive temperature coefficient (PTC) effect, followed by a
negative temperature coefficient (NTC) effect. Similar PTC
effects were reported in other conductive composites based on
semicrystalline polymers, such as polyethylene,37,38 polyure-
thane,32 and poly(vinylidene fluoride)39 near the melting tem-
perature of the matrix.
Kohler40 indicated that the PTC mechanism is a function of
the difference in the thermal expansion coefficients between the
matrix and filler. The sudden conversion occurring at the melting
temperature of the polymer matrix results in the breakup of the
conducting channels or networks with a consequent anomalous
decrease in the conductivity. The appearance of the NTC effect
is due to the movement of GO nanoplatelets in the PAA melt
(37) Xi, Y.; Bin, Y.; Chiang, C.; Matsuo, M. Carbon 2007, 45, 1302-1309.
(38) Zhang, C.; Ma, C.; Wang, P.; Sumita, M. Carbon 2005, 43, 2544-2553.
(39) Xu, H.; Dang, Z. Chem. Phys. Lett. 2007, 438, 196-202.
(40) Kohler, F. U.S. Patent 3,243,753, 1966.
Wu et al.
Figure 8. AFM images of (a) CTAB on GO and (b) the arrangement
of GO on a PAA layer. (a) The scan area is 279 × 279 nm2, the bright
regions in A correspond to the headgroups, and the darker strand
corresponds to the gaps between the terminal methyls of the tail-
to-tail alkyl chains. (b) The scan area is 5 × 5 µm2.
and the formation of new conducting pathways.41 For the (PAA/
GO)n films, it is observed that the electrical conductivities of
(PAA/GO)n films with bilayer numbers of 13 and 21 decrease
by 69.19 and 76.59%. As previously discussed (Table 1), with
the increase in the bilayer number (n) in (PAA/GO)n films, the
C-C interlayer height in each GO sheet increases, and the average
carbon layer number (N) in each GO sheet decreases. In other
words, the PAA content increases. (PAA/GO)n films with larger
bilayer number will contain more PAA, and more melted PAA
bring about a decrease in conductivity.
AFM Observation. The periodic structure of CTAB on the
GO surface was imaged using AFM at ambient conditions. Figure
8a shows an image of a CTAB ordered structure on GO. In this
image, CTAB self-assembles into a 2-D lamella with two domains
occurring in two different orientations (indicated as 1 and 2).
The interlaced angle between two head-to-head CTAB molecules
situated interior and on the edge of GO nanoplatelets is about
33.69°, which shows different CTAB molecule arrangements on
the interior and the edge of GO nanoplatelets. Within the two
domains, parallel and extended stripes can be resolved. This
results show that CTAB forms a 2-D ordered structure on the
GO surface, although there are both nonpolar and polar groups
in one molecule.
Figure 8b offers the morphology of GO modified with CTAB
absorbed on the PAA layer. The stacking GO nanoplatelets are
embedded in the PAA layer. The formation of CTAB self-
assembled lamella on the GO surface results in the headgroups
pointing toward the exterior. Under the drive of CTA+ ions,
PAA chains are adsorbed to construct the LbL film. In GO/
CTAB solution, the tail groups of CTAB adsorb strongly on the
GO surface in a tail-to-tail assembly as a result of the interactions
between the alkyl tail in the registry and the GO lattice. The
stable ...GO-PAA-GO-PAA... structure can be explained by the
following reaction equation.
self-assembly
CTAB + PAA 98 PAA/CTA + KBr
In the preparation, washing is introduced after each PAA/GO
bilayer formation. In this case, most of the K+ and Br- ions are
rinsed, which results in the reversible equilibrium moving to the
right and a stable PAA/CTA appearing.
Conclusions
GO nanoplatelets were prepared from natural graphite powder
by oxidizing, ball milling, exfoliating, and modifying with cationic
surfactant CTAB. The (PAA/GO)n composite films were designed
by alternating the electrostatic adsorption of polyelectrolyte PAA
and the GO nanoplatelets. The (PAA/GO)n films were LbL grown
(41) Panwar, V.; Sachdev, V.; Mehra, R. Eur. Polym. J. 2007, 43, 573-585.
Conducting (PAA/GO)n Film Langmuir, Vol. 24, No. 9, 2008 4805

by Langmuir-Blodgett self-assembly. The compositions, the PAA melting temperature. The AFM images reveal that CTAB
structure, thermal, and electric properties, and the micromor- molecules stack in a well-ordered head-to-head structure on both
phology of the (PAA/GO)n films were investigated by FTIR, surfaces of GO nanoplatelets and the GO nanoplatelets are
XRD, DSC, AFM, and conductivity measurement, respectively. embeded between PAA layers.
It is found that PAA and GO stack LbL orderly and repeatedly
in the (PAA/GO)n films and about three carbon molecular layers
Acknowledgment. We are grateful for support from the
are superposed in each GO sheet. FTIR spectra offer evidence
National Natural Science Foundation of China (no. 50572030),
for the interaction between the carboxylic groups in PAA and
the Specialized Projiect of Fujian Province (nos. 2005HZ01-4
the CTAB on the surface of GO nanoplatelets. The electro-
and 2007HZ0001-3), the Key Project of the Chinese Ministry
chemistry measurements show that the conductivity of (PAA/
of Education (no. 206074), and the Specialized Research Fund
GO)n film depends on the C-C interlayer height of GO sheet.
for the Doctoral Program of Chinese Higher Education (no.
In our experiment conditions, the interlayer threshold height for
20060385001)
the (PAA/GO)n films with a PAA degree of neutralization of
60% is more than 5.82 Å. A typical PTC effect occurs above the LA800095Z