You are on page 1of 7

Facile synthesis of graphenesilicon nanocomposites with an

advanced binder for high-performance lithium-ion battery anodes


Da Chen
1
, Ran Yi, Shuru Chen, Terrence Xu, Mikhail L. Gordin, Donghai Wang
Department of Mechanical & Nuclear Engineering, The Pennsylvania State University, University Park, PA 16802, USA
a b s t r a c t a r t i c l e i n f o
Article history:
Received 12 August 2013
Received in revised form 7 November 2013
Accepted 10 November 2013
Available online xxxx
Keywords:
Silicon/graphene nanocomposites
Binder
Cycling performance
Lithium ion batteries
In this work, the nanocomposite of silicon nanoparticles dispersed on conducting graphene (Si/graphene) was
successfully synthesized using high-energy ball milling followed by thermal treatment, and Xanthan gum was
developed for the rst time as a novel advanced binder for Si-based lithium-ion battery anodes. Compared to
the pristine Si anode, the Si/graphene composite anode showed an enhanced reversible capacity, excellent cyclic
performance and rate capability, highlighting the advantages of dispersing Si nanoparticles on graphene sheets.
The signicant enhancement on electrochemical performance could be ascribed to the fact that the Si/graphene
composite anode could maintain excellent electronic contact and accommodate the large volume change of Si
during the lithiation/delithiation process. In addition, the Si/graphene anode with the gum binder exhibited im-
proved cycling and rate performances compared to that with the conventional carboxymethyl cellulose (CMC)
binder. Such an enhancement was ascribed to the high binder stiffness and the strong adhesion of the binder
to Si-based particles due to the binder's specic chemical structure and properties, which helps maintain the in-
tegrity of the electrode and accommodate the volume change of Si. This work demonstrates that the Si/graphene
nanocomposite with an advanced binder offers great advantages to enhance the lithium storage capacity, cyclic
stability, and rate capability, making it a promising candidate as an anode material for high-performance lithium
ion batteries (LIBs).
2013 Elsevier B.V. All rights reserved.
1. Introduction
Lithiumion batteries (LIBs) have been recognized as an enabling en-
ergy storage technology for many emerging applications, including
electric vehicles (EVs) and plug-in hybrid electric vehicles (PHEVs)
[14]. However, such a goal cannot be reached without designing alter-
native materials both as positive and as negative electrodes to increase
the capacity of LIBs. While studies on positive-electrode materials have
led to a limited increase of their capacity, the development of silicon-
based negative electrodes has already been shown to enhance the ca-
pacity of standard graphite electrodes by nearly one order of magnitude
[57]. Besides the high theoretical capacity (4200 mAh g
1
) with a low
potential window, silicon-based anode materials are also environment-
friendly, easily available and produced in widely distributed areas [8].
Despite all of these advances, silicon-based anodes suffer from severe
pulverization triggered by their more than 300% volume change upon
lithiuminsertion and extraction, which, if not addressed, causes serious
capacity fading during charge/discharge cycling, especially at high
current densities [9,10].
On the one hand, many attempts have been made on tailoring Si to
address this issue, including the use of nanostructuredsiliconcompounds
[11,12], alloying with metals [13,14], and dispersing the silicon in
a carbon matrix [1517]. Among these, Si/C composites have attracted
particular interest. Different kinds of carbon materials have been demon-
strated to accommodate the Si volume change and improve the cycling
stability of Si-based materials, including amorphous carbon (pyrolysis
[18,19] or hydrothermal [20]), graphite [7,21], carbon nanotubes [22,23]
and mesoporous carbon [24]. Graphene, a new kind of carbon material,
is an excellent substrate to host active nanomaterials for energy applica-
tions due to its high conductivity, large surface area, exibility, and
chemical stability [2527]. Recently, graphene has been demonstrated
as an active matrix for the preparation of Si/graphene nanocomposites
to improve the electrochemical performance of Si-based materials
[2840]. The enhanced electrochemical performance can be attributed
to the fact that graphene can not only provide a support for dispersing
Si nanostructures and work as a highly conductive matrix for enabling
good contact between them, but can also effectively prevent the vol-
ume expansion/contraction and aggregation of Si nanostructures dur-
ing Li charge/discharge process. We note that in these earlier reports,
Si/graphene nanocomposites were prepared by two approaches:
physically blending pre-synthesized Si nanocrystals and graphene
[28,3032,34,35,3840], and chemical reaction process (such as vapor
deposition methods [33], covalent binding via aromatic linkers
through diazonium chemistry [36], in situ magnesiothermic reduction
of SiO
2
/graphene oxide composites [37]). The former physical approach
is simple and economical, but difcult to obtain homogeneous and stable
Solid State Ionics 254 (2014) 6571
Corresponding author. Tel.: +1 814 863 1287.
E-mail address: dwang@psu.edu (D. Wang).
1
Present address: China Jiliang University, Hangzhou 310018, China.
0167-2738/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ssi.2013.11.020
Contents lists available at ScienceDirect
Solid State Ionics
j our nal homepage: www. el sevi er . com/ l ocat e/ ssi
Si/graphene nanocomposites, which is believed to be a key issue for a
composite to exhibit good performances. In spite of the noticeably
enhanced homogeneity and stability, the latter chemical approach is
complicated, time-consuming and costly. Fromthe viewpoint of industri-
al application, the former physical approach will be even more promising
than the latter chemical approach. Thus, it is crucial to develop a facile
physical approachtoprepare homogeneous andstable Si/graphene nano-
composites as an advanced anode material for high-performance LIBs
with potential industrial applications.
On the other hand, much attention has also been paid to the devel-
opment of selected binders for anode and cathode electrodes, as they
can help to alleviate the structural changes and improve the cycling
performance of the electrodes [41]. Poly(vinylideneuoride) (PVDF),
the most conventional binder, is attached to Si particles via weak van
der Waals forces only, and fails to accommodate large changes in spac-
ing between the particles [42]. Recently, carboxymethyl cellulose
(CMC), one of bio-derived polymers containing carboxy groups, has
been demonstrated as an alternative effective binder for Si-based an-
odes, which typically shows a better battery performance than the
PVDF binder [4346]. However, reasonably stable anode performance
could only be achieved when Si volume changes were accommodated
by using extra-large binder content [46,47], which lowers the resulting
anode capacity. Development of more efcient binders is therefore an
important task for the realization of high capacity anodes with stable per-
formance. Recently Kovalenko and coworkers [48] reported that mixing
Si nanopowder with sodium alginate (SA), a natural polysaccharide
extracted from brown algae, yielded a stable battery anode possessing
an enhanced reversible capacity compared to the conventional PVDF
binder and the CMC binder.
Herein we demonstrate that the electrochemical performance of
Si-based anode can be signicantly improved by combining the advan-
tages offered by both design of Si/graphene nanocomposite materials
and utilization of a novel binder. Si/graphene nanocomposites are pre-
pared by using high-energy ball milling followed by thermal treatment,
and Xanthan gum is employed for the rst time as an advanced binder
for Si/graphene nanocomposite anodes. The as-prepared Si/graphene
composite anode exhibits signicantly improved cycling performance
compared to the Si anode when CMC is used as binders for both anodes.
With the Xanthan gumbinder the performance of Si/graphene compos-
ite anode has been further improved, comparable to that with the
alginate binder.
2. Experimental
Graphene oxide (GO) was prepared by a modied Hummers meth-
od [49], and Si nanoparticles was synthesized by the magnesiothermic
reduction according to the literature [50]. To prepare the Si/graphene
nanocomposites, a mixture of as-prepared Si nanoparticles and GO
was blended in a 4 wt.% N-Methyl-2-pyrrolidone (NMP) solution of
PVDF. The composition of the resulting mixture was 48 wt.% Si nano-
particles, 28 wt.% GO and 24 wt.% PVDF. The obtained mixture was
then ball milled for 10 h with agate balls as the mixing media, and nal-
ly vacuum-dried at 100 C for 6 h. To reduce GO and carbonize PVDF,
the dried composite was further calcined at 700 C in an argon-owing
tube furnace for 3 h with a heating ramp of 5 C min
1
.
Transmission electron microscope (TEM) measurements were con-
ducted on a JEOL 1200 microscope at 200 keV. X-ray diffraction (XRD)
patterns of the samples were measured on a Powder X-ray diffractom-
eter (MiniFlex II, Rigaku) using Cu K radiation. Thermogravimetric
analysis (TGA) was carried out using a thermogravimetric analyzer
(SDT 2960, TA instruments) with a heating rate of 5 C min
1
in air.
The Si-based anodes before and after cycling were characterized by a
eld-emission scanning electron microscope (FE-SEM, Zeiss LEO 1530).
Delithiated Si-based anodes after cycling were taken out of the coin
cells inside a glovebox, washed with acetonitrile to remove the residual
electrolyte and lithium salts and dried at room temperature before
further SEM investigation.
Electrochemical properties of the products were measured using
coin cells. The working electrodes were prepared by casting a slurry
consisting of 80 wt.% of active material (as-prepared Si nanoparticles,
or Si/graphene nanocomposites), 10 wt.% of conductive Super P carbon,
and 10 wt.% of a binder (CMC (Alfa Aesar), SA (MP Biomedicals)
or Xanthan gum (Tokyo Chem.)) onto copper foil. The electrolyte
consisted of a solution of 1 M LiPF
6
in ethylene carbonate (EC)/diethyl
carbonate (DEC) (1:1 v/v). Lithium foil was used as counter electrodes.
The mass loading of active materials was 1.0 mg/cm
2
. These cells
were assembled in an argon-lled glovebox (MBraunLabstar) and
galvanostatically cycled between 0.01 V and 1.5 V on a multi-channel
battery cycler (Arbin Instruments). The AC impedance was measured at
an Autolab electrochemical workstation (CH Instruments), with the
frequency range and voltage amplitude set as 100 kHz to 0.01 Hz and
10 mV, respectively.
3. Results and discussion
Fig. 1 shows the typical TEM images of as-prepared Si nanoparticles,
GO and Si/graphene nanocomposites. Si nanoparticles (Fig. 1A)
were irregular-shaped particles about 50150 nm in size and prone to
aggregation. For the sample of as-prepared GO (Fig. 1B), it was found
that GO was almost transparent, and had one or few layer 2D sheet
structure with some obvious wrinkles. These corrugated wrinkles indi-
cated that the 2D sheet structure becomes thermodynamically stable
during bending [51]. As for the sample of as-prepared Si/graphene
nanocomposite (Fig. 1C), the graphene sheets and Si nanoparticles
were distinguished clearly, and Si nanoparticles with the size of about
50150 nm were well dispersed and attached to the transparent
graphene sheets. This indicates that the high-energy ball milling process
could prevent Si nanoparticles from agglomerating and enable rm
attachment of these nanoparticles to the graphene support.
The XRD patterns obtained from the as-prepared Si nanoparticles
and Si/graphene nanocomposites are shown in Fig. 2. All of the peaks
of as-prepared Si nanoparticles can be indexed as well-crystallized Si
with cubic structure (JCPDS No. 27-1402) indicating that high-purity
Si was successfully obtained by magnesiothermic reduction. After
mixing with graphene, besides the sharp peaks (2 = 28.5, 47.4,
56.0, 69.2 and 76.4) corresponding to the Si crystalline phase, the dif-
fraction peaks at 2 = 26.0 and 42.8 attributed to the graphite-like
(002) and (100) structure from graphene can also be observed [52],
which implies that GO was reduced to graphene during the calcination
process. In addition, the broad character of the peak at 26.0 proves that
graphene was homogeneously distributed in the nanocomposites with-
out signicant stacking or agglomeration [32]. Fig. 3 shows the Raman
spectrum of the Si/graphene nanocomposite. Two peaks at 1333 and
1614 cm
1
can be observed, which correspond to the D (disordered)
band and the G (graphite) band of carbon, respectively. These two
peaks conrm the presence of graphene in the nanocomposite.
To determine the content of Si in the Si/graphene composites, TGA
measurements were performed in air from room temperature to 900 C
and the results are shown in Fig. 4. A small weight loss (10 wt.%) around
120 C and a major weight loss (33 wt.%) around 200 C were found in
GO due to the loss of adsorbed water and pyrolysis of the labile oxygen-
containing functional groups, respectively, while the largest weight
loss occurs at around 600 C due to the complete oxidation reaction of
GO. For Si nanoparticles, the weight increases slightly from room-
temperature to 600 C due to the formation of SiO
x
, indicating that
the oxidation of Si powder in air was not signicant at 600 C. Thus,
it's reasonable to determine the content of Si in the nanocomposites
from the largest weight loss in the TGA curve of Si/graphene nanocom-
posites. It can be estimated that the largest weight loss in nanocompos-
ites is about 46 wt.% at 600 C. Correspondingly, the Si content in the
66 D. Chen et al. / Solid State Ionics 254 (2014) 6571
composite is about 54 wt.%, close to the loading ratio of Si (48 wt.%)
during the ball milling process.
In order to understand the electrochemical performance of these ma-
terials, the as-prepared Si nanoparticles and Si/graphene nanocompos-
ites were tested as LIB anodes for electrochemical evaluation. Fig. 5A
shows the voltage proles of the as-prepared Si anode and Si/graphene
composite anode with the CMC binder cycled at a current den-
sity of 400 mA g
1
with the voltage range of 0.011.5 V. Both of Si
and Si/graphene composite anodes exhibit similar lithiation and
delithiation features to those of previously reported silicon anodes
[11], in which a long at plateau during the rst discharge (lithiation)
was observed. The rst discharge and charge (delithiation) capacities
were 3170 and 2204 mAh g
1
for the pristine Si anode, with an initial
Coulombic efciency of 69.5%. The irreversible capacity ratio of
30.3% can be assigned to the decomposition of electrolyte, forming a
solid/electrolyte interphase (SEI) on the electrode surface [53], and to
the irreversible insertion of Li ions into silicon particles. Although the
capacity for the rst cycle was relatively high, the discharge capacity
dropped rapidly to 2078 mAh g
1
(only 65.6% of the initial capacity)
for the second cycle with a Coulombic efciency of 74.4%. This could
be ascribed to the large volume change (N300%) during Li alloying (to
form Li
x
Si) and dealloying (to restore to Si) processes, leading to
an electrical disconnection between particles [16,54]. In contrast,
the Si/graphene composite anode displayed a capacity of 1553 and
1314 mAh g
1
for the rst discharge and charge processes, respective-
ly. Compared with the Si anode (74.4%), the initial Coulombic efciency
of the Si/graphene nanocomposites electrode was improved to 84.6%. In
this case, the initial irreversible capacity loss was mainly attributed to
the reductive decomposition of the electrolyte and the SEI formation
due to the large surface area, and the irreversible reaction of Li-ions
with silicon particles and the residual oxygen-containing functional
group of the Si/graphene composites [55,56]. For the second cycle,
the Si/graphene anode had the excellent capacity retention of
1446 mAh g
1
(~93.1% of the initial discharge capacity) with an ef-
ciency of 92.5%. The capacity retention and initial Coulombic efciency
of the Si/graphene composite anode represent a signicant improve-
ment over those of the Si anode. The dQ/dV proles of Si anode and
Si/graphene composite anode are shown in Fig. 5B. For both anodes, a
main peak was observed at around 0.07 V with an onset potential of
~0.12 Vduring the rst discharge corresponding to the long at plateau
in the rst discharge voltage prole, which was due to the phase transi-
tion of crystalline Si (c-Si) to amorphous lithium silicide (a-Li
x
Si)
[57,58]. Upon the second discharge, there were two peaks (at 0.24 V
and 0.09 V) for the Si anode, while three peaks (at 0.24, 0.09 and
0.04 V) were observed for the Si/graphene composite anode due to the
contribution of graphene [30]. The peaks at 0.24 and 0.09 V may come
fromthe phase transitions between amorphous Li
x
Si, i.e., the phase tran-
sition from the P-I (LiSi) phase to the P-II (Li
7
Si
3
) phase and the subse-
quent transition to the P-III (Li
15
Si
4
) phase [59]. On the other hand, the
differential capacity plot of the 1st charge and 2nd charge for both of
the samples (Si and Si/graphene) shows that the de-alloying (charging)
reaction from P-III to P-II and P-II to P-I occurred corresponding to the
peak potential of ~0.28 V and 0.45 V, respectively [57,58].
To study the cycling stability of the as-prepared Si and Si/graphene
anodes with the CMC binder, their dischargecharge cycling perfor-
mance was evaluated for the initial fty cycles at a current density of
10 20 30 40 50 60 70 80
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
2 Theta (degree)
Si
Si/graphene
0
0
2
graphene
Fig. 2. XRD patterns of as-prepared Si nanoparticles and Si/graphene nanocomposite.
1000 1200 1400 1600 1800
Raman shift (cm
-1
)
I
n
t
e
n
s
i
t
y

(
a
.
u
.
)
G-band D-band
Fig. 3. Raman spectrum of the Si/graphene nanocomposite.
Fig. 1. TEM images of (A) as-prepared Si nanoparticles, (B) graphene oxide and (C) Si/graphene nanocomposite.
67 D. Chen et al. / Solid State Ionics 254 (2014) 6571
400 mA g
1
, as shown in Fig. 6A. It can be seen that the pristine
Si anode displayed severe capacity decay, and the specic capacity
dropped rapidly from the initial high value of 3170 mAh g
1
to about
424 mAh g
1
(13.4% capacity retention) and 105 mAh g
1
(3.3%
capacity retention) after 10 cycles and 50 cycles, respectively. In
contrast, the Si/graphene composite anode exhibited much better
cycling stability. The Si/graphene anode showed the initial discharge
capacity of 1553 mAh g
1
with the Coulombic efciency of 84.6%.
After 10 and 50 cycles, the discharge capacities are 1353 mAh g
1
(~87.2% of the initial capacity) and 484 mAh g
1
(~31.2% of the initial
capacity), respectively. It should be noted that though the capacity re-
tention of Si/graphene was better than that of Si, the discharge capacity
of Si/graphene faded a lot over 50 cycles. The decay of reversible capac-
ity of the Si/graphene over 50 cycles could be due to the pulverization
of original Si nanoparticles during Li insertion/extraction process,
thus leading to the gradual damage of the intimate attachment between
graphene and Si nanoparticles and the loss of electrical connectivity
between them. In addition, the rate capabilities of the as-prepared
Si and Si/graphene anodes were also investigated. As shown in Fig. 6B,
the Si/graphene composite anode exhibited a much better rate perfor-
mance. Particularly signicant is that the composite anode still retained
a capacity of 766 mAh g
1
and 458 mAh g
1
when the current density
reached 2000 mA g
1
and 4000 mA g
1
, respectively. At these
two high rates, however, the Si anode showed only 109 mAh g
1
and 60 mAh g
1
, respectively. Furthermore, to demonstrate the favor-
able electrical connection and charge transport between graphene and
Si nanoparticles, the AC impedance for the as-prepared Si anode and
Si/graphene composite anode was investigated. Fig. 7 shows the
Nyquist plots of the as-prepared Si anode and Si/graphene composite
anode at a discharged potential of 0.1 V vs. Li/Li
+
after three cycles.
The impedance curves show one compressed semicircle in the
medium-frequency region, which could be assigned to the charge-
transfer resistance (R
ct
), and an inclined line in the low-frequency
range, which could be considered as Warburg impedance [30]. It is
clear that the Si/graphene composite anode exhibited a smaller charge-
transfer resistance (R
ct
) than the Si anode, indicating faster charge trans-
fer of the Si/graphene composite anode. This enhancement could be as-
cribed to the high electrical conductivity as well as the high surface area
andporous structure of the graphene, whichcanfacilitate the penetration
of the electrolyte. From the above electrochemical results, it is clearly
shown that the attachment of Si nanoparticles with graphene plays an
important role in improving the electrochemical performance. The high
reversible capacity, improved cycle stability and rate performance could
be attributedto the porosity betweenthe graphene sheets that was favor-
able for Li ion transport, the interleaved electron transfer highways built
up from graphene nanosheets, and the exible connement ability of
graphene to enwrap Si nanoparticles for inhibiting their volume change,
alleviating mechanical stress on the nanoparticles and preventing the
0 100 200 300 400 500 600 700 800 900
0
20
40
60
80
100
120
graphene oxide
Si/graphene
W
e
i
g
h
t

(
%
)
Temperature (
o
C)
Si
Fig. 4. TGA curves of as-prepared Si nanoparticles, graphene oxide and Si/graphene
nanocomposite.
(A)
0 500 1000 1500 2000 2500 3000
0.0
0.5
1.0
1.5
2.0
V
o
l
t
a
g
e

v
s
.

L
i
/
L
i
+

(
V
)
Specific Capacity (mAh/g)
1st cycle_Si/graphene
2nd cycle_Si/graphene
1st cycle_Si
2nd cycle_Si
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
-50000
-40000
-30000
-20000
-10000
0
10000
20000
30000
40000
charging
d
Q
/
d
V

(
m
A
h

g
-
1

V
-
1
)
Voltage vs. Li/Li
+
(V)
1st cycle_Si
2nd cycle_Si
1st cycle_Si/graphene
2nd cycle_Si/graphene
discharging
(B)
Fig. 5. (A) The initial two chargedischarge voltage proles of as-prepared Si nanoparticles and Si/graphene nanocomposite using CMC as the binder. The current density is 400 mA g
1
.
(B) Differential capacity dQ/dV curves of the initial two cycles of the as-prepared Si nanoparticles and Si/graphene nanocomposite.
(A) (B)
0 10 20 30 40 50
0
500
1000
1500
2000
3500
Si_charge
Si_discharge
Si/graphene_charge
Si/graphene_discharge
Si_charge
Si_discharge
Si/graphene_charge
Si/graphene_discharge
S
p
e
c
i
f
i
c

c
a
p
a
c
i
t
y

(
m
A
h
/
g
)
Cycle number
0 4 8 12 16 20 24 28
0
500
1000
1500
2000
3500
4000
S
p
e
c
i
f
i
c

c
a
p
a
c
i
t
y

(
m
A
h
/
g
)
Cycle number
100mA/g
4000mA/g
2000mA/g
1000mA/g
500mA/g
200mA/g
100mA/g
Fig. 6. (A) Comparison of the cycling performance of as-prepared Si anode and Si/graphene composite anode with the CMC binder. The current density is 400 mA g
1
. (B) Capacity per-
formances at various cycling rates for as-prepared Si anode and Si/graphene composite anode with the CMC binder.
68 D. Chen et al. / Solid State Ionics 254 (2014) 6571
detachment and agglomeration of pulverized Si nanoparticles during
cycling.
As is well known, binders play an important role in LIBs, especially
for the anode materials with huge volume change, such as Si, during
lithiation and delithiation process. It has been demonstrated that
the electrochemical performance of anode can be greatly improved
by simply employing novel binders without further processing the
active materials [48]. Thus to nd a new type of binder for enhanced
electrochemical performance of Si is attractive but challenging. We
demonstrate here that by using Xanthangumas a novel binder the elec-
trochemical performance of Si/graphene composite anode could be
much improved. To highlight the superiority of Xanthan gumas a bind-
er for Si/graphene composite anode, the Xanthan gumbinder was com-
pared with the conventional CMC binder and the newly-discovered
favorable SA binder [48]. Fig. 8A shows the dischargecharge cycling
performance of the Si/graphene anode with these three different
binders for the fty cycles at a current density of 400 mA g
1
. Although
Si/graphene anodes with the three binders of CMC, SA and Xanthan
gum exhibited similar decay trend, those with Xanthan gum and SA
have higher specic capacity and better capacity retention than the
one with CMC. To be more specic, after fty cycles, the Si/graphene an-
odes with the SA and gum binders retained the discharge capacity of
780 mAh g
1
(~43.8% of the initial capacity) and 725 mAh g
1
(~43.1% of the initial capacity), respectively, while the anode with the
CMC binder retained only 484 mAh g
1
(~31.2% of the initial capacity).
In addition, the rate performance measurements further demonstrated
the enhancement of the electrochemical performance of Si/graphene
anodes by using Xanthan gum instead of CMC as the binder. Fig. 8B
shows the rate performances of Si/graphene composite anodes with
CMC, SA and Xanthan gum binders. It can be clearly observed that the
Si/graphene anodes with the SA and Xanthan gum binders show
a much better rate performance than that with the CMC binder. At a
low current rate (such as: 100 and 200 mA g
1
), the discharge capaci-
ties of these three Si/graphene anodes were nearly the same for
the initial four cycles; while at a high current rate (500 mA g
1
),
the Si/graphene anodes with SA and Xanthan gum binders exhibited
much higher capacities than that with the CMC binder. For example,
at a high current rate of 4000 mA g
1
, the Si/graphene anodes with
the SA and Xanthan gum binders possessed the initial discharge capac-
ities of 754 and 769 mAh g
1
, respectively, while the initial discharge
capacity of the anode with CMC binder was only 458 mAh g
1
.
Based on the above results, the conventional CMC binder showed
the worst cycling and rate performance evidenced by low capacity
and rapid capacity fading, and the gum binder exhibited a signicantly
improved cycling and rate performance comparable to the newly-
discovered SA binder. We speculate that this enhancement in cycling
and rate performance could be ascribed to the high binder stiffness
[48] and the strong adhesion of the binder to Si-based particles, which
is closely related to the chemical structure and properties of the binder.
It is knownthat CMC is a cellulose derivative withcarboxymethyl groups
(\CH
2
\COOH) bound to the cellulose backbone. These carboxymethyl
groups are synthetically induced and their distribution is random,
0 50 100 150 200 250 300
0
50
100
150
200
250
300
-

Z
'
'

(
o
h
m
)
Z' (ohm)
Si/graphene
Si
Fig. 7. Nyquist plots of as-prepared Si, Si/C and Si/graphene anodes at a discharged poten-
tial of 0.1 V (vs. Li/Li
+
) from 100 kHz to10 mHz. CMC was used as the binder.
(A)
(C)
0 10 20 30 40 50
0
400
800
1200
1600
2000
S
p
e
c
i
f
i
c

c
a
p
a
c
i
t
y

(
m
A
h
/
g
)
Cycle number
gum binder
CMC binder
SA binder
0 4 8 12 16 20 24 28
0
400
800
1200
1600
2000
100mA/g
4000mA/g
2000mA/g
1000mA/g
500mA/g
200mA/g
100mA/g
CMC binder
SA binder
gum binder
S
p
e
c
i
f
i
c

c
a
p
a
c
i
t
y

(
m
A
h
/
g
)
Cycle number
(B)
Xantham gum
Fig. 8. (A) Cycling performance of Si/graphene composite electrodes with CMC, sodium alginate (SA) and Xanthan gum binders cycled at 400 mA g
1
. (B) Rate performances of
Si/graphenecomposite anodes with CMC, SA and Xanthan gum binders. (C) Chemical structure of Xanthan gum.
69 D. Chen et al. / Solid State Ionics 254 (2014) 6571
where some monomeric units may have more than one carboxylic
group, and others have none. In contrast, Xanthan gum, similar to SA,
is also a natural polysaccharide, which contains many functional groups,
such as carboxyl, hydroxyl and ester, in each of the polymer's monomer-
ic units (Fig. 8C). These functional groups are naturally present and
evenly distributed in the polymer chain. The higher concentration and
more uniform distribution of the carboxylic and hydroxyl groups
along the chain in the binder should lead to a larger number of possible
binder-Si bonds (such as hydrogen bonds, C\O\Si ester bonds,
C\O\C ester bonds, C\C bonds), and thus better Si electrode stability
[46,48]. In addition, Xanthan gummacromolecules are also much more
polar than the CMC polymer chains, which can ensure better interfacial
interaction between the polymer binder and the particles, and stronger
adhesion between the electrode layer and Cu substrate [48], thus
leading to an improved cycling performance. To verify this speculation,
the morphological changes of the Si/graphene composite anodes with
the CMC, SAand gumbinders before and after 50 cycles were examined
by ex-situ SEM (Fig. 9). It can be clearly seen that these Si/graphene
anodes revealed a similar surface morphology before and after cycling.
Notably, after cycling the particle size was signicantly enlarged
without the formation of any crack. These results indicate that the
microstructure of the Si/graphene anode with the binder (CMC, SA or
Xanthan gum) was effective in buffering the volume expansion of Si
during cycling. However, it is worthwhile to note that the surface mor-
phology of the Si/graphene anodes with the Xanthan gum binder
(Fig. 9F) and SA (Fig. 9D) after cycling was more uniform, compact and
Fig. 9. SEM images of Si/graphene composite anodes with CMC (A, B), SA (C, D) and gum (E, F) binders before and after 50 cycles at 400 mA g
1
.
70 D. Chen et al. / Solid State Ionics 254 (2014) 6571
coherent than that of the Si/graphene anode with the CMC binder
(Fig. 9B), indicating that Xanthan gum and SA possess higher stiffness
and stronger adhesion to Si/graphene nanocomposites, which thus
leads to the enhanced structural stability and electrochemical perfor-
mance of the Si/graphene composite anodes during cycling.
4. Conclusions
In summary, homogeneous Si/graphene nanocomposites were
successfully prepared using high-energy ball milling combined with
thermal treatment, and their anode performance with CMC, SA and
Xanthan gum as the binder was evaluated. Compared to the pristine
Si anode, the Si/graphene nanocomposite anode showed an enhanced
reversible capacity, excellent cyclic performance and rate capability,
highlighting the advantages of dispersing Si nanoparticles on graphene
sheets. The improved electrochemical performance of the Si/graphene
anode can be attributed to the following factors. First, the Si/graphene
nanocomposites can accommodate large strains and provide good elec-
tronic contact because of the excellent mechanical properties and con-
ductivity offered by graphene. Second, the high surface area and porous
structure of graphene can facilitate electrolyte penetration and alleviate
the large volume changes of Si.
Moreover, Xanthan gum was employed for the rst time as a binder
for Si-based anodes. The Si/graphene anode with the gumbinder exhib-
ited a signicantly improved electrochemical performance comparable
to that with the newly-discovered SA binder, and much better than that
with the conventional CMC binder. We proposed that such enhance-
ment in electrochemical performance was ascribed to the high binder
stiffness and the strong adhesionof the gumbinder to Si-based particles
due to the specic chemical structure and properties of the gumbinder.
In this case, Xanthan gum could ensure excellent interfacial interaction
between the polymer binder and the particles, and strong adhesion of
the electrode layer to the Cu substrate, which further helps maintain
the integrity of the electrode and accommodate the volume change of
Si. This work demonstrates that Si/graphene composite anode with a
novel advanced binder offers greatly enhanced lithium storage capacity,
cyclic stability, and rate capability, indicating its great potential for use
as an anode material for high-performance LIBs.
Acknowledgments
This work was supported by the Assistant Secretary for Energy
Efciency and Renewable Energy, Ofce of Vehicle Technologies of the
U.S. Department of Energy under Contract No. DE-AC02-05CH11231,
Subcontract No. 6951378 under the Batteries for Advanced Transporta-
tion Technologies (BATT) Program.
References
[1] A.S. Arico, P. Bruce, B. Scrosati, J.M. Tarascon, W. Van Schalkwijk, Nat. Mater. 4 (2005)
366.
[2] M. Armand, J.M. Tarascon, Nature 451 (2008) 652.
[3] B. Kang, G. Cedar, Nature 458 (2009) 190.
[4] A. Magasinski, P. Dixon, B. Hertzberg, A. Kvit, J. Ayala, G. Yushin, Nat. Mater. 9 (2010)
353.
[5] M.N. Obrovac, L. Christensen, Electrochem. Solid-State Lett. 7 (2004) A93.
[6] S.R. Chen, M.L. Gordin, R. Yi, G. Howlett, S.H. Sohn, D.H. Wang, Phys. Chem. Chem. Phys.
14 (2012) 12741.
[7] C. Martin, M. Alias, F. Christien, O. Crosnier, D. Blanger, T. Brousse, Adv. Mater. 21
(2009) 4735.
[8] K.L. Lee, J.Y. Jung, S.W. Lee, H.S. Moon, J.W. Park, J. Power Sources 129 (2004) 270.
[9] L.Y. Beaulieu, T.D. Hatchard, A. Bonakdarpour, M.D. Fleischauer, J.R. Dahn,
J. Electrochem. Soc. 150 (2003) A1457.
[10] X.W. Zhang, P.K. Patil, C.S. Wang, A.J. Appleby, F.E. Little, D.L. Cocke, J. Power Sources 125
(2004) 206.
[11] C.K. Chan, H. Peng, G. Liu, K. McIlwrath, X.F. Zhang, R.A. Huggins, Y. Cui, Nat.
Nanotechnol. 3 (2008) 31.
[12] R. Yi, F. Dai, M.L. Gordin, H.S. Sohn, D.H. Wang, Adv. Energy Mater. 3 (2013) 1507.
[13] M.S. Park, S. Rajendran, Y.M. Kang, K.S. Han, Y.S. Han, J.Y. Lee, J. Power Sources 158
(2006) 650.
[14] J.R. Dahn, R.E. Mar, M.D. Fleischauer, M.N. Obrovac, J. Electrochem. Soc. 153 (2006)
A1211.
[15] R. Yi, F. Dai, M.L. Gordin, S.R. Chen, D.H. Wang, Adv. Energy Mater. 3 (2013) 295.
[16] J. Saint, M. Morcrette, D. Larcher, L. Lafont, S. Beattie, J.P. Peres, D. Talaga, M. Couzi,
J.M. Tarascon, Adv. Funct. Mater. 17 (2007) 1765.
[17] L.F. Cui, L. Hu, J.W. Choi, Y. Cui, ACS Nano 4 (2010) 3671.
[18] S.H. Ng, J.Z. Wang, D. Wexler, K. Konstantinov, Z.P. Guo, H.K. Liu, Angew. Chem. Int.
Ed. 45 (2006) 6896.
[19] R. Yi, J.T. Zai, F. Dai, M.L. Gordin, D.H. Wang, Electrochem. Commun. 36 (2013) 29.
[20] Y.S. Hu, R. Demir-Cakan, M.M. Titirici, J.O. Muller, R. Schlogl, M. Antonietti, J. Maier,
Angew. Chem. Int. Ed. 47 (2008) 1645.
[21] S. Cahen, R. Janot, L. Laffont-Dantras, J.M. Tarascon, J. Electrochem. Soc. 155 (2008)
A512.
[22] J. Shu, H. Li, R.Z. Yang, Y. Shi, X.J. Huang, Electrochem. Commun. 8 (2006) 51.
[23] J.Y. Eom, H.S. Kwon, ACS Appl. Mater. Interfaces 3 (2011) 1015.
[24] H. Kim, J. Cho, Nano Lett. 8 (2008) 3688.
[25] D. Chen, L.H. Tang, J.H. Li, Chem. Soc. Rev. 39 (2010) 3157.
[26] Y. Xu, R. Yi, B. Yuan, X.F. Wu, M. Dunwell, Q.L. Lin, L. Fei, S.G. Deng, P. Andersen, D.H.
Wang, H.M. Luo, J. Phys. Chem. Lett. 3 (2012) 309.
[27] D.P. Lv, M. Gordin, R. Yi, T. Xu, J.X. Song, Y.B. Jiang, D. Choi, D.H. Wang, Adv. Funct. Mater.
(2013), http://dx.doi.org/10.1002/adfm.201301882.
[28] J.K. Lee, K.B. Smith, C.M. Hayner, H.H. Kung, Chem. Commun. 46 (2010) 2025.
[29] X.L. Wang, W.Q. Han, ACS Appl. Mater. Interfaces 2 (2010) 3709.
[30] S.L. Chou, J.Z. Wang, M. Choucair, H.K. Liu, J.A. Stride, S.X. Dou, Electrochem.
Commun. 12 (2010) 303.
[31] J.Z. Wang, C. Zhong, S.L. Chou, H.K. Liu, Electrochem. Commun. 12 (2010) 1467.
[32] H.F. Xiang, K. Zhang, G. Ji, J.Y. Lee, C. Zou, X.D. Chen, J.S. Wu, Carbon 49 (2011) 1787.
[33] K. Evanoff, A. Magasinski, J.B. Yang, G. Yushin, Adv. Energy Mater. 1 (2011) 495.
[34] X. Zhao, C.M. Hayner, M.C. Kung, H.H. Kung, Adv. Energy Mater. 1 (2011) 1079.
[35] H.C. Tao, L.Z. Fan, Y.F. Mei, X.H. Qu, Electrochem. Commun. 13 (2011) 1332.
[36] S.N. Yang, G.R. Li, Q. Zhu, Q.M. Pan, J. Mater. Chem. 22 (2012) 3420.
[37] X. Xin, X.F. Zhou, F. Wang, X.Y. Yao, X.X. Xu, Y.M. Zhu, Z.P. Liu, J. Mater. Chem. 22
(2012) 7724.
[38] X.S. Zhou, Y.X. Yin, L.J. Wan, Y.G. Guo, Chem. Commun. 48 (2012) 2198.
[39] A.M. Chockla, M.G. Panthani, V.C. Holmberg, C.M. Hessel, D.K. Reid, T.D. Bogart, J.T.
Harris, C.B. Mullins, B.A. Korgel, J. Phys. Chem. C 116 (2012) 11917.
[40] B.P.N. Nguyen, N.A. Kumar, J. Gaubicher, F. Duclairoir, T. Brousse, O. Crosnier, L.
Dubois, G. Bidan, D. Guyomard, B. Lestriez, Adv. Energy Mater. 3 (2013) 1351.
[41] H. Buqa, M. Holzapfel, F. Krumeich, C. Veit, P. Novk, J. Power Sources 161 (2006) 617.
[42] A. Magasinski, B. Zdyrko, I. Kovalenko, B. Hertzberg, R. Burtovyy, C.F. Huebner, T.F.
Fuller, I. Luzinov, G. Yushin, ACS Appl. Mater. Interfaces 2 (2010) 3004.
[43] W.R. Liu, M.H. Yang, H.C. Wu, S.M. Chiao, N.L. Wu, Electrochem. Solid-State Lett. 8
(2005) A100.
[44] J. Li, R.B. Lewis, J.R. Dahn, Electrochem. Solid-State Lett. 10 (2007) A17.
[45] D. Mazouzi, B. Lestriez, L. Roue, D. Guyomard, Electrochem. Solid-State Lett. 12
(2009) A215.
[46] J.S. Bridel, T. Azais, M. Morcrette, J.M. Tarascon, D. Larcher, Chem. Mater. 22 (2010)
1229.
[47] S.D. Beattie, D. Larcher, M. Morcrette, B. Simon, J.M. Tarascon, J. Electrochem. Soc.
155 (2008) A158.
[48] I. Kovalenko, B. Zdyrko, A. Magasinski, B. Hertzberg, Z. Milicev, R. Burtovyy, I.
Luzinov, G. Yushin, Science 334 (2011) 75.
[49] Y.X. Xu, H. Bai, G.W. Lu, C. Li, G.Q. Shi, J. Am. Chem. Soc. 130 (2008) 5856.
[50] Y. Yu, L. Gu, C. Zhu, S. Tsukimoto, P.A. van Aken, J. Maier, Adv. Mater. 22 (2010) 2247.
[51] G.X. Wang, X.P. Shen, J. Yao, J. Park, Carbon 47 (2009) 2049.
[52] K. Raidongia, A. Nag, K.P.S.S. Hembram, U.V. Waghmare, R. Datta, C.N.R. Rao, Chem.
Eur. J. 16 (2010) 149.
[53] Y. Liu, K. Hanai, J. Yang, N. Imanishi, A. Hirano, Y. Takeda, Electrochem. Solid-State
Lett. 7 (2004) A369.
[54] M. Holzapfel, H. Buqa, W. Scheifele, P. Novak, F.M. Petrat, Chem. Commun. (2005) 1566.
[55] W. Xing, J. Dahn, J. Electrochem. Soc. 144 (1997) 1195.
[56] H. Kim, Z.H. Wen, K.H. Yu, O. Mao, J.H. Chen, J. Mater. Chem. 22 (2012) 15514.
[57] M.N. Obrovac, L.J. Krause, J. Electrochem. Soc. 154 (2007) A103.
[58] C.K. Chan, R. Ruffo, S.S. Hong, R.A. Huggins, Y. Cui, J. Power Sources 189 (2009) 34.
[59] M.K. Datta, P.N. Kumta, J. Power Sources 194 (2009) 1043.
71 D. Chen et al. / Solid State Ionics 254 (2014) 6571