Available online at www.sciencedirect.com

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

Nano Energy (2014) 6, 211–218

http://dx.doi.org/12211-2855/& 2014 E

nCorresponding aufax: +1 8148634848.

E-mail address: d

RAPID COMMUNICATION

Dual conductive network-enabledgraphene/Si–C composite anode with highareal capacity for lithium-ion batteries

Ran Yi, Jiantao Zai, Fang Dai, Mikhail L. Gordin, Donghai Wangn

Department of Mechanical & Nuclear Engineering, The Pennsylvania State University, University Park,PA 16802, USA

Received 28 January 2014; accepted 16 April 2014Available online 24 April 2014

KEYWORDSBatteries;Microstructures;Electrodes;Graphene/silicon/carbon composites

0.1016/j.nanoen.2lsevier Ltd. All rig

thor. Tel.: +1 814

wang@psu.edu (D

AbstractSilicon has been regarded as one of the most promising alternatives to the current commercialgraphite anode for Li-ion batteries due to its high theoretical capacity and abundance. Althoughhigh gravimetric capacity (mAh/g) of Si-based materials can be achieved, areal capacity (mAh/cm2), an indication of the energy stored at the electrode level, has rarely been discussed.Herein, a novel micro-sized graphene/Si–C composite (G/Si–C) is reported, in which micro-sizedSi–C particles are wrapped by graphene sheets. Owing to dual conductive networks both withinsingle particles formed by carbon and between different particles formed by graphene, lowelectrical resistance can be maintained at high mass loading, which enables a high degree ofmaterial utilization. Areal capacity thus increases almost linearly with mass loading. As a result,G/Si–C exhibits a high areal capacity of 3.2 mAh/cm2 after 100 cycles with high coulombicefficiency (average 99.51% from 2nd to 100th cycle), comparable to that of commercial anodes.The current findings demonstrate the importance of building a conductive network at theelectrode level to ensure high material utilization at high mass loading and may shed light onfuture designs of Si-based anodes with high areal capacity.& 2014 Elsevier Ltd. All rights reserved.

014.04.006hts reserved.

8631287;

. Wang).

Introduction

Lithium-ion batteries (LIBs) have been intensively studiedbecause of their relatively high energy density, which makesthem attractive for use in many electronic devices [1].Recently, the emerging market for electric vehicles requires

R. Yi et al.212

LIBs with higher energy density [2,3]. Developing new anodematerials with high specific capacity is an effective way toincrease the energy density of LIBs. Due to its hightheoretical capacity (3579 mAh/g) and abundance, siliconhas been regarded as one of the most promising alternativesto the currently-used graphite anode, which has a lowtheoretical capacity of 372 mAh/g [4,5]. However, there isa major barrier to the practical application Si: its largevolume change during charge/discharge causes to severepulverization of Si particles and degradation of Si electro-des, leading to poor cycling stability [6]. Extensive effortshave been devoted to improving the cyclability of Si-basedanodes with encouraging results, including development ofvarious Si nanostructures/nanocomposites, novel bindersand electrolyte additives [7–16].

However, most of the advances in Si-based materials todate were reported in the form of gravimetric capacity(mAh/g), while areal capacity (mAh/cm2) has rarely beendiscussed. Gravimetric capacity describes the capacitythat a material can deliver. However, in practical applica-tions the performance of anodes is evaluated at theelectrode level, and areal capacity is an indication of theenergy that an electrode can store. Although high gravi-metric capacity of Si-based materials can be achieved,it usually comes with very low mass loading, which in turnleads to electrodes with limited areal capacity [11,13].In the few reports involving high mass loading, higher arealcapacity was demonstrated with: (1) limited cycle num-bers; (2) fixed gravimetric capacity/reduced voltagerange; (3) low coulombic efficiency; or (4) electrodesfabricated by special techniques which are not compatiblewith industrial slurry coating approaches, such as binder-free electrodes [17–21].

We previously demonstrated a micro-sized Si–C compositecomposed of interconnected Si nanoscale building blocksand carbon conductive network [22–24]. However, theconductive network is only within the micro-sized particlesand no such a network exists between particles, potentiallyleading to large interparticle contact resistance. Due to itsextraordinary electronic conductivity and two-dimensionalmorphology [25,26], graphene has been identified as anexcellent conductive additive in various composites toenhance the electrochemical performance of electrodematerials by improving electron transport and maintainingelectrical contact in electrodes [27–31]. However, graphenehas majorly been incorporated with nanomaterials to formnanocomposites and graphene-containing micro-sized com-posites have been less reported.

Herein, we report a novel micro-sized graphene/Si–Ccomposite (G/Si–C), in which micro-sized Si–C particles arewrapped by graphene sheets. The two-dimensional conduc-tive graphene sheets act as a conductive network betweenparticles and thus decrease the contact resistance of thewhole electrode. Thanks to it having conductive networksboth within single particles and between different particles,G/Si–C shows a higher degree of material utilization at highmass loading compared to the raw micro-sized Si–C compo-site (Si–C). An areal capacity of 3.2 mAh/cm2 after 100cycles and high coulombic efficiency (average 99.51% from2nd to 100th cycle) are achieved by G/Si–C, comparable tothat of commercial LIBs [32], making it a promising anodematerial for practical applications in LIBs.

Material and methods

Synthesis of G/Si–C

Graphite oxide (GO) was firstly prepared from naturalgraphite powder (300 mesh, Alfa Aesar) by a modifiedHummers method [33,34]. In a typical process, 1 g of SiO(2 μm) was dispersed in 150 mL water under stirring for30 min. Then 3 g of PDDA (Sigma-Aldrich) was added and themixture was stirred for another 30 min. Afterwards, as-prepared GO solution was added dropwise to the PDDA–SiOsuspension with a mass ratio of 28:1 between SiO and GO.After 2 h stirring, the GO/PDDA–SiO was obtained by vacuumfiltration followed by washing sequentially with water andethanol, and finally drying in vacuum oven at 80 1C over-night. The GO/PDDA–SiO was then transferred to a horizon-tal quartz tube. Ar/H2 (95:5 v/v) was introduced at a flowrate of 1500 sccm for 20 min to purge the system. After-wards the flow rate was reduced to 100 sccm and the tubewas heated to 950 1C with a ramping rate of 10 1C/min andkept for 5 h. The samples were taken out of the tube attemperatures below 40 1C and immersed in 20 wt% HFsolution (H2O:ethanol=5:1 by volume) at room temperaturefor 3 h to remove SiO2. The obtained porous Si was collectedby filtration and washed with distilled water and ethanol insequence several times. The final product was dried in avacuum oven at 60 1C for 4 h. Carbon coating of G/Si wasdone by thermal decomposition of acetylene gas at 800 1Cfor 10 min in a quartz furnace. The mixture of acetylene andhigh-purity argon (argon:acetylene=9:1 by volume) is intro-duced at a flow rate of 100 sccm. Si–C was preparedsimilarly without the addition of GO.

Characterization

The obtained samples were characterized on a RigakuDmax-2000 X-ray powder diffractometer (XRD) with Cu Kαradiation (λ=1.5418 Å). The operating voltage and currentwere kept at 40 kV and 30 mA, respectively. The size andmorphology of the as-synthesized products were deter-mined by a JEOL-1200 transmission electron microscope(TEM), FEI Nova NanoSEM 630 scanning electron microscope(SEM). Raman spectroscopy was conducted with a WITecCMR200 confocal Raman instrument.

Electrochemical measurements

Electrochemical experiments were performed using 2016-type coin cells, which were assembled in an argon-filled dryglovebox (MBraun, Inc.) with the G/Si–C and Si–C electrodesas the working electrode and the Li metal as the counterelectrode. The working electrodes were prepared by castingthe slurry consisting of 80 wt% of active material and 20 wt%of poly(acrylic acid) (PAA) binder. Six different electrodetypes with different mass loadings of active materials(around 1.2, 2.0 and 3.2 mg/cm2; see Table S1 in SupportingInformation for details) were fabricated using G/Si–C andSi–C. 1 mol/L LiPF6 in a mixture of ethylene carbonate, diethylcarbonate and dimethyl carbonate (EC:DEC:DMC, 2:1:2 by vol%) and 10 wt% fluoroethylene carbonate (FEC) was used as theelectrolyte (Novolyte Technologies, Independence, OH). The

213High areal capacity of Si-based anode

electrochemical performance was evaluated by galvanostaticcharge/discharge cycling on an Arbin BT-2000 battery tester atroom temperature under different current densities in thevoltage range between 1 and 0.01 V versus Li+/Li. The currentdensity and specific capacity are calculated based on the massof the composites.

Results and discussion

Figure 1 shows the preparation process of the G/Si–Ccomposite. According to the previous report [35], poly(diallydimethylammonium chloride) (PDDA) is employed tomake a GO/PDDA–SiO assembly, with PDDA acting as apositively charged medium to attract (graphite oxide) GOand SiO, both of which are negatively charged (see Figure S1in Supporting Information for more on the influence ofPDDA) [35,36]. The GO/PDDA–SiO assembly is then heatedin an H2/Ar atmosphere. During this process, GO is reducedto graphene and at the same time SiO is disproportionatedto Si and SiO2. After removal of SiO2 by HF treatment, G/Siis obtained. Finally, carbon filling by decomposition ofgaseous carbon precursor leads to formation of G/Si–C, inwhich both of graphene sheets and porous Si are coated bycarbon.

The phase and crystallinity of G/Si–C were investigatedby XRD. As shown in Figure 2a, the peaks of the pattern canbe indexed to those of crystalline face-centered cubic Si(JCPDS Card no. 27-1402). Figure 2b shows the Ramanspectrum of G/Si–C, with peaks at 1333 and 1614 cm�1,corresponding to the D (disordered) band and the G(graphitic) band of carbon, respectively [37]. These twopeaks confirm the presence of carbon in the composite. Thecarbon content was determined by elemental analysisbefore and after carbon coating to distinguish grapheneand carbon from acetylene decomposition. The carboncontent is around 30 wt% in total, with 6 wt% and 24 wt%in the form of graphene and carbon from acetylene decom-position, respectively. The intensity of the D band iscomparable to that of G band, indicating an overallamorphous carbon structure derived from the majority ofthe carbon formed by acetylene decomposition.

The morphology, size, and structure of the compositehave been investigated by transmission electron microscopy(TEM) and scanning electron microscopy (SEM). Figure 2cshows a typical TEM image of GO/PDDA–SiO. It is clear thatSiO particles are covered by graphene sheets (the edge ismarked by white arrows). TEM images of products obtainedat different steps are available in Figure S2 (see SupportingInformation). Similar morphology to these is observed in

Figure 1 Preparation process of G/Si–C.

G/Si–C (Figure 2d), which indicates that the assembledgraphene/Si structure is preserved during the HF etchingand carbon coating processes. The high-magnification TEMimage in Figure 2e shows that micro-sized particles arecomposed of interconnected nanoparticles with a size ofabout 10–15 nm. In addition, wrinkles, a characteristicfeature of graphene sheets [28,31,38,39], can also beclearly seen (marked by white arrows). Figure 2f shows anSEM image of G/Si–C, in which micro-sized particleswrapped by graphene sheets are found, consistent withthe TEM observations.

Electrochemical studies of G/Si–C and Si–C

Galvanostatic charge–discharge tests were carried out toevaluate the electrochemical performance. To determinethe influence of the incorporation of graphene, Si–C com-posite with similar carbon content (30 wt%) was also pre-pared and tested as the control sample. Electrodes withthree different active material mass loadings (around 1.2,2.0 and 3.2 mg/cm2) were fabricated for each composite,and were designated G/Si–C@low, G/Si–C@medium, G/Si–C@high, Si–C@low, Si–C@medium and Si–C@high, where low,medium, and high referred to mass loading. Note that allelectrodes were prepared without any external carbonadditives (electrodes used only composites and binder inan 8:2 ratio). Considering the potential increase in elec-trode resistance with increasing mass loading, lower activa-tion current densities were used for electrodes with mediumand high mass loading (see Table S1 in Supporting Informa-tion for details).

Figure 3a–d shows cycling performance of G/Si–C and Si–Celectrodes with different mass loadings in both gravimetricand areal forms. It is clear that increasing mass loading hasno obvious effect on gravimetric capacity of G/Si–C for theinitial cycles. All three G/Si–C electrodes exhibit similargravimetric capacities of about 1100 mAh/g (Figure 3a),indicating a similar degree of material utilization. Arealcapacity thus increases almost linearly with mass loading. Asshown in Figure 3b, G/Si–C@low and G/Si–C@medium exhi-bits about 1.3 and 2.1 mAh/cm2, respectively, while G/Si–C@high can deliver an areal capacity of 3.2 mAh/cm2 after100 cycles, comparable to that of commercial LIBs [32]. Bycontrast, the gravimetric capacity of Si–C is sharplydecreased by high mass loading. Although Si–C@low has asimilar gravimetric capacity to that of G/Si–C@low, theincrease in mass loading dramatically decreases the gravi-metric capacity. Si–C@medium shows a low gravimetriccapacity of about 740 mAh/g after the initial two cycles,only about 65% of that of Si–C@low. As a result, the arealcapacity of Si–C@medium is close to that of Si–C@low.Further increase in mass loading leads to even lowergravimetric capacity. Si–C@high delivers only 580 mAh/g,barely above half that of Si–C@low. This gravimetric capa-city drop means that only a slight increase in areal capacityof Si–C, from 1.3 to 1.8 mAh/cm2, was achieved when themass loading was almost tripled. It is also clear that Si–C@high showed capacity fading. We ascribe this to highmass loading and high degree of material utilization. Highermass loading, generally accompanied by larger volume,results in slow relaxation of stress. Added to this, higher

a

Inte

nsity

(a.u

.)

2θ (deg.)10 20 30 40 50 60 70 80 1000 1200 1400 1600 1800

b G-bandD-band

Raman shift (cm-1)

Inte

nsity

(a.u

.)

Figure 2 (a) XRD pattern and (b) Raman spectrum of G/Si–C. (c) TEM of GO/PDDA–SiO. (d) Low- and (e) high-magnification TEMimages, and (f) SEM image of G/Si–C.

R. Yi et al.214

degree of material utilization is directly related to largervolume change, which generates more stress. These twofactors together lead to larger stress in the electrode.Further material and electrode optimization such as intro-ducing built-in void into materials and developing newbinders may help to stabilize the cyclability.

Figure 3e and f shows the first cycle voltage profiles ofthe extreme cases – G/Si–C and Si–C electrodes with low andhigh mass loadings. G/Si–C@low delivered charge (delithia-tion) and discharge (lithiation) capacities of 1314 and1939 mAh/g, corresponding to a coulombic efficiency (CE)of 67% (Figure 3e). A similar CE of 64.0% was obtained for

G/Si–C@high as 1834 and 2864 mAh/g were achieved(Figure 3e). In comparison, the first cycle CE decreasedpronouncedly for Si–C at high mass loading. As shown inFigure 3f, the charge and discharge capacities of Si–C@loware 1320 and 1950 mAh/g, giving a CE of 68%. However, Si–C@high has a lower CE of 57% with charge and dischargecapacities of 1423 and 2471 mAh/g, respectively. Aside fromthe first cycle CE, another noticeable difference betweenG/Si–C and Si–C is the lithiation plateau at high massloading. G/Si–C@high has a lithiation plateau of about0.15 V, higher than 0.11 V of Si–C@high. This is indicativeof a lower polarization of the G/Si–C electrode than that of

0

200

400

600

800

1000

1200

1400

1600

1800

2000 G/Si-C@high G/Si-C@medium G/Si-C@low

Gra

vim

etric

cap

acity

(mAh

/g)

Cycle number

0.00.51.01.52.02.53.03.54.04.55.05.56.06.5

G/Si-C@high G/Si-C@medium G/Si-C@low

Are

al c

apac

ity (m

Ah/

cm2 )

Cycle number

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Gra

vim

etric

cap

acity

(mA

h/g)

Cycle number

Si-C@high Si-C@medium Si-C@low

0.00.51.01.52.02.53.03.54.04.55.05.56.06.5

Si-C@high Si-C@medium Si-C@low

Are

al c

apac

ity (m

Ah/

cm2 )

Cycle number

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Pot

entia

l vs

Li+ /L

i (V

)

Specific capacity (mAh/g)

G/Si-C@high50 mA/g

G/Si-C@low400 mA/g

0 20 40 60 80 100 0 20 40 60 80 100

0 20 40 60 80 100 0 20 40 60 80 100

0 400 800 1200 1600 2000 2400 2800 0 400 800 1200 1600 2000 24000.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Specific capacity (mAh/g)

Pot

entia

l vs

Li+ /L

i (V

)

Si-C@high50 mA/g

Si-C@low400 mA/g

Figure 3 (a–d) Cycling performance of G/Si–C and Si–C. First cycle voltage profiles of (e) G/Si–C and (f) Si–C at low and high massloadings.

215High areal capacity of Si-based anode

the Si–C one [31]. Such difference is due to the incorpora-tion of graphene sheets as a conductive network [28,39],which provides additional electron transport pathwaysbetween Si–C particles. It is worth noting that the presenceof graphene, unlike in many previous reports [40–43], doesnot significantly decrease the first CE (only from 68% to67%), which is probably due to the low content of graphene(6 wt%). G/Si–C@high also has high CE after the initial cycle.Figure 4a shows the CE from 2nd to 100th cycle, with thered line marking 99.5%. The CE increases above 99.5% in 15cycles and remains that high afterwards. The average CEfrom the 2nd to 100th cycle is 99.57%, which is rarelyreported for Si-based electrodes with high mass loading.

The difference in electrochemical performance was furtherstudied by electrochemical impedance spectroscopy (EIS)

measurements. Figure 4b shows the Nyquist plots of G/Si–Cand Si–C electrodes with low and high mass loadings, as twoextremes, after five cycles at delithiated state. All spectraconsist of a depressed semicircle in the high-to-mediumfrequency range and a straight line in the low frequency range.The Randles equivalent circuit (upper inset) was employed toanalyze the impedance spectra [31]. According to the fittingresults (Table S2, Supporting Information), the charge transferresistance of G/Si–C@high is 10.9Ω, similar to that of G/Si–C@low (9.14Ω, middle inset). By contrast, Si–C@high hasa much higher charge transfer resistance of 59.98Ω thanSi–C@low (16.83 Ω). The difference in resistance, especially athigh mass loading, serves strong evidence of improvement inelectronic conductivity through forming conductive network bygraphene at both low and high mass loadings.

92.092.593.093.594.094.595.095.596.096.597.097.598.098.599.099.5

100.0100.5

Cou

lom

bic

effic

ienc

y (%

)

Cycle number

0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 800

10

20

30

40

50

60

70

80

4 6 8 10 12 14 16 18 20 22 24

2

4

6

8

10

12

14 G/Si-C@low G/Si-C@high Si-C@low Si-C@high

Figure 4 (a) Coulombic efficiency (2nd to 100th cycle) of G/Si–C@high. (b) Impedance spectra of G/Si–C and Si–C at low andhigh mass loadings. Insets in (b) are the Randles equivalent circuit and enlarged spectra showing difference between G/Si–C@lowand G/Si–C@high, respectively.

Figure 5 (a) SEM and (b) TEM images of G/Si–C@high electrode after 100 cycles.

R. Yi et al.216

To better understand the reason for the high degree ofmaterial utilization, post-cycling SEM and TEM investigationwas carried out on delithiated Si–C@high electrode after 100cycles. Acetonitrile and 1 M hydrochloric acid solution wereused to remove electrolyte, SEI layer, and binder of the TEMsample. As shown in Figure 5a, the structure of electrodemaintains well as only small cracks are present without anyactive material peeling off, similar to our previous report[23]. TEM observation (Figure 5b) reveals change of theactive material at nanoscale level. The micro-sized compo-site did break into smaller pieces during cycling. However,these pieces were still connected by graphene sheets,similar to the structure before cycling (Figure 2d), whicheffectively decreased the contact resistance between smal-ler pieces and thus ensured good material utilization.

Conclusions

In summary, a graphene-wrapped micro-sized Si–C compo-site has successfully been prepared via a facile approach.

The graphene provides additional electron pathways for thewhole electrode by forming a conductive network thatconnects different Si–C particles, giving the G/Si–C conduc-tive networks both within and between particles. As aresult, low electrical resistance can be maintained at highmass loading, which enables a high degree of materialutilization. Correspondingly, the graphene-wrapped Si–Ccomposite exhibits a high areal capacity of 3.2 mAh/cm2

after 100 cycles with high coulombic efficiency. Thesefindings demonstrate the importance of building a conduc-tive network at the electrode level for high materialutilization at high mass loading and may shed light onfuture designs of Si-based anodes with high areal capacity.

Acknowledgments

This work was supported by the Assistant Secretary forEnergy Efficiency and Renewable Energy, Office of VehicleTechnologies of the U.S. Department of Energy underContract no. DE-EE0006447.

217High areal capacity of Si-based anode

Appendix A. Supporting information

Supplementary data associated with this article can befound in the online version at http://dx.doi.org/10.1016/j.nanoen.2014.04.006.

References

[1] M. Armand, J.M. Tarascon, Nature 451 (2008) 652–657.[2] M.S. Whittingham, Chem. Rev. 104 (2004) 4271–4302.[3] B. Kang, G. Ceder, Nature 458 (2009) 190–193.[4] B.A. Boukamp, G.C. Lesh, R.A. Huggins, J. Electrochem. Soc.

128 (1981) 725–729.[5] M. Winter, J.O. Besenhard, M.E. Spahr, P. Novák, Adv. Mater.

10 (1998) 725–763.[6] L.Y. Beaulieu, K.W. Eberman, R.L. Turner, L.J. Krause, J.

R. Dahn, Solid-State Lett. 4 (2001) A137–A140.[7] A. Magasinski, P. Dixon, B. Hertzberg, A. Kvit, J. Ayala,

G. Yushin, Nat. Mater. 9 (2010) (461–461).[8] 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) 3004–3010.

[9] I. Kovalenko, B. Zdyrko, A. Magasinski, B. Hertzberg,Z. Milicev, R. Burtovyy, I. Luzinov, G. Yushin, A. Major, Science334 (2011) 75–79.

[10] N.-S. Choi, K.H. Yew, K.Y. Lee, M. Sung, H. Kim, S.-S. Kim, J.Power Sources 161 (2006) 1254–1259.

[11] H. Wu, G. Chan, J.W. Choi, I. Ryu, Y. Yao, M.T. McDowell, S.W. Lee, A. Jackson, Y. Yang, L. Hu, Y. Cui, Nat. Nanotechnol. 7(2012) 310–315.

[12] C. Wang, H. Wu, Z. Chen, M.T. McDowell, Y. Cui, Z. Bao, Nat.Chem. 5 (2013) 1042–1048.

[13] N. Liu, H. Wu, M.T. McDowell, Y. Yao, C. Wang, Y. Cui, A. Yolk-Shell, Nano Lett. 12 (2012) 3315–3321.

[14] S. Chen, M.L. Gordin, R. Yi, G. Howlett, H. Sohn, D. Wang,Phys. Chem. Chem. Phys. 14 (2012) 12741–12745.

[15] J.K. Lee, K.B. Smith, C.M. Hayner, H.H. Kung, Chem. Com-mun. 46 (2010) 2025–2027.

[16] M. Holzapfel, H. Buqa, W. Scheifele, P. Novak, F.-M. Petrat,Chem. Commun. 12 (2005) 1566–1568.

[17] L.-F. Cui, Y. Yang, C.-M. Hsu, Y. Cui, Nano Lett. 9 (2009)3370–3374.

[18] L. Hu, H. Wu, Y. Gao, A. Cao, H. Li, J. McDough, X. Xie,M. Zhou, Y. Cui, Adv. Energy Mater. 1 (2011) 523–527.

[19] 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) 1351–1357.

[20] Q. Xiao, Y. Fan, R. Susantyoko, X. Wang, Q. Zhang, EnergyEnviron. Sci. 7 (2014) 655–661.

[21] J. Yin, M. Wada, K. Yamamoto, Y. Kitano, S. Tanase, T. Sakai, J.Electrochem. Soc. 153 (2006) A472–A477.

[22] R. Yi, F. Dai, M.L. Gordin, S. Chen, D. Wang, Adv. Energy Mater.3 (2013) 295–300.

[23] R. Yi, F. Dai, M.L. Gordin, H. Sohn, D. Wang, Adv. Energy Mater.3 (2013) 1507–1515.

[24] R. Yi, J. Zai, F. Dai, M.L. Gordin, D. Wang, Electrochem.Commun. 36 (2013) 29–32.

[25] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183–191.[26] X. Li, G. Zhang, X. Bai, X. Sun, X. Wang, E. Wang, H. Dai, Nat.

Nanotechnol. 3 (2008) 538–542.[27] Z. Song, T. Xu, M.L. Gordin, Y.-B. Jiang, I.-T. Bae, Q. Xiao,

H. Zhan, J. Liu, D. Wang, Nano Lett. 12 (2012) 2205–2211.[28] D. Wang, D. Choi, J. Li, Z. Yang, Z. Nie, R. Kou, D. Hu,

C. Wang, L.V. Saraf, J. Zhang, I.A. Aksay, J. Liu, ACS Nano 3(2009) 907–914.

[29] D. Wang, R. Kou, D. Choi, Z. Yang, Z. Nie, J. Li, L.V. Saraf,D. Hu, J. Zhang, G.L. Graff, J. Liu, M.A. Pope, I.A. Aksay, ACSNano 4 (2010) 1587–1595.

[30] H. Wang, L.-F. Cui, Y. Yang, H. Sanchez Casalongue, J.T. Robinson, Y. Liang, Y. Cui, H. Dai, J. Am. Chem. Soc. 132(2010) 13978–13980.

[31] D. Lv, M.L. Gordin, R. Yi, T. Xu, J. Song, Y.-B. Jiang, D. Choi,D. Wang, Adv. Funct. Mater. 24 (2014) 1059–1066.

[32] B.A. Johnson, R.E. White, J. Power Sources 70 (1998) 48–54.[33] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958)

(1339–1339).[34] N.I. Kovtyukhova, P.J. Ollivier, B.R. Martin, T.E. Mallouk, S.

A. Chizhik, E.V. Buzaneva, A.D. Gorchinskiy, Chem. Mater. 11(1999) 771–778.

[35] X. Zhou, Y.-X. Yin, L.-J. Wan, Y.-G. Guo, Adv. Energy Mater. 2(2012) 1086–1090.

[36] D. Li, M.B. Muller, S. Gilje, R.B. Kaner, G.G. Wallace, Nat.Nanotechnol. 3 (2008) 101–105.

[37] J.-I. Lee, N.-S. Choi, S. Park, Energy Environ. Sci. 5 (2012)7878–7882.

[38] Y. Xu, R. Yi, B. Yuan, X. Wu, M. Dunwell, Q. Lin, L. Fei, S. Deng,P. Andersen, D. Wang, H. Luo, J. Phys. Chem. Lett. 3 (2012)309–314.

[39] D. Chen, R. Yi, S. Chen, T. Xu, M.L. Gordin, D. Wang, SolidState Ion. 254 (2014) 65–71.

[40] M. Zhou, F. Pu, Z. Wang, T. Cai, H. Chen, H. Zhang, S. Guan,Phys. Chem. Chem. Phys. 15 (2013) 11394–11401.

[41] J.-Z. Wang, C. Zhong, S.-L. Chou, H.-K. Liu, Electrochem.Commun. 12 (2010) 1467–1470.

[42] X. Zhao, C.M. Hayner, M.C. Kung, H.H. Kung, Adv. EnergyMater. 1 (2011) 1079–1084.

[43] A. Hu, X. Chen, Y. Tang, Q. Tang, L. Yang, S. Zhang, Electro-chem. Commun. 28 (2013) 139–142.

Ran Yi received his B.S. (2007) and M.S.(2010) in Materials Science from the CentralSouth University, China. Now he is a Ph.D.candidate in Department of MechanicalEngineering, Penn State University workingin Dr. Donghai Wang's Energy Nanostructure(E-Nano) lab. His research focuses on devel-opment of functional materials for energyconversion and storage.

Jiantao Zai received his Bachelor (2007) andPh.D. (2012) degrees in Applied Chemistryfrom the Shanghai Jiao Tong University. Thenhe worked with Prof. Donghai Wang at thePenn State University. He joined ShanghaiJiao Tong University as a research associate.His research interests include the controlledsynthesis of functional inorganic materialsand their potential applications in energystorage, photovoltaic and photocatalytic

areas.

Fang Dai received his Ph.D. degree inInorganic Chemistry from the University ofDelaware in 2012, and since then worked asa postdoc fellow at the Pennsylvania StateUniversity. His research interest is focusedon materials synthesis for energy storageand energy conversion.

R. Yi et al.218

Mikhail L. Gordin is currently a doctoralstudent working with Dr. Donghai Wang'sgroup at Penn State University. He receivedhis BSE degree in Mechanical Engineeringfrom the Duke University in 2009. Hisresearch interests include development andcharacterization of new battery materialsand in-situ battery analysis.

Dr. Donghai Wang is currently AssistantProfessor at Department of Mechanical Engi-neering at The Pennsylvania State Univer-sity. Before joining Penn State in 2009, hewas a postdoc and subsequently became astaff scientist at Pacific Northwest NationalLaboratories. He received a B.S. and Ph.D.degree in Chemical Engineering from theTsinghua University and Tulane University in

1997 and 2006, respectively. Dr. Donghai Wang's research interestshave been related to design and synthesis of nanostructuredmaterials for a variety of applications. His recent research isfocused on materials development for energy storage technologiessuch as Li-ion batteries and beyond Li-ion batteries.