Xe

JPa

b

e

a

ARRAA

KLCGSEX

1

ibcapcsslcarep

a

0d

Applied Surface Science 257 (2011) 9110– 9119

Contents lists available at ScienceDirect

Applied Surface Science

jou rn al h om epa g e: www.elsev ier .com/ locate /apsusc

PS, XRD and SEM characterization of a thin ceria layer deposited onto graphitelectrode for application in lithium-ion batteries

olanta Swiatowskaa,∗, Virginie Lairb, Catarina Pereira-Nabaisa, Gérard Coteb,hilippe Marcusa, Alexandre Chagnesb,∗

Laboratoire de Physico-Chimie des Surfaces, CNRS (UMR 7045), Ecole Nationale Supérieure de Chimie de Paris (Chimie ParisTech), 11 Rue Pierre et Marie Curie, 75005 Paris, FranceLaboratoire d’Electrochimie, Chimie des Interfaces et Modélisation pour l’Energie, CNRS (UMR 7575), Ecole Nationale Supérieure de Chimie de Paris (Chimie ParisTech), 11 Rue Pierret Marie Curie, 75005 Paris, France

r t i c l e i n f o

rticle history:eceived 1 December 2010eceived in revised form 23 May 2011ccepted 24 May 2011vailable online 1 June 2011

eywords:ithium ion batteryeria

a b s t r a c t

Thin ceria layer deposited by electro-precipitation onto graphite was synthesised and characterized by X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and scanning electron microscopy (SEM).The electro-precipitated ceria has a cubic structure with nanocrystallites of about 6 nm. The SEM analysesshows that the ceria layer reflects the morphology of the graphite electrode, exhibits small cracks usuallyfound on the electro-precipitated films but covers almost completely the surface of the graphite. Theceria layer is composed of 75% Ce(IV) and 25% Ce(III) oxides as indicated by the XPS analyses. Cyclicvoltammetry and galvanostatic charge–discharge tests in ethylene carbonate/dimethyl carbonate (1/1)(wt/wt) in the presence of 1 M LiPF6 show that reversible lithium insertion and deinsertion occurs in

raphiteEI layerlectro-precipitationPS

the graphite/ceria electrode and that the ceria layer on the graphite electrode prevents from the loss ofcapacity during the first four cycles. The reduction of the electrolyte occurs at about 0.7 V vs Li/Li+ onboth electrodes but XPS and SEM analyses show that the SEI layer is thin and not as homogenous onthe graphite as on the graphite/ceria electrode. The composition of the SEI layer on the graphite/ceriaelectrode, mainly composed of Li2CO3, ROCO2Li, R–CH2OLi and LiF, is different than those obtained onthe graphite.

. Introduction

Efficient lithium-ion batteries must exhibit high charge capac-ty, high rate capability and also long cycle life. It is generallyelieved that one of the most important prerequisites for goodycling stability of Li-ion batteries is the formation of a completend stable passivating layer, called also solid electrolyte inter-hase (SEI) layer [1] at the negative electrode during the initialharge–discharge cycles. In case of a graphite electrode, a good pas-ivating layer avoids an exfoliation of the graphite by removing theolvation shell around lithium ions. The quality of the passivatingayer formed on the graphite electrode depends on the electrolyteomposition. Nowadays, ethylene carbonate (EC) is commonly useds a co-solvent in electrolytes for lithium-ion batteries because itseduction leads to the formation of a good passivating layer. Nev-

rtheless, during the charge–discharge cycles, the ageing of thisassivating layer induces a drop of charge capacity.

∗ Corresponding authors.E-mail addresses: jolanta-swiatowska@chimie-paristech.fr (J. Swiatowska),

lexandre-chagnes@ens.chimie-paristech.fr (A. Chagnes).

169-4332/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.apsusc.2011.05.108

© 2011 Elsevier B.V. All rights reserved.

In order to improve the quality of the SEI layer and the electro-chemical performances of the Li-ion batteries, numerous studiesreported in the literature are related to the development of newelectrolytes. For instance, Chagnes et al. studied the influence ofvarious electrolytes on the quality of the SEI and the cycling abilityat graphite electrodes [2–6].

Recently, some studies have shown an improvement of theelectrochemical performances of the graphite electrodes aftermodification of the surface structure by mild oxidation [7,8], depo-sition of metals oxides [9–11], polymer coatings [12,13] and coatingwith other kinds of carbons [14]. In fact, metal or metal oxide depo-sition onto graphite electrodes improves the reversible capacity bylimiting the exposure of the active edge sites to the electrolyte [15].For instance, metal oxides containing Sn, Cu, Ni, Fe or Pb depositedon the surface of graphite display higher reversible capacity, betterrate capability and longer cycle life in comparison with bulk metaloxides [16–18].

Another approach for improving the electrical performances ofnegative material for lithium-ion batteries could be a deposition of

a “synthetic” passivating layer on the graphite electrode by electro-precipitation. This “synthetic” passivating layer should have a goodadhesion to the graphite, allow for a reversible intercalation anddeintercalation of lithium and prevent a co-intercalation of solvent

rface S

mdoews

---

-

gc

fTtgi

btstacpgat

2

2

epAcee

faMidbp−tsfl

gwd[

h

J. Swiatowska et al. / Applied Su

olecules into graphite. Ceria (CeO2) seems to be a good candidateue to its insulating properties that should prevent the reductionf the electrolyte responsible for irreversible capacity fading. Sev-ral authors have proposed a modification of the graphite surfaceith ceria by the immersion of a graphite electrode in a cerium (IV)

ulfate solution [19–23], which leads to:

the formation of a dense oxide layer onto the graphite surface the increase of the number of nanochannels, micropores the inhibition of the movement of the graphene sheets along a-axis

the decrease of the imperfection of the graphite such as sp3-hydridized carbon atoms, carbon chains and carbon radicals

It was demonstrated that the deposition of ceria layer on theraphite by this method improves the stability and the electro-hemical performances of the graphite electrode [24].

Electro-precipitation is another method that can be appliedor the deposition of thin ceria layer on the electrode surface.o best of our knowledge, no work concerns the investigation ofhe physicochemical and electrochemical properties of negativeraphite electrodes modified by electro-precipitation for lithium-on batteries.

In this paper, a thin ceria layer was deposited onto graphitey electro-precipitation and the physicochemical properties ofhis material were characterized by X-ray diffraction (XRD),canning electron microscopy (SEM) and X-ray photoelectron spec-roscopy (XPS) before and after lithium insertion/deinsertion in

solution of lithium hexafluorophosphate (LiPF6) in ethylenearbonate/dimethyl carbonate (1/1) (wt/wt). The electrochemicalroperties (the reversible and the irreversible capacities) of theraphite/ceria electrode were investigated by cyclic voltammetrynd galvanostatic charge–discharge and compared to graphite elec-rode.

. Experimental

.1. Electrochemical experiments

Electro-precipitation of ceria was performed in a three-lectrodes electrochemical cell by using graphite (0.7–0.8 cm2)rovided by SAFT Company (France), a platinum grid and ang/AgCl electrode (EAg/AgCl = 0.2 V vs NEH) as working electrode,ounter electrode and reference electrode, respectively. Thelectro-precipitation of ceria was performed with a rotating disclectrode at 300 rpm.

The electro-precipitation of ceria onto graphite was per-ormed at room temperature (20 ± 2 ◦C) in 0.1 mol L−1 NaNO3nd 0.1 mol L−1 Ce(NO3)3 solution prepared with ultra pureillipore® water (resistivity >18 M� cm) and reagent grade chem-

cals (Aldrich). The electrochemical cell was bubbled with pureioxygen (Alpha Gaz 1, 99.995%, Air Liquide) during 30 minefore and during the electro-precipitation of ceria. The electro-recipitation was performed by applying a constant potential of0.8 V vs Ag/AgCl for 90 min using an EG&G PAR model 263A poten-

iostat [25,26]. After the electro-precipitation, the graphite/ceriaample was rinsed with ultra pure water and dried in an argonow at room temperature.

The study of the insertion/deinsertion of lithium ions into theraphite/ceria electrode was performed by cyclic voltammetryith an Autolab (AUT30) potentiostat/galvanostat in a glove-box

irectly connected to the ultra-high vacuum XPS analysis chamber27].

The electrochemical tests were carried out in Swagelokalf-cells (graphite electrode/Li or graphite/ceria electrode/Li) con-

cience 257 (2011) 9110– 9119 9111

taining a polypropylene separator (Celgard® 2000) to separatethe anodic and cathodic compartments. The cyclic voltammetryof the graphite/ceria electrode was performed at 0.1 mV s−1 fromthe open circuit potential (3.2 V vs Li/Li+) to 0.005 V vs Li/Li+ atroom temperature in a solution of 1 M LiPF6 (purity > 99.99%, bat-tery grade, Aldrich) in ethylene carbonate (EC, purity = 99%, Aldrich)and dimethyl carbonate (DMC, purity = 99%, Aldrich) [EC/DMC (1/1)(wt/wt)]. The cyclic voltammetry of the graphite electrode was per-formed between 1.4 and 0.005 V vs Li/Li+.

The galvanostatic charge–discharge tests of half and full-cells ofthe graphite/ceria and graphite electrode were operated at C/20charge and discharge during 20 h in the same electrolyte. Half-cells were charged at constant current to a cut-off voltage of 0.01 Vvs Li+/Li. Discharge was operated at constant current, to a cut-offvoltage of 3 V vs Li+/Li.

2.2. X-ray diffraction (XRD)

X ray diffractograms were recorded with an X’Pert Pro appara-tus from PANalytical using Cu K� radiation (� = 1.5406 A), equippedwith an incident beam Ge (1 1 1) monochromator and a linear Pix-Cell detector (active length 14 mm). XRD data were recorded in ascanning mode between 15◦ and 0◦ at every 0.026◦ 2�. X-ray gen-erator worked at 45 kV and 40 mA. All the data were processed byX’Pert HighScore software with his commercial databases (back-ground deduction, profile fitting, and peaks identification).

2.3. X-ray photoelectron spectroscopy (XPS)

XPS analyses were carried out with a VG ESCALAB 250 spec-trometer equipped with a UHV preparation chamber directlyconnected to the glove box [27]. Base pressure during analy-sis was ∼1 × 10−9 mbar. An Al K� monochromatized radiation(h� = 1486.6 eV) was employed as X-ray source. For all analyses,the take-off angle of the photoelectrons was 90◦. High resolu-tion spectra of the Ce 3d, O 1s, C 1s, N 1s and Li 1s core levelregions were recorded with a pass energy of 20 eV. The data pro-cessing (peak fitting) was performed with the Avantage softwareprovided by Thermo Electron Corporation, using a Shirley typebackground subtraction and Gaussian/Lorentzian peak shapes. Thebinding energies were corrected by setting the C 1s hydrocarbon(–CH2–CH2–bonds) peak at 285.0 eV for the graphite electrodes andthe uIII line at 917.0 eV for the graphite/ceria samples. The XP spec-tra were recorded on the pristine graphite, graphite/ceria samplesand after two cycles of lithium insertion/deinsertion in a deinsertedstate.

2.4. Scanning electron microscopy (SEM)

SEM pictures were obtained from a scanning electronic micro-scope LEO DSM 892 Gemini equipped with energy-dispersive X-ray(EDX) analyzer. The samples were washed with dry DMC and theremaining solvent was evaporated at room temperature.

3. Results and discussion

3.1. Mechanisms of electro-precipitation of ceria

During the electro-precipitation of ceria onto graphite at −0.8 Vvs Ag/AgCl in an aqueous solution of NaNO3 and Ce(NO3)3,the following electrochemical reactions occur in addition to the

electro-precipitation of ceria:

O2 + 2H2O + 4e− = 4OH−, E◦ = 0.40 V/NHE (1)

H2O + e− = 1/2H2 + OH−, E◦ = −0.83 V/NHE (2)

9 rface

N

riotedp

oet

4

C

a

C

C

dtoif

3S

cegpcdc

ibscst(cyact

eopptooa

g(

112 J. Swiatowska et al. / Applied Su

O3− + H2O + 2e− = NO2

− + 2OH−, E◦ = 0.01 V/NHE (3)

The formation of hydroxide in the vicinity of the electrode isesponsible for a local increase of the pH of the solution lead-ng to the precipitation of cerium oxide or cerium hydroxidento the electrode. This method was widely used for the elec-rochemical preparation of zinc oxide [28]. The mechanisms oflectro-precipitation are not yet completely understood as the oxi-ation of Ce(III)–Ce(IV) can occur either in the solution, before therecipitation step [29] or during the drying process [30].

Recently, different groups [31–33] have suggested that Ce(III) isxidized to Ce(OH)2

2+ by the molecular dioxygen dissolved in thelectrolyte followed by the precipitation of Ce(OH)2

2+–CeO2 due tohe local alkaline pH near the electrode:

Ce3+ + O2 + 4OH− + 2H2O = 4Ce(OH)22+ (4)

e(OH)22+ + 2OH− = CeO2 + 2H2O (5)

The precipitation of cerium(III) hydroxide further oxidized maylso occur according to the following reactions:

e3+ + 3OH− = Ce(OH)3 (6)

e(OH)3 = CeO2 + H3O+ + e− (7)

Then, the electro-precipitation in the presence of molecularioxygen leads to formation of a solid thin film containing Ce(III) inhe form of Ce(OH)3 and/or Ce2O3 (vide infra) and Ce(IV) in the formf CeO2. The aim of this paper is not to elucidate the mechanismsnvolved in the electro-precipitation of ceria but to use this methodor depositing ceria onto graphite electrodes.

.2. Characterizations of the graphite/ceria electrode by XRD,EM and XPS

X-ray diffraction (XRD) analyses of the graphite/ceria samplesonfirm the presence of ceria (mainly CeO2) onto graphite afterlectro-precipitation (Fig. 1). The presence of a ceria layer ontoraphite is responsible for a decrease of the intensity of the XRDeaks related to graphite (Fig. 1, inset). The XRD pattern is typi-al for cubic ceria according to the JCPDS-International Centre foriffraction data (43-1002 file). The estimated size of the crystallitesalculated by the Debye–Scherrer formula was about 6 nm.

The morphology of the graphite presented in Fig. 2A is rough ast consists of graphite flakes. The EDX mapping of cerium and car-on on the graphite electrode after electro-precipitation of ceriahows that the distribution of cerium (red color) and carbon (blueolor) is not homogeneous (Fig. 2B). Indeed, the nucleation of ceriaeems to depend strongly on the morphology of the graphite. Then,he process of precipitation may occur more easily in the hollowsbetween flakes) than on the top of the graphite flakes. Even if noerium was observed in the EDX mapping (Fig. 2B), the EDX anal-ses performed at longer acquisition time confirm the presence of

thin layer of cerium in the blue regions. Therefore, it can be con-luded, that the cerium covers the surface of the graphite but thehickness of the electro-precipitated layer is not homogeneous.

Fig. 3C displays SEM image of graphite electrode after cerialectro-precipitation. For the sake of comparison, the SEM imagef pristine graphite (before electro-precipitation) has been alsoresented in Fig. 3A. The electro-precipitated ceria reflects the mor-hology of the graphite electrode and exhibits small cracks onhe surface. This kind of porous or cracked ceria layer previouslybserved by Kulp et al. [34] is most likely due to the formationf loosely adhered nanoparticles generated by oxygen evolution,

long with the thin film deposited electrochemically.

Fig. 4 presents the XP spectra of ceria electro-precipitated ontoraphite. Binding energies (BE) and full widths at half-maximumFWHM) are compiled in Table 1. The Ce 3d core level shows the

Science 257 (2011) 9110– 9119

presence of two types of cerium oxides: Ce(III) and Ce(IV) (Fig. 4A).The shape of the Ce 3d spectra reveals the presence of Ce(III) oxideas observed previously in the literature [35,36]. Both CeO2 andCe2O3 spectra are composed of two multiplets (v and u). These mul-tiplets correspond to the spin–orbit split 3d5/2 and 3d3/2 core holes,which is around 18 eV. The intensity ratio I(3d5/2)/I(3d3/2) was fixedto 1.5. For the CeO2, the Ce 3d3/2,5/2 spectra are composed of threemultiplets which were already observed by other authors [37–51].The lowest binding energy peaks v, u and v′′, u′′ located at 882.7,901.3, 888.9 and 907.7 eV, respectively are the results of Ce 3d94f2

O 2p4 and Ce 3d94f1 O 2p5 final states. The highest binding energypeaks, v′′′ and u′′′ located at about 898.6 and 917.0 eV, respectively,are the result of a Ce 3d94f0 O 2p6 final state. The satellite peaku′′′ associated to the Ce 3d3/2 is characteristic of the presence oftetravalent Ce (Ce4+ ions) in Ce compounds.

For the Ce2O3 (Fig. 4A), the Ce 3d3/2,5/2 spectra are composedof two multiplets (v and u) corresponding to split 3d5/2 and 3d3/2core holes [37–42,45–48]. The lowest binding energy peaks v◦ andu◦ located at 880.8 and 899.0 eV, respectively, correspond to Ce3d94f2 O2p5 final state. The highest binding energy peaks v′ and u′

located at about 884.6 and 902.6 eV, respectively, can be attributedto Ce 3d94f1 O2p6 final state.

The O 1s core level peak (Fig. 4B and Table 1) detected at about529.5 eV is related to the lattice oxygen in CeO2. For Ce(III) com-pounds, the O 1s core level peak is shifted to higher binding energyat 530.7 eV (Fig. 4B and Table 1) [52]. The major part of ceriumoxide is constituted of Ce(IV) as the areas of the O 1s peaks foroxygen bonded with Ce(IV) and Ce(III) are equal to 75% and 25%,respectively. The amounts of Ce(IV) and Ce(III) calculated from theCe 3d core level show almost the same values. The calculation wasperformed using the following equations [35,36,53]:

Ce(III) = v◦ + u◦ + v′ + u′ (8)

Ce(IV) = v + u + v′′ + u′′ + v′′′ + u′′′ (9)

Ce(III) = Ce(III)Ce(III) + Ce(IV)

(10)

The O 1s core level spectrum exhibits also a peak at 532.8 eV (O1sD) attributed to hydroxide, ether oxygen and/or nitrates whichremain on the surface after the ceria electro-precipitation in NaNO3and Ce(NO3)3 aqueous solution. The XP spectra performed on theCe(NO3)3 reference sample (not presented here), show the O 1speak almost at the same binding energy. The presence of nitrateson the graphite/ceria sample is also confirmed by the N 1sB peakfound at 407.2 eV and described in Fig. 4C. The O 1sC peak at531.6 eV attributed to C O and/or C–O corresponds to organiccontaminations always present on the surface. The presence of amixture of Ce(III) and Ce(IV) on the graphite/ceria electrode canbe explained by the application of molecular dioxygen during theelectro-precipitation in the electrolytic solution which allows theoxidation of Ce(III)–Ce(IV) during the electro-precipitation accord-ing to reaction (4) or (6) and (7) reported above.

Two N 1s peaks A and B observed at 403.5 and 407.2 eV (Fig. 4Cand Table 1) can be attributed to –NO2 and –NO3 species, respec-tively. The nitrates present on the surface originate from theelectrolyte used for the electro-precipitation of ceria. The –NO3was not completely removed from the sample even though thesample was abundantly rinsed with distilled water after electro-precipitation of ceria in NaNO3–Ce(NO3)3 electrolyte. The –NO2may come from the reduction of –NO3 as the electro-precipitationoccurs at negative potential (−0.8 V vs Ag/AgCl).

The C 1s signal is attributed to the contamination always

detected on the extreme surface (Fig. 4D and Table 1). The C 1s corelevel was fitted with five peaks assigned to graphite at 284.4 eV,–CH2–CH2– bond at 285 eV (C 1sA peak), C–O and/or C–N bonds at286.1 eV (C 1sB peak), O C bond (carboxyl group) at 288.7 eV (C 1sC

J. Swiatowska et al. / Applied Surface Science 257 (2011) 9110– 9119 9113

Fig. 1. XRD spectra of graphite electrode before and after electro-precipitation of ceria. Graphite peaks are marked by asterisks.

Fig. 2. (A) SEM image of graphite/ceria surface, (B) EDX mapping of graphite/ceria sample (Ce red and C blue). (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of the article.)

Fg

ig. 3. SEM images: (A) graphite electrode pristine (without a ceria layer) before andraphite/ceria sample pristine and (D) after two cycles (CV) of lithium insertion/deinserti

(B) after two cycles (CV) of lithium insertion/deinsertion in EC/DMC+LiPF6, (C)on in EC/DMC+LiPF6.

9114 J. Swiatowska et al. / Applied Surface Science 257 (2011) 9110– 9119

A

870880890900910920930

2.0x104

2.5x104

3.0x104

3.5x104

4.0x104

4.5x104

5.0x104(CeO

2)

(Ce2O

3)

experi men tal lin e

fi tting lineIn

ten

sity / c

ps

Binding energy /eV

V0

V

V'

V"

V'''

U

U0

U'

U"

U'' 'Ce3d

(Ce2O

3)

(CeO2)

B

524526528530532534536538540542

5.0x103

1.0x104

1.5x104

2.0x104

2.5x104

Inte

nsity / c

ps

expe rimen tal lin e

fi tting lin e

O1s CeO2

Ce2O

3

C=O-C

C-O

hydro xides

C

400402404406408410

3.0x103

3.2x103

3.4x103

3.6x103

3.8x103

4.0x103

Inte

nsity / c

ps

Binding energy / eV

exp erime ntal line

fitting line

N1s

-NO2

-NO3

D

4x103

5x103

6x103

7x103

8x103

9x103

1x104

1x104

C-O

-CO3

Inte

nsity / c

ps

experimental lin e

fitt ing li ne

C1s

C-C

C-H

graph ite

O=C

phite

pcot

TBip

Binding energy /eV

Fig. 4. XP spectra of ceria electro-precipitated on gra

eak) and carbonates at 290.0 eV (C 1sD peak). The intensity of thearbon peak corresponding to graphite is significantly lower thann the pristine graphite electrode confirming that the graphite elec-rode is almost completely covered by the ceria film as indicated

able 1inding energy (BE) and full-width at half-maximum (FWHM) of the Ce 3d, O 1s, C 1s,

nsertion/deinsertion. The element marked in bold and italics (C or O) stands for the bondresence of the XPS peak at a given binding energy.

Pristine graphite/ceria

Compounds BE (eV) FWHM (eV)

Ce 3d V CeO2 882.7 1.8

Ce 3d U CeO2 901.3 1.6

Ce 3d V′′ CeO2 888.9 4.4

Ce 3d U′′ CeO2 907.7 4.2

Ce 3d V′′′ CeO2 898.6 1.8

Ce 3d U′′′ CeO2 917.0 2.0

Ce 3d V◦ Ce2O3 880.8 2.3

Ce 3d U◦ Ce2O3 899.0 2.0

Ce 3d V′ Ce2O3 884.6 4.3

Ce 3d U′ Ce2O3 902.6 4.5

O 1sA CeO2 529.5 1.1

O 1sB Ce2O3 530.7 1.3

O 1sC C O, C–O 531.6 1.5

O 1sD OH, –NO3, R–O–R 532.8 1.8

O 1sE – – –

C 1sA′ Graphite 284.4 1.1

C 1sA –CH2–CH2– 285.0 1.3

C 1sB C–O and/or C–N 286.1 1.5

C 1sC C O 288.7 1.4

C 1sD –CO3 290.0 1.5

N 1sA –NO2 403.5 1.5

N 1sB –NO3 407.2 1.5

F 1sA – – –

F 1sB – – –

Li 1sA – – –

Li 1sB – – –

280282284286288290292294

Binding energy / eV

: (a) Ce 3d, (b) O 1s, (c) N 1s and (d) C 1s core levels.

previously by EDX (see above). The presence of the graphite peakcan be justified by the formation of cracks in the ceria layer and anon homogenous coverage of the rough graphite electrode by thethin ceria layer during the electro-precipitation.

F 1s and Li 1s peaks for a pristine graphite/ceria sample and after two cycles ofing of this element to a group compound (i.e. carbonate group) responsible for the

Graphite/ceria after 2CV

Compounds BE (eV) FWHM (eV)

– – –– – –– – –– – –– – –– – –– – –– – –– – –– – –CeO2 529.1 1.0Ce2O3 530.9 1.2–CO3 532.0 1.4R–CH2OLi,R–CH2OCO2Li,–C–O–C– (e.g. PEO)

532.9 1.5

R–CH2OCO2Li 533.8 1.5– – ––CH2–CH2– 285.0 1.3–C–O–C– (e.g. PEO) 286.7 1.4R–CH2OLi,R–CH2OCO2Li

288.7 1.3

–CO3, R–CH2OCO2Li 290.3 1.3– – –– – –LiF 685.3 1.5LiFP6 687.4 1.8LiF 55.0 1.7LiFP6 56.0 1.7

J. Swiatowska et al. / Applied Surface Science 257 (2011) 9110– 9119 9115

A

-1.5E-04

-1.0E-04

-0.5E-04

0.0E+00

0.5E-04

1.0E-04

1.5E-04

0.00 0.2 0 0.4 0 0.60 0.80 1.00 1.20 1.40

E vs Li/ Li+

/ V

2nd c ycle

1st cycleE2'

E2

E1

-8E-04

-6E-04

-4E-04

-2E-04

0E+00

2E-04

4E-04

6E-04

8E-04

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50

E vs Li/ Li+ / V

1st c ycle2nd cycle

E2

E2'

E1E3

E3'

B

j / A

.cm

-2j /

A.c

m-2

FtE

3

gttpB

(tvase[

lweect

gcfi

c

w(

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0

100.0

0 1 2 3 4

Grap hite/ce ria electrode

Grap hite electrode

Ch

arg

e e

ffic

ien

cy / %

Cycle number

ig. 5. Two cycles (cyclic voltammetry, scan rate v = 0.1 mV s−1) of lithium inser-ion/deinsertion performed on (A) graphite and (B) graphite/ceria electrode inC/DMC+LiPF6.

.3. Lithium insertion/deinsertion through ceria layer

Lithium insertion/deinsertion into the graphite andraphite/ceria electrodes has been investigated in EC/DMC inhe presence of 1 M LiPF6. The voltammograms of the graphite andhe graphite/ceria electrodes have the same shape in the range ofotential comprised between 1 V and 0.005 V vs Li/Li+ (Fig. 5A and) [3].

The increase of cathodic current from 0.2 V to 0.005 V vs Li/Li+

E2) shows that insertion of lithium occurs into the graphitehrough the ceria layer (Fig. 5B). An anodic peak observed at 0.39 Vs Li/Li+ (E2

′) on the graphite or the graphite/ceria electrode (Fig. 5And B) indicates the process of lithium deinsertion is reversible. Amall cathodic bump observed at around 0.7 V vs Li/Li+ (E1) for eachlectrode can be attributed to the formation of a passivating layer3].

In a case of graphite/ceria electrode, the small cathodic peakocated around 1.5 V vs Li/Li+ (E3

′) and on the reverse scan the

ide peak between 0.8 and 2 V vs Li/Li+ (E3) may correspond tolectrochemical reactions between lithium and ceria as no suchlectrochemical phenomena occur in the absence of ceria [3]. Theyclic voltammetry confirms that lithium ions can diffuse throughhe ceria layer and intercalate reversibly into the graphene layers.

Galvanostatic charge–discharge cycles have been performed onraphite and graphite/ceria electrodes at C/20 (charge and dis-harge during 20 h). The variation of the charge efficiency duringour charge–discharge cycles have been calculated and displayedn Fig. 6:

harge efficiency = Crev(n)Crev(i)

× 100 (11)

here Crev(n) and Crev(i) denote the reversible capacity during then) cycle and the initial cycle (i), respectively.

Fig. 6. Variation of the charge efficiency vs cycle number during galvanostaticcharge–discharge of lithium into graphite and graphite/ceria electrodes at C/20.

Fig. 6 shows that the presence of the ceria layer on the graphiteelectrode prevents from the capacity fading during the first fourcycles. The lost of capacity charge is 2%/cycle for the graphite/ceriaelectrode whereas for the graphite electrode is 6%/cycle duringthe first four cycles. Nevertheless, the formation of the passivat-ing film is still observed on the graphite/ceria electrode as well ason the graphite electrode. The irreversible capacity evaluated fromthe galvanostatic charge–discharge at C/20 is 133 mAh/g (1st cycle)and 11 mAh/g (4th cycle) for the graphite/ceria electrode whereasfor the graphite electrode 51 (1st cycle) and 6 mAh/g (4th cycle).Then, the reduction of electrolyte seems to be more important onthe graphite/ceria electrode. The high decrease of the irreversiblecapacity between the first and the second cycle shows that the SEIonto the graphite/ceria electrode prevents from further reductionof the electrolyte.

Then, this preliminary study of lithium insertion/deinsertioninto graphite/ceria electrode shows that the deposition of the thinceria layer on the graphite electrode could improve the electro-chemical performances of lithium-ion batteries after some furtherimprovements.

3.4. Characterizations of the graphite/ceria electrode afterlithium insertion/deinsertion by XRD, SEM and XPS.

Unmodified XRD patterns presented in Fig. 7 confirm thatneither exfoliation nor significant structural modification of thegraphite/ceria electrode occurs after lithium insertion/deinsertion.Only a slight shift of the XRD graphite peaks to higher val-ues is observed after lithium insertion/deinsertion which can beattributed to an insignificant decrease of the lattice parameters ofthe graphite.

Fig. 3D shows a SEM image of graphite/ceria electrode aftertwo cycles of insertion/deinsertion. For the sake of comparison,SEM images of the graphite electrode (without ceria layer) beforeand after two lithium insertion/deinsertion cycles are presented inFig. 3A and B, respectively. The comparison of these two figuresshows that a passivating layer (SEI layer) is formed on the surfaceof the graphite electrode after lithium insertion/deinsertion. Thisfilm looks porous with small nuclei and does not uniformly coverthe surface of the graphite electrode. Contrary to the graphite elec-trode, the SEI layer formed on the graphite/ceria electrode seems tocover homogenously the surface of the electrode and the size of the

nuclei constituting the SEI layer seems to be bigger (Fig. 3D). Ceriais known as an insulator material but the presence of a mixture ofCe(III) and Ce(IV) may get this layer conductive and then, the elec-

9116 J. Swiatowska et al. / Applied Surface Science 257 (2011) 9110– 9119

efore

ts

ii1afanXt(ttsa[owrt

Ftd

graphite/ceria electrode after two cycles of insertion/deinsertionwhere four peaks can be distinguished: a component at 285 eV (C1sA) which corresponds to –CH2–CH2– bonds, a peak at 286.7 eV

A

3

5x103

6x103

7x103

8x103

9x103

C-C

C-H

Inte

nsity /

cps

experimental lin e

fitt ing line

C1s

-C-O-C-

-CO3

R-CH2OCO

2Li

R-CH2OLi

R-CH2OCO

2Li

Fig. 7. XRD spectra of graphite/ceria electrode performed b

rolyte (LiPF6 in EC/DMC) is reduced and forms the SEI layer on theurface of the graphite/ceria electrode.

The XPS survey spectra (Fig. 8) shows significant mod-fication of the graphite/ceria electrode after two cycles ofnsertion/deinsertion. The presence of high intensity C 1s and Os peaks and appearance of new peaks i.e. Li 1s, F 1s and P 2p (Fig. 8nd Table 1) confirms the formation of the SEI layer on the sur-ace of the electrode. The thick SEI layer (more than 5 nm) stronglyttenuates the XP signal of Ce 3d core level (inset in Fig. 8). No sig-ificant modification of the graphite/ceria electrode observed byRD proves that the low intensity Ce 3d signal may be only due to

he formation of a thick SEI layer and not due to the destructioni.e. an exfoliation) of the ceria layer. The effect of attenuation ofhe Ce 3d signal due to formation of the thick SEI layer after elec-rochemical insertion/deinsertion is consistent with our previoustudies performed on negative electrode materials [54–56] as wells the studies performed by the group of Gonbeau [57] or Edström58]. The XPS results are in a good agreement with the SEM imagesf the graphite/ceria electrode where the formation of the SEI layer

as also observed. One can observe a small intensity signal in the

ange of Ce 3d core level at lower binding energy that can indicatehe reduction of Ce(IV) and the formation of Ce(III) oxide. However,

1200 1100 1000 900 800 700 600 500 400 300 200 100 0

940 93 0 92 0 91 0 90 0 89 0 88 0

Inte

nsity / a

.u.

Binding energy / eV

Ce3 d

graph ite/ce ria aft er

Li i nse rtion /deinse rtion

graph ite/ce ria pristine

Inte

nsity

/ a.

u.

Binding energy / eV

graphi te/ce ria after

Li insertion/d einse rtio n

graphi te/ce ria p ristin e

O1s

C1s

F1s

N1s Ce4p

P2s

P2p

Ce4d

Ce Auge r

Ce Auge r

F Auge r

Ce 3dO Auge r

Li1s

ig. 8. XP survey spectra of the graphite/ceria electrode performed before (pris-ine) and after two cycles of lithium insertion/deinsertion. The Ce 3d core level isisplayed in the inset of the Figure.

and after two cycles (CV) of lithium insertion/deinsertion.

no conclusion about the ceria reduction can be deduced due to thelow intensity signal and the possible superposition of the peak offluorine Auger (F KL3 and F KL2) that can be also observed at thesame binding energy.

Fig. 9A (see also, Table 1) shows the C 1s core level for the

278280282284286288290292294

1x103

2x103

3x103

4x10

Binding energ y / eV

B

524526528530532534536538540542

0.0

5.0x103

1.0x104

1.5x104

2.0x104

R-CH2OLi

R-CH2OCO

2Li

-C-O-C-

R-CH2OCO

2Li

-CO3

Inte

nsity /

cps

Bin ding energy /eV

exp eri mental li ne

fi tting li ne

O1s

CeO2

Ce2O

3

Fig. 9. XP spectra of the graphite/ceria electrode of C 1s (a) and O 1s (b) core levelsafter two cycles of lithium insertion/deinsertion.

rface Science 257 (2011) 9110– 9119 9117

(spR1ipD

1dpsppgctnbii

aw(p1

(e(aat(pt

gapaaeTg2a(t

ata2tR–

iccerc

A

278280282284286288290292294296

0.0

5.0x103

1.0x104

1.5x104

2.0x104

2.5x104

3.0x104

graphi te

C-C

C-H

Inte

nsity / c

ps

Bindin g energ y / eV

exper imental line

fitting lin eC1s

C-O

O=C

B

278280282284286288290292294296

0

1x103

2x103

3x103

4x103

5x103

6x103

ROLi

-CO3

R-CH2OCO

2Li

R-CH2OLi

R-CH2OCO

2Li

-C-O-C-

graphite

C-C

C-H

Inte

nsity / c

ps

Bindi ng energ y / eV

experimen tal li ne

fitting lin eC1s

-CF2

J. Swiatowska et al. / Applied Su

C 1sB peak) attributed to –C–O–C– bonds (polymeric compounduch as poly(ethylene oxide) PEO (–CH2–CH2–O–)n), and twoeaks at 288.7 and 290.3 corresponding to Li-alkyl carbonates:–CH2OCO2Li, R–CH2OLi (C 1sC peak) and –CO3, R–CH2OCO2Li (CsD peak), respectively [58–61]. The carbonate species have been

dentified several times as the main components of the SEI layerresent on graphite electrode due to reductive decomposition ofMC and EC [62–64].

In peculiar, a significant increase can be observed for the CsB peak corresponding to polymeric species resulted from theecomposition and the polymerization of EC [65–68]. The secondrincipal composition of the SEI layer is the C 1sD peak corre-ponding to –CO3, R–CH2OCO2Li. The less intense peak is the C 1sCeak corresponding to R–CH2OLi and R–CH2OCO2Li species. Theeak corresponding to graphite is not anymore observed on theraphite/ceria electrode after two cycles of insertion/deinsertiononfirming that thickness of the SEI layer is higher than the detec-ion limit of the XPS. It was also verified that the SEI layer doesot form spontaneously on the graphite/ceria electrode as no car-on peak characteristic for the SEI layer was observed by XPS after

mmersion of the sample in the electrolyte of LiPF6 in EC/DMC dur-ng 24 h.

In the O 1s core level region, the very low intensity peaks O 1sAnd O 1sB correspond to CeO2 and Ce2O3, respectively (Table 1),hich is in agreement with the weak Ce 3d signal. The O 1sD

R–CH2OLi, R–CH2OCO2Li, –C–O–C–) and O 1sE (R–CH2OCO2Li)eaks can be attributed mainly to Li-alkyl carbonates and the OsC to lithium carbonates (Li2CO3) [62–68].

The presence of fluorine and phosphorus containing compounds–PF6 at and LiF) is still observed on the graphite/ceria electrodeven after rinsing the sample with a pure dimethyl carbonateFig. 8). Two peaks were distinguished at F 1s core level region: onet 685.3 eV and another one at 687.4 eV that can be attributed to LiFnd LiPF6, respectively [69–71]. The binding energy of the F 1s spec-rum corresponding to the LiF is little higher than usually observedat around 686 eV) on graphite electrodes. The displacement of thiseak at higher binding energy could be due to abundant rinsing ofhe sample before the XPS analysis as observed previously [72].

For the purpose of comparison of the graphite and theraphite/ceria electrodes, the C 1s spectra for the pristine graphitend for the graphite after two cycles of insertion/deinsertion areresented in Fig. 10A and B, respectively. Binding energies (BE)nd full-widths at half-maximum (FWHM) for graphite electrodere compiled in Table 2. The narrow (FWHM = 0.6 eV) low bindingnergy peak observed at 284.5 ± 0.1 eV corresponds to graphite.hree other peaks can be observed on the surface of the pristineraphite: C 1sA present at 285.0 eV, C 1sB at 286.9 eV and C 1sC at88.5 eV which correspond to hydrocarbons, C–O contaminationsnd CO2 environment from the sodium carboxymethyl celluloseCMC) binder. The Na 1s peak observed at around 1072 eV confirmshe presence of CMC in the graphite electrode.

The thin graphite peak (C 1sA′ ) is still present on the surfacefter two cycles of lithium insertion/deinsertion, which indicateshat the SEI layer is thin and/or not as homogenous on the graphites on the graphite/ceria electrode. The new peaks observed at 286.5,87.6, 288.6, 290.1 and 292.0 eV after two cycles can be attributedo –C–O–C– bonds (e.g. PEO)–(C 1sB), ROLi (e.g. LiOCH3)–(C 1sC),–CH2OCO2Li and/or R–CH2OLi–(C 1sD), R–CH2OCO2Li–(C 1sE) andCF2–(C 1sF), respectively [62–68].

The most significant peak present on the graphite electrode sim-larly to graphite/ceria electrode is the peak at 286.5 eV (C 1sB)orresponding to the polymeric compounds (PEO). Completely new

ompound (C 1sC peak), not observed before on the graphite/cerialectrode, is the lithium methoxide (LiOCH3). This compound,eported previously by Zhuang et al. [73], can overlap with the peakorresponding to Li-carboxylate. Lithium alkoxides are known as

Fig. 10. XP spectra of C 1s core level of the graphite electrode before after two cyclesof lithium insertion/deinsertion.

not very stable components of the SEI layer under the UHV condi-tions of the XPS due to rapid reaction with water and formation ofthe corresponding alcohols. The other important products of elec-trolyte decomposition are Li-alkyl carbonates observed regularlyon the graphite electrodes after cycling. The second carbon peaknot observed on the graphite/ceria electrode is the peak at 292.0 eV(C 1sD) which can be assigned to fluorinated carbons (–CF2–). Thefluorinated carbon was not observed on the pristine graphite elec-trode (Fig. 10A). According to Stjerndahl et al. [74], the formation ofthis product on the cycled graphite results from reaction betweenthe PVDF binder and the electrolyte.

The O 1s core level region shows the presence of three peaks.The strongest O 1s peak at approximately 532.8 eV originatesfrom Li-alkyl carbonates (R–CH2OCO2Li), Li-alcoxides (R–CH2OLi)and polymer compounds (–C–O–C–). At the higher bindingenergy, the peak at 533.8 eV originates from Li-alkyl carbon-ates (R–CH2OCO2Li). The third C 1s peak at 535.2 eV (O 1sD)can be assigned to LixPFyOz compound coming from degrada-tion of LiPF6 electrolyte salt [65,75]. The oxygen species foundat so high binding energy can also correspond to polymer com-pounds like organic-fluorinated and/or organofluoro-phosphorous(i.e. C2H4OF2P) compounds that have been already reported by Aur-bach et al. [76]. The high binding energy peaks found in the area ofLi 1s and F 1s core levels can be attributed to polymeric compoundsissue of electrolyte decomposition. The lithium peak Li 1s at 57.4 eV(Table 2) is quite symmetric thus a decomposition and peaks con-tribution could not be reliable. It can be assumed that the lithium

peak corresponds to a mixture of organic and inorganic (princi-pally LiFP6) lithium compounds. However, the binding energy ofthe lithium peak corresponds to lithium graphite (LiC6) [77]. The

9118 J. Swiatowska et al. / Applied Surface Science 257 (2011) 9110– 9119

Table 2Binding energy (BE) and full-width at half-maximum (FWHM) of the O 1s, C 1s, F 1s and Li 1s peaks for a pristine graphite sample and after two cycles of insertion/deinsertion.

Pristine graphite Graphite after 2CV

Compounds BE (eV) FWHM (eV) Compounds BE (eV) FWHM (eV)

O 1sD C O and/or C–O (CMC binder) 532.0 1.6 R–CH2OLi, R–CH2OCO2Li, –C–O–C (e.g. PEO), ROLi 532.8 1.6O 1sE OH, R–O–R 533.6 1.6 R–CH2OCO2Li 533.8 1.5O 1sF – – – LixPFyOz 535.2 1.8C 1sA

′ Graphite 284.5 0.55 Graphite 284.6 0.7C 1sA –CH2–CH2– 285.0 1.3 –CH2–CH2– 285.0 1.2C 1sB –C–O–C– (e.g. PEO) 286.5 1.3C 1sC C–O 286.9 1.3 ROLi (e.g. LiOCH3) 287.6 1.3C 1sD C O and/or CO2 (CMC binder) 288.5 1.2 R–CH2OLi, R–CH2OCO2Li 288.6 1.3C 1sE –CO3, R–CH2OCO2Li 290.1 1.6C 1sF – – – –CF2 292.0 1.6

Mi

Mi

flaoao1me

gXtptLrtrnam

4

aTetc

gcoitlgecp

ohgtar

[

[

[[

[[

[

F 1sA – – –

F 1sB – – –

Li 1s – – –

uorine peaks placed also at high binding energies (F 1sA and F 1sBt 687.0 and 689.0 eV, respectively) indicate on a polymerizationf organic solvent molecules such as polyethers and polycarbon-tes [65,75,78]. The presence of –CF2 compound found in the areaf C 1s core level region can be confirmed by the presence of Fs peak found at 689.0 eV. However, the presence of other poly-eric compounds (–CHFCHF–)n [79], (–CHFCH2–)n [80] cannot be

xcluded.The formation of the SEI layer is significantly different on the

raphite/ceria than on the graphite electrode as observed by thePS analysis. The composition of both layers formed on these two

ypes of electrodes is quite complex and attributions of the XPSeaks are not evident. The SEI layer formed on the graphite elec-rode is more polymer-like layer with the presence of not stablei-alkoxides whereas the layer on the graphite/ceria electrode isich in Li-carbonates and Li-alkyl carbonates. The composition andhe morphology of the SEI layer as observed from the XPS and SEMesults, respectively, and then discussed thoroughly, depend sig-ificantly on the composition of the electrode surface and haven influence on the electrochemical performance of the electrodeaterial.

. Conclusions

A thin layer of cerium oxides constituted of nanocrystallites ofbout 6 nm was deposited onto graphite by electro-precipitation.his film reflects the morphology of the graphite electrode andxhibits small cracks on the surface but it covers the surface ofhe graphite. XPS analyses show that the cerium oxide layer isomposed of 75% Ce(IV) and 25% Ce(III) oxides.

Reversible insertion and deinsertion of lithium into theraphite through the ceria layer occurs during the first, fourharge–discharge cycles in the electrolyte of LiPF6 in EC/DMC with-ut altering the structure of the material as indicated by XRD. Thensertion/deinsertion of lithium is accompanied by the reduction ofhe electrolyte at 0.7 V vs Li/Li+ leading to the formation of the SEIayer as confirmed by XPS analyses. The galvanostatic tests on theraphite/ceria electrode show that the ceria layer on the graphitelectrode prevents from the loss of capacity during the first, fourycles but the reduction of the electrolyte still occurs despite of theresence of ceria onto graphite.

According to the XPS and SEM results, the SEI layer formedn the graphite/ceria electrode is thicker, more homogenous andas a different composition than the SEI layer formed on the

raphite electrode. The SEI layer formed on the graphite/ceria elec-rode is mainly composed of Li2CO3, ROCO2Li and/or R–CH2OLind LiF. The presence of lithium carbonate was difficult to beevealed on the graphite electrode as no oxygen peak correspond-

[

[[[

xture polyethers and polycarbonates, –CF2 687.0 1.6689.0 1.8

xture (LiPF6 + LiC6+· · ·) 57.4 1.5

ing to –CO3 species was observed. The XP spectra performedon the graphite and graphite/ceria electrodes after two lithiuminsertion–deinsertion show few differences in the composition ofthe SEI layer like the absence of LixPFyOz, ROLi (e.g. LiOCH3) andCF2 compounds in the SEI layer formed on the graphite/ceria whencomparing to the SEI layer on the graphite electrode.

On the basis of the results obtained by different techniques(XRD, SEM and XPS), it can be concluded that the ceria deposited ongraphite negative electrode by electro-precipitation favors the for-mation of a thicker SEI layer which is more homogenous and richerin carbonaceous species than the SEI layer formed on the graphiteelectrode.

Acknowledgements

The authors thank Philippe Biensan, Fabrice Vigier and CecileTessier from SAFT Company for providing the graphite electrodes,Odile Majerus from Chimie Paristech for the useful and fruitfuldiscussions about XRD results and Stephan Borensztajn from Labo-ratoire Interfaces et Systèmes Electrochimiques (Université Pierreet Marie Curie) for the SEM and EDX analyses.

References

[1] E. Peled, J. Electrochem. Soc. 126 (1979) 2047.[2] A. Chagnes, B. Carré, D. Lemordant, P. Willmann, J. Power Sources 109 (2002)

203.[3] I. Geoffroy, A. Chagnes, B. Carré, D. Lemordant, P. Biensan, S. Herreyre, J. Power

Sources 112 (1) (2002) 191.[4] A. Chagnes, B. Carré, P. Willmann, R. Dedryvère, D. Gonbeau, D. Lemordant, J.

Electrochem. Soc. 159 (9) (2003) A1255.[5] A. Chagnes, M. Diaw, B. Carré, P. Willmann, D. Lemordant, J. Power Sources 145

(2005) 82–88.[6] M. Diaw, A. Chagnes, B. Carré, P. Willmann, D. Lemordant, J. Power Sources 146

(2005) 682.[7] H. Buka, P. Golob, M. Winter, J.O. Bensenhard, J. Power Sources 97/98 (2001)

122.[8] Y.P. Yu, C.Y. Jiang, C.R. Wan, E. Tsuchida, Electrochem. Commun. 2 (2000) 272.[9] J. Lee, R. Zhang, Z. Liu, J. Power Sources 90 (2000) 70.10] H. Momose, H. Honbo, S. Takeuchi, K. Nishimura, T. Horiba, Y. Mulranaka, Y.

Kozono, J. Power Sources 90 (2000) 70.11] S. Kim, Y. Kadoma, H. Ikuta, Y. Uchimoto, M. Wakihara, Electrochem. Solid State

Lett. 4 (2001) A109.12] Y.P. Wu, E. Rahm, R. Holze, J. Power Sources 114 (2003) 228.13] S. Kuwabata, N. Tsumura, S. Goda, C.R. Martin, H. Yoneyama, J. Electrochem.

Soc. 145 (1998) 1415.14] H. Wang, M. Yoshio, T. Abe, Z. Ogumi, J. Electrochem. Soc. 149 (2002) A499.15] L.J. Fu, H. Liu, C. Li, Y.P. Wu, E. Rahm, R. Holze, H.Q. Wu, Solid State Sci. 8 (2006)

113.16] T. Takamura, K. Sumiya, J. Suzuki, C. Yamada, K. Sekine, J. Power Sources 81/82

(1999) 368.

17] J.K. Lee, D.H. Ryu, J.B. Ju, Y.G. Shul, B.W. Cho, D. Park, J. Power Sources 107 (2002)

90.18] H. Huang, E.M. Kelder, J. Shoonman, J. Power Sources 97/98 (2001) 114.19] Y. Ein-Eli, V.R. Roch, J. Electrochem. Soc. 144 (1997) 2968.20] Y.P. Wu, C. Jiang, C. Wan, R. Holze, J. Power Sources 111 (2002) 329.

rface S

[[[[

[[[

[[[[[

[[

[

[[

[

[[[

[[[[[

[[[[

[[

[[

[

[

[

[

[

[

[

[[

[

[[

[[[[

[

[

[[

[

[

[

[

J. Swiatowska et al. / Applied Su

21] Y.P. Wu, C. Jiang, C. Wan, R. Holze, J. Appl. Electrochem. 32 (2002) 1011.22] Y.P. Wu, C. Jiang Wan, E. Tsuchida, J. Mater. Chem. 11 (2001) 1233.23] Y.P. Wu, C. Jiang Wan, R. Holze, Electrochem. Commun. 4 (2002) 483.24] Y.P. Wu, R. Holze, in: 53rd International Meeting of Electrochemical Society,

Düsseldorf, Germany, September. 15-20, 2002, p. 278.25] V. Lair, A. Ringuedé, P. Vermaut, S. Griveau, Phys. Stat. Sol. C 5 (2008) 3492.26] H. Elbelghiti, V. Lair, A. Ringuedé, M. Cassir, ECS Trans. 7 (2007) 2261.27] J. Swiatowska-Mrowiecka, V. Maurice, S. Zanna, L. Klein, P. Marcus, Electrochim.

Acta 52 (2007) 5644.28] S. Peulon, D. Lincot, J. Electrochem. Soc. 145 (1998) 864.29] P.L. Chen, I.W. Chen, J. Am. Ceram. Soc. 76 (1993) 1577.30] Y. Zhou, J.A. Switzer, J. Alloys Compd. 237 (1996) 1.31] L. Arurault, B. Daffos, F.X. Sauvage, Mater. Res. Bull. 43 (2008) 796.32] Y. Hamloui, F. Pedraza, C. Remazeilles, S. Cohendoz, C. Rebere, L. Tifouti, J. Creus,

Mater. Chem. Phys. 113 (2009) 650.33] W.F. Ng, M.H. Wong, F.T. Cheng, Mater. Chem. Phys. 119 (2009) 384.34] E.A. Kulp, S.J. Limmer, E.W. Bohannan, J.A. Switzer, Solid State Ionics 178 (2007)

749.35] J. E1 Fallah, L. Hilaire, M. Roméo, F. Le Normand, J. Electron. Spectrosc. Relat.

Phenom. 73 (1995) 89.36] R. Leppelt, B. Schumacher, V. Plzak, M. Kinne, R.J. Behm, J. Catal. 244 (2006) 137.37] P. Burroughs, A. Hamnett, A.F. Orchard, G.J. Thornton, Chem. Soc., Dalton Trans.

17 (1976) 1686.38] G. Praline, B.E. Koel, R.L. Hance, H.I. Lee, J.M. White, J. Electron. Spectrosc. Relat.

Phenom. 21 (1980) 17.39] E. Paparazzo, G.M. Ingo, N.J. Zacchetti, Vac. Sci. Technol. A 9 (1991) 1416.40] A. Pfau, K.D. Schierbaum, Surf. Sci. 312 (1994) 171.41] M. Romeo, K. Bak, J. El Fallah, F. Le Normand, L. Hilaire, Surf. Interface Anal. 20

(1993) 508.42] O. Gunnarsson, K. Schonhammer, Phys. Rev. B 28 (1983) 4315.43] A. Fujimori, Phys. Rev. B 28 (1983) 4489.44] A. Kotani, T. Jo, J.C. Parlebas, Adv. Phys. 37 (1988) 37.45] D.R. Mullins, S.H. Overbury, D.R. Huntley, Surf. Sci. 409 (1998) 307.46] Y.A. Teterin, A.Y. Teterin, A.M. Lebedev, I.O. Utkin, J. Electron. Spectrosc. Relat.

Phenom. 88–91 (1998) 275.47] L. Qiu, F. Liu, L. Zhao, Y. Ma, J. Yao, Appl. Surf. Sci. 252 (2006) 4931.48] F. Larachi, J. Pierre, A. Adnot, A. Bernis, Appl. Surf. Sci. 195 (2002) 236.49] R.J. Chung, M.F. Hsieh, R.N. Panda, T.S. Chin, Surf. Coat. Technol. 165 (2003) 194.50] V. Chauvaut, V. Albin, H. Schneider, M. Cassir, H. Ardelean, A. Galtayries, J. Appl.

Electrochem. 30 (2000) 1405.51] H. Ardelean, I. Frateur, P. Marcus, Corr. Sci. 50 (2008) 1907.

52] E. Beche, P. Charvin, D. Perarnau, S. Abanades, G. Flamant, Surf. Interface Anal.

40 (2008) 264.53] F. Zhang, P. Wang, J. Koberstein, S. Khalid, S.-W. Chan, Surf. Sci. 563 (2004) 74.54] J.-T. Li, J. Swiatowska, A. Seyeux, L. Huang, V. Maurice, S.-G. Sun, J. Power Sources

195 (2010) 8251.

[

[

cience 257 (2011) 9110– 9119 9119

55] J.-T. Li, J. Swiatowska, V. Maurice, A. Seyeux, L. Huang, S.-G. Sun, P. Marcus, J.Phys. Chem. C 115 (2011) 7012.

56] J.-T. Li, V. Maurice, J. Swiatowska-Mrowiecka, A. Seyeux, S. Zanna, L. Klein, S.-G.Sun, P. Marcus, Electrochim. Acta 54 (2009) 3700.

57] S. Leroy, H. Martinez, R. Dedryvere, D. Lemordant, D. Gonbeau, Appl. Surf. Sci.253 (2007) 4895.

58] M. Herstedt, A.M. Andersson, H. Rensmo, H. Siegbahn, K. Edström, Electrochim.Acta 49 (2004) 4939.

59] A. Benayad, H. Martinem, A. Gies, B. Pecquenard, A. Lavasseur, D. Gonbeau, J.Electron. Spectrosc. 150 (2006) 1.

60] G.R. Zhuang, K. Wang, Y. Chen, P.N. Ross Jr., J. Vac. Sci. Technol. A 16 (1998)3041.

61] D. Aurbach, B. Markovsky, A. Shechter, Y. Ein-Eli, H. Cohen, J. Electrochem. Soc.143 (1996) 3809.

62] R. Fong, U. Von Sacken, J.R. Dahn, J. Electrochem. Soc. 137 (1990) 2009.63] D. Aurbach, Y. Ein-Eli, B. Markovsky, A. Zaban, S. Luski, Y. Carmeli, H. Yamin, J.

Electrochem. Soc. 142 (1995) 2882.64] S. Laruelle, S. Pilard, P. Guenot, S. Grugeon, J.-M. Tarascon, J. Electrochem. Soc.

151 (2004) 1202.65] A.M. Andersson, K. Edström, J. Electrochem. Soc. 148 (2001) AA1100.66] E. Peled, D. Bar Tow, A. Merson, A. Gladkich, L. Burstein, D. Golodnitsky, J. Power

Sources 97-98 (2001) 52.67] L. Vogdanis, W. Heitz, Macromol. Chem. Rapid Commun. 7 (1986) 543.68] J.-C. Lee, M. Litt, Macromolecules 33 (2000) 1618.69] D. Aurbach, Nonaqueous Electrochemistry, Marcel Dekker Inc., New York, 1999.70] D. Aurbach, A. Zaban, Y. Gofer, Y. Ein Ely, I. Weissman, O. Chusid, O. Abramson,

J. Power Sources 54 (1995) 76.71] K. Kanamura, S. Shiraishi, H. Tamura, Z.-i. Takehara, J. Electrochem. Soc. 141

(1994) 2379.72] A.M. Andersson, M. Herstedt, A.G. Bishop, K. Edström, Electrochim. Acta 47

(2002) 1885.73] G.R. Zhuang, P.N. Ross Jr., Electrochem. Solid State Lett. 6 (2003) A136.74] M. Stjerndahl, H. Bryngelsson, T. Gustafsson, J.T. Vaughey, M.M. Thackeray, K.

Edström, Electrochim. Acta 52 (2007) 4947.75] M. Herstedt, D.P. Abrahamb, J.B. Kerr, K. Edström, Electrochim. Acta 49 (2004)

5097.76] D. Aurbach, B. Markovsky, A. Rodkin, M. Cojocaru, E. Levi, H.-J. Kim, Electrochim.

Acta 00 (2002) 1.77] G.K. Wertheim, P.M.Th.M.Van Attekum, S. Basu, Solid State Commun. 33 (1980)

1127.78] T. Eriksson, A.M. Andersson, A.G. Bishop, C. Gejke, T. Gustafsson, J.O. Thomas, J.

Electrochem. Soc. 149 (2002) A69.79] D.T. Clark, W.J. Feast, D. Kilcast, W.K.R. Musgrave, J. Polym. Sci. Polym. Chem.

Ed. 11 (1973) 389.80] G. Beamson, D. Briggs, High Resolution XPS of Organic Polymers: The Scienta

ESCA300 Database, 1992.