Effects of Vacuum Annealing on the Charge–Discharge ... · Effects of Vacuum Annealing on the ChargeDischarge Characteristics of Eutectic AlSi/Al Thin Film as Anode Material for

Effects of Vacuum Annealing on the Charge­Discharge Characteristicsof Eutectic Al­Si/Al Thin Film as Anode Material for Li-Ion Batteries

Chao-Han Wu, Truan-Sheng Lui, Fei-Yi Hung+ and Li-Hui Chen

Department of Materials Science and Engineering, National Cheng Kung University, Tainan, Taiwan 701, R. O. China

In this study, radio frequency magnetron sputtering was used to prepare eutectic Al­Si/Al bi-layered films as anode materials and the effectof vacuum annealing in the charge­discharge capacity characteristics at different temperatures were discussed. For the purpose of 400 nm Al­Sifilm can possess the lowest crystallization temperature, the eutectic composition was adopted. The pre-sputtered 40 nm Al thin film not onlyreduced the resistivity of the composite anode film, but also diffused to prevent peeling between the Al­Si films and Cu foils after vacuumannealing. While the annealing temperatures were elevated (RT ³ 400°C), indexes of crystalline (IOC) and resistivities of specimens werechanged. The properties of materials containing ASEC-400 (at RT) and ASEC-200 (at 55°C) had outstanding charge­discharge characteristics.The morphology transformation at the surface and cross section resulted from annealing at different temperatures and cycling testing wereexamined by Focus Ion Beam (FIB). Besides, the relationship between cycling performances and electrochemical characteristics of Al­Si/Alfilm anodes were also investigated by Cyclic Voltammetry and Electrochemical AC Impedance Spectroscopy (EIS).[doi:10.2320/matertrans.M2012086]

(Received March 6, 2012; Accepted June 13, 2012; Published August 25, 2012)

Keywords: aluminum­silicon, anode material, charge­discharge

1. Introduction

Many alloy systems are being developed to replacegraphite as the anode in lithium rechargeable batteries dueto their better capacity (Sn­Cu1­4) · Li­Sn5) · Cu­Sb6) · Mg­Si7) · Li­Si8)). Si-based intermetallic compounds possessmarvelous capacity (Li4.4Si: 4200mAh/g) and have beeninvestigated continuously. In addition to Si, Al has becomeattractive gradually owing to its excellent capacity (Li2.25Al:2235mAh/g). Comparing with Si, the ratio of Li/Al is 2.25when the maximum storage for Li of Al has been reached. Itmeans that Al suffers less impact resulted from the insertingof Li ions than Si does. But Al still hasn’t possessed wellcycleability yet and low irreversible capacity without anydoping because the anode has to be inserted/extracted by Liions at a single voltage. Doping the second element in Alcan improve the above problems very well. Some Al-basedsystem like Al­C,9) Al­Fe10) and Al­Sn11) has been revealedso far.

On account of the above mentioned reasons, the anodematerial was prepared by sputtering Al-based binary film onthe Cu foil. Si was adopted as the second element becauseof its wonderful capacity. The main lithiation product of Siis the IMC, Li22Si5. Comparing that of Al(LiAl), Si has toundergo more violent volume variation (4.4 : 1 vs. 1 : 1)during main lithiation/delothiation process than Al does.Although Si is more unfavorable as the anode matrix mate-rial than Al is, it’s still a suitable choice for the secondelement in the active-active system like Al­Si. The systemwon’t only suffer too much volume variation at a singlevoltage, but also let both elements contribute to overallcapacity.

According to the Ref. 12), it revealed that Al filled thechinks at the Al/Cu interface after annealing. It means thatenhancing interface joint and the vertical conductivity of the

anode material are reached after annealing. That’s the reasonwhy the 40 nm pre-sputtered Al layer was adopted. Besides,this paper used lower annealing temperatures to elevate IOCof samples because the Al­Si eutectic-composed materialspossess lower crystallization temperature. Also, the relation-ship among IOC, resistivities, cycled morphorlogies andelectrochemical properties were investigated.

2. Experimental Procedures

In this study, Al/eutectic Al­Si dual layered films weresputtered on 10 µm Cu foil. The thicknesses of unary andbinary film were 40 and 400 nm. ASEC-AD (Al­Si(400 nm)/Al(40 nm)/Cu(10 µm)) is defined as the as-deposited sample.Some ASEC-AD films were annealed at 200 and 400°C for1 h in vacuum and are designated ASEC-200 and ASEC-400.Each film was cut for the charge­discharge testing. Thecomposition of the electrolyte was LiPF6 + EC + DEC(EC : DEC = 1 : 1 vol).

The micro-morphology and interface characteristics ofthe pre- and post-cycling samples were investigated by SEMand FIB (focused ion beam). The phases and IOC of theun-annealed and annealed films were analyzed by thin-filmXRD. The angle of incidence was 1°. The velocity ofscanning was 4°/min and the range was from 20 to 100°.A constant current rate, 0.1C, was used for electrochemicaltest with 20 cycles (1st­10th at room temperature, 11th­20that high temperature, 55°C). The voltage was limited to therange 0.01­1.5V. Electrochemical AC Impedance Spec-troscopy (EIS) analysis was conducted at the voltage of0.01V as the end of 10th and 20th lithiation stage. Thepotential amplitude was set to be 10mV and the scanningfrequency was ranged from 100 kHz to 10mHz. CyclicVoltammetry measurements were performed with the scan-ning rate of 0.05mV/s. In addition, the resistivity wasmeasured using a four-point probe and each datum wasaverage of 10 test results.+Corresponding author, E-mail: [email protected]

Materials Transactions, Vol. 53, No. 9 (2012) pp. 1669 to 1673©2012 The Japan Institute of Metals

3. Results and Discussion

Figure 1 shows the surface and cross-section character-istics of ASEC-AD and dense deposition are observed onimages. EDS analysis on the upper binary layer of ASEC-ADis shown in Table 1. In Fig. 2, grazing-incident XRDconfirmed the peaks of Al and Cu foil but the peaks of Sididn’t appear. The result also took place in some formerstudies. Some explanations for it are that Si within the film isnanostructure and its grain size is too tiny to be detected.13)

The others consider that the amorphous structure of Si causesthe situation.14,15) Based on the Ref. 16), the indexes ofcrystalline (IOC) of materials were proportional to integra-tions of XRD diffraction peaks. In Fig. 2 and Table 2, itrevealed that IOC of Al within samples was raised withincreased annealing temperatures. The diffraction peaks of Si

appeared after 400°C vacuum annealing. It means that thepartial Si in ASEC-400 matrix transformed into the crystallinestate from amorphous state.

The resistivities of samples are shown in Fig. 3. IOC ofthin film was inversely proportional to its resistivity.17) Itrevealed that the resistivity of ASEC-AD was higher than thatof others. Largest standard deviation means there’s still lotsof amorphous areas in as-deposited matrix. After 200°C-1 hvacuum annealing, both resistivity and standard deviationchanged obviously. While 400°C-1 h vacuum annealing wascarried out, not only resistivity but also standard deviation

(a)

(b)

Pt

Cu foil

AlSi

Al

Fig. 1 SEM photographs: (a) surface and (b) cross section of ASEC-AD.

Table 1 EDS analysis of Al­Si layer.

Al at% Si at%

ASEC-AD 87.7 12.3

20 40 60 802θ

Inte

nsi

ty

ASEC-AD

ASEC-200

ASEC-400

AlCuSi

Fig. 2 GI-XRD analysis of specimens.

Table 2 Detail data of Al(111) in Fig. 2.

FWHM(°)

Average grainsize (nm)

Integration areaof Al(111)

ASEC-AD 0.38 21.7 108.6

ASEC-200 0.34 24.7 158.3

ASEC-400 0.40 21.6 506.6

0

0.0002

0.0004

0.0006

Res

isti

vity

(Ω

−cm

)

ASEC-AD ASEC-200 ASEC-400

Fig. 3 Resistivities of specimens.

C.-H. Wu, T.-S. Lui, F.-Y. Hung and L.-H. Chen1670

were reduced startlingly. Based on the above reason, it’ssupposed that the area of crystalline shows the overwhelmingmajority in ASEC-400 matrix.

The cycling performances of specimens at differenttemperatures are plotted in Fig. 4. The delithiation capacityand the retention of ASEC-400 specimen were outstanding atRT but there was an evident decay in both of them at 55°C.Different from ASEC-400, ASEC-200 presented very well athigher temperature. Figure 5 shows that the three samples allhad an abrupt decay with various degrees at 11th cycle inCoulombic efficiency. The ultra-low resistivity and enhanceddriving force (a higher temperature is helpful to the lithiumions migration through SEI and the charge transfer) weresupplied by higher temperature to make lithiation reactionsgo beyond the limit of ASEC-400 specimen. AppropriateIOC and resistivity may be the reasons that the ASEC-200specimen had better charge­discharge characteristics at 55°Cbecause they prevent sample from disintegration whichoverload lithiation led to.

The electrochemical behaviors of all samples were studiedby Cyclic Voltammetry as shown in Fig. 6. Based on theRefs. 15, 18), the lithiation/delithiation voltage of Al is0.2V/0.48V and that of Si is 0.05V/0.3V. ASEC-400 filmand ASEC-200 film possessed more intensive redox peaksthan others at different temperatures. It was implied that thetwo samples lithiated and delithiated more actively thanothers at RT and 55°C. The fact is also corresponsive inconcert with their performances on capacity, retention andcoulombic efficiency. In Fig. 7, it shows that both ASEC-ADfilm and ASEC-200 film have more obvious redox peaks at55°C. The main delithiation peak of ASEC-400 film at 55°Cdecayed about 58% comparing with the specimen of RT.It indicated that it’s difficult for ASEC-400 film to delithiateat reactive voltages in this stage. The above fact could besupposed the reason why the capacity of specimen faded.Besides, the fewer potential differences between lithiationand delithiation peaks for all specimens at 55°C than those atRT. So, lower electrode polarization and higher lithium-iondiffusivity appeared in anode material at 55°C.19)

Figure 8 shows the comparison of 10th and 20th cycle’sNyquist plots (ZA vs. ZAA) at 0.01V. ZA and ZAAwithin the plotsindicated the real and imaginary parts of the cell impedance.Nyquist plots contained an obvious semicircle in high

0 4 8 12 16 20Cycles

0

200

400

600

800

Del

ith

iati

on

Cap

acit

y (m

Ah

/g)

RT HT

ASEC-ADASEC-200ASEC-400

Fig. 4 Delithiation capacities as a function of cycle number.

0 4 8 12 16 20Cycles

50

60

70

80

90

100

Co

ulo

mb

Eff

icie

ncy

(%

)

RT HT

ASEC-ADASEC-200

ASEC-400

Fig. 5 Coulomb efficiency as a function of cycle number.

0 0.4 0.8 1.2 1.6Potential (V)

-0.0001

-5E-005

0

5E-005

0.0001

Cu

rren

t (A

)

ASEC-AD

ASEC-200ASEC-400

(a)

0 0.4 0.8 1.2 1.6Potential (V)

-0.0002

0

0.0002

Cu

rren

t (A

)

ASEC-AD

ASEC-200

ASEC-400

(b)

10th cycle at RT

20th cycle at 55°C

Fig. 6 Cyclic voltammograms of specimens: (a) RT, (b) 55°C.

Effects of Vacuum Annealing on the Charge­Discharge Characteristics of Eutectic Al­Si/Al Thin Film 1671

frequency, a vague one in middle frequency and an obliquestraight line in low frequency. The three parts represented theresistance of solid electrolyte interface (SEI), Rsei, the charge-transfer resistance, Rct, and Warburg impedance whichreflected the solid state diffusion of Li-ions into the bulk ofsample. The slope of the fitting line plotted in ZA or ZAA vs.½¹1/2 (the augular frequency) in low frequency was known

as Warburg factor ·. It’s inversely proportional to the halfsquare of diffusion coefficient. Rsei is related to the migrationof lithium-ions within SEI.20) At RT, the Rsei fitted of threesamples were 165³ (ASEC-AD), 437³ (ASEC-200) and680³ (ASEC-AD). It seems that higher IOC would promotethe formation of SEI and lead to obstacle to lithium-ionsmigration in this stage. The measurement of Rct showedthe consequence: ASEC-200 > ASEC-400 > ASEC-AD. It’ssuggested that crystalline Al and amorphous Si are beneficialto lithiation and delithiation. ASEC-400 film possessed thelowest · at RT and ASEC-200 film owned the least Rct at55°C. The · values listed in Table 3, it revealed that theenvironment with higher temperature was helpful to lithium-ions diffusion within electrodes.

The surface morphologies of samples after charge­discharge cycling test are shown in Fig. 9. The change onsurface morphology owing to continuous redox reactions ofASEC-AD film is much less than those of others. Higherresistivity might disturb the migration of electrons and letlithiation-delithiation reaction was only close to surface ofanodes. The ASEC-400 film transformed much more aftercycling test and even appeared a bit powdered. It’s probablyrelated to ASEC-400’s poor electrochemical performance at55°C.

-0.4 0 0.4 0.8 1.2 1.6Potential (V)

-0.00012

-8E-005

-4E-005

0

4E-005

8E-005C

urr

ent

(A)

ASEC-AD

10th cycle20th cycle

(a)

-0.4 0 0.4 0.8 1.2 1.6Potentail (V)

-0.0002

0

0.0002

Cu

rren

t (A

)

ASEC-200

10th cycle20th cycle

(b)

-0.4 0 0.4 0.8 1.2 1.6Potentail (V)

-0.0001

-5E-005

0

5E-005

Cu

rren

t (A

)

ASEC-400

10th cycle20th cycle

(c)

Fig. 7 Comparison of Cyclic voltammograms: (a) ASEC-AD, (b) ASEC-200, (c) ASEC-400.

(a)

0 40 80 120Z' (Ω)

0

40

80

120

-Z"

( Ω)

20th at HT-0.01VASEC-AD

ASEC-200

ASEC-400

(b)

0 400 800 1200 1600Z' (Ω)

0

400

800

1200

1600

-Z"

(Ω)

10th at RT-0.01VASEC-AD

ASEC-200

ASEC-400

Fig. 8 Nyquist plots of specimens at different voltages and temperatures(a) RT (b) HT.

C.-H. Wu, T.-S. Lui, F.-Y. Hung and L.-H. Chen1672

4. Conclusion

IOC of eutectic Al­Si/Al dual layered film increased andtheir resistivities decreased obviously without ultra-highannealing temperature. The ASEC-400 film had excellentperformance in cycling test at room temperature but didn’toperate well at higher temperature due to gradual destructionon the microstructure. The ASEC-200 film had higherresistivity and limited the redox reactions at RT but hadhuge volumetric variation at 55°C. Both enhanced drivingforce and appropriate resistivity had contributed to improveand stable the performance of eutectic Al­Si/Al thin filmanode materials.

Acknowledgements

The authors are grateful to National Cheng KungUniversity, the Center for Micro/Nano Science and Tech-nology (NCKU Project of Promoting Academic Excellence& Developing World Class Research Center: D101-2700)and the Chinese National Science Council for its financialsupport (NSC 100-2221-E-006-094; NSC 100-2622-E-006-030-CC3).

REFERENCES

1) W. Pu, X. He, J. Ren, C. Wan and C. Jiang: Electrochim. Acta 50(2005) 4140­4145.

2) N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitani, M. Kamino and I.Yonezu: J. Power Sources 107 (2002) 48­55.

3) K. D. Kepler, J. T. Vaughey and M. M. Thackeray: J. Power Sources81­82 (1999) 383­387.

4) G. X. Wang, L. Sun, D. H. Bradhurst, S. X. Dou and H. K. Liu:J. Alloy. Compd. 299 (2000) L12­L15.

5) J. Chouvin, J. Olivier-Fourcade, J. C. Jumas, B. Simon and O.Godiveau: Chem. Phys. Lett. 308 (1999) 413­420.

6) L. M. L. Fransson, J. T. Vaughey, R. Benedek, K. Edstrom, J. O.Thomas and M. M. Thackeray: Electrochem. Commun. 3 (2001) 317­323.

7) H. Sakaguchi, H. Honda, Y. Akasaka and T. Esaka: J. Power Sources119­121 (2003) 50­55.

8) M. Inaba, T. Uno and A. Tasaka: J. Power Sources 146 (2005) 473­477.

9) Z. Chen, J. Qian, X. Ai, Y. Cao and H. Yang: Electrochim. Acta 54(2009) 4118­4122.

10) M. J. Lindsay, G. X. Wang and H. K. Liu: J. Power Sources 119­121(2003) 84­87.

11) R. Z. Hu, L. Zhang, X. Liu, M. Q. Zeng and M. Zhu: Electrochem.Commun. 10 (2008) 1109­1112.

12) S. Chen, F. Ke, M. Zhou and Y. Bai: Acta Mater. 55 (2007) 3169­3175.13) H. Guo, H. Zhao, C. Yin and W. Qiu: Mater. Sci. Eng. B 131 (2006)

173­176.14) H. Ahn, Y. Kim, W. Kim, Y. Sung and T. Seong: J. Power Sources 163

(2006) 211­214.15) L. B. Chen, J. Y. Xie, H. C. Yu and T. H. Wang: Electrochim. Acta 53

(2008) 8149­8153.16) C. Y. Yang, B. C. Wang and J. D. Wu: J. Mater. Sci. Mater. Med. 6

(1995) 249­257.17) C. H. Wu, F. Y. Hung, T. S. Lui and L. H. Chen: Mater. Trans. 50

(2009) 381­387.18) M. J. Lindsay, G. X. Wang and H. K. Liu: J. Power Sources 119­121

(2003) 84­87.19) Z. Wang, G. Chen, J. Xu, Z. S. Lv and W. Q. Yang: J. Phys. Chem.

Solids 72 (2011) 773­778.20) H. Usui, Y. Yamamoto, K. Yoshiyama, T. Itoh and H. Sakaguchi:

J. Power Sources 196 (2011) 3911­3915.

ASEC-AD

ASEC-400

ASEC-200

Fig. 9 Surface observations of specimens after cycling testing.

Table 3 Fitted data from Nyquist plots in Fig. 8.

T Rsei (³) Rct (³) · (³Hz1/2)

ASEC-ADRT 165 1200 250

HT 42 500 55

ASEC-200RT 437 600 254

HT 19 8 46

ASEC-400RT 680 830 151

HT 30 37 41

Effects of Vacuum Annealing on the Charge­Discharge Characteristics of Eutectic Al­Si/Al Thin Film 1673