Preparation and characterization of three-dimensional tin thin-film anode with good cycle performance

Article abstract of DOI:10.1016/j.electacta.2010.01.065

Three-dimensional tin thin-film anode was prepared by electroless plating tin onto three-dimensional (3D) copper foam (which served as current collector), and characterized physically by SEM, EDS and XRD. Its electrochemical property and mechanism were studied by charge-discharge test, cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The SEM and EDS results indicated that tin film with 500 nm thickness was formed over the whole surface of copper branches. The XRD results suggested that a new phase of Cu₆Sn₅ was formed between copper and tin. Besides the tin microflake structure of 500 nm thickness, the interaction effects of the copper foam and Cu₆Sn₅ phase formed between copper and tin resulted in good cycle performance with first discharge capacity of 737 mAh g^-1, 97% capacity retention after 20 cycles and still 84% after 40 cycles.

Full text of DOI:10.1016/j.electacta.2010.01.065

Electrochimica Acta 55 (2010) 3537–3541
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Preparation and characterization of three-dimensional tin thin-film anode with
good cycle performance
Zhijia Du, Shichao Zhang∗, Tao Jiang, Zhiming Bai
School of Materials Science and Engineering, Beihang University, Xueyuan Road, Haidian District, Beijing 100191, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Three-dimensional tin thin-film anode was prepared by electroless plating tin onto three-dimensional
(3D) copper foam (which served as current collector), and characterized physically by SEM, EDS and XRD.
Its electrochemical property and mechanism were studied by charge–discharge test, cyclic voltammetry
(CV) and electrochemical impedance spectroscopy (EIS). The SEM and EDS results indicated that tin film
with 500 nm thickness was formed over the whole surface of copper branches. The XRD results suggested
that a new phase of Cu6Sn5 was formed between copper and tin. Besides the tin microflake structure of
500 nm thickness, the interaction effects of the copper foam and Cu6Sn5 phase formed between copper
and tin resulted in good cycle performance with first discharge capacity of 737 mAh g−1, 97% capacity
retention after 20 cycles and still 84% after 40 cycles.
Received 17 November 2009
Received in revised form 14 January 2010
Accepted 17 January 2010
Available online 25 January 2010
Keywords:
Lithium-ion battery
Tin anode
Thin film
Three-dimension
Electroless plating
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
good combination between the active materials and the substrate
for preventing the electrode failure [14].
As
a
successful application in lithium-ion batteries, car-
Simultaneously, in order to control the severe volume change,
some researchers dispersed tin into inactive components to form
the active/inactive system and strengthen the interfacial binding
force, such as Sn–Cu, Sn–Ni, Sn–Co, Sn–Mo, Sn–C, etc. [8,15–19].
In Tamura’s study, the capacity retention after 10 cycles was
improved from 20% to 94% due to the formation of tin–copper
intermetallic layers between copper and tin layers by heat treat-
ment which enhanced the interface strength [20]. Although two- or
multi-component alloys improved the battery’s cyclability to some
extent, they are still far from commercialization.
In the present work, copper foam was used as current collector.
The electrode was prepared by electroless plating tin onto copper
foam. Electroless tin plating is a replacement reaction. Eventually,
the good electrochemical behavior of the electrode and the proba-
ble mechanism are discussed.
bonaceous materials have many advantages such as stable
charge–discharge properties and low-cost [1]. However, since
their theoretical capacity is only 372 mAh g−1, it is important
to develop new anode materials with higher capacity. Sn and
Li can form Li4.4Sn in lithiation process with theoretical capac-
ity of 994 mAh g−1, which has attracted researchers’ interest [2].
Previously, people carried out many methods to prepare tin
such as rolling tin pieces to foil, wire electric explosion, con-
stant current electrodeposition, pulse electrodeposition, spray
pyrolysis, etc. [3–8]. In these reports, however, severe volume
expansion occurred due to lithiation/delithiation process dur-
ing charge–discharge cycles. Tin gradually cracked, pulverized
and shed from the current collector, which caused poor cycle
performance and inhibited its application in lithium-ion batter-
ies.
In order to enhance the cyclability of active materials, 3D sub-
strates were used as current collectors, such as carbon paper,
nickel foam, copper foam, and copper cellular architecture [9–13].
Compared to compact metal, the foamy metal possesses several
advantages. Firstly, it offers a good conductive environment for
the active materials. Moreover, its stress absorbability is benefi-
cial to construct a stretchy electrode system which would insure
2. Experimental
2.1. Preparation and characterization of the 3D tin thin-film
anode
The copper foam was fabricated by electrodepositing copper
onto the PU (polyurethane) foam template. And then the PU tem-
plate was burnt out. In this process, copper foam became brittle
and was partially oxidized. Subsequently, the resulting copper
foam was heat treated at 700 ◦C for 1 h under hydrogen/nitrogen
(vol.% = 1:3) mixed gas protection.
∗
Corresponding author. Tel.: +86 10 82338148; fax: +86 01 82339319.
E-mail address: csc@buaa.edu.cn (S. Zhang).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.01.065

3538
Z. Du et al. / Electrochimica Acta 55 (2010) 3537–3541
The 3D tin thin-film anodes (named 3D-TTA) were prepared by
3. Results and discussion
electroless plating tin onto copper foam at 45 ◦C for 45 min. The
plating bath contained 0.1 M SnSO4, 1 M thiourea, 0.5 M sodium
hypophosphite and 0.85 M concentrated sulfuric acid. For compar-
ison, 2D tin thin-film anodes (named 2D-TTA) were also prepared
by electroless plating tin onto flat copper foil under the same condi-
tions. All of the electrodes were dried at 60 ◦C for 60 min in vacuum.
The loading mass of tin on copper foam and flat copper foil was 2.9
and 1.1 mg cm−2 respectively.
The morphologies, composition and phase structure of elec-
trodes were investigated by field emission scanning electron
microscopy (FE-SEM, Hitachi S-4800 SEM system), Hitachi S-530
SEM and MAC Science X-ray diffraction (Cu K␣ radiation), respec-
tively.
The copper foam is composed of many hollow branches with
rough surface (seen in Fig. 1a) with the sectional image of branch
shown in inset of Fig. 1a. Fig. 1b and c shows the cross-section
morphology of 3D-TTA prepared by electroless plating tin onto the
copper foam. The element composition of 3D-TTA is investigated
by EDS as shown in Fig. 1d, which demonstrates the presence of Sn
and Cu without other element. The sectional SEM image in Fig. 1b
and the spectrograms of points A and B in Fig. 1d suggest that there
are deposits overlaid both inside and outside copper branches: tin
is deposited everywhere wetted by solution due to the electroless
plating. Fig. 1c shows the tin film outside is around 500 nm thick
with a dense microstructure.
The charge–discharge curves of 3D-TTA shown in Fig. 2a sug-
gest a poor efficiency during the first cycle with 737 mAh g−1 initial
discharge capacity and 72% columbic efficiency, which is assigned
to the compact microstructure of tin film which severely hinders
the lithiation/delithiation process. Besides, the irreversible capacity
above 0.8 V is attributed to the decomposition of the electrolyte. In
the following cycles, charge–discharge curves have the same pro-
file with good cyclability and efficiency. Fig. 2b shows CV of 3D-TTA
at 0.2 mV s−1. A small shoulder at ∼0.63 V followed by a sharp peak
at ∼0.19 V is observed during Li intercalation while a small hump at
∼0.40 V accompanied by a huge peak divided into two anodic peaks
at ∼0.75 and ∼0.87 V is observed during Li de-intercalation. The 1st
and 2nd cycles have the same CV profile except an irreversible peak
at 1.13 V assigned to the decomposition of the electrolyte and SEI
formation [21].
2.2. Cell assemble and electrochemical measurements
Electrochemical charge–discharge behaviors were investigated
in simulant cells assembled with the as-prepared anode, lithium
foil and Celgard 2300 membrane in an Ar-filled glove box. 1 M
LiPF6/EC-DEC (1:1 by vol.) was used as the electrolyte. All cells were
galvanostatically charged and discharged in a battery test system
(NEWARE BTS-610, Neware Technology Co., Ltd., China) for a cut-off
potential of 0.02–1.5 V (vs. Li/Li+) at ambient temperature.
CV test was performed between 0 and 2.0 V (vs. Li/Li+) at
0.2 mV s−1. After the electrodes were charged up to 1.5 V (vs. Li/Li+),
then left on open-circuit for 2 h to obtain equilibrium, EIS mea-
surements were carried out over frequency range from 60 kHz to
10 mHz with ac amplitude of 5 mV. CV and EIS were performed
using an electrochemical measurement unit (PARStat 2273).
Fig. 2c shows the cycling behavior of anodes with current
density of 0.75 mA mg−1 (=0.75C rate), indicating that 3D-TTA pos-
sesses good capacity and efficiency cycling performance while
Fig. 1. (a): SEM images of copper foam; (b) and (c): cross-section view of electroless plated 3D-TTA; (d): EDS spectrograms of 3D-TTA.

Z. Du et al. / Electrochimica Acta 55 (2010) 3537–3541
3539
Fig. 2. (a): Galvanostatic charge–discharge curves of 3D-TTA at 0.75C charge–discharge rate; (b): cyclic voltammetry of 3D-TTA at 0.2 mV s−1; (c): capacity and efficiency
cycling performance of as-deposited anodes at 0.75C charge–discharge rate; (d): capacity and efficiency cycling performance of 3D-TTA at different charge–discharge rates.
at each rate are 812, 688, 628, 523, 290 mAh g−1, respectively. Com-
pared to the first cycle at each rate, reversible capacity retention
after 10 cycles reaches 101%, 103%, 100%, 89%, 96%, respectively.
The average columbic efficiency is above 95% at 2C and higher rate.
The SEM micrographs of 3D-TTA and 2D-TTA after 40 gal-
vanostatical charge–discharge cycles were shown in Fig. 3a and
b respectively. On the rough surface of the copper branches, tin
film cracks gradually and forms microflake structure after 40 cycles
2D-TTA shows poor cyclability and efficiency after the first 10
cycles. The reversible capacity retention of 3D-TTA reaches 97%
after 20 cycles, 91% after 30 cycles and still 84% after 40 cycles
which is higher than 2D-TTA and other batteries with as-prepared
tin anode ever reported. The average columbic efficiency of 3D-
TTA is above 95% in the second and following cycles. Fig. 2d shows
the high rate capability of 3D-TTA. The charge–discharge rate was
increased stepwise up to 8C. The reversible capacity in the first cycle
Fig. 3. SEM images of the electrodes after 40 cycles at 0.75C charge–discharge rate. (a1) and (a2): top and cross-section views of 3D-TTA; (b): top view of 2D-TTA.

3540
Z. Du et al. / Electrochimica Acta 55 (2010) 3537–3541
Fig. 4. XRD patterns of 3D-TTA. A: as-prepared; B: after 10 cycles.
(shown in Fig. 3a1). Fig. 3a1 also shows that repeated lithia-
tion/delithiation process causes some pulverization but nearly no
shedding. The cross-section morphology in Fig. 3a2 shows that
both the inside and outside tin films are retained in 3D-TTA after
cyclic volume changes. However, the tin film on flat copper foil
emerged severe pulverization and shed from the substrate under
cyclic charge–discharge as shown in Fig. 3b.
film deposited over the whole surface of copper branches, which
provide more active sites and short transfer distance for rapid lithi-
ation/delithiation. Besides, the 3D conductive and stretchy network
structure would also be favorable to the good cyclability and high
rate capability [13].
To further understand the reason for the good cyclability of
3D-TTA, EIS measurements were conducted after charging the elec-
trodes up to 1.5 V (vs. Li/Li+), a namely 100% delithiation state.
The EIS graphs after 1st, 10th, 20th, 30th and 40th cycles are
presented in Fig. 5. The Nyquist plots all comprise two semicir-
cles in high-frequency (HF) and medium-frequency (MF) ranges
and a line inclined at an approximate 45◦ angle to the real axis
in low frequency. According to the previous studies [22,23], the
HF semicircle is ascribed to SEI film, the MF semicircle to inter-
facial charge-transfer and the linear part to Li+ diffusion process
within electrode. As discussed above, CV result suggests that SEI
film formed only in the 1st charge process and retained in the
XRD pattern of 3D-TTA is shown in Fig. 4, which indicates the
presence of two basic phases: the main deposited product ␤-Sn
and Cu substrate. The peak at 30.2◦ is assigned to (1 0 1) plane of
Cu6Sn5 phase with minor quantity. In electroless plating solution,
the potential of [Cu(NH2CSNH2)4]2+/Cu redox pair is more nega-
tive than Sn2+/Sn redox pair. This leads to copper dissolving while
tin is plating on current collector and finally a layer of Cu6Sn5 is
formed along the interface. The formation of Cu6Sn5 layer enhances
the bonding force between active materials and current collector,
which restrains tin’s shedding from the current collector and leads
to good cycling performance. Accordingly, the cyclability of both
3D-TTA and 2D-TTA in the first 10 cycles is considerably good.
The generation of Cu6Sn5 was somewhat different from Tamura’s
previous report in which the intermetallic compound layers were
formed by heat treatment [20]. In Tamura’s article the cyclability
was good in the first 20 cycles, but deteriorated in the following
cycles. The XRD pattern of 3D-TTA after 10 cycles shown in Fig. 4B
indicates that the Cu6Sn5 phase was inherited probably after cycles.
The thickness of tin film is merely around 500 nm which
also reduces volume expansion and improve adhesion strength
between active material and substrate. Moreover, in 3D-TTA the
uniform tin microflake structure (seen in Fig. 3a1) of 500 nm thick-
ness formed on the rough surface of copper branches provides
appropriate space for tin to expand in either upper or side directions
so that it could strongly bind to the substrate under about 360%
volume expansion. However, in 2D-TTA the tin film did not form
microflake structure but pulverized severely. Huge stress–strain
caused by volume change accumulated along the flat copper foil
and led ultimately to the electrode failure while the stress–strain
released on the rough surface of copper branches and resulted in
the well-behaved microflake structure. High rate cycling capabil-
ity of 3D-TTA is attributed to microflake structure and the thin tin
Fig. 5. Electrochemical impedance spectroscopies of 3D-TTA after 1st, 10th, 20th,
30th and 40th cycles.

Z. Du et al. / Electrochimica Acta 55 (2010) 3537–3541
3541
following cycles, so the HF resistance has little change. The MF
resistance keeps almost similar before the 20th cycle. It suggests
the structure of 3D-TTA is highly stable, which corresponds to the
good cyclability that the capacity retention reaches 97% after 20
cycles. However, the MF resistance changes slightly in the 30th
and maintains similar in the 40th cycle. It probably associates with
tin film’s cracking into microflakes and some pulverization due to
the cyclic expansion of crystal structure, which lead to increase of
contact resistance between active materials and hinder interfacial
charge-transfer. This can also be confirmed from the cycling per-
formance in Fig. 2a, in which the capacity fades after the 30th cycle
due to the microstructure changes of the electrode.
Program of China (2007CB936502) and National 863 Program
(2006AA03Z230, 2008AA03Z208).
References
[1] M. Winter, R.J. Brodd, Chem. Rev. 104 (2004) 4245.
[2] M. Winter, J.O. Besenhard, Electrochim. Acta 45 (1999) 31.
[3] S. Yang, P.Y. Zavalij, M.S. Whittingham, Electrochem. Commun.
587.
5 (2003)
[4] W. Choi, J.Y. Lee, B.H. Jung, H.S. Lim, J. Power Sources 136 (2004) 154.
[5] M. Inaba, T. Uno, A. Tasaka, J. Power Sources 146 (2005) 473.
[6] H. Morimoto, S. Tobishima, H. Negishi, J. Power Sources 146 (2005)
469.
[7] L. Balan, R. Schneider, J. Ghanbaja, P. Willmann, D. Billaud, Electrochim. Acta
51 (2006) 3385.
[8] S.H. Ju, H.C. Jang, Y.C. Kang, J. Power Sources 189 (2009) 163.
[9] C. Arbizzani, S. Beninati, M. Lazzari, M. Mastragostino, J. Power Sources 141
(2005) 149.
4. Conclusions
[10] M. Yoshio, T. Tsumura, N. Dimov, J. Power Sources 146 (2005) 10.
[11] J.Y. Xiang, J.P. Tu, X.L. Wang, X.H. Huang, Y.F. Yuan, X.H. Xia, Z.Y. Zeng, J. Power
Sources 185 (2008) 519.
[12] L. Huang, H.B. Wei, F.S. Ke, X.Y. Fan, J.T. Li, S.G. Sun, Electrochim. Acta 54 (2009)
2693.
[13] T. Jiang, S. Zhang, X. Qiu, W. Zhu, L. Chen, Electrochem. Commun. 9 (2007) 930.
[14] G.J. Davies, S. Zhen, J. Mater. Sci. 18 (1983) 1899.
[15] W. Pu, X. He, J. Ren, C. Wan, C. Jiang, Electrochim. Acta 50 (2005)
4140.
[16] H. Guo, S. Zhao, H. Zhao, Y. Chen, Electrochim. Acta 54 (2009) 4040.
[17] P.A. Connor, J.T.S. Irvine, Electrochim. Acta 47 (2002) 2885.
[18] M. Martos, J. Morales, L. Sánchez, Electrochim. Acta 46 (2000) 83.
[19] G.X. Wang, J. Yao, H.K. Liu, S.X. Dou, J. Ahn, Electrochim. Acta 50 (2004)
517.
[20] N. Tamura, R. Ohshita, M. Fujimoto, S. Fujitani, M. Kamino, I. Yonezu, J. Power
Sources 107 (2002) 48.
[21] N. Tamura, R. Ohshita, M. Fujimoto, M. Kamino, S. Fujitani, J. Electrochem. Soc.
150 (2003) A679.
[22] S. Yang, H. Song, X. Chen, Electrochem. Commun. 8 (2006) 137.
[23] S. Yang, H. Song, X. Chen, J. Power Sources 173 (2007) 487.
A 3D tin thin-film anode was successfully prepared by electro-
less plating and applied in lithium-ion batteries. Tin distributed
both inside and outside the copper hollow branches and the thick-
ness of tin film was around 500 nm. X-ray diffraction demonstrated
that Cu6Sn5 phase was formed between tin and copper substrate,
which strengthened the bonding force between active materials
and substrate. Furthermore, the well-behaved microflake structure
also restrained tin’s shedding from the current collector. Although
3D-TTA displayed only 72% initial columbic efficiency, it showed
good cycling performance with capacity retention 97% after 20
cycles and 84% after 40 cycles, which would be potential electrode
in commercial lithium-ion batteries.
Acknowledgements
This work was supported by the National Natural Science Foun-
dation of China (50574008, 50954005), National Basic Research