Electrochimica Acta 117 (2014) 426–430

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

A New Approach to Synthesis of Porous SiOx Anode for Li-ion

Batteries via Chemical Etching of Si Crystallites
Byeong-Chul Yu a , Yoon Hwa a , Jae-Hun Kim b , Hun-Joon Sohn a,∗
a
Department of Materials Science and Engineering, Seoul National University, Seoul, 151-744, Korea
b
School of Advanced Materials Engineering, Kookmin University, Seoul, 136-702, Korea

a r t i c l e i n f o a b s t r a c t

Article history: Unlike previous HF etching of Si oxides to produce porous Si a new concept for the preparation of porous
Received 24 September 2013 SiOx was suggested adopting Si as a pore generating agent and Si oxides as template using NaOH solution.
Received in revised form The heat treatment of pristine SiO at 900 ◦ C provided nano-crystalline Si within the SiOx matrix. The nano-
25 November 2013
crystalline Si was dissolved to provide pore size of 200∼500 nm, while most of the SiOx matrix remained
Accepted 26 November 2013
as a template during NaOH etching. The surface area of the etched SiO was increased more than 5 times
Available online 11 December 2013
compared with not-etched SiO. The porous SiOx anode exhibited a stable reversible capacity of about
1240 mAh g−1 over 100 cycles at 0.2 C.
Keywords:
Porous SiOx © 2013 Elsevier Ltd. All rights reserved.
Self-template
Lithium batteries
Anode
NaOH etching

1. Introduction
There is growing demand for next-generation Li-ion batteries
with high energy capacity and high power for various applications,
such as portable electronic devices, satellites, and electric vehi-
cles [1–4]. The low theoretical capacity of current commercialized
graphite anodes (372 mAh g−1 ) cannot satisfy the demand of high
energy density, and there are interests in developing new anode
materials with higher capacities. Si is an attractive anode material
that is being closely scrutinized for use in Li-ion batteries because
of its experimentally verified specific capacity of 3579 mAh g−1
(Li15 Si4 ) at room temperature [5,6]. However, the development of
Si anodes for Li-ion batteries has been limited, mostly because of
the large volumetric change (of up to ∼300%) that occurs upon the
insertion and extraction of Li ions, and the large electrochemically
induced stress, which results in electrode pulverization, the loss
of electrical contact, and early capacity fading of the battery cell
[7–10].
In recent years, more attention has been paid to Si-based
anode materials such as SiO and SiO2 in order to prevent the
common problem of a large volume expansion. These Si oxides
react with Li, forming Li2 O, Li-silicates, and Li-Si alloy, during
∗ Corresponding author. Tel.: +82 2 880 7226.
E-mail address: hjsohn@snu.ac.kr (H.-J. Sohn).
the first discharge. Although Li2 O and Li silicates phases are
irreversible (except Li2 Si2 O5 ), they act as buffering phases dur-
ing cycling. Due to these unique microstructural combinations,
they show a high capacity with relatively stable cycle performance
[11–21].
Many studies have successfully demonstrated that such a high
capacity could be stably maintained by the fabrication of porous
structure, which can provide sufficient free space to absorb the
large volume expansion [21–26]. Moreover, the open structure of
the porous particles is favorable for the fast transport of Li+ , and
gives rise to a high rate performance. Porous-structured SiO has
rarely been reported, because the preparation of SiO is usually car-
ried out by vapor deposition techniques [27], which has hindered
the application of synthesis for various porous structures.
We report a facile route for preparing porous SiOx . Since SiO
is thermodynamically unstable at all temperature, it dispropor-
tionates into a nano-crystalline Si and amorphous Si oxide during
heat treatment above 800 ◦ C [13,16,28,29]. After heat treatment at
900 ◦ C, crystalline Si and Si oxide can be pore-generating agent and
self-template, respectively, since etching rates of nanosize crys-
talline Si and amorphous Si oxides are different when etched with
NaOH solution [30,31]. This new method is different from previ-
ous HF etching of SiO2 to provide a porous structure of Si [22,25].
The porous SiOx materials obtained were tested as an anode for
Li storage, and showed a significantly improved electrochemical
performance without carbon coating.

0013-4686/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2013.11.158
 B.-C. Yu et al. / Electrochimica Acta 117 (2014) 426–430
Fig. 1. Schematic views of the preparation of porous SiOx .
2. Experimental
2.1. Sample Preparation.
Porous SiOx (P-SiOx ) powders were synthesized using the fol-
lowing procedures. Bulk SiO (Aldrich, -325 mesh) was heat-treated
at 900 ◦ C for 3 h (H-SiO). To reduce the size of the H-SiO pow-
der, high-energy mechanical milling (HEMM) was performed under
Ar atmosphere at 800 rpm for 6 h (mH-SiO). To obtain the micro-
porous structure, the mH-SiO was etched by 1 M NaOH solution for
4 h at room temperature. The P-SiOx powders were washed with
distilled-deionized water several times and dried in a vacuum oven
at 60 ◦ C for 4 h.
2.2. Material Characterization.
All of the samples were examined using X-ray diffraction tech-
niques (Rigaku, D-MAX2500). A field emission scanning electron
microscope (FESEM, Hitachi, SU70) was employed to observe the
surface and microstructure of each sample, and a high resolution
transmission electron microscopy (HRTEM, JEOL 3000F) analyses
were carried out to identify the intermediate and final product
during cycling. Brunauer-Emmett-Teller (BET, ASAP2010) analyses
were performed to measure the surface area using the nitrogen
adsorption-desorption isotherms before and after etching process
by the Barrett–Joyner–Halenda (BJH) method. To confirm the Si/O
ratio of mH-SiOx and P-SiOx samples, X-ray fluorescence (XRF,
Bruker AXS, S4 Poineer) was carried out three times for each
Fig. 2. (a) XRD patterns of SiO, H-SiO, mH-SiO and P-SiOx powders. FESEM images of (b) mH-SiO and (c) P-SiOx . HRTEM and TEM bright field images with SAED patterns of
(d,e) mH-SiO and (f,g) P-SiOx .
427
sample. To prepare ex situ TEM samples, the electrodes were
detached from the coin-type cell and washed with diethyl carbon-
ate (DEC). Dilute suspensions of the samples were dropped onto
a carbon-coated TEM grid. The ex situ HRTEM sample preparation
processes were conducted in an Ar-filled glove box.
2.3. Electrochemical Cell Test.
The electrodes consisted of an active material powder (70 wt.%),
Ketchen black (10 wt.%) as a conducting agent, and poly amide
imide (PAI, 20 wt.%) as a binder. These were dissolved in N-methyl
pyrrolidinone (NMP) and thoroughly mixed using a magnetic stir-
rer to form a slurry at 60 ◦ C, which was coated onto copper foil.
The samples were pressed and dried under a vacuum at 200 ◦ C
for 4 h. A laboratory-made coin-type cell was used with Li foil as
the counter and reference electrodes. The electrolyte contained 1 M
LiPF6 and 10% fluoro ethylene carbonate (FEC) in ethylene carbon-
ate (EC)/diethylene carbornate (DEC) (3:7 vol.%, PANAX). The cells
were tested galvanostatically between 0.0 and 2.0 V (vs. Li/Li + )
using a Maccor automated tester in an Ar-filled glove box. Li was
inserted into the electrode during discharging, and extracted from
the working electrode during charging.
3. Results and discussion
Fig. 1 illustrates a schematic of the experimental process to
synthesize porous SiOx (P-SiOx ). When bulk SiO was heated at
900 ◦ C under Ar atmosphere (H-SiO), disproportionation reaction
428 B.-C. Yu et al. / Electrochimica Acta 117 (2014) 426–430
Fig. 3. N2 adsorption/desorption isotherm of mH-SiO and P-SiOx .
occurred, and nanosize crystalline Si (nc-Si, 10∼20 nm) was dis-
persed into the SiOx matrix [13]. After HEMM, the particle size
dropped to less than 20 ␮m, and more nc-Si was exposed (mH-SiO).
Finally, the obtained mH-SiO was etched in NaOH solution (P-SiOx ).
However, silicon (sub) oxides could be etched very slowly, while nc-
Si was attacked vigorously by NaOH etching [30]. This indicated the
possibility of the self-template process in which the exposed nc-Si
could act as a main pore-generating agent to provide fine pores
during etching, while the template SiOx was dissolved slightly. The
SiO reacted with Li, resulting in the formation of various Li-Silicates
(Li4 SiO4 , Li6 Si2 O7 and Li2 Si2 O5 ) and Li-Si alloy (Li21 Si8 ) [20], which
induced the stress due to the volume expansion. In order to ease
Fig. 4. (a) DCP (inset graph: voltage profile) of P-SiOx at the first cycle. Ex situ HRTEM images with SAED patterns of P-SiOx electrode at various potential indicated in DCP.
(b) Discharge at 0.4 V, (c) discharge at 0.15 V, (d) discharge at 0 V, (e) charge at 0.4 V and (f) charge at 2 V.
the volume change, mH-SiO samples were etched for 4 h to obtain
pore size greater than 100 nm.
Samples from each step were examined by X-ray diffraction
(XRD), as shown in Fig. 2(a). The pristine SiO sample showed an
amorphous nature, and mainly consisted of amorphous Si and SiO2
surrounded by sub-oxide matrix [32,33]. The XRD pattern of H-SiO 歧化反应
powders indicated that crystalline Si was formed by the dispro-
portionation reaction. After the milling process (mH-SiO), Si peaks
became broadened, although there was no change in microstruc-
ture [20]. After etching, a part of the Si within the particles still
remained, as can be seen in the XRD pattern. The surface mor-
phology of the mH-SiO sample was compared before and after the
etching process. Unlike the surface of mH-SiO (Fig. 2(b)), the surface
of P-SiOx showed a well-developed porous structure with pore size
of 200∼500 nm (Fig. 2(c)), in which the exposed nc-Si and part of
the SiO was dissolved, providing voids during etching. To support
the XRD data, the microstructures of samples were investigated
using HRTEM. Figs. 2(d-g) show the TEM bright field and HRTEM
images with SEAD patterns before and after the etching processes.
The powders of mH-SiO were dense without voids in the matrix
(Fig. 2(b) and 2(d)), and the nanocrystalline Si of the same size was
dispersed in the SiOx matrix after the milling process (Fig. 2(e)). In
the case of etched SiOx (P-SiOx , Fig. 2(f)), the porous SiOx matrix
template remained, although some of the nc-Si within the SiOx
matrix was still present, which supported the XRD data shown in
Fig. 2(a). 等温线
The N2 adsorption-desorption isotherms of mH-SiO and P-SiOx
samples are compared in Fig. 3. The large adsorption-desorption
isotherm of P-SiOx was attributed to the pores generated by the dis-
solution of nc-Si. The BET surface area was found to increase from
19.57 m2 g−1 for mH-SiO to 105.41 m2 g−1 for P-SiOx , an increase in
surface area by a factor of more than five. And the average pore size
 B.-C. Yu et al. / Electrochimica Acta 117 (2014) 426–430 429

Table 1 were formed (Fig. 4(b)). At 0.15 V (Fig. 4(c)), Li4 SiO4 phase was
X-ray fluorescence (XRF) data of mH-SiO and P-SiOx .
also observed in addition to the aforementioned two Li silicates.
Si/O ratio 1st 2nd 3rd Average Si/O ratio As the potential was reduced to 0.0 V, crystalline Li21 Si8 phase was
mH-SiO 0.995 0.995 0.995 0.995 detected (Fig. 4(d)). During discharge, the Li21 Si8 phase disappeared
P-SiOx 0.961 0.964 0.964 0.963 at 0.4 V (Fig. 4(e)). At 2.0 V, the Li2 Si2 O5 phase disappeared, while
the Li4 SiO4 and Li6 Si2 O7 phases remained as irreversible phases
(Fig. 4(f)). These ex situ analyses agreed well with the previous
in the P-SiOx samples is approximately 75.2 nm (Barrett-Joyner- report on the reaction mechanism of SiO with Li [20].
Halenda (BJH) analysis). The average Si/O ratios of mH-SiO and The electrochemical performance of four samples used as anode
P-SiOx samples from XRF data were 0.995 and 0.963, respectively, materials for Li storage was tested. Fig. 5a shows a comparison of
which confirmed that the exposed nc-Si (2.04 wt.%) was dissolved the cycle performance of each sample. The rate of all samples was
during the etching process (Table 1). 0.1 C (150 mAh g−1 ) for the first cycle, and 0.2 C for later cycles.
Fig. 4 shows the differential capacity plot (DCP) and voltage pro- The discharge capacities of SiO and H-SiO were 2612 mAh g−1 and
file (inset graph) of P-SiOx . To investigate the mechanism during 1853 mAh g−1 , respectively. After disproportionation reaction at
the first cycle of P-SiOx , ex situ HRTEM was conducted at selected high temperature, the SiO2 ratio increased which hindered the Li-
potentials indicated in DCP as shown in Fig. 4(b-f). When the poten- ion diffusion into the SiOx matrix [13]. Hwa et al. reported that
tial reached 0.4 V during discharge, Li2 Si2 O5 and Li6 Si2 O7 phases disproportionated SiO at 1200 ◦ C was made active to Li-ions after
milling, because the Si oxide layer on the surface was partially bro-
ken, and active Si was exposed on the surface [16]. With similar
behavior, the discharge capacity of mH-SiO increased again to 2653
mAh g−1 . The first Coulombic efficiencies of SiO, H-SiO, mH-SiO,
and P-SiOx were 56.3%, 28.8%, 60.8%, and 64.4%., respectively. The
Coulombic efficiency of P-SiOx increased by 8%. The P-SiOx without
a carbon coating exhibited a reversible capacity of 1242 mAh g−1
over 100 cycles, while cycles of other materials decreased rapidly.
This was attributed to the porous structure of P-SiOx , which could
accommodate the large volume expansion during Li insertion. The
rate capability of mH-SiO and P-SiOx were also tested at various
current rates, as shown in Fig. 5(b). Without a carbon coating, the
discharge capacity at the 1 C rate was 890 mAh g−1 due to the fast
Li-ion diffusion within pores filled with electrolyte.
To confirm the structural integrity of the P-SiOx during cycling,
ex situ HRTEM was conducted after 100 cycles, and the results
are presented in Fig. 5(c) and 5(d). The pores in the SiOx matrix
were well maintained as shown in bright field image (Fig. 5(c)) and
were almost the same with those before cycling as illustrated in
Fig. 2(f). This indicated that the pores present in the P-SiOx were
effective in reducing the volume change, which led to a good cycle
performance. Fig. 3(d) shows the HRTEM image with SAED pattern
after 100 cycles which Li6 Si2 O7 and Li4 SiO4 crystallites were still
remained as irreversible phases.

4. Conclusions

Unlike previous HF etching of silicon oxide to produce porous Si

anode, a new concept for producing porous SiOx material through
facile synthesis has been proposed. The disproportionated SiO sub-
jected to milling process was etched by NaOH solution, in which
crystalline Si acted as the main pore-generating agent, and the SiOx
matrix remained as a template. The porous SiOx was tested as an
anode for Li-ion batteries, and it showed excellent electrochemical
performance without any carbon coating. This was due to the alle-
viation of volume expansion during cycles and facilitated Li-ion
transport through the pores. We anticipate that this approach of
facile porous SiOx preparation via self-template etching technique
can be extended to the synthesis of other porous materials.

References

[1] J.B. Goodenough, Y. Kim, Chem. Mater. 22 (2010) 587–603.

Fig. 5. Electrochemical behaviors of various samples. (a) cycle performance (0.1 C for
first cycle, then 0.2 C for later cycles) of SiO, H-SiO, mH-SiO and P-SiOx powders and
(b) rate capability of mH-SiO and P-SiOx . Ex situ (c) TEM bright field and (d) HRTEM
images with SAED patterns of P-SiOx electrode, respectively, after 100 cycles.
[2] M. Armand, J.M. Tarascon, Nature 451 (2008) 652–657.
[3] C.-M. Park, J.-H. Kim, H. Kim, H.-J. Sohn, Chem. Soc. Rev. 39 (2010) 3115–3141.
[4] G. Jeong, Y.-U. Kim, H. Kim, Y.-J. Kim, H.-J. Sohn, Energy Environ. Sci. 4 (2011)
1986–2002.
[5] A.H. Mayne, S.C. Bayliss, P. Barr, M. Tobin, L.D. Buckberry, Phys. Status Solidi (A)
Appl. Res 182 (2000) 505–513.
[6] U. Kasavajjula, C. Wang, A.J. Appleby, J. Power Sources 163 (2007) 1003–1039.
430 B.-C. Yu et al. / Electrochimica Acta 117 (2014) 426–430

[7] J.R. Szczech, S. Jin, Energy Environ. Sci. 4 (2011) 56–72. [20] B.-C. Yu, Y. Hwa, C.-M. Park, H.-J. Sohn, J. Mater. Chem. A 1 (2013) 4820–4825.
[8] R.A. Huggins, W.D. Nix, Ionics 6 (2000) 57–63. [21] J.-I. Lee, S. Park, Nano Energy 2 (2013) 146–152.
[9] J.H. Ryu, J.W. Kim, Y.E. Sung, S.M. Oh, Electrochem Solid State Lett. 7 (2004) [22] J. Cho, J. Mater. Chem. 20 (2010) 4009–4014.
A306–A309. [23] Y. Yu, L. Gu, C. Zhu, S. Tsukimoto, P.A. van Aken, J. Maier, Adv. Mater. 22 (2010)
[10] B.A. Boukamp, G.C. Lesh, R.A. Huggins, J. Electrochem. Soc. 128 (1981) 725–729. 2247–2250.
[11] J.-H. Kim, H.-J. Sohn, H. Kim, G. Jeong, W. Choi, J. Power Sources 170 (2007) [24] D. Chen, X. Mei, D. Ji, M. Lu, J. Xie, J. Lu, J.Y. Lee, Angew. Chem. Int. Ed. 51 (2012)
456–459. 2409–2413.
[12] A. Veluchamy, C.-H. Doh, D.-H. Kim, J.-H. Lee, D.-J. Lee, K.-H. Ha, H.-M. Shin, [25] Y. Yao, M.T. McDowell, I. Ryu, H. Wu, N. Liu, L. Hu, W.D. Nix, Y. Cui, Nano Lett.
H.-S. Kim, S.-I. Moon, C.-W. Park, J. Power Sources 188 (2009) 574–577. 11 (2011) 2949–2954.
[13] C.-M. Park, W.C. Choi, Y. Hwa, J.-H. Kim, G. Jeong, H.-J. Sohn, J. Mater. Chem. 20 [26] B.M. Bang, J.-I. Lee, H. Kim, J. Cho, S. Park, Adv. Energy Mater 2 (2012) 878–883.
(2010) 4854–4860. [27] P.R. Abel, Y.M. Lin, H. Celio, A. Heller, C.B. Mullins, ACS Nano 6 (2012)
[14] J. Wang, H. Zhao, J. He, C. Wang, J. Wang, J. Power Sources 196 (2011) 2506–2516.
4811–4815. [28] M. Mamiya, M. Kikuchi, H. Takei, J. Cryst. Growth 237 (2002) 1909–1914.
[15] J.-I. Lee, N.-S. Choi, S. Park, Energy Environ. Sci. 5 (2012) 7878–7882. [29] M. Mamiya, H. Takei, M. Kikuchi, C. Uyeda, J. Cryst. Growth 229 (2001) 457–461.
[16] Y. Hwa, C.-M. Park, H.-J. Sohn, J. Power Sources 222 (2013) 129–134. [30] K. Sakaino, Y. Kawabata, S. Adachi, J. Electrochem. Soc. 47 (2000) 1530–1534.
[17] G. Jeong, J.-H. Kim, Y.-U. Kim, Y.-J. Kim, J. Mater. Chem. 22 (2012) 7999–8004. [31] B.-C. Yu, Y. Hwa, C.-M. Park, J.-H. Kim, H.-J. Sohn, RSC Adv 3 (2013) 9408–9413.
[18] F. Dai, R. Yi, M.L. Gordin, S. Chen, D. Wang, RSC Adv. 2 (2012) 12710–12713. [32] A. Hohl, T. Wieder, P.A. Aken, T.E. Weirich, G. Denninger, M. Vidal, S. Oswald, C.
[19] W.-S. Chang, C.-M. Park, J.-H. Kim, Y.-U. Kim, G. Jeong, H.-J. Sohn, Energy Envi- Deneke, J. Mayer, H. Fuess, J. Non-Cryst. Solids 230 (2003) 255–280.
ron. Sci. 5 (2012) 6895–6899. [33] K. Schulmeister, W. Mader, J. Non-Cryst. Solids 320 (2003) 143–150.