Electrochimica Acta 168 (2015) 403–413
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrochemical lithiation of thin silicon based layers potentiostatically
deposited from ionic liquid
Codruta Aurelia Vlaic a , Svetlozar Ivanov a, * , Ralf Peipmann a , Anja Eisenhardt b ,
Marcel Himmerlich b , Stefan Krischok b , Andreas Bund a
a
b
Electrochemistry and Electroplating Group, Technische Universität Ilmenau, Gustav-Kirchhoff-Str. 6, 98693 Ilmenau, Germany
Institute of Physics and Institute of Micro- and Nanotechnologies, Technische Universität Ilmenau, PF 100565, 98684 Ilmenau, Germany
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 3 December 2014
Received in revised form 29 March 2015
Accepted 31 March 2015
Available online 3 April 2015
Thin silicon layers containing about 20% carbon and 20% oxygen were deposited on copper substrates by
potentiostatic electroreduction from a 1 M SiCl4 1-butyl-1-methyl-pyrrolidinium bis (trifluoromethyl)
sulfonylimide [BMP][TFSI] electrolyte. The electrodeposition process was investigated by means of
voltammetric techniques, coupled with in-situ microgravimetry (quartz crystal microbalance, QCM). The
electrochemical and QCM data suggest a possible contribution of a partial Si4+ to Si2+ reduction and/or a
restructuring of the metallic substrate. Considerable impact of side reactions parallel to the deposition
process was indicated by QCM measurements performed under potentiostatic and potentiodynamic
conditions. The deposition of silicon-based films was confirmed by energy dispersive X-ray analysis
(EDX). Analysis of the chemical composition of the deposit and its elemental distribution were achieved
by depth profiling X-ray photoelectron spectroscopy (XPS). The electrodeposited silicon containing layers
showed stable lithiation and delithiation with capacity values of about 1200 mAhg1 and 80% capacity
retention after 300 cycles in standard EC/DMC electrolytes. In ionic liquid (IL) the material displayed
lower capacity of ca. 500 mAhg1, which can be attributed to the higher viscosity of this electrolyte and
deposition of IL decomposition products during lithiation.
ã 2015 Elsevier Ltd. All rights reserved.
Keywords:
silicon electrodeposition
ionic liquid
Li ion battery
quartz crystal microbalance
X-ray photoelectron spectroscopy
1. Introduction
There is a great expectation worldwide that lithium ion
batteries (LIBs) will power the hybrid and electric vehicles of
the future. Accordingly, commercially available LIBs require
reduction in cost, enhanced safety and especially, much higher
energy and power densities [1,2]. Consequently, one major aspect
of LIB research is focused on finding novel electrode materials with
enhanced energy and power densities.
The development of thin-film and microbatteries based on Li ion
technology established a further main direction for the progress of
the electrochemical energy storage. Microbatteries are necessary for
a wide range of applications, comprising small medical devices,
sensors, microelectronics and micro-electromechanical-systems
(MEMS) [3–8]. The research work in these fields is motivated by
the possibility to miniaturize electrochemical power sources with
specific geometries, which can be integrated into electronic and
* Corresponding author. Tel.: +49 3677 69 2842; fax: +49 3677 69 3104.
E-mail address: svetlozar-dimitrov.ivanov@tu-ilmenau.de (S. Ivanov).
http://dx.doi.org/10.1016/j.electacta.2015.03.216
0013-4686/ ã 2015 Elsevier Ltd. All rights reserved.
medical devices for special application [3,4]. Due to their narrow
geometry, a space and shape optimization of the battery is required.
Therefore, deposition of stable and thin layers is a constructive
approach for attaining simultaneously the required geometry and
high energy density. Thin film all-solid (micro) battery technology is
usually based on a consecutive deposition of micrometer sized
layers of a cathode material, a solid electrolyte (e.g. lithium–
phosphorous oxynitride-LIPON) and an anode material (e.g. lithium
or silicon tin oxynitride anode). The active materials and the solid
electrolytes are usually deposited by means of magnetron sputtering or other vacuum techniques. Even though during the recent
years there is a significant progress in this field, still the major
disadvantages of all-solid-state thin film batteries are the low
rate capabilities and the sluggish Li–ion transport in the solid
electrolyte [3].
One useful alternative approach for boosting the performance
of microbattery systems is a novel 3D microbattery technology
[3,5–8]. Beside their specific geometry, 3D microbatteries have
further advantages related to electrolyte development. Instead of a
solid electrolyte, a hybrid polymer electrolyte (HPE) consisting of a
polymer soaked with a conventional carbonate based Li ion
404
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
electrolyte can be used [3]. Thus, the ionic conductivity and ion
mobility of the electrolyte and therefore the battery performance
can be substantially enhanced.
Along with the conventional carbonate electrolytes incorporated in polymers, ionic liquids have their special application in the
HPE development as well [9]. Advantages are their electrochemical
stability, low vapor pressure and non-flammability [10], offering
additional safety and reliability of the battery. Based on the abovementioned breakthrough in the field of microbattery technology, it
can be concluded that fundamental studies on cyclability of thin
film materials in conventional and ionic liquid based electrolytes
are necessary. Before their eventual implementation into microbatteries with specific geometry and architecture basic characterization of the novel materials at standard electrochemical
conditions, where the maximum of their performance can be
extracted is essential. The obtained results can be further used as a
benchmark for (micro) battery development at technological and
engineering levels.
Silicon, an extensively studied material suitable for thin film
deposition has a theoretical capacity of 4200 mAhg1, more than
ten times higher compared to the capacity of the currently used
graphite [11]. Its main disadvantage is the huge volume expansion
(>300%) upon alloying with lithium, inducing disintegration of the
electrode material, loss of electrical contact, poor reversibility and
a rapid capacity fade [12]. Different approaches including nanostructuring [12,13], use of silicon in amorphous form [14–16] as
well as forming composites with carbon [11] or other materials
[11,17] resulted in improved performance of the electrode. The
nanostructured silicon anodes accommodate mechanical stress
more efficiently, display diminished crack formation and decreased structural instability. Furthermore, nanodimensional
materials, organized in the form of thin films or dispersed
nanopowders, provide shorter diffusion lengths and additional
surface storage sites for lithium [11].
Electrodeposition is one promising alternative to vacuum
physical deposition techniques, due to its technical simplicity,
the chance for low cost processing and its attractiveness for further
technological scale-up [18]. Similarly to physical vapor deposition,
the Si electrodeposition process yields a thin layer [19], preferable
for thin film and microbattery applications. Furthermore, the
electrodeposition is a versatile technique, allowing control of the
morphology by adjusting the experimental conditions and
obtaining deposits on different substrate geometries. Concerning
this advantage electrodeposition of Si as a stage for 3D microbattery technology was recently achieved from propylene carbonate based electrolyte [5]. Additionally, silicon electrodeposition is
potentially important not only for LIB applications but also for the
deposition of materials in photovoltaic cells [20], electrical,
electronic [21], and optical devices [22] as well as for corrosionresistant coatings [23]. Due to the negative standard electrode
potential of Si (approx. 1.7 V vs. NHE [24]) and the high reactivity
of its precursors (e.g. halides) only some non-aqueous electrolytes
can be used to electrodeposit Si. These include a variety of organic
electrolytes [5,18,20,25–35] and inorganic molten salts [36]. The
room temperature electrolytes are based on propylene carbonate
[5,18,28,29], acetonitrile [30], dichloromethane [25] and ionic
liquids [19,22,31–35]. The electrolytes containing volatile solvents
display certain drawbacks, including high vapor pressure, flammability and need a supporting electrolyte, which is often
incorporated into the deposited film [27]. Molten salts require
high temperatures and are extremely reactive. It is worth noting
that ionic liquids combine the advantages of having a large
potential window, do not require supporting electrolyte and can be
be handled safely (low volatility and flammability). Furthermore,
in the recent years ionic liquids have been tested as electrolytes in
LIBs [16,37–39]. However, the high viscosity, considerable price
and in many cases insufficient cathodic passivation, still limit their
application as commercial electrolytes for electrodeposition and
for implementation in LIBs.
The main goal of this work is to explore the electrodeposition of
thin Si layers on copper substrates from SiCl4 containing [BMP]
[TFSI] with the central intention to shed more light on the
deposited layer composition and structure. This is achieved by the
correlation of experimental data obtained through electrochemical, microgravimetric and spectroscopic methods of investigation.
The interest in the performance of the deposited Si layers as Li
alloying materials is related to their possible direct implementation in thin film and 3D microbatteries. It was correspondingly
shown that Si thin films deposited on TiO2 nanotube template
arrays via magnetron sputtering [39] and plasma enhanced
physical vapor deposition (PECVD) [40] have a good perspective
to be implemented as advanced anodes for microbatteries. This
paper reports on electrochemical deposition of Si containing layer
as a prerequisite for TiO2 nanotube modification by means of Si
electroreduction, which will be addressed in our further work.
Targeting practical application of the obtained Si coatings in
LIBs the Si/Cu electrodes are investigated as anodes in standard and
in environmentally benign, non-flammable ionic liquid electrolytes with improved safety.
2. Experimental
2.1. Reagents and chemicals
Ethylene carbonate (EC), dimethyl carbonate (DMC), SiCl4,
LiTFSI and LiPF6 were purchased from Alfa Aesar. The reagents
were used as received without further purification. The ionic liquid
[BMP][TFSI] was obtained from IoLiTec (Heilbronn, Germany). It
was dried at 50 C under argon atmosphere and vacuum until a
value of 15 ppm H2O was reached. The H2O content in the
electrolyte solutions was monitored by Karl - Fischer titration,
using a Metrohm 831 KF Coulometer.
2.2. Silicon electrodeposition
Silicon electrodeposition was carried out in a home-made PTFE
electrochemical cell designed for quartz crystal microbalance
(QCM) measurements (see section 2.4). The three-electrode
system consisted of a Cu plate as working electrode (0.23 cm2
geometric area) and two Pt wires as counter and quasi-reference
electrodes (QRE), respectively. The Pt electrodes were thoroughly
cleaned with HNO3 prior to use. Before silicon electrodeposition,
the Cu electrodes were freshly polished and sonicated in ethanol
for 10 minutes. The cell assembly and electrochemical measurements were performed at room temperature (20 C) in a glove-box
(MBRAUN UNIlab LMF auto) maintaining H2O and O2 levels below
0.1 ppm. The potentiostatic deposition of silicon was performed at
2 V vs. Pt QRE for 2 hours from 1 M SiCl4 in [BMP][TFSI].
Immediately after the deposition, the samples were thoroughly
rinsed with DMC.
2.3. Chemical and structural characterization
The surface morphology of electrodeposited Si samples was
investigated using an ultrahigh resolution scanning electron
microscope (FE-SEM, Hitachi S-4800) equipped with EDX analysis.
The chemical composition of the electrodeposited films was
investigated by X-ray photoelectron spectroscopy (XPS) depth
profiling using a Specs SAGE system (base pressure <1 108
mbar) equipped with a Phoibos 150 electron analyzer including a
1D delay line detector. AlKa radiation was generated by a XR 50 M
X-ray source in combination with a Focus 500 monochromator. For
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
405
the sample transfer from the glove box to the surface analysis
system an inert gas carrier box was used. An exposure to
atmospheric conditions for approximately 60 s could not be
avoided during the insertion into the load lock chamber. Sequential
XPS depth profiling was performed utilizing a differentially
pumped IQE 12/38 ion source operated with Argon (ion source
Ar pressure 2.2 103 mbar, ion energy 3 keV, emission current
10 mA, ion current 6 mA) and laterally scanning the Ar+ ions
across the sample surface (scan area 4 4 mm2). X-ray diffraction
of the samples was performed by means of X-Ray diffractometer,
Bruker D5000 with Göbel mirror, using CuKa radiation.
2.4. QCM measurements
QCM measurements were carried out in the same electrochemical setup described above, replacing the Cu plate by a quartz
crystal (f0 = 10 MHz, AT cut). Technical details for the setup of
electrochemically coupled QCM can be found in Ref. [41]. Cu thin
film electrodes (200 nm) on a Cr adhesion layer (5 nm) were
thermally evaporated onto the bare quartz crystal resonators
(Vectron International). The admittance of the quartz crystal was
monitored using a Saunders 250B Network Analyzer PCI computer
card (Saunders & Associates Inc.) and a lab-made software,
extracting on line the resonance frequency f and the damping
parameter w (full width at half height).
2.5. Electrochemical characterization
The electrodeposited silicon samples were electrochemically
characterized using a potentiostat/galvanostat Biologic VMP3 by
cyclic voltammetry and galvanostatic experiments. The cyclic
voltammetry experiments were carried out in a lab-made PTFE
electrochemical cell equipped with Li metal counter and reference
electrodes at 0.5 mV s1 scan rate. The galvanostatic measurements were carried out using CR2032 coin cells, sealed with a
hydraulic crimping machine (MSK 110 from MTI Corporation). The
electrolytes used for the electrochemical experiments consisted of
1 M LiPF6 dissolved in EC:DMC (1:1 vol. %) and 1 M LiTFSI dissolved
in [BMP][TFSI].
3. Results and discussion
3.1. Electrodeposition of silicon
3.1.1. Voltammetric measurements
In order to study the silicon reduction in 1 M SiCl4 [BMP][TFSI]
and to determine the cathodic limit of IL stability linear sweep
voltammetry (LSV) was performed, coupled with QCM measurements. EQCM is a useful tool for in-situ characterization of the
microgravimetric behavior of the system. This technique is
particularly helpful to discriminate electrochemical reactions
related to mass changes that take place on the electrode surface.
The voltammetric curves in both pure ionic liquid and 1 M SiCl4
[BMP][TFSI] electrolytes are shown in Fig. 1a.
The absence of voltammetric waves in the pristine IL at
potentials E > 2 V underpins its good cathodic stability. Decomposition can be observed at potentials more negative than 2 V.
The voltammetry performed in SiCl4 containing IL displayed
several cathodic waves. The first, observed at approximately
0.37 V can be attributed to the reduction of a surface oxide layer
on the copper electrode. This assumption is supported by the
simultaneous increase in frequency of the QCM (Fig. 1b) which
corresponds to a mass decrease. Furthermore, the damping of
the QCM (w) hardly changes indicating that this is a process
involving a rigid and smooth surface. Between 0.8 and 1.9 V, a
peak at 1.13 V followed by a plateau for more negative voltages
Fig. 1. Linear sweep voltammetry measured in pure [BMP][TFSI] (- - -) and in 1 M
SiCl4 [BMP][TFSI] (___) (a). Frequency and damping changes measured in situ with
linear sweep voltammetry in 1 M SiCl4 [BMP][TFSI] (b).
is observed. Since the frequency change around 1.13 V is rather
weak, bulk deposition can be ruled out. However, possible partial
reduction of Si4+ to Si2+ and/or restructuring of the surface [42]
(adsorption) can explain the observed behavior.
Further reduction of Si species is observed at more cathodic
potentials, displayed by an intense peak at 1.95 V, which is
accompanied by a significant frequency decrease. In parallel to the
frequency change the damping increases approximately twice the
frequency shift. Therefore, at these conditions an accurate mass
evaluation of the deposited silicon from the frequency change
using the Sauerbrey equation cannot be performed, as shown in
the studies describing QCM application and theoretical principles
of the method [43,44].
In further experiments the electrochemical behavior of freshly
polished Cu electrodes in SiCl4 [BMP][TFSI] was studied by means
of cyclic voltammetry (Fig. 2) performed between 0 V and 2.2 V.
The voltammetric data clearly indicate the irreversibility of the
silicon electrodeposition. The observed about two times higher
cathodic currents compared to the LSV scan on Fig. 1 are related to
the influence of the larger active surface of the mechanically
polished Cu substrates. The peak at 1.9 V (labelled 1 in Fig. 2) is
related to Si deposition and shifts to more positive potentials while
its intensity decreases with increasing cycle number.
Fig. 2. First (black), second (red) and third (blue) voltammetric scans measured in
1 M SiCl4 [BMP][TFSI]. The numbers denote the sequence of voltammetric scans.
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
406
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
The positive shift in the peak potential is probably related to the
high activity of the already formed Si clusters, on which further
electrodeposition is energetically preferred. Additionally, the
decreasing currents in the consecutive scans can be ascribed to
a decrease in the active surface of the deposit, succeeding the
process of electrochemical nucleation on the foreign Cu substrate.
Alternatively, the observed current decrease can be caused by
depletion of the active species near the electrode surface.
Furthermore, a positive potential shift in the following scans is
also present for the peak at 1.13 V (labelled 10 ), indicating that the
above described phenomena related to this peak are facilitated in
the subsequent voltammetric scans.
3.1.2. Potentiostatic deposition of silicon
Silicon was potentiostatically deposited onto Cu electrodes
from 1 M SiCl4 [BMP][TFSI] electrolyte for 2 h at 2 V vs. Pt QRE.
The electrodeposition showed initially a high current followed
after about 60 seconds by a stable steady-state at approximately
0.35 mA cm2 and by further slow current decrease (Fig. S1). The
deposited Si layer has a transparent, slightly yellowish-green color
similar to other reports of Si electrodeposited on Cu [18] or Ni [20].
As for the LSV experiments, there is an increase in the damping
of the QCM during Si electrodeposition, suggesting a considerable
deviation from the formation of a smooth and rigid thin layer, not
allowing application of the Sauerbrey relation for further analysis.
It has been established that the increased resonator damping can
be caused by changes in the morphology and rigidity of the layer.
For example, the physical reason for the high damping in case of
electrodeposited conducting polymer layers was interpreted as an
individual or mutual contribution of roughness and viscoelastic
effects [44]. The EQCM data analysis can give a further contribution
to the identification of the supposed chemical nature of the
deposit. The frequency change depending on the charge accumulated during the potentiostatic silicon deposition is presented in
Fig. 3. In case that the deposit consists of a rigid and smooth silicon
layer, the slope of the dependence should be proportional to the
apparent molar mass of the deposit (Mapp = MSi, z = 4), where z is
the number of transferred electrons. However, the experimental
graph shows a much higher slope than the theoretically expected
one for the four-electron-transfer (Si4+ to Si0) with 100% current
efficiency (dashed line). Furthermore, the rather high damping
during the microgravimetric measurements indicates that the
frequency decrease is not only caused by a mass increase but it is
also affected by surface roughness and viscoelastic effects.
Moreover, additional mass in form of ionic liquid could be trapped
in the deposit and/or decomposed in parallel to the Si deposition.
Fig. 3. Frequency - charge dependence during the electrodeposition of Si from 1 M
SiCl4 [BMP][TFSI] on Cu-coated quartz crystal resonator (___). Theoretical line
demonstrating frequency - charge dependence under assumption of 100% Si
deposition efficiency and 4 electron transition (- - -).
Consequently, the IL decomposition products can be incorporated
into the film, affecting the apparent molar mass (Mapp > MSi, z > 4).
These aspects are studied in more detail using XPS analyses
discussed below. For most experiments the deposition ceased at a
frequency change of 30 kHz as can be observed from the noisy
horizontal part of the graph. This can be explained by the higher
electrical resistance of the deposited Si layer, which is consistent
with the decreasing currents observed during the potentiostatic
deposition (Fig. S1). The effect of increased QCM damping was
reported in a recent work on Si electrodeposition from trimethylhexyl ammonium bis(trifluorosulfonyl) imide [36]. In this case
however the results display Si deposition at much lower potential
(2.6 V). Furthermore, the deviation from the mass increase
expected for pure Si deposition was attributed to integration of
non-decomposed IL species in the layer structure.
3.2. Surface morphology and chemical composition of the layers
The surface morphology of the Si layers was characterized by
SEM. The resulting micrographs are presented in Fig. 4. The top
view micrographs (Fig. 4a, b) show a deposit with a superimposed
silicon layered structure. At a sub-micrometer scale, the material
has a globular surface morphology consisting of spherical units
between 100 and 600 nm in diameter (Fig. 4b). Furthermore, this
particular structure of the deposit can facilitate the trapping of
ionic liquid or its decomposition products in between the layers
(Fig. 4c, d). The observed morphology most likely results from the
internal stress accumulated inside the layer during its deposition
[17]. The thickness of the obtained coating, estimated from the
cross section of the SEM images, was 1.0–1.2 mm (Fig. 4c, d). Similar
morphologies of electrodeposited silicon have been reported in
earlier studies [19,30,32].
The observed morphological properties of the layers as studied
by SEM are in a good accordance with the results from the
microgravimetric characterization. Roughness can be the main
factor for the damping increase for earlier deposition times (thin
films). Furthermore, Si in elemental or oxidized state is the main
component of the layer with physical properties of a solid. On the
other hand as already observed, the cross-sectional SEM images
showed laterally located cavities, where ionic liquid or its
decomposition products can be trapped and thus to affect the
rigid behavior of the layer.
EDX analysis was performed to qualitatively investigate the
elemental composition of the prepared samples (an example is
shown in the supplementary material Figure S2). Since the
electrochemically deposited layers have a low thickness
(1 mm) typical for thin films, most of the EDX signal originates
from the Cu substrate and its belonging natural oxide layer.
Nevertheless, next to the corresponding features of the main film
constituents (silicon and oxygen), trace signals from nitrogen,
carbon, fluorine and chlorine were detected as well. However, due
to the sample structure, a quantitative analysis is rather difficult
and we refer to the XPS analysis below as it presents more
assessable analytical data about the thin film composition. The
additionally performed XRD measurements (not shown here) do
not indicate crystallinity, suggesting that the silicon deposit is
amorphous or consists of particles with diameter of a few
nanometers.
The elemental composition and chemical structure of the
deposited films was analyzed by XPS. The red lines in Fig. 5a
represent the O1s, Si2p, C1s and N1s XPS core level spectra directly
measured after the sample was inserted into the UHV analysis
chamber. Since slight charging of the sample occurred during the
measurements, the binding energy scale of the red spectra is
corrected (DEbin = 1.1 eV) with respect to the Si2p core level of a
naturally oxidized Si wafer (black) leading to a binding energy (BE)
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
407
Fig. 4. SEM images of the electrodeposited layer (a,b) top view, (c,d) cross-sectional view at different magnifications.
of 689.9 eV for the F1s signal (not shown), which is in the expected
range for the CF3 component of [TFSI] [45]. The Si2p spectrum of
the deposited film includes two components at 103.9 and
100.0 eV that can be assigned to SiO2 (Si4+) and Si
Si (Si0)
bonds, respectively. The Si4+ component clearly dominates this
spectrum, indicating a strong oxidation of the surface. The O1s
spectrum with a single peak at 533 eV BE is also comparable to
the O1s reference of the Si wafer.
Besides the silicon and oxygen signals, carbon (Fig. 5a (c)),
nitrogen (Fig. 5a (d)) as well as fluorine (not shown) were detected.
The C1s spectrum includes three peaks at 292.9, 286.3 and
285.2 eV, respectively. Comparing the spectrum with the C1s
shape and features of the neat [BMB][TFSI] ionic liquid [45],
the two peaks at low BE can be clearly assigned to the [BMP]+
cation, while the carbon atoms bound in the [TFSI] anion
would explain the weak feature at 292.9 eV. However the
C1s[BMP]/C1s[TFSI] intensity ratio is much higher compared to the
neat IL, since the amount of CF3 groups from [TFSI] is drastically
lower on the electrodeposited Si film surface. At the same time, the
N1s spectrum is composed of a strong feature at 402.3 eV and a
shoulder at 399.5 eV BE. The signature at 402.3 eV is caused by
nitrogen atoms in the pyrrolidinium ring of [BMP] [45]. N atoms in
the [TFSI] anion configuration would lead to a second peak at
3.2 eV lower energy [45], equivalent to a BE of 399.1 eV as
indicated by the dotted vertical line in Fig. 5a (d). Again, no
considerable contribution of the [TFSI] anion is found when
compared to spectra of the neat IL. The detected weak shoulder at
399.5 eV could be originating from fragments of [TFSI], altered
by the electrodeposition process or could be related to NOx bonds
formed at the surface during contact with the ambient. The F1s
signal (not shown) is at 689.9 eV BE, in the energy range where the
F1s signal of the neat [TFSI] is reported [45]. No sulfur, also a
component of the [TFSI] anion, was found within the detection
limit of XPS (0.1 at. %). These surface characteristics allow a partial
analysis of the processes during electrodeposition in the IL. The
absence of significant [TFSI]-related signals in the C1s and N1s
spectra points to a fragmentation of the [TFSI] anions at the
electrode, enabling incorporation of oxygen, carbon, nitrogen and
fluorine species into the growing film, while the [BMP]+ cations are
more stable and remain accumulated at the surface after removal
of the sample from solution and thorough washing with dimethyl
carbonate.
The elemental composition at the surface in at. % is represented
by the first data points in Fig. 5b at 0 min sputter time (light red
vertical line). The additionally detected chlorine signal can be
explained by SiCl4 dissolved in [BMP][TFSI] used as precursor for Si
deposition.
Ar+ ion sputtering of the sample reveals the elemental
composition of the bulk of the film. The results of the quantitative
XPS depth profile analysis are presented in Fig. 5b. Corresponding
core level spectra after a sputter time of 48 min, representing the
bulk of the deposited film (indicated by the light blue vertical line
in Fig. 5b), are plotted for comparison in Fig. 5a. During the first
sputter cycles, the reduction of the oxygen amount and the
increase of the silicon signal, combined with the strong reduction
of the Si4+ component in the Si2p signal (see Fig. 5a), can be
explained by the removal of the silicon oxide surface layer that was
formed during exposure of the sample to ambient conditions
before insertion into the UHV system.
In addition, a decrease in the amount of fluorine and carbon as
well as the obvious changes in the C1s and N1s core level binding
energies (see Fig. 5a) indicate the removal of IL residuals
accumulated at the surface. Although the amount of nitrogen is
comparable at the sample surface and in the bulk of the deposited
film, the distinct change in core level BE (see Fig. 5a) indicates that
the bulk N is incorporated in a different chemical state compared to
the N species existing at the sample surface, where it mainly
belongs to the [BMP]+ cation.
408
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
Fig. 5. (a) XPS O1s (a), Si2p (b), C1s (c) and N1s (d) core level spectra of the electrochemically deposited films; red: sample surface as received (BE corrected by 1.1 eV), blue:
measurements acquired during the depth profiling after 48 min of Ar+ ion sputtering, black: spectra of a naturally oxidized Si wafer. All spectra are normalized to their
respective peak maximum. (b) Depth profile of the elemental composition in electrochemically deposited Si films on Cu, determined by XPS analysis. The film preparation
stages for which the detail spectra are presented in Fig. 5a are indicated by light red (as received) and light blue (sputtered film - approx. half of the Si film removed) vertical
lines. The dashed line indicates the sputtering time, where the detected Cu concentration is equal to that of Si (the film substrate interface).
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
Apart from the outermost surface, in the applied quantification
model the distribution of silicon, oxygen, carbon, nitrogen
and fluorine throughout the film is almost constant at levels of
48–49 at. % for Si, 21–22 at. % for O, 18–20 at. % for C, 2.0–2.5 at. % for
N and 1.0–1.4 at. % for F, respectively. The proportion of
incorporated Cl remains almost constant (2.5 at.%) through the
whole film, while the copper substrate signal increases as expected
with increasing sputter time. It has to be mentioned that the
elemental depth distribution in Fig. 5b exhibits a very broad
interface region between the deposited film and the Cu substrate.
Furthermore, the Cu signal already slightly increases in the early
stages of material sputter removal. We assign these characteristics
to the granular structure and the rough topography, which affect
the sputter uniformity of the film, as well as to the possible
existence of cracks, where the Cu substrate is uncovered within.
Since carbon, oxygen, nitrogen and fluorine are components of
the IL used as electrolyte for Si deposition, we attribute the
detected impurities to fragments of the IL that decompose during
the deposition process and are incorporated into the film due to the
equivalent potential range for Si deposition and IL decomposition.
The amount of incorporated oxygen might be also attributed to the
high reactivity of electrodeposited Si and a possible diffusion of
oxygen into the film during contact to ambient conditions. In this
context it is important to note that the found quantity of included
oxygen is considerably lower compared to earlier studies on Si
based porous electrode materials [32,46].
A thorough analysis of the core level spectra from the bulk of
the deposited films reveals that the incorporated elements exist in
various chemical states. The very broad Si2p spectrum around
102 eV includes several oxidation states of Si. While the Si
O2 (Si4
+
) and Si
Si (Si0) components represent minor margins of the
signal, a convolution of sub-stoichiometric Si configurations
(oxidation states Si+, Si2+ and Si3+) mainly contribute to the
spectral shape. Si
C, Si
Ox (x < 2), Si
N and SiFx (x < 2) bonds
are typically found in the binding energy ranges of 100–101 eV,
101–103 eV, 102 eV and 101–103 eV, respectively [21,47–49].
However, the amounts of incorporated nitrogen and fluorine are
rather low for a significant contribution to the Si2p spectrum.
Furthermore, the C1s signal at 283.7 eV, the O1s peak at 532.3 eV
and the N1s state at 398.0 eV also point to the existence of silicon
carbide, sub-stoichiometric silicon oxide and silicon nitride bonds
[49–52]. In relation with the amount of incorporated oxygen and
carbon, it is concluded that the films basically consist of an
amorphous network of silicon oxycarbide (SiOxCy) with N, F and
Cl impurities. Due to the observed characteristics of chemical film
properties, we attribute the bulk incorporation of oxygen as well as
the C, N, F and Cl atoms to the fragmentation of [TFSI] anions and the
existence of Cl ions in solution during the electrodepositionprocess.
3.3. Lithiation behavior of electrodeposited silicon
The Li ion insertion/extraction in the electrodeposited silicon
was measured initially by means of cyclic voltammetry in 1 M LiPF6
EC/DMC electrolyte. The voltammetric cycling of the samples was
performed between 1.5 V and 0.02 V vs. Li, Li+ with a scan rate of
0.5 mV s1. The deposited silicon showed the characteristic
voltammetric peaks for the process of alloying and de-alloying
with lithium (Fig. 6a).
The first voltammetric cycle displayed a very sharp cathodic
current shoulder corresponding to Li – Si alloying reactions. The
initial Li–ion insertion is shifted to negative potentials, starting at
0.2 V vs Li,Li+. This can be explained by the lower conductivity of
the non-lithiated silicon layer. Consequently, the Li+ insertion in
the structure during the first cathodic sweep is very inhibited,
requiring a high overpotential for alloying with lithium [53]. The
high cathodic current during the first scan might partially be due to
409
Fig. 6. First (black), second (red), 15th (green) and 20th (blue) voltammetric curves
of electrodeposited Si layers measured in: 1 M LiPF6 EC/DMC (a) and 1 M LiTFSI
[BMP][TFSI] (b) electrolytes. v = 0.5 mVs1. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
the passivation of the electrolyte on the electrode surface
(formation of the solid-electrolyte interfacial layer (SEI)), irreversible reaction of Li ions with dangling silicon bonds and/or with SiOx
present at the surface of the deposited silicon [16]. In the
subsequent voltammetric scans the cathodic current decreases
and reaches saturation after the 10th cycle. This suggests that after
a certain number of cycles the process of Li+ exchange reaches a
steady state and a constant amount of silicon takes part in the
alloying/de-alloying process. The initial decrease of the cathodic
current might be explained by the loss of contact between the
particles, which have low mechanical stability after the electrodeposition. On the other hand, the current drop in the first
voltammetric cycles can be related to the high volume change
and a mechanical instability during the initial lithium insertion.
Nevertheless, stabilization of the current in terms of obtaining a
steady state signal suggests that the major part of the deposited Si
remains stable during voltammetric cycling. The first anodic sweep
showed two peaks at 0.34 V and at 0.49 V corresponding to the
phase transition from crystalline Li–Si phase to amorphous silicon
[53]. The next cycles displayed more or less the same behavior,
characteristic for amorphous silicon with a slight increase in the
current values, probably due to an increased conductivity of the
silicon lithium matrix.
Besides the conventional organic carbonate based electrolyte,
the samples were also cycled in 1 M LiTFSI [BMP][TFSI]. Voltammetric curves of electrochemically deposited Si on Cu electrodes
measured in the Li+ containing IL electrolyte are shown in Fig. 6b
for comparison. Voltammetric features of Si lithiation/delithiation
are observed for the IL electrolyte as well. In contrast to the sample
410
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
cycled in 1 M LiPF6 EC/DMC, in IL electrolyte the lithiation current
keeps a relatively constant maximum values with the cycle
number and the delithiation current gradually increases, reaching
a saturation around the 10th cycle. The lower voltammetric
currents in IL media indicate hindered lithiation/delithiation,
involving lower silicon amount. Furthermore, the absence of
current decrease in the following (anodic and cathodic) voltammetric scans points toward a good mechanical stability of the
electrode. It is worth noting that the de-lithiation of Si in IL media
occurs at markedly positive potentials, suggesting an inhibition of
Li electrochemical extraction in comparison with Si studied in the
1 M LiPF6 EC/DMC electrolyte. A possible explanation for this
observation is the hindered dynamics of the Li ion transport
between the Si containing grains and in the layer pores and cavities
due to the high viscosity of the IL media.
Electrochemically deposited Si containing layers were finally
tested by constant current cycling (Fig. 7). The potential profiles
measured in standard (Fig 7a) and the IL based electrolyte (Fig 7b)
show that in both media there is a high irreversible capacity,
yielding 76% for EC/DMC and 67% for the [BMP][TFSI] based
electrolytes. The large irreversible capacity during the first cycle
can be caused by a partial active material disintegration,
conversion of SiOx to LiySi and Li2O [30] as well as SEI formation.
After a number of cycles (30 for EC/DMC and 20 for IL electrolyte),
the current efficiency increases to almost 100% in both electrolytes. The voltage profiles of Li insertion and extraction have been
further analyzed in relative capacity coordinates for more
appropriate comparison at identical lithiation state (Fig. 7c). It
was found that the samples cycled in the EC/DMC based
electrolyte show a gradual, nearly linear decrease of the potential
corresponding to the process of SEI formation until 0.15 V,
displaying a distinct point where the Si alloying with Li begins. In
contrast, the cycling in the IL electrolyte demonstrates a smooth
and delayed transition into the alloying plateau (Fig. 7c). The
observed effect can be attributed to the incomplete electrode
passivation in the IL, overlapping with the process of Si lithiation.
In this way, the IL reduction continues to contribute to the charge
capacity in the subsequent cycles. As already observed for the
experiments performed in the voltammetric mode, the galvanostatic discharge transient in the IL displayed a considerable
inhibition of the de-lithiation process, evidenced by a potential
polarization of more than 170 mV. The observed difference in the
anodic potential profiles can be related to the influence of
additional passivating products deposited during the charging
process and hindered Li+ transport out of the bulk. The inhibition
of the lithiation/delithiation phenomena in IL media can be
further accelerated by the significant viscosity of the IL and slow
Li+ transport through the Si/electrolyte interphase.
The evolution of charge and discharge capacity values with the
cycle number measured in both electrolytes is presented in Fig. 8.
The specific capacities were determined by taking into account the
mass of the deposit calculated from the potentiostatic transients
considering 50% silicon in the electrodeposited composite,
obtained by XPS analysis.
There is a high irreversible capacity during the first cycle for
both electrolytes (Fig. 8). Stable capacity values of about 1200 mAh
g1 at 0.4 A/g and 1000 mAh g1 at 1.6 A/g were obtained in 1 M
LiPF6 EC/DMC for more than 100 cycles. The electrochemical
parameters of the layers appear to be promising when compared
with literature reports summarized in Table 1.
It can be seen that specific discharge capacities obtained in
carbonate based electrolyte are comparable [18,28,29] or even
superior [46] to already reported ones. A capacity retention value
of about 80% after 350 cycles performed at different cycling
currents has been observed, displaying a good cycling stability
(Fig. S3). The effective preservation of the structural integrity of the
Fig. 7. Galvanostatic cycling profiles of Si containing layers measured in 1 M LiPF6 EC/DMC (a) and 1 M LiTFSI [BMP][TFSI] (b) electrolytes. Comparison of the potential profiles
is after capacity normalization. 1 M LiPF6 EC/DMC (black), 1 M LiTFSI [BMP][TFSI] (red) (c). j = 0.4 A g1. The numbers denote the sequence of galvanostatic cycles.
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
Fig. 8. Evolution of charge (square) and discharge (circle) capacities of the
electrodeposited layers with cycle number, measured in 1 M LiPF6 EC/DMC (full
symbols) and 1 M LiTFSI [BMP][TFSI] (open symbols). The numbers denote the
cycling current densities and C-rates.
layer during electrochemical cycling can be related to the organic IL
decomposition products that are incorporated during Si deposition, acting as a mechanical buffer against the strain accumulation.
In contrast to the cycling in a conventional carbonate based
electrolyte, the galvanostatic measurements in the IL showed
lower capacities (Fig. 8). The capacity values obtained at current
densities in the interval 0.4–1,6 A/g were in the range between
630 and 460 mAh g1. The lower capacities observed in the IL
confirm that the electrochemical lithiation of Si is hindered in
comparison with that measured in the EC/DMC based electrolyte.
Nevertheless, the capacity retention of about 500 mAh g1 for
more than 120 cycles remains comparatively stable (Fig. S4). To our
knowledge there is a lack of research on electrochemical behavior
of electrodeposited silicon or silicon composites in Li–ion
containing [BMP][TFSI]. A preliminary study in this field showed
the voltammetric behavior of electrodeposited Si in [BMP][TFSI],
emphasizing on the future applicability of electrodeposited thin Si
411
films and the necessity of further research [53]. Furthermore,
composite electrodes based on dispersed silicon powders have not
been studied in [BMP][TFSI] and their performance in other IL
based electrolytes is presented in a limited number of contributions [19,54–56]. It was shown that the cycling performance of Si
strongly depends on the ionic liquid type. One of the most
promising results was found for LaSi2/Si composites cycled in Nmethyl-N-propylpiperidinium bis(fluorosulfonyl) imide [54]. It
was further observed that the performance of Si electrodes
significantly improves after exchanging the ionic liquid anion bis
(trifluoromethanesulfonyl) amide (TFSA) by bis(fluorosulfonyl)
amide (FSA), which can be explained by an easier desolvation of Li
and FSA ions [57]. Generally, the performance of thick Si containing
films based on powder material in ionic liquid is higher than for the
electrochemically deposited Si layers. On the other hand, our
investigation on electrochemical activity of magnetron sputtered
Si structures in LiTFSI [BMP][TFSI] showed much higher capacities
in the range of 2200–1200 mAhg1 for 200 cycles [39]. This
comparison implies that beside the ionic liquid type as well the
chemical composition and morphology of the Si containing layer
play an important role for the electrochemical performance of the
material.
For both electrolytes the current efficiencies were high (more
than 99%, Table 1), except for the initial cycles, when the formation
and stabilization of the SEI layer or initial degradation phenomena
lowered the current efficiency of the galvanostatic cycles.
The SEM images (Fig. 9) of the deposits after 300 galvanostatic
cycles, performed in both types of electrolytes, reveal changes in
the morphology. The layer cycled in 1 M LiPF6 EC/DMC still appears
to be stable and compact (Fig. 9a,b). The observed structural
difference between the pristine and electrochemically cycled
material is caused by the process of alloying and de-alloying with
lithium. The morphological transformation exhibits a loss of the
initially obtained globular structure, presumably due to amorphisation and SEI layer formation (Fig. 9a,b). Furthermore, the silicon
layer cycled in organic carbonate electrolyte displays a uniform
structure, whereas the layer tested in IL electrolyte shows macrocracks, probably resulting from the internal stress during the
deposition (Fig. 9c,d). Unlike the sample tested in organic
Table 1
Comparison of synthesis conditions and cycling performance of electrodeposited Si layers.
Electrodeposition conditions
Cycling electrolyte
C-rate /
Current density
Capacity /
mAh g1
Current efficiency / %
Electrolyte:0.5 M SiCl4 in propylene carbonate (PC)
and 0.1 M tetrabutylammonium chloride (TBACL); Si on 3D virusstructured nickel; galvanostatic
and potentiostatic;
Electrolyte:0.5 M SiCl4; 0.1 M TBACL in PC; amorphous Si on Cu;
galvanostatic;
1 M LiPF6
EC/DMC (1:1) (v:v)
0.25 C
1200 after
160 cycles
77,5% in the first cycle [28]
and 99.5% in the eighth
cycle at 0.25C
1 M LiPF6
EC/DEC (1:2)
(v:v)
1 M LiPF6 EC/DMC
(1:1) (v:v)
0.4 A g1
1300
99% up to 100 cycles at [18]
0.4 Ag-1
50–200 mA cm-2
1100–800
[46]
95% in the first cycle
and 99% for the further
cycles at 0.1-C rate
96% at 0.36 A g-1
[30]
Electrolyte:0.5 M SiCl4, 0.5 M TBAClO4 in PC; Si composite on Ni micronanocones; galvanostatic;
Electrolyte:0.3 M SiCl4 and 0.1 M TBACL in CH3CN; Si film on Ni foam;
potentiostatic;
Electrolyte:0.5 M SiCl4 and 0.5 M TBAP/PC; SiOC composite on Cu foil;
galvanostatic
Electrolyte:1 M SiCl4 in [BMP][TFSI] SiOC composite on Cu foil
potentiostatic
0.36–18 A g1
1 M LiPF6 EC/DMC
(1:1)
(v:v)
1 M LiClO4 PC/EC (1:1) 0.25 A cm-2
(v:v)
(0.4–1.6) A g1
1 M LiPF6 (EC/DMC)
(1:1) (v:v)
1 A g1
1 M LiTFSI [BMP][TFSI]
2800–1000
1045 at 2000th
cycle;
842 at 7200th
cycle;
1200 For
300 cycles
500 for
120 cycles
Ref.
29% in the 1st cycle
>98% for the further
cycles
[29]
25% in 1 M LiPF6 EC/
DMC (1st cycle)
33% in 1 M LiTFSI
[BMP][TFSI](1st cycle)
99% in both
electrolytes for the
further cycles
This
work
412
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
Fig. 9. SEM imaging of the Si-containing layers after 300 galvanostatic cycles taken in 1 M LiPF6 EC/DMC (a,b) and 1 M LiTFSI [BMP][TFSI] (c,d) at low and high magnifications.
carbonate electrolyte, the layer cycled in 1 M LiTFSI [BMP][TFSI]
displayed a partially preserved globular surface morphology.
Apparently, the observed after deposition silicon containing
spherical objects are enlarged and merged due to the volume
increase upon lithiation. Most probably the lack of a complete
electrolyte passivation in IL and the involvement of smaller
amount of electrochemically active material retain the surface
morphology partially globular.
4. Conclusions
Thin silicon containing composite layers were deposited on
copper substrates by potentiostatic deposition from 1 M SiCl4
[BMP][TFSI]. EQCM measurements suggested a significant contribution of side reactions taking place parallel to the process of Si
electroreduction. SEM imaging showed a compact and uniform
globular morphology of the layers with thickness of about 1 mm.
XPS measurements were performed to analyze the elemental
composition and chemical structure of the deposited films. The
analysis suggests a joint major contribution of Si, C and O during
the electrodeposition process, whereas N and F are present in
lower amounts. The distribution of silicon, oxygen, carbon,
nitrogen and fluorine throughout the film is almost constant at
levels of 50 at. % for Si, 20 at. % for O, 20 at. % for C, 2 at. % for
N and 1 at. % for F. The absence of [TFSI] anions related signals in
the C1s and N1s spectra indicates a fragmentation of the [TFSI] at
the electrode, enabling incorporation of oxygen, carbon, nitrogen
and fluorine species into the growing film, while the [BMP]+
cations are more stable and remain accumulated at the surface.
The electrodeposited silicon containing layers showed
stable galvanostatic cycling reaching capacity values of about
1200 mAh g1 and 80% capacity retention after 300 cycles in
standard electrolytes. In ionic liquid media, the material displayed
lower capacity (about 500 mAh g1). The effective preservation of
the structural integrity of the layer during electrochemical cycling
can be related to incorporation of side organic products caused by
IL decomposition, acting as mechanical buffer against strain
accumulation.
The galvanostatic discharge transients in IL displayed considerable inhibition of the de-lithiation process, evidenced by a
polarization of ca. 170 mV. The observed difference in the anodic
potential profiles can be related to the influence of additional
passivating products deposited during the charging process and
hindered Li+ transport out of the bulk. The inhibition of the
lithiation/delithiation phenomena in IL media can be further
accelerated by the significant viscosity of the IL and slow Li+
transport through the Si/electrolyte interphase.
Acknowledgements
Financial support by the DFG funded program WeNDeLIB
(project 6: “Linking of model and commercial active materials for
lithium ion batteries by in situ determination of thermodynamic
and kinetic data”) and by a grant (NanoBatt TNA VII-1/2012) from
the state of Thuringia (TMWAT by LEG Thuringen), co-financed by
the European Union within the frame of the European Funds for
Regional Development (EFRD) is gratefully acknowledged. We are
furthermore thankful for research co-funding by the state of
Thuringia and the European Union (ESF and EFRD) under grants
2012 FGR 0231 and 12021-715.
The authors are thankful to Dr. A. Ispas for her assistance in
accomplishment of the EDX measurements.
C.A. Vlaic et al. / Electrochimica Acta 168 (2015) 403–413
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
electacta.2015.03.216.
References
[1] B. Scrosati, J. Hassoun, Y.-K. Sun, Lithium–ion batteries. A look into the future,
Energy & Environmental Science 4 (2011) 3287.
[2] V. Etacheri, R. Marom, R. Elazari, G. Salitra, D. Aurbach, Challenges in the
development of advanced Li–ion batteries: a review, Energy Environmental
Science 4 (2011) 3243–3262.
[3] D. Golodnitsky, V. Yufit, M. Nathan, I. Shechtman, T. Ripenbein, E. Strauss,
S. Menkin, E. Peled, Advanced materials for 3D microbattery, J. Power Sources
153 (2006) 281–287.
[4] G. Pistoia, Lithium Batteries-science and technology, in: G.-A. Nazri (Ed.),
Springer, 2009.
[5] H. Liu, H.-M. Cho, Y. Meng, Q. Li, Engineering Three-Dimensionally
Electrodeposited Si-on-Ni Inverse Opal Structure for High Volumetric
Capacity Li–Ion Microbattery Anode, Appl. Mater. Interfaces 6 (2014) 9842–9849.
[6] W. Wei, G. Oltean, C.-W. Tai, K. Edström, F. Björefors, L. Nyholm, High energy
and power density TiO2 nanotube electrodes for 3D Li–ion microbatteries, J.
Mater. Chem. A 1 (2013) 8160–8169.
[7] M. Valvo, M. Roberts, G. Oltean, B. Sun, D. Rehnlund, D. Brandell, L. Nyholm,
T. Gustafsson, K. Edström, Electrochemical elaboration of electrodes and
electrolytes for 3D structured batteries, J. Mater. Chem. A 1 (2013) 9281–9293.
[8] S. Cheah, E. Perre, M. Rooth, M. Fondell, A. Harsta, L. Nyholm, M. Boman,
T. Gustafsson, J. Lu, P. Simon, K. Edström, Self-supported three-dimensional
nanoelectrodes for microbattery applications, Nano Letters 9 (2009)
3230–3233.
[9] W. Zhai, H. Zhu, L. Wang, X. Liu, H. Yang, Study of PVDF-HFP/PMMA blended
micro-porous gel polymerelectrolyte incorporating ionic liquid [BMIM]BF4 for
Lithium ion batteries, Electrochim. Acta 133 (2014) 623–630.
[10] J. Liu, J. Jönsson, G. Jiang, Application of ionic liquids in analytical chemistry,
Trends in Analytical Chemistry 24 (2005) 20–27.
[11] U. Kasavajjula, C. Wang, A.J. Appleby, Nano- and bulk-silicon-based insertion
anodes for lithium–ion secondary cells, J. Power Sources 163 (2007) 1003–
1039.
[12] H. Wu, Y. Cui, Designing nanostructured Si anodes for high energy lithium ion
batteries, Nano Today 7 (2012) 414–429.
[13] L. Ji, Z. Lin, M. Alcoutlabi, X. Zhang, Recent developments in nanostructured
anode materials for rechargeable lithium–ion batteries, Energy &
Environmental Science 4 (2011) 2682–2699.
[14] T.D. Hatchard, J.R. Dahn, In Situ XRD and Electrochemical Study of the Reaction
of Lithium with Amorphous Silicon, J. Electrochem. Soc. 151 (2004) A838.
[15] J.W. Wang, Y. He, F. Fan, X.H. Liu, S. Xia, Y. Liu, et al., Two-phase electrochemical
lithiation in amorphous silicon, Nano Lett. 13 (2013) 709–715.
[16] V. Baranchugov, E. Markevich, E. Pollak, G. Salitra, D. Aurbach, Amorphous
silicon thin films as a high capacity anodes for Li–ion batteries in ionic liquid
electrolytes, Electrochem. Commun. 9 (2007) 796–800.
[17] M.-S. Park, Y.-J. Lee, S. Rajendran, M.-S. Song, H.-S. Kim, J.-Y. Lee,
Electrochemical properties of Si/Ni alloy–graphite composite as an anode
material for Li–ion batteries, Electrochim. Acta. 50 (2005) 5561–5567.
[18] R. Epur, M. Ramanathan, F.R. Beck, A. Manivannan, P.N. Kumta,
Electrodeposition of amorphous silicon anode for lithium ion batteries,
Mater. Sci. Eng. B. 177 (2012) 1157–1162.
[19] Y. Nishimura, Y. Fukunaka, T. Nishida, T. Nohira, R. Hagiwara, Electrodeposition
of Si Thin Film in a Hydrophobic Room-Temperature Molten Salt, Electrochem.
Solid-State Lett. 11 (2008) D75.
[20] J. Gobet, H. Tannenberger, Electrodeposition of silicon from a nonaqueous
solvent, J. Electrochem. Soc. 10 (1988) 109–112.
[21] P. Ram, J. Singh, T.R. Ramamohan, S. Venkatachalam, V.P. Sundarsingh, Surface
properties of electrodeposited a-Si: C: H: F thin films by X-ray photoelectron
spectroscopy, J. Mater. Sci. 2 (1997) 6305–6310.
[22] F. Martineau, K. Namur, J. Mallet, F. Delavoie, F. Endres, M. Troyon, M. Molinari,
Electrodeposition at room temperature of amorphous silicon and germanium
nanowires in ionic liquid, IOP Conf. Ser. Mater. Sci. Eng. 6 (2009) 012012.
[23] A. Krishnamurthy, D.H. Rasmussen, I.I. Suni, Aqueous Room Temperature
Electrochemical Deposition of Compact Si Films, Electrochem. Solid-State Lett.
14 (2011) D99.
[24] D.R. Lide, CRC Handbook of Chemistry and Physics, 87 ed., CRC Press, Boca
Raton, FL, 2006 ISBN 0-8493-0487-3.
[25] M. Bechelany, J. Elias, P. Brodard, J. Michler, L. Philippe, Electrodeposition of
amorphous silicon in non-oxygenated organic solvent, Thin Solid Films 520
(2012) 1895–1901.
[26] T. Munisamy, A.J. Bard, Electrodeposition of Si from organic solvents
and studies related to initial stages of Si growth, Electrochim. Acta. 55 (2010)
3797–3803.
[27] J.P. Nicholson, Electrodeposition of Silicon from Nonaqueous Solvents, J.
Electrochem. Soc. 152 (2005) C795.
[28] X. Chen, K. Gerasopoulos, J. Guo, A. Brown, C. Wang, R. Ghodssi, J.N. Culver,
A Patterned 3D Silicon Anode Fabricated by Electrodeposition on a VirusStructured Current Collector, Adv. Funct. Mater 21 (2011) 380–387.
413
[29] H. Nara, T. Yokoshima, T. Momma, T. Osaka, Highly durable SiOC composite
anode prepared by electrodeposition for lithium secondary batteries, Energy
Environ. Sci. 5 (2012) 6500–6505.
[30] G. Zhao, Y. Meng, N. Zhang, K. Sun, Electrodeposited Si film with excellent
stability and high rate performance for lithium–ion battery anodes, Mater.
Lett. 76 (2012) 55–58.
[31] S.Z. El Abedin, N. Borissenko, F. Endres, Electrodeposition of nanoscale silicon
in a room temperature ionic liquid, Electrochem. Commun. 6 (2004) 510–514.
[32] C. Fournier, F. Favier, Zn, Ti and Si nanowires by electrodeposition in ionic
liquid, Electrochem. Commun. 13 (2011) 1252–1255.
[33] J. Mallet, M. Molinari, F. Martineau, F. Delavoie, P. Fricoteaux, M. Troyon,
Growth of silicon nanowires of controlled diameters by electrodeposition in
ionic liquid at room temperature, Nano Lett. 8 (2008) 3468–3474.
[34] X. Liu, Y. Zhang, D. Ge, J. Zhao, Y. Li, F. Endres, Three-dimensionally ordered
macroporous silicon films made by electrodeposition from an ionic liquid,
Phys. Chem. Chem. Phys. 14 (2012) 5100–5105.
[35] J. Komadina, T. Akiyoshi, Y. Ishibashi, Y. Fukunaka, T. Homma, Electrochemical
quartz crystal microbalance study of Si electrodeposition in ionic liquid,
Electrochim. Acta. 100 (2013) 236–241.
[36] G.M. Haarberg, L. Famiyeh, A.M. Martinez, K.S. Osen, Electrodeposition of
silicon from fluoride melts, Electrochim. Acta. 100 (2013) 226–228.
[37] A. Lewandowski, A. Swiderska-Mocek, Ionic liquids as electrolytes for Li–ion
batteries - An overview of electrochemical studies, J. Power Sources 194 (2009)
601–609.
[38] S. Ivanov, L. Cheng, H. Wulfmeier, D. Albrecht, H. Fritze, A. Bund,
Electrochemical behavior of anodically obtained titania nanotubes in
organic carbonate and ionic liquid based Li ion containing electrolytes,
Electrochim. Acta 104 (2013) 228–235.
[39] S. Ivanov, R. Grieseler, L. Cheng, P. Schaaf, A. Bund, Electrochemical lithiation of
Si modified TiO2 nanotube arrays, investigated in ionic liquid electrolyte, J.
Electroanal. Chem. 731 (2014) 6–13.
[40] J. Brumbarov, J. Kunze-Liebhäuser, Silicon on conductive self organised TiO2
nanotubes – A high capacity anode material for Li ion batteries, J. Power
Sources 258 (2014) 129–133.
[41] R. Bund, A. Peipmann, Application of PEDOT layers for the electrogravimetric
detection of sulphate and phosphate in aqueous media, Electrochim. Acta 53
(11) (2008) 3772–3778.
[42] Electrodeposition from ionic liquids, in: F. Endres, A. Abbott, D. McFarlane
(Eds.), Wiley-VCH, 2008.
[43] G. Sauerbrey, Verwendung von Schwingquarzen zur Wägung dünner
Schichten und zur Mikrowägung, Zeitschrift für Phys. 155 (2) (1959) 206–222.
[44] A. Bund, M. Schneider, Characterization of the Viscoelasticity and the Surface
Roughness of Electrochemically Prepared Conducting Polymer Films by
Impedance Measurements at Quartz Crystals, J. Electrochem. Soc. 149 (2002)
E331–E339.
[45] S. Men, K.R.J. Lovelock, P. Licence, X-ray photoelectron spectroscopy of
pyrrolidinium-based ionic liquids: cation–anion interactions and a
comparison to imidazolium-based analogues, Phys. Chem. Chem. Phys. 13
(2011) 15244–15255.
[46] T. Hang, H. Nara, T. Yokoshima, T. Momma, T. Osaka, Silicon composite thick
film electrodeposited on a nickel micro-nanocones hierarchical structured
current collector for lithium batteries, J. Power Sources 222 (2013) 503–509.
[47] Z.H. Lu, S.P. Tay, R. Cao, P. Pianetta, The effect of rapid thermal N2O nitridation
on the oxide/Si(100) interface structure, Appl. Phys. Lett. 67 (1995) 2836–
2838.
[48] R. Alfonsetti, L. Lozzi, M. Passacantando, P. Picozzi, S. Santucci, XPS studies on
SiOx thin films, Appl. Surf. Sci. 222 (1993) 70–71.
[49] I. Bertóti, Characterization of nitride coatings by XPS, Surface and Coatings
Technology 151/152 (2002) 194–203.
[50] J.S. Pan, A.T.S. Wee, C.H.A. Huan, H.S. Tan, K.L. Tan, Argon incorporation and
silicon carbide formation during low energy argon-ion bombardment of Si
(100), J. Appl. Phys. 79 (1996) 2934–2941.
[51] J. Finster, D. Schulze, F. Bechstedt, A. Meisel, Interpretation of XPS core
level shifts and structure of thin silicon oxide layers, Surf. Sci. 152/153 (1985)
1063–1070.
[52] J.-W. He, X. Xu, J.S. Corneille, D.W. Goodman, X-ray photoelectron
spectroscopic characterization of ultra-thin silicon oxide films on a Mo
(100) surface, Surf. Sci. 279 (1992) 119–126.
[53] M. Schmuck, A. Balducci, B. Rupp, W. Kern, S. Passerini, M. Winter, Alloying of
electrodeposited silicon with lithium - a principal study of applicability as
anode material for lithium ion batteries, J.Solid State Electrochem. 14 (2010)
2203–2207.
[54] H. Usui, M. Shimizu, H. Sakaguchi, Applicability of ionic liquid electrolytes to
LaSi2/Si composite thick-film anodes in Li–ion battery, J. Power Sources 235
(2013) 29–35.
[55] T. Sugimoto, Y. Atsumi, M. Kono, M. Kikuta, E. Ishiko, M. Yamagata, M. Ishikawa,
Application of bis(fluorosulfonyl) imide-based ionic liquid electrolyte to
silicon–nickel–carbon composite anode for lithium–ion batteries, J. Power
Sources 195 (2010) 6153–6156.
[56] H. Usui, Y. Yamamoto, K. Yoshiyama, T. Itoh, H. Sakaguchi, Application of
electrolyte using novel ionic liquid to Si thick film anode of Li–ion battery, J.
Power Sources 196 (2011) 3911–3915.
[57] H. Usui, T. Masuda, H. Sakaguchi, Li-insertion/extraction Properties of Si
Thick-film Anodes in Ionic Liquid Electrolytes Based on Bis(fluorosulfonyl)
amide and Bis(trifluoromethanesulfonyl) amide Anions, Chem. Lett. 41
(2012) 521.