View Article Online

View Journal

Nanoscale
Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: T. Brezesinski, B.
Breitung, P. Baumann, H. Sommer and J. Janek, Nanoscale, 2016, DOI: 10.1039/C6NR03575B.

This is an Accepted Manuscript, which has been through the

Royal Society of Chemistry peer review process and has been
accepted for publication.

Accepted Manuscripts are published online shortly after

acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

Information for Authors.

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.

www.rsc.org/nanoscale
 Page 1 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

In Situ and Operando Atomic Force Microscopy of High-Capacity

Nano-Silicon Based Electrodes for Lithium-Ion Batteries

Ben Breitung,*a Peter Baumann,b Heino Sommer,a,b Jürgen Janek*a,c and Torsten
Brezesinski*a
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

a
Battery and Electrochemistry Laboratory, Institute of Nanotechnology, Karlsruhe
Institute of Technology, Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-

Nanoscale Accepted Manuscript

Leopoldshafen, Germany
b
BASF SE, 67056 Ludwigshafen, Germany
c
Institute of Physical Chemistry, Justus-Liebig-University Giessen, Heinrich-Buff-Ring
17, 35392 Giessen, Germany

E-mail: ben.breitung@kit.edu; Tel: +49 (0)721 60826439

E-mail: juergen.janek@phys.chemie.uni-giessen.de; Tel: +49 (0)641 9934500
E-mail: torsten.brezesinski@kit.edu; Tel: +49 (0)721 60828827

†Electronic supplementary information (ESI) available.

1
 Nanoscale Page 2 of 19
View Article Online
DOI: 10.1039/C6NR03575B

Abstract

Silicon is a promising next-generation anode material for high-energy-density lithium-

ion batteries. While the alloying of nano- and micron size silicon with lithium is
relatively well understood, the knowledge of mechanical degradation and structural
rearrangements in practical silicon-based electrodes during operation is limited. Here,
we demonstrate, for the first time, in situ and operando atomic force microscopy
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

(AFM) of nano-silicon anodes containing polymer binder and carbon black additive.
With the help of this technique, the surface topography is analyzed while

Nanoscale Accepted Manuscript

electrochemical reactions are occurring. In particular, changes in particle size as well
as electrode structure and height are visualized with high resolution. Furthermore, the
formation and evolution of the solid-electrolyte interphase (SEI) can be followed and
its thickness determined by phase imaging and nano-indentation, respectively. Major
changes occur in the first lithiation cycle at potentials below 0.6 V with respect to
Li/Li+ due to increased SEI formation – which is a dynamic process – and alloying
reactions. Overall, these results provide insight into the function of silicon-based
composite electrodes and further show that AFM is a powerful technique that can be
applied to important battery materials, without restriction to thin film geometries.

Introduction

The ever-increasing energy demands raise the need for storage systems which not
only offer high volumetric and gravimetric energy densities but also show high
reversibility. The Li-ion battery (LIB) is currently the most frequently used and most
important electrochemical energy storage technology.1–5 This is because LIBs usually
have good longevity, are safe (very small rate of failure) and can provide tailorable
energy outputs. Over the past several years, much effort has been invested to
develop novel materials to increase energy density.6–15 The chemical interactions of
different cell components during operation are, unfortunately, rather complex and can
strongly vary for different electrode materials, electrolytes and binders.16–18 Also, it
should be noted that the redox chemistry of a specific storage phase does not control
the electrochemical properties of the electrode material alone. Because lithiation and
delithiation are heterogeneous reactions leading to volume changes, both size and
morphology also affect the reversibility of the Li-storage process.19,20 To enhance the
electrochemical properties of anode or cathode materials, a detailed knowledge of

2
 Page 3 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

the chemical and structural development during cycling is required. Such knowledge
can be obtained, in part, through ex situ measurements,21–23 although they provide
only limited information on the cause or source of capacity degradation and other
phenomena. Moreover, the interpretation of data from ex situ experiments and post
mortem analyses should be always carried out with care. Nevertheless, a variety of
ex situ techniques has been used to gain insight into cell reactions and the function
of battery materials and components in general.24–26 Despite the progress made in
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

studying different battery chemistries, many mechanisms are not yet well understood.

Nanoscale Accepted Manuscript

Thus, improved in situ and operando characterization techniques are needed to
better understand the processes occurring in operating LIBs. In recent years, a
number of in situ and operando techniques have been developed and successfully
implemented, including differential electrochemical mass spectrometry (DEMS),27,28
transmission electron microscopy (TEM),29,30 X-ray diffraction (XRD),31–34 atomic
force microscopy (AFM)35–37 and others. However, unlike AFM, most of them are
ensemble averaged techniques, i.e., they blur out important local or microscopic
details by providing solely spatially averaged results.
While there are only few methods to image the surface of materials during charge
and discharge, there are even less available for LIB electrodes with realistic
dimensions and construction. Here, we demonstrate in situ and operando AFM of
silicon (Si) electrodes containing polymer binder and carbon black additive.
Specifically, the focus is on nanoscale Si, which serves as a model system in this
study. However, we note that AFM can be applied to a wide variety of battery
materials, without restriction to thin films. Si has been shown to have very high
theoretical specific and volumetric capacities of >4000 mAh/g and 8000 mAh/cm3 (for
Li22Si5), respectively.38 These values are around 10 times higher than those of
graphite,39 which is currently the state-of-the-art anode material. Both the high natural
abundance and low toxicity are further properties making Si one of the most
promising negative electrode materials for next-generation LIBs.40,41 In this regard, it
should be noted that recent LIB cells already contain a few wt.% Si in the anode.
Si electrodes have been extensively studied in the past and the underlying alloying
reactions with Li are relatively well understood. In particular, in situ TEM, XRD and
nuclear magnetic resonance (NMR) spectroscopy provided important information on
the lithiation and delithiation of crystalline and amorphous Si as well as structural and
morphological changes during cycling.31,42–44 However, there are also excellent

3
 Nanoscale Page 4 of 19
View Article Online
DOI: 10.1039/C6NR03575B

reports on in situ AFM of thin film electrodes available in the literature.45–48 In

particular, Tokranov et al. recently developed a detailed picture of the solid-electrolyte
interphase (SEI) formation on patterned (planar) Si islands in carbonate-based
electrolytes. They proposed a multiphase SEI with two-layer structure, which evolves
during initial cycling but undergoes significant changes in the electrochemical
properties through the first 20 cycles.49
Here, we investigate laboratory scale electrodes – based on commercial Si
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

nanoparticles – with reasonably high loading and excellent cycling performance in a

Nanoscale Accepted Manuscript

state-of-the-art electrolyte (containing fluoroethylene carbonate) by means of in situ
and operando AFM. This technique allows visualizing changes in particle size,
hardness and surface structure during cycling. Furthermore, as we will show, it also
allows tracking the formation of the SEI, which is known to play a major role in the
cyclability and stability of LIBs.50–53 We emphasize that the present work focuses on
the study of practical nano-Si anodes rather than understanding of the SEI on thin
film model electrodes. To our knowledge, such an experimental approach has not yet
been reported, and we aim to demonstrate that AFM offers valuable information even
in the case of electrodes with complex surfaces. This is highlighted by the fact that
our nano-Si anodes show competitive cycling performance.

Experimental section

Cell assembly and AFM studies

Nano-Si electrodes were prepared by casting a water-based slurry containing 63
wt.% Si particles of size <100 nm (≥98%, Sigma-Aldrich), 22 wt.% Super C65 carbon
black (Timcal) and 15 wt.% poly(vinyl alcohol) Selvol 425 (Sekisui) onto copper foil
(Gould Electronics), followed by drying in vacuum at 80 °C for 12 h. Typical Si
loadings were in the range of 0.6-1.5 mg/cm2. Conventional coin-type cells for long
term cycling tests were assembled in an argon-filled glovebox from MBraun by
stacking 600 µm-thick Li foil (China Lithium Ltd.), GF/D separator (GE Healthcare
Life Sciences, Whatman) and nano-Si electrode. For atomic force microscopy (AFM),
an open ring of 600 µm-thick Li was used as the counter electrode. This design
allowed the AFM head to be lowered onto the top surface of the nano-Si electrodes.
The head also sealed the cell, thus ensuring stable environmental conditions. The
electrolyte used was 1 M LiPF6 in a 1:1 weight ratio of fluoroethylene carbonate
(FEC, Solvay) and ethyl methyl carbonate (EMC, BASF SE). Electrochemical testing
4
 Page 5 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

was performed at 25 °C at rates ranging from C/20 to C/10, with 1C = 4008 mA/gSi,
using either a BioLogic SP-150 potentiostat or MACCOR Series 4000 (Tulsa)
multichannel battery cycler. For in situ AFM, the potential was kept constant until a
current drop of 90% was achieved (CC/CV mode) before probing the electrode
surface. The AFM measurements were conducted on a Bruker AFM (Dimension Icon)
housed inside an argon-filled glovebox by use of a slightly modified ECAFM cell
(Bruker closed electrochemical cell for ECAFM) and SCM-PIT probes with spring
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

constant of 2.8 N/m and Pt/Ir coating. In general, AFM was performed in peak force

Nanoscale Accepted Manuscript

tapping mode in liquid environment at a scan rate of 0.25 Hz and applied tip force of
6 nN. Scanning electron microscopy (SEM) and energy-dispersive X-ray
spectroscopy (EDS) were performed on a LEO 1530 microscope operated at 5 and
10 keV, respectively.

Results and discussion

Cycling performance
Prior to in situ and operando AFM measurements, the nano-Si electrodes employed
in this work were tested electrochemically in Li half-cells to ensure good capacity
retention upon cycling. A detailed overview of the slurry preparation as well as the
choice of polymer binder, electrolyte and Si material can be found elsewhere.16 The
performance was evaluated at a rate of C/10. However, the first (formation) cycle was
performed at C/20. The cut-off voltages were set to 10 mV and 1000 mV and 30 mV
and 600 mV for the cycling at C/20 and C/10, respectively. Fig. 1a shows the
performance over 150 cycles (see also capacity retention and Coulombic efficiency
over 400 cycles in Fig. S1†). A large capacity drop is observed in the first 3 cycles.
This, however, is characteristic of Si-based electrodes and is the result of different
effects (including increase in C-rate and change in potential range), which will be
described in detail below.54 Nevertheless, the nano-Si electrodes demonstrate
excellent stability, with specific capacities of ~2200 mAh/gSi. The capacity fades by
only 2.9% between the 25th and 150th cycles (~0.023% per cycle). Also, the
Coulombic efficiency stabilizes quickly above 99% (Fig. S1†). Overall, the Li-storage
properties of our model electrodes are very good, so that investigations by means of
AFM are worthwhile and can lead to interesting results.
Fig. 1b shows the voltage profiles for the first two cycles at C/20 and C/10,
respectively. While the alloying reaction of Si with Li occurs below 0.2 V with respect
5
 Nanoscale Page 6 of 19
View Article Online
DOI: 10.1039/C6NR03575B

to Li/Li+ in the initial cycle, the first lithiation plateau is shifted towards higher
potentials in the second cycle. This shift is due to changes in electrode structure and
the different lithiation behavior of crystalline and amorphous Si. We note that the
lithiation/delithiation (charge/discharge) potentials in Fig. 1b cannot be directly
compared to those measured during AFM because of the different cell geometry.
However, the profiles and general trends are similar.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

Nanoscale Accepted Manuscript

Fig. 1 Cycling performance of nano-Si electrodes in Li half-cells. (a) Capacity
retention at C/10. (b) Voltage profile of the first and 2nd cycle at C/20 and C/10,
respectively.

Mechanical degradation
One of the major issues of Si-based electrodes is mechanical degradation due to
expansion and shrinkage of Si upon lithiation and delithiation, respectively. These
volume changes are accompanied by the generation of cracks, which may lead to
loss of electrical contact of active electrode regions. Because Si itself is an insulating
material, an interconnected and electrically conductive matrix is needed to ensure
good electron transport through the electrode.55,56 Usually, conductive carbon
additives such as Super C65 or graphite are used to achieve this.

6
 Page 7 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

Fig. 2 shows white light interference microscope images of the top surface of nano-Si
electrodes before and after cycling. The mechanical degradation upon cycling is
clearly visible. The black spots represent surface cracks and pinholes. The area
fraction of “intact” electrode was calculated by use of a graphical program and found
to decrease from 99.5% (as prepared) to 96% (after 10 cycles). Fig. 2 also presents
images of two different electrodes after 150 cycles at C/10. While one of them
showed stable behavior, the other one exhibited significant capacity fading (Fig.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

S2†). The performance degradation was probably due to binder failure (or poor

Nanoscale Accepted Manuscript

electrode adhesion), which can happen in rare cases. The results indicate that the
different cycling behavior is correlated with the fraction of defects: the stable
electrode showed 92% of “intact” electrode surface, whereas only 85% of the surface
was free of cracks and pinholes in the case of the unstable electrode.
Overall, these data confirm the direct correlation between capacity decay and
macroscopic mechanical degradation in Si-based electrodes. It should be noted,
nevertheless, that the mechanical degradation does not only affect the integrity of the
whole electrode. As mentioned above, the alloying process is accompanied by a
large volume increase of approx. 300%. This change in volume has been shown to
generate severe mechanical strain inside the particles (during delithiation this strain
is reversed),6,57 which may lead to fracture and pulverization of the material, and thus
to formation of reactive surfaces and new SEI (and eventually battery failure).

Fig. 2 White light interference microscope images of nano-Si electrodes (a) before
and after (b) 10 and (c, d) 150 cycles. (c, d) Comparison of the top surface of

7
 Nanoscale Page 8 of 19
View Article Online
DOI: 10.1039/C6NR03575B

electrodes with (a) stable and (b) unstable cycling behavior. The black spots
represent cracks and pinholes.

In situ AFM (1st cycle)

For in situ AFM measurements, a fixed area of the top surface was investigated as a
function of potential during charge and discharge. The SEI formation begins before
the lithiation and the corresponding volume expansion of the Si particles occurs. In
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

recent years, electrolytes containing fluoroethylene carbonate (FEC) have proven to

Nanoscale Accepted Manuscript

be particularly effective in enhancing the cycling stability of Si-based electrodes
through formation of a relatively stable and flexible SEI.50–53 Typically, different
lithiation states can be observed, which depend on the choice of binder, electrolyte
and Si material.16
In the following, we focus specifically on the first two charge/discharge cycles of Li-
half cells with FEC-based electrolyte. In general, both particle expansion and
changes in the overall electrode structure and morphology due to SEI formation and
alloying of Si with Li at low potentials are expected. Fig. 3 summarizes these effects.

Fig. 3 Schematic showing the setup used for in situ and operando AFM of nano-Si
electrodes containing polymer binder and carbon black additive. Upon lithiation
(charging), the Si particles expand in volume and probably fracture to some extent. In
addition, SEI formation on the free electrode surface occurs.

Fig. 4 presents AFM height images obtained on a nano-Si electrode. Only a small
selection of images taken during the initial lithiation and delithiation cycle is shown for
clarity (see also images of the lithiation process in Fig. S3†). The size of the Si
particles is in the range of 50-100 nm before cycling (i.e., at open circuit voltage,
OCV). This finding is in agreement with results from SEM (Fig. S4†). In accordance
with the work of Tokranov et al., the first apparent changes – due to SEI formation –
are observed at potentials around 0.6 V.36 After this, there is no further increase in
8
 Page 9 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

particle size until a potential of ~0.2 V is reached and the lithiation of Si occurs. The
expanded particles can be seen particularly well in the image obtained at the cut-off
potential of 0.01 V. Interestingly, the structure of the top surface remains virtually
unaltered during delithiation and also the total electrode height decreases only
slightly (see also Fig. 7). We attribute this to SEI formation on the freshly formed
surfaces, because the Si particles are likely to fracture to some extent during charge
and discharge. Overall, it seems that the SEI can partially fill the cavities created
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

during shrinkage of the particles.

Nanoscale Accepted Manuscript

Fig. 4 AFM height images of a nano-Si electrode obtained in situ during the first (a-e)
lithiation and (e-h) delithiation cycle. Each image shows the same area of the top
surface.

To gain more information on the first lithiation cycle and the behavior of particles, the
topographical images were transformed into SEM-like images by use of a Monte
Carlo method (see ESI for additional details).58 The data in Fig. 5 show again a fixed
area of the surface at different potentials. Two regions can be clearly identified, in
which the electrode structure changes significantly. The first major change is
observed when a potential of about 0.6 V is reached. The particles remain at their
initial positions, but the surface roughness increases due to increased SEI formation.
Upon further charging (lithiation), the surface does not change much until a potential
of 0.2 V is reached. Especially the particles indicated by the dashed circle in Fig. 5
seem to increase in size and tend to form bigger agglomerates. At 0.01 V, they have
9
 Nanoscale Page 10 of 19
View Article Online
DOI: 10.1039/C6NR03575B

eventually fully merged into a large cluster. This behavior is observed in many other
regions as well, thereby indicating good electrical contact of the Si particles in the
electrode.
Collectively, the data in Figs. 4, 5 emphasize the importance of both a flexible binder
and a stable and flexible SEI to maintain the electrode integrity during cycling and to
keep the Li loss and electrolyte decomposition at a minimum.6,59,60 We note that these
results cannot be obtained with thin film electrodes and give direct support for our
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

experimental approach of studying practical Si-based electrodes.

Nanoscale Accepted Manuscript

Fig. 5 SEM-like images of a nano-Si electrode at different potentials. The dashed
circle indicates a surface region with significant structural changes during the first
lithiation cycle.

In situ AFM (2nd cycle)

The 2nd lithiation cycle was also investigated by means of in situ AFM. Interestingly,
the structure, including particle size, did not change significantly anymore after the
initial charge/discharge cycle (Fig. 6). This is likely because the surface is already
covered to a large extent by an SEI layer, so that the volume expansion cannot be
clearly seen, which is in agreement with findings on Si nanopillars by Becker et
al.46,61 Nevertheless, we noticed a slight increase of the total electrode height upon
lithiation (see also Figs. 7, S5†). However, size-change effects appear to be mainly
driven by SEI reactions in later cycles. Overall, these data indicate a stabilized and
somewhat flexible structure, which is capable of accommodating the volume changes
of the Si particles. We note that nanoparticles are known to better resist stresses.
10
 Page 11 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

This means they can withstand stress-induced cracking, unlike micrometer-sized Si

(shows faster capacity decay due to fracturing).57 The major reason for this is the
high ratio of surface to bulk atoms (surface energy).62 Moreover, McDowell et al.
have shown by in situ TEM that amorphous Si has a much more favorable fracture
behavior, with larger critical fracture diameter, than crystalline Si.42 Consequently,
pulverization effects are more likely to severely affect the performance of electrodes
containing “large” and crystalline Si. So overall, this implies that if the nanoparticles
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

used in this work break upon lithiation/delithiation, they will do so predominantly in

Nanoscale Accepted Manuscript

the initial cycle.

Fig. 6 AFM height images of a nano-Si electrode obtained in situ (a) before cycling
and in a (b, first cycle; d, 2nd cycle) “fully” lithiated and (c, first cycle) delithiated state.
Each image shows the same area of the top surface.

Electrode height
The height change during cycling was measured by monitoring the z-sensor position
of the cantilever. The total thickness of the nano-Si electrode (including current
collector) used for this kind of experiments was 34 µm. Fig. 7 shows the height in
arbitrary units versus the potential. As is seen, the height increase upon lithiation (first
cycle) is not linear in nature. The rate of thickness increase is highest at around 0.7 V
and below 0.25 V with respect to Li/Li+. These potentials correspond to the SEI
formation and onset of the alloying reaction, respectively. Furthermore, it is evident
that the most significant changes occur in the first lithiation cycle; the height change

11
 Nanoscale Page 12 of 19
View Article Online
DOI: 10.1039/C6NR03575B

during delithiation is relatively small (~2.5%). In the 2nd lithiation cycle, the thickness
increases by only 20% compared to the first one (i.e., by 1/5th of the height change
in the first lithiation) due to the reasons mentioned above. These findings are
confirmed by cross-sectional SEM images (Fig. S5†), showing that the out-of-plane
expansion – perpendicular to the plane of the current collector – is more than 100%
after the first cycle (from approx. 14 µm to 32 µm). In contrast, the thickness increase
in the subsequent cycle is only about 4 µm. Furthermore, elemental maps from SEM-
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

EDS (Fig. S6†) demonstrate the homogeneous distribution of C and Si across large

Nanoscale Accepted Manuscript

areas before and after cycling. This also helps explain the good cyclability of the
nano-Si electrodes.

Fig. 7 Changes in electrode height measured in the first lithiation cycle – indicated by
the arrow – and after the first delithiation (blue) and 2nd lithiation (orange).

In operando AFM (1st cycle)

For the above measurements, the potential was always kept constant while taking
AFM images. However, it is also possible to conduct similar studies under in
operando conditions. Fig. 8 presents a selection of height images obtained on a
nano-Si electrode in the first lithiation cycle. Most of them reveal the structural
evolution of the top surface during the alloying reaction (longest time period). Figs.
8a-d show the electrode in pristine state at OCV and in the SEI formation process, as
indicated by the small increase in particle size. The residual images represent early
(Figs. 8e-h) and later (Figs. 8i-l) stages of lithiation. From these it seems that Si in
the bulk of the electrode reacts first, or in other words, undergoes alloying with Li at
lower overpotential. This is likely due, in part, to the electrode porosity, which allows
good electrolyte penetration. The volume expansion of the subsurface particles leads
to structural distortions on the micrometer level and eventually to the generation of

12
 Page 13 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

cracks. However, apparently, such cracks can somewhat heal – at least during the
lithiation cycle – because the polymer binder used in electrodes for LIBs enables
some flexibility.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

Nanoscale Accepted Manuscript

Fig. 8 Top surface of a nano-Si electrode studied in operando by AFM during the first
lithiation cycle. (a-d) OCV and onset of SEI formation (3.2 V to ~0.6 V). (e-h) Early
and (i-l) later stages of lithiation.

Phase imaging
AFM phase images of the nano-Si electrodes were also taken as a function of cycle
number, because they provide additional information on the surface composition.
Phase imaging as such has been shown to be a powerful tool to detect material
differences, and thus should be suited to visualize the SEI. Fig. 9 shows phase
images collected before and after 1, 5 and 10 cycles. We note that pre-cycled
electrodes were used to allow comparable measurements. The corresponding AFM
height images are shown in Fig. S7†.
The surface of pristine electrodes appears to be uniform and smooth (Fig. 9a). After
the first cycle, the particles have expanded in volume and the surface roughness

13
 Nanoscale Page 14 of 19
View Article Online
DOI: 10.1039/C6NR03575B

increased significantly. A granular structure is clearly visible in Fig. 9b, which

indicates areas of different hardness. These probably contain the SEI. According to
Nie et al., the reductive decomposition of FEC is also accompanied by the formation
of LiF nanoparticles on the surface of the Si particles.63 With increasing cycle number,
the surface is increasingly covered by a mixture of organic and inorganic species
which, however, appear as one material with distinct hardness in the phase images.
This is in agreement with the cycling data, showing that the Coulombic efficiency
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

stabilizes after a few charge/discharge cycles. After 5 cycles (Fig. 9c), the granular

Nanoscale Accepted Manuscript

structure is still visible, but much less apparent than after the first one. And after 10
cycles (Fig. 9d), the surface appears to be smooth again, which indicates that the
particles are uniformly coated. These observations confirm that the SEI formation in
Si-based electrodes is in fact a dynamic process (i.e., the SEI forms over several
cycles).

Fig. 9 AFM phase images of different nano-Si electrodes (a) before and after (b) 1,
(c) 5 and (d) 10 cycles.

Nano-indentation
Lastly, the thickness of the SEI layer on surface particles was determined by in situ
nano-indentation measurements. These are known to be very sensitive to any kind of
changes, so that a large number of measurements was performed and then
averaged. In particular, we examined the surface after the first lithiation cycle. Fig. 10
shows representative force-distance curves obtained on two different regions of the

14
 Page 15 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

electrode. A difference in slope is generally a direct reflection of the difference in

material stiffness (Young's modulus). The tracing and retracing curves in Fig. 10a lie
perfectly on top of each other, whereas in Fig. 10a the curves show hysteretic
behavior and indicate that the AFM tip penetrated through a layer of around 45 nm
thickness. We believe that this layer of softer material is the SEI. The fact that the
surface is not yet fully covered after the initial lithiation cycle is in agreement with the
results from phase imaging.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

Nanoscale Accepted Manuscript

Fig. 10 Force-distance curves obtained on two different surface regions of a “fully”
lithiated nano-Si electrode (a) with and (b) without SEI layer. A diamond probe with
spring constant of 80 N/m was utilized, and the approaching and retracting speed
was 1 µm/s.

The growth of a patterned, multiphase SEI is certainly related to the volume changes
during cycling. These changes lead to the fact that SEI formation in Si and other
conversion-type electrodes proceeds in a dynamic and continuous manner.6 In a
recent paper, Tokranov et al. examined in detail the formation of the SEI on planar Si
islands. They showed that the SEI thickness stabilizes during the first cycle and the
properties strongly depend on the cycling conditions.49 Nie et al. reported
inhomogeneous deposition of decomposition products on the surface of Si
nanoparticles over the first few cycles, with LiF, LixSiOy and an insoluble polymeric
species as the primary SEI components when using an LiPF6/FEC electrolyte.63
However, we cannot rule out slight differences between electrodes cycled in
conventional coin cells and the ECAFM cell used here, which contained a much
larger volume of electrolyte. Nevertheless, we believe that our results closely
resemble the behavior of realistic electrodes in practical cells.

15
 Nanoscale Page 16 of 19
View Article Online
DOI: 10.1039/C6NR03575B

Taken collectively, the data presented demonstrate the potential of in situ and
operando AFM as a characterization technique in the field of energy storage and
conversion. The intriguing possibility of simultaneously visualizing surface processes,
measuring the stiffness/hardness and following height changes during cycling makes
it very powerful for studying battery materials.

Conclusions
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

This study reveals structural rearrangements in nano-Si electrodes containing

Nanoscale Accepted Manuscript

polymer binder and carbon black additive during operation. Using atomic force
microscopy, the most significant changes are observed in the first lithiation cycle at
potentials around 0.6 V and below 0.2 V vs. Li/Li+ due to increased solid-electrolyte
interphase formation and alloying of Si with Li, respectively. The latter reactions result
in the generation of surface cracks and pinholes which correlates with the
performance degradation. We show that electrode “breathing” is significantly reduced
in the 2nd cycle, and the polymer binder enables some flexibility. Phase imaging
reveals that the solid-electrolyte interphase formation occurs over several cycles, in
agreement with the Coulombic efficiency data. And nano-indentation indicates that
the deposited surface layer is already a few tens of nanometers thick after the initial
lithiation cycle.
Overall, atomic force microscopy is a powerful technique that is suitable for
investigating practical battery electrodes under both in situ and operando conditions.
The major limitation is the lack of chemical information, and thus it needs to be
combined with other state-of-the-art techniques to gain a complete picture of the
processes occurring in lithium-ion batteries during charge and discharge.

Acknowledgements

Katja Graf is acknowledged for performing the AFM measurements. We thank Artur
Schneider for graphical assistance. This study is part of projects being funded within
the BASF International Network for Batteries and Electrochemistry.

References

1 P. G. Bruce, B. Scrosati and J.-M. Tarascon, Angew. Chem. Int. Ed. Engl.,
2008, 47, 2930–2946.

16
 Page 17 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

2 J. B. Goodenough and Y. Kim, Chem. Mater., 2010, 22, 587–603.

3 M. Armand and J.-M. Tarascon, Nature, 2008, 451, 652–657.
4 B. Scrosati, J. Hassoun and Y.-K. Sun, Energy Environ. Sci., 2011, 4, 3287–
3295.
5 D. Aurbach, J. Power Sources, 2000, 89, 206–218.
6 W.-J. Zhang, J. Power Sources, 2011, 196, 13–24.
7 S.-Y. Chung, J. T. Bloking and Y.-M. Chiang, Nat. Mater., 2002, 1, 123–128.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

8 B. Dunn, H. Kamath and J.-M. Tarascon, Science, 2011, 334, 928–935.

9 A. Débart, J. Bao, G. Armstrong and P. G. Bruce, J. Power Sources, 2007, 174,
1177–1182.

Nanoscale Accepted Manuscript

10 X. Ji and L. F. Nazar, J. Mater. Chem., 2010, 20, 9821–9826.
11 H.-G. Jung, J. Hassoun, J.-B. Park, Y.-K. Sun and B. Scrosati, Nat. Chem.,
2012, 4, 579–585.
12 Y.-G. Guo, J.-S. Hu and L.-J. Wan, Adv. Mater., 2008, 20, 2878–2887.
13 V. Etacheri, O. Haik, Y. Goffer, G. A. Roberts, I. C. Stefan, R. Fasching and D.
Aurbach, Langmuir, 2012, 28, 965–976.
14 M. Winter and J. O. Besenhard, Electrochim. Acta, 1999, 45, 31–50.
15 M. Anji Reddy and M. Fichtner, J. Mater. Chem., 2011, 21, 17059–17062.
16 C. Erk, T. Brezesinski, H. Sommer, R. Schneider and J. Janek, ACS Appl.
Mater. Interfaces, 2013, 5, 7299–7307.
17 N. S. Hochgatterer, M. R. Schweiger, S. Koller, P. R. Raimann, T. Wöhrle, C.
Wurm and M. Winter, Electrochem. Solid-State Lett., 2008, 11, A76–A80.
18 M. A. Reddy, B. Breitung, V. S. K. Chakravadhanula, C. Wall, M. Engel, C.
Kübel, A. K. Powell, H. Hahn and M. Fichtner, Adv. Energy Mater., 2012, 5, 1–
6.
19 A. S. Aricò, P. Bruce, B. Scrosati, J.-M. Tarascon and W. van Schalkwijk, Nat.
Mater., 2005, 4, 366–377.
20 B. Hertzberg, A. Alexeev and G. Yushin, J. Am. Chem. Soc., 2010, 132, 8548–
8549.
21 I. A. Bobrikov, A. M. Balagurov, C.-W. Hu, C.-H. Lee, T.-Y. Chen, S. Deleg and
D. A. Balagurov, J. Power Sources, 2014, 258, 356–364.
22 S.-W. Kim, H.-W. Lee, P. Muralidharan, D.-H. Seo, W.-S. Yoon, D. K. Kim and
K. Kang, Nano Res., 2011, 4, 505–510.
23 M. Xu, N. Tsiouvaras, A. Garsuch, H. A. Gasteiger and B. L. Lucht, J. Phys.
Chem. C, 2014, 118, 7363–7368.
24 E. J. Jeon, Y. W. Shin, S. C. Nam, W. Il Cho and Y. S. Yoon, J. Electrochem.
Soc., 2001, 148, A318–A322.
25 I. Bezza, M. Kaus, R. Heinzmann, M. Yavuz, M. Knapp, S. Mangold, S. Doyle,
C. P. Grey, H. Ehrenberg, S. Indris and I. Saadoune, J. Phys. Chem. C, 2015,
119, 9016–9024.
26 E. Pollak, G. Salitra, V. Baranchugov and D. Aurbach, J. Phys. Chem. C, 2007,
111, 11437–11444.
17
 Nanoscale Page 18 of 19
View Article Online
DOI: 10.1039/C6NR03575B

27 B. B. Berkes, A. Jozwiuk, H. Sommer, T. Brezesinski and J. Janek,

Electrochem. Commun., 2015, 60, 64–69.
28 M. Holzapfel, A. Würsig, W. Scheifele, J. Vetter and P. Novák, J. Power
Sources, 2007, 174, 1156–1160.
29 L. Cui, R. Ruffo, C. K. Chan, H. Peng and Y. Cui, Nano Lett., 2009, 9, 491–495.
30 M. Gu, Y. Li, X. Li, S. Hu, X. Zhang, W. Xu, S. Thevuthasan, D. R. Baer, J.-G.
Zhang, J. Liu and C. Wang, ACS Nano, 2012, 6, 8439–8447.
31 T. D. Hatchard and J. R. Dahn, J. Electrochem. Soc., 2004, 151, A838–A842.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

32 J. Park, S. S. Park and Y. S. Won, Electrochim. Acta, 2013, 107, 467–472.

33 S. Misra, N. Liu, J. Nelson, S. S. Hong, Y. Cui and M. F. Toney, ACS Nano,

Nanoscale Accepted Manuscript

2012, 6, 5465–5473.
34 M. Kaus, I. Issac, R. Heinzmann, S. Doyle, S. Mangold, H. Hahn, V. Sai, K.
Chakravadhanula, C. Ku, H. Ehrenberg and S. Indris, J. Phys. Chem. C, 2014,
118, 17279–17290.
35 L. Martin, H. Martinez, M. Ulldemolins, B. Pecquenard and F. Le Cras, Solid
State Ionics, 2012, 215, 36–44.
36 A. Tokranov, B. W. Sheldon, C. Li, S. Minne and X. Xiao, ACS Appl. Mater.
Interfaces, 2014, 6, 6672–6686.
37 X. Liu, X. Deng, R. Liu, H. Yan, Y. Guo, D. Wang and L. Wan, ACS Appl. Mater.
Interfaces, 2014, 6, 20317–20323.
38 B. Key, R. Bhattacharyya, M. Morcrette, V. Sezne, J. Tarascon and C. P. Grey,
J. Am. Chem. Soc., 2009, 131, 9239–9249.
39 M. Yoshio, H. Wang, K. Fukuda, T. Umeno, N. Dimov and Z. Ogumi, J.
Electrochem. Soc., 2002, 149, A1598–A1603.
40 J. Graetz, C. C. Ahn, R. Yazami and B. Fultz, Electrochem. Solid-State Lett.,
2003, 6, A194–A197.
41 U. Kasavajjula, C. Wang and A. J. Appleby, J. Power Sources, 2007, 163,
1003–1039.
42 M. T. McDowell, S. W. Lee, J. T. Harris, B. A. Korgel, C. Wang, W. D. Nix and Y.
Cui, Nano Lett., 2013, 13, 758–764.
43 F. Wang, L. Wu, B. Key, X.-Q. Yang, C. P. Grey, Y. Zhu and J. Graetz, Adv.
Energy Mater., 2013, 3, 1324–1331.
44 K. Ogata, E. Salager, C. J. Kerr, A. E. Fraser, C. Ducati, A. J. Morris, S.
Hofmann and C. P. Grey, Nat. Commun., 2014, 5, 1–11.
45 I. T. Lucas, E. Pollak and R. Kostecki, Electrochem. Commun., 2009, 11, 2157–
2160.
46 C. R. Becker, K. E. Strawhecker, Q. P. Mcallister and C. A. Lundgren, ACS
Nano, 2013, 7, 9173–9182.
47 L. Y. Beaulieu, T. D. Hatchard, A. Bonakdarpour, M. D. Fleischauer and J. R.
Dahn, J. Electrochem. Soc., 2003, 150, A1457–A1464.
48 Q. P. McAllister, K. E. Strawhecker, C. R. Becker and C. A. Lundgren, J. Power
Sources, 2014, 257, 380–387.

18
 Page 19 of 19 Nanoscale
View Article Online
DOI: 10.1039/C6NR03575B

49 A. Tokranov, R. Kumar, C. Li, S. Minne, X. Xiao and B. W. Sheldon, Adv.

Energy Mater., 2016, 1502302.
50 U. S. Vogl, S. F. Lux, E. J. Crumlin, Z. Liu, L. Terborg, M. Winter and R.
Kostecki, J. Electrochem. Soc., 2015, 162, A603–A607.
51 M. A. McArthur, S. Trussler and J. R. Dahn, J. Electrochem. Soc., 2012, 159,
A198–A207.
52 S. Dalavi, P. Guduru and B. L. Lucht, J. Electrochem. Soc., 2012, 159, A642–
A646.
Published on 17 May 2016. Downloaded by University of Leeds on 19/05/2016 08:17:56.

53 A. L. Michan, M. Leskes and C. P. Grey, Chem. Mater., 2016, 28, 385–398.

54 M. N. Obrovac and L. J. Krause, J. Electrochem. Soc., 2007, 154, A103–A108.

Nanoscale Accepted Manuscript

55 Y.-S. Hu, R. Demir-Cakan, M.-M. Titirici, J.-O. Müller, R. Schlögl, M. Antonietti
and J. Maier, Angew. Chem. Int. Ed. Engl., 2008, 47, 1645–1649.
56 Y.-S. Hu, P. Adelhelm, B. Smarsly and J. Maier, ChemSusChem, 2010, 3, 231–
235.
57 X. Su, Q. Wu, J. Li, X. Xiao, A. Lott, W. Lu, B. W. Sheldon and J. Wu, Adv.
Energy Mater., 2014, 4, 1–23.
58 D. Nečas and P. Klapetek, Cent. Eur. J. Phys., 2012, 10, 181–188.
59 J. Zheng, H. Zheng, R. Wang, L. Ben, W. Lu, L. Chen, L. Chen and H. Li, Phys.
Chem. Chem. Phys., 2014, 16, 13229–13238.
60 G. M. Veith, M. Doucet, J. K. Baldwin, R. L. Sacci, T. M. Fears, Y. Wang and J.
F. Browning, J. Phys. Chem. C, 2015, 119, 20339–20349.
61 C. R. Becker, S. M. Prokes and C. T. Love, ACS Appl. Mater. Interfaces, 2016,
8, 530–537.
62 J. Li, A. K. Dozier, Y. Li, F. Yang and Y.-T. Cheng, J. Electrochem. Soc., 2011,
158, A689–A694.
63 M. Nie, D. P. Abraham, Y. Chen, A. Bose and B. L. Lucht, J. Phys. Chem. C,
2013, 117, 13403–13412.

Graphical abstract

AFM is a powerful tool suitable for in situ and operando studies of structural changes
and solid-electrolyte interphase formation in practical battery electrodes containing
polymer binder and carbon additive.

19