Copyright WILEY-VCH Verlag GmbH & Co. KGaA, 69469 Weinheim, Germany, 2018.

Supporting Information
for Adv. Energy Mater., DOI: 10.1002/aenm.201803121

Robust Pitch on Silicon Nanolayer–Embedded Graphite for

Suppressing Undesirable Volume Expansion
Seong-Hyeon Choi, Gyutae Nam, Sujong Chae, Donghyuk
Kim, Namhyung Kim, Won Sik Kim, Jiyoung Ma, Jaekyung
Sung, Seung Min Han, Minseong Ko,* Hyun-Wook Lee,* and
Jaephil Cho*
Supplementary Information

Robust pitch on silicon-nanolayer-embedded graphite for suppressing

undesirable volume expansion

Seong-Hyeon Choi1†, Gyutae Nam1†, Sujong Chae1†, Donghyuk Kim2, Namhyung Kim1,
Won Sik Kim2, Jiyoung Ma1, Jaekyung Sung1, Seung Min Han2, Minseong Ko3*, Hyun-
Wook Lee1*, and Jaephil Cho1*

1
Department of Energy Engineering, School of Energy and Chemical Engineering, Ulsan
National Institute of Science and Technology (UNIST), Ulsan 44919, South Korea
2
Department of Material Science and Engineering, Korea Advanced Institute of Science and
Technology (KAIST), Daejeon 34141, South Korea
3
Department of Metallurgical Engineering, Pukyung National University, Busan 48547,
South Korea
*Corresponding author: msko876@pknu.ac.kr, hyunwooklee@unist.ac.kr, jpcho@unist.ac.kr
† These authors equally contributed.

1
Preparation of the SG, SGCacetylene, SGCpitch, and SGCsucrose. The fabrication of SG and
SGCacetylene is described in our previous report.[1] To synthesize SGCpitch, 30 g of SG and 2.14
g of pitch (POSCO Chemtech) were well dispersed in 25 ml of tetrahydrofuran (THF, Junsei).
The mixture was stirred vigorously in a planetary centrifugal mixer until the THF was
completely evaporated. The as-obtained powder was placed into a tube furnace. Under the
inert atmospheric conditions, the powder was carbonized as following profiles: Heat to
300 °C at 2 °C min−1; hold at 300 °C for 3 h; heat to 900 °C at 4 °C min−1; hold at 900 °C for
2 h. In the case of SGCsucrose, 30 g of SG and 9.56 g of sucrose (Sigma-Aldrich) were first
dispersed in a mixture of distilled water and methanol. With vigorous stirring, the dispersed
solution was dried using a spray dryer (Mini Spray Dryer B-290, BUCHI Labortechnik) with
inlet and outlet temperature of 200 and 120 °C, respectively. The spray-dried powder was
collected and carbonized at 900 °C for 1 h under inert atmospheric conditions.

Preparation of the acetylene- and pitch-coated Si nanoparticles. To prepare the acetylene-

coated Si nanoparticles, 0.3 g of Si nanopowder (Alfa-aesar) was placed into the tube furnace.
Under inert atmospheric conditions, high-purity acetylene gas was supplied with flow rate of
1.5 L min−1 for 5 min at 900 °C. For the pitch-coated Si nanoparticles, 0.6 g of Si nanopower
and 0.05 g of pitch were dissolved in THF. The mixing method and heating profiles for
carbonization are same with those of SGCpitch.

Material characterization. The molecular structure for pitch was investigated using X-ray
photoelectron spectroscopy (K-alpha, Thermo Fisher), Thermogravimetric analysis (Q500,
TA), 13C-Nuclear magnetic resonance (VNMRS 600, Agilent), X-ray diffraction (D/Max2000,
Rigaku), Elemental analysis (Flash 2000, Thermo), Matrix-assisted laser
desorption/ionization time-of-flight mass spectroscopy (Ultraflex III, Bruker), and Electron
energy loss spectroscopy (Quantum 965, Gatan). A Mettler Toledo (DP 70, MBI KOREA)
was used to measure the softening point of pitch.

The structural characterization for SGCpitch was carried out using scanning electron
microscopy (Verios 460, FEI), Dual Beam Focused Ion Beam (Helios 450HP, FEI), Ion
milling system (Hitachi IM4000, Hitachi High-technologies), and High-resolution
transmission electron microscopy (JEM-2100F, FEI). In particular, the cross-sectioned
electrode could be obtained by using ion milling system to observe the change of particles’
morphologies after cycling concretely (Figure S16) or by cutting the cycled electrodes using

2
straw cutter to just compare the electrode thickness change (Figure S19). To prepare samples
using the FIB workstation, epoxy soaking, and carbon deposition were performed prior to
analysis to protect the intact morphology of the sample from the permeation of Ga ion into
the particles. EDS was utilized in combination with HR-TEM (Aztec, Oxford). The specific
surface areas of SG and SGCs were obtained with a surface area and porosity analyzer
(TriStar II, micromeritics) based on BET theory. To observe the changes in electrode
thickness upon cycling, cells were disassembled, and the electrodes were rinsed with excess
dimethyl carbonate (DMC) in an Ar-filled glove box. Electrochemical impedance
spectroscopy (VSP300, Bio-Logics) was used to obtain the Nyquist plots.

In situ TEM lithiation analysis. The in situ TEM lithiation experiment was carried out in an
FEI Tecnai G2 F20 X-Twin TEM at an acceleration voltage of 200 kV. Nanofactory
Instruments Dual-Probe STM-TEM in situ TEM holder was employed to investigate the
lithiation process of acetylene- and pitch-coated Si nanoparticles. Li metal was used as a Li
source as well as the counter electrode. When Li metal was transferred to the TEM, the
electrode was exposed to air for about 10 s to form a thin lithium oxide solid electrolyte layer.
Relative bias of −3 and −7 V for acetylene- and pitch-coated Si nanoparticles, respectively,
were applied between the two electrodes, causing Li+ ions to be transferred to the Si
nanoparticles.

In situ SEM indentation analysis. Acetylene- and pitch-coated Si nanoparticles were

dispersed in ethanol and ultrasonicated for 1 h to break up any agglomerated nanoparticles
and obtain a homogenous dispersion. Then, the dispersion was drop-casted onto a Si wafer
substrate (≈300 nm of native SiO2 layer). The compression axis was arranged perpendicular
to the substrate, whereas the imaging axis was 85° relative to the compression axis for in in
situ observation. The Hitachi SU5000 SEM was operated at 30.0 kV. A Hysitron PI 87 SEM
Picoindenter in displacement control mode was used to conduct in situ compression
experiments. A 1.0 μm diamond flat punch tip at a loading rate of 2 nm s−1 was used to
compress the nanoparticles. Compression was terminated when the particle was observed to
be completely compressed.

Electrochemical characterization. The working electrodes of SG, SGCpitch, SGCacetylene, and

SGCsucrose were fabricated by mixing active material, carbon black (Super P, Imerys),
carboxymethyl cellulose (CMC, Nippon paper), and styrene butadiene rubber (SBR, Zeon) in

3
distilled water at the mass ratio of 96:1:1.5:1.5. The homogeneous blended slurry was loaded
on 18 m of Cu current collector at a loading level of ≈7 mg cm−2 and dried at 80 °C for 30
min. Afterward, the electrodes were calendared in order to increase the electrode density to
1.6 g cc−1. The calendared electrodes were vacuum-dried at 110 °C for 8 h. 2032R coin-type
half-cells were fabricated in a glove box (Threeshine, H2O < 1 ppm, O2 < 1 ppm) with Li
metal (>99%, Honjo metal) as the counter electrode. Microporous polyethylene (15 m,
Celgard), 1.3 M LiPF6 in ethylene carbonate (EC)/ethyl methyl carbonate (EMC)/diethyl
carbonate (DEC) (=3/5/2, v/v/v) containing 0.2% LiBF4, 10% FEC, 0.5% VC, and 1% PS
(Panax Starlyte) were used as separator and liquid electrolyte, respectively. Upon fabricating
coin-cell, the input amounts of electrolyte were approximately 100 l. The electrochemical
characterization of half-cells was measured over the voltage range 0.005–1.5 V at the first
cycle at 0.1C, and 0.005–1.0 V for the cycling test at 0.5C. The electrochemical tests were
conducted with TOSCAT-3100 battery cycler (TOYO SYSTEM).

The slurry for the Ni-rich cathode was prepared with 96 wt% of active materials (90% of
NCM and 10% of NCA, Samsung SDI), 2 wt% of carbon black and 2 wt% of polyvinylidene
fluoride (PVDF, Solvay) in N-methyl-2-pyrrolidinone (NMP). The homogeneous slurry was
cast on the 15 m of Al current collector at a loading level of ≈18.5 mg cm−2 and dried at
120 °C for 2 h. Then, the electrode was calendared up to 3.0 g cc−1. The subsequent cell
assembly and other cell components were used same as those in above electrochemical
characterizations. The electrochemical characterization of half-cells was carried out over the
voltage range 2.5–4.3 V with TOSCAT-3100 battery cycler (TOYO SYSTEM). The rate was
0.1C at the first cycle and then 0.5C for the rest of cycles.

For the full-cell test, SGCpitch and SGCacetylene anodes and Ni-rich cathodes were utilized to
assemble pouch-type full-cells. The ratio of negative to positive electrode capacity (N/P ratio)
was fixed at 1.15. The pouch-type full-cells were assembled in a dry room with a humidity of
less than 1%. The separator and liquid electrolyte as those used above were used. The full-
cell tests were run at 2.5–4.2 V with a TOSCAT-3100 battery cycler (TOYO SYSTEM). The
formation cycle was run at 0.1C and the rest of cycles were run at 0.5C for 200 cycles.

4
Figure S1. SEM image of the pristine Si nanoparticles.

5
Figure S2. Intermediate time-lapse images of the (a) acetylene- and (b) pitch-coated Si

nanoparticles upon in situ TEM internal pressurization lithiation.

6
Figure S3. The molecular structure of (a) sucrose, (b) citric acid, and (c) PVP, respectively.

7
Figure S4. MALDI-TOF MS spectra of pristine (black) and carbonized pitch (red)

8
Figure S5. Morphology comparison with pristine graphite, SG, and SGCpitch. The low and

high magnification SEM images of (a,d) pristine graphite, (b,e) SG, and (c,f) SGCpitch,

respectively, which reveal their different surface texture.

9
Figure S6. Statistical distribution of SG and SGCpitch particle size.

10
Figure S7. Magnified STEM images of SGCpitch with EDS mapping for Si and C. Pitch can

penetrate into the porous SG and cover the inner Si nanolayer.

11
Figure S8. The electrochemical performances of the SGCpitch which the coating amounts of
pitch are increased. a) Voltage profiles of SGC10wt% pitch at the formation cycle. b) Discharge
capacities and cycling CEs of SGC10wt% pitch for 50 cycles. c) Rate capabilities under
increasing current densities from 0.2C to 5C. d) The changes in electrode thickness for
anodes in their lithiated states over 50 cycles. The yellow box indicates the commercially
acceptable limit for electrode swelling.

Figure S8 presents the electrochemical results regarding increased amounts of pitch on

SGCpitch over twice (denoted as SGC10wt% pitch). As shown in Figure S8, SGC10wt% pitch exhibit
the initial specific capacity of 488 mAh g1 with initial CE of 87.4%, compared to SGCpitch
showing 523 mAh g1 and 90.9%. In the cycling test under discharging rate of 0.5C, the
capacity retention of SGC10wt% pitch is 97.9% after 50 cycles. While SGC10wt% pitch exhibited
slightly lower rate capability in the high current density, the level of the electrode swelling of
SGC10wt% pitch is comparable to that of SGCpitch. In particular, both SGCpitch and SGC10wt% pitch
outperform other carbon coated SGCs in rate capability and electrode swelling. In conclusion,
the coating amount of pitch for SGCpitch influences in the initial specific capacity and the rate
capability. However, other electrochemical properties such as cycling stability and the
12
swelling behavior of electrodes are confirmed similar.

13
Figure S9. (a) The specific surface area and (b) N2 sorption isotherms of SG and three

different SGCs based on BET theory.

14
Figure S10. Extended cycling performance of SGCpitch with cycling Coulombic efficiencies
with error bars.

15
Figure S11. Various voltage profiles of (a) SGCpitch, (b) SGCacetylene, and (c) SGCsucrose at the

10th, 30th, and 50th cycle.

16
Figure S12. Rate capabilities of three SGCs under increasing current densities from 0.2C to

5C with error bars.

17
Figure S13. Nyquist plots for SGCpitch and SGCacetylene after 1st cycle.

18
Figure S14. Electrochemical characterization of Ni-rich cathode. (a) Voltage profiles of Ni-

rich cathode at the 1st cycle. (b) Reversible discharge capacity of Ni-rich cathode for 100

cycles with error bars.

19
Figure S15. Voltage profiles of (a) SGCpitch and (b) SGCacetylene in full-cell with Ni-rich

cathode plotted as number of cycles (10th, 50th, 100th, and 200th cycle).

20
Figure S16. Cross-sectional SEM images of pristine electrodes for (a) SGCpitch, (b)

SGCacetylene and (c) SGCsucrose with inner magnified images (inset).

21
Figure S17. Cross-sectional images of SGCpitch, SGCacetylene, and SGCsucrose of EDS mapping
at C (b,f,j), Si (c,g,k), and F (d,h,l) after 50 cycles.

22
Figure S18. XPS spectra of SGCpitch, SGCacetylene and SGCsucrose after 50 cycles for (a-c) C

and (d-f) F. All XPS spectra were calibrated by C-C peak at the binding energy of 284.5 eV.

The least amounts of F in SGCpitch indicate the smallest side reaction with electrolyte.

23
Figure S19. Series of SEM images of three different SGCs before and after 50 cycles. The

cross-sectional SEM images of (a-c) SGCpitch, (d-f) SGCacetylene, and (g-i) SGCsucrose electrodes

before cycling, 50th lithiated state, and 50th delithiated state, respectively.

24
 C wt% H wt% H/C ratio

Pristine pitch 94.87 ± 0.55 5.36 ± 0.05 0.67

Carbonized pitch at
94.70 ± 0.53 5.24 ± 0.04 0.66
400 °C

Carbonized pitch at
97.16 ± 0.44 2.91 ± 0.02 0.36
600 °C

Carbonized pitch at
98.75 ± 0.64 1.01 ± 0.02 0.12
800 °C

Carbonized pitch at
99.34 ± 0.55 0.56 ± 0.02 0.07
900 °C
Table S1. Chemical composition of pristine and carbonized pitch. The atomic composition of
C and H, and H/C ratio for pristine and carbonized pitch at 400, 600, 800, and 900 °C for 2 h.

25
 C wt% H wt% O wt% N wt%

94.87 ± 5.36 ± 1.45 ±

Pitch 0.00
0.55 0.05 0.19

42.67 ± 6.41 ± 48.89 ±

Sucrose 0.00
0.49 0.01 0.50

37.94 ± 4.18 ± 53.88 ±

Citric acid 0.00
0.19 0.01 0.30

60.41 ± 8.08 ± 20.46 ± 11.56 ±

PVP
0.53 0.11 0.66 0.09
Table S2. Chemical composition of pitch and other carbonaceous materials. Elemental
analysis results for atomic composition of pitch, sucrose, citric acid, and PVP.

26
 Average CEs for 50 Discharge capacities at
Figure 4b 1st cycle CE (%)
cycles (%) 50th cycles
SG 92.4 ± 0.08 99.10 ± 0.07 469.7 ± 11.0
SGCpitch 90.9 ± 0.22 99.57 ± 0.05 506.8 ± 0.60

SGCacetylene 91.4 ± 0.30 99.38 ± 0.06 505.6 ± 0.36

SGCsucrose 86.7 ± 0.51 98.79 ± 0.16 472.4 ± 4.4
Table S3. The summary of electrochemical performances of anodes in the half-cell
configuration.

27
 Cell dimension
Cathode (Width*Length) 1.6 cm*2.2 cm
Anode (Width*Length) 1.8 cm*2.4 cm
Material selection
Cathode Anode
Specific
capacity:
Specific
523 mAh
capacity: SGCpitch
g1
183.2 mAh g1
Initial CE:
Ni-rich 90.9%
Active material
cathode Specific
capacity:
523 mAh
Initial CE: 93.4% SGCacetylene
g1
Initial CE:
91.4%
Binder PVDF SBR and CMC
Conductive agent Super P Super P
Electrode engineering
Cathode Anode
Electrode composition
(Active
96:2:2 96:3:1
material:Binder:Conductive
agent)
Loading level (mg cm2) 18.5 7
Areal charge capacity SGCpitch 3.855
3.365
(mAh cm2) SGCacetylene 3.845
Areal discharge capacity SGCpitch 3.505
3.196
(mAh cm2) SGCacetylene 3.515
N/P ratio 1.19
Electrode density
3 1.6
(g cc1)
Table S4. Detailed electrochemical cell designs for Ni-rich cathode and SGCs

28
 Average CEs for Discharge capacities at
1st cycle CE (%)
200 cycles (%) 200th cycle
SGCpitch 83.6 ± 0.05 99.8 ± 0.02 8.30 ± 0.03
SGCacetylene 86.1 ± 0.05 99.4 ± 0.02 6.74 ± 0.04
Table S5. The summary of electrochemical performances of the full-cells.

29
 SGCpitch SGCacetylene SGCsucrose
C 91.0 77.0 81.5
Si 6.9 7.6 4.2
Ga 1.5 3.6 1.1
O - 6.9 -
F - 4.3 6.4
P - - 0.6
Other atoms (e.g. Mo, Cr) 0.6 0.6 6.2
Total 100 (wt%) 100 (wt%) 100 (wt%)
Table S6. The atomic contents of the anodes derived from the EDS mapping after 50 cycles.

30
Supplementary Videos

Video S1. Lithiation of acetylene-coated Si nanoparticles. Note that the expansion and severe
pulverization could be observed.

Video S2. Lithiation of pitch-coated Si nanoparticles. During lithiation, particles are

expanded. Nevertheless, pitch could withstand the internal pressure and Si nanoparticles
retain original morphologies.

Video S3. External load testing of acetylene-coated Si nanoparticles via the flat punch
diamond tip. The indentation was stopped when particles are completely compressed.

Video S4. External load testing of pitch-coated Si nanoparticles via the flat punch diamond
tip. Indentation was equally processed with that of acetylene-coated Si nanoparticles.

Video S5. The measurement of softening point of pitch through Mettler Toledo method. Pitch
starts to convert into viscous liquid at about 240 °C.

Reference

[1] M. Ko, S. Chae, J. Ma, N. Kim, H.-W. Lee, Y. Cui, J. Cho, Nat. Energy 2016, 1,
16113.
31