High performance stoichiometric Cu2S cathode on carbon fiber current collector for lithium batteries

Article abstract of DOI:10.1016/j.electacta.2016.12.058

In this study, stoichiometric Cu₂S with monoclinic crystal structure with space group P2₁/c was prepared via spray pyrolysis with followed heat treatment. Its electrochemical properties were investigated after coating on a carbon fib

Full text of DOI:10.1016/j.electacta.2016.12.058

Accepted Manuscript
Title: High performance stoichiometric Cu S cathode on
2
carbon fiber current collector for lithium batteries
Author: Gulnur Kalimuldina Izumi Taniguchi
PII:
S0013-4686(16)32604-4
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.electacta.2016.12.058
EA 28530
To appear in:
Electrochimica Acta
Received date:
Revised date:
Accepted date:
14-10-2016
7-12-2016
9-12-2016
Please cite this article as: Gulnur Kalimuldina, Izumi Taniguchi, High performance
stoichiometric Cu2S cathode on carbon fiber current collector for lithium batteries,
Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2016.12.058
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

High performance stoichiometric Cu S cathode on carbon fiber
2
current collector for lithium batteries
a
b,
Gulnur Kalimuldina , Izumi Taniguchi *
a
Department of Chemical Engineering, Tokyo Institute of Technology, Tokyo 152-8552, Japan
b
Department of Chemical Science and Engineering, Tokyo Institute of Technology, Tokyo
1
52-8552, Japan
Abstract
In this study, stoichiometric Cu
group P2 /c was prepared via spray pyrolysis with followed heat treatment. Its
electrochemical properties were investigated after coating on a carbon fiber paper
CFP) current collector and cycling at various cut-off voltages. The reversible stable
2
S with monoclinic crystal structure with space
1
(
−
1
capacity of about 330 mAh g (98 % of its theoretical capacity) exhibited by the cell
−1
for 20 cycles at 0.1 C and slightly dropped to 311 mAh g after 50 cycles. At higher
−1
1
C rate, a discharge capacity of 280 mAh g was retained for 100 cycles with
approximately 100 % Coulombic efficiency between 1.0-3.0 V. While the gradual
capacity decrease is observed with the increase of the discharge cut-off voltage to
st
avoid solid electrolyte interface (SEI) formation during the 1 discharge. The cells
−1
th
showed capacities of 307 and 280 mAh g after the 5 cycle at 0.1 C when cycled
between 1.1-3.0 V and 1.2-3.0 V, respectively, due to the lowered lithium ion
diffusion. Ex situ XRD analysis of the recovered electrodes after cycling with near
theoretical capacity allowed discussing the actual reaction mechanism of
+
stoichiometric Cu
2
S with Li .
Keywords: copper sulfide, carbon fiber paper, lithium-ion battery, cathode,
spray pyrolysis
_
________________________
*
Corresponding author.
Tel. & Fax: +81-3-5734-2155
E-mail: taniguchi.i.aa@m.titech.ac.jp
1

1
. Introduction
The increased demand for the sustainable energy is driving an enormous impact
on the development of advanced storage systems by exploring new types of battery
chemistries and configurations [1]. Traditional cathode materials based on the lithiated
transition-metal oxide and phosphate are unable to satisfy the need for the
combination of high energy density, safety and reduced cost required for EVs and
smart grid [2,3]. As an alternative cathode material Cu
2
S is gaining an interest due to
−1
−1
its high capacity of 337 mAh g and energy density of 553 Wh kg . However, it has
the hindrance of partial lithium polysulfide dissolution in organic electrolyte during
cycling, which causes rapid capacity decay, loss of active material and poor cycle life
2
[4,5]. So far, several Cu S preparation methods have been reported [6-10] and a few
researchers [4,6] highlighted the effect of Al and Cu foil current collectors on the
electrochemical performance of copper sulfides. In our previous work, the details of
the failure mechanism of stoichiometric Cu
due to the corrosion during cycling and its improvement on Cu foil were reported [11].
However, up to now the electrochemical properties of stoichiometric Cu S coated on
2
S coated on an Al foil current collector
2
carbon fiber paper (CFP) have not been investigated. In this work, CFP used as a
+
current collector to increase the electron flow and Li ion diffusion, to facilitate active
electrode mass utilization and create a preferable condition for the investigation of the
2
actual Cu S reaction mechanism with lithium. Moreover, carbon fiber will avoid
corrosion problems tend to occur on Al and Cu current collectors.
Carbon materials are well known for their applicability for the improvement of
electronic conductivity of cathode materials [12-14]. Carbon fiber is well-connected,
2

and its conductive structure and appropriate pore channels create an effective electron
path as well as an effective condition for the electrolyte to access the active material
2
[15]. Here we report simple preparation method of stoichiometric Cu S with
monoclinic crystal structure by spray pyrolysis (SP) with followed heat treatment and
that stable high capacity with a long cycle life can be achieved using CFP as a current
collector without an involvement of the complex synthesis or the surface
modifications.
Copper sulfide has varying stoichiometries, which significantly affect its
structural and electrochemical properties [6,11]. Thus, we verified chemical
2
composition of monoclinic Cu S by X-ray diffraction (XRD) and inductively coupled
plasma-optical emission spectroscopy (ICP-OES) analyses before electrochemical
testing. In addition, this study will provide insight into the fully charged/discharged
stoichiometric Cu
2
S electrode’s behavior coated on CFP by ex situ XRD analysis after
cycling at various cut-off voltage windows (1.0, 1.1, 1.2 to 3.0 V).
2
. Experimental
2
2
.1. Preparation of stoichiometric Cu S
Stoichiometric Cu
The precursor solution was prepared from copper nitrate trihydrate (Cu(NO
and thiourea (CS(NH ) dissolved in distilled water in the molar ratio of 1:1 at 0.05
2
S was prepared using SP method with further heat treatment.
3
)
2
∙ 3H O)
2
2 2
)
M. All chemicals were purchased from Wako Pure Chemical Industries Ltd. The
schematic diagram of the SP facility is described elsewhere [16]. The atomized
precursor solution at a frequency of 1.7 MHz by ultrasonic nebulizer was carried
3

inside the vertical reactor by N
2
+ 3% H
fixed at 2 L min and reactor temperature was 400 °C. The as-prepared sample was
heat treated at the temperature of 460 °C in N +3% H atmosphere for 2 h to obtain
stoichiometric Cu S [11].
2
gas atmosphere. The flow rate of the gas was
−1
2
2
2
2
.2. Material characterization
2
The crystalline phase of stoichiometric Cu S was characterized by powder
X-ray diffraction (Ultima IV with D/tex Ultra, Rigaku) with Cu-Kα radiation. The
lattice parameters were determined by Rietveld refinement of the XRD data using
PDXL (Rigaku, Ver. 1.8.1.0) software. To verify Cu/S molar ratio in the Cu
ICP-OES (Shimadzu, ICPS-7510) was used and the morphology of as-prepared Cu
2
S
S
2
was examined by field-emission scanning electron microscopy (FE-SEM, Hitachi,
S4500).
2
.3. Cell assembly and electrochemical measurements
Electrochemical measurements were performed with coin-type CR2032 cells
2
with stoichiometric Cu S cathodes and lithium-metal foils. 1 M lithium
bis(trifluoromethane sulfonamide) (LiTFSI) in 1,3-dioxolane (DOL) and
dimethoxyethane (DME) with a volume ratio of 1:1 (Tomiyama Pure Chemical Co.,
Ltd.) was used as an electrolyte. The electrode consisting of 80:10:10 wt. % of Cu
acetylene black (AB) and polyvinylidene fluoride (PVDF) was dispersed in
-methyl-2-pyrrolidinone (NMP). Then the as-made slurry was coated onto CFP
2
S,
1
(
Toray Paper TGP-H-030 5% Wet Proofing) by using doctor blade technique and
4

dried at 60 °C for 4 h in a vacuum oven. Finally, circular disks were punched, and the
2
mass loadings of Cu
2
S were 1.5-2.0 mg per scrapped electrode square (100 mm ). All
cells were fabricated inside an Ar filled glove box (99.9995 % purity).
The cells were galvanostatically cycled at the range of 1.0-3.0 V, 1.1-3.0 V, and
+
1
.2-3.0 V versus Li/Li at 0.1 to 5 C on multichannel battery testers (Hokuto Denko,
HJ1010mSM8A), respectively. The C-rates were based on the theoretical capacity of
−1
Cu
2
S (337 mAh g ).
Cyclic voltammetry (CV) measurements were performed in the potential
+
−1
windows of 1.0, 1.1, 1.2 and 3.0 V (vs. Li/Li ) at a scanning rate of 0.05 mV s .
Electrochemical impedance spectroscopy (EIS) was carried out using a Solartron
1
255B frequency response analyzer connected to a Solartron SI 1287 electrochemical
interface. The amplitude of the ac signal was 10 mV in the frequency range from 100
kHz to 0.1 Hz. The electrochemical measurements were performed at room
temperature.
For the ex situ XRD studies, the recovered electrodes after cycling were washed
with DOL, dried inside an Ar gas atmosphere glove box at 60 °C for 2 h and protected
®
from the air atmosphere by a Kapton foil.
3
. Results and discussion
2
The XRD pattern of prepared stoichiometric Cu S is shown in Fig. 1. The
obtained XRD peaks are fitting with ICDD database pattern (00-033-0490) for the
monoclinic Cu S crystal structure with space group P2 /c. No peaks arising from
impurities can be detected. The lattice constants of a= 15.237 Å, b= 11.884 Å, and c=
2
1
1
3.492 Å obtained by Rietveld refinement are matching with the values reported in
5

the literature [17]. This sample was analyzed by ICP-OES, and it revealed that the
Cu/S molar ratio equals to two, which confirms stoichiometric ratio. The morphology
of spherical particles less than 1 µm in size demonstrated in FE-SEM image within
Fig. 1.
2
Electrochemical properties of stoichiometric Cu S coated on a CFP current
collector at cut-off voltage window of 1.0-3.0 V were investigated by galvanostatic
discharge-charge measurements at a current rate of 0.1 C as depicted in Fig. 2a. The
first discharge exhibited two plateaus at 1.70 and 1.10 V, while the first charge process
showed two potential plateaus, which are short and prolonged at around 1.85 and 2.25
th
V, respectively. After the 5 cycle, only one pair of discharge and charge plateaus
were observed at 1.73 and 1.85 V, and they are consistent with the reported potential
st
2
plateaus for Cu S [4]. Whereas the discharge plateau at around 1.10 V during the 1
cycle, which is not characteristic for copper sulfides, can be associated with the
electrolyte decomposition, thus causing the formation of a passivation film or a solid
electrolyte interface (SEI) known to occur on carbon materials [8]. The initial
−1
discharge and charge capacities of Cu
2
S on CFP are 532 and 379 mAh g ,
th
−1
respectively. On the following 5 cycle the cell exhibited value of 330 mAh g (98%
of its theoretical capacity) with approximately 100% Coulombic efficiency.
A CV profile in Fig. 2b shows lithiation and delithiation processes at a scan rate
−1
st
of 0.05 mV s . The 1 cycle cathodic scan defines two reduction peaks at 1.63 V
th
coming from lithium insertion and at 1.05 V from SEI, respectively. After the 5 cycle,
mainly high intensity cathodic peak at about 1.63 V can be observed. Whereas, on the
st
anodic scan at the 1 cycle, two distinct peaks less intense at 1.87 V and sharp at 2.30
th
V changed after the 5 cycle to the main peak at about 1.87 V and broad at 2.10 V.
The results correspond to the charge-discharge voltage profiles in Fig. 2a.
6

Ex situ XRD analysis of fully discharged and charged electrodes were
investigated after disassembling cells to understand actual reaction mechanism of
st
th
Cu
are shown in Figs. 2c and 2d. After the first discharge, we can confirm the formation
of Cu and Li S peaks [18,19], where Li S appeared on the overlap with carbon peaks
2
S. The XRD peaks after discharge and charge processes of the 1 and 5 cycles
2
2
by slightly broadening of CFP peak from its original position at 27 °. The high
th
repetitiveness of peaks can be observed after the completed 5 discharge indicating
reversibility of reactions.
st
As for the charge process in Fig. 2d, after the 1 cycle, we can clearly detect
crystal structure change from monoclinic Cu
2
S with space group P2
1
/c to tetragonal
Cu1.96S with space group P4 2 2 [11]. That might be the reason for the flattening of
3 1
st
the 1 oxidation plateau from 2.25 to 1.85 V shown in Fig. 2a. The formed less
st
crystalized Cu1.96S during the 1 charge results in the higher voltage plateau at 2.25 V
[20]. Furthermore, along with newly formed Cu1.96S peaks, we can also detect Cu
st
th
metal spectra in XRD patterns after the 1 and 5 cycles, and that confirms the
oxidation reaction is not reversible to its initial phase of monoclinic Cu
2
S. The formed
th
tetragonal Cu1.96S peaks appeared with higher intensity on the following 5 cycle,
where the plateau flattened to 1.85 V implies that different product with less energy
barrier than Cu
2
S is formed [20]. The obtained results suggest that the occurred
st
capacity fading during the initial five cycles comes from the irreversibility of the 1
cycle reaction due to the formation of the tetragonal Cu1.96S as depicted in XRD
+
patterns. The reaction mechanism of Cu
2
S with Li between 1.0 and 3.0 V cut-off
st
range can be expressed for the 1 discharge process, when the lithium ions intercalate
between active sites of sulfur and trigger movement of the highly-mobile copper
cations, as follows,
7

+
−
Cu
2
S + 2Li + 2e → Li
2
S + 2Cu
(1)
and for the following cycles as,
+
−
Li
2
S + 2Cu ⇆ Cu1.96S + 0.04Cu + 2Li + 2e
(2)
which are reported correspond to the displacement reactions [4,11,18]. The actual
reaction mechanism for stoichiometric Cu S on CFP defined through reaching near
2
theoretical capacity and consequently efficient consumption of active mass. However,
th
in the fully discharged 5 cycle XRD pattern series of fading Cu1.96S reflections were
detected, which might relate to a minor amount of unutilized discharge-charge
2
products of Li S and Cu1.96S during the discharge-charge processes as capacity was
around 98 % of its theoretical value.
In order to gradually decrease SEI formation causing irreversible capacity
st
during the 1 cycle and study reaction mechanism at different cut-off voltage
windows the cells were cycled between 1.1 and 3.0 V as demonstrated in Fig. 3a. The
st
−1
1
1
discharge capacity reached 421 mAh g with exhibiting two plateau regions at
.70 and 1.10 V as in 1.0-3.0 V cut-off range, and the plateau at 1.10 V occurred only
st
at the 1 discharge. While the charge curve had two plateaus at 1.85 and 2.25 V with
−1
th
the capacity of 332 mAh g , which decreased after the 5 cycle to 1.85 V with
−
1
th
followed gradual increase to 2.1 V and reached 307 mAh g . The difference of the 5
charge curve from 1.0-3.0 V might be due to the slower phase transformation from
2
monoclinic Cu S to tetragonal Cu1.96S than that of at lower discharge cut-off voltage
+
since Li diffusion was lowered by reduced depth of discharge. The consistent profile
of CV measurements is displayed in Fig. 3b.
8

Ex situ XRD analysis of fully discharged and charged electrodes in Figs. 3c and
st
th
3
d after the 1 and 5 cycles at 1.1-3.0 V did not show a significant difference from
the 1.0-3.0 V cut-off range. The formation of Cu and Li
2
S after discharge was
st
th
th
demonstrated after the 1 and 5 cycles in Fig. 3b. After the 5 charge in Fig. 3d it
st
can be seen that the intensity of Cu1.96S peaks is much sharper than in the 1 cycle.
st
This might result in the 1 charge potential plateau flattening from 2.25 V to 1.85 and
2
.1 V as well as larger capacity fading compared with that of at 1.0-3.0 V.
Charge-discharge profile of stoichiometric Cu S at further increased discharge
cut-off voltage of 1.2-3.0 V is shown in Fig. 4a. Only one pair of the plateaus is
2
st
th
shown after the 1 and 5 cycles for discharge at 1.70 V and charge at 2.25 V, which
th
is flattened after the 5 cycle to 1.85 V similar to other cut-off voltage ranges.
Through increasing the discharge cut-off voltage to 1.2 V the formation of SEI plateau
st
−1
at 1.10 V disappeared and the 1 discharge capacity exhibited a value of 338 mAh g .
th
After the 5 cycle, we can see the deterioration of the discharge capacity to 280 mAh
−
1
g .
The demonstrated CV results in Fig. 4b revealed similar cathodic and anodic
scan peaks as in the charge-discharge potential profiles in Fig. 4a. The intensities of
th
the reduction and oxidation peaks were higher after the 5 cycle at 1.2-3.0 V than
those of at 1.0-3.0 V and 1.1-3.0 V in Figs. 2b and 3b. Furthermore, we can detect a
st
th
small shift of the cathodic peak from 1.63 V at the 1 cycle to 1.65 V after the 5
cycle. The observed differences from the lower discharge cut-off voltages could be
linked to the slight changes in the reaction products during discharge and charge
processes.
st
th
XRD patterns of electrodes after the 1 and 5 discharge-charge are shown in
st
Figs. 4c and 4d. After the 1 discharge, we can observe the formation of Cu and Li
2
S
9

peaks along with small but clear Cu1.96S peaks at 31.3, 32.6 and 39.1 ° in Fig. 4c.
th
Particularly, after the 5 cycle Cu1.96S peaks are sharpening, while Li
2
S peak at 27.0 °
is shifted to 27.2 ° closer to the another Cu1.96S peak. That might relate to the insertion
+
of Li into Cu
2
S structure and followed formation of Li2-xCu
x
S [11,21,22]. The
+
increasing discharge cut-off voltage to 1.2 V did not drive enough Li ions and
electrons to trigger movement of Cu from Cu
Li
2
S structure to form separated Cu and
th
2
S. The increase of intensity of Cu1.96S peaks after the 5 discharge indicates that at
the cut-off voltage of 1.2-3.0 V the electrode’s active mass gradually decrease its
utilization.
st
On the 1 charge process, the phase transformation to tetragonal Cu1.96S also
was observed, but the difference from the lower cut-off voltage ranges (1.0, 1.1-3.0 V)
is that faded Li
2
S and some Cu
2
S (Li2-xCu
x
S) peaks can be additionally detected at
st
1
.2-3.0 V voltage window. This might suggest that after the 1 charge phase
transformation to tetragonal Cu1.96S happened slower than at 1.0-3.0 V and 1.1-3.0 V.
st
Therefore, the
capacity of Cu
1 discharge and charge capacities were about theoretical
−1
th
2
S, which drastically dropped to 280 mAh g after the 5 cycle.
th
Consequently, after the 5 charge all peaks are matching with tetragonal Cu1.96S. The
capacity fading during the initial five cycles might relate to the intensity of the formed
products of Cu1.96S after charge. The capacity fading was experienced to be less at
1
.0-3.0 V in contrast with 1.1-3.0 and 1.2-3.0 V, where the intensity of formed Cu1.96S
st
th
after the 1 and 5 cycles were higher and sharper as revealed in Figs. 3d and 4d.
The cyclability of the cell at different cut-off voltage windows demonstrated in
Fig. 5a for 50 cycles at 0.1 C rate. Cu
2
S coated on CFP exhibited high initial
−1
−1
discharge capacity of 532 mAh g , which dropped to 330 mAh g (98% of its
theoretical capacity) after the third cycle and maintained stable for 20 cycles around
10

this value with about 100 % Coulombic efficiency between 1.0 and 3.0 V. Further, at
th
−1
the end of the 50 cycle, capacity slightly decreased to 311 mAh g . On the other
hand, with higher discharge cut-off voltages cells exhibited less charge/discharge
−1
capacities. At cut-off voltage range of 1.1-3.0 V discharge capacity was 307 mAh g
−1
th
th
and at 1.2-3.0 V 280 mAh g after the 5 cycle. To the end of the 50 cycle, cells
−1
showed 284 and 266 mAh g for 1.1-3.0 and 1.2-3.0 V, respectively. Coulombic
efficiency decreased as well from 100 % at 1.0-3.0 V to 98 % at 1.2-3.0 V, which
could be related to the increased internal resistance, since with the increase of
discharge voltage energy efficiency delivered by the cells decreased as well. Moreover,
2
due to the phase change from monoclinic Cu S to tetragonal Cu1.96S the discharge
capacity stabilized only after the third cycle for cut-off range 1.0-3.0 V, while it took 5
cycles for 1.1-3.0 V and 1.2-3.0 V. This tendency is consistent with the ex situ XRD
results suggesting the slower phase change at higher discharge cut-off voltages.
The slight difference in the discharge capacities can be highlighted when cells
were cycled at 1.1-3.0 V and 1.2-3.0 V in Fig. 5a, which might imply that irreversible
st
capacities from SEI formed during the 1 discharge did not significantly affect the
2
electrochemical performance of stoichiometric Cu S. From the results, we observed
that CFP facilitates Li ions transfer into the electrode, which significantly enhanced
utilization of the electrode’s active mass and near the theoretical capacity has been
exhibited at 1.0-3.0 V cut-off voltage without significant capacity deterioration for 50
cycles. Stoichiometric Cu
and Cu foils [11] as corrosion of current collector was omitted and exhibited
capacities were higher than most of the previously reported results on Cu S [4,6-9,20].
To evaluate further electrode stability, we cycled cells at various current rates
and Fig. 5b shows the cycling performance of Cu S on CFP at 1 C for 100 cycles. The
2
S coated on CFP showed superior results than that of on Al
2
2
11

−1
cell at 1.0-3.0 V delivered an initial discharge capacity of 415 mAh g and after the
rd
−1
3
cycle capacity decreased to 280 mAh g and retained approximately at this value
for 100 cycles with 100 % Coulombic efficiency. The discharge capacity of the cell
decreased slightly with increasing discharge cut-off voltage. When the cut-off
voltages were set at 1.1-3.0 and 1.2-3.0 V the discharge capacities reached an
−1
th
approximately similar value of 250 mAh g and slightly increased after the 20 cycle
−1
to 260 mAh g in the case of 1.2-3.0 V. This might happen due to the increased
utilization of active material over a long cycling time [11]. To confirm it we carried
st
th
out ex situ XRD analysis of cycled electrodes after the 1 and 50 discharge at 1 C as
th
shown in Figs. 5c and 5d. After the 50 discharge, Cu1.96S peaks formed along with
2
Cu and Li S almost disappeared at the cut-off voltage of 1.2-3.0 V compared with that
st
of at the 1 discharge. The discharge reaction became close to the completed reaction
st
of Cu1.96S with Li. While small Cu1.96S peaks had similar intensity after the 1 and
th
5
0 discharge at 1.0-3.0 and 1.1-3.0 V, which is consistent with the cycling stability of
the cells.
Capacity retention of cells cycled at 1 C rate for 100 cycles was compared in
Fig. 6. As demonstrated stoichiometric Cu S showed excellent capacity retention at
various cut-off voltage ranges with the highest value exhibited when it was cycled
2
between 1.0-3.0 V, approximately 90 %. The obvious decrease observed when Cu
2
S
was cycled between 1.1-3.0 and 1.2-3.0 V to about 83 and 80 %, respectively.
The cycling behavior at different charge-discharge rates between 0.1 and 5 C
presented in Fig. 7. The 5-cycle performances at higher C-rates are shown in Fig. 7a
and discharge capacity without severe decline can be observed during cycling at
various
cut-off voltages. The overall discharge capacity between 0.1 and 5 C was
−1
approximately around 300 mAh g demonstrating excellent high rate capability of
12

stoichiometric Cu
2
S coated on CFP at 1.0-3.0 V. The cells at higher discharge cut-off
voltages also showed stable capacity retention with very similar discharge capacities
−1
for both in overall around 250 mAh g , when cycled between 1.1-3.0 V and 1.2-3.0
V.
The first oxidation/reduction curves at different current rates have similar flat
st
one-plateau regions except for the 1 cycle in Figs. 7(b-d). While the initial specific
−1
capacity at 0.3 C was 322 mAh g and at 5 C it still showed the high reversible
−
1
capacity of 290 mAh g at 1.0-3.0 V cut-off range. The increase of discharge cut-off
+
voltage cannot benefit for the diffusion ability of Li , since the voltage difference
between discharge and charge is increased. Therefore, charge-discharge capacities
−1
−1
significantly reduced at higher C-rates from 248 mAh g at 0.3 C to 213 mAh g at 5
−1
−1
C between 1.1-3.0 V. Also from 244 mAh g at 0.3 C to 217 mAh g at 5 C between
.2-3.0 V. Additionally, the charge and discharge plateau voltage difference was
mostly negligible with only slight increase from approximately 0.13 V at 1.0-3.0 V to
.15 V at 1.1-3.0 and 1.2-3.0 V. The flat and prolonged plateaus at 1.70 V for
1
0
discharge and 1.85 V for charge was kept unchanged at different C-rates for cells
tested at different cut-off voltages, which implies that the process has good kinetics
even at high current rates.
Toward a further understanding of the influences of discharge cut-off voltages
on electrochemical performance, electrochemical impedance spectroscopy (EIS)
measurements were performed on the cells (1.0~1.2 and 3.0 V) before cycling and
after five discharge-charge cycling tests at 0.1 C as shown in Figs. 8a and 8b. The
four Nyquist curves are all composed of semicircles in the high-to-medium region and
inclined line at the low frequency region. The semicircles are known to be related to
the charge transfer resistance (Rct), which comes from the difficulty of charge transfer
13

between cathode and electrolyte interface, while slope line corresponds to Warburg
+
th
impedance (R
w
) over Li diffusion in the solid materials. The Rct values after the 5
discharge are demonstrated in Table 1, where different discharge capacity values were
st
th
th
also presented after the 1 , 20 and 50 cycles at various cut-off voltage windows.
The EIS results confirmed that the smallest interface barrier occurred at a cut-off
th
voltage of 1.0-3.0 V after the 5 discharge as Rct reached the lowest value compared
with those of other cells (1.1, 1.2-3.0 V). The higher Rct resistance was consistent with
th
the higher capacity drop after the 50 cycle. The Nyquist plots were also measured
th
after the 5 charge process as shown in Fig. 8b. We can observe that the charge
transfer resistance reduced due to the faster kinetics products of the process during
charge (Cu1.96S) in comparison with discharge (Li
2
S). The data difference of Rct at
discharge and charge clearly shown in Fig. 8c, where charge process Rct is lower at
1
.0-3.0, 1.1-3.0 V and gradually increased at 1.2-3.0 V due to the reduced electron
+
transfer. Moreover, Li diffusion could be observed from the R
w
slope line, where it
was steep at 1.0-3.0 V than those of at 1.1-3.0 V and 1.2-3.0 V, which is in agreement
with obtained high capacity results during the high C-rate testing in the case of 1.0-3.0
V cut-off voltage.
4
. Conclusions
2
Successfully synthesized stoichiometric Cu S with monoclinic crystal structure
coated on CFP exhibited high capacity, excellent cycling stability, and good rate
capability. The minor capacity drop and plateau flattening from 2.25 to 1.85 V during
the initial cycles could be linked to the phase change from monoclinic Cu
2
S to
tetragonal Cu1.96S, which was detected by ex situ XRD analysis. Effective electron
14

path and corrosion prevention on CFP enhanced active material utilization, and near
−1
2
theoretical capacity of Cu S 330 mAh g was stable for 50 cycles at 0.1 C and 1.0-3.0
V cut-off voltage window. When the cut-off voltage was increased to 1.2-3.0 V to
st
avoid SEI formation and irreversible discharge capacity at the 1 cycle, the cell still
−1
−1
exhibited relatively high capacity of 280 mAh g at 0.1 C and 250 mAh g at 1 C.
Ex situ XRD analysis showed that the phase transformation during charge, from
monoclinic Cu
.0-3.0 V. The reason could come from the incomplete electrochemical reactions,
which caused higher charge transfer resistance and lowered kinetics. At higher current
rates, stoichiometric Cu S on CFP showed excellent capacity retention without severe
2
S to tetragonal Cu1.96S, progresses slower at 1.2-3.0 V than that of at
1
2
decay between 0.1 and 5 C at various cut-off voltage windows.
Acknowledgements
This research was partially supported by the Ministry of Education, Culture,
Sports, Science and Technology of Japan (Grant No. 15H04251). The authors are also
grateful to Mr. J. Koki of the Center for Advanced Materials Analysis (Tokyo
Institute of Technology, Japan), for the FE-SEM observation of the material.
References
[
1] H. Li, Z. Wang, L. Chen, X. Huang, Research on Advanced Materials for Li-ion
Batteries, Adv. Mater. 21 (2009) 4593-4607.
2] B. Scrosati, Recent advances in lithium ion battery materials, Electrochim. Acta
5 (2000) 2461-2466.
3] N. Nitta, F. Wu, J. T. Lee, G. Yushin, Li-ion battery materials: present and future,
Mater. Today 18 (2015) 252-264.
[
4
[
15

[4] B. Jache, B. Mogwitz, F. Klein, P. Adelhelm, Copper sulfides for rechargeable
lithium batteries: Linking cycling stability to electrolyte composition, J. Power
Sources 247 (2014) 703-711.
[5] C. Shi, X. Li, X. He, J. Zhao, New insight into the interaction between
Carbonate-based Electrolyte and Cuprous Sulfide Electrode Material for Lithium Ion
Batteries, Electrochim. Acta 174 (2015) 1079-1087.
[
6] X. Wang, Y. Wang, X. Li, B. Liu, J. Zhao, A facile synthesis of copper sulfides
composite with lithium-storage properties, J. Power Sources 281 (2015) 185-191.
7] S. Ni, X. Lv, T. Li, X. Yang, Fabrication of Cu S cathode for Li-ion battery via a
low dry thermal sulfuration method, Mater. Chem. Phys. 143 (2013) 349-354.
8] X. Meng, S. C. Riha, J. A. Libera, Q. Wu, H. H. Wang, A. B. F. Martinson, J. W.
Elam, Tunable core-shell single-walled carbon nanotube-Cu S networked
[
2
[
2
nanocomposites as high-performance cathodes for lithium-ion batteries, J. Power
Sources 280 (2015) 621-629.
[9] C. -H. Lai, K. -W. Huang, J. -H. Cheng, C. -Y. Lee, B. -J. Hwang, L. -J. Chen,
Direct growth of high-rate capability and high capacity copper sulfide nanowire array
cathodes for lithium-ion batteries, J. Mater. Chem. 20 (2010) 6638-6645.
[10] W.-Y. Kim, B. M. Palve, H. M. Pathan, O.-S. Joo, Spray pyrolitic deposition of
polycrystalline Cu
2
S thin films, Mater. Chem. Phys. 131 (2011) 525-528.
[11] G. Kalimuldina, I. Taniguchi, Synthesis and electrochemical characterization of
stoichiometric Cu
2
S as cathode material with high rate capability for rechargeable
lithium batteries, J. Power Sources 331 (2016) 258-266.
12] M. Konarova, I. Taniguchi, Synthesis of carbon-coated LiFePO
with high rate performance in lithium secondary batteries, J. Power Sources 195
2010) 3661-3667.
[
4
nanoparticles
(
16

[
13] S. -H. Chung, A. Manthiram, Low-cost, porous carbon current collector with
high sulfur loading for lithium-sulfur batteries, Electrochem. Comm. 38 (2014) 91-95.
14] Y. Liu, B. Jin, Y. F. Zhu, X. Z. Ma, X. Y. Lang, Synthesis of Cu S/carbon
composite with improved lithium storage performance, Int. J. Hydrogen Energy 40
[
2
(2015) 670-674.
[15] L. Yang, S. Cheng, Y. Ding, X. Zhu, Z. L. Wang, M. Liu, Hierarchical Network
Architectures of Carbon Fiber Paper Supported Cobalt Oxide Nanonet for
High-Capacity Pseudocapacitors, Nano Lett. 12 (2012) 321-325.
[
16] I. Taniguchi, C. K. Lim, D. Song, M. Wakihara, Particle morphology and
electrochemical performances of spinel LiMn powders synthesized using
ultrasonic spray pyrolysis method, Solid State Ionics 146 (2002) 239-247.
17] R. W. Potter II, H. T. Evans Jr., Definitive X-ray powder data for covellite,
anilite, djurleite, and chalcocite, J. Research U.S. Geol. Surv. 4 (1976) 205-212.
2
O
4
[
[
18] A. Debart, L. Dupont, R. Patrice, J. -M. Tarascon, Reactivity of transition metal
Co, Ni, Cu) sulphides versus lithium: The intriguing case of copper sulphide, Solid
(
State Sci. 8 (2006) 640-651.
19] J. -S. Chung, H. -J. Sohn, Electrochemical behaviors of CuS as a cathode
material for lithium secondary batteries, J. Power Sources 108 (2002) 226-231.
20] Y. Fu, A. Manthiram, Electrochemical properties of Cu S with ether-based
electrolyte in Li-ion batteries, Electrochim. Acta 109 (2013) 716-719.
21] N. Yamakawa, M. Jiang, C. P. Grey, Investigation of the Conversion Reaction
[
[
2
[
Mechanisms for Binary Copper (II) Compounds by Solid-State NMP Spectroscopy
and X-ray Diffraction, Chem. Mater. 21 (2009) 3162-3176.
17

[
22] A. Beleanu, Ternary Lithium Based Compounds Used for Technological
Applications, Pharmazie und Geowissenschhaften der Johannes
Gutenberg-Universitat Mainz, 2014. Doctoral dissertation.
2
Table 1 Discharge capacity and charge transfer resistance of stoichiometric Cu S
electrode on CFP at 0.1 C rate and various cut-off voltages.
st
th
th
Cut-off range / V
1 cycle
20 cycle
50 cycle
Rct / ohm
−1
−1
−1
/
mAh g
/ mAh g
/ mAh g
1
1
.0-3.0
.1-3.0
532
421
321
298
311
284
8.84
11.69
1
.2-3.0
338
285
266
12.01
18

Figure captions
2
Fig. 1. XRD spectra of stoichiometric Cu S. Inset: SEM image of powders.
Fig. 2. (a) Charge-discharge voltage profiles vs. capacity of stoichiometric Cu2S
coated on a CFP at a cut-off voltage of 1.0-3.0 V at 0.1 C; (b) Cyclic voltammetry at a
-1
scan rate of 0.05 mV s between 1.0-3.0 V; (c) XRD patterns of the fully discharged
st
th
electrode at the 1 and 5 cycles; (d) XRD patterns of the fully charged electrode at
st
th
the 1 and 5 cycles.
2
Fig. 3. (a) Charge-discharge voltage profiles vs. capacity of stoichiometric Cu S
coated on a CFP at a cut-off voltage of 1.1-3.0 V at 0.1 C; (b) Cyclic voltammetry at a
-1
scan rate of 0.05 mV s between 1.1-3.0 V; (c) XRD patterns of the fully discharged
st
th
electrode at the 1 and 5 cycles; (d) XRD patterns of the fully charged electrode at
st
th
the 1 and 5 cycles.
2
Fig. 4. (a) Charge-discharge voltage profiles vs. capacity of stoichiometric Cu S
coated on a CFP at a cut-off voltage of 1.2-3.0 V at 0.1 C; (b) Cyclic voltammetry at a
-1
scan rate of 0.05 mV s between 1.2-3.0 V; (c) XRD patterns of the fully discharged
19

st
th
electrode at the 1 and 5 cycles; (d) XRD patterns of the fully charged electrode at
st
th
the 1 and 5 cycles.
2
Fig. 5. Cycle performance of stoichiometric Cu S coated on a CFP at various cut-off
voltage ranges between 1.0, 1.1, 1.2 and 3.0 V: (a) at 0.1 C for 50 cycles and (b) at 1
st
C for 100 cycles with Coulombic efficiencies; (c) XRD after the 1 discharge at 1 C;
th
(
d) XRD after the 50 discharge at 1 C.
2
Fig. 6. Capacity retention of stoichiometric Cu S at various cut-off voltage windows
at 1 C rate for 100 cycles.
Fig. 7. (a) Rate performance of stoichiometric Cu
voltage ranges; Charge-discharge voltage profiles vs. capacity between: (b) 1.0-3.0 V;
c) 1.1-3.0 V; (d) 1.2-3.0 V.
Fig. 8. Electrochemical impedance spectra (EIS) of stoichiometric Cu
2
S from 0.1 to 5 C at various cut-off
(
2
S coated on a
th
th
CFP (a) after the 5 discharge, (b) after the 5 charge and (c) relationship between
charge-transfer resistance Rct and lower cut-off voltage.
20

Cu S (ICDD:00-033-0490)
2
2
0
30
40
50
2
θ/degree
Fig.1
21

3
2
1
(a) 1.0-3.0 V
(b) 1.0-3.0 V
2
0
2
st
1
5
cycle
cycle
th
st
1
cycle
cycle
th
-
5
0
100
200
300
400
500
1
2
3
-
1
Voltage / V
Capacity / mAh g
(c)
1.0-3.0 V
Cu (ICDD-4-0836)
(d)
1.0-3.0 V
Cu (ICDD-4-0836)
th
5
discharge
th
5
charge
st
st
1
discharge
1 charge
fresh electrode
carbon paper
fresh electrode
carbon paper
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Li2 S(ICDD-01-077-2145)
20
25
30
35
40
45
50
20
25
30
35
40
45
50
2
/degree
2 /degree
Fig.2
22

3
2
1
(a) 1.1-3.0 V
(b) 1.1-3.0 V
2
0
2
st
1
5
cycle
cycle
th
st
1
cycle
cycle
th
-
5
0
100
200
300
400
500
1
2
3
-
1
Voltage / V
Capacity / mAh g
(c)
1.1-3.0 V
Cu (ICDD-4-0836)
(d)
1.1-3.0 V
Cu (ICDD-4-0836)
th
5
discharge
th
5
charge
st
st
1
discharge
1 charge
fresh electrode
carbon paper
fresh electrode
carbon paper
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Li2 S(ICDD-01-077-2145)
20
25
30
35
40
45
50
20
25
30
35
40
45
50
2
/degree
2 /degree
Fig.3
23

3
2
1
(b) 1.2-3.0 V
(a) 1.2-3.0 V
2
0
2
st
1
5
cycle
cycle
th
st
1
cycle
cycle
th
5
-
0
100
200
300
400
500
1
2
3
-
1
Voltage / V
Capacity / mAh g
(c)
1.2-3.0 V
Cu (ICDD-4-0836)
(d)
1.2-3.0 V
Cu (ICDD-4-0836)
th
th
5
discharge
5
charge
st
1
discharge
st
1
charge
fresh electrode
fresh electrode
carbon paper
carbon paper
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Li2 S(ICDD-01-077-2145)
Li2 S(ICDD-01-077-2145)
20
25
30
35
40
45
50
20
25
30
35
40
45
50
2
/degree
2 /degree
Fig.4
24

500
400
300
200
100
0
100
80
60
40
20
0
500
400
300
200
100
0
100
80
60
40
20
0
1
1
.0-3.0 V
.1-3.0 V
1
1
1
.0-3.0 V
.1-3.0 V
.2-3.0 V
1.2-3.0 V
(
a) 0.1C-rate
(b) 1C-rate
10
20
30
40
50
20
40
60
80
100
Cycle number
Cycle number
st
th
Cu (ICDD-4-0836)
Cu (ICDD-4-0836)
1.2-3.0 V
1
1
discharge
50 discharge
1C-rate
(c)
(d)
C-rate
1.2-3.0 V
1.1-3.0 V
1.1-3.0 V
1
.0-3.0 V
1.0-3.0 V
fresh electrode
fresh electrode
carbon paper
carbon paper
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Cu1.96 S(ICDD-029-0578)-Tetragonal
Cu2 S(ICDD-033-0490)-Monoclinic
Li2 S(ICDD-01-077-2145)
Li2 S(ICDD-01-077-2145)
20
25
30
35
40
45
50
20
25
30
35
40
45
50
2
/degree
2 /degree
Fig.5
25

1
00
80
60
40
20
0
1
1
1
.0-3.0 V
.1-3.0 V
.2-3.0 V
1
C-rate
20
40
60
80
100
Cycle number
Fig.6
26

6
5
4
3
2
1
00 (a)
3
.5
2
(
b) 1.0-3.0 V
00
0
0
0
.1C
.3C
.5C
2
1
charge
0.1C 0.3C 0.5C 1C
2C
3C
5C
0.1C
00
00
00
00
1C
2
3
5
C
C
C
.5
1
discharge
1
.0-3.0 V
1.1-3.0 V
1
.2-3.0 V
0
10
20
30
40
0
100
200
300
400
500
600
-
1
Cycle number
Capacity / mAh g
3
.5
2
(c) 1.1-3.0 V
3 (d) 1.2-3.0 V
0
0
0
1
2
3
.1C
0.1C
0.3C
0.5C
1C
charge
charge
2
1
2.5
2
.3C
.5C
C
C
2C
C
3C
5C
.5
1
5C
1.5
1
discharge
discharge
100
0
100
200
300
400
500
600
0
200
300
400
500
600
-
1
-1
Capacity / mAh g
Capacity / mAh g
Fig.7
27

th
-
-
-
20 (a)After 5 discharge
fresh cell
15
1
1
1
.0-3.0 V
.1-3.0 V
.2-3.0 V
10
-
5
0
5
10
15
20
Z' /ohm
th
-
-
-
20 (b) After 5 charge
15
10
fresh cell
1
1
1
.0-3.0 V
.1-3.0 V
.2-3.0 V
-
5
0
5
10
15
20
Z' /ohm
Fig.8
28