Access to this full-text is provided by IOP Publishing.
Content available from Journal of The Electrochemical Society
This content is subject to copyright. Terms and conditions apply.
Novel In Situ Gas Formation Analysis Technique Using a
Multilayer Pouch Bag Lithium Ion Cell Equipped with Gas
Sampling Port
Jan-Patrick Schmiegel,
1,
*Marco Leißing,
1
Franz Weddeling,
1
Fabian Horsthemke,
1,
*
Jakub Reiter,
2
Quan Fan,
3
Sascha Nowak,
1,
** Martin Winter,
1,4,
***
,z
and
Tobias Placke
1,z
1
University of Münster, MEET Battery Research Center, Institute of Physical Chemistry, 48149 Münster, Germany
2
BMW Group, Petuelring 130, 80788 Munich, Germany
3
Contemporary Amperex Technology Limited, Ningde 352100, Fujian Province, People’s Republic of China
4
Helmholtz Institute Münster, IEK-12, Forschungszentrum Jülich GmbH, 48149 Münster, Germany
Parasitic gas evolution in lithium ion battery (LIB) cells especially occurs within the first charge cycle, but can also take place
during long-term cycling and storage, thereby, negatively affecting the cell performance. Gas formation is influenced by various
factors, such as the cell chemistry and operating conditions, thus, demanding fundamental studies in terms of interphase and gas
formation (gas volume and composition) and electrolyte consumption. Gas analyses in terms of mass spectrometry of gaseous
products are regularly performed, however, usually using custom-made cell designs with a high excess of electrolyte. Here, a gas
sampling port (GSP) is incorporated in a commercial small-scale multilayer pouch cell in a simple post-production process and
systematically evaluated as proof-of-principle approach towards effective electrolyte additive research under practically relevant
conditions, i.e., when applying a limited amount of electrolyte per cell capacity. The GSP-based LIB pouch cell design allows the
voltage-dependent identification and separation of formed gases, while a clear correlation between electrolyte reduction peaks,
observed in differential capacity profiles, and the onset of gas evolution is demonstrated. In summary, the novel GSP-based pouch
cell setup benefits from the possibility of multiple time-, cell voltage- or state-of-charge-dependent gas measurements, without
significantly influencing the original cell performance.
© 2020 The Author(s). Published on behalf of The Electrochemical Society by IOP Publishing Limited. This is an open access
article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/
by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/
1945-7111/ab8409]
Manuscript submitted February 13, 2020; revised manuscript received March 10, 2020. Published April 9, 2020.
Supplementary material for this article is available online
Lithium ion batteries (LIBs) are the predominant rechargeable
energy storage system to replace fossil fuels in conventional means
of transportation. Within recent years, the need for LIBs has grown
significantly by the demand for millions of electric vehicles (EVs),
plug-in hybrid electric vehicles (PHEVs) or hybrid electric vehicles
(HEVs) as well as consumer electronics and various other applica-
tion purposes.
1–3
However, for application in EVs targeting driving
ranges above 500 km while remaining cost-competitive with internal
combustion engine-based cars, energy densities above 500 Wh l
−1
and costs below 125 US$ kWh
−1
(both at battery pack level) are
required. Consequently, performance and cost improvements along
the whole battery value chain, i.e., in terms of the active materials
and components, such as the electrolyte, are obligatory for being
cost-competitive and to achieve the above-mentioned energy den-
sities while still exhibiting a sufficient cycle/calendar life.
4–6
On material level, these improvements can be achieved by
replacing the state-of-the-art electrode materials with new electrode
materials with increased capacity and higher operation voltage.
7
The
capacity of the negative electrode can be raised by replacing
intercalation-type graphite by “alloying-type”silicon, which is able
to provide an almost 10 times higher specific capacity, compared to
graphite. Furthermore, Ni-rich or high-voltage NCM-based cathode
materials are established as next generation of positive electrode
materials, delivering higher discharge capacities. However, this
increase in discharge capacity with both electrode materials is at the
detriment of the charge/discharge cycling stability.
8–13
Incorporating electrolyte additives in conventional state-of-the-
art carbonate-based electrolytes, i.e., ethylene carbonate (EC) in
combination with linear carbonates and LiPF
6
, offers a straightforward
approach to improve the electrochemical performance in these LIB
cell chemistries, e.g., by formation of an effective solid electrolyte
interphase (SEI) at the negative electrode.
14–19
However, many LIB
cell chemistries in combination with the electrolyte evolve gases
during cycling or storage, while impacting the cycling performance
and introducing a safety risk.
20–25
Significant gas evolution in LIB cells generally occurs within the
very first charge cycle due to reductive decomposition of the
electrolyte during the SEI formation, but can also take place during
long-term cycling or storage.
26–31
In particular, high-capacity
negative electrode materials such as silicon suffer from an ineffec-
tive SEI, leading to continuous electrolyte decomposition during
prolonged cycling, which might result in severe gassing.
32
In this
respect, electrolyte additives might be able to reduce the gassing
behavior of LIB cells significantly, which has been shown in various
reports.
16,32–34
However, since different LIB cell chemistries and
various electrolytes and combinations of additives are applied, a
deeper understanding on the underlying gas evolution mechanism is
important. In general, a different gassing behavior can be expected
for different cell chemistries, which can be further influenced by the
operating conditions (cell voltage range, temperature) and by cross-
talk phenomena.
26
Furthermore, especially Ni-rich NCM cathode
materials tend to generate CO
2
by chemical rather than electro-
chemical oxidation of the electrolyte, due to the release of reactive
lattice oxygen species.
8,35,36
In order to elucidate the gassing behavior of electrolytes, mass
spectrometry (MS) analysis of gaseous products is typically per-
formed in literature, usually using a custom-made cell setup or cell
housing for these kinds of measurements.
37–42
Even though these
studies have provided a deeper understanding of the gas evolution
behavior of different active materials or electrolyte formulations,
there is still a lack of gas evolution analysis in LIB full-cell setups
43
under commercially relevant conditions, i.e., when applying a
limited amount of electrolyte per cell capacity. In this respect, a
practical electrolyte to cell capacity ratio of <4gAh
−1
(e.g., in the
z
E-mail: martin.winter@uni-muenster.de;tobias.placke@uni-muenster.de
*Electrochemical Society Student Member.
**Electrochemical Society Member.
***Electrochemical Society Fellow.
Journal of The Electrochemical Society, 2020 167 060516
range of 1.3 to 1.5 g Ah
−1
) has been proposed for high-energy LIB
cells, while laboratory cells typically use large excess of electrolyte
(i.e., >13 g Ah
−1
), which in turn can significantly increase the cell
cycle life and, thus, does not give practically relevant data.
44,45
Recently, a pouch cell/pouch bag method was developed by
Dahn et al. to study gas generation and consumption at the negative
and positive electrode separately, in order to study the “cross-talk”
between the electrodes within an LIB cell.
30,32,35,36
With this
method, the gas sampling is performed in a gas-tight brass chamber,
while the pouch cell is penetrated with a sharp needle, allowing a
one-time measurement of the gases, in order to extract the permanent
gases from the pouch cell.
41,46
In contrast to this cell-destructive method, we include a gas
sampling port (GSP) in a commercial small-scale multilayer pouch
cell in a simple post-production process, which benefits from the
non-destructiveness as well as the possibility of multiple time-,
voltage- or state-of-charge (SOC)-dependent measurements without
significantly influencing the electrochemical performance of the
pouch cell (Fig. 1). In this report, the GSP is introduced as a proof-
of-concept and used to demonstrate the effect of the electrolyte
additive fluoroethylene carbonate (FEC) on the amount and compo-
sition of the formed gases, whereas the practicality of the pouch cell
setup itself is supported by literature data.
17,33,41,47,48
Experimental
Pouch cell setup and custom-made cell holder design.—Small-
scale (35 ×20 ×4 mm) wound LIB pouch cells with a nominal
capacity of 240 mAh and an NCM-811 ∣∣ artificial graphite (AG) cell
chemistry were obtained from Li-Fun Technology (Xinma Industry
Zone, Golden Dragon Road, Tianyuan District, Zhuzhou City,
Hunan Province, PRC, 412000). The negative electrode was based
on 94.8 wt% artificial graphite (AG) as active material, a mixture of
carboxymethyl cellulose (CMC) and styrene-butadiene rubber
(SBR) as binder (1.3 wt% CMC +2.5 wt% SBR) and carbon black
(CB) as conductive agent (1.4 wt%). The positive electrode
consisted of NCM-811 as active material (96.4 wt%), poly(vinyli-
dene) difluoride (PVdF) as binder (2.0 wt%) and CB as conductive
additive (1.6 wt%). The cells were capacity-balanced to an operation
cut-off voltage of 4.3 V. The negative and positive electrode mass
loadings were ≈10.0 mg cm
−2
and ≈13.9 mg cm
−2
, respectively.
The separator was based on polyethylene (PE), which was one-side
coated with Al
2
O
3
facing the positive electrode. The cells were
vacuum-sealed in a dry state (without electrolyte) and subsequently
shipped. Before electrolyte filling, the cells were opened in a dry-
room (dew point: −80 °C, <0.55 ppm water) and dried at 80 °C
under reduced pressure (<1 mbar) for at least 12 h to remove any
residual moisture from the cell. The cells were filled with 0.75 ml
(≈0.9 g) electrolyte and vacuum sealed by heat-crimping at 165 °C
for 5 s at a relative pressure of −90 kPa (≈100 mbar (≈10 kPa)
absolute pressure) using a vacuum sealer (GN-HS200V, Gelon LIB
Group). All pouch cells used in this study were filled with 1 M LiPF
6
in ethylene carbonate (EC): ethyl methyl carbonate (EMC) (3:7, by
weight, Solvionic, battery grade) electrolyte. As a benchmark
electrolyte additive, 2.0 wt% fluoroethylene carbonate (FEC,
Sigma Aldrich, battery grade) was used. Once sealed, the pouch
cells were clamped into a specifically designed cell holder that
ensures a defined and reproducible pressure (≈2 bar) on the cell
stack by the usage of a torque screwdriver (4 cNm, cf. Fig. 2).
Electrochemical cycling procedure.—The cells were connected to
aMaccor 4000 series charger and held at 1.5 V for 20 h in a
temperature-controlled chamber at 20 °C to ensure a complete electrode
wetting and to prevent copper dissolution, as reported in previous
studies.
16,17,34,38,49,50
For long-term cycling, a modified cycling proce-
dure was used, which was introduced elsewhere.
16,17,34,38
First, the cells
were charged with a constant current of 10 mA to 3.5 V, followed by a
constant voltage (CV) step for 1 h and a subsequent discharge at 10 mA
to 2.8 V, in order to allow a homogeneous SEI formation at the
negative electrode at low current rates. Two subsequent cycles at 40
mA between 2.8 V and 4.3 V, with a CV step at 4.3 V until the current
droppedbelow4mA,wereappliedtocompletetheformation
procedure. In order to calculate the Coulombic efficiencies (C
Eff
)in
the first cycle, the pre-cycle (up to 3.5 V) and the subsequent full cycle
(up to 4.3 V) were combined in the following. The cells were then
transferred into a dry-room to be opened, and degassed by vacuum re-
sealing as described above, to remove any gas formed during the
formation cycles. The pouch cells were then re-clamped in the cell
holder (same conditions as described above) and connected to the
battery tester. For long-term cycling, the cells were cycled at 100 mA
between 2.8 V and 4.3 V with a constant voltage step at 4.3 V until the
current dropped below 4 mA.
For in situ gas sampling, a modified formation procedure was
used, which will be discussed in detail below.
Introduction of in situ gas sampling port to pouch cell.—The
in situ analysis of the gases formed within the NCM-811 ∣∣ AG
pouch cells made use of a polypropylene (PP)-based cannula system
which was heat-sealed into the PP-based inner layer of the pouch
foil. Prior to electrolyte filling, a flexible PP-based cannula/dispen-
sing tip (Gonano Dosiertechnik GmbH; length: 2.54 cm; inner Ø:
0.26 mm; outer Ø: 0.8 mm, cf. Fig. 3,part 1) was heat-sealed into
the pouch bag foil using a custom-made copper bracket with a 1 mm
drilling. The gas sampling port (GSP) was fused to the pouch foil by
heat-crimping at 195 °C for 20 s at a relative pressure of −90 kPa
(≈100 mbar absolute pressure) using a vacuum sealer. The sealing
temperature and time were increased, compared to the regular
sealing process to compensate the heat hysteresis of the copper
sealing bracket. In order to prevent a possible fusing of the cannula
itself, a stainless-steel wire (0.26 mm diameter) was inserted into the
cannula temporarily during heat sealing. A schematic illustration of
the process can be seen in Fig. 3and an image of the pouch cell with
GSP is shown in Fig. S1 (available online at stacks.iop.org/JES/167/
060516/mmedia).
Figure 1. Schematic illustration of the gas sampling process and the corresponding voltage-dependent gas chromatograms.
Journal of The Electrochemical Society, 2020 167 060516
The cells were filled with electrolyte using the just introduced
Luer-Lock cannula port and vacuum-sealed under the same condi-
tions as the reference cells (without gas-port). The Luer-Lock port
itself was sealed with a commercially available Luer-Lock to M10
adapter (PP, RCT Reichelt Chemietechnik GmbH +Co., cf. Fig. 3,
part 2), equipped with a ultra-low bleed GC injector septa
(Thermogreen LB-2 for Shimadzu,SUPELCO, cf. Fig. 3, part 3),
which was fixed with a GC-vial screw cap (PP, 9 mm, PTFE septa,
VWR International; cf. Fig. 3,part 4).
In situ gas chromatography—barrier ionization discharge
detector.—The analysis of the gas formed in these pouch cells
followed the gas chromatography—barrier ionization discharge (GC-
BID) technique described by Horsthemke et al.
51
All samples were
measured with a GC-2010 Plus system, equipped with a BID-2010 Plus
detector (all Shimadzu). The GC-system was equipped with a PLOT
gas separation column RT
®
-Msieve 5Å (30 m ×0.32 mm ×30 μm,
Restek). The analysis was carried out with the help of LabSolutions
software (version 5.90, Shimadzu). A gas mixture containing a variety
of different gaseous decomposition products, which were previously
found in LIB cells, was used as external calibration standard. Further
details on the analytical method and the experimental details have been
reported in a previous publication.
51
After electrolyte filling, the cells with GSP were processed in the
same way as the reference cells (all measurements at 20 °C).
However, for the purpose of a cell voltage-dependent measurement
of the total volume of gas and of the composition of the formation
gases likewise, a modified formation procedure was applied. In
contrast to the long-term cycling procedure, the formation procedure
for the in situ gas sampling was performed in a single charge/
discharge cycle.
Subsequent to the wetting process at 1.5 V for 20 h, 200 μlof
argon (Westfalen; Argon 5.0; Purity: ⩾99.999%) were added via a
gastight syringe through the rubber GC-septum into the pouch cell
(cf Fig. 4). After complete mixing (≈5 min), 5 μl of gas were
extracted and analysed by GC-BID as a blank sample. Afterwards,
the cells were charged at 40 mA to 3.1 V, followed by a 10 min CV
step to decrease overpotentials and subsequently, 5 μl of gas were
extracted and analysed. Thereupon, the cells were further charged at
40 mA up to 4.3 V with 10 min CV steps at 3.4, 3.7, 4.0 and 4.3 V
and subsequently discharged to 2.8 V for gas sampling (see also
Fig. 8). In each step, 5 μl of gas were extracted and analysed.
Figure 2. Schematic illustration of the wound pouch cells and the custom-designed cell holder.
Figure 3. Schematic illustration of the GSP setup and sealing process of the pouch cell.
Journal of The Electrochemical Society, 2020 167 060516
In situ cell volume measurements.—In situ gas volume measure-
ments during the formation cycle were performed using a method
introduced by Dahn and co-workers.
33
The apparatus measured the gas
evolution in pouch cells non-destructively during operation. The
Archimedes′in situ gas analyser (AISGA)
33,52
uses a small-scale thin
film load cell (S256, 10 g, Strain Measurement Devices,Ltd.)to
measure the buoyant force of the cells suspended in a fluid (MilliQ
water; Merck Millipore Milli-Q Advantage A10; resistivity at 25 °C =
18.2 MΩcm; total organic carbon (TOC) ⩽5 ppb). When applying a
force on the load cell, a deformation of the strain gauges resulted in a
resistance alteration, which could be measured as a change in voltage
using a USB data acquisition module (OMB-DAQ-2408 Series,
OMEGA Engineering GmbH). Through calibration with precise
masses, the output of the load cell could be translated to the mass of
the displaced water and, therefore, the volume change of the pouch
cell.
33,52
Gas tightness of the GSP-based pouch cell.—The air tightness
or a possible leakage of the newly introduced GSP was investigated
by several techniques. The cells were filled with ≈1 ml Ar, while the
buoyant force was recorded for several days. Therefore, a possible
leakage or gas diffusion through the system could be correlated to
the mass of the displaced water.
Furthermore, leak detection by means of vacuum assisted
detection of carbonates by mass spectrometry (MS) was applied.
For this purpose, the pouch cells with the GSP were vacuum sealed
prior to the measurement (as described above) and loaded with pure
DMC via a syringe through the rubber GC-septum into the pouch
cell. Subsequently, the cell was placed in the vacuum chamber.
The system (Inficon GmbH) itself consisted of two main
components, a vacuum chamber into which the sealed GSP-based
pouch cell was introduced and a detection unit that was based on a
quadrupole-MS. The system itself was calibrated with dimethyl
carbonate (DMC) as a volatile organic carbonate. DMC was used
since the fragmentation reaction within the MS-system yields
descent fragments with known m/z-ratios (m/z 59). When using
unsymmetrical carbonates like EMC or mixtures of carbonates like
in electrolytes, these multi-component systems can yield in addi-
tional fragments or can overlap with those of DMC and, therefore,
disturb the quantification of these ions.
Since the leakage rate was proportional to the amount of DMC
that could be detected by the MS, a quantification of the leakage rate
could be performed. Through a calibration with a DMC filled
cylinder with a defined leakage rate, the leakage rate of the sealed
GSP-based pouch cell was determined (all measurements were
carried out at 20 °C).
Analysis of the liquid (portion of the) electrolyte.—The analysis
of the electrolyte of cycled NCM-811 ∣∣ AG pouch cells was
followed the procedure reported by Horsthemke et al. in order to
evaluate possible contaminations by the GSP.
53,54
After the forma-
tion procedure, the cells were transferred into a glovebox and opened
in order to extract the cell stack from the cell housing. Afterwards,
the cell stack was unwound, separating the anode and the cathode,
while the anode and separator were transferred into a centrifuge vial.
The liquid phase was later separated from the solid phase and
analysed using gas chromatography (GC) coupled with mass
spectrometry (MS). Experimental details and parameters can be
found in previous publications.
53,54
Results and Discussion
Electrochemical performance survey.—In order to investigate a
possible impact of the GSP on the electrochemical performance of
the NCM-811 ∣∣ AG pouch cells, two cells with GSP and two cells
without GSP were assembled and evaluated via charge/discharge
cycling. Besides the GSP, all cells contained the same amount of
electrolyte and were cycled under the same conditions while being
clamped in the custom-made cell holder.
Figure 4. Schematic illustration of the gas sampling during operation using the GSP and a gastight syringe.
Figure 5. Long-term charge/discharge cycling performance of two NCM-
811 ∣∣ AG pouch cells with and without GSP using the reference electrolyte
(1 M LiPF
6
in EC:EMC, 3:7 by wt.). Voltage range: 2.8–4.3 V; operation
temperature: 20 °C; Error range of the capacity: ⩽1 mAh.
Journal of The Electrochemical Society, 2020 167 060516
Figure 5shows the long-term cycling performance of cells with
and without GSP using the reference electrolyte (1 M LiPF
6
in EC:
EMC (3:7, by wt.). No significant difference in terms of discharge
capacity or capacity fading between the different cells can be
observed. Pouch bag cells with an included GSP show a slightly
lower initial discharge capacity compared to the reference cells,
however, it has to be mentioned, that the difference between both
cell systems is only ≈2 mAh, which is a deviation of below ≈1.0%
of the initial charge capacity.
Especially for parasitic reactions during charge/discharge, the
Coulombic efficiency (C
Eff
) can be used to determine any reductive/
oxidative processes during operation. During the measurement and
even throughout long-term cycling, no significant difference in terms
of C
Eff
is noticeable for cells with or without GSP (cf. Table SI, Fig.
S2). It has to be noticed, that for cells without GSP at cycles >180, a
slightly lower C
Eff
is obvious. However, the difference in C
Eff
is
below 0.06%. As gas sampling and subsequent analysis is especially
performed within the formation cycles, the effect of the GSP on the
electrochemical performance will be neglected in the following.
Considering the direct physical contact between the canula of the
GSP and the electrolyte, a high chemical resistance of these
materials towards the carbonate-based solvents along with the
conductive salt LiPF
6
and possible additives is of utmost impor-
tance. Appropriately, a polypropylene-based cannula is used to
minimize the possible contamination by the used materials.
However, it cannot be excluded, that other components of the GSP
(like plasticizer or other additives) can be dissolved into the
electrolyte, especially at high operating temperatures.
Within the literature, dQ/dV-analysis is applied as a technique to
identify electrochemical reactions (oxidation/reduction) or decom-
position reactions of electrolyte additives.
15
Hence, dQ/dV-analysis
was performed to exclude the electrochemical reaction of the GSP or
cell components during operation. In addition, the dQ/dVanalysis
can be used to estimate the onset of gas evolution in these cells for
the subsequent in situ gas sampling. However, it has to be
mentioned, that there is typically a small offset between the
electrochemical signal (here cell voltage of the dQ/dVpeak) and
the gas evolution.
29,30,41
The first change in cell volume (i.e. gas
evolution) is shifted to higher voltages compared to the dQ/dVpeak
(i.e., reduction of EC or components such as FEC) and, therefore
will not necessarily be at the same cell voltage. While the underlying
reason or mechanism for this phenomenon is unclear, this shift,
according to Dahn et al., might be related to the gas generation
within a secondary reaction and not the initial reduction of several
additives, or the slow nucleation kinetics of the gaseous products.
41
Figure 6shows the cell differential capacity plot for the
formation cycle with and without GSP with two almost congruent
curves. Both differential capacities show an overlapping peak at
≈3.0 V, which can be attributed to the reduction of ethylene
carbonate (EC) at the graphite surface due to solid electrolyte
interphase (SEI) formation.
30,55–57
Furthermore, no additional peaks
are visible within the dQ/dVplot, indicating no electrochemical
activity or decomposition of the GSP within these cells.
Nevertheless, a possible contamination of the electrolyte due to
the GSP cannot be excluded with this technique, hence, analytical
techniques must be utilized to exclude a possible “bleeding”of the
GSP. Neither within the first cycle (cf. Fig. 6), nor during long-term
cycling (cf. Figs. S3 and S4) a difference between cells with and
those without GSP is visible in the differential capacity plots.
In order to determine whether the GSP leads to a contamination
of the electrolyte by “bleeding”or reactions with the cell compo-
nents during operation, GC-MS of the electrolyte was performed.
Within the operation, no differences between the electrolyte that was
in contact with the GSP or the pristine electrolyte were visible.
Leakage evaluation of the GSP-based pouch cell.—Leak detec-
tion is performed by the application of a vacuum leakage detection,
which is applied on this GSP-based pouch cell and explained in detail
within the experimental section. As a result of the external vacuum of
≈4 mbar that is applied onto the pouch cells with GSP, even the
smallest leaks can be detected (up to 1∙10
−8
mbar L s
−1
)bycauseof
the instant evaporation of DMC, that is detected within the mass
spectrometry. In industrial processes leakage rates <1∙10
−6
mbar L s
−1
were defined as very tight systems.
58
For these cells, a leakage rate as small as 1.44 ± 0.33 ∙10
−7
mbar
Ls
−1
was detected. Based on these results and the specification of a
leaky system with leakage rates >1∙10
−6
mbar L s
−1
, by the
equipment, the gas tightness of the GSP-based cell is confirmed.
Furthermore, the permeability of the GSP-based cell towards
permanent gases was investigated by the Archimedes’In Situ Gas
Analyzer (AISGA) method.
33,59
By inflating the pouch cells with a
gas and surveying the buoyant force and, therefore, solely the
alteration of the volume of the cell, the permeability of the GSP-
based cell towards gas diffusion could be determined. As can be seen
in Fig. 7, the cell, which was previously filled with ≈1 ml Ar gas,
does not suffer from a significant amount of volume loss within
several days. The deviation of 22 μl within four days can be mainly
be attributed to the signal-to-noise ratio of the measurement, rather
than a volume decrease due to diffusion.
Figure 6. Differential cell capacity vs cell voltage for the first formation
cycle of NCM-811 ∣∣ AG pouch cells with or without GSP (cell voltage
range: 1.5 V to 3.5 V; operation temperature: 20 °C).
Figure 7. Representative volume change data of the GSP-based pouch cell
measured as a function of days, indicating no volume loss due to diffusion.
The cell was previously filled with Ar, while the buoyant force was measured
each day.
Journal of The Electrochemical Society, 2020 167 060516
Voltage-dependent analysis of the gassing mechanism in the
formation cycle.—Figure 8shows the voltage profile of the NCM-
811 ∣∣ AG pouch bag cells with an incorporated GSP, cycled at 40
mA between 2.8 and 4.3 V with 10 min CV steps at 3.1, 3.4, 3,7, 4.0,
4.3 and 2.8 V for gas sampling. In the voltage profiles shown here,
the reference electrolyte and 2.0 wt% FEC-containing cells are
shown.
In carbonate-based electrolytes, the reductive instabilities of the
cyclic and linear carbonates are the main reason for gas evolution in
the formation procedure.
22,23,37,47
In particular, the reductive decom-
position of EC within the SEI formation produces large amounts of
different gases, such as ethene (C
2
H
4
), carbon monoxide (CO) or
hydrogen (H
2
).
27,32,37
Since EC reduction and, therefore, gas evolution starts at a cell
voltage of ≈3.0 V (cf. Fig. 6), only a negligible amount of gas
generation is expected below this threshold voltage (≈3.0 V).
Nonetheless, as discussed in literature,
41
an offset between the cell
voltage of the dQ/dVpeak and the gas evolution is observed, which
can cause in an even higher shift of the onset of gas generation
towards higher cell voltages.
In Fig. 9, a representative cell voltage resolved GC-BID
chromatogram of the gases formed within the formation cycle is
shown. Based on the previous work by Horsthemke et al. and the
classification of the different peaks by mass spectrometry,
51
all
peaks can be identified as permanent gases and light hydrocarbons
evolving during cell formation and are in good agreement with
literature reports.
37,39,47,48,59–61
After the tap charge to 1.5 V and the complete wetting process, a
blank gas sample was extracted and injected into the GC system. At
this cell voltage, only argon (matrix), oxygen and nitrogen can be
detected. Thereby, oxygen and nitrogen appear as contaminants from
the atmosphere during the manual sampling from the GSP to the GC
injector port. However, this contamination is reproducible and
almost constant for all measurements and cell voltages (cf. Figs. 9
and 11). In the subsequent charge up to 3.1 V, H
2
, methane (CH
4
),
CO and C
2
H
4
are detected within the gas sample and are, thus, in good
agreement with the literature.
37,38,62,63
Considering the dQ/dVanalysis
in Fig. 6, the decomposition of the electrolyte starts at a cell voltage of
≈3.0 V, whereas the peak of total gas volume can be shifted towards
higher voltages.
30
The formation of H
2
is usually attributed to a
decomposition reaction of trace amounts of water at the negative
electrode surface,
37,39,60
while C
2
H
4
and CO originate from reductive
decomposition of EC via a ring-opening reaction.
40,42,64,65
In contrast,
CH
4
evolution is assumed to be generated by reductive decomposition
of the linear carbonates, such as EMC, and subsequent radical reaction
with hydrogen.
42,65–67
Upon a consecutive charge to 3.4 V, an
immense increase of the C
2
H
4
and CO concentrations was visible,
while H
2
and CH
4
are increasing likewise, but less severe.
Furthermore, ethane (C
2
H
6
) as another gaseous decomposition product
was detected.
27,40,65
In the following charging step up to a cut-off
voltage of 3.7 V, a further, but rather modest increase in intensity was
detected for all gaseous decomposition products. With increasing cell
voltage (>3.7 V), the intensities for all gases measured in this
experiment, remain almost constant, confirming the results reported
in literature, which showed that the majority of gas generation in
carbonate-based electrolytes occurred at voltages below 3.5 V.
33,68
Furthermore, this again confirms no or negligible gas leakage or
diffusion of the GSP during operation.
In the following, FEC as a well-known SEI-forming electrolyte
additive was added to the reference electrolyte in order to investigate
the gassing behaviour of the electrolyte by means of the described
in situ gas formation analysis in multilayer pouch bag cells,
equipped with a GSP. By the addition of FEC, the reductive
decomposition of EC is prevented almost entirely (cf. Fig. 10)
and, therefore, the total amount of gas evolved upon operation is
greatly decreased, as known from literature.
47,69,70
The addition of
FEC to the electrolyte leads to a broad reduction feature in the cell
voltage range of ≈2.60 V to ≈3.10 V. As discussed in literature, the
peak broadening can be explained by FEC reduction at the negative
electrode, resulting in CO
2
evolution and subsequent CO
2
consump-
tion at lower anode potentials (=higher cell voltages).
47,48
Furthermore, the reductive decomposition mechanism of FEC
usually yields CO
2
, LiF, Li
2
O, Li
2
CO
3
,H
2
and partially cross-linked
polymeric structures, such as polyvinylene carbonate.
17,41,47,48,70
Additionally, it has to be mentioned, that the separation and
detection of CO
2
is not possible with the used GC column system. A
suitable approach is the systematic gas analysis using different GC
columns to enable a good separation of the different gases and, thus,
to allow for a clear detection and quantification of the individual
gases. However, these comprehensive analyses will be addressed in
future research studies.
Figure 11 depicts a representative voltage resolved GC-BID
chromatogram of the gas formed within the formation procedure in
FEC-containing cells. In agreement with the reference electrolyte
data, the strongest increase in intensities is observed during the
charging step from 3.1 to 3.4 V, while the intensities stay almost
constant afterwards up to 4.3 V. Based on reported literature data, a
strong decrease in intensity for light hydrocarbons (i.e., C
2
H
4
) and
CO by the addition of FEC is assumed and confirmed by this
method, whereas an increase of CO
2
(decomposition of FEC) is
foreseen, but has to be confirmed using a different GC-system in
consecutive works.
When comparing both chromatograms (cf. Fig. 9vs Fig. 11), as
expected, C
2
H
4
, CO, and C
2
H
6
are effectively decreased, compared
to the reference electrolyte,
33,37,39,47,66
as they are formed by
reductive decomposition of EC, while H
2
(likely originating from
reduction of trace water),
37
N
2
and O
2
(sampling contaminants) stay
almost constant, since these aspects remain consistent for both
electrolyte formulations. Furthermore, based on this measurement,
the total amount of formation gas evolved within these cells could be
roughly estimated by measuring the Ar intensity during operation.
When comparing the Ar signals in both chromatograms (cf. Fig. 9vs
Fig. 11), intensities within the tap charge at 1.5 V were comparable
for both electrolyte formulations (with and without FEC). However,
when increasing cell voltage, the Ar intensities for FEC-containing
electrolyte remain at a higher level compared to the reference
electrolyte. Since the total amount of Ar is assumed to remain almost
constant (200 μl Ar matrix) during gas sampling, a decrease in
intensity, as seen in Fig. 9(3.1 V to 3.4 V), can be associated with a
dilution of the matrix gas (Ar) due to the evolving formation gases
and vice versa. Based on these findings, the addition of FEC to the
reference electrolyte greatly reduces the amount of gas formed
Figure 8. Cell voltage vs cell capacity plot for the modified formation cycle
of NCM-811 ∣∣ AG pouch cells. Gas sampling is performed after a 10 min
CV step at the desired cell voltage and indicated by the dashed lines. Cell
voltage range: 1.5 V to 3.5 V; operation temperature: 20 °C.
Journal of The Electrochemical Society, 2020 167 060516
during formation (cf. Fig. 9vs Fig. 11). The exact amount of gas
generated during the formation cycle was measured by means of the
well-established AISGA method.
In Fig. 12, the total formation gas volume at different cell
voltages for the pouch cells during the formation cycle applying the
reference electrolyte and the 2.0 wt% FEC containing electrolyte are
shown. Subsequent to electrolyte filling, the total cell volume was
measured by means of AISGA. Interestingly, after the 20 h wetting
step at 1.5 V, the total cell volume decreases, which might be related
to the wetting process, during which the active material and
separator pores are filled with electrolyte, leading to a decrease in
“free”electrolyte volume and, therefore, a consecutive decrease in
total cell volume. However, further physical processes, i.e., electrode
swelling, may also influence the total cell volume. Certainly, this
observation is independent of the electrolyte composition, as no
difference between the additive-containing electrolyte and the
reference electrolyte can be seen.
71
In the subsequent first charge to 3.1 V, a gas volume of ≈120 μl
is generated in reference electrolyte-containing cells, while only ≈15
μl of gas is generated in FEC-containing cells. As the addition of
FEC prevents the reductive decomposition of EC to a large extent
(cf. Fig. 10), the total gas volume is greatly reduced and is in good
agreement with the GC-BID results and the literature.
33
Upon a
consecutive charge to 3.4 V, an immense increase of ≈300% in total
gas volume is detected for both electrolyte formulations, i.e., ≈350
and ≈67 μl, respectively. A further, but rather modest increase of
gas volume is measured during ongoing charging up to 3.7 V,
amounting to ≈440 μl (reference electrolyte) and ≈88 μl (FEC),
respectively.
Within the completed first charge process (up to 4.3 V) a total gas
generation of ≈480 μl in cells using the reference electrolyte and
≈100 μl in FEC-containing cells is observed. Consequently, it can
be concluded that the majority of gas generating reactions arise at a
cell voltage below 3.7 V, considering that ≈90% of total gas volume
is already evolved until that cell voltage for both electrolyte
formulations.
Towards the fundamental understanding of the reaction me-
chanism of several (novel) electrolyte additives and electrolyte
formulations, both methods (the GSP setup and AISGA) are suitable
to either investigate the decomposition mechanism (i.e., gas evol-
ving decomposition) by GC-BID, or to quantify the gas volume
formed by electrolytes/additives. Especially, additive-induced gas
evolution during storage or long-term charge/discharge cycling
should be greatly reduced for practical LIB cells. Therefore,
in situ gas formation analysis combined with AISGA is a powerful
approach to achieve a fundamental understanding for the effective-
ness of novel electrolyte formulations in terms of their gassing
behavior.
Conclusions
Within this work, we introduced a gas sampling port (GSP) into a
commercial small-scale multilayer LIB pouch cell in a simple post-
production process and systematically evaluated this cell setup
towards practicality for effective electrolyte additive research. The
Figure 9. GC-BID chromatogram of the gases formed during the formation cycle at different cell voltages, using the reference electrolyte (1 M LiPF
6
in EC:
EMC, 3:7 by wt.).
Figure 10. Differential cell capacity vs cell voltage for the first formation
cycle for NCM-811 ∣∣ AG pouch cells without and with 2 wt% FEC (1 M
LiPF
6
in EC:EMC, 3:7 by wt.; cell voltage range: 1.5 V to 3.5 V; operation
temperature: 20 °C).
Journal of The Electrochemical Society, 2020 167 060516
GSP-based LIB pouch cell design allows the analysis of the gassing
mechanism by in situ gas formation studies under commercially
relevant conditions, i.e., by use of a limited electrolyte amount per
cell capacity.
A sufficient gas-tightness, no significant impact of the GSP on
the electrochemical performance of the LIB pouch cell and the
suitable application in electrolyte additive research could be
demonstrated. In this proof-of-principle study, we evaluated an
NCM-811 ∣∣ artificial graphite (AG) LIB cell chemistry in combina-
tion with a state-of-the-art carbonate-based electrolyte, i.e., 1 M
LiPF
6
in EC:EMC (3:7 by wt.) without or with 2.0 wt% FEC as
additive. A clear correlation between the electrolyte reduction peaks
within the dQ/dVprofiles and the onset of gas evolution was
observed for the first charge/discharge cycle. Furthermore, the
practicality of the GSP setup was investigated with the use of FEC
as well-known additive and subsequently compared to literature
dealing with gas formation analysis. Simultaneously to the in situ
gas analysis, gas formation and subsequent gas consumption/fixa-
tion, i.e., CO
2
reduction, can occur in these cells. A time-dependent
gas analysis can elucidate the complex working mechanism of novel
electrolyte additives and/or mixtures and, thus, will be the focus of
future works on self-designed electrolyte additives.
In summary, the novel GSP-based pouch cell setup benefits from
the possibility of multiple, time-, cell voltage- or state-of-charge
(SOC)-dependent gas measurements, without significantly influen-
cing the cell performance as only minor gas amounts need to be
extracted. Furthermore, the developed GSP system can be adopted to
different pouch-type battery cells and used in online electrochemical
mass spectrometry (OEMS) or differential electrochemical mass
spectrometry (DEMS). A clear advantage of this cell system that it is
very close to commercially relevant conditions, i.e., using a limited
amount of electrolyte per cell capacity, however, at the detriment of
missing separation of gases evolving at the anode and cathode.
Acknowledgments
The authors of WWU Münster wish to thank the BMW Group and
CATL for financial support. Furthermore, we highly appreciate
financial support by the Federal Ministry of Education and
Research (BMBF) within the project OptiZellForm (03XP0071B)
within the ProZell cluster. In addition, we greatly acknowledge
graphical support by Andre Bar.
ORCID
Jan-Patrick Schmiegel https://orcid.org/0000-0002-9624-1364
Fabian Horsthemke https://orcid.org/0000-0003-3267-0458
Martin Winter https://orcid.org/0000-0003-4176-5811
Tobias Placke https://orcid.org/0000-0002-2097-5193
References
1. M. Winter, B. Barnett, and K. Xu, Chem. Rev.,118, 11433 (2018).
2. Y. Liang et al., InfoMat.,1, 6 (2019).
3. J. Betz, G. Bieker, P. Meister, T. Placke, M. Winter, and R. Schmuch, Adv. Energy
Mater.,9, 1803170 (2019).
4. T. Placke, R. Kloepsch, S. Dühnen, and M. Winter, J. Solid State Electrochem.,21,
1939 (2017).
5. R. Schmuch, R. Wagner, G. Hörpel, T. Placke, and M. Winter, Nat. Energy,3, 267
(2018).
Figure 12. Cell volume change vs sampling spot for NCM-811 ∣∣ AG pouch
cells during the formation cycle using the reference electrolyte (1 M LiPF
6
in
EC:EMC, 3:7 by wt.) and the 2.0 wt% FEC-containing electrolyte.
Figure 11. GC-BID chromatogram of the gases formed within the pouch cells during the formation procedure at different cell voltages, using the reference
electrolyte (1 M LiPF
6
in EC:EMC, 3:7 by wt.) +2.0 wt% FEC.
Journal of The Electrochemical Society, 2020 167 060516
6. X. Zeng, M. Li, D. Abd El-Hady, W. Alshitari, A. S. Al-Bogami, J. Lu, and
K. Amine, Adv. Energy Mater.,9, 1900161 (2019).
7. R. Wagner, N. Preschitschek, S. Passerini, J. Leker, and M. Winter, J. Appl.
Electrochem.,43, 481 (2013).
8. R. Jung, M. Metzger, F. Maglia, C. Stinner, and H. A. Gasteiger, J. Electrochem.
Soc.,164, A1361 (2017).
9. D. Andre, S.-J. Kim, P. Lamp, S. F. Lux, F. Maglia, O. Paschos, and B. Stiaszny,
J. Mater. Chem. A,3, 6709 (2015).
10. H.-H. Ryu, K.-J. Park, C. S. Yoon, and Y.-K. Sun, Chem. Mater.,30, 1155 (2018).
11. H.-J. Noh, S. Youn, C. S. Yoon, and Y.-K. Sun, J. Power Sources,233, 121 (2013).
12. J. Kasnatscheew, S. Röser, M. Börner, and M. Winter, ACS Appl. Energy Mater.,2,
7733 (2019).
13. M. Winter and J. O. Besenhard, Handbook of Battery Materials (WILEY-VCH
Verlag GmbH, Weinheim, Germany) 383 (1999).
14. J.-P. Schmiegel et al., J. Electrochem. Soc.,166, A2910 (2019).
15. R. Nölle, A. J. Achazi, P. Kaghazchi, M. Winter, and T. Placke, ACS Appl. Mater.
Interfaces,10, 28187 (2018).
16. L. Ma, L. Ellis, S. L. Glazier, X. Ma, Q. Liu, J. Li, and J. R. Dahn, J. Electrochem.
Soc.,165, A891 (2018).
17. X. Ma, J. E. Harlow, J. Li, L. Ma, D. S. Hall, S. Buteau, M. Genovese, M. Cormier,
and J. R. Dahn, J. Electrochem. Soc.,166, A711 (2019).
18. I. Cekic-Laskovic, N. von Aspern, L. Imholt, S. Kaymaksiz, K. Oldiges, B. R. Rad,
and M. Winter, Topics in Current Chemistry (Cham),375, 37 (2017).
19. Y. Arinicheva et al., Advanced Ceramics for Energy Conversion and Storage,
ed. O. Guillon (Elsevier, Amsterdam) p. 549 (2020).
20. B. Michalak, B. B. Berkes, H. Sommer, T. Bergfeldt, T. Brezesinski, and J. Janek,
Anal. Chem.,88, 2877 (2016).
21. I. Belharouak, G. M. Koenig, and K. Amine, J. Power Sources,196, 10344
(2011).
22. B. B. Berkes, A. Schiele, H. Sommer, T. Brezesinski, and J. Janek, J. Solid State
Electrochem.,20, 2961 (2016).
23. L. D. Ellis, J. P. Allen, L. M. Thompson, J. E. Harlow, W. J. Stone, I. G. Hill, and J.
R. Dahn, J. Electrochem. Soc.,164, A3518 (2017).
24. D. J. Xiong, T. Hynes, L. D. Ellis, and J. R. Dahn, J. Electrochem. Soc.,164, A3174
(2017).
25. H. Hahn, R. Wagner, F. Schappacher, M. Winter, and S. Nowak, J. Electroanal.
Chem.,772, 52 (2016).
26. B. Michalak, B. B. Berkes, H. Sommer, T. Brezesinski, and J. Janek, J. Phys. Chem.
C,121, 211 (2017).
27. H. Ota, Y. Sakata, A. Inoue, and S. Yamaguchi, J. Electrochem. Soc.,151, A1659
(2004).
28. B. Michalak, H. Sommer, D. Mannes, A. Kaestner, T. Brezesinski, and J. Janek,
Sci. Rep.,5, 15627 (2015).
29. M. Winter, R. Imhof, F. Joho, and P. Novák, 15th International Meeting on Lithium
Batteries (IMLB),81–82, 818 (1999).
30. M. R. Wagner, P. R. Raimann, A. Trifonova, K.-C. Moeller, J. O. Besenhard, and
M. Winter, J. Electrochem. Soc.,7, A201 (2004).
31. M. R. Wagner, P. R. Raimann, A. Trifonova, K.-C. Möller, J. O. Besenhard, and
M. Winter, Anal. Bioanal. Chem.,379, 272 (2004).
32. R. Nölle, J.-P. Schmiegel, M. Winter, and T. Placke, Chem. Mater.,32, 173 (2020).
33. C. P. Aiken, J. Xia, D. Y. Wang, D. A. Stevens, S. Trussler, and J. R. Dahn,
J. Electrochem. Soc.,161, A1548 (2014).
34. X. Ma, R. S. Young, L. D. Ellis, L. Ma, J. Li, and J. R. Dahn, J. Electrochem. Soc.,
166, A2665 (2019).
35. A. T. S. Freiberg, M. K. Roos, J. Wandt, R. de Vivie-Riedle, and H. A. Gasteiger,
J. Phys. Chem. A,122, 8828 (2018).
36. R. Jung, M. Metzger, F. Maglia, C. Stinner, and H. A. Gasteiger, J. Phys. Chem.
Lett.,8, 4820 (2017).
37. R. Bernhard, M. Metzger, and H. A. Gasteiger, J. Electrochem. Soc.,162, A1984
(2015).
38. S. L. Glazier, J. Li, A. J. Louli, J. P. Allen, and J. R. Dahn, J. Electrochem. Soc.,
164, A3545 (2017).
39. M. Metzger, B. Strehle, S. Solchenbach, and H. A. Gasteiger, J. Electrochem. Soc.,
163, A798 (2016).
40. M. Onuki, S. Kinoshita, Y. Sakata, M. Yanagidate, Y. Otake, M. Ue, and
M. Deguchi, J. Electrochem. Soc.,155, A794 (2008).
41. R. Petibon, V. L. Chevrier, C. P. Aiken, D. S. Hall, S. R. Hyatt,
R. Shunmugasundaram, and J. R. Dahn, J. Electrochem. Soc.,163, A1146 (2016).
42. X. Teng, C. Zhan, Y. Bai, L. Ma, Q. Liu, C. Wu, F. Wu, Y. Yang, J. Lu, and
K. Amine, ACS Appl. Mater. Interfaces,7, 22751 (2015).
43. R. Nölle, K. Beltrop, F. Holtstiege, J. Kasnatscheew, T. Placke, and M. Winter,
Mater. Today,32, 131 (2020).
44. M. N. Obrovac and V. L. Chevrier, Chem. Rev.,114, 11444 (2014).
45. J. Liu et al., Nat. Energy,4, 180 (2019).
46. R. Petibon, L. M. Rotermund, and J. R. Dahn, J. Power Sources,287, 184 (2015).
47. R. Jung, M. Metzger, D. Haering, S. Solchenbach, C. Marino, N. Tsiouvaras,
C. Stinner, and H. A. Gasteiger, J. Electrochem. Soc.,163, A1705 (2016).
48. K. U. Schwenke, S. Solchenbach, J. Demeaux, B. L. Lucht, and H. A. Gasteiger,
J. Electrochem. Soc.,166, A2035 (2019).
49. M. Zhao, S. Kariuki, H. D. Dewald, F. R. Lemke, R. J. Staniewicz, E. J. Plichta, and
R. A. Marsh, J. Electrochem. Soc.,147, 2874 (2000).
50. S. J. An, J. Li, Z. Du, C. Daniel, and D. L. Wood, J. Power Sources,342, 846
(2017).
51. F. Horsthemke, M. Leißing, V. Winkler, A. Friesen, L. Ibing, M. Winter, and
S. Nowak, Electrochim. Acta,338, 135894 (2020).
52. R. Gauthier, D. S. Hall, T. Taskovic, and J. R. Dahn, J. Electrochem. Soc.,166,
A3707 (2019).
53. F. Horsthemke, A. Friesen, X. Mönnighoff, Y. Stenzel, M. Grützke, J. T. Andersson,
M. Winter, and S. Nowak, RSC Adv.,7, 46989 (2017).
54. F. Horsthemke, V. Winkler, M. Diehl, M. Winter, and S. Nowak, Energy Technol.,
8, 1801081 (2020).
55. K. Xu, Y. Lam, S. S. Zhang, T. R. Jow, and T. B. Curtis, J. Phys. Chem. C,111,
7411 (2007).
56. M. Winter, Z. Phys. Chem.,223, 1395 (2009).
57. G. V. Zhuang, K. Xu, H. Yang, T. R. Jow, and P. N. Ross, J. Phys. Chem. B,109,
17567 (2005).
58. Leybold GmbH, (2016), Vakuum Technologie in Print und Online—Leybold
Rottländer H., Umrath W., Voss G Editor : Leybold GmbH; Cat. No. 199
79_VA.02, https://leybold.com/de/downloads/download-von-dokumenten/
broschueren/.
59. J. Self, C. P. Aiken, R. Petibon, and J. R. Dahn, J. Electrochem. Soc.,162, A796
(2015).
60. R. Imhof, J. Electrochem. Soc.,145, 1081 (1998).
61. B. Zhang, M. Metzger, S. Solchenbach, M. Payne, S. Meini, H. A. Gasteiger,
A. Garsuch, and B. L. Lucht, J. Phys. Chem. C,119, 11337 (2015).
62. L. Bai, J. Smuts, P. Walsh, H. Fan, Z. Hildenbrand, D. Wong, D. Wetz, and
K. A. Schug, J. Chromatogr. A,1388, 244 (2015).
63. D. Aurbach (ed.), Nonaqueous Electrochemistry (Marcel Dekker Inc, New York,
Basel) (1999).
64. R. Mogi, M. Inaba, Y. Iriyama, T. Abe, and Z. Ogumi, J. Power Sources,119–121,
597 (2003).
65. J.-S. Shin, C.-H. Han, U.-H. Jung, S.-I. Lee, H.-J. Kim, and K. Kim, J. Power
Sources,109, 47 (2002).
66. D. Aurbach, Y. Ein-Eli, B. Markovsky, A. Zaban, S. Luski, Y. Carmeli, and
H. Yamin, J. Electrochem. Soc.,142, 2882 (1995).
67. H. Yoshida, T. Fukunaga, T. Hazama, M. Terasaki, M. Mizutani, and M. Yamachi,
J. Power Sources,68, 311 (1997).
68. D. Y. Wang and J. R. Dahn, J. Electrochem. Soc.,161, A1890 (2014).
69. D. J. Xiong, L. D. Ellis, R. Petibon, T. Hynes, Q. Q. Liu, and J. R. Dahn,
J. Electrochem. Soc.,164, A340 (2017).
70. A. L. Michan, B. S. Parimalam, M. Leskes, R. N. Kerber, T. Yoon, C. P. Grey, and
B. L. Lucht, Chem. Mater.,28, 8149 (2016).
71. J. B. Habedank, F. J. Günter, N. Billot, R. Gilles, T. Neuwirth, G. Reinhart, and
M. F. Zaeh, Int. J. Adv. Manuf. Technol.,102, 2769 (2019).
Journal of The Electrochemical Society, 2020 167 060516
Content uploaded by Jan-Patrick Schmiegel
Author content
All content in this area was uploaded by Jan-Patrick Schmiegel on Apr 13, 2020
Content may be subject to copyright.