ArticlePDF Available

Thermal Runaway of a Li-Ion Battery Studied by Combined ARC and Multi-Length Scale X-ray CT

IOP Publishing
Journal of The Electrochemical Society
Authors:

Abstract and Figures

Lithium ion battery failure occurs across multiple length scales. In this work, the properties of thermal failure and its effects on electrode materials were investigated in a commercial battery using a combination of accelerating rate calorimetry (ARC) and multi-length scale X-ray computed tomography (CT). ARC measured the heat dissipated from the cell during thermal runaway and enabled the identification of key thermal failure characteristics such as onset temperature and the rate of heat generation during the failure. Analysis before and after failure using scanning electron microscopy (SEM) and X-ray CT were performed to reveal the effects of failure on the architecture of the whole cell and microstructure of the cathode material. Mechanical deformations to the cell architecture were revealed due to gas generation at elevated temperatures (>200 °C). The extreme conditions during thermal runaway caused the cathode particles to reduce in size by a factor of two. Electrode surface analysis revealed surface deposits on both the anode and cathode materials. The link between electrode microstructure and heat generation within a cell during failure is analysed and compared to commercially available lithium ion cells of varying cathode chemistries. The optimisation of electrode designs for safer battery materials is discussed.
This content is subject to copyright. Terms and conditions apply.
Thermal Runaway of a Li-Ion Battery Studied by Combined ARC
and Multi-Length Scale X-ray CT
Drasti Patel,
1
James B. Robinson,
1,2
Sarah Ball,
3
Daniel J. L. Brett,
1,2
and
Paul R. Shearing
1,2,
*
,z
1
Electrochemical Innovation Laboratory, Department of Chemical Engineering, University College London, WC1E 7JE,
United Kingdom
2
The Faraday Institution, Harwell Science and Innovation Campus, Didcot, OX11 0RA, United Kingdom
3
Johnson Matthey Battery Materials, Oxford Science Park, Oxford, OX4 4GB, United Kingdom
Lithium ion battery failure occurs across multiple length scales. In this work, the properties of thermal failure and its effects on
electrode materials were investigated in a commercial battery using a combination of accelerating rate calorimetry (ARC) and
multi-length scale X-ray computed tomography (CT). ARC measured the heat dissipated from the cell during thermal runaway and
enabled the identication of key thermal failure characteristics such as onset temperature and the rate of heat generation during the
failure. Analysis before and after failure using scanning electron microscopy (SEM) and X-ray CT were performed to reveal
the effects of failure on the architecture of the whole cell and microstructure of the cathode material. Mechanical deformations to
the cell architecture were revealed due to gas generation at elevated temperatures (>200 °C). The extreme conditions during
thermal runaway caused the cathode particles to reduce in size by a factor of two. Electrode surface analysis revealed surface
deposits on both the anode and cathode materials. The link between electrode microstructure and heat generation within a cell
during failure is analysed and compared to commercially available lithium ion cells of varying cathode chemistries. The
optimisation of electrode designs for safer battery materials is discussed.
© 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/ab7fb6]
Manuscript submitted December 19, 2019; revised manuscript received February 6, 2020. Published April 2, 2020. This paper is
part of the JES Focus Issue on Battery Safety, Reliability and Mitigation.
With high energy densities and favourable ageing characteristics,
the rechargeable Li-ion battery is rapidly gaining traction for high-
energy applications such as power sources in electric vehicles.
1
However, when subjected to unfavourable conditions, Li-ion cells
are liable to undergo failure owing to their highly energetic and
ammable constituent materials;
2
this has been demonstrated
recently by several high-prole failures.
34
As such, improvements
to their safety is critical, especially with the increasingly demanding
range of applications proposed.
5,6
Overcharging or internal short circuits (ISCs) can lead to battery
temperatures well above manufacturer ratings.
7,8
At a critical
temperature, a chain of exothermic reactions can be triggered and
the battery will undergo rapid self-heating. If the battery is
inadequately designed to dissipate the heat generated, it will
catastrophically fail. This failure mechanism is known as thermal
runaway and can be characterised across multiple length scales: from
electrode microstructure to the whole cell and pack level.
910
Several investigations to predict battery failure behaviour and the
thermal runaway mechanism have been conducted both on the
component and cell level, using accelerating rate calorimetry (ARC).
Based on kinetic parameters determined from ARC tests, Dahn
et al.
11
developed models to study reaction mechanisms and thermal
stabilities of small amounts of various materials at a component
level. Kinetic parameters for the decomposition of the SEI lm layer,
the reaction between lithiated graphite and electrolyte and the
cathode decomposition reaction
12
revealed that the onset tempera-
tures and quantity of heat generated differs with electrode
composition.
13,14
Furthermore, Spotnitz and Franklin
15
summarised
the kinetics of anode, cathode and electrolyte decomposition
reactions at elevated temperatures. This was extended to a cell level
by Hatchard and Kim
16,17
who established a model to predict the
behaviour of large format cylindrical Li-ion batteries.
A number of exothermic reactions which occur at different
temperatures have been determined as major sources of heat during
failure. At 90 °C, the solid electrolyte interphase (SEI) layer begins
to decompose and self-heating is initiated.
11,18
This is closely
followed by a reaction between the anode active material with the
electrolyte and binder at approximately 100 °C. The polymeric
separator, which melts between 120 °C140 °C results in a small
decrease in cell temperature.
19
At elevated temperatures, 170 °C
235 °C,
15
the cathode active material begins to decompose and
generate oxygen.
20,23
The released oxygen may oxidise the electro-
lyte and exothermically react with the anode active material,
substantially increasing the overall temperature of the battery.
Gases (such as H
2
, CO, CH
4
,C
2
H
6
,C
2
H
4
)
21
released from these
reactions may then begin to combust at high temperatures if an
ignition source is present, resulting in a re or explosion.
To decipher the sequence and severity of these reactions, ARC
measurements of high-power Li-ion cells in an 18650 format have
been widely reported. These studies quantify failure by determining
the onset temperature (the critical temperature at which thermal
runway is initiated) and the rate of heat generation during failure.
Experiments in literature are primarily conducted on commercial
cells with cathode materials such as LiCoO
2
(LCO),
Li(Ni
x
Mn
x
Co
x
)O
2
(NMC) and LiFePO
4
(LFP)
2223
each showing
varying onset temperatures of 150 °C, 170 °C and 195 °C,
respectively.
21
Studies have also revealed that high states of charge
(SOCs) aggravate exothermic reactions, where the self-heating rate
increases exponentially with SOC.
24
For a LiNi
0.8
Co
0.15
Al
0.05
O
2
(NCA), 1 Ah capacity cell, Abraham et al.
25
reported self-heating
initiated at the anode at 84 °C, and surface deposits on both the
anode and cathode surfaces were found using X-ray photoelectron
spectroscopy (XPS). Changes to the crystal structures, i.e. evidence
of oxygen loss, were not identied.
In addition to the thermodynamic stability of components and
cell size, the composition and morphology of the electrode materials
directly inuence cell behaviour during failure.
13,14,21,26
Surface
dependent features such as specic surface area play a large role in
determining the degree of heat generation during failure.
27
Smaller
particles have been shown to release oxygen at a higher rate during
decomposition, corresponding to an increase in heat generation, and
greater loss of mass during failure. At high temperatures the
decomposition rate of the cathode increases linearly with great
particle surface area.
28
Jiang and Dahn
29
have demonstrated a
z
E-mail: p.shearing@ucl.ac.uk
*Electrochemical Society Member.
Journal of The Electrochemical Society, 2020 167 090511
reduction in particle size of LiCoO
2
from 5 μm to 0.8 μm resulted in
a lower onset temperature.
The fast kinetics of the exothermic reactions leading up to
thermal runaway (within hundreds of seconds) require high-speed
techniques to record developments within a battery during failure.
High speed operando X-ray imaging has previously been used to
track structural deformations leading up to and during thermal
runway in 18650 format cells.
9,10
Additionally, X-ray CT with a
multi-length scale approach has been used to analyse Li-ion battery
materials
30
and demonstrated as an effective diagnostic tool for cells
during.
31,32
and after failure.
10
In this work, ARC in combination with lab-based multi-length
scale X-ray CT imaging is used to understand the failure of a
commercial Li-ion battery. Image based analysis is used to quantify
the changes in particle morphologies of the cathode, before and after
failure, to investigate the effect of this design feature on battery
safety.
Experimental
Thermal runaway.A commercial 2.2 Ah 18650 battery
LGDAS31865 P313K083A6 LG Chem, South Korea), consisting of a
Li(Ni
0.6
Mn
0.2
Co
0.2
)O
2
cathode, polymer separator and graphite
anode, was fully charged to the maximum rated voltage, 4.2 V at
1 C via a constant-current constant-voltage (CCCV) charging
protocol. Prior to conducting the ARC experiments, the OCV was
checked to ensure no capacity fade had occurred. The thermal abuse
test was performed using ARC inside a calorimeter (Phitec Battery
Test Calorimeter, HEL Group, Herts., UK) using the heat-wait-
search method. Once the battery had reached a start temperature
(25 °C) the following procedure was initiated: the calorimeter
increased the temperature in discrete steps (5 °C), waited for any
thermal transients to decay and subsequently monitored the battery
temperature. If in this time, the temperature remained unchanged, up
to a threshold value (0.02 °C min
1
), the calorimeter continued to
increase the temperature by 5 °C until self-heating was detected.
Multi-scale lab-based X-ray tomography.Tomographic recon-
structions of varying length scales and sample sizes were produced
using three different lab-based X-ray CT systems; a Nikon XT 225
(Nikon Metrology, Tring, UK, a Zeiss Xradia Versa 510 and a Zeiss
Xradia Ultra 810 (Carl Zeiss XRM, Pleasanton, CA, USA).
Pre- and post-failure X-ray CT images of the whole battery were
captured using a Nikon XT 225: the geometric conguration of
the radiographic scans resulted in a pixel resolution of 38 μm and
35 μm, for the pre- and post-failure scans, respectively. An
accelerating voltage of 210 kV with a tungsten target was used to
generate 3176 projections. The acquired datasets were reconstructed
using CT Pro 3D software with a built in ltered back projection
algorithm with subsequent data was visualised using Avizo Fire 9.2.
Prior to the pre- and post-failure X-ray micro-CT scans and SEM,
the cylindrical cell casing was removed using a pipe cutter inside an
argon glovebox. The spiral wound layers were carefully unrolled and
the separator, positive and negative electrodes were isolated from
one another. The electrodes were washed with dimethyl carbonate to
remove any electrolyte and left for two days to dry. The fresh cell
was discharged to 0 V before being dismantled. Electrode samples
were taken from the centre of the unrolled cell. SEM samples were
prepared by cutting a 5 mm electrode square while micro-CT
samples were prepared by cutting a 3 mm triangle from the corner
of an electrode and attaching (5-minute epoxy, ITW Devcon, USA)
it to the end of a pin attached to the sample holder. Details of the
parameters used to capture the microscale images and the pixel
resolutions achieved are summarised in Table I.
Nanoscale post-failure analysis of the positive electrode was
imaged using an X-ray nano-CT system (Zeiss Xradia Ultra 810).
Electrode samples for nanoscale characterisation were prepared
using a laser micro-machining procedure to achieve the desired
sample size.
33
A resolution of 63.1 nm was achieved using a Cr
target with an accelerating voltage of 35 kV and a tube current of
25 mA. Further details of the image scanning parameters are
summarised in Table I. For both micro- and nano-scale CT,
tomographic reconstruction was achieved using the XM
Reconstructor software suite (Zeiss, Pleasanton, CA, USA).
Reconstructed data was processed using Avizo Fire 9 software
(FEI VSG, France). After reconstruction a non-local means lter was
applied to the images to reduce noise and preserve phase boundaries
as shown in Fig. 1. Phases were separated based on their grey scale
values, where weakly attenuating materials are displayed in shades
of grey and highly attenuating materials in white. Binary images and
volume renderings as shown in Fig. 2, are subsequently used for
measurements of porosity and PSD using Avizo Fire 9s label
analysis tool. Image processing in the context of lithium ion battery
materials requires preprocessing to enhance particular features by
using a set of lters. This procedure is often performed to prepare
data for threshold-based binarization. The second step is to
distinguish phases by assigning each voxel to a particle or pore
phase based on the measured grayscale value. This is often sensitive
to variations such as enhanced edge contrast and so the choice of
thresholding approach becomes important for particular data sets and
their features. Detailed descriptions of various image analysis
techniques are provided specically for lithium ion battery elec-
trodes in the following reviews.
34,35
Analysis programmes such as
ImageJ (National Institutes of Health, USA) or Drishti (https://
github.com/nci/drishti) provide suitable open source image segmen-
tation routines for battery electrode materials.
36
All SEM micrographs are obtained from a Zeiss EVO 10 SEM,
using the SE1 signal at 12 kV accelerating voltage with a
magnication of approximately 5,000 yielding a pixel sizes ranging
between 5769 nm for all images.
Results and Discussion
Accelerated rate calorimetry was performed to initiate thermal
runway in a commercial 18650 cell. The temperature proles
obtained for the cell are shown in Fig. 3. The initial exothermic
event observed in Fig. 3a between 30 °C50 °C is a result of the
breakdown and reformation of the SEI layer,
37
while the second
visible exotherm at 175 °C indicates the onset of thermal runaway. It
is possible that the discontinuity between the two exotherms is a
result of the cell rupturing during failure, initiating the Joule-
Thompson effect, as previously reported in literature.
26
Further
evaluation of the thermal failure process reveals the reaction rate of
initial degradation and self-heating rate of thermal runway to be
7.82 ×10
3
min
1
and 8.5 ×10
2
min
1
, respectively. The plateau
after the initial SEI layer exotherm observed in Fig. 3b, may be
attributed to the shutdown separator incorporated into the cell
design. The tri-layer composite nature of the polypropylene (PP)
melting at approximately 135 °C and polyethylene (PE) at 165 °C,
arranged as PPPEPP, is shown to have delayed the onset of thermal
runway by approximately 1,750 min. This is outlined further in
Fig. 3a, at 165 °C, where the separator melts and possible rupture of
the cell may have occurred. It is considered that the thermal runaway
is only delayed, in this case, due to the ARC operating in the heat-
wait-search mode.
38
While the self-heating rate determined from this study reveals
important information regarding heat dissipation within the cell,
further studies are required to compare the inuence of different cell
chemistries
26
and designs. A cylindrical cell with tightly wound
layers (like an 18650 format) will have a different heat dissipation
time-constant when compared to a pouch cell of equivalent capacity.
Robinson et al. have previously reported a thermal runaway rate of
0.213 min
1
using ARC with similar parameters for a 2.5 Ah pouch
cell.
38
NMC is amongst the most widely used cathode materials to date,
and its demand is projected to plateau until 2025.
39
Results from this
work correlate well with literature values for commercial 18650 cells
with similar cathode chemistries and capacities.
26
While these
Journal of The Electrochemical Society, 2020 167 090511
comparisons offer valuable insights into the thermal runaway
process regarding the onset temperature, and rate of heat generation,
the specic effects of whole cell structure and electrode morphology
on this behaviour are not well understood. Understanding heat
dissipation at the whole cell level as well as a component level are
essential for accurately comparing the thermal runaway mechanism
in commercial Li-ion batteries.
Macroscale X-ray CT results of the fully charged cell pre- and
post-failure are shown in Fig. 4. The non-destructive nature of the
technique enables visualisation of the architectural changes to the
structure of the cell after failure. A 3D reconstruction of the cell
prior to failure is presented in Fig. 4a; where orthogonal slices in the
(X, Z) and (Y, Z) planes are also shown, similarly orthogonal slices
of the failed cell are shown in Fig. 4b for comparison. Figure 4b
shows the propagation of gas formation due to the decomposition of
the SEI layer during failure. For example, Arrow 1, highlights
possible delamination of an electrode layer where a gas pocket has
formed.
40
Arrows 2 and 3 show changes both to the outer casing of
the cell as well as the arrangement of electrode layers. As the
internal cell pressure increased, the ridge depicted by arrow 2, has
expanded and changes to the tightly wound electrode layers are
visible.
At 130 °C, the positive electrode reacts with electrolyte and
oxygen loss from the cathode is initiated. At temperatures above
200 °C thermal decomposition of the cathode occurs, where changes
to the crystal structure begin and heat, in addition to CO
2
and H
2
O
are released.
41
The effects of this phenomenon can be visualised in
Fig. 4b where the active cathode material has delaminated from the
aluminium current collector. Furthermore, changes to the outer
casing, particularly at the positive terminal are shown, although
there is expansion, there were no visible ruptures to the cell during
post-failure analysis. Similarly, Finegan et al. reported that the cell
casing and positive cap of a Li(Ni
0.33
Mn
0.33
Co
0.33
)O
2
2.6 Ah 18650
cell at 100% SOC remained intact as it experienced thermal
runaway.
9
Although gas formation within the cell increases the
internal pressure, it does not necessarily lead to cell rupture. In this
case, the internal cell structure collapsed on itself as observed in
Fig. 4b. A cylindrical mandrel at the centre of the electrode layers
has been reported to play an important role in the mechanical failure
of the cell; providing both mechanical strength and a route for gasses
to reach the vent during failure.
4
For microscale, pre- and post-failure analysis, the cells were
disassembled and samples of the electrode layers were extracted as
close to the centre of the cell as possible. The cathode of the failed
cell was noted to be brittle with weak adhesion to the aluminium
current collector, in contrast to the fresh cathode, but in a similar
manner to that reported by both Finegan et al. & Robinson et al.
10,38
The fresh and failed anodes on the other hand, were both relatively
intact macroscopically. Figure 5shows the SEM images obtained of
the anode (a, b) and cathode (c, d) before and after failure. Changes
to the anode surface are evident. Continuous exothermic decom-
position and reformation of the SEI layer is believed to occur up to
approximately 220 °C.
42
Reaction products such as lithium-alkyl
carbonates and lithium carbonate species make up the small
precipitates of a surface lm that are visible on the carbon particles.
Exfoliation of the graphitic phase can also be seen. Although the
particle size seems to have remained the same after failure, the
porosity is observed to have reduced, in contrast to the observations
by Robinson et al. in hard carbon electrodes in Na-ion batteries.
38
SEM images of the cathode after failure show a rougher surface than
compared to the fresh electrode; this could be attributed to the
delamination and agglomeration of Co containing species as
suggested by Finegan et al.
10
Furthermore, it suggests that the
temperature at the centre of the cell reached >350 °C. This cannot be
Table I. Summary of the multi-scale X-ray CT scanning parameters used in this work.
Macroscale Microscale
Nanoscale
Parameter Pre-failure Post failure Pre-failure Post failure Post failure
Voltage (kV) 210 80 100 35
Projections 3176 1601 1301 1601
Exposure time (s) 1 38 35
Pixel size (μm) 38 35 0.48 0.37 0.0631
Figure 1. Orthogonal slices obtained using nano-scale CT of the failed cathode showing (a) unltered image and (b) image after applying a non-local means
lter.
Journal of The Electrochemical Society, 2020 167 090511
Figure 2. Volume renderings of (a) bulk cathode structure, (b) segmented bulk cathode structure, and (c) individual cathode particles after label analysis.
Figure 3. ARC self-heating rate proles of a commercial 18650 cell. (a) The start temperature of 50 °C was increased step-wise by 5 °C. Onset temperature of
175 °C is shown. (b) Plateau in self-heating rate can be attributed to the shutdown separator.
Journal of The Electrochemical Society, 2020 167 090511
seen in the ARC prole since the temperature is recorded by
thermocouples attached to the outer casing of the cell only. While
the electrode thickness remained the same after failure, particle sizes
are visibly reduced in the volume reconstructions and PSD sug-
gesting an increase in specic surface area and consequently a higher
rate of heat generation.
43
The onset temperature for oxygen evolution is highest for
particles with smaller diameters.
28
Under abusive conditions,
the liberation of large amounts of oxygen is undesirable, therefore
nding an optimum particle size distribution that minimises the
cathode decomposition rate is critical. The cathode particles are
expected to reduce both in mass and volume after undergoing
Figure 4. X-ray macro-CT results of a commercial 18650 cell (a) showing a 3D reconstruction of the whole cell and orthogonal slices in the XY and YZ planes
before thermal runaway and (b) showing orthogonal slices in the XY and YZ planes after thermal runaway using ARC. Arrows depict areas of deformation
within cell architecture.
Figure 5. SEM images of (a) fresh, and (b) failed negative electrode from the commercial 18650 cell, (c) fresh, and (d) failed positive electrode after thermal
runaway has occurred using ARC.
Journal of The Electrochemical Society, 2020 167 090511
thermal runaway.
10
A PSD shown in Fig. 6, comparing tomography
data of the fresh and failed cathode reveals there is a signicant
reduction in the mean diameter of the post-failure particles,
approximately half their original size (from 9.6 to 6.5 μm). The
sub-volumes used for the PSD of each sample were 86,132 μm
3
and
41,612 μm
3
for the fresh and failed samples, respectively. The
spread of data for the post-failure particles differs greatly from the
fresh sample. For example, there is a signicant shift in the peak
diameter from approximately 10 μmto5μm. Furthermore, the
porosity of the failed sample was almost half of the fresh sample; i.e.
0.24 and 0.44, suggesting particle cracking and mass loss may have
occurred. An evaluation of whether the considered sample volume
fully encompasses all heterogeneities in the material is
necessary.
44,45
In this work, only a sample taken from the centre
most point of the cell is analysed. Since it cannot be assumed that the
cell reacts in a uniform way,
9
comparing pre- and post-failure
samples from separate points in the cell may give a more
comprehensive understanding of electrode morphology changes.
Furthermore, a representative volume element (RVE) analysis may
be performed on the extracted sub-volumes to dene the minimum
volume which statistically represents the cathode material.
46,47
The links between cathode microstructure and thermal failure
behaviour can be analysed further at ner lengths to accurately
identify features which may not be visible at lower resolutions. In
this study, X-ray nanoscale CT providing a pixel resolution of
63.1 nm, which enabled ne structural features to be seen. Micro-
cracks and fractures within the failed cathode particles are shown in
Figs. 7c7d. Cracks in the particles increase the surface area
available for reactions with the electrolyte during failure.
Additionally, the presence of smaller cracked particles may result
Figure 6. Particle size distribution for particles from the fresh (top) and failed cell (bottom) with complementing XZ orthogonal slices obtained from X-ray
micro-CT. For the PSD analysis a sub-volume was extracted from each electrode sample.
Figure 7. Multi-scale X-ray CT results of a commercial 18650 cell after thermal failure induced by ARC showing (a) an orthogonal slice in the XY plane where
gas generation and delamination of the electrode layers is visible, (b) a microscale orthogonal slice of the failed cathode in the XZ plane, and (c) showing a
nanoscale orthoslice of the failed cathode highlighting cracked particles with (d) showing a 3D reconstructed sub-volume.
Journal of The Electrochemical Society, 2020 167 090511
in a lower onset temperature
29
and accelerated thermal runaway rate.
The application of ARC in combination with X-ray CT imaging has
been successfully demonstrated in this work. The safetyand
reliability of a commercial lithium ion battery has been characterised
by determining the onset temperature and rate of heat generation. In
addition, X-ray CT imaging at a nanoscale has revealed features
such as particle cracks that could be useful for optimising electrode
manufacturing processes for safer batteries with lower onset
temperatures and heat generation rates.
Conclusions
The thermal failure of a commercial Li-ion battery with an 18650
format was analysed across multiple length scales. ARC was used to
induce thermal runaway and the temperature at which the SEI layer
breakdown occurs (30 °C50 °C) and the onset temperature (175 °C)
was revealed. It was found that the melting of the shutdown
separator signicantly delayed the time taken to reach the onset
temperature.
Pre- and post-failure examination of the cell using lab-based
X-ray CT imaging revealed severe deformation of the electrode
layers. Image analysis techniques applied to the bulk cathode both
pre- and post-failure demonstrated that both the porosity of and PSD
within the cathode nearly halved after failure. Nanoscale X-ray CT
was more representative of the material showing evidence of particle
cracking which is difcult to visualise in the SEM images of the
cathode surface. Changes to the anode surface were also observed,
suggesting its interaction with decomposed electrolyte may play a
role in the thermal runaway process. In this work, electrode
morphology has been considered when analysing battery failure
behaviour. While it is suggested that larger particle sizes may
increase the thermal stability, they can drastically reduce the overall
power density of a battery.
28
A critical balance between performance
and safety exits, and future work considering electrochemical
characteristics of a material alongside its contribution towards
battery failure is imperative.
This work demonstrates how multi-length scale X-ray CT
imaging can be used as a diagnostic tool to decipher the series of
events that a Li-ion battery undergoes during failure. The investiga-
tion of various microstructural parameters of the cathode material
and their possible inuence on battery failure severity have been
discussed.
Acknowledgments
This work was supported by the EPSRC (EP/N032888/1, EP/
R020973/1, EP/K005030/1, EP/M028100/1), DP acknowledges
funding from the EPSRC CASE Award scheme with Johnson
Matthey, PRS acknowledges support from the Royal Academy of
Engineering (CiET1718/59) and the Faraday Institution (Faraday.ac.
uk; EP/S003053/1), grant number FIRG001.
ORCID
Drasti Patel https://orcid.org/0000-0003-4288-2825
Paul R. Shearing https://orcid.org/0000-0002-1387-9531
References
1. Y. Miao, P. Hynan, A. von Jouanne, and A. Yokochi, Energies,12, 1074 (2019).
2. D. H. Doughty and E. P. Roth, Journal of Power Sources,128, 308 (2012).
3. J. Christman, J. Case Stud.,30, 88 (2012), http://sfcrjcs.org/index.php/sfcrjcs/
article/view/28.
4. D. P. Finegan et al., Energy Environ. Sci.,10, 1377 (2017).
5. V. Ruiz, A. Pfrang, A. Kriston, N. Omar, P. Van den Bossche, and L. Boon-Brett,
Renew. Sustain. Energy Rev.,81, 1427 (2018).
6. H. P. Jones, J. T. Chapin, and M. Tabaddor, Proceedings of the Fourth IAASS
Conference, Making Safety Matter, Huntsville, Alabama, USA ESA SP-680,
September 2010 (ESA Communications, ESTEC, Noordwijk, The Netherlands)
680 (2010).
7. P. Biensan, B. Simon, J. P. Pérès, A. De Guibert, M. Broussely, J. M. Bodet, and
F. Perton, J. Power Sources,8182, 906 (1999).
8. S. J. Harris, D. J. Harris, and C. Li, J. Power Sources,342, 589 (2017).
9. D. P. Finegan et al., Nat. Commun.,6, 6924 (2015).
10. D. P. Finegan, M. Scheel, J. B. Robinson, B. Tjaden, M. Di Michiel, G. Hinds, D. J.
L. Brett, and P. R. Shearing, Phys. Chem. Chem. Phys.,18, 30912 (2016).
11. M. N. Richard and J. R. Dahn, J. Power Sources,79, 135 (1999).
12. J. R. Dahn, E. W. Fuller, M. Obrovac, and U. von Sacken, Solid State Ionics,69,
265 (1994).
13. Y. Baba, S. Okada, and J. I. Yamaki, Solid State Ionics,148, 311 (2002).
14. S. P. Ong, A. Jain, G. Hautier, B. Kang, and G. Ceder, Electrochem. Commun.,12,
427 (2010).
15. R. Spotnitz and J. Franklin, J. Power Sources,113, 81 (2003).
16. G. H. Kim, A. Pesaran, and R. Spotnitz, J. Power Sources,170, 476 (2007).
17. T. D. Hatchard, D. D. Macneil, A. Basu, and J. R. Dahn, J. Electrochem. Soc.,148,
A755 (2001).
18. Q. Wang, J. Sun, X. Yao, and C. Chen, J. Electrochem. Soc.,153, A329 (2006).
19. X. Feng, J. Sun, M. Ouyang, X. He, L. Lu, X. Han, M. Fang, and H. Peng, J. Power
Sources,272, 457 (2014).
20. G. Gachot, S. Grugeon, G. G. Eshetu, D. Mathiron, P. Ribière, M. Armand, and
S. Laruelle, Electrochim. Acta,83, 402 (2012).
21. A. W. Golubkov, D. Fuchs, J. Wagner, H. Wiltsche, C. Stangl, G. Fauler, G. Voitic,
A. Thaler, and V. Hacker, RSC Adv.,4, 3633 (2014).
22. C. Y. Jhu, Y. W. Wang, C. Y. Wen, and C. M. Shu, Appl. Energy,100, 127 (2012).
23. W. C. Chen, Y. W. Wang, and C. M. Shu, J. Power Sources,318, 200 (2016).
24. T. Ohsaki, T. Kishi, T. Kuboki, N. Takami, N. Shimura, Y. Sato, M. Sekino, and
A. Satoh, J. Power Sources,146, 97 (2005).
25. D. P. Abraham, E. P. Roth, R. Kostecki, K. McCarthy, S. MacLaren, and
D. H. Doughty, J. Power Sources,161, 648 (2006).
26. B. Lei, W. Zhao, C. Ziebert, N. Uhlmann, M. Rohde, and H. J. Seifert, Batteries,3,
1 (2017).
27. D. D. MacNeil, J. Electrochem. Soc.,146, 3596 (1999).
28. J. Geder, H. E. Hoster, A. Jossen, J. Garche, and D. Y. W. Yu, J. Power Sources,
257, 286 (2014).
29. J. Jiang and J. R. Dahn, Electrochim. Acta,49, 2661 (2004).
30. P. R. Shearing, N. P. Brandon, J. Gelb, R. Bradley, P. J. Withers, A. J. Marquis,
S. Cooper, and S. J. Harris, J. Electrochem. Soc.,159, 1023 (2012).
31. M. Ebner, F. Marone, M. Stampanoni, and V. Wood, Science (80-.).,342, 716
(2013).
32. D. S. Eastwood et al., Adv. Energy Mater.,4, 1300506 (2014).
33. J. J. Bailey et al., J. Microsc.,267, 384 (2017).
34. B. Sankur, J. Electron. Imaging,13, 146 (2004).
35. P. Pietsch and V. Wood, Annu. Rev. Mater. Res.,47, 451 (2017).
36. E. Maire and P. J. Withers, Int. Mater. Rev.,59, 1 (2014).
37. Z. Chen, Y. Qin, Y. Ren, W. Lu, C. Orendorff, E. P. Roth, and K. Amine, Energy
Environ. Sci.,4, 4023 (2011).
38. J. Robinson, D. Finegan, T. Heenan, K. Smith, E. Kendrick, D. Brett, and
P. R. Shearing, J. Electrochem. Energy Convers. Storage,15, 1 (2017).
39. G. Berckmans, M. Messagie, J. Smekens, N. Omar, L. Vanhaverbeke, and
J. V. Mierlo, Energies,10, 1314 (2017).
40. V. Yut, P. Shearing, R. W. Hamilton, P. D. Lee, M. Wu, and N. P. Brandon,
Electrochem. Commun.,13, 608 (2011).
41. D. D. MacNeil and J. R. Dahn, J. Electrochem. Soc.,148, A1211 (2001).
42. V. Agubra and J. Fergus, Materials (Basel).,6, 1310 (2013).
43. J. Geder, H. E. Hoster, A. Jossen, J. Garche, and D. Y. W. Yu, J. Power Sources,
257, 286 (2014).
44. D. Kehrwald, P. R. Shearing, N. P. Brandon, P. K. Sinha, and S. J. Harris,
J. Electrochem. Soc.,158, A1393 (2011).
45. S. J. Harris and P. Lu, J. Phys. Chem. C,117, 6481 (2013).
46. S. J. Cooper et al., J. Power Sources,247, 1033 (2014).
47. O. O. Taiwo, D. P. Finegan, D. S. Eastwood, J. L. Fife, L. D. Brown, J. A. Darr,
P. D. Lee, D. J. Brett, and P. R. Shearing, J. Microsc.,263, 280 (2016).
48. E. Darcy, J. Power Sources,174, 575 (2007).
49. F. Tariq et al., ECS Electrochem. Lett.,3, A76 (2014).
50. H. F. Xiang, H. Wang, C. H. Chen, X. W. Ge, S. Guo, J. H. Sun, and W. Q. Hu,
J. Power Sources,191, 575 (2009).
51. T. Y. Lu, C. C. Chiang, S. H. Wu, K. C. Chen, S. J. Lin, C. Y. Wen, and C. M. Shu,
J. Therm. Anal. Calorim.,114, 1083 (2013).
Journal of The Electrochemical Society, 2020 167 090511
... Non-energetic failures such as increased cell resistance, Li + ion loss, cell swelling, electrolyte leakage, and the consequent cell dry-out are the most common failure modes for batteries. Non-energic failures show no thermal runaway but compromise the battery reliability [8]. However, off-nominal operating conditions can lead to continuous heat generation, gas release, and critical cell failures [9]. Figure 4 lists different abuse conditions, including thermal, electrical (overcharge, overdischarge, short circuit, etc.), have a preferred operation condition with a voltage range of 2.5-4.3 ...
... Non-energetic failures such as increased cell resistance, Li + ion loss, cell swelling, electrolyte leakage, and the consequent cell dry-out are the most common failure modes for batteries. Non-energic failures show no thermal runaway but compromise the battery reliability [8]. However, off-nominal operating conditions can lead to continuous heat generation, gas release, and critical cell failures [9]. Figure 4 lists different abuse conditions, including thermal, electrical (overcharge, overdischarge, short circuit, etc.), Non-energic failures show no thermal runaway but compromise the battery reliability [8]. ...
... Non-energic failures show no thermal runaway but compromise the battery reliability [8]. However, off-nominal operating conditions can lead to continuous heat generation, gas release, and critical cell failures [9]. Figure 4 lists different abuse conditions, including thermal, electrical (overcharge, overdischarge, short circuit, etc.), Non-energic failures show no thermal runaway but compromise the battery reliability [8]. However, off-nominal operating conditions can lead to continuous heat generation, gas release, and critical cell failures [9]. Figure 4 lists different abuse conditions, including thermal, electrical (overcharge, overdischarge, short circuit, etc.), and mechanical (nail penetration, crush, impact, etc.) abuses [10]. ...
Article
Full-text available
Stationary battery energy storage systems (BESS) have been developed for a variety of uses, facilitating the integration of renewables and the energy transition. Over the last decade, the installed base of BESSs has grown considerably, following an increasing trend in the number of BESS failure incidents. An in-depth analysis of these incidents provides valuable lessons for improving the safety of BESS. This paper discusses multiple safety layers at the cell, module, and rack levels to elucidate the mechanisms of battery thermal runaway and BESS failures. We further provide insights into different safety aspects of BESS, covering the system architecture, system consideration, safety standards, typical quality issues, failure statistics, and root causes. Various mitigation strategies are recommended and summarized. We highlight the importance of multi-disciplinary approaches in knowing, managing, and mitigating the risks associated with BESS. In general, this review paper serves as a guide for understanding the safety of BESS.
... 19 It also offers an efficacious solution and theoretical foundation for non-destructive testing, failure analysis, and quality control of the internal structure and defects in LIBs. 20,21 ...
... CT technology enables a comprehensive 3D visualization of battery components, facilitating the examination of electrode pore structure, particle morphology and surface degradation, dendrite formation, intra-granular cracking, and short-circuit dendrite morphology in solid electrolytes, as well as other alterations in material structure that may contribute to accelerated capacity decline and impact the operational reliability of batteries. [21][22][23][24] Paul et al. detected and quantified defective particles by using low-resolution imaging, which has greatly improved material statistics (Fig. 3). 34 Daniele et al. proposed a study on nano-CuO anodes of LIBs using a combination of electrochemical measurements and off-site x-ray CT at the nano-scale level (Fig. 4). ...
Article
As the global lithium-ion batteries (LIBs) market continues to expand, the necessity for dependable and secure LIBs has reached an all-time high. However, the use of batteries is associated with a number of significant risks, including the potential for thermal runaway and explosions. The meticulous inspection of LIBs is not only essential for guaranteeing their quality and functionality, but also for ensuring their safety. This underscores the criticality of advanced inspection technologies. In contrast to traditional inspection technologies, industrial x-ray computed tomography (CT) scanning technology affords a non-destructive comprehensive, three-dimensional insight into the interior structure of a battery without the need for disassembly. It can make the inner LIBs structures visible through the housing and even batteries already installed in devices can be examined safely and accurately without being removed or opened. This capability is of critical importance for the identification of defects that could lead to battery failure or safety issues, and guide the optimization of LIBs with better safety and performance. This perspective review briefly summarize the comprehensive application of industrial CT in LIBs including battery materials, cells and modules. Finally, we further discuss the challenges and prospects of industrial CT for energy storage.
Article
The creation of a circular economy is seen as one of the key challenges in recycling spent Li-ion batteries and would vastly diminish pressures faced in the initial extraction stage of the life cycle. Molten salts (MS) possess a set of excellent electrochemical properties and have been used to recycle metals and non-metals in the battery, metallurgical, nuclear and planetary science sectors. However, an in-depth and clear visual understanding of the electrochemical reduction process is still lacking. Here, we have overcome this challenge by developing a bespoke, miniaturised electrochemical cell enabling real-time X-ray imaging studies. A combination of X-ray radiography and tomography provide an opportunity to non-destructively reveal detailed microstructural evaluation of the electrochemical cell during the pyro-chemical process. Moreover, we have found that significant amounts of CO/CO2 accumulated at the anode surface may lead to undesired operational consequences.
Article
Lithium-ion batteries, one of the most important energy storage technologies, are widely used in portable electronic devices, electric vehicles, and energy storage systems due to their high energy density and long cycle life. However, the degradation of the batteries causes many safety hazards. The degraded batteries show some different characteristics compared with fresh ones, such as temperature distribution and evolution which can be tested and recorded by infrared (IR) thermography. This study investigates the differences in temperature distribution and average temperature evolution of degraded and fresh 18,650 Li-ion batteries at various discharging and charging currents. It is apparent that the temperature rise rate of degraded batteries is much higher than that of fresh ones in the first 30 s at a discharging current of 9 A. Also, the temperature distribution begins to change from the very beginning of the discharging, changes fast, and becomes stable before the DoD of 0.5 in the degraded batteries, while in the fresh one, it changes slowly and becomes stable after reaching a DoD of 0.5. Results show that there are obvious differences in temperature distribution and average temperature evolution between degraded and fresh batteries at high currents, while the difference at low currents is inconspicuous, which might be caused by the increase in the internal resistance in degraded batteries. Altogether, these findings suggest the potential for early detection of degraded batteries through IR thermography technology during high-current discharging.
Article
Full-text available
Over the past several decades, the number of electric vehicles (EVs) has continued to increase. Projections estimate that worldwide, more than 125 million EVs will be on the road by 2030. At the heart of these advanced vehicles is the lithium-ion (Li-ion) battery which provides the required energy storage. This paper presents and compares key components of Li-ion batteries and describes associated battery management systems, as well as approaches to improve the overall battery efficiency, capacity, and lifespan. Material and thermal characteristics are identified as critical to battery performance. The positive and negative electrode materials, electrolytes and the physical implementation of Li-ion batteries are discussed. In addition, current research on novel high energy density batteries is presented, as well as opportunities to repurpose and recycle the batteries.
Article
Full-text available
Thermal runaway is a phenomenon that occurs due to self-sustaining reactions within batteries at elevated temperatures resulting in catastrophic failure. Here, the thermal runaway process is studied for a Li-ion and Na-ion pouch cells of similar energy density (10.5 Wh, 12 Wh, respectively) using accelerating rate calorimetry (ARC). Both cells were constructed with a z-fold configuration, with a standard shutdown separator in the Li-ion and a low-cost polypropylene (PP) separator in the Na-ion. Even with the shutdown separator, it is shown that the self-heating rate and rate of thermal runaway in Na-ion cells is significantly slower than that observed in Li-ion systems. The thermal runaway event initiates at a higher temperature in Na-ion cells. The effect of thermal runaway on the architecture of the cells is examined using X-ray microcomputed tomography, and scanning electron microscopy (SEM) is used to examine the failed electrodes of both cells. Finally, from examination of the respective electrodes, likely due to the carbonate solvent containing electrolyte, it is suggested that thermal runaway in Naion batteries (NIBs) occurs via a similar mechanism to that reported for Li-ion cells.
Article
Full-text available
The negative impact of the automotive industry on climate change can be tackled by changing from fossil driven vehicles towards battery electric vehicles with no tailpipe emissions. However their adoption mainly depends on the willingness to pay for the extra cost of the traction battery. The goal of this paper is to predict the cost of a battery pack in 2030 when considering two aspects: firstly a decade of research will ensure an improvement in material sciences altering a battery’s chemical composition. Secondly by considering the price erosion due to the production cost optimization, by maturing of the market and by evolving towards to a mass-manufacturing situation. The cost of a lithium Nickel Manganese Cobalt Oxide (NMC) battery (Cathode: NMC 6:2:2 ; Anode: graphite) as well as silicon based lithium-ion battery (Cathode: NMC 6:2:2 ; Anode: silicon alloy), expected to be on the market in 10 years, will be predicted to tackle the first aspect. The second aspect will be considered by combining process-based cost calculations with learning curves, which takes the increasing battery market into account. The 100 dollar/kWh sales barrier will be reached respectively between 2020-2025 for silicon based lithium-ion batteries and 2025–2030 for NMC batteries, which will give a boost to global electric vehicle adoption.
Article
Full-text available
Lithium ion batteries are a proven technology for automotive applications and their continued use in the future electric vehicle fleet is undeniable. In addition to battery performance and durability, battery safety is paramount to ensure confidence and widespread adoption of electromobility in our society. This comprehensive review aims at presenting the various international standards and regulations for safety testing of lithium ion batteries in automotive applications under various abusive environments. Safety tests are presented and analysed including mechanical, electrical, environmental and hazards of chemical nature. The intention of this review is compiling the most relevant standards and regulations to identify shortcomings and areas for future improvement.
Article
Full-text available
A robust and versatile sample preparation technique for the fabrication of cylindrical pillars for imaging by X-ray nano-computed tomography (nano-CT) is presented. The procedure employs simple, cost-effective laser micro-machining coupled with focused-ion beam (FIB) milling, when required, to yield mechanically robust samples at the micrometre length-scale to match the field-of-view (FOV) for nano-CT imaging. A variety of energy and geological materials are exhibited as case studies, demonstrating the procedure can be applied to a variety of materials to provide geometrically optimised samples whose size and shape are tailored to the attenuation coefficients of the constituent phases. The procedure can be implemented for the bespoke preparation of pillars for both lab- and synchrotron-based X-ray nano-CT investigations of a wide range of samples.
Article
Full-text available
In this work, commercial 18650 lithium-ion cells with LiMn 2 O 4 , LiFePO 4 , and Li(Ni 0.33 Mn 0.33 Co 0.33)O 2 cathodes were exposed to external heating in an accelerating rate calorimeter (es-ARC, Thermal Hazard Technology (THT), Bletchley, UK), to investigate the thermal behavior under abuse conditions. New procedures for measuring the external and internal pressure change of cells were developed. The external pressure was measured utilizing a gas-tight cylinder inside the calorimeter chamber, in order to detect the venting of the cells. For internal pressure measurements, a pressure line connected to a pressure transducer was directly inserted into the cell. During the thermal runaway experiments, three stages (low rate, medium rate, and high rate reactions) were observed. Both the pressure and temperature change indicated different stages of exothermic reactions, which produced gases or/and heat. The onset temperature of the thermal runaway was estimated according to the temperature and pressure changes. Moreover, the different activation energies for the exothermic reactions could be derived from Arrhenius plots.
Article
Full-text available
Lithium-ion batteries are being used in increasingly demanding applications where safety and reliability are of utmost importance. Thermal runaway presents the greatest safety hazard, and needs to be fully understood in order to progress towards safer cell and battery designs. Here, we demonstrate the application of an internal short circuiting device for controlled, on-demand, initiation of thermal runaway. Through its use, the location and timing of thermal runaway initiation is pre-determined, allowing analysis of the nucleation and propagation of failure within 18650 cells through the use of high-speed X-ray imaging at 2000 frames per second. The cause of unfavourable occurrences such as sidewall rupture, cell bursting, and cell-to-cell propagation within modules is elucidated, and steps towards improved safety of 18650 cells and batteries are discussed.
Article
Full-text available
Catastrophic failure of lithium-ion batteries occurs across multiple length scales and over very short time periods. A combination of high-speed operando tomography, thermal imaging and electrochemical measurements is used to probe the degradation mechanisms leading up to overcharge-induced thermal runaway of a LiCoO2 pouch cell, through its interrelated dynamic structural, thermal and electrical responses. Failure mechanisms across multiple length scales are explored using a post-mortem multi-scale tomography approach, revealing significant morphological and phase changes in the LiCoO2 electrode microstructure and location dependent degradation. This combined operando and multi-scale X-ray computed tomography (CT) technique is demonstrated as a comprehensive approach to understanding battery degradation and failure.
Article
X-ray tomography has emerged as a powerful technique for studying lithium ion batteries, allowing nondestructive and often quantitative imaging of these complex systems, which contain solid components with length scales spanning orders of magnitude and which are in-filled with liquid electrolyte. Over the past decade, X-ray tomography has allowed interrogation of structure and material composition, providing quantitative or qualitative insight into battery operation and degradation. In this review, we first provide an overview of X-ray tomography and explore what types of experiments can yield insight into open questions in the lithium ion battery research field. In the second half of the review, we discuss the aspects a researcher must consider, and we summarize challenges and approaches to sample preparation, experimental setup, and data analysis. Finally, we describe both outstanding challenges and promise in using X-ray tomography for lithium ion battery research. Expected final online publication date for the Annual Review of Materials Research Volume 47 is July 1, 2017. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
Article
There are relatively few publications that assess capacity decline in enough commercial cells to quantify cell-to-cell variation, but those that do show a surprisingly wide variability. Capacity curves cross each other often, a challenge for efforts to measure the state of health and predict the remaining useful life (RUL) of individual cells. We analyze capacity fade statistics for 24 commercial pouch cells, providing an estimate for the time to 5% failure. Our data indicate that RUL predictions based on remaining capacity or internal resistance are accurate only once the cells have already sorted themselves into “better” and “worse” ones. Analysis of our failure data, using maximum likelihood techniques, provide uniformly good fits for a variety of definitions of failure with normal and with 2- and 3-parameter Weibull probability density functions, but we argue against using a 3-parameter Weibull function for our data. pdf fitting parameters appear to converge after about 15 failures, although business objectives should ultimately determine whether data from a given number of batteries provides sufficient confidence to end lifecycle testing. Increased efforts to make batteries with more consistent lifetimes should lead to improvements in battery cost and safety.