Thermal-runaway experiments on consumer Li-ion batteries with metal-oxide and olivin-type cathodes

Article abstract of DOI:10.1039/c3ra45748f

Li-ion batteries play an ever-increasing role in our daily life. Therefore, it is important to understand the potential risks involved with these devices. In this work we demonstrate the thermal runaway characteristics of three types of commercially available Li-ion batteries with the format 18650. The Li-ion batteries were deliberately driven into thermal runaway by overheating under controlled conditions. Cell temperatures up to 850 °C and a gas release of up to 0.27 mol were measured. The main gas components were quantified with gas-chromatography. The safety of Li-ion batteries is determined by their composition, size, energy content, design and quality. This work investigated the influence of different cathode-material chemistry on the safety of commercial graphite-based 18650 cells. The active cathode materials of the three tested cell types were (a) LiFePO₄, (b) Li(Ni_0.45Mn _0.45Co_0.10)O₂ and (c) a blend of LiCoO ₂ and Li(Ni_0.50Mn_0.25Co_0.25)O ₂.

Full text of DOI:10.1039/c3ra45748f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Thermal-runaway experiments on consumer Li-ion
batteries with metal-oxide and olivin-type
cathodes
Cite this: RSC Adv., 2014, 4, 3633
a
a
b
c
Andrey W. Golubkov,* David Fuchs, Julian Wagner, Helmar Wiltsche,
d
d
e
a
Christoph Stangl, Gisela Fauler, Gernot Voitic, Alexander Thaler
e
and Viktor Hacker
Li-ion batteries play an ever-increasing role in our daily life. Therefore, it is important to understand the
potential risks involved with these devices. In this work we demonstrate the thermal runaway
characteristics of three types of commercially available Li-ion batteries with the format 18650. The Li-ion
batteries were deliberately driven into thermal runaway by overheating under controlled conditions. Cell
ꢀ
temperatures up to 850 C and a gas release of up to 0.27 mol were measured. The main gas
components were quantified with gas-chromatography. The safety of Li-ion batteries is determined by
their composition, size, energy content, design and quality. This work investigated the influence of
different cathode-material chemistry on the safety of commercial graphite-based 18650 cells. The active
Received 11th October 2013
Accepted 26th November 2013
DOI: 10.1039/c3ra45748f
4 2
cathode materials of the three tested cell types were (a) LiFePO , (b) Li(Ni0.45Mn0.45Co0.10)O and (c) a
www.rsc.org/advances
blend of LiCoO and Li(Ni0.50Mn0.25Co0.25)O
2
2
.
quantify possible hazards of exothermic Li-ion battery over-
temperature reactions, tests with complete batteries should be
performed. Such experiments were undertaken with commer-
1
Introduction
1
Li-ion batteries have been commercially available since 1991.
3
–11
As of 2013, Li-ion batteries are in wide use for portable elec- cial Li-ion batteries produced for consumer electronics
and
12–16
tronics, such as cell phones and notebook computers. They are with Li-ion batteries fabricated in the laboratory.
also gaining traction as a power source in electried vehicles.
This work investigated the thermal stability of three types of
Li-ion batteries have a high specic energy and favourable commercially available Li-ion batteries for consumer elec-
ageing characteristics compared to NiMH and lead acid tronics. Particular attention was given to (1) the dynamics of the
batteries. However, there are concerns regarding the safety of thermal responses of the cells, (2) the maximum temperatures
Li-ion batteries. Abuse conditions such as overcharge, over- reached, (3) the amount of gases produced and (4) to the
discharge and internal short-circuits can lead to battery production rates of the gases. To further assess the hazard
temperatures far beyond the manufacturer ratings. At a critical potential of the released gases, samples were taken and ana-
temperature, a chain of exothermic reactions can be triggered. lysed with a gas chromatography system.
The reactions lead to a further temperature increase, which in
turn accelerates the reaction kinetics. This catastrophic self-
2
Experimental
accelerated degradation of the Li-ion battery is called thermal
2
2.1 Brief description of the test rig
runaway.
ꢀ
During thermal runaway, temperatures as high as 900 C can To carry out unrestricted thermal-runaway experiments, a
3
be reached, and the battery can release a signicant amount of custom-designed test stand was built (Fig. 1). The main
4
burnable and (if inhaled in high concentrations) toxic gas. To component of the test rig is a heatable reactor with electric
feedthroughs for the temperature measurement and the inner
sample heating. The reactor has gas feedthroughs that connect
it to an inert gas ask, to a gas sampling station and to a cold
trap with an attached vacuum pump. The pressure inside the
reactor is recorded by a pressure transmitter. The whole struc-
a
VIRTUAL VEHICLE Research Center, Inffeldgasse 21a, 8010 Graz, Austria. E-mail:
andrej.golubkov@alumni.tugraz.at; Fax: +43-316-873-9002; Tel: +43-316-873-9639
b
Graz Centre for Electron Microscopy, Steyrergasse 17, 8010 Graz, Austria
c
Institute of Analytical Chemistry and Food Chemistry, Graz University of Technology,
ture is hosted inside a fume hood to prevent any escaping of
Stremayrgasse 9/III, 8010 Graz, Austria
d
gases and electrolyte vapours.
Varta Micro Innovation GmbH, Stremayrgasse 9, 8010 Graz, Austria
e
A removable sample holder is placed inside the reactor. The
sample holder consists of a metal structure, which houses a
Institute of Chemical Engineering and Environmental Technology, Graz University of
Technology, Inffeldgasse 25/C/II, 8010 Graz, Austria
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 3633–3642 | 3633

View Article Online
RSC Advances
Paper
Fig. 1 (a) The reactor and its principal elements. (b) The reactor is the main component of the test stand.
heating sleeve and the thermocouples. A Li-ion battery with the the cooling trap and the vacuum-pump into the fume hood. The
dimensions 18650 (cylindrical geometry with d ¼ 18 mm and reactor and the gas tubes between the reactor and the cooling
ꢀ
l ¼ 65 mm) can be tted into the centre of the heating sleeve. trap are heated above 130 C to avoid gas condensation.
The inside wall of the heating sleeve is thermally insulated. The
By following this procedure, most liquid residue in the
role of the thermal insulation layer is to provide the thermal reactor is passed from the reactor to the cooling trap. The liquid
connection between the heating sleeve and the sample. Due to residue can be easily removed from the cooling trap before the
the low thermal conductivity of the insulation layer, a thermal next experiment run.
runaway reaction can proceed in adiabatic-like conditions. Ten
thermocouples measure the temperature at different positions
2
.3 Gas analysis
inside the reactor: three thermocouples are directly attached to
the sample housing, three thermocouples are attached to the
heating sleeve and four thermocouples measure the gas
temperature inside the reactor.
The compositions of the sampled gases were analysed using a
gas chromatograph (GC, Agilent Technologies 3000 Micro GC,
two columns, Mol Sieve and PLOTU). A thermal conductivity
detector (TCD) was used to detect permanent gases. The GC was
calibrated for H , O , N , CO, CO , CH , C H , C H and C H .
2
2
2
2
4
2
2
2
4
2
6
2.2 Testing method
Ar and He were used as carrier gases.
Initially, the sample battery is CC/CV charged to the respective
cut-off voltage. Then, the plastic envelope is removed from the
cell and the cell mass and cell voltage are recorded. Three
thermocouples are welded to the cell housing, and the whole
package is inserted into the heating sleeve of the sample holder.
The sample holder is placed inside the reactor. The reactor is
Note, that the current test set-up cannot detect HF, which
can be a major source of toxicity of gas released by Li-ion
batteries during thermal runaway.
4
2.4 Cell-components identication
evacuated and ushed with argon gas twice. The heaters are set In order to identify the components of each cell species, several
to constant power, and the pressure and temperature signals cells were disassembled: the cells were discharged to 2.0 V, and
are recorded. In order to trace fast temperature and pressure the cell casings were then carefully removed without causing
changes, each signal is recorded with a high sampling rate of short circuits. The exposed jelly rolls were subject to
5000 samples per second.
several tests.
When a critical temperature is reached, the cell goes into
For electrolyte identication, the jelly rolls were immersed in
rapid thermal runaway: it produces gas and heat. During the asks with CH
thermal runaway, the temperature of the cell increases by The solutions were then analysed using a gas-chromatography-
several hundred degree Celsius in a few seconds. Aer the mass spectrometry system (GC-MS: Agilent 7890 MS
2 2
Cl solution immediately aer casing removal.
&
thermal-runaway event, the cell cools down slowly. Gas samples 5975MSD) with the ChemStation soware and the NIST spec-
are taken and analysed with the gas chromatograph. In the next trum library. To analyse the solid materials of the cells, the
step, the vacuum pump is switched on, and the cooling trap is extracted jelly rolls were separated into the anode, cathode and

lled with liquid nitrogen. The gas is carefully released through separator layers. Aer drying in a chemical fume hood, anode
3634 | RSC Adv., 2014, 4, 3633–3642
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
and cathode-foil samples were taken for identication of the Li(Ni0.50Mn0.25Co0.25)O
2
. A clean cut through the sample was
electrochemically active materials. Microwave-assisted sample done with a focused ion beam (FIB). Subsequently, EDX
digestion followed by inductively coupled plasma optical measurements of the bulk materials of individual cathode
emission spectrometry (ICP-OES, Ciros Vision EOP, Spectro, particles were performed. The ratio of LCO and NMC layered
Germany) was used to obtain the gross atomic compositions of oxide particles was estimated by comparing the SEM-EDX and
the cathode active masses. A scanning-electron microscope with ICP-OES results. The resulting ratio of LCO and NMC was
energy-dispersive X-ray spectroscopy (SEM/EDX: Zeiss Ultra55 & LCO : NMC ¼ (66 : 34) with 5% uncertainty. The cells with LCO/
EDAX Pegasus EDX) was used to conrm the ICP-OES results for NMC blended cathodes are a compromise to achieve high rate
the compositions of the cathodes and to validate the anode capability of LCO material and to maintain acceptable safety
17
materials.
and high capacity of the NMC material. The average voltage of
For the mass-split calculation, the following procedure was this cell was ꢂ3.8 V.
followed for each cell type: positive and negative electrode
ꢁ The NMC cell had a Li(Ni0.45Mn0.45Co0.10)O layered oxide
2
samples were extracted from the jelly roll. The samples were cathode. The properties of the NMC mixed oxide cathodes
rinsed with diethyl carbonate (DEC) and then dried again, in depended on the ratios of nickel, manganese and cobalt
order to remove the remaining electrolyte residues from the material. In general, NMC cells have an average voltage of
18
active materials. The samples were weighed, and the geometries ꢂ3.8 V and high specic capacity.
of the electrode foils were recorded, so that the mass split could cathode with olivine structure.
ꢁ The LFP cell had a LiFePO
4
be calculated. The amount of electrolyte was estimated as the This cathode type is known for featuring good safety charac-
mass difference between the initial cell mass and the calculated teristics. Commercial LiFePO cathode material for high power
4
dry mass for each cell. The thickness of the active material layers Li-ion batteries consists of carbon-coated LiFePO nano-scale
4
on the electrode substrates was extracted from SEM images. The particles. The cathode material is readily available and non-
thicknesses of the aluminium and copper substrates were hazardous. Commercially available LFP cells have a lower
calculated from the measured area density. The thickness of the operating voltage (ꢂ3.3 V) than cells with LCO and NMC
18
separator foils was measured with a micrometer.
cathodes.
The active anode materials consisted only of carbonaceous
material for all cells, as veried by SEM/EDX. The exact types of
graphite materials could not be identied.
2.5 Li-ion cells
18650 consumer cells with three types of chemistry were
purchased for the experiments. The cells were produced by
three well-known companies. For simplicity, the samples will be
2.6 Electrical characterisation
referred to as LFP, NMC and LCO/NMC cells, in order to reect An electrical characterisation of the cells was done with a
their respective cathode material. Despite the simple naming BaSyTec CTS cell test system. In the rst step, the cells were
scheme, please note that the cells do not differ in the types of discharged to their respective minimum voltage. In the second
their cathode material alone. Naturally, they also have different step, the cells were charged using a pulse-pause protocol, until
layer geometries (Table 2) and different ratios of their compo- the voltage of the cells stayed above their respective maximum
nent masses (Fig. 2), and there are differences in the composi- voltage during a pause. The current pulses were set to 100 mA
tion of the active masses as well (Table 1).
and 30 s. The duration of the pauses was set to 50 s. The open-
ꢁ
The LCO/NMC cell had a blended cathode with two circuit voltage (OCV) at the end of each pause and the charge
types of electrochemically active particles LiCoO
2
and capacity were recorded (Fig. 3). For the NMC cell, the cell
Fig. 2 Mass split (m%) of the main components of the three cell species.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 3633–3642 | 3635

View Article Online
RSC Advances
Paper
Table 1 Overview of the three cells species used in the experiments. All ratios in this table are given as mol ratios. The electrolyte solvents are
dimethyl carbonate (DMC), ethyl methyl carbonate (EMC), ethylene carbonate (EC) and propylene carbonate (PC)
LCO/NMC
NMC
LFP
Cell mass
g
44.3
43.0
38.8
Capacity
A h
V
V
2.6
3.0
4.2
1.5
3.0
4.1
1.1
2.5
3.5
Minimum voltage
Maximum voltage
Electrolyte solvents
DMC : EMC : EC (6 : 2 : 1)
DMC : EMC : EC : PC
DMC : EMC : EC : PC
(4 : 2 : 3 : 1)
(
7 : 1 : 1 : 1)
Cathode material
Anode material
LiCoO
2
: Li
Li(Ni0.45Mn0.45Co0.10)O
2
LiFePO
4
(
2
Ni0.50Mn0.25Co0.25)O (2 : 1)
Graphite
Graphite
Graphite
manufacturer did not provide the voltage ratings. For safety
reasons, 4.1 V was selected as the maximum voltage.
3
Results and discussion
3.1 Typical course of a thermal runaway experiment
In order to illustrate the events during the heat-up process and
the thermal runaway itself, one experiment with a NMC cell is
described here in detail:
The NMC sample cell was prepared as described above. At
the start of the test, the cell heater sleeve was set to constant
ꢀ
heating power. The sample was slowly heated, starting at 25 C,
ꢀ
ꢃ1
ꢀ
with a heat-rate of ꢂ2 C min . Aer reaching 220 C, the cell
went into rapid thermal runaway. The cell temperature rose
ꢀ
ꢀ
from 220 C to 687 C in a few seconds. When the exothermic
reaction ended, the cell cooled down slowly (Fig. 4a).
The amount of gas produced inside the pressure vessel was
calculated by applying the ideal gas law:
Fig. 3 OCV profiles of the three cell species.
Joule–Thomson effect. The vent opening was then probably
(1) clogged until, at 230 C, concurrent with the rapid thermal
pV
ꢀ
n ¼ Rq ꢃ n
0
gas
runaway, the cell vented for a second time. The second venting
3
where p is the recorded pressure in the reactor, V ¼ 0.0027 m is was the major venting: an additional 0.15 mol of vent gas were
the reactor volume, R is the gas constant, q is the recorded gas produced (Fig. 4b).
gas
temperature in the reactor (in K), and n is the initial amount of
Note that the amount of gas in the reactor decreased shortly
0
gas in the reactor at the start of the experiment.
aer venting. This effect can be explained by the condensation
ꢀ
At 160 C, the safety vent device of the battery housing of gas at the reactor walls. Since the reactor walls had a lower
ꢀ
opened, and 0.02 mol of gas were released by the cell. The cell temperature (ꢂ150 C) than the cell in full thermal runaway (up
ꢀ
ꢀ
cooled down by 10 C during the release process because of the to 687 C), the walls could act as a gas sink.
Table 2 Mass (m), area (A), thickness (d) and volume (V) of the main components of the three cell species. The geometrical volume of a standard
3
1
8650 cell is 16.5 cm
LCO/NMC
NMC
LFP
2
3
2
3
2
3
m (g)
A (cm )
d (mm)
V (cm )
m (g)
A (cm )
d (mm)
V (cm )
m (g)
A (cm )
d (mm)
V (cm )
Separator
1.2
1.7
18.3
2.9
8.1
4.6
942
403
715
402
739
19
16
91
8
1.8
0.6
6.5
0.3
6.0
1.4
3.1
11.3
7.5
6.2
4.4
944
389
654
418
695
23
30
67
20
60
2.2
1.1
4.4
0.8
4.2
1.2
2.1
9.7
3.9
5.2
940
396
793
396
793
20
19
70
17
50
1.9
0.7
5.5
0.7
4.0
Cathode Al foil
Cathode active material
Anode Cu foil
Anode active material
Electrolyte
81
6.4
Housing
Sum
7.5
44.3
9.2
43.1
10.5
39.0
15.2
12.7
12.8
3636 | RSC Adv., 2014, 4, 3633–3642
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 4 (a) Temperature versus time plot of all temperature sensors in the pressure vessel. The whole duration of the experiment is shown. (b)
Amount of produced gas versus time plot. Cell temperature is shown in arbitrary units. (c) Temperature rate of the cell versus cell temperature.
Overview of a whole experiment duration. (d) Temperature rate of the cell versus cell temperature. The straight lines are fitted to the heat-up
o
stage and to the quasi-exponential stage. The intersection of the two lines marks the onset point q of the thermal runaway reaction. A sharp
increase in the temperature rate marks the onset of the rapid thermal runaway q
r
.
ꢀ
ꢀ
In order to visualise subtle changes in thermal behaviour of Between 170 C and 220 C, the temperature rate increase fol-
the cell during the experiment, rate diagrams are utilized. lowed a nearly straight line in the logarithmic plot (Fig. 4d). At
ꢀ
Contrary to a common temperature versus time diagram (q vs. t), 220 C, a sharp increase in temperature rate marked the end of
the temperature rate is plotted versus temperature (dq/dt vs. q) the quasi-exponential heating stage.
ꢀ
in a rate diagram. This type of diagram is oen used to visualise
(3) Rapid thermal runaway stage (q
r
< q < q
m
): At 220 C, q/dt
accelerating rate calorimetry (ARC) results as well. Three increased sharply and initiated the rapid thermal runaway. The
distinct experiment stages can be seen in the rate diagram for transition to thermal runaway was accompanied by a venting
the NMC cell (Fig. 4c):
event. The thermal runaway ended when all reactants had
(
1) Heat-up stage (q < q ): In the temperature range from been consumed. At this point, the maximum temperature
o
ꢀ
ꢀ
room temperature to q
The heater sleeve was the only heat source in this phase. The
negative peak at 130 C is associated with endothermic sepa- and 2. Several endothermic events oen occurred near the onset
rator melting. (It is analogous to a negative endothermic peek in temperature q
o
at ꢂ170 C, the cell generated no heat.
q
m
¼ 687 C was reached.
It is difficult to pinpoint the exact transition between stage 1
ꢀ
ꢀ
o
: the separator melt temperature was 130 C, the
ꢀ
a differential scanning calorimetry (DSC) diagram during the cell safety vent usually opened at 160 C and some material was
phase change of a sample). The temperature q at which a cell released from the cell, causing a slight cool-down due to the
o
starts to generate heat is commonly called the onset tempera- Joule–Thomson effect. Thus, the exact value of qo can be
ture of the thermal runaway.
2) Quasi-exponential heating stage (q
atures higher than q , the battery became a heat source. heating-rate curve switches from constant to quasi-exponential
obscured by the intermediate cell cool-down.
(
o
< q < q
r
): At temper-
To keep it simple, q was dened as the point at which the
o
o
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 3633–3642 | 3637

View Article Online
RSC Advances
Paper
rising. One line is tted to the heat-up part and one line to the Table 3 Characteristic temperatures and venting parameters in the
thermal-runaway experiments. Here, q
the transition temperature into rapid thermal runaway, q
maximum recorded temperature, n is the total amount of gas
o
is the onset temperature, q
r
is
quasi-exponential part of the rate curve in the logarithmic rate
m
is the
plot. The onset temperature q can be further dened as the
o
temperature at which the two lines cross (Fig. 4d).
produced as measured in the reactor at a reactor temperature of
ꢀ
1
50 C, and Dt is the typical venting duration
3.2 Thermal-runaway experiments
LCO/NMC
NMC
LFP
At least 3 thermal-runaway experiments were conducted with
each of the three cell species. A temperature prole overview of
all experiments is shown in Fig. 5a. Each species had its unique
ꢀ
ꢀ
ꢀ
q
q
q
o
C
C
C
149 ꢄ 2
208 ꢄ 2
853 ꢄ 24
265 ꢄ 44
0.8
168 ꢄ 1
223 ꢄ 3
678 ꢄ 13
149 ꢄ 24
0.2
195 ꢄ 8
—
404 ꢄ 23
50 ꢄ 4
30.0
r
m
thermal-runaway characteristics. The high capacity, cobalt rich
n
Dt
mmol
s
ꢀ
LCO/NMC cells reached the highest q
m
at (853 ꢄ 24) C during
thermal runaway. The cobalt poor NMC cells had a lower q
m
of
ꢀ
(
678 ꢄ 13) C. The LFP cells showed a less pronounced thermal
ꢀ
runaway and reached a moderate qm of (404 ꢄ 23) C. The withstand the highest temperature before going into
temperature curves showed small variations from sample to thermal runaway.
sample. It is likely that the variations were caused by different
Both metal oxide cells showed the three stages described
burst times of the rupture plates, which, together with subtle above (heat-up, quasi exponential heating, rapid thermal
effects of venting, Joule–Thomson cool-down and clogging of runaway). In contrast, the thermal runaway prole of the LFP
the vent openings, inuence the thermal-runaway reaction- cell lacked a distinct quasi-exponential stage. For the LFP cell, it
pathways.
was difficult to nd a clear distinction between q
is not given for the LFP species.
During the thermal runaway, the cells produced a signicant
o r
and q .
For the sake of completeness, two additional LFP experi- Therefore, q
r
ꢀ
ments with different heater-sleeve heating-rates (1.5 and 3.5 C
ꢃ1
min ) were also included in the analysis (Fig. 5a). The thermal amount of gas (Table 3). The amount of gas strongly depended
runaway characteristics of the LFP cell (q , q and n) did not on the cell type. The highest amount of gas was released by the
r
m
depend on the heater-sleeve heating rate in the given heat-rate LCO/NMC cell, followed by the NMC cell. The LFP cell yielded
range. The two additional experiments contributed to the mean the least amount of gas.
values in table 3 and Fig. 6
Two successive gas production events were evident in all
For clarity, only one representative curve for each cell species experiments (Fig. 7):
is shown in the following graphs.
1 In the rst venting event, prior to rapid thermal runaway,
Each cell species had distinctive kinetic thermal-runaway the burst plate of the battery opened, and ꢂ20 mmol were
characteristics (Table 3 and Fig. 5b). Of the three specimen, released by all three cell types.
the LCO/NMC cell showed the lowest qo and q , hence the
2 In the second venting event, at the start of rapid thermal
LCO/NMC cell was the cell most vulnerable to over-heating runaway, both metal-oxide cells released a high amount of
conditions. For the NMC cell, q and q
were shied to additional gas at a high rate (Fig. 8). In contrast, the LFP cell
r
o
r
higher temperatures. Transition temperatures of the LFP released only a small amount of additional gas at a low
specimen were noticeably higher than those of both metal/ production rate. In the case of the metal-oxide cells the gas was
oxide cells (LCO/NMC and NMC). The LFP cell was able to released in very short time. The NMC cell produced the main
Fig. 5 (a) Overview of the time–temperature profiles for the cells tested. Data for the whole experiment durations and for the whole experiment
sets is shown. For the sake of completeness, one LFP test with a higher (1) and one with a lower (2) heating rate of the heater sleeve are included.
(
b) Temperature rates from three representative experiments.
3638 | RSC Adv., 2014, 4, 3633–3642
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 6 Detected components of the produced gases (mol%).
decreased, as the released gas came into contact with the walls
and condensed. In contrast, the gas production duration of the
second venting for the LFP cell was ꢂ30 s. Because of the
gradual release, the gases of the LFP cell were in better
temperature equilibrium with the reactor walls and the gas
condensation effect was not noticeable.
3.3 Gas analysis
At least one gas analysis was performed for each cell species.
Each cell type showed a unique gas composition footprint
2 2
(Fig. 6). The main components were H and CO . Both metal-
oxide cells produced a signicant amount of CO. Additionally,
smaller fractions of CH , C H , and C H were identied. As
4
2
4
2
6
mentioned before, HF was not measured.
Most components of the gases are ammable. The gases can
Fig. 7 Temperature-vent gas profiles. Note that the x-axis is limited to
the relevant temperature range.
be toxic due to the presence of CO.
3.4 Gas producing reactions
amount of gas in just 0.2 s, and the LCO/NMC in 0.8 s. Aer
release, the hot gas was not in thermal equilibrium with the During the thermal runaway gases are released by thermally and
cooler walls of the reactor, and therefore the amount of gas electrochemically driven reactions of the electrode active
Fig. 8 Time-vent gas profiles. Note: to make the curves comparable, each curve was moved on the time axis, so that the second venting event
starts at time zero. (a) The first 100 seconds and (b) the first 2 seconds of the second venting event are shown.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 3633–3642 | 3639

View Article Online
RSC Advances
Paper
materials, the intercalated lithium, the binder, the solid elec- oxidation with O
2
freed from the cathode may be the dominant
trolyte interface (SEI), the electrolyte and the separator.
CO producing reaction for the LCO/NMC and NMC cell. The
2
Evolution of H . One possible source of hydrogen is the cathode material of the LFP cell is thermally more stable and
2
0
42
reaction of the binder with Li . Common binder materials are does not release oxygen.
polyvinylidene uoride (PVdF) and carboxymethyl cellulose Evolution of CO. One possible mechanism of carbon
CMC). At temperatures above 230 C graphite particles of the monoxide is the reduction of CO2 with intercalated Li at the
19
ꢀ
(
2
6,43,44
anode defoliate and Li is exposed to the surrounding electrolyte anode:
20
ꢀ
and binder. Above 260 C PVdF may react with Li at the anode
+
ꢃ
21
2CO
2
+ 2Li + 2e / Li
CO
2
3
+ CO
(12)
and release H
2
:
DH
1
On the other hand, as shown in the case of an overcharged
LCO cell, the main contribution of CO gas may come from the
cathode side and not from the anode side. We suggest, that
another source of CO may be incomplete combustion of carbon
-
CH -CF
2
2
- þ Li ƒƒƒƒ ƒƒ ! LiF þ -CH]CF- þ H
2
(2)
2
41
A similar reaction of CMC and Li may take place above
ꢀ
22
2
50 C:
containing material with a limited amount of O that is freed
2
DH
CMC-OH þ Li ƒƒƒƒ ƒƒ ! CMC-OLi þ H
1
from the cathode.
2
(3)
2
Evolution of CH
4 2
. In the presence of H methane can be
produced by reduction of the electrolyte to lithium
Evolution of CO . Many SEI and electrolyte mechanisms can
lead to carbon dioxide generation. The SEI can decompose in
2
45–47
carbonate
C H O (DMC) + 2Li + 2e + H / Li CO + 2CH
4
e.g.
23,24
thermally driven reactions,
+
ꢃ
(13)
3
6
3
2
2
3
DH
1
2
ðCH
2
OCO
2
LiÞ ƒƒƒƒ ƒƒ ! Li
2
CO
3
þ C
2
H
4
þ CO
2
þ
O
2
(4)
2
Evolution of C
tion of EC at the lithiated anode
2 4
H . Ethylene can be produced by the reduc-
2
5,26,26–29
or by reactions with traces of water or HF
31,39,47
ROCO Li + HF / ROH + CO + LiF
(5)
(6)
2
2
2
Li + C
3
H
4
O
3
(EC) / Li
2
CO
3
+ C
2
H
4
(14)
(15)
(16)
2
ROCO
2
Li + H
2
O / 2ROH + Li
2
CO
3
+ CO
2
24,48,49
and
30
Li CO may be present in the cathode and/or can be
2
3
2Li + 2C H O (EC) / (CH OCO Li) + C H
3 4 3 2 2 2 2 4
3
1
produced by two-electron reduction of EC at the anode. Li CO
2
3
24,30
23
reacts with traces of HF with CO evolution:
or by SEI decomposition
2
DH
2
Li + (CH OCO Li) / 2Li CO + 2C H
Li
2
CO
3
þ 2HF ƒƒƒƒ ƒƒ ! 2LiF þ CO
2
þ H
2
O
(7)
2
2
2
2
3
2
4
EC solvent reduction through SEI (re)formation at the
3
2,33
ꢀ
carbon surface of the anode can release CO
2
.
Above 263 C
Evolution of C
2
6
H . In an analogous reaction ethane can be
3
4
31,39,48
pure EC can thermally decompose and produce CO
2
.
Linear produced by the reduction of DMC at the lithiated anode:
carbonate solvents can decompose with CO release in the
2
31
2Li + C
3
6
H O
3
(DMC) / Li
2
CO
3
+ C
H
2 6
(17)
presence of CH OLi.
3
In the presence of impurities LiPF may react to POF that in
6
3
turn reacts with the electrolyte in a decarboxylation reaction
31,35–38
2
with CO release:
4
Conclusion and outlook
LiPF / LiF + PF
(8)
(9)
6
5
Three types of consumer Li-ion batteries with the format 18650
with different cathode materials were evaluated in thermal
runaway tests. The cells were brought into thermal runaway by
PF
5 3
+ ROH / HF + RF + POF
POF + solvent / CO + phosphate
(10) external heating. All tests were performed in a pressure-tight
3
2
5
reactor in an argon atmosphere. In agreement with literature,
In the presence of oxygen, combustion of the carbonate the cell containing LFP showed the best safety characteristics.
2
8,34,39
ꢀ
based electrolyte solvents takes place,
e.g.
The LFP cell had the highest onset temperature (ꢂ195 C), the
smallest temperature increase during the thermal runaway
5
2
DH
ðECÞ ƒƒƒƒ ƒƒ ! 3CO
ꢀ
O
2
þ C
3
H
4
O
3
2
þ 2H
2
O
(11) (ꢂ210 C), the lowest amount of produced gas (ꢂ50 mmol) and
the lowest percentage of toxic CO in the gas (ꢂ4%). Unfortu-
A plausible source of oxygen is the structural breakdown of nately, it was also the cell with the lowest working voltage (3.3 V)
delithiated metal oxide cathodes of the LCO/NMC and NMC and the lowest energy content (3.5 W h).
40
cell. It was shown, that CO
2
is mainly produced on the cathode
side of an overcharged LCO cell. Therefore the electrolyte performed worse in safety tests. The onset temperature shied
Batteries with higher energy content (5.7 W h and 9.9 W h)
41
3640 | RSC Adv., 2014, 4, 3633–3642
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
ꢀ
ꢀ
down to ꢂ170 C and ꢂ150 C, the temperature increase during
3 C.-Y. Jhu, Y.-W. Wang, C.-M. Shu, J.-C. Chang and H.-C. Wu,
J. Hazard. Mater., 2011, 192, 99–107.
4 P. Ribi `e re, S. Grugeon, M. Morcrette, S. Boyanov, S. Laruelle
and G. Marlair, Energy Environ. Sci., 2012, 5, 5271.
5 D. Doughty and E. P. Roth, Electrochem. Soc. Interface, 2012,
21, 37–44.
ꢀ
ꢀ
thermal runaway rose to ꢂ500 C and ꢂ700 C, the amount of
gas released was ꢂ150 mmol and ꢂ270 mmol, and signicant
percentages of CO (13% and 28%) were found for the NMC and
NMC/LCO cells, respectively.
2
All cells released high amounts of H and hydrocarbons.
These gases are highly ammable. Even though the gas could
6 C.-Y. Jhu, Y.-W. Wang, C.-Y. Wen and C.-M. Shu, Appl. Energy,
2012, 100, 127–131.
7 S.-I. Tobishima and J.-I. Yamaki, J. Power Sources, 1999, 81–
82, 882–886.
not burn in the inert atmosphere inside the reactor, the surface
ꢀ
of the high-energy cells reached temperatures of up to 850 C
during the experiments.
Modern devices are equipped with battery temperature and
voltage monitoring. If a state beyond specication is detected,
the devices shut down automatically to prevent battery abuse.
If system shut-down can not prevent a thermal runaway in all
8 D. Belov and M.-H. Yang, Solid State Ionics, 2008, 179, 1816–
1821.
9 H. Maleki, G. Deng, A. Anani and J. Howard, J. Electrochem.
Soc., 1999, 146, 3224.
18
cases, data in this work may be a valuable source for the spec- 10 C.-Y. Jhu, Y.-W. Wang, C.-Y. Wen, C.-C. Chiang and
ication of a robust energy-storage system which can withstand
conceivable abuse events.
To reduce possible damage from thermal-runaway events in
consumer devices, we suggest the following design optimiza-
C.-M. Shu, J. Therm. Anal. Calorim., 2011, 106, 159–163.
11 C.-Y. Wen, C.-Y. Jhu, Y.-W. Wang, C.-C. Chiang and
C.-M. Shu, J. Therm. Anal. Calorim., 2012, 109, 1297–
1302.
tion targets: (1) increase the temperature endurance and heat 12 E. P. Roth and C. J. Orendorff, Electrochem. Soc. Interface,
absorption capability of used materials; (2) minimize heat 2012, 21, 45–49.
propagation to neighbouring burnable elements; (3) minimize 13 D. Abraham, E. P. Roth, R. Kostecki, K. McCarthy,
gas ignition probability (e.g. mechanical separation of electric
components from the gas release position).
S. MacLaren and D. Doughty, J. Power Sources, 2006, 161,
648–657.
This work has shown that the kinetics of the thermal- 14 Z. Chen, Y. Qin, Y. Ren, W. Lu, C. Orendorff, E. P. Roth and
runaway process strongly depend on the energy content of the K. Amine, Energy Environ. Sci., 2011, 4, 4023.
Li-ion battery. Future work will focus on the thermal runaway 15 D. H. Doughty, E. P. Roth, C. C. Cras, G. Nagasubramanian,
triggered by over-heating at different states of charge (SOC) and
G. Henriksen and K. Amine, J. Power Sources, 2005, 146, 116–
the thermal runaway caused by overcharge. Emphasis will be
120.
given to assessment of HF gas evolution, to gas analysis with 16 G. Nagasubramanian and C. J. Orendorff, J. Power Sources,
GC-MS, and to the analysis of the liquid residues that are
collected in the cooling trap.
2011, 196, 8604–8609.
17 K.-S. Lee, S.-T. Myung, D.-W. Kim and Y.-K. Sun, J. Power
Sources, 2011, 196, 6974–6977.
1
8 Z. J. Zhang and P. Ramadass, Encyclopedia of Sustainability
Science and Technology, Springer, New York, 2012.
Acknowledgements
The present work was conducted in the K2 project “New 19 J.-H. Lee, U. Paik, V. a. Hackley and Y.-M. Choi,
Component Simulation Models for HEVs” in the task “Li-Ion J. Electrochem. Soc., 2005, 152, A1763.
Battery Safety in Automotive Environment (LISAE)”. The 20 O. Haik, S. Ganin, G. Gershinsky, E. Zinigrad, B. Markovsky,
authors would like to acknowledge the nancial support of the
COMET K2 – Competence Centres for Excellent Technologies
D. Aurbach and I. Halalay, J. Electrochem. Soc., 2011, 158,
A913.
“
Programme” of the Austrian Federal Ministry for Transport, 21 A. D. Pasquier, J. Electrochem. Soc., 1998, 145, 472.
Innovation and Technology (BMVIT), the Austrian Federal 22 W. Haiyan, A. Tang and W. Kelong, Chin. J. Chem., 2011, 29,
Ministry of Economy, Family and Youth (BMWFJ), the Austrian
Research Promotion Agency (FFG), the Province of Styria and 23 M. N. Richard and J. Dahn, J. Electrochem. Soc., 1999, 146,
the Styrian Business Promotion Agency (SFG). We would 2068.
furthermore like to express our thanks to our supporting 24 M. Onuki, S. Kinoshita, Y. Sakata, M. Yanagidate, Y. Otake,
27–32.
scientic project partners, namely Graz Centre for Electron
M. Ue and M. Deguchi, J. Electrochem. Soc., 2008, 155,
Microscopy, Varta Micro Innovation GmbH and the Graz
A794.
University of Technology. This work was made possible by the 25 J.-S. Hong, H. Maleki, S. A. Hallaj, L. Redey and J. R. Selman,
principal industrial project partners BASF SE and MAGNA
STEYR Battery Systems GmbH & Co OG.
J. Electrochem. Soc., 1998, 145, 1489.
26 D. Aurbach, A. Zaban, Y. Gofer, Y. E. Ely, I. Weissman,
O. Chusid and O. Abramson, J. Power Sources, 1995, 54,
7
6–84.
References
2
7 D. Aurbach, J. Electrochem. Soc., 1995, 142, 2882.
1
T. Nagaura and K. Tozawa, Prog. Batteries Sol. Cells, 1990, 9, 28 W. Kong, H. Li, X. Huang and L. Chen, J. Power Sources, 2005,
09. 142, 285–291.
J. Wen, Y. Yu and C. Chen, Mater. Express, 2012, 2, 197–212. 29 D. Aurbach, J. Electrochem. Soc., 1991, 138, 3529.
2
2
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 3633–3642 | 3641

View Article Online
RSC Advances
Paper
30 C. Doh, D. Kim, J. Lee, D. Lee, B. Jin, H. Kim, S. Moon, 40 I. Belharouak, W. Lu, D. Vissers and K. Amine, Electrochem.
Y. Hwang and A. Veluchamy, Bull. Korean Chem. Soc., 2009,
0, 783.
1 G. Gachot, P. Ribi `e re, D. Mathiron, S. Grugeon, M. Armand,
Commun., 2006, 8, 329–335.
3
41 T. Ohsaki, T. Kishi, T. Kuboki, N. Takami, N. Shimura,
Y. Sato, M. Sekino and A. Satoh, J. Power Sources, 2005,
146, 97–100.
3
J.-B. Leriche, S. Pilard and S. Laruelle, Anal. Chem., 2011, 83,
478–485.
42 A. K. Padhi, J. Electrochem. Soc., 1997, 144, 1188.
32 C. Yang, Y. WangandC. Wan, J. PowerSources, 1998, 72, 66–70. 43 A. D. Pasquier, F. Disma and T. Bowmer, J. Electrochem. Soc.,
3
3 S. Mori, H. Asahina, H. Suzuki, A. Yonei and K. Yokoto, J.
Power Sources, 1997, 68, 59–64.
1998, 145, 472.
44 S. E. Sloop, J. B. Kerr and K. Kinoshita, J. Power Sources, 2003,
119–121, 330–337.
45 K. Kumai, J. Power Sources, 1999, 81–82, 715–719.
3
3
3
3
4 G. G. Botte and T. J. Bauer, J. Power Sources, 2003, 119–121,
815–820.
5 A. Hammami, N. Raymond and M. Armand, Nature, 2003, 46 J.-S. Shin, C.-H. Han, U.-H. Jung, S.-I. Lee, H.-J. Kim and
24, 635–636. K. Kim, J. Power Sources, 2002, 109, 47–52.
6 C. L. Campion, W. Li and B. L. Lucht, J. Electrochem. Soc., 47 H. Yoshida, T. Fukunaga, T. Hazama, M. Terasaki,
4
2005, 152, A2327.
M. Mizutani and M. Yamachi, J. PowerSources, 1997, 68,
311–315.
7 C. L. Campion, W. Li, W. B. Euler, B. L. Lucht, B. Ravdel,
J. F. DiCarlo, R. Gitzendanner and K. M. Abraham, 48 G. Gachot, S. Grugeon, G. G. Eshetu, D. Mathiron, P. Ribi `e re,
Electrochem. Solid-State Lett., 2004, 7, A194.
8 T. Kawamura, A. Kimura, M. Egashira, S. Okada and
J.-I. Yamaki, J. Power Sources, 2002, 104, 260–264.
M. Armand and S. Laruelle, Electrochim. Acta, 2012, 83, 402–
409.
49 D. Aurbach, A. Zaban, Y. Ein-Eli, I. Weissman, O. Chusid,
B. Markovsky, M. Levi, E. Levi, A. Schechter and E. Granot,
J. Power Sources, 1997, 68, 91–98.
3
3
9 R. Spotnitz and J. Franklin, J. Power Sources, 2003, 113, 81–
100.
3642 | RSC Adv., 2014, 4, 3633–3642
This journal is © The Royal Society of Chemistry 2014