Journal of Power Sources 119–121 (2003) 902–905
Study of life evaluation methods for Li-ion batteries
aNTT Telecommunications Energy Laboratories, 3-1 Morinosato, Wakamiya, Atsugi-shi,
bNTT-BTI, 3-1 Morinosato, Wakamiya, Atsugi-shi, Kanagawa-ken 243-0198, Japan
The backup characteristics of lithium-ion batteries were investigated using commercial prismatic lithium-ion cells with a LiCoO2/graphite
cell system. An accelerated method of estimating the lifetime of lithium-ion batteries was developed. It was found that higher temperaturesand voltages accelerate the degradation of the cells: a 15 8C increase in temperature cuts the cell life in half, and about 0.1 V increase incharging voltage also cut the cell life in half. # 2003 Elsevier Science B.V. All rights reserved.
Keywords: Lithium-ion batteries; Calendar life; Backup use; Accelerated test; Impedance; Capacity fade
Many stationary backup batteries are needed to supply
power to telecommunications equipment during a poweroutage. In addition, telecommunications networks are now
Cells for backup use are subjected to a peculiar pattern of
being converted to broadband systems, like Fiber to the
usage in that they are infrequently charged or discharged,
Home (FTTH); and there is an explosive increase in traffic.
and remain in a fully or partially charged state for most of
So, the amount of power used is also increasing dramati-
cally. This has led to a search for backup batteries with a
The special methods illustrated in were used to
higher energy density that can provide the full backup time
examine the backup characteristics. In the continuous float
and occupy only a limited space. The high energy density of
charging test, the cells were charged at a constant voltage
lithium-ion batteries makes them a very attractive replace-
from 4.0 to 4.3 V and a rate of 1 CmA, and the voltage was
ment for current ones, which are mainly valve regulated lead
maintained continuously. The cells in each test were dis-
charged every month at the rate of 1 CmA to 2.75 V and then
One of the requirements for backup batteries is a very long
recharged to 4.1 V and discharged to check their capacity.
life, typically 15 years or more. Time compression of testing
The cells were kept continuously at constant temperatures
duration is an indispensable means of evaluating life of
developed batteries for a given periods. However, there have
All tests were conducted on commercialized prismatic
been some studies on the accelerated cycle and storage
lithium-ion cells of 900 mAh capacity (type LP10) from
life tests of lithium-ion batteries the accelerated
calendar life in continuous float charging have not beenproposed yet.
In this study, we aimed to develop an accelerated method
of estimating the calendar life of backup batteries. We
Electrochemical impedance spectroscopy (EIS) was car-
investigated the two accelerated life tests with the stress
ried out on charged cells in the frequency range of 10 kHz to
factors, temperature and state of charge.
0.05 Hz using a low ac voltage of 10 mV to minimizeperturbations to the system. The experiments were carriedout at various temperatures with a Solatron 1286 electro-
chemical interface and a 1255 frequency-response analyzer.
0378-7753/03/$ – see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0378-7753(03)00208-8
K. Asakura et al. / Journal of Power Sources 119–121 (2003) 902–905
activation energy (Ea) was 34.9 kJ/mol. For VRLA batteries,temperature strongly affects lifetime, and it is known that
3.1. Dependence of cell capacity on temperature
their Ea is about 60 kJ/mol and a 10 8C increase in tem-perature cuts the lifetime in half But for lithium-
shows the results of continuous charging at 4.1 V at
ion batteries, we confirmed that an elevated temperature
various temperatures, with periodic discharging to check the
accelerates degradation and the slope of the Arrhenius
capacity. The capacity retention decreased gradually with an
plots shows that it takes a 15 8C increase to cut the lifetime
increase in the charging duration and the operating tem-
perature. When the end of life was defined to be a decrease incapacity to 70%, the lifetime of the test cells was estimated
3.2. Dependence of cell capacity on charging voltage
We examined the influence of the temperature on lifetime
The effect of charging voltage on the backup character-
of the cells. shows Arrhenius behavior and the
istics was also examined. shows the results of
Fig. 2. Capacity retentions during continuous float charging.
Fig. 3. Temperature dependences on lifetime in continuous float chargingtests.
Fig. 4. Capacity retentions in continuous float charging characteristics with various charging voltages at 45 8C. (a) Normal continuous float charging test andcapacity check (recharged at floating voltage), (b) capacity check (recharging at 4.1 V).
K. Asakura et al. / Journal of Power Sources 119–121 (2003) 902–905
An analysis was made of how the impedance parameters
change during continuous float charging at various tempera-tures. We investigated that elevated temperatures and highercharging voltages were found to enhance the increasein the electrolyte impedance and the two charge-transferimpedances. Some possible reasons: (i) electrolyte decom-positions on the anode/cathode electrodes and (ii) cathodedegradation, are generally suggested for larger increase inthe impedance.
In the electrolyte impedance increased markedly as
the charging duration increased. An increase in the thicknessand a slight weight loss of the cells was also observed. These
Fig. 5. Dependences of states of charge on lifetime in continuous float
phenomena are probably due to the generation of gas, which
is attributable to decomposition of the electrolyte on thesurface of the electrodes.
continuous charging, with periodic discharging to check the
In addition, the lower frequency semicircle becomes
capacity; the parameter is charging voltage. In , 100%
markedly larger, and the higher frequency one becomes a
capacity is the initial cell capacity on the same condition of
little larger. It has been reported that the higher frequency
the capacity check in before they were float charged
semicircle corresponds in large part to the anode reaction,
continuously. When the cell was charged at higher voltage,
while the lower frequency one mainly corresponds to the
the cell showed large capacity at the beginning of the tests.
cathode reaction . Aurbach et al. reported that the surface
However they degraded rapidly as the charging duration
firms formed on not only anode but also cathode along with
electrolyte decompositions During backup use, cell
Each of the cells was also recharged at 4.1 V and dis-
degradation may also involve decomposition of the electro-
charged to check the capacity. shows that cell
lyte and formation of the surface film on the electrodes, thus
degradation is accelerated at higher charging voltages; and
increasing their cell polarization impedance and charge-
shows that cell life drops exponentially as the charging
voltage increases. We also confirmed about 0.1 V increase in
Other degradation factors must also be considered, such
charging voltage cut the cell life in half.
as dissolution of the cathode and the disordering of thecrystal structure. It has been reported that LiCoO2 dissolves
slightly during aging tests shows the amountof Co dissolution in the electrolyte after 1 month conti-
is a typical change in Nyquist plots of batteries in
nuous float charging at 55 8C. As shown in the test
continuous float charging tests. The plots contain two semi-
for only 1 month caused the slight Co dissolution from
circles. An equivalent circuit model similar to that reported
the LiCoO2 cathode. During long charging duration, these
by Monma et al. was devised. Simulations performed
Co dissolution phenomena might also increase the charge-
with the Solatron impedance software ZView 2TM (Scribner
transfer impedance of the cathode. On the other hand,
Associates, Inc.) on the equivalent impedance model pro-
there may be little disordering of the crystal structure
duced results that agree quite well with the experimental
during backup use because the infrequency of charging
and discharging should result in little damage to the crystalstructure.
From these data, elevated temperature and higher char-
ging voltage accelerated the cell degradation and thesestress factors were appropriate to accelerate the calendarlife experimentally. However, further work needs to chooseof the degradation mechanism(s) which influence on the
Table 1The amount of Co dissolution after continuous float charging at 55 8C
Fig. 6. Nyquist plots of the batteries in continuous float charging test at45 8C.
a The charge-transfer impedance related to the lower semi-circle (R).
K. Asakura et al. / Journal of Power Sources 119–121 (2003) 902–905
decrease in the cell capacity, in order to establish universal
methods for accelerated life evaluation with all types ofcells.
[1] M. Broussely, S. Herreyre, P. Biensan, P. Kasztejna, K. Nechev, R.J.
Staniewicz, J. Power Sources 97–98 (2001) 13.
[2] K. Takei, K. Kumai, Y. Kobayashi, H. Miyashiro, N. Terada, T.
Iwahori, T. Tanaka, J. Power Sources 97–98 (2001) 697.
[3] I. Bloom, B.W. Cole, J.J. Sohn, S.A. Jones, E.G. Polzin, V.S. Battaglia,
G.L. Henriksen, C. Motloch, R. Richardson, T. Unkelhaeuser, D.
We investigated the basic backup characteristics of
Ingersoll, H.L. Case, J. Power Sources 101 (2001) 238.
lithium-ion batteries and two accelerated evaluation meth-
[4] R.F. Nelson, in: Proceedings of the INTELEC’89, 1989, pp.12–16.
ods were experimentally adopted to quickly predict the
[5] T. Momma, K. Tsuchiya, T. Osaki, in: Proceedings of the 200th
lifetime of a cell. The cell degradation is accelerated at
Meeting of Electrochemical Society, 2001, p. 135.
elevated temperatures and at higher charging voltages. A
[6] G. Nagasubramanian, J. Power Sources 87 (2000) 226. [7] D. Aurbach, K. Gamolsky, B. Markovsky, G. Salitra, Y. Gofer, U.
15 8C increase in temperature cuts the cell life in half, and
Heider, R. Oesten, M. Schmidt, J. Electrochem. Soc. 147 (2000) 1322.
about 0.1 V increase in charging voltage also cut the cell life
[8] G.G. Amatucci, J.M. Tarascon, L.C. Klien, Solid State Ionics 83
medicine.ox.ac.uk Swots Corner: What is an odds ratio? Bandolier readers wil know that we favour the number-needed-to-treat (NNT) as a way of describing the benefits (or harms) of treatments, both in individual trials and in systematic reviews. Few papers report results using this easily interpretable measure, so Bandolier has had to get its head around how to do the calculations. NNT cal
EDUCATION Ph.D. September 1998. University of California Davis, Department of History B.A January 1992. Smith College. Magna cum laude with high honors in American Studies. EMPLOYMENT 2010- Professor, Department of History, University of Cincinnati 2005-2010 Associate Professor, Department of History, University of Cincinnati 2000-2005 Assistant Professor, Department of History, Univer