A transmission electron microscopy study of cycled LiCoO
2
H. Gabrisch, R. Yazami
*,1
, B. Fultz
Division of Engineering and Applied Science, Mail 138-78, California Institute of Technology, Pasadena, CA 91125, USA
Abstract
A sample of LiCoO
2
was prepared as a cathode for a lithium battery, and subjected to a large number of charge/discharge cycles that
induced an irreversible loss in capacity. A transmission electron microscopy study of this material showed that the initial O3 crystal structure
transformed partially to a cubic spinel phase, especially on the surfaces of the particles. We suggest that this spinel phase formation could be
responsible in part for the irreversible capacity fade.
# 2003 Elsevier Science B.V. All rights reserved.
Keywords: Electron microscopy; LiCoO
2
; Spinel; Phase transformation
1. Introduction
LiCoO
2
heat-treated above 700 8C (HT-LiCoO
2
) is used
in positive electrodes (cathodes) of rechargeable Li-ion
batteries. Compared to other cathode materials it has a high
energy density, high cycle life and good capacity retention.
The structure of HT-LiCoO
2
is trigonal (also referred to as
hexagonal) of R
3m symmetry. The lattice is formed by
oxygen atoms in ABC stacking with alternating layers of
Li and Co ions in octahedral interstitial sites between the
oxygen planes. Lithium is removed from the structure (and
moved to the anode) when the battery is charged. During
charge–discharge cycling the Li-content in Li
x
CoO
2
is
cycled between x ¼ 0:5 (charged) and x ¼ 1 (discharged).
During delithiation of Li
x
CoO
2
(charging), the following
changes occur in the crystal structure of Li
x
CoO
2
. Between
x ¼ 1 and 0.75, two hexagonal lattices co-exist, followed by a
regime of one phase (hexagonal) for x ¼ 0:75–0.25. Around
x ¼ 0:5, a monoclinic unit cell is observed, but it is stable only
in the very narrow concentration range of x ¼ 0:44–0.49
[1,2]. There have been many studies on how the crystal struc-
ture evolves over one charge–discharge cycle. On the other
hand, there have been few studies on how the crystal structure
evolves after extensive electrochemical cycling [3,4]. Never-
theless, irreversible capacity fade after cycling is an important
phenomenon that is not well understood, and likely involves
some change in the crystal structure of the LiCoO
2
.
Here, we report the results of a transmission electron micro-
scopy (TEM) investigation on the microstructural evolution of
LiCoO
2
after extensive electrochemical cycling, where the
cathode has undergone a capacity loss. Although the starting
material was entirely the O3 structure, we found cubic spinel
phase in the cycled LiCoO
2
. Since the cubic spinel phase is
less active electrochemically, its formation could be a source
of the capacity fade. Furthermore, we observed degradation
of crystal quality at the surface of LiCoO
2
particles after
cycling, and this deterioration may affect electrical contact
between the LiCoO
2
particles in the electrode.
2. Experimental
Virgin LiCoO
2
powder and LiCoO
2
that was electroche-
mically cycled in a battery between 3.0 and 4.2 V at room
temperature, were provided by the courtesy of ENAX, Inc.,
Japan. The electrochemically cycled material was subjected
to 334 charge–discharge cycles, and left in the discharged
(lithiated) state. The cycled LiCoO
2
powder was retrieved
by scraping the carbon/binder (PVDF: polyvinylidene
difluoride)/LiCoO
2
mix from the cathode and washing it
in n-methyl-pyrrolidinone (NMP) to dissolve the binder.
Specimens for TEM examination were prepared from a
suspension of LiCoO
2
powder in methanol. A droplet was
placed on a holey carbon film supported by a copper grid. The
TEM observations were performed with a Philips EM420
instrument operated at 120 kV. Several particles of the
uncycled and cycled material were viewed in low-index zone
axes. Images were taken along with their corresponding dif-
fraction pattern. The experimental studies were accompanied
by simulations of diffraction patterns for various crystal
structures using the software package ‘‘Desktop Microsco-
pist’’. For these simulations we used Co Ka rays with
l ¼ 0:17904 nm.
Journal of Power Sources 119–121 (2003) 674–679
*
Corresponding author. Tel.: þ1-626-395-4496; fax: þ1-626-795-6132.
E-mail address: yazami@caltech.edu (R. Yazami).
1
Present address: LEPMI, INPG-CNRS 5631, BP 75, 38402 St. Martin
d’Heres, France.
0378-7753/03/$ – see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0378-7753(03)00234-9