See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/244374216

Nanomaterials for lithium ion batteries

Article in Nano Today November 2006

DOI: 10.1016/S1748-0132(06)70114-1

CITATIONS READS

186 296

3 authors, including:

Chunhai Jiang Haoshen Zhou

Xiamen University of Technology National Institute of Advanced Industrial Science and Technology
59 PUBLICATIONS 1,302 CITATIONS 309 PUBLICATIONS 11,569 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Chunhai Jiang

Retrieved on: 16 September 2016
 Nanomaterials for
lithium ion batteries
Nanostructured materials are currently of interest for lithium ion
storage devices because of their high surface area, porosity, etc. These
characteristics make it possible to introduce new active reactions,
decrease the path length for Li ion transport, reduce the specific
surface current rate, and improve stability and specific capacity.
Moreover, composite nanostructured materials designed to include
electronic conductive paths could decrease the inner resistance of
lithium ion batteries, leading to higher specific capacities even at high
charge/discharge current rates.

Chunhai Jiang, Eiji Hosono, and Haoshen Zhou*

Energy Technology Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Umezono, 1-1-1, Tsukuba,
305-8568 Japan
*E-mail: hs.zhou@aist.go.jp

Nanomaterials have been widely applied in the life sciences,
information technology, the environment, and other related fields.
Recently, nanostructured materials have also attracted attention
for application in energy storage devices1,2, especially for those
with high charge/discharge current rates such as lithium ion
batteries3. The development of next-generation energy storage
devices with high power and high energy density is key to the
success of electric and hybrid electric vehicles (EVs and HEVs,
respectively)4,5, which are expected to at least partially replace
conventional vehicles and help solve the problems of air pollution
and climate change. These energy storage technologies will rely
on innovative materials science, i.e. developing electrode materials
capable of being charged and discharged at high current rates.
Lithium ion rechargeable batteries consist of a positive electrode
(cathode), Li ion-containing electrolyte, and negative electrode (anode)
(Fig. 1). The positive and negative electrode materials of the most
common, commercial lithium ion batteries are LiCoO2 and graphite,
respectively. Both LiCoO2 and graphite are Li ion insertion hosts.
During the charging process, Li ions are extracted from the LiCoO2
electrode and simultaneously inserted into the graphitic carbon
electrode, coupling with negatively charged electrons to keep overall
charge neutrality. During the discharging process, Li ions are reversibly
extracted from the negative electrode and simultaneously inserted into
the positive electrode.
This electrochemical insertion/extraction process is a solid-state
redox reaction involving electrochemical charge transfer coupled with
insertion/extraction of mobile guest ions into/from the structure of an
electronic and ionic conductive solid host. The major structural
features of the host are kept after the insertion/extraction of the
guests.

28 NOVEMBER 2006 | VOLUME 1 | NUMBER 4 ISSN:1748 0132 Elsevier Ltd 2006

 Nanomaterials for lithium ion batteries
Fig. 1 Schematic of an insertion electrode-based rechargeable lithium ion battery.
It is usually desirable that the amount of energy stored in a given
mass or volume of the lithium ion battery is as high as possible. To
compare the energy content of lithium ion batteries, the terms specific
energy density (expressed in Wh/kg) and energy density (in Wh/L)
are used, whereas the rate capability is expressed as specific power
density (in W/kg) and power density (in W/L). For HEVs, it is thought
that the required specific energy density and specific power density
of lithium ion batteries should be above 50 Wh/kg and 3 kW/kg,
respectively. Moreover, much greater values will be necessary for EVs.
The development of nanostructured electrodes is considered to be the
most promising route to achieve these goals.
Generally, the potential advantages of nanostructured active
electrode materials can be summarized as follows2: new reactions can
be used that are not possible with bulk materials; a larger electrode/
electrolyte contact area, leading to higher charge/discharge rates;
short path lengths for both electronic and Li ion transport (permitting
operation even with low electronic or low Li ion conductivity, or at
higher power); etc. Here, we review some recent experimental results
that show the advantages of nanostructured active electrode materials.
New reactions
In recent years, great efforts have been put into the reinvestigation
of materials that were thought of as being electrochemically inactive
in bulk form but that could present improved electrochemical
performance at the nanoscale. One good example is the demonstration
that nanoparticles of some simple transition metal oxides, sulfides,
fluorides, and nitrides can provide innovative anode materials for
lithium ion batteries. The reaction of such transition metal compounds
with Li leads to the in situ formation of metal nanoparticles embedded
in an LizX matrix (where X = O, S, F, or N).
REVIEW
The transition metals do not alloy with Li, so the mechanism of
Li reactivity differs from classical Li insertion/extraction or Li-alloying
processes. The classical Li insertion/extraction process is shown in Fig.1
while the Li-alloying process can be written as:
M + xLi+ + xe- LixM (1)
where M is Sn, Si, Pb, Bi, Sb, Ag, Al, or a multicomponent alloy.
Instead, the mechanism for the reaction of transition metal
compounds with Li during charge and discharge involves the formation
and decomposition of LizX accompanying the reduction and oxidation
of metal nanoparticles. The overall mechanism is described by the
follow equation4,6,7:
MX + zLi+ + ze- LizX + M (2)
where M represents the transition metals Fe, Co, Ni, Cu, etc. As can be
seen, the major difference between eqs. 1 and 2 is the involvement of
formation and decomposition of LizX or a Li alloy.
In a systematic study, Poizot and coauthors7 showed that electrodes
made of nanoparticles of transition metal oxides can deliver specific
capacities of 700 mAh/g with 100% capacity retention for up to
100 charge/discharge cycles and high recharging current rates when
being charged/discharged in the potential window of 3.5 V to
0.01 V (versus Li+/Li). The high Li storage capacity in nanometer-sized
transition metal oxides at low potential has recently been explained by
an interfacial charge storage mechanism8,9. According to this model,
Li ions are stored on the oxide side of the interface, while electrons (e-)
are localized on the metallic side, resulting in a charge separation. On
this basis, the electrochemically driven size confinement of the metal
particles is believed to enhance their electrochemical activity toward
the formation/decomposition of Li2O. With decreasing particle size, an
increasing proportion of the total number of atoms lies near or on the
surface, making the electrochemical reactivity of the particles more
NOVEMBER 2006 | VOLUME 1 | NUMBER 4 29
REVIEW Nanomaterials for lithium ion batteries
and more important. Such considerations can explain why the cycling
performance of such materials is extremely sensitive to their degree
of aggregation. For instance, Kim et al.10 recently showed that SnO2
particles 3 nm in size have a superior capacity and cycling stability than
4 nm and 8 nm diameter particles because they show less aggregation
of Sn nanoparticles into large clusters since the nanoparticles are better
dispersed in the Li2O matrix.
Nanostructured electrodes may not only introduce innovative
reaction mechanisms but also improve electrochemical properties,
such as specific energy storage capacity, high current charge/discharge
ability, and cycle stability, over their bulk counterparts. This results
from the short diffusion length and high contact area between the
active materials and electrolyte. Li ion diffusion is highly dependent
on the transport length and accessible sites on the surface of active
materials. Those compounds exhibiting low Li ion diffusion coefficients
usually show low Li ion storage capacities in bulk form, especially
at high current rates. This is the case for rutile TiO2, which can only
accommodate a negligible amount of Li ions at room
temperature11-13. Li ion diffusion is highly anisotropic in rutile TiO2
and shows rapid diffusion along the c-axis channels. Meanwhile, a large
distortion of the rutile framework makes Li ion diffusion in the a-b
planes very slow at low temperature (Dab~10-15 cm2/s). This prevents
Li ions from reaching thermodynamically favorable octahedral sites
residing in the a-b planes and separates Li ions in the
c channels14-16. However, at the nanoscale the situation is significantly
different. For rutile TiO2 particles with an average size of 15 nm, a full
loading of lithium (x > 1 in LixTiO2) has been observed in our recent
3.5
3.0
Voltage (V vs. Li+/Li)
2.5
2.0
1.5
1.0
Fig. 2 First and second discharge/charge profiles for nanometer-sized rutile TiO2 electrode with a particle size of 15 nm. The charge and discharge current densities
are both 50 mA/g.
30 NOVEMBER 2006 | VOLUME 1 | NUMBER 4
work17. Moreover, about 0.7 Li per rutile TiO2 can be reversibly loaded
and unloaded in subsequent cycling (Fig. 2). A similar result has also
been reported by Hu et al.18 and Reddy et al.19.
It is believed that the limits to Li ion diffusion in the a-b planes
have been weakened because of the very short diffusion length within
the nanometer-sized rutile TiO2 electrode. That is, more octahedral
sites in the a-b planes can be reached by Li ions within a certain time.
Besides that, theoretical work by Stashans et al.20 has demonstrated
that, in the case of the most stable (110) surface of rutile TiO2, lithium
intercalation seems to be mainly a surface effect since the Li atom
does not penetrate into the bulk. Because of the large specific surface
area of nanometer-sized rutile TiO2 electrodes, the amount of surface-
stored Li is significantly increased, contributing substantially to the
lithium storage capacity.
Larger electrode/electrolyte contact area
As described above, surface Li storage plays an important role in the
overall capacity of nanoelectrodes. In addition, as has been observed
on many anode materials, a higher electrode/electrolyte contact area
can also lead to improved high current rate charge/discharge capacities.
This can be explained based on two considerations.
First, on the one hand, the fine particle size, i.e. the short transport
length, makes full Li diffusion possible within a short diffusion time,
i.e. at high charge/discharge current rates. On the other, surface Li
storage only depends on the surface area but not seriously on the
diffusion time. A higher electrode/electrolyte contact area is thus
beneficial to high current rate performance.
 Nanomaterials for lithium ion batteries
Voltage (Li+/Li)
Fig. 3 Voltage profile of a 6 nm anatase TiO2 electrode cycled at 0.1 A/g, 1 A/g, and 10 A/g. Inset shows the cycle performance at 10 A/g.
Second, using nanoelectrodes can greatly reduce the specific
current density of the active material because of the very large
surface area. The low specific current density can effectively stabilize
the electrode and preserve a high capacity at high charge/discharge
current densities3. For example, Poizot et al.7 have shown that a CoO
nanoelectrode could sustain 85% of its total capacity at a rate of
2 C (C is the current rate needed to charge or discharge the theoretical
capacity in 1 hour). Also, besides the recently demonstrated high
specific capacity, an excellent high-rate performance has been observed
for rutile TiO2 nanoelectrodes18. This finding is also valid for the
spinel lithium titanate Li4Ti5O12. Li4Ti5O12 is a very attractive anode
material because its volume change upon lithium insertion/extraction
is almost negligible, resulting in excellent cycle stability. However, its
semiconductor nature means that the high current charge/discharge
performance of Li4Ti5O12 is rather poor in bulk form. Kavan et al.21
have reported that electrodes of nanocrystalline Li4Ti5O12 exhibit
excellent activity toward Li insertion, even at charging rates as high
as 250 C (where 1 C = 175 mA/g). The materials with surface areas
between ~20-100 m2/g can be charged/discharged nearly to the
nominal capacity of Li4Ti5O12 (175 mAh/g) within a wide range of
charge/discharge current rates (from 2 C to 250 C).
In recent work, we have systematically investigated the particle size
dependence of lithium storage capacity and high rate performance of
nanocrystalline anatase TiO2 electrodes22. We found that by decreasing
the particle size of the anatase electrode, the narrowing of the Li
insertion/extraction plateaus at high charge/discharge current rates was
retarded. Meanwhile, the contribution of the surface Li storage almost
did not depend on the current rates and cycle number. This resulted in
REVIEW
very stable cycle performance in an anatase TiO2 nanoelectrode even
at high charge/discharge current rates (Fig. 3).
Short path lengths
Generally, there is a reduction or oxidation reaction in the lithium
insertion/extraction (discharge/charge) process, where Li ion and
electron transport take an important role, especially in high-rate
charging and discharging. Nanostructured materials can provide
short path lengths for both Li ion and electron transport. In contrast,
commercial batteries are mostly based on micrometer-sized electrode
materials, i.e. powders containing particles in the micrometer range
and having a low surface area (<10 m2g-1). From the point of view of
Li ion diffusion, these micrometer-sized electrode materials are not
beneficial for high-rate charge/discharge because of the long path
length for Li ion transport and low contact area between the electrode
and electrolyte.
Diffusion of lithium ions is complex because the nature of the
electrolyte phase, the solid-liquid interface, the tortuosity of diffusion
paths, and the size of the nanoparticles need to be considered23. If
we only consider the overall process by assuming that the diffusion
coefficient is independent of these factors, the diffusion length L can be
estimated as D, where D and are the diffusion coefficient and time,
respectively. The specific capacity Q of the battery is obtained by i ,
where i is the specific charge/discharge current density in A/kg or
mA/g. For a certain specific capacity, increasing i will decrease rapidly.
So the effective specific capacity will depend on the volume ratio
[r3 - (r - L)3]/r3, where r is the radius of the active particle3. The
required diffusion length L should be larger than r to obtain the
NOVEMBER 2006 | VOLUME 1 | NUMBER 4 31
REVIEW Nanomaterials for lithium ion batteries
maximum specific capacity. If we assume that the rapid charge/
discharge process is complete within one minute and that D is of the
order of 10-16 cm2s-1 in the solid state23, particles with a size of 2r
should be about 2 nm. This suggests that nanometer-sized electrode
materials are necessities for high power and high energy density energy
storage and conversion.
This concept has been shown to hold for nanoporous TiO2 active
materials3. The nanoporous TiO2 had a two-dimensional hexagonal
mesostructure containing uniform pores with a diameter of 4 nm and
5 nm-wide pore walls of nanocrystalline anatase TiO2. A high specific
capacity (about 260 mAh/g) at high current density (10 A/g) has been
obtained for these materials3. Similar results have also been observed
in nanocrystalline anatase TiO2 with a diameter of 6 nm (Fig. 3)22.
A short electronic transport path is also necessary for improving
high current rate charge/discharge performance. Carbon black has
commonly been used as a conductive assistant material in lithium
ion batteries. However, there are some problems in the mechanical
process of mixing the conductive assistant material with the active
electrode material, such as contact area, surface contamination, etc. So
decreasing resistance in high rate charging/discharging by shortening
the electronic transport path is still a challenge. Some chemical
synthesis methods have been developed to synthesize active electrode
materials directly, such as V2O5 24, TiO2 25, MnO2 26 on acetylene black
carbon, for high-rate performance.
Recently, we developed a new method to synthesize nanoporous
active materials such as NiO27, Fe2O328, and Co3O429 on a metal Ni
mesh to form shell-core micro/nanostructured materials for high-rate
performance (see Fig. 4). Tarascon et al.6 first reported that negative
electrodes consisting of NiO, FeO, or CoO have large specific capacities
of ~700 mAh/g at low charge/discharge rates. For shell-core micro/
nanostructured materials, we found that both the discharge and charge
profiles show similar high specific capacities even at high charge/
discharge rates. Very good high-rate capacities of around 695 mAh/g
Fig. 4 Electrode concept for a lithium storage device with both high power density and high energy density. (Reprinted with permission from27. 2006 Elsevier.)
32 NOVEMBER 2006 | VOLUME 1 | NUMBER 4
(at 10 A/g) and 780 mAh/g (at 13 A/g) were achieved using NiO-Ni
and Fe2O3-Ni shell-core active materials, respectively (Fig. 5).
In shell-core micro/nanostructured active materials, the diameter
of the Ni wire is too thick for practical use. Conductive nanowires or
nanotubes with diameters of several nanometers to several tens of
nanometers would be more suitable as a core for electronic transport.
Active nanocrystalline electrode materials synthesized on carbon
nanotubes have been investigated for high-rate lithium ion batteries.
They have shown some improved charge/discharge profiles at high
current densities30. Currently, synthesizing active nanostructured
materials on conductive nanowires and nanotubes is still one of the
most promising areas of research.
Nanostructured electrodes for stable cycle
performance
Besides their excellent high current rate charge/discharge performance,
nanostructured electrode materials may also deliver good cycle
stability3,17,18,22,31. The capacity fading of lithium ion batteries upon
cycling is usually caused by the large volume expansion/contraction
associated with Li insertion/extraction or Li alloying/de-alloying. For
example, Si as the negative electrode of lithium ion batteries has
the highest theoretical capacity of 4200 mAh/g32-34. However, its
commercial use has been hindered by the drastic volume change during
lithium insertion/extraction, which inevitably introduces cracking
or crumbling of the anode material and thus causes loss in capacity
during cycling34. Nanostructured electrodes can absorb this large
volume expansion/contraction, preserving the integrity of the electrode
and leading to stable cycle performance. For example, a new type of
nanostructured Si/C composite electrode shows a very high reversible
capacity (about 1000 mAh/g) and good capacity retention (~99.8%)
from the second cycle35. It is believed that the nanosized composite
electrode has accommodated the large volume change of the Si during
charge and discharge processes, keeping the integrity of the electrode.
 Nanomaterials for lithium ion batteries
Fig. 5 Nanostructure and charge/discharge profiles of NiO-Ni and Fe2O3-Ni shell-core micro/nanostructured materials. (Reprinted with permission from27,28.
2006 Elsevier and The Electrochemical Society.)
Summary
We have reviewed some advantages of nanostructured electrode
materials for high-rate energy storage devices, especially for high-rate
lithium ion batteries. Of course, there are some disadvantages, such as a
REFERENCES
1. Maier, J., et al., Nat. Mater. (2005) 4, 805
2. Arico, A. S., et al., Nat. Mater. (2005) 4, 366
3. Zhou, H., et al., Angew. Chem. Int. Ed. (2005) 44, 797
4. Scrosati, B., Nature (1995) 373, 557
5. Nelson, R. F., J. Power Sources (2000) 91, 2
6. Tarascon, J.-M., et al., Nature (2001) 414, 359
7. Poizot, P., et al., Nature (2000) 407, 496
8. Jamnik, J., et al., Phys. Chem. Chem. Phys. (2003) 5, 5215
9. Zhukovskii, Y. F., et al., Phys. Rev. Lett. (2006) 96, 058302
10. Kim, C., et al., Chem. Mater. (2005) 17, 3297
11. Ohzuku, T., et al., Electrochim. Acta (1979) 24, 219
12. Bonino, F., et al., J. Power Sources (1981) 6, 261
13. Kavan, L., et al., J. Electrochem. Soc. (1999) 146,1375
14. Koudriachova, M. V., et al., Phys. Rev. B (2002) 65, 235423
15. Koudriachova, M. V., et al., Phys. Rev. Lett. (2001) 86, 1275
16. Koudriachova, M. V., et al., Solid State Ionics (2003) 157, 35
17. Jiang, C. H., et al., (2006), submitted
REVIEW
more complex synthesis process for the nanomaterials, which will
increase the cost of lithium ion batteries. Therefore, the next challenge
will be to develop simple synthesis methods for large-scale production
of nanostructured active materials.
18. Hu, Y.-S., et al., Adv. Mater. (2006) 18, 1421
19. Reddy, M. A., et al., Electrochem. Commun. (2006) 8, 1299
20. Stashans, A., et al., Phys. Rev. B (1996) 53, 159
21. Kavan, L., et al., J. Electrochem. Soc. (2003) 150, A1000
22. Jiang, C. H., et al., (2006), submitted
23. Winter, M., et al., Adv. Mater. (1998) 10, 725
24. Kudo, T., et al., Solid State Ionics (2002) 152-153, 833
25. Hibino, M., et al., J. Power Sources (2004) 126, 139
26. Kawaoka, H., et al., J. Power Sources (2004) 125, 85
27. Hosono, E., et al., Electrochem. Commun. (2006) 8, 284
28. Hosono, E., et al., J. Electrochem. Soc. (2006) 153, A1273
29. Hosono, E., et al., Presented at International Meeting on Lithium Batteries, Biarritz,
France, June 18-23, 2006
30. Moriguchi, I., et al., Adv. Mater. (2006) 18, 69
31. Liu, P., et al., Adv. Mater. (2002) 14, 27
32. Sharma, R. A., et al., J. Electrochem. Soc. (1976) 123, 1763
33. Boukamp, B. A., et al., J. Electrochem. Soc. (1981) 128, 725
34. Yang, J., et al., Solid State Ionics (1996) 90, 281
35. Holzapfel, M., et al., Chem. Commun. (2005), 1566
NOVEMBER 2006 | VOLUME 1 | NUMBER 4 33