Unconventional Mechanisms in Niobium Tungsten Oxides for High-rate
Lithium-ion Charge Storage
Kent J. Griffith
1
, Kamila M. Wiaderek
2
, Giannantonio Cibin
3
, Lauren E. Marbella
1
, Clare P.
Grey
1
*
1
Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, UK.
2
X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, Argonne, IL
60439, USA.
3
Diamond Light Source, Harwell Science and Innovation Campus, Didcot, Oxfordshire OX11
0DE, U.K.
*Correspondence to: cpg27@cam.ac.uk
The maximum power output and minimum charging time of a lithium-ion battery – key
parameters for its use in, for example, transportation applications – depend on mixed ionic–
electronic diffusion. While the discharge/charge rate and capacity can be tuned by varying
the composite electrode structure, ionic transport within the active particles represents a
fundamental limitation. Thus, to achieve high rates, particles are frequently reduced to
nanosize dimensions despite this being disadvantageous in terms of volumetric packing
density as well as cost, stability, and sustainability considerations. As an alternative to
nanoscaling, we show that complex niobium tungsten oxides with topologically frustrated
polyhedral arrangements and dense
µ
m-scale particle morphologies can rapidly and
reversibly intercalate large quantities of lithium. Multielectron redox, buffered volume
expansion, and extremely fast lithium transport approaching that of a liquid lead to
extremely high volumetric capacities and rate performance for both crystallographic shear
structure and bronze-like niobium tungsten oxides. The active materials Nb
16
W
5
O
55
and
Nb
18
W
16
O
93
offer new strategies toward designing electrodes with advantages in energy
density, scalability, electrode architecture/complexity and cost as alternatives to the state-of-
the art high-rate anode material Li
4
Ti
5
O
12
. The direct measurement of solid-state lithium
diffusion coefficients (D
Li
) with pulsed field gradient NMR demonstrates room temperature
D
Li
values of 10
–12
–10
–13
m
2
×
s
-
1
in the niobium tungsten oxides, which is several orders-of-
magnitude faster than typical electrode materials and corresponds to a characteristic
diffusion length of ~10
µ
m for a 1 minute discharge. Materials and mechanisms that enable
lithiation of
µ
m particles in minutes have implications for high power applications, fast
charging devices, all-solid-state batteries, and general approaches to electrode design and
material discovery.
2
New high-rate lithium-ion battery electrode materials that can store large quantities of charge in a
few minutes, rather than hours, are required to increase the power and decrease the charging time
to help alleviate technological challenges associated with the adoption of electric vehicles and
grid-scale batteries, and to enable new power-intensive devices. The most intuitive and commonly
used approach to increase the rate performance is to create nanosized or porous (and often
hierarchical) structures, which minimize Li
+
solid-state diffusion distances, enable more rapid Li
+
transport through the composite electrode and increase the surface areas of electrode materials in
contact with electrolyte. Carbonaceous hierarchical structures and carbon-coating are also
frequently employed to improve electronic conductivity, which is another prerequisite for the
application of high current densities.
In practice, despite excellent lithium mobility, graphite cannot be used at high-rates due to particle
fracture and the risk that Li dendrites form, leading to short circuits and potentially fires and
explosions
1–3
. The latter issue inherently limits the use of low voltage anodes in high-rate
applications, since the electrode inhomogeneity or any source of increased overpotential can lead
to Li plating potentials on the surface of the electrode
3
. Li
4
Ti
5
O
12
, with an average voltage of 1.55
V (Fig. S1), enables high-rate (de)intercalation without the risk of Li dendrites or substantial solid–
electrolyte interphase (SEI) formation albeit with an undesirable but necessary decrease in full-
cell voltage and thus energy density. In this well-established “high”-voltage/high-rate anode, the
capacity of 1 µm particles from solid-state synthesis reaches only 60–65 mA·h·g
–1
at a rate of
10C
4
, where C-rate is defined as the inverse of the number of hours to reach a defined theoretical
capacity e.g., 10C corresponds to a 6 min discharge or charge time (see Methods). In contrast,
through two decades of research, present carbon-coated nanoparticles of LTO can reach at least
150 mA·h·g
–1
at 10C,
5,6
150 mA·h·g
–1
corresponding to approximately 0.5 lithium ions per
transition metal (Li
+
/TM). However, using nano- and porous materials for electrochemical energy
storage applications inherently results in a severe penalty in terms of volumetric energy density.
Furthermore, these carefully designed porous and nano-architectures are time consuming and
expensive to synthesize, characterize, and manufacture, synthesis methods often resulting in
relatively low yields and/or the generation of large quantities of chemical waste
7
while also
unfortunately being more susceptible to degradation during electrochemical cycling (from
processes such as e.g. catalytic decomposition of electrolyte,
8
morphological changes that result
in loss of nanostructuring,
9
and higher first cycle capacity loss
10
).
In this work, we break from the conventional strategy of nanoscaling and nanostructuring of
electrode materials to overcome poor diffusion and electronic properties (found in e.g., TiO
2
and
LTO), and demonstrate that, with the appropriate host lattice, strictly none of the usual criteria
discussed above are required to achieve a practical high-rate battery electrode. Instead we leverage
insight obtained from prior investigations of complex binary niobium oxides such as T-Nb
2
O
5
11
and of superionic conductors such as lithium lanthanum titanate (LLTO) perovskite
12
, to identify
structural motifs that should exhibit favorable Li diffusion properties, and thus exhibit superior
performance, allowing micrometer-sized particles to be used at extremely high rates. We show that
when multi-redox 4d and 5d transition metals are used with the appropriate three-dimensional
oxide structure, we can achieve extremely high volumetric energy densities and impressive rates.
The bulk compounds studied are a series of complex “block” or “bronze-like” oxide structures
(Fig. 1) largely comprised of corner- and edge-sharing NbO
6
and WO
6
octahedra, both oxides
prepared via gram-scale solid-state synthesis. The unusual electrochemical performance is first
illustrated by studying large (3-10 µm primary, 10-30 µm agglomerate) dense particles of the
3
block structure Nb
16
W
5
O
55
(Fig. 1a–c). On a mass normalized basis, the lithium storage
performance of Nb
16
W
5
O
55
exceeds nanostructured versions of heavily studied Li
4
Ti
5
O
12
5,13–15
,
TiO
2
(B)
16–18
, and T-Nb
2
O
5
19–21
under similar loading conditions. Given the high density of the
crystal structure and the high tap density of bulk Nb
16
W
5
O
55
vs. nanomaterials, this leads to
exceptionally high volumetric performance. We further demonstrate the generality of this bulk
phenomenon by exploring another new electrode material, bronze-like Nb
18
W
16
O
93
(Fig. 1d–f).
Figure 1 | Crystal structure and particle morphology of Nb
16
W
5
O
55
and Nb
18
W
16
O
93
. a–c, Nb
16
W
5
O
55
is built up from blocks of 4 × 5 octahedra with the blocks adjoined forming crystallographic shear planes.
d–f, Nb
18
W
16
O
93
is a 1 × 3 × 1 superstructure of the tetragonal tungsten bronze with pentagonal tunnels
partially filled by –W–O– chains that form pentagonal bipyramids.
Nb
16
W
5
O
55
is a metastable member within the Nb
2
O
5
–WO
3
phase diagram
22
with a monoclinic
structure comprised of subunits of corner-shared octahedra arranged into ReO
3
-like blocks, four
octahedra wide by five octahedra long, and infinite in the third dimension (Fig. 1a).
23
The block
subunits are connected by crystallographic shear planes along the edges and by tetrahedra at each
corner leading to the notation (4 × 5)
1
where, in (m × n)
p
, m and n denote block length in units of
octahedra and p relates to the connectivity of the blocks which may also be joined in pairs (p = 2)
or infinite ribbons (p = ∞). To the best of our knowledge, this is the first reported application, of
any kind, for Nb
16
W
5
O
55
since its discovery in 1965.
23,24
Nb
18
W
16
O
93
is orthorhombic, a
1 × 3 × 1 superstructure of the classic tetragonal tungsten bronze (Fig. 1d, S2). The superstructure
25
results from partial filling of pentagonal tunnels by –M–O– chains to form pentagonal bipyramids
in addition to the distorted octahedra of the tetragonal tungsten bronzes. Nb
16
W
5
O
55
and
Nb
18
W
16
O
93
were prepared via co-thermal oxidation of pellets of NbO
2
and WO
2
. Details and
alternative synthetic routes are described in the Methods (Fig. S3–7).
Reaction of Nb
16
W
5
O
55
with lithium (Fig. 2a) proceeds in three regions from 2.5 V to 1.0 V, with
an average voltage of 1.57 V (Fig. S1), comparable to the average voltage of Li
4
Ti
5
O
12
of 1.55
V
26
. The three regions, more easily observed in the derivative plot (Fig. 2b), are characterized by
their slope and are reminiscent of the three regions observed in other crystallographic shear
5 μm
10 μm
5 μm
10 μm
d
b
e
c
f
a
4
structures
27
, e.g., H-Nb
2
O
5
28
, PNb
9
O
25
29
, TiNb
2
O
7
30
, and Nb
12
WO
33
31
. When the kinetics were
examined over a range of current densities from C/5 (34.3 mA×g
–1
) up to 60C (10.3 A×g
–1
),
Nb
16
W
5
O
55
showed unprecedented bulk rate performance in standard electrode formulations (See
Methods) (Fig. 2a–b,e, S8). At C/5, around 1.3 Li
+
can be reversibly intercalated per transition
metal for a gravimetric capacity of ca. 225 mA·h·g
–1
. When the rate is increased by a factor of 25
to 5C, Nb
16
W
5
O
55
maintains a capacity of 1.0 Li
+
/TM (171 mA·h·g
–1
). At 20C, which corresponds
to a three minute discharge, it is still possible to exchange 0.86 Li
+
/TM and access 148 mA·h·g
–1
.
Rate tests on Nb
16
W
5
O
55
were measured with a potentiostatic hold at the top of charge to ensure a
reliable starting point for discharge. To test the performance under more demanding conditions,
1000 cycles were measured with fixed galvanostatic discharge and charge conditions of 10C for
250 cycles followed by 20C for 750 cycles with no voltage hold (Fig. 2f). Under these conditions,
0.90 Li
+
/TM (avg. 155 mA·h·g
–1
) were reversibly intercalated at 10C with 95% capacity retention
after 250 cycles on non-optimized or calendared electrodes. At 20C, the capacity was 0.75 Li
+
/TM
(avg. 128 mA·h·g
–1
); the capacity retention was again 95% over the 750 cycles at 20C.
In an extension to another niobium tungsten oxide with distinct structural motifs, excellent
electrochemical energy storage was also discovered with µm-scale particles of the bronze-like
phase Nb
18
W
16
O
93
(Fig. 1c–f), with enhanced rate performance at the highest rates (Fig. 2e–f). The
average voltage of Nb
18
W
16
O
93
is 1.67 V (Fig. S1). In terms of gravimetric capacity, Nb
18
W
16
O
93
stores ca. 20 mA×h×g
–1
less than Nb
16
W
5
O
55
at C/5 and 1C due to the higher molar mass of the
tungsten-rich bronze phase. However, at 20 C, Nb
18
W
16
O
93
is still able to accommodate a full unit
Li
+
/TM for a capacity of ca. 150 mA·h·g
–1
. At 60C and 100C (14.9 A×g
–1
), the capacity is still 105
and 70 mA·h·g
–1
, respectively.
Other cycling conditions such as long term cycling at C/5 and the effect of current collectors,
which cannot be ignored at high rates
32
, were examined (Fig. S9–10). As a control, Li || Li
symmetric cells were cycled at current densities corresponding to those for C/5 to 100C in Fig. 2c
(Fig. S11). The overpotentials in the symmetric cell closely match those observed in the
electrochemical cycling curves of Fig. 2a–d. This suggests that the extremely high rates for a bulk
electrode are approaching the limits of Li metal plating/stripping and/or lithium-ion desolvation
and transport in carbonate ester electrolytes at room temperature, i.e., a significant fraction of the
ohmic drop during fast charging results from the Li metal and electrolyte rather than the complex
oxide electrode materials.
5
Figure 2 | Electrochemistry of Nb
16
W
5
O
55
and Nb
18
W
16
O
93
. Galvanostatic discharge and charge curves
and dQ/dV plots of bulk a–b, Nb
16
W
5
O
55
and c–d, Nb
18
W
16
O
93
from C/5 to 100C. e, Rate performance
summary based on gravimetric capacity. f, High-rate cycling of 250 cycles at 10C followed by 750 cycles
at 20C. Dense electrodes of large particles with 2–3 mg×cm
-
2
active mass loading were tested at current
densities corresponding to discharge times of several hours to tens of seconds. Nb
16
W
5
O
55
was charged with
a 1 h constant voltage step at the top of charge to ensure a comparable starting point on discharge;
Nb
18
W
16
O
93
was cycled without this step and stored over 100 mA×h×g
-
1
at 60C (i.e., in <60 s). High-rate
cycling for 1000 cycles was performed on both oxides at 10C/20C constant current without any
potentiostatic step.
The extremely high mobility of Li
+
ions in these systems enabled the direct measurement of lithium
diffusion with the pulsed field gradient (PFG) NMR technique, which has previously only been
performed on liquids or diamagnetic superionic solid electrolytes (Table S2). To combat the short
T
2
(spin–spin) relaxation times of the Li ions, which generally prevent these measurements in
electrode materials, experiments were performed from 333–453 K. Analysis of the data (Fig. S12–
14, Methods and Supplementary Text) for Li
x
Nb
16
W
5
O
55
(x = 6.3, 8.4) and Li
x
Nb
18
W
16
O
93
(x =
3.4, 6.8, 10.2) showed lithium transport above 10
–13
m
2
×s
-1
at 333 K and 10
–12
m
2
×s
-1
at 373 K.
Assuming Arrhenius behavior, and the extremely low measured activation energies of 0.10–0.30
eV, the room temperature lithium diffusion coefficients are estimated to be > 1×10
–13
m
2
×s
-1
for
all the materials (Table 1). The different samples exhibited similar diffusion coefficients and
activation energies, consistent with the transport measurements made via the use of the
galvanostatic intermittent titration technique (GITT) (Supplementary Materials; Fig. S15) at low
to moderate lithium contents. Lithium diffusion in both niobium tungsten oxide structures is
a
d
b
e
c
f
