of 21
A High Energy Density Li-ion Battery Cathode
Using Only Industrial Elements
Eshaan S. Patheria,
Pedro Guzman,
Leah S. Soldner,
Michelle D.
Qian,
Colin T. Morrell,
Seong Shik Kim,
Kyle Hunady,
Elena R. Priesen
Reis,
Nicholas V. Dulock,
James R. Neilson,
Johanna Nelson Weker,
§
Brent Fultz,
and Kimberly A. See
,
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena,
California 91125, United States
Department of Applied Physics and Materials Science, California Institute of Technology,
Pasadena, California 91125, United States
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523-1872,
United States
§
Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, Menlo
Park, California 94025, United States
E-mail: ksee@caltech.edu
Abstract
Li-ion batteries are crucial for the global energy
transition to renewables, but their scalability is
limited by the supply of key elements used in com-
mercial cathodes (e.g., Ni, Mn, Co, P). There-
fore, there is an urgent need for next-generation
cathodes composed of widely available and in-
dustrially scalable elements. Here, we introduce
a Li-rich cathode based on the known material
Li
2
FeS
2
, composed of low-cost elements (Al, Fe,
S) that are globally mined and refined at indus-
trial scale. The substitution of redox-inactive Al
3+
for Fe
2+
achieves remarkably high degrees of an-
ion redox, which in turn yields high gravimet-
ric capacity (
450 mAh
·
g
1
) and energy den-
sity (
'
1000 Wh
·
kg
1
). We show that Al
3+
en-
ables high degrees of delithiation by stabilizing
the delithiated state, suppressing phase transfor-
mations that would otherwise prevent deep delithi-
ation and extensive anion redox. This mechanis-
tic insight offers new possibilities for developing
scalable, next-generation Li-ion battery cathodes
to meet pressing societal needs.
Introduction
It is estimated that between 100 to 400 TWh of
energy storage are needed to decarbonize/electrify
global transport and energy sectors by 2050.
1–3
To
achieve that goal with commercial Li-ion batteries
with LiNi
x
Mn
y
Co
z
O
2
(NMC
xyz
) cathodes, Ni and
Co production must double their respective maxi-
mum historical compound annual growth rates for
every year until 2050.
2
Although Mn production
is greater than Ni and Co production, limited re-
fining capacity for ‘battery-grade’ Mn forecasts
supply shortages by 2030.
3,4
Even with LiFePO
4
(LFP), refinement bottlenecks for battery-grade P
imply supply shortages by 2030.
5
Cathodes reliant
only on industrial metals, or industrial elements,
would alleviate the supply challenges that impede
the ‘net zero by 2050’ goal. We classify ‘indus-
trial elements’ as elements with global production
of at least
10
7
metric tons in 2023 in primarily el-
emental form with
'
90 wt% purity. For exam-
ple, Al, Fe, and S all meet these criteria, while Ni,
Co, Mn, and P do not.
6,7
While Li itself does not
meet the criteria, ‘beyond Li-ion’ batteries (e.g.,
Na-ion, aqueous Zn-ion, etc.) without any Li re-
1
https://doi.org/10.26434/chemrxiv-2024-vvwn4
ORCID:
https://orcid.org/0000-0002-0133-9693
Content not peer-reviewed by ChemRxiv.
License:
CC BY-NC-ND 4.0
quire new infrastructure, time, and investment to
reach scale.
4,8,9
By contrast, next-generation Li-
ion battery cathodes that contain only industrial
elements, except for Li, could scale faster and at
lower capital expenditure (CapEx) into Li-ion bat-
teries by using existing infrastructure, just as Si
anodes have already entered the market.
4
Fe is the most globally produced transition
metal, motivating research to develop high-
performance cathodes that leverage Fe redox.
The resurgence of LFP in commercial applica-
tions stems from its lower cost and more industrial
element-like composition compared to NMC,
10
despite LFP’s low energy density (
580 Wh
·
kg
1
,
2068 Wh
·
L
1
)
11
compared to, for example,
NMC811 (
950 Wh
·
kg
1
,
4500 Wh
·
L
1
).
12
Just over a decade ago, efforts to develop Fe-
based cathodes that outperform LFP sought to
increase the voltage of Fe
2+/3+
redox. By means of
iono-covalency/inductive effects, the voltage can
be shifted by over
1.1 V
13
to a maximum of 3.9
V vs. Li/Li
+
in triplite LiFeSO
4
F.
14
However, the
energy density of LiFeSO
4
F remained close to that
of LFP, limiting its commercial viability.
15
More
recently, Heo et al. over-discharged amorphous
LiFeSO
4
F, achieving 906 Wh
·
kg
1
.
16
However,
this required converting LiFeSO
4
F to Li
2
O, Fe
0
,
and LiSO
3
F, and also required Li
+
at the anode in
the as-assembled cell – requirements incompatible
with current manufacturing techniques. Overall,
high voltage Fe
2+/3+
redox in Fe-based cathodes
has been unable to match the energy density of
NMC, and high energy density requires conver-
sion reactions.
Multielectron transition metal and anion redox
processes in Li-rich materials invoke both inter-
calation and bond-forming/breaking reactions,
17
surpassing capacity limits of traditional single-e
transition metal redox. However, stabilizing the
delithiated, oxidized state remains a key challenge.
Multielectron redox increases energy density by
increasing capacity, requiring reversible redox re-
actions even at deep delithiation levels. Initial
delithiation involves ‘transition metal oxidation,’
emptying associated covalent
d
-
p
states.
18,19
Deep
delithiation, however, yields under-coordinated
anions, creating associated nonbonding
p
states
near the Fermi level.
18,20–23
Often, the empty
d
-
p
states lie below the filled nonbonding
p
states, trig-
gering anion to metal charge transfers that, in ox-
ides, create reactive O peroxides/superoxides and
promote O
2(
g
)
release.
20,21,24–27
The electronic re-
organization and structural changes hinder elec-
trochemically mediated anion redox involving the
nonbonding
p
states. This issue is acute in Li-rich
Fe-based oxides, specifically Li
1.17
Ti
0.33
Fe
0.5
O
2
21
and Li
1.33
Fe
0.33
Sb
0.33
O
2
,
24
where deep delithiation
incurs charge transfers from O
2 –
to Fe
3+/4+
, asso-
ciated with large hysteresis (
1.4 V) and capacity
fade (
20% per cycle).
Figure 1:
The crystal structure of (a) Li
2
FeS
2
projected along the
c
-axis
(top) and
b
-axis (bottom), and the crystal structure of (b) Li
5
AlS
4
projected
along the
c
-axis (top) and
b
-axis (bottom). In each panel, the solid black
line indicates the unit cell of the structure shown, while the dashed black
line indicates the unit cell of the other structure for comparison.
Here, to develop a Li-ion battery cathode
entirely composed of industrial elements that
achieves high energy density through multielec-
tron redox, we target Li-rich, Fe-based sulfides
derived from Li
2
FeS
2
.
Sulfides must have ex-
tremely high capacities and thus high Li content
to match the energy density of NMC because S
is heavier and less electronegative than O. For
example, Li
1.13
Ti
0.57
Fe
0.3
S
2
achieves only up to
600 Wh
·
kg
1
,
28
despite having greater capac-
ity in mole e
per f.u. than NMC cathodes. The
crystal structure of Li
2
FeS
2
,
29
shown in Figure 1a
along the
c
and
b
axes, adopts the P
̄
3
m1 trigonal
space group with a hexagonal close-packed (HCP)
sulfide anion framework and cations alternating
between octahedral and tetrahedral sites in lay-
2
https://doi.org/10.26434/chemrxiv-2024-vvwn4
ORCID:
https://orcid.org/0000-0002-0133-9693
Content not peer-reviewed by ChemRxiv.
License:
CC BY-NC-ND 4.0