New Route to Battery Grade NaPF6 for Na-Ion Batteries: Expanding the Accessible Concentration

Article abstract of DOI:10.1002/anie.202111215

Sodium-ion batteries represent a promising alternative to lithium-ion systems. However, the rapid growth of sodium-ion battery technology requires a sustainable and scalable synthetic route to high-grade sodium hexafluorophosphate. This work demonstrates a new multi-gram scale synthesis of NaPF₆ in which the reaction of ammonium hexafluorophosphate with sodium metal in THF solvent generates the electrolyte salt with the absence of the impurities that are common in commercial material. The high purity of the electrolyte (absence of insoluble NaF) allows for concentrations up to 3 M to be obtained accurately in binary carbonate battery solvent. Electrochemical characterization shows that the degradation dynamics of sodium metal-electrolyte interface are different for more concentrated (>2 M) electrolytes, suggesting that the higher concentration regime (above the conventional 1 M concentration) may be beneficial to battery performance.

Full text of DOI:10.1002/anie.202111215

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: New Route to Battery Grade NaPF6 for Na-Ion Batteries:
Expanding the Accessible Concentration
Authors: Darren M. C. Ould, Svetlana Menkin, Christopher A. O’Keefe,
Fazlil Coowar, Jerry Barker, Clare P. Grey, and Dominic
Wright
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202111215
Link to VoR: https://doi.org/10.1002/anie.202111215

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
New Route to Battery Grade NaPF
the Accessible Concentration
6
for Na-Ion Batteries: Expanding
Darren M. C. Ould,[a,b] Svetlana Menkin,[a,b] Christopher A. O’Keefe,[a,b] Fazlil Coowar, Jerry Barker,
[c]
[c]
Clare P. Grey*[a,b] and Dominic S. Wright*[a,b]
[
a]
Dr. D. M. C. Ould, Dr. S. Menkin, Dr. C. A. O’Keefe, Prof. C. P. Grey and Prof. D. S. Wright
Yusuf Hamied Department of Chemistry
University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, U.K.
E-mail: cpg27@cam.ac.uk dsw1000@cam.ac.uk
[
b]
c]
Dr. D. M. C. Ould, Dr. S. Menkin, Dr. C. A. O’Keefe, Prof. C. P. Grey and Prof. D. S. Wright
The Faraday Institution, Quad One, Harwell Science and Innovation Campus, Didcot, U.K.
Dr. F. Coowar and Dr. J. Barker
[
Faradion Limited, The Innovation Centre, 217 Portobello, Sheffield, S1 4DP, U.K.
Supporting information for this article is given via a link at the end of the document.
Abstract: Sodium-ion batteries represent a promising alternative to
lithium-ion systems. However, the rapid growth of sodium-ion battery
technology requires a sustainable and scalable synthetic route to
high-grade sodium hexafluorophosphate. This work demonstrates a
safety of the battery, for which direct sodium analogues of lithium
electrolytes are normally used.[8–12] From the selection of sodium
salts currently available, the present benchmark is sodium
hexafluorophosphate,
although optimum
solvents
and
concentrations are still widely debated.[ Organic carbonate-
based solvents are typically preferred, in part due to their high
dielectric constant and low viscosity, which aids solubility and
gives high ionic mobility. Recent work by Younesi and colleagues
has shown that ethylene carbonate: diethyl carbonate (EC:DEC)
may be advantageous as a solvent due to aiding a more stable
solid-electrolyte interphase (SEI), whereas propylene carbonate
(PC) leads to a less stable SEI due to the higher solubility of SEI
13]
new multi-gram scale synthesis of NaPF
6
in which the reaction of
ammonium hexafluorophosphate with sodium metal in THF solvent
generates the electrolyte salt with the absence of the impurities that
are common in commercial material. The high purity of the electrolyte
(absence of insoluble NaF) allows for concentrations up to 3 M to be
obtained accurately in binary carbonate battery solvent.
Electrochemical characterization shows that the degradation
dynamics of sodium metal–electrolyte interface are different for more
concentrated (>2 M) electrolytes, suggesting that the higher
concentration regime (above the conventional 1 M concentration) may
be beneficial to battery performance.
2 3
components, such as NaF and Na CO
.[14]
The ever-growing demand for electrical energy production and
large-scale grid storage has led to a surge in research efforts in
recent years into sodium-ion batteries (SIBs).[1–5] Although the
lithium-ion battery (LIB) field currently dominates and is well
established, arguably one of the most attractive features of
moving towards SIBs is the significantly higher abundance of
sodium compared to lithium. Moreover, these sodium deposits
are evenly distributed worldwide so are less susceptible to
geopolitical price fluctuations and are much cheaper than
lithium.[6] Although the inherently lower energy density has
previously plagued SIBs, for large-scale or stationary storage
applications this is less of a critical consideration as here, the low
cell cost dominates. Given the low abundance and relatively high
cost of lithium, it is envisioned that the development of SIBs for
grid storage will be critical to relieve the pressure on LIBs, as the
surging rise of portable electronic devices and electric powered
vehicles is likely to place an insurmountable burden on lithium
supplies.[7] Furthermore, the use of sodium has other
sustainability implications, which are particularly important for
grid-scale applications, as it allows cobalt-free cathodes to be
used and the copper current collectors at the anode to be
replaced by aluminium.
6
Figure 1. Top: Previous synthetic methods of producing NaPF . Bottom: This
work using ammonium hexafluorophosphate and sodium metal.
As sodium sits below lithium in Group 1 of the periodic table,
knowledge of the LIB field can help accelerate research and
development of SIBs as the general principles are the same for
both batteries. This is true in particular of the electrolyte, which
plays a crucial role for the overall lifetime, rate capability and
Sodium hexafluorophosphate has a number of attractive
features as a salt compared to the other commonly used sodium
salts. Nevertheless, studies have shown that NaPF
hydroscopic and readily undergoes hydrolysis to produce NaF,
HF and other phosphate species (such as POF
).[15,16]
6
is highly
3
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
Remarkably, this hydrolysis can still take place in battery-grade
electrolytes which have <20 ppm of water present,[15] highlighting
(NH
4
PF
6
) was instead employed, which has the additional benefit
of reduced cost compared to NOPF
synthesized NaPF
).[28]
4 6
NaPF was synthesized by firstly dissolving NH PF in THF
6
(see SI for cost analysis of
the highly hydroscopic nature of NaPF
6
. Consequently, this
may be
6
means that the quality of commercially purchased NaPF
6
6
variable and depends on the degree of hydrolysis that has
occurred (either through storage or as a result of the synthetic
route used to make it). The presence of impurities has a
detrimental effect on use in battery research, as these highly
insoluble hydrolysis products crucially affects the real electrolyte
concentration. Bhide and Adelhelm in recent work reported that
solvent. Once dissolved, this was added dropwise to freshly cut
metallic sodium suspended in THF, which immediately gave
effervescence due to the release of dihydrogen and ammonia gas
as the only by-products. The reaction was heated to reflux for six
hours, then left to stir for a further sixteen hours at 50 °C, which
after removing the excess sodium by filtration and drying under
when using a high purity (99%) supply of NaPF
6
, an insoluble
6
vacuum produced NaPF in an excellent yield of 89% (Scheme 1).
fraction appeared beyond 0.4 M concentration when dissolving in
ethylene carbonate: dimethyl carbonate (EC:DMC 30:70 wt%).
Subsequent X-ray diffraction of the mixture found evidence of NaF
Two equivalents of sodium metal were found to be optimal for the
synthesis as using stoichiometric or a small excess (1.2
equivalents) failed to consistently give complete conversion using
the above stated conditions. Furthermore, the scalability of this
(
which is also an important solid electrolyte interphase (SEI)
[
17]
component) as the insoluble impurity. Resolving this solubility
issue of NaPF is therefore of primary importance for the SIB field
6
protocol has been shown by producing 24 g of NaPF (See SI).
6
as fundamental properties of the electrolyte are dependent upon
the ions concentration.[13,17]
6
As commercially purchased NaPF may contain insoluble
impurities, it would be desirable to synthesize it under anhydrous
conditions and immediately use it, thus preventing NaF formation.
However, the literature contains few methods to do this and most
involve using HF, which has significant safety concerns. One
reported synthesis by Woyski is the fluorination of a mixture of
Scheme 1. Synthesis of sodium hexafluorophosphate from metallic sodium and
ammonium hexafluorophosphate.
with liquid HF,[18] whereas Clifford’s synthesis
6
Initial characterization of the NaPF salt was performed
using solution-state NMR spectroscopy (Figures S2.1.1–S2.1.3).
Analysis of both the 31P NMR and 19F NMR spectra revealed the
sodium and PCl
involves the formation of PF
5
5
from red phosphorus, HF and ClF
3
and subsequent reaction with NaF.[19] A patented method from
expected septet and doublet signals at −144.6 ppm and −73.2
South Africa Nuclear Energy involves the reaction of pyridinium
hexafluorophosphate, synthesized from HF, phosphoric acid and
pyridine, with sodium hydroxide,[20] whereas Astruc has shown
−
ppm respectively, confirming the presence of the PF
6
anion.
1
More informative was the H NMR spectrum, which showed the
absence of an ammonium signal, which appears at ca. 5.9 ppm.
Furthermore, on account of the strong donor ability of the THF
solvent, the 1H NMR spectrum showed residual THF solvent
signals at 3.64 ppm and 1.79 ppm, despite drying under vacuum
for 16 hours. The addition of 1,4-dimethoxybenzene as an internal
standard showed these THF signals account for ca. 3% of the
sample (Figure S2.1.4).
NaPF
hexafluorophosphate with sodium amalgam at −20 °C (Figure
).[21] Clearly these synthetic routes all contain a number of
hazardous or expensive steps that make the synthesis of NaPF
inaccessible without special control measures and specialized
equipment. Alternatively, electrolyte solutions of NaPF in
6
may be produced from reduction of CpFe(II)(arene)
1
6
6
carbonate solvent may be commercially purchased, but their high
cost can be prohibitive.
Attempts were made to fully remove the residual solvent by
drying under vacuum at both ambient temperature and at 160 °C
for 72 hours and 24 hours respectively, but both cases still
showed THF solvent signals in the solution-state 1H NMR
,[22,23] where
Similar synthesis routes exist for producing LiPF
early methods involved the addition of LiF to PF in anhydrous
6
5
HF.[24] This protocol may be used in acetonitrile solvent, but gives
a tetra-coordinated solvated lithium cation.[25] The solvated
6
spectrum. However, precipitating NaPF from a concentrated
solution of THF with pentane and drying under vacuum at 120 °C
for 24 hours was found to reduce the intensity of the THF signals,
albeit not fully removing them (see SI and Figure S2.1.7).
Additionally, the scope of this synthetic procedure was also
Li(CH
3
CN)
4
PF
6
salt can also be produced from the reaction of
with LiOH.[
23]
HPF
6
6
Alternatively, unsolvated LiPF may be
synthesized from direct fluorination from LiF and phosphorus in a
fluorine atmosphere.[26]
investigated by synthesizing KPF
6
. This proceeded in analogous
PF to
Herein, we address the urgent need to produce high grade
fashion to the NaPF synthesis, with the addition of NH
6
4
6
NaPF
simple
6
that is suitable to act as an electrolyte salt, showcasing a
and inexpensive method using ammonium
metallic potassium in THF solvent; solution-state NMR
−
spectroscopy again confirmed the presence of the PF
loss of NH PF
With NaPF
6
anion and
hexafluorophosphate and metallic sodium. This method does not
involve the use of HF or aqueous conditions and can be readily
performed in a laboratory setting without specialist or expensive
equipment.
Previously our group has reported that magnesium
hexafluorophosphate can be synthesized from nitrosonium
4
6
starting material (SI and Figures S2.1.8–S2.1.10).
in hand, solid-state NMR (SSNMR)
6
spectroscopy was used to assess the purity of the salt as it allows
the identification of insoluble chemical species, even if they are
1
present at low concentrations (Figure 2). Figure 2a shows the H
solid-state NMR spectrum, which consists of several signals with
low signal-to-noise ratios, suggesting a low concentration of
proton-containing species. These signals are assigned to the
minor amount of THF that is incorporated into the material, with
hexafluorophosphate (NOPF
6
) with magnesium turnings in
acetonitrile solvent.[27] However, this synthetic route is
inappropriate for producing NaPF as sodium metal reacts with
acetonitrile and other solvents were found to react with NOPF . In
search of a new protocol, ammonium hexafluorophosphate
6
6
the appearance of several peaks being
a
result of
crystallographically distinct protons in the THF molecules. The 19F
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
spectrum has a doublet centered at –81.1 ppm with overlapping,
low-intensity multiplets underneath (Figure 2b). The additional
signals are likely due to minor amounts of hydrolysis product
more vigorous stirring for 30 mins was required to obtain a clear
solution. Saturation of the solution occurs at higher concentration,
with the 3.5 M concentration electrolyte having very high viscosity
2–
–
impurities (e.g., PO
2
F
2
, POF
3
); however, the overlap with the
6
and gel-like appearance, as well as containing undissolved NaPF .
NaPF
6
signal precludes quantification of these impurities using
The electrolyte solutions become pale yellow in colour and less
transparent as the concentration increases beyond 1 M (Figure
S4.1.1).
1
9F NMR spectroscopy. On the other hand, in the 31P SSNMMR
spectrum (Figure 2d), the signals for NaPF
impurities (16 ppm) are well separated. Integration of the two
signals gives a NaPF : impurity ratio of 1:0.0071, indicating a high
purity sample. The 23Na SSNMR spectrum has a single peak at –
0.6 ppm, consistent with NaPF (Figure 2c), with no evidence for
NaF. Conversely, multinuclear SSNMR spectra recorded on a
6
(–145 ppm) and the
6
2
6
commercial sample of NaPF
tentatively assigned to NaHF
6
showed the presence of impurities;
(Figures S2.2.5–S2.2.7).
2
6
Figure 3. Electrolyte conductivity vs. concentration of NaPF electrolytes.
Measured in RHD instruments TSC 1600 Closed conductivity cell, the cell
constant (K) is 13 cm-1 at T = 35 °C using impedance spectroscopy 1MHz–1Hz
10 mV amplitude.
6
Figure 4. Impedance for 2 M NaPF electrolyte (EC:DEC 1:1 v/v) from Na metal
symmetric cell in coin cell.
The absence of insoluble impurities in our synthesized
NaPF allows for the accurate study of ultra-concentrated
6
electrolytes, with the maximum electrolyte conductivity obtained
with 1 M electrolyte solution (19.0 mS/cm at T = 35 °C), larger
than the conductivity values of 8.2, 16.1 and 11.0 mS/cm for 0.5
M, 2.0 M and 3.0 M electrolyte solutions, respectively (Figure 3).
Electrolyte conductivity and the aging of the native solid-
electrolyte interphase (SEI) were then measured in Na-Na
symmetric cells in the concentration range 1–3 M using
electrochemical impedance spectroscopy (EIS) measurements
between 1 MHz and 0.1 Hz (Figures 4 and S3.4–3.6).
Consecutive impedance spectra were recorded for 70 hours after
cell assembly and Nyquist plots accumulated during the first 70
hours from cell assembly are shown in Figure 4. It is assumed that
the main contribution to the low-frequency conductivity is the
sodium SEI, thus the evolution of the low-frequency conductivity
is attributed to the aging, reorganization and dissolution of the SEI
Figure 2. (a) 1H, (b) 19F, (c) 23Na, and (d) 31P solid-state NMR spectra of
synthesized NaPF . The inset in panel (d) shows an expanded view with the
6
signal of the hydrolysis products visible.
As the level of hydrolysis and degradation products present
in the NaPF
achieved, we looked to determine the concentration range that
could be produced from our high purity NaPF salt in a binary
6
salt affects the electrolyte concentration that can be
6
mixture of ethylene carbonate: diethyl carbonate (EC:DEC 1:1
v/v) solvent. A 1 M solution could easily be obtained with minimal
stirring at room temperature, giving a transparent electrolyte
solution within minutes. A similar story was found when making a
2
M solution, however upon increasing the concentration to 3 M
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
during rest (Figure S3.7). The interface impedance for the ultra-
concentrated (>2 M) electrolytes is around 300 Ω (6.3 µS/cm),
while for 1 M it is around 500 Ω (3.8 µS/cm) at the same storage
time. More importantly, the interface stabilization is slower for the
6
aluminium current collectors as opposed to 1 M NaPF electrolyte.
Moreover, the 2 M and 3 M electrolyte solutions resulted in lower
impedance of the SEI on sodium compared to the impedance of
the SEI in 1 M electrolyte which increased with time. This trend
indicates a change in the break down and repair mechanism of
the SEI and possibly lower SEI solubility with the increase in
1
M electrolyte (Figures 4, S3.6), which could be the result of
different SEI compositions, structures, or dissolution rate.
The electrochemical stability window (ESW) of our
6
NaPF concertation. However, since the bulk conductivity of the 2
synthesized NaPF
voltammetry (CV). Mixtures of 1–3 M NaPF
were tested in three-electrode cells using glassy carbon, stainless
steel, and aluminium as working electrodes (WE); sodium metal
was used as the counter and quasi-reference electrode (Figures
6
electrolyte was determined using cyclic
in EC:DEC (1:1 v/v)
M and 3 M electrolytes is lower compared to 1 M electrolytes, the
choice of electrolyte concentration presents a potential trade-off
between SEI and current collector stability, and cell power
capability (further discussed in SI, S7).
6
Lastly, in order to showcase the high-grade nature of our
5
, S3.1–S3.3). The current density of both oxidation (>4 V) and
reduction (<1 V) waves decrease with the cycle number (Figure
S3.1). The electrochemical stability of NaPF vs. aluminium was
synthesized NaPF
were produced independently by Faradion Limited, UK. 1 M
electrolyte solutions of both our synthesized NaPF and
commercial Faradion baseline NaPF were prepared in an
6
, commercial 2- and 3-electrode pouch cells
6
6
investigated (Figure 5) and it was found that the measured current
is three times lower for 2 M and 3 M electrolytes, in turn showing
that the electrolyte electrochemical stability towards the current
collectors is noticeably affected by the concentration of NaPF
.
6
ethylene carbonate, diethyl carbonate and propylene carbonate
mixture (with no additives); the anode active material was a
commercially available hard carbon and the cathode active
material was sourced from Haldor-Topsoe A/S, synthesized
according to Faradion’s specification for a mixed phase O3/P2
layered oxide.[36,37] For the 3-electrode cell testing the reference
electrode comprised an appropriately positioned tab of Na metal
on aluminium. The results of these 3- and 2-electrode sodium-ion
pouch cells showed comparable gravimetric capacity, rate
capability and cycle life between our synthesized and Faradion
6
6 6
baseline NaPF electrolytes. In turn this shows that our NaPF salt
is compatible and stable electrochemically with both anode and
cathode (Figure 6 and SI). Lastly, these results confirm the quality
is as good as the commercial baseline electrolyte and the NaPF
produced is suitable for battery use.
6
6
Figure 5. Cyclic voltammetry of 1–3 M NaPF (EC:DEC 1:1 v/v) in three-
electrode cell (WE - aluminium, CE- sodium metal, RE- sodium metal). 3rd cycle,
measured at 1 mV/sec between 0.01 to 4.2 V.
The onset oxidation potential on glassy carbon, 304
stainless steel and aluminium for 1–3 M electrolyte solutions was
observed at 3.7 V vs. sodium metal (Figures 5 and S3.1–3.3).
However, EC oxidation onset potential above 5 V vs. sodium was
recorded in the literature.[29–31] Traces of water in the electrolyte
could be a possible cause for lower oxidation potentials.[32–34]
While in the case of lithium, lithium hydroxide, which is generated
by water reduction on the anode, tends to precipitate and
accumulate in the SEI, sodium hydroxide remains in solution due
to its high solubility (similarly to other sodium-SEI components).[35]
This higher solubility results in less electrode passivation in
sodium electrolytes and thus a larger water reduction current at
lower potentials. Additionally, a minor reduction peak was
Figure 6. Cell voltage (V) and Anode and Cathode (V vs. Na) vs. cumulative
active material specific capacity. Collected from 2nd cycle at an approximate
constant current rate of C/7 for charge and discharge using cell voltage limits of
1.0 and 4.2 V. The cell voltage trace is in blue, the cathode voltage profile versus
a Na metal reference electrode in red and the anode voltage profile in black.
6
Electrolyte is our synthesized NaPF (described above).
In conclusion this work demonstrates a simple, low cost and
st
observed at 1 V vs. sodium on the 1 cycle (Figures 5, S3.1, S3.3)
scalable synthetic route for producing high-grade NaPF that is as
6
and disappeared on the 3rd cycle (Figure S3.1). The most
probable reaction responsible for this peak is the reduction of
water, resulting in hydroxide formation and hydrogen evolution,
consistent with the above proposals. The cathodic wave that
starts around 0.6 V and peaks at 0.2–0.3 V is attributed to SEI
formation.
pure as the highest-grade battery electrolyte available, and has
identical electrochemical performance at 1 M concentration, as
demonstrated in commercial 2- and 3-electrode pouch cells by
Faradion. Our protocol avoids the use of HF or aqueous
conditions and the absence of insoluble impurities such as NaF
means that a 1 M concentration solution can easily be achieved
and up to 3 M solutions can be prepared accurately in standard
Comparing the different electrolyte concentrations, 2 M and
3
M solutions of NaPF
6
were found to be more stable versus
electrochemical
solvents.
Importantly,
electrochemical
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
measurements reveal potential benefits of using higher
concentration electrolyte solutions on SEI characteristics, which
may be of value in the optimization of battery performance and
lifetime.
[29] L. T. M. Le, T. D. Vo, K. H. P. Ngo, S. Okada, F. Alloin, A. Garg, P. M. L.
Le, J. Mol. Liq., 2018, 271, 769–777.
[
[
[
[
[
[
[
[
30] B. L. D. Rinkel, D. S. Hall, I. Temprano, C. P. Grey, J. Am. Chem. Soc.,
2020, 142, 15058–15074.
31] M. L. P. Le, N. A. Tran, H. P. K. Ngo, T. G. Nguyen, V. M. Tran, J. Solution
Chem., 2015, 44, 2332–2343.
32] N. Dubouis, A. Serva, E. Salager, M. Deschamps, M. Salanne, A.
Grimaud, J. Phys. Chem. Lett., 2018, 9, 6683–6688.
Acknowledgements
33] C. Cometto, G. Yan, S. Mariyappan, J.-M. Tarascon, J. Electrochem.
Soc., 2019, 166, A3723–A3730.
The authors would like to acknowledge the Faraday Institution
FIRG018 Next Generation Na-ion Batteries for funding.
34] M. Metzger, B. Strehle, S. Solchenbach, H. A. Gasteiger, J. Electrochem.
Soc., 2016, 163, A1219–A1225.
35] D. Aurbach, M. Daroux, P. Faguy, E. Yeager, J. Electroanal. Chem.,
1991, 297, 225–244.
Keywords: Battery • Electrolyte • NMR • Sodium-ion batteries •
36] J. Barker, R. Heap, WO Application, 2019197812, 2020 (priority date
SEI
April 9, 2018).
37] R. Sayers, J. Barker, R. Heap, US Patent 10550007, 2020 (priority date
[
[
[
[
[
[
[
[
1]
2]
3]
4]
5]
6]
7]
8]
9]
N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Chem. Rev., 2014, 114,
1636–11682.
X. Pu, H. Wang, D. Zhao, H. Yang, X. Ai, S. Cao, Z. Chen, Y. Cao, Small,
019, 15, 1805427.
P. K. Nayak, L. Yang, W. Brehm, P. Adelhelm, Angew. Chem. Int. Ed.,
018, 57, 102–120.
May 22, 2014).
1
2
2
T. Perveen, M. Siddiq, N. Shahzad, R. Ihsan, A. Ahmad, M. I. Shahzad,
Renew. Sustain. Energy Rev., 2020, 119, 109549.
D. Kundu, E. Talaie, V. Duffort, L. F. Nazar, Angew. Chem. Int. Ed., 2015,
5
4, 3431–3448.
C. Vaalma, D. Buchholz, M. Weil, S. Passerini, Nat. Rev. Mater., 2018,
, 18013.
T. Kim, W. Song, D.-Y. Son, L. K. Ono, Y. Qi, J. Mater. Chem. A, 2019,
, 2942–2964.
3
7
A. Ponrouch, D. Monti, A. Boschin, B. Steen, P. Johansson, M. R.
Palacín, J. Mater. Chem. A, 2015, 3, 22–42.
[
[
Z. Lin, Q. Xia, W. Wang, W. Li, S. Chou, InfoMat, 2019, 1, 376–389.
10] G. G. Eshetu, G. A. Elia, M. Armand, M. Forsyth, S. Komaba, T. Rojo, S.
Passerini, Adv. Energy Mater., 2020, 10, 2000093.
[11] C. Bommier, X. Ji, Small, 2018, 14, 1703576.
[12] H. Che, S. Chen, Y. Xie, H. Wang, K. Amine, X.-Z. Liao, Z.-F. Ma, Energy
Environ. Sci., 2017, 10, 1075–11011.
[
[
[
13] A. Ponrouch, E. Marchante, M. Courty, J.-M. Tarascon, M. R. Palacín,
Energy Environ. Sci., 2012, 5, 8572–8583.
14] L. A. Ma, A. J. Naylor, L. Nyholm, R. Younesi, Angew. Chem. Int. Ed.,
2021, 60, 4855–4863.
15] P. Barnes, K. Smith, R. Parrish, C. Jones, P. Skinner, E. Storch, Q. White,
C. Deng, D. Karsann, M. L. Lau, J. J. Dumais, E. J. Dufek, H. Xiong, J.
Power Sources, 2020, 447, 227363.
[
[
[
16] L. Terborg, S. Nowak, S. Passerini, M. Winter, U. Karst, P. R. Haddad,
P. N. Nesterenko, Anal. Chim. Acta, 2012, 714, 121–126.
17] A. Bhide, J. Hofmann, A. K. Dürr, J. Janek, P. Adelhelm, Phys. Chem.
Chem. Phys., 2014, 16, 1987–1998.
18] M. M. Woyski, W. J. Shenk, E. R. Pellon, Inorganic Syntheses, 1950, 3,
111–117.
[19] A. F. Clifford, A. G. Morris, J. Inorg. Nucl. Chem., 1957, 5, 71–75.
[20] M. D. S. Lekgoathi, J. P. Le Roux, WO 2015/150862 A1, 2015.
[21] J.-R. Hamon, D. Astruc, J. Am. Chem. Soc., 1983, 105, 5951–5952.
[22] R. Younesi, G. M. Veith, P. Johansson, K. Edström, T. Vegge, Energy
Environ. Sci., 2015, 8, 1905–1922.
[23] Fluorinated Materials for Energy Conversion, eds. T. Nakajima, H. Groult,
Elsevier, 1st edn, 2005.
[24] R. D. W. Kemmitt, D. R. Russell, D. W. A. Sharp, J. Chem. Soc., 1963,
4408–4413.
[25] R. A. Wiesboeck., US Pat., 3654330, 1972.
[26] J.-H. Kim, K. Nagahara, S. Yonezawa, M. Takashima, Chem. Lett., 2004,
33, 884–885.
[
27] E. N. Keyzer, H. F. J. Glass, Z. Liu, P. M. Bayley, S. E. Dutton, C. P.
Grey, D. S. Wright, J. Am. Chem. Soc., 2016, 138, 8682–8685.
28] D. Barr, R. Snaith, D. S. Wright, R. E. Mulvey, K. Wade, J. Am. Chem.
Soc., 1987, 109, 7891–7893.
[
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.202111215
COMMUNICATION
Entry for the Table of Contents
This work addresses the urgent need to produce high grade sodium hexafluorophosphate electrolyte salt for sodium-ion batteries. Our
synthesis involves the addition of ammonium hexafluorophosphate to sodium metal under anhydrous conditions, which prevents the
formation of NaF and other commonly found hydrolysis products seen in commercial samples. The high purity allows for up to 3 M
electrolyte concentrations to be achieved.
Institute and/or researcher Twitter usernames: @darren_ould, @SvetlanaMenkin, @ChemCambridge @GreyGroupCam
1
This article is protected by copyright. All rights reserved.