Low-Temperature Tailoring of Copper-Deficient Cu3- xP - Electric Properties, Phase Transitions, and Performance in Lithium-Ion Batteries

Article abstract of DOI:10.1021/acs.chemmater.8b02950

A convenient approach for a controlled and high-yield synthesis of copper-deficient Cu_3-xP (0.1 < x < 0.7) is reported that makes use of ionic liquids with highly nucleophilic naked halide anions. Halide anions drastically enhance the reactivity of the white phosphorus precursor and kinetically disfavor the formation of phosphorus-rich side products. Cu_3-xP shows a high degree of tolerance for cation vacancies without major structural reorganization, as evidenced by X-ray diffraction and solid-state nuclear magnetic resonance spectroscopy. Measurements of the electric properties reveal that Cu_3-xP is a bad metallic p-type conductor. The resistivity is composition-dependent and displays a distinct anomaly from a phase transition, leading to the discovery and structural characterization of two hitherto unknown low temperature polymorphs. Electrochemical evaluation of copper-deficient Cu_3-xP as anode material for lithium ion batteries reveals a drastic change in the cycling mechanism leading to an increase of the initial capacities by about 70%. This work gives a comprehensive insight into the chemical and structural features of copper-deficient Cu_3-xP and should lead to an improved understanding of its properties, not just for battery applications.

Full text of DOI:10.1021/acs.chemmater.8b02950

Subscriber access provided by Kaohsiung Medical University
Article
3-x
Low-Temperature Tailoring of Copper-Deficient CuP – Electric
Properties, Phase Transitions and Performance in Lithium-Ion Batteries
Alexander Wolff, Thomas Doert, Jens Hunger, Martin Kaiser, Julia Pallmann, Romy Reinhold,
Sivatmeehan Yogendra, Lars Giebeler, Jörg Sichelschmidt, Walter Schnelle, Rachel Whiteside,
H. Q. Nimal Gunaratne, Peter Nockemann, Jan J. Weigand, Eike Brunner, and Michael Ruck
Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.8b02950 • Publication Date (Web): 17 Sep 2018
Downloaded from http://pubs.acs.org on September 18, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 16
Chemistry of Materials
1
2
3
4
5
6
Low-Temperature Tailoring of Copper-Deficient Cu P – Electric
3
−x
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Properties, Phase Transitions and Performance in Lithium-Ion
Batteries
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
†
†
Alexander Wolff, Thomas Doert, Jens Hunger, Martin Kaiser, Julia Pallmann, Romy
‡
†
‡
∥
∥
Reinhold, Sivatmeehan Yogendra, Lars Giebeler, Jörg Sichelschmidt, Walter Schnelle, Rachel
Whiteside, H. Q . Nimal Gunaratne, Peter Nockemann, Jan J. Weigand, Eike Brunner and
†
†
#
#
#
,†,
∥
Michael Ruck*
†
Faculty of Chemistry and Food Chemistry, Technische Universität Dresden, 01062 Dresden, Germany.
Leibniz Institute for Solid State and Material Research Dresden e. V., 01069 Dresden, Germany.
Max Planck Institute for Chemical Physics of Solids, 01187 Dresden, Germany.
‡
∥
#
School of Chemistry and Chemical Engineering, Queen’s University of Belfast, Belfast BT71NN, United Kingdom.
ABSTRACT: A convenient approach for a controlled and high-yield synthesis of copper-deficient Cu3−xP (0.1 < x < 0.7) is
reported that makes use of ionic liquids with highly nucleophilic “naked” halide anions. Halide anions drastically enhance
the reactivity of the white phosphorus precursor and kinetically disfavour the formation of phosphorus-rich side
products. Cu3−xP shows a high degree of tolerance for cation vacancies without mayor structural reorganisation, as
evidenced by X-ray diffraction and solid-state nuclear magnetic resonance spectroscopy. Measurements of the electric
properties reveal that Cu3−xP is a bad metallic p-type conductor. The resistivity is composition-dependent and displays a
distinct anomaly from a phase transition, leading to the discovery and structural characterisation of two hitherto
unknown low temperature polymorphs. Electrochemical evaluation of copper-deficient Cu3−xP as anode material for
lithium ion batteries reveals a drastic change in the cycling mechanism leading to an increase of the initial capacities by
about 70 %. This work gives a comprehensive insight into the chemical and structural features of copper-deficient Cu3−x
and should lead to an improved understanding of its properties, not only for battery applications.
P
INTRODUCTION
the impact of this structural feature is not yet widely
considered in many investigations, although there are
even a few studies using the copper deficiency for
application. It was found, for example, that copper
vacancies are responsible for plasmonic absorption
features of Cu3−xP nanocrystals in the near-infrared, which
Metal phosphides are well-known for their interesting
structural diversity and their large variation of useful
1
–4
physical properties. Among this group of phosphides,
copper(I) phosphide Cu3−xP is one binary representative.
Although its crystal structure has been already reported
3
4,36,37
5
were used for constructing flexible photodetectors.
in the 1970s by Olofsson, scientific interest has only
st
The full potential of this material, however, can only be
explored if Cu3−xP with defined and chosen composition is
accessible. Unfortunately, robustly reproducible synthetic
protocols have not been reported so far.
emerged at the beginning of the 21 century, and
subsequently several applications have been developed.
Cu3−xP can be used, for example, as an oxygen scavenger
6–9
during the production of copper and its alloys , as anode
1
0–18
material in lithium ion batteries , and as catalyst for the
Recently, we reported an ionothermal route to
synthesize single-phase Cu2.95(4)P by direct reaction of
copper metal and red phosphorus (Pred) in ionic liquids
1
9–32
splitting of water . Especially the latter example has
lately attracted a lot of attention.
(
ILs). Although this synthesis is simple and capable of
Despite this emerging interest, full characterisation and
understanding of the structural features of this phosphide
are yet to be completed. In particular, it is still frequently
producing large quantities of high-quality copper-rich
Cu3−xP in a comparatively short time, the drawback of this
method is the formation of X-ray amorphous CuP as a
2
considered to have the idealized “Cu P” composition,
3
35
second phase when Cu3−xP with x ≥ 0.1 is targeted. As
although Olofsson had postulated partially occupied
5
evidenced by nuclear magnetic resonance (NMR)
spectroscopy, a crucial step during this ionothermal
synthesis is the chemical activation of Pred. In particular,
small quantities of P molecules are formed when P is
copper positions in the crystal structure. This evidence
33
was
further
substantiated
experimentally
and
34
35
theoretically , and confirmed by us recently. However,
4
red
ACS Paragon Plus Environment

Chemistry of Materials
Page 2 of 16
heated in trihexyltetradecylphosphonium chloride
expected to yield quantitative chloride-to-acetate
exchange.
3
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
[P66614]Cl). This initial finding suggested that white
phosphorus (Pwhite) might be an even better phosphorus
precursor in the synthesis of Cu3−xP, which we
subsequently evaluated in this work.
Synthesis of Cu3−xP (0.1 < x < 0.7). In a glovebox,
stoichiometric amounts of copper and Pwhite (total mass ≈
00 mg) were carefully weighed directly into a glass flask
2
Here we report our results on the syntheses of copper-
deficient Cu3−xP (x > 0.1) via the reaction of copper and
Pwhite in different ILs. The use of Pwhite in chloride-
containing ILs suppresses the formation of CuP2
impurities thus making single-phase Cu3−xP (0.1 < x < 0.7)
accessible. Furthermore, we shed light on the relationship
containing a magnetic stirring bar. The starting materials
were covered with 1300 mg of [P66614]Cl (7.5 mmol) and
the flask was subsequently sealed with a glass stopper.
The flask was rapidly heated to 200 °C in an oil bath and
stirred vigorously for 24 h. After cooling to room
temperature, the mixture was dispersed in ethanol using
an ultrasonic bath and subsequently centrifuged. The
solvent was then decanted and discarded. The washing
procedure was repeated three times. The resulting fine
greyish powder was dried under ambient conditions. To
avoid surface oxidation, the material was alternatively
washed in a glovebox with dried dichloromethane,
provided by a solvent purification system (M. Braun), and
dried under argon atmosphere. Cu3−xP is obtained in
yields of 99+ %.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
between the crystal structure and
a selection of
composition-dependent physical properties, hence
demonstrating that the control of the Cu3−xP composition
is crucial for possible applications to lithium ion batteries.
EXPERIMENTAL SECTION
Starting Materials. All compounds were handled in an
0
argon-filled glovebox (M. Braun; p(O )/p < 1 ppm,
2
0
p(H O)/p < 1 ppm). The ILs trihexyltetradecylphosphon-
2
ium chloride ([P66614]Cl, IoLiTec, >95 %), trihexyltetra-
decylphosphonium bromide ([P66614]Br, IoLiTec, >95 %),
Trihexyltetradecylphosphonium bis(trifluoromethylsul-
Crystal Growth of Cu3−xP. Single-crystals suitable for
crystal structure determination were obtained by
annealing polycrystalline Cu3−xP at high temperature. A
fonyl)imide ([P66614][NTf ], IoLiTec, >98 %) and trihexyl-
2
sample of Cu3−x
P
(≈ 100 mg) with the nominal
tetradecylphosphonium acetate ([P66614][OAc], own
synthesis) were dried at 110 °C for 12 h under dynamic
vacuum. Copper (ABCR, 99.9 % and 99.999 %, spherical,
composition of Cu P was synthesized and dried under
2.7
argon as described above. In a glovebox, it was transferred
into a silica tube, which had previously been dried under
dynamic vacuum at 1000 °C for 1 h. After sealing the silica
tube under vacuum, the sample was annealed at 900 °C
for 2 weeks in a chamber furnace. Adding a small amount
100 mesh) was treated with H at 400 °C before use.
2
Following a procedure described by Brauer, Pred (ABCR,
9
9.999 %) was washed with sodium hydroxide, rinsed
[39]
with refluxing water, and finally dried under vacuum.
of I notably accelerated the recrystallization process. The
2
Pwhite was obtained by heating the purified Pred in an
evacuated silica tube to 450 °C. The opposite side of the
ampoule protruded from the tubular furnace to allow
condensation of Pwhite. The crude product was collected
under argon and sublimed twice in an evacuated silica
tube at 100 °C until a transparent waxy solid was
obtained. Pwhite is a hazardous form of the element. It is
spontaneously combustible upon exposure to air and
extremely toxic (estimated human lethal dose: 50-100
mg).
resulting solid was gently ground to obtain shiny black
polyhedra of Cu3−xP. Energy dispersive X-ray (EDX)
spectroscopy showed no iodine in the Cu3–xP crystals.
Elemental Analysis. The Cu3−x
P
samples were
dissolved in aqua regia with the assistance of microwave
irradiation and subsequently diluted with water. The
determination of the Cu/P ratio was performed with
inductively coupled plasma optical emission spectroscopy
(ICP-OES) using
a
5100 SVDV (Agilent). The
measurement of every sample was independently
repeated at least three times.
Synthesis of [P66614][OAc]. The IL was synthesized via
anion metathesis. [P66614]Cl (20.8 g, 40 mmol) and K[OAc]
(4.9 g, 50 mmol) were dissolved separately in 20 mL and
Powder X-ray Diffraction. Powder X-ray diffraction
(PXRD) was performed at 296(1) K on an X’Pert Pro MPD
diffractometer (PANanalytical) equipped with a curved
5
0 mL ethanol, respectively. Upon mixing of both
solutions, potassium chloride precipitated immediately as
a fine white solid. The solution was allowed to stir over
night before KCl was filtered off leading to a transparent
yellow solution. Subsequently, the ethanol was removed
with a rotary evaporator. In the course of this procedure,
the excess of K[OAc] precipitates forming a cloudy
dispersion. Due to the low viscosity the solid can easily be
filtered, leading to a transparent light-yellow filtrate.
Finally, the product was dried as stated above. It must be
noted, however, that this anion metathesis is not
Ge(111) monochromator using CuKα
pm). Le Bail analysis was performed with the software
radiation (λ = 154.056
1
4
0,41
package Jana2006
using an internal LaB standard
6
(660b, National Institute of Standards and Technology,
USA).
Scanning Electron Microscopy. The dried as-prepared
samples were dispersed in ethanol, and a few drops were
placed on a polished silicon wafer. After evaporation of
the ethanol, the wafer was fixed on a sample holder with a
graphite pad. Scanning electron microscopy (SEM) was
2
ACS Paragon Plus Environment

Page 3 of 16
Chemistry of Materials
resonance frequency of 202.45 MHz by using a 5 mm
performed by using an SU8020 (Hitachi) instrument with
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a triple detector system for secondary and low-energy
high-resolution CryoProbe Prodigy probe head. Solid-
state magic angle spinning (MAS) NMR experiments were
performed on a Bruker Ascend 800 MHz spectrometer
using a double-resonance 1.3 mm MAS NMR probe
operating at the resonance frequency of 323.9 MHz and
the MAS frequency of 40 kHz. The chemical shifts were
backscattered electrons (U = 3 kV).
e
Energy-Dispersive X-ray Analysis. The compositions
of the samples were semi-quantitatively analyzed with
EDX spectroscopy (U = 20 kV) using an SU8020 (Hitachi)
e
instrument equipped with a Silicon Drift Detector X-
N
referenced externally to δ(H PO ; 85 %) = 0.00 ppm. For
3 4
Max (Oxford Instruments). Samples were embedded into
TM
the determination of the chemical shift of signals in the
an epoxy polymer (EpoThin 2, Buehler) under vacuum.
31
P MAS NMR spectra, the centers of gravity of the signals
The pellets were then wet-ground with a MetaServ 250
were used
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
Buehler, SiC grinding paper) and subsequently polished
with a fleece on the same machine using an alumina
Electron Spin Resonance. For the characterization
with electron spin resonance (ESR) spectroscopy, it was
required to synthesize Cu3−xP without paramagnetic
impurities. The material was therefore prepared with
freshly reduced Cu (99.999 %) and the samples were
washed and stored in a glovebox. The ESR experiments
were carried out with a standard continuous-wave
spectrometer at temperatures between 5 K and 300 K. We
measured the power (P) absorbed by the sample from a
transverse magnetic microwave field at X-band frequency
(~9.4 GHz) as a function of an external, quasi-static
magnetic field (B). A lock-in technique (modulating the
field with 0.5 mT at 100 kHz) was used to improve the
signal-to-noise ratio, which yields the derivative of the
resonance signal dP/dB.
TM
suspension (MasterPrep , Buehler). After coating the
polished pellets with carbon, they were fixed on the
sample holder with a graphite pad. With respect to the
preparation method, the elements C, O and Cl (both of
the polymer and the coating) as well as Al (from the
polishing
suspension)
were
omitted
in
EDX
quantifications.
Crystal Structure Determination. Suitable single-
crystals (shiny black polyhedra) were selected with the
help of a light microscope in a glovebox. The crystals were
attached to 0.2 mm glass capillaries and sealed in larger
ones of 0.5 mm diameter to prevent oxidation during the
measurements. Temperature dependent screening
measurements and complete data sets were recorded on a
SuperNova diffractometer (RigakuOxford Diffraction)
using graphite-monochromated MoKα radiation (λ =
Electrical Resistivity and Hall Effect Measure-
ments. For the determination of the electronic transport
properties, two samples with the compositions Cu2.79(2)P
71.073 pm) with a microfocus source. Lorentz factor,
polarization, and multi-scan absorption corrections were
and Cu2.43(5)P were selected. The powders were checked
for purity by PXRD and SEM, their compositions were
determined by elemental analysis. The fine ground
powders were compacted with spark plasma sintering
(SPS) using an SPS-515 ET setup (Fuji Electronic
Industrial) under a pressure of 100 MPa with the heating
applied. The structure at 300 K (Cu3−xP (hP24)) was solved
4
2,43
by charge-flipping ; atomic coordinates of the low-
temperature phases at 200 K (Cu3−xP (hP96)) and 80 K
(Cu3−xP (tP288)) were developed by group-subgroup
relations, cf. discussion. Structure refinements were
4
0,41
−1
performed with Jana2006.
Graphical representations of
rate of 50 K∙min to the final temperature of 350 °C,
[44]
the structures were drawn with the program Diamond.
which was subsequently held for 10 min until the sample
was cooled to RT. PXRD and SEM investigations proved
the material to be unchanged after the SPS. Rods of about
8 mm length were cut from the pellets. Thin Pt wires (25
μm) were attached to the samples by spot welding.
Crystallographic data of the three polymorphs is
summarized in Table S6, the respective atom parameters
are stated in Tables S7 – S9. Further details about the
crystal structures can be obtained from the
Fachinformationszentrum Karlsruhe, 76344 Eggenstein-
Leopoldshafen (Germany) on quoting the depository
numbers CSD-434583 (Cu3−xP (hP24), 400 K), CSD-434584
Resistivity (ρ) was measured in a conventional four-probe
arrangement while for the Hall resistivity (ρxy) a five-wire
configuration was used. Measurements were performed
(
Cu3−xP (hP24), 300 K), CSD-434585 (Cu3−xP (hP96),
using low-frequency alternating current in
a
200 K) and CSD-434586 (Cu3−xP (tP288)), 80 K).
measurement system PPMS 9 (Quantum Design, ACT
option) with magnetic fields up to B = 9 T and
temperatures between 2 K and 300 K. The transverse
Nuclear Magnetic Resonance Spectroscopy. Liquid
samples
were
transferred
into
tubes
with
magnetoresistance
ρ(B) and the Hall resistivity ρxy(Β)
polytetrafluorethylene valves (Deutero) in a glovebox.
The tubes were evacuated and the valves were closed
prior to the measurement. To avoid any solvation
were measured at fixed temperatures. The ρxy(B) was
found to vary linearly with field up to ± 9 T.
influences, dimethyl sulfoxide-d was provided inside a
6
Heat Capacity Measurements. For the heat capacity
measurements, the SPS sintered sample with the
composition of Cu2.79(2)P was chosen. The sample was
sealed capillary in the NMR tubes as external lock. The
liquid-state NMR experiments were performed using a
Bruker AVANCE III HDX, 500 MHz Ascend spectrometer
crushed
and
a
thin
platy
piece
3
1
at 333 K. The P NMR spectra were recorded at the
(1.00 mm x 0.83 mm x 8.24 mm) was selected with the
3
ACS Paragon Plus Environment

Chemistry of Materials
Page 4 of 16
help of a microscope. Heat capacity measurements were
conducted on 9ꢀT CFMS micro-chip calorimeter
Cryogenic Ltd.) using the a. c. method (f = 5 Hz, U =
to larger agglomerates (Figure S2, Supporting
Information).
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
(
0.8 V) with Apiezon N grease as thermal contact.
Electrochemical Characterization. To minimize the
influence of a possible surface oxidation of Cu3−xP, all
samples for the electrochemical characterization were
washed and handled exclusively under argon atmosphere
in a glovebox. For the preparation of the electrodes, 80
wt.% of Cu3−xP and 10 wt.% Super P (Timcal) were mixed
in a stainless-steel shaker for 20 min at 25 Hz. The binder
solution was prepared by dissolving 10 wt.% PVDF 1013
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
0
Solvay) in N-methylpyrrolidone (NMP). Subsequently,
.3 ml of the binder solution and 100 mg powder were
mixed and sonicated for 30 min. After doctor blading
onto a copper foil in a glovebox, the samples were pre-
dried overnight at 80 °C. After cutting electrodes of 12
mm in diameter, the electrodes were again dried
overnight at 80 °C under vacuum. The Swagelok cells
were assembled in an argon-filled glovebox using a two-
electrode configuration. Two glass fiber separators
(Whatman) and 250 μL electrolyte were placed between a
lithium metal disc (Chempur, 250 μm) as counter
electrode and the copper phosphide as working electrode.
Figure 1. PXRD patterns of Cu3−
P from the reaction of Cu
x
and Pwhite in [P66614]Cl with nominal compositions.
A mixture of 1 M LiPF in dimethyl carbonate/ethylene
6
carbonate (1:1 v/v) (Selectilyte LP30, BASF) was employed
as electrolyte. Electrochemical tests were conducted at a
constant temperature of 25 °C in a climate chamber using
a
multichannel
VMP3
potentiostat
(BioLogic).
Galvanostatic cycling with potential limitation (GCPL)
+
was performed between 0.7 and 2.0 V vs. Li/Li at a
−
1
current density of 12 mA∙g (C/10 for differential capacity
plots). The current density and the specific capacity were
calculated based on the mass of copper phosphide.
Figure 2. SEM micrographs of Cu3−xP (after washing and
drying).
As discussed, the ionothermal synthesis of Cu3−xP (x ≥
RESULTS AND DISCUSSION
0
.1) from Pred suffers from the formation of X-ray
3
5
Synthesis and material characterization. The
reaction of elemental copper with white phosphorus in
amorphous CuP as a byproduct. As shown in Figure 3, a
2
sample of the nominal composition Cu2.5P synthesized
with Pred exhibits a large amount of the amorphous CuP
as impurity, which appears dark relative to the Cu3−x
[
(
(
P66614]Cl yielded a greyish powder of single-phase Cu3−x
0.1 ≤ x ≤ 0.7) according to PXRD (Figure 1) and SEM
Figure S1, Supporting Information). Elemental analysis
P
2
P
main phase in SEM images captured with a back-
scattered electron detector (BSE) and shows a notable
confirmed that the final composition of the material is in
accordance with the targeted nominal composition of the
respective synthesis (Table S1, Supporting Information). It
was observed that the pristine material is slightly more
coarse-grained in comparison to the Cu3−xP obtained from
difference in composition as evidenced by EDX
35
spectroscopy. In contrast, CuP
is not detectable if the
2
synthesis is conducted with Pwhite leaving all other
reaction parameters unchanged. This finding is supported
35
31
the syntheses with Pred , but can be easily ground to a
by the results of the solid-state P NMR experiments and
fine powder. SEM imaging of the samples reveal
agglomerated particles with an average size of
approximately 50 μm (Figure 2, left). These particles
consist of tiny intergrown platelets (Figure 2, right),
which rationalizes the observation that the sample can be
crushed easily. It was also found that the tendency of
agglomeration increases with the ratio of starting material
to IL; hence increasing the total mass of the starting
materials whilst keeping the amount of IL constant lead
the Le Bail fittings of the X-ray diffraction data. As
reported recently, Cu3−x
P
exhibits
a
composition
dependent Knight shift, which is proportional to the
change of the unit cell dimensions. Both can easily be
3
5,45
used to track the change of the Cu3−xP composition.
Comparing Cu3−xP samples from Pred and Pwhite precursors
with the same nominal composition reveals a significant
downfield shift of the solid-state NMR signal and a
smaller unit cell volume for all samples synthesized with
Pwhite (Figure 4). This indicates that Cu3−xP samples
4
ACS Paragon Plus Environment

Page 5 of 16
Chemistry of Materials
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. SEM (BSE) micrographs of Cu3−xP synthesized from a molar ratio Cu:P = 2.5:1 with Pred (left) and Pwhite (right). Bright
areas: Cu3−xP; dark areas: CuP ; black areas: organic polymer.
2
3
1
Figure 4. P MAS NMR spectra (left) and unit cell volumes fitted from the PXRD data (right) of Cu3−xP with different nominal
compositions. Black symbols: samples synthesized with Pred. Blue symbols: samples synthesized with Pwhite. For detailed data see
Table S2, Supporting Information.
obtained from reactions with Pwhite exhibit a higher Cu:P
ratio than their counterparts synthesized from Pred
because no phosphorus is consumed by the formation of
S3, Supporting Information). The copper-poor limit of
Cu3−xP seems to be reached at x = 0.7 with the synthesis
method presented in this work. This is, on the one hand,
indicated by the P NMR shift, which does not change
significantly beyond this composition. Additionally,
fluctuations in the NMR shifts and in the lattice constants
were found if Cu3−xP (x > 0.7) was targeted leading to
poorly reproducible results, however, the exact reason for
this behavior requires further studies. Attempts to
,
3
1
the undesired phosphorus-rich CuP impurity. The shapes
2
3
1
of the P solid-state NMR signals of the Cu3−xP samples
differ significantly depending on which allotrope of
phosphorus was used as starting material. Cu3−x
P
synthesized from Pred exhibits narrow but asymmetric
signals whilst samples obtained from Pwhite show broad
isotropic signals. This broadening can most likely be
attributed to the presence of Cu3−xP phases with slightly
different compositions in one batch, pointing towards a
high compositional homogeneity if the synthesis is
conducted with Pred and rather low homogeneity if Pwhite is
used. This is in agreement with the results of the EDX
analysis. While the Cu3−xP from Pred shows a very
synthesize Cu P using P lead to the formation of CuP2
white
2.1
impurities in the product (Figure S4 and Table S4,
Supporting Information) as it was previously observed for
the synthesis with Pred
.
The impact of different phosphorus allotropes on the
phase formation. Our results demonstrate that the co-
formation of CuP inevitably leads to copper-richer (i.e.
2
phosphorus-poorer) Cu3−xP phases. Therefore, the key
challenge for the synthesis of copper-deficient Cu3−xP is to
35
homogenous composition across the sample, we found
the composition of different Cu3−xP crystallites varying
slightly if Pwhite is used as indicated by a higher standard
deviation of the average composition (Figure S3 and Table
suppress the CuP formation. This applies not only to IL-
2
5
ACS Paragon Plus Environment

Chemistry of Materials
Page 6 of 16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 5. Mobilization of the discussed phosphorus allotropes in [P66614]Cl at elevated temperatures as evidenced by NMR
spectroscopy (s = solid; l = liquid).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Scheme of the proposed mechanism of the Cu3−xP and CuP formation in [P66614]Cl.
2
5
4
based syntheses at low temperatures, as similar phase
relations have also been observed in high-temperature
larger fragments, e.g. phosphorus nanoparticles , which
are currently not detectable with NMR spectroscopy
undergone in this work.
5
,33,46
solid-state approaches.
Our results suggest that
under ionothermal conditions CuP2 is effectively
suppressed if Pred is substituted by Pwhite, which should be
attributed to their different allotropic structures and,
consequently, their chemical properties. White
These results demonstrate the impact of the reactivity
of the different precursors on the phase formation of
Cu3−xP, which is mainly attributed to the rate mobile P4
molecules are provided. Furthermore, we assume that
these different reaction behaviors are also related to the
phosphorus, which consists of P molecules, is highly
4
reactive (e.g. spontaneous ignition in air) and soluble in
47–50
crystal structures of Cu3−xP and CuP . The crystal
2
many organic solvents.
White phosphorus, which
3−
structure of Cu3−xP hosts isolated P atoms (or P ions if
consists of P4 molecules, is highly reactive (e.g.
5
we consider the ionic limit) without P–P bonds, and the
spontaneous ignition in air) and soluble in many organic
47–50
solvents.
formation of the phase requires a consecutive splitting of
In contrast to that, the polymeric structure
the phosphorus precursors into “P ” moieties at the
1
of Pred is far less reactive, stable in air, and almost
51
reaction sites. This requires less energy when starting
insoluble in a majority of common solvents. However, it
from P molecules than from the polymeric phosphorus
4
must be considered that our ionothermal reactions are
conducted at approximately 200 °C in an ionic liquid
medium. It is therefore crucial to discuss the ways in
which the phosphorus precursors are mobilized and
changed under the impact of temperature and the IL.
network of Pred, and we consequently assume that Cu3−x
P
is preferably formed by P molecules (Figure 6, path a).
4
On the other hand, the comparatively slow
decomposition of Pred creates a “bottle neck” effect in the
reaction leading to considerable amounts of unreacted
Pred amongst the emerging Cu3−xP particles. We assume
that the presence of unreacted Pred is causing the
At the reaction temperature P
is already molten
white
5
1
(
T (Pwhite) = 44.1 °C) and the P molecules are partially
m
4
3
8,52
dissolved in the IL as evidenced by NMR spectroscopy.
formation of CuP , as an extended covalent phosphorus
2
This means, the P4 molecules are homogenously
distributed within the reaction medium and diffusion is
not rate-determining, especially if of the reaction mixture
54,55
network is found in the crystal structure of this phase,
and it seems reasonable that CuP is preferably formed if
2
existing P–P bonds are available (Figure 6, path b).
Moreover, it is known from the synthesis of
is continuously stirred. Extensive heating of P , on the
4
other hand, initiates its polymerization to Pred (Figure S5
and Figure S6, Supporting Information), hence decreasing
the reactivity of the precursor. Vice versa, it was found
polyphosphides that Cu3−xP can be further phosphidized
56
by Pred
, which makes a reaction of Cu3−xP with unreacted
Pred also probable (Figure 6, path c). The formation of
CuP at high P concentrations, on the other hand, is
that Pred dissolves slowly as P if heated inside the reaction
4
2
4
medium, which consequently increases the reactivity of
3
8,52
reasonably
explained
by
the
aforementioned
the precursor (Figure 5).
It should be noted, that we
polymerization of P , which leads to the extended
4
only have experimental evidence for P4 molecules as
mobile species, but we cannot exclude the presence of
formation of P–P bonds. However, the rate of the CuP2
6
ACS Paragon Plus Environment

Page 7 of 16
Chemistry of Materials
Evidently, chloride ions facilitate the reaction and act as
formation in all these cases is limited due to slower solid-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
solid diffusions involved in these reactions, which may
explain why Cu3−xP nevertheless appears as the main
phase.
a mineralizer in the same way as has been reported for
halide ions in the solid-state synthesis of the
5
6
aforementioned polyphosphides.
Apparently, this
mineralizer concept can be transferred to the ionothermal
syntheses as well. Similar findings have been already
reported by the group of Dehnen, who demonstrated the
positive impact of chloride on the crystallization of
The impact of the ionic liquid anion on the phase
formation and the activation of phosphorus. The
question remains what excatly the role of the IL is during
Cu3−xP formation and how it affects the proposed
mechanism. Our previous results indicated that the
cation does not have a notable influence on the phase
formation. In contrast to that, the anions appear to have
57
chalcogenidometalate superspheres.
Besides this mineralizing effect, high concentrations of
chloride ions seem to prevent the formation of P–P bonds
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
35
considerable impact. This can be clearly demonstrated if
and, thus, inhibit the formation of the CuP impurities.
2
the Cu3−xP synthesis is attempted in [P66614][NTf ] (Figure
The latter effect is closely related to what we have
discussed in the previous section. The phosphorus
precursor has to be split at the reaction site, which
obviously requires a certain amount of chloride as
catalyst. If the P–P bonds are not sufficiently activated,
the reaction follows the pathways for polyphosphides,
ultimately leading to the formation of CuP2.
2
7). The reaction of copper and Pred in this IL is almost
completely inhibited, evidenced by the detection of only
elemental copper in the PXRD of the solid reaction
product. Only the sample surface turns slightly dark
during the reaction, most likely due to the formation of
an X-ray amorphous phosphide coating. The reaction of
copper with Pwhite under the same conditions, on the
other hand, leads to the formation of Cu3−xP, although
elemental copper is still present in the product.
Figure 7. PXRD patterns of the products of reacting Cu and
Pred (top, molar ratio Cu:P = 2.95:1) as well as Cu and Pwhite
(
bottom, molar ratio Cu:P = 2.7:1) in [P66614][NTf2].
o further evaluate the differences of these two anions, a
series of Cu3−x model reactions with Pwhite were
Figure 8. PXRD patterns (left) and SEM-BSE micrographs
(
right) of the products of reacting Cu and Pwhite (molar ratio
T
Cu:P = 2.7:1) in different [P66614]Cl/[P66614][NTf ] mixtures. The
2
P
molar ratios of the anions in these mixtures are given in the
figure.
conducted in [P66614]Cl/[P66614][NTf2] mixtures (Figure 8,
left). The results of the PXRD measurements prove that
already 5 mol% of chloride in the reaction mixture lead to
a drastic improvement of the phase formation, in
particular, no elemental copper can be detected in the
products anymore. Remarkably, SEM investigations of the
reaction products (Figure 8, right) reveal the formation of
X-ray amorphous CuP2, indicated by dark spots in the
back-scattered electron images, which only vanish if 60
mol% or more chloride-anion IL is used in the reaction.
Conducting
the
same
model
reaction
in
[
P66614]Br/[P66614][NTf ] mixtures reveals the mineralizing
2
effect also for bromide anions, although CuP was found
2
in the product already after reducing the bromide content
in the reaction mixture to 80 mol% (Figure S7, Supporting
Information). Hence, bromide seems less effective in
activating P–P bonds.
7
ACS Paragon Plus Environment

Chemistry of Materials
Page 8 of 16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
1
Figure 9. Sections of the P NMR spectra of different ILs after heating with Pred (left) and SEM micrographs of Pred particles
after heat treatment in different ILs (right). The red arrow marks the NMR resonance of P4 in both ILs. For full spectra, see
Figure S8 and S9, Supporting Information.
−
The different reactivities of chloride, bromide and [NTf2]
may be explained by their different nucleophilicity or
Lewis basicity. P , which is known for its electrophilic
character, can readily be attacked by nucleophiles. This
reactivity is, however, not limited to P but is also found
for Pred, where nucleophilic attacks on the phosphorus
network lead to a cascade of disproportionation reactions
and rearrangements. For example, Pred is degraded by
be strongly etched in [P66614]Cl, as indicated by the rough
surface of the red phosphorus particles (Figure 9, right).
In contrast, the surface of Pred remains unaffected after
4
4
9
being heated in [P66614][NTf ], explaining the distinctively
2
hindered Cu3−xP formation in this IL. Notably, no
differences were observed with NMR spectroscopy when
Pred was heated in [P66614]Br or [P66614]Cl. However, the
formation of CuP2 in the product after reducing the
amount of [P66614]Br in the reaction mixture demonstrates
the lower nucleophilicity of bromide compared to
chloride.
4
3
−
2−
−
K[OEt] and the polyphosphide anions P , P16 , and P5
21
5
8
are obtained. The nucleophilicity of dissolved anions, on
the other hand, depends on the nature of the solvent. In
protic solvents, the nucleophilicity of halide anions
decreases in the series I > Br > Cl > F due to the strong
solvation of the smaller halide ions. In aprotic solvents,
however, this trend is inverted and fluoride becomes the
strongest nucleophile. This trend for aprotic solvents is
in general also found for ILs, as demonstrated mainly for
alkyl-substituted imidazolium compounds
P66614] cation was found to interact even less with anions
than any imidazolium cation, owing to its increased steric
hindrance and decreased potential to form hydrogen
bonds. Regarding the different natures of the anions,
chloride can be considered to be one of the strongest and
NTf2] one of the weakest nucleophiles within the
−
−
−
−
These results suggest that the reactivity might be
influenced or even tuned by other anions as well.
However, the application of stronger nucleophiles in ILs
is limited by the stability of the organic cation, especially
59
63
at higher temperature. Nevertheless, we found that Pred
60–62
can be degenerated by [P66614][OAc] yielding a dark red-
The
2−
+
colored solution, which is caused by traces of the [P16]
[
polyanions as evidenced by NMR spectroscopy (Figure
S10, Supporting Information).
6
1
Electrical transport properties. Cu3−xP with small x
was found to be
temperature already several decades ago.
confirmed the metallic nature of this compound down to
a metallic conductor at room
−
9,64
[
We recently
6
1,62
commercially available ILs.
3
5
4 K. However, due to a lack of high-quality data, we
decided to reinvestigate the electronic transport
The influence of different anions on the decomposition
of red phosphorus has been tested, as the aforementioned
decomposition yields P4 which can qualitatively be
followed by NMR spectroscopy. Indeed, we found that
the degradation of Pred upon heating in the IL under
vacuum for several days is far less pronounced in
P66614][NTf ] compared to [P66614]Cl, evidenced by a
smaller P signal in the P NMR spectrum (Figure 9, left).
Moreover, SEM micrographs reveal the surface of Pred to
properties and to extend our characterization to Cu3−x
P
38
with lower copper content. The temperature and
magnetic field dependence of the electrical resistivity
ρ(B,T) and of the Hall resistivity ρxy(B,T) were determined
for compacted polycrystalline samples of Cu2.79(2)P and
Cu2.43(5)P.
[
2
31
4
8
ACS Paragon Plus Environment

Page 9 of 16
Chemistry of Materials
Both samples were found to exhibit metallic p-type cannot always be detected at the X-band frequencies due
6
5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
conductivity with electrical resistivity values at room
temperature of 100 and 140 μꢀ∙cm, respectively. Upon
cooling (Figure 10), an anomalous decrease of the
resistivity is observed to below 190 K (Cu2.79(2)P) or 180 K
(Cu2.43(5)P) indicating a phase transition. The electric
resistivity is lower below the phase transition (more
pronounced for the copper-richer phase), suggesting a
decrease of the scattering of the charge carriers. Upon
heating, the anomalies in both samples are observed at
slightly higher temperatures (Figure 10, inset left), hinting
at a first order structural phase transition. The transition
to spin-phonon interactions.
Temperature-dependent structure investigations.
Structural phase transitions of Cu3−xP below ambient
temperature, as indicated by the resistivity and Hall
measurements, have not yet been reported in literature.
Thus, single crystals were grown to investigate the low
temperature polymorphs.
Temperature-dependent diffraction data indicated the
presence of two low-temperature phases (Figure S16,
Supporting Information). For that, diffraction images
were taken in steps of 10 K and complete data sets were
recorded at 400, 300, 200, 100, and 80 K. The
reinvestigation of the crystal structure at 400 K and 300 K
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
is accompanied by an anomaly of the heat capacity C (T)
at this temperature (for Cu2.79(2)P, see Figure S13,
Supporting Information).
p
confirms the reported Na As-type structure in the
3
hexagonal space group P6 cm (no. 185) with lattice
3
parameters a = 693.5(1) pm and c = 713.0(1) pm (at 295 K)
5
and six formula unit per unit cell. According to its
Pearson symbol we call this polymorph Cu3−xP (hP24) in
the following.
Cu3−xP forms a three-dimensional network structure
with Cu–P distances between 232 pm and 249 pm and
relatively short Cu–Cu distances of 255 pm to 273 pm.
Both, Cu–Cu distances and Cu coordination numbers
(Table S5, Supporting Information) bear a resemblance to
66
typical intermetallic phases, like AuCu3 . The phosphorus
atom exhibits an unusual high coordination number of 11.
For simplicity, the structure can formally be divided in
two different layers that alternate along the [001]
direction. Slightly puckered honeycomb nets are formed
by Cu1, Cu2 (Wyckoff site 2a and 4b, respectively) and
phosphorus atoms. The Cu3 and Cu4 atoms (Wyckoff
sites 6c) form a strongly corrugated honeycomb net,
topologically resembling the structure of grey arsenic
Figure 10. Normalized electrical resistivity of two different
Cu3−xP samples. The inset shows the derivative of the
resistivity for both samples while cooling (symbols) and
heating (lines). For more detailed resistivity curves, see
Figure S11 and S12, Supporting Information.
The magnetoresistance is positive and small and
increases with decreasing temperature. At T = 2 K it
3
4
(Figure 11). In agreement with de Trizio et al. , only the
latter Cu positions were found to be partially occupied
(Table S7, Supporting Information).
attains +5.8 % in a field of 9 T. The Hall coefficient R (T)
H
(
Figure S14, Supporting Information) as derived from the
slopes of ρxy(B,T) is very similar for both compositions
–1
–9
3
and varies weakly with temperature between 1∙10 m ∙C
and 2∙10 m ∙C . Within a one-band model this indicates
hole concentrations between 6∙10 cm at T = 2 K and
–9
3
–1
21
–3
21
–3
3∙10 cm at T = 300 K, respectively. An impact of the
structural phase transition on the carrier concentration is
not visible in the Hall coefficient data, which is, however,
also due to the small number of data points.
Cu3–xP samples with different compositions were probed
by ESR spectroscopy to check if an increasing amount of
copper vacancies is mirrored in a partial oxidation of
+
2+
some Cu ions to Cu . In accordance with the metallic
2+
conductivity we found no evidence for Cu in the X-band
ESR spectra in the range of 300 K to 5 K (Figure S15,
Supporting Information). Thus Cu3–xP probably adopts
the variable composition by creating itinerant charge
2+
carriers. In this context we must admit, though, that Cu
9
ACS Paragon Plus Environment

Chemistry of Materials
Page 10 of 16
symmetry reduction of index 4 and a starting model for
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the this superstructure was established using the group-
subgroup formalism starting from the Cu3−xP (hP24)
67–69
structure model (Figure S17, Supporting Information)
.
The resulting low-temperature structure is called
Cu3−xP (hP96) in the following.
Noticeably, the Wyckoff positions of Cu1 and Cu2 of the
Cu3−xP (hP24) structure split into two non-equivalent sites
in the low-temperature Cu3−xP (hP96) structure and the P,
Cu3, and Cu4 positions split into three sites each.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The Cu3−xP (hP96) structure model displays similar large
and elongated displacement parameters for the Cu atoms
so that an anharmonic refinement was performed again.
The main difference between Cu3−xP (hP24) and
Cu3−xP (hP96) structures is, however, that the Cu deficit
accumulates on one copper site in the low-temperature
polymorph Cu3−xP (hP96): the refined occupancy for this
position, Cu3a, is only 25(1) % whereas all other Cu
position are fully occupied within a 3σ uncertainty. In
other words, a rearrangement of vacancies in the Cu
substructure occurs upon cooling (Figure 12).
Figure 11. The Cu3−xP (hP24) crystal structure (300 K data).
The ellipsoids represent 90 % probability.
As visible from Figure 11, Cu3 and Cu4 atoms have large
displacement parameters in the hexagonal ab-plane
whereas the displacements of Cu1 and Cu2 atoms are
conspicuous along [001]. The conventional refinement
with anisotropic displacement parameters taking the
statistical copper vacancies on the Cu3 and Cu4 positions
into account resulted in reasonable R-values but exhibited
large atomic displacement values for all Cu atoms and
two pronounced peaks in the difference Fourier maps
close to the Cu1 and Cu2 positions. The displacement
parameters of all Cu atoms are elongated, those of Cu1
and Cu2 along [001], Cu3 and Cu4 in the ab-plane (Figure
The composition was computed to Cu2.81(1)P for the
Cu3−xP (hP96) structure. The coordination polyhedra and
interatomic distances of Cu3a (occupancy 25%) and Cu3b
(fully occupied), both derived from Cu3 in the
Cu3−xP (hP24) structure and adopting the same site
symmetry ..m (Wyckoff site 6c), are hardly
distinguishable. As main difference, Cu3a exhibits four
short Cu-Cu distances of less than 250 pm, whereas for
Cu3b, all Cu-Cu distances are larger than 260 pm (Figure
S19, Supporting Information).
1
1). The extrapolation of the principal mean square
displacement parameters to T = 0 K indicates mainly
dynamic contributions so that an anharmonic refinement
th
with up to 4 rank tensors were considered. To minimize
correlations, tensor components were set to zero if the
refined parameters were within a standard uncertainty of
3
σ. As a results, the R-values dropped considerably and
the difference Fourier map became featureless (Table S6,
ESI†). The displacement and occupational parameters of
P are unsuspicious. The occupations of the Cu3 and Cu4
positions refined to 88(1) % and 90(1) %, respectively,
leading to a composition Cu2.79(1)P in good agreement to
the initial composition of the Cu3–xP powder used for
growing the single crystals. It should also be noted, that
Figure 12. The As-like Cu substructure in Cu3−xP (hP96),
200 K data. The Cu atoms marked in dark blue exhibit an
occupation of 25(1) %; the others are fully occupied. The
ellipsoids represent 90 % probability. For an image of the
unit cell content, see Figure S18, Supporting Information.
(
twin) refinements in subgroups of P6 cm without c-glide
3
planes (P6 , P3m1, etc.) do not lead to smooth Fourier
3
Upon further cooling below 190 K a second structural
reorganization takes place as evidenced by additional
satellite reflections at slightly incommensurate positions
along c* (Figure S16, right, Supporting Information). To
establish a first structure model of this second low-
temperature polymorph, the satellites were indexed using
a modulation wave vector q = (0, 0, 0.31(1)). Additional
diffuse streaks running between the reflections along the
maps and inconspicuous displacement parameters.
Upon cooling below 260 K additional reflections
indicated a doubling of the hexagonal a and b lattice
parameters (Figure S16, center, Supporting Information).
As the symmetry of the diffraction image and the
reflection conditions remain unchanged, the cell
enlargement can be described by an isomorphic
10
ACS Paragon Plus Environment

Page 11 of 16
Chemistry of Materials
l-direction were ignored. A starting model was created in
occupational modulations. Cu2a is generated from Cu2 of
the Cu3−xP (hP24) structure in the course of the symmetry
reductions, which was the atom with the largest
displacement parameters along [001] in the Cu3−xP (hP24)
phase. This is probably due to a certain amount of static
disorder underlying the dynamic effects, i.e. local
situations where Cu2 is located at slightly different z-
positions with respect to the plane of the phosphorus
atoms.
70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the superspace group P6 cm(00γ)000 (no. 185.1.24.1) by
3
taking the atomic positions of the Cu3−xP (hP96) structure
model as average structure. The final refinements were
performed using the commensurate option with a
threefold c-axis resulting in the 3D space group symmetry
P31m. The resulting structure is thus called Cu3−xP (tP288)
in the following. One harmonic positional modulation
wave for all atom positions and for the displacement
parameters were refined together with one harmonic
wave for the occupation parameters of Cu3a. The electron
density distribution around the atomic domain of Cu3a is
clearly disrupted along the modulation direction,
mirroring the occupational modulation (Figure 13).
Considering the omitted diffuse intensities, the
refinements converged to acceptable R-values for the
main reflections and a featureless difference Fourier map.
The refined composition of Cu2.85(1)P is in reasonably good
agreement with the results of the Cu3−xP (hP96) and
Cu3−xP (hP24) refinements.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As a consequence, the resulting puckered grey As-like
Cu exhibit vacancies in some heights, whereas others are
not affected by the positional modulation (Figure 14). In
layers with vacancies, the response of the surrounding
atoms in terms of positional modulations is much
stronger than in fully occupied layers.
The phase transitions are fully reversible. Heating the
crystal reveals the same diffraction images and structure
models as observed before and during the cooling
procedure.
To conclude the crystal structure results, we were able
to confirm the proposed copper depleted sites in the
Cu3−xP (hP24)
structure
experimentally.
Atomic
displacement parameters of the Cu atoms indicate a
considerable thermal motion. Upon cooling, two
polymorphs were discovered that show substantial
redistribution of the copper vacancies coupled to a subtle
reorganization of the copper partial structure.
Electrochemical characterization. Cu3−xP has been
intensively studied as anode material for lithium ion
th
batteries since the beginning of the 21 century, primarily
because its volumetric capacity exceeds graphite by a
1
1–14
factor of three.
This advantage initiated a series of
detailed mechanistic studies of the Cu3−xP reaction with
1
4,15,40,71–73
lithium.
The following discussion is based on the
3
1
most recent study of the mechanism with in situ P and
Li NMR spectroscopy by Poli et al., which also nicely
links in the results of the previous PXRD studies.
7
15
4
6,71
Figure 13. Fourier maps (Fo) around Cu3a of the Cu3−x
P
(
(
tP288) structure indicating the occupational modulation
left) and Cu2a with a strong displacement (right); contour
Based on this recent investigation, copper-rich Cu3−xP
reacts with lithium to form three different products,
–
3
lines in steps of 10 e∙Å , thick black lines represent the
calculated atomic positions.
depending on the applied electrochemical potential: (a)
33,74
Li Cu P (crystal structure unknown); (b) Li CuP
(
and
0
.2
2.8
75,76
2
c) Li3P . The chemistry behind this substitution can be
understood by comparing the known crystal structures in
this system (Figure 15). Formally, Li CuP is obtained if the
2
+
Cu ions on the Cu3 and Cu4 positions (Wyckoff site 6c)
of Cu3−xP are substituted by Li . These Cu positions, which
+
2–
are located between layers of buckled [CuP] hexagonal
nets (including Cu1 and Cu2 positions), were also
identified by our crystal structure investigations to host
the Cu vacancies of Cu3−xP. Following this argument, the
structurally unknown Li Cu P should consist of an
Figure 14. Impact of the Cu modulation on the As-like Cu
substructure of Cu3−xP (tP288).
0
.2
2.8
intermediate structure, where the Cu atoms between the
+
Additionally, Cu2a exhibits
a
strong positional
nets are only partially exchanged by Li . Upon further
modulation along the c direction, which may also be
interpreted as two Cu positions with alternating
lithiation of Li CuP, the Cu atoms of the hexagonal nets
2
are fully substituted, leading to the formation of Li P.
3
11
ACS Paragon Plus Environment

Chemistry of Materials
Page 12 of 16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 15. The crystal structures of Cu3−xP and its relation to lithiated phases found within the lithium ion batteries. The
chemical bonds outside the hexagonal nets are omitted for clarity. For consistency with literature, the discharging refers to the
lithiation and the charging to the delithiation process.
Figure 16. Differential capacity plot of the first cycle (left) and specific capacity plot (right) of different Cu3−xP electrodes. All
compositions given are nominal compositions. The processes marked in the differential capacity plot correspond to the reaction
equations given in the discussion. For the corresponding voltages profile see Figure S20, Supporting Information.
Lithiation of Li CuP, however, requires a more complex
reactions are found less pronounced or missing for Cu3−x
with lower copper content.
P
2
reorganization of the crystal structure, which was also
found to be disadvantageous in regards to cycling
1
4
Process 1, assigned to the formation of Li0.2Cu2.8P,
corresponds to the signal found at 0.89 V for Cu2.95P.
Following the proposed mechanism, this phase should
not be formed with Cu P and Cu P, as the copper
stability. Subsequently, the cut-off potential in our
experiments was set to 0.7 V, to prevent the formation of
Li P and to enable a reversible cycling between Cu3−xP and
3
2.5
2.3
Li CuP. Thus, the following processes should take place:
2
content of these compounds is below that of the targeted
phase. This is indeed proven by the absence of the
corresponding signal at 0.89 V. Instead, a broad signal is
observed for the copper-deficient compounds, which
might be caused by small amounts of copper-rich phases
present in the investigated samples. Upon further
Cu3−xP + 0.2 Li
Li Cu P + 0.2 Cu (process 1)
0.2 2.8
Li Cu P + 1.8 L
Li CuP + 1.8 Cu (process 2)
2
0
.2
2.8
0
Li CuP + x Cu
2
Li2−xCu P – Cu + x Li (process 2*)
1+x
0
Li2−xCu P – Cu + x Cu
Cu3−xP + x Li (process 1* + 0*)
1+x
lithiation, the Li CuP formation (process 2) is observed
2
We found that electrodes produced from Cu3−xP with
various compositions differ significantly (Figure 16). The
reaction steps proposed in the mechanism stated above
for all materials at 0.81 V.
The delithiation of Li CuP (process 2*) is found at 1.05 V
2
for all materials, as expected from the results of the first
discharge. In the following reaction, further delithiation
can be accurately assigned to material of a Cu2.95
P
composition. The differential capacity plots (Figure 16,
left) obtained for this material matches those reported in
(
process 1* and 0*) is only observed for Cu2.95P at 1.09 V
and 1.26 V, respectively. Apparently, there is not enough
copper available to form copper-rich phases, which agrees
well with the absence of process 1 during the first
discharge.
1
0,11,14,46
literature.
This supports the assumption, that the
stoichiometric Cu:P = 3:1 starting ratio in the syntheses
ofthese material resulted in the formation of copper-rich
Cu3−xP. In contrast to that, some features of the lithiation
12
ACS Paragon Plus Environment

Page 13 of 16
Chemistry of Materials
Our results clearly indicate that copper-rich and
AUTHOR INFORMATION
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
copper-deficient Cu3−xP reacts differently with lithium.
These differences lead to very interesting results in terms
of the specific capacity and the cycle stability. The initial
capacity of Cu P during the first discharge is about 70 %
Corresponding Author
*E-mail: michael.ruck@tu-dresden.de
2.3
Author Contributions
higher than the capacity of Cu2.95P, vice versa the cycling
stability of Cu2.95P is much better (Figure 16, right). The
performance of our cells can, however, not compete with
All authors have given approval to the final version of the
manuscript.
1
4
the excellent results reported by Stan et al. This behavior
Funding Sources
is most likely caused by the coarse-grained material used
This work was supported by the Deutsche Forschungsge-
meinschaft (DFG) within the Priority Program SPP 1708.
J.J.W. thanks the DFG for funding a Rigaku Oxford
Diffraction SuperNova system with a dual source (INST
269/618-1).
1
0
in our work which could probably be optimized, but was
not in the scope of this work. As the higher capacity of
the copper-deficient material is counterintuitive to the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5
mechanism discussed in literature, a different lithiation
pathway of the copper-deficient material needs to be
considered, which should be investigated by more
elaborated experiments in the future. Our results,
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
however, suggest that the optimal composition of Cu3−x
P
for battery applications should not be too close to the
We would like to thank Dr. F. Haarmann for helpful
discussions of the solid-state NMR results. We gratefully
acknowledge Dr. G. Auffermann (MPI CPfS, Dresden) for
elemental analysis, Dr. I. Veremchuk (MPI CPfS, Dresden)
for SPS and V. Charbonneau (IFW, Dresden) for
experimental assistance.
stoichiometric Cu P composition to enable a high specific
3
capacity and reversible cycling.
CONCLUSION
In summary, we have developed a convenient approach
to precisely synthesize different Cu3−xP (0.1 < x < 0.7)
phases in high yields with excellent reproducibility. Our
ABBREVIATIONS
IL, ionic liquid; Pred
,
red phosphorus; Pwhite, white
mechanistic
investigations
reveal
the
different
phosphorus; [P66614], trihexyltetradecylphosphonium; [NTf2],
bis(trifluoromethylsulfonyl)imide; EDX, energy dispersive X-
ray; PXRD; powder X-ray diffraction; SEM, scanning electron
microscopy; BSE, back-scattered electron; ICP, inductively
coupled plasma; OES, optical emission spectrometry; MAS,
magic-angle-spinning; ESR, electron spin resonance; SPS,
spark plasma sintering; NMP, N-methylpyrrolidone.
mineralizing and activating properties of halide anions in
ionic liquids. This is in particular interesting for the
development of ionothermal approaches for other
inorganic materials, as rules for choosing appropriate
ionic liquids are only rarely found in literature. The
physical characterization of the Cu3−x
P
shows
a
composition-dependent change in the
electric
conductivity, which is accompanied by a change of the
Hall coefficient. The low-temperature phase transition
indicated by the non-linear behavior of the electric
properties was followed by single-crystal X-ray diffraction
elucidating two hitherto unknown Cu3−xP polymorphs.
Our analysis of the structural relations of these
polymorphs confirms the Cu vacancy accumulation
REFERENCES
(
1) von Schnering, H.-G.; Hönle, W. Bridging Chasm with
Polyphosphides. Chem. Rev. 1988, 88, 243–273.
2) Pöttgen, R.; Hönle, W.; von Schnering, H.-G. Phosphides:
(
Solid-State Chemistry. In Encyclopedia of Inorganic Chemistry,
2nd ed.; King, R. B., Ed.; John Wiley & Sons, Ltd.: Chichester,
U.K., 2005; Vol. 8, pp 4255–4308.
10
d
3
4
(
3) Bawohl, M.; Nilges, T. Phosphorus Rich
Polyphosphides and Selected Materials. Z. Anorg. Allg. Chem.
015, 641, 304–310.
4) Chen, J.-H.; Whitmire, K. H. A structural survey of binary
Ion
between the Cu/P-layers as suggested in literature. The
application of different Cu3−xP phases as anode material in
lithium ion batteries evidences a composition-dependent
change in the lithiation pathway, which might be used to
further optimize these devices.
2
(
transition metal phosphides and arsenides of the d-block
elements. Coord. Chem. Rev. 2018, 355, 271–327.
(5) Olofsson, O. The Crystal Structure of Cu P. Acta Chem.
3
Scand. 1972, 26, 2777–2787.
SUPPORTING INFORMATION
(6) Solovyov, S. E. Oxygen Scavengers. In Kirk-Othmer
SEM images of Cu3−xP, EDX data of Cu3−xP, data of elemental
analysis of Cu3−xP, results of lattice parameter fitting, solid-
state NMR data of Cu3−xP, photography and solid-state NMR
spectra of polymerized Pwhite, PXRD data of Cu3−xP, liquid-
state NMR spectra of P/IL mixtures, resistivity and Hall data
of Cu2.79(2)P and Cu2.43(5)P, heat capacity data of Cu2.79(2)P, ESR
spectra of Cu3−xP, crystallographic data, voltage profiles of
Cu3−xP. This material is available free of charge via the
Internet at http://pubs.acs.org.
Encyclopedia of Chemical Technology; Seidel, A., Bickford, M.,
Eds.; John Wiley and Sons, Inc.: Hoboken, NJ, 2014; pp 1–31.
(7) Kato, T.; Adachi, S.; Aoyagi, T.; Naito, T.; Yamamoto, H.;
Nojiri, T.; Yoshida, M. A Crystalline Metallic Copper Network
Application Film Produced by High Temperature Atmospheric
Sintering. IEEE J. Photovoltaics 2012, 2, 499–505.
(8) Adachi, S.; Yoshida, Y.; Nojiri, T.; Kato, T.; Watanabe, S.;
Yoshida, M. A reaction mechanism of atmospheric sintering for
13
ACS Paragon Plus Environment

Chemistry of Materials
Page 14 of 16
copper–phosphorus alloy electrode. J. Alloys Compd. 2017, 695,
3353–3359.
heterojunction nanophotocatalyst for hydrogen evolution from
water. Nanoscale 2016, 8, 17516–17523.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(9) Robertson, D. S.; Snowball, G.; Webber, H. The
(25) Read, C. G.; Callejas, J. F.; Holder, C. F.; Schaak, R. E.
General Strategy for the Synthesis of Transition Metal Phosphide
Films for Electrocatalytic Hydrogen and Oxygen Evolution. ACS
Appl. Mater. Interfaces 2016, 8, 12798–12803.
(26) Wei, S.; Qi, K.; Jin, Z.; Cao, J.; Zheng, W.; Chen, H.; Cui,
X. One-Step Synthesis of a Self-Supported Copper Phosphide
Nanobush for Overall Water Splitting. ACS Omega 2016, 1, 1367–
1373,
(27) Hou, C.-C.; Chen, Q.-Q.; Wang, C.-J.; Liang, F.; Lin, Z.;
3
Fu, W. F.; Chen, Y. Self-Supported Cedarlike Semimetallic Cu P
Nanoarrays as a 3D High-Performance Janus Electrode for Both
Oxygen and Hydrogen Evolution under Basic Conditions. ACS
Appl. Mater. Interfaces 2016, 8, 23037–23048.
Preparation and Properties of Single Crystal Copper Phosphide.
J. Mater. Sci. 1979, 15, 256–258.
(10) Bichat, M.-P.; Politova, T.; Pfeiffer, H.; Tancret, F.;
Monconduit, L.; Pascal, J.-L.; Brousse, T.; Favier, F. Cu P as
3
Anode Material for Lithium Ion Batteryꢀ: Powder Morphology
and Electrochemical Performances. J. Power Sources 2004, 136,
80–87.
(11) Pfeiffer, H.; Tancret, F.; Bichat, M.-P.; Monconduit, L.;
Favier, F.; Brousse, T. Air stable copper phosphide (Cu3P): a
possible negative electrode material for lithium batteries.
Electrochem. Commun. 2004, 6, 263–267.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
12) Pfeiffer, H.; Tancret, F.; Brousse, T. Synthesis,
Characterization and Electrochemical Properties of Copper
Phosphide (Cu P) Thick Films Prepared by Solid-State Reaction
(28) Ma, L.; Shen, X.; Zhou, H.; Zhu, J.; Xi, C.; Ji, Z.; Kong, L.
Synthesis of Cu P nanocubes and their excellent electrocatalytic
3
efficiency for the hydrogen evolution reaction in acidic solution.
RSC Adv. 2016, 6, 9672–9677.
(29) Tian, J.; Liu, Q.; Cheng, N.; Asiri, A. M.; Sun, X. Self-
3
at Low Temperature: A Probable Anode for Lithium Ion
Batteries. Electrochim. Acta 2005, 50, 4763–4770.
(13) Chandrasekar, M. S.; Mitra, S. Thin Copper Phosphide
Films as Conversion Anode for Lithium-Ion Battery Applications.
Electrochim. Acta 2013, 92, 47–54.
3
Supported Cu P Nanowire Arrays as an Integrated High-Perfor-
mance Three-Dimensional Cathode for Generating Hydrogen
from Water. Angew. Chem. Int. Ed. 2014, 53, 9577–9581.
(30) Feng, L.; Xue, H. Advances of transition metal phosphide
application in electrochemical energy storage and catalysis.
ChemElectroChem 2017, 4, 20–34.
(31) Callejas, J. F.; Read, C. G.; Roske, C. W.; Lewis, N. S.;
Schaak, R. E. Synthesis, Characterization, and Properties of
Metal Phosphide Catalysts for the Hydrogen-Evolution Reaction.
Chem. Mater. 2016, 28, 6017–6044.
(32) Xiao, P.; Chen, W.; Wang, X. A Review of Phosphide-
Based Materials for Electrocatalytic Hydrogen Evolution Adv.
Energy Mater. 2015, 5, 1500985.
(33) Schlenger, H.; Jacobs, H.; Juza, R. Ternäre Phasen des
Lithiums mit Kupfer und Phosphor. Z. Anorg. Allg. Chem. 1971,
385, 177–201.
(34) De Trizio, L.; Gaspari, R.; Bertoni, G.; Kriegel, I.; Moretti,
L.; Scotognella, F.; Maserati, L.; Zhang, Y.; Messina, G. C.; Prato,
M.; Marras, S.; Cavalli, A.; Manna, L. Cu3−xP Nanocrystals as a
Material Platform for Near-Infrared Plasmonics and Cation
Exchange Reactions. Chem. Mater. 2015, 27, 1120–1128.
(35) Wolff, A.; Pallmann, J.; Boucher, R.; Weiz, A.; Brunner, E.;
Doert, T.; Ruck, M. Resource-Efficient High Yield Ionothermal
Synthesis of Microcrystalline Cu3−xP. Inorg. Chem. 2016, 55,
8844–8851.
(36) Manna, G.; Bose, R.; Pradhan, N. Semiconducting and
Plasmonic Copper Phosphide Platelets. Angew. Chem. Int. Ed.
2013, 52, 6762–6766.
(37) Sun, T.; Wang, Y.; Yu, W.; Wang, Y.; Dai, Z.; Liu, Z.;
Shivananju, B. N.; Zhang, Y.; Fu, K.; Shabbir, B.; Ma, W.; Li, S.;
Bao, Q. Flexible Broadband Graphene Photodetectors Enhanced
by Plasmonic Cu3−xP Colloidal Nanocrystals. Small 2017, 13,
1701881.
(14) Stan, M. C.; Klöpsch, R.; Bhaskar, A.; Li, J.; Passerini, S.;
Winter, M. Cu P Binary Phosphide: Synthesis via a Wet
3
Mechanochemical Method and Electrochemical Behavior as
Negative Electrode Material for Lithium-Ion Batteries. Adv.
Energy Mater. 2013, 3, 231–238.
(
15) Poli, F.; Wong, A.; Kshetrimayum, J. S.; Monconduit, L.;
Letellier, M. In Situ NMR Insights into Electrochemical Reaction
of Cu P Electrodes in Lithium Batteries. Chem. Mater. 2016, 28,
1787–1793.
(16) Liu, S.; Liu, C.; Guo, J.; Yan, W. Microstructure and
Superior Electrochemical Activity of Cu P/Reduced Graphene
3
3
Oxide Composite for an Anode in Lithium-Ion Batteries.
J. Electrochem. Soc. 2017, 164, A2390–A2397.
(17) Chen, Y.-C.; Chen, Z.-B.; Lin, Y.-G.; Hsu, Y.-K. Synthesis
of Copper Phosphide Nanotube Arrays as Electrodes for
Asymmetric Supercapacitors. ACS Sustainable Chem. Eng. 2017,
5, 3863–3870.
3
(18) Liu, S., He, X.; Zhu, J.; Xu, L.; Tong, J. Cu P/RGO
Nanocomposite as a New Anode for Lithium-Ion Batteries. Sci.
Rep. 2016, 6, 35189.
(19) Shi, Y.; Zhang, B. Recent advances in transition metal
phosphide nanomaterials: synthesis and application in hydrogen
evolution reaction. Chem. Soc. Rev. 2016, 45, 1529–1541.
(20) Bai, Y.; Fang, L.; Xu, H.; Gu, X.; Zhang, H.; Wang, Y.
Strengthened Synergistic Effect of Metallic M P (M = Co, Ni,
x y
and Cu) and Carbon Layer via Peapod-Like Architecture for Both
Hydrogen and Oxygen Evolution Reactions. Small 2017, 13,
1603718–1603718.
(
21) Du, H.; Zhang, X.; Tan, Q.; Kong, R.; Qu, F. A Cu P–CoP
3
hybrid nanowire array: a superior electrocatalyst for acidic
hydrogen evolution reactions. Chem. Commun. 2017, 53, 12012–
12015.
(38) Lê Anh, M.; Wolff, A.; Kaiser, M. ; Yogendra, S.; Weigand,
J. J.; Pallmann, J.; Brunner, E. ; Ruck, M.; Doert, T. Mechanistic
exploration of the copper(I) phosphide synthesis in
phosphonium-based and phosphorus-free ionic liquids. Dalton
Trans. 2017, 46, 15004–15011.
(39) Brauer, G. Handbuch der Präparativen Anorganischen
Chemie, 3rd ed.; F. Enke: Stuttgart, 1962.
(40) Petříček, V.; Dušek, M.; Palatinus, L. Crystallographic
Computing System JANA2006: General features. Z. Kristallogr.-
Cryst. Mater. 2014, 229, 345–352.
(22) Han, A.; Zhang, H.; Yuan, R.; Ji, H.; Du, P. Crystalline
Copper Phosphide Nanosheets as an Efficient Janus Catalyst for
Overall Water Splitting. ACS Appl. Mater. Interfaces 2017, 9,
2240–2248.
(23) Lu, C.; Wang, J.; Czioska, S.; Dong, H.; Chen, Z.
Hierarchically Structured Cu-Based Electrocatalysts with
Nanowires Array for Water Splitting. J. Phys. Chem. C 2017, 121,
2
5875–25881.
24) Yue, X.; Yi, S.; Wang, R.; Zhang, Z.; Qiu, S. A novel and
highly efficient earth-abundant Cu with TiO2 “P–N”
(
3
P
14
ACS Paragon Plus Environment

Page 15 of 16
Chemistry of Materials
(
41) Petříček, V.; Dušek, M.; Palatinus, L. Jana2006, The
(60) Schmeisser, M.; van Eldik, R. Elucidation of inorganic
reaction mechanisms in ionic liquids: the important role of
solvent donor and acceptor properties. Dalton Trans. 2014, 43,
15675–15692.
(61) Holzweber, M.; Lungwitz, R.; Doerfler, D.; Spange, S.;
Koel, M.; Hutter, H.; Linert, W. Mutual Lewis Acid–Base
Interactions of Cations and Anions in Ionic Liquids. Chem. Eur. J.
2013, 19, 288–293.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Crystallographic Computing System Version 25/10/2015, Institut
of Physics, Academy of Sciences of the Czech Republik, Prague,
2
015.
42) Oszlányi, G.; Sütő, A. The charge flipping algorithm. Acta
Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 123–134.
43) Oszlányi, G.; Sütő, A. Ab initio structure solution by
(
(
charge flipping. Acta Crystallogr., Sect. A: Found. Crystallogr.
2004, 60, 134–141.
(44) Brandenburg, K. Diamond 3.2k - Crystal and Molecular
Structure Visualization, Crystal Impact GbR, Bonn, 2014.
(62) Schmeisser, M.; Illner, P.; Puchta, R.; Zahl, A.; van Eldik,
R. Gutmann Donor and Acceptor Numbers for Ionic Liquids.
Chem. Eur. J. 2012, 18, 10969–10982.
(
45) Furó, I.; Bakonyi, I.; Tompa, K.; Zsoldos, É.; Heinmaa, I.;
Alla, M.; Lippmaa, E. Nuclear Magnetic Resonance Knight Shift
and Linewidth in Ni P and Cu Pꢀ: A Magic-Angle Spinning Study.
(63) Wang, B.; Qin, L.; Mu, T.; Xue, Z.; Gao, G. Are Ionic
Liquids Chemically Stable? Chem. Rev. 2017, 117, 7113–7131.
(64) Juza, R.; Bär, K. Leiter und Halbleiter unter den
Phosphiden der ersten und zweiten Nebengruppe. Z. Anorg.
Allg. Chem. 1956, 283, 230–245.
(65) Kochelaev, B. I. Spin-Phonon Interaction and the EPR
2 4
Linewidth in La CuO and Related Cuprates J. Supercond. 1999,
12, 53–55.
(66) Flinn, P. A.; McManus, G. M.; Rayne, J. A. Elastic
3
constants of ordered and disordered Cu Au from 4.2 to 300°K
J. Phys. Chem. Solids 1960, 15, 189–195.
(67) Bärnighausen, H. Group-subgroup relationships between
space groups: A useful tool in crystal chemistry. MATCH
Commun. Math. Comput. Chem. 1980, 9, 139–175.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
J. Phys. Condens. Matter 1990, 2, 4217–4225.
(46) Mauvernay, B.; Doublet, M.-L.; Monconduit, L. Redox
mechanism in the binary transition metal phosphide Cu
J. Phys. Chem. Solids 2006, 67, 1252.
3
P.
(
47) Jones, R. O.; Hohl, D. Structure of phosphorus clusters
using simulated annealing–P to P . J. Chem. Phys. 1990, 92,
710–6721.
48) Schoeller, W. W.; Staemmler, V.; Rademacher, P.; Niecke,
2
8
6
(
E. Theoretical studies on inorganic ring systems.
Tetraphosphabicyclobutane, cyclotriphosphane, and white
phosphorus: ring strain and hybridization. Inorg. Chem. 1986,
2
5, 4382–4385.
49) Scheer, M.; Balázs, G.; Seitz, A. P
Group Elements and Compounds. Chem. Rev. 2010, 110, 4236–
256.
50) Holthausen, M. H.; Weigand, J. J. The chemistry of
cationic polyphosphorus cages syntheses, structure and
reactivity. Chem. Soc. Rev. 2014, 43, 6639–6657.
51) Perry, D. L. Handbook of Inorganic Compounds; CRC
Press: Boca Raton, 1995; p. 293.
52) Boros, È.; Earle, M. J; Gîlea, M. A.; Metlen, A.; Mudring,
(68) Wondratschek, H.; Müller, U. (Eds.), International
(
4
Activation by Main
Tables
for
Crystallography;
International
Union
of
Crystallography: Chester, 2011.
4
(69) Kroumova, E.; M. Perez-Mato, J.; Aroyo, M. I.
WYCKSPLIT: a computer program for determination of the
relations of Wyckoff positions for a group-subgroup pair.
J. Appl. Crystallogr. 1998, 31, 646.
(70) Stokes, H. T.; Campbell, B. J.; van Smaalen, S. Generation
of (3+ d)-dimensional superspace groups for describing the
symmetry of modulated crystalline structures. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2011, 67, 45–55.
(71) Mauvernay, B.; Bichat, M.-P.; Favier, F. ; Monconduit, L.
Morcrette, M.; Doublet, M.-L. Progress in the Lithium Insertion
3
Mechanism in Cu P. Ionics 2005, 11, 36–45.
(72) Bichat, M.-P.; Politova, T.; Pascal, J. L.; Favier, F.;
Monconduit, L. Electrochemical Reactivity of Cu3P with
Lithium. J. Electrochem. Soc. 2004, 151, A2074–A2081.
(73) Crosnier, O.; Nazar, L. F. Facile Reversible Displacement
Reaction of Cu3P with Lithium at Low Potential. Electrochem.
Solid-State Lett. 2004, 7, A187–A189.
(74) Crosnier, O.; Mounsey, C.; Herle, P. S.; Taylor, N.; Nazar,
2
L. F. Crystal Structure and Electrochemical Behavior of Li CuP: a
Surprising Reversible Crystalline-Amorphous Transformation
Chem. Mater. 2003, 15, 4890–4892.
(75) Dong, Y.; DiSalvo, F. J. Reinvestigation of trilithium
3
phosphide, Li P. Acta Crystallogr., Sect. E: Struct. Rep. Online
2007, 63, i97–i98.
(76) Nazri, G. Preparation, structure and ionic conductivity of
lithium phosphide. Solid State Ionics 1989, 34, 97–102.
(
–
(
(
A.-V.; Rieger, F.; Robertson, A. J.; Seddon, K. R.; Tomaszowska,
A. A.; Trusov, L.; Vyle, J. S. On the dissolution of non-metallic
solid elements (sulfur, selenium, tellurium and phosphorus) in
ionic liquids. Chem. Commun. 2010, 46, 716–718.
(53) Groh, M. F.; Paasch, S.; Weiz, A.; Ruck, M.; Brunner, E.
Unexpected Reactivity of Red Phosphorus in Ionic Liquids. Eur.
J. Inorg. Chem. 2015, 3991−3994.
(54) Olofsson, O. The Crystal Structures of CuP
2
and AgP
2
with some Phase Analytical Data of the Cu-P and Ag-P Systems.
Acta Chem. Scand. 1965, 19, 229−241.
(55) Möller, M. H.; Jeitschko, W. Darstellung, Eigenschaften
und Kristallstruktur von Cu
CuP und AgP . Z. Anorg. Allg. Chem. 1982, 491, 225−236.
56) Lange, S.; Bawohl, M.; Weihrich, R.; Nilges, T.
P
2 7
und Strukturverfeinerungen von
2
2
(
Mineralization Routes to Polyphosphides: Cu
Angew. Chem. Int. Ed. 2008, 47, 5654–5657.
2
P20 and Cu5InP16.
(57) Santner, S.; Yogendra, S.; Weigand, J. J.; Dehnen, S.
Exploring the Chemical ReactionSpace at the Formation of
Chalcogenidometalate Superspheres in Ionic Liquids. Chem. Eur.
J. 2017, 23, 1999–2004.
(
58) Dragulescu-Andrasi, A.; Miller, L. Z.; Chen, B,; McQuade,
D. T.; Shatruk, M. Facile Conversion of Red Phosphorus into
Soluble Polyphosphide Anions by Reaction with Potassium
Ethoxide. Angew. Chem. Int. Ed. 2016, 55, 3904−3908.
(59) Pietro Oliveri, I.; Di Bella, S. Lewis basicity of relevant
monoanions in a non-protogenic organic solvent using a zinc(II)
Schiff-base complex as a reference Lewis acid. Dalton Trans.
017, 46, 11608–11614.
2
15
ACS Paragon Plus Environment

Chemistry of Materials
Page 16 of 16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Insert Table of Contents artwork here
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
ACS Paragon Plus Environment