High-Pressure Metathesis of the M 1- xPO3+4 xN1-4 x (x ≈ 0.05) and M 0.75PO4 (M = Zr, Hf) Orthophosphates

Article abstract of DOI:10.1021/acs.inorgchem.8b00373

We describe the oxonitridophosphates M_1-xPO_3+4xN_1-4x (x -0.05) and the isotypic oxophosphates M_0.75PO₄ (M = Zr, Hf) obtained by high-pressure metathesis. The structures (ZrSiO₄-type, space group I4₁/amd (no. 141), a = 6.5335(7)-6.6178(12), c = 5.7699(7)-5.8409(9) ?, Z = 4) were refined from single-crystal X-ray diffraction data, and the powder samples were examined with quantitative Rietveld refinement. Infrared spectroscopy did not indicate the presence of X-H (X = O, N) bonds. The optical band gaps, between 3.5 and 4.3 eV, were estimated from UV-vis data using the Kubelka-Munk function under assumption of a direct band gap. Temperature-dependent powder X-ray diffraction showed a phase transformation of the M_0.75PO₄ (M = Zr, Hf) compounds to ambient pressure polymorphs at 780 (Zr) and 900 °C (Hf). The preparation of the nitrogen containing compounds exemplifies the systematic access to the new class of group 4 nitridophosphates granted by high-pressure metathesis. Moreover, we show that high-pressure metathesis can also be used for the preparation of metastable oxophosphates.

Full text of DOI:10.1021/acs.inorgchem.8b00373

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
High-Pressure Metathesis of the M1−xPO3+4xN1−4x (x ≈ 0.05) and
M0.75PO4 (M = Zr, Hf) Orthophosphates
Simon D. Kloß, Andreas Weis, Sophia Wandelt, and Wolfgang Schnick*
Department of Chemistry, University of Munich (LMU), Butenandtstraße 5-13, 81377 Munich, Germany
S
* Supporting Information
ABSTRACT: We describe the oxonitridophosphates
M
M
1−xPO3+4xN1−4x (x ≈ 0.05) and the isotypic oxophosphates
0.75PO4 (M = Zr, Hf) obtained by high-pressure metathesis.
The structures (ZrSiO4-type, space group I41/amd (no. 141), a
= 6.5335(7)−6.6178(12), c = 5.7699(7)−5.8409(9) Å, Z = 4)
were refined from single-crystal X-ray diffraction data, and the
powder samples were examined with quantitative Rietveld
refinement. Infrared spectroscopy did not indicate the
presence of X−H (X = O, N) bonds. The optical band gaps,
between 3.5 and 4.3 eV, were estimated from UV−vis data
using the Kubelka−Munk function under assumption of a
direct band gap. Temperature-dependent powder X-ray
diffraction showed a phase transformation of the M0.75PO4
(M = Zr, Hf) compounds to ambient pressure polymorphs at 780 (Zr) and 900 °C (Hf). The preparation of the nitrogen
containing compounds exemplifies the systematic access to the new class of group 4 nitridophosphates granted by high-pressure
metathesis. Moreover, we show that high-pressure metathesis can also be used for the preparation of metastable oxophosphates.
control over the formed anion network. High-pressure
metathesis may remedy these problems. It starts from reaction
of readily available metal halides with LiPN2 under pressures of
several GPa (eq 1), a concept which is in principle transferable
to almost all metals of the periodic system:1
INTRODUCTION
■
High-pressure metathesis using the multianvil technique was
the first synthesis route to offer fast and efficient access to
oxonitridophosphates with trivalent rare-earth metals and,
moreover, providing the means to systematically direct the
structures’ anion networks.1−4 Recently, high-pressure meta-
thesis was successfully expanded to nitridophosphates with the
group 4 metal Hf. As the first instance of this class of
compounds the oxonitridophosphate Hf9−xP24N52−4xO4x (x ≈
1.84), which consists of two interpenetrating diamond-like
anion frameworks, has been prepared.5 While this compound
has been a first proof of principle, one strong-point of high-
pressure metathesis for nitridophosphates, however, is the
granted control over the formed anion network. For the rare
earths, for example, the flexible anion networks range from
highly condensed frameworks down to noncondensed tetrahe-
dra.1−4 This systematic tuning of the network type has yet to be
proven for group 4 nitridophosphates.
The inherent challenge during synthesis of transition metal
nitridophosphates is a competitive redox reaction by which the
nitride ion is oxidized and transition metal phosphides are
formed.5 Previous synthesis routes to transition metal
nitridophosphates led only to scarce success most likely due
to unsuitable reaction conditions, e.g. lack of high-pressure
methods, or unsuitable starting materials, e.g. lack of metal
azides or reactive nitrides.6−9 The existence of nitridosodalites,
which were prepared at ambient pressures, showed stability of
transition metals embedded in a nitridophosphate network.10,11
No synthesis route, however, was discovered yet that grants
MHalx + xLiPN2 → MP N2x + xLiHal
(1)
x
The starting material LiPN2 constitutes a preformed
tetrahedra network made up of PN4-units, which breaks
down and reforms the new network.12 The high-pressure
conditions suppress the elimination of N2 and preserve the
formed metal nitridophosphate. Moreover, the Li halide
produced during the reaction is at the same time the
thermodynamic driving force and flux aiding the crystallization
process of the nitridophosphate.1 Flexibility and further control
over stoichiometry and structure can be achieved by addition of
oxides or nitrides such as Li2O, Li3N, P3N5, or PON as starting
materials.2−4 The targeted degree of condensation κ, i.e. the
atomic ratio of tetrahedra centers to tetrahedra corners,
influences the anionic network formed and can be controlled
by the stoichiometry of the starting materials. Naturally, the low
end of κ is 1/4, representing noncondensed tetrahedra, while
the high end is determined by electrostatics. For the element
combination P and N the highest κ is 3/5, that being the binary
parent compound P3N5.13
Received: February 9, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a-PON. Amorphous PON (a-PON) was prepared by condensation
of OP(NH2)3 in a constant flow of ammonia.27 A fused silica tube was
prepared by drying at 1000 °C for 8 h under dynamic vacuum <0.1 Pa.
The above obtained 1:3 mixture of OP(NH2)3 and NH4Cl was loaded
on a fused silica boat in an Ar-filled glovebox (O2 and H2O < 1 ppm)
and conveyed into the tube furnace in an Ar counterflow via
transporter. The gas flow was switched to ammonia and the starting
materials fired at 300 °C for 5 h to sublimate the NH4Cl; then the
temperature was increased to 620 °C with a 12 h dwell. All
temperature ramps were set to 5 °C/min. The colorless a-PON was
characterized by PXRD (check for residual NH4Cl) and IR.
cri-PON. Cristobalite phosphorus oxonitride (cri-PON) was
prepared by crystallizing amorphous PON.27,28 A dried fused silica
tube with fused silica boat (tube furnace setup, 1000 °C for 8 h under
dynamic vacuum <0.1 Pa) was loaded with a-PON. The condensation
was carried out for 168 h in dynamic vacuum (<0.1 Pa) at 750 °C with
temperature ramps set to 5 °C/min. cri-PON was obtained as a
colorless powder and characterized by PXRD and IR.
Oxo(nitrido)phosphates. Zr1−xPO3+4xN1−4x (x ≈ 0.05),
Hf1−xPO3+4xN1−4x (x ≈ 0.05), Zr0.75PO4, and Hf0.75PO4 were prepared
by high-pressure metathesis using a hydraulic 1000t press
(Voggenreiter, Mainleus, Germany) and the multianvil technique.
The setup included a modified Walker-type module (Voggenreiter)
with an octahedron-within-cubes load driven by steel wedges. The
here used 18/11 assembly consisted of Co-doped (7%) tungsten
carbide cubes (Hawedia, Marklkofen, Germany) with truncated edges
(11 mm), a Cr2O3-doped (6%) MgO octahedron as pressure medium
(18 mm edge length), and pyrophyllite gaskets (Ceramic Substrates &
Components, Isle of Wight, U.K.). The samples resided inside a h-BN
crucible (Henze, Kempten, Germany), and they were heated through
resistance heating with a graphite sleeve (Schunk Kohlenstofftechnik
GmbH, Zolling, Germany). Additional information regarding the
employed high-pressure technique can be found in the literature.29−33
Starting materials were mixed and ground in a glovebox (O2 and
H2O < 1 ppm). All samples were prepared by high-pressure metathesis
starting from stoichiometric mixtures of ZrCl4 (99.99%, Sigma-
Aldrich), HfCl4 (99.9%, abcr GmbH), P4O10 (99.99% Sigma-Aldrich),
cri-PON, and Li2O. The samples were recovered as colorless products
and washed with H2O to remove the LiCl byproduct. Additional
information regarding synthesis can be found in the Synthesis section.
Spectroscopy. A Helios Nanolab G3 Dualbeam UC (FEI,
Hillsboro, OR, United States) equipped with an X-Max 80 SDD
detector (Oxford Instruments, Abingdon, United Kingdom) was used
for scanning electron microscopy and the recording of EDX spectra.
The samples were placed on an adhesive conducting carbon foil affixed
to a metal carrier and coated with carbon.
In this contribution we show that high-pressure metathesis
may offer a systematic access to oxonitridophosphates
containing a group 4 element by preparation of Zr and Hf
compounds with noncondensed tetrahedra and κ = 1/4, namely
Zr1−xPO3+4xN1−4x and Hf1−xPO3+4xN1−4x (x ≈ 0.05).
As these nitridophosphates have vacancies on the metal
position, the end members of the O/N solid solution series, the
oxophosphates Zr0.75PO4 and Hf0.75PO4, are also fathomable.
These are missing members of group 4 phosphates, which is
curious since this group was intensively investigated because of
their ion exchange and ion conductivity properties.14 The
subsolidus phase diagrams of the pseudobinary system ZrO2−
P2O5 and several pseudoternary ZrO2−P2O5−MOx have
extensively been investigated in the literature.15,16 Hitherto,
only three ternary structure-types, the diphosphate MP2O7 and
the orthophosphate oxide M2[PO4]2O with M = Zr, Hf, as well
as the chain phosphate Zr(PO3)4 were structurally charac-
terized.17−21 ZrP2O7, among other tetravalent metal phos-
phates, was discussed as a H+ ion conductor for fuel cells.22 We
were able to prepare these curious Zr0.75PO4 and Hf0.75PO4
compounds with the high-pressure metathesis route and so
advance the chemistry of the group 4 oxophosphates.
Zr0.75PO4 and Hf0.75PO4 are isotypic to Zr1−xPO3+4xN1−4x
and Hf1−xPO3+4xN1−4x (x ≈ 0.05) and crystallize in a defect
variant of the ZrSiO4 structure type. The structural feature of
noncondensed tetrahedra is reminiscent of the important class
of NASICON materials (Na1−xZr2P3−xSixO12 (0 ≤ x ≤ 3)),
which are known for their ion conduction properties and
nuclear waste immobilization.14,23,24 Incorporation of Na into
oxonitridophosphates or metastable oxophosphates through
high-pressure metathesis could lead to compounds with similar
properties.
In the following, we report the structures and properties of
M1−xPO3+4xN1−4x (x ≈ 0.05), and M0.75PO4 (M = Zr, Hf) by
means of single-crystal and powder X-ray diffraction (SCXRD,
PXRD), scanning electron microscopy (SEM) with energy
dispersive X-ray (EDX) spectroscopy, infrared (IR) spectros-
copy, and UV−vis spectroscopy. We discuss stabilities of the
compounds based on high-temperature XRD (HTXRD) and
possible alternative synthesis routes of the M0.75PO4 (M = Zr,
Hf) to determine reasons for their late discovery.
Infrared spectra were recorded on a Spectrum BX II spectrometer
(PerkinElmer Waltham, MA, United States) with a DuraSampler ATR-
unit in the range of 650−4500 cm−1.
EXPERIMENTAL SECTION
■
Li2O. Li2O was prepared by thermal decarboxylation of Li2CO3
(Sigma-Aldrich 99.99%) under dynamic vacuum conditions (<10−3
mbar).25 For this purpose, a fused silica tube residing in a tube furnace
was loaded with a Li2CO3-filled Ag boat. The tube was connected to
an Ar/vacuum Schlenk line and subsequently heated to 700 °C for 72
h in dynamic vacuum with temperature ramps set to 5 °C/min.
Moisture sensitive Li2O was recovered and then stored in a glovebox.
The purity of the colorless product was confirmed by PXRD to check
for residual Li2CO3.
Phosphoric Triamide. The phosphoric triamide OP(NH2)3 is a
starting material for the synthesis of PON and was prepared by
ammonolysis of phosphorus oxychloride POCl3 in liquid ammonia.26
A three-necked flask equipped with one connection to an inert gas/
vacuum/ammonia line and one connection to an overpressure valve
was dried in vacuum. Approximately 200 mL NH3 (Air Liquide 5.0)
was condensed into the flask with an aceton/CO2 freezing mixture.
Approximately 10 mL of POCl3 was added dropwise through a septum
into the stirred liquid ammonia. Excess ammonia was evaporated
overnight and the product OP(NH2)3 was obtained as a colorless
powder in a 1:3 mixture with NH4Cl. The product was characterized
by PXRD.
UV−vis spectra were recorded in reflection geometry on a V-650
UV−vis spectrophotometer (JASCO, Gross-Umstadt, Germany)
equipped with a photomultiplier tube detector and a single
monochromator with 1200 lines/mm. The detected spectral range
was from 240 to 800 nm, scanned with a deuterium lamp (240−330
nm) and a halogen lamp (330−800 nm) with a 1 nm resolution and a
scan speed of 400 nm/min. The device was controlled with the Spectra
Manager II software. Samples were loaded onto a sample holder
consisting of a fused silica slide and a BaSO4-coated stamp. The
baseline was recorded prior each measurement.
Solid State NMR. 31P NMR spectra of M0.75PO4 (M = Zr, Hf)
were recorded on a DSX Avance spectrometer (Bruker) at a magnetic
field of 11.7 T. The samples were loaded into a 2.5 mm rotor made of
ZrO2. The rotor was mounted on a commercial MAS probe (Bruker).
Spectra were recorded at a rotation frequency of 20 kHz and at room
temperature. The raw data were analyzed with the device-specific
software.
Single-Crystal Diffraction. X-ray diffraction on single-crystals was
carried out on a D8 Venture diffractometer (Bruker, Billerica, MA,
United States) with a fine-focus sealed tube X-ray source. Cell
determination, data reduction, and multiscan absorption correction
B
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(SADABS), as well as calculation of planes through reciprocal space
were carried out using the APEX3 software.34,35 Crystals were affixed
on a MicroMount (MiTeGen, Ithaca, NY, United States). The space
groups were determined based on systematic absences analyzed with
XPREP.36 Structure solution and refinement were carried out with
SHELX-97.37,38 Crystal structures were visualized with VESTA.39
Further details on the crystal structure analysis can be obtained from
the Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leopold-
shafen, Germany (fax: + 49-7247-808-666; e-mail: crysdata@fiz-
karlsruhe.de) on quoting the depository numbers CSD-433854,
-433855, -433856, and -433857. CCDC 1816010−1816013 contain
the supplementary crystallographic data for this paper. The data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/structures.
Powder Diffraction. PXRD on microcrystalline samples was
carried out on a Stadi P diffractometer (Stoe & Cie GmbH,
Darmstadt, Germany) in parafocusing Debye−Scherrer geometry
with Mo Kα1 radiation. The instrument was equipped with a Ge single-
crystal monochromator, singling out the Kα1 emission line, and a
MYTHEN 1K silicon strip detector (Dectris, Baden, Switzerland).
Samples were loaded into glass capillaries (Hilgenberg GmbH,
Malsfeld, Germany) with diameters ranging from 0.1 to 0.3 mm,
depending on absorption, and a wall thickness of 0.01 mm.
Diffractograms were recorded in the angular range of 2θ = 2−76°.
Rietveld refinement was carried out with the program TOPAS-
Academic V4.1.40,41 Peak shapes were modeled using a fundamental
parameters approach featuring a direct convolution of source emission
profiles, axial instrument contributions, crystallite size, and microstrain
effects. The background was handled using a shifted Chebychev
polynom.
High-temperature PXRD was recorded on a Stadi P diffractometer
(Stoe & Cie GmbH, Darmstadt, Germany) with Mo Kα1 radiation
source and equipped with a graphite furnace and an image-plate
position-sensitive detector. The diffractograms were recorded in the
range of 2θ = 3−65° in temperature steps of 20 K starting from room
temperature to a maximum of 1000 °C. Samples were loaded into
fused silica capillaries (Hilgenberg GmbH, Malsfeld, Germany) with
0.5 mm diameter and 0.01 mm wall thickness.
Figure 1. SEM micrographs of representative crystals of (a)
Zr1−xPO3+4xN1−4x, (b) Zr0.75PO4, (c) Hf1−xPO3+4xN1−4x, and (d)
Hf0.75PO4.
Information (Figures S1 to S4, Table S1). The thermodynamic
stability of respective metal dioxides impeded the preparation
of phase pure compounds, but their formation can be, to a good
amount, circumvented by careful grinding of the starting
materials. Structure analysis (see Structure Determination)
showed that the nitrogen containing compounds both have
defects on the heavy atom positions, necessitating a surplus of
O for charge balancing and thus leading to sum formulas
M
1−xPO3+4xN1−4x (x ≈ 0.05, M = Zr, Hf). Infrared spectroscopy
did not indicate the presence of N−H or O−H bonds (Figure
S5). EDX of the compounds showed that the experimentally
determined composition is in good agreement with the
t h e o r e t i c a l v a l u e s : e x p e r i m e n t / t h e o r y
Zr1.17(1)P1.08(7)O2.99(17)N0.72(23)/Zr0.95P1O3.2N0.8 (three points),
Zr0. 85(6)P0.85(9)O4. 04(15)/Zr0.75P1O4 (three points),
Hf1.08(10)P1.01(4)O3.11(16)N0.72(13)/Hf0.95P1O3.2N0.8 (nine points),
Hf0.87(10)P0.94(8)O3.94(16)/Hf0.75P1O4 (eight points). EDX and
single-crystal analysis indicate that the amount of Zr/Hf and N/
O incorporated in M1−xPO3+4xN1−4x seems to favor a structure
with vacancies on the heavy atom position. We found no
experimental evidence for the existence of the end members
ZrPO3N and HfPO3N. This could have electrostatic causes, in
which for energy minimization vacancies are required. Single-
crystal diffraction and EDX, however, are not methods for
generating statistics; hence, the real compositional variance can
only be estimated. The composition certainly varies to some
degree around the postulated value of x ≈ 0.05, and in domains
the end members ZrPO3N and HfPO3N might be present. The
analysis of this, however, is beyond the scope of this report.
SYNTHESIS
■
The title compounds were prepared by high-pressure meta-
thesis starting from the metal chlorides, Li2O, and PON or
P2O5, respectively. Reaction equations are as follows:
ZrCl4 + PON + 2Li2O → ZrPO3N + 4LiCl
(2)
3ZrCl4 + 2P O5 + 6Li2O → 4Zr0.75PO + 12LiCl
(3)
(4)
(5)
2
4
HfCl4 + PON + 2Li2O → HfPO3N + 4LiCl
3HfCl4 + 2P O5 + 6Li2O → 4Hf0.75PO + 12LiCl
2
4
The conditions for each reaction are listed in Table 1. All
compounds were obtained as colorless and transparent single-
Table 1. Reaction Conditions for Reactions inEqs 2 to 5
STRUCTURE DETERMINATION
■
Reaction Pressure/GPa Temp/°C Dwell/min Temp up/down/min
The crystal structures of all compounds were solved and refined
from single-crystal data. Crystallographic data is summarized in
Table S2, atom positions in Table S3, lists of anisotropic
displacement parameters, bond lengths, and angles can be
found in the Supporting Information (Tables S4−S6). The
structures were solved in space group I41/amd (no. 141),
determined from systematically absent reflections, yielding the
positions of all heavy atoms including P. O and, respectively, N
were determined through difference Fourier maps. The
occupancy of the single Zr and Hf positions in
eq 2
eq 3
eq 4
eq 5
6
6
6
6
1200
1200
1100
1100
160
160
160
160
30/150
30/150
30/150
30/150
crystals large enough for XRD (Figure 1). Zr0.75PO4 and
Hf0.75PO4 were prepared as crystallographically phase-pure
materials, Zr1−xPO3+4xN1−4x had a minor, not-quantifiable, side
phase of ZrO2 and Hf1−xPO3+4xN1−4x has an 8 wt % side phase
of HfO2.42 Rietveld refinements are shown in the Supporting
M
1−xPO3+4xN1−4x (x ≈ 0.05, M = Zr, Hf) was freely refined
and the O/N occupancy adjusted to maintain charge neutrality.
C
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The M0.75PO4 structures were tested for additional vacancies on
the P and M positions; however, none were found, and
therefore, both occupancies were fixed to 1 and 3/4,
respectively (details in Supporting Information). As the heavy
atom positions of all products are not fully occupied, they
might form an ordered vacancy distribution. Such an ordering
would require either a lowering of the space group symmetry,
as only one Zr/Hf Wyckoff-position exists in I41/amd (no.
141), or an enlarged unit cell. Reconstructed planes through
reciprocal space (Figures S6 to S9) indicate neither the
existence of superstructure reflections nor reflections breaking
the systematic absences, as would be required for the maximal
translationengleiche subgroups of I41/amd (no. 141). Hence, a
random distribution of vacancies was assumed. The structure
models with a random distribution of vacancies were also
corroborated by 31P solid state NMR measurements carried out
on the M0.75PO4 compounds (Figure S10).
CaSO4 and scheelite CaWO4.44−46 A mathematical bisdi-
sphenoid is a deltahedron with symmetry D2d in Schonflies
̈
notation and vertex symbol (34)4(35)4; in the M(O/N)8
polyhedra of presented orthophosphates this symmetry is
retained (Figure 3b) but two different M−(O/N) distances
lead to irregular triangles as faces of the polyhedra.47
A comparison of the title compounds with ZrSiO4 and
HfSiO4 shows that the Hf compounds have a smaller unit cell
than their Zr counterparts (Figure 4).48,49 This shrinkage is due
STRUCTURE DISCUSSION
■
All title compounds crystallize in the ZrSiO4 structure-type,
space group I41/amd (no. 141), albeit being defect variants as
the heavy atom positions are not fully occupied. The crystal
structure comprises noncondensed P(O/N)4 tetrahedra, as
illustrated in projection along [100], alternatingly pointing up-
and downward (Figure 2). The tetrahedra are separated by
Figure 4. Comparison of lattice parameters a and c, unit cell volumes
V, and P−(X) and M−(X) (X = O/N) distances of all title compounds
as well as ZrSiO4 and HfSiO4.48,49 Cell parameters of the title
compounds were extracted from Rietveld refinement and interatomic
distances from single-crystal refinement. Standard deviations are
displayed for interatomic distances, not for lattice parameters as they
are too small to be visualized.
to the increased relativistic contraction of the 5d Hf atoms
compared to the lighter 4d Zr atoms, resulting in slightly
smaller Hf atoms.50 Chemically, Hf and Zr are very similar,
unlike, e.g. Mo and W; the relativistic contraction and the shell-
structure expansion cancel each other out, which might explain
why Hf and Zr adopt the same structure-type.51 Since the
heavy-atom positions in M0.75PO4 are occupied to 3/4, their
unit cell volume is expectedly smaller than that of the
oxonitridophosphates. The P−(O/N) bond lengths are longer
for the nitrogen containing compounds than in the pure
oxophosphates, but all bond lengths are shorter than the
respective Si−O interatomic distances of zircon and hafnon.
The P−(O/N) bond lengths found here are in good
agreement with values found in β-K2SO4-type Sr2PO3N (d(P−
(O/N)) = 1.569−1.574).52 For the M−(O/N) distances no
trend can be postulated, and the values of the two different
distances range from 2.135 to 2.163 Å and from 2.250 to 2.264
Å, respectively.
Figure 2. Unit cell of the Zr and Hf orthophosphates. Figure based on
structure model of Zr1−xPO3+4xN1−4x, with metal atoms in green, P in
black and N in orange: (a) random orientation of the unit cell; (b)
projection along [100]. Ellipsoids are set to a 90% probability level.
heavy atoms, which are coordinated in a M(O/N)8 bisdi-
sphenoid (also called dodecadeltahedron, Johnson solid J84,
Figure 3).43 The bisdisphenoid is common in structures with
noncondensed tetrahedra and has been observed in the room
temperature modification of AgClO4 as well as anhydrite
THERMAL PROPERTIES
■
High-temperature PXRD diffractograms (Figures S12 to S15)
were recorded to a maximum temperature of 1000 °C to
investigate the thermal stability of the compounds in air. The
oxophosphates M0.75PO4 seem to undergo a phase trans-
formation to ambient pressure polymorphs with the same
composition at ca. 900 °C for Zr0.75PO4 and at ca. 780 °C for
Hf0.75PO4. The diffraction pattern of the Zr0.75PO4 phase
transformation matches a unit cell reported by Alamo et al.
Figure 3. Heavy-atom M(O/N)8 coordination polyhedron based on
structure model of Zr1−xPO3+4xN1−4x, with metal atom in green and N
in orange: (a) polyhedron in random orientation; (b) polyhedron in
projection along the 4 rotoreflection. Ellipsoids are set to a 90%
̅
probability level.
D
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reported for Zr orthophosphate Zr2.25(PO4)3. This Zr2.25(PO4)3
was reported to enter the NASICON structure-type as the unit
cell is similar to that of NaZr2(PO4)3 (Figure S16).21 The phase
transformation of Hf0.75PO4 yields a similar powder pattern
(Figure S17), but no corresponding Hf2.25(PO4)3 structure has
been reported. Since the unit cell reported for Zr2.25(PO4)3 also
matches the reflection pattern after the phase transformation, a
similar structure can be assumed.
While no decomposition was observed for the
Hf1−xPO3+4xN1−4x up to 1000 °C, the oxonitridophosphate
Zr1−xPO3+4xN1−4x decomposes at ca. 680 °C. Most of the
diffraction pattern can be explained by the Zr2.25(PO4)3
reference; however, several reflections could not be assigned
to any known phase (Figure S18). From the chemical
composition the unknown phase should be rich in Zr, but
the pattern did not match that of ZrO2.42
of Eg, 3.6 and 4 eV, respectively. This absorption might not
only be caused by the valence to conduction band transition
but could also be influenced by local defects. Hence, a true
optical band gap cannot be derived here, only the beginning of
absorption, which is at ≈3.6 eV.
ALTERNATIVE SYNTHESIS METHODS
■
Metathesis reactions were successfully applied in the synthesis
of temperature labile compounds of, e.g., metal borides,
nitrides, and carbides.55 The energy for the product formation
is generated through the exothermic nature of the metathesis
reaction, allowing for lower furnace temperatures. Since
Hf0.75PO4 and Zr0.75PO4 are ternary phosphates they might
be accessible without applying high pressures. Analogous
metathesis reactions were therefore carried out in alumina
crucibles in sealed Ar ampules at 700 °C. This temperature was
chosen since the HTXRD data indicate that both phosphates
do not start to undergo a phase transformation at that
temperature. The reactions, however, did not yield the desired
orthophosphate as M2(PO4)2O (M = Hf, Zr), and unidentified
side phases were formed.19 Hence, formation of the
orthophosphate, with its defect ZrSiO4 structure, seemed to
be stabilized under high-pressure conditions. As mentioned in
the HTXRD section, the phosphates undergo a phase
transformation to the ambient pressure form of
Zr2.25(PO4)3.21 Hence the oxophosphates reported here might
be the high-pressure polymorphs of this structure-type and
metastable under ambient pressure conditions. Hence they
were not previously discovered despite the thorough
investigation of the group 4 phosphates. Thus high-pressure
metathesis opens a route to compounds that are thermally
labile and potentially metastable at ambient pressures.
OPTICAL PROPERTIES
■
The UV−vis spectra recorded for each compound reveal
valence to conduction band transitions in the UV range (Figure
S19). To obtain an estimate of the optical band gap, the
reflectance spectra were transformed into the Kubelka−Munk
function and plotted in a Tauc-plot (Figure 5).53,54 A direct
CONCLUSION
■
The preparation of the oxonitridophosphates with group 4
metals comprising noncondensed P(N/O)4 tetrahedra,
M1−xPO3+4xN1−4x (M = Zr, Hf) shows that high-pressure
metathesis can grant a systematic access to this class of
compounds. Naturally, more types of anion network have to be
prepared to underline the capability of the high-pressure
metathesis route.
Moreover, high-pressure metathesis could be a viable route
to unknown oxophosphates, as shown with the preparation of
Zr0.75PO4 and Hf0.75PO4. The latter are inaccessible by ambient
pressure synthesis methods and are probable high-pressure
polymorphs of the M2.25(PO4)3 phases crystallizing in the
NASICON structure-type. Zr0.75PO4 and Hf0.75PO4 have, most
likely, not been discovered before since they are metastable at
ambient pressures and could not be prepared without high-
pressure techniques. Hence, high-pressure metathesis could
also be used to target metastable systems of oxophosphates,
which were previously only accessible via kinetic control of
ambient pressure synthesis methods.
Figure 5. Tauc-plots of all compounds, generated from UV−vis
reflectance data. Kubelka−Munk function in black, linear regression
(blue) fit to data points highlighted in red. Special case for Hf0.75PO4
as two slopes are discernible; two regression lines (blue and green)
were fit to the data points (red and orange) yielding adsorption onset
values of 3.6 and 4 eV, respectively.
band gap was assumed for the structures. A linear regression
was fit to the regions of steep slope and the estimated band gap
value read off at the x-axis intersection. The values for the
M1−xPO3+4xN1−4x (M = Zr, Hf) are Eg ≈ 3.5 and 4.3 eV,
respectively, while Zr0.75PO4 has a value of Eg ≈ 4.2 eV. The
band gap of Hfr0.75PO4 cannot unambiguously be derived since
the UV−vis spectra shows no clear beginning of the valence to
conduction band transition (Figure S19d), which leads to a
Kubelka−Munk function with an additional point of inflection
in the region of steepest slope (Figure 5, bottom). Two
different slopes are discernible, both leading to different values
The orthophosphates reported here are an intriguing starting
point for the search for ion conduction properties within the
group 4 (nitrido)phosphates, as are known from the
structurally related class of NASICON materials.
E
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(10) Schnick, W.; Lucke, J. Zn [P N ]Cl - A Sodalite with a
Phosphorus−Nitrogen Framework. Angew. Chem., Int. Ed. Engl. 1992,
̈
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b00373.
7
12 24
2
■
S
31, 213−215.
(11) Stock, N.; Lucke, J.; Volkmann, M.; Jansen, M.; Schnick, W.
̈
Nitrido-Sodalites. II. Synthesis, Crystal Structure, and Properties of
M(6+(y/2)‑x)H2x[P12N24]Zy with M = Fe, Co, Ni, Mn; Z = Cl, Br, I; 0 <
X < 4; Y < 2. Z. Anorg. Allg. Chem. 1995, 621, 987−992.
Rietveld refinement data, infrared spectroscopy, crystallo-
graphic tables, details on structure determination, NMR
measurements, high-temperature powder diffraction,
UV−vis spectroscopy (PDF)
(12) Schnick, W.; Lucke, J. Zur Kenntnis von Lithium-phosphor(V)-
̈
nitrid. Reindarstellung und Verfeinerung der Kristallstruktur von
LiPN2. Z. Anorg. Allg. Chem. 1990, 588, 19−25.
(13) Horstmann, S.; Irran, E.; Schnick, W. Synthesis and Crystal
Structure of Phosphorus(V) Nitride α-P3N5. Angew. Chem., Int. Ed.
Engl. 1997, 36, 1873−1875.
Accession Codes
CCDC 1816010−1816013 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(14) Scheetz, B. E.; Agrawal, D. K.; Breval, E.; Roy, R. Sodium
Zirconium Phosphate (NZP) as a Host Structure for Nuclear Waste
Immobilization: A Review. Waste Manage. 1994, 14 (6), 489−505.
(15) Martinek, C. A.; Hummel, F. A. Subsolidus Equilibria in the
System ZrO2−WO3−P2O5. J. Am. Ceram. Soc. 1970, 53, 159−161.
(16) Laud, K. R.; Hummel, F. A. Subsolidus Relations in the System
ZrO2−ThO2−P2O5. J. Am. Ceram. Soc. 1971, 54, 407−409.
(17) Stinton, G. W.; Hampson, M. R.; Evans, J. S. O. The 136-Atom
Structure of ZrP2O7 and HfP2O7 from Powder Diffraction Data. Inorg.
Chem. 2006, 45, 4352−4358.
(18) Gebert, W.; Tillmanns, E. Die Kristallstruktur von Dizirkon-
diorthophosphat. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst.
Chem. 1975, 31, 1768−1770.
(19) Wallez, G.; Souron, J.-P.; Quarton, M. Crystal Structure and
Thermal Evolution of Inedite α-Zr2O(PO4)2 and α-Hf2O(PO4)2. Solid
State Sci. 2006, 8, 1061−1066.
(20) Gorbunova, Y. E.; Ilyukhin, V. V.; Kuznetsov, V. G.; Lavrov, A.
V.; Linde, S. A. Crystal Structure of Zirconium Polyphosphate
Zr(PO3)4-I. Dokl. Akad. Nauk SSSR 1976, 228, 1329−1331.
(21) Alamo, J.; Rustum, R. Revision of Crystalline Phases in the
System ZrO2−P2O5. J. Am. Ceram. Soc. 1984, 67, 80−82.
(22) Paschos, O.; Kunze, J.; Stimming, U.; Maglia, F. A Review on
Phosphate Based, Solid State, Protonic Conductors for Intermediate
Temperature Fuel Cells. J. Phys.: Condens. Matter 2011, 23, 234110.
(23) Anantharamulu, N.; Koteswara Rao, K.; Rambabu, G.; Vijaya
Kumar, B.; Radha, V.; Vithal, M. A Wide-Ranging Review on Nasicon
Type Materials. J. Mater. Sci. 2011, 46, 2821−2837.
AUTHOR INFORMATION
Corresponding Author
*E-mail: wolfgang.schnick@uni-muenchen.de.
ORCID
Wolfgang Schnick: 0000-0003-4571-8035
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
Financial support by the Deutsche Forschungsgemeinschaft
(DFG), Grant SCHN377/18-1, is gratefully acknowledged.
The authors thank the Fonds der Chemischen Industrie (FCI)
for a Ph.D. fellowship for S.D.K. We thank Christian Minke and
Dr. Thomas Brauniger (both at LMU Munich) for the NMR
̈
measurements and helpful discussion.
REFERENCES
■
(1) Kloß, S. D.; Schnick, W. Rare-Earth-Metal Nitridophosphates
through High-Pressure Metathesis. Angew. Chem., Int. Ed. 2015, 54,
11250−11253.
(24) Dacheux, N.; Podor, R.; Brandel, V.; Genet, M. Investigations of
Systems ThO2−MO2−P2O5 (M = U, Ce, Zr, Pu). Solid Solutions of
thorium−uranium (IV) and thorium−plutonium (IV) phosphate−
diphosphates. J. Nucl. Mater. 1998, 252, 179−186.
(2) Kloß, S. D.; Weidmann, N.; Niklaus, R.; Schnick, W. High-
Pressure Synthesis of Melilite-Type Rare-Earth Nitridophosphates
RE2P3N7 and a Ba2Cu[Si2O7]-Type Polymorph. Inorg. Chem. 2016, 55,
9400−9409.
(25) Zintl, E.; Harder, A.; Dauth, B. Gitterstruktur der Oxyde,
Sulfide, Selenide und Telluride des Lithiums, Natrium und Kaliums. Z.
Elektrochem. 1934, 40, 588−593.
(3) Kloß, S. D.; Weidmann, N.; Schnick, W. Anti-Perovskite
Nitridophosphate Oxide Ho3[PN4]O by High-Pressure Metathesis.
Eur. J. Inorg. Chem. 2017, 2017, 1930−1937.
(26) Klement, R.; Koch, O. Phosphoroxy-Triamid Und Phosphor-
thio-Triamid. Chem. Ber. 1954, 87, 333−340.
(4) Kloß, S. D.; Neudert, L.; Doblinger, M.; Nentwig, M.; Oeckler,
̈
(27) Baumann, D.; Niklaus, R.; Schnick, W. A High-Puressure
Polymorph of Phosphorus Oxonitride with the Coesite Structure.
Angew. Chem., Int. Ed. 2015, 54, 4388−4391.
O.; Schnick, W. Puzzling Intergrowth in Cerium Nitridophosphate
Unraveled by Joint Venture of Aberration-Corrected Scanning
Transmission Electron Microscopy and Synchrotron Diffraction. J.
Am. Chem. Soc. 2017, 139, 12724−12735.
́
(28) Leger, J. M.; Haines, J.; Chateau, C.; Bocquillon, G.; Schmidt,
M. W.; Hull, S.; Gorelli, F.; Lesauze, A.; Marchand, R. Phosphorus
oxynitride PON, a silica analogue: structure and compression of the
cristobalite-like phase; P-T phase diagram. Phys. Chem. Miner. 2001,
28, 388−398.
(5) Kloß, S. D.; Wandelt, S.; Weis, A.; Schnick, W. Accessing
tetravalent transition metal nitridophosphates through high-pressure
metathesis. Angew. Chem., Int. Ed. 2018, 57, 319210.1002/
anie.201712006.
(29) Huppertz, H. Multianvil High-Pressure/High-Temperature
Synthesis in Solid State Chemistry. Z. Kristallogr. - Cryst. Mater.
2004, 219, 330−338.
(6) Marchand, R.; Schnick, W.; Stock, N. Molecular, Complex Ionic,
and Solid State PON Compounds. Adv. Inorg. Chem. 2000, 50, 193−
233.
I
III
I
(30) Rubie, D. C. Characterising the Sample Environment in
Multianvil High-Pressure Experiments. Phase Transitions 1999, 68,
431−451.
̈
(7) Feldmann, W. Uber Nitridophosphate M 3M P3O9N (M Na;
K, MIII = Al; Ga; Cr, Fe; Mn). Z. Chem. 1987, 27, 182−183.
(8) Pucher, F. J.; Hummel, F.; Schnick, W. CuPN2: Synthesis, Crystal
Structure, and Electronic Properties. Eur. J. Inorg. Chem. 2015, 2015,
1886−1891.
(31) Walker, D.; Carpenter, M. A.; Hitch, C. M. Some Simplifications
to Multianvil Devices for High Pressure Experiments. Am. Mineral.
1990, 75, 1020−1028.
(9) Pucher, F. J.; Karau, F. W.; Schmedt auf der Gunne, J.; Schnick,
W. CdP2N4 and MnP2N4 - Ternary Transition-Metal Nitridophos-
phates. Eur. J. Inorg. Chem. 2016, 2016 (10), 1497−1502.
̈
(32) Walker, D. Lubrication, Gasketing, and Precision in Multianvil
Experiments. Am. Mineral. 1991, 76, 1092−1100.
F
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(33) Kawai, N.; Endo, S. The Generation of Ultrahigh Hydrostatic
Pressures by a Split Sphere Apparatus. Rev. Sci. Instrum. 1970, 41,
1178−1181.
(34) Bruker AXS, Inc. SADABS, Madison, Wisconsin, USA, 2001.
(35) Bruker AXS, Inc. APEX3, Madison, Wisconsin, USA, 2016.
(36) Bruker AXS, Inc. XPREP, Karlsruhe, Germany, 2001.
(37) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2008, 64, 112−122.
(38) Sheldrick, G. M. SHELXS - A Program for Crystal Structure
Solution; University of Gottingen: Gottingen, Germany, 1997.
̈
̈
(39) Momma, K.; Izumi, F. VESTA 3 for Three-Dimensional
Visualization of Crystal, Volumetric and Morphology Data. J. Appl.
Crystallogr. 2011, 44, 1272−1276.
(40) Rietveld, H. M. A Profile Refinement Method for Nuclear and
Magnetic Structures. J. Appl. Crystallogr. 1969, 2, 65−71.
(41) Coelho, A. A., TOPAS-Academic V4.1; Coelho Software:
Brisbane, Australia, 2007.
(42) Adam, J.; Rogers, M. D. The Crystal Structure of ZrO2 and
HfO2. Acta Crystallogr. 1959, 12, 951−951.
(43) Johnson, N. W. Convex Polyhedra with Regular Faces. Canad. J.
. Math. 1966, 18, 169−200.
́
(44) Errandonea, D.; Gracia, L.; Beltran, A.; Vegas, A.; Meng, Y.
Pressure-Induced Phase Transitions in AgClO4. Phys. Rev. B: Condens.
Matter Mater. Phys. 2011, 84, 1−11.
(45) Hawthorne, F. C.; Ferguson, R. B. Anhydrous Sulphates. II.
Refinement of the Crystal Structure of Anhydrite. Can. Mineral 1975,
13, 289−292.
(46) Kay, M. I.; Frazer, B. C.; Almodovar, I. Neutron Diffraction
Refinement of CaWO4. J. Chem. Phys. 1964, 40, 504−506.
(47) Blatov, V. A.; O’Keeffe, M.; Proserpio, D. M. Vertex-, Face-,
Point-, Schlafli-, and Delaney-Symbols in Nets, Polyhedra and Tilings:
̈
Recommended Terminology. CrystEngComm 2010, 12, 44−48.
́
(48) Krstanovic, I. R. Redetermination of the Oxygen Parameters in
Zircon (ZrSiO4). Acta Crystallogr. 1958, 11, 896−897.
(49) Speer, J. A.; Cooper, B. J. Crystal Structure of Synthetic Hafnon,
HfSiO4, Comparison with Zircon and the Actinide Orthosilicates. Am.
Mineral. 1982, 67, 804−808.
(50) Pyykko, P. Relativistic Effects in Structural Chemistry. Chem.
Rev. 1988, 88, 563−594.
(51) Pyykko, P.; Desclaux, J.-P. Relativity and the Periodic System of
̈
Elements. Acc. Chem. Res. 1979, 12, 276−281.
(52) Marchuk, A.; Schultz, P.; Hoch, C.; Oeckler, O.; Schnick, W.
M2PO3N (M = Ca, Sr): Ortho-Oxonitridophosphates with β-K2SO4
Structure Type. Inorg. Chem. 2016, 55, 974−982.
(53) Tauc, J.; Grigorovici, R.; Vancu, A. Optical Properties and
Electronic Structure of Amorphous Germanium. Phys. Status Solidi B
1966, 15, 627−637.
(54) Davis, E. A.; Mott, N. F. Conduction in Non-Crystalline
Systems V. Conductivity, Optical Absorption and Photoconductivity in
Amorphous Semiconductors. Philos. Mag. 1970, 22, 903−922.
(55) Meyer, H.-J. Solid state metathesis reactions as a conceptual tool
in the synthesis of new materials. Dalton Trans. 2010, 39, 5973−5982.
G
DOI: 10.1021/acs.inorgchem.8b00373
Inorg. Chem. XXXX, XXX, XXX−XXX