Chemistry and physical properties of the phosphide telluride Zr 2PTe2

Article abstract of DOI:10.1002/ejic.200900346

The synthesis of the phosphide telluride Zr₂PTe₂ by solidstate reaction from the elements at 850 °C or by thermitetype reaction of Zr and Te₈O₁₀(PO₄)₄ was accomplished. Crystals were grown

Full text of DOI:10.1002/ejic.200900346

FULL PAPER
DOI: 10.1002/ejic.200900346
Chemistry and Physical Properties of the Phosphide Telluride Zr2PTe2
Kristina Tschulik,[a] Michael Ruck,[a] Michael Binnewies,[b] Edgar Milke,[b]
Stefan Hoffmann,[c] Walter Schnelle,[c] Boniface P. T. Fokwa,[d] Michael Gilleßen,[d] and
Peer Schmidt*[a]
Dedicated to Prof. Dr. Martin Jansen on the occasion of his 65th birthday
Keywords: Phosphorus / Tellurium / Zirconium / Chemical vapor transport / Thermodynamics / Structure elucidation
The synthesis of the phosphide telluride Zr2PTe2 by solid-
state reaction from the elements at 850 °C or by thermite-
type reaction of Zr and Te8O10(PO4)4 was accomplished.
Crystals were grown by chemical vapour transport by using
iodine from 800 °C in the direction of higher temperatures up
to 900 °C. The chemical bonding in Zr2PTe2 was investigated
by the linear muffin-tin orbital (LMTO) method. The non-
vanishing DOS of zirconium and tellurium states at the Fermi
level (EF) complies with the metallic conductivity (ρ ≈
40 µΩcm at 300 K) and temperature-independent Pauli para-
magnetism.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
stand the complex mechanisms of phase formation. We suc-
ceeded in synthesising the new metal phosphide telluride
Zr2PTe2 through several pathways and investigated its ther-
mal properties by using experimental methods and thermo-
dynamic modelling. On the basis of these results single-
crystals became available for the determination of the crys-
tal structure and the physical properties. Results of mea-
surements of electrical resistivity and magnetic suscep-
tibility are discussed together with band-structure calcula-
tions. In this way it is possible to give a comprehensive view
on the chemistry and the physical properties of Zr2PTe2.
Introduction
Quite a large number of zirconium pnictide chalcogen-
ides have been described so far as nitride sulfides,[1] arsenide
selenides,[2] arsenide tellurides,[2,3] antimonide selenides[4]
and antimonide tellurides.[5] In contrast, ternary phosphide
tellurides seem to be extraordinary compounds, because
only a small number of those are known by now. Two of
them consist of isolated phosphide and telluride anions: the
uranium phosphide telluride UPTe[6,7] and the recently
found compound Ti2PTe2.[8] Four other phosphorus- and
tellurium-containing compounds are known: the three tel-
luro-phosphides IrPTe,[9] OsPTe[10] and RuPTe,[10] con-
sisting of M3+ cations besides [P–Te]3– dumbbells, and the
Results and Discussion
[11]
compound BaP4Te2
built from Ba2+ cations and
1
ϱ[P4Te2]2– anions.
Methods of Synthesis
Remarkably, no binary phosphorus and tellurium com-
pounds are known, which additionally reflects the low af-
finity of the two elements as well as kinetic restraints.
Therefore, syntheses of compounds containing phosphorus
and tellurium are quite challenging and detailed studies of
their thermal properties are indispensable in order to under-
We achieved the synthesis of Zr2PTe2 in three different
ways. Single-phase products can be obtained performing
common solid-state reactions. Besides the possibility of
using the elements as starting materials [Equation (1)],
which require strictly controlled thermal conditions, the bi-
naries ZrP and ZrTe2 can be used as reagents under easily
controllable conditions [Equation (2)].
[a] Inorganic Chemistry, Dresden University of Technology,
Helmholtzstraße 10, 01069 Dresden, Germany
Fax: +49-351-463-37287
E-mail: peer.schmidt@chemie.tu-dresden.de
http://www.peer-schmidt.de
[b] Inorganic Chemistry, Leibniz University Hannover,
Callinstr. 9, 30167 Hannover, Germany
2Zr + P + 2Te Ǟ Zr2PTe2
(1)
ZrP + ZrTe2 Ǟ Zr2PTe2
(2)
[c] Max Planck Institute for Chemical Physics of Solids,
Nöthnitzer Str. 40, 01187 Dresden, Germany
[d] Inorganic Chemistry, RWTH Aachen,
Landoltweg 1, 52074 Aachen, Germany
Sublimation of phosphorus and tellurium is not negligi-
ble at temperatures above 500 °C. Yet, below this tempera-
ture an adjustment of the ternary equilibrium is not ac-
3102
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 3102–3110

Chemistry and Physical Properties of Zr2PTe2
complishable in reasonable reaction times. Actually, control
Because the crystallisation of zirconium phosphide tellu-
of the composition of a solid under evaporation conditions ride through this reaction path is even worse than in those
of its components is a general problem of synthesis of phos- paths mentioned before, chemical vapour transport reac-
phorus- and tellurium-containing compounds. Hence, it is tions were performed to gain (single) crystals.
extremely difficult to ensure a complete incorporation of
tellurium and phosphorus – which sublime readily – into
the ternary compound at the reaction temperature of
800 °C, according to Equation (1).
Thermal Properties
Thermogravimetric Measurements
However, starting with ZrP and ZrTe2 [Equation (2)] di-
minishes this problem, as only the latter compound decom-
poses remarkably at the reaction temperature. Conse-
quently, the equilibrium state is easier to access. Nonethe-
less, slow cooling of the reaction mixture is required to en-
sure complete incorporation of tellurium and phosphorus
into the solid to form Zr2PTe2. Although no explicit kinetic
studies were carried out, no single-phase product was ob-
served at higher cooling rates.
The thermal decomposition of Zr2PTe2 does not start be-
low 1000 °C; its degradation temperature is higher than for
ZrTe2, whereas ZrP is stable under the given experimental
conditions. The ternary compound decomposes in one sin-
gle step that is not finished at 1500 °C when applying a
heating rate of 10 Kh–1. Thus, another decomposition pro-
cess can be observed along the cooling curve (– – –, Fig-
ure 2) of the TG measurement down to 1300 °C.
A long reaction time is also necessary to obtain a well-
crystallised product. Reduction of reaction time by increas-
ing the reaction temperature above 850 °C results in side
products due to reactions with the silica wall of the reaction
tube.
The third method of synthesis is a thermite-type reaction
of elementary zirconium and the tellurium oxide phosphate
Te8O10(PO4)4 leading to the products ZrO2 and Zr2PTe2
[Equation (3)].
21/4Zr + 1/4Te8O10(PO4)4 Ǟ Zr2PTe2 + 13/4ZrO2
(3)
This way, the kinetic restraints of the other methods of
synthesis can be avoided. The use of Te8O10(PO4)4, which
remains solid at temperatures below 800 °C,[8] is preferred
over TeO2 and P4O10, which would sublime at lower tem-
peratures.
The possibility to form Zr2PTe2 besides ZrO2 through a
redox reaction with elementary zirconium originates from
differences in the oxygen partial pressures between tel-
lurium and phosphorus oxides and elemental zirconium,
respectively. These differences represent a potential gradient
in the sense of an electromotive series of solids.[12] The
equalisation of this gradient results in a thermodynamic
equilibrium, represented by zirconium oxide besides ele-
mental phosphorus and tellurium (Figure 1).
Figure 2. Thermal decomposition of Zr2PTe2 applying a heating
rate of 10 Kmin–1 in comparison with the degradation of the bi-
nary compound ZrTe2.
In contrast to the thermal decomposition of Ti2PTe2,[8]
which decomposes in several steps at much lower tempera-
tures to form Ti2P (s) with release of gaseous tellurium spe-
cies only, decomposition of Zr2PTe2 occurs with release of
gaseous species of phosphorus and tellurium. The residue
is a ternary phase with increased Zr content [Equation (4)].
(1 + δ/2)Zr2PTe2(s) = Zr2+δPTe2(s) + δ/8P4(g) + δ/2Te2(g)
(4)
Unfortunately, total pressure measurements with the use
of a silica zero-point membrane manometer,[13] as per-
formed for Ti2PTe2,[8] are not useful in the case of Zr2PTe2,
as sufficient partial pressures are not reached.
Thermodynamic Modelling of the System Zr/P/Te
In order to gain a better understanding of the complex
mechanism of the formation and the thermal behaviour of
Zr2PTe2 and to optimise the synthesis, a detailed thermo-
dynamic modelling and description of the ternary system
Zr/P/Te was established by using the computer program
Tragmin.[14] This was hampered by lacking thermodynamic
data for the binary systems Zr/P and Zr/Te. Nevertheless,
we succeeded by considering those condensed species for
which thermodynamic data are known [Zr5Te4(s), ZrTe2(s),
Zr(s), P(s,red), Te(s)] and adding ZrP(s) and Zr2PTe2(s) as
essential species, whose thermodynamic data had to be esti-
mated.
Figure 1. Oxygen partial pressure and electrochemical potential
above the oxides of zirconium, phosphorus and tellurium calcu-
lated for T = 1000 K {calculation of log[p(O2)/bar] and E based on
data in ref.[31] see ref.[12]}, the arrow marks the level of equalisation
of p(O2).
Eur. J. Inorg. Chem. 2009, 3102–3110
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3103

P. Schmidt et al.
FULL PAPER
All gaseous species that have been found by mass spec-
trometric analysis of the gas-phase during the decomposi-
tion of Zr2PTe2 were taken into account: P4(g), P2(g), P(g),
Te2(g) and Te(g).
The results of the modelling of the thermal decomposi-
tion of Zr2PTe2 over a temperature range of 700–1000 °C
are visualised in Figure 3. In agreement with Equation (4)
P4(g) and Te2(g) represent the dominating gaseous species
besides P2(g) and Te(g), whereas the calculated partial pres-
sure of Zr(g) is negligible {p[Zr(g)] ≈ 10–28 bar}. Conse-
quently the calculations confirm the assumption that the
mechanism of the thermal decomposition of Zr2PTe2 differs
fundamentally from the decomposition mechanism of
Ti2PTe2 as a result of the formation of a homogeneity range
Zr2+δPTe2(s).
Figure 4. Results of the mass spectrometric analysis of a precipitate
of Zr2PTe2 heated to ϑ = 520 °C in the presence of iodine.
Because all components of Zr2PTe2 form sufficient
amounts of gaseous species in the quaternary system Zr/P/
Te/I, chemical vapour transport could be possible, if the
solubility of all these components can be lowered signifi-
cantly by applying a reasonable temperature gradient. For
this reason, mass spectrometric analysis of the gas phase
over Zr2PTe2 containing iodine was performed at different
temperatures. Actually, the relative intensities of gaseous
species of zirconium (ZrI3+), phosphorus (P4+, P2+) and tel-
lurium (Te+, Te2+) decrease with increasing temperature, as
it becomes obvious by comparison of Figures 4 (ϑ =
520 °C) and 5 (ϑ = 600 °C). We assume that the chemical
vapour transport of Zr2PTe2 proceeds in the quaternary
system Zr/P/Te/I in the direction of increasing temperatures
from source to sink.
Figure 3. Temperature dependence of the gas-phase equilibrium of
Zr2PTe2.
Chemical Vapour Transport
Because single crystals, needed for investigation of the
crystal structure of the new compound, were not accessible
by the methods of synthesis introduced before, and isola-
tion of Zr2PTe2 from the product mixture of the thermite-
type reaction was required, chemical vapour transport of
the zirconium phosphide telluride was aspired.
An attempt of chemical vapour transport of Zr2PTe2 into
the direction of lower temperatures by adding TeCl4 – by
using the parameters for chemical vapour transport of
Ti2PTe2 – failed. Detailed investigations of the appropriate
transport agents and transport directions/temperature gra-
dients became necessary.
Figure 5. Results of the mass spectrometric analysis of a precipitate
of Zr2PTe2 heated to ϑ = 600 °C in the presence of iodine.
Chemical Vapour Transport Experiments of Zr2PTe2
Knudsen Cell Mass Spectrometry
For all attempts of chemical vapour transport, 300 mg of
As shown in Figure 4, the results of Knudsen cell mass Zr2PTe2 powder and 15 mg of iodine were stacked in a large
spectrometric analysis of gas phase Zr2PTe2 containing io- ampoule (l = 12 cm, A = 1.2 cm2). By using a horizontal
dine reflect the fact that iodine is able to dissolve Zr2PTe2 two-zone furnace, the residual was heated to ϑ = 800 °C
by reactions forming gaseous species at temperatures much (source), and the top of the ampoule was heated to ϑ =
lower than 1000 °C. Obviously, binary compounds of zirco- 900 °C (sink) for five days. Near the top of the ampoule
nium and iodine, like ZrI4 are the gaseous species of zirco- shiny black hexagonal platelets of Zr2PTe2 resulted. Unfor-
nium with the highest partial pressure, yet neither binary tunately, these platelets were intergrown crystals and hence
species of phosphorus nor tellurium can be found. The lat- not suitable for crystal investigations.
+
ter two elements are detectable as P4+, P2 and P+ as well
In order to optimise the conditions for growth of single
+
as Te2 and Te+, respectively, whereas Te+ is the second crystals, chemical vapour transport was performed at vari-
most frequently detected species, surpassed only by I+.
ous source temperatures and temperature gradients. All
3104
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 3102–3110

Chemistry and Physical Properties of Zr2PTe2
other parameters were kept constant and equal to those Te2(g), P4(g) and ZrI4(g) are the dominating gaseous spe-
mentioned above. These experiments were carried out by cies, which was not directly visible by the results of mass
using a transport balance[15] that permits in situ measure- spectrometric analysis due to fragmentation and unequal
ments of the mass flow by time. Figure 6 shows the trans- tendency of ionisation of the different gaseous species.
ported mass for three different temperature gradients:
Figure 7. Temperature dependence of the gas-phase equilibria and
Figure 6. Rate of mass transport by vapour transport experiments
composition of gas phase for reaction of Zr2PTe2(s) with iodine;
of Zr2PTe2, starting temperature ϑsource = 800 °C, precipitation
temperature ϑsink = ϑsource + ∆T.
[p(i) Ͻ 10–5 bar for i = ZrI, ZrI3, P, TeI4 not shown].
Basically, transports are feasible from the source at ϑ =
800 °C in a wide range of sink temperature ϑsink = ϑsource
+ ∆T with ∆T = 20–100 K. Large gradients result in a large
amount of deposited polycrystalline material and are thus
suitable for purification and separation. A small gradient
leads to formation of a diminutive number of crystals. Some
of them were convenient for single-crystal investigations.
The rates of mass transport displayed in Table 1 were deter-
mined as the slopes of linear fits of the different m-t-tran-
sients.
Furthermore, the calculated transport efficiencies of all
gaseous species for a chemical vapour transport from 800
(source) to 900 °C (sink) enables a clarification of the trans-
port mechanism and specification of the transport reaction.
The transport efficiency is defined as the difference of the
partial pressure of a compound in the source and its partial
pressure in the sink normalised to the partial pressure of a
solvent L at both temperatures {here inert nitrogen gas, see
[Equation (5)]}. This is attributed to the fact that transport
effective species exhibit positive and transport agents nega-
tive transport efficiencies, respectively, whereas transport
efficiencies are zero for species that are not involved in the
transport reaction.[16]
Table 1. Rate of mass transport by vapour transport experiments
of Zr2PTe2, starting temperature ϑsource = 800 °C, precipitation
temperature ϑsink = ϑsource + ∆T.
ϑsource / °C
∆T / K
Rate of mass transport / mgh–1
800
800
800
20
50
100
0.1
0.3
1.0
(5)
Admittedly, as a result of the temperature difference be-
tween source and sink, secondary gas phase equilibria
[Equation (6)] are included in the calculated transport effi-
ciencies as well, which have to be deducted.
Thermodynamic Modelling of the Chemical Vapour
Transport of Zr2PTe2
In order to verify the predicted conditions of chemical
vapour transport of Zr2PTe2 as well as to reveal its mecha-
nism, thermodynamic modelling of the quaternary system
Zr/P/Te/I was performed by using the program Tragmin.[14]
In addition to the species considered in the modelling of
the ternary system Zr/P/Te the iodine containing species
ZrI4(s), ZrI2(s), TeI4(s), I2(s), ZrI4(g), ZrI3(g), ZrI2(g),
ZrI(g), PI3(g), TeI4(g) and TeI2(g) were taken into account
(Table 3). Zr(g) was not taken into account, as isothermal
calculations proved its influence on the chemical vapour
transport to be negligible {p[Zr(g)] ≈ 10–28 bar}.
(6)
The effective transport efficiencies received this way are
visualised in Figure 8. Hence Zr2PTe2 is transported, ac-
cording to the transport reaction [Equation (7)].
4Zr2PTe2(s) + 32I(g) = 8ZrI4(g) + P4(g) + 4Te2(g)
(7)
The composition of gas phase (Figure 7) displays only
species with partial pressures exceeding p(i) = 10–5 bar, as
Additional calculations indicate that an enrichment of
this region of partial pressures is said to be relevant for zirconium by chemical vapour reaction is possible by ad-
chemical vapour transport reactions. Obviously all elements justment of different parameters: high temperature of the
can be found in the gas phase, as already known from mass source, major temperature gradient, zirconium-rich precipi-
spectrometric analysis. Yet the calculation shows that tate in the source and a small amount of iodine.
Eur. J. Inorg. Chem. 2009, 3102–3110
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3105

P. Schmidt et al.
FULL PAPER
sequence, but in these compounds octahedral voids between
the sandwich layers are partially or completely filled.[3]
In Zr2PTe2 the interatomic distance Zr–Te is
288.98(2) pm, which is slightly larger than the one in the
layered binary compound ZrTe2 (282 pm; CdI2 structure
type[21]). This might be due to the fact that in Zr2PTe2 the
zirconium cations are coordinated by three (smaller) phos-
phide and three (larger) telluride anions and therefore
slightly move out of the centre of the octahedra in the direc-
tion of the phosphide anions, so that a trigonal-antipris-
matic coordination results. On the contrary, zirconium ions
occupy the central position of the octahedral voids in
ZrTe2, as they are coordinated by six equivalent telluride
ions. The interatomic distances Zr–P in Zr2PTe2
Figure 8. Effective transport efficiencies of gaseous species above a
precipitate of Zr2PTe2.
[262.98(2) pm] and ZrP (263 pm; NaCl structure type[22]
)
Crystal Structure of Zr2PTe2
are about equal. Telluride ions in adjacent layers are sepa-
rated by 380.52(2) pm from each other in Zr2PTe2, which is
slightly shorter than that in the similar compound
Ti2PTe2.[20] This can be explained by the fact that – as zirco-
nium cations [r[6](Zr4+) = 72 pm[23]] are larger than titanium
cations [r[6](Ti4+) = 61 pm[23]] – they expand the layers in
the hexagonal plane and hence the distance between tel-
lurium atoms in one layer. In consequence, telluride anions
in adjacent layers lie closer to each other.
Zr2PTe2 crystallises in the Bi2STe2 structure type.[17] For
better recognition of structural motives the hexagonal set-
ting will be described in the following (Figure 9, prepared
with DIAMOND[18]). Phosphide as well as telluride anions
are arranged in hexagonal layers, which are stacked along
the hexagonal c axis in a sequence with two layers of tellu-
ride ions followed by one layer of phosphide ions. All octa-
hedral vacancies between phosphide and telluride layers are
filled by zirconium cations. Because the voids between
neighbouring telluride layers remain completely empty,
sandwich layers [Te–Zr–P–Zr–Te] result. Thereby, the
anions form an ordered close sphere packing with stacking
order [ABABCBCAC], which is known from samarium
metal.[19] The zirconium cations fill two thirds of the octa-
hedral vacancies of this packing. The zirconium positions
as well as the phosphorus and tellurium positions were
found to be fully occupied.
Obviously the compound Zr2PTe2 can be described
with
two
different
ionic
boundary
formulae:
[(Zr4+)(Zr3+)(P3–)(Te2–)2] or [(Zr4+)2(P3–)(Te2–)2e–]. Here the
crystal structure might be indicative of the latter to be pre-
ferred, as only one zirconium position was found and not
several positions as might be expected for different oxi-
dation states of zirconium. In order to verify this assump-
tion physical properties of Zr2PTe2 were determined, as the
compound should be metallic according to the second for-
mula.
Physical Properties of Zr2PTe2
Measurements of Electrical Resistivity
Figure 10 shows the results of electrical resistivity mea-
surements in a temperature range of 4 K to 320 K. Zr2PTe2
exhibits a resistivity of 40 µΩcm at 300 K and a linear
increase in resistivity with temperature. This typical
metallic behaviour corresponds to the ionic formula
Figure 9. From left to right: crystal structure of Zr2PTe2 [projection
along the b axis (unit cell outlined)] with [PZr6] octahedra (grey
drawn polyhedra) and stacking sequence of the hexagonal layers of
anions (Latin letters) and cations (Greek letters); trigonal-antipris-
matic coordination of Zr (light grey) by three Te (black) and three
P (white) ions; µ3-bridging of [PZr6] octahedra by Te ions. Plotted
ellipsoids represent 99.99% probability at 293(2) K.
The same structural principle was reported for
Ti2PTe2.[20] Furthermore, the zirconium arsenide tellurides
Figure 10. Temperature dependence of the electrical resistivity of
Zr0.29Zr2AsTe2 and NaZr2AsTe2 exhibit the same stacking Zr2PTe2.
3106
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 3102–3110

Chemistry and Physical Properties of Zr2PTe2
[(Zr4+)2(P3–)(Te2–)2e–], that is, delocalised electrons. The low orbitals have the greatest contributions and are therefore
residual resistivity indicates the absence of large grain mainly responsible for the metallic behaviour of the com-
boundary resistances.
pound. Above the Fermi level, Zr valence orbitals show a
greater contribution to the DOS.
Measurements of Magnetic Susceptibility
Measurements of magnetic properties of Zr2PTe2 were
performed for various magnetic fields between 20 Oe and
70 kOe in the temperature range 1.8–400 K. Neither phase
transitions nor superconductivity were detected within
these ranges.
Figure 11 exemplifies the temperature dependence of cor-
rected magnetic susceptibility. The visible decrease in mag-
netic susceptibility is due to a paramagnetic contamination
of the analysed powder specimen, whereas its main compo-
nent Zr2PTe2 is characterised by temperature-independent
Pauli paramagnetism. The value of magnetic susceptibility
at 0 K was extrapolated to χmol,0 ≈ –54ϫ10–6 emumol–1.
Subtraction of the diamagnetic increments of the included
ions (Σχdia ≈ –164ϫ10–6 emumol–1 [24]) reveals the Pauli
paramagnetic term χp ≈ 110ϫ10–6 emumol–1. As this
equals a density of states of about 3.4 stateseV–1 at the
Fermi level Zr2PTe2 could be proved to be a metallic com-
pound by means of its magnetic properties.
Figure 12. Total and partial DOS curves for Zr2PTe2 obtained from
spin-polarised calculations. The Fermi level has been set to zero.
A chemical bonding analysis is possible on the basis of
the shapes and integrated values of COHP curves (ICOHP
= integrated value of COHP) for various interatomic inter-
actions in the solid. The COHP curve in Figure 13a indi-
cates that Zr–P orbital interactions are optimised in
Zr2PTe2: bonding orbitals are filled, antibonding orbitals
are empty. However, the Zr–Te orbital interactions are not
fully optimised, as some antibonding orbitals are present at
the Fermi level (Figure 13b), which surely weaken the bond-
ing compared to Zr–P. Both P–P and Te–Te interactions
(not shown) reveal antibonding character around the Fermi
level. The Zr–Zr orbital interactions show virtual bonding
levels in the conduction band (Figure 13c) indicating that
this unit could host even more electron density, in agree-
ment with the slightly lowered DOS (pseudogap) just above
the Fermi level. This COHP analysis suggests that the het-
eroatomic Zr–P and Zr–Te bonds are responsible of the
structural stability of Zr2PTe2, as expected. In agreement
with this assessment, the largest ICOHP values are found
for the Zr–P (–0.82 eVbond–1) and Zr–Te (–0.59 eVbond–1)
contacts. This is reflected by the fact that both bond
lengths, 262.98(2) and 288.98(2) pm for Zr–P and Zr–Te, lie
close to the sum of their covalent radii (255 and
282 pm[25,26]). Also, as already mentioned above, Zr–P
Figure 11. Temperature dependence of the corrected magnetic
susceptibility of a powder sample of Zr2PTe2.
These results are consistent with those reported by
Philipp et al.[20] for the related compound Ti2PTe2.
Quantum Chemical Calculations
The chemical bonding in Zr2PTe2 was investigated by the [262.98(2) pm] in Zr2PTe2 almost equals the distance found
linear muffin-tin orbital (LMTO) method. Figure 12 illus- in ZrP (263 pm; NaCl structure type[22]), and Zr–Te
trates the total DOS curve calculated for Zr2PTe2 and its [288.98(2) pm] in Zr2PTe2 is just slightly larger than the one
decomposition into contributions from the ionic constitu- observed in the similarly layered binary compound ZrTe2
ents. The non-vanishing DOS matches at the Fermi level (282 pm; CdI2 structure type[21]). The average ICOHP value
(EF) indicate metallic character, and thus confirm measured for Zr–Zr contacts amounts to –0.041 eVbond–1 suggesting
electrical resistivity and Pauli paramagnetism. The Fermi weak interactions, and this result is in line with the long
level lies ca. 0.5 eV below a small pseudogap in the DOS. average Zr–Zr distance of 374.92(5) pm, which is 0.72 Å
Bands located between –6 and –3 eV arise from valence or- longer than the sum of the metallic radius of Zr for CN 12
bitals of all three constituents. Bands beginning at ca. –3 eV (303 pm[25,26]). In fact, these weak Zr–Zr interactions can
and extending to the vicinity of the Fermi level involve sig- be easily understood because most of the electron density
nificant combinations of Zr and Te valence orbitals: on Zr is engaged in the heteroatomic Zr–X (X = P, Te)
Throughout this region of the total DOS curve, Te valence bonds. No significant interactions based on ICOHP values
orbitals dominate. At the Fermi level Zr and Te valence were found for P–P and Te–Te contacts in the structure.
Eur. J. Inorg. Chem. 2009, 3102–3110
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3107

P. Schmidt et al.
FULL PAPER
the starting materials as well as Zr2PTe2 are sensitive to air and
moisture, all reagents were handled in an argon-filled glove box
[c(O2, H2O) Ͻ 0.1 ppm, M. Braun LAB 130].
Syntheses were performed in evacuated closed silica tubes of small
volume (V ≈ 1 mL) to diminish sublimation of phosphorus and
tellurium and the thereby resulting changes in the composition of
the reaction mixture during the syntheses. Stoichiometric amounts
of the well-mixed starting materials were filled into ampoules that
had been freed from adherent moisture by heating under dynamic
vacuum prior to use. Using the elements as starting materials the
reaction requires heating of the elements from room temperature
up to 700 °C within 90 h, further heating to 800 °C within the next
48 h and keeping this temperature for 120 h. In the following 150 h
the reaction mixture must be cooled down to a temperature of
400 °C and quenched afterwards. In contrast to this quite complex
regime of temperatures the synthesis of Zr2PTe2 by reaction of ZrP
und ZrTe2 allows fast heating of the starting materials up to
850 °C, holding the reagents at this temperature for 120 h and cool-
ing them down to 400 °C with a cooling rate of 5 Kh–1 before
quenching. A third method of synthesis is a thermite-type reaction
of elementary zirconium and the tellurium oxide phosphate
Te8O10(PO4)4 leading to the products ZrO2 and Zr2PTe2. The tem-
perature regime is identically to the one used for reaction of the
Figure 13. COHP curves for (a) Zr–P, (b) Zr–Te and (c) Zr–Zr
contacts in Zr2PTe2 obtained from spin-polarised calculations. The
Fermi level has been set to zero.
Conclusions
The new phosphide telluride Zr2PTe2 was prepared by
different methods of synthesis. Its thermochemical proper-
ties were investigated and chemical vapour transport was
performed. On the basis of experimental data, thermo- binary starting materials.
dynamic modelling of thermal decomposition and chemical
Thermogravimetric Measurements: Performed under argon atmo-
vapour transport of Zr2PTe2, estimation of its thermo-
dynamic data was possible. Furthermore, it has been ascer-
tained that Zr2PTe2 crystallises in a layered structure and
exhibits delocalised electrons according to the boundary
formula [(Zr4+)2(P3–)(Te2–)2e–], which has been validated by
resistivity and magnetic measurements as well as quantum
mechanical calculations.
sphere with the use of a Netzsch STA 449 C apparatus situated in
an argon-filled glove box; the heating rate applied was 10 Kmin–1
in a temperature range of 25–1500 °C, same cooling rate down-
wards.
Investigations of the composition of the gas phase in the quater-
nary system Zr/P/Te/I were performed by using the mass spectrom-
eter Varian MAT 212 coupled to a Knudsen effusion cell.[30] The
sample was heated isothermally to 460, 520, 600, 700 and 800 °C.
Evaporated species were ionised by electron impact ionisation with
an energy of 70 eV and detected by a quadrupole mass spectrome-
ter with m/zՅ600.
Experimental Section
General: Starting materials used were red amorphous phosphorus
(Fluka, 99%, washed with NaOH according to ref.[27]), tellurium
(Acros, 99.8%), zirconium powder (freshly filed from a rod, Chem-
Pur, 99.8%), iodine (Merck, doubly sublimated over BaO2), tel-
lurium oxide phosphate (synthesised according to ref.[28]), ZrP (pre-
pared by heating of an equimolar mixture of Zr and P at 800 °C
for 7 d) and ZrTe2 (synthesised according to ref.[29]). As some of
Thermodynamic Modelling: The phase equilibria of the ternary sys-
tem Zr/P/Te as well as of the quaternary system Zr/P/Te/I were
studied by using Calphad methods based on the Eriksson–Gibbs
energy minimiser implemented in the computer program Trag-
min.[14] The solid state/gas phase equilibria in the ternary and qua-
ternary systems Zr/P/Te/(I) were calculated for different composi-
Table 2. Thermodynamic data of condensed species used for modelling of the system Zr/P/Te/I.
Species
∆H0298 / kJmol–1
∆S0298 / Jmol–1 K–1
Cp / Jmol–1 K–1[a]
Ref.
a
b
c
[31]
[31]
[31]
[31]
[31]
[31]
Zr(s)
P(s,red)
Te(s)
0
38.996
22.801
49.706
22.856
16.735
19.121
37.660
70.120
65.178
39.590
69.693
212.091
109.275
124.097
102.137
79.369
30.124
8.968
14.897
22.094
0
–0.712
0
0
–17.488
0
12.545
Te(l)
55.578
0
ZrO2(s)
TeO2(s)
ZrP(s)
ZrTe2(s)
Zr5Te4(s)
Zr2PTe2(s)
ZrI4(s)
ZrI2(s)
TeI4(s)
I2(s)
–1051.496
–272.220
–293.076
–294.102
–828.999
–607.086
–488.692
–277.815
–49.999
0
132.869
160.840
58.615
124.239
329.999
180.032
250.203
150.206
226.087
116.133
7.021
–1.423
–0.502
–0.712
0.134
3.161
–0.586
–3.098
–5.359
0
14.560
23.865
15.060
44.409
38.937
15.106
–6.586
185.354
81.630
this work
[33]
[34]
this work
[31]
[31]
[31]
[31]
0
[a] Cp = a + b(10–3)T + c(106)T–2.
3108
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 3102–3110

Chemistry and Physical Properties of Zr2PTe2
Table 3. Thermodynamic data of gaseous species used for modelling of the system Zr/P/Te/I.
Species
∆H0298 / kJmol–1
∆S0298 / Jmol–1 K–1
Cp / Jmol–1 K–1[a]
b
Ref.
a
c
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[31]
[32]
[32]
[31]
[31]
Zr(g)
Te2(g)
Te(g)
P4O6(g)
P4O10(g)
P4(g)
P2(g)
P(g)
I2(g)
I(g)
ZrI4(g)
ZrI3(g)
ZrI2(g)
ZrI(g)
TeI4(g)
TeI2(g)
PI3(g)
N2(g)
601.241
160.371
211.710
–2214.290
–2902.076
59.118
144.302
333.885
62.191
106.776
–355.246
–128.866
–85.277
402.921
61.953
181.339
262.152
182.699
345.745
406.697
279.980
218.128
163.201
260.159
180.782
446.556
397.788
344.787
275.810
443.716
333.206
374.372
191.609
22.994
34.646
19.414
216.355
292.830
81.839
36.295
20.670
37.254
20.394
107.969
83.104
58.176
37.384
90.0
3.052
6.615
1.842
8.665
19.192
0.678
0.800
0.172
0.779
0.402
0.075
0.025
0.017
0.929
0
3.638
–0.251
0.754
–6.795
–10.715
–13.440
–4.145
0
–0.498
–0.289
–3.010
–2.052
–0.842
–0.670
0
81.627
–17.991
0
54.962
82.843
30.417
0
0
0
–0.385
–2.378
2.546
[a] Cp = a + b(10–3)T + c(106)T–2.
[19]
¯
tions isothermally in the temperature range 800–1300 K with ∆T
= 100 K. The thermodynamic data of condensed phases used are
displayed in Table 2. The data sets of all gaseous species containing
the components Zr, P, Te and I were adopted from ref.[31,32]
(Table 3).
space group R3m (no. 166) by direct methods
and refined by
2
full-matrix least-squares on Fo (SHELX-97[37]). Details of data
collection and structure solution are gathered in Tables 4 and 5.
Table 4. Crystallographic data and details of structure determi-
nation for Zr2PTe2.
Estimation and Optimisation of Standard Data: First, the standard
enthalpy of formation of ZrP had to be estimated, which was done
by using the particular value of the analogous compound TiP
{∆H0298 [TiP(s)] ≈ –250 kJmol–1} as a starting point. Optimisation
of this value was accomplished by varying the standard enthalpy
of formation of ZrP until modelled coexistences of condensed
phases were consistent with the experimental determined ones and
reveals ∆H0298[ZrP(s)] ≈ –290 kJmol–1. As for all unknown standard
entropies of formation and functions of heat capacity presented in
this work, approximations of S0298 (ZrP) and Cp(T) (ZrP) were per-
formed according to the Neumann–Kopp rule as the stoichiometric
sum of the particular values of the elements.
Formula
Zr2PTe2
Crystal system
Space group
rhombohedral
¯
R3m (no. 166), hexagonal setting
Formula units per cell
Temperature / K
a / pm
3
293(2)
381.17(3)
c / pm
2918.9(3)
Cell volume / pm3
Calculated density / gcm–3
Crystal size / mm3
Measurement modii
367.27(2)106
6.356
0.2ϫ0.15ϫ0.01
0–180° (ω = 0°) and
0–90° (ω = 30°); ∆φ = 1°
2θՅ59.90°;–5 Յ h, kՅ5; –40 Յ l Յ 40
Measurement limits
µ / mm–1
The standard enthalpy of formation of zirconium phosphide tellu-
ride could be estimated on the basis of the results of the thermo-
gravimetric measurements that indicate the decomposition pressure
of Zr2PTe2 equals 1 bar at a temperature of about 1100 °C and
optimised according to experimental determined coexistences of
16.1 (Mo-Kα)
isotropic, empirical
0.0093(6)
176
4776
0.092; 0.042
10; 0
Extinction correction
Extinction parameter
Independent reflections
Measured reflections
Rint; Rσ
Parameters; restrictions
Residual e– density / e–pm–3
R1 (for all Fo)
condensed species. The obtained value is ∆H0298(Zr2PTe2)
=
–607 kJmol–1. A comparison of the optimised standard enthalpy
of formation of Zr2PTe2 with the particular value of the analogous
titanium compound (∆H0298 = –450 kJmol–1) shows the latter one
to be smaller, what is consistent with the lower thermal stability
of Ti2PTe2 that already decomposes at temperatures above 700 °C.
Anyway, all thermodynamic data used for modelling of the ternary
system Zr/P/Te taken from the literature as well as estimated data
are presented in Table 2.
1.99ϫ10–6 to –1.32ϫ10–6
0.027
0.058
1.88
2
wR2 (for all Fo )
GooF
Table 5. Wyckoff positions (W.p.), coordinates, and coefficients
Uij/pm2 of the anisotropic displacement parameters according to
exp[–2π2(h2a*2U11 + ... + 2hka*b*U12 + ...)] for Zr2PTe2; U13 = U23
= 0.
Single crystals of Zr2PTe2 suitable for structural investigations were
selected on the basis of precession photographs. Intensity data were
collected with an imaging plate diffractometer (Stoe IPDS-II) by
using graphite-monochromated Mo-Kα. (λ = 71.073 pm) radiation.
The microscopic crystal description was optimised by using sets of
symmetry equivalent reflections[35] and used afterwards in numeri-
cal absorption corrections.[36] The crystal structure was solved in
Atom W.p.
x
y
z
U11 = U22
U33
U12
P
Te
Zr
3a
6c
6c
0
0
0
60(20)
90(6)
78(6)
80(30) 29(8)
75(7) 45(3)
80(9) 39(4)
2/3 1/3 0.11376(2)
1/3 2/3 0.04943(3)
Eur. J. Inorg. Chem. 2009, 3102–3110
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3109

P. Schmidt et al.
FULL PAPER
[11] S. Jörgens, D. Johrendt, A. Mewis, Chem. Eur. J. 2003, 9, 2405.
[12] P. Schmidt, Habilitation, TU Dresden, 2007; http://nbn-resol-
ving.de/urn:nbn:de:bsz:14-ds-1200397971615-4054.
[13] H. Oppermann, O. Schneider, E. Wittig, Chem.-Tech. (Heidel-
berg) 1966, 18, 433.
[14] G. Krabbes, W. Bieger, K.-H. Sommer, T. Söhnel, U. Steiner,
GMIN Version 5.0, Package TRAGMIN for Calculation of
Thermodynamic Equilibrium, Dresden, 2007.
All sample handling for physical properties measurements was
done under Ar or He atmospheres. The electrical resistance of a
microcrystalline sample, which was cold compressed in a cylindrical
sapphire cell, was measured in the temperature range 4–320 K by
a DC method (van der Pauw). As a result of the nature of the
sample the estimated inaccuracy of the absolute resistivity is
Ϯ20%.
[15] M. Lenz, R. Gruehn, Chem. Rev. 1997, 97, 2967.
[16] H. Oppermann, W. Reichelt, G. Krabbes, E. Wolf, Krist. Tech.
1977, 12, 717.
Magnetisation measurements were performed between 1.8 and
400 K for selected magnetic field strengths with a SQUID magne-
tometer (MPMS XL-7, Quantum Design). About 100 mg of a
microcrystalline sample were contained in a precalibrated quartz
tube. Because the sample contained a small amount of ferromag-
netic impurities the susceptibility data were corrected by the
Honda–Owen method [extrapolation of high-field χ(H) data for
1/HǞ0 for all temperatures].
[17] D. Harker, Z. Kristallogr. 1934, 89, 175.
[18] K. Brandenburg, DIAMOND 3.1e, Crystal and Molecular
Structure Visualization, Crystal Impact GbR, Bonn, 2007.
[19] A. H. Daane, R. E. Rundle, H. G. Smith, F. H. Spedding, Acta
Crystallogr. 1954, 7, 532.
[20] F. Philipp, P. Schmidt, M. Ruck, W. Schnelle, J. Solid State
Chem. 2008, 181, 2859.
[21] T. Matkovic, P. Matkovic, Metalurgija 1992, 31, 107.
[22] A. R. Moodenbaugh, D. C. Johnston, R. Viswanathan, Mater.
Res. Bull. 1974, 9, 1671.
The quantum-mechanical calculations were carried out by using
the linear muffin-tin orbital (LMTO) method[38–40] in its tight-bind-
ing representation[41] through the LMTO-ASA 4.7 program.[42] The
chemical bonding was investigated by using the Crystal Orbital
Hamilton Population (COHP) concept,[43] which is an energy-re-
solved partitioning technique of the band-structure energy (sum
of the Kohn–Sham eigenvalues) in terms of atomic and bonding
contributions.[44] For convenience, the COHP plots are set up such
that bonding contributions are given to the right and antibonding
contributions are given to the left. The figures were generated with
the program wxDragon.[45]
[23] R. D. Shannon, C. T. Prewitt, Acta Crystallogr., Sect. B 1969,
25, 925.
[24] P. W. Selwood, Magnetochemistry, 2nd ed., Interscience Pub-
lishers, New York, 1956.
[25] R. D. Shannon, Acta Crystallogr., Sect. A 1976, 32, 751.
[26]
[27]
[28]
L. Pauling, B. Kamb, Proc. Natl. Acad. Sci. (USA) 1986, 83,
3569.
G. Brauer, Handbuch der Präparativen Anorganischen Chemie,
3rd ed., Enke Verlag, Stuttgart, 1975.
H. Mayer, G. Pupp, Z. Kristallogr. 1977, 145, 321.
Further details of the crystal-structure investigation may be ob-
tained from the Fachinformationszentrum Karlsruhe, 76344 Egg-
enstein-Leopoldshafen, Germany, on quoting the depository
number CSD-420650.
[29] H. Hahn, P. Ness, Z. Anorg. Allg. Chem. 1959, 302, 136–154.
[30] M. Binnewies, K. Rinke, H. Schäfer, Z. Anorg. Allg. Chem.
1973, 50, 395.
[31] O. Knacke, O. Kubaschevski, K. Hesselmann, Thermochemical
Properties of Inorganic Substances, 2nd ed., Springer, Berlin,
1991.
Acknowledgments
[32] H. Oppermann, G. Stöver, E. Wolf, Z. Anorg. Allg. Chem. 1976,
419, 200.
The authors thank Mrs. S. Müller, Max Planck Institute for Chemi-
cal Physics of Solids, Dresden, for performing the thermogravime-
tric measurements. Financial support by the Deutsche Forschungs-
gemeinschaft (DFG) is gratefully acknowledged.
[33] G. Chattopadhyay, J. M. Juneja, J. Nucl. Mater. 1993, 202, 10.
[34] R. de Boer, E. H. P. Cordfunke, J. Chem. Thermodyn. 1998, 30,
87.
[35] X-Shape 1.06, Program for Crystal Optimization for Numerical
Absorption Correction, STOE & Cie GmbH, Darmstadt, 1999.
[36] X-Red-32 1.01, Program for Data Reduction, STOE & Cie
GmbH, Darmstadt, 2001.
[37] G. M. Sheldrick, SHELX-97, Program for Crystal Structure
Determination, University of Göttingen, 1997.
[38] O. K. Andersen, Phys. Rev. B 1975, 12, 3060.
[39] H. Skriver, The LMTO Method, Springer, Berlin, 1984.
[40] O. K. Andersen, The Electronic Structure of Complex Systems,
Plenum, New York, 1984.
[41] O. K. Andersen, O. Jepsen, Phys. Rev. Lett. 1984, 53, 2571.
[42] G. Krier, O. Jepsen, A. Burkhardt, O. K. Andersen, The TB-
LMTO-ASA Program, version 4.7, Max-Planck-Institut für
Festkörperforschung, Stuttgart, Germany.
[1] C. Stoltz, K. Ramesha, S. A. Sirchio, Z. S. Gönen, B. W. Eich-
horn, L. Salamanca-Riba, J. Gopalakrishnan, J. Am. Chem.
Soc. 2003, 125, 4285.
[2] A. Schlechte, R. Niewa, M. Schmidt, G. Auffermann, Yu.
Prots, W. Schnelle, D. Gnida, T. Cichorek, F. Steglich, R.
Kniep, Sci. Technol. Adv. Mater. 2007, 8, 341.
[3] C. Wang, C. Eylem, T. Hughbanks, Inorg. Chem. 1998, 37, 390.
[4] H. Kleinke, B. Harbrecht, Z. Anorg. Allg. Chem. 1999, 625,
1873.
[5] N. Soheilnia, K. M. Kleinke, H. Kleinke, Chem. Mater. 2007,
19, 1482.
[6] A. Zygmunt, A. Murasik, S. Ligenza, J. Leciejewicz, Phys.
Status Solidi A 1974, 22, 75.
[7] W. B. Pearson, Z. Kristallogr. 1985, 171, 23.
[8] F. Philipp, P. Schmidt, E. Milke, M. Binnewies, S. Hoffmann,
J. Solid State Chem. 2008, 181, 758.
[9] G. Kliche, Z. Naturforsch., Teil B 1986, 41, 130.
[10] H. D. Lutz, Th. Schmidt, G. Wäschenbach, Z. Anorg. Allg.
Chem. 1988, 562, 7.
[43] R. Dronskowski, P. E. Blöchl, J. Phys. Chem. 1993, 97, 8617.
[44] R. Dronskowski, Computational Chemistry of Solid State Ma-
terials, Wiley-VCH, Weinheim, 2005.
[45] B. Eck, wxDragon 1.5.1, Aachen, 1994–2009; available at:
http://www.ssc.rwth-aachen.de.
Received: April 16, 2009
Published Online: June 17, 2009
3110
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 3102–3110