Cluster harvesting in the WBr6-P System

Article abstract of DOI:10.1021/ic502333u

A combined thermal scanning-X-ray diffraction (XRD) approach was performed for the WBr₆-P system to detect and analyze phases in this system, including metal-rich phases generated with increasing amounts of elemental (red) phosphorus under partial PBr₃ release. Phases were characterized by powder XRD. A black crystalline powder of W₄(PBr)Br₁₀ was obtained after reduction of WBr₆ with elemental phosphorus at 450 °C. The crystal structure of the new compound was found to be isotypic with the structure of W₄(PCl)Cl₁₀ on the basis of powder XRD data. The structure of W₄(PBr)Br₁₀ is represented by a cyclobutadiene-like tetranuclear tungsten cluster interconnected into a layered (W₄(μ₄-PBr)Br₆ⁱ)Br_8/2^a-a arrangement via outer bromide ligands. The μ₄-capping bromophosphinidene ligand was verified by solid-state magic-angle spinning ³¹P NMR spectroscopy.

Full text of DOI:10.1021/ic502333u

Article
pubs.acs.org/IC
Cluster Harvesting in the WBr6−P System
Markus Strobele, Klaus Eichele, and Hans-Jurgen Meyer*
̈
̈
Section of Solid State and Theoretical Inorganic Chemistry, Institute of Inorganic Chemistry, Eberhard Karls University Tubingen,
̈
Auf der Morgenstelle 18, D-72076 Tubingen, Germany
̈
S
* Supporting Information
ABSTRACT: A combined thermal scanning−X-ray diffraction (XRD)
approach was performed for the WBr6−P system to detect and analyze phases
in this system, including metal-rich phases generated with increasing amounts
of elemental (red) phosphorus under partial PBr3 release. Phases were
characterized by powder XRD. A black crystalline powder of W4(PBr)Br10 was
obtained after reduction of WBr6 with elemental phosphorus at 450 °C. The
crystal structure of the new compound was found to be isotypic with the
structure of W4(PCl)Cl10 on the basis of powder XRD data. The structure of
W4(PBr)Br10 is represented by a cyclobutadiene-like tetranuclear tungsten
i
a‑a
cluster interconnected into a layered (W4(μ4-PBr)Br6 )Br8/2 arrangement via
outer bromide ligands. The μ4-capping bromophosphinidene ligand was
verified by solid-state magic-angle spinning 31P NMR spectroscopy.
contribution, was explored by the thermal scanning method.
INTRODUCTION
■
Phases appearing during this study were characterized by
powder X-ray diffraction (XRD), and the presence of
phosphorus in one compound was verified by magic-angle-
spinning (MAS) 31P NMR spectroscopy.
A thermal scanning approach has been developed to detect
phases with limited thermal stability that are usually hard to
find by straightforward solid-state reactions. This approach
involves the in situ detection of thermal effects caused by the
formation of compounds by means of differential thermal
analysis or differential scanning calorimetry (DSC) during
heating of a reaction mixture. At the same time, a concentration
gradient is implemented in the reaction by using an excess of
one reaction partner, which in this case is a reducing agent, to
generate metal-rich compounds.
RESULTS AND DISCUSSION
■
The P−W−Br system was investigated for a mixture of WBr6
and red phosphorus, charged into a modified gold-plated DSC
sample holder. The amount of reacting agents was balanced
according to reaction (1) to allow the successive reduction of
tungsten from WVI to WII sequentially with oxidation stages in
between.
This thermal scanning approach has been successfully
employed to detect numerous compounds in the M−W−Cl
system with M = Fe,1 Co,2 and Mn,3 and phases in the Sb−W−
Br system.1,4
6WBr6 + 8P → W Br + 8PBr
(1)
6
12
3
Recently, the reduction of WCl6 with phosphorus in a fused-
silica tube yielded the new compounds W6PCl17 and W4(PCl)-
Cl10,5 of which the first compound represents a new type of
hexanuclear phosphorus-centered W6 cluster, which can be
considered as a structural intermediate between the (empty)
W6Cl18,6 and the carbon-centered W6CCl18-type cluster.7 The
discovery of W6PCl17 was indeed guided by serendipity (and
persistence) because its thermal stability range (under the
reaction conditions used) is approximately between 410 °C
(formation) and 440 °C (decomposition). This example clearly
emphasizes that phases that exist in a similarly small
temperature window can be regarded as undiscovered in
most cases of classical solid-state synthesis. The thermal
scanning approach was not applied to the P−W−Cl system
because the reactivity and pressure of the coproduced PCl3
offer a serious stability challenge to most container materials.
Meanwhile, we have introduced gold-plated steel as a robust
and pressure-resistant container material. Hence, the inves-
tigation of the P−W−Br system, which is reported in this
A total of five thermal effects were obtained in the DSC
measurement during heating of the mixture from room
temperature to 550 °C at a rate of 2 K/min (Figure 1). The
reaction container remained intact in spite of the significant
pressure caused by the increasing formation of PBr3. In the
following process, separate reactions were performed and
assigned to each single thermal effect. Reaction products were
inspected by powder XRD, with the corresponding powder
XRD patterns displayed in Figure 2. The first and second
thermal effects in the DSC measurement were assigned to the
formation of WBr5 and WBr4. No known compound could be
assigned to the endothermic effect near 350 °C. The
exothermic effect above 350 °C was due to the new
W4(PBr)Br10 compound, whose structure was refined on the
Special Issue: To Honor the Memory of Prof. John D. Corbett
Received: September 29, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic502333u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a temperature interval of about 30 °C, although the structure
remains unknown up to now.
The crystal structure of W4(PBr)Br10, as refined on the basis
of powder XRD data, appears to be isotypic to the structure of
W4(PCl)Cl10. The basic cluster features a Jahn−Teller-distorted
rectangular W4 core following the motif of the butadiene
structure (Figure 3). The distances between tungsten atoms
Figure 1. DSC of the successive reduction of WBr6 with elemental
phosphorus powder with increasing temperature (corrected for
background).
i
Figure 3. Building block of (W4(μ4-PBr)Br6 )Br8/2a‑a with atom labels.
Bra‑a ligands, bridging between adjacent clusters, are indicated with
bond sticks.
involve short contacts, dW1−W1 = 2.649(3) and dW2−W2
=
2.665(4) Å, and long contacts of 2 × 2.855(4) Å for the W1−
W2 bonds. These distances are slightly (on average 0.038 and
0.046 Å) longer than those in the corresponding chloride
compound. Nearest intercluster contacts within layers of
(W4(μ4-PBr)Br6 )Br8/2 are established via bridging Bra‑a
ligands and involve W−W distances of 4.031(4) and
4.048(4) Å. The average W−Br distance with bridging Br5a‑a
and Br7a‑a ligands is 2.649 Å, and that with terminal Bri is 2.575
Å.
i
a‑a
Figure 2. Powder XRD patterns of compounds obtained during the
reduction of WBr6 with elemental phosphorus powder at certain
reaction temperatures (given in the figure).
Following this connectivity pattern, the rectangular clusters
are interconnected by bridging bromide ligands into a 44 layer
structure displayed in Figure 4. Alternating layers are linked by
van der Waals attractions, whereupon empty rectangular spaces
within one layer are stuffed by bromophosphinidene ligands of
adjacent layers.
basis of the powder XRD pattern. Finally, the known
compound W6Br12 was formed, beginning at about 450 °C
(onset).
Compared to the P−W−Cl system, the endothermic effect
near 350 °C was expected to relate to formation of the
unknown W6PBr17. However, the XRD pattern obtained from a
reaction with this composition (Figure 2) could not be related
to the corresponding XRD pattern of W6PCl17 or to any other
known tungsten bromide compound.
A comparative DSC study performed in the P−W−Cl system
revealed a pattern similar to that with the formation of WCl5
and WCl4 in the lower-temperature regime, below 200 °C. At
higher temperatures, two endothermic and two exothermic
effects were obtained, most likely related to the formation of
PCl3 adducts of W6PCl17 (adduct onset at 375 °C; W6PCl17
onset at 410 °C) and W4(PCl)Cl10 (adduct onset at 450 °C;
W4(PCl)Cl10 onset at 480 °C). A behavior similar to that with
adduct formation has been seen in the course of the reduction
of WBr6 with elemental antimony (e.g., W4Br10·2SbBr3 before
W4Br10 formation).8
The most remarkable feature of W4(PBr)Br10 is the presence
of the μ4-capping bromophosphinidene ligand, having W−P
distances of 2.448(8) and 2.466(8) Å and a P−Br distance of
2.09(1) Å, which is shorter than the reported P−Br distance in
PBr3 [2.212(3)−2.216(4) Å]. This finding is in agreement with
the result of the NMR measurement.
In the MAS 31P NMR spectrum of a powder sample of
W4(PBr)Br10 (Figure 5), direct dipolar and indirect spin−spin
3
coupling interactions with the spin /2 isotopes 79Br and 81Br
result in characteristic nonsymmetric multiplets9,10 centered at
an isotropic chemical shift of 345.7 ppm. The observed splitting
can be simulated11 using the following parameters for the
79Br−31P isotopomer: a direct dipole−dipole coupling constant
of 1320 Hz (close to the value calculated from the P−Br
distance, 1340 Hz), an indirect spin−spin coupling constant of
−500 Hz, and a 79Br nuclear quadrupole coupling constant of
250 MHz. The asymmetry parameter of the electric field
With our thermal scanning approach, we demonstrate the
detection of an endothermic compound that only exists within
B
dx.doi.org/10.1021/ic502333u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(glovebox) into a gold-plated steel container (V = 100 μL; Fa. Bachler
̈
Feintech, CH) and sealed with a gold plate (99.99% gold). The loaded
sample holder and a reference container were inserted into a
differential thermal analyzer (STA 204 F1 Phoenix, Fa. Netzsch)
and heated at a rate of 2 °C/min, while the thermal effects of the
progressive reduction steps were monitored.
Preparation of Single Compounds. Compounds were prepared
from WBr6 (300 mg, 0.452 mmol) and elemental red phosphorus
(18.7 mg, 0.6 mmol; Alfa Aesar, 99%), which were mixed under an
argon atmosphere (glovebox), sealed in an evacuated silica tube, and
then heated in a tube furnace (heating and cooling rate of 2 K/min).
The temperature and duration of the reactions were optimized to
obtain a high X-ray purity for each compound, leading to the following
conditions: WBr5, 5 min at 100 °C, black powder; WBr4, 30 min at
175 °C, dark-brown powder; unknown compound (W6PBr17), 9 h at
335 °C, black powder; W4(PBr)Br10, 72 h at 450 °C, dark-green
powder; W6Br12, 72 h at 500 °C, dark powder.
An excess of elemental phosphorus, which is X-ray amorphous, was
present in all reaction products, except for the product targeting the
synthesis of W6Br12. Coproduced PBr3 was separated by sublimation.
Powder XRD: W4(PBr)Br10, Mr = 1645.3 g/mol, monoclinic space
group C2/m (No. 12), a = 14.0933(3) Å, b = 10.3228(2) Å, c =
11.8169(2) Å, β = 93.906(3)°, V = 1715.17(6) Å3, Z = 4, ρcalcd
=
Figure 4. Projection of one layer of the crystal structure of (W4(μ4-
6.37131 g/cm3, θ range = 2.5−50°, Cu Kα1 (λ = 1.54060 Å), T =
293(2) K, 1020 measured data (STOE StadIP), structure refinement
by global refinement (FullProf Suite), refined parameters 47, final R
indices of Rp = 0.042, Rwp = 0.0565, RBragg = 0.0409, and χ2 = 1.345.
Solid-State NMR. Solid-state MAS 31P NMR spectra were
acquired at 4.7 T using a Bruker DSX-200 super-wide-bore
spectrometer equipped with a double-bearing broad-band cross-
polarization/MAS probe head, operating at 81.01 MHz for 31P
NMR. A powdered sample was packed into a 4 mm (o.d.) zirconia
rotor sealed with a Kel-F cap. Spectra were acquired after single-pulse
excitation with a spinning frequency of 4 kHz, a recycle delay of 30 s,
and 2406 scans. 31P chemical shifts are reported with respect to
external 85% aqueous phosphoric acid. Spectral simulations were
carried out using WSolids1.11
i
a‑a
PBr)Br6 )Br8/2
.
CONCLUSION
■
A combined thermal scanning−XRD approach has shown
several phases in the P−W−Br system and their stability ranges
under given conditions. On this basis, the new compound
W4(PBr)Br10 was separately prepared and structurally charac-
terized. The crystal structure is isotypic with W4(PCl)Cl10 and
comprises a bromophosphinidene-capped tungsten cluster, as
refined from powder XRD data and confirmed by MAS 31P
NMR spectra. The compound appears quite stable in air and
water, showing an unchanged powder XRD pattern after
exposure to air for 1 week.
Thermal scanning−XRD has shown once more its capability
to detect and subsequently prepare thermally labile cluster
moieties with no more than six cluster atoms (the missing
children in cluster chemistry). An expansion of this methodology
to other metal halides that are known from cluster compounds
is likely.
Figure 5. MAS 31P NMR spectrum of W4(PBr)Br10 including spinning
sidebands (*, top) and the calculated spectrum including coupling
interactions of 31P with 79/81Br (bottom).
gradient about bromine has been assumed to be zero. The
corresponding values of the 81Br−31P isotopomer have been
calculated by taking into account the magnetogyric ratios,
γ(81Br)/γ(79Br) = 1.077, or the ratios of the nuclear quadrupole
moments, Q(81Br)/Q(79Br) = 0.8355. In summary, the
multiplet structure observed in the MAS 31P NMR spectrum
corroborates the presence of one bromine atom bonded to
phosphorus at a relatively short distance. Because of the great
line widths, ca. 1 kHz, in the MAS 31P NMR spectrum, satellites
due to coupling with 183W (spin −1/2, n.a. 14.3%) cannot be
resolved.
ASSOCIATED CONTENT
■
S
* Supporting Information
X-ray crystallographic information for W4(PBr)Br10 in CIF
format. This material is available free of charge via the Internet
at http://pubs.acs.org.
EXPERIMENTAL SECTION
DSC. A mixture of WBr6 and elemental red phosphorus (Alfa
Aesar, 99%) was combined in a molar ratio corresponding to the
composition W6Br12 (plus the corresponding PBr3) to allow successive
reduction of tungsten from oxidation state 6+ to 2+. The starting
mixture (39.6 mg) was loaded under a dry argon atmosphere
■
8
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: juergen.meyer@uni-tuebingen.de.
C
dx.doi.org/10.1021/ic502333u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Strobele, M.; Mos, A.; Meyer, H.-J. Inorg. Chem. 2013, 52, 6951−
̈
6956.
(2) Mos, A.; Strobele, M.; Meyer, H.-J. J. Cluster Chem. 2014.
̈
(3) Mos, A.; Strobele, M.; Meyer, H.-J. In preparation.
̈
(4) Strobele, M.; Meyer, H.-J. Z. Anorg. Allg. Chem. 2012, 638 (6),
̈
945−949.
(5) Strobele, M.; Eichele, K.; Meyer, H.-J. Eur. J. Inorg. Chem. 2011,
̈
26, 4063−4068.
(6) Siepmann, R.; von Schnering, H. G.; Schafer, H. Angew. Chem.
1967, 79, 650. Nagele, A.; Glaser, J.; Meyer, H.-J. Z. Anorg. Allg. Chem.
̈
̈
2001, 627, 244−249.
(7) Zheng, Y.-Q.; von Schnering, H. G.; Chang, J. H.; Grin, Y.;
Engelhardt, G.; Heckmann, G. Z. Anorg. Allg. Chem. 2003, 629, 1256−
1264.
(8) Strobele, M.; Meyer, H.-J. Russ. J. Coord. Chem. 2012, 38 (3),
̈
178−182.
(9) Rabis, A.; Grimmer, A.-R.; Thomas, B.; Brendler, E.; Beck, S.;
Meisel, M. Solid State Nucl. Magn. Reson. 2005, 28, 57−63.
(10) Thomas, B.; Paasch, S.; Steuernagel, S.; Eichele, K. Solid State
Nucl. Magn. Reson. 2001, 20, 108−117.
(11) Eichele, K. WSolids1, version 1.20.21; Universitat Tubingen:
̈
̈
Tubingen, Germany, 2013.
̈
D
dx.doi.org/10.1021/ic502333u | Inorg. Chem. XXXX, XXX, XXX−XXX