ARTICLE IN PRESS
Y. Wu, W. Bensch / Journal of Solid State Chemistry 180 (2007) 2166–2174
2167
s
K Ta S are isostructural adopting the orthorhombic
4
heated at 450 1C for 6 days in a glass ampoule (Duran ).
After cooling with 2 1C/h, transparent red platelets were
obtained in a yield of about 90%. The compound is stable
in dry air for several weeks. An EDX analysis of single
crystals indicated the presence of all four elements (K, Nb,
Ta, S) in an atomic ratio of 6:1:3:22.
2 11
space group Pca2 . Therefore, it should be possible to
1
replace Nb by Ta (or vice versa) yielding the orthorhombic
structure type. Interestingly, there is a triclinic modification
of K Ta S [14], which was not observed for Nb. Another
4
2 11
interesting finding is that only one structural modification
was reported for K Nb S (space group C2/c), whereas
K Nb2.97Ta1.03S25 (3): The procedure for the synthesis of
6
6
4
22
K Ta S is dimorphic and crystallizes in space groups
4 22
3 is the same as that for 2 except that the molar ratio was
changed to 3:3:1:14. The melt contained red polyhedral
crystals in a 90% yield. The crystals are stable in dry air for
several weeks. The EDX analysis of selected single crystals
indicated the presence of all four elements (K, Nb, Ta, S) in
an atomic ratio of 6:3:1:25.
6
C2/c and P2 /c under different reaction conditions [17]. A
1
similar observation was made for A M S . The compound
6
4 25
Rb Ta S crystallizes in C2/c whereas space group P2 /n
1
6
4
25
was adopted by K Nb S . These examples demonstrate
6
4 25
the different structural behavior of Nb and Ta sulfides
mainly prepared in polysulfide fluxes. After the successive
substitution on the Nb sites (i.e. K Nb0.96Ta1.04S11 [40]), we
K Cu Nb0.98Ta1.02S8 (4) and KCu Nb0.53Ta0.47S4 (5):
3
3
2
Crystals were prepared by reacting a mixture of K S , Cu,
2 3
4
started to investigate the effect of partial substitution based
on K TaS , K Nb S , K Nb S , K Cu Nb S , and
KCu NbS structure types. In this paper we report the
Nb, Ta, S with a molar ratio of 2:4:1:1:7. After being sealed
ꢀ
3
in a silica tube under vacuum (ꢁ2 ꢂ 10 mbar), the
starting material was heated to 600 1C, held there for 6
days, and cooled down to 100 1C at a rate of 2 1C/h,
followed by cooling to room temperature in 4 h. After
removing excess K S flux with DMF followed by washing
3
4
6
4
22
6
4
25
3
3
2 8
2
4
syntheses, crystal structures, and characterizations of
K V0.32Ta0.68S , K Nb1.07Ta2.93S , K Nb2.97Ta1.03S ,
K Cu Nb0.98Ta1.02S , and KCu Nb0.53Ta0.47S .
3
4
6
22
6
25
3
3
8
2
4
2 x
with acetone, transparent brown needles (4: 40%) and
transparent green needles (5: 40%) were obtained. Both
compounds are stable in dry air for several months.
The EDX analyses of single crystals indicated the
presence of all five elements (K, Cu, Nb, Ta, S) in an
atomic ratio of 3:3:1:1:8 for compound 4 and 2:4:1:1:8 for
compound 5.
2
. Experimental
2
.1. Reagents
The following reagents were used as obtained unless
noted: (i) K metal, 99.95%, ABCR GmbH & Co.KG;
ii) S, 99.99%, Heraeus; (iii) Nb powder, 99.8%, 325 mesh,
(
Alfa Aesar; (iv) Ta powder, 99.98%, 325 mesh, ABCR
GmbH & Co.KG; (v) Cu powder, 99.5%, 200 mesh, Alfa
Aesar; (vi) V powder, 99.7%, 325 mesh, Heraeus.
2.3. X-ray crystallography
All single crystal X-ray investigations were performed
using an imaging plate diffraction system (IPDS) (MoKa-
˚
2
.2. Synthesis
radiation; l ¼ 0.71073 A) equipped with a low-temperature
device from Oxford Cryosystem. The crystals were
mounted on top of glass fibers and bathed in cold nitrogen
stream during data collection. The raw intensities were
treated in the normal way applying Lorentz, polarization,
and numerical absorption corrections. All structures
The compound K S was prepared from the reaction of
2
3
stoichiometric amounts of elemental K and S powder in
liquid ammonia under an argon atmosphere.
K V0.32Ta0.68S (1): K S , V, Ta, and S in a 4/2/1/8
3
4
2 3
molar ratio were thoroughly mixed in a nitrogen-filled
glove box. The mixture was then loaded into a glass
were solved with direct methods using SHELXS-97 and
2
refined against F using SHELXL-97 of Shelxtl program
o
s
ampoule (Duran ), which was subsequently evacuated
package [41].
On the basis of the systematic absences the space groups
Pnma for 1, P2 /c for 2, P2 /n for 3, C2/c for 4, and Ama2
ꢀ
3
(
ꢁ2 ꢂ 10 mbar) and flame sealed. The ampoule was
heated from 25 to 500 1C in 24 h and kept at this
temperature for 6 days, cooled to 100 1C at a rate of
1
1
for 5 were chosen. After successful assignments of the high
electron density peaks as Ta, (Cu), K, and S atoms, the
displacement parameters and occupancy on each atomic
site were examined. For all compounds the S atom sites are
fully occupied with reasonable displacement parameters.
The refined occupancies of heavy metal sites, initially
assigned to Ta, were significantly low indicating that lighter
atoms V or Nb are also involved in this site. Therefore,
disorder models were applied by introducing V or Nb
atoms so that the sum of the occupancies were set equal to
full occupancy to maintain charge balance. The other sites
were found suitable for K atoms and/or Cu atoms based
on the observed electron densities. After successive
2
1C/h followed by cooling to room temperature in 4 h.
The resulting black melt was washed with DMF and
acetone and the product was dried in vacuum. The product
consists of black platelet crystals (yield ꢁ50%). The
crystals are stable in dry air for several weeks, but
decompose slowly under humid conditions. An EDX
analysis of single crystals indicated the presence of all four
elements (K, V, Ta, S) in an approximate atomic ratio of
9
:1:2:12.
K Nb1.07Ta2.93S22 (2): The preparation and isolation
6
process of 2 were similar to those of 1. However, the
mixture of K S , Nb, Ta, S (molar ratio: 3:1:3:14) was
2
3