Reactions in molten alkalimetal polychalcogenides: What happens in the melt? A study of the reactions in the system K-Nb-S using differential scanning calorimetry, infrared spectroscopy, and X-ray powder diffraction

Article abstract of DOI:10.1515/znb-2002-1207

The reactions of potassium polysulfides with elemental Nb were investigated with different analytical techniques. The amount of the polysulfide applied has no influence onto product formation, i. e. the ratio K₂S_x: Nb is not important. The length of the polsysulfide chain, i. e. the value of x in K₂S_x determines what product is formed. In sulfur-poor melts, K₃NbS₄ is observed. Increasing x to 5-6, K₄Nb₂S₁₁ is formed with a structure containing S₂^2- anions. Finally, applying a melt with x > 6, K₆Nb₄S₂₅ is found as the product with a crystal structure containing the S₅^2- polysulfide anion. When K₂S_x (x < 5) is heated with sulfur in the first step the pentasulfide K₂S₅ is formed. Immediately after melting of K₂S₅ a reaction with elemental Nb occurs. The results of FT-IR and X-ray investigations have demonstrated that after oxidation the anion [Nb₂S₁₁]^4- is formed relatively fast, and after a short time crystalline K₄Nb₂S₁₁ can be detected. After 24 h the reaction is complete.

Full text of DOI:10.1515/znb-2002-1207

Reactions in Molten Alkalimetal Polychalcogenides: What Happens in
the Melt? A Study of the Reactions in the System K-Nb-S Using Differential
Scanning Calorimetry, Infrared Spectroscopy, and X-Ray Powder Diffraction
Peter Du¨richena and Wolfgang Benschb
a
Infineon Technologies, Dresden, Ko¨nigsbruckerstr. 180, D-01099 Dresden, Germany
b Institut fu¨r Anorganische Chemie, Christian-Albrechts-Universita¨t Kiel,
Olshausenstrasse 40, D-24098 Kiel, Germany
Reprint requests to Prof. Wolfgang Bensch. Fax: +49(0)431 880 1520.
E-mail:wbensch@ac.uni-kiel.de
Dedicated to Professor Albrecht Mewis on the occasion of his 60th birthday
Z. Naturforsch. 57 b, 1382–1386 (2002); received August 14, 2002
Alkalimetal Niobium Polychalcogenides, Infrared Spectroscopy,
Differential Scanning Calorimetry
The reactions of potassium polysulfides with elemental Nb were investigated with different
analytical techniques. The amount of the polysulfide applied has no influence onto product
formation, i. e. the ratio K2Sx: Nb is not important. The length of the polsysulfide chain, i. e.
the value of x in K2Sx determines what product is formed. In sulfur-poor melts, K3NbS4 is
2
observed. Increasing x to 5 - 6, K4Nb2S11 is formed with a structure containing S2 anions.
Finally, applying a melt with x > 6, K6Nb4S25 is found as the product with a crystal structure
2
containing the S5 polysulfide anion. When K2Sx (x < 5) is heated with sulfur in the first step
the pentasulfide K2S5 is formed. Immediately after melting of K2S5 a reaction with elemental
Nb occurs. The results of FT-IR and X-ray investigations have demonstrated that after oxidation
the anion [Nb2S11]4 is formed relatively fast, and after a short time crystalline K4Nb2S11 can
be detected. After 24 h the reaction is complete.
Introduction
gated with differential scanning calorimetry (DSC)
and X-ray powder diffractometry. The reaction of
elemental Bi with Cs2Sx melts (ratio: 1:4.5; x = 2.5
- 3.0) in the temperature range of 250 C to 300 C
was found to yield exclusively the modification.
Changing the Bi to Cs2Sx ratio to 1:4, increasing
x to 3.0 - 3.7 and using a temperature of 350 C
lead to the formation of -CsBiS2. The results have
demonstrated that for each modification a distinct
temperature and composition range is required. The
knowledge of the influence of the different synthesis
parameters onto product formation is a prerequisite
for a more rational synthesis strategy. Reactions be-
tween Bi and Li2Sx resp. Na2Sx (x = 2.2 - 5) in the
temperature range from 290 to 345 C always yield
LiBiS2 and NaBiS2. Of special interest is also the
observation made by the authors that reactions per-
formed with commercially available Cs2S result in
the formation of Bi2S3 whereas CsBiS2 is formed
if freshly prepared Cs2S is used. It was argued that
the variations in reactivity are due to the different
particle sizes of Cs2S.
During the last decade many new multinary
chalcogen compounds were prepared applying the
low melting alkali metal polychalcogenides. Com-
prehensive collections of the compounds are re-
ported in several review articles [1 - 8].
We focussed our syntheses on group 4 and 5
metals with the aim to prepare low-dimensional
compounds exhibiting interesting physical proper-
ties. In the K-Nb-S system we synthesised several
new compounds with compositions K4Nb2S11 [9],
K4Nb2S14 [10], K6Nb4S22 [11], and K6Nb4S25 [12].
Until now the synthetic work can still be de-
scribed as a ’trial and error’ endeavour and only
very few rules of thumb are at hand allowing
the prediction of the composition of the final
product.
To the best of our knowledge there is only
one contribution in which the synthesis of com-
pounds was systematically investigated using dif-
ferent educts and synthesis conditions [13]. The
formation of -CsBiS2 and -CsBiS2 was investi-
c
¨
¨
0932–0776/02/1200–1382 $ 06.00
2002 Verlag der Zeitschrift fur Naturforschung, Tubingen www.znaturforsch.com
K

P. Du¨richen – W. Bensch · Reactions in Molten Alkalimetal Polychalcogenides
1383
Table 1. Reactions of Nb with K2S3 and different amounts Table 2. Reactions of Nb with K2Sx, x = 1, 3, 5 and S.
of S.
No. Mixture Ratio
K2Sx Product Temperature
No.
Mixture
Ratio Product
Temperature
5
6
7
K2S:S:Nb 2:8:1 K2S5 K4Nb2S11 300 - 400 C
K2S3:S:Nb 2:4:1 K2S5 K4Nb2S11 300 - 400 C
K2S5:S:Nb 2:0:1 K2S5 K4Nb2S11 300 - 400 C
1
2
3
4
K2S3 : S : Nb 2 : 0 : 1 K3NbS4
K2S3 : S : Nb 2 : 2 : 1 K3NbS4
K2S3 : S : Nb 2 : 4 : 1 K4Nb2S11
K2S3 : S : Nb 2 : 6 : 1 K4Nb2S11
350 C
350 C
350 C
350 C
Here we report the results of a systematic study
of the processes occurring in the melts of the K-Nb-
S system. The system was chosen because the five
well characterised compounds K3NbS4, K4Nb2S11,
K4Nb2S14, K6Nb4S22, and K6Nb4S25 are formed
depending on the actual preparation conditions.
Experimental Section
The potassium polysulfides were prepared by the
reaction of stoichiometric amounts of the elements
(K 99.95%, Strem) in liquid ammonia under argon at-
mosphere. Elemental Nb (99.97 %, Fluka) was used as
supplied. Differential scanning calorimetry (DSC) was
conducted on a Netzsch device. All measurements were
performed in evacuated and sealed glass ampoules using
heating and cooling rates of 3 K min 1. The composi-
tions of all reaction products were determined by elemen-
tal analysis and X-ray powder diffractometry. The FT-IR
spectra were recorded on an ISF-66 device (Bruker) in the
Fig. 1. Reaction products in the system K/Nb/S for vary-
ing compositions of the melt (T = 350 C). a: K6Nb4S25
is the main phase; by-product: K4Nb2S14. b: K4Nb2S11
and K3NbS4 c: K3NbS4. d in the melt solidified sulfur is
present. Symbols: squares: K3NbS4; circles: K4Nb2S11;
triangles: K6Nb4S25. The lines join melts with identical
composition.
1
1
range from 80 to 550 cm with a resolution of 2 cm
.
with a fixed amount of K2S3 and Nb and a vari-
able amount of S (Table 1). The composition of
the products was investigated with X-ray powder
diffractometry.
The samples were prepared as polyethylene pellets. X-
ray powder diffractometry experiments were performed
with a D5000 diffractometer (Siemens AG) in reflection
geometry and with the STADI P diffractometer (STOE)
Obviously, less S rich melts like K2S3 or K2S4
lead to the formation of K3NbS4 which contains
no polysulfide anions in the structure, and the most
S rich reaction mixtures favour the formation of
operating in transmission geometry (CuK radiation,
1.54056 A).
=
˚
Results and Discussion
2
K4Nb2S11 which contains several S2 anions in the
structure. The same results were obtained at 300 C
and 400 C. Therefore, the composition of the melt
determines the final product within a temperature
range of 300 C to 400 C.
The next set of experiments was performed with
different K2Sx (x = 1, 3, 5) chalcogenides (Table 2).
For reactions between polychalcogenides and
certain very reactive metals it was assumed that the
metal reacts with the original starting material A2Qx
(A = alkali metal, Q = S, Se, Te) before a polychalco-
genide flux containing different polychalcogenides
has formed [4]. Therefore, it cannot be excluded
that different products are formed when a mixture
Within the temperature range 300 - 400 C no
of K2S and sulfur or the already synthesised poly- influence of the chalcogenide K2Sx supplied onto
sulfide e. g. K2S5 is supplied. Hence, in the first part product formation could be observed. In all syn-
of the experiments the influence of different poly- theses only K4Nb2S11 was formed. Based on the
chalcogenides (K2Sx) onto the product formation amount of Nb used in the syntheses the yields were
was investigated. The experiments were performed identical. In the next series of experiments two pa-

1384
P. Du¨richen – W. Bensch · Reactions in Molten Alkalimetal Polychalcogenides
rameters were varied: x in K2Sx as well as the Hence, we are not surprised that this polysulfide
ratio K2Sx : Nb. For syntheses with a polysulfide was not formed during our experiments.
with x
5 we used K2S5 plus additional sulfur
when necessary and syntheses with x < 5 were done
with K2S and the appropriate amount of S. All ex-
periments were conducted at 350 C. The results
of the experiments are displayed in Fig. 1. In the
graph the amount of S is plotted vs. the amount
of K with the concentration of Nb set to one. The
different symbols mark the products formed with
the different mixtures. The lines are guides for the
eyes and connect the K2Sx samples with different
x values.
We note that as major phases only 3 different
compounds with composition K3NbS4, K4Nb2S11
and K6Nb4S25 were formed. As a by-product
K4Nb2S14 was obtained using a polysulfide melt
with formal composition “K2S7”. In the X-ray pow-
der pattern of the products marked with d elemental
sulfur could be identified. Hence, there is an upper
limit where additional sulfur plays no role for the
syntheses.
One important point should be mentioned here:
Using a K:Nb ratio of 2:1 and low amounts of S
(symbols marked with a and b) no “true” meltcake
is formed but rather a dense and crumbly product
is obtained. Treating the product with DMF and di-
ethylether nearly no excess polysulfide is dissolved
and the crystals are small and of poor quality. This
observation may indicate that the product consists
of a phase mixture. Obviously, when no true melt is
formed the diffusion is low at the low temperatures
leading to local inhomogeneities. On the other hand
it cannot be excluded that inhomogeneities within
the mixture of the educts are present that are not
removed during the reaction.
The most important result is that the amount of
the polysulfide applied has no influence onto prod-
uct formation, i. e. the ratio K2Sx: Nb is not im-
portant. The length of the polysulfide chain, i. e.
the value of x in K2Sx determines what product is
formed. For x = 3 - 4 only K3NbS4 is observed
which only contains S2 anions. Increasing x to 5
- 6 K42Nb2S11 is formed with a structure contain-
ing S2 anions. Finally, applying a melt with x >
Ex-situ Investigations
The reactions occurring in the melt were also in-
vestigated with DSC, FT-IR spectroscopy and X-ray
diffractometry (XRD). The DSC experiments were
conducted in sealed and evacuated glass ampoules
with about 20 mg of K2Sx. The material was heated
to 400 C with a heating rate of 3 C/min., held at this
temperature for 2 h and cooled to room temperature
with 3 C/min. This heating cycle was performed
two times. At the end of the treatment the products
were characterised with XRD.
For a better understanding of the DSC curves of
the reactions between Nb and K2Sx the pure polysul-
fides K2S3 und K2S5 were characterised with DSC
in advance. In addition, mixtures of S with K2S
were also investigated with DSC (K2S:S = 1:4 and
K2S3:S = 1:2).
For K2S3 and K2S5 the melting points (Tonset val-
ues) were 294 C and 206 C respectively. Both val-
ues are in good agreement with data published in
the literature (K2S3: 292 C; K2S5: 211 C [14, 15].
Upon heating a mixture of K2S/S with a ratio of
1:4 several thermal events were observed during
the first heating cycle. But on heating the product a
second time only one melting point at 206 C was
observed. The X-ray powder pattern of the product
contained only reflections of K2S5. For the reaction
between K2S3 and S (ratio 1:2) again several ther-
mal events were observed in the first heating curve.
But only one endothermic signal at 206 C in the
second heating cycle indicated the melting of K2S5
formed during the first treatment. In the X-ray pow-
der pattern of the product only the reflections of
K2S5 were seen.
The results of our DSC as well as XRD investiga-
tions of reactions between K2Sx and sulfur suggest
that in all cases the pentasulfide is formed in-situ.
Hence, the reaction of K2S5 with Nb in the 2:1 ratio
was studied with DSC, XRD, and FT-IR. According
to the results presented above the only product of
the reaction is K4Nb2S11.
6 K6Nb4S25 is found as the product with a crystal
The DSC curve for the reaction of Nb with pure
2
structure containing S5 polysulfide anion. It must K2S5 leading to K4Nb2S11 is displayed in Fig. 2.
be noted that the polysulfide K6Nb4S22 was synthe- A weak endothermic event occurs at 190 C fol-
sised using NbS2 instead of Nb [11] and all attempts lowed by the melting of K2S5 at 206 C. The melt-
to prepare this compound in the usual way failed. ing is followed by two exothermic processes (216

P. Du¨richen – W. Bensch · Reactions in Molten Alkalimetal Polychalcogenides
1385
Fig. 2. DSC curve for a mixture of K2S5 with Nb.
Fig. 3. FT-IR spectra of the products formed at the differ-
ent temperatures indicated in Scheme 1.
sulfide starts. In the spectrum a weak absorption at
335 cm 1 starts to grow. This band is characteristic
for K4Nb2S11. At 217 C the bands of K2S5 have
Scheme 1. Temperature program for the quenching ex-
periments. The numbers are the steps where ampoules are
removed and immediately quenched. 1: 20 C, 2: 210 C,
3: 213 C, 4: 217 C, 5: 230 C, 6: 290 C, 7: 350 C,
8: 350 C after 24 h, 9: 300 C, 10: 230 C, 11: final
product after cool to room temperature.
1
nearly disappeared and the mode at 335 cm be-
comes more intense. In the spectra of the samples
quenched from even higher temperatures K2S5 can-
not be identified and the intensities of the typical
absorptions of the product are more pronounced. At
the end of the heating cycle (350 C) the charac-
teristic pattern of K4Nb2S11 located between 260
and 228 C) that are due to the oxidation of Nb. No
further thermal events occur up to 400 C and in the
cooling curve.
For the XRD and FT-IR investigations a reaction
mixture K2S5 + Nb was prepared and distributed
over 9 ampoules. The heat treatment of the am-
poules was identical with the heat procedure used
for the syntheses of K4Nb2S11.
During the heating procedure ampoules were re-
moved from the furnace at distinct temperatures (see
numbers in Scheme 1). These ampoules were im-
mediately quenched in liquid nitrogen, opened in a
1
and 360 cm is found. After annealing for 24 h
the IR spectrum is very similar to that of the pure
product K4Nb2S11. Differences of the intensities are
due to the fact that the spectrum was recorded with
the as-prepared mixture whereas the spectrum of
K4Nb2S11 was measured after purification, i. e. re-
moval of unreacted educts.
The last two spectra obtained with samples
glove-box and characterised with XRD and FT-IR quenched during the cooling cycle show no signif-
icant changes. A similar picture was obtained with
X-ray diffraction experiments [8]. The metal is not
in the as received state. The spectra are displayed
in Fig. 3. In the figure the spectra of pure K2S5
measured at 20 C and of pure K4Nb2S11 are also consumed within the first exothermic reaction and
reflections of elemental Nb could be detected up to
shown.
At 210 C which is the melting temperature of 350 C. After 24 h the Nb had completely reacted
K2S5 the absorption modes of the pentasulfide are with the sulfide melt. Interestingly, reflections of the
seen. Upon increasing the temperature to about crystalline product K4Nb2S11 develop immediately
213 C the intensities of the bands strongly decrease. after the reaction of Nb with the K2S5 melt. It is
According to the DSC experiments at this tempera- highly likely that the species formed at the surface
ture the reaction between Nb and the molten poly- of the Nb crystallites diffuse away to crystallise as

1386
P. Du¨richen – W. Bensch · Reactions in Molten Alkalimetal Polychalcogenides
K4Nb2S11. We have no hints for the formation of a
crystalline intermediate compound.
When K2Sx (x < 5) is heated with sulfur, in the
first step the pentasulfide K2S5 is formed which
The results of the present experiments yield the melts at about 206 C. Immediately after melting
following picture for the reaction ofNb in potassium the polysulfide reacts with elemental Nb. The FT-IR
polysulfide melts: The amount of the polysulfide and X-ray investigations demonstrate that after oxi-
applied has no influence onto product formation. dation the anion [Nb2S11]4 is formed relatively fast
That means the ratio K2Sx : Nb is not important. and after a short time crystalline K4Nb2S11 appears.
As discussed above the length of the polysulfide After 24 h the reaction is complete as is evidenced
chain determines which product is formed. Exclu- by the very small changes of the intensities of the
sive crystallisation of K3NbS4 is observed in K2Sx absorptions in the FT-IR spectra and of the reflection
melts with x = 3 - 4. When x is increased to 5 - 6, intensities in the X-ray powder patterns [8].
the polysulfide K4Nb2S11 is formed with a structure
2
Acknowledgements
containing S2 anions. The compound K6Nb4S25
2
with a structure containing the S5 polysulfide an-
Financial support by the State of Schleswig-Holstein
ion is the only product upon applying melts with and the Deutsche Forschungsgemeinschaft (DFG) is
gratefully acknowledged.
x > 6.
[10] W. Bensch, P. Du¨richen, Inorg. Chim. Acta 261, 103
(1997).
[11] P. Stoll, C. Na¨ther, W. Bensch, Z. Anorg. Allg.
[1] M. G. Kanatzidis, Chem. Mater. 2, 353 (1990).
[2] M. A. Ansari, J. M. McConnachie, J. A. Ibers, Acc.
Chem. Res. 26, 574 (1993).
Chem. 628, 2489 (2002).
[12] W. Bensch, P. Du¨richen, Eur. J. Solid State Inorg.
Chem. 33, 1233 (1996).
[3] M. G. Kanatzidis, Phosphorous, Sulfur, Silicon 93-
94, 159 (1994).
[4] M. G. Kanatzidis, A. C. Sutorik, Progr. Inorg. Chem.
43, 151 (1995).
[5] M. G. Kanatzidis, S. P. Huang, Coord. Chem. Rev.
130, 509 (1994).
[6] M. A. Pell, J. A. Ibers, Chem. Ber. 130, 1 (1997).
[7] J. A. Cody, M. F. Mansuetto, S. Chien, J. A. Ibers,
Materials Science Forum 152-153, 35 (1994).
[8] W. Bensch, P. Du¨richen, C. Na¨ther, Solid State Sci. 1,
85 (1999).
[13] T. J. McCarthy, S-P. Ngeyi, J-H. Liao, D. C.
DeGroot, T. Hogan, C. R. Kannewurf, M. G.
Kanatzidis, Chem. Mater. 5, 331 (1993).
[14] N. N. Greenwood, A. Earnshaw, Chemie der Ele-
mente, p. 892, VCH 1990.
[15] Holleman-Wiberg, Lehrbuch der anorganischen
Chemie, 101. Auflage, p. 561, de Gruyter, Berlin
& New York (1995).
[9] W. Bensch, P. Du¨richen, Eur. J. Solid State Inorg.
Chem. 33, 527 (1996).