Synthesis, structure, magnetic and optical properties of ternary thio-germanates: Ln4(GeS4)3 (Ln = Ce, Nd)

Article abstract of DOI:10.1002/zaac.200700511

Single crystals of two ternary thio-germanates containing rare-earth metals, Ln₄(GeS₄)₃ (Ln = Ce (I), Nd (II)), have been isolated from the reaction of anhydrous rare-earth trichloride (LnCl ₃) and ternary sodium thio-germanate (Na₂GeS₃) in evacuated quartz ampoules. We have determined the crystal structure of the compounds, which are isostructural to La₄(GeS₄) ₃ and crystallize in trigonal system in the space group R3c with the cell dimensions: I, a = b = 19.375(3)A, c = 8.028(2) A, Z = 6; II, a = b = 19.250(3) A, c = 7.949(2) A, Z = 6. The structure is built with the complex network of two independent tricapped trigonal prisms of CeS₉, in which Ge atoms occupy tetrahedral holes of sulfur atoms. The bulk synthesis of the two compounds has also been achieved by the stoichiometric combination of the elements. The magnetic and optical properties of the compounds have been investigated. The magnetic moments of 2.32 and 3.49 μ_B for I and II, respectively, are in good agreement with the theoretical magnetic moments of Ce and Nd in the +3 oxidation state. The optical band gap of I is found to be located around 2.3 eV, while the optical band gap of II lies around 2.62 eV. In addition, Raman spectroscopic characterizations have also been performed for I, II, and La4(GeS₄)₃.

Full text of DOI:10.1002/zaac.200700511

DOI: 10.1002/zaac.200700511
Synthesis, Structure, Magnetic and Optical Properties of Ternary
Thio-germanates: Ln4(GeS4)3 (Ln ؍ Ce, Nd)
Amitava Choudhury and Peter K. Dorhout*
Fort Collins, Colorado/USA, Department of Chemistry, Colorado State University
Received October 9th, 2007.
Abstract. Single crystals of two ternary thio-germanates containing
rare-earth metals, Ln4(GeS4)3 (Ln ϭ Ce (I), Nd (II)), have been
isolated from the reaction of anhydrous rare-earth trichloride
(LnCl3) and ternary sodium thio-germanate (Na2GeS3) in evacu-
ated quartz ampoules. We have determined the crystal structure of
the compounds, which are isostructural to La4(GeS4)3 and crys-
tallize in trigonal system in the space group R3c with the cell di-
achieved by the stoichiometric combination of the elements. The
magnetic and optical properties of the compounds have been inves-
tigated. The magnetic moments of 2.32 and 3.49 μB for I and II,
respectively, are in good agreement with the theoretical magnetic
moments of Ce and Nd in the ϩ3 oxidation state. The optical band
gap of I is found to be located around 2.3 eV, while the optical band
gap of II lies around 2.62 eV. In addition, Raman spectroscopic
characterizations have also been performed for I, II, and
La4(GeS4)3.
˚
˚
mensions: I, a ϭ b ϭ 19.375(3) A, c ϭ 8.028(2) A, Z ϭ 6; II, a ϭ
˚
˚
b ϭ 19.250(3) A, c ϭ 7.949(2) A, Z ϭ 6. The structure is built with
the complex network of two independent tricapped trigonal prisms
of CeS9, in which Ge atoms occupy tetrahedral holes of sulfur
atoms. The bulk synthesis of the two compounds has also been
Keywords: Lanthanides; Thiogermanates; Crystal structure;
Magnetic property; Optical property
Introduction
Recently, we have been exploring the possibility of mak-
ing new multinary chalcogenides by the reaction of anhy-
drous rare-earth halides and ternary chalcogenides of main
group and alkali metals [35, 36]. Often this method yields
good quality single crystals, likely a result of the formation
of an alkali halide byproduct, which can also act as a flux,
thus making it an alternative route for the exploratory syn-
thesis of new structures. In order to extend the above con-
cept, we have attempted to stabilize new compounds from
thio-germanate building blocks with Ce and Nd through
the exploration of LnCl3/Na2GeS3 (Ln ϭ Ce, Nd) systems.
Our efforts have yielded good quality single crystals of two
compounds of the composition Ln4(GeS4)3 (Ln ϭ Ce, Nd),
the existence of which were reported by Flahaut et al. [37]
and structure type was reported for La4(GeS4)3 from single
crystal data [38]. In this article we report the synthesis,
structure, magnetic and optical properties of these two com-
pounds.
The research in the area of rare-earth chalcogenides is thriv-
ing not only through the rich structural varieties observed
in the materials but also through the widely varying physi-
cal properties, which promise potential applications. A very
powerful chemical approach that involves the use of a low
melting alkali metal poly-chalcogenide flux has widened the
scope of exploratory synthesis of new chalcogenide com-
pounds with rare-earth and main group elements [1, 2].
Thus, a large number of ternary and quaternary rare-earth
containing chalcogenides have been synthesized by the
poly-chalcogenide flux method [3Ϫ8], and to a lesser extent
by the use of alkali halide fluxes [9, 10]. The physical
properties of these compounds are impacted by both the
rare-earth element, which has inner lying f-orbitals with lo-
calized energy levels and narrow optical absorption and em-
ission bands, high magnetic moments (Eu2ϩ, Gd3ϩ
Ϫ
Er3ϩ), as well as the effect of high covalency, high refractive
index, and low energy phonons as a result of the heavy chal-
cogen atoms [11Ϫ15]. For these reasons, rare-earth chalcog-
enides have been investigated as luminescent [16Ϫ30], in-
dustrial pigment [31], magneto-optical [32], optical window
[33], and optical amplifier [34] materials both in crystalline
[16Ϫ32] and glassy matrix materials [33, 34].
Experimental Section
Synthesis
Good quality single crystals of the compounds Ln4(GeS4)3 (Ln ϭ
Ce (I), Nd (II)) were obtained from the reaction of anhydrous rare-
earth chloride (LnCl3) and ternary sodium thio-germanate,
Na2GeS3. Na2GeS3 was prepared by heating a stoichiometric com-
bination of Na2S2, Ge, and S in a sealed fused silica ampoule at
725 °C for 3 hours. Na2S2 was prepared from the stoichiometric
combination of the elements in liquid ammonia as described pre-
viously for several alkali polychalcogenides [39, 40]. CeCl3 and
NdCl3 (Sigma-Aldrich, 99.99 % powder) were used as received.
The original aim of the reaction was to prepare NaLn3GeS7
* Prof. Dr. Peter K. Dorhout
Department of Chemistry
Colorado State University
Fort Collins, CO 80523/USA
Fax: ϩ 1 970-491-1801
E-mail: pkd@lamar.colostate.edu
Z. Anorg. Allg. Chem. 2008, 634, 649Ϫ656
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
649

A. Choudhury, P. K. Dorhout
(CuLa3SiS7-structure type) through a metathesis related route [41].
Typically 0.36 mmol of LnCl3 (Ln ϭ Ce, Nd) and 0.28 mmol of
Na2GeS3 (9 LnCl3:7 Na2GeS3) were mixed and ground in an agate
mortar inside a N2-filled glove-box. The mixture was then trans-
ferred to a graphite crucible, which was then placed inside a fused
silica tube; flame sealed in vacuum and placed vertically into a
temperature-controlled furnace. The furnace was ramped to 825 °C
at a rate of 20 °C/h, and held constant at that temperature for 96
hours. The furnace was then slowly cooled to ambient temperature
at a rate of 5 °C/h. The as-synthesized product, in the case of
CeCl3, contained orange crystals (major product), a few dark col-
ored hexagonal rods, and a small amount of colorless/white crystal-
line material, while in the case of NdCl3, the color of the major
product was dark orange and the minor products looked similar to
the CeCl3 reaction. The product mixture was washed with DMF
containing 10 % H2O and dispersed in a sonocation bath for 1
minute, which removed some white product, presumably NaCl,
while leaving some colorless plates, orange, and dark colored crys-
tals unaffected. Subsequent single-crystal X-ray analysis indicated
that the colorless plates from both reactions were GeS2, while or-
ange/dark orange and dark colored rod crystals were Ce4(GeS4)3
(I) and NaCe3GeS7, respectively for the CeCl3 reaction and
Nd4(GeS4)3 (II) and NaNd3GeS7, respectively for the NdCl3 reac-
tion [42]. We have also rationalized the synthesis of bulk polycrys-
talline powders of I and II along with La4(GeS4)4 for physical and
optical property measurements. The powder samples were prepared
by heating a stoichiometric combination of pure elements, or GeS2
as the Ge source, in sealed fused silica ampoules in a furnace. The
ampoules were first slowly heated to 550 °C (5 °C/h) and kept at
that temperature for 24 hours followed by a ramp to 1000 °C and
kept at that temperature for 8 hours, cooled down to 600 °C at a
rate of 35 °C/h, and then annealed at that temperature for 5 days
before cooling down to room temperature. The powder X-ray dif-
fraction (PXRD) of the as-synthesized product for all the reactions
indicated the presence of minor amounts of unknown impurities.
Table 1 Crystal data and structural refinement parameters for
compounds I and II.
I
II
Chemical formula
Formula weight
Crystal System
Space group
Ce4(GeS4)3
1162.97
trigonal
R3c (no. 161)
293
19.375(1)
8.029(2)
90.0
120
2610.2(3)
6
4.439
16.76
Nd4(GeS4)3
1179.45
trigonal
R3c (no. 161)
293
19.250(2)
7.949(1)
90.0
120
2551.1(1)
6
4.606
18.66
Temperature /K
˚
a ϭ b /A
c /A
˚
Ͱ ϭ β /°
γ /°
3
˚
Volume /A
Z
ρcalc /MgmϪ3
μ /mmϪ1
Reflections collected
Unique Reflections /Rint.
R1 [I >2σ(I)]a)
wR2 (F2)(all data)b)
7044
1188 (0.075)
0.043
7500
1405 (0.074)
0.0276
0.0712
0.0964
2
a) R1 ϭ Σ ʈFoΗϪΗFcʈ/ Σ ΗFoΗ. b) wR2 ϭ {Σ [w(F0 Ϫ Fc2)2] / Σ [w(F02)2]}1/2, w ϭ
2
1/[σ2(FO)2 ϩ (aP)2 ϩ bP] where P ϭ [FO ϩ 2FC2]/3; a ϭ 0.0736 and b ϭ
4.49 for I; a ϭ 0.046 and b ϭ 2.035 for II.
Table 2 Final coordinates and equivalent isotropic displacement
parameters of the atoms for compounds I and II (denoted in
italics).
2
˚
Atom
Wyckoff
x
y
z
U(eq) /A
Ge1
Ge1
Ce1
Nd1
Ce2
Nd2
S1
S1
S2
S2
S3
18b
6a
0.20008(5)
0.19974(4)
0
0.18738(6)
0.18717(4)
0
0.15401(8)
0.15641(9)
0
0.0138(3)
0.0147(2)
0.0151(2)
0.0166(1)
0.0185(2)
0.0195(1)
0.0178(7)
0.0189(4)
0.0157(7)
0.0161(4)
0.0180(7)
0.0186(5)
0.0176(6)
0.0184(4)
0
0
0
18b
18b
18b
18b
18b
0.00341(3)
0.00434(2)
0.15470(5)
0.15389(9)
0.12421(4)
0.12346(9)
0.11404(4)
0.11357(8)
0.39601(4)
0.39581(9)
0.23122(3)
0.23224(2)
0.37867(4)
0.37839(9)
0.06300(4)
0.06122(8)
0.20038(4)
0.20100(8)
0.05940(4)
0.05854(9)
0.20416(8)
0.20714(6)
0.1615(3)
0.1608(2)
0.2502(3)
0.2500(2)
0.9977(3)
0.9984(2)
0.1844(3)
0.1889(2)
S3
S4
S4
Single crystal X-ray diffraction
U(eq) ϭ 1/3 of the trace of the orthogonalized U Tensor
Intensity data sets for the compounds (I؊II) were collected on a
Bruker Smart CCD diffractometer. These data were integrated with
SAINT [43], the program HABITUS was used for the numerical
absorption correction [44]. The structures were solved by direct
methods using SHELXS-97 [45] and difference Fourier syntheses.
Full-matrix least-squares refinement against ΗF2Η was carried out
using the SHELXTL-PLUS [46] suit of programs. Based on the
systematic absences, the R3c space group was chosen for both the
compounds, which immediately revealed that I and II are isostruc-
tural to La4(GeS4)3. Rare-earth cations can be easily located on the
6a and 18b sites, while subsequent refinement identified Ge on the
18b site and four S atoms on the 18b site. The last cycles of refine-
ment for I and II included anisotropic thermal parameter refine-
ment for all the atoms. Details of the final refinements and the
cell parameters for I and II are given in Table 1. The final atomic
coordinates and important inter-atomic distances and angles for I
and II are given in Tables 2 and 3, respectively.
Physical Property Measurements
The bulk solid-state Raman spectra were collected on a Nicolet
Magna-IR 760 spectrometer with a FT-Raman Module attachment
by use of a Nd:YAG excitation laser (1064 nm). Diffuse reflectance
measurement for the compounds were done on a Varian Cary 500
UV-vis-NIR spectrophotometer equipped with a Praying Mantis
accessory. A Teflon standard was used as a reference. The Kubelka-
Munk function was applied to obtain band-gap information [47,
48]. Magnetic susceptibility measurements of I and II were made
at 2 Tesla and 1 Tesla, respectively, over the temperature range 1.8
Ϫ 300 K and isothermal magnetization at 1.8 K up to a field of 5
Tesla was measured with a Quantum Design SQUID magnet-
ometer.
Results and Discussion
Further details of the crystal structure investigations may be
obtained from the Fachinformationszentrum Karlsruhe, 76344
Eggenstein-Leopoldshafen, Germany, (fax: (49) 7247-808-666;
e-mail: crysdata@fiz.karlsruhe.de) on quoting the depository num-
ber CSD-418634 (Ce4(GeS4)3) and CSD-418635 (Nd4(GeS4)3).
Structure description of Ln4(GeS4)3 (Ln ؍ Ce, Nd)
The structures of Ce4(GeS4)3 (I) and Nd4(GeS4)3 (II) are
isostructural to La4(GeS4)3 [38] and are made up of com-
650
www.zaac.wiley-vch.de
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2008, 649Ϫ656

Ternary Thio-germanates
˚
˚
3.5528(1) and 3.6785(1) A, respectively, for II. The Ge atom
Table 3 Selected interatomic distances /A and angles /° for the
coordination polyhedra of I and II.
sits in the center of a tetrahedron surrounded by four S
atoms with the GeϪS distances in the range
2.200(2)Ϫ2.241(3) A and 2.200(1)Ϫ2.242(1) A, respectively,
˚
˚
Moiety
Distance /A Moiety
Distance /A
˚
˚
for and II; SϪGeϪS angles are in the range
I
I
101.1(9)Ϫ125.33(10)° and 101.22(7)Ϫ125.43(7)°, respec-
tively, for I and II. The deviation of SϪGeϪS angles from
the ideal value indicates that the GeS4 tetrahedra are
slightly distorted.
Ge(1) Ϫ S(3)#1
Ge(1) Ϫ S(4)#2
Ge(1) Ϫ S(1)#3
Ge(1) Ϫ S(2)
2.200(2)
2.201(2)
2.203(2)
2.241(3)
Ce(2) Ϫ S(3)#1
2.997(3)
3.005(3)
3.074(3)
3.537(1)
3.7223(1)
Angle (°)
Ce(2) Ϫ S(1)#12
Ce(2) Ϫ S(3)#5
Ce(2) Ϫ S(1)#13
Ce(2) Ϫ S(4)#14
Moiety
(#4, #5, #6)
(#7, #8)
Ce(1) Ϫ S(2) ϫ 3
Ce(1) Ϫ S(2) ϫ 3
Ce(1) Ϫ S(3) ϫ 3
2.893(3)
2.894(3)
3.373(1)
S(3)#1 Ϫ Ge(1) Ϫ S(4)#2 125.33(10)
S(3)#1 Ϫ Ge(1) Ϫ S(1)#3 109.71(11)
S(4)#2 Ϫ Ge(1) Ϫ S(1)#3 103.24(10)
S(3)#1 Ϫ Ge(1) Ϫ S(2) 101.10(9)
S(4)#2 Ϫ Ge(1) Ϫ S(2) 112.16(10)
S(1)#3 Ϫ Ge(1) Ϫ S(2) 103.53(9)
(#9, #10)
Ce(2) Ϫ S(4)#7
Ce(2) Ϫ S(2)#7
Ce(2) Ϫ S(1)
2.857(2)
2.899(2)
2.915(3)
2.956(2)
Ce(2) Ϫ S(4)#11
II
˚
˚
Distance (A)
Moiety
Distance (A) Moiety
Ge(1) Ϫ S(3)#1
Ge(1) Ϫ S(1)#2
Ge(1) Ϫ S(4)#3
Ge(1) Ϫ S(2)
2.2005(19)
2.2009(18)
2.2050(18)
2.2423(18)
Nd(2) Ϫ S(1)#12
2.9616(19)
2.9666(18)
3.0400(19)
3.5528(1)
3.6785(1)
Angle (°)
Nd(2) Ϫ S(3)#1
Nd(2) Ϫ S(3)#6
Nd(2) Ϫ S(1)#13
Nd(2) Ϫ S(4)#14
Moiety
(#4, #5, #6)
(#7, #8)
Nd(1) Ϫ S(2) ϫ 3
Nd(1) Ϫ S(2) ϫ 3
Nd(1) Ϫ S(3) ϫ 3
2.8608(18)
2.8613(18)
2.3601(1)
S(3)#1 Ϫ Ge(1) Ϫ S(1)#2 109.25(8)
S(3)#1 Ϫ Ge(1) Ϫ S(4)#3 125.43(7)
S(1)#2 Ϫ Ge(1) Ϫ S(4)#3 102.91(7)
S(3)#1 Ϫ Ge(1) Ϫ S(2) 101.22(7)
S(1)#2 Ϫ Ge(1) Ϫ S(2) 103.42(7)
S(4)#3 Ϫ Ge(1) Ϫ S(2) 112.78(7)
(#9, #10)
Nd(2) Ϫ S(4)#7
Nd(2) Ϫ S(1)
2.8323(18)
2.8703(18)
2.8789(17)
2.9167(18)
Nd(2) Ϫ S(2)#7
Nd(2) Ϫ S(4)#11
Symmetry transformations used to generate equivalent atoms: #1 x, y, z Ϫ 1; #2
Ϫ
Fig. 1 Perspective view of the 3-D structure of I along the c-axis
showing the complex networks of rare-earth polyhedra and GeS4
tetrahedra. The connectivity around one of the rare-earth atoms
(Ce1) has been drawn as a ball and stick representation to show
how they are located into the 1-dimensional channel.
y ϩ 1/3, Ϫ x ϩ 2/3, z ϩ 1/6; #3 Ϫ y ϩ 2/3, Ϫ x ϩ 1/3, z Ϫ 1/6; #4 Ϫ y, Ϫ x, z Ϫ
1/2; #5 Ϫ x ϩ y, y, z Ϫ 1/2; #6 x, x Ϫ y, z Ϫ 1/2; #7 Ϫ y, x Ϫ y, z; #8 Ϫ x ϩ y, Ϫ x,
z; #9 Ϫ y, x Ϫ y, z Ϫ 1; #10 Ϫ x ϩ y, Ϫ x, z Ϫ 1; #11 Ϫ x ϩ y ϩ 1/3, Ϫ x ϩ 2/3, z
Ϫ 1/3; #12 Ϫ x ϩ y Ϫ 1/3, Ϫ x ϩ 1/3, z ϩ 1/3; #13 x Ϫ 1/3, 1/3 ϩ x Ϫ y, z Ϫ 1/6;
#14 Ϫ 1/3 ϩ x, 1/3 ϩ y, 1/3 ϩ z.
plex 3-dimensional (3-D) networks of tricapped trigonal
prisms (ttps) of LnS9 and GeS4 tetrahedra (Fig. 1). The
asymmetric unit contains two crystallographically distinct
lanthanide atoms, one germanium atom, and four sulfur
atoms, where the Ln atoms are surrounded by 9 sulfur
atoms in a tricapped trigonal prismatic coordination and
the Ge atoms are surrounded by 4 sulfur atoms in a tetra-
hedral coordination (Fig. 2). Ln1 is located in the center of
a three-fold axis, surrounded by 6 sulfur atoms in an almost
regular trigonal prismatic environment with two equilateral
triangular bases formed by S2 atoms (Table 3). Moreover,
the rectangular faces of the prism are capped by three S
atoms (S3) lying at a greater distance, 3.373(1) and
˚
3.3601(1) A, respectively for I and II. Ln2, on the other
hand, sits on a general position, surrounded by 6 sulfur
atoms in a very distorted trigonal prismatic coordination
˚
with LnϪS distances in the range 2.857(2)Ϫ3.074(3) A for
˚
I and 2.832(1)Ϫ3.040(1) A for II; in addition, the two tri-
angular bases do not form equilateral triangles. There are
three capping S atoms from the three rectangular faces of
the prism, with one normal and two very long Ln2ϪS dis-
Fig. 2 The asymmetric unit of I showing the tricapped trigonal
prismatic coordination of two crystallographically distinct rare-
earth atoms and tetrahedral coordination of Ge atom. The thermal
ellipsoids are given at 50 % probability.
˚
tances, 3.537(1) and 3.7223(1) A, respectively, for I and
Z. Anorg. Allg. Chem. 2008, 649Ϫ656
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
651

A. Choudhury, P. K. Dorhout
Ln1S9 ttps share their triangular faces along the three-
fold axis that forms a chain along the c-axis (Fig. 3). Ln2S9
ttps, on the other hand, share the edges (S1 and S4) and
form a three-fold helix along the c-axis (Fig. 4). Apart from
the helix, the Ln2S9 ttps also share corners to form a com-
plex 3-D network of Ln2S9 polyhedra, which create 1-D
channels along the c-axis (Fig. 1). The face-shared chain of
Ln1S9 ttps fills those channels, and while doing so, each
Ln2S9 ttp shares edges with two Ln1S9 ttps. Moreover, each
Ln1S9 ttps shares edges with six Ln2S9 ttps. Thus the con-
nectivity between rare earth polyhedra creates a 3-D array
and Ge cations lie inside the tetrahedral holes of four S
atoms resulting from such an assembly. The LnϪS distances
in both I and II are in agreement with the reported litera-
ture for Ce- and Nd containing sulfide compounds where
they are in ϩ3 oxidation state [49, 50]. The oxidation states
of different elements have also been established from bond-
valence sum calculations [51], which yield valence sums at
Ge1 ϭ 4.102, Ce1 ϭ 3.256, Ce2 ϭ 2.994, S1 ϭ 1.935, S2 ϭ
2.370, S3 ϭ 1.841, and S4 ϭ 2.034 for compound I and
Ge1 ϭ 4.090, Nd1 ϭ 3.258, Nd2 ϭ 3.011, S1 ϭ 1.962, S2 ϭ
2.360, S3 ϭ 1.837, and S4 ϭ 2.028 for compound II.
Therefore, based on the LnϪS distances and the bond-
valence sum calculations a charge balance formula of
It should be mentioned at this point that while the exist-
ence of these compounds has been realized in the larger
lanthanides, La Ϫ Gd [37], there is no report of the isotypic
compounds with the smaller, later lanthanides (Tb Ϫ Yb).
Interestingly, in the thiosilicates, there seem to exist two po-
lymorhs with the Ln4(SiS4)3 composition. In Ln4(SiS4)3,
those with Ln ϭ Ce Ϫ Gd, crystallized in trigonal system
[52], as evident by the structure of Ce4(SiS4)3 refined from
the powder data using the coordinates of the isostructural
La4(GeS4)3 [28], while Hatscher and Urland have reported
Ln4(SiS4)3 (Ln ϭ Gd Ϫ Dy) [53Ϫ55] with a different struc-
ture type crystallizing in monoclinic system in the space
group P21/n. From these studies, we see there is at least one
compound, Gd4(SiS4)3, which exists in both polymorphic
modifications. In the monoclinic form of Ln4(SiS4) (Ln ϭ
Gd Ϫ Dy), the lanthanide ions are surrounded by 7Ϫ8 sulf-
ide ions, fewer compared to the 9-fold coordination of the
lanthanide in the trigonal form, Ln4(GeS4)3, but the lower
coordination number of the former conforms with the
smaller size of the heavier lanthanide.
Other than the difference in the coordination numbers of
the rare-earth atoms, the degree of distortion of the TS4
(T ϭ Si, Ge) tetrahedra in the two polymorphs also differ.
For example, by employing the concept of continuous sym-
metry measurements (S(tet)) [56], we have calculated the dis-
tortion of the GeS4 tetrahedron in the trigonal phases
(S(tet)(Ge1) ϭ 3.252, 3.326, 2.588, and 3.282 for La, Ce, Pr,
and Nd containing phases, respectively), which are much
higher than the distortions of the three crystallographically
distinct SiS4 tetrahedra in the monoclinic polymorphs
(S(tet)(Si1) ϭ 0.735, S(tet)(Si2) ϭ 0.439, and S(tet)(Si3) ϭ
0.440 for Tb4(SiS4)3 and S(tet)(Si1) ϭ 0.757, S(tet)(Si2) ϭ
0.471, and S(tet)(Si3) ϭ 0.438 for Dy4(SiS4)3). So one can
speculate that with decreasing size of the lanthanides, it be-
comes increasingly difficult to accommodate the more dis-
torted GeS4 in the tetrahedral holes of the trigonal struc-
ture, or it may just be the lower coordination of the heavier
lanthanide that prefers the monoclinic form over the trig-
onal. It would be interesting to explore the possibility of
the formation of monoclinic polymorph in the thiogerman-
ate series with the later lanthanides and the role of the te-
trelate ions (Si4ϩ and Ge4ϩ) in stabilizing a particular form.
(Ln3ϩ)4(Ge4ϩ)3(S2Ϫ
II.
)
is reasonable for compounds I and
12
Fig. 3 The face-shared chain of Ce1S9 polyhedra along the c-axis.
Raman spectroscopy
Raman spectra of compound I and La4(GeS4)3 are given in
Fig. 5. The Raman spectrum for compound II is dominated
by the luminescence of f Ǟ f levels of Nd, which masks the
Raman peaks. The main structural moiety that is of interest
in this region of the Raman spectra of these compounds is
the tetrahedral GeS4 unit. For an ideal tetrahedral point
symmetry (Td), one should observe the following internal
modes i) ν1 (A1) Ϫ nondegenerate symmetric bond stretch-
ing; ii) ν2 (E) Ϫ doubly degenerate bond bending (defor-
mation); iii) ν3 (F2) Ϫ triply degenerate antisymmetric
stretching and iv) ν4 (F2) Ϫ triply degenerate bond bending
vibrations [57]. Among all the modes of Raman active vi-
Fig. 4 A view of how the Ce2S9 polyhedra are joined together to
form the three-fold helix along the c-axis. The three capping
S-atoms of the Ce2S6 trigonal prisms have been omitted from the
figure for the sake clarity.
652
www.zaac.wiley-vch.de
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2008, 649Ϫ656

Ternary Thio-germanates
brations, the totally symmetric A1 mode appears as the
most intense line in several thio-germanates containing iso-
lated or connected GeS4 tetrahedra. Consequently, the
strong line at 372 cmϪ1 in I and La4(GeS4)3 can be assigned
to the A1 mode. This value is close to the ν1(A1) mode of
many thio-germanates containing isolated GeS4, e.g,
ment of the Raman peaks is qualitative and based on the
similarities with other GeS4-containing compounds. How-
ever, the spectra may contain some lattice phonon modes
of GeS4 and possibly the rare-earth polyhedra (LnS9) in the
low energy region (328 Ϫ 121 cmϪ1). The weak Raman
modes due to the phonons of the lanthanide cations would
likely appear at even lower energies (90 Ϫ 20 cmϪ1) and are
probably not observed under the current spectroscopic con-
figuration.
4Ϫ
386 cmϪ1 (GeS4 in solution) [58], 380 cmϪ1 (KEuGeS4)
[5], 393 cmϪ1 (Ba2GeS4) [59Ϫ61] 382 cmϪ1 (Sr2GeS4) [60,
61], 375 cmϪ1 (Pb2GeS4) [62] and 389 cmϪ1 (GeS2) [61]. As
the GeS4 tetrahedra are not ideal but distorted by the sur-
rounding atoms both in terms of bond angles (SϪGeϪS)
and distances (GeϪS), the degeneracy of the ν2, ν3, ν4 vi-
brations are removed and expected to decompose into sev-
eral components. Thus we can attribute peaks at 399 and
414 cmϪ1 to the ν3 bond stretching vibrations, while peaks
between 199 Ϫ 280 cmϪ1 and 121 Ϫ 171 cmϪ1 can be tenta-
tively assigned to the low frequency bond bending ν2 and
ν4 vibrations, respectively, of GeS4 tetrahedra. The assign-
Magnetic Measurements
The plot of the temperature dependence of the magnetic
susceptibility (χ) and inverse-susceptibility (χϪ1) of I and II
in the range 1.8 Ϫ 300 K are given in Figs. 6 and 7, respec-
tively. The asymptotic nature of the χ(T) plots at 2 Tesla
and 1 Tesla for I and II, respectively, under Field-Cooled
and Zero-Field Cooled (FC and ZFC) conditions indicate
paramagnetic behavior of the compounds and no other
magnetic transitions are observed under these conditions.
The absence of any transition in I and II was also estab-
lished by the FC and ZFC χ(T) measurements under low
applied field (10 Oe), which are again asymptotic (data not
shown). The plots of χϪ1(T) at high fields (2 T and 1 T for
I and II, respectively) show linear behavior above 100 K
and the data in the range 150 Ϫ 300 K could be fitted to
the Curie-Weiss law, thus yielding the values of the effective
moment (μeff) of 2.32 and 3.49 μB, and the paramagnetic
Curie temperature (θp) as Ϫ8.3 and Ϫ18.3 K for I and II,
respectively. The effective moments are slightly lower than
the theoretical magnetic moments of 2.54 μB and 3.62 μB
expected for Ce and Nd, respectively, in the ϩ3 oxidation
state. However, these values are in agreement with data re-
ported for the many Ce3ϩ and Nd3ϩ containing chalcogen-
ides and intermetallics; for example, KCeGeSe4 (μeff
ϭ
2.3 μB) [36], CePd3S4 (μeff ϭ 2.14 μB) [63], and Nd2Pd2Cd
(μeff ϭ 3.49 μB) [64]. Below 100 K, the inverse susceptibility
Fig. 5 The Raman spectra of compounds I and La4(GeS4)3.
Fig. 6 The plots show the magnetic susceptibility (χ) and inverse susceptibility (χϪ1) of compound I as a function of temperature. The
inset shows the isothermal magnetization at 1.8 K.
Z. Anorg. Allg. Chem. 2008, 649Ϫ656
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
653

A. Choudhury, P. K. Dorhout
Fig. 7 The plots show the magnetic susceptibility (χ) and inverse susceptibility (χϪ1) of compound II as a function of temperature. The
inset shows the isothermal magnetization at 1.8 K.
curves of
I
and II start to deviate from linearity
tion below 420 nm (above 2.95 eV) and 480 nm (above
2.58 eV) for I and II, respectively. The absorption edges lie
around 2.3 eV and 2.62 eV for I and II, respectively. The
absorption edge of 2.3 eV in the case of I can be attributed
to the 4f Ǟ 5d intra-atomic Ce transition, as has also been
observed in the isotypic thio-silicate, Ce4(SiS4)3 [28] while
in the case of II the absorption edge of 2.62 eV represents
the optical band gap of the material and can be assigned to
the electronic transition from predominantly anionic or-
bitals (valence band (VB)) to the conduction band (CB).
The bottom of CB can be assumed to be made of predomi-
nantly 5d-Ln orbitals based on previous calculations in iso-
typic Ce-containing thio-silicate, Ce4(SiS4)3 [28]. It can be
mentioned here that the 4f Ǟ 5d gap of I is slightly nar-
rower than the isostructural cerium thio-silicate, Ce4(SiS4)3
(2.38 eV) [28]. Besides the band edge of II, there are several
absorptions characteristic of the f Ǟ f transitions of Nd3ϩ
(CurieϪWeiss behavior), indicating crystal field splitting of
the J ϭ 5/2 and J ϭ 9/2 ground states of the Ce3ϩ and
Nd3ϩ, respectively [36, 63, 65]. The negative sign and small
magnitude of θp indicate weak antiferromagnetic interac-
tions between the rare-earth ions. The isothermal magnetiz-
ation (M) measured at 1.8 K for both the compounds exhi-
bit a very sluggish increase in magnetization with increasing
field; moreover, M does not seem to saturate even at 5 T,
and the value of the saturation moments at 5 T are about
0.85 μB/Ce (see inset of Fig. 6) and 1.3 μB/Nd (see inset of
Fig. 7) for I and II, respectively. The saturation moment of
0.85 μB/Ce for I at 5 T and 1.8 K is less than that expected
for a fully ordered Ce3ϩ moment (g ϫ J ϭ 2.14 μB/Ce) but
is in close agreement with Ce3ϩ-containing chalcogenide,
CePd3S4 (0.83 μB at 5T/2K) [63] and intermetallic com-
pounds, e.g., CeCuAs2 (0.85 μB at 12T/2K) [66]. The sub-
stantial reduction of the saturation moment can be attri-
buted to the crystal field splitting of the J ϭ 5/2 ground
state and the fact that the magnetic moment is less than
1 μB, suggests that the ground state may be a crystal field
split doublet [63, 66]. The saturation moment of 1.3 μB/Nd
for II at 1.8 K/5 T is smaller than the theoretical value of
3.27μB/Nd for fully ordered Nd3ϩ moment according to
g ϫ J. The saturation magnetization II is even smaller than
several Nd3ϩ containing intermetallic compounds, e.g.,
Nd2Pd2Cd (2 μB at 8T/2K) [64]; the substantial reduction
of the magnetization in II can also be due to the splitting
(4I9/2
Ǟ
2Sϩ1Lj) observed between 1800 Ϫ 450 nm before
the interband transition (Fig. 8b). These transitions can be
tentatively assigned with the help of previous reports to
4
2
originate from the ground state I9/2 to P1/2 (459 nm),
2G9/2 (477 nm) K13/2
ϩ ϩ
4G7/2,9/2 (522, 533 nm), G5/2
2
4
2G7/2 (596 nm), H11/2 (632 nm), F9/2 (693 nm), F7/2 ϩ S3/2
4
4
4
4
2
4
4
(759 nm), H9/2, F5/2 (815 nm) and F3/2 (891 nm) [67Ϫ69].
In addition, we have observed absorption at longer wave-
lengths (1550 Ϫ 1800 nm) due to the multiplet manifolds of
4I15/2 (inset of Fig. 8b).
4
of the I9/2 ground state.
Conclusion
We have successfully synthesized and determined the struc-
tures of two ternary thio-germanates in Ln4(GeS4)3 family.
We have also determined their magnetic and preliminary
optical properties. These compounds are stable in air for a
long time, so their luminescence properties could be useful
as well as the effect of substitution chemistry on optical
Optical Spectroscopy
The absorption data calculated from diffuse reflectance
spectra (DRS) using the Kubelka-Munk function for com-
pounds I and II are given in Figure 8a. The diffuse reflec-
tance spectra of both the compounds show strong absorp-
654
www.zaac.wiley-vch.de
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2008, 649Ϫ656

Ternary Thio-germanates
[5] C. R. Evenson, P. K. Dorhout, Inorg. Chem. 2001, 40, 2409.
[6] R. G. Iyer, J. A. Aitken, M. G. Kanatzidis, Solid State Sci.
2004, 6, 451.
[7] B. R. Martin, P. K. Dorhout, Inorg. Chem. 2004, 43, 385.
[8] B. R. Martin, L. A. Polyakova, P. K. Dorhout, J. Alloys
Compds. 2006, 408Ϫ412, 490.
[9] I. Hartenbach, T. Schleid, Z. Anorg. Allg. Chem. 2002, 628,
1327.
[10] I. Hartenbach, T. Schleid, Z. Anorg. Allg. Chem. 2003, 629,
394.
[11] Spectroscopic Properties of Rare Earths in Optical Materials;
G. Liu, B. Jacquier, Eds.; Tsinghua University Press and
Springer-Verlag Berlin, Tsinghua, 2005, Vol. 83.
[12] S. Cotton, Lanthanide and Actinide Chemistry, John Wiley &
Sons, Ltd, Chichester, 2006.
[13] R. Reisfeld, A. Bornstein, Chem. Phys. Lett. 1977, 47, 194.
[14] R. Reisfeld, A. Bornstein, J. Flahaut, A. M. LoireauLazac’h,
in: The Rare Earths in Modern Science Technology, G. J.
McCarthy et al. (eds), Plenum, New York, 1980, Vol. 2.
[15] R. Reisfeld, Ann. Chim. Fr. 1982, 7, 147.
[16] T. E. Peters, J. A. Baglio, J. Electrochem. Soc. 1972, 119, 230.
[17] A. Garcia, C. Fouassier, P. Dougier, J. Electrochem. Soc. 1982,
129, 2063.
[18] R. Iban˜ez, A. Garcia, C. Fouassier, P. Hagenmuller, J. Solid
State Chem. 1984, 53, 406.
[19] R. Ibanez, P. Garavereau, A. Garcia, C. Fouassier, J. Solid
State Chem. 1988, 73, 252.
[20] M. R. Davolos, A. Garcia, C. Fouassier, P. Hagenmuller, J.
Solid State Chem. 1989, 83, 316.
[21] W. A. Barrow, R. E. Coovert, E. Dickey, C. N. King, C. S.
Laakso, S.-S. Sun, R. T. Tuenge, R. C. Wentross, J. Kane,
Society of Information Display International Symposium-
Digest 93 1993, p761.
[22] S.-S. Sun, R. T. Tuenge, J. Kane, M. Ling, J. Electrochem. Soc.
1994, 141, 2877.
[23] P. Benalloul, C. Barthou, J. Benoit, L. Eichenauer, A. Zeinert,
Appl. Phys. Lett. 1993, 63, 1954.
[24] I. Ronot-Limousin, A. Garcia, C. Fouassier, C. Barthou, P.
Fig. 8 (a) Optical absorption spectra (taken on the crushed single
crystals) of I and II, transformed from the diffuse reflectance data.
(b) Optical absorption spectra of II indicating the characteristic
fǞf transition of Nd. Insets on the right top corner is the blow up
of the low energy absorption lines of II in the 1500-1800 nm region.
Benalloul, J. Benoit, J. Electrochem. Soc. 1997, 144, 687.
´
[25] P. Benalloul, C. Barthou, J. Benoit, J. Alloys Compds. 1998,
275Ϫ277, 709.
`
[26] G. Gauthier, S. Jobic, F. Boucher, P. Macaudiere, D. Hug-
uenin, J. Rouxel, R. Brec, Chem. Mater. 1998, 10, 2341.
[27] R. Riccardi, D. Gout, G. Gauthier, F. Guillen, S. Jobic, A.
`
Garcia, D. Huguenin, P. Macaudiere, C. Fouassier, R. Brec, J.
properties can be performed. For example, small amount of
Ce3ϩ can be substituted in the La4(GeS4)3 to investigate
how the chromatic and luminescence properties vary.
Solid State Chem. 1999, 147, 259.
[28] G. Gauthier, S. Jobic, M. Evain, H.-J. Koo, M.-H. Whangbo,
C. Fouassier, R. Brec, Chem. Mater. 2003, 15, 828.
[29] C. Chartier, C. Barthou, P. Benalloul, J. M. Frigerio, J. Lumin.
2005, 111, 147.
[30] P. Dorenbos, J. Lumin. 2007, 122Ϫ123, 315.
[31] P. Maestro, D. Huguenin, J. Alloys Compds. 1995, 225, 520.
[32] P. Fumagalli, A. Schirmeisen, R. J. Gambino, J. Appl. Phys.
1996, 79, 5929.
Acknowledgement. The authors wish to acknowledge financial
support provided by the National Science Foundation, NSF-
DMR-0343412.
[33] P. N. Kumta, S. H. Risbud, J. Mater. Sci. 1994, 29, 1135.
[34] A. Zakery, S. R. Elliott, J. Non-cryst. Solids 2003, 330, 1.
[35] A. Choudhury, L. A. Polyakova, S. Strobel, P. K. Dorhout, J.
Solid State Chem. 2007, 180, 1381.
References
[1] J. A. Cody, M. F. Mansuetto, S. Chien, J. A. Ibers, Mater. Sci.
Forum 1994, 35, 152.
[2] M. G. Kanatzidis, A. C. Sutorik, Prog. Inorg. Chem. 1995,
43, 151.
[3] P. Wu, J. A. Ibers, J. Solid State Chem. 1993, 107, 347.
[4] J. A. Aitken, G. A. Marking, M. Evain, L. Iordanidis, M. G.
Kanatzidis, J. Solid State Chem. 2000, 153, 158.
[36] A. Choudhury, L. A. Polyakova, I. Hartenbach, T. Schleid, P.
K. Dorhout, Z. Anorg. Allg. Chem. 2006, 632, 2395.
[37] A. Michelet, P. Laurelle, J. Flahaut, C. R. Acad. Sci. Ser. C
1966, 262, 753.
[38] P. A. Mazurier, J. Etienne, Acta Crystallogr. 1974, B30, 759.
Z. Anorg. Allg. Chem. 2008, 649Ϫ656
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
655

A. Choudhury, P. K. Dorhout
[39] I. Schewe-Miller, Metallreiche hauptgruppenmetall-Chalkogen-
[54] S. T. Hatscher, W. Urland, Z. Anorg. Allg. Chem. 2002, 628,
1673.
verbindungen: Synthese, Strukturen and Eigenschaften, Ph.D.
[55] S. T. Hatscher, W. Urland, J. Solid State Chem. 2003, 172, 417.
[56] S. Alvarez, P. Alemany, D. Casanova, J. Cirera, M. Llunell, D.
Avnir, Coord. Chem. Rev. 2005, 249, 1693.
[57] K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, John Wiley & Sons, 1986.
[58] S. Pohl, W. Schiwy, N. Weinstoc, B. Krebs, Z. Naturforschung
1973, 28B, 565.
[59] M. Ribes, M. Maurin, Rev. Chim. Miner. 1970, 7, 75.
[60] M. Meyrand, M. Ribes, E. Philippot, M. Maurin, Rev. Chim.
Miner. 1975, 12, 406.
Thesis,
Stuttgart, 1990.
Max-Planck-Institut
fur
Festkorperforschung,
[40] J.-H. Liao, M. G. Kanatzidis, Inorg. Chem. 1992, 31, 431.
[41] G. Collin, P. Laruelle, Bull. Soc. Fr. Miner. Crystallogr. 1971,
94, 175.
[42] The structure and properties of the compounds NaCe3GeS7
and NaNd3GeS7 will be published in future.
[43] SAINT, Data processing software for the SMART system;
Bruker Analytical X-ray Instruments, Inc.: Madison, WI,
1995.
[44] W. Herrendorf, HABITUS, Program for the Optimization of
the Crystal Shape for Numerical Absorption Correction, Uni-
versity of Gießen: Gießen, Germany, 1995.
[45] G. M. Sheldrick, SHELXS-97, Programs for X-ray Crystal
Structure Solution, University of Göttingen, Göttingen, Ger-
many, 1997.
[46] G. M. Sheldrick, SHELXL-97, Programs for X-ray Crystal
Structure Refinement, University of Göttingen, Göttingen,
Germany, 1997.
[47] W. W. Wendlandt, H. G. Hecht, Reflectance Spectroscopy; In-
terscience Publishers, New York, 1966.
[48] G. Kortüm, Reflectance Spectroscopy, Springer-Verlag, New
York, 1969.
[49] I. Hartenbach, T. Schleid, J. Alloys Compds. 2006, 18, 5.
[50] S. T. Hatscher, W. Urland, Z. Anorg. Allg. Chem. 2001, 627,
2198.
[61] Z. V. Popovic, Fizika 1983, 15, 11.
[62] D. I. Bletskan, Y. V. Voroshilov, L. M. Durdinets, P. P.
Migalko, V. A. Stefanovich, V. N. Kabatsii, Crystallography
Reports 2003, 48, 573.
[63] M. Wakeshima, Y. Hinatsu, J. Solid State Chem. 1999, 146,
226.
[64] S. Rayaprol, A. Dogan, R. Pottgen, J. Phys.: Condens. Matter
2006, 18, 5473.
[65] A. Sasaki, M. Wakeshima, Y. Hinatsu, J. Phys.: Condens.
Matter 2006, 18, 9031Ϫ9046.
[66] K. Sengupta, E. V. Sampathkumaran, T. Takano, M. Hedo,
M. Abliz, N. Fujiwara, Y. Uwatoko, S. Rayaprol, K. Shigetoh,
T. Takabatake, T. Doert, J. P. F. Jemetio, Phys. Rev. B 2004,
70, 64406.
[67] B. Viana, M. Palazzi, O. LeFol, J. Non-Cryst. Solids 1997,
215, 96.
[68] K. Kadono, M. Shojiya, M. Takahashi, H. Higuchi, Y. Kawa-
[51] I. D. Brown, D. Altermatt, Acta Crystallogr. 1985, B 41,
244.
[52] G. Perez, M. Duale, C. R., Acad. Sci. Ser. C 1969, 269, 984.
[53] S. T. Hatscher, W. Urland, Acta Crystallogr. 2002, E 58, i74.
moto, J. Non-Cryst. Solids 1999, 259, 39.
[69] S. M. Lima, J. A. Sampaio, T. Catunda, A. S. S. de Camargo,
L. A. O. Nunes, M. L. Baesso, D. W. Hewak, J. Non-Cryst.
Solids 2001, 284, 274.
656
www.zaac.wiley-vch.de
© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2008, 649Ϫ656