Low-dimensional compounds containing cyano groups. XIV. Crystal structure, spectroscopic, thermal and magnetic properties of [CuL2][Pt(CN)4] complexes (L=ethylenediamine or N,N-dimethylethylenediamine)

Article abstract of DOI:10.1016/j.jssc.2006.03.033

Violet crystals of [Cu(en)₂][Pt(CN)₄] and blue crystals of [Cu(dmen)₂][Pt(CN)₄] were crystallized from the water-methanol solution containing CuCl₂·2H₂O, ethylenediamine (en) or N,N-dimethylethylenediamine (dmen) and K₂[Pt(CN)₄]·3H₂O. Both compounds were characterized using elemental analysis, infrared and UV-VIS spectroscopy, magnetic measurements, specific heat measurements and thermal analysis. X-ray structure analysis revealed chain-like structure in both compounds. The covalent chains are built of Cu(II) ions linked by [Pt(CN)₄]^2- anions in the [111] and [101] direction, respectively. The Cu(II) atoms are hexacoordinated by four nitrogen atoms in the equatorial plane from two molecules of bidentate ligands L with average Cu-N distance of 2.022(2) and 2.049(4) A, respectively. Axial positions are occupied by two nitrogen atoms from bridging [Pt(CN)₄]^2- anions at longer Cu-N distance of 2.537(2) and 2.600(5) A, respectively. Both materials are characterized by the presence of weak antiferromagnetic exchange coupling. Despite the one-dimensional (1D) character of the structure, the analysis of magnetic properties and specific heat at very low temperatures shows that [Cu(en)₂][Pt(CN)₄] behaves as two-dimensional (2D) spatially anisotropic square lattice Heisenberg magnet, while more pronounced influence of interlayer coupling is observed in [Cu(dmen)₂][Pt(CN)₄].

Full text of DOI:10.1016/j.jssc.2006.03.033

ARTICLE IN PRESS
Journal of Solid State Chemistry 179 (2006) 1965–1976
www.elsevier.com/locate/jssc
Low-dimensional compounds containing cyano groups. XIV. Crystal
structure, spectroscopic, thermal and magnetic properties of
[CuL2][Pt(CN)4] complexes (L ¼ ethylenediamine or
N,N-dimethylethylenediamine)
a,
Ã
a
b
b
b
ˇ
ˇ
ˇ
´
´
´
´
´ ˇ ´
ˇ beta Orendacova ,
Ivan Potocnak , Martin Vavra , Erik Ciz
ˇ
mar , Katarına Tibenska , Alz
ekd, Karla Fejfarovad, Harry Schmidtc,
Thomas Muller , Martin Orenda
ˇ b
, Alexander Feherb
Dirk Steinbornc, Christoph Wagnerc, Michal Dus
ˇ
´
c
+
´
c
a
ˇ
Department of Inorganic Chemistry, Institute of Chemistry, P.J. Safarik University, Moyzesova 11, SK-04154 Kosice, Slovakia
´
ˇ
b
ˇ
´
ˇ
Centre of Low Temperature Physics of the Faculty of Science of P.J. Safarik University and IEP SAS, Park Angelinum 9, SK-04154 Kosice, Slovakia
cInstitute of Inorganic Chemistry, Martin-Luther-University, Halle-Wittenberg, Kurt-Mothes-Street 2, D-06120 Halle, Germany
dInstitute of Physics, Na Slovance 2, CZ-182 21 Praha 8, Czech Republic
Received 28 February 2006; accepted 24 March 2006
Available online 1 April 2006
Abstract
Violet crystals of [Cu(en)2][Pt(CN)4] and blue crystals of [Cu(dmen)2][Pt(CN)4] were crystallized from the water–methanol
solution containing CuCl2 ꢀ 2H2O, ethylenediamine (en) or N,N-dimethylethylenediamine (dmen) and K2[Pt(CN)4] ꢀ 3H2O.
Both compounds were characterized using elemental analysis, infrared and UV-VIS spectroscopy, magnetic measurements,
specific heat measurements and thermal analysis. X-ray structure analysis revealed chain-like structure in both compounds.
The covalent chains are built of Cu(II) ions linked by [Pt(CN)4]2ꢁ anions in the [111] and [101] direction, respectively. The Cu(II)
atoms are hexacoordinated by four nitrogen atoms in the equatorial plane from two molecules of bidentate ligands L with average
˚
Cu–N distance of 2.022(2) and 2.049(4) A, respectively. Axial positions are occupied by two nitrogen atoms from bridging
[Pt(CN)4]2ꢁ anions at longer Cu–N distance of 2.537(2) and 2.600(5) A, respectively. Both materials are characterized by the presence of
˚
weak antiferromagnetic exchange coupling. Despite the one-dimensional (1D) character of the structure, the analysis of magnetic
properties and specific heat at very low temperatures shows that [Cu(en)2][Pt(CN)4] behaves as two-dimensional (2D) spatially
anisotropic square lattice Heisenberg magnet, while more pronounced influence of interlayer coupling is observed in
[Cu(dmen)2][Pt(CN)4].
r 2006 Elsevier Inc. All rights reserved.
Keywords: 1D crystal structure; [Pt(CN)4]2ꢁ anion; Hydrogen bonds; Thermal studies; Spectroscopic studies; Heisenberg square lattice
1. Introduction
and permits a wide number of binding modes in cyanide-
bridged complexes. By far the most commonly observed
binding modes are the terminal C-bond mode and the
linear bridged arrangement. Differently bended bridged
arrangements as well as various types of coordination to
the three metals are much less common [1].
Cyanide is an efficient ligand for the stabilization of the
transition metals in either low or high oxidation states.
This high electronic and coordinative versatility prompts
its terminal coordination to almost all transition metals,
Because tetracyanoplatinate(II) anions can use a differ-
ent number of cyano groups to bind to other metal atoms,
they exist in a wide range of solid-state structures. When all
Ã
Corresponding author. Fax: +421 55 62 221 24.
E-mail address: ivan.potocnak@upjs.sk (I. Potocnak).
ˇ ˇ ´
0022-4596/$ - see front matter r 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2006.03.033

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1966
four cyano groups are bound as monodentate ligands, it
leads to the separation of ion pairs [2,3] (Scheme 1a). One
structure has been reported very recently in which one of
the four cyano groups acts as a bridging ligand and the
resulting structure is a binuclear compound [4] (Scheme
1b). The bridging character of two cyano groups of
[Pt(CN)4]2ꢁ anion has been shown to yield several types
of structures in the solid state. When two bridging cyano
groups are in cis-arrangement three [5] (Scheme 1c) or four
nuclear molecular compounds [6] (Scheme 1d) and zigzag
chains [7] (Scheme 1e) are formed. When two bridging
cyano groups are in trans-arrangement molecular [8]
(Scheme 1f) or one-dimensional (1D) [9] (Scheme 1g)
compounds are produced. According to our knowledge,
only one compound exists containing two bridging cyano
groups both in cis- and trans-arrangement resulting in
three-dimensional (3D) structure [10] (Scheme 1h). Re-
cently, the first example of three-dimensional structure in
which three of the four cyano groups bridge two metal
atoms was reported [11] (Scheme 1i). When all four cyano
ligands are bridging, a molecular five nuclear complex
cation [4] (Scheme 1j) or 3D structures [12,13] (Scheme 1k)
can be isolated.
2. Experimental
2.1. Materials
Copper chloride dihydrate (CuCl2 ꢀ 2H2O), en (C2H8N2)
and dmen (C4H12N2) were of Merck quality and used as
received. K2[Pt(CN)4] ꢀ 3H2O was prepared according to
the literature [20] using H2[PtCl6] ꢀ 6H2O from Chempur as
starting platinum containing material.
2.2. Synthesis
[Cu(en)2][Pt(CN)4] (1). Into stirring water–methanol
solution (1:1) of CuCl2, ethylenediamine (en) (4 mmol)
was added in one portion and after 30 min, aqueous
solution of K2[Pt(CN)4] (1 mmol) was added in one
portion, too. In a few days, violet crystals of 1 were
filtrated off and dried on air. Yield 82%. Anal. Calc. for
C8H16N8CuPt: C, 19.90; H, 3.34; N, 23.19. Found: C,
20.21; H, 3.79; N, 22.87%.
[Cu(dmen)2][Pt(CN)4] (2). Similar to the synthesis of 1,
into stirring water–methanol solution (1:1) of CuCl2
Recently, cyano complexes with various degrees of
dimensionality have often been the subject of magnetic
studies. In this case, the cyano group or cyano complex
anion in addition to its structural function also exhibits an
important electronic function: it forms an exchange path
mediating the interaction among spins localized on
paramagnetic centers. Dimensionality of a magnetic sub-
system is fundamental for cooperation processes, but it
does not mean that the magnetic and structural dimension-
ality must be the same. The increase of magnetic
dimensionality can be caused by the influence of hydrogen
bonds (HBs), p–p and also dipole–dipole interactions
[14,15].
2-
N
C
N
C
M
N
C
Pt
C N
N
C
Pt
C N
C
N
C
N
(1a)
(1b)
N
N
C
C
C
Pt
C
Previous studies on the structure and magnetic proper-
ties of [Cu(en)2][Ni(CN)4] [16,17] and also the isostructural
[Cu(en)2][Pd(CN)4] [18] revealed that despite the 1D
character of their structure they behave as two-dimensional
(2D) magnets. 2D character of the magnetic subsystem can
be explained by the formation of HBs linking the
neighboring chains and thus serving as exchange paths
for magnetic interactions. The compounds with general
formula [Cu(L)2][Ni(CN)4], where L ¼ N-methylethylene-
diamine or N,N-dimethylethylenediamine (dmen) were
prepared [19] with the aim to suppress the formation of
HBs. However, their study has shown that the formation of
HBs was reduced only partially. Therefore, we have
decided to change [Ni(CN)4]2ꢁ anion by larger [Pt(CN)4]2ꢁ
anion with the expectation to suppress or at least to modify
existing HBs. In this paper, we describe the preparation
of [Cu(en)2][Pt(CN)4] (1) and [Cu(dmen)2][Pt(CN)4] (2)
complexes along with their characterization using infrared
(IR) and UV-VIS spectroscopy, thermal and X-ray
crystal structure analyses, magnetic and specific heat
measurements.
N
N
N
M
M
M
M
N
N
N
C
C
C
C
C
C
C
Pt
Pt
C
N
N
N
N
N
(1c)
(1d)
N
C
N
C
C
C
C
Pt
M
N
C
Pt
C
N
M
C
N
N
N
N
M
M
(1e)
(1f)
Scheme 1. Bonding modes of [Pt(CN)4]2ꢁ anion. Dotted lines represent a
propagation of the molecular or ionic structure to 1D (Scheme 1e and g)
or 3D, structure (Scheme 1h, i and k).

ARTICLE IN PRESS
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
ˇˇ ´
1967
N
C
N
N
C
Pt
C
M
N
M
C
N
(1g)
N
N
N
M
M
N
M
N
C
N
N
N
N
C
C
C
C
C
C
Pt
N
C
Pt
C N
M
Pt
C
C
C
N
N
N
N
N
M
M
(1h)
(1i)
M
n+
M
N
C
N
C
N
M
N
C
Pt
C
N
M
M
N
C
Pt
C
N
M
N
C
N
C
N
M
M
(1j)
(1k)
Scheme 1. (Continued)
(0.5 mmol), dmen (2 mmol) was added and finally after
30 min aqueous solution of K2[Pt(CN)4] (0.5 mmol) was
added. As a result a blue precipitate was immediately
formed. This precipitate was dissolved by addition of
20 mL of concentrated ammonia solution to the reaction
mixture. After 3 days, blue crystals of 2 were filtrated
off and dried on air. Yield 93%. Anal. Calc. for
C12H24N8CuPt: C, 26.74; H, 4.49; N, 20.78. Found: C,
26.68; H, 5.05; N, 20.60%.
Elmer UV/VIS/NIR Spectrometer Lambda 19. The ther-
mal investigations were performed using NETZSCH STA
409C/CD thermal analyzer under air conditions with a
heating rate of 10 1C/min to 900 1C. The electron spin
resonance (ESR) experiment has been conducted at low
temperatures in a home-built ESR spectrometer operating
at 10 GHz. DPPH (2,2-diphenyl-1-picrylhydrazyl) was used
as g-marker. Measurements of magnetic susceptibility and
magnetization have been carried out on powdered samples
using a commercial SQUID magnetometer in the tempera-
ture range from 2 to 300 K and in magnetic fields up to 7 T.
The background contribution from the gelcap and core
diamagnetic contribution to the magnetic moment of the
sample have been subtracted. Specific heat of both
materials was studied in the temperature range from
95 mK to 2.1 K in zero magnetic field using a dual-slope
technique [21] in dilution refrigerator TLE 2000 made by
2.3. Physical measurements
Elemental analyses for C, H and N were carried out
using an LECO CHNS-932. IR spectra were recorded on a
MATTSON 5000 FTIR spectrometer by the method of
KBr pellets in the range from 4000 to 400 cmꢁ1
.
Reflectance UV-VIS spectra were measured with Perkin

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1968
Oxford Instruments. RuO2 resistor of Dale type with the
nominal value of 1 kO calibrated against the commercial
Lake Shore thermometer GR200A-30 was used for
monitoring the temperature of the sample. The experi-
mental inaccuracy of the specific heat measurements was
less than 6%. The ESR and magnetic measurements have
been carried out on powdered samples, while in thermo-
dynamic experiments pelletized powder samples have been
used. In addition, a small needle-like single crystal of
complex 1 was used in the ESR investigations.
performed using SHELXL97 and PARST [25]. DIA-
MOND [26] was used for molecular graphics.
3. Results and discussions
3.1. Infrared spectroscopy
The IR spectra of 1 and 2 comprise bands confirming the
presence of all characteristic functional groups in the
prepared complexes. In accordance with the substitution of
two hydrogen atoms of one amino group in en (1) by two
methyl groups in dmen (2), the IR spectrum of 1 contains
more absorption bands originating from n(N–H) stretching
vibrations and less absorption bands of stretching n(C–H)
vibrations than the IR spectrum of 2. Moreover, d, o and
s(CH3) deformation vibrations are absent in 1. Other
bands confirming the presence of en and dmen in 1 and 2,
respectively, are listed in Table 2. Individual bands were
assigned according to [27]. The presence of [Pt(CN)4]2ꢁ
anion in the prepared complexes is proved by n(CꢂN)
stretching vibrations whose positions are an important tool
to distinguish between terminal and bridging character of
cyano group. This band is observed at 2080 cmꢁ1 in ionic
KCN [28]. Upon coordination to a metal the n(CꢂN) shift
to higher frequencies and the range of n(CꢂN) for
cyanoplatinates(II) with terminal cyano ligands extends
from 2120 to 2140 cmꢁ1 [29]. Because cyano nitrogen lone
pair resides in a mostly CꢂN antibonding orbital, an
increase of n(CꢂN) in bridging cyanides is found and for
bridged cyanoplatinates(II) it ranges from 2150 to
2210 cmꢁ1 [29]. According to this, the bands recorded at
2125 and 2130 cmꢁ1 in 1 and 2, respectively, can be
2.4. X-ray data collection and structure refinement
A summary of crystal data and structure refinement for 1
and 2 is presented in Table 1. The crystal structure of 1 was
determined using an Oxford Diffraction diffractometer
equipped with the CCD detector. Crysalis CCD [22]
was used for data collection while Crysalis RED [22] was
used for cell refinement, data reduction and absorption
correction.
The crystal structure determination of 2 was carried out
with Stoe Stadi-4 diffractometer. STADI4 [23] was used for
data collection and cell refinement, X-RED [23] was used
for data reduction. Both structures were solved by the
direct method with SHELXS97 [24] and subsequent
Fourier syntheses using SHELXL97 [24]. Anisotropic
displacement parameters were refined for all non-H
atoms. The H atoms were placed in calculated positions
and refined riding on their parent C or N atoms with
˚
C–H distances of 0.97 for methylene and of 0.96 A for
˚
methyl H atoms and of 0.90 A for amino H atoms, and
Uiso(H) ¼ 1.2Ueq(C or N). A geometric analysis was
Table 1
Crystal data and structure refinement of 1 and 2
Empirical formula
Formula weight
Temperature
C8H16CuN8Pt
482.92
120 K
C12H24CuN8Pt
539.01
293(2) K
˚
0.71073 A
Triclinic
¯
P1
˚
0.71073 A
Wavelength
Crystal system
Space group
Monoclinic
P21/n
˚
˚
a ¼ 6.5254(3) A, a ¼ 106.919(5)1
a ¼ 6.7819(8) A
˚
˚
Unit cell dimensions
b ¼ 7.2159(5) A, b ¼ 91.223(4)1
b ¼ 13.8444(17) A, b ¼ 97.618(11)1
˚
˚
c ¼ 8.1024(4) A, g ¼ 106.881(5)1
c ¼ 9.2152(10) A
3
3
˚
346.93(3) A
˚
857.59(17) A
Volume
Z
1
2
Density (calculated)
Absorption coefficient
F(000)
2.311 g/cm3
2.087 g/cm3
11.598 mmꢁ1
227
9.396 mmꢁ1
518
Crystal size
y range for data collection
Index ranges
0.27 ꢃ 0.12 ꢃ 0.06 mm3
0.35 ꢃ 0.35 ꢃ 0.19 mm3
3.28–27.491
2.67–24.991
ꢁ8php8, ꢁ9pkp9, ꢁ10plp10
6484
ꢁ8php8, 0pkp16, ꢁ10plp10
2995
Reflections collected
Independent reflections
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I42s(I)]
R indices (all data)
Largest diff. peak and hole
1564 [R(int) ¼ 0.0197]
Full-matrix least-squares on F2
1564/0/85
1508 [R(int) ¼ 0.0378]
Full-matrix least-squares on F2
1508/0/104
1.115
R1 ¼ 0.0120, wR2 ¼ 0.0275
R1 ¼ 0.0120, wR2 ¼ 0.0275
1.100
R1 ¼ 0.0265, wR2 ¼ 0.0743
R1 ¼ 0.0327, wR2 ¼ 0.0803
3
3
˚
˚
0.634 and ꢁ0.922 e/A
1.532 and ꢁ1.606 e/A

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1969
Table 2
IR spectra of 1 and 2
1
[Cu(en)2][Pt(CN)4] (1)
[Cu(dmen)2][Pt(CN)4] (2)
n(N–H)
n(C–H)
3335 vs
3312 vs
3281 vs
3254 sh
3160 m
2982 w
2961 m
2932 m
2878 w
3337 vs
3273 vs
3143 m
2
3028 m
2973 m
2930 sh
2895 m
2854 sh
2816 m
2130 vs
2088 w, sh
1584 s
n(CꢂN)
2125 vs
2089 w, sh
1597 s
d(NH2)
d(CH2)
1467 w
1456 w
d(CH3)
o(CH2)
o(NH2)
t(CH2)
o(CH3)
o(NH2)
t(CH3)
1462 s
1366 w
1323 w
1289 m
1248 w
1190 m
1134 vs
1116 sh
1063 s
1002 s
940 s
1370 w
1324 m
1277 m
250
350
450
Wavelength (nm)
550
Fig. 1. UV-VIS spectra of 1 and 2.
1165 m
t(NH2)
n(C–N)
n(C–C)
s(CH2)
s(NH2)
s(CH3)
1088 m
1033 vs
975 s
six-coordinated deformed octahedron with CuN6 chromo-
phoric group. The spectrum of 2 also consists of one
broadband with the maximum at 435 nm and is assigned to
2B1g-2Eg transition in consistence with CuN6 chromo-
phore, likewise in 1.
881 w
725 s
894 m
783 s
654 s
d(chelate ring)
n(Pt–C)
536 m
499 w
473 w
402 m
552 m
502 m
476 w
401 s
3.3. Thermal studies
Thermal decomposition of 1 and 2 is a multistage
process. The thermal decomposition of 1 consists of two
distinguished stages corresponding to the separated release
of en molecules and cyano groups decomposing. Complex
1 is stable up to 246 1C. In the 246–417 1C temperature
range, a weight loss of 25.3% corresponding to the release
of two en molecules (calculated 24.9%) is observed during
endothermic two-stage process (these stages are over-
lapped). The second step in the 417–448 1C temperature
range, with observed weight loss of 21.2%, involves the
decomposition of cyanides (calculated 21.6%). Their
decomposition includes oxidation of four CNꢁ groups to
form two cyanogen molecules during reduction of one
Pt(II) atom and one half of oxygen molecule from the air
ambient. This process is accompanied by a strong
exothermic effect as a consequence of the cyanogen
molecules pyrolysis. The final thermal decomposition
product is a mixture of CuO and metallic Pt (solid residue
56.8%; calculated 56.9%). The most probable thermal
decomposition scheme for 1 is
n—stretching, d—scissoring, o—wagging, t—twisting, s—rocking vibra-
tions, vs—very strong, s—strong, m—middle, w—weak vibrations, sh—
shoulder.
assigned to n(CꢂN) stretching vibrations of bridging cyano
groups, while those at 2089 and 2088 cmꢁ1 in 1 and 2,
respectively, to terminal cyano groups. Surprisingly,
although there are two bridging and two terminal cyano
groups in the structures of 1 and 2, the later bands are
much weaker. On the other hand, in the case of
[Cu(en)2][Ni(CN)4] [17,30] and [Cu(dmen)2][Pd(CN)4] [31]
compounds only one n(CꢂN) absorption band has been
recorded. The same was observed in the spectra of
[Cu(L)2][Ag2(CN)4] compounds where L ¼ en [32] or 1,2-
diaminopropane [33] although both compounds contain
both bridging and terminal cyano groups.
3.2. UV-VIS spectroscopy
246ꢁ417 ꢅ
½CuðenÞ2ꢄ½PtðCNÞ4ꢄ ꢁ! C CuPtðCNÞ4
ꢁ2en
The reflectance UV-VIS spectra of 1 and 2 (Fig. 1) were
taken as described earlier. A broad asymmetric band with a
maximum at 354 nm, observed in the spectrum of 1, which
could be assigned to 2B1g-2Eg transition is supportive of a
417ꢁ448 ꢅC ðO2Þ
ꢁ!
CuO þ Pt:
ꢁ2ðCNÞ2

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1970
The thermal decomposition of 2 is similar to the thermal
Table 3
Selected bond lengths (A) and bond angles (deg) of 1
˚
decomposition of 1 and consists of two distinguished steps.
In the first endothermic two-stage step a weight loss of
30.3% corresponding to the release of two dmen molecules
(calculated 32.7%) is observed in the 193–427 1C tempera-
ture range. The decomposition proceeds continuously into
the second strongly exothermic step which is finished at
461 1C with observed weight loss of 22.3% which
corresponds to the decomposition of cyanides (calculated
19.3%). Their decomposition includes the oxidation of
four CNꢁ groups to form two cyanogen molecules during
the reduction of one Pt(II) atom and one half of oxygen
molecule from the air ambient. The final thermal decom-
position product is a mixture of CuO and metallic Pt (solid
residue 51.5%; calculated 51.0%). The most probable
thermal decomposition scheme for 2 is
Pt–C2
Pt–C1
1.994(2)
2.000(2)
C2–N2
N1–C1
1.153(3)
1.154(3)
N2–Cu
Cu–N4
Cu–N3
2.5379(19)
2.0170(19)
2.0274(17)
87.75(8)
C2–Pt–C1
N2–C2–Pt
N1–C1–Pt
C2–N2–Cu
N4–Cu–N3
N4–Cu–N2
N3–Cu–N2
177.1(2)
176.01(19)
120.64(15)
84.39(7)
91.44(7)
87.80(7)
193ꢁ427 ꢅ
½CuðdmenÞ2ꢄ½PtðCNÞ4ꢄ ꢁꢁ2d!menC CuPtðCNÞ4
Table 4
Hydrogen bonds of 1 (A and deg)
˚
427ꢁ461 ꢅC ðO2Þ
ꢁ!
CuO þ Pt:
D–H?A
D–H
H?A
D?A
D–H?A
ꢁ2ðCNÞ2
N4–H4A?N1
N3–H3B?N2i
N4–H4B?N1ii
0.90
0.90
0.90
2.38
2.42
2.31
3.250(3)
3.220(3)
3.126(3)
161.6
147.6
151.2
3.4. X-ray crystallography
Symmetry transformations used to generate equivalent atoms: i: ꢁx,
ꢁy+1, ꢁz, ii: ꢁx+1, ꢁy, ꢁz.
X-ray analysis has revealed that the crystal structure of 1
consists of 1D zigzag chains running in [111] direction. An
ORTEP view of the complex 1 is depicted in Fig. 2 with the
atomic numbering scheme. In the structure, both Cu(II)
and Pt(II) ions occupy sites of inversion centers. The Cu(II)
ion is six-coordinate, complexed with the four nitrogen
atoms of en ligands in the equatorial plane and the two
nitrogen atoms from bridging cyano groups in the axial
positions, respectively. The Cu–N bond distances in
matter whether they are located in the equatorial plane or
in the axial positions, are 180.01. By contrast, the Pt(II) ion
possesses a square planar coordination surrounding in
which it is coordinated to the four carbon atoms from four
cyano groups. Despite the division of Pt–C bonds into two
pairs, depending on CN groups (bridging or terminal) their
bond lengths are similar with the average value of
˚
the equatorial plane have an average value of 2.022(7) A
(Table 3), while weak axial interactions, a characteristic of
˚
1.997(4) A. Alike the Cu(II) atom, the C–Pt–C bond angles
are of 180.01 due to the presence of the inversion center.
Two of the four cyano groups from a tetracyanoplatina-
te(II) ion are connected to two different Cu(II) moieties in
a trans mode that leads to a zigzag chain with the bond
angle of 120.6(2)1 for the Cu–N2–C2. The interchain HBs
N3–H3B?N2i and N4–H4B?N1ii with length of 3.220(3)
9
d systems, give rise to a Cu–N bond length of 2.537(2) A.
˚
The Cu–N bonds in the axial positions are slightly tilted
from the normal to the CuN4 plane (2.741), forming a
slightly distorted and tetragonally elongated octahedral
geometry. Because of the inversion center, all N–Cu–N
bond angles formed by two face to face nitrogen atoms, no
˚
and 3.126(3) A, respectively (Table 4), connect the neigh-
boring chains resulting in a 3D network. Moreover, the
˚
N1
N4–H4A?N1 distance of 3.250(3) A indicates the HB
N2
interaction within the chain. The HB system in 1 is shown
in Fig. 3. If we compare the HB system in 1 with the HB
system observed in the reference [Cu(en)2][Ni(CN)4]
compound [17] we can see that the HB system in 1 was
not reduced because both the intrachain HB and the
interchain HB between N3 and N2i atoms connecting the
chains into sheets are present in the reference compound,
too. Moreover, interchain HB between N4 and N1ii atoms
connecting the sheets in the 3D network in 1 was not
observed in the reference compound at all.
N4
C1
N3
C2
C3
Cu
C4
C4
C3
Pt
N3
Cu
N4
C2
N2
C1
N2
N1
Complex 2 exhibits an analogous solid-state structure as
complex 1, see Fig. 4. The Cu(II) ion also exhibits a slightly
Fig. 2. The crystal structure of 1 with displacement ellipsoids (75%
probability) and atom labeling.

ARTICLE IN PRESS
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
ˇˇ ´
1971
N1
Pt
C
H4A
H4B
N
N2
N4
Cu
H
N3
H3B
a
c
b
Fig. 3. HB system in 1 viewed along the chain direction. The HBs connecting chains into sheets are represented by black short dashed lines, the HBs
connecting sheets into 3D structure are represented by black long dashed lines while intrachain HBs are represented by empty dashed lines. Only hydrogen
atoms of amino groups are shown because of clarity.
C6
N3
Table 5
Selected bond lengths (A) and bond angles (deg) of 2
˚
N2
C5
Cu
N1
C1
Pt
C2
N4
N2
Cu–N4
Cu–N3
1.984(4)
2.115(4)
2.600(5)
1.989(6)
1.996(5)
1.146(8)
1.147(6)
85.00(16)
88.55(17)
93.00(15)
90.0(2)
C4
Cu
N2
Cu–N2
Pt–C2
C1
N1
C3
C2
Pt–C1
C2–N2
C1–N1
C4
C3
N3
N4
N4–Cu–N3
N4–Cu–N2
N3–Cu–N2
C2–Pt–C1
N2–C2–Pt
N1–C1–Pt
C2–N2–Cu
C5
C6
Fig. 4. The crystal structure of 2 with displacement ellipsoids (45%
probability) and atom labeling.
179.3(5)
179.0(4)
137.5(5)
distorted (the tilting angle between the apical axis and the
normal to the CuN4 plane is 3.461) and tetragonally
elongated octahedral coordination geometry. In contrast to
1, the Cu–N bond distances in the equatorial plane are not
Table 6
Hydrogen bonds of 2 (A and deg)
˚
˚
of the same length as the Cu–N3 bond of 2.115(4) A is
D–H?A
D–H
0.90
H?A
2.61
D?A
D–H?A
˚
significantly longer than the Cu–N4 bond of 1.984(4) A.
This difference is caused by the steric repulsion of two
methyl groups bonded to nitrogen N3 atom. The Cu–N
N4–H4A?N1i
3.351(6)
140.0
Symmetry transformation used to generate equivalent atoms: i: ꢁx+2,
ꢁy+1, ꢁz+1.
˚
bond distances in the axial positions are 2.600(5) A due to
Jahn–Teller effect. The bond distances and angles for 2 are
given in Table 5. Pt(II) ion has the same coordination
environment as in 1 with the average Pt–C bond lengths of
type of the HB system was observed also in the reference
[Cu(dmen)2][Ni(CN)4] compound [19].
˚
1.993(5) A. The basic features of the molecular structure of
2 are the same as that of 1 consisting of zigzag chains
running in [101] direction with the Cu–N2–C2 bond angle
of 137.5(5)1. Similar to 1, the interchain HB N4–H4A?N1
3.5. Magnetic properties
˚
with length of 3.351(6) A (Table 6), connects the chains
The ESR spectra obtained at 4.2 K on powdered samples
have been analyzed within a simple model including
dipolar interactions, magnetic field and isotropic exchange
coupling [34]. The best fit to the experimental data using
into 2D sheets (Fig. 5). However, the formation of
intrachain HB is suppressed due to the presence of two
methyl groups bonded on nitrogen N3 atom. The same

ARTICLE IN PRESS
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
ˇˇ ´
1972
N1
N4
H4A
b
Cu
N
H
C
a
c
Pt
Fig. 5. HB system in 2 viewed along the chain direction. The HBs connecting chains into sheets are represented by black short dashed lines. Methyl groups
are omitted and only hydrogen atoms of N4 amino group are shown because of clarity.
(also in Table 7) are consistent with the axial distortion of
[Cu(en) ][Pt(CN) ]
1.0
0.8
0.6
0.4
0.2
0.0
2
4
the coordination octahedron and are typical for spin 1/2
Cu(II) ions. The deviations between the fit and data might
be attributed to the low-dimensional character of the
magnetic subsystem and/or orthorhombic anisotropy of
the g-factor resulting from a deformation of the local
octahedron.
The angular dependence of g-factor of the complex 1
shown in Fig. 7 has been obtained from resonance fields of
ESR spectra measured on a single crystal. Magnetic field
was oriented within the a0c0 plane. The c0 direction is the
longest edge of the crystal, while a0 is a shortest edge
approximately perpendicular to the c0 direction. The
experimental data have been fitted to the formula
experiment
fit
[Cu(dmen) ][Pt(CN) ]
2
4
experiment
fit
DPPH
2
2
g2 ¼ gjj cos2ðy ꢁ y0Þ þ g?2 sin ðy ꢁ y0Þ,
(1)
0.30
0.32
0.34
B (T)
0.36
0.38
0.40
where y is an angle between the magnetic field and the local
z-axis and y0 is the angle between the c0 direction and local
z-axis (Fig. 7). The best agreement was achieved for the
Fig. 6. Comparison of the experimental ESR spectra of complex 1
(squares) and complex 2 (circles) with the simple theoretical prediction [34]
(dashed line for complex 1, while solid line is for complex 2). The values of
resonance fields and halfwidths obtained from the fits are in Table 7.
values of y0 ¼ 231, g|| ¼ 2.2470.01 and g ¼ 2.0770.01
?
close to the ones obtained from the analysis of powder
spectra. Additional X-ray experiments were performed to
find a proper orientation of the single crystal. Analysis of
the X-ray data provided the estimation of the angle
between the local z-axis and c0 direction, being equal to
about 201. This value is in good agreement with the angle
least-squares method (Fig. 6) has been obtained for the
values of resonance fields and the halfwidth of the
resonance line, DB presented in Table 7. Slightly aniso-
tropic g-factor values corresponding to the resonance fields

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1973
Table 7
Resonance fields and g-factor values obtained from powder ESR spectra
B|| (T)
B
(T)
DB (mT)
g||
g
?
?
[Cu(en)2][Pt(CN)4]
[Cu(dmen)2][Pt(CN)4]
0.334670.001
0.335070.001
0.353570.001
0.354570.001
2.070.2
1.970.2
2.2170.01
2.2070.01
2.1070.01
2.0870.01
[Cu(en) ][Pt(CN) ]
2
4
5.0
2.0
1.5
1.0
0.5
0.0
g
g
= 2.24
||
z
15
10
5
= 2.07
⊥
g = 2.14 0.01
4.8
c'
ꢀ =-0.73 0.10 K
4.6
4.4
xy
0
0
20
40
60
a'
T (K)
4.2
0
25
50
75
100
125
150
0
100
200
300
angle (deg)
T (K)
Fig. 7. Angular dependence of g2-factor value obtained from the ESR
spectra of single crystal of the complex 1 (full circles). The solid line
represents the least-squares fit to Eq. (1).
Fig. 8. Effective magnetic moment of complex 2. Inset shows the
temperature dependence of inverse susceptibility (circles) with a fit to
the Curie–Weiss law (solid line).
231 found from the fitting procedure using Eq. (1) and
corresponds to the position of the maximum g-factor value
found in the a0c0 plane. Consequently, the position of a
minimum g-factor value in the same plane might be
ascribed to the orientation within the equatorial xy-plane
of the local octahedron. The results confirm that the
electronic ground state of Cu(II) ion is described by a wave
function of dz2 symmetry and the exchange paths will
propagate along the directions determined by the lobes of
the dx2ꢁy2 orbital.
quently, only an estimate of the effective strength of
the exchange coupling zJ/kB, where z is the number of the
nearest neighbors, by fitting the Curie–Weiss law to the
experimental data in the temperature range 2–75 K is
obtained. The fit yields the values zJ/kB ¼ ꢁ1.6 K and
g ¼ 2.15 for complex 1 [35], while zJ/kB ¼ ꢁ2.92 K and
g ¼ 2.14 for complex 2 (inset in Fig. 8).
In the work [35], it has been shown that the field
dependence of the magnetization of the complex 1 is close
to the saturation at 7 T and indicates a free spin behavior at
2 K, while the magnetization of complex 2 measured at
4.75 K shows no saturation up to 6 T (Fig. 9). The
comparison of the experimental data with the free spin
model suggests the presence of weak antiferromagnetic
exchange coupling. The results of magnetization measure-
ments are consistent with the estimation of the effective
exchange coupling from the inverse susceptibility, where
the magnitude of effective exchange coupling in complex 2
is higher than in complex 1.
Magnetic susceptibilities of powdered samples of both
materials measured at 0.1 T were corrected for the
diamagnetic contribution of the materials which was
estimated, using Pascal’s constants, to be wDIA
ꢁ2.9586 ꢃ 10ꢁ9 m3/mol for complex 1, while wDIA
¼
¼
ꢁ2.396 ꢃ 10ꢁ9 m3/mol for complex 2. From susceptibility
at T ¼ 300 K, the effective magnetic moment may be
quantified and yields values that are typical for a Cu(II)
cation with d9 configuration, namely meff/mB ¼ 1.95 for
complex 1 [35] and meff/mB ¼ 1.99 for complex 2 (Fig. 8).
No difference between the susceptibility of zero-field
cooled (ZFC) and field-cooled sample has been observed
down to 2 K for both materials, which suggests that no
long-range order appears above 2 K. The temperature
dependence of the susceptibility is characterized by Curie-
like behavior without the short-range order maximum
expected at low temperatures for the low-dimensional
system with antiferromagnetic exchange coupling. Conse-
The specific heat of both materials has been investigated
in zero magnetic field in the temperature range
95 mK–2.0 K and 95 mK–0.80 K, for complex 1 and
complex 2, respectively, with the aim to resolve the
magnetic structure and dimensionality of the systems.
The temperature dependence of the magnetic specific heat
of complex 1 displays a round maximum at Tmax ¼ 0.28 K
and l-like anomaly, associated with long-range ordering, at
Tl ¼ 0.18 K, while the complex 2 shows only an indication

ARTICLE IN PRESS
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
ˇˇ ´
1974
Table 8
Magnetic entropy of studied materials calculated from experimental
specific heat
[Cu(dmen) ][Pt(CN) ]
2
4
4
3
2
1
0
experimental data atT=4.75 K
Brillouin function,g=2.14
Sm (J/K/mol)
Sm/Stheor
Entropy
removed
above Tl
[Cu(en)2][Pt(CN)4]
[Cu(dmen)2][Pt(CN)4]
5.75
6.1
0.91
1.06
4.61 (80%)
3.64 (60%)
values of the magnetic entropy, Sm, with respect to the
theoretically predicted value Stheor ¼ R ln(2S+1) ¼ 5.76 J/
K/mol for magnetic system with spin S ¼ 1/2 are shown in
Table 8. The magnetic entropy removed above Tl in both
materials represents more than 50% of the total magnetic
entropy suggesting low-dimensional character of the
studied systems. However, the magnetic dimensionality of
the complex 2 with 60% of total entropy removed above Tl
is more likely close to the 3D magnetic system.
0
1
2
3
4
5
6
B (T)
Fig. 9. Field dependence of the magnetization of complex 2 measured at
4.75 K (circles) compared to the Brillouin function for g ¼ 2.14 obtained
from Curie–Weiss fit of the inverse susceptibility (solid line).
Considering the topology of covalent and possible HBs
the magnetic specific heat of the complex 1 has been
compared to the theoretical models for Heisenberg
antiferromagnetic (HAF) linear chain [36] and isotropic
HAF square lattice [37,38] (Fig. 10). The values of
exchange coupling parameters obtained from the compar-
ison were J/kB ¼ ꢁ0.62 and ꢁ0.43 K for HAF linear chain
model and HAF square lattice model, respectively. These
values are in agreement with the estimation of the effective
strength of the exchange coupling from the Curie constant.
The height of the maximum at Tmax ¼ 0.28 K suggests that
the system might be described by the 2D model of the
spatially anisotropic square lattice [39], where linear chains
are coupled by the interchain exchange coupling J0. This
model coincides well with the geometry of the HBs net in
the structure of complex 1, i.e. intrachain interaction J
might be mediated by doubled Cu–N4–H4A?
N1?H4B–N4–Cu exchange path, while J0 runs along
Cu–N3–H3B?N2–Cu pathway. However, the theoretical
prediction for the temperature dependence of the magnetic
specific heat of the aforementioned model is not available
yet, thus, we are unable to estimate the interchain coupling
J0 in complex 1 from the measured data set.
1.2
[Cu(dmen) ][Pt(CN) ]
2
4
[Cu(en) ][Pt(CN) ]
2
4
1.0
0.8
0.6
0.4
0.2
0.0
square lattice, J/k = -0.43 K
B
linear chain, J/k = -0.62 K
B
quasi-two-dimensional lattice
J/k = -0.27 K, J /J = 0.5
B
⊥
0.1
1
T (K)
Fig. 10. Magnetic specific heat of complex 1 compared to the HAF square
lattice model with J/kB ¼ ꢁ0.43 K (dotted line), while dashed line
represents the HAF linear chain model with J/kB ¼ ꢁ0.62 K. The solid
line represents quasi-two-dimensional square lattice HAF model with
intralayer J/kB ¼ ꢁ0.27 K and interlayer J /J ¼ 0.5 describing the
?
behavior of magnetic specific heat of the complex 2.
The magnetic specific heat of the complex 2 can be
described by the theoretical model for quasi-two-dimen-
sional HAF square lattice [40] with in-plane coupling J and
of the round maximum overlapped with a dominating l-
like anomaly at Tl ¼ 0.2 K (Fig. 10). The lattice contribu-
tion to the total specific heat of both materials was
subtracted after fitting high temperature region to the
equation C ¼ aTꢁ2 þ bT3, where aTꢁ2 describes the high-
temperature expansion of magnetic specific heat, while bT3
represents the low-temperature description of the lattice
contribution in the Debye approximation. After subtrac-
tion of the lattice contribution, the magnetic entropy Sm
has been calculated from the experimental data with simple
approximations used for T-0 and T-N. The calculated
varying interplane coupling J oJ. From the comparison
?
of the experimental data to the theoretical model we
have obtained the values J/kB ¼ ꢁ0.27 K and J /J ¼ 0.5
?
(Fig. 10) suggesting strong influence of interplane coupling
in complex 2, driving the system into transition to the
three-dimensional long range order. This result is sup-
ported by the 60% ratio of the magnetic entropy removed
above the Tl. The strength of exchange coupling is also in
agreement with the estimation of effective exchange
coupling from the Curie–Weiss fit of the susceptibi-
lity. The difference between the experimental data and

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1975
theoretical prediction can be ascribed to the presence of
slightly stronger interlayer coupling J in complex 2 than
20-005204 and by the Grant Agency of the Czech Republic,
Grant 202/05/0421. One of the authors (M.V.) thanks
Socrates/Erasmus stipend program for financial support
?
that taken into account in the available theoretical
prediction and finite size calculations [40].
ˇ
and hospitality of Martin-Luther-University and E.C.
thanks DAAD for financial support and hospitality of
University of Bayreuth during the magnetic measurements.
The financial support of US Steel—DZ Energetika Kosice
ˇ
4. Conclusions
The X-ray analysis of violet crystals of [Cu(en)2]
[Pt(CN)4] (1) and blue crystals of [Cu(dmen)2][Pt(CN)4]
(2) revealed that both structures are formed by infinite
zigzag chains. The covalent chains are built of Cu(II) atoms
linked by [Pt(CN)4]2ꢁ anions in the [111] and [101]
direction, respectively. The Cu(II) atoms are hexacoordi-
nated by four nitrogen atoms in the equatorial plane from
two molecules of bidentate ligands L with average Cu–N
distance of 2.022(2) and 2.049(4), respectively. Axial
positions are occupied by two nitrogen atoms from
bridging [Pt(CN)4]2ꢁ anions at longer Cu–N distance of
is acknowledged.
Supplementary material
CCDC 299194 & 299195 contains the supplementary
crystallographic data for this paper. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.htmL (or from the CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033;
e-mail: deposit@ccdc.cam.ac.uk).
References
˚
2.537(2) and 2.600(5) A, respectively.
The chains are bound into sheets in both complexes by
relatively strong HBs of N–H?N type. Moreover, another
N–H?N HB is found between these sheets in 1 and thus
3D structure is finally formed. Contrary to the 3D
character of this structure the specific heat analysis suggests
that the magnetic structure in this complex can be
approximated by the spatially anisotropic square lattice
Heisenberg model with J/kB E ꢁ0.5 K similar as in the
compound [Cu(en)2][Ni(CN)4] [17]. However, the inter-
chain coupling J0 is weaker in complex 1. The 2D magnetic
structure can be explained by the fact that the magnetic
dx2ꢁy2 orbital is oriented within the equatorial plane of the
local octahedron due to the Jahn–Teller distortion and is
not involved in the covalent bonds. Consequently, the
exchange path between Cu(II) atoms is preferred through
HBs between the chains and not through the covalent
bonds within the chain.
The structure of 2 remains 2D as the second N–H?N
bond binding the neighboring sheets is missing. On the
other hand, the replacement of two hydrogen atoms by
methyl groups on N2 atom in en ligands supports the
creation of a dense network of weak C–H?N HBs
increasing the magnetic dimensionality. Consequently, the
2D magnetic behavior is strongly overlapped by interlayer
effects due to the weak HBs network.
[1] J. McCleverty, T. Meyer (Eds.), Comprehensive Coordination
Chemistry II, vol. 1, Elsevier, Amsterdam, 2004.
[2] P. Vitoria, J.I. Beitia, J.M. Gutierrez-Zorrilla, E.R. Saiz, A. Luque,
M. Insausti, J.J. Blanco, Inorg. Chem. 41 (2002) 4396.
[3] J.A. Brozik, B.L. Scott, B.I. Swanson, Inorg. Chim. Acta 294 (1999)
275.
[4] M.-L. Flay, H. Vahrenkamp, Eur. J. Inorg. Chem. (2003) 1719.
[5] I. Muga, J.M. Gutierrez-Zorrilla, P. Vitoria, P. Roman, F. Lloret,
Polyhedron 21 (2002) 2631.
[6] L.R. Falvello, M. Tomas, Chem. Commun. (1999) 273.
[7] M.M. Olmstead, M.A. Lee, J.R. Stork, Acta Crystallogr. E 61 (2005)
m1048.
[8] G.N. Richardson, U. Brand, H. Vahrenkamp, Inorg. Chem. 38
(1999) 3070.
[9] M. Munakata, J.C. Zhong, I. Ino, T. Kuroda-Sowa, M. Maekawa,
Y. Suenaga, N. Oiji, Inorg. Chim. Acta 317 (2001) 268.
[10] D.W. Knoeppel, S.G. Shore, Inorg. Chem. 35 (1996) 1747.
[11] D.W. Knoeppel, S.G. Shore, Inorg. Chem. 35 (1996) 5328.
[12] M.-L. Flay, H. Vahrenkamp, Z. Anorg. Allg. Chem. 630 (2004) 2508.
[13] R.W. Gable, B.F. Hoskins, R. Robson, Chem. Commun. (1990) 762.
[14] J.P. Zhang, R.S. Wang, M. Baumgarten, Mol. Cryst. Liq. Cryst. 306
(1997) 119.
[15] R. Boc
ˇ
a, Coord. Chem. Rev. 173 (1998) 167.
, A. Sopkova, J. Sivy, V. Kettmann, V. Vra
Acta Crystallogr. C 47 (1991) 2447.
[16] J. Lokaj, K. Gyoryova
´
´
´
´
bel,
¨
ˇ
´ ˇ ´ ˇ ´ ´
[17] M. Orendac, A. Orendacova, J. Cernak, A. Feher, Solid State
Commun. 94 (1995) 833.
ˇ
[18] J. Cerna
A. Orenda
[19] J. Kuchar, J. Cerna
Phenom. 90 (2003) 323.
[20] G.B. Kaufmann, Inorg. Synth. 9 (1967) 182.
[21] S. Riegel, G. Webber, J. Phys. E 19 (1986) 790.
´
k, J. Skors
˘
epa, K.A. Abboud, M.W. Meisel, M. Orendac,
´ ˇ
´
c
ˇ
ova, A. Feher, Inorg. Chim. Acta 326 (2001) 3.
´
ˇ
ˇ
k, Z. Mayerova, P. Kubacek, Z. Zak, Solid State
´ ´ ´ ˇ ´
´
Finally we can conclude that although the use of bulky
[Pt(CN)4]2ꢁ anion instead of [Ni(CN)4]2ꢁ anion resulted in
a different HBs system, HBs were not suppressed at all.
Therefore, we intend to use in our next work other ligands
L which are definitely unable to form strong hydrogen
bridges, like 1,10-phenanthroline, 2,20-bipyridine or
N,N,N0,N0-tetramethylethylenediamine.
[22] Oxford Diffraction, Crysalis CCD and Crysalis RED, Oxford
Diffraction Ltd., Oxford, UK, 2004.
[23] Stoe & Cie, STADI4 (Version 1.06) and X-RED (Version 1.07), Stoe
& Cie, Darmstadt, Germany, 1996.
[24] G.M. Sheldrick, SHELXS97 and SHELXL97, University of Gottin-
¨
¨
gen, Gottingen, Germany, 1997.
[25] M. Nardelli, J. Appl. Crystallogr. 28 (1995) 659.
[26] K. Brandenburg, DIAMOND (Release 2.1e) Crystal Impact GbR,
Bonn, Germany, 2000.
Acknowledgments
This work was supported by the grants of the Slovak
Grant Agency VEGA Nos. 1/2470/05 and 1/3027/06, by
Research and Development Support Agency APVV No.
[27] Z. Gabelica, Spectrochim. Acta 32A (2) (1976) 337.
[28] K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Part B: Applications in Coordination,

ARTICLE IN PRESS
ˇˇ ´
I. Potocnak et al. / Journal of Solid State Chemistry 179 (2006) 1965–1976
1976
Organometallic, and Bioinorganic Chemistry, fifth ed., Wiley, New
York, 1997, p. 105.
[29] A.M. Golub, H. Kohler, V.V. Skopenko, Chemistry of Pseudoha-
¨
[34] J.A. Ibers, J.D. Swalen, Phys. Rev. 127 (1962) 1914.
ˇ
ˇ
[35] K. Tibenska
M. Orenda
M.W. Meisel, Phys. Stat. Sol. B 243 (2006) 281.
[36] L.J. de Jongh, A.R. Miedema, Adv. Phys. 23 (1974) 1.
´
, A. Orenda
´
c
ˇ
ova
´
, E. Ciz
ˇ
ma
´
r, J.-H. Park, S. Lanc
aric,
ˇ ˇ
´
c
ˇ
, A.G. Anders, M. Vavra, I. Potoc
k, A. Feher,
˘
na
ˇ ´
lides, Elsevier, Amsterdam, 1986.
[30] K. Seits, S. Peschel, D. Babel, Z. Anorg. Allg. Chem. 627 (2001) 929.
ˇ
[31] J. Kucha
´
r, J. Cerna
k, J. Chomic
, A. Feher, C. Kappenstein, Inorg. Chim. Acta 281 (1998) 134.
[33] L’. Tris kova, I. Potoc k, J. Chomic, P. Baran, Trans. Met. Chem.
(2006), accepted for publication.
´
k, K.A. Abboud, Acta Crystallogr. C 60 (2004) m492.
[37] M.S. Makivic, H.-Q. Ding, Phys. Rev. B 43 (1991) 3562.
[38] B. Bernu, G. Misguich, Phys. Rev. B 63 (2001) 134409.
[39] Y.J. Kim, R.J. Birgenau, Phys. Rev. B 62 (2000) 6378.
[40] P. Sengupta, A.W. Sandvik, R.R.P. Singh, Phys. Rev. B 68 (2003)
094423.
ˇ
[32] J. Cerna
Kova
´
ˇ
, P. Gravereau, A. Orenda ova, M. Orenda , J.
´
c
ˇ
´
c
´ ˇ
c
´ ˇ
˘
c
ˇ
ı
´
´
ˇ
n
ˇ
a
´
ˇ