Mercury(II) complexes of heterocyclic thiones. Part 1. Preparation of 1:2 complexes of mercury(II) halides and pseudohalides with 3,4,5,6-tetrahydropyrimidine-2-thione. X-ray, thermal analysis, and NMR studies

Article abstract of DOI:10.1016/S0020-1693(00)00162-6

A series of complexes HgX₂(H₄pymtH)₂ (X = Cl^-, Br^- I^-, SCN^-, CN^- ; H₄pymtH = 3,4,5,6-tetrahydropyrimidine-2-thione) has been obtained by the reaction of H₄pymtH with mercury(II) halides and pseudohalides in the molar ratio 2:1. X-ray diffraction studies revealed tetrahedral coordination of mercury with S-bound H₄pymtH. The exception is Hg(CN)₂(H₄pymtH)₂ where the coordination is 2 + 2 with two strongly bound CN^- ligands and weaker H-S bonds with H₄pymtH. One- and two-dimensional ¹H and ¹³C NMR measurements in dimethylsulfoxide solution confirmed the complexation of mercury to sulfur. The greatest complexation effects on chemical shifts were detected for the C-2, C-5 and H-1,3 atoms, i.e. two, four and five bonds away from mercury atom. The complexation effects in Hg(SCN)₂(H₄pymtH)₂ and Hg(CN)₂(H₄pymtH)₂ are in agreement with the strongest intramolecular H-bonding in the former and the weakest Hg-S bonds in the latter as compared to other complexes here. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0020-1693(00)00162-6

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 306 (2000) 142–152
Mercury(II) complexes of heterocyclic thiones.
Part 1. Preparation of 1:2 complexes of mercury(II) halides and
pseudohalides with 3,4,5,6-tetrahydropyrimidine-2-thione. X-ray,
thermal analysis and NMR studies
a,
a
a
a
&
&
Zora Popovic´ *, Gordana Pavlovic´ , Dubravka Matkovic´-Calogovic´ , Zeljka Soldin ,
Masˇa Rajic´ , Drazˇen Vikic´-Topic´ c, Damir Kovacˇek d
b
a Laboratory of General and Inorganic Chemistry, Chemistry Department, Faculty of Science, Uni6ersity of Zagreb, Ul. kralja Z6onimira 8,
HR-10000 Zagreb, Croatia
b Laboratory for Thermal Analysis, Brodarski Institute-Marine Research and Special Technology, A6. V. Holje6ca 20,
HR-10000 Zagreb, Croatia
-
c Ruder Bosˇko6ic´ Institute, NMR Centre, Bijenicˇka cesta 54, HR-10000 Zagreb, Croatia
d Faculty of Food Technology and Biotechnology, Pierottije6a 6, HR-10000 Zagreb, Croatia
Received 27 January 2000; accepted 10 May 2000
Abstract
A series of complexes HgX2(H4pymtH)2 (X=Cl−, Br−, I−, SCN−, CN−; H4pymtH=3,4,5,6-tetrahydropyrimidine-2-thione)
has been obtained by the reaction of H4pymtH with mercury(II) halides and pseudohalides in the molar ratio 2:1. X-ray
diffraction studies revealed tetrahedral coordination of mercury with S-bound H4pymtH. The exception is Hg(CN)2(H4pymtH)2
where the coordination is 2+2 with two strongly bound CN− ligands and weaker HꢀS bonds with H4pymtH. One- and
1
two-dimensional H and 13C NMR measurements in dimethylsulfoxide solution confirmed the complexation of mercury to sulfur.
The greatest complexation effects on chemical shifts were detected for the C-2, C-5 and H-1,3 atoms, i.e. two, four and five bonds
away from mercury atom. The complexation effects in Hg(SCN)2(H4pymtH)2 and Hg(CN)2(H4pymtH)2 are in agreement with the
strongest intramolecular H-bonding in the former and the weakest HgꢀS bonds in the latter as compared to other complexes here.
© 2000 Elsevier Science S.A. All rights reserved.
Keywords: Crystal structures; Thermal analysis; Mercury complexes; Heterocyclic thione complexes; NMR studies
1. Introduction
ligand is strongly solvated, their ability to form com-
plexes is hampered. The complex formation thermody-
namics, kinetic and structural properties of mercury(II)
halide and pseudohalide systems in non-aqueous and
aqueous solutions are available [1,7–11]. These data
have to be taken into account while planing the reac-
tion. The fact that mercury(II) ions interact with many
biological molecules through coordination with depro-
tonated thiol, imidazole, disulfide, thioether, amino or
carboxylate groups is well known and a great deal of
effort has been devoted to the characterisation of these
interactions in model molecules and in proteins [12].
Mercury(II) is in neutral mercury(II) halides a typical
soft electron pair acceptor which forms strong bonds of
a covalent character to soft donor atoms. In the past
few decades many papers have dealt with solvated
mercury(II) halides in solvents with different coordinat-
ing properties [1–6]. The solvation of the involved
species plays an important role for all complex forma-
tion reactions in solutions. If the metal ion or the
The complexes of HgCl2 with L-cysteine show an inter-
esting example where a complex with covalently
bonded chlorine atoms can be easily converted into an
* Corresponding author. Tel.: +385-1-460 6652; fax: +385-1-455
4960.
E-mail address: zpopovic@zagreb.zoak.pmf.hr (Z. Popovic´).
0020-1693/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0020-1693(00)00162-6

Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
143
essentially ionic chloride which is the more stable com-
plex at physiological pH [13]. There are also evidences
pointing to the significance of mercury–halide complex-
ation in biological systems. Both neutral mercuric
halides and halogenomercurate anions have been used
as probes of biological activity in proteins since intact
mercurates can interact electrostatically with the
protein in a specific site [14]. As a part of a more
general program designed to prove certain aspects of
mercury–halide binding in competition with other lig-
ands particularly heterocyclic thiones [15–17] we have
investigated the reactions of mercury(II) halides and
pseudohalides with 3,4,5,6-tetrahydropyrimidine-2-
thione under various conditions. There are no reports
on the donor characteristics of H4pymtH toward mer-
cury in solid or in solution, although the metal com-
plexes of the related ligands, namely ethylene thiourea
[18,19], pyrimidine-2-thione and its derivatives, have
been reported [20–33]. On the contrary, the structure of
the ligand itself has been known for many years [34].
We report here the synthesis of the 1:2 complexes of
mercury(II) halides and pseudohalides with H4pymtH
and spectroscopic, thermal and X-ray diffraction
studies.
19.5 Hz per point in F2 and F1 dimensions, respec-
tively. The NOESY spectra were recorded in phase-sen-
sitive mode with mixing time 0.5 s. All other
measurement parameters were as for COSY spectra.
The thermal measurements were performed using a
simultaneous TGA–DTA analyzer (TA Instruments,
SDT model 2960). The samples were placed in small
aluminium oxide sample pans. The TGA and DTA
curves were obtained by placing the samples of about 5
mg in mass, in open sample pans, with a heating rate of
10°C min−1 and nitrogen (purity above 99.996%) pour-
ing at a flow rate of 50 ml min−1. The SDT was
calibrated with indium.
2.2. Preparation of the complexes
In solvents with strong donor properties, the mer-
cury(II) halides react immediately with the solvent
forming adduct compounds in the solid phase. There-
fore we have chosen methanol as the media for prepar-
ing these complexes. Mercury(II) complexes 1–5 were
prepared by reacting 1 equiv. of the corresponding
mercury(II) salt and 2 equiv. of the H4pymtH ligand in
methanol. The crystalline solids, which formed on
standing for several days, were filtered off, washed with
methanol and dried. The isolated complexes are crys-
talline and colourless substances, except for the iodo
complex which is yellow. The thiocyanato complex is
photosensitive and should be kept in the dark. All of
the complexes are generally insoluble in common or-
ganic solvents, but soluble in solvents with pronounced
donor properties such as DMSO, DMF, g-picoline or
pyridine.
2. Experimental
2.1. Materials and measurements
Mercury(II) halides and cyanide as well as the
H4pymtH ligand were obtained from Aldrich and used
without further purification. Mercury(II) thiocyanate
was prepared by the reaction of mercury(II) nitrate
monohydrate and ammonium thiocyanate. The IR
spectra were recorded as KBr pellets on the Perkin–
Elmer FTIR 1600 spectrometer over the range 4000–
2.2.1. Preparation of HgCl2(H4pymtH)2 (1)
Yield: 71%. Anal. Calc. for C8H16Cl2HgN4S2: Hg,
39.81; N, 11.12; S, 12.73. Found: Hg, 40.02; N, 11.22;
S, 12.65%.
450 cm−1
.
The 1H and 13C NMR measurements were performed
with a Varian Gemini 300 spectrometer, operating at
75.46 MHz for the 13C nucleus. Concentrations of
samples, dissolved in DMSO-d6 in 5 mm tubes, were
2.2.2. Preparation of HgBr2(H4 pymtH)2 (2)
Yield: 92%. Anal. Calc. for C8H16Br2HgN4S2: Hg,
33.84; N, 9.45; S, 10.82. Found: Hg, 34.09; N, 9.54; S,
10.72%.
1
0.05 and 0.25 M in H and 13C measurements, respec-
tively. Spectra were recorded at 20°C. All chemical
shifts, in parts per million (ppm), are referred to TMS.
2.2.3. Preparation of HgI2(H4 pymtH)2 (3)
Yield: 69%. Anal. Calc. for C8H16HgI2N4S2: C, 13.99;
H, 2.34; N, 8.16. Found: C, 14.07; H, 2.60; N, 8.12%.
1
Digital resolution was 0.32 Hz per point in H and 0.70
Hz per point in 13C one-dimensional spectra. The fol-
1
lowing spectra were measured: standard H, 13C broad-
2.2.4. Preparation of Hg(SCN)2(H4pymtH)2 (4)
Yield: 67%. Anal. Calc. for C10H16HgN6S4: Hg,
36.53; N, 15.30; S, 23.36. Found: Hg, 36.77; N, 15.32;
S, 23.46%.
band proton decoupled, 13C gated decoupled, COSY45
and NOESY. For proton decoupling the Waltz-16 se-
quence was used. The COSY45 spectra were measured
in the magnitude mode with 1024 points in F2 dimen-
sion and 256 increments in F1 dimension, latter zero-
filled to 1024 points. Each increment was taken with 16
scans, 5000 Hz spectral width and a relaxation delay of
0.8 s. The corresponding digital resolution was 9.8 and
2.2.5. Preparation of Hg(CN)2(H4 pymtH)2 (5)
Yield: 81%. Anal. Calc. for C10H16HgN6S2: N, 17.33;
S, 13.22; Hg, 41.36. Found: Hg, 41.50; N, 17.55; S,
13.09%.

144
Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
2.3. Crystal structure studies
(Stoe upgraded) with graphite-monochromated Mo Ka
,
radiation (u=0.71073 A). Lattice constants were ob-
Crystal data, experimental conditions, details of
structure determination and final refinement parameters
are given in Table 1. Data were collected at room
temperature on a four-circle PW 1100 diffractometer
tained by least-squares refinement from the settings of
46 reflections (10.35q518.7°) for 1, 29 reflections
(8.95q516.5°) for 2, 58 reflections (10.05q518.4°)
for 3, 64 reflections (10.25q516.5°) for 4 and 55
Table 1
General and crystal data and summary of intensity data collection and structure refinement
Compound
1
2
3
4
5
Formula
Formula weight
Crystal system and habit
C8H16Cl2HgN4S2
503.86
orthorhombic,
irregular block
Pbca
C8H16Br2HgN4S2
592.78
monoclinic, prism
C8H16I2HgN4S2 C10H16HgN6S4
C10H16HgN6S2
485.00
monoclinic, prism
686.76
monoclinic,
prism
549.12
triclinic, prism
(
Space group
P21/c
0.30×0.22×0.08
I2/a
P1
P21/a
0.60×0.20×0.12
Crystal dimensions (mm3)
Unit cell parameters
0.53×0.51×0.50
0.26×0.15×0.12 0.65×0.50×0.35
,
a (A)
14.8340(7)
12.8340(3)
16.8530(5)
7.8698(7)
15.644(2)
12.8821(10)
14.7371(17)
7.8550(5)
15.0797(11)
7.6508(7)
8.4976(7)
14.175(2)
105.968(10)
89.941(9)
92.003(10)
885.44(18)
2
12.5082(7)
9.0302(5)
14.0857(10)
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
98.79(2)
96.345(8)
95.583(6)
,
V (A)
3208.47(19)
8
2.086
10.173
1904
1567.4(3)
4
2.512
15.170
1096
1734.9(3)
4
2.629
12.659
1240
1583.45(17)
4
2.034
9.980
920
Z
Dcalc (g cm−3
)
2.060
9.164
524
v (mm−1
F(000)
)
2q Range for data collection
Range h, k, l
4–56
4–60
4–60
4–54
4–54
−12 to 0,
−1 to 10,
0 to 10
ꢀ/q
−11 to 0,
0 to 21,
0 to 18
ꢀ
−20 to 20,
−4 to 11,
0 to 21
ꢀ/q
−9 to 9,
−10 to 10,
0 to 18
ꢀ
−15 to 15,
0 to 11,
0 to 17
ꢀ/q
Scan type
Measured reflections
2006
4635
2613
3977
3509
Independent reflections (Rint
Refined parameters
)
898 (0.006)
155
4456 (0.0554)
156
2437 (0.0366)
89
3825 (0.0231)
202
3374 (0.0790)
173
Observed reflections, I]2|(I)
Range of transmission
797
2142
0.0785, 0.3353
1168
0.0252, 0.0808
3160
0.0265, 0.1359
2150
0.1303, 0.3500
0.0301, 0.0627
factors, min., max. a
Face indices from centroid (mm)
(−1 1 0) 0.109
(1 −1 0) 0.109
(0 1 1) 0.038
(0 −1 −1) 0.038
(−1 0 1) 0.150
(1 0 −1) 0.150
(0 1 −1) 0.150
(0 −1 1) 0.150
0.0465, 0
(1 0 0) 0.173
(−1 0 0) 0.173
(0 0 1) 0.250
(0 0 −1) 0.250
(0 1 −1) 0.323
(0 −1 1) 0.323
(0 2 1) 0.113
(1 0 0) 0.300
(−1 0 0) 0.300
(0 0 1) 0.060
(0 0 −1) 0.060
(0 −1 1) 0.090
(0 1 1) 0.098
(0 −1 1) 0.150
b
g1, g2 in w
0.0710, 8.1007
0.0380, 0.0955
0.0443, 0.1020
1.178
0.0318, 0
0.0950, 0
0.0795, 0
c
d
R , wR [I]2|(I)]
0.0397, 0.0802
0.1255, 0.1007
0.870
0.0297, 0.0634
0.0931, 0.0768
0.854
0.0512, 0.1221
0.0640, 0.1300
1.059
0.0464, 0.1069
0.0767, 0.1216
0.983
R, wR [all data]
Goodness-of-fit on F2, S e
Max., min/ electron density
0.758, −1.294
0.784, −1.525
0.633, −0.509
2.680, −2.933
1.694, −3.321
−3
f
,
(e A
)
Extinction coefficients
0.0025(2)
0.001
0.0035(2)
0.001
0.00086(9)
0.001
0.0030(10)
0.001
0.0209(9)
0.001
Maximum Z/|
a Absorption correction type: numerical Gaussian for 2, 4 and 5 and „-scan semi-empirical for 1 and 3.
b w=l/[|2(F2o)+[g1P+g2P] where P=(Fo2+2F2c)/3.
c R=SꢀꢀFoꢀ−ꢀFcꢀꢀ/SwꢀFoꢀ.
d wR=[S(F2o−Fc2)2/Sw(F2o)2]1/2
.
e S=S[w(F2o−F2c)2/(Nobs−Nparam)]1/2
f In all structures the maximum electron density in the last difference Fourier map is near Hg.
.

Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
of the ꢀNHCSNHꢀ moiety and their assignment b
145
Table 2
a
Characteristic frequencies (cm−1
)
wNH
waCꢀN+lNH
lNH+waCꢀN
lNH+wsCꢀN
wCH2
yNH
wCꢁS
H4pymtH
3158s,br 3099s
3229s,br 3143m
3239vs,br 3146s
3300s,br 3125m
3236vs,br 3145vs
3371s 3204s,br
1563vs
1593vs
1595vs
1586vs
1589vs
1580vs,br
1554vs
1561vs
1561vs
1553vs
1564vs
1556vs,br
1208vs
1228s
1229vs
1207s
1224vs
1222s
1193s
1199m-s
1208s
1207s
1190S
1208vs
768s,br
706m,br
704s,br
679m-s,br
741s,br
749s,br
644s
624m
HgCl2(H4pymtH)2
HgBr2(H4pymtH)2
HgI2(H4pymtH)2
Hg(SCN)2(H4pymtH)2
Hg(CN)2(H4pymtH)2
607m c
634s
625m
a Abbreviations: wa, asymmetric stretching; wa, symmetric stretching; l, in plane bending; y, out of plane bending; w, wagging; vs, very strong;
s, strong; m, medium; br, broad.
b The assignment according to Ref. [39].
c Very broad band (ꢀ100 cm−1) centred at 607 cm−1
.
reflections (9.35q514.7°) for 5. Reference reflections
were monitored periodically (every 60 min in 1 or every
90 min in all others) during intensity collection to check
crystal stability. The intensity variation of standard
reflections was as follows: 91.4 in 1, 92.2 in 2, 91.6
in 3, 91.4 in 4, 96.5% in 5. The data were corrected
for Lorentz-polarisation and absorption effects by the
X-RED [35] program. The Gaussian numerical integra-
tion based on regular shapes of crystals was performed
for 2, 4 and 5 and the „-scan semi-empirical method
for 1 and 3 [35]. The absorption correction significantly
improved the crystal structure parameters in all struc-
tures (lower R values, e.s.d. values and residual density
near Hg in last difference Fourier maps). Heavy atoms
were located by Patterson (1) or direct methods (2–5).
The remaining non-hydrogen atoms were found in sub-
sequent difference Fourier maps. Refinement in which
non-hydrogen atoms were treated anisotropically was
performed by full-matrix least-squares methods based
on F2 values against all reflections. Positional disorder
of the methylene C3 atom was found in 3 and 4. The
disordered atoms were refined anisotropically with re-
straints on the Uij components to have the same values
as the neighbouring C atoms. The sum of the occu-
pancy factors of the split C atoms was restrained to
unit and resulted in a value of 0.67 for the major
component in 3 and 0.72 in 4. All methylene H atoms
3. Results and discussion
3.1. IR spectra
The most significant IR data of the H4pymtH ligand
and its mercury(II) complexes 1–5 are summarised in
Table 2. In the spectrum of H4pymtH the in-phase and
out-of-phase NꢀH stretching modes are assigned to a
strong band at 3099 cm−1 and a broad one centred at
3158 cm−1. In the region 1500–1600 cm−1 there is a
very strong broad band, split at the bottom (1563, 1554
cm−1), which can be assigned to coupled modes of
asymmetric CꢀN stretching and NꢀH bending. The
contribution from NꢀH groups is confirmed by deuter-
ation studies [39]. The broad absorption bands of
strong intensity at 768 cm−1 are assigned to the out-of-
plane NH bending, while symmetric CꢀN stretching
coupled with NH bending can be associated with strong
bands at 1361 and 1208 cm−1, respectively. The present
band of strong intensity at 644 cm−1 in the spectrum of
free H4pymtH is assigned to CꢁS stretching on the basis
of its sensitivity to S-methylation (in its S-methyl
derivative this band is shifted to 596 cm−1) [39]. In the
spectra of investigated complexes 1–5, CꢀS stretching is
shifted to lower frequencies due to HgꢀS bonding.
From the structural studies of H4pymtH the predomi-
nance of the thione form over thiol was revealed in
solid and solution phases although a weak band at 2658
cm−1 in the IR spectrum of solid H4pymtH is ob-
served. In the complexes the NH stretching frequency
increases by approximately 70–200 cm−1 on going
from chloro to cyano complex. This displacement of
the NH modes can be explained by means of different
hydrogen bonding present in complexes. The ligand
bands at 1563 and 1554 cm−1 shifted to higher frequen-
cies due to an increase in the CꢀN bond order in the
complexes. The bands assigned to the NH out-of-plane
vibrations shifted to lower frequencies by approxi-
mately 90 cm−1 confirming the coordination through
,
were constrained to ride at CꢀH 0.97 A with the
temperature factor Uij=1.2Ueq (Csp3). The last Fourier
maps revealed positions of all H atoms bonded to N
atoms but mostly with poor geometry. Their positions
were therefore geometrically optimised (only in 2 the
found coordinates had good geometry and were fixed
during refinement, while the thermal isotropic parame-
ters were allowed to refine) assuming sp2 hybridisation
of N atoms and applying the riding model with NꢀH
,
0.86 A and Uij=1.2Ueq. Calculations were done by
SHELXS-97 [36] and SHELXL-97 [37], and the molecular
graphics with PLATON-98 [38].

146
Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
the exocyclic sulfur atom as revealed by NMR and
X-ray structure analysis. The lowest shift is observed
for the cyano complex, which is consistent with the
weakest mercuryꢀsulfur bonds in complexes 1–5.
single-crystal X-ray analysis the influence of the X
ligand on the length of the HgꢀS bond in the 1:2
mercury complexes HgX2(H4pymtH)2.
The values of selected bond distances and angles are
listed in Table 3 and the hydrogen bonding geometry in
Table 4.
3.2. Description of the crystal structures
The stereochemistry of mercury in the solid state is
characterised by a few typical modes of ligand binding,
i.e. characteristic coordination (m) which is formed by
covalent HgꢀL bonds with m ligands, and effective
coordination (m+n) with n additional ligands at dis-
tances shorter than the sum of the van der Waals radii
of Hg and L [40]. In a tetrahedral environment the
extension of coordination number is usually not accom-
plished, but the deviation from regular tetrahedral ge-
ometry occurs.
3.2.1. HgX2(H4 pymtH)2, X=Cl (1), Br (2) and I (3)
The mean HꢀS distances are similar in the chloro
,
(2.450 A) and bromo complex (2.449) and are by 0.07
,
A shorter than the sum of the covalent radii of S (1.04
,
,
A [41]) and tetrahedral Hg (1.48 A [40]). The longest
,
HgꢀS bond is found in the iodo complex (2.612(2) A)
being significantly longer than the calculated covalent
bond. The molecular structures of 1, 2 and 3 are shown
in Figs. 1–3, respectively.
The sum of the covalent radius of Hg with Cl, Br and
,
The deformation from regular polyhedral geometry
can be influenced by crystal packing forces in the
crystalline state, mostly by hydrogen bonds that involve
ligand-donor atoms. In this work we examined by
I amounts to 2.52, 2.62 and 2.81 A, respectively. It can,
therefore, be calculated that the mean HgꢀX bond
,
,
length (HgꢀCl 2.640 A) in 1 and (HgꢀBr 2.733 A) in 2
,
is by 0.12 and 0.11 A longer than the calculated sum.
Table 3
,
Selected bond distances (A) and angles (°)
HgCl2(H4pymtH)2 HgBr2(H4pymtH)2 HgI2(H4pymtH)2
Hg(SCN)2(H4pymtH)2 Hg(CN)2(H4pymtH)2
a
HgꢀX1
2.647(4)
2.633(4)
2.414(4)
2.485(4)
1.746(15)
1.748(15)
1.296(6)
1.353(15)
1.465(16)
2.7336(10)
2.7315(8)
2.431(2)
2.4674(19)
1.756(8)
1.752(8)
1.312(10)
1.315(9)
1.482(10)
1.475(12)
1.527(12)
1.451(10)
1.311(9)
1.297(9)
1.448(9)
1.430(6)
1.467(12)
1.436(11)
2.7123(5)
2.6119(16)
1.729(5)
2.556(2)
2.608(2)
2.5260(19)
2.487(2)
1.753(8)
1.730(9)
1.303(10)
1.311(10)
1.449(13)
2.046(7)
2.065(7)
2.8558(18)
2.9742(17)
1.723(6)
1.710(7)
1.331(8)
1.309(8)
1.471(9)
1.493(13)
1.495(14)
1.470(10)
1.328(8)
1.321(8)
1.454(9)
1.448(15)
1.456(15)
1.459(8)
HgꢀX2 b
HgꢀS1
HgꢀS2
S1ꢀC1 (SꢀC1 in 3)
S2ꢀC21
C11ꢀN11 (C1ꢀN1 in 3)
C11ꢀN12 (C1ꢀN2 in 3)
N11ꢀC12 (N1ꢀC2 in 3)
1.303(7)
1.323(7)
1.461(8)
C12ꢀC13 (C2ꢀC3 in 3) c 1.506(19)
1.457(12), 1.501(12) 1.456(18)
1.428(10), 1.455(12) 1.458(17)
1.453(8)
C13ꢀC14 (C3ꢀC4 in 3)
C14ꢀN12 (C4ꢀN2 in 3)
C21ꢀN21
C21ꢀN22
N21ꢀC22
C22ꢀC23
C23ꢀC24
C24ꢀN22
1.472(17)
1.502(16)
1.308(16)
1.331(16)
1.444(16)
1.51(2)
1.469(11)
1.318(11)
1.315(10)
1.458(12)
1.479(13), 1.474(17)
1.467(17), 1.457(13)
1.467(12)
1.659(10)
1.139(13)
1.43(2)
1.481(17)
S3ꢀC3
C3ꢀN3
S4ꢀN4
C4ꢀN4
1.151(8)
1.119(8)
1.656(10)
1.150(13)
S1ꢀHgꢀS2
S1ꢀHgꢀX1 a
S1ꢀHgꢀX2 b
S2ꢀHgꢀX1
S2ꢀHgꢀX2
X2ꢀHgꢀX1
128.31(12)
107.66(13)
110.97(13)
107.11(12)
104.28(11)
92.71(14)
138.82(8)
93.38(5)
105.93(6)
103.44(6)
101.28(5)
105.56(3)
98.67(9)
108.56(4)
108.56(4)
107.64(3)
107.64(3)
123.06(3)
112.45(7)
114.01(9)
103.39(9)
112.34(9)
112.97(8)
100.80(9)
98.02(5)
97.46(19)
93.35(19)
96.43(19)
91.5(2)
165.6(3)
a X1=C1 in 1, Br in 2, I in 3, S3 (from SCN) in 4 and C3 (from CN) in 5.
b X2=C1 in 1, Br in 2, S4 (from SCN) in 4 and C4 (from CN) in 5.
c Values in the second column in 3 are for the disordered C3 atom of smaller occupancy; the same is for C23 in 4. Symmetry transformations
used to generate equivalent atoms: (i) −x−1/2, y, −z.

Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
147
Table 4
a
,
Hydrogen bond geometry (A, °)
DꢀH···A
DꢀH
d(DꢀH)
d(H···A)
ÚDHA
d(D···A)
HgCl2(H4pymtH)2
N11ꢀH11···C12i
N12ꢀH12···C11
N21ꢀH21···C11
N22ꢀH22···C12ii
N11ꢀH11
N12ꢀH12
N21ꢀH21
N22ꢀH22
0.86
0.86
0.86
0.86
2.43
2.39
2.41
2.55
170.4
165.8
175.5
146.1
3.278(14)
3.227(11)
3.264(13)
3.300(13)
HgBr2(H4pymtH)2
N11ꢀH11···S2iii
N12ꢀH12···Br2
N21ꢀH21···Br1
N22ꢀH22···Br2
N11ꢀH11
N12ꢀH12
N21ꢀH21
N22ꢀH22
1.07
0.95
1.00
1.01
2.69
2.52
2.50
2.55
155.2
175.9
150.5
148.7
3.691(7)
3.466(6)
3.403(6)
3.454(6)
iv
v
HgI2(H4pymtH)2
N2ꢀH2···S vi
N2-H2
N1-H1
0.86
0.86
2.89
2.89
141.1
164.2
3.599(6)
3.725(5)
NIꢀH1···I vii
Hg(SCN)2(H4pymtH)2
N11ꢀH11···N3 viii
N21ꢀH21···N4 iii
N12ꢀH12···S1 ix
N22ꢀH22···N4
N11ꢀH11
N21ꢀH21
N12ꢀH12
N22ꢀH22
0.86
0.86
0.86
0.86
2.06
2.12
2.59
2.39
172.6
158.9
165.2
157.3
2.918(13)
2.933(13)
3.424(8)
3.199(13)
Hg(CN)2(H4pymtH)2
N11ꢀH11···N4 x
N22ꢀR22···N3 xi
N12ꢀH12···S2 xii
N21ꢀH21···S1 xiii
N11ꢀH11
N22ꢀH22
N12ꢀH12
N21ꢀH21
0.86
0.86
0.86
0.86
2.36
2.29
2.49
2.55
134.2
146.6
168.9
170.6
3.020(9)
3.041(9)
3.335(6)
3.399(6)
a Symmetry transformations used to generate equivalent atoms: (i) 0.5−x, 0.5−y; −z; (ii) 0.5−x, 0.5+y, z; (iii) 1+x, y, z; (iv) x, 0.5−y,
0.5+z; (v) 1−x, l−y, −z; (vi) −x−0.5, y, −z; (vii) −0.5+x, 1−y, z; (viii) x, −1+y, z; (ix) −x, 1−y, 1−z; (x) 0.5−x, −0.5+y, −z; (xi)
0.5−x, 0.5+y, 1−z; (xii) −0.5+x, 0.5−y, z; (xiii) 0.5+x, 0.5−y, z.
,
A, respectively, with a corresponding increase in the
On the contrary, the bond length HgꢀI in 3 is even
,
shorter than HgꢀBr in 2 and is 0.10 A shorter than the
SꢀHgꢀS angle [103.3(1), 107.2(2) and 109.1(2)] and a
decrease in the XꢀHgꢀX angle [113.2(1), 112.4(1) and
111.8(2)]. The mean HgꢀCl (2.481 A) and HgꢀBr (2.592
length of the corresponding calculated bond. Therefore
the strongest bond, HgꢀI, is formed by elements having
the smallest electronegativity difference.
,
,
A) bond lengths are shorter than in 1 and 2 but HgꢀI
Mercury has a strong tendency to preserve linear
coordination and the largest angle will always be the
one with the strongest bonds. This is in agreement with
the experimental values since the IꢀHgꢀI angle of
123.06(3) is the greatest XꢀHgꢀX angle in complexes
1–3 [BrꢀHgꢀBr 105.56(3), ClꢀHgꢀCl 92.71(14)]. Corre-
spondingly, the weakest HgꢀS bond and the smallest
SꢀHgꢀS angle of 98.67(9) is in 3. It is interesting,
however, that the tetrahedron in 2 is the one with
greatest distortions — the SꢀHgꢀS angle is 138.82(8) in
2 and 128.31(12) in 1 whereas the HgꢀS bonds are of
similar strength. This could be influenced by crystal
packing and a hydrogen bond to sulfur in 2.
,
(2.743 A) is longer than in 3. These three complexes
show a much smaller distortion of the tetrahedron and
we suppose that it could be explained by hydrogen
bonding which is much stronger in 1–3 than in the
mentioned complexes where only weak CꢀH···X hydro-
gen bonds exist. In 1–3 both nitrogen atoms from both
H4pymtH ligands are involved in extensive hydrogen
bonding which interconnects the molecules. The hydro-
gen bonds are both intra- and intermolecular (1 and 2)
There is only one type of tetrahedral HgX2L2 com-
plex (L being a S bound monodentate ligand) where
crystal structures of all three halide complexes (X=Cl,
Br, I) have been determined up to now: the structures
with L=N,N-dimethylthioformamide [42]. In these
complexes the HgꢀS bonds are weaker than in 1 and 2
and stronger than in 3. The mean value increases from
the chloro to the iodo complex, 2.532, 2.542 and 2.578
Fig. 1. PLATON-98 drawing of HgCl2(H4pymtH)2 (1) with the atom
numbering scheme. The thermal ellipsoids are at the 50% probability
level for non-H atoms. H atoms are shown as spheres of arbitrary
radii.

148
Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
3.2.2. Hg(SCN)2(H4 pymtH)2 (4)
In the crystal structure of Hg(SCN)2(H4pymtH)2 the
mercury atom is tetrahedrally surrounded with four
sulfur atoms. Two stronger HgꢀS bonds are formed
with the H4pymtH ligands [HgꢀS1 2.526(2) and HgꢀS2
,
2.487(2) A] and two weaker ones with thiocyanate
,
anions [HgꢀS3 2.556(2) and HgꢀS4 2.608(2) A] (Fig. 4).
The bond angles vary between S3ꢀHgꢀS4 of 100.80(9)°
and S1ꢀHgꢀS3 of 114.01(9)° and have the smallest
deviation from a regular tetrahedron in all complexes
1–4. The HgꢀS(thiocyanate) bond lengths are slightly
longer than the sum of covalent radii for Hg and S,
while those from the H4pymtH ligand, are equal to or
,
shorter than 2.52 A. The HgꢀS(thiocyanate) bond val-
Fig. 2. PLATON-98 drawing of HgBr2(H4pymtH)2 (2) with the atom
numbering scheme. The thermal ellipsoids are at the 50% probability
level for non-H atoms. H atoms are shown as spheres of arbitrary
radii.
ues are among the larger ones in tetrahedral
[Hg(SCN)4]2− which are in the range from 2.496 to
2.610 A according to the Cambridge Structural Data-
,
base (CSD) [43]. The analogous bond model [shorter
HgꢀS(thione) and longer HgꢀS(thiocyanate)] is found
in the structures of two polymorphs of Hg(SCN)2-
(C5H5NS)2 [16] containing HgꢀS(thiocyanate) bonds in
,
the range 2.568(1)–2.685(2) A, while the HgꢀS(thione)
,
distances are in the range 2.4580(12)–2.5123(12) A. A
survey of the CSD retrieved only one other monomeric
complex with mercury in the tetrahedral surrounding
built up of S atoms (two from SCN− anions and two
from any other ligand): (dithiocyanato-S)-(N,N%-dicy-
clohexyldithio - oxamide)mercury(II) (YUBZOK [44]).
In this structure the HgꢀSCN bond lengths are the
Fig. 3. PLATON-98 drawing of HgI2(H4pymtH)2 (3) with the atom
numbering scheme. The thermal ellipsoids are at the 50% probability
level for non-H atoms. H atoms are shown as spheres of arbitrary
radii. The two possible orientations of C3 atom are shown with the
major component labelled. The symmetry transformation used to
generate equivalent atoms: (i) −x−1/2, y, −z.
,
shorter ones (2.494(2) and 2.500(2) A) whereas the
,
HgꢀS values are longer (2.577(2) and 2.551(2) A) since
the ligand is chelating and forms a very small angle
SꢀHgꢀS of 86.27(6)°.
The crystal structure is stabilised by NꢀH···N and
NꢀH···S hydrogen bonds (Table 4). Interestingly, atom
N3 from the SCN− anion is involved in only one linear
intermolecular hydrogen bond, but atom N4 takes part
as a proton acceptor forming two hydrogen bonds
which depart from linearity. Such a hydrogen bond
pattern is not unexpected, i.e. a statistical analysis of
the hydrogen bonding geometry at the thiocyanate
nitrogen acceptor atom showed that the SCN− anion
usually accepts more than one non-linear hydrogen
bond [45].
3.2.3. Hg(CN)2(H4 pymtH)2 (5)
This structure is quite different from the previous
ones since two strong linear HgꢀC bonds are formed
with the CN− ligands [HgꢀC3 2.046(7), HgꢀC4 2.065(7)
Fig. 4. PLATON-98 drawing of Hg(SCN)2(H4pymtH)2 (4) with the
atom numbering scheme. The thermal ellipsoids are at the 50%
probability level for non-H atoms. H atoms are shown as spheres of
arbitrary radii. The positional disorder of C23 atom is shown with the
major component labelled.
,
A, C3ꢀHgꢀC4 165.6(3)°] (Fig. 5). The HgꢀS distances
,
[HgꢀS1 2.8558(18) and HgꢀS2 2.9742(17) A] are the
longest ones in 1–5. They are longer than covalent
HgꢀS bond but shorter than the sum of van der Waals
,
or only intermolecular as in 3. In all complexes they are
of the type NꢀH···X while in 1 and 3 bonds of the type
NꢀH···S are also formed (Table 4).
radii for Hg (1.55, 1.54 or 1.55–1.65 A [46–48]) and S
,
(1.60–2.03 A [49]). Therefore, the effective coordination
is 2+2 [40]. The HgꢀC values relate to strong covalent

Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
149
Scheme 1.
All available NꢀH protons are included in inter-
molecular hydrogen bond formation. There are no in-
tramolecular hydrogen bonds. Two NꢀH groups are
involved in hydrogen bonding with the cyano N atoms.
Two others form NꢀH···S hydrogen bonds with thione
S atoms.
Fig. 5. PLATON-98 drawing of Hg(CN)2(H4pymtH)2 (5) with the atom
numbering scheme. The thermal ellipsoids are at the 50% probability
level for non-H atoms. H atoms are shown as spheres of arbitrary
radii. The Hg···S contacts are shown by dashed lines indicating the
2+2 coordination of Hg.
1
3.3. H and 13C NMR spectra
NMR data of parent H4pymtH and its complexes
with mercury(II) halides and pseudohalides 1–5 are
1
collected in Table 5. H and 13C NMR spectra show
mercury to carbon bonds and according to the X-ray
crystallographic criterion could be regarded only as
slightly longer than in the structure of Hg(CN)2 where
that in dimethylsulfoxide (DMSO) solution the
H4pymtH molecule exists in the thione form as previ-
ously found in the solid phase [34,39]. The thione and
thiol tautomers of H4pymtH and the enumeration of
,
they are 2.015(3) A [50]. The review of the CSD re-
1
vealed no structures containing such a HgS2(CN)2 coor-
carbon atoms are displayed in Scheme 1. In H spectra
dination moiety as in 5.
the greatest complexation effect (Dl) was observed at
Table 5
1H and 13C NMR data for the ligand H4pymtH and its mercury(II) complexes 1–5. Chemical shifts (l/ppm) a, H–H spin–spin coupling constants
(JHH/Hz) b, mercury induced 1H shifts (Dl/ppm) c and mercury induced 13C shifts (Dl/ppm) d
e
f
Compound
H-1,3
H4pymtH HgCl2(H4pymtH)2 HgBr2(H4pymtH)2 HgI2(H4pymtH)2 Hg(SCN)2(H4pymtH)2
Hg(CN)2(H4pymtH)2
8.22 (2H)
l
7.86 (2H) 8.97 (2H)
8.72 (2H)
(s)
8.59 (2H)
(s)
9.01 (2H)
(s)
JHH (s) (s)
Dl
(s)
1.11
0.86
0.73
1.15
0.36
3.18 (4H)
l
3.09 (4H) 3.26 (4H)
3.25 (4H)
3.25 (4H)
3.29 (4H)
H-4,6
JHH
Dl
l
(m)
(m)
0.17
(m)
0.16
1.82 (2H)
(m)
0.16
1.83 (2H)
(m)
0.20
1.84 (2H)
0.09
1.77 (21-1)
1.71 (2H) 1.82 (2H)
H-5
C-2
JHH
Dl
5.70 (qn) 5.22 (t) g
0.11
4.78 (t) g
0.11
4.98 (t) g
0.12
5.07 (t) g
0.13
5.41 (qn)
0.06
l
175.95
39.88
19.29
168.66
−7.29
168.90
−7.05
169.74
−6.21
167.10
−8.85
172.99
−2.96
Dl
C-4,6
C-5
l
40.06
0.18
40.10
0.22
40.11
0.23
40.21
0.33
39.99
0.11
Dl
l
18.25
18.30
18.50
18.03
18.79
Dl
−1.04
−0.99
−0.79
−1.26
−0.50
a Measured in DMSO-d6 solution. Referred to TMS. Number of protons in brackets.
b Digital resolution 90.32 Hz; (s) singlet, (t) triplet, (qn) quintet, (m) multiplet splitting.
c Mercury-induced 1H shifts are defined as the difference of proton chemical shifts in mercurated and parent molecule. Minus sign denotes
shielding.
d Mercury-induced 13C shifts are defined as the difference of carbon chemical shifts in mercurated and parent molecule. Minus sign denotes
shielding. Digital resolution 90.70 Hz.
e In Hg(SCN)2(H4pymtH)2 the 13C chemical shift of SCN is 122.79 ppm. Dl is 6.91 ppm as referred to Hg(SCN)2 in DMSO-d6, l=115.88 ppm.
f In Hg(CN)2(H4pymtH)2 13C chemical shift of CN is 144.50 ppm. It is somewhat broadened. Dl is −0.25 ppm as referred to Hg(CN)2 in
DMSO-d6, l=144.75 ppm..
g Signal appears as a triplet, since quintet splitting is not completely resolved.

150
Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
bonded to sulfur, since the C-2 is the closest carbon to
the complexation site. This is in agreement with length-
ening of the CꢀS bond in complexes as also obtained by
PM3 calculation [52]. In 1–5 the C-2 is shielded from
−2.96 to −8.85 ppm, as compared to the C-2 in the
parent H4pymtH. The difference in shielding at the C-2
is consistent with the difference of the HgꢀS bond
strength which arises from the back donation of mer-
cury to sulfur and is influenced by the degree of cova-
lency of the HgꢀX bonds. As a consequence, the
shielding at the C-2 is lowered from the chloro to the
iodo complex, while deshielding at the H-1, 3 is lowered
in same sequence. The large downfield shift (6.91 ppm)
of the C(thiocyanate) in 4, relative to the chemical shift
of this C-atom in free mercuric thiocyanate is in agree-
ment with intramolecular H-bonding and the tetrahe-
dral coordination of the mercury atom. The decrease in
shielding at the C-2 in 5 is connected with the 2+2
coordination. This means that due to longer HgꢀS
bonds, the influence of the Hg(CN)2 moiety on the
electron distribution in the ligand is small. Therefore,
among all complexes investigated here, the 13C chemical
shifts in 5 are the most similar to those of the free
H4pymtH molecule. The smallest complexation 13C
shift at C-2, amounting to −2.96 ppm, was detected in
5 while the greatest one (−8.85 ppm) was found in 4
(Table 5). Although the C-4, 6 are four bonds and the
C-5 are even five bonds away from mercury, the com-
plexation effects at the latter carbon (−0.50 to −1.26
ppm) are approximately five times greater than at the
former one. In addition, Dl at the C-5 is a shielding
effect (as at the C-2), while that at the C-4, 6 is a
deshielding one. The subtle deshielding in the latter is
related to an increase of the p-electron density in the
CꢀN bond. Dl at C-2 and C-5 are different by an order
of magnitude, which arises from a different through-
bond distance between mercury and these carbons. The
complexation effects at C-5 show that even long-range
effects on chemical shifts (through five bonds), might be
used as a complexation probe.
the H-1, 3 similarly as in the mercury(II) complexes
with 1,3-imidazole-2-thione [17]. These NH protons,
four bonds away from the mercury atom, are 0.36–1.15
ppm more deshielded in the complexes than in the free
H4pymtH molecule. This is due to delocalization of
p-electrons between N-atoms and accompanies shorten-
ing of the CꢀN and lengthening of the CꢀS bonds in the
complexes. In addition, deshielding of NH protons is
also caused by differences of H-bonding in complexes
and in free H4pymtH, which was supported by variable
temperature measurements. Dl at the H-1, 3 is related
to the strength of the HgꢀS bond, which is influenced
by the character of the HgꢀX bond [40,51]. Thus,
shielding of the H-1, 3 is greater in the iodo than in the
bromo and chloro complexes. The greatest complexa-
1
tion H shift (1.15 ppm) in a series of complexes 1–5
was found at the H-1, 3 in the thiocyanato complex,
due to strong intramolecular H-bonding of these pro-
tons with the nitrogen atoms of the SCN groups.
Contrary to this, in the cyano complex the smallest
complexation effect at H-1, 3 (0.36 ppm) was observed
(Table 5) suggesting that in the cyano complex the
H4pymtH ligand is bound much weaker (longest HgꢀS
bonds) than in the other complexes. Deshielding com-
plexation effects at the H-5 in 1–5 are the smallest ones
observed (0.06–0.13 ppm). This is due to six-bond
separation between Hg and the H-5 protons, but also
because of considerable distance from the nitrogen
positive charge. The greatest change on complexation
in 13C chemical shifts was observed for the thione
carbon atom (C-2). This confirms that mercury is
Table 6
Thermoanalytical data for thermal analysis of the mercury(II) com-
plexes
Complex
Process
Temperature
of peak (°C)
HgCl2(H4pymtH)2
melting (endo)
I endo
II endo
179.43
328.42
386.29
One has to mention that the conformational flexibil-
ity of the ring system might also be important for the
long-range transmission. Satellite 199Hgꢀ1H and
199Hgꢀ13C couplings were not observed. This is usually
encountered when the HgꢀC coupling pathway includes
heteroatoms like S, N or O [16,17].
HgBr2(H4pymtH)2
HgI2(H4pymtH)2
melting (endo)
I endo
II endo
147.78
370.85
426.86
melting (endo)
I endo
II endo
134.81
368.07
430.88
3.4. Thermal analysis (TGA and DTA)
Hg(SCN)2(H4pymtH)2
melting (endo)
I
II endo
III exo
133.43
243.89
369.70
407.21
The samples of mercury complexes 1–5 were heated
from room temperature up to 700°C (see thermoanalyt-
ical data shown in Table 6). In each DTA curve there is
a typical endothermic minimum, which corresponds to
the melting process of the complex. It is represented by
the first single endothermic peak at a temperature be-
tween 130 and 180°C. The endothermic peak of melting
Hg(CN)2(H4pymtH)2
melting (endo)
I
II endo
III exo
147.02
206.84
304.63
407.08

Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
151
of the halide complexes progressively decreases with
References
increasing formula weight. Two overlapping endother-
mic peaks, the first one being small and broad and the
second one larger and sharper, represent the decompo-
sition of the halide complexes. The DTA curve contains
endothermic minima at 368.07 and 430.88°C in 3, at
370.85 and 426.86°C in 2, and at 328.42 and 386.29°C
in 1. Two different decomposition steps can be
classified as decomposition of the ligand and dehalo-
genation of the residue. From the first decomposition
peak in the DTA curve the following order of complex
stability can be established: iodo:bromo\chloro.
From the temperature of the second decomposition
step, the following order of dehalogenation processes is
established: iodo\bromo\chloro. The shapes of the
DTA curves of the halide complexes suggest an analogy
in the decomposition steps due to their similar struc-
tures. The decomposition of the thiocyanato complex is
represented by a broad exothermic peak at 243.89°C,
an endothermic minimum at 369.70°C, and a broad
exothermic peak at 407.21°C. The DTA curve of
Hg(CN)2(H4pymtH)2 is characterised by three peaks,
two exothermic ones, a sharp one at 206.84°C and a
small, broad one at 407.08°C, and the third one being a
broad endothermic peak at 304.63°C. The first step of
decomposition, determined by exo(I) and endo(II)
peaks, represents elimination of the ligand. One can
conclude from the TGA–DTA curves of the reactants
that the second step, exo(III), is connected with degra-
dation of the corresponding mercury(II) pseudohalides.
Parallel occurrence of different steps of the decomposi-
tion process, resulting in the overlapping of DTA
peaks, makes partial determination of the weight loss of
particular elimination compounds impossible.
[1] I. Persson, K.C. Dash, Y. Kinjo, Acta Chem. Scand. 44 (1990)
433.
[2] M. Sandstro¨m, I. Persson, P. Persson, Acta Chem. Scand. 44
(1990) 653.
[3] I. Persson, M. Sandstro¨m, P.L. Goggin, Inorg. Chim. Acta 129
(1987) 183.
[4] I. Persson, M. Landgren, A. Marton, Inorg. Chim. Acta 116
(1986) 135.
[5] I. Persson, M. Sandstro¨m, P.L. Goggin, A. Mosset, J. Chem.
Soc. Dalton Trans. (1985) 1597.
[6] S. Ahrland, Pure Appl. Chem. 54 (1982) 1451.
[7] L.G. Sille´n, Acta Chem. Scand. 3 (1949) 539.
[8] Y. Marcus, Acta Chem. Scand. 11 (1957) 599.
[9] P.K. Gallagher, E.L. King, J. Am. Chem. Soc. 82 (1960) 3510.
[10] S. Ahrland, L. Kullberg, R. Portanova, Acta Chem. Scand. A 32
(1978) 251.
[11] J.J. Christiansen, R.M. Izatt, L.D. Hansen, J.D. Hale, Inorg.
Chem. 3 (1964) 130.
[12] T.L. Blundell, J. Jenkins, Chem. Soc. Rev. (1977) 139.
[13] N.J. Taylor, A.J. Carty, J. Am. Chem. Soc. 99 (1977) 6143.
[14] L. Book, A.J. Carty, C. Chieh, Can. J. Chem. 59 (1981) 138.
&
[15] N. Davidovic´, D. Matkovic´-Calogovic´, Z. Popovic´, I. Vedrina-
Dragojevic´, Acta Crystallogr., Sect. C 54 (1998) 574.
&
[16] Z. Popovic´, D. Matkovic´-Calogovic´, J. Hasic´, D. Vikic´-Topic´,
Inorg. Chim. Acta 285 (1999) 208.
&
&
[17] Z. Popovic´, D. Matkovic´-Calogovic´, Z. Soldin, G. Pavlovic´, N.
Davidovic´, D. Vikic´-Topic´, Inorg. Chim. Acta 294 (1999) 35.
[18] S.L. Hold, R.L. Carlin, J. Am. Chem. Soc. 86 (1964) 3017.
[19] F.A. Devillanova, G. Verani, Inorg. Chim. Acta 30 (1978) 209.
[20] M.Y.W. Yu, J. Sedlak, R.H. Lindsay, Arch. Biochem. Biophys.
155 (1973) 111.
[21] I.P. Khullar, U. Agarwalla, Indian J. Chem. 12 (1974) 1096.
[22] I.P. Khullar, U. Agarwalla, Aust. J. Chem. 28 (1975) 1529.
[23] J. Abbot, D.M.L. Goodgame, J. Jeeves, J. Chem. Soc., Dalton
Trans. (1978) 880.
[24] D.M.L. Goodgame, G.A. Leach, J. Chem. Soc., Dalton Trans.
(1978) 1705.
[25] R. Battistuzzi, G. Peyronel, Transition Met. Chem. 3 (1978) 345.
[26] F.A. Cotton, W.H. Ilsley, Inorg. Chem. 20 (1981) 614.
[27] D.A. Stuart, L.R. Nassimbeni, A.T. Hutton, K.R. Koch, Acta
Crystallogr., Sect. B 36 (1980) 2227.
4. Supplementary material
[28] R. Battistuzzi, G. Peyronel, Spectrochim. Acta 36A (1980) 113.
[29] J.M. Bret, P. Castan, J.P. Laurent, Inorg. Chim. Acta 51 (1981)
103.
[30] M.L. Godino-Salido, M.D. Gutie´rrez-Valero, R. Lo´pez-Garzo´n,
J.M. Moreno-Sa´nchez, Inorg. Chim. Acta 221 (1994) 177.
[31] J.G. Contreras, G.V. Seguel, J.B. Alderete, Spectrochim. Acta
50A (1994) 371.
[32] R. Lo´pez-Garzo´n, M.L. Godino-Salido, M.D. Gutie´rrez-Valero,
P. Arranz-Mascaro´s, J.M. Moreno, Inorg. Chim. Acta 247
(1996) 203.
[33] A.K. Das, S. Seth, J. Inorg. Biochem. 65 (1997) 207.
[34] H.W. Dias, M.R. Truter, Acta Crystallogr. 17 (1964) 937.
[35] Stoe & Cie, X-RED, Data Reduction Program for Windows,
1995, Darmstadt, Germany.
Structure factor tables are available from the au-
thors. Atomic coordinates and equivalent isotropic dis-
placement parameters, calculated hydrogen atom
parameters, anisotropic thermal parameters and bond
lengths and angles have been deposited at the Cam-
bridge Crystallographic Data Centre (deposition Nos.
137430–1374334). Copies of this information may be
obtained free of charge from: The Director, CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK (Fax: +44-
1223-336-033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
[36] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[37] G.M. Sheldrick, SHELXL-97, Program for the Refinement of
Crystal Structures, University of Go¨ttingen, Germany, 1997.
[38] A.L. Spek, PLATON-98 for Windows, University of Utrecht, The
Netherlands, 1998.
[39] S. Manogaran, D.N. Sathyanarayana, Bull. Soc. Chim. Belg. 90
(1981) 427.
[40] D. Grden´ıc´, Quart. Rev. 19 (1965) 303.
Acknowledgements
This research was supported by the Ministry of Sci-
ence and Technology of the Republic of Croatia
(Grants Nos. 119408 and 00980802).

152
Z. Popo6ic´ et al. / Inorganica Chimica Acta 306 (2000) 142–152
[41] L. Pauling, The Nature of the Chemical Bond, 3rd ed., Cornell
University Press, Ithaca, NY, 1960.
[47] D. Grdenic´, Connections in the crystal structures of mercury
compounds, in: G. Dodson, J.P. Glusker, D. Sayre (Eds.),
Structural Studies of Molecules of Biological Interest, Clarendon
Press, Oxford, 1981, p. 207.
,
[42] C.M.V. Sta˚lhandske, I. Persson, M. Sandstro¨m, M. Aberg,
Inorg. Chem. 36 (1997) 4945.
&
[43] Cambridge Structural Database, Version 5.17, Cambridge Crys-
tallographic Data Centre, 12 Union Road, Cambridge, UK,
1998.
[44] R. Baggio, M.T. Garland, M. Perec, J. Chem. Soc. Dalton
Trans. (1995) 987.
[45] J.P.M. Lornmerse, J.C. Cole, Acta Crystallogr., Sect. B 54
(1998) 316.
[46] A. Bondi, J. Phys. Chem. 68 (1964) 441.
[48] D. Matkovic´-Calogovic´, Ph.D. Thesis (in Croatian; Abstract in
English), University of Zagreb, 1994.
[49] S.C. Nyburg, C.H. Faerman, Acta Crystallogr., Sect. B 41 (1985)
274.
[50] R.C. Seccombe, C.H.L. Kennard, J. Organomet. Chem. 18
(1969) 134.
[51] C. Hansch, A. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165.
[52] D. Kovacˇek, D. Vikic´-Topic´, to be published.
.