Simultaneous formation of both novel bis(μ-sulfur)- and bis(μ-oxo)-bridged copper(II) dimers by reaction of N-benzoyl-N′,N′-dimethylthiourea (HL) with CuCl2. Syntheses, crystal structures and magnetic properties of [{CuL(HL)Cl}2] and

Article abstract of DOI:10.1039/a903517f

Reaction of N-benzoyl-N′,N′-dimethylthiourea (HL) with copper(II) chloride yielded simultaneously two novel dimers: a pale green bis(μ-sulfur) bridged copper(II) dimer 1, [{CuL(HL)Cl}₂], and a dark green bis(μ-oxo) bridged copper(II) dimer 2, [

Full text of DOI:10.1039/a903517f

View Article Online / Journal Homepage / Table of Contents for this issue
Simultaneous formation of both novel bis(ꢀ-sulfur)- and bis(ꢀ-oxo)-
bridged copper(II) dimers by reaction of N-benzoyl-NЈ,NЈ-dimethyl-
thiourea (HL) with CuCl2. Syntheses, crystal structures and
magnetic properties of [{CuL(HL)Cl}2] and [{CuL2}2]†
De-Ji Che,* Gang Li, Xiao-Lan Yao, Yu-Zhu and Da-Peng Zhou
Department of Chemistry, Zhengzhou University, Henan 450052, P. R. China
Received 4th May 1999, Accepted 29th June 1999
Reaction of N-benzoyl-NЈ,NЈ-dimethylthiourea (HL) with copper() chloride yielded simultaneously two novel
dimers: a pale green bis(µ-sulfur) bridged copper() dimer 1, [{CuL(HL)Cl}2], and a dark green bis(µ-oxo) bridged
copper() dimer 2, [{CuL2}2], instead of the mononuclear copper() complexes, [CuL2], usually formed from other
dialkyl-substituted benzoylthiourea ligands. The molecular structures of 1 and 2 have been established by single-
crystal X-ray analyses. In dimer 1 the central copper() ion has a distorted tetrahedral co-ordination environment
with one chloride and three sulfur atoms derived from one terminal and two bridging ligands. The two sulfur atoms
of both bridging ligands with two copper() ions form a strictly planar Cu2S2 bridging core, including two short
[2.330(2) Å] and two longer [2.549(2) Å] Cu–S distances. The dimer 2 may be considered as comprising two
mononuclear [CuL2] complexes bridged by one of the acyl oxygen atoms of each to the copper atom of the other,
resulting in a distorted tetragonal pyramidal geometry at each copper() center, and forming a planar four-membered
Cu2O2 bridging unit which consists of two short [1.944(3) Å] and two longer Cu–O distances [2.668(3) Å], bridging
bond angle Cu–O–Cu* [97.9(1)Њ]. Magnetic measurements for both dimers showed weak antiferromagnetic coupling
between the copper() centers through bis-sulfur or bis-oxo bridges. The best fitting of the experimental magnetic
susceptibility data gave g = 2.110, J = Ϫ33 cmϪ1 for dimer 1 and g = 2.107, J = Ϫ4.5 cmϪ1 for 2.
Much attention has been paid to the investigation of a variety
Experimental
of dinuclear copper() complexes,1 especially those ligated by
Reagent grade chemicals were used as received unless stated
sulfur donor atoms, because of their resemblance to the active
otherwise. Elemental analyses were carried out on a MOD
centers of some binuclear copper proteins, such as hemocyanins
1106 analyzer. Infrared spectra were recorded on a Shimadzu
and tyrosinase, which mediate redox reactions and electron
IR-435 instrument using KBr pellets in the 400–4000 cmϪ1
region. X-Band EPR spectra were recorded with a Bruker 200
D-SRC ESR spectrometer at room temperature. In a typical
transfer in biological systems. Therefore, research into the syn-
thesis, structure and properties of dinuclear copper() com-
plexes of thiolate-containing ligands is expected to shed light
experiment the instrument settings were: modulation frequency,
on the structure and mechanism of action of such metallo-
100 kHz; field modulation intensity, 5 G; receiver gain, 2 × 105;
microwave power, 10 mW.
Variable-temperature magnetic susceptibility data were
proteins. On the other hand, magnetostructural properties of
copper() complexes have also attracted considerable atten-
tion2 due to the potential applications of magnetic exchange
obtained on
a SQUID susceptomer (Quantum Design,
properties in designing new magnetic materials, such as
molecular-based magnets.
MPMS-5) in the temperature range 4.5–300 K with an applied
field of 1 T. All data have been corrected for diamagnetism by
using Pascal’s constants.4
In view of the versatility of co-ordination of acylthioureas
with transition metals, recently we treated N-ferrocenecarb-
onyl-NЈ,NЈ-dimethylthiourea (HLЈ) with copper() chloride
and obtained a bis-sulfur-bridged copper() dimer, [{CuLЈ-
(HLЈ)Cl}2],3 which has a similar co-ordination configuration
to that of the CuA site of cytochrome c oxidase and other
heme copper oxidases in which two distorted tetrahedral
co-ordination units are bridged by two cysteinate ligands.
We now report an extension of our prior investigations to
the reaction of N-benzoyl-NЈ,NЈ-dimethylthiourea (HL) with
copper() chloride. Interestingly, two novel dimers, a bis-
sulfur-bridged copper() dimer 1, [{CuL(HL)Cl}2], and a bis-
oxo-bridged copper() dimer 2, [{CuL2}2], were afforded
simultaneously. Now we present the crystal structures of
both dimers together with their magneto-structural properties.
Preparation of the ligand
N-Benzoyl-NЈ,NЈ-dimethylthiourea was prepared using a simi-
lar procedure to that reported5 by the reaction of benzoyl chlor-
ide with KSCN in anhydrous acetone, and then condensation
with dimethylamine. The crude product was recrystallized from
ethanol to obtain a white crystal, mp 146 ЊC (Found: C, 57.21;
H, 5.64; N, 13.10. Calc. for C10H12N2OS: C, 57.69; H, 5.77; N,
13.46%).
Preparation of the dimers 1 and 2
To an ethanol solution (4 cm3) of ligand HL (0.02 mmol),
copper() chloride (0.04 mmol) in ethanol (3.5 cm3) was added
dropwise at about one drop per 2 h at room temperature. At
first a pale green crystal dimer 1 began to appear within several
days. After about two weeks a second dark green crystal dimer 2
was formed simultaneously. Both crystals were of adequate size
and quality for X-ray studies and easily separated manually.
† Supplementary data available: IR and X-band EPR spectra, experi-
mental and calculated susceptibilities. For direct electronic access see
http://www.rsc.org/suppdata/dt/1999/2683/, otherwise available from
BLDSC (No. SUP 57596, 2 pp.) or the RSC Library. See Instructions
for Authors, 1999, Issue 1 (http://www.rsc.org/dalton).
J. Chem. Soc., Dalton Trans., 1999, 2683–2687
2683

View Article Online
Table 1 Crystal parameters, data collection and refinement for dimers
Dimer 1, [{CuL(HL)Cl}2]: mp 150 ЊC (Found: C, 46.53; H,
4.77; N, 10.53. Calc. for C20H23ClCuN4O2S2: C, 46.69; H, 4.47;
N, 10.89%). Dimer 2, [{CuL}2]: mp 216 ЊC (Found: C, 49.67;
H, 4.60; N, 11.44. Calc. for C20H22CuN4O2S2: C, 50.20; H, 4.60;
N, 11.71%).
1 and 2
1
2
Formula
C40H46Cl2Cu2N8O4S4 C40H44Cu2N8O4S4
M
1029.07
956.18
Crystallography
Crystal system
Space group
a/Å
b/Å
orthorhombic
Pccn(no. 56)
21.637(4)
26.290(5)
9.109(2)
monoclinic
P21/n(no. 14)
10.19(1)
17.138(4)
12.732(2)
102.900(4)
2166(1)
Crystal intensity data were collected on a Rigaku RAXIS-IV
imaging plate area detector using graphite monochromated
Mo-Kα (λ = 0.71070 Å) radiation for unit-cell determination
and data collection. Summaries of the crystallographic data,
structure solution and refinement are given in Table 1.
A total of 30 3.00Њ oscillation frames with an exposure time
of 12.0 min for dimer 1 and 45.0 min for dimer 2 per frame were
used. The structures were obtained by direct methods and
expanded using Fourier techniques. The data were corrected for
Lorentz-polarization effects, the non-hydrogen atoms refined
anisotropically and hydrogen atoms included but not refined.
The data were refined by the SHELXL 93 program6 using all
data for dimer 1 and TEXSAN for dimer 2.6
c/Å
β/Њ
U/Å3
5181.4(18)
4
18 (2)
1.319
1.129
2825/2825
Z
2
15
T/ЊC
1
Dc/g cmϪ3
1.465
1.224
3764
µ(Mo-Kα)/mmϪ1
Reflections collected/
unique
No. observations
(I > 3.00σ(I))
No. variables
Final R1, wR2 [I > 2σ(I)]
(all data)
None
3005
263
0.038, 0.056
0.1070, 0.1737
0.1635, 0.1945
CCDC reference number 186/1545.
See http://www.rsc.org/suppdata/dt/1999/2683/ for crystallo-
graphic files in .cif format.
Table 2 Selected bond lengths (Å) and angles (Њ) for dimer 1
Results and discussion
Formation of the dimers 1 and 2
Metal co-ordination sphere
Cu(1)–S(1)
Cu(1)–S(2)
Cu(1)–S(2*)
2.273(2)
2.330(2)
2.549(2)
Cu(1)–Cl(1)
Cu(1) ؒ ؒ ؒ Cu(1*)
2.331(3)
2.619(2)
Recently we have found that co-ordination of the dimethyl-
substituted ferrocenecarbonylthiourea HLЈ to copper() ion
differs substantially from that of the corresponding diethyl-
substituted analogue HLЉ. The former yields a bis-sulfur-
bridged copper() dimer, [{CuLЈ(HLЈ)Cl}2],3 while the latter
only co-ordinates in a bidentate manner to yield a single
mononuclear copper() complex, [CuLЉ2].7
S(1)–Cu(1)–S(2)
S(1)–Cu(1)–Cl(1)
S(1)–Cu(1)–S(2*)
Cl(1)–Cu(1)–S(2)
112.46(10)
115.92(10)
102.72(8)
108.36(9)
Cl(1)–Cu(1)–S(2*)
S(2)–Cu(1)–S(2*)
Cu(1)–S(2)–Cu(1*)
101.81(9)
115.24(6)
64.76(6)
Terminal ligands
Bridging ligand
In this study, we attempted to replace the ferrocenyl in HLЈ
by a phenyl group to minimize the steric hindrance of the
acylthiourea ligand in order to realize the numerous modes of
co-ordination to copper() ions. Just as expected, the reaction
of N-benzoyl-NЈ,NЈ-dimethylthiourea (HL) with copper()
chloride affords simultaneously both dinuclear copper() com-
plexes with different bridging atoms (sulfur or oxygen) and
different geometries (distorted tetrahedron or square pyramid).
Therefore, for such metal ions as copper with remarkably flex-
ible co-ordination and versatile ligands such as acylthioureas
having both potential O/S donor atoms, decreasing the steric
hindrance of the substituents in the ligand might be favorable
to realize various co-ordination modes.
S(1)–C(3)
N(1)–C(3)
N(2)–C(3)
N(2)–C(4)
O(1)–C(4)
C(4)–C(5)
1.694(8)
1.306(8)
1.385(10)
1.392(9)
1.204(10)
1.515(13)
S(2)–C(12)
N(3)–C(12)
N(4)–C(12)
N(4)–C(14)
O(2)–C(14)
C(14)–C(15)
1.677(9)
1.334(9)
1.410(10)
1.379(9)
1.197(8)
1.518(11)
S(1)–C(3)–N(2)
S(1)–C(3)–N(1)
N(1)–C(3)–N(2)
O(1)–C(4)–N(2)
O(1)–C(4)–C(5)
N(2)–C(4)–C(5)
118.7(5)
122.9(6)
118.4(7)
122.9(8)
120.3(7)
116.7(8)
S(2)–C(12)–N(4)
S(2)–C(12)–N(3)
N(3)–C(12)–N(4)
O(2)–C(14)–N(4)
O(2)–C(14)–C(15)
N(4)–C(14)–C(15)
120.8(5)
122.7(7)
116.5(8)
123.4(7)
119.9(7)
116.6(6)
The synthetic methodology adopted in this work is to control
the relative concentration between the reaction components
by the dropwise addition of metal to ligand solution, or the
reverse, to make the reaction proceed under the conditions of
one component (ligand or metal) in excess. In this way, dimer 1
formed first in the presence of an excess of ligand HL; under
this condition sufficient HL exists in the solution to bridge the
copper() ions, and the dimer 2 formed subsequently only in the
presence of a lower concentration of HL which is favorable to
the bidentate co-ordination of HL. At the same time, the drop-
ping method made the reaction proceed under mild conditions
in favor of the formation of single crystals suitable for X-ray
studies.
Molecular structures
Dimer 1, [{CuL(HL)Cl}2]. An ORTEP8 drawing of dimer 1 is
shown in Fig. 1 together with the numbering scheme. Selected
bond distances and angles are given in Table 2. Each copper()
ion is bonded to one chloride and three sulfur atoms from one
terminal and two bridging acylthiourea ligands, displaying a
distorted tetrahedral geometry. The terminal ligands exhibit
Fig. 1 An ORTEP diagram of the molecular structure of dimer 1,
showing the atomic numbering.
2684
J. Chem. Soc., Dalton Trans., 1999, 2683–2687

View Article Online
an ionic form (L) deprotonated from the thioamido (NHC(᎐S))
view of the dimeric entity is given in Fig. 2 together with the
᎐
group and the bridged species are neutral (HL). The two
copper() ions and the bridging sulfur atoms lie strictly in
plane, forming a Cu2S2 core with two short (2.330(2) Å) and
two longer (2.549(2) Å) Cu–S distances. The short one is nearly
the same as the sum of the tetrahedral covalent radii (2.39 Å).
The core also contains a relatively short Cu ؒ ؒ ؒ Cu separation
of 2.619(2) Å, a narrow bridging Cu–S–Cu* angle of 64.76(6)Њ
and a wide S(2)–Cu–S(2)* angle of 115.24(6)Њ. The dihedral
angle τ (56.60Њ) between the core and the remaining co-
ordination plane, Cl(1)Cu(1)S(1) or Cl(1Ј)Cu(1Ј)S(1Ј), is smaller
than the right angle (90Њ) of a normal tetrahedron, indicating
a rather distorted tetrahedral co-ordination geometry around
copper().
numbering scheme. Selected bond distances and angles are
given in Table 3.
Each copper() center has an elongated square pyramidal
(4ϩ1) environment; the basal plane comprises two oxygen
and two sulfur atoms in cis position from both quasiplanar
deprotonated ligands (LϪ) and the axial co-ordination site is
occupied by a second oxygen atom belonging to the other half
of the dimer. The out-of-plane Cu–O distance of 2.668(3) Å is
somewhat longer than the in-plane distances (bridged Cu–O
1.944(3), non-bridged Cu–O 1.929(3) Å). The bridging Cu2O2
unit is constrained to be planar by the presence of the crystallo-
graphic inversion center in the middle of the dimer. The
intradimeric Cu ؒ ؒ ؒ Cu distance is 3.51 Å and bridging Cu–O–
Cu* angle is 97.9(1)Њ.
Dimer 2, [{CuL2}2]. Dimer 2 is built by two mononuclear
[CuL2] units related through a molecular inversion center. A
From Fig. 2, it is striking that each mononuclear moiety of
the dimer is nearly planar, the greatest deviations from the
mean plane being just 0.629(5) (C(18)) and 0.466(5) Å (C(7)) for
the phenyl rings, while the interval (2.668(3) Å) between both
moieties is large enough to form the bis-oxo-bridging bonds
without steric hindrance. This may result from the substitution
of planar phenyl and smaller methyl groups on the N,NЈ sites
of the ligand.
IR spectroscopic properties
The characteristic IR absorption bands of ligand HL and both
dimers are presented in Table 4. The broad and weak band
around 2900–3200 cmϪ1 observed for free HL may be attributed
to the combined ν(CH) vibrations of the benzene portion and
δ(NH) vibration which is split into two broad bands at 3457
and 3110 cmϪ1 for both dimers. The lack of a ν(SH) band
around 2500 cmϪ1 and the presence of two ν(NH) bands at ca.
3457 and 3110 cmϪ1 unequivocally confirm the S-co-ordination
mode of the ligand in dimers 1 and 2. The presence of a strong
Fig. 2 An ORTEP diagram of the molecular structure of dimer 2,
showing the atomic numbering.
band at 1245–1257 cmϪ1 is tentatively assigned to ν(C᎐S). The
᎐
ν(N–C᎐S) band at 1190 cmϪ1 for free HL shifts to 1120 cmϪ1 for
᎐
dimer 1 and 1130 cmϪ1 for 2; this further evidences that the
ligand is co-ordinated to copper() via the sulfur atom. The
strong stretch at 1685 cmϪ1 assigned to the carbonyl stretch of
the C(O)NH moiety in the uncomplexed ligand is nearly
unchanged at 1690 cmϪ1 for dimer 1, because in this dimer
the carbonyl group does not participate in the co-ordination.
In dimer 2 this vibration virtually disappears due to the form-
Table 3 Selected bond lengths (Å) and angles (Њ) for dimer 2
Metal co-ordination sphere
Cu(1)–S(1)
Cu(1)–S(2)
Cu(1)–O(1)
Cu(1)–O(2)
2.245(1)
2.248(2)
1.929(3)
1.944(3)
Cu(1)–O(2*)
Cu(1) ؒ ؒ ؒ Cu(1*)
2.668(3)
3.51
ation of bis-oxo-bridging to make the ν(C᎐O) vibration
weak.
᎐
S(1)–Cu(1)–S(2)
S(2)–Cu(1)–O(2)
O(1)–Cu(1)–O(2)
O(1)–Cu(1)–S(1)
89.14(7)
93.15(10)
84.8(1)
S(1)–Cu(1)–O(2)
S(2)–Cu(1)–O(1)
O(2)–Cu(1)–O(2*)
Cu(1)–O(2)–Cu(1*)
177.70(8)
167.0(1)
82.1(1)
Further characteristics of the IR spectrum of dimer 2 are the
occurrence of strong and sharp bands at 1420 and 1380 cmϪ1
which are assignable to the C–CH3 symmetric and asymmetric
deformation vibrations, respectively,9 and other strong bands at
1574 and 1498 cmϪ1 may result from the skeletal vibrations of
phenyl rings.10 These four absorption bands are strengthened
considerably in intensity in comparison with those of the “free”
ligand and dimer 1. These observations are consistent with the
molecular structures of dimers 1 and 2.
92.90(10)
97.9(1)
Ligand
S(1)–C(3)
O(1)–C(4)
C(3)–N(2)
N(2)–C(4)
1.734(4)
1.275(4)
1.348(5)
1.321(5)
S(2)–C(13)
O(2)–C(14)
C(13)–N(4)
N(4)–C(14)
1.730(4)
1.278(4)
1.347(5)
1.315(4)
Cu(1)–O(1)–C(4)
O(1)–C(4)–N(2)
C(3)–N(2)–C(4)
S(1)–C(3)–N(2)
O(1)–C(4)–C(5)
S(1)–C(3)–N(1)
133.3(2)
129.1(3)
124.8(3)
128.9(3)
114.9(3)
116.1(3)
Cu(1)–O(2)–C(14)
O(2)–C(14)–N(4)
C(13)–N(4)–C(14)
S(2)–C(13)–N(4)
O(2)–C(14)–C(15)
S(2)–C(13)–N(3)
128.8(2)
128.9(3)
124.9(3)
127.8(3)
115.7(3)
116.8(3)
EPR Spectra
The X-band (9.78 GHz) powder EPR spectra for dimers 1 and
2 have been recorded at 300 K (Fig. 3). The spectrum of 1
exhibits well resolved resonance bands in the middle magnetic
Table 4 Main characteristic IR spectral bands of HL and dimers (cmϪ1
)
Compound
ν(CH), δ(NH)
ν(C᎐S)
ν(N–C᎐S)
ν(C᎐O)
ν(C–CH3)
ν(phenyl ring)
᎐
᎐
᎐
HL
Dimer 1
Dimer 2
2900, 3200
3457, 3110
3456, 3108a
1245, 660
1246, 658
1257, 680a
1190
1120
1130a
1685
1690
—
1460, 1385
1468, 1392
1420, 1380c
1550, 1460
1562, 1468
1574, 1498c
b
a Reduced in intensity. b Disappeared. c Strengthened in intensity considerably.
J. Chem. Soc., Dalton Trans., 1999, 2683–2687
2685

View Article Online
Fig. 4 Thermal variation of χm and χmT for dimer 1. Crosses (×)
represent the total susceptibility, filled circles (᭹) that after subtracting
a diamagnetic correction (see text). Solid line (––) corresponds to the
best fit.
Ng2β2
Fig. 3 X-Band EPR spectra for powdered dimers 1 (a) and 2 (b) at
300 K, ν = 9.78 GHz.
χm =
[1 ϩ ¹ exp(Ϫ2J/kT)]Ϫ1(1 Ϫ P) ϩ
¯
³
3k(T Ϫ θ)
Ng2β2
4kT
(1)
P ϩ χimpurity
field region, indicative of a typical copper() anisotropic (axial)
g tensor, with g⊥ = 2.23 and g|| = 2.05, the average value (taken
as gav = (2g||ϩg⊥)/3) being 2.110. These values allow us to con-
sider the existence of a distorted four-co-ordinated geometry
around the CuII (g|| < g⊥) in dxy ground electronic state. The
spectrum for dimer 2 is typical for a very weakly coupled
copper() ion in an axially elongated square pyramidal sur-
rounding with g⊥ = 2.04 and g|| = 2.14. The average g value is
2.107. From the observed trend of g values (g|| > g⊥ > 2.0023)
the ground electronic state is suggested to be associated with the
derived from the isotropic (Heisenberg) exchange Hamilton
(H = Ϫ2JS1ؒS2) for two interacting S = 1/2 centers, where χm is
expressed per mole of copper atoms, p is the fraction of mono-
meric impurity, and θ a corrective term for interdimer inter-
actions, and the correction term χimpurity = Ϫ696 × 10Ϫ5 emu
molϪ1. The best data fit with eqn. (1) (solid line in Fig. 4) gave
g = 2.110, J = Ϫ33 cmϪ1, p = 0.008 and θ = Ϫ3.5 K. The dis-
crepancy factor σ = [Σ(χobs Ϫ χcalc)2/Σχobs
fits was 1.57 × 10Ϫ3.
]
1/2 in the least-squares
2 Ϫ y2
dx
orbitals.
In the literature two kinds of copper dimers with Cu2S2
bridging cores have been reported as: diamagnetically bis-
sulfur-bridged monovalent copper() dimers [{CuI(HL1)Cl}2]13
and [{CuI(H2L2)Cl}2]14 (HL1 = 2(1H)-pyridinethione, H2L2 =
1-methylimidazoline-2-thione), and paramagnetic divalent
copper() dimers, such as [{CuII(L3)Cl}2]15 (denoted as dimer 3)
and [{CuIILЈ(HLЈ)Cl}2]3 (denoted as dimer 4) (HL3 = pyridine-
2-carbaldehyde thiosemicarbazone, HLЈ = N-ferrocenecarb-
onyl-NЈ,NЈ-dimethylthiourea). To our knowledge, only dimers
3 and 4 have been studied magnetostructurally, and can be
compared with 1.
Magnetic properties
Dimer 1. In the measurements of the variable-temperature
susceptibility for dimer 1, negative diamagnetic values appeared
over the temperature range 50–300 K. Further lowering the
temperature (<50 K) gave positive values which increased with
temperature decrease. The χm vs. T curve still exhibits the
feature of antiferromagnetic exchange as a whole (Fig. 4a). This
situation may be attributed to the lower susceptibility of dimer 1
and the presence of a large amount of diamagnetic impurities
which may stem from the reduction of CuII to CuI by the
acylthiourea ligand, or to the existence of different crystal
forms varying in magnetic interaction. In order to facilitate the
simulation, a diamagnetic correction (χimpurity) was applied to
the experimental values by simple shift along the χm axis (Fig.
4b). However, this does not make any changes in the general
trend of the magnetic response, and does not affect the mag-
netic parameters. The temperature dependence of the molar
magnetic susceptibility χmT for dimer 1 in Fig. 4 illustrates that
at room temperature (300 K) the value of 0.53 emu K molϪ1
(µeff = 2.06 µB) is near to the value of 1.9–2.0 µB normally
observed for copper() compounds. On lowering the temper-
ature, the χmT value gradually decreases to reach a minimum
at 75 K with χmT = 0.30 emu K molϪ1 (µeff = 1.549 µB). Such
behavior reveals an intramolecular antiferromagnetic inter-
action between copper() ions with a spin-triplet ground state.
However, on further decreasing the temperature below 50 K,
another discontinuous χmT vs. T curve appears, which has a
different magnetic exchange behavior. This may be due to
the presence of small amounts of uncoupled paramagnetic
impurity.
In dimer 3 the copper() center has a distorted square-
pyramidal (4ϩ1) geometry, the bridging S atom is axial to one
of the two Cu atoms bridged and orthogonal to the magnetic
2
2
orbital (dx Ϫ y ), resulting in a weak antiferromagnetic inter-
action, J = Ϫ4.7 cmϪ1. In 4 the copper() center is in a distorted
tetrahedral co-ordination geometry, the magnetic orbital (dxy) is
roughly toward the bridging S orbital, leading to a larger
J value of Ϫ196.3 cmϪ1. However, 1 has a similar tetrahedral
co-ordination sphere, but the J value, Ϫ33 cmϪ1, is substantially
smaller than that of 4. The weak magnetic interaction in dimer
1 may be related to the geometry of the copper co-ordination
sphere as shown in Scheme 1. The apical co-ordination distance
Cu–S(2)*, 2.549(2) Å, is apparently longer than that of
three others: Cu–S(1), 2.273(2); Cu–S(2), 2.330(2) and Cu–
Cl(1), 2.331(3) Å, respectively, and the copper center is only
0.156 Å apart from the trigonal basal plane, S(1)S(2)Cl(1).
Apparently, the copper center is in an elongated tetrahedral
environment. In addition, the dihedral angle τ between the
bridging core plane Cu2S2 and the remaining co-ordinating
plane Cu(1)S(1)Cl(1) is 56.50Њ in dimer 1, which is smaller than
88.55Њ in 43 and the normal tetrahedral τ value of 90Њ, con-
sequently in 1 the distortions from idealized tetrahedral
symmetry (Td) are so great that orbital overlap is probably
A preliminary analysis of the experimental susceptibility was
performed by use of the modified Bleaney–Bowers eqn. (1),11,12
2686
J. Chem. Soc., Dalton Trans., 1999, 2683–2687

View Article Online
Conclusion
Acylthiourea is a versatile complexing agent having potential
O/S donor atoms bonded in mono- or bi-dentate fashion,
and copper() is a flexible cation exhibiting plasticity of co-
ordination geometry. In order to research the different possible
co-ordination modes between acylthiourea and copper()
ion, we think that diminishing the steric hindrance of the
substituents at the ligand may be a crucial factor. Our recent
investigations demonstrated that dimethyl-substituted ferro-
cenecarbonylthiourea can be co-ordinated to CuII in an unusual
bis-sulfur bridged manner, while the dimethyl- and phenyl-
substituted acylthioureas show a noteworthy versatility giving
rise to both bis-sulfur- and bis-oxo-bridged co-ordination.
Therefore, the substituent groups at the ligand play an essential
role in the stereochemistry of the co-ordination compounds.
These results may provide a useful channel to study the co-
ordination modes of sulfur-containing binuclear copper com-
plexes in biological systems, or to seek ferromagnetically
coupled copper() dimers for new magnetic materials.
Scheme 1 The geometry of bis-sulfur-bridged co-ordination sphere in
dimer 1.
In respect of the methodology used in this work, a dropwise
addition was adopted to make the reaction occur in a regular
sequence and under mild conditions in favor of formation of
different products in different concentration ranges and advan-
tageous for the formation of single crystals suitable for X-ray
studies.
Acknowledgements
Financial support from the Henan Science and Technology
Committee is gratefully acknowledged.
References
1 E. I. Solomon, U. M. Sundaram and T. E. Machonkin, Chem. Rev.,
1996, 96, 2563; N. Kilajima and Y. Moro-Oka, Chem. Rev., 1994, 94,
737; K. A. Magnus, H. Ton-that and J. E. Carpenter, Chem. Rev.,
1994, 94, 727; E. I. Solomon, M. J. Baldwin and M. D. Lowery,
Chem. Rev., 1992, 92, 521; Bioinorganic Chemistry of Copper, eds.
K. D. Karlin and Z. Tyeklar, Chapman & Hall, New York, London,
1993; T. G. Spiro, Copper Proteins, Wiley, New York, 1981.
2 W. E. Hatfield, in Magneto structure Correlations in Exchange
Coupled Systems, eds. R. D. Willett, D. Gatteschi and O. Kahn,
Reidel, Dordrecht, 1985, p. 555; O. Kahn, Angew. Chem., Int. Ed.,
1998, 24, 834.
3 D. J. Che, X. L. Yao, G. Li and Y. H. Li, J. Chem. Soc., Dalton
Trans., 1998, 1853.
4 R. L. Carlin, Magnetochemistry, Springer, Berlin, Heidelberg, New
York, Tokyo, 1986.
5 Y. F. Yuan, L. Y. Zhang, J. P. Cheng and J. T. Wang, Transition Met.
Chem., 1997, 22, 281.
6 G. M. Sheldrick, SHELXL 93, Program for refining crystal
structures, University of Göttingen, 1993; TEXSAN, Crystal
Structure Analysis Package, Molecular Structure Corporation,
Houston, TX, 1985 and 1992.
7 O. Seidelman, L. Bayer and R. Richler, Z. Naturforsch., Teil B,
1995, 50, 1679.
8 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
9 N. Yamazaki and Y. Hohokable, Bull. Chem. Soc. Jpn., 1971, 44, 63.
10 R. M. Silverstein, G. C. Bassler and T. C. Morrill, Spectrometric
Identification of Organic Compounds, 4th edn., Wiley, New York,
1981.
11 K. T. McGregor, J. A. Barnes and W. E. Hatfield, J. Am. Chem. Soc.,
1973, 95, 7993.
12 S. Sikorov, I. Brouche-Waksman and O. Kahn, Inorg. Chem., 1984,
23, 490.
13 E. C. Constable and P. R. Raithby, J. Chem. Soc., Dalton Trans.,
1987, 2281.
14 J. R. Creighton, D. J. Gardiner and A. C. Goruin, Inorg. Chim. Acta,
1985, 103, 195.
15 G. T. Javier, M. K. Urtiaga and R. Cortes, J. Chem. Soc., Dalton
Trans., 1994, 2233.
Fig. 5 Thermal variation of χm and χmT for dimer 2. Solid line (––)
corresponds to the best theoretical fit.
greatly diminished, and hence the coupling interaction is
lowered.
Dimer 2. The magnetic susceptibilities were determined over
the temperature range 4.5–300 K. As seen in Fig. 5 the suscepti-
bilities increase slightly as the temperature decreases from 300
to 100 K, and then increase rapidly through a maximum at
about 8 K, and eventually decrease. The shape of the plot
reveals behavior typical of weakly antiferromagnetic exchange
interactions. The temperature dependence of the χmT product
for dimer 2 could be fitted very well by the Bleaney–Bowers
model from the temperature 4.5 to 300 K. The χmT values grad-
ually decrease from 1.104 emu K molϪ1 (µeff = 2.97 µB) at 300 K
to 0.095 emu K molϪ1 (µeff = 0.87 µB) at 4.5 K. The feature of
this curve indicates an antiferromagnetic interaction between
the coper() centers. The magnetic data were fitted well by the
Bleaney–Bowers eqn. (1) with θ = 0, and Nα = 240 × 10Ϫ5 emu
molϪ1, as indicated by the solid line curve in Fig. 5. The best
fit parameters are J = Ϫ4.5 cmϪ1, g = 2.107, p = 0.015. The
discrepancy σ in the least-square fits was 1.318 × 10Ϫ3. The
weak character of the antiferromagnetic interaction can be
interpreted as due to the fact that the copper center has a
distorted square pyramidal co-ordination geometry and
2 Ϫ y2
dx
ground state. The bridging Cu–O–Cu–O plane is no
longer the xy plane and the orbitals containing the unpaired
spins (which lie in the xy planes) overlap only very poorly
with the bridging orbitals. Hence, the magnitude of J is much
smaller in dimer 2.
It should be mentioned again that the average g values result-
ing from these fits for both dimers agree well with the values
determined from the EPR spectra.
Paper 9/03517F
J. Chem. Soc., Dalton Trans., 1999, 2683–2687
2687