Synthesis, characterisation and biological aspects of copper(II) dithiocarbamate complexes, [Cu{S2CNR(CH2CH 2OH)}2], (R = Me, Et, Pr and CH2CH 2OH)

Article abstract of DOI:10.1016/j.molstruc.2010.11.048

Cu(II) dithiocarbamates, [Cu{S₂CNR(CH₂CH ₂OH)}₂], R = Me (1), Et (2), Pr (3) and CH ₂CH₂OH (4), have been prepared from HNR(CH ₂CH₂OH) (R = Me, Et, Pr and CH₂CH ₂OH), CS₂ and Cu(OAc)₂. Characterisation of the complexes were generally achieved by infrared and EPR spectroscopies and, in addition, for (2) and (3), by X-ray crystallography at 120 K. Complex (2) crystallises as a Cu-S linked dimer, in which the CH₂CH₂OH groups have a cis arrangement in each monomer but are trans to those in the other monomer partner. On the other hand complex (3) exists in the solid state in the form of two similar and independent centrosymmetric monomers. The weak antiferromagnetic coupling, present in similar complexes, was absent in complexes (1)-(3). The in vitro activity of (1)-(4) was investigated against colonies of Candida albicans, Sthaphyloccocus aureus and Pseudomonas aeruginosa. They all displayed MIC (minimal inhibitory concentration) values against C. albicans close to those found for Fluconazole. All complexes were inert towards Gram-negative or Gram-positive bacteria, S. aureus and P. auruginosa, respectively.

Full text of DOI:10.1016/j.molstruc.2010.11.048

Journal of Molecular Structure 988 (2011) 1–8
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Synthesis, characterisation and biological aspects of copper(II) dithiocarbamate
complexes, [Cu{S CNR(CH CH OH)} ], (R = Me, Et, Pr and CH CH OH)
2
2
2
2
2
2
a,
⇑
a
a
a
Geraldo M. de Lima , Daniele C. Menezes , Camila A. Cavalcanti , Jaqueline A.F. dos Santos ,
a
a
a
b
c
Isabella P. Ferreira , Eucler B. Paniago , James L. Wardell , Solange M.S.V. Wardell , Klaus Krambrock ,
Isolda C. Mendes , Heloisa Beraldo
Departamento de Química, ICEx, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil
CHEMSOL, 1 Harcourt Road, Aberdeen AB15 5NY, Scotland, United Kingdom
Departamento de Física, ICEx, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil
a
a
a
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 August 2010
Received in revised form 17 November 2010
Accepted 17 November 2010
Available online 13 December 2010
Cu(II) dithiocarbamates, [Cu{S CNR(CH CH OH)} ], R = Me (1), Et (2), Pr (3) and CH CH OH (4), have been
2
2
2
2
2
2
prepared from HNR(CH
complexes were generally achieved by infrared and EPR spectroscopies and, in addition, for (2) and (3), by
X-ray crystallography at 120 K. Complex (2) crystallises as a Cu–S linked dimer, in which the CH CH OH
2 2 2 2 2 2
CH OH) (R = Me, Et, Pr and CH CH OH), CS and Cu(OAc) . Characterisation of the
2
2
groups have a cis arrangement in each monomer but are trans to those in the other monomer partner. On
the other hand complex (3) exists in the solid state in the form of two similar and independent centro-
symmetric monomers. The weak antiferromagnetic coupling, present in similar complexes, was absent in
complexes (1)–(3). The in vitro activity of (1)–(4) was investigated against colonies of Candida albicans,
Sthaphyloccocus aureus and Pseudomonas aeruginosa. They all displayed MIC (minimal inhibitory concen-
tration) values against C. albicans close to those found for Fluconazole. All complexes were inert towards
Gram-negative or Gram-positive bacteria, S. aureus and P. auruginosa, respectively.
Keywords:
Copper(II) complexes
Dithiocarbamate complexes
Spectroscopic studies
Candida albicans
Sthaphyloccocus aureus
Pseudomonas aeruginosa
Ó 2010 Elsevier B.V. All rights reserved.
1
. Introduction
include thermolysis [16–19], uses in analytical chemistry [20–24],
and interactions with NO and their biological applications
1
2 ꢀ
Dithiocarbamate ligands, [S
2
CNR R ] , have found extensive
[25,26]. It is apparent that biological studies of similar derivatives
have not really attracted much attention. This has been so despite
the possibility that the increased hydrophilicity of these com-
pounds, over compounds not bearing hydroxyl groups, may have
significant biological implications.
use in coordination chemistry [1,2]. Their wide range of applica-
tions, e.g. in industry, agriculture and medicine, has generated a
large collection of crystallographic data for their metal complexes
1
2
[
3]. The majority of the complexes studied have simple R and R
groups, such as methyl, ethyl and phenyl. Complexes with dithio-
carbamate ligands having functional substituted organic groups,
such as R and or R equal to CH CH OH, are increasingly being
2 2
studied, with crystal structures reported for a number of metal
complexes including those of alkali metals [4] copper [5,6], nickel
Following our general interest in the biological properties of
metal complexes [27–30], we now report results on the anti-fungal
and anti-bacterial properties of complexes (1)–(4), as well as their
IR and EPR spectra. In addition, the crystal structures (2) and (3)
are described.
1
2
[
7–9], zinc [10], mercury [11] and antimony [12,13]. Clearly the ex-
tra structural aspects generated by the functional groups [4–13]
have encouraged interest in these compounds.
2
. Experimental
By far the best studied of the reported [Cu{S
OH)} ] compounds, where R = Me, Et and CH CH OH is the latter
14–26]. Very limited reports have been made on (1) and (2),
16]. Apart from informations on the synthesis, spectroscopy
14,15] and crystal structure of (4) [5,6], other areas of interest
2 2 2
CNR(CH CH -
2
2
.1. Chemistry
2
2
2
[
[
[
.1.1. Materials and methods
All starting materials were purchased from Aldrich, Merck or
Synth and used as received. Infrared spectra were recorded in the
range of 4000–400 cm
ꢀ
1
with samples in KBr pellets using a
Perkin–Elmer 283B spectrometer. Carbon, hydrogen and nitrogen
analyses were performed on a Perkin–Elmer PE-2400 CHN-analysis
using tin sample-tubes.
⇑
Corresponding author.
E-mail address: delima.geraldo@gmail.com (G.M. de Lima).
0
022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.11.048

2
G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
2.1.2. Synthesis of (1) (R = Me)
2 2
2.1.5. Synthesis of (4) (R = CH CH OH)
To a solution of HN(Me)(CH
2
CH
2
OH) (2.50 g, 0.017 mol) in cold
Prepared accordingly by reacting HN(CH
0.03 mol), CS (1.32 g, 0.03 mol) and Cu(OAc) (2.99 g, 0.015 mol).
Yield 71%. Mp. 164.0–166.4 °C.
2
CH
2
OH)
2
(3.0 g,
ethyl ether at ꢀ10 °C, was successively added carbon disulfide
0.75 g, 0.017 mol) and a suspension of Cu(OAc) O (1.70 g,
ꢁH
.009 mol) in cold ether. The reaction mixture was allowed to
2
2
(
0
2
2
ꢀ
1
IR (
Analysis for C10
5.8 (5.6), N 7.5 (7.8).
m
/cm ): 1517 (
m
C–N) and 993 (
mC–S).
reach room temperature and after stirring for 2 h, the solid product
was collected. The dark brown residue was re-crystallised from
methanol. Yield 68%. Mp. 196.4–198.5 °C.
20 2 4 4
H N S O Cu found% (calc.%): C 34.1 (33.4), H
ꢀ1
IR (
Analysis for C
5.3), N 8.9 (9.3).
m
/cm ): 1507 (
m
C–N) and 989 (
mCS).
2.1.6. Electron paramagnetic resonance
8
H16CuN
2
O
S
2 4
found% (calc.%): C 32.3 (32.1), H 5.6
Electron paramagnetic resonance spectra were recorded on a
spectrometer constructed with a 500 mW klystron (Varian), a com-
mercial cylindrical resonance cavity (Bruker), an electromagnet
(
(
Varian) with maximum field amplitude of 800 mT and a He flux
2
.1.3. Synthesis of (2) (R = Et)
Prepared similarly using HN(Et)(CH
(0.88 g, 0.02 mol) and Cu(OAc)
ꢁH
cryosystem (Oxford) for low temperature measurements. Spectra
were recorded as first-derivates using common 100 kHz field
modulation. For g factor calibration, 1,1-diphenyl-2-picrylhydrazyl
was used as the standard (g = 2.0037). Polycrystalline samples
were measured as fine powders in cylindrical borosilicate tubes
2
CH
2
OH) (2.74 g, 0.02 mol),
CS
6
2
2
2
O (1.99 g, 0.01 mol). Yield
8%. Mp. 171.1–171.9 °C.
ꢀ
1
IR (
m
/cm ): 1499 (
m
C–N) and 985 (
20CuN found% (calc.%): C 36.9 (36.6), H
.5 (6.1), N 8.4 (8.5).
mCS).
Analysis for C10
H
2 2 4
O S
(
Wilmad).
6
2.1.7. X-ray crystallography
2
.1.4. Synthesis of (3) (R = Pr)
Similarly prepared employing HN(Pr)(CH
.014 mol), CS (0.62 g, 0.014 mol) and Cu(OAc)
Intensity data for [Cu{S
(3), were collected at 120 K with Mo Ka
2
CNR(CH
2
CH
2
OH)}
2
], R = Et (2) and Pr
radiation using the
j-goniostat Bruker–Nonius CCD camera of the EPSRC crystallo-
2
CH
2
OH) (2.45 g,
0
2
2
2
ꢁH O (1.4 g, 0.007
mol). Yield 65%. Mp. 111.7–122.0 °C.
graphic service, based at the University of Southampton, UK. Data
collection was carried using the program COLLECT [31] and data
reduction and unit cell refinement were achieved with the COL-
LECT and DENZO [32] programs. Corrections for absorption, by
ꢀ
1
IR (
Analysis for C12
.1 (6.8), N 7.6 (7.9).
m
/cm ): 1512 (
m
C–N) and 997 (
mC–S).
H
24CuN O S
2 2 4
found% (calc.%): C 41.2 (40.5), H
7
Table 1
Crystallographic data for complexes (2) (R = Et) and (3) (R = Pr).
Compound
(2) R = Et
20CuN
392.06
120(2)
(3) R = Pr
Empirical formula
Formula weight
Temperature, K
C
10
H
2
O
S
2 4
12 2 2 4
C H24CuN O S
420.11
120(2)
0.71073
0
0.71073
Wavelength, ÅA
Crystal system
Space group
Monoclinic
P21/n
Triclinic
P-1
Unit cell dimensions
0
a, ÅA
0
9
1
1
.4925(2)
0.9232(3)
5.7429(4)
6.42860(10)
11.5308(3)
12.6699(4)
b, ÅA
0
c, ÅA
a
, °
b, °
, °
90
92.0870(10)
97.863(2)
90.401(2)
929.66(4)
102.2230(15)
90
1595.35(7)
c
0
3
Volume, ÅA
Z
4
2
3
Density (calculated), Mg/m
Absorption coefficient, mm
F(0 0 0)
1.632
1.890
812
1.501
1.628
438
ꢀ
1
Crystal size, m
Theta range for data coll., °
Index ranges
0.28 ꢂ 0.16 ꢂ 0.05
0.10 ꢂ 0.10 ꢂ 0.01
3.20–27.48
ꢀ8 ꢃ h ꢃ 8
ꢀ14 ꢃ k ꢃ 14
ꢀ16 ꢃ l ꢃ 16
14,004
3.24–25.00
ꢀ10 ꢃ h ꢃ 11
ꢀ
ꢀ
12 ꢃ k ꢃ 12
18 ꢃ l ꢃ 18
Reflections collected
Independent reflections
18,609
2799
4244
[
R(int) = 0.0445]
[R(int) = 0.0432]
3662
Reflections obd. (>2sigma)
Data completeness
2480
0.998
0.993
Absorption correction
Refinement method
None
None
Full-matrix least-squares on F2
Full-matrix least-squares on F2
4244/0/197
1.108
Data/restraints/parameters
2799/0/185
1.053
2
Goodness-of-fit on F
Final R indices [I > 2sigma(I)]
R indices (all data)
R
1
= 0.0525
wR = 0.1442
= 0.0584
wR = 0.1495
3.247 and ꢀ0.641
R
1
= 0.0374
wR = 0.0739
= 0.0482
wR = 0.0796
0.375 and ꢀ0.366
2
2
R
1
R
1
2
2
0
3
Largest diff. peak and hole, e. ÅA

G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
3
comparison of the intensities of equivalent reflections, were ap-
plied using the program SADABS [33]. The structures were solved
2
by direct methods using SHELXS-97 [34] and refined against F
with SHELXL-97 [35]. Hydrogen atoms were placed in calculated
positions and refined as riding. The program ORTEP-3 for Windows
[
36] was used in the preparation of Figures and PLATON [37] in the
calculation of molecular geometry. Crystal data and structure
refinement details are listed in Table 1.
2.2. Anti-fungal activity
The standard method for anti-fungal susceptibility testing pro-
posed by the NCCLS M27-A2 [38] was used for in vitro susceptibil-
ity tests. Candida albicans ATCC18804, stored in Sabouraud broth,
was sub-cultured for testing in the same medium and grown at
6
ꢀ1
3
5 °C for 24 h. An inoculum of 10 cells mL was prepared by a
spectrophotometric method. Serial dilutions of the compounds dis-
solved in DMSO were prepared to provide final concentrations of
ꢀ
1
6
1.5–0.12 lg mL . A 24 h old inoculum was added to each tube
3
ꢀ1
to provide a final concentration of 10 cells mL . Minimum inhib-
itory concentration (MIC), the lowest concentration to give 100%
inhibition of growth, was determined visually after incubation
for 24 h at 35 °C. Tests using fluconazole as a negative control
and DMSO as a positive control were carried out in parallel. All
tests were performed in triplicate with full agreement between
the results.
2.3. Anti-bacterial activity
The standard method for anti-bacterial susceptibility testing pro-
posed by the NCCLS M7-A6 [39] was used for in vitro susceptibility
tests. Pseudomonas aeruginosa ATCC 25853 and Sthaphyloccocus
aureus ATCC 6538 stored in Müller–Hilton broth were sub-cultured
for testing in the same medium and incubated at 37 °C for 24 h and
Fig. 1. Complex (1) (R = Et) (a) atom numbering system and atom arrangements:
i
0
0
only, O1a and O1a , of the pair of disordered OH sites, [O1a + O1c] and [O1a + O1c ]
are shown; probability ellipsoids are drawn at the 50% level and hydrogen atoms
have been omitted for clarity; (b) packing of the molecules. Symmetry operations
are shown in Table 2.
6
h, respectively. The bacterial cells of the two strains were sus-
8
ꢀ1
pended in Müller–Hilton broth to produce inocula of 10 CFU mL
,
determined by a spectrophotometric method. Serial dilutions of
the compounds, previously dissolved in DMSO, were prepared in
test tubes with Müller–Hilton broth to final concentrations of
The
000 and 1490–1520 cm , respectively, for all complexes. The
values of the (CAS) indicate a bidentate chelating form of the li-
gand towards the metal cation [40], according to the X-ray results.
The infrared spectra of (2), which display an extra intermolecular
CuAS interaction, does not show any difference if compared to
m(CAS) and m(CAN) were observed in the ranges of 980–
ꢀ1
1
ꢀ1
ꢀ1
2
70–1 lg mL for P. aeruginosa and 160–0.6 lg mL for S. aureus.
m
An old inoculum of each strain was added to the tubes to obtain final
5
concentrations of 10 CFU. Minimum inhibitory concentration
(
MIC), the lowest concentration to give 100% inhibition of growth,
were determined visually after incubation for 24 h at 37 °C. Tests
using tetracycline as a negative control and DMSO as a positive
control were carried out in parallel. All tests were performed in
triplicate with full agreement between the results.
the spectra of (3) in the region of m(CAS) and m(CAN) absorption.
3.2. Crystallographic determination
3
. Results and discussion
The structures of [Cu{S
3), were determined by X-ray crystallography using data collected
at 120 K.
Generally [Cu{S
2 2 2 2
CNR(CH CH OH)} ], R = Et (2) and Pr
(
3.1. General
1
2
2
2
CNR R } ], displays two distinct structural
All complexes were obtained from Cu(OAc)
HNR(CH CH OH), see Eq. (1), as crystalline and stable brown-
2 2
coloured solids, readily soluble in polar solvents such as ethanol,
methanol and acetone.
2
,
CS
2
and
arrangements: (a) a square planar monomer, arising from chelating
dithiocarbamate ligands and (b) a 5-coordinate dimer, resulting
from square planar monomers linking by mutual inter-dimer CuAS
coordination. However a less common infinite polymeric chain has
ð1Þ

4
G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
Fig. 3. EPR spectra for the complexes (black solid lines) measured at room
temperature with microwave frequency of 9.39 GHz. Calculated spectra based on
the electronic Zeeman interaction of isolated Cu(II) ions are shown as coloured
lines.
Fig. 2. Complex (3) (R = Pr): (a) atom arrangement and numbering scheme for
Molecule A:Molecule B is numbered similarly with suffix b in place of a; probability
ellipsoids are drawn at the 50% level and hydrogen atoms as spheres of arbitrary
radius; (b) a view of the OAHꢁ ꢁ ꢁO hydrogen bonding arrangements and formation of
the cyclic tetramers. Symmetry operations are listed in Table 3.
indicated in Fig. 2a. Within each of the monomeric units, the
2
-hydroxyethyl groups are in a cis arrangement. Both these groups
been also described in the literature [40]. Structures (a) and (b)
were determined in this study: complex (2) was isolated as a di-
mer, see Fig. 1, while (3) had the monomeric structure, see Fig. 2.
It is notable that very recently a monomeric form of (4)
R = CH CH OH) [6] has been reported in contrast to the dimeric
2 2
form known for some years [5]. The new crystallographic determi-
nation revealed a distorted square planar geometry comprising the
have a trans relationship to the 2-hydroxy groups in the other
monomeric unit.
Complex (3) was isolated as a centrosymmetric monomeric spe-
cies, with the asymmetric unit consisting of two similar but inde-
pendent fragments, each containing one dithiocarbarmate ligand
and a copper ion, Fig. 3. In each of the centrosymmetric molecules
of (3), Mol.A and Mol.B, the 2-hydroxyethyl groups are in a trans
arrangement, which has also been observed in other unsymmetri-
4
CuS environment. In both cases, the crystals used in the structural
1
2
determinations, based on data collected at 298 K, were grown from
methanol solutions: from the scant synthetic and crystal growth
details in the two articles, it is impossible to ascertain any
differences in the procedures used to obtain the different crystals.
Generally the steric bulk of the organic groups on nitrogen is an
important factor, with smaller groups leading to dimers [40–45], in
contrast to the bulkier units providing monomers [40,46–53]. With
bulk intermediate, the form can be solvent dependent, as with
2 2
cal substituted monomeric compounds, e.g., [Cu{S CNR R } ]
1
2
(R , R = MePh) [50].
Selected geometric parameters for (2) and (3) are shown in
Tables 2 and 3, respectively. The dithiocarbamate ligands are
slightly asymmetrically bonded to Cu, with the basal CuAS bond
lengths in the regions found for other copper dithiolates.
As generally found for dimeric copper(II) dithiocarbamates, the
bridging S atom in (2) is involved with the longest CuAS basal bond
length. The length of the bridging CuAS bond and the Cuꢁ ꢁ ꢁCu sep-
aration in copper dithiolate dimers vary with the organic groups on
nitrogen groups, see Table 4. The closest Cuꢁ ꢁ ꢁCu distance in mono-
meric (3) was calculated to be 9.3206(2) Å compared to the CuACu
distances in dimers being between 3.4 and 3.9 Å.
1
2
1
2
[
Cu{S
EtOH/CHCl
meric [40,46] and dimeric forms [46,55], respectively. Also both
2
CNR R }
2
] (R , R = Bu
2
), for which recrystallisation from
3
or CHCl
3
/petroleum ether led to formation of mono-
1
2
monomeric and dimeric forms of [R , R = Pr, CH
2
(Pr-cyclo)] and
] were found within the same crystal [46,48].
In this study, complex (2) was isolated in the dimeric form, the
1
2
[
2 6
R , R = cyclo-(CH )
While the hydroxyl groups in (2) and (3) have little influence on
the geometries at Cu, they do have major impacts on the packing of
the molecules and do lead to packing difference with compounds
asymmetric unit being composed of one molecule, Fig. 2, and as
such is similar to that found for dimeric (4) [5]. Disorder is appar-
ent in one of the 2-hydroxyethyl groups in each of the monomeric
units making up the dimer: two almost equally populated sites,
O1a and O1c, were found: only one each of these sites, O1a is
1
2
2 2
[Cu{S CNR R } ], having unsubstituted alkyl groups. Unfortu-
nately, the disorder of one HO group in (2) prevents a readily
describable account of the hydrogen bonding arrangements of

G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
5
Table 2
Selected crystallographic parameters (Å, °) for complex (2) (R = Et).
(
a) Bond angles and lengths
Cu1AS1A
Cu1AS1B
Cu1AS2A
2.3113(13)
2.3157(13)
2.7936(13)
1.220(14)
1.284(15)
160.07(5)
76.92(5)
102.41(5)
101.04(5)
95.07(5)
Cu1AS2B
Cu1AS2A
S2AACu1i
C3BAO1B
2.3154(13)
2.3295(13)
2.7936(13)
1.366(8)
i
C3AAO1A
C3AAO1Ca
S1AACu1AS2B
S2BACu1AS1B
S2BACu1AS2A
S1AACu1AS2A
S1BACu1AS2A
Cu1AS2AACu1
S1AACu1AS1B
S1AACu1AS2A
S1BACu1AS2A
S2BACu1AS2Ai
S2AACu1AS2Ai
101.21(5)
76.75(4)
172.31(5)
98.89(4)
92.60(4)
i
i
i
87.40(4)
O1AAC3AAO1C
O1C AC3AAC2A
106.6(12)
112.5(12)
O1AAC3AAC2A
O1BAC3BAC2B
116.8(7)
114.5(5)
a
(
b) Hydrogen bonding parameters
DAHꢁ ꢁ ꢁA
Symmetry operation
DAH
Hꢁ ꢁ ꢁA
Dꢁ ꢁ ꢁA
DAHꢁ ꢁ ꢁA
O1AAH1Aꢁ ꢁ ꢁS2A
O1BAH1Bꢁ ꢁ ꢁO1B
C2AAH2A2ꢁ ꢁ ꢁS2A
C2BAH2Bꢁ ꢁ ꢁS2B
C2BAH2B2ꢁ ꢁ ꢁS1A
C4AAH4A2ꢁ ꢁ ꢁS1A
C4BAH4B1AS1B
C5BAH5B2ꢁ ꢁ ꢁO1A
0.84
0.84
0.99
0.99
0.99
0.99
0.99
0.98
2.75
2.13
2.59
2.58
2.85
2.57
2.62
2.59
3.463(14)
2.921(9)
3.062(6)
3.085(5)
3.837(6)
3.069(5)
3.032(6)
3.468(16)
144
157
109
111
173
111
105
149
1 ꢀ x, 2 ꢀ y, 1 ꢀ z
1/2 + x, 3/2 ꢀ y, 1/2 + z
x, ꢀ1 + y, z
(
c) CAHꢁ ꢁ ꢁ.
p interaction
CAHꢁ ꢁ ꢁCg
HꢁꢁCg
H
perp
C
CAHꢁ ꢁ ꢁCg
Cꢁ ꢁ ꢁCg
CAH, p
C5BAH5B3ꢁ ꢁ ꢁCg2
1/2 ꢀ x, ꢀ1/2 + y, 1/2 ꢀ z
2.90
2.708
21.09
139
3.695(7)
70
(
d)
pꢁ ꢁ ꢁp interactions
Cg(I)ꢁ ꢁ ꢁCg(J)
Cgꢁ ꢁ ꢁCg
a
b
CgIperp
CgJperp
Slippage
i
Cg(1)ꢁ ꢁ ꢁCg(1 )
1 ꢀ x, 2 ꢀ y, -z
1 ꢀ x, 2 ꢀ y, ꢀz
1 ꢀ x, 2 ꢀ y, ꢀz
3.7322(18)
3.9384(18)
3.9384(18)
0.00
20.39
20.39
42.73
34.93
41.90
2.742
41.90
34.93
2.741
2.931
3.229
2.533
3.229
2.931
i
Cg(1)ꢁ ꢁ ꢁCg(2 )
i
Cg(2)ꢁ ꢁ ꢁCg(1 )
(
e) Cuꢁ ꢁ ꢁp interactions
Cg(1)ꢁ ꢁ ꢁCu
3.274
Cu-Perp
2.744
b
Cg(1)ꢁ ꢁ ꢁCu
1 ꢀ x, 2 ꢀ y, ꢀz
33.06
Cg(2) is the centroid of the ring defined by Cu1B, S1B, C1B, S2B; Gamma is the angle at H between the vectors Hꢁ ꢁ ꢁCg and Hperp. CAH,
p is an estimate of the significance of the
interaction.
Cg(1) and Cg(2) are the centroids of the rings defined by Cu1, S1A, C1A, S2A; and Cu1, S1B, C1B, S2B, respectively Alpha is the dihedral angle between planes I and J and is
identically zero because the overlapping rings are related to one another by unit cell translation. Beta is the angle between Cg(I)ꢁ ꢁ ꢁCg(J) and CgIperp where CgIperp is the
perpendicular from Cg(I) to ring J. Slippage is the distance between Cg(J) and perpendicular projection of Cg(I) on ring J.
a
O1C is the alternative site to O1A of the hydroxyl oxygen atom.
Symmetry operation: i = ꢀx + 1, ꢀy + 2, ꢀz.
i
the hydroxyl groups. However it is clear that the OAHꢁ ꢁ ꢁO hydro-
gen bonds link molecules, see Table 2. In addition to the OAHꢁ ꢁ ꢁO
hydrogen bonds, molecules are also linked by CAHꢁ ꢁ ꢁS, and
based only on the electron Zeeman interaction. The calculations re-
veal that all Cu(II) centres have nearly axial g tensors. The spin
Hamiltonian parameters are listed in Table 5.
CAHA
p
and other intermolecular interactions, see Table 2. Packing
The EPR spectra are typical for isolated Cu(II) with S = ½ and the
of molecules of (2) is shown in Figs. 2b.
EPR lines are due to the allowed
tions, corresponding to coupled Cu(II) ions at half-field (transitions
= ±2 of a spin triplet), were not observed. From Fig. 4 and
s
Dm = ±1 transitions. EPR transi-
A much simpler picture can be drawn for (3), see Fig. 3b. Here,
the molecules, with trans 2-hydroxyethyl groups, are linked into
cyclic tetrameric arrays by OAHꢁ ꢁ ꢁO hydrogen bonds, see Table 3.
Further contacts between the monomers are engendered by
Dm
s
Table 5 one may infer that the EPR spectra of compounds (1)–(3)
present very similar spin Hamiltonian parameters. The medium g
values hgi of these compounds are about 2.050 and the parallel val-
CAHꢁ ꢁ ꢁS and CAHꢁ ꢁ ꢁ
p intermolecular interactions leading to a
three dimensional arrangement.
ues greater than the perpendicular values, i.e. g > g > g , where g
e
k
e
(
=2.0023) is the g factor of the free electron. According to the liter-
3.3. EPR spectra and magnetic properties
ature if g > g it implies in a distortion of the complex by elonga-
k
?
tion in z [54]. Therefore the obtained results for complexes (1)–(3)
are consistent with a distorted octahedral or square pyramidal
symmetry for which it is expected the ground state to be the
Fig. 3 shows the first-derivative EPR spectra (black lines) for
compounds (1)–(4) as polycrystalline samples measured at room
temperature in X-band (9.39 GHz). The spectra exhibit axial sym-
metric EPR lines with half widths ranging from 1.0 to 3.5 mT. In
Fig. 3 the colour lines correspond to calculated Cu(II) EPR spectra
dx2–y2 orbital. It is consistent with the X–ray results obtained for
complexes (2) and (3). The medium g factors of all compounds
and also the axial components are within the ranges found for

6
G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
Table 3
Selected crystallographic parameters (Å, °) for complex (3) (R = Pr).
(a) Bond angle and lengths
Molecule A
Molecule B
Cu1AAS2A
Cu1AAS2A
Cu1AAS1A
Cu1AAS1A
S1AAC1A
2.2891(7)
2.2891(7)
2.3105(6)
2.3105(6)
1.733(3)
1.726(3)
180
102.25(2)
77.75(2)
77.75(2)
102.25(2)
180
Cu1BAS2B
2.2757(6)
2.2757(6)
2.3172(6)
2.3172(6)
1.728(2)
1.726(2)
180
102.07(2)
77.93(2)
77.93(2)
102.07(2)
180
i
ii
Cu1BAS2B
Cu1BAS1B
Cu1BAS1B
S1BAC1B
S2BAC1B
i
ii
S2AAC1A
S2AACu1AAS2A
i
i
ii
ii
S2BACu1BAS2B
S2AACu1AAS1A
S2BACu1BAS1B
i
i
ii
ii
S2A ACu1AAS1A
S2AACu1AAS1A
S2B ACu1BAS1B
S2BACu1BAS1B
i
ii
S2A ACu1AAS1A
S2B ACu1BAS1B
i
ii
S1A ACu1AAS1A
S1BACu1BAS1B
(
b) Hydrogen bonding parameters)
D—Hꢁ ꢁ ꢁA
Symmetry operation
D—H
Hꢁ ꢁ ꢁA
Dꢁ ꢁ ꢁA
D—Hꢁ ꢁ ꢁA
O1AAHO1Aꢁ ꢁ ꢁO1B
O1BAHO1Bꢁ ꢁ ꢁO1A
C2AAH2A2ꢁ ꢁ ꢁS2A
C2BAH2B1ꢁ ꢁ ꢁS2B
C3BAH3B2ꢁ ꢁ ꢁS1A
C4AAH4A2ꢁ ꢁ ꢁS1A
C4BAH4B1ꢁ ꢁ ꢁS1B
C4BAH4B1ꢁ ꢁ ꢁS2B
ꢀx, 1 ꢀ y, 1 ꢀ z
0.84
0.84
0.99
0.99
0.99
0.99
0.99
0.99
1.87
1.90
2.72
2.56
2.86
2.61
2.64
2.87
2.705(3)
2.704(3)
3.053(3)
3.075(3)
3.760(3)
3.051(2)
3.043(2)
3.825(3)
175
160
100
112
152
107
104
162
ꢀx, 1 ꢀ y, 1 ꢀ z
ꢀ1 + x, y, z
(
c) .CAHꢁ ꢁ ꢁ.
p interactions
CAHꢁꢁCg
HꢁꢁCg
H
perp
c
CAHꢁ ꢁ ꢁCg
Cꢁ ꢁ ꢁCg
CAH, p
C3BAH3B2ACg1
C3BAH3B2ACg2
C6AAH6A3ACg3
C6AAH6A3ACg4
ꢀx, 1 ꢀ y, 1 ꢀ z
2.83
2.83
2.92
2.92
2.702
2.702
2.800
2.800
17.07
17.07
16.15
16.15
178
178
126
126
3.816(3)
3.816(3)
3.580(3)
3.580(3)
75
75
52
52
1 + x, ꢀ1 + y, z
ꢀx, 1 ꢀ y, ꢀz
x, y, z
i
i
i
ii
ii
ii
Cg(1)ACg(4) are the centroids of the rings defined by Cu1A, S1A, C1A, S2A; Cu1A, S1A , C1A , S2A ; Cu1B, S1B, C1B, S2B; Cu1B, S1B , C1B , S2B , respectively: Gamma is the
angle at H between the vectors Hꢁ ꢁ ꢁCg and Hperp. CAH,
p is an estimate of the significance of the interaction.
i
Symmetry operation: i = ꢀx ꢀ 1, ꢀy + 2, ꢀz + 1.
ii
Symmetry operation: ii = ꢀx, ꢀy + 1, ꢀz.
analogous compounds, which ranges between 2.044 and 2.050,
respectively [55,56]. The calculated g values for complexes (1)–
Table 5
Spin Hamiltonian parameters for complexes (1)–(4), g factors and g and g , the full
k
?
k
width at half maximum in units of mT and the medium hgi which is hgi = 1/3
(
3), which are between 2.02 and 2.10, indicate a strong covalent
(g
þ 2g ).
2+
k
?
bonding between the Cu and sulfur ions [57]. The compounds
1
2
1 2
[
Cu{S
2
CNR(CH
2
CH
2
OH)}
2
]
g
g }
?
D
H
pp (mT)
hgi
[
Cu{S
2
CNR R }
2
] (R R = simple alkyl) investigated in a previous
k
work [58,59] display very similar EPR spectra such as in (1)–(3).
In those compounds the CuACu distances varied between 3.4 and
(
1) (R = Me)
2.099(1)
2.072(1)
2.101(1)
2.025(1)
2.032(1)
2.032(1)
2.027(1)
2.0565(5)
2.0
3.5
2.2
1.0
2.054
2.045
2.075
2.036
(2) (R = Et)
(
(
3) (R = Pr)
3) (R = CH
7
.6 Å and a weak antiferromagnetic coupling was observed. The
latter was not the case for the four compounds investigated in this
work. Compound (4), [Cu{S CNR(CH CH OH)} ] (R = CH CH OH),
2
2
CH OH)
2
2
2
2
2
2
behaves somewhat differently. Although the medium g value hgi
is similar to that of the other complexes, the parallel and perpen-
dicular g values are inverted. It might be a consequence of the
geometry at the Cu(II) centre which in view of the crystallographic
determination [6], is not symmetrical as in the case of (3). This sub-
tle deviation of the square planar geometry might transform dz2
orbital in the ground state. EPR measurements at low temperatures
3.4. Pharmacological results
It was expected that the presence of CH
hydrophilic group, could increase biologic activity. However, only
complexes (3) and (4) displayed biocidal activity. The best results
2
2
CH OH, a more
ꢀ3
were found towards C. albicans. The MIC values 26.5 ꢂ 10 and
ꢀ
3
ꢀ1
(
down to 10 K) do not modify significantly the line shape and spin
36.3 ꢂ 10 mmol L respectively, are close to that of the control
ꢀ3 ꢀ1
Hamiltonian parameters of any of the investigated compounds.
drug Fluconazole 32.9 ꢂ 10 mmol L
, see Table 6. In the
Table 4
Bridging CuAS bond lengths and CuACu separations in dimeric copper dithiocarbamates.
Compound
Bridging CuAS bond length, Å
d (Cuꢁ ꢁ ꢁCu) Å
T (K)
Ref.
(
(
[
[
[
[
[
2) (R = Et)
4) (R = CH
2.793(13)
2.773(4)
2.740(1)
2.844(1)
3.230(3)
2.888(2)
2.902(1)
3.5552(8)
3.451(2)
3.418
120
293
295
295
295
295
295
This study
[5]
[40,41]
[42,43]
[44]
2
2
CH OH))
1
2
1
2
Cu{S
Cu{S
Cu{S
Cu{S
Cu{S
2
2
2
2
2
CNR R }
2
2
2
2
2
] (R , R = Pr
2
)
)
1
2
1
2
CNR R }
] (R , R = Et
2
3.572
1
2
1
2
CNR R }
] R , R = (2-pyridinyl)]
1
2
1
2
CNR R }
] (R , R = allyl
2
)
[45]
[52]
1
2
1
2
CNR R }
] (R , R = Bu
2
)
3.783(1)

G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
7
Table 6
Acknowledgements
Minimal inhibitory concentrations (MIC) for complexes (1)–(4), and starting materials
towards C. albicans, S. aureus and P. aeruginosa.
This work was supported by CNPq and FAPEMIG – Brazil. The
authors thank the EPSRC X-ray Crystallographic Service, based at
the University of Southampton, England for the data collections.
ꢀ
3
ꢀ1
[
Cu{S
2
CNR(CH
2 2
CH OH)}
2
]
MIC 10 mmol L
C. albicans
S. aureus
P. aeruginosa
(
(
(
(
1) (R = Me)
2) (R = Et)
3) (R = Pr)
–
–
–
–
17.7
>187
2.2
–
–
–
–
–
References
26.5
36.3
–
32.9
–
4) (R = CH
2
CH
2
OH)
[1] P.J. Heard, Prog. Inorg. Chem. 53 (2005) 1.
[2] G. Hogarth, Prog. Inorg. Chem. 53 (2005) 71.
[3] F.H. Allen, Acta Crystallogr. B 58 (2002) 71.
[4] R.A. Howie, G.M. de Lima, D.C. Menezes, J.L. Wardell, S.M.S. Wardell, D.J. Young,
E.R.T. Tiekink, Chem. Eng. Commun. 10 (2008) 1626.
Tetracycline
Fluconazole
>96.7
–
–
[
Cu(OAc)
2
]ꢁ2H
2
O
–
[
5] A. Radha, M. Seshadayee, K. Radha, D.G. Aravamudan, C.H. Subramanyam, Acta
Crystallogr. E C41 (1985) 1166.
[
[
6] L.-F. Hou, Y. Zhong, Y. Mei, J. Fan, Acta Crystallogr. E E65 (2009) m1694.
7] K. Ramalingam, K. Radha, G. Aravamudan, C. Mahadevan, C. Subramanyam, M.
Seshasayee, Acta Crystallogr. E C40 (1984) 1838.
presence of the Gram-negative S. Aureus both complexes showed
smaller activity than tetracycline and were inert against Gram-po-
sitive P. auruginosa bacteria.
In view of the structures one should not expect great differences
between complexes (1–3), in terms of structure–reactivity
relationship. However, complexes (1) and (2) are inactive towards
the microorganisms used in this study. On the other hand, com-
pound (3) is more effectiveagainst C. albicans than (4) or fluconazole.
[8] R. Nomura, S. Nakamoto, Trans. Mater. Res. Soc. Jpn. 26 (2001) 483.
9] E.R.T. Tiekink, S.M.S.V. Wardell, J.L. Wardell, J. Chem. Crystallogr. 37 (2007)
39.
[
4
[
[
[
10] R.E. Benson, C.A. Ellis, C.E. Lewis, E.R.T. Tiekink, Chem. Eng. Comm. 9 (2007)
930.
11] R.A. Howie, E.R.T. Tiekink, J.L. Wardell, S.M.S.V. Wardell, J. Chem. Crystallogr.
39 (2009) 293.
12] E.R.T. Tiekink, D.J. Young, Acta Crystallogr. E E62 (2006) m1887.
One should expect that the CH
cidal activity, and complex (4), [Cu{S
efficient. However our results show another reality. Complex (3),
2
CH
2
OH group, should enhance bio-
[13] E.R.T. Tiekink, D.J. Young, J. Chem. Crystallogr. 38 (2008) 419.
[
[
[
14] B. Annuar, J.O. Hill, R.J. Magee, J. Inorg. Nucl. Chem. 36 (6) (1974) 1253.
15] B.G. Sejekan, G. Aravamudan, Monatsh. Chem. 109 (1) (1978) 165.
16] S. Singhal, A.N. Garg, K. Chandra, J. Therm. Anal. Calorim. 78 (3) (2004) 941.
2
CN(CH CH OH)} ] the more
2
2
2
[
Cu{S
2
CNR(CH CH
2 2
OH)}
2
] which possesses the less polar R group
[17] D.C. Menezes, G.M. de Lima, F.A. Carvalho, M.G. Coelho, A.O. Porto, R. Augusti,
J.D. Ardisson, Appl. Organomet. Chem. 24 (2010) 650.
(
propyl) is more active than complex (4). We believe it just confirms
[
18] S.T. Breviglieri, E.T.G. Cavalheiro, G.O. Chierice, Thermochim. Acta 356 (2000)
9.
19] B. Annuar, J.O. Hill, R.J. Magee, Thermochim. Acta 8 (1974) 439.
the importance of liposolubility in the drug-cell interaction. Even
though, it is not an easy task to explain the biocidal activities of
metallic complexes such as (1)–(4) frustrating any attempt to
establish a mechanism of action. The anti-fungal activity might be
explained by new interaction of the Cu-based complexes with the
cytoplasmatic membrane. Such interactions with other rich elec-
tronic donor centres, amino-acids, proteins, nucleosides, carbohy-
drates and steroids can increase lipid-solubility assisting the
complexes to cross the cell membrane [60]. In spite of the low activ-
ity towards C. Albicans, the use of metal-based compounds might
represent an alternative therapeutic route to overcome resistance
of the most used drugs such as fluconazole, amphotericin, among
others [60].
7
[
[20] E.F. Hilder, M. Macka, P.R. Haddad, Analyst 123 (1998) 2865.
[
[
21] E.F. Hilder, M. Macka, D. Bogan, P.R. Haddad, Anal. Commun. 34 (1997) 63.
22] J.N. Barisci, G.G. Wallace, E.A. Wilke, M. Meaney, M.R. Smyth, J.G. Vos,
Electroanalysis 1 (1989) 245.
[23] E. Beinrohr, K. Blazsekova, J. Garaj, Collect. Czech. Chem. C. 50 (1985) 1661.
[
[
24] S. Hikime, H. Yoshida, M. Yamamoto, Bunseki Kagaku 10 (1961) 508.
25] A. Cachapa, A. Mederos, P. Gilli, R. Hernandez-Molina, Pedro, S. Dominguez, E.
Chinea, M.L. Rodriguez, M. Feliz, R. Llusar, F. Brito, C.M.R. de Galarreta, C.
Tarbraue, G. Gallardo, Polyhedron 25 (2006) 3366.
[
[
26] A. Mederos, A. Cachapa, R. Hernandez-Molina, M.T. Armas, P. Gili, M. Sokolov, J.
Gonzalez-Platas, F. Brito, Inorg. Chem. Commun. 6 (2003) 498.
27] D.C. Menezes, Main Group Met. Chem. 31 (2008) 21.
[28] D.C. Menezes, F.T. Vieira, G.M. de Lima, A.O. Porto, M.E. Cortés, J.D. Ardisson,
T.E. Albrecht-Schmitt, Eur. J. Med. Chem. 40 (2005) 1277.
[
29] D.C. Menezes, F.T. Vieira, G.M. de Lima, J.L. Wardell, M.E. Cortés, G.A.B. Silva
Ferreira, A. Vilas Boas, J.R. da S. Maia, Appl. Organometal. Chem. 22 (2008) 433.
4
. Conclusions
[30] D.C. Menezes, F.T. Vieira, G.M. de Lima, J.L. Wardell, M.E. Cortes, M.P. Ferreira,
M.A. Soares, A. Vilas Boas, Appl. Organometal. Chem. 22 (2008) 221.
[
31] R.W.W. Hooft, COLLECT, Data Collection Software, Nonius BV, Delft, The
Netherlands, 1998.
2 2 2 2
Complexes [Cu{S CNR(CH CH OH)} ], R = Me (1), Et (2), Pr (3)
and CH
2
CH
2
OH (4), have been prepared and characterised. The
[32] Z. Otwinowski, W. Minor, Processing of X-ray diffraction data collected in
oscillation mode, in: C.W. Carter Jr., R.M. Sweet (Eds.), Methods in
Enzymology, Macromolecular Crystallography, Part A, vol. 276, Academic
Press, New York, 1997, pp. 307–326.
[33] G.M. Sheldrick, SADABS. Version 2.10, Bruker AXS Inc., Madison, Wisconsin,
USA, 2003.
infrared results confirmed the bidentate coordination form of the
ligand. The EPR spectra of (1)–(3) were consistent with a square
pyramidal or square planar geometry with the ground state being
the orbital dx2–y2. Compound (4) has been studied previously by
X-ray [5,6]. The inversion in the g values found in our study can
only be justified if compound (4) has been crystallised as a
distorted square planar molecule, as found previously [6]. Complex
[
34] G.M. Sheldrick, SHELXS-97. Program for the Solution of Crystal Structures,
University of Göttingen, Germany, 1997.
[35] G.M. Sheldrick, SHELXL-97. Program for Crystal Structure Refinement,
University of Göttingen, Germany, 1997.
[
[
36] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
37] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.
(
2) crystallises as a dimer linked by intermolecular CuAS bonds,
and (3) exists in the solid state in the form of two similar and
independent centrosymmetric monomers. The in vitro activity of
[38] National Committee for Clinical Laboratory Standards, Reference Method for
Broth Dilution Antifungal Susceptibility Testing of Yeasts, second ed.,
Approved Standard. NCCLS Document M27-A2, NCCLS, Pennsylvania, USA,
(
1)–(4) was investigated against colonies of C. albicans, S. aureus
2002, ISBN 1-56238-469-4.
and P. aeruginosa. The biocidal results of (3) and (4) against C. albi-
cans, in terms of MIC (Minimal Inhibitory Concentration), were
quite promising. All complexes were inert in the presence of P.
aeruginosa.
[39] National Committee for Clinical Laboratory Standards, Reference Method for
Broth Dilution Antimicrobial Susceptibility Testing for Bacteria that Grow
Aerobically, sixth ed., Approved Standard. NCCLS Document M7-A6, NCCLS,
Pennsylvania, USA, 2003, ISBN 1-56238-486-4.
[
40] (a) F.W.B. Einstein, J.S. Field, Acta Crystallogr. E B30 (1974) 2928;
b) F. Jian, Z. Wang, Z. Bai, Z. You, H.-K. Fun, K. Chinnakali, I.A. Razak,
(
Polyhedron 18 (1999) 3401.
Supplementary data
[
[
[
41] A. Pignedoli, G. Peyronel, Gazz. Chim. Ital. 92 (1962) 745.
42] G. Peyronel, A. Pignedoli, L. Antolini, Acta Crystallogr. B 28 (1972) 3596.
43] M. Bonamico, G. Dessy, A. Mugnoli, A. Vaciago, L. Zambonelli, Acta Crystallogr.
E 19 (1965) 886.
44] A.M.M. Lanfredi, F. Ugozzoli, A. Camus, J. Chem. Crystallogr. 26 (1996) 141.
[45] E. Kello, V. Kettman, J. Garaj, Coll. Czech. Chem. Commun. 49 (1984) 2210.
Crystallographic data are available on request from: Cambridge
Crystallographic Data Centre on quoting the deposition numbers
CCDC 754825 and 754826 for complexes (2) and (3), respectively.
[

8
G.M. de Lima et al. / Journal of Molecular Structure 988 (2011) 1–8
[
[
46] S.C. Ngo, K.K. Banger, M.J. DelaRosa, P.J. Toscano, J.T. Welch, Polyhedron 22
2003) 1575.
47] T.A. Rodina, O.A. Bredyuk, A.V. Ivanov, A.S. Zaeva, A.V. Gerasimenko, Russ. J.
Inorg. Chem. 54 (2009) 1243.
[53] P.W.G. Newman, C.L. Raston, A.H. White, J. Chem. Soc. Dalton Trans. (1973)
(
1332.
[54] A. Aboukais, F. Delmaire, M. Rigole, R. Hubaut, G. Mairesse, Chem. Mater. 5
(1993) 1819.
[
[
[
48] H. Iwasaki, K. Kobayashi, Acta Crystallogr. B 36 (1980) 1655.
49] W.E. Hatfield, P. Singh, F. Nepveu, Inorg. Chem. 29 (1990) 4214.
50] J.M. Martin, P.W.G. Newman, B.W. Robinson, A.H. White, J. Chem. Soc. Dalton
Trans. (1972) 2233.
51] V. Kettman, J. Garaj, S. Kudela, Coll. Czech. Chem. Commun. 42 (1977) 402.
52] J.-P. Lang, J.-M. Lei, G.-Q. Bian, J.-H. Cai, B.-S. Kang, X.-Q. Xin, Chem. Abstr. 123
[55] P.D.W. Boyd, S. Mitra, C.L. Raston, G.L. Rowbottom, A.H. White, J. Chem. Soc.
Dalton Trans. (1981) 13.
[56] S.N. Choi, J.R. Wasson, Inorg. Chem. 14 (1975) 1964.
[57] D. Kivelson, R. Neiman, J. Chem. Phys. 35 (1961) 149.
[58] M. Tumer, H. Koksal, M.K. Sener, S. Serin, Trans. Metal Chem. 24 (1999) 414.
[59] A.V. Ivanov, O.V. Loseva, A.V. Gerasimenko, Russ. J. Coord. Chem. 34 (2008) 413.
[60] D.P. Kontoyiannis, R.E. Lewis, Lancet 359 (2002) 1135.
[
[
(
1995) 131141.