Metal complexes as anticancer agents 2. Iron(III) and copper(II) bio-active complexes with N6-benzylaminopurine derivatives

Article abstract of DOI:10.1016/S0020-1693(01)00611-9

Iron(III) complexes with 2-(3-hydroxypropylamino)-6-benzylamino-9-isopropylpurine (Bohemin, HL₁), in its protonized form, of the composition (H⁺HL₁)₂[FeCl₅]·2H ₂O (1), (H⁺HL₁)₂[FeCl₅]·3H ₂O (2) have been prepared by two different routes. A new way for synthesis of copper(II) complex with 6-(2-chlorobenzylamino)purine (HL₂), [Cu₂(μ-Cl)₂(μ-HL₂)₂(HL ₂)₂Cl₂]·2H₂O (3), together with the preparation of copper(II) complex with 6-(3-chlorobenzylamino)purine (HL₃), [Cu₂(μ-Cl)₂(μ-HL₃)₂Cl ₂] (4), is also reported. The characterization have been based on elemental analysis, electronic, infrared, ES+ mass and ⁵⁷Fe Mo?ssbauer spectra, conductivity data and magnetic susceptibility temperature measurements over the 4.5-300 K for 1-3, and 35-300 K for 4, temperature range, respectively. Molecular structure of an electroneutral form of the HL₂ ligand, (HL₂·2H₂O), and a protonized form of the HL₃ ligand, (H⁺HL₃-Cl), have been determined by a single-crystal X-ray analysis. A mononuclear trigonal-bipyramidal (for 1 and 2), binuclear octahedral (for 3) and binuclear trigonal-bipyramidal (for 4) structures of the complexes were proposed mainly on the basis of spectral and magnetic properties. An S=3/2-5/2 spin-admixed state in 1 and 2 was found to be related to the presence of [FeCl₅]^2- (S = 3/2) and [FeCl₅(H₂O)]^2- (S = 5/2) complex anions in 1 and 2, as found by ⁵⁷Fe Mo?ssbauer spectroscopy. Cytotoxic activity of the complexes was determined by a calcein AM assay and IC₅₀ values were also estimated. For testing, human malignant melanoma cell line G-361, human osteogenic sarcoma cell line HOS, human chronic myelogenous leukaemia K-562 and human breast adenocarcinoma cell line MCF7 were used. The inhibition of p34^cdc2 kinase by the complexes 1 and 2, which is known to be one of the important mechanisms responsible for cytotoxicity of cytokinin-derived compounds, was also studied.

Full text of DOI:10.1016/S0020-1693(01)00611-9

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 323 (2001) 119–129
Metal complexes as anticancer agents 2. Iron(III) and copper(II)
bio-active complexes with N6-benzylaminopurine derivatives
a
Michal Malonˇ , Zdeneˇk Tra´vn´ıcˇek a,e,*, Miroslav Marysˇko b, Radek Zborˇil c,
c
Miroslav Masˇla´nˇ , Jarom´ır Marek d, Karel Dolezˇal e, Jakub Rolcˇ´ık e,
Vladim´ır Krysˇtof e, Miroslav Strnad e
a Department of Inorganic and Physical Chemistry, Palacky´ Uni6ersity, Krˇ´ızˇko6ske´ho 10, CZ-771 47 Olomouc, Czech Republic
b Institute of Physics ASCR, Cukro6arnicka´ 10, CZ-162 53 Prague 6, Czech Republic
c Department of Experimental Physics, Palacky´ Uni6ersity, Trˇ. S6obody 26, CZ-771 46 Olomouc, Czech Republic
d X-ray Laboratory of Faculty of Science, Masaryk Uni6ersity, Kotla´rˇska´ 2, CZ-611 37 Brno, Czech Republic
e
&
Laboratory of Growth Regulators, Institute of Experimental Botany ASCR and Palacky´ Uni6ersity, Slechtitelu˚ 11,
CZ-783 71 Olomouc, Czech Republic
Received 5 March 2001; accepted 6 August 2001
Abstract
Iron(III) complexes with 2-(3-hydroxypropylamino)-6-benzylamino-9-isopropylpurine (Bohemin, HL1), in its protonized form,
of the composition (H+HL1)2[FeCl5]·2H2O (1), (H+HL1)2[FeCl5]·3H2O (2) have been prepared by two different routes. A new
way for synthesis of copper(II) complex with 6-(2-chlorobenzylamino)purine (HL2), [Cu2(v-Cl)2(v-HL2)2(HL2)2Cl2]·2H2O (3),
together with the preparation of copper(II) complex with 6-(3-chlorobenzylamino)purine (HL3), [Cu2(v-Cl)2(v-HL3)2Cl2] (4), is
also reported. The characterization have been based on elemental analysis, electronic, infrared, ES+ mass and 57Fe Mo¨ssbauer
spectra, conductivity data and magnetic susceptibility temperature measurements over the 4.5–300 K for 1–3, and 35–300 K for
4, temperature range, respectively. Molecular structure of an electroneutral form of the HL2 ligand, (HL2·2H2O), and a protonized
form of the HL3 ligand, (H+HL3ꢀCl), have been determined by a single-crystal X-ray analysis. A mononuclear trigonal-bipyrami-
dal (for 1 and 2), binuclear octahedral (for 3) and binuclear trigonal-bipyramidal (for 4) structures of the complexes were
proposed mainly on the basis of spectral and magnetic properties. An S=3/2–5/2 spin-admixed state in 1 and 2 was found to
be related to the presence of [FeCl5]2− (S=3/2) and [FeCl5(H2O)]2− (S=5/2) complex anions in 1 and 2, as found by 57Fe
Mo¨ssbauer spectroscopy. Cytotoxic activity of the complexes was determined by a calcein AM assay and IC50 values were also
estimated. For testing, human malignant melanoma cell line G-361, human osteogenic sarcoma cell line HOS, human chronic
myelogenous leukaemia K-562 and human breast adenocarcinoma cell line MCF7 were used. The inhibition of p34cdc2 kinase by
the complexes 1 and 2, which is known to be one of the important mechanisms responsible for cytotoxicity of cytokinin-derived
compounds, was also studied. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Iron(III) and copper(II) complexes; 6-Benzylaminopurine derivatives; Cytotoxic activity; CDK inhibitor; Magnetic properties; Crystal
structures
1. Introduction
(HL2) and 6-(3-chlorobenzylamino)purine (HL3) [1].
We have showed that the composition of the prepared
complexes chiefly depends on the molar reactants ratio,
solvent and pH-medium used. When we have allowed
to react CuCl2·2H2O with HL2 in a molar ratio of 1:2
or 2:1 in ethanol we have always obtained the octahe-
dral binuclear copper(II) complex of the composition
[Cu2(v-Cl)2(v-HL2)2(HL2)2Cl2]·2H2O. Likewise, the re-
action of CuCl2·2H2O with the organic ligands HL2 and
In our previous work we have described the prepara-
tion, properties and biological activity of several cop-
per(II) complexes with 6-(2-chlorobenzylamino)purine
* Corresponding author. Tel.: +420-68-563 4361; fax: +420-68-
522 5737.
E-mail address: trav@risc.upol.cz (Z. Tra´vn´ıcˇek).
0020-1693/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0020-1693(01)00611-9

120
M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
2. Experimental
HL3, in molar ratios 1:1, led in ethanol to formation of
probably tetrahedral [Cu2(v-Cl)2(HL2)2Cl2], and
[Cu2(v-Cl)2(HL3)2Cl2], respectively. Whereas, the com-
plexes [Cu2(v-Cl)2(v-HL2)2(H2O)2] and [Cu2(v-Cl)2(v-
HL3)2(H2O)2] were prepared by the reaction of
CuCl2·2H2O with the appropriate organic ligand in
ethanol in the presence of Et3N, in a molar ratio 1:1:1.
The above-mentioned compounds belong to a category
of copper(II) complexes containing either Cu(v-Cl)2(v-
L)2Cu (L, bidentate nitrogen-donor ligand, e.g. 1,8-
naphthyridine [2]) or Cu(v-Cl)2Cu core [3–5].
These compounds are not only studied from a basis
research point of view for their interesting magnetic
properties but also because of their possible both tech-
nological and biological use [6,7]. Some of the discussed
complexes, actually dichloro bridged binuclear cop-
per(II) containing the Cu(v-Cl)2Cl core, were also
structurally classified [7].
This paper represents results in our continuing work
of transition metal complexes with 6-benzylaminop-
urine derivatives. Ligands representing different groups
of 6-benzylaminopurine derivatives having different
roles in cell cycle regulation were chosen for this study.
We report here the synthesis and characterization of
iron(III) compounds with 2-(3-hydroxypropylamino)-6-
benzylamino-9-isopropylpurine (Bohemin, HL1) and
copper(II) complexes with 6-(2-chlorobenzylamino)-
purine (HL2) or 6-(3-chlorobenzylamino)purine (HL3)
(see Scheme 1). Apart from other thinks, one of the
findings of our research was discovery of a new manner
in preparation of the compound [Cu2(v-Cl)2(v-HL2)2-
(HL2)2Cl2] (3). The other aim of the present study was
to test biological activity, especially cytotoxicity of the
prepared complexes and ligands used, and their ability
to inhibit p34cdc2 kinase. Special respect was given to
study possible influence of complex formation on bio-
logical activity of studied compounds.
2.1. Starting materials
CuCl2·2H2O, FeCl3·6H2O, 6-chloropurine, 2-chloro-
benzylamine, 3-chlorobenzylamine and solvents used
were purchased from commercial sources (Fluka co.,
Aldrich Co.) and used as received. Bohemin (HL1) was
supplied from Olchemim Co. (Olomouc, Czech Repub-
lic).
2.2. Preparation of HL2 and HL3
The compounds were prepared by slightly modified
procedure as described in the literature [8] and were
characterized by elemental analysis (C, H, N) and ES+
mass spectra. Anal. Calc for HL2 C12H10N5Cl1 (259.7):
C, 55.5; H, 3.9; N, 27.0. Found: C, 55.6; H, 3.7; N,
27.1%. ES+MS: m/z=260 (based on 35Cl) [M]+, 224
[MꢀCl]+, 125 [ClꢀC6H4ꢀCH2]+, 91 [C6H4ꢀCH2]+, 78
[C6H5]+. Anal. Calc. for HL3 C12H10N5Cl1 (259.7): C,
55.5; H, 3.9; N, 27.0. Found: C, 55.3; H, 3.9; N, 26.8%.
ES+MS: m/z=260 (based on 35Cl) [M]+, 125
[ClꢀC6H4ꢀCH2]+, 90 [C6H4ꢀCH2]+, 78 [C6H5]+.
2.3. Preparation of the iron(III) and copper(II)
complexes
2.3.1. (H+HL1)2[FeCl5]·2H2O (1)
A total of 1 mmol (340 mg) of HL1 was added to a
solution of FeCl3·6H2O (0.5 mmol, 135 mg) in 2 M HCl
(20 cm3). The reaction mixture was kept in ultrasonic
for 5 min. Yellow solution was refluxed for 10 h and
left for 5 weeks during which yellow microcrystals
formed. The solid was filtered off, washed with ethanol
and diethyl ether, and dried under an infra lamp at
40 °C. m.p. 113–116 °C. Yields ranged from 30 to
40%. Anal. Calc. for C36H50N12O2Cl5Fe1·(H2O)2: C,
Scheme 1.

M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
121
45.4; H, 5.7; N, 17.7; Cl, 18.6. Found: C, 45.3; H, 5.6;
N, 17.5; Cl, 18.4%. FAB+MS (m/z): 341, 250, 163,
93, 75, 45. IR (cm−1): 3200, 3060, 2740, 1840, 1684,
1602, 1532, 1508, 1446, 1408, 1376, 1278, 1242, 1200,
1172, 1060, 1040, 964, 924, 846, 756, 722, 702, 652, 604,
486.
(cm−1): 3292, 3224, 3140, 2800–2600, 1632, 1600, 1578,
1536, 1508, 1426, 1404, 1346, 1312, 1270, 1230, 1206,
1186, 1144, 1114, 1076, 1030, 976, 928, 892, 868, 856,
796, 784, 698, 684, 676, 600, 572, 552, 444.
2.4. Elemental analyses
2.3.2. (H+HL1)2[FeCl5]·3H2O (2)
Elemental analyses (C, H, N) were performed on an
EA1108 CHN analyser (Fisons Instruments). The chlo-
rine content was determined by the Scho¨niger method.
A total of 1 mmol (340 mg) of HL1 was added to a
solution of FeCl3·6H2O (1 mmol, 270 mg) in 2 M HCl
(20 cm3). The resulting light-yellow solution was
refluxed for 4 h and left for 2 weeks during which
yellow acicular microcrystals formed. Crystals were sep-
arated by filtration, washed with 2 M HCl, water,
ethanol and diethyl ether, and dried under an infra
lamp at 40 °C. m.p. 114–118 °C. Yields ranged from
50 to 60%. Anal. Calc. for C36H50N12O2Cl5Fe1·(H2O)3:
C, 44.6; H, 5.8; N, 17.3; Cl, 18.3. Found: C, 44.2; H,
5.9; N, 17.1; Cl, 18.5%. ES+MS (m/z): 681, 419, 341,
247, 157, 78. IR (cm−1): 3220, 3050, 1686, 1632, 1604,
1542, 1510, 1444, 1392, 1376, 1276, 1230, 1210, 1170,
1060, 1040, 950, 922, 852, 756, 702, 654, 604, 486.
2.5. Spectroscopic studies
Electronic absorption spectra (diffuse-reflectance)
were recorded on a SPECORD M40 (Carl Zeiss Jena)
within the 40 000–1100 cm−1 range. IR spectra were
recorded on a SPECORD IR 80 instrument (Carl Zeiss
Jena) in Nujol mulls (4000–400 cm−1).
ES+ mass spectra were recorded for 1, 3 and 4 using
direct probe on a Waters ZMD 2000 mass spectrome-
ter. The mass-monitoring interval was 10–1500 amu.
The spectra were collected using 3.0 second cyclical
scans and applying sample cone voltage 20, 30 or 40 V
at source block temperature 80 °C, desolvation temper-
2.3.3. [Cu2(v-Cl)2(v-HL2)2(HL2)2Cl2]·2H2O (3)
To a solution of CuCl2·2H2O (1 mmol, 170 mg) in 2
M HCl (20 cm3) was added the HL2 ligand (1 mmol,
260 mg). The reaction mixture was kept in ultrasonic
for 5 min and subsequently refluxed for 6 h. A colour
of the reaction mixture changed gradually from light
green to green–yellow. The precipitate which separated
was filtered and washed with 2 M HCl, water, ethanol
and diethyl ether, and dried under an infra lamp at
40 °C. m.p. 258 °C. Yields ranged from 60 to 80%.
Anal. Calc. for C48H40N20Cl8Cu2·(H2O)2: C, 42.9; H,
3.3; N, 20.8; Cl, 21.2. Found: C, 42.7; H, 3.1; N, 20.1;
Cl, 21.1%. ES+MS (m/z): 1305, 1051, 972, 750, 654,
617, 582, 513, 260, 78. IR (cm−1): 3260, 3190, 3125,
3040, 2660, 1644, 1608, 1582, 1536, 1504, 1432, 1410,
1342, 1320, 1280, 1230, 1210, 1190, 1150, 1126, 1114,
1056, 1030, 970, 932, 868, 856, 756, 722, 684, 660, 610,
588, 572, 528.
ature 150 °C and desolvation gas flow rate 200 l h−1
.
Mass spectrometer was directly coupled to a MassLynx
data system. FAB+ mass spectra for 2 were obtained
using direct probe on double-focusing magnetic sector
mass spectrometer (JMS-SX 102, JEOL). The ion
source temperature was 55 °C. Desorption was gener-
ated with a beam of 3 kV xenon atoms at an emission
current of 10 mA. Full scan analyses were performed
for a mass range 10–1400 m/z. Data were processed by
a JEOL MD-7010 data system. All interpretations of
m/z are based on 35Cl and 63Cu.
The transmission Mo¨ssbauer spectra were collected
at 300 and 30 K using a Mo¨ssbauer spectrometer in
constant acceleration mode with a 57Co(Rh) source. A
cryostat with closed He-cycle (Janis Research Com-
pany, USA) was used as a refrigerating system that
allowed to work in a temperature range of 12–300 K.
2.3.4. [Cu2(v-Cl)2(v-HL3)2Cl2] (4)
2.6. Magnetic data
To a solution of CuCl2·2H2O (1 mmol, 170 mg) in 2
M HCl (20 cm3) was added the organic ligand HL3 (1
mmol, 260 mg). The reaction mixture was kept in
ultrasonic for 5 min and light green solution was
refluxed. The light green precipitate developed after 2 h
and reaction mixture was refluxed for next 4 h. No
change was observed and the solid was isolated by
filtration, washed with 2 M HCl, water, ethanol and
diethyl ether, and dried under an infra lamp at 40 ° C.
m.p. 262 °C. Yields ranged from 70 to 80%. Anal. Calc.
for C24H20N10Cl6Cu2: C, 36.6; H, 2.6; N, 17.8; Cl, 27.0.
Found: C, 36.8; H, 2.6; N, 17.3; Cl, 27.1%. ES+MS
(m/z): 1051, 972, 750, 654, 617, 582, 513, 260, 78. IR
The susceptibility studies were performed on pow-
dered samples with the average weight 60 mg using a
Quantum Design MPM-5S SQUID magnetometer. All
measurements were conducted in a field of 0.9 T.
Measurements were made in the range 35.0–300 K for
3 and 4.5–300 K for all remaining compounds. The
molar susceptibilities were corrected for diamagnetism
using Pascal’s constants [9]. Conductivity measurements
were made with an OK 102 l−1 conductometer
(Radelkis, Budapest) in DMFA at 25 °C. The concen-
tration of the complexes in solutions was 10−3 mol
dm−3 [10].

122
M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
respectively, where P=(F2o+2F2c)/3. All hydrogens in
both structures were located in difference Fourier maps
and all their parameters were refined. The largest peak
Table 1
Crystal data and structure refinement for HL2·2H2O and H+HL3ꢀCl
HL2·2H2O
H+HL3ꢀCl
,
and hole on the final difference map were 0.36 [0.89 A
−3
,
,
Empirical formula
Formula weight
Temperature (K)
Wavelength (A)
Crystal system
Space group
C12H14ClN5O2
295.73
C12H11Cl2N5
296.16
from O(1)] and −0.30 [0.91 A from C(10)] e A
for
,
HL2·2H2O, and 0.36 [1.02 A from Cl(1)] and −0.59
120(2)
293(2)
−3
[0.82 A from Cl(1)] e A
for H+HL3ꢀCl, respectively.
,
,
,
0.71073
triclinic
0.71073
monoclinic
P21/c
Full crystallographic data may be found in Table 1.
Selected bond distances and angles are given in Table 2.
(
P1
Unit cell dimensions
,
a (A)
6.9289(7)
8.8180(11)
11.5462(12)
69.751(10)
84.838(8)
80.017(9)
651.50(12)
2
18.791(2)
8.8630(8)
8.1371(9)
90
96.513(9)
90
1346.4(2)
4
1.461
2.8. Biological acti6ity testing
,
b (A)
,
c (A)
Human malignant melanoma cell line G-361, human
osteogenic sarcoma cell line HOS, human chronic myel-
ogenous leukaemia cell line K-562 and human breast
adenocarcinoma cell line MCF7 were used for a cyto-
toxicity determination of the prepared complexes by
calcein AM assay, as it was described previously in
more details [1,13]. Briefly, the tumour cells were main-
tained in plastic tissue culture flasks and grown on
Dulbecco’s modified Eagle’s cell culture medium
(DMEM) at 37 °C in 5% CO2 atmosphere and 100%
humidity. The cell suspension of approximate density
1.25×105 cells ml−1 was redistributed into 96-well
microtitre plates (Nunc, Denmark). After 12 h of prein-
cubation, the tested compounds (in the 0.5–200 mM
range) were added. Incubation lasted for 72 h. At the
end of this period, the cells were incubated for 1 h with
calcein AM and fluorescence of the live cells was mea-
sured at 485/538 nm (ex/em) with a Fluoroskan Ascent
(Labsystems, Finland). IC50 values, the drug concentra-
tions lethal to 50% of the tumour cells, were estimated.
Kinase p34cdc2/cyclin B was produced in Sf9 insect
cells co-infected with baculoviral constructs [12]. Kinase
activity was assayed with 1 mg ml−1 histone H1 in the
presence of 15 mM [k-32P] ATP (500–1000 cpm
pmol−1) in final volume of 10 ml. After 15 min the
reactions were stopped by addition of 3×SDS sample
buffer and separated electrophoretically on SDS poly-
acrylamide gel [1,12]. Final measurement of phosphory-
lated proteins was done using a digital imaging system.
The stability of tested complexes in the media used
was monitored by measurement of UV–Vis spectra. No
significant changes in the maxima of the absorption
peaks were observed after 24 h.
h (°)
i (°)
k (°)
,
V (A)
Z
Dcalc (g cm−3
)
1.508
0.303
308
v (mm−1
F(000)
)
0.475
608
Crystal size (mm)
q Range (°)
Index ranges
0.50×0.30×0.20 0.50×0.40×0.15
3.59–25.00
−85h57
−105k510
−135l513
3558
3.41–24.99
−225h522
−105k510
−95l56
7917
Reflections collected
Independent reflections
2220
2354
[Rint=0.0342]
0.9418 and
0.8632
[Rint=0.0323]
0.9322 and
0.7971
Max./min. transmission
Refinement method
full-matrix least-squares on F2
Data/restraints/parameters 2220/0/229
2354/0/216
0.992
Goodness-of-fit on F2
0.983
Final R indices [I\2|(I)]
R1
wR2
0.0324
0.0854
0.0451
0.1293
R indices (all data)
R1
wR2
0.0363
0.0879
0.004(3)
0.0521
0.1347
none
Extinction coefficient
2.7. X-ray data collection and structure refinement for
6-(2-chlorobenzylamino)purine dihydrate sol6ate,
HL2·2H2O, and 6-(3-chlorobenzylamino)purinium
chloride, H+HL3ꢀCl
X-ray data were collected at 120 K (for HL2·2H2O)
and 293 K (for H+HL3ꢀCl) on a four-circle s-axis
KUMA KM-4 diffractometer equipped with an Oxford
Cryostream cooler using graphite-monochromated Mo
Ka radiation. Data collections for the organic ligands
were performed using a CCD detector (KUMA Dif-
fraction, Wroclaw). KUMA KM4RED software was
used for data reduction. The structures were solved
using the direct methods [SHELXS-97] [11] and refined
on F2 using full-matrix least-squares procedure
3. Results and discussion
The reaction between FeCl3·6H2O and HL1 in 2M
HCl in a molar ratio 1:1 or in a molar ratio 1:2 yielded
mononuclear complexes (H+HL1)2[FeCl5]·xH2O (x=2
for
1 and x=3 for 2), while the reaction of
[
SHELXL-97] [12] with weight: w=1/[|2(Fo2)+
(0.0481P)2+0.4289P] for HL2·2H2O, and w=1/
[|2(F2o)+(0.0890P)2+0.5986P] H+HL3ꢀCl,
for
CuCl2·2H2O with HL2 or HL3 (in a molar ratio 1:1) in
2M HCl led to formation of binuclear complexes
[Cu2(v-Cl)2(v-HL2)2(HL2)2(Cl)2]·2H2O (3) and [Cu2(v-

M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
123
cm−1 for 1 and 32 100 cm−1 for 2 are due to CT
transitions. In the case of 3 and 4, the maxima observed
at 12 600 and 12 400 cm−1, respectively, may be con-
nected with the d–d transitions, while the shoulders
near 28 000 cm−1 are attributable to the y “
copper(II) LMCT transitions. The maxima ranging
from 28 400 to 35 000 cm−1 are probably due to inter-
nal ligand or ligand-to-metal charge-transfer transitions
[14,15].
In IR spectra of all complexes, the bands observed
between 1630 and 1644 cm−1 may be assigned to the
w(CꢁN) stretch vibration of the heterocyclic ring. The
bands at approximately 1604, 1510 and 1420–1440
cm−1 are assignable to the w(CꢁC) vibration, while the
maxima appearing in the region 910–660 cm−1 are due
to skeletal vibrations of aromatic rings and thus
confirm the presence of the HL1, HL2 or HL3 ligands in
the compounds 1–4. The presence of chlorine attached
directly to the aromatic ring is apparent from the band
at 1114–1150 cm−1 found in the IR spectra of com-
pounds 3 and 4. The bands appearing in spectra of all
complexes near 3050 and 3240 cm−1 can be connected
with w(CꢀH) and w(NꢀH) vibrations, respectively
[16,17].
Table 2
+
,
Selected bond lengths (A) and angles (°) for HL2·2H2O and H
HL3ꢀCl
HL2·2H2O
H+HL3ꢀCl
Bond lengths
Cl(1)ꢀC(12)
N(1)ꢀC(6)
N(1)ꢀC(2)
N(3)ꢀC(2)
N(3)ꢀC(4)
N(6)ꢀC(6)
N(6)ꢀC(10)
N(7)ꢀC(8)
N(7)ꢀC(5)
N(9)ꢀC(8)
N(9)ꢀC(4)
C(6)ꢀC(5)
1.7441(17) Cl(1)ꢀC(13)
1.745(3)
1.361(3)
1.306(3)
1.328(3)
1.356(3)
1.323(3)
1.452(3)
1.339(3)
1.362(3)
1.322(3)
1.356(3)
1.410(3)
1.373(3)
1.508(4)
1.371(4)
1.383(4)
1.370(4)
1.358(5)
1.354(6)
1.395(6)
1.350(2)
1.347(2)
1.333(2)
1.351(2)
1.343(2)
1.453(2)
1.318(2)
1.381(2)
1.365(2)
1.374(2)
1.415(2)
1.380(2)
1.515(2)
1.388(2)
1.397(2)
1.385(3)
1.387(3)
1.386(3)
1.391(3)
N(1)ꢀC(6)
N(1)ꢀC(2)
N(3)ꢀC(2)
N(3)ꢀC(4)
N(6)ꢀC(6)
N(6)ꢀC(10)
N(7)ꢀC(8)
N(7)ꢀC(5)
N(9)ꢀC(8)
N(9)ꢀC(4)
C(6)ꢀC(5)
C(4)ꢀC(5)
C(4)ꢀC(5)
C(10)ꢀC(11)
C(11)ꢀC(16)
C(11)ꢀC(12)
C(12)ꢀC(13)
C(13)ꢀC(14)
C(14)ꢀC(15)
C(15)ꢀC(16)
C(10)ꢀC(11)
C(11)ꢀC(16)
C(11)ꢀC(12)
C(12)ꢀC(13)
C(13)ꢀC(14)
C(14)ꢀC(15)
C(15)ꢀC(16)
Bond angles
C(6)ꢀN(1)ꢀC(2)
C(2)ꢀN(3)ꢀC(4)
C(6)ꢀN(6)ꢀC(10)
C(8)ꢀN(7)ꢀC(5)
C(8)ꢀN(9)ꢀC(4)
N(6)ꢀC(6)ꢀN(1)
N(6)ꢀC(6)ꢀC(5)
N(1)ꢀC(6)ꢀC(5)
N(3)ꢀC(2)ꢀN(1)
N(3)ꢀC(4)ꢀN(9)
N(3)ꢀC(4)ꢀC(5)
N(9)ꢀC(4)ꢀC(5)
C(4)ꢀC(5)ꢀN(7)
C(4)ꢀC(5)ꢀC(6)
N(7)ꢀC(5)ꢀC(6)
N(7)ꢀC(8)ꢀN(9)
N(6)ꢀC(10)ꢀC(11)
118.81(14) C(6)ꢀN(1)ꢀC(2)
111.32(14) C(2)ꢀN(3)ꢀC(4)
123.97(15) C(6)ꢀN(6)ꢀC(10)
103.68(15) C(8)ꢀN(7)ꢀC(5)
106.19(14) C(8)ꢀN(9)ꢀC(4)
120.33(15) N(6)ꢀC(6)ꢀN(1)
121.74(16) N(6)ꢀC(6)ꢀC(5)
117.93(15) N(1)ꢀC(6)ꢀC(5)
128.48(17) N(3)ꢀC(2)ꢀN(1)
127.90(15) N(3)ꢀC(4)ꢀN(9)
126.65(15) N(3)ꢀC(4)ꢀC(5)
105.45(15) N(9)ꢀC(4)ꢀC(5)
111.14(14) C(4)ꢀC(5)ꢀN(7)
116.81(15) C(4)ꢀC(5)ꢀC(6)
132.06(16) N(7)ꢀC(5)ꢀC(6)
113.54(16) N(7)ꢀC(8)ꢀN(9)
114.74(14) N(6)ꢀC(10)ꢀC(11)
119.10(19)
117.6(2)
123.0(2)
106.8(2)
102.60(18)
118.7(2)
123.7(2)
117.55(19)
126.0(2)
127.91(19)
119.7(2)
112.40(19)
104.41(19)
120.06(19)
135.53(19)
113.8(2)
The molar conductance values for the complexes 1
and 2 as formulated in Table 1 show on considerable
dissociation of the compounds in DMFA solution indi-
cating thus that they behave as electrolytes of the 2:1
type [10]. The values of uM found for 3 and 4 clearly
indicate a non-ionic feature of the complexes in the
solvent used.
3.1. Magnetic and 57Fe Mo¨ssbauer data
The magnetic properties of complexes 1 and 2 have
been studied in the temperature range from 4.5 to 300
K. The temperature variations of magnetic susceptibil-
ity and magnetic moment for 1 are shown in Fig. 1. The
112.28(19)
C(16)ꢀC(11)ꢀC(12) 116.96(16) C(16)ꢀC(11)ꢀC(12) 118.6(3)
C(16)ꢀC(11)ꢀC(10) 123.37(15) C(16)ꢀC(11)ꢀC(10) 120.9(3)
C(12)ꢀC(11)ꢀC(10) 119.67(15) C(12)ꢀC(11)ꢀC(10) 120.5(2)
C(13)ꢀC(12)ꢀC(11) 122.56(16) C(13)ꢀC(12)ꢀC(11) 120.2(3)
C(13)ꢀC(12)ꢀCl(1) 118.29(13) C(12)ꢀC(13)ꢀCl(1) 118.7(2)
C(11)ꢀC(12)ꢀCl(1) 119.15(13) C(14)ꢀC(13)ꢀCl(1) 119.6(3)
C(14)ꢀC(13)ꢀC(12) 119.08(16) C(14)ꢀC(13)ꢀC(12) 121.7(3)
C(13)ꢀC(14)ꢀC(15) 119.80(17) C(13)ꢀC(14)ꢀC(15) 118.5(3)
C(14)ꢀC(15)ꢀC(16) 120.09(17) C(14)ꢀC(15)ꢀC(16) 121.4(3)
C(11)ꢀC(16)ꢀC(15) 121.50(16) C(11)ꢀC(16)ꢀC(15) 119.6(4)
effective magnetic moment was calculated as veff
=
2.828(M T)1/2 vB. It can be seen from the figure, that
the effective magnetic moment is nearly constant in the
temperature interval 15–300 K with the values of 4.21–
4.23 veff/vB for 1, and 4.51–4.48 veff/vB for 2, respec-
tively. Below 15 K, a slight decrease can be related to
the presence of a weak antiferromagnetic interaction
for both cases. The values of C and q were determined
using Curie–Weiss law, M=C/(T−q), and are equal
to 2.343(3) cm3 K mol−1 and −0.87(1) K for 1, and
2.691(5) cm3 K mol−1 and 0.90(1) K for 2, respectively.
At room temperature M T value is equal to 2.24 cm3 K
mol−1 (for 1) and 2.51 cm3 K mol−1 (for 2), which is
significantly higher than anticipated for a pure S=3/2
state (:1.876 cm3 K mol−1), but considerable lower
than anticipated for a pure S=5/2 state (:4.377 cm3
K mol−1). On cooling, M T value is nearly constant
down to 15 K, and then smoothly decreases. At 4.5 K,
Cl)2(v-HL3)2(Cl)2]
(4)
{HL2=6-(2-chlorobenzy-
lamino)purine, HL3=6-(3-chlorobenzylamino)purine}.
Their electronic spectral, magnetic and conductivity
data are listed in Table 3. The diffuse-reflectance elec-
tronic spectra of complexes 1 and 2 exhibit band at
26 800 and 27 000 cm−1, respectively, which can be
assigned to d–d transitions. The maxima at 32 000

124
M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
Table 3
Colours, magnetic, electronic spectral and molar conductivity data for the complexes
c
Compound Colour
veff/vB
J
g
Absorption maxima b ×103
(cm−1
uM
(S cm2 mol−1
)
(cm−1
)
)
1
2
3
Yellow
Yellow
Green–yellow
4.23 (300 K)–3.95 (4.5K)
4.48 (300 K)–4.22 (4.5 K)
1.61 (295 K)–0.49 (35
K) a
26.8, 32.0
27.0, 32.1
12.6, 23.0, 29.0, 34.0
122
126
24
−61.6(6)
2.00
4
Light–green
1.93 (300 K)–0.13 (4.5
−40.89(7)
1.982(4)
12.4, 22.8, 28.4, 35.0
14
K) a
a Value per one Cu(II) ion.
b Diffuse-reflectance spectra measured by Nujol technique.
c Measured in DMFA.
Fig. 1. Thermal dependence of  (z) and veff/vB (5) for 1. The solid line represents the best-fit to the Curie–Weiss law. The insert shows the
M
thermal dependence of 1/
.
M
Fe2+ doublet. The parameters [l=0.30, Z=0.45 mm
−1 for 1, l=0.31, Z=0.45 mm s−1 for 2] of the Fe3+

M T is equal to 1.95 cm3 K mol−1 (for 1), and 2.23
cm3 K mol−1 (for 2), respectively. From the above
mentioned magnetochemical data for 1 and 2 we may
conclude that both complexes belong to the group of
spin-admixed iron(III) compounds with S=3/2–5/2
[18]. Thus, we suppose that the coordination geometry
in [FeCl5]2− is dominantly trigonal-bipyramidal (within
the S=3/2 state) encountered with six-coordinate octa-
hedral [FeCl5(H2O)]2− with an S=5/2 ground state.
This conclusion may be supported by 57Fe Mo¨ssbauer
spectroscopy. Taking into account the low content of
iron in compounds 1 and 2, the spectra were recorded
not only at room temperature but also at 30 K to
improve the recoil-less fraction. The spectra measured
at 30 K are depicted in Fig. 2, while the hyperfine
parameters are given in Table 4. Both spectra consist of
three subspectra, i.e. Fe3+ doublet, Fe3+ singlet and
s
doublet clearly characterize the presence of the
[FeCl5]2− complex anion with D3h symmetry in both
compounds. The data are also consistent with the re-
sults obtained on similar systems containing trigonal-
bipyramidal [FeCl5]2− anion, such as I[Fe(III)Cl5]
(l=0.21, Z=0.42
azinium) [19], II[Fe(III)Cl5] (l=0.312, Z=0.50 mm
s−1
II=tetramethylimidazolium) [19] and II-
mm
s
−1; I=tetramethylpiper-
;
I[Fe(III)Cl5] (l=0.27, Z=0.34 mm s−1; III=diquat-
ernized cation obtained from diazabicyclo[2.2.2]-octane,
DABCO) [20]. The Fe3+ singlet with l=0.21 mm s−1
may be probably connected with the presence of octa-
hedrally coordinated iron(III) in [FeCl5(H2O)]2−. A
molar ratio of [FeCl5]2−: [FeCl5(H2O)]2− calculated
from areas of component subspectra was found to be

M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
125
2Ni2g2
3kT [1+1/3 exp(−2J/kT)]
1−z
Ni2g2z
2kT
51.5: 29.7% in 1, and 50.0: 39.1% in 2, respectively. The
M=
+
+2Nh.
percentages of subspectra areas indicate that [FeCl5]2−
with trigonal-bipyramidal arrangement dominates in
both complexes. This conclusion is in good agreement
with magnetic data. A doublet with parameters corre-
sponding to Fe2+ observed in both Mo¨ssbauer spectra
is probably connected with admixture of iron(II) phase
in the starting material FeCl3·6H2O.
(1)
Keeping g fixed at 2.0 and Nh at 60×10−6 cm3 mol−1
the J, and z parameters were determined by non-linear
regression analysis. The best-fit parameters are: J=
−61.6(6) cm−1 and z=0.017(3).
,
Variable temperature magnetic susceptibility mea-
surements over the temperature range 4.5–300 K were
used to study of magnetic properties of complex 4. A
plot of the molar magnetic susceptibilities (M) and the
effective magnetic moment (veff/vB) versus temperature
is shown in Fig. 4. The veff value of 4 at room temper-
ature is 1.93 vB per Cu. As the temperature is de-
creased, the magnetic moment for 1 is slowly reduced
to 1.48 vB at approximately 100 K, and then sharply
decreases to a minimum value of 0.13 vB at 4.5 K. As
can be seen from the Fig. 4, a maximum in the mag-
netic susceptibility is observed at 70 K (the Neel’s
point) indicating thus a medium strength antifferomag-
netic interaction between the paramagnetic Cu(II) cen-
The room temperature magnetic moment of 3 per Cu
is equal to 1.61 veff/vB at 295 K and gradually decrease
with decreasing temperature down to 34 K, where the
veff=0.49 vB (see Fig. 3), suggesting on the presence of
an antiferromagnetic interaction between the copper(II)
ions. Against to this, the  values per dimer increase
M
up to a maximum at approximately 110 K (the Neel’s
temperature) and consequently decrease up to 35 K.
The experimentally observed molar magnetic suscepti-
bility data (per dimer) were treated using the (Eq. (1)),
which includes a corrective term, z, for non-coupled
copper(II) impurity and the temperature-independent
paramagnetism, Nh.
Fig. 2. The Mo¨ssbauer spectra for 1 (left) and 2 (right) at measured 30 K. A – the best-fit of experimental data, B – Fe3+ doublet in [FeCl5]2−
C – Fe3+ singlet in [FeCl5(H2O)]2−, D – Fe2+ doublet.
,
Table 4
Mo¨ssbauer parameters for 1 and 2
Compound Doublet of [FeCl5]2−
Singlet of [FeCl5(H2O)]2−
Doublet of Fe2+
l (mm s−1
)
Z (mm s−1
)
A (%)
l (mm s−1
)
Z (mm s−1
)
A (%)
l (mm s−1
)
Z (mm s−1
)
A (%)
1
2
0.30
0.31
0.45
0.45
51.5
50.0
0.31
0.31
0
0
29.7
39.1
1.21
1.33
3.00
3.27
18.8
10.9
l, isomer shift related to iron; Z, quadrupole splitting; A, percentage of subspectrum area related to iron content.

126
M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
Fig. 3. Thermal dependence of M (z) and veff/vB (5) for 3. The solid line represents the best-fit to the Eq. (1) (see text). Values of M and veff/vB
are given per molecule.
Fig. 4. Thermal dependence of M (z) and veff/vB (5) for 4. The solid line represents the best-fit to the Eq. (1) (see text). Values of M and veff/vB
are given per molecule.
presence of the organic part of the molecules (HL2 or
HL3) in the complexes 3 and 4 is evident from the
dominant peaks at 260 m/z, representing the molecular
peaks of HL2 and HL3. Consecutively, the peaks at
125, 91 and 78 m/z were also observed and they may be
connected with the fragments [ClꢀC6H5ꢀCH2]+,
[C6H5ꢀCH2]+, and [C6H5]+, respectively. The peaks
observed in ES+ mass spectra of 3 and 4 at
654, 617, 582, 512 m/z may be related to fragments of
tres. Below this temperature, the  value decreases to
M
a minimum value at 12.5 K. Consequential increasing
in the magnetic susceptibility below 12.5 K is at-
tributable to a small amount of monomeric impurity.
The observed magnetic susceptibility data for dimer
were fitted to the (Eq. (1)). The best-fit parameters of
g, J, z and Nh using (Eq. (1)) were obtained by a
non-linear fitting procedure with the Marquardt opti-
mization algorithm. The following parameters were ob-
the
mononuclear
species
[Cu1(HL)2Cl2]+,
tained:
g=1.982(4),
J= −40.89(7)
cm−1
,
.
[Cu1(HL)2Cl1]+, [Cu1(HL)2]+, and [Cu1(HLꢀCl)2]+
(where HLꢁHL2 or HL3). Whilst, the peak at 752 m/z
belongs probably to a dicopper(II) fragment [Cu2Cl2-
(HL)(HLꢀCl)]+. The observation of peaks at 750 and
387 m/z, assignable to dicopper(II) fragments
[Cu2(HL)2Cl3]+ or [Cu2Cl2(HL)(HLꢀCl)]+, and
[Cu2(HL)1]+, respectively, is typical for binuclear com-
z=0.0003(1), and Nh=4.1(1)×10−5 cm3 mol−1
3.2. Mass spectroscopy
ES+ mass spectra were measured for the complexes
2, 3 and 4. The ES+ mass spectrum of 2, likewise the
FAB+ mass spectrum of 1, revealed a molecular peak
at 341 m/z clearly identifying the HL1 molecule. The
plexes
3 and 4. The presence of the fragment

M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
127
[Cu2Cl4(HL)4]+ in the ES+ spectrum of 3 is evident
from the peak at 1304 m/z.
complexes 3 and 4 have the binuclear structures as
shown in Figs. 5 and 6, respectively. We suppose that
both copper(II) atoms in complex 3 are six-coordinated
and bridged by two chloride ions and two HL2 ligands
which connect two central atoms through N(3) and
N(9) atoms of HL2. The remaining two HL2 and two
chloride anions are bonded as terminal. In contrast,
both copper(II) ions in complex 4 are five-coordinated
probably in trigonal-bipyramidal arrangement. Thus,
the central atoms in 4 are bridged by two chlorides and
by two HL3 molecules, whereas two remaining chlo-
rides are again bonded as terminal. This assumption
both coordination geometry around the Cu-atoms and
bridging mode in complexes 3 and 4 may be supported
by the values of exchange integral, J= −61.6(6) and
−40.89(7) cm−1, as found for 3 and 4, respectively.
This value is higher than was observed in dichloro-
bridged dicopper(II) systems containing the Cu(v-
Cl)2Cu core which are the most frequently either
slightly antiferromagnetic {J= −8 cm−1 in [Cu2(v-
Cl)2(TMSO)4Cl2], where TMSO=tetramethylene sulf-
oxide [4]; J= −3.7 cm−1 in [Cu(2-pic)2Cl2]2, where
2-pic=2-methylpyridine [21]} or ferromagnetic {J= +
3.15 and +0.31 cm−1 in [Cu(dmgH)Cl2]2, where
dmgH=dimethylglyoxime, as referred in [22,23], re-
spectively; J= +40 cm−1 in (Ph4As)2[Cu2Cl6] [24]} as
described and compared in the literature [7]. Thus, we
suppose that the complex 4 has the same structure as
determined by a single crystal X-ray analysis for
[Cu2(v-Cl)2(v-ntphd2)2Cl2] (where ntphd=1,8-naph-
thyridine) [2], even if the last-mentioned structure
shows a strong antiferromagnetic coupling with J= −
139 cm−1 [25].
We can conclude, on the basis of results following
from physical measurements, that the complexes 1 and
2 are mononuclear consisting of the trigonal-bipyrami-
dal complex anion [FeCl5]2− and two protonized or-
ganic molecules H+HL1 situated outside of the
coordination sphere of the Fe(III) ions. Whereas, the
Fig. 5. Tentative structure for 3.
Fig. 6. Tentative structure for 4.
3.3. Structural description of 6-(2-chlorobenzyl-
amino)purine dihydrate sol6ate, HL2·2H2O, and
6-(3-chlorobenzylamino)purinium chloride, H+HL3ꢀCl
The structure of HL2·2H2O in its electroneutral form
is depicted in Fig. 7, while a drawing of a protonized
form of HL3, H+HL3ꢀCl, is shown in Fig. 8. As can be
seen from Table 2, selected bond lengths and angles in
both structures are quite similar. However, significant
differences in some bond angles of the purine rings
(Table 2) are due to proton shifts on the atoms N(3),
N(7) and N(9). In an electroneutral form of HL2·2H2O,
the hydrogen atoms are bonded to C(2) and N(9) atoms
of a purine ring. Moreover, the H+HL3ꢀCl molecule is
also protonized on the N(3) atom in acidic medium and
the hydrogen atom bonded to N(9) in the former
structure is moved to N(7) in the this structure. There
are no significant deviations from planarity both in
purine (plane 1) and phenyl (plane 2) rings in both
structures [26]. The dihedral angle between the plane 1
and plane 2 is equal to 79.5(2)° in HL2·2H2O and 82(1)°
in H+HL3ꢀCl.
Fig. 7. Molecular structure of 6-(2-chlorobenzylamino)purine dihy-
drate solvate (HL2·2H2O). Thermal ellipsoids are plotted at the 50%
probability level.

128
M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
HL1, HL2 and HL3, was determined. The data obtained
from a LIVE/DEAD viability/cytotoxicity assay are
presented in Table 5. All the tested complexes are
shown to be potent growth inhibitors of different can-
cer cell lines, such as human malignant melanoma
G-361, human osteogenic sarcoma HOS, human
chronic myelogenous leukaemia K-562 and human
breast adenocarcinoma MCF7. The IC50 values for
FeCl3·6H2O and HL1 obtained on all used cell lines
were found to be remarkably higher than those for the
appropriate complexes. This indicates that the cytotoxic
activity of the HL1 ligand increases even more after the
formation of the iron(III) complexes. Cell lines G-361
and MCF7 were the most sensitive to the assayed
iron(III) complexes, displaying cytotoxicity with IC50 in
the range from 21 to 24 mM. Cell lines HOS and K-562
were less sensitive in this case, with IC50 ranging from
27 to 31 mM.
Surprisingly, formation of Cu(II) complexes had even
stronger impact on antitumor activity of non-toxic
compounds HL2 and HL3. The IC50 found for prepared
complexes 3 and 4 were in range from 17 to 39 mM,
whereas the lowest IC50 value determined for starting
compounds, i.e. HL2, HL3 and CuCl2·2H2O, was 95
mM.
Fig. 8. Molecular structure of 6-(3-chlorobenzylamino)purinium chlo-
ride, (H+HL3ꢀCl). Thermal ellipsoids are plotted at the 30% proba-
bility level.
The inhibition of crude p34cdc2 kinase by the most
soluble complexes (1, 2) was also estimated. The data
obtained are shown in Table 6. As can be seen, very
strong inhibitory effect was found for both studied
compounds, indirectly pointing out about identical
composition of complexes 1 and 2. Unfortunately, the
obtained results do not show any clear influence of
complex formation on p34cdc2 kinase inhibition, due to
high activity of starting HL1. However, found dis-
crepancy that complex formation affect cytotoxicity of
used ligands but not their ability to inhibit p34cdc2
kinase can be explained by the finding that the cellular
effect of cytotoxic 6-benzylaminopurine derivatives are
not dependent only on their ability to inhibit CDKs
and could be mediated by several other factors such as
a decrease in the protein synthesis and/or glycolysis
which in turn diminishes the ability of cancer cells to
function [27].
Table 5
IC50 (mM) assessed by calcein AM assay of surviving tumour cells
Compound
Cell line
G-361
HOS
31
K-562
MCF7
1
2
3
4
23
22
17
36
55
29
26
31
32
96
24
21
24
39
58
27
18
38
H+HL1ꢀCl
HL2
60
95
\167
\167
\200
\100
96
129
HL3
FeCl3·6H2O
CuCl2·2H2O
\167
\200
100
\167
\200
\100
\167
\200
\100
The tumor cell lines (G-361, human malignant melanoma; HOS,
human osteogenic sarcoma; K-562, human chronic myelogenous
leukaemia; MCF7, human breast adenocarcinoma) were treated with
solution of the tested compound in the 0.5–200 mM range for 72 h at
37 °C, 5% CO2.
Unfortunately, our attempts to prepare crystals of
compounds 1–4 suitable for a single crystal X-ray
analysis, which could be positively to determine struc-
ture of the reaction products, were unsuccessful up to
now.
Table 6
IC50 values for tested compounds added to crude p34cdc2 kinase
Compound
IC50 (mM)
2-(3-Hydroxypropylamino)-6-(benzylamino)-9-
1.2
isopropylpurine (Bohemin, HL1)
1
2
1.0
0.9
3.4. Biological acti6ity
Enzyme activities were assayed as described in the Section 2, in the
presence of increasing concentrations of tested compounds. IC50
values were substracted from the dose-response curves.
Antitumor activity of all prepared complexes and the
starting compounds, i.e. CuCl2·2H2O, FeCl3·6H2O,

M. Malonˇ et al. / Inorganica Chimica Acta 323 (2001) 119–129
129
[8] J.A. Kuhnle, G. Fuller, J. Corse, B.E. Mackey, Physiol. Plant 41
(1977) 14.
4. Supplementary material
[9] E. Ko¨nig, Magnetic Properties of Coordination and
Organometallic Transition Metal Compounds, Springer, Berlin,
1966.
[10] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81.
[11] G.M. Sheldrick, Acta Crystallogr. A46 (1990) 467 SHELXS-97.
[12] G.M. Sheldrick, SHELXL-97, Program for crystal structure refine-
ment, University of Go¨ttingen, Germany, 1997.
[13] Z. Tra´vn´ıcˇek, M. Malonˇ, M. Biler, M. Hajdu´ch, P. Brozˇ, K.
Dolezˇal, J. Holub, V. Krysˇtof, M. Strnad, Transition Met.
Chem. 25 (2000) 265.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC Nos. 156251 for compound
HL2·2H2O and 156252 for compound H+HL3ꢀCl.
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 http://www.ccdc.
cam.ac.uk).
[14] A.B.P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Am-
sterdam, 1968.
[15] (a) K.H. Reddy, P.S. Reddy, P.R. Babu, J. Inorg. Biochem. 77
(1999) 169;
(b) R.J. Parker, L- Spiccia, B. Moubaraki, K.S. Murray, B.W.
Skelton, A.H. White, Inorg. Chim. Acta 302 (2000) 922.
[16] C.J. Pouchert, The Aldrich Library of Infrared Spectra (Edition
III), Aldrich Chemical Co., INC, 1001 W. St. Paul. Ave., Mil-
waukee, Wisconsin, 53233, 1981.
[17] T. Fujita, T. Sakaguchi, Chem. Pharm. Bull. 25 (9) (1977) 2419.
[18] O. Kahn, Molecular Magnetism, Wiley-VCH Inc, New York,
1993.
Acknowledgements
This research was supported by the Grant Agency of
the Czech Republic (grants no. 203/00/0152 and 203/
99/0067) and MSMT (grant no. MSM153100008).
[19] B.D. James, J. Liesegang, M. Balakova, W.M. Reiff, B.W.
Skelton, A.H. White, Inorg. Chem. 34 (1995) 2054.
[20] L. Takacs, W.M. Reiff, B.D. James, Hyperfine Interact. 28
(1986) 693.
[21] D.Y. Jeter, D.J. Hodgson, W.E. Hatfield, Inorg. Chim. Acta 5
(1972) 257.
[22] D.J. Hodgson, Prog. Inorg. Chem. 19 (1975) 225.
[23] Ch. Chow, R.D. Willet, B.C. Gerstein, Inorg. Chem. 14 (1975)
205.
References
&
[1] Z. Tra´vn´ıcˇek, M. Malonˇ, Z. Sindela´rˇ, K. Dolezˇal, J. Rolcˇ´ık, V.
Krysˇtof, M. Strnad, J. Marek, J. Inorg. Biochem. 84 (2001) 23.
[2] C. Mealli, F. Zanobini, J. Chem. Soc., Chem. Commun. (1982)
97.
[3] I. Søtofte, K. N, Acta Chem. Scan. A 35 (1981) 733.
[4] D.D. Swank, G.F. Needham, R. Willet, Inorg. Chem. 18 (1979)
[24] M. Me´gnamisi-Be´lombe´, M.A. Novotny, Inorg. Chem. 19 (1980)
2470.
761.
[5] J.C. Dayson, L.M. Engelhardt, P.C. Healy, C. Pakawatchai,
[25] K. Emerson, A. Emad, R.W. Brookers, R.L. Martin, Inorg.
Chem. 12 (1973) 978.
A.H. White, Inorg. Chem. 24 (1985) 1950.
[6] J. Reedijk, Bioinorganic Catalysis, Marcel Dekker, New York,
[26] M. Nardelli, J. Appl. Cryst. 28 (1995) 659.
[27] H. Kova´rˇova´, M. Hajdu´ch, G. Korˇ´ınkova´, P. Halada, S.
Krupicˇkova´, A. Gouldsworthy, N. Zhelev, M. Strnad, Elec-
trophoresis 21 (2000) 3757.
1993.
[7] M. Rodriguez, A. Llobet, M. Corbella, Polyhedron 19 (2000)
2483.