EPR spectroscopic study of a dinuclear copper(II) complex of tolfenamic acid

Article abstract of DOI:10.1016/j.ica.2004.06.064

The copper(II) complex with tolfenamic acid [Cu(tolf)₂(H ₂O)]₂ was studied by X-band and K-band EPR spectroscopies in the temperature range from 90 to 300 K. The Cu²⁺ ions in the dinuclear complex show a strong antiferromagnetic exchange interaction with |J| = 292 cm^-1. The spin-Hamiltonian parameters indicate that the local environment around the copper species is distorted tetragonal pyramid. The copper(II) complex with tolfenamic acid [Cu(tolf)₂(H ₂O)]₂ was studied by X-band and K-band EPR spectroscopies in the temperature range from 90 to 300 K. The Cu²⁺ ions in dinuclear complex show a strong antiferromagnetic exchange interaction with |J| = 292 cm^-1. The EPR spectra, which were observed for [Cu(tolf) ₂(H₂O)]₂, are typical powder spectra of the copper pairs. The spectra exhibit the hyperfine structure in low temperature range. The values of the spin-Hamiltonian parameters were determined on the basis of the best fit for the simulated spectra at both K-band (0.75 cm ^-1) at T = 298 K and X-band (0.3 cm^-1) at T = 93 K as compared with the experimentally observed spectra. These values show that the local environment around the copper species is distorted tetragonal pyramid. This EPR evidence is consistent with the crystallographic data.

Full text of DOI:10.1016/j.ica.2004.06.064

Inorganica Chimica Acta 358 (2005) 186–190
www.elsevier.com/locate/ica
Note
EPR spectroscopic study of a dinuclear copper(II) complex
of tolfenamic acid
a
D. Kovala-Demertzi , D. Skrzypek , B. Szymanska , A. Galani , M.A. Demertzis
b,
b
a
a
*
´
a
Inorganic and Analytical Chemistry, Department of Chemistry, University of Ioannina, 45 100 Ioannina, Greece
b
Chełkowski Institute of Physics, University of Silesia, Uniwersytecka 4, 40-007 Katowice, Poland
Received 5 March 2004; accepted 30 June 2004
Available online 18 September 2004
Abstract
The copper(II) complex with tolfenamic acid [Cu(tolf)2(H2O)]2 was studied by X-band and K-band EPR spectroscopies in the
temperature range from 90 to 300 K. The Cu2+ ions in dinuclear complex show a strong antiferromagnetic exchange interaction
with |J| = 292 cmꢀ1. The EPR spectra, which were observed for [Cu(tolf)2(H2O)]2, are typical powder spectra of the copper pairs.
The spectra exhibit the hyperfine structure in low temperature range. The values of the spin-Hamiltonian parameters were deter-
mined on the basis of the best fit for the simulated spectra at both K-band (0.75 cmꢀ1) at T = 298 K and X-band (0.3 cmꢀ1) at
T = 93 K as compared with the experimentally observed spectra. These values show that the local environment around the copper
species is distorted tetragonal pyramid. This EPR evidence is consistent with the crystallographic data.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Copper; Anti-inflammatory drugs; EPR
1. Introduction
ence of two polymorphic forms containing the
molecule as different conformers (Fig. 1) [8].
Synthesis and study of metal compounds with non-
steroidal anti-inflammatory drug (NSAID), which
exhibit synergistic activity, have concentrated much
attention as an approach to new drug development. A
great deal is known about the role of copper complexa-
tion in enhancing the pharmacological profile of
NSAIDs activity and reducing toxicity. A range of
Cu(II) complexes have been studied as anti-inflamma-
tory agents [1–8].
Tolfenamic acid (i.e., N-(3-chloro-2-methylphe-
nyl)anthranilic acid, Htolf) is a potent, well-tolerated
NSAID with a low gastroulcerogenicity, a low overall
toxicity and high therapeutic indications [8]. X-ray crys-
tal structure investigations of Htolf revealed the exist-
The synthesis, crystal structure, IR and UV–Vis spec-
troscopies and electrochemical studies of Cu(II) com-
plexes of Htolf were reported by us recently [8]. The
molecular structure of [Cu(tolf)2(DMF)]2 consists of
two Cu atoms linked by four tolf groups in a fashion
similar to that of [Cu(diclofenac)2(DMF)]2 [2] and
[Cu(indomethacin)2(DMF)]2 Æ 1.6 DMF [6]. The super-
oxide dismutase activity was measured and IC50 value
was determined by the Fridovich test. Following the
usual criteria, the present complexes are considered as
potent SOD mimics.
The analysis of electronic and magnetic properties is
important to establish the structure of the active compo-
nents of the pharmaceutical preparations. In this paper,
we report the EPR spectroscopic study of a dinuclear
copper(II) complex with tolfenamic acid [Cu(tolf)2-
(H2O)]2 (Fig. 2).
*
Corresponding author.
E-mail address: dskrzyp@us.edu.pl (D. Skrzypek).
0020-1693/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.06.064

D. Kovala-Demertzi et al. / Inorganica Chimica Acta 358 (2005) 186–190
187
Fig. 1. The structural representation of Htolf [8].
2. Experimental
Fig. 3. K-band EPR spectrum of [Cu(tolf)2(H2O)]2 observed at room
temperature (in simulated spectrum (b) M-signal is not included).
2.1. Synthesis of the complex
The reagents (Aldrich, Merck) were used as supplied,
while the solvents were purified according to standard
procedures. Tolfenamic acid was a gift from ‘‘ELPEN
A.E’’. The dinuclear complex of tolfenamic acid [Cu-
(tolf)2(H2O)]2 was prepared according to the reaction
in methanolic solution in the pH (apparent) range of
6–7
magnetic field strength was monitored by an NMR gauss
meter. The X-band measurements were performed in
temperature range from 90 to 300 K with a liquid nitro-
gen variable temperature device. For calculation of the
EPR spectra of powder computer simulation program
WIN-EPR SimFonia (Bruker) was involved.
2CuCl2 þ 4Htolf þ 4NaOCH3 þ 2H2O
3. Results and discussion
! 2½CuðtolfÞ ðH2OÞꢁ þ 4NaCl þ 4CH3OH
2
2
Typical powder EPR spectra for [Cu(tolf)2(H2O)]2
complex are shown in Figs. 3(a) and Fig. 4. The spectra
are composed of: (i) signals denoted by P, which are
attributed to copper(II) ion pairs, (ii) signal denoted
by M, which is attributed to isolated Cu2+ ions.
Slow crystallization of [Cu(tolf)2(H2O)]2 from dimethyl-
formamide solution afforded green crystals of
[Cu(tolf)2(DMF)]2.
2.2. EPR spectroscopy
Electron paramagnetic resonance spectra of the solid
[Cu(tolf)2(H2O)]2 were recorded with Radiopan EPR
Spectrometer operating at X-band (9 GHz) and K-band
(22.3 GHz) frequencies, using 100 kHz field modulation.
The microwave frequencies were measured using Hewlett
Packard 534 microwave frequency counter and the
Fig. 4. X-band EPR spectrum of [Cu(tolf)2(H2O)]2 observed at room
temperature.
Fig. 2. The representation of the dinuclear complex [Cu(tolf)2(H2O)]2.

188
D. Kovala-Demertzi et al. / Inorganica Chimica Acta 358 (2005) 186–190
from the temperature dependence of the intensity of P-
type signals in relation to paramagnetic copper sample.
The value of J is J = ꢀ292 cmꢀ1. Dex = 0.672 cmꢀ1
,
Ddd = ꢀ0.326 cmꢀ1 and r = 0.227 nm were derived from
equations occurred in [10]. The value of the inter-ion
distance r does not correspond to r = 0.260 nm esti-
mated from X-ray studies. This is because we deduced
r-parameter from the temperature dependence of only
the part of EPR spectrum and from approximate theory.
The large negative J value shows that the [Cu(tolf)2-
(H2O)]2 complex has a singlet ground state with a triplet
excited state [9], which is 292 cmꢀ1 higher in energy. The
observed value of J is comparable to those of related
dinuclear Cu(II) carboxylates, e.g., [Cu(Indo)2(H2O)]2
[6], which possess similar structure to our complex. In
the case when J is large, we can describe the EPR spectra
in terms of the triplet state alone (S = 1) [9]. In the
K-band spectrum (Fig. 3, D < hm), there are two very
strong peaks (P^) and on both sides of these peaks there
are signals (Pi) whose the intensity is significantly smal-
ler. The X-band EPR spectrum (D > hm) consists of one
strong line and two weak lines: one in low field and one
in high field (P^ and Pi in Fig. 4).
Fig. 5. Variation in EPR spectrum intensity (measured by double
integration) for copper pairs in [Cu(tolf)2H2O)]2 vs. the temperature.
3.1. The EPR spectra of Cu–Cu pairs in [Cu(tolf)2-
(H2O)]2 complex
The spin-Hamiltonian for a pair of interaction ions 1
and 2 may be written [9–13] as
The EPR spectrum of [Cu(tolf)2(H2O)]2 may be inter-
pretable in terms of the following spin-Hamiltonian:
H ¼ Hi þ SDS þ JS1S2;
where
ð1Þ
H ¼ g l BzSz þ g l ðBxSx þ BySyÞ þ A SzðI1z þ I2zÞ
k
k
B
?
B
X
þ A fSxðI1x þ I2xÞ þ SyðI1y þ I2yÞg
?
Hi ¼
ðlBgjSi;jBj þ AjSi;jIi;jÞ; i ¼ 1; 2:
þ DfS2z ꢀ 1=3SðS þ 1Þg þ EðSx2 ꢀ S2y Þ;
ð3Þ
j¼xi;yi;zi
where gi, Ai and g^, A^, D and E have their usual
meaning.
S1, S2, S, I, I, A, J and D have their usual meaning. The
magnitude of D is a function of two terms [10]: a dipolar
contribution Ddd, which is related to the Cu–Cu dis-
tance, and the anisotropic exchange contribution, Dex.
The nature of exchange coupling within Cu–Cu pairs
is evident from the thermal dependence of EPR intensity
(Fig. 5) [12]. From these data, we can see that the triplet
state becomes depopulated as the temperature is lowered
and so the signal intensity drops. The population of the
triplet state is governed by Boltzmann distribution and
the Curie-law
This Hamiltonian was used to simulate the experi-
mental spectrum of Fig. 3. The simulation refers only
to the allowed transitions (DMS = 1, DMI = 0). The
weak signal, denoted as F, corresponds to the forbidden
transition (DMS = 2). The best-fit parameters are sum-
marized in Table 1.
We have assumed that the two ions are in identical
sites of axial symmetry with small rhombic distortion
(see Fig. 6, at low temperature the 0resonance peak near
420 mT splits into doublet, g? ꢀ g? ffi 0:005). The inter-
nuclear vector r is assumed to be parallel to the z-axis of
the system.
I=I0 ¼ 1=f3 þ expðJ=kTÞg:
ð2Þ
The Cu–Cu pairs in [Cu(tolf)2(H2O)]2 complex are
antiferromagnetically coupled. The J value, in agree-
ment with Eq. (2), was deduced as a fitting parameter
D and g parameters (Table 1) are similar to those for
Cu(II) carboxylate complexes where unpaired electron is
Table 1
EPR parameters of [Cu(tolf)2(H2O)]2 complex
a
b
Signal
T (K)
gi
g^
Aia
77.6
77.6
152
152
A^
Da
Ea
DBib
DB^
P (Cu–Cu pairs)
298
93
2.48
2.48
2.06
2.09
13.5
13.5
3458
3458
–
34
80
30
80
30
M (isolated Cu2+
)
298
97
2.32
2.32
2.06
2.07
25
25
–
–
–
–
60
40
27
18
a
In 10ꢀ4 cmꢀ1
In Gs.
.
b

D. Kovala-Demertzi et al. / Inorganica Chimica Acta 358 (2005) 186–190
189
Fig. 6. The high field part of X-band EPR spectrum of
[Cu(tolf)2(H2O)]2 at T = 93 K.
Fig. 7. X-band EPR spectra for isolated Cu2+ ions in [Cu(tolf)2(H2O)]2
at T = 97 K.
in a dx2ꢀy2 orbital. The observed D value, 0.3458 cmꢀ1, is
comparable to those found for dinuclear Cu(II) com-
pounds with exclusively oxygen donors (D = 0.34
cmꢀ1) [6]. The best-fit E value is very small, in keeping
with the strong axial symmetry observed in related
X-band and K-band spectra. From this it may be rea-
sonably proposed that [Cu(tolf)2(H2O)]2 has dimeric
structure [Cu2(tolf)2L]2 with a distorted tetragonal-
pyramidal configuration around each copper(II) atom.
Each of these is coordinated to four oxygen atoms of
the tolfenamic anions (in plane) and donor atom O
(from L = H2O) at the apex. The apical ligand is one
which plays an important role in determining the J value
in dinuclear compounds, however, in the literature exists
conflicting evidence concerning this influence [6].
The effect of the hyperfine interaction with two equiv-
alent copper ions, for which nuclear spin I = 3/2, is
observable with the lowering of the temperature
(Fig. 6). In axial symmetry, the resolved structure with
the seven line pattern with intensities 1: 2: 3: 4: 3: 2: 1
arises from those pairs for which B is parallel to the z-
axis. The strong lines, with generally no hyperfine struc-
ture, arise from those molecules for which B is aligned in
the xy plane. The lack of hyperfine structure is a result of
the smaller hyperfine component in the xy plane com-
pared with that along the z axis.
interpretable in terms of the following spin-
Hamiltonian:
H ¼ lBfgkBzSz þ g?ðBxSx þ BySyÞg þ AIIIzSz
þ A ðIxSx þ IySyÞ;
ð4Þ
?
where S = 1/2 and I = 3/2. The pattern (b) of Fig. 7 shows
the simulation of the experimental spectrum with the
parameters listed in Table 1. In the perpendicular region,
the spectrum is not well resolved, though some hyperfine
structure is observed. The peak–peak width of the deriv-
ative spectrum (roughly 11 mT) suggests a value of about
3 mT for A^. The lack of clear resolution in this spectral
region is presumably due to the large number of overlap-
ping DMI = 0 and DMI > 0 transitions and to the large
intrinsic widths of the individual transitions. The cop-
per(II) spectra are also characterized by the strong ‘‘ex-
tra’’ absorption, which has its origin in the angular
anomalies in the MI = +3/2 hyperfine structure compo-
nent. In Fig. 7 we can see the splitting of the parallel com-
ponents of the spectrum. The shoulders near parallel
hyperfine components arise from 63Cu components.
The 65Cu parallel components are displaced in relation
to the 63Cu components with Ai (65Cu)/Ai (63Cu) = 1.07.
4. Conclusions
3.2. The EPR spectra of isolated Cu2+ ions in
[Cu(tolf)2(H2O)]2 complex
The EPR parameters estimated for copper complex
of [Cu(tolf)2(H2O)]2 are characteristic for distorted
tetragonal pyramid. This EPR evidence is consistent
with the crystallographic studies where we obtained four
short Cu–ORCOO (0.1944–0.1951 nm) bond lengths
and long solvent Cu–O (0.2140 nm) bond length [8].
The values of hyperfine splitting of the dimer are two
times smaller than for the monomer. This indicates that
the corresponding Cu(II) ions are located in the same
The observation of EPR spectra in [Cu(tolf)2(H2O)]2
shows that small amounts of isolated copper ions exist
in this compound. Mononuclear signals are commonly
found for dinuclear copper carboxylate complexes [6].
EPR spectrum in Fig. 7 (ꢂ300 mT) displays the typical
features of a powder-like EPR spectra for copper ions
having dx2ꢀy2 as the ground state [14,15] and may be

190
D. Kovala-Demertzi et al. / Inorganica Chimica Acta 358 (2005) 186–190
[6] J.E. Weder, T.W. Hambley, B.J. Kennedy, P.A. Lay, D.
MacLachlan, R. Bramley, C.D. Delfs, K.S. Murray, B. Mouba-
raki, B. Warwick, J.R. Biffin, H.L. Regtop, Inorg. Chem. 38
(1999) 1736, and references therein.
type of environment. The intensity ratio of isolated cop-
per ion signals to pair copper signals IM/ID = 0.003
means that isolated ion species are impurities in sample.
The investigations of the local symmetry around the
copper species are important because the axial ligands
of complexes are likely to exchange during drug
metabolism.
[7] M. Konstandinidou, A. Kourounaki, M. Yiangou, L. Hadjipe-
trou, D. Kovala-Demertzi, S.K. Hadjikakou, M.A. Demertzis, J.
Inorg. Biochem. 70 (1998) 63.
[8] D. Kovala-Demertzi, A. Galani, M.A. Demertzis, S. Skoulika,
C. Kotoglou, J. Iorg. Biochem. 98 (2004) 358, and references
therein.
[9] A. Abragam, B. Bleaney, Electron Paramagnetic Resonance of
Transition Ions, Clarendon Press, Oxford, 1970.
References
[10] A. Aboukais, A. Bennani, C.F. Aissi, G. Wrobel, M. Guelton,
J.C. Vedrine, J. Chem. Soc., Faraday Trans. 88 (1992) 615.
[11] J. Stankowski, M. Mackowiak, Prog. Phys. 23 (1972) 515 (in
Polish).
[1] J.E. Weder, C.T. Dillon, T.W. Hambley, B.J. Kennedy, P.A.
Lay, J.R. Biffin, H.L. Regtop, N.M. Davies, Coord. Chem. Rev.
232 (2002) 95, and references therein.
[12] T.D. Smith, J.R. Pilbrow, Coord. Chem. Rev. 13 (1974) 173.
[13] A. Bencini, D. Gatteschi, Electron Paramagnetic Resonance of
Exchange Coupled Systems, Springer, Berlin, 1990.
[14] S.K. Hoffmann, in: J. Stankowski (Ed.), Solid State Radiospectr-
oscopy, PWN, Warsaw, 1975 (in Polish).
[2] D. Kovala-Demertzi, A. Theodorou, M.A. Demertzis, C. Rap-
topoulou, A. Terzis, J. Inorg. Biochem. 65 (1997) 151.
[3] D. Kovala-Demertzi, J. Inorg. Biochem. 79 (2000) 153.
[4] J.R. Sorenson, Recl. Trav. Chim. Pay. B 106 (1987) 378.
[5] R. Cini, Comments Inorg. Chem. 22 (2000) 151.
[15] J. Stankowski, J. Mol. Struct. 597 (2001) 109.