Spectroscopic, Magnetic, and Electrochemical Studies of a Dimeric N-Substituted-Sulfanilamide Copper(II) Complex. X-ray and Molecular Structure of the Cu2(sulfathiazolato)4 Complex

Article abstract of DOI:10.1021/ic960968p

A copper(II) complex of formula Cu₂(stz)₄ (stz^- = sulfathiazolato) has been synthesized and characterized by single-crystal X-ray diffraction. The compound crystallizes in the orthorhombic system, space group P2₁cn with a = 10.595(7) ?, b = 14.274(3) ?, c = 29.65(1) ?, and Z = 4. The structure consists of dinuclear copper(II) units which contain four sulfathiazolato ligands bridging the metal ions through a nonlinear NCN group. The copper atoms are four-coordinated, the chromophore being CuN₄. The Cu?Cu bond distance is 2.671(2) ?. Magnetic susceptibility data in the temperature range 7-300 K show the occurrence of intramolecular antiferromagnetic coupling with 2J = -61.5 cm^-1. This low exchange energy value has been analyzed and rationalized through extended Hückel calculations. EPR spectra at X- and Q-band frequencies show the signals corresponding to a dinuclear entity, being the zero-field splitting parameter, D = 0.230 cm^-1.

Full text of DOI:10.1021/ic960968p

2052
Inorg. Chem. 1997, 36, 2052-2058
Spectroscopic, Magnetic, and Electrochemical Studies of a Dimeric
N-Substituted-Sulfanilamide Copper(II) Complex. X-ray and Molecular Structure of the
Cu2(sulfathiazolato)4 Complex
Javier Casanova, Gloria Alzuet, Julio Latorre, and Joaqu´ın Borra´s*
Departamento de Qu´ımica Inorga´nica, Universidad de Valencia, Avenida Vicent Andre´s Estelle´s, s/n,
46100 Burjassot, Valencia, Spain
ReceiVed August 8, 1996X
A copper(II) complex of formula Cu2(stz)4 (stz- ) sulfathiazolato) has been synthesized and characterized by
single-crystal X-ray diffraction. The compound crystallizes in the orthorhombic system, space group P21cn with
a ) 10.595(7) Å, b ) 14.274(3) Å, c ) 29.65(1) Å, and Z ) 4. The structure consists of dinuclear copper(II)
units which contain four sulfathiazolato ligands bridging the metal ions through a nonlinear NCN group. The
copper atoms are four-coordinated, the chromophore being CuN4. The Cu‚‚‚Cu bond distance is 2.671(2) Å.
Magnetic susceptibility data in the temperature range 7-300 K show the occurrence of intramolecular
antiferromagnetic coupling with 2J ) -61.5 cm-1. This low exchange energy value has been analyzed and
rationalized through extended Hu¨ckel calculations. EPR spectra at X- and Q-band frequencies show the signals
corresponding to a dinuclear entity, being the zero-field splitting parameter, D ) 0.230 cm-1
.
Introduction
Sulfanilamide and its N-substituted derivatives are well-
known antibacterial drugs. The metal complexes of these
substances have been extensively studied.1-7 From the spec-
troscopic properties of the compounds, Bult8 has suggested that
the drugs behave as bidentate ligands through the Nsulfonamido
and the Nheterocyclic atoms.
We have undertaken several studies on the characterization
of some N-substituted sulfanilamide derivatives, mainly the
sulfathiazole (Hstz) (Figure 1).
Until our research, no crystal structures of sulfathiazole metal
complexes have been reported. In previous papers9-13 we have
obtained crystal structures of copper(II) and zinc(II) sulfathiazole
compounds. In our study about the behavior as ligand of the
Hstz, we found different forms of interaction with the metals.
As a neutral ligand, the Hstz acts as monodentate, binding the
Figure 1. Sulfathiazole.
metal ion through the Namino atom.10,11 As a deprotonated ligand,
stz- has a variety of coordination behavior, in the [Zn(stz)2]‚H2O
complex, it binds to one Zn(II) through the Namino and to another
Zn(II) through the Nthiazole atom in a polymeric structure;9 in
the [Cu(stz)(py)3Cl] compound, the anionic ligand binds to the
Cu(II) through the Nthiazole atom,12 and in the [Cu(stz)2(Him)2]‚
MeOH and [Cu(stz)2(mim)2]‚H2O complexes, one of the stz-
exhibits bidentate behavior linking the metal ion through the
Nthiazole and the Nsulfonamido atoms.13
Here, we report a binuclear copper complex. To our
knowledge, this compound is the first structurally characterized
example of a dimer of sulfathiazole with the ligand bridging
the two copper ions through the Nthiazole and Nsulfonamido atoms.
The Cu2(stz)4 complex is a dimer one, classified in the type
IV of copper dimers14 with triatomic NCN bridging groups. An
interesting aspect of the ligand in this compound is the presence
of a nonlinear bridging NCN, also found in molecules of
biological importance like adenine,15-20 and that provides an
efficient pathway for coupling magnetism. Considerable at-
tention has focused on the bridging capabilities of a closely
related system containing the NCN group such as the copper
complexes of the adenine. The synthesis of these dimers is
important for studying by magnetic and spectroscopic techniques
the nature of the spin-spin interactions which occur intramo-
lecularly between the metal atoms.
* Author to whom correspondence should be addressed.
X Abstract published in AdVance ACS Abstracts, April 1, 1997.
(1) Bult, A. Pharm. Weekbl., Sci. Ed. 1981, 3, 1.
(2) Bult, A. Metal Complexes of Sulfanilamides. In Metal Ions in
Biological Systems; Sigel, Ed.; Dekker: New York, 1983; Vol. 16.
(3) Narang, K. K.; Gupta, J. K. Transition Met. Chem. (London) 1977, 2,
83.
(4) Narang, K. K.; Gupta, J. K. Transition Met. Chem. (London) 1977, 2,
181.
(5) Malecki, F.; Staroscik, R.; Weiss-Gradzinska, W. Pharmazie 1984,
39, 158.
(6) Pezeshk, A.; Pezeshk, V. J. Inorg. Biochem. 1990, 38, 185.
(7) Bonamartini Corradi, A.; Gozzoli, E.; Menabue, L.; Saladini, M.;
Battaglia, L. P.; Sgarabotto, P. J. Chem. Soc., Dalton Trans. 1994,
273.
(8) Bult, A.; Uitterdijk, J. D.; Klasen, H. B. Transition Met. Chem.
(London) 1979, 4, 285.
(9) Casanova, J.; Alzuet, G.; Ferrer, S.; Borra´s, J.; Garc´ıa-Granda, S.;
Perez-Carren˜o, E. J. Inorg. Biochem. 1993, 51, 689.
(10) Casanova, J.; Alzuet, G.; Borra´s, J.; Amigo´, M.; Debardemaeker, T.
Z. Kristallogr. 1993, 209, 271.
(11) Casanova, J.; Alzuet, G.; Borra´s, J.; Timoneda, J.; Garc´ıa-Granda, S.;
Ca´ndano-Garc´ıa, I. J. Inorg. Biochem. 1994, 56, 65.
(12) Casanova, J.; Alzuet, G.; Borra´s, J.; Latorre, J.; Sanau´, M.; Garc´ıa-
Granda, S. J. Inorg. Biochem. 1995, 60, 219.
(14) Hathaway, B. J. Copper. In ComprehensiVe Coordination Chemistry;
Wilkinson, G., Gillard, R. D., Cleverty, J. A., Eds.; Pergamon Press:
New York, 1987; Vol. 5.
(15) Sletten, E. Acta Crystallogr. 1969, B25, 1480.
(16) de Meester, P.; Goodgame, D. M. L.; Angela Price, K.; Skapski, A.
C. Nature 1971, 229, 191.
(17) de Meester, P.; Skapski, A. C. J. Chem. Soc. A 1971, 2167.
(18) Terzis, A.; Beauchamp, A. L.; Rivest, R. Inorg. Chem. 1973, 12, 1170.
(19) Brookes, R. W.; Martin, R. L. Inorg. Chem. 1975, 14, 528.
(20) Sonnenfroh, D.; Kreilick, R. W. Inorg. Chem. 1980, 19, 1259.
(13) Casanova, J.; Alzuet, G.; Borra´s, J.; Carugo, O. J. Chem. Soc., Dalton
Trans. 1996, 2239.
S0020-1669(96)00968-8 CCC: $14.00 © 1997 American Chemical Society

X-ray and Molecular Structure of Cu2(stz)4
Inorganic Chemistry, Vol. 36, No. 10, 1997 2053
Table 1. Crystallographic Data for the Cu2(stz)4 Complex
Experimental Section
formula
C36H32Cu2- cryst size (nm)
N12O8S8
1144.3
0.15 × 0.10 ×
Materials. Copper fluoride dihydrate and sulfathiazole were
reagents of grade and used without purification. Analytical data (C,
H, N) were obtained in a Carlo Erba Instrument Model MOG 1106 of
the Inorganic Chemistry Department of the University of Salamanca
(Spain). The copper content was determined by atomic absorption
spectroscopy.
0.15
fw
µ(Mo KR) (cm-1
λ(Mo KR) (Å)
)
32.30
0.710 73
space group P21cn
a (Å)
b (Å)
c (Å)
10.595(7) orientation reflcns number 25
10.274(3) range (deg)
15 < θ < 25
1 < θ < 25
4145
3243
0.035
29.65(1)
4
4483(4)
2776
data collcn range (deg)
no. of unique data
Z
Synthesis of the Complex. A methanolic solution (25 mL)
containing copper(II) fluoride (0.10 g, 1 mmol) was added to a warm
methanolic solution (50 mL) containing Nastz (0.55 g, 2 mmol) with
continuous stirring. The solution became dark blue, and immediately
a purple precipitate was obtained. The solid was filtered off and was
dried (yield: 90%). Single crystals were obtained through the diffusion
method. Anal. Calcd for C36H32Cu2N12O8S8: C, 37.78%; H, 2.82%;
N, 14.69%; Cu, 11.10%. Found: C, 37.10%; H, 2.90%, N, 14.41%;
Cu, 11.32%. Selected IR bands (KBr cm-1): 3480m, 3390m, NH;
1500m, thiazole ring; 1320s, 1130s, 550-570s, SO2. It should be
pointed out that when other copper (II) salts (chloride, nitrate,
perchlorate, etc.) were used, the same dinuclear compound is obtained,
but only single crystals suitable for X-ray diffraction measurements
were formed with copper(II) fluoride.
V (Å3)
F(000)
Dc (g‚cm-3
total with Fo > 4σ(Fo)
a
R1
b
)
1.695
Rw2
0.1086
a R1 ) ∑||Fo| - |Fo||/∑|Fo|. Rw2 ) ∑ω(Fo2 - Fc2)2/ω(Fo )2. ω )
b
2
2
2
1/[σ2(Fo ) + (0.1000P); P ) [(max (Fo ,0) + 2(Fc)2]/3.
additional empirical absorption correction was applied using DIFABS.25
The maximum and minimum absorption correction factors were
respectively 1.07 and 0.92. Hydrogen atoms were geometrically placed.
During the final stages of the refinement the positional parameters
and the anisotropic thermal parameters of the non-hydrogen atoms were
refined. The hydrogen atoms were isotropically refined with a common
thermal parameter. The final conventional agreement factors were R1
Physical Techniques. IR spectra were recorded on a Perkin-Elmer
843 instrument. Samples were prepared using the KBr technique. Vis-
UV spectra was recorded on a Perkin-Elmer Lambda 15 spectropho-
tometer. EPR spectra of ground crystals were carried out at X-band
frequencies with a Bruker ER 200D spectrometer and at Q-band
frequencies with a Bruker-ESP300 equipped with an Oxford continuous-
flow cryostat. Room-temperature EPR spectra of copper complex
solutions were taken in a flat cell purchased from Wilmad Glass Co.
Magnetic susceptibility measurements were carried out in the 7-300
K temperature range with a fully automatized AZTEC DSM8 pendulum-
type susceptometer equipped with a TBT continuous-flow cryostat and
a Bruker BE15 electromagnet, operating at 1.8 T. Mercury tetrakis-
(thyocianato)cobaltate(II) was used as a susceptibility standard. Cor-
rections for the diamagnetism of the complex was estimated from
Pascal’s constants to -562 × 10-6 emu. Magnetic susceptibility data
were also corrected for temperature-independent paramagnetism [60
× 10-6 emu/Cu(II)] and magnetization of the sample holder.
) 0.035 and Rw ) 0.1086 for the 3243 “observed” reflections and
2
631 parameters. The maximum shift to the estimated standard deviation
(esd) ratio in the last full matrix least squares cycle was 0.01. The
final difference Fourier map showed no peaks higher than 0.49 e‚A-3
nor deeper than 0.62 e‚A-3. Atomic scattering factors were taken from
International Tables for X-ray Crystallography.26 The crystallographic
plots were made with ORTEP.27 All calculations were made on an
IBM PS Pentium computer. All crystallographic details are listed in
Table 1.
Results and Discussion
Description of the Structure. Final atomic coordinates for
all non-hydrogen atoms are listed in Table 2. Figure 2 shows
an ORTEP drawing of the structure of the dimer unit with the
atomic numbering scheme. Selected bond distances and angles
are collected in Table 3. The structure consists of dimer units
which contain stacking interactions between the aromatic rings
and tridimensionally connected by means of hydrogen bonds
(see Figure 3).
Electrochemical measurements were carried out in a three-electrode
cell; the working and auxiliary electrodes were platinum, and the
reference one was a saturated calomel electrode (SCE) electrically
connected to the solution via a “salt bridge” containing a saturated
solution of supporting electrolyte and the solvent. Cyclic voltammo-
grams were obtained with a BAS potentiostat Model CV-27 and
recorded on a Omnigraphic 100 recorder.
Each copper atom of the dinuclear species [Cu1 and Cu2]
is four-coordinated, with a slightly distorted square-planar
environment: the Cu1 is linked to N1 and N8 thiazole nitrogen
atoms from two sulfathiazolato ligands and N2 and N3
sulfonamido nitrogen atoms from the other two sulfathiazolato
anions; the Cu2 is bonded to N5 and N6 of the thiazole rings
and N4 and N7 of the sulfonamido groups. The metal-ligand
distances Cu-Nthiazole are in the range from 1.93 to 1.98 Å,
and the Cu-Nsulfonamido ones are in the range from 2.01 to 2.09
Å. These distances are similar to those observed in the
previously reported sulfathiazole complexes.9,11,13 The bond
angles in both CuN4 chromophores are nearly regular, ranging
from 88 to 92°. In both chromophores, the N thiazole atoms
are in the trans position. The deviations of N1, N2, N3, and
N8 atoms from the mean plane Cu1N4 are -0.124, +0.225,
+0.218, and -0.125 Å, respectively, and the Cu1 is displaced
-0.194 Å from this plane in the opposite direction to the other
X-ray Crystal Structure Determination. A red-purple well-formed
crystal was of approximate size 0.15 × 0.10 × 0.15 mm. Mo KR
radiation monochromated with a graphite crystal was used on an Enraf-
Nonius CAD-4 single-crystal diffractometer (λ ) 0.170 73 Å). The
unit-cell dimensions were determined from the angular settings of 25
reflections with θ between 15 and 25°. The space group was P21cn.
A set of 4419 reflections was measured, in the hkl range (0, 0, 0) to
(12, 16, 35) between θ limits: 1 < θ < 25°. The ω-2θ scan technique
and a variable scan rate with a maximum scan time of 60 s/reflection
were used. The intensity of the primary beam was checked throughout
the data collection by monitoring three standard reflections every 3600
s. The final drift correction factors were between 0.97 and 1.01. On
all reflections profile analysis was performed.21,22 Some double
measured reflections were averaged: 4145 “unique” reflections from
which were 3243 with Fo > 4σ(Fo). Lorentz and polarization
corrections were applied, and the data were reduced to |Fo| values.
The structure was solved by the Patterson method using the program
SHELXS86.23 Isotropic least-squares refinement was performed by
means of SHELXS93,24 converging to R ) 0.095. At this stage
(24) Sheldrick, G. M. SHELXS-93, Program for Crystal Structure Refine-
ment; Institute fu¨r Anorganishe Chemie de Universitat: Go¨ttingen,
Germany, 1993.
(25) Walker, N.; Stuart, D. Acta Crystallogr. 1983, A39, 158.
(26) International Tables for X-ray Crystallography; Kynoch Press: Bir-
mingham, England, 1974; Vol. IV. Present distributor: Kluwer
Academic Publishers, Dordrecht, The Netherlands,
(27) Johnson, C. K. ORTEP, Report ORNL-3794; Oak Ridge National
Laboratory: Oak Ridge, TN, 1971.
(21) Lehman, M. S.; Larsen, F. K. Acta Crystallogr. 1974, A30, 580.
(22) Grant, D. F.; Gabe, E. J. J. Appl. Crystallogr. 1978, 11, 114.
(23) Sheldrick, G. M. SHELX86. In Crystallographic Computing 3;
Sheldrick, G. M., Kruger, C., Goddard, R., Eds.; Clarendon Press:
Oxford, U.K., 1985.

2054 Inorganic Chemistry, Vol. 36, No. 10, 1997
Casanova et al.
Table 2. Atomic Coordinates (×104) and Equivalent Isotropic
Displacement Parameters (Å2 × 103) for Cu2(stz)4, with Their Esd’s
in Parenthesesa
Å from the mean plane in the opposite direction to the Cu1-
N4 plane. The dihedral angle between the two CuN4 mean
planes is 0.10°, indicating a parallel arrangement. The very
short Cu1‚‚‚Cu2 distance [2.671(2) Å] appears to be primarily
determined by the width of the “bite” of the bridging ligand. It
is noteworthy that in spite of the Cu1 and Cu2 atoms being
placed below and above the Cu1-N4 and Cu2-N4 mean planes,
the metal-metal distance is very short, comparable with the
shortest Cu‚‚‚Cu distances found in copper-adenine
complexes.15-20
Tetrahedrality for any tetracoordinate copper complex can
be characterized from the angle subtended by two planes, each
encompassing the copper and two adjacent donor atoms.28 For
strictly square-planar complexes with D4h symmetry, the tetra-
hedrality is 0°. For tetrahedral complexes with D2d symmetry,
the tetrahedrality equals 90°. The values of these dihedral angles
for Cu2(stz)4 are 14 and 11°, respectively, suggesting a slight
distortion from a square-planar geometry.
atom
x
y
z
U(eq)
Cu1
Cu2
S1
S2
S3
S4
S5
S6
S7
S8
N1
N2
N3
N4
N5
N6
N7
N8
N9
N10
N11
N12
O1
O2
O3
O4
O5
-429(1)
1572(1)
-2606(2)
2660(2)
281(2)
1992(2)
-1517(2)
3350(2)
-939(3)
1494(3)
-977(6)
-1140(5)
926(5)
992(6)
2699(6)
506(6)
1619(6)
223(6)
342(11)
-3181(8)
3805(8)
696(8)
-3438(6)
-2698(5)
3201(9)
3472(7)
-1022(6)
937(6)
1392(8)
2984(8)
-2798(7)
-2275(8)
-2435(9)
-3085(8)
-3580(8)
-3458(8)
1914(7)
645(8)
2091(1)
1852(1)
1003(1)
-109(2)
3541(1)
3690(1)
723(2)
3173(2)
4396(2)
-361(1)
3119(4)
1201(4)
2934(4)
3202(4)
2360(4)
1336(4)
575(4)
1034(3)
-2720(7)
-3127(5)
7078(5)
7488(5)
1393(4)
1354(4)
-698(6)
508(6)
3247(4)
3429(4)
3994(4)
2988(4)
-217(5)
-742(5)
-1707(5)
-2165(5)
-1601(5)
-648(5)
-828(5)
-1014(5)
-1626(6)
-2069(6)
-1849(6)
-1254(5)
4726(5)
5315(5)
6229(6)
6572(5)
5977(5)
5068(5)
4696(5)
4634(6)
5406(5)
6295(5)
6345(5)
5564(5)
1128(5)
860(6)
883(1)
1418(1)
1235(1)
1300(1)
170(1)
1935(1)
2255(1)
213(1)
1868(1)
254(1)
1263(2)
1358(2)
562(2)
1596(2)
944(2)
1880(2)
1082(2)
527(2)
2602(3)
1243(3)
982(3)
759(3)
1569(2)
779(2)
966(3)
1563(3)
202(2)
-256(2)
2345(2)
1963(2)
1235(2)
884(2)
33(1)
36(1)
39(1)
62(1)
44(1)
56(1)
50(1)
51(1)
66(1)
66(1)
43(1)
38(1)
36(1)
42(1)
40(1)
39(1)
44(1)
37(1)
90(3)
63(2)
60(2)
75(2)
54(1)
50(1)
113(3)
110(3)
63(2)
59(2)
78(2)
83(2)
37(2)
46(2)
52(2)
50(2)
49(2)
51(2)
43(2)
49(2)
57(2)
56(2)
58(2)
56(2)
45(2)
49(2)
55(2)
54(2)
54(2)
47(2)
50(2)
59(2)
54(2)
46(2)
50(2)
49(2)
36(2)
61(2)
55(2)
35(1)
50(2)
46(2)
45(2)
67(3)
54(2)
41(2)
66(2)
44(2)
The sulfathiazolato anions act as a bridge linking to one Cu-
(II) ion through the Nthiazole atom and to the other Cu(II) by
means of the Nsulfonamido atom. It must be noted that the bridge
NCN formed by each of the four ligands is very unusual. The
internal geometry of the sulfathiazole ligand is normal and
agrees with numerous studies performed previously.9,12,29,30
Crystal Packing. The dimer units are stabilized by two weak
stacking interactions between the thiazole [C100-S5-C101-
C102-N6] and phenyl [C20-C21-C22-C23-C24-C25]
rings (shortest distance 2.78 Å, angle between planes 25.9°)
and between phenyl rings [C30-C31-C32-C33-C34-C35]
and [C41-C42-C43-C44-C45] (shortest distance 3.23 Å,
angle between plane 25.06°).
O6
O7
O8
C10
C11
C12
C13
C14
C15
C20
C21
C22
C23
C24
C25
C30
C31
C32
C33
C34
C35
C40
C41
C42
C43
C44
C45
C100
C101
C102
C200
C201
C202
C300
C301
C302
C400
C401
C402
890(3)
Dimer units are connected by weak hydrogen bonds involving
the Namino and the Osulfonamido atoms (see Table 4).
1238(3)
1577(2)
1580(2)
1697(2)
1682(3)
1990(3)
2319(3)
2328(3)
2025(3)
322(2)
136(3)
272(3)
614(3)
803(3)
Infrared Spectrum. Previous studies on sulfathiazole
complexes have shown that useful information on the mode of
coordination of the ligand may be obtained from an examination
of infrared spectra. The ν(N-H) vibrations at 3480 and 3390
cm-1 are shifted to higher frequencies with respect to the
equivalent ones in the uncoordinated ligand (3320 and 3280
cm-1). This shift is consistent with that observed for other
metal-sulfathiazole complexes. The bands attributed to the SO2
group that appear as sharp peaks at 1320, 1130, and 550-570
cm-1 (corresponding to νas, νs, and scissors and wagging
vibrations, respectively) remain unchanged. According to the
deprotonation and coordination of the ligand, the characteristic
band of the thiazole ring at 1540 cm-1 in the free ligand is
strongly reduced and shifted to lower frequencies (1500 cm-1).
123(9)
852(9)
2127(10)
2660(9)
375(8)
1328(7)
1386(8)
624(8)
-289(9)
-395(8)
2594(8)
3513(9)
3922(8)
3418(7)
2477(8)
2044(8)
-690(6)
-167(9)
795(9)
2212(7)
4493(8)
3993(8)
-186(8)
-2327(10)
-2178(8)
1109(7)
382(10)
-164(7)
648(2)
1656(3)
1328(3)
1112(3)
1204(2)
1531(2)
1758(2)
1789(2)
2574(3)
2333(3)
599(2)
546(3)
907(3)
1560(2)
1578(3)
1286(3)
670(2)
Electronic Spectrum. The reflectance diffuse spectrum
shows a very broad and intense band centered around 19 000
cm-1, characteristic of a CuN4 chromophore with the metal ion
in a distorted square-planar arrangement.14 A linear correlation
between the dihedral angle with the highest energy d-d
transition in the electronic spectra of some CuN4 chromophores
has been reported.31 From this correlation the maximum for a
CuN4 chromophore is at 19 960 cm-1 for a regular square-planar
geometry. Taking into account the dihedral angles (11 and 14°)
obtained by X-ray crystallography for Cu2(stz)4, the values of
1175(6)
2823(4)
2663(5)
2256(5)
3508(4)
4219(6)
3507(6)
461(5)
ν
max found from this approximation are 19 050 and 18 900 cm-1
98(7)
836(5)
-104(3)
87(2)
a U(eq) is defined as one-third of the trace of the orthogonalized Uij
tensor.
(28) Battaglia, L. P.; Bonamartini-Corradi, A.; Marcotrigiano, G.; Menabue,
L.; Pellacani, G. C. Inorg. Chem. 1979, 18, 148.
(29) Kruger, G. J.; Garner, G. Acta Crystallogr. 1971, B27, 326.
(30) Kruger, G. J.; Garner, G. Acta Crystallogr. 1972, B28, 272.
(31) Ravichandran, V.; Chacko, K. K.; Aoki, A.; Yamazaki, H.; Ruiz-
Sanchez, J.; Suarez-Verela, J.; Lopez-Gonzalez, J. D.; Salas-Peregrin,
J. M.; Colacio-Rodriguez, E. Inorg. Chim. Acta 1990, 173, 107.
Cu2-N4 plane. The deviations of N4, N5, N6, and N7 atoms
from the mean plane Cu2-N4 are -0.226, +0.154, +0.153,
and -0.227 Å, respectively, and the Cu2 is displaced +0.146

X-ray and Molecular Structure of Cu2(stz)4
Inorganic Chemistry, Vol. 36, No. 10, 1997 2055
Figure 2. ORTEP drawing of the Cu2(stz)4 molecule. Thermal ellipsoids are drawn at the 25% probability level.
Table 3. Selected Bond Lengths (Å) and Angles (deg) for
Cu2(stz)4
Cu1-N1
Cu1-N2
Cu1-Cu2
Cu2-N5
Cu2-N7
1.938(6)
2.041(5)
2.671(2)
1.982(6)
2.077(6)
Cu1-N8
Cu1-N3
Cu2-N6
Cu2-N4
1.967(5)
2.100(6)
1.922(5)
2.090(6)
N1-Cu1-N8
N8-Cu1-N2
N8-Cu1-N3
N1-Cu1-Cu2
N2-Cu1-Cu2
N6-Cu2-N5
N5-Cu2-N4
N5-Cu2-N7
N6-Cu2-Cu1
N4-Cu2-Cu1
176.0(3)
91.3(2)
87.6(2)
89.4(2)
78.6(2)
178.7(3)
91.0(2)
88.1(2)
90.4(2)
78.4(2)
N1-Cu1-N2
N1-Cu1-N3
N2-Cu1-N3
N8-Cu1-Cu2
N3-Cu1-Cu2
N6-Cu2-N4
N6-Cu2-N7
N4-Cu2-N7
N5-Cu2-Cu1
N7-Cu2-Cu1
87.6(2)
91.9(2)
157.1(2)
86.7(2)
78.5(2)
90.1(2)
91.1(2)
159.5(2)
90.5(2)
81.1(2)
for the Cu1-N4 and Cu2-N4 chromophores, respectively, which
agree well with the experimental results.
Electrochemistry. The electrochemical properties of the
dinuclear compound were studied by cyclic voltammetry in 10-3
M acetone solution containing 0.1 M NBu4PF6 as supporting
electrolyte in the potential range from +1.0 to -1.5 V vs
saturated calomel electrode (SCE) (Figure 4a).
The cyclic voltammograms involve two clearly distinguish-
able redox processes corresponding to two well-defined one-
electron reduction irreversible peaks A1 (-0.05 V) and A2
(-0.37 V) that tentatively can be assigned as follows:
CuIICuII -0.05 V8 CuIICuI -0.37 V8 CuICuI
If the sweep is reversed after peak A2 is scanned, A′2 (0.40
V) and A′1 (0.75 V) peaks are observed, so ∆Ep1 ) Ep(A′1) -
Ep(A1) and ∆Ep2 ) Ep(A′2) - Ep(A2) are 0.80 and 0.77 V,
respectively. According to that, the CuI-CuI species generated
after reduction in A2 are unstable, probably due to the dissocia-
tion of the copper-copper bond. Tentatively, we can assume
that A′2 and A′1 peaks correspond to the oxidation of mono-
nuclear species from Cu0 to CuI and from CuI to CuII,
respectively. The formation of the Cu0 is clearly observed in
the voltammogram of the title compound in MeCN (Figure 4b).
The picture shows two reduction peaks like the previous
Figure 3. Crystal packing of the Cu2(stz)4 complex showing the
stacking interactions and the hydrogen bonds.
voltammogram in acetone, and the reverse scan shows a
characteristic reoxidation stripping peak A′2 (-0.11 V) of the
Cu0 species.
EPR Measurements. The dimetallic nature of the Cu2(stz)4
complex is also indicated by EPR data. The X- and Q-band
EPR spectra taken on ground crystals at room temperature and

2056 Inorganic Chemistry, Vol. 36, No. 10, 1997
Table 4. Distances (Å) and Angles (deg) between the Atoms Involved in the Observed Hydrogen Bonds for the Cu2(stz)4 Complex
donor-H donor‚‚‚acceptor H‚‚‚acceptor donor-H‚‚‚acceptor
1.080(0.015)
Casanova et al.
N9-H2
N9‚‚‚O1a
3.046(0.012)
3.101(0.010)
3.205(0.011)
3.236(0.012)
H2‚‚‚O1a
2.104(0.012)
2.105(0.010)
2.128(0.010)
2.193(0.012)
N9-H2‚‚‚O1a
144.33(1.05)
152.14(0.72)
174.34(0.79)
161.76(0.85)
N10-H3
N11-H5
N11-H6
1.080(0.012)
1.080(0.012)
1.080(0.011)
N10‚‚‚O6b
N11‚‚‚O6c
N11‚‚‚O3d
H3‚‚‚O6b
H5‚‚‚O6c
H6‚‚‚O3d
N10-H3‚‚‚O6b
N11-H5‚‚‚O6c
N11-H6‚‚‚O3d
1
1
1
1
1
a (x + /2, y - /2, -z + /2). b (x - /2, -y, -z). c (x + /2, -y + 1, -z). d (x, y + 1, z).
Figure 6. Q-band EPR spectrum of the Cu2(stz)4 complex at 100 K.
Table 5. EPR Data for the Cu2(stz)4 Complexa
X-band
Q-band
∆ms ) (2
Hz1
Hxy1
Hz2
Hxy2
b
886
1873
4900
4180
5 595
8 860
10 715
c
Figure 4. Cyclic voltammograms in the scan range +1.0 to -1.5 V
for Cu2(stz)4 in acetone solution (a) at a scan rate of 0.15 V‚s-1 and in
acetonitrile solution (b) at a scan rate of 0.05 V‚s-1
.
13 015
a Resonance fields in gauss ()10-4 T). b Not resolved from nearby
∆ms ) (1 band. c Hidden by stronger Hxy2 band.
the case in compounds containing copper dimers, and E ) 0,
the solution of spin Hamiltonian yields four allowed (∆ms )
(1) at resonance fields Hz1, Hz2, Hxy1, and Hxy2 given by the
equations
hν ) D - g|âHz1
hν ) -D + g|âHz2
hν ) -D/2 + [D2/4 - (g âHxy1)2]1/2
hν ) -D/2 + [D2/4 + (g âHxy2)2]1/2
Figure 5. X-band EPR spectrum of the Cu2(stz)4 complex at room
temperature.
In addition, the formally forbidden transition (∆ms ) (2) the
so-called half-field spin transition, is often observed. The
position of the low-field edge of the half-field signal in a powder
spectrum is given by the equation34
100 K, respectively, are shown in Figures 5 and 6. Solution of
the spin Hamiltonian for the triplet state of dimetallic copper
complexes is given by
H ) gâBS + DSz2 + E(Sx2 - Sy2) - 2D/3
Hmin(∆ms ) 2) ) 1/2gâ[(hν)2 - 4(D2/3 + E2)]1/2
where D and E are zero-field splitting parameters. As shown
by Wasson et al.,32 two allowed transitions by the selection rule
(∆ms ) (1) will result in each principal direction, and six
resonance fields can be determined. Furthermore a transition
with (∆ms ) (2) is also obtained. The six resonance fields
are Hx1, Hy1, Hz1, Hx2, Hy2, and Hz2 whose equations are proposed
by Wasserman et al.33 When D is lower than hν, as is usually
The observed transitions are summarized in Table 5, from
these the calculated parameters are g| ) 2.33, g ) 2.09, and
D ) 0.230 cm-1. The values of Hz2 and Hxy2 must be virtually
coincident at the Q-band frequency, and consequently the low-
intensity Hz2 is hidden by the much stronger Hxy2 band.
Changing the microwave frequency to X-band enables a clear
separation of these transitions to be made (Figure 5). Con-
(32) Wasson, J. R.; Chin-I Shyr; Trapp, C. Inorg. Chem. 1968, 7, 469.
(33) Wasserman, E.; Snyder, L. C.; Yager, W. A. J. Chem. Phys. 1964,
41, 1763.
(34) Eaton, S. A.; More, K. M.; Sawant, B. M.; Eaton, G. R. J. Am. Chem.
Soc. 1983, 105, 6550.

X-ray and Molecular Structure of Cu2(stz)4
Inorganic Chemistry, Vol. 36, No. 10, 1997 2057
Table 6. Magnetic Exchange Parameters and Distances of Copper
Dimers
2J (cm-1
)
Cu‚‚‚Cu (Å)
[Cu2(Aden)4(H2O)2]‚6H2Oa
[Cu2(Aden)4]‚4H2Ob
-179
-257
-285
-389d
-61.5
2.75
2.95
2.81
[Cu2(Aden)4(Cl)2]Cl2‚6H2Oa
[Cu2(7-Azaindole)4‚4DMSOc
[Cu2(stz)4]e
2.67
a Reference 20. b Reference 15. c Reference 19. d The authors indi-
cated that they are unable to distinguish whether both of the dimethyl
sulfoxide (DMSO) molecules per copper are contained within the lattice
or whether one of these is involved in axial coordination to copper.
e This work.
Bowers equation for the exchange coupled copper(II) dimers:
Figure 7. Plot of molar magnetic susceptibility (0) and øMT (9) for
the Cu2(stz)4 compound as a function of temperature. The solid line
represents the best fit to the data.
øM ) Ng2â2{kT[3 + exp(-2J/kT)]}-1(1 - F) +
FNg2â2/(4kT) + NR
which results from a consideration of the eigenvalues of H )
-2JS1S2, the Heisenberg exchange Hamiltonian for two inter-
acting S ) 1/2 centers [øM is the molar magnetic susceptibility,
NR is the temperature independent paramagnetism (60 × 10-6
emu/copper), and F is the fraction of monomeric impurity]. A
nonlinear regression analysis of the data was carried out with F
as a floating parameter. The best fit line shown in Figure 7 is
for g ) 2.19, -2J ) 61.53 cm-1, F ) 0.118, and R ) [∑(øobs
versely, the Hz1, Hxy1, and ∆ms ) (2 transitions could not have
been unambiguously assigned from the X-band results alone
because of appreciable band overlap at this frequency. How-
ever, these transitions are clearly resolved and readily assigned
at Q-band frequency.
The D value observed ()0.230 cm-1) is slightly higher than
the range of D values reported for copper(II) dimers with
nitrogen donor atoms in the bridges.35
- øcalc)2/∑(øobs)2]1/2 ) 6.9 × 10-3
.
In order to obtain more information about the monomeric
copper(II) complex also present in a small amount in the sample,
the purple complex was obtained in solid state doped in the
diamagnetic matrix of the complex [Zn(stz)2]‚H2O. The EPR
of the magnetically diluted complexes of copper(II) (Cu:Zn )
1:4 and 1:9) show clearly the anisotropic spectra of a monomeric
copper(II) complex with an axial symmetry and the signals of
the dimer weaker with respect to those observed in the EPR of
the Cu2(stz)4 complex. The calculated EPR parameters of the
Table 6 shows the magnetic exchange parameters in copper-
(II) complexes which exhibit pairwise exchange. From the table
we can observe that the values of -2J for the copper adenine
complexes are higher than that for the Cu2(stz)4 complex, which
is noteworthy because the Cu‚‚‚Cu bond distance in this complex
is the lowest. This is probably due to the effect of the axial
ligand in the other complexes which affects the relative
2
2
2
separation between the dx -y and dz orbitals of the two copper-
(II) ions.19 Sonnenfroh and Kreilick20 found for the copper-
adenine complexes that an increase in the axial bond distance
signifies a decrease in the exchange energy value. In the Cu2-
(stz)4 complex there is not an axial ligand as a consequence
because the value of -2J is the lowest.
monomer are g| ) 2.35, g ) 2.07, and A| ) 132 × 10-4 cm-1
,
suggesting a tetrahedral arrangement around the metal ion.36
As the crystal structure of the [Zn(stz)2]‚H2O was previously
determined, we know that the Zn(II) ions is tetrahedrally
surrounded by two Nthiazole and two Namino, so we suggest that
the Cu(II) in the doped powder presents a coordination
polyhedron similar to that of the Zn(II) and not the planar CuN4
found in the dimer copper complex.
Molecular Orbital Calculations. In order to understand the
reasons of the different values of the exchange energies between
the copper-adenine complexes and the Cu2(stz)4 complex, we
have performed molecular orbital calculations.
Solutions of the Cu2(stz)4 in MeOH-ethylene glycol (1:1)
that have the same electronic spectrum as the reflectance one
of Cu2(stz)4 exhibit an EPR spectrum at room temperature with
an isotropic signal at 3230.4 G, giving a g0 value of 2.16 and
a half-signal at 1502.9 G, suggesting the existence of the dimer
in this solution.
Magnetic Measurements. Susceptibility values per mole
of dimer as a function of the temperature have been plotted in
Figure 7. The dominant features of the data are a maximum
near 50 K and a rapid decrease to zero at lower temperatures.
An increase in susceptibility at the lowest temperatures of the
type seen here is often observed in antiferromagnetically coupled
copper complexes and has been accounted for by the presence
of small amounts of paramagnetic impurities.37 The solid curve
in Figure 7 is the best fit of the data to the modified Bleaney-
1
It is well-known that the interaction between two S ) /2
metal atoms in a dimer leads to two molecular states: a spin
singlet (S ) 0) and a triplet (S ) 1) separated by 2J. The
interaction will be antiferromagnetic (J < 0) if S ) 0 is the
ground state; on the contrary if S ) 1, the interaction will be
ferromagnetic (J > 0). Whereas ferromagnetic contributions
are usually small, antiferromagnetic ones can be considered as
proportional to the square of the gap between the molecular
orbitals constructed from the magnetic orbitals.38 Extended
Hu¨ckel molecular orbital (EHMO) calculations (atomic param-
eters used are given in Table 7) can be employed to evaluate
such interactions. The EHMO calculations performed in this
work have been made, by means of the CACAO program.39
For a copper(II) dimer with one unpaired electron on each metal
atom two molecular orbitals are generated from the combination
of magnetic orbitals (symmetric and antisymmetric forms).
(35) Goodgame, D. M. L.; WAggett, S. V. Biochem. Biophys. Res.
Commun. 1971, 42, 63.
(36) Bertini, I. ESR and NMR of Paramagnetic Species in Biological and
Related Systems; Bertini, I., Drago, R. S., Eds.; Reidel: Dordrecht,
The Netherlands, 1980.
(37) Otieno, T.; Rettig, S. J.; Thompson, R. C.; Trotter, J. Inorg. Chem.
1993, 32, 4384.
(38) Kahn, O. Angew. Chem., Int. Ed. Engl. 1985, 24, 834.
(39) Mealli, C.; Proserpio, D. Computer Aided Composition of Atomic
Orbitals (CACAO program) PC version, July 1992. See also: J.
Chem. Educ. 1990, 67, 399.

2058 Inorganic Chemistry, Vol. 36, No. 10, 1997
Casanova et al.
Table 7. Orbital Exponents (Contraction Coefficients in Double-ú
Expansion Given in Parentheses) and Energies Used in the
Extended Hu¨ckel Calculations
atom
O
orbital
êi(ci)
2.275
2.275
1.950
1.950
1.625
1.625
1.300
1.900
1.900
2.200
2.200
5.950(0.5933)
2.300(0.5744)
Hii (eV)
2s
2p
2s
2p
2s
2p
1s
3s
3p
4s
4p
3d
-32.30
-14.80
-26.00
-13.40
-21.40
-11.40
-13.60
-20.30
-14.00
-11.40
-6.06
N
C
H
S
Cu
-14.00
Figure 9. Idealized form of the copper-adenine complexes.
2
exchange interaction because the dz orbital overlaps more
2
2
efficiently and must be mixed with the dx -y orbital. These
results are in agreement with that reported by Sonnenfroh and
Kreilick. However, the axial bond distances observed in the
copper-adenine complexes (from 2.20 to 2.43 Å) are rather
2
2
long; as a consequence we suggest that the mixture of the dx -y
2
and dz orbitals must be small and that it is not the only factor
that contributes to the differences of -2J values.
We have studied the influence of the bridging ligand nature,
proposing another model for copper complexes with a NCN
bridge (see Figure 9), where the selected Cu‚‚‚Cu bond distance
was 3.00 Å, which has been reported for one of the copper-
adenine complexes.19 The EH calculation program indicates
that the SOMO’s of the copper-adenine model has a difference
of energy of 212 meV, while, for the copper-sulfathiazole one,
the value is 104 meV. These results are according to the
experimental values of the exchange energies observed in the
Table 6. The difference could be explained from the most
favorable electron transfer that takes place through the aromatic
rings in the copper-adenine complexes.15
Figure 8. Idealized form of the Cu2(stz)4 complex.
2
2
2
Table 8. Energy of the dx -y and dz Orbitals of the Cu2(stz)4
Model from the Variation of the Cu-L Axial Bond Distance
2
2
2
Cu-L bond distance
E(dx -y
)
E(dz )
2.10
2.20
2.30
2.40
2.50
2.60
-10.664
-10.664
-10.664
-10.664
-10.664
-10.664
-10.664
-12.419
-12.550
-12.589
-12.887
-13.059
-13.087
-13.152
Acknowledgment. We thank Drs. F. Lloret (University of
Valencia) and T. Rojo and L. Lezama (University of Pais Vasco)
for their assistance in the magnetic and EPR measurements. We
greatly appreciate financial support from the Comisio´n Inter-
ministerial de Ciencia y Tecnolog´ıa (SAF 94-075). J.C. thanks
the Generalitat Valenciana (Spain) for a Doctoral fellowship.
With the aim of understanding the influence of the axial
ligand, we calculate the variation of the energy of dx -y and dz
2
2
2
Supporting Information Available: Tables listing crystal data and
structure refinement, atomic coordinates and equivalent isotropic
displacement parameters, bond lengths and angles, hydrogen bonds,
anisotropic displacement parameters, calculated hydrogen coordinates,
isotropic displacement parameters, and torsion angles (14 pages).
Ordering information is given on any current masthead page.
orbitals with respect to the Cu-L axial bond distance in one
model for the Cu2(stz)4 complex being the Cu‚‚‚Cu bond
distance of 2.60 Å (see Figure 8). Table 8 shows the results of
the calculation, where one can appreciate that the energy
differences diminish when the Cu-L axial bond distance is
shorted. These variations will affect the magnitude of the
IC960968P