Studies of pyridinyl-containing 14-membered macrocyclic copper(II) complexes

Article abstract of DOI:10.1002/ejic.200390182

Six copper(II) complexes of tetracoordinating, pyridinyl-containing 14-membered macrocycles with varying ratios of nitrogen and oxygen donor atoms were prepared and characterized by IR, UV/Vis, and EPR spectroscopy and cyclic voltammetry. A distorted tetragonal coordination of the copper center in the solid-state was established by X-ray crystallography for the tetraazamacrocyclic complex Cu-3 carrying a methoxybenzyl pendent arm and the trioxaaza complex Cu-6. The superoxide dismutase-like activity of the Cu^II complexes was investigated by inhibition of NADH oxidation. Although the UV/Vis and EPR spectra of the complexes were strongly affected when the coordinating nitrogen atoms were successively replaced by oxygen atoms, no significant change in their reactivity towards superoxide was observed. In all cases a 1:1 or 1:2 stoichiometry for the reaction with superoxide was found, with the exception of the methoxybenzyl-substituted tetraaza complex, which showed a low catalytic activity with a turnover number of about 10. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200390182

FULL PAPER
Studies of Pyridinyl-Containing 14-Membered Macrocyclic Copper(II
Complexes
)
Sabrina Autzen,[a] Hans-Gert Korth,[a] Roland Boese,[b] Herbert de Groot,*[c] and
Reiner Sustmann*[a]
Keywords: Macrocyclic ligands / Copper / Enzyme models / Superoxide dismutase
Six copper(II) complexes of tetracoordinating, pyridinyl-con-
taining 14-membered macrocycles with varying ratios of
nitrogen and oxygen donor atoms were prepared and charac-
though the UV/Vis and EPR spectra of the complexes were
strongly affected when the coordinating nitrogen atoms were
successively replaced by oxygen atoms, no significant
terized by IR, UV/Vis, and EPR spectroscopy and cyclic vol- change in their reactivity towards superoxide was observed.
tammetry. A distorted tetragonal coordination of the copper In all cases a 1:1 or 1:2 stoichiometry for the reaction with
center in the solid-state was established by X-ray crystallo- superoxide was found, with the exception of the methoxy-
graphy for the tetraazamacrocyclic complex Cu-3 carrying a
methoxybenzyl pendent arm and the trioxaaza complex Cu-
6. The superoxide dismutase-like activity of the CuII com-
plexes was investigated by inhibition of NADH oxidation. Al-
benzyl-substituted tetraaza complex, which showed a low
catalytic activity with a turnover number of about 10.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Numerous transition metal (preferably Cu, Fe, Mn) com-
plexes of various kinds have been investigated with regard
to their reactivity toward superoxide. The use of copper
complexes as SOD mimics has been reviewed.[3,4] Although
several investigated compounds exhibit good SOD-like ac-
tivities, systematic studies concerning the factors which gov-
ern the enzyme-mimicking reactivities are scarce.[5] We re-
port here on the copper() complexes Cu-1ϪCu-6 of li-
gands 1Ϫ6 in which the sp3-nitrogen atoms of the parent
macrocycle 1 are successively replaced by oxygen atoms.
Spectroscopic properties were compared to determine the
influence of the coordinating atoms on the SOD-like activ-
ities of the complexes. Complexes Cu-1 and Cu-2 have pre-
viously been described,[6Ϫ9] but their SOD activity was not
investigated. Complex Cu-3, carrying a methoxybenzyl pen-
dent arm, was included in this study to determine whether
a possible coordination by the phenolic oxygen atom to the
fifth coordination site at the copper atom might influence
the catalytic activity. Coordination of phenolate groups to
the fifth position in square-planar CuII complexes has been
reported before;[10] however, to the best of our knowledge,
coordination of phenyl ether substituents to the axial posi-
tion has not been reported. There is, however, a recent re-
port[11] in which phenyl ether moieties within the chain of
a 20-membered macrocyclic ligand bind to the axial posi-
tion of a tetra-N-coordinated copper() complex. X-ray
crystal structures of the copper() complexes Cu-3 and Cu-
6 are presented. X-ray structures of eight 14-membered
Introduction
Our present studies are concerned with the preparation
of transition metal complexes of 14-membered, pyridine-
containing macrocycles and the examination of their spec-
troscopic properties to elucidate their potential enzyme-mi-
metic qualities. We have recently reported the catalase-mi-
metic activity of some non-heme iron() complexes.[1] Cop-
per is another biologically significant metal, being found in
various enzymes, such as the important Cu-Zn superoxide
dismutase (Cu2Zn2-SOD), which is used in living organisms
to detoxify the superoxide radical anion, O2·Ϫ, [Equa-
tion (1)].[2]
(1)
[a]
Institut für Organische Chemie, Universität Essen,
Universitätsstr. 5, 45117 Essen, Germany
Fax: (internat.) ϩ 49-201/183-4259
E-mail: reiner.sustmann@uni-essen.de
[b]
Institut für Anorganische Chemie, Universität Essen,
Universitätsstr. 5, 45117 Essen, Germany
Institut für Physiologische Chemie, Universitätsklinikum Essen,
[c]
Hufelandstr. 55, 45122 Essen, Germany
Fax: (internat.) ϩ 49-201/723-5943
E-mail: h.de.groot@uni-essen.de
Ϫ
endo-pyridine CuII complexes, including the PF6 salt of
1401
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ1948/03/0407Ϫ1401 $ 20.00ϩ.50/0

S. Autzen, H.-G. Korth, R. Boese, H. de Groot, R. Sustmann
FULL PAPER
Cu-1,[9] can be found in the Cambridge Structural Data-
Facile formation of the copper() complexes was
achieved by addition of an equimolar ethanolic solution of
copper() perchlorate to an ethanolic solution of the li-
gands at room temperature and subsequent crystallization.
As expected, for the ligands 1, 2, and 4Ϫ6 the correspond-
ing copper complexes Cu-1, Cu-2, and Cu-4ϪCu-6
(Scheme 2) with two perchlorate counterions were ob-
tained.
base.[12]
Results and Discussion
Synthesis
Macrocycle 2 was prepared from its nickel complex as
described by Karn and Busch.[13] Compound 5 was pre-
pared from ditosylated pyridine-2,6-diyldimethanol and
the alkanediol, compound 6[14] from pyridine-2,6-diyldi-
methanol and the appropriate alkanediyl ditosylate. Macro-
cycles 1[6] and 3[1] were prepared according to literature pro-
cedures in a copper-template reaction. The preparation of
ligands 1, 3 and 4 was devised starting from pyridine-2,6-
dicarbaldehyde, which was treated with the appropriate di-
amine in a copper-template reaction to give the macrocycle.
Diimine reduction by sodium borohydride and removal of
copper() as sulfide provided the free ligands. Unfortu-
nately, attempts to prepare the macrocycles 7 and 8, in
which the position of the nitrogen and oxygen atoms was
exchanged with respect to compounds 4 and 5, respectively,
were not successful (Scheme 1).
Scheme 2
Unexpectedly, X-ray structural and elemental analysis re-
vealed that the copper complex Cu-3 of ligand 3 contained
one chloride ion as a counterion, although no chloride was
involved in any steps of the synthesis. Thus, perchlorate
must have been reduced during complex formation. Al-
though perchlorate (which is a weakly chelating, relatively
large anion) is a strong oxidant it is generally kinetically
stable to reductants and is therefore frequently used to
regulate the ion strength in kinetic and thermodynamic
studies.[15] Because perchlorate reduction was not observed
in the preparation of the related complexes Cu-1 and Cu-2,
it is very likely that the methoxybenzyl side chain of 3 facil-
itated this reduction. This conclusion is supported by the
rapid formation of a CuIIϩ3 complex of 1:2 stoichiometry,
[CuII(3)2](ClO4)2, after mixing of copper() perchlorate and
3 (see Exp. Sect.). When kept in the reaction mixture, this
complex slowly (within 12 h) converted into complex Cu-3.
We propose that the initial 1:2 complex may have been
formed by a kinetically controlled, exocyclic complexation,
probably involving the oxygen atom of the methoxybenzyl
Scheme 1
1402
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410

Pyridinyl-Containing 14-Membered Macrocyclic Copper(II) Complexes
FULL PAPER
substituent. In solution, this exocyclic complex then re- 3 copper() has a fivefold coordination environment. The
arranges to the thermodynamically more stable complex four in-plane distances are of lengths commonly found in
Cu-3, thereby re-releasing one ligand molecule. Concomit-
antly, partial reduction of perchlorate to chloride takes
place. Given the rather mild reaction conditions used here
it is not clear how the perchlorate might have been reduced,
i.e. which chemical entity would have been the reductant.
The copper() ion is unlikely to be oxidized under the ap-
plied conditions (compare the inertness of Cu-1, Cu-2),
whereas the electron-rich phenyl group of the side arm
might be a chemically reasonable target. A reduction of per-
chlorate to chloride during the preparation of a macrocyclic
iron complex has been described by Goedken et al.[16] who
discussed the formation of an unspecified, yet very reactive,
intermediate. Elucidation of the mechanism of this interest-
ing reaction is beyond the scope of this paper and will be
the subject of a separate study.
square-based pyramidal structures, with the CuϪNpyr dis-
˚
tance (1.94 A) being somewhat shorter than the other CϪN
˚
distances (2.02Ϫ2.05 A). This bond-length pattern seems to
be characteristic of complexes of tetraazamacrocyclic li-
gands containing a pyridine moiety (see refs.[1,9] and refer-
ences cited therein). The copper() ion in Cu-3 is lifted out
of the plane of the four coordinating ligand atoms by ρ(Cu-
II
˚
˚
3) ϭ 0.311 A with a Cu ϪCl distance of 2.564 A. The tetra-
˚
hedral distortion of 0.311 A is significantly increased com-
[9]
˚
pared with that in Cu-1·(PF6)2 (ρ ϭ 0.133 A), which may
be related to the coordination of ClϪ. The methoxybenzyl
group, C(15)ϪC(21) and O(1), with the riding hydrogen
atoms and the methylene hydrogen atoms H(14A/14B) is
disordered over two sites with occupancies of 0.6 and 0.4
together; O(31) and C(32) of the methoxy group are disor-
dered over two sites with occupancies of 0.5. In the crystal,
the oxygen atom of the 2-methoxybenzyl group is not co-
ordinated to the central ion. The second counterion is es-
sentially noncoordinating perchlorate. Unlike Cu-3, in Cu-
6 the copper atom is essentially located in-plane [ρ(Cu-6) ϭ
Crystal Structure
Ϫ
The crystal structure of Cu-1 with two PF6 counterions
instead of perchlorate has been reported.[9] The crystal
structures and the unit-cell diagrams of Cu-3 and Cu-6 are
displayed in Figures 1 and 2. Selected bond lengths and
˚
˚
0.003 A], with a CuϪN distance of 1.89 A and CuϪO
˚
angles are listed in the captions. It can be seen that in Cu- lengths of 1.93Ϫ1.96 A. One of the perchlorate anions is
Table 1. Crystallographic data of Cu-3 and Cu-6
Compound
Cu-3
Cu-6
Empirical formula
Formula mass
C21H30ClCuN4O·ClO4·0.5[CH4O]
568.95
203(2)
C13H19CuNO3·2[ClO4]
499.73
Temperature [K]
203(2)
0.71073
monoclinic
P21
9.739(2)
7.1341(17)
13.785(3)
˚
Wavelength [A]
0.71073
triclinic
Crystal system
Space group
¯
P1
˚
a [A]
6.8494(12)
10.764(2)
17.819(3)
˚
b [A]
˚
c [A]
α [°]
β [°]
γ [°]
87.336(4)
79.461(4)
73.991(4)
1241.4 (4)
90
105.772(4)
90
921.7(4)
3
˚
Volume [A ]
Z
2
2
Density (calculated) [g cmϪ3
]
1.522
1.137
592
1.801
1.535
510
Absorption coefficient [mmϪ1
F(000)
]
Crystal size [mm]
Crystal color
Crystal description
θ range of data collection [°]
Index range
0.28 ϫ 0.14 ϫ 0.12
blue
rod
1.97Ϫ28.33
Ϫ9 Յ h Յ 6
0.28 ϫ 0.24 ϫ 0.05
blue
plate
2.17Ϫ28.40
Ϫ12 Յ h Յ 13
Ϫ9 Յ k Յ 9
Ϫ6 Յ k Յ 14
Ϫ14 Յ l Յ 23
10779
Ϫ18 Յ l Յ 18
11593
Reflections collected
Independent reflections
Absorption correction
Max./min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I Ͼ 2σ(I)]
R indices (all data)
4805 [Rint ϭ 0.0627]
Siemens SADABS program multi-scan
1.00/0.69
4565 [Rint ϭ 0.0338]
Bruker AXS Saint program Vers. 6.0
Ϫ/Ϫ
full-matrix least-squares on F2
2970/0/364
full-matrix least-squares on F2
3533/1/263
0.933
0.985
R1 ϭ 0.0571, wR2 ϭ 0.1332
R1 ϭ 0.0928, wR2 ϭ 0.1515
0.811/Ϫ0.440
R1 ϭ 0.0606, wR2 ϭ 0.1574
R1 ϭ 0.0774, wR2 ϭ 0.1770
0.993/Ϫ0.577
Ϫ3
˚
Largest diff. peak/hole [e·A
]
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410
1403

S. Autzen, H.-G. Korth, R. Boese, H. de Groot, R. Sustmann
FULL PAPER
Figure 2. Molecular structure of Cu-6 in the crystal; selected bond
Figure 1. Molecular structure of Cu-3 in the crystal; selected bond
˚
˚
lengths [A] and angles [°]: Cu1ϪN1 1.885, Cu1ϪO1 1.955,
lengths [A] and angles [°]: Cu1ϪCl1 2.564, Cu1ϪN1 1.938,
Cu1ϪO2 1.931, Cu1ϪO3 1.964, Cu1ϪO4 2.425; O1ϪCu1ϪN1
81.5, O2ϪCu1ϪN1 178.7, O3ϪCu1ϪN1 81.6, O4ϪCu1ϪN1 89.6,
O2ϪCu1ϪO1 98.1, O3ϪCu1ϪO1 163.0, O4ϪCu1ϪO1 91.9,
O3ϪCu1ϪO2 98.8, O4ϪCu1ϪO2 91.6, O4ϪCu1ϪO3 86.9
Cu1ϪN2 2.023, Cu1ϪN3 2.026, Cu1ϪN4 2.054; N1ϪCu1ϪCl1
92.8, N2ϪCu1ϪCl1 100.8, N3ϪCu1ϪCl1 101.1, N4ϪCu1ϪCl1
97.4, N2ϪCu1ϪN1 81.9, N3ϪCu1ϪN1 166.1, N4ϪCu1ϪN1 81.0,
N3ϪCu1ϪN2 95.6, N4ϪCu1ϪN2 155.6, N4ϪCu1ϪN3 96.7
coordinated to two CuII centers as a bridging ligand with Table 2. Characteristic infrared frequencies [cmϪ1] of complexes
II
˚
Cu-1ϪCu-6; frequencies of the free ligands 1Ϫ6 are given in paren-
theses; the most intense lines are in italics
Cu ϪO distances of 2.425 and 2.433 A, respectively, in a
˚
range (2.2Ϫ2.9 A) commonly observed for axial CuϪO
bond lengths.[17] Thus, the coordination geometry of Cu-
6 is essentially tetragonal-octahedral. Oxygen atom O(5) is
disordered over two sites with occupancies of 0.68Ϫ0.32. In
both complexes, the pyridine ring adopts an offset face-to-
face orientation.
Ϫ
Complex ν˜(NϪH)
ν˜(CϭN)
ν˜(ClO4 )
Cu-1
Cu-2
Cu-3
Cu-4
Cu-5
Cu-6
3250, 3221
1609, 1583
1098, 1068, 624
(3242, 3220) (1588, 1576)
3240, 3223 1605, 1582
(3247, 3218) (1589, 1571)
3225, 3163
(3378)
3243
1122, 1064, 624
Infrared Spectra
1600, 1583
(1598, 1560)
1608, 1585
(1593, 1576)
1607, 1585
1107, 1054, 623
The spectra of the free ligands 1Ϫ5 show sharp, medium-
intensity bands at 3250Ϫ3300 cmϪ1 for the NϪH valence
vibrations. All ligands showed two bands around 1590
cmϪ1 for the CϭN valence vibration of the pyridine ring
(Table 2). On CuII complexation, the NϪH valence bands
are shifted by about 15Ϫ150 cmϪ1 to lower frequencies. For
Cu-1 and Cu-2 two NϪH valence vibrations are observed,
1108, 1066,[a] 624
1147, 1111, 1091,[a] 627
1085, 1061, 1038,[a] 621
(3310)
3354
(3369, 3288) (1594, 1578)
Ϫ
Ϫ
1608, 1585
(1590, 1574)
[a]
Assignment ambiguous due to superposition with ν˜(CϪO).
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410
1404

Pyridinyl-Containing 14-Membered Macrocyclic Copper(II) Complexes
FULL PAPER
in line with the different chemical environment of the NH hitherto unknown ligands 7 and 8 should have absorption
groups located in cis and trans positions with respect to the maxima around 700 and 740 nm, respectively.
pyridine nitrogen atom. As the higher wavenumber vibra-
tion is missing in the spectra of Cu-3 and Cu-4 but found
in Cu-5 (for which the lower wavenumber is absent), we
assign this band to the trans-oriented NH group. In the
Table 3. Visible electronic absorptions of CuII complexes in aque-
ous solution (298 K)
complexes, the pyridine ν(CϭN) bands are uniformly Complex
λmax [nm]
lgε [Ϫ1cmϪ1
]
shifted to higher frequencies by 10Ϫ20 cmϪ1, but virtually
Cu-1
565[a]
560[b]
563
603
757
2.56
2.29
2.13
2.18
1.42
1.36
unaffected by the variation of the coordinating ring atoms.
Cu-2
The typical ν3 (antisymmetric stretch) and ν4 (antisym-
Cu-3
metric bent) modes of noncoordinated perchlorate give rise
Cu-4
to single, strong absorptions. In complexes Cu-1ϪCu-6 the Cu-5
perchlorate absorption at about 1100 cmϪ1 (ν3) is split into
Cu-6
793
730
Cu-EDTA[19]
two or three components, whereas the sharp strong band at
620 cmϪ1 (ν4) remained a singlet. Thus, in the solid state,
the symmetry of the perchlorate ions in these complexes is
lowered.[7] However, it is not certain whether this is the re-
sult of coordination or of crystal packing effects. The latter
appears to be more likely because the X-ray structure of
Cu-3 revealed that here the perchlorate is not coordinated
to CuII.
Cu(ClO4)2
820
680
1.04
2.1
Cu2Zn2SOD[20]
[a]
[b]
Ref.[21] 544 (2.24) nm, in nitromethane. Ref.[7] 560 (2.25) nm,
in methanol; ref.[8] 560 (1.87) nm, in 0.1  NaNO3.
Electron Paramagnetic Resonance
To check whether the characteristics of the X-ray struc-
tures, the UV/Vis spectra, and/or the SOD-like activities
(see below) of our Cu complexes are reflected by their EPR
spectral features, the EPR spectra of Cu-1ϪCu-6 were ex-
amined in liquid (Figure 3, Table 4) and frozen (77 K) aque-
Electronic Absorption
Complexes Cu-1ϪCu-6 show broad electronic absorp-
tions in the 550Ϫ800-nm region (Table 3), typical for dϪd
transitions of CuII in a weak tetragonal field.[24] The ab-
sorption maxima show a red-shift (i.e. decreasing ligand-
field strength) with increasing number of coordinating oxy-
gen atoms, ranging from λmax ϭ 560 nm for the tetra-N-
substituted complexes Cu-1ϪCu-3 to 793 nm for the tris-O
complex Cu-6. This red-shift is accompanied by a similar
decrease of the molar extinction coefficient ε. With respect
to the parent complex Cu-1, the methyl and methoxybenzyl
substituents in Cu-2 and Cu-3, respectively, have virtually
no effect on the absorption maxima but appear to decrease
slightly the molar extinction. However, the red-shift of the
absorption maxima is markedly different for the various po-
sitions with respect to the pyridine ring. Replacement of the
nitrogen atom trans to the pyridine ring by an oxygen atom,
viz. Cu-1ϪCu-3 Ǟ Cu-4 and Cu-5 Ǟ Cu-6, shifts the ab-
sorption maximum by ∆λ ഠ 40 nmϪ1 toward longer wave-
lengths, whereas replacement of the cis nitrogen atoms, viz.
Cu-1ϪCu-3 Ǟ Cu-5 and Cu-4 Ǟ Cu-6, induces an approx-
imate fivefold stronger shift of ∆λ ഠ 190Ϫ200 nmϪ1. This
discontinuity in the λmax values on cis-NǞO replacement is
paralleled by a decrease of the extinction coefficient ε and
a ‘‘jump’’ in the g values of the EPR spectra (see below),
both indicating a significant change of the coordination
geometry from a distorted tetragonal (square-based pyram-
idal) to a more octahedral tetragonal structure. In line with
the stronger effect exerted by a cis-oxygen atom, the CuII-
EDTA complex, having a cis-2N-2O coordination arrange-
ment, shows a λmax value of ca. 730 nm.[19] For the
observed dependencies of the dϪd transitions on the
Figure 3. EPR spectra of the copper complexes Cu-1ϪCu-6 in
aqueous solution (298 K)
Table 4. EPR parameters of CuII complexes in aqueous solution
(298 K)
Complex[a]
gapp
Aiso [mT]
Cu-1
2.101
2.099
2.102
2.112
2.176
2.197
2.134
7.80
8.05
8.32
8.18
ca. 4.9
n.r.[b]
n.r.
Cu-2
Cu-3
Cu-4
Cu-5
Cu-6
Cu-EDTA
[a]
Instrument settings: scan range 150 mT, modulation amplitude
0.5 mT, modulation frequency 100 kHz, microwave power 20 mW,
ligand substitution pattern, possible CuII complexes of the microwave frequency 9.77Ϫ9.79 GHz. Not resolved.
[b]
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410
1405

S. Autzen, H.-G. Korth, R. Boese, H. de Groot, R. Sustmann
FULL PAPER
strikingly parallel the above-mentioned shifts of the UV/Vis
absorption maxima. In addition, the apparent g factor of
CuII-EDTA,[23] which has a 2N-2O coordination pattern
similar to the arrangement in the macrocycle 8, lies in
between the g factors of Cu-1ϪCu-4 and Cu-5 and Cu-6.
The small spectral features between 330 and 370 mT in the
EPR spectrum of Cu-6 are probably due to a CuII-con-
taining impurity. Because these signals could not be dimin-
ished by repeated recrystallization, they might belong to a
copper complex formed from, and in equilibrium with, Cu-
6 in the aqueous phase. The uncommonly low g value of
about 1.951 [a(63/65Cu) ഠ 8.9 mT] of this feature would in-
dicate a dz2 ground state (gЌ Ͼ g|| ഠ 2)[18,22] for this species.
In frozen (77 K) aqueous solution (Figure 4) a 63/65Cu
hyperfine splitting can no longer be resolved in any of the
spectra. The spectra of Cu-1ϪCu-3, Cu-5, and Cu-6 all con-
sist of a single, broad line with apparent g values as given
in Table 5. The spectral resolution could not be improved
by tenfold dilution and lower microwave power and modu-
lation settings, in agreement with observations by Costa
and Delgado[8] who noted for Cu-1 and related complexes
the loss of resolution in solutions of low ionic strength
(Table 5, footnote[b]). The anisotropic parameters (Table 5)
were evaluated by spectral simulation and are approximate
values because of the low resolution. Nevertheless, they are
typical for CuII in a distorted tetragonal environment. As
was observed for liquid aqueous solution, the apparent g
factors are shifted to higher values, from 2.085 to about
2.140, with increasing number of oxygen atoms in the mac-
rocyclic ring. However, two distinct differences compared
with the liquid-phase spectra are apparent. First, the spec-
tra of Cu-5 and Cu-6 now show approximately the same
apparent g factor (ca. 2.134Ϫ9), similar to that of Cu-
EDTA; second, the spectrum of Cu-4 displays a clear sep-
aration of the parallel and perpendicular components of the
g tensor and also has a smaller linewidth than the other
spectra. We have no straightforward explanation for this
Figure 4. EPR spectra of the copper complexes Cu-1ϪCu-6 in
frozen aqueous solution (77 K)
Table 5. EPR parameters of Cu complexes in frozen aqueous solu-
tion (77 K)
Complex[a]
gapp
gЌ
g||
A|| [mT]
Cu-1
2.090
2.085
2.082
Ϫ
2.139
2.134
2.136
Ϫ
2.086[b]
2.084[c]
2.082
2.056
Ϫ
2.128
Ϫ
2.073
2.107
2.105
2.200
2.209
Ϫ
2.180
Ϫ
2.267
Յ 5.5
Յ 5.5
17.4
Յ 1.5
Ϫ
8.4
Ϫ
13.3
Cu-2
Cu-3
Cu-4
Cu-5
Cu-6
Cu-EDTA
Cu2Zn2SOD[24]
[a]
Instrument settings: scan range 150 mT, modulation amplitude
0.5 mT, modulation frequency 100 kHz, microwave power 20 mW,
[b]
microwave frequency 9.52Ϫ9.55 GHz.
Ref.[8] g1 ϭ 2.040, g2
ϭ
2.110, g3 ϭ 2.178, A1 ϭ 4.8, A2 ϭ 9.8, A3 ϭ 199.3 cmϪ1; in 1 
[c]
[7]
NaClO4. Ref.
gЌ ϭ 2.052, g|| ϭ 2.19; polycrystalline sample,
298 K.
ous solutions (Figure 4, Table 5). The spectra show the ex- unique feature of Cu-4. As the spectrum of Cu-4 is strik-
pected characteristics of distorted (axially elongated) tetra- ingly similar to those observed for polycrystalline samples
gonal coordination of CuII (S ϭ1/2) with g|| Ͼ gЌ,[18,22] thus of Cu-2 and related complexes at ambient temperature,[4] it
indicating a dx Ϫy ground state. EPR data for CuII-EDTA might indicate that the coordination sphere in frozen water
and Cu2Zn2-SOD are given for comparison. In liquid solu- is very similar to that in the solid state.
tion, Cu-1ϪCu-3 show virtually identical, nearly isotropic
2
2
EPR spectra with a well-resolved 63/65Cu hyperfine splitting
in the range of 8 mT (Figure 3), confirming that the addi-
Electrochemical Properties
Cyclic voltammograms of complexes Cu-1ϪCu-4 and Cu-
tional ring substituents in Cu-2 and Cu-3 have only a mar- 6 were examined in oxygen-free aqueous solution at room
ginal effect on the spectral properties, in agreement with the temperature. Potassium chloride was employed as sup-
absorption spectral properties.
porting electrolyte to achieve a better correspondence with
Replacement of the trans-nitrogen atom by an oxygen the conditions of the SOD activity measurements (see be-
atom in Cu-4 yields a similarly resolved spectrum with only low). All complexes showed a quasi-reversible redox wave
the apparent g factor slightly shifted to higher values. Re- (E1) in the range ϩ0.37 to ϩ0.42 V vs. NHE and an appar-
placement of the cis-nitrogen atoms by oxygen atoms (Cu- ent irreversible cathodic peak (E2pc) between Ϫ0.13 and
5), however, results in both a stronger g shift to higher Ϫ0.36 V (Table 6). The latter potential might be quasi-re-
values and an increased anisotropy of the spectra, as well versible but the corresponding anodic peak was overlayed
as a reduced copper splitting [a(63/65Cu) ഠ 4.9 mT]. These by a strong adsorption peak which appeared between 0 and
effects are further pronounced for Cu-6. Thus, a replace- ϩ0.15 V on the subsequent oxidation cycle. Very similar
ment of a cis-N by O has a stronger effect on the g factor voltammograms were obtained for hydrated CuII ions from
than replacement of the trans-N by O. These observations solutions of copper() perchlorate and copper() sulfate un-
1406
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410

Pyridinyl-Containing 14-Membered Macrocyclic Copper(II) Complexes
FULL PAPER
der the same conditions. The data of Table 6 display no rate is not proportional to the concentrations of copper
dramatic variation of the potentials with the nature of the complexes but rather follows an exponential dependence.
ligands. The quasi-reversible potentials E1 of Cu-1ϪCu-3 The IC50 value, that is the amount of complex reducing the
1/2
differ only by 10 mV, the cathodic peak potentials E2 by rate of the NADH oxidation by 50%, was extracted from
pc
ca. 50 mV. Replacement of N by O in the macrocyclic li- the graphs (Table 7). Since catalase does not interfere in the
gand (Cu-4, Cu-6) shifted the E1 potentials toward slightly above reaction, this enzyme was used to remove the pro-
(20Ϫ40 mV) more positive values, approaching the values duced hydrogen peroxide so as to determine whether the
for CuSO4 and Cu(ClO4)2. The same is true for E2 with peroxide and/or its possible reaction products affect the as-
pc
the exception of Cu-4 for which a cathodic shift of ca. 200 say. As the catalytic activity in the presence of catalase[26]
mV was observed. However, inspection of Table 6 clearly did not change, inhibition by hydrogen peroxide and its
reveals that neither the E1 nor the E2 potential would ac- possible products can be excluded.
pc
count for the unique, though low SOD activity of Cu-3 (see
below, Table 7), as its anodic redox potential does not differ
from those of Cu-1 and Cu-2, and, also, the SOD activities
of all copper compounds (see below) do not parallel the
variation of the E2 values. Thus, the redox potentials of
pc
the CuII complexes are not essentially associated with their
SOD activities.
Table 6. Cyclic voltammetric data [V vs. NHE] of the copper()
complexes
[b]
[c]
Compound[a]
E1
E1
E1
E2
pc
pa
pc
1/2
Cu-1
0.439
0.438
0.465
0.491
0.466
0.463
0.446
0.322
0.317
0.287
0.318
0.366
0.335
0.314
0.381
0.378
0.376
0.405
0.416
0.400
0.414
Ϫ0.164
Ϫ0.138
Ϫ0.214
Ϫ0.358
Ϫ0.056
Ϫ0.068
Ϫ0.005
Figure 5. Inhibition of NADH oxidation as a function of Cu-3
concentration; inset: kinetic traces of NADH oxidation monitored
at 340 nm at different concentrations of Cu-3
Cu-2
Cu-3
Cu-4
Cu-6
Table 7. Comparison of superoxide dismutase activity at 296 K
Cu(ClO4)2
CuSO4
Compound
IC50 [µ]
Cu/NADH ratio
[a] At 1 m and a scan rate of 0.1 V sϪ1 [b] Quasi-reversible, E1/2 ϭ
.
(Epa ϩ Epc)/2. Irreversible or quasi-reversible.
[c]
Cu-2
247
28
206
126
247
1:1.07
1:9.46
1:1.29
1:2.10
1:1.07
Cu-3
Cu-4
Superoxide Dismutase Activity
Cu-6
Cu(ClO4)2
Copper() complexes Cu-2؊Cu-4, and Cu-6 were exam-
ined with regard to their Superoxide Dismutase (SOD) mi-
metic activity by the method of Paoletti et al.[25] For com-
The IC50 values and the Cu/NADH ratios listed in
parison purposes, hydrated CuII [employed as Cu(ClO4)2] Table 7 reveal that complexes Cu-2 and Cu-4 do not show
·Ϫ
was also investigated. The applied method is based on the any true catalytic activity, rather, they convert O2 in a
oxidation of β-nicotinamide adenine dinucleotide (β- roughly 1:1 stoichiometric ratio, similar to copper() per-
NADH) by mercaptoethanol/EDTA/Mn2ϩ at pH ϭ 7.4. In chlorate. Likewise, Cu-6 does not appear to be a catalyst,
the presence of molecular oxygen, this system generates su- reacting with O2·Ϫ in a 1:2 stoichiometry. Hence, simple ex-
peroxide (O2·Ϫ), which, in turn, oxidizes NADH to NADϩ. change of the coordinating atoms, i.e. N vs. O, in the macro-
Based on the kinetics of the oxidation of NADH in the cyclic ring does not effect the SOD-like activity of this type
presence and absence (control) of copper() complexes, the of CuII complexes. However, for complex Cu-3, which car-
SOD-mimetic potential was determined. Addition of SOD ries a methoxybenzyl pendent arm, a low but unequivocal
or a SOD-active compound inhibits the oxidation of catalytic activity is found. The data indicate a turnover of
NADH, which is monitored UV-spectroscopically by the re- about 10, i.e. one molecule of the complex decomposes
tarded decay of the absorption of NADH at 340 nm. After about ten molecules of superoxide before being deactivated.
an initially progressive course, the decline of the absorption
Investigation of the SOD-mimetic activity of complexes
follows apparent zero-order kinetics (Figure 5, inset). These by employing indirect methods like the one chosen here are
parts of the decay traces were used for the subsequent frequently hampered because ‘‘free’’ (hydrated) CuII ions
evaluation of the rates. To determine the SOD activity, the also react with O2·Ϫ, as is demonstrated here by Cu(ClO4)2.
percentage of the rate of inhibition of NADH oxidation Thus, the question arises as to whether the present copper
(control: absence of SOD-mimic ϭ 100%) was plotted complexes or low amounts of free CuII ions, deriving from
against the initial concentration of the copper complexes dissociation of the complexes or impurities, are responsible
(Figure 5). For all complexes the inhibition of the oxidation for the observed reaction. The latter possibility can be ex-
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410
1407

S. Autzen, H.-G. Korth, R. Boese, H. de Groot, R. Sustmann
FULL PAPER
cluded on the basis of the control experiments with cop- might be a crucial feature for the catalytic activity of the
per() perchlorate. Further, the stability of our complexes complex. However, from the currently available data, no
was confirmed by EPR experiments in which a tenfold ex- further conclusions concerning the reaction mechanism can
cess of EDTA added to aqueous solutions of the complexes be made. Further work is necessary for a deeper insight into
produced no detectable changes in the EPR spectra of Cu- the mechanism of superoxide dismutation.
1ϪCu-6.
The catalytic activity of Cu-3 is rather surprising as the
above spectroscopic and electrochemical data do not indic-
ate any unique physicochemical properties in solution com-
pared with the analogues Cu-1 and Cu-2. Indeed, the oxy-
gen-substituted systems Cu-4؊Cu-6 which indeed show
marked variations in their UV/Vis and EPR spectra, behave
very similarly to Cu-1 and Cu-2. Exploratory experiments
were performed with regard to the mechanistic background
for the enhanced activity of Cu-3. In complex Cu-3, the
fifth coordination site in the solid state is occupied by a
chloride ion, whereas in the nonactive complexes Cu-2, Cu-
4, and Cu-6 this site is occupied by a second perchlorate
ion. Thus, the chloride coordination might be responsible
for the catalytic activity of Cu-3. Experiments were, there-
fore, performed in which large amounts (50 m) of potas-
sium chloride were added to the reaction mixture. However,
no changes in the catalytic activities of all complexes were
detected. As neither contamination by ‘‘free’’ CuII nor dif-
ferent coordinating counterions account for the different
activity of Cu-3, it appears that the methoxybenzyl side
chain in Cu-3 is responsible. The X-ray structure of Cu-3
showed that the methoxybenzyl substituent is not coordin-
ated via the oxygen atom to the copper center but rather
directed outward (Figure 3). Such an arrangement would
make a significant influence of the pendent arm on the
SOD activity very unlikely. However, in solution the situ-
ation might change due to free rotational movement of the
methoxybenzyl group. Trial PM3(tm) calculations indeed
favor (at least in the gas phase) an arrangement of the sub-
stituent where the methoxy oxygen atom is coordinated to
the central CuII ion (Figure 6).
Experimental Section
1
Instrumentation: H and 13C NMR spectra: Bruker DRX 500 and
Varian Gemini 200. Melting points (uncorrected): Elektrothermal
9100. IR: Bio-Rad Series FTS 135 FT-IR. UV/Vis: J&M TIDAS
Detector NMC 301. EPR: Bruker ESP 300E. Elemental analysis:
Carlo Erba 1010 CHNSO. Cyclic voltammetry: ECO-Chemie
Autolab GPES.
Materials: The following compounds were prepared according to
published procedures: pyridine-2,6-dicarbaldehyde,[27] 3,7,11,17-
teraazabicyclo[11.3.1]heptadeca-1(17),13,15-triene (1),[6] {2,12-di-
methyl-3,7,11,17-tetraazabicyclo[11.3.1]heptadeca-1(17),2,11,13,15-
pentaene}nickel() perchlorate,[13] {2,12-dimethyl-3,7,11,17-tetra-
azabicyclo[11.3.1]heptadeca-1(17),13,15-triene}nickel() perchlor-
ate,[13] (2R,12S)-2,12-dimethyl-3,7,11,17-tetraazabicyclo[11.3.1]-
heptadeca-1(17),13,15-triene monohydrate (2),[28,29] N,N-bis(2-
cyanoethyl)-2-methoxybenzylamine,
N,N-bis(3-aminopropyl)-2-
methoxybenzylamine,[1]
7-(2-methoxybenzyl)-3,7,11,17-tetraaza-
bicyclo[11.3.1]heptadeca-1(17),13,15-triene (3),[1] 2,6-bis[(tosyl-
oxy)methyl]pyridine,[30] dipropanolamine,[31] 4-oxa-1,7-heptane-
diol,[14] 4-oxaheptane-1,7-diyl ditosylate,[14] 3,7,11-trioxa-17-azabi-
cyclo[11.3.1]heptadeca-1(17),13,15-triene (6).[14] Reduced adenine
dinucleotide (β-NADH, disodium salt) and diethyleneaminotetra-
acetic acid (EDTA) were obtained from Sigma, while manganese()
chloride dihydrate and 2-mercaptoethanol were supplied by Merck
and 3,3Ј-oxydipropionitrile by Fluka. Reagents and solutions: tri-
ethanolamine/diethanolamine (Tea-Dea) buffer (100 m each,
pH ϭ 7.4); NADH (7.5 m); EDTA/MnCl2 (100 m/50 m); mer-
captoethanol (10 mm).
3,3Ј-Oxydipropaneamine: The literature synthesis[32] was improved
as follows: Raney nickel (3 g) was added to a solution of 3,3Ј-oxydi-
propionitrile (2.06 g, 16.6 mmol), dissolved in methanol (40 mL),
containing anhydrous ammonia (5.2 g, 0.31 mol). The nitrile was
hydrogenated under pressure (8 bar) at room temperature, until no
drop in pressure was observed (10 h). The residue was filtered off
and the product obtained by distillation under reduced pressure
1
(2.0 g, 15.1 mmol, 91%). H NMR (500 MHz, CD3OD): δ ϭ 1.66
3
(m, 4 H, 2 ϫ CH2), 2.66 (t, J ϭ 6.5 Hz, 4 H, 2 ϫ CH2), 3.44 (t,
3J ϭ 6.2 Hz, 4 H, 2 ϫ CH2) ppm. 13C NMR (75 MHz, CD3OD):
δ ϭ 33.77 (2 ϫ CH2), 40.11 (2 ϫ CH2), 70.13 (2 ϫ CH2) ppm.
7-Oxa-3,11,17-triazabicyclo[11.3.1]heptadeca-1(17),13,15-triene (4):
A solution of pyridine-2,6-dicarbaldehyde (3.25 g, 24.0 mmol) in
ethanol was added to a stirred solution of copper nitrate trihydrate
(5.80 g, 24.0 mmol) in water (65 mL), resulting in a pale green mix-
ture. A solution of 3,3Ј-oxydipropioamine (3.24 g, 24.5 mmol) in
ethanol (10.0 mL) was then added dropwise with rapid stirring over
30 min and the resulting dark blue solution was refluxed for 2 h,
then cooled to 5 °C in an ice/brine bath and sodium borohydride
(2.30 g, 60.4 mmol) added. After 30 min, the solution was stirred
at room temperature for 2 h, heated at 60 °C for 1 h, and stirred
further at room temperature for several hours. The copper() was
removed by treating the mixture with sodium sulfide nonahydrate
(14.41 g, 60 mmol) followed by stirring at 60 °C for 1 h, and the
Figure 6. Optimized PM3 structure for Cu-3
Therefore, in aqueous solution the methoxybenzyl sub-
stituent might ‘‘protect’’ one of the axial positions against
further, deactivating reactions (e.g. from attack of a second
superoxide molecule) when the substrate superoxide has as-
sociated to the opposite side. Such a blocking function solution was then cooled and the copper() sulfide removed by
1408 Eur. J. Inorg. Chem. 2003, 1401Ϫ1410

Pyridinyl-Containing 14-Membered Macrocyclic Copper(II) Complexes
FULL PAPER
filtration through Celite. Ethanol was evaporated, the residue ex-
tracted three times with dichloromethane, and the combined ex-
tracts dried with MgSO4. After removal of the dichloromethane
by evaporation, the resultant light brown residue was purified by
chromatography (chloroform, neutral aluminium oxide). The prod-
uct was obtained as a colourless solid (3.16 g, 13.4 mmol, 56%). 1H
Cu-4: Blue crystals (0.83 g, 73%), decomp. 290 °C. IR (KBr): ν˜ ϭ
3242 (NH), 3092 (CHPh), 2951 (CHaliph), 1608 (NH), 1108 (COC),
624 (ClO4) cmϪ1. C13H21Cl2CuN3O9 (497.78): calcd. C 31.37, H
4.25, Cu 12.77, N 8.44; found C 31.72, H 4.26, Cu 12.66, N 8.52.
FAB (glycerol): m/z (%) ϭ 397 (22), 298 (30). UV (water): λmax (lg
ε) ϭ 255 (3.82), 603 (2.15) nm.
3
NMR (500 MHz, CDCl3): δ ϭ 1.82 (m, 4 H, CH2), 2.59 (t, J ϭ
Cu-5: Light blue crystals (0.24 g, 65%), decomp. 291 °C. IR (KBr):
3
5.5 Hz, 4 H, CH2), 3.54 (t, J ϭ 5.5 Hz, 4 H, CH2), 3.70 (s, 2 H,
ν˜ ϭ 3078 (CHPh), 2957 (CHaliph), 1605 (NH), 627 (ClO4) cmϪ1
.
3
NH), 3.89 (s, 4 H, CH2), 6.99 (d, J ϭ 7.6 Hz, 2 H, CH), 7.52 (dd,
C13H20Cl2CuN2O10 (498.76): calcd. C 31.31, H 4.04, Cu 12.74, N
5.62; found C 30.98, H 3.94, Cu 12.89, N 5.47. UV (water): λmax
(lg ε) ϭ 757 (1.42) nm.
1 H, 3J ϭ 7.6 Hz, CH) ppm. 13C NMR (75 MHz, CDCl3): δ ϭ
29.28 (CH2), 45.39 (CH2), 53.91 (CH2), 68.34 (CH2), 120.81 (CH),
136.72 (CH), 158.53 (C) ppm. C13H21N3O (235.33): calcd. C 66.35,
H 8.99, N 17.86; found C 66.05, H 9.17, N 17.61.
Cu-6: Turquoise crystals (0.36 g, 72%), decomp. 292 °C. IR (KBr):
ν˜
ϭ
3084 (CHPh), 2957 (CHaliph), 621 (ClO4) cmϪ1
.
3,11-Dioxa-7,17-diazabicyclo[11.3.1]heptadeca-1(17),13,15-triene
C13H19Cl2CuNO11 (499.73): calcd. C 31.24, H 3.83, Cu 12.72, N
2.80; found C 31.52, H 3.82, Cu 12.43, N 2.99. FAB (glycerol):
m/z (%) ϭ 300 (100), 238 (83). UV (water): λmax (lg ε) ϭ 264 (3.59),
793 (1.36) nm.
(5): Macrocycle 5 was prepared from (1.81 g, 4.0 mmol) 2,6-bis[(to-
syloxy)methyl]pyridine and dipropanolamine (0.54 mg, 4.0 mmol)
in a similar fashion to the procedure of Bradshaw et al.[14] The
product thus obtained was a colourless solid (0.51 g, 2.16 mmol,
1
54%). H NMR (200 MHz, CDCl3): δ ϭ 1.91 (m, 4 H, CH2), 3.05
Aqueous solutions of the complexes all had pH values between 7
and 8.
(br. m, 8 H, CH2), 4.67 (s, 4 H, CH2), 7.20 (d, 2 H, CH), 7.68 (dd,
1 H, CH) ppm. 13C NMR (50 MHz, CDCl3): δ ϭ 30.12 (CH2),
49.15 (CH2), 65.83 (CH2), 72.72 (CH2), 117.12 (CH), 123.52 (CH),
158.88 (C) ppm. C13H20N2O2 (236.31): calcd. C 66.07, H 8.53, N
11.86; found C 66.33, H 8.55, N 11.92.
Kinetic Measurements: The principle of the method is based on the
oxidation of NADH, mediated by superoxide radical, in a purely
chemical system developed by Paoletti et al.[25] To monitor the
assay we added the following solutions (see materials) into a
cuvette (1-mm cell): 400 µL Tea-Dea buffer (100 m each), 20 µL
NADH solution (7.5 m), 12.5 µL EDTA/MnCl2 solution
(100 m/50 m) and 50 µL of sample or buffer were mixed thor-
oughly. In a second series of experiments the reaction mixture was
supplemented with 50 m KCl. The reaction was started by rapid
mixing with 50 µL of mercaptoethanol (10 m). Rates of NADH
oxidation, observed at 340 nm, were initially low, then increased
progressively to yield good linear kinetics and an 8-min window
was used for our calculations. The maximal rates obtained are ex-
pressed as a percentage of the control and plotted against the com-
plex concentracion. One unit of the SOD-mimic is the amount of
copper complex capable of inhibiting by 50% the rate of NADH
oxidation observed in the control.
Copper Complexes Cu-1ϪCu-6: The preparation of Cu-1[6] and Cu-
2[7] has been reported before. All solid complexes were obtained in
good yields (Ͼ 65%) by mixing ethanolic solutions of the appropri-
ate ligands with equimolar solutions of cupric perchlorate hexahy-
drate in ethanol at room temperature. The precipitated complexes
were isolated by filtration and purified by washing several times
with cold ethanol. Whereas complexes Cu-1, Cu-2, and Cu-4ϪCu-
6 were instantaneously formed on mixing of the reactants, addition
of an equimolar amount of copper() perchlorate in ethanol to an
ethanolic solution of ligand 3 initially led to rapid formation of a
yellow precipitate which according to elemental analysis and EPR
spectroscopy represented a CuIIϪligand complex of 1:2 stoichi-
ometry, [CuII(3)2](ClO4)2 (60% yield relative to 3). When the reac-
tion mixture without prior separation of this complex was allowed
to stand for 12 h at ambient temperature, violet crystals of complex
Cu-3 were obtained in 68% yield.
EPR Spectroscopy: X-band (9.5 GHz) EPR spectra were recorded
from 1Ϫ3 m solutions of the complexes in oxygen-free, deionized
water in a TM110 wide-bore cavity (Bruker). For liquid solutions, a
0.4 mm inner diameter quartz flat cell (Wilmad) was used; meas-
urements at 77 K were performed in 4 mm Suprasil quartz tubes
immersed in a liquid nitrogen dewar insert. Spectral simulations
were carried out with the Simfonia program (Bruker). The concen-
tration of the copper complexes in liquid and frozen solution was
CAUTION: Metal perchlorate salts containing organic ligands are
potentially explosive, and should be handled with care!
Characterization of Complexes Cu-1ϪCu-6
Cu-1: Blue-purple powder (0.39 g, 68%). IR (KBr): ν˜ ϭ 3242 (NH),
3084 (CHPh), 2936 (CHaliph), 624 (ClO4) cmϪ1. C13H22Cl2CuN4O8 checked by numerical double integration (WinEPR software,
(496.80): calcd. C 31.43, H 4.46, Cu 12.79, N 11.28; found C 31.14,
Bruker) of the EPR spectra and comparison with the integrated
H 4.41, Cu 12.58, N 10.99. UV (water): λmax (lg ε) ϭ 565 (2.56) nm. EPR signal of a 1 m CuII-EDTA solution. In all cases the integra-
tion accounted for Ͼ 90% purity of the complexes.
Cu-2: Violet-blue powder (0.76 g, 76%), decomp. 254 °C. IR (KBr):
ν ϭ 3223 (NH), 3094 (CHPh), 2943 (CHaliph), 624 (ClO4) cm-1.
Cyclic Voltammetry: Cyclic voltammetric measurements were per-
˜
C15H26Cl2CuN4O8 (524.86): calcd. C 34.33, H 4.99, Cu 12.11, N formed at 20 °C in aqueous solution under nitrogen. Complex con-
10.68; found C 34.29, H 4.97, Cu 11.84, N 10.39. UV (water): λmax
(lg ε) ϭ 254 (3.90), 560 (2.29) nm.
centrations of 1 m were employed, and 100 m KCl was used as
supporting electrolyte. A 3 mm diameter glassy-carbon working
electrode, an Ag/AgCl reference electrode, and a Pt-wire counter
Cu-3: Violet-blue crystals (0.39 g, 68%), decomp. 231 °C. IR (KBr):
ν˜ ϭ 3226 (NH), 3068 (CHPh), 2934 (CHaliph), 2893 (CHmethyl ether),
electrode were used. Scan rates were varied from 0.01 to 1.0 V sϪ1
.
Electrochemical potentials were converted into the normal hydro-
gen electrode (NHE) scale by addition of 0.222 V.
1600 (NH), 1248 (COaryl
ether), 622 (ClO4) cmϪ1
.
alkyl
C21H30Cl2CuN4O5 (553.73): calcd. C 45.61, H 5.47, Cu 11.49, N
10.13; found C 45.31, H 5.55, Cu 11.41, N 9.84. FAB (glycerol):
X-ray Crystallographic Study: CCDC-188550 (Cu-3) and -188551
m/z (%) ϭ 516 (3), 417 (15), 355 (34). UV (water): λmax (lg ε) ϭ (Cu-6) contain the supplementary crystallographic data for
256 (3.89), 563 (2.13) nm.
this paper. These data can be obtained free of charge at
1409
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410

S. Autzen, H.-G. Korth, R. Boese, H. de Groot, R. Sustmann
FULL PAPER
Chi, F.-L. Liao, C.-S. Chung, Anal. Sci., 2000, 16, 441Ϫ442.
K. Mochizuki, S. Miyashita, Chem. Lett. 1996, 899Ϫ900. H.
Keypour, S. Salehzadeh, R. G. Pritchard, R. V. Parish, Inorg.
Chem. 2000, 39, 5787Ϫ5790.
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge
CB2 1EZ, UK; Fax: (internat.)
deposit@ccdc.cam.ac.uk].
ϩ 44-1223/336-033; E-mail:
[13]
[14]
J. L. Karn, D. H. Busch, Inorg. Chem. 1969, 8, 1149Ϫ1153.
J. S. Bradshaw, J. M. Guynn, S. G. Wood, B. E. Wilson, N. K.
Dalley, R. M. Izatt, J. Heterocycl. Chem. 1987, 24, 415Ϫ419.
[15]
´
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft
(SFB, 452).
M. M. Abu-Omar, L. D. McPherson, J. Arias, V. M. Bereau,
Angew. Chem. 2000, 112, 4480Ϫ4483; Angew. Chem. Int. Edit.
2000, 39, 4310Ϫ4313.
V. L. Goedken, Y.-A. Park, J. Chem. Soc., Chem. Commun.
1975, 214Ϫ215.
T. Murakami, S. Hatakeyama, S. Igarashi, Y. Yukawa, Inorg.
Chim. Acta 2000, 310, 96Ϫ 102.
B. J. Hathaway in Comprehensive Coordination Chemistry, vol.
5, chapter 53 (Ed.: G. Wilkinson), Pergamon Press, Oxford,
1987, p. 594Ϫ744.
C. K. Jørgensen, Acta Chem. Scand. 1955, 9, 1362Ϫ1377.
M. Linss, U. Weser, Inorg. Chim. Acta 1986, 125, 117Ϫ121.
N. W. Alcock, K. P. Balakrishnan, P. Moore, G. A. Pike, J.
Chem. Soc., Dalton Trans. 1987, 889Ϫ894.
J. R. Pilbrow Transition Metal Ion Electron Paramagnetic Res-
onance, Clarendon Press, Oxford, 1990, pp. 9, 30, 154, 484.
F. S. Stephens, J. Chem. Soc. A 1969, 1723Ϫ1734.
R. A. Lieberman, R. H. Sands, J. A. Fee, J. Biol. Chem. 1982,
257, 336Ϫ344.
F. Paoletti, D. Aldinucci, A. Mocali, A. Caparrini, Anal. Bio-
chem. 1986, 154, 536Ϫ541.
F. Paoletti, A. Mocali, Methods Enzymol. 1990, 186, 209Ϫ220.
N. W. Alcock, R. G. Kingston, P. Moore, C. Pierpoint, J.
Chem. Soc., Dalton Trans. 1984, 1937Ϫ1943.
A. C. Melnyk, N. K. Kildahl, A. R. Rendina, D. H. Busch, J.
Am. Chem. Soc. 1979, 101, 3232Ϫ3240.
C. J. Cairns, R. A. Heckman, A. C. Melnyk, W. M. Davis, D.
H. Busch, J. Chem. Soc., Dalton Trans. 1987, 2505Ϫ2510.
J. S. Bradshaw, P. Huszthy, C. W. McDaniel, C. Y. Zhu, N. K.
Dalley, R. M. Izatt, J. Org. Chem. 1990, 55, 3129Ϫ3137.
C. Granier, R. Guilard, Tetrahedron 1995, 51, 1197Ϫ1208.
P. F. Wiley, J. Am. Chem. Soc. 1946, 68, 1867Ϫ1868.
Received July 12, 2002
[16]
[17]
[18]
[1]
S. Autzen, H.-G. Korth, H. de Groot, R. Sustmann, Eur. J.
Org. Chem. 2001, 3119Ϫ3125.
[2]
B. Halliwell, New Phytol. 1974, 73, 1075Ϫ1086.
[3]
T. Nagano, Yuki Gosei Kagaku 1989, 47, 843Ϫ854. K. Jitsu-
kawa, M. Harata, H. Arii, H. Sakurai, H. Masuda, Inorg.
Chim. Acta 2001, 324, 108Ϫ116.
[19]
[20]
[21]
[4]
J. R. Sorenson, Prog. Med. Chem. 1989, 26, 437Ϫ568.
[5]
D. P. Riley, Chem. Rev. 1999, 99, 2573Ϫ2587.
K. P. Balakrishnan, H. A. A. Omar, P. Moore, N. W. Alcock,
[6]
[22]
G. A. Pike, J. Chem. Soc., Dalton Trans. 1990, 2965Ϫ2969.
L. F. Lindoy, N. E. Tokel, L. B. Anderson, D. H. Busch, J.
[7]
[23]
[24]
Coord. Chem. 1971, 1, 7Ϫ16.
[8]
J. Costa, R. Delgado, Inorg. Chem. 1993, 32, 5257Ϫ5265.
[9]
[25]
V. Felix, M. J. Calhorda, J. Costa, R. Delgado, C. Brito, M. T.
Duarte, T. Arcos, M. G. B. Drew, J. Chem. Soc., Dalton Trans.
1996, 4543Ϫ4553.
[26]
[27]
[10]
P. Chaudhuri, K. Wieghardt, Progr. Inorg. Chem. 2001, 50,
151Ϫ216.
[11]
[28]
[29]
[30]
R. R. Fenton, R. Gauci, P. C. Junk, L. F. Lindoy, R. C.
Luckay, G. V. Meehan, J. R. Price, P. Turner, G. Wei, J. Chem.
Soc., Dalton Trans. 2002, 2185Ϫ2193.
E. V. Rybak- Akimova, A. Y. Nazarenko, S. S. Silchenko, Inorg.
[12]
Chem. 1999, 38, 2974Ϫ2980. M. Mikuriya, K. Hamada, S.
Kida, I. Murase, Bull. Chem. Soc. Jpn, 1985, 58, 1839Ϫ1840.
J. Costa, R. Delgado, M. G. B. Drew, V. Felix, J. Chem. Soc.,
Dalton Trans. 1999, 4331Ϫ4339. M. Di Casa, L. Fabbrizzi, M.
Licchelli, A. Poggi, D. Sacchi, M. Zema, J. Chem. Soc., Dalton
Trans. 2001, 1671Ϫ1675. K. Panneerselvam, T.-H. Lu, T.-Y.
[31]
[32]
[I02379]
1410
Eur. J. Inorg. Chem. 2003, 1401Ϫ1410