Complexes of CuII with mixed-donor phenanthroline-containing macrocycles: Analysis of their structural, redox and spectral properties in the context of Type-1 blue copper proteins biomimetic models

Article abstract of DOI:10.1016/j.ica.2005.01.014

The macrocycles L¹-L³ having N₂S ₂O-, N₂S₂-, and N₂S ₃-donor sets, respectively, and incorporating the 1,10-phenanthroline unit interact in EtOH and MeCN solutions with Cu^II to give 1:1 [M(L)]²⁺ complex species. The compounds [Cu(L¹)(ClO ₄)]ClO₄ (1), [Cu(L²)(ClO₄)]ClO ₄·1/2H₂O (2) and [Cu(L³)](ClO ₄)₂ (3) were isolated at the solid state and the first two also characterised by X-ray diffraction studies. The conformation adopted by L¹ and L² in the cation complexes reveals the aliphatic portion of the rings folded over the plane containing the heteroaromatic moiety with the ligands encapsulating the metal centre within their cavity by imposing, respectively, a square-based pyramidal and a square planar geometry. In both complexes, the metal ion completes its coordination sphere by interacting with a ClO₄^- ligand. The compound [Cu(L³) ₂](PF₆)₂ (4) containing a 1:2 cation complex was also isolated at the solid state: EPR spectroscopy measurements suggest the presence of a CuN₄ chromophore in this complex. The EPR and electronic spectral features of 1-4 have been studied and their redox properties examined in comparison with those observed for Type-1 blue copper proteins. The reactivity of L¹-L³ has also been tested toward stoichiometric amounts of the Cu^I salt [CuCl(PPh₃) ₃].

Full text of DOI:10.1016/j.ica.2005.01.014

Inorganica Chimica Acta 358 (2005) 2403–2412
www.elsevier.com/locate/ica
Complexes of CuII with mixed-donor phenanthroline-containing
macrocycles: analysis of their structural, redox and spectral
properties in the context of Type-1 blue copper proteins
biomimetic models
a
Massimiliano Arca , Gholamhassan Azimi , Francesco Demartin ,
b
c
a
Francesco A. Devillanova , Lluis Escriche , Alessandra Garau , Francesco Isaia ,
d
a
a
e
a,
d
a
Raikko Kivekas , Vito Lippolis *, Vicent Muns , Alessandro Perra ,
f
Mojtaba Shamsipur , Luigi Sportelli , Abdollah Yari
g
f
a
`
Dipartimento di Chimica Inorganica ed Analitica, Universita degli Studi di Cagliari, S.S. 554 Bivio per Sestu, 09042 Monserrato, Cagliari, Italy
b
Department of Chemistry, Arak University, Shariati sp., Arak, Iran
c
`
Dipartimento di Chimica Strutturale e Stereochimica Inorganica, Universita di Milano, Via G. Venezian 21, 20133 Milano, Italy
d
´ `
Departement de Quımica (Unit Inorganica), Universitat Autonoma de Barcelona, 08193 Bellaterra, Barcelona, Spain
e
Department of Chemistry, University of Helsinky, P.O. Box 55, FIN 00014 Helsinky, Finland
f
Department of Chemistry, Razi University, Kermanshah, Iran
g
` `
Laboratorio di Biofisica Molecolare, Dipartimento di Fisica and Unita INFM, Universita della Calabria, 87036 Rende, Cosenza, Italy
Received 20 December 2004; accepted 21 January 2005
Abstract
The macrocycles L1–L3 having N2S2O-, N2S2-, and N2S3-donor sets, respectively, and incorporating the 1,10-phenanthroline unit
interact in EtOH and MeCN solutions with CuII to give 1:1 [M(L)]2+ complex species. The compounds [Cu(L1)(ClO4)]ClO4 (1),
[Cu(L2)(ClO4)]ClO4 Æ 1 H2O (2) and [Cu(L3)](ClO4)2 (3) were isolated at the solid state and the first two also characterised by X-
2
ray diffraction studies. The conformation adopted by L1 and L2 in the cation complexes reveals the aliphatic portion of the rings
folded over the plane containing the heteroaromatic moiety with the ligands encapsulating the metal centre within their cavity
by imposing, respectively, a square-based pyramidal and a square planar geometry. In both complexes, the metal ion completes
its coordination sphere by interacting with a ClO4 ligand. The compound [Cu(L3)2](PF6)2 (4) containing a 1:2 cation complex
ꢀ
was also isolated at the solid state: EPR spectroscopy measurements suggest the presence of a CuN4 chromophore in this complex.
The EPR and electronic spectral features of 1–4 have been studied and their redox properties examined in comparison with those
observed for Type-1 blue copper proteins.
The reactivity of L1–L3 has also been tested toward stoichiometric amounts of the CuI salt [CuCl(PPh3)3].
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Macrocycles; Phenanthroline; Type-1 blue copper proteins; X-ray diffraction
1. Introduction
The spectral and redox properties of Type-1 blue cop-
per proteins [1–6] are strongly determined by the partic-
ular chemical environment at the metal centre in the
*
Corresponding author. Tel.: +39 070 675 4467; fax: +39 070 675
4456.
E-mail address: lippolis@unica.it (V. Lippolis).
0020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.01.014

2404
M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
active site, which in general comprises two N- (from his-
tidine residues) and two S-donors (from a cysteine and a
methionine residue) arranged in a distorted tetrahedral
geometry [1–6]. In azurin, one oxygen atom from a gly-
cine residue is also strongly coordinated to CuII to give a
distorted trigonal bipyramidal coordination geometry
[5].
N
N
N
N
N
N
S
S
S
S
S
S
O
S
Many CuII complexes have been synthesised over sev-
eral decades with the aim of reproducing Type-1 coordi-
nation geometry and addressing the structure/spectral
and redox properties relationship [7–9]. In this respect,
macrocyclic ligands incorporating N/S/O donors have
been used extensively to model the active sites of blue
copper proteins [10–16] due to their restricted frame-
work conformability, which can be considered responsi-
ble for unusual constrained coordination geometries at
the metal centre, and consequently for peculiar elec-
tronic spectra and redox potentials of the resulting
complexes.
For example, copper(II) complexes of macrocyclic li-
gands, such as [14]aneSxN(4 ꢀ x) (x = 1–4), have been
extensively studied [13,14,16]; substitution of a thioe-
ther for an amine donor leads to an increase in E1/2
for the CuII/CuI couple and parallely to a shift in the
absorption spectrum to longer wavelengths and higher
absorption coefficients. In some cases, the spectroscopic
and redox properties of CuII complexes of these ligands
are similar to those of Type-1 native blue copper pro-
teins, though the coordination geometry around CuII
tends to be square-based. Macrocyclic ligands with re-
stricted flexibility have also been synthesised by intro-
ducing aromatic moieties in the cyclic structure with
the aim of having CuII and CuI complexes with identi-
cal donor sets and closely related geometries [14]. Addi-
son and co-workers [15] have reported closely related
pairs of CuII and CuI complexes with Schiff-base mac-
rocycles derived from the condensation of 1,4-bis(2-
formylphenyl)-1,4-dithiabutane and aliphatic a,x-dia-
mine; on changing the length of the N–N linkage a
tetrahedral environment can be provided to CuI, thus
making the reduction potential for the CuII/CuI couple
more positive.
L1
L2
L3
2. Experimental
All melting points are uncorrected. Microanalytical
data were obtained using a Fisons EA CHNS-O instru-
ment (T = 1000 ꢁC). FAB mass spectra (3-NOBA ma-
trix) were recorded at the Institut fur Organische
¨
Chemie der Technischen Universita¨t, Braunschweig.
1
The H and 13C NMR spectra were recorded on a Var-
ian VXR300 spectrometer (operating at 75.4 MHz). The
spectrophotometric measurements were carried out
using a Cecil 9000 spectrophotometer.
Electron paramagnetic resonance (EPR) measure-
ments were carried out using a Bruker Mod. ER 200D
SRC X-band spectrometer equipped with the TE102
standard cavity and the ESP 1600 Data System. Cyclic
voltammetry was performed using a conventional
three-electrode cell, with a platinum double-bead elec-
trode and Ag/AgCl reference electrode (in 3.5 M KCl;
0.2050 V vs SHE). All measurements were taken in a
0.1 M MeCN solution of Bu4NBF4. A stream of argon
was passed through the solution prior to the scan. Data
were recorded on a computer-controlled EG&G (Prince-
ton Applied Research) potentiostat-galvanostat Model
273 EG&G, using model 270 electrochemical analysis
software.
Reagent grade copper perchlorate, copper chloride
dihydrate, acetonitrile (MeCN), and ethanol (EtOH)
(all from Aldrich) were of the highest purity available
and used without any further purification. The salt
[CuCl(PPh3)3] was synthesised according to the litera-
ture [22]. The ligands L1, L2 and L3 were synthesised
as previously reported [17–21].
Recently, we have reported the synthesis and coor-
dinating properties towards some d8 and d10 transi-
tion metal ions of the mixed aza-thioether crowns
L1–L3 incorporating the 1,10-phenanthroline sub-unit
[17–21]. These ligands have shown peculiar chelating
properties strongly determined by conformational
constraints imposed by the phenanthroline unit on
the rest of the aliphatic chain of the ring. Now, we
have investigated the coordination chemistry of L1,
L2, and L3 towards CuII. The effect of the phenan-
throline unit on the spectral and redox properties of
the resulting complexes has also been analysed in
the context of Type-1 blue copper protein biomimetic
models.
Caution! Perchlorate salts are potentially explosive;
these compounds must be handled with great care.
2.1. Synthesis of [Cu(L1)(ClO4)]ClO4 (1), [Cu(L2)
(ClO4)]ClO4 Æ 1 H2O (2) and [Cu(L3)](ClO4)2 (3)
2
Addition of one equivalent of Cu(ClO4)2 Æ 2H2O in
EtOH (5 ml) to a solution of L1–L3 (10ꢀ3 mol) in
EtOH–CH2Cl2 (10 ml, 2:1 v/v ratio) maintained at room
temperature under N2 afforded blue-green (1), light blue
(2), and purple (3) solids after four hours stirring.

M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
2405
1: Re-crystallisation of the crude product by slow dif-
fusion of Et2O into a MeCN solution gave [Cu(L1)-
(ClO4)]ClO4 (51% yield) as well shaped blue prisms:
m.p. 260 ꢁC with decomposition. Elemental Analysis:
Anal. Calc. for C18H18Cl2CuN2O9S2: C, 35.7; H, 3.0;
N, 4.6; S, 10.6. Found: C, 36.1; H, 3.1; N, 4.6; S,
10.6%. FAB mass spectrum (3-noba matrix): m/z 504,
405; calc. for [63Cu(L1)(ClO4)]+ and [63Cu(L1)]+ 504
and 405, respectively. Electronic spectrum (MeCN):
Anal. Calc. for C36H33ClCuN2PS3: C, 60.1; H, 4.6; N,
3.9; S, 13.4. Found: C, 60.3; H, 4.5; N, 3.8; S, 13.4%.
FAB mass spectrum (3-noba matrix): m/z 421; calc. for
1
[63Cu(L3)]+ 421. H-NMR (300 MHz, 298 K, CD2Cl2):
dH 8.26 (d, 2H, J = 7.8 Hz), 7.76 (s, 2H), 7.71 (d, 2H,
J = 7.8 Hz), 7.26–7.05 (m, 15H), 4.5 (s, 4H), 3.10 (m,
4H), 2.7 (m, 4H). 13C-NMR (75.4 MHz, 298 K,
CD2Cl2): dC 160.6, 142.9, 137.7, 134.1, 133.9, 129.2,
128.3, 128.2, 127.7, 126.2, 125.0, 38.3, 34.5, 33.4. Elec-
tronic spectrum (CH2Cl2): kmax = 273 (emax = 26 000),
427 nm (650 Mꢀ1 cmꢀ1).
k
max = 230 (emax = 48 000), 278 (29 600), 361 (4400),
639 nm (400 Mꢀ1 cmꢀ1).
2: Re-crystallisation of the crude product by slow dif-
fusion of Et2O into
a
MeCN solution gave
2.4. Spectrophotometric titrations
[Cu(L2)(ClO4)]ClO4 Æ 1 H2O (60% yield) as well shaped
2
light blue prisms: m.p. 270 ꢁC with decomposition. Ele-
mental Analysis: Anal. Calc. for C19H21Cl2CuN2O8.5S2:
C, 37.3; H, 3.5; N, 4.6; S, 10.5. Found: C, 37.5, H, 3.3,
N, 4.6, S, 10.6%. FAB mass spectrum (3-noba matrix):
m/z 502, 403; calc. for [63Cu(L2)(ClO4)]+ and
[63Cu(L2)]+ 502 and 403, respectively. Electronic spec-
In order to monitor complex formation between L1–
L3 and CuII, 2 ml of the copper solution in MeCN or
EtOH (4 · 10ꢀ4–5 · 10ꢀ4 M) was placed in a 1 cm opti-
cal path quartz cell thermostated at 25 ꢁC. A known
amount of the ligand (L/CuII = 0.2–4) was added to
the cell and the spectra were recorded soon after addi-
tion of the ligand to the copper solution.
trum (MeCN): kmax = 230 (emax = 45 000), 278
(29 100), 360 (4400), 630 nm (400 Mꢀ1 cmꢀ1).
3: Crystals of good quality were not obtained upon re-
crystallisation (61% yield) (m.p. = 250ꢁC with decompo-
sition). Elemental Analysis: Anal. Calc. for C18H18Cl2
CuN2O8S3: C, 34.8; H, 2.9; N, 4.5; S, 15.5. Found: C,
34.8; H, 2.8; N, 4.5; S, 15.4%. FAB mass spectrum (3-
noba matrix): m/z 521, 421; calc. for [63Cu(L3)(ClO4)]+
and [63Cu(L3)]+ 521 and 421, respectively. Electronic
spectrum (MeCN): kmax = 230 (kmax = 34 540), 278
(28 300), 355 (4000), 592 nm (250 Mꢀ1 cmꢀ1).
2.5. Crystallography
Crystal data and refinement details of all structure
determinations are summarised in Table 1. Only special
features of the analyses are pointed out here. Single crys-
tal data collections for L2 Æ H2O and [Cu(L1)(ClO4)]ClO4
(1) were performed at room temperature on a CAD-4
diffractometer using x scans. For [Cu(L2)(ClO4)]-
ClO4 Æ 1 H2O (2) and [Cu(L3)(Cl)(PPh3)] (5), data were
2
acquired at room temperature on a SMART CCD dif-
fractometer (x scans) and a Rigaku AFC5S diffractom-
eter (xꢀh scans), respectively. All data sets were
corrected for Lorentz-polarisation effects and for
absorption as specified in Table 1. The structures were
solved by direct methods using the SHELXS program
[24] followed by difference Fourier synthesis and refined
by full matrix least-squares on F2 [25]. All non-H atoms
were refined anisotropically and H atoms were intro-
duced at calculated positions and thereafter incorpo-
rated into a riding model with Uiso(H) = 1.2Ueq(C). In
L2 Æ H2O, the carbon atoms C(18) and C(19) in the ali-
phatic portion of the ring between the two S-donors
were found to be disordered. The disorder was modelled
setting for each carbon atom the occupancy of the two
disordered sites to 0.65 and 0.35, respectively. Appropri-
ate restraints were applied to the bond lengths of the dis-
ordered portion of the macrocyclic ligand, both
components being refined isotropically. The H atoms
could not be included in the refinement model. The posi-
tions of the H atoms of the water molecule were seen in
a difference Fourier map and were refined. For
2.2. Synthesis of [Cu(L3)2](PF6)2 (4)
A mixture of L3 (50 mg, 0.139 mmol) and CuCl2 Æ 2-
H2O (23.8 mg, 0.139 mmol) in MeCN/H2O (20 ml, 1:1
v/v) was refluxed under N2 for 2 h. Addition of a large
excess of NH4PF6 to the resulting solution afforded a
brick red solid of [Cu(L3)2](PF6)2 (20 mg, 14% yield):
m.p. 230 ꢁC with decomposition. Elemental Analysis:
Anal. Calc. for C36H36CuF12N4P2S6: C, 40.4; H, 3.4;
N, 5.2; S, 18.0. Found: C, 40.4; H, 3.6; N, 5.7; S,
17.7%. FAB mass spectrum (3-noba matrix): m/z 421;
calc. for [63Cu(L3)]+ 421. Electronic spectrum (MeCN):
k
max = 275 (emax = 10 800), 377 (3400), 467 (1100),
750 nm (140 Mꢀ1 cmꢀ1).
2.3. Synthesis of [Cu(L3)(Cl)(PPh3)] (5)
A solution of L3 (20 mg, 0.06 mmol) in CH2Cl2
(1.5 ml) was added to a suspension of [CuCl(PPh3)3]
(50 mg, 0.06 mmol) in EtOH (2 ml) maintained at room
temperature under N2. After filtration of the solution
and about one hour, [Cu(L3)(Cl)(PPh3)] (8.7 mg,
22.4% yield) was obtained as prism orange crystals:
m.p. 270 ꢁC with decomposition. Elemental Analysis:
[Cu(L2)(ClO4)]ClO4 Æ 1 H2O (2), the space group cannot
2
be uniquely determined from systematic absences being
the C2/m, C2 and Cm space groups possible. The

2406
M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
Table 1
Summary of crystallographic data for L2 Æ H2O, [Cu(L1)(ClO4)]ClO4 (1), [Cu(L2)(ClO4)]ClO4 Æ 1 H2O (2), and [Cu(L3)(Cl)(PPh3)] (5)
2
Compound
Formula
M
L2 Æ H2O
C19H22N2OS2
358.51
[Cu(L1)(ClO4)]ClO4 (1)
C18H18Cl2CuN2O9S2
604.90
[Cu(L2)(ClO4)]ClO4 Æ 21 H2O (2)
C19H21Cl2CuN2O8.5S2
611.94
[Cu(L3)(Cl)(PPh3)] (5)
C36H33ClCuN2PS3
719.78
Crystal system
Space group
monoclinic
P21/n (No. 14)
7.653(1)
19.644(3)
11.979(2)
90.72(1)
1800.8(5)
4
monoclinic
P21/c (No. 14)
9.063(3)
monoclinic
C2/m (No. 12)
13.242(2)
monoclinic
P21/c (No. 14)
9.915(2)
˚
a (A)
˚
b (A)
11.872(3)
20.658(10)
93.03(4)
11.623(1)
15.950(2)
33.468(2)
10.374(2)
99.72(1)
˚
c (A)
b (ꢁ)
98.32(1)
2429.1(5)
3
˚
V (A )
2219.6(14)
4
1.810
3393.0(10)
4
1.409
Z
4
1.673
Dc (g cmꢀ3
)
1.322
0.304
l (Mo Ka) (mmꢀ1
Unique reflection (Rint
)
1.469
3890 (0.018)
2669
1.342
2285 (0.036)
1440
0.983
5996 (0.034)
3490
)
3145 (0.023)
1367
Observed reflections [I > 2r(I)]
Absorption correction
Tmin, Tmax
w-scans
0.94, 1.00
0.0707
w-scans
0.91, 1.00
SADABS [23]
0.81, 1.00
0.0596
w-scans
0.719, 0.813
R1
0.0385
0.1049
0.0511
0.1189
wR2 [all data]
0.2010
0.1894
refinement carried out in the space group C2/m gives a
model with atom C(18) of the aliphatic moiety disor-
dered over two positions. The corresponding refinement
in the non centrosymmetric space group C2 provides a
model with no disorder in the aliphatic moiety, but
absolutely not reliable interatomic distances and dis-
placement parameters for all the ‘‘mirror-related’’ atoms
of the phenanthroline moiety, due to strong correlation
in the refinement procedure. Any attempt to solve the
structure in the space group Cm failed. Therefore, we
consider space group C2/m as the most appropriate for
describing the structure of (2).
After many attempts, we have also been able to grow
single crystals of L2 Æ H2O from a MeCN/Et2O solution.
The crystal structure confirms the nature of the com-
pound and shows the presence of disorder in the ali-
phatic portion of the ring between S(1) and C(17)
(Fig. 1), analogously to what was found for one of the
two independent molecules in the asymmetric unit of
L1 Æ 12 H2O [20]. Molecular mechanics (MM) calculations
previously performed on both L1 and L3 [20] have now
been extended to L2, with the aim to understand the ef-
fect on the flexibility of the aliphatic chain on removing
one donor atom. The eight most stable conformers cal-
In [Cu(L3)(Cl)(PPh3)] (5), the carbon atoms C(15)
and C(16) between the S-donors S(1) and S(3) in the ali-
phatic portion of the ring were found to be disordered.
In this case, the disorder was modelled by a partial occu-
pancy over two sites for each of the carbon atoms, con-
verging with factors of 0.58 and 0.42. Both components
bearing H atoms positioned geometrically were refined
anisotropically with appropriate restraints imposed to
the bond lengths of the disordered portion of the ligand.
2.6. Molecular mechanics calculations
MM calculations were performed using the Spartan
5.0 program [26] as previously described [21]. The calcu-
lated conformers have been examined using the Molekel
4.3 program [27].
3. Results and discussion
3.1. Ligands
Fig. 1. ORTEP view of L2 in the crystal structure of L2 Æ 21 H2O. Atoms
C(19A) and C(18A) comprise the major disordered component.
We reported previously the synthesis of the ligands
L1–L3 and the X-ray crystal structure of L1 [17–21].

M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
2407
culated for L2, which differ in energy by less than
0.6
12.5 kJ molꢀ1, and their torsion angle have been depos-
ited as ESI (Fig. S1, Table S1). Analogously to what was
found for L1 and L3 [20], two features are common to all
calculated conformations of L2: (a) the aliphatic portion
of the ring is folded over the plane of the phenanthroline
unit, (b) the lone pairs on both S-donors adopt exoden-
tate orientation pointing out of the macrocyclic cavity.
Compared to L1 and L3 [20], the replacement of the
O(S)-donor with a methylene group in L2 causes a slight
increase in the flexibility of the aliphatic chain connected
to the phenanthroline unit. In fact, the torsion angles
S(2)–C–C–X and X–C–C–S(1), which in the case of L1
(X = O) and L3 (X = S) assume anti arrangements in
the most stable calculated conformers [20], can also as-
sume gauche disposition in L2 (X = CH2). Furthermore,
the torsion angles about the C–X bonds, which are one
generally gauche and one anti in L1 and L3, can be both
anti and both gauche for the most stable calculated con-
formers for L2. A less regular trend is also observed for
the torsion angles about the C–S(1) and C–S(2) bonds,
which are generally gauche in the most stable calculated
conformers of L1 and L3 [20]. Interestingly, the confor-
mation adopted by the major disorder component in L2
(Fig. 1) is very similar to the most stable calculated con-
former, while that adopted by the minor disorder com-
ponent is practically similar to that calculated for the
sixth most stable conformer.
0.4
0.2
0.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
[L]/[CuII]
Fig. 2. Absorbance (370 nm) vs L/CuII molar ratio plots for EtOH
solutions at 25 ꢁC of CuII (2 · 10ꢀ4 M) and L1 (r) L2 (m) or L3 (d)
immediately after mixing the reactants.
ever, the transformation of the [Cu(L)]2+ species is slow
enough to allow confirmation of their 1:1 stoichiometry
by continuous variations plots in EtOH under similar
experimental conditions and to allow their isolation at
the solid state.
In fact, whatever the CuII/L molar ratio used (1:1 or
1:2), the reaction between L1–L3 and Cu(ClO4)2 Æ 2H2O
in EtOH/CH2Cl2 always resulted in solid products con-
taining the 1:1 complexes, as demonstrated by elemen-
tal analysis, FAB mass spectrometry (see Section 2),
and X-ray structural analysis for [Cu(L1)(ClO4)]ClO4
3.2. Complexation
(1) and [Cu(L2)(ClO4)]ClO4 Æ 1 H2O (2). Several at-
The reaction of L1–L3 with CuII has been studied
spectrophotometrically both in EtOH and in MeCN.
Immediately upon mixing the reactants, a blue-green
colour is observed (kmax = 650 nm). In time, this colour
disappears and solutions turn to yellow. This behav-
iour is observed for all three ligands in EtOH and
MeCN, and for solutions having different L/CuII molar
ratios (from 0.2 to 4). However, the blue-green colour
persists for several hours if the L/CuII molar ratio is
lower than 1.
In order to understand the stoichiometry of the CuII
species responsible for the blue-green colour, due to
their instability in the presence of excess ligands, we car-
ried out fast spectrophotometric titrations of CuII with
the ligands in EtOH or MeCN by recording the spectra
immediately after mixing the reactants. The absorbance
(370 nm)/molar ratio plots show a sharp inflection point
at the 1:1 (L/CuII) molar ratio (Fig. 2) and thus indicate
a quantitative formation of [Cu(L)]2+ (L = L1–L3) com-
plexes. It is interesting to observe that further addition
of the ligands beyond the 1:1 L/CuII molar ratio led to
a gradual small decrease in absorbance at 370 nm,
according to the observed instability of the species
[Cu(L)]2+ in the presence of excess ligands; this effect is
particularly evident in the case of L2 (see Fig. 2). How-
2
tempts were made to isolate 1:2 CuII/L complexes at
the solid state using different starting copper(II) metal
salts and solvents. Only from the reaction of L3
with CuCl2 Æ 2H2O, and after addition of excess of
NH4PF6 to the reaction mixture, was the complex
[Cu(L3)2][PF6]2 (4) separated as a red microcrystalline
solid. Unfortunately, all attempts to grow good quality
single crystals for this compound failed. The complex-
ing ability of L1–L3 was also tested toward stoichiome-
tric amounts of the CuI salt [CuCl(PPh3)3] in an EtOH/
CH2Cl2 mixture, and only in the case of L3 was it pos-
sible to isolate the complex [Cu(L3)(Cl)(PPh3)] (5), of
which good quality crystals suitable for X-ray diffrac-
tion studies were grown by slow evaporation of the
reaction mixture.
3.3. X-ray diffraction analysis
In 1 (Fig. 3, Table 2), L1 acts as an N2S2O-donor,
encapsulating the copper(II) ion within a cavity having
a square-based pyramidal geometry with the equatorial
positions occupied by the two N-donors of the phenan-
throline moiety [Cu–N 1.969(3) and 1.964(3) A] and by
the two S-donors of the aliphatic linker [Cu–S 2.361(1)
˚

2408
M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
Table 2
1
˚
Selected bond lengths (A) and angles (ꢁ) for [Cu(L )(ClO4)]ClO4 (1),
[Cu(L2)(ClO4)]ClO4 Æ 21 H2O (2) and [Cu(L3)(Cl)(PPh3)] (5)a
[CuL1(ClO4)]
ClO4 (1)
[Cu(L2)(ClO4)]
ClO4 Æ 12 H2O (2)b
[CuL3(Cl)
(PPh3)] (5)
Cu–S(1)
Cu–S(2)
Cu–N(1)
Cu–N(2)
Cu–X(1)
Cu–X(2)
2.361(1)
2.364(1)
1.969(3)
1.964(3)
2.304(3)
2.648(5)
2.347(2)
1.956(5)
2.750(12)
2.140(4)
2.120(4)
2.226(2)
2.263(2)
N(1)–Cu–N(2)
N(1)–Cu–N(1i)
N(1)–Cu–S(1)
N(1)–Cu–S(2)
N(1)–Cu–S(1i)
N(1)–Cu–X(1)
N(1)–Cu–X(2)
N(2)–Cu–S(1)
N(2)–Cu–S(2)
N(2)–Cu–X(1)
N(2)–Cu–X(2)
S(1)–Cu–S(2)
S(1)–Cu–S(1i)
S(1)–Cu–X(1)
S(1)–Cu–X(2)
S(2)–Cu–X(1)
S(2)–Cu–X(2)
X(1)–Cu–X(2)
82.48(13)
77.60(16)
83.0(3)
82.9(1)
82.82(10)
165.93(10)
163.7(2)
112.7(2)
84.7(2)
99.42(12)
82.09(14)
164.32(10)
83.45(10)
95.58(12)
98.46(13)
111.21(5)
102.04(11)
118.21(12)
117.88(12)
115.21(11)
109.8(9)
108.4(2)
74.0(2)
81.49(8)
84.86(10)
82.19(9)
Fig. 3. ORTEP of the [Cu(L1)(ClO4)]+ complex cation in 1 with the
adopted numbering scheme. Displacement ellipsoids are drawn at 30%
probability and hydrogen atoms are omitted for clarity.
99.76(11)
165.94(11)
42.4(4)
118.26(6)
No significant interactions between CuI and S-donors are present in 5:
˚
and 2.364(1) A], and the axial position occupied by the
˚
Cu–S(1) 4.728(2), Cu–S(2) 4.689(2) A.
a
II
O-donor [Cu–O 2.304(3) A]. The Cu ion lies in the
˚
X(1) = O(1), O(5)ClO3 (2), PPh3 (5); X(2) = O(1)ClO3 (1),
O(5i)ClO3 (2), Cl (5).
plane defined by the atoms N(1), N(2), S(1), and S(2),
and an overall pseudo octahedral coordination sphere
is reached at the metal centre by interaction of a per-
chlorate ion acting as a monodentate ligand at the
coordination site left free by the pentadentate macrocy-
b
The complex cation [Cu(L2)(ClO4)]+ in 2 lies across a mirror plane,
i = x, ꢀy, z.
˚
˚
clic ligand [Cu–O(1) 2.648(5) A, O–Cu–O(1)
0.140 A out of the mean plane defined by thꢀe
165.94(11)ꢁ]. L1 therefore adopts a folded conformation
typical for this type of macrocycles with the aliphatic
chain of the ring tilted over the plane containing the
phenanthroline unit. Interestingly, the L1 donation to
CuII is comparable with that of [15]aneN2OS2 observed
in [Cu([15]aneN2OS2)]2+: in this latter complex, the
more flexible aliphatic macrocyclic ligand imposes a
similar square-based pyramidal N2S2O environment at
the metal centre with metal–donor atom distances
atoms N(1), N(1i), S(1), and S(1i) towards the ClO4
ligand.
The reaction of L3 with CuCl(PPh3)3 in EtOH/
CH2Cl2 afforded the complex [Cu(L3)(Cl)(PPh3)] (5)
(Fig. 5, Table 2). The CuI ion is coordinated by the
3
˚
nitrogen donors of L [Cu–N 2.140(4), 2.120(4) A], a
˚
PPh3 unit [Cu–P 2.226(2) A], and a chloride ligand
˚
[Cu–Cl 2.263(2) A] to complete a distorted tetrahedral
coordination sphere. The S-donors in the aliphatic chain
of L3 remain uncoordinated. The rigidity of the phenan-
throline unit forces the angle N(1)–Cu–N(2) to assume a
value of 77.60(16)ꢁ, thus allowing the P–Cu–Cl angle to
open to the value of 118.26(6)ꢁ. The metal ion is located
under the plane of the phenanthroline moiety with the
formation of a five-membered chelate ring showing a
pseudo-boat conformation. The dihedral angle between
the plane of the phenanthroline moiety and the plane
containing the atoms N(1), Cu, and N(2) is 21ꢁ as com-
pared to values not greater than 8ꢁ observed for other
CuI complexes with phenanthroline-based ligands (from
a recent search of the Cambridge Crystallographic
Database) [29].
˚
(Cu–N 1.97, Cu–S 2.31, Cu–O 2.28 A) very similar to
those found in 1 [28].
Fig. 4 shows the structure of the complex cation
[Cu(L2)(ClO4)]+ in 2 obtained from the reaction of cop-
per(II) perchlorate with L2 in EtOH/CH2Cl2. Selected
bond distances and angles are given in Table 2. The
complex cation sits on a mirror plane and the CuII ion
resides in a distorted square-based pyramidal N2S2O2
environment with the basal positions occupied by the
four donors of the macrocyclic ligand and the apical site
occupied by a bidentate perchlorate group. The Cu–N
and Cu–S bond lengths are comparable to those ob-
served in [Cu(L1)]2+ and the metal centre is displaced

M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
2409
Fig. 5. ORTEP of [Cu(L3)(Cl)(PPh3)] (5) with the adopted numbering
scheme. Displacement ellipsoids are drawn at 30% probability and
hydrogen atoms are omitted for clarity.
tal (LMCT) charge-transfer transition, respectively
[16,30,31].
Fig. 4. ORTEP of the [Cu(L2)(ClO4)]+ complex cation in 2 with the
adopted numbering scheme. Displacement ellipsoids are drawn at 30%
probability and hydrogen atoms are omitted for clarity.
The blue band at about 600 nm can be attributed to a
d–d electronic transition and the fairly high molar
extinction coefficient values observed (e = 250–
400 Mꢀ1 cmꢀ1, see Table 3) can be justified by the low
symmetry of the complexes. However, it is opportune
to remember that a band at about 600 nm with a much
higher molar extinction coefficient (e = 2000–
6000 Mꢀ1 cmꢀ1)1 is typical of the electronic spectra of
Type-1 blue copper proteins and was assigned to a
3.4. Spectral properties of complexes 1–4
The electronic spectral features of complexes 1, 2, and
3 in MeCN solution are very similar (Table 3). The
intense band observed at 278 nm (e = 28 300–
29 600 Mꢀ1 cmꢀ1
) and the one at about 360 nm
(e = 4000–4400 Mꢀ1 cmꢀ1) can be assigned to the phe-
nanthroline p ! p* and to a S(r) ! CuII ligand-to-me-
S
Cys(p) ! CuII transition [16,30,31]. Therefore, a contri-
bution of a S(p) ! CuII LMCT transition to the band at
Table 3
Spectral and electrochemical data for the isolated CuII complexes 1–4 in MeCN solutiona
Complex
Electronic spectra k/nm (e/Mꢀ1cmꢀ1
)
EPR spectrab
E1/2 (V)
Powder
Frozen MeCN solutionc
[Cu(L1)(ClO4)]ClO4(1)
278 (29 600)
361 (4400)
639 (400)
gi 2.181
g^ 2.049
gi 2.196
g^ 2.038
Ai 170.2
+0.33
[Cu(L2)(ClO4)]ClO4 Æ 21 H2O (2)
[Cu(L3)](ClO4)2 (3)
278 (29 100)
360 (4400)
630 (400)
gi 2.168
g^ 2.054
gi 2.182
g^ 2.033
Ai 171.7
+0.33
+0.42
+0.44
278 (28 300)
355 (4000)
592 (250)
gi 2.151
g^ 2.042
gi 2.191
g^ 2.036
Ai 168.6
[Cu(L3)2](PF6)2 (4)
275 (10 800)
377 (3400)
467sh (1100)
750 (140)
gi 2.169
g^ 2.065
Ai 173
gi 2.175
g^ 2.040
Ai 180
a
All measurements were carried out in MeCN rather than in EtOH solutions for solubility reasons.
Ai values are in units of 10ꢀ4 cmꢀ1
b
c
.
MeCN solutions obtained by mixing L1, L2 or L3 and Cu(ClO4) Æ 2H2O in 1:1 molar ratio; in the case of 4, the solution was prepared by dissolving
the solid complex in MeCN.

2410
M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
about 600 nm observed for 1–3 in MeCN might also be
hypothesised.
nated to the metal centre, probably due to the coordina-
tion of MeCN molecules.
The UV–Vis spectrum [Cu(L3)2](PF6)2 (4) in
MeCN is quite different from those recorded for 1–3,
in fact besides a d–d electronic transition at 750 nm
The gi- and Ai-values recorded for 1, 2, and 3 as fro-
zen MeCN glasses can be compared with those recorded
as frozen EtOH/H2O glasses for CuII complexes with ali-
phatic N2S2-donor macrocycles with a distorted square-
planar coordination geometry at the metal centre: i.e.,
cis-[12]aneN2S2 (gi = 2.174, Ai = 152), cis-[14]aneN2S2
(gi = 2.145, Ai = 170), and cis-[16]aneN2S2 (gi = 2.172,
Ai = 151) [16].
In any case, these EPR parameters are significantly
different from those reported for Type-1 blue copper
proteins (gi = 2.21–2.29, Ai = 35–63) [1], thus suggesting
a higher delocalisation for the unpaired d-electron in the
CuII-macrocyclic systems including those reported here.
In fact, the structures of native blue copper proteins typ-
ically show a much longer Cu–SMet bond distance than
that shown by Cu–SCys, thus creating a peculiar tetrahe-
dral distorted coordination geometry around the metal
centre.
(e = 140 Mꢀ1 cmꢀ1
)
a
shoulder at 467 nm (e =
1100 Mꢀ1 cmꢀ1) is also present.
The X-band EPR spectra of 1, 2, 3, and 4 were re-
corded as solid powders and as frozen (77 K) MeCN
glasses (Table 3). The EPR spectra of 1, 2, and 3 mea-
sured as powder are clearly axial and no hyperfine cou-
pling to the metal centre is observed [32–34]. In the case
of 4, the axial EPR spectrum shows three of the four ex-
pected hyperfine signals in the parallel region, with the
fourth being hidden under the g^ region lines (Fig. 6).
Furthermore, for 4, a superhyperfine coupling can be
observed in the perpendicular region of the spectrum,
which is consistent with the presence of not less than
three N-donors in the coordination sphere of the metal
centre. Although very useful, this information does not
allow to fully elucidate the coordination sphere around
the metal centre in 4. Interestingly, the gi-value de-
creases on passing from 1 to 3, which reflects the substi-
tution of an O-donor with a S-donor in the coordination
sphere around the metal centre [35]. The EPR spectra
measured as frozen MeCN glasses show isotropic signals
with giso values of 2.093, 2.092, and 2.077 for 1, 2, and 3,
respectively. In the case of 4, the EPR spectrum as fro-
zen MeCN glass is very similar to that recorded as pow-
der (Table 3). In order to better understand the nature of
complexes 1, 2, and 3 in MeCN solution, the EPR spec-
tra for mixtures of L1, L2 or L3 and Cu(ClO4)2 in molar
ratio 1:1 were recorded. For all three cases, the recorded
EPR spectrum is typical of mononuclear CuII complexes
with a tetragonal or square-based pyramidal coordina-
tion geometry at the metal centre (Table 3). Only three
of the four expected hyperfine lines are observed in the
parallel region; for 1 and 2, and less marked for 3, a
superhyperfine coupling is also present in the perpendic-
ular region, which accounts for the presence of two N
atoms in 1 and at least three N atoms in 2 and 3 coordi-
3.5. Cyclic voltammetry
Cyclic voltammetry of [Cu(L1)(ClO4)]ClO4 (1),
[Cu(L2)(ClO4)]ClO4 Æ 1 H2O (2), [Cu(L3)](ClO4)2 (3) and
2
[Cu(L3)2](PF6)2 (4) at platinum electrodes in MeCN
shows for each of them a quasi-reversible one-electron
(coulometrically tested) reduction at E1/2-values (vs
Ag/AgCl) of +0.33 (Epa ꢀ Epc = 90), +0.33 (Epa ꢀ Epc
=
=
90), +0.42 (Epa ꢀ Epc = 70), and +0.44 V (Epa ꢀ Epc
100 mV) for 1, 2, 3, and 4, respectively, that can be
assigned to the CuII/CuI couple (Table 3).
Despite the different coordination environment at the
metal centre in 1 and 2, the CuII/CuI redox potentials for
these two complexes are the same. The E1/2-value re-
corded for 3, instead, is significantly higher than those
recorded for 1 and 2, according to the increased number
of coordinated S-donors in 3 that stabilises the metal
centre in its reduced CuI form [14]. An increase in the
CuII/CuI redox potential has also been observed for
complexes with aliphatic macrocycles on passing from
a N4- to a N2S2-, and to a S4-donor set [14]. The redox
potential recorded for 2 is considerably higher than the
E1/2-values recorded in MeCN for CuII complexes with
the aliphatic N2S2-donor macrocycles cis-[12]aneN2S2
(E1/2 = 0.164 V vs SHE), cis-[14]aneN2S2 (E1/2
=
0.012 V vs SHE), and cis-[16]aneN2S2 (E1/2 = 0.368 V
vs SHE) [16], presumably due to the p-acceptor proper-
ties of the phenanthroline sub-unit, which stabilise the
reduced CuI form. However, the E1/2-value for 2 falls
in the range 0.184–0.680 V (vs SHE in water solution)
observed for the Eꢁ of a series of Type-1 blue-copper
proteins having an N2S2 trigonal pyramidal environ-
ment at the metal centre [6]. In the case of 1, the value
of E1/2 is out of the range 0.285–0.310 V (vs SHE in
water solution) generally reported for the Eꢁ of Type-1
Fig. 6. X-band EPR spectra recorded at the solid state for
[Cu(L3)2](PF6)2 (4). Ai hyperfine splitting and gi value are shown.
Superhyperfine resonance lines are observed (see text).

M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
2411
azurin proteins characterised by an N2S2O trigonal
bipyramidal environment at the metal centre [6]. The
E1/2-value recorded for 4 and assigned to a CuII/CuI
reduction process is very close to that of +0.603 (vs
SHE in water solution) recorded for [Cu(2,9-
dmphen)2]2+ (2,9-dmphen = 2,9-dimethylphenanthro-
line) [36]. The mono-electron reduction of 4 was also
monitored by UV–Vis spectroscopy. The in situ electro-
chemical reduction of [Cu(L3)2](PF6)2 (4) was carried
properties and the flexibility of the aromatic fragments
considered.
Acknowledgements
`
We thank the Ministero Italiano dellꢀUniversita e
della Ricerca Scientifica e Tecnologica (MIUR) for
financial support (FIRB project RBAU01HFNS). L.E.
and V.M. thank the financial assistance of the Spanish
Government provided by the Project BQU2000-0233.
out in MeCN (0.1 M NBun BF4) at 298 K. The reduc-
4
tion proceeds with increasing of the band at 467 nm
and a decreasing of the band at 750 nm (see above) re-
corded for the starting solution of 4.
Appendix A. Supplementary data
4. Conclusions
Full crystallographic data for the reported com-
pounds have been deposited with the Cambridge Crys-
tallographic Data Centre, CCDC No. 213075-213078.
Copy may be obtained free of charge from The director,
CCDC, 12 Union Road, Cambridge CB2 1Ez, UK (fax:
+44-1223-336-033; e-mail: deposit@ccdc.cam.ac.uk or
http://www.ccdc.cam.ac.uk). The eight most stable con-
formers calculated for L2 and their torsion angles are
also available as supplementary material (Fig. S1, Table
S1). Supplementary data associated with this article can
be found, in the online version at doi:10.1016/
j.ica.2005.01.014.
The coordination chemistry towards CuII of L1, L2,
and L3, having, respectively, N2S2O-, N2S2-, and N2S3-
donor sets, and a phenanthroline sub-unit within the
macrocyclic framework, has been studied. All three li-
gands form 1:1 CuII/L complexes both in solution
(MeCN and EtOH) and in the solid state. X-ray diffrac-
tion analysis shows that the metal centre encapsulated
within the macrocyclic ligands in a distorted square-
based pyramidal or pseudo-octahedral coordination
geometry, due to the interaction with also counteranion
molecules. The ligands L1 and L2 have the right donor
set to mimic the coordination environment of Type-1
blue copper proteins. However, the corresponding 1:1
CuN2S2O and CuN2S2 complexes tend to have square-
based geometries in solution and in the solid state,
because of the interaction with solvents and counter-an-
ions, and because of the rigidity imposed by the phenan-
throline framework to the cyclic systems.
Despite this, 1:1 CuII complexes with L1 and L2 show
quasi-reversible redox potentials comparable to those
observed for blue copper proteins. Due to its p-acceptor
properties, the phenanthroline moiety certainly plays an
important role in stabilising the metal centre in its CuI
oxidation form. Because of the square-based nature of
the CuII complexes with L1 and L2, the recorded hyper-
fine splitting constant (Ai) is comparable to that ob-
served for complexes with aliphatic NxS4 ꢀ x-donating
macrocycles, but different from that observed for blue
copper proteins.
References
[1] A.G. Sykes, Adv. Inorg. Chem. 36 (1991) 377, and references
therein.
[2] P.M. Colman, H.C. Freeman, J.M. Guss, M. Murata, V.A.
Norris, J.A.M. Ramshaw, M.P. Ventakappa, Nature 272 (1978)
319.
[3] J.M. Guss, E.A. Merritt, R.P. Phizackerley, B. Hedman, M.
Murata, R.O. Hodgson, H.C. Freeman, Science 241 (1988) 806.
[4] E.T. Adman, S. Turley, R. Branson, K. Petratos, D. Banner, D.
Tsernoglov, D. Beppo, H. Watanabe, J. Biol. Chem. 264 (1989)
87.
[5] F.N. Baker, J. Mol. Biol. 203 (1988) 1071.
[6] H.B. Gray, B.G. Malmstro¨m, R.J.P. Williams, J. Biol. Inorg.
Chem. 5 (2000) 551.
[7] K.B. Yatsimirskii, V.V. Pavlishchuk, J. Coord. Chem. 37 (1996)
341, and references therein.
[8] P.L. Holland, W.B. Tolman, J. Am. Chem. Soc. 122 (2000) 6331.
[9] M. Vaidyanathan, R. Balamurugan, U. Sivagnanam, M. Pala-
niandar, J. Chem. Soc., Dalton Trans. (2001) 3498.
[10] H.C. Freeman, in: J.P. Laurent (Ed.), Coordination Chemistry,
vol. 21, Pergamon, Oxford, 1981, p. 29.
So far, the proposed models for blue copper proteins
which comprise macrocyclic ligands have been somehow
successful in reproducing the coordination environment
around the metal; however, a single CuII complex with
macrocyclic ligands, which shows redox, electronic,
and EPR spectral properties reasonably similar to those
of the native proteins, is still not available. The choice of
macrocyclic ligands containing hetero-aromatic sub-
units seems promising but it must be very well weighed
so as to reach an appropriate balance between the donor
[11] O. Farver, I. Pecht, Coord. Chem. Rev. 94 (1989) 17, and
references therein.
[12] A.J. Blake, M. Schro¨der, Adv. Inorg. Chem. 35 (1990) 1, and
references therein.
[13] G. Reid, M. Schro¨der, Chem. Soc. Rev. 19 (1990) 239.
[14] V. McKee, Adv. Inorg. Chem. 40 (1993) 323.
[15] K.K. Nanda, A.W. Addison, R.J. Butcher, M.R. McDevitt, T.
Nageswara Rao, E. Sinn, Inorg. Chem. 36 (1997) 134.
[16] L. Siegfried, T.A. Kaden, Helv. Chim. Acta 67 (1984) 29.

2412
M. Arca et al. / Inorganica Chimica Acta 358 (2005) 2403–2412
[17] A.J. Blake, F. Demartin, F.A. Devillanova, A. Garau, F. Isaia,
V. Lippolis, M. Schro¨der, G. Verani, J. Chem. Soc., Dalton
Trans. (1996) 3705.
[26] Spartan Version 5.0, Wavefunction Inc., 18401 Von Karman
Ave., Ste. 370, Irvine, CA 92612, USA.
[27] S. Potmann, H.P. Luthi, Chimia 54 (2000) 766.
[28] F. Arnaud-Neu, M.J. Shwing-Weill, R. Louis, R. Weiss, Inorg.
Chem. 18 (1979) 2956.
[18] F. Contu, F. Demartin, F.A. Devillanova, A. Garau, F. Isaia, V.
Lippolis, A. Salis, G. Verani, J. Chem. Soc., Dalton Trans. (1997)
4401.
[29] (a) F.H. Allen, O. Kennard, Chem. Des. Autom. News 8 (1)
(1993) 1;
`
[19] A.J. Blake, J. Casabo, F.A. Devillanova, L. Escriche, A. Garau,
F. Isaia, V. Lippolis, R. Kivekas, V. Muns, M. Schro¨der, R.
Sillanpa¨a¨, G. Verani, J. Chem. Soc., Dalton Trans. (1999)
1085.
(b) F.H. Allen, O. Kennard, Chem. Des. Autom. News 8 (1)
(1993) 31, Cambridge Structural database Version 5.23, released
April 2002.
`
[20] M. Arca, A.J. Blake, J. Casabo, F. Demartin, F.A. Devillanova,
[30] J.V. Dagdigian, V. McKee, C.A. Reed, Inorg. Chem. 21 (1982)
1332.
A. Garau, F. Isaia, V. Lippolis, R. Kivekas, V. Muns, M.
Schro¨der, G. Verani, J. Chem. Soc., Dalton Trans. (2001) 1180.
[21] M.C. Aragoni, M. Arca, F. Demartin, F.A. Devillanova, A.
Garau, F. Isaia, V. Lippolis, F. Jalali, U. Papke, M. Shamsipur,
L. Tei, A. Yari, G. Verani, Inorg. Chem. 40 (2002) 323.
[22] L.-Y. Chunng, E.C. Constable, M.S. Khan, J. Lewis, P.R.
Raithby, M.D. Vargas, J. Chem. Soc., Chem. Comm. (1984) 1425.
[23] SADABS Area-Detector Absorption Correction Program, Bruker
AXS, Inc., Madison, WI, USA, 2000.
[31] V.M. Miskowski, J.A. Thich, R. Solomon, H.J. Schugar, J. Am.
Chem. Soc. 98 (1976) 8344.
[32] I.M. Procter, B.J. Hathaway, D.E. Billing, R. Dudley, P. Nicholls,
J. Chem. Soc. A (1969) 1192.
[33] D.E. Billing, R. Dudley, B.J. Hathaway, A.A.G. Tomlinson, J.
Chem. Soc. A (1971) 691.
[34] A.W. Addison, E. Sinn, Inorg. Chem. 22 (1983) 1225.
[35] J. Peisach, W.E. Blumberg, Arch. Biochem. Biophys. 165 (1974)
691.
[24] G.M. Sheldrick, SHELXS 97, Acta Crystallogr., Sect A 46 (1990)
467.
[36] B.J. Hathaway, in: G. Wilkinson (Ed.), Comprehensive Coordi-
nation Chemistry, vol. 5, 1987, p. 533.
[25] G.M. Sheldrick, SHELXL 97, Universita¨t Go¨ttingen, 1997.