Exchange coupling in a bis(heterodinuclear) [CuIINiII]2 and a linear heterotrinuclear complex CoIIICuIINiII. Synthesis, structures and properties

Article abstract of DOI:10.1039/a908426f

Two heterometallic complexes [{Cu^II(HLOX)Ni^II(N₃)}₂] and [(tmtacn)Co^III(μ-OH)Cu^II(LOX)Ni^II(OH ₂)₂]-[ClO₄]₂ (tmtacn = 1,4,7-trimethyl-1,4,7-triazacyclononane) containing the same [Cu^IINi^II] core embedded in an unsymmetrical dicompartmental imine-oxime ligand H₄LOX have been synthesized and characterized. Their crystal structures show that the Cu^II resides at the N(oxime)₂O(phenolate)₂ site and assumes a planar geometry. The Ni_II is six-co-ordinated and bound to an N₂O₄ donor array comprising two iminonitrogens, two phenolate oxygens and two axially co-ordinated H₂O molecules. The cobalt(III) is low spin and six-co-ordinated. In DMF solution at 10 K the EPR spectra of the complexes exhibit a spin-doublet ground state with inverted g values which demonstrate the delocalization of the unpaired electron over the CuNi core. Magnetic susceptibility measurements over the range 2-290 K confirm that the paramagnetic nickel(II) and copper(II) centres are antiferromagnetically coupled, with values for the exchange coupling constant J through the phenolate oxygens of -115 cm^-1 and -130 cm^-1 respectively. Considering the dimer as a single tetranuclear unit, the coupling constant J′ through the path Cu-N-O-Ni is very small (≈1 cm^-1) but positive (ferromagnetic coupling). The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/a908426f

View Article Online / Journal Homepage / Table of Contents for this issue
Exchange coupling in a bis(heterodinuclear) [CuIINiII]2 and a linear
heterotrinuclear complex CoIIICuIINiII. Synthesis, structures and
properties
Cláudio Nazari Verani, Eva Rentschler, Thomas Weyhermüller, Eckhard Bill and
Phalguni Chaudhuri*
Max-Planck-Institut für Strahlenchemie, Stiftstraße 34-36, D-45470 Mülheim an der Ruhr,
Germany. E-mail: chaudh@mpi-muelheim.mpg.de
Received 22nd October 1999, Accepted 3rd December 1999
Two heterometallic complexes [{CuII(HLOX)NiII(N3)}2] and [(tmtacn)CoIII(µ-OH)CuII(LOX)NiII(OH2)2]-
[ClO4]2 (tmtacn = 1,4,7-trimethyl-1,4,7-triazacyclononane) containing the same [CuIINiII] core embedded in an
unsymmetrical dicompartmental imine-oxime ligand H4LOX have been synthesized and characterized. Their
crystal structures show that the CuII resides at the N(oxime)2O(phenolate)2 site and assumes a planar geometry.
The NiII is six-co-ordinated and bound to an N2O4 donor array comprising two iminonitrogens, two phenolate
oxygens and two axially co-ordinated H2O molecules. The cobalt() is low spin and six-co-ordinated. In DMF
solution at 10 K the EPR spectra of the complexes exhibit a spin-doublet ground state with “inverted” g
values which demonstrate the delocalization of the unpaired electron over the CuNi core. Magnetic susceptibility
measurements over the range 2–290 K confirm that the paramagnetic nickel() and copper() centres are
antiferromagnetically coupled, with values for the exchange coupling constant J through the phenolate oxygens
of Ϫ115 cmϪ1 and Ϫ130 cmϪ1 respectively. Considering the dimer as a single tetranuclear unit, the coupling
constant JЈ through the path Cu–N–O–Ni is very small (≈1 cmϪ1) but positive (ferromagnetic coupling).
As an extension of the strategy of using “metal complexes”
Introduction
as bridging groups, we envisaged the possibility of preparing a
series of linear trinuclear complexes containing three different
metals MAMBMC,15 where two metal ions, e.g. MA and MB, are
embedded in a dicompartmental imine-oxime ligand H4LOX,
synthesized by stepwise metal template condensation of
the dialdehyde 2,6-diformyl-4-methylphenol and 1,3-diamino-
propane, followed by oxime formation through hydroxylamine.
Such a building block used in this work is shown below. Some
Exchange-coupled polynuclear complexes,1–7 particularly
heterometallic systems,4–6 are of special interest due to their
relevance to different branches of natural sciences, physics,
chemistry and biology. There is an impressive diversity of spin-
coupled structures in biology,8,9 e.g. dicopper sites in hemo-
cyanin and tyrosinase, diiron() sites in methemerythrin,
ribonucleotide reductase, and heterobimetallic sites of FeIII and
CuII in respiratory cytochrome oxidase which catalyses the 4e
reduction of dioxygen to water in the mitochondria of eukary-
otic cells. Active sites involving more than one metal centre
in metalloproteins have elicited the interest of bioinorganic
chemists in such interactions. In another area of research
the intimate relationship between spin coupling and mol-
ecular structure has fostered the emergence of molecular
magnetism2–4 as a multidisciplinary field. The fundamental
understanding regarding the factors that determine the spin
states of polynuclear transition metal complexes owes much to
the study of model compounds where magnetostructural corre-
lations can be established in a systematic way. Thus exchange-
coupled polymetallic systems constitute a common ground for
at least two areas of current interest, molecular magnetism and
metal sites in biology. Of particular concern in this context is
the development of directed routes to specific complexes, par-
ticularly those in which the metals are connected by “robust”
ligands so that fragmentation of the complexes in their sub-
sequent reactions is inhibited. One particularly successful route
involves the use of “metalloligands”10,11 in which the ligands
already bound to one metal have free co-ordination sites that
can bind a second metal of the same or different kind. We have
previously used the same strategy utilizing “metal oximates”
building blocks as ligands, to synthesize various homo- and
hetero-metal complexes containing two,12 three13 or four14
metal centres. Thus we have reported a series of complexes of
the type MAMB, MAMBMA, MAMBMBMA and (MA)2(µ3-O)2-
(MB)2, which have been proved to be ideal materials for the
investigations of exchange mechanisms.
examples of dicopper complexes with this dicompartmental
Schiff base oxime ligand have been reported.16 Homodinuclear
complexes of the dioxime ligand 2-hydroxy-5-methylbenzene-
1,3-dicarbaldehyde dioxime17 and heterodinuclear NiIICuII
complexes of some related ligands have also been described.18–22
The above synthetic strategy has successfully been applied to
prepare several linear trinuclear heterometal complexes and
in a short communication15 we have reported the dication
[(tmtacn)FeIII(Cl)CuII(CH3OH)(LOX)NiII(CH3OH)2]2ϩ
1,
where tmtacn is the tridentate macrocyclic amine ligand 1,4,7-
trimethyl-1,4,7-triazacyclononane. Complex 1 exhibits a nearly
linear FeIIICuIINiII core in which the terminal metal centres FeIII
J. Chem. Soc., Dalton Trans., 2000, 251–258
This journal is © The Royal Society of Chemistry 2000
251

View Article Online
Scheme 1
and NiII adopt an octahedral co-ordination geometry, whereas
Treatment of the heterotrinuclear complex FeIIICuIINiII
1
with an excess of azide results also in the formation of 2 and
[FeIII(tmtacn)(N3)3] in reasonable yield (Scheme 1).
the central CuII is in a square-pyramidal environment. The
subject of this paper is the syntheses, structures, magnetic
and spectroscopic characterization of tetranuclear [{CuII-
(HLOX)NiII(N3)}2] 2 and trinuclear [(tmtacn)CoIII(µ-OH)-
CuII(LOX)NiII(OH2)2][ClO4]2 3. Complex 2, although present
as a “dimer of dimer”, [CuIINiII]2, in the solid state, dissociates
to the heterodimetallic CuIINiII species in solution, as will be
evident later from the EPR measurements. Complex 3 contains
the same dimetallic CuIINiII core as in 1 and 2, but the capping
paramagnetic (tmtacn)FeIII fragment in 1 has been replaced by
the diamagnetic (tmtacn)CoIII capping agent.
The red-brown solution obtained from cobalt() chloride
and the macrocyclic amine in methanol–water reacting with
[NiIICuII(HLOX)]Cl affords upon addition of the perchlorate
counter ion in the presence of air dark green crystals of 3,
containing the CoIIICuIINiII core. The function of added tri-
ethylamine is to provide a basic medium needed for the
deprotonation of the O ؒ ؒ ؒ H ؒ ؒ ؒ O groups present in solid [NiII-
CuII(HLOX)]ϩ and for the oxidation of CoII to CoIII by air. In
the infrared spectrum of 3 a broad strong band is observed at
3407 cmϪ1 due to the ν(O–H) stretching mode. The strong band
at 1081 cmϪ1 and the sharp band at 626 cmϪ1, due to the
antisymmetric stretch and antisymmetric bend of perchlorate
ions, are indicative of their unco-ordinated nature. The other
important strong bands are at 1636 and 1121 cmϪ1 and are
assignable to ν(CN) and ν(NO) vibrations, respectively.
The optical spectra of complexes 1, 2 and 3 have been meas-
ured in the range 200–1400 nm in dry dimethylformamide. On
the basis of their high absorption coefficients and the relative
sharpness, the bands can be ascribed to charge-transfer transi-
tions within the ligands or metal-to-ligand charge transfer
(MLCT). The bands at 1070, 583 nm and a shoulder at 698 nm
for 2, 983, 750 and 585 nm for 3 are thought to be due to ligand-
field (d–d) transitions. For 1, only one d-d transition at 743 nm
has been observed.
Results and discussion
A straightforward, clean and relatively high-yield synthetic
route to the pure heteronuclear complex [{CuII(HLOX)NiII-
(N3)}2] 2 is outlined in Scheme 1. More than a 5-fold excess of
NH2OH leads to the formation of intractable matter and hence
was avoided. Similarly the reaction temperature was kept below
60 ЊC assuring the avoidance of metal scrambling. All attempts
to isolate the dioxime H2LOX were unsuccessful. The IR spec-
trum of 2 exhibits a strong sharp band at 2042 cmϪ1 which is
assigned to the antisymmetric ν(N–N) stretching frequency of
the azide groups, and a medium peak at 1314 cmϪ1 (νsym(N–N)),
which has not been identified unambiguously due to over-
lapping bands in this region of the co-ordinated imine-oxime
ligand. The strong band at 1636 cmϪ1 is unambiguously assign-
able to the ν(CN) vibration. The bands of medium strength at
1113, 1086 and 1069 cmϪ1 are attributed to the N–O stretching
vibrations from the oxime groups.
Description of the structures
FeIIICuIINiII 1. The structure has been reported in a previous
publication.15
252
J. Chem. Soc., Dalton Trans., 2000, 251–258

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for [{Cu(HLOX)-
Ni(N3)}2]ؒ2CH3OH
Cu–N(4)
Cu–O(4)
Ni–N(3)
Ni–O(3)
Ni–N(5)
O(1)–N(1)
1.9402(12)
1.9534(10)
2.0036(12)
2.0374(11)
2.2072(13)
1.386(2)
Cu–N(1)
Cu–O(3)
Ni–N(2)
Ni–O(4)
Ni–O(2)
O(2)–N(4)
1.9504(12)
1.9566(10)
2.0012(12)
2.0373(10)
2.1814(11)
1.3803(14)
N(4)–Cu–N(1)
N(1)–Cu–O(4)
N(1)–Cu–O(3)
N(2)–Ni–N(3)
N(3)–Ni–O(4)
N(3)–Ni–O(3)
N(2)–Ni–O(2)
O(4)–Ni–O(2)
N(2)–Ni–N(5)
O(4)–Ni–N(5)
O(2)–Ni–N(5)
Cu–O(3)–Ni
97.28(5)
91.22(4)
171.60(4)
99.53(5)
168.18(4)
91.79(4)
90.67(5)
92.62(4)
91.85(5)
84.31(4)
176.09(4)
101.14(4)
N(4)–Cu–O(4)
N(4)–Cu–O(3)
O(4)–Cu–O(3)
N(2)–Ni–O(4)
N(2)–Ni–O(3)
O(4)–Ni–O(3)
N(3)–Ni–O(2)
O(3)–Ni–O(2)
N(3)–Ni–N(5)
O(3)–Ni–N(5)
Cu–O(4)–Ni
166.03(4)
90.40(4)
80.59(4)
91.98(4)
168.67(4)
76.72(4)
89.94(4)
88.99(4)
92.59(5)
87.95(5)
101.25(4)
Fig. 2 Structure of the dication containing the CoIIICuIINiII units in
crystals of complex 3.
copper is bound to two nitrogens, N(1) and N(4), of the
oxime groups, one of which is protonated at O(1) and the other
deprotonated oxygen O(2) occupies the axial position of the
crystallographically related second nickel by an inversion
centre. The azide ions are bound only to nickel ions. In this
regard 2 differs from the reported23 [{Cu(oxpn)Ni(µ-NCS)-
(OH2)(tmen)}2]2ϩ, in which two dinuclear [Cu(oxpn)Ni(OH2)-
(tmen)] entities [oxpn = N,NЈ-bis(3-aminopropyloxamide)] are
linked by two Cu–SCN–Ni bridging units, related by an inver-
sion centre. In contrast, the bis(heterodinuclear) complex 2 is
built up of two heterodinuclear [Cu(HLOX)Ni(N3)] units, which
are held together by two deprotonated oxime oxygens. Each of
the oxygens, O(4) and O(2), is bound respectively to an axial
position of nickel ions. The bridging network NiO(3)O(4)Cu is
nearly planar with NiOCu angles of 101.2Њ and a Cu ؒ ؒ ؒ Ni
separation of 3.085 Å.
The nickel centre is co-ordinated to two azomethine nitrogen
atoms, N(2) and N(3), and two phenolate oxygen atoms O(3)
and O(4) in the equatorial plane. The Ni–N (azomethine) (aver-
age 2.003(2) Å) and Ni–O (phenolate) (average 2.037(1) Å)
bond lengths are very similar to those of the analogous com-
plexes reported previously by us14b and others.17a The deviation
of the nickel ion from the mean plane comprising NiN(2)-
N(3)O(3)O(4) is only 0.019 Å. The dihedral angle between the
nickel plane and the plane CuN(1)N(4)O(3)O(4) is 5.1Њ.
The copper centre adopts a square planar geometry by co-
ordinating to two oxime nitrogen atoms, N(1) and N(4), with
an average Cu–N bond length of 1.945(5) Å and to bridging
phenolate oxygens O(3) and O(4) with an average Cu–O dis-
tance of 1.955(3) Å. These values are in good agreement with
those in compounds containing the same ligand H4LOX with
cores like CuIICuII,16 FeIIICuIINiII 115 and CoIIICuIINiII 3 and
other comparable structures.20 The copper ion is displaced by
0.115 Å from the mean basal plane comprising N(1)N(4)-
O(3)O(4). The nearest Cu ؒ ؒ ؒ Ni distance is 3.085 Å in the ligand
plane, while the same between the ligand planes is 4.039 Å.
Fig. 1 (a) Molecular structure of [{CuII(HLOX)NiII(N3)}2] 2, showing
the protonated oximic oxygen. (b) A perspective view of 2 highlighting
the connectivity between the two CuIINiII units, together with two
solvent methanol molecules.
The N–O (average 1.383(3) Å) and C᎐N (average 1.286(6) Å)
᎐
bond lengths and C–N–O bond angle (average 114.2(9)Њ) of the
bridging oxime groups are found to be very similar to those of
other comparable structures.12–14
[{CuII(HLOX)NiII(N3)}2] 2. Complex 2, [{CuII(HLOX)NiII-
(N3)}2], shows a centrosymmetric tetranuclear structure [CuNi]2
(Fig. 1(a) and (b)) with both nickel() centres having six-
co-ordinated geometries NiN3O3, whereas both CuII in square-
planar CuN2O2 stereochemistries. A crystallographic centre of
inversion lies at the midpoint of the Ni ؒ ؒ ؒ Ni vector. Selected
bond lengths and angles are listed in Table 1. The neighbouring
nickel and copper ions in the ligand plane are bridged together
through the phenolate oxygen atoms, O(3) and O(4). Each
[(tmtacn)CoIII(ꢀ-OH)CuII(LOX)NiII(OH2)2][ClO4]2ؒCH3OHؒ
2H2O 3. Fig. 2 shows the structure of the complex dication in 3
containing the trinuclear core CoIII(l.s.)CuIINiII (l.s. = low spin).
Selected bond lengths are listed in Table 2. The co-ordination
geometry of the terminal cobalt, Co(1), is distorted octahedral
with three nitrogen atoms, N(1), N(2) and N(3), from the
facially co-ordinated macrocyclic amine, two oxygen atoms
O(2) and O(5) from the oxime ligands and a bridging hydroxide
ion, O(1) resulting in the fac CoN3O3 core. The average Co–O
J. Chem. Soc., Dalton Trans., 2000, 251–258
253

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for [(tmtacn)-
Co(µ-OH)Cu(LOX)Ni(OH2)2][ClO4]2ؒCH3OHؒ2H2O 3
Ni(1)–N(5)
Ni(1)–O(4)
Ni(1)–O(6)
Co(1)–O(5)
Co(1)–O(1)
Co(1)–N(1)
Cu(1)–N(7)
Cu(1)–O(3)
Cu(1)–O(1)
N(4)–O(2)
1.997(5)
2.012(4)
2.109(5)
1.880(4)
1.913(4)
1.994(5)
1.951(5)
1.967(4)
2.432(4)
1.367(6)
Ni(1)–N(6)
Ni(1)–O(3)
Ni(1)–O(7)
Co(1)–O(2)
Co(1)–N(3)
Co(1)–N(2)
Cu(1)–O(4)
Cu(1)–N(4)
N(7)–O(5)
2.002(6)
2.035(4)
2.113(5)
1.886(4)
1.992(5)
2.003(5)
1.958(4)
1.969(5)
1.366(6)
N(5)–Ni(1)–N(6)
N(6)–Ni(1)–O(4)
N(6)–Ni(1)–O(3)
N(5)–Ni(1)–O(6)
O(4)–Ni(1)–O(6)
N(5)–Ni(1)–O(7)
O(4)–Ni(1)–O(7)
O(6)–Ni(1)–O(7)
N(7)–Cu(1)–O(3)
N(7)–Cu(1)–N(4)
O(3)–Cu(1)–N(4)
O(4)–Cu(1)–O(1)
N(4)–Cu(1)–O(1)
O(5)–Co(1)–O(1)
O(5)–Co(1)–N(3)
O(1)–Co(1)–N(3)
O(2)–Co(1)–N(1)
N(3)–Co(1)–N(1)
O(2)–Co(1)–N(2)
N(3)–Co(1)–N(2)
Co(1)–O(1)–Cu(1)
Cu(1)–O(3)–Ni(1)
Cu(1)–O(4)–Ni(1)
97.8(2)
93.4(2)
170.4(2)
91.3(2)
89.3(2)
89.5(2)
89.8(2)
178.9(2)
169.9(2)
99.7(2)
89.6(2)
118.9(2)
75.3(2)
92.2(2)
91.1(2)
92.2(2)
88.2(2)
86.3(2)
92.1(2)
86.5(2)
98.1(2)
100.6(2)
101.8(2)
N(5)–Ni(1)–O(4)
N(5)–Ni(1)–O(3)
O(4)–Ni(1)–O(3)
N(6)–Ni(1)–O(6)
O(3)–Ni(1)–O(6)
N(6)–Ni(1)–O(7)
O(3)–Ni(1)–O(7)
N(7)–Cu(1)–O(4)
O(4)–Cu(1)–O(3)
O(4)–Cu(1)–N(4)
N(7)–Cu(1)–O(1)
O(3)–Cu(1)–O(1)
O(5)–Co(1)–O(2)
O(2)–Co(1)–O(1)
O(2)–Co(1)–N(3)
O(5)–Co(1)–N(1)
O(1)–Co(1)–N(1)
O(5)–Co(1)–N(2)
O(1)–Co(1)–N(2)
N(1)–Co(1)–N(2)
168.8(2)
91.5(2)
77.2(2)
89.1(2)
93.0(2)
91.7(2)
86.1(2)
90.0(2)
80.1(2)
164.7(2)
76.1(2)
110.4(2)
89.8(2)
93.2(2)
174.4(2)
87.9(2)
178.5(2)
174.3(2)
93.1(2)
86.8(2)
Fig. 3 Simulated and experimental X-band EPR spectra for a solid
sample of complex 2 at 10 K (microwave frequency 9.650 GHz; power
100 µW; modulation amplitude 11.43 G).
Fig. 4 X-Band EPR spectrum of complex 2 in DMF at 10 K (micro-
wave frequency 9.644 GHz; power 100 µW; modulation amplitude
11.43 G) together with the simulated spectrum.
and Co–N distances at 1.996(5) and 1.893(18) Å, respectively,
are consistent with a d6 low-spin electron configuration of the
cobalt centre.24 The Co–Cu–Ni skeleton is not linear with an
angle of 155.6Њ, whereas the angle Fe–Cu–Ni in 1 is 174Њ.15 The
intramolecular separations between the metal centres Co ؒ ؒ ؒ Cu
3.299, Cu ؒ ؒ ؒ Ni 3.081 and Co ؒ ؒ ؒ Ni 6.236 Å are in conformity
with the values observed earlier for comparable structures.15
The dihedral angle between the basal copper plane Cu(1)N(4)-
N(7)O(3)O(4) and the equatorial Ni plane Ni(1)N(5)N(6)-
O(4)O(3) is 175.4Њ. The corresponding dihedral angle in 1 is
very similar, 169.4Њ. Accordingly, the Cu ؒ ؒ ؒ Ni separations in 1
and 3 are not different; Cu ؒ ؒ ؒ Ni in 1 3.087 Å.15
The geometry of the central coper() centre is distorted
square pyramidal, with the elongated fifth bond (2.432 Å) to
axially co-ordinated oxygen atom O(1) of the µ-OH ligand. The
angles O(1)Cu(1)N(7) and O(1)Cu(1)N(4) are noticeably less
(76.1 and 75.3Њ) than 90Њ, thus making the CoN3O3 octahedron
tipped toward the equatorial copper plane Cu(1)N(4)N(7)-
O(4)O(3). Accordingly, the Co–Cu–Ni skeleton in 3 is not
linear in contrast to the Fe–Cu–Ni skeleton of 1. The metrical
parameters for the copper centre are very similar to those of the
copper dimer with the same Schiff-base oxime ligand.16
The terminal nickel ion, Ni(1), is co-ordinated equatorially
to the azomethine nitrogen atoms, N(5) and N(6), and two
bridging phenolate groups, O(3) and O(4), from the Schiff
base oxime ligand. The nickel centre adopts a six-co-ordinated
environment by interacting with two trans axially disposed
water molecules, O(6) and O(7). The nickel centre is displaced
by 0.022 Å from the mean basal plane comprising O(3)O(4)-
N(5)N(6). The equatorial Ni–N and Ni–O (phenolate) dis-
tances are nearly equal, average 2.000(6) and 2.024(12) Å,
respectively. The Ni–OH2 distances are rather long, average
2.111(5) Å, as has been observed earlier.17a,25 The bridging
angles Cu(1)–O(3)–Ni(1) and Cu(1)–O(4)–Ni(1) are nearly
equal 100.6(2) and 101.8(2)Њ, respectively. Thus the CuNi
centres embedded in the Schiff base oxime ligand can be
described as edge sharing between a square pyramid and an
octahedron.
EPR spectroscopy
Complex 2, [{CuII(HLOX)NiII(N3)}2], both as a solid and in
frozen DMF solution was subjected to EPR spectroscopy and
the spectra are shown in Figs. 3 and 4. Fig. 3 was obtained for a
powdered sample of 2 at 10 K and X-band frequency. A broad
single line emerges which can be simulated with parameters
given below. In DMF glass at 10 K the spectral resolution is
considerably enhanced. The EPR spectrum (Fig. 4) is of axial
pattern with slightly rhombic distortion with g⊥ seen at ≈2.26
together with a g|| ≈ 2.1 signal split by four copper hyperfine
lines. Both solid and solution spectra were simulated with
gx = 2.285, gy = 2.239, gz = 2.111, Ax = Ay = 7 × 10Ϫ4 cmϪ1 and
Az = 58 × 10Ϫ4 cmϪ1. The presence of hyperfine structure only
in the solution spectrum strongly suggests decomposition of the
tetranuclear species (dimer of dimer) into two dinuclear units in
the presence of a strongly co-ordinating solvent like DMF.
It is noteworthy that the relation g|| > g⊥ is typical of axially
symmetric d9 CuII(SCu = ¹) having one unpaired electron in a
¯
²
orbital26 whereas g|| < g⊥ is observed for CuII with the
2 Ϫ y2
dx
2
unpaired electron in a dz orbital. Axially symmetric 2 in solu-
tion exhibits features of S = ¹ with inverted g values (g|| < g⊥),
¯
²
keeping in mind in this regard that we are observing the doub-
let ground state (St = ¹) arising from the antiferromagnetic
¯
²
exchange coupling between SCu = ¹ and SNi = 1 in a CuIINiII-
¯
²
containing dinuclear unit. The molecular gz axis of 2, defined as
the Cu ؒ ؒ ؒ Ni axis in an idealized Oh symmetry, is perpendicular
to the local gz axis of the CuII, thus causing the inversion of the
observed g values, i.e., four unequally spaced intense features in
the high-field region. The four hyperfine lines in spin-coupled
CuIINiII are easy to understand, as only Cu has isotopes 63Cu
254
J. Chem. Soc., Dalton Trans., 2000, 251–258

View Article Online
Fig. 6 Schematic representation of the interactions JAF and JF between
the two hetero-dimeric units CuIINiII in complex 2.
Fig. 5 Temperature dependence of the magnetic moments for complex
2. The solid line represents the best least-squares fit of the experimental
data by the Heisenberg–Dirac–Van Vleck model.
(69.1%) and 65Cu (30.9%) with ICu = ³, whereas Ni has isotope
¯
²
61Ni with INi = ³, but its natural abundance is very low, 1.25%.
¯
²
Thus the hyperfine structure due to only the copper nucleus is
observed for exchange-coupled CuIINiII. For the spin-doublet
ground state of 2 in solution the hyperfine coupling constant
AЈCu is related to the hyperfine coupling constant ACu for a
mononuclear copper() complex by the expression AЈCu
=
ϪACu/3. The hyperfine coupling constant Az = 58 × 10Ϫ4 cmϪ1
evaluated for 2, which is indeed one-third of the ACu range
observed generally for mononuclear d9 copper() systems,27
confirms therefore the binuclear nature of 2 in solution. The
EPR spectrum of 2 exhibiting copper hyperfine structure in the
high field regions clearly indicates that the unpaired electron in
Fig. 7 Effective magnetic moment as a function of temperature for
complex 3. The solid line represents the simulation with the spin
Hamiltonian (see text).
2
the ground state for 2 resides on the molecular orbital of dz
character comprising dz2(Cu) and dz2(Ni) and is delocalized
over the CuNi core.
SCu(1) = SCu(2) = ¹, where J corresponds to the coupling between
¯
²
neighbouring NiII and CuII through the phenolate bridges
(intra-dimer) and JЈ to the coupling through the oximato
bridges (inter-dimer). The experimental magnetic data were
simulated using a least-squares fitting computer program28
with a full-matrix diagonalization approach including exchange
The CoIIICuIINiII complex, 3, in water–CH3OH glass exhibits
an EPR spectrum at 10 K which is nearly identical with that of
2 in solution. Like 2, this signal can be assigned to a transition
|¹, Ϫ¹> → |¹, ¹> of the St = ¹ ground state for 3. The
¯
²
¯
²
¯
²
¯
²
¯
²
simulated parameters for 3 are g1 = 2.273, g2 = 2.236, g3 = 2.099
and a clear four-line hyperfine structure with A3 = 54 × 10Ϫ4
cmϪ1 and A⊥ ≤ 2 × 10Ϫ4 cmϪ1 is superposed on the g3 compon-
coupling, Zeeman splitting and zero-field interactions (DS2 ), if
z
necessary. The best fit shown as the solid line in Fig. 5 results in
J = Ϫ114.9 cmϪ1, JЈ = ϩ0.96 cmϪ1, gNi = 2.16 (fixed), gCu = 2.12
(fixed), D = 0 (fixed) and temperature independent paramagnet-
ism (TIP) = 134 × 10Ϫ6 cm3 molϪ1. These data correspond to a
strong antiferromagnetic coupling, J, between the neighbouring
CuII/NiII centres (intra-dimer) as well as a weak ferromagnetic
coupling, JЈ, between the CuII/NiII pairs through the oximato
bridges. We have found that the experimental data could not be
fitted with only J; JЈ is absolutely necessary to fit the data satis-
factorily. Additionally, as a variation of D between 0 and 10
cmϪ1 has no influence on the quality of the fitting, we have kept
D = 0 (fixed) during the simulation. The spin interactions, J
and JЈ, for complex 2 are shown schematically as JAF and JF in
Fig. 6. It is interesting that a ferromagnetic coupling through
oximato bridges is not very common.29 Indeed, bridging
oximate groups are known to be very efficient in mediating very
strong antiferromagnetic interactions, which are provided by
an orbital exchange pathway essentially of σ nature.12,13 As the
2
ent due to the copper nucleus, thus strongly favouring the dz
character of the molecular orbital of the St = ¹ ground state
for 3.
¯
²
Magnetic susceptibility studies
Magnetic susceptibility data (SQUID) for polycrystalline
samples of complexes 2 and 3 were collected over the temper-
ature range 2–290 K. Fig. 5 shows the variable-temperature
magnetic susceptibility for 2 in the form of a µeff vs. T plot. At
290 K the µeff value of 3.86 µB (χmT = 1.873 cm3 molϪ1 K) is
considerably lower than that of 4.69 µB expected for a com-
pound containing two S = ¹ (CuII) and two S = 1 (NiII)
¯
²
uncoupled spins. When the complex is cooled to about 60 K,
µeff decreases monotonically reaching a plateau defined by
µeff = 2.66 µB (χmT = 0.888 cm3 molϪ1 K). Below 40 K the mag-
netic moment decreases very slowly to a value of 2.54 µB
(χmT = 0.813 cm3 molϪ1 K) at 10 K and then starts to decrease
rapidly reaching a value of 2.12 µB (χmT = 0.564 cm3 molϪ1 K)
at 2 K. This magnetic behaviour is completely in accord with
antiferromagnetically interacting spins of S = 1 and ¹, leading
2
doublet–quartet A1↔4A1 separation is large (345 cmϪ1) for 2,
the influence of the zero-field splitting on the magnetic
behaviour of 2 is negligible, justifying D = 0 kept fixed during
the simulation.
The cryomagnetic property of complex 3 is shown in Fig. 7 in
the form of a µeff vs. T plot. The magnetic moment of 2.85 µB
(χmT = 1.017 cm3 molϪ1 K) at 290 K decreases steadily with
decreasing temperature, reaching a value of 2.03 µB (χmT =
0.5145 cm3 molϪ1 K) at 80 K which remains nearly constant
until 15 K with a value of 1.97 µB (χmT = 0.4833 cm3 molϪ1 K)
and then starts to decrease reaching a value of 1.86 µB
¯
²
to a spin-doublet (2A1 in C2v symmetry) ground state. Below
60 K the excited quartet 4A1 state is totally depopulated.
We applied the usual Heisenberg–Dirac–van Vleck (HDvV)
exchange Hamiltonian in the form (1) with SNi(1) = SNi(2) = 1 and
ˆ
H = Ϫ2J(SNi(1)ؒSCu(1) ϩ SNi(2)ؒSCu(2)) Ϫ
2JЈ(SNi(1)ؒSCu(2) ϩ SNi(2)ؒSCu(1)
)
(1)
J. Chem. Soc., Dalton Trans., 2000, 251–258
255

View Article Online
Table 3 Selected magnetic and structural parameters for doubly phenoxide-bridged heterodinuclear CuIINiII complexes (H = Ϫ2JS1ؒS2)
Compounda
Cu ؒ ؒ ؒ Ni/Å
J/cmϪ1
Cu–O–Ni/Њ
Ref.
[Ni((prp)2en)Cu(hfa)2]
[CuNi((fsa)2en)(OH2)2]ؒH2O
[{Cu(salen)Ni(hfa)2}2]
[CuNiL][ClO4]2ؒH2Oؒ2DMF
1 FeIIICuIINiII
≈3.0
Ϫ48
Ϫ71
Ϫ11.8
Ϫ90
Ϫ118.6
Ϫ115
Ϫ130
100.8
98.9
93.9
18
20
21
22
2.975
2.897
3.038
3.087
3.085
3.081
99.9
101.6
101.1
100.7
15
2 CuIINiII
This work
This work, 15
3 CoIIICuIINiII
a (prp)2en = N,NЈ-ethylenebis(2-hydroxypropiophenone iminato)(2Ϫ); hfa = hexafluoroacetylacetonate; H4(fsa)2en = N,NЈ-bis(3-carboxy-2-hydroxy-
benzylidene)-1,2-diaminoethane, H2salen = N,NЈ-bis(salicylidene)-1,2-diaminoethane; H2L = a dinucleating macrocycle derived from the [2:1:1]
condensation of 2,6-diformyl-4-methylphenol, 1,2-diaminoethane and 1,4-diaminobutane.
(χmT = 0.4321 cm3 molϪ1 K) at 2 K. This magnetic behaviour
is quite characteristic of antiferromagnetic coupling between
the paramagnetic nickel() and copper() centres. The least-
squares fitting, shown as the solid line in Fig. 7, of the experi-
mental data leads to J = Ϫ130 cmϪ1, gCu = 2.24 (fixed) and
gNi = 2.27 (fixed). A temperature-independent paramagnetism
of 360 × 10Ϫ6 cm3 molϪ1 was also considered to achieve the
excellent fit shown in Fig. 7. Thus the copper and nickel centres
in 3 are antiferromagnetically exchange coupled with a doublet
ground state, and the excited quartet state lying 375 cmϪ1 above
the ground state. Thus, as is expected, the low-lying electronic
states for 2 and 3 are nearly identical.
stronger than that in 2. These results together with the com-
pounds in Table 3 emphasize the necessity for a more intricate
explanation for a complete rationalization of superexchange
through two phenoxide bridges, not only for the homometallic
but also for the heterometallic systems.
Experimental
Chemicals
Reagent or analytical grade materials were obtained from
commercial suppliers and used without further purification.
The macrocycle 1,4,7-trimethyl-1,4,7-triazacyclononane (L =
C9H21N3) was prepared as described previously.34
The relatively small linewidth and the lack of hyperfine
59
7
–
splitting by the Co (I = 2) nucleus in the EPR spectrum
of complex 3 at 10 K are clear indications that 3 is a CoIII-
CuIINiII species with localized CoIII(SCo = 0), CuII(SCu = ¹),
Physical measurements
¯
²
NiII(SNi = 1) oxidation states.
Fourier transform infrared spectroscopy on KBr pellets was
performed on a Perkin-Elmer 1720X FT-IR instrument. Elec-
tronic absorption spectra of solution were measured on a
Perkin-Elmer Lambda 19 spectrophotometer (range: 220–1400
nm). Temperature-dependent (2–298 K) magnetization data
were recorded on a SQUID magnetometer (MPMS, Quantum
Design) in a field of 1 T. The experimental susceptibility data
were corrected for underlying diamagnetism using tabulated
Pascal’s constants. The X-band EPR spectra of the polycrystal-
line material either as solid or in solution were recorded at
various temperatures between 3 and 100 K with a Bruker ESP
300 spectrometer equipped with a standard TE 102 cavity, an
Oxford Instruments liquid helium continuous-flow cryostat,
an NMR gaussmeter, and a frequency meter.
Spin multiplets in the complexes 2 and 3 are well separated by
exchange interactions. Under this condition, the g values of the
different spin multiplets can be related to local values of indi-
vidual metal ions with spin-dependent weight factors derived
from the Wigner–Eckart theorem.30,31 The g values for the spin
states St = ¹ and ³ of 2 and 3 are related to those of the local
¯
²
¯
²
copper and nickel ions by eqns. (2) and (3). The ground-state
1
2
–
–
g3/2 = (3gCu ϩ 3gNi)
(2)
(3)
1
4
3
–
–
g1/2 = (Ϫ3gCu ϩ gNi)
properties of 2 and 3 are dominated by NiII as is shown clearly
by the relation for g1/2 given above. Using the simulated gCu and
gNi values from the susceptibility data of 3 in the above relation
for g1/2, one obtains g1/2 of 2.280 which deviates appreciably
from the gav of 2.203 obtained from the simulation of the EPR
spectrum at 10 K. On the contrary g1/2 of 2.212 from the EPR
for 2 is not far from that of 2.17 obtained from susceptibility
measurements on 2.
Preparation of the compounds
2,6-Diformyl-4-methylphenol.35 The dialdehyde, described
earlier,36 was prepared by a completely different way. To a solu-
tion of p-cresol (10.8 g; 0.1 mol) in acetic acid (50 cm3) were
added hexamethylenetetraamine (28.2 g; 0.2 mol) and para-
formaldehyde (30 g; 1.0 mol). The system was stirred until a
light brown viscous solution was formed and then heated (70–
90 ЊC) for 2 h. The solution was cooled to room temperature
and conc. H2SO4 (10 cm3) carefully added. The resulting solution
was refluxed again for 0.5 h and then on treatment with distilled
water (400 cm3) resulted in the formation of a light yellow pre-
cipitate, which was stored at 4 ЊC overnight. The yellow product
was isolated by filtration and washed with a small amount of
cold CH3OH. A more pure product was obtained by means
of recrystallization from toluene. Yield: 5.7 g (35%), mp 130–
134 ЊC (lit., 133.5 ЊC). EI mass spectrum: m/z 164 (73.3), 136
(100%), calc. for [C9H8O3]ϩ 164. 1H NMR [250.13 MHz,
CDCl3, 300 K]: δ 2.36 [s, 3 H (CH3)], 7.75 [s, 2 H (aryl H)], 10.19
[s, 2 H (CHO)] and 11.43 [s, 1 H (OH)]. IR(KBr, cmϪ1): 1682s,
1666vs, 1216m and 2871w.
The mechanism of the leading phenomenon of isotropic
exchange for the CuIINiII pairs is now well understood.4,20
The net exchange is composed of two competing individual
2
2
antiferro- and ferro-magnetic exchange, J[dx Ϫ y (Cu)↔
2
2
2
2
2
dx Ϫ y (Ni)] and J[dx Ϫ y (Cu)↔dz (Ni)]. The dominant pathway
2
2
2
2
2
2 Ϫ y2
dx Ϫ y /dx
is antiferromagnetic, while the pathway dx Ϫ y /dz
involving orthogonal magnetic orbitals is weak ferromagnetic,
thus resulting in a significant overall antiferromagnetic inter-
action for all reported CuIINiII complexes.
In the case of binuclear copper()32 and nickel()33 com-
plexes linked through two phenoxide bridges, the net J is found
empirically to depend on the bridging angle θ. It has been
pointed out recently17a that these empirical relations should be
treated with caution. There is far less work published on hetero-
dinuclear CuIINiII complexes and no such empirical relation
between J and θ can be drawn up, as is evident from Table 3.
It is noteworthy that although the structures of complexes 2
and 3 are very similar, the strength of spin coupling in 3 is
[Ni(L)] A (Scheme 1). To a solution of 2,6-diformyl-4-
methylphenol (3.0 g; 18 mmol) in warm dry dimethylformamide
256
J. Chem. Soc., Dalton Trans., 2000, 251–258

View Article Online
Table 4 Crystallographic data for [{Cu(HLOX)Ni(N3)}2]ؒ2CH3OH 2
and [(tmtacn)Co(µ-OH)Cu(LOX)Ni(OH2)2][ClO2]2ؒCH3OHؒ2H2O 3
(40 cm3) under constant stirring was added dropwise 1,3-
diaminopropane (81.13 g; 15 mmol). Solid NiAc2ؒ4H2O (2.3 g;
9 mmol) was added and the solution was kept at 50 ЊC for 1 h
when a brown-yellow solid precipitated; the solid was warm-
filtered and washed with i-PrOH and diethyl ether. Yield: 1.74 g
(45%).
2
3
Formula
M
C44H50Cu2N14Ni2O10
1179.48
C31H54Cl2CoCuN7Ni
1064.88
Crystal size/mm
Crystal system
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
0.53 × 0.46 × 0.35
Monoclinic
P21/n
8.726(1)
15.024(3)
0.70 × 0.25 × 0.14
Triclinic
P1
[Cu(L)Ni]Cl2ؒ3H2O B (Scheme 1). To a vigorously stirred
suspension of complex A (0.84 g; 2.0 mmol) in methanol (20
cm3) a methanolic solution (10 cm3) of CuCl2ؒ2H2O (0.34 g;
2 mmol) was added dropwise and slowly. Within 20 min the
solution started to become turbid, when it was cooled at 4 ЊC.
Addition of diethyl ether initiated precipitation of a greenish
yellow powder, which was isolated by filtration and washed
thoroughly with water, i-PrOH and ether. Yield: 0.96 g (79%).
Calc. for C21H26Cl2CuN2NiO7: C, 41.24; H, 4.28; Cu, 10.39; N,
4.58; Ni, 9.60. Found: C, 41.5; H, 4.5; Cu, 10.2; N, 4.7; Ni,
9.7%. IR(KBr, cmϪ1): 3373m (br); 1638s, str.; 1553s, str.;
1409ms; 1340ms (s = sharp, str = strong).
¯
7.443(1)
14.281(2)
21.582(3)
104.04(2)
96.90(2)
104.31(2)
2116.5(5)
2
17.649(3)
100.18(3)
γ/Њ
V/Å3
2277.3(7)
2
1.812
7860
Z
µ/mmϪ1
1.532
7050
No. independent
reflections
(F > 4.0σ(F ))
Data/restraints/
parameters
RF (F > 4σ(F )), wR2
T/K
7669/0/335
6705/0/560
[Cu(HLOX)Ni]Clؒ2H2O C (Scheme 1). Compound B (0.61 g;
1 mmol) dissolved in methanol (30 cm3) was treated with a
methanolic solution (10 cm3) of NH2OHؒHCl (0.21 g; 3 mmol)
and Et3N (0.4 g; 4 mmol) and heated at 60 ЊC for 1 h with
continuous stirring. A green-brown solid precipitated and was
collected after filtration and washing with methanol and ether.
Yield: 0.43 g (73%). Calc. for C21H26ClCuN4NiO: C, 42.96; H,
4.29; Cu, 10.83; N, 9.54; Ni, 10.0. Found: C, 43.1; H, 4.4; Cu,
10.9; N, 9.4; Ni, 9.7%. IR(KBr, cmϪ1): 3423m (br); 1628s, str.;
1560s, str.; 1450s, str.; 1316ms.
0.0275, 0.0698
100
0.0623, 0.1635
293
was located from Fourier-difference maps and refined with
restrained distances. All non-hydrogen atoms were refined with
anisotropic thermal parameters.
Some parts of the structure of complex 3 were found to be
disordered. The disorder of the carbon atoms of the tmtacn
backbone was successfully modelled by a split atom model to
yield the δδδ and λλλ form of the macrocycle. The carbon
atoms were isotropically refined with occupancy factors of 0.5
for each set. Hydrogen atoms were geometrically attached to
both parts and refined as riding atoms.
One of the two perchlorate anions is also disordered over two
sites. Splitting of the chlorine and one oxygen atom with a ratio
of 0.75:0.25 gave a satisfactory model of two face sharing
tetrahedral parts of the anion. The mean component was aniso-
tropically refined while the minor positions (Cl(1X) and
O(10X)) were treated isotropically.
[{Cu(HLOX)Ni(N3)}2]ؒ2CH3OH 2. A solid sample of NaN3
(0.13 g; 2 mmol) was added to a warm solution of compound C
(0.587 g; 1 mmol) in methanol (30 cm3) at 60 ЊC with stirring.
The solution was kept at 60 ЊC for 0.5 h without stirring, when
crystallization started. The green crystals were collected by fil-
tration and air-dried. Yield: 0.95 g (80%). Calc. for C22H25-
CuN7NiO5: C, 44.81; H, 4.27; Cu, 10.78; N, 16.63; Ni, 9.95.
Found: C, 44.6; H, 4.3; Cu, 10.8; N, 16.5; Ni, 9.8%. IR(KBr,
cmϪ1): 2042s, str.; 1636s, str.; 1557ms; 1453ms; 1314ms; 1113,
1086 and 1069ms. UV-vis in DMF: λ/nm (ε/MϪ1 cmϪ1
372(25200), 583(249), 698 (sh)(≈190) and 1070(41).
)
The disorder of the solvent molecules is less well understood.
Cations of complex 3 are hydrogen bridged by two water mol-
ecules (O(40), O(50)) which were anisotropically refined with
occupancies of 1 and 0.75, respectively. Since the crystals were
grown from a water–methanol mixture residual electron density
in potential solvent areas suggested two more positions of
methanol molecules. An occupancy factor of 0.25 was used for
methanol O(51)–C(51) near water molecule O(50) and 0.5 was
given to O(60)–C(60).
[(tmtacn)Co(ꢀ-OH)Cu(LOX)Ni(OH2)2][ClO4]2ؒCH3OHؒ2H2-
O 3. A solid sample of compound C (0.29 g; 0.5 mmol) was
added in small portions to a solution of 1,4,7-trimethyl-1,4,7-
triazacyclononane (0.085 g; 0.5 mmol) in water–methanol
(15:30 cm3) containing CoCl2ؒ6H2O (0.13 g; 0.5 mmol). After
addition of a small portion of Et3N (0.5 cm3) the resulting
solution was warmed at 70 ЊC on a water-bath for 1 h. The
solution kept at ambient temperature after addition of NaClO4ؒ
H2O yielded green crystals. Yield: 0.28 g (53%). Calc. for
C31H54Cl2CoCuN7NiO18: C, 34.97; H, 5.11, Co, 5.53; Cu, 5.97;
N, 9.21; Ni, 5.51. Found: C, 34.8; H, 5.10; Co, 5.7; Cu, 6.0; N,
9.4; Ni, 5.7%. IR(KBr, cmϪ1): 3407 (br), str.; 1636s, str.;
1556ms; 1448ms; 1308ms; 1144ms; 1121s, str.; 1081s, vs; 626ms.
UV-vis in DMF: λ/nm (ε/MϪ1 cmϪ1): 379(16510), 585(429),
750(87) and 983(42).
CCDC reference number 186/1763.
See http://www.rsc.org/suppdata/dt/a9/a908426f/ for crystal-
lographic files in .cif format.
Acknowledgements
We thank Deutsche Akademischer Austauschdienst for a
stipend (to C. N. V.), the Fonds der Chemischen Industrie and
Max-Planck-Society for support.
X-Ray crystallography
The crystallographic data of [CuIINiII]2 2 and CoIIICuIINiII
3
References
1 E. Sinn, Coord. Chem. Rev., 1970, 5, 313.
2 Magneto-Structural Correlations in Exchange Coupled Systems,
eds. R. D. Willett, D. Gatteschi and O. Kahn, Kluwer Academic
Publishers, Dordrecht, 1985.
3 Magnetic Molecular Materials, eds. D. Gatteschi, O. Kahn, J. S.
Miller and F. Palacio, Kluwer Academic Publishers, Dordrecht,
1991.
are summarized in Table 4. Graphite monochromated Mo-Kα
X-radiation (λ = 0.71073 Å) was used throughout. Intensity
data were corrected for Lorentz-polarization and absorption
effects using the program SADABS.37 The structures were
solved by direct methods by using SHELXTL PLUS.38 The
function minimized during full-matrix least-squares refinement
was Σw(|Fo| Ϫ |Fc|)2. Hydrogen atoms attached to carbon were
placed at calculated positions and refined as riding atoms
with isotropic thermal parameters. The oximic hydrogen OH
4 O. Kahn, Molecular Magnetism, VCH Verlagsgesellschaft, Wein-
heim, 1993.
J. Chem. Soc., Dalton Trans., 2000, 251–258
257

View Article Online
17 (a) D. Black, A. J. Blake, K. P. Dancey, A. Harrison, M. McPartlin,
5 O. Kahn, Adv. Inorg. Chem., 1995, 43, 179.
6 K. S. Murray, Adv. Inorg. Chem., 1995, 43, 261.
S. Parsons, P. A. Tasker, G. Whittaker and M. Schröder, J. Chem.
Soc., Dalton Trans., 1998, 3953; (b) K. K. Nanda, A. W. Addison,
N. Paterson, E. Sinn, L. K. Thompson and U. Sakaguchi, Inorg.
Chem., 1998, 37, 1028; (c) H. Okawa, T. Tokii, Y. Muto and S. Kida,
Bull. Chem. Soc. Jpn., 1973, 46, 2464.
7 Research Frontiers in Magnetochemistry, ed. C. J. O’Connor, World
Scientific, Singapore, 1993.
8 See for example: R. H. Holm and E. I. Solomon (Guest Editors),
Chem. Rev., 1996, 96, No. 7; A. L. Feig and S. J. Lippard, Chem.
Rev., 1994, 94, 759; L. Que and A. E. True, Prog. Inorg. Chem., 1990,
38, 98.
18 C. J. O’Connor, D. P. Freyberg and E. Sinn, Inorg. Chem., 1979, 18,
1077.
9 S. J. Lippard and J. M. Berg, Principles of Bioinorganic Chemistry,
University Science Books, Mill Valley, CF, 1994; W. Kaim and
B. Schwederski, Bioanorganische Chemie, B. G. Teubner, Stuttgart,
1991; Bioinorganic Chemistry of Copper, eds. K. D. Karlin and
Z. Tyeklár, Chapman & Hall, New York, 1993; Mechanistic
Bioinorganic Chemistry, eds. H. Holden Thorp and V. L. Pecoraro,
American Chemical Society, Washington, 1995.
10 S. J. Gruber, C. M. Harris and E. Sinn, J. Inorg. Nucl. Chem., 1968,
30, 1805; J. Selbin and L. Ganguly, J. Inorg. Nucl. Chem. Lett., 1969,
5, 815; R. L. Linkvedt, L. S. Kramer, G. Ranger, P. W. Corfield and
M. D. Glick, Inorg. Chem., 1983, 22, 3580.
11 D. Luneau, H. Oshio, H. Okawa and S. Kida, J. Chem. Soc., Dalton
Trans., 1990, 2283; S. Pal, R. Mukherjee, M. Tomas, L. R. Falvello
and A. Chakravorty, Inorg. Chem., 1986, 25, 200; E. Colacio,
J. M. Dominguez-Vera, A. Escuer, R. Kivekäs and A. Romerosa,
Inorg. Chem., 1994, 33, 3914.
19 S. L. Lambert, C. L. Spiro, R. R. Gagné and D. N. Hendrickson,
Inorg. Chem., 1982, 21, 68.
20 I. Morgenstern-Badarau, M. Rerat, O. Kahn, J. Jaud and J. Galy,
Inorg. Chem., 1982, 21, 3050.
21 Y. Journaux, O. Kahn, I. Morgenstern-Badarau, J. Galy, J. Jaud,
A. Bencini and D. Gatteschi, J. Am. Chem. Soc., 1985, 107, 6305.
22 T. Aono, H. Wada, Y. Aratake, N. Matsumoto, H. Okawa and
Y. Matsuda, J. Chem. Soc., Dalton Trans., 1996, 25.
23 J. Ribas, C. Diaz, R. Costa, J. Tercero, X. Solans, M. Font-Bardía
and H. Stoeckli-Evans, Inorg. Chem., 1998, 37, 233.
24 Comprehensive Coordination Chemistry, eds. G. Wilkinson, R. D.
Gillard and J. A. McCleverty, Pergamon, Oxford, 1987.
25 S. Mohanta, K. K. Nanda, L. K. Thompson, U. Flörke and K. Nag,
Inorg. Chem., 1998, 37, 1465 and references therein.
26 B. A. Goodman and J. B. Raynor, Adv. Inorg. Chem. Radiochem.,
1978, 35, 701.
12 P. Chaudhuri, M. Winter, U. Flörke and H.-J. Haupt, Inorg. Chim.
Acta, 1995, 232, 125; F. Birkelbach, M. Winter, U. Flörke, H.-J.
Haupt, C. Butzlaff, M. Lengen, E. Bill, A. X. Trautwein,
K. Wieghardt and P. Chaudhuri, Inorg. Chem., 1994, 33, 3990;
D. Burdinski, F. Birkelbach, M. Gerdan, A. X. Trautwein,
K. Wieghardt and P. Chaudhuri, J. Chem. Soc., Chem. Commun.,
1995, 963.
13 P. Chaudhuri, M. Winter, B. P. C. Della Vedova, E. Bill, A. X.
Trautwein, S. Gehring, P. Fleischhauer, B. Nuber and J. Weiss,
Inorg. Chem., 1991, 30, 2148; D. Burdinski, F. Birkelbach, T.
Weyhermüller, U. Flörke, H.-J. Haupt, M. Lengen, A. X. Trautwein,
E. Bill, K. Wieghardt and P. Chaudhuri, Inorg. Chem., 1998, 37,
1009; F. Birkelbach, U. Flörke, H.-J. Haupt, C. Butzlaff, A. X.
Trautwein, K. Wieghardt and P. Chaudhuri, Inorg. Chem., 1998, 37,
2000 and references therein.
14 (a) P. Chaudhuri, F. Birkelbach, M. Winter, V. Staemmler, P.
Fleischhauer, W. Haase, U. Flörke and H.-J. Haupt, J. Chem. Soc.,
Dalton Trans., 1994, 2313; (b) C. Krebs, M. Winter, T. Weyher-
müller, E. Bill, K. Wieghardt and P. Chaudhuri, J. Chem. Soc.,
Chem. Commun., 1995, 1913 and references therein.
15 C. N. Verani, T. Weyhermüller, E. Rentschler, E. Bill and P.
Chaudhuri, Chem. Commun., 1998, 2475.
27 E. I. Solomon, M. D. Lowery, D. E. Root and B. L. Hemming,
Adv. Chem. Ser. 1995, 246, 121.
28 V. Staemmler, F. Birkelbach and C. Krebs, Bochum, 1997.
29 F. Lloret, R. Ruiz, M. Julve, J. Fans, Y. Journaux, I. Castro and
M. Verdaguer, Chem. Mater., 1992, 4, 1150.
30 C. C. Chao, J. Magn. Reson., 1973, 10, 1; R. P. Scaring, D. J.
Hodgson and W. E. Hatfield, Mol. Phys., 1978, 35, 701.
31 A. Bencini and D. Gatteschi, EPR of Exchange Coupled Systems,
Springer, Berlin, 1990.
32 L. K. Thompson, S. K. Mandal, S. S. Tandon, J. N. Bridson and
M. K. Park, Inorg. Chem., 1996, 35, 3117.
33 K. K. Nanda, L. K. Thompson, J. N. Bridson and K. Nag, J. Chem.
Soc., Chem. Commun., 1994, 1337.
34 K. Wieghardt, P. Chaudhuri, B. Nuber and J. Weiss, Inorg. Chem.,
1982, 21, 3086.
35 S. Gallert, personal communication.
36 R. R. Gagné, C. L. Spiro, T. J. Smith, C. A. Hamann, W. R. Thies
and A. K. Shiemke, J. Am. Chem. Soc., 1981, 103, 4073; F. Ullmann
and K. Brittner, Chem. Ber., 1909, 42, 2539.
37 G. M. Sheldrick, SADABS, Universität Göttingen, 1994.
38 G. M. Sheldrick, SHELXTL PLUS, Siemens Analytical X-Ray
Instruments, Madison, WI, 1990.
16 E. V. Rybak-Akimova, D. H. Busch, P. K. Kahol, N. Pinto, N. W.
Alcock and H. J. Clase, Inorg. Chem., 1997, 36, 510.
Paper a908426f
258
J. Chem. Soc., Dalton Trans., 2000, 251–258