Diversity in the reactions of unsymmetric dinucleating Schiff base ligands with Cu11 and Ni11

Article abstract of DOI:10.1039/b001395l

A diversity of reaction products have been found in the reactions of two related unsymmetrical Schiffbase dinucleating ligands, HL1 and HL2, derived from 3-chIoromethyl-5-methylsalicylaldehyde 1, with copper(n) and nickel(n) salts. The ligands remain intact in the copper(n) complexes to give the homodinuclear complexes [Cu2L'Br3] 2 and [Cu2L2(OH)(C1O4)]C1O4 3, the crystal structures of which have been solved. The reactions of HL1 and HL2 with nickel perchlorate led to hydrolysis of the imine bond. With HL1 the homodinuclear complex [NiLA(OH2)]2[ClO4]2'4H2O was formed and with HL2 hydrolysis was followed by elimination of C2H4 from the terminal NEt2 of the iminic side arm to leave an NHEt group and the dinuclear complex [NiLc(OH2)]2[ClO4]2-3CH3OH. The crystal structures of the two nickel complexes are also reported. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b001395l

View Article Online / Journal Homepage / Table of Contents for this issue
Diversity in the reactions of unsymmetric dinucleating Schiff base
ligands with CuII and NiII
Harry Adams,a David E. Fenton,*a Shabana R. Haque,a Sarah L. Heath,a Masaaki Ohba,b
Hisashi Okawab and Sharon E. Speya
a Department of Chemistry, Dainton Building, The University of Sheffield, Sheffield,
UK S3 7HF
b Department of Chemistry, Faculty of Science, Kyushu University, Hakozaki, Higashiku,
Fukuoka 812, Japan
Received 21st February 2000, Accepted 20th April 2000
Published on the Web 23rd May 2000
A diversity of reaction products have been found in the reactions of two related unsymmetrical Schiff base dinucleating
ligands, HL1 and HL2, derived from 3-chloromethyl-5-methylsalicylaldehyde 1, with copper() and nickel() salts.
The ligands remain intact in the copper() complexes to give the homodinuclear complexes [Cu2L1Br3] 2 and [Cu2L2-
(OH)(ClO4)]ClO4 3, the crystal structures of which have been solved. The reactions of HL1 and HL2 with nickel
perchlorate led to hydrolysis of the imine bond. With HL1 the homodinuclear complex [NiLA(OH2)]2[ClO4]2ؒ4H2O
was formed and with HL2 hydrolysis was followed by elimination of C2H4 from the terminal NEt2 of the iminic side
arm to leave an NHEt group and the dinuclear complex [NiLC(OH2)]2[ClO4]2ؒ3CH3OH. The crystal structures of the
two nickel complexes are also reported.
Introduction
Current awareness of the asymmetric nature of a number of
homodinuclear or heterodinuclear transition metal-derived
metallobiosites and of the ability of the individual metal ions to
have quite distinct roles in the functioning of the metallo-
enzyme concerned has led to a search for carefully designed
unsymmetric dinucleating ligands which will give dinuclear
complexes capable of acting as models for the metallobiosites.1,2
The biosites in question have been classified in four distinct
groupings as follows (Fig. 1):3 (a) symmetric, in which an iden-
tical number of donor atoms of the same type are bound to
Fig. 2 Schematic representations of dinucleating ligands. Mono- or
bi-bracchial pendant arms may be attached to the N atoms in the “end-
each metal atom in similar geometries; (b) donor asymmetry,
where different types of donor atom co-ordinate to each metal
atom; (c) geometric asymmetry, where there are inequivalent
geometric arrangements of the donor atoms about each metal
atom; and (d) co-ordination number asymmetry, where a differ-
ent number of donor atoms are co-ordinated to each metal
atom. To a first approximation the nature of the donor atom
may be restricted to simply O, N, S, etc. but a more accurate
definition would specify the functional grouping associated
off” compartmental ligands and to the X atoms in the isolated donor
sets. The spacers in the isolated donor sets do not provide bridging
atoms.
with the donor atom and so differentiate between an oxygen
atom in water and carboxylate or a sulfur atom in a thiolate or a
thioether. A combination of different types of asymmetry may
also occur at a dinuclear centre.
Although many examples of co-ordination complexes
derived from symmetric acyclic dinucleating ligands have been
prepared and investigated as potential model complexes for
metallobiosites, polydentate ligand systems that would give
necessarily asymmetric dinuclear complexes remain rare; site
asymmetry was often only accessed through good fortune.
Complexes of dinucleating ligands have been divided into two
general classes (Fig. 2).4 The first group consists of those
complexes in which the metals share at least one donor atom
in species containing adjacent sites in which the central donor
atom(s) provide a bridge; the ligands giving these ‘bridging
donor sets’ have collectively been termed compartmental lig-
ands. The second group consists of those complexes in which
donor atoms are not shared and so isolated donor sets exist. If
the arms are constituted of different donor atoms then an
unsymmetric ligand results.
Fig. 1 A classification of metal co-ordination environments found at
transition metal-derived dinuclear centres present in metallobiosites (M
is a transition metal and W, X, Y and Z are ligand donor atoms such as
N, O, S, etc.).
Early studies on the derivation of unsymmetric “end-off”
compartmental proligands involved the introduction of a single
pendant arm into a 5-substituted salicylaldehyde, using the
DOI: 10.1039/b001395l
J. Chem. Soc., Dalton Trans., 2000, 1849–1856
This journal is © The Royal Society of Chemistry 2000
1849

View Article Online
Mannich reaction, followed by condensation of the resulting
aminomethyl salicylaldehyde with a primary amine to give a
range of unsymmetric dinucleating Schiff base proligands
and their dinuclear copper() complexes.5–7 More recently a
synthetic route based on the reaction of 3-chloromethyl-5-
methylsalicylaldehyde (1) with functionalised secondary amines
followed by Schiff base condensation with functionalised pri-
mary amines has provided a further range of unsymmetric pro-
ligands.8,9 In the current work the functionalised secondary
amines N-ethyl-NЈ,NЈ-dimethylethane-1,2-diamine and N,N-
diethyl-NЈ-methylethane-1,2-diamine were used to prepare the
proligands HLA and HLB; N,N-dimethylethane-1,2-diamine
was then used to prepare the donor asymmetric compartmental
ligands HL1 and HL2, the asymmetry arising from the mixture
of sp2 and sp3 N atoms available for metal co-ordination. The
ligands were then treated with copper() and nickel() to derive
homodinuclear complexes.
ethanol (25 cm3) for 10 minutes and N,N-dimethylethane-1,2-
diamine (0.092 ml, 0.836 mmol) added to the boiling solution
via a microsyringe. The solution instantly changed from a pale
to bright yellow and was stirred for 10 minutes at room
temperature. The solution was cooled in ice and the sol-
vent removed in vacuo giving a bright orange-yellow oil. This
was dried under high vacuum. Yield = 95%. Found (required
for C19H34N4O): C, 67.51 (68.22); H, 10.57 (10.25); N,
1
16.49 (16.75)%. H NMR (250 MHz, CDCl3): δ 8.35 (1 H,
s, CHN), 7.20 (1 H, s, aryl CH), 7.00 (1 H, aryl s, CH), 3.70
(2 H, t, EtNCH2CH2NMe2), 3.65 (2 H, s, CH2C6H2), 2.65 (2 H,
t, imine CH2CH2NMe2), 2.65 (2 H, t, imine CH2CH2NMe2),
2.60 (2 H, t, NEtCH2CH2NMe2), 2.50 (2 H, t, EtNCH2-
CH2NMe2), 2.50 (4 H, q, NCH2CH3), 2.40 (6 H, s, NMe2,
unsaturated pendant arm), 2.30 (3 H, s, CH3C6H2), 2.15 (6 H,
s, NMe2, saturated pendant arm), 2.20 (3 H, s, NMe) and 1.10
(3 H, t, NCH2CH3). 13C NMR (250 MHz, CDCl3): δ 164.7
(CN), 157.2 (aryl), 133.7 (aryl), 129.5 (aryl), 126.3 (aryl), 118.6
(aryl), 60.0, 57.9, 57.3, 52.1, 51.3, 47.9, 45.8 (NMe2), 20.4
(C6H2Me) and 11.2 (NEt). IR (NaCl plates): 2940, 2857, 2815
Experimental
(νC–H, aliphatic) and 1632 cmϪ1 (νC᎐N). MS (FABϩ): m/z 335 (95%,
Elemental analyses were carried out by the University of
Sheffield microanalytical service. Infrared spectra were
recorded as KBr discs or as liquid films between NaCl plates,
using a Perkin-Elmer 297 (4000–600 cmϪ1) or a 1600 (4000–400
᎐
[C19H35N4O]ϩ.
Precursor HLB. 3-Chloromethyl-5-methylsalicylaldehyde
(1.42 g, 0.772 mmol) was dissolved in THF (50 cm3) at 0 ЊC.
N,N-Diethyl-NЈ-methylethane-1,2-diamine (2.50 cm3, 1.54
mmol) was predissolved in THF (15 cm3), and then added
dropwise to the stirring chloroaldehyde over a period of
approximately one hour. On addition of the amine there was
an immediate change from colourless to yellow and a cream
precipitate deposited on the sides of the flask. The solution
was stirred for a further hour at 0 ЊC. The solvent was removed
in vacuo giving a red solid. The flask containing the solid was
first washed with distilled water (20 cm3), and then chloroform
(20 cm3). In each case the washings were poured into a beaker
giving rise to two layers; an aqueous top layer and an organic
bottom layer. Both layers were then acidified with HCl (2 M),
causing a small amount of the precipitate found in the aqueous
phase to dissolve. The aqueous phase was removed and made
strongly basic with NaOH (10 M) (pH 10–11). On addition of
base a yellow solid precipitated. The bright yellow solution was
then extracted with chloroform (4 × 25 cm3), the organic
extracts were combined, dried over anhydrous magnesium
sulfate, filtered and then the filtrate was evaporated to dryness
in vacuo giving a dark orange oil which was dried under high
vacuum prior to purification by flash column chromatography
using ethyl acetate–triethylamine (90%:10%) as the eluent.
Yield = 63%. Found (required for C8H13NO): C, 68.59 (69.03);
H, 9.34 (9.41); N, 10.34 (10.06)%: 1H NMR (250 MHz, CDCl3):
δ 10.30 (1 H, s, CHO), 8.40 (1 H, broad s, phenolic proton), 7.40
(1 H, s, aryl CH), 7.00 (1 H, s, aryl CH), 3.55 (2 H, s, CH2C6H2),
2.60 (2 H, t, MeNCH2CH2NEt2), 2.55 (2 H, t, MeNCH2CH2-
NEt2), 2.55 (4 H, q, NCH2CH3), 2.50 (4 H, q, NCH2CH3), 2.20
(3 H, s, NMe), 2.15 (3 H, s, CH3C6H2) and 0.90 (6 H, t,
NCH2CH3). 13C NMR (250 MHz, CDCl3): δ 191.8 (CO), 159.6
(aryl), 136,6, 127.9, 127.8, 124.6, 122.7, 58.4, 54.3, 50.0, 46.7,
42.1 (NEt2), 20.2 (NMe) and 11.0 (C6H2Me). IR (NaCl plates):
1
cmϪ1) infrared spectrophotometer, H and 13C NMR spectra
using either a Bruker ACF-250, AM-250 or WH-400 spec-
trometer and positive ion fast atom bombardment (FAB) mass
spectra using a Kratos MS80 or a VG PROSPEC spectrometer
(the matrix used was 4-nitrobenzyl alcohol).
Ligand synthesis
3-Chloromethyl-5-methylsalicylaldehyde was prepared by the
method of reference 8.
Precursor HLA. 3-Chloromethyl-5-methylsalicylaldehyde (1.0
g, 5.43 mmol) was stirred at 0 ЊC in THF (75 ml). N-Ethyl-
NЈ,NЈ-dimethylethane-1,2-diamine (1.71 cm3, 10.9 mmol) was
added dropwise, over approximately 45 minutes, via a dropping
funnel. The solution immediately turned bright yellow and was
stirred for a further hour at 0 ЊC. The orange reaction mixture
was filtered to remove any hydrochloride salt, then THF was
removed from the filtrate in vacuo to give an oily orange solid
which was first dissolved in distilled water (25 cm3) to which
chloroform (25 cm3) was added. The washings were poured into
a beaker and both layers acidified with HCl (2 M) to pH 1–2,
causing the aqueous layer to become decolourised. The aque-
ous layer was removed and the pale yellow chloroform layer
then made strongly basic with NaOH (10 M). The addition of
base caused a bright yellow solid to be precipitated. The prod-
uct was extracted with chloroform (4 × 25 cm3), the organic
extracts were combined, dried with anhydrous magnesium
sulfate, filtered and the filtrate evaporated to dryness giving a
dark orange-yellow oil which was dried under vacuum. The oil
was characterised using NMR and this revealed that there was
no need for further purification. Yield = 65%. Found (required
for C15H24N2O2): C, 67.69 (68.41); H, 9.25 (8.80); N, 10.49
(10.64)% . 1H NMR (250 MHz, CDCl3): δ 10.40 (1 H, s,CHO),
8.45 (1 H, broad s, phenolic proton), 7.45 (1 H, s, aryl CH), 7.00
(1 H, s, aryl CH), 3.60 (2 H, s, CH2C6H2), 2.60 (4 H, q,
NCH2CH3 and 2 H, t, EtNCH2CH2NMe2), 2.55 (2 H, t,
EtNCH2CH2NMe2), 2.50 (2 H, q, NCH2CH3), 2.20 (3 H, s,
3480 (νOH), 2968, 2804 (νC–H, aliphatic) and 1679 cmϪ1 (νC᎐0). MS
᎐
(FABϩ): m/z 278 (100%, [C16H26N2O2]ϩ).
Unsymmetrical Schiff base proligand HL2. The proligand
HLB (0.20 g, 0.746 mmol) was gently refluxed in ethanol (25
cm3) for approximately 10 min and N,N-dimethylethane-1,2-
diamine (0.08 cm3, 0.746 mmol) added dropwise to the boiling
solution via a microsyringe. The resulting solution was stirred
for 10 minutes; the bright yellow solution was then cooled in ice
before removal of the solvent in vacuo to give a bright yellow oil
which was dried under a vacuum. Yield = 83%. Found (required
for C20H36N4O): C, 68.28 (68.92); H, 10.47 (10.47); N, 16.21
NMe2 and 3 H, s, CH3C6H2) and 1.05 (3 H, t, NCH2CH3). 13
C
NMR (250 MHz, CDCl3): δ 191.4 (CO), 159.9 (aryl), 136.4
(aryl), 127.7 (aryl), 127.4 (aryl), 124.6 (aryl), 56.5, 55.1, 50.0,
47.3, 45.3 (NMe2), 20.2 (NMe) and 10.9 (C6H2Me). IR (NaCl
plates): 3480 (νOH), 2969, 2819 (νC–H, aliphatic) and 1678 cmϪ1
(νC᎐O). MS (FABϩ): m/z 264 (25%, [C15H24N2O2]ϩ).
᎐
Unsymmetrical Schiff base proligand HL1. The proligand
HLA (0.22 g, 0.836 mmol) was gently refluxed in absolute
1
(16.07)%. H NMR (250 MHz, CDCl3): δ 8.35 (1 H, s, CHN),
1850
J. Chem. Soc., Dalton Trans., 2000, 1849–1856

View Article Online
Table 1 Crystal data, structure solution and refinement
[Cu2L1Br3] 2
[Cu2L2(OH)][ClO4]2 3
[NiLA(OH2)]2[ClO4]2ؒ4H2O 4
[NiLC(OH2)]2[ClO4]2ؒ3CH3OH 5
Chemical formula
Formula weight
T/K
C19H33Br3Cu2N4O
700.30
150(2)
C20H36Cl2Cu2N4O10
690.51
298(2)
C30H58Cl2N4Ni2O18
951.12
150(2)
C31H58Cl2N4Ni2O17
947.13
150(2) K
¯
Crystal system, space group
Orthorhombic, Pbca
Monoclinic, P21/c
Monoclinic, C2/c
Triclinic, P1
a/Å
b/Å
c/Å
α/Њ
15.282(3)
11.3423(19)
28.242(5)
13.138(3)
12.800(4)
18.132(5)
22.938(6)
10.148(2)
17.916(4)
10.3948(15)
10.6991(15)
10.8277(14)
73.874
β/Њ
107.870(10)
111.963(10)
85.768
62.618
1025.2(2)
1
1.124
γ/Њ
V/Å3
4895.4(14)
8
6.653
2902.1(14)
4
1.705
3867.8(16)
4
1.193
Z
µ/mmϪ1
Reflections collected
Independent reflections
Final R1, wR2 indices
[I > 2σ(I)]
30453
5961 (Rint = 0.3510)
0.0678, 0.1520
6404
12499
4584 (Rint = 0.1299)
0.1264, 0.3124
4539
5118 (Rint = 0.0734)
0.0688, 0.1724
2938 (Rint = 0.0953)
0.0892, 0.2436
(all data)
0.1135, 0.1770
0.1055, 0.1988
0.2391, 0.3780
0.1126, 0.2946
7.15 (1 H, s, aryl CH), 6.95 (1 H, s, aryl CH), 3.65 (2 H, t,
imineCH2CH2NMe2), 3.55 (2 H, s, CH2C6H2), 2.60 (2 H, t,
imineCH2CH2NMe2), 2.60 (2 H, t, MeNCH2CH2NEt2), 2.55
(2 H, t, MeNCH2CH2NEt2), 2.50 (4 H, q, NCH2CH3), 2.20
(6 H, s, NMe2), 2.20 (3 H, s, NMe2), 2.20 (3 H, s, CH3C6-
H2), 0.90 (6 H, t, NCH2CH3). 13C NMR (250 MHz, CDCl3):
δ 165.1 (CN), 150.6 (aryl), 134.3, 130.2, 128.9 123.2, 121.8,
59.9, 57.7, 55.8, 50.3, 45.8 (NMe2), 42.8 (NEt2), 20.4 (NMe)
and 11.2 (C6H2Me). IR (NaCl plates): 3405 (νOH), 2968, 2816
X-Ray crystallography
The details of the crystal data, and of the structure solution
and refinement, are given in Table 1. Measurements were made
on a Siemens SMART CCD area diffractometer for the copper
complex 2 and the two nickel complexes 4 and 5, and using a
Siemens P4 four-circle diffractometer for 3. The programs used
were Siemens SMART and SAINT for control and integration
software and SHELXTL as implemented on the Viglen
Pentium computer; XSCANS was used for the P4.10
CCDC reference number 186/1945.
(νC–H, aliphatic) and 1635 cmϪ1 (νC᎐N). MS (FABϩ): m/z 348 (30%,
᎐
[C20H36N4O]ϩ).
See http://www.rsc.org/suppdata/dt/b0/b001395l/ for crystal-
lographic files in .cif format.
Complexation reactions
Proligand HL1 and CuBr2. The proligand HL1 (0.05 g, 0.15
mmol) was stirred at room temperature in methanol (20 cm3)
with CuBr2 (0.067 g, 0.3 mmol) for 1 hour. The dark green
solution was hot filtered and the filtrate allowed slowly to evap-
orate at room temperature. A dark green crystalline solid was
obtained, washed with light petroleum and left to dry in air at
room temperature. Found for bulk sample (required for
[C19H33Br2Cu2N4O][CuBr2]): C, 27.41 (27.05); H, 4.34 (4.94);
Results and discussion
Synthesis and structural characterisation
The proligands (HL1 and HL2) were synthesized from com-
pound 1 by reaction with either N-ethyl-NЈ,NЈ-dimethylethane-
1,2-diamine or N,N-diethyl-NЈ-methylethane-1,2-diamine and
N,N-dimethylethane-1,2-diamine according to Scheme 1 and
N, 6.62 (6.64)%. IR (KBr disc): 1636 cmϪ1 (νC᎐N).
᎐
Proligand HL2 and Cu(ClO4)2. The complex was prepared in
a similar manner to that above using Cu(ClO4)2 (0.21 g, 0.575
mmol) and proligand HL2 (0.10 g, 0.287 mmol). Found for bulk
sample (required for [C20H37Cu2N4O2][ClO4]2ؒ3H2O): C, 32.14
(32.26); H, 5.32 (5.56); N, 7.10 (7.52)%. IR (KBr disc): 3446
(νO–H), 1637 (νC᎐N), 1090, 625 cmϪ1 (νClO ).
Ϫ
4
᎐
Proligand HL1 and Ni(ClO4)2. Proligand HL1 (0.05 g, 0.15
mmol) and nickel() perchlorate (0.110 g, 0.299 mmol) were
stirred at room temperature in methanol (20 cm3) for 1 hour.
The deep orange solution was hot filtered and the filtrate
left slowly to evaporate at room temperature. A crystalline
brown solid was obtained, washed with light petroleum (bp 60–
80 ЊC) and dried in air. Recrystallisation from methanol gave
green crystals suitable for structural analysis. Found (required
for C15H29ClN2NiO9): C, 37.58 (37.89); H, 5.59 (6.15); N, 6.34
(5.89)%. IR (KBr disc): 3448 (νOH), 1628 (νC᎐N), 1083, 669 cmϪ1
᎐
Ϫ
(νClO ).
4
Scheme 1
Proligand HL2 and Ni(ClO4)2. The complex was prepared in
a similar manner to the above using Ni[ClO4]2ؒ6H2O (0.21 g,
0.575 mmol) and proligand HL2 (0.10 g, 0.287 mmol). Found
(required for C31H58Cl2N4Ni2O17): C, 38.61 (39.31); H, 5.67
(6.17); N, 5.28 (5.92)%. IR (KBr disc): 3446 (νOH), 1636 (νC᎐N),
used to prepare homodinuclear copper() and nickel() com-
plexes. Problems were encountered in obtaining reproducible
pure bulk samples and so the complexes were recrystallised and
the resulting products used in both structural and subsequent
physico-chemical determinations.
1121, 624 cmϪ1 (νClO ).
Ϫ
4
J. Chem. Soc., Dalton Trans., 2000, 1849–1856
1851

View Article Online
The reaction of HL1 with CuBr2 in methanol gave a complex
the bulk sample of which analysed as [Cu3L1Br4]. This sug-
nϪ
gested that the product might contain the anion [CuBr2]n to
give either [Cu2L1Br2][CuBr2] or the dimer [Cu2L1Br2]2[Cu2Br4]
by analogy with the crystallographically characterised product
of the reaction of the unsymmetrical ligand L with CuBr2 and
triethyl orthoformate in methanol, [CuLBr(HCO2)]2[Cu2Br4].5
In order to obtain this anion it was suggested that bromide ions
present in solution reduce copper() to copper() and that the
bromide ions are in turn oxidised to bromine. Copper() can
then combine with bromide to give a linear anion, [CuBr2]Ϫ,
which can reversibly dimerise to give [Cu2Br4]2Ϫ i.e. 2CuII ϩ
Fig. 3 The molecular structure of [Cu2L1Br3] 2. Selected bond lengths
(Å) and angles (Њ) at the metal atoms: Cu(1)–O(1), 1.960(5); Cu(1)–
N(2), 2.036(6); Cu(1)–N(1), 2.049(6); Cu(1)–Br(1), 2.4334(11); Cu(1)–
Br(2), 2.6786(11); Cu(2)–O(1), 1.971(5); Cu(2)–N(4), 1.972(6); Cu(2)–
N(3), 2.057(6); Cu(2)–Br(2), 2.4441(11); Cu(2)–Br(3), 2.6245(12);
Cu(1)–Cu(2), 3.221; O(1)–Cu(2)–N(4), 88.4(2); O(1)–Cu(1)–N(2),
172.7(2); O(1)–Cu(2)–N(3), 159.4(2); O(1)–Cu(1)–N(1), 86.4(2); N(4)–
Cu(2)–N(3), 83.3(2); N(2)–Cu(1)–N(1), 86.5(2); O(1)–Cu(2)–Br(2),
86.80(14); O(1)–Cu(1)–Br(1), 90.75(14); N(4)–Cu(2)–Br(2), 162.66(19);
N(2)–Cu(1)–Br(1), 96.40(16); N(3)–Cu(2)–Br(2), 95.56(18); N(1)–
Cu(1)–Br(1), 152.39(18); O(1)–Cu(2)–Br(3), 99.82(15); O(1)–Cu(1)–
Br(2), 80.70(14); N(4)–Cu(2)–Br(3), 95.58(18), N(2)–Cu(1)–Br(2),
99.88(18); N(3)–Cu(2)–Br(3), 99.77(19); N(1)–Cu(1)–Br(2), 109.57(18);
Br(2)–Cu(2)–Br(3), 101.65(4); Br(1)–Cu(1)–Br(2), 97.01(4); Cu(2)–
Br(2)–Cu(1), 77.76(3); Cu(1)–O(1)–Cu(2), 110.1(2)Њ.
6BrϪ → 2[CuBr2]Ϫ ϩ Br2; 2[CuBr2]Ϫ
[Cu2Br4]2Ϫ. Both
anions can be isolated from the solutions of copper() bromide
depending on the nature of the cation present.11,12 A further
possibility was that the product should be formulated as
Cu2L1Br3ؒCuBr. In order to resolve the ambiguity the bulk
sample was recrystallised from methanol and crystal structure
determinations were carried out which showed that the
recrystallised complex was [Cu2L1Br3] 2. This suggests that the
ordination site is supplied by a non-bridging bromide, Br(3).
The two pyramids are not coplanar but are twisted at an angle of
90Њ to each other and the two non-bridging bromides Br(1) and
Br(3) adopt a “trans” configuration with respect to each other.
Selected bond lengths and angles are shown in the caption to
Fig. 3. Those associated with the ligand are normal, as has been
found throughout the structural studies. The Cu(1)–O(1)–Cu(2)
bridging angle is 110.1Њ for the endogenous bridge and there is a
smaller angle of 77.8Њ for the Cu(1)–Br(2)–Cu(2) exogeneous
bridge. The Cu(1)–O(1) distance of 1.96 Å is slightly shorter
than the Cu(2)–O(1) bond length of 1.97 Å. There is quite a
marked difference between the two values of Cu–Br(2) for the
two copper atoms ; Cu(1)–Br(2) has a bond length of 2.68 Å
compared to the Cu(2)–Br(2) bond length of 2.44 Å indicating
an asymmetry found in the (µ-phenoxy)(µ-bromo) bridging
unit. The two apical copper–bromide interactions are longer,
Cu(1)–Br(2) 2.68 and Cu(2)–Br(3) 2.63 Å, than the equatorial
copper–bromide distances, Cu(1)–Br(1) 2.43 and Cu(2)–Br(2)
2.44 Å. The Cu–N(iminic) bond length of 1.97 Å is 0.77 Å
shorter than the Cu–N(aminic) bond and this closer approach
may reflect a co-ordination preference that copper() may have
for an unsaturated imine nitrogen compared to its saturated
nitrogen counterpart. The index of the degree of trigonality (τ)
within the structural continuum between square pyramidal,
(τ = 0) and trigonal bipyramidal (τ = 1) geometries14 is 0.054
at Cu(2) and 0.34 at Cu(1) indicating that Cu(1) has a slightly
more square pyramidal character than Cu(2).
bulk sample was [Cu2L1Br3]CuBr and that CuBr had been
loosely associated in the original precipitation.
The crystal structure of complex 2 was solved both at room
temperature and at 150 K using different crystals from the
recrystallised sample. The gross molecular features are found to
be the same at both temperatures whereas there is a difference
in the space group (P21/c at 293 K† and Pbca at 150 K) suggest-
ing that a phase transition has occurred and that 2 is capable of
existing in at least two polymorphic forms. The higher temper-
ature structure is less accurate and so only the details of the
structure solved at 150 K are discussed further.
The structure, illustrated in Fig. 3, confirms that complex 2
has the molecular formulation Cu2L1Br3. The Cu ؒ ؒ ؒ Cu separ-
ation of 3.221 Å is slightly longer than found in the structure of
a related symmetrical dinuclear complex, 3.151 Å.13 The co-
ordination environment at each copper atom is a square pyram-
idal arrangement derived from donor atoms (N2Br2O) from the
ligand and the bromides. At Cu(1) the base of the pyramid
stems from the N2O compartment derived from the aminic side-
arm of the ligand and a counter anionic non-bridging bromide,
Br(1), with the bridging bromide, Br(2), providing the apex. At
Cu(2) the base is derived from the N2O compartment provided
by the iminic side-arm of the ligand,with the fourth corner of
the base occupied by the bridging bromide, Br(2); the apical co-
Although crystals suitable for structural study were not
recovered from the reaction of HL1 with Cu[ClO4]2ؒ6H2O,
crystals were obtained when the substitution pattern on the
secondary amine was juxtaposed and N,N-diethyl-NЈ-methyl-
ethane-1,2-diamine was used to produce HL2. A structure was
obtained and confirmed that the product of the complexation
reaction was [Cu2L2(OH)(ClO4)]ClO4 3. The crystal structure,
† a = 14.790(16), b = 15.400(13), c = 11.360(8) Å, β = 101.75(7)Њ,
V = 2533(4) Å3.
1852
J. Chem. Soc., Dalton Trans., 2000, 1849–1856

View Article Online
Fig.
4
The molecular structure of the cation in [Cu2L2(OH)-
(ClO4)]ClO4 3. Selected bond lengths (Å) and angles (Њ) at the metal
atoms: Cu(1)–O(2), 1.910(4); Cu(1)–N(2), 2.037(6); Cu(1)–O(1),
1.974(4); Cu(1)–N(1), 1.928(6); Cu(1) ؒ ؒ ؒ Cu(2), 2.9596(13); Cu(2)–
O(2), 1.940(4); Cu(2)–N(4), 2.005(5); Cu(2)–O(1), 1.979(4); and Cu(2)–
N(3) 2.030(5); O(2)–Cu(1)–N(1), 169.5(2); O(2)–Cu(2)–O(1), 78.33(19);
O(2)–Cu(1)–O(1), 79.16(19); O(2)–Cu(2)–N(4), 171.0(2); N(1)–Cu(1)–
O(1), 91.7(2); O(1)–Cu(2)–N(4), 93.1(2); O(2)–Cu(1)–N(2), 103.1(2);
O(2)–Cu(2)–N(3), 99.2(2); N(2)–Cu(1)–N(1), 86.0(2); O(1)–Cu(2)–
N(3), 169.1(2); O(1)–Cu(1)–N(2), 177.6(2); N(4)–Cu(2)–N(3), 88.8(2);
Cu(1)–O(1)–Cu(2), 96.96(19); and Cu(1)–O(2)–Cu(2), 100.5(2).
Fig. 5 The atom connectivity in the cation from [NiLA(OH2)]2-
[ClO4]2ؒ4H2O 4.
lisation gave the complexes 4 and 5. Instead of recovering
µ-hydroxo-bridged dinuclear complexes, by analogy with the
reaction of HL1 and HL2 with Cu(ClO4)2ؒ6H2O, hydrolysis of
the pendant Schiff base arm in the proligand has occurred.
Recrystallisation of the complex from the reaction of HL1 with
Ni(ClO4)2ؒ6H2O gave a batch of green crystals.
The crystal structure showed that a dinuclear nickel() com-
plex had been formed and that the imine bond of the unsym-
metrical unsaturated proligand had undergone hydrolysis to
give the corresponding aldehyde HLA and that the complex was
[NiLA(OH2)]2[ClO4]2ؒ4H2O 4. The two molecules of the hydro-
shown in Fig. 4 together with the numbering scheme, reveals the
presence of a dinuclear core structure with a Cu ؒ ؒ ؒ Cu inter-
nuclear separation of 2.96 Å. This is a closer internuclear
approach than that found in the (µ-bromo) bridged complex 2
and reflects the smaller spatial requirements of the hydroxide
anion relative to the bromide anion.
Selected bond lengths and angles are shown in the caption to
Fig. 4. Those associated with the ligand are normal. Each
copper atom has a square pyramidal arrangement of N2O3
donor atoms; the τ values are 0.13 for Cu(1) and 0.03 for Cu(2).
The two nitrogens at Cu(1) are supplied by the iminic side-arm
and at Cu(2) the two nitrogens are from the aminic pendant
arm. The co-ordination environment at each metal is completed
by the oxygen atoms from an endogenous µ-phenoxy bridge
derived from the compartmental ligand itself, an exogenous
µ-hydroxy bridge and an oxygen atom from a bridging biden-
tate perchlorate anion; Cu(1)–O(7) 2.53 and Cu(2)–O(8) 2.59
Å. Both of the perchlorate counter anions are somewhat dis-
ordered with a 47:53% occupancy at Cl(1) and a 46:54% occu-
pancy at Cl(2). The double bridged (µ-phenoxy)(µ-hydroxy)
unit was found to be slightly asymmetric with Cu(1)–O(1) 1.97,
Cu(2)–O(1) 1.98, Cu(1)–O(2) 1.91 and Cu(2)–O(2) 1.94 Å.
The close approach of the two metals gives rise to a small
µ-phenoxy bridging angle of 97.0Њ and a hydroxy bridging angle
of 100.5Њ.
lysed ligand then assumed a typical octahedral arrangement
around each nickel atom, resulting in a dimeric dinuclear nickel
structure with a Ni ؒ ؒ ؒ Ni distance of 3.088 Å. The molecular
structure is shown in Fig. 5.
Each nickel atom is co-ordinated by the two nitrogen atoms
from the saturated pendant arm moiety and the octahedral co-
ordination at each metal is completed by interaction with the
bridging phenolic oxygen atom, the carbonyl oxygen atom and
an oxygen atom from a water molecule. The water molecules lie
on either side of the ligand plane and so are “trans” to each
other. The phenolic oxygen atoms serve as non-symmetric
di(µ-phenoxy) bridges [Ni(1)–O(1)#1 2.017(5), O(1)–Ni(1)
2.029(7) Å, Ni(1)#1–O(1)–Ni(1) 99.5(3)Њ] between the two nickel
atoms, and the perchlorate anions are not co-ordinated. There
is considerable disorder in the cation and although this has
caused difficulty with refinement of the bond lengths and
angles to an acceptable level and has inhibited accurate
determination of the molecular planes and dihedral angles
pertinent to full interpretation of the magnetic data, the gross
In contrast to the reactions with CuII the reaction of HL1 and
HL2 with Ni(ClO4)2ؒ6H2O in methanol followed by recrystal-
J. Chem. Soc., Dalton Trans., 2000, 1849–1856
1853

View Article Online
stereochemical features of the complex are secure. The co-
ordinating aminic arm, the methyl substituent on the phenyl
ring and the keto-oxygen atom are disordered over two sites
with a 56% occupancy of the major component; it is this major
component that is depicted in Fig. 5. The perchlorate anion is
disordered over two positions (52%:48%) and one water of
solvation is also disordered.
The reason for the hydrolysis of the nickel complex on
recrystallisation is not immediately apparent as the correspond-
ing copper complexes recrystallise intact. It may be that the
combination of a dinickel() metal centre, which could act as a
focus for Lewis acid activity as in the proposed mechanism for
the hydrolysis of urea by urease, and the use of a protic solvent
like methanol encourages the hydrolysis. The acidity of the pro-
ton from any water molecule co-ordinated to the nickel would
be enhanced so generating a nucleophile such as a hydroxide
anion which could then attack the positive carbon of the imine
bond and thus initiate a retro-Schiff base reaction. Further-
more the nickel() with its d8 configuration strongly favours an
octahedral environment and so in the presence of weakly co-
ordinating perchlorate anions water can readily add to the
empty axial sites provided by coplanar co-ordination by the
ligand; this may then serve as a driving force for the hydrolysis.
It is therefore possible that use of a polar aprotic solvent such as
acetonitrile would discourage the hydrolysis reaction.
Fig. 6 The molecular structure of the cation in [NiLC(OH2)]2-
[ClO4]2ؒ3CH3OH 5. Selected bond lengths (Å) and angles (Њ) at the
metal atoms: Ni(1)–O(1), 2.041(5); Ni(1)–O(1)#1, 2.028(5); Ni(1)–
O(2)#1, 2.045(5); Ni(1)–N(1), 2.090(6); Ni(1)–N(2), 2.100(6); Ni(1)–
O(3), 2.143(5); and Ni(1) ؒ ؒ ؒ Ni(1)#1 3.107; Ni(1)#1–O(1)–Ni(1), 99.6(2);
O(1)#1–Ni(1)–O(1), 80.4(2); O(1)#1–Ni(1)–O(2)#1, 90.0(2); O(1)–Ni(1)–
O(2)#1, 169.12(19); O(1)#1–Ni(1)–N(1), 171.2(2); O(1)–Ni(1)–N(1),
91.0(2); O(2)#1–Ni(1)–N(1), 98.6(2); O(1)#1–Ni(1)–N(2), 94.6(2); O(1)–
Ni(1)–N(2), 94.9(2); O(2)#1–Ni(1)–N(2), 91.1(2); N(1)–Ni(1)–N(2),
83.9(3); O(1)#1–Ni(1)–O(3), 87.2(2); O(1)–Ni(1)–O(3), 88.7(2); O(2)#1
–
Ni(1)–O(3), 85.6(2); N(1)–Ni(1)–O(3), 94.7(2); and N(2)–Ni(1)–O(3),
176.2(2).
The reaction of HL2 with Ni[ClO4]2ؒ6H2O in methanol fol-
lowed by recrystallisation gave a further unexpected result. The
preparation of the bulk sample followed that for the reaction of
nickel with HL1 with the IR spectrum of the product showing
a peak at 1636 cmϪ1, indicative of an imine bond. Recrystallis-
ation of the bulk sample from methanol yielded crystals as
green blocks suitable for X-ray diffraction studies. The crystal
structure shown in Fig. 6 indicated that the sample has under-
gone hydrolysis giving a product 5 with a similar structure to
that found for complex 4 but with the notable difference in
that there is the loss of a C2H4 moiety from each of the
terminal NEt2 groups on the aminic pendant arm to give a
secondary aminic HNEt functionality. The formulation of the
complex is therefore [NiLC(OH2)]2[ClO4]2ؒ3CH3OH. The
reproducibility of the C2H4 loss is evidenced by the preparation
and structural characterisation of an analogous complex,
[NiLC(NCCH3)]2[BF4]2ؒCH3CN, from the reaction of HL2 with
Ni(BF4)2ؒnH2O in methanol followed by recrystallisation from
acetonitrile.15
Selected bond lengths and angles for the cation in complex 5
are given in the caption to Fig. 6. There is again an octahedral
arrangement of N2O4 donors around each nickel atom supplied
by two molecules of hydrolysed ligand and a water molecule.
There are also three methanol molecules of solvation. The
Ni ؒ ؒ ؒ Ni internuclear distance is 3.107 Å which compares
favourably with 3.088 Å for complex 4. Each metal is doubly
bridged by a non-symmetric di(µ-phenoxy) bridge from the
hydrolysed ligand (Ni–O, 2.041(5) and 2.028(5) Å). The co-
ordination is completed by two sp3 nitrogen donors from the
aminic pendant arm (Ni–N, 2.090(6) and 2.100(6) Å), an
oxygen from the carbonyl group (Ni–O, 2.045(5) Å) and an
oxygen from a water molecule (Ni–O, 2.143(5) Å).
Although the precise mechanism for this ligand degradation
is not yet known it is possible to draw a parallel with observ-
ations made on the stability of the copper complexes of
N,N,NЈ,NЈ-tetraethylethane-1,2-diamine, [Cu(Et2NCH2CH2-
NEt2)Cl2]16 and N,N,NЈ,NЈ-tetrabenzylethane-1,2-diamine,
[Cu(Bz2NCH2CH2NBz2)Cl2].17 These complexes have been
found to undergo similar loss of a C2H4 moiety on recrystallis-
ation to generate [Cu(Et2NCH2CH2NHEt)Cl2] and [Cu(Bz2-
NCH2CH2NHBz)Cl2]. Acetaldehyde and benzaldehyde were
detected as the respective accompanying products and a
reduction-oxidation reaction proposed to occur. Interestingly
this same reaction was not detected for N,N,NЈ,NЈ-tetra-
methylethane-1,2-diamine [Cu(Me2NCH2CH2NMe2)Cl2] so
paralleling the lack of activity at the NMe2 pendant in complex
4. It has also been suggested that the bulkier nature of the ethyl
group relative to the methyl group can induce strain in the
molecule and hence influence the reactivity. A further possi-
bility is that either in the reaction carried out in methanol or
during the process of slow crystallisation, with exposure to the
atmosphere, the conditions have been set up for a Hofmann
elimination reaction to occur. Protonation of the aminic arm,
followed by nucleophilic attack by hydroxide anion, would then
yield a secondary amine as observed (Scheme 2). The plausi-
bility of the protonation step is illustrated by the ready proton-
ation of the morpholino-arm in a mononuclear copper()
complex of the unsymmetrical dinucleating ligand 4-bromo-2-
(2-hydroxyethyliminomethyl)-6-(morpholin-4-yl-methyl)phenol
(HL3).18
1854
J. Chem. Soc., Dalton Trans., 2000, 1849–1856

View Article Online
Bowers equation (1) where χA is the magnetic susceptibility per
Nβ2g2
Nβ2
χA =
[3 ϩ exp(Ϫ2J/kT)]Ϫ1(1 Ϫ ρ) ϩ ρ(
) ϩ Nα
(1)
kT
kT
Cu atom, N is Avogadro’s number, β the Bohr magneton, k the
Boltzmann constant, J the exchange integral, T the absolute
temperature, Nα the temperature-independent paramagnet-
ism and ρ the fraction of paramagnetic impurity. The para-
magnetic impurity is presumed to be a monomeric copper()
species. A tolerable magnetic fitting is obtained with this
equation, using the magnetic parameters of J = Ϫ215 cmϪ1
,
g = 2.10, Nα = 60 × 10Ϫ6 cm3 molϪ1 and ρ = 0.1. The result
clearly indicates a strong antiferromagnetic spin exchange in 3.
The distinct magnetic properties of 2 and 3 can be explained by
their different core structures. In the case of 2 both Cu(1) and
Cu(2) have a square-pyramidal geometry sharing the phenolic
oxygen O(1) and the bridging bromide ion Br(2). The basal
plane for Cu(1) is formed by O(1), N(1), N(2) and Br(1), and
Br(2) occupies the apical site but the basal plane for Cu(2)
is formed by O(1), Br(2), N(3) and N(4) with terminal Br(3)
occupying the apical site. The O(1) can contribute to an anti-
ferromagnetic exchange but Br(1) contributes to a ferro-
magnetic exchange between the two copper ions. This is an
example of counter-complementarity of the bridging groups in
a doubly bridged dinuclear copper() complex, and the overall
magnetic interaction of this complex is very weakly antiferro-
magnetic.
Fig. 7 µeff vs. T plots for compounds 2–5. Solid lines are drawn based
on eqn. (1) for 2 and 3 and eqn. (2) for 4 and 5, using the magnetic
parameters given in the text.
In the case of complex 3 the two bridging oxygens, O(1) and
O(2), are shared at the equatorial positions of Cu(1) and Cu(2).
This is an example of complementarity of bridging groups, and
the two oxygens contribute to an antiferromagnetic spin
exchange to produce a strong antiferromagnetic interaction
(J = Ϫ215 cmϪ1). Both complexes showed only a broad EPR
signal centered around g = 2.0 when measured as powdered
samples; the broadening is assigned to a dipolar interaction in
2 and to spin exchange in 3.
Scheme 2
Complexes 2 and 3 are also differentiated by our preliminary
cyclic voltammetric studies (using a glassy carbon electrode in
dmf and a Ag–AgCl reference electrode). Complex 2 shows a
quasi-reversible couple at Ϫ0.03 V (vs. Ag–AgCl) [Epc = Ϫ0.10
V and Epa = ϩ0.04 V] and a reversible couple at Ϫ0.85 V
[Epc = Ϫ0.90 V and Epa = Ϫ0.80 V]. These can be attributed to
stepwise reductions at the metal centre, CuIICuII → CuIICuI
and CuIICuI → CuICuI, respectively. Complex 3 shows a
weak couple at Ϫ0.30 V and an irreversible couple near Ϫ0.9 V.
The former can be ascribed to the monomeric copper impurity.
Owing to the irreversible nature of the wave near Ϫ0.9 V and
the presence of the monomeric impurity, we could not deter-
mine the number of electrons transferred at Ϫ0.9 V. The
facile reduction of 2 relative to 3 certainly relates to the
square-pyramidal geometry about the Cu atoms in 2 and the
involvement of the soft donor BrϪ anion. The reduction of the
Cu atoms of 3 is more difficult because the square-planar
environment about the metals is an unfavourable geometric
environment for CuI.
The room-temperature magnetic moments for complexes 4
and 5 (per Ni atom) are 2.85 and 2.94 µB, respectively, which are
compared to the spin-only value for isolated nickel() atoms of
2.83 µB. The moments gradually decreased with decreasing
temperature to 1.85 and 2.39 µB, respectively, at 82 K, suggest-
ing an antiferromagnetic spin-exchange interaction within each
molecule (Fig. 7). Magnetic analyses were carried out using the
magnetic susceptibility expression (2) for an (S1 = 1)–(S2 = 1)
The nature of the products confirmed by the structural
studies draws attention to the instability of the nickel() com-
plexes when left to stand in solution under ambient conditions
or during redissolution for recrystallisation; in order to grow
crystals of the complexes described herein it was necessary to
leave the samples standing for many weeks. In the hydrolysis
process the precursor aldehyde is regenerated and the nickel()
complex of the precursor ligand is produced.
Properties
The physical measurements were carried out on recrystallised
samples. The magnetic moment of complex 2 (per Cu atom) is
1.85 µB at room temperature and the moment was practically
independent of temperature down to liquid nitrogen temper-
ature (1.83 µB at 82 K) (see Fig. 7). Evidently, no appreciable
magnetic spin exchange operates between the pair of copper()
ions in 2. On the other hand, the magnetic moment of 3 is
subnormal at room temperature (1.19 µB per Cu) and decreases
with decreasing temperature to 0.63 µB at 82 K (Fig. 7). The
result suggests the operation of an antiferromagnetic inter-
action between the copper() ions. Contamination with a sig-
nificant amount of paramagnetic impurity is suggested from
the χA vs. T curve showing a minimum around 120 K and an
increasing tendency below this temperature. Magnetic analyses
for 3 were therefore carried out using the modified Bleaney–
(Ng2β2/3kT)[5 ϩ exp(Ϫ4J/kT)]
χA =
ϩ Nα
(2)
[5 ϩ 3exp(Ϫ4J/kT) ϩ exp(Ϫ6J/kT)]
system based on the Heisenberg model. Fairly good magnetic
J. Chem. Soc., Dalton Trans., 2000, 1849–1856 1855

View Article Online
Conclusion
The reaction of copper() with the unsymmetric Schiff bases
HL1 and HL2 gave homodinuclear complexes. The structure of
[Cu2L1Br3] 2 revealed that the asymmetry in the molecule was
not restricted to that implicit to the ligand (donor asymmetry)
but that the two square-pyramidal arrangements at the copper
atoms were subtly different with the bridging Br(2) atom
serving as the apex for the geometrical arrangement at Cu(1)
and a corner of the pyramidal base at Cu(2) leading to a non-
coplanarity of the two square pyramids. This distortion was
reflected in the magnetic properties of 2, no antiferromagnetic
coupling being detected between the two Cu atoms. A strong
antiferromagnetic coupling (J = Ϫ215 cmϪ1) was observed in
complex [Cu2L2(OH)(ClO4)]ClO4 3, where the presence of the
hydroxy bridge, complementary to the phenoxy bridge, aids
coupling. The reactions of HL1 and HL2 with nickel() led to
hydrolysis reactions in which the pendant Schiff base was
cleaved; with HL1 the homodinuclear complex [NiLA(OH2)]2-
[ClO4]2ؒ4H2O was formed and with HL2 hydrolysis was
followed by elimination of an C2H4 moiety from one C2H5
of the terminal NEt2 group of the iminic side arm to leave a
NHEt group and the dinuclear complex [NiLC(OH2)]2[ClO4]2ؒ
3CH3OH.
Acknowledgements
Fig. 8 The dinickel complexes viewed along the Ni ؒ ؒ ؒ Ni vector: (a) 5
and (b) the major component of 4.
We thank the EPSRC for a studentship (to S. R. H.) and for
funds towards the purchase of the diffractometer; we also
thank the British Council and the Daiwa Anglo-Japanese
Foundation for their support and the Japanese Ministry of
Education for the award of a Monbusho Research Experience
Fellowship for Young Foreign Researchers (to S. R. H.).
simulations are obtained with this equation, as seen in Fig. 7,
using magnetic parameters of J = Ϫ31 cmϪ1, g = 2.15, Nα =
200 × 10Ϫ6 cm3 molϪ1 for 4 and J = Ϫ17 cmϪ1, g = 2.15, Nα =
200 × 10Ϫ6 cm3 molϪ1 for 5.
Thompson and co-workers have studied the magnetostruc-
tural correlation for di(µ-phenoxo)dinickel() complexes
derived from the macrocyclic ligand H2L4 and reported a linear
References
1 D. E. Fenton and H. Okawa, Chem. Ber./Recueil, 1997, 130, 433 and
references therein.
2 T. N. Sorrell, Tetrahedron, 1989, 45, 3.
3 J. H. Satcher, Jr., M. W. Droege, T. J. R. Weakley and R. T. Taylor,
Inorg. Chem., 1995, 34, 3317.
4 D. E. Fenton, Adv. Inorg. Bioinorg. Mech., 1983, 2, 187.
5 J. D. Crane, D. E. Fenton, J. M. Latour and A. J. Smith, J. Chem.
Soc., Dalton Trans., 1991, 2979.
6 M. Lubben, R. Hage, A. Meetsma, K. Byma and B. L. Feringa,
Inorg. Chem., 1995, 34, 2217.
7 J. Reim and B. Krebs, J. Chem. Soc., Dalton Trans., 1997, 3793.
8 E. Lambert, B. Chabut, S. Chardon-Noblat, A. Deronzier, G.
Chottard, A. Bousseksou, J.-P. Tuchagues, J. Laugier, M. Dardet
and J.-M. Latour, J. Am. Chem. Soc., 1997, 119, 9424.
9 S. Uozumi, M. Ohba, H. Okawa and D. E. Fenton, Chem. Lett.,
1997, 673.
10 SHELXTL, An integrated system for solving and refining crystal
structures from diffraction data (Revision 5.1), Bruker AXS LTD,
Madison, WI, USA, 1997.
11 M. Asplund and S. Jagner, Acta Chem. Scand., Ser. A, 1984, 38, 135.
12 M. Asplund, S. Jagner and N. Nilsson, Acta Chem. Scand., Ser. A,
1983, 37, 57.
13 C. J. O’Connor, D. Firmin, A. K. Pant, B. R. Babu and E. D.
Stevens, Inorg. Chem., 1986, 25, 3000.
14 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C.
Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349.
15 S. Clunas and D. E. Fenton, unpublished work.
16 R. C. E. Belford, D. E. Fenton and M. R. Truter, J. Chem. Soc.,
Dalton Trans., 1972, 2345.
17 K. C. Patel, S. R. Shirali and D. E. Goldberg, J. Inorg. Nucl. Chem.,
1975, 37, 1659.
18 H. Adams, N. A. Bailey, D. E. Fenton and G. Papageorgiou,
J. Chem. Soc., Dalton Trans., 1995, 1883.
19 K. K. Nanda, L. K. Thompson, J. N. Bridson and K. Nag, J. Chem.
Soc., Chem. Commun., 1994, 1337.
relationship between the Ni–O–Ni bridging angle and the
exchange integral (ϪJ); the larger the Ni–O–Ni angle the
stronger is the antiferromagnetic interaction.19 The complexes 4
and 5 significantly differ in exchange integral (Ϫ31 and Ϫ17
cmϪ1 respectively) (in spite of their similar Ni–O–Ni angles (99.5
and 99.6Њ respectively). The equatorial ligating atoms and
nickel atoms in 5 are essentially coplanar (Fig. 8a) and the
exchange integral can be incorporated into the above linear
relationship. However for 4 it is possible that there are other
factors influencing the deviation from the relationship. The dis-
ordered nature of the structure precludes an accurate determin-
ation of the dihedral angles and planarities at the dinuclear
centre but there is a distinct “stepping” of the conformation of
4 which leads to distortion of the equatorial environments at
the nickel atoms (Fig. 8b).
1856
J. Chem. Soc., Dalton Trans., 2000, 1849–1856