Copper(II) complexes of thioether-substituted salcyen and salcyan derivatives and their silver(I) adducts

Article abstract of DOI:10.1039/b505972k

New syntheses are reported of 5-tert-butyl-2-hydroxy-3- methylsulfanylbenzaldehyde, 5-tert-butyl-2-hydroxy-3-phenylsulfanyl- benzaldehyde, and salcyen (H₂L¹-H₂L ³) and salcyan (H₂L⁴-H₂L ⁶)-type ligands derived from these aldehydes and from 5-tert-butyl-2-hydroxybenzaldehyde. The complexes [CuL] (L^2- = [L¹]^2--[L⁶]^2-) bearing sulfanyl substituents each show two distinct voltammetric ligand-based oxidations under the same conditions, the first of which is chemically reversible. The first oxidation product is much longer lived by coulometry for the salcyen than for the salcyan ligand complexes, despite the latter having a substantially lower oxidation potential. The lifetimes of all the ligand oxidation products in this system are substantially smaller than for similar compounds derived from 3,5-di(tert-butyl)-2-hydroxybenzaldehyde (Dalton Trans., 2004, 2662). Attempted chemical oxidation of the Schiff base compounds using AgBF₄ yielded instead stable silver(I) adducts. A crystal structure of one such compound showed that the Ag atom was coordinated in a slightly bent geometry by the two ligand sulfanyl groups, with two additional long-range Ag ... O interactions to the phenoxide donors. EPR spectra showed that some of these silver adducts dimerise in CH₂Cl₂, probably through basal, apical intermolecular Cu-O ... Cu bridging. In contrast the parent copper(II) complexes are all monomeric in this solvent by EPR. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b505972k

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Copper(II) complexes of thioether-substituted salcyen and salcyan
derivatives and their silver(I) adducts
Isabelle Sylvestre,a Joanna Wolowska,b Colin A. Kilner,a Eric J. L. McInnesb and
Malcolm A. Halcrow*a
a School of Chemistry, University of Leeds, Woodhouse Lane, Leeds, UK LS2 9JT.
E-mail: M.A.Halcrow@leeds.ac.uk
b EPSRC CW EPR Service, School of Chemistry, University of Manchester, Oxford Road,
Manchester, UK M13 9PL
Received 28th April 2005, Accepted 2nd August 2005
First published as an Advance Article on the web 26th August 2005
New syntheses are reported of 5-tert-butyl-2-hydroxy-3-methylsulfanylbenzaldehyde, 5-tert-butyl-2-hydroxy-
3-phenylsulfanyl-benzaldehyde, and salcyen (H2L1–H2L3) and salcyan (H2L4–H2L6)-type ligands derived from these
aldehydes and from 5-tert-butyl-2-hydroxybenzaldehyde. The complexes [CuL] (L2− = [L1]2−–[L6]2−) bearing sulfanyl
substituents each show two distinct voltammetric ligand-based oxidations under the same conditions, the first of
which is chemically reversible. The first oxidation product is much longer lived by coulometry for the salcyen than for
the salcyan ligand complexes, despite the latter having a substantially lower oxidation potential. The lifetimes of all
the ligand oxidation products in this system are substantially smaller than for similar compounds derived from
3,5-di(tert-butyl)-2-hydroxybenzaldehyde (Dalton Trans., 2004, 2662). Attempted chemical oxidation of the Schiff
base compounds using AgBF4 yielded instead stable silver(I) adducts. A crystal structure of one such compound
showed that the Ag atom was coordinated in a slightly bent geometry by the two ligand sulfanyl groups, with two
additional long-range Ag · · · O interactions to the phenoxide donors. EPR spectra showed that some of these silver
adducts dimerise in CH2Cl2, probably through basal, apical intermolecular Cu–O · · · Cu bridging. In contrast the
parent copper(II) complexes are all monomeric in this solvent by EPR.
Introduction
There is continuing interest in the spectroscopic and structural
characterisation of metalcomplexes of phenoxylradical centres.1
In large part, this is driven by the goal of understanding the
chemistry of proteinaceous tyrosyl radicals, which often occur
in proximity to a metal centre.2 One such phenoxyl biosite is
found in the enzyme galactose oxidase (“GOase”), whose active
site contains a mononuclear copper(II) complex of an oxidisable
tyrosinate residue.3 This reactive tyrosine bears two unusual
structural features.4 First, it has been chemically modified by
an oxidative cross-linking to a neighbouring cysteine side-
chain, yielding an ortho-alkylsulfanyl substituent.5 Second, it
is protected by a p–p interaction to a tryptophan residue, which
sterically shields the modified tyrosine ring from exogenous
attack. Many groups, including ourselves, have synthesised new
copper(II) phenoxide complexes, with or without thioether sub-
stitution at the phenoxide ligand, with the aim of understanding
the unusual structural features of the GOase copper biosite.1,6,7
For our own part, we have separately studied the effects of
thioether substituents,8–11 and of p–p interactions,11–13 on the
stability and spectra of copper complexes of oxidisable arenes.
As part of this work, we obtained the first copper(II)/phenoxyl
radical species whose absorption spectrum resembled that of the
During our own work, three other groups have reported
oxidised GOase protein.8,9
that the closely related complexes [Cu(L7)]18,19 and [Ni(L8)]20
both yield long-lived phenoxyl radical species following electro-
chemical or chemical oxidation. This contrasts with this study,
As our next goal in this area, we decided to target cop-
per(II)/phenoxide complexes, with biologically relevant sub-
stituents at the phenoxyl centre(s), that show low oxidation
potentials and so might be amenable to isolation. In the light
of our earlier results7,8 and some work from Stack’s group,14,15
we reasoned that copper(II) complexes of the salcyen derivatives
H2L2 and H2L3, and their reduced salcyan congeners H2L5 and
H2L6, might show these particular properties. We describe here
the synthesis of H2L1–H2L6 and their copper(II) complexes,
including the unsubstituted derivatives H2L1 and H2L4 that
were also prepared for purposes of comparison. The complex
[Cu(L1)] has been synthesised previously, although its oxidative
electrochemistry and crystal structure were not reported.16,17
where we have found the oxidation products [CuL]+ (L2−
=
[L1]2−–[L6]2−) to be stable only at very low temperatures on the
voltammetric timescale.
Results and discussion
Ligand and complex syntheses and structures
The synthesis of several 5-tert-butyl-2-hydroxy-3-organosulfa-
nylbenzaldehydes has been previously described, by electrophilic
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9
3 2 4 1
©

View Article Online
◦
˚
Table 1 Selected bond lengths (A) and angles ( ) for the two solvates
of 1 in this study
attack of tetralithiated 2-(5-tert-butyl-2-hydroxyphenyl)-1,3-
dimethylimidazoline at the appropriate diorganodisulfide, fol-
lowed by deprotection of the aldehyde by aqueous acid.14 We
have used this method successfully in the past to prepare 3-
substituted 2-hydroxy-5-methylbenzaldehyde derivatives.8 How-
ever, it gave very low yields in our hands when applied to the 5-
tert-butyl-2-hydroxybenzaldehydes required for this work, and
we therefore sought an alternative synthesis. We were able to
prepare the aldehydes shown in Scheme 1, in viable yields, by
a two-step procedure. The first step of this process21 was very
sensitive to the identity of the disulfide reagent, and did not
give useful yields when diethyl- or dibutyl-disulfides were used.
The choice of Lewis acid was also important, since similar
reactions using AlCl3 rather than FeCl3 gave only starting
material following the usual work-up. That may imply that
the iron salt, which is used in excess, acts as an oxidising
agent towards the disulfide reagent as well as a Lewis acid
during the reaction. Step (ii) in Scheme 1 is a very reliable
method for the synthesis of salicylaldehyde derivatives that
was first reported by Hofslokken and Skattebol.22 The three
benzaldehydes in Scheme 1 were converted to the Schiff bases
H2L1–H2L3 by condensation with half an equivalent of rac-trans-
1,2-diaminohexane; H2L1–H2L3 were then reduced to H2L4–
H2L6 using NaBH4.
1·H2O
1·CH2Cl2
Cu(1)–N(8)
Cu(1)–O(16)
Cu(1)–N(21)
Cu(1)–O(29)
1.9398(18)
1.9070(15)
1.9475(18)
1.9028(16)
1.9464(18)
1.9096(15)
1.9444(18)
1.8931(15)
N(8)–Cu(1)–O(16)
N(8)–Cu(1)–N(21)
N(8)–Cu(1)–O(29)
O(16)–Cu(1)–N(21)
O(16)–Cu(1)–O(29)
N(21)–Cu(1)–O(29)
93.57(7)
84.40(7)
166.55(8)
168.88(7)
90.01(7)
94.48(7)
93.81(7)
83.96(7)
170.62(7)
168.39(7)
89.86(6)
94.12(7)
crystals contain four-coordinate, approximately square-planar
Cu centres, with no apical Cu · · · O interaction shorter than
i
i
˚
Cu(1) · · · O(29 ) = 4.4683(16) A (1·H2O) or Cu(1) · · · O(16 ) =
˚
4.6146(16) A (1·CH2Cl2; symmetry code i: 1–x, 1–y, –z for both
structures) (Table 1, Fig. 1). The bond lengths in each molecule
are identical to within 3 sus, while the bond angles at Cu show
only small differences. However, the chiral twist of the [L4]2−
ligand in 1·H2O is slightly greater than in 1·CH2Cl2 according to
the dihedral angle between the two [L4]2− phenyl rings [C(10)–
C(15)] and [C(23)–C(28)], which is 27.44(9)◦ in 1·H2O and
21.03(8)◦ in 1·CH2Cl2. We tentatively suggest that this small
structural difference is the most likely origin of the different
colours of the two solvates. Importantly, since both solvates of 1
show no basal, apical Cu–O · · · Cu bridging between molecules,
intermolecular interactions of this type cannot be the cause of
the different colours of their crystals.23,26
Scheme 1 Synthesis of salicylaldehyde derivatives used in this work.
Reagents and conditions: (i) RꢀSSRꢀ, FeCl3, toluene, reflux 6 h; then dil.
HCl. (ii) (CH2O)x, MgCl2, NEt3, MeCN, reflux 4 h; then dil. HCl.
The complexes [CuL] (L2− = [L1]2−, 1; L2− = [L2]2−, 2; L2−
=
[L3]2−, 3; L2− = [L4]2−, 4; L2− = [L5]2−, 5; L2− = [L6]2−,
6) precipitated in good yields upon reaction of the relevant
•
H2L ligand with 1 equiv. of Cu(O2CMe)2 H2O in refluxing
MeCN. The ES mass spectra of 1–6 in MeOH all show significant
dimer peaks in addition to the expected molecular ions, implying
that significant association of the compounds takes place under
these conditions. This presumably occurs through intermolec-
ular basal/apical Cu–O · · · Cu bridging interactions.23 For the
salcyan complexes 4–6, significant mass peaks corresponding to
the demetallated ligands H3L+ (H2L = H2L4 − H2L6) and, for 5
and 6, from the species [63Cu2(L)]+ (L2− = [L5]2− or [L6]2−) were
also observed.
The UV/vis spectra of these compounds in CH2Cl2, and
their EPR spectra in this solvent, closely resemble those of
other Cu(II) salcyen and salcyan derivatives (see Experimental
section).23–25 This implies that 1–6 adopt the same approximately
square planar coordination geometries that are exhibited by
other compounds of these types. Unfortunately the X- and Q-
band EPR spectra of 1–6 as neat powders are very broad at
293 K and 110 K, and it was only for 3 and 4 that accurate g-
values could be measured. Importantly the g-values of 3 and 4,
and A1{63,65Cu} for 4, in the solid state and in frozen solution are
very similar to each other. That strongly implies that the solid
state and solution structures of these compounds, at least, are
essentially the same.
Fig. 1 View of the asymmetric unit in the crystal structure of 1·H2O,
showing the atom numbering scheme employed. All C-bound H atoms
have been omitted for clarity. Other details as for Fig. 1. The complex
molecule in 1·CH2Cl2 is visually indistinguishable from the one shown
here, and uses the same atom numbering scheme.
The hydrate crystals contain a water molecule of crystalli-
sation that is not associated with the Cu ion, but which
hydrogen bonds to both [L4]2− O atoms of the same complex
molecule (Fig. 1). The solvent molecule in 1·CH2Cl2 occupies a
similar position in the lattice, and also forms weak C–H · · · O
interactions to both O donors of the same complex molecule.
Single crystals of two different solvates of 1 were obtained
during the course of this work, 1·H2O and 1·CH2Cl2. Inter-
estingly the hydrate crystals are green in colour, while the
CH2Cl2 solvate crystals are red; similar variations in colour have
also been noted in different solvates of [Cu(salen)].23,26 Both
3 2 4 2
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9

View Article Online
Table 2 Selected voltammetric data at 293 K, and coulometric measurements at 196 K, for the complexes in this study. All measurements were
performed in CH2Cl2/0.5 M NnBu4BF4. Potentials quoted are referenced to an internal ferrocene/ferrocenium standard, and were obtained at
100 mV s−1
E1/2 {L/L•}, V [reversibilitya]
n
E1/2{L•/L••}, V [reversibilitya]
n
Epc{Cu(II/I)}, V
Epc{Cu(I/0)}, V
[Cu(L1)] (1)
[Cu(L2)] (2)
[Cu(L3)] (3)
[Cu(L4)] (4)
[Cu(L5)] (5)
[Cu(L6)] (6)
+0.57 [irrev]c
+0.43 [rev]
+0.45 [rev]
+0.08 [irrev]c
+0.03 [rev]
+0.05 [rev]
0.90
—
—
—
−1.55d
−1.81
−1.86
−1.66
−1.43
−1.48
+0.56 [rev]
+0.58 [rev]
—
1.95
2.09
—
0.92
1.03
—
0.3–0.5e
0.93
1.13
−2.06
−2.07
−2.16
+0.18 [irrev]c
+0.34 [irrev]c
a Chemically reversible [rev], quasi-reversible [qrev] or irreversible [irrev]. b From ref. 9. c Epa value quoted. d Lit. E1/2 = −1.86 V.15 e Variable results
obtained.
to 2 and 3, 1-electron oxidised [5•]+ and [6•]+ are stable on
Electrochemistry
the voltammetric timescale of a few seconds, but decompose
The cyclic voltammograms of 1–6 in CH2Cl2/0.5 M NnBu4BF4
rapidly enough in the coulometry experiment (ca. 1 h) to
at 293 K are summarised in Table 2. As for [Cu(L7)]18,19 and
prevent the second oxidation from taking place. The apparently
[Cu(L8)],18 the first oxidations of 4–6 are 0.40–0.45 V less positive
than those of 1–3. The two compounds lacking thioether sub-
stituents, 1 and 4, both show one irreversible ligand oxidation.
The reversibility of these oxidations for both compounds does
not improve significantly on cooling to 196 K. A coulometric
determination of the oxidation of 1 yielded n ≈ 1, while the first
oxidation of 4 gave variable values of n < 1; in both cases, re-
reduction of the oxidised solution at 0 V resulted in near-zero
current flow. The 1-electron nature of the oxidations of 1 and
4 was also confirmed, by comparing their voltammetric peak
heights with those of an equimolar amount of ferrocene. This
demonstrates that the 1-electron oxidised complexes [1•]+ and
[4•]+ are short-lived under these conditions, as expected for 2-
unsubstituted phenoxyl species,1 and decompose too quickly
to allow a second oxidation product to be formed on the
voltammetric timescale.
In contrast, the sulfanyl-substituted complexes 2, 3, 5 and 6
exhibit two distinct oxidations (Table 2). Importantly, the rela-
tive intensities of these two oxidation peaks for these compounds
are the same in CVs run at two different concentrations (0.03
and 0.06 mol dm−3). This implies that both oxidations originate
from the same species, and are not evidence of a monomer/dimer
equilibrium at the electrode23,26,27 (cf. the mass spectra of these
compounds, see above). For 2 and 3, the two oxidations lie at
very similar potentials and are chemically reversible at 293 K
(Table 2). Coulometric determinations of 2 and 3 at 196 K, at
a potential 0.2 V positive of the second oxidation peak, both
yielded n ≈ 2. Cyclic voltammograms of the oxidised solutions
were different from those of the starting materials, showing a
number of poorly defined reduction waves. Consistent with this,
subsequent re-reduction of the oxidised solutions at 0 V gave
n = 0.6–0.7 for both compounds. These data show that the
phenoxide groups of 2 and 3 are oxidised in 1-electron steps.
The one-electron oxidised species [2•]+ and [3•]+ are long-lived
enough to allow [2••]2+ and [3••]2+ to be generated quantitatively;
but, these doubly oxidised final products are unstable under
our coulometry conditions. Unfortunately the potentials of
the first and second oxidations of 2 and 3 are so close as
to make it impossible to prepare [2•]+ and [3•]+ selectively by
controlled potential electrolysis, so that spectroelectrochemical
characterisation of these species was not possible.
shorter lifetimes of [5•]+ and [6•]+, compared to [2•]+ and [3•]+,
are consistent with the previous observation that chemically
•
•
generated [Cu(L8 )]+ decomposes more rapidly than [Cu(L7 )]+.18
Reactions of complexes 1–6 with silver salts
As a first attempt to generate [1•]+–[6•]+ by chemical oxidation,
we treated 1–6 with 1 molar equivalent of dry AgBF4 in CH2Cl2
(E1/2 = +0.65 V28), since a different silver(I) salt was used to
•
successfully prepare [Cu(L7 )]+.18 In contrast, however, none of
these six reactions gave rise to a silver metal precipitate. Rather,
−
they yielded air-stable products that contained the BF4 anion
by IR spectroscopy. The products obtained in this way using
4–6 were both poorly soluble and very photosensitive, which
prevented their full characterisation although the latter property
implies that they contain silver(I). However, the silver-containing
products from 1–3 were soluble crystalline solids, which were
formulated as [CuAg(l-L)]BF4 (L2− = [L1]2−, 7; L2− = [L2]2−,
8; L2− = [L3]2−, 9) by ES mass spectrometry, IR, EPR (see below)
and from a crystal structure of an acetonitrile solvate of 9 which
is described below. Solid 7–9 are also slightly photosensitive,
which is a possible cause of their slightly variable microanal-
yses. This silver adduct formation contrasts with [Cu(L7)] and
[Cu(L8)], which instead undergo one-electron oxidation when
treated with AgSbF6.17 This should be a consequence of the ortho
tert-butyl ligand substituents in the latter complexes, which will
sterically prevent binding of silver(I) to their O donor atoms.
The crystal structure of 9·xCH3CN (x ≈ 1.45) shows a
[Cu(L3)] moiety, complexed to an Ag(I) ion through the two
thioether S atoms and by bridging interactions to the phenoxide
O donors (Table 3, Fig. 2). The Ag–O and Ag–S bond lengths
˚
are similar at 2.411(3)–2.5508(10) A although, given the smaller
˚
˚
covalent radius of O (0.73 A) compared to S (1.02 A), the
Ag–O interactions are likely to be relatively weak. This results
in an unusual, approximately planar [2 + 2] stereochemistry
at Ag(2), that is probably best described as a distorted linear
geometry [S(22)–Ag(2)–S(42) = 151.53(3)◦] with two additional
long-range Ag · · · O interactions. The coordination geometry at
Cu(1) in 9 shows some small differences compared to the two
solvates of 1. Most notably, the Cu(1)–N and Cu(1)–O bonds
in 9 are all slightly shorter than in 1, by up to 0.028(3) and
6
2−
˚
The two oxidations shown by 5 and 6 are more widely spaced
than those of 2 and 3. The first of these oxidations is chemically
reversible at 293 K, but the second is irreversible for both
compounds; scanning past the second oxidation destroys the
reversibility of the first. Coulometric determinations of the first
and second oxidation waves of both compounds at 196 K, were
each consistent with n = 1. As before, cyclic voltammograms
of their oxidised solutions did not resemble those of the initial
samples, while subsequent re-reduction of the oxidised solutions
at 0 V yielded almost no current flow. This shows that, in contrast
0.031(3) A respectively, while the [L ] ligand shows a reduced
dihedral angle between its two phenyl rings [C(11)–C(16)] and
[C(31)–C(36)], of12.90(14)◦. Thesedifferences may reflect ligand
conformational constraints imposed by the coordinated Ag ion.
The (disordered) BF4− anion in the structure forms a weak apical
interaction to Cu(1), the distances between Cu(1) and the closest
˚
partial F atoms lying between 2.500(14)–2.565(11) A (Table 3,
Fig. 2).
The EPR spectra of 9 as a powder and in frozen CH2Cl2 show
near-identical g and A1 values (see Experimental section). Apart
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9
3 2 4 3

View Article Online
◦
˚
Table 3 Selected bond lengths (A) and angles ( ) for 9·xCH3CN. Data
−
are listed for all three orientations of the disordered BF4 donor F(50)
Cu(1)–N(9)
1.919(3)
1.879(3)
1.919(4)
1.889(3)
2.565(11), 2.549(9), 2.500(14)
3.5020(5)
2.541(3)
2.5392(10)
2.411(3)
2.5508(10)
Cu(1)–O(17)
Cu(1)–N(29)
Cu(1)–O(37)
Cu(1)–F(50[A, B, C])
Cu(1) · · · Ag(2)
Ag(2)–O(17)
Ag(2)–S(22)
Ag(2)–O(37)
Ag(2)–S(42)
N(9)–Cu(1)–O(17)
N(9)–Cu(1)–O(37)
93.92(13)
168.79(15)
N(9)–Cu(1)–N(29)
86.05(14)
98.7(3), 82.2(2), 112.3(4)
176.55(15)
85.37(12)
101.3(3), 88.0(2), 86.3(4)
93.99(14)
82.1(3), 95.4(3), 96.9(4)
92.4(3), 108.9(2), 78.8(4)
70.28(6)
N(9)–Cu(1)–F(50[A, B, C])
O(17)–Cu(1)–N(29)
O(17)–Cu(1)–O(37)
O(17)–Cu(1)–F(50[A, B, C])
N(29)–Cu(1)–O(37)
N(29)–Cu(1)–F(50[A, B, C])
O(37)–Cu(1)–F(50[A, B, C])
O(17)–Ag(2)–S(22)
O(17)–Ag(2)–O(37)
O(17)–Ag(2)–S(42)
S(22)–Ag(2)–O(37)
62.04(9)
132.31(6)
132.30(7)
S(22)–Ag(2)–S(42)
O(37)–Ag(2)–S(42)
151.53(3)
73.14(7)
Fig. 3 Top: Q-band EPR spectrum of 9 in CH2Cl2 at 110 K. Bottom:
X-band EPR spectrum of 8 in CH2Cl2 at 110 K.
solution the spectra of 8 are characteristic of a spin-triplet
system, showing fine (zero-field splitting) and hyperfine (to
63,65Cu) structure and a well-resolved formally spin-forbidden
“half-field” transition (Fig. 3, bottom). The resolution at X-
band of seven-line hyperfine patterns in the parallel and half-
field regions, with hyperfine coupling about half that expected
for a monomeric square-planar Cu(II) species, suggests that
this spectrum is dominated by a dimeric copper(II) compound.
Attempts to simulate these spectra at different frequencies were
unsuccessful, which may reflect the presence of more than
one paramagnetic species. However, the spectra are clearly
dominated by a spin-triplet dimer. The zero-field splitting within
the triplet is relatively small (|D| ≈ 0.05 cm−1), which allows us to
˚
estimate an intra-dimer Cu · · · Cu distance of ≤4.0 A, depending
on the relative orientations of the g-matrix and D-tensor.29 That
distance is consistent with dimerisation through basal, apical
Cu–O · · · Cu bridging, which would lead to a Cu · · · Cu distance
of between 3.1–3.7 A.23,26,27 The fact that 8 appears to oligomerise
˚
in solution while 9 does not may simply reflect the greater steric
bulk of the [L3]2− ligand in 9 compared to [L2]2− in 8. Since 8
dimerises in solution while 2 does not, the silver ion in 8 may
also play a role in the bridging interactions. It is unclear what
that might be, however.
The powder EPR spectra of 7 are very broad with no resolved
hyperfine coupling, although g-anisotropy can be resolved at
W-band. The frozen solution spectra of 7 are also poorly
resolved but two features are clear: A||(63,65Cu) is about half
that expected for a mononuclear copper(II) centre; and, there
are at least six hyperfine lines in this region. As for 8, this
suggests oligomerisation in solution. However, in contrast to
8 the spectra contain no obvious fine structure nor any half-field
signal. Hence, if 7 has dimerised in solution then the resultant
spin-triplet must show |D| significantly smaller than the spectral
linewidth. Given these ambiguities, in the absence of a crystal
structure the detailed molecular structure of 7 is uncertain. One
related adduct of silver(I), [{Cu(l-salen)}2Ag]ClO4, has also
been isolated but not structurally characterised.30 However the
ability of 1,15 or other [Cu(salen)]-type derivatives lacking ortho
substituents at their phenoxide rings,30,31 to form adducts with
other metal ions by bridging through the phenoxide O atoms is
well known.
Fig. 2 View of the [CuAg(l-L3)]BF4 moiety in the crystal structure
of 9·1.45CH3CN, showing the atom numbering scheme employed. For
clarity only the major orientations of the disordered cyclohexenyl moiety
C(3A)–C(8A), tert-butyl group C(38A)–C(41A) and BF4 anion are
shown. Other details as for Fig. 1.
−
from a slightly reduced g1 and a small rhombicity (apparent at
Q- and W-bands), the spectra are very comparable to those
of 3 (Fig. 3, top). This strongly implies the copper/silver
adduct 9 retains its integrity in solution. In the solid state
the EPR spectrum of 8 is similar to that of 9 (although the
spectrum appears strictly axial, even at W-band). That implies
the molecular structure of 8 is probably analogous to that of 9
in the solid state, as expected. In contrast, in frozen CH2Cl2
3 2 4 4
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9

View Article Online
In the light of these results, the reactions of 2, 3, 5 and 6 with
other oxidising agents were pursued. The imine complexes 2
and 3 did not react with [NH4]2[Ce(NO3)6] in CH2Cl2, probably
because of the poor solubility of the cerium reagent in that
solvent. However, treatment of the same compounds with 1 or
2 equiv. of SbCl5 in CH2Cl2 under N2 at 196 K afforded dark
brown solutions, that were indefinitely stable at that temperature
but which rapidly decomposed to intractable brown precipitates
upon warming. These solutions afforded EPR spectra that were
different from 2 and 3, but still typical of S = 1/2 copper(II)
compounds9,11,33,34 and galactose oxidase.3,5 Since the oxidation
potentials of 5 and 6 are 0.4–0.5 V less positive than those of 2
and 3, the poorer kinetic stability of [5•]+ and [6•]+ compared to
[2•]+ and [3•]+ is also a little surprising, although it has literature
precedent.18,20
It is also notable that, despite their almost identical oxidation
potentials, the electrochemical lifetimes of [2•]+ and [3•]+, and
•
of [2••]2+ and [3••]2+, are rather shorter than those of [Cu(L7 )]+
••
and [Cu(L7 )]2+.19 This relative instability must be kinetic in
origin, and presumably reflects the lower steric protection of
the phenoxyl centre afforded by an ortho-sulfanyl substituent
compared to a tert-butyl group. The poor kinetic stability of
the monooxidation products in this work must reflect some
aspect of the chemistry of the salan and salcyan ligand systems,
since other copper(II) complexes are known of polydentate
phenoxides bearing a 4-tert-butyl-2-sulfanyl substitution pat-
tern, but lacking any other steric protection, which do afford
long-lived phenoxyl radical species at room temperature.34,35 In
conclusion we have characterised an unusual system in copper
phenoxyl chemistry, where a sulfanyl substituent confers almost
no additional thermodynamic stabilisation, and limited kinetic
stabilisation, on a series of phenoxyl radical species.
centres with tetragonal stereochemistries (gꢁ = 2.18, g|⊥
=
2.04, A|| = 180
5 G for both compounds at X-band at
120 K). This rules out their formulation as copper(II) phenoxyl
radical species, which are usually EPR-silent.6 Rather, the brown
solutions probably contain adducts between 2 and 3 and an
unidentified antimony centre, with structures similar to 8 and 9.
Owing to their thermal instability, no further characterisation
of these species was attempted.
Treatment of 5 and 6 with 1 equiv. of the milder oxidants
acetylferrocenium tetrafluoroborate (E1/2 = + 0.27 V28) or DDQ
(E1/2 = + 0.13 V28) in CH2Cl2 under N2 at 196 K yields black
solutions, which deposit black powders upon standing. These
solids appear to contain intact 5 or 6 by mass spectrometry
but could not otherwise be unambiguously identified. Clearly,
[5•]+ and [6•]+ decompose rapidly under these conditions, in
agreement with the electrochemical results.
Experimental
All reactions were carried out in air using non-pre-dried
AR-grade solvents unless otherwise stated. Acetylferrocenium
tetrafluoroborate was made by the literature procedure,28 while
4-tert-butylphenol, rac-trans-1,2-diaminocyclohexane, DDQ,
SbCl5 (1.0 M solution in CH2Cl2) and all metal salts were used
as supplied. EPR data for the compounds are listed in Table 4.
Conclusions
It is interesting that the first oxidation potentials of 1–3, and
[Cu(L7)],18,19 all lie within 0.14 V of each other while those
of 4–6 and [Cu(L8)]17 differ by ≤0.05 V. We are only aware
of one literature system that shows similar behaviour,32 as
most previous studies have shown that introduction of an
ortho-alkylsulfanyl substituent onto a phenoxide ring lowers
its oxidation potential by 0.25–0.55 V, both in synthetic
Synthesis of 4-tert-butyl-2-methylsulfanylphenol
A mixture of 4-tert-butylphenol (10 g, 0.072 mol), dimethyld-
isulfide (9.0 g, 0.22 mol) and FeCl3 (17.9 g, 0.11 mol) in toluene
(100 cm3) was heated under reflux for 6 h. After cooling to
Table 4 EPR data for the compounds in this study, obtained by simulation of spectra obtained at X-band, Q-band and (for 7–9) W-band. All
hyperfine coupling constants are in G. Fluid and frozen solution spectra were all obtained from 10 : 1 CH2Cl2 : toluene, at 293 K and 110 K,
respectively, while powder spectra were run at 110 K. Isotropic g and A values are quoted for the fluid solution spectra
14
14
1
g1
g2
g3
A1{63,65Cu}
A2,3
{
63,65Cu}
A1{ N} A2,3{ N}
A{ H}
[Cu(L1)] (1)
Powdera
Fluid solution
Frozen solution
Powdera
Fluid solution
Frozen solution
Powder
Fluid solution
Frozen solution
Powder
Fluid solution
Frozen solution
Powdera
Fluid solution
Frozen solution
Powdera
Fluid solution
Frozen solution
Powder
—
—
—
—
92
202
—
92
2.092
2.192
—
14
7
7
7
2.040
—
2.040
—
30
30
30
23
28
25
16
[Cu(L2)] (2)
2.097
2.195
2.203
2.097
2.198
2.226
2.118
2.233
—
2.104
2.227
—
2.107
2.220
2.167
2.093
2.17
14
2.043
2.044
2.043
2.044
200
17
[Cu(L3)] (3)
92
202
174
78
182
—
88
187
—
14
2.043
2.063
2.043
2.063
16
[Cu(L4)] (4)
2.052
—
2.052
—
[Cu(L5)] (5)
2.052
—
2.052
—
[Cu(L6)] (6)
86
188
—
2.046
2.045
2.046
2.045
[CuAg(l-L1)]BF4 (7)
[CuAg(l-L2)]BF4 (8)
[CuAg(l-L3)]BF4 (9)
Fluid solution
Frozen solution
Powder
86
2.04
2.043
2.04
2.043
∼90
200
86
∼90
200
92
2.175
2.092
2.22
2.175
2.096
2.183
Fluid solutionb
Frozen solutionb
Powder
2.04
2.050
2.04
2.037
Fluid solutionb
Frozen solutionb
14
7
7
2.050
2.032
206
28
14
a Broad spectrum, that could not be accurately simulated. b Oligomeric species in solution. See text for more details.
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9
3 2 4 5

View Article Online
room temperature, 10% aqueous HCl was added until all the
solid dissolved. The mixture was extracted with diethyl ether
(6 × 50 cm3), the combined organic layers dried over MgSO4,
and the volatiles removed to give brown oil. Silica flash column
chromatography in pentane : diethyl ether (9 : 1) provided the
product as a yellow oil. Yield 8.9 g, 63%. Found C, 66.7; H,
8.4%. Calcd for C11H16OS C, 67.3; H, 8.2%. EI mass spectrum:
m/z 196 [M]+, 181 [M − CH3]+, 149 [M − SCH3]+. NMR spectra
(CDCl3): 1H d 1.31 (s, 9H, C{CH3}3), 2.34 (s, 3H, SCH3), 6.48
(br s, 1H, OH), 6.90 (d, 1H, J = 8.8 Hz, Ph H6), 7.27 (dd, 1H,
J = 2.3 and 8.8 Hz, Ph H5), 7.48 (d, 1H, J = 2.3 Hz, Ph H3).
13C d19.9 (SCH3), 31.4 (C{CH3}3), 34.1 (C{CH3}3), 114.2 (C6),
119.3 (C2), 127.9, 131.4 (C3 + C5), 143.8 (C4), 159.4 (C1).
70.8; H, 5.9%. Calcd for C17H18O2S C, 71.3; H, 6.3%. ES mass
spectrum: m/z 287 [M + H]+, 271 [M − CH3]+. NMR spectra
(CDCl3): 1H d 1.24 (s, 9H, C{CH3}3), 7.25–7.33 (m, 5H, C6H5),
7.48 (d, 1H, J = 2.3 Hz, H4), 7.54 (d, 1H, J = 2.3 Hz, H6), 9.93 (s,
1H, CHO). 13C d 31.1 (C{CH3}3), 34.4 (C{CH3}3), 119.1 (C3),
125.0 (C6) 127.4 (C1), 129.0 (Ph C4ꢀ), 129.2 (Ph C3ꢀ/5ꢀ), 130.6 (Ph
C
2ꢀ/6ꢀ), 133.3 (C4), 137.6 (Ph C1ꢀ), 143.4 (C5), 158.4 (C2), 196.4
(CHO).
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxybenzylidene)-
cyclohexane-1,2-diamine (H2L1)
A solution of 5-tert-butyl-2-hydroxybenzaldehyde (1.00 g, 5.6 ×
10−3 mol) and trans-cyclohexane-1,2-diamine (0.32 g, 2.8 ×
10−3 mol) in methanol (20 cm3) was refluxed for 1 h, yielding
a yellow precipitate. The product was recovered by filtration,
washed with cold methanol and dried in vacuo. Yield 1.2 g,
95%. Mp 169–171 ◦C. Found C, 77.5; H, 8.9; N, 6.5%. Calcd for
C28H38N2O2 C, 77.7; H, 8.4; N, 6.5%. ES mass spectrum: m/z 435
Synthesis of 4-tert-butyl-2-phenylsulfanylphenol
Method as for 4-tert-butyl-2-methylsulfanylphenol, using
diphenyldisulfide (48.0 g, 0.22 mol). The yellow oil product
was purified by silica flash column chromatography in pentane :
diethyl ether (9 : 1). Yield 5.6 g, 31%. Found C, 73.9; H, 7.1%.
Calcd for C16H18OS C, 74.3; H: 7.0%. EI mass spectrum: m/z
258 [M]+, 243 [M − CH3]+, 134 [M − SC6H5]+, 77 [C6H5]+. NMR
spectra (CDCl3): 1H d 1.27 (s, 9H, C{CH3}3), 6.30 (s, 1H, OH),
6.93 (d, 1H, J = 8.8 Hz, Ph H6), 7.04–7.24 (m, 5H, C6H5), 7.41
(dd, 1H, J = 2.3 and 8.8 Hz, Ph H5), 7.53 (d, 1H, J = 2.3 Hz, Ph
H3). 13C d 31.4 (C{CH3}3), 34.5 (C{CH3}3), 114.9 (C6), 119.3
(Cꢀ2), 125.9 (1C, C5), 127.1 (Ph C4ꢀ), 129.0 (1C, C3), 129.1 (Ph
+
t
+
=
[M + H] , 259 [M − {N CH(C6H3-OH-2- Bu-5}] , 218 [M +
2H]2+. NMR spectra (CDCl3): H d 1.19 (s, 18H, C{CH3}3),
1
1.47–1.89 (m, 8H, Cy CH2), 3.30 (m, 2H, Cy CH), 6.94 (d, 2H,
J = 8.8 Hz, H3), 7.12 (d, 2H, J = 2.3 Hz, H6), 7.28 (dd, 2H, J =
2.3 and 8.8 Hz, H ), 8.25 (s, 2H, N CH). 13C d 24.2 (Cy C4/5),
4
=
31.3 (C{CH3}3), 33.2 (Cy C3/6), 33.8(C{CH3}3), 72.8 (Cy C1/2),
116.2 (C3), 117.9 (C1), 127.8, 129.4 (C4 + C6) 141.2 (C5), 158.6
2
=
(C ), 165.0(N CH).
C
3 /5ꢀ), 129.5 (Ph C2ꢀ/6ꢀ), 133.2 (Ph C1ꢀ), 137.0 (C4), 156.3 (C1).
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxy-3-methyl-
Synthesis of 5-tert-butyl-2-hydroxybenzaldehyde
sulfanylbenzylidene)cyclohexane-1,2-diamine (H2L2)
A modification of the Hofslokken and Skattebol’s procedure was
followed.19 To a solution of 4-tert-butylphenol (3.0 g, 0.022 mol)
in acetonitrile (100 cm3) was added dry triethylamine (7.6 g,
0.075 mol), paraformaldehyde (4.05 g, 0.14 mol) and anhydrous
MgCl2 (2.85 g, 0.030 mol). The mixture was refluxed for 4 h, then
cooled to room temperature. 5% Aqueous HCl was slowly added
to the mixture until all the solid dissolved, then the mixture
was extracted with diethyl ether (5 × 50 cm3). The combined
organic layers were dried over MgSO4 and filtered. Evaporation
of the volatiles gave a brown oil that was distilled under reduced
pressure to afford a colourless oil. Yield 3.5 g, 90%. Found C,
74.0; H, 8.1%. Calcd. for C11H14O2 C, 74.1; H: 7.9%. EI mass
Method as for H2L4, using 5-tert-butyl-2-hydroxy-3-
methylsulfanylbenzaldehyde (1.26 g, 5.6 × 10−3 mol). The
product was a yellow solid. Yield 1.3 g, 91%. Mp 145–
147 ◦C. Found C, 67.3; H, 8.0; N, 5.2%. Calcd for C30H42O2S2N2
C, 67.4; H, 8.0; N, 5.3%. ES mass spectrum: m/z 527 [M +
H]+, 264 [M + 2H]2+. NMR spectra (CDCl3): 1H d 1.19 (s, 18H,
C{CH3}3), 1.26–1.85 (m, 8H, Cy CH2), 2.39 (s, 6H, SCH3),
3.26 (m, 2H, Cy CH), 6.94 (d, 2H, J = 2.3 Hz, H4), 7.22 (d, 2H,
6
J = 2.3 Hz, H ), 8.19 (s, 2H, N CH). 13C d 15.5 (SCH3), 24.1
=
(Cy C4/5), 31.3 (C{CH3}3), 33.1 (Cy C3/6), 34.1 (C{CH3}3), 72.4
(Cy C1/2), 116.9 (C3), 124.4 (C6), 125.8 (C1) 128.6 (C4), 141.3
5
2
=
(C ), 157.3 (C ), 165.9 (N CH).
t
spectrum: m/z 179 [M + H]+, 149 [M − CHO]+, 122 [M − Bu +
1
H]+. NMR spectra (CDCl3): H d 1.31 (s, 9H, C{CH3}3), 6.90
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxy-3-phenyl-
(d, 1H, J = 8.8 Hz, H3), 7.54 (d, 1H, J = 2.3 Hz, H6), 7.58 (dd,
1H, J = 2.3 and 8.8 Hz, H4), 9.83 (s, 1H, CHO), 10.83 (br s, 1H,
OH). 13C d 31.3 (C{CH3}3), 34.0 (C{CH3}3), 117.2 (C3), 121.3
(C1) 129.7 (C6), 134.6 (C4), 143.6 (C5), 159.5 (C2), 196.7 (CHO).
sulfanylbenzylidene)cyclohexane-1,2-diamine (H2L3)
Method as for H2L4, using 5-tert-butyl-2-hydroxy-3-phenyl-
sulfanylbenzaldehyde (1.60 g, 5.6 × 10−3 mol). The product was
a yellow solid. Yield 1.6 g, 87%. Mp 77–79 ◦C. Found C, 73.5; H,
7.7; H, 3.9%. Calcd for C40H46N2O2S2 C, 73.7; H, 7.7; N, 4.1%.
ES mass spectrum: m/z 651 [M + H]+, 326 [M + 2H]2+. NMR
spectra (CDCl3): 1H d 1.18 (s, 18H, C{CH3}3), 1.23–1.92 (m, 8H,
Cy CH2), 3.29 (m, 2H, Cy CH), 7.21 (d, 2H, J = 2.3 Hz, H4),
7.20–7.28 (m, 12H, d, 2H, J = 2.3 Hz, H6 + C6H5), 8.24 (s, 2H,
Synthesis of 5-tert-butyl-2-hydroxy-3-
methylsulfanylbenzaldehyde
Method as for HL1, using 4-tert-butyl-2-methylsulfanylphenol
(4.3 g, 0.022 mol). The crude product was purified by silica flash
column chromatography in pentane : diethyl ether (98 : 2), to
yield a yellow solid. Yield 3.7 g, 76%. Mp 42–44 ◦C. Found C,
63.3; H, 5.8%. Calcd for C12H16O2S C, 63.2; H, 6.2%. ES mass
spectrum: m/z 225 [M + H]+. NMR spectra (CDCl3): 1H d 1.31
(s, 9H, C{CH3}3), 2.49 (s, 3H, SCH3), 7.26 (br s, 1H, OH), 7.38
(d, 1H, J = 2.3 Hz, H4), 7.55 (d, 1H, J = 2.3 Hz, H6), 9.91 (s,
1H, CHO). 13C d 15.5 (SCH3), 31.4 (C{CH3}3), 34.3 (C{CH3}3),
119.5 (C3), 126.1 (C6) 127.4 (C1), 133.3 (C4), 143.4 (C5), 158.4
(C2), 196.7 (CHO).
N CH). 13C d24.1 (Cy C4/5), 31.3 (C{CH3}3), 33.1 (Cy C3/6),
=
34.1 (C{CH3}3), 72.4 (Cy C1/2), 117.8 (C3), 126ꢀ.3 (C1), 128.1
(C6) 128.9 (Ph C4ꢀ), 129.6 (Ph C3ꢀ/5ꢀ), 131.6 (Ph C2 /6ꢀ), 133.6 (Ph
1ꢀ
4
5
2
=
C ), 141.6 (C ), 149.0 (C ), 157.3 (C ), 164.8 (N CH).
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxybenzyl)cyclo-
hexane-1,2-diamine (H2L4)
NaBH4 (0.54 g, 0.014 mol) was slowly added to a solution of
H2L1 (1.54 g, 3.5 × 10−3 mol) in a mixture of 1 : 1 THF/MeOH
(30 cm3). The mixture was stirred for 2 h then evaporated to
dryness. The white solid residue was redissolved in water and
extracted with CH2Cl2 (32 × 5 cm3). The combined organic
layers were dried over MgSO4, filtered and the volatiles were
removed to give a white solid that was recrystallised from
pentane/diethyl ether. Yield 1.34 g, 86%. Mp 83–85 ◦C. Found
C, 76.4; H, 9.6; N, 6.5%. Calcd for C28H42N2O2 C, 76.7; H, 9.6; N,
Synthesis of 5-tert-butyl-2-hydroxy-3-
phenylsulfanylbenzaldehyde
Method as for HL1, using 4-tert-butyl-2-phenylsulfanylphenol
(5.3 g, 0.022 mol). The crude product was purified by silica flash
column chromatography in pentane : diethyl ether (95 : 5), to
yield a yellow solid. Yield 5.2 g, 83%. Mp 85–87 ◦C. Found C,
3 2 4 6
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9

View Article Online
6.4%. ES mass spectrum: m/z 439 [M]+. NMR spectra (CDCl3):
1H d 1.29 (s, 18H, C{CH3}3), 1.73–2.52 (m, 10H, C6H10), 4.01
(m, 4H, NHCH2), 6.79 (d, 2H, J = 8.8 Hz, H3), 7.19 (dd, 2H,
J = 2.3 and 8.8 Hz, H4), 7.22 (d, 2H, J = 2.3 Hz, H6). 13C d 24.1
(Cy C4/5), 30.4 (Cy C3/6), 31.6 (C{CH3}3), 33.9 (C{CH3}3), 50.1
(NHCH2), 59.8 (Cy C1/2), 115.8 (C3), 122.2 (C1), 125.1, 125.5
(C4 + C6), 141.8 (C5), 155.5 (C2).
spectrum: m/z 999 [63Cu2(L4)2 + H]+, 938 [63Cu(L4)(H2L4) +
H]+, 500 [63Cu(L4) + H]+, 439 [H3L4]+. UV/vis (dmf): kmax, nm
(emax, M−1 cm−1) 596 (300), 406 (835), 281 (65 000).
For 5. Yield 0.54 g, 79%. Found C, 60.4; H, 7.4; N, 4.6%.
Calcd for C30H44CuN2O2S2 C, 61.0; H, 7.4; N: 4.7%. ES
mass spectrum: m/z 1183 [63Cu2(L5)2 + H]+, 654 [63Cu2(L5)]+,
592 [63Cu(L5) + H]+, 531 [H3L5]+. UV/vis (dmf): kmax, nm
(emax, M−1 cm−1) 600 (380), 420 (930), 321 (20 000), 259 (36 000).
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxy-3-methyl-
sulfanylbenzyl)cyclohexane-1,2-diamine (H2L5)
For 6. Yield 0.58 g, 65%. Found C, 66.5; H, 7.2; N,
3.9%. Calcd for C40H48CuN2O2S2 C, 67.1; H, 6.7; N, 3.9%. ES
mass spectrum: m/z 1431 [63Cu2(L6)2 + H]+, 778 [63Cu2(L6)]+,
716 [63Cu(L6) + H]+, 655 [H3L6]+. UV/vis (dmf): kmax, nm
(emax, M−1 cm−1) 600 (300), 420 (550), 325 (24 000).
Method as for H2L4, using H2L2 (1.84 g, 3.5 × 10−3 mol). The
product formed pale yellow microcrystals from pentane/diethyl
ether. Yield 1.55 g, 83%. Mp 125–127 ◦C. Found C, 67.9; H,
9.0; N, 5.2%. Calcd for C30H46N2O2S2 C, 67.8; H, 8.7; N, 5.3%.
ES mass spectrum: m/z 531 [M + H]+. NMR spectra (CDCl3):
1H d 1.19 (s, 18H, C{CH3}3), 1.70–2.40 (m, 8H, Cy CH2), 2.42
(s, 6H, SCH3), 2.46 (m, 2H, Cy CH), 3.92 (m, 4H, NHCH2),
Syntheses of [CuAg(l-L)]BF4 (L2− = [L1]2−, 7; L2− = [L2]2−,
8; L2− = [L3]2−, 9)
6.91 (d, 2H, J = 2.3 Hz, H4), 7.23 (d, 2H, J = 2.3 Hz, H6). 13
C
The same method, as given here for 7, was followed for all three
compounds. A solution of 1 (0.50 g, 1.01 × 10−3 mol) and AgBF4
(0.20 g, 1.01 × 10−3 mol) in pre-dried CH2Cl2 (25 cm3) was
stirred at room temperature for 15 min. The deep red solution
was filtered and concentrated to ca. 5 cm3. Diffusion of Et2O
vapour into this solution yielded dark red–black microcrystals.
Similar reactions employing 2 (0.59 g, 1.01 × 10−3 mol) or 3
(0.72 g, 1.01 × 10−3 mol) gave red microcrystalline 8 and 9,
respectively. The microanalyses of all three compounds varied
somewhat between samples, which may reflect either solvent loss
or their mild photosensitivity.
d 16.6 (SCH3), 24.2 (Cy C4/5), 30.4 (Cy C3/6), 31.5 (C{CH3}3),
34.2 (C{CH3}3), 49.7 (NHCH2), 60.3 (Cy C1/2), 122.4 (C3), 122.9
(C6), 124.1 (C1), 125.9 (C4), 142.3 (C5), 153.3 (C2).
Synthesis of N,Nꢀ-bis(5-tert-butyl-2-hydroxy-3-phenyl-
sulfanylbenzyl)cyclohexane-1,2-diamine (H2L6)
Method as for H2L4, using H2L3 (2.28 g, 3.5 × 10−3 mol). The
product formed pale yellow microcrystals from pentane/diethyl
ether. Yield 2.04 g 89%. Mp 141–143 ◦C. Found C, 72.8; H, 7.6;
N, 4.2%. Calcd for C40H50N2O2S2 C, 73.2; H, 7.7; N, 4.3%. ES
mass spectrum: m/z 655 [M + H]+. NMR spectra (CDCl3): 1H
d 1.22 (s, 18H, C{CH3}3), 1.23–1.92 (m, 8H, Cy CH2), 2.36 (m,
2H, Cy CH), 3.94 (m, 4H, NHCH2), 7.05 (d, 2H, J = 2.3 Hz,
For 7. Yield 0.22 g, 22%. Found C, 49.6; H, 5.4; N, 4.3%.
Calcd for C28H36AgBCuF4N2O2 C, 48.7; H, 5.3; N, 4.1%. ES
mass spectrum: m/z 1097 [63Cu2107Ag(L1)2]+, 991 [63Cu2(L1)2 +
H]+, 602 [63Cu107Ag(L1)]+, 496 [63Cu(L1) + H]+, 435 [H3L1]+.
UV/vis (CH2Cl2): kmax, nm (emax, M−1 cm−1) 555 (1,150), 382
(61 500).
H4), 7.06–7.26 (m, 10H, C6H5), 7.27 (d, 2H, J = 2.3 Hz, H6). 13
C
d24.3 (Cy C4/5), 30.6 (Cy C3/6), 31.4 (C{CH3}3), 34.1(C{CH3}3),
49.7 (NHCH2), 60.6 (Cy C1/2), 118.4 (C3), 123.6 (C1), 125.8 (C6),
128.4 (Ph C4ꢀ), 128.5 (Ph C3ꢀ/5ꢀ), 128.9 (Ph C2ꢀ/6ꢀ), 130.8 (C4),
136.5 (Ph C1ꢀ), 142.6 (C5), 154.7 (C2).
For 8. Yield 0.85 g, 65%. Found C, 44.8; H, 5.5; N, 3.4%.
1
Calcd. for C30H40AgBCuF4N2O2S2· CH2Cl2 C, 44.4; H, 5.1;
N, 3.4%. ES mass spectrum: m/z 2694 [63Cu107Ag(L2)]+, 587
[63Cu(L2) + H]+. UV/vis (CH2Cl2): kmax, nm (emax, M−1 cm−1)
495 (970), 378 (46 000).
Syntheses of [CuL] (L2− = [L1]2−, 1; L2− = [L2]2−, 2; L2− = [L3]2−,
3; L2− = [L4]2−, 5; L2− = [L6]2−, 6; L2− = [L7]2−, 8)
The same method, as given here for 1, was followed for all six
compounds. A mixture of Cu(O2CMe)2·H2O (0.23 g, 1.15 ×
10−3 mol) and H2L1 (0.50 g, 1.15 × 10−3 mol) in MeCN (10 cm3)
was refluxed for 2 h. The resultant green precipitate was isolated
by filtration, washed with cold MeCN and Et2O, and dried
in vacuo. Similar reactions employing H2L2 (0.60 g, 1.15 ×
10−3 mol), H2L3 (0.75 g, 1.15 × 10−3 mol), H2L4 (0.50 g, 1.15 ×
10−3 mol), H2L5 (0.60 g, 1.15 × 10−3 mol) or H2L6 (0.75 g, 1.15 ×
10−3 mol) afforded 2–6 respectively.
For 9. Yield 0.92 g, 46%. Found C, 50.4; H, 5.4; N: 2.6%.
Calcd. for C40H44AgBCuF4N2O2S2·1/2CH2Cl2 C, 51.2; H, 4.8;
N, 2.9%. ES mass spectrum: m/z 818 [63Cu107Ag(L3)]+. UV/vis
(CH2Cl2): kmax, nm (emax, M−1 cm−1) 483 (852), 389 (sh), 379
(46 000).
Single crystal X-ray structure determinations
Single crystals of 1·H2O and 1·CH2Cl2 were, respectively, grown
by slow evaporation of a solution of 1 in EtOH, and by diffusion
of pentane vapour into a CH2Cl2 solution of the complex.
Diffusion of Et2O vapour into a MeCN solution of 9 yielded
dark red prisms of formula 9·1.45CH3CN. Experimental details
from the structure determinations are given in Table 5. Data
were collected using a Nonius KappaCCD area detector diffrac-
tometer fitted with an Oxford cryosystems low-temperature
device. All structures in this work were solved by direct methods
(SHELXS 9736) and refined by full matrix least-squares on F2
(SHELXL 9737).
Single crystal X-ray structure determination of [Cu(L1)]·H2O
(1·H2O). No disorder was detected during refinement of this
structure. All non-H atoms were refined anisotropically, while
all C-bound H atoms were placed in calculated positions and
refined using a riding model. The water H atoms H(34A) and
H(34B) were located in the difference map, and refined with a
common thermal parameter subject to the refined restraints O–
For 1. Yield 0.35 g, 62%. Found C, 67.2; H, 7.3; N,
5.6%. Calcd. for C28H36CuN2O2 C, 67.8; H, 7.3; N, 5.6%. ES
mass spectrum: m/z 991 [63Cu2(L1)2 + H]+, 496 [63Cu(L1)+H]+.
UV/vis (CH2Cl2): kmax, nm (emax, M−1 cm−1) 566 (305), 372 (4000),
279 (14 000), 250 (sh), 230 (27 700).
For 2. Yield 0.37 g, 54%. Found C, 61.3; H, 6.9; N, 4.6%.
Calcd for C30H40CuN2O2S2 C, 61.4; H, 6.8; N, 4.6%. ES mass
spectrum: m/z 1175 [63Cu2(L2)2 + H]+, 588 [63Cu(L2) + H]+.
UV/vis (CH2Cl2): kmax, nm (emax, M−1 cm−1) 565 (323), 393 (3900),
307 (9400), 266 (sh), 232 (sh).
For 3. Yield 0.46 g, 56%. Found C, 67.4; H, 6.2; N, 3.9%.
Calcd for C40H44CuN2O2S2 C, 67.2; H, 6.2; N, 3.9%. ES mass
spectrum: m/z 1423 [63Cu2(L3)2 + H]+, 1155 [63Cu2(L3)(L3-
CHC6H2O-2-SPh-3-tBu-5) + 2H]+, 712 [63Cu(L3) + H]+. UV/vis
(CH2Cl2): kmax, nm (emax, M−1 cm−1) 570 (335), 390 (5,400), 306
(sh), 281 (sh), 250 (sh), 232 (sh).
˚
˚
H = 0.72(1) A and H · · · H = 1.18(2) A. The highest residual
−3
−3
˚
˚
˚
For 4. Yield 0.48 g, 84%. Found C, 67.7; H, 8.3; N, 5.6%.
Calcd for C28H40CuN2O2 C, 67.2; H, 8.1; N, 5.6%. ES mass
Fourier peak (0.9 e A ) and hole (−1.0 e A ) both lie ≤1.0 A
from Cu(1).
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9
3 2 4 7

View Article Online
Table 5 Experimental details for the single crystal structure determinations in this study
[Cu(L1)]·H2O (1·H2O)
[Cu(L1)]·CH2Cl2 (1·CH2Cl2)
[CuAg(l-L3)]BF4·xCH3CN (9·xCH3CN; x ≈ 1.45)
Formula
Mr
Crystal System
C28H38CuN2O3
514.14
C29H38Cl2CuN2O2
581.05
C42.90H48.35AgBCuF4N3.45O2S2
966.64
Triclinic
Triclinic
Triclinic
¯
¯
¯
Space Group
P1
P1
P1
˚
a/A
10.4650(1)
11.0662(2)
11.2949(2)
91.5998(8)
98.6155(8)
97.9804(6)
1279.17(3)
2
8.2379(2)
12.0392(2)
14.8235(3)
99.3685(9)
95.7750(7)
99.3788(10)
1418.90(5)
2
8.4806(1)
14.8652(2)
17.4574(3)
89.8338(6)
84.3792(6)
79.3241(8)
2152.03(5)
2
˚
b/A
˚
c/A
a/◦
b/◦
c /◦
3
˚
V/A
Z
l/mm−1
0.885
0.986
1.103
T/K
150(2)
26629
150(2)
30377
150(2)
39767
Measured reflections
Independent reflections
5804
0.159
0.055
0.147
6503
0.107
0.042
0.115
9801
0.069
0.052
0.155
Rint
R(F)a
wR(F2)b
2
4
1/2
.
a R = R[|Fo| − | Fc|]/R|Fo|. b wR = [Rw(Fo − Fc2)/RwFo
]
Single crystal X-ray structure determination of [Cu(L1)]·
CH2Cl2(1·CH2Cl2). Although the thermal parameters on the
solvent atoms C(34) and Cl(35) are a little high, attempts to
refine a disorder model involving these atoms afforded two
orientations (with refined occupancies of 0.50 each) whose
atoms had thermal ellipsoids larger than those of the crys-
tallographically ordered solvent model. The residuals R1 and
wR2 were also virtually identical in the two models. Therefore,
an ordered CH2Cl2 molecule was included in the final least
squares cycles, since the inclusion of a disorder model led to no
improvement in the overall refinement. Presumably, the larger
thermal parameters on Cl(35) in particular reflect librational
motion of this group in the crystal. No other disorder was
detected during refinement. All non-H atoms were refined
anisotropically, while all H atoms were placed in calculated
positions and refined using a riding model.
CCDC reference numbers 270588–270590.
See http://dx.doi.org/10.1039/b505972k for crystallographic
data in CIF or other electronic format.
Other measurements
Elemental microanalyses were performed by the University
of Leeds School of Chemistry microanalytical service. NMR
spectra were obtained on a Bruker ARX250 spectrometer,
operating at 250.1 MHz (1H) and 62.9 MHz (13C). Infra-red
spectra were obtained as Nujol mulls sandwiched between KBr
windows, between 400–4000 cm−1 using a Nicolet Avatar 360
spectrophotometer. Electron impact mass spectra were per-
formed on a Kratos MS50 spectrometer, while electrospray mass
spectra were obtained on a Micromass LCT TOF spectrometer,
employing a MeOH matrix. UV/Vis measurements were per-
formed using a Perkin Elmer Lambda 900 spectrophotometer,
in 1 cm quartz solution cells. X-band (ca. 9.3 GHz) and Q-
band (34 GHz) EPR spectra of 1–9 were obtained using a
Bruker ESP300E spectrometer; X-band spectra employed an
ER4102ST resonator and ER4111VT cryostat, while for Q-band
spectra an ER5106QT resonator and ER4118VT cryostat were
used. W-band (95 GHz) EPR spectra of 7–9 were measured on a
Bruker E600 ELEXSYS spectrometer. Spectral simulations were
performed using in-house software which has been described
elsewhere.38 Other X-band EPR spectra were obtained using a
Bruker EMX spectrometer fitted with an ER4119HS resonator
and ER4131VT cryostat.
Single crystal X-ray structure determination of [CuAg(l-
L3)]BF4·xCH3CN (9·xCH3CN; x ≈ 1.45). The following moi-
eties in the structure are disordered. First, the cyclohexyl ring
C(3)–C(8) was modelled over two orientations, labelled ‘A’ and
‘B’, with refined occupancies of 0.72 and 0.28, respectively.
Second, the tert-butyl group C(38)–C(41) was modelled over
three orientations with fixed occupancies of 0.40, 0.40 and 0.20.
All disordered C–C and C–N bonds in both these residues were
˚
˚
restrained to 1.54(2) A and 1.47(1) A respectively, and 1,3-
C · · · C distances within a given disorder orientation to 2.51(2)
−
˚
A. Third, the BF4 anion B(49)–F(53) was modelled over three
Electrochemical measurements were carried out using an
Autolab PGSTAT20 voltammetric analyser, in pre-dried dmf
or CH2Cl2 containing 0.1 M or 0.5 M NnBu4BF4 (prepared
from NnBu4OH and aqueous HBF4), respectively, as supporting
electrolyte. Voltammetry experiments used a Pt disk working
electrode, a Pt wire counter electrode and an Ag/AgCl reference
electrode. All potentials quoted are referenced to an internal
ferrocene/ferrocenium standard and were obtained at a scan
rate of 100 mV s−1. Coulometric determinations were obtained
using a 3-compartment cell, with a Pt basket working electrode,
and Pt wire and Ag/AgCl counter and reference electrodes as
before.
orientations with fixed occupancies of 0.50, 0.25 and 0.25. The
first two of these partial anions share a common B atom B(49A),
which therefore has a total occupancy of 0.75. All B–F distances
˚
were restrained to 1.37(2) A, and F · · · F distances within a given
˚
disorder orientation to 2.24(2) A. Finally, a disordered region
of solvent was modelled using five partial CH3CN molecules
with fixed occupancies leading to a total stoichiometry of 1.45
CH3CN molecules per asymmetric unit. All C–C bonds within
˚
these molecules were restrained to 1.46(1) A, C≡N distances to
˚
˚
1.16(1) A and 1,3-C · · · N distances to 2.62(1) A. Five residual
Fourier peaks of 1.0–1.6 e A−3 in the final model all lie within this
˚
disordered solvent region. All non-H atoms with occupancies
>0.5 were refined anisotropically, and all H atoms were placed
in calculated positions and refined using a riding model. The
torsions of the crystallographically ordered methyl groups were
allowed to refine freely, while the disordered methyl groups were
fixed in staggered conformations.
Acknowledgements
Funding of this work by the EPSRC and the University of Leeds
is acknowledged.
3 2 4 8
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9

View Article Online
24 See e.g. J. Beretka, B. O. West and M. J. O’Connor, Aust. J. Chem.,
1964, 17, 192; R. S. Downing and F. L. Urbach, J. Am. Chem. Soc.,
1969, 91, 5977; D. F. Rohrbach, W. R. Heineman and E. Deutsch,
Inorg. Chem., 1979, 18, 2536; J. Csaszar, Acta Phys. Chem., 1984,
30, 61; R. Atkins, G. Brewer, E. Kokot, G. M. Mockler and E. Sinn,
Inorg. Chem., 1985, 24, 127; T. Tanaka, K. Tsurutani, A. Komatsu, T.
Ito, K. Iida, Y. Fujii, Y. Nakano, Y. Usui, Y. Fukuda and M. Chikira,
Bull. Chem. Soc. Jpn., 1997, 70, 615.
25 E. H. Charles, L. M. L. Chia, J. Rothery, E. L. Watson, E. J. L.
McInnes, R. D. Farley, A. J. Bridgeman, F. E. Mabbs, C. C.
Rowlands and M. A. Halcrow, J. Chem. Soc., Dalton Trans., 1999,
2087.
References
1 P. Chaudhuri and K. Wieghardt, Prog. Inorg. Chem., 2001, 50, 151.
2 J. Stubbe and W. A. van der Donk, Chem. Rev., 1998, 98, 705; R. P.
Pesavento and W. A. van der Donk, Adv. Protein Chem., 2001, 58,
317.
3 J. W. Whittaker, Chem. Rev., 2003, 103, 2347.
4 N. Ito, S. E. V. Phillips, C. Stevens, Z. B. Ogel, M. J. McPherson, J. N.
Keen, K. D. S. Yadav and P. F. Knowles, Nature (London), 1991,
350, 87.
5 M. A. Halcrow, Angew. Chem., Int. Ed., 2001, 40, 346.
6 S. Itoh, M. Taki and S. Fukuzumi, Coord. Chem. Rev., 2000, 198,
3; B. A. Jazdzewski and W. B. Tolman, Coord. Chem. Rev., 2000,
200–202, 633; J.-L. Pierre, Chem. Soc. Rev., 2000, 29, 251.
7 For recent examples see: Y. Shimazaki, S. Huth, S. Hirota and O.
Yamauchi, Inorg. Chim. Acta, 2002, 331, 168; L. Benisvy, A. J. Blake,
D. Collison, E. S. Davies, C. D. Garner, E. J. L. McInnes, J. McMaster,
G. Whittaker and C. Wilson, Dalton Trans., 2003, 1975; A. Philibert,
F. Thomas, C. Philouze, S. Hamman, E. Saint-Aman and J.-L. Pierre,
Chem. Eur. J., 2003, 9, 3803; O. Se´ne`que, M. Campion, B. Douziech,
M. Giorgi, Y. Le Mest and O. Renaud, Dalton Trans., 2003, 4216;
F. Michel, F. Thomas, S. Hamman, E. Saint-Aman, C. Bucher and
J.-L. Pierre, Chem. Eur. J., 2004, 10, 4115.
26 E. N. Baker, D. Hall and T. N. Waters, J. Chem. Soc. A, 1970, 406
and refs. therein.
27 See e.g. C. Freiburg, W. Reichert, M. Melchers and B. Engelen, Acta
Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1980, B36,
1209; A. Berkessel, J. W. Bats and C. Schwarz, Angew. Chem., Int.
Ed., 1990, 29, 106; I. Morgenstern-Badarau, D. Laroque, E. Bill, H.
Winkler, A. X. Trautwein, F. Robert and Y. Jeannin, Inorg. Chem.,
1991, 30, 3180.
28 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877.
29 S. S. Eaton, K. M. More, M. Sawant and G. R. Eaton, J. Am. Chem.
Soc., 1983, 105, 6560; S. S. Eaton, G. R. Eaton and C. K. Chang,
J. Am. Chem. Soc., 1985, 107, 3155.
8 M. A. Halcrow, L. M. L. Chia, X. Liu, E. J. L. McInnes, L. J.
Yellowlees, F. E. Mabbs and J. E. Davies, Chem. Commun., 1998,
2465.
30 S. J. Gruber, C. M. Harris and E. Sinn, J. Inorg. Nucl. Chem., 1968,
30, 1805.
9 M. A. Halcrow, L. M. L. Chia, X. Liu, E. J. L. McInnes, L. J.
Yellowlees, F. E. Mabbs, I. J. Scowen, M. McPartlin and J. E. Davies,
J. Chem. Soc., Dalton Trans., 1999, 1753.
10 I. Sylvestre, J. Wolowska, E. J. L. McInnes, C. A. Kilner and M. A.
Halcrow, Inorg. Chim. Acta, 2005, 358, 1337.
11 X. Liu, S. A. Barrett, C. A. Kilner, M. Thornton-Pett and M. A.
Halcrow, Tetrahedron, 2002, 58, 603.
12 X. Liu, L. M. L. Chia, C. A. Kilner, L. J. Yellowlees, M. Thornton-
Pett, S. Trofimenko and M. A. Halcrow, Chem. Commun., 2000,
1947.
31 See e.g. J. M. Epstein, B. N. Figgis, A. H. White and A. C. Willis,
J. Chem. Soc., Dalton Trans., 1974, 1954; C. J. O’Connor, D. P.
Freyberg and E. Sinn, Inorg. Chem., 1979, 18, 1077; A. Bencini,
C. Benelli, A. Caneschi, R. L. Carlin, A. Dei and D. Gatteschi,
J. Am. Chem. Soc., 1985, 107, 8128; G. A. Brewer and E. Sinn, Inorg.
Chem., 1987, 26, 1529; I. Morgenstern-Badarau, D. Laroque, E. Bill,
H. Winkler, A. X. Trautwein, F. Robert and Y. Jeannin, Inorg. Chem.,
1991, 30, 3180; J.-P. Costes, F. Dahan, A. Dupuis and J.-P. Laurent,
New J. Chem., 1997, 22, 1525; I. Ramade, O. Kahn, Y. Jeannin and F.
Robert, Inorg. Chem., 1997, 36, 930; M. L. Kahn, T. M. Rajendiran,
Y. Jeannin, C. Mathoniere and O. Kahn, C. R. Acad. Sci., Ser. IIc,
2000, 3, 131.
13 L. M. L. Chia, A. E. H. Wheatley, N. Feeder, J. E. Davies and M. A.
Halcrow, Polyhedron, 2000, 19, 109.
14 Y. Wang and T. D. P. Stack, J. Am. Chem. Soc., 1996, 118, 13097.
15 Y. Wang, J. L. DuBois, B. Hedman, K. O. Hodgson and T. D. P.
Stack, Science, 1998, 279, 537.
32 J. A. Halfen, B. A. Jazdzewski, S. Mahapatra, L. M. Berreau, E. C.
Wilkinson, L. Que, Jr. and W. B. Tolman, J. Am. Chem. Soc., 1997,
119, 8217.
33 D. Zurita, C. Scheer, J.-L. Pierre and E. Saint-Aman, J. Chem. Soc.,
Dalton Trans., 1996, 4331; S. Itoh, S. Takayama, S. Arakawa, A.
Furuta, M. Komatsu, A. Ishida, S. Takamuku and S. Fukuzumi,
Inorg. Chem., 1997, 36, 1407.
34 S. Itoh, M. Taki, H. Kumei, S. Takayama, S. Nagamoto, T. Kitagawa,
N. Sakaruda, R. Arakawa and S. Fukuzumi, Inorg. Chem., 2000, 39,
3708.
35 M. Taki, H. Kumei, S. Nagamoto, T. Kitagawa, S. Itoh and S.
Fukuzumi, Inorg. Chim. Acta, 2000, 300–302, 622; M. Taki, H.
Hattori, T. Osako, S. Nagatomo, M. Shiro, T. Kitagawa and S. Itoh,
Inorg. Chim. Acta, 2004, 357, 3369.
16 M. Nakamura, H. Okawa and S. Kida, Inorg. Chim. Acta, 1982, 62,
201.
17 Y. N. Belokon, M. North, T. D. Churkina, N. S. Ikonnikov and V. I.
Maleev, Tetrahedron, 2001, 57, 2491; Y. N. Belokon, R. G. Davies,
J. A. Fuentes, M. North and T. Parsons, Tetrahedron Lett., 2001, 42,
8093.
18 R. C. Pratt and T. D. P. Stack, J. Am. Chem. Soc., 2003, 125, 8716.
19 F. Thomas, O. Jarjayes, C. Duboc, C. Philouze, E. Saint-Aman and
J.-L. Pierre, Dalton Trans., 2004, 2662.
20 Y. Shimazaki, F. Tani, K. Fukui, Y. Naruta and O. Yamauchi, J. Am.
Chem. Soc., 2003, 125, 10512.
21 C. Vottero, Y. Labat and J.-M. Poirier, Eur. Pat. 19890531, 1989, 5
36 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, A46, 467.
37 G. M. Sheldrick, SHELXL 97 Program for the Refinement of Crystal
Structures, University of Go¨ttingen, Germany, 1997.
38 F. E. Mabbs and D. Collison, Electron Paramagnetic Resonance of d
Transition Metal Compounds, Elsevier, Amsterdam, 1992, ch. 7.
pp.
22 N. U. Hofslokken and L. Skattebol, Acta Chem. Scand., 1999, 53,
258.
23 M. M. Bhadbhade and D. Srinivas, Inorg. Chem., 1993, 32, 6122.
D a l t o n T r a n s . , 2 0 0 5 , 3 2 4 1 – 3 2 4 9
3 2 4 9