1,5-Diarylbiguanides and their nickel(ii) complexes

Article abstract of DOI:10.1039/c2dt32483k

1,5-Diarylbiguanides, where the aryl groups are phenyl (HL1), 3,5-dimethylphenyl (HL2), 3,5-dimethoxyphenyl (HL3), 4-t-butylphenyl (HL4) or 4-bromophenyl (HL5), have been prepared and characterised. HL3 and HL5 have been structurally characterised by X-ray crystallography, which shows them to adopt the expected tautomeric form for biguanides. They have extensive hydrogen-bonding interactions in the solid state, involving the biguanide NH groups supported by, in the case of HL3, the OCH₃ aryl substituents or, in the case of HL5, Br?Br interactions. Reactions of HL1-HL4 with Ni(BF₄)₂ gives complexes of the type [Ni(HL) ₂](BF₄)₂, while reactions of HL1-HL4 with Ni(BF₄)₂ and triethylamine give neutral complexes of the type [Ni(L)₂], where the biguanide ligand has been deprotonated at the N_ring nitrogen. Both series of complexes were characterised in solution and the solid state. Cyclic voltammetry shows a largely irreversible Ni(ii)/Ni(iii) oxidation which becomes easier by ca. 70 mV upon ligand deprotonation, with more subtle variations resulting from the changes in aryl ring substituents. Infrared and ¹H NMR spectroscopies both provide evidence for ligand deprotonation leading to the chelate ring becoming increasingly aromatised. X-ray crystallographic analyses of five of the complexes also show changes in bond lengths and angles within the chelate ring, consistent with increased electron delocalisation. A variety of hydrogen bonding motifs involving the complex ions, counterions and solvent molecules are found. The results of DFT calculations on both cationic and neutral complexes provide calculated structures consistent with the experimental ones and these, along with the results of vibrational spectroscopic studies, provide further evidence for increased aromatisation upon deprotonation. The potential for the complexes to act as tectons for the rational assembly of hydrogen bonded metallosupramolecules is discussed and the X-ray structure of such an assembly, between [Ni(L3)₂] and 1,8-naphthalimide, is presented.

Full text of DOI:10.1039/c2dt32483k

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
1,5-Diarylbiguanides and their nickel(II) complexes†
Cite this: Dalton Trans., 2013, 42, 2948
David A. McMorran,* C. John McAdam, Holly van der Salm and Keith C. Gordon
1,5-Diarylbiguanides, where the aryl groups are phenyl (HL1), 3,5-dimethylphenyl (HL2), 3,5-dimethoxy-
phenyl (HL3), 4-t-butylphenyl (HL4) or 4-bromophenyl (HL5), have been prepared and characterised. HL3
and HL5 have been structurally characterised by X-ray crystallography, which shows them to adopt the
expected tautomeric form for biguanides. They have extensive hydrogen-bonding interactions in the
solid state, involving the biguanide NH groups supported by, in the case of HL3, the OCH3 aryl substitu-
ents or, in the case of HL5, Br⋯Br interactions. Reactions of HL1–HL4 with Ni(BF4)2 gives complexes of
the type [Ni(HL)2](BF4)2, while reactions of HL1–HL4 with Ni(BF4)2 and triethylamine give neutral com-
plexes of the type [Ni(L)2], where the biguanide ligand has been deprotonated at the Nring nitrogen.
Both series of complexes were characterised in solution and the solid state. Cyclic voltammetry shows a
largely irreversible Ni(II)/Ni(III) oxidation which becomes easier by ca. 70 mV upon ligand deprotonation,
with more subtle variations resulting from the changes in aryl ring substituents. Infrared and 1H NMR
spectroscopies both provide evidence for ligand deprotonation leading to the chelate ring becoming
increasingly aromatised. X-ray crystallographic analyses of five of the complexes also show changes in
bond lengths and angles within the chelate ring, consistent with increased electron delocalisation.
A variety of hydrogen bonding motifs involving the complex ions, counterions and solvent molecules are
found. The results of DFT calculations on both cationic and neutral complexes provide calculated struc-
tures consistent with the experimental ones and these, along with the results of vibrational spectroscopic
studies, provide further evidence for increased aromatisation upon deprotonation. The potential for the
complexes to act as tectons for the rational assembly of hydrogen bonded metallosupramolecules is dis-
cussed and the X-ray structure of such an assembly, between [Ni(L3)2] and 1,8-naphthalimide, is
presented.
Received 18th October 2012,
Accepted 7th December 2012
DOI: 10.1039/c2dt32483k
www.rsc.org/dalton
area of research, new ligands which incorporate hydrogen
bonding motifs are required. Biguanide (1) was first reported
Introduction
The rational assembly of component molecules into larger in 18924 and, since then, various methods for the synthesis of
supramolecular species continues to be a major theme in 1 and its derivatives have been reported and several reviews
modern chemistry. In particular, the use of non-covalent inter- have appeared.5 While these molecules have received signifi-
actions to drive the assembly of transition-metal containing cant interest due to their varied biological properties,6–10 the
components into new species with interesting structural and coordination chemistry of 1 itself has also been extensively
functional properties has received increasing attention.1 Of the studied, particular by Ray and co-workers.11,12 These studies
various non-covalent interactions available, the most attractive, aimed to better understand the biological function of biguanide
in terms of strength and directionality, are hydrogen bonds. and also explore its ability to stabilise transition metals in
We,2 and others,3 have shown that, by incorporating suitable high oxidations states (e.g. Ni(III)13 and Ag(III)14). While a wide
hydrogen bond donor and/or acceptor groups into the ligands, variety of transition metal ions are found to form stable com-
rational assembly of complex ions can be achieved, either plexes with 1, it was only more recently, once X-ray structural
directly with each other or to other molecular fragments with data for 1 and its complexes were obtained, that the nature of
complementary hydrogen bonding groups. To advance this the coordination itself was resolved. It is now known that, in
the solid state, 1 adopts the tautomer shown in Scheme 1a,5 in
which an intramolecular hydrogen bond stabilises a largely
planar conformation, whereas in complex ions, the neutral
ligand coordinates as shown in Scheme 1b.
Department of Chemistry, University of Otago, PO Box 56, Dunedin, New Zealand.
E-mail: davidm@chemistry.otago.ac.nz; Fax: +64 03 479 7906; Tel: +64 03 479 7934
†Electronic supplementary information (ESI) available. CCDC 843769–843775.
Metal complexes of 1 act as weak acids, with the hydrogen
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c2dt32483k
on the chelate ring nitrogen (Nring) being readily lost12
2948 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Scheme 1 Tautomeric forms of free (a) and coordinated (b) biguanide 1.
Scheme 2 Neutral (a) and deprotonated (b) forms of coordinated biguanide 1.
(Scheme 2). The resulting delocalisation of charge within the
chelate ring upon deprotonation has been demonstrated by a
number of X-ray crystallographic studies, which show that the
coordinated neutral ligand has C–N bond lengths in the order
C–Ndonor < C–Nterminal < C–Nring (with the C–Ndonor length
often less than 1.3 Å), whereas coordinated deprotonated
ligand has the order C–Ndonor < C–Nring < C–Nterminal, with the
C–Ndonor lengthening and becoming more similar to the
C–Nring length.12 A characteristic decrease in the C–Nring–C
bond angle from ca. 126° to ca. 120° is also observed.
Scheme 3 Possible hydrogen bonding motifs (green arrows) for metal com-
plexes of neutral {(a) and (b)} and deprotonated {(c) and (d)} diarylbiguanide
ligands. Red arrows indicate hydrogen-bonding sites potentially blocked by
adjacent aryl rings.
Much literature exists on the coordination chemistry of 1,
and, to a lesser extent, derivatives of 1, typically obtained by
substitution at the donor or terminal nitrogen atoms.13–15
Only one report of complexes involving 1,5-diaryl substituted
biguanides exists, however, in which Masuda and coworkers
describe the preparation of Ni(II) and Cu(II) complexes of
deprotonated 1,5-diphenylbiguanide and demonstrate their
potential as tectons for supramolecular species.16 Given the
obvious structural similarities of these complexes with mel-
amine, a hydrogen bonding tecton par excellence,17 it seems
that a thorough exploration of the solution and solid-state
chemistries of such compounds, so as to ascertain their poten-
tial for the construction of assemblies with interesting struc-
tural and functional properties, is warranted.
Consideration of the structures in Scheme 2 show that, in
both cases, the biguanide ligand has great potential for acting
as a tecton for assembly by hydrogen bonding.18–20 Scheme 3
depicts the possible hydrogen bonding motifs that might be
expected for the 1,5-diarylbiguanides in the current complexes.
With the aryl rings in an anti–anti conformation (a) and (c),
donor–donor–donor (DDD) and donor–acceptor–donor (DAD)
triple hydrogen bonding motifs, respectively, are possible. In
blocked by the two aryl rings. Syn–anti conformations might
be expected to give intermediate situations.
In the current paper, a series of 1,5-diarylbiguanides and
their nickel(II) complexes are reported. Based on previous
studies,16,21,22 it was expected that the use of Ni(II) would
reliably give square planar species. Complexes, which contain
either the ligands in their neutral form (giving complex salts)
or in their deprotonated form (giving neutral complex
species), have been synthesised and characterised, both in
solution and in the solid state. DFT calculations are also used
to characterise their electronic properties. As such, this is, to
our knowledge, the first time that, within the one study, a
thorough comparison of the chemistries of a series of com-
plexes with both neutral and deprotonated biguanide ligands
has been reported.
Results and discussion
these cases, the Ndonor–H hydrogens may not be able to hydro- The ligands HL1–HL5 were prepared in a straightforward
gen bond due to the proximity of the aryl rings. Conversely, manner following the method described by Wuest and co-
with the aryl rings in a syn–syn conformation, DDDD motifs, workers.5a Two equivalents of the hydrochloride salt of the
involving the Ndonor–H and Naryl–H hydrogens, are possible for respective anilines were refluxed with sodium dicyanamide
complexes incorporating either the neutral and the deproto- overnight, with the free ligand being then obtained from the
nated ligands. In this case, the Nring position is potentially resulting hydrochloride salt by treatment with sodium
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2949

View Article Online
Paper
Dalton Transactions
Scheme 5 Synthesis of nickel(II) complexes. (i) 1/2 equiv. Ni(BF4)2·6H2O,
Scheme 4 Syntheses of 1,5-biarylbiguanides. (i) NaN(CN)2, dil. HCl, reflux; (ii)
acetone, RT, (ii) 1/2 equiv. Ni(BF4)2·6H2O, 1 equiv. (CH3CH2)3N, methanol, RT.
NaOCH3, methanol, RT.
methoxide (Scheme 4). Ligands were obtained in good to high suggesting that, in the DMF/CH3CN solution used for the
yield and in excellent purity directly from the reaction mix- measurements, DMF molecules can readily hydrogen bond to
tures. The new ligands HL3 and HL4 were characterised by the NH protons in the complex ions (Scheme 5).
microanalysis, ESI-MS and IR and 1H NMR spectroscopies.
While the infrared spectra of a number of simple biguanide
Reactions of HL1 to HL4 with 1/2 equiv. of Ni(BF4)2·6H2O in complexes have been reported, there are few detailed compara-
acetone gave an orange-red solution which, upon addition of tive studies on pairs of complexes containing both the neutral
diethyl ether, gave the analytically pure [Ni(HL)2](BF4)2 salts in and deprotonated ligands, respectively.24,25 It might be
good yield. Microanalyses and ESI-MS confirmed the desired expected that, upon deprotonation, the bond orders of the CN
1 : 2 Ni : L ratio. Under the experimental conditions used for bonds in the chelate ring will become more similar, and this
the ESI-MS, deprotonation of the ligands in the cationic should result in the CvN stretches moving to lower energy
nickel(II) complexes was observed.
and the C–N stretches moving to higher energy. Infrared
Reactions between Ni(BF4)2·6H2O and HL5 under these spectra were recorded on both series of the present nickel
conditions did not give precipitates in the same way. Orange complexes. In all cases, a series of weak peaks in the region
solids could be isolated by addition of diethyl ether or petro- ca. 3400–3050 cm−1 were observed, corresponding to N–H
leum ether to the orange reaction solutions, however micro- stretches. In the complexes containing neutral ligands, strong
analyses consistent with the desired complex could not be CvN bands and medium to strong C–N stretches were
obtained. This may be due to nickel species formed in the observed at ca. 1670–1600 cm−1 and 1050–990 cm−1, respecti-
solution reacting with the C–Br bonds in the ligands.23 The vely. This is consistent with spectra reported previously.24 In
natures of the products of these reactions were not explored contrast, spectra of the deprotonated ligands had no strong
further. It is further noted that attempts to prepare these com- peaks in these areas (except for [Ni(L3)2] which had strong
plexes in methanol gave products for which correct micro- C–O bands at ca. 1150 cm−1). Instead, strong peaks at
analyses could not be obtained – the results suggested that ca. 1600–1450 cm−1 are observed. This is consistent with the pro-
partial ligand deprotonation occurs in this solvent, giving posal that, upon deprotonation, the chelate ring becomes
samples with non-stoichiometric numbers of counterions.
increasing aromatised, thus lowering the bond order of the
The reaction of each ligand with 1/2 equiv. of Ni(BF4)2· CvN bonds and increasing the bond order of the C–N bonds.
6H2O in methanol, followed by addition of one equivalent of A more detailed analysis of these effects is provided by DFT
triethylamine, gave orange-red solutions which deposited calculations and Raman spectra (vide infra).
orange precipitates. Microanalyses of these precipitates
1H NMR spectra were recorded for the ligands and the com-
showed them to have a 1 : 2 Ni : L ratio and confirmed that the plexes in various solvents. For comparison, spectra of all com-
ligands had been successfully deprotonated as there were no pounds were recorded in d7-DMF, but the spectra of the
counterions present. Again, reactions with HL5 appeared to ligands were also recorded in CDCl3 and some of the complex
proceed differently and pure compounds could not be obtained. spectra were also recorded in d6-DMSO. It has been reported
In the ESI-MS spectra, each pair of complexes ([Ni(HL)2](BF4)2 previously that signals corresponding to the NH protons in the
and [Ni(L)2]) gave essentially the same spectra, with the major ligands are poorly resolved5a and this was found to be the case
peak corresponding to [Ni(HL)(L)]+. Further, in some cases, in the current study. In d7-DMF, extremely broad signals could
peaks corresponding to DMF adducts were observed, be identified at ca. 6.4–7.2 ppm, while in d6-DMSO and CDCl3,
2950 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
no peaks could be seen at all. This is presumably due to the
d
xy → dx2−y2 (1A1g → 1A2g) transition.25,26 UV/Visible spectra of the
NH protons being involved in hydrogen bonding interactions current complexes were recorded in DMF and, in each case,
with solvent molecules. It has further been reported that the spectra consistent with a square planar d8 system were found.
protonated salts of the ligands give much better resolved NH For the cationic nickel(II) complexes,
a
single band at
1
peaks in the H NMR spectra.5a Similar results were found for ca. 445 nm is observed (extinction coefficients 100–120 mol−1
L
the spectra of the nickel(II) complexes of the neutral ligand. cm−1), while for the neutral complexes, this band shifts to
A peak attributable to the Nring–H proton is observed at ca. 436 nm (extinction coefficients 110–170 mol−1 L cm−1) and a
ca. 9.6 ppm for each of the complexes. Two other peaks attribu- low energy shoulder at ca. 490 nm is also observed. This small
table to the other NH protons, are also observed; one of these shift to higher energy implies that the deprotonated ligand is a
remains at ca. 7.6 ppm for all the complexes, while the pos- better donor than the neutral ligand, as has previously been
ition of the second one varies from 7.27 ppm (for [Ni(HL4)2]2+) proposed for complexes of 1.25 Calculated spectra (vide infra)
to 6.17 ppm (for [Ni(HL3)2]2+). The former peaks are therefore confirm this and show that the transitions are in fact metal–
assigned to the Ndonor–H hydrogens, while the latter are ligand charge transfer in nature.
assigned to the Naryl–H hydrogens, which would be expected to
The electrochemistry of the complexes was probed using
be more influenced by the natures of the substituents on the cyclic voltammetry in DMF solution. Results are presented in
aryl rings. In each of the spectra, small peaks due to free Table 1 and representative voltammograms in Fig. 2 and S1.†
ligands were also present, showing that, in DMF solvent, Previous literature studies on similar biguanide systems have
some degree of ligand dissociation occurs. In the case of been performed in aqueous solution, making comparison with
[Ni(HL2)2]2+ a variable temperature study was carried out and this our results difficult.27 For all complexes an anodic sweep at
showed neither an increase in the amount of free ligand 100 mV s−1 gives rise to an irreversible feature attributed to
present, nor a coalescence of the peaks due to free and coordi- the Ni3+/2+ oxidation couple. The reversibility of this process
nated ligand.
improved with increasing scan rate (Fig. S1†). Sweeping to
1
In H NMR spectra of the neutral nickel(II) complexes, the higher anodic potentials reveals another irreversible process
peak at ca. 9.6 ppm is lost, as expected. The peak at ca. 1.2 V typical of amine oxidation.28 Removal of the protons
ca. 7.6 ppm, due to the coordinated Ndonor–H hydrogens, shar- from [Ni(HL)2]2+ shifts Epa for [Ni(L)2] approximately 70 mV
pens and moves downfield to ca. 8 ppm (Fig. 1), consistent cathodically, concordant with the increase in electron density
with the chelate ring becoming increasingly aromatic. at the nickel centre (Fig. 2).
However, peaks due to the Naryl–H hydrogens could not be
resolved (as was the case in the free ligands). The reason for
this is not clear but it may be that these NH protons are
Table 1 Electrochemistry data
involved in hydrogen bonding interactions with the DMF
solvent molecules, consistent with the observations from the
Epa/V
[Ni(HL1)2](BF4)2
[Ni(HL2)2](BF4)2
[Ni(HL3)2](BF4)2
[Ni(HL4)2](BF4)2
[Ni(L1)2]
[Ni(L2)2]
[Ni(L3)2]
0.79
0.77
0.80
0.76
0.72
0.69
0.75
0.69
ESI-MS spectra. There was no evidence of free ligand in these
solutions, suggesting that the deprotonated ligand is more
strongly coordinated than the neutral ligand.
Previous studies on the UV/Visible spectra of cationic
nickel(II) biguanide complexes report
a single band at
ca. 445 nm, consistent with a square planar geometry and, based
[Ni(L4)2]
on solvent studies, this was assigned to a metal centred
Solutions ∼1 × 10−3 mol L−1 in DMF with 0.1 M Bu4NPF6. Values
referenced to [Fc*]+/0 = 0.00 V.
Fig. 1 1H NMR spectra of [Ni(HL3)2](BF4)2 (top) and [Ni(L3)2] (bottom) in
d7-DMF solution. Arrow indicates movement of the Ndonor–H hydrogen peak
upon deprotonation. # solvent, * free ligand.
Fig. 2 Cyclic voltammograms of [Ni(HL3)2](BF4)2 and [Ni(L3)2] (ca.
1
×
=
10−3 mol L−1 in DMF, 0.1 mol L−1 Bu4NPF6, 100 mV s−1, referenced to [Fc*]+/0
0.00 V).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2951

View Article Online
Paper
Dalton Transactions
Fig. 3 X-ray structures of HL3 (a) and HL5 (b), with ellipsoids drawn at the 50% level. (c) Hydrogen bonding interactions joining adjacent HL3 molecules. (d) Hydro-
gen bonding interactions, which assemble adjacent HL5 molecules parallel to the b axis; (e) Br⋯Br interactions, which generate helical chains of HL5 molecules par-
allel to the c axis. Non-NH hydrogens omitted for clarity.
Table 2 Selected bond lengths (Å) and angles (°) for HL3, HL5 and HL1a
The various phenyl substituents on HL1–HL4 provide a
subtle tuning of Epa commensurate with their predicted induc-
tive effect. Thus the 3,5-dimethylphenyl derivatives have nickel
oxidation potentials ca. 30 mV lower than the phenyl deriva-
tives, and the inductively negative 3,5-dimethoxyphenyl deriva-
tives raise Epa a similar amount.
HL3
HL5
HL1
N1–C1
C1–N2
N2–C2
C2–N3
C1–N4
C2–N5
N1–C1–N2
N1–C1–N4
C1–N2–C2
N4–C1–N2
N2–C2–N3
N2–C2–N5
N3–C2–N5
1.3580(18)
1.3724(18)
1.3408(17)
1.3494(19)
1.3168(18)
1.3390(18)
113.06(13)
122.66(13)
121.25(12)
124.24(12)
116.57(13)
124.41(13)
118.98(13)
1.329(5)
1.322(5)
1.378(5)
1.395(5)
1.372(5)
1.298(5)
126.4(4)
115.4(4)
120.8(3)
118.1(3)
110.0(3)
126.9(4)
123.0(4)
1.321
1.305
1.378
1.424
1.388
1.321
126.73
115.52
120.75
117.73
110.59
127.23
121.87
X-Ray crystallography
The solid state structures of a number of the nickel(II) com-
plexes, as well as two of the ligands were obtained by X-ray
crystallography, in order to confirm the structures and explore
the nature of the bonding within the chelate rings. Charac-
terisation of the hydrogen bonding motifs present provides
understanding of the potentials of these molecules for
forming higher assemblies.
a From ref. 5a.
The structures of the ligands HL3 and HL5 are both found
to exist in the expected tautomer (Scheme 1) with the central
C2N5 unit being planar and stabilised by intramolecular
motifs, involving hydrogen atoms on the N3 atoms of two
molecules and the O atoms in the OCH3 groups in two mole-
hydrogen bonding. HL3 crystallises in the triclinic space group cules. The chains propagate parallel to the crystallographic c
ˉ
axis. The remaining aryl rings, not involved in this hydrogen
bonding system, point out from the chain and interdigitate
P1 with the asymmetric unit containing one ligand molecule
(Fig. 3a, Table 2). An internal R11(6)29 hydrogen bond between
N4 and the hydrogen atom on N5 stabilises the arrangement with related rings from the adjacent chains, stabilised by π–π
stacking interactions (intercentroid distance is 3.840 Å, closest
(Table 5). The OCH3 groups are arranged in a syn–anti confor-
C⋯C distance 3.804 Å). In this way the chains are arranged
mation on one of the aryl rings and in an anti–anti confor-
mation on the other. Pairs of molecules are hydrogen bonded into sheets lying parallel to the crystallographic ab diagonal.
2
The ligand HL5 crystallises in the non-centric ortho-
rhombic space group P212121, with the asymmetric unit con-
to each other by a R2 (8) cyclic motif, involving N2 and a
hydrogen atom on N3 in each molecule (Fig. 3c). These pairs
are further hydrogen bonded into chains by R4 (24) cyclic tains one ligand molecule (Fig. 3b, Table 2). The conformation
4
2952 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
of the HL5 is very similar to that found for HL1,5a with an helical chains, all of which have the same handedness due to
1
internal R1 (6) hydrogen bond between N5 and one of the the non-centric space group.
hydrogen atoms on N1 stabilising this arrangement (Table 5).
X-ray structures of two of the cationic complexes were
2
Cyclic R2 (8) hydrogen bonding motifs, involving NH and NH2 obtained, and in each, the anions were found to hydrogen
protons on one molecule and the central N and an NH2 (acting bond to the complexes. In each case the pattern of CN bond
as an acceptor) on a second molecule, generate chains which lengths within the C2N5 core is the same as that previously
run parallel to the crystallographic b axis (Fig. 3d, Table 5). reported12 for coordinated neutral ligands with Ndonor–C <
Individual HL5 molecules are also assembled into one-dimen-
Naryl–C < Nring–C and the C–Nring–C angle being ca. 126°
sional helical chains by Br⋯Br interactions (Br⋯Br distance is (Table 3). [Ni(HL3)](BF4)2(C3H6O)2 crystallises in the triclinic
ˉ
3.59 Å), which run parallel to the crystallographic c axis and space group P1. The asymmetric unit contains a Ni atom, posi-
−
have a pitch of 26.8 Å (Fig. 3e). (The two aryl rings are at an tioned on a two-fold axis, one HL3 ligand and a BF4 anion,
angle of 82.76° to each other, giving the molecule a plane of along with an acetone solvate molecule (Fig. 4a, Table 3). The
chirality.) The combined effect of these two, orthogonal, inter- ligand adopts a syn–syn conformation, with the aryl rings lying
molecular interactions is to generate chiral sheets of nested almost parallel to each other. The OCH3 groups on each of
the rings are themselves in a syn–anti arrangement, with the
groups on each ring arranged in the same way and lying in the
plane of the aryl rings (Fig. 4a). The syn–syn conformation of
the ligands means that the Nring–H hydrogens are not available
for hydrogen bonding. However, both the Ndonor–H hydrogens
and the Naryl–H hydrogens are involved in relatively weak
Table 3 Selected bond lengths (Å) and angles (°) for [Ni(HL3)2](BF4)2·C3H6O
and [Ni(HL2)2](ClO4)2·(CH3CN)2
[Ni(HL3)2](BF4)2·(C3H6O)2 [Ni(HL2)2](ClO4)2·(CH3CN)2
−
hydrogen bonds to the BF4 counterions (Table 5). Two of the
Ni1–N1
Ni1–N3
N1–C1
C1–N2
N2–C2
C2–N3
C1–N4
C2–N5
1.854(4)
1.859(3)
1.286(5)
1.362(5)
1.363(5)
1.294(5)
1.356(5)
1.360(5)
1.8582(16)
1.8612(16)
1.295(2)
1.368(2)
1.371(2)
1.295(2)
1.350(2)
1.349(3)
89.66(7)
−
1
fluorine atoms in the BF4 forming contiguous R2 (6) hydro-
gen bonds to two Ndonor–H hydrogens and to one Ndonor–H
and one Naryl–H hydrogen, respectively (Table 5). The acetone
oxygen hydrogen bonds to the remaining Naryl–H hydrogen
and then is further hydrogen bonded (more weakly) to the
related Naryl–H hydrogen on an adjacent [Ni(HL3)2]2+ cation.
This, along with relatively strong π–π stacking interactions
between aryl rings on adjacent cations (closest C⋯C distance
is 3.48 Å) links the cations into chains that run parallel to the
crystallographic b axis.
N1–Ni1–N3 90.16(16)
Ni1–N1–C1
N1–C1–N2
N1–C1–N4
C1–N2–C2
N4–C1–N2
N2–C2–N3
N2–C2–N5
130.8(3)
120.7(4)
123.0(4)
126.5(4)
116.2(3)
121.3(4)
115.5(4)
129.45(13)
121.06(16)
125.24(17)
125.27(16)
113.69(16)
120.83(17)
113.96(17)
129.58(13)
125.19(17)
A small number of crystals of [Ni(HL2)2](ClO4)2(CH3CN)2
were obtained after reaction of HL2 with Ni(ClO4)2 rather than
Ni(BF4)2. The complex crystallised in the triclinic space group
C2–N3–Ni1 129.9(3)
N3–C2–N5 123.1(4)
Fig. 4 X-ray structures of [Ni(HL3)2](BF4)2·C3H6O (a) and [Ni(HL2)2](ClO4)2·(CH3CN)2 (b), with ellipsoids drawn at the 50% level. Disordered aryl rings, perchlorate
oxygens and acetonitrile carbons in [Ni(HL2)2](ClO4)2·(CH3CN)2 are omitted for clarity.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2953

View Article Online
Paper
Dalton Transactions
ˉ
P1 with the asymmetric unit containing a nickel(II) cation, ac diagonal. The acetonitrile solvate molecules are hydrogen
lying on a two-fold rotation axis, one complete HL2 ligand, bonded via the nitrogen atom to the remaining Naryl–H hydro-
one perchlorate anion and an acetonitrile solvate molecule gens, one per ligand (Table 5). The Ndonor–H hydrogens are not
(Fig. 4b, Table 3). Parts of the HL2 ligand, the perchlorate involved in any hydrogen bonding interactions. The CH3 group
anion and the acetonitrile molecule were disordered and were of each acetonitrile molecule lies above the nickel(II) ion of an
modelled appropriately (Experimental section). In contrast to adjacent [Ni(L1)2] hydrogen bonds between unit, forming a
the structure found for [Ni(HL3)2](BF4)2(C3H6O)2, the HL2 C–H⋯Ni anagostic interaction,30 with an H⋯Ni distance of
ligands adopt an anti–anti conformation, meaning that the 2.83 Å and a C–H⋯Ni angle of 171.23°. The arrangement of
N
donor–H protons are not involved in hydrogen bonding inter- the ligands is such that two acetonitrile molecules lie on the
actions. Rather all three remaining NH groups act as donors, same side of the NiN4 planes of adjacent [Ni(L1)2] units, brid-
with adjacent Nring–H and Naryl–H groups hydrogen bonding ging the nickel(II) ions and the NH donors. Thus adjacent
2
to two oxygens of the perchlorate anion, forming a R2 (8) [Ni(L1)2] units in the chain are assembled not only by the
2
motif. The third Naryl–H hydrogen bonds to the nitrogen atom R2 (8) cyclic hydrogen bonding but also by the hydrogen
of the acetonitrile solvate molecule (Table 5).
The structures of two of the neutral complexes, containing (Fig. 5d).
deprotonated ligands, were also obtained. In the absence of [Ni(L2)2]·(DMSO)4 crystallised in the triclinic space group
bonding/anagostic bonding of pairs of acetonitrile molecules
ˉ
anions, the complexes are found to hydrogen bond to solvent P1 with the asymmetric unit containing a nickel(II) ion, lying
molecules and, in one case, to each other. In both cases the on a two-fold rotation axis, one deprotonated HL2 ligand and
pattern of CN bond lengths within the C2N5 core is found to two DMSO solvate molecules (Fig. 5b, Table 4). The ligand now
be the same as that previously reported for coordinated depro- adopts a syn–anti conformation. One of the DMSO molecules
tonated ligands,12 with Ndonor–C < Nring–C < Naryl–C, and is hydrogen bonded to one of the Naryl–H hydrogens while the
2
the C–Nring–C angle is now decreased to ca. 120° (Table 4). other forms a bifurcated R2 (6) hydrogen bond to the second
[Ni(L1)2] is found to crystallise in the monoclinic space group
Naryl–H hydrogen and an adjacent Ndonor–H hydrogen
C2/c, with the asymmetric unit containing a nickel(II) ion, (Table 5). The remaining Ndonor–H hydrogen is not involved in
lying on a two-fold rotation axis, two halves of a deprotonated any hydrogen bonding interactions. Due to the presence of the
HL1 ligand and an acetonitrile solvate molecule (Fig. 5a, DMSO molecules, there are no hydrogen bonding interactions
Table 4). The ligand adopts an anti–anti conformation, with between the [Ni(L2)2] units themselves.
aryl rings on different ligands within each [Ni(L1)2] arranged
Consideration of the packing of the molecules in the crystal
in a pseudo-orthogonal fashion. [Ni(L1)2] molecules are hydro- shows that each of the DMSO molecules which are hydrogen
2
2
gen bonded to each other by a R2 (8) motif, involving the bonded to the Naryl–H hydrogen (only) form an R2 (8) cyclic
deprotonated Nring nitrogen and one of the other Naryl–H hydrogen bonding motif with a related DMSO molecule,
hydrogens (Fig. 5c). Adjacent molecules are not coplanar but which is, in turn, hydrogen bonded to an adjacent [Ni(L2)2]
rather lie with an angle of 45.49° between the NiN4 planes, pre- unit. In this way, chains of [Ni(L2)2]·(DMSO)4 units are gen-
sumably due to the steric interactions between aryl rings. This erated, which run parallel to the crystallographic ab diagonal
cyclic hydrogen bonding motif joins adjacent molecules (Fig. 5e).
to generate chains which run along the crystallographic
In an initial set of experiments to ascertain the ways in
which the deprotonated complexes might act as DAD hydrogen
bonding motifs towards complementary organic molecules,
each of the complexes were mixed with two equivalents of 1,8-
naphthalimide in concentrated DMSO solutions. After stand-
ing at room temperature for some weeks, the solution contain-
ing [Ni(L3)2] gave some crystals suitable for X-ray analysis. The
[Ni(L3)2]·(1,8-naphthalimide)2 assembly crystallised in the tri-
Table 4 Selected bond lengths (Å) and angles (°) for [Ni(L1)2]·(CH3CN)2,
[Ni(L2)2]·(DMSO)4 and [Ni(L3)2]·(1,8-naphthalimide)2
[Ni(L1)2]·
(CH3CN)2
[Ni(L2)2]·
(DMSO)4
[Ni(L3)2]·
(1,8 naphthalimide)2
Ni1–N1
Ni1–N3
N1–C1
C1–N2
N2–C2
C2–N3
C1–N4
C2–N5
N1–Ni1–N3
Ni1–N1–C1
N1–C1–N2
N1–C1–N4
C1–N2–C2
N4–C1–N2
N2–C2–N3
N2–C2–N5
C2–N3–Ni1
N3–C2–N5
1.851(2)
1.846(2)
1.315(3)
1.338(3)
1.346(3)
1.319(3)
1.380(3)
1.363(3)
90.76(9)
129.01(17)
126.1(2)
121.9(2)
119.99(2)
111.9(2)
125.7(2)
112.0(2)
128.79(17)
122.3(2)
1.8539(11)
1.8596(13)
1.3290(15)
1.3389(15)
1.3501(15)
1.3201(15)
1.3854(15)
1.3774(15)
89.36(5)
1.8497(17)
1.8388(17)
1.312(3)
1.352(2)
1.352(2)
1.316(3)
1.378(2)
1.377(2)
91.32(8)
130.24(15)
125.98(18)
120.79(18)
118.56(16)
113.22(17)
126.46(18)
113.14(17)
129.97(14)
120.40(18)
ˉ
clinic space group P1, with the asymmetric unit containing
half a nickel(II) ion, two halves of a deprotonated L3 ligand
and one molecule of 1,8-naphthalimide (Fig. 6, Table 4). The
pattern of CN bond lengths within the C2N5 core is again the
same as that previously reported for coordinated deprotonated
ligands. The aryl arms of the L3 ligand are in an anti–anti con-
formation, (as required for a DAD interaction) and the OCH3
groups on each ring are in an anti–anti conformation with
respect to their respective aryl rings.
The Naryl–H hydrogens and the Nring are involved in a DAD
triple hydrogen bond with the 1,8-naphthalimide group, which
is not coplanar to the NiN4 plane, as found in related
systems,21,31 but rather is at an angle of 41.54°. The triple
129.00(8)
126.60(11)
116.21(10)
119.15(10)
117.17(10)
126.76(10)
111.88(10)
128.88(9)
121.35(11)
2954 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 5 X-ray structures of [Ni(L1)2]·(CH3CN)2 (a) and [Ni(L2)2]·(DMSO)4, (b) with ellipsoids drawn at the 50% level. (c) Hydrogen bonding interactions joining adja-
cent [Ni(L1)2] molecules. Non-NH hydrogens and acetonitrile molecules omitted for clarity. (d) Hydrogen bonding and anagostic interactions joining adjacent [Ni-
(L1)2] molecules together via acetonitrile molecules. Aryl rings omitted for clarity. (e) Hydrogen bonding interactions joining adjacent [Ni(L2)2]·(DMSO)4 molecules.
Non-NH hydrogens omitted for clarity.
hydrogen bonding system can be described as a contiguous compounds have been crystallographically characterised, com-
2
pair of R2 (8) cyclic motifs. Adjacent [Ni(L3)2]·(1,8-naphthal- parison of experimental and calculated vibration spectra pro-
imide)2 units are held together by strong π–π stacking inter- vides a secondary way of validating calculations. Calculated FT
actions between 1,8-naphthalimide units, which are parallel to Raman and IR spectra were compared to collected experi-
each other (by symmetry) and 3.37 Å apart. These assemble mental spectra and were found to have a good correlation, with
the molecules into chains that run parallel to the crystallo- Mean Average Deviations (MAD) of 7–13 cm−1 (Fig. 7, S2–S9,
1
graphic b axis (Fig. 6b, c). It is noted that a H NMR titration Table S1†). Calculated, bond order changes with protona-
experiment between [Ni(L3)2] and 1,8-naphthalimide was tion/deprotonation are supported by observed band shifts in
carried out in d7-DMF and no evidence for association in this experimental FT-Raman and IR spectra for each pair of com-
solution was observed, perhaps due to their being outcom- pounds. For [Ni(L2)2] and [Ni(HL2)2]2+, selected vibrational
peted by the solvent molecules.
modes are shown in Fig. 8 and vibrational frequencies listed
in Table 7.
Calculations
Mode 129 of [Ni(HL2)2]2+ and mode 133 of [Ni(L2)2] are the
To provide further insight into the electronic and thermo- same type of vibration, but the frequency of the vibration is
dynamic properties of the complexes, DFT calculations have been shifted due to the change in bond order around the chelating
carried out. N–C bond lengths in the calculated structures of nitrogens, as these bonds are involved in this vibration. In the
the complexes are listed in Table 6. These data clearly show FT-Raman spectrum of [Ni(HL2)2]2+, the band attributed to
that the characteristic patterns of bond lengths found in the mode 129 is observed at 1014 cm−1, while in the FT-Raman
X-ray crystals structures for the current [Ni(HL)2]2+ and [Ni(L)2] spectrum of [Ni(L2)2], the band attributed to this vibration
complexes, and in others in the literature,12 are successfully (mode 133) is observed at 1036 cm−1. Similarly in the IR
reproduced by the calculations. Given that not all of the spectra, the bands due to modes 209 and 222 of [Ni(HL2)2]2+
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2955

View Article Online
Paper
Dalton Transactions
Table 5 Hydrogen bond lengths (Å) and angles (°)a
d(D⋯A)
∠(D–H⋯A)
HL3
N(5)–H(5)⋯N(4)
N(1)–H(1A)⋯O(4)#1
N(1)–H(1B)⋯O(4)#2
N(3)–H(3A)⋯N(2)#3
N(3)–H(3B)⋯O(4)#4
HL5
2.5940(17)
3.0438(18)
3.2441(17)
2.9501(19)
3.1944(16)
137.0(19)
144.9(14)
143.4(14)
175.0(17)
146.1(14)
N1–H1A⋯N5
2.674(5)
3.000(5)
3.027(5)
129(4)
176(4)
170(5)
N1–H1B⋯N2#5
N4–H4B⋯N3#5
[Ni(HL3)2](BF4)2·(C3H6O)2
N1–H1⋯F3
3.116(5)
3.298(8)
2.893(5)
2.957(5)
2.714(7)
140(3)
163(4)
149(4)
160(4)
162(5)
N1–H1⋯F4
N4–H4⋯F3
N3A–H3A⋯F4#6
N5–H5⋯O20
[Ni(HL2)2](ClO4)2·(CH3CN)2
N2–H2⋯O2
2.878(3)
2.806(4)
3.038(3)
171(2)
157(2)
129(2)
N4–H4⋯O1
N5–H5⋯N20
[Ni(L1)2]·(CH3CN)2
N4–H4⋯N20
3.219(3)
2.979(3)
162(2)
174(2)
N5–H5⋯N2#7
[Ni(L2)2]·(DMSO)4
N1–H1⋯O1
3.0742(17)
2.8553(19)
2.9486(17)
154.4(15)
164.0(15)
158.7(14)
N4–H4⋯O1
N5–H5⋯O2
[Ni(L3)2]·(1,8 naphthalimide)2
N4–H4⋯O6#8
2.868(2)
2.898(2)
3.015(2)
165(2)
158(2)
168(2)
N5–H5⋯O5
N6–H6B⋯N2
a Symmetry transformations: #1 x, y, z − 1; #2 −x, −y + 1, −z; #3 −x + 1,
−y + 2, −z; #4 −x + 1, −y + 2, −z + 1; #5 −x + 1, y − 1/2, −z + 1/2; #6 −x,
−y, −z + 1; #7 −x + 2, y, −z + 3/2; #8 −x + 1, −y + 1, −z.
Fig. 6 (a) X-ray structure of [Ni(L3)2]·(1,8-naphthalimide)2, with ellipsoids
drawn at the 50% level. Views down the a axis (b) and the c axis (c) of [Ni-
(L3)2]·(1,8-naphthalimide)2, showing the stacking interactions between 1,8-
naphthalimide units. Non-NH hydrogens omitted for clarity.
are observed at 1508 and 1619 cm−1 respectively, and these
correspond to modes 212 and 226 of [Ni(L2)2] which are
observed at 1564 and 1677 cm−1
,
respectively (Fig. 7).
Vibrational modes which significantly involve bonds which for each of the complexes three major transitions are found
(Table S3†). These all have lower energies than those observed
experimentally and are all predicted to be weak (oscillator
change order with protonation are expected to show a greater
change in frequency.
TD-DFT calculations allow the visualisation of the mole- strength ∼0.0001) but of similar strength, making it difficult
to assign which transition exactly corresponds to each experi-
cular orbitals for the complexes (Fig. S11–S12†). In each case
mental transition. The fact that the experimental data are col-
the HOMOs for the cationic complexes appear more diffuse
than for the neutral complexes, with appreciable wavefunction lected in DMF solution, where there is the possibility of the
complexes hydrogen bonding to solvent molecules, may go some
way to explain the discrepancies between the calculated and
amplitude across the aryl rings as well as the C2N3M rings,
whereas the neutral systems show HOMOs which are confined
to the C2N3M rings (Fig. 9). The calculations therefore experimental values. Nonetheless, the main findings from ana-
lysing these data are that: (1) the configurations that contrib-
ute significantly to the transitions all contain some HOMO
confirm that, upon deprotonation, delocalisation of charge
through the chelate ring to the metal ion occurs.
TD-DFT calculations also allow assignment of important and LUMO character (Fig. 9, S11, S12†), with all transitions
terminating on the LUMO (which is quite similar between
compounds); (2) the donor orbital set also possesses contri-
transitions in the electronic spectra of the complexes. In DMF
solution the cationic complexes were found to have a single
band at ca. 445 nm, whereas in the neutral complexes, a band bution from lower energy occupied orbitals, such as H-9, H-8
from Ni(HL1)2 – these orbitals are by and large metal-based;
at ca. 436 nm, with a lower energy shoulder at ca. 490 nm, was
found (above). In each case, extinction coefficients in the
(3) the deprotonated systems show a greater LUMO delocalisa-
range ca. 100–160 mol−1 L cm−1 were found. TD-DFT calcu- tion than the protonated counterpart; (4) one of the transitions
for each of the neutral complexes are significantly blue-shifted
with respect to the corresponding cationic complexes.
lations generally offset the absolute wavelength of transitions,
but predict the nature of the transitions reasonably well. Thus,
2956 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 6 Calculated bond lengths (experimental bond lengths in parentheses)a
Table 7 Comparison of experimental and calculated corresponding vibrational
modes for [Ni(HL2)2]2+ and [Ni(L2)2], which display the consequences of the
changes in bond order
Complex
Nring–C (Å)
Ndonor–C (Å)
Naryl–C (Å)
[Ni(HL1)2](BF4)2
[Ni(HL2)2](BF4)2
1.38595
1.38572
(1.370)b
1.39174
(1.363)
1.38449
1.34095
(1.342)
1.34187
(1.345)
1.34425
(1.352)
1.34210
1.30361
1.30364
(1.295)b
1.32608
(1.290)
1.30459
1.32095
(1.317)
1.32094
(1.325)
1.31815
(1.316)
1.32013
1.35609
1.35695
(1.350)b
1.36914
(1.358)
1.35609
1.39181
(1.372)
1.39144
(1.381)
1.39856
(1.378)
1.39277
Experimental
Calculated
Compound
Mode
frequency/cm−1
frequency/cm−1
[Ni(HL2)2]2+
[Ni(HL2)2]2+
[Ni(HL2)2]2+
[Ni(L2)2]
[Ni(L2)2]
[Ni(L2)2]
129
209
222
133
212
226
1014
1508
1619
1036
1564
1677
1006
1514
1633
1020
1547
1670
[Ni(HL3)2](BF4)2
[Ni(HL4)2](BF4)2
[Ni(L1)2]
[Ni(L2)2]
[Ni(L3)2]
[Ni(L4)2]
a Experimental values are averages of the two bond lengths found in
the X-ray structure. b Data from the X-ray structure of [Ni(HL2)2](ClO4)2.
Fig. 9 Frontier molecular orbitals for [Ni(HL2)2](BF4)2 (left) and [Ni(L2)2]
(right). Note that the HOMO for [Ni(HL2)2](BF4)2 is one of a pair of degenerate
orbitals.
Conclusions
The work reported herein shows that 1,5-diarylbiguanides are
readily prepared and effective ligands for nickel(II). Complexes
containing either neutral or deprotonated biguanide ligands
give complexes which can be characterised in solution and in
the solid state. Various techniques show that, upon deprotona-
tion, electron delocalisation within the chelate ring increases.
Vibrational spectroscopy shows shifts in the band positions of
the CN stretches, 1H NMR spectroscopy shows shifts in the
position of the peak corresponding to the Ndonor–H hydrogens,
Fig. 7 Experimental and calculated Raman spectra for [Ni(L2)2].
and X-ray crystallography shows, upon deprotonation,
a
lengthening of the C–Ndonor bonds and a lengthening of the
C–Nring bonds, along with a tightening of the C–Nring–C bond
angle, all of which are consistent with increased aromaticity
in the chelate ring. UV-Visible spectroscopy shows small shifts of
the absorption maximum upon deprotonation, consistent with
the deprotonated ligands being better donors than the neutral
ones and this is confirmed by cyclic voltammetry, which shows
a stabilisation of the Ni(III) oxidation state upon deprotona-
tion. X-ray crystal structures of the complexes allow the hydro-
gen bonding potential of these complexes to be assessed. The
results presented here show that, while the electronic natures
of the aryl ring substituents do play a subtle role in the elec-
tronic properties of the compounds, it is the conformation of
the aryl rings themselves which is important in controlling the
way in which the complex can hydrogen bond to other mole-
cules. An anti orientation of the aryl ring with respect to the
Fig. 8 Selected corresponding vibrational modes for [Ni(L2)2] and [Ni-
(HL2)2]2+
.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2957

View Article Online
Paper
Dalton Transactions
central C2N5 core enables the associated Naryl–H hydrogen to analysed using GRAMS/AI 8.00 (Thermo Electron Corporation)
act as a donor, while the associated Ndonor–H hydrogen and OriginPro 8.0 (Origin Lab Corporation). Calculations were
becomes blocked by the aryl ring and unavailable for hydrogen performed using the Gaussian09 package34 and spectra
bonding, whereas a syn orientation of the aryl ring blocks the extracted from calculations using GaussSum v2.2.5.35 Geo-
Nring hydrogen (or acceptor site in the deprotonated ligand) metry optimisation and frequency calculations utilised density
but enables hydrogen bonding of both the Naryl–H and Ndonor
–
functional theory (DFT) with a B3LYP functional, where a
H hydrogens. This work shows that these nickel(II) complexes LANL2DZ basis set was used for nickel, and 6-31g(d) was used
have much potential as tectons for hydrogen-bonded assem- for all other atoms. Single point energy calculations in vacuo
blies, and that manipulation of the aryl ring substituents does used the same parameters, while single point energy calcu-
offer promise for fine tuning the hydrogen bonding inter- lations in water required the inclusion of a solvent field, using
actions. However, it is also clear that solvents such as DMF the keyword “SCRF = (CPCM, solvent = water)”. Calculated
and DMSO readily compete for the NH donors, so current Raman and IR frequencies were scaled by 0.975, as DFT
work is focused on assemblies with more soluble organic com- methods typically slightly overestimate the energy of
ponents and also other transition metal containing com- vibrations. Calculated Raman intensities were scaled by
ponents with complementary hydrogen bonding motifs. In eqn (5).36 The “goodness of fit” between a calculation and
this way, it is hoped that assemblies with interesting chemical experiment is quantified by the mean absolute deviation
properties, as well as structural ones, might be realised.
(MAD) between bands in the calculated and experimental IR
and Raman spectra.
Syntheses
Experimental section
Methods and materials
HL3. 3,5-Dimethoxyaniline (1.45 g, 9.5 mmol) was dissolved
in 1 M HCl(aq) (10 mL). Sodium dicyanamide (0.40 g,
All reagents and metal salts were purchased commercially and 4.5 mmol) was added and the solution stirred at 100 °C for
used as received. 1H NMR spectra were recorded on a 400 MHz 18 hours. The solution was then cooled to room temperature
Varian UNITYINOVA spectrometer at 298 K, referenced to the and the resulting pale purple precipitate filtered off, washed
internal solvent signal. IR spectroscopy was carried out using a with water and dried, to give 1.3 g of crude HL3·HCl. This was
Bruker ALPHA FT-IR spectrometer with an ALPHA P ATR then suspended in dry methanol (10 mL) and a solution of
measurement module. ESI Mass Spectra were collected on a sodium methoxide (1.5 mL, 25 wt%) was added. After stirring
Bruker micro-TOF-Q spectrometer. UV/Visible spectra were at room temperature for 1 hour, water (5 mL) was added and
recorded on a Perkin Elmer lambda 950 UV/VIS/NIR spectro- the resulting off-white solid filtered off, washed with water and
meter. Microanalyses were performed at the Campbell dried in vacuo. Yield 0.92 g (55%). Mp 176–178 °C. Found:
Microanalytical Laboratory at the University of Otago. The C, 57.78; H, 6.06; N, 18.84. Calc. for C18H23N5O4: C, 57.90; H,
ligands HL1–HL5 were prepared following the general method 6.21; N, 18.75%. HRESI-MS (CH3OH) m/z 374.1818 [HL3 + H]
described by Wuest.5a HL1, HL2 and HL5 have been reported (calc. for
C18H24N5O4 374.1823 [HL3 +
H]+). 1H NMR
previously.5a Cyclic voltammetric experiments in DMF were (300 MHz, CDCl3) δ 6.29 (m, 4H, H2, H6), 6.25 (m, 2H, H4),
performed at 20 °C on solutions degassed with nitrogen. A 3.79 (s, 12H, CH3). IR νmax/cm−1 3430w, 3347w, 3083w, 2994w,
three-electrode cell was used with Cypress Systems 1.4 mm 2918w, 2835w, 1575m, 1538m, 1454m, 1419m, 1355m, 1313m,
diameter glassy carbon working, Ag/AgCl reference and plati- 1241m, 1203m, 1191m, 1150s, 1096m, 1061m, 1044m, 990m,
num wire auxiliary electrodes. The solution was ∼10−3 M in 959m, 927m, 868m, 853m, 837m, 812m, 766m, 724m, 679m,
electroactive material and contained 0.1 M [Bu4N][PF6] as the 644m, 610m.
supporting electrolyte. Voltammograms were recorded with
HL4. This was prepared as for HL3, using 4-t-butylaniline
the aid of a Powerlab/4sp computer-controlled potentiostat. (1.50 g, 10 mmol) and sodium dicyanamide (0.424 g,
Potentials are referenced to the reversible formal potential 4.8 mmol). The product was isolated as an off-white solid.
(taken as Eo = 0.00 V) for the decamethylferricenium/deca- Yield 1.581 g (90%). Mp 171–173 °C. Found: C, 72.10; H, 8.22;
methylferrocene ([Fc*]+/0) process,32 where Eo was calculated N, 19.43. Calc. for C22H31N5: C, 72.30; H, 8.55; N, 19.15%.
from the average of the oxidation and reduction peak poten- HRESI-MS (CH3OH) m/z 366.2633 [HL4 + H]+ (calc. for
1
tials under conditions of cyclic voltammetry. Under the same C22H31N5 366.2652 [HL4 + H]+). H NMR (300 MHz, CDCl3) δ
conditions, Eo calculated for [FeCp2]+/0 was 0.48 V versus 7.36 (d, J = 8.4 Hz, 4H, H2, H6), 7.05 (d, J = 8.4 Hz, 4H, H3,
[Fc*]+/0 33 FT-Raman spectra were recorded on solid-state samples H5), 3.50 (s, 1H, NH), 1.32 (s, 18H, CH3). IR νmax/cm−1 3506w,
.
at room temperature using 184 scans per sample, at 4 cm−1 spec- 3430w, 3403w, 2958m 1671m, 1598s, 1548s, 1497m, 1389s,
tral resolution and 50 mW power. The excitation wavelength 1362m, 1256s, 841m, 752m.
was 1064 nm, from a Nd:YAG laser source. A Bruker Equinox
[Ni(HL1)2](BF4)2. Ni(BF4)2·6H2O (41 mg, 0.13 mmol) was
IFS-55 interferometer was used with a FRA/106 S attachment, dissolved in acetone (2 mL) and added to a solution HL1
and was controlled using the Bruker Opus v5.5 software (100 mg, 0.27 mmol) in acetone (4 mL) to give a clear orange-
package. Scattered Raman photons were detected using a red solution. The solution was stirred for 1 hour, then concen-
liquid nitrogen cooled Ge diode (D418T). Spectral data was trated and diethyl ether was added. The resulting pale orange
2958 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
precipitate was filtered off, washed with diethyl ether and
[Ni(L2)2]. This was prepared by the method described for
dried in vacuo. Yield 93 mg (64%). Found: C, 45.74; H, 4.13; N, [Ni(L1)2]. Yield 72%. Found: C, 64.04; H, 6.54; N, 20.71. Calc.
19.12. Calc. for C28H30N10NiB2F8: C, 45.51; H, 4.09; N, 18.95%. for C36H44N10Ni: C, 64.01; H, 6.57; N, 20.73%. HRESI-MS
HRESI-MS (DMF/MeCN) m/z 563.1927 [Ni(HL1)(L1)]+ (calc. for (DMF/MeCN) m/z 675.3131 [Ni(HL2)(L2)]+ (calc. for C36H45-
1
C28H29N10Ni 563.1925 [Ni(HL1)(L1)]+). 1H NMR (300 MHz, N10Ni 675.3177 [Ni(HL2)(L2)]+). H NMR (300 MHz, DMF-d7) δ
DMF-d7) δ 9.70 (s, 2H, NH), 7.60 (s, 4H, NH), 7.39 (d, 8H, CH), 7.95 (s, 4H, NH), 6.96 (s, 8H, CH), 6.57 (s, 4H, CH), 2.15 (s,
7.33 (t, 8H, CH), 7.16 (t, 4H, CH), 7.07 (s, 4H, NH). IR νmax
/
24H, CH3). IR νmax/cm−1 3417w, 3202w, 2914w, 1601m, 1507s,
cm−1 3378w, 3348w, 3318w, 3168w, 1660s, 1593s, 1548s, 1494s, 1459s, 1279s, 1177s, 1083m, 826m, 679s 665s. UV/Vis (DMF)
1461m, 1297m, 1247s, 1122m, 992s, 758s, 693s. UV/Vis (DMF) 437 nm (108 mol−1 L cm−1).
448 nm (108 mol−1 L cm−1).
[Ni(L3)2]. This was prepared by the method described for
[Ni(HL2)2](BF4)2. This was prepared by the method [Ni(L1)2]. Yield 80%. Found: C, 52.68; H, 5.59; N, 17.06. Calc.
described for [Ni(HL1)2]. Yield 33%. Found: C, 50.66; H, 5.55; for C36H44N10O8Ni·H2O: C, 52.63; H, 5.64; N, 17.04%. HRESI-MS
N, 16.19. Calc. for C36H46N10NiB2F8: C, 50.80; H, 5.45; N, (DMF/MeCN) m/z 803.2788 [Ni(HL3)(L3)]+ (calc. for C36H45-
1
16.45%. HRESI-MS (DMF/MeCN) m/z 675.3223 [Ni(HL2)(L2)]+ N10NiO8 803.2770 [Ni(HL3)(L3)]+). H NMR (300 MHz, DMF-d7)
(calc. for C36H45N10Ni 675.3177 [Ni(HL2)(L2)]+). 1H NMR δ 8.15 (s, 4H NH), 6.57 (s, 8H, CH), 6.10 (s, 4H, CH), 3.68 (s,
(300 MHz, DMF-d7) δ 9.50 (s, 2H, NH), 7.52 (s, 4H, NH), 6.99 24H, CH3). IR νmax/cm−1 3407w, 3342w, 3199w, 2930w, 2836w,
(s, 4H, CH), 6.89 (s, 8H, CH), 6.70 (s, 4H, NH), 2.18 (s, 24H, 1592s, 1509s, 1449s, 1429s, 1285m, 1192s, 1152s, 1065m, 801m,
CH3). IR νmax/cm−1 3349m, 3325m, 3009w, 2917w, 1672s, 702m, 681m. UV/Vis (DMF) 435 nm (137 mol−1 L cm−1).
1618m, 1601m, 1567s, 1467s, 1325m, 1240m, 1144m, 1036s,
991s, 686s. UV/Vis (DMF) 446 nm (110 mol−1 L cm−1).
[Ni(L4)2]. This was prepared by the method described for
[Ni(L1)2]. Yield 45%. Found: C, 62.48; H, 7.79; N, 16.39. Calc.
[Ni(HL3)2](BF4)2. This was prepared by the method for C44H60N10Ni·3H2O: C, 62.78; H, 7.90; N, 16.63%. HRESI-MS
described for [Ni(HL1)2]. Yield 52%. Found: C, 45.89; H, 5.24; (DMF/MeCN) m/z 787.4361 [Ni(HL4)(L4)]+ (calc. for C44H61-
1
N, 13.52. Calc. for C36H46N10O8NiB2F8·2C3H6O C, 46.06; H, N10Ni 787.4429 [Ni(HL4)(L4)]+). H NMR (300 MHz, DMF-d7) δ
5.34; N, 12.78%. HRESI-MS (DMF/MeCN) m/z 803.2743 [Ni- 8.05 (s, 4H, NH), 7.29 (d, 8H, J = 8.0 Hz, CH), 7.21 (d, 8H, J =
(HL3)(L3)]+ (calc. for C36H45N10O8Ni 803.2770 [Ni(HL3)(L3)]+). 8.0 Hz, CH), 1.28 (s, 36H, CH3). IR νmax/cm−1 3605w, 3422w,
1H NMR (300 MHz, DMF-d7) δ 9.66 (s, 2H, NH), 7.62 (s, 4H, 3311w, 3050m, 2950w, 2899w, 1612m, 1583m, 1543m, 1504s,
NH), 6.61 (s, 4H, CH), 6.51 (s, 8H, CH), 6.18 (s, 4H, NH), 3.69 1461s, 1304m, 1269m, 1242m, 1195m, 1065w, 823m, 681m.
(s, 24H, CH3). IR νmax/cm−1 3330w, 2938w, 1666s, 1596s, UV/Vis (DMF) 437 nm (153 mol−1 L cm−1).
1562s, 1458s, 1207m, 1154s, 1049s, 926m, 835m, 720m, 686m.
X-ray crystallography
UV/Vis (DMF) 440 nm (116 mol−1 L cm−1).
[Ni(HL4)2](BF4)2. This was prepared by the method Crystals of HL3 and HL5 were obtained by slow evaporation of
described for [Ni(HL1)2]. Yield 42%. Found: C, 54.41; H, 6.58; methanol/water solutions. Crystals of [Ni(HL3)2](BF4)2·(C3H6O)
N, 14.18. Calc. for C44H62N10NiB2F8: C, 54.86; H, 6.49; N, were obtained by diffusion of diethyl ether into an acetone
14.53%. HRESI-MS (DMF/MeCN) m/z 787.4396 [Ni(HL4)(L4)]+ solution. Crystals of [Ni(HL2)2](ClO4)2·(CH3CN)2 were obtained
(calc. for C44H61N10Ni 787.4429 [Ni(HL4)(L4)]+). 1H NMR by diffusion of diethyl ether into an acetonitrile solution. One
(300 MHz, DMF-d7) δ 9.61 (s, 2H, NH), 7.54 (s, 4H, NH), 7.37 of the HL2 aryl rings was disordered over two coplanar posi-
(d, 8H, J = 8.8 Hz, CH), 7.31 (d, 8H, J = 8.8 Hz, CH), 7.27 (s, 4H, tions, while the perchlorate anion and the acetonitrile solvate
NH), 1.27 (36H, CH3). IR νmax/cm−1 3354m, 2959w, 1655s, were also disordered over two sites, respectively. Crystals of
1600m, 1552s, 1511s, 1462m, 1298m, 1263m, 1115m, 1016s, [Ni(L1)2]·(CH3CN)2 were obtained from slow evaporation of an
840m, 750m, 713m. UV/Vis (DMF) 449 nm (119 mol−1 L cm−1). acetonitrile solution of [Ni(HL1)2](BF4)2. Crystals of [Ni(L2)2]·
[Ni(L1)2]. Ni(BF4)2·6H2O (67 mg, 0.19 mmol) dissolved in (DMSO)4 were obtained from a dimethyl sulfoxide solution at
warm CH3OH (1 mL) was added to HL1 (100 mg, 0.39 mmol) room temperature. Crystals of [Ni(L3)2]·(1,8-naphthalimide)2
dissolved in warm CH3OH (5 mL) to give a clear, deep red solu- were obtained from a dimethyl sulfoxide solution of the two
tion. Triethylamine (55 μL, 0.39 mmol) was added via syringe reactants at room temperature. The crystal data, data collection
and the solution stirred for ca. 1 hour, resulting in the for- and refinement parameters are listed in Table 8. All measure-
mation of a deep orange precipitate. This was filtered off, ments were made with a Bruker Kappa ApexII area detector
washed with methanol and diethyl ether and dried in vacuo. using graphite monochromatised Mo Kα (λ = 0.71073 Å) radia-
Yield 106 mg (95%). Found: C, 56.34; H, 5.37; N, 22.94. Calc. tion. Intensities were corrected for Lorentz and polarisation
for C28H28N10Ni·2H2O: C, 56.12; H, 5.38; N, 23.36%. HRESI-MS effects37 and for absorption using SADABS.38 The structures
(DMF/MeCN) m/z 563.1967 [Ni(HL1)(L1)]+ (calc. for C28H29- were solved by direct methods using SHELXS39 or SIR97,40
N10Ni 563.1925 [Ni(HL1)(L1)]+). 1H NMR (300 MHz, DMF-d7) running within the WinGX package41 and refined on F2 using
δ 8.17 (s, 4H, NH), 7.35 (d, 8H, J = 7.6 Hz, CH), 7.20 (t, 8H, all data by full-matrix least-squares procedures with
J = 7.6 Hz, CH), 6.93 (t, 4H, J = 6.8 Hz, CH). IR νmax/cm−1 SHELXL-97.42 All non-hydrogen atoms were refined with aniso-
3404w, 3184w, 3036w, 2929w, 1583m, 1494s, 1449s, tropic displacement parameters. Hydrogen atoms on carbon
1275m, 1207m, 745s, 698s, 658s. UV/Vis (DMF) 436 nm atoms were included in calculated positions with isotropic dis-
(169 mol−1 L cm−1).
placement parameters. Hydrogen atoms on nitrogen atoms
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2959

Table 8 Details of X-ray crystallographic data collection
[Ni(HL3)2](BF4)2·
(C3H6O)2
[Ni(HL2)2](ClO4)2·
(CH3CN)2
[Ni(L3)2 (1,8-
naphthalimide)2
HL3
HL5
[Ni(L1)2]·CH3CN
[NiL2)2·(DMSO)4
Empirical formula
Formula weight
C18H23N5O4
373.41
C14H13Br2N5
411.11
C39H46B2F8N10NiO9
1031.19
C20H26ClN6Ni0.5O4
479.27
C32H34N12Ni
645.42
C44H68N10Ni4S4
988.03
C60H58N12NiO12
1197.89
Crystal system
Space group
Triclinic
P1
Orthorhombic
P212121
Triclinic
P1
Triclinic
P1
Monoclinic
C2/c
Triclinic
P1
Triclinic
P1
ˉ
ˉ
ˉ
ˉ
ˉ
a/Å
b/Å
c/Å
α/°
8.961(2)
6.351(5)
8.876(5)
26.832(5)
90
90
90
1512.6(15)
1.805
4
5.360
808
7.9360(9)
8.5760(10)
17.863(2)
93.507(4)
90.073(4)
111.697(4)
1127.1
1.519
1
0.529
532
8.059(5)
8.919(5)
17.985(6)
10.573(4)
18.103(6)
90
114.717(19)
90
3127.1
1.371
4
8.701(5)
7.3986(3)
14.5430(8)
14.5749(12)
116.861(4)
96.539(4)
98.148(3)
1356.50(15)
1.466
10.019(2)
10.989(2)
75.650(11)
75.016(11)
80.808(13)
918.5(4)
1.350
2
0.098
396
11.840(5)
13.424(5)
92.325(5)
104.102(5)
110.911(5)
1240.4(10)
1.323
16.632(5)
101.045(5)
101.543(5)
98.542(5)
1127.7(10)
1.411
β/°
γ/°
V/Å3
Dc/Mg m−3
Z
2
1
1
μ/mm−1
F(000)
0.613
502
0.664
1352
0.610
526
0.437
626
Dimensions/mm
θ range/°
Index ranges
0.29 × 0.21 × 0.06
2.11 to 27.55
−11 ≤ h ≤ 11
−13 ≤ k ≤ 13
−14 ≤ l ≤ 14
31 150
0.45 × 0.37 × 0.05
2.42 to 26.51
−7 ≤ h ≤ 7
−8 ≤ k ≤ 10
−32 ≤ l ≤ 33
26 016
0.50 × 0.30 × 0.10
1.14 to 24.79
−9 ≤ h ≤ 8
−10 ≤ k ≤ 9
−20 ≤ l ≤ 21
18 212
0.26 × 0.20 × 0.17
2.43 to 27.24
−9 ≤ h ≤ 10
−11 ≤ k ≤ 11
−21 ≤ l ≤ 21
38 990
0.60 × 0.16 × 0.12
2.48 to 24.94
−21 ≤ h ≤ 21
−12 ≤ k ≤ 12
−21 ≤ l ≤ 17
20 854
0.54 × 0.36 × 0.13
2.60 to 34.89
−13 ≤ h ≤ 13
−18 ≤ k ≤ 14
−19 ≤ l ≤ 21
34 673
0.58 × 0.23 × 0.20
1.64 to 25.64
−8 ≤ h ≤ 8
−17 ≤ k ≤ 17
−17 ≤ l ≤ 17
40 825
Reflections collected
Reflections unique
Rint
4201
0.0564
2639
0.0675
3771
0.0368
4999
0.469
2715
0.0686
10 423
0.0443
5095
0.0528
Completion (%)
99.2
4201/0/272
1.061
0.0658, 0.1195
0.0433, 0.1092
0.266, −0.245
95.7
2639/0/205
1.029
0.0414, 0.0635
0.0303, 0.0610
0.578, −0.495
0.005(11)
97.9
3771/0/351
1.064
0.0777, 0.1815
0.0600, 0.1633
1.298, −1.226
98.9
4999/0/390
1.070
0.0456, 0.0980
0.0367, 0.0929
0.565, −0.406
98.8
2715/0/222
1.049
0.0511, 0.0873
0.0361, 0.0806
0.361, −0.346
96.4
10 423/0/310
1.055
0.0548, 0.1109
0.0409, 0.1037
0.730, −0.636
99.6
5095/0/405
1.037
0.0482, 0.0872
0.0362, 0.0806
0.354, −0.331
Data/restraints/parameters
Goodness of fit on F2
R1, wR2 (all data)
R1, wR2 (I > 2σ(I))
Largest diff. peak and hole/e Å−3
Flack parameter

View Article Online
Dalton Transactions
Paper
were found from the difference map and refined appropriately.
7 H. Tsubouchi, K. Ohguro, K. Yasumura, H. Ishikawa and
M. Kikuchi, Bioorg. Med. Chem. Lett., 1997, 7, 1721.
8 M. Pollack, Cancer Prev. Res., 2010, 3(9), 1060.
9 N. Thakkar, V. Pirrone, S. Passic, S. Keogan, W. Zhu,
V. Kholodovych, W. Welsh, R. Rando, M. Labib, B. Wigdahl
and F. Krebs, Antimicrob. Agents Chemother., 2010, 11, 1965.
10 (a) K. M. Huttunen, A. Mannila, K. Laine, E. Kemppainen,
J. Leppanen, J. Vepsalainen, T. Jarvinen and J. Rautio,
J. Med. Chem., 2009, 52, 4142; (b) C. J. Bailey and
R. C. Turner, N. Engl. J. Med., 1996, 334, 574.
The functions minimised were Σw (Fo − Fc ), with w = [σ2(Fo )
2
2
2
+ aP2]−1, where P = [max(Fo)2 + 2Fc ]/3. Crystallographic metrics
2
were analysed using Mercury43 and figures were produced
using ORTEP44 and POV-Ray.45
Acknowledgements
Dr James Crowley and Assoc Prof. Allan Blackman are thanked
for helpful discussions, referees are thanked for helpful com- 11 P. Rây, Chem. Rev., 1961, 61, 313.
ments and the University of Otago is thanked for financial 12 R. K. Ray and G. B. Kauffman, Metal and Non-metal Bigua-
support.
nide Complexes, New Age International Publishers, New
Delhi, 1999.
13 (a) M. Zhu, L. Lu, P. Yang and X. Jin, Acta Crystallogr., Sect.
E: Struct. Rep. Online, 2002, 58, m272; (b) M. Zhu, L. Lu,
P. Yang and X. Jin, Acta Crystallogr., Sect. E: Struct. Rep.
Online, 2002, 58, m217; (c) F. Bentefrit, B. Viossat,
A. Tomas, D. Nguyen-Huy and G. Morant, Ann. Pharm. Fr.,
2002, 60, 93; (d) F. Bentefrit, G. Morgant, B. Viossat,
S. Leonce, N. Guilbaud, A. Pierre, G. Atassi and D. Nguyen-
Huy, J. Inorg. Biochem., 1997, 68, 53; (e) P. Lemione,
M. Chiadmi, V. Bissery, A. Yomas and B. Viossat, Acta Cryst-
allogr., Sect. C: Cryst. Struct. Commun., 1996, 52, 1430;
(f) B. Viossat, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 1995, 51, 878; (g) B. Viossat, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun., 1995, 51, 213.
References
1 (a) D. Fujita, A. Takahashi, S. Sato and M. Fujita, J. Am.
Chem. Soc., 2011, 133, 13317; (b) M. Wang, Y.-R. Zheng,
T. R. Cook and P. J. Stang, Inorg. Chem., 2011, 50, 6107;
(c) C. A. Otter, P. J. Patty, M. A. K. Williams,
M. R. Waterland and S. G. Telfer, Nanoscale, 2011, 3, 941;
(d) K. J. Kilpin, M. L. Gower, S. G. Telfer, G. B. Jameson and
J. D. Crowley, Inorg. Chem., 2011, 50, 1123; (e) M.-J. Lin,
A. Jouaiti, P. Grosshans, N. Kyritsakas and M. W. Hosseini,
Chem. Commun., 2011, 7635; (f) M.-C. Dul, E. Pardo,
R. Lescouezec, Y. Journaux, J. Ferrando-Soria, R. Ruiz-
Garcia, J. Cano, M. Julve, F. Lloret, D. Cangussu, 14 (a) F. Bentefrit, P. Lemione, D. Nguyen-Huy, G. Morgant
C. L. M. Pereira, H. O. Stumpf, J. Pasan and C. Ruiz-Perez,
Coord. Chem. Rev., 2010, 254, 2281; (g) M. Albrecht,
M. Fiege and O. Osetska, Coord. Chem. Rev., 2008, 252, 812;
(h) P. J. Steel, Chem. N.Z., 2003, 67, 57; (i) M. Schutte,
C. Stolle and D. G. Kurth, Supramol. Chem., 2003, 15, 549.
2 D. A. McMorran, Inorg. Chem., 2008, 47, 592.
and B. Viossat, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 2003, 59, m331; (b) F. Bentefrit, A. Tomas,
G. Morgant, D. Nguyen-Huy, P. Lemione, B. Viossat and
Z. Kristallogr, New Cryst. Struct., 2002, 217, 505;
(c) P. Spacu, C. Gheorghiu and A. Nicolaescu, Inorg. Chim.
Acta, 1968, 2, 413.
3 (a) D. Pogozhev, S. A. Baudron and M. W. Hosseini, Cryst- 15 (a) A. De, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
EngComm, 2010, 12, 2238; (b) S. G. Telfer and J. D. Wuest,
Cryst. Growth Des., 2009, 9, 1923; (c) C. Carpanese, S. Ferlay,
N. Kyritsakas, M. Henry and M. W. Hosseini, Chem.
Commun., 2009, 6786; (d) K. Adachi, Y. Sugiyama,
K. Yoneda, K. Yamada, K. Nozaki, A. Fuyuhiro and
S. Kawata, Chem.–Eur. J., 2005, 11, 6616; (e) D. Braga,
L. Brammer and N. R. Champness, CrystEngComm, 2005, 7,
1990, 46, 1004; (b) A. De, J. Crystallogr. Spectrosc. Res., 1990,
20, 279; (c) D. L. Ward, C. N. Caughlan and G. D. Smith,
Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem.,
1971, 27, 1541; (d) B. L. Holian and R. E. Marsh, Acta Cryst-
allogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1970, 26,
1049; (e) N. R. Kunchur and M. Matthew, Chem. Commun.,
1966, 86.
1; (f) S. Kitagawa and K. Uemura, Chem. Soc. Rev., 2005, 34, 16 (a) H. Kitamura, T. Ozawa, K. Jitsukawa, H. Masuda and
109; (g) A. D. Burrows, C.-W. Chan, M. M. Chowdhry,
J. E. McGrady and D. M. P. Mingos, Chem. Soc. Rev., 1995,
24, 329.
4 E. Bamberger and W. Dieckmann, Ber., 1892, 25, 543.
5 (a) O. LeBel, T. Maris, H. Duval and J. D. Wuest,
Can. J. Chem., 2005, 83, 615; (b) A. R. Katritzky, S. R. Tala
H. Einaga, Mol. Cryst. Liq. Cryst. Sci. Technol. Sect. A, 2000,
342, 69; (b) H. Kitamura, T. Ozawa, K. Jitsukawa,
H. Masuda and H. Einaga, Chem. Lett., 1999, 1225;
(c) H. Kitamura, T. Ozawa, K. Jitsukawa, H. Masuda and
H. Einaga, Mol. Cryst. Liq. Cryst. Sci. Technol. Sect. A, 1996,
285, 281.
and A. Singh, ARKIVOC, 2010, 76; (c) O. Lebel, T. Maris and 17 E. E. Simanek, X. Li, I. S. Choi and G. M. Whitesides, in
J. D. Wuest, Can. J. Chem., 2006, 84, 1426.
Comprehensive Supramolecular Chemistry, ed. J. L. Atwood,
J. E. D. Davies, D. D. MacNichol, F. Vögtle and J.-M. Lehn,
Pergamon Press, New York, 1996, ch. 17, vol. 9, pp.
595–621.
6 (a) D. Sweeney, M. L. Raymer and T. D. Lockwood, Biochem.
Pharmacol., 2003, 66, 663; (b) N. P. Jensen, A. L. Ager,
R. A. Bliss, C. J. Canfield, B. M. Kotecka, K. H. Rieckmann,
J. Terpinski and D. P. Jacobus, J. Med. Chem., 2001, 44, 18 (a) S. R. Saha, A. Bagchi and A. Das, J. Indian Chem. Soc.,
3925.
1990, 67, 234; (b) P. Spacu, C. Gheorghiu and
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 2948–2962 | 2961

View Article Online
Paper
Dalton Transactions
A. Nicolaescu, Inorg. Chim. Acta, 1968, 2, 411;
(c) S. N. Poddar, Sci. Cult., 1963, 29, 308.
membered ring motif comprising one donor and one
acceptor.
19 (a) W. Levason and M. D. Spicer, Coord. Chem. Rev., 1987, 30 (a) M. Brookhart, M. L. Gree and G. Parkin, Proc. Natl.
76, 45; (b) C. R. Saha, J. Inorg. Nucl. Chem., 1976, 38, 1635;
(c) A. K. Banerjee and S. P. Ghosh, J. Indian Chem. Soc.,
1976, 53, 719.
Acad. Sci. U. S. A., 2007, 104, 6908; (b) W. I. Sundquist,
D. P. Bancroft and S. J. Lippard, J. Am. Chem. Soc., 1990,
112, 1590.
20 P. Hubberstey and U. Suksangpanya, Struct. Bonding, 2004, 31 A. Houlton, D. M. P. Mingos and D. J. Williams, Transition
111, 33. Met. Chem., 1994, 19, 653.
21 M. M. Bishop, L. F. Lindoy, B. W. Skelton and A. H. White, 32 I. Noviandri, K. N. Brown, D. S. Fleming, P. T. Gulyas,
J. Chem. Soc., Dalton Trans., 2002, 377.
22 (a) M. M. Bishop, L. F. Lindoy, M. MaPartlin, A. Parkin,
P. A. Lay, A. F. Masters and L. Phillips, J. Phys. Chem. B,
1999, 103, 6713.
O. T. Thorn-Seshold and P. Turner, Polyhedron, 2007, 26, 33 F. Barriere and W. E. Geiger, J. Am. Chem. Soc., 2006, 128,
415; (b) M. M. Bishop, S. J. Coles, L. F. Lindoy and
A. Parkin, Inorg. Chim. Acta, 2006, 359, 3565.
23 (a) J. Tsuji, Transition Metal Reagents and Catalyst:
3980.
34 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma,
V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman,
J. V. Ortiz, J. Cioslowski and D. J. Fox, GAUSSIAN 09 (Revision
A.02), Gaussian, Inc., Wallingford, CT, 2009.
Innovations in Organic Synthesis, John Wiley
& Sons,
England, 2000; (b) L. Brandsma, S. F. Vasilevsky and
H. D. Verkruijsse, Application of Transition Metal Catalysts
in Organic Synthesis, Springer, Berlin, 1998.
24 Cobalt(III): (a) T. Taha, Y. Kushi and H. Yoneda, Bull.
Chem. Soc. Jpn., 1982, 55, 1063; (b) M. R. Snow, Acta Crys-
tallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1974, 30,
1850; (c) L. Coghi, M. Lanfranchi, G. Pelizzi and
P. Tarascono, Transition Met. Chem., 1978, 3, 69; Chro-
mium(III): (d) G. R. Brubaker and L. E. Webb, J. Am.
Chem. Soc., 1969, 91, 7199; (e) L. Coghi, M. Nardelli and
G. Pelizzi, Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem., 1976, 32, 842; Manganese(IV): (f) G. Das,
P. K. Bharadwaj, D. Ghosh, B. Chaudhuri and R. Banerjee,
Chem. Commun., 2001, 323; (g) L.-P. Lu, M.-L. Zhu and
P. Yang, Acta Crystallogr., Sect. C: Cryst. Struct. Commun.,
2004, 60, m18; (h) R. O. C. Hart, S. G. Bott, J. L. Atwood 35 N. M. O’Boyle, A. L. Tenderholt and K. M. Langner,
and S. R. Cooper, J. Chem. Soc., Chem. Commun., 1992, J. Comput. Chem., 2008, 29, 839.
894; Nickel(II) (i) Y.-L. Su, L.-P. Lu and M.-L. Zhu, Acta 36 R. Horvath and K. C. Gordon, Coord. Chem. Rev., 2010, 254,
Crystallogr., Sect. E: Struct. Rep. Online, 2005, 61, m2531; 2505.
( j) T. C. Creitz, R. Gsell and D. L. Wampler, Chem. 37 (a) SAINT V4, Area Detector Control and Integration Software,
Commun., 1969, 1371.
Siemens Analytical X-ray Systems Inc., Madison, WI, 1996;
(b) Z. Otwinowski and W. Minor, in Methods in Enzymology:
Macromolecular Crystallography Part A, ed. C. R. Carter Jr.
and R. M. Sweet, Academic Press, 1997, vol. 276, p. 307.
38 G. M. Sheldrick, SADABS, Program for Absorption Correction,
University of Göttingen, Germany, 1996.
25 (a) P. V. Babykutty, C. P. Prabhakaran, R. Anantaraman and
C. G. R. Nair, J. Inorg. Nucl. Chem., 1974, 36, 3685;
(b) R. H. Skabo and P. W. Smith, Aust. J. Chem., 1969, 22,
659.
26 A. B. P. Lever, Inorganic Electronic Spectroscopy. Elsevier,
Amsterdam, The Netherlands, 2nd edn, 1984.
27 (a) C. P. Joshua, A. George and M. Alaudeen, J. Ind. Chem.
39 G. M. Sheldrick, SHELXS and SHELXL, Institüt für Anorga-
nische Chemie der Universität, Göttingen, Germany, 1996.
Soc., 2004, 81, 357; (b) B. Dutta and D. Sen, Ind. J. Chem., A, 40 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano,
1980, 19A, 669; (c) S. Sen Gupta and D. Banerjea, J. Ind.
Chem. Soc., 1976, 53, 741; (d) D. Sen and C. Saha, J. Chem.
Soc., Dalton Trans., 1976, 776.
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115.
41 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
28 A. Adenier, M. M. Chehimi, I. Gallardo, J. Pinson and 42 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
N. Vila, Langmuir, 2004, 20, 8243. logr., 2008, A64, 112.
29 M. C. Etter, Acc. Chem. Res., 1990, 23, 120. For cyclic hydro- 43 Mercury: visualization and analysis of crystal structures;
gen bonding motifs, the designator R indicates a ring,
the number of donors (d) and acceptors (a) found in
the motif are assigned as subscripts and superscripts,
C. F. Macrae, P. R. Edgington, P. McCabe, E. Pidcock,
G. P. Shields, R. Taylor, M. Towler and J. van de Streek,
J. Appl. Crystallogr., 2006, 39, 453.
respectively, and the number of atoms in the motif is 44 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
1
indicated in parentheses. Thus R1 (6) indicates a six 45 http://www.povray.org/
2962 | Dalton Trans., 2013, 42, 2948–2962
This journal is © The Royal Society of Chemistry 2013