Influence of the R-substituents on the properties of [Ni(R 2pipdt)(dmit)] complexes and crystal structure where R = CH 2C6H5

Article abstract of DOI:10.1039/b712243h

[Ni(R₂pipdt)(dmit)], (R₂pipdt = 1,4-disubstituted- piperazine-3,2-dithione, R = CH₂C₆H₅; dmit = 1,3-dithiole-2-thione-4,5-dithiolate) (1b) has been prepared and characterised and the properties compared with those of the known complex 2b belonging to the same class where R = Prⁱ. Cyclic voltammetry of 1b and 2b was carried out and compared with that of the respective R₂pipdt ligand precursors (1a and 2a). The nature of the R-groups of the pipdt ligand exerts an effect on the redox potentials and confirmed the position of the LUMO as mainly on the R₂pipdt part of the complex. Accordingly the low frequency absorption, assigned to the HOMO-LUMO transition which has inter-ligand charge-transfer character, is found for 1b at lower frequency when compared to the corresponding transition of 2b. In situ EPR was carried out on electroreduced radical species of the R₂pipdt ligand precursors (1a, 2a) and corresponding complexes (1b, 2b). This revealed considerable delocalisation of the unpaired electron on the R₂pipdt ligand in 1b and 2b with coupling constants to N and H comparable with those of 1a and 2a. Complex 1b crystallised in the space group Pnma and shows an essentially planar complex (with out-of-plane R groups) π stacked at a distance of 3.65(1) A. Such a one-dimensional structure is not achieved in the case of 2b, where the complex units are almost parallel and head-to-tail with each other forming dimers and this difference in solid-state packing is apparent in the diffuse reflectance spectrum of each. Plane-wave DFT calculations for 1b revealed a highly one-dimensional band structure with considerable band dispersion along the direction of greatest molecular interaction via π-stacking. The Royal Society of Chemistry.

Full text of DOI:10.1039/b712243h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Influence of the R-substituents on the properties of [Ni(R2pipdt)(dmit)]
complexes and crystal structure where R = CH2C6H5†‡§
Elaine A. M. Geary,a Lesley J. Yellowlees,a Simon Parsons,a Luca Pilia,b Angela Serpe,b M. Laura Mercuri,b
Paola Deplano,*b Stewart J. Clarkc and Neil Robertson*a
Received 9th August 2007, Accepted 18th September 2007
First published as an Advance Article on the web 4th October 2007
DOI: 10.1039/b712243h
[Ni(R2pipdt)(dmit)], (R2pipdt = 1,4-disubstituted-piperazine-3,2-dithione, R = CH2C6H5; dmit =
1,3-dithiole-2-thione-4,5-dithiolate) (1b) has been prepared and characterised and the properties
compared with those of the known complex 2b belonging to the same class where R = Pri. Cyclic
voltammetry of 1b and 2b was carried out and compared with that of the respective R2pipdt ligand
precursors (1a and 2a). The nature of the R-groups of the pipdt ligand exerts an effect on the redox
potentials and confirmed the position of the LUMO as mainly on the R2pipdt part of the complex.
Accordingly the low frequency absorption, assigned to the HOMO–LUMO transition which has
inter-ligand charge-transfer character, is found for 1b at lower frequency when compared to the
corresponding transition of 2b. In situ EPR was carried out on electroreduced radical species of the
R2pipdt ligand precursors (1a, 2a) and corresponding complexes (1b, 2b). This revealed considerable
delocalisation of the unpaired electron on the R2pipdt ligand in 1b and 2b with coupling constants to N
and H comparable with those of 1a and 2a. Complex 1b crystallised in the space group Pnma and
˚
shows an essentially planar complex (with out-of-plane R groups) p stacked at a distance of 3.65(1) A.
Such a one-dimensional structure is not achieved in the case of 2b, where the complex units are almost
parallel and head-to-tail with each other forming dimers and this difference in solid-state packing is
apparent in the diffuse reflectance spectrum of each. Plane-wave DFT calculations for 1b revealed a
highly one-dimensional band structure with considerable band dispersion along the direction of
greatest molecular interaction via p-stacking.
Unsymmetrical [Ni(II)(dithione)(dithiolate)] complexes, where
Introduction
the dithione is formally a neutral donor ligand and the dithiolate
is formally a dianionic species, are of interest as NIR dyes and
NLO materials due to the intense LLCT transition that these
display.8,9,4 These have been assigned as “push-pull” complexes
where the “push” is due to the electron donating dithione ligand
and the “pull” is due to the electron withdrawing dithiolate ligand.9
In the case of [Ni(R2pipdt)(dithiolate)] complexes (R2pipdt = 1,4-
disubstituted-piperazine-3,2-dithione) (Fig. 1, with dithiolate =
dmit), this was illustrated through DFT calculations,9 wherein the
HOMO in these complexes was assigned as based mostly on the
dithiolate ligand and the LUMO based mostly on the dithione,
although both have components over the entire molecule. This
results in a highly delocalised structure with a small HOMO–
LUMO gap and therefore an intense NIR band in their absorption
spectrum. Changing the characteristics of the R groups may be
expected to change the electronic characteristics of a “push-pull”
complex, resulting in a redistribution of the electron density and
greater emphasis being placed on one of the resonance forms.
Metal dithiolenes have been intensely studied since the 1960s for
their novel properties that lead to many applications in the fields
of conducting and magnetic materials, dyes, non-linear optics and
catalysis.1,2,3 Metal dithiolene complexes possess highly delocalised
frontier orbitals and this results in properties including redox
activity with one or more reversible redox processes (depending
on the metal), low energy absorption in the visible/NIR region
and strong intermolecular interactions through the S atoms in the
delocalised core. The electronic and photophysical properties of
dithiolene complexes can be tuned by variation of the attached
ligand set4 and of particular interest, the ligand 1,3-dithiole-2-
thione-4,5-dithiolate (dmit)5 has been widely used in the field of
conducting materials.6,7
aSchool of Chemistry, University of Edinburgh, King’s Buildings, West Mains
Road, Edinburgh, UK EH9 3JJ. E-mail: neil.robertson@ed.ac.uk; Fax: +44
131 6504743; Tel: +44 131 6504755
bDipartimento di Chimica Inorganica e Analitica, S. S. 554, Bivio per Sestu,
09042 Monserrato, Cagliari, Italy. E-mail: deplano@unica.it, mercuri@
unica.it; Fax: +39 070 6754456; Tel: +39 070 6754680
cDepartment of Physics, University of Durham, South Road, Durham, UK
DH1 3LE
† The HTML version of this article has been enhanced with colour images.
‡ CCDC reference numbers 656892. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b712243h
§ Electronic supplementary information (ESI) available: UV/Vis and EPR
spectra of compounds 1a−, 1b−, 2a− and 2b−; Fig. S1–S6 and Table S1. See
DOI: 10.1039/b712243h
Fig. 1 Structure of complexes studied in this work. R = CH2C6H5 (Bz),
1b; R = Pri, 2b.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5453–5459 | 5453
©

View Article Online
Studies of the [Ni(R2pipdt)(dithiolate)] series of complexes
to date have focussed largely on their optical properties in the
context of NLO and NIR dyes. Ni-bis(dithiolene) complexes,
however, often show attractive superconducting, metallic and
photoconducting properties,6,7,10 that depend crucially on the
solid-state packing of the complexes as well as the intrinsic
molecular characteristics. We present in this paper spectroscopic,
structural and computational results that compare the influence
of the R-substituents of the R2pipdt ligand on the solid-state as
well as the molecular properties of [Ni(R2pipdt)(dmit)], 1b (R =
CH2C6H5), and 2b (R = Pri).
˚
the dmit (1.349(3) A) corresponding to the electron poor nature of
the pipdt and electron rich dmit ligand. The crystal lattice shows
the complex forms stacks (Fig. 3) with an interplanar distance of
˚
3.65(1) A, which is typical of p-stacked metal dithiolene complexes.
In contrast, such a one-dimensional structure is not observed in
the case of 2b. In this case the complex units are almost parallel
and head-to-tail with each other forming dimers through short
contacts between the S-atoms of dmit and the two carbon atoms
i
9
˚
of the C2S2 moiety of Pr2 pipdt (3.42 and 3.34 A) (Fig. 4).
Results and discussion
The novel mixed dithiolene complex 1b (Fig. 1) was prepared by an
analogous method to [Ni(Pri2pipdt)(dmit)] 2b which, despite the
interesting nature of these complexes and the general importance
of the dmit ligand, is the only previously structurally characterised
example of the series shown in Fig. 1.4,9 (Other examples, not
structurally characterised, have been reported previously in a PhD
thesis.) Synthesis involved reaction of [TBA]2[Ni(dmit)2] with the
dicationic symmetric Ni dithione molecules, [Ni(R2pipdt)2][BF4]2.
Crystals of 1b were grown by slow diffusion of diethyl ether into
a saturated solution of 1b in DMF. For 1b the central Ni atom
is coordinated by four S atoms in a square planar environment
˚
with Ni–S bonds ranging from 2.1477(5)–2.1628(4) A (Fig. 2).
The molecule is essentially planar except for the Bz substituents
at the R2pipdt N and the torsion angle of only 5.88◦ around S12,
C22, N32, C62 suggests that the Npp-lone pair is almost parallel
to the C2S2 p-system as previously reported for 2b. The Ni–S
bond lengths are similar in value to those previously published
for other “push-pull” coordination complexes9 (and references
therein) which suggests that the Ni–S r bonds are more or less
equivalent irrespective of the nature of the ligand and that the Ni–
S p interactions have only minor importance or none at all.9 The
Fig. 3 View along the b-axis showing packing motif of 1b.
˚
C–S bond lengths are shorter (ca. 0.04 A) in the R2pipdt ligand
than in dmit (since it has p donors adjacent to the C2S2 core). The
˚
C–C bond is significantly longer in the R2pipdt (1.477(3) A) than
Fig. 4 X-Ray structure showing head-to-tail dimer packing for 2b.9
The electrochemistry of 1b and 2b, along with the corresponding
R2pipdt ligand precursors (1a and 2a) was studied by cyclic
voltammetry in a solution of 0.1 M TBABF4 in DMF at
293 K. Redox potentials for the processes are listed in Table 1. Each
of the two pipdt ligand precursors, 1a and 2a shows one reduction
process within the solvent window (Fig. 5), chemically reversible
for 1a and quasireversible for 2a. Reduction of the ligands is
noticeably easier for 1a than 2a, indicating that the substituent
on the N of the pipdt ligand influences the reduction potential
of the ligand. Both ligands show one chemically irreversible
Table 1 Redox potentials, E1/2, of the (R2pipdt) ligand precursors and
[Ni(R2pipdt)(dmit)] complexes in DMF solution. All processes are chem-
ically reversible unless otherwise indicated: *irreversible, §quasi reversible
Fig. 2 X-Ray crystal structure of 1b showin◦ g the atom labeling
E1/V
E2/V
E3/V
˚
scheme. Selected bond lengths (A) and angles ( ): Ni–S(12) 2.1477(5),
1a
1b
2a
2b
Bz2pipdt
—
−1.24
−0.41
−1.40§
−0.49
1.13∗
Ni–S(11) 2.1627(4), S(11)–C(21) 1.7306(16), S(12)–C(22) 1.6908(16),
C(22)–C(22ꢀ) 1.477(3), C(21)–C(21ꢀ) 1.349(3), S(12)–Ni–S(11) 87.544(16),
S(11)–Ni–S(11ꢀ) 94.07(2), S(11ꢀ)–Ni–S(12ꢀ) 87.541(16), S(12ꢀ)–Ni–S(12)
90.84(3).
[Ni(Bz2pipdt)(dmit)]
−0.91§
0.69∗
1.01∗
0.65∗
Pri2pipdt
—
[Ni(Pri2pipdt)(dmit)]
−1.04
5454 | Dalton Trans., 2007, 5453–5459
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
[Ni(dmit)2]1−. Complex 1b shows a peak at 1430 cm−1, in agreement
with the assignement of a −2 formal charge to the dmit ligand.
These results are indicative of “pull” character (dithiolate) for
the dmit ligand and hence “push” character (dithione) for the
Bz2pipdt ligand, with significant separation of charges in the
resulting complex.4,9
The UV/Vis spectra of compounds 1b and 2b (Fig. 7, Table 2)
show an intense peak in the UV, a visible peak near 490 nm
and two peaks which partially overlap in the 700–850 nm region.
Table 2 shows the kmax values and molar extinction coefficients.
According to previous assignments, the lowest energy band is
due to a HOMO–LUMO transition. This transition has a certain
LLCT character due to the unbalanced contribution to the
frontier p-orbitals of the two ligands: a greater weight of the
R2pipdt to the LUMO (dithione) and a greater weight of dmit
to the HOMO (dithiolate) and its nature is consistent with the
predicted and observed negative solvatochromism and molecular
first hyperpolarisability.4,8,9 The value of kmax follows the order
1b > 2b (858 nm vs. 829 nm) indicating that the HOMO–
LUMO gap is smaller for 1b in agreement with electrochemical
results which showed that E3–E2 is smaller for 1b (Table 1)
and in keeping with the more electron rich Pri groups raising
the energy of the LUMO more than that of the HOMO. An
inspection of the frontier orbitals, based on approximate Extended
Hu¨ckel methods,11 allows assignment of the peak (shoulder) which
appears near 700 nm. This absorption, which is particular to the
dmit class of the [Ni(R2pipdt)(dithiolate)] family, may be assigned
to the HOMO → LUMO + 5 transition which involves p orbitals
with a predominant contribution of the dmit ligand (see Fig. 8).
Note that in this complex due to the presence of four additional
closely-spaced p-orbitals of the benzyl substituents the LUMO +
5 orbital is the next highest orbital based on the dmit moiety.
Fig. 5 Cyclic voltammogram of the reduction (left) and oxidation (right)
of 1a (solid line) and 2a (broken line) in a solution of 0.1 M TBABF4 in
DMF at 293 K.
oxidation within the solvent window and the ease of oxidation
is approximately converse to that of the ease of reduction of
the ligands, with oxidation easier for 2a than 1a (Fig. 5). After
coordination of each ligand to a metal centre the reduction
potentials of the resulting complexes is shifted to a less negative
potential (Fig. 6). Both complexes 1b and 2b show two chemically
reversible or quasireversible reductions within the solvent window.
The electrochemistry of 2b shows similar results to the previously
reported data in acetonitrile.9 The ease of reduction for the
complexes follows the order 1b > 2b, analogous to the ligand
precursors, and the reduction potentials of the [Ni(R2pipdt)(dmit)]
complexes are again shown to be dependant on the R substituent
on the pipdt ligand. This can be seen more clearly in the second
reduction of the complexes, where the second reduction potential
of 2b is significantly more negative than that of 1b.
Fig. 6 Cyclic voltammogram of the reductions (left) and oxidation (right)
of 1b (solid line) and 2b (broken line) in a solution of 0.1 M TBABF4 in
DMF at 293 K.
Both complexes 1b and 2b show one chemically irreversible
oxidation process, and 2b is easier to oxidise (Fig. 6). Previous
DFT calculations on [Ni(H2pipdt)(dithiolate)]9 and consideration
of the bond lengths in 1b and 2b9 suggest that although the frontier
orbitals are spread over the entire molecule, the LUMO has a
greater component on the R2pipdt ligand and the HOMO has a
greater component on the dithiolate ligand (dmit). Our results with
varying R indicate a greater variation in the reduction potential
(0.13 V for E1, 0.08 V for E2) than the variation in the oxidation
potential (0.04 V for E3) and this adds experimental evidence to
support the calculated locations of the frontier orbitals. In all the
electrochemical results 1a/1b is more electron rich than 2a/2b and
it might be envisaged that even larger changes could be engineered
by using a wider variety of R groups.
Fig. 7 UV/Vis/NIR spectra of 1b (solid red line) and 2b (broken black
line) in DMF solution at 293 K.
To gain insight into the influence of the solid-state structure on
the electronic characteristics, the diffuse reflectance spectra of 1b
and 2b were recorded (Table 2, Fig. 9). At higher energies, the
bands observed corresponded to equivalent absorptions for the
solution species and the principal differences from the solution
spectra occurred for the lower energy bands. For both 1b and
2b, the band arising from the HOMO–LUMO transition was
shifted to kmax = 960 nm in the solid state, indicative of the
interaction with neighbouring molecules. For 2b, however, an
additional shoulder at 1450 nm was evident. The presence of this
Raman spectra have previously been collected to provide
information on the electron distribution in related dithiolene
9
=
complexes. Typically, the C C stretch shows a shift to higher
frequency as the negative charge on the complex increases: this
peak is found at 1435 cm−1 in [Ni(dmit)2]2− and at 1390 cm−1 in
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5453–5459 | 5455
©

View Article Online
Fig. 8 Molecular frontier orbitals of 1b.
Fig. 9 The diffuse reflectance spectra of [Ni(Bz2pipdt)(dmit)] (dotted line)
and [Ni(Pripipdt)(dmit)] (continuous line). The absorbances are reported
in arbitrary units.
additional band presumably arises from a new absorption within
the dimer units due to the particularly strong dimerisation present
in 2b (Fig. 4) with short S · · · C interactions between molecules
˚
of 3.42 and 3.34 A, compared with the regular stacks of weaker
˚
interacting molecules for 1b (interplanar separation of 3.65 A).
Thus, the diffuse reflectance spectra reflect the differences in solid-
state electronic characteristics that arise from the different packing
of 1b and 2b.
UV/Vis/NIR spectroelectrochemistry was performed on lig-
ands 1a and 2a (Table S1, Fig. S1 and S2§) for comparison
with complexes 1b and 2b, but poor solubility of the latter upon
reduction at the low temperatures required for chemical stability
precluded measurement of these so no further interpretations
could be made. In situ EPR spectroelectrochemistry was per-
formed on 1a, 1b, 2a and 2b (Fig. 10 and Fig. S3–S6§). The
distribution of the unpaired electron density on the reduced species
was studied by analysing the couplings to compare the location
of the first reduction electron in 1b and 2b to its location on the
corresponding uncoordinated ligand precursor, 1a and 2a. Each
species was reduced in a solution of 0.1 M TBABF4 in DMF at
233 K to give an EPR active solution and at a higher temperature
(293 K) the spectra were more resolved (Tables 3 and 4). Simulation
of the spectra in terms of the g-factor, coupling constants and line
width was achieved for each. In every case, on introduction of a
second electron to form the dianionic species the signal collapsed
5456 | Dalton Trans., 2007, 5453–5459
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 3 EPR coupling constants for 1a− and 1b− in 0.1 M TBABF4–
DMF at 293 K. All hyperfine coupling constants are given in G. D =
linewidth
Due to the extended intermolecular interactions via p-stacking
shown by 1b in the X-ray structure, we sought to further explore
the bulk material properties of this solid. Accordingly plane-wave
DFT calculations were carried out to investigate the molecular
overlap and resulting band structure. The result obtained for
the distribution of the frontier orbitals on individual molecules
was consistent with those of previous calculations on isolated
molecules using localised basis functions.9 The band structure
is shown along the principle reciprocal-space axes (Fig. 11) (for
a tetragonal unit cell, these are parallel to the direction of the
corresponding real axes).
Parameter
1a−
1b−
A (H × 4)
A (H × 4)
A (N × 2)
D
2.211
2.211
2.82
3.68
2.211
2.722
1.4
0.8
g
1.9987
2.0013
Table 4 EPR coupling constants for 2a− and 2b− in 0.1 M TBABF4–
DMF at 233 K. All hyperfine coupling constants are given in G. D =
linewidth
Parameter
2a−
2b−
a (H × 2)
a (H × 4)
a (N × 2)
D
0.619
3.05
2.392
0.5
0.619
4.206
4.206
3.0
g
1.9982
1.9988
Fig. 11 Band structure calculation for 1b through plane-wave DFT
calculations showing bands around the Fermi level (dotted line).
There are four molecules per unit cell and this is reflected in the
presence of four HOMO bands around 1.5–2 eV and four LUMO
bands around 2.5 eV. For such DFT calculations, the gap between
filled and vacant orbitals is typically underestimated and this is
the case here, as shown by the diffuse reflectance spectrum for
1b where the low-energy tail extending to around 1 eV gives an
experimental estimate of the gap.
Relevant to the potential conductivity properties, we observe
significant band dispersion along the (0, 0, 0.5) direction which
corresponds to the direction of greatest molecular overlap as
shown in the X-ray structure (Fig. 3, c-axis direction). The band
dispersion of around 0.25 eV for both HOMO and LUMO
bands is comparable to that observed in many [Ni(dmit)2]x− salts12
which suggests that [Ni(R2pipdt)(dmit)] complexes, when the R
substituents have the suitable steric and electronic features (e.g.
1b), may be able to show electro-conductivity when suitably doped.
Almost no intermolecular interaction is determined for 1b along
the (0, 0.5, 0) direction, however, consistent with the separation
of the molecular units caused by the benzyl groups along the
b-direction of the crystal. Thus, 1b forms an essentially one-
dimensional material in contrast with the typical two-dimensional
characteristics of many [Ni(dmit)2]x− salts.
Fig. 10 EPR spectrum of 1a1− in solution of 0.1 M TBABF4 in DMF at
293 K. Egen = −1.70 V.
as the second electron enters the same orbital as the first to give a
diamagnetic ion.
In each case for
a
ligand, and its corresponding
[Ni(R2pipdt)(dmit)] complex, it can be seen that the reduction
electron couples to the ring N atoms, ring H atoms and the H
atoms on the N-substituent, illustrating the considerable electron
delocalisation in all cases. The comparable magnitude of the
coupling constants in the free ligands and in the complexed ligands
strongly suggest that for the [Ni(R2pipdt)(dmit)] complexes (1b
and 2b) the reduction electron locates itself primarily on the
dithione ligand. Indeed, in the case of 2b−, coupling to the R2pipdt
N-atoms and H-atoms is increased in comparison with 2a−. In
addition, the g-factor in the complexes 1b and 2b are very close
to the free electron value typical of organic radicals, suggesting
that the LUMO has little contribution from the Ni atom and
is largely ligand based. The results of this EPR study therefore
support assignment of the LUMO of these complexes as highly
delocalised and R2pipdt ligand based, consistent with the findings
of the previous theoretical studies.4,9
Conclusions
The work reported has provided the first insight into the role of the
R-groups in controlling the molecular and solid-state electronic
and structural properties of the complexes [Ni(R2pipdt)(dmit)].
Electrochemical, spectroscopic and spectroelectrochemical evi-
dence for the isolated molecules has confirmed that the LUMO is
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5453–5459 | 5457
©

View Article Online
primarily based on the R2pipdt ligand and that the LUMO energy
(and to a lesser extent the HOMO energy) can be varied by the
choice of suitable R. Structural, spectroscopic and computational
work has shown the influence of the R-groups on the solid state
structure and properties of the materials. Variation of the R-group
from Pri (2b) to benzyl (1b) leads to a uniform one-dimensional
stack rather than the strongly dimerised motif of the former. This
is reflected in the diffuse reflectance spectra of the complexes
and in the similar band dispersion for 1b to that observed for
example in [Ni(dmit)2]. This suggests potential for these complexes
as conducting materials when suitably doped. Future work will
include study of the conductivity properties of 1b in crystalline
and thin film form when doped chemically or in a field-effect
transistor.
are described using ultrasoft pseudo potentials. Integrations over
the Brillouin zone are performed on a grid which also converges
the total energy of the system to 1 meV atom−1. Geometry
optimizations are performed by relaxing the positions and unit
cell parameters under the influence of the Hellmann–Feynman
forces and stresses respectively.
All chemicals were used as supplied by Sigma-Aldrich.
[TBA]2[Ni(dmit)2] was synthesised by a previously published
method.5
Bz2pipdt (1,4-dibenzyl-piperazine-2,3-dithione) (1a) was syn-
thesised as previously reported.15 The synthesis departed from that
of the original synthesis for the conversion of the diketone product
to the dithione. Lawessons reagent was used for this conversion
as previously described for a related ligand.16 Yield 51.0%. MS
(FABMS) m/z: 327 (M+). Anal calcd for C18H18N2S2: C, 66.2; H,
5.6; N, 8.6. Found C, 66.7; H, 6.1; N, 9.3.
Experimental
[Ni(Bz2pipdt)2][BF4]2 was synthesised by a similar method to
that previously reported for Pt analogues of the complex.17
NiCl2·6(H2O) (0.2 g, 0.84 mmol) was dissolved in ca. 50 mL EtOH
and 1a (0.55 g, 1.68 mmol) was dissolved in 50 mL DCM and the
two solutions were added together and stirred at RT for 30 min.
The solvent was removed and the crude product was dissolved in
EtOH before filtering to remove excess ligand. NaBF4 (0.184 g,
1.68 mmol) was added as a solid to the solution and the mixture
was stirred at RT until a solid product precipitated. The product
was recrystallised by dropwise addition of diethyl ether to a hot
solution of the product in MeCN. Yield 57.4% MS (FABMS) m/z:
797 ([{Ni(Bz2pipdt)2}(BF4)]+). Anal calcd for C36H36N4S1b2F8Ni:
C, 48.8; H, 4.1; N, 6.3. Found C, 48.6; H, 4.1; N, 6.2.
Electrochemical studies were carried out using a DELL GX110
PC with General Purpose Electrochemical System (GPES), ver-
sion 4.8 software connected to an autolab system containing a
PGSTAT 20 potentiostat. The techniques used a three electrode
configuration, with a 0.5 mm diameter Pt disc working electrode,
a Pt rod counter electrode and an Ag/AgCl (saturated KCl) ref-
erence electrode against which the ferrocenium/ferrocene couple
was measured to be +0.55 V. The supporting electrolyte was 0.1 M
tetrabutylammonium tetrafluoroborate (TBABF4).
In situ EPR spectra were recorded on an X-Band Bruker
ER200D-SCR spectrometer, connected to PC running EPR
Acquisition System, version 2.42 software. Species were electro-
generated using a BAS CV-27 Voltammograph and temperature
controlled using a Bruker ER111VT unit.
[Ni(Bz2pipdt)(dmit)] (1b) was synthesised by modifying the
previously reported method for 2b.4 Yield 92.0% MS (FABMS)
m/z: 581 (M+). Anal calcd for C21H18N2S7Ni: C, 43.4; H, 3.1; N,
4.8. Found C, 43.6; H, 1.6; N, 4.7.
Diffuse reflectance spectra (2000–300 nm) were recorded on
KBr pellets by using a Cary 5 spectrophotometer equipped with a
diffuse reflectance accessory.
[Ni(Pripipdt)2][BF4]2
was
synthesised
as
described
Raman spectra were taken at room temperature on a single crys-
tal by using a Raman microscope (BX 40, Olimpus) spectromete◦r
(ISA xy 800) equipped with an Ar+ laser (k = 514.15 nm). A 180
reflective geometry was adopted. The samples were mounted on
a glass microscope slide and the scattering peaks were calibrated
against a Si standard (k = 520 cm−1). A typical spectrum was
collected with a 300 s time constant at a 1 cm−1 resolution and was
averaged over 2 scans. No sample decomposition was observed
during the experiments.
For the UV/Vis spectroelectrochemistry study, the quartz cell
used was 0.5 mm thick, the working electrode a Pt/Rh gauze, the
counter electrode a Pt wire and the reference electrode Ag/AgCl.
A Perkin-Elmer Lambda 9 spectrophotometer, linked to a PC
running UV/Winlab software was used to record the spectra. In
every case after recording the final spectrum the potential was
adjusted so that the neutral starting material was regenerated and
each absorption spectrum was observed to return exactly to that
of the starting species, thus the monoreduced species 1a− and 2a−
are all stable at 213 K.
previously.4 MS (FABMS) m/z: 259 ([Ni(Pripipdt)2]2+),
605 ([{Ni(Pripipdt)2}(BF4)]+). Yield 66.0%. Anal calcd for
C20H36N4S4B2F8Ni: C, 34.7; H, 5.2; N, 8.1. Found C, 34.5; H, 5.2;
N, 7.7.
[Ni(Pripipdt)(dmit)] (2b) was synthesised according to the
literature method.4 Yield 71.4%. MS (FABMS) m/z: 484 (M+).
Anal calcd for C13H18N2S7Ni: C, 32.2; H, 3.7; N, 5.8. Found C,
32.3; H, 3.65; N, 5.3.
X-Ray crystallography:‡ Green lath-like needles of 1b (dimen-
sions 0.58 × 0.19 × 0.07 mm3) were grown by slow diffusion of
diethyl ether into a saturated solution of 1b in DMF. Single crystal
X-ray diffraction data were collected using Mo-Ka radiation on
a Smart APEX CCD diffractometer equipped with an Oxford
Cryosystems low-temperature device operating at 150 K. An
absorption correction was applied using the multi-scan procedure
SADABS.18 The structure was solved by Patterson methods
(DIRDIF)19 and refined by full-matrix least squares against |F|2
using all data (SHELXL-97).20 Figures were prepared using the
programme Mercury.21 The molecule lies with its long axis in
a crystallographic mirror plane. C42/C52 are disordered about
the mirror, with occupancies both equal to 0.5. N–C and C–C
DFT plane wave calculations were performed using the density
functional formalism within the generalized gradient approxima-
tion using the CASTEP code.13,14 The electronic wavefunctions are
expanded in a plane wave basis set up to a kinetic energy cut off
of 380 eV which converges the total energy of the system to better
than 1 meV atom−1. The valence electron and ion interactions
˚
were lightly restrained to 1.45 and 1.52 A. The part-weight atoms
were refined with isotropic displacement parameters, all other
non-H atoms were refined with anisotropic displacement param-
eters. H-atoms were placed in idealized positions. C21H18N2S7Ni,
5458 | Dalton Trans., 2007, 5453–5459
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
T/K = 150, space group Pnma, a = 19.1586(6), b = 19.5819(5),
7 P. Cassoux, L. Valade, H. Kobayashi, A. Kobayashi, R. A. Clark and
A. E. Underhill, Coord. Chem. Rev., 1991, 110, 115; P. Cassoux, Coord.
Chem. Rev., 1999, 186, 213.
8 F. Bigoli, P. Cassoux, P. Deplano, M. L. Mercuri, M. A. Pellinghelli, G.
Pintus, A. Serpe and E. F. Trogu, J. Chem. Soc., Dalton Trans., 2000,
4639.
9 S. Curreli, P. Deplano, C. Faulmann, A. Ienco, C. Mealli, M. L.
Mercuri, L. Pilia, G. Pintus, A. Serpe and E. F. Trogu, Inorg. Chem.,
2004, 43, 5069.
3
˚
˚
c = 6.1048(2) A, V = 2290.29(12) A , no. reflections for cell =
7912, hmax(◦) = 28.75, Z = 4, Dc = 1.686 Mg m−3, l = 1.500 mm−1,
reflections collected = 25 384, unique [Rint] = 2941 [0.0376], no.
I > 2r = 2487, Tmin/Tmax = 0.684/0.900, parameters = 144, R1
−3
˚
[F > 4r(F)] = 0.0298, wR = 0.0758, S = 1.046, Dqmax/e A
=
−3
˚
0.50, Dqmin/e A = −0.55.
10 M. C. Aragoni, M. Arca, M. Caironi, C. Denotti, F. A. Devillanova,
E. Grigiotti, F. Isaia, F. Laschi, V. Lippolis, D. Natali, L. Pala, M.
Sampietro and P. Zanello, Chem. Commun., 2004, 1882.
11 C. Mealli and D. Proserpio, J. Chem. Educ., 1990, 67, 39.
12 E. Canadell, Coord. Chem. Rev., 1999, 185–186, 629.
13 M. D. Segall, P. J. D. Lindan, M. J. Probert, C. J. Pickard, P. J. Hasnip,
S. J. Clark and M. C. Payne, J. Phys.: Condens. Matter, 2002, 14, 2717.
14 S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. J. Probert, K.
Refson and M. C. Payne, Z. Kristallogr., 2005, 220, 567.
15 R. Isaksson, T. Liljefors and J. Sandstroem, J. Chem. Res. (S), 1981,
43.
Acknowledgements
For financial support we thank the University of Edinburgh
Christina Miller fund, and the British Council for UK–Italy
collaborative support, the MIUR, through the PRIN project
2005033820_002 “Molecular Materials with Magnetic, Optical
and Electrical Properties” and the COST D-35 working group
on Multifunctional and Switchable Molecular Materials.
16 M. A. Pellinghelli, J. M. Williams, E. F. Trogu, F. Bigoli, P. Deplano,
F. A. Devillanova, J. R. Ferraro, V. Lippolis, P. J. Lukes and M. L.
Mercuri, Inorg. Chem., 1997, 36, 1218.
17 F. Bigoli, P. Deplano, M. L. Mercuri, M. A. Pellinghelli, L. Pilia, G.
Pintus, A. Serpe and E. F. Trogu, Inorg. Chem., 2002, 41, 5241.
18 G. M. Sheldrick, SADABS, Version 2004/1, University of Go¨ttingen,
Germany.
19 P. T. Beurskens, G. Beurskens, R. de Gelder, S. Garcia-Granda, R. O.
Gould, R. Israel and Jan M. M. Smits, The DIRDIF-99 program system,
Crystallography Laboratory, University of Nijmegen, The Netherlands,
1998.
References
1 U. T. Mueller-Westerhoff and B. Vance, Comp. Coord. Chem., 1987,
595.
2 N. Robertson and L. Cronin, Coord. Chem. Rev., 2002, 227, 93.
3 ed. E. I. Stiefel, Dithiolene Chemistry, Synthesis, Properties and
Applications, Prog. Inorg. Chem., 2004, 52, 1–738.
4 F. Bigoli, C.-T. Chen, W.-C. Wu, P. Deplano, M. L. Mercuri, M. A.
Pellinghelli, L. Pilia, G. Pintus, A. Serpe and E. F. Trogu, Chem.
Commun., 2001, 2246.
20 G. M. Sheldrick, SHELXL97, University of Go¨ttingen, Germany,
2007.
5 G. Steimecke, H. J. Sieler, R. Kirmse and E. Hoyer, Phosphorus, Sulfur
Relat. Elem., 1979, 7, 49.
6 T. Akutagawa and T. Nakamura, Coord. Chem. Rev., 2000, 198, 297.
21 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, 453–457.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5453–5459 | 5459
©