Synthesis, X-ray crystal structures, optical properties and modelling data of neutral bis(1,2-dithiolene) nickel complexes of the non-cyclic SR family

Article abstract of DOI:10.1039/c2nj40398f

Three nickel-bisdithiolene-based compounds were synthesized and characterized by X-ray single crystal (for two of them), electrochemical and spectroscopic analyses. A minor change in the alkyl chain structure surrounding the nickel-bisdithiolene core induces dramatic changes in molecular packing: complex 1 crystallizes in a triclinic (P1) space group while complex 3 crystallizes in a monoclinic (C2/c) space group. No such differences are to be noted concerning electrochemical or spectroscopic characteristics, the alkyl chains induce no influence on electronic parameters. Furthermore, theoretical calculations were conducted to obtain theoretical insight of the behaviour of the complexes. The three complexes were investigated by DFT calculations using the PBE0 functional.

Full text of DOI:10.1039/c2nj40398f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 2033–2041
www.rsc.org/njc
PAPER
Synthesis, X-ray crystal structures, optical properties and modelling
data of neutral bis(1,2-dithiolene) nickel complexes of the
‘‘non-cyclic SR’’ familyw
ab
Thanh-Tuan Bui,ab Minh Ha Vuong,ab Benedicte Garreau-de Bonneval,
Fabienne Alary,c Jacob Kane,ab Carine Duhayon,ab Alix Sournia-Saquetab and
Kathleen I. Moineau-Chane-Ching*ab
´
´
Received (in Victoria, Australia) 16th May 2012, Accepted 13th July 2012
DOI: 10.1039/c2nj40398f
Three nickel-bisdithiolene-based compounds were synthesized and characterized by X-ray single
crystal (for two of them), electrochemical and spectroscopic analyses. A minor change in the alkyl
chain structure surrounding the nickel-bisdithiolene core induces dramatic changes in molecular
%
packing: complex 1 crystallizes in a triclinic (P1) space group while complex 3 crystallizes in a
monoclinic (C2/c) space group. No such differences are to be noted concerning electrochemical
or spectroscopic characteristics, the alkyl chains induce no influence on electronic parameters.
Furthermore, theoretical calculations were conducted to obtain theoretical insight of the
behaviour of the complexes. The three complexes were investigated by DFT calculations
using the PBE0 functional.
Introduction
to different ways of packing in the condensed state – crystalline or
liquid crystalline phases – depending on the nature and position of
functional groups grafted on the molecule periphery.8
During the past decade, electron accepting small molecules
have attracted considerable attention due to a number of
promising applications in organic electronics: organic field-
effect transistors,1 light emitting diodes,2 and sensors.3 In the
field of organic photovoltaics (OPV), different classes of
electron accepting small molecules have been explored. The
fullerene derivatives, and especially [6,6]-phenyl-C61-butyric
acid methyl ester (PCBM), are the most studied ones.4 In
general, a drawback of employing PCBM for organic photo-
voltaics is that it absorbs poorly in the visible part of the solar
spectrum.5 Other organic small molecules have been emerging
recently as an attractive alternative to fullerenes due to the
ease of tuning their band structure by chemical modifications:
perylenes, vinazenes, diketopyrrolopyrroles and benzothiadia-
zoles gave rise to promising results.6 Previously, we have
reported on tetra-phenyl substituted nickel bisdithiolene
complexes.7 They are efficient electron acceptors and give rise
Regarding the other known electron accepting small mole-
cules, neutral square planar nickel organic complexes constitute
a promising class of molecules for OPV applications: (i) they
absorb strongly in the vis-NIR region of the solar spectrum,
(ii) they present a great stability towards oxygen, water or
relatively high temperatures (at least up to 350 1C), (iii) their
neutral character is essential for such applications. Moreover,
some of them exhibit a p-stacked columnar structure with a
remarkably high electron mobility of 2.8 cm2 Vꢀ1.9 All these
specificities confirm their interest for organic electronics. The
key step in the development of such molecules for such applica-
tions involves developing structure–property relationships for
tuning their absorption, their electronic properties or their
physical properties. In that aim, and in the continuation of
our former work, we focus on the rational design of nickel
bisdithiolenes: we investigate here three derivatives possessing
alkyl chains grafted on the dithiolene core. They are part of the
family commonly known as the ‘‘non-cyclic SR family’’10
(Fig. 1). The pendant alkyl chains introduced in the periphery
of the molecule may influence the way of packing which
therefore significantly differs from that of complexes with
cyclic ligands.11 To the best of our knowledge, crystallo-
graphic data of molecules with odd-numbered chain lengths
are presented here for the first time.
a CNRS, LCC (Laboratoire de Chimie de Coordination), 205,
route de Narbonne, BP 44099, F-31077 Toulouse, France
b
´
Universite de Toulouse, UPS, INPT, LCC, F-31077 Toulouse,
France. E-mail: kathleen.chane@lcc-toulouse.fr;
Tel: +33 5 61 33 31 52
c CNRS, IRSAMC, Laboratoire de chimie et physique quantiques,
118 route de Narbonne, F-31062 Toulouse, France
w Electronic supplementary information (ESI) available: X-Ray struc-
tures, schemes and computed RAMAN data. CCDC 646782 (for
complex 1) and CCDC 646783 (for complex 3). For ESI and crystal-
lographic data in CIF or other electronic format see DOI: 10.1039/
c2nj40398f
In addition to this experimental contribution, theoretical
investigations are carried out in parallel to fully describe these
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2033–2041 2033

View Article Online
Fig. 1 General structure of neutral bis-dithiolene complexes of the
SR family.
complexes: in recent work,12 we have shown by using a reliable
model system that DFT calculations provide high quality
calculations of structural parameters and vibrational spectra.
On the contrary, computation of absorption spectra of such
nickel dithiolene molecules needs elaborate multireference
approaches to achieve accurate results in the reproduction of
the very intense electronic transition displayed by these sets of
complexes in the UV-VIS-NIR region. The multideterminantal
character of the ground state wave function prohibits use of
DFT calculations and the size of the complexes investigated in
this work renders ab initio studies unrealistic. On the one hand,
we performed DFT calculations to explore the incidence of large
Fig. 2 Molecular structure of complex 1 and numbering scheme.
Thermal displacement ellipsoids are drawn at the 50% probability level.
%
crystallizes in a triclinic system c, space group P1, with one
molecule per unit cell (the Ni atom lies in a centre of symmetry
and thus the asymmetric unit contains half a molecule). Fig. 2
displays the molecular structure of complex 1.
The central [Ni(S2C2S2)2] core of the molecule is planar
(the maximum deviation from the mean plane is 0.09 A for atom
C9). The C2–C8 side chain is almost located in the plane of this
central core, their two respective planes forming a dihedral angle
of 8.411. Conversely the C10–C16 side chain is significantly
twisted as its mean plane makes a dihedral angle of 87.381 with
the plane of the central core (a side view of the molecule is given in
Fig. S1 (ESIw)). Measured bond lengths and angles of complex 1
are listed in Table 1. For comparison and subsequent discussion,
computed values are also gathered in Table 1.
alkyl chains on the Raman spectra of Ni(dmit)2 (dmit2ꢀ
=
2-thioxo-1,3-dithiole-4,5-dithiolato) derivatives. On the other
hand, theoretical calculations are performed and systematically
compared to experimental results concerning parameters of
the molecular structures, energy levels HOMO and LUMO,
electronic and Raman scattering.
Since the unit cell contains only one molecule, the crystal
structure is a simple periodic pattern in which the molecules
are repeated by the parameters a, b, c. The overall packing
Results and discussion
can be described as
a set of columns of molecules
running along the [100] axis (Fig. 3). The shortest distance
between two adjacent Ni atoms is thus the a parameter
(5.335 A), but since the molecules are strongly tilted with
respect to this direction, the distance between two
[Ni(S2C2S2)2] mean planes is 3.66 A. There is no molecular
overlap and no short contacts.
Scheme 1 illustrates the final step of the synthesis of the three
complexes. The preparation of the bis(alkylthio)-substituted
1,3-dithiole-2-thione precursors ligands 1 to 3 was carried out
according to a published procedure13 but with optimized reflux
times and purification by column chromatography instead of
recrystallization, allowing a quantitative yield to be systemati-
cally reached. Subsequent complexes 1 to 3 were obtained with
improved yields (33, 69 and 62%, respectively) in comparison to
the reference procedure (15%), thanks to longer times of stirring
and refluxing. While complexes 1 and 3 were obtained as dark
green crystals, complex 2 was isolated as a viscous oil.
Complex 3 crystallises in the monoclinic system, space group
C2/c. As for complex 1, the molecule [Ni(S2C2(SC4H9)2)2] is
centrosymmetric with the Ni atom lying on a centre of
symmetry (Fig. 4).
The central [Ni(S2C2S2)2] core is planar, the C2–C6 side
chain lies at the central plane while the C8–C12 side chain
is much more twisted (a side view of the molecule is given in
Fig. S2 (ESIw)). Bond lengths and angles are reported in
Table 2. For comparison and subsequent discussion, computed
values are also gathered in Table 2.
Crystal structure descriptions
X-ray-quality crystals of complex 1 and complex 3 were grown by
slow evaporation from dichloromethane–methanol solutions. The
molecular structures were analyzed by X-ray structural analysis at
160 K and 150 K, respectively. As previously described for parent
compounds containing linear alkyl chains [Ni(S2C2(SC4H9)2)2]
(complex A) and [Ni(S2C2(SC6H13)2)2] (complex B),13 complex 1
The molecules are stacked in columns along the [010]
direction, the shortest distance between two Ni atoms is so
the b parameter (6.464 A). As the central plane [Ni(S2C2S2)2]
of the molecules is strongly tilted with regard to the [010] direction,
the shortest distance between two repeated [Ni(S2C2S2)2]
planes in a column is 3.75 A. As before, owing to this
arrangement in the crystal, there is no short contact or
molecular overlap (Fig. 5).
In a previous study,13 it has been observed that the melting points
of parent linear-chain-containing complexes fluctuate depending on
the odd or even number of carbons constituting the chain. The
following explanation was tentatively proposed: the alkyl
chains influence the interaction between the [Ni(S2C2S2)2] cores.
In order to assess this hypothesis, we compare crystallographic data
Scheme 1 Schematic synthetic path leading from ligands 1–3 to
complexes 1–3.
c
2034 New J. Chem., 2012, 36, 2033–2041
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Table 1 Comparative experimental (X-ray) and computed values of bond lengths (A) and angles (1) for complex 1
Distance
Distance
Experimental
Computed
Experimental
Computed
S1–Ni
S1–C1
S3–Ni
S2–C9
S3–C1
S3–C2
S4–C9
S4–C10
C1–C9
C2–C3
C3–C4
2.1207(13)
1.712(5)
2.1319(13)
1.699(5)
1.743(5)
1.799(5)
1.752(5)
1.810(5)
1.368(7)
1.511(7)
1.523(7)
2.131
1.710
2.133
1.710
1.757
1.821
1.761
1.820
1.398
1.519
1.522
C4–C5
C5–C6
C6–C7
C7–C8
C10–C11
C11–C12
C12–C13
C13–C14
C14–C15
C15–C16
1.509(8)
1.517(7)
1.503(8)
1.512(8)
1.520(7)
1.518(7)
1.516(7)
1.512(7)
1.519(7)
1.514(8)
1.521
1.521
1.521
1.519
1.522
1.518
1.521
1.521
1.521
1.519
Angle
Angle
Experimental
Computed
Experimental
Computed
S1–Ni–S2
S1–C1–S3
S1–C1–C9
S2–C9–C1
S3–C1–C9
S3–C2–C3
S4–C10–C11
C1–S1–Ni
C1–S3–C2
C2–C3–C4
91.94(5)
120.8(3)
119.2(4)
120.1(4)
120.0(4)
109.0(4)
115.8(4)
104.48(19)
103.9(3)
111.5(5)
92.1
C3–C4–C5
C4–C5–C6
C5–C6–C7
C9–S2–Ni
C9–S4–C10
C10–C11–C12
C11–C12–C13
C12–C13–C14
C13–C14–C15
C14–C15–C16
114.0(5)
113.4(5)
113.8(5)
104.20(18)
104.7(3)
110.4(4)
113.8(5)
113.7(5)
113.4(4)
113.1(5)
113.2
113.5
113.7
104.3
103.8
112.0
113.1
113.6
113.7
113.4
121.8
119.8
119.7
118.4
109.0
114.5
104.2
103.2
112.0
Fig. 4 X-Ray structure of complex 3 showing the molecular unit and
Fig. 3 Perspective view of the packing of complex 1 molecules.
numbering scheme.
previously reported for two of the studied complexes (complex A
and complex B),13 with even-numbered chain lengths, with those
of complex 1 (see Table 3). The dihedral angles measured for the
twisted chains are used to compare their deviations from the core
plane: F1 and F2 are the S2–C9–S4–C10 and the C9–S4–C10–C11
dihedral angles, respectively, of complex 1. The same F1 and F2
angles were measured for complexes A and B. The shortest Ni–Ni
distance and the distance between two adjacent [Ni(S2C2S2)2]
mean planes may give information about possible central cores
interactions. Data for complex 3 are also reported in Table 3.
The three linear-chain-containing complexes crystallize in
complex 1 if compared to complex B. There is thus no clear
explanation for the rising of the melting point of complex 1
(82 1C) in comparison to that of complex B (70 1C). For
complex B and complex 1, the atomic deviation of the end-
chain carbon relative to the [Ni(S2C2S2)2] mean plane is of the
same order of magnitude (2.748 and 2.338 A, respectively). It is
noteworthy that the torsion of the chain occurs at the first
carbon (C10) of the chain for complexes A and 1 whereas it is
observed at the third one (C12) for complex B. Thus, the
fluctuation over a temperature range of 20 1C for the n = 4
to n = 11 derivatives observed by Underhill et al.13 probably
relies on interactions between the alkyl chains. This is clearly
the case for complex 3 which presents a high melting point of
168 1C and for which numerous Sꢁ ꢁ ꢁH short contacts between
two adjacent molecules can be observed (see ESIw).
%
the P1 space group and they show similar crystallographic data,
in particular concerning the distance between [Ni(S2C2S2)2]
mean planes, from 3.50 to 3.69 A. No supplementary specific
interaction through either Niꢁ ꢁ ꢁS or Sꢁ ꢁ ꢁS contact is formed in
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2033–2041 2035

View Article Online
Table 2 Comparative experimental (X-ray) and computed values of bond lengths (A) and angles (1) for complex 3
Distance
X-Ray
Computed
X-Ray
Computed
S1–Ni
S1–C1
S2–Ni
S2–C7
S3–C2
S4–C7
S4–C8
C1–C7
2.1330(7)
1.703(3)
2.1325(7)
1.701(3)
1.804(3)
1.746(3)
1.814(3)
1.404(4)
2.133
1.700
2.132
1.700
1.804
1.743
1.807
1.400
C2–C3
C3–C4
C4–C5
C4–C6
C8–C9
C9–C10
C10–C11
C10–C12
1.518(4)
1.523(4)
1.516(5)
1.514(5)
1.521(4)
1.523(4)
1.514(4)
1.527(4)
1.529
1.529
1.523
1.522
1.521
1.527
1.522
1.523
Angle
X-Ray
Computed
X-Ray
Computed
S1–Ni–S2
S1–C1–S3
S2–C7–S4
C1–S1–Ni
C1–S3–C2
C1–C7–S2
C1–C7–S4
C2–C3–C4
C3–C2–S3
C3–C4–C5
92.41(3)
92.0
C3–C4–C6
C5–C4–C6
C7–S2–Ni
C7–S4–C8
C7–C1–S1
C7–C1–S3
C8–C9–C10
C9–C8–S4
C9–C10–C11
C9–C10–C11
C11–C10–C12
110.1(3)
112.9(5)
104.46(9)
103.58(13)
119.7(2)
118.26(19)
115.1(2)
113.5(2)
110.0(3)
111.9(3)
110.5(3)
109.88
110.6
104.6
104.1
119.6
118.4
115.9
115.1
109.9
109.9
110.5
121.99(15)
122.14(15)
104.15(10)
104.05(13)
119.2(2)
118.6(2)
113.6(2)
107.41(18)
112.9(3)
122.0
122.8
104.5
130.4
119.3
117.9
113.6
108.5
112.4
these results, we propose to describe the core of complex 2 through
computed structural parameters. They are collected after geometry
optimisation without symmetry constraints in Fig. 6. The geo-
metrical parameters proposed in Fig. 6 have expected values:
the bonds lengths as well as the angles are comparable to those
observed in the crystal structures of complex 1 and complex 2.
Cyclic voltammetry
Electrochemistry of complexes 1 and 2 has already been
described,14 showing three electrochemical processes attributed to
electron exchange on the core of the complexes. Table 4 reports the
measured potential values associated with the first reduction and
the first oxidation of complexes 1 to 3, respectively. A formula
proposed by Bre
´
das et al.15 on the basis of a comparison between
Fig. 5 Perspective view of the packing of complex 3 molecules.
effective Hamiltonian calculations and experimental electrochemical
measurements on the conjugated system allows their HOMO and
LUMO energy levels to be estimated.
Computed geometrical parameters
As can be seen in Tables 1 and 2, geometry optimizations lead to
computed geometrical parameters in very good agreement with
experimental data for the two complexes. The computed distances
as well as the angles are very well reproduced. For the dithiolene
core, the values of the C–C bond lengths as well as of the C–S bond
lengths clearly indicate their delocalized character. Starting from
HOMO = ꢀ(E1/2(ox) + 4.8) eV and
LUMO = ꢀ(E1/2(red) + 4.8) eV
They give rise to the value of the electrochemical bandgap
denoted Egel. Experimental values are gathered in Table 4, and
calculated values are included for comparison:
Table 3 Comparative data obtained by X-ray structural analysis on complexes containing either linear alkyl chains (complexes A, B and 1) or
branched alkyl chains (complex 3)
[Ni(S2C2(SC4H9)2)2]
Complex Aa
[Ni(S2C2(SC6H13)2)2]
Complex Ba
[Ni(S2C2(SC7H15)2)2]
Complex 1b
[Ni(S2C2(SC5H11)2)2]
Complex 3b
%
P1
%
P1
%
P1
Space group
Distance 1c
Distance 2d
F1e
F2
C2/c
a = 7.753 A
3.69 A
10.191
a = 5.254 A
3.50 A
2.071
a = 5.335 A
3.66 A
7.261
b = 6.464 A
3.75 A
21.091
e
79.931
177.721
83.621
76.171
a
b
Ref. 13. This work. Shortest Ni–Ni distance. Distance between [Ni(S2C2S2)] mean planes. Defined in the text (see above).
c
d
e
c
2036 New J. Chem., 2012, 36, 2033–2041
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Fig. 7 Electronic spectra of complex 1, complex 2 and complex 3
recorded in dichloromethane at room temperature.
spectra of the complexes 1, 2 and 3 show strong absorption bands
in dichloromethane at l = 1007 nm (e B 34 000 L molꢀ1 cmꢀ1),
l = 1010 nm (e B 34 000 L molꢀ1 cmꢀ1) and l = 1007 nm
(e B 38 000 L molꢀ1 cmꢀ1), respectively. Their absorption
bands are offset to the NIR region in comparison to analogues
substituted by phenyl rings.7–9,20 We observe a minor solvato-
chromic effect as their absorption is shifted by only 15 nm
towards lower wavelengths in heptanes, in comparison with
dichloromethane. The three complexes exhibit relatively low
absorption from 450 to 700 nm. For the three complexes, the
optical bandgaps measured in solution are equal to 1.15 eV.
Thin films of complexes 1 and 2 present enlarged absorption
bands (in comparison with solutions) leading to a slightly
lower value of 1.08 eV. Complex 3 did not give rise to reliable
data on thin films for film quality reasons.
Fig. 6 Computed bond lengths (A) and angles (1) for complex 2.
Table 4 Comparative experimental (electrochemistry) and computed
values of the HOMO–LUMO bandgaps
Electrochemical dataa
E1/2 E1/2 LUMO/ HOMO/ Egel/ LUMO/ HOMO/
Complex (red)/V (ox)/V eV eV eV eV eV
Computed data
1
2
3
ꢀ0.13 0.73 ꢀ4.67 ꢀ5.53
ꢀ0.13 0.73 ꢀ4.67 ꢀ5.53
ꢀ0.14 0.74 ꢀ4.66 ꢀ5.54
0.86 ꢀ3.42 (ag) ꢀ4.96(au)
0.86 ꢀ3.42 ꢀ4.96
0.88 ꢀ3.45(ag) ꢀ4.98(au)
a
Reduction and oxidation potentials (V/SCE) obtained by SQW in
CH2Cl2 with (n-Bu4)[PF6] (0.1 mol Lꢀ1) on Pt microdisk.
The electrochemical redox potentials are very similar for the
three complexes. The eigenvalues of the computed HOMO–
LUMO are in good agreement with the experimental ones: the
computed HOMOs are around 0.6 eV lower in energy than the
values deduced from electrochemistry. The difference for LUMO
energies is slightly higher (around 1.3 eV). The calculated frontier
orbitals of the three complexes are identical in shape and in
energy (see Fig. S4 in the ESIw). HOMOs are ligand in character
while the LUMOs are mainly S (3p) in character and display
antibonding interactions with a 3d orbital of the nickel atom.
Ligand valence orbitals lie at higher energy than the Ni-d
orbital manifold. The different alkyl chains have no effect on
the electronic properties of the three complexes. The electro-
chemical bandgaps (Egel) are found to be similar, around
0.9 eV whereas calculated values are found to be significantly
Raman spectroscopy and computed Raman data
Resonance Raman excitation profiles have been determined
for crystals of complex 1 and complex 3 and for an oil of
complex 2 in the 200–1500 nm region. These spectra (Fig. 8)
are fairly similar to those obtained for the [Ni(S2C2S2)2] core
of neutral bis(1,2-dithiolene) complexes.21,22 For complex 3,
two distinct low frequency bands are observed at 382 cmꢀ1
and 372 cmꢀ1 (see Fig. 8, top) whereas a single signal is
observed in this region for the two other compounds.
We have calculated Raman modes and assigned their
respective representations to bands vibrations (see ESIw,
Fig. S5). The spectral features for the three compounds are quite
similar. Due to interaction with neighbouring molecules in the
crystal state, some deviation from the calculated frequencies of
the free complexes is expected for complex 1 and complex 3.
Analysis of vibration modes of complex 2 is more difficult due to
the lack of molecular symmetry. It is anticipated that the oily
nature of complex 2 may induce a great variety of lateral chain
conformations. This is probably the reason why the Raman
spectrum of complex 2 displays numerous poorly resolved bands.
For both complexes 1 and 3, the Raman spectra have a limited
number of vibration modes because of the rule of mutual
exclusion. Only molecular vibrations symmetric with regard to
the centre of symmetry (symmetry Ag) are present. Among them
we find characteristic modes of the core of these complexes.
cal
larger: Eg = 1.53–1.54 eV. It is worth noting that our
calculations give the same value for the three molecules since
the HOMO and LUMO energies are not affected by the
presence of the different alkyl chains.
Electronic spectra
Neutral bis(1,2-dithiolene) complexes absorb strongly in the
NIR region.16–18 This absorption is usually assigned to a p–p*
transition between orbitals delocalized over the [Ni(S2C2S2)2]
core of the complex.9,19 As can be seen in Fig. 7, the electronic
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2033–2041 2037

View Article Online
This value is exactly the same for the three complexes, maybe
because this mode is not mixed with other vibrations.
Fig. 9 displays selected normal vibrations of the different
alkyl chains. Calculated values for H–C–H, H–C–C and
C–C–C bending motions occur in the 1000–1500 cmꢀ1 region.
For instance, complex 1 shows twisting modes of the
CH2 group with frequencies of 1322 cmꢀ1, 1323 cmꢀ1 and
1325 cmꢀ1. Theoretical calculations predict an increase in the
vibration intensity per added CH2 group. Experimental results
confirm this prediction. In complex 3, the frequency of
1334 cmꢀ1 is attributed to the bending of the C–CMe–C
linkage. Agreement between theoretical and experimental data
appears satisfactorily good and our assignments are in accordance
with those proposed for other neutral bisdithiolene complexes
in recent work.23
Experimental
Synthetic procedures
All solvents used were dried by refluxing over magnesium
turnings, followed by distillation, and then stored under argon
before use.
Preparation of [Ni(S2C2(SC7H15)2)2] 1. Sodium methanolate
freshly prepared from 0.23 g of sodium (10 mmol) dissolved in
dry methanol was added under a dry argon atmosphere to a
solution of 4,5-bis(benzoylthio)-1,3-dithiole-2-thione (2.00 g,
4.92 mmol) in methanol (10 mL). After 1 h of stirring, 1.6 mL
of 1-bromoheptane (1.82 g, 10 mmol) was added dropwise to
the dark purple solution. The mixture was then refluxed
(78 1C) for 5 h, under vigorous agitation. The dark orange
solution was concentrated in vacuo to afford an orange oil.
Addition of pentane (20 mL) to this oil induced a white solid
to precipitate (NaBr), which was filtered and washed with
pentane (10 mL). The collected filtrates were concentrated
under vacuum and the resulting orange oil was purified by a
chromatography column (silica gel Fluka 60, eluant: hexa-
ne–acetone (95/5)) to afford 1.93 g of ligand 1 (quantitative
Fig. 8 Raman spectra of complexes 1, 2 and 3 (a) in the 200–500 nm
region (top) and (b) in the 400–1500 nm region (bottom).
Fundamental vibrations of the nickel core are weakly
influenced by the presence of alkyl chains. The low frequency
band near 380 cmꢀ1 corresponds to a combination of Ni–S
stretching bond and CQC–S bending modes. In the case
of complex 3, an additional band is observed. As shown in
Fig. S-5 (see ESIw), it could be attributed to a combination of
symmetric Ni–S bond stretching with C–CMe–C bending. As
expected, the intense band near 985 cmꢀ1 corresponds
to the C–S bond stretching mode. The weak intense
mode at 1393 cmꢀ1 corresponds to the C–C stretching mode.
1
yield) as a pale yellow oil. H NMR (CD2Cl2) d, ppm: 0.92
(t, 6H, 3J(H,H) = 6.7 Hz), 1.32 to 1.47 (m, 16H), 1.70 (q, 4H,
3
3J(H,H) = 7.5 Hz), 2.92 (t, 4H, J(H,H) = 7.5 Hz); GCMS:
m/z = 394 (M+).
Pentane-washed sodium (0.23 g, 10 mmol) dissolved in
methanol (10 mL) was added to the preceding oil (4.9 mmol)
dissolved in 10 mL of dry methanol. The orange mixture was
refluxed (78 1C) for 2.5 h under vigorous stirring and then
allowed to cool to room temperature. A solution of [Bu4N][Br]
(3.22 g, 10 mmol) in methanol (10 mL) was added. After 1 h of
stirring, NiCl2ꢁ6H2O (0.59 g, 2.5 mmol) in methanol (50 mL)
was added dropwise. The mixture was then maintained at
room temperature overnight with stirring. After filtration of
the mixture, the solvent was removed under vacuum to yield a
black-brown oily solid which was redissolved in acetone
(50 mL).
To the dark brown solution was added I2 (0.63 g, 2.5 mmol)
and NaI (1.12 g, 7,5 mmol) dissolved in acetone (40 mL),
resulting in an instantaneous colour change to dark green.
Reduction of the solvent volume induced a dark coloured solid
Fig. 9 Normal vibrations of the alkyl chains of complex 1: CH2 twisting
mode at 1323 cmꢀ1 (top left); CH2 twisting mode at 1325 cmꢀ1 (top right)
and complex 3: C–C–C bending mode at 1334 cmꢀ1 (bottom).
c
2038 New J. Chem., 2012, 36, 2033–2041
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
to precipitate, which was filtered off, washed with methanol
and dried.
spectrometer (DYLOR XY micro Raman) equipped with
a helium–neon laser (l = 6328 A, laser power density:
Recrystallisation from CH2Cl2–IprOH afforded the pure
dark green crystalline product (0.62 g, yield 33%); mp
82 1C. IR (KBr) n, cmꢀ1: 490 (NI–S), 803 (C–S), 1020, 1098,
1257 (C–C) + (C–S), 1418 (CQC); near-IR absorption
spectrum (CH2Cl2) lmax, nm (eM, L molꢀ1 cmꢀ1): 1007
(33 600); 1H NMR (CD2Cl2) d, ppm: 0.93 (t, 12H,
3J(H,H) = 6.9 Hz ), 1.35 to 1.57 (m, 32H), 1.88 (m, 8H),
3.42 (m, 8H); Raman spectrum (on crystal) n, cmꢀ1: 230 (m),
380 (s), 440 (w), 478 (w), 495 (w), 508 (w), 555 (w), 940 (m),
1065 (w), 1316 (s), 1332 (s) and 1411 (m); anal calc. for
C32H60S8Ni: C, 50.57; H, 7.96; S, 33.75; Ni, 7.72%. Found:
C, 50.45; H, 7.93; S, 34.02; Ni, 7.76%.
9 ꢂ 104 W cmꢀ2). 1H NMR spectra were recorded on a Bruker
¨
DPX 300 spectrometer operating at 300.13 MHz. Dichloro-
methane-d2 (CD2Cl2) was used as the solvent and tetramethyl-
silane (TMS) as the internal standard. Melting points were
determined by DSC thermograms obtained on a DSC 204
NETZSCH system using 2–5 mg samples in 30 ml sample pans
and a scan rate of 10 1C minꢀ1. Mass analyses were performed
at the MS Service of the Paul Sabatier University. Elemental
analyses were carried out at the CNRS Central Analyses
Service in Lyon. Electronic spectra were recorded on a Perkin
Elmer Lambda 35.
Voltammetric measurements were carried out with an
Autolab PGSTAT100 potentiostat controlled by GPES 4.09
software. Experiments were performed at room temperature
in a homemade airtight three-electrode cell connected to a
vacuum/argon line. A saturated calomel electrode (SCE)
separated from the solution by a bridge compartment was
used as a reference electrode. The counter electrode was a
platinum wire of ca. 1 cm2 apparent surface. Working electrodes
were: a Pt microdisk (0.5 mm diameter). The supporting electro-
lyte, (n-Bu4N)[PF6] (Fluka, 99% puriss electrochemical grade),
was used as received. For studies in solution, electrochemical
media containing 0.1 mol Lꢀ1 of supporting electrolyte, and
10ꢀ3 mol Lꢀ1 of complex were used. Before each measure-
ment, the solutions were deaerated by bubbling argon and the
working electrode was polished with a Presi P230 polishing
machine. Square Wave Voltammetry (SWV) was performed
at room temperature, and the parameters are: SW frequency
f = 20 Hz, SW amplitude ESW = 20 mV, and scan increment
dE = 5 mV.
Preparation of [Ni(S2C2(SC8H17)2)2] 2. The neutral complex
2 was prepared in the same way. The ligand 2 was first
synthesized, purified by a chromatography column (silica
gel Fluka 60, eluant: hexane–dichloromethane (90/10)) and
isolated as a yellow oil in quantitative yield (2.07 g); 1H NMR
(CD2Cl2): d, ppm: 0.92 (m, 12H), 1.29 to 1.55 (m, 16H), 1.55 to
1.67 (m, 2H), 2.92 (d, 4H, 3J(H,H) = 6.3 Hz); GCMS: m/z =
422 (M+). It was then totally engaged in the synthesis of the
green oily complex 2, which was purified by a chromatography
column (silica gel Fluka 60, eluant: hexane–dichloromethane
(99/1)); (1.4 g, yield 69%); near-IR absorption spectrum
1
(CH2Cl2) lmax, nm (eM, L molꢀ1 cmꢀ1): 1010 (33 600); H NMR
(CD2Cl2) d ppm: 0.93 to 0.99 (m, 24H), 1.30 to 1.61 (m, 32H),
1.88 (m, 4H), 3.42 (m, 8H); anal calc. for C36H68S8Ni: C,
52.98; H, 8.40; S, 32.01; Ni, 7.19%. Found: C, 53.30; H, 8.41;
S, 31.14; Ni, 6.84%.
Preparation of [Ni(S2C2(SC5H11)2)2] 3. The neutral complex
3 was prepared in the same way. The ligand 3 was first
synthesized, purified by a chromatography column (silica
gel Fluka 60, eluant: hexane–dichloromethane (95/5)) and
isolated as a yellow oil in quantitative yield (1.80 g);
Structure determination
Single-crystal X-ray diffraction data were collected at 160 K
for complex 1 and at 150 K for complex 3 on a Xcalibur
Oxford Diffraction diffractometer using graphite-monochro-
mated Mo Ka radiation. Crystals were cooled by a nitrogen
gas flow using an Oxford Cryosystems Cryostream Cooler
Device. The structures were solved by direct methods using
SIR92,24 and refined by means of least-squares procedures on
F using the programs of the PC version of CRYSTALS.25
Atomic scattering factors were taken from the International
Tables for X-ray Crystallography.26 All non-hydrogen atoms
were refined anisotropically. Hydrogen atoms were located on
difference Fourier maps and repositioned geometrically during
the refinement process using the riding model. The programs
MERCURY,27 ORTEP III28 and CAMERON29 were used
for drawings.
3
1H NMR (CD2Cl2): d, ppm: 0.95 (d, 12H, J(H,H) = 6.6 Hz),
1.55 to 1.63 (m, 2H), 1.73 to 1.76 (m, 4H), 2.91 to 2.96 (m, 4H);
GCMS: m/z = 338 (M+). It was then totally engaged in the
synthesis of complex 3, which was purified by recrystallisation
from hexane–acetone to give the pure dark green crystalline
product (1.0 g, 52%); mp 168 1C; IR (KBr) n/cmꢀ1: 490 (NI–S),
803 (C–S), 1020, 1098, 1257 (C–C) + (C–S), 1418 (CQC); near-
IR absorption spectrum (CH2Cl2) lmax, nm (eM, L molꢀ1 cmꢀ1):
1007 (37 600); 1H NMR (CD2Cl2) d, ppm: 1.02 (d, 24H,
3J(H,H) = 6.6 Hz), 1.55 (s, 4H), 1.76 to 1.81 (m, 8H), 3.40 to
3
3.46 (t, 8H, J(H,H) = 7.5 Hz); anal calc. for C24H44S8Ni: C,
44.50; H, 6.84 ; S, 39.60; Ni, 9.06%. Found C, 44.19; H, 6.75 %;
S, 39.37%; Ni, 8.66%.
Crystal structure determination of complex 1 at 160 K
Crystal data. C32H60Ni1S8, M = 760.07, triclinic, a =
5.3358(6) (A), b = 13.3538(13) (A), c = 13.8940(18) (A),
a = 101.260(10)1, b = 90.396(10)1, g = 101.111(19)1, V =
Chemical analyses
GCMS analysis was realised on a Hewlett Packard 6890 CPG
combined with a Hewlett Packard 5973 mass spectrometer.
3
%
951.72(19) (A) , T = 160 K, space group P1, Z = 1, m(MoKa) =
Physical measurements
2.463, 9034 reflections measured, 5044 independent reflections
(Rint = 0.06) of which 1929 reflections with I 4 1.7s(I) were
used for the refinement of 187 parameters. The final R and
wR(F) are 0.046 and 0.051, respectively.
IR spectra and near-infrared spectra were recorded on a
Perkin Elmer GX FT-IR spectrophotometer. Raman measure-
ments were carried out at room temperature using a Raman
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2033–2041 2039

View Article Online
Crystal structure determination of complex 3 at 150 K
Crystal data. C24H44Ni1S8, M = 647.85, monoclinic, a =
26.182(2) (A), b = 6.4642(6) (A), c = 19.9353(11) (A), b =
107.729(6)1, V = 3213.7(5) (A)3, T = 150 K, space group
C2/c, Z = 4, m(MoKa) = 2.463, 30 379 reflections measured,
4287 independent reflections (Rint = 0.04) of which 2201
reflections with I 4 3s(I) were used for the refinement of
151 parameters. The final R and wR(F) are 0.032 and 0.036,
respectively.
Conclusions
Three neutral nickel complexes of the ‘‘non-cyclic SR’’ family
have been investigated and fully characterized. Their crystal-
line organization, their behaviour in electrochemistry and their
electronic spectra (UV-vis and Raman) were fully investigated.
The nature of the alkyl chains has no influence on the
electronic properties, i.e. electrochemistry and electronic
spectra (UV-vis and Raman) of the three investigated com-
plexes but shows significant impact on the physical properties,
i.e. melting point and crystalline organization. All the experi-
mental data were systematically discussed in the light of
computed values. Using these theoretical tools, we show a
direct link between measured and calculated molecular parameters
and we demonstrate the accuracy of theoretical calculations for
these three neutral nickel coordination compounds. This reinforces
our wish to combine experimental and theoretical approaches
and encourages us to continue in this way. This approach
opens the way toward predictive calculations, a very ambitious
topic in the field of organic electronics. Further works will
concern the design of such molecules for organic electronics,
with a new challenge that consists of taking into account
charge transfer considerations.
Calculations
Two different procedures were applied depending on the
complexes.
Complex 1 and complex 3. Calculations were performed
using the Gaussian 03 package. The PBE030 functional was
used to perform geometry optimisations followed by analytical
Raman frequency calculations.
Starting from crystallographic data, geometry optimization
led to an electronic structure of Ci symmetry. Raman frequencies
were determined via DFT using the PBE0 functional. Calculations
are performed with a triple-z plus polarisation basis set (6-311G,
(2df, 2pd))31 for the hydrogen and carbon atoms, a sulfur atom is
described with a triple-z plus polarisation basis set (6-311++G,
(3df, 3pd)).32 The inner 10 core electrons of the Ni atom were
described using the Stuttgart relativistic effective core potential
with the corresponding basis set (6s5p3d1f).33
Acknowledgements
The authors thank French National Research Agency for financial
support as part of ANR CPA-FALOIR 2007–2010 and ANR
Spir-Wind 2009–2012 Projects. The authors also thank Professor
Colin J. Marsden for critically reading the manuscript.
Complex 2. Due to the size of this complex and in the
absence of crystallographic data another strategy was used to
study complex 2. Electronic structure calculations and Raman
frequency calculations of complex 2 were carried out with PC
Gamess.34 Geometry optimizations and Raman frequency
analyse were run without symmetry constraints. The PBE0
functional was employed in combination with scalar relativistic
core potential SBKJC and their corresponding basis sets35 for
all atoms.
References
1 Y. Wen and Y. Liu, Adv. Mater., 2010, 22, 1331.
2 L. Duan, J. Qiao, Y. Sun and Y. Qiu, Adv. Mater., 2011, 23, 1137.
3 S. Shanmugaraju, S. A. Joshi and P. S. Mukherjee, Inorg. Chem.,
2011, 50, 11736.
4 (a) T. M. Clarke and J. R. Durrant, Chem. Rev., 2010, 110, 6736;
(b) D. Credgington, R. Hamilton, P. Atienzar, J. Nelson and
J. R. Durrant, Adv. Funct. Mater., 2011, 21, 2744.
5 A. Weston and M. Murthy, Carbon, 1996, 34, 1267.
6 J. E. Antony, Chem. Mater., 2011, 23, 583.
For comparison purposes, data proposed in Fig. 6 were
obtained from a single point energy calculation with the same
procedure as that used for the complexes 1 and 3.
7 T. T. Bui, B. Garreau-de Bonneval and K. I. Moineau-Chane
Ching, New J. Chem., 2010, 34, 337.
8 T. T. Bui, O. Thiebaut, E. Grelet, M. F. Achard, B. Garreau-de
Bonneval and K. I. Moineau-Chane Ching, Eur. J. Inorg. Chem.,
2011, 17, 2663.
9 J.-Y. Cho, B. Domercq, S. C. Jones, J. Yu, X. Zhang, Z. An,
M. Bishop, S. Barlow, S. R. Marder and B. Kippelen, J. Mater.
Chem., 2007, 17, 2642.
10 B. Garreau-de Bonneval, K. I. Moineau-Chane Ching, F. Alary,
T. T. Bui and L. Valade, Coord. Chem. Rev., 2010, 254, 1457 and
references therein.
11 G. Soras, N. Psaroudakis, M. J. Manos, A. J. Tasiopoulos,
D. G. Liakos and G. A. Mousdis, Polyhedron, 2009, 28, 3340.
12 F. Alary, J. L. Heully, A. Scemama, B. Garreau-de-Bonneval,
K. I. Chane-Ching and M. Caffarel, Theor. Chem. Acc., 2010,
126, 243 and references therein.
13 A. Charlton, C. A. S. Hill, A. E. Underhill, K. M. A. Malik,
M. B. Hursthouse, A. I. Karaulov and J. Moller, J. Mater. Chem.,
1994, 4, 1861.
14 A. Sournia-Saquet, B. Garreau-de Bonneval, K. I. Chane-Ching
and L. Valade, J. Electroanal. Chem., 2008, 624, 84.
15 J. L. Bredas, R. Silbey, D. S. Boudreaux and R. R. Chance, J. Am.
´
Chem. Soc., 1983, 105, 6555.
16 U. T. Mueller-Westerhoff, B. Vance and D. I. Yoon, Tetrahedron,
1991, 47, 909.
Prediction of Raman intensities. The Raman activities (Si)
calculated were converted to relative Raman intensities (Ii)
using the following relationship derived from the basic theory
of Raman scattering.36
4
f ðnꢀ0 ꢀ nꢀiÞ Si
Ii ¼
nꢀi½1 ꢀ expðhcnꢀi=kTÞꢃ
where nꢀ0 is the exciting frequency (in cmꢀ1), nꢀi is the vibra-
tional wavenumber of the ith mode, h, c and k are the
fundamental constants and f is a suitably chosen common
normalization factor for all peak intensities. Herein the
conversion was carried out with nꢀ0 = 15802.8 cmꢀ1 and
T = 298 K.
The Raman frequencies were not corrected for anharmoni-
city or other additional effects by scaling factors.
c
2040 New J. Chem., 2012, 36, 2033–2041
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
17 F. Bigoli, P. Deplano, F. A. Devillanova, V. Lippolis, P. J. Lukes,
M. L. Mercuri, M. A. Pellinghelli and E. F. Trogu, J. Chem. Soc.,
Chem. Commun., 1995, 371.
18 F. Bigoli, P. Deplano, M. L. Mercuri, M. A. Pellinghelli, G. Pintus,
E. F. Trogu, G. Zonnedda, H. H. Wang and J. M. Williams, Inorg.
Chim. Acta, 1998, 273, 175.
26 D. T. Cromer, International Tables for X-ray Crystallography, the
Kynoch Press, Birmingham, England, vol. IV, 1974.
27 Copyright r 2005 The Cambridge Crystallographic Data Centre,
Registered Charity No 800579.
28 M. N. Burnett and C. K. Johnson, ORTEP III, Report ORNL-6895,
Oak Ridge National Laboratory, Tennessee, USA, 1996.
29 D. M. Watkin, L. Pearce and C. K. A. Prout, CAMERON,
University of Oxford, England, 1993.
19 C. L. Kean, D. O. Miller and P. G. Pickup, J. Mater. Chem., 2002,
12, 2949.
20 P. Basu, A. Nigam, B. Mogesa, S. Denti and V. N. Nemykin,
Inorg. Chim. Acta, 2010, 363, 2857.
21 Dithiolene chemistry Synthesis, Properties and Applications, Prog.
Inorg. Chem. 52, ed. E. I. Stiefel, J. Wiley and Sons, Hoboken,
New Jersey, 2004.
22 K. I. Pokhodnya, C. Faulmann, I. Malfant, R. Andreu-Solano,
P. Cassoux, A. Mlayah, D. Smirnov and J. Leotin, Synth. Met.,
1999, 103, 2016.
23 O. Siimann and J. Fresco, Inorg. Chem., 1971, 10, 297.
24 A. Altomare, G. Cascarano, C. Giacovazzo and A. Guagliardi,
J. Appl. Crystallogr., 1993, 26, 343.
30 C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158.
31 R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem.
Phys., 1980, 72, 650.
32 A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72, 5639.
33 M. Dolg, H. Stoll, H. Preuss and R. M. Pitzer, J. Phys. Chem.,
1993, 97, 5852.
34 A.V. Nemukhin, B.L. Grigorenko, A.A. Granovsky, Molecular
modeling by using the PC GAMESS program: From diatomic
molecules to enzymes, Moscow University Chemistry Bulletin, 2004,
45, 75.
35 W. J. Stevens, M. Krauss, H. Basch and P. G. Jasieu, Can. J. Chem.,
1992, 70, 612.
36 P. L. Polavarapu, J. Phys. Chem., 1990, 94, 8106.
25 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and
D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 2033–2041 2041