Role of the aromatic bridge on radical ions formation during reduction of diphosphaalkenes

Article abstract of DOI:10.1039/c1nj20314b

Two molecules containing two phenylphosphaalkene moieties linked by an anthracene (1) or by a naphthalene (2) ring have been synthesized and their crystal structures have been determined. While electrochemical measurements show that these two systems are

Full text of DOI:10.1039/c1nj20314b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 2510–2520
www.rsc.org/njc
PAPER
Role of the aromatic bridge on radical ions formation during reduction
of diphosphaalkeneswz
Manuel Lejeune, Philippe Grosshans, Theo Berclaz, Helena Sidorenkova,
a
a
a
a
´
Celine Besnard, Phil Pattison and Michel Geoffroy*a
b
cd
´
Received (in Montpellier, France) 9th April 2011, Accepted 13th July 2011
DOI: 10.1039/c1nj20314b
Two molecules containing two phenylphosphaalkene moieties linked by an anthracene (1) or by a
naphthalene (2) ring have been synthesized and their crystal structures have been determined.
While electrochemical measurements show that these two systems are easily reduced, EPR spectra
ꢁ
indicatꢁe that, at room temperature, the electronic structures of the two reduction compounds 1ꢀ
and 2ꢀ are quite different. In 1ꢀꢁ, in good accordancꢁe with DFT predictions, the unpaired
electron is delocalized on the full molecule while in 2ꢀ it is confined on a single phosphaalkene
moiety. Thisꢁdifference is attributed to the short distance between the two phenylphosphaalkene
groups in 2ꢀ which hinders their reorientation after addition of an electron. The role of this
motion is consistent with the fact that two additional paramagnetic species are detected at 145 K:
the dianion 22ꢁ characterized by a rather small exchange coupling constant and the radical
ꢁ
monoanion 2*ꢀ resulting from the formation of a one electron P–P bond. These two species are
probably reaction intermediates which can lead to the formation of biphosphane.
two TTF moieties linked by an organic bridge, the ability of
the system to give rise to mixed valence species after oxidation
depends on the nature of the spacer.5
1. Introduction
Molecular and supramolecular systems containing several
redox switches have received much attention in recent years
because of their potential applications in modern material
science and microelectronics.1 They play, for example, an
important role in the design of nanoelectromechanical devices2
and of molecular magnets.3 In most of these systems, the
nature of the bridge between two redox centers is often
determining for the response of the compound to external
stimuli. For example, the shuttling of a crown ether along a
redox-active rotaxane is affected by the flexibility of the bridge
between viologen recognition sites.4 In compounds containing
There is currently much interest in compounds containing
diarylethene switches bonded through delocalized bridges.6 In
diarylethene-capped sexithiophene wires, electrochemical ring
cyclization and cycloreversion are influenced by the participation
of the thiophene linker in the frontier orbitals of the resulting
paramagnetic dication.7
The purpose of the present work is to assess the role of the
bridge in the reduction mechanism of a system containing two
a Department of Physical Chemistry, University of Geneva,
30 Quai Ernest Ansermet, CH-1211 Geneva, Switzerland.
E-mail: Michel.Geoffroy@unige.ch
b Laboratory of Crystallography, University of Geneva,
24 quai Ernest-Ansermet, CH-1211 Geneva, Switzerland.
E-mail: Celine.Besnard@unige.ch
c
´ ´
Laboratory of Crystallography, Ecole Polytechnique Federale de
Lausanne, CH 1015 Lausanne, Switzerland
d Swiss Norwegian Beamline, European Synchrotron Radiation
Facility, Grenoble, France
w To the memory of Pascal Le Floch (1958–2010).
z Electronic supplementary information (ESI) available: Syntheses and
spectroscopic characterization of 10 0, 10, 20 0, 20, 3, 40, 4; geometrical
parameters from the crystal structures of 1 and 2; geometrical parameters,
ꢁ
energies and coupling constants for DFT optimized structures of 1ꢀ
;
experimental and simulated EPR spectra, cyclic voltammograms for 3.
CCDC reference numbers 815021 and 815022. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/
c1nj20314b
Chart 1
c
2510 New J. Chem., 2011, 35, 2510–2520
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
phosphaalkene groups. PQC containing compounds are
well-known to accept easily an additional electron in their
p* orbitals8 and sterically encumbered diphosphaalkenes have
already been reported as potential multiredox-active molecular
switches.9 We report the syntheses and crystal structures of
1 and 2 (Chart 1) which only differ by the size of the aromatic
bridge. By means of electrochemical and EPR measurements
the paramagnetic reaction intermediates formed during
reduction of these two diphosphaalkenes are identified. In
order to facilitate the interpretation of these results in terms
of the radical structure, the hyperfine coupling constants
observed after reduction of 1 and 2 are compared with those
of the four monophosphaalkenes 3, 4, 5,8a 610 and of the two
diphosphaalkenes 7 and 810 (Chart 1).
which could then react at ꢁ78 1C with the appropriate aldehyde
(10 or 20), thus leading to the formation of compounds
1 and 2 in 48% and 78% yield, respectively. It should be noted
that these reactions were carried out in the dark in order to
favour the E-configuration around the PQC bond. The
stereospecificity of the reaction is, besides, confirmed by
31P NMR studies which showed a unique doublet with a JPH
coupling constant lower than 30 Hz (JPH(1) = 25.2 Hz; JPH(2) =
25.1 Hz), corresponding to the (E,E)-isomer.13 No (Z,Z)- and
(Z,E)-isomers were observed in the reaction mixtures.
The new phosphaalkenes (E)-3 and (E)-4 were synthesized
according to the same procedure described above. Synthesis of
4 required the preparation of 1-(4-formylphenyl)-8-phenyl
naphthalene 40 which was obtained via double Suzuki cross-
couplings14 between 1,8-diiodonaphthalene and the boronic
acid derivatives.
2. Results
2.1. Syntheses
2.2. Crystal structures of 1 and 2
The syntheses of the new mono- or diphosphaalkenes are
outlined in Scheme 1.
For 1, fine platelets were formed which diffracted very weakly.
Measurements were carried out at the Swiss-Norwegian
Beamline at the European Synchrotron Radiation Facility.
For 2, the measurements were performed using Mo-radiation.
Details of the refinement are reported in ESI.z
The dialdehydes 4,40-(anthracene-1,8-diyl)dibenzaldehyde
(10) and 4,40-(naphthalene-1,8-diyl)dibenzaldehyde (20) were
obtained by hydrolysis of the corresponding acetal 10 0 and 200,
prepared via Negishi cross-coupling11 between 2-(4-bromo-
phenyl)-1,3-dioxane and the corresponding diiodide molecule.
Bisphosphaalkenes 1 and 2 were then synthesized using the
one pot procedure developed by Yoshifuji et al.12 After a first
lithiation reaction, Mes*PHLi was converted into its silyl
Compound 1 crystallizes in the triclinic system, space group
%
P1; compound 2 crystallizes in the monoclinic system, space
group P21/n. As reported in ESI,z for both compounds, the
two Mes*PQC(H)-C6H5 moieties exhibit the geometrical
characteristics reported for this type of phosphaalkene:15 the
CPC(H)C groups are practically planar, and are oriented
perpendicular to the benzene ring of Mes* (Ph3a or Ph3b here-
after, see Fig. 1), the PC(H)C planes make an angle of B301
t
derivative by reaction with BuMe2SiCl at low temperature.
t
The deprotonation of Mes*PH(SiMe2 Bu) by the second
t
lithiation step afforded the silylphosphide Mes*P(SiMe2 Bu)Li
Scheme 1
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2510–2520 2511

View Article Online
Fig. 1 Crystal structures of 1 and 2 with numbering of the aromatic rings. Hydrogen atoms are omitted for clarity. Thermal ellipsoids are shown
at 50% probability for 1 and 25% probability for 2 (Diffraction measurement was performed at 100 K for 1 and at 293 K for 2, which accounts for
the different degrees of thermal motion.).
Table 1 Reduction potentialsa for some mono- and diphosphaalkenes
with the adjacent phenyl rings (Ph2a or Ph2b hereafter, see
Fig. 1) linked to the anthracene (or naphthalene) ring.
Compound
Epc/V
Epa/V
Epc ꢁ Epa/V
0.10
0.10
Ered/V
The main differences between the conformations of 1 and 2
result from the steric interactions between the phenyl rings
Ph2a and Ph2b: the distance between the centroids of these
two phenyl rings is equal to 5.30 A in 1 and only 3.66 A in 2. In
1 the constraints are, of course, less pronounced than in the
naphthalene containing system: while the x interplane angles
(Ph1a, Ph2a) and (Ph1b, Ph2b) are equal to 52.61 and 47. 91 in
A1, they reach 61.81 and 65.01 in 2. Strong hindrance between
the two encumbered Ph2C(H)PQMes* groups is also revealed
by the fact that the angle between the C1aꢂ ꢂ ꢂC8a and
C1bꢂ ꢂ ꢂC8b directions is more pronounced in 2 (23.41) than
in 1 (7.51). Moreover, the distances between the two phos-
phaalkene bonds are appreciably shorter in 2 (distance
P1aꢂ ꢂ ꢂP1b = 5.620 A, distance C1aꢂ ꢂ ꢂC1b = 4.957 A) than
in 1 (distance P1aꢂ ꢂ ꢂP1b = 5.844 A, distance C1aꢂ ꢂ ꢂC1b =
5.643 A).
1
ꢁ1.82
ꢁ2.00
ꢁ1.72
ꢁ1.90
ꢁ1.89
ꢁ1.84
ꢁ1.89
ꢁ1.82
ꢁ1.80
ꢁ1.59
ꢁ1.69
ꢁ1.77
ꢁ1.95
2
3
4
ꢁ1.97
ꢁ2.01
ꢁ2.15
ꢁ1.93
ꢁ1.69
ꢁ1.78
0.13
0.12
0.33
0.13
0.10
0.09
ꢁ1.91
ꢁ1.95
ꢁ1.99
ꢁ1.87
ꢁ1.64
ꢁ1.74
58a
610
710
810
a
V vs. SCE.
2.4. DFT calculations
ꢁ
The fact that our attempts to get an optimized structure of 2ꢀ
with no imaginary frequency remained unsuccessful prevented
us from using DFT for assessing the reduction mechanism of 2.
2.3. Cyclic voltammetry
2.4.1. Radical anion 1ꢀꢁ. Due to hindered rotation around
the C1a–P1a and C1b–P1b bonds as well as aroundꢁthe
C2a–C1a and C2b–C1b bonds, the radical anion 1ꢀ is
expected to adopt 6 main conformations. Rotations around
the carbon–phosphorus bonds lead to the E and Z isomers,
while rotations around the carbon–carbon bonds lead to
isomers called C (cis, both phosphorus atoms on the same
side of the anthracene plane) or T (trans, a phosphorus atom
on each side of the anthracene plane). The structures of the six
rotamers CEE, CEZ, CZZ, TEE,TEZ, TZZ have been optimized,
conformations of CZE and TZZ are represented in Fig. 4; all
the energies are comprised in a 2.35 kcal molꢁ1 range; TEE is
found as being the most stable rotamer. These energies
together with the geometry parameters are given in ESI.z
Three reduction waves are observed on the voltammogram
obtained with 1 (Fig. 2a). The two first waves are quasi-reversible
and their parameters are given in Table 1. The voltammogram
obtained with 2 (Fig. 2b) is more complex; a first broad
reduction wave is partially overlapped by a second irreversible
wave which precludes a precise determination of Epc. The
voltammogram obtained with 4 is shown in Fig. 3. The
electrochemical parameters of these compounds are given
in Table 1, together with those of 3 whose voltammogram
is shown in ESI.z For the sake of comparison the
values previously reported for the monophosphaalkenes
5 and 6 and the diphosphaalkenes 7 and 8 are also shown in
Table 1.
c
2512 New J. Chem., 2011, 35, 2510–2520
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 2 Cyclic voltammograms of diphosphaalkenes 1 (a) and 2 (b) recorded on a Pt working electrode at 100 mV sꢁ1 in THF with 0.1 M
Bu4NPF6, ref. SCE.
adjacent CCP planes (dihedral angle less than 51 instead of 301
for the crystal structure of neutral 1). The Ph2a and Ph2b rings
make an angle of B 431 with the anthracene plane (instead of
52.61 and 47.91 in the crystal structure of neutral 1). The
distances between the two phosphorus atoms range from
7.518 A (for CZE) to 9.429 A (for TEE), and are considerably
longer than the distance measured for the neutral compound
(5.84 A).
The 31P and 1H isotropic hyperfine coupling constants
have been calculated; as expected they are sensitive to the
conformation of the radical anion and are given in ESIz for
the six rotamers. In Table 2 we report only the couplings larger
than 1.0 G; it is clear that several protons (H11a, H11b, H12,
H9a and H9b, Chart 2) are likely to participate in the observed
1H hyperfine pattern (a(1H) = 2.5 G, vide infra).
Fig. 3 Voltammogram obtained with the monophosphaalkene 4.
2.5. EPR study
2.5.1. Anthracene-containing diphosphaalkene 1. At room
It can be remarked that, in the crystal, both neutral com-
pounds 1 and 2 adopt a CEE conformation.
temperature, reduction of a THF solution of 1 with Na
naphthalenide leads to the EPR spectrum shown in Fig. 5a.
The 1-2-1 triplet structure is attributed to hyperfine interaction
with two 31P nuclei; an additional substructure, probably due
to protons, is characterized by splittings of B2.5 G. The low
For all rotamers the benzene rings of the Mes* groups
remain oriented perpendicular to the adjacent CPC moieties
and the Ph2a and Ph2b rings are almost coplanar with the
ꢁ
Fig. 4 Optimized structures for the conformations CZE and TZZ of 1ꢀ
.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2510–2520 2513

View Article Online
ꢁ
Table 2 Ranges of the calculated isotropic hyperfine splitting constants (Gauss) for the six rotamers of 1ꢀ
1
1
1
Nuclei
31P1a, 31P1b
[9.4 to 13.1]
1H11a, H11b
1H9a, H9b
1H12
1H1a, H1b
1H13
Coupling range
[ꢁ2.9 to ꢁ3.1]
[ꢁ2.2 to ꢁ2.4]
[ꢁ3.1 to ꢁ3.5]
[ꢁ0.6 to ꢁ1.2]
[ꢁ0.9 to ꢁ1.5]
proton structure; this observation seems more in accordance
with the second hypothesis. Moreover, a rapid exchange
between the various rotamers is consistent with DFT results
which indicate that energy differences between these rotamers
are very low. At 110 K, only a very broad line is detected
(Fig. 5b). Simulation of the frozen solution spectrum is given
in ESI.z
The hyperfine pattern observed after electrochemical reduction
of a solution of 1 in THF is quite similar to the structure
observed in Fig. 5a; however, the line-shape appears to be
altered by dynamics effects and perhaps by interaction with the
electrode (spectrum shown in ESIz).
2.5.2. Naphthalene-containing diphosphaalkenes
Compound 2. At 295 K, the spectrum obtained by electro-
chemical reduction of 2 in solution in THF exhibits two lines
of similar intensity and separated by 50.4 G (Fig. 6a). This
EPR response disappears when the voltage is switched off. The
solution turns to blue when electrolyzed. Rapid cooling
(130 K) of the paramagnetic solution hinders molecular motion
and, as shown in Fig. 6b, reveals the anisotropic components of
the EPR tensors. This spectrum is simulated by using the
g and 31P hyperfine tensors, of axial symmetry, reported in
Table 3 (see ESIz). The resulting isotropic hyperfine splitting
constant (48.3 G) agrees with the value measured in the liquid
phase (50.4 G). The same spectra are observed by chemical
reduction at 295 K with Na naphthalenide, the colour of the
resulting solution is dark blue.
Chart 2
field pattern clearly shows that this hyperfine substructure is
composed of 7 transitions.
Since DFT calculations on 1ꢀ indicate that this radical
ꢁ
anion can adopt several conformations of similar energy, it is
rather difficult to accurately analyse the observed hyperfine
structure. Nevertheless, as shown in Fig. 5a, overlaps of the
lines hardly occur on the lateral parts of the spectrum obtained
after reaction with a small amount of reduction agent. The low
field part of the spectrum was therefore used for the determi-
nation of the coupling constants. As shown by spectrum
simulation (see ESIz), the observed pattern (1-6-15-20-15-6-1)
can be explained by assuming the coexistence of two non-
exchanging structures characterized by a coupling with two
31P nuclei (a(31P) = 16.0 G for isomer 1, 13.5 G for isomer 2)
and with five equivalent protons (a(1H) E 2.5 G for isomer 1
and isomer 2). However, as shown in ESI,z the experimental
spectrum can also be simulated by assuming that a rapid
exchange between several rotamers leads to an averaged
coupling a(1H) E 2.5 G with six protons and of 14.9 G with
two 31P nuclei. Decreasing temperature causes a broadening of
the lines but never modifies the relative intensities inside the
When drops of a solution of Na naphthalenide are added, at
243 K, to a solution of 2 in THF the colour of the solution
becomes violet; then, if temperature is increased or if the tube
is stirred the colour turns from violet to blue. The EPR
spectrum obtained with the violet solution exhibits three types
of lines (Fig. 7): two signals L1, L2 separated by 50 G,
a central line marked M and two lines, marked T1 and T2,
separated by 20 G. Increasing temperature causes a rapid
disappearance of the T1 and T2 lines but not of the lines L1,
L2 and M. When Na naphthalenide is added to this solution at
Fig. 5 EPR spectra obtained after reduction at RT of a solution of 1 with Na naphthalenide. (a) Spectrum recorded at 295 K, (b) spectrum
recorded at 110 K (peak-to-peak linewidth = 30 G).
c
2514 New J. Chem., 2011, 35, 2510–2520
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 6 EPR spectra obtained after electrochemical reduction of a solution of 2 in THF (a) spectrum recorded at 295 K, (b) spectrum recorded at
130 K.
Table 3 EPR parameters obtained after reduction of phosphaalkenes containing a naphthalene group 2, 3, and 4
Liquid solution
Frozen solution
g
a(31P)
g>
gJ
31P-T>(G)
31P-TJ
31P-t>
31P-tJ
a(31P)
Reduced solution
2
3
4
2.0075
2.0078
2.0076
50.4
54.0
50.4
2.0085
2.0087
2.0085
2.0055
2.0060
2.0057
3.0
3.0
3.0
142.0
156.0
148.0
ꢁ46.8
ꢁ51.5
ꢁ48.8
93.7
103.0
97.7
48.3
53.0
50.3
Values of a, T and t expressed in Gauss.
3. Discussion
3.1. Reduction potentials of mono- and diphosphaalkenes
The potentials of the first reduction wave of compounds
containing a single phosphaalkene group linked to a phenyl
ring (e.g. 5,8a 6,10 4, Table 1) range from ꢁ1.87 to ꢁ1.99 V.
As shown by the reduction potentials measured for 7 (ꢁ1.64 V)
and 8 (ꢁ1.74 V), the presence of a second *Mes–PQCH group
causes a decrease of at least 0.13 V of the absolute values of
Ered. Consistent with a diphosphaalkene, 1 is reduced at
ꢁ1.77 V. For 2, however, the behaviour seems to be quite
different. Addition of a second Mes*–PQCH moiety to 4 does
not affect the value of Epa and the shape of the voltammogram
does not indicate a decrease of the absolute value of Ered when
passing from 4 to 2. The electrochemical behaviour of 2 seems
therefore to be more related to a monophosphaalkene than to
a diphosphaalkene; we will now check if this property is
corroborated by EPR.
Fig. 7 EPR spectrum obtained at 243 K after reaction, at the same
temperature, of a solution of 2 in THF with Na naphthalenide.
295 K, only the two L lines remain present on the spectrum.
The frozen solution spectrum obtained with the violet solu-
tions not only exhibits two lines similar to PJ1 and PJ2 as
shown in Fig. 6b, but also a broad and intense central pattern.
3.2. Spin delocalisation in mono- and diphosphaalkene radical
anions
Compounds 3 and 4. Both electrochemical reduction (295 K)
and chemical reduction (Na naphthalenide, 285 K) of a
solution of 3 in THF lead to a two-line EPR spectrum
characterized by g = 2.0077 and a 31P hyperfine splitting
constant equal to 54 G. The frozen solution spectrum,
recorded at 110 K, after chemical reduction is readily simulated
by using the parameters reported in Table 3. Electrochemical
and chemical reductions of 4 lead to the same observations as
for 3. Both reduction compounds have very similar EPR
parameters. Decomposition of the 31P hyperfine tensor
(T(31P)) into its isotropic (a(31P)) and anisotropic (t(31P))
components is shown in Table 3, together with the g
anisotropy.
In the ‘‘classical’’ phosphaalkene monoradical anion 5ꢀ
unpaired electron is mainly confined in the PQC p* bond, with
31P hyperfine interaction (a(31P) = 54 G, tJ (31P) = 108 G)
consistent with phosphorus spin density around 45%.
,
ꢁ 8a the
a
a
Increasing the size of the aromatic moiety linked to the Ph
ring in Mes*PQC(H)Ph hardly affects the spin delocalization:
in 6ꢀꢁ 10 and 3ꢀꢁ, the 31P isotropic hyperfine splitting constant
remains close to 54 G; moreover, in 4ꢀꢁ, the presence of a
second phenyl ring bound to the naphthalene has no effect on
the phosphorus spin density. As already reported for 7ꢀꢁand
ꢁ 10
8ꢀ
,
the presence of a second phosphaalkene group causes
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2510–2520 2515

View Article Online
a marked diminution of the 31P coupling (a(31P) = 43 G and
34 G, respectively). In these two radical anions, the unpaired
electron is mainly shared by the two phosphaalkene groups, a
residual part of the spin remains on the bridge and depends
upon the nature of this bridge. In the next paragraph we
examine how this spin sharing between the two phosphaalkene
groups operates for radical anions formed from 1 and 2.
As shown by cyclic voltammetry (Fig. 2a), the first and
second reduction waves of 1 are clearly separated and their
quasi-reversible character suggests that the radical monoanion
can be observed by EPR in solution. Indeed, independently of
the model used for the simulation of the spectrum, the experi-
mental coupling constants are close to the values predicted by
DFT for 1ꢀꢁ. The fact that the calculated Aiso(31P) values are
slightly smaller than the experimental constants has often been
mentioned for other phosphorus radical ions.10,16
Scheme 2
electron by each phosphaalkene moiety and (2) only one of the
two phosphaalkene groups is reduced, the additional electron
remains localized on this group.
The hyperfine interaction (a(31P) E 13 G)ꢁwith twꢁo 31P
The fact that the frozen solution spectrum obtained with
2 corresponds to that of a single monophosphaalkene radical
anion, without any distortion or modification due to dipolar
interaction with the neighbouring reduced (PQC) group is less
compatible with the first hypothesis. In accord with the second
hypothesis, cyclic voltammetry shows that an irreversible
ꢁ
nuclei demonstrates that, as for 7ꢀ and 8ꢀ , in 1ꢀ the
unpaired electron is delocalized on the two phosphaalkene
groups; nevertheless, couplings with protons of the anthracene
ring indicate that an appreciable part of the spin also lies on
this part of the molecule. Clearly, the dihedral angle between
anthracene and the adjacent phenylene rings is sufficiently
small to allow electron communication between the two
phosphaalkene groups. These experimental results are consistent
with tꢁhe SOMO shown, for example, for the TEE conformation
of 1ꢀ (Fig. 8) and are confirmed by the calculated spin
densities: r = 0.20 on each PQC bond, r = 0.10 on each
Ph2 ring, r = 0.40 on the anthracene bridge and zero on the
Mes* groups. It can be remarked that such a delocalization of
the unpaired electron is quite suitable for the trapping of a
Na+ ion inside the ‘‘cavity’’; the formation of this type of
complex is perhaps important for the stabilization of some
conformations.
reduction process, likely associated with the formation of 22ꢁ
,
occurs very close to the first reduction wave (quasi-reversible,
formation of 2ꢀꢁ). This suggests that during the relaxation
process which follows the electron capture by one of the
two phosphaalkene groups (say PA), the steric interactions
between the two (C6H4)–CQP–Mes* chains considerably slow
down the required conformational changes. The large value of
the initial torsion angle x probably hinders spin delocalisation
and ꢁcontributes to an increase of the lifetime of the
PAꢀ configuration and explains why a single 31P coupling
contributes to the hyperfine pattern. At room temperature, as
soon as the second phosphaalkene group (say PB) is reduced,
the system becomes diamagnetic, probably due to the formation
The situation is quite different when the anthracene moiety
is replaced by a naphthalene ring. Only a coupling with a
single 31P nucleus is observed on the spectrum recorded at
room temperature after reduction of 2 and no proton hyperfine
interaction is detected. This 31P isotropic hyperfine splitting
constant (B50 G) is almost equal to that observed after
reduction of 5 and the anisotropic coupling constants confirm
a large participation of a phosphorus p-orbital to the SOMO.
Two hypotheses are consistent with this observation: (1)
formation of a biradical with an exchange coupling constant
J equal to zero; this species would result from the capture of an
of
a P–P bond and a subsequent protonation of the
phosphaalkene carbons. Such a mechanism is illꢁustrated in
Scheme 2 which uses the limit mesomeric form – C(H)–ꢀP–
for the phosphaalkene monoradical anion.
3.3. Reaction intermediates during reduction of 2 at low
temperature
In contrast with 1 which leads to a single type of EPR
spectrum, the nature of the signals detected with 2 is dependent
upon the reaction temperature. We will now examine to what
extent the second hypothesis mentioned above for the
ꢁ
structure of 2ꢀ is consistent with experiments performed at
245 K. We have previously shown, with a two phosphinine
containing macrocycle, that when two unsaturated carbon–
phosphorus bonds are maintained close to each other with an
orientation allowing overlap of the phosphorus pp orbitals, a
reduction process can lead to the formation of a one-electron
P–P bond (Chart 3).17 In the resulting radical anion, the a(31P)
value is small (o2.5 G) and, likely, this coupling could remain
unresolved in a less rigid system.
As shown by the crystal structure, in 2 the initial Pꢂ ꢂ ꢂP
distance is only 5.6 A and the angle between the two CPC
planes is only 181; likely, a rather small change of conformation,
Fig. 8 Representation of the SOMO for the TEE conformation
ꢁ
of 1ꢀ
.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
2516 New J. Chem., 2011, 35, 2510–2520

View Article Online
Chart 3
Fig. 10 Simulation of the spectrum expected for a solution contain-
ꢁ
ꢁ
ing a mixture of 2ꢀ (64%), 2*ꢀ (18%) and 22ꢁ (18%).
species—18%, 64%, 18%, respectively—were estimated by
optimizing the weighting coefficients (see ESIz).
The a(31P) values of the biradical responsible for the signals
T1 and T2 (Fig. 7ꢁ) are closeꢁto the constant measured for the
ꢀ
ꢀ
Scheme 3
monoradicals C1 and B1 and indicate that these signals
.
can be assigned to the dianion 22ꢁ
subsequent to an electron capture, can lead to the formation of
a one-electron P–P bond (species 2*ꢀꢁ) as shown in Scheme 3.
The central line M (Fig. 7), observed at 245 K by chemical
reduction of 2, agrees with this mechanism which gives rise to
the radical 2*ꢀꢁ. It should be remarked, however, that this
central line could also result from a hopping of the unpaired
electron between the two phosphaalkene sites; in this hypothesis
the phosphorus hyperfine splitting constant equal to 25 G (half
the value found for [–PQC]ꢀꢁ) would lead to a 1-2-1 pattern
Working at low temperature increases therefore the life
times of two paramagnetic intermediates, 2*ꢀꢁ and 22ꢁ, which
can lead, after protonation, to the diamagnetic reduction
compound P.
4. Conclusion
The chemical behaviour of a system containing two identical
reduction centers does not only result from the nature of these
centers but is also determined by both their relative orientation
and the orientation of the bridging group. In this context, all
the factors which can affect the conformation of the molecule
can therefore be determining for the evolution of the reduction
process. This is specially the case for fluxional systems containing
cumbersome groups; as shown with 1 and 2, replacement of a
naphthalene group by anthracene considerably modifies the
communicability between the two phosphaalkene moieties,
determines the structure of the monoradical anion and influences
the stability of the biradical dianion. The present work shows
that all these effects affect the spin delocalisation in systems
containing two redox sites bonded through delocalized
bridges and that the nature of these bridges can be crucial
for the control of electronic devices containing multiple
switching units.
ꢁ
with the two external lines hidden by the two lines of 2ꢀ
.
Nevertheless, we favour the formation of 2*ꢀꢁ which is a direct
precursor of the final reduction compound P.
At 245 K, two additional signals, marked T1 and T2 (Fig. 7)
and separated by 20 G, are also detected. They are reminiscent
of unidentified signals already mentioned in a previous study
on systems containing two phosphaalkene groups.10 In the
present study, using the X-Sophe program,18 we have simulated
the spectra of a biradical whose exchange coupling value J is
close to the hyperfine couplings (Aiso) with two 31P nuclei.
Using Aiso values equal to 47 ꢃ 10ꢁ4 cmꢁ1 (49.8 if expressed in
Gauss) and a J value equal to 50 ꢃ 10ꢁ4 cmꢁ1 leads to the
spectrum shown in Fig. 9.
ꢁ
Addition of this spectrum to those simulated for 2ꢀ and
ꢁ
2*ꢀ leads to the spectrum shown in Fig. 10 which is a very
satisfactory simulation of the experimental spectrum shown
in Fig. 7. The relative abundances of the corresponding
5. Experimental
5.1. General
All syntheses were performed in Schlenk-type flasks under dry
nitrogen. Solvents were dried by conventional methods and
distilled immediately prior to use. Tetrabutylammonium
hexafluorophosphate (TBAPF6) was recrystallized twice from
ethanol/H2O (1 : 1) and dried under vacuum. Phenylboronic
acid and 4-formylphenylboronic acid were purchased from
1
Aldrich and used as received. Routine H, 13C{1H} and 31P
spectra were recorded at 25 1C with FT Bruker instruments
(Bruker AMX-300 or Bruker DRX-500 MHz). 1H NMR
spectra were referenced to residual protiated solvents (7.26 ppm
for CDCl3), 13C chemical shifts are reported relative
Fig. 9 Simulation of the spectrum due to the biradical 22ꢁ by
assuming J = 50 ꢃ 10ꢁ4 cmꢁ1
.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2510–2520 2517

View Article Online
3
to deuterated solvents (77.0 ppm for CDCl3), and the
31P NMR data are given relative to external H3PO4. Multi-
plicity is indicated as follows: s (singlet), d (doublet), t (triplet),
m (multiplet), dd (doublet of doublet). In the 13C NMR data
Cq denotes a quaternary carbon atom and C0s indicates a
secondary carbon atom. The infra-red spectra were measured
on a Perkin Elmer Spectrum 100 FT-IR spectrometer with
ATR sampling accessory. Electron impact (EI) mass spectra were
obtained using a Varian CH-4 instrument and Electrospray
ionization (ESI) mass spectra were recorded on a Finnigan
SSQ 7000 spectrometer at the Laboratory of Mass
Spectroscopy of the University of Geneva. Melting points
125.73 (d, o-C of C6H4), JPH = 21.9 Hz), 123.37 (s, C0s of
anthracene), 121.82 (s, m-C of Mes*), 38.32 (s, o-C(CH3)3),
4
35.03 (s, m-C(CH3)3), 33.97 (d, o-C(CH3)3, JP-C = 6.4 Hz),
31.46 (s, m-C(CH3)3). 31P NMR (121.5 MHz CDCl3, 298 K): d
260.95 (d, C(H)QP, 2JPH = 25.2 Hz). EI-MS: m/z (%) 905 (24)
[M]+, 793 (6) [M-2tBu]+, 631 (39) [M-PMes*]+, 357 (46)
[M-2PMes*]+.
5.2.2. 1,8-Bis((4-(E)-(2,4,6-tri-tert-butylphenyl)phosphaethenyl)
n
phenyl)naphthalene (2). A solution of BuLi in hexane (1.6 M,
0.48 mL, 0.77 mmol) was added to a solution of 2,4,6-tri-tert-
butylphosphine (0.215 g, 0.77 mmol) in THF (20 mL) at ꢁ78 1C
in the dark. After 15 min, the solution of tBuMe2SiCl
(0.116 g, 0.77 mmol) in THF (10 mL) was added dropwise at
0 1C and the resulting mixture was warmed up to room
(mp) were measured on a Buchi 510 melting point apparatus
¨
and were uncorrected
n
5.2. Syntheses
temperature for 45 min. BuLi (1.6 M, 0.48 mL, 0.77 mmol)
was added again at 0 1C to the pale yellow reaction mixture.
After 15 min at room temperature, the solution was cooled to
ꢁ78 1C and 20 (0.l00 g, 0.29 mmol) in THF (10 mL) was added
dropwise. After addition, the mixture was warmed up to room
temperature and the solution was stirred 1 h in the dark. The
solvent was evaporated under reduced pressure and the residue
was subjected to column chromatography on silica gel in the
dark (hexane then hexane/CH2Cl2 (9 : 1, v/v)) to give 2 as a
pale yellow powder (Rf = 0.34, hexane/CH2Cl2 (9 : 1, v/v)).
Yield: 78%, 0.200 g, 0.23 mmol. mp = 203–204 1C. 1H NMR
(300 MHz, CDCl3, 298 K): d 7.93 (dd, 2H, H of naphthalene,
Pd(PPh3)4,19 1,8-diiodonaphthalene,20 1,9-diiodoanthracene,21
2-(4-bromophenyl)-1,3-dioxane22 and 1-(4-formylphenyl)-
naphthalene23 were prepared according to methods reported
in the literature. Syntheses and spectroscopic characterizations
of the following compounds are given in ESIz: 1,8-bis(4-(1,3-
dioxan-2-yl)phenylanthracene (10 0), 4,40-(anthracene-1,8-diyl)-
dibenzaldehyde (10), 1,8-bis(4-(1,3-dioxan-2-yl)phenyl)naphthalene
(200), 4,40-(naphthalene-1,8-diyl)dibenzaldehyde (20), (E)-1-[4-
(2-(2,4,6-tri-tert-butylphenyl)phosphaethenyl)phenyl]naphthalene
(3), 1-iodo-8-phenylnaphthalene, 4-(8-phenylnaphthalen-1-yl)-
benzaldehyde (40), and (E)-1-[4-(2-(2,4,6-tri-tert-butylphenyl)-
phosphaethenyl)phenyl]-8-phenyl-naphthalene (4)
4
3JHH = 8.2 Hz, JHH = 1.2 Hz), 7.92 (d, 2H, C(H)QP,
2JPC = 25.3 Hz), 7.53 (t, 2H, H of naphthalene, 3JHH = 7.6 Hz),
7.41 (s, 4H, m-H of Mes*), 7.37 (dd, 2H, H of naphthalene,
5.2.1. 1,8-Bis((4-(E)-(2,4,6-tri-tert-butylphenyl)phosphaethenyl)
phenyl)anthracene (1). A solution of nBuLi in hexane (1.6 M,
0.75 mL, 1.20 mmol) was added to a solution of 2,4,6-tri-tert-
butylphosphine (0.281 g, l.01 mmol) in THF (20 mL) at ꢁ78 1C in
4
3JHH = 7.1 Hz, JHH = 1.2 Hz), 7.08 (dd, 4H, H of C6H4,
3JHH = 8.1 Hz, 4JHH = 2.9 Hz), 6.79 (d, 4H, H of C6H4), 1.49
(s, 36H, o-tBu), 1.43 (s, 18H, p-tBu). 13C{1H} NMR (75.46 MHz,
1
t
CDCl3, 298 K): d 175.99 (d, C(H)QP, JP-C = 34.8 Hz), 153.96
the dark. After 15 min, the solution of BuMe2SiCl (0.197 g,
(s, o-Mes*(C)), 149.45 (s, p-Mes*(C)), 143.17 (d, i-C of C6H4,
2JPC = 7.7 Hz), 140.57 (s, Cq of naphthalene), 139.31 (d, i-C of
Mes*, 1JPC = 53.4 Hz), 138.05 and 137.86 (2s, Cq of naphthalene),
135.24 (s, Cq of C6H4), 130.43 (s, m-C of C6H4), 128.32 and
1.31 mmol) in THF (10 mL) was added dropwise at 0 1C and the
resulting mixture was warmed up to room temperature for 45 min.
nBuLi (1.6 M, 0.75 mL, 1.20 mmol) was added again at 0 1C to the
pale yellow reaction mixture. After 15 min at room temperature,
the solution was cooled to ꢁ78 1C and 10 (0.l50 g, 0.38 mmol) in
THF (10 mL) was added dropwise. Then, the mixture was
warmed up to room temperature and the solution was stirred
1 h in the dark. The solvent was evaporated under reduced
pressure and the residue was subjected to column chromatography
on silica gel in the dark (hexane then hexane/CH2Cl2 (9 : l, v/v)) to
give 1 as a pale yellow powder (SiO2, Rf = 0.3, hexane/CH2Cl2
(9 : l, v/v)). Yield: 48%, 0.170 g, 0.19 mmol. mp = 242–243 1C.
1H NMR (300 MHz, CDCl3, 298 K): d 9.05 (s, 1H, H of
anthracene), 8.57 (s, 1H, H of anthracene), 8.20 (d 2H,
125.15 (2s, C’s of naphthalene), 124.59 (d, o-C of C6H4, 3JPC
=
21.3 Hz), 121.68 (s, m-C of Mes*), 38.24 (s, o-C(CH3)3), 35.03
4
(s, m-C(CH3)3), 33.88 (d, o-C(CH3)3, JPC = 6.4 Hz), 31.51
(s, m-C(CH3)3).31P NMR (121.5 MHz, CDCl3, 298 K): d 256.23
2
(d, C(H)QP, JPH = 25.1 Hz). EI-MS: m/z (%) 305 (20) [M]+,
743 (8) [M-2tBu]+, 611 (50) [M-Mes*]+, 585 (47) [M-PMes*]+,
308 (47) [M-2PMes*]+.
5.3. Crystal structures
5.3.1. Crystals data for 1. Crystals suitable for X-ray
diffraction were grown as yellow elongated plates by slow
diffusion of MeOH into a chloroform solution of the
compound at room temperature. Data were collected at
100 K using synchrotron X-ray radiation (l = 0.70000 A) at
the Swiss-Norwegian Beam Lines (SNBL), ESRF, Grenoble,
France with a marresearch mar345 image plate detector.
Formula: C64H76P2; Mr = 907.25 g molꢁ1; triclinic, space
2
C(H)QP, JPC = 25.3 Hz), 8.04 (d, 2H, H of anthracene,
3
3JHH = 8.5 Hz), 7.62 (dd, 4H, o-H of C6H4, JHH = 8.4,
4JPH = 2.5 Hz), 7.57–7.52 (m, 6H, m-H of C6H4 and H of
anthracene), 7.47–7.44 (m, 6H, H of anthracene and m-H of
Mes*), 1.51 (s, 36H, o-tBu), 1.36 (s, 18H, p-tBu). 13C{1H}
NMR (75.46 MHz, CDCl3, 298 K): d 175.28 (d, C(H)QP,
1JPC = 34.1 Hz), 154.08 (s, o-Mes*(C)), 149.65 (s, p-Mes*(C)),
140.28 (d, i-C of C6H4, 2JPC = 7.8 Hz), 139.98–139.43 (3s, Cq of
%
group P1, a = 9.834(5), b = 10.871(5), c = 25.800(10) A,
1
anthracene), 138.92 (d, i-C of Mes*, JPC = 54.1 Hz), 132.01
(s, Cq of C6H4), 130.43 (d, m-C of C6H4, 4JPC = 1.9 Hz), 129.65
a = 88.42(2), b = 83.77(2), g = 79.06(2)1, V = 2692(2) A3;
Z = 2; Dx = 1.119 g cmꢁ3, m = 0.12 mmꢁ1; F(000) = 980.
Crystal dimensions: 0.20 ꢃ 0.05 ꢃ 0.05. 8700 independent
(s, Cq of anthracene), 127.91–126.97 (3s, C0s of anthracene),
c
2518 New J. Chem., 2011, 35, 2510–2520
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
reflections, 6992 with I 4 2s(I). Goodness-of-fit on F2
=
calculated with the Gaussian03 package29 using the B3LYP
functional30 and the 6-31 + G* basis set.
1.1263; R(I) 4 2s(I) = 0.066; wR2 = 0.094, 595 parameters;
maximum/minimum residual density 0.52/ꢁ0.72 e Aꢁ3. The
crystal structure was solved with SIR200424 and refined with
CRYSTALS25 by full matrix least-squares using anisotropic
thermal displacement parameters for all non-hydrogen atoms.
Hydrogen atoms were included with a riding model. CCDC
reference number: 815021.
Acknowledgements
The authors thank the Swiss National Science Foundation for
financial support.
5.3.2. Crystals data for 2. Single colourless crystals were
obtained by slow diffusion of MeOH into a chloroform
solution of the compound at room temperature: crystal dimensions
References
1 (a) Y.-W. Zhong, N. Villa, J. C. Henderson and H. D. Abruna,
Inorg. Chem., 2009, 48, 991; (b) W. R. Browne and B. L. Feringa,
Chimia, 2010, 64, 398; (c) G. Chen, I. Mahmud, L. N. Dawe,
L. M. Daniels and Y. Zhao, J. Org. Chem., 2011, 76, 2701;
(d) J. Areephong, H. Logtenberg, W. R. Browne and
B. L. Feringa, Org. Lett., 2010, 12, 2132.
2 (a) A. Trabolsi, M. Hmadeh, N. M. Khashab, D. C. Friedman,
M. E. Belowich, N. Humbert, M. Elhabiri, H. A. Khatib,
A.-M. Albrecht-Gary and J. F. Stoddart, New J. Chem., 2009,
33, 254; (b) Q. Zhang, Y. Tu, H. Tian, Y.-L. Zhao, J. F. Stoddart
and H. Agren, J. Phys. Chem. B, 2010, 114, 6561; (c) S. Durot,
F. Reviriego and J.-P. Sauvage, Dalton Trans., 2010, 39, 10557.
3 (a) E. Beer, J. Daub, C. Palivan and G. Gescheidt, J. Chem. Soc.,
Perkin Trans. 2, 2002, 1605; (b) A. W. Misiolek, A. S. Ichimura,
R. A. Gentner, R. H. Huang, V. P. McCaffrey and J. E. Jackson,
Inorg. Chem., 2009, 48, 9005.
4 K. Nikitin, E. Lestini, J. K. Stolarczyk, H. Muller-Bunz and
¨
D. Fitzmaurice, Chem.–Eur. J., 2008, 14, 1117.
5 (a) F. Riobe, N. Avarvari, P. Grosshans, H. Sidorenkova,
´
T. Berclaz and M. Geoffroy, Phys. Chem. Chem. Phys., 2010,
12, 9650; (b) I. Danila, F. Biaso, H. Sidornkova, M. Geoffroy,
M. Fourmigue
2009, 28, 3691; (c) M. Iyoda, M. Hasegawa and Y. Miyake, Chem.
Rev., 2004, 104, 5085; (d) F. Riobe, P. Grosshans, H. Sidorenkova,
0.70 ꢃ 0.50 ꢃ 0.30 mm, C60H74P2, M = 857.20 g molꢁ1
,
monoclinic, space group P121/n1,
a
=
10.185(10),
b = 17.940(5), c = 29.06(2) A, a = g = 90, b = 96.79(5)1,
V = 5272(7) A3, Z = 4, Dx = 1.080 g cmꢁ3, m = 0.12 mmꢁ1
,
F(0000) = 1856. Data were collected at 293 K on a STOE IPDS
2 diffractometer using a Mo Ka X-ray source (l = 0.71073 A)
and a graphite monochromator. 10372 reflections collected with
5 o y o 291, 7342 with I 4 2s(I). Data were integrated using
the program XDS.26 The structure was solved by the charge-
flipping method using Superflip27 and refined anisotropically
on F2 using the CRYSTALS package.25 Hydrogen atoms were
t
included with a riding model. One Bu group is disordered
over two positions. Final results: R(I 4 2s(I)) = 0.050,
wR2 = 0.081, goodness-of-fit on F2 = 0.9830, 587 parameters,
maximum/minimum residual density 0.29/ꢁ0.32
e .
Aꢁ3
´
, E. Levillain and N. Avarvari, Organometallics,
CCDC reference number: 815022.
´
5.4. Electrochemistry and EPR measurements
M. Geoffroy and N. Avarvari, Chem.–Eur. J., 2009, 15, 380.
6 (a) D. Jacquemin, E. A. Perpete, F. Maurel and A. Perrier, J. Phys.
Chem. Lett., 2010, 1, 434; (b) D. Jacquemin, C. Michaux,
E. A. Perpete, F. Maurel and A. Perrier, Chem. Phys. Lett.,
2010, 488, 193.
7 (a) J. Areephong, J. H. Hurenkamp, M. T. W. Milder,
A. Meetsma, J. L. Herek, W. R. Browne and B. L. Feringa, Org.
Lett., 2009, 11, 721; (b) A. Staykov, J. Areephong, W. R. Browne,
B. L. Feringa and K. Yoshizawa, ACS Nano, 2011, 5, 1165.
8 (a) M. Geoffroy, A. Jouaiti, G. Terron, M. Cattani-Lorente and
Y. Ellinger, J. Phys. Chem., 1992, 96, 8241; (b) M. Washington,
J. L. Payton, M. C. Simpson and J. D. Protasiewicz, Organo-
metallics, 2011, 30, 1975; (c) V. B. Gudimetla, L. Ma,
M. P. Washington, J. L. Payton, M. C. Simpson and
J. Protasiewicz, Eur. J. Inorg. Chem., 2010, 854.
9 C. Dutan, S. Shah, R. C. Smith, S. Choua, T. Berclaz, M. Geoffroy
and J. D. Protasiewicz, Inorg. Chem., 2003, 42, 6241.
10 A. Al Badri, A. Jouaiti and M. Geoffroy, Magn. Reson. Chem.,
1999, 37, 735.
11 (a) A. O. King, N. Okukado and E. Negishi, Chem. Commun.,
1977, 683; (b) R. Jana, T. P. Pathak and M. S. Sigman, Chem. Rev.,
2011, 111, 1417; (c) E. Negishi, Q. Hu, Z. Huang, G. Wang and
N. Yin, The Chemistry of Organozinc Compounds, ed.
Z. Rappoport and I. Marek, John Wiley & Sons Ltd., Chichester,
England, 2006, ch. 11, p. 457; (d) E. Negishi, Bull. Chem. Soc. Jpn.,
2007, 80, 233.
Cyclic voltammetry measurements were performed on a
BAS-CV-50W analyzer with a homebuilt one-compartment
cell using a Pt working electrode (1 mm +), a Pt counter
electrode and a Ag/AgNO3 reference electrode. The experiments
were carried out in dry THF using a concentration of
approximately 10ꢁ3 M of the corresponding compound and
0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6)
as a supporting electrolyte. Ferrocene was used as an internal
reference.
EPR spectra were recorded on a Bruker ESP-300 spectrometer
(100 KHz field modulation) equipped with a B-VT 2000 variable
temperature controller. Samples were prepared in a glovebox
using freshly degassed THF (three successive ‘‘Freeze–
Pump–Thaw’’ cycles). Solutions of Na naphthalenide in
THF were used for chemical reductions. Electrochemical
reductions were carried out by electrolysis in situ in the EPR
cavity using a two-Pt electrodes cell together with a Princeton
Applied Research (Model 362) potentiostat. TBAPF6 (0.1 M)
was used as a supporting electrolyte. Electrochemical reductions
at a controlled potential were performed by using a home made
EPR quartz cell made up of a Pt working electrode, a Pt
counter electrode and an Ag wire as a pseudo-reference.
12 M. Yoshifuji, I. Shima and N. Inamoto, J. Am. Chem. Soc., 1981,
103, 4587.
13 H. Kawanami, K. Toyota and M. Yoshifuji, J. Organomet. Chem.,
1997, 535, 1.
5.5. Computational methods
14 (a) N. Miyaura and A. Suzuki, Chem. Rev., 1979, 95, 2457;
(b) A. Suzuki, Angew. Chem., Int. Ed., 2011, 50, 2; (c) J. Hassan,
M. Sevignon, C. Gozzi, E. Schultz and M. Lemaire, Chem. Rev.,
´
ꢁ
Geometries of the six rotamers of the radical anion Aꢀ were
optimized with the Turbomole program28 using the B-P86
functional and the SV(P) standard basis set. Minima were
characterized with harmonic frequency calculations (all
frequencies real). The hyperfine coupling constants were
2002, 102, 1359.
15 R. Appel, J. Menzel, F. Knoch and P. Volz, Z. Anorg. Allg. Chem.,
1986, 534, 100.
16 M. Chentit, H. Sidorenkova, M. Geoffroy and Y. Ellinger, J. Phys.
Chem. A, 1998, 102, 10469.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2510–2520 2519

View Article Online
17 L. Cataldo, S. Choua, T. Berclaz, M. Geoffroy, N. Me
´
zailles,
L. Ricard, F. Mathey and P. Le Floch, J. Am. Chem. Soc., 2001,
123, 6654; S. Choua, C. Dutan, L. Cataldo, T. Berclaz,
27 L. Palatinus and G. Chapuis, J. Appl. Crystallogr., 2007, 40, 786.
28 R. Ahlrichs, M. Bar, M. Haser, H. Horn and C. Kolmel, Chem.
¨
Phys. Lett., 1989, 162, 165.
¨
¨
M. Geoffroy, N. Me
´
Chem.–Eur. J., 2004, 10, 4080.
zailles, A. Moores, L. Ricard and P. Le Floch,
29 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi,V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala,
K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain,
O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
Gaussian 03, Revision B.03, Gaussian, Inc., Pittsburgh, PA, 2003.
30 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 6548; (b) C. Lee,
W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
18 G. R. Hanson, K. E. Gates, C. Noble, M. Griffin, A. Mitchell and
S. Benson, J. Inorg. Biochem., 2004, 98, 903. XSophe-eprview.
A computer simulation software suite for the analysis of EPR
spectra. Bruker BioSpin Karlsruhe, Germany.
19 F. Tellier, R. Sauvetre and J.-F. Normant, J. Organomet. Chem.,
1985, 292, 19.
20 L. Sabater, R. Guillot and A. Aukauloo, Tetrahedron Lett., 2005,
46, 2923.
21 L. Goichi, K. Segawa, S. Suzuki and S. Toyota, Synthesis, 2005,
13, 2116.
22 F. Voegtle, H. Koch and K. Rissanen, Chem. Ber., 1992, 125, 2129.
23 S. L. Goodman, G. Holtzeman, G. A. G. Sulyok and H. Kessler,
¨
J. Med. Chem., 2002, 45, 1045.
24 M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini,
G. L. Cascarano, L. Caro, C. Giacovazzo, G. Polidori and
R. J. Spagna, J. Appl. Crystallogr., 2005, 38, 381.
25 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and
D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
26 W. Kabsch, J. Appl. Crystallogr., 1993, 26, 795.
c
2520 New J. Chem., 2011, 35, 2510–2520
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011