Towards a better understanding of photo-excited spin alignment processes using silole diradicals

Article abstract of DOI:10.1039/b606593g

The synthesis of two nitroxide-based diradicals connected to a 2,3,4,5-tetraphenylsilole (TPS) unit, especially designed to present high spin photo-excited states, is reported. While the bisnitronylnitroxide (NN) silole-based diradical, TPSNN, is unstable

Full text of DOI:10.1039/b606593g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Towards a better understanding of photo-excited spin alignment
processes using silole diradicalsw
Nans Roques,a Philippe Gerbier,*a Yoshio Teki,*b Sylvie Choua,c Petra
´
Lesniakova, Jean-Pascal Sutter, Philippe Guionneau and Christian Guerin
c
d
e
a
Received (in Montpellier, France) 9th May 2006, Accepted 3rd July 2006
First published as an Advance Article on the web 1st August 2006
DOI: 10.1039/b606593g
The synthesis of two nitroxide-based diradicals connected to a 2,3,4,5-tetraphenylsilole (TPS) unit,
especially designed to present high spin photo-excited states, is reported. While the
bisnitronylnitroxide (NN) silole-based diradical, TPSNN, is unstable and experiences a
spontaneous fragmentation of its imidazolinic ring into a iso-butylammomium salt, the
corresponding bisiminonitroxide (IN), TPSIN, is stable both in solution and in the solid state.
This diradical crystallizes in the triclinic P-1 space group with a = 10.984(1), b = 11.474(1),
c = 17.492(1) A, a = 81.10(1), b = 89.01(1), and g = 65.71(2)1. Ground state magnetic properties
of TPSIN have been investigated by means of SQUID and ESR measurements: the diradical
displays weak intramolecular antiferromagnetic interactions (J/kB E ꢀ1 K), in agreement with its
topology and with the molecular packing observed in its crystal structure. In order to investigate
the magnetic photo-excited states of TPSIN, time-resolved ESR experiments (TRESR) have been
performed on this radical species. Despite the presence of both an appropriate topology for the
diradical and a triplet photo-excited state for the TPS coupler, no TRESR signal was observed
for this molecule within the timescale of the measurement. In addition to the work already
published in this field, this result clearly indicates that besides the radical nature, the p-topological
requirements and the need of photo-tunable spin-states for the coupler, the flexibility of the
molecule also plays a crucial role in the achievement of photo-induced spin alignment processes.
yielding a singlet or low spin ground state, have been generally
Introduction
discredited. The recent observation, for purely organic p-
conjugated diradicals, of photo-excited quintet (S = 2) states
starting from singlet ground state has relaunched the interest
in designing systems with the ‘‘wrong’’ topology.6–13 Such
systems are generated through photo-induced spin alignment
using the p-conjugation between dangling iminonitroxide or
verdazyl radicals (S = 1/2) and the photo-excited triplet (S =
1) state of 9,10-diphenylanthracene (Fig. 1). Excited high-spin
systems arising from the radical-triplet pairs have also been
reported in the pioneering works of Corvaja et al.14 and
Yamauchi et al.15 However, the number of references found
in the literature is fairly limited, and in almost all the studies,
stable nitroxide radicals have been used as spin bearing units.
For purely organic excited high-spin systems, only fullerene,
anthracene and pyrene derivatives have been reported as
couplers. The search for novel photo-excited high-spin organic
systems constructed from both different radicals and triplet
couplers is therefore an important research target.
Control of intramolecular spin alignment and exchange inter-
actions in purely organic spin systems are essential topics in
the field of molecule-based magnetism.1,2 Since most studies
are limited to the magnetic ground state, synthetic efforts are
currently devoted to the design of high-spin compounds, in
which an appropriate topology gives rise to ferromagnetic
interactions between the spin bearing units.3–5 Therefore,
topologies expected to allow antiferromagnetic interactions,
a
´
Laboratoire de Chimie Moleculaire et Organisation du Solide-UMR-
CNRS/UM2 5637, Universite´ Montpellier II, C.C.007, Place E.
Bataillon, 34095 Montpellier Cedex 5, France. E-mail:
gerbier@univ-montp2.fr; Fax: +33 4 67 14 3852;
Tel: +33 4 67 143 972
b Department of Material Science, Graduate School of Science, Osaka
City University, 3-3-138 Sugimoto Sumiyoshi-ku, Osaka 558-8585,
Japan. E-mail: teki@sci.osaka-cu.ac.jp; Fax: +81 6605 8585
c
´
Institut Charles Sadron—UPR-CNRS 22, Universite Louis Pasteur,
6 Rue Boussingault, BP 40016, 67083 Strasbourg Cedex, France.
E-mail: turek@ics.u-strasbg.fr; Fax: +33 3 88 414 099;
Tel: +33 3 88 414 000
In this respect, we decided to investigate the potentiality of
the silole to act as a photo-active magnetic coupler. With this
idea, we first studied and evidenced the presence of an
accessible photo-excited triplet state for the 2,3,4,5-tetra-
phenyl-1,1-dimethylsilole (TPS).16 We also reported pre-
viously the synthesis, crystal structure, and ground state
magnetic properties of a series of siloles bearing two nitroxide
radicals,17,18 as well as the magnetic behaviour upon light
irradiation of one of them, the bis tert-butylnitroxide
d Laboratoire de Chimie de Coordination—UPR-CNRS 8241,
´
Universite Paul Sabatier, 205 route de Narbonne, 31077 Toulouse
Cedex 4, France. E-mail: sutter@lcc-toulouse.fr
e
`
Institut de Chimie de la Matiere Condensee—UPR-CNRS 9048,
Universite Bordeaux I, 87, Av. Dr. Schweitzer, 33608 Pessac,
´
´
France. E-mail: guio@icmcb-bordeaux.cnrs.fr;
Tel: +33 5 40 002 279
w The HTML version of this article has been enhanced with colour
images.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1319–1326 | 1319

View Article Online
photo-excited state of the molecule.19 In photo-excited high-
spin states the nature of the magnetic exchange coupling
between two dangling radical spins through the spin coupler
changes from antiferromagnetic to ferromagnetic after photo-
excitation. The key process, which is an enhanced intersystem
crossing (ISC) mechanism is directly related to the attachment
of the radical species. ISC mechanism may be expected if the
spin bearing units are connected through spin nodal sites to
high spin bearing positions of the linker in the photo-excited
T1 state. Since the positions adjacent to the silicon atom are
the best spin populated ones in the T1 state,16 TPSNN and
TPSIN (Fig. 2) were designed and synthesized by considering
the above-mentioned requirements.
Fig. 1 Organic spin systems derived from 9,10-diphenylanthracene in
which the photo-excited spin alignment has been observed.
derivative (TPSNO).16 Briefly, the ground state magnetic
behaviour of TPSNO is characterized by weak intramolecular
antiferromagnetic interactions leading to a singlet ground
state. Surprisingly, and despite the existence of a photo-excited
triplet state for the parent TPS, no higher spin state was
observed for the TPSNO diradical. Assuming that tert-butyl-
nitroxide groups were not adequate for this purpose, we
decided to graft nitronylnitroxide and iminonitroxide spin-
bearing units on the silole ring. The choice of these groups was
motivated by the fact that they have already allowed the
successful spin alignment in the photo-excited states of p-
conjugated diradical systems derived from the 1,9-diphenylan-
thracene pattern.7–13 The design, synthesis, structural and
electronic characterizations of a new silole-linked iminonitr-
oxide diradical are presented here in detail. Its magnetic
behaviour has been investigated both in the ground and
photo-excited states to explore the ability of the silole to act
as a photo-active magnetic coupler.
Synthesis of TPSNN and TPSIN
TPSNN was prepared following the synthetic methodology
reported in Scheme 1.20 It involves a cross coupling reaction
between the organozinc derivative 2 and the Me3Si-protected
dihydroxylamine 4, followed by acidic hydrolysis allowing the
removal of the protecting groups. Oxidation of 5 by phase
transfer reaction using sodium periodate in water/dichloro-
methane afforded the crude TPSNN diradical as a very
hygroscopic deep green solid in 78% yield. Unfortunately
TPSNN was unstable either in concentrated dichloromethane
solutions or in the solid state. Concentrated TPSNN solutions
yield a mixture of products in which a white compound
crystallizes that has been clearly identified as an iso-butyl-
ammonium salt. Some examples of nitronylnitroxide photo-
chemical degradation have already been reported by Ullman
et al.,21,22 but in all cases the 2,3-dimethylbutane pattern has
been identified in the products. To our knowledge, the only
skeleton rearrangement that may be invoked to explain the
formation of iso-butylamine has been proposed to support the
fragmentation mechanism of nitronylnitroxide radicals in
electrospray mass spectrometry experiments.23 As shown in
Scheme 2, the fragmentation pathway upon electrospray
ionization possibly involves the reductive removal of the
oxygen atoms followed by a rearrangement of the transient
Results
Molecular design
p-Conjugated spin systems constructed from aromatic hydro-
carbons and dangling stable radicals are ideal models to study
the relationship between p topology and spin alignment in
photo-excited states. The design of the delocalized p-orbital
network is of the utmost importance in order to try to
anticipate the spin state for both the ground state and the
tetramethylimidazolium cycle into
a dimethylaziridinium
cation. The further cleavage of the benzylic bond yields a
dimethylaziridinium radical cation that may experience a ring-
opening transformation to afford the observed iso-butylamine.
Fig. 2 Expected photo-induced spin alignment and sign inversion of the effective exchange between two dangling radical spins. TPSNN: X is an
oxygen atom, and TPSIN: X is a lone pair. The phenyl rings present at the 3- and 4-positions and the methyl groups borne by the silicon atom have
been removed for the sake of clarity.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1320 | New J. Chem., 2006, 30, 1319–1326

View Article Online
Scheme 1 (i) 4 Np/Li, THF; 4 ZnCl2-TMEDA; (ii) 3 Me3SiCl, 3 Et3N, THF; (iii) PdCl2(PPh3)2, THF; (iv) HCl 0.1 M; (v) NaIO4, CH2Cl2/H2O;
(vi) NaNO2, AcOH, CH2Cl2, H2O; NaIO4, CH2Cl2/H2O.
It is worthy of note that we have no evidence for the validity of
this mechanism in our case, but it has merit in explaining this
very unexpected fragmentation.
As shown in Fig. 4, two C2–H2ꢂ ꢂ ꢂO2–N4 weak hydrogen
bonds (2.51(2) A, 134.2(1)1) lead to the formation of a head-
to-tail silole dimer.24,25 The other nitrogen atoms of the same
iminonitroxide groups are involved in the formation of two
C12–H12ꢂ ꢂ ꢂN3 weak hydrogen bonds (2.64(2) A, 164.8(1)1),
that are connecting the dimers together with C38–H38Aꢂ ꢂ ꢂp
interactions with the phenyl rings connected to C2 to form
infinite supramolecular stair-like tapes.26 It is worthy to note
that N1 and N2 containing iminonitroxide groups are not
involved in any supramolecular interaction with neighbouring
molecules and that the shortest N–Oꢂ ꢂ ꢂN–O distance (6.78(2)
A, N2–O1ꢂ ꢂ ꢂN4–O2) is observed in a dimer. Packing of the
tapes is assumed through C38–H38Bꢂ ꢂ ꢂp and C38–H38Cꢂ ꢂ ꢂp
interactions with phenyl rings in 3,4-positions (four by
molecule, see Fig. 5).
TPSIN has been prepared in good yield by treatment of 5
with sodium nitrite in slightly acidic conditions followed by
oxidation with sodium periodate. In contrast to TPSNN, the
bisimononitroxide radical TPSIN that is stable both in the
solid state and in concentrated solution, was purified by
preparative TLC. Orange spearhead monocrystals suitable
for X-ray analysis were obtained by slow diffusion from a
dichloromethane solution of TPSIN layered with pentane.
X-Ray structure description
TPSIN crystallizes in the P-1 triclinic space group with two
molecules of TPSIN packed in the unit cell. An ORTEP view
is presented in Fig. 3. The silole displays a propeller-like
arrangement of the four benzene rings as usually observed
with the other tetraarylsiloles. The dihedral angles between the
mean plane of the central silole and the 2,5-benzene rings
bearing the nitroxide radicals have values of 54.2(1)1 and
59.8(1)1 which are in the range of those previously reported
for related siloles. The two iminonitroxide rings are nearly
coplanar with the 2,5-phenyl rings and make dihedral angles
with them of 2.3(1)1 and 7.4(1)1, respectively. The N–Oꢂ ꢂ ꢂN–O
intramolecular distance is of 15.14(3) A.
UV-Visible absorption spectra for siloles 5, TPSNN and TPSIN
The UV-Visible spectra of these compounds have been mea-
sured in chloroform. Their data are summarized in Table 1. In
the UV-Visible absorption spectra of siloles, it is known that
the absorption maximum, ascribed to the p–p* transition of
the 2,5-diarylsilole p-conjugated moieties, significantly de-
pends on the nature of the 2,5-diaryl groups and on the nature
Fig. 3 ORTEP view of the molecular structure of TPSIN (thermal
Scheme 2 Proposed electrospray ionization-induced fragmentation
mechanism for nitronyl nitroxide radicals after ref. 23.
ellipsoids set at 50% of probability).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1319–1326 | 1321

View Article Online
Table 1 UV-Visible absorption spectral data for diradicals 5, 6
and 7ab
Silole
p - p*ArꢀNO
p - p*Silole
n - p*NO
5
—
262 (5.47)
258 (4.27)
372 (5.05)
375 (5.16)
370 (3.93)
—
603 (3.79)
483 (2.84)
TPSNN
TPSIN
a
b
10ꢀ3 M solutions in CHCl3. lmax in nm, (log e).
respectively. The spectrum is in agreement with a nitroxide
diradical in which the exchange coupling parameter J is sub-
stantially larger than the nitrogen hyperfine coupling
(|J| c aN). At 4 K, ESR spectra gave broad signals due to
2
4
3
5
C
T
1
ꢀ
ꢁ
IESR
¼
ð1Þ
DETꢀS
3 þ exp
kB
T
Fig. 4 Side view of a dimer, showing NOꢂ ꢂ ꢂH weak hydrogen bonds
(dashed lines) and additional Nꢂ ꢂ ꢂH weak hydrogen bonds yielding a
stair-like arrangement of the dimers to form infinite supramolecular
chains.
weak dipolar coupling of the unpaired electrons including
Dms = 2 transitions at about 1715 G. To determine the nature
of the magnetic interactions, the intensity of the half-field transi-
tion was measured as a function of the temperature (4–30 K). As
the temperature was elevated, the signal due to the triplet
increased in intensity in accordance with the Curie law, indicating
that both the singlet ground state and the thermally populated
triplet state are nearly degenerated in TPSIN diradical (Fig. 6).
The best data fit according to the Bleaney–Bowers model
gives a singlet–triplet energy gap DET–S/kB of 0.9 K, using eqn
(1) to describe the Boltzmann distribution between the two
states where C is a proportionality constant.
and position of the substituents on them. The lmax ascribed to
the p–p* transition vary from the UV region to the visible
region. For TPSNN and TPSIN, the formation of the diradi-
cal species is accompanied by the appearance of two absorp-
tion bands ascribed to the Ar–NO p–p* transitions and to the
N–O n–p* transitions, respectively. Small variations are ob-
served for the siloles p–p* transitions, indicating weak differ-
ences in electronic effects between the parent dihydroxylamine
5 and the corresponding imino and nitronyl nitroxide radicals.
ESR spectra of TPSIN
Magnetic susceptibility of TPSIN
ESR spectra of the nitroxide diradical TPSIN in degassed
dichloromethane solutions were measured in the range
4–298 K. At 298 K, it shows a characteristic thirteen line
spectrum due to hyperfine coupling between the two non-
equivalent pairs of nitrogen atoms. Hyperfine coupling cons-
tants were determined to be aN1/2 = 4.61 G and aN2/2 = 2.26 G,
The static magnetic susceptibility of a polycrystalline sample
of TPSIN was measured in the range 2–300 K with a SQUID
susceptometer at a constant magnetic field of 5000 Gauss. The
temperature dependence of the molar magnetic susceptibility
(wmol) is shown in the form wmol ꢂ T vs. T in Fig. 7. The observed
wmol ꢂ T value is 0.72 cm3 K molꢀ1 at 300 K, suggesting that the
singlet and the triplet states are nearly statistically populated
at ambient temperature and that a slight amount of mono-
radical is present in the sample. However, the wmol ꢂ T product
value is in good agreement with the theoretical value of
0.75 cm3 K molꢀ1 expected for two uncorrelated S = 1/2
Fig. 5 Top view of a dimer interacting through C38–H38ꢂ ꢂ ꢂp inter-
actions (dashed lines) with two siloles belonging to two different
dimers.
Fig. 6 Temperature dependence of the ESR signal intensities of the
DMs = 2 transition for diradical TPSIN.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1322 | New J. Chem., 2006, 30, 1319–1326

View Article Online
Fig. 7 Temperature dependence of the molar magnetic susceptibility
(wmol) as expressed by wmolT vs. T plots for TPSIN. The experimental
data (open squares) and the theoretical behaviour (solid line, see text)
are overlapping.
Fig. 8 Time resolved ESR of TPS in EPA rigid glass 0.4 ms after a
Nd:YAG pulse laser excitation (l = 355 nm). (a) Observed spectrum
at 30 K. (b) Simulation. The spin Hamiltonian parameters used in the
simulation are described in Table 2.
units with g = 2.0.
2Ng2m2B
kBT
1
Table 2 Spin Hamiltonian parameters and relative polarization of
Ms sublevels in zero-field for TPS
w
¼ f ꢂ
ꢂ
ð2Þ
mol
3 þ expðꢀ2J=kBTÞ
Fine-structure
parameters
Relative
polarizations
g value
On decreasing the temperature, the wmol ꢂ T value remains
almost constant in the temperature range 300–10 K, and then
decreases to reach 0.58 cm3 K molꢀ1 as the temperature is
lowered down to 2 K. The temperature dependence of the
wmol ꢂ T value was analysed in terms of a modified singlet–
triplet two state model where the magnetic exchange coupl-
ing constant J corresponds to an Hamiltonian of the form
H = ꢀ2JS1S2. A purity factor f was introduced for
microcrystalline samples of the diradical used for the
magnetic measurements.27 The best fit parameters were J/kB
= ꢀ1.0 K, and f = 0.96 (eqn 2).
2.010
D = 0.076 cmꢀ1
E = 0.002 cmꢀ1
,
PX = 0.2, PY = 0.8,
PZ = 0.0
Discussion
Structurally speaking, diradical TPSIN belongs to a class of
diradicals termed as doubly disjoint by Lahti et al.28 It is
mainly constituted by a non-alternating heteropentacyclic
system (the silole ring) in which spin bearing units are con-
nected to the central ring through sites with low spin density:
weak antiferromagnetic interactions and nearly degenerated
singlet and triplet states are predicted from both valence-bond
and spin polarization theory for this kind of molecule.29 The
intramolecular distance between the two radical centers is
greater than 15 A. In dilute solutions, this distance is too
remote to allow through-space intramolecular magnetic inter-
actions.30 As a consequence, the magnetic interactions ob-
served following the half field signal intensity by ESR traduce
the existence of very weak antiferromagnetic interactions that
Time-resolved ESR experiments
Since the existence of a photo-accessible triplet state for the
silole itself remained elusive, we measured the photo-excited
state of the parent TPS by time-resolved ESR. A typical
TRESR spectrum of the parent TPS is shown in Fig. 8(a).
The observed TRESR spectrum with well-resolved fine struc-
ture splitting has been unambiguously analysed to be an
excited triplet state by the spectral simulation shown in Fig.
8(b). The determined g value, fine-structure parameters, and
relative populations of the Ms sublevels are listed in Table 2.
The observed TRESR spectrum shows the E/A pattern of the
dynamic electron polarization (DEP) (eeaa) (e/a: emission/
absorption of microwave). This indicates that a selective
intersystem crossing by the spin–orbit interaction (SO-ISC)
occurs toward the spin-sublevel, Y, of the zero-field wavefunc-
tions. Despite multiple attempts to observe TRESR signals
characteristics of high-spin photo-excited states for TPSIN, no
signal was obtained studying this diradical in contrast to the
results reported when 9,10-diphenylanthracene was used as a
spin coupler.8,10,11,13
are propagated through the skeleton of the molecule (JESR
1
=
ꢀ /2(DET–S/kB) E ꢀ0.5 K). The slight difference observed
between the JSQUID (ꢀ1.0 K) and JESR values may traduce the
existence of weak additional antiferromagnetic intermolecular
interactions in the solid state. The N–Oꢂ ꢂ ꢂN–O shortest dis-
tances observed in a dimer (6.78 A) are too remote to allow
short contact type magnetic interactions. Magnetic interac-
tions may be expected through hydrogen bonding considering
C–Hꢂ ꢂ ꢂp interactions and C–Hꢂ ꢂ ꢂN or C–Hꢂ ꢂ ꢂO–N hydrogen-
bonded motives. The well known low spin density delocalisa-
tion on methyl groups in imino or nitronyl nitroxide radicals
together with the low spin density delocalisation on the
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1319–1326 | 1323

View Article Online
3,4-phenyl rings allow us to discard this hypothesis.31 Another
way to explain such a slight difference is to consider that in the
solid state the molecular conformation of TPSIN allows better
magnetic exchanges than in frozen solution, in which numer-
ous and less favourable conformations are possible. Whatever
the explanation, J values are in agreement with the values
ascribed to disjoint systems, with the values encountered in the
case of TPSNO (ꢀ4 K) and with the ones obtained when
diphenylanthracene is used as a coupler (ꢀ2.9 K).11,17,18
Turning back to TRESR experiments, we can consider the
following reasons to explain the lack of high-spin photo-
excited states in the case of silole-based diradicals: (1) the
lifetime of the triplet state of the parent silole (TPS) is much
shorter than the ones observed for anthracene derivatives (the
triplet lifetime of anthracene is ca. 55 ms and the triplet
lifetime of TPS is about a few ms), and (2) the efficiency of
the enhanced SO-ISC is much smaller than the one encoun-
tered in diphenylanthracene-bisiminonitroxide. Concerning
the last point, the SO-ISC efficiency highly depends on the
nature of the spin bearing units and the electronic structure of
the triplet photo-excited spin coupler. Since TPSIN has been
designed to present the appropriate topology, and the mole-
cule is built with a coupler that presents both a triplet photo-
excited state and iminonitroxide radicals, it is reasonable to
postulate that the enhanced SO-ISC mechanism is also oper-
ating in the present diradical. Therefore, the only way to
explain the lack of high-spin photo-excited state is to consider
that the strong non-radiative relaxation processes from the
triplet spin coupler make the triplet lifetime shorter, leading to
the unsuccessful detection of high-spin photo-excited states
within the timescale of TRESR measurement. Both factors
arising from the molecular and electronic structures prevent
the effective generation of the expected result. Therefore it is of
crucial importance that the p-conjugated spin coupler pos-
sesses a very rigid skeleton to restrict to a minimum the non-
radiative deactivation channel that may be opened if sets of
molecular motion or vibrations are available. This is why rigid
fused polyaromatic structures such as fullerene, porphyrin,
pyrene, and of course, anthracene, are used to achieve photo-
excited spin alignment.
high-spin states was possible within the timescale of TRESR
measurement. Therefore, TPS-derivatives clearly provide evi-
dence that besides the need of a photo-tunable coupler, specific
radical species, and p-topological requirements, the flexibility
of the coupler skeleton also plays an important role in the
achievement of photo-excited high spin states. Taking into
account that the silole ring, with its unique electronic structure
and its triplet photo-excited state, is all the same a good
candidate to achieve the photo-induced spin alignment, syn-
thetic work is currently under progress to increase the rigidity
of the whole structure by fusing the spin-bearing units to the
central organometallic linker.
Experimental
Materials and methods
The photo-excited states of the siloles were examined by Time-
Resolved ESR (TRESR) at the Department of Material
Science of the Osaka City University. A conventional X-band
ESR spectrometer (JEOL TE300) was used in the measure-
ments of TRESR spectra without field modulation. Excitation
was carried out at 355 nm light using a Nd:YAG pulse laser
(Continuum Surelite II-10, pulse width o7 ns). The typical
laser power used in the experiments was ca. 2–5 mJ. EPA glass
matrix was used for the TRESR experiments. The measure-
ments were carried out at 30 K. The typical microwave power
was ca. 10 mW. The spectral simulation was carried out by the
eigenfield/exact-diagonalization hybrid method,32,33 taking
dynamic electron spin polarization (ESP) into account. The
following ordinary spin Hamiltonian given in eqn (3) was used
for the analysis.
H0spin = beHgS + SDS
(3)
The resonance field BMs2Ms+1(y, f) for each transition was
directly calculated by the eigenfield method.32 The transition
probabilities I (y, f, j) were evaluated by numerically diag-
onalization of the spin Hamiltonian matrix at the calculated
resonance field. Since the resonance field is independent of the
third Euler angle, j, the above procedure practically saves the
computing time for the simulation. The line-shape function of
the TRESR spectrum in the glass matrix is given by
Conclusions
R
R
R
g(B) = NS dj df dy siny PMs2Ms+1(y, f)
I (y, f, j) f[B ꢀ BMs2Ms+1(y, f)]
In this article we have reported the synthesis of two nitroxide-
based diradicals linked by a 2,3,4,5-tetraphenylsilole (TPS)
pattern. These two molecules have been especially designed
and synthesized with the aim of accessing high-spin photo-
excited states for both of these species. Whereas the bisimino-
nitroxide diradical TPSIN is stable either in solution or in the
solid state, its corresponding bisnitronylnitroxide TPSNN
experiences a spontaneous fragmentation to afford a iso-
butylammomium salt. Magnetic behaviour of TPSIN has been
analysed by means of ESR and SQUID measurements com-
bined with structural considerations. As expected from its
doubly disjoint character, the TPSIN diradical displays weak
(4)
In the simulation, the dynamic electron polarization (ESP),
Ms2Ms+1(y, f), on each spin sublevel in a zero magnetic field
P
was given as parameters. Thus, the relative populations of the
Ms sublevels were taken into account as parameters. The
simulation was carried out using a program written by the
author on a personal computer. The details of the simulation
procedures for the high-spin TRESR spectra were described in
our previous articles.
1H, 13C and 29Si NMR spectra were recorded on a Bruker
Advance 200 DPX spectrometer, the FT-IR spectra on a
Thermo Nicolet Avatar 320 spectrometer, the UV-Visible
spectra on a Secomam Anthelie instrument and the MS
spectra on a Jeol JMS-DX 300 spectrometer. The ESR spectra
have been recorded on X-band Bruker Elexsys spectrometer.
intramolecular antiferromagnetic interactions with J/kB
=
ꢀ1.0 K. Although the combination of the silole core with
iminonitroxide radicals would be expected to lead to photo-
induced spin alignment and to high spin states, no detection of
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1324 | New J. Chem., 2006, 30, 1319–1326

View Article Online
Magnetic susceptibility measurements were obtained with a
Quantum Design MPMS-5S SQUID magnetometer. All reac-
tions were carried out routinely under nitrogen using standard
Schlenck techniques. Solvents were distilled prior to use. THF
was dried over sodium/benzophenone, and distilled under
Argon. All the commercial reagents were used as received.
Bis(phenylethynyl)dimethylsilane was prepared by the reac-
tion of dimethyldichlorosilane and phenylethynyllithium,
which was prepared from nBuLi and phenylacetylene in ether.
TPSNN. To a solution of silole 5 (100 mg, 0.14 mmol) in 20
mL of freshly distilled dichloromethane, was added NaIO4
(130 mg, 0.6 mmol) as a solution in 20 mL of distilled water.
The mixture was stirred for 1 h and the phases were separated.
The aqueous phase was extracted with dichloromethane (3 ꢃ
20 mL). The organic layers were mixed and dried other
MgSO4. The solvent was removed under vacuum and the
crude product was purified by preparative thin-layer chroma-
tography (silica gel, eluant; pentane : ethyl acetate 60 : 40) to
give 79 mg of TPSNN (58 mmol) as a very hygroscopic deep-
green solid (78%). Mp: not determined. IR (KBr, cmꢀ1): 1363
(nN–O). UV/Vis (CHCl3): lmax (log e) 262 (5.47, p - p*
arylnitroxide), 375 (5.16, p - p* silole), 603 (3.79, n - p*
N–O). HRMS (FAB+, m-nitrobenzyl alcohol matrix): m/z:
calcd for C44H50N4O4Si [M+ + 3 H]: 726.3601; found
726.3594.
Syntheses
1,3-Bis(trimethylsilyloxylamino)-2-(4-bromophenyl)-4,4,5,5-
tetramethylimidazolidine (4). To a solution of 1,3-dihydroxy-2-
(4-bromophenyl)-4,4,5,5-tetramethylimidazolidine34 (3.15 g,
10 mmol) in THF (30 mL) was added an excess of triethyla-
mine (8.3 mL, 60 mmol). To the reaction mixture was added
chlorotrimethylsilane (7.60 mL, 60 mmol) in THF (50 mL).
After stirring at 45 1C for 20 h, the solvents were evaporated to
yield a residue that was treated with 200 mL of pentane and
filtered. The filtrate was then evaporated in vacuo to yield 1.92
g of 4 as white crystals (42%). Mp: 87–89 1C. IR (CHCl3,
Chemical degradation in solution. Typically, a solution of
TPSNN (0.100 g, 0.14 mmol) in dichloromethane (5 mL)
yields white crystals of iso-butylamine within three days at
room temperature. The crystals are collected on a frit, washed
with dichloromethane and dried in air. Mp: 175 1C. IR (KBr,
cmꢀ1): 3060 (nN–H), 2922 (nC–H), 1572 (nN–C). 1H NMR
1
cmꢀ1): 1373 (nN–O). H NMR (CDCl3, 25 1C): d = ꢀ0.22 (s,
3
3
18H), 1.18 (s, 12H), 4.60 (s, 1H), 7.32 (d, JHH = 8 Hz, 2 H),
(CDCl3, 25 1C): d = 1.02 (d, JHH = 6.8 Hz 6H), 1.63–1.75
(m, 1H), 3.25 (d, 3JHH = 6.5 Hz 2H), 7.3 (s, 2H). MS (FAB+,
m-nitrobenzyl alcohol matrix): [M+]: m/z: 74 (100%).
3
7.48 (d, JHH = 8 Hz, 2 H). 13C NMR (CDCl3, 25 1C): d =
ꢀ0.36, 16.95, 24.20, 61.06, 92.37, 119.18, 121.49, 130.37,
131.71. 29Si NMR (DMSO-d6, 25 1C): d = 22.66. HRMS
(FAB+, m-nitrobenzyl alcohol matrix): m/z: calcd for
C19BrH36N2O2Si2 [M++H]: 459.1481; found 459.1495.
TPSIN. To a solution of silole 5 (100 mg, 0.14 mmol) in 20
mL of freshly distilled dichloromethane, was added NaNO2
(38 mg, 0.6 mmol) as a solution in 20 mL of distilled water.
The mixture was stirred and a few drops of acetic acid were
added to obtain a pH value near 6. After 15 min, NaIO4 (38
mg, 0.6 mmol) was added as a solid and the mixture stirred for
15 min more. The two phases were separated and the aqueous
layer was extracted three times with 20 mL of dichloro-
methane. The organic layers were mixed and dried over
MgSO4. The solvent was removed under vacuum and the
crude product was purified by column chromatography (neu-
tral aluminium oxide, eluant: pentane–dichloromethane 80 :
20) to give 38 mg of TPSIN as red crystals (42%). Mp: 182 1C.
IR (KBr, cmꢀ1): 1368 (nN–O), 1538 (nCQN). UV/Vis (CHCl3):
Silole 5. A mixture of lithium (0.1 g, 14.5 mmol) and
naphthalene (1.85 g, 14.5 mmol) in THF (15 mL) was stirred
at room temperature under argon for 5 h to form a deep green
solution of lithium naphthalenide. To this mixture was added
bis(phenylethynyl)dimethylsilane 1 (1 g, 3.85 mmol) in THF
(10 mL). After stirring for 10 min, the reaction mixture was
cooled to 0 1C and [ZnCl2(tmen)] (tmen = N,N,N0,N0-tetra-
methylenediamine) (3.9 g, 14.5 mmol) was added as a solid to
form organozinc derivative 2. After stirring for 1 h at room
temperature, a solution of 4 (3.5 g, 7.65 mmol) in THF (10mL)
and [PdCl2(PPh3)2] (0.14 g, 0.2 mmol) were successively added.
The mixture was heated under reflux and stirred for 24 h. After
hydrolysis by acetic acid (35%), the mixture was extracted
with Et2O several times. The combined organic layers were
washed with brine, saturated solutions of Na2CO3, dried over
MgSO4 and concentrated. The resulting residue was subjected
to column chromatography (neutral silica, eluant; pentane :
dichloromethane 80 : 20) to give 1.57 g of 5 as a yellow solid
(56%). Dec. temp.: 192 1C. IR (KBr, cmꢀ1): 3529, 3242 (nO–H),
1364 (nN–O). UV/Vis (CHCl3): lmax (log e): 372 (5.05, p - p*
lmax (log e) 258 (4.27, p - p* arylnitroxide), 370 (3.93, p - p*
silole), 483 (2.84, n - p* N–O). HRMS (FAB+, m-nitroben-
zyl alcohol matrix): m/z: calcd for C44H51N4O2Si [M+ + 3 H]:
695.3770; found 695.3770.
Experimental and crystal data for TPSIN. Spearhead single
crystals of approximate dimensions 0.15 ꢃ 0.08 ꢃ 0.08 mm3
were selected on a polarized microscope and mounted on a
Bruker-Nonius k-CCD diffractometer, Mo-Ka radiation
(0.710 73 A). Data collection was performed using mixed f
and o scans, 179 frames of 1.51, 285 seconds per frame and a
distance crystal-detector of 30 mm. The structural determina-
tion by direct methods and the refinement of atomic para-
meters based on full-matrix least squares on F2 were
performed using the SHELX-97.35,36 Results: a = 10.984(1)
A, b = 11.474(1) A, c = 17.492(1) A, a = 81.10(1)1, b =
89.01(1)1, g = 65.70(2), V = 1982.6(9) A3, density(calc.) =
1.161, triclinic P-1, 97.7% completeness to theta 26.481, 12 809
collected data, 8110 independent reflections (Rint = 0.032) for
1
silole). H NMR (CDCl3, 25 1C): d = 0.47 (s, 6H), 1.02 (s,
12H), 1.06 (s, 12H), 4.40 (s, 2H) 6.83–7.06 (m, 8H), 7.09 (m,
6H), 7.25 (d, 3JHH = 8 Hz, 4 H), 7.75 (s, 4H, OH). 13C NMR
([D6]DMSO, 25 1C): d = ꢀ2.80, 17.87, 25.25, 66.89, 91.05,
126.76, 127.17, 128.48, 128.56, 129.01, 130.20, 138.78, 139.79,
140.32, 141.43, 154.29. 29Si NMR ([D6]DMSO): d = 8.05.
HRMS (FAB+, m-nitrobenzyl alcohol matrix): m/z: calcd for
C44H55N4O4Si [M+ + H]: 731.3993; found 731.3995.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1319–1326 | 1325

View Article Online
605 refined parameters, Robs = 0.069, wR2obs = 0.147,
(D/s)max = 0.001, largest difference peak and hole 0.46/
ꢀ0.42 e Aꢀ3, max.
12 Y. Teki and S. Nakajima, Chem. Lett., 2004, 33, 1500.
13 Y. Teki, M. Nakatsuji and Y. Miura, Mol. Phys., 2002, 100,
1385.
14 C. Corvaja, M. Maggini, M. Prato, G. Scorrano and M. Venzin,
J. Am. Chem. Soc., 1995, 117, 8857.
CCDC reference number 603216.
15 K. Ishii, J. Fujisawa, Y. Ohba and S. Yamauchi, J. Am. Chem.
Soc., 1996, 118, 13079.
For crystallographic data in CIF or other electronic format
see DOI: 10.1039/b606593g
16 N. Roques, P. Gerbier, S. Nakajima, Y. Teki and C. Gue
J. Phys. Chem. Solids, 2004, 64, 759.
17 N. Roques, P. Gerbier, J.-P. Sutter, P. Guionneau, D. Luneau and
C. Guerin, Organometallics, 2003, 22, 4833.
18 N. Roques, P. Gerbier, U. Schatzneider, J.-P. Sutter, P. Guion-
neau, J. Vidal-Gancedo, J. Veciana, E. Rentschler and C. Guerin,
´
rin,
Acknowledgements
´
This work was supported by the French CNRS, the European
Commission through the Network of Excellence MAGMA-
Net (NMP3/CT/2005/515767), and the Grant-in-Aid for
Scientific Research on Priority Area ‘‘Application of Molecu-
lar Spin’’ ‘Area 769, Prop. No. 15087208) from MEXT, Japan.
Ph. G. and N. R. want to warmly thank Prof. Philippe Turek
´
Chem.–Eur. J., 2006, DOI: 10.1002/chem.200501280.
19 Y. Teki, Polyhedron, 2005, 24, 2299.
20 S. Yamaguchi, T. Endo, M. Uchida, T. Izumizawa, K. Furukawa
and K. Tamao, Chem.–Eur. J., 2000, 6, 1683.
21 E. F. Ullman, L. Call and S. S. Tseng, J. Am. Chem. Soc., 1975, 75,
1677.
22 L. Call and F. E. Ullmann, Tetrahedron Lett., 1971, 42, 3935.
23 C. Smith, J. Bartley and S. Bottle, J. Mass Spectrom., 2002, 37,
897.
24 G. Desiraju, Crystal Engineering the Crystal as a Supramolecular
Entity Perspectives in Supramolecular Chemistry, Wiley, Chi-
chester, 1996.
(Universite
´
Louis Pasteur, Strasbourg, France), and Prof.
Claude Bernard-Lyon I,
Dominique Luneau (Universite
´
Lyon, France) for their assistance in magnetic characterization
and fruitful scientific discussions.
25 G. Desiraju, Chem. Commun., 2005, 2995.
26 M. Nishio, CrystEngComm, 2004, 6, 130.
27 T. Matsumoto, T. Ishida, N. Koga and H. Iwamura, J. Am. Chem.
Soc., 1992, 114, 9952.
28 P. M. Lahti and A. S. Ichimura, J. Org. Chem., 1991, 56,
3030.
References
1 O. Kahn, Molecular Magnetism, VCH, New York, 1993.
2 P. M. Lahti, Magnetic Properties of Organic Materials, Marcel
Dekker, Inc., New York, 1999.
3 R. Ziessel, C. Stroh, H. Heise, F. Kholer, P. Turek, N. Claiser, M.
¨
Souhassou and C. Lecomte, J. Am. Chem. Soc., 2004, 126, 12604.
4 O. Benedi Borodia, P. Guionneau, H. Heise, F. H. Kholer, L.
¨
Ducasse, J. Vidal-Gancedo, J. Veciana, S. Golhen, L. Ouahab and
J.-P. Sutter, Chem.–Eur. J., 2005, 11, 128.
29 D. A. Shultz, in Magnetic Properties of Organic Materials, ed. P.
M. Lahti, Marcel Dekker, Inc., New York, 1999, pp. 103–125.
30 P. Ferruti, D. Gill, M. P. Klein, H. H. Wang, G. Entine and M.
Calvin, J. Am. Chem. Soc., 1970, 92, 3704.
31 C. Rancurel, H. Heise, F. H. Kohler, U. Schatzschneider, E.
¨
Rentchler, J. Vidal-Gancedo, J. Veciana and J.-P. Sutter, J. Phys.
Chem. A, 2004, 108, 5903.
5 A. Ito, Y. Nakano, M. Urabe, T. Kato and K. Tanaka, J. Am.
Chem. Soc., 2006, 128, 2948.
6 O. Sato, S. Hayami, Y. Einaga and Z.-Z. Gu, Bull. Chem. Soc.
Jpn., 2003, 76, 443.
32 G. G. Belford, R. L. Belford and J. F. Burkhalter, J. Magn. Reson.,
1973, 11, 251.
7 Y. Teki, T. Toichi and S. Nakajima, Chem.–Eur. J., 2006, 12, 2329.
8 Y. Teki, Polyhedron, 2001, 20, 1163.
33 Y. Teki, I. Fujita, T. Takui, T. Kinoshita and K. Itoh, J. Am.
Chem. Soc., 1994, 116, 11499.
9 Y. Teki, M. Kimura, S. Narimatsu, K. Ohara and K. Mukai, Bull.
Chem. Soc. Jpn., 2004, 77, 95.
10 Y. Teki, S. Miyamoto, K. Imura, M. Naktsuji and K. Miura, J.
Am. Chem. Soc., 2000, 122, 984.
11 Y. Teki, S. Miyamoto, M. Naktsuji and Y. Miura, J. Am. Chem.
Soc., 2001, 123, 294.
34 L. Catala, P. Turek, J. Le Moigne, A. De Cian and N. Kyritsakas,
Tetrahedron Lett., 2000, 41, 1015.
35 G. M. Sheldrick, Programs for Crystal Structure Analysis, release
97-2, Institut fur Anorganische Chemie der Universitat, Gottingen,
¨
Germany, 1998.
¨
¨
¨
36 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1326 | New J. Chem., 2006, 30, 1319–1326