Conformational profile, energy barriers and optical properties of quinquethiophene-S,S-dioxides

Article abstract of DOI:10.1016/S0040-4020(02)01398-4

Theoretical calculations, dynamic NMR experiments and absorption and photoluminescence data in solution are reported for a series of quinquethiophene S,S-dioxides substituted with alkyl groups of variable size and steric hindrance. Ab initio B3LYP/6-31G* and force field MM3 theoretical calculations show that the energy barriers for rotation around the inter-ring C-C bonds amount to a few kcal/mol even in the presence of very bulky substituents such as the cyclohexyl group. Dynamic NMR data were in agreement with the results of theoretical calculations. It was found that changing the steric hindrance of the substituents leaves the emission and photoluminesce properties unaltered. However, the photoluminesce intensities and wavelengths of all compounds were found to be very sensitive to solvent variations.

Full text of DOI:10.1016/S0040-4020(02)01398-4

Tetrahedron 58 (2002) 10151–10158
Conformational profile, energy barriers and optical properties
of quinquethiophene-S,S-dioxides
a,*
b
Alessandro Bongini, Giovanna Barbarella, Laura Favaretto, Giovanna Sotgiu,
b
b
b
Massimo Zambianchi and Daniele Casarini
c
a
Dipartimento Chimico G. Ciamician, Universit a` di Bologna,Via Selmi 2, 40126 Bologna, Italy
CNR-ISOF, Via Gobetti 101, 40129 Bologna, Italy
Dipartimento Chimico, Universit a` della Basilicata, Via N. Sauro 85, 85100 Potenza, Italy
b
c
Received 19 July 2002; revised 4 October 2002; accepted 24 October 2002
Abstract—Theoretical calculations, dynamic NMR experiments and absorption and photoluminescence data in solution are reported for a
p
series of quinquethiophene S,S-dioxides substituted with alkyl groups of variable size and steric hindrance. Ab initio B3LYP/6-31G and
force field MM3 theoretical calculations show that the energy barriers for rotation around the inter-ring C–C bonds amount to a few kcal/mol
even in the presence of very bulky substituents such as the cyclohexyl group. Dynamic NMR data were in agreement with the results of
theoretical calculations. It was found that changing the steric hindrance of the substituents leaves the emission and photoluminesce properties
unaltered. However, the photoluminesce intensities and wavelengths of all compounds were found to be very sensitive to solvent
variations. q 2002 Elsevier Science Ltd. All rights reserved.
Thiophene oligomers are among the most studied conju-
gated materials both as models for the corresponding
polymers and as materials interesting in themselves for
might play an important role in determining the photo-
luminescence behavior.
1
–6
application in a variety of organic devices.
However, while many accurate studies on conventional
9
thiophene oligomers have been reported, very little is
Selective oxidation of one or more of the thienyl sulfurs of
conventional thiophene oligomers deeply affects their
known so far about the conformational properties of
oligothiophene-S,S-dioxides. Since a deeper understanding
of the relationships between photoluminescence properties
and molecular structure is crucial to the design and synthesis
of new, higher performance materials, we report in this
paper a theoretical and experimental study of the confor-
mational properties of a homogeneous series of these
compounds. In particular, we have explored the effect
exerted by the introduction of bulky substituents into the
molecular backbone on the inter-ring energy barriers and
conformational mobility.
7
electronic properties, in particular their electron affinity
and their photoluminescence quantum yield in the solid
8
a
state. Recently, widely tunable electroluminescence
emission and good electroluminescence efficiency have
been demonstrated for these types of compounds, owing to
the increased photoluminesce efficiency and electron
8
b–d
affinity.
It has been found that while in the solid state the
photoluminescence quantum yields of oligothiophene-S,S-
8
a
dioxides may attain values up to 70%, in solution their
photoluminescence quantum yields drop by one or two
To this purpose, compounds 1–4—whose molecular
structure is reported in Scheme 1—were taken into account.
Their conformational profile was estimated by the aid of
8d
orders of magnitude. This indicates that the non-radiative
mechanisms for energy relaxation from the excited state of
oligothiophene-S,S-dioxides are different from those of the
corresponding conventional oligothiophenes (which display
high photoluminescence efficiencies in solution but not in
the solid state) and suggests that conformational mobility
p
B3LYP/6-31G ab initio and MM3 theoretical calculations
Keywords: conformation; thiophenes; theoretical studies; optical
properties.
0
Scheme 1. Oligothiophene-S,S-dioxides 1–4. R¼R ¼methyl (1); R¼
0
0
0
*
Corresponding author. Tel.: þ39-051-2099589; fax: þ39-051-2099456;
methyl, R ¼n-hexyl (2); R¼cyclohexyl, R ¼n-hexyl (3); R¼R ¼neopentyl
(4).
e-mail: bongini@ciam.unibo.it
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01398-4

10152
A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
p
9a
on shorter sub-systems and the energy barriers to confor-
mational interconversion compared to those obtained from
low temperature proton NMR experiments.
calculated by us at the B3LYP/6-31G level. We have thus
calculated the inter-ring torsional profile of TOMe at the
same calculation level and the results are illustrated in Fig. 1
together with those of TTMe for comparison. The inter-
2
Finally, absorption and photoluminescence data in two
solvents of different viscosity (methylene chloride and
decalin) are reported for 1–4 and the optical properties
of these compounds discussed in the light of their
conformational characteristics.
ring torsional angle is varying from 08 for fully planar syn
conformation to 1808 for fully planar anti conformation.
The critical points were checked by a frequency analysis.
The minimum points showed no imaginary frequency.
The maximum points showed one imaginary frequency,
which resulted consistent with the inter-ring torsional path.
Fig. 1 shows that, for TOMe, there is a principal minimum
for v¼1588 corresponding to an anti orientation between
the two rings and a less populated minimum for v¼368
corresponding to a syn orientation.
1
. Results
1
1
.1. Theoretical calculations
.1.1. Ab initio torsional profiles and MM3 geometries
and energies of model dimers and trimers. We have
p
recently demonstrated by B3LYP/6-31G ab initio calcu-
lations that, for a correct description of the torsional
potential profiles of thiophene oligomers, correlation energy
contributions have to be taken into account. However,
the size of compounds 1–4 prevents the possibility to
perform ab initio calculations on the entire molecule at such
a level and some approximations are required.
The region between the two anti forms (v¼þ1588 and
21588) is very flat indicating a very low energy barrier to
interconversion between the two forms. The energy barriers
for the interconversion between the two syn forms (v¼þ368
and 2368) and between the syn and anti forms are larger but
still small (1.14 and 3.19 kcal/mol, respectively).
9
a
The comparison of the energy profile of TOMe with that of
TTMe , which is described by two almost equally populated
2
syn and anti conformations, shows that the latter is
characterized by a markedly more twisted geometry than
the former and by a different profile of the energy barriers
Looking at the molecular structure of 1–4 (Scheme 1) it is
seen that there are two different types of junctions between
the aromatic rings, namely the one linking the terminal
thiophene moieties and the one linking the oxygenated
moiety to the adjacent thiophene rings. Such junctions can
(for TTMe the largest energy barrier is that between the
2
two anti forms. The direct rotation from þsyn to 2syn
through a planar cis conformation would necessitate the two
methyl groups passing one into the other and is not a valid
torsional path). However, both TTMe and TOMe do not
2
display energy barriers higher than 5 kcal/mol.
0
0
be modeled by the dimers 3,3 -dimethyl-2,2 -bithiophene
0
(
(
TTMe2) and 3-methyl-2,2 -bithiophene-1,1-dioxide
TOMe), respectively (Scheme 2).
The inter-ring torsional profile of TTMe has already been
2
In order to estimate the trend of variation in energies and
0
geometries as the size of the alkyl groups R and R increases
progressively, MM3 mechanical calculations—which are
known to give reliable geometry and energy values—were
0
carried out for fragments TTR and TOTR (Scheme 3).
2
2
Since the standard MM3 package does not contain the
parametrization constants for the torsional angles between
the thienyl and sulfonyl rings we had to derive such
constants first. The parametrization was carried out in such a
way as to have the best fitting with the ab initio results
shown in Fig. 1. We found that by assigning to the torsional
constants V1, V2 and V3 the values 0,0,5 27,4,5 and 0,99,0
for the atom types 18-2-2-42 (Sox–C–C–S), 42-2-2-42
Scheme 2. Dimers used in ab initio calculations.
3
2
2
2
(
S–C–C–S) and 1-2-2-2 (Csp –Csp –Csp –Csp ),
respectively, a satisfactory reproduction of ab initio
structures was obtained. Indeed, the torsional inter-ring
angles and the relative energies of the two minima of TOMe
Scheme 3. Dimers and trimers used for MM3 calculations. R¼methyl,
0
Figure 1. Ab initio torsional profile of TOMe and TTMe
2
(Scheme 2).
cyclohexyl, neopentyl; R ¼methyl, n-hexyl, neopentyl.

A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
10153
0
Table 1. MM3 energies (E, kcal/mol) of TOTR
2
as a function of the inter-ring angle (v, 8)
0
0
0
0
R methyl
v,v
E
v,v
E
v,v
E
0,0 fixed
79.9
0,0 fixed
86.7
0,0 fixed
95.0
^(38,38)
75.2
^(47,47)
82.4
248,248
83.8
^(38,150)
74.3
^(47,154)
81.3
^(150,150)
73.3
^(154,154)
80.5
180,180 fixed
74.3
180,180 fixed
81.0
180,180 fixed
91.1
0
R n-hexyl
0
R neopentyl
248,þ134
þ134,þ134
83.2
82.6
the two anti–anti forms is very low, not exceeding
0.5 kcal/mol, whereas the energy barrier between the two
syn–syn forms amounts to more than 2 kcal/mol. Due to the
Table 2. MM3 energies (E, kcal/mol) of TTR
angle (v, 8)
2
as a function of the inter-ring
R methyl
v
E
v
E
v
E
v
E
^48
42.0
^61
52.1
þ70
48.5
^111
41.5
180 fixed
49.3
0
larger size of the substituents, for the trimer with R ¼
neopentyl, both energy barriers are markedly higher,
amounting to 5–6 kcal/mol. This behavior can be explained
by taking into account that the neopentyl group has a
preferred conformation in which the t-butyl moieties of the
neopentyl substituents are placed perpendicularly up and
down the plane of the sulfonyl ring, as depicted in Scheme
R cyclohexyl-(eqeq)
R neopentyl (A)
R neopentyl (B)
^110
51.3
180 fixed
66.1
2132
52.5
180 fixed
61.0
^108
48.8
180 fixed
61.1
4
. The rotation of the terminal ring generates two different
energy minima þanti and 2syn corresponding to the forms
characterized by the C–H bonds (2syn) or S atoms (þanti)
of the thiophene ring being on the other side to the face
where the t-butyl group is located.
and TTMe calculated by MM3 are very similar to those
2
calculated ab initio (TOMe conformations: syn v¼328;
trans v¼1648, DE¼1.4 kcal/mol). Moreover, the relative
trends of the energy barriers for the planar conformations of
TOMe and TTMe are reproduced. As a further control, an
2
The þanti to 2syn profile is of course equi-energetic with
the enantiomeric 2anti to þsyn profile in which the t-butyl
residues are on the other side of the sulfonyl ring. As the two
t-butyl residues cannot stay on the same side for steric
reasons, a synchronous rotation of the two neopentyl groups
must occur to pass from one profile to the other.
ab initio calculation on the syn,syn and anti,anti confor-
p
mations of TOTMe was performed at B3LYP/6-31G level
2
and the results (syn,syn v¼þ41,241; anti,anti v¼
þ153,2153; DE¼2.3 kcal/mol) are in good agreement
with those obtained by MM3 calculation (Table 1). MM3
failed in the reproduction of the syn–anti interconversion
energy barriers. The MM3 calculated profile between the
syn and the anti conformers is very flat, and on imposing
fixed inter-ring angles v¼908 the values of the energy
barrier are too low to be meaningful and then are not
reported.
Table 2 shows that also in dimers TTR the anti orientation
2
is always energetically favored over the syn one.
Due to greater steric interactions between the substituents,
the inter-ring junction is more twisted and the energy barrier
between the two anti forms is higher than in the trimers.
Concerning the cyclohexyl substituents, which can be either
linked equatorially or axially to the thiophene rings, Table 2
only reports the energies relative to the diequatorial dimer,
as axial substitutions merely increases the energy of all
maxima and minima by 4 kcal/mol (8 kcal/mol for axial
disubstitution).
0
The results of MM3 calculations on TOTR and TTR are
2
2
reported in Tables 1 and 2, in which the minimum energy
conformations are shown, together with the planar confor-
mations obtained by imposing to the inter-ring angles the
fixed values of 08 or 1808, respectively. Table 1 shows that
in trimers TOTR , the anti,anti conformers are always the
0
2
most stable ones. The energy barrier between the þsyn and
2
syn forms (or between the þanti and 2anti forms) was
The dimer bearing two neopentyl substituents displays two
different torsional profiles depending on whether the t-butyl
moieties, which are placed perpendicularly to the plane of
the thienyl rings, are on the same side (case A) or on
opposite sides (case B), as shown in Scheme 4. In this case,
the two neopentyl groups do not interfere with each other,
obtained dividing by two the difference between the energy
of the syn,syn (or anti,anti) conformation and the energy of
the planar 0,0 (or 180,180) conformation.
0
0
For R ¼methyl and R ¼n-hexyl, the energy barrier between
Scheme 4. Relative orientations of the t-butyl moieties of the neopentyl substituents.

10154
A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
ground state and a CI-Singles correlation method for the
first excited state.
The ground state was found to have an almost planar
geometry, with an inter-ring torsional angle v¼1688 and
˚
an inter-ring distance of 1.428 and 1.445 A for the
thienyl dioxide–thienyl and thienyl–thienyl moieties,
respectively.
The excited state was found to have a full planar geometry
with a quinoid structure mainly localized on the three
central rings as shown, inter alia, by the marked decrease of
the inter-ring distance between the thienyl dioxide and the
thienyl rings, which changes from 1.428 A in the ground
˚
state to 1.403 A in the excited state whilst the thienyl–
thienyl distance is unchanged (Scheme 5).
Scheme 5. Ab initio optimized geometries of the preferred conformation
for the ground and excited states of unsubstituted quinquethiophene-S,S-
dioxide.
˚
and it is possible to pass from the A to the B profile simply
by rotation of one single neopentyl group.
Moreover, the excited state is characterized by a greater
polar character as shown by the change in dipole moments,
which vary from 3.78 D in the ground state to 4.65 D in the
excited state.
1
.1.2. ZINDO/S calculations of absorption and emission
wavelengths of unsubstituted quinquethiophene-S,S-
dioxide. To have an insight into the optical properties of
compounds 1–4 a ZINDO/S analysis was performed on the
ab initio optimized geometry of the all-anti preferred
The ZINDO/S calculated maximum absorption wavelength,
p
l
max¼507 nm, originates from a HOMO–LUMO p–p
transition with the coefficients of the orbitals mainly
localized on the skeleton carbon atoms. In consequence,
the absorption band is very sensitive to torsional dis-
tortions of the ring chain. The lmax value wavelength falls to
8
f
conformation
dioxide (see Scheme 1, R¼R ¼H) taken as model
of unsubstituted quinquethiophene-S,S-
0
compound.
4
40 nm when the calculations are performed on imposing
p
The ab initio calculations were performed at the 6-31G
level, using a hybrid density functional B3LYP for the
to the ab initio calculated ground geometry the MM3
calculated inter-ring torsion angles typical of the substituted
1
Figure 2. Left: experimental H NMR spectra at 400 MHz of the aliphatic region of 4 in CD
2
Cl
2
at different temperatures. Right: simulated spectra.

A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
10155
compounds (v¼1508 and 1108 for the thienyl dioxide–
Table 3. Maximum absorption (lmax, nm) and emission(lPL, nm)
wavelengths, molar extinction coefficients (1, cm
photoluminescence intensities (I) of compounds 1–4 in methylene chloride
2
1
21
M ) and relative
thienyl and thienyl–thienyl junctions, respectively).
(
MC) and decalin (D)
Solvent
The ZINDO/S calculated wavelength emission is l ¼
PL
p
5
74 nm and the emission band has the same p–p character
l
max
1
l
PL
I
as that of the absorption band.
.2. Dynamic NMR
Samples in CHF Cl–CF Cl (1:1) were prepared for
compounds 1–3 and H VT experiments carried out, but
in all cases the signals pertaining to the thienyl hydrogens
remained substantially unchanged even at 21408C except
that for 1, which precipitated below T¼2808C.
1
2
3
4
MC
D
MC
D
MC
D
MC
D
445
444
453
451
452
450
444
439
22387
25704
23988
22387
24547
26915
19498
19953
591
570
590
569
581
554
580
557
1
20
2
32
4
35
3
22
1
2
2
2
1
molar absorption coefficients (1) and the photoluminescence
5
(
PL) intensities measured for 102 M solutions. The two
2
4
Compound 4 was dissolved in CD Cl (<10 M) and
2
2
solvents were chosen because of their very different
polarities and viscosities (dielectric constant 8.9 and 2.2;
dipole moment 1.6 and 0 D; viscosity 0.4 and 2–3 cp,
respectively).
slowly cooled in several steps down to 2958C. The changes
observed in the proton spectra are shown in Fig. 2.
At þ258C, the proton spectrum of 4 showed only half
signals for the ring hydrogens and the neopentyl groups in
the molecule, thus indicating the presence of symmetry
element that, bisecting the central ring, divides the molecule
in two equivalent halves.
The most remarkable feature shown in Table 3 is that there
is one order of magnitude increase in the photoluminescence
intensities of 1–4 on going from methylene chloride to
decalin while the wavelengths, l , decrease by 20–30 nm,
PL
contrary to lmax which is unaffected by the change of
solvent. Since the concentrations of 1–4 are the same in
both solvents and the molar absorption coefficients are also
very similar, it is likely that the increase in photolumi-
nescence intensities is mainly due to the increase in
photoluminescence quantum yields.
The two sharp signals at 2.38 and 2.40 ppm were assigned to
the methylenes of the neopentyl groups bonded to the outer
thienyl rings by comparison with the observed shifts of the
0
0
neopentyl methylenes in the 3,3 -dineopentyl-2,2 -bithio-
8
a
phene. The remaining broad signal at 2.80 ppm was
assigned to the methylene of the neopentyl bonded to the
central ring.
On lowering the temperature, the broad methylene signal at
2
2. Discussion
.8 ppm decoalesces at 2158C and splits into a sharp AB
quartet (J¼14.0 Hz) below 2608C. In addition, the other
Theoretical calculations show that quinquethiophene-S,S-
dioxides are characterized by very high conformational
mobilities. These kinds of compounds are more planar than
the corresponding conventional oligothiophenes but the
energy barriers for conversion from one form to the other
are very low and they exist in solution as families of rapidly
interconverting conformers of nearly the same energies.
two methylene signals broaden, at 2408C, and split, below
2
608C, into two sharp and overlapped AB quartets (J¼
1
3.8 Hz).
This indicates that the neopentyl groups are tilted from
coplanarity with the averaged molecular plane of the
thiophene rings and that the barriers to rotation about the
bonds methylene–thienyl are quite high. Indeed, by
According to theoretical calculations, the energy barriers for
ring rotation around the C–C bonds linking the thienyl
dioxide moiety and the adjacent thienyl rings are very low.
The energy barriers for ring rotation around the C–C bond
linking the terminal thienyl rings are higher, but these
values are still low enough to allow the molecules to be
conformationally mobile at room temperature despite the
presence of bulky substituents.
simulating the changes of the CH line shapes with the
2
temperature the kinetic constants were achieved and then an
energy barrier of 11.50^0.15 kcal/mol involved in the
observed rotational process of the neopentyl groups was
obtained.
1
In order to slow down other rotational processes H spectra
of 4 in CHF Cl–CF Cl (1:1) were taken at much lower
2
2
2
temperatures. At 21458C, the minimum temperature
reached owing to solubility problems, the aromatic and
the t-butyl signals remained substantially unchanged
suggesting that such rotational barriers are lower than
Low temperature dynamic proton NMR experiments
confirmed the results of theoretical calculations. Indeed,
even at temperatures as low as 21408C no freezing of the
conformational equilibria was observed. The only rotations
frozen at low temperature were the rotations of the
methylene groups of the neopentyl substituents, shown in
Fig. 2 and easily interpreted with the aid of MM3
calculations. The spectra of Fig. 2 show that the four
neopentyl substituents of compound 4 experience similar
steric environments and MM3 calculations allow it to be
established that the origin of the energy barriers measured
5
kcal/mol.
1
.3. Optical properties
Table 3 shows the absorption and photoluminescence
wavelengths of compounds 1–4 in two different solvents,
methylene chloride (MC) and decalin (D), together with the

10156
A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
by proton NMR is due both to steric interactions occurring
between the bulky neopentyl groups and the thienyl rings
and to steric interactions occurring between the neopentyl
substituents themselves. Although these steric interactions
also contribute to the distortions of the rings from
coplanarity, conferring on the molecule a helix-like shape,
they do not prevent the molecule from interchanging
between the different forms of the whole family of syn–
anti arrangements of adjacent rings.
excited state is mainly localized on the oxygenated
8
e
moiety. Ab initio calculations confirm this finding also
for 1–4 compounds, but, even if the quinoid form assumed
in the excited state is more rigid than the aromatic structure
of the ground state, these compounds are still sufficiently
conformationally mobile around the central oxygenated part
to have an efficient non-radiative relaxation. Only the
increase in the viscosity of the medium has the effect of
slowing down the conformational mobility and therein
strongly increases the PL efficiency. Energy barriers
induced by the substituents are not sufficiently high to
hinder the conformational mobility. Only strongly hindering
groups as t-butyl can freeze the conformational mobility,
but they are known to prevent the planarity, and hence the
conjugation, in the molecule and cut off all luminescent
Table 3 shows that the different size of the substituents does
not substantially affect the spectroscopic properties of 1–4
compounds.
The high values of the molar extinction coefficients of the
p
absorption bands indicate the p–p nature of the transition
8a
effects.
that ZINDO/S calculations show to be HOMO–LUMO in
g,10
8
character, as found in trimers analogues.
The wave-
lengths of the maximum absorption of p bands is very
sensitive to distortions from planarity of the conjugated
system. The invariance of lmax in 1–4 shows that the inter-
ring twisting induced in these compounds by the sub-
stituents is almost the same.
3. Conclusions
Theoretical and experimental data show that the confor-
mational mobility of substituted quinquethiophene-S,S-
dioxides is not blocked even by the use of bulky substituents
such as cyclohexyl or neopentyl groups. As oligothiophene-
S,S-dioxides are known to be relaxed from the excited state
by IC pathways, the conformational mobility of 1–4
compounds makes this mechanism very efficient and
strongly decreases the photoluminescence of these com-
pounds in solution. The size of the alkyl substituents is
therefore not very important for the photoluminescence
efficiency of these compounds. Our data suggest that it is the
viscosity of the solvent which, reducing the conformational
mobility, plays the most important role in determining the
photoluminescence properties.
In compounds 1–4, the similarity of lPL in the same solvent
shows that also in the excited state the different size of the
substituents does not affect the inter-ring energy barriers and
the geometry of the molecular backbones. The strong red
shift (5600 cm2 ) found between absorption and emission
bands indicates a more planar structure in the excited state,
in agreement with ab initio calculations which predict a
planar quinoid geometry for the excited state of the model
unsubstituted pentamer (Scheme 5). Moreover, the strict
concordance observed between the ZINDO/S lPL value
calculated on the ab initio geometry (574 nm) and the
experimental values found for 1–4 in decalin (550–570 nm)
further supports the planar structure of the excited state.
1
4. Experimental
4.1. Materials
The red shift of about 600 cm2 of lPL found on going from
decalin to methylene chloride for all compounds indicates
the occurrence of some interaction between the solute and
the solvent in the excited state. Ab initio calculations show
that the excited state is more polar than the ground state.
Upon excitation, the smaller and more polar molecules of
dichloromethane have time to rearrange around the excited
1
All compounds were prepared by reacting 2,5-dibromo-
8
f
thiophene-S,S-dioxide or 2,5-dibromo-3,4-dialkyl-thio-
d,f
8
phene-S,S-dioxide
with the appropriate monostannane
according to the modalities described in Ref. 2. Compounds
1,4 are described in Refs. 2,4, while compound 2 is
described in Ref. 8d. The synthesis of compound 4 and of
the corresponding intermediates, compounds 5–8, is
reported below.
1
1
molecule, some energy is dissipated and lPL increases.
The most important variation observed for all compounds
on passing from methylene chloride to decalin is the marked
increase in PL intensities, independent of the size of the
substituents. It is known that the dominant non-radiative
relaxation pathway for oligothiophene-S,S-dioxides is
internal conversion (IC) rather than inter system crossing
4.1.1. 3-Cyclohexyl-thiophene (5). The title compound
was prepared by the Grignard reaction between 3-bromo-
thiophene and 2-cyclohexyl-magnesium bromide (from
2-cyclohexyl bromide, Aldrich) in the presence of nickel
diphenylphosphinopropane, Ni(dppp)Cl (Aldrich). Yield
2
65%; pale yellow oil; MS (70 eV, EI): m/e 166 (M ). H
8
e,10
(
ISC).
The IC mechanism is strongly dependent on the
zþ
1
coupling with vibro-rotational quanta of the excited state.
In these molecules, the inter-ring torsional twisting mode
is the coupling type that is to be considered. The low
photoluminescence intensities measured in dichloro-
methane for 1–4 show that in these compounds the IC
mechanism is very efficient. It is known that, when PL
depends on IC, PL quantum yields of flexible molecules
increase in viscous media as decalin. In a previous work we
could infer that in oligothiophene-S,S-dioxides the first
NMR (CDCl ,TMS): d (ppm)¼7.23 (m, 1H); 6.98 (m, 1H);
3
6.91 (m, 1H), 2.6 (m, 1H); 2.0 (m, 2H); 1.75 (m, 3H); 1.3
(m, 5H). C NMR (CDCl , TMS): d (ppm)¼149.1; 127.0;
1
3
3
124.9; 118.3; 39.6; 34.3; 26.6; 26.2. Anal. calcd for
C H S: C, 72.23; H, 8.49. Found: C, 72.31; H, 8.52.
1
0 14
4.1.2. 2-Bromo-3-cyclohexyl-thiophene (6). The title
compound was prepared by reacting compound 5 with

A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
10157
1
2
N-bromosuccinimide (Lancaster) in acetic acid–methylene
chloride 1:1 (v:v). Yield 88%; pale yellow oil; MS (70 eV,
EI): m/e 244, 246 (M ). H NMR (CDCl , TMS): d
3
series of programs. The geometries were fully optimized
by standard gradient techniques and the critical points
checked by frequency analysis. Molecular Mechanics
zþ
1
1
3
(
(
ppm)¼7.18 (d, 1H, J¼5.8 Hz); 6.82 (d, 1H, J¼5.8 Hz); 2.7
calculations were carried out using the MM3(96) program
1
3
m, 1H); 1.80 (m, 5H); 1.3 (m, 5H). C NMR (CDCl ,
3
parametrized as described in Section 1. UV transitions
were calculated by ZINDO/S-C.I. (10x10) single point
calculations on ab initio geometries using HyperChem
TMS): d (ppm)¼146.7; 126.0; 125.3; 107.7; 39.0; 33.2;
2
Found: C, 49.12; H, 5.36.
6.6; 26.1. Anal. calcd for C H BrS: C, 48.99; H, 5.34.
10 13
1
4
integrated package.
0
0.0025 mol) of magnesium turnings contained into a dry
0
4
.1.3. 3,3 -Di-cyclohexyl-2,2 -bithiophene (7). To 0.06 g
4.3. NMR measurement
(
flask under N stream was added first anhydrous diethyl
ether (3 mL) and then 6 (1.0 g, 0.004 mol) in diethyl ether
1
The H NMR spectra at variable temperatures were
2
recorded with a Varian-Mercury spectrometer operating at
400 MHz. The samples were prepared under vacuum by
condensing the appropriate quantity of CHF Cl and CF Cl
2
(
Ni(dppp)Cl (0.33 g, 0.0006 mol) was added stepwise and
1 mL) and the mixture was refluxed for 1 h. Then
2
2
2
the mixture refluxed for about 4 h. Afterwards the mixture
was washed with water, the organic layer separated, dried
over Na SO and chromatographed on silica gel using
in an NMR tube immersed in liquid nitrogen and then
sealing the tube with a flame. In the VT experiments the
samples were dissolved in CD Cl and cooled to the
2
4
2
2
cyclohexane as the eluant. 0.37 g (55% yield) of pure 7 were
obtained. White solid, mp 968C; MS (70 eV, EI): m/e 330
appropriate temperature by a flow of dry nitrogen precooled
in the standard spectrometer heat exchanger immersed in
liquid nitrogen. The temperatures were checked by
substituting the sample before the measurements with a
calibrated thermocouple in order to reduce the experimental
error, that is estimated to be in the range of ^0.58C. Line
zþ
1
(
(
(
M ); l
(CH Cl )¼267 nm. H NMR (CDCl , TMS): d
max
2
2
3
ppm)¼7.31 (d, 2H, J¼5.2 Hz); 7.04 (d, 2H, J¼5.2 Hz); 2.6
1
3
m, 2H); 1.80 (m, 10H); 1.3 (m, 10H). C NMR (CDCl ,
3
TMS): d (ppm)¼147.5; 127.5; 126.1; 125.4; 38.1; 34.5;
1
2
Found: C, 80.37; H, 8.81.
6.7; 26.1. Anal. calcd for C H S : C, 80.48; H, 8.78.
2
shape simulations of the H traces were performed by using
0 26 2
1
5
a PC version of the DNMR-6 programme and the best fit
was visually judged by superimposing the simulated and
experimental traces.
0
0
4
.1.4. 5-Tributylstannyl-3,3 -di-cyclohexyl-2,2 -bithio-
phene (8). To a solution of 7 (0.4 g, 0.0012 mol) in
anhydrous THF (3.5 mL) at T¼2708C was added dropwise
BuLi (2.5 M in hexane, 0.48 mL, 0.0012 mol) and the
mixture stirred for 1 h. Then the temperature was raised to
room temperature and, after 2 h, 2-tributylstannyl chloride
4.4. Absorption and photoluminescence measurements
Absorption and photoluminescence spectra were recorded
using a Perkin–Elmer Lambda 20 and a Perkin–Elmer LS
50 B spectrometer, respectively. The concentrations were in
(
0.406 g, 0.00125 mol) was added dropwise and the mixture
2
6
stirred overnight. Then the crude product was chromato-
graphed on Al O using hexane as the eluant. 371 mg (50%
yield) of a yellow oil was obtained. MS (70 eV, EI): m/e 619
the range 2–5£10 M in CH Cl (absorbance 0.1–0.2).
2
2
The excitation wavelengths were in the range 440–450 nm
for all compounds.
2
3
zþ
1
(
(
M ); l
(CH Cl )¼275 nm. H NMR (CD Cl , TMS): d
max
2
2
2
2
ppm)¼7.27 (d, 1H, J¼5.0 Hz); 7.01 (s, 1H), 6.99 (d, 1H,
1
3
J¼5.0 Hz); 1.8–1.1 (m, 38H); 0.9 (m, 9H). C NMR
Acknowledgements
(
CDCl , TMS): d (ppm)¼148.3; 147.0; 136.9; 134.5; 133.2;
3
1
2
6
28.3; 126.2; 125.0; 38.1; 34.6; 34.5; 29.1; 27.3; 26.8; 26.7;
6.2; 26.1; 13.9; 11.0. Anal. calcd for C H S Sn: C,
The present work was partially supported by the project
‘Nuovi emettitori di luce a semiconduttore organico’ (CNR-
3
2 52 2
2.03; H, 8.46. Found: C, 61.88; H, 8.48.
5% Nanotecnologie).
0 000 0000 00 00
.1.5. 3,3 ,4 ,3 Tetracyclohexyl-3 ,4 -di(n-hexyl)-
,2 :5 ,2 :5 ,2 :5 ,2 -quinquethiophene-1 ,1 -dioxide
4
2
0
0
00 00 000 000 0000
00 00
(
coupling of 8 with 2,5-dibromo-3,4-dihexyl-thiophene-1,1-
dioxide according to the method described in Ref. 8g.
Yield: 90%. Orange solid, mp 1858C; MS (70 eV, EI): m/e
4). The title compound was obtained from the cross
References
8
g
1. Handbook of Conductive Polymers; Skotheim, T. A.,
Elsenbaumer, R. L., Reynolds, J. R., Eds.; Marcel Dekker:
New York, 1998.
zþ
1
9
40 (M ); l
(CH Cl )¼455 nm. H NMR (CDCl3,
max
2
2
TMS): d (ppm)¼7.69 (s, 2H), 7.35 (d, 2H, J¼5.2 Hz); 7.04
2. Electronic Materials: The Oligomer Approach; M u¨ llen, K.,
Wegner, G., Eds.; Wiley-VCH: New York, 1998.
3. Handbook of Oligo and Polythiophenes; Fichou, D., Ed.;
Wiley-VCH: New York, 1999.
(
6
1
3
d, 2H, J¼5.2 Hz); 2.65 (m, 6H); 1.9–1.1 (m, 56H); 0.9 (m,
1
3
H). C NMR (CDCl , TMS): d (ppm)¼148.6; 148.2;
3
36.4; 130.8; 129.9; 128.8; 127.8; 126.5; 126.3; 126.2; 38.3;
8.2; 34.6; 34.3; 31.5; 29.8; 28.5; 27.4; 26.6; 26.1; 26.0;
2.7; 14.2. Anal. calcd for C H O S : C, 82.70; H, 9.42.
4. Bao, Z.; Rogers, J. A.; Katz, H. E. J. Mater. Chem. 1999, 9,
1895–1904.
5. Garnier, F. Acc. Chem. Res. 1999, 32, 209–215.
2
5
6 76 2 5
Found: C, 82.55; H, 9.39.
6
7
. Mitschke, U.; B a¨ uerle, P. J. Mater.Chem. 2000, 10,
1471–1507.
. (a) Charas, A.; Morgado, J.; Marthinho, J. M. G.; Alc a´ cer, L.;
Cacialli, F. Chem. Commun. 2001, 1216–1217. (b) Barbarella,
4
.2. Computational methods
Ab Initio calculations were performed using the Gaussian98

10158
A. Bongini et al. / Tetrahedron 58 (2002) 10151–10158
G.; Favaretto, L.; Zambianchi, M.; Pudova, O.; Arbizzani, C.;
Bongini, A.; Mastragostino, M. Adv. Mater. 1998, 10,
51–554.
10. Della Sala, F.; Heinze, H. H.; G o¨ rling, A. Chem. Phys. Lett.
2001, 339, 343–350.
11. Lanzani, G.; Cerullo, G.; De Silvestri, S.; Barbarella, G.;
Sotgiu, G. J. Chem. Phys. 2001, 115, 1623–1625.
5
8
. (a) Barbarella, G.; Favaretto, L.; Sotgiu, G.; Zambianchi, M.;
Bongini, A.; Arbizzani, C.; Mastragostino, M.; Anni, M.;
Gigli, G.; Cingolani, R. J. Am. Chem. Soc. 2000, 122,
12. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A. Jr; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.;
Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian98,
Revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998.
1
1971–11978. (b) Gigli, G.; Ingan a¨ , O.; Anni, M.; De
Vittorio, M.; Cingolani, R.; Barbarella, G.; Favaretto, L.
Appl. Phys. Lett. 2001, 78, 1493–1495. (c) Gigli, G.;
Barbarella, G.; Favaretto, L.; Cacialli, F.; Cingolani, R.
Appl. Phys. Lett. 1999, 75, 439–441. (d) Barbarella, G.;
Favaretto, L.; Sotgiu, G.; Zambianchi, M.; Fattori, V.; Cocchi,
M.; Cacialli, F.; Gigli, G.; Cingolani, R. Adv. Mater. 1999, 11,
1375–1379. (e) Barbarella, G.; Favaretto, L.; Sotgiu, G.;
Antolini, L.; Gigli, G.; Cingolani, R.; Bongini, A. Chem.
Mater. 2001, 13, 4112–4122. (f) Barbarella, G.; Favaretto, L.;
Sotgiu, G.; Zambianchi, M.; Antolini, L.; Pudova, O.; Bongini,
A. J. Org. Chem. 1998, 63, 5497–5506. (g) Barbarella, G.;
Favaretto, L.; Sotgiu, G.; Zambianchi, M.; Arbizzani, C.;
Bongini, A.; Mastragostino, M. Chem. Mater. 1999, 11,
2
533–2541.
. (a) Bongini, A.; Bottoni, A. J. Phys. Chem. A 1999, 103,
800–6804, and references therein. (b) Diaz-Quijada, G. A.;
Weinberg, N.; Holdcroft, S.; Pinto, B. M. J. Phys. Chem. A
002, 106, 1266–1276, and referencestherein. (c) Diaz-Quijada,
G. A.; Weinberg, N.; Holdcroft, S.; Pinto, B. M. J. Phys. Chem. A
002, 106, 1277–1285, and references therein.
1
3. Allinger, N. L. QCPE; Indiana University: Bloomington, IN
7405, USA.
9
4
6
1
4. HyperChem rel 7.0 from HyperCube: Waterloo, Ont., Canada.
5. QCPE program n. 633; Indiana University: Bloomington,
USA.
1
2
2