Toward a switchable molecular rotor. Unexpected dynamic behavior of functionalized overcrowded alkenes

Article abstract of DOI:10.1021/jo962210t

In an approach toward a photochemically bistable molecular rotor the synthesis of cis-1a and trans-1b isomers of 2-(2,6-dimethy1phenyl)-9-(2',3'- dihydro-1'H-naphtho[2,1-b]thiopyran-1'-ylidene)-9H-thioxanthene (1), being sterically overcrowded alkenes functionalized with an o-xylyl group as a rotor, is described. The key stops in the synthesis are a Suzuki coupling to attach the xylyl moiety and a diazo-thioketone coupling with subsequent desulfurization to introduce the central olefinic bond in 1. The X-ray structure of cis-1a revealed an anti-folded helical conformation. Dynamic NMR studies on both isomers were performed, to elucidate the kinetics of their rotation processes and to investigate the possibility to control the biaryl rotation by photochemical cis-trans isomerization. A rotation barrier was found by coalescence spectroscopy for cis-1a: ΔG(c)(+) = 22 ± 1 kcal mol^{- 1} in DMSO-d₆. 2D exchange spectroscopy (EXSY) showed barriers for cis-1a: ΔG₃₀₃(+) 19.0 ± 0.2 kcal mol^-1 and trans-1b: ΔG₃₀₃(+) = 19.7 ± 0.2 kcal mol^-1 in DMSO-d₆, respectively. Molecular mechanics and semiempirical calculations support the unexpected higher barrier for trans- 1b. Since the rotation barriers are different for the cis and trans isomer, it can be concluded that control of a second mechanical effect, e.g. the rate of rotation of an attached biaryl rotor, is feasible in a photochemically switchable system.

Full text of DOI:10.1021/jo962210t

J . Org. Chem. 1997, 62, 4943-4948
4943
Tow a r d a Sw itch a ble Molecu la r Rotor . Un exp ected Dyn a m ic
Beh a vior of F u n ction a lized Over cr ow d ed Alk en es
Anne Marie Schoevaars,† Wim Kruizinga,† Robert W. J . Zijlstra,† Nora Veldman,‡
Anthony L. Spek,‡ and Ben L. Feringa*,†
Department of Organic and Molecular Inorganic Chemistry, Groningen Center for Catalysis and
Synthesis, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands, and
Department of Crystal & Structural Chemistry, Bijvoet Center for Biomolecular Research,
Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands
Received November 26, 1996X
In an approach toward a photochemically bistable molecular rotor the synthesis of cis-1a and trans-
1b isomers of 2-(2,6-dimethylphenyl)-9-(2′,3′-dihydro-1′H-naphtho[2,1-b]thiopyran-1′-ylidene)-9H-
thioxanthene (1), being sterically overcrowded alkenes functionalized with an o-xylyl group as a
rotor, is described. The key steps in the synthesis are a Suzuki coupling to attach the xylyl moiety
and a diazo-thioketone coupling with subsequent desulfurization to introduce the central olefinic
bond in 1. The X-ray structure of cis-1a revealed an anti-folded helical conformation. Dynamic
NMR studies on both isomers were performed, to elucidate the kinetics of their rotation processes
and to investigate the possibility to control the biaryl rotation by photochemical cis-trans
q
isomerization. A rotation barrier was found by coalescence spectroscopy for cis-1a : ∆Gc ) 22 (
1 kcal mol-1 in DMSO-d6. 2D exchange spectroscopy (EXSY) showed barriers for cis-1a : ∆G303
q
19.0 ( 0.2 kcal mol-1 and trans-1b: ∆G303 ) 19.7 ( 0.2 kcal mol-1 in DMSO-d6, respectively.
q
Molecular mechanics and semiempirical calculations support the unexpected higher barrier for
trans-1b. Since the rotation barriers are different for the cis and trans isomer, it can be concluded
that control of a second mechanical effect, e.g. the rate of rotation of an attached biaryl rotor, is
feasible in a photochemically switchable system.
In tr od u ction
nent devices is an important challenge as well. An
example of a gated molecular system based on chiroptical
molecular switches in which fluorescence is reversibly
modulated was demonstrated recently.10 Functionaliza-
tion of the optical switches with a rotor would open the
possibility to control a mechanical effect (e.g. the rate of
rotation around a single bond) by the photochemical cis-
trans isomerization, due to the difference in steric
hindrance in the cis and the trans isomer. This concept
in fact represents the molecular analog of an accelerator
in an ordinary machine or a brake, if it is possible to block
the rotation completely in one of the isomers at a specific
temperature.
In this paper we describe the synthesis of a photo-
chemically bistable molecular rotor and dynamic NMR
studies on both isomers, to elucidate the kinetics of their
rotation processes and to investigate the feasibility to
control this mechanical effect by a chiroptical molecular
switch. Although an effective photochemical cis-trans
isomerization could not be achieved in this model system
since the UV/vis absorption spectra of the isomers are
nearly identical,11 unexpected dynamic behavior was
observed.
The development of functional devices at the molecular
and supramolecular level has attracted considerable
interest during the last decade.1,2 Elegant examples of
functional molecular devices were reported recently,
among them a “molecular brake” by Kelly,3 a “molecular
turnstile” by Moore,4 a “light-switchable catalyst” by
Rebek,5 “photoresponsive crown ethers” by Shinkai,6
“photoswitchable fluorophores” by Lehn,7 and “chiroptical
molecular switches”.8 Extensive research in our group
on sterically overcrowded alkenes provided the basis for
the application of these compounds as optical switches
to control chirality of molecules and mesoscopic materi-
als.8 These molecular switches are based on a photo-
chemical cis-trans isomerization of helically shaped
alkenes which simultaneously inverts the helicity of the
molecules. Optical data storage9 is one major objective
for such photochromic bistable organic molecules, but
photochemical control of other functions in multicompo-
† University of Groningen.
‡ Utrecht University.
X Abstract published in Advance ACS Abstracts, J uly 1, 1997.
(1) Feynman, R. P. In Miniaturization; Gilbert, H. D., Ed.; Rein-
hold: New York, 1961; p 282.
(2) (a) Drexler, K. E. Nanosystems: Molecular Machinery, Manu-
facturing and Computation; Wiley: New York, 1992. (b) Lehn, J .-M.
Supramolecular Chemistry; VCH: Weinheim, 1995.
(3) Kelly, T. R.; Bowyer, M. C.; Bhaskar, K. V.; Bebbington, D.;
Garcia, A.; Lang, F.; Kim, M. H.; J ette, M. P. J . Am. Chem. Soc. 1994,
116, 3657.
(4) Bedard, T. C.; Moore, J . S. J . Am. Chem. Soc. 1995, 117, 10662.
(5) Wu¨rthner, F.; Rebek, J ., J r. Angew. Chem., Int. Ed. Engl. 1995,
34, 446.
Resu lts a n d Discu ssion
The principle of a photochemical switchable molecular
rotor is shown in Scheme 1. Compound 1, a thioxan-
thene-based overcrowded ethylene which can adopt a cis-
(6) Shinkai, S.; Ogawa, T.; Kusano, Y.; Manabe, O.; Kikukawa, K.;
Goto, T.; Matsuda, T. J . Am. Chem. Soc. 1982, 104, 1960.
(7) Tsivgoulis, G. M.; Lehn, J .-M. Angew. Chem., Int. Ed. Engl. 1995,
34, 1119.
(8) (a) Feringa, B. L.; J ager, W. F.; de Lange, B.; Meijer, E. W. J .
Am. Chem. Soc. 1991, 113, 5468. (b) J ager, W. F.; de J ong, J . C.; de
Lange, B.; Huck, N. P. M.; Meetsma, A.; Feringa, B. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 348. (c) Feringa, B. L.; Huck, N. P. M.; van
Doren, H. A. J . Am. Chem. Soc. 1995, 117, 9929.
(9) For reviews see: Feringa, B. L.; J ager, W. F.; de Lange, B.
Tetrahedron 1993, 49, 8267. Feringa, B. L.; Huck, N. P. M.; Schoe-
vaars, A. M. Adv. Mater. 1996, 8, 681.
(10) Huck, N. P. M.; Feringa, B. L. J . Chem. Soc., Chem. Commun.
1995, 1095.
(11) Currently this concept is extended to an effectively switchable
system by the attachment of electron-donating and -withdrawing
groups in the molecule in order to increase the difference of the UV
spectra of the cis and the trans isomer.8b
S0022-3263(96)02210-4 CCC: $14.00 © 1997 American Chemical Society

4944 J . Org. Chem., Vol. 62, No. 15, 1997
Schoevaars et al.
Sch em e 1
The crystal structure of cis-1a is shown in Figure 1.
The arrangement about the central double bond is nearly
planar (dihedral angles C12-C11-C14-C26 ) 7.1(11)°;
C10-C11-C14-C15 ) 4.7(11)°). The bond length obtained
for the central double bond (1.345(12) Å, C11-C14) is a
characteristic value for a simple nonconjugated double
bond (1.337 Å).18 The anti-folded helical structure in
which the top and bottom halves are tilted down and
upward, respectively, relative to the plane of the central
double bond, and by which severe nonbonding interac-
tions are avoided, is clearly shown. Both heterocyclic
rings in cis-1a adopt twisted-boat conformations relieving
the steric interactions present in this molecule. In the
crystal structure the xylyl moiety is rotated strongly out
of the plane of the adjacent aryl group resulting in a
torsion angle of this biaryl unit of -86.6(10)° (C18-C17-
C27-C32).
or trans-geometry and is extended with a xylyl group as
a rotor, was designed as a model system. The presence
of the o-methyl groups of the rotor are expected to cause
restricted rotation around the biaryl bond due to steric
interactions with the upper part of the molecule. The
influence of these steric effects on the rate of rotation is
presumed to be different in the cis and the trans isomers
of 1. Therefore the rate of rotation might be controlled
by a photochemical cis-trans isomerization. Inspection
of molecular models suggested severe steric hindrance
for biaryl rotation in the cis isomer compared to the trans
isomer, as a result of interactions with the upper part of
the molecule due to the influence of the large naphtha-
lene unit facing the rotor in the first case compared to
the small methylene group in the latter. Advantageous
are also the singlets in the 1H-NMR spectra arising from
the methyl groups of the xylene rotor, allowing dynamic
NMR studies of the rotation process.
A successful synthetic route to 1 is shown in Scheme
2. Key steps in the synthetic scheme are the Suzuki
coupling12 to attach the xylyl rotor to the thioxanthone
and the coupling of the upper and lower half of the helical
alkene using a diazo-thioketone method.13 Boronic acid
derivative 6 was prepared from 2-bromo-1,3-dimethyl-
benzene 5 via lithiation, quenching with trimethyl borate
and subsequent hydrolysis of the boronic ester. 2-Ni-
trothioxanthone (2)14 was reduced to the corresponding
amine 315 followed by conversion into iodide 4 via a
diazotization procedure. Boronic acid 6 was connected
to the thioxanthone via a palladium-catalyzed coupling16
to provide the xylyl-substituted thioxanthone 7. All
reactions proceed in good yields. Reaction of ketone 7
with P2S5 in toluene resulted in thioketone 8 in 82% yield,
which was reacted with hydrazone 9 in a diazo-thioketone
coupling13 to provide episulfide 10 as a cis-trans mixture.
Desulfurization of the episulfide with copper-bronze in
p-xylene under reflux conditions failed, and the use of
LiAlH4 in refluxing toluene/ether was not successful
either. Finally we were able to desulfurize 10 using PPh3
in boiling toluene.17 The desired alkene 1 was obtained
as a cis-trans mixture with a cis:trans ratio of 1.3:1. The
geometrical isomers and the enantiomers (with opposite
helicities) of the cis as well as the trans forms of 1 were
readily separated by HPLC using a chiral stationary
phase (see Experimental Section).
In the 1H-NMR spectra the isomers can be distin-
guished on the basis of the resonance of the methyl
groups of the xylyl moiety. Furthermore, both isomers
cis-1a and trans-1b show two separate methyl signals
1
in their H-NMR spectra at 30 °C, indicating slow biaryl
rotation on the NMR time scale. The chemical shifts of
the methyl groups in DMSO-d6 are 0.64 and 1.62 ppm
for cis-1a and 1.91 and 2.18 ppm for trans-1b, respec-
tively. The large chemical shift difference between the
methyl signals of cis-1a can be explained by the difference
in electronic environment of both groups. From the X-ray
structure of cis-1a (Figure 1) and molecular modeling
studies, it becomes evident that one methyl group is
pointing toward the aromic upper part of the alkene,
while the other methyl is not influenced by this naph-
thalene moiety. The rotation barriers for cis-1a and
trans-1b were examined first using temperature depend-
1
ent H-NMR,19 focussing on the coalescence behavior of
the methyl signals. From the difference in chemical
shifts of the two methyl absorptions at both sites (∆ν)
plotted against the temperature, the rate constant (kc)
for rotation at the coalescence temperature can be
determined by means of the formula kc ) 2-1/2π∆νc.
Substitution of kc in the Eyring equation gives the free
energy of activation for the rotation, i.e. the rotation
q
barrier ∆Gc .
A series of 1H-NMR spectra were recorded in the
temperature range 313 to 443 K with increments of 10
K. The measurements were carried out for both isomers
simultaneously to ensure exact identical conditions using
a mixture of cis-1a and trans-1b in DMSO-d6. The
results are presented in Figure 2. For the cis isomer ∆ν
decreases with increasing temperature but not to com-
plete coalescence.
Extrapolation of the graph, however, resulted in an
estimated coalescence temperature of 473 ( 5 K and a
rotation barrier ∆Gc of 22 ( 2 kcal mol-1. Much to our
q
surprise, for the trans isomer we did not observe a
significant change in ∆ν in the temperature range
studied; therefore, no estimation of the coalescence
temperature could be made. Although a lower rotation
barrier of the trans isomer was expected, these experi-
ments show that the barrier of the trans isomer is even
higher than that of the cis isomer.
(12) Miyaura, N.; Yanagi, T.; Suzuki, A. Synth. Commun. 1981, 11,
513.
(13) (a) Barton, B. H. R.; Willis, B. J . J . Chem. Soc., Perkin Trans
1 1972, 305. (b) Kellogg, R. M.; Buter, J .; Wassenaar, S. J . Org. Chem.
1972, 37, 4045. (c) Scho¨nberg, A.; Ko¨nig, B.; Singer, E. Chem. Ber.
1967, 100, 767.
(14) Amstutz, E. D.; Neumoyer, C. R. J . Am. Chem. Soc. 1947, 69,
1922.
(15) Mann, F. G.; Turnbull, H. J . J . Chem. Soc. 1951, 747.
(16) Yang, Y. Synth. Commun. 1989, 19, 1001. Procedure as
described in this reference, except that 1.5 equiv of Ba(OH)2 was used
instead of 3 equiv of NaHCO3.
(18) Handbook of Chemistry and Physics; Astle, M. J ., Weast, R.
C., Eds.; CRC Press: Boca Raton, FL, 1980; F-218.
(19) (a) Oki, M. Applications of Dynamic NMR Spectroscopy to
Organic Chemistry. In Methods of Stereochemical Analysis; Marchand,
A. P., Ed.; VCH: Weinheim, 1985; Vol. 4. (b) Sandstro¨m, J . Dynamic
NMR Spectroscopy; Academic Press: New York, 1982. (c) Binsch, G.;
Kessler, H. Angew. Chem., Int. Ed. Engl. 1980, 19, 411.
(17) Seitz, G.; Hoffmann, H. Synthesis 1977, 201.

Toward a Switchable Molecular Rotor
J . Org. Chem., Vol. 62, No. 15, 1997 4945
Sch em e 2
F igu r e 2. Difference in chemical shifts of the two methyl
absorptions (∆ν) of 1a and 1b in DMSO-d6 as a function of
the temperature.
F igu r e 1. PLUTON diagram of the X-ray structure of cis-1a
with adopted numbering scheme.
Since quantitative comparison of the rotation barriers
was not possible using coalescence studies, we studied
the rotation process by 2D EXSY spectroscopy, a powerful
technique for determining exchange rate constants (and
via the Eyring equation free energies of activation) in the
region of slow exchange on NMR time scale.20
As the dynamic behavior of the xylene rotor, within
either cis-1a or trans-1b, resembles the situation of a two-
site exchange system (of two methyl groups on a xylyl
substituent) with equal populations and spin-lattice
relaxation times within each others error margins, the
exchange rate () rotation rate) k can be calculated
directly from the ratio of cross- and auto-peak integra-
tions (aAA/aAB) and the mixing time (tm), using the formula
(aAA/aAB) ) (1 - ktm)/ktm.20c The experiments were carried
out using a mixture of cis and trans isomers of 1 in
DMSO-d6 at temperatures from 303 to 363 K. The
results are presented in Figure 3.
As expected, a linear relationship was found for the
rotation barrier as a function of the temperature. Via
an Eyring plot (ln[kh/kBT] plotted against 1/T), the values
for the enthalpies and entropies of activation were cal-
culated and presented together with the rotation barriers
at 303 K in Table 1. The estimated rotation barrier (at
Tc) of cis-1a in DMSO-d6 from the coalescence studies
(∆G473q) of 22 ( 1 kcal mol-1 is indeed in the same order
of magnitude compared to the results for cis-1a from the
q
exchange spectroscopy in DMSO-d6 in which ∆G473
)
23.6 ( 0.2 kcal mol-1 (calculated from ∆Hq and ∆Sq). In
accordance with the results of the coalescence studies the
barrier of trans-1b is higher than that of cis-1a at
ambient temperature. At 303 K a small but significant
difference of 0.7 kcal mol-1 in rotation barriers for the
cis and trans isomers was determined.
(20) (a) J eener, J .; Meier, B. H.; Bachmann, P.; Ernst, R. R. J . Chem.
Phys. 1979, 71, 4546. (b) Macura, S.; Ernst, R. R. Mol. Phys. 1980,
41, 95. (c) Bodenhausen, G.; Ernst, R. R. J . Am. Chem. Soc. 1982,
104, 1304.
For a better understanding of the unexpected higher
rotation barrier of trans-1b , we have employed both

4946 J . Org. Chem., Vol. 62, No. 15, 1997
Schoevaars et al.
F igu r e 4. AM1-optimized lowest energy conformation of
cis-1a .
F igu r e 3. Rotation barriers (∆Gq) of the biaryl rotation in
cis-1a and trans-1b based on the EXSY experiments in DMSO-
d6.
Ta ble 2. Ca lcu la ted Min im a l En er gies a n d Rota tion
Ba r r ier s of cis-1a a n d tr a n s-1b
Ta ble 1. Kin etic Da ta Exch a n ge Sp ectr oscop y
TRIPOS
AM1
∆Hq
∆Sq
(cal/mol K)
∆G303
(kcal/mol)
q
Emin
∆Erot
Emin
∆Erot
(kcal/mol)
compound
(kcal/mol)
(kcal/mol)
(kcal/mol)
(kcal/mol)
cis-1a
trans-1b
14.3
16.2
-15.5
-11.4
19.0 ( 0.2
19.7 ( 0.2
cis-1a
trans-1b
15.7
15.7
17.0
18.7
138.0
137.3
16.6
17.8
molecular mechanics calculations using the TRIPOS21
force field as implemented in SYBYL and semiempirical
AM122 calculations with MOPAC93.23 It should be noted
that especially in the case of the AM1 calculations caution
has to be taken in defining the reaction path. Since AM1
takes each last optimized coordinate as a starting point
for the next step, biased reaction paths are likely to occur.
It is therefore necessary to investigate the reaction path
over a 360° range, although C2-symmetry is present in
the xylyl rotor. In fact conformations in the vicinity of
maximum rotational energy are recognized by significant
distortions leading to symmetry breaking in this part of
the molecule, which implies that the total reaction path
of the rotation has to be investigated.
The structure of the AM1 optimized lowest energy
conformation of cis-1a (Figure 4) is in agreement with
the X-ray structure (Figure 1); the twisted boat confor-
mations of the heterocyclic rings are confirmed in the
calculated structure. Only a minor difference is found
in the torsion angles of the biaryl units, -86.6(10)° in
the X-ray structure (C18-C17-C27-C32) versus 90° in the
calculated lowest energy conformation. This difference
might be explained by crystal packing effects.
The results of the computational calculations are
presented in Table 2. Both the molecular mechanics and
semiempirical results show the same trend. In the case
of trans-1b an increase of the rotation energy barrier of
1.2 kcal mol-1 (AM1) to 1.7 kcal mol-1 (TRIPOS) in
comparison with cis-1a is predicted.
Analysis of the AM1 transition state (TS) structures
for the biaryl rotations provides a satisfactory explana-
tion for the surprisingly higher barrier of rotation in,
what we expected to be the less sterically hindered, trans-
1b. As can be seen in Figure 5, in the trans-1b transition
state (TS) the methyl group (C34) of the o-xylyl rotor gets
entangled between the two hydrogens at C13. The
distance between the methyl and the ring hydrogens is
2.1 Å, which is well inside their van der Waals radii and
is indicative of a significant repulsive interaction between
the two halves of the molecule.
Examination of the cis-1a TS structure (Figure 5) does
not lead to the discovery of such intimate contacts,
although one of the o-xylyl methyl groups is pointing
toward the naphthyl moiety in the upper part of the
molecule. Typical distances between the methyl hydro-
gens and naphthyl carbon atoms are 3.0 Å. Careful
analysis of the geometry of this TS structure confirms
this interaction to be the sterically demanding one. The
relatively large naphthalene unit in the upper part of the
molecule, which causes the hindrance in the cis isomer,
bends away from the central part of the molecule as is
also clearly seen in the X-ray structure (Figure 1), leaving
enough space for faster rotation of the xylyl unit in cis-
1a compared to trans-1b. Furthermore the C19-C18-
C17-C27 dihedral angle shifts from 180.0° in the lowest
energy conformation to 162.3° in the TS structure (Figure
5). In other words, the o-xylyl group appears to bend
away from the naphthyl, thus reducing the steric interac-
tions between its methyl groups and the opposite naph-
thyl moiety in the other half of the molecule.
From the values of ∆Hq and ∆Sq (Table 1) it can be
seen that the higher rotation barrier of trans-1b is caused
by a higher ∆Hq. The different nature of the interactions
in the transition states for the biaryl rotation in cis-1a
and trans-1b also provides a reasonable explanation for
the inverse trend in the observed ∆Sq of rotation (Table
1). Although the contacts in the case of trans-1b are
more pronounced and thus leading to a higher ∆Hq, a
relatively small region of the molecule (the ethylene
bridge C12-C13) is involved in the interaction with the
o-xylyl rotor. In the case of cis-1a , however, the much
larger naphthyl moiety is involved in the interaction with
the rotor, thus restricting the conformational freedom in
a significantly larger part of the molecule in the TS. The
confinement of the larger part of the molecule in the
(21) (a) Lowe, J . P. Barriers to Internal Rotation about single bonds
in Progress in Physical Organic Chemistry; Wiley: New York, 1968;
Vol. 6, p 1. (b) Dixon, D. A.; Zeroka, D.; Wendoloski, J . J .; Wasserman,
Z. R. J . Phys. Chem. 1985, 89, 5334.
(22) (a) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J .
P. J . Am. Chem. Soc. 1985, 107, 3902. (b) Dewar, M. J . S.; Yuan, Y.-
C. Inorg. Chem. 1990, 29, 3881.
(23) Stewart, J . J . P. Quant. Chem. Prog. Exch. 1990, 10, 86.

Toward a Switchable Molecular Rotor
J . Org. Chem., Vol. 62, No. 15, 1997 4947
F igu r e 5. AM1-optimized cis-1a (left) and trans-1b transition state structures for biaryl rotation.
latter case should lead to a larger ∆Sq of rotation, which
is in agreement with the experimental findings.
each of the 512 increments was the accumulation of 16 scans.
The relaxation delay was 1.5 s. The measurements were
carried out at each temperature with several mixing times
(0.025-1.5 s) to ensure an appropriate choice of the mixing
time. Before Fourier transformation, the FID’s were multi-
plied by p/2 shifted sine-bell function in the F2 domain and in
the F1 domain. The data file was zero-filled, resulting in a
spectrum of 1K × 1K real data points. A base-plane correction
was applied to the 2D spectrum. Peak volumes were deter-
mined by integration of the peaks of interest.
Com p u ta tion a l Deta ils. In all rotation barrier calcula-
tions, the geometry of the whole molecule was optimized as a
function of the rotation around the o-xylyl C-C bond. The
rotational energy barriers were obtained by 5° increments on
the dihedral angle defined. In the case of the AM1 calcula-
tions, the step size was reduced to 1° increments in the regime
of the potential energy barrier maximum. TRIPOS force field
calculations were performed within SYBYL on a SG Indy.
BFGS gradient minimization with convergence criterion of
0.001 cal/mol Å was applied. AM1 calculations were performed
within MOPAC93 on a HP Apollo. For the AM1 geometry (and
reaction path) optimizations, the BFGS converger was used
together with the tightest possible convergence criterion of the
SCF energy (keyword GNORM ) 0). For strained structures
at reaction coordinates in the vicinity of the rotational barrier
maximum, Pulay’s converger was used to improve the conver-
gence of the SCF procedure.
In summary, the assumption that cis-alkenes are the
sterically more hindered structures when compared with
their trans counterparts does not seem to hold for such
helically shaped overcrowded alkenes. As can be seen
in Figure 1, the considerable pitch of these helical
structures introduces a large separation between upper
and lower halve of the alkene in its cis configuration,
especially in the regions remote from the central olefinic
bond. This leads to more conformational freedom then
expected on the basis of a cis-alkene geometry.
On the basis of these results and the additional
observation of unexpected dynamical behavior in several
other helicenes with related structural features,24 it is
suggested that the “classical picture” regarding steric
hindrance in cis and trans isomers is at least incomplete
for such helically shaped inherently dissymetric alkenes.
Con clu sion s
It can be concluded that control of a second mechanical
effect, the rate of rotation of a rotor, by cis-trans
isomerization of a sterically overcrowded alkene is fea-
sible, since the rates of rotation are different for the cis
and trans isomer. In contrast with expectation the
rotation barrier of the trans isomer proved to be higher
than that of the cis isomer. Modifications to increase the
difference in UV/vis absorptions and to enhance the
difference in rotation barriers between the cis and the
trans isomer are necessary to achieve an effective pho-
tochemically switchable system. Such studies are cur-
rently in progress.
Cr ysta l Da ta for cis-1a . C34H26S2, monoclinic, space group
P21/c, a ) 12.9026(10), b ) 14.8056(15), c ) 18.0919(15) Å, â
) 133.733(10), V ) 2497.3(6) Å3, Z ) 4, dx ) 1.327 g cm-3
.
X-ray data for a colorless, plate-shaped crystal cut to size (0.05
× 0.38 × 0.38 mm) were collected on an ENRAF-NONIUS
CAD4T on rotating anode [Mo KR, graphite monochromated,
λ ) 0.71073 Å, Θmax ) 25°, ω-scan, T ) 150 K]. The structure
was solved by direct methods (SHELXS86) and refined on F2
(SHELXL-93). Hydrogen atoms were taken into account at
calculated positions. Final R1 value 0.0829 [2018 reflections
with I > 2σ(I)], wR2 ) 0.1565 [4376 reflections], S ) 1.018,
327 parameters. Atomic coordinates, bond lengths and angles,
and thermal parameters have been deposited at the Cambridge
Crystallographic Data Centre and can be obtained, on request,
from the Director, Cambridge Crystallographic Data Centre,
12 Union Road, Cambridge, CB2 1EZ, UK.
Exp er im en ta l Section
Gen er a l Rem a r k s. Melting points are uncorrected. 1H
NMR data were recorded on a Varian Gemini 200, a Varian
VXR 300, or a Varian Unity 500 (at 200, 300, or 500 MHz,
respectively). 13C NMR data were recorded on a Varian
Gemini 200 or a Varian VXR 300 (at 50.32 or 75.48 MHz,
respectively). All shifts are denoted relative to the solvent
used. Merck silica gel 60 (230-400 mesh ASTM) was used
for chromatography. Solvents were purified using standard
procedures. 2-Bromo-1,3-dimethylbenzene (5) was purchased
from Across Chimica and used without purification. All other
reagents are commercially available and were used without
purification.
(2,6-Dim eth ylp h en yl)bor on ic Acid (6). A stirred solu-
tion of 1.00 g (5.41 mmol) of 2-bromo-1,3-dimethylbenzene (5)
in 50 mL of THF was cooled to -78 °C under a nitrogen
atmosphere, 5.94 mmol of sec-BuLi (4.57 mL of 1.3 M solution
in hexane) was added via a syringe, and the temperature of
the mixture was maintained at -78 °C for 1 h. The temper-
ature was gradually raised to -50 °C, and 0.62 mL (5.41 mmol)
(24) Harada, N.; Saito, A.; Koumura, N.; Roe, C. D.; J ager, W. F.;
Zijlstra, R. W. J .; de Lange, B.; Feringa, B. L. J . Am. Chem. Soc.,
submitted for publication.
(25) States, D. J .; Haberkorn, R. A.; Ruben, D. J . J . Magn. Reson.
1982, 48, 286.
Exch a n ge Sp ectr oscop y Exp er im en ts. The EXSY spec-
tra were recorded at 300 MHz on a Varian VXR 300 with the
NOESYPH pulse sequence supplied by the Varian software.
For phase cycling, the States-Haberkorn method25 was used;

4948 J . Org. Chem., Vol. 62, No. 15, 1997
Schoevaars et al.
of B(OMe)3 was added. The mixture was warmed to room
temperature, and 200 mL of ether was added. The organic
layer was washed with 5% HCl solution (100 mL) and water
(100 mL) and dried over MgSO4. Evaporation of the solvents
gave the crude product which was crystallized from ether and
hexanes to give 6 as white needles (0.41 g, 2.73 mmol, 51%):
to -5 °C, whereupon MgSO4 (approximately 0.5 g), Ag2O (0.6
g, 2.6 mmol), and a saturated solution of KOH in MeOH (0.3
mL) were successively added. After 15 min, the mixture
turned deep red, and after another 30 min of stirring at -5
°C, the solution was filtered into another ice-cooled flask, and
the remaining residue was washed with cold CH2Cl2 (15 mL).
To this clear solution was added thioketone 8 in small portions.
Evolution of nitrogen was observed, and the deep red color
disappeared. The thioketone was added until the evolution
of nitrogen had ceased; a total of 360 mg (1.08 mmol) was
necessary. After stirring for 3 h at room temperature, the
solvents were evaporated. Purification of the crude product
by column chromatography (silica gel, CH2Cl2 and subse-
quently CH2Cl2/hexanes, 1/1) and crystallization from ethanol
and CH2Cl2 afforded 10 as white crystals (0.47 g, 0.89 mmol,
82% based on the thioketone): mpcis ) 200.8 °C; 1H NMR
(CDCl3, cis isomer): δ 0.89 (s, 3H), 1.75 (s, 3H), 2.19-2.27 (m,
1H), 2.49-2.69 (m, 3H), 6.63 (dd, J ) 7.8, 1.8 Hz, 1H), 6.80
(d, J ) 7.3 Hz, 1H), 6.92 (d, J ) 6.4 Hz, 1H), 6.99 (d, J ) 7.3
Hz, 1H), 7.05 (d, 1H), 7.11-7.55 (m, 16 H), 7.60 (d, J ) 8.4
Hz, 1H), 8.13 (dd, J ) 5.9, 3.6 Hz, 1H), 9.10 (d, J ) 8.54 Hz,
1H), methyl signals of the trans isomer: δ 2.11(s) and 2.19(s);
13C NMR (CDCl3, mixture of cis and trans isomers): δ 19.78,
20.66, 26.36, 26.76, 36.61, 37.21, 57.90, 58.60, 59.80, 60.50,
122.67, 123.16, 124.01, 124.14, 124.38, 124.49, 125.05, 125.15,
125.51, 125.59, 125.72, 125.83, 125.98, 126.20, 126.51, 126.62,
127.19, 127.21, 127.31, 127.41, 127.49, 127.55, 127.70, 127.73,
127.80, 128.12, 128.20, 128.51, 128.71, 129.69, 130.34, 131.02,
131.16, 131.48, 131.64, 131.91, 132.33, 133.04, 135.69, 135.80,
135.84, 136.26, 136.61, 137.48, 138.17, 138.55, 140.37, 140.97.
Anal. Calcd for C34H26S3: C 76.94, H 4.94, S 18.12. Found:
C 76.53, H 5.11, S 18.07; HRMS calcd 530.120, found 530.120.
cis- a n d tr a n s-2-(2,6-Dim eth ylp h en yl)-9-(2′,3′-d ih yd r o-
1′H -n a p h t h o[2,1-b]t h iop yr a n -1′-ylid e n e )-9H -t h ioxa n -
th en e (1a , 1b). A mixture of 0.20 g (0.38 mmol, mixture of
cis and trans isomers) of episulfide 10, 0.12 g (0.46 mmol) PPh3,
and 20 mL of dry toluene was stirred and refluxed for 18 h
and then cooled to room temperature. The solvent was
evaporated, and the crude product was purified by column
chromatography (silica gel, toluene) to give alkene 1 as a cis/
1
mp ) 121 °C; H NMR (acetone d6): δ 2.31 (s, 6H), 6.90 (d, J
) 7.3 Hz, 2H), 7.04 (t, J ) 7.3 Hz, 1H), 7.20 (s, 2H); 13C NMR
(acetone-d6): δ 36.13, 140.38, 142.25, 153.46, 220.10; HRMS
calcd 150.085, found 150.085.
2-Am in oth ioxa n th en -9-on e (3). This compound was pre-
pared from 2-nitro-9H-thioxanthen-9-one 2 (2.00 g, 7.78 mmol)
following the procedure described by Mann and Turnbull.15
Amine 3 was obtained as a yellow powder (1.30 g, 5.73 mmol,
74%): mp ) 227-228 °C 1H NMR (CDCl3): δ 3.95 (bs, 2H),
7.06 (dd, J ) 8.6, 3.0 Hz, 1H), 7.42 (d, J ) 8.5 Hz, 1H), 7.46-
7.61 (m, 3H), 7.92 (d, J ) 3.0 Hz), 8.63 (dd, J ) 6.9, 0.9 Hz);
HRMS calcd 227.040, found 227.040.
2-Iod oth ioxa n th en -9-on e (4). A suspension of amine 3
(0.55 g, 2.42 mmol), 1.2 mL concentrated HCl, and 1 g of ice
was stirred and cooled with an ice bath. An ice-cooled solution
of NaNO2 (0.19 g, 2.78 mmol) in 1 mL of water was added
dropwise, the temperature being kept at 0-5 °C. After stirring
the suspension for 15 min, it was slowly poured into a solution
of 1.43 g (8.61 mmol) of KI in 5 mL of water. The mixture
was warmed till the evolution of gas ceased and allowed to
cool to room temperature. CH2Cl2 (100 mL) was added, and
the organic layer was separated, washed successively with 10%
aqueous NaOH, water, and 5% aqueous NaHSO3, and then
dried over MgSO4. Purification of the crude product by column
chromatography (silica gel, CH2Cl2) afforded 4 as a yellow solid
1
(0.51 g, 1.51 mmol, 62%): mp ) 147.6 °C; H NMR (CDCl3):
δ 7.31 (dd, J ) 7.3, 2.6 Hz, 1H), 7.47-7.71 (m, 3H), 7.88 (dd,
J ) 8.6, 2.1 Hz, 1H), 8.60 (d, J ) 8.1 Hz, 1H), 8.93 (d, J ) 1.7
Hz, 1H); 13C NMR (CDCl3): δ 90.76, 126.04, 126.52, 127.42,
128.40, 129.89, 130.40, 132.50, 136.73, 138.49, 140.55, 178.20;
HRMS calcd 337.926, found 337.926.
2-(2,6-Dim eth ylp h en yl)th ioxa n th en -9-on e (7). A mix-
ture of 4 (1.00 g, 2.96 mmol), Pd(PPh3)4 (0.09 g, 0.08 mmol),
and DME (20 mL) was stirred under a nitrogen atmosphere
for 10 min. Boronic acid 6 (0.49 g, 3.26 mmol) and a
suspension of 1.54 g (4.90 mmol) of Ba(OH)2‚8H2O in 20 mL
of water were subsequently added. The mixture was stirred
and refluxed for 18 h and allowed to cool to room temperature.
Ether (100 mL) was added, and the water layer was separated
and extracted with ether (50 mL) and dichloromethane (50
mL). The combined organic layers were dried over MgSO4,
and the solvents were evaporated. Purification of the crude
product by column chromatography (silica gel, CH2Cl2/hexanes,
1/1) afforded 7 as a yellow solid (0.77 g, 2.44 mmol, 82%): mp
trans mixture in
a 1/1 ratio (0.10 g, 0.21 mmol, 55%).
Crystallization from ethanol/CH2Cl2 afforded crystals of cis-
1a suitable for X-ray analysis (see Supporting Information).
Resolution of the enantiomers of the cis and trans isomers was
performed with chiral HPLC employing a (+)-poly(triphenyl-
methyl methacrylate) column (OT+) using hexanes/2-propanol
9/1 as eluent (flow 1 mL/min) retention times: tcisI ) ttransI
)
7.9 min, ttransII ) 10.5 min, tcisII ) 13.4 min; mpcis ) 236.7 °C;
1H NMR (CDCl3, the data for cis-1a and trans-1b in a 2.75/
1.00 ratio are given): δ 0.71 (s, 2.20H), 1.72 (s, 2.20H), 1.98
(s, 0.80H), 2.11-2.16 (m, 1H), 2.23 (s, 0.80H), 3.38-3.70 (m,
3H), 6.47 (d, J ) 2.0 Hz, 0.73H), 6.48 (m, 0.27H), 6.53 (d, J )
7.8 Hz, 0.27H), 6.59 (dd, J ) 7.8, 2.0 Hz, 0.73H), 6.80 (d, J )
7.8 Hz, 0.73H), 6.82 (m, 0.27H), 6.89 (d, J ) 7.3 Hz, 0.73 H),
6.98 (t, J ) 7.6 Hz, 0.73H), 7.03 (t, J ) 7.8 Hz, 0.27H), 7.09-
7.23 (m, 2.54H), 7.30-7.44 (m, 3.73H), 7.55-7.63 (m, 3H),
7.66-7.70 (m, 1H), 7.79 (d, J ) 8.3 Hz, 0.73 H); 13C NMR
(CDCl3, the data for cis-1a are given): δ 19.73, 20.09, 29.60,
30.02, 124.49, 124.60, 125.83, 126.11, 126.49, 126.65, 126.71,
126.78, 126.86, 127.42, 127.86, 127.92, 127.81, 129.11, 131.95,
132.43, 132.44, 133.99, 134.10, 134.40, 135.90, 136.22, 138.62,
139.00, 140.07; HRMS calcd 498.148, found 498.148.
1
) 163.1 °C; H NMR (CDCl3): δ 2.05 (s, 6H), 7.12-7.20 (m,
3H), 7.43-7.49 (m, 2H), 7.62-7.68 (m, 3H), 8.45 (d, J ) 1.8
Hz, 1H), 8.64 (d, J ) 8.1 Hz); 13C NMR (CDCl3): δ 20.70,
125.91, 126.11, 126.19, 127.37, 127.44, 129.16, 129.22, 129.79,
130.07, 132.15, 133.39, 135.51, 135.89, 137.14, 139.28, 140.06,
179.92; HRMS calcd 316.092, found 316.092.
2-(2,6-Dim eth ylp h en yl)th ioxa n th en e-9-th ion e (8).
A
solution of 7 (0.72 g, 2.28 mmol) in 25 mL of toluene was
stirred under a nitrogen atmosphere whereupon P2S5 (1.01 g,
4.56 mmol) was added, and the mixture was subsequently
heated at reflux. After 10 min the dark green mixture was
filtered hot, the residue was washed with toluene and CH2Cl2
till the filtrate was slightly green, and the solvents were
evaporated. Purification of the crude product by column
chromatography (silica gel, CH2Cl2) and crystallization from
ethanol afforded 8 as dark green crystals (0.62 g, 1.87 mmol,
82%): mp ) 100.8 °C; 1H NMR (CDCl3): δ 2.08 (s, 6H), 7.13-
7.18 (m, 3H), 7.43-7.46 (m, 2H), 7.60-7.67 (m, 3H), 8.86 (d,
Su p p or tin g In for m a tion Ava ila ble: 1H and 13C NMR
spectra of compounds 1, 4, 6, 7, and 8; data of coalescence
studies and EXSY studies (12 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J
) 1.46 Hz, 1H), 9.00 (d, J ) 8.42 Hz, 1H); 13C NMR
(CDCl3): δ 20.82, 125.86, 125.98, 126.85, 127.44, 130.14,
131.49, 131.75, 132.80, 133.21, 133.61, 135.93, 137.58, 137.62,
139.76, 140.07, 211.07; HRMS calcd 332.069, found 332.069.
cis- a n d tr a n s-Disp ir o[2,3-d ih yd r o-1H-n a p h th o[2,1-b]-
t h iop yr a n -1,2′-t h iir a n e-3′,9′′-[2′′-(2,6-d im et h ylp h en yl)-
9′′H-th ioxa n th en e]] (10). A stirred solution of hydrazone 9
(340 mg, 1.52 mmol) in anhydrous CH2Cl2 (30 mL) was cooled
Ack n ow led gm en t. Financial support was provided
by the Netherlands Organization for Scientific Research
(NWO/SON) and Stichting Technische Wetenschappen
(STW).
J O962210T