An enantioselective synthetic route toward second-generation light-driven rotary molecular motors

Article abstract of DOI:10.1021/jo902348u

(Chemical Equation Presented) Controlling the unidirectional rotary process of second-generationmolecular motors demands access to these motors in their enantiomerically pure form. In this paper, we describe an enantioselective route to three new second-generation light-driven molecular motors. Their synthesis starts with the preparation of an optically active R-methoxy-substituted upper-half ketone involving an enzymatic resolution. The subsequent conversion of this ketone to the corresponding hydrazone by treatment with hydrazine led to full racemization. However, conversion to a TBDMS-protected hydrazone by treatment with bis-TBDMS hydrazine, prepared according to a new procedure, proceeds with nearly full retention of the stereochemical integrity. Oxidation of the TBDMS-protected hydrazone and subsequent coupling to a lower-half thioketone followed by recrystallization provided the molecular motors with >99% ee. As these are the first molecular motors that have a methoxy substituent at the stereogenic center, the photochemical and thermal isomerization steps involved in the rotary cycle of one of these new molecules were studied in detail with various spectroscopic techniques.

Full text of DOI:10.1021/jo902348u

pubs.acs.org/joc
An Enantioselective Synthetic Route toward Second-Generation
Light-Driven Rotary Molecular Motors
Thomas C. Pijper, Dirk Pijper, Michael M. Pollard, Fr ꢀe d ꢀe ric Dumur, Stephen G. Davey,
Auke Meetsma, and Ben L. Feringa*
Stratingh Institute for Chemistry, Zernike Institute for Advanced Materials, and Center for Systems
Chemistry, University of Groningen, Nijenborgh 4, 9747 AG, Groningen, The Netherlands
b.l.feringa@rug.nl
Received November 6, 2009
Controlling the unidirectional rotary process of second-generation molecular motors demands access
to these motors in their enantiomerically pure form. In this paper, we describe an enantioselective
route to three new second-generation light-driven molecular motors. Their synthesis starts with the
preparation of an optically active R-methoxy-substituted upper-half ketone involving an enzymatic
resolution. The subsequent conversion of this ketone to the corresponding hydrazone by treatment
with hydrazine led to full racemization. However, conversion to a TBDMS-protected hydrazone by
treatment with bis-TBDMS hydrazine, prepared according to a new procedure, proceeds with nearly
full retention of the stereochemical integrity. Oxidation of the TBDMS-protected hydrazone and
subsequent coupling to a lower-half thioketone followed by recrystallization provided the molecular
motors with >99% ee. As these are the first molecular motors that have a methoxy substituent at the
stereogenic center, the photochemical and thermal isomerization steps involved in the rotary cycle of
one of these new molecules were studied in detail with various spectroscopic techniques.
4
irradiation withlight. Rotary molecular motors that are fueled
Introduction
5
by chemical energy are also known: Kelly and co-workers and
The bottom-up construction of nanoscale devices in which
motion can be induced and/or controlled at the molecular
6
Branchaud and co-workers have presented rotors which are
capable of unidirectional 120 and 180° rotation, respectively,
whereas we have reported a rotary motor which is capable of
1
level is a topic of great contemporary interest. Inspired by
the highly sophisticated and effective molecular devices that
are present in nature, such as the bacterial flagellum and the
7
continuous, unidirectional 360° rotation. In addition, two
catenane systems which are able to undergo unidirectional
2
ATP synthase rotary motors, a wide array of such devices
8
rotation have been described by Leigh and co-workers.
Two types of light-driven rotary molecular motors that are
derived from overcrowded alkenes have been designed: the
3
has been designed and studied in recent years. Among these
are various types of rotary molecular motors, which are
capable of converting energy into rotary motion and which
3
have a preference for a unique sense of rotation. Our group
(4) For a comprehensive review, see: Feringa, B. L. J. Org. Chem. 2007,
has reported a series of rotary molecular motors that per-
form unidirectional rotation around an olefin axis upon
72, 6635–6652.
(5) (a) Kelly, T. R.; De Silva, H.; Silva, R. A. Nature 1999, 401, 150–152.
b) Kelly, T. R.; Silva, R. A.; De Silva, H.; Jasmin, S.; Zhao, Y. J. Am. Chem.
(
Soc. 2000, 122, 6935–6949. (c) Kelly, T. R.; Cavero, M. Org. Lett. 2002, 4,
2653–2656. (d) Kelly, T. R.; Cai, X.; Damkaci, F.; Panicker, S. B.; Tu, B.;
Bushell, S. M.; Cornella, I.; Piggott, M. J.; Salives, R.; Cavero, M.; Zhao, Y.;
Jasmin, S. J. Am. Chem. Soc. 2007, 129, 376–386.
(
1) (a) Molecular Devices and Machines: Concepts and Perspectives for the
Nanoworld; Balzani, V., Credi, A., Venturi, M., Eds.; Wiley-VCH: Weinheim,
Germany, 2008. (b) van den Heuvel, M. G. L.; Dekker, C. Science 2007, 317, 333–
3
36. (c) Whitesides, G. M. Sci. Am. 2001, 285, 78–83.
2) Molecular Motors Schliwa, M., Ed.; Wiley-VCH: Weinheim,
Germany, 2003.
3) For reviews, see: (a) Kay, E. R.; Leigh, D. A.; Zerbetto, F. Angew.
(6) Lin, Y.; Dahl, B. J.; Branchaud, B. P. Tetrahedron Lett. 2005, 46,
8359–8362.
(7) Fletcher, S. P.; Dumur, F.; Pollard, M. M.; Feringa, B. L. Science
2005, 310, 80–82.
(
(
Chem., Int. Ed. 2007, 46, 72–191. (b) Browne, W. R.; Feringa, B. L. Nat.
Nanotechnol. 2006, 1, 25–35. (c) Kottas, G. S.; Clarke, L. I.; Horinek, D.;
Michl, J. Chem. Rev. 2005, 105, 1281–1376.
(8) (a) Leigh, D. A.; Wong, J. K. Y.; Dehez, F.; Zerbetto, F. Nature 2003,
424, 174–179. (b) Hern ꢀa ndez, J. V.; Kay, E. R.; Leigh, D. A. Science 2004,
306, 1532–1537.
DOI: 10.1021/jo902348u
r 2010 American Chemical Society
Published on Web 01/07/2010
J. Org. Chem. 2010, 75, 825–838 825

Pijper et al.
JOCArticle
molecular motors have been successfully immobilized on
1
4
15
gold nanoparticles and quartz and silica surfaces, used to
influence the speed of rotation of an attached secondary
1
rotor, and applied to induce and switch a preference of the
6
17
helical screw sense of an attached helical polymer. In
addition, they were explored as chiral dopants in liquid
crystals in order to induce cholesteric phases in a reversible
manner and control the helical organization of these meso-
1
8
19
phases and to induce microscopic rotary motion.
In order to study and harness the unidirectionality of the
rotary motion and the intrinsic helical shape of these motors,
most of the aforementioned studies required the use of these
motors in their enantiopure form. Procedures for the
enantioselective synthesis of first-generation molecular
2
0,21
motors have previously been developed.
However,
a procedure which can provide enantiomerically pure
second-generation motors directly has so far not been avail-
FIGURE 1. (a) First-generation molecular motor 1 and second-
generation molecular motor 2. (b) Molecular motor 3 with various
substituents, in order of the enhanced rate of the thermal isomeri-
zation step.
2
2
able. Second-generation motors have therefore always
been synthesized as a racemate, after which preparative
chiral stationary phase HPLC (CSP-HPLC) was used to
separate the enantiomers. The amount of material that can
be separated in a single “run” is usually limited to only a few
milligrams. This amount is typically enough to allow the
performance of spectroscopic measurements but often
insufficient when one would like to examine other material
properties or study these molecules in complex systems
such as surfaces, polymers, liquid crystals, and Langmuir-
Blodgett films. Multiple chiral HPLC runs are needed in
these cases, meaning a significant investment in both equip-
ment and solvents as well as time. Additionally, the separa-
tion of these molecular motors by CSP-HPLC is often
challenging due to the small difference in the retention times
of the two enantiomers and their low solubility in the
required eluent. Because of these drawbacks, preparative
CSP-HPLC is a less attractive method to provide enantio-
pure motors. Consequently, future research on functional
second-generation motors would benefit considerably from
a procedure for their enantioselective preparation.
9
first-generation type 1, which has an identical upper and
1
0
lower half, and the second-generation type 2, which has a
dissimilar upper and lower half (Figure 1a). Key features of
these motors are the helical shape, the central olefin which is
prone to stilbene-type photoisomerization and which serves
as the axis of rotation, and the presence of a stereogenic
center (two stereogenic centers in the case of the first-
generation system) next to the central olefin. Rotation in
these systems takes place through two photochemically
driven, reversible cis-trans isomerizations of the central
olefin, which are each followed by a thermally driven,
irreversible helix inversion. The direction of this rotation is
dictated by the absolute configuration of the stereogenic
center(s). The rotational speed of these motors could be
influenced by a series of structural modifications. Adjusting
the size of atoms X and Y, for example, resulted in a change
in the extent of steric interactions between the upper and
lower half of the system, thereby either accelerating or
Second-generation molecular motors are usually con-
structed by connecting an upper-half diazo compound to a
1
1
decelerating the thermal helix inversion steps. Varying
the size of the substituent R also affected the rate of the
thermal step, although the result of this modification was
counterintuitive: when the size of R on motor 3 was in-
creased, an increase of the rate of the thermal step was
(
14) van Delden, R. A.; ter Wiel, M. K. J.; Pollard, M. M.; Vicario, J.;
Koumura, N.; Feringa, B. L. Nature 2005, 437, 1337–1340.
15) Pollard, M. M.; Lubomska, M.; Rudolf, P.; Feringa, B. L. Angew.
Chem., Int. Ed. 2007, 46, 1278–1280.
16) ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.; Feringa, B. L.
(
1
2,13
observed (Figure 1b).
In addition to studies which
(
involve influencing the motor function itself, investigations
have been performed toward the functioning of these mole-
cular motors in larger systems. So far, second-generation
Org. Biomol. Chem. 2005, 3, 4071–4076.
(17) Pijper, D.; Feringa, B. L. Angew. Chem., Int. Ed. 2007, 46, 3693–3696.
(18) (a) van Delden, R. A.; Koumura, N.; Harada, N.; Feringa, B. L.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 4945–4949. (b) Eelkema, R.; Feringa,
B. L. Org. Biomol. Chem. 2006, 4, 3729–3745. (c) Eelkema, R.; Feringa, B. L.
Chem. Asian J. 2006, 1, 367–369.
(19) (a) Eelkema, R.; Pollard, M. M.; Vicario, J.; Katsonis, N.; Ramon,
B. S.; Bastiaansen, C. W. M.; Broer, D. J.; Feringa, B. L. Nature 2006, 440,
163. (b) Eelkema, R.; Pollard, M. M.; Katsonis, N.; Vicario, J.; Broer, D. J.;
Feringa, B. L. J. Am. Chem. Soc. 2006, 128, 14397–14407. (c) Bosco, A.;
Jongejan, M. G. M.; Eelkema, R.; Katsonis, N.; Lacaze, E.; Ferrarini, A.;
Feringa, B. L. J. Am. Chem. Soc. 2008, 130, 14615–14624.
(20) (a) Harada, N.; Koumura, N.; Feringa, B. L. J. Am. Chem. Soc. 1997,
119, 7256–7264. (b) Fujita, T.; Kuwahara, S.; Harada, N. Eur. J. Org. Chem.
2005, 4533–4543.
(21) For a catalytic asymmetric synthesis of overcrowded alkenes, see:
Hojo, D.; Noguchi, K.; Tanaka, K. Angew. Chem., Int. Ed. 2009, 48, 8129–
8132.
(
9) Koumura, N.; Zijlstra, R. W. J.; van Delden, R. A.; Harada, N.;
Feringa, B. L. Nature 1999, 401, 152–155.
10) (a) Koumura, N.; Geertsema, E. M.; Meetsma, A.; Feringa, B. L.
(
J. Am. Chem. Soc. 2000, 122, 12005–12006. (b) Koumura, N.; Geertsema,
E. M.; van Gelder, M. B.; Meetsma, A.; Feringa, B. L. J. Am. Chem. Soc.
2
002, 124, 5037–5051.
11) For a comprehensive review, see: Pollard, M. M.; Klok, M.; Pijper,
D.; Feringa, B. L. Adv. Funct. Mater. 2007, 17, 718–729.
12) (a) Vicario, J.; Meetsma, J.; Feringa, B. L. Chem. Commun. 2005,
910–5912. (b) Vicario, J.; Walko, M.; Meetsma, A.; Feringa, B. L. J. Am.
Chem. Soc. 2006, 128, 5127–5135.
13) A similar finding about the effect of substituents on rotational
(
(
5
(
barriers was found by Kelly and co-workers in their research on molecular
ratchets increasing the size of the “pawl” of a ratchet lowered the barrier to
rotation. Their explanation for this observation is that a larger pawl increases
the energy of the ground state of the molecule, thus bringing it closer to the
rotation energy barrier. See: Kelly, T. R.; Sestelo, J. P.; Tellitu, I. J. Org.
Chem. 1998, 63, 3655–3665.
(22) While this paper was under consideration, Tietze et al. published an
attractive asymmetric catalytic synthesis of chiral overcrowded alkenes that
are structurally analogous to second-generation molecular motors. See:
Tietze, L. F.; D u€ fert, A.; Lotz, F.; S o€ lter, L.; Oum, K.; Lenzer, T.; Beck,
T.; Herbst-Irmer, R. J. Am. Chem. Soc. 2009, 131, 17879–17884.
8
26 J. Org. Chem. Vol. 75, No. 3, 2010

Pijper et al.
JOCArticle
SCHEME 1. (a) Key Disconnections in the Retrosynthesis of
a Second-Generation Molecular Motor, (b) Conversion of an
Enantiopure Upper-Half Ketone to the Corresponding Hydrazone
with Accompanying Racemization
SCHEME 2. Conversion of an Enantiomerically Pure Upper-
Half Ketone to the Corresponding TBDMS Hydrazone As a Key
Step toward Second-Generation Molecular Motors
ketone without causing racemization (Scheme 2). After
oxidation of the hydrazone to the diazo compound, this
compound is then to be used in a Barton-Kellogg coupling
procedure during which steps we did not expect any loss of
enantiomeric excess (ee).
lower-half thioketone through a Barton-Kellogg reaction
2
3-25
(Scheme 1a).
The diazo compound is typically gener-
ated in situ by oxidation of the corresponding hydrazone,
which in turn can be obtained from the corresponding ketone
by treatment with hydrazine. Since the stereogenic center
that is situated in the upper half is already present in the
ketone precursor, the most straightforward approach for an
enantioselective synthesis of the motor would be to prepare
the ketone using an enantioselective procedure and then
perform the coupling to a lower half without affecting the
stereochemistry. However, because of the steric hindrance
around the carbonyl group, conversion of the ketone to the
hydrazone usually needs to be performed by heating at reflux
for several hours. Combined with the basicity of the medium,
these conditions allow the ketone to undergo keto-enol
tautomerization, causing racemization and thus providing
the product hydrazone as a racemate (Scheme 1b).
We anticipated that this racemization could be avoided by
converting the ketone to a TBDMS-protected hydrazone by
treatment with bis-TBDMS hydrazine and a catalytic
amount of scandium(III) triflate. This procedure, introduced
2
6
by Furrow and Myers, employs much milder conditions
than the ordinary hydrazone formation procedure described
above as it can be carried out at lower temperatures with
short reaction times, whereas the protected hydrazine used in
this reaction is a sterically hindered base. Although Furrow
and Myers have demonstrated their procedure only with
relatively unhindered ketones and aldehydes that were either
achiral or not prone to tautomerization, it was anticipated
that this procedure would provide the corresponding
TBDMS hydrazone from the sterically hindered upper-half
FIGURE 2. Synthesized upper-half ketone (S)-4 and alkenes (S)-5,
(
S)-6, (S)-7, and 8.
To test this hypothesis, we have attempted this TBDMS
hydrazone formation as well as the subsequent Bar-
ton-Kellogg reaction with the enantiomerically enriched
ketone (S)-4 (Figure 2). This ketone, which bears a methoxy
substituent at the R-position, was prepared via a synthetic
route which involved enzymatic kinetic resolution as the key
step. In order to obtain the TBDMS-protected hydrazine
necessary for the conversion of this ketone to the corres-
ponding TBDMS hydrazone, it proved to be necessary to
develop a new, safer procedure to make TBDMS-protected
hydrazine. This compound was then used in the preparation
of three different second-generation molecular motors,
namely, alkenes (S)-5, (S)-6, and (S)-7. As second-generation
motors with a methoxy substituent at the stereogenic center
(
23) (a) Barton, D. H. R.; Willis, B. J. J. Chem. Soc. D 1970, 1225–1226.
b) Barton, D. H. R.; Smith, B. J.; Willis, B. J. J. Chem. Soc. D 1970, 1226.
24) (a) Kellogg, R. M.; Wassenaar, S. Tetrahedron Lett. 1970, 11, 1987–
990. (b) Kellogg, R. M. Tetrahedron 1976, 32, 2165–2184.
(
(
1
(
25) Staudinger, H.; Siegwart, J. Helv. Chim. Acta 1920, 3, 833–840.
26) (a) Furrow, M. E.; Myers, A. G. J. Am. Chem. Soc. 2004, 126, 5436–
(
445. (b) Furrow, M. E.; Myers, A. G. J. Am. Chem. Soc. 2004, 126, 12222–
5
1
2223.
J. Org. Chem. Vol. 75, No. 3, 2010 827

Pijper et al.
JOCArticle
have previously not been made in our laboratories, the photo-
chemical and thermal isomeric behavior of one of these motors,
SCHEME 3. Asymmetric Synthesis of Upper-Half Ketone (S)-4
(S)-7, was studied in detail. Finally, alkene 8, a desymmetrized
version of (S)-7, was synthesized in order to study the uni-
directionality of the rotary motion displayed by this system.
Results and Discussion
The enantioselective synthesis of ketone (S)-4 started from
2
7
the unsubstituted ketone 9, which was first converted to an
unstable R-hydroxy dimethyl acetal by treatment with
2
bis(acetoxy)iodo]benzene followed by in situ hydrolysis
8
[
to provide ketone 10 in 62% yield (Scheme 3). Following a
2
9
procedure by Saha-M o€ ller and co-workers, enzymatic
kinetic resolution was then applied to achieve a selective
3
0
esterification of the R enantiomer, providing the corres-
ponding acetate. The lipase used for this conversion, Amano
lipase PS, performed this task with high selectivity, allowing
us to obtain ketone (S)-10 in 43% yield out of a maximum
theoretical yield of 50%, and with an ee of >99%. (S)-10 was
then treated with silver(I) oxide, iodomethane, and calcium
sulfate in order to methylate the hydroxyl function under
mild conditions, avoiding racemization and providing
ketone (S)-4 in 67% yield and 98% ee.
Bis-TBDMS hydrazine is typically prepared by mixing
anhydrous hydrazine and TBDMS chloride, either in the
SCHEME 4. Conversion of Ketone (S)-4 to TBDMS Hydra-
zone (S)-11
2
6,31
presence or in the absence of a solvent.
However, as the
sale of anhydrous hydrazine in Europe is prohibited and the
in-house preparation of anhydrous hydrazine through dis-
3
2
tillation is very hazardous, we have developed a new
procedure for the preparation of bis-TBDMS hydrazine
which omits the use of anhydrous hydrazine. In this method,
bis-TBDMS hydrazine is prepared from hydrazine mono-
hydrochloride by stirring this compound with TBDMS
chloride in the presence of triethylamine at 110 °C for
3
3,34
4
.5 h.
After a quick workup by extraction with n-pen-
used in the conversion of ketone (S)-4 to the corresponding
TBDMS hydrazone (S)-11, a major drop in ee was observed
unless the bis-TBDMS hydrazine was distilled a third time.
Additional distillations did not improve the ee any further.
Following this method, typically 3.5 g of bis-TBDMS
hydrazine (2.4 mL) was isolated from 3.4 g of hydrazine
monohydrochloride, which corresponds to a yield of 27%.
Using the three-fold distilled bis-TBDMS hydrazine,
ketone (S)-4 was converted to TBDMS hydrazone (S)-11
by heating a mixture of (S)-4, 2 equiv of bis-TBDMS
hydrazine, and 2 mol % of scandium(III) triflate to 70 °C
tane, purification of the crude bis-TBDMS hydrazine is then
performed by multiple distillations under reduced pressure.
We found that two distillations were enough to provide bis-
TBDMS hydrazine that was pure by H NMR spectroscopy.
However, when the bis-TBDMS hydrazine was subsequently
1
(
(
27) Mayer, F.; M u€ ller, P. Ber. Dtsch. Chem. Ges. 1927, 60, 2278–2283.
28) Moriarty, R. M.; Hu, H.; Gupta, S. C. Tetrahedron Lett. 1981, 22,
1
283–1286.
29) Adam, W.; Dı
Asymmetry 1996, 7, 2207–2210.
30) The absolute configuration of the enantiomerically pure ketone 10
was assumed to be S, based on the substrate preference of Amano lipase PS
(
´
az, M. T.; Fell, R. T.; Saha-M o€ ller, C. R. Tetrahedron:
(
(Scheme 4). The conversion was complete after 45 min, as
29
as reported by Saha-M o€ ller and co-workers. The absolute configuration
of ketone 10 (as well as that of the successive compounds ketone 4, TBDMS
hydrazone 11, and alkenes 5, 6, and 7) was confirmed by X-ray crystallo-
graphy with refinement of the Flack parameter of alkene 5.
1
determined by H NMR spectroscopy. Analysis of the
reaction mixture with CSP-HPLC revealed that the conver-
sion had proceeded with only limited decrease in enantio-
purity, as the ee of (S)-11 was 84%.
(
31) West, R.; Ishikawa, M.; Bailey, R. B. J. Am. Chem. Soc. 1966, 88,
648–4652.
32) The Merck Index, An Encyclopaedia of Chemicals, Drugs, and
4
(
Following the progress of the reaction by CSP-HPLC, we
observed that the slight decrease of the ee is caused by the
gradual racemization of ketone (S)-4 under the reaction
conditions: shortly after the reaction had started, the ee of
the formed TBDMS hydrazone (S)-11 was 98%; however, as
the ee of (S)-4 slowly decreased, so did the ee of (S)-11. After
the completion of the reaction, no further drop in ee
was observed, ruling out that TBDMS hydrazone (S)-11
racemizes under the applied reaction conditions.
Biologicals, 13th ed.; O’Neil, M. J., Smith, A., Heckelman, P. E., Budavari,
S., Eds.; Merck & Co., Inc.: Whitehouse Station, NJ, 2001; pp 851-852.
(
33) The presented procedure does not provide bis-TBDMS hydrazine in
26
a yield better than the procedure of Furrow and Myers, as the formation of
bis-TBDMS hydrazine under the described conditions only goes to 80%
completion, and at least two distillations are needed to obtain pure bis-
TBDMS hydrazine. The presented procedure, however, does not involve the
heating of anhydrous hydrazine. Therefore, it can be regarded as a safer
alternative, more suitable for the large-scale preparation of bis-TBDMS
hydrazine in a laboratory environment.
(
34) Caution: Although the presented procedure avoids the use of anhy-
drous hydrazine, suitable safety precautions should still be taken. It is
strongly recommended to perform the preparation of bis-TBDMS hydrazine
behind a safety blast shield.
Two additional investigations have been conducted to
determine the cause of the decrease of the ee. First, a mixture
8
28 J. Org. Chem. Vol. 75, No. 3, 2010

Pijper et al.
JOCArticle
TABLE 1. Optimization of the Formation of TBDMS Hydrazone
hydrazone formation, trifluoroacetic acid was used as an
additive because it should neutralize any basic impurities
formed during the reaction. However, we found that the
addition of near-equimolar amounts (0.3 and 0.8 equiv) of
trifluoroacetic acid did not lead to an improvement of the ee,
and that the use of more trifluoroacetic acid (3 equiv)
inhibited the formation of (S)-11.
Having been able to obtain TBDMS hydrazone (S)-11 in
84% ee, we subsequently employed (S)-11 in the synthesis
of three different second-generation molecular motors
(
S)-11
bis-TBDMS
hydrazine
(equiv)
Sc(OTf)
(mol %)
3
entry
a
temp (°C)
(S)-11 (% ee)
1
2
2
2
8
8
8
2
2
2
2
2
2
8
0.4
0.1
40
70
60
45
70
70
70
70
70
b
84
84
80
83
84
77
80
55
a
2
a
3
a
4
a
5
b
6
b
7
(Scheme 5). Because of the propensity of second-generation
a
8
molecular motors as well as their episulfide precursors to
crystallize, we expected to be able to readily improve the ee
of these motors by crystallization. In each synthesis, the
Barton-Kellogg reaction was carried out with a freshly
prepared sample of (S)-11, which had been heated in vacuo
at 80 °C for a few hours to remove the bulk of the remaining
bis-TBDMS hydrazine as well as most of the TBDMS
alcohol that was formed during the reaction.
a
Conversion was completed within 45 min. Conversion was com-
pleted overnight.
of ketone (S)-4 and 2 equiv of bis-TBDMS hydrazine was
stirred overnight at elevated temperature in the absence of
scandium(III) triflate. With this experiment, overnight stirr-
ing at 70 and 100 °C resulted only in a marginal loss of
enantiopurity of (S)-4 (97% ee at 70 °C, 94% ee at 100 °C),
far too small to account for the observed decreased ee of (S)-
The Barton-Kellogg reaction of TBDMS hydrazone (S)-
11 with thioketone 12, in which [bis(acetoxy)iodo]benzene
was used to convert (S)-11 to the corresponding diazo
compound, gave episulfide (S)-13 in 33% yield (based on
ketone (S)-4) and 84% ee, thereby confirming our expecta-
tion that this reaction would proceed without any further
decrease in ee. Afterward, the easy crystallization of (S)-13
allowed us to raise the ee up to >99% through two con-
secutive recrystallizations from ethyl acetate. Subsequent
desulfurization by treatment with triphenylphosphine pro-
vided alkene (S)-5. (S)-5 was purified by column chroma-
tography using a 9:1 mixture of silica gel and silver nitrate as
1
1 since the formation of (S)-11 can be completed in 45 min.
However, at 125 °C, the ee of (S)-11 decreased to 23%. A
possible explanation for this observation is the dispropor-
tionation of the bis-TBDMS hydrazine at this temperature,
which would lead to the presence of monoprotected or
unprotected hydrazine in the reaction mixture (vide supra).
In the second experiment, (S)-4 and 20 mol % of scandium-
(III) triflate were dissolved in dichloroethane (without bis-
TBDMS hydrazine) and stirred at 70 °C. Hereby, we obser-
ved a rapid loss of enantiopurity of (S)-4 (50% ee after 3 h).
These findings suggest that, under the applied reaction
conditions, keto-enol tautomerization catalyzed by the
Lewis acid scandium(III) triflate is the major cause of the
3
7
the stationary phase. In this way, (S)-5 was obtained in
87% yield and >99% ee. X-ray crystallography of (S)-5 with
refinement of the Flack parameter was used to confirm that
the absolute configuration of the stereogenic center was
3
5
observed racemization of (S)-4.
3
0
Attempts to optimize the formation of TBDMS hydra-
zone (S)-11 are summarized in Table 1. Lowering the reac-
tion temperature did not improve the ee of the product
indeed S and that (S)-5 showed an M helicity (Figure 3;
see also the Supporting Information). Additionally, it was
found that the methoxy substituent adopts a pseudoaxial
orientation.
(entries 2 and 3) nor did the use of more equivalents of
bis-TBDMS hydrazine (entry 4). When both the amount of
bis-TBDMS hydrazine and scandium(III) triflate was in-
creased, no change in ee was observed (entry 5). However, an
increase in the amount of bis-TBDMS hydrazine combined
with a decrease in the amount of scandium triflate resulted in
much longer reaction times needed for full conversion as well
as a concomitant drop in ee (entry 6). Finally, when the
amount of scandium(III) triflate was decreased 20 times, the
ee was marginally lower, whereas an increase by the same
factor resulted in a substantially lower ee (entries 7 and 8).
Additional attempts to improve the formation of TBDMS
hydrazone (S)-11 were made with the use of cosolvents and
additives. Two cosolvents, DMF and dioxane, were
examined, as we expected that changing the polarity of
the reaction medium would have an effect on the inter-
action between the scandium(III) triflate and ketone (S)-4.
However, with both solvents, no formation of (S)-11 was
observed. In another attempt to optimize the TBDMS
Improvement of the ee of the target alkene by recrystal-
lization of the episulfide was also used in the synthesis of
alkene (S)-6, which has two ester chains attached to the lower
half, thus allowing this motor to be tethered to a nano-
1
4,15
particle, surface, or another molecule.
Episulfide (S)-15
was obtained from the coupling reaction of TBDMS hydra-
zone (S)-11 with thioketone 14. However, after workup, it
was found that we were not able to separate (S)-15 from the
remaining thioketone 14 by column chromatography as their
R was identical under all of the separation conditions
f
examined. To facilitate their separation, we therefore treated
crude (S)-15 with hydrazine monohydrate in order to convert
the unwanted thioketone to the corresponding hydrazone,
after which this hydrazone was easily removed by column
chromatography, providing (S)-15 in 56% yield (based on
ketone (S)-4) and 84% ee. (S)-15 was then recrystallized
twice from methanol in order to improve the ee up to >99%.
Desulfurization by treatment with triphenylphosphine
finally provided alkene (S)-6 in 52% yield and
>99% ee.
3
6
(
35) Furrow and Myers screened a broad range of different Lewis acid
catalysts and found Sc(OTf) to be the most effective; see ref 26.
36) Furrow and Myers have successfully employed diethyl ether as a
3
(
cosolvent in some of their reactions to prepare TBDMS-protected hydrazone
derivatives.
(37) We were not able to find conditions for the separation of (S)-5 and
triphenylphosphine sulfide with regular silica gel or aluminum oxide.
J. Org. Chem. Vol. 75, No. 3, 2010 829

Pijper et al.
JOCArticle
SCHEME 5. Synthesis of Alkenes (S)-5, (S)-6, and (S)-7
The third second-generation molecular motor that has
been successfully prepared using an enantioselective proce-
dure is alkene (S)-7. This compound was synthesized in order
to compare its rotary behavior with that of alkenes 3a-d,
hereby elucidating the effect of a methoxy substituent on a
motor’s rotary action. The Barton-Kellogg reaction was
carried out by subsequently treating TBDMS hydrazone (S)-
in part spontaneously desulfurize; therefore, no efforts have
been made to isolate and fully characterize (S)-17. Instead,
the mixture was treated with triphenylphosphine in order to
desulfurize all of the remaining (S)-17, providing (S)-7
with 84% ee. During our attempts to purify (S)-7, we
observed that (S)-7 would slowly decompose in the presence
of silica gel, which was most likely caused by the acidic nature
of silica gel. It was established, however, that the addition
of triethylamine to the eluent effectively neutralized the
silica gel, thereby minimizing the decomposition of (S)-7
during column chromatography. Finally, the ee of (S)-7
was improved to >99% by a single crystallization from
n-propanol, which, remarkably, provided nearly racemic
crystals and a mother liquor containing enantiomerically
pure (S)-7.
1
1 with [bis(acetoxy)iodo]benzene and an excess of freshly
prepared thioketone 16, which yielded a mixture of episulfide
S)-17 and alkene (S)-7. We observed that, during workup
and purification by column chromatography, (S)-17 would
(
As alkene (S)-7 has a symmetric lower half, the first part of
a full 360° rotation, comprising one photochemically driven
and one thermally driven isomerization step, is identical to
the second part, making only two steps of the full rotary cycle
1
distinguishable. In order to enable us to identify by H NMR
spectroscopy the four distinct steps that make up one full
rotary cycle, the lower half of the motor molecule therefore
had to be desymmetrized. Toward this goal, alkene 8, a
methoxy-substituted analogue of (S)-7, was synthesized
(Scheme 6). The R-methoxy-substituted ketone 4 was
obtained from the unsubstituted ketone 9 by successive
formation of the R-hydroxy dimethyl acetal, methylation,
and hydrolysis, providing 4 in 62% yield. Ketone 4 was
subsequently converted to the corresponding hydrazone 18
by treatment with hydrazine monohydrate at reflux.
After formation of the hydrazone had gone to completion,
overnight cooling of the reaction mixture at -12 °C
FIGURE 3. Perspective PLUTO drawing of alkene (S)-5.
30 J. Org. Chem. Vol. 75, No. 3, 2010
8

Pijper et al.
JOCArticle
SCHEME 6. Synthesis of Alkenes (Z)-8 and (E)-8 (Relative Configuration and Helicity Indicated)
conveniently caused the reaction product to crystallize,
providing 18 in 50% yield. Ketone 19 was obtained in 75%
yield through the methylation of 2-hydroxyfluorenone by
TLC plate provided (Z)-8 in 5% yield. Alkenes (E)-8 and
1
(Z)-8 were identified by the differences in their H NMR
spectra: the methoxy groups of (E)-8 are found at 2.78 and
3.05 ppm, whereas those of (Z)-8 are found at 3.03 and
3.59 ppm. Furthermore, (E)-8 has an distinct aromatic
proton (the lower half aromatic proton that is spatially the
closest to the upper half methoxy substituent) with an
absorption at 8.30 ppm, which is shifted downfield from
the other aromatic protons.
3
8
treatment with KOH and iodomethane, after which it was
converted to thioketone 20 by treatment with Lawesson’s
reagent. It was found that when the formation of 20 was
performed under reflux conditions the desired product was
not obtained, most likely due to degradation of 20 under
3
9
these conditions. When the reaction temperature was
lowered to 80 °C, 20 was obtained, although we found that
The photochemical and thermal isomerization pro-
cesses of alkene (S)-7 were studied by UV-vis and CD
spectroscopy (as well as by H NMR spectroscopy, vide
2
0 rapidly decomposed after workup, even when stored
1
at -12 °C. Therefore, this compound was used directly in a
Barton-Kellogg reaction with hydrazone 18.
The Barton-Kellogg reaction of hydrazone 18 and thio-
ketone 20 provided a mixture consisting of two episulfides
-
5
infra) using a sample of (S)-7 in n-hexane (1.1 ꢀ 10 M)
at -20 and -10 °C, respectively. Irradiation at 365 nm
resulted in a bathochromic shift of the absorption at
391 nm in the UV-vis spectrum to a slightly broader
absorption centered at 402 nm (Figure 4). This spectral
change, which we attribute to an increased twist angle of
the central olefin, is indicative of the formation of the
unstable isomer of (S)-7 by cis-trans photoisomerization
(the E- and Z-isomer) and their corresponding alkenes. After
workup and filtration over a short column of silica gel, the
mixture was directly treated with triphenylphosphine in
toluene for 3 h to provide alkenes (Z)-8 and (E)-8. The
purification of these alkenes proved to be challenging as it
was observed that, upon column chromatography, these
alkenes displayed an even greater extent of degradation on
silica gel than their symmetric analogue, alkene (S)-7. Also,
no separation conditions were found when aluminum oxide
was used as the stationary phase. It was, however, observed
that (Z)-8 and (E)-8 were stable on silica gel TLC plates when
1
0b,41
(Scheme 7).
The changes in the UV-vis spectrum
were accompanied by an inversion of the major absorp-
tions in the CD spectrum (Figure 4), consistent with the
inversion of the helicity that is the result of the photo-
chemically induced cis-trans isomerization. When the
samples were subsequently heated, the absorptions in the
UV-vis and CD spectra returned to their initial states,
which is indicative for the selective conversion of the
unstable isomer to the stable isomer by a thermally
4
0
a heptane/ethyl acetate mixture was used as the eluent and
that preparative silica gel TLC allowed for the complete
separation and purification of (Z)-8 and (E)-8. In this way,
1
0,12
(
(
E)-8 was obtained in 7% yield while further purification of
Z)-8 by preparative TLC using a silver nitrate impregnated
induced helix inversion (Scheme 7).
Characterization of the stable and unstable isomers of
1
alkene (S)-7 by H NMR spectroscopy was performed with a
42
sample of stable (S)-(M)-7 in toluene-d (Figure 5). First, a
8
(
38) Dei, S.; Teodori, E.; Garnier-Suillerot, A.; Gualtieri, F.; Scapecchi,
S.; Budriesi, R.; Chiarini, A. Bioorg. Med. Chem. 2001, 9, 2673–2682.
39) Scheibye, S.; Shabana, R.; Lawesson, S.-O.; Rømming, C. Tetrahe-
dron 1982, 38, 993–1001.
40) We believe that the observed higher stability with TLC silica gel
Analtech UNIPLATE, 2000 μm) could be the result of its higher degree of
1
H NMR spectrum of stable (S)-(M)-7 was recorded
(
(
(41) ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.; Feringa, B. L.
J. Am. Chem. Soc. 2003, 125, 15076–15086.
(
purity, which would cause it to contain less acidic impurities than silica gel
used for column chromatography (SiliCycle SiliaFlash P60, 40-63 μm).
(42) The small absorption observed at 4.10 ppm originates from a trace
amount of dichloromethane.
J. Org. Chem. Vol. 75, No. 3, 2010 831

Pijper et al.
JOCArticle
FIGURE 4. UV-vis absorption spectra at -20 °C (left) and CD spectra at -10 °C (right) of alkene (S)-7 in n-hexane before (solid) and after
dashed) irradiation at 365 nm.
(
SCHEME 7. Photochemical cis-trans Isomerization and
Thermal Helix Inversion of Alkene (S)-7
substituent at the stereogenic center increases. By performing
the measurement at temperatures in the range of 20-50 °C, we
obtained rate constants for various temperatures. Using an
Eyring plot, we found the values of enthalpy of activation
q
q
-1
(
ΔH ) and entropy of activation (ΔS ) to be 90 kJ mol and
-
1
-1
-
3.8 J mol
K , respectively. The Gibbs free energy of
-1
q
activation (ΔG ) at 20 °C was calculated to be 91 kJ mol
.
1
Careful examination of the H NMR spectra of the sample
of alkene (S)-7 before irradiation and the same sample after
irradiation and subsequent heating revealed that both spec-
tra contain a very small amount of the unstable isomer of (S)-7,
even when the sample is kept in the dark for a long time.
4
3
at -30 °C, after which the sample was irradiated overnight
1
1
at -50 °C. Analysis of the H NMR spectrum of the
irradiated sample showed, among others, an upfield shift
Furthermore, we observed by H NMR spectroscopy that
the amount of unstable isomer increases slightly with tempera-
ture, rising from 3% at 20 °C to as much as 5% at 80 °C.
These observations imply that the thermal isomerization step,
which is typically found to be an irreversible process, is actually
reversible with (S)-7, allowing interconversion between the
stable and unstable isomer to a small extent. This implies that
the major forward light-driven rotary process is competing
with a minor, much slower, backward rotation. A rationale for
this finding is that the presence of a methoxy substituent, which
is sterically less demanding than the Ph, Me, i-Pr, and t-Bu
substituents that were present on alkene 3, might lower
the energy of the unstable isomer enough to make it thermally
accessible. It is not the first time this observation has been
made with this type of molecular motor; displacement of
the substituent of a second-generation molecular motor
of the -CH(OCH )- signal from 5.36 to 4.88 ppm, corres-
3
ponding with the expected pseudoaxial to pseudoequatorial
conformational change of the methoxy group that takes
place during the conversion of the (S)-(M)-7 stable isomer
1
2
to the (S)-(P)-7 unstable isomer. The stable/unstable iso-
mer ratio at the photostationary state was determined to be
3
0:70 by integration of the CH signals of these protons at the
stereogenic center. When the irradiated sample was heated to
room temperature, the original spectrum was regenerated,
indicating a thermal (S)-(P)-7 to (S)-(M)-7 conversion.
In order to study the kinetic parameters of the thermal helix
inversion of alkene (S)-7, a sample of (S)-7 in n-hexane was
irradiated at rt until no further change of the UV-vis absorption
spectrum was observed. The temperature was then increased,
after which the change in absorption at this temperature was
monitored at various wavelengths. Using this method, the rate
constant (k) of the first-order thermal helix inversion process at
45
from the R-position to the β-position gave a related result.
Additionally, directional preference (as opposed to uni-
directionality) has also been observed for other motor designs,
7
such as the chemically driven motor reported by our group
-
4 -1
2
0 °C was determined to be 3.47 ꢀ 10
s , which corresponds
8a
to a half-life of 33 min. Comparing this value with that observed
for the analogous alkene 3, it is evident that the rate by which the
thermal step of (S)-7 proceeds is decreased compared to that of
the methyl-substituted molecular motor 3a (half-life=10 min).
This behavior fits into the trend that the half-life of the unstable
and the [3]catenane reported by Leigh and co-workers, and
even for the biological rotary motor F -ATPase in ATP
1
synthase, which is known to occasionally rotate 120° in
12
46
1
the wrong direction. It should be noted that the H NMR
spectra of alkenes (S)-5 and (S)-6 did not show any unstable
isomer, not even at elevated temperatures. This implies
11,12,44
isomers of 3a-d decreases as the relative bulk
of the
(
43) In addition, a 2D NOESY spectrum of (S)-(M)-7 was recorded,
which can be found in the Supporting Information.
44) F o€ rster, H.; V o€ gtle, F. Angew. Chem., Int. Ed. Engl. 1977, 16, 429–
41.
(45) van Delden, R. A.; ter Wiel, M. K. J.; de Jong, H.; Meetsma, A.;
Feringa, B. L. Org. Biomol. Chem. 2004, 2, 1531–1541.
(46) Yasuda, R.; Noji, H.; Kinosita, K., Jr; Yoshida, M. Cell 1998, 93,
1117–1124.
(
4
8
32 J. Org. Chem. Vol. 75, No. 3, 2010

Pijper et al.
JOCArticle
SCHEME 8. Rotary Cycle of Alkene 8
48
performed in the same way as it was done for alkene 7.
A
sample of (Z)-8 in toluene-d was irradiated at -50 °C,
8
showing the formation of the unstable (E)-8 isomer. By
integration of the signals of the stereogenic CH of the stable
(
Z)-8 isomer and unstable (E)-8 isomer at 5.24 and 4.90 ppm,
respectively, the ratio of stable/unstable was determined to
be 41:59 (at PSS). Minor amounts of stable (E)-8 and
unstable (Z)-8 isomers were also observed, indicating, as
was observed with alkene 7, that interconversion between the
stable and unstable isomers is possible due to reversibility of
the thermal step. The amount of stable (E)-8 and unstable
(
Z)-8 was, however, relatively small, indicating that 8 still
strongly prefers one direction over the other in the overall
isomerization pathway. Heating of the sample resulted in the
disappearance of the unstable (E)-8 isomer and the appear-
ance of the stable (E)-8 isomer. Irradiation of a sample of (E)-
8
led to the formation of the unstable (Z)-8 isomer; however,
we also observed that irradiation for a longer period (several
hours) caused degradation of these compounds, significantly
lowering the intensity of the signals that represent the
individual isomers of 8 to the point where these isomers were
no longer observed. The stereogenic CH signals of the stable
(
E)-8 and unstable (Z)-8 were observed at 5.38 and 4.75 ppm,
respectively, and on the basis of their integration, the stable/
unstable ratio was determined to be 45:55. Heating of the
irradiated sample again resulted in the disappearance of the
unstable (Z)-8 isomer and the appearance of the stable (Z)-8
isomer. These findings confirm that a 360° unidirectional
four-step rotary pathway exists, in which each photochemi-
cal isomerization is followed by a thermal isomerization.
Although there are minor competing pathways, for the
methoxy-substituted motors, the major isomerization pro-
cesses are photochemical isomerization from stable to
unstable forms (Z f E and E f Z) and thermal isomeriza-
tion from unstable to stable forms (E f Z and Z f E).
1
FIGURE 5. Partial H NMR spectra of alkene (S)-7 at -30 °C in
toluene-d before irradiation (a), after irradiation at 365 nm (b), and
after irradiation and subsequent heating (c).
8
that the minor thermal conversion toward the unstable
isomer therefore seems to be a characteristic of only (S)-7
and not of methoxy-substituted second-generation motors in
47
general.
Conclusions
Monitoring of the rotary cycle of the unsymmetri-
1
cal alkene 8 (Scheme 8) by H NMR spectroscopy was
Three second-generation light-driven molecular motors
were successfully prepared through an enantioselective
synthesis. In this route, the enantiomerically enriched
(47) For a recent extensive analysis of unidirectionality, competing
processes, and efficiency of rotary motors, see: Geertsema, E. M.; van der
Molen, S. J.; Martens, M.; Feringa, B. L. Proc. Natl. Acad. Sci. U.S.A. 2009,
1
1
(48) H NMR spectra can be found in the Supporting Information.
06, 16919–16924.
J. Org. Chem. Vol. 75, No. 3, 2010 833

Pijper et al.
JOCArticle
upper-half ketone (S)-4 was converted to the corresponding
TBDMS-protected hydrazone (S)-11 with only a slight
drop in ee. We are currently investigating if this approach
is also compatible with other upper-half ketones. Preli-
minary results suggest this might indeed be the case, as an
enantiomerically enriched sample of the methyl-substi-
tuted analogue of (S)-4, used as the upper half of mole-
cular motor 3b, was converted to the corresponding
TBDMS-protected hydrazone with nearly complete reten-
tion of its stereochemical integrity. As the slight drop in ee
could not be avoided, the present approach can only
provide enantiomerically pure second-generation mole-
cular motors if the ee can be improved by recrystallization
of the target alkene or its episulfide precursor. However, as
this class of compounds generally displays a strong pro-
pensity to crystallize, improvement of the ee via this
method is readily achieved for the majority of these
compounds.
(S)-2-Hydroxy-2,3-dihydrocyclopenta[a]naphthalen-1-one (10).
Ketone (()-10 (1.80 g, 9.08 mmol) was dissolved in a mixture of
tert-butyl methyl ether (90 mL) and THF (30 mL). Isopropenyl
acetate (9.12 g, 91.1 mmol) and Amano lipase PS from Pseudo-
monas cepacia (1.36 g) were added, and the mixture was stirred
for 16 h at 22-26 °C. The mixture was subsequently filtered and
concentrated in vacuo. Purification by column chromatography
(
ketone (S)-10 as a yellow solid (781 mg, 43%, > 99% ee): mp
silica gel, heptane/ethyl acetate/methanol 12:4:1) yielded
2
0
1
meric purity was determined by HPLC analysis (Chiralcel OB-
52.7-153.3 °C; [R]
D
=þ132 (c=0.33, ethanol). The enantio-
-
1
H, heptane/isopropanol 90:10, flow rate=0.5 mL min , t =
R
22.9 min (R), t
(
R
=33.5 min (S)).
S)-2-Methoxy-2,3-dihydrocyclopenta[a]naphthalen-1-one (4).
Ketone (S)-10 (780 mg, 3.68 mmol, >99% ee) was dissolved in
iodomethane (10 mL). Calcium sulfate (2.00 g, 14.7 mmol) and
silver(I) oxide (2.45 g, 10.6 mmol) were added, and the resulting
mixture was stirred for 16 h at rt. The mixture was filtered, and
the residue was washed with acetone, after which the filtrate was
concentrated in vacuo. Purification by column chromatography
Studying the photochemistry of alkene (S)-7 as well as that
of its desymmetrized analogue alkene 8, we found that these
compounds display rotational behavior identical to that of
earlier reported second-generation molecular motors. How-
ever, with the introduction of a methoxy substituent, a limit
in the design of these motors seems to have been reached, as
the small size and associated steric effect of this methoxy
substituent lowers the energy of the unstable isomer far
enough to make it thermally accessible. The result of this
modification is that the rotary direction displayed by (S)-7
and 8 is not exclusive, but rather a strong preference for
directionality is seen. These observations provide important
guidelines for the design of future generations of functional
molecular motors.
(
silica gel, dichloromethane) yielded ketone (S)-4 as an orange
1
solid (561 mg, 67%, 98% ee): mp 60-62 °C; H NMR (400
MHz, CDCl ) δ 9.07 (d, J=8.4 Hz, 1H), 8.08 (d, J=8.4 Hz, 1H),
7.89 (d, J=8.1 Hz, 1H), 7.68 (t, J=7.7 Hz, 1H), 7.57 (t, J=7.5
Hz, 1H), 7.48 (d, J=8.4 Hz, 1H), 4.28 (dd, J=7.3, 4.0 Hz, 1H),
3
3
.70 (s, 3H), 3.59 (dd, J=17.1, 7.2 Hz, 1H), 3.13 (dd, J=17.0, 3.9
1
3
Hz, 1H); C NMR (100 MHz, CDCl
36.7 (d), 132.8 (s), 129.3 (s), 129.2 (d), 129.0 (s), 128.3 (d), 126.8
d), 124.1 (d), 123.8 (d), 81.4 (d), 58.2 (q), 33.6 (t); IR (KBr) 3055,
3
) δ 204.2 (s), 153.8 (s),
1
(
2
2
2
-1
þ
930, 2825, 1704, 1518, 1174, 1095 cm ; m/z (EI, %) 212 (M ,
5), 182 (100); HRMS (EI) calcd for C14 212.0837, found
=þ33.5 (c=0.33, ethanol). The enantiomeric
12 2
H O
2
0
D
12.0829; [R]
purity was determined by HPLC analysis (Chiralcel OB-H,
-
1
heptane/isopropanol 90:10, flow rate = 0.5 mL min , t =
R
20.5 min (R), t
1
R
=24.4 min (S)).
,2-Bis(tert-butyldimethylsilyl)hydrazine. Caution: Because of
the instable nature of hydrazine, it is strongly recommended to
perform this procedure behind a safety blast shield. Hydrazine
monohydrochloride (3.43 g, 50.0 mmol) and tert-butyldimethyl-
silyl chloride (15.1 g, 100 mmol) were placed in a round-
bottomed flask equipped with a reflux condenser. Triethylamine
(25 mL) was slowly added at rt, after which the mixture was
stirred at 110 °C for 4.5 h. During the reaction, additional
triethylamine was added after 5, 50, 90, and 140 min (10 mL
each time) of reaction time. The mixture was extracted with
n-pentane (5 ꢀ 100 mL), and the combined layers were concen-
trated in vacuo. The resulting liquid was distilled three times
under reduced pressure (85 °C at 1 mmHg), yielding the target
Experimental Section
(()-2-Hydroxy-2,3-dihydrocyclopenta[a]naphthalen-1-one (10).
,3-Dihydrocyclopenta[a]naphthalen-1-one (9, 5.00 g, 27.4
2
mmol) was dissolved in methanol (180 mL) and added dropwise
to a solution of KOH (4.60 g, 82.0 mmol) in methanol (75 mL) at
5
°C over a period of 15 min. The mixture was stirred for 10 min
and [bis(acetoxy)iodo]benzene (9.60 g, 29.8 mmol) was added in
small portions over a period of 10 min, after which the mixture
was stirred for 16 h at rt. The majority of the methanol was
removed in vacuo. Water (60 mL) was added, and the mixture
was extracted with ethyl acetate (2 ꢀ 50 mL). The combined
organic layers were washed with water (4 ꢀ 50 mL), dried over
1
hydrazine as a colorless liquid (3.5 g, 27%, 2.4 mL): H NMR
MgSO
4
, and concentrated in vacuo. The residue was dissolved in
3
(400 MHz, CDCl ) δ 2.34 (s, 2H), 0.88 (s, 18H), -0.02 (s, 12H);
1
3
THF (20 mL) and acidified with 15% aqueous HCl (40 mL).
After stirring for 1 h, the mixture was extracted with ethyl
acetate (3 ꢀ 30 mL). The combined organic layers were washed
C NMR (100 MHz, CDCl
(EI, %) 260 (M , 83), 203 (48), 130 (48), 83 (100); HRMS (EI)
3
) δ 26.9 (q), 18.1 (s), -5.6 (q); m/z
þ
32 2 2
calcd for C12H N Si 260.2104, found 260.2103.
with water (4 ꢀ 80 mL), dried over MgSO , and concentrated in
(S)-1-(tert-Butyldimethylsilyl)-2-(2-methoxy-2,3-dihydrocyclo-
penta[a]naphthalen-1-ylidene)hydrazine (11). Ketone (S)-4 (40.0
mg, 0.188 mmol, >98% ee), 1,2-bis(tert-butyldimethylsilyl)-
hydrazine (98 mg, 0.380 mmol), and scandium(III) triflate (1.8
mg, 3.7 μmol) were stirred at 70 °C for 45 min. The mixture was
heated in vacuo (1 mmHg) at 80 °C in order to remove the bulk
of the remaining bis-TBDMS hydrazine as well as most of the
4
vacuo. Purification by column chromatography (silica gel,
heptane/ethyl acetate/methanol 12:4:1) yielded ketone (()-10
1
as a yellow solid (3.35 g, 62%): mp 143-145 °C; H NMR (400
MHz, CD Cl ) δ 8.93 (d, J=8.4 Hz, 1H), 8.09 (d, J=8.4 Hz,
2
2
1
(
4
H), 7.91 (d, J=8.1 Hz, 1H), 7.67 (dt, J=8.2, 1.1 Hz, 1H), 7.57
dt, J=7.5, 1.1 Hz, 1H), 7.53 (d, J=8.4 Hz, 1H), 4.57 (dd, J=7.5,
.2 Hz, 1H), 3.64 (dd, J=16.9, 7.3 Hz, 1H), 3.12 (s, 1H), 3.07 (dd,
4
9
TBDMS alcohol, affording the hydrazone as a mixture of syn
and anti diastereoisomers, which were used in the next reaction
1
3
J=17.0, 4.4 Hz, 1H); C NMR (125 MHz, CD Cl ) δ 206.9 (s),
2
2
1
step without further purification: H NMR (400 MHz, CDCl
9.30 (d, J=8.1 Hz, 1H), 7.81 (d, J=8.1 Hz, 1H), 7.68 (d, J=8.4Hz,
1
(
54.4 (s), 136.8 (d), 132.9 (s), 129.3 (s), 129.1 (d), 128.4 (d), 126.9
d), 124.1 (d), 123.8 (d), 74.3 (d), 35.7 (t), one aromatic (s) was
not observed due to overlap with one of the aromatic (d) signals;
) δ
3
-
1
IR (KBr) 3398, 3057, 2918, 1703, 1518, 1176, 1084 cm ; m/z
(
1
49) It was observed, by H NMR, that prolonged concentrating in vacuo
(
þ
EI, %) 198 (M , 100); HRMS (EI) calcd for C13
98.0681, found 198.0675.
H
11
O
2
at this temperature would lead to partial deprotection of the hydrazone;
however, this partial deprotection did not result in lower ee’s or yields.
1
8
34 J. Org. Chem. Vol. 75, No. 3, 2010

Pijper et al.
JOCArticle
1
7
3
H), 7.54 (dt, J=7.5, 1.5 Hz, 1H), 7.46 (dt, J=7.5, 1.3 Hz, 1H),
.31 (d, J = 8.1 Hz, 1H), 6.86 (s, 1H), 5.23 (dd, J = 8.1,
.3 Hz, 1H), 3.44 (dd, J=18.0, 8.1 Hz, 1H), 3.29 (s, 3H), 3.12
133.7 (s), 133.3 (s), 130.4 (d), 129.0 (d), 128.9 (s), 128.4 (d), 128.1
(d), 127.9 (d), 127.7 (d), 126.9 (d), 126.8 (d), 126.5 (d), 126.36 (d),
126.35 (d), 125.1 (d), 124.4 (d), 123.7 (d), 80.3 (d), 55.3 (q), 39.8
þ
(dd, J = 17.6, 3.3 Hz, 1H), 1.02 (s, 9H), 0.31 (s, 6H, syn
diastereoisomer), 0.29 (s, 6H, anti diastereoisomer); m/z
(t); m/z (EI, %) 392 (M , 56), 361 (76), 212 (100); HRMS (EI)
calcd for C27H20OS 392.1235, found 392.1238. The enantio-
meric purity was determined by HPLC analysis (Chiralpak AD,
þ
(
EI, %) 340 (M , 100), 283 (67), 251 (76); HRMS (EI) calcd
-
1
for C20
excess (84%) was determined by HPLC analysis (Chiralpak AD,
heptane/isopropanol 99.5:0.5, flow rate=1.0 mL min , t
min (S), t =5.6 min (R)).
S)-Dispiro[2-methoxy-2,3-dihydro-1H-cyclopenta[a]naphtha-
lene-1,2 -thiirane-3 ,9 -thioxanthene] (13). Hydrazone (S)-11,
freshly prepared from ketone (S)-4 (40.0 mg, 0.188 mmol),
was dissolved in DMF (0.5 mL), and the resulting solution
was cooled to -50 °C. While stirring, a solution of thiox-
anthene-9-thione (14, 43.0 mg, 0.188 mmol) in a mixture of
DMF (0.5 mL) and dichloromethane (0.5 mL) and a solution of
H
28
N
2
OSi 340.1971, found 340.1974. The enantiomeric
R
heptane/isopropanol 99:1, flow rate=1.0 mL min , t =10.7
min (S), t =14.0 min (R)). The structure and absolute config-
R
uration were determined by X-ray crystallographic analysis
using Flack’s refinement (see Supporting Information).
-
1
R
=4.6
R
0
(
(S)-Dispiro[dimethyl 3,3 -(2-methoxy-2,3-dihydro-1H-cyclo-
00
0
0
00
0
0
00
00
penta[a]naphthalen-1,2 -thiirane-3 ,9 -(9 ,10 -dihydroanthra-
0
0
00
cene-9 ,9 -diyl)dipropanoate)] (15). Hydrazone (S)-11, freshly
prepared from ketone (S)-11 (40.0 mg, 0.188 mmol), was
dissolved in DMF (0.5 mL), and the resulting solution
was cooled to -50 °C. While stirring, a solution of dimethyl
0
3,3 -(10-thioxo-9,10-dihydroanthracene-9,9-diyl)dipropanoate
[
bis(acetoxy)iodo]benzene (60.5 mg, 0.188 mmol) in dichloro-
(14, 72.0 mg, 0.188 mmol) in a mixture of DMF (0.5 mL) and
dichloromethane (0.5 mL) and a solution of [bis(acetoxy)-
iodo]benzene (60.5 mg, 0.188 mmol) in dichloromethane
(0.5 mL) were subsequently added (both solutions were cooled
to -50 °C prior to their addition). The mixture was allowed to
slowly warm to rt, after which stirring was continued for 1 h.
Water (5 mL) was added, and the mixture was extracted with
ethyl acetate (2 ꢀ 10 mL). The combined organic layers were
washed with water (3 ꢀ 15 mL), dried over Na SO , and
methane (0.5 mL) were subsequently added (both solutions were
cooled to -50 °C prior to their addition). The mixture was
allowed to slowly warm to rt, after which stirring was continued
for 1 h. Water (5 mL) was added, and the mixture was extracted
with ethyl acetate (2 ꢀ 10 mL). The combined organic layers
were washed with water (3 ꢀ 15 mL), dried over Na
2 4
SO , and
concentrated in vacuo. Purification by column chromatography
(
(
(
silica gel, toluene) yielded episulfide (S)-13 as a yellow solid
26.4 mg, 33% from ketone (S)-4, 84% ee). Part of this product
14.8 mg) was recrystallized twice from ethyl acetate, thereby
2
4
concentrated in vacuo. The crude product was redissolved in
chloroform (2 mL), and hydrazine monohydrate (0.1 mL) was
added, after which the mixture was stirred at rt for 1 min. After
concentration in vacuo, the crude product was purified by
column chromatography (silica gel, heptane/ethyl acetate 3:1),
yielding episulfide (S)-15 as a yellow solid (61 mg, 56% from
ketone (S)-4, 84% ee). This product was recrystallized twice
from methanol to afford episulfide (S)-15 as a white solid (32
affording episulfide (S)-13 as a white solid (6.1 mg, 13.8% from
ketone (S)-4, >99% ee): mp 212-214 °C; H NMR (400 MHz,
3
CDCl ) δ 8.92 (dd, J=8.8, 2.2 Hz, 1H), 7.92 (dd, J=7.7, 1.4 Hz,
1
1
(
H), 7.80 (dd, J=7.7, 1.1 Hz, 1H), 7.55 (d, J=8.4 Hz, 1H), 7.51
dd, J=6.6, 2.9 Hz, 1H), 7.50 (dd, J=7.5, 1.3 Hz, 1H), 7.34 (dt,
J=7.5, 1.5 Hz, 1H), 7.21-7.14 (m, 3H), 6.98 (dd, J=7.7, 1.1 Hz,
1
1
3
1
H), 6.93 (dt, J=7.6, 1.3 Hz, 1H), 6.77 (dt, J=7.5, 1.5 Hz, 1H),
.43 (dd, J=16.3, 4.6 Hz, 1H), 3.25 (s, 3H), 2.95 (d, J=4.4 Hz,
H), 2.87 (d, J=16.5 Hz, 1H), one signal was not observed due to
mg, 29% from ketone (S)-4, >99% ee): mp 144-145 °C; H
3
NMR (400 MHz, CDCl ) δ 9.32 (d, J=8.8 Hz, 1H), 8.26 (dd, J=
7.9, 1.3 Hz, 1H), 8.05 (dd, J=7.3, 1.8 Hz, 1H), 7.53-7.32 (m,
6H), 7.24 (t, J=7.0 Hz, 1H), 7.02-6.99 (m, 2H), 6.93 (dt, J=7.4,
1.3 Hz, 1H), 6.85 (dt, J=7.3, 1.3 Hz, 1H), 3.65 (s, 3H), 3.58 (s,
3H), 3.32 (s, 3H), 3.25 (d, J=5.1 Hz, 1H), 2.93 (dd, J=17.8, 5.0
Hz, 1H), 2.76 (d, J=17.6 Hz, 1H), 2.64-2.51 (m, 2H), 2.14-2.07
1
3
overlap with the solvent signal; C NMR (100 MHz, CDCl ) δ
1
(
1
1
3
3
41.4 (s), 138.9 (s), 136.8 (s), 136.1 (s), 134.3 (s), 133.0 (s), 131.4
s), 131.0 (d), 130.9 (s), 129.8 (d), 128.8 (d), 127.8 (d), 127.3 (d),
27.1 (d), 126.9 (d), 126.89 (d), 126.82 (d), 126.2 (d), 124.9 (d),
24.6 (d), 124.4 (d), 123.3 (d), 86.5 (d), 67.5 (s), 60.8 (s), 57.0 (q),
6.8 (t); m/z (EI, %) 424 (M , 16), 392 (100), 359 (29); HRMS
1
3
3
(m, 1H), 1.86-1.48 (m, 5H); C NMR (100 MHz, CDCl ) δ
þ
174.2 (s), 173.3 (s), 142.6 (s), 141.2 (s), 140.7 (s), 137.6 (s), 133.6
(s), 133.4 (s), 132.4 (d), 131.6 (s), 131.3 (s), 129.8 (d), 128.7 (d),
127.9 (d), 127.74 (d), 127.68 (d), 126.3 (d), 125.9 (d), 125.57 (d),
125.55 (d), 124.9 (d), 124.8 (d), 124.3 (d), 123.0 (d), 83.9 (d), 69.4
(s), 60.0 (s), 56.6 (q), 51.7 (q), 51.6 (q), 45.4 (s), 39.0 (t), 36.0 (t),
(EI) calcd for C27
antiomeric purity was determined by HPLC analysis (Chiralpak
2
H20OS 424.0956, found 424.0943. The en-
-1
AD, heptane/isopropanol 99:1, flow rate=1.0 mL min , t
3.7 min (S), t =15.9 min (R)).
R
=
1
R
þ
0
0
0
0
(
len-1 -ylidene)-9H-thioxanthene (5). A solution of episulfide (S)-
S)-9-(2 -Methoxy-2 ,3 -dihydro-1 H-cyclopenta[a]naphtha-
0
29.9 (t), 29.4 (t), 29.2 (t); m/z (EI, %) 578 (M , 5), 546 (59), 459
þ
(100); m/z (CI, NH
(EI) calcd for C36
antiomeric purity was determined by HPLC analysis (Chiralpak
, %) 596 (M þ NH
, 100), 564 (39); HRMS
3
4
1
3 (6.1 mg, 14.4 μmol, >99% ee) and triphenyl phosphine (37.7
mg, 0.144 mmol) in p-xylene (1 mL) was stirred for 16 h at 125
C, after which the mixture was concentrated in vacuo. The
34 5
H O S 578.2127, found 578.2116. The en-
-
1
°
OD, heptane/isopropanol 95:5, flow rate=1.0 mL min , t
12.0 min (S), t =26.7 min (R)).
=
R
crude product was redissolved in dichloromethane (5 mL), and
iodomethane (2 mL) was added, after which the mixture was
stirred at rt for 1 h. After concentration in vacuo, the residue was
filtered over a short column of silica gel that was impregnated
with 10% silver nitrate (heptane/ethyl acetate 1:1), concentrated
in vacuo, and purified by column chromatography (silica gel,
R
0
0
0
0
0
(S)-Dimethyl-3,3 -(10-(2 -methoxy-2 ,3 -dihydro-1 H-cyclo-
0
penta[a]naphthalen-1 -ylidene)-9,10-dihydroanthracene-9,9-diyl)-
dipropanoate (6). A solution of episulfide (S)-15 (31.8 mg, 54.9
μmol, >99% ee) and triphenyl phosphine (144 mg, 0.549 mmol)
in p-xylene (2.2 mL) was stirred for 16 h at 125 °C, after which
the mixture was concentrated in vacuo. The crude product was
redissolved in dichloromethane (5 mL), and iodomethane (2
mL) was added, after which the mixture was stirred at rt for 1 h.
After concentration in vacuo, the residue was filtered over a
short column of silica gel (diethyl ether), concentrated in vacuo,
and purified by column chromatography (silica gel, heptane/
ethyl acetate 3:1). Recrystallization from isopropanol yielded
heptane/ethyl acetate 8:1), affording alkene (S)-5 as a white solid
(
CDCl
1
5.0 mg, 87%, >99% ee): mp 156-158 °C; H NMR (400 MHz,
3
) δ 7.74 (d, J=8.1 Hz, 1H), 7.70 (d, J=8.8 Hz, 1H), 7.64
apparent d, J=8.1 Hz, 3H), 7.47 (d, J=8.1 Hz, 1H), 7.38 (dt, J=
(
7
1
6
.5, 1.5 Hz, 1H), 7.28 (dt, J=7.5, 1.3 Hz, 1H), 7.17 (dt, J=7.3,
.5 Hz, 1H), 7.06 (dt, J=7.5, 1.5 Hz, 1H), 6.86-6.79 (m, 2H),
.75 (dd, J=7.7, 1.5 Hz, 1H), 6.67 (dt, J=7.2, 1.1 Hz, 1H), 5.67
(
1
d, J=4.0 Hz, 1H), 3.61 (dd, J=15.9, 3.8 Hz, 1H), 3.03 (d, J=
alkene (S)-6 as a yellow solid (16 mg, 52%, >99% ee): mp
) δ 7.75-7.70 (m, 3H),
1
6.1 Hz, 1H), 2.83 (s, 3H); C NMR (125 MHz, CDCl
3
1
3
) δ 144.9
173-174 °C; H NMR (400 MHz, CDCl
3
(
s), 139.5 (s), 139.4 (s), 137.6 (s), 136.0 (s), 135.4 (s), 134.8 (s),
7.53-7.47 (m, 3H), 7.39-7.33 (m, 2H), 7.19-7.11 (m, 2H),
J. Org. Chem. Vol. 75, No. 3, 2010 835

Pijper et al.
JOCArticle
6
Hz, 1H), 3.70 (s, 3H), 3.60 (dd, J=15.6, 3.9 Hz, 1H), 3.50 (s, 3H),
.82-6.73 (m, 3H), 6.60 (dt, J=7.5, 0.9 Hz, 1H), 5.93 (d, J=3.7
7.06-7.02 (m, 4H), 6.65 (t, J=7.5 Hz, 1H), 5.43 (d, J=3.7 Hz,
1H), 3.03 (s, 3H), 2.97 (dd, J=16.4, 3.7 Hz, 1H), 2.91 (d, J=16.3
Hz, 1H); C NMR (125 MHz, CDCl ) δ 146.3 (s), 142.4 (s), 140.6
3
13
3
2
.03 (d, J = 15.8 Hz, 1H), 2.91-2.76 (m, 2H), 2.85 (s, 3H),
1
3
.67-2.46 (m, 4H), 2.14 (t, J=8.3 Hz, 2H); C NMR (125 MHz,
) δ 174.4 (s), 173.4 (s), 144.9 (s), 141.54 (s), 139.9 (s), 139.6
s), 139.4 (s), 138.9 (s), 135.7 (s), 133.3 (s), 132.8 (s), 130.2 (d),
(s), 140.4 (s), 139.9 (s), 137.2 (s), 136.5 (s), 135.3 (s), 133.1 (s), 131.5
(d), 129.2 (s), 128.9 (d), 128.1 (d), 128.0 (d), 127.7 (d), 127.6 (d),
127.1 (d), 126.4 (d), 126.3 (d), 125.7 (d), 125.1 (d), 123.6 (d), 119.7
CDCl
(
3
þ
1
1
1
3
29.0 (s), 128.7 (d), 128.5 (d), 128.3 (d), 127.1 (2d), 126.4 (d),
26.1 (d), 126.0 (d), 125.8 (d), 125.7 (d), 124.7 (d), 124.4 (d),
23.8 (d), 80.53 (d), 56.1 (q), 51.9 (q), 51.8 (q), 47.0 (s), 39.9 (t),
(d), 119.4 (d), 87.1 (d), 54.5 (q), 38.1 (t); m/z (EI, %) 360 (M , 46),
329 (100); HRMS (EI) calcd for C H O 360.1514, found
27
20
360.1496. The enantiomeric purity was determined by HPLC
analysis (Chiralpak OD, heptane/isopropanol 99:1, flow rate=
þ
6.7 (t), 30.6 (t), 29.5 (t), 27.6 (t); m/z (EI, %) 546 (M , 59), 459
-1
(
36), 427 (100); HRMS (EI) calcd for C H O 546.2406, found
5
1.0 mL min , t =10.6 min (S), t =12.5 min (R)).
R R
3
6
34
20
5
purity was determined by HPLC analysis (Chiralpak OD,
heptane/isopropanol 95:5, flow rate=1.0 mL min , t =16.3
R
46.2389; [R] =-108 (c=0.75, CH Cl ). The enantiomeric
(()-2-Methoxy-2,3-dihydrocyclopenta[a]naphthalen-1-one (4).
2,3-Dihydrocyclopenta[a]naphthalen-1-one (9, 4.22 g, 23.2
mmol) was dissolved in methanol (115 mL) and added dropwise
to a solution of KOH (3.90 g, 69.5 mmol) in methanol (60 mL)
over a period of 15 min at 5 °C. The mixture was stirred for 10
min, and [bis(acetoxy)iodo]benzene (8.10 g, 25.1 mmol) was
added in small portions over a period of 10 min, after which the
mixture was stirred for 16 h at rt. The majority of the methanol
was removed in vacuo. Water (100 mL) was added, and the
mixture was extracted with ethyl acetate (2 ꢀ 75 mL). The
combined organic layers were washed with water (4 ꢀ 50 mL),
D
2
2
-
1
min (S), t =39.5 min (R)).
R
0
0
0
0
(S)-9-(2 -Methoxy-2 ,3 -dihydro-1 H-cyclopenta[a]naphtha-
0
len-1 -ylidene)-9H-fluorene (7). Lawesson’s reagent (286 mg,
0
.707 mmol) was added to a solution of fluoren-9-one (84.9
mg, 0.471 mmol) in toluene (1.5 mL), and the resulting mixture
was refluxed for 2.5 h. The mixture was concentrated in vacuo,
after which the residue was purified by flash column chroma-
tography (silica gel, pentane/dichloromethane 1:1, R = 0.8),
f
thereby affording thioketone 16 as a green solid (48.4 mg, 52%).
Hydrazone (S)-11, freshly prepared from ketone (S)-4 (40.0 mg,
4
dried over MgSO , and concentrated in vacuo. The residue was
dissolved in THF (30 mL) and cooled to 0 °C, after which
sodium hydride (1.10 g, 45.8 mmol) was added. After the
resulting mixture was stirred for 5 min, methyl iodide (1.4 mL)
was added and stirring was continued for 16 h. The mixture was
acidified with 10% aqueous HCl (50 mL) and stirred for 3 h,
after which it was extracted with ethyl acetate (3 ꢀ 30 mL). The
combined organic layers were washed with water (4 ꢀ 80 mL),
0
.188 mmol), was dissolved in DMF (0.5 mL), and the resulting
solution was cooled to -50 °C. While stirring, a solution of
bis(acetoxy)iodo]benzene (60.5 mg, 0.188 mmol) in dichloro-
[
methane (0.5 mL), cooled to -50 °C, was added. After stirring
for 20 s, a solution of thioketone 16 (48.4 mg, 0.247 mmol) in a
mixture of DMF (0.5 mL) and dichloromethane (0.5 mL),
cooled to -50 °C, was added. While stirring, the mixture was
allowed to slowly warm to rt, after which the mixture was stirred
for an additional hour. Water (5 mL) was added, and the
mixture was extracted with ethyl acetate (2 ꢀ 10 mL). The
combined organic layers were washed with water (3 ꢀ 15 mL),
dried over Na SO , and concentrated in vacuo. Purification by
4
dried over MgSO , and concentrated in vacuo. Purification
by column chromatography (silica gel, heptane/ethyl acetate/
methanol 12:4:1) yielded ketone (()-4 as a yellow solid (3.05 g,
62%): mp 85.8-86.1 °C.
(()-(2-Methoxy-2,3-dihydrocyclopenta[a]naphthalen-1-ylidene)-
hydrazine (18). Ketone (()-4 (1.00 mg, 4.71 mmol) was dissolved
in a stirred mixture of hydrazine monohydrate (10 mL) and
n-propanol (10 mL) and heated to reflux for 24 h. Stirring was
ceased, and the mixture was cooled to -12 °C, thereby allowing
the product to crystallize from the reaction mixture. Filtration
2
4
column chromatography (silica gel, toluene/heptane 4:1)
yielded 23 mg of a yellow solid, which was identified as a mixture
of episulfide (S)-17 (R =0.35) and alkene (S)-7 (R =0.4) by H
1
f
f
1
NMR spectroscopy. Episulfide (S)-17: H NMR (400 MHz,
CDCl ) δ 9.67 (d, J=8.4 Hz, 1H), 7.81 (d, J=8.1 Hz, 1H),
.71-7.57 (m, 5H), 7.48-7.35 (m, 3H), 7.18 (d, J=8.1 Hz, 1H),
.15 (d, J=8.1 Hz, 1H), 7.08 (t, J=7.5 Hz, 1H), 6.60 (t, J=7.7
3
afforded hydrazone (()-18 as brown solid (536 mg, 50%): mp
1
7
7
76-78 °C; H NMR (400 MHz, CDCl
3
) δ 9.19 (d, J=8.4 Hz,
1H), 7.83 (d, J=8.1 Hz, 1H), 7.76 (d, J=8.4 Hz, 1H), 7.57 (dt, J=
7.7, 1.1 Hz, 1H), 7.48 (dt, J=7.3, 1.1 Hz, 1H), 7.34 (d, J=8.4 Hz,
1H), 6.26 (s, 2H), 5.16 (dd, J=7.0, 2.6 Hz, 1H), 3.49 (dd, J=17.4,
7.5 Hz, 1H), 3.42 (d, J=1.5 Hz, 3H), 3.15 (dd, J=17.6, 3.3 Hz,
Hz, 1H), 4.55 (d, J=4.0 Hz, 1H), 3.57 (s, 3H), 2.77 (d, J=16.1
Hz, 1H), 2.63 (dd, J=15.4, 3.3 Hz, 1H); m/z (EI, %) 392 (M , 2),
þ
3
60 (43), 329 (100); HRMS (EI) calcd for C27H20OS 392.1235,
1
3
found 392.1239. The mixture of episulfide and alkene was
dissolved in toluene (2.5 mL) with triphenyl phosphine (150
mg, 0.571 mmol) and refluxed for 16 h, after which the mixture
was concentrated in vacuo. The crude product was redissolved
in dichloromethane (1 mL), and iodomethane (1 mL) was
added, after which the mixture was stirred at rt for 1 h. After
concentration in vacuo, the residue was filtered over a short
column of silica gel (dichloromethane), concentrated in vacuo,
and filtered over a short column of silica gel that was impreg-
nated with 10% silver nitrate (heptane/ethyl acetate 1:1). Puri-
fication was performed by column chromatography (silica gel,
toluene/heptane 3:1 þ 3% triethylamine) and afforded alkene
1H); C NMR (100 MHz, CDCl ) δ 153.9 (s), 142.4 (s), 133.0
3
(s), 131.1 (s), 129.8 (d), 128.5 (s), 128.1 (d), 127.0 (d), 125.51 (d),
125.50 (d), 123.1 (d), 77.0 (d), 54.3 (q), 34.8 (t); IR (KBr) 3402,
-
1
;
3300, 3240, 3051, 2931, 2823, 1701, 1597, 1514, 1196, 1086 cm
m/z (EI, %) 226 (M , 100), 165 (96); HRMS (EI) calcd for
þ
C
14
H
14
2
N O 226.1106, found 226.1105.
2-Methoxyfluoren-9-one (19). KOH (286 mg, 5.10 mmol) was
grounded and suspended in dimethyl sulfoxide (2.5 mL). After
stirring for 10 min, 2-hydroxyfluoren-9-one (250 mg, 1.27
mmol) and iodomethane (0.25 mL) were added and stirring
was continued for 20 min. Water (10 mL) was added, and the
resulting mixture was extracted with dichloromethane (2 ꢀ 10
mL). The combined organic layers were washed with water (2 ꢀ
(
S)-7 as a yellow oil (5.1 mg, 8% from ketone (S)-4, 84% ee) with
1
only a minor impurity remaining (visible by H NMR spectro-
scopy in the region of 6.7-7.0 ppm). The compound was
crystallized from n-propanol, after which the mother liquor
was concentrated in vacuo, yielding alkene (S)-7 as a yellow
2 4
10 mL), dried over Na SO , and concentrated in vacuo. Pur-
ification by column chromatography (silica gel, pentane/di-
chloromethane 1:1) yielded ketone 19 as a orange solid (200
1
mg, 75%): mp 78.7-79.0 °C; H NMR (400 MHz, CDCl
3
) δ
1
solid (3.2 mg, 5% from ketone (S)-4, >99% ee); H NMR (500
MHz, toluene-d ) δ 8.35 (d, J=7.7 Hz, 1H), 7.87 (d, J=8.5 Hz,
7.59 (d, J=7.3 Hz, 1H), 7.44-7.38 (m, 3H), 7.21-7.18 (m, 2H),
6.97 (dd, J=8.1, 2.2 Hz, 1H), 3.85 (s, 3H); C NMR (100 MHz,
13
8
1
7
H), 7.63 (apparent t, J=7.5 Hz, 2H), 7.59 (d, J=8.2 Hz, 1H),
.54 (d, J=7.5 Hz, 1H), 7.33 (t, J=7.4 Hz, 1H), 7.27 (t, J=7.4
CDCl ) δ 193.6 (s), 160.7 (s), 144.7 (s), 136.7 (s), 135.7 (s), 134.7
(d), 134.1 (s), 127.7 (d), 124.1 (d), 121.2 (d), 119.9 (d), 119.5 (d),
109.3 (d), 55.5 (q); IR (KBr) 3055, 2937, 2835, 1716, 1603, 1230
3
Hz, 1H), 7.22 (d, J = 8.4 Hz, 1H), 7.19 (t, J = 7.6 Hz, 1H),
36 J. Org. Chem. Vol. 75, No. 3, 2010
8

Pijper et al.
JOCArticle
CD Monitoring of the Rotary Cycle. A molar CD spectrum of
-1
þ
cm ; m/z (EI, %) 210 (M , 100); HRMS (EI) calcd for
210.0681, found 210.0670.
-
5
C
14
H
10
O
2
alkene (S)-7 in n-hexane (1.1 ꢀ 10 M) was taken at -10 °C
before and after irradiation at 365 nm, thereby showing forma-
tion of the unstable isomer through cis-trans isomerization by
inversion of the CD signal. When the sample was warmed to rt,
another inversion of the CD signal was observed and the
original spectrum was regained, indicating the formation of
0
0
0
0
(
()-(E)-2-Methoxy-9-(2 -methoxy-2 ,3 -dihydro-1 H-cyclo-
0
penta[a]naphthalen-1 -ylidene)-9H-fluorene (8). Lawesson’s re-
agent (335 mg, 0.829 mmol) was added to a solution of ketone 19
(
116 mg, 0.552 mmol) in toluene (2 mL), and the resulting
mixture was stirred at 80 °C for 3 h. The mixture was concen-
trated in vacuo, after which the residue was filtered over a
-
1
-1
the stable isomer. (S)-7: λmax [nm] (Δε [L mol cm ]) (S)-7st,
column with silica gel (pentane/dichloromethane 1:1, R
to provide a red oil which contained thioketone 20. Hydrazone
()-18 (50.0 mg, 0.221 mmol) was dissolved in DMF (0.5 mL),
f
=0.8)
218 (-26.7), 234 (þ93.7), 275 (-85.9); (S)-7unst, 218 (þ62.9), 239
(-56.6), 268 (þ29.1), 279 (þ39.8).
(
UV Monitoring of the Rotary Cycle. A molar UV spectrum of
-
5
and the resulting solution was cooled to -50 °C. While stirring,
a solution of [bis(acetoxy)iodo]benzene (71.2 mg, 0.221 mmol)
in dichloromethane (0.5 mL), cooled to -50 °C, was added.
After stirring for 20 s, a solution of the crude thioketone 20 in a
mixture of DMF (0.5 mL) and dichloromethane (0.5 mL),
cooled to -50 °C, was added. While stirring, the mixture was
allowed slowly to warm to rt, after which the mixture was stirred
for an additional hour. Water (5 mL) was added, and the
mixture was extracted with ethyl acetate (2 ꢀ 10 mL). The
combined organic layers were washed with water (3 ꢀ 15 mL),
alkene (S)-7 in n-hexane (1.1 ꢀ 10 M) was taken at -20 °C
before and after irradiation at 365 nm, thereby showing forma-
tion of the unstable isomer through cis-trans isomerization by a
bathochromic shift of the absorption spectrum. When the
sample was warmed to rt, the original spectrum was regained,
indicating the formation of the stable isomer: λmax [nm] (ε [L
-
1
-1
mol cm ]): 7st, 391 (25 800); 7unst, 402 (22 300).
Determination of the Kinetic Parameters of the Thermal Helix
Inversion by UV Monitoring. A sample of alkene (S)-7 in n-
-
5
hexane (1.1 ꢀ 10 M) was irradiated at 365 nm at rt until the
PSS was reached. The sample was inserted into the sample
holder of the UV spectrometer, which had been kept at a
temperature of 20, 30, 40, or 50 °C. The change in the absorp-
tion, associated with the conversion of unstable isomer into
stable isomer, was monitored over time at 430, 433, 437, 440,
450, and 500 nm. The change in the absorption at 500 nm was
used as the baseline. A clear isosbestic point was observed at 407
nm (see Supporting Information).
2 4
dried over Na SO , and concentrated in vacuo. After filtration
over a short column of silica gel (toluene), the resulting oil and
triphenyl phosphine (175 mg, 0.753 mmol) were dissolved in
toluene (3 mL) and the mixture wsa stirred at 80 °C for 3 h,
followed by concentration in vacuo. The crude product was
redissolved in dichloromethane (1 mL), and iodomethane (1
mL) was added, after which the mixture was stirred at rt for 1 h.
After concentration in vacuo, the residue was filtered over a
short column of silica gel (dichloromethane), concentrated in
vacuo, and filtered over a short column of silica gel that was
impregnated with 10% silver nitrate (heptane/ethyl acetate 1:1).
Purification and separation of the diastereomers was performed
by preparative TLC (silica gel, hexane/ethyl acetate 20:1 þ 0.1%
1
H NMR Monitoring of the Rotary Cycle. A sample of (S)-7,
(E)-8, or (Z)-8 in toluene-d was cooled to -50 °C in an
8
1
ethanol bath and irradiated overnight at 365 nm. H NMR
spectra of the sample, taken at -30 °C before and after
irradiation, showed the formation of a second compound of
1
triethylamine), thereby affording alkene (()-(E)-8 as a yellow oil
which the H NMR spectra correspond to that of the unstable
1
(
5.9 mg, 7%): H NMR (500 MHz, toluene-d ) δ 8.30 (d, J=6.4
isomer, indicating the conversion of stable isomer into un-
stable isomer. In the case of (S)-7, subsequent heating of the
sample resulted in the regeneration of the original spectrum.
In the case of (Z)-8 (or (E)-8), subsequent heating of the
sample resulted in the disappearance of the unstable isomer
8
Hz, 1H), 7.84 (d, J=8.3 Hz, 1H), 7.60-7.55 (m, 3H), 7.43 (d, J=
.3 Hz, 1H), 7.30-7.25 (m, 2H), 7.21 (d, J=8.1 Hz, 1H), 7.12 (t,
J=6.8 Hz, 1H), 7.00 (d, J=8.1 Hz, 1H), 6.83 (dd, J=8.2, 2.3 Hz,
8
1
3
2
H), 6.53 (d, J=2.4 Hz, 1H), 5.46 (d, J=3.9 Hz, 1H), 3.05 (s,
H), 2.97 (dd, J=16.5, 4.0 Hz, 1H), 2.92 (d, J=16.4 Hz, 1H),
1
and the formation of stable (E)-8 (or (Z)-8). 7st: H NMR (500
13
.78 (s, 3H); C NMR (125 MHz, toluene-d
8
) δ 159.2 (s), 146.6
8
MHz, toluene-d ) δ 8.39 (d, J=7.8 Hz, 1H), 7.90 (d, J=8.6 Hz,
(
1
(
s), 142.7 (s), 141.2 (s), 140.3 (s), 138.9 (s), 136.9 (s), 136.2 (s),
34.2 (s), 133.2 (s), 131.2 (d), 129.1 (s), 129.0 (d), 128.3 (d), 128.2
d), 127.2 (d), 126.5 (d), 125.7 (d), 125.5 (d), 123.4 (d), 120.3 (d),
1H), 7.63-7.61 (m, 2H), 7.57 (d, J=8.1 Hz, 1H), 7.52 (d, J=
7.6 Hz, 1H), 7.38 (t, J=7.6 Hz, 1H), 7.30 (t, J=7.5 Hz, 1H),
7.22-7.18 (m, 2H), 7.08 (d, J=7.8 Hz, 1H), 7.03 (d, J=8.1 Hz,
1H), 6.65 (t, J=7.6 Hz, 1H), 5.36 (s, 1H), 2.99 (s, 3H), 2.87 (s,
2H), one aromatic H signal was not observed due to overlap
1
m/z (EI, %) 390 (M , 57), 359 (100).
19.2 (d), 116.2 (d), 110.8 (d), 86.6 (d), 53.9 (q), 53.5 (q), 38.2 (t);
þ
0
0
0
0
(
()-(Z)-2-Methoxy-9-(2 -methoxy-2 ,3 -dihydro-1 H-cyclo-
with the solvent signal.
0
1
penta[a]naphthalen-1 -ylidene)-9H-fluorene (8). After separation
of the diastereomers by preparative TLC as described above,
additional purification was performed by preparative TLC
using a silica gel TLC plate that was impregnated with 10%
silver nitrate (hexane/ethyl acetate 8:1 þ 0.1% triethylamine),
7
unst: H NMR (500 MHz, toluene-d
8
) δ 7.94 (d, J=8.5 Hz,
1H), 7.76 (d, J=7.7 Hz, 1H), 7.64-59 (m, 2H), 7.50 (d, J=7.4
Hz, 1H), 7.40 (t, J=7.4 Hz, 1H), 7.29 (t, J=7.1 Hz, 1H), 7.17 (t,
J=7.6 Hz, 1H), 6.66 (t, J=7.1 Hz, 1H), 4.88 (t, J=6.2 Hz, 1H),
3.04 (dd, J=14.8, 4.9 Hz, 1H), 3.01 (s, 3H), 2.93 (dd, J=15.4, 7.4
Hz, 1H), five aromatic H signals were not observed due to
thereby affording alkene (()-(Z)-8 as a yellow oil (4.2 mg, 5%):
1
8
H NMR (500 MHz, toluene-d ) δ 7.94 (d, J=2.2 Hz, 1H), 7.91
overlap with the solvent signal.
(E)-8st: H NMR (500 MHz, toluene-d
1
(
d, J=8.6 Hz, 1H), 7.64 (d, J=8.1 Hz, 1H), 7.60 (d, J=8.1 Hz,
8
) δ 8.35 (d, J=7.6 Hz,
1
8
6
H), 7.52 (d, J=8.2 Hz, 1H), 7.48 (d, J=7.5 Hz, 1H), 7.23 (d, J=
.2 Hz, 1H), 7.19 (t, J = 7.2 Hz, 1H), 7.07-7.03 (m, 2H),
.96-6.93 (m, 2H), 6.61 (t, J = 7.3 Hz, 1H), 5.37 (d, J = 4.0
1H), 7.85 (d, J=8.5 Hz, 1H), 7.57-7.55 (m, 2H), 7.53 (d, J=8.1
Hz, 1H), 7.42 (d, J=8.1 Hz, 1H), 7.35 (t, J=7.3 Hz, 1H), 7.31 (t,
J=7.2 Hz, 1H), 7.19 (d, J=8.1 Hz, 1H), 7.10 (t, J=7.4 Hz, 1H),
6.94-6.89 (m, 2H), 6.59 (d, J=2.2 Hz, 1H), 5.38 (s, 1H), 3.01 (s,
Hz, 1H), 3.59 (s, 3H), 3.03 (s, 3H), 2.98 (dd, J=16.6, 4.1 Hz, 1H),
2
1
.92 (d, J=16.3 Hz, 1H); C NMR (125 MHz, toluene-d
3
8
) δ
3H), 2.88 (s, 2H), 2.71 (s, 3H).
1
1
(
60.4 (s), 146.5 (s), 142.7 (s), 142.1 (s), 141.3 (s), 137.0 (s), 136.2
s), 134.1 (s), 133.3 (s), 131.4 (d), 129.5 (s), 128.2 (d), 128.0 (d),
(Z)-8unst: H NMR (500 MHz, toluene-d ) observed signals δ
8
7.98 (d, J=8.2 Hz, 1H), 6.60 (s, 1H), 4.75 (s, 1H), 3.60 (s, 3H),
3.00 (s, 3H).
1
1
27.2 (d), 126.5 (d), 125.7 (d), 125.4 (d), 123.3 (d), 120.5 (d),
18.9 (d), 114.7 (d), 111.2 (d), 87.1 (d), 55.1 (q), 53.5 (q), 38.0 (t),
1
(Z)-8 : H NMR (500 MHz, toluene-d ) δ 7.95 (d, J=8.4 Hz,
st
8
two signals were not observed due to overlap with the solvent
signal; m/z (EI, %) 390 (M , 52), 359 (100); HRMS (EI) calcd
for C28H O 390.1620, found 390.1618.
22 2
1H), 7.92 (s, 1H), 7.63 (d, J=8.2 Hz, 1H), 7.57 (d, J=8.2 Hz,
1H), 7.52 (d, J=8.2 Hz, 1H), 7.48 (d, J=7.7 Hz, 1H), 7.23 (d, J=
8.2 Hz, 1H), 7.19 (t, J=7.4 Hz, 1H), 7.08 (apparent s, 1H), 7.02
þ
J. Org. Chem. Vol. 75, No. 3, 2010 837

Pijper et al.
JOCArticle
(
3
t, J=7.8 Hz, 1H), 6.62 (t, J=7.5 Hz, 1H), 5.24 (s, 1H), 3.54 (s,
H), 2.97 (s, 3H), 2.87 (s, 2H), two signals were not observed due
to overlap with the solvent signal.
Acknowledgment. Financial support from The Netherlands
Organization for Scientific Research (NWO-CW) and
NanoNed is gratefully acknowledged.
1
(
E)-8unst: H NMR (500 MHz, toluene-d ) δ 7.90 (d, overlaps
8
with s signal from (Z)-8st, 1H), 7.74 (d, J = 7.7 Hz, 1H),
.58-7.57 (m, 3H), 7.44 (d, J = 8.2 Hz, 1H), 7.37 (t, J = 7.2
Hz, 1H), 7.30 (t, J=7.3 Hz, 1H), 6.48 (s, 1H), 4.90 (s, 1H), 3.05 (s,
Supporting Information Available: General experimental,
spectroscopic data for compounds 4-8, 10, 11, 13, 15, 18, and
19, X-ray crystallographic data for alkene 5, and UV-vis, CD,
7
1
3
H), 3.05-2.97 (m, overlaps with s signals from (Z)-8 and (E)-
and H NMR spectra of the stable and unstable isomers of
st
8unst, 2H), 2.68 (s, 3H), four signals were not observed due to
overlap with the solvent signal.
alkenes 7 and 8. This material is available free of charge via the
Internet at http://pubs.acs.org.
8
38 J. Org. Chem. Vol. 75, No. 3, 2010