Investigation of bicyclic thioketones as triggers for liquid crystal optical switches

Article abstract of DOI:10.1021/jo0264675

The axially chiral bicyclic thioketones 11 and 15 were prepared and investigated for suitability as chiroptical triggers in a liquid crystal optical switch. Irradiation of partially resolved 15 with unpolarized light leads to its conversion to the racemic form (photoracemization). However, irradiation of racemic thioketones 11 and 15 with circularly polarized light does not lead to detectable photoresolution. The lack of photoresolution was traced to inefficiency in intramolecular, throughbond triplet energy transfer. These thioketones are not suitable for use as phototriggers.

Full text of DOI:10.1021/jo0264675

In vestiga tion of Bicyclic Th iok eton es a s Tr igger s for Liqu id
Cr ysta l Op tica l Sw itch es
Rochelle Fisher Bradford and Gary B. Schuster*
School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332
gschuster@cosdean.cos.gatech.edu
Received September 23, 2002
The axially chiral bicyclic thioketones 11 and 15 were prepared and investigated for suitability as
chiroptical triggers in a liquid crystal optical switch. Irradiation of partially resolved 15 with
unpolarized light leads to its conversion to the racemic form (photoracemization). However,
irradiation of racemic thioketones 11 and 15 with circularly polarized light does not lead to detectable
photoresolution. The lack of photoresolution was traced to inefficiency in intramolecular, through-
bond triplet energy transfer. These thioketones are not suitable for use as phototriggers.
usually produce a cholesteric texture.1
7-21
In a function-
In tr od u ction
ing chiroptical switch, irradiation of the racemic trigger
with CPL in a nematic liquid crystal will convert the
liquid crystal to its cholesteric texture; irradiation of the
trigger with normal light will racemize it and result in
the reformation of the nematic texture. The difference
between nematic and cholesteric liquid crystals is readily
sensed nondestructively by optical methods.2
There is widespread interest in the development of
materials that can be switched by light between inde-
pendent, stable states and can be detected optically with-
out their destruction or interconversion.1 Such materi-
als have potential for incorporation in devices that may
be useful for display or memory applications. We have
been developing a series of chiroptical triggers for liquid
crystalline materials that may be useful for such
devices.1
A chiroptical trigger is an optically active compound
whose enantiomers are interconverted by irradiation with
visible or ultraviolet light. Irradiation of one enantiomer
of such a compound with unpolarized light will convert
it to the racemic mixture (photoracemization). Irradiation
of the racemic mixture with circularly polarized light
-8
2
We have developed chiroptical triggers based on these
principles and we have shown their utility in the devel-
,9-11
1
opment of an optical switch. In this earlier report, the
chiroptical trigger compound is a ketone linked by a
bicyclic unit to a photoisomerizable olefin, see structure
1
. Irradiation of the ketone chromophore with either CPL
or unpolarized light in a liquid crystal solution causes
through-bond triplet energy transfer and isomerization
of the olefinic group, which triggers the reversible
conversion of nematic and cholesteric textures. In this
system, the lowest energy nπ* ketone chromophore
absorbs light at ca. 300 nm, which is a spectral region
where most liquid crystalline materials also absorb. This
(CPL) will cause a partial photoresolution with the
enantiomeric excess at the photostationary state ([γ]pss
determined by the Kuhn anisotropy factor (g ) ∆ꢀ
according to [γ]pss ) g /2, where ∆ꢀ
)
)
λ
λ
/ꢀλavg
λ
λ
is the difference
between molar extinction coefficients for an enantiomer
with right- and left-handed CPL at wavelength λ, deter-
mined by circular dichroism (CD) spectroscopy, and ꢀλavg
is the average extinction coefficient of the two enanti-
(1) Burnham, K.; Schuster, G. B. J . Am. Chem. Soc. 1999, 121, 1.
(2) Anzai, N.; Machida, S.; Horie, K. Chem. Lett. 2001, 9, 888-889.
(
(
(
(
(
(
3) Goto, H.; Yashima, E. J . Am. Chem. Soc. 2002, 124, 7943-7949.
4) Ball, P.; Garwin, L. Nature 1992, 355, 76.
1
2-16
omers determined by absorption spectroscopy.
5) Willner, I.; Willne, B. Adv. Mater. 1995, 7, 587.
6) Tamaoki, N. Adv. Mater. 2001, 13, 1135.
7) Ruslim, C.; Ichimura, L. Adv. Mater. 2001, 13, 37.
8) Chen, C. T.; Chou, Y. C. J . Am. Chem. Soc 2000, 122, 7662-
7
672.
(
(
(
(
(
9) Zhang, Y.; Schuster, G. B. J . Org. Chem. 1995, 60, 7192.
10) Zhang, M.; Schuster, G. B. J . Am. Chem. Soc. 1994, 116, 4852.
11) Li, J .; Schuster, G. B. J . Am. Chem. Soc. 2000, 122, 2603.
12) LeBel, J . A. Bull. Soc. Chim. (Paris) 1874, 22, 337.
13) van’t Hoff, J . H. Die Lagerung der Atome im Raume, 2nd ed.;
Vieweg & Sohn: Braunschweig, Germany, 1894.
(
(
(
(
(
14) Kuhn, W.; Braun, E. Naturwissenshaften 1929, 17, 227.
15) Kuhn, W.; Braun, E. Angew. Chem. 1929, 42, 828.
16) Kuhn, W.; Braun, E. Angew. Chem. 1929, 42, 296.
17) Demus, D. Mol. Cryst. Liq. Cryst. 1988, 165, 45.
18) Solladie, G.; Zimmermann, G. Angew. Chem., Int. Ed. Engl.
1
984, 23, 348.
(
(
(
(
19) Reinitzer, F. Monatsh. Chem. 1888, 9, 421.
20) Lehmann, O. Z. Phys. Chem. 1889, 4, 462.
21) Friedel, G. Ann. Phys. 1922, 18, 273.
Incorporation of an optically active form of such a
trigger in an achiral liquid crystalline material will
22) Huck, N. P. M.; J ager, W. F.; deLange, B.; Feringa, B. L. Science
1996, 273, 1686.
1
0.1021/jo0264675 CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/04/2003
J . Org. Chem. 2003, 68, 1075-1080
1075

Bradford and Schuster
property of the known chiroptical triggers limits their
utility because overlap of the trigger and liquid crystal-
line material absorption spectra at the irradiation wave-
length cannot be tolerated in a switch. Consequently, we
have sought chiroptical triggers that absorb in the visible
region of the spectrum where few liquid crystalline
materials absorb light. The attempt to develop such a
2. Design a n d Syn th esis of Th iok eton e-Con ta in -
in g Ch ir op tica l Tr igger s. Our first attempt to develop
a thioketone-containing trigger compound focused on
bicyclic compound 4, which incorporates a substituted
styrene group appended to the 6-position of a bicyclo-
[3.2.1]octane framework. This structure was selected as
a model for the conversion of a rigid aliphatic ketone to
a thioketone because it mimics the bicyclic core of
previously prepared triggers.
trigger based on the thiocarbonyl chromophore (λmax
)
4
80 nm for the nπ* absorption band) is reported here.
Ketone 5 was prepared according to a procedure that
has been described previously.11 Its treatment with
Resu lts
. Th iok et on es a s P h ot osen sit izer s for Olefin
Lawesson’s reagent, C14
14 2 2 4
H O P S , or any of a number
1
of other thionating reagents, yields primarily the dimeric
Isom er iza tion . It is generally known that triplet-triplet
energy transfer from a sensitizer to an olefinic compound
occurs efficiently and causes double bond isomerization
when the triplet energy of the sensitizer is greater than
that of the olefin by at least ca. 2-3 kcal/mol.23 The triplet
energy of the thiocarbonyl chromophore in aliphatic
compounds is ca. 57 kcal/mol.24 The spectroscopic (verti-
cal) triplet energy of trans-â-methylstyrene, a model for
the isomerizable component of previously studied chi-
roptical triggers, is estimated to be ca. 60.5 kcal/mol by
sulfide 6 (eq 1) or no reaction at all. Formation of dimeric
2
O perturbation of its absorption spectrum, but the triplet
energy of its relaxed, twisted conformation is estimated
to be 53 kcal/mol.25 Hence, we studied the ability of
2
-adamantanethione (a model for the bicyclic thioketone
chromophore of possible chiroptical triggers) to sensitize
the isomerization of â-methylstyrene by a “non-vertical
energy transfer” mechanism.
Irradiation of 2-adamantanethione (0.04 M) in a deoxy-
6
genated benzene-d solution (for ease of analysis by NMR
sulfides is a common decomposition route for aliphatic
thioketones having enolizable protons.2
6,27
Consequently,
we modified the structure of the target compounds to
avoid this problem. The 7-thiocarbonylbicyclo[3.2.1]oc-
tane structure shown in eq 2 will not dimerize to the sul-
fide readily because formation of the enol violates Bredt’s
rule.28 This skeleton forms the core of a family of potential
trigger compounds we prepared and investigated.
spectroscopy) containing trans-â-methylstyrene (0.2 M)
with visible light for 3 h did not result in the forma-
tion of a detectable amount of cis-â-methylstyrene. A con-
trol experiment showed that irradiation (at 300 nm) of a
solution containing 2-adamantanone (triplet energy ) 80
2
4
kcal/mol ) in place of the thioketone results in efficient
isomerization of the olefin. Evidently, sensitization of the
styrene isomerization by 2-adamantanethione is inef-
ficient, probably because its triplet energy is too low.
The vertical triplet energies of trans-piperylene and cis-
piperylene are reported to be 54 and 53 kcal/mol,
respectively, based on triplet sensitization experiments.24
We find that irradiation of 2-adamantanethione in solu-
tions containing piperylene causes the isomerization of
the cis isomer but not the trans form. From this result,
we assume that dienes with vertical triplet energies of
5
3 kcal/mol or lower will be isomerizable by triplet
sensitization with thioketones. To confirm this hypoth-
esis, we studied the sensitized isomerization of a mixture
of E and Z isomers of 1-phenyl-4-methyl-1,3-butadiene
(structure 2 and 3). We find that this mixture of olefins
is isomerized by both 2-adamantanethione and 2-ada-
mantanone. On this basis, we conclude that aryl-
substituted dienes have suitable triplet energies for
sensitization by thioketones and we designed a series of
possible chiroptical triggers that incorporate these groups
into a bicyclic structure.
(
23) Turro, N. J . Modern Molecular Photochemistry; University
Science Books: Sausalito, CA, 1991.
24) Murov, S.; Carmichael, I.; Hug, G. Handbook of Photochemistry;
Marcel Dekker: New York, 1993.
25) Ni, T.; Caldwell, R.; Melton, L. J . Am. Chem. Soc. 1989, 111,
57-464.
Our synthesis of the racemic bicyclo[3.2.1]octane-
containing triggers begins with monoprotected diketone
(
(
(26) Selzer, T.; Rappoport, Z. J . Org. Chem. 1996, 61, 5462.
(27) Lawesson, S. Tetrahedron 1982, 38, 1002.
4
1
076 J . Org. Chem., Vol. 68, No. 3, 2003

Thioketones as Triggers for Liquid Crystal Optical Switches
SCHEME 1
7
, which has been prepared previously.11 The Horner-
Similarly, attempts to separate diastereomeric deriva-
tives of the ketone precursor to 11 failed to provide
material that could be transformed into optically active
thioketone.
Wadsworth-Emmons reaction of 7 to form the benzo-
substituted diene was followed by deprotection and
conversion to the thioketone 8 with Lawesson’s Reagent,
as is shown in eq 2. Other methods to convert the ketone
to the thioketone gave mostly trithiolane byproducts.29
We designed and prepared a potential chiroptical
trigger containing a benzofuran group on the hypothesis
that heteroatoms in the aryl-substituted diene portion
would provide an additional interaction for resolution
using chiral chromatographic methods and increase the
likelihood of its success. In addition, the synthetic route
selected for preparation of this trigger proceeds through
an intermediate carboxylic acid that provides an op-
portunity for the classical resolution of enantiomers by
formation of diastereomeric salts. The target compound
(15) and the synthetic route are shown in Scheme 1. It
was discovered that thioketone 15 could not be resolved
into enantiomers by chromatographic procedures. Con-
sequently, carboxylic acid 12 was partially resolved by
forming its diastereomeric salts with quinine and the
preferential precipitation of one compound. The CD
spectrum of partially resolved 12 is shown in Figure 1.
Optically active 12 was converted to optically active 15
by following the route shown in Scheme 1.
2
Unfortunately, thioketone 8 is prone to O and light-
initiated oxidation to the naphthalene compound, 9,
which made its isolation in pure form difficult and
reduces its value as a potential chiroptical trigger (eq 3).
Hence, we turned our attention to indene-containing
potential trigger compounds which cannot oxidize in this
way.
The reaction of tetraethylmethylenediphosphonate
(TEMPD) with 2-indanone and NaH gives diethylphos-
phonate ester (10) in 76% yield (eq 4). The Horner-
Wadsworth-Emmons reaction of 10 with the protected
diketone (7) followed by deprotection and thionation with
Lawesson’s reagent gives racemic thioketone 11 in 40%
overall yield (eq 5). Optically active thioketones are
3
. Op tica l Ch a r a cter iza tion a n d P h otoch em istr y
of P oten tia l Tr igger s 11 a n d 15. One objective for the
preparation of thioketone-containing triggers is to move
the action spectrum of a chiroptical switch to the visible
region. Figure 2 shows the region of the spectrum of
thioketone 11 from 350 to 600 nm. As is typical of
aliphatic thioketones, it exhibits a weak (nπ*) band with
λ
max at ca. 480 nm. The absorption spectrum of 15 is
similar.
A successful chiral trigger will be partially photore-
solved by irradiation with CPL. The irradiation of race-
mic thioketones 11 and 15 with CPL (λ > 400 nm) in
deoxygenated benzene solution was monitored by absorp-
tion and CD spectroscopy. These thioketones were con-
sumed very slowly during the course of the irradiation
(by monitoring its absorption spectrum), but even after
several days of irradiation the CD spectrum showed no
required for assessment of their utility as chiroptical
triggers. We attempted to resolve thioketone 11 by chiral
chromatographic methods; however, all attempts failed.
(
28) Bredt, J . J ustus Liebigs Ann. Chem. 1924, 437, 1-13.
(29) Steliou, K.; Mrani, M. J . Am. Chem. Soc. 1982, 104, 3104.
J . Org. Chem, Vol. 68, No. 3, 2003 1077

Bradford and Schuster
F IGURE 1. CD spectrum of partially resolved 12 (4.4 mM in
CH Cl solution in a 1.0 cm path length cell).
2
2
F IGURE 3. A photoracemization of partially resolved 15 (1.4
mM deuterated benzene solution in a 1.0 cm path length
cell): (-) before irradiation and (--) after irradiation with
unpolarized light for 24 h.
will cause the trigger to be useless.31 We find that the
Φ
rac of the thioketone-containing triggers is approxi-
mately an order of magnitude less than their ketone-
containing counterparts, which explains why no photo-
resolution was detected when 11 or 15 was irradiated
with CPL.
It is worthwhile to speculate briefly about the reasons
that underlie the inefficient triplet-triplet energy trans-
fer in the thioketone triggers. We demonstrated that
2
-adamantanethione is an effective sensitizer of diene 2,
which shows that the energy criterion for energy transfer
is met. However, this model reaction proceeds by an
exchange mechanism, which requires a collision between
F IGURE 2. Absorption spectrum of 11 (30 mM in deuterated
benzene solution in a 1.0 cm path length cell).
3
2
the triplet sensitizer and the diene acceptor. In fact,
exchange energy transfer cannot lead to photoresolution
because its efficiency is independent of the handedness
of the CPL absorbed. Triggers 11 and 15 were designed
to avoid this problem by providing a rigid, through-bond
framework for intramolecular energy transfer. Evidently,
the through-bond transfer mechanism is less effective for
thioketones than for ketones. This may be a result of the
fact that sulfur is a second-row element and that the
orbital overlap of its nπ* state with carbon-carbon bond
orbitals of the bridging bicyclic group, which is required
for the through bond process, is less.
sign of partial photoresolution. The failure to detect
photoresolution could be a consequence of either an
inadequate g value or inefficient intramolecular triplet
λ
energy transfer from the thioketone to the isomerizing
group.
We determined the quantum efficiency for the photo-
racemization of optically active thioketone 15. Irradiation
of a deoxygenated benzene solution prepared from par-
tially resolved 15 in the visible spectral region results in
the slow loss of optical activity that is revealed by
analysis of the CD spectrum in Figure 3. The quantum
yield for photoracemization (Φrac) of 15 at 450 nm was
determined to be 0.04 by using a Reinecke’s salt acti-
nometer.30 This is about 10% of the value measured for
Exp er im en ta l Section
Gen er a l. 1H and 13C NMR spectra were recorded at 300
and 75 MHz, respectively, in CDCl unless otherwise specified.
3
intramolecular through-bond triplet-triplet energy trans-
fer when the sensitizer is a ketone.9
Chemical shifts are reported in parts per million (δ) downfield
from TMS. The following compounds were synthesized accord-
ing to procedures reported previously: 2-adamantaneth-
Discu ssion
ione,3
2
3,34
1-phenyl-4-methyl-1,3-butadiene (2 and 3), methyl
-[4-(3-oxo-bicyclo[3.2.1]oct-8-ylidenemethyl)phenyl]acetate, 5,
35
The goals of this work were to prepare and characterize
a chiroptical trigger that absorbs in the visible spectral
region. Thioketone-containing compounds were prepared
that could have fulfilled this objective. These compounds
are chiral and are theoretically capable of partial pho-
toresolution and photoracemization. However, the ef-
ficiency of photoresolution is exponentially dependent on
the Φrac, which means that a low value for this parameter
(30) Shizuka, H.; deMayo, P. Creation and Detection of the Excited
State: Measurement of Reaction Quantum Yields; Marcel Dekker: New
York, 1976; Vol. 4.
(31) Stevenson, L. L.; Verdieck, J . F. Mol. Photochem. 1969, 1, 271.
(
(
(
32) Dexter, D. L. J . Chem. Phys. 1953, 21, 836.
33) Griedanaus, J . W. Can. J . Chem. 1970, 48, 3593.
34) Griedanaus, J . W. Can. J . Chem. 1970, 48, 3530.
(35) Mesenheimer J ustus Liebigs Ann. Chem. 1926, 446, 84.
1
078 J . Org. Chem., Vol. 68, No. 3, 2003

Thioketones as Triggers for Liquid Crystal Optical Switches
1
1
bicyclo[3.2.1]octane-3,8-dione 8-ethylene ketal, 7, and 2-(bro-
momethyl)-3,4-dihydronaphthalene.
stirred for 15 min and then a THF solution of 2-indanone (2
mL, 0.2 g, 1.5 mmol) was added dropwise. The mixture was
stirred overnight at room temperature and then water (ca. 3
mL) was added and the product was extracted into ether. The
combined organic layers were washed with brine and then
3
6
Gen er a l Meth od for Dep r otection of Bicyclic Keta ls.
An excess of formic acid was added to the ketal and the
mixture was heated at reflux for 4 h. The reaction mixture
was cooled to room temperature and poured over ice water.
The solution was neutralized with 2 M NaOH and extracted
into ether. The combined organic layers were washed with
4
dried with MgSO . The crude product was purified by column
chromatography (silica gel 1:3 EtOAc/hexane, followed by
MeOH) to give a yellow oil in 68% yield. This material was
1
brine and dried with MgSO
4
to give the ketone.
used without further purification. H NMR δ 1.2-1.4 (t, 6H),
3
.0-3.1 (d, 2H), 3.5 (s, 2H), 4.0-4.2 (m, 4H), 6.7-8.8 (d, 1H),
Gen er a l Meth od for th e Con ver sion of Bicyclic Ke-
ton es to Th iok eton es. Lawesson’s reagent (0.5 mmol) was
added slowly to a room temperature solution of the bicyclic
ketone in toluene (3 mL, 0.8 mmol). For preparation of racemic
thioketones, the mixture was heated at reflux until the
reaction was complete as determined by thin-layer chroma-
tography. If optically active thioketone was sought, the mixture
was kept at 70 °C for 6-12 h until the starting ketone had
been consumed. The crude product mixture was separated by
chromatography on a silica gel column eluted with a mixture
of EtOAc and hexane (1:3). The isolated thioketone was
purified by sublimation, which yields bright orange crystals.
13
7.1-7.5 (m, 4H); C NMR δ 16.6, 28.8, 30.6, 42.3, 62.2, 120.6,
123.7, 124.5, 126.5, 130.9, 138.8, 143.7, 145.0; P (121 MHz,
CDCl ) 26.3; GC/MS m/z 266 (M ), 238, 210, 156, 128.
3
3
1
+
P r ep a r a tion of Th iok eton e 11. A THF solution of phos-
phonate ester 10 (2 mL, 0.35 g, 1.3 mmol) was added dropwise
to a THF slurry of NaH (2.5 mL, 0.04 g, 1.7 mmol) and the
mixture was stirred at room temperature for 15 min. A THF
solution of ketal 7 (2 mL, 0.2 g, 1.1 mmol) was slowly added
to this mixture and then it was heated at reflux for 8 h. Water
(ca. 2 mL) was added to the reaction mixture and the product
was extracted into ether. The combined organic layers were
The reaction of ketone 5 under these conditions gives sulfide
washed with brine and dried with MgSO . The crude material
4
1
6
2
3
, which was not purified further. H NMR δ 1.2 (s, 6H), 1.8-
was purified by column chromatography (silica gel 1:6 EtOAc/
hexane) to give the protected ketone as a white solid in 84%
.0 (m, 8H), 2.2-2.4 (m, 2H), 2.5-2.9 (m, 4H), 3.1-3.2 (m, 1H),
1
.3 (m, 1H), 3.5 (d, 2H), 3.7 (d, 2H), 6.1-6.3 (d, 4H), 7.1-7.3
yield. Mp 136 °C; H NMR δ 1.2-2.2 (m, 7H), 2.5-2.9 (m, 3H),
+
(
m, 8H); GC/MS m/z 566 (M ), 537, 507, 403, 337, 267, 239,
2.8-2.9 (m, 2H), 3.4-3.7 (m, 2H), 4.0 (s, 4H), 6.25 (s, 1H), 6.65
(s, 1H), 7.1-7.4 (m, 4H); 13C NMR δ 25.30, 25.56, 34.33, 39.68,
40.01, 41.54, 41.88, 63.88, 64.93, 116.80, 120.39, 123.47,
124.17, 126.43, 130.14, 137.83, 143.04, 144.98, 145.31; GC/MS
2
07, 179, 155, 117, 91.
Dim eth yl 3,4-Dih yd r on a p h th a len -2-ylm eth ylp h osp h o-
n a te. Trimethyl phosphite (4.5 g, 36 mmol) was added to
-bromomethyl-1,2-dihydronaphthalene (1 g, 4.5 mmol) and
+
3
294 (M ), 266, 249, 221, 205, 179, 151, 128, 99. The ketal
stirred overnight at 50 °C. The crude product was purified by
column chromatography (silica gel EtOAc/hexane 1:3, followed
by MeOH) to yield 0.74 g (65%) of phosphonate ester. H NMR
protecting group was removed by means of the standard
procedure to give the ketone in quantitative yield. Mp 103-
1
1
104 °C; H NMR δ 1.5-2.0 (m, 4H), 2.5-2.7 (s, 2H), 2.9-3.0
δ 2.5-2.6 (m, 2H), 2.9-3.0 (m, 4H), 4.8 (s, 3H), 4.9 (s, 3H), 6.5
(m, 1H), 3.1 (m, 2H), 3.2-3.4 (m, 1H), 3.4-3.7 (m, 2H), 6.3 (s,
+
13
(
d, 1H), 7.1-7.4 (m, 4H); GC/MS m/z 252 (M ), 141, 124.
1H), 6.7 (s,1H), 7.0-7.4 (m, 4H); C NMR δ 22.99, 23.26, 39.78,
3
-(3,4-Dih ydr on aph th alen -2-ylm eth ylen e)bicyclo[3.2.1]-
41.41, 44.13, 44.94, 47.96, 120.81, 123.50, 124.76, 126.10,
126.65,131.79, 134.43, 143.07, 144.31, 144.57, 221.60; GC/MS
octa n -8-on e. A THF solution of dimethyl 3,4-dihydronaph-
thalen-2-ylmethylphosphonate (ca. 1 mL, 0.28 g, 1.1 mmol) was
added slowly to a slurry of NaH in THF (2 mL, 0.052 g, 2.2
mmol) at room temperature. The mixture was stirred for 15
min and then a THF solution (3 mL, 0.1 g, 0.55 mmol) of
bicyclo[3.2.1]octane-3,8-dione 8-ethylene ketal was added drop-
wise. When addition was completed, the reaction mixture was
heated at reflux for 8 h. Water (ca. 1 mL) was added slowly to
the reaction mixture and then the product was extracted into
ether. The combined organic layers were washed with brine
+
m/z 250 (M ), 207, 167, 152, 130, 115, 91. The ketone was
converted to the thioketone in 40% yield using the standard
1
procedure. H NMR δ 1.4-2.0 (m, 4H), 2.5-2.7 (m, 2H), 2.8-
3.0 (m, 1H), 3.1 (s, 2H), 3.2-3.3 (m, 1H), 3.4-3.7 (m, 2H), 6.4
+
(1, 1H), 6.7 (s, 1H), 7.1-7.4 (m, 4H); GC/MS m/z 266 (M ),
233, 191, 167, 152, 137, 115.
3
H-Spir o[bicyclo[3.2.1]octan e-8,2′-[1,3]dioxolan ]-3-ylide-
n ea cetic Acid (12). Solid NaH (0.18 g, 7.8 mmol) was added
slowly at room temperature to a benzene solution of trimeth-
ylphosphonoacetate (10 mL, 1.35 g, 7.4 mmol). The mixture
was stirred for 10 min and then a toluene solution of ketone
4
and dried with MgSO . The crude material was purified by
column chromatography (silica gel 1:3 EtOAc/hexane) to give
1
a white solid in 15% yield. H NMR δ 1.2-2.8 (m, 14H), 3.9
7
(3 mL, 0.5 g, 2.8 mmol) was added over a 15-min period.
(
3
m, 4H), 5.9 (s, 1H), 6.3 (s, 1H), 6.9-7.2 (m, 4H); GC/MS m/z
The mixture was heated at reflux overnight then water (ca. 4
mL) was added and the solution was neutralized with 1 M
HCl. The organic layer was extracted into methylene chloride,
washed with brine, and then dried with MgSO . The crude
4
product was purified by column chromatography (silica gel 1:3
+
08 (M ), 280, 235, 193, 181, 142, 129, 99. The ketal protecting
group was removed by means of the standard procedure to
1
give the ketone in 52% yield. H NMR δ 1.4-2.0 (m, 4H), 2.2-
2
.6 (m, 6H), 2.7-2.9 (m, 3H), 3.0-3.2 (m, 1H), 6.1 (s, 1H), 6.3
+
(s, 1H), 7.0-7.2 (m, 4H); GC/MS m/z 264 (M ), 221, 181, 165,
EtOAc/hexane) to give methyl 3H-spiro[bicyclo[3.2.1]octane-
1
28, 91.
8
,2′-[1,3]dioxolan]-3-ylideneacetate as a white solid in 83%
1
P r ep a r a tion of Th iok eton e 8. The thioketone was pre-
yield. Mp 135-136 °C; H NMR δ 1.3-1.5 (m, 2H), 1.8 (m 2H),
2.0 (m, 2H), 2.1-2.2 (m, 1H), 2.4-2.6 (m, 1H), 2.8-3.0 (m, 1H),
3.6 (m, 1H), 3.7 (s, 3H), 4.0 (s, 4H), 5.8 (s, 1H); GC/MS m/z
pared by means of the standard procedure and isolated in 33%
1
yield. H NMR δ 1.2-3.2 (m, 14H), 6.1 (s, 1H), 6.3 (s, 1H),
+
+
7
9
.0-7.3 (m, 4H); GC/MS m/z 280 (M ), 247, 181, 168, 150, 115,
1. Upon exposure to air and/or light, 8 oxidizes to give 3-(2-
238 (M ), 223, 206, 179, 164, 134, 119, 99, 79. This ester was
hydrolyzed by reaction in 3 M NaOH for 12 h at room
temperature. The reaction mixture was acidified with 6 M HCl
and the organic layer was extracted into methylene chloride,
4
naphthylmethylene)bicyclo[3.2.1]octane-8-thione (9), which
was identified by NMR and mass spectroscopy and not
1
characterized further. H NMR δ 1.4-2.8 (m, 10H), 6.3 (s, 1H),
washed with brine, and then dried with MgSO to give
+
1
7
.0-7.4 (m, 7H); GC/MS m/z 278 (M ), 245, 128.
carboxylic acid 12 in racemic form in 75% yield. H NMR δ
Dieth yl 1H-In d en e-2-ylm eth ylp h osp h on a te (10). A THF
1.3-1.5 (m, 2H), 1.8-1.9 (m, 2H), 2.0 (m, 2H), 2.1-2.2 (d, 1H),
2.4-2.6 (d, 1H), 2.8-3.0 (d, 1H), 3.6 (m, 1H), 4.0 (s, 4H), 5.8
solution of tetraethylmethylenediphosphonate (10 mL, 0.52 g,
.8 mmol) was added slowly at room temperature to a THF
slurry of NaH (ca. 2 mL, 0.04 g, 1.6 mmol). The mixture was
(s, 1H); 13C (75 MHz, CDCl
) δ 25.66, 25.82, 34.37, 40.01, 40.44,
1
3
42.31, 64.48, 65.43, 116.59, 117.42, 162.50, 171.72; GC/MS m/z
+
2
24 (M ), 206,179, 150, 134, 105, 99, 79.
Bicyclic Keton e 14. Solid (2-hydroxybenzyl)phosphonium
(36) Miyashi, T.; Nishizawa, Y.; Fujii, Y.; Yamakawa, K.; Kamata,
M. J . Am. Chem. Soc. 1986, 108, 1617-1632.
bromide (1.24 g, 2.8 mmol) and dicyclohexylcarbodiimide (0.5
J . Org. Chem, Vol. 68, No. 3, 2003 1079

Bradford and Schuster
g, 2.5 mmol) were added to a CH
mL, 0.44 g, 2 mmol). Solid (dimethylamino)pyridine (0.04 g,
.3 mmol) was added all at once to the reaction solution and
the mixture was stirred at room temperature until acid 12 was
consumed (5-10 h) as indicated by thin-layer chromatography.
The solvent was removed from the reaction mixture and
toluene (5 mL) was added. Triethylamine (1.53 mL) was added
to the toluene solution, which was then heated at reflux until
2
Cl
2
solution of acid 12 (10
(20 mL, 7.8 g, 24.0 mmol). The mixture was stirred at room
temperature for ca. 1 h. Most of the methanol was evaporated
and 10 mL of acetone was added to the concentrated solution.
The resulting solution was heated at reflux for 2 h and cooled
overnight, and the white precipitate that formed was isolated
by filtration. The partially resolved acid 12 was liberated from
its quinine salt by addition of HCl (2 M, final pH ca. 2) and
0
2 2 4
extracted into CH Cl . The solution was dried with MgSO and
1
3 was consumed (3-5 h) as indicated by TLC. The cooled
the resolved acid (ca. 2.5 g) was isolated by evaporation of the
solvent. Resolution was confirmed by recording the CD spec-
trum of 12. The crude, partially resolved acid was recrystal-
lized from a methanol/acetone solution. The resolution proce-
dure was repeated until an enantiomeric excess sufficient for
conversion to optically active 15 was obtained.
reaction mixture was filtered and purified by column chroma-
tography (silica gel 1:3 EtOAc/hexane) to give the coupled ketal
1
in 67% yield as a white solid. Mp 106 °C; H NMR δ 1.2-1.6
(
2
m, 2H), 1.6-1.8 (m, 2H), 1.9-2.0 (m, 2H), 2.1-2.2 (m, 1H),
.5-2.7 (m, 1H), 2.8-3.0 (m, 1H), 3.0-3.2 (m, 1H), 3.8-4.0
(
2
1
m, 4H), 6.2 (s, 1H), 6.4 (s, 1H), 7.0-7.2 (m, 2H), 7.3-7.5 (m,
P h otor a cem iza tion of Keton e 14 a n d Th iok eton e 15.
Deoxygenated benzene (d , to permit analysis by NMR spec-
6
+
H); GC/MS m/z 296 (M ), 268, 251, 223, 207, 181, 165, 152,
31, 99, 77. The ketal was added to an excess of 2 M HCl in
troscopy) solutions of ketone 14 or thioketone 15 (absorbance
at 300 nm ) 2) were irradiated with UV (in the case of 14, λ
> 300 nm) or visible (in the case of 15, λ > 400 nm) light from
an Oriel 1000 W Hg-Xe arc lamp. The racemization reaction
was monitored by CD spectroscopy and the chemical composi-
tion of the solutions was confirmed by GC/MS and NMR
spectroscopy. Ketone 14 was fully racemized after 36 h of
irradiation and thioketone was racemized after 24 h. There
was no significant decomposition of either the ketone or the
thioketone. The quantum yield for racemization of thioketone
15 was determined by comparison with a Reinecke’s salt
actinometer.30
dioxane, and the mixture was stirred at 50 °C until starting
material was consumed as indicated by TLC to give ketone
1
2
1
1
4 in 87% yield. Mp 73-74 °C; H NMR δ 1.6-2.0 (m, 4H),
.4 (s, 2H), 2.6 (m, 1H), 2.8 (m, 1H), 3.0 (m, 1H), 3.6-3.8 (m,
H), 6.4 (s, 1H), 6.6 (s, 1H), 7.2-7.3 (m, 2H), 7.4 (d, 1H), 7.5
+
(d, 1H); GC/MS 252 (M ), 224, 209, 195, 183, 169, 157, 144,
1
16 2
31, 115. Anal. Calcd for C17H O : C, 80.91; H, 6.39. Found:
C, 80.64, 6.40. A similar procedure was followed with partially
resolved 12 except that the esterification to form 13 and the
Wittig coupling reaction were carried out at 50 °C.
P r ep a r a tion of Th iok eton e 15. The thioketone was
prepared by means of the standard procedure and isolated in
1
7
3
1
2
1
5
2% yield. H NMR δ 1.5-2.0 (m, 4H), 2.6-2.8 (m, 1H), 2.9-
Ack n ow led gm en t. This work was supported by a
grant from the National Science Foundation, for which
we are grateful.
.1 (m 2H), 3.1 (s, 2H), 3.5-3.7 (m, 1H), 6.4 (s, 1H), 6.6 (s,
H), 7.1-7.4 (m, 2H), 7.4-7.6 (m, 2H); GC/MS m/z 268 (M ),
36, 221, 195, 170, 131, 91. When preparing optically active
5 from partially resolved 14 the reaction was carried out at
0 °C. Formation of partially resolved 15 was confirmed by
+
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
1
0-12, 14, and 15. This material is available free of charge
CD spectroscopy.
Resolu tion of 12. A solid portion of racemic acid 12 (5.4 g,
via the Internet at http://pubs.acs.org.
2
4.0 mmol) was added to a solution of quinine in methanol
J O0264675
1
080 J . Org. Chem., Vol. 68, No. 3, 2003