An efficient photo-SET-induced cleavage of dithiane-carbonyl adducts and its relevance to the development of photoremovable protecting groups for ketones and aldehydes

Article abstract of DOI:10.1021/jo981697y

Irradiation of dithiane-aldehyde/ketone adducts in the presence of benzophenone leads to C-C bond cleavage regenerating the carbonyl compounds. It is established that the mechanism of this reaction involves photochemically induced single electron transfer from the dithiane moiety to the excited molecule of ET-photosensitizer, accompanied by mesolytic C-C cleavage in the generated cation-radical, which is assisted by the anion- radical of benzophenone. This mechanism is confirmed by a Hammett plot study of the cleavage in the dithiane adducts of substituted aromatic aldehydes and a deuterium kinetic isotope effect study. Ab initio computations at UHF/6- 31G* and MP2/6-31G* levels of theory in conjunction with self-consistent reaction field (self-consistent isodensity-polarized continuum model), to account for the solvent effect, also support the experimental findings. The reaction is most efficient for protection of aromatic aldehydes and ketones and aliphatic ketones, and is a novel method for protecting carbonyl functionalities with a photoremovable group.

Full text of DOI:10.1021/jo981697y

9924
J . Org. Chem. 1998, 63, 9924-9931
An Efficien t P h oto-SET-In d u ced Clea va ge of Dith ia n e-Ca r bon yl
Ad d u cts a n d Its Releva n ce to th e Develop m en t of P h otor em ova ble
P r otectin g Gr ou p s for Keton es a n d Ald eh yd es
William A. McHale and Andrei G. Kutateladze*
Department of Chemistry and Biochemistry, University of Denver, Denver, Colorado 80208
Received August 20, 1998
Irradiation of dithiane-aldehyde/ketone adducts in the presence of benzophenone leads to C-C
bond cleavage regenerating the carbonyl compounds. It is established that the mechanism of this
reaction involves photochemically induced single electron transfer from the dithiane moiety to the
excited molecule of ET-photosensitizer, accompanied by mesolytic C-C cleavage in the generated
cation-radical, which is assisted by the anion-radical of benzophenone. This mechanism is confirmed
by a Hammett plot study of the cleavage in the dithiane adducts of substituted aromatic aldehydes
and a deuterium kinetic isotope effect study. Ab initio computations at UHF/6-31G* and MP2/6-
31G* levels of theory in conjunction with self-consistent reaction field (self-consistent isodensity-
polarized continuum model), to account for the solvent effect, also support the experimental findings.
The reaction is most efficient for protection of aromatic aldehydes and ketones and aliphatic ketones,
and is a novel method for protecting carbonyl functionalities with a photoremovable group.
In tr od u ction
One significant problem with this method is that carbo-
nyls are often protected to avoid an undesired reaction
with organometallic compounds or hydrides, and the
presence of the nitro group is, in our view, a limiting
factor in such synthetic sequences.
Another example of a true photochemical deprotection
of carbonyls is the photoreverse of the Scho¨nberg-
Mustafa reaction, although it can only be used for
photodeprotection of masked o-quinones.3
To some extent the thioacetal protection of carbonyls,
which is normally removed by Hg2+-assisted hydrolysis,
can also qualify as a photochemically removable protec-
tion. It appears from analysis of the literature, however,
that thioacetal-based methods for photochemical depro-
tection of carbonyls would fall under the category pho-
tooxidative hydrolysis of thioacetals in the presence of
oxygen.4 Recent efforts in our laboratories have been
focused on developing a somewhat orthogonal technique
that utilizes not a 1,3-propanedithiol but rather a 1,3-
dithiane ring “as a whole”.
It was shown by Corey and Seebach5 that 2-lithio-1,3-
dithiane adds to carbonyl compounds to furnish 2-(1-
hydroxyalkyl)-substituted dithianes in excellent yields
(the reported yields were nearly quantitative for many
of the cases studied), Scheme 1. Over the years this
Recent developments in the chemistry of photoremov-
able protecting groups show the immense potential of
light as a reagent. Most commonly it is nucleophilic
functional groups, e.g., alcohol, thiol, or amino groups,
which are protected with 2-nitrobenzyl, phenacyl, 2-ben-
zoylbenzoic acid, 3,5-dimethoxybenzoin, substituted ben-
zyloxycarbonyl, R-keto carbamates, etc.1 We note, how-
ever, that a true photoremovable carbonyl protection is
lacking from the representative list of functionalities for
which photolabile protecting groups are found. A thor-
ough search of the literature revealed several attempts
to develop photoremovable protecting groups for alde-
hydes and ketones based on an acetal formation with
o-nitrobenzyl alcohol or o-nitrophenylethylene glycol.2
(1) (a) For a review on 2-nitrobenzyl, see: Morrison, H. A. Chemistry
of the Nitro and Nitroso Groups; J ohn Wiley: New York, 1970; pp 185-
191. (b) Misetic, A.; Boyd, M. K. Tetrahedron Lett. 1998, 39, 1653-
1656. (c) Banerjee, A.; Falvey, D. E. J . Org. Chem. 1997, 62, 6245-
6251. (d) Hasan, A.; Stengele, K.-P.; Giegrich, H.; Cornwell, P.; Isham,
K. R.; Sachleben, R. A.; Pfleiderer, W.; Foote, R. S. Tetrahedron 1997,
53, 4247-4264. (e) Givens, R. S.; J ung, A.; Park, C.-H.; Weber, J .;
Bartlett, W. J . Am. Chem. Soc. 1997, 119, 8369-8370. (f) Gee, K. R.;
Kueper, L. W., III; Barnes, J .; Dudley, G.; Givens, R. S. J . Org. Chem.
1996, 61, 1228-1233. (g) J ones, P. B.; Pollastri, M. P.; Porter, N. A. J .
Org. Chem. 1996, 61, 9455-9461. (h) Cameron, J . F.; Willson, C. G.;
Fre´chet, J . M. J . J . Am. Chem. Soc. 1996, 118, 12925-12937. (i)
Schilling, M. L.; Katz, H. E.; Houlihan, F. M.; Kometani, J . M.; Stein,
S. M.; Nalamasu, O. Macromolecules 1995, 28, 110-115. (j) Pirrung,
M. C.; Huang, C.-Y. Tetrahedron Lett 1995, 36, 5883-5884. (k) Pirrung,
M. C.; Shuey, S. W. J . Org. Chem. 1994, 59, 3890-3897. (l) Pirrung,
M. C.; Lee, Y. R. J . Org. Chem. 1993, 58, 6961-6963. (m) Binkley, R.
W.; Liu, X. J . Carbohydr. Chem. 1992, 11, 183-188. (n) Liu, X. 1992,
450 pp. (o) Pirrung, M. C.; Nunn, D. S. Bioorg. Med. Chem. Lett. 1992,
2, 1489-1492. (p) Oliveira-Campos, A. M.; Queiroz, M. J . R. P.;
Shannon, P. V. R. Chem. Ind. 1991, 352-353. (q) Binkley, R. W.;
Koholic, D. J . J . Org. Chem. 1989, 54, 3577-3581. (r) Nishida, A.;
Oishi, S.; Yonemitsu, O. Chem. Pharm. Bull. 1989, 37, 2266-2268.
(s) Masnovi, J .; Koholic, D. J .; Berki, R. J .; Binkley, R. W. J . Am. Chem.
Soc. 1987, 109, 2851-2853. (t) Haridasan, V. K.; Pillai, V. N. R. Proc.
Indian Natl. Sci. Acad., Part A 1987, 53, 717-728. (u) Rajasekharan
Pillai, V. N. Org. Photochem. 1987, 9, 225-323. (v) Epling, G. A.;
Walker, M. E. Tetrahedron Lett 1982, 23, 3843-3846. (w) Ohtsuka,
E.; Tanaka, S.; Ikehara, M.; Townsend, L. B.; Tipson, R. S. Nucleic
Acid Chem. 1978, 1, 410-414. (x) Kalbag, S. M.; Roeske, R. W. J . Am.
Chem. Soc. 1975, 97, 440-441.
(2) (a) Aurell, M. J .; Boix, C.; Ceita, M. L.; Llopis, C.; Tortajada, A.;
Mestres, R. J . Chem. Res., Synop. 1995, 452-453. (b) Gravel, D.;
Murray, S.; Ladouceur, G. J . Chem. Soc., Chem. Commun. 1985, 1828-
1829. (c) Gravel, D.; Hebert, J .; Thoraval, D. Can. J . Chem. 1983, 61,
400-410.
(3) Wilson, R. M.; Venkitachalam, S.; Layton, M.; Heckmann, J .;
Barra, M.; Scaiano, J . C. In 10th Winter Conference of Inter-American
Photochemical Society; IAPS: Clearwater Beach, FL, 1998; p C7.
(4) (a) Schmittel, M.; Levis, M. Synlett 1996, 315-316. (b) Tane-
mura, K.; Dohya, H.; Imamura, M.; Suzuki, T.; Horaguchi, T. Chem.
Lett. 1994, 965-968. (c) Kamata, M.; Murakami, Y.; Tamagawa, Y.;
Kato, M.; Hasegawa, E. Tetrahedron 1994, 50, 12821-12828. (d)
Epling, G. A.; Wang, Q. Tetrahedron Lett. 1992, 33, 5909-5912. (e)
Epling, G. A.; Wang, Q. Synlett 1992, 335-336.
(5) Corey, E. J .; Seebach, D. Angew. Chem., Int. Ed. Engl. 1965, 4,
1075-1077.
(6) For a review, see: Gro¨bel, B.-T.; Seebach, D. Sythesis 1977, 357-
402.
10.1021/jo981697y CCC: $15.00 © 1998 American Chemical Society
Published on Web 12/09/1998

Photo-SET-Induced Cleavage of Dithiane-Carbonyl Adducts
J . Org. Chem., Vol. 63, No. 26, 1998 9925
Ta ble 1. P h otoch em ica l Dep r otection Yield sa
Sch em e 1
reaction has been developed into a potent synthetic
methodology used in many celebrated synthetic se-
quences.6
It is easy to see that the [-C(OH)C(SR)-] fragment
bears a striking resemblance to vicinal diols and amino
alcohols which are capable of cleaving under oxidative
conditions (formation of cation-radicals). It is known that
the removal of an electron from a ground state organic
molecule frequently leads to molecular fragmentation due
to the weakening of specific C-C bonds. The mechanistic
aspects of such cleavage in cation-radicals has been
studied extensively by several research groups, leading
to the development of many novel organic reactions.7 In
contrast, organosulfur systems of this type were not
studied in depth. In fact, we found only two publications
in the literature where the C-C cleavage in â-phenylthio-
alkanols was utilized as a synthetic method of indirect
cleavage of olefins8 and also in carbohydrate synthesis.9
This lack of experimental data may be attributed to the
propensity of organosulfur systems to cleave the C-S
bond rather than the C-C bond.
We now report a novel approach to photoreversibly
protecting carbonyls based on SET-initiated C-C frag-
mentation in R-substituted hydroxymethyldithianes.
Resu lts a n d Discu ssion
We found that irradiation of Corey-Seebach dithiane-
aldehyde/ketone adducts in the presence of benzophenone
in acetonitrile leads to efficient C-C bond cleavage,
regenerating the carbonyl compounds nearly quantita-
tively.
a
Pyrex-filtered irradiation of a medium-pressure Hanovia lamp
b
was utilized as a UV source. Yields are reported as determined
by calibrated GC; benzophenone is used as a SET sensitizer.
The reaction is most efficient for aromatic/aliphatic
ketones and aromatic aldehydes (Scheme 2). The yields
and irradiation times of the studied adducts are pre-
sented in Table 1.
Sch em e 2
Mech a n ism . Our rationale of the reaction mechanism
includes photochemically induced single electron transfer
from the dithiane moiety to the excited molecule of ET-
photosensitizer accompanied by mesolytic C-C cleavage
in the generated cation-radical. Excited benzophenone is
certainly capable of oxidizing 1,3-dithianes: the one-
electron reduction potential of triplet benzophenone is
-1.68 V (vs SCE in acetonitrile),10 whereas various
2-substituted dithianes oxidize in the range of +0.73 to
+1.18 V in the same solvent.11
To predict the “polarity” of the mesolytic fragmentation
in the generated cation radical, we conducted ab initio
computations for the benzaldehyde adduct (2) (see Table
2). The mp2/6-31g* energy difference obtained with fully
optimized geometries (the same level of theory) showed
a 14.5 kcal/mol preference for the 1,3-dithian-2-yl radical
and protonated benzaldehyde (entries 1 and 2) over the
{1,3-dithian-2-yl cation}-{phenyl-hydroxymethyl radi-
(7) (a) Bockman, T. M.; Hubig, S. M.; Kochi, J . K. J . Am. Chem.
Soc. 1998, 120, 6542-6547. (b) Sankararaman, S.; Perrier, S.; Kochi,
J . K. J . Am. Chem. Soc. 1989, 111, 6448. (c) Han, D. S.; Shine, H. J .
J . Org. Chem. 1996, 61, 3977. (d) Penn, J . H.; Duncan, J . H. J . Org.
Chem. 1993, 58, 2003-2008. (e) Penn, J . H.; Lin, Z. J . Org. Chem.
1990, 55, 1554-1559. (f) Penn, J . H.; Deng, D. L.; Aleshire, S. K. J .
Org. Chem. 1988, 53, 3572-3582. (g) Popielarz, R.; Arnold, D. R. J .
Am. Chem. Soc. 1990, 112, 3068-3082. (h) Arnold, D. R.; Lamont, L.
J . Can. J . Chem. 1989, 67, 2119-2127. (i) Okamoto, A.; Snow, M. S.;
Arnold, D. R. Tetrahedron 1986, 42, 6175-6187. (j) Gaillard, E. R.;
Whitten, D. G. Acc. Chem. Res. 1996, 29, 292-297. (k) Gan, H.;
Leinhos, U.; Gould, I. R.; Whitten, D. G. J . Phys. Chem. 1995, 99,
3566-3573. (l) Kellett, M. A.; Whitten, D. G. Res. Chem. Intermed.
1995, 21, 587-611. (m) Leon, J . W.; Whitten, D. G. J . Am. Chem. Soc.
1995, 117, 2226-2235. (n) Whitten, D. G.; Kellett, M. A.; Leon, J .; Gan,
H.; Tian, Z. W. Photochem. Photoelectrochem. Convers. Storage Sol.
Energy, Proc. Int. Conf., 9th 1993, 257-275. (o) Kellett, M. A.; Whitten,
D. G. Mol. Cryst. Liq. Cryst. 1991, 194, 275-280. (p) Ci, X.; Whitten,
D. G. J . Am. Chem. Soc. 1989, 111, 3459-3461. (q) Cl, X.; Whitten, D.
G.; Fox, M. A.; Chanon, M. Photoinduced Electron Transfer 1988, Part
C, 553-578. (r) Lee, L. Y. C.; Ci, X.; Giannotti, C.; Whitten, D. G. J .
Am. Chem. Soc. 1986, 108, 175-177. (s) Maslak, P.; Chapman, W. H.,
J r. J . Org. Chem. 1996, 61, 2647. (t) Maslak, P.; Chapman, W. H., J r.;
Vallombroso, T. M.; Watson, B. A. J . Am. Chem. Soc. 1995, 117, 12380-
12389. (u) Maslak, P. Top. Curr. Chem. 1993, 168, 1. (v) Wang, Y.;
Lucia, L. A.; Schanze, K. S. J . Phys. Chem. 1995, 99, 1961-1968. (w)
Burton, R. D.; Bartberger, M. D.; Zhang, Y.; Eyler, J . R.; Schanze, K.
S. J . Am. Chem. Soc. 1996, 118, 5655-5664.
(8) Gravel, D.; Farmer, L.; Ayotte, C. Tetrahedron Lett 1990, 31,
63-66.
(9) Gravel, D.; Farmer, L.; Denis, R. C.; Schultz, E. Tetrahedron Lett.
1994, 35, 8981-8984.
(10) Roth, H. D.; Lamola, A. A. J . Am. Chem. Soc. 1974, 96, 6270-
6275.
(11) Glass, R. S.; Petsom, A.; Wilson, G. S. J . Org. Chem. 1986, 51,
4337-4342.

9926 J . Org. Chem., Vol. 63, No. 26, 1998
McHale and Kutateladze
Ta ble 2. Ab In itio Resu lts
with that of Whitten’s KIE experiments in acetonitrile,
which in their case were inconclusive due to low quantum
efficiency of amino alcohol cleavage.12 (In benzene their
reported KIE values were much highersfrom 1.26 to
4.01).
Plotting the logarithms of relative quantum efficiencies
of the dithiane adduct cleavage versus Hammett’s sub-
stituent constants σ gave us the second experimental
evidence for the benzophenone-assisted mechanism. The
small value of F (F ) -0.2, r2 ) 0.994), although indicative
of some positive charge accumulation in the transition
state, is in our view not nearly enough to explain the
development of a full positive charge as in protonated
benzaldehyde. For one thing, basicities of para-substi-
tuted benzaldehydes13 do not correlate with σ, they
correlate with σ+, with F+ being about -1.9 (we obtained
cal} pair (entries 3 and 4). We estimated a possible
impact of the solvent (acetonitrile, ꢀ ) 36.64) at mp2/6-
31g* level employing the self-consistent isodensity polar-
ized continuum model. This produced additional stabi-
lization for the corresponding cations, but the total energy
difference stayed approximately the same, about 15 kcal/
mol favoring 1,3-dithiane-2-yl radical and protonated
benzaldehyde.
At the same time UHF/6-31g* computations showed
that the energy of the predissociated cation-radical,
-1294.312 280 92 hartree, is below the sum of the
energies of the dithiane radical (-950.530 933 7 hartree)
and protonated benzaldehyde (-343.766 345 3 hartree),
a total of -1294.297 278 964 hartree, by about 10 kcal/
mol. This would argue that an unassisted C-C cleavage
in the cation-radical is an uphill process. Thus the actual
mechanism for this fragmentation would likely involve
a benzophenone anion-radical accelerating the cleavage
via deprotonation of the hydroxy group (Scheme 3).
+
this value by plotting pKBH values from ref 13 against
the latest σ+ values found in ref 14). Depending on
whether the transition state of the unassisted cleavage
is late or early, the actual F value for such cleavage may
vary. Our ab initio (UHF) results seem to point to the
late transition state (an uphill process) in unassisted
fragmentation. Although one can expect the absolute
value of F to be less than 1.9, it is unlikely that it could
be as insignificant as 0.2.
To summarize the mechanistic part, we believe that
there is substantial evidence pointing to the involvement
of a benzophenone anion-radical in the hydrogen abstrac-
tion during the cleavage of dithiane adducts. Admittedly,
in acetonitrile this effect is not overly noticeable. Also,
the mere fact that vicinal bis(dialkylamines) do cleave
under similar conditions should indicate that slow unas-
sisted cleavage is also a possibility, especially in polar
solvents.7j
Sch em e 3
P r otection of Alip h a tic Ald eh yd es. The yields for
deprotection of aliphatic aldehydes were modest. We
attempted to improve the yield by utilizing lithiated
2-phenyl-1,3-dithiane as a protecting reagent in place of
the unsubstituted dithiane. Scheme 4 shows a deprotec-
tion reaction that gave 65% of deprotected propanal. This
2-fold improvement over the result listed in Table 1
demonstrates that the lack of aryl/alkyl stabilization at
one end of the to-be-cleaved bond can be partially
compensated with enhanced aryl stabilization at the
other end.
Whitten12 showed in a mechanistic study that a similar
C-C cleavage in amino alcohol cation-radicals is indeed
assisted by a benzophenone anion-radical. To prove that
such assistance is in fact present in our case, we have
conducted the following mechanistic study.
We first synthesized dithiane adducts of para-substi-
tuted benzaldehydes and studied the kinetic isotope effect
of their cleavage in 3% H2O(D2O) acetonitrile. Our
findings are summarized in Table 3.
Sch em e 4
Ta ble 3
Generally, 2-lithio-2-phenyl-1,3-dithiane addition to
carbonyls cannot be utilized to protect ketones. The yields
for the protection step may not be practical for more
hindered carbonyls. Note, however, that we suggest using
phenyldithiane as a remedy to boost deprotection yields
for aliphatic aldehydes. It is known that aldehydes do
The relatively small values of the observed isotope
effects may be attributed to fast hydrogen exchange in
aqueous acetonitrile. It should also be noted that the
trend, {MeO/1.09} - {H/1.13} - {CN/1.14}, is in keeping
with the expected change in electronic demand for the
cleavage. Unfortunately we cannot compare our values
(12) Ci, X.; Kellett, M. A.; Whitten, D. G. J . Am. Chem. Soc. 1991,
113, 3893-3904.
(13) Yates, K.; Stewart, R. Can. J . Chem. 1959, 37, 664-671.
(14) Extner, O. In Correlation Analysis in Chemistry; Chapman, N.
B., Shorter, J ., Eds.; Plenum Press: New York, 1978; Chapter 10.

Photo-SET-Induced Cleavage of Dithiane-Carbonyl Adducts
J . Org. Chem., Vol. 63, No. 26, 1998 9927
F igu r e 1. Schematic representation of major NOE interactions in the mono- and bis-dithiane adducts of camphorquinone.
F igu r e 2. AM1 optimized geometry of camphorquinone-dithiane monoadduct cation-radical (11+•, left) and UHF/6-31g* optimized
geometry of benzaldehyde-dithiane adduct cation-radical (2+•, right). The desired (anti- to the dithiane moiety) approach of
benzophenone anion-radical allowing to minimize the back electron transfer is severely hindered in the case of camphorquinone
adduct.
react with the phenyldithiane lithio derivative almost
quantitatively.15
norbornane ring prevents the kind of unhindered ap-
proach available for benzophenone anion-radical in the
case of benzaldehyde adduct 2. This should increase the
probability of back electron transfer (BET) and effectively
kill the fragmentation channel.
Scop e. Camphorquinone reacted with lithiodithiane
to give either the mono- or bisadduct depending on the
molar ratio of the reagents. Both compounds were fully
characterized by NMR, including an exhaustive NOE
study to determine the stereo- and regiochemistry (Figure
1). As expected, the nucleophilic attack was endo leading
to exo alcohols 11 and 12.
Admittedly, the presence of the carbonyl electron-
withdrawing group in the monoadduct of camphorquino-
ne makes the steric hindrance argument weaker. If,
To our disappointment, the yield of photodeprotected
camphorquinone was essentially zero in both cases. Even
after extended irradiation of 11 or 12 in acetonitrile in
the presence of benzophenone we did not detect any
significant amount of camphorquinone. In our view this
fact may further illustrate the necessity of benzophenone
anion-radical assistance. We optimized the geometry of
the cation-radical 11+• at the AM1 level of theory and
contend that in order for the benzophenone anion-radical
to abstract the proton from the hindered OH group of
11+• it should approach/pass in the immediate vicinity
of sulfur atoms bearing the positive charge. As shown in
Figure 2, the syn-methyl group in position 7 of the
(16) Kessar, S. V.; Singh, T.; Vohra, R. Tetrahedron Lett. 1987, 28,
5323-5326. Kessar, S. V.; Singh, T.; Vohra, R. Indian J . Chem. 1991,
30B, 999-1005.
(17) As we tried to optimize the reaction yield, we found the
following. Upon extended irradiation (over 4 h) GC-MS analysis of the
reaction mixture showed a growing peak of a secondary photoproduct
with a molecular ion of 272 and a fragmentation pattern consistent
with the 4-formyltriphenylmethane structure. It is conceivable that
under reaction conditions trityl alcohol 18 photochemically ionizes to
produce the corresponding trityl cation, which is subsequently reduced
to the triphenylmethane. We also cannot rule out a radical mechanism
for such reduction.
(18) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, Revision E.2; Gaussian, Inc.: Pittsburgh, PA, 1995.
(15) Stowell, M. H. B.; Rock, R. S.; Rees, D. C.; Chan, S. I.
Tetrahedron Lett. 1996, 37, 307-310.

9928 J . Org. Chem., Vol. 63, No. 26, 1998
McHale and Kutateladze
found in the literature,16 the photostationary intermo-
lecular equilibrium between the carbonyl compounds we
studied and 1,3-dithiane favors dissociated products. For
example, the GC-MS analysis of the reaction between 1,3-
dithiane and benzophenone after 30 min and 1 h shows
no traces of 2-(diphenylhydroxymethyl)-1,3-dithiane. At
the same time, irradiation of 2-(diphenylhydroxymethyl)-
1,3-dithiane in the presence of an ET-sensitizer gives 97%
of benzophenone back.
Sch em e 5
however, the BET vs fragmentation partition is indeed
controlled by the facility of anion-radical approach, one
also should expect the dithiane-camphor adduct to be
reluctant to cleave under the reaction conditions. In fact
we found this to be exactly the case. Upon extended
irradiation of the adduct 13 in the presence of benzophe-
none, it did not produce any detectable amount of
camphor (Scheme 5). Except for the hindered approach,
we can think of no other reason for this failure to cleave.
Syn th etic Ap p lica tion ssCom p a tibility w ith Gr ig-
n a r d Rea gen ts. One of the most important features of
any carbonyl protecting group should, in our view, be its
compatibility with organometallic reagents and hydrides.
Our dithianyl protection is perfectly compatible with, for
example, Grignard reagents. We synthesized the follow-
ing monoformyl-substituted trityl alcohol from ethyl
4-formylbenzoate with an overall yield of 87% (excluding
the final deprotection step, it is a one-pot reaction),
Scheme 8.17
Sch em e 6
Con clu sion
An efficient photoremovable protecting group based on
utilization of a dithiane ring “as a whole” has been
developed. Although some limitations for this protection
reaction resulting from either the sterically hindered
approach or due to fast triplet energy transfer were
documented in the course of this research, the overall
reaction is a general way to photoreversibly protect
various aldehydes and ketones. The protection is compat-
ible with organometallic reagents, which makes it a
valuable addition to the arsenal of carbonyl protecting
groups. It also makes it the only practical photoremovable
protecting group for aldehydes and ketones.
We also have encountered a case when triplet energy
transfer from excited benzophenone was faster than the
desired single electron transfer. Upon extended irradia-
tion, trans-cinnamaldehyde-dithiane adduct undergoes
trans-cis isomerization with little or no cleavage ob-
served (Scheme 6).
Rever se Rea ction : C-C Bon d F or m a tion . Another
issue that needs to be addressed here is the reverse
reaction (C-C bond formation). It is well-known that
R-hydrogen abstraction in alkyl sulfides by excited car-
bonyls proceeds most commonly via an electron-transfer
mechanism. However, in all the cases we studied, we did
not observe any reverse reaction of the following type
(Scheme 7).
Exp er im en ta l Section
Melting points are uncorrected. Common solvents were
purchased from Aldrich and used as is, except for THF, which
was refluxed over and distilled from potassium benzophenone
ketyl prior to use. n-BuLi (as a 2.0 M solution in n-pentane,
n-hexane, or hexanes), 1,3-dithiane, 2-phenyl-1,3-dithiane,
acetophenone, benzaldehyde and derivatives (p-MeO, CN),
benzophenone, propanal, heptanal, trans-cinnamaldehyde,
cyclohexanone, camphor, camphorquinone, 5R-cholestan-3-one,
and ethyl 4-formylbenzoate were all purchased from Aldrich
and used without additional purification. 1H and 13C NMR
spectra were recorded at 25 °C on a Varian Mercury 400 MHz
instrument in CDCl3 with TMS as an internal standard (unless
otherwise noted). Column chromatography was performed on
silica gel, 70-230 mesh ASTM, using ethyl acetate-hexane
mixtures as eluent. HP 6890 gas chromatograph (flame
ionization detector) was used to determine the yields of the
deprotected ketones and aldehydes with dodecane as an
internal standard. Another HP 6890 with MSD detector was
used to monitor the progress of photoreactions and to identify
components of complex reaction mixtures. A Pyrex-filtered
Sch em e 7
Although some rare (intramolecular) cases of the
“backward” reaction, namely, a photochemically induced
coupling of carbonyl compounds with 1,3-dithianes are
Sch em e 8

Photo-SET-Induced Cleavage of Dithiane-Carbonyl Adducts
J . Org. Chem., Vol. 63, No. 26, 1998 9929
output of a medium-pressure Hanovia lamp was utilized as
the UV source.
1H), 2.03 (s, 1H), 2.0-0.6 (m, H), 0.89 (d, 3H, J ) 6.5 Hz),
0.86 (d, 3H, J ) 6.5 Hz), 0.85 (d, 3H, J ) 6.5 Hz), 0.76 (s, 3H),
0.64 (s, 3H); 13C NMR with C-type assignments based on DEPT
experiment (CDCl3) δ 74.053 (4°), 61.282 (3°), 56.445 (3°),
56.194 (3°), 53.911 (3°), 42.555 (4°), 40.791 (3°), 39.999 (2°),
39.486 (2°), 37.674 (2°), 36.139 (2°), 35.768 (3°), 35.686 (4°),
35.473 (3°), 33.775 (2°), 31.924 (2°), 30.952 (2°), 30.707 (2
carbons, 2°), 28.403 (2°), 28.217 (2°), 27.988 (3°), 25.957 (2°),
24.171 (2°), 23.811 (2°), 22.801 (1°), 22.544 (1°), 20.999 (1°),
Ab initio computations were performed on a dual mips
R10000 processor SGI Octane workstation equipped with 1 G
of memory using the Gaussian 94 Revision E.2 computational
package.18 The input geometries were created and preopti-
mized using a force field geometry optimization as imple-
mented in Chem3D (Cambridgesoft). Full geometry optimiza-
tions were then performed at UHF/6-31G* and/or MP2/6-31G*
levels of theory. Self-consistent reaction field (SCRF) computa-
tions were performed to account for solvent polarity effects
utilizing the self-consistent isodensity polarized continuum
model (SCIPCM). These were run as single-point calculations
without further geometry optimization (MP2-SCIPCM/6-31G*//
MP2/6-31G*).
Gen er a l Meth od of th e Ad d u ct P r ep a r a tion . A generic
method by Corey and Seebach was used to prepare the desired
dithiane-carbonyl adducts.19 A total of 5 mmol of dithiane was
dissolved in 40 mL of freshly distilled THF, and the solution
was cooled to -20 °C under nitrogen. Then 5.6 mmol (2.8 mL)
of n-butyllithium (a 2 M solution in hexanes) was added
dropwise upon stirring. The resulting mixture was stirred for
2-2.5 h. The temperature was then lowered to -78 °C, and 5
mmol of an appropriate carbonyl compound dissolved in 10
mL of THF was added to the vigorously stirred solution of the
dithianyl anion. The reaction mixture was stirred for 2-4 h
at this temperature and then stored in a freezer at -25 °C
overnight.20 The subsequent aqueous workup included quench-
ing the reaction mixture with a 1 M solution of ammonium
chloride, extracting twice with ether, and drying the combined
organic extracts over sodium sulfate. The solvent was then
removed with a rotary evaporator, and the residue was purified
using either column chromatography (silica gel, ethyl acetate-
hexane) or recrystallization.
18.662 (1°), 12.072 (1°), 11.204 (1°). Anal. Calcd for C31H54
OS2: C, 73.45; H, 10.74. Found: C, 73.73; H, 10.59.
-
2-(1-H yd r oxy-1-(p -cya n op h en yl)m et h yl)-1,3-d it h ia n e
(8): 1H NMR (CDCl3) δ 7.57 (d, 2H, J ) 9.4 Hz), 7.48 (d, 2H,
J ) 9.4 Hz), 5.0 (dd, 1H, J ) 8.2, 1.5 Hz), 3.98 (d, 1H, J ) 8.2
Hz), 3.1 (d, 1H, J ) 1.6 Hz), 3.0-2.85 (m, 2H), 2.8-2.65 (m,
2H), 2.15-1.95 (m, 2H); MS (EI) m/z 251 (M+), 119 (100%).
Anal. Calcd for C12H13NOS2: C, 57.33; H, 5.21. Found: C,
57.22; H, 5.40.
2-(1-H y d r o x y -1-(p -m e t h o x y p h e n y l )m e t h y l )-1 ,3 -
d ith ia n e (9): 1H NMR (CDCl3) δ 7.33 (d, 2H, J ) 9.5 Hz),
6.89 (d, 2H, J ) 9.5 Hz), 4.83 (dd, 1H, J ) 7.6, 1.9 Hz), 4.06
(d, 1H, J ) 7.8 Hz), 3.80 (s, 3H), 2.85 (d, 1H, J ) 2.0 Hz),
2.95-2.80 (m, 2H), 2.73-2.62 (m, 2H), 2.10-1.90 (m, 2H); MS
(EI) m/z 256 (M+), 137 (100%), 119. Anal. Calcd for
C
12H16O2S2: C, 56.22; H, 6.29. Found: C, 56.20; H, 6.46.
2-(1-Hydr oxypr opyl)-2-ph en yl-1,3-dith ian e (10): 1H NMR
(CDCl3) δ 7.96 (d, 2H, J ) 7.5 Hz), 7.40 (t, 2H, J ) 7.5 Hz),
7.29 (t, 1H, J ) 7.5 Hz), 3.73 (d, 1H, J ) 10.4 Hz), 2.78-2.63
(m, 4H), 2.17 (s, 1H), 1.97-1.89 (m, 2H), 1.69-1.57 (m, 1H),
1.26-1.13 (m, 1H), 0.90 (t, 3H, J ) 7.3 Hz); 13C NMR (CDCl3)
δ 138.400, 129.725, 128.301, 127.120, 79.960, 65.936, 27.067,
26.898, 24.940, 24.388, 10.991; MS (EI) m/z 254 (M+), 195
(100%). Anal. Calcd for C13H18OS2: C, 61.37; H, 7.13. Found:
C, 61.54; H, 7.29.
2-(Dip h en ylh yd r oxym eth yl)-1,3-d ith ia n e (1):19 1H NMR
(CDCl3) δ 7.58-7.30 (m, 10H), 5.16 (s, 1H), 3.29 (s, 1H), 2.80-
2.90 (m, 4H), 2.04-1.87 (m, 2H); MS (EI) m/z 302 (M+), 119
(100%), 105, 77; mp 135 °C (lit.19 mp 136.0-136.5 °C).
2-(Hyd r oxy-p h en ylm eth yl)-1,3-d ith ia n e (2):21 1H NMR
(CDCl3) δ 7.44-7.30 (m, 5H), 4.91 (dd, 1H, J ) 7.5, 2.2 Hz,),
4.08 (d, 1H, J ) 7.5 Hz), 2.97 (d, 1H, J ) 2.2 Hz), 2.99-2.89
(m, 2H), 2.76-2.68 (m, 2H), 2.11-1.92 (m, 2H); MS (EI) m/z
226 (M+), 119 (100%); mp 74-75 °C (lit.21 mp 73-74 °C).
2-(1-Hyd r oxy-1-p h en yleth yl)-1,3-d ith ia n e (3):22 1H NMR
(CDCl3) δ 7.55-7.26 (m, 5H), 4.44 (s, 1H), 2.88-2.70 (m, 5H),
2.4-2.0 (m, 1H), 1.82-1.76 (m, 1H), 1.74 (s, 3H); MS (EI) m/z
240 (M+), 119, 43 (100%).
en do-3-(1,3-Dith ian -2-yl)-exo-3-h ydr oxy-1,7,7-tr im eth yl-
2-bicyclo[2.2.1]h ep ta n on e (11): 1H NMR (CDCl3) δ 4.37 (s,
1H), 3.11 (s, 1H), 2.95-2.83 (m, 4H), 2.19 (d, 1H, J ) 4.3 Hz),
2.10-2.02 (m, 1H), 2.00-1.85 (m, 1H), 1.85-1.76 (m, 1H),
1.75-1.66 (m, 1H), 1.62-1.53 (m, 1H), 1.05 (s, 3H), 1.01 (s,
3H), 0.92 (s, 3H); 13C NMR with C-type assignments based on
DEPT experiment (CDCl3) δ 215.18 (CdO); 79.401 (4°), 58.743
(4°), 54.117 (3°), 52.373 (3°), 45.456 (4°), 29.880 (2°), 29.698
(2°), 29.181 (2°), 25.163 (2°), 22.432 (2°), 22.220 (1°), 20.491
(1°), 9,753 (1°); MS (EI) m/z 286 (M+), 119 (100%). Anal. Calcd
for C14H22O2S2: C, 58.70; H, 7.74. Found: C, 58.41; H, 7.92.
en d o,en d o-2,3-Bis(1,3-d ith ia n -2-yl)-1,7,7-tr im eth yl-exo,
exo-2,3-bicyclo[2.2.1]h ep ta n ed iol (12): mp 153-154 °C; 1H
NMR (CDCl3) δ 5.83 (s, 1H), 5.52 (s, 1H), 4.23 (s, 1H), 4.07 (s,
1H), 3.10-2.90 (m, 8H), 2.25-2.05 (m, 3H), 2.04 (d, 1H, J )
4.8 Hz), 2.01-1.90 (m, 2H), 1.76-1.64 (m, 1H), 1.58-1.48 (m,
1H), 1.46-1.37 (m, 1H), 1.21 (s, 3H), 1.06 (s, 3H), 0.81 (s, 3H);
13C NMR with C-type assignments based on DEPT experiment
(CDCl3) δ 86.399 (4°), 85.928 (4°), 61.267 (3°), 58,628 (3°),
55.299 (4°), 53.615 (3°), 48.974 (4°), 35.149 (2°), 34.535 (2°),
31.532 (2°), 30.971 (2°), 29.143 (2°), 26.504 (2°), 26.118 (2°),
22.849 (1°), 22.584 (1°), 21.803 (2°), 12.649 (1°). Anal. Calcd
for C18H30O2S4: C, 53.16; H, 7.44. Found: C, 52.88; H, 7.58.
en d o-2-(1,3-Dith ia n -2-yl)-1,7,7-tr im eth yl-exo-2-bicyclo-
[2.2.1]h ep ta n ol (13):25 mp 128-129 °C (lit.25 mp 130-131 °C);
1H NMR (CDCl3) δ 4.26 (s, 1H), 3.02-2.80 (m, 4H), 2.31 (s,
1H), 2.12-2.03 (m, 2H), 1.90-1.62 (m, 5H), 1.52-1.43 (m, 1H),
1.15-1.09 (m, 1H), 1.08 (s, 3H), 1.05 (s, 3H), 0.83 (s, 3H); MS
(EI) m/z 272 (M+), 119 (100%).
2-(1-Hyd r oxycycloh exyl)-1,3-d ith ia n e (4):19 1H NMR
(CDCl3) δ 4.19 (s, 1H), 3.0-2.80 (m, 5H), 2.10-1.89 (m, 2H),
1.85-1.10 (m, 10H); MS (EI) m/z 218 (M+), 120 (100%).
2-(1-Hydr oxypr opyl)-1,3-dith ian e (5):23,24 1H NMR (CDCl3)
δ 3.90-3.78 (m, 2H), 2.81-2.75 (m, 4H), 2.30 (m, 1H), 2.15-
1.85 (m, 4H), 0.89 (t, 3H, J ) 7.6 Hz).
2-(1-Hyd r oxyh ep tyl)-1,3-d ith ia n e (6):24 1H NMR (CDCl3)
δ 3.93-3.85 (m, 2H), 2.98-2.90 (m, 2H), 2.84-2.70 (m, 2H),
2.37 (d, 1H), 2.14-2.04 (m, 1H), 2.04-1.92 (m, 1H), 1.85-1.38
(m, 10H), 0.96 (t, 3H); MS (EI) m/z 234 (M+), 119 (100%).
3-(1,3-Dith ia n -2-yl)-5r-ch olesta n -3-ol (7). A 5:4 mixture
of R-dithianyl to â-dithianyl compounds was obtained. The
photodeprotection reaction was carried out with this original
mixture of isomers. A small quantity of the R-isomer was also
isolated using column chromatography (silica gel, ethyl acetate-
hexane 1:3). 3R-(1,3-Dit h ia n -2-yl)-5R-ch olest a n -3-ol: 1H
NMR (CDCl3) δ 4.12 (s, 1H), 2.97-2.83 (m, 4H), 2.15-2.05 (m,
tr a n s-1-(1,3-Dith ia n -2-yl)-3-p h en yl-2-p r op en ol (14):26
1H NMR (CDCl3) δ 7.41 (d, 2H, J ) 7.45 Hz), 7.32 (t, 2H, J )
7.45 Hz), 7.24 (t, 1H, J ) 7.5 Hz), 6.74 (d, 1H, J ) 16.1 Hz),
6.35 (dd, 1H, J ) 16.1, 6.4 Hz), 4.59-4.52 (m, 1H), 4.02 (d,
1H, J ) 6.8 Hz), 3.00-2.91 (m, 2H), 2.82-2.72 (m, 2H), 2.68
(d, 1H, J ) 3.4 Hz), 2.14-2.04 (m, 1H), 2.04-1.93 (m, 1H).
(19) Seebach, D.; Corey, E. J . J . Org. Chem. 1975, 40, 231-237.
(20) The reaction with aldehydes was normally completed within 2
h. More hindered and/or aromatic ketones required 12+ h in the freezer
to react.
(21) Einborn, J .; Luche, J . L. J . Org. Chem. 1987, 52, 4124-4126.
(22) Katzenellenbogen, J . A.; Bowlus, S. B. J . Org. Chem. 1973, 38,
627-632.
(25) Watanabe, Y.; Ono, Y.; Hayashi, S.; Ueno, Y.; Toru, T. J . Chem.
Soc., Perkin Trans. 1 1996, 1879-1885.
(26) Andersen, N. H.; McCrae, D. A.; Grotjahn, D. B.; Gabhe, S. Y.;
Theodore, L. J .; Ippolito, R. M.; Sarkar, T. K. Tetrahedron 1981, 37,
4069-4079.
(23) Bel-Rhlid, R.; Renard M. F.; Veschambre, H. Bull. Soc. Chim.
Fr. 1996, 133, 1011-1021.
(24) Guanti, G.; Banfi, L.; Narisano, E. Tetrahedron Lett. 1986, 27,
3547-3550.

9930 J . Org. Chem., Vol. 63, No. 26, 1998
McHale and Kutateladze
Ta ble 4
Gen er a l P r oced u r e for P h otolysis of th e Ad d u cts in
th e P r esen ce of Ben zop h en on e. A 10-2 M solution of a
dithiane-carbonyl adduct was irradiated in acetonitrile con-
taining benzophenone as a sensitizer (the same 10-2
M
concentration) for the period of time listed in Table 1. A
medium-pressure Hg Hanovia lamp with a Pyrex sleeve was
used as the UV source. After irradiation, an appropriate
amount of dodecane was injected into the reaction mixture as
an internal standard and the yield of the liberated carbonyls
was determined by gas chromatography using a calibration
curve set up for corresponding aldehydes and ketones. The
calibration procedure utilized the same internal standards
dodecane. Several freeze-thaw degassing cycles were found
to accelerate the reaction, without affecting the preparative
yield of deprotection. The times listed in Table 1, however, are
of nondegassed solutions to demonstrate that even in the
presence of oxygen the reaction is completed within 2-3 h.
Quantum yield determinations (see below) were carried out
with thoroughly degassed solutions.
* Average of two measurements.
Ta ble 5
Ta ble 6
Ta ble 7
P h otolysis of Cin n a m a ld eh yd e Ad d u ct 14: Tr a n s-Cis
Isom er iza tion . A degassed (four freeze-thaw cycles) 0.01 M
acetonitrile-d3 solution of trans-1-(1,3-dithian-2-yl)-3-phenyl-
2-propenol (14) was irradiated in a Pyrex NMR tube with 0.01
M benzophenone present. The reaction progress was monitored
by NMR. Before irradiation: 1H NMR (CD3CN) δ 7.8-7.2 (m,
∼15H), 6.66 (d, 1H, J ) 15.9 Hz), 6.32 (dd, 1H, J ) 15.9, 6.6
Hz), 4.45-4.39 (m, 1H), 4.16 (d, 1H, J ) 6.1 Hz), 3.43 (d, 1H,
J ) 4.7 Hz), 2.93-2.77 (m, 4H), 2.09-2.00 (m, 1H), 1.86-1.74
(m, 1H). After 25 min of irradiation a 35:65 mixture of the
trans-cis isomers was obtained: 1H NMR (CD3CN) δ 7.8-7.2
(m, ∼15H), 6.67 (d, 0.65H, J ) 11.6 Hz), 6.66 (d, 0.35H, J )
15.9 Hz), 6.32 (dd, 0.35H, J ) 15.9, 6.6 Hz), 5.77 (dd, 0.65H J
) 11.6, 9.5 Hz), 4.63-4.56 (m, 0.65H), 4.45-4.39 (m, 0.35H),
4.16 (d, 0.35H, J ) 6.1 Hz), 4.14 (d, 0.65H, J ) 6.6 Hz), 3.43
(d, 0.35H, J ) 4.7 Hz) 3.39 (d, 0.65H, J ) 5.1 Hz), 2.94-2.71
(m, 4H), 2.09-1.97 (m, 1H), 1.86-1.70 (m, 1H). We also
observed trace amounts of cis- and trans-cinnamaldehydes
(<3% by NMR integration): 9.94 (d, J ) 8.2 Hz) and 9.68 (d,
J ) 7.6 Hz) in a ratio of 2:7, respectively. After extended
irradiation for 1.5 h, the ratio of cis and trans adducts did not
change. The amount of free aldehydes also did not increase.
Rela tive Qu a n tu m Yield Stu d y. Ad d u cts 2, 8, a n d 9.
All samples were irradiated in Pyrex test tubes. Since the
absolute yields of deprotection of substituted benzaldehydes
were found to approach 100%, the quantum yield study was
based on monitoring the disappearance of the starting mate-
rial, measured by GC as an adduct:sensitizer ratio. First, 56.6
mg of the unsubstituted adduct 2 was weighed into a 25 mL
volumetric flask and diluted to volume with the sensitizer
stock solution (0.02 M benzophenone). Next, 62.8 mg of the
p-CN-substituted adduct 8 was weighed into another 25 mL
volumetric flask and diluted to volume with the sensitizer
stock solution. These amounts give 0.01 M solutions of each
adduct. Each sample was injected into the GC to determine
the prephotolysis ratio of adduct:sensitizer. After degassing
by four freeze-pump-thaw cycles, each adduct was irradiated
in duplicate (total of four samples) for 5 min. To ensure even
distribution of light, a carousel Rayonet photoreactor was used
in these experiments. After the irradiation, each sample was
injected into the GC again to obtain the new ratios. The same
procedure was used for comparison of the p-methoxy adduct
9: 64.1 mg of the p-methoxybenzaldehyde adduct 9 was
weighed into a 25 mL volumetric flask and diluted to volume
with the sensitizer solution. Degassed solutions of 2 and 9 were
irradiated in the carousel Rayonet photoreactor. Tables 4-6
contain the actual measured ratios and calculated percent
conversion.
aqueous solutions. The sensitizer (benzophenone) was used
again as an internal standard. The actual measured ratios, %
conversions, and KIE are presented in Table 7.
4-(Hyd r oxy-(1,3-d ith ia n -2-yl)m eth yl)tr ityl a lcoh ol (17).
A total of 0.6 g (5 mmol) of dithiane was dissolved in 40 mL of
freshly distilled THF, and the solution was cooled to -20 °C
under nitrogen. Then 2.8 mL of n-butyllithium (a 2 M solution
in hexanes) was added dropwise upon stirring. The resulting
mixture was stirred for 2.5 h. After that the temperature was
lowered to -78 °C and 0.82 g (5 mmol) of ethyl 4-formylben-
zoate dissolved in 2 mL of THF was added upon stirring. The
reaction mixture was stirred for 1 h at this temperature and
then was allowed to warm to 0 °C and stirred for 1 h. At this
point a small portion (0.2 mL) of the reaction mixture was
quenched with aqueous ammonium chloride, extracted with
1 mL of ether, dried over anhydrous Na2SO4, and analyzed
with GC-MS. The GC-MS analysis showed only one major
product present with an M+ of 284, which corresponds to
adduct 16: MS (EI) m/z 284 (M+), 266, 253, 165, 147, 134, 119
(100%), 106, 105, 91, 85, 77, 59, 45. The reaction sequence was
then continued without isolation of 16. To the reaction mixture
containing mainly 16 at 0 °C and under N2 was added 12 mL
of 1 M PhMgBr (12 mmol) in THF slowly. The reaction was
allowed to warm to room temperature and stirred for another
2 h. It was then quenched with a 1 M aqueous solution of NH4-
Cl, extracted twice with ether, and dried over Na2SO4. The
solvent was then removed with a rotary evaporator, and the
residue was purified using column chromatography (silica gel,
ethyl acetate-hexane, 1:3) furnishing 1.861 g (91%) of a white
crystalline solid. Upon minimal heating, the compound turn
brown and melted with decomposition, becoming a dark red-
brown liquid: 1H NMR (CDCl3) δ 7.44-7.03 (m, 14H), 4.93
(d, 1H, J ) 7.6 Hz), 4.06 (d, 1H, J ) 7.6 Hz), 3.07 (s, 1H),
3.02-2.93 (m, 2H), 2.91 (s, 1H), 2.78-2.67 (m, 2H), 2.14-1.95
(m, 2H); 13C NMR (CDCl3) δ 146.736, 146.538, 139.029, 127.754
(several Cs), 127.087, 126.377, 81.780, 74.114, 52.211, 27.837,
27.170, 25.240; MS (EI) m/z 390 (M+ - H2O), 119 (100%). Due
to thermal instability of 17, we were unable to completely
remove traces of ethyl acetate from the solid and obtained the
following analytical data (low in carbon); HRMS calcd for
Kin etic Isotop e Effect Stu d y. Using the same stock
solutions prepared for the relative quantum yield study for
each of the three benzaldehyde adducts 2, 8, and 9, we
removed four 4 mL portions and placed them all in separate
Pyrex tubes. Taking two tubes of each adduct solution, we
added 100 µL of H2O. To the remaining six tubes we added
100 µL of D2O. These tubes were allowed to stand overnight
for isotopic exchange to take place. The samples were 2.5%
C
C
24H24O2S2 408.1218, found 408.1053 ( 0.005. Anal. Calcd for
24H24O2S2: C, 70.55; H, 5.92. Found: C, 66.85; H, 5.60.
P h otolysis of 17 in th e P r esen ce of Ben zop h en on e. A
total of 204 mg (0.5 mmol) of 17 was dissolved in 50 mL of

Photo-SET-Induced Cleavage of Dithiane-Carbonyl Adducts
J . Org. Chem., Vol. 63, No. 26, 1998 9931
acetonitrile containing 91 mg (0.5 mmol) of benzophenone, and
the solution was irradiated with a Pyrex sleeve filter for 2.5
h. The solvent was evaporated, and the residual reaction
mixture was column-chromatographed (silica gel, ethyl acetate-
hexane, 1:3) to give 139 mg (96%) 18 as a faintly yellow solid:
1H NMR (CDCl3) δ 9.99 (s, 1H), 7.83 (d, 2H, J ) 8.4 Hz), 7.52
(d, 2H, J ) 8.4 Hz), 7.38-7.15 (m, 10H), 2.58 (s, 1H); MS (EI)
m/z 288 (M+), 241 (100%). Anal. Calcd for C20H16O2: C, 83.31;
H, 5.59. Found: C, 82.96; H, 5.83.
Ack n ow led gm en t. Thanks are due to Mrs. Lisa X.
Deng for assistance at the initial stage of this project.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR, 13C NMR,
1D NOE differences, and MS spectra for compounds 11, 12,
16 (MS only), 17, 18, and 7a (26 pages). See any current
masthead page for ordering or Internet access information.
J O981697Y