Mechanistic organic chemistry in a microreactor. Zeolite-controlled photooxidations of organic sulfides

Article abstract of DOI:10.1021/jo0261808

The intrazeolite and solution photooxygenations of a series of sulfides have been compared. The unusual zeolite environment enhances the rates of reaction, it suppresses the Pummerer rearrangements, and it has a dramatic effect on the sulfoxide/sulfone ratio. A detailed kinetic study utilizing trapping experiments and intramolecular competition provides evidence for cation complexation to a persulfoxide intermediate as the underlying phenomenon for the unique intrazeolite behavior. For example, the enhanced rate of reaction is traced to the cation stabilization of the persulfoxide toward unproductive decomposition to substrate and triplet oxygen.

Full text of DOI:10.1021/jo0261808

Mech a n istic Or ga n ic Ch em istr y in a Micr or ea ctor .
Zeolite-Con tr olled P h otooxid a tion s of Or ga n ic Su lfid es
Edward L. Clennan,* Wenhui Zhou, and J acqueline Chan
Department of Chemistry, University of Wyoming, Laramie, Wyoming 82071
Clennane@uwyo.edu
Received J uly 10, 2002
The intrazeolite and solution photooxygenations of a series of sulfides have been compared. The
unusual zeolite environment enhances the rates of reaction, it suppresses the Pummerer
rearrangements, and it has a dramatic effect on the sulfoxide/sulfone ratio. A detailed kinetic study
utilizing trapping experiments and intramolecular competition provides evidence for cation
complexation to a persulfoxide intermediate as the underlying phenomenon for the unique
intrazeolite behavior. For example, the enhanced rate of reaction is traced to the cation stabilization
of the persulfoxide toward unproductive decomposition to substrate and triplet oxygen.
Remarkable progress in the control of regio- and
stereoselectivity in organic transformations has been
achieved in the past 25 years.1 However, the further
development of new methodology to control the introduc-
tion of stereogenic elements is essential in order to devise
economically efficient syntheses of complex organic mol-
ecules. A fundamental feature of many successful regio-
and/or stereoselective synthetic strategies is the simul-
taneous control of the conformational flexibility and of
the approach geometry of the reacting components.
Synthetically useful levels of control utilizing this feature
have been achieved by either careful reaction component
design or by the use of supramolecular systems to impart
“enzyme-like” organization to the substrates and acti-
vated complexes.2 For example, Adam and co-workers,3-5
have used the reaction component design approach to
cleverly couple allylic strain (A1,3 strain) with hydrogen-
bonding effects to control both the conformation of allylic
alcohol 1 and the direction of singlet oxygen attack to
give stereoselectively the S*S* diastereomer.
zeolites as supramolecular hosts to influence the regio-
chemistry and stereochemistry14 of the singlet oxygen ene
reaction. Despite the advantages of this supramolecular/
zeolite approach, rationale implementation is impeded
by the paucity of information on how the zeolite environ-
ment can influence reaction mechanisms. To help rectify
this situation, we describe here a mechanistic study of
the reactions of singlet oxygen with organic sulfides in
the interior of zeolite Y.15
Zeolites are crystalline solids consisting of catenated
silicon and aluminum tetrahedra that enclose regular
repeating cavities or channels of well-defined size and
shape. Zeolite Y (NaY) is a member of the faujasite family
with a typical unit cell composition of Na56(AlO2)56-
(SiO2)136‚264H2O.16 It is characterized by a honeycomb
structure consisting of eight supercages per unit cell each
approximately 13 Å in diameter and each accessible
through tetrahedrally arranged windows approximately
7.4 Å in diameter (Scheme 1). The supercages of zeolite
Y are electrostatically charged as a result of seven
counterions that balance the negative charges on the
tetravalent framework aluminum atoms.
(7) Robbins, R. J .; Ramamurthy, V. J . Chem. Soc. Chem. Commun.
1997, 1071-1072.
(8) Shailaja, J .; Sivaguru, J .; Robbins, R. J .; Ramamurthy, V.; Sunoj,
R. B.; Chandrasekhar, J . Tetrahedron 2000, 56, 6927-6943.
(9) Stratakis, M.; Froudakis, G. Org. Lett. 2000, 2, 1369-1372.
(10) Stratakis, M.; Rabalakos, C. Tetrahedron Lett. 2001, 42, 4545-
4547.
(11) Stratakis, M.; Kosmas, G. Tetrahedron Lett. 2001, 42, 6007-
6009.
The supramolecular approach has also been exten-
sively utilized. For example, several groups6-13 have used
(12) Clennan, E. L.; Sram, J . P. Tetrahedron Lett. 1999, 40, 5275-
5278.
(13) Clennan, E. L.; Sram, J . P. Tetrahedron 2000, 56, 6945-6950.
(14) J oy, A.; Robbins, R. J .; Pitchumani, K.; Ramamurthy, V.
Tetrahedron Lett. 1997, 38, 8825-8828.
* Corresponding author. Fax: 307-766-2807.
(1) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
J ohn Wiley & Sons: New York, 1994; Chapter 12, pp 835-990.
(2) Clennan, E. L. Tetrahedron 2000, 56, 9151-9179.
(3) Adam, W.; Nestler, B. J . Am. Chem. Soc. 1992, 114, 6549-6550.
(4) Adam, W.; Nestler, B. J . Am. Chem. Soc. 1993, 115, 5041-5049.
(5) Adam, W.; Gevert, O.; Klug, P. Tetrahedron Lett. 1994, 35, 1681-
1684.
(15) For a preliminary account of this research, see: Zhou, W.;
Clennan, E. L. J . Chem. Soc. Chem. Commun. 1999, 2261-2262m.
Zhou, W.; Clennan E. L. Org. Lett. 2000, 2, 437-440. Zhou, W.;
Clennan, E. L. J . Am. Chem. Soc. 1999, 121, 2915-2916.
(16) Ramamurthy, V.; Garcia-Garibay, M. A. In Solid-State Supro-
molecular Chemistry: Two- and Three-Dimensional Inorganic Net-
works; Alberti, G., Bein, T., Eds.; Elsevier Science Ltd.: Tarrytown,
New York, 1996; Vol. 7, pp 693-719.
(6) Li, X.; Ramamurthy, V. J . Am. Chem. Soc. 1996, 118, 10666-
10667.
10.1021/jo0261808 CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/03/2002
9368
J . Org. Chem. 2002, 67, 9368-9378

Zeolite-Controlled Photooxidations of Sulfides
SCHEME 1
1,2-shift of the hydroxy group (OH) from oxygen to sulfur
to ultimately give sulfone.24
Experimental evidence that supports the solution-
phase mechanism depicted in Scheme 2 includes (1)
formation of Ph2SO2 during cophotooxidations with Ph2-
SO, which is itself inert to singlet oxygen (kSO in Scheme
2); (2) formation of Ph2SO during cophotooxidations with
Ph2S, which is also inert to singlet oxygen (kPhS in Scheme
2); (3) the kinetic demonstration that Ph2S but not Ph2-
SO competes with the sulfide substrate for a reaction
intermediate; (4) the kinetic demonstration that Ph2SO
but not Ph2S increases the quantum yield of the reaction
by competing with physical quenching, kq; (5) the use of
substituted diaryl sulfide and sulfoxide trapping agents,
which demonstrate that the first intermediate is nucleo-
philic and the second intermediate is electrophilic; and
(6) product isotope effects, which require the removal of
an R-hydrogen on the reaction surface for formation of
the sulfoxide product. Our experimental approach in this
study was to first survey the intrazeolite photooxidations
of a variety of sulfide structural types (e.g. dialkylsul-
fides, cyclic sulfides, acyclic sulfides, and aryl sulfides)
and then to focus on a mechanistic examination of one
of the sulfide substrates. The results from this experi-
mental approach are described below.
SCHEME 2. A Gen er ic Mech a n ism for th e
Rea ction of Sin glet Oxygen w ith Su lfid es in
Ap r otic Solven ts
Resu lts
The sulfide singlet oxygen system was chosen for our
initial investigations since dialkyl and aryl sulfide pho-
tooxidations are expected to be dramatically affected by
the extraordinary environment found in the interior of
zeolite Y. This expectation is a direct result of extensive
studies over the past 20 years that have culminated in a
mechanistic suggestion (Scheme 2) that invokes two
charge-separated intermediates, the persulfoxide 2 and
the hydroperoxysulfonium ylide 3, on the potential
energy surface.17-19
The addition of 1O2 to the sulfide occurs on a very flat
potential energy surface to give persulfoxide 2 as the first
stationary point. It is the decomposition of this interme-
diate along a physical quenching pathway, kq, that is
responsible for the very low experimentally observed
quantum yields (Φ < 0.05) in these reactions. The
persulfoxide 2, in competition with physical quenching,
also undergoes a very novel intramolecular hydrogen
abstraction from the R-carbon to give the second observed
stationary point on the reaction surface, the hydroper-
oxysulfonium ylide 3. This second intermediate then
reacts with the sulfide substrate to give two molecules
of sulfoxide. Not depicted in Scheme 2 are two very
interesting rearrangements of 3 that have been observed
as competing processes with some substrates: (1) a 1,2-
shift of the hydroperoxy group (OOH) to the anionic
R-carbon in a Pummerer-like rearrangement to ulti-
mately give S-C bond cleavage products20-23 and (2) a
P r od u ct Stu d ies. The homogeneous (solution) and
heterogeneous (intrazeolite) photooxidations of 11 sul-
fides have been examined. A comparison of the results
in the two media and the details of the intrazeolite
photooxidations are given in Figure 1 and Table 1,
respectively.
The intrazeolite photooxidations were conducted by
irradiations of oxygen-saturated hexane slurries contain-
ing the sulfide substrate and methylene blue-doped NaY
with a 600-W tungsten-halogen lamp through a 400-nm
cutoff filter (saturated NaNO2). The methylene blue-
doped zeolite (NaMBY) was prepared by adding NaY to
an aqueous solution of methylene blue and stirring until
the blue color of the water completely faded and the
zeolite powder turned blue. The NaMBY was then filtered
and dried at 100 °C at 10-4 Torr for 24 h. The loading
level of methylene blue was on average one molecule per
every 100 supercages ( S ) 0.01) or 15 molecules per
every 10 000 supercages ( S ) 0.0015). In most of the
work, the higher loading level was used, since sensitizer
bleaching was a minor annoyance at the lower loading
level. The reactions were worked up by filtration followed
by mild digestion of the zeolite with 10% aqueous HCl
and extraction with acetonitrile, tetrahydrofuran, or
chloroform. The percent recovery with this workup
protocol was in excess of 95%, except where noted (Table
1).
Gas chromatographic monitoring of the hexane com-
ponent of the slurries demonstrated that migration of the
sulfides into the interior of the zeolite was nearly
(17) J ensen, F.; Greer, A.; Clennan, E. L. J . Am. Chem. Soc. 1998,
120, 4439-4449.
(18) Toutchkine, A.; Clennan, E. L. J . Org. Chem. 1999, 64, 5620-
5625.
(19) Toutchkine, A.; Clennan, E. L. Tetrahedron Lett. 1999, 40,
6519-6522.
(20) Takata, T.; Hoshino, K.; Takeuchi, E.; Tamura, Y.; Ando, W.
Tetrahedron Lett. 1984, 25, 4767-4770.
(21) Takata, T.; Tamura, Y.; Ando, W. Tetrahedron 1985, 41, 2133-
2137.
(22) Takata, T.; Ishibashi, K.; Ando, W. Tetrahedron Lett. 1985, 26,
4609-4612.
(23) Clennan, E. L.; Aebisher, D. J . Org. Chem. 2002, 67, 1036-
1037.
(24) Ishiguro, K.; Hayashi, M.; Sawaki, Y. J . Am. Chem. Soc. 1996,
118, 7265-7271.
J . Org. Chem, Vol. 67, No. 26, 2002 9369

Clennan et al.
F IGURE 1. A Comparison of the solution and intrazeolite photooxidations of sulfides 4-14.
TABLE 1. In tr a zeolite P h otooxid a tion s of Su lfid es
4-14a
hν
%
%
%
time convn SOd,f % SO2
cleavage
b
c
e,f
sulfide
S
S
MB
S
4
5
6
7
8
0.01
0.23 1 h
100
100
100
100
46.5
70.7
89
98.7
35.5
36.2
44
46
67
69
93
97
94
89
78
88
58
56
54
33
25
7
3
6
11
2
1
0.0015 0.26 1 h
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
1.06 1 h
1.06 1 h
3
1.3
1.3
1.3
1 h
1 h
1 h
9
10
11
12
13
14
0.26 1 h
1.3
1.3
1.3
1 h
1 h
21
11
30 min 100
42
a
Conducted in a hexane slurry as described in the Experimental
b
Section. Number of methylene blue sensitizers per supercage.
c Number of substrate molecules per supercage. Sulfoxide. e S-
d
ulfone. f See Figure 1 for structures of substrates and products.
quantitative (>96-98%), even at the highest loading
levels ( S ) 1.3 sulfide molecules per supercage) utilized
in this study. Such high loading levels were not feasible
in all solvents. For example, a plot (Figure 2) of the
number of millimoles (T) of sulfide 5 in the interior of
the zeolite versus equilibration time for both hexane and
acetonitrile demonstrates that quantitative migration
F IGURE 2. Absorption profile for migration of sulfide 5 into
NaMBY ( S
) 0.01) in CH3CN and hexane.
MB
interior of the zeolite. Consequently, hexane was used
in all the studies described in this paper.
Under these carefully controlled hexane slurry reaction
conditions, interactions between singlet oxygen and the
sulfides occur exclusively in the interior of the zeolite.
This was established by a series of quenching experi-
ments with the quenchers â-carotene and 2,3-dimethyl-
(>98%) into the zeolite to give a loading level of S
)
1.4 occurred within 2 min in hexane. In contrast, even
after extended periods of time in the acetonitrile slurries,
the majority of 5 resided in solution rather than in the
9370 J . Org. Chem., Vol. 67, No. 26, 2002

Zeolite-Controlled Photooxidations of Sulfides
2-butene and the substrate diphenyl sulfide, 11. â-Car-
otene (kq ) 1.4 × 1010 M-1 s-1),25 which is too large to
migrate into the zeolite, has no effect on the percent
conversion of 11; however, 2,3-dimethyl-2-butene, which
can migrate into the zeolite, competitively inhibits the
reaction of 11 (see plots S1 and S2 of the Supporting
Information).
A comparison of the results from the homogeneous and
heterogeneous photooxidations (Figure 1) reveals several
unusual features: (1) the yields of sulfone were enhanced
by moving the photooxidations into the zeolite, (2)
sulfur-carbon bond cleavages via the Pummerer reaction
were dramatically suppressed during intrazeolite pho-
tooxidations of 4, 6, 7, 9, 10, and 12, and (3) diphenyl
sulfide, 11, was completely inert in solution but reacted
nearly quantitatively within 1 h in the zeolite. In fact,
most of the intrazeolite reactions occurred more rapidly
than their solution counterparts. This reactivity differ-
ence was especially evident for the aryl-substituted
sulfides.
Mech a n istic Stu d ies. As a direct result of the dra-
matic reactivity enhancement observed for the intrazeo-
lite photooxidation of diphenyl sulfide, 11, we have
focused much of our initial mechanistic work on this
compound. However, in a select number of experiments,
we have used thiane, 5, as a representative for dialkyl
sulfides to compare to the results obtained with 11. For
example, we have examined the effect of substrate
loading level on product composition during the intrazeo-
lite photooxidation of 5 and compared it to the results
obtained with 11. In both cases a linear increase (Figure
3) in the sulfoxide/sulfone ratio was observed as a
function of increasing loading level, S . (See Tables S1
and S2 of the Supporting Information for the raw data
used to generate Figure 3.) The sensitivity of the sulfox-
ide/sulfone ratio on
S was approximately 2.3 times
greater for 11 than 5, reflecting the smaller amount of
sulfone formed in the 11 intrazeolite photooxidation at
low concentrations and perhaps the larger molecular
volume of Ph2S, 11 (236.4 Å3), in comparison to thiane,
5 (148.4 Å3).26 This effect of substrate concentration is
unprecedented in solution, where little or no effect is
observed on the sulfoxide/sulfone ratios over a fairly large
substrate concentration range.27,28
The sulfoxide/sulfone ratio was also a sensitive function
of the degree of hydration of the zeolite. This phenomenon
is illustrated for 5 in Table 2, where it is compared to
the effect of water on the sulfoxide/sulfone ratio in
solution. When the zeolite was air-dried at atmospheric
pressure for 24 h, no photooxidation of 5 was observed,
even after 1 h of irradiation. Ramamurthy and co-
workers29 have previously reported that water promotes
the formation of thiazine dye dimers in NaY. The
formation of dimers in our 24 h air-dried NaMBY sample
was readily confirmed by comparison of solution and
diffuse reflectance UV-vis spectra, which demonstrated
F IGURE 3. The sulfoxide/sulfone product ratio as a function
of Ph2S (11, plot a) and thiane (5, plot b) loading levels.
TABLE 2. Effect of Wa ter on th e In tr a zeolite a n d
Solu tion P h otooxid a tion s of 5
drying time/[H2O], mol/L % conversion % 5-SOa % 5-SO2
b
zeolite (1 h hν)c
24 h (rt, 1 atm)
no reaction
78.5
72.2
1 he
27.8
60.5
81.5
72.2
39.5
18.5
2 he
8 he
77.6
solution (1 h hν)d
0
100
100
100
100
100
100
93.6
96.5
95.7
96.2
89.4
100
6.4
3.5
4.3
3.8
10.6
trace
0.05
0.1
0.5
1.0
5.0
a
Thiane sulfoxide. b Thiane sulfone. c In hexane slurry contain-
d
ing 1% ( S
) 0.01) NaMBY. In acetonitrile containing 6 ×
MB
10-4 M MB. e At 100 °C and 6 × 10-4 Torr.
a dramatic 69-nm hypsochromic shift of λMAX from 672
nm in solution to 603 nm in NaY. The absence of reaction
promoted by this dimeric Na(MB)2Y is consistent with
preferential self-quenching rather than singlet oxygen
formation. However, drying of the NaMBY sample at 100
°C and 6 × 10-4 Torr for as little as 1 h resulted in rapid
photooxidation. The efficiency of the photooxidation, as
(25) Wilkinson, F.; Helman, W. P.; Ross, A. B. J . Phys. Chem. Ref.
Data 1995, 24, 663-1021.
(26) PC Model, Serena Software, Box 3076, Bloomington, IN 47402-
3076.
(27) Clennan, E. L. In Advances in Oxygenated Processes; Baum-
stark, A. L., Ed.; J AI Press: Greenwich, CT, 1995; Vol. IV, pp 49-80.
(28) Clennan, E. L. Sulfur Rep. 1996, 19, 171-214.
(29) Ramamurthy, V.; Sanderson, D. R.; Eaton, D. F. J . Am. Chem.
Soc. 1993, 115, 10438-10439.
J . Org. Chem, Vol. 67, No. 26, 2002 9371

Clennan et al.
TABLE 3. P h otooxid a tion s of 11 a s a F u n ction of
Ca tion -Exch a n ged Zeolitea
zeolite
% conversion
[11-SO]/[11-SO2]
NaMBY
LiMBY
KMBY
RbMBY
CsMBY
BaMBY
95.8
95.2
38.3
4.8
0.5
50.8
89.1/10.9
91.6/8.4
95.2/4.8
100/0
100/0
98.6/1.4
a
Photooxidations conducted with MB loading levels of S ) 0.01
in 5-mL hexane slurries for 1 h. The concentrations of 11 in the
5-mL hexane samples were all 0.02 M prior to the addition of the
cation-exchanged zeolite.
F IGURE 4.
revealed by percent conversion, was independent of
drying time (from 1 to 8 h), but the amount of sulfone
dramatically decreased from 72.2% to 18.5% as the water
content decreased (Table 2). The diffuse reflectance
spectra also revealed a bathochromic shift of λMAX from
603 nm to a near solution-like λMAX of 676 nm after the
1 h drying period. No additional shift in λMAX was
observed, even after 8 h of drying under vacuum at 100
°C. In dramatic contrast, little or no change was observed
in the sulfoxide/sulfone ratio in acetonitrile over a wide
range of water concentrations (Table 2).
F IGURE 5.
The influence of the intrazeolite charge-balancing
counterion was explored with a series of ion-exchanged
zeolites. The zeolite framework atoms are ineffective at
providing extensive shielding from the high electrostatic
fields surrounding the cations.30 Consequently, these
electrostatic fields can play large roles in the stabilization
of charge-separated intermediates.30-47 These electro-
static fields, which are typically on the order of 1 to 10 V
nm-1 decrease in the order BaY > LiY > NaY > KY >
RbY > CsY.30 The data in Table 3 reveal a dramatic
electrostatic field effect during the intrazeolite photooxi-
dation of 11 on the percent conversions but only a minor
effect on the sulfoxide/sulfone ratio. The low percent
conversion in BaY in comparison to LiY or NaY, despite
the higher electrostatic field generated by Ba2+, reflects
the dichotomy between the size of Ba2+, which decreases
the supercage free volume, and the influence of the
electrostatic field.
The dramatic change in the intrazeolite % conversions
of 11 depicted in Table 3 can be attributed to either
electrostatic-field-induced enhancement of the rate of
persulfoxide (Ph2S+OO-) formation or to electrostatic-
field-induced stabilization of the persulfoxide, leading to
a decrease in physical quenching (i.e. Ph2S+OO- f Ph2S
+ 3O2). To distinguish between these two possibilities and
to develop a greater understanding of intrazeolite sulfide
dynamics, we have examined the intrazeolite photooxi-
dations of a series of aryl sulfides, 15a -f (Figure 5),
which are tethered to alkene linkages.48 Sulfide 15a ,
2-methyl-5-phenylthio-2-pentene, was synthesized by
treatment of 5-bromo-2-methyl-2-pentene49 with sodium
phenylthiolate. Sulfides 15b-f were synthesized by
treatment of 5-bromo-2,3-dimethyl-2-pentene with the
corresponding thiolate as shown in Figure 4.
(30) Vasenkov, S.; Frei, H. In Solid State and Surface Photochem-
istry; Ramamurthy, V., Schanze, K. S., Eds.; Marcel Dekker: New
York, 2000; Vol. 5, pp 295-323.
(31) Blatter, F.; Frei, H. J . Am. Chem. Soc. 1993, 115, 7501-7502.
(32) Blatter, F.; Frei, H. J . Am. Chem. Soc. 1994, 116, 1812-1820.
(33) Blatter, F.; Moreau, F.; Frei, H. J . Phys. Chem. 1994, 98,
13403-13407.
The photooxidations of sulfides 15a -f were examined
both in solution and in zeolite Y. The electron-rich
tetrasubstituted-alkene aryl sulfides 15b-e reacted in
solution exclusively at the double bond to give a mixture
of the allylic hydroperoxides 16-OOH and 18-OOH
(Figure 5 and Table 4). In all cases hydrogen abstraction
occurred to a small extent preferentially at the geminal-
dimethyl end of the double bond. As anticipated, the
substituent on sulfur had little or no influence over the
regiochemistry of the ene reaction at the remote olefinic
linkage. Oxidation at sulfur competed with reaction at
the double bond only in the trisubstituted alkene 15a and
in the dialkyl sulfide 15f, presumably as a result of
decreased double bond nucleophilicity and of enhanced
steric access to and/or enhanced nucleophilicity of sulfur,
(34) Blatter, F.; Sun, H.; Frei, H. Catal. Lett. 1995, 35, 1-12.
(35) Blatter, F.; Sun, H.; Frei, H. Chem. Eur. J . 1996, 2, 385-389.
(36) Sun, H.; Blatter, F.; Frei, H. J . Am. Chem. Soc. 1994, 116,
7951-7952.
(37) Sun, H.; Blatter, F.; Frei, H. J . Am. Chem. Soc. 1996, 118,
6873-6879.
(38) Frei, H. In Studies in Surface Science and Catalysis. 3RD World
Congress on Oxidation Catalysis; Grasselli, R. K., Oyama, S. T.,
Gaffney, A. M., Lyons, J . E., Eds.; Elsevier Science: New York, 1997;
Vol. 110, pp 1041-1050.
(39) Myli, K. B.; Larsen, S. C.; Grassian, V. H. Catal. Lett. 1997,
48, 199-202.
(40) Vasenkov, S.; Frei, H. J . Phys. Chem. B 1998, 102, 8177-8182.
(41) Xiang, Y.; Larsen, S. C.; Grassian, V. H. J . Am. Chem. Soc.
1999, 121, 5063-5072.
(42) Blatter, F.; Sun, H.; Vasenkov, S.; Frei, H. Catal. Today 1998,
41, 297-309.
(43) Panov, A. G.; Larsen, R. G.; Totah, N. I.; Larsen, S. C.; Grassian,
V. H. J . Phys. Chem. B 2000, 104, 5706-5714.
(44) Frei, H.; Blatter, F.; Sun, H. Chemtech 1996, 26, 24-30.
(45) Sun, H.; Blatter, F.; Frei, H. Catal. Lett. 1997, 44, 247-253.
(46) Vasenkov, S.; Frei, H. J . Phys. Chem. B. 1997, 101, 4539-4543.
(47) Larsen, R. G.; Saladino, A. C.; Hunt, T. A.; Mann, J . E.; Xu,
M.; Grassian, V. H.; Larsen, S. C. J . Catal. 2001, 204, 440-449.
(48) Zhou, W.; Clennan, E. L. Org. Lett. 2000, 2, 437-440.
(49) Hiranuma, S.; Shibata, M.; Hudlicky, T. J . Org. Chem. 1983,
48, 5321-5326.
9372 J . Org. Chem., Vol. 67, No. 26, 2002

Zeolite-Controlled Photooxidations of Sulfides
TABLE 4. P h otooxid a tion s of 15a -f in Solu tion a ,b
substrate % conversionc 15-SO 16-OOH 17-OOH 18-OOH
15a
15b
47
64
20
62
87
48
41
27
17
29
46
14
18
46
42
55
55
53
53
53
54
34
36
21
40
40
45
45
47
47
47
46
23
24
18
15c
15d
15e
15f
44
39
45
a
In CDCl3 at room temperature using 0.1 M 15a -f and 3 ×
10-4 M tetraphenylporphyrin (TPP) as sensitizer. Product ratios
were measure directly by proton NMR in the photooxidation
mixtures and are reproducible to (4%. c To get larger % conver-
sions the time of irradiation was extended.
b
respectively.50 Hydrogen abstraction from the tether to
give 17-OOH only occurred in 15a . The preferred 90°
dihedral angle (H-CH-CdC) for hydrogen abstraction
is energetically inaccessible as a result of the eclipsing
interaction between the CH2CH2 bond of the tether and
the vinyl geminal-methyl group, R, in 15b-f. This
eclipsing interaction is absent in 15a .
F IGURE 6.
TABLE 5. P r od u ct Distr ibu tion s in In tr a zeolite
P h otooxid a tion s
The allylic hydroperoxide products were indefinitely
stable in solution. Reductive decomposition, however, was
enhanced by removal of the solvent. Consequently, prior
to the concentration of the reaction mixtures, triph-
enylphosphine, PPh3, was added to all the samples to
quantitatively generate the corresponding alcohols and
triphenylphosphine oxide, Ph3PO. The different allylic
alcohols and hydroperoxides could easily be distinguished
by their characteristic proton NMR spectra (Experimen-
tal Section).
substrate
15a
15b
15c
15f
% conversion
15-SO
16-OH
16-OH(SO)
17-OH
17-OH(SO)
18-OH
26
73
13
14
a
30
57
19
10
26
53
10
18
40
68
14
9
a
6
8
7
12
3
6
18-OH(SO)
a
Formed at higher conversions.
The intrazeolite photooxidations of 15a -f proved to be
more complex than their counterparts in homogeneous
media. In addition to the sulfoxide, 15-SO, three different
allylic alcohols and their corresponding sulfoxides were
also isolated from the intrazeolite photooxidation mix-
tures (Figure 6). Remarkably, however, in stark contrast
to the homogeneous photooxidations, no allylic hydrop-
eroxides were isolated from the reaction mixtures. This
was verified by treating these reaction mixtures with
PPh3 and noting the lack of formation of Ph3PO. The
absence of allylic hydroperoxides was not due to a low
mass balance, which exceed 97% in all cases, nor was it
due to decomposition during workup. Control experi-
ments demonstrated that the hydroperoxides formed in
the solution photooxidation of 15a when subjected to the
same gentle acid wash and extraction procedures (see
Experimental Section) survived unchanged. In addition,
the proton NMR spectra of the reaction mixtures clearly
demonstrated the absence of ketones or aldehydes from
acid-catalyzed51 or intrazeolite cleavages31 of the hydro-
peroxides.
generated by adding a mixture consisting of 15a (58%),
15a -SO (9%), 16a -OOH (14%), 17a -OOH (13%), 16a -OH
(2.1%), 17a -OH (1.7%), and several ene sulfoxides (2.9%)
to a hexane slurry of NaY. This reaction mixture was
then subjected to the same isolation procedure afforded
the intrazeolite photooxidation mixtures. Analysis of the
results by NMR demonstrated that the hydroperoxy
groups are completely reduced to the alcohols in NaY in
the absence of methylene blue. In addition, 15a decreased
by 9.7% concomitant with a 9.7% increase in the concen-
tration of 15a -SO, consistent with an intermolecular
reduction pathway. The feasibility of an intermolecular
pathway was also demonstrated by the oxidation of
coadsorbed thioanisole. However, the inability to detect
a second diastereomer corresponding to 16-OH(SO)
provides the distinct possibility that this compound may
have formed via an intramolecular cyclic six-membered
ring transition state.
The intrazeolite photooxidation product ratios for
15a -c and 15f are given in Table 5 at the lowest percent
conversions convenient for analytical analysis. Overoxi-
dation of sulfoxide 15-SO and sulfides 16-OH, 17-OH,
and 18-OH is a problem at high conversion. A comparison
of the data in Tables 4 and 5 reveals a remarkable
increase in sulfoxide 15-SO formation as the reaction is
moved from solution into the zeolite. For example, 15a -
SO increases from 14% in solution to 73% in the zeolite.
We suggest that the sulfide moiety, by an intrazeolite
intermolecular process with a small intramolecular con-
tribution, spontaneously reduced the allylic hydroperox-
ides. Experimental evidence for this suggestion was
(50) It is also possible that physical quenching, kq in Scheme 2,
competes more effectively in aryl-substituted sulfides.
(51) Porter, N. A. In Organic Peroxides; Ando, W., Ed.; J ohn Wiley
& Sons Ltd: New York, 1992; pp 101-156.
J . Org. Chem, Vol. 67, No. 26, 2002 9373

Clennan et al.
TABLE 6. Rela tive Ra te Con sta n ts for In tr a zeolite
Rea ction s a t Su lfu r a n d th e Dou ble Bon d a s a F u n ction
of Cou n ter ion
substrate
MY
% conversionb
[kr(sulfide)]/[kr(olefin)]
15a
LiYa
NaYa
KY
RbY
CsY
LiYa
NaYa
KY
RbY
CsY
LiYa
NaYa
KY
30.2
25.9
36.8
23.2
9.5
34.0
30.7
26.1
35.2
21.5
45.4
39.5
36.0
30.2
14.8
1.68
2.31 ( 0.22
0.65
0.92
0.48
0.56
15b
15f
0.87 ( 0.10
0.27
0.14
0.04
1.68
2.26 ( 0.41
3.08
2.70
1.15
RbY
CsY
a
These samples were only irradiated for 5 min. All other
b
samples were irradiated for 15 min. See Table S3 of the Sup-
porting Information for product ratios.
In fact, 73% may be an underestimate, even at the
modest percent conversion reported in Table 5, of the
contribution of 15a -SO to the product ratio, because of
nefarious overoxidation.
F IGURE 7. Diphenylsulfoxide/thiolane sulfone product ratio
as a function of both the concentrations of Ph2S, 11, and
thiolane sulfoxide.
A lower limit for the relative rate constants for reaction
at sulfur and the olefinic linkage [kr(sulfide)/kr(olefin)]
can be calculated both in solution and in the interior of
the zeolite for 15a and 15f, which give products at both
sites in both media. The calculation uses the data in
Tables 4 and 5 in conjunction with eq 1.52
at sulfur than those generating smaller electrostatic
fields. This trend is very evident for aryl-substituted
sulfides 15a and 15b but less evident for dialkyl sulfide
15f.
To examine the possibility that an intermediate could
be trapped in these experiments, the co-intrazeolite
photooxidations of 11 and thiolane sulfoxide were exam-
ined. Control experiments demonstrated that thiolane
sulfoxide in the absence of 11 is inert under the reaction
conditions. However, coinclusion with 11 resulted in
formation of thiolane sulfone, providing irrefutable evi-
dence for an oxidatively active intermediate. A quantita-
tive treatment of these trapping results shown in Figure
7 demonstrated that the (diphenyl sulfoxide/thiolane
sulfone) ratio was a function of both the thiolane sulfoxide
and 11 concentrations. This result is consistent with
competitive trapping of the intermediate with thiolane
sulfoxide and the substrate 11.53
kr(sulfide) ln[Co(sulfide)/Ct(sulfide)]
)
(1)
kr(olefin)
ln[Co(olefin)/Ct(olefin)]
The success of this approach is a direct result of the
presence of the tether linking the sulfide and olefinic
sites, which guarantees the close proximity of the two
functional groups and as a consequence equal access to
singlet oxygen. In this situation [15]o ) Co(sulfide) ) Co-
(olefin). However, Ct(sulfide) must be corrected for the
amount of 15-SO formed by intermolecular reduction of
the hydroperoxides in the zeolite and is therefore set
equal to [15]o - [15-SO] + ([16-OH] + [17-OH] + [18-
OH]) for the zeolite photooxidations. The concentration
of the olefin at time t, Ct(olefin), is set equal to [15]o -
∑[ene products]. This treatment gives a [kr(sulfide)/kr-
(olefin)] of 0.14 ( 0.02 and 2.31 ( 0.22 for 15a and 0.70
( 0.07 and 2.26 ( 0.41 for 15f in solution and in the
zeolite, respectively. These data indicate that by moving
the photooxidations of 15a and 15f from solution to the
interior of the zeolite, oxidation rate constants at sulfur
are enhanced in comparison to the rate constants at the
olefinic sites by factors greater than 16- and 3-fold,
respectively.
Discu ssion
The unique features of the sulfide intrazeolite photo-
oxidations that distinguish them from their solution
counterparts include (1) an enhanced rate of reaction, (2)
the ability of water to influence the sulfoxide/sulfone
ratio, (3) the suppression of the Pummerer rearrange-
ment (cleavage), (4) the ability of the substrate concen-
tration to influence the sulfoxide/sulfone ratio, and (5)
the ability of counterions to influence the efficiency of
the reactions. We suggest that these unusual results can
be rationalized by the mechanism shown in Scheme 3
for the intrazeolite photooxidation of diphenyl sulfide, 11.
The key element of this new intrazeolite mechanism
is the formation of a counterion complexed persulfoxide
intermediate, I. It differs from the solution mechanism
(Scheme 2) by invoking only one rather than two inter-
The relative rate constants [kr(sulfide)/kr(olefin)] were
also measured as a function of counterion for the in-
trazeolite photooxidations of 15a , 15b, and 15f (Table 6).
In general, counterions generating high electrostatic
fields, Na+ and Li+, show a higher propensity for reaction
(52) Higgins, R.; Foote, C. S.; Cheng, H. In Advances in Chemistry
Series; Gould, R. F., Ed.; American Chemical Society: Washington,
DC, 1968; Vol. 77, pp 102-117.
(53) Liang, J .-J .; Gu, C.-L.; Kacher, M. L.; Foote, C. S. J . Am. Chem.
Soc. 1983, 105, 4717-4721.
9374 J . Org. Chem., Vol. 67, No. 26, 2002

Zeolite-Controlled Photooxidations of Sulfides
SCHEME 3
mediates. The complexation with the counterion (Na+ in
NaMBY) enhances the lifetime of the persulfoxide, in
comparison to solution, allowing its capture by water
SCHEME 4
(kH O) and inhibiting its decomposition via the physical
2
quenching channel (kq) and its interconversion to the
second intermediate (kX) (Scheme 2). In addition, the
complexation diminishes the negative charge on the
peroxy appendage, allowing nucleophilic attack by the
sulfide substrate (kS) and transfer of oxygen. At the same
time the diminution of charge decreases the rate of
oxygen transfer to advantageous sulfoxide product (kSO
or added electrophilic trapping agent R′2SO (kSO′).
)
In Scheme 3, diphenyl sulfide, 11, and sulfoxide R′2-
SO compete for the same intermediate. Consequently, at
low conversions, where kSO does not contribute to diphe-
nyl sulfoxide formation, treating persulfoxide I as a
steady-state intermediate allows derivation of eq 2.
concentration) of the sulfide is also consistent with the
mechanism shown in Scheme 354 (eq 3).
[Ph2SO]
2kS[Ph2S]
) 1 +
(2)
[R′2SO2]
kSO′[R′2SO2]
[Ph2SO]
2kS[Ph2S]
)
(3)
This equation predicts that [Ph2SO]/[R′2SO2] should be
linearly related to 1/[R′2SO] (eq 2) exactly as observed
in Figure 7.
[Ph2SO2] kH O[H2O]
2
The ability of the sulfide concentration to influence the
sulfoxide/sulfone ratio is direct evidence that the sulfide
is trapping an intermediate that also serves as a precur-
sor to the sulfone.
A similar equation was derived by Foote and co-
workers for the reaction of singlet oxygen with diethyl
sulfide in methanol53 (Sch em e 4). In this protic solvent,
Et2S and an added sulfoxide compete for a single trap-
pable intermediate formulated as either a hydrogen-
bonded persulfoxide, II, or sulfurane, III. The value of
kS/kSO′ derived from the slopes of the lines in Figure 7 is
0.64 ( 0.12, considerably smaller than the kS/kSO′ of 2.77
measured by Foote and co-workers for diethyl sulfide in
CH3OH (Scheme 4). This difference can be attributed to
(i) an enhanced nucleophilicity of Et2S in comparison to
Ph2S, (ii) a difference in reactivity of thiolane sulfoxide
in comparison to Ph2SO, or (iii) a greater negative charge
on the peroxy linkage in the intrazeolite intermediate,
(I in Scheme 3) than on the peroxy linkage in the solution
intermediate (II or the sulfurane III in Scheme 4).
The linear relationship depicted in Figure 3 between
the sulfoxide/sulfone ratio and the doping level (i.e.
The intrazeolite-induced increase in reaction at the
sulfide moiety in comparison to the olefinic linkage in
tethered substrates 15a -f can also be accommodated
within the framework of the mechanism depicted in
Scheme 3. The enhancement in sulfur oxidation can be
a result of either an increase in the rate of formation of
persulfoxide I (kT) or an inhibition of its decomposition
via the physical quenching channel (kq) (Scheme 3). Both
effects can be attributed to the stabilization of the
persulfoxide in the electrostatic field of the associated
(54) The nonzero intercept and lower quality linear relationship
observed for 11 in comparison to 5 (Figure 3) can be attributed to the
very minor amount of sulfone formed in the reaction of 11 and the
corresponding inaccuracy in measuring the sulfoxide/sulfone ratio.
J . Org. Chem, Vol. 67, No. 26, 2002 9375

Clennan et al.
SCHEME 5
counterion. In principle, the contributions of these two
effects can be separated by independently measuring the
rate constants for disappearance of singlet oxygen and
the rate constants for product formation. If the only effect
operating in the zeolite is suppression of physical quench-
ing, the rate constant of substrate-induced singlet oxygen
disappearance should be identical in solution and in the
zeolite. The rate constant for disappearance of singlet
oxygen can be measured in solution by following the
substrate quenching of the time-resolved emission of
singlet oxygen at 1270 nm.55 Unfortunately, all our
attempts to use this method to follow intrazeolite singlet
oxygen decay have failed because of our inability to
adequately account for the massive amounts of scattered
light observed in these heterogeneous sample slurries.
If suppression of physical quenching is the only effect
operating by moving these reactions into the interior of
the zeolite, the relative rate constants [kr(sulfide)/kr-
(olefin)] derived from eq 1 could never exceed the ratio
of solution kT values for reaction at the sulfide and
olefinic sites. These kT ratios cannot be measured directly
with the tethered substrates; however, they can be
estimated using suitable monofunctionalized model
substrates.56-58 This analysis predicts kT(sulfide)/kT(ole-
fin) values of 3.84, 0.66, 0.28, and 6.0 for 15a , 15b, 15c,
and 15f, respectively. These calculated values are suf-
ficiently close to the experimental values (2.31 ( 0.22,
0.87 ( 0.10, 0.70 ( 0.07, and 2.26 ( 0.41, for 15a , 15b,
15c, and 15f, respectively) to suggest that suppression
of physical quenching has a major role, if not the
exclusive role, in the ability of the zeolite to promote
sulfide photooxidation.
Th e in tr a zeolite beh a vior of th ia n e, 5, mimics in
many respects the intrazeolite behavior observed for
diphenyl sulfide, 11. This similarity argues for analogous
“single intermediate” intrazeolite mechanisms for the two
substrates. The complete suppression of the Pummerer
rearrangement in the five- and seven-membered ring
sulfides, 4 and 6, respectively, also lends support to this
suggestion, since the second intermediate, the hydrop-
eroxysulfonium ylide (3 in Scheme 2), serves as a direct
precursor for the Pummerer product.
The “single intermediate” mechanism depicted in
Scheme 3, however, cannot be a general mechanism. In
acyclic sulfides, 7, 9, 10, 12, and 13, sulfur-carbon bond
cleavage products from the Pummerer rearrangements
are less important but, nevertheless, are still observed
in the zeolite. These results require formation of the
hydroperoxy sulfonium ylide in the zeolite. In addition,
the intramolecular hydrogen-abstraction product, ethyl
vinyl sulfoxide, is formed to the same extent in solution
and in intrazeolite photooxidations of 14.19 This suggests
that in this case complexation of sodium to the persul-
foxide cannot suppress the very rapid kX interconversion
(Scheme 2) to the extent that allows competitive trapping
of the persulfoxide by the sulfide substrate (kS in Scheme
3).
This analysis suggests that cation complexation de-
creases, but does not completely abolish, the negative
charge on the peroxy linkage of the persulfoxide, and as
a result, hydrogen abstraction (kX in Scheme 2) is
suppressed but not completely eliminated. Consequently,
the complete suppression of hydroperoxy sulfonium ylide
formation in the intrazeolite reactions of cyclic sulfides
4 and 6 is surprising. We suggest that in these cyclic
sulfides a change in the conformation of the persulfoxide
upon inclusion into the zeolite also contributes to the
complete competitive inhibition of kX. Extensive quantum
chemical calculations have demonstrated that the per-
sulfoxide adopts a bisected conformation (A in Scheme
5) in which the oxygen-oxygen bond bisects the carbon-
sulfur-carbon bond angle.62 This conformation places the
pendant oxygen in the persulfoxide in an ideal location
to abstract an R-hydrogen to form the second intermedi-
ate. The cation complexed persulfoxide, however, will
likely adopt the extended conformation B rather than C
(Scheme 5) in order to minimize cation-sulfide ring
interactions. Hydrogen abstraction is geometrically pre-
cluded in the thermodynamically more stable intrazeolite
persulfoxide B. In contrast, in the acyclic sulfides greater
An alternative mechanism that invokes an initiation
step involving an intrazeolite-promoted electron transfer
from the sulfide substrate, 11, to singlet oxygen can be
ruled out on thermodynamic grounds. The free energy
for electron-transfer calculated using the Rehm-Weller
equation59 (∆G° ) 23.06 (kcal/mol)/V [E°(11/11+•) - E°-
(O2/O2-•) - e2/ꢀa] - ∆E°°) is very endothermic (29.7 kcal/
mol).60 Oxidation and reduction potentials are likely to
be different in the zeolite than in solution; however, the
magnitude of the change is unlikely to be sufficient to
convert this endothermic process into an exothermic
one.61
(55) Clennan, E. L.; Noe, L. J .; Szneler, E.; Wen, T. J . Am. Chem.
Soc. 1990, 112, 5080-5085.
(56) 15a model system: PhSCH3, k ) 2.3 × 106 M-1 s-1, and
2-methyl-2-pentene, 1.2 × 106 M-1 s-1. 15b model system: PhSCH3
and 2,3-dimethyl-2-butene, k ) 7.0 × 106 M-1 s-1. 15c model system:
pClC6H4SCH3, k ) 1.0 × 106 M-1 s-1, and 2,3-dimethyl-2-butene. 15f
model system: diethyl sulfide, 2 × 107 M-1 s-1, and 2,3-dimethyl-2-
butene. Manring, L. E.; Kanner, R. C.; Foote, C. S. J . Am. Chem. Soc.
1983, 105, 4707-4710.
(57) Monroe, B. M. Photochem. Photobiol. 1979, 29, 761-764.
(58) Wilkinson, F. Pure Appl. Chem. 1997, 69, 851-856.
(59) Rehm, D.; Weller, A. Isr. J . Chem. 1970, 8, 259-271.
(60) The terms in the Rhem-Weller equation E°(11/11+•), E°(O2/
O2-•), e2/ꢀa, and ∆E°° represent the oxidation potential of 11 (1.483 V
versus SCE in CH3CN; Wendt, H. Angew. Chem., Int. Ed. Engl. 1982,
21, 256-270), the reduction potential of triplet oxygen (-0.82 V vs
SCE in CH3CN; Clennan, E. L.; Noe, L. J .; Szneler, E.; Wen, T. J . Am.
Chem. Soc. 1990, 112, 5080-5085), the work term (0.04 V in CH3CN)
for bringing two ions to the encounter distance a in a solvent of
dielectric constant ꢀ, and the singlet-singlet excitation energy of
oxygen (22.5 kcal/mol). The work term, e2/ꢀa, was determined from
the dielectric constant for acetonitrile and the encounter distance, a
) 5 Å, calculated from the radii for singlet oxygen and diphenyl sulfide.
(61) Yoon, K. B.; Kochi, J . K. J . Am. Chem. Soc. 1988, 110, 6586-
6588.
(62) Greer, A.; Chen, M.-F.; J ensen, F.; Clennan, E. L. J . Am. Chem.
Soc. 1997, 119, 4380-4387.
9376 J . Org. Chem., Vol. 67, No. 26, 2002

Zeolite-Controlled Photooxidations of Sulfides
sulfone, 10SO2;74 diphenyl sulfone, 11-SO2;69 benzyl phenyl
sulfide, 12;71 benzyl phenyl sulfoxide, 12-SO;71 benzyl phenyl
sulfone, 12-SO2;71 2-phenylethyl phenyl sulfide, 13;75 2-phe-
nylethyl phenyl sulfoxide, 13-SO;76 2-phenylethyl phenyl sul-
fone, 13-SO2;68 2-chloroethyl ethyl sulfide, 14;19 2-chloroethyl
ethyl sulfoxide, 14-SO;19 and ethyl vinyl sulfoxide19 are all
known and their structures are consistent with their published
physical data. NaY was purchased commercially and converted
to NaMBY as described.
Diffuse reflectance spectra of MB-zeolites were obtained on
a UV/vis/NIR spectrometer. Proton NMR spectra were ob-
tained in CDCl3 at 400.13 MHz and are referenced to TMS.
An HP-5 [30 m × 0.25 mm × 0.25 µm (length × inside
diameter × film thickness)] capillary column and a 5%
diphenyl-95% dimethyl polysiloxane (30 m × 0.32 mm × 1.0
µm) fused silica column were used.
Zeolite P h otolysis. The standard intrazeolite photooxida-
tion experiment was conducted by addition of 0.3 g of dried
NaMBY to 5 mL of hexane containing the substrate. This
mixture was then stirred for 15 min and saturated with dry
(CaCl2) oxygen gas. These samples were then irradiated for 1
h with continuous oxygen bubbling with a 600 W tungsten-
halogen lamp through a 12 M NaNO2 400-nm cutoff filter.
After irradiation the hexane-zeolite slurry is centrifuged and
the solvent decanted. The products were isolated by continuous
extraction with tetrahydrofuran for 3 h or overnight using
acetonitrile or methanol and analyzed by GC or in some cases
by proton NMR.
Tr a p p in g Stu d ies. The trapping studies were done using
the standard photolysis protocol (vide supra) in hexane slurries
using various concentrations of thiolane sulfoxide. Thiolane
sulfoxide was 100% incorporated into the zeolite at all con-
centrations used in these studies. Using 5 mL of a 0.1 M
solution of diphenyl sulfide, 11, the intrazeolite concentration
was 1.93 M, representing an average incorporation of 81% of
the sulfide from solution. Using 5 mL of a 0.05 M solution of
diphenyl sulfide, 11, the intrazeolite concentration was 1.18
M, representing an average incorporation of 99% of the sulfide
from solution. The concentrations were calculated using a void
volume of 0.21 mL in 300 mg of the zeolite. The void volume
was calculated using a value of 1.5 × 104 Å3 as the volume of
the unit cell and 12 700 as the molar mass of the NaY unit
cell. This calculation is an approximation, since solvent does
not have access into the sodalite cages and the framework also
takes up volume in the unit cell. An alternative, and perhaps
better, means of calculating the void volume is to use the
volume of the supercage (827 Å3) and multiplying by eight
supercages per unit cell. This gives a void volume of 0.095 mL
in 300 mg of the zeolite and a corresponding higher concentra-
tion of thiolane sulfoxide and Ph2S in the interior of the zeolite.
This gives slopes in Figure 7 that are 2.21 times larger (0.21
mL/0.095 mL), but since the concentrations of 11 are also 2.21
conformational flexibility in the cation-complexed per-
sulfoxide still allows adoption of the bisected conforma-
tion without severe destabilizing steric interactions.
Con clu sion
We have examined the photooxidations of a series of
sulfides in the interior of zeolite Y. Dramatic changes in
reaction efficiency and product ratios were observed.
Unfortunately, direct spectroscopic observation of in-
trazeolite singlet oxygen and other transient intermedi-
ates failed because of yet unsolved technical problems
in this heterogeneous environment. Nevertheless, very
revealing kinetic studies demonstrate that the potential
energy surfaces for sulfide photooxidations are perturbed
but not completely altered by moving the reaction from
solution into the highly charged intrazeolite environment.
The interstitial cations provide stabilization for the
pivotal persulfoxide intermediates. As a consequence of
this “complexation/stabilization”, the negative charge is
diminished on the peroxy linkage, direct attack by the
sulfide substrate is promoted, and decomposition via a
physical quenching channel is inhibited. On a practical
level, in comparison to their solution counterparts, the
reactions are more rapid in the zeolite and the product
ratios are dramatically more sensitive to traces of water.
In these photooxidations the zeolite appears to be
acting as an authentic microreaction vessel. No direct
(covalent) participation of the zeolite framework has been
detected. However, participation of the zeolite framework
has been suggested in other cases. We anticipate that as
the scope of intrazeolite mechanistic studies expand, the
factors that control a more active participation of the
zeolite supercage will be delineated.
Exp er im en ta l Section
Compounds thiolane sulfoxide, 4-SO; thiolane sulfone,
4-SO2; thiane, 5; diphenyl disulfide; thioanisole, 8; phenyl
methyl sulfoxide, 8-SO; diphenyl sulfide, 11; and diphenyl
sulfoxide, 11SO, were all obtained commercially and used
without further purification. Thiolane, 4, was obtained com-
mericially and used without further purification. The com-
pounds 4-butanal disulfide;22 thiane sulfoxide, 5-SO;63 thiane
sulfone, 5-SO2;63 hexamethylene sulfide, 6;63 hexamethylene
sulfoxide, 6-SO;63 hexamethylene sulfone, 6SO2;63 6-hexanal
disulfide;64 phenyl allyl sulfide, 7;65 phenyl allyl sulfoxide,
7-SO;66 phenyl allyl sulfone, 7-SO2;67 phenyl methyl sulfone,
8-SO2;68 phenyl ethyl sulfide, 9;69 phenyl ethyl sulfoxide,
9-SO;70 phenyl ethyl sulfone, 9-SO2;71 phenyl isopropyl sulfide,
10;72 phenyl isopropyl sulfoxide, 10-SO;73 phenyl isopropyl
times larger, this method gives the same values of kS/kSO
.
P r od u ct Ra tios a s a F u n ction of Su bstr a te Loa d in g
Levels. The plots in Figure 3 were generated by adding 300
mg of dry NaMBY ( S
) 0.01) to a hexane solution
MB
containing the sulfide substrate. The percent absorptions were
determined by monitoring the concentration of the substrate
in hexane by gas chromatography. The maximum absorption
value was attained almost immediately upon addition of the
zeolite to the hexane solution. The sulfoxide/sulfone ratios
changed slightly as a function of percent conversion. The data
in Figure 3 were generated at high conversions (27.9-98.7%;
Table S1 in Supporting Information). In general, longer
(63) Block, E.; Bazzi, A. A.; Lambert, J . B.; Wharry, S. M.; Andersen,
K. K.; Dittmer, D. C.; Patwardhan, B. H.; Smith, D. J . H. J . Org. Chem.
1980, 45, 4807-4810.
(64) Cox, J . M.; Owen, L. N. J . Chem. Soc. C 1967, 1130-1134.
(65) Furuta, K.; Ikeda, Y.; Meguriya, N.; Ikeda, N.; Yamamoto, H.
Bull. Chem. Soc. J pn. 1984, 57, 2781-2790.
(66) Ali, M. H.; Stevens, W. C. Synthesis 1997, 764-768.
(67) Schreiber, K. C. Anal. Chem. 1949, 21, 1168-1172.
(68) Pine, S. H.; Shen, G.; Bautista, J .; Sutton J r., C.; Yamada, W.;
Apodaca, L. J . Org. Chem. 1990, 55, 2234-2237.
(69) Pouchert, C. J .; Behnke, J ., Eds. The Aldrich Library of 13C
and 1H FT NMR Spectra; 1st ed.; Aldrich Chemical Co.: Milwaukee,
WI, 1993.
(70) Ali, M. H.; Bohnert, G. J . Synthesis 1998, 1238-1240.
(71) Hepworth, H.; Clapham, H. W. J . Chem. Soc. 1921, 119, 1188-
1198.
(73) Barbieri, G.; Cinquini, M.; Colonna, S.; Montanari, F. J . Chem.
Soc. C. 1968, 659.
(74) Lamothe, M.; Anderson, M. B.; Fuchs, P. L. Synth. Commun.
1991, 21, 1675-1693.
(75) Baciocchi, E.; Del Giacco, T.; Ferrero, M. L.; Rol, C.; Sebastiani,
G. V. J . Org. Chem. 1997, 62, 4015-4017.
(72) Iptieff, N.; Pines, H.; Friedman, B. S. J . Am. Chem. Soc. 1938,
60, 2731-2734.
(76) Mashraqui, S. H.; Mudaliar, C. D.; Karnik, M. A. Synth.
Commun. 1998, 28, 939-943.
J . Org. Chem, Vol. 67, No. 26, 2002 9377

Clennan et al.
irradiation times were needed at the higher loading levels to
get comparable percent conversions. For experiments in which
irradiation times of Ph2S (11) and NaMBY slurries were kept
constant (all 1 h), the percent conversions were very low at
high loading levels and led to a lower quality linear plot.
Absor p tion Stu d ies. The absorption studies depicted in
Figure 2 were conducted by adding 2.5 mL of a 0.04 M hexane
solution of 5 to 0.15 g of dried NaY. The amount of 5 remaining
in solution was periodically analyzed by capillary column gas
chromatography. The amount of 5 absorbed by the zeolite was
then calculated using the following equation
100 supercages). This hexane slurry was saturated with
oxygen and stirred for 15 min and then irradiated under
continuous oxygen agitation with a 600-W tungsten lamp
thorough 1 cm of a 12 M NaNO2 filter solution. The zeolite
powder was then filtered and placed in a Soxhlet cup and
continuously extracted overnight with chloroform. The zeolite
was then mildly digested with 5 mL of 10% HCl and the
mixture extracted with four 10-mL portions of chloroform. The
combined organic extracts were then washed with aqueous
Na2CO3 and dried with MgSO4, and the solvent was removed
under reduced pressure. The reaction mixtures were then
characterized by proton NMR. 1H NMR (CDCl3): 16a -OOH δ
5.03 (bs, 2H), 4.49 (dd, J ) 7.9, 7.8 Hz 1H), 3.0 (m, 2H), 1.7-
2.0 (m, 2H), 1.68 (s, 3H); 17a -OOH δ 5.70 (dt, J ) 15.7, 7.1
Hz, 1H), 5.50 (bd, J ) 15.7 Hz, 1H), 3.53 (dd, J ) 7.1, 1.2 Hz,
2H), 1.21 (2, 6H); 16a -OH δ 4.97 (s, 1H), 4.88 (s, 1H), 4.22 (t,
J ) 6.1 Hz, 1H), 3.01 (m, 2H), 1.86 (q, J ) 7.0 Hz, 2H), 1.65
(s, 3H); 17a -OH δ 5.62 (d, J ) 15.4 Hz, 1H), 5.69 (dt, J ) 15.4,
6.5 Hz, 1H), 3.51 (d, J ) 6.5 Hz, 2H), 1.20 (s, 6H); 16b-OOH
δ 5.02 (bs, 1H), 4.98 (bs, 1H), 2.95 (m, 2H), 2.01 (m, 2H), 1.38
(s, 3H); 18b-OOH δ 5.26 (s, 1H), 5.08 (bs, 1H), 3.18 (t, J ) 7.3
Hz, 2H), 2.52 (t, J ) 7.1 Hz, 2H), 1.38 (s, 6H); 16b-OH δ 5.04
(bs, 1H), 4.90 (bs, 1H), 2.8-3.0 (m, 2H), 1.96 (m, 2H), 1.72 (bs,
3H), 1.32 (s, 3H); 18b-OH δ 5.18 (s, 1H), 4.87 (bs, 1H), 3.10 (t,
J ) 8.0 Hz, 2H), 2.45 (t, J ) 8.0 Hz, 2H), 1.34 (s, 3H).
T(mmol/g) ) {[1 - (At/Ais)/(Ao/Ais)]CoVo}/Wz
where T ) the amount of 5 absorbed per gram of zeolite, At )
the GC area of 5 in hexane at time t, Ais ) the GC area of
internal standard (Ph3CH), Ao ) the GC area of sulfide at t )
0, Wz ) the weight of the zeolite in grams, Co ) the concentra-
tion of 5 in the hexane at t ) 0 in mmol/mL, and Vo ) the
volume of the hexane (mL).
The loading level at any time t was calculated using the
following equation
S
) [(g5)/(MW5)]/(Msupercage)
-1(Wz)
5
where, S 5 ) the molecules of 5 per supercage, g5 ) the grams
of 5 absorbed into zeolite, MW5 ) the molecular weight of 5,
Msupercage ) the grams of dry zeolite per moles of supercage
(calculated for a unit cell composition of Na56(AlO2)56(SiO2)136
and using eight supercages per unit cell to give 1595.32 g/mol
supercage).
Ack n ow led gm en t. We thank the National Science
Foundation for their generous support of this research.
Su p p or tin g In for m a tion Ava ila ble: Data for â-carotene
and 2,3-dimethyl-2-butene quenching of Ph2S oxidation, prod-
uct ratios as a function of loading level, and kinetic data for
15a , 15b, and 15f. This material is available free of charge
via the Internet at http://pubs.acs.org.
In tr a zeolite P h otooxid a tion of 15a -f. The reaction
mixtures were prepared by taking 5 mL of a hexane solution,
either 0.05 or 0.1 M in 15a -f, and adding 0.3 g of zeolite Y
(occupancy S
) 0.01; one molecule of methylene blue per
J O0261808
MB
9378 J . Org. Chem., Vol. 67, No. 26, 2002