Gauging the significance of atomic oxygen [O(3P] in sulfoxide photochemistry. A method for hydrocarbon oxidation

Article abstract of DOI:10.1021/jo0266487

A liquid-phase photolysis of 1,2-benzodiphenylene sulfoxide, 1, and dibenzothiophene sulfoxide, 2, was used to generate atomic oxygen [O(³P)] or an equivalent active oxygen species. The reaction for sulfoxide photodeoxygenation was similar to a microwave discharge method for generating O(³P) atoms in the condensed phase (Zadok, E.; Rubinraut, S.; Mazur, Y. J. Org. Chem. 1987, 52, 385-90). Sulfoxide photodeoxygenation is a potentially clean method for O(³P) production compared to the microwave discharge method. With Argon purging of the sulfoxide sample before photolysis, the method can preclude a secondary oxidation process involving molecular oxygen. Our study focused on the results of oxidation products in the reaction of styrene, 3, and on the dependence of substrates that provided an opportunity to vary the electronic and steric effects. The sulfoxide photochemistry is rationalized with the primary formation of O(³P) in which a charge-transfer interaction between O(³P) and substrate precedes oxidation. Functionalization of hydrocarbons takes place under mild photolysis conditions of 1 and 2, which leads to an interesting possibility for the synthetic use of atomic oxygen, O(³P). Alkanes give principally alcohols. Alkenes give principally epoxides and ketones. For comparison, hydroxyl radicals are more reactive and less selective toward hydrocarbons compared to O(³P) atoms. On the other hand, O(³P) atoms balance reactivity and selectivity and involve the oxidation of inert alkanes typically inaccessible to peracid, dioxirane, ozone, and singlet molecular oxygen chemistry. The findings from this study may be useful to those interested in generating high-value oxygenated compounds from readily available petroleum components.

Full text of DOI:10.1021/jo0266487

3
Ga u gin g th e Sign ifica n ce of Atom ic Oxygen [O( P )] in Su lfoxid e
P h otoch em istr y. A Meth od for Hyd r oca r bon Oxid a tion
Keith B. Thomas and Alexander Greer*
Department of Chemistry, Graduate School and University Center & The City University of New York
(CUNY), Brooklyn College, Brooklyn, New York 11210
agreer@brooklyn.cuny.edu
Received November 2, 2002
A liquid-phase photolysis of 1,2-benzodiphenylene sulfoxide, 1, and dibenzothiophene sulfoxide, 2,
3
was used to generate atomic oxygen [O( P)] or an equivalent active oxygen species. The reaction
3
for sulfoxide photodeoxygenation was similar to a microwave discharge method for generating O( P)
atoms in the condensed phase (Zadok, E.; Rubinraut, S.; Mazur, Y. J . Org. Chem. 1987, 52, 385-
3
9
0). Sulfoxide photodeoxygenation is a potentially clean method for O( P) production compared to
the microwave discharge method. With Argon purging of the sulfoxide sample before photolysis,
the method can preclude a secondary oxidation process involving molecular oxygen. Our study
focused on the results of oxidation products in the reaction of styrene, 3, and on the dependence of
substrates that provided an opportunity to vary the electronic and steric effects. The sulfoxide
3
photochemistry is rationalized with the primary formation of O( P) in which a charge-transfer
3
interaction between O( P) and substrate precedes oxidation. Functionalization of hydrocarbons takes
place under mild photolysis conditions of 1 and 2, which leads to an interesting possibility for the
3
synthetic use of atomic oxygen, O( P). Alkanes give principally alcohols. Alkenes give principally
epoxides and ketones. For comparison, hydroxyl radicals are more reactive and less selective toward
3
3
hydrocarbons compared to O( P) atoms. On the other hand, O( P) atoms balance reactivity and
selectivity and involve the oxidation of inert alkanes typically inaccessible to peracid, dioxirane,
ozone, and singlet molecular oxygen chemistry. The findings from this study may be useful to those
interested in generating high-value oxygenated compounds from readily available petroleum
components.
In tr od u ction
SCHEME 1
In the presence of light, sulfoxides 1 and 2 can react
with alkanes to afford alcohols.1 Oxygen is introduced
in a selective manner in 2-methylbutane (Scheme 1A).
The mechanism is postulated as the loss of oxygen from
-3
3
sulfoxide to yield atomic oxygen [O( P)] or an equivalent
active oxygen species capable of oxygen transfer to the
hydrocarbon.1 Sulfoxide photodeoxygenation methods
may enable one to address questions about atomic oxygen
in the condensed phase. Building a context for the
chemistry would be desirable. Atomic oxygen is a selec-
-3
3
5,8-11
previous work on condensed phase O( P) atoms,
and
organic
1
2-14
despite its potential as an oxidizing agent,
chemists do not often view O( P) atoms as a reagent for
synthetic transformations. Little is known of O( P) atoms
in the condensed phase compared to the gas phase. The
reaction may be complicated by solvent effects. Few
studies have brought together results on condensed phase
O( P) atoms, which address questions about the influence
of solvent. Complexes or ions are commonly encountered
3
3
3
tive agent in the ground state, O( P), and a nonselective
1
4-6
agent in the excited state, O( D) (Scheme 1B).
The selective oxidation of hydrocarbon bonds is a
7
seminal and longstanding problem in organic chemistry.
3
3
A reagent, such as atomic O( P), that possesses the ability
to transform hydrocarbon bonds to oxygenated functional
groups would have useful applications. Yet despite the
(
8) Zadok, E.; Amar, D.; Mazur, Y. J . Am. Chem. Soc. 1980, 102,
6
369-6370.
(1) Lucien, E.; Greer, A. J . Org. Chem. 2001, 66, 4576-4579.
(2) Wan, Z.; J enks, W. S. J . Am. Chem. Soc. 1995, 117, 2667-2668.
(3) Gregory, D. D.; Wan, Z.; J enks, W. S. J . Am. Chem. Soc. 1997,
(9) Zadok, E.; Rubinraut, S.; Mazur, Y. J . Org. Chem. 1987, 52, 385-
390.
(10) Rudich, Y.; Lifson, S.; Naaman, R. J . Am. Chem. Soc. 1991,
1
19, 944-102.
113, 7077-7078.
(
(
(
4) Cvetanovic, R. J . J . Phys. Chem. Ref. Data 1987, 16, 261-326.
5) Bucher, G.; Scaiano, J . C. J . Phys. Chem. 1994, 98, 12471-12473.
6) Kajimoto, O.; Yamasaki, H.; Fueno, T. Chem. Lett. 1977, 349-
(11) Rudich, Y.; Hurwitz, Y.; Lifson, S.; Naaman, R. J . Chem. Phys.
1993, 98, 2936-2940.
(12) Wang, C.; Weiss, R. G. Macromolecules 1999, 32, 7032-7039.
(13) Paz, Y.; Trakhtenberg, S.; Naaman, R. J . Am. Chem. Soc. 1994,
116, 10344-10345.
(14) Paz, Y.; Trakhtenberg, S.; Naaman, R. J . Phys. Chem. 1994,
98, 13517-13523.
3
52.
(
7) A complete listing is not possible. Two recent references in-
clude: J ohnson, J . A.; Sames, D. J . Am. Chem. Soc. 2000, 122, 6321-
322. Barton, D. H. R. Tetrahedron 1998, 54, 5805-5817.
6
1
0.1021/jo0266487 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/06/2003
1
886
J . Org. Chem. 2003, 68, 1886-1891

Significance of Atomic Oxygen in Sulfoxide Photochemistry
in solution studies of sulfur,15 chlorine,16,17 and fluorine
atoms.18 One would expect that complexation and ioniza-
tubes. Argon was flushed through the reaction mixtures prior
to irradiation. The reagents were checked by GC or GCMS to
establish that no oxidation products were present before
photolysis. A typical experiment was conducted using 385 nm
3
tion pathways of O( P) may differ in solution compared
to the gas phase.19,20
light in anhydrous acetonitrile, which contained 5-10 mM 1,
We sought to establish the mechanistic criteria to learn
-4
0
.05 M styrene 3, and 5 × 10 M biphenyl as an internal
3
of a possible sulfoxide method to generate O( P) atoms
standard. The photodeoxygenation studies were conducted in
an identical fashion for 2 with the exception of using 350-nm
light. Product distributions are the average of 2-4 runs. Both
epoxide and rearranged byproducts were monitored. The
percent conversion of the reaction components was kept below
or an equivalent active oxygen species. Microwave dis-
3
charge methods generate O( P), but often in the presence
1
1
8,21,22
2 3
of other reactive species, such as O( D), O , and O .
In developing the research problem on sulfoxide photo-
chemistry, it was important to choose a system whose
1
6 4 2
0% since some products, in particular (p-X-C H ) SO, can be
further oxidized. Relative rates were determined by competi-
tion of substrates with benzene, styrene, and/or diphenyl
sulfide, where the appearance of the oxidation products were
followed. Under the photolysis conditions there is negligible
rearrangement of the product styrene oxide. Control experi-
ments demonstrate that styrene is inert to 1 and 2 in the
absence of light. Acetonitrile solvent was chosen because of
3
solution phase O( P) reaction was well-studied. One such
reaction previously studied in the condensed phase is that
3
8,9
between O( P) and styrene 3. Answers to questions
about the formation of rearranged styrene byproducts,
discussed in this study, have yielded encouraging lessons
3
about the viability of sulfoxides 1,2 to produce O( P)
3
5,24
its low relative reactivity toward O( P) atoms
and for the
atoms. Mechanistic information was also provided by
adding substituents to the structure of the substrate to
vary the electronic and steric effects.
reason that charge-transfer processes are often investigated
in this solvent.25
Th eor et ica l Met h od s. Density functional theoretical
(
DFT) and PM3 semiempirical calculations were performed
2
6,27
Exp er im en ta l Section
with the Gaussian-94 or Gaussian-98 program packages.
Geometries were optimized with the B3LYP method along with
Ch em ica ls a n d In str u m en ta tion . Reagents and solvents
were obtained commercially [1,2-benzodiphenylene sulfide,
dibenzothiophene sulfide, benzene (anhydrous), phenol, bi-
phenyl, diphenyl sulfoxide, acetonitrile (anhydrous), and ac-
etonitrile-d ] and were used as received. Diphenyl sulfide was
3
distilled under reduced pressure (bp 124 °C/6 mmHg). Com-
the 6-31G(d) basis set, or were optimized with the PM3
28,29
method. Polarized continuum model (PCM)
and self-
consistent reaction field (SCRF) single-point calculations at
the B3LYP/6-31(d) level were performed on the B3LYP/6-31G-
(d) or PM3 derived stationary points to model solvent effects.
A dielectric constant of 37.5 was used to simulate the solvent
acetonitrile.
pounds 1 and 2 were prepared either by using a literature
2
3
method or by reaction with m-chloroperoxybenzoic acid.
Samples were stored in the dark at 25 °C and replaced every
Resu lts a n d Discu ssion
3
-6 weeks. Gas chromatographic data were acquired on one
of two gas chromatographs, a Hewlett-Packard GC/MS instru-
ment consisting of a 5890 series GC and a 5988A series mass
selective detector, or a Shimadzu-17A auto-sampler capillary
gas chromatograph equipped with a flame ionization detector.
Additional measurements were carried out on a Bruker (250
MHz) FT-NMR spectrometer. Photolyses were conducted with
a 75-W Xenon Model L-201 Arc lamp (Photon Technology
International) focused on a tunable monochromator to obtain
monochromatic light over the range 280-400 nm (linear
dispersion equal to ca. (12 nm).
Da ta An a lysis. Concentrations of 1,2-benzodiphenylene
sulfoxide, 1,2-benzodiphenylene sulfide, dibenzothiophene sul-
foxide, dibenzothiophene sulfide, phenylacetaldehyde, ace-
tophenone, benzaldehyde, phenol, di-tert-butyl sulfoxide, di-
tert-butyl sulfide, diisopropyl sulfoxide, diisopropyl sulfide,
This study concentrated on two topics regarding the
photolysis of 1 and 2. First, on the results of the oxidation
of styrene 3. Second, on the dependence of substrates that
provided an opportunity to vary the electronic and steric
effects. The experiments and computations provided
3
evidence for the formation of O( P) from 1,2-photodeoxy-
genation in which a charge-transfer interaction between
3
O( P) and substrate precedes oxidation.
Oxid a tion of Styr en e. Styrene 3 forms five products
3
from a liquid-phase reaction of O( P) produced by micro-
(
24) Mielke, Z.; Hawkins, M.; Andrews, L. J . Phys. Chem. 1989, 93,
5
58-564.
(25) Kavarnos, G. J .; Turro, N. J . Chem. Rev. 1986, 86, 401-449.
(26) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
diethyl sulfoxide, para-substituted styrenes (p-X-C
CH ), para-substituted styrene oxides, and para-substituted
4 2
arylsulfoxides (p-X-C H ) SO, where X ) OMe, Me, H, Cl, were
6 4
H CHd
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. 1994,
Gaussian 94; Gaussian, Inc.: Pittsburgh, PA.
2
6
detected by reference to calibration curves constructed from
authentic samples. Excellent linear correlations were observed
2
in the calibration curves (r g 0.999). The study was conducted
at room temperature. Sulfoxide photodeoxygenation reactions
were carried out with 1.5-mL airtight GC vials or 5-mm NMR
(27) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
J ohnson, B.; Chen, W.; Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J . A. 1998, Gaussian 98; Gaussian,
Inc.: Pittsburgh, PA.
(
15) Nau, W. M.; Bucher, G.; Scaiano, J . C. J . Am. Chem. Soc. 1997,
1
19, 1961-1970.
(
16) Chateauneuf, J . E. J . Am. Chem. Soc. 1990, 112, 442-444.
(17) Bunce, N. J .; Ingold, K. U.; Landers, J . P.; Lusztyk, J .; Scaiano,
J . C. J . Am. Chem. Soc. 1985, 107, 5464-5472.
18) Bucher, G.; Scaiano, J . C. J . Am. Chem. Soc. 1994, 116, 10076-
0079.
19) Hurwitz, Y.; Stern, P. S.; Naaman, R.; McCoy, A. B. J . Chem.
Phys. 1997, 106, 2627-2633.
20) McCoy, A. B.; Lufaso, M. W.; Veneziani, M.; Atrill, S.; Naaman,
(
1
(
(
R. J . Chem. Phys. 1998, 108, 9651-9657.
(
(
985, 50, 2647-2649.
(
21) Zadok, E.; Mazur, Y. J . Org. Chem. 1982, 47, 2223-2225.
22) Zadok, E.; Rubinraut, S.; Frolow, F.; Mazur, Y. J . Org. Chem.
(28) Miertus, S.; Scrocco, E.; Tomasi, J . J . Chem. Phys. 1981, 55,
117-129.
(29) Miertus, S.; Tomasi, J . J . Chem. Phys. 1982, 65, 239-245.
1
23) Takata, T.; Ando, W. Tetrahedron Lett. 1983, 24, 3631-3634.
J . Org. Chem, Vol. 68, No. 5, 2003 1887

Thomas and Greer
SCHEME 2
involving 6 and 7. Formaldehyde (8) was not detected
with our GC and GCMS analyses.
A similar styrene oxidation pattern is observed with
3
the microwave discharge method of generating O( P)
atoms.8 The N
,9
/N
styrene gave epoxide 4 and phenylacetaldehyde 5 in a
2:31 ratio along with the byproducts acetophenone 6 and
3
2
2
O discharge reaction of O( P) with
6
benzaldehyde 7 (entry 3, Table 1). The epoxide 4 poten-
tially forms in a concerted or a stepwise process (Scheme
2
I,II). The yield of 6 and 7 increased when an O
2
dis-
charge was used, suggesting O contributes to their for-
2
mation (compare entries 3-5). Acetophenone 6 and benz-
aldehyde 7 byproducts were suggested to result from a
3
3
-step mechanism involving oxidative attack of O( P)
TABLE 1. Com p a r ison of P r od u ct Distr ibu tion s (%) for
Styr en e Oxid a tion in Su lfoxid e P h otod eoxygen a tion
Rela tive to Micr ow a ve Disch a r ge Meth od s (MDM) th a t
with styrene followed by addition of O
2
, to yield B or C
8
(Scheme 2II). A dependence on O
2
concentration is con-
sistent with promoting its attack on an initially formed
3
Gen er a te O( P ) Atom s
8
diradical A. In addition, a reaction may proceed where
3
5
O( P) reacts with O
Scheme 2III),30 then ozonide C, and ultimately 6-8.
A previously studied system provides a possible context
2
at diffusion control to yield ozone
(
3
8,9
for the data in Table 1. J enks and Mazur observed the
randomization of epoxide stereochemistry, which sug-
3
gested a possible stepwise reaction of O( P) with the
double bond of â-methyl styrene (9 and 10, Scheme 3).
3
The photolysis of 2 yielded an epoxide ratio similar to
8
,9
2 2
that obtained by using microwave discharge (N /N O)
a
Relative yields determined by GC or GCMS. b Formaldehyde
3
c
to produce O( P) in the condensed phase (Scheme 3A,B).
Unfortunately, data are unavailable in the J enks and
8
1
was not detected in the GC and GCMS analyses. Photolysis of
(10 mM) was carried out in the presence of 385-nm light in
MeCN. Photolysis of 2 (10 mM) was carried out in the presence of
50-nm light in MeCN. Product distributions are averaged since
identical results are observed, to within experimental error ((4%),
when 1 and 2 are used as the oxygen source. Product distribu-
3
Mazur studies8 to compare possible similarities between
ketone products resulting from 1,2-H shifts, alcohol
products, or products arising from an oxidative cleavage.
A resulting triplet diradical intermediate with rotation
about the C-C bond was suggested to explain the mix-
ture of cis and trans epoxides observed.3 The isomer-
ization efficiency indicates that ring-closure of the diradi-
cal intermediate must be rapid since isomerization is
,9
3
d
e
tions are the average of 2-4 runs. Argon was flushed through
the reaction mixture for 15-30 min prior to irradiation. f Argon
,8,9
was flushed through the reaction mixture for 30 s to 5 min prior
g
3
to irradiation. The microwave discharge methods generate O( P)
in the presence of other active oxygen species, such as O( D), O2,
O2, and O3.
1
1
3
3
incomplete or else there is a competitive reaction with
concerted addition of O( P) (Scheme 3B). Ab initio UHF
3
31
wave discharge and the difference in regioselectivity is
pronounced (4-8, Scheme 2).8,9 The O( P) styrene oxida-
tion pattern is potentially useful since rearrangement
effects can be compared with those arising from the active
oxygen species produced in the photolysis of 1 and 2. The
results of the photooxidation reaction of 1 or 2 with
styrene are shown in Table 1. Styrene partially rear-
ranged during the oxidation. In the presence of 385-nm
light and styrene (0.05 M) in argon-saturated acetonitrile
at room temperature, 1 (10 mM) deoxygenated to give
styrene oxide (4), phenylacetaldehyde (5), and 1,2-ben-
zodiphenylene sulfide. The chemistry is identical to
within experimental error ((4%) when dibenzothiophene
sulfoxide 2 is used as the oxygen source along with 350-
nm light. The mixture of epoxide 4 to phenylacetaldehyde
calculations predicted a low ∼2 kcal/mol barrier to rota-
3
3
tion about the C-C bond, where the (σπ) state can
1
undergo surface crossing to the (σσ) state to ring close
3
2,33
to yield the epoxide.
Surface-crossing studies have
1
suggested that O( D) intercedes on the path to products
3
in the reaction of O( P) with cyclohexane clusters in the
liquid phase.10 Complexing reactions of O( P) have been
3
3
reported. A triplet CHCl - -O complex has been charac-
3
terized in the atomic O( P) oxidation of chloroform in a
matrix infrared study.34
Since the oxidation reactions of 3, 9, and 10 are not
stereospecific this suggests that radicals or even radical
ions may be involved.35 The 1,2-H shift data on pheny-
lacetaldehyde in Table 1 are reminiscent of an electron-
transfer reaction observed between high-valent iron
5
is 58:42 and the yield is 68% (entry 1, Table 1). In the
photodeoxygenation of 1 and 2 the formation of pheny-
lacetaldehyde 5 is competitive with epoxide 4 formation,
but not with acetophenone (6) nor benzaldehyde (7).
However, 6 and 7 can be formed when Ar purging of the
sample is incomplete and molecular oxygen O
in solution (entry 2, Table 1). We find that when the
acetonitrile medium contains dissolved O , 1 and 2 react
photochemically with styrene through a mechanism
(30) Klaning, U. K.; Sehested, K.; Wolff, T. J . Chem. Soc., Faraday
Trans. 1 1984, 80, 2969-2979.
(
31) Cvetanovic, R. J . Adv. Photochem. 1963, 1, 115-182.
(32) Dupuis, M.; Wendoloski, J . J .; Takada, T.; Lester, W. A., J r. J .
Chem. Phys. 1982, 76, 481
(
33) Yamaguchi, K.; Yabushita, S.; Fueno, T.; Kato, S.; Morokuma,
K.; Iwata, S. Chem. Phys. Lett. 1980, 71, 563-568.
34) Schriver-Mazzuoli, L.; Schriver, A.; Hannachi, Y. J . Phys. Chem.
998, 102, 10221-10229.
35) Kim, C.; Traylor, T. G.; Perrin, C. L. J . Am. Chem. Soc. 1998,
120, 9513-9516.
2
remains
(
1
2
(
1
888 J . Org. Chem., Vol. 68, No. 5, 2003

Significance of Atomic Oxygen in Sulfoxide Photochemistry
SCHEME 3
3
TABLE 2. Rela tive Ra tes for O( P ) or Su lfoxid e P h otooxygen a tion of Su bstr a tes a lon g w ith Liter a tu r e Ha lf-Wa ve a n d
Ion iza tion P oten tia ls a n d DF T Com p u ted Ion iza tion P oten tia ls
c
E1/2 (V)d
entry
substrate
log krel
IPexpta (eV)
IPgas/DFTb (eV)
9.1
IPMeCN/DFT (eV)
6.7
1
2
3
4
5
6
7
8
9
.
.
.
.
.
.
.
.
.
Cl-
2.28 (2.3)
e
f
e
P(OMe)3
2.10
8.8
9.0
e
g
cyclopentene
(p-MeOC6H4)2S
(p-MeC6H4)2S
(C6H5)2S
1.73 (1.8)
f
3.15
7.79
7.83
8.04
8.13
8.15
8.38
8.9
6.8
7.0
7.3
7.4
7.3
7.7
8.0
7.9
4.5
5.0
5.1
5.2
5.1
5.3
5.5
5.5
3.07f
f
2.62
(p-ClC6H4)2S
p-MeOC6H4CHdCH2
p-MeC6H4CHdCH2
C6H5CHdCH2
p-ClC6H4CHdCH2
tBu2S
2.44f
g
1.38
g
1.15
g
1
1
1
1
1
1
1
1
0.
1.
2.
3.
4.
5.
6.
7.
1.03
g
i
0.94
9.1
g
j
3.09
8.13
1.65
1.63
g
j
iPr2S
3.41
8.15
g
j
nBu2S
3.76
8.23
g
j
Et2S
3.87
8.45
1.65
h
j
8.6k
5.6k
Me2S
3.61
8.71
C6H6
0
9.9
a
Experimental adiabatic ionization potentials from ref 52. b Geometries optimized with the PM3 semiempirical method followed by a
single-point calculation at the B3LYP/6-31G(d) and UB3LYP/6-31G(d) levels. c PCM calculations at the B3LYP/6-31G(d) and UB3LYP/
d
6
-31G(d) levels from PM3 optimized minima. A dielectric constant of 37.5 is used to simulate acetonitrile. Half-peak potentials vs SCE
in acetonitrile (ref 45). Value from the irradiation of pyridine N-oxide (ref 5). f Value from the irradiation of 1,2-benzodiphenylene sulfoxide
e
g
1
(ref 1). This work. Relative rates were determined by competition with benzene, styrene, or diphenyl sulfide in the photolysis of 1.
h
i
Value from the gas-phase mercury-sensitized decomposition of N2O (ref 50). IP estimated as 9.1 eV from the IPs of a series of para-
substituted diaryl sulfides. j IPs estimated from the corresponding oxygenated molecules (refs 52 and 53). Geometry optimized at the
k
B3LYP/6-31G(d) level.
porphyrins and styrene,36-38 where styrene rearrange-
3
9
ment occurs via a radical cation. Resolving questions
about the influence of electronics in the sulfoxide pho-
tolysis process seemed like an important next step. The
influence of solvation diminishes the energy required to
form radical cations in styrene and diaryl sulfide trap-
ping agents according to polarized continuum model
UB3LYP calculations (Table 2).
In flu en ce of Rin g Su bstitu ten ts. We have examined
a series of para-substituted styrenes in the photodeoxy-
genation of 1 and 2. The experiments provide evidence
that sulfoxide photolysis underpins the production of
3
O( P), which provides mechanistic insight on the process
that gives rise to hydrocarbon oxidation. Competition
between substituted styrenes for the oxygenating inter-
mediate produced in the photolysis of 1 and 2 yielded a
+
2
linear correlation with σ (r )0.996, Figure 1). The
+
+
magnitude of the reaction constant, F ) -0.48, is
X H
F IGURE 1. Plot for log k /k vs the Hammett-Brown σ for
consistent with trapping of an electrophilic oxidant40 and
para-substituted styrene trapping of an intermediate in the
+
photodeoxygenation of 1 and 2 in acetonitrile [slope (F ) )
suggests a build-up of positive charge on the alkene in
the transition state. The excellent correlation of the rate
-
p
0.48, r2 ) 0.996]. Inset: Plot for log k
X H
/k vs the Hammett
σ
for para-substituted styrene trapping of an intermediate
in the photodeoxygenation of 1 and 2 in acetonitrile [slope (F)
-0.82, r2 ) 0.926].
+
2
data with σ is less satisfactory against σ (r ) 0.926,
p
)
(
36) Groves, J . T.; Myers, R. S. J . Am. Chem. Soc. 1983, 105, 5791-
796.
37) Groves, J . T.; Nemo, T. E. J . Am. Chem. Soc. 1983, 105, 5786-
791.
5
5
Figure 1, Inset). Data sets which correlate with σ+ but
not with σ are typically thought to possess an electron-
deficient site in the transition state, which is in conjuga-
(
p
(
38) Traylor, T. G.; Xu, F. J . Am. Chem. Soc. 1988, 110, 1953-1958.
(39) Traylor, T. G.; Kim, C.; Richards, J . L.; Xu, F.; Perrin, C. L. J .
4
1,42
tion with an electron-donating substituent.
It is
Am. Chem. Soc. 1995, 117, 3468-3474.
40) Cvetanovic, R. J . Can. J . Chem. 1960, 38, 1678-1687.
(
tempting to suggest that through-bond resonance is
J . Org. Chem, Vol. 68, No. 5, 2003 1889

Thomas and Greer
SCHEME 4
favored in the para-substituted styrenes, where C-C
overlap is achieved between the ring and side chain to
yield the resonance contributions shown in Scheme 4A.43
A distinction may be made between conjugated and
nonconjugated transition states in the substrate due to
4
4
relative contributions of resonance by using the rate
data of para-substituted styrenes and diaryl sulfides
(
Table 2). In contrast to the para-substituted styrenes,
the para-substituted diaryl sulfides yielded a linear
correlation with σ values, but was less satisfactory with
p
+
σ . The correlation coefficient of the Hammett plot of (p-
X-C
k
σ (r ) 0.913). We propose that through-bond resonance
is not favored in the substituted diaryl sulfides (Scheme
X H
F IGURE 2. Correlation of log k /k with the adiabatic
ionization potentials for the oxidation of substituted styrene
and diaryl sulfide trapping agents. Oxidation of para-substi-
tuted styrenes (X ) MeO, Me, H, Cl) from the photodeoxy-
6
H
4
)
p
2
S determined from the slope of a plot of log k
is enhanced (r ) 0.968) compared to a plot vs
X
/
2
1
H
+
vs σ
2
genation of 1 and 2 in acetonitrile (A, open squares, slope )
2
-
0.41, r ) 0.954). Oxidation of para-substituted diaryl sulfides
4
2 2
B) since, as with RC dSR ylides, the C-S overlap is
(
X ) MeO, Me, H, Cl) from the photodeoxygenation of 1 and 2
poor. In a similar vain, the behavior of para-substituted
2
in acetonitrile (B, solid squares, slope ) -2.10, r ) 0.999).
Each point represents the average of the product distribution
from four runs. Two of the runs contained 1 as the oxygen
source, the other two contained 2. The experimental error is
diaryl sulfides in the photodeoxygenation of 1 and 2 may
1
be analogous to that of the reaction of O
2
with para-
substituted aryl sulfides.45 In the p-X-C
1
6
5 3 2
H SCH s O
(
4%.
study, Foote and Kacher found that a Hammett plot vs
+
σ gives an excellent correlation, while use of σ gives a
45
p
The oxygen donor properties are similar in the sulfox-
poor correlation.
Little data are available in the literature regarding
1-3
ide 1,2-photodeoxygenation
oxide photodeoxygenation. We have explored reactions
of tetrabutylammonium chloride and cyclopentene in the
photolysis of 1 and 2. The results suggest a remarkably
and in the pyridine N-
5
3
46
Hammett F values for reactions involving O( P) atoms.
1
The Hammett data by itself provides limited evidence for
a discrete O( P) intermediate in the photolysis of 1,2. It
3
similar behavior between our product-based and Sca-
is possible that sulfoxide may act as an oxygen donor in
the excited state. In fact, some sulfoxides are known to
undergo bimolecular photochemical reductions in the
presence of alkoxides in alcoholic solvents.47 Our results
suggest that electronics of the substrate is in part
responsible for the oxygen donor properties of 1 and 2.
Ch a r ge-Tr a n sfer In ter a ction . In the photolysis of
iano’s kinetic-based selectivities5 for chloride ion and
cyclopentene (compare entries 1 and 3, Table 2). How-
ever, this close agreement may be partly fortuitous upon
scrutinizing the data collected for diphenyl sulfide. The
k
rel value for diphenyl sulfide was obtained in benzene.
The concentrations were chosen to reduce the possibility
for sulfoxides to form dimeric structures in a bimolecular
reaction (5-10 mM). The krel value was obtained from
observing a small relative concentration of phenol com-
pared to a high concentration of Ph
precision of only (25%. In contrast, the k
determined within the Hammett series (p-X-C
a precision of (4%. The same situation applied with
styrene. Precision was low on competition of benzene with
styrene, but high with competition between pairs of the
1
and 2, the interaction of the oxygen intermediate as
measured by k(fastest)/k(slowest) is more sensitive to electronic
effects for the diaryl sulfides (5.1) compared to the
styrenes (2.7) (entries 4-11, Table 2). Even though
2
SO, which gave a
X
/k
H
H
value
4 2
) S had
3
knowledge is limited on the O( P) intermediate in solu-
6
tion, one would expect reaction rates to be enhanced with
substrates of higher nucleophilicity or reduced ionization
potentials (IP) (Table 2). In the past, gas-phase studies
3
of O( P) have correlated a dependence with IPs of the
6 4 2
Hammett series, p-X-C H CHdCH .
1
,31,40,46,49,50
5
acceptor molecules.
Bucher and Scaiano sug-
How efficient a charge-transfer or an electron-transfer
process will be in the oxygenation process may be inferred
from the observed slopes of log k vs IP for the para-
substituted styrenes and para-substituted diaryl sulfides
3
gested that O( P) reactions in acetonitrile solution may
yield rates above diffusion-control with the electron-rich
substrates bromide ion and chloride ion.
(
Figure 2). Here we used an analysis similar to that of
(
41) Brown, H. C.; Okamoto, Y. J . Am. Chem. Soc. 1958, 80, 4979-
35
Perrin and co-workers. If a full electron is transferred
at the transition state, and if the energy level of that
electron is raised by 1 eV (23 kcal/mol), then the rate
constant would be expected to increase by a factor of exp-
4
987.
(42) Gawley, R. E. J . Org. Chem. 1981, 46, 4595-4597.
43) Fleming, I. Frontier Orbitals and Organic Chemical Reactions;
(
J ohn Wiley & Sons: Cambridge, UK, 1976.
44) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in
Organic Chemistry; Harper & Row: New York, 1987.
45) Foote, C. S.; Kacher, M. L. Photochem. Photobiol. 1979, 29, 765-
69.
(
1
7
35
(
23/RT), or 1.6 × 10 at 20 °C. Figure 2 shows that a
(
plot of log k vs IP is linear with a slope that is dependent
on the IP of the substrate. The sensitivity of the slope
with IP of the substrate is consistent with an electron
transfer at the transition state, only to the extent of 2%
7
(
(
46) Cvetanovic, R. J . J . Chem. Phys. 1959, 30, 19-26.
47) Cubbage, J . W.; Tetzlaff, T. A.; Groundwater, H.; McCulla, R.
D.; Nag, M.; J enks, W. S. J . Org. Chem. 2001, 66, 8621-8628.
1
890 J . Org. Chem., Vol. 68, No. 5, 2003

Significance of Atomic Oxygen in Sulfoxide Photochemistry
for para-substituted styrene (Figure 2A) and 12% for
para-substituted diaryl sulfide (Figure 2B). The slope for
photolysis of 1 and 2 was used to generate atomic oxygen
[O( P)] or an equivalent active oxygen species capable of
3
k
X
/k
H
is more than 5 times larger for diaryl sulfide than
oxygen transfer to the hydrocarbons. Our study focused
on the results of oxidation products in the reaction of
styrene 3 and on substrates that provided an opportunity
to vary the electronic and steric effects. The reaction was
similar to sulfoxide photodeoxygenation in a microwave
3
styrene. This is consistent with the idea that O( P) acts
between nonconjugated and conjugated transition states
in the substrate due to relative contributions of resonance
(Scheme 4). One might suggest that electron donation
3
9
enhances the formation of product due to a developing
charge in p-MeO-C
C
discharge method for generating O( P) atoms. Sulfoxide
photodeoxygenation is a potentially clean method for
δ+
δ-
6
H
4
CHdCH
2
- -O
and (p-MeO-
δ+
δ-
3
6
H
4
)
2
S
- -O , where solvation in acetonitrile is in-
O( P) production compared to the microwave discharge
method.8 With Argon purging of the sulfoxide sample
before photolysis, the method can preclude a secondary
oxidation process involving molecular oxygen. Microwave
,9
creased compared to less polar reagents and products.
The data are consistent with a charge-transfer depen-
dence of the active oxygen species and the electron
density on the alkene and sulfide trapping agent.
In flu en ce of Ster ics vs Electr on ics. The substrates
in Table 2 provide the first opportunity to compare the
steric and electronic effects in the photodeoxygenation
of 1 and 2. How the substrate structure effects the inter-
mediate along the sulfoxide photoreduction pathway was
sought. The interaction of the oxygen intermediate with
3
discharge methods produce O( P), but do so typically in
the presence of O
2
, which can lead to the production of
singlet molecular oxygen and ozone along with ozonides
8
,9
and carbonyl oxides in reactions with alkenes.
The findings from this study may establish a way to
generate high-value oxygenated compounds from readily
available petroleum molecules. Functionalization of hy-
drocarbons takes place under mild photolysis conditions
of 1 and 2, which leads to a possibility for the synthetic
o
substrate as measured by k/k appears to be more sensi-
tive to electronic effects than to steric effects. The reduced
importance of steric interactions in substituted alkyl
sulfides (entries 12-16) is shown by the observation of a
poor linear correlation in a Taft plot when log(k /k ) is
plotted versus E
3
use of atomic oxygen, O( P). Electrophilic triplet oxygen
atoms react mainly at tertiary C-H bonds of alkanes,
R
Me
based on relative bond strengths, 3° < 2° < 1°, to give
2
1-4
(δ ) 0.435; r ) 0.871). Steric effects
the corresponding alcohol molecules. Functionalization
s
appear to play a minor role in determining the outcome
of the reaction. The alkyl sulfides, diethyl sulfide, diiso-
propyl sulfide, and di-tert-butyl sulfide, possess nearly
identical oxidation potentials [E1/2 (V)], and yet the
decrease in rate is less than 4-fold (Table 2). This com-
of alkenes, such as styrene 3, gives rise to epoxide and
ketone products where a preference exists for oxygenation
of the side chain rather than the ring of styrene. It may
be noted that hydroxyl radicals are more reactive and
3
less selective toward hydrocarbons compared to O( P)
5
3
pares to the approximate 110-fold decrease in rate in the
atoms. Oxygen ( P) atoms balance reactivity and selec-
tivity and involve the oxidation of inert alkanes typically
inaccessible to the oxidants peracid, dioxirane, ozone, and
singlet molecular oxygen. The sulfoxide system provides
support for the application of this photooxygenation
procedure in other substrates. Future research will be
based on the idea that a catalyst present at the outset
may aid in the transfer of oxygen from sulfoxide to a
same series for 1
O
.
45
Steric interactions are known to
2
1
45
decrease the yield of alkyl sulfides with O
2
by affecting
physical quenching and chemical reaction pathways.
However, only the latter pathway would be available in
3
a mechanism involving the ground-state agent O( P). The
relative rates for tBu
monitoring the loss of starting material (entries 12 and
3, Table 2) given the fact branched alkyl sulfides have
2 2
S and iPr S were determined by
5
4
1
hydrocarbon. Polyoxomolybdates and polyoxotung-
4
8
55,56
a potential to oxidatively cleave. In doing so, enhanced
experimental errors ((20%) were observed for these
measurements.
states
have been studied in a similar manner for
catalysis under photochemical conditions.
Ack n ow led gm en t. This work was supported by the
Petroleum Research Fund, administered by the Ameri-
can Chemical Society, with start-up funds from the City
University of New York, and by a research award from
PSC-CUNY. We thank Professors Charlie Perrin and
Spiro Alexandratos for stimulating discussions and
Chengguo Dong for work on p-chlorostyrene in an early
phase of the project.
The indirect trapping data are rationalized by sug-
gesting that it is the electron-donating ability of the
substrates and not the steric interaction with the ap-
proaching oxidant that dictates the magnitude of k.
3
Atomic O( P) is a reasonable candidate for the electro-
philic species trapped in the Hammett, IP, and Taft
3
studies. Laser studies involving atomic oxygen O( P) have
also recorded a dependence on the ionization potential
Su p p or tin g In for m a tion Ava ila ble: Descriptions of the
computed geometries of all stationary points (Cartesian coor-
dinates) and absolute energies. This material is available free
of charge via the Internet at http://pubs.acs.org.
of the substrate.51
Con clu sion
J O0266487
We have studied the mechanism of sulfoxide photo-
chemistry and provide evidence for the oxygenation of
saturated and unsaturated hydrocarbons. A liquid-phase
(52) Levin, R. D. Ionization potential and appearance potential
measurements; Department of Commerce, National Bureau of Stan-
dards: Washington, DC, 1982; Vol. IV.
1
(
53) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photo-
(
48) Fox, M. A.; Abdel-Wahab, A. A. Tetrahedron Lett. 1990, 31,
533-4536.
49) Cvetanovic, R. J .; Singleton, D. L.; Irwin, R. S. J . Am. Chem.
Soc. 1981, 103, 3530-3539.
50) Nip, W. S.; Singleton, D. A.; Cvetanovic, R. J . J . Am. Chem.
Soc. 1981, 103, 3526-30
51) Guidoni, A. G.; Paladini, A.; Veneziani, M.; Naaman, R.; Di
Palma, T. M. Appl. Surf. Sci. 2000, 154-155, 186-191.
chemistry, 2nd ed., Revised and Expanded; Marcel Dekker Inc.: New
York, 1993; p 420.
(54) Renneke, R. F.; Hill, C. L. J . Am. Chem. Soc. 1986, 108, 3528-
4
(
9.
(
(55) Renneke, R. F.; Pasquali, M.; Hill, C. L. J . Am. Chem. Soc. 1990,
112, 6585-94.
(
(56) Renneke, R. F.; Hill, C. L. J . Am. Chem. Soc. 1988, 110, 5461-
70.
J . Org. Chem, Vol. 68, No. 5, 2003 1891