Visible light-promoted selective oxidation of sulfides to sulfoxides catalyzed by ruthenium porphyrins with iodobenzene diacetate

Article abstract of DOI:10.1016/j.apcata.2014.04.014

Under visible light irradiation, the carbonyl ruthenium(II) porphyrin complexes efficiently catalyze the selective oxidation of sulfides to sulfoxides with iodobenzene diacetate [PhI(OAc)₂] as the oxygen source. Various thioanisoles and allylic

Full text of DOI:10.1016/j.apcata.2014.04.014

Applied Catalysis A: General 478 (2014) 275–282
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Visible light-promoted selective oxidation of sulfides to sulfoxides
catalyzed by ruthenium porphyrins with iodobenzene diacetate
Tse-Hong Chen, Zhibo Yuan, Aaron Carver, Rui Zhang∗
Department of Chemistry, Western Kentucky University, 1906 College Heights Blvd no. 11079, Bowling Green, KY 42101, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Under visible light irradiation, the carbonyl ruthenium(II) porphyrin complexes efficiently catalyze the
selective oxidation of sulfides to sulfoxides with iodobenzene diacetate [PhI(OAc)2] as the oxygen source.
Various thioanisoles and allylic sulfides were oxidized to the corresponding sulfoxides without overoxi-
dation to sulfones. The high selectivity of this unprecedented oxygen-transfer process is mechanistically
rationalized by a low-reactivity ruthenium(IV)-oxo species which can be detected in the reaction of car-
bonyl ruthenium(II) porphyrin with iodobenzene diacetate. To the best of our knowledge, this is the first
demonstration of a mild and high-yield method for the highly selective sulfoxidations by ruthenium
porphyrins and PhI(OAc)2.
Received 18 December 2013
Received in revised form 3 April 2014
Accepted 5 April 2014
Available online 15 April 2014
Keywords:
Ruthenium porphyrin
Oxidation
© 2014 Elsevier B.V. All rights reserved.
Sulfide
Iodobenzene diacetate
Visible light
1. Introduction
In the past decades, many transition metal complexes [5] have
been synthesized to mimic the predominant oxidation catalysts in
The selective oxidation of sulfides to sulfoxides (sulfoxida-
tion) is of great importance in organic synthesis [1]. Organic
sulfoxides are valuable synthetic reagents for the production of
a variety of chemically and biologically significant molecules.
Optically active sulfoxides are also useful intermediates in medic-
inal and pharmaceutical chemistry to prepare the therapeutic
agents such as antiulcer (proton pump inhibitors), antibacterial,
antifungal, antiatherosclerotic, antihypertensive and cardiotonic
agents, as well as psychotonics and vasodilators [2]. A large vari-
ety of electrophilic reagents such as peracids, hydrogen peroxide,
hypochlorite, sodium periodate, iodosobenzene, peroxyacids, and
highly toxic oxo metal oxidants have been utilized for the oxidation
of conventional sulfides with the aim of obtaining high selectivity
for sulfoxide over sulfone [3,4]. However, the methods that are cur-
rently reported rarely provide the ideal combination of selectivity,
fast reaction kinetics, and high product yields. The search for the
efficient and selective catalytic oxidation reactions for the sulfoxide
preparation has continued to be the interest of chemical research.
Nature, namely the cytochrome P450 enzymes [6]. In biomimetic
catalytic oxidations, a transition metal catalyst is oxidized to a
high-valent metal-oxo species [7] by a sacrificial oxidant, and
then the reactive transition metal-oxo intermediate oxidizes the
substrate [8]. In this regard, ruthenium porphyrin complexes are
among the most extensively studied biomimetic oxidation cat-
alysts in view of their rich coordination and redox chemistry
[9,10]. Ruthenium porphyrins can be used to catalyze the oxida-
tion of a wide variety of organic substrates, particularly for various
alkenes, benzylic hydrocarbons and arenes [11]. Product turnovers
of over 10,000 can be reached by recycling the catalysts several
times or by performing the oxidation at low catalyst loadings, as
demonstrated in a number of selected cases [12,13]. The synthetic
ruthenium porphyrins associated with various sacrificial oxidants
such as iodosobenzene (PhIO), tert-butyl hydroperoxide (TBHP) and
m-chloroperoxybenzoic acid (m-CPBA) can catalyze a wide vari-
ety of oxidation reactions including epoxidation, hydroxylation
and oxidation of amines, sulfides, alcohols and aldehydes [11,14].
In particular, with heteroaromatic N-oxides as oxygen source,
ruthenium porphyrin complexes exhibit high region-, chemo-
and stereoselectivity in the catalytic oxidation of a variety of
hydrocarbons [13,15,16]. Notably, the well-characterized trans-
dioxoruthenium(VI) porphyrins, formulated as [RuVI(Por)O2], have
been shown to catalyze the clean aerobic epoxidation of olefins in
the absence of a reducing agent under mild conditions [17,18]. In
addition, stoichiometric oxidation of alkenes and C H bonds by
Abbreviations: Por, porphyrin dianion; TMP, 5,10,15,20-tetramesitylporphyrin
dianion; TPFPP, 5,10,15,20-tetrakispentafluorophenylporphyrin dianion; PhI(OAc)2,
iodobenzene diacetate; PhIO, iodosobenzene; TBHP, tert-butyl hydroperoxide; m-
CPBA, m-chloroperoxybenzoic acid; 2,6-Cl2PyNO, 2,6-dichloropyridine N-oxide.
∗
Corresponding author. Tel.: +1 270 745 3803; fax: +1 270 745 5361.
E-mail address: rui.zhang@wku.edu (R. Zhang).
http://dx.doi.org/10.1016/j.apcata.2014.04.014
0926-860X/© 2014 Elsevier B.V. All rights reserved.

276
T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
Scheme 1. Catalytic oxidation of sulfides by ruthenium porphyrins (1) in the presence of PhI(OAc)2 and visible light.
isolable [RuVI(Por)O2] revealed a useful insight into the mecha-
nisms of the catalytic oxidation processes [9,14,19].
Iodosobenzene (PhIO) and tert-butyl hydroperoxide (TBHP) were
purchased from the TCI America Co. and was used as obtained. All
reactive substrates of organic sulfides for catalytic oxidations were
passed through a dry column of active alumina (Grade I) before
use. 5,10,15,20-Tetrakis(pentafluorophenyl)porphyrin (H2TPFPP)
was commercially available from Aldrich and used as received.
5,10,15,20-Tetramesitylporphyrin (H2TMP) was prepared accord-
ing to the known methods [28]. The corresponding ruthenium(II)
carbonyl complexes RuII(Por)(CO), RuIV(Por)Cl2 and RuVI(Por)O2
used for catalytic sulfoxidadtions were prepared by literature
methods[14] and characterized by 1H NMR, IR and UV–vis spec-
troscopies, matching those reported [14,17,29].
UV–vis spectra were recorded on an Agilent 8453 diode array
spectrophotometer. IR spectra were obtained on a Bio-Rad FT-IR
spectrometer. 1H NMR was performed on a JEOL ECA-500 MHz
spectrometer at 298 K with tetramethylsilane (TMS) as inter-
nal standard. Chemical shrifts (ppm) are reported relative to
TMS. Gas chromatograph analyses were conducted on an Agilent
GC6890/MS5973 equipped with a flame ionization detector (FID)
using a DB-5 capillary column. The above GC/MS system is also cou-
pled with an auto sample injector. Reactions of RuII(Por)(CO) (1)
with excess of PhI(OAc)2 were conducted in a chloroform solution
at 23 2 ◦C.
Although a large number of reports on the catalytic behav-
ior of ruthenium porphyrins have appeared in the past two
decades with major focus on alkene epoxidation and activated
alkane hydroxylation, only very few studies have been reported
with limited success on oxidation of sulfides by these well stud-
ied ruthenium porphyrins [20]. Recently, we have reported the
kinetic studies of sulfide oxidation by the well characterized
trans-dioxoruthenium(VI) porphyrins [21]. We have also shown
that trans-dioxoruthenium(VI) porphyrin complexes can be readily
produced by irradiation of porphyrin–ruthenium(IV) dichlorate
complexes with visible light [22,23]. In this work, we aim to fully
explore the potential of ruthenium porphyrins toward catalytic
sulfoxidation reactions with iodobenzene diacetate, abbreviated
as [PhI(OAc)2], which is commercially available and easy to han-
dle. In contrast to the sacrificial oxidants in common use for
metalloporhyrin catalyzed reactions, PhI(OAc)2 does not show
appreciable reactivity toward organic substrates or dose not dam-
age the porphyrin catalysts under the usual catalytic conditions.
Due to the mild oxidizing ability, PhI(OAc)2 has been less often
employed in the metalloporphyrin-catalyzed oxidations. Coll-
man and Nam reported, respectively, the use of PhI(OAc)2 as
terminal oxidant for the iron(III) porphyrin catalyzed oxidation
of hydrocarbons [24,25]. Adam and coworkers also described a
highly selective oxidation of alcohols by chromium(III) salen with
PhI(OAc)2 [26]. In addition, Nishiyama and co-workers showed that
PhI(OAc)2 is a better oxidant than PhIO in ruthenium–pyridine-
2,6-dicarboxylate complex-catalyzed epoxidation of trans-stilbene
[27]. In the present study, we report that PhI(OAc)2 is an effi-
cient oxygen source associated with ruthenium porphyrins for the
selective oxidation of sulfides to sulfoxides, which of catalytic effi-
ciency is greatly enhanced by visible light irradiation (Scheme 1).
In all cases, quantitative conversions of sulfides and exclusive
selectivities for sulfoxides were obtained. Meanwhile, we show
that a low-reactivity ruthenium(IV)–oxo porphyrin intermediate is
detected by the oxidation of the ruthenium(II) carbonyl precursor
with PhI(OAc)2, which is ascribed to the observed excellent selec-
tivity for sulfoxide. To the best of our knowledge, using PhI(OAc)2
for the selective sulfoxidation reactions catalyzed by ruthenium
porphyrin complexes is unprecedented.
2.2. General procedure for photocatalytic sulfoxidations
In general, a Rayonet photoreactor (RPR-100) with a wave-
length range of 400–500 nm (ꢀmax = 420 nm) from 300 W mercury
lamps (RPR-4190 × 12) was used for the photocatalytic reactions.
The photochemical reactions typically consisted of 0.5–1.0 mg of
catalyst (approximate 0.5–1 ␮mol) in 2 mL of chloroform contain-
ing 0.5 mmol of organic substrates. 1.5 equivalent of PhI(OAc)2
(0.75 mmol) was added to the reaction solution as it was irradiated
at 25 2 ◦C. Aliquots of the reaction solution at constant time inter-
val were analyzed by 1H NMR or GC/MS to determine the formed
products and yields with an internal standard (diphenylmethane).
All reactions were run at least in duplicate, and the data reported
represent the average of these reactions. Monitoring reaction by
UV–vis spectroscopy before and after reactions indicated that no
significant degradation of ruthenium catalyst was found after 24 h
photolysis.
2.3. Catalytic competitive oxidations
2. Experimental
A
CHCl3 solution containing equal amounts of two sub-
2.1. Materials and instrumentation
strates, e.g. thioanisole (0.5 mmol) and substituted thioanisole
(0.5 mmol), ruthenium(II) porphyrin catalyst (1 ␮mol) and an inter-
nal standard of diphenylmethane (0.1 mmol) was prepared (final
volume = 2 mL). The internal standard was shown to be stable to the
oxidation conditions in control reactions. PhI(OAc)2 (0.4 mmol) as
limiting reagent was added, and the mixture was irradiated under
visible lights at ambient temperature (25 2 ◦C) until the reaction
was complete. Relative rate ratios for oxidations were determined
based on the amounts of products (sulfoxides) by 1H NMR or GC
All commercial reagents were of the best available purity
and were used as supplied unless otherwise specified. Iodoben-
zene diacetate or (diacetoxyiodo)benzene, [PhI(OAc)2], was
purchase from Aldrich Chemical Co. and used as such. m-
Chloroperoxybenzoic acid (m-CPBA) (77%) from Aldrich Chemical
Co. was purified by precipitation–crystallization from methy-
lene chloride and n-hexane, and then dried in vacuum.

T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
277
(1a) catalyzed the oxidation of thioanisole very slowly, only ca.
10% conversion was observed after 24 h. Gratifyingly, under visi-
ble light irradiation (ꢀmax = 420 nm), the reaction proceeded much
rapidly and thioanisole 2a was smoothly and quantitatively oxi-
dized to corresponding methyl phenyl sulfoxide (3a) in high yield
and short reaction times (entry 2 in Table 1 and Fig. 1). The sulfox-
ide was the only identifiable oxidation product (>99% by GC) and
the usual undesirable over oxidation of sulfoxide (3a) to sulfone
(4a) was not observed, which manifests the high chemoselectivity
of this reaction.
It has been well established that ruthenium(II) carbonyl por-
phyrins undergo photo-induced decarbonylation reactions [30].
The observed remarkable photo-stimulation with visible light was
consistent with photoejection of the carbonyl ligand to generate the
ruthenium(II) porphyrin that is much more active form of catalyst
to react with PhI(OAc)2. Therefore, all reactions were carried out
under visible light irradiation in a Rayonet photoreactor. The con-
trol experiments show that no formation of sulfoxide (<5%) was
detected when the reaction was carried out under light irradiation
in the absence of either ruthenium porphyrin catalyst or PhI(OAc)2.
In addition, monitoring catalytic reaction by UV–vis spectroscopy
indicated that the no significant degradation of porphyrin
catalyst was observed under visible light photolysis (see Fig. S1 in
the Supplementary material).
Fig. 1. Time courses of oxidation of thioanisole (0.5 mmol) with PhI(OAc)2
(0.75 mmol) in CDCl3 (2 mL) at room temperature catalyzed by
1 (1 ␮mol):,
RuII(TMP) (CO) (1a) without visible light (diamond); RuII(TMP)(CO) (1a) with visible
light (circle); RuII(TPFPP)(CO) (1b) with visible light (cube). Aliquots were taken at
selected time intervals for product analyses with 1H NMR.
(FID, DB-5). The values reported in Table 5 are the averages of 2–3
runs with a minor error (<5%).
As expected, the catalytic activity (100% convn in 4 h) was
enhanced with more PhI(OAc)2 present (Table 1, entry 3). Instead
of CDCl3, the use of CD2Cl2 as the solvent resulted in a similar
yield (entry 4). The reaction was also carried out in other sol-
vents like toluene, diethyl ether, ethyl acetate, acetonitrile. Due
to the poor solubility of PhI(OAc)2 in all these solvents, much
slower reactions were obtained with up to 30% conversions into
products (data not shown). In terms of different oxidation states
form of ruthenium catalyst, strikingly, the RuIV(TMP)Cl2 proved
to be completely inactive in the sulfoxidation reaction (entry
5), which is in marked contrast to the epoxidation of unfunc-
tionalized alkenes with 2,6-dichoropyridine N-oxide, for which
dichlororuthenium(IV) porphyrins are considerably more efficient
compared to the carbonyl ruthenium(II) catalyst [16]. Much lower
conversion was also observed in the oxidation of thioanisole
catalyzed by RuVI(TMP)O2 (entry 6). These results indicate that
ruthenium(IV) and ruthenium(VI) are not accessible for the gen-
eration of active intermediate in the catalytic oxidation of sulfides
by PhI(OAc)2. The electron-deficient catalyst RuII(TPFPP)(CO) (1b),
3. Results and discussions
3.1. Screening the catalytic oxidation of thioanisole
Although the iodobenzene diacetate, PhI(OAc)2, has been widely
used as a mild oxidant for some time [25,27], apparently it had
hitherto not been used for ruthenium porphyrin-catalyzed sul-
foxidation reactions. Therefore, we have first investigated the
PhI(OAc)2 as oxygen source in the catalytic oxidation of thioanisole
(2a) with different ruthenium porphyrin catalysts. Our immedi-
ate goal in this section was to identify the most efficient systems
and optimal conditions from the screening reactions. In order
to work under homogeneous conditions and follow the reaction
course via 1H NMR, the reactions were performed in CDCl3 at room
temperature using a 0.2 mol% catalyst loading and a 1:1.5 ration
of thioanisole and PhI(OAc)2. Our results collected in Fig. 1 and
Table 1 show the carbonyl ruthenium(II) tetramesitylporphyrin
Table 1
Catalytic oxidation of thioanisole by ruthenium porphyrins with iodobenzene diacetate under visible light irradiationa
.
Entry
Catalyst
t (h)
Convn (%)b
mb (%)b
Selectivity (3a:4a)b
1c
2
24
6
4
10
94
100
92
<5
15
>95
>95
>95
>95
>95
>95
>95
>99:1
>99:1
>99:1
>99:1
n.d.f
RuII(TMP)(CO)
3d
4e
5
6
RuIV(TMP)Cl2
RuVI(TMP)O2
RuII(TPFPP)(CO)
24
24
24
6
7
95:5
>99:1
90
a
All reactions were conducted with visible light irradiation (ꢀmax = 420 nm) in a Rayonet reactor or otherwise noted. Reactions were carried out in CDCl3 at ca. 25 ◦C with
a 1:1.5 molar ratio of thioanisole versus PhI(OAc)2 and 0.2 mol% catalyst at an initial substrate concentration of 0.25 M.
b
Conversions (convn), mass balance (mb) and product selectivity were determined by 1H NMR (JEOL 500 M) and/or quantitative GC–MS analysis on the crude reaction
mixture after the reaction is quenched by sodium hydroxide solution (consuming all oxygen source).
c
Without visible light irradiation.
2 equivalent of PhI(OAc)2 (1 mmol) was used.
In CD2Cl2.
n.d. = Not determined.
d
e
f

278
T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
Table 2
Catalytic oxidation of thioanisole by the ruthenium(II) poprhyrin (1a) with various oxygen sources in the presence of visible lighta
.
Entry
Oxygen source
t (h)
Convn (%)b
mb (%)b
Selectivity (3a:4a)b
1
PhI(OAc)2
PhIO
m-CPBA
TBHP
6
12
12
12
24
94
89
100
90
>95
96
92
>95
>95
>99:1
91:9
45:55
93:7
2
3c
4
5
2,6-Cl2PyNO
33
>99:1
a
All reactions were conducted with visible light irradiation (ꢀmax = 420 nm) in a Rayonet reactor. Reactions were carried out in CDCl3 at ca. 25 ◦C with a 1:1.5 molar ratio
of thioanisole versus oxygen source and 0.2 mol% catalyst at an initial substrate concentration of 0.25 M.
b
Conversions (convn), mass balance (mb) and product ratios were determined by 1H NMR (JEOL 500 M) or by quantitative GC–MS analysis with an internal standard
(diphenylmethane) on the crude reaction mixture after the reaction is quenched by sodium hydroxide solution (consuming all oxygen source).
c
In the absence of the ruthenium(II) catalyst, control experiment shows m-CPBA gave 100% conversion with 70:21 ratio of sulfoxide versus sulfone over 24 h.
which has been used a robust catalyst in many oxidation reactions,
also gave the sulfoxide exclusively. Compared to 1a, 1b was less
reactive and a longer reaction time (24 h) was needed to observe a
complete substrate conversion (entry 7).
We also followed the time courses of para-substituted
thioanisoles oxidations by RuII(TMP)(CO) (1a) and PhI(OAc)2 under
visible light photolysis (Fig. 2). The introduction of electron-
donating and electron-demanding substituents in the aryl ring of
substrates gave a noticeable effect. As is evident, the introduction of
electron-donating groups like methyl or methoxy groups resulted
in increased reactivities compared to thioanisole (entries 5, 6, 11,
12 in Table 3), while reduced reactivities were observed in the pres-
ence of electron-demanding groups like Cl or Br groups. However,
the log krel [krel = k(substituted thioanisole)/k(thioanisole)] versus
Hammett substituent constant (ꢁp) did not give a linear correlation
for the catalytic oxidation of substituted sulfides under compet-
itive conditions. Previously, we showed that the stoichiometric
thioanisoles oxidation by trans-dioxoruthenium(VI) porphyrins in
which cases both electron-donating and -withdrawing substituents
moderately slow down the reactions [21]. Apparently, involvement
of trans-dioxoruthenium(VI) species as the premier reactive inter-
mediate in present study is implausible.
3.2. Comparison of various oxygen sources in the catalytic
oxidation of thioanisole
The promising results with the PhI(OAc)2 in Table 1 prompted
us to evaluate other common oxygen sources in the rutheniu(II)
porphyrin-catalyzed oxidation of thioanisole for the intended pur-
pose of comparison. A screening of diverse oxygen source under
identical experimental conditions disclosed that the mild oxygen
source PhI(OAc)2 was especially effective for selective oxidation
of sulfide to sulfoxide, as representative results are shown in
Table 2. The use of more oxidizing oxygen source like iodosoben-
zene (PhIO), m-chloroperoxybenzoic acid (m-CPBA) or tert-butyl
hydroperoxide (TBHP) gave quantitative conversions albeit with
minor to significant sulfone formation (Table 2, entries 2–4).
The most likely explanation is that these oxygen sources, having
more oxidizing abilities, might generate more reactive oxidizing
intermediates with ruthenium porphyrin catalyst than that with
PhI(OAc)2.[29,31] Similar to the catalytic expoxidation of unfunc-
tionlized alkenes [13], 2,6-dichloropyridine N-oxide (2,6-Cl2PyNO)
as the oxygen source afforded excellent selectivity to sulfoxide,
however, with a very poor conversion of 33% after 24 h (entry 5).
Besides the high chemoselectivity, an additional advantage of the
PhI(OAc)2 compared to the other oxidants, is the fact that it will
not cause the degradation of the porphyrin catalysts. Thus, a sim-
ilar catalytic activity with excellent selectivity was achieved even
at the very low catalyst loading of only 0.05 mol% (data not shown
here).
The present results demonstrate that the ruthenium porphyrin
with PhI(OAc)2 is an efficient catalytic system for the selective oxi-
dation of sulfides. Its efficacy for epoxidation was also tested with
cis-cyclooctene as substrates under the optimized reaction con-
ditions. Surprisingly, no appreciable reactivity (<1% convn) was
observed after 24 h for the unfunctionalized alkene. This result
stimulates us to test this protocol for the oxidation of allylic sulfides.
3.3. Catalytic oxidation of substituted thioanisoles and allylic
sulfides by PhI(OAc)2
The substrate scope of the unprecedented catalytic sulfoxida-
tions was explored under optimized conditions. Table 3 lists the
oxidized products and corresponding selectivities using the 1a and
1b as the catalyst, respectively. In analogy with what we observed
for 2a, all catalytic oxidation of substituted thioanisoles proceeded
with a quantitative conversion into the corresponding sulfoxides
with excellent selectivity (100%) and mass balance (>95%), and in
all cases no traces of sulfones were detected. Significant for prepar-
ative purposes, the reactions gave comparable yields in isolated
products (entries 1 and 7).
Fig. 2. Time courses of oxidation of para-substituted (4-X) thioanisoles (0.5 mmol)
with PhI(OAc)2 (0.75 mmol) in CDCl3 (2 mL) at room temperature catalyzed by 1a
(1 ␮mol) under visible light irradiation. X = H (dot), MeO (triangle), Me (circle), Cl
(square) and Br (diamond).

T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
279
Table 3
Catalytic oxidation of substituted thioanisoles by ruthenium porphyrin (1) and PhI(OAc)2 with visible lighta.
Entry
Catalyst 1
Substrate 2
Product 3
Convn (%)b
8
Yield (%)b
100
Time (h)
1
100 (90)c
100
RuII(TMP)(CO) 1a
2
3
4
5
6
8
100
100
100
100
100
100
100
12
5
6
7
8
4
24
6
100
100
100
100
100 (93)c
100
RuII(TMP)(CO) 1b
9
10
11
8
6
6
6
100
100
100
100
100
100
100
100
12
a
Unless otherwise specified, all reactions were carried out in CDCl3 at ca. 25 ◦C in the presence of 1.5 equiv. of PhI(OAc)2 and 0.2 mol% catalyst; only sulfoxide was detected
by 1H NMR or GC/MS analysis of the crude reaction mixture.
b
Based on the conversion of thioanisoles and determined by 1H NMR or GC/MS analysis of the crude reaction mixture; material balance > 95%.
Isolated material after silica-gel chromatography with methylene and n-hexane as eluent.
c
The oxidation of sulfides in the presence of electron-rich double
bonds is often problematic with many traditional oxidants such as
m-CPBA, NaIO4, and catalytic systems because of interference with
epoxidations [3]. With ruthenium(II) catalyst and PhI(OAc)2, no
epoxidation took place and excellent chemoselectivities for sulfox-
ide were obtained albeit with somewhat low conversions (Table 4).
Similarly, presence of hydoxy group did not disturb the selec-
tive oxidation of sulfide to sulfoixde and no alcohol oxidation was
observed (entry 3 in Table 4).
preparative utility and synthetic value of the new catalytic system
presented above are indisputable, but mechanistic understand-
ing of the complex oxygen-transfer processes is important for the
design of still more effective and selective oxidants with general
applicability.
In fact, trans-dioxoruthenium(VI) porphyrin is capable of oxi-
dizing olefins and activated hydrocarbons to undergo reduction
[11]. However, the fact that no expoxidation or hydroxylation
was observed in the catalytic oxidation of allylic/hydroxy sul-
fides by ruthenium porphyrin and PhI(OAc)2 strongly suggests that
trans-dioxoruthenium(VI) porphyrin is not a viable oxidizing inter-
mediate in above catalytic cycle, even if it does have the ability to
effect substrate oxidations.
3.4. Mechanistic studies of active oxidizing species
Prior to the present studies, the use of ruthenium porphyrins
for metal-catalyzed sulfoxidations has met with limited success
in view of the low reactivity and, in most cases, undesirable
over-oxidation to sulfones. We now show that ruthenium por-
phyrins catalyze the highly selective oxidation of thioanisoles and
allylic sulfides by PhI(OAc)2 in the presence of visible light. The
To accentuate this point, the competitive sulfoxidation reac-
tions catalyzed by carbonyl ruthenium(II) porphyrin complexes (1)
with PhI(OAc)2 were conducted as described in Section 2. The pur-
pose of the competition studies is to evaluate whether the same
species was active in the two sets of conditions by comparing the

280
T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
Table 4
Catalytic oxidation of allylic or hydoxy sulfides by ruthenium porphyrin (1) and PhI(OAc)2 with visible lighta.
Entry
1
Catalyst
Substrate
Product
Convn (%)b
40
Yield (%)b
100
RuII(TMP)(CO) 1a
2
3c
4
100
32
100
100
100
RuII(TPFPP)(CO) 1b
5
64
Unless otherwise specified, all reactions were carried out in CDCl3 at ca. 25 ◦C in the presence of 1.0 equiv. of PhI(OAc)2 and 0.2 mol% catalyst; only sulfoxide was detected
by 1H NMR analysis of the crude reaction mixture.
a
Based on the conversion of thioanisoles and determined by 1H NMR analysis of the crude reaction mixture; mass balance > 95%. The products were fully characterized
b
by 1H NMR after column chromatography.
c
1.5 equiv. of PhI(OAc)2 was used.
ratios of products formed under catalytic turnover conditions to the
ratios of rate constants measured in the direct kinetic studies [32].
If the same oxidant was present in both cases, the ratios of abso-
lute rate constants from direct kinetic measurements and relative
rate constants from the competition studies should be similar [33],
although a coincident similarity for two different oxidants could
not be excluded. When the ratios were not similar, however, the
active oxidants under the two sets of conditions must be different
[21,34].
Table 5 contains the results of kinetic results with trans-
dioxoruthenium(VI) porphyrin complexes from our recent study
[21] and competition reactions where PhI(OAc)2 was employed
as sacrificial oxygen source, respectively. Each sulfide substrate
was oxidized to the corresponding sulfoxide in nearly quantita-
tive yield based on the oxidant consumed. In this study, a limiting
amount of PhI(OAc)2 was used to keep the conversion less than
20% to avoid the effect of substrate concentration on the prod-
uct ratios. As evident in Table 5, the results from the competition
studies suggest that an unusually different behavior is apparent
during catalytic turnover conditions with porphyrin–ruthenium(II)
carbonyl complexes. For most of the cases, the ratios of absolute
rate constants found in direct kinetic studies differed from the
oxidation ratios for competition oxidation reactions of the two
substrates. The differences appear to be statistically significant in
the ratios of substrate oxidation rates. The oxidation rate ratios
for each pair of substrates were found to be much more larger
(more selective) when PhI(OAc)2 employed compared to that from
the absolute rate constants ratio of RuVI(Por)O2, implying that the
active oxidant generated under catalytic turnover condition with
PhI(OAc)2 should be less reactive than RuVI(Por)O2. Therefore, a
trans-dioxoruthenium(VI) is unlikely to be the active oxidant under
turnover conditions.
To further probe the identity of the active oxidizing species
under catalytic conditions, we conducted the chemical oxidation
reaction of ruthenium(II) catalyst 1 by PhI(OAc)2 in CHCl3 in the
absence of sulfides (Fig. 3). As shown in Fig. 3A, with five equivalent
of PhI(OAc)2, the precursor RuII(TMP)(CO) (1a) was quantitatively
converted to a species 5a with clearly resolved isosbestic points.
The absorption spectrum of 5a with a weak Soret band at 406 nm
and broad absorption bands around 550–750 nm was essentially
identical to that of the known ruhenium(IV) mono–oxo porphyrin,
which was independently prepared from a reported method[29]
(the Supplementary material, Fig. S2). Further addition of PhI(OAc)2
with up to 20 equivalent resulted in conversion of 1a to another
species 6a with a red shift of the Soret band to 422 nm and blue shift
of Q band to 518 nm, again with clean isosbestic points (Fig. 3B). The
later signal of the final product is characteristic of the well-known
trans-dioxoruthenium(VI) poprhyrin, i.e. RuVI(TMP)O2 [31]. These
Table 5
Comparison of competition catalytic oxidations with ratios of absolute rate constants of trans-dioxoruthenium(VI) porphyrinsa.
b
Porphyrin system
TPFPP
Substrates
Method
krel
Kinetic results
PhI(OAc)2
Kinetic results
PhI(OAc)2
Kinetic results
PhI(OAc)2
Kinetic results
PhI(OAc)2
Kinetic results
PhI(OAc)2
Kinetic results
PhI(OAc)2
0.94
1.49
0.71
3.22
0.66
2.23
0.64
3.7
0.375
1.47
1.01
3.12
p-Fluorothioanisole/thioanisole
TPFPP
TPFPP
TPFPP
TMP
p-Chlorothioanisole/thioanisole
p-Methylthioanisole/thioanisole
p-Methoxythioanisole/thioanisole
p-Fluorothioanisole/thioanisole
p-Chlorothioanisole/thioanisole
TMP
a
A reaction solution containing equal amounts of two substrates, e.g., thioanisole (0.5 mmol) and substituted thioanisole (0.5 mmol), ruthenium(II) porphyrin catalyst
(1 ␮mol) and an internal standard of diphenylmethane (0.1 mmol) was prepared in CH3Cl (5 mL). PhI(OAc)2 (0.4 mmol) was added, and the mixture was stirred overnight
(12 h) under visible light irradiation at 25 ◦C.
b
Relative ratios of absolute rate constants from kinetic results with trans-dioxoruthenium(VI) porphyrin complexes[21] and for competitive oxidations with carbonyl
ruthenium(II) porphyrin catalysts (1a and 1b). All competition ratios are averages of 2–3 determinations with standard deviations smaller than 10% of the reported values.

T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
281
Fig. 3. Time-resolved spectrum for the oxidation of 1a (5 ×10−6 M) with PhI(OAc)2 in CHCl3 (A) 5 equivalent of PhI(OAc)2 over 40 min. (B) 20 equivalent of PhI(OAc)2 over
20 min.
results unambiguously demonstrate that PhI(OAc)2 can be used as a
single oxygen atom source in generating trans-dioxoruthnium(VI)
porphyrin via a monoruthenium(IV)–oxo intermediates (Eq. (1)). Of
note, the formation of trans-dioxoruthenium(VI) porphyrin species
(6a) was not observed for the same reaction but in the presence of
organic sulfides.
(1)
It is well known that ruthenium(IV)–oxo porphyrins are much
less reactive than the corresponding trans-dioxoruthenium(VI)
porphyrins as the latter can oxidize the alkenes to expoxides and
the former does not oxidize the alkenes but react with powerful
reductants like triphenylposphine [35]. In this study, we found that
ruthenium(IV)-oxo species (5a) when produced independently was
competent oxidant and reacted rapidly with thioanisoles to gen-
Scheme 2. A proposed catalytic cycle for the catalytic oxidations by the ruthe-
nium(II) carbonyl porphyrin in the presence of PhI(OAc)2 and visible light.
erate the corresponding ruthenium(II) compound (Supplementary
material, Fig. S3). Similarly, the electron-demanding RuVI(TPFPP)O2
(6b) can be prepared by oxidation of the carbonyl precursor 1b with
which could explain the observed lower activity for the electron-
30 equivalent of PhI(OAc)2, as shown by the close spectral agree-
demanding system 1b in comparison to 1a.
ment with the TMP analogue (Supplementary material, Fig. S4). Due
to the electron-demanding nature, the oxidation of RuII(TPFPP)(CO)
4. Conclusions
to RuVI(TPFPP)O2 is much slower than that of TMP analogue.
On the basis of the present experimental facts, a catalytic
cycle is proposed in Scheme 2. First, visible light photolysis of
carbonyl precursor 1 gives a more active ruthenium(II) species
RuIIPor. Second, oxidation of RuIIPor with PhI(OAc)2 generated
the ruthenium(IV)–oxo porphyrin (5) as the major oxidizing
intermediate which subsequently oxidizes an organic sulfide selec-
tively to sulfoxide without over oxidation to sulfone due to its
low reactivity. After the oxygen transfer, the ruthenium(IV)-oxo
species 5 then regenerates the precursor RuIIPor (Scheme 2).
The pivotal formation of the low-reactivity ruthenium(IV)–oxo
poprhyrin was unequivocally confirmed by the oxidation of 1a
with PhI(OAc)2 in the absence of sulfide substrate (Fig. 3A). The
notable absence of reactivity in the case of expoxidation and
hydroxylation compared to sulfoxidation for the oxidation of allylic
or hydroxy sulfides illustrates that the formation of more reac-
tive trans-dioxoruthenium(VI) species is not directly involved in
above catalytic cycle. We further assume that the formation of
the ruthenium(IV)-oxo intermediate is the rate-determined step,
In conclusion, an efficient method for the highly selective
oxidation of sulfides to sulfoxides with ruthenium porphrin as
catalyst in the presence of PhI(OAc)2 and visible light has been
developed. All the factors that effected the catalytic sulfoxida-
tion reactions were well investigated and understood. Various
thioanisoles and allylic sulfides could be successfully oxidized with
good conversions (most quantitative) and excellent selectivities.
A plausible mechanism which relies on the generation of low-
reactivity ruthenium(IV)-oxo intermediate has been proposed to
account for the observed high selectivities. Further studies to define
synthetic applications and characterize the observed transients
more fully are underway in our laboratory.
Acknowledgments
This work was supported by the NSF grant (CHE-1213971) and
an internal grant from WKU Office of Research (RCAP 12-8012).

282
T.-H. Chen et al. / Applied Catalysis A: General 478 (2014) 275–282
[14] C.-M. Che, J.-L. Zhang, R. Zhang, J.-S. Huang, T.-S. Lai, W.-M. Tsui, X.-G. Zhou,
Appendix A. Supplementary data
Z.-Y. Zhou, N. Zhu, C.K. Chang, Chem. Eur. J. 11 (2005) 7040–7053.
[15] J.T. Groves, M. Bonchio, T. Carofiglio, K. Shalyaev, J. Am. Chem. Soc. 118 (1996)
8961–8962.
[16] R. Zhang, W.-Y. Yu, K.-Y. Wong, C.-M. Che, J. Org. Chem. 66 (2001) 8145–8153.
[17] J.T. Groves, R. Quinn, J. Am. Chem. Soc. 107 (1985) 5790–5792.
[18] T.-S. Lai, R. Zhang, K.-K. Cheung, C.-M. Che, H.-L. Kwong, Chem. Commun. (1998)
1583–1584.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.apcata.
2014.04.014.
References
[19] R. Zhang, W.-Y. Yu, H.-Z. Sun, W.-S. Liu, C.-M. Che, Chem. Eur. J. 8 (2002)
2495–2507.
[20] N. Rajapakse, B.R. James, D. Dolphin, Catal. Lett. 2 (1989) 219–226.
[21] C. Abebrese, Y. Huang, A. Pan, Z. Yuan, R. Zhang, J. Inorg. Biochem. 105 (2011)
1555–1561.
[22] Y. Huang, E. Vanover, R. Zhang, Chem. Commun. 46 (2010) 3776–3778.
[23] R. Zhang, Y. Huang, C. Abebrese, H. Thompson, E. Vanover, C. Webb, Inorg. Chim.
Acta 372 (2011) 152–157.
[24] J.P. Collman, A.S. Chien, T.A. Eberspacher, J.I. Brauman, J. Am. Chem. Soc. 122
(2000) 11098–11100.
[25] J.H. In, S.E. Park, R. Song, W. Nam, Inorg. Chim. Acta 343 (2003) 373–376.
[1] P. Kowalski, K. Mitka, K. Ossowska, Z. Kolarska, Tetrahedron 61 (2005)
1933–1953.
[2] S. Caron, R.W. Dugger, S.G. Ruggeri, J.A. Ragan, D.H.B. Ripin, Chem. Rev. 106
(2006) 2943–2989.
[3] J.E. Baeckvall, Modern Oxidation Methods, Wiley-VCH Verlag, Weinheim,
2004.
[4] E. Wojaczynska, J. Wojaczynski, Chem. Rev. 110 (2010) 4303–4356.
[5] R.A. Sheldon (Ed.), Metalloprophyrins In Catalytic Oxidations, Marcel Dekker,
New York, NY, 1994.
[6] P.R. Ortiz de Montellano (Ed.), Cytochrome P450 Structure, Mechanism, and
Biochemistry, third ed., Kluwer Academic/Plenum, New York, NY, 2005.
[7] B. Meunier (Ed.), Metal-Oxo and Metal-Peroxo Species in Catalytic Oxidations,
Springer-Verlag, Berlin, 2000.
[26] W. Adam, S. Hajra, M. Herderich, C.R. Saha-Moller, Org. Lett.
2773–2776.
2 (2000)
[27] H. Nishiyama, Y. Motoyama, Chem. Commun. (1997) 1863–1864.
[28] J. Lindsey, R.D. Wagner, J. Org. Chem. 54 (1989) 828–836.
[29] W.-H. Leung, C.-M. Che, J. Am. Chem. Soc. 111 (1989) 8812–8818.
[30] M. Hoshino, Y. Kashiwagi, J. Phys. Chem. 94 (1990) 673–678.
[31] J.T. Groves, R. Quinn, Inorg. Chem. 23 (1984) 3844–3846.
[32] R. Zhang, M. Newcomb, Acc. Chem. Res. 41 (2008) 468–477.
[33] R. Zhang, J.H. Horner, M. Newcomb, J. Am. Chem. Soc. 127 (2005) 6573–6582.
[34] Z. Pan, R. Zhang, M. Newcomb, J. Inorg. Biochem. 100 (2006) 524–532.
[35] Y. Watanabe, H. Fujii, in: B. Meunier (Ed.), Metal-Oxo and Metal-Peroxo Species
in Catalytic Oxidations, Springer-Verlag, Berlin, 2000.
[8] B. Meunier, Chem. Rev. 92 (1992) 1411–1456.
[9] C.-M. Che, W.-Y. Yu, Pure Appl. Chem. 71 (1999) 281–288.
[10] J.T. Groves, K. Shalyaev, J. Lee, in: K.M. Kadish, K.M. Smith, R. Guilard (Eds.), The
Porhyrin Handbook, Academic Press, 2000, pp. 17–40.
[11] C.-M. Che, J.-S. Huang, Chem. Commun. (2009) 3996–4015.
[12] X.-Q. Yu, J.-S. Huang, W.-Y. Yu, C.-M. Che, J. Am. Chem. Soc. 122 (2000)
5337–5342.
[13] H. Ohtake, T. Higuchi, M. Hirobe, Heterocycles 40 (1995) 867–903.