Mechanism of the selective sulfide oxidation promoted by HNO 3/FeBr3

Article abstract of DOI:10.1021/jo9015248

(Chemical Equation Presented) The oxidation of aryl methyl sulfides containing electron withdrawing and electron donating groups (p-NO₂, p-CHO, p-NCS, p-Br, H, p-CH₃, p-OCH₃) was carried out in homogeneous solution in acetonitrile in the presence of catalytic amounts of HNO₃ and FeBr₃. The FeBr₃ is required for the reaction to proceed for compounds with strongly electron withdrawing groups (p-NO₂ and p-CHO) but is not necessary in the case of all the other compounds, although in the latter cases the yield decreases considerably. The rate of the reaction was measured as a function of substrate, FeBr₃ and HNO₃ concentration. From the experimental data a mechanism is suggested where there are two reaction pathways, one involving the formation of a ternary complex between the substrate, FeBr₃ and NO ₃^- and one involving a complex formed between the sulfide and the HNO₃. From these complexes HNO₂ is generated, which then combines with HNO₃ to yield N₂O₄, initiating a catalytic cycle where the sulfide is oxidized and O₂ from the air is stoichiometrically consumed.

Full text of DOI:10.1021/jo9015248

pubs.acs.org/joc
Mechanism of the Selective Sulfide Oxidation Promoted by HNO3/FeBr3
Claudio O. Kinen, Laura I. Rossi,* and Rita H. de Rossi*
´
´
ꢀ
´
ꢀ
Instituto de Investigaciones en Fısico Quımica de Cordoba (INFIQC), Departamento de Quımica Organica,
´
Facultad de Ciencias Quımicas, Universidad Nacional de Cordoba, Ciudad Universitaria, X5000HUA
ꢀ
ꢀ
Cordoba, Argentina
ritah@fcq.unc.edu.ar; lauraros@fcq.unc.edu.ar
Received July 15, 2009
The oxidation of aryl methyl sulfides containing electron withdrawing and electron donating groups
(p-NO2, p-CHO, p-NCS, p-Br, H, p-CH3, p-OCH3) was carried out in homogeneous solution in
acetonitrile in the presence of catalytic amounts of HNO3 and FeBr3. The FeBr3 is required for the
reaction to proceed for compounds with strongly electron withdrawing groups (p-NO2 and p-CHO)
but is not necessary in the case of all the other compounds, although in the latter cases the yield
decreases considerably. The rate of the reaction was measured as a function of substrate, FeBr3 and
HNO3 concentration. From the experimental data a mechanism is suggested where there are two
reaction pathways, one involving the formation of a ternary complex between the substrate, FeBr3
-
and NO3 and one involving a complex formed between the sulfide and the HNO3. From these
complexes HNO2 is generated, which then combines with HNO3 to yield N2O4, initiating a catalytic
cycle where the sulfide is oxidized and O2 from the air is stoichiometrically consumed.
Introduction
chemistry because oxidations of other functional groups can
take place simultaneously.5 Furthermore, sulfoxides can un-
dergo overoxidation to sulfones and therefore it is important
that the catalyst has a low reactivity toward the sulfoxides.6
Considering the interest in the development of synthetic
methods for the selective conversion of sulfides to sulfoxides,7,8
Sulfoxides and other organosulfur compounds are impor-
tant synthetic intermediates in organic chemistry1 and are
valuable in the preparation of biologically and pharmaceuti-
cally relevant materials.2,3 One of the oxidation reactions
found in pharmaceutical research and production is that of a
sulfide toa sulfoxide, which is achieved witha very wide variety
of reagents.3 When there are several different functional
groups present in a molecule as in esomeprazole,4 a sulfox-
ide-containing drug, chemoselective transformations are of
great significance. This has often been difficult in sulfoxidation
(5) Shukla, V. G.; Salgaonkar, P. D.; Akamanchi, K. G. J. Org. Chem.
2003, 68, 5422.
(6) (a) Dell’Anna, M. M.; Mastrorilli, P.; Nobile, C. F.; Taurino, M. R.;
ꢁ
Calo, V.; Nacci, A. J. Mol. Catal. A: Chem. 2000, 151, 61. (b)
Venkataramanan, N. S.; Kuppuraj, G.; Rajagopal, S. Coord. Chem. Rev.
2005, 249, 1249.
(7) (a) Hudlickꢀy, M. Oxidations in Organic Chemistry, ACS Monograph
186; American Chemical Society: Washington, DC, 1990. (b) Mata, E. G.
Phosphorus, Sulfur, Silicon Relat. Elem. 1996, 117, 231. (c) Romanelli, G. P.;
(1) (a) Cremlyn, R. J. An Introduction to Organosulfur Chemistry; John
Wiley & Sons: Chichester, UK, 1996. (b) Page, P. Organosulfur Chemistry:
Synthetic and Stereochemical Aspects; Academic Press, London, UK, 1998; Vol.
2. (c) Hiroi, K.; Suzuki, Y.; Abe, I.; Kawagishi, R. Tetrahedron 2000, 56 4701.
ꢀ
Bennardi, D. O.; Palermo, V.; Vazquez, P. G.; Tundo, P. Lett. Org. Chem. 2007, 4,
544.
(8) (a) Kowalski, P.; Mitka, K.; Ossowska, K.; Kolarska, Z. Tetrahedron
2005, 61, 1933. (b) Mahamuni, N. N.; Gogate, P. R.; Pandit, A. B. Ultrason.
Sonochem. 2007, 14, 135. (c) Yuan, Y.; Bian, Y. Tetrahedron Lett. 2007, 48,
8518. (d) Kumar, A.; Kanksha, A. Tetrahedron Lett. 2007, 48, 7857. (e)
~
ꢀ
(2) (a) Carreno, M. C. Chem. Rev. 1995, 95, 1717. (b) Fernandez, I.;
Khiar, N. Chem. Rev. 2003, 103, 3651.
(3) Caron, S.; Dugger, R. W.; Ruggeri, S. G.; Ragan, J. A.; Brow Ripin,
D. H. Chem. Rev. 2006, 106, 2943.
ꢀ
Sathicq, A. G.; Romanelli, G. P.; Palermo, V.; Vazquez, P. G.; Thomas, H. J.
€
(4) Cotton, H.; Elebring, T.; Larsson, M.; Li, L.; Sorensen, H.; Von Unge,
S. Tetrahedron: Asymmetry 2000, 11, 3819.
Tetrahedron Lett. 2008, 49, 1441. (f) Matsumoto, K.; Yamaguchi, T.;
Katsuki, T. Chem. Commun. 2008, 1704.
7132 J. Org. Chem. 2009, 74, 7132–7139
Published on Web 08/19/2009
DOI: 10.1021/jo9015248
r
2009 American Chemical Society

Kinen et al.
JOCArticle
we are currently engaged in the study of reaction methodologies
to achieve chemoselective sulfoxidation reactions.9,10 In a pre-
vious paper,11 we reported on the synthesis, characterization,
and catalytic activity of several cyclodextrin-FeBr3 (CD-Fe)
complexes. We found that CD-Fe complexes, in the presence
of a catalytic amount of [Fe(NO3)3 9H2O] as oxidant, reacted
3
with sulfides giving excellent yields of the corresponding sulf-
oxides. Furthermore, these complexes can be reused several
times. These reactions were performed by recycling the solid
CD-Fe complexes while the substrate, iron(III) nitrate, and the
organic solvent were renewed.11b
Complexes with cyclodextrins are of particular interest
because of the possibility of inducing enantiomeric excess.12
We reported that substrates with various highly oxidizable
functions such as isothiocyanate or aldehyde remain un-
changed under the reaction conditions used.10 In addition,
the methodology proposed for the oxidation of sulfides
fulfills several of the green chemistry principles since the
oxidant is oxygen from the air, the reactions are highly
selective producing minimum waste, and the catalyst is a
noncontaminating metal.13
To obtain more insight into the mechanism of these
reactions, we performed a kinetic study of the sulfoxida-
tion reaction of sulfides 1a-g (eq 1) in acetonitrile using
catalytic amounts of FeBr3 and HNO3 as oxidation
promoters. The results clearly indicate that the oxygen
from the air is the species used in stoichiometric amounts
whereas FeBr3 plays a prominent role in the initiation
of the process. On the other hand, HNO3 is involved in
the initiation and in the catalytic cycle. Furthermore,
FeBr3 is needed for substrates with strongly electron
withdrawing groups while the reactions of sulfides with
donor groups proceeds to products even in the absence
of FeBr3.
FIGURE 1. Spectra of 1a (X = p-NO2) as a function of time. The
arrow indicates increasing time. The reaction contains 1a =1.3 ꢀ
10-3 M, HNO3 (10%), and FeBr3 (5%) in 2.6 mL of acetonitrile
at 25 °C. The spectra were recorded after 500-fold dilution of the
original solution in dichloromethane.
completion. The reactions carried out in open vessels did not
give quantitative yields of the oxidation product. We suspect
that this is because the nitrogen oxides, which are formed and
which are required for the reaction, leave the solution in open
vessels and therefore cannot take part in the catalytic cycle
(see below). In all cases, the reaction gave the sulfoxide
without any contamination by the corresponding sulfone
or other products. The evolution of the reaction was deter-
mined by gas chromatography in some cases but in most
instances the measurements were done by UV-vis spectro-
metry, following the decrease in absorption at the wave-
length maximum of the sulfide (Figure 1 is representative).
At the end of the reaction, the yield of product was deter-
mined by GC analysis. All reactions were run in duplicate
and when the difference between the two runs was more than
10% they were discarded. Some of the reactions were
followed by UV-vis and gas chromatography as well and
the rate constants calculated were the same within experi-
mental error.
Results and Discussion
Effect of the Substituents. The sulfoxidation reaction of
sulfides 1a-g was studied in the presence of FeBr3 and in its
absence, using 13% of HNO3 as oxidation promoter. The
results are summarized in Table 1. It can be seen that in the
presence of FeBr3 all substrates gave 100% reaction (entries
1-7, Table 1). On the other hand, in otherwise identical
conditions but in the absence of FeBr3, substrates with
strongly electron withdrawing groups, such as p-NO2 and
p-CHO, did not react at all (entries 8 and 9, Table 1). The
other substrates (entries 10-13, Table 1) reacted but the yield
of the reaction was less than 100%, except for the one with
the strongest electron donating group (entry 14, Table 1). It is
interesting to note that for the substrates that reacted in the
presence and in the absence of FeBr3, the rate in the presence
of the metal increased very little, so the rate constant ratios
are 1.37, 1.09, 1.76, 1.51, and 1.55 for compounds with
p-NCS (entries 3 and 10, Table 1), p-Br (entries 4 and 11,
The kinetics of the sulfoxidation reaction of substrates
1a-g in the presence of HNO3/FeBr3 in catalytic quantities
was measured in different reaction vessels open or closed. It
was determined that for the reaction to proceed to comple-
tion it was necessary to have a closed vessel with a volume
that contained the required amount of oxygen for the
stoichiometric oxidation reaction. If the closed vessel con-
tained a small amount of O2, the reaction did not proceed to
(9) Rossi, L. I.; Martin, S. Appl. Catal., A 2003, 250, 271.
(10) Kinen, C. O.; Rossi, L. I.; de Rossi, R. H. Appl. Catal., A 2006, 312,
120.
(11) (a) Rossi, L. I.; de Rossi, R. H. J. Supramol. Chem. 2002, 2, 509. (b)
Rossi, L. I.; de Rossi, R. H. Appl. Catal., A 2004, 267, 267.
(12) (a) Qiu, H. B.; Yang, C.; Inoue, Y.; Che, S. N. Org. Lett. 2009, 11,
1793. (b) Malta, L. F. B.; Cordeiro, Y.; Tinoco, L. W.; Campos, C. C.; Medeiros,
M. E.; Antunes, O. A. C. Tetrahedron Asymmetry 2008, 19, 1182 and references
cited therein.
(13) Kinen, C. O.; Rossi, L. I.; de Rossi, R. H. Green Chem. 2009, 11, 223.
J. Org. Chem. Vol. 74, No. 18, 2009 7133

Kinen et al.
JOCArticle
TABLE 1. Effect of the Substituents on the Oxidation of Aryl Methyl
SCHEME 1. Schematic Representation of the Two Pathways
Suggested for the Sulfoxidation Reaction
Sulfides 1a-ga
kobs
(10-2
Hammett conver-
σb
FeBr3
(10-2M)
entry substituent
sion (%)c min-1 d
)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
p-NO2
p-CHO
p-NCS
p-Br
þ0.78
þ0.42
þ0.38
þ0.23
0.00
-0.17
-0.27
þ0.78
þ0.42
þ0.38
þ0.23
0.00
100
100
100
100
100
100
100
0
0
50
53
63
1.36
1.56
1.67
1.63
2.15
2.65
2.80
N.R.
N.R.
1.22
1.52
1.22
1.76
1.81
2.5
2.5
2.5
2.5
2.5
2.5
2.5
H
p-CH3
p-OCH3
p-NO2
p-CHO
p-NCS
p-Br
place with the transfer of one electron from the sulfur atom
to the oxidant species.16
H
p-CH3
p-OCH3
-0.17
-0.27
90
100
To gain more insight into the mechanism of the reaction,
the effect of change in the concentration of the reagents
participating in the reaction on the rate constant was studied
with substrates 1-methoxy-4-(methylsulfanyl)benzene 1g
and 4-(methylsulfanyl)benzaldehyde 1b as representative of
compounds with donor and acceptor groups, respectively.
aSolvent: acetonitrile, HNO3 (13%), sulfide 0.509 M. bHammett σ
constant taken from ref 15. cYield calculated by GC analysis. dObserved
rate constant.
Table 1), H (entries 5 and 12, Table 1), p-CH3 (entries 6 and
13, Table 1), and p-OCH3 (entries 7 and 14, Table 1),
respectively.
FIGURE 3. Log of the ratio of the rate constants for the sulfoxida-
tion of 1 vs. oxidation potentials of the sulfides. (Oxidation poten-
tials taken from the following: Watanabe, Y.; Iyanagi,T.; Oae, S.
Tetrahedron Lett. 1980, 21, 3685).
Effect of the Reactant Concentrations on the Rate Con-
stant. We took as the reference state the reactions conducted
with the following concentrations: [substrate] = 0.509 M,
[HNO3]=6.83 ꢀ 10-2 M, [FeBr3]=2.54 ꢀ 10-2 M, solvent
(acetonitrile)=2.6 mL. We performed experiments changing
the concentrations of one of the reagents and keeping all the
others constant.
FIGURE 2. Hammett plot for the oxidation of sulfides 1a-g 0.509
M in acetonitrile with 13% HNO3 and 5% FeBr3. The lines have
been drawn as a visual aid.
A Hammet plot is shown in Figure 2 and it is clear that the
plot is nonlinear. The nonlinearity of Hammett plots may
indicate a change in mechanism or a change in rate determin-
ing step. Reactions that show an upward curvature indicate
that more than one mechanism is operating.14 We suggest
that the reaction takes place by two pathways, one that
involves FeBr3 and one that does not, as shown in Scheme 1.
On the other hand, the plot of the log of the rate constants
vs. the oxidation potentials of the sulfur compounds is linear,
as shown in Figure 3. It has been suggested that this type of
dependence implies that the sulfoxidation reaction takes
Nitric Acid. Several experiments for each substrate were
carried out changing the concentration of nitric acid in a
range that gave measurable reaction rates, the maximum
concentration being 0.1 M. It can be seen in Figure 4 that
in both cases linear dependence of the observed rate constant
with HNO3 concentration was obtained, but the slope
is slightly different, being smaller for the less reactive
compound 1b.
Effect of the Sulfide Concentration. In this case the con-
centration of the substrate was changed from 0.3 to 1 M. The
data are collected in Table 2 and in Figure 5.
(14) Williams, A. Free Energy relationships in Organic and Bio-organic
Chemistry; The Royal Society of Chemistry: Cambridge, UK, 2003; p 166.
(15) Hansch, C.; Leo, A.; Taft, R. Chem. Rev. 1991, 91, 165.
(16) Baciocchi, E.; Intini, D.; Piermattei, A.; Rol, C.; Ruzziconi, R. Gazz.
Chim. Ital. 1989, 119, 649.
7134 J. Org. Chem. Vol. 74, No. 18, 2009

Kinen et al.
JOCArticle
TABLE 3. Effect of the FeBr3 Concentration on the Sulfoxidation Reac-
tion of Sufides 1b and 1ga
kobs (10-2 min-1 c
)
[FeBr3], 10-2
M
% FeBr3
sulfide 1g
sulfide 1b
b
Entry
1
2
3
4
5
0.00
1.26
2.54
5.05
7.57
0.00
2.50
5.00
10.0
15.0
1.81
2.20
2.70
2.14
2.90
no reaction
1.92
1.56
2.03
2.34
a[HNO3] = 6.83 ꢀ 10-2 M, [substrate] = 0.509 M; acetonitrile = 2.6
mL. The yield was 100%, except in entry 1. bPercentage of FeBr3 in
moles with respect to the substrate. cObserved rate constant determined
by UV-vis spectrometry measurement of the disappearance of the
substrate. The values are averages of at least two determinations and
the deviation among them was less than 10%.
FIGURE 4. Effect of the concentration of HNO3 on the rate of the
reaction of compounds 1g (X = p-OCH3; blue line, slope 0.365) and
1b (X = p-CHO; red line, slope 0.269). [Substrate] = 0.509 M,
[FeBr3] = 2.54 ꢀ 10-2 M.
requirement of the formation of a ternary complex that
involves the sulfide and the oxidant on the coordination
sphere of the metal (see below).
It is interesting to note that in both cases the reaction rate
decreases as the concentration of substrate increases in a
nonlinear fashion, and the effect is even more noticeable for
the less reactive substrate.
TABLE 2. Effect of the Concentration of the Sulfide (1b or 1g) on the
Rate Constant for the Sulfoxidation Reaction
kobs (10-2 min-1 c
)
Effect of the FeBr3 Concentration. In Table 3 the data are
collected. It can be seen that for both substrates there was
little variation of the rate when the percentage of FeBr3
changed from 2.5% to 15%. However, the less reactive
substrate 1b required the presence of the metal in order to
proceed, although once it was available the rate did not
depend significantly on its concentration. These results show
that compound 1g can be oxidized without the participation
of the metal, indicating that there is an alternative pathway
as shown in Scheme 1.
Proposed Mechanism. Some relevant mechanisms pro-
posed in the literature for the oxidation of sulfides can be
grouped as follows:17 (a) oxidation initiated by electron
transfer generating cation radicals,18 (b) concerted oxygen
transfer from the oxidant species without the participation of
intermediates,19 (c) mechanism initiated by the formation of
a complex between the nonbonding electrons of the sulfur
atom and the d orbitals of a transition metal,20 and (d)
reaction initiated by the interaction of the nonbonding
electron pair of the sulfur atom and the π* orbitals of the
transition metal oxo-metallic complex.21 The last one is the
more frequent mechanism when the oxidant species are
organic or inorganic peroxides in the presence of transition
metals as catalyst.
b
entry
[sulfide], M
% HNO3
sulfide 1g
sulfide 1b
1
2
3
4
5
0.327
0.409
0.509
0.654
1.089
20.00
16.00
13.00
10.00
6.00
7.11
4.68
2.80
1.75
0.65
5.01
3.08
1.56
0.08
a[HNO3] = 6.83 ꢀ 10-2 M, [FeBr3] = 2.54 ꢀ 10-2 M; acetonitrile =
2.6 mL. The yield was in all cases 100% (determined by GC analysis).
bPercentage of HNO3 in moles with respect to the substrate. cRate
constant determined by measuring the disappearance of the substrate
with time; the value is the average of two or more determinations and the
deviation is less than 10%.
The sulfide oxidation to sulfoxide with use of nitrates as
oxidant agent was reported several times. In 1961 the use of
HNO3 as oxidant was reported22 and the authors emphasize
the use of HNO3 as oxidant for its low cost and the possibility
that the nitric oxides liberated could be reused in future
oxidations. The proposed mechanism involves the direct
reaction of the HNO3 with the sulfide generating NO as
product.
(17) Lee, D.; Gai, H. Can. J. Chem. 1995, 73, 49.
FIGURE 5. Effect of the substrate concentration on the rate of
oxidation of sulfides (data from Table 2).
(18) Lepage, C.; Mihichuk, L.; Lee, D. Can. J. Chem. 2003, 81, 75.
(19) Hanson, P.; Hendrick, R.; Smith, J. Org. Biomol. Chem. 2008, 6, 745.
(20) Lee, D.; Chen, T. J. Org. Chem. 1991, 56, 5346.
(21) (a) Bryliakov, K.; Talsi, E. Chem.;Eur. J. 2007, 13, 8045. (b)
Venkataramanan, N.; Kuppuraj, G.; Rajagopal, S. Coord. Chem. Rev.
2005, 249, 1249.
The observed inhibition of the reaction as the substrate
concentration increases may be attributed to the facility of
the sulfides to coordinate with the metal center and the
(22) Goheen, D.; Bennett, C. J. Org. Chem. 1961, 26, 1331.
J. Org. Chem. Vol. 74, No. 18, 2009 7135

Kinen et al.
JOCArticle
SCHEME 2. Mechanism Proposed for the Oxidation of Sulfides
SCHEME 3. Mechanism Proposed for the Oxidation of Sulfides
with HNO3 and P2O5 Supported on Silica Gel27
by Nitric Acid Catalyzed by KBr28
SCHEME 4. Proposed Mechanism for the Aerobic Oxidation of
Sulfides with Catalytic Quantities of NO2
29
In a later work23 it was determined that for the HNO3
oxidation of sulfides the stoichiometric ratio sulfoxide/oxi-
dant was three in the presence of oxygen and one in its
absence (reactions carried out under N2). In addition, they
found that a product resulting from nitration of the aromatic
ring of the substrate is formed and that the NO2 substituted
product24 does not undergo oxidation under the reaction
conditions employed, leading them to the conclusion that the
oxidant species has an electrophilic nature.
It was reported that the oxidation of sulfides by HNO3
under oxygen is catalyzed by Au halides25 with very good
chemoselectivity in the presence of several other functional
groups. Dialkyl and diaryl sulfides were oxidized with cerium
nitrate in acetic acid16 and it was found that the reactivity
correlated very well with the oxidation potential of the
sulfides, leading the authors to propose that sulfur radical
cations were involved in the mechanism of the reaction.
The combination of BiBr3 and Bi(NO3)3 was proposed as
the catalyst for the aerobic oxidation of sulfides26 and the
formation of a brownish gas suggested the formation of NO2
as reduction product of the nitrate.
Nitric oxide (NO) has been widely studied since it parti-
cipates in many processes of great biological importance.
Several iron complexes are involved in the metabolism of
nitric oxide and many of them contain sulfur as ligands. It is
well-known that S-nitrosothiols are responsible for the
storage and transport of nitric oxide and related compounds
and many NO receptors contain Fe and thiol groups that
form ternary complexes Fe-sulfur-nitrosyl.30
Transition metal bromides were proposed to act as redox
mediators in the sulfoxidation of sulfides catalyzed by
nitrates.9
It was reported that sulfides are selectively oxidized with 1
equiv of 64% HNO3 in the presence of P2O5 supported
on silica gel under solvent-free conditions. The suggested
mechanism is outlined in Scheme 2.27
Recently, a catalytic system that involves nitric acid
supported on silica gel or polyvinylpyrrolidone (PVP) and
sodium or potassium bromide was reported for the oxidation
of sulfides to sulfoxides.28 The authors proposed the
mechanism shown in Scheme 3. It involves the autoxidation
of nitric acid to yield nitronium ions (NO2þ), which in the
presence of bromide ions yield bromonium ions.
On the basis of the literature information mentioned
above and our experimental results, we suggest the mecha-
nism presented in Scheme 5. The proposed mechanism
comprises an initiation step that may or may not involve
FeBr3, depending on the electronic nature of the substrates,
and a catalytic cycle that is common to all substrates. In the
initiation step the only oxidant present in the system is nitric
acid, which must therefore be involved in this step.31 The
relative participation of the pathways involving intermedi-
ates 3A and 3B depends on the electronic nature of the
substituents on the aromatic ring. Compounds with strongly
electron withdrawing groups such as NO2 react only through
the pathway involving intermediate 3B. On the other hand,
compounds with electron donating groups may react via
intermediates 3A and 3B as well.
The sulfides are Lewis bases and their basicity increases
with the presence of electron releasing groups in the aromatic
ring. They can coordinate to the proton of the nitric acid,
transforming the sulfur into an electrophilic species that may
react with the nucleophilic nitrate generating intermediate
3A. The later intermediate gives the sulfoxidation product
and nitrous acid as reduction product derived from nitric
acid.32
The aerobic oxidation of sulfides in the presence of
catalytic quantities of NO2 takes place as shown in Scheme 4,
and involves various intermediates derived from NO2.29 The
authors suggest that the heterolytic dissociation of N2O4 to
-
give NOþNO3 is favored in polar solvents and in the
presence of sulfides in the reaction system.
(23) Ogata, Y.; Kamei, T. Tetrahedron 1970, 26, 5667.
(24) The substrate used in this study (ref 23) was diphenyl sulfide and the
product of the nitration of the substrate was 4-nitrodiphenyl sulfide
(25) Gasparrini, F.; Giovannoli, M.; Misiti, D. Tetrahedron 1983, 19,
1381.
(26) Komatsu, N.; Uda, M.; Suzuki, H. Chem. Lett. 1997, 1229.
(27) Hajipour, A.; Booshki, B.; Ruoho, A. Tetrahedron Lett. 2005, 46,
5503.
(30) Szacilowski, K.; Chmura, A.; Stasicka, Z. Coord. Chem. Rev. 2005,
249, 2408.
(31) It is important to note that the sulfides are not oxidized in the absence
of HNO3.
(28) Zolfigol, M.; Amani, K.; Ghorbani-Choghamarani, A.; Hajjami,
M.; Ayazi, R.; Jafari, S. Catal. Commun. 2008, 9, 1739.
(29) Bosch, E.; Kochi, J. J. Org. Chem. 1995, 60, 3172.
(32) The formation of HNO2 in these reactions was further supported by
the fact that compounds having an aromatic amine produce azo derivatives.
Unpublished results from our lab.
7136 J. Org. Chem. Vol. 74, No. 18, 2009

Kinen et al.
JOCArticle
SCHEME 5. Suggested Mechanism for the Oxidation of Sulfi-
des Promoted by Catalytic Amounts of HNO3 and FeBr3
SCHEME 6. Displacement of NO3- by the Sulfide in the Ternary
Complex
When the volume was small, such that the amount of O2
present was less than the stoichiometric amount, the reaction
did not proceed to completion. The formation of NO2
was evident from the observation of the evolution of a
brownish gas.
The observed inhibition of the reaction when the sulfide
concentration increased may be explained by considering the
reaction shown in Scheme 6 where the sulfide displaces the
-
NO3 from the ternary complex forming an unreactive
complex. In addition, when the substrate concentration
increases, the amount of free HNO3 decreases and therefore
it is not available to initiate the catalytic cycle; this may be the
reason for the decrease in rate in the reactions that do not
require FeBr3.
The formation of a ternary complex between the sulfide,
FeBr3, and nitrate was demonstrated spectroscopically by
measuring the change in absorption with time at the wave-
length maximum of FeBr3 (388 nm). Since the absorbance of
FeBr3 is smaller than that of any sulfide used in this work, in
order to do measurements at this wavelength it was necessary
to dilute the solutions 5 times less than for solutions where
the disappearance of the sulfide was followed (259-314 nm).
Therefore the change in absorption cannot be measured at
the two wavelengths (259-314 and 388 nm) in the same
solution. With use of sulfide 1g (X=p-OCH3), during the
first 25 min there is a significant increase in absorbance at
λ=388 nm. There was no change in the absorbance between
25 and 100 min and thereafter the absorbance decreases
(see Figure 6).
On the other hand, the sulfides that contain electron
attracting groups are probably not basic enough to be
protonated by nitric acid, but they can coordinate to the
Fe center. The coordinated sulfur is an electrophilic center
that is able to react with the nitrate (intermediate 3B) leading
to the sulfoxidation product and nitrous acid (Scheme 5).
It is important to note that the sulfides have negative
electron density on the sulfur atom but the sulfoxides have
positive electron density. The change in charge on the sulfur
may lead to its liberation from the coordination with the
metal and therefore it becomes free to bind another sulfide
molecule. Thus, only catalytic quantities of FeBr3 are needed
for the reaction to proceed.
Both pathways generate nitrous acid, which in the pre-
sence of nitric acid forms the colorless gas N2O4,23 the first
intermediate produced in the catalytic cycle. This gas is in
equilibrium with the product of homolytic dissociation
(2NO2) or heterolytic dissociation (NOþNO3-), the latter
being the predominant species in polar solvents.23
The oxidation of sulfides by nitrosonium ion has been
reported and the proposed mechanism for these reactions
involves the coordination of the NOþ to the sulfide followed
by an electron transfer to generate the radical cation of the
sulfide29 (intermediate 4). The intermediate decomposes
generating the sulfoxide, NO, and NO2. The NO combines
with O2 from the air present in the system yielding NO2(g).
The combination of two molecules of NO2 gives N2O4, which
continues the catalytic cycle.
These results indicated that during the first 25 min a new
species accumulated and it remained unchanged until the
first 100 min; at this time, the sulfoxidation reaction finished
and the absorbance started to decay.
The species absorbing at 388 nm could be a complex
between sulfide and FeBr3, a complex formed between nitric
acid and FeBr3, or a ternary complex formed with the three
compounds. To determine the nature of the species we mixed
solutions of FeBr3 and sulfide at concentrations similar to
those used in the experiment described above and we did not
see any change in the spectrum at λ=388 nm, which discounts
the formation of a complex between sulfide and FeBr3. On the
other hand, when solutions of nitric acid and FeBr3 were
mixed, the absorbance at λ=388 nm decreased in a similar
way to that observed for the reaction described above after
about 100 min (see Figure 6). The observed decrease in
absorbance at 388 nm is attributed to the ionic exchange
between FeBr3 and HNO3 to yield Fe(NO3)3 and HBr.33
The involvement of O2 in the reaction is evident from the
fact that the yield of reactions carried out in closed systems is
very much dependent on the volume of the reaction system.
´
ꢀ
(33) Cotton, F. A.; Wilkinson G. Quımica Inorganica Avanzada, 3rd ed.:
Editorial Limusa: Mexico City, Mexico, 1990; pp 905-924.
J. Org. Chem. Vol. 74, No. 18, 2009 7137

Kinen et al.
JOCArticle
The reaction was followed by UV-vis spectroscopy and
TLC. Both techniques indicated that most of the substrate
remained unchanged for 120 min; however, there were
absorption bands that appeared 5 min after the start of the
reaction with λmax=412 and 435 nm (Figure S2, Supporting
Information). These bands are coincident with those corres-
ponding to the trityl cation.35 Upon the addition of nucleo-
philes such as HO- or I-, the bands at 412 and 435
disappeared, which gives further support to the presence of
the trityl cation in the solution.
For the reactions that require the presence of FeBr3, it may
be that the metal is required in the initiation step as shown in
Scheme 5 and/or in the catalytic cycle. To check for the latter
possibility we carried out an experiment using NO2 as
oxidant for sulfide 1b (see the Supporting Information). In
this case the sulfide was obtained in quantitative yield
indicating that the FeBr3 is not needed for the catalytic cycle.
FIGURE 6. Change in absorbance with time for the sulfoxidation
reaction at 388 nm. Solvent: acetonitrile (2.6 mL), HNO3 (13%),
FeBr3 (5%), sulfide 1g (X = p-OCH3) 0.509 M. The reaction
solution was diluted 100-fold before measurement.
From experiments mentioned above we conclude that a
ternary complex is formed.
In a similar reaction carried out with 1b (X=p-CHO)
there is also an increase in the absorbance at λ=388 nm but
the time required to reach the maximum is approximately
45 min, i.e., almost twice that required for the other
substrate. This observation is consistent with the mecha-
nism suggested where substrates having electron withdraw-
ing groups react only through the pathway that involves
the ternary complex and therefore it is consumed as soon
as it is formed. (See Figure S1 in the Supporting
Information).
Comparing the behavior of the two substrates, we can
see that in the substrate with an electron donor group (1g) the
ternary complex is formed at the beginning of the reaction
and its concentration stays constant until the oxidation
reaction is complete. With the substrate having an electron
withdrawing group (1b) the concentration of the ternary
complex increases slowly and reaches a maximum, but then it
decreases as the reaction proceeds.
Conclusions
It was shown that the oxidation of aryl methyl sulfides
with strong electron donating groups takes place in the
presence of catalytic amounts of HNO3 in acetonitrile. The
same reaction takes place with compounds having electron
withdrawing groups but in this case the presence of a
catalytic amount of FeBr3 is required in addition to the nitric
acid. It is suggested that the mechanism has an initiation step
where HNO2 is formed and that this species enters into a
catalytic cycle that involves the formation of a radical cation
of the sulfide. The oxidant species which is consumed in
stoichiometric amounts is O2 from the air. The Fe(III) metal
is not involved in the catalytic cycle; it only participates in the
initiation step and this may be the reason why the use of
complexes of FeBr3 with several chiral ligands, including
cyclodextrin, did not lead to enantioselective oxidation. The
oxidation reaction of sulfides with catalytic amounts of
FeBr3 and HNO3 is highly chemoselective, proceeds at a
convenient rate at 25 °C, and must be carried out in a closed
system to prevent the gaseous species derived from HNO3
from leaving the system.
These results are consistent with the proposed mechanism
since the main reaction pathway for the substrates with
electron donating group involves the complex formed with
nitric acid. On the other hand, for the substrates with
electron withdrawing groups the main reaction pathway is
that involving the ternary complex; therefore, it is formed
and degraded during the oxidation reaction.
Experimental Section
The formation of radical cations has been postulated
on the basis of the linear dependence of the oxidation of
aryl methyl sulfides with oxidation potentials.29 To find
additional evidence for the participation of radical cations
in the reaction mechanism, we carried out an experi-
ment under the same reaction conditions (5% FeBr3 and
13% HNO3 in acetonitrile) but using triphenyl benzyl
sulfide as a substrate. It is known that the radical
cation of this substrate 5 gives mainly the fragment-
ation products 6 and 7 due to the stability of the trityl
cation 6 (eq 2).34
All reagents were commercial materials and their purity was
checked by NMR spectroscopy with a 400 MHz instrument.
Kinetic Procedures. The reactions were carried out in a bottle
containing a side arm with a septum to take samples. The
required amount of FeBr3 was weighed in the reaction vessel
and the sulfur compound dissolved in 1.3 mL of acetonitrile was
added. The HNO3, also dissolved in 1.3 mL of acetonitrile, was
then added and this time was taken as the zero time for the
reaction. All solutions were kept in a temperature-controlled
bath at 25 °C. The reaction was continuously stirred with a
magnetic stirrer.
~ ~
(34) Penenory, A.; Adam, W.; Arguello, J. J. Org. Chem. 1998, 63, 3905.
€
(35) Acar, M.; Yagci, Y.; Schnabel, W. Polym. Int. 1998, 46, 331.
7138 J. Org. Chem. Vol. 74, No. 18, 2009

Kinen et al.
JOCArticle
To follow the reactions, two different techniques were used:
UV-vis and gas chromatography. For the UV-vis experi-
ments, 2 μL of the reaction was dropped on 10 mL of dichloro-
methane. The dilution stops the reaction and the spectrum was
determined at the selected times. The reactions were followed by
measuring the decrease in absorbance of the substrate at its
wavelength maximum. The data were fitted to a monoexponen-
tial equation and the pseudo-first-order rate constant was
calculated.
For the gas chromatography method, 0.1 mL of the reaction
solution was taken at different reaction times and dropped
on 0.5 mL of dichloromethane containing an internal stan-
dard previously selected. Then 0.5 mL of water was added
to eliminate the FeBr3 and HNO3. The two layers were
separated, the organic layer was dried with anhydrous MgSO4,
and the samples were analyzed by gas chromatography using a
capillary column of ZB-5 (5% phenyl polysiloxane) 30 m ꢀ
0.25 mm ꢀ 0.25 μm.
The two methods gave comparable results, but the UV-vis
method is simpler and therefore most of the reactions were
followed by this method.
Acknowledgment. This research was supported in part by
the Consejo Nacional de Investigaciones Cientıficas y
´
ꢀ
ꢀ
Tecnicas (CONICET), Agencia Nacional de Promocion
ꢀ
Cientıfica y Tecnica (FONCYT), Secretarıa de Ciencia y
´ ´
ꢀ
ꢀ
Tecnica (SECyT) of the Universidad Nacional de Cordoba,
Argentina. C.O.K. is a grateful recipient of a fellowship from
CONICET. We thank Prof. Mino Caira University of Cape
Town, South Africa for the critical reading of the manuscript
and helpful suggestions.
Supporting Information Available: UV-vis spectrum of the
reaction of 6 and experimental details. This material is available
free of charge via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 74, No. 18, 2009 7139