Oxidation of sulfides with a silica-supported peracid in supercritical carbon dioxide under flow conditions: Tuning chemoselectivity with pressure

Article abstract of DOI:10.1002/ejoc.201000826

Supercritical carbon dioxide is a convenient medium for performing the selective oxidation of sulfides 1 to either sulfoxides 2 or sulfones 3 with [2-percarboxyethyl]-functionalized silica (4) under flow conditions. The chemoselectivity of the reaction, which results from the different diffusion rates of sulfide and sulfoxide over the reagent bed, can be controlled by adjusting the pressure and the hydration of the silica surface as both the solvating power of the mobile phase and the surface activity of the stationary phase determine the migration rates of sulfide 1 and sulfoxide 2 over the supported peroxide. The results elucidate the impact of surface phenomena on the course of chemical reactions carried out under flow conditions. The oxidation of sulfides 1 with hydrated [2-percarboxyethyl]-functionalized silica (4) in scCO₂ under flow conditions can be tuned to give either sulfoxides 2 or sulfones 3 by adjusting the pressure.

Full text of DOI:10.1002/ejoc.201000826

FULL PAPER
DOI: 10.1002/ejoc.201000826
Oxidation of Sulfides with a Silica-Supported Peracid in Supercritical Carbon
Dioxide under Flow Conditions: Tuning Chemoselectivity with Pressure
Rossella Mello,[a,b] Andrea Olmos,[a] Ana Alcalde-Aragonés,[a] Alba Díaz-Rodríguez,[a]
María Elena González Núñez,*[a] and Gregorio Asensio[a]
Keywords: Supercritical CO2 / Oxidation / Surface diffusion / Supported catalysts / Adsorption
Supercritical carbon dioxide is a convenient medium for per-
forming the selective oxidation of sulfides 1 to either sulfox-
ides 2 or sulfones 3 with [2-percarboxyethyl]-functionalized
silica (4) under flow conditions. The chemoselectivity of the
reaction, which results from the different diffusion rates of
sulfide and sulfoxide over the reagent bed, can be controlled
by adjusting the pressure and the hydration of the silica sur-
face as both the solvating power of the mobile phase and the
surface activity of the stationary phase determine the mi-
gration rates of sulfide 1 and sulfoxide 2 over the supported
peroxide. The results elucidate the impact of surface phen-
omena on the course of chemical reactions carried out under
flow conditions.
catalysts, reagents, products or transition states, which can
alter the reaction course.[3] In order to explore this possibil-
ity, we studied the oxidation of sulfides with a silica-sup-
ported peracid in scCO2 under flow conditions. In this reac-
tion, the peracid performs two consecutive oxygen-transfer
steps to the substrate [Equation (1)], the first being faster
than the second. The distinct interactions of the competing
sulfide 1 and sulfoxide 2 with the silica surface under dif-
ferent reaction conditions could overcome this kinetic pref-
erence and thereby change the reaction chemoselectivity.
Introduction
Supercritical carbon dioxide[1] (scCO2) is a convenient
medium for performing heterogeneous chemical reactions[2]
because its low surface tension, low viscosity, and high dif-
fusivity[1] allow the substrate solution to reach the reactive
sites on the active surface much better than conventional
liquid solvents. These properties of scCO2, along with its
readily accessible critical conditions (Tc = 31.0 °C; Pc =
73.8 bar), benign character, low cost, and the existence of a
technology platform for the application of scCO2 in large-
scale continuous-flow processes have extended the potential
of this medium for application as a green solvent for chemi-
cal reactions.
The density and transport properties of scCO2 depend
on the pressure, which allows for continuous adjustment of
the solvating ability of the medium without changing the
nature of the solute–solvent interactions.[1] This unique tun-
able character of supercritical fluids has proven most useful
for solid nucleation, extractive and chromatographic pro-
cesses in scCO2 with applications in the food and pharma-
ceutical industries.[1] Heterogeneous reactions in which so-
lid matrices are used as either catalysts or insoluble sup-
ports for reagents or catalysts should also be sensitive to
pressure in scCO2, as changes in the solvating ability of the
solvent modify the interactions of the active surface with
(1)
We report herein that the oxidation of sulfides 1 with
hydrated [2-percarboxyethyl]-functionalized silica (4) in
scCO2 under flow conditions can be tuned to selectively
yield sulfoxides 2 or sulfones 3 by adjusting the pressure.
The results are interpreted in terms of the distinct transport
of the competing sulfide 1 and sulfoxide 2 over the silica
surface under the specific reaction conditions, which de-
pends on the solvating power of scCO2 and the adsorbing
ability of the silica surface. The reaction chemoselectivity
observed in conventional solvents under flow conditions
differs from that obtained in scCO2, suggesting that surface
diffusion plays a major role in the migration of the sub-
strates along the reagent bed. The results show the impor-
tance of surface phenomena in reactions carried out under
flow conditions over supported reagents or catalysts, and
illustrate the application of the tunable character of scCO2
to control the reaction course.
[a] Departamento de Química Orgánica, Universidad de Valencia,
Avda. Vicente Andrés Estellés s.n., 46100 Burjassot, Valencia,
Spain
Fax: +34-96-3544939
E-mail: elena.gonzalez@uv.es
[b] Fundación General de la Universidad de Valencia (FGUV),
Amadeo de Saboya 4, 46010 Valencia, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000826.
View this journal online at
wileyonlinelibrary.com
6200
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6200–6206

Oxidation of Sulfides with an SiO2-Supported Peracid in Supercritical CO2
Results and Discussion
Silica-supported peracid 4 was prepared[4] by acid hy-
drolysis of [2-cyanoethyl]-functionalized silica, obtained by
a standard sol–gel method, followed by treatment of the
resulting [2-carboxyethyl]-functionalized silica with 70%
hydrogen peroxide in acidic medium (Scheme 1). Lyophili-
zation of the solid reagent yielded a 22–26% w/w hydrated
material. Analysis of the hybrid materials by conventional
1H NMR apectroscopy according to a reported procedure[5]
showed a load of 2.5–3.5 mmol of organic ligands per gram.
Iodometric titration of [2-percarboxyethyl]-functionalized
silica (4) showed loads of 0.8–1.7 mmol of peracid/g. The
silica-supported peracid 4 was stored at –20 °C for weeks
without a noticeable loss of peroxidic titer. The reagent can
be recycled[2a,4] by reaction of the hybrid material with
aqueous hydrogen peroxide under acidic conditions.
(2)
Table 1. Oxidation of sulfides 1 with the silica-supported peracid 4
in scCO2 under flow conditions.
Scheme 1. Synthesis of the supported peracid (4).[4]
[a] Reactions at 250 bar, 40 °C and flow rate 0.10–0.12 mL
scCO2 min–1; load of peroxide: 1 mmol of peracid g–1; hydration of
4: 25% w/w; molar ratio of 1/4 was 1:2. [b] Reactions in scCO2 at
100 bar, with anhydrous 4, and a 1:1 molar ratio of 1/4; products
were eluted with a solution of methanol in scCO2.
Oxidations in scCO2
The first series of experiments was carried out with a 1:2
initial molar ratio of sulfide 1/peracid 4 in order to verify
the reactivity of the supported peracid towards sulfides 1
The results we had previously obtained[2a,2k] in the oxi-
and sulfoxides 2. Oxidations were carried out by flowing dation of organic substrates with supported peracids in
scCO2 (0.10–0.12 mL scCO2 min–1) at 250 bar and 40 °C scCO2 under flow conditions showed that an excess of oxi-
for 3 h through a reservoir containing the substrate 1 dant was required to achieve the complete conversion of the
(0.5 mmol), and then through a column packed with 25% substrate. This fact was attributed to the shorter contact
w/w hydrated reagent 4 [Equation (2)]. The system was de- time of the substrate solution with the supported reagent in
pressurized through a micrometric valve and the reaction the flow reactions compared to the batch reactions. The
products were collected in a trap cooled to –78 °C. The resi- results in Table 1 reveal that the reactions of both sulfides
due obtained was dissolved in dichloromethane and ana- 1 and sulfoxides 2 with the supported peracid 4 are fast
lyzed by gas chromatography. Products were characterized enough to obtain the corresponding sulfone 3 as the main
by gas chromatography/mass spectrometry, 1H and 13C product in all cases. The oxygen transfer to the substrates
NMR spectroscopy. Mass balances were correct in all cases, 1, though generally high, was in no case quantitative. The
indicating that the reaction products were not retained on substrate conversion and the sulfone 3/sulfoxide 2 ratio ob-
the silica surface under these conditions. The results are tained for the different sulfides 1a–g indicate that the reac-
shown in Table 1 (Method A). The silica-supported reagent tion efficiency depends on the substrate structure (Table 1,
recovered from the column had a white color and a loose Method A). Thus, sulfides 1 with electron-withdrawing sub-
appearance in all cases. Iodometric titration of the con- stituents showed lower conversions and lower sulfone/sulf-
sumed reagent showed a loss of peroxidic titer correspond- oxide ratios than those substrates with electron-releasing
ing to the oxygen-transfer performed to the substrate. The substituents (Table 1). In addition, sterically demanding
control experiments carried out with a 2:1 initial molar ra- substituents on sulfide 1 led to less efficient and less selec-
tio of methyl phenyl sulfide (1a)/peracid 4 led to the com- tive reactions. For instance, the substrate conversion and
plete consumption of the peracid, showing that all the per- the sulfone 3/sulfoxide 2 ratios were 95% and 29.8 for
oxidic ligands on the silica surface are accessible to the sub- methyl phenyl sulfide (1a), 93% and 5.8 for dibutyl sulfide
strate.
(1g) and 85% and 4.4 for diphenyl sulfide (1f), respectively.
Eur. J. Org. Chem. 2010, 6200–6206
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6201

M. E. González Núñez et al.
FULL PAPER
These results are in agreement with the reaction mechanism bonded to the acidic silica surface, then sulfide 1, which is
proposed for the oxygenation of sulfides 1 and sulfoxides 2 less retained because of its weaker Lewis-base character,
with organic peracids in solution.[6,7]
would be transported preferentially towards the unreacted
The oxidation of methyl phenyl sulfide (1a) with a stoi- peroxide. As a result, sulfoxide 2 would be the main product
chiometric amount of 25% w/w hydrated oxidant 4 in of the reaction. Thus, the efficiency of the mobile phase in
scCO2 was carried out at 250 bar and 40 °C under flow con- transporting sulfoxide 2 along the reactive stationary phase
ditions (0.11 mL scCO2 min–1), according to the same pro- would determine the chemoselectivity of the reaction. As
cedure described above [Equation (3)]. Analysis of the or- scCO2 is a tunable solvent, which permits increasing the
ganic material recovered in the cold trap showed a substrate number of solute–solvent interactions by raising the pres-
conversion of 49% and a 3a/2a ratio of 2.3. A thorough sure,[1] the reaction chemoselectivity should be pressure-de-
washing of the solid reagent with dichloromethane did not pendent.
allow the recovery of additional amounts of organic mate-
rial.
The nature of the silica surface contributes also to estab-
lishing the magnitude of the adsorption equilibrium con-
stant.[8] Our previous results in the oxidation of organic
substrates with supported peroxides[2a,2k] have shown that
coordination of water molecules to the acidic sites on the
active surface diminishes its acidic strength, and thereby its
ability to adsorb polar substrates. The 1H NMR analysis of
the 25% hydrated silica-supported peracid 4 (ca. 14 mmol
of water per gram of hybrid material) showed a load of
2.5 mmol of organic ligands and 5.6 mmol of silanol groups
per gram of material.[5] The hydration layer in this case con-
sists of ca. 2 molecules of water for each acidic position
(3)
This result does not follow the reactivity pattern of sul- (silanol, carboxylic acid or peracid). Further hydration of
fides 1 and sulfoxides 2 towards organic peracids in solu- this solid material provides less strongly bonded water mo-
tion,[6,7] in which sulfoxides 2 are the major products in the lecules and diminishes accordingly the ability of the silica
reaction of sulfides 1 with stoichiometric amounts of per- surface to adsorb polar substrates.
acid. Considering that the oxidation of methyl phenyl sul-
Since the solvating ability of scCO2 and the adsorbing
fide (1a) to methyl phenyl sulfone (3a) with the organic per- ability of the silica surface are tunable by changing the pres-
acid 4 implies two successive oxygen transfer steps [Equa- sure or by hydration, respectively, we can predict that higher
tion (1)] and that methyl phenyl sulfoxide (2a) is less reac- pressures and higher hydration of the supported reagent 4
tive than sulfide 1a towards 4,[6,7] the formation of 3a as the would favor the formation of sulfone 3 in these reactions.
main product strongly suggests that the reaction conditions
determine a competitive advantage for sulfoxide 2a to react the chemoselectivity of the reaction, we examined the effect
with peracid 4 when compared to sulfide 1a. of pressure on the oxidation of methyl phenyl sulfide (1a)
In order to verify the influence of these parameters on
The main features of the oxidation of methyl phenyl sul- in scCO2 under flow conditions with differently hydrated
fide (1a) under continuous flow over a supported peracid 4 supported peracids 4. The oxidation of methyl phenyl sul-
are: (i) substrate and products migrate along the reagent fide (1a) was carried out by flowing scCO2 (0.12 mL
bed driven by the flow vector, (ii) the migration rates for scCO2 min–1) at 40 °C and at pressures ranging from 100 to
sulfide 1 and sulfoxide 2 are determined by their interac- 300 bar, for 3 h through a reservoir containing the substrate
tions with the silica surface, which are stronger for the latter (0.5 mmol), and then through a column packed with the
given its higher basicity and polarity, (iii) the intermediate silica-supported peracid 4 (initial molar ratio sulfide/per-
sulfoxide 2 competes with sulfide 1 for peracid 4, and (iv) acid 1:1). The reactions were carried out with anhydrous,
peracid 4 is consumed in the oxidation steps. In a simplified 25% and 50% w/w hydrated supported peracid 4. The an-
view, the migration of the substrates along the reagent bed hydrous peracid 4 was prepared by drying the solid reagent
would involve a series of successive desorption/convection under vacuum at room temperature until it reached a con-
transport/adsorption steps.[8] Thus, if the intermediate sulf- stant weight. Hydration was performed by adding the de-
oxide 2 does not significantly adsorb on the silica surface, sired amount of water to this reagent and then allowing the
the concentration of sulfide 1 in the stream of scCO2 enter- mixture to equilibrate at 4 °C for 12 h. The products were
ing the reagent bed would rapidly be converted to the corre- collected by depressurization in a trap cooled to –80 °C and
sponding sulfoxide 2 as the solution advances over the sup- by moisturizing the solid reagent recovered from the col-
ported peracid 4. The flow of scCO2 would transport the umn and then washing thoroughly with dichloromethane.
sulfoxide 2 toward a region of the reagent bed with a higher Substrate conversions and product distributions were deter-
load of unreacted peroxidic ligands, where it would be oxid- mined by GC analysis of the combined residues. Mass bal-
ized in the absence of competing sulfide 1. Under these con- ances were satisfactory in all the cases.
ditions, sulfone 3 would be the main product of the reac-
The results, which are shown in Figure 1, fit nicely with
tion. Conversely, if the sulfoxide 2, formed by oxygen trans- the model described above. Thus, the anhydrous silica-sup-
fer from the peracid to the sulfide 1, remains hydrogen- ported peracid 4 strongly retains sulfoxide 2a over the
6202
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6200–6206

Oxidation of Sulfides with an SiO2-Supported Peracid in Supercritical CO2
whole range of pressures used and, consequently, the ratio scCO2 and the physicochemical processes taking place on
3a/2a is consistently low along the series. In this case, the the silica surface. The experiments under flow conditions
main reaction product was sulfoxide 2a, which was com- were carried out at 40 °C by eluting a 0.02  solution of 1a
pletely desorbed from the silica surface by flowing a solu- in dichloromethane or hexane through a column packed
tion of methanol in scCO2 over the supported reagent once with reagent 4 in the same solvent at 0.12 mLmin–1. The
the reaction had been completed. The series of reactions solid reagent in the column was thoroughly washed with
with the 25% w/w hydrated reagent 4 showed that an in- the solvent, and the solution was analyzed by gas
creased pressure of up to 250 bar reversed the selectivity of chromatography. The results are shown in Scheme 2.
the reaction. A further increase of the pressure to 300 bar
enhanced the overoxidation of the sulfide up to 3a/2a = 2.5.
This trend became more evident with the 50% w/w hydrated
silica-supported peracid 4 (ca. 7 water molecules for each
acidic ligand). In this case, the selectivity 3a/2a of the reac-
tion could be tuned from 0.34 to 12.6 by increasing the
pressure from 100 bar to 300 bar. In these experiments, the
reaction products were collected exclusively from the cold
trap.
Scheme 2. Oxidation of sulfide 1a with 50% w/w hydrated silica-
supported peracid 4 in conventional solvents under flow condi-
tions.
The reactions in the liquid solvents under flow condi-
tions gave methyl phenyl sulfoxide (2a) as the main reaction
product (Scheme 2), in clear contrast to the results obtained
in scCO2 at high pressures (Figure 1). These experimental
results cannot be interpreted in terms of a simplified trans-
port mechanism exclusively involving successive desorption/
convection transport/adsorption steps. In fact, this model
predicts that increasing the solvating ability of the solvent
will diminish the adsorption equilibrium constant for sulf-
oxide 2a, thus facilitating its convection transport towards
the unreacted peracids on the stationary phase and increas-
ing sulfone 3a formation Accordingly, sulfone 3a formation
should be more efficient not only in liquid solvents than
Figure 1. Oxidation of methyl phenyl sulfide (1a) in scCO2 under
flow conditions (0.12 mL scCO2 min–1) at different pressures with
anhydrous (diamonds), 25% w/w (squares) and 50% w/w (tri-
angles) supported peracid 4.
in scCO2, but also in dichloromethane than in n-hexane.
However, the reverse is observed experimentally, indicating
the involvement of an alternative transport mechanism for
sulfoxide 2a that is favored in the less-solvating media.
The results can be rationalized by considering that the
migration of sulfoxide 2a along the reagent bed also takes
place by surface diffusion,[8,9] in which the adsorbate mi-
grates by successive shifts from its initial adsorption site to
an adjacent acidic ligand without leaving the surface, pro-
moted by thermal activation or collisions with the solvent.
Hydration of the silica surface facilitates surface diffusion
as the water molecules lower the acidity of the surface li-
gands and provide a continuous protic layer that intercon-
nects the acidic sites on the silica surface.[9] Under these
conditions, the activation energy for surface diffusion is
lower than for desorption,[8,9] because the adsorbed sub-
strate remains hydrogen-bonded to the silica surface in the
former, while it breaks the hydrogen-bonding interaction as
it is transferred into the solution in the latter. Accordingly,
the migration of the adsorbed sulfoxide 2a along the silica
According to these results, the selective oxidation of sul-
fides 1 to the corresponding sulfoxides 3 was performed by
flowing scCO2 (0.12–0.10 mL scCO2 min–1) at 100 bar and
at 40 °C for 3 h through a reservoir containing the substrate
1 (0.5 mmol), and then through a column packed with
1 equiv. of the anhydrous silica-supported peracid 4. Sulfox-
ides 3 were recovered in the cold trap by flowing a solution
of methanol in scCO2 at 250 bar over the reagent bed for
1 h (Table 1, Method B). The process was successful even
for the most reactive sulfides.
Oxidations in Conventional Solvents
Oxidation of methyl phenyl sulfide (1a) with 50% w/w surface would take place by successive adsorption/surface
hydrated silica-supported peracid 4 in conventional solvents diffusion/desorption/convection transport steps. Adsorp-
provides further information on the solvent properties of tion/desorption equilibrium and rate constants determine
Eur. J. Org. Chem. 2010, 6200–6206
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6203

M. E. González Núñez et al.
FULL PAPER
both the efficiency of the long-range convection transport
and the residence time of the substrate on the silica surface,
while the surface diffusion coefficient determines the dis-
tance covered by the substrate during its stay on the silica
surface. In competitive reactions with silica-supported rea-
gents or catalysts, the substrate that better adsorbs and dif-
fuses on the reagent surface will have a competitive advan-
tage in colliding with the reagent or catalyst immobilized
on the silica surface.
The chemoselectivity observed in the oxidation of sulfide
1a with 50% w/w hydrated silica-supported peracid 4 in dif-
ferent solvents and under flow conditions results from the
balance of different surface phenomena under the specific
reaction conditions. The adsorption equilibrium constant is
larger for sulfoxide 2a than for sulfide 1a, and the residence
time at the silica surface is longer for the former. As the
solvating ability of the solvent diminishes, both parameters
increase. Consequently, less-solvating media enhance the
competitive advantage of sulfoxide 2a over sulfide 1a to re-
act with the silica-supported peroxide 4. This is the pattern
observed experimentally (Figure 1, Scheme 2). For the an-
hydrous silica-supported peracid 4, the lack of a continuous
protic network on the silica surface minimizes the surface
diffusion coefficient for sulfoxide 2a, because the stronger
acid character of the active surface and the distance be-
tween the ligands make the required hydrogen-bond inter-
changes difficult.
Conclusions
The oxidation of sulfides 1 with hydrated [2-percarboxy-
ethyl]-functionalized silica (4) in scCO2 under flow condi-
tions can be tuned to give either sulfoxides 2 or sulfones 3
by adjusting the pressure. The reaction chemoselectiviy,
which results from the different diffusion rates of the com-
peting sulfide 1 and sulfoxide 2 over the reagent bed, de-
pends on the surface activity of the supported reagent and
the solvating power of scCO2. The results indicate the im-
portance of surface phenomena in reactions carried out un-
der flow conditions over supported reagents or catalysts,
and the potential of the tunable transport properties of
scCO2 for the control of the selectivity in these cases. This
heterogeneous procedure improves the product isolation
step with regard to the conventional homogeneous reaction
conditions, as it involves a simple depressurization and avo-
ids the wastes associated with the carboxylic acid neutral-
ization. The silica-supported peracid 4 can be recycled by
treatment with hydrogen peroxide under acidic conditions.
Experimental Section
CAUTION! The experiments described in this paper involve the
use of relatively high pressures and require equipment with the ap-
propriate pressure rating.
General: [2-Percarboxyethyl]-functionalized silica (4) was prepared
as described elsewhere.[2a,4] Dichloromethane was purified accord-
ing to described procedures.[12] The glassware used in the reactions
with 70% hydrogen peroxide was carefully cleaned and washed be-
fore use with a solution of EDTA in ultrapure water (0.25 gL–1) to
remove metal traces. The peroxidic content of the supported rea-
gent was determined by standard iodometric titration. Hydration
of the silica-supported peracid 4 was performed by adding to a
weighed sample of anhydrous functionalized silica the desired
amount of ultrapure water, and allowing the mixture to equilibrate
at 4 °C in a tightly closed vial for 12 h. The high-pressure equip-
ment consisted of a 250-mL AISI 316 stainless-steel jacketed auto-
clave, a diaphragm pump (Orlita MHS 30/8) with a maximum theo-
retical flow of 8.44 L/h of liquid CO2, and a set of high-pressure
valves, pressure and temperature probes and safety rupture-discs
suitably placed to control the flow of CO2 along the system.
Surface diffusion under flow conditions is directional as
adsorbed sulfoxide 2a diffuses along the reagent bed driven
by collisions with the solvent. Accordingly, the effect of
pressure on the reaction chemoselectivity observed in the
oxidation of sulfide 1a with the 50% w/w hydrated sup-
ported peracid 4 in scCO2 reveals that the efficiency of the
surface transport increases as the solvent density increases.
The experimental data also show that surface diffusion is
more efficient in scCO2 at high pressures than it is in con-
ventional solvents. This result can be attributed to the high
diffusivity of scCO2, which minimizes solvent stagnation in
the less accessible regions of the silica surface and improves
the transport of the substrates by both convection and sur-
face diffusion along the reagent bed.
The kinetic effect of scCO2 on the relative reaction rates
of sulfide and sulfoxide with organic peracids has not been
determined in this study. The higher reactivity of sulfides 1
over sulfoxides 2 toward organic peracids observed in solu-
tion arises[6b,7] from the differences in the solvation of sub-
strates and transition states. This kinetic effect is stronger
Oxidation of Sulfides (1) with [2-Percarboxyethyl]-Functionalized
Silica (4) in scCO2. Method A. General Procedure: Methyl phenyl
sulfide (1a) (0.12 mL, 1 mmol) was placed in a U-shaped stainless-
steel tubing, which was connected through suitable fittings to a
stainless steel 8 mm ID column packed with 25% w/w hydrated
silica-supported peracid 4 (2 g, 1.0 mmolg–1, 2 equiv.). Two filters
for polar solvents than for apolar solvents, although the placed at either end of the column prevented displacement of the
trend is maintained even in CCl4 or n-hexane.[6b] By con-
solid reagent throughout the experiment. The outlet of the column
was connected to a high-pressure micrometric valve which was con-
[10]
sidering that the quadrupole moment of CO2
permits
nected to a trap cooled with a dry-ice bath through a 1/8 inch Tef-
lon tubing. The pressure in the trap was equilibrated with a flow
of nitrogen. The inlet of the substrate reservoir was connected with
a high-pressure valve to a 250 mL autoclave set at 40 °C. The auto-
clave was closed, charged with CO2 and pressurized to 250 bar.
Then both the substrate reservoir and the column were placed into
a water bath heated to 40 °C, and the system was allowed to pres-
surize by carefully opening the inlet valve. The stroke volume of
the pump and the aperture of the high-pressure micrometric outlet
Lewis acid–Lewis base interactions with polar functional
groups and imparts “polar” attributes to scCO2 as a sol-
vent,[11] and that reactions take place in the polar protic
environment of the silica surface, the kinetic effect of scCO2
on the oxidation of sulfide 1a with the silica-supported per-
acid 4 is expected to play a minor role in the preferential
formation of methyl phenyl sulfone (3a) under our reaction
conditions.
6204
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6200–6206

Oxidation of Sulfides with an SiO2-Supported Peracid in Supercritical CO2
W. Leitner), Wiley-VCH, Weinheim, New York, 1999; d) Chem.
Rev. 1999, 99, thematic issue; e) M. McHugh, V. Krukonis,
Supercritical Fluid Extraction, 2nd ed., Butterworth-Heinem-
ann, Boston, 1994.
valve were regulated to achieve steady continuous flow conditions
at 250 bar. The CO2 flow at the outlet of the system was monitored
with a bubble flow meter. The system was left to operate for 3 h.
The inlet valve was then closed, and the system was allowed to
[2]
See, for instance: a) R. Mello, A. Olmos, J. Parra-Carbonell,
M. E. González-Núñez, G. Asensio, Green Chem. 2009, 11,
994–999; b) F. Jutz, J. D. Grunwaldt, A. Baiker, Ind. Eng.
Chem. Res. 2008, 47, 4561–4585; c) F. Zayed, L. Greiner, P. S.
Schulz, A. Lapkin, W. Leitner, Chem. Commun. 2008, 79–81;
d) T. Seki, J. D. Grunwaldt, N. vanVegten, A. Baiker, Adv.
Synth. Catal. 2008, 350, 691–705; e) Z. Hou, N. Theyssen, W.
Leitner, Green Chem. 2007, 9, 127–132; f) M. I. Burguete, A.
Cornejo, E. García-Verdugo, M. J. Gil, S. V. Luis, J. A.
Mayoral, V. Martínez-Merino, M. Sokolova, J. Org. Chem.
2007, 72, 4344–4350; g) P. Clark, M. Poliakoff, A. Wells, Adv.
Synth. Catal. 2007, 349, 2655–2659; h) T. Seki, J. D. Grun-
waldt, A. Baiker, Chem. Commun. 2007, 3562–3564; i) P. Ste-
phenson, B. Kondor, P. Licence, K. Scovell, S. K. Ross, M. Pol-
iakoff, Adv. Synth. Catal. 2006, 348, 1605–1610; j) M. Caravati,
J. D. Grunwaldt, A. Baiker, Appl. Catal. A 2006, 50, 50–56; k)
M. E. González-Núñez, R. Mello, A. Olmos, G. Asensio, J.
Org. Chem. 2006, 71, 6432–6436; l) M. E. González-Núñez, R.
Mello, A. Olmos, R. Acerete, G. Asensio, J. Org. Chem. 2006,
71, 1039–1042.
a) A. Salameh, A. Baudouin, J.-M. Basset, C. Copéret, Angew.
Chem. Int. Ed. 2008, 47, 2117–2120; b) R. Berthoud, A. Bau-
douin, B. Fenet, W. Lukens, K. Pelzer, J.-M. Basset, J.-P. Candy,
C. Copéret, Chem. Eur. J. 2008, 14, 3523–3526; c) D. A. Ruddy,
T. D. Tilley, Chem. Commun. 2007, 3350–3352; d) R. L. Brut-
chey, D. A. Ruddy, L. K. Andersen, T. D. Tilley, Langmuir
2005, 21, 9576–9583; e) R. Anwander, Chem. Mater. 2001, 13,
4419–4438; f) J. Bu, H.-K. Rhee, Catal. Lett. 2000, 66, 245–
249; g) M. B. Dámore, S. Schwarz, Chem. Commun. 1999, 121–
122; h) J.-M. Basset, F. Lefebvre, C. Santini, Coord. Chem. Rev.
1998, 178–180, 1703–1723; i) A. Corma, M. Domine, J. A.
Gaona, J. L. Jordá, M. T. Navarro, F. Rey, J. Pérez-Pariente, J.
Tsuji, B. McCulloch, L. T. Nemeth, Chem. Commun. 1998,
2211–2212; j) T. Tatsumi, K. A. Koyano, N. Igarashi, Chem.
Commun. 1998, 325–326; k) S. A. Holmes, F. Quignard, A.
Choplin, R. Teissier, J. Kervennal, J. Catal. 1998, 176, 182–
191.
depressurize. The trap was warmed to room temperature, and the
colorless residue was dissolved in deuterated chloroform and ana-
1
lyzed with the aid of GC, GC–MS, H, and 13C NMR techniques.
The solid reagent recovered from the column was washed four
times with 15 mL of dichloromethane in a round-bottomed flask
with magnetic stirring. The filtered solution was analyzed by means
of GC and then concentrated under vacuum at 0 °C.
Oxidation of Sulfides (1) with [2-Percarboxyethyl]-Functionalized
Silica (4) in scCO2. Method B. General Procedure: Methyl phenyl
sulfide (1a) (0.12 mL, 1 mmol) was placed in a 0.2 mL loop of a
Rheodyne valve, which was connected through suitable fittings to
a stainless steel 8 mm ID column packed with anhydrous silica-
supported peracid 4 (0.83 g, 1.2 mmolg–1, 1 equiv.). Two filters
placed at either end of the column prevented displacement of the
solid reagent throughout the experiment. The outlet of the column
was connected to a high-pressure micrometric valve, which was
connected to a trap cooled with a dry-ice bath through a 1/8 inch
Teflon tubing. The pressure in the trap was equilibrated with a flow
[3]
of nitrogen. The inlet of the Rheodyne valve was connected with a
high-pressure valve to a 250 mL autoclave set at 40 °C. The auto-
clave was closed, charged with CO2 and pressurized to 100 bar.
Both the Rheodyne loop and the column were then placed into a
water bath heated to 40 °C, and the system was allowed to pressur-
ize by carefully opening the inlet valve. The stroke volume of the
pump and the aperture of the high-pressure micrometric outlet
valve were regulated to achieve steady continuous flow conditions
at 100 bar. The CO2 flow at the outlet of the system was monitored
with a bubble flow meter. The substrate was injected, and the sys-
tem was left to operate for 3 h. Afterwards, the pressure was raised
to 250 bar, and the CO2 flow was fixed again to 0.10–0.12 mL of
scCO2 min–1. Six portions of 0.2 mL of methanol were successively
injected through the Rheodyne valve at 10 min intervals. The sys-
tem was then allowed to depressurize. The trap was warmed to
[4]
a) A. Lambert, J. A. Elings, D. J. Macquarrie, G. Carr, J. H.
Clark, Synlett 2000, 7, 1052–1054; b) D. J. Macquarrie, D. B.
Jackson, J. E. G. Mdoe, J. H. Clark, New J. Chem. 1999, 23,
539–544; c) J. A. Elings, R. Ait-Meddour, J. H. Clark, D. J.
Macquarrie, Chem. Commun. 1998, 2707–2708.
R. Mello, A. Olmos, T. Varea, M. E. González-Núñez, Anal.
Chem. 2008, 80, 9355–9359.
a) J. Skerjanc, A. Regent, B. Plesnicar, J. Chem. Soc., Chem.
Commun. 1980, 1007–1009; b) R. Curci, R. A. DiPrete, J. O.
Edwards, G. Modena, J. Org. Chem. 1970, 35, 740–745; c) C. G.
Overberger, R. W. Cummins, J. Am. Chem. Soc. 1953, 75,
4250–4254.
room temperature, and the solution was concentrated under vac-
uum. The colorless residue was dissolved in deuterated chloroform
and analyzed with the aid of GC, GC–MS, 1H, and 13C NMR
techniques. The solid reagent recovered from the column was
washed four times with 15 mL of dichloromethane in a round-bot-
tomed flask with magnetic stirring. The filtered solution was ana-
[5]
lyzed by means of GC and then concentrated under vacuum at
0 °C.
[6]
Supporting Information (see footnote on the first page of this arti-
1
cle): GC–MS, H and 13C NMR spectra of the reaction mixtures.
[7]
R. D. Bach, J. E. Winter, J. J. W. McDouall, J. Am. Chem. Soc.
1995, 117, 8586–8593.
Acknowledgments
[8]
I. N. Levine, Physical Chemistry, 4th ed., McGraw-Hill, New
York, 1995.
Financial support from the Spanish Dirección General de In-
vestigación and Fondos FEDER (CTQ2007-65251/BQU) and Con-
solider Ingenio 2010 (CSD2007-00006), is gratefully acknowledged.
A. O, and A. A. A. thank the Spanish Ministerio de Educación y
Ciencia for fellowships. We thank Solvay Química S.L. for a gener-
ous gift of 70% hydrogen peroxide. We also thank the SCSIE (Uni-
versidad de Valencia) for allowing us access to their instruments
and facilities.
[9]
a) S.-C. Li, L.-C. Chu, X.-Q. Gong, U. Diebold, Science 2010,
328, 882–884; b) F. Pizzitutti, M. Marchi, F. Sterpone, P. J. Ros-
sky, J. Phys. Chem. B 2007, 111, 7584–7590; c) R. Valiullin, P.
Kortunov, J. Kärger, V. Timoshenko, J. Phys. Chem. B 2005,
109, 5746–5752; d) J. P. Pevlin, N. Uras, J. Sadlej, V. Buch, Na-
ture 2002, 417, 269–271; e) E. Gedat, A. Schreiber, G. H. Find-
enegg, I. Shenderovich, J.-H. Limbach, G. Buntkowsky, Magn.
Reson. Chem. 2001, 39, 149–157; f) M. Bjelopavlic, P. K. Singh,
H. El-Shall, B. M. Moudgil, J. Colloid Interface Sci. 2000, 226,
159–165.
[1] a) W. Leitner, Acc. Chem. Res. 2002, 35, 746–756; b) S. Wells,
J. M. DeSimone, Angew. Chem. Int. Ed. 2001, 40, 519–527; c)
Chemical Synthesis Using Supercritical Fluids (Eds.: P. Jessop,
[10]
a) C. Graham, D. A. Imrie, R. E. Raab, Mol. Phys. 1998, 93,
49–56; b) J. H. Williams, Acc. Chem. Res. 1993, 26, 593–598;
Eur. J. Org. Chem. 2010, 6200–6206
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
6205

M. E. González Núñez et al.
FULL PAPER
c) A. D. Buckingham, R. L. Disch, Proc. R. Soc. London 1963,
A273, 275–289.
102, 7860–7863; g) J. F. Kauffman, J. Phys. Chem. A 2001, 105,
3433–3442; h) S. G. Kazarian, M. F. Vincent, F. V. Bright, C. L.
Liotta, C. A. Eckert, J. Am. Chem. Soc. 1996, 118, 1729–1736;
i) J. C. Meredith, K. P. Johnston, J. M. Seminario, S. G. Kaz-
arian, C. A. Eckert, J. Phys. Chem. 1996, 100, 10837–10848.
D. D: Perrin, W. L. F. Armarego, Purification of Laboratory
Chemicals, 3rd ed., Pergamon, New York, 1988.
Received: June 8, 2010
[11] a) P. Raveendran, Y. Ikushima, S. L. Wallen, Acc. Chem. Res.
2005, 38, 478–485; b) Y. Enokida, O. Tomika, S. C. Lee, A.
Rustenholtz, C. M. Wai, Ind. Eng. Chem. Res. 2003, 42, 5037–
5041; c) P. Raveendran, D. L. Wallen, J. Am. Chem. Soc. 2002,
124, 12590–12599; d) M. A. Blatchford, P. Raveendran, S. L.
Wallen, J. Am. Chem. Soc. 2002, 124, 14818–14819; e) P. Ra-
veendran, S. L. Wallen, J. Am. Chem. Soc. 2002, 124, 7274–
7275; f) M. R. Nelson, R. F. Borkman, J. Phys. Chem. A 1998,
[12]
Published Online: October 7, 2010
6206
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 6200–6206