Remarkable efficiency of iron(III) versus manganese(III) tetraphenylporphyrins as catalysts for fast and quantitative oxidation of sulfides into sulfones by hydrogen peroxide

Article abstract of DOI:10.1139/v98-050

The efficiency of various metallo-phtalocyanines (Pht) and - tetraphenylporphyrins (TPP) as catalysts for the H₂O₂ oxidations of dibenzylsulfide, phenylchloroethylsulfide, and thioanisole is investigated in ethanol and acetonitrile, using imidazole as a cocatalyst. Neither PhtNi(II) nor TPPCo(II) exhibits any catalytic activity. PhtMn(II) and TPPMn(III)Cl accelerate markedly these reactions but do not promote quantitative oxidations, at most 70% of the sulfides being transformed into sulfoxides. In contrast, with PhtFe(II) sulfoxides are obtained with a 100% yield from sulfides. Finally, the only catalyst able to oxidize sulfides rapidly (<5 min), completely and quantitatively (100% sulfone) is TPPFe(III)Cl in EtOH. The absence of any by-product, disulfide in particular, suggests that a free sulfenium radical cation is not an active intermediate in these reactions. The marked differences in the behaviour of TPPMn(III)Cl and TPPFe(III)Cl are analyzed by comparing the rates of the catalyst decomposition, of the sulfoxide and sulfone formation as a function of the hydrogen peroxide concentration. The results are discussed in terms of a competition between the several oxidative pathways and a possible mechanism for the oxygen transfer to sulfides.

Full text of DOI:10.1139/v98-050

770
Remarkable efficiency of iron(III) versus
manganese(III) tetraphenylporphyrins as
catalysts for fast and quantitative oxidation of
sulfides into sulfones by hydrogen peroxide
Antonio Marques, Massimo di Matteo, and Marie-Françoise Ruasse
Abstract: The efficiency of various metallo-phtalocyanines (Pht) and -tetraphenylporphyrins (TPP) as catalysts for the H2O2
oxidations of dibenzylsulfide, phenylchloroethylsulfide, and thioanisole is investigated in ethanol and acetonitrile, using
imidazole as a cocatalyst. Neither PhtNiII nor TPPCoII exhibits any catalytic activity. PhtMnII and TPPMnIIICl accelerate
markedly these reactions but do not promote quantitative oxidations, at most 70% of the sulfides being transformed into
sulfoxides. In contrast, with PhtFeII sulfoxides are obtained with a 100% yield from sulfides. Finally, the only catalyst able to
oxidize sulfides rapidly (<5 min), completely and quantitatively (100% sulfone) is TPPFeIIICl in EtOH. The absence of any
by-product, disulfide in particular, suggests that a free sulfenium radical cation is not an active intermediate in these reactions.
The marked differences in the behaviour of TPPMnIIICl and TPPFeIII Cl are analyzed by comparing the rates of the catalyst
decomposition, of the sulfoxide and sulfone formation as a function of the hydrogen peroxide concentration. The results are
discussed in terms of a competition between the several oxidative pathways and a possible mechanism for the oxygen transfer
to sulfides.
Key words: sulfide, oxidation, H2O2, manganese(III) or iron(III) tetraphenylporphyrin.
Résumé : Opérant dans l’éthanol et l’acétonitrile et utilisant l’imidazole comme cocatalyseur, on a étudié l’efficacité de divers
métallo-phtalocyanines (Pht) et de -tétraphénylporphyrines (TPP) comme catalyseurs des oxydations par le peroxyde
d’hydrogène du sulfure de dibenzyle, du sulfure de chloroéthyle et de phényle et du thioanisole. Ni le PhtNiII ni le TPPCoII ne
présente d’activité catalytique. Le PhtMnII et le TPPMnIIICl accélèrent beaucoup ces réactions, mais ils ne conduisent pas à
des oxydations quantitatives; au mieux, 70% des sulfures sont transformés en sulfoxydes. Par ailleurs, avec le PhtFeII, les
sulfures conduisent aux sulfoxydes avec des rendements de 100%. Enfin, le TPPFeIIICl dans l’éthanol est le seul catalyseur à
pouvoir oxyder les sulfures rapidement (moins de 5 min), complètement et quantitativement (100% de sulfone). L’absence de
sous-produit, en particulier de disulfure, suggère que le cation radicalaire sulfénium libre n’est pas un intermédiaire actif dans
ces réactions. On a analysé les différences marquées de comportement du TPPMnIIICl et TPPFeIIICl en procédant à une
comparaison des vitesses de décomposition du catalyseur ainsi que les vitesses de formation du sulfoxyde et de la sulfone en
fonction de la concentration du peroxyde d’hydrogène. On discute des résultats en fonction d’une compétition entre plusieurs
voies d’oxydation et un mécanisme possible pour le transfert d’oxygène aux sulfures.
Mots clés : sulfure, oxydation, peroxyde d’hydrogène, manganèse(III) ou fer(III) tétraphénylporphyrine.
[Traduit par la rédaction]
sulfones (1). Therefore, strong oxidants, hypochlorite, or N-
chloramines for
Introduction
Toxic sulfides and, in particular, mustard (ClCH2-CH2)2S, a
warfare agent, are markedly detoxified by transforming them
into their corresponding oxidized derivatives, sulfoxides, and
[O]
[O]
[1]
R2S F R2SO F R2SO2
example, were currently used2 as decontaminants (1b, 2, 3). In
contrast, hydrogen peroxide, a cheap, nonpolluting, and not
strongly corrosive oxidant, has not been considered3 until now
for this objective (4). This is not surprising, since it is well
known that uncatalyzed hydrogen peroxide oxidations4 of
Received December 18, 1997.
This paper is dedicated to Professor Erwin Buncel in
recognition of his contributions to Canadian chemistry.
2
For synthetic, mechanistic, and other purposes, oxidation of
sulfides into sulfoxides but not sulfones by a large variety of
oxidants is extensively investigated (3).
In contrast, the reactivity of H2O2 under basic conditions towards
VX, a toxic organophosphorous thioester, has been investigated
(4) using the high nucleophilicity of the perhydroxyl anion.
A. Marques, M. di Matteo, and M.-F. Ruasse.1 Institut de
Topologie et de Dynamique des Systèmes de l’Université Paris
7, Denis Diderot, associé au CNRS, URA 34, 1, rue Guy de la
Brosse, 75005 Paris, France.
3
1
Author to whom correspondence may be addressed.
4
Telephone: 33 1 44 27 68 05. Fax: 33 1 44 27 68 14. E-mail:
ruasse@paris7.jussieu.fr
Sulfide half-lives of several hours for reagent concentrations in
the 10 mmol range (5).
Can. J. Chem. 76: 770–775 (1998)
© 1998 NRC Canada

771
Marques et al.
Table 1. Sulfide oxidationa by hydrogen peroxide in the presence
of metallo-phtalocyanines (Phth) and -tetraphenylporphyrins (TPP).b
sulfides are slow (5) whereas decontamination procedures re-
quire very fast chemical transformations. Consequently, hy-
drogen peroxide can be viewed as an efficient oxidant only if
its reaction is markedly catalyzed. Biomimetic catalysts and,
in particular, metalloporphyrins have been extensively devel-
oped for olefin epoxidations and alkane hydroxylations (6).
These catalysts should be efficient also for sulfide oxidations,
since these latter are oxidized more readily than unsaturated
and saturated hydrocarbons (7). However, with the currently
investigated catalysts, strong oxidants (iodosobenzene, per-
acids, etc.) are generally preferred to hydrogen peroxide, in
order to avoid the catalyst destruction by hydroxyl radicals
readily released by homolytic H2O2 decomposition, itself cata-
lyzed by metal complexes (8). Nevertheless, hydrogen perox-
ide can be used successfully using specially robust porphyrin
ligands (9). More recently, metallo-phtalocyanines were
shown to be also powerful catalysts for H2O2 oxidation of
resistant chlorophenols (10). Moreover, depending on the sub-
strate and on the objectives of the authors, iron or manganese
or any other transition metal can be the most efficient for oxy-
gen transfer (11).
With regards to sulfides, much less work has been carried
out (3b and c, 12) and, to the best of our knowledge, there is
no data that allows for a reasonable choice of a given catalyst
(metal and ligand) for promoting their fast and quantitative
oxidation into sulfones. In this paper, we report a preliminary
investigation of the efficiency of various commercially avail-
able biomimetic catalysts for the hydrogen peroxide oxidation
of some alkyl sulfides often used to simulate the reaction of
mustard (2).
Catalyst Time (min)c
Sulfide
Solvent %Sd %SOa %SO2a
f
None
2800
2760
2760
(PhCH2)2S
EtOH
5
5
5
95
95
95
f
f
f
f
f
PhthNiIIe
TPPCoII
PhthMnII
5
5
5
5
5
5
5
5
30 70
93
PhthFeII
7
TPPMnIIICl
TPPFeIIICl
TPPFeIIICl
TPPFeIIICl
TPPMnIIICl
TPPFeIIICl
30 70
0
0
0
f
99
100
f
PhSCH2CH2Cl
PhSCH3
(PhCH2)2S CH3CNg 25 75
f
100
f
0
0
99
a Sulfoxide, SO, and sulfone, SO2, are the only products. Product yields
are reproducible at ±3%.
b Catalyst conc., 2 × 10–3 M; [Im], 5 × 10–2 M; sulfide conc., 8.5 × 10–2 M;
[H2O2]Tot, 0.476 M, added by fraction of 10 µL; room temperature.
c Between the last addition of H2O2 and the end of the reaction.
d Unchanged sulfide.
e Catalyst conc., 5 × 10–4 M, because of the poor solubility of the catalyst in
EtOH.
f This product is not detectable in the corresponding chromatogram.
g In this solvent, metallo-phthalocyanines are not soluble at the
concentration used in ethanol.
comparison of their retention times to those of authentic sam-
ples that were commercially available. When sulfide is not
completely consumed, aliquots of the reaction mixture are ana-
lyzed at regular time intervals until the sulfide disappearance
does not progress significantly.
Results and discussion
As shown in the three first rows of Table 1, neither CoIITPP
nor NiIIPht catalyzes significantly the oxidation of 1, since the
time required for its almost complete disappearance is similar
with and without these catalysts. In contrast, Mn(II) or Fe(II)
phtalocyanines exhibit significant catalytic power; 5 min after
the complete addition of H2O2, most of dibenzylsulfide is
transformed into the corresponding sulfoxide. However,
Mn(II) is less efficient than Fe(II), since with Mn(II) the reac-
tion stops when 70% of sulfide is oxidized whereas 7% only
subsists with Fe(II). The behavior of Mn(II) porphyrin is quite
similar to that of the two metallo-phtalocyanines. Compound
1 is significantly oxidized into sulfoxide only. With all these
catalysts, no sulfone was obtained. Finally, the only catalyst
able to totally oxidize dibenzylsulfide into sulfone is FeIIITPPCl.
The reaction is very fast under the reaction conditions, sulfide
being completely oxidized in less than 5 min. Moreover, oxi-
dation products, other than sulfoxide and sulfone, are never
observed.
In Table 1, are also shown results for (2-chloroethyl)phenyl
sulfide 2 and thioanisole 3. Again, FeIIITPPCl is able to pro-
mote the complete and quantitative transformation of these
sulfides into the corresponding sulfones directly. At variance
with dibenzylsulfide, MnIIITPPCl is not very efficient, 2 being
oxidized at most 30% with this catalyst.
In these experiments, ethanol is preferred to acetonitrile in
order to minimize sulfide oxidation by OH radicals, formed
by metal-mediated homolysis of H2O2, which would lead to
unwanted oxidation by-products. It is expected that these radi-
cals react with ethanol more rapidly than with sulfide, since
In the two last columns of Table 1 are shown the yields in
sulfoxide and sulfone obtained by oxidation of dibenzyl
(PhCH2SCH2Ph), (2-chloroethyl)phenyl sulfides (PhSCH2CH2Cl),
and thioanisole (PhSCH3), 1, 2, and 3, respectively, by hydro-
gen peroxide in ethanol and acetonitrile in the presence of
catalysts with phtalocyanine, Pht, and meso-tetrakis-
phenylporphyrin, TPP, as ligands5 complexing transition met-
als in various oxidation states, Co(II), Ni(II), Fe(II), Fe(III),
Mn(II), Mn(III).
In all the experiments, the initial concentrations of sulfide
and catalyst are 8.5 × 10–2 and 2 × 10–3 M, respectively (these
values correspond to a substrate/catalyst molar ratio of 42.5).
The catalyst concentration is significantly larger than those
currently used to compensate for its decomposition, expected
because of the well-known fragility of the ligands in the ab-
sence of particular substituents. As usual (6–9, 11–13), imida-
zol (Im) is employed as a cocatalyst in a [Im]:catalyst conc.
ratio of 25. Five sulfide equivalents of hydrogen peroxide
(35% in water) are added by small portions (10 µL every
5 min) to minimize dismutation of the oxidant and (or) the
destruction of the catalyst by reaction of the active metallic
species with H2O2. Five minutes after the completion of the
addition, an aliquot of the reaction mixture is analyzed by GC.
The products are identified by GC–MS analysis and (or) by
5
On one hand, phtalocyanines, which are important industrial
dyes, are very cheap. On the other hand, there is extensive work
on metallo-porphyrins as biomimetic catalysts of oxidation (6–9,
11, 12).
© 1998 NRC Canada

772
Can. J. Chem. Vol. 76, 1998
Fig. 1. The progress of the hydrogen peroxide oxidation of
dibenzylsulfide with the oxidant addition in the presence of M(III)
porphyrins; (a) M = Fe; (b) M = Mn; n, sulfide; u, sulfoxide; d,
sulfone. (Room temperature. Initial concentrations in ethanol:
0.085 M sulfide; 0.05 M imidazole; 0.002 M TPPMIII Cl. 40 µL
H2O2 (35% in water) corresponds to 1 sulfide conc. equiv.).
ethanol is an OH trapping agent more efficient than acetoni-
trile (13). Nevertheless, when the catalyzed oxidations are car-
ried out in acetonitrile (last rows of Table 1), the product
distribution is essentially the same as that observed in ethanol.
A reasonable conclusion is, therefore, that free OH radicals do
not play a significant role in these sulfide oxidations under the
present reaction conditions.
Another remarkable result of this investigation is the ab-
sence of any by-product arising from the cleavage of sulfide
radical cations as potential intermediates,6 namely aldehyde
(a)
%
100
A-TPPFe
RS-SR
80
60
40
20
0
[2]
.
.
+
+
R2S
R2S
R + RS
RCHO
or alcohol and disulfide (eq. [2]). These two latter products are
currently obtained in sulfide oxidation by typical electron
transfer (ET) oxidants but not when the H2O2 oxidation is cata-
lyzed by metal porphyrins or cytochrome P-450 (14). In par-
ticular, our results agree fairly well with the conclusions of
Baciocchi et al. (14) in their extensive comparison of sulfide
oxidation by ET oxidants and by hydrogen peroxide in the
presence of biomometic catalyst. Whereas biomometic oxida-
tions give mainly sulfoxide and small amounts of sulfone,
these two compounds are very minor products when the oxi-
dant is a CoIIIW complex (3c).
0
20 40
1 eq.uiv.
60
80 100 120 140 160
L H O
These results are in contrast with previous observations of
Oae et al. (12) who obtained some disulfide with an analogous
catalytic system. However, disulfides were obtained, only
when the starting sulfide (PhSCH2CN or PhSCH2COPh) pre-
sents an α-acidic hydrogen favoring the cleavage of the carb-
on-sulfur bond and the formation of a phenylthiyl radical.
In agreement with Baciocchi results (14), a free sulfur-cen-
tered radical cation is not suggested as an active intermediate
by our results. Therefore, a mechanism (eq. [3], route a) in-
volving a direct oxygen transfer from oxene 4 (6d, 8, 15) to
sulfide and then to sulfoxide, in a process similar to that for
catalyzed epoxidation (9, 11), can be reasonably proposed
when the catalyst is FeIIITPPCl. Nevertheless, an oxygen trans-
fer from metal-oxo radical 5 formed by homolysis of the
perhydroxymetal species (9) (route b of eq. [3]):
µ
2
2
%
B-TPPMn
(b)
100
80
60
40
20
0
• +
+R S
)
a)
2
→
Por M
PorM-OOH
-OH• (b)
PorMO•
→
Por -
M
= O
→
R2 SO
4
[3]
+ 4
+ R S
2
→
R2 SO
R2 SO2
5
0
40
80
120 160 200 240 280
L H O
cannot be ruled out a priori, in particular when M is Mn(III).
However, further results (vide supra) do not support necessar-
ily route b.
µ
2
2
1 equ. iv.
The markedly different behavior of the Mn(III) and Fe(III)
porphyrins is rather surprising, since these two catalysts be-
have quite similarly in olefin epoxidation (11). In the view of
better understanding the origin of this difference, we investigated
in more detail the progress of the oxidation of 1 as hydrogen
peroxide is added. The results are shown in Figs. 1a and b.
With FeIIITPPCl, the very first additions of H2O2 quantita-
tively transform sulfide into sulfoxide. The further additions
are still very efficient, since after the addition of 1 equiv. of
H2O2, 75% of H2O2 is used in sulfide–sulfoxide oxidation, the
remaining 25% of sulfide being unchanged. With 1.5 equiv. of
H2O2, sulfide is totally consumed, the sulfoxide yield reaches
6
However, the formation of this sulfur radical cation by electron
transfer from oxene (eq. [3]) and its very fast recombination with
4 cannot be totally excluded.
© 1998 NRC Canada

773
Marques et al.
Fig. 2. Catalyst decomposition during dibenzylsulfide oxidation with H2O2 addition: d, TPPFeIIICl; m, TPPMnIIICl. The open marks (s and n) are
obtained by correcting the added H2O2 from the amount consumed by sulfide oxidation. Reaction conditions are identical to those in Fig. 1.
% Cat.
100
80
60
With TPPMn 15% of H2 O2
are used for S oxidation
40
20
With TPPFe, 50% of H2 O2
are used for S oxidation
0
0
20
40
60
80
100 120 140 160
L H O
µ
2
2
a maximum at about 95% whereas some sulfone is present.
The last additions of H2O2 are used for sulfoxide oxidation into
sulfone. Finally, after the incorporation of 4 equiv. of H2O2,
the yield in sulfone is quantitative. There are several conclu-
sions from this experiment. Firstly, oxidation of sulfide into
sulfoxide and of sulfoxide into sulfone are consecutive reac-
tions, sulfoxide being an intermediate. This implies that sul-
foxide oxidation is slower than that of sulfide, so that sulfone
appears only when sulfide is almost completely transformed
into sulfoxide. Secondly, at every stage of the reaction pro-
gress, the transformation of sulfide and sulfoxide is quantita-
tive, without formation of any other oxidation products.
Thirdly, hydrogen peroxide is not quantitatively used for sul-
fide oxidation, since 4 equiv., as compared to sulfide, are nec-
essary for the complete transformation in sulfone, which
requires 2 equiv. only. Clearly, other H2O2-consuming proc-
esses are involved.
monitored spectroscopically at their Soret band (λmax =
466 nm and 415 nm in ethanol for Mn(III) and FeIIITPPCl,
respectively), during oxidation of 1 under concentrations iden-
tical to those used in the previous experiments. In Fig. 2 is
shown the decrease in the concentration of the two catalysts
as the reaction progresses. MnIIITPPCl is destroyed rapidly
with the first H2O2 additions, and then its disappearance is
progressively slowed down. FeIIITPPCl is also consumed, al-
though in the first H2O2 additions its concentration decreases
less rapidly. With 140 µL of H2O2 (i.e., 3.5 sulfide equiv.),
about 25% of the two catalysts are still preserved, i.e., the
catalyst concentration is still 8 × 10–5 M with an initial con-
centration of 2 × 10–3 M. This remaining amount is enough to
promote further substrate oxidation, since sulfide, added at this
stage of the reaction, is transformed into sulfoxide. Neverthe-
less, there is a very large difference between the two catalysts
insofar as after the addition of 140 µL H2O2, 57% of the added
H2O2 has been consumed by sulfide, in the presence of
FeIIITPPCl, whereas with MnIIITPPPCl, 17% of H2O2 has been
used for this reaction. Figure 3 shows the disappearance of
FeIIITPPCl consecutive to successive additions of H2O2 in ex-
periments similar to those of Figs. 1 and 2 but in the absence
of sulfide. The comparison of Figs. 2 and 3 exhibits that the
catalyst is markedly less consumed when sulfide is present. In
other words, it seems that the presence of sulfide exerts some
“protection” of the catalyst. However, this conclusion is arti-
ficial, since H2O2 is used partly in sulfide oxidation. When the
amount of added H2O2 is corrected for the sulfide uptake, the
rates of the ligand destruction by H2O2 are not markedly dif-
ferent with Mn(III) and Fe(III).
The results with MnIIITPPCl are completely different
(Fig. 1b). Sulfide is still exclusively transformed into sulfoxide
but the efficiency of the oxidant is very small, since after the
addition of 1 equiv. of H2O2, only 20% of the sulfide is oxi-
dized. After a more and more rapid efficiency decrease, the
reaction stops when about 30% of sulfide still subsists.
Therefore, the difference between the two metal porphyrins
is that most of hydrogen peroxide is consumed by processes
in which sulfide is not involved when MnIIITPPCl is the cata-
lyst whereas with FeIIITPPCl, the H2O2 consumption by sul-
fide is an important route. This difference explains also why
no sulfone is obtained with MnIIITPPCl, since sulfoxide oxi-
dation can start only when there is no more sulfide.
To gain some insights on the competing processes that con-
sume hydrogen peroxide, the destruction of the catalysts was
Therefore, the oxidative destruction of the two catalysts is
promoted by hydrogen peroxide via similar processes,
© 1998 NRC Canada

774
Can. J. Chem. Vol. 76, 1998
Fig. 3. Comparison of TPPFeIIICl decomposition by hydrogen
peroxide in the presence (d) and in the absence (s) of
dibenzylsulfide. Reaction conditions are identical to those in Fig. 1.
complex to sulfides is probably faster with Fe(III) than with
Mn(III) catalyst.
In conclusion, this preliminary work points out that among
the investigated biomimetic catalysts, the most efficient cata-
lyst for quantitative and fast oxidation of sulfide into sulfone
by hydrogen peroxide is FeIIITPPCl. The efficiency of the
analogous Mn(III) complex is markedly smaller. Regarding
the selective biomimetic oxidations of sulfide into sulfoxide,
Fe(II) phtalocyanine can be used successfully. Therefore, with
readily oxidizable sulfides, a significant role of both the metal
and its ligand on the oxygen-transfer ability of these catalysts
is evidenced. More work, in particular using more robust li-
gands (16) and aqueous media, is in progress with the objective
of designing performant catalytic systems for toxic sulfide oxi-
dation.
% Cat.
100
80
60
40
20
0
Experimental section
Physical measurements
The UV-vis spectra were recorded on a Perkin–Elmer lambda
II spectrophotometer. Gas chromatography analysis was car-
ried out on a Delsi–Nermag 200 gas chromatograph equipped
with a fid detector and a CP-Sil 5 column. The products were
identified by comparison of their retention times with authen-
tic samples or by GC–MS (I.T.D. Finigan 800).
0
20
40
60
80 100 120 140 160
L H O
µ
2
2
Materials
independently on the catalyzed sulfide oxidation route. Since
the efficiencies of the two catalysts are very different, there
must be a much more efficient route for the H2O2 consumption
in the presence of MnIIITPPCl than with FeIIITPPCl. This route
is most likely the dismutation of hydrogen peroxide, which is
well known to be metal-catalyzed and which would be fa-
voured, particularly by the Mn(III) catalyst. Our results sug-
gest, therefore, that among the various possible fates of the
oxene species, which is assumed to be the active form of the
catalyst for sulfide oxidation as for alkene epoxydation (9, 11),
route c (eq. [4]) is of comparable importance when M is
Mn(III) or Fe(III).
Acetonitrile and ethanol (HPLC grade, Prolabo) were used as
received. Hydrogen peroxide (H2O2 35% in water, Acros) was
stored at 5°C and titrated every month. Sulfides, sulfoxides,
and sulfones were purchased from Aldrich or Acros and used
as received. The absence of any oxidation product in the com-
mercial sulfides was checked before use by GC. All the cata-
lysts were commercial (Aldrich) and used without further
purification.
Oxidation procedure
To 5 mL of a reaction mixture (0.085 M sulfide, 0.05 M imi-
dazole, and 0.002 M catalyst, in ethanol or in CH3CN) was
added directly the commercial solution of hydrogen peroxide
by small amounts with a microsyringe. The oxidant is added
10 µL per 10 µL with a delay of 5 min between each addition,
to prevent hydrogen peroxide dismutation. The GC analysis
was performed on aliquots withdrawn directly from the reac-
tion mixture. Yields were measured by GC with the internal
standard method. All the reactions were carried out at room
temperature, without any buffer and under aerobic conditions.
• +
(b)
(a)
M
H2O + O2
←
Por -
= O
→
R SO
2
+H O
+R S
2
2
2
Im
[4]
(c)
Porphyrin Destruction
Measurements of the catalyst destruction
In contrast, route a is highly favoured as compared to route b
with iron but not with manganese catalyst. There are two pos-
sible interpretations for the observed metal dependence of the
competition between these two routes. The H2O2 dismutation
could be catalyzed by manganese complex more efficiently
than by iron complex. This is unlikely in the view of previous
data on alkene epoxidation (6a), which show that H2O2 con-
sumption is similarly fast with the two catalysts in the presence
of imidazole, i.e., the rate of route b does not markedly depend
on the nature of the metal. Since our results suggest that route
a is favored with iron but not with manganese, a reasonable
conclusion is that the oxygen transfer from the high-valent oxo
The disappearance of the porphyrins were monitored by UV-
vis spectrophotometry by the decrease of their Soret band
(415 nm for TPPFeIIICl, 466 nm for TPPMnIIICl in ethanol).
Eight microliters of the reaction mixture was taken with a mi-
crosyringe and diluted directly in the cell containing 2 mL of
ethanol. The percentage of remaining catalyst was then calcu-
lated.
Acknowledgements
A.M. is indebted to DRET (Direction des Recherches et Études
Techniques) for financial support (1996). The authors gratefully
© 1998 NRC Canada

775
Marques et al.
and D. Mansuy. J. Am. Chem. Soc. 110, 8462 (1988); (b) T.C.
Bruice. Acc. Chem. Res. 24, 243 (1991); (c) B. Meunier. Chem.
Rev. 92, 1411 (1992); (d) T.G. Traylor, C. Kim, J.L. Richards,
F. Xu, and C.L. Perrin. J. Am. Chem. Soc. 117, 3468 (1995).
7. J.T. Groves and P. Viski. J. Org. Chem. 55, 3628 (1990).
8. T.G. Traylor and F. Xu. J. Am. Chem. Soc. 112, 178 (1990).
9. T.G. Traylor, S. Tsuchiya, Y.S. Byun, and C. Kim. J. Am. Chem.
Soc. 115, 2775 (1993).
10. A. Sorokin and B. Meunier. Chem. Eur. J. 2, 1308 (1996).
11. D. Ostovic and T.C. Bruice. Acc. Chem. Res. 25, 314 (1992).
12. (a) S. Oae, Y. Watanabe, and K. Fujimori. Tetrahedron Lett. 23,
1189 (1982); (b) Y. Watanabe, T. Iyanagi, and S. Oae. Tetrahe-
dron Lett. 23, 533 (1982).
13. J.O. Edwards and R. Curci. In Catalytic Oxidation with hydrogen
peroxide as oxidant. Edited by G. Strukul. Kluwer, Dordrecht.
1992. p. 112.
14. (a) E. Baciocchi, M. Crescenzi, E. Fasella, and M. Mattioli. J.
Org. Chem. 57, 4684 (1992); (b) E. Baciocchi, O. Lanzalunga,
and F. Marconi. Tetrahedron Lett. 35, 9771 (1994).
15. T.G. Traylor and J.P. Ciccone. J. Am. Chem. Soc. 111, 8413
(1989).
acknowledge the fruitful suggestions of Dr. P. Battioni, Dr. E.
Baciocchi, and Dr. C.L. Perrin.
References
1. (a) E.E. Reid. In Chemistry of bivalent sulfur. Chemical Publish-
ing Co., New York. 1960. Chap. 5; (b) Y.C. Yang, J.A. Baker,
and J.R. Ward. Chem. Rev. 92, 1729 (1992).
2. (a) J.H. Ramsden, R.S. Drago, and R. Riley. J. Am. Chem. Soc.
111, 3958 (1989); (b) Y.C. Yang, L.L. Szafraniec, and W.T.
Beaudry. J. Org. Chem. 55, 3664 (1990).
3. (a) M. Madesclaire. Tetrahedron, 42, 5459 (1986); (b) F. Pautet,
R. Barret, L. Favre, and M. Daudon. Pharmazie, 43, 437 (1988);
(c) V. Conte, F. Di Furia, and G. Modena. In Organic peroxides.
Edited by W. Ando. Wiley, New York. 1992. p. 559; (d) E.
Baciocchi, O. Lanzalunga, and B. Pirozzi. Tetrahedron, 53,
12287 (1997).
4. Y.C. Yang, L.L. Szafraniec, W.T. Beaudry, and C.A. Bunton. J.
Org. Chem. 58, 6964 (1993).
5. M.A.P. Dankleff, R. Curci, J.O. Edwards, and H.Y. Pyun. J. Am.
Chem. Soc. 90, 3209 (1968), and refs. cited therein.
16. D. Dolphin, T.G. Traylor, and L.Y. Xie. Acc. Chem. Res. 30, 251
(1997).
6. (a) P. Battioni, J.P. Renaud, J. Bartoli, M. Reina-Artiles, M. Fort,
© 1998 NRC Canada