Oxidation of tertiary diaryl alcohols catalyzed by a water-soluble metalloporphyrin: Caliph-Caliph versus Caliph-CAr bond cleavage

Article abstract of DOI:10.1039/a705713j

In water and at room temperature, tertiary diaryl alcohols were easily cleaved by monopersulfate oxidation using a water-soluble metalloporphyrin catalyst to give mainly aryl methyl ketones and to a lower extent diaryl ketones, probably resulting from a β-fragmentation of the diarylethoxy radical Ar₂MeC-O-; a small amount of unexpected ester derivatives was also detected indicating at least one alternative pathway.

Full text of DOI:10.1039/a705713j

View Article Online / Journal Homepage / Table of Contents for this issue
Oxidation of tertiary diaryl alcohols catalyzed by a water-soluble
metalloporphyrin: Caliph–Caliph versus Caliph–C bond cleavage
Ar
Karine Wietzerbin, Bernard Meunier*† and Jean Bernadou*‡
Laboratoire de Chimie de Coordination du CNRS, 205 route de Narbonne, 31077 Toulouse Cedex 4, France
In water and at room temperature, tertiary diaryl alcohols
were easily cleaved by monopersulfate oxidation using a
water-soluble metalloporphyrin catalyst to give mainly aryl
methyl ketones and to a lower extent diaryl ketones,
probably resulting from a b-fragmentation of the diaryl-
and 4 show that cleavage of the Caliph–CAr bond (with release of
a phenyl or a carboxyphenyl group) occurred more than 10
times faster than that of Caliph–Caliph bond (with release of a
methyl group). The relative scission of phenyl and 4-carboxy-
phenyl groups differed only by a factor of 2 (ratio 2/3 ≈ 2).
Thus the relative rates of cleavage with formation of 2, 3 and 4
were in the ratios 35:17.5:1.
2
ethoxy radical Ar MeC–O·; a small amount of unexpected
ester derivatives was also detected indicating at least one
alternative pathway.
These results are reminiscent of the oxidative cleavage of
some tertiary alcohols by cerium(iv) salts in aqueous aceto-
7
As a part of our mechanistic studies on C–H bond oxidation
nitrile at reflux, where tert-butyl, benzyl and allyl radicals were
1–3
catalyzed by water-soluble metalloporphyrins, we wished to
determine how the oxidation of some intermediate tertiary
alcohols can give ketones through cleavage of a C–C bond. In
addition, little is known about oxidative C–C bond cleavage by
biomimetic catalysts, whereas this is a key step in the
degradation of the lateral chain of cholesterol by cytochrome
shown to be cleaved rapidly from the appropriate alcohol (the
relative rates of formation of the tert-butyl:benzyl:allyl
radicals were 20–63:4.4:1). In the present case, the formation
of the ketones 2–4 suggests a mechanism involving the
formation of a diarylethoxyl radical intermediate (Scheme 2)
which is cleaved by b-fragmentation to give the three different
ketones 2, 3 and 4 and carboxyphenyl (route a), phenyl (route b)
or methyl (route c) radicals. The diarylethoxyl radical could be
formed from 1 in two steps by e loss followed by H+
4
P-450. Tertiary alcohols are usually resistant to oxidation and
their oxidative conversion generally requires a sufficient
2
intrinsic reactivity and the use of strong oxidants (e.g.
5
6
·
concentrated H
2
O
2
in sulfuric acid, chromium(vi) or cer-
abstraction, or directly by H abstraction. Benzoquinone 6 might
7
ium(iv) reagents). We chose to develop this study on the two
diarylmethyl alcohols 4-(1-hydroxy-1-phenylethyl)benzoic
be an end-product deriving from one or both aryl radicals
acid 1§ and 1,1-diphenylethanol, using KHSO
5
as oxidant
Table 1 Yields of products obtained from oxidation of 1 catalyzed by Mn-
TMPyPa
(hydrogen peroxide was also tested in similar conditions but
proved to be inactive) and Mn-TMPyP¶ as catalyst. This water-
soluble metalloporphyrin is well-known for its efficiency in
epoxidation, hydroxylation or peroxidasic oxidation reac-
tions in aqueous media. The reaction products were identified
by HPLC co-injection of authentic samples, and by NMR and
GC–MS analyses of the product mixtures.
Products
Yield (%)b
Selectivity (%)c
8
2,3
9
2
3
4
5
6
27.9 ± 1.1
14.0 ± 0.5
0.8 ± 0.1
7.7 ± 0.3
9.2 ± 0.4
31.3 ± 0.7
14.0 ± 0.4
0.9 ± 0.1
10.6 ± 0.3
8.7 ± 0.2
Among the compounds produced by the oxidation of 1, two
resulted from the loss of one of the two aromatic rings (2, 3),
two others from the loss of the methyl group (4, 5) and the last
one (6) was identified as benzoquinone (Scheme 1). Conversion
and product yields are given in Table 1. More than 90% of the
starting material was converted in 10 min at room temperature,
with only 2% of catalyst (introduced in five consecutive
additions). Such levels of conversion could not be obtained
under similar conditions with cerium(iv) salts, which are well-
Conversion (%) 90.5 ± 4.0
a
The reaction medium (1 ml of acetonitrile–water, 1:9) contained 66 mm
phosphate buffer, pH 5 and 500 mm 1 (introduced as a 25 mm solution in
acetonitrile) and 200 mm of p-diacetylbenzene as internal standard;
[
5
KHSO ] was 10 mm, and [Mn-TMPyP] was 10 mm (introduced in five
consecutive 2 nmol additions, every 2 min; the first addition initiated the
reaction). The reaction was monitored by HPLC using a Nucleosil C18 10
mm column and 65 or 75% aqueous acetonitrile containing 0.1 m
7
known for their ability to oxidize some tertiary alcohols, even
2
1
triethylammonium acetate as eluent (flow rate: 1 ml min ); the detection
was performed at 240 nm and experimental errors were obtained from three
in acetonitrile–water at reflux and/or raising the concentrations
of substrate and oxidant by a factor of 100. The pool 2 + 3 + 4
b
independent experiments. Yields were calculated after 10 min of reaction
at room temperature. Selectivity is defined as the ratio of product yield
divided by substrate conversion.
+
5 represented about 50% yield (or 56% yield with respect to
c
the starting material converted). Relative yields of ketones 2, 3
O
O
O
O
C
O
C
+
C
O
C
+
C
Me
Me
OH
OH
OH
O
KHSO5
2
3
4
C
C
OH
Mn-TMPyP
O
Me
O
C
O
1
C
+
OH
O
5
6
Scheme 1
Chem. Commun., 1997
2321

View Article Online
agreement with the above data concerning the oxidation of 1
Me
C
(oxidation products from 1,1-diphenylethanol were: aceto-
Ar
Ar′
phenone, benzophenone, p-benzoquinone and phenyl benzo-
ate), the quantitative results differ significantly. Especially, the
yield of phenyl benzoate was very low (about 1–2%), indicating
that the carboxylic group present in 1 presumably plays a crucial
role in orienting the reaction towards the ester 5.
•
OH
+
– H +
–
e–
2
3
Me
C
a
Me
C
c
b
c
Ar
Ar′
Ar
b
Ar′
–
H •
a
In summary, a highly efficient oxidative cleavage of a tertiary
diaryl alcohol has been observed. Scission of the Caliph–CAr
bond occurred preferentially compared to the Caliph–CAliph
bond, yielding mainly acetophenone derivatives through a
probable b-fragmentation mechanism. The presence of an ester
derivative whose formation is oxygen dependent indicates that
an alternative pathway is involved. An overall comprehensive
mechanism of this oxidative cleavage reaction is presently
under investigation using isotopic labelling experiments.
OH
1
O
•
4
Scheme 2 b-Fragmentation of a possible alkoxyl radical intermediate
Ar = phenyl and ArA = p-carboxyphenyl)
(
resulting from the b-scission (routes a, b), since we checked that
phenol, benzoic acid and p-hydroxybenzoic acid, three putative
evolution products of the released aryl radicals, could be
quickly oxidized to 6 by the catalytic system. With this
substrate, aryl (phenyl or carboxyphenyl) rather than methyl is
lost, yielding acetophenone rather than benzophenone deriva-
tives. This direction of cleavage differs from investigations on
the decomposition of cumyloxyl radical
recent results on oxidation of 4-(1-hydroxy-1-methylethyl)ben-
zoic acid where the loss of methyl is predominant. A plausible
Footnotes and References
† E-mail: bmeunier@lcc-toulouse.fr
‡ E-mail: bernadou@lcc-toulouse.fr
10,11
and from our
§
Compound 1 was synthesized from benzoylbenzoic acid 4 and
3
methylmagnesium bromide according to ref. 13, and 4-benzoyloxybenzoic
acid 5 was obtained by esterification of benzoyl chloride with 4-hydroxy-
benzoic acid.
explanation of this special behaviour could be a combination of
resonance stabilization of the resulting radicals (methyl less
stable than aryl) and of the enhanced stability of acetophenone
¶
Mn-TMPyP denotes the pentaacetate of the diaqua-manganese(iii)
derivative of meso-tetrakis(1-methylpyridinium-4-yl)porphyrin, and was
prepared according to ref. 12.
1
0
compared to acetone or to benzophenone derivatives.
In the course of catalytic oxidation of 1, the unexpected
product 5 was detected. It was identified as benzoyloxybenzoic
acid by MS and NMR analyses of a collected sample, and by
comparison of its chromatographic behaviour with that of an
authentic sample.§ Product 5 represented only about 8% yield
when 10% acetonitrile was used in the reaction medium but this
yield was increased up to 18% when the amount of acetonitrile
was raised to 30%. At the same time, the yields of all other
identified products were slightly lowered. Furthermore, the
formation of 5 was dependent on oxygen concentration (yield of
1 B. Meunier, Chem. Rev., 1992, 92, 1411.
2 M. Piti e´ , J. Bernadou and B. Meunier, J. Am. Chem. Soc., 1995, 117,
2935.
3
4
R. J. Balahura, A. Sorokin, J. Bernadou and B. Meunier, Inorg. Chem.,
997, 36, 3488.
P. R. Ortiz de Montellano, in Cytochrome P-450: Structure, Mechanism
and Biochemistry, 2nd edn., ed. P. R. Ortiz de Montellano, Plenum, New
York, 1995, p. 279.
1
5
M. Hudlicky, in Oxidations in Organic Chemistry, ACS Monograph
1
86, ed. M. Hudlicky, American Chemical Society, Washington, DC,
5
was 1.5% under nitrogen and 7.5% under oxygen atmosphere,
1990, p. 150.
under experimental conditions slightly different from that of
Table 1), while the formation of all the other products did not
significantly depend on the atmospheric composition. It is also
remarkable that the isomeric compound monophenyl tere-
phthalate was not detected. A conceivable pathway to afford the
ester 5 would be a Baeyer–Villiger oxidation of benzoylbenzoic
acid (which is effectively produced in the reaction, product 4)
but this assumption can be discarded since, in a control
experiment, we showed that benzoylbenzoic acid was stable
under the catalytic oxidation conditions used. The results
clearly demonstrate that pathway for formation of 5 is different
from that of the other products 2–4 and 6 and, especially, that
dioxygen probably plays a crucial role. The exact mechanism
explaining the formation of this unexpected reaction product,
which needs the cleavage of two C–C bonds, is presently
unknown.
6 T. G. Waddell, A. D. Carter and T. J. Miller, J. Org. Chem., 1992, 57,
381; J. J. Cawley and V. T. Spaziano, Tetrahedron Lett., 1973, 47,
4
719.
W. S. Trahanovsky and D. B. Macaulay, J. Org. Chem., 1973, 38,
497.
7
8
1
J. Bernadou, A. S. Fabiano, A. Robert and B. Meunier, J. Am. Chem.
Soc., 1994, 116, 9375; J. T. Groves, J. Lee and S. S. Marla, J. Am. Chem.
Soc., 1997, 119, 6269.
9
M. Vidal, M. Bonnafous, S. Defrance, P. Loiseau, J. Bernadou and
B. Meunier, Drug Metab. Dispos., 1993, 21, 811; M. N. Carrier,
P. Battioni and D. Mansuy, Bull. Soc. Chim. Fr., 1993, 130, 405; T. Ohe,
T. Mashino and M. Hirobe, Arch. Biochem. Biophys., 1994, 310, 402.
1
1
0 C. Walling and A. Padwa, J. Am. Chem. Soc., 1963, 85, 1593.
1 E. Baciocchi, F. d’Acunzo, C. Galli and M. Ioele, J. Chem. Soc., Chem.
Commun., 1995, 429.
1
2 J. Bernadou, G. Pratviel, F. Bennis, M. Girardet and B. Meunier,
Biochemistry, 1989, 28, 7268.
1
3 M. M. J. Kubitscheck and W. A. Bonner, J. Org. Chem., 1961, 26,
Similar experiments as with 1 were performed with 1,1-di-
phenylethanol. Here also, results show a high conversion of the
substrate (more than 75%) and, while they are in qualitative
2
194.
Received in Liverpool, UK, 5th August 1997; 7/05713J
2322
Chem. Commun., 1997