Mechanistic study on benzylic oxidations catalyzed by bismuth(III) salts: X-ray structures of two bismuth-picolinate complexes

Article abstract of DOI:10.1016/j.tetlet.2008.04.044

We report studies that are consistent with benzylic oxidation reactions catalyzed using bismuth(III) salts proceeding via a radical mechanism. Additionally, X-ray structures of two potential bismuth-picolinate complex intermediates are reported.

Full text of DOI:10.1016/j.tetlet.2008.04.044

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 3709–3712
Mechanistic study on benzylic oxidations catalyzed
by bismuth(III) salts: X-ray structures of two
bismuth–picolinate complexes
a
b
a
a,
*
Emmanuel Callens , Andrew J. Burton , Andrew J. P. White , Anthony G. M. Barrett
a
Department of Chemistry, Imperial College London, Exhibition Road, London SW7 2AZ, England, United Kingdom
Chemical Development, GlaxoSmithKline, Medicines Research Centre, Stevenage SG1 2NY, England, United Kingdom
b
Received 8 February 2008; revised 21 March 2008; accepted 8 April 2008
Available online 11 April 2008
Abstract
We report studies that are consistent with benzylic oxidation reactions catalyzed using bismuth(III) salts proceeding via a radical
mechanism. Additionally, X-ray structures of two potential bismuth–picolinate complex intermediates are reported.
Ó 2008 Elsevier Ltd. All rights reserved.
The oxidation of benzylic C–H bonds is of fundamental
importance in the scale-up synthesis of arene carboxylic
acids, aryl ketones and other compounds. Traditionally,
OH
OOH
1
these reactions involve the use of stoichiometric quantities
of potent oxidants such as potassium permanganate or
1
2
2
potassium dichromate. In the last decade, chemists have
3
4
been developing alternative processes using Co, Cr,
Fig. 1.
5
6
7
8
9
10
Mn, Rh, Ru, Zn and Fe species among others. With
increasing environmental concerns, focus on non-toxic,
benign catalysts has assumed greater importance. Since,
bismuth(III) salts are of low toxicity and often Lewis
acidic, numerous synthetic procedures have been reported
prop-2-yl hydroperoxide (2) were consistent with a free
radical mechanism operating during the reaction (Fig. 1).
Nevertheless, when benzylcyclopropane (3) was used as
a substrate, no rearranged product 5 (R=H, Ot-Bu) was
detected and only the corresponding ketone 4 was isolated
in 73% yield (Scheme 1).
12
1
1
using such salts. Recently, we have developed a bis-
muth-catalyzed oxidation of alkylarenes to the correspond-
ing ketones or carboxylic acids using t-butyl hydroperoxide
Consequently, we decided to further investigate this
apparent mechanistic inconsistency. It has been well docu-
mented that ring-opening reactions of cyclopropylalkyl
radicals are generally dependent on enthalpic, polar and
1
2
as the stoichiometric oxidant. In this Letter, we describe
some mechanistic studies of this reaction.
Previously, we reported that experiments on bismuth-
catalyzed benzylic oxidation reactions in the presence of
13
steric effects. Indeed, the predominant reaction pathway
2
,6-di-t-butyl-4-methylphenol (1) and 2-methyl-1-phenyl-
is strongly dependent on the substituents present on the
cyclopropane (Scheme 2).
1
3a
In the case of cyclopropane 3, Ingold and co-workers
*
demonstrated that the ring opening of this radical is a
reversible process with the cyclized form predominant at
22 °C. Based on these results, it is reasonable to suggest
Corresponding author. Tel.: +44 (0)20 75945767; fax: +44 (0)20
7
5945805.
E-mail address: agmb@imperial.ac.uk (A. G. M. Barrett).
0
040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.04.044

3710
E. Callens et al. / Tetrahedron Letters 49 (2008) 3709–3712
Bi (20 mol%),
O-t-Bu
O
O
picolinic acid (20 mol%),
Ph
t
-BuOOH (6 eq),
[O]
Ph
Ph
Ph
10
Ph
+
pyridine /AcOH
O
(
9:1),
11 (2%)
12 (15%)
3
1
00 ºC, 16-18 h
+
by-products
O
Scheme 4.
OR
Ph
+ Ph
Bi (20 mol%),
picolinic acid (20 mol%),
4
73%
5 (not observed)
Ph
O
t
-BuOOH (6 eq),
Scheme 1.
Ph
pyridine /AcOH
Ph
Ph
(
9:1),
9
100 °C, 16-18 h
13 (52%)
Scheme 5.
Y
kopen
kcycle
X
X
Y
Encouraged by these results, we explored the scope of
our bismuth-catalyzed oxidation to determine whether
the addition of oxygen to an alkene could occur to form
the corresponding oxirane (Scheme 5).
6
7
Scheme 2.
Bismuth-catalyzed oxidation of trans-1,3-diphenylpro-
pene (9) gave only isolated chalcone (13) in 52% yield along
with the recovered starting material (42%). The NMR spec-
trum of the reaction mixture did not show the presence of
an epoxide derivative. This result showed that the oxida-
tion procedure is selective towards the benzylic position
over the alkene.
that in the oxidation of cyclopropane 3, the formation of
ketone 4 is preferred and this substrate is not an optimum
probe of radical intermediates. Thus, we examined trans-
1
-benzyl-2-phenylcyclopropane (10) since the derived
radical 6 (X =Y= Ph) is predominantly ring opened as 7
1
4
(
Scheme 3).
Following literature precedent, phenylacetaldehyde (8)
The synthesis and characterization of bismuth(III) pico-
linates was examined to gain a better understanding of the
inner sphere ligands during the catalysis. Several pyridine
and pyrazinecarboxylates of bismuth(III) have been
1
5
was condensed under alkaline conditions to give trans-1,3-
diphenylpropene (9) in 37% yield. Subsequently, Simmons–
1
6
17
Smith cyclopropanation gave the desired substrate 10.
Cyclopropane 10 was subjected to a benzylic oxidation
catalyzed by bismuth(0) and picolinic acid with excess of
t-butyl hydroperoxide. Based on the recovery of starting
material 10, conversion in this reaction was less than
reported in the literature, including bismuth picolinate.
All these insoluble, polymeric complexes were character-
1
7a
ized mainly by IR spectroscopy. However, Postel et al.
have reported the crystal structure of the mononuclear bis-
muth 2,6-pyridinedicarboxylate complex.
1
7b
7
0%. Interestingly, the corresponding ketone was not
The common procedure to prepare these complexes is
by heating the ligand with bismuth oxide in water under
reflux for 1–2 days. Unfortunately, the resulting products
from this procedure are highly insoluble in most solvents.
Thus, we prepared the bismuth–picolinate complex via bis-
observed (by NMR and MS of the reaction mixture) but
two other products, peroxide 11 and dione 12, were iso-
lated in poor yields (Scheme 4).
The formation of products 11 and 12 was consistent
with the operation of a radical rearrangement reaction dur-
ing oxidation. Of particular note, the isolation of peroxide
1
8
muth(III) ethoxide, which was allowed to react with
excess picolinic acid. This gave a white powder that was
partially soluble in hot DMSO and to our delight gave
white crystals, from DMSO and acetic acid, suitable for
1
1 was consistent with the presence of t-butyl peroxyl rad-
Å
icals (t -BuOO ). Secondly, these results underlined the
importance of the nature of the substrate as a mechanistic
probe.
1
9
an X-ray crystallographic study (Fig. 2).
The X-ray structure of 14 revealed an eight-coordinate
bismuth center with four chelating picolinate ligands
(Fig. 2); the coordination at the metal is bi-capped trigonal
O
KOH
prismatic (see Fig. S2 in the Supplementary data). Adjacent
anionic units are linked by the sodium cation (not drawn in
the picture) to form a chain polymer along the crystallo-
graphic b axis direction (see Fig. S3).
To gain further insight on the bismuth species involved
in the benzylic oxidation reaction, we sought to character-
ize the powder resulting from the reaction of bismuth(0)
with picolinic acid in the presence of the oxidant but with-
2
Ph
Ph
Ph
EtOH
H
37%
Et2Zn, TFA
CH2I2
Ph
Ph
8
6%
Scheme 3.

E. Callens et al. / Tetrahedron Letters 49 (2008) 3709–3712
3711
In conclusion, we have shown that bismuth-catalyzed
benzylic oxidation most reasonably proceeds via the for-
mation of the radical of the stoichiometric oxidant, t-butyl
hydroperoxide and benzylic radicals. In addition, we also
report the first examples of two modestly soluble bismuth
picolinate complexes, which may be involved in the cata-
lytic cycle.
Acknowledgements
We thank GlaxoSmithKline for the generous endow-
ment (to A.G.M.B.) and for research support, the Royal
Society and the Wolfson Foundation for a Royal Society
Wolfson Research Merit Award (to A.G.M.B.), the Wolf-
son Foundation for establishing the Wolfson Centre for
Organic Chemistry in Medical Sciences at Imperial College
London, the Engineering and Physical Sciences Research
Council for the generous support of our studies and the
Thorpe Bequest for a studentship (to E.C.). We thank R.
Murray McKinnell for developing the improved procedure
for the preparation of Bi(OEt)3.
Fig. 2. The molecular structure of the complex anion present in the
crystals of 14.
1
9
out the substrate present. After considerable efforts, we
isolated a novel polymeric complex, the structure of which
was determined by X-ray crystallography (Fig. 3).
The X-ray analysis of crystals 15 showed only three
picolinate ligands per bismuth (cf. four in 14), and so there
is no need for any additional cation; the coordination at
the metal is again bi-capped trigonal prismatic (see Fig.
S7). The structure is an extended coordination polymer
Supplementary data
(
Fig. 3) with ligand bridges between the adjacent bismuth
Supplementary data (full details of the X-ray crystal
structures for 14 and 15) associated with this article can
be found, in the online version, at doi:10.1016/j.tetlet.
centers (cf. the sodium bridges in 14). The bismuth has
an oxidation state of three and each metal is coordinated
by three picolinate ligands.
2008.04.044.
Interestingly, when the benzylic oxidation reaction was
carried out using complex 15 in the presence of tetrahydro-
naphthalene, pyridine/acetic acid and t-BuOOH, a-tetra-
lone was obtained in 76% yield, which is similar to the
yield from the usual oxidation reaction. This result is con-
sistent with picolinic acid acting as an inner sphere ligand
for bismuth in our benzylic oxidation reactions.
References and notes
1
. Sheldon, R. A.; Kochi, J. K. In Metal-Catalyzed Oxidation of Organic
Compounds; Academic Press: New York, 1981.
2
. Hudlicky, M. Oxidations in Organic Chemistry; ACS Monograph 186;
American Chemical Society: Washington, DC, 1990.
. Lei, P.; Alper, H. J. Mol. Catal. 1990, 61, 51.
3
4
. (a) Muzart, J.; Ajjou, A. N. A. J. Mol. Catal. 1991, 66, 155; (b)
Muzart, J. Tetrahedron Lett. 1987, 28, 2131; (c) Muzart, J. Tetra-
hedron Lett. 1986, 27, 3139.
5
6
. Lee, N. H.; Lee, C.-S.; Jung, D.-S. Tetrahedron Lett. 1998, 39, 1385.
. Catino, A. J.; Nichols, J. M.; Choi, H.; Gottipamula, S.; Doyle, M. P.
Org. Lett. 2005, 7, 5167.
7
8
9
. Murahashi, S.-I.; Komiya, N.; Oda, Y.; Kuwabara, T.; Naota, T. J.
Org. Chem. 2000, 65, 9186.
. Gupta, M.; Paul, S.; Gupta, R.; Loupy, A. Tetrahedron Lett. 2005, 46,
4
957.
. (a) Kim, S. S.; Sar, S. K.; Tamrakar, P. Bull. Korean Chem. Soc. 2002,
3, 937; (b) Nakanishi, M.; Bolm, C. Adv. Synth. Catal. 2007, 861.
2
1
0. (a) Ueyama, N.; Yoshinaga, N.; Okamura, T.; Zaima, H.; Nakamura,
A. J. Mol. Catal. 1991, 64, 247; (b) Khan, A. T.; Goswami, P.;
Choudhury, L. H. Tetrahedron Lett. 2006, 47, 2751; (c) Shaikh, T. M.
A.; Emmanuvel, L.; Sudalai, A. J. Org. Chem. 2006, 71, 5043; (d)
Silvestre, S. M.; Salvador, J. A. R. Tetrahedron 2007, 63, 2439.
1. (a) Postel, M.; Dunach, E. Coord. Chem. Rev. 1996, 155, 127; (b)
Leonard, N. M. A.; Wieland, L. C.; Mohan, R. S. Tetrahedron 2002,
1
5
8, 8373.
1
2. Bonvin, Y.; Callens, E.; Larrosa, I.; Henderson, D.; Burton, A. J.;
Barrett, A. G. M. Org. Lett. 2005, 7, 4549.
1
3. (a) Bowry, V. W.; Lusztyk, J.; Ingold, K. U. J. Chem. Soc., Chem.
Commun. 1990, 923; (b) Ajjou, A. N.; Riahi, A.; Chatterjee, D.; J.
Muzart Catal. Commun. 2006, 7, 563.
Fig. 3. Part of one of the extended polymer chains present in the crystals
of 15.

3712
E. Callens et al. / Tetrahedron Letters 49 (2008) 3709–3712
1
4. Hollis, R.; Hughes, L.; Bowry, V. W.; Ingold, K. U. J. Org. Chem.
992, 57, 4284.
5. Stoermer, R.; Thier, C.; Laage, E. Chem. Ber. 1925, 58, 2607.
6. Lorenz, J. C.; Long, J.; Yang, Z.; Xue, S.; Xie, Y.; Shi, Y. J. Org.
Chem. 2004, 69, 327.
a syringe to a dry, pre-weighed Schlenk tube. Evaporation of the
solvent at room temperature, under reduced pressure, gave bismuth
ethoxide (3.48 g, 41%) as a white solid.
1
1
1
4
Preparation of Na[Bi(picolinate) ] 14. Picolinic acid (3.48 g, 31
mmol) in pyridine and AcOH (8:1 ratio; 45 mL) was added to bismuth
ethoxide (3.48 g, 9.9 mmol) in a Schlenk tube via a syringe under Ar.
The mixture was stirred overnight at room temperature, the resulting
precipitate was filtered off and washed with pyridine leaving a white
powder (550 mg). This powder was partially dissolved in hot DMSO
and filtered. The resulting filtrate was diluted with a mixture of PhMe
and DMSO (9:1) and allowed to stand affording white crystals. In the
X-ray structure (CCDC 675457) shown in Fig. 2, the sodium cation
and the 2.5 water molecules per metal have been omitted for clarity.
Preparation of bismuth–picolinate complex 15. Picolinic acid (0.49 g,
17. (a) Zevaco, T.; Postel, M.; Benali Cherif, N. Main Group Met. Chem.
1992, 15, 217; (b) Zevaco, T.; Guilhaume, N.; Postel, M. New J.
Chem. 1991, 15, 927; (c) Andrews, P. C.; Deacon, G. B.; Junk, P. C.;
Kumar, I.; Silberstein, M. Dalton Trans. 2006, 40, 4852; (d) Stavila,
V.; Davidovich, R. L.; Gulea, A.; Whitmire, K. H. Coord. Chem. Rev.
2006, 250, 2782.
1
1
8. (a) Kucheiko, S. I.; Kessler, U. G.; Turova, N. Y. Sov. J. Coord.
Chem. 1987, 13, 1043; (b) McKinnell, R. M. Ph.D. Dissertation,
Imperial College London, 2000.
9. Procedures for the syntheses of bismuth picolinate complexes:
2
4 mmol) and t-BuOOH in H O (70%; 16 mL, 120 mmol) were added
to a suspension of Bi(0) (0.83 g, 4 mmol) in distilled pyridine (10 mL)
and AcOH (1.25 mL). The mixture was heated at 100 °C for 16 h in a
1
8
Preparation of Bi(OEt)
EtOH (50 mL) at 0 °C and the suspension was stirred until gas
evolution had ceased. BiCl (7.67 g, 24.3 mmol) in distilled THF
60 mL) was added resulting in the formation of a white precipitate.
This was heated to reflux for 16 h and concentrated to approximately
0% of its original volume (by distillation) under reduced pressure.
The residue was centrifuged and the supernatant liquid transferred via
3
.
Na (1.67 g, 72.6 mmol) was added to
3
sealed vessel. The precipitate was washed with distilled H
2
O (3 Â 10
(
mL) and dried leaving a white powder (650 mg), which was partially
dissolved in hot DMSO and filtered. The resulting filtrate was diluted
with a mixture of PhMe and DMSO (9:1) and allowed to stand
affording white crystals (CCDC 675458).
5