Selective ortho hydroxylation of nitrobenzene with molecular oxygen catalyzed by the H5PV2Mo10O40 polyoxometalate

Article abstract of DOI:10.1021/ja051286g

Nitrobenzene was regioselectively oxidized to 2-nitrophenol with oxygen in a reaction catalyzed by the H₅PV₂Mo₁₀O₄₀ polyoxometalate. The reaction was first order in oxygen and catalyst. ¹⁵N NMR showed the interaction between nitrobenzene and the polyoxometalate. Use of labeled ¹⁸O₂, H₂¹⁸O, a competitive kinetic isotope experiment, and use of phenyl-tert-butylnitrone as a spin-trap and identification by EPR provided evidence for formation of a radical intermediate involving a selective intramolecular interaction at the ortho position due to formation of a H₅PV₂Mo₁₀O₄₀-nitrobenzene complex. Copyright

Full text of DOI:10.1021/ja051286g

Published on Web 06/21/2005
Selective Ortho Hydroxylation of Nitrobenzene with Molecular Oxygen
5 2
Catalyzed by the H PV Mo10O
40 Polyoxometalate
†
‡
,†
Alexander M. Khenkin, Lev Weiner, and Ronny Neumann*
Department of Organic Chemistry and Chemical Research Support Unit, Weizmann Institute of Science,
RehoVot, Israel 76100
Received March 1, 2005; E-mail: Ronny.Neumann@weizmann.ac.il
The selective liquid phase catalytic oxidation of hydrocarbons
with molecular oxygen is an important research goal. Especially
difficult is the selective hydroxylation of arenes to the corresponding
phenols because aromatic substitution reactions are usually not
regioselective, and the initial phenol products are more reactive
than the arene substrate. Thus, arene hydroxylation often leads to
formation of regioisomers, polyhydroxylated arenes, quinones, and
1
tars. Even rarer, in fact, apparently undocumented, is the selectiVe
hydroxylation of a deactivated arene substituted with electron-
withdrawing moieties, such as nitrobenzene. Thus, nitrobenzene
has been hydroxylated to a mixture of nitrophenol regioisomers
Figure 1. Reaction profile for selective oxygenation of nitrobenzene; 0.01
M H5PV2Mo10O40‚34H2O in PhNO2, 140 °C, 2 bar O2.
2
using various hydroxy radical sources, including hydrogen per-
3
oxide. There is also an old report of nitrobenzene hydroxylation
4
inactive polyoxometalate catalysts for this reaction as was [(n-
butyl) N] PV Mo10O40. This gives a preliminary indication that, for
4 5 2
catalytic activity, incorporation of vanadium in the polyoxomolyb-
date framework and the presence of an acid are required for
catalysis. In fact, the relative rates of reaction for 0.01 M
to a mixture of products under basic aerobic conditions, as well
as a report on hydroxylation by anodic electrolysis.5
There has been significant interest in selective oxidation and the
6
activation of hydrocarbons catalyzed by polyoxometalates, and in
this context, the use of molecular oxygen as the terminal oxidant
in oxydehydrogenation and oxygenation reactions catalyzed by
H PV Mo10O40 is especially relevant to this report. Thus, it has
5 2
been previously found that in the activation of arenes with relatively
low oxidation potentials, for example, anthracene or 4-methoxy-
5 2
toluene, H PV Mo10O40 mediates the reaction by initially catalyzing
H
1
3+xPV
x
Mo12-x
O
40 in nitrobenzene at 140 °C under 2 bar O
.0, 3.0, and 5.1 for x from 1 to 3, respectively (relative to
PVMo11
Further kinetic experiments included the following: (i) A kinetic
isotope experiment (0.01 M H PV 40 in 1:1 C NO
NO at 140 °C under 2 bar O ) yielded a k /k ) 2.5 ( 0.1.
(ii) A reaction using labeled (0.01 M H
NO
at 140 °C under 2 bar 18O
2
were
H
4
O40).
5
2
Mo10
O
6
H
5
2
/
1
0
C
6
D
5
2
2
H
D
electron transfer from the hydrocarbon to the catalyst through
1
8
7
O
2
5
PV
2
Mo10
O
40 in
formation of a radical cation intermediate. Therefore, one would
18
18
C
6
H
5
2
2
(96.1% O)) yielded O-
expect that nitrotoluene with a significantly higher oxidation
1
8
labeled 2-nitrophenol with 86% O at the phenolic oxygen. (iii) A
reaction in the presence of H
5 2
potential would be inactive in H PV Mo10O40-mediated oxidations.
1
8
2
O (0.01 M H 40, 0.5 M
at 140 °C under 2 bar 16O
) yielded
2
5
PV
2
Mo10
O
The unexpected oxygenation by dioxygen of 2- and 4-nitrotoluene
1
8
18
H
2
O (94.3% O) in C
H
6 5
NO
2
catalyzed by H
5
PV
2
Mo10
O
40 versus the inactivity of toluene clearly
1
8
8
no O-labeled 2-nitrophenol. (iv) Nitrobenzene at 140 °C under 2
bar O for 24 h in the presence of 1 mol % 1,1′-azobis(cyclo-
points to a change in mechanism for such an oxidation.
Perhaps even more surprising is the present finding that H
2
5 2
PV -
hexanecarbonitrile) as radical initiator yielded a mixture of 2-, 3-,
and 4-nitrophenol (0.056 M) in a nearly statistical distribution
Mo10O
40 catalyzed the selectiVe aerobic oxygenation of nitrobenzene
to 2-nitrophenol. Thus, heating a 0.01 M solution of H PV
4H O in neat nitrobenzene at 140 °C under 2 bar O selectively
5
2 40
Mo10O ‚
(
∼1:0.9:0.45 ratio).
3
2
2
9
We sought to forward an explanation for the described unusual
(>99%) yielded 2-nitrophenol at a ∼5% maximum yield (no other
selective hydroxylation to 2-nitrophenol. Considering its high
oxidation potential, it would appear unlikely that activation of
nitrobenzene proceeded by its oxidation via electron transfer. In
fact, nitrobenzene has been observed to be a mild oxidative
product was observed by NMR or GC) with a kinetic reaction
profile as shown in Figure 1. There is no reaction under Ar. Since
the substrate was in large excess, it is not surprising that the reaction
is pseudo (observed) zero-order in nitrobenzene. After about 30-
1
1
dehydrogenating agent under both acidic and basic conditions.
4
0 h, the reaction is inhibited, although the catalyst appears to
31
Additionally, the possibility of obtaining hydroxylation in the
presence of a radical reaction initiator, the value of the kinetic
isotope effect in a competitive experiment, and the finding, using
labeled oxygen reagents, that the source of the phenolic oxygen in
the product was derived from molecular oxygen indicated the
possibility of initial formation of a radical intermediate formed
through an intramolecular regioselectiVe process and oxidation as
opposed to a nonregioselective autoxidation in the presence of 1,1′-
azobis(cyclohexanecarbonitrile). A rough scheme of such a reaction
scenario is provided in Scheme 1.
remain intact ( P NMR). Further kinetic studies using different
initial catalyst concentrations (H PV 40 from 5 to 20 mM)
showed also that the reaction was first-order in H PV 40 (see
Figure S1). Similarly, reactions carried out at oxygen pressures from
.2 (air) to 8 bar O showed that the reactions were also approxi-
mately first-order in molecular oxygen (Figure S2). Isostructural
5
2
Mo10O
5
2
Mo10O
0
2
3 3
polyoxometalates, such as H PW12O40 and H PMo12O40, were
†
Department of Organic Chemistry.
Chemical Research Support Unit.
‡
9988
9
J. AM. CHEM. SOC. 2005, 127, 9988-9989
10.1021/ja051286g CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Cartoon Representation of the Interaction of
Nitrobenzene with the Acidic H5PV2Mo10O40 Polyoxometalate (just
one of the five H5PV2Mo10O40 isomers is presented for
simplification)
The reaction orders observed in the kinetic experiments would
support a reaction between the proposed polyoxometalate-
nitrobenzene intermediate and molecular oxygen in the rate-
determining step. The pathway for the completion of the reaction,
nitrobenzene to 2-nitrophenol, after the formation of the intermediate
and its reaction with oxygen remains unclear. Notably, aerobic
oxidation usually is accompanied by water formation; in this case,
the reaction stoichiometry shows that no protons are generated in
the oxidative transformation, and it is possible that this is the cause
of the reaction stopping after ∼50 TON. In this context, it is
valuable to note that the addition of 4 Å molecular sieves to a
reaction mixture to remove water completely inhibited the reaction
(Figure S3). Unfortunately, however, addition of water was also
not helpful as it significantly slowed the reaction (Figure S3).
To the best of our knowledge, this is probably the first example
of an (aerobic) regioselective hydroxylation of an arene; it has been
achieved through an intramolecular interaction between a poly-
oxometalate and the substrate.
This scenario is supported first of all by the known formation
+
+
2 n 2 3
of PhNO /H or PhNO /H O intermediates when nitrobenzene
12
Acknowledgment. The research was supported by Israel
Science Foundation and the Helen and Martin Kimmel Center for
Molecular Design. R.N. is the Rebecca and Israel Sieff Professor
of Organic Chemistry. Leonid Konstantinovski is thanked for his
help with the 15N NMR.
is mixed with Bronsted acids. In confirmation of such an
intermediate, a 15N NMR (50.69 MHz) of ∼10% Ph NO
15
2
showed
line broadening from ∆ν1/2 ) 1.24 to 1.71 Hz and a downfield
shift of 8 Hz upon addition of H
calculations by various methods and microwave measurements
5 2
PV Mo10O40 (0.04 M). Second,
13
have shown that the C-H bond length at the ortho position is either
shorter by a fraction of a picometer or practically the same as the
bond lengths at the meta and para positions. Therefore, since the
C-H bond strengths are similar at all positions, the regioselective
hydroxylation should be the result of an intramolecular preference
with the resulting thermokinetic advantage for reaction at the ortho
position.
Supporting Information Available: Full details of the experimental
methods and additional kinetic results. This material is available free
of charge via the Internet at http://pubs.acs
References
(
1) A known exception is the oxidation of benzene by nitrous oxide cf.: (a)
Panov, G. I. CATTECH 2000, 4, 18-32. (b) Panov, G. I.; Uriate, A.;
Rodkin, M. A.; Sobolev, V. I. Catal. Today 1998, 41, 365-385.
2) (a) Marusawa, H.; Ichikawa, K.; Narita, N.; Murakami, H.; Ito, K.; Tezuka,
T. Biorg. Med. Chem. 2002, 10, 2283-2290. (b) Tezuka, T.; Mausawa,
H.; Ichikawa, K. Chem. Lett. 1984, 2245-2148. (c) Tezuka, T.; Narita,
N.; Ando, W.; Oae, S. J. Am. Chem. Soc. 1981, 103, 3045-3049.
Direct evidence for the formation of a radical intermediate
was obtained by an EPR spin-trap experiment. Thus, mixing a
(
solution of 0.01 M H
nitrone in nitrobenzene at room temperature under 2 bar O
yielded after 10 min an EPR signal at g ) 2.0 of a spin-trapped
radical intermediate with splitting constants of A ) 14.06 G
) 2.55 G (Figure 2).14 The trapped radical may be
and A
5 2
PV Mo10O40 and 0.1 M phenyl-tert-butyl-
2
(3) (a) Bianchi, D.; Bertoli, M.; Tassinari, R.; Ricci, M.; Vignola, R. J. Mol.
Catal. A: Chem. 2003, 200, 111-116. (b) Nomiya, K.; Hashino, K.;
Nemoto, Y.; Watanabe, M. J. Mol. Catal. A: Chem. 2001, 176, 79-86.
(4) Wohl, A. Ber. Deut. Chem. Ges. 1899, 32, 3486-3488.
(5) Fujimoto, K.; Tokuda, Y.; Maekawa, H.; Matsubara, Y.; Mizuno, T.;
Nishiguchi, I. Tetrahedron 1996, 52, 3889-3896.
N
H
associated to either a carbon- or oxygen-centered radical inter-
mediate15 formed via a polyoxometalate-nitrobenzene inter-
(
6) (a) Kozhevnikov, I. V. Catalysis by Polyoxometalates; Wiley: Chichester,
England, 2002. (b) Hill, C. L.; Prosser-McCartha, C. M. Coord. Chem.
ReV. 1995, 143, 407-455. (c) Okuhara, T.; Mizuno, N.; Misono, M. AdV.
Catal. 1996, 41, 113-252. (d) Neumann, R. Prog. Inorg. Chem. 1998,
47, 317-370.
5 2
action. Replacing the H PV Mo10O40 with catalytically inactive
H PW12O40 or H PMo12O40 did not yield the formation of a spin-
3 3
(7) (a) Khenkin, A. M.; Neumann, R. Angew. Chem., Int. Ed. 2000, 39, 4088-
trapped intermediate; the solution remained EPR silent.
4
090. (b) Khenkin, A. M.; Weiner, L.; Wang, Y.; Neumann, R. J. Am.
Chem. Soc. 2001, 123, 8531-8542.
(8) Solutions of H
5
2
PV Mo10O40‚34H
2
O (10 mM) in 2- or 4-nitrotoluene at
1
2
40 °C under 2 bar O
2
for 12 h yielded selective formation of
-nitrobenzaldehyde (17 TON) and 4-nitrobenzaldehyde (19 TON),
respectively.
(
9) A ∼10% maximum yield is possible at 20 mM polyoxometalate, which
represents the boundary of solubility of the catalyst. Dilution of the reaction
medium inhibits the reaction.
(
H D
10) A k /k ) 4.3 was measured in a comparative experiment. Since the rates
are low, the data for a competitive experiment were deemed more accurate.
11) Cristiano, M. L. S.; Gago, D. J. P.; Rocha Gonsalves, A. M. d’A.;
Johnstone, R. A. W.; McCaron, M.; Varej a˜ o, J. M. T. B. Org. Biomol.
Chem. 2003, 1, 565-574 and references therein.
(
(
12) (a) Olah, G. A.; Laali, K. K.; Wang, Q.; Prakash, G. K. S. Onium Ions;
Wiley: New York, 1998. (b) Laali, K. K. Coord. Chem. ReV. 2000, 210,
47-71.
(
13) (a) Clarkson, J.; Smith, W. E. J. Mol. Struct. 2003, 655, 413-422. (b)
Urbanowicz, P.; Kupta, T.; Wrzalik, R.; Pasterny, K. J. Mol. Struct. 1999,
4
82-483, 409-414. (c) Shlyapochnikov, V. A.; Khaikin, L. S.; Grikina,
O. E.; Bock, C. W.; Vilkov, L. V. J. Mol. Struct. 1994, 326, 1-16.
(
14) There is also a broader superimposed and somewhat distorted 8 line signal,
5
1
V(IV), that is associated with a reduced polyoxometalate.
(
15) Buettner, G. R. Free Radical Bio. Med. 1987, 3, 259-303.
Figure 2. EPR spectrum of a radical intermediate trapped by phenyl-tert-
butylnitrone.
JA051286G
J. AM. CHEM. SOC.
9
VOL. 127, NO. 28, 2005 9989