Oxidation of alkylarenes by nitrate catalyzed by polyoxophosphomolybdates: Synthetic applications and mechanistic insights

Article abstract of DOI:10.1021/ja031710i

Alkylarenes were catalytically and selectively oxidized to the corresponding benzylic acetates and carbonyl products by nitrate salts in acetic acid in the presence of Keggin type molybdenum-based heteropolyacids, H_3+xPV_xMo_12-xO₄₀ (x = 0-2). H ₅PV₂Mo₁₀O₄₀ was especially effective. For methylarenes there was no over-oxidation to the carboxylic acid contrary to what was observed for nitric acid as oxidant. The conversion to the aldehyde/ketone could be increased by the addition of water to the reaction mixture. As evidenced by IR and ¹⁵N NMR spectroscopy, initially the nitrate salt reacted with H₅PV₂Mo₁₀O ₄₀ to yield a N^VO₂⁺[H ₄PV₂Mo₁₀O₄₀] intermediate. In an electron-transfer reaction, the proposed N^VO₂ ⁺[H₄PV₂Mo₁₀O₄₀] complex reacts with the alkylarene substrate to yield a radical-cation-based donor-acceptor intermediate, N^IVO₂[H₄PV ₂Mo₁₀O₄₀]-ArCH₂R^+.. Concurrent proton transfer yields an alkylarene radical, ArCHR^., and NO₂. Alternatively, it is possible that the N^VO ₂⁺[H₄PV₂Mo₁₀O ₄₀] complex abstracts a hydrogen atom from alkylarene substrate to directly yield ArCHR^. and NO₂. The electron transfer-proton transfer and hydrogen abstraction scenarios are supported by the correlation of the reaction rate with the ionization potential and the bond dissociation energy at the benzylic positions of the alkylarene, respectively, the high kinetic isotope effect determined for substrates deuterated at the benzylic position, and the reaction order in the catalyst. Product selectivity in the oxidation of phenylcyclopropane tends to support the electron transfer-proton transfer pathway. The ArCHR^. and NO₂ radical species undergo heterocoupling to yield a benzylic nitrite, which undergoes hydrolysis or acetolysis and subsequent reactions to yield benzylic acetates and corresponding aldehydes or ketones as final products.

Full text of DOI:10.1021/ja031710i

Published on Web 05/01/2004
Oxidation of Alkylarenes by Nitrate Catalyzed by
Polyoxophosphomolybdates: Synthetic Applications and
Mechanistic Insights
Alexander M. Khenkin and Ronny Neumann*
Contribution from the Department of Organic Chemistry, Weizmann Institute of Science,
RehoVot, Israel 76100
Received December 15, 2003; E-mail: Ronny.Neumann@weizmann.ac.il
Abstract: Alkylarenes were catalytically and selectively oxidized to the corresponding benzylic acetates
and carbonyl products by nitrate salts in acetic acid in the presence of Keggin type molybdenum-based
heteropolyacids, H3+xPVxMo12-xO40 (x ) 0-2). H5PV2Mo10O40 was especially effective. For methylarenes
there was no over-oxidation to the carboxylic acid contrary to what was observed for nitric acid as oxidant.
The conversion to the aldehyde/ketone could be increased by the addition of water to the reaction mixture.
As evidenced by IR and 15N NMR spectroscopy, initially the nitrate salt reacted with H5PV2Mo10O40 to yield
a NVO2+[H4PV2Mo10O40] intermediate. In an electron-transfer reaction, the proposed NVO2+[H4PV2Mo10O40]
complex reacts with the alkylarene substrate to yield a radical-cation-based donor-acceptor intermediate,
NIVO2[H4PV2Mo10O40]-ArCH2R+•. Concurrent proton transfer yields an alkylarene radical, ArCHR•, and NO2.
Alternatively, it is possible that the NVO2+[H4PV2Mo10O40] complex abstracts a hydrogen atom from alkylarene
substrate to directly yield ArCHR• and NO2. The electron transfer-proton transfer and hydrogen abstraction
scenarios are supported by the correlation of the reaction rate with the ionization potential and the bond
dissociation energy at the benzylic positions of the alkylarene, respectively, the high kinetic isotope effect
determined for substrates deuterated at the benzylic position, and the reaction order in the catalyst. Product
selectivity in the oxidation of phenylcyclopropane tends to support the electron transfer-proton transfer
pathway. The ArCHR• and NO2 radical species undergo heterocoupling to yield a benzylic nitrite, which
undergoes hydrolysis or acetolysis and subsequent reactions to yield benzylic acetates and corresponding
aldehydes or ketones as final products.
at the benzylic position yielding either R-nitroalkylbenzenes,5
Introduction
coupling products,6 or oxygenated products.7
Nitrogen oxides, specifically nitrogen dioxide and its dimer,
and nitric acid have long been used as reagents in organic
chemistry for nitration and oxidation.1 Also, in the biological
realm, nitrate has been shown to be an oxidant for anaerobic
enzymatic oxidation of saturated and aromatic hydrocarbons.2
The selectivity of reactions with nitrogen oxides or nitric acid
is mostly determined by the nature of the substrates. Nitration
is usually observed for aromatic substrates, whereas oxidation
is more predominant for aliphatic substrates. The most well-
known use of nitric acid in oxidation is that of cyclohexanol/
cyclohexanone for the production of adipic acid.3 Our interest
in the research described herein was in the liquid phase oxidation
of alkylarenes. Thus, while aromatic substrates typically undergo
nitration,4 alkylbenzenes are preferentially nitrated or oxidized
Polyoxomolybdates of the Keggin structure, H3+xPVxMo12-xO40
(especially x ) 0-2), have been attracting considerable interest
over recent years as oxidation catalysts for both gas phase and
liquid oxidation reactions.8 In the area of liquid-phase oxidation,
the earliest use was as cocatalysts in Wacker type oxidations
of terminal alkenes,9 followed later on by various catalytic
(4) Although these reaction are well-known, the mechanism is still controversial;
see for example ref 1d and Esteves, P. M.; Walkimar de M. Carneiro, J.;
Cardoso, S. P.; Barbosa, A. G. H.; Laali, K. K.; Rasul, G.; Surya Parkash,
G. K.; Olah, G. A. J. Am. Chem. Soc. 2003, 125, 4836-4849 and references
therein.
(5) (a) Shechter, H.; Brain, D. K. J. Am. Chem. Soc. 1963, 85, 1806-1812.
(b) Nishiwaki, Y.; Sakaguchi, S.; Ishii, Y. J. Org. Chem. 2002, 67, 5663-
5668.
(6) (a) Puskas, I.; Fields, E. K. J. Org. Chem. 1966, 31, 4204-4210. (b) Puskas,
I.; Fields, E. K. J. Org. Chem. 1967, 32, 589-592. (c) Puskas, I.; Fields,
E. K. J. Org. Chem. 1967, 32, 3924-3928.
(1) (a) Riebsomer, J. L. Chem. ReV. 1945, 36, 157-233. (b) Gray, P.; Yoffe,
A. D. Chem. ReV. 1955, 55, 1069-1154. (c) Ogata, Y. In Oxidation in
Organic Chemistry, Part C; Trahanovsky, W. S., Ed.; Academic Press:
New York, 1978; pp 295-342. (d) Bosch, E.; Kochi, J. K. In N-Centered
Radicals; Alfassi, Z. B., Ed.; John Wiley & Sons: Chichester, 1998; pp
69-128.
(2) (a) Widdel, F.; Rabus, R. Curr. Opin. Biotechnol. 2001, 12, 259-275. (b)
Spormann, A. M.; Widdel, F. Biodegradation 2000, 11, 85-105. (c)
Kniemeyer, O.; Heider, J. J. Biol. Chem 2001 276, 21381-21386.
(3) (a) Castellan, A.; Bart, J. C. J.; Cavallero, S. Catal. Today 1991, 9, 285-
299. (b) Godt, H. C., Jr; Quinn, J. F. J. Am. Chem. Soc. 1956, 78, 1461-
1464.
(7) (a) Ogata, Y.; Tezuka, H.; Kamei, T. Tetrahedron 1969, 25, 4913-4918.
(b) Ogata, Y.; Tezuka, H.; Kamei, T. Tetrahedron 1969, 25, 4919-4922.
(c) Ogata, Y.; Tezuka, H.; Kamei, T. J. Org. Chem. 1969, 34, 845-847.
(d) Ogata, Y.; Tezuka, H.; Kamei, T. Tetrahedron 1970, 26, 4313-4318.
(e) Zaugg, E.; Rapala, T. Org. Synth., Collect. Vol. 3 1955, 820-821. (f)
Tulry, F. W.; Marvel, C. S. Org. Synth., Collect. Vol. 3 1955, 822-823.
(g) Crandall, W.; Beasly, R.; Lambing, L. L.; Moriconi, R. J. Org. Chem.
1967, 32, 134-136. (h) Ogata, Y.; Tezuka, H. Bull. Chem. Soc. Jpn. 1970,
43, 3285-3286. (i) Srinivasan, B.; Chandalic, S. B. Indian J. Technol.
1971, 9, 274-275. (j) Minisci, F.; Recupero, F.; Gambarotti, C.; Punta,
C.; Paganelli, R. Tetrahedron Lett. 2003, 44, 6919-6922. (k) Nyberg, K.
Acta Chem. Scand. 1970, 24, 473-481.
9
6356
J. AM. CHEM. SOC. 2004, 126, 6356-6362
10.1021/ja031710i CCC: $27.50 © 2004 American Chemical Society

Alkylarene Oxidation by NO /Polyoxophosphomolybdates
A R T I C L E S
3
oxidations of halides,10 alcohols, ketones and amines,11 phe-
nols,12 (alkyl)aromatics,13 dienes,14 alkanes,15 and sulfur contain-
ing compounds.16 From a mechanistic point of view there is
now quite abundant information indicating that the polyoxo-
molybdates act as electron-transfer oxidants toward one of the
reaction components, commonly, the organic substrate or
oxidant. The specific complete reaction pathway, thus, much
depends on the both the targeted oxidative transformation and
the oxidant employed. We have, in the past, studied the
mechanism of polyoxomolybdate-catalyzed reactions using
various oxidants including molecular oxygen,13e,14b nitrous
oxide,11e sulfoxides,13a and alkylhydroperoxides.15 Now we
present our studies on a novel system; the oxidation of
alkylaromatic substrates to acetates and aldehydes using nitrate
as the primary oxidant/oxygen donor.17 Synthetic aspects and
mechanistic insights are presented.
Figure 1. Oxidation of durene by nitrate in acetic acid with different
catalysts. 0.5 M durene, 0.1 M lithium nitrate, and 0.01 M polyoxometalate
in AcOH, 80 °C, Ar. Product yields defined as % mol alkylarene reacted/
mol nitrate vary slightly, (2%. Q ) tetrabutulammonium.
Results and Discussion
Synthetic Aspects. The synthetic objective of this research
was to carry out the selective oxidation of alkyl aromatic
substrates at the benzylic position with emphasis on preventing
over oxidation, i.e., avoiding the formation of the corresponding
carboxylic acid, eq 1. Classical aerobic Co2+/3+/Mn2+/3+/Br-
aerobic oxidations typically yield carboxylic acids as major
products.18 Toward this goal, the activity of different Keggin
type heteropolyacids, namely H3+xPVxMo12-xO40 and H3-
PW12O40, as catalysts was initially tested for the oxidation of
1,2,4,5-tetramethylbenzene (durene) as a model substrate with
nitrate salts to yield 2,4,5-trimethylbenzyl acetate and 2,4,5-
trimethylbenzaldehyde under the following reaction condi-
tions: 0.5 M durene, 0.1 M lithium nitrate, and 0.01 M
polyoxometalate in acetic acid at 80 °C under argon, Figure 1.
As may be observed, the vanadium-containing polyoxomolyb-
dates were slightly more active compared to the simple
H3PMo12O40 compound, while H3PW12O40 was not effective
nor was Mo(O)2acac2 or the neutral Q3PMo12O40 (Q ) tetrabu-
tylammonium). One may preliminarily conclude that both a
sufficiently high oxidation potential and acidic conditions are
needed for catalytic activity. No reaction takes place in the
absence of nitrate. In addition, it should also be noted that the
use of nitric acid in place of a nitrate salt (NaNO3 and LiNO3
gave virtually the same results) yielded 2,4,5-trimethylbenzoic
acid instead of the acetate or aldehyde.
(8) (a) Pope, M. T. Isopoly and Heteropoly Anions; Springer: Berlin, Germany,
1983. (b) Mu¨ller, A. Polyoxometalate Chemistry; Kluwer Academic:
Dordrecht, The Netherlands, 2001. (c) Kozhevnikov, I. V. Catalysis by
Polyoxometalates; Wiley: Chichester, U.K., 2002. (d) Hill, C. L.; Prosser-
McCartha, C. M. Coord. Chem. ReV, 1995, 143, 407-455. (e) Mizuno,
N.; Misono, M. Chem. ReV. 1998, 98, 171-192. (f) Neumann, R. Prog.
Inorg. Chem. 1998, 47, 317-370.
(9) (a) Matveev, K. I. Kinet. Catal. 1977, 18, 716-729. (b) Matveev, K. I.;
Kozhevnikov, I. V. Kinet. Catal. 1980, 21, 855-861. (c) Grate, J. R.;
Mamm, D. R.; Mohajan, S. Mol. Eng. 1993, 3, 205. (d) Grennberg, H.;
Bergstad, K.; Ba¨ckvall, J.-E. J. Mol. Catal. 1996, 113, 355-358. (e) Yokota,
T.; Fujibayashi, S.; Nishyama, Y.; S. Sakaguchi, Y. Ishii, J. Mol. Catal.
1996, 114, 113-119. (f) Bergstad, K.; Grennberg, H.; Ba¨ckvall, J.-E.
Organometallics 1998, 17, 45-50.
(10) (a) Neumann, R.; Assael, I.; J. Chem. Soc., Chem. Commun. 1998, 1285-
1287. (b) Branytskaya, O.; Neumann, R. J. Org. Chem. 2003, 68, 9510-
9512.
(11) (a) Bre´gault, J.-M.; El Ali, B.; Mercier, J.; Martin, J.; Martin, C. C. R.
Acad. Sci. II 1989, 309, 459-462. (b) El Ali, B.; Bre´gault, J.-M.; Mercier,
J.; Martin, J.; Martin, C.; Convert, O. J. Chem. Soc., Chem. Commun. 1989,
825-826. (c) Atlamsani, A.; Ziyad, M.; Bre´gault, J.-M. J. Chim. Phys.,
Phys.-Chim. Biol. 1995, 92, 1344-1364. (d) A. M. Khenkin, A. M.;
Neumann, R. J. Org. Chem. 2002, 67, 7075-7079. (e) Ben-Daniel, R.;
Neumann, R. Angew. Chem., Int. Ed. 2003, 42, 92-95. (f) Neumann, R.;
Levin, M. J. Org. Chem. 1991, 56, 5707-5710. (g) Ben-Daniel, R.; Alsters,
P. L.; Neumann, R. J. Org. Chem. 2001, 66, 8650-8653. (h) El Aakel, L.;
Launay, F.; Atlamsani, A.; Bregeault, J.-M. Chem. Commun. 2001, 2218-
2219. (i) Khenkin, A. M.; Vigdergauz, I.; Neumann, R. Chem.sEur. J.
2000, 6, 875-882.
(12) (a) Kholdeeva, O. A.; Golovin, A. V.; Maksimovskaya, R. I.; Kozhevnikov,
I. V. J. Mol. Catal. 1992, 75, 235-244. (b) Lissel, M.; Jansen van de Wal,
H.; Neumann, R. Tetrahedron Lett. 1992, 33, 1795-1798. (c) Kolesnik, I.
G.; Zhizhina, E. G.; Matveev, K. I. J. Mol. Catal. A: Chem. 2000, 153,
147-154.
(13) (a) Khenkin, A. M.; Neumann, R. J. Am. Chem. Soc. 2002, 124, 4198-
4199. (b) Neumann, R.; de la Vega, M. J. Mol. Catal. 1993, 84, 93-108.
(c) Nomiya, K.; Yanagibayashi, H.; Nozaki, C.; Kondoh, K.; Hiramatsu,
E.; Shimizu, Y. J. Mol. Catal. A 1996, 114, 181-190. (d) Khenkin, A.
M.; Neumann, R. Angew. Chem., Int. Ed. 2000, 39, 4088-4090. (e)
Khenkin, A. M.; Weiner, L.; Wang, Y.; Neumann, R. J. Am. Chem. Soc.
2001, 123, 8531-8542.
After the initial experiments on durene, the oxidation reaction
was tested on a series of substrates as presented in Table 1. As
a rule the acetate ester of the benzylic alcohol was observed to
be the major product, in most cases >80%, while the aldehyde
was the minor product. For substrates susceptible to aromati-
zation, such as tetralin and dihydroanthracene, the formation
of the aromatic product, naphthalene and anthracene, respec-
tively, was significant. There was virtually no nitration at either
the aromatic ring or benzylic position for the substrates noted,
except for highly ring-activated substrates such as 4-methylani-
sole where 10 mol % ring nitration was observed. The use of
H3PMo12O40 as catalyst yielded by and large similar acetate/
aldehyde ratios; however, there was sometimes considerable
nitration, ∼20-30% at the aromatic ring, presumably because
H3PMo12O40 is a weaker oxidant compared to H5PV2Mo10O40.
(14) (a) Neumann, R.; Lissel, M. J. Org. Chem. 1989, 54, 4607-4610. (b)
Neumann, R.; Levin, M. J. Am. Chem. Soc. 1992, 114, 7278-7286.
(15) Khenkin, A. M.; Neumann, R. J. Am. Chem. Soc. 2001, 123, 6437-6438.
(16) (a) Gall, R. D.; Faraj, M.; Hill, C. L. Inorg. Chem. 1994, 33, 5015-5021.
(b) Kozhevnikov, I. V.; Simagina, V. I.; Varnakova, G. V.; Matveev, K. I.
Kinet. Catal. 1979, 20, 506-511. (c) Dzhumakaeva, B. S.; Golodov, V.
A. J. Mol. Catal. 1986, 35, 303-307. (d) Harrup, M. K.; Hill, C. L. Inorg.
Chem. 1994, 33, 5448-5455. (e) Harrup, M. K.; Hill, C. L. J. Mol. Catal.
A 1996, 106, 57-66. (c) Gall, R. D.; Hill, C. L.; Walker, J. E. J. Catal.
1996, 159, 473-478.
(17) Recently, there has been a report on the oxidation of sulfides to sulfoxides
catalyzed by polyoxometalate/nitrate catalysts: Okun, N. M.; Anderson,
T. M.; Hardcastle, K. I.; Hill, C. L. Inorg. Chem. 2003, 42, 6610-6612.
(18) Sheldon, R. A.; Kochi, J. K. Metal-Catalyzed Oxidations of Organic
Compounds; Academic Press: New York, 1981.
9
J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004 6357

A R T I C L E S
Khenkin and Neumann
Table 1. Oxidation of Alkylarenes by Nitrate Catalyzed by H5PV2Mo10O40 (Reaction Conditions: 0.5 mmol of Substrate, 0.3 mmol of
NaNO3, 0.01 mmol of H5PV2Mo10O40, Ar, 1 mL of AcOH, 14 h, 80 °C)
a Product yield ) % mol alkylarene reacted/mol of nitrate; selectivity ) % mol product/all products. b 120 °C, 28 h. c 10% nitration. d 5% nitration.
31P NMR of the reaction mixtures indicated about a 10% loss
after 14 h of H5PV2Mo10O40 to H4PVMo11O40 and H3PMo12O40
coupled with formation of insoluble NaVO3. The latter was
identified by elemental analysis. H3PMo12O40 was stable under
reaction conditions. To maximize conversions to the organic
product, stepwise addition of NaNO3 seems to be the most
effective method compared to use of one large portion. For
example, addition of three portions of 2.5 mmol of NaNO3 at
8 h intervals to 5 mmol of p-xylene in 1 mL of AcOH in the
presence of 0.01 mmol of H5PV2Mo10O40 gave a 98% conver-
sion of p-xylene to 9:1 4-tolyl acetate/4-tolylaldehyde. It is also
possible to partially control the ratio of the acetate versus
aldehyde product by adding water to the solvent, Figure 2,
although total product yields are reduced approximately 2-fold
going from 100% AcOH to 50% AcOH%/50% H2O. Water
decreased both the rate of consumption of alkylarene and the
rate of formation of the acetate.
benzylic acetates or benzaldehyde derivatives using nitrate as
the primary oxidizing agent and acidic polyoxomolybdate
catalysts.
Mechanistic Insights. In the context of this research, we were
also interested in gaining insight into the mechanism of these
reactions. In general, surprisingly little has been reported on
the mechanism of nitrate oxidations. The first obvious important
factor is the fate of the nitrate salt in these reactions. Thus,
various alkylarenes (0.5 M) were reacted with Na15NO3 (0.1
M)19 in AcOH at 80 °C under 1 atm of Ar in the presence of
H5PV2Mo10O40 (0.01 M). After 8 h the gas phase was sampled
with a gastight syringe and analyzed by MS. Typically, ratios
of 15NO/15N2O/15N2 of 20-25:1:1 were measured for substrates
such as dihydroanthracene, diphenylmethane, and 1,2,4-trimeth-
ylbenzene; NO is clearly the major product from the nitrate salt.
Thus, the nitrate is a three-electron oxidant in this system. In
principle, a general mechanistic scenario for such a catalytic
Thus, a rather convenient method has been developed for the
oxidation of alkyl aromatic substrates to the corresponding
(19) 15N labeled NaNO3 was used to allow differentiation from background N2
and O2.
9
6358 J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004

Alkylarene Oxidation by NO /Polyoxophosphomolybdates
A R T I C L E S
3
cm-1. A crystal grown using Na15NO3 as reagent absorbed at
1352 cm-1, which is in excellent agreement with the expected
isotope shift to 1359 cm-1 calculated from the reduced mass
ratio for N-O. Finally the electrochemical behavior of H5PV2-
Mo10O40 in the presence of LiNO3 was measured by cyclic
voltametry. Thus, a solution containing 3.3 mM H5PV2Mo10O40,
66 mM LiNO3, and 100 mM LiClO4 in acetic acid showed an
approximately 3-fold increase in the cathodic current from 60
µA to 186 µA upon electrochemical reduction of the H5PV2-
Mo10O40 polyoxometalate. There was no shift in the cathodic
potential and only a slight decrease in the anodic current. At
higher nitrate concentrations there is a further significant increase
in the cathodic current. Thus, one may conclude that nitrate
would oxidize any reduced polyoxometalate species at relatively
very fast rates. Indeed during the oxidation of the alkylarene
substrates described above, there is no evidence of formation
of reduced (green or blue-green) polyoxometalate species, and
chemical oxidation of reduced polyoxometalate species with
nitrate is immediate.
Figure 2. Oxidation of various alkylarenes by nitrate in different concentra-
tions of acetic acid. Reaction conditions: 0.5 mmol of substrate, 0.3 mmol
of NaNO3, 0.01 mmol of H5PV2Mo10O40, 1 atm of Ar, 1 mL of AcOH/
H2O, 14 h, 80 °C. Selectivity ) mol % aldehdye/all products.
oxidation could first involve either the activation of the
alkylarene or the nitrate anion by the H5PV2Mo10O40 catalyst.
Since, under the reaction conditions studied, but in the absence
of nitrate, the oxidation of alkylarenes to benzylic acetates did
not take place, we surmised that the initial activation mechanism
was by an interaction between the nitrate salt and the H5PV2-
Mo10O40 polyoxometalate. This interaction was studied in three
ways. First, 15N NMR experiments showed that by mixing 0.1
M Na15NO3 with 0.1 M H5PV2Mo10O40 in CD3COOD one peak
at 339 ppm, ∆ν1/2 ) 1.82 was observed (NH3 external standard)
versus a peak at 370 ppm, ∆ν1/2 ) 1.64 Hz for Na15NO3 alone
in CD3COOD. Importantly, use of noncatalytically active acids
such as H3PW12O40 (339 ppm) and H2SO4 (341 ppm) instead
of H5PV2Mo10O40 yielded almost identical 15N NMR spectra.
The source of the oxygen atom(s) in the reaction product was
investigated by using isotopically labeled NaN18O3 and H5PV2-
Mo10
O
and by addition of 18O2. When using 70% 18O in
18
40
18
13e
H5PV2Mo10
O
or 96% 18O in 18O2 for the oxidation of
40
durene, no 18O containing product was observed. However,
when using 75.2% 18O in NaN18O3, approximately 10% 18O
incorporation in both 2,4,5-trimethylbenzyl acetate and 2,4,5-
trimethylbenzaldehyde was observed. The isotope labeling effect
was much diluted due to a secondary competitive reaction
whereby acetate and nitrate exchange oxygen atoms under acid
conditions.22 Apparently, oxygen is transferred to the product
from the nitrate anion. To understand how a NO2+-polyoxo-
metalate complex could activate an alkylarene substrate, kinetic
measurements were carried out to see if the reaction rate could
be correlated with a property of the organic substrate. The rate
of oxidation of various alkylarenes, measured as disappearance
of the organic substrate, showed a reasonable correlation (r2 )
0.81) between the rate constants and the ionization potential of
the alkylarenes,23 Figure 3 (left). Since relative gas-phase
ionization potentials are proportional to liquid-phase oxidation
potentials, the rate-potential correlation notable in Figure 3 (left)
would indicate activation of the alkylarenes by electron transfer
to a NO2+-polyoxometalate complex. On the other hand, Figure
3 (right), there was also a similar correlation of the rate as a
function of the C-H bond strength (r2 ) 0.82, after normaliza-
tion to take into account the number of reactive C-H bonds).
This would indicate the possibility of a hydrogen abstraction
mechanism for the activation of the alkylarene substrate.
The isotope effect in the electron-transfer step was measured
by comparing the rate of oxidation of several nondeuterated
and deuterated alkylarenes. For xanthene/xanthene-9-d2, diphen-
ylmethane/diphenylmethane-d2, and ethylbenzene/ethylbenzene-
d10, ratios kH/kD ) 9.3, 5.9, and 5.4, respectively, were measured
for formation of xanthone and the benzylic acetates of diphen-
ylmethane and ethylbenzene. Such significant kinetic isotope
effects are consistent with reactions proceeding either by a
hydrogen abstraction mechanism or with electron-transfer
oxidations where there is oxidative C-H activation of hydro-
This shift in the 15N NMR indicated the formation of a NO2
+
+
species.20 The formation of the NO2 cation in such a system
would indicate probable ion pairing between NO2 and the
+
polyoxometalate anion. In fact, addition of 20 µL of a saturated
solution of NaNO3 in AcOH to 2 mL of 25 µM H5PV2Mo10O40
in AcOH lead to an ∼7% increase in the optical density at λmax
) 310 nm (from 1.028 to 1.098 au) despite slight dilution of
the solution. Attempts to structurely describe this proposed ion
pair were not successful. Thus, the cocrystallization of 0.1 M
NaNO3 and 0.1 M H5PV2Mo10O40 from acetic acid yielded
typical orange cubic single crystals. X-ray diffraction analysis
clearly showed the presence of the polyanion, but neither solvent
+
atoms nor countercations including the proposed NO2 cation
could be resolved due to disorder.21 Acetate/acetic acid was also
not observed. Crystallization of H5PV2Mo10O40 with authentic
+
NO2 from nitronium tetrafluorborate gave similar results.
However, the IR spectrum of the crystal clearly showed a peak
at 1384 cm-1 attributable to a N-O stretch, in addition to the
fingerprint peaks of H5PV2Mo10O40 between 1100 and 700
(20) Prakash, G. K. S.; Heiliger, L.; Olah, G. A. Inorg. Chem. 1990, 29, 4965-
4968.
(21) Crystal data and structure refinement. Empirical formula PMo10V2O40
;
formula weight 1732.23; space group monoclinic, P2(1)/m; temperature
120 K; λ 0.710 73 Å; a 13.062(3) Å; b 20.422(4) Å; c 20.213(4) Å;
â 107.59(3)°; V 5139.8(19) Å3; Z 4; dcalcd 2.355 mg/cm3; µ 2.935 mm-1
;
Theta range for data collection 2.58° e θ e 17.22°; Limiting indices
-10 e h e 10, 0 e k e 16, -16 e l e 0; reflections collected/unique
3196/3196 [R(int) ) 0.0000]; completeness to θ ) 17.22-98.6%; refine-
ment method full-matrix least-squares on F2; data/restraints/parameters
3196/0/513; goodness-of-fit on F2 1.151; final R indices [I > 2σ(I)] R1 )
0.1107, wR2 ) 0.3306; R indices (all data) R1 ) 0.1270, wR2 ) 0.3422;
largest difference peak and hole -2.686 and -1.224 e/A3.
(22) Heating 100 µmol of NaN18O3 (75, 2 % 18O) and 5 µmol of H5PV2Mo10O4
with 300 µmol of AcOH at 80 °C for 8 h gave acetic acid that was 23 mol
% enriched with 18O as measured by GC-MS.
(23) Taken from the NIST database at http://webbook.nist.gov/chemistry.
9
J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004 6359

A R T I C L E S
Khenkin and Neumann
Figure 3. Initial rate of oxidation of alkylarenes as a function of the ionization potential. Reaction conditions: 0.5 mmol of substrate, 0.1 mmol of LiNO3,
0.01 mmol of H5PV2Mo10O40, 1 atm of Ar, 1 mL of AcOH, 80 °C. Observed rate constants were obtained from initial reaction rates (pseudo first conditions,
measured from disappearance of the organic substrate). For the correlation with bond dissociation energies, the rate constants were normalized taking into
account the number of reactive C-H bonds.
Scheme 1. Reaction Pathway for the Oxidation of Alkylarenes by Nitrate Catalyzed by H5PV2Mo10O40
carbons involving endergonic electron-transfer steps and where
the reverse electron-transfer step is competitive with the proton
transfer.24 The proton transfer is concurrent with the electron
transfer.
the corresponding carbonyl compounds. Alternatively, the
benzylic nitrite intermediate may directly undergo acetolysis
to yield the benzylic acetate. Nitrous acid is quickly decomposed
under acidic conditions to NO, nitrate, and water. NaNO2 when
added to 0.01 M H5PV2Mo10O40 in AcOH is violently decom-
By combining the data presented above, reaction schemes
consistent with the experimental findings may be formulated
as follows, Scheme 1. Initially there is a reaction (a) that is
dehydration of the nitrate anion by the polyoxometalate to yield
the NVO2+[H4PV2Mo10O40] complex as borne out mainly by
the IR and 15N NMR evidence upon interaction of Na15NO3
and H5PV2Mo10O40. Note also the addition of water inhibited
the reaction. In one scenario, the rate-determining step is the
activation of the alkylarene substrate by NVO2+[H4PV2Mo10O40]
via electron transfer (b) followed by concurrent proton transfer
to yield the reduced NIVO2 radical and the benzylic radical
possibly within a “radical cage” (c).25 Alternatively, the rate-
determining step is the activation of the alklyarene to directly
yield the NIVO2 and the benzylic radicals (b-c′).26 In both cases,
fast heterocoupling27 of the benzylic radical and NO2 followed
by hydrolysis yields the benzylic alcohol and nitrous acid as
initial products (d).28,29 The benzylic alcohols either undergo
esterification to yield the benzylic acetates or are oxidized to
(25) A rate equation developed on the basis of a steady-state concentration of
the intermediate NIVO2[H4PV2Mo10O40]-ArCH2R+• and concurrent proton
transfer step can be written as follows, where [I] ≡ NIVO2[H4PV2Mo10
-
O
40]-ArCH2R+•
,
SA
≡
NVO2+[H4PV2Mo10O40], SB
≡
ArCH2R, and
P ≡ ArCHR• f products. dI/dt ) kETSASB - k-ETI - kPTI ) 0. I )
[kET/(k-ET + kPT)]SASB. dP/dt ) kPTI ) [(kPTkET)/(k-ET + kPT)]SASB. Since
the polyoxometalate is the limiting reagent with the nitrate and alkylarene
in excess, the obserVed reaction rate should be first order in SA. Since in
order to obtain SA 2 equiv of the H5PV2Mo10O40polyoxometalate are
required (see Scheme 1), the reaction should be in fact second-order in
H5PV2Mo10O40. A log dP/dt versus log[H5PV2Mo10O40] plot for the
oxidation of durene (0.5 M durene, 0.1 M LiNO3, and 0.003-0.02 M H5-
PV2Mo10O40 in AcOH at 80 °C) yielded a linear curve of slope 2.03, r2 )
0.98 indicating the reaction is indeed second-order in H5PV2Mo10O40
.
(26) A rate equation based on a hydrogen abstraction rate-determining step can
be written as follows where SA ≡ NVO2+[H4PV2Mo10O40], SB ≡ ArCH2R,
and P ≡ ArCHR• f products. P/dt ) kHabsSASB.
(27) There was no significant homocoupling of benzylic radicals to yield dimeric
dibenzylic products. See also: Weiner, B.; Barhard, K. I. N-Centered
Radicals; Alfassi, Z. B., Ed.; John Wiley & Sons: Chichester, 1998; pp
39-67.
(28) In a separate reaction it was shown that 1-pentylnitrite reacts immediately
in acetic acid to yield mostly >95% 1-pentanol.
(29) A referee suggested the possibility of fast further oxidation of the benzylic
radical to a benzyl cation followed by reaction of the benzylic cation with
nitrate to yield benyzl nitrate. We found no evidence for formation of nitrate
esters that have some stability under reaction conditions. Also this pathway
is not compatible with the results observed for the oxidation of phenylcy-
clopropane (see below).
(24) (a) Bockmam, T. M.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. 1998,
120, 2826-2830. (b) Eberson, L.; Jonsson, L. Acta Chem. Scand. B 1986,
40, 79-91.
9
6360 J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004

Alkylarene Oxidation by NO /Polyoxophosphomolybdates
A R T I C L E S
3
Scheme 2. Reaction of Phenylcyclopropane with Nitrate Catalyzed by H5PV2Mo10O40
Table 2. Comparative Oxidation of Alkylarenes and
Corresponding Alcohols by Nitrate Catalyzed by H5PV2Mo10O40
0.3 M NaNO3, 0.01 M H5PV2Mo10O40 in AcOH, 1 atm of Ar,
80 °C, 8 h) yielded 5-phenylisoxazoline (59% selectivity) and
5-phenylisoxazole (29% selectivity) as major products at a
conversion of 41%. Benzaldehyde (10% selectivity and R-nitro-
R-methylstyrene (2% selectivity) were formed as minor prod-
ucts. It appears from the results that the major products were
obtained via a ring-opened intermediate to yield first 5-phenyl-
isoxazoline, and then further by oxydehydrogenation 5-phenyl-
isoxazole. Hydrogen abstraction to yield the neutral benzylic
radical could reasonably lead to the formation of the minor
products observed according to Scheme 2. It has been often
concluded that ring opening products in the oxidation of
phenylcyclopropane are good probes for an electron-transfer
mechanism,31 lending support to such a pathway in the present
case; however, it has also been suggested that the ring opening
from the neutral radical may also be fast31c,d and therefore
competitive with radical coupling. Therefore, the product
distribution observed could conceivably be a result of hydrogen
abstraction. In the context of the present inconclusive mecha-
nistic research, it should be noted that the long standing
discussion of the mechanism of aromatic nitration4 has yet to
yield a consensus mechanism. In this case as for aromatic
nitration, the mechanism may be a function of the properties of
the substrate with lower oxidation potentials and higher bond
dissociation energies favoring an electron transfer pathway and
higher oxidation potentials and lower bond dissociation energies
favoring a hydrogen abstraction mechanism.
a
products, selectivity mol %
initial rate,
mM/min
substrate
acetate
carbonylate
diphenylmethane
diphenylcarbinol
ethylbenzene
1-phenylethanol
1,4-dimethylbenzene
4-methylbenzyl alcohol
72
75
76
82
87
71
28
25
24
18
13
29
0.3
72
0.12
40
0.05
6.6
a Reaction conditions: 0.5 mmol of substrate, 0.1 mmol of LiNO3,
0.01 mmol of H5PV2Mo10O40 in 1 mL of AcOH, 1 atm of Ar, 80 °C.
Selectivity ) mol % product/all products.
posed. The NO formed does not appear to react with the
alklyarene substrate. Control experiments with NO and NO+
yielded a very different product not observed in the reactions
with nitrate. Thus a solution of 0.5 M durene and 0.01 M H5-
PV2Mo10O40 in AcOH under 1 atm of NO at 80 °C for 14 h
showed an 18 mol % conversion of durene, but mostly (95%
selectivity) to the homo-radical coupling product (dimer), 1,2-
bis-(2,4,5-trimethylphenyl)ethane. Similarly a reaction of durene
(0.5 M) and NO+BF4- (0.5 M) in AcOH at 80 °C under 1 atm
Ar gave 23 mol % conversion after 14 h to the same dimer
(93% selectivity).
The question of the fate of the initial benzylic nitrite inter-
mediate was also investigated. One possibility is a nitrite f
alcohol f acetate or carbonylate pathway; another possibility
is a nitrite f acetate or alcohol pathway, where the alcohol is
then oxidized to the carbonylate.30 Comparative oxidations of
diphenylmethane and diphenylcarbinol, ethylbenzene and 1-
phenylethanol, and 1,4-dimethylbenzene and 4-methylbenzyl
alcohol showed the rates of the benzylic alcohol oxidation were
significantly higher than that of the corresponding alkylarene
but the product selectivity was similar, Table 2.30 This would
at least indicate that the benzylic alcohol could be a common
intermediate in the obviously faster transformation of benzylic
alcohols to the benzylic acetate and carbonyl products. Finally,
as noted above the known fast decomposition of nitrite under
acidic conditions was verified under the present reaction
conditions.
Conclusion
In this new catalytic application of the phosphovanado-
molybdate acid, H5PV2Mo10O40, it has been shown that the
reaction of the polyoxometalate with the nitrate anion leads to
formation of a nitronium complex of the polyanion, NVO2
-
+
[H4PV2Mo10O40], which does not lead to nitration of alkylarenes
but rather to oxidation at the benzylic position. It would appear
that the nitronium complex reacts with the alkylarene substrate
by either electron and concurrent proton transfer or hydrogen
abstraction leading to formation of an alkylarene radical and
NO2 that undergo radical heterocoupling to yield benzylic
nitrites. The latter are not observed, but rather they undergo
Using phenylcyclopropane as a probe substrate, we attempted
to gain additional information regarding an electron transfer
versus hydrogen abstraction mechanism,31 Scheme 2. Reaction
of phenylcyclopropane with nitrate (0.5 M phenylcyclopropane,
(31) (a) Riley, P.; Hanzlik, R. P. Tetrahedron Lett. 1989, 30, 3015-3018. (b)
Kim, E. K.; Kochi, J. K. J. Am. Chem. Soc. 1991, 113, 4962-4974. (c)
Wang, Y.; Tanko, J. M. J. Am. Chem. Soc. 1997, 119, 8201-8208. (d)
Wang, Y.; Tanko, J. M. J. Chem. Soc., Perkin Trans. 2 1998, 2705-2711.
(e) Dinnocenzo, J. P.; Simpson, T. R.; Zuilhof, H.; Todd, W. P.; Heinrich,
T. J. Am. Chem. Soc. 1997, 119, 987-993. (f) Dinnocenzo, J. P.; Zuilhof,
H.; Lieberman, D. R.; Simpson, T. R.; McKechney, M. W. J. Am. Chem.
Soc. 1997, 119, 994-1001.
(30) The oxidation of benzylic alcohols with nitric acid is known; cf. Strazzolini,
P.; Runcio, A. Eur. J. Org. Chem. 2003, 526-536 and references therein.
9
J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004 6361

A R T I C L E S
Khenkin and Neumann
dissolved in water to H5PV2Mo10O40 also dissolved in water. The
precipitate formed was filtered and dried overnight in a vacuum oven
at 80 °C. The thermogravimetric analysis showed no water was present.
Elemental analysis: found (calculated) % C, 32.09 (32.64); H, 5.99
(6.16); N, 2.13 (2.38). Solvents and substrates available from com-
mercial sources were of the highest purity available and used without
further purification. 18O2 (Enritech) was 96.1% 18O labeled. Sodium
nitrate-15N, 98% enrichment was from Cambridge Isotope Laboratories
and sodium nitrate, 18O 75.2% enrichment was from D-Chem Ltd..
Diphenylmethane-d2 and xanthene-9-d2 were prepared by reduction of
benzophenone and xanthone, respectively, by reduction with AlCl3-
hydrolysis or acetolysis and subsequent reactions leading to the
formation of benzylic acetates and carbonyl derivatives as final
products. The possible synthetic outcome of such a procedure
has been noted for a series of alkylarene substrates.
Experimental Part
Instruments and Measurements. 31P NMR spectra (101.27 MHz,
85% H3PO4 external standard) were taken on a Bruker Avance 250
DPX instrument and 15N NMR spectra (40.55 MHz, NH3 external
standard) were taken on a Bruker Avance 400 instrument. GC and GC-
MS measurements were recorded on HP 6890 (FID detector) and HP
5973 instruments equipped with a 5% phenyl methylsilicone 0.32 mm
ID, 0.25 µm coating, 30 m column (Restek 5MS) using helium as eluant.
IR spectra for (2% samples in KBr) were measured on a Nicolet Protege
460 FTIR instrument. Samples were prepared by mixing 1 mmol of
H5PV2Mo10O40 and 0.3 mmol of NaNO3 or Na15NO3 which were stirred
in 0.4 mL of acetic acid at 80 °C for 30 min. The acetic acid was
evaporated under vacuum, and the solid obtained was redissolved in
0.5 mL of acetonitrile. To this solution 1 mL of decane was added,
and the reaction mixture was slowly evaporated. After a week, nice
yellow crystals appeared, which were analyzed by X-ray diffraction
(typical Keggin structure was observed) and IR. Cyclic voltammetry
was carried on a BAS-1 instrument. The electrochemical data were
collected using a glass carbon cathode, a platinum anode, and a calomel
reference electrode. The measurements were of solutions of 3.3 mM
POM and 66 mM lithium nitrate in 100 mM lithium perchlorate in
acetic acid at 100 mV/s; I/V ) 500 µA/V.
1
LiAlD4. H NMR and GC-MS were used to verify the absence of
benzylic hydrogen atoms and the presence of deuterium. 18O labeled
H5PV2Mo10O40 was prepared by drying the polyoxometalate at 120 °C
for 24 h and then by adding 50 equiv of H218O (96.2%) in dry CH3CN
and mixing for 6 days in order to obtain a fully equilibrated compound.
Oxidation Reactions. Reactions were carried out under anaerobic
conditions in 20 mL Schlenk tubes equipped with a vacuum/pressure
valve. Typically, the tubes were loaded with the catalyst, nitrate, solvent,
and the substrate and degassed by three successive “freeze-pump-
thaw” cycles and loaded with Ar. The solution was brought to the
appropriate temperature in a thermostated oil bath. The progress of
the reactions was then followed periodically by sampling via a gastight
syringe; analysis was by GC and GC-MS. Kinetic isotope effects were
measured in competitive reactions by reacting equimolar concentrations
of deuturated and protonated substrates. Peaks were quantified by
comparing molecular peak intensities of the benzylic acetate products.
Materials. The H5PV2Mo10O40 × 34H2O polyoxometalate was
prepared using a known literature method.32 Thermogravimetric analysis
(Mettler 50) indicated 34 water molecules per polyoxometalate and
the elemental analysis was the following: found (calculated) % P, 1.31
(1.34); V, 4.38 (4.41); Mo, 41.32 (41.56). IR 1057, 960, 865, and 774
Acknowledgment. This research was supported by the 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.
cm-1 31P NMR (CD3COCD3, 85% H3PO4 external standard) -3.96
.
Supporting Information Available: 15N NMR spectra. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(6), -3.42 (1), -3.37 (4), -3.31 (2), and -3.22 (2) ppm (area of peak).
Q5PV2Mo10O40 (Q ) (C4H9)4N) was prepared mixing 10 equiv of QBr
(32) Tsigdinos, G. A.; Hallada, C. J. Inorg. Chem. 1968, 7, 437-441.
JA031710I
9
6362 J. AM. CHEM. SOC. VOL. 126, NO. 20, 2004