Oxidations by the reagent "O2-H2O2-vanadium derivative-pyrazine-2-carboxylic acid". Part 12. Main features, kinetics and mechanism of alkane hydroperoxidation

View Article Online / Journal Homepage / Table of Contents for this issue

Oxidations by the reagent “O₂–H₂O₂–vanadium derivative–

pyrazine-2-carboxylic acid’. Part 12.¹Main features, kinetics and

mechanism of alkane hydroperoxidation†

2

Georgiy B. Shul’pin,*^aYuriy N. Kozlov,^aGalina V. Nizova,^aGeorg Süss-Fink,^b

Sandrine Stanislas,^bAlex Kitaygorodskiy^cand Vera S. Kulikova^d

^aSemenov Institute of Chemical Physics, Russian Academy of Sciences, ul. Kosygina, dom 4,

Moscow 119991, Russia. E-mail: Shulpin@chph.ras.ru; WEB site: http://Shulpin.newmail.ru

^bInstitut de Chimie, Université de Neuchâtel, Avenue de Bellevaux 51, CH-2000, Neuchâtel,

Switzerland

^cChemistry Department, Clemson University, Clemson, SC 29634-0973, USA

^dInstitute of Problems of Chemical Physics, Russian Academy of Sciences,

Chernogolovka 142432, Moscow Region, Russia

Received (in Cambridge, UK) 13th February 2001, Accepted 4th June 2001

First published as an Advance Article on the web 19th July 2001

Various combinations of vanadium derivatives (n-Bu₄NVO₃is the best catalyst) with pyrazine-2-carboxylic acid

(PCA) catalyse the oxidation of saturated hydrocarbons, RH, with hydrogen peroxide and air in acetonitrile solution

to produce, at temperatures <40 ЊC, alkyl hydroperoxides, ROOH, as the main primary products. These compounds

are easily reduced with triphenylphosphine to the corresponding alcohols, which can then be quantitatively deter-

mined by GLC. Certain aminoacids similar to PCA can play the role of co-catalyst; however the oxidation rates and

ﬁnal product yields are lower for picolinic and imidazole-4,5-dicarboxylic acids, while imidazole-4-carboxylic and

pyrazole-3,5-dicarboxylic acids are almost inactive. The oxidation is induced by the attack of a hydroxyl radical on

ؒ

the alkane, RH, to produce alkyl radicals, R . The latter further react rapidly with molecular atmospheric oxygen.

ؒ

The peroxyl radicals, ROO , thus formed can be converted to alkyl hydroperoxides. We conclude on the basis of our

kinetic investigation of the oxidation of cyclohexane that the rate-limiting step of the reaction is the monomolecular

decomposition of the complex containing one coordinated PCA molecule: V^V(PCA)(H₂O₂) → V^IV(PCA) ϩ

ϩ

HOO ϩ H . The V^IVspecies thus formed reacts further with a second H₂O₂molecule to generate the hydroxyl radical

ؒ

IV

V

Ϫ

ؒ

according to the equation V (PCA) ϩ H₂O₂→ V (PCA) ϩ HO ϩ HO . The concentration of the active species in

the course of the catalytic process has been estimated to be as low as [V(PCA)H₂O₂] ≈ 3.3 × 10^Ϫ6mol dm^Ϫ3. The

eﬀective rate constant for the cyclohexane oxidation (d[ROOH]/dt = k_eff[H₂O₂]₀[V]₀) is k_eff= 0.44 dm³mol^Ϫ1s^Ϫ1at

40 ЊC, the eﬀective activation energy is 17 2 kcal mol^Ϫ1. It is assumed that the accelerating role of PCA is due

to its facilitating the proton transfer between the oxo and hydroxy ligands of the vanadium complex on the one

hand and molecules of hydrogen peroxide and water on the other hand. For example: (pca)(O᎐)V ؒ ؒ ؒ H O →

᎐

2

(pca)(HO–)V–OOH. Such a “robot’s arm mechanism” has analogies in enzyme catalysis.

temperature the reaction with alkanes gives rise predominantly

to the formation of alkyl hydroperoxides. Substantially lower

amounts of the corresponding alcohols and ketones (alde-

hydes) are formed simultaneously. It was shown that atmos-

pheric oxygen participates in this reaction, and when it is absent

no oxygenation occurs. The oxidation of cyclohexane in an

atmosphere of ¹⁸O₂provided convincing evidence that the

oxidation products contain a considerable amount of labeled

oxygen, and it has been concluded that hydrogen peroxide is a

promoter in the oxidation of alkanes, while atmospheric oxygen

plays the role of oxidant. This reagent gives a convenient

method for eﬃcient hydroperoxidation and oxygenation of

alkanes under very mild conditions (for alkane oxygenation

see books and reviews^2a,8). It should be noted that recently,

in analogous alkane oxidations, encapsulated vanadium com-

plexes have been used.⁹An accelerating inﬂuence of PCA has

been also observed in H₂O₂oxidations catalysed by methyl-

trioxorhenium.¹⁰The aim of the present work is to investigate

the vanadium-catalysed alkane oxygenation with H₂O₂in

more detail. We will particularly i) demonstrate that various

vanadium derivatives can serve as catalysts in addition to the

previously studied tetra-n-butylammonium vanadate, ii) com-

Introduction

Soluble vanadium derivatives are known to be eﬃcient reagents

and catalysts in oxidations of various organic compounds.²

Vanadium plays an important role in living nature;³particularly

some enzymes containing vanadium ions are capable of oxidiz-

ing C–H bonds in organic compounds. For example, vanadium-

containing haloperoxidases⁴catalyse the halogenation of

organic compounds, RH, in the presence of hydrogen peroxide

and the indirect disproportionation of H₂O₂in the presence of

a halide anion.⁵

We have previously demonstrated that hydrogen peroxide

oxidizes some organic substances, mainly alkanes (including

very inert methane) in air, if a combination of tetrabutyl-

ammonium vanadate and pyrazine-2-carboxylic acid (PCA) is

used as a catalyst.^1,6,7Almost no reaction occurs in the absence

of PCA. Acetonitrile has been used as a solvent. At ambient

† Electronic supplementary information (ESI) available: Appendix 1.

On the possibility of radical-chain oxygenation of alkanes. Appendix 2.

Comparison with other alkane-oxygenating systems. Additional refer-

ences. See http://ww w .rsc.org/suppdata/p2/b1/b101442k/

DOI: 10.1039/b101442k

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

This journal is © The Royal Society of Chemistry 2001

1351

View Article Online

pare the accelerating action of various chelating nitrogen-

containing ligands with that of PCA, iii) check possible

solvents for the reaction, iv) follow the formation of oxo and

peroxo complexes of vanadium in the course of the oxidation

by an NMR method, v) prove the formation of free radicals,

vi) study the kinetics of the alkane oxygenation and propose the

mechanism of the process.

Results and discussion

The system investigated in this work contains a vanadium

derivative as a catalyst, VL_n, where L are various ligands,

obligatory co-catalyst (pyrazine-2-carboxylic acid), promoter

(H₂O₂) and “true” oxidant (molecular oxygen from air). An

alkane, RH, was used as a substrate. The oxidation of an alkane

by the system under consideration gives, according to GLC

analysis, the corresponding ketone (aldehyde) and alcohol. If,

prior to GLC analysis, an excess of solid triphenylphosphine is

added to a sample of the reaction solution, the amount of the

ketone (aldehyde) decreases, while that of the alcohol increases.

This diﬀerence is due to the complete decomposition of the

primary product alkyl hydroperoxide in the GLC injector to

give the corresponding ketone (aldehyde) and alcohol. It should

be noted that in some cases, when a GLC with quartz-lined

injector and quartz capillary column was used, it is possible to

detect peaks in the chromatograms due to alkyl hydroper-

oxides.^6f,gTriphenylphosphine reduces the alkyl hydroperoxide

easily and quantitatively to the corresponding alcohol. By

comparing the concentrations of the ketone (aldehyde) and of

the alcohol measured before and after the treatment of the

sample with PPh₃, it is possible to estimate the real concentra-

tions, not only of the carbonyl compound and of the alcohol,

but also that of the alkyl hydroperoxide (for this method,

see refs. 6, 7, 8c, d, 11). In order to precisely determine the rates

of the reactions, however, it is more convenient to measure

by GLC the concentrations only of ketones (aldehydes) and

alcohols after the treatment of the sample with PPh₃. The sum

of the amounts of all products determined after the reduction

corresponds to the reaction rate. In this kinetic study we are not

interested in the alkyl hydroperoxide/ketone (aldehyde)/alcohol

ratio.

Vanadium-based catalysts

In our previous work on alkane oxidation with H₂O₂, we used

tetra-n-butylammonium vanadate, n-Bu₄NVO₃, as the starting

vanadium-containing component of the catalyst. This salt is

readily soluble in acetonitrile and can be easily prepared

starting from V₂O₅and n-Bu₄NOH.¹²It was important to

explore the possibility of using other vanadium derivatives as

catalysts, especially those which do not contain oxo ligands.

The oxidation reactions with all the vanadium-containing

catalysts were carried out using PCA as a co-catalyst and cyclo-

hexane as a substrate in acetonitrile at 50 or 40 ЊC. We have

found that an eﬃcient catalytic system can be prepared in the

form of a complex of oxovanadate with PCA. For this purpose,

vanadium() salt NH₄VO₃(insoluble in acetonitrile), and PCA

were stirred in hot acetonitrile before the catalytic oxidation

(see Experimental section for details). The reaction course of

the cyclohexane oxidation at various V : PCA ratios is shown in

Fig. 1. It can be seen that the highest rate of the reaction is

attained when the following concentrations of the components

are used: 1.0 × 10^Ϫ4mol dm^Ϫ3of NH₄VO₃and 5.0 × 10^Ϫ4mol

dm^Ϫ3of PCA (Fig. 1d). When the ratio V : PCA = 1 : 2 the

reaction rate is noticeably lower (Fig. 1a). If an eﬃcient trap for

alkyl radicals, CCl₃Br, is added to the reaction solution, cyclo-

hexyl bromide is formed at the expense of oxygenated products

(Fig. 1c).

Fig. 1 Proﬁles of cyclohexane (initial concentration 0.4 mol dm^Ϫ3

)

oxidations by hydrogen peroxide (initial concentration 0.2 mol dm^Ϫ3) in

acetonitrile at 50 ЊC catalysed by salt NH₄VO₃(1.0 × 10^Ϫ4mol dm^Ϫ3) in

the presence of PCA at various concentrations: 2.0 × 10^Ϫ4(a),

4.0 × 10^Ϫ4(b, c), and 5.0 × 10^Ϫ4mol dm^Ϫ3(d ). Compound CCl₃Br (1.0

mol dm^Ϫ3) was added to the reaction solution in the experiment shown

in graph 1(c). Concentrations, c, of cyclohexanol (curve A),

cyclohexanone (curve B) and cyclohexyl bromide (curve C) were

measured after the addition of an excess of solid triphenylphosphine.

case (Fig. 2) the reaction is slightly more rapid in comparison

with the system based on NH₄VO₃(compare with Fig. 1b).

Fig. 3 presents results on the oxidation catalysed by derivatives

of vanadium(), VCl₃, and vanadium(), VO(acac)₂which are

An analogous system can be prepared using a derivative of

vanadium(), VOSO₄, which is insoluble in acetonitrile. In this

1352

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Table 1 Oxidation of cyclohexane with H₂O₂in acetonitrile (concen-

tration of products, × 10³, is given) using diﬀerent samples of stock

a

solution of n-Bu₄NVO₃

Stock solution of (n-Bu₄N)^ϩ[VO₃]^Ϫ

Freshly

prepared prior

to experiment

Kept at

Ϫ5 ЊC for

1 month

Kept at room

temperature

for 2 weeks

Time/min

5

10

15

2.5

4.5

6.7

2.5

4.3

6.8

0.4

0.6

1.0

^aConditions: cyclohexane, 0.46 mol dm^Ϫ3, H₂O₂, 0.2 mol dm^Ϫ3, n-

Bu₄NVO₃, 1.0 × 10^Ϫ4mol dm^Ϫ3, PCA, 4.0 × 10^Ϫ4mol dm^Ϫ3, 40 ЊC.

Concentrations of cyclohexanol and cyclohexanone were measured

after the addition of an excess of solid triphenylphosphine.

Fig. 2 Accumulation of cyclohexanol (A) and cyclohexanone (B)

(measured after the addition of an excess of solid triphenylphosphine)

during cyclohexane (0.4 mol dm^Ϫ3) oxidation by hydrogen peroxide (0.2

mol dm^Ϫ3) in acetonitrile at 50 ЊC catalysed by salt VOSO₄(1.0 × 10^Ϫ4

mol dm^Ϫ3) in the presence of PCA (4.0 × 10^Ϫ4mol dm^Ϫ3).

soluble in acetonitrile and turned out to be less eﬃcient than

n-Bu₄NVO₃. It is important to note that in all cases the presence

of PCA as a co-catalyst is obligatory. Indeed, the concentration

of all oxygenates after 3 h under catalysis with VO(acac)₂in the

absence of PCA was only 0.5 × 10^Ϫ4mol dm^Ϫ3. The vanadium-

containing cluster (p-MeC₆H₄-iPr)₄Ru₄V₆O₁₉(the synthesis of

which has been described recently in ref. 13) catalyses the cyclo-

hexane oxygenation in the presence of PCA (Fig. 4). Finally,

a catalyst can be prepared in situ from V₂O₅, H₂O₂and PCA

(Fig. 5). Details of this experiment are described in the Experi-

mental section.

It can be concluded that n-Bu₄NVO₃is one of the most eﬃ-

cient and available vanadium-containing catalysts. This com-

pound was used in this work in almost all experiments. The

catalyst was introduced into the reaction mixture in the form of

an aliquot of a stock solution of n-Bu₄NVO₃in acetonitrile.

Before conducting precise kinetic measurements, it was import-

ant to study the dependence of the eﬃciency of this stock solu-

tion on the time of its storage prior to the experiment. Table 1

summarises the initial rates of cyclohexane oxygenation. It can

be seen that the activity of the catalyst is not changed if the

stock solution is kept at Ϫ5 ЊC for a period of 1 month.

Eﬀect of pyrazine-2-carboxylic acid and other possible co-

catalysts

If a vanadium derivative is used at a concentration of

1.0 × 10^Ϫ4mol dm^Ϫ3and no PCA is added the rate of alkane

oxidation is negligible. In the case when n-Bu₄NVO₃is intro-

duced into the reaction mixture in a higher concentration

(1.0 × 10^Ϫ3mol dm^Ϫ3) some amount of cyclohexanol can be

detected even in the absence of PCA as a co-catalyst (Fig. 6).

However, the turnover number (TON, moles of products per

mole of a catalyst) only slightly exceeds unity, that is to say the

reaction is close to stoichiometric. The concentration of

cyclohexanol after 10 min is only 0.26 × 10^Ϫ3mol dm^Ϫ3. The

amount of cyclohexanone formed is negligible, which shows

that in this case pure cyclohexyl hydroperoxide is formed. When

a relatively small amount of PCA (4.0 × 10^Ϫ4mol dm^Ϫ3) is

added to this solution, the concentration of the alcohol rises up

to 1.5 × 10^Ϫ2mol dm^Ϫ3, which means an acceleration of the

reaction by a factor of 55.

It is reasonable to assume that the vanadium derivative

employed as a catalyst forms, under the reaction conditions,

oxo and oxo-peroxo complexes (see below). We investigated the

cyclohexane oxidation catalysed by isolated complexes of this

type in the absence of free PCA. Fig. 7a shows the accumulation

of oxygenates during the oxidation catalysed by a dioxo com-

plex of vanadium(), [V(O)₂(pca)₂]^Ϫ(1), where pcaH = PCA.

The oxidations catalysed by oxo-peroxo complexes [V(O₂)-

(O)(pca)₂]^Ϫ(2) (for preparation, see ref. 7f ) and [V(O₂)(O)-

(pca)]ؒ2H₂O^14aare illustrated by Fig. 8a,b. It is evident that the

Fig. 3 Cyclohexane (0.4 mol dm^Ϫ3) oxidation by hydrogen peroxide

(0.2 mol dm^Ϫ3) in acetonitrile at 50 ЊC in the presence of PCA

(4.0 × 10^Ϫ4mol dm^Ϫ3) catalysed by certain vanadium derivatives

(1.0 × 10^Ϫ4mol dm^Ϫ3): n-Bu₄NVO₃(a), VCl₃(b), and VO(acac)₂(c).

Concentrations of cyclohexanol (A) and cyclohexanone (B) were

measured after the addition of an excess of solid triphenylphosphine.

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1353

View Article Online

Fig. 4 Accumulation of cyclohexanol (A) and cyclohexanone (B)

(measured after the addition of an excess of solid triphenylphosphine)

during cyclohexane (0.4 mol dm^Ϫ3) oxidation by hydrogen peroxide (0.5

mol dm^Ϫ3) in acetonitrile at 40 ЊC catalysed by cluster complex (p-

MeC₆H₄-iPr)₄Ru₄V₆O₁₉(1.0 × 10^Ϫ4mol dm^Ϫ3) in the presence of PCA

(4.0 × 10^Ϫ4mol dm^Ϫ3).

Fig. 5 Cyclohexane (0.46 mol dm^Ϫ3) oxidation by hydrogen peroxide

(0.17 mol dm^Ϫ3) in acetonitrile at 40 ЊC catalysed by a mixture prepared

previously starting from V₂O₅(concentration of V ions 1.0 × 10^Ϫ4mol

dm^Ϫ3), PCA (4.0 × 10^Ϫ4mol dm^Ϫ3) and H₂O₂(see Experimental

section). Concentration of cyclohexanol (᭹) was measured after the

addition of an excess of solid triphenylphosphine. Concentration of

cyclohexanone was less than 0.8 × 10^Ϫ4mol dm^Ϫ3in all cases.

Fig. 7 Cyclohexane (0.4 mol dm^Ϫ3) oxidation by hydrogen peroxide

(0.5 mol dm^Ϫ3) in acetonitrile at 40 ЊC catalysed by complex (n-

Bu₄N)^ϩ[V(O)₂(pca)₂]^Ϫ(1) (1.0 × 10^Ϫ4mol dm^Ϫ3) in the absence of

HClO₄or free PCA (a), in the presence of HClO₄(2.0 × 10^Ϫ4mol dm^Ϫ3

)

(b), and in the presence of free PCA (2.0 × 10^Ϫ4mol dm^Ϫ3) (c).

Concentrations of cyclohexanol (A) and cyclohexanone (B) were

measured after addition of an excess of solid triphenylphosphine.

oxygenates, getting close to that obtained with “n-Bu₄NVO₃ϩ

4PCA” (compare Figs. 7a,c, 8c). This clearly demonstrates that

the system needs free PCA in addition to the amount which

is necessary for the formation of a complex between an oxo-

vanadium derivative and PCA.

We have also found that perchloric acid can be used as a

co-catalyst; however the accelerating eﬀect of HClO₄is much

less pronounced than that for PCA (compare Fig. 9a and

Fig. 8c). The dependence of the initial reaction rate on the

HClO₄concentration is shown in Fig. 10. When PCA is added

to the mixture containing HClO₄, the reaction proceeds much

faster (Fig. 9c,d), especially for a PCA concentration of

4.0 × 10^Ϫ4mol dm^Ϫ3(Fig. 9d). However, the latter experiment

demonstrates that the addition of HClO₄to the “standard”

catalyst composition (n-Bu₄NVO₃ϩ 4PCA) does not enhance

the rate of oxygenation (this rate is even slightly lower in com-

parison with the experiment in the absence of perchloric acid:

compare Fig. 9d and Fig. 8c). Unlike perchloric acid, the weak

acetic acid does not aﬀect the oxidation at all.

Fig. 6 Accumulation of cyclohexanol (measured after the addition of

an excess of solid triphenylphosphine) during cyclohexane (0.46 mol

dm^Ϫ3) oxidation by hydrogen peroxide (0.2 mol dm^Ϫ3) in acetonitrile at

40 ЊC in the presence of Bu₄NVO₃(1.0 × 10^Ϫ3mol dm^Ϫ3) in the absence

of PCA.

complexes 1 and 2 used in a concentration of 1.0 × 10^Ϫ4mol

dm^Ϫ3(in the absence of added PCA) catalyse the oxidation,

however the eﬃciency of this catalysis is relatively low (com-

pare Figs. 7a and 8a,b with Fig. 8c; the latter graph presents the

oxidation catalysed with the combination “n-Bu₄NVO₃ϩ

4PCA” under the same conditions). It is noteworthy that the

addition of free PCA to the reaction mixture gives rise to

an acceleration of the oxidation and enhances the yield of

1354

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Fig. 9 Proﬁles of cyclohexane (0.4 mol dm^Ϫ3) oxidations by hydrogen

peroxide (0.5 mol dm^Ϫ3) in acetonitrile at 40 ЊC catalysed by salt n-

Bu₄NVO₃(1.0 × 10^Ϫ4mol dm^Ϫ3) in the presence of HClO₄(4.0 × 10^Ϫ4

mol dm^Ϫ3) (a), PCA (2.0 × 10^Ϫ4mol dm^Ϫ3) (b), HClO₄(3.0 × 10^Ϫ4mol

dm^Ϫ3) ϩ PCA (2.0 × 10^Ϫ4mol dm^Ϫ3) (c), and HClO₄(4.0 × 10^Ϫ4mol

Fig. 8 Cyclohexane (0.4 mol dm^Ϫ3) oxidation by hydrogen peroxide

(0.5 mol dm^Ϫ3) in acetonitrile at 40 ЊC catalysed by complexes

NH₄[V(O₂)(O)(pca)₂] (2) (a) and [V(O₂)(O)(pca)]ؒ2H₂O (b) (1.0 × 10^Ϫ4

mol dm^Ϫ3) in the absence of free PCA and catalysed by n-Bu₄NVO₃

(1.0 × 10^Ϫ4mol dm^Ϫ3) in the presence of certain co-catalysts (4.0 × 10^Ϫ4

mol dm^Ϫ3): PCA (c), pyrazine-2,3-dicarboxylic acid (d ), and picolinic

acid (e). Concentrations of cyclohexanol (A) and cyclohexanone (B)

were measured after the addition of an excess of solid triphenyl-

phosphine.

dm^Ϫ3) ϩ PCA (4.0 × 10^Ϫ4mol dm^Ϫ3

)

(d ). Concentrations of

cyclohexanol (A) and cyclohexanone (B) were measured after the

addition of an excess of solid triphenylphosphine.

acids closest to PCA, pyrazine-2,3-dicarboxylic acid, PDCA,

(Fig. 8d) and picolinic acid (Fig. 8e) also play the role of

eﬃcient co-catalysts, although the initial rates of the oxidation

in the case of these compounds are lower in comparison with

that of PCA (Fig. 8c). When PCA as a co-catalyst was replaced

It was important to check the co-catalytic activity of

other compounds whose structure is similar to that of PCA. It

turned out that the chelating nitrogen-containing carboxylic

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1355

View Article Online

Table

2

Selectivity parameters for competitive oxygenations of

cycloalkanes^a

Reagent

c-C₅/c-C₆

c-C₈/c-C₆

c-C₁₂/c-C₆

b

O₂–H₂O₂–NH₄VO₃–PCA

O₂–H₂O₂–hν^c

0.73

0.96

1.6

1.0

0.9

^aNormalized (i.e., taking into account the number of hydrogen atoms

in each hydrocarbon) reactivities of hydrogen atoms in cycloalkanes,

C_N, relative to that in cyclohexane (c-C₆= 1.0); the reactions were in

acetonitrile. ^bFor c-C₈/c-C₆and c-C₁₂/c-C₆at 50 ЊC, for c-C₅/c-C₆at

30 ЊC; 1 h. ^cUV Irradiation at 25 ЊC; 20 min.

Fig. 10 Plot of the initial rate of cyclohexane (0.46 mol dm^Ϫ3

oxidation by hydrogen peroxide (0.2 mol dm^Ϫ3) in acetonitrile at 40 ЊC

)

hydrogen peroxide, especially in the presence of a metal-

complex catalyst.^14b,cFor example, acetonitrile, in the presence

of a base in large amounts (pH of 10) as well as trichloro-

acetonitrile,^14d,egives peroxycarboximidic acids. Acetone forms

tetrameric acetone peroxide.^14fHowever, we can neglect the

chemical reaction between hydrogen peroxide and the nitrile

group under our conditions (i.e., in the absence of a base).

We have also obtained preliminary results on the possibility of

carrying out this oxidation in a protic solvent. The oxidation of

cyclohexane under “standard” conditions in acetic acid is

shown in Fig. 11. It can be concluded that in this solvent the

vanadium-catalysed oxidation proceeds less eﬃciently in com-

parison with acetonitrile, and the addition of PCA does not

accelerate the reaction. Since cyclohexane is insoluble in water

we checked this solvent using benzene (in a very low concentra-

tion) as a substrate. The oxidation is not eﬃcient under these

conditions. Indeed, the yield of phenol after 5 h attains 14%

based on benzene and TON = 14 based on [V]₀.

catalyzed by salt n-Bu₄NVO₃(1.0 × 10^Ϫ4mol dm^Ϫ3

) and PCA

(4.0 × 10^Ϫ4mol dm^Ϫ3) versus initial concentration of HClO₄added.

Selectivity of alkane oxygenation

In order to gain insight into the nature of the oxidizing species

and evaluate the mechanism of the process we determined vari-

ous selectivity parameters for the reaction under discussion

with alkanes as well as for oxygenations by some other systems.

These data are summarized in Tables 2–5.

Competitive reactions for pairs of cycloalkanes allowed us to

determine the reactivities¹⁷of the hydrogen atoms in cyclo-

pentane, cyclooctane and cyclododecane relative to those in

cyclohexane (Table 2). The corresponding parameters for the

oxygenation with H₂O₂in acetonitrile under UV irradiation are

also given in Table 2, since the mechanism of the action of this

system is known: hydroxyl radicals are reactive intermediates in

the reaction.¹⁸It is clear that a close similarity exists between

the parameters for the two systems, “O₂–H₂O₂–NH₄VO₃–PCA”

and “O₂–H₂O₂–hν”. Both these oxidizing systems exhibit low

selectivities in oxygenations of linear and branched alkanes

(Table 3). Fenton’s reagent, “O₂–H₂O₂–FeSO₄”, and the system

“O₂–H₂O₂–Fe(ClO₄)₃” similarly oxidize with low selectivities.

Fenton’s reagent is believed to produce hydroxyl radicals¹⁹

(see, however, refs. 20). Experiments on the oxidation of disub-

stituted cyclohexanes demonstrated that vanadium-catalysed

reaction occurs without retention of conﬁguration, i.e., the

trans/cis ratio is approximately the same for oxidation of both

cis- and trans-isomers of the substituted cyclohexane and this

value is usually more than unity (Table 5). A similar situation

was found for the reaction with the “O₂–H₂O₂–FeSO₄” and

“O₂–H₂O₂–FeSO₄” systems. Various selectivity parameters for

the oxidations with participation of PDCA and perchloric acid

as co-catalysts are similar to those obtained for PCA.

Fig. 11 Cyclohexane (0.45 mol dm^Ϫ3) oxidation by hydrogen peroxide

(0.5 mol dm^Ϫ3) in a mixture acetic acid–water (99 : 1 v/v) at 40 ЊC

catalysed by NaVO₃(1.0 × 10^Ϫ4mol dm^Ϫ3) in the presence of PCA

(4.0 × 10^Ϫ4mol dm^Ϫ3

) (a) and in the absence of PCA (b).

Concentrations of cyclohexanol (A) and cyclohexanone (B) were

measured after the addition of an excess of solid triphenylphosphine.

The catalyst (NaVO₃) was introduced as a stock solution in water.

by imidazole-4,5-dicarboxylic acid, the yield of oxygenates

was found to be 2–3 times lower. It is very interesting that the

similar compounds pyrazole-3,5-dicarboxylic acid, imidazole-

4-carboxylic acid and 5-methyl-2-phenyl-1,2,3-triazole-4-

carboxylic acid turned out to be almost inactive in the cyclo-

hexane oxidation. One can assume that the eﬃcient accelerating

eﬀect of PCA, and cyclic amino acids very similar to it, is

associated with the ability of these compounds to form

chelating six-membered cycles with vanadium ion and possibly

also to form zwitterionic structures (see below for more detailed

discussion).

It is important to note that the systems known to oxidize

without the production of hydroxyl radicals exhibited diﬀerent

values of the selectivity parameters. For example, the oxid-

ations of alkanes with H₂O₂in CF₃COOH and with

C₆H₅COOOH in benzene exhibit much higher selectivities

compared to the oxidation with the “O₂–H₂O₂–n-Bu₄NVO₃–

PCA” system (Tables 3 and 5). The oxo-peroxo complex

Suitable solvents

Acetonitrile is the solvent of choice for the reaction under

discussion. The oxidation in acetone proceeds less eﬃciently.^6e

It is known that these solvents are not completely inert towards

1356

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Table 4 Oxidation of cyclohexane-d₁₂, adamantane and ethylbenzene/

cyclohexane^a

KIE,^b

k_H/k_D

Ethylbenzene/

cyclohexane

c

d

Reagent

2Њ : 3Њ

O₂–H₂O₂–n-Bu₄NVO₃–PCA^e

O₂–H₂O₂–hν ^f

1.2

2.2

1.9

2.8^h

0.75

0.7

0.25

0.5

0.8

f

O₂–H₂O₂–Fe(ClO₄)₃

[V(O)(O₂)(pic)(H₂O)₂]^g

1.6

C₆H₅COOOH in benzeneⁱ

0.09

^aSamples were reduced with triphenylphosphine before GLC analysis.

The reactions were in acetonitrile. ^bKinetic isotope eﬀect, determined

by comparison of oxidations of cyclohexane and cyclohexane-d₁₂.

^cRelative normalized reactivities of hydrogen atoms at secondary and

tertiary carbons of adamantane. ^dThe normalized reactivities of

methylene hydrogen atoms in ethylbenzene relative to those in cyclo-

hexane. ^eAt 50 ЊC. ^fAt room temperature. ^gStoichiometric oxidation at

room temperature by the complex with picolinate anion (pic); the syn-

thesis of this complex has been described in ref. 14a. ^hData from ref.

14a. ⁱPeroxybenzoic acid in benzene at 100 ЊC; data from ref. 15b.

[V(O)(O₂)(pic)(H₂O)₂], which contains only one molecule of

chelating ligand (picolinate, pic), oxidizes alkanes stoichio-

metrically with high selectivity (Table 3), and the reaction of

this complex with disubstituted cyclohexanes occurs with par-

tial retention of conﬁguration (cis-disubstituted cyclohexanes

give rise to the predominant formation of cis-tert-alcohol, see

Table 5). The reaction of alkanes with H₂O₂catalysed by the

combination “[L₂Mn₂O₃](PF₆)₂(L = 1,4,7-trimethyl-1,4,7-tri-

azacyclononane)–MeCOOH”,¹⁶which is believed to proceed

without formation of free hydroxyl radicals demonstrated

selectivities diﬀerent from those found for the “O₂–H₂O₂–n-

Bu₄NVO₃–PCA” system. Unlike the complex [V(O)(O₂)(pic)-

(H₂O)₂], the vanadium oxo-peroxo derivative 2, which contains

two anions of pyrazine-2-carboxylic acid as ligands, exhibits

stereoselectivity neither in stoichiometric nor in catalytic reac-

tions. This shows that the oxidations in the presence of complex

[V(O)(O₂)(pic)(H₂O)₂] and of compound 2 proceed via diﬀerent

mechanisms.

Production of hydroxyl radicals from hydrogen peroxide

Since the selectivity parameters obtained for oxidation by the

“O₂–H₂O₂–n-Bu₄NVO₃–PCA” system indicate that hydroxyl

radicals could be reactive intermediates in this reaction, we

undertook an additional study to prove the formation of

hydroxyl radicals. It should be noted that hydroxyl radicals

are believed to play a very important role in various chemical

and biochemical processes.^19a,d,g,21Hydroxyl radicals can be

generated from hydrogen peroxide under the action of

various transition metal ions such as iron(),¹⁹iron(),^22aand

copper(),^22b,ccopper(),^22dchromium(),^22e(see also ref. 22f ),

uranium(),^22gcobalt().^22hSome chemical and biochemical

systems involving vanadium are also known to generate

hydroxyl radicals.^22g,23

The spin trap method is widely used for the ﬁxation of

radicals that arise in various chemical and biochemical

reactions.^{21c,d,f,22d–h,24}In order to prove the formation of hydroxyl

radicals and their participation in hydrocarbon oxidations by

the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” system we used N-(benzyl-

idene)-tert-butylamine N-oxide (nitrone 3). The interaction of

nitrone 3 with a hydroxyl radical gives rise to the formation of

the radical adduct 4 [eqn. (1)].

We have found that a signal having a triplet structure with

β

g = 2.0057 0.0002, a_N= 13.75 mT and a_H= 2.25 mT appears

in the EPR spectrum (Fig. 12) of a solution containing n-

Bu₄NVO₃(1.0 × 10^Ϫ4mol dm^Ϫ3), PCA (4.0 × 10^Ϫ4mol dm^Ϫ3),

cyclohexane (0.46 mol dm^Ϫ3) and the nitrone 3 (0.05 mol dm^Ϫ3)

in acetonitrile after the addition of a 30% aqueous solution of

hydrogen peroxide (resulting concentration was 0.1 mol dm^Ϫ3).

These parameters of the EPR spectrum agree satisfactorily with

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1357

View Article Online

a

Table 5 Oxidation of disubstituted cyclohexanes (trans/cis ratio) in acetonitrile

Oxidation of

cis-

Decalin

trans-

Decalin

cis-1,2-

DMCH

trans-1,2-

DMCH

cis-1,4-

DMCH

trans-1,4-

DMCH

Reagent

O₂–H₂O₂–n-Bu₄NVO₃–PCA^b

2.1

0.9

1.0

5.3

5.9

1.3

3.4

1.15

0.24

3.0

2.5

2.4

5.0

3.8

5.4

1.4

2.7

8.8

6.0

0.75

0.85

0.8

0.7

1.45

1.5

1.6

1.2

O₂–H₂O₂–n-Bu₄NVO₃in MeCOOH^b

O₂–H₂O₂–n-Bu₄NVO₃–PCA in MeCOOH^b

O₂–H₂O₂–n-Bu₄NVO₃–PDCA^c

c

O₂–H₂O₂–n-Bu₄NVO₃–HClO₄

O₂–H₂O₂–hν^c

0.9

1.3

1.0

0.26

0.9

0.8

RC

1.0

1.2

0.9

ca. 8

0.6

0.8

1.5

1.65

1.0

1.6

1.3

1.4

1.5

3.4

1.5

1.3

c

O₂–H₂O₂–FeSO₄

O₂–H₂O₂–FeSO₄–PCA^c

[V(O)(O₂)(pic)(H₂O)₂]^d

NH₄[V(O₂)(O)(pca)₂] (compound 2, stoichiometrically)^c

O₂–H₂O₂–compound 2^c

4.5

H₂O₂in CF₃COOH^e

RC

C₆H₅COOOH in benzene^f

0.03

0.12

32

33

H₂O₂–[L₂Mn₂O₃]–MeCOOH^g

0.34

4.09

0.50

2.15

^aThe trans/cis ratio of isomers of tert-alcohols formed in the oxidation of disubstituted cyclohexanes was measured after treatment of the reaction

sample with triphenylphosphine. DMCH, dimethylcyclohexane; PDCA, pyrazine-2,3-dicarboxylic acid. ^bAt 40 ЊC. ^cAt room temperature. ^dStoi-

chiometric oxidation at room temperature by the complex with picolinate anion (pic); the synthesis of this complex has been described in Ref. 14a.

^eAt room temperature. Data from Ref. 15c. RC = retention of conﬁguration. ^fPeroxybenzoic acid in benzene at 100 ЊC; data from Ref. 15b. ^gFor the

catalysis with complex [L₂Mn₂O₃](PF₆)₂(L = 1,4,7-trimethyl-1,4,7-triazacyclononane) in the presence of carboxylic acid see Refs. 16.

petitive oxidation of benzene and various aliphatic alcohols.²⁵

According to this method, the [PhOH]₀/[PhOH] ratio is pro-

portional to k_A[ROH] for the reaction between hydroxyl

radicals and a mixture benzene–aliphatic alcohol, ROH. Here

[PhOH]₀and [PhOH] are phenol concentrations at a certain

moment chosen at the beginning of the reaction in the absence

(1)

and presence of the alcohol, ROH, respectively. The parameter

k_Ais the rate constant for the interaction of the hydroxyl radical

and ROH via hydrogen atom abstraction from the α-position of

ROH. Finally, [ROH] is the concentration of the alcohol added.

The existence of a linear correlation between the [PhOH]₀/

[PhOH] ratio and the constants k_Aof several diﬀerent alcohols

supports the participation of hydroxyl radicals in the key stage

of this reaction. We have found such a correlation^6cin the com-

petitive oxidation of benzene with methanol, ethanol and

n-propanol by the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” system and

using some parameters from ref. 25b.

Alkane transformations with participation of free radicals

Hydroxyl radicals are well known to be very reactive species,

and it is reasonable to assume that radicals produced from

hydrogen peroxide under the action of the “V–PCA” com-

bination will attack the hydrocarbon molecules.²⁶Hydroxyl

radicals easily abstract hydrogen atoms from alkanes according

to eqn. (2):

ؒ

HO ϩ RH → H₂O ϩ R

(2)

The rate constants for the reaction (2) lie in the interval

between 3 × 10⁸dm³mol^Ϫ1s^Ϫ1for the relatively inert cyclo-

hexane and ca. 10⁹dm³mol^Ϫ1s^Ϫ1for hydrogen abstraction from

the methyl group of methanol at ambient temperature. The rate

constant for the abstraction of methyl hydrogen atoms from

acetonitrile by hydroxyl radicals is much lower, 3.5 × 10⁶dm³

mol^Ϫ1s^Ϫ1. More detailed discussion of the rates of reaction (2)

Fig. 12 EPR Spectrum of a solution containing nitrone 3 (see the

text).

ؒ

will be given in the next sections. Alkyl radicals R react rapidly

with the molecular oxygen present in the solution²⁷(see also

refs. 2a, p. 21; 8a, p. 99; 8c, pp. 373, 382, 384, 406, 414, 417)

according to eqn. (3) to produce the corresponding alkylperoxyl

the corresponding values for the adduct of nitrone 3 with

hydroxyl radicals generated in the photolysis of hydrogen per-

oxide in acetonitrile (g = 2.0054 0.0002, a_N= 15.5 mT and

β

a_H= 2.35 mT), and with the parameters for adduct 4 prepared

by the photolysis of H₂O₂in an aqueous solution^24a(g = 2.0057,

ؒ

R ϩ O₂→ ROO

(3)

β

a_N= 15.3 mT and a_H= 2.75 mT).

27d,e

ؒ

We obtained additional evidence for hydroxyl radical forma-

tion in the system under discussion using the method of com-

radicals, ROO . The rate constants

for reaction (3) for

carbon-centred radicals range from 4 × 10⁶up to 5 × 10⁹dm³

1358

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

mol^Ϫ1s^Ϫ1. For example, values of k₃(dm³mol^Ϫ1s^Ϫ1)^27dat

ambient temperature have been reported as follows: benzyl,

2.5 × 10⁷; cyclohexyl, (3.4 0.6) × 10⁹; hydroxymethyl, 4.8 ×

10⁹; trichloromethyl, 3.3 × 10⁹. Although it is generally

accepted that at ambient temperature recombination (3) is the

sole pathway, in principle, disproportionation of two radicals

temperatures, with simultaneous formation of the alkyl hydro-

peroxide, is in accordance with reaction (12). However, it is

more reasonable to propose that the stable ketone (aldehyde)

and alcohol are produced in the alkyl hydroperoxide decom-

position under the action of a metal complex present in the

solution via reactions (14) and (15) (see, e.g., ref. 8c, pp. 375, 376

and ref. 29f ).

ؒ

(R and O₂) can occur to yield an alkene and the hydroperoxyl

radical:^27k

Ϫ

(nϩ1)ϩ

ROOH ϩ M^nϩ→ RO ϩ HO ϩ M

(14)

(15)

ؒ

R ϩ O₂→ R(ϪH) ϩ HOO

(4)

ROOH ϩ M^(nϩ1)ϩ→ ROO ϩ H ϩ M

ϩ

nϩ

ؒ

In addition, in the case of long hydrocarbon chains, iso-

merisation and cyclisation reactions of peroxyalkyl radicals are

possible.^28aFor example, self-abstraction in aliphatic long-chain

hydroperoxyl radicals (intramolecular H-atom abstraction) can

proceed, as shown in reaction (5):^28b

Other reactions of radical species possible in the solution,

especially at high temperatures, are depicted as eqns. (16)–

(22).^1a,8,29gSome of these reactions are very fast, for example,

ؒ

HO ϩ H₂O₂→ H₂O ϩ HOO

(16)

(17)

(18)

(19)

(20)

(21)

(22)

ؒ

CH₃CH₂CH₂OO → CH₃C HCH₂OOH

(5)

ؒ

HOO ϩ H₂O₂→ H₂O ϩ O₂ϩ HO

It is interesting that reactions of such a type occur much

more rapidly in the case of analogous organosilicon peroxy

radicals.^28cIn the presence of a donor of bromine atom, CCl₃Br,

there is a competition between recombination of R with O₂

and bromine atom abstraction, reaction (6), which readily takes

ؒ

HO ϩ HOO → H₂O ϩ O₂

ؒ

ROO ϩ R → ROOR

place in our system (see Fig. 1c).

ؒ

R ϩ H₂O₂→ RH ϩ HOO

ؒ

R ϩ CCl₃Br → RBr ϩ CCl₃

(6)

ؒ

2 HOO → H₂O₂ϩ O₂

It is important to remember here that alkane oxidation by

the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” reagent at low (<50 ЊC)

temperature gives rise to the clean formation of the alkyl

hydroperoxide as the sole product, at least in the initial period

of the reaction. Consequently, we have to assume that at low

temperatures the alkylperoxyl radical is smoothly converted

into the corresponding alkyl hydroperoxide [eqn. (7)]:

ؒ

RO ϩ ROOH → ROH ϩ ROO

the rate constants for processes (16)^26fand (18)^26aare 2.7 × 10⁷

and 9 × 10⁹dm³mol^Ϫ1s^Ϫ1, respectively, and k₂₂= 2.5 × 10⁸

dm³mol^Ϫ1s^Ϫ1for R = t-Bu at 22 ЊC (ref. 29g). Reaction (17)

proceeds very slowly, if at all.

However, since the current concentrations of free radicals

should be low, in our further consideration we will ignore many

of these processes, restricting the analysis to the initial period

of the oxidation reaction which occurs at low temperature. The

possibility of a radical chain mechanism is discussed in the

Electronic Supplementary Information (Appendix 1).

Ϫ

ϩ

ؒ

ROO ϩ e ϩ H → ROOH

(7)

To be transformed into the relatively stable alkyl hydro-

peroxide, the radical ROO should add a hydrogen atom or

consecutively add an electron and a proton. Hydrogen peroxide

could play the role of hydrogen atom donor, reaction (8), and

the alkyl hydroperoxide can be also produced via recombination

reactions of peroxyl radical, eqn. (9).

ؒ

Cleavage of C–C bonds in normal alkanes

In order to obtain additional evidence for the formation of

alkyl hydroperoxyl radicals in the oxidation with the reagent

under discussion, we decided to investigate more carefully the

composition of the products of the reaction with long-chain

normal saturated hydrocarbons. Destruction of the hydro-

carbon chain is known for alkane oxygenation processes which

involve the formation of alkoxyl radicals³⁰(see also ref. 8a,

pp. 12, 347). Usually this destruction occurs at high (>100 ЊC)

temperatures. We studied the oxidation of propane and n-

hexane with the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” reagent at room

temperature. The data obtained are summarized in Table 6. It

turned out that the content of the oxygenate mixture depends

on the concentrations of the catalyst and hydrogen peroxide.

Thus when the vanadium-containing catalyst is used at a con-

centration of 1.0 × 10

system A, only the products derived from the C–H bond split-

ting (i.e., propan-1-ol and propan-2-ol) are formed even after

24 h. When the initial concentrations of H₂O₂and the catalyst

are higher (systems B and C) some products of C–C bond

splitting (i.e., ethanol and methanol) can be detected in the

reaction mixture. The [C–C]/[C–H] ratio attains 7.6% for system

B. Similar results were obtained for the oxidation of propane

with hydrogen peroxide catalysed by Fe(ClO₄)₃, systems E and

F. It is interesting that, unlike propane, n-hexane is oxidized

with the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” reagent at room tem-

perature without C–C bond breakage. In contrast, the “O₂–

H₂O₂–Fe(ClO₄)₃” system oxidizes this alkane with extensive

ؒ

ROO ϩ H₂O₂→ ROOH ϩ HOO

(8)

(9)

ؒ

ROO ϩ HOO → ROOH ϩ O₂

IV

ؒ

The radical ROO could alternatively be reduced with V -

containing species produced in the catalytic cycle [see: reactions

(10) and (11)]:

ROO ϩ V^IV→ ROO^Ϫϩ V^V

ROO^Ϫϩ H^ϩ→ ROOH

(10)

(11)

ؒ

At low temperatures the reactions between peroxyl radicals

aﬀording intermediate or ﬁnal stable products, i.e., so-called

Russell termination^29b(12) and Vaughan termination^29c(13)

^Ϫ4mol dm^Ϫ3and [H₂O₂]₀= 0.16 mol dm^Ϫ3,

2 R¹R²CHOOH → R¹R²CHOOOOCHR¹R²→

R¹R²C᎐O ϩ O ϩ R¹R²CHOH (12)

᎐

2

ؒ

2 ROO

ROOOOR →

RO ϩ O₂ϩ OR → ROOR (13)

ؒ

are not probable in our case, because they do not correspond

with the products found in our oxidation (see also refs. 29d, e).

Generation of ketone (aldehyde) and alcohol at higher (>50 ЊC)

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1359

View Article Online

Table 6 Oxidation of n-alkanes by certain H₂O₂-containing systems under various conditions (products, concentration in mmol dm^Ϫ3

)

Systems

nBu₄NVO₃–

Products

PCA^a(A)

PCA^b(B)

PCA^c(C)

PCA^d(D)

Fe(ClO₄)₃^e(E)

Fe(ClO₄)₃^f(F)

Oxidation of propane

Propan-2-ol

Propan-1-ol

3.6

1.6

0

33

8

2.1

1.0

7.6

101

54

1.8

3.3

45^g

26^g

2.1^g

2.0^g

5.8

63

22

4.6

3.2

9.2

Ethanol

Methanol

Ratio [C–C]/[C–H] (%)^h

Oxidation of n-hexane

Hexan-2-ol

Hexan-3-ol

11

8

15

12

19

21

16

18

Hexan-1-ol

Pentan-1-ol

Butan-1-ol

Propan-1-ol

2.0

3.2

3.8

0.1

1.0

0.8

3.5

0.2

3.8

3.6

0.3

1.0

5.5

3.0

32

<0.05

traces

<0.5

0

<0.05

traces

<0.5

0

Ethanol

Methanol

Ratio [C–C]/[C–H] (%)^h

Conditions: solvent, MeCN; room temperature; in all cases, concentration of the products was determined after reduction of the sample with

NaBH₄; [propane]₀, 1.4 mol dm^Ϫ3; [hexane]₀, 0.3 mol dm^Ϫ3in the oxidations catalysed by Fe(ClO₄)₃and 0.5 mol dm^Ϫ3in the oxidation catalysed by

nBu₄NVO₃. ^a[nBu₄NVO₃], 0.1 mmol dm^Ϫ3; [PCA], 0.4 mmol dm^Ϫ3; [H₂O₂]₀, 160 mmol dm^Ϫ3; 24 h. ^b[nBu₄NVO₃], 0.5 mmol dm^Ϫ3; [PCA], 2.0 mmol

dm^Ϫ3; [H₂O₂]₀, 400 mmol dm^Ϫ3; 2 h. ^c[nBu₄NVO₃], 0.5 mmol dm^Ϫ3; [PCA], 2.0 mmol dm^Ϫ3; [H₂O₂]₀, 600 mmol dm^Ϫ3; 4 h. ^d[nBu₄NVO₃], 0.5 mmol

dm^Ϫ3; [PCA], 2.0 mmol dm^Ϫ3; [H₂O₂]₀, 400 mmol dm^Ϫ3; 24 h. ^e[Fe(ClO₄)₃], 0.4 mmol dm^Ϫ3; [H₂O₂]₀, 550 mmol dm^Ϫ3; 60 min. ^f[Fe(ClO₄)₃], 4.0 mmol

dm^Ϫ3; [H₂O₂]₀, 360 mmol dm^Ϫ3; 10 min. ^gReaction time, 10 min. ^hParameter [C–C]/[C–H] is the ratio of the total concentration of products obtained

via C–C bond cleavage to total concentration of products derived from C–H bond splitting.

formation of lower alcohols (pentanol, butanol, propanol,

ethanol and methanol), and at higher concentration of iron-

containing catalyst, system F, the [C–C]/[C–H] ratio is larger.

The autoxidation of styrene, initiated by free radicals, gives

benzaldehyde as a main product (ref. 8a, p. 356). It is known

that (E)-stilbene reacts with hydroxyl radicals in the presence of

oxygen in aqueous acetonitrile to aﬀord a high yield of benz-

aldehyde.^30eRecently we have found^6kthat the oxidation of

styrene and stilbene with the “O₂–H₂O₂–n-Bu₄NVO₃–PCA”

reagent at room temperature also proceeds with C–C bond

splitting and gives benzaldehyde and benzoic acid. These

data and the data given in Table 6 are in accordance with

our assumption that the vanadium-based reagent generates

hydroxyl radicals which are capable of converting the alkane

into the alkyl hydroperoxyl radical.

Intermediate vanadium oxo and peroxo complexes

In order to obtain additional insight into the mechanism

of alkane oxygenation with the “O₂–H₂O₂–n-Bu₄NVO₃–PCA”

reagent, we investigated by NMR spectroscopy the transform-

ations of various vanadium complexes under conditions close

to those used in the oxidation reaction. It turned out that the

⁵¹V NMR spectrum of the starting compound n-Bu₄NVO₃con-

sists of three signals at Ϫ539, Ϫ571, and Ϫ514 ppm. It is

important to note that the relative intensities of these signals

depended on the CD₃CN batch used to prepare the sample: for

wet acetonitrile the two outer signals predominated, whereas

for freshly dried acetonitrile the spectrum showed the central

line only. Both the H and the ⁵¹V NMR spectra showed that

treatment of n-Bu₄NVO₃with PCA in acetonitrile-d₃resulted in

the formation of a new product. The reaction is complete at

about 2 equiv. of PCA. Higher concentrations of PCA do not

lead to further chemical transformations. The product formed

from the initial vanadate species was the same whatever solvent

batch was used. The ﬁnal product has a single vanadium signal

at δ Ϫ532 and three proton signals of equal intensity in the

aromatic region from coordinated pca ligands: δ 9.22 (d,

J = 1.5, 2 H), 8.77 (d, J = 2.7, 2 H), and 8.45 (dd, J = 2.7, 1.5,

2 H) (see the Figs. 13a, 14a). The aliphatic signals due to

1

Fig. 13 500 MHz ¹H NMR spectra of (a) n-Bu₄NVO₃ϩ PCA (4

equiv.); (b) (n-Bu₄N)^ϩ[V(O)₂(pca)₂]^Ϫ(1); (c) n-Bu₄NVO₃ϩ PCA (4

equiv.) ϩ H₂O₂(1 equiv.); (d ) 1 ϩ H₂O₂(1 equiv.); (e) 1 ϩ H₂O₂(50

equiv.). Peaks indicated by asterisks are due to free PCA. Concentration

of the vanadium species was 5.0 × 10^Ϫ3mol dm^Ϫ3

.

tetrabutylammonium cation were observed at δ 3.11 (m, 8 H),

1.62 (m, 8 H), 1.37 (sextet, J = 7.4, 8 H), 0.98 (t, J = 7.3, 12 H).

It should be noted that in all systems containing NBu₄^ϩcation

we observed one set of aliphatic signals with chemical shifts

equal to population-weighted means of all charged species

1360

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

at δ Ϫ535 and Ϫ550 with the ratio 0.3 : 1 (Fig. 14c, d). Upon

increasing the concentration of H₂O₂these signals showed

small (a few ppm) upﬁeld shifts and their ratio changed to

0.14 : 1 at 0.5 mol dm^Ϫ3of H₂O₂(Fig. 14e). Our earlier study^7f

revealed the existence of two isomers for the peroxo complex 2,

with the two coordinated nitrogen atoms from the two N,O-

coordinated pca ligands (2b and 2c) being cis and trans. We

assume that the two vanadium signals belong to these isomers.

The smooth changes in the ⁵¹V chemical shifts and in the iso-

meric ratio of 2b and 2c observed upon increasing the concen-

tration of H₂O₂can be explained by gradual changes in solvent

composition. The lack of two separate sets of signals for 2b and

2c in the proton NMR could result from some form of dynamic

exchange which leads to time-averaged pca signals with signs of

exchange broadening. It is noteworthy that 2b and 2c are the

only vanadium species existing in the studied systems in the

presence of hydrogen peroxide. The ⁵¹V NMR spectra show

that complexes 2b and 2c are stable under catalytic conditions.

Even at very low vanadium concentrations (about 10^Ϫ4mol

dm^Ϫ3) and at high hydrogen peroxide concentrations (up to

0.5 mol dm^Ϫ3), which correspond to real catalytic conditions,

there are no signs of dissociation of 2 or further reaction of 2

with H₂O₂resulting in the formation of diperoxo complexes.

We have also studied the spectra of the reaction mixture

models containing perchloric acid instead of PCA. A sample

for ⁵¹V NMR was prepared in the following way. First, an

appropriate amount of 30% H₂O₂was added to a solution of

n-Bu₄NVO₃in acetonitrile-d₃. Then a diluted acetonitrile solu-

tion of HClO₄(0.12 mol dm^Ϫ3) was added to this mixture to

obtain the ﬁnal NMR sample with concentrations 10^Ϫ3mol

dm^Ϫ3of n-Bu₄NVO₃ϩ 4 × 10^Ϫ3mol dm^Ϫ3of HClO₄ϩ 0.1 mol

dm^Ϫ3H₂O₂. The resulting spectrum consisted of at least two

overlapping broad signals with almost identical chemical shifts

at about Ϫ600 ppm. These signals probably represent diﬀerent

peroxo species formed in acetonitrile solution. Recently, a

wide range of peroxovanadate complexes formed in water–

acetonitrile mixtures has been characterized by ⁵¹V NMR.^31a

The ⁵¹V signals for all peroxovanadate species moved down-

ﬁeld as the water concentration decreased, with particularly

dramatic changes being observed at lower water concentrations.

The minimum content of water in those studies was 2%. In the

systems we have examined, the content of water (introduced

with H₂O₂) was less than 1%, and it is not surprising that the

two signals we observed were located at somewhat lower ﬁeld

than most of the signals of the peroxovanadate species studied

in ref. 31a. We believe that our peaks belong to similar species

to those observed in ref. 31a when acid was added to the

[VO(O₂)₂]^Ϫsolution of low water concentration. Among those

species were [VO(O₂)₂]^Ϫ, [V₂O₂(O₂)₃] and some unidentiﬁed

complexes. We showed that the system “VO(acac)₂ϩ 4PC-

A ϩ H₂O₂(excess)” exhibits a moderate catalytic activity,

which is somewhat lower than for “n-Bu₄NVO₃ϩ 4PCA ϩ H₂O₂

(excess)”. It is of considerable interest to determine whether the

presence of acetylacetonato ligands and the diﬀerent oxidation

state of the metal ion in the starting vanadium complex result

in diﬀerent types of catalytically active species and hence in

diﬀerent types of catalytic mechanisms in these two systems.

Compound VO(acac)₂is a paramagnetic V^IVcomplex, and no

⁵¹V NMR signals can be observed in acetonitrile solution at

4 × 10^Ϫ3mol dm^Ϫ3of VO(acac)₂. The proton NMR spectrum

of VO(acac)₂does not reveal any signals in the diamagnetic

region. In the presence of 4 equiv. of PCA the ⁵¹V NMR spec-

trum of VO(acac)₂remains “blank”. However, the proton

NMR spectrum shows that PCA molecules replace acetyl-

acetonato ligands in the coordination sphere of vanadium.

Indeed, two sets of diamagnetic signals appear in the ¹H NMR

spectrum of a mixture of VO(acac)₂and PCA (4 equiv.); these

signals can be assigned to the keto and enol forms of free

uncoordinated acetylacetone molecules. Their integral inten-

sities correspond to a ketone/enol ratio of about 2 : 3. The

Fig. 14 131.5 MHz ⁵¹V NMR spectra of (a) n-Bu₄NVO₃ϩ PCA (4

equiv.); (b) (n-Bu₄N)^ϩ[V(O)₂(pca)₂]^Ϫ(1); (c) 1 ϩ H₂O₂(1 equiv.); (d ) n-

Bu₄NVO₃ϩ PCA (4 equiv.) ϩ H₂O₂(1 equiv.); (e) n-Bu₄NVO₃ϩ PCA

(4 equiv.) ϩ H₂O₂(100 equiv.). Concentration of the vanadium species

was 5.0 × 10^Ϫ3mol dm^Ϫ3

.

present in solution. The positions of these signals did not

provide any relevant information, therefore we shall omit the

aliphatic proton signals from our further discussion.

1

As can be seen from Figs. 13b and 14b, similar H and ⁵¹V

spectra with identical chemical shifts were obtained for

acetonitrile-d₃solutions of complex (n-Bu₄N)^ϩ[V(O)₂(pca)₂]^Ϫ

(anion 1),^7fand we can conclude that 1 is the only vanadium

complex existing in acetonitrile solutions of salt n-Bu₄NVO₃

and PCA when [PCA]₀/[n-Bu₄NVO₃] > 2. Addition of hydrogen

peroxide (30% aqueous) to a solution of 1 or a mixture of

(n-Bu₄N)^ϩ[VO₃]^Ϫϩ 4PCA produced a reddish solution. The

NMR spectra showed that 1–2 equiv. of H₂O₂were enough for

complete disappearance of 1. The three aromatic proton signals

mentioned above were replaced by ﬁve aromatic signals at

δ 9.86 (1 H), 9.40 (1 H), 9.32 (1 H), 9.24 (2 H) and 9.08 (1 H)

(see Fig. 14c–e). The signals are broadened, and information

about coupling constants is not readily available from the

spectra. It was shown earlier^7fthat H₂O₂can easily react with

compound 1 to yield the oxo-peroxo complex containing anion

[V(O₂)(O)(pca)₂]^Ϫ(2), and it is reasonable to assume that the

above signals belong to 2. The two pca ligands in 2 are not

equivalent, and one might expect to observe six diﬀerent

aromatic signals for this compound. Obviously, two of these

signals have very close chemical shifts in acetonitrile and are

not resolved even at 500 MHz.

In the presence of a slight excess of hydrogen peroxide the

signal observed at δ Ϫ532 in the ⁵¹V NMR spectra of 1 or of a

mixture of n-Bu₄NVO₃ϩ PCA was replaced by two new signals

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1361

View Article Online

Table 7 Oxidation of cyclohexane in the presence of naphthols^a

spectra demonstrate that the acac ligands in VO(acac)₂are

readily replaced in the presence of PCA by the stronger pca

ligands, with vanadium retaining its paramagnetic oxidation

state . In addition to the free acacH signals, the diamagnetic

region of the proton spectrum of the mixture contains three

broadened aromatic signals of equal integral intensities with

chemical shifts corresponding to free PCA (9.28, 8.84 and 8.72

ppm) which can be assigned to uncoordinated molecules of

PCA. Each vanadium ion coordinates two pca ligands, and

half of the PCA molecules in a mixture VO(acac)₂ϩ 4PCA

remain uncoordinated. These molecules, however, can form

outer-sphere adducts with paramagnetic vanadium species,

and this may be the reason for the excessive broadening of

their signals.

10²[Products]/mol dm^Ϫ3

10⁴[Additive]/

mol dm^Ϫ3

Additive

Cyclohexanone

Cyclohexanol

None

α-Naphthol

—

0.19

0.22

<0.01

0.13

5.2

6.5

0.43

4.7

0.49

1.0

5.0

1.0

5.0

β-Naphthol

<0.01

^aConditions: Bu₄NVO₃, 1.0 × 10^Ϫ4mol dm^Ϫ3; PCA, 4.0 × 10^Ϫ4mol

dm^Ϫ3; cyclohexane, 0.46 mol dm^Ϫ3; hydrogen peroxide, 0.5 mol dm^Ϫ3

;

in acetonitrile at 40 ЊC; 4 h. Concentrations of cyclohexanol and

cyclohexanone were measured after the addition of an excess of solid

triphenylphosphine.

Addition of H₂O₂to the mixture VO(acac)₂ϩ 4PCA pro-

duced a reddish solution containing diamagnetic vanadium

species with a detectable NMR spectrum. In the presence of a

10-fold excess of H₂O₂the ⁵¹V NMR spectrum consists of three

signals at Ϫ522, Ϫ542, and Ϫ545 ppm with the ratio of integral

intensities about 1 : 2 : 0.5. At higher concentrations of H₂O₂,

the relative intensity of the most downﬁeld component grows,

and at a 40-fold excess of H₂O₂the ratio of integrals is close

to 1 : 0.5 : 0.2. The aromatic region of the proton spectrum

reveals at least nine new broad partially overlapped signals with

chemical shifts from 9.2 to 9.9 ppm, which clearly belong to

coordinated pca ligands. The ⁵¹V and ¹H NMR signals

observed in systems “n-Bu₄NVO₃ϩ PCA (4 equiv.) ϩ H₂O₂

(excess)” and “VO(acac)₂ϩ PCA (4 equiv.) ϩ H₂O₂(excess)”

have diﬀerent chemical shifts, and at this moment we do not

know the exact nature of the species formed from VO(acac)₂in

the presence of PCA and H₂O₂. It is clear that a more detailed

study is required in order to identify all of the vanadium

complexes formed in this system. The conclusion that may be

drawn from the above experiments is that these species are dia-

magnetic complexes of vanadium() in which acetylacetonato

ligands are replaced by pca ligands.

We have found that pyridine-2,6-dicarboxylic acid (PyDCA),

which coordinates as a tridentate ligand (vanadium complexes

have been described^31b,c), is absolutely inactive as a co-catalyst

for cyclohexane oxidation in acetonitrile. To explain this fact,

we prepared a vanadium-containing complex of PyDCA by

mixing n-Bu₄NVO₃and PyDCA in a 1 : 1 ratio at elevated tem-

perature and isolated the resulting product. The vanadium

spectrum of this complex, which possibly is an oxo derivative,

in CD₃CN consists of a single sharp signal at Ϫ519 ppm, and

its proton spectrum in the aromatic region consists of two

signals: δ 8.46 (1 H, t, J = 7.7) and 8.15 (2 H, d, J = 7.7).

Addition of 1 equivalent of H₂O₂does not generate any

changes in the spectrum. Upon increasing the concentration of

H₂O₂to 50 equivalents, the aromatic signals shift slightly

downﬁeld (δ 8.48 and 8.18 ppm, correspondingly), and along

with the old signals we can clearly see small signals of a new

product at 8.39 ppm (1 H, t, J = 7.8) and 8.15 (2 H, d, J = 7.8).

Its vanadium signal is observed at Ϫ554 ppm. Immediately

after the addition of hydrogen peroxide the concentration of

the new complex (possibly an oxo-peroxo derivative) is low and

does not exceed 5% of the concentration of the starting species.

However, in 2 hours the spectrum of the same solution corre-

sponds already to a 2 : 3 mixture of the old and the new species,

and the spectrum obtained after 16 hours shows the complete

conversion of the starting species into the new complex. Obvi-

ously, an additional study is necessary for better understanding

of this process. Nevertheless, the above results already demon-

strate that PyDCA forms much stronger vanadium complexes

than PCA, thus hindering the coordination of hydrogen per-

oxide. Therefore, for systems with PyDCA the formation of

peroxo complexes becomes much more diﬃcult and proceeds

much more slowly. In addition, when a rigid tridentate PyDCA

ligand is coordinated to vanadium, a transformation of peroxo

ligands becomes impossible because this requires a certain

ﬂexibility of the coordination sphere. As a result, no catalysis is

observed.

Eﬀect of phenols on the oxidation of cyclohexane

It was useful for the understanding of the radical mechanism to

study the inﬂuence of certain “radical traps”^8a,32on the oxid-

ation process. It turned out that addition of 2,6-di-tert-butyl-

4-methylphenol (BHT, 0.01 mol dm^Ϫ3) to the reaction mixture

(cyclohexane, 0.23 mol dm^Ϫ3; hydrogen peroxide, 0.2 mol dm^Ϫ3;

n-Bu₄NVO₃, 1.0 × 10^Ϫ4mol dm^Ϫ3; PCA, 4.0 × 10^Ϫ4mol dm^Ϫ3;

50 ЊC, 1 h, acetonitrile) even slightly accelerates the oxidation:

the total yield of oxygenates was 0.076 mol dm^Ϫ3, while in the

absence of the additive the yield was 0.064 mol dm^Ϫ3. Hydroxyl

radicals are known to react with BHT,^8agiving a variety of

products. The known rate constants for the reaction of BHT

with other oxygen-centred radicals are as follows: 2.0 × 10⁷dm³

32a

mol^Ϫ1s^Ϫ1for Me₃CO at 295 K in acetonitrile and 2.5 × 10⁴

ؒ

dm³mol^Ϫ1s^Ϫ1for PhCH₂OO .^32fIf we accept the rate constant

ؒ

3 × 10⁸dm³mol^Ϫ1s^Ϫ1for the reaction between cyclohexane and

hydroxyl radicals and ∼10⁸dm³mol^Ϫ1s^Ϫ1for the interaction

between hydroxyl radicals and BHT, and take into account

the concentrations of cyclohexane and BHT in our solution

(0.23 and 0.01 mol dm^Ϫ3, respectively) we will see that hydrogen

atom abstraction from cyclohexane is one or two orders of

magnitude more rapid than that from BHT.^32bThus it is not

surprising that BHT in relatively low concentrations does not

retard the oxidation of cyclohexane. Moreover, since the BHT

concentration is much lower than that of cyclohexane, all of the

added BHT is consumed in the beginning of the process, after

which only the oxidation of cyclohexane can continue. Also no

retardation of cyclohexane oxidation was detected when very

low concentrations of other known “radical traps”, α- and β-

naphthols,^32awere added (1.0 × 10^Ϫ4mol dm^Ϫ3of the naphthol

per 1.0 × 10^Ϫ4mol dm^Ϫ3of n-Bu₄NVO₃) as shown in Table 7.

This is not surprising if we take into account the above-

mentioned calculations, assuming the rate constants for the

oxidation of cyclohexane and the naphthol to be comparable.

However, an unexpectedly sharp drop in the oxidation rate was

noticed when the naphthol was added in higher concentrations

(5 equivalents of the naphthol per 1 equivalent of vanadate

anion).

We assume that the naphthol enters the coordination sphere

of the vanadium complex and thus prevents coordination of

a hydrogen peroxide molecule to the vanadium ion. Indeed,

we have found that phenol itself interacts with the vanadium

peroxo species formed in acetonitrile solution, leading to rather

slow changes in the UV-visible spectrum of this solution

(Fig. 15, curves a, b, c and d). These changes might be associ-

ated with pca substitution by phenol in the vanadium coordi-

nation sphere and its subsequent oxidation. It is noteworthy

that BHT, when added to the solution, causes only negligible

changes (compare Fig. 15, curves a and e). Such a small eﬀect

could be explained if we take into account the very strong

spatial shielding of the hydroxy group in BHT by its two bulky

1362

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Fig. 17 Plot of the initial cyclohexane oxidation rate, W₀, versus

cyclohexane initial concentration (curve A) and its anamorphosis

(curve B) which was obtained using eqn. (26). Reaction conditions: [n-

Bu₄NVO₃]₀= 1.0 × 10^Ϫ4mol dm^Ϫ3[PCA]₀= 4.0 × 10^Ϫ4mol dm^Ϫ3

; ;

[H₂O₂]₀= 0.20 mol dm^Ϫ3, 40 ЊC, acetonitrile.

Fig. 15 UV-visible spectra of a solution of n-Bu₄NVO₃(2.0 × 10^Ϫ3

mol dm^Ϫ3), PCA (8.0 × 10^Ϫ3mol dm^Ϫ3) and H₂O₂(2.5 × 10^Ϫ2mol dm^Ϫ3

)

Information, Appendix 1) is impossible. The rate found experi-

mentally for the cyclohexane oxidation by the reagent

under consideration is 3 × 10^Ϫ6mol dm^Ϫ3s^Ϫ1. Due to the low

reactivity of the cyclohexyl peroxyl radicals, the value for the

oxidizability parameter will be consequently rather low,^33a

6.6 × 10^Ϫ5mol^Ϫ1/2dm^3/2s^Ϫ1/2at 60 ЊC. For such a value of the

oxidizability parameter, the experimentally measured rate of

the oxidation will be attained if the initiation rate equals

2 × 10^Ϫ3mol dm^Ϫ3s^Ϫ1at [CyH]₀= 1 mol dm^Ϫ3. Thus the chain

length for such a process is much less than unity, i.e., the process

is not a chain one. At low temperature and high hydrocarbon

concentration, when the oxidation rate does not depend on

[CyH]₀, the yield of alkyl hydroperoxide approximately corre-

sponds^6fto one half of the amount of H₂O₂consumed (Fig. 16),

which also supports the non-chain mode of the oxidation. In

our experiments at low temperatures, the yield of oxygenates

never exceeded 1 mole per 2 moles of hydrogen peroxide. At

relatively high temperature (e.g., 50 ЊC) some non-productive

H₂O₂decomposition to O₂and H₂O occurs (the so-called

catalase activity of the vanadium catalyst) (Fig. 16).

in the absence of any additive (a), in the presence of phenol (1.0 × 10^Ϫ2

mol dm^Ϫ3) after 20 (b), 60 (c) and 120 (d ) min at room temperature, as

well as the same solution in the presence of BHT (1.0 × 10^Ϫ2mol dm^Ϫ3

)

after 120 min (e). For comparison, spectra of quinone solution

(1.0 × 10^Ϫ2mol dm^Ϫ3) ( f ) and the solution containing n-Bu₄NVO₃

(0.4 × 10^Ϫ3mol dm^Ϫ3) and H₂O₂(0.625 × 10^Ϫ3mol dm^Ϫ3) (g) are also

shown.

Fig. 17 (curve A) demonstrates the dependence of the cyclo-

hexane oxidation rate on the cyclohexane initial concentration.

Such a behavior corresponds to the competition between cyclo-

hexane, RH, and acetonitrile for the same oxidative species,

i.e., hydroxyl radical [eqns. (23)–(26)]:

Fig. 16 Proﬁles of cyclohexane (0.46 mol dm^Ϫ3) oxidations by

hydrogen peroxide (0.24 mol dm^Ϫ3) in acetonitrile at 20 (a) and 50 ЊC

(b) catalysed by salt n-Bu₄NVO₃(1.0 × 10^Ϫ4mol dm^Ϫ3) and PCA

(4.0 × 10^Ϫ4mol dm^Ϫ3). Consumption of H₂O₂(curves 1) and

accumulation of the sum of the products (measured after addition of

an excess of solid triphenylphosphine) (curves 2) are shown.

(23)

(24)

(25)

tert-butyl groups. Due to this spatial hindrance BHT is unable

to coordinate to a vanadium centre via its hydroxy group and

thus to inhibit the oxidation. Simple phenols, which have

no steric hindrance around the hydroxy group, can possibly be

oxidized on the vanadium-containing reaction centre, but

these substrates have to be coordinated to vanadium prior to

oxidation. Such a coordination suggests a diﬀerent reaction

mechanism from that proposed for alkane oxidation.

(26)

Treatment of the data presented in Fig. 17, curve A using

eqn. (26) allows us to determine value W_g= 2.5 × 10^Ϫ6mol

dm^Ϫ3s^Ϫ1and the rate constant ratio k₂₅/k₂₄= 8.3 × 10^Ϫ3. Value

W_gcoincides with the maximum cyclohexane oxidation rate

at its high concentration. The k₂₅/k₂₄ratio is close to the

ratio of rate constants for the reactions between acetonitrile

and cyclohexane with hydroxyl radical. These parameters, taken

from the literature,^33b,care as follows: k₂₅= 3.5 × 10⁶dm³mol^Ϫ1

Kinetics of the oxygenation (hydroxyl radical generation)

Discussing the kinetics and mechanism of cyclohexane, CyH,

oxidation by the “O₂–H₂O₂–n-Bu₄NVO₃–PCA” reagent, we

ﬁrst of all must note that at temperatures <60 ЊC, the classical

radical-chain mechanism (see Electronic Supplementary

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1363

View Article Online

s^Ϫ1, k₂₄^cyclohexane= 3.0 × 10⁸dm³mol^Ϫ1s^Ϫ1and k₂₅/k₂₄= 1.2 ×

10^Ϫ2. It is of interest to calculate the relative reactivities of

methane and acetonitrile in their reactions with hydroxyl

radicals. Since k₂₄for CH₄is k₂₄^methane= 1.1 × 10⁸dm³mol^Ϫ1s^Ϫ1

(taken from ref. 33d) and accepting [CH₄] ≈ 1.0 mol dm^Ϫ3(at

reasonably high pressures) we obtain k₂₄^methane[CH₄] ≈ 1.1 × 10⁸

s^Ϫ1and k₂₅[MeCN] ≈ 0.5 × 10⁸s^Ϫ1. This means that the rate of

hydroxyl attack on methane will be approximately two times

higher than that on acetonitrile and, consequently, in the

acetonitrile solution methane will be oxidized, although oxy-

genates from acetonitrile can also be expected in the mixture of

products.^6h,j

It has been shown in previous sections and ref. 7f that

vanadate anions in acetonitrile solution are capable of forming

complexes with one and two PCA molecules. In the presence

of hydrogen peroxide, peroxo derivatives of vanadium() are

generated, as well as peroxo complexes of vanadium and

PCA. Based upon these data we can propose the following

formal sequence [eqns. (27)–(32)] of complexing stages in the

system under consideration:

Fig. 18 Dependence of the oxidation rate on the initial concentration

of PCA: A—plot of the experimentally obtained dependence; B—curve

for the dependence A obtained by calculation in accordance with eqn.

(37); C—linear anamorphosis of curve A in coordinate Γ Ϫ [PCA]₀,

where Γ is governed by eqn. (42). Reaction conditions are as described

in the caption to Fig. 17.

(27)

(28)

(29)

(30)

(31)

Fig. 18 shows the experimental dependence of the cyclo-

hexane oxidation rate on the initial PCA concentration. An

increase of the rate upon addition of the co-catalyst at low

concentration unambiguously shows that hydroxyl radical

generation is associated with the formation of complexes

between vanadium and PCA. The bell-shaped curve A observed

in Fig. 18 can be interpreted in terms of a mechanism corre-

sponding to eqns. (27)–(32) if a vanadium complex containing

one PCA molecule takes part in the rate-determining stage

of the oxidation. Vanadium complexes containing none or

two PCA molecules are much less eﬃcient as catalysts. The

dependence of the reaction rate on the initial H₂O₂concentra-

tion also exhibits a maximum (Fig. 19, curve A). In special

experiments we showed that water added to the reaction solu-

tion retards the oxidation process. The lowering of the oxida-

tion rate at high hydrogen peroxide concentration is due to the

increase in concentration of water introduced into the reaction

mixture simultaneously with hydrogen peroxide (usually we

employed 30 or 35% aqueous solution of H₂O₂). Taking this

inﬂuence into consideration we can see that at constant H₂O

concentration the reaction rate approaches an ultimate value

(curve B in Fig. 19). This means that a monoperoxo vanadium

complex is a catalytically active species.

(32)

It has been demonstrated in previous sections that, at [PCA]₀

> [V]₀(these parameters are initial concentrations of PCA and

a vanadium derivative, respectively), vanadium ions not bound

to PCA are practically absent in the solution. Taking this into

account we can neglect the concentration of V, compared with

the concentration of PCA–V, complexes present in the equation

of balance. The ﬁrst two equations simplify the situation since

the real stoichiometry of these stages should be written as eqns.

(27a) and (28a):

n-Bu₄N^ϩVO₃^Ϫϩ pcaH

VO₂(pca) ϩ n-Bu₄NOH (27a)

Given the fact that the cyclohexane oxidation is a non-chain

process and considering the data presented in Figs. 18 and 19,

we conceive that the rate-determining step of the oxidation

process is the monomolecular transformation of complex

V(PCA)H₂O₂. Therefore, the generation of hydroxyl radicals

can be described by the sequence shown as eqns. (33)–(35):

VO₂(pca) ϩ pcaH ϩ nBu₄NOH

n-Bu₄N^ϩ[VO₂(pca)₂]^Ϫϩ H₂O (28a)

In accordance with eqn. (27a) the interaction of vanadate

anion tetrabutylammonium salt with PCA gives rise to the for-

mation of n-Bu₄NOH, which will be further involved in reac-

tion (28a) when the second PCA molecule enters the vanadium

coordination sphere. We have demonstrated in special experi-

ments that the tetrabutylammonium cation (added in small

amounts in the form of the PCA salt) does not aﬀect the hydro-

carbon oxidation rate. Water present in substantial con-

centrations slightly retards the alkane oxidation (see below).

However the amount of water liberated in reaction (28a) should

not exceed 10^Ϫ4mol dm^Ϫ3(at the standard concentration of

vanadium derivative which we used in almost all the experi-

ments). Taking into account all these arguments, in our kinetic

calculations we can replace eqns. (27a) and (28a) by the more

simple eqns. (27) and (28), respectively, in which we consider

PCA as a indivisible ligand. This ligand can be conventionally

added or replaced as a whole in various reactions.

(33)

IV

V

Ϫ

ؒ

V (PCA) ϩ H₂O₂→V (PCA) ϩ HO ϩ HO (34)

V

ؒ

V (PCA)H₂O₂ϩ HOO →

V

ؒ

V (PCA) ϩ HO ϩ H₂O ϩ O₂(35)

Reaction (33) is a rate-limiting step. Because at high cyclo-

hexane concentrations all the hydroxyl radicals generated in the

system will interact with this hydrocarbon, the reaction rate can

be expressed by eqn. (36):

(36)

1364

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

(37)

described by eqn. (37) at [PCA]/[n-Bu₄NVO₃] ≥ 4 if we use

corresponding values of the equilibrium and rate constants

(see below). Secondly, the dependence of the reciprocal reaction

rate on the reciprocal hydrogen peroxide concentration [eqn.

(38)] should give a straight line:

(38)

More simply, in accordance with eqn. (37) the dependence

of function Π [eqn. (39)] on [H₂O₂]₀should be linear. Indeed,

(39)

the straight line D in Fig. 19 demonstrates the very close

correspondence of the model, i.e., eqn. (37), with experimental

data. Thirdly, the dependence of the reaction rate on the PCA

concentration is also in accordance with the experimental data.

Really, in accordance with eqn. (37) the maximum value for the

reaction rate at the ﬁxed H₂O₂concentration is attained at the

PCA concentration magnitude governed by expression (40).

(40)

Using eqn. (40) and the kinetic expression for the reaction

rate (37) we can obtain eqn. (41). It follows from this expression

that the left-hand side value of eqn. (41), which can be easily

(41)

Fig. 19 Dependence of the oxidation rate on the initial concentration

of hydrogen peroxide: A—plot of the experimentally obtained

dependence of the initial rate W₀on the initial H₂O₂concentration; B—

curve for the same dependence taking into account the retarding

inﬂuence of water; C—curve obtained from curve B using eqn. (37);

D—linear anamorphosis of curve B in coordinate Π—[H₂O₂]₀, where Π

is governed by eqn. (39). Reaction conditions are as described in the

caption to Fig. 17.

calculated using the experimental data, should be a linear func-

tion of the PCA concentration at a ﬁxed H₂O₂concentration.

Such a dependence of parameter Γ [eqn. (42)] on the PCA

(42)

concentration has been observed in reality and shown in the

form of line C in Fig. 18.

Thus, in order to obtain a kinetic expression for the oxidation

rate, it is suﬃcient to establish the dependence of [V(PCA)-

H₂O₂] on the initial concentrations of the reagents. In the

steady-state approximation the value [V^V(PCA)H₂O₂] can be

easily obtained if we accept [PCA]₀/[V]₀> 4, i.e., [PCA]₀Ϫ

[V]₀≈ [PCA]₀, [V]₀/[PCA]₀<< 1 and [V^V] >> [V^IV]. Using this

expression for the steady-state concentration of species

V^V(PCA)H₂O₂we will come to the kinetic equation for the

initial reaction rate given by eqn. (37).

To check the applicability of the proposed kinetic expression

to the experimental data we studied the dependence d[ROOH]/

dt on [V]₀under the condition [V]₀/[PCA]₀≤ 1/4 and quantita-

tively analysed the data presented in Figs. 18 and 19. Firstly,

in accordance with kinetic eqn. (37) the reaction rate is pro-

portional to the vanadium concentration (Fig. 20). The experi-

mental dependences depicted in Figs. 18 and 19 are tolerably

We can state that the reaction rate dependences on [V]₀,

[H₂O₂]₀and [PCA]₀are adequately characterized, if one accepts

the following values (at 40 ЊC) for the equilibrium constants:

K₂₈≤ 10 dm³mol^Ϫ1; K₂₉= 0.55 dm³mol^Ϫ1; K₃₀= 8 × 10^Ϫ3dm³

mol^Ϫ1; K₃₁K₂₈= 0.9 × 10³dm⁶mol^Ϫ2, and for the rate constant

k₃₃= 0.4 s^Ϫ1. For our typical experimental conditions

([H₂O₂]₀= 0.1 mol dm^Ϫ3, [V]₀= 1.0 × 10^Ϫ4and [PCA]₀=

1.0 × 10^Ϫ3mol dm^Ϫ3) we can calculate the following concentra-

tions of transient species (mol dm^Ϫ3): [V(PCA)₂] ≤ 6 × 10^Ϫ7,

[V(PCA)] ≈ 6 × 10^Ϫ5, [V(PCA)H₂O₂] ≈ 3.3 × 10^Ϫ6, [VH₂O₂] ≈

2.8 × 10^Ϫ5, [V(PCA)₂H₂O₂] ≈ 5.6 × 10^Ϫ6. The results of the cal-

culations are illustrated in the forms of curve B in Fig. 18 and

curve C in Fig. 19. Thus, it can be concluded that the proposed

kinetic scheme for this process adequately describes the whole

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1365

View Article Online

of true activation energy for the monomolecular decomposition

of complex V(PCA)H₂O₂and enthalpy for equilibrium (29),

E_eff= 17 2 kcal mol^Ϫ1; ii) eﬀective value for preexponent

factor A_eff= 7.2 × 10¹¹dm³mol^Ϫ1s^Ϫ1, which corresponds to

iii) eﬀective rate constant k_eff= 2k₃₃K₂₉= 0.44 dm³mol^Ϫ1s^Ϫ1at

40 ЊC.

Catalytic cycle for hydroxyl radical generation

In accordance with our experiments we assume that in the

initial period of the reaction any vanadium derivative under the

action of hydrogen peroxide (and possibly also dioxygen and

other components of the reaction mixture) is rapidly trans-

formed into a reactive form of the catalyst. Because the most

eﬃcient reaction is the oxidation catalysed by vanadate anion,

such transformations of ligands at the vanadium centre for the

case VCl₃can be conventionally presented as eqn. (46):

(46)

Fig. 20 Plots of the initial cyclohexane oxidation rate, W₀, versus the

[PCA]/[n-Bu₄NVO₃] ratio at constant [PCA] = (1.0 × 10^Ϫ3mol dm^Ϫ3

(curve A) and versus the concentration of the catalyst composition [V]₀

(1 equiv. of n-Bu₄NVO₃ϩ 4 equiv. of PCA) (curve B). Initial

concentration of H₂O₂was 0.2 mol dm^Ϫ3. Other conditions are as

described in the caption to Fig. 17.

)

It has been shown above that the interaction between the

᎐

vanadate anion and a PCA (᎐pcaH) molecule results in the

᎐

formation of the neutral dioxo complex (5) (Scheme 1) of

vanadium() having one pca ligand. This complex is in equi-

librium with the anionic dioxo complex (1) containing two pca

ligands [eqn. (28)]. However, our kinetic analysis established

that complex 5 is the precursor of the catalytically active species

responsible for generating hydroxyl radicals. Generation and

subsequent transformations of complex 5 were expressed for-

mally by eqns. (27), (29) and (33). In more detail these reactions

are depicted by eqns. (27a) and (28a). Complex 5 can coordi-

nate a H₂O₂molecule as shown previously in the formal

eqn. (29). More detailed presentation of this reaction is

depicted by Step 1 in Scheme 1. The following transformation is

the reduction of vanadium() to vanadium() by a hydrogen

peroxide molecule. This reaction, which is a rate-determining

stage, corresponds to a hydrogen atom abstraction by the ligand

set of the experimental data and properly expresses the main

features of the considered oxidation reaction. It is necessary to

emphasize that the conclusion about the direct proportionality

between the reaction rate (i.e., hydroxyl radical generation rate)

and concentration of the complex containing only one PCA

molecule is by no means in contradiction with our experimental

data. Indeed, on the one hand, at [PCA]₀> [V]₀practically all

the vanadium ions exist as complexes with PCA. According to

our NMR data, mixing vanadate, PCA and H₂O₂in a ∼1 : 4:1

ratio gives rise to the formation of the complex containing two

pca ligands per one vanadium. On the other hand, the max-

imum reaction rate has been observed at [PCA]₀/[V]₀> 4 rather

than at [PCA]₀/[V]₀= 1. This excess of the PCA concentration

over the vanadium concentration, necessary to reach the max-

imum reaction rate, is due to the competition between H₂O₂and

PCA for the sites in the vanadium coordination sphere. The

equilibrium constant K₃₁is reasonably high (≈10⁴dm³mol^Ϫ1)

and it is also evident that the concentration of the catalytically

active species V^V(PCA)H₂O₂[eqns. (29) and (32)] should be low

(≈3 × 10^Ϫ6mol dm^Ϫ3as stated above).

V

ؒ

of a V complex, which gives hydroperoxyl radical, HOO , and

oxo-hydroxy V^IVderivative 8. It is reasonable to assume that the

reduction occurs in two steps: proton transfer from coordinated

H₂O₂to one of the two oxo ligands with simultaneous form-

ation of a new covalent V–O bond (Step 2) and the subsequent

homolysis of this V–OOH bond (Step 3). The oxidation of

hydrogen peroxide by metal complexes is not surprising. For

example, according to the Haber–Weiss mechanism for the

“Fe^III–Fe^II–H₂O₂–H₂O” system (see also Electronic Supplemen-

tary Information, Appendix 2), the initiation stage of hydrogen

peroxide decomposition [reaction (47)] proceeds^8b,19awith the

When the concentrations of hydrogen peroxide and PCA are

not high, i.e., [H₂O₂]₀≤ 0.1 mol dm^Ϫ3and [PCA]₀≤ 2 × 10^Ϫ3

mol dm^Ϫ3and 10 ≥ [PCA]₀/[V]₀≥ 4, kinetic eqn. (37) can be

simpliﬁed to eqn. (43). Taking into consideration the constants

given above, we can transform eqn. (43) into expression (44),

ϩ

II

H₂O₂ϩ Fe^III→ HOO ϩ H ϩ Fe

(47)

ؒ

(43)

(44)

eﬀective rate constant k₄₇= 2 × 10^Ϫ4dm³mol^Ϫ1s^Ϫ1at 25 ЊC and

[H^ϩ] = 1.0 × 10^Ϫ2mol dm^Ϫ3.

The second hydrogen peroxide molecule adds to a vacant site

of complex 8, forming in Step 4 the V^IVcomplex 9. The other

proton transfer (Step 5) results in the formation of a dihydroxy-

hydroperoxy V^IVderivative which can exist in two forms,

10a and 10b. Both forms are known for various metals. We

assume that complex 10a decomposes via the homolysis of the

O–O bond in the hydroperoxy ligand (Step 6), generating a free

hydroxyl radical. This reaction seems to be favorable because it

which can be easily used for the evaluation of the expected

cyclohexane oxidation rates. If the hydrogen peroxide concen-

tration is low, [PCA]₀> K₂₉K₃₀[H₂O₂]₀, and the equation for the

reaction rate will be even simpler [eqn. (45)]. These conditions

IV

ؒ

leads to an oxygen-centred radical V –O (11a) which is a

mesomeric form of the stable oxo-vanadium() complex (11b).

The formation of a stable species can be “a driving force” for

the liberation of hydroxyl radicals. The mesomeric forms 11a

can be considered as the excited state of species 11b, existing

in the ground state. Oxo complexes of the type 11b can be

transformed into their excited states of the type 11a by light

(45)

have been used to study the temperature dependence of the

cyclohexane oxidation rate. The following parameters were

obtained: i) eﬀective activation energy, which equals the sum

1366

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

irradiation. Excited oxo complexes are stronger oxidizing re-

agents and easily abstract hydrogen atoms from alkanes;^{8b–d,34a–d}

this has been utilised in photocatalytic aerobic alkane

oxygenations.^{8b–d,34e–k}Vanadium oxo-peroxo complexes are also

more active in alkane oxidation if irradiated with light.^34lIn all

these cases light irradiation stimulates the electron transfer

from a ligand to a high-valent metal ion. A transformation

analogous to 10a → 11a (Step 6) with the O–O cleavage, for the

case of the V^Vderivative 7 looks less probable, because such a

decomposition of 7 could lead to an unfavorable V –O . This is

why complex 7 decomposes with V–O bond rupture (Step 3).

Finally, in order to transform complex 11b to the starting

species 5 active in catalysis, we need to transfer a proton from

one oxo ligand of 11b to another one and then to eliminate a

water molecule (Steps 7 and 8).

In this discussion we have not considered the role PCA in the

proposed transformations. However, it should be emphasized

that in the absence of PCA no catalytic oxidation takes place.

Moreover, PCA is the most eﬃcient co-catalyst and some

PCA analogues turned out to be very poor co-catalysts. We

therefore propose that the pca ligand plays the role of “a robot-

manipulator’s arm”, facilitating the proton transfer within the

vanadium complex. Scheme 2A illustrates this mechanism using

the transformation 6 → 7 (Step 2 in Scheme 1). The nitrogen

atom of the pca ligand can be de-coordinated from vanadium

and abstract a proton from the coordinated H₂O₂molecule to

form an ammonium base which, after rotation of “the robot’s

very poor co-catalysts.^7fTridentate pyridine-2,6-dicarboxylic

acid strongly binds to the vanadium ion and cannot rotate,

which leads to the inactivity of this compound as a co-catalyst.

We can assume that the pca-assisted proton transfer between

an H₂O₂molecule and the nitrogen atom of the pca ligand

(Scheme 2C) as well as between the protonated nitrogen and the

oxo ligand (Scheme 2D) proceeds via four-membered transition

states 13 and 14, respectively.

We can also propose that PCA facilitates the proton transfer

participating in the process either in the zwitterionic form

(structure 15 in Scheme 3) or as an amino acid with an inter-

molecular hydrogen bond (structure 16 in Scheme 3). In both

cases the close geometrical conformity between the parameters

of the vanadium complex and PCA molecule is very important.

It is not clear, however, whether in catalytic reactions PCA

molecules exist in solution as monomers or as H-bonded self-

associates. In order to detect the potential intermolecular self-

association we studied the concentration dependence of the

chemical shifts of the PCA aromatic protons in acetonitrile

solutions. One could expect that, in the case of intermolecular

hydrogen bonding, dilution will lead to the destruction of the

H-bonded self-associates, accompanied by changes in the

NMR chemical shifts of interacting molecules. It turned out

that the chemical shifts remained constant upon 200-fold

dilution of PCA from 1.2 × 10^Ϫ2to 6 × 10^Ϫ5mol dm^Ϫ3: δ 9.28

(d, J = 1.4, 1 H), 8.83 (d, J = 2.4, 1 H) and 8.72 (dd, J = 2.4, 1.4,

1 H). The fact that the chemical shifts of PCA are not con-

centration dependent can be explained by considering that

self-association for PCA is negligible. At the same time, PCA

molecules have both donor and acceptor groups capable of

strong hydrogen bonding. The most plausible interpretation is

that the hydroxy hydrogen of the PCA carboxylate group

is involved in intramolecular H-bonding with the adjacent

nitrogen atom (structure 16) and therefore is not available for

intermolecular complexation.

V

ؒ

arm”, approaches another ᎐O ligand of the complex and pro-

᎐

tonates it to produce an –OH ligand. After re-coordination of

the nitrogen to vanadium, complex 7 is formed. Analogously,

a pca ligand can facilitate the transformation 9 → 10a (Step 5

in Scheme 1). Another possible role of this ligand is proton

transfer between two hydroxy ligands of the vanadium com-

plex, resulting in water molecule extrusion from complex 11b

and formation of the catalytically active starting species 5

(Scheme 2B). Obviously, a close spatial correspondence should

be realized, i.e., all distances and angles between atoms must

accommodate these atoms properly to provide for the possi-

It is known that in some biological systems vanadium()

derivatives play the role of catalysts for hydroxyl radical

generation, which causes these compounds to be toxic. The

following processes [reactions (48)–(52)] occurring in the

presence of the biological reducing agent, NADPH, have been

proposed^23h,35a,bfor these systems:

bility of the H^ϩ→ :N , H^ϩ→ Ö᎐ and H^ϩ→ HÖ– transfers.

᎐

For example, imidazole-4-carboxylic acid and pyrazole-3,5-

dicarboxylic acid as ligands do not fulﬁl this condition and are

Scheme 1 Proposed catalytic cycle for hydroxyl radical generation.

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1367

View Article Online

Scheme 2 Possible role of PCA in proton transfer (“robot’s arm mechanism”) between coordinated H₂O₂molecule and OH ligand (A) and between

two OH ligands (B).

V^Vϩ O₂→ V –OO

(48)

Ϫ

IV

ؒ

IV

ؒ

V –OO ϩ NADPH → V –OOH ϩ NADP (49)

V^IV–OOH ϩ H^ϩ→ V^Vϩ H₂O₂

(50)

(51)

(52)

ϩ

Ϫ

ؒ

NADP ϩ O₂→ NADP ϩ O₂

V

Ϫ

V^IVϩ H₂O₂→ V ϩ HO ϩ HO

ؒ

The mechanism of step (52)^23ehas not been discussed in

detail.^23h,35a,bHowever, it is clear that there is some similarity

between the reactions (48)–(52) occurring in living cells and the

steps of the catalytic cycle depicted in Scheme 2. We can assume

that the accelerating role of PCA in the generation of hydroxyl

radicals which we have found for the “O₂–H₂O₂–n-Bu₄NVO₃–

PCA” reagent is played in living organisms by amino acids of

peptides or free amino acids of various cell components. Thus

we can conclude that processes that occur under the action of

the reagent mimic some biological reactions and this reagent

can be considered as a biomimetic system. Proton transfer

processes are of great importance for biochemical systems (see,

e.g., refs. 35c–h).

Creating the total catalytic cycle for the system under dis-

cussion, we should take into consideration not only the process

of hydroxyl radical production but also the interaction of

these radicals with the alkane molecules and all subsequent

reactions. These reactions were discussed above; see eqns. (2)–

Scheme 3 Facilitation of proton transfer by PCA molecules via six-

membered transition states.

(4), (7)–(22). The closed catalytic cycle is depicted in Scheme 4.

All starting compounds coming in to the cycle are shown

1368

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Scheme 4 Simpliﬁed catalytic cycle for alkane hydroperoxidation.

beneath the cycle and all outgoing, relatively stable products are

Inerton AW-HMDS and DANI-86.10; fused silica capillary

shown above. We realize, of course, that this is only a very

simpliﬁed formal scheme combining the various processes that

proceed in reality.

column 25 m × 0.32 mm × 0.25 µm, CP-WAX52CB; integrator

SP-4400) twice: before and after addition of an excess of solid

triphenylphosphine. Triphenylphosphine reduces hydrogen

peroxide to water and the alkyl hydroperoxide to the corre-

sponding alcohol, from which it is possible to determine the real

concentrations of the alkyl hydroperoxide, alcohol, and ketone

(aldehyde). This addition of PPh₃allows us also to avoid the

simultaneous injection into the GC of a substrate and hydrogen

peroxide.

Conclusions

The reagent described in the present paper consists of any

derivative of vanadium(), () or even () and should contain

PCA as an obligatory co-catalyst. Such a combination very

eﬃciently produces hydroxyl radicals from H₂O₂at relatively

low temperatures (<50 ЊC) if acetonitrile is used as the solvent.

These radicals can be used to initiate the aerobic oxidation of

saturated hydrocarbons. At higher temperatures extensive

decomposition of hydrogen peroxide occurs under the action of

V–PCA to give molecular oxygen and water. PCA turns out to

be a unique co-catalyst and only pyrazine-2,4-dicarboxylic acid

and picolinic acid can be compared with PCA. The role of PCA

in the process is thought to facilitate the proton transfer

between an H₂O₂molecule coordinated to vanadium and oxo

ligands. Similar mechanisms operate in enzymatic catalysis.

¹H and ⁵¹V NMR spectra were recorded on a JEOL

Eclipseϩ500 NMR spectrometer at room temperature, with

neat VOCl₃as external standard for ⁵¹V NMR (δ = 0). The

instrument was operated at 131.46 MHz for recording ⁵¹V

NMR spectra. Normally, a total of 4096 points were collected

over a spectral width of 52 630 Hz. The pulse width was 45Њ.

Vanadium spectra of 5 × 10^Ϫ3mol dm^Ϫ3species could be

obtained with good signal-to-noise ratio by accumulating 2000–

4000 transients, at about 11 per second. Spectra of diluted solu-

tions with 10^Ϫ4mol dm^Ϫ3concentration of vanadium species

required the accumulation of 100 000–500 000 transients. The

processing of the vanadium spectra included exponential

multiplication with the line broadening 15–50 Hz to improve

signal-to-noise ratio. All NMR spectra were obtained in

acetonitrile-d₃. Unless it is stated otherwise, the concentration

of vanadium species in the NMR samples was kept between

3 × 10^Ϫ3and 9 × 10^Ϫ3mol dm^Ϫ3.

Experimental

The compound n-Bu₄NVO₃, easily soluble in acetonitrile, was

prepared¹²by adding V₂O₅(4.0 g) to 200 mL of aqueous n-

Bu₄NOH solution (0.4 mol dm^Ϫ3), stirring the solution for 18 h,

ﬁltering oﬀ the small amount of insoluble material, and then

evaporating the solution to complete dryness under vacuum at

60 ЊC. This and other compounds soluble in acetonitrile,

such as VCl₃and VO(acac)₂, were used as stock solutions in

acetonitrile. Aliquots of these vanadium-containing stock solu-

tions and stock solutions of PCA or other co-catalysts were

introduced into the reaction mixture and the reaction started by

the addition of hydrogen peroxide. In the cases of the vanadium

derivatives insoluble in acetonitrile, the stock solution con-

taining both catalyst and co-catalyst was prepared prior to the

oxidation reaction. Thus, stirring a suspension of NH₄VO₃(2.9

mg) or VOSO₄ؒ5H₂O (6.1 mg) and PCA (12.5 mg) in 100 mL of

hot acetonitrile over a few hours gave a stock solution, which

was used (2 mL) to prepare the reaction mixture (5 mL) con-

taining vanadium (1.0 × 10^Ϫ4mol dm^Ϫ3) and PCA (4.0 × 10^Ϫ4

mol dm^Ϫ3). The oxidations of higher hydrocarbons were carried

out in air in thermostatted Pyrex cylindrical vessels with

vigorous stirring. The total volume of the reaction solution was

5 or 10 mL.

Acknowledgements

We thank the Russian Basic Research Foundation (Projects

Nos. 98-03-32015a and 98-03-32079) and the Swiss National

Science Foundation (Grant No. 21-50430.97) for support. The

work at Clemson University was supported by the National

Science Foundation (CHE-9700278). The authors are grateful

to Professor Bernard Meunier and Professor Ulf Schuchardt

for valuable discussions.

References

1 For parts 1–11 see refs. 6(a)–(k), respectively. Preliminary

communications of previous parts and related papers have also been

published (ref. 7).

2 (a) R. A. Sheldon and J. K. Kochi, Metal-Catalyzed Oxidations

of Organic Compounds, Academic Press, New York, 1981;

(b) G. B. Shul’pin, Organic Reactions Catalysed by Metal Complexes,

Nauka, Moscow, 1988; (in Russian); (c) A. Butler, M. J. Clague

and G. E. Meister, Chem. Rev., 1994, 94, 625; (d ) I. Yamanaka,

K. Morimoto, M. Soma and K. Otsuka, J. Mol. Catal. A: Chem.,

1998, 133, 251.

The samples of the reaction solutions were analysed by

GC (instruments LKhM 80/6; column 2 m; 5% Carbowax on

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1369

View Article Online

3 (a) C. Orvig, K. H. Thompson, M. Battell and J. H. McNeill, in

Metal Ions in Biological Systems, eds. H. Sigel and A. Sigel, vol. 31,

Vanadium and Its Role in Life, Marcel Dekker, Inc., New York, 1995,

p. 575; (b) D. C. Crans, A. D. Keramidas, H. Hoover-Litty,

O. P. Anderson, M. M. Miller, L. M. Lemoine, S. Pleasic-Williams,

M. Vanderberg, A. J. Rossomando and L. J. Sweet, J. Am. Chem.

Soc., 1997, 119, 5447; (c) Vanadium Compounds: Chemistry,

Biochemistry, and Therapeutic Applications, eds. A. S. Tracey

and D. C. Crans, Oxford University Press, New York, 1998;

(d ) D. Rehder, Coord. Chem. Rev., 1999, 182, 297; (e) A. T.

Kotchevar, P. Ghosh, D. D. DuMez and F. M. Uckun, J. Inorg.

Biochem., 2001, 83, 151.

( f ) H. Jiang, G. Chu, H. Gong and Q. Qiao, J. Chem. Res. (S),

1999, 288.

15 (a) N. C. Deno, E. J. Jedziniak, L. A. Messer, M. D. Meyer,

S. G. Stroud and E. S. Tomezsko, Tetrahedron, 1977, 33, 2503;

(b) J. Fossey, D. Lefort, M. Massoudi, J.-Y. Nedelec and J. Sobra,

Can. J. Chem., 1985, 63, 678; (c) G. A. Hamilton, J. R. Giacin,

T. M. Hellman, M. E. Snook and J. W. Weller, Ann. N. Y. Acad. Sci.,

1973, 212, 4.

16 (a) J. R. Lindsay Smith and G. B. Shul’pin, Tetrahedron Lett., 1998,

39, 4909; (b) G. B. Shul’pin and J. R. Lindsay Smith, Russ. Chem.

Bull., 1998, 47, 2379; (c) G. B. Shul’pin, G. Süss-Fink and

J. R. Lindsay Smith, Tetrahedron, 1999, 55, 5345; (d ) G. B. Shul’pin,

G. Süss-Fink and L. S. Shul’pina, J. Mol. Catal. A: Chem., 2001, 170,

17.

4 (a) W. Plass, Angew. Chem., Int. Ed. Engl., 1999, 38, 909;

(b) S. Macedo-Ribeiro, W. Hemrika, R. Renirie, R. Wever and

A. Messerschmidt, JBIC, 1999, 4, 209.

17 M. Newcomb and P. A. Simakov, Tetrahedron Lett., 1998, 39,

5 A. Butler, Coord. Chem. Rev., 1999, 187, 17.

965.

6 (a) G. B. Shul’pin, D. Attanasio and L. Suber, Russ. Chem. Bull.,

1993, 42, 55; (b) G. B. Shul’pin, A. N. Druzhinina and G. V. Nizova,

Russ. Chem. Bull., 1993, 42, 1327; (c) G. V. Nizova and G. B.

Shul’pin, Russ. Chem. Bull., 1994, 43, 1146; (d ) G. B. Shul’pin

and G. Süss-Fink, J. Chem. Soc., Perkin Trans. 2, 1995, 1459;

(e) G. B. Shul’pin, R. S. Drago and M. Gonzalez, Russ. Chem.

Bull., 1996, 45, 2386; ( f ) M. C. Guerreiro, U. Schuchardt and

G. B. Shul’pin, Russ. Chem. Bull., 1997, 46, 749; (g) G. B. Shul’pin,

M. C. Guerreiro and U. Schuchardt, Tetrahedron, 1996, 52, 13051;

(h) G. V. Nizova, G. Süss-Fink and G. B. Shul’pin, Tetrahedron,

1997, 53, 3603; (i) U. Schuchardt, M. C. Guerreiro and G. B.

Shul’pin, Russ. Chem. Bull., 1998, 47, 247; (j) G. Süss-Fink,

G. V. Nizova, S. Stanislas and G. B. Shul’pin, J. Mol. Catal. A:

Chem., 1998, 130, 163; (k) G. B. Shul’pin, Y. Ishii, S. Sakaguchi and

T. Iwahama, Russ. Chem. Bull., 1999, 48, 887.

7 (a) G. B. Shul’pin and G. V. Nizova, React. Kinet. Catal. Lett., 1992,

48, 333; (b) G. B. Shul’pin, D. Attanasio and L. Suber, J. Catal.,

1993, 142, 147; (c) G. V. Nizova, G. Süss-Fink and G. B. Shul’pin,

Chem. Commun., 1997, 397; (d ) G. Süss-Fink, H. Yan, G. V. Nizova,

S. Stanislas and G. B. Shul’pin, Russ. Chem. Bull., 1997, 46, 1801;

(e) G. V. Nizova, G. Süss-Fink, S. Stanislas and G. B. Shul’pin,

Chem. Commun., 1998, 1885; ( f ) G. Süss-Fink, S. Stanislas,

G. B. Shul’pin, G. V. Nizova, H. Stoeckli-Evans, A. Neels,

C. Bobillier and S. Claude, J. Chem. Soc., Dalton Trans., 1999, 3169;

(g) G. Süss-Fink, S. Stanislas, G. B. Shul’pin and G. V. Nizova, Appl.

Organomet. Chem., 2000, 14, 623; (h) A. Hoehn, G. Suess-Fink,

G. B. Shul’pin and G. V. Nizova, Ger. Oﬀen, DE 19,720,344,

1997 (Chem. Abstr., 1998, 128, 49802p); (i) Yu. N. Kozlov, G. V.

Nizova and G. B. Shul’pin, Russ. J. Phys. Chem., 2001, 75, 770;

8 (a) N. M. Emanuel, E. T. Denisov and Z. K. Maizus, Chain Reactions

of Hydrocarbon Oxidation in the Liquid Phase (in Russian), Nauka,

Moscow, 1965; (b) A. E. Shilov and G. B. Shul’pin, Activation and

Catalytic Reactions of Hydrocarbons (in Russian), Nauka, Moscow,

1995; (c) A. E. Shilov and G. B. Shul’pin, Activation and Catalytic

Reactions of Saturated Hydrocarbons in the Presence of Metal

Complexes, Kluwer Academic Publishers, Dordrecht, Netherlands

2000; (d ) A. E. Shilov and G. B. Shul’pin, Chem. Rev., 1997, 97, 2879.

9 (a) A. Kozlov, K. Asakura and Y. Iwasama, J. Chem. Soc., Faraday

Trans., 1998, 94, 809; (b) A. Kozlov, A. Kozlova, K. Asakura and

Y. Iwasama, J. Mol. Catal. A: Chem., 1999, 137, 223.

18 J. H. Baxendale and J. A. Wilson, Trans. Faraday Soc., 1957, 53, 344.

19 (a) C. Walling, Acc. Chem. Res., 1975, 8, 125; (b) T. Kurata,

Y. Watanabe, M. Katoh and Y. Sawaki, J. Am. Chem. Soc., 1988,

110, 7472; (c) J. M. C. Gutteridge, I. Zs.-Nagy, L. Maidt and

R. A. Floyd, Arch. Biochem. Biophys., 1990, 277, 422; (d ) E. A.

Karakhanov, S. Yu. Narin and A. G. Dedov, Appl. Organomet.

Chem., 1991, 5, 445; (e) I. Yamazaki and L. H. Piette, J. Am. Chem.

Soc., 1991, 113, 7588; ( f ) J. Jiang, J. F. Bank and C. P. Scholes,

J. Am. Chem. Soc., 1993, 115, 4742; (g) A. Ya. Sychev and V. G. Isak,

Usp. Khim., 1995, 64, 1183; (h) B. R. Van Dyke, D. A. Clopton and

P. Saltman, Inorg. Chim. Acta, 1996, 242, 57; (i) J. S. B. Park,

P. M. Wood, M. J. Davies, B. C. Gilbert and A. C. Whitwood, Free

Radical Res., 1997, 27, 447; (j) X. Zhang, D. Zhang, D. H. Busch

and R. van Eldik, J. Chem. Soc., Dalton Trans., 1999, 2751;

(k) J. S. B. Park, P. M. Wood, B. C. Gilbert and A. C. Whitwood,

J. Chem. Soc., Perkin Trans. 2, 1999, 923; (l) B. C. Gilbert,

S. Silvester and P. H. Walton, J. Chem. Soc., Perkin Trans. 2,

1999, 1115; (m) B. C. Gilbert, S. Silvester, P. H. Walton and

A. C. Whitwood, J. Chem. Soc., Perkin Trans. 2, 1999, 1891;

(n) I. L. Yurkova, H.-P. Schuchmann and C. von Sonntag, J. Chem.

Soc., Perkin Trans. 2, 1999, 2049.

20 (a) D. T. Sawyer, A. Sobkowiak and T. Matsushita, Acc. Chem. Res.,

1996, 28, 409; (b) C. Walling, Acc. Chem. Res., 1998, 31, 155;

(c) P. A. MacFaul, D. D. M. Wayner and K. U. Ingold, Acc. Chem.

Res., 1998, 31, 159; (d ) S. Goldstein and D. Meyerstein, Acc. Chem.

Res., 1999, 32, 54; (e) S. H. Bossmann, E. Oliveros, S. Göb,

S. Siegwart, E. P. Dahlen, L. Payawan, Jr., M. Straub, M. Wörner

and A. M. Braun, J. Phys. Chem. A, 1998, 102, 5542; ( f ) M. L.

Kremer, Phys. Chem. Chem. Phys., 1999, 1, 3595.

21 (a) D. M. Miller, G. R. Buettner and S. D. Aust, Free Radical Biol.

Med., 1990, 8, 95; (b) M. H. Schoenberg and H. G. Beger, Chem.

Biol. Interact., 1990, 76, 141; (c) I. Saito, M. Takayama,

T. Matsuura, S. Matsugo and S. Kawanishi, J. Am. Chem. Soc,,

1990, 112, 883; (d ) I. Yamazaki, L. H. Piette and T. A. Grover,

J. Biol. Chem., 1990, 265, 652; (e) X.-M. Pan, M. N. Schuchmann

and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2, 1993, 289, 1021;

( f ) H. Noda, K. Oikawa, H. Ohya-Nishiguchi and H. Kamada,

Bull. Chem. Soc. Jpn., 1994, 67, 2031; (g) W. Adam, J. Cadet,

F. Dall’Acqua, B. Epe, D. Ramaiah and C. R. Saha-Möller, Angew.

Chem., Int. Ed. Engl., 1995, 34, 107; (h) B. M. Aveline,

I. E. Kochevar and R. W. Redmond, J. Am. Chem. Soc., 1996, 118,

289; (i) J. S. Daniels and K. S. Gates, J. Am. Chem. Soc., 1996, 118,

3380.

22 (a) K. Sakurai, T. Miura and T. Ogiso, Chem. Pharm. Bull., 1990, 38,

993; (b) M. K. Eberhardt, G. Ramirez and E. Ayala, J. Org. Chem.,

1989, 54, 5922; (c) H. C. Sutton, J. Chem. Soc., Faraday Trans. 1,

1989, 85, 883; (d ) T. Ozawa and A. Hanaki, J. Chem. Soc., Chem.

Commun., 1991, 330; (e) X. Shi and N. S. Dalal, Arch. Biochem.

Biophys., 1990, 281, 90; ( f ) P. O’Brien and G. Wang, Inorg. Chim.

Acta, 1989, 162, 27; (g) M. M. Hamilton, J. W. Ejnik and

A. J. Carmichael, J. Chem. Soc., Perkin Trans. 2, 1997, 2491; (h)

G. V. Rumyantseva, L. M. Vayner, M. Yu. Tuvin, G. N. Novodarova

and M. E. Vol’pin, Izv. Akad. Nauk SSSR, Ser. Khim., 1989, 2679.

23 (a) R. J. Keller, R. A. R. P. Sharma, T. A. Grover and L. H. Piette,

Arch. Biochem. Biophys., 1989, 271, 40; (b) R. J. Keller, R. A. R. P.

Sharma, T. A. Grover and L. H. Piette, Free Radical Biol. Med.,

1989, 6, 15; (c) T. Ozawa and A. Hanaki, Chem. Pharm. Bull., 1989,

37, 1407; (d ) X. Shi and N. S. Dalal, Arch. Biochem. Biophys., 1990,

278, 288; (e) A. J. Carmichael, Free Radical Res. Commun., 1990, 10,

37; ( f ) A. J. Carmichael, FEBS Lett., 1990, 261, 165; (g) T. Ozawa,

A. Hanaki and F. Takazawa, Inorg. Chim. Acta, 1991, 188, 163;

(h) X. Shi and N. S. Dalal, Arch. Biochem. Biophys., 1991, 289, 355;

(i) K. Kan, J. Inorg. Biochem., 1999, 77, 157.

10 (a) U. Schuchardt, D. Mandelli and G. B. Shul’pin, Tetrahedron

Lett., 1996, 37, 6487; (b) H. Q. N. Gunaratne, M. A. McKervey,

S. Feutren, J. Finlay and J. Boyd, Tetrahedron Lett., 1998, 39, 5655;

(c) R. Saladino, P. Carlucci, M. C. Danti, C. Crestini and

E. Mincione, Tetrahedron, 2000, 56, 10031.

11 (a) G. B. Shul’pin and A. N. Druzhinina, Bull. Acad. Sci. USSR,

Div. Chem. Sci. (Engl. Transl.), 1992, 41, 346; (b) G. B. Shul’pin,

A. N. Druzhinina and L. S. Shul’pina, Pet. Chem. (Russia), 1993,

33, 321; (c) G. B. Shul’pin, M. M. Bochkova and G. V. Nizova,

J. Chem. Soc., Perkin Trans. 2, 1995, 1465; (d ) G. B. Shul’pin,

G. V. Nizova and Yu. N. Kozlov, New J. Chem., 1996, 20, 1243.

12 V. W. Day, W. G. Klemperer and A. Yagasaki, Chem. Lett., 1990,

1267.

13 G. Süss-Fink, L. Plasseraud, V. Ferrand, S. Stanislas, A. Neels,

H. Stoeckli-Evans, M. Henry, G. Laurenczy and R. Roulet,

Polyhedron, 1997, 17, 2817.

14 (a) H. Mimoun, L. Saussine, E. Daire, M. Postel, J. Fischer and

R. Weiss, J. Am. Chem. Soc., 1983, 105, 3101; (b) R. Curci and

J. O. Edwards, in Catalytic Oxidations with Hydrogen Peroxide as

Oxidant, ed. G. Strukul, Kluwer Academic Publishers, Dordrecht,

1992; (c) W. Adam and L. Hadjiarapoglou, in Organic Peroxygen

Chemistry, ed. W. A. Herrmann, Springer-Verlag, Berlin, 1993;

(d ) G. B. Payne, P. H. Deeming and P. H. Williams, J. Org. Chem.,

1961, 26, 659; (e) A. M. Rocha Gonsalves, R. A. W. Johnstone,

M. M. Pereira and J. Shaw, J. Chem. Res. (S), 1991, 208;

24 (a) J. R. Harbour, V. Chow and J. R. Bolton, Can. J. Chem., 1974, 52,

3549; (b) E. G. Janzen, Y. Y. Wang and R. V. Shetty, J. Am. Chem.

1370

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

View Article Online

Soc., 1978, 100, 2923; (c) M. Kohno, M. Yamada, K. Mitsuta,

Y. Mizuta and T. Yoshikawa, Bull. Chem. Soc. Jpn., 1991, 64,

1447; (d ) J.-L. Clément, B. C. Gilbert, W. F. Ho, N. D. Jackson,

M. S. Newton, S. Silvester, G. S. Timmins, P. Tordo and

A. C. Whitwood, J. Chem. Soc., Perkin Trans. 2, 1998, 1715.

25 (a) S. Ito and K. Sasaki, in The Role of Oxygen in Chemistry and

Biochemistry, eds. W. Ando and Y. Moro-oka, Elsevier, Amsterdam,

1988, p. 115; (b) S. Ito, K. Ueno, A. Mitarai and K. Sasaki, J. Chem.

Soc., Perkin Trans. 2, 1993, 255.

26 (a) L. S. Villata, J. A. Rosso, M. C. Gonzalez and D. O. Mártire,

J. Chem. Res. (S), 1997, 172; (b) I. Aliagas and S. Gronert, J. Phys.

Chem. A, 1998, 102, 2609; (c) F. Kramp and S. E. Paulson, J. Phys.

Chem. A, 1998, 102, 2685; (d ) N. I. Butkovsaya and D. W. Setser,

J. Chem. Phys., 1998, 108, 2434; (e) N. M. Donahue, J. G. Anderson

and K. L. Demerjian, J. Phys. Chem. A, 1998, 102, 3121;

( f ) P. Ulanski and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2,

1999, 165; (g) K. H. Becher, C. M. F. Dinis, H. Geiger and P. Wiesen,

Phys. Chem. Chem. Phys., 1999, 1, 4721; (h) F. Louis,

C. A. Gonzalez, R. E. Huie and M. J. Kurylo, J. Phys. Chem. A,

2000, 104, 2931.

27 (a) F. R. Mayo, Acc. Chem. Res., 1968, 1, 193; (b) K. U. Ingold, Acc.

Chem. Res., 1969, 2, 1; (c) J. A. Howard, Adv. Free Radical Chem.,

1972, 4, 49; (d ) B. Maillard, K. U. Ingold and J. C. Scaiano, J. Am.

Chem. Soc., 1983, 105, 5095; (e) S. L. Boyd and L. R. C. Barclay,

J. Am. Chem. Soc., 1990, 112, 5724; ( f ) G. E. Quelch, M. M. Gallo

and H. F. Schaefer III, J. Am. Chem. Soc., 1992, 114, 8239;

(g) O. Dobis and S. W. Benson, J. Am. Chem. Soc., 1993, 115, 8798;

(h) R. D. Bach, P. Y. Ayala and H. B. Schlegel, J. Am. Chem. Soc.,

1996, 118, 12758; (i) V. D. Knyazev and I. R. Slagle, J. Phys. Chem.

A., 1998, 102, 1770; (j) C.-J. Chen and J. W. Bozzelli, J. Phys.

Chem. A., 1999, 103, 9731; (k) M. Kranenburg, M. V. Ciriano,

A. Cherkasov and P. Mulder, J. Phys. Chem. A., 2000, 104, 915.

28 (a) R. K. Jensen, S. Korcek and M. Zinbo, J. Am. Chem. Soc., 1992,

114, 7742; (b) W.-T. Chan, I. P. Hamilton and H. O. Pritchard,

J. Chem. Soc., Faraday Trans., 1998, 94, 2303; (c) V. A. Radtsig and

S. N. Kostritsa, Kinet. Catal., 2000, 41, 242.

29 (a) V. L. Antonovskii, Kinet. Catal., 1995, 36, 335; (b) G. A. Russell,

J. Am. Chem. Soc., 1957, 79, 3871; (c) E. R. Bell, J. H. Raley,

F. F. Rust, F. H. Seubold, Jr. and W. E. Vaughan, Discuss. Faraday

Soc., 1951, 10, 242; (d ) Q. J. Niu and G. D. Mendenhall, J. Am.

Chem. Soc., 1992, 114, 165; (e) G. S. Tyndall, T. J. Wallington and

J. C. Ball, J. Phys. Chem. A, 1998, 102, 2547; ( f ) G. Lauterbach,

W. Pritzkow, T. D. Tien and V. Voerckel, J. Prakt. Chem., 1988, 330,

933; (g) A. Bravo, F. Fontana, F. Minisci and A. Seri, Chem.

Commun., 1996, 1843.

(e) M. K. Eberhardt and W. Valasco, Tetrahedron Lett., 1992, 33,

1165.

31 (a) C. Slebodnick and V. L. Pecoraro, Inorg. Chim. Acta, 1998, 283,

37; (b) H. Mimoun, P. Chaumette, M. Mignard, L. Saussine,

J. Fischer and R. Weiss, Nouv. J. Chim., 1983, 7, 467; (c) H. Mimoun,

M. Mignard and L. Saussine, J. Am. Chem. Soc., 1986, 108, 3711.

32 (a) M. Foti, K. U. Ingold and J. Lusztyk, J. Am. Chem. Soc., 1994,

116, 9440; (b) I. W. C. E. Arends, K. U. Ingold and D. D. M. Wayner,

J. Am. Chem. Soc., 1995, 117, 4710; (c) J. Kim, R. G. Harrison,

C. Kim and L. Que, Jr., J. Am. Chem. Soc., 1996, 118, 4373;

(d ) P. A. MacFaul, K. U. Ingold, D. D. M. Wayner and L. Que, Jr.,

J. Am. Chem. Soc., 1997, 119, 10594; (e) L. Valgimigli, J. T. Banks,

J. Lusztyk and K. U. Ingold, J. Org. Chem., 1999, 64, 3381;

( f ) A. L. Buchachenko and A. M. Vasserman, Stable Radicals,

Khimiya, Moscow, 1973, p. 378 (in Russian); (g) V. W. Bowry and

K. U. Ingold, Acc. Chem. Res., 1999, 32, 27.

33 (a) A. E. Semenchenko, V. M. Solyanikov and E. T. Denisov,

Neftekhimiya, 1971, 11, 555; (b) T. V. Khar’kova, I. L. Arest-

Yakubovich and V. V. Lipes, Kinet. Katal., 1989, 30, 954;

(c) M. Anbar, D. Meyerstein and P. Neta, J. Chem. Soc. (B),

1966, 742; (d ) G. V. Buxton, C. L. Greenstock, W. P. Helman and

A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513.

34 (a) G. B. Shul’pin, P. Lederer and E. Mácová, Bull. Acad. Sci.

USSR, Div. Chem. Sci., 1986, 35, 2422; (b) G. B. Shul’pin,

E. Mácová and P. Lederer, Russ. J. Gen. Chem., 1989, 59, 2329;

(c) G. B. Shul’pin and A. N. Kitaigorodskii, Russ. J. Gen. Chem.,

1989, 59, 2335; (d ) M. M. Kats and G. B. Shul’pin, Bull. Acad. Sci.

USSR, Div. Chem. Sci., 1990, 39, 2233; (e) G. B. Shul’pin and

M. M. Kats, Russ. J. Gen. Chem., 1991, 61, 684; ( f ) G. B. Shul’pin

and M. M. Kats, Bull. Acad. Sci. USSR, Div. Chem. Sci., 1989,

38, 2202; (g) G. B. Shul’pin and M. M. Kats, Russ. J. Gen. Chem.,

1989, 59, 2447; (h) G. B. Shul’pin, A. N. Druzhinina and

G. V. Nizova, Bull. Acad. Sci. USSR, Div. Chem. Sci., 1991, 40,

2145; (i) G. B. Shul’pin and A. N. Druzhinina, React. Kinet. Catal.

Lett., 1991, 44, 387; (j) G .B. Shul’pin and G. V. Nizova, React.

Kinet. Catal. Lett., 1991, 45, 7; (k) G. V. Nizova, J. Muzart

and G. B. Shul’pin, React. Kinet. Catal. Lett., 1991, 45, 173;

(l) D. Attanasio, L. Suber and G. B. Shul’pin, Bull. Russ. Acad. Sci.,

1992, 41, 1502.

35 (a) S. I. Liochev and I. Fridovich, Free Radical Biol. Med., 1989, 6,

617; (b) S. I. Liochev and I. Fridovich, Arch. Biochem. Biophys.,

1990, 279, 1; (c) M. Dion and E. C. Webb, Enzymes, Longman

Group Ltd., London, 1979; (d ) H. Dugas and C. Penney, Bioorganic

Chemistry: A Chemical Approach to Enzyme Action, Springer-

Verlag, New York, 1981; (e) M. L. Bender, R. J. Bergeron and

M. Komiyama, The Bioorganic Chemistry of Enzymatic Catalysis,

J. Wiley & Sons, New York, 1984; ( f ) J. Bernadou and B. Meunier,

Chem. Commun., 1998, 2167; (g) W. Nam, H. J. Lee, S.-Y. Oh,

C. Kim and H. G. Jang, J. Inorg. Biochem., 2000, 80, 219;

(h) Y. Nishida, Inorg. Chem. Commun., 2000, 3, 310.

30 (a) A. L. Perkel’, S. G. Voronina and B. G. Freidin, Usp. Khim., 1994,

63, 793; (b) W.-T. Chan, H. O. Pritchard and I. P. Hamilton, Phys.

Chem. Chem. Phys., 1999, 1, 3715; (c) M. Weber and H. Fischer,

J. Am. Chem. Soc., 1999, 121, 7381; (d ) C. Fittschen, H. Hipper

and B. Viskolcz, Phys. Chem. Chem. Phys., 2000, 2, 1677;

J. Chem. Soc., Perkin Trans. 2, 2001, 1351–1371

1371

Article Doi

Oxidations by the reagent "O2-H2O2-vanadium derivative-pyrazine-2-carboxylic acid". Part 12. Main features, kinetics and mechanism of alkane hydroperoxidation

DOI: 10.1039/b101442k

Source and publish data:

Authors:

Article abstract of DOI:10.1039/b101442k

Full text of DOI:10.1039/b101442k

Products guided by the article

R&D Labs maybe for 33936-17-3

Relevant to this article

Hot Product