Reactivity of C-H bonds in cyclohexanone and 1-tert-butylperoxycyclohexanol toward the tert-butylperoxyl radical

Article abstract of DOI:10.1134/S0023158413020122

The kinetics of oxygen uptake and the composition of the cyclohexanone oxidation products in the azobisisobutyronitrile-initiated oxidation of cyclohexanone in the presence of tert-butyl hydroperoxide have been investigated by the Howard-Ingold method. The partial rate constants of the reaction of the tert-butylperoxyl radical with the C-H bonds of cyclohexanol and 1-tert-butylperoxycyclohexanole at 333 K have been determined. The carbonyl group of cyclohexanone activates the C-H bonds in the 2- and 6-positions (α) and deactivates the C-H bonds in the 3- and 5-positions (β) compared to the C-H bonds in the 4-position (γ), whose reactivity is similar to that of the methylenic C-H bonds in cyclohexane. Evaluation of the joint effect of the hydroxyl and tert-butylperoxyl groups in 1-tert-butylperoxycyclohexanol suggests a considerable deactivation of the C-H bonds in the 2- and 6-positions (β) and, to a lesser extent, in the 3- and 5-positions (γ).

Full text of DOI:10.1134/S0023158413020122

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2013, Vol. 54, No. 2, pp. 139–148. © Pleiades Publishing, Ltd., 2013.
Original Russian Text © S.V. Puchkov, Yu.V. Nepomnyashchikh, E.S. Kozlova, A.L. Perkel’, 2013, published in Kinetika i Kataliz, 2013, Vol. 54, No. 2, pp. 148–157.
Reactivity of C–H Bonds in Cyclohexanone
and 1ꢀtertꢀButylperoxycyclohexanol
toward the tertꢀButylperoxyl Radical
S. V. Puchkov, Yu. V. Nepomnyashchikh, E. S. Kozlova, and A. L. Perkel’*
Faculty of Chemical Engineering, Kuzbass State Technical University, Kemerovo, 650026 Russia
*eꢀmail: perkel2@rambler.ru
Received March 26, 2012
Abstract—The kinetics of oxygen uptake and the composition of the cyclohexanone oxidation products in
the azobisisobutyronitrileꢀinitiated oxidation of cyclohexanone in the presence of tertꢀbutyl hydroperoxide
have been investigated by the Howard–Ingold method. The partial rate constants of the reaction of the tert
butylperoxyl radical with the C–H bonds of cyclohexanol and 1ꢀtertꢀbutylperoxycyclohexanole at 333 K have
been determined. The carbonyl group of cyclohexanone activates the C–H bonds in the 2ꢀ and 6ꢀpositions
) and deactivates the C–H bonds in the 3ꢀ and 5ꢀpositions ( ) compared to the C–H bonds in the 4ꢀposiꢀ
tion ( ), whose reactivity is similar to that of the methylenic C–H bonds in cyclohexane. Evaluation
of the joint effect of the hydroxyl and tertꢀbutylperoxyl groups in 1ꢀtertꢀbutylperoxycyclohexanol suggests a
considerable deactivation of the C–H bonds in the 2ꢀ and 6ꢀpositions ( ) and, to a lesser extent, in the 3ꢀ and
5ꢀpositions ( ).
ꢀ
(α
β
γ
β
γ
DOI: 10.1134/S0023158413020122
INTRODUCTION
ical with the αꢀС–Н bond of cyclohexanone is 6 times
higher than that of the reaction between the same radꢀ
ical and the C–H bond of cyclohexane [6].
Cyclohexanone is the target product of the indusꢀ
trialꢀscale cyclohexane oxidation in caprolactam proꢀ
duction via the benzene oxidation technology [1, 2].
The process is conducted to a low hydrocarbon conꢀ
version (4–5%) because cyclohexanone has a higher
oxidizability compared to cyclohexane [1, 2]. The furꢀ
ther transformations of cyclohexanone are usually
It is usually difficult to study the reactivity of the
C–H bonds in cyclohexanone and other ketones by
the Howard–Ingold method because of the formation
of the product of reversible hydroperoxide addition to
the carbonyl group (hemiperketal) [5], which
decreases the effective ketone concentration. Data on
the reactivity of hemiperketals and on the chemical
mechanism of their oxidation are lacking in the literaꢀ
ture. This makes it difficult to estimate the errors
introduced by hemiperketal formation and oxidation
into the overall process. At the same time, a kinetic
analysis shows that data on the reactivity of all types of
C–H bonds in cyclohexanone and hemiperketal
(1ꢀtertꢀbutylperoxycyclohexanol) can be obtained by
investigating the influence of the ketone and/or
hydroperoxide concentration on the oxidation kinetꢀ
associated with its oxidation at the
α
ꢀС–Н bonds
nearest to the functional group. As a rule, the pubꢀ
lished data on the reactivity and composition of the
products refer to this position [1, 3–8]. Data on the
reactivity of the
βꢀ
and ꢀС–Н bonds of cyclohexꢀ
γ
anone and on the chemism and composition of the
resulting products are lacking, and, for this reason,
these bonds are ignored in the examination of the oxiꢀ
dation mechanism [5–8].
The reactivity of the ꢀС–Н bonds of cyclohexꢀ
α
anone and other ketones toward their own radicals is
12–16 times higher than that of the methylenic C–H
bonds of hydrocarbons [4]. The relative reactivities of ics and composition of the products related to all reacꢀ
the same bonds toward the standard tertꢀbutyl and
cumylperoxyl radicals differ noticeably. On the one
hand, the reactivity of the cyclohexanone molecule in
its reactions with tertꢀbutylperoxyl radicals at 353 K is
13 times higher than that of the cyclohexane molecule
[5]. Therefore, the tertꢀbutylperoxyl radical reacts
tive positions in the cyclohexanone and hemiperketal
molecules.
The purpose of this work is to estimate the reactivꢀ
ity of the methylenic C–H bonds of cyclohexanone
and 1ꢀtertꢀbutylperoxycyclohexanone in their reacꢀ
tions with the tertꢀbutylperoxyl radical. These data are
expected to be useful for modeling of byꢀproduct forꢀ
mation in industrial cyclohexanone synthesis by
with the
α
ꢀС–Н bond of cyclohexanone 39 times
more rapidly than with the methylenic C–H bond of
cyclohexane. On the other hand, at 348 K the partial
rate constant of the reaction of the cumylperoxyl radꢀ cyclohexane oxidation.
139

140
PUCHKOV et al.
EXPERIMENTAL
The peroxide content was determined by the coloꢀ
rimetric method based on the treatment of the sample
with the reagent containing Fe2+ ions and
ꢀdimeꢀ
thylꢀ ꢀphenylenediamine ( r = 0.04–0.07, = 8,
0.95) [15].
Cyclohexanone free of other products of cyclohexꢀ
N,N
n
ane oxidation was isolated from its bisulfite derivative.
According to GLC data, the purity of cyclohexanone
thus obtained was at least 99.5 0.5%.
p
s
P =
The Howard–Ingold method was used to deterꢀ
mine the empirical rate constant for the reaction of
cyclohexanone with the tertꢀbutylperoxyl radical [16].
tertꢀButyl hydroperoxide (TBHP) was synthesized
via a standard procedure [9]. According to iodometꢀ
ric data, the product contained 99.3 0.3% hydroꢀ
peroxide.
In the presence of high initial concentrations of tert
ꢀ
butyl peroxide (>0.4 mol/L), the oxidation of the subꢀ
strate results in the complete exchange of the peroxyl
radicals of the substrate and initiator for tertꢀbutylperꢀ
oxyl radicals [12, 16, 17]. Since it is difficult to deterꢀ
mine the initiation rate due to hydroperoxide decomꢀ
position in these reaction mixtures, the mixed initiaꢀ
tion method based on the introduction of an
additional initiator with a known initiation rate is
commonly used to increase the reliability of determiꢀ
nation.
2ꢀHydroxycyclohexanone, 1,2ꢀcyclohexanedione,
and 2ꢀcyclohexenone were also synthesized using
standard procedures [10, 11].
Azobisisobutyronitrile (AIBN) was purified by
recrystallization from ethanol to a constant rate of
substrate oxidation. Chlorobenzene (highꢀpurity
grade) was distilled.
The gas uptake kinetics in cyclohexanone oxidaꢀ
tion with dioxygen in chlorobenzene was studied using
a manometric setup at 333 K. The concentration of
tertꢀbutyl hydroperoxide was varied in the range from
0.5 to 2.5 mol/L. The stirring intensity used in the
experiments ensured that the process was kinetically
controlled. The mixed initiation method was used
(AIBN and TBHP served as initiators). Oxidized
cyclohexanone samples used in the study of the comꢀ
position and kinetics of product accumulation were
obtained under the same conditions.
The optimum values of parameters of the kinetic
equations were determined using a program based on
the leastꢀsquares method. The objective function was
optimized by the Fletcher–Reeves conjugate gradient
method.
RESULTS AND DISCUSSION
The oxidation products were quantitatively deterꢀ
mined by GLC on a column packed with 20% diethylꢀ
ene glycol succinate on Chromaton NꢀAWꢀHMDS.
Oxidation products were identified by comparing the
retention times of the sample components to those of
the authentic compounds. The retention times of
4ꢀacetoxycyclohexanone and 2,3ꢀdihydroxycyclohexꢀ
anone in acetylated oxidized cyclohexanone were
compared with the retention times of the tracers
obtained by the oxidation of cyclohexyl acetate and
1,2ꢀcyclohexanediol, respectively, with dioxygen at
373 K [12].
For solving the stated problem (determination of
partial rate constants of the reaction of the tertꢀbutylpꢀ
eroxyl radical with the C–H bonds of cyclohexanone),
it was necessary to determine the empirical rate conꢀ
stant for the reaction of the tertꢀbutylperoxyl radical
with ketone and hemiperketal molecules and the comꢀ
position of the products that characterize the direction
of cooxidation of these compounds using the
Howard–Ingold method.
In the reaction mixture, tertꢀbutyl hydroperoxide
adds to cyclohexanone (R1H) to form 1ꢀtertꢀbutylperꢀ
oxycyclohexanol (R2H). Under the conditions of
mixed initiation by AIBN and TBHP, the cooxidation
of a mixture of R1H and R2Н occurs according to the
following scheme:
2ꢀHydroperoxycyclohexanone and 2ꢀhydroxycyꢀ
clohexanone were determined as their combination in
acetate form [13]. 1,2ꢀCyclohexanedione, (4ꢀhydroxꢀ
ycyclohexanone + 4ꢀhydroperoxycyclohexanone),
and (2,3ꢀdihydroperoxycyclohexanone + 2,3ꢀdihyꢀ
droxycyclohexanone) were determined similarly. Prior
to the determination of 2ꢀcyclohexenone, the samples
were reduced with triphenylphosphine. 2,3ꢀEpoxycyꢀ
clohexanone was analyzed in unreduced samples,
because it is reducible to 2ꢀcyclohexenone [14].
(CH3)2(CN)CN=NC(CN)(CH3)2
+2O2
w
i, AIBN
•
2(CH3)2(CN)C
2(CH3)2(CN)COO•
,
–N
2
(I)
(CH3)3CO• + HO•
(CH3)3COOH
w
i, eff, AIBN
(CH3)3CO• + (CH3)3COO• + H2O
R2C(OH)O• + (CH3)3CO•
(CH3)3COOH
(II)
R2C=O
,
R2C(OH)OOC(CH3)3
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

REACTIVITY OF C–H BONDS IN CYCLOHEXANONE
141
(
CH3)2
(CH3)3CO•
R2C(OH)O•
(CH3)3COO•
(
CN
)
COO•
(CH3)2
(
CN)
COOH + (CH3)3COO•
(CH3)3COH + (CH3)3COO•
(III)
(CH3)3COOH
R2C(OH)OH –H O R2C=O,
2
–
emp
emp
k
where
k
and
k
are the empirical rate conꢀ
1
2.1
2.2
R1H + (CH3)3COOH
R2H,
(IV)
(V)
k
stants of the chain propagation reactions (V) and (VI)
involving cyclohexanone and 1ꢀtertꢀbutylperoxycycloꢀ
–1
emp
2.1
R1H + (CH3)3COO•–(CH ) COOHR•1
k
+O
2
R1OO•
,
hexanol, respectively;
and
are the equilibꢀ
[R2H]
[R1H]
3 3
rium concentrations (mol/L) of cyclohexanone and
hemiperketal, respectively. The equilibrium concenꢀ
trations of tertꢀbutyl hydroperoxide and 1ꢀtertꢀbutylpꢀ
eroxycyclohexanol are related to the equilibrium conꢀ
stant (K1) of step (IV) by the equation
emp
2.2
R2H + (CH3)3COO•–(CH ) COOHR•2
k
+O
2
R2OO•
,
(VI)
3 3
(CH ) COOH
3 3
R1OO•
R2OO•
(CH3)3COO•
,
(VII)
–R OOH
1
(CH ) COOH
3 3
(2)
K1 = k1 k−1 =[R2H] [R1H][(CH3)3COOH].
(CH3)3COO•
,
(VIII)
–R OOH
2
From the material balance equations for ketone and
hydroperoxide (Eqs. (3) and (4)),
k
t
2
(CH3)3COO•
(CH3)3COOC(CH3)3 + O2, (IX)
(3)
(4)
[R1H]0 = [R1H] + [R2H],
1
k
(CH3)3COO•
2
(CH3)3CO• + O2.
(X)
2
[(CH3)3COOH]0 = [(CH3)3COOH] + [R2H],
The following assumptions were made when derivꢀ
where [R1H]0, [(CH3)3COOH]0, and
[(CH3)3COOH]
ing the kinetic equations: (1) the current concentraꢀ
tions of cyclohexanone, hemiperketal, and hydroperꢀ
oxide remain almost unchanged, since the ketone
conversion does not exceed 3%, as found by the study
of the gas uptake kinetics; (2) at 333 K, according to
kinetic data [18], the equilibrium in reaction (IV) is
established within less than 5 min, so the current conꢀ
centrations of ketone, tertꢀbutyl hydroperoxide, and
hemiperketal are equal to the equilibrium concentraꢀ
tions of these compounds.
are the initial concentration of cyclohexanone, initial
concentration of tertꢀbutyl hydroperoxide, and equiꢀ
librium concentration of tertꢀbutyl hydroperoxide,
respectively, and Eq. (1), one can obtain the reaction
rate equation
emp
2.1
emp
2.2
wrc = ((k − k )[ R 1H ]
(5)
emp
2.2
+ k [R1H]0)[(CH3)3COOi ].
Using Eqs. (2)–(4), we obtain
Therefore, according to reactions (V) and (VI), the
overall rate of the radical chain oxidation of the equiꢀ
librium mixture (wrc) is
(6)
(7)
[R1H]0 = [R1H] + K1[R1H][(CH3)3COOH],
[(CH3)3COOH]0 = [(CH3)3COOH] +
+ K1[R1H][(CH3)3COOH].
wrc = k [R1H][(CH3)3COOi ] +
emp
2.1
The solution of the system of Eqs. (6) and (7) for
(1)
emp
+ k [R2H][(CH3)3COOi ],
gives the expression
[R1H]
2.2
−(1 + K1([(CH3)3COOH]0 −[R1H]0))
[R1H] =
(1 + K1([(CH3)3COOH]0 − [R1H]0)2 + 4K1[R1H]0
2K1
(8)
.
Only the following equation has a physical meaning for К1 = 0.22 (333 K) L/mol [18]:
−(1 + K1([(CH3)3COOH]0 −[R1H]0)) + (1 + K1([(CH3)3COOH]0 − [R1H]0)2 + 4K1[R1H]0
[R1H] =
(9)
.
2K1
The equilibrium concentrations of tertꢀbutyl hydroperoxide and hemiperketal from Eqs. (2) and (7) are
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

142
PUCHKOV et al.
Table 1. Equilibrium composition of the reaction mixture (cyꢀ
clohexanone + tertꢀbutyl hydroperoxide + 1ꢀtertꢀbutylperoxyꢀ
cyclohexanol) calculated for different initial hydroperoxide
concentrations (333 K, [R1H]0 = 3.2 mol/L)
wrc will be related to the gas uptake rate ( ) by the
equation
w
O2
wO = wрц − 0.5wi,АИБН
AIBN(k' kt −1) −
2
rc
(15)
− 0.5(wi,ТБГП + wi,АИБН)(k' kt + 1) − 0.5wi,АИБН e,
TBHP
AIBN
AIBN
[(CH3)3COOH]0
[(CH3)3COOH]
[R1H]
[R2H]
where
tertꢀbutylperoxyl radicals without chain termination;
at 323 K, '/kt = 8.1 [21].
k' is the recombination rate constant for the
k
0.5
1.0
1.5
2.0
2.5
3.00
2.82
2.65
2.50
2.35
0.30
0.62
0.95
1.29
1.65
0.20
0.38
0.55
0.70
0.85
The conditions for correct application of Eqs. (1),
(14), and (15) are the completeness of exchange of the
peroxyl radicals of the oxidized substance, initiator,
and tertꢀbutyloxyl radicals for tertꢀbutylperoxyl radiꢀ
cals in reactions (III) and the stability of the peroxide
products of the oxidized compound (absence of
degenerate branching). For the oxidation of cyclohexꢀ
anol [17], cyclohexyl acetate [12], methyl cyclohexꢀ
anoate [22], aromatic ketones, and esters [23], the
completeness of exchange of the “foreign” peroxyl
and oxyl radicals for the tertꢀbutylperoxyl radicals is
attained at a tertꢀbutyl hydroperoxide concentration of
0.35–0.4 mol/L and is almost independent of the
nature of the compound being oxidized. For this reaꢀ
son, TBHP concentrations of [TBHP]0 = 0.5 mol/L
or higher were used in experiments on cyclohexanone
oxidation.
The kinetics of accumulation of the sum of peroxꢀ
ide compounds in the AIBNꢀinitiated oxidation of
cyclohexanone (333 K, [R1H]0 = 3.2 mol/L, chloꢀ
robenzene, [AIBN] = 0.01 mol/L) (Fig. 1) was studied
to estimate the stability of the peroxide compounds
formed from oxidized cyclohexanone. Figure 1 shows
that the concentration of the peroxide compounds
depends linearly on the oxidation time and is only
insignificantly lower than the concentration of
absorbed oxygen. Therefore, the decomposition of the
peroxide compounds resulting from cyclohexanone
oxidation at 333 K can be neglected.
[(CH3)3COOH]0
1 + K1[R1H]
(10)
(11)
[(CH3)3COOH] =
,
[R2H] = K1[R1H][(CH3)3COOH].
The equilibrium composition of the reaction mixꢀ
ture (cyclohexanone + tertꢀbutyl hydroperoxide +
1ꢀtertꢀbutylperoxycyclohexanol) at [R1H]0 = 3.2 mol/L
and [(CH3)3COOH]0 = 0.5–2.5 mol/L (Table 1) can
be calculated via Eqs. (9)–(11) using the above value
of K1. It can be seen from Table 1 that the hemiperketal
content of the equilibrium mixture is significant, being
about 30% for an initial hydroperoxide concentration
of 2.5 mol/L.
When peroxyl and oxyl radicals of the substrates
and initiators are completely exchanged for tert
ꢀ
butylperoxyl radicals, chain termination proceeds via
reaction (IX) and its rate is determined by the equation
wt = 2kt[(CH3)3COOi ]2,
(12)
An analysis of Eq. (14) shows that the effective
emp
emp
(k2e.m1p −k2.2 )[ R1H] k2.2 [R1H]0,
where kt is the rate constant of recombination of tert
butylperoxyl radicals (reaction (IX)) causing chain termiꢀ
nation; at 333 K,
[19].
Under the conditions of mixed initiation by tert
butyl hydroperoxide and AIBN, the total initiation
rate is determined by the equation
ꢀ
parameter
+
that characꢀ
terizes the oxidizability of the reaction mixture can be
determined from oxidation experiments in which the
2k
t = (2.90 0.05)
×
103 L mol –1 s–1
initiation rate is varied by means of AIBN (
w
i,AIBN).
(k2e.m1p −k2.2 )[ R 1H]
emp
ꢀ
According to Eqs. (9) and (14), the
+
emp
parameter depends on both the cyclohexꢀ
)[ R1H]0
k
2.2
anone and tertꢀbutyl hydroperoxide concentrations.
emp
emp
2.2
(k2e.m1p −k2.2 )[ R1H]
k
(13)
wi = wi,ТБГП + wi,АИБН
,
AIBN
TBHP
The values of
+
at differꢀ
)[ R 1H]0
where wi,АИБН and wi, TBHP are the reaction rates of iniꢀ
tiation due to the decomposition of AIBN (reaction (I))
and TBHP (reaction (II)), respectively (mol L–1 s–1).
ent initial tertꢀbutyl hydroperoxide concentrations
were obtained (Table 2) to determine the constants
emp and emp. It can be seen from Table 2 that the oxiꢀ
k2.1
k2.2
dizability (а) of the reaction mixture decreases with an
For the quasiꢀsteady state (wi
=
wt), Eq. (1) appears
increasing tertꢀbutyl hydroperoxide concentration.
Therefore, the hemiperketal molecules are more diffiꢀ
cult to oxidize than the cyclohexanone molecules.
as
wrc = ((k −k )[R1H] + k [R1H]02kt−0.5wi.
If corrections are applied to the nonchain oxygen
uptake (reaction (I)), oxygen evolution (reactions (IX)
and (X)), and nitrogen evolution in reaction (I) [20],
emp
2.1
emp
2.2
emp
2.2
(14)
emp
2.1
emp
2.2
Introducing the designation
+
(k −k )[ R1H]
emp
2.2
= a and using the data from Tables 1 and 2,
k
)[ R1H]0
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

REACTIVITY OF C–H BONDS IN CYCLOHEXANONE
143
–1
С, mol/L
a
, s
1
2
0.03
0.6
0.4
0.2
0.02
0.01
2.4
2.6
2.8
3.0
[R H], mol/L
1
200
400
600
Time, min
Fig. 2. Oxidizability of the reaction mixture (
a
) versus the
Fig. 1. Kinetics of the (
lation of the peroxide compounds (total content) for the
AIBNꢀinitiated oxidation of cyclohexanone (333 K,
1
) oxygen uptake and (
2
) accumuꢀ
equilibrium concentration of cyclohexanone (
) for
[R1H]
the cooxidation of cyclohexanone and 1ꢀtertꢀbutylperoxꢀ
ycyclohexanol (333 K, [R H] = 3.2 mol/L, chlorobenꢀ
[R H]
0.01 mol/L).
= 3.2 mol/L, chlorobenzene, [AIBN] =
1
0
1
0
zene, [AIBN] = 0.001–0.01 mol/L).
we established a linear relationship between
a
and
O
O
O
(Fig. 2). The ordinate intercept corresponds to
[R1H],
OOH
OH
•
+RO , O , RH
2
[H]
2
emp
, and the slope of the straight line is
−k2.2 )[ R1H]0
emp
(k2e.m1 p − k2.2
⎯
. The processing of this relationship
)
1
emp
(XI)
O
O
gave the following values of rate constants:
=
k
2.1
emp
(21.9 2.5)
×
10–2 L mol–1 s–1 and
= (10 1.3)
×
k2.2
[H]
10–3 L mol–1 s–1.
.
2
As was mentioned above, the composition of the
oxidation products should be known to estimate the
partial rate constants for the reactions of C–H bonds
of different types in cyclohexanone and 1ꢀtertꢀbutylpꢀ
OOH
OH
eroxycyclohexanol molecules. If the possibility of tert
ꢀ
The reduction of compounds
1 and 2 with tripheꢀ
butyl radical oxidation in hemiperketal is neglected (as
is assumed in the Howard–Ingold method), the cycloꢀ
hexanone and 1ꢀtertꢀbutylperoxycyclohexanol moleꢀ
cules will have ten reactive C–H bonds each. Cycloꢀ
nylphosphine allows them to be converted into stable
products (2ꢀ and 4ꢀhydroxycyclohexanones) and to be
analyzed by GLC (scheme XI).
It is likely that no monohydroperoxide results from
hexanone contains four
bonds, and two ꢀС–Н bonds. 1ꢀtertꢀButylperoxycyꢀ
clohexanol contains four ꢀС–Н bonds, four ꢀС–Н
αꢀС–Н bonds, four
β
ꢀС–Н
the oxidation of cyclohexanone at the
It was shown that the oxidation of the carbonylꢀconꢀ
taining compounds (including ketones) at their ꢀC–
H bonds resulted in the transformation of the intermeꢀ
diate peroxyl radical into ꢀunsaturated compounds
βꢀC–H bonds.
γ
β
γ
bonds, and two
δꢀС–Н bonds:
β
O
1
OOC(CH3)3
HO
α,β
1
[22, 24], their oxides [22, 25], and dihydroperoxide
derivatives [22]. In the case of cyclohexanone, these
6
5
2
3
6
2
3
β
β
α
α
γ
γ
β
β
δ
γ
5
.
transformations of radical
enone and 2,3ꢀepoxycyclohexanone (scheme XII,
and , respectively) and 2,3ꢀdihydroperoxycyclohexꢀ
anone, whose reduction yields 2,3ꢀdihydroxycycloꢀ
hexanone (scheme XII,
3 should yield 2ꢀcyclohexꢀ
4
4
a
b
The oxidation of cyclohexanone at the
ꢀС–Н bonds mainly yields hydroperoxides (
stable at the experimental temperature (Fig. 1):
α
ꢀ
and
γ
1
and 2)
c)
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

144
PUCHKOV et al.
Table 2. Effect of the TBHP concentration on the rate of the AIBNꢀinitiated oxidation of cyclohexanone (333 K, [R1H]0 =
3.2 mol/L, chlorobenzene)
–0.5
107,
wiAIBN
×
107,
wrc
×
106,
a
(2kt
)
×
103,
×
106,
[TBHP]0, wiTBHP
mol/L
×
[AIBN],
mol/L
wO
Chain
length
, s–1
2
a
mol L–1 s–1
mol L–1 s–1
mol L–1 s–1
mol0.5 L0.5 0.5
s
mol L–1 s–1
0.5
1.0
1.5
2.0
2.5
0.5 0.2
0.001
0.003
0.006
0.008
0.01
0.19
0.58
1.16
1.54
1.93
3.3 0.5
3.6 0.4
3.7 0.2
3.8 0.1
3.8 0.3
3.8 0.5
4.6 0.4
5.4 0.4
5.9 0.5
6.4 0.5
55.4
42.5
32.4
29.0
26.3
13.4 0.3
0.72 0.06
0.70 0.08
0.61 0.08
0.59 0.09
0.59 0.07
2.5 0.1
8.0 0.3
9.5 0.5
12.5 0.4
0.001
0.003
0.006
0.008
0.01
0.19
0.58
1.16
1.54
1.93
4.4 0.5
4.5 0.4
4.5 0.2
4.4 0.2
4.4 0.4
6.3 0.6
6.9 0.4
7.7 0.3
8.1 0.3
8.7 0.5
23.6
22.6
21.1
20.1
19.5
13.0 0.3
0.001
0.003
0.006
0.008
0.01
0.19
0.58
1.16
1.54
1.93
5.2 0.2
9.8 0.4
12.0
11.9
11.7
11.6
11.5
11.3 0.2
5.2 0.1 10.2 0.4
5.1 0.3 10.7 0.5
5.0 0.4 11.1 0.4
4.9 0.3 11.4 0.5
0.001
0.003
0.006
0.008
0.01
0.19
0.58
1.16
1.54
1.93
5.8 0.3 10.7 0.5
5.7 0.2 11.0 0.4
5.5 0.1 11.4 0.3
5.4 0.2 11.7 0.4
5.3 0.8 11.9 0.9
11.1
10.9
10.7
10.6
10.4
10.9 0.2
0.001
0.003
0.006
0.008
0.01
0.19
0.58
1.16
1.54
1.93
6.3 0.1 12.7 0.3
6.1 0.2 12.9 0.5
5.9 0.4 13.3 0.6
5.7 0.3 13.5 0.5
5.6 0.2 13.7 0.3
10.0
9.9
9.510.9 0.2
9.7
9.6
10.0
O
O
O
O
•
a
–HO
b
•
O
•
–HO
2
•
OO
OOH
3
(XII)
O
O
OOH
OOH
OH
c
[H]
.
OH
The 1ꢀtertꢀbutylperoxycyclohexanol molecule also
contains three types of C–H bonds. Hydroperoxides ical with its
4
–
6
form via the reaction of the tertꢀbutylperoxyl radꢀ
ꢀ, , and ꢀС–Н bonds.
β
γꢀ
δ
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

REACTIVITY OF C–H BONDS IN CYCLOHEXANONE
145
O
OOC(CH3)3
OOC(CH3)3
OOH
HO
HO
OH
•
+R , O , RH
2
[H]
2
4
O
O
OOC(CH3)3
HO
HO
[H]
.
–H O
2
(XIII)
OOH
5
OOC(CH3)3
OH
O
[H]
OOH
OH
6
The reduction of compounds
4–6 with tripheꢀ to the additivity principle and the number of C–H
nylphosphine yields 2ꢀ, 3ꢀ, and 4ꢀhydroxycyclohexꢀ bonds in the equivalent reactive positions of the subꢀ
anones. As the samples are treated, 3ꢀhydroxycycloꢀ
hexanone readily loses water and turns into 2ꢀcycloꢀ
hexenone under the determination conditions [26].
strates, the empirical rate constants for chain propagaꢀ
emp
tion,
(k2.1
and emp, are related to the partial rate conꢀ
k2.2
stants of hydrogen atom abstraction by the tertꢀbutylpꢀ
eroxyl radical from different positions of
cyclohexanone and 1ꢀtertꢀbutylperoxycyclohexanol
through the equations
The above listed compounds and 1,2ꢀcyclohexꢀ
anedione were indeed observed (Fig. 3) in the products
of the AIBNꢀinitiated oxidation of cyclohexanone in
the presence of tertꢀbutyl hydroperoxide (333 K,
[R1Н]0 = 3.2 mol/L, chlorobenzene, [TBHP]0 =
0.5 mol/L, [AIBN] = 0.01 mol/L). 1,2ꢀCyclohexꢀ
anedione was observed in small amounts and formed,
most probably, by 2ꢀhydroperoxycyclohexanone
decomposition initiated by the tertꢀbutylperoxyl radiꢀ
cal [4]:
emp
2.1
H;2,6
2.1
H;3,5
2.1
H;4
(1)
(2)
(k
= 4k
+ 4k
+ 4k
+ 2k
+ 2k
,
2.1
emp
2.2
H;2,6
2.2
H;3,5
2.2
H;4
.
2.2
k
)= 4k
Taking into account Eqs. (16) and (17) and introꢀ
ducing the corresponding parameters a2, 6 a3, 5, and a4
, ,
we obtain
O
OOH
+
(CH3)3COO•
2
С
×
10 , mol/L
С × 10, mol/L
(XIV)
1
6
0.8
0.6
0.4
0.2
1.6
1.2
0.8
0.4
O
O
OOH
O
–(CH ) COOH
3 3
•
3
–HO•
.
Figure 3 shows that the concentrations of all prodꢀ
ucts depend linearly on the reaction time under the
experimental conditions. This indicates their formaꢀ
tion via parallel reactions. Therefore, the composition
of the products is nearly independent of the cyclohexꢀ
anone conversion within the conversion range examꢀ
ined.
4
5
2
2
4
6
8
Time, h
It can be seen from the above structural formulas of
cyclohexanone and hemiperketal and from reactions
Fig. 3. Accumulation kinetics of the products of the
AIBNꢀinitiated oxidation of cyclohexanone (333 K,
(XIII) and (XIV) that the
the ꢀposition in the hemiperketal are identical to the
2ꢀ and 6ꢀpositions; the ꢀposition in the ketone and
the ꢀposition in the hemiperketal, to the 3ꢀ and
5ꢀpositions; the ꢀposition in the ketone and the ꢀposiꢀ
tion in the hemiperketal, to the 4ꢀposition. According
α
ꢀposition in the ketone and
[R H]
=
3.2 mol/L, chlorobenzene, [AIBN]
0.01 mol/L, [TBHP] = 0.5 mol/L): ( ) (2ꢀhydroperoxyꢀ
cyclohexanone + 2ꢀhydroxycyclohexanone), ( ) 1,2ꢀcycloꢀ
hexanedione, ( ) 2ꢀcyclohexenone, ( ) 2,3ꢀepoxycycloꢀ
hexanone, ( ) (2,3ꢀdihydroperoxycyclohexanone +
2,3ꢀdihydroxycyclohexanone), and ( ) (4ꢀhydroꢀperoxyꢀ
cyclohexanone + 4ꢀ hydroxycyclohexanone).
=
1
0
1
β
0
2
β
3
4
γ
5
γ
δ
6
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

146
PUCHKOV et al.
H;4
H;3,5
k2H.1;2,6, k2H.1;3,5, k2.1 , k2H.2;2,6, k2.2
Table 3. Effect of the TBHP concentration on the fraction
of the products of cyclohexanone oxidation in the ꢀ, ꢀ,
and
ꢀpositions (333 K, [R1H]0 = 3.2 mol/L, chlorobenꢀ and H;4 were determined from the dependences of the
The rate constants
,
α
β
γ
k2.2
zene, [AIBN] = 0.01 mol/L)
total fractions of the products of the AIBNꢀinitiated
cooxidation of cyclohexanone and 1ꢀtertꢀbutylperoxꢀ
ycyclohexanol on the tertꢀbutyl hydroperoxide conꢀ
centration (333 K, [R1Н]0 = 3.2 mol/L, chlorobenꢀ
zene, [AIBN] = 0.01 mol/L, [TBHP]0 varied between
0.5 and 2.5 mol/L). The experimental results are given
Fraction of products, mol %
[TBHP]0, Converꢀ
position
mol/L
sion, %
α
β
γ
in Table 3. Using the values obtained for a at different
0.5
1.0
1.5
2.0
2.5
4.4 0.4 91.9 0.3 4.4 0.7 3.8 0.4
5.0 0.3 91.2 0.2 4.6 0.6 4.2 0.5
4.1 0.2 91.1 0.1 4.5 0.5 4.4 0.3
4.6 0.4 90.8 0.3 4.4 0.4 4.8 0.4
4.4 0.2 90.2 0.2 4.7 0.7 5.1 0.5
TBHP concentrations (Table 2) and the fractions of
cyclohexanone oxidation products under the
same conditions (Table 3), we plotted the depenꢀ
dences of a2, 6 a3, 5, and a4 on the equilibrium cycloꢀ
,
hexanone concentration (Fig. 4). According to
Eqs. (19)–(21), the dependences of a2, 6 a3, 5, and a2, 6
,
,
a4 on the equilibrium cyclohexanone concentration
are indeed linear. The values of the parameters a2, 6 and
a3, 5 increase with an increasing equilibrium ketone
concentration, whereas a4 remains unchanged. Thereꢀ
fore, the reactivities of the C–H bonds of the hemiꢀ
perketal in the 2ꢀ and 6ꢀpositions and in the 3ꢀ and
5ꢀpositions are lower than those of the same bonds in
the corresponding positions of ketone, whereas the
reactivities of the C–H bonds in the 4ꢀposition are
similar for these compounds.
H;2,6
2.1
H;3,5
2.1
H;4
2.1
a2,6 + a3,5 + a4 = ((4k
+ 4k
+ 2k ) −
H;2,6
2.2
H;3,5
H;4
− (4k
+ 4k
+ 2k ))[R1H] +
(3)
2.2
2.2
H;2,6
2.2
H;3,5
2.2
H;4
2.2
+ (4k
+ 4k
+ 2k )[R1H]0.
It follows from Eqs. (16)–(18) that
H;2,6
2.1
H;2,6
H;2,6
2.2
a2,6 = (4k
− 4k
)[R1H] + 4k
)[R1H] + 4k
[R1H]0,
(4)
(5)
(6)
2.2
H;3,5
2.1
H;3,5
H;3,5
2.2
a3,5 = (4k
− 4k
[R1H]0,
2.2
Based on Eq. (21), from the ordinate intercept of
H;4
H;4
H;4
2.2
a4 = (2k − 2k )[R1H] + 2k [R1H]0.
k2H.2;4[R1H]0
2.1
2.2
line
3
(Fig. 4), the value of
was determined
H;4
H;4
and the value of
1.3)
=
was estimated at (4.0
k2.1 ,
k2.2
10–3 L mol–1 s–1). Using Eq. (20) and the data
a3, 5 a4, s–1
,
a2, 6, s–1
×
presented in Fig. 4, we determined H;3,5 = (2.75 0.5)
×
k2.1
1
10–3 L mol–1 s–1 and H;3,5 = (0.L mol–1 s–1 from line
2
k2.2
0.6
0.06
0.04
0.02
H;2,6
k2.1
(Fig. 4). The
= (50 5)
10–3 L mol–1 s–1 values (Table 4)
were calculated from Eqs. (16) and (17) using the valꢀ
×
10–3 L mol–1 s–1 and
H;2,6 = (0.23 0.02)
×
k2.2
H;4
H;4
k2.2
ues of
H;3,5, and
k2H.1;3,5, k2.1 , k2.2
.
0.4
0.2
H;4
k2.1
H;4
of
k2.2
2
3
The values of the partial constants
and
the C–H bonds of cyclohexanone and 1ꢀtertꢀbutylperꢀ
oxycyclohexanol in the 4ꢀposition (Table 4) coincide
(within the experimental error) with kHCH for the reacꢀ
tion of the methylenic C–H bond in cyclohexane with
the tertꢀbutylperoxyl radical: kH (333 K) = 3.8
×
CH
10–3 L mol–1 s–1 [27]. Therefore, the functional group
of the ketone does not noticeably affect the ꢀС–Н
γ
bonds, as was also observed for methyl hexanoate oxiꢀ
dation [22]. At the same time, the reactivity of the
C⎯H bonds of hemiperketal in the 3ꢀ and 5ꢀpositions
2.4
2.6
2.8
3.0
[R1H], mol/L
is lower than that of the methylenic C–H bonds in
cyclohexane, as was observed for the oxidation of
cyclohexanol [17] and cyclohexyl acetate [12]. The
data of this study show that the deactivation effect of
the carbonyl and carboxyalkyl groups is extended over
a shorter distance than that of the hydroxyl and alkoxyl
groups.
Fig. 4. Oxidizability of the reaction mixture at positions
2,6, 3,5, and 4 vs. equilibrium concentration of cyclohexꢀ
anone (
) for the cooxidation of cyclohexanone and
1ꢀtertꢀbutylperoxycyclohexanol (333 K, [R H]
[R1H]
=
1
0
3.2 mol/L, chlorobenzene, [AIBN] = 0.001–0.01 mol/L):
2,6;
3,5
4
(1)
a
(2) a ; and (3) a .
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

REACTIVITY OF C–H BONDS IN CYCLOHEXANONE
147
Table 4. Relative reactivities and partial rate constants for the reactions of the tertꢀbutylperoxyl radical with the C–H bonds
of the oxygenꢀcontaining compounds (333 K, [TBHP]0 = 0.5–2.5 mol/L)
H
k2.i
×
103 (per one C–H bond),
Substance
Position
Relative reactivity
–1 –1
L mol
s
Cyclohexanone
α
β
γ
50
5
12.5
0.7
2.8 0.5
4.0 0.8
1
1ꢀtertꢀButylperoxycyclohexanol
β
γ
0.23 0.04
0.33 0.05
4.0 0.8
0.058
0.083
1
δ
α
β
γ
Cyclohexanol [17]
420.0 2.9
0.26 0.04
0.33 0.05
4.0 0.4
110.5
0.068
0.087
1
δ
α
β
γ
Cyclohexyl acetate [12]
Methyl hexanoate* [22]
18.2 1.5
0.31 0.05
0.52 0.07
3.7 0.4
5.0
0.087
0.21
1
δ
α
β
γ
83 16
0.59
0.81
0.98
1
114 13
138 23
δ
ε
140 27
12
2
0.085
0.425
0.63
0.085
Butanoic acid ** [28]
* 373 K.
α
β
γ
–
–
–
** Reaction with its own radicals at 373 K; relative to the methylenic δꢀC–H bond of methyl hexanoate.
Table 4 shows that the partial constant for the peroxyl radicals with the C–H bonds of the substrate
ꢀС–Н bonds of cyclohexanone (2ꢀ and 6ꢀpositions) [4, 12, 17, 22]. On the one hand, the polar effect
α
(inductive effect + field effect) of the electronꢀaccepꢀ
tor functional group decreases the electron density on
the hydrogen atom of the attacked C–H bond, thus
decreasing the probability of its interaction with the
electrophilic peroxyl radicals [4]. This deactivation
effect of the electronꢀacceptor functional groups natꢀ
is nearly 13 times larger than that for the C–H bond in
the 4ꢀposition. This value is in fairly good agreement
with the aboveꢀpresented values for the reactions of
ketones with their radicals. For the methylenic C–H
bonds in 1ꢀtertꢀbutylperoxycyclohexanol in the 2ꢀ and
H;2,6
k2.2
6ꢀpositions,
is almost 17.2 times smaller than
urally decreases on passing from the
ꢀposition and, particularly strongly, to the
α
ꢀ
position to the
position.
that for its C–H bonds in the 4ꢀposition. The reactivꢀ
ities of the ꢀС–Н bonds of both cyclohexanone (3ꢀ
β
γ
β
On the other hand, due to the conjugation effect, the
carbonyl, hydroxyl, alkoxyl, and peroxyl groups can
stabilize the transition state, which is similar to carꢀ
and 5ꢀpositions) and 1ꢀtertꢀbutylperoxycyclohexanol
(2ꢀ and 6ꢀpositions) are, respectively, 1.44 and
12 times lower than the reactivity of the methylenic
C–H bonds of hydrocarbons (Table 4).
bonꢀcentered radical
7
[4]:
δ
δ
–O
–O
A similar influence of electronꢀacceptor functional
δ
+
δ
+
groups on the reactivity of the C–H bonds in the
and ꢀpositions was observed for the oxidation of
α
ꢀ
•
(XV)
β
.
cyclohexanol [17], cyclohexyl acetate [12], butanoic
acid [28], and methyl hexanoate [22] (Table 4), which
can be explained in terms of the dual influence of these
7
It is the net effect of the two opposite electronic
ꢀС–Н
functional groups on the reaction of the electrophilic effects that determines the reactivity of the
α
KINETICS AND CATALYSIS Vol. 54
No. 2
2013

148
PUCHKOV et al.
13. Perkel’, A.L., Freidin, B.G., and Voronina, S.G., Zh.
bonds in the ketone (the hemiperketal has no αꢀС–Н
bonds). The conjugation effect cannot exert a stabilizꢀ
ing influence on the or ꢀcarbonꢀcentered radical.
Therefore, only the polar effect affects the C–H bonds
in these positions, and they are deactivated. It can be
seen from Table 4 that the simultaneous effect of the
hydroxyl and tertꢀbutylperoxyl groups leads to a stronꢀ
Anal. Khim., 1993, vol. 48, no. 10, p. 1697.
14. Advances in Organic Chemistry: Methods and Results
,
β
ꢀ
γ
Raphael, R.A., Taylor, E.C., and Böhme, H., Eds.,
New York: Interscience, 1960, vol. 1.
15. Perkel’, A.L., Voronina, S.G., and Perkel’, R.L., Zh.
Anal. Khim., 1991, vol. 46, no. 11, p. 2283.
16. Schwalm, W.J. and Ingold, K.U., Int. Oxidation Symp.
San Francisco, 1967, vol. 1, p. 3.
,
ger deactivation of the
hemiperketal than in cyclohexanol.
βꢀ
and ꢀС–Н bonds in the
γ
17. Puchkov, S.V., Buneeva, E.I., and Perkel’, A.L., Kinet.
Catal., 2002, vol. 43, no. 6, p. 756.
18. Terent’ev, V.A. and Antonovskii, V.L., Uspekhi khimii
organicheskikh perekisnykh soedinenii i avtookisleniya
(Advances in the Chemistry of Organic Peroxides and
Autooxidation), Moscow: Khimiya, 1969.
REFERENCES
1. Berezin, I.V., Denisov, E.T., and Emanuel’, N.M.,
Okislenie tsiklogeksana (Cyclohexane Oxidation),
Mosk. Gos. Univ., 1962.
2. Proizvodstvo kaprolaktama (Caprolactam Production), 19. Landolt–Börnstein Lahlwerte u. Funktionen aus Naturꢀ
Ovchinnikov, I.V. and Ruchinskii, V.R., Eds., Moscow:
Khimiya, 1977.
wissenschaften u. Technik, Neue Serie, Group II, vol. 13,
Berlin: Springer, 1984, p. 431.
3. Denisov, E.T., Mitskevich, N.I., and Agabekov, B.E.,
Mekhanizm zhidkofaznogo okisleniya kislorodꢀ
soderzhashchikh soedinenii (Mechanism of the Liquidꢀ
Phase Oxidation of OxygenꢀContaining Compounds),
Minsk: Nauka i Tekhnika, 1975.
4. Perkel’, A.L., Voronina, S.G., and Freidin, B.G., Russ.
Chem. Rev., 1994, vol. 63, no. 9, p. 75.
20. Denisov, E.T. and Kovalev, G.I., Okislenie i stabiliꢀ
zatsiya reaktivnykh topliv (Oxidation and Stabilization
of Jet Fuels), Moscow: Khimiya, 1983.
21. Mill, T. and Hendry, D.G., LiquidꢀPhase Oxidation
,
Comprehensive Chemical Kinetics, vol. 16, Amsterꢀ
dam: Elsevier, 1980, p. 1.
22. Nepomnyashchikh, Yu.V., Puchkov, S.V., Perkel’, A.L.,
and Arnatskaya, O.I., Kinet. Catal., 2012, vol. 53, no. 2,
p. 155.
5. Hendry, D.G., Gould, C.W., Schuetzle, D., Syz, M.G.,
and Mayo, F.R., Org. Chem., 1976, vol. 41, no. 1, p. 1.
6. Kucher, R.V., Timokhin, V.I., and Kravchuk, N.A.,
23. Buneeva, E.I., Voronina, S.G., Perkel’, A.L., and Freiꢀ
Dokl. Akad. Nauk SSSR, 1987, vol. 294, no. 6, p. 1411.
din, B.G., Kinet. Catal., 1990, vol. 31, no. 6, p. 1163.
7. Hermans, I., Jacobs, P.A., and Peeters, J., J. Mol. Catal.
A: Chem., 2006, vol. 251, nos. 1–2, p. 221.
24. Nepomnyashchikh, Yu.V., Nosacheva, I.M., and Perꢀ
kel’, A.L., Kinet. Catal., 2004, vol. 45, no. 6, p. 768.
8. Hermans, I., Nguyen, T.L., Jacobs, P.A., and
Peeters, J., Chem. Phys. Chem., 2005, vol. 6, p. 637.
9. Karnojitzki, V., Les peroxydes organiques, Paris: Herꢀ
25. Perkel’, A.L., Borkina, G.G., Voronina, S.G.,
Kotel’nikova, T.S., Nepomnyashchikh, Yu.V., Puchꢀ
kov, S.V., and Revkov, O.A., Polzunov. Vestn., 2008,
no. 3, p. 113.
mann, 1958.
26. Puchkov, S.V., Buneeva, E.I., and Perkel’, A.L., Kinet.
Catal., 2005, vol. 46, no. 3, p. 340.
10. Gumulka, W., Korosinski, J., and Hancyk, B., Proc.
Inst. Przem. Org., 1972, vol. 4, p. 21.
27. Chenier, J.H.B., Tong, S.B., and Howard, J.A., Can. J.
Chem., 1978, vol. 56, no. 24, p. 3047.
11. De Borger, L., Anteunis, M., Lammens, H., and Verꢀ
rele, M., Bull. Soc. Chim. Belg., 1964, vol. 73, no. 1,
p. 73.
12. Puchkov, S.V., Buneeva, E.I., and Perkel’, A.L., Kinet.
Catal., 2001, vol. 42, no. 6, p. 751.
28. Nepomnyashchikh,
Puchkov, S.V., and Perkel’, A.L., Vestn. KuzGTU, 2011,
no. 2, p. 81.
Yu.V.,
Borkina,
G.G.,
KINETICS AND CATALYSIS Vol. 54
No. 2
2013