Mild oxidation of styrene epoxide in the presence of trace perchloric acid

Article abstract of DOI:10.1134/S0965544111060144

Perchloric acid in a tert-butanol solution with 10 vol % chlorobenzene exhibits an almost three orders of magnitude higher activity in comparison with para-toluenesulfonic acid (TSA) as a catalyst for the parallel styrene epoxide (SE) heterolysis and homo

Full text of DOI:10.1134/S0965544111060144

ISSN 0965ꢀ5441, Petroleum Chemistry, 2011, Vol. 51, No. 6, pp. 458–464. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © L.V. Petrov, V.M. Solyanikov, 2011, published in Neftekhimiya, 2011, Vol. 51, No. 6, pp. 467–472.
Mild Oxidation of Styrene Epoxide in the Presence of Trace
Perchloric Acid
L. V. Petrov and V. M. Solyanikov
Institute of Problems of Chemical Physics, Russian Academy of Sciences,
pr. Akademika Semenova 1, Chernogolovka, Moscow oblast, 142432 Russia
eꢀmail: plv@icp.ac.ru
Received September 21, 2010
Abstract—Perchloric acid in a tertꢀbutanol solution with 10 vol % chlorobenzene exhibits an almost three
orders of magnitude higher activity in comparison with paraꢀtoluenesulfonic acid (TSA) as a catalyst for the
parallel styrene epoxide (SE) heterolysis and homolysis processes. There is a substantial difference between
the reactions mediated by these catalysts: rate curves for the overall consumption in the presence of perchloric
acid (SE heterolysis) yield straight lines in the log [SE]–time coordinates, but the firstꢀorder rate constant
drops with an increase in [SE]. However, the oxygen uptake rate increases with [SE]. In the case of TSA, neiꢀ
ther overall consumption nor oxidation rate depended on [SE] at [SE] > [TSA]; i.e., the reaction was zeroꢀ
order in SE.
DOI: 10.1134/S0965544111060144
This work is in continuation of the investigation chromatography. The solvents were distilled tert
ꢀ
into the mechanism of conversion of phenylꢀsubstiꢀ butanol and chlorobenzene. The catalyst, aqueous
tuted ethylene oxides in the presence of acids. Earlier, perchloric acid with a concentration of ~13 mol/l, was
it has been shown that styrene epoxide (SE) is decomꢀ prepared from commercial technicalꢀgrade HClO4 by
posed mainly in the heterolytic mode by the action of double distillation in a vacuum. Styrene epoxide purꢀ
paraꢀtoluenesulfonic acid (TSA), but there is a simulꢀ chased from Aldrich had a base substance content of
taneous (parallel to heterolysis) generation of tranꢀ 97% and was used without additional purification.
sient species reactive toward oxygen [1–3]. Their
existence is confirmed by the ability of TSA to catalyze
RESULTS AND DISCUSSION
SE oxidation with oxygen at a low temperature
(343 K). Generally, the radical chain oxidation of SE
proceeds quite slowly even at a 60 K higher temperaꢀ
ture in the presence of a radical initiator [4]. A detailed
kinetic study has shown the parallel occurrence of the
heterolytic and homolytic processes of SE conversion
in the presence of TSA, the rates of both processes are
proportional to [TSA] and do not depend on [SE] at
[SE] > [TSA]; i.e., they are firstꢀorder in acid and
zeroꢀorder in SE. The study of SE transformation in
the presence of HClO4 has also revealed that, like TSA
and sulfuric acid [2, 5], perchloric acid (PCA) mediꢀ
ates both heterolysis and oxygen absorption in a
BUCH (90% tertꢀbutanol and 10% chlorobenzene by
volume) solution; however, the transformation reacꢀ
tion in the presence of HClO4 is kinetically quite difꢀ
ferent from that in the presence of sulfonic acids.
Figure 1 shows the oxygen uptake rate curves for
acidified solutions of SE in the reactor of the manoꢀ
metric unit. There are two distinct differences from the
earlier data obtained in experiments with TSA. First,
the activity of HClO4 in oxidation is approximately
three orders of magnitude higher than that of TSA.
Second, there is no induction period in the oxidation
reaction with HClO4, which was detectable in the
experiments with the SE–TSA binary system at a low
TSA concentration, was clearly distinguished in
experiments with H2SO4, and disappeared when a catꢀ
alytic amount of a copper salt was introduced into the
joint SE and H2SO4 solution [5]. Figure 2 depicts the
dependence of the initial oxidation rate on [HClO4
]
and the initial epoxide concentration [SE]0. We again
observe a noticeable contrast with the earlier data.
Oxygen uptake in the SE + TSA and SE + H2SO4
binary systems was described by a simple kinetic equaꢀ
EXPERIMENTAL
tion for the oxidation rate
[acid] [2]. From Fig. 2, it is seen that the reaction
order with respect to HClO4 is greater than unity;
of Fig. 2 gives
4 0.15. We can suggest the cause of the
v
= k[SE]0 [acid]1 at [SE] >
The experimental procedures used were similar to
those described in [1–3]. The absorption of oxygen
was measured on a manometric unit, and the rates of
n
overall SE consumption and benzaldehyde (BA) calculation by the data points in curve
1
buildup in a bubble reactor were measured using liquid a value of
458
n≈1.

MILD OXIDATION OF STYRENE EPOXIDE
459
12
10
8
4
1
2
5
6
6
3
4
2
0
10
20
30
40
50
60
, min
t
–5
–5
Fig. 1. Rate curves of oxygen uptake in BUCH at 343 K: (
1.05
ning of measurements (0 min) refers to the 30th min.
1
–
3
) [SE] = 0.87 mol/l and [HClO ] = 2.8
×
10 , 2.1
× 10 mol/l, respectively; (4–6) [HClO ] = 2.1 × 10 mol/l and [SE] = 1.74, 0.44, and 0.21 mol/l, respectively; the beginꢀ
4
× 10 , and
0
4
–5
–5
0
overestimated order in acid. Comprehensively analyzꢀ Fig. 2 also differs from the relevant curve for TSA. The
oxidation rate at low values of [SE] increases with an
increase in [SE]0, then the curve becomes less steep,
and the order decreases. A distinct zero order in
epoxide was observed at [SE] > [TSA] in the reactions
of SE and its two halogenated derivatives. We believe
that the partial orders of the oxidation reaction in both
components are unity in the SE + HClO4 binary sysꢀ
ing our data on the SE + HClO4 system and considerꢀ
ing possible causes of the difference of the order in
acid from unity, we assume the following. It should not
be ruled out that the working epoxide solutions had
primordially contained a small amount, on the order
of 10–5 mol/l, of an organic base that neutralizes part
of the acid under the given experimental conditions at
= k [SE]1 [HClO4]1. The aforemenꢀ
[
HClO4] of about 10–5 mol/l and decreases the oxidaꢀ
tem; that is,
v
tioned assumption on the presence of a trace base
impurity explains the overestimated order in acid, the
decrease in the rate order in SE at a high epoxide conꢀ
tion rate: the lower the [HClO4], the more pronounced
the decrease. In the previous study with TSA, the
amount of acid was at least 0.01 mol/l, i.e., two to
three orders of magnitude higher, and the effect of
trace base impurities could quite definitely be unnoꢀ
ticeable. The amount of a strong base can be estimated
as follows. The dashed line in Fig. 2 refers to the first
order in acid, it corresponds to the intercept of
centration, and the fact that the
v
–[SE]0 curve
(curve in Fig. 2) passes through the origin. Forcedly
2
looking ahead, we note that the data on the PCAꢀ
mediated overall consumption of SE are consistent
with the assumption on the presence of an impurity
base in SE. The drop in the SE conversion rate conꢀ
stant with an increase in [SE]0 (Fig. 4) is hard to
explain another way, if at all. Calculation of the effecꢀ
tive bimolecular rate constant of the oxidation of the
SE + PCA system from the slope of the dashed line in
Δ
[HClO4] 5 ×
10–6 mol/l on the abscissa. Assuming
≈
that the amount of the base that in the reactor neutralꢀ
izes this amount of the acid is even five times greater
(i.e., 2.5
epoxide should be 2.5
×
10–5 mol/l), its concentration in the initial
×
10–4 mol/l. Under the assumpꢀ
Fig. 2 (
= 0.84 l/(mol s)) and the slope of the tanꢀ
k
bO2
tion that the molecular weights of SE and the hypoꢀ
thetical impurity base are equal, this is ~0.003 wt %,
gent the initial portion of curve
an amount that is quite likely to be present in SE. 0.71 l/(mol s)) gives an average value of
2
in Fig. 2 (
=
=
k
bO2
k
bO2
Curve
2
of the dependence of
v
on [SE]0 plotted in
0.77 l/(mol s). Since the oxidation rates of the SE +
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011

460
PETROV, SOLYANIKOV
[SE], mol/l
1.0
0
0.5
1.5
2.0
28
24
20
16
12
1
2
8
4
0
1
2
3
4
5
5
[НСlО ] × 10 , mol/l
4
Fig. 2. Dependence of the oxygen uptake rate curves in BUCH at 343 K on (1) [HClO ] at [SE] = 0.87 mol/l and (2) [SE] at
4 0 0
–5
[HClO ] = 2.1 × 10 mol/l.
4
TSA and SE + PCA systems are equal at [PCA] = high degrees of SE conversion suggests the firstꢀorder
rate. The BA buildup kinetics is also firstꢀorder, as we
have found from the data of curve in Fig. 3a. Under
2
.65
0.5 mol/l in both cases,
10⎯6 mol/(l s) ( = 5.7 × 10–4 s–1 [2]),
b[PCA][SE]0 = 10.2
10–6 mol/(l s), it follows that
×
10–5 mol/l and [TSA] = 0.02 mol/l for [SE]0 =
[TSA] = 10.2 ×
1
'
v
= k
TSA
the assumption of firstꢀorder BA buildup during the
conversion of SE at a constant BA yield of
= [BA]/Δ[SE] and its independence from the
k
v
=
PCA
k
×
α
Δ
the efficiency of perchloric acid as an SE oxidation
catalyst is higher than that of TSA by a factor of ~750
(almost three orders of magnitude): at equal rates of
oxidation mediated by PCA and PSA, the
[PCA]/[TSA] ratio is 1/750.
degree of conversion of epoxide, we get d[BA]/dt =
αk
[SE]t
=
αk
[SE]0e–kt to give by integrating [BA]∞ –
[BA]t = [BA]∞e– , log
m
m = const – (kt/2.3), where
= [BA]∞ – [BA]t. The BA buildup data for calculatꢀ
kt
ing
k
are given in the table.
The results of measuring the overall consumption
of SE in the bubble reactor in the presence of PCA are
presented in Figs. 3 and 4. Figure 3a shows the typical
SE consumption and benzaldehyde buildup rate
curves. Benzaldehyde does not seem to form in an
argon atmosphere: a barely noticeable rise in the initial
BA concentration in argon is most likely due to the
carryover of part of the solvent during argon bubbling.
As in the case of TSA [1], the oxidation of SE with
PCA yields benzaldehyde as a product. However, the
SE consumption and BA buildup kinetics are differꢀ
ent; both reactions in the presence of PCA follow the
firstꢀorder law, not zeroꢀorder observed in the TSA
catalysis [1]. Figure 3b presents semilogarithmic
transformations of the rate curves for SE consumption
From the slope of the straight line in the log
m
– t
coordinates, a value of
lated; it is very close to
from the slope of curve
dicted first order in concentration, i.e., the proporꢀ
tionality of the SE conversion rate to its initial concenꢀ
tration is experimentally not observed, the firstꢀorder
rate constant calculated from the slopes of straight
kBA= 2.53
×
×
10–3 s–1 was calcuꢀ
kSE= 3.15
10–3 s–1 calculated
1
in Fig. 3b. However, the preꢀ
lines in the log[SE]
with an increase in [SE]0 (Fig. 4, curve
more, the initial rate [SE]0 weakly depends on
[SE]0; there is no certain tendency toward a rise or fall
in . For example, five runs at [SE]0 of 0.29, 0.43, 0.61,
–
t
coordinates (Fig. 3b) drops
1
). Furtherꢀ
v
= k
v
in the presence of PCA. The steady linearization until 0.87, and 1.30 mol/l to yield curve
1 in Fig. 4 correꢀ
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011

MILD OXIDATION OF STYRENE EPOXIDE
461
0.7
0.6
0.5
0.4
0.3
0.2
0.1
14
12
10
8
(а)
2
1'
6
4
2
'
2
1
0
32
0
4
8
12
16
20
24
28
t, min
1.6
1.6
(b)
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.4
1.2
1.0
1
0.8
0.6
5
0.4
2
0.2
3
4
0.0
–0.2
–0.4
–0.2
–0.4
–0.6
–0.6
0
4
8
12
16
20
24
28
32
36
t
, min
Fig. 3. (a) Rate curves of (
gen and ( 2' ) argon atmospheres at (
and [HClO ] = 3.6 10 mol/l; (b) semilogarithmic transforms of (
1
,
2
) styrene epoxide consumption and (
1
',
2
') benzaldehyde buildup in a BUCH solution in (
1
,
1
') oxyꢀ
–5
2,
1,
1
') [SE] = 0.43 mol/l and [HClO ] = 4.4
×
10 mol/l and (
) SE consumption curves and
10 mol/l under argon; ( ) [SE] = 0.87 mol/l and [HClO ] = 4.4 ×
2, 2 ') [SE] = 0.87 mol/l
2 in Fig. 3a, respectively,
0
4
0
–5
1
,
2
1
4
–5
and those at (
3
) [SE] = 0.87 mol/l and [HClO ] = 5.3
×
4
0
4
0
4
–5
–5
10 mol/l, oxygen; and (5) [SE] = 1.3 mol/l and [HClO ] = 4.4 × 10 mol/l, oxygen; BUCH, 343 K.
0 4
103 values of 1.26, 1.37, 1.50, 1.15, and
in Fig. 4 illustrates
pendence of the rate from [SE]0; this is another differꢀ
ence between the two acid catalysts for the overall conꢀ
version of SE. A comparison of the bimolecular rate
constants of SE oxidation in the presence of PCA
spond to
0.95 mol/(l s), respectively. Curve
the dependence of the pseudounimolecular rate conꢀ
stant on [PCA]. The data of experiments in oxygen
v
×
2
k
and argon atmospheres are quite close; the bimolecuꢀ
lar rate constant calculated from the slope of the
dashed line in Fig. 4 is kb = 30 l/(mol s). In the experꢀ
iments with TSA, a concerted time and concentration
order in SE was observed [1]: the invariability of the SE
(
= 0.77 l/(mol s) and the overall conversion (kb =
k
bO2
30 l/(mol s) shows that the contribution of oxidation
to the overall conversion is approximately 2.5%; i.e.,
there is one oxidation event per 40 events of disappearꢀ
ance of SE molecules. The proportion of oxidation for
consumption rate with time corresponded to the indeꢀ the SE + TSA pair was ~4% or one event per 25 events.
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011

462
PETROV, SOLYANIKOV
[SE] , mol/l
0
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
5
2.0
1.5
1.0
0.5
4
3
2
1
0
2
1
0
2
4
6
5
[HClO ] × 10 , mol/l
4
Fig. 4. Dependence of the overall consumption pseudounimolecular rate constant of styrene epoxide in a BUCH solution at
–5
343 K on (
1) [SE] at [HClO ] = 4.4 × 10 mol/l in an oxygen atmosphere and (2) [HClO ] at [SE] = 0.87 mol/l in (circles)
0
4
4
0
oxygen and (triangles) argon atmospheres.
SEH+ + ROH equilibrium to the left. But oxidation in
BUCH is faster by a factor of ~2 [2]. Here is a conꢀ
trasting example from a close field of chemistry,
The phenomenon of mild oxidation of SE in the presꢀ
ence of TSA was first described in [2]. Comparative
experiments on oxidation in alcohol–chlorobenzene
(BUCH) and acetonitrile (MeCN) solutions raise the the perchloric acidꢀcatalyzed radical degradation of
tertꢀbutyl hydroperoxide. The initiation rate at workꢀ
ing concentrations of [HClO4] = 0.1 and [ROOH] =
question as to why the oxidation rates are close in solꢀ
vents that are so different in nature [2]. Why does the
reaction in BUCH proceeds faster than in MeCN? 0.01 mol/l in an isopropyl alcohol solution is 16 times
lower than in a MeCN solution at 343 K [6, 7]. This is
the expected, normal ratio. However, a special study of
the role of solvent in the TSAꢀcatalyzed oxidation of
SE has shown that the higher the polarity of the
medium and the greater the concentration of protoꢀ
philic OH groups in the solution, the faster the oxidaꢀ
tion [8]; this is surprising. For comparison, we have
shown in [8] that the introduction of alcohol in as low
an amount as 10 vol % into MeCN almost stops the
classical reaction of acidꢀcatalyzed decomposition of
cumyl hydroperoxide. These data illustrate a distinct
abnormality of the acidꢀcatalyzed oxidation of SE
amongst other acidꢀcatalyzed reactions. This abnorꢀ
mality is manifested to even a greater extant in the case
of perchloric acid. It does not catalyze oxidation in
MeCN, although its activity in the alcohol medium is
almost three orders of magnitude higher than that of
TSA. A comparison of the kinetic data for TSA and
PCA reveals a noticeable difference between these catꢀ
alysts in action mechanism. The reactions of overall
The acid type of oxidation catalysis has been clearly
demonstrated: the oxidation rate in both solvents is
strictly proportional to the acid concentration. Proton
transfer from an acid to the substrate, which usually
involves the solvent, is the familiar necessary step in an
acidꢀcatalyzed reaction. In a protophilic alcohol soluꢀ
tion, the concentration of the reactive protonated
form should be much lower than in aprotic MeCN
+
because of displacement of the SE +
ROH2
Data for calculating
k: a run with [SE]0 = 0.43 mol/l and
[HClO4] = 4.4
×
10–5 mol/l under O2 in BUCH at 343 K
tmin
1
2.5
5
8
12
16
21
1.7 1.63 1.54 1.31 1.02 0.78 1.18
1 + logm
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011

MILD OXIDATION OF STYRENE EPOXIDE
463
consumption of SE with the two acids differ in rate complexation between TSA and SE. The results of the
order with respect to epoxide, which are zero for TSA
and unity for PCA. The first order is explained by the
displacement of the equilibrium to the left:
study of the SE–PCA system force us to recur to this
subject. The formal scheme with equilibrium (2)
shifted to the right:
K
k
1
+
k
SE + HAn
O
X
carbene
SE + ROH+2
SEH+ + ROH
(1)
–
k
2
k'
carbonyl oxide
K
SE + HAn
X
(ROH is tertꢀbutanol and SEH+ is protonated
k
2
overall conversion products
epoxide). The SEH+ formation rate is defined by
d
k
suggests that the epoxide and the acid form an interꢀ
mediate complex X, which is consumed in reactions
yielding carbene and heterolysis products. In the
quasiꢀsteadyꢀstate approximation at [SE]0 > [HAn]0
[SEH+]/
dt =
k
+[SE][ROH2 ] –
k
–[SEH+][ROH] –
+
'[SEH+]; in the steady state, [SEH+]
=
+
(
k
+[SE][ROH2 ])/(k+
+
k
–[ROH]) and the overall
consumption rate is
v
=
k
'[SEH+]
=
and [HAn]0 = [HAn] + [X], we get
[X]/[SE]0([HAn]0 – [X]), [X] = [SE]0[HAn]0/(1 +
[SE]0); at
K
=
K
+[SE][ROH2 ]/(k+
+
k
–[ROH])). When equilibꢀ
+
(
k'
k
K
[ROH+2
–[ROH]), the
b/[PCA].
] and
K
[SE]0 ӷ 1, [X] = [HAn]0 and the overall SE
rium (1) is shifted to the left, [PCA]
b[SE][PCA], kb = (
+)/(k+
experimental firstꢀorder rate constant is
≈
consumption rate will be proportional to the acid conꢀ
centration raised to a power of unity (first order) and
independent from [SE] (zero order),
v
=
k
k'
k
+
k
k
=
k
v
= (k1 + k2)
Considering equilibrium (1) in the presence of an
impurity base N, which can be protonated via the
[HAn]. A weakness of this scheme as a whole is the
lack of any independent piece of evidence for comꢀ
plexation. There are two experimental findings that
can be interpreted in favor of complexation. The first
is the distinct independence of the SE oxidation rates
in the presence of TSA or H2SO4 (zero order in SE).
The second is the existence of induction periods of
oxidation for SO4ꢀcontaining acids, especially, sulfuric
acid [5]. It is likely that the intermediate reactive
epoxide–acid complex is formed during the induction
period. Quite an embarrassing feature of the acid
catalysis of epoxide oxidation with oxygen that comꢀ
plicates its investigation is a variety of specific actions
of the chemicals—solvents, epoxides, acids, etc.—
involved in the reactions; here are examples. paraꢀTolꢀ
uenesulfonic acid catalyzes oxidation in both MeCN
and BUCH. Perchloric acid catalyzes only in BUCH.
Hydrochloric acid does not catalyze the process altoꢀ
gether. The oxidation of the SE + H2SO4 pair is charꢀ
acterized by a long induction period, the SE + HClO4
reaction with an acid, forming NH+
,
N + ROH+2
SE + ROH+2
NH+ + ROH
K
1
SEH+ + ROH
k'
where
the mass balance equation [PCA] =
K / ) is the equilibrium constant, and
1 = (k+ k–
+
+
ROH2
SEH+ + NH+ gives a simple expression for the overall
SE consumption rate that explains the drop in the rate
with an increase in [SE]0 and the conservation of the
first order in SE:
v
=
K
k
' [SEH+]=
1[SE]0) (2).
k'K1 [SE]0([PCA] –
[NH+])/([ROH] +
As [SE]0 increases, the concentrations of N and its
protonated form NH+ increase and the actual concenꢀ
tration of PCA in the solution decreases. The rate is
not proportional to [SE]0, but the consumption kinetꢀ
ics is firstꢀorder in epoxide, since the rate curves are binary system is oxidized without any induction
linearized in the semilogarithmic coordinates over the period, and they are detectable only at very low reacꢀ
entire range of epoxide concentrations (Fig. 3a); i.e., tant concentrations in experiments with SE + TSA.
K1[SE]0 Ӷ [ROH] in the denominator of Eq. (2). The However, the paraꢀsubstituted chloro and fluoro
zero order in SE in the reaction with TSA suggested derivatives of styrene epoxide are oxidized with disꢀ
the scheme with equilibrium (1) shifted to the right tinct induction periods. All these peculiarities cannot
[2]. If equilibrium (1) is shifted to the right in the case obscure the main point: catalytic amounts of strong
of TSA, the same must be true in the solution of PCA, protic acids cause fast oxidation of barely oxidizable
which is a stronger acid. Thus, zero order in SE should styrene epoxide and its halogenated derivatives under
be observed then in experiments with PCA, whereas it mild conditions. The main oxidation products are
is firstꢀorder in actuality. The contradiction is elimiꢀ aldehydes and hydrogen peroxide. The formation of
nated if we assume that SE forms a reactive complex aldehydes unequivocally indicates that the conversion
with acids containing the SO4 group, such as TSA and of epoxides involves C–C bond cleavage, the mechaꢀ
sulfuric acid [5]. In [2] we skated over the feasibility of nism of which is of interest. In an inert gas atmoꢀ
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011

464
PETROV, SOLYANIKOV
4. L. V. Petrov, B. L. Psikha, and V. M. Solyanikov,
Neftekhimiya 49, 263 (2009) [Pet. Chem. 49, 245
(2009)].
sphere, aldehydes are not produced, but intermediate
species decompose hydroperoxides at rates that are
quite close to the rate of oxygen absorption in the presꢀ
ence of aldehydes [3].
5. L. V. Petrov and V. M. Solyanikov, Neftekhimiya 40
,
438 (2000) [Pet. Chem. 40, 399 (2000)].
6. V. M. Solyanikov and E. T. Denisov, Izv. AN SSSR. Ser.
Khim., No. 6, 1391 (1968).
REFERENCES
1. L. V. Petrov and V. M. Solyanikov, Neftekhimiya 39
,
7. L. V. Petrov, V. M. Solyanikov, and E. T. Denisov, Izv.
Akad. Nauk SSSR. Ser. Khim., No. 4, 739 (1977).
107 (1999) [Pet. Chem. 39, 89 (1999)].
2. L. V. Petrov and V. M. Solyanikov, Dokl. Akad. Nauk
350, 357 (1996).
8. L. V. Petrov, B. L. Psikha, and V. M. Solyanikov,
Neftekhimiya 47, 379 (2007) [Pet. Chem. 47, 348
(2007)].
3. L. V. Petrov and V. M. Solyanikov, Neftekhimiya 45
,
225 (2005) [Pet. Chem. 45, 202 (2005)].
PETROLEUM CHEMISTRY Vol. 51
No. 6
2011