Liquid-phase noncatalytic butene oxidation with nitrous oxide

Article abstract of DOI:10.1007/s11172-005-0339-8

The kinetics and mechanism of noncatalytic liquid-phase oxidation of but-1-ene and but-2-ene with nitrous oxide in a benzene solution in the temperature range from 180 to 240°C were studied. Oxidation proceeds via the 1,3-dipolar cycloaddition mechanism to form carbonyl compounds. Both of these reactions occur with close rates and activation energies and have the first orders with respect to the alkene and N₂O. A considerable fraction (39%) of but-1-ene involved in oxidation undergoes cleavage at the double bond yielding propanal and an equivalent amount of methylene, the latter producing ethylcyclopropane and cycloheptatriene. The oxidation of but-2-ene proceeds with a minimum bond cleavage and affords methyl ethyl ketone with 84% selectivity. Regularities of the oxidation of terminal and internal alkenes C ₂-C₈ with nitrous oxide were analyzed using the previously published data.

Full text of DOI:10.1007/s11172-005-0339-8

948
Russian Chemical Bulletin, International Edition, Vol. 54, No. 4, pp. 948—956, April, 2005
Liquidꢀphase noncatalytic butene oxidation with nitrous oxide
†
ꢀ
S. V. Semikolenov, K. A. Dubkov, E. V. Starokon´, D. E. Babushkin, and G. I. Panov
G. K. Boreskov Institute of Catalysis, Russian Academy of Sciences,
5 prosp. Lavrent´eva, 630090 Novosibirsk, Russian Federation.
Fax: +7 (383 3) 30 8056. Eꢀmail: dubkov@catalysis.nsk.su
The kinetics and mechanism of noncatalytic liquidꢀphase oxidation of butꢀ1ꢀene and butꢀ
2ꢀene with nitrous oxide in a benzene solution in the temperature range from 180 to 240 °C
were studied. Oxidation proceeds via the 1,3ꢀdipolar cycloaddition mechanism to form carboꢀ
nyl compounds. Both of these reactions occur with close rates and activation energies and have
the first orders with respect to the alkene and N2O. A considerable fraction (39%) of butꢀ1ꢀene
involved in oxidation undergoes cleavage at the double bond yielding propanal and an equivaꢀ
lent amount of methylene, the latter producing ethylcyclopropane and cycloheptatriene. The
oxidation of butꢀ2ꢀene proceeds with a minimum bond cleavage and affords methyl ethyl
ketone with 84% selectivity. Regularities of the oxidation of terminal and internal alkenes
C2—C8 with nitrous oxide were analyzed using the previously published data.
Key words: oxidation, butenes, nitrous oxide, carbonyl compounds, methyl ethyl ketone,
methylene, carbenes.
Scheme 1
Nitrous oxide (N2O) attracts attention as an oxidizing
agent for the gasꢀphase catalytic oxidation. Reactions of
this type are strikingly exemplified by the selective oxidaꢀ
tion of benzene and other aromatic compounds to the
corresponding phenols in the presence of ironꢀcontaining
zeolites FeZSMꢀ5. The works in this area are considered
in several reviews.1—6
It has previously been shown that nitrous oxide can
also be an efficient oxidizing agent in liquidꢀphase oxidaꢀ
tion. These reactions are mainly alkene epoxidation and
alcohol oxidation, occurring with ~100% selectivity in
the presence of the Ru porphyrin complexes,7,8 metalꢀ
containing polyoxometallates,9,10 and some metal—oxide
systems.11 Meanwhile, the most remarkable feature of
nitrous oxide is, probably, its capability for noncatalytic
liquidꢀphase alkene oxidizing to form carbonyl comꢀ
pounds.12—15 Cyclopentene and cyclohexene are oxidized
with N2O at 150—250 °C, transforming into the correꢀ
sponding cyclic ketones with 96—99% selectivity.14,15 The
reaction proceeds via the nonradical 1,3ꢀdipolar cycloadꢀ
dition mechanism to form an intermediate 1,2,3ꢀoxaꢀ
diazoline complex (Scheme 1).
R is hydrocarbon radical or hydrogen atom
and their derivatives. The rate and selectivity of the reacꢀ
tion depend strongly on the substrate composition and
structure. In particular, those effects are especially proꢀ
nounced for linear alkenes C2—C8, whose behavior difꢀ
fers considerably depending on the position of the double
bond. However, it is difficult to compare quantitatively
the results obtained for different alkenes, because even
under strictly standardized experimental conditions (reꢀ
actant concentration, reaction temperature, and pressure)
the real conditions of alkene oxidation are different due
to different boiling points of alkenes. For instance, for the
reaction temperature 220 °C, the main amount of octꢀ1ꢀ
ene (b.p. 121 °C) exists in solution, whereas a considerꢀ
able portion of ethylene (b.p. –107 °C) goes to the gas
phase and exists outside the reaction zone, thus resulting
in a seemingly decreased activity in the reaction with N2O.
To avoid these difficulties, in this work, we studied
the oxidation of butene isomers with the terminal and
internal double bonds. Since the physical properties of
The energy route of cyclohexene oxidation was comꢀ
pletely described16,17 by the quantumꢀchemical method.
The results agree well with the experimental data.
The recent screening of substrates shows18 that nitrous
oxide can noncatalytically oxidize alkenes of different
classes, including linear, cyclic, and heterocyclic alkenes
† Deceased.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 4, pp. 925—933, April, 2005.
1066ꢀ5285/05/5404ꢀ0948 © 2005 Springer Science+Business Media, Inc.

Liquidꢀphase oxidation of butenes with N2O
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
949
After the reactants were loaded, the reactor was heated with a
rate of 6 °C min–1 to the experimental temperature (180—240 °C)
at which the reactor was stored for a specified time. After the
end of the experiment, the reactor was cooled to room temperaꢀ
ture. The gas phase was analyzed by gas chromatography. Furꢀ
ther, after the pressure was slowly discharged, the liquid phase
was analyzed by GLC. The composition of the gas phase (N2O,
N2, О2, СО, СО2) was analyzed on a Tsvet 500 M chromatoꢀ
graph at room temperature using a heatꢀconductivity detector
and a column packed with Porapak Q. For a higher measureꢀ
ment accuracy of small amounts of CO and CO2, they were preꢀ
hydrogenated over a nickel catalyst and then analyzed as methꢀ
ane on a flameꢀionization detector. Molecular nitrogen was the
only gaseous reaction product, whose amount corresponded to
that of the oxygenꢀcontaining compounds formed in the liquid
phase. In all cases, the COх concentration did not exceed
0.01 mol.%.
Liquid reaction products were analyzed by chromatography
on a Kristal 2000 instrument equipped with a capillary column
(SЕꢀ50, 50 m×0.2 mm) and a flameꢀionization detector. Caliꢀ
brating solutions of authentic compounds were used to identify
the products and determine their concentration. The products
were additionally identified by 1Н and 13С NMR spectroscopy
(Bruker DPXꢀ250, 1Н, 250.13 MHz; 13C, 62.9 MHz; Bruker
MSLꢀ400, 1Н, 400.13 MHz; 13C, 100.61 MHz) and GCꢀMS
(VG Analytical Ltd. 7070HS and Varian Saturn 2000).
butꢀ1ꢀene and butꢀ2ꢀene are similar, this study should
certainly answer the question about the influence of the
alkene structure on the reaction mechanism and compoꢀ
sition of the products. The reaction can also be of practiꢀ
cal interest, because the main expected product is such a
valuable substance as methyl ethyl ketone (1), whose
methods for preparation are far from perfection.
In the final part of the present work, we consider the
results in comparison with the earlier obtained data for
linear alkenes18 to reveal general regularities of oxidation
with nitrous oxide.
Experimental
Experiments on butene oxidation were carried out in benꢀ
zene solution using a stainless steel autoclaveꢀtype 100ꢀcm3 reꢀ
actor (Parr). To decrease a possible condensation of compoꢀ
nents of the reaction mixture in cold parts of the reactor, the
stirrer was removed and the solution was stirred with a glassꢀ
sealed metallic rod, which was rotated (300 rpm) with an alterꢀ
nating magnetic field providing the reaction to occur in the
kinetic region.
The scheme of the setup is shown in Fig. 1. The reactor was
loaded with benzene (50 cm3) and tightly closed. Then, air was
removed for 1 min from the reactor through shutoff valves 1
and 2, and butene and N2O were conveyed. Butene was supplied
by portions from metallic cylinder 5 through intermediate 1ꢀL
bypass cylinder 6. Upon loading the reactor, butene was abꢀ
sorbed with benzene, and the amount of the loaded alkene was
determined from a pressure change in the vessel. If necessary,
this procedure was repeated several times to provide the reꢀ
quired initial amount of butene. In the most part of experiꢀ
ments, the butene load was N 0R = 0.08 mol.
The butene conversion (ХR) was calculated from the amount
of the products taking with stoichiometric coefficients. The reꢀ
action rate was determined by the equation
W = N 0RXR/(100Vt),
(1)
where N 0 is the initial amount of butene, V is the solution
R
volume, and t is the reaction duration.
Butꢀ1ꢀene and butꢀ2ꢀene (99%, Aldrich) and medical 99.8%
nitrous oxide (Azot Plant, Cherepovets, Russia), which is delivꢀ
ered in the fluidized state in metallic cylinders under ~50 atm,
were used in experiments. Benzene (99%) was used as solvent.
Nitrous oxide was supplied from a 0.5ꢀL bypass cylinder 7.
The amount of loaded nitrous oxide usually was N 0
=
N O
2
0.12 mol. Nitrous oxide is well dissolved in benzene, and the
initial N2O pressure in the reactor for this load is ~10 atm.
Results and Discussion
9
2
3
3
Butꢀ1ꢀene oxidation. The results on butꢀ1ꢀene oxidaꢀ
tion in the 180—240 °C temperature region are presented
in Table 1. The main reaction products are carbonyl comꢀ
pounds 1—3, being ~75% in total, and two compounds
containing no oxygen: ethylcyclopropane (4) and cycloꢀ
heptatriene (5). A small amount of unidentified products
(1.4—2.5 mol.%) is also present in the solution.
8
13
14
1
7
1
3
6
1
10
11
The duration of experiments at different temperatures
was chosen in such a way that a relatively low conversion
of butꢀ1ꢀene (<25%) was provided. Entry 4 with a higher
conversion (220 °C, 12 h) was carried out to compare
butꢀ1ꢀene with other linear alkenes, whose oxidation has
previously been studied under these conditions.18 The
data in Table 1 show that the reaction rate increases by
more than an order of magnitude with the temperature
increase from 180 to 240 °C. The activation energy calcuꢀ
4
5
C4H8
N2O
12
13
Fig. 1. Scheme of the experimental setup: 1—3, shutoff valves;
4 and 5, gas cylinders; 6 and 7, bypass cylinders; 8, reference
pressure gauge; 9, pressure gauge; 10, reactor (autoclave);
11, capillary for gas supply; 12, oven; 13, magnetic stirrer; and
14, vacuum pump.
lated from the Arrhenius plot (Fig. 2) is 21 1 kcal mol–1
,

950
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
Semikolenov et al.
Table 1. Temperature effect on the parameters of butꢀ1ꢀene oxidation*
Entry T/°C τ/h Conversion
of butꢀ1ꢀene,
Composition of reaction products in liquid phase (mol.%)
W
/µmol cm–3 h–1
1
Butanal
Propaꢀ
Ethylcycloꢀ
Cyclohepꢀ
Other
XR (%)
(2)
nal (3)
propane (4)
tatriene (5)
products
1
2
3
4
5
180
200
220
220
240
12
12
3
12
2
6.9
19.8
11.9
37.9
17.6
34.7
35.3
34.1
33.3
32.7
33.8
11.1
11.6
12.7
13.4
14.2
12.8
30.5
26.4
30.6
29.4
30.0
29.5
16.3
12.4
12.7
12.6
12.3
13.1
6.0
12.4
7.4
9.5
8.7
1.4
2.0
2.5
1.8
2.1
1.8
9.2
26.4
63.5
50.5
140.8
Averaged concentrations:
9.0
* Reaction conditions: N 0R = 0.08 mol, N 0N O = 0.12 mol, solvent benzene (50 mL).
2
which slightly differs from a value of 24 kcal mol–1 calcuꢀ
lated by the quantumꢀchemical study of this reaction.17
Reaction mechanism. The composition of the products
of butꢀ1ꢀene oxidation (see Table 1) agrees well with the
mechanism of 1,3ꢀdipolar cycloaddition of N2O to the
alkene double bond.12,17 The general reaction mechaꢀ
nism is shown in Scheme 2.
lnW
5
4
3
2
As can be seen, two reaction routes (М1 and М2) are
possible, depending on which the carbon atom (first or
second) is bound to the oxygen atom of the N2O molꢀ
ecule. According to this, the intermediate oxadiazoline
complex can exist in two configurations: I and II. Correꢀ
spondingly, this leads to the formation of compounds 2
(route i) and 1 (route ii). Complex II can also decompose
through the carbon—carbon bond cleavage to form
propanal (3) and an equivalent quantity of methylene
(route iii). Methylene, being a highly reactive carbene
1
2
2.0
2.1
2.2
T –1•103/K–1
Fig. 2. Arrhenius plots for the oxidation of butꢀ1ꢀene (1) and
butꢀ2ꢀene (2) by nitrous oxide (according to the data in
Tables 1 and 3).
Scheme 2

Liquidꢀphase oxidation of butenes with N2O
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
951
species,19,20 reacts rapidly either with butꢀ1ꢀene to yield
ethylcyclopropane (4, route iv), or with benzene affordꢀ
ing cycloheptatriene (5, route v).
thylene with butꢀ1ꢀene (kiv) and benzene (kv) can be preꢀ
sented as follows:
kiv/kv = C4CC H /(C5CC H8).
(4)
It cannot be excluded that the cleavage primarily proꢀ
duces not methylene but diazomethane12 (CH2N2), which
can rapidly decompose to methylene and nitrogen or reꢀ
acts directly with butꢀ1ꢀene and benzene. Since diazoꢀ
methane should be very unstable under the reaction conꢀ
ditions and produces the same products as methylene
does, it is difficult to establish the exact nature of a species
formed by the cleavage. In this case, we can speak about
the formation of methylene in the reaction.
6
6
4
Calculation by Eq. (4) gives the ratio kiv : kv = 10,
which indicates a much higher reactivity of carbene toꢀ
ward alkene. The real difference in activities should be
higher, because a considerable fraction of butꢀ1ꢀene is in
the gas phase, i.e., outside the solution where the methylꢀ
eneꢀgenerating oxidation occurs, due to the low boiling
point of butꢀ1ꢀene (–6.3 °C).
Note that, according to Scheme 2, the yield of propanal
should be equal to the sum of the yields of compounds 4
and 5 formed from methylene. However, the experimenꢀ
tal yield of propanal (3) is somewhat higher (see Table 1).
This can be related to the consumption of methylene in
reactions with butꢀ1ꢀene and benzene in other routes. We
assumed that methylene can react with N2O to form formꢀ
aldehyde. However, detailed analysis using special chroꢀ
matographic procedures and NMR found neither formalꢀ
dehyde nor additional products of methylene conversion.
Nevertheless, it cannot be excluded that formaldehyde is
formed but polymerizes, under the reaction conditions,
to produce an insoluble product, which is not detected by
the above analyses.
Quantitative comparison of the composition of the
reaction products with Scheme 2 makes it possible to
determine (a) contribution of routes М1 and М2 to the
overall oxidation rate, (b) fraction of the reaction leading
to the cleavage of molecules at the double bond, and
(c) reactivity of methylene toward butꢀ1ꢀene and benzene.
The results presented in Table 1 indicate that the comꢀ
position of the reaction products is almost independent of
the temperature and duration of experiments within the
experimental accuracy.
As can be seen from the data in Scheme 2, the contriꢀ
bution of route М1 to the overall oxidation rate (FM ) can
1
be expressed through the concentrations of the correꢀ
sponding products:
The ratio between the competitive reactions of methꢀ
ylene can be controlled by the variation of the amounts of
alkene and benzene in the reaction system. The results of
experiments, where the amount of benzene was constant
(50 cm3) and the amount of butꢀ1ꢀene was increased from
0.04 to 0.16 mol, are presented in Table 2. It is seen that
an increase in the butꢀ1ꢀene amount in the system results,
in fact, in a proportional increase in the ratio of comꢀ
pounds 4 : 5 in the reaction products. The data presented
in Table 2 also indicate that the reaction rate is proporꢀ
tional to the amount of introduced butꢀ1ꢀene, i.e., the
reaction has the first order with respect to alkene.
It has previously been shown12,14 that nitrous oxide is
inert toward organic compounds containing no С=С bond
(cyclohexane, benzene, acetonitrile, ethanol, and other).
Therefore, alkenes can be oxidized by nitrous oxide only
in the presence of different solvents usually used in orꢀ
ganic synthesis. The solvent nature exerts no considerable
effect on the alkene conversion and selectivity with reꢀ
spect to carbonyl compounds.18 At the same time, a solꢀ
vent can affect the character and ratio of the products of
methylene conversion, if the latter forms. For instance,
when toluene is used instead of benzene as a solvent, the
conversion of butꢀ1ꢀene and selectivity with respect to
the main products remain virtually unchanged, but
methylcycloheptatriene is formed instead of cycloheptaꢀ
triene (5).
FM = M1/(M1 + M2) = C2/(C1 + C2 + C3).
(2)
1
Calculation with the averaged data in Table 1 shows
that FM is 17%. Thus, the reaction mainly proceeds via
1
route M2 through intermediate complex II.
The fraction of the reaction leading to the cleavage of
butꢀ1ꢀene molecules at the double bond (Fcleav) can be
found from the equation
Fcleav = C3/(C1 + C2 + C3),
(3)
which gives Fcleav = 39%.
As already mentioned in earlier works,12,18 the "cleavꢀ
age" mechanism of terminal alkene oxidation affords a
stoichiometric amount of methylene (or diazomethane)
and aldehyde with a smaller number of carbon atoms.
Methylene is grouped with highly reactive carbene interꢀ
mediates that are used in organic synthesis. Several methꢀ
ods for methylene generation are known.19,20 The simꢀ
plest of them is based on the thermal, catalytic, or photoꢀ
decomposition of diazomethane, which is a very expenꢀ
sive, toxic, and explosive substance. The generation of
:СН2 by alkene oxidation using N2O is a simple method,
which provides, in the case of butꢀ1ꢀene, a considerable
yield of this carbene (39%).
Knowing the contents of butꢀ1ꢀene and benzene (with
which methylene reacts), one can estimate the relative
reactivity of methylene to these substrates. Based on
Scheme 2, the ratio of the reaction rate constants of meꢀ
Butꢀ2ꢀene oxidation. The results of experiments on
butꢀ2ꢀene oxidation, which were carried out under the

952
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
Semikolenov et al.
Table 2. Influence of the initial amount of butꢀ1ꢀene on the oxidation parameters*
Entry
N 0
/mol
XR (%)
Composition of the reaction products in solution (mol.%)
4 : 5
W
R
/µmol cm–3 h–1
1
2
3
4
5
Other
products
1
2
3
0.04
0.08
0.16
11.1
11.9
10.8
31.9
34.1
33.9
13.0
12.7
12.5
28.0
30.6
31.7
12.3
12.7
14.8
11.7
7.4
4.6
3.1
2.5
2.5
1.1
1.7
3.2
29.6
63.5
115.2
* Reaction conditions: N 0
= 0.12 mol, solvent benzene (50 mL), 220 °C, 3 h.
N O
2
Table 3. Temperature effect on the parameters of butꢀ2ꢀene oxidation
Entry
T
/°C
τ/h
XR
(%)
Composition of reaction products in liquid phase (mol.%)
W
/µmol cm–3 h–1
1
Isobuꢀ Acetalꢀ
tanal dehyde
1,2,3ꢀTrimeꢀ
thylcycloꢀ
7ꢀMethylcycloꢀ
heptaꢀ1,3,5ꢀ
triene (9)
Other
products
(6)
(7)
propane (8)
1
2
3
4
5
180
200
220
220
240
12
4.0
82.0
84.6
84.5
85.1
83.9
84.0
2.8
3.4
3.5
3.3
3.6
3.3
9.1
6.1
7.0
7.2
7.7
7.4
4.5
4.7
3.6
3.4
3.3
3.9
0.2
0.2
0.2
0.1
0.2
0.2
1.4
1.1
1.2
0.9
1.3
1.2
5.3
17.9
34.1
30.8
100.0
12 13.4
6.4
12 23.1
12.5
3
2
Averaged concentrations:
* Reaction conditions: N 0R = 0.08 mol, N 0N O = 0.12 mol, solvent benzene (50 mL).
2
conditions similar to those of butꢀ1ꢀene oxidation, are
presented in Table 3. In this case, the main products
are also carbonyl compounds (methyl ethyl ketone (1),
isobutanal (6), and acetaldehyde (7)), being >90% in
total. In addition, 1,2,3ꢀtrimethylcyclopropane (8) and
7ꢀmethylcycloheptaꢀ1,3,5ꢀtriene (9) are formed. As in
the case of butꢀ1ꢀene, the composition of the products
of butꢀ2ꢀene oxidation is virtually temperatureꢀindeꢀ
pendent.
Noteworthy that the reaction is highly selective with
respect to methyl ethyl ketone (1): the fraction of 1 in the
reaction products is 84% on the average. Compound 1 is a
valuable chemical product widely use as solvent in vanꢀ
ishꢀandꢀpaint and petroleum processing industries. At
present, this compound is mainly produced from a mixꢀ
ture of butꢀ1ꢀene and butꢀ2ꢀene through a complicated
and poorly selective threeꢀstep process: sulfonation of
butenes, hydrolysis of isobutyl sulfate, and dehydrogenaꢀ
tion of the butanꢀ2ꢀol that formed to yield 1. Since niꢀ
trous oxide, which can be obtained by the selective oxidaꢀ
tion of ammonia with oxygen,21,22 is cheap, the results
presented on the oxidation of butꢀ2ꢀene with nitrous oxꢀ
ide are worth of attention as a new alternative method for
methyl ethyl ketone (1) preparation.
cal calculation is somewhat higher than the experimental
value: 26 kcal mol–1 17
.
The results of experiments on butꢀ2ꢀene oxidation
at different initial amounts of the reactants are listed
in Table 4. The amount of N2O was varied within
0.04—0.19 moles, and that of butꢀ2ꢀene changed from
0.04 to 0.16 moles. Based on these data, we plotted the
functions of the reaction rate vs. amounts of N2O and
butꢀ2ꢀene (Fig. 3). The linear character of these plots
indicates the first reaction order with respect to each reꢀ
actant. Thus, the kinetic equation of butene oxidation
can be presented as follows:
W = k0exp[–E/(RT )]CC H CN O
,
(5)
4
8
2
where k0 is the preꢀexponential factor, Е is the activation
energy, and CC H and CN O are the concentrations of the
4
2
reactants in solu8tion. The exact concentrations of the
dissolved reactants, under experimental conditions, are
unknown but they are assumed to be proportional to the
amounts of introduced substrates.
Reaction mechanism. The mechanism of butꢀ2ꢀene oxiꢀ
dation is presented in Scheme 3. Unlike butꢀ1ꢀene, the
butꢀ2ꢀene molecule is symmetrical relatively to the posiꢀ
tion of the С=С bond. Therefore, one can propose only
one configuration of the intermediate oxadiazoline comꢀ
plex for which the oxygen atom is always bonded to the
second carbon atom. As in the case of butꢀ1ꢀene, this
The Arrhenius plot for the reaction rate of butꢀ2ꢀene
oxidation, which corresponds to an activation energy of
22 1 kcal mol–1, is presented in Fig. 2. The theoretical
activation energy value obtained by the quantumꢀchemiꢀ

Liquidꢀphase oxidation of butenes with N2O
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
953
Table 4. Influence of the initial amount of the reactants on the parameters of butꢀ2ꢀene oxidation at 220 °C (solvent
benzene, 50 mL)
Entry
N 0
N 0
τ/h
XR Composition of reaction products in liquid phase (mol.%)
W
R
N O
2
(%)
/µmol cm–3 h–1
mol
1
6
7
8
9
Other
products
1
2
3
4
5
6
7
0.08
0.08
0.04
12
6
6
8.6
7.0
13.5
23.6
81.6
82.1
84.0
84.0
81.2
79.2
84.7
3.4
3.2
4.0
3.4
2.9
3.3
3.8
3.3
8.1
8.3
6.7
7.6
7.8
9.8
9.7
7.1
3.1
4.0
4.3
3.5
2.6
2.4
2.6
2.6
0.3
0.3
0.5
0.1
0.2
0.3
0.6
0.4
2.6
2.7
2.4
1.4
2.5
3.1
4.1
2.1
11.5
18.7
36.0
62.9
10.2
20.8
81.1
0.06
0.12
0.19
0.06
0.12
0.12
0.08
0.08
0.04
0.04
0.16
6
12 15.3
6
6
15.6
15.2
82.4
Averaged concentrations:
W/µmol cm–3 h–1
(route i); in the second case, acetaldehyde (7) and
ethylidene :CH—Me are formed (route ii). The latter reꢀ
acts subsequently with butꢀ2ꢀene and benzene to afford
1,2,3ꢀtrimethylcyclopropane (8, route iii) and 7ꢀmethylꢀ
cycloheptaꢀ1,3,5ꢀtriene (9, route iv), respectively.
1
80
60
40
20
2
The formation of butanal 6 in step i can be understood
by the detailed consideration of the mechanism of "nonꢀ
cleavage" decomposition of the oxadiazoline complex
shown in Scheme 4. It can be proposed that the eliminaꢀ
tion of an N2 molecule in this complex affords a comꢀ
pound (А), whose further rearrangement proceeds as eiꢀ
ther hydride 1,2ꢀshift to form 1, or the 1,2ꢀshift of the Me
group to form isobutanal via the mechanism similar to the
pinacoline rearrangement. The ratio of the amounts of
these products (see Table 3) shows that hydrogen transfer
is ~25ꢀfold faster than the transfer of the Me group.
Although the rearrangement of compound А to alkene
oxide (2,3ꢀepoxybutane) cannot basically be excluded,
this compound is not formed in the reaction. This fact
agrees with the experimental data12,14,15 and quantumꢀ
0
0.04
0.08
0.12
0.16
N 0/mol
Fig. 3. Plots of the reaction rate (W ) vs. initial amounts of butꢀ
2ꢀene (1) and N2O (2) (according to the data in Table 4).
complex can decompose via two routes: without and with
С—С bond cleavage in the complex. In the first case,
methyl ethyl ketone (1) and isobutanal (6) are formed
Scheme 3

954
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
Semikolenov et al.
Scheme 4
chemical calculations,16,17 which show that alkene oxides
are neither intermediates, nor reaction products between
alkenes and nitrous oxide.
It seems interesting to estimate the ratio of reaction
rates i and ii (see Scheme 3), i.e., the reactions with and
without C—C bond cleavage in the oxadiazoline complex
of butꢀ2ꢀene. The fraction of step ii, which proceeds via
the cleavage mechanism (Fcleav), can be expressed through
the concentrations of the corresponding reaction products
estimation, most likely, is incorrect. We have already menꢀ
tioned that the lower the boiling point of alkene, the
greater its fraction occurrence in the gas phase, i.e., outꢀ
side the reaction zone, under the experimental condiꢀ
tions. The effect of unfavorable phase distribution should
be especially pronounced for ethylene and weaken as the
boiling point of the alkenes approaches that of the solvent
(benzene, 80.1 °C). Taking into account this fact, we
conclude that the data in Table 5 indicate a monotonic
decrease in the reactivity of terminal alkenes with an
increase in the number of carbon atoms in the series
С2—С8. This conclusion is valid, probably, for internal
alkenes.
The position of the double bond in an alkene molꢀ
ecule can also affect its reactivity. This question can be
answered quite exactly by a comparison of two butene
isomers. The results presented in Tables 1 and 3 and in
Fig. 2 show that the reactivity of terminal butꢀ1ꢀene is by
~1.5 times higher than that of internal butꢀ2ꢀene.
Ratio of the reaction routes. As can be seen for butꢀ1ꢀ
ene (see Scheme 2), two reaction routes (М1 and М2) are
possible. The probability of this or another route depends
on the configuration of the intermediate oxadiazoline
complex in which the oxygen atom is bonded to the first
(complex I) or second (complex II) carbon atom. The
contribution of each route can be estimated from the
composition of the resulting products. The results of the
estimation for the studied alkenes are given in Table 5 as a
Fcleav = C7/(C1 + C6 + C7).
(6)
Calculation using the averaged data in Table 3 shows
that the fraction of "cleavage" mechanism Fcleav for butꢀ2ꢀ
ene is only 8%, which is manifold lower than Fcleav for
butꢀ1ꢀene that is equal to 39%.
Ethylidene :СН—Me formed by the C—C bond cleavꢀ
age is much less reactive than methylene :СН2 and almost
incapable of reacting with benzene. Estimation taking into
account the amount of butꢀ2ꢀene and benzene in the
reaction system shows that the reactivity of this carbene
toward benzene is at least two orders of magnitude lower
than its reactivity toward alkene.
Regularities of the oxidation of linear alkenes. The data
of the previous work18 on the oxidation of linear alkenes
were supplemented by the results on butene oxidation
and are presented in Table 5. All the experiments were
carried out at 220 °C for 12 h at similar initial amounts of
alkenes and nitrous oxide. This set of substrates, includꢀ
ing five terminal and two internal alkenes, makes it posꢀ
sible to analyze qualitatively the influence of the compoꢀ
sition and structure of alkenes on the reaction course.
Reactivity of alkenes. Since the experiments were carꢀ
ried out under similar conditions, it seems reasonable to
use the alkene conversion as a characteristic of the reacꢀ
tivity toward nitrous oxide. As can be seen from the data
in Table 5, when the number of carbon atoms increases in
the case of terminal alkenes, the conversion passes through
a maximum and reaches the highest value for butꢀ1ꢀ
ene (38%). However, taking into account high differences
in the physical properties of the alkenes under study, this
value of FM , which is the fraction of route М1 in the
1
overall oxidation rate.
Naturally, in the case of ethylene when only comꢀ
plex I can form, FM is 100%. When the formation of
complex II becomes1 probable, the fraction of М1 deꢀ
creases sharply, being only 30% for propylene and 17%
for butꢀ1ꢀene. The FM value remains approximately unꢀ
1
changed with the further elongation of the carbon chain.
This implies that the formation of complex II is energetiꢀ
cally more favorable than that of complex I, which agrees
with the theoretical calculations.17 This regularity can be
due, probably, to the electrophilic nature of oxygen, due
to which oxygen is preferentially bound to the carbon

Liquidꢀphase oxidation of butenes with N2O
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
955
Table 5. Oxidation of linear alkenes*
Entry
Alkene
B.p.
XR
FM
Fcleav
Product
Yield
1
/°C
(mol.%)
%
Terminal alkenes
1
2
Ethylene
–107.3
–47.8
27
26
100
7
Acetaldehyde
Cyclopropane
Cycloheptatriene
Propanal
91
4
3
Propylene
30
29
23
31
22
4
15
13
34
29
13
9
15
32
27
13
10
13
28
26
9
Acetone
Acetaldehyde
Methylcyclopropane
Cycloheptatriene
Butanal
Methyl ethyl ketone
Propanal
Ethylcyclopropane
Cycloheptatriene
Hexanal
Hexanꢀ2ꢀone
Pentanal
Butylcyclopropane
Cycloheptatriene
Octanal
Octanꢀ2ꢀone
Heptanal
3
4
5
Butꢀ1ꢀene
Hexꢀ1ꢀene
Octꢀ1ꢀene**
–6.25
63.5
38
35
30
17
21
19
39
38
39
121.3
Hexylcyclopropane
Cycloheptatriene
18
Internal alkenes
1
2
Butꢀ2ꢀene
3.7
23
22
0
8
Methyl ethyl ketone
Isobutanal
85
3
7
Acetaldehyde
Trimethylcyclopropane
Pentanꢀ2ꢀone
Pentanꢀ3ꢀone
Acetaldehyde
Propanal
3
Pentꢀ2ꢀene
36.9
0
8
41
47
5
3
Cyclopropanes
1
* Reaction conditions: N 0R = 0.08 mol, N 0N O = 0.12 mol, solvent benzene (50 mL), 220 °C, and 12 h.
2
** The initial amount of octꢀ1ꢀene is 0.06 moles.
atom having an alkyl substituent and, hence, an enhanced
electronꢀreleasing ability.
virtually does not affect the composition of the reaction
products.
Double bond cleavage. When studying butenes, we
found that a molecule is cleaved at the double bond only
if the reaction proceeds via route М2, i.e., through interꢀ
mediate complex II. However, the data for ethylene
(see Table 5) show that a small fraction of cleavage
(Fcleav = 7%) can also be observed for complex I, i.e., for
route М1. The fact that this cleavage is not observed for
the oxidation of other terminal alkenes can be related to
its insignificant contribution to the overall reaction rate.
In fact, if we accept that the fraction of cleavage for route
M1 is 7% in all cases, the contribution of this cleavage to
the overall oxidation rate can be estimated for other alkꢀ
enes. The estimation shows that it should be 2.1% for
propylene and ~1.4% for other terminal alkenes, which
As can be seen from the data in Table 5, for terminal
alkenes, the contribution of the mechanism with С=С
bond cleavage (Fcleav) to the overall oxidation rate inꢀ
creases sharply on going from ethylene (7%) to propylene
(29%) and butꢀ1ꢀene (39%) and further remains almost
unchanged. Internal alkenes are much weaker prone to
cleavage (Fcleav = 8%). The reason for so high difference
in the behavior of terminal and internal alkenes remains
unclear. At first glance, this can be related to the dissipaꢀ
tion of the energy that evolves upon the oxidative attack
to the double bond, if assuming that the efficiency of this
process is higher for the more remote position of the
double bond from the "end" of the molecule. However,
the fraction of cleavage for the "doubly" terminal ethylene

956
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 4, April, 2005
Semikolenov et al.
10. R. BenꢀDaniel, L. Weiner, and R. Neuman, J. Am. Chem.
Soc., 2002, 124, 8788.
11. T. L. Stuchinskaya and I. V. Kozhevnikov, Catal. Commun.,
2003, 4, 609.
is as low (7%) as that in the case of internal butꢀ2ꢀene and
pentꢀ2ꢀene (8%). Therefore, this assumption is rather imꢀ
probable.
12. F. S. BridsonꢀJones, G. D. Buckley, L. H. Cross, and A. P.
Driver, J. Chem. Soc., 1951, 2999.
13. F. S. BridsonꢀJones and G. D. Buckley, J. Chem. Soc.,
1951, 3009.
14. G. I. Panov, K. A. Dubkov, E. V. Starokon, and V. N.
Parmon, React. Kinet. Catal. Lett., 2002, 76, 401.
15. K. A. Dubkov, G. I. Panov, E. V. Starokon, and V. N.
Parmon, React. Kinet. Catal. Lett., 2002, 77, 197.
16. V. I. Avdeev, S. Ph. Ruzankin, and G. M. Zhidomirov,
Chem. Commun., 2003, 42.
The authors thank A. Ermakova and V. M. Tormyshev
for discussion of the results and V. A. Utkin for GCꢀMS
analysis of the reaction products.
This work was financially supported in part by
the Presidium of the Russian Academy of Sciences
(Grant 9.3, 2003).
References
17. V. I. Avdeev, S. F. Ruzankin, and G. M. Zhidomirov, Kinet.
Katal., 2005, 46, No. 2, 191 [Kinet. Catal., 2005, 46, No. 2
(Engl. Transl.)].
18. E. V. Starokon, K. A. Dubkov, D. E. Babushkin, V. N.
Parmon, and G. I. Panov, Adv. Synth. Catal., 2004, 346, 268.
19. W. Kirmse, Carbene Chemistry, Academic Press, New
York, 1971.
20. O. M. Nefedov, A. I. Ioffe, and L. G. Menchikov, Khimiya
karbenov [The Chemistry of Carbenes], Khimiya, Moscow,
1990, 304 pp. (in Russian).
21. A. K. Uriarte, Stud. Surf. Sci. Catal., 2000, 130, 743.
22. A. S. Noskov, I. A. Zolotarskii, S. A. Pokrovskaya, V. N.
Korotkikh, E. M. Slavinskaya, V. V. Mokrinskii, and V. N.
Kashkin, Chem. Eng. J., 2003, 91, 235.
1. G. I. Panov, A. S. Kharitonov, and V. I. Sobolev, Appl.
Catal. A: General, 1993, 98, 1.
2. A. Reizman, M. Häfele, and G. Emig, Trends Chem. Eng.,
1996, 3, 63.
3. G. Panov, A. Uriarte, M. Rodkin, and V. Sobolev, Catal.
Today, 1998, 4, 365.
4. G. Genti, S. Perathoner, and F. Vazzana, CHEMTECH,
1999, 48.
5. G. Panov, CATTECH, 2000, 4, 8.
6. A. V. Leont´ev, O. A. Fomicheva, M. V. Proskurina, and
N. S. Zefirov, Usp. Khim., 2001, 70, 107 [Russ. Chem. Rev.,
2001, 70 (Engl. Transl.)].
7. T. Yamada, K. Hashimoto, Y. Kitaichi, K. Suzuki, and
T. Ikeno, Chem. Lett., 2001, 268.
8. K. Hashimoto, Y. Kitaichi, H. Tanaka, T. Ikeno, and
T. Yamada, Chem. Lett., 2001, 992.
9. K. Hashimoto, H. Tanaka, T. Ikeno, and T. Yamada, Chem.
Lett., 2002, 582.
Received October 3, 2004;
in revised formed January 14, 2005