Synthesis of α-hydroxyalkyl peroxide esters and ethers as initiators of radical processes

Article abstract of DOI:10.1134/S1070427214120167

α-Hydroxyalkyl peroxide esters and ethers containing, along with the peroxy bond, also acyloxy and alkoxy groups, are effective initiators of radical processes. The initiation efficiency is due to the capability of these molecules to generate simultaneously oxygen- and carbon-centered radicals capable of hydrogen abstraction from the substrate and of addition to C=C bonds. The selectivity of the synthesis of α-hydroxyalkyl peroxide esters and ethers is mainly determined by the steps of the synthesis and isolation of the intermediate α-chloroalkyl peroxides, because of their high reactivity. Experimental conditions allowing control of the synthesis of α-hydroxyalkyl peroxide esters and ethers were found.

Full text of DOI:10.1134/S1070427214120167

ISSN 1070-4272, Russian Journal of Applied Chemistry, 2014, Vol. 87, No. 12, pp. 1890−1900. © Pleiades Publishing, Ltd., 2014.
Original Russian Text © T.P. Aleinikova, V.A. Navrotskii, 2014, published in Zhurnal Prikladnoi Khimii, 2014, Vol. 87, No. 12, pp. 1816−1827.
ORGANIC SYNTHESIS AND INDUSTRIAL
ORGANIC CHEMISTRY
Synthesis of α-Hydroxyalkyl Peroxide Esters and Ethers
as Initiators of Radical Processes
T. P. Aleinikova and V. A. Navrotskii
Volgograd State Technical University, pr. Lenina 28, Volgograd, 400131 Russia
e-mail: aleynikova_tp@vstu.ru
Received July 14, 2014
Abstract—α-Hydroxyalkyl peroxide esters and ethers containing, along with the peroxy bond, also acyloxy
and alkoxy groups, are effective initiators of radical processes. The initiation efficiency is due to the capability
of these molecules to generate simultaneously oxygen- and carbon-centered radicals capable of hydrogen ab-
straction from the substrate and of addition to >С=С< bonds. The selectivity of the synthesis of α-hydroxyalkyl
peroxide esters and ethers is mainly determined by the steps of the synthesis and isolation of the intermediate
α-chloroalkyl peroxides, because of their high reactivity. Experimental conditions allowing control of the synthesis
of α-hydroxyalkyl peroxide esters and ethers were found.
DOI: 10.1134/S1070427214120167
(CH3)3COOCH(CH3)OR
(CH3)3CO + CH3CO + ROH.
α-Hydroxyalkyl peroxide (OP) esters and ethers
containing, along with the O–O bond, also alkoxy
and acyloxy groups are effective initiators of radical
processes: (co)polymerization [1, 2], vulcanization,
addition to multiple bonds [3], and substitution of
С–Н bonds [4]; they are also used as comonomers in
copolymerization with nonperoxy monomers [1, 5] and in
polymer compounds [1, 6–10]. The polyfunctionality of
these molecules is responsible for their ability to generate
both oxygen-centered radicals, abstracting hydrogen
from a substrate in most cases, and carbon-centered alkyl
radicals, undergoing addition to multiple bonds. Owing to
this ability, OP esters and ethers find diverse applications
as initiators of radical processes.
·
·
The generated tert-butoxy radical abstracts the
hydrogen atom from the substrate to form tert-butanol
(TB) or decomposes to acetone and methyl radical:
The acetyl radicals undergo decarbonylation to form
methyl radicals:
·
·
CH3CO
CH3 + CO.
Thermal decomposition of OP esters (ethers)
apparently occurs by the mechanism of concerted
cleavage of two bonds, peroxy and ester (ether) bonds,
with the formation of tert-butoxy and acetyl radicals and
of the corresponding carboxylic acids [11, 12]
The use of these initiators in radical processes is
restricted by the lack of technology for their production.
The synthesis methods described in the literature
[6, 9, 13] are preparative and multistep; the synthesis
is complicated by high reactivity and instability of the
starting compounds, intermediates, and target products.
CH3)3COOCH(CH3)OC(O)R
·
·
(CH3)3CO + CH3CO + RC(O)OH
From the technological viewpoint, the most promising
route to these initiators is the reaction of α-chloroalkyl
or alcohols
1890

SYNTHESIS OF α-HYDROXYALKYL PEROXIDE ESTERS AND ETHERS
1891
peroxides (CPs) with alcohols and complexes of
trialkylamines with carboxylic acids. However, CPs
themselves, prone to diverse chemical transformations,
are studied insufficiently. This particularly concerns
methods of their synthesis.
can “choose” the pathways of further transformations,
reacting either with HCl or with hydroperoxide (HP).
Synthesis of a CPunder definite conditions always yields
DP as by-product along with the target product.
Analysis of scheme (2) from the viewpoint of
optimization of the CPyield and selectivity allows making
apparent assumptions how to optimize the process. First,
it is necessary to take into account the nucleophilic
reactivity of HCl and HP. If their concentrations will be
comparable, there will be virtually no competition for
the carbocation because of high nucleophilicity of HP.
Therefore, it seems necessary and justified to ensure
excess of HCl by rapid saturation of the reaction mixture
with it. Second, to reduce the HP concentration, it is
necessary to ensure conditions for shifting the equilibrium
toward OP formation. For this purpose, the aldehyde
should also be taken in excess. The reagents are taken in
excess also because of the fact that the certain fraction of
the aldehyde and HCl inevitably passes into the aqueous
layer forming a separate phase with the progress of the
reaction.
This study was aimed at finding experimental
conditions allowing control of the reactions of CP
synthesis and transformation into OP esters and ethers.
SYNTHESIS OF CPs
It is known [14, 15] that CPs are synthesized by the
replacement of hemiacetal hydroxyl in OP by chlorine
under the action of excess HCl in the presence of a solvent.
Because OPs dissociate into the starting compounds on
heating, they are hydrochlorinated without isolation from
the reaction mixture:
(CH3)3COOH + RCHO + HCl
(CH3)3COOCH(R)Cl + H2O,
(1)
where R = Me, Et, Pr, i-Pr.
Thus, to increase the yield and selectivity with respect
to CP, it is necessary to take the aldehyde and HCl in
excess relative to HP. However, this apparent advantage
in the step of the CP synthesis transforms into a serious
disadvantage in the step of target product isolation from
the reaction mixture. Therefore, the excess of these
reagents should be minimum necessary. To find the
optimum reactant ratio in reaction (2), we performed a
special study and determined the optimum HP: aldehyde :
HCl molar ratio to be 1 : (1.01–1.05) : (1.05–1.30),
respectively. The use of the reactants in this ratio allows
CP to be obtained in a yield as high as 97% based on HP,
with no DP by-product found in the reaction mixture.
Reaction (1) is not selective. Along with CP, sym-
metrical di-tert-butylperoxy acetal (DP) is formed; it is
difficult to separate it from the target peroxide because
of their close boiling points:
2(CH3)3COOH + RCHO
+
+
H
H
(CH3)3COOCH(R)OOC(CH3)3 + H2O.
Synthesis of CPs via OPs. Synthesis of CPs in
the presence of a solvent seems promising. This fact
stimulated us to study in more detail reaction (1) as the
basis of the future industrial process. Synthesis of CPs by
reaction (1) can be described by a more detailed scheme:
The experimental substantiation of the above reactant
ratio was proved by experiments on studying the effect
of the excess of acetaldehyde and HCl on the yield of
α-chloroethyl tert-butyl peroxide (I). The yield of I was
estimated from the residual content of available oxygen,
AO, in the reaction mixture after its decomposition with
triethylamine under the conditions recommended in [16].
We found that the yield of I was virtually independent of
the excess of acetaldehyde and reached 95 and 97% with
the 5 and 20% molar excess, respectively, although the
accumulation of I at 20% excess occurred faster.
H+
H2O
HCl
HCl
(2)
(СH3)3СOOH
It evidently follows from scheme (2) that the reaction
is complex and that an OP, transforming in acid solution
into a carbocation, is necessarily formed as intermediate.
Specifically the reactions of the carbocation determine the
selectivity of the whole process, because this substrate
The curves of HCl accumulation in the reaction
mixture, obtained by titration of hydrolyzable chlorine
with a 0.5 N alcoholic KOH solution, are shown in
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

1892
ALEINIKOVA, NAVROTSKII
m, mol
m, mol
τ, min
τ, min
Fig. 1. Effect of excess acetaldehyde on the accumulation m
of (1, 2) I and (3, 4) HCl in the reaction mixture. (τ) Time; the
same for Fig. 2. Molar excess of acetaldehyde, %: (1, 3) 5 and
(2, 4) 20.
Fig. 2. Influence of the HCl feeding rate on the accumulation m
of (2, 4, 6) HCl and (1, 3, 5) I in the reaction mixture. Feeding
rate, mL min–1: (1, 2) 0.3, (3, 4) 2.9, and (5, 6) 5.2.
Fig. 1. As can be seen, the reaction of HCl with the
intermediate OP starts without induction period, which
confirms the carbocation formation in reaction (2). Also,
it can be noted that the HCl concentration in the reaction
mixture is higher than the concentration of I and that this
difference increases with time as the reaction products
are accumulated.
distillation of the reaction mixture with the preliminary
removal of the solvent. However, vacuum distillation is
insufficiently efficient because of inevitable losses and
is practically unfeasible because of thermal instability of
CPs. Therefore, we chose another approach consisting in
stabilization of CP by binding excess HCl. The choice
of stabilizers for CP is restricted by its high reactivity.
Amines, alkali and alkaline-earth metal hydroxides,
and alkali metal carbonates and hydrocarbonates are
unsuitable for this purpose, because they react with CP
with the cleavage of the О–О bond [15]. Urea and calcium
carbonate, which do not react with CP, were suggested as
stabilizers. As we found, unstabilized solutions of I can
be stored at 8–10°С for no more than 1 h, after which the
compound starts to decompose via cleavage of the О–О
bond (Table 1), and in 5 h the compound completely
decomposes with tarring of the reaction mixture.
Storage of solutions of I after preliminary treatment with
urea or calcium carbonate does not lead to noticeable
decomposition of CP, even after storage for 24 h at room
temperature. The AO content in a chloroform solution
of I was determined by the procedure described in [16].
The influence of the HCl feeding rate on the yield
of I was studied at the initial amounts (mole) of HP and
acetaldehyde of 0.3 and 0.315, respectively (Fig. 2). As
seen from Fig. 2, initially the rate of accumulation of I
is essentially the same as the rate of feeding HCl, but at
deeper conversion the accumulation of I in the reaction
mixture noticeably decelerates. At lower rate of feeding
HCl, the deceleration is observed earlier. For example,
in the curves of accumulation of I at HCl feeding rates
of 0.3 and 2.9 mL min–1, the deceleration is apparent and
corresponds to 60 and 95% yield of I, i.e., the formation
of I directly depends on the rate of feeding HCl. The
highest yield of I (97%) is observed when HCl is fed at
a rate of 2.9 mL min–1.
Isolation of CPs from the reaction mixture is a
separate problem. First, these compounds are thermally
unstable; second, the reagents present in excess are
capable not only to catalyze the CP decomposition, but
also to react with agents used in further work-up. The
classical approach to solution of this problem is vacuum
After treatment of the reaction mixture with urea or
calcium carbonate, taken in a molar ratio of (0.05–0.1) : 1
to CP, it is unnecessary to isolate CP by distillation, and
the stabilized solution can be used for the synthesis of,
e.g., OPethers [6, 9, 10, 17] and esters. Calcium carbonate
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

SYNTHESIS OF α-HYDROXYALKYL PEROXIDE ESTERS AND ETHERS
1893
is preferable as stabilizer, because it does not require
Table 1. Stabilization of chloroform solutions of I
AO content of I, %
preliminary treatment before use and, in contrast to
urea, is nonhygroscopic, which is particularly important,
because CPs are readily hydrolyzed in the presence of
moisture, i.e., treatment of CP solutions with urea or
calcium carbonate allows their reliable storage for 24 h.
stabilized solution of I
Storage
time, h
at 20°C
unstabilized
solution of I at
with
8–10°C
with urea
calcium
The temperature at which the CP synthesis is
performed is mainly determined by the stability of the С–
Сl bond, which, in turn, is determined by the structure of
the radical in the α-position to the О–О bond, i.e., by the
structure of the starting aldehyde. For aliphatic aldehydes,
the synthesis temperature should not exceed 5–10°С.
carbonate
0
1
6.45
6.30
6.26
6.18
6.06
6.50
6.45
6.40
6.42
6.30
6.70
6.12
4.25
2
Thus, to prepare CP, it is necessary to take the aldehyde
and HCl in no less than 5% and no more than 20% molar
excess, respectively, relative to HP. Isolation of the target
products should be performed after treatment of the
reaction mixture with urea or calcium carbonate, followed
by solvent removal if necessary.
5
Vigorous evolution
of HCl, tarring
24
intermediate carbocations. This assumption is confirmed
in an experiment on using HP as reagent in reaction (4),
with no less than 94% yield of the corresponding CPs
(Table 2):
Thus, synthesis of CPs using aliphatic aldehydes
consists of five main steps: (1) cooling of an aldehyde
solution in chloroform to ~10°С, (2) addition of HPto the
aldehyde solution at a temperature no higher than 10°С,
(3) cooling of the reaction mixture to –10°С, (4) feeding
of HCl to the reaction mixture at a temperature no higher
than 5°С, and (5) isolation of the target product.
(CH3)3COOH + [RC+HCl]
[(CH3)3COOC+H(R)Cl]
(4)
(CH3)3COOCH(R)Cl + H+.
Because the solubility of HCl in the reaction mixture
is limited, initially it can be introduced in an amount of
≤30 wt % relative to the required amount. Further feeding
of HCl leads to its breakthrough and nonproductive
consumption. Therefore, the addition of HP should be
started only after introducing 30 wt % of HCl, with
simultaneous feeding of the remaining amount of HCl.
By so doing, it becomes possible, first, to increase the
HCl solubility in the reaction mixture, second, to shift
the equilibrium of reaction (3) toward formation of
chlorohydrin, and, third, to efficiently utilize the starting
reactants.
Synthesis of CPs via geminal chlorohydrins.
Geminal chlorohydrins are intermediates in reactions
of nucleophilic addition of Сl– to carbonyl compounds
under the conditions of acid catalysis and practically are
not used as reagents for organic synthesis because of their
extremely low stability:
H+
Cl
H+
(3)
Reaction (3) starts with addition of proton to the
С=О bond and yields a carbocation RHC+–OH, which
is attacked by the Cl– ion to form a chlorohydrin and
then a carbocation RHC+–Cl. Analyzing reaction (3),
we can conclude that a series of equilibrium processes
in reaction of an aldehyde with HCl inevitably results
in formation of two carbocations, with each of them
present in an equilibrium concentration. It is natural to
assume that introduction of stronger nucleophiles into
this system will favor their efficient interaction with
Table 2. Physicochemical characteristics and yields of
(CH3)3COOCHRCl
AO, %
Yield, Tb, °C
(3 mmHg)
R
nD20
d420
%
found calcd.
Me(I) 97.0
25–26 1.4202 0.9241 9.95 10.05
Et
94.5
94.0
95.4
34–36 1.4221 0.9720 9.60
42–43 1.4293 0.9660 8.82
45–46 1.4224 0.9656 8.84
9.65
8.90
8.90
i-Pr
Pr
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

1894
ALEINIKOVA, NAVROTSKII
After adding HP, one can expect parallel occurrence of
aldehyde : HCl : HPmolar ratio should be 1 : (1.2–1.3) : 1,
respectively.
several reactions, because formally the reaction mixture
contains aldehyde, HCl, and carbocations. However, this
is not the case. If HP reaction (1) between HP and the
aldehyde took place, under the conditions of reaction (4)
the major product would be DP, but actually it is not
formed in noticeable amounts. Apparently, CP is mainly
formed by reaction (4), but also the parallel reaction of
HPwith the carbocation RHC+–OH [reaction (3)] cannot
be ruled out. This reaction yields OP, which undergoes
hydrochlorination under the conditions of reaction (4) to
form CP by reaction (2). Because both pathways yield
the same target product, it makes no sense to estimate the
contribution of the reaction of HP with the carbocation
RHC+–OH.
CP analysis. From the technological viewpoint, it
is more appropriate to use a CP without its isolation
from the reaction mixture. For this purpose, we
developed an analytical method for the determination
of CP concentration in a mixture with other peroxides,
which would allow more efficient use of CP and other
reagents for preparing functional dialkyl peroxides
and quantitative estimation of the DP impurity in CP
transformation products [16].
CPs are more reactive than DPs, primarily because
of the presence of the С–Сl bond. For example, CPs
are readily hydrolyzed with water with the release of
HCl whose content can be determined by classical
methods. However, it is impossible to calculate the
CP concentration from the HCl content, because the
reaction mixture always contains a certain excess of
HCl. Therefore, the choice of conditions for separate
determination of the concentrations of CP and DP in
their mixture is based on the reaction of CP with TEA,
involving all the three reaction sites (О–О, С–Н, С–Сl)
of CP [15, 18]. TEA reacts with CP with the cleavage of
the О–О bond, leaving the O–O bond in DP intact. The
reaction starts with the formation of a complex of CPwith
TEA via С–Сl bond:
It follows from the above analysis that HP does not
react with the aldehyde under these conditions, because by
the moment of HPappearance the effective concentration
of the reactive carbocations RHC+–Cl is already present
in the reaction mixture. It is quite obvious that the HP as
a “supernucleophile” will primarily react specifically with
this carbocation and shift the equilibrium in reaction (3)
toward CP formation.
Thus, we have shown for the first time that the reaction
of geminal chlorohydrins with HP can yield CP and
can be performed in four steps instead of five, with the
second and third steps practically coinciding in time:
(1) cooling of an aldehyde solution in chloroform to 5°С,
(2) feeding of 30 wt % of HCl into the aldehyde solution
at a temperature no higher than 10°С, (3) gradual feeding
of HP and the remaining HCl at a temperature no higher
than 15°С, and (4) isolation of the target product.
(CH3)3COOCH(R)Cl + N
[(CH3)3COOCH(R)N ]+Cl–,
(5)
After that, the amine hydrochloride is eliminated with
the cleavage of the С–Н bond:
In CPsynthesis by reaction (4), the process monitoring
is simplified. The maximal heat release is observed in the
second step, when the reaction mixture contains no peroxy
compounds and, therefore, there are no problems with
their possible decomposition. When performing synthesis
by reaction (1), the reaction of HP with an aldehyde in
the second step is accompanied by strong heat release
(142.7 kJ mol–1) and requires additional cooling to avoid
acid-catalytic and thermal decomposition of peroxides.
Therefore, it becomes possible in the third step to increase
the temperature to 15°С and to perform the process more
intensely than via reaction (1).
[(CH3)3COOCH(R)N ]+Cl–
:
N·HCl + [(CH3)3COO(R)C ].
The tert-butylperoxycarbene arising in the process is
unstable and transforms into nonperoxy compounds via
cleavage of the О–О bond:
(6)
where MHBC is a mixture of high-boiling components.
In CP synthesis via chlorohydrin, it becomes
unnecessary to use excess aldehyde, but excess HCl is
necessary for shifting the equilibrium in reaction (3)
toward formation of the RHC+–Cl carbocation. The
The iodometric analysis of the DP (prepared by the
method of [19]), performed before and after its treatment
with TEA for 1 h, showed that the AO concentration did
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

SYNTHESIS OF α-HYDROXYALKYL PEROXIDE ESTERS AND ETHERS
1895
where R = CH3, H; R' = CH3, C2H5, C3H7, C7H15–C9H19,
C6H5, CF3, CF3(CF2)n; M = K, Na.
not change noticeably, e.g., DP did not decompose in
the course of the analysis. Thus, after the treatment of
a mixture of peroxides with TEA, only one peroxide,
DP, remains in the reaction mixture, so its concentration
can be determined by iodometric titration, and the CP
concentration can be calculated as the difference between
the total content of peroxides in the mixture and the DP
content.
The reaction of CPs with sodium or potassium salts of
carboxylic acids, which are insoluble in organic solvents,
occurs at the interface at 40–50°С for 1.5–3.0 h with the
target product yield of 41–60%.Apparently, to accelerate
the process, it should be performed under homogeneous
conditions. To this end, we searched for an acylating
agent soluble in organic solvents. It could be expected
that triethylammonium complexes of carboxylic acids
would be such agents. However, actually acylation with
triethylammonium complexes occurred unselectively
and was accompanied by the cleavage of the O–O bond
in the CP. Iodometric analysis showed that the total
content of AO in the reaction mixture by the end of the
synthesis decreased by 35–40% relative to the initial
content. For example, α-pelargonyloxyethyl tert-butyl
peroxide (II) was obtained in 60% yield. The following
by-products were detected in the reaction products
by gas–liquid chromatography (mole per mole of I):
tert-butyl acetate (0.23), TB (0.11), pelargonic acid
(0.35), and TEA hydrochloride (0.78). The formation
of tert-butyl acetate and TB and a decrease in AO in the
reaction mixture indicate that compound I decomposes
under the action of TEA via intermediate carbene
formation by reactions (5) and (6). The lack of balance
with respect to TEA hydrochloride suggests that, along
with the above-indicated decomposition pathway of
I, there also can be another pathway yielding unstable
dimethylvinylamine whose existence was proved by
spectroscopy in [21]. Thus, the results of the experiment
suggest the presence of free TEAin its equimolar mixtures
with monocarboxylic acids. To prove this hypothesis, we
mixed equimolar amounts of acetic acid (AA, рKа 4.75)
and TEA, after which we added an equimolar amount
of the weaker pelargonic acid (рKа 4.90) and acylated
I with the resulting mixture. The reaction products
contained α-acetyloxyethyl tert-butyl peroxide III and
compound II in a molar ratio of 1 : 0.36 (Rf 0.6 and 0.75,
respectively). The AO content in the mixture decreased
by 23%. Similar result was obtained when adding AA
under the same conditions. When AA was taken in a
twofold excess relative to TEA, the yield of III increased
by 15% (Table 3). When the acylation was performed in
AA medium (AA–TEA ratio 57 : 1), the AO content in
the reaction mixture after the end of the reaction did not
change noticeably. Thus, complete binding of TEA is
possible only in the presence of a large excess of the acid.
It should be noted that TEAshould be dried for the use
in the analysis, because the CPhydrolysis is accompanied
by the release of HP [15], which is decomposed by TEA
more slowly, distorting the analysis results. The weight
ratio of TEA and the mixture being analyzed should be
0.7 : 1.0, respectively. DPis a difficultly reducible peroxy
compound; therefore, the procedure for quantitative
determination ofAO in it involves more severe conditions
than in the analysis of OPor CP.As found in [20], peroxy
acetals are quantitatively reduced in 15 min at 70°С in
the presence of hydroiodic acid generated in situ by the
reaction of KI with a 40% HBr solution.
The decomposition time for I is 25 min at 25°С; in
chloroform solution, which is used most frequently in
the CP synthesis, it completely decomposes in 20 min.
In hexane, this reaction occurs in 2 h. Therefore, when
CPs are analyzed in hydrocarbon solvents, the solutions
should be diluted with an equal amount of chloroform
prior to adding TEA.
Thus, the method developed allows sufficiently
accurate determination of the CP and DP concentrations
in their mixture. This method allows monitoring of the
CP concentration, opening the possibility of using CPs
as starting compounds for preparing functional dialkyl
peroxides without preliminary isolation of CPs from the
reaction mixture and of appreciably simplifying the CP
work-up technology by reducing the number of process
steps.
SYNTHESIS OF OP ESTERS
AND ETHERS
Synthesis of OP esters. OP esters are prepared by
OP acylation with acid anhydrides or chlorides [13] or
by alkylation of carboxylic acid salts with CPs. The latter
procedure is the most promising, because it allows CPs to
be used without their isolation from the reaction mixture:
R'C(O)OM + ClCH(R)OOC(CH3)3
R'C(O)OCH(R)OOC(CH3)3 + MCl,
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

1896
ALEINIKOVA, NAVROTSKII
HN+R3
A + HN+R3.
Table 3. Effect of solvent and AA : TEA ratio on the time of
acylation of I and on the yield of III at 20°С
AH + NR3
AH NR3
SH NR3
A
A
(7)
AA : TEA Time, Yield,
Solvent
Hexane
ε25
Pj
Ej
ratio
min
%
From the viewpoint of acylation, the conditions
favoring the formation of ionic complexes, up to free
ions, are the most favorable.
1.9
0
1 : 1
300
63.0
2 : 1
1 : 1
2 : 1
1 : 1
1 : 1
2 : 1
60
4
72.0
60.0
75.0
35.0
41.0
70.0
The equilibrium in reaction (7) largely depends on the
solvent [22]. In proton-donor solvents, the equilibrium is
shifted toward the ion pair owing to hydrogen bonding
of the solvent (S–H) with the carbonyl group of the
molecular complex. Therefore, the acylation of CP with
the ionic complex occurs within 3–5 min. The yields
of product III in different proton-donor media are
essentially different. For example, in water the process
is complicated by the hydrolysis of CP; therefore, the
yield does not exceed 35%. In TB, the yield of III was
also low. However, TB does not react with I, which is
confirmed by the absence of reaction products different
from those obtained by the acylation in chloroform.
This is apparently due to the competition of TB with
AA for TEA, which probably leads to the formation of
species of more complex composition than in chloroform.
An increase in the yield of III in TB to 70% suggests
that the twofold excess of AA virtually suppresses the
competition of TB for TEA, and the amount of the free
amine decreases.
Chloroform 4.7 0.35
4
4
Water
TB
78.5 0.55
12.2 0.58
4
5
AA
6.2 1.2
20.0
57 : 1
1 : 1
2 : 1
1 : 1
3
90
60
5
85.0
62.0
75.0
80.0
Acetone
0.93
1.17
DMF
36.7
However, product III was isolated in only 85% yield,
which may be due to its loss in the course of isolation from
the acetic acid solution. The experimental data obtained
are given in Table 3. As solvent characteristics we used
the acidity factor P (reference phenol, Рi = 1) or basicity
factor Е (standard diethyl ether, Еj =1) [22].
In inert solvents, the equilibrium in reaction (7)
is shifted toward formation of molecular complexes;
therefore, the acylation of I in hexane occurs within
5 h. However, the use of a twofold excess of AA allows
the reaction time to be decreased by a factor of 5. This
is apparently due to the shift of the equilibrium in
reaction (7) toward formation of the ionic complex, so
that the yield of III increases.
It is known [23, 24] that the capability of acid
molecules for self-association leads to the formation,
along with the complexes RСООН·NR3, of more stable
higher complexes (RСООН)n·NR3. Therefore, a certain
amount of the amine remains unbound in equimolar
mixtures of a carboxylic acid with TEA, which leads to
the side reaction of CP decomposition and to a decrease
in the yield of the target product.
The acylation time of I in proton-acceptor solvents is
different. For example, in DMF the process is complete
in 5 min, whereas in acetone it lasts for 1.5 h. Similarly
to the reaction in hexane, with a twofold excess of AA
in acetone the reaction occurs faster (in 1 h). The highest
yield of III was obtained in DMF, because this solvent
creates the most favorable conditions for the formation of
the complexes СН3СООН·N(С2Н5)3 and for a decrease in
the content of bound TEA in the reaction mixture.
The experimental data (Table 3) demonstrate a
considerable effect of solvents on the time and selectivity
of the acylation. For example, in hexane the reaction takes
5 h, whereas in polar solvents it is complete in 4 min.
This large difference in the reaction times is probably
determined by the reactivity of both reactants.
The formation of the structure of the complex is
determined by a number of factors, primarily by the
proton-donor properties of the acids, proton-acceptor
properties of the amines, and solvating power of the
solvent. By varying these properties, it is possible to
control reaction (7):
The polarity of the medium favors ionization not
only of the acylating complex [reaction (7)], but also of
CP [reaction (8)], which is solvated with chloroform by
hydrogen bonding with the leaving group:
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

SYNTHESIS OF α-HYDROXYALKYL PEROXIDE ESTERS AND ETHERS
1897
Table 4. Physicochemical characteristics and yields of OP esters (CH3)3COOCH(СН3)ОC(О)R
MM
AO, %
calculated
R
Rfa
nD20
d420
Yield, %
found calculated
found
8.76
CH3
0.60
0.74
0.64
0.62
0.67
0.75
0.70
0.71
0.60
174
215
202
200
227
268
402
374
224
176
218
204
204
232
274
400
398
238
9.09
7.34
7.84
7.84
6.89
5.83
4.00
4.02
6.72
0.9402
0.9125
0.9920
0.9324
0.9386
0.9050
0.8991
0.9007
1.0402
1.4080
1.4109
1.4090
1.4115
1.4200
1.4250
1.4395
1.4490
1.4783
95
95
94
95
95
92
85
85
90
C(CH3)3
CH(CH3)3
(CH2)2CH3
(CH2)4CH3
(CH2)7CH3
C17H35
7.53
7.88
7.80
6.89
5.80
3.90
3.95
6.60
C17H33
C6H5
C10H15
(adamantly)
0.69
293
296
5.40
5.41
1.0252
1.4760
86
a Silufol UV-254, eluent hexane–diethyl ether (15 : 5 by volume), visualization with iodine vapor.
(CH3)3COOCH(R)Cl HCCl3
[(CH3)3COOC+HR Cl– HCCl3].
is not fully excluded in this case. Hence, none of the two
approaches can ensure successful acylation of CP.
(8)
In our studies we found the conditions for the
formation of the 1 : 1 complex owing to the artificially
created excess of the amine by dosing the acid into a
chloroform solution of TEA (1 M) in the adiabatic mode
at such a rate that the temperature of the reaction mixture
changed by no more than 5°C. Under these conditions,
the yields of OP esters in chloroform increased to 95%
(Table 4). Thus, we suggested for the first time the use
of a new acylating agent in the form of a complex of a
carboxylic acid with a tertiary amine, with chloroform
as a solvent. We were able to perform the CP acylation
within 4 min under adiabatic conditions at equimolar
ratio of the reactants.
In the process, the negative charge is concentrated
on the leaving group, and the strength of such hydrogen
bonds in the activated state is considerably higher than in
the initial state. Such specific solvation leads to additional
stabilization of the activated state of CP in proton-
containing media, increasing the degree of its ionization
and decreasing the acylation time. In chloroform, AA–
pyridine complexes exist in the molecular form [22] but
acylate CP within 5 min; therefore, it can be stated that
the acylation time is mainly determined by the extent of
CP ionization, rather than by the structure (molecular or
ionic) of the acylating complex.
A study of the regular trends in CP acylation with
carboxylic acid–TEA complexes has shown that the
composition of the acylating complex determines mainly
the reaction selectivity, whereas the solvent polarity
determines the acylation time.
Thus, we found that the CP acylation should be
performed at an equimolar ratio of CP and acylating
complex in a chloroform solution, that the acylating
complex should be prepared under adiabatic conditions
by dosing the acid to TEA in a molar ratio of 1 : 1 in
chloroform, and that the reaction progress should be
monitored by variation of pH of the reaction mixture
(the reaction is considered to be complete at pH 6.0–7.0).
The optimum conditions are those under which 1 : 1
complexes are formed. The conditions reducing the self-
association of acids and hence favoring the formation
of the complexes RСООН·NR3 are dilution and excess
amine [25]. The use of 9 × 10–3 M solutions of AA in
chloroform for preparing the acylating complex leads to
the preparation of III in 80% yield, i.e., the side reaction
of CP decomposition under the action of unbound TEA
The synthesized OP esters are colorless liquids
with ester odor, readily soluble in organic solvents and
insoluble in water. They do not change their properties
after 1-year storage. The structure of the synthesized
peroxides was proved by the analysis for AO and by
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

1898
ALEINIKOVA, NAVROTSKII
Table 5. Physicochemical characteristics and yields of OP ethers (CH3)3COOCH(СН3)ОR
MM
calculated found
AO, %
calculated
Tb, °C
(P, mmHg)
R
nD20
d420
Yield, %
found
180
CН2СН(CH3)2
C6H11-cyclo
(CH2)2CH3
CH3
30 (1)
60 (1)
–
190
216
176
148
8.45
7.50
8.98
8.84
8.42
7.41
9.09
9.09
0.9125
0.9199
0.8748
0.8782
1.4078
1.4370
1.4060
1.4085
94
94
95
90
210
178
31 (13)
150
molecular mass determination and was confirmed by the
IR and 1H NMR spectra. The IR spectra contain a strong
band at 1720–1740 cm–1, characteristic of С=О stretching
vibrations, and a band at 860–880 cm–1, characteristic of
O–O stretching vibrations. In the 1H NMR spectra, there
are the following characteristic signals (δ, ppm): 1.22 s
[9Н, (CH3)3С], 4.65 q, J 5.4 Hz (1Н, О–О–СН), 1.22 d,
J 11.7 Hz (3Н, О–О–СН–CH3); these signals are present
in the spectra of all the molecules. The 1H NMR spectra
of the synthesized OP esters differ in the proton signals
of the radical in the acyl moiety.
alkylation at 20–30°С within 10–30 min, to increase the
yield of the peroxides to 95% (Table 5), and to simplify
the isolation of OP ethers by eliminating the step of
reaction mixture neutralization and reducing the amount
of wastewater. The starting alcohols, compound I, and
calcium carbonate were taken in a molar ratio of 1 : 1 :
(0.51–0.58), respectively [10, 17]. The physicochemical
characteristics of the synthesized OP ethers are in
agreement with the published data.
Using various hydroperoxides as nucleophiles instead
of alcohols, we obtained symmetrical and unsymmetrical
diperoxides.
The new acylating agent, a 1 : 1 complex of a
carboxylic acid with a tertiary amine, is effective in
reactions not only with CPs, but also with chloro ethers,
benzyl chloride, allyl chloride, and alkyl halides in which
the chlorine atom is bound to a tertiary carbon atom.
EXPERIMENTAL
The IR absorption spectra of OPethers and esters were
recorded in the range 500–4000 cm–1 with a Specord UV-
VIS spectrophotometer from thin films, and the 1H NMR
spectra, with a Varian Mercury Plus-300 spectrometer
operating at 300 MHz, with HMDS as internal reference.
The content of AO in CPs was determined by the
standard procedure, and that in ethers and esters, by the
procedure described in [20]. Carboxylic acids, alcohols,
and chloroform were purified and dried by the standard
procedures [26].
Synthesis of OP ethers. Alcoholysis of CPs occurs
at 30–70°С within 2.5–6.0 h in an inert solvent with
continuous removal of HCl with a stream of dry nitrogen.
The yield of the target peroxides is 53–75% [6, 9]:
ROOCH(R')Cl + HOR''
ROOCH(R')OR'' + HCl,
R = CH3, C(CH3)3; R' = H, CH3, C2H5; R'' = CH2CH2CH3,
CH2CH(CH3)2, cyclo-C6H11, CH2CH(NO2)CH3,
CH2(CH2OCH2)2CH2OH, CH2CH(OH)CH2OH,
–[CH2–CH–]n–.
Synthesis of I and III. A reactor equipped with a
thermometer and a gas-feeding tube with a porous plate
was charged with 9.2 g (0.21 mol) of acetaldehyde in
50 mL of chloroform, after which 18 g (0.2 mol) of HP
was added at a temperature not exceeding 10°С. The
reaction mixture was cooled to –10°С, and a flow of
8.0 g (0.22 mol) of dry HCl was passed. In the process,
the temperature no higher than 5°С was maintained. The
organic payer was separated from the aqueous layer and
treated with 1 g (0.01 mol) of calcium carbonate.
Relatively low yields of OP ethers are due to the
presence of HCl in the reaction mixture, because HCl,
being partially soluble in chloroform, is removed with
an inert gas stream incompletely, and also to elevated
temperature and long time of the reaction, favoring
catalytic decomposition of the starting CP. Therefore, as
acceptor of the released HCl we took calcium carbonate
used for stabilization of chloroform solutions of CPs.
With calcium carbonate, we were able to perform the CP
The resulting chloroform solution of I was filtered
and mixed with 32.2 g (0.2 mol) of the complex of
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

SYNTHESIS OF α-HYDROXYALKYL PEROXIDE ESTERS AND ETHERS
1899
triethylamine with acetic acid in 100 mL of chloroform.
CONCLUSIONS
After the lapse of 5 min, the reaction mixture was washed
with water, a 2% KOH solution, and again with water.
After removing the solvent, 32.9 g (92%) of the product
was obtained, nD20 1.4078, AO 8.79%.
(1) The reaction of hydroperoxides with aldehydes
and hydrogen chloride as a route to α-chloroalkyl
peroxides was studied. Conditions were found for
preparing α-chloroalkyl peroxides via chlorohydrin;
this intermediate can be used without isolation from the
reaction mixture.
Synthesis of α-chloropropyl tert-butyl peroxide
and α-acetyloxypropyl tert-butyl peroxide. A reactor
equipped with a dropping funnel, a thermometer, and
a gas-feeding tube with a porous plate was charged
with 11.62 g (0.2 mol) of propionaldehyde in 50 mL of
chloroform. The solution was cooled to 10°С, and 2.6 g of
dry HCl was passed.After that, at a temperature no higher
than 10°С, 18 g (0.2 mol) of HPwas added dropwise, with
simultaneous feeding of the residual amount (6.2 g) of dry
HCl in the course of 18–20 min. The total amount of dry
HCl was 8.3 g (0.24 mol). After that, the organic layer
was separated, dried over calcium chloride, and analyzed
for the CP content. Yield 31.6 g (95% based on HP). The
chloroform solution of CP was filtered and then mixed
with 32.2 g (0.2 mol) of the complex of triethylamine
with acetic acid, dissolved in 100 mL of chloroform,
at 30°С. After the lapse of 5 min, the reaction mixture
was washed with water. The solvent was removed, and
α-acetyloxypropyl tert-butyl peroxide was obtained; yield
32 g (91% based on HP), nD20 1.4120, AO 8.39%.
(2) A new procedure was developed for preparing
α-hydroxyalkyl peroxide esters from chloroalkyl
peroxides using a new acylating agent, a complex of
a carboxylic acid with triethylamine. Conditions were
found for the formation of the 1 : 1 complex. The acylation
selectivity in solvents of different polarity is determined
by the composition of the complex, and the reaction time,
by the extent of ionization of α-chloroalkyl peroxides.
REFERENCES
1. Ivanchev, S.S., Radikal’naya polimerizatsiya (Radical
Polymerization), Leningrad: Khimiya, 1985.
2. Antonovskii, V.L. and Khursan, S.L., Fizicheskaya khimiya
organicheskikh peroksidov (Physical Chemistry of Organic
Peroxides), Moscow: Akademkniga, 2003.
3. RF Patent 2246487, Publ. 2005.
4. Kurdyukov, A.M., Izv. Vyssh. Uchebn. Zaved., Ser.: Khim.
Khim. Tekhnol., 2007, vol. 50, no. 8, pp. 97–104.
Synthesis of I. The compound was prepared similarly
from 8.8 g (0.2 mol) of acetaldehyde. Yield of I, according
to analysis results, 29.3 g (96% based on HP). The solvent
was distilled off, and the reaction mixture was distilled
with the collection of the fraction boiling at 40–42°С
(12 mmHg). Yield of I 24.4 g (80%), nD20 1.4190, AO
10.05%.
5. Kudryavtseva, T.V., Ivanchev, S.S., Kirillova, E.I., et al.,
Plast. Massy, 1977, no. 9, p. 45.
6. Kovalets, I.Yu.,Aleinikova, T.P., and Derbisher, V.E., Russ.
J. Appl. Chem., 2002, vol. 75, no. 8, pp. 1371–1372).
7. RF Patent 2298054, Publ. 2007.
8. RF Patent 2295598, Publ. 2007.
Synthesis of α-butoxyethyl tert-butyl peroxide.
A reactor equipped with a stirrer, a thermometer, and a
dropping funnel was charged with 4.32 g (0.0432 mol)
of calcium carbonate, a solution of 5.9 g (0.08 mol) of
butanol in 12 mLof chloroform was added, and a solution
of 12.2 g (0.08 mol) of I in 25 mLof chloroform was added
from the dropping funnel with stirring. The mixture was
stirred at 20–30°С for 30 min until pH 7.0 was attained,
after which the calcium chloride precipitate was filtered
off, the filtrate was washed with water and dried over
calcined magnesium sulfate, the chloroform was distilled
off in a water-jet-pump vacuum, and the residual solvent
was removed by evacuation at 30–40°С/100 mmHg.
Yield of α-butoxyethyl tert-butyl peroxide 14.4 g (95%),
nD20 1.4056, d420 0.9174, AO 8.50%.
9. Vasil’eva, V.D., Derbisher, V.E., Kovalets, I.Yu., et al.,
Russ. J. Appl. Chem., 2007, vol. 80, no. 8, pp. 1409–1412.
10. Aleinikova, Z.S., Khardina, I.A., Aleinikova, T.P., and
Vasil’eva, V.D., Izv. Vyssh. Uchebn. Zaved., Ser.: Tekhnol.
Legk. Prom–sti., 2011, vol. 12, no. 2, pp. 38–40.
11. Turovskii, A.A., Navrotskii, V.A., Turovskii, N.A., et al.,
Zh. Obshch. Khim., 1985, vol. 55, no. 1, pp. 173–176.
12. Kushch, O.V., Opeida, I.A., Turovskii, N.A., and
Navrotskii, V.A., Russ. J. Phys. Chem., 2005, vol. 79, no. 8,
pp. 1238–1241.
13. Kozlov, E.M., Shreibert, A.I., Khardin, A.P., and
Ermachenko, V.I., Zh. Org. Khim., 1969, vol. 5, p. 427.
14. Khardin,A.P., Shreibert,A.I., and Kokhan, L.M., Zh. Org.
Khim., 1968, vol. 4. no. 4, pp. 719–720.
15. Zaburdyaeva, S.N., Dodonov, V.A., and Razuvaev, G.A.,
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014

1900
ALEINIKOVA, NAVROTSKII
22. Epshtein, L.M., Russ. Chem. Rev., 1979, vol. 48, no. 8,
Zh. Org. Khim., 1972, vol. 8, no. 3, pp. 552–555.
pp. 655–657.
16. Aleinikova, T.P., Brailovskaya, A.A., Kargina, L.V., et al.,
Zh. Analit. Khim., 1984, vol. 39, no. 6, pp. 1120–1123.
23. Gur’yanova, E.N., Gol’dshtein, I.P., and Perepelkova, T.I.,
Russ. Chem. Rev., 1976, vol. 45, no. 9, p. 792.
17. RF Patent 2284321, Publ. 2006.
24. Lipovskii,A.A. and Kuzina, M.G., Zh. Neorg. Khim., 1967,
18. Razuvaev, G.A., Zaburdyaeva, S.N., and Dodonov, V.A.,
vol. 12, p. 236.
Zh. Org. Khim., 1971, vol. 7, no. 11, pp. 2484–2486.
25. Gusakova, G.V., Denisov, G.S., and Smolyanskii, A.L.,
19. Navrotskii, V.A., Pil’dus, I.E., and Shreibert,A.I., Zh. Org.
Khim., 1973, vol. 9, no. 11, pp. 2422–2423.
Zh. Prikl. Spektrosk., 1972, no. 16, p. 503.
26. Becker, H., Berger, W., Domschke, G., et al., Organikum.
Organisch-chemisches Grundpraktikum, Berlin:
Wissenschaften, 1976.
20. RF Patent 2296989, Publ. 2007.
21. Markaryan, A.Sh., Zh. Org. Khim., 1982, vol. 18, no. 2,
p. 252.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 87 No. 12 2014