Kinetics of thermoinitiated oligomerization of 3,3-(2,4,6-triethyl-1,3- phenylene)bis(5-methyl-1,2,4-oxadiazole)

Article abstract of DOI:10.1007/s11172-013-0206-y

The crystal structure of 3,3-(2,4,6-triethyl-1,3-phenylene)bis(5-methyl-1, 2,4-oxadiazole), which models the cross-linking fragment of polynitriles by benzodinitrile oxides, was deter- mined by X-ray diffraction analysis. The kinetic regularities of its t

Full text of DOI:10.1007/s11172-013-0206-y

Russian Chemical Bulletin, International Edition, Vol. 62, No. 6, pp. 1434—1441, June, 2013
1434
Kinetics of thermoinitiated oligomerization
of 3,3´ꢀ(2,4,6ꢀtriethylꢀ1,3ꢀphenylene)bis(5ꢀmethylꢀ1,2,4ꢀoxadiazole)
T. K. Goncharov, A. I. Kazakov, Z. G. Aliev, G. V. Lagodzinskaya, E. L. Ignat´eva, and L. S. Kurochkina
Institute of Problems of Chemical Physics, Russian Academy of Sciences,
1 prosp. Akad. Semenova, 142432 Chernogolovka, Moscow Region, Russian Federation.
Fax: +7 (496) 522 3507. Eꢀmail: tel@icp.ac.ru
The crystal structure of 3,3´ꢀ(2,4,6ꢀtriethylꢀ1,3ꢀphenylene)bis(5ꢀmethylꢀ1,2,4ꢀoxadiazole),
which models the crossꢀlinking fragment of polynitriles by benzodinitrile oxides, was deterꢀ
mined by Xꢀray diffraction analysis. The kinetic regularities of its thermoinitiated oligomerizaꢀ
tion were studied by isothermal calorimetry and spectrophotometry. The process proceeds with
selfꢀacceleration, and the Arrhenius dependences of the initial and maximum reaction rates
were determined. The reaction accelaration is determined by an increase in the viscosity of the
reaction medium. Possible mechanisms were proposed for oligomerization.
Key words: 3,3´ꢀ(2,4,6ꢀtriethylꢀ1,3ꢀphenylene)bis(5ꢀmethylꢀ1,2,4ꢀoxadiazole), Xꢀray difꢀ
fraction analysis, kinetics, thermoinitiated oligomerization, dynamic calotimetry, NMR spectroꢀ
scopy, mass spectrometry.
The method of lowꢀtemperature curing of rubbers with
in the reaction of benzodinitrile oxide with polymers conꢀ
taining nitrile groups.
dinitrile oxides1 was further developed2—4 after sterically
shielded benzodinitrile oxides of the (R)3—C6H—(CNO)2
type stable at room temperature were synthesized. The
wide use of this efficient method of curing depends on the
thermal stability of the isoxazoline ring formed as a fragꢀ
ment (FR) of transversal crossꢀlinking upon the addition
of dinitrile oxides to olefins.
Experimental
1
H and 13C NMR spectra were recorded on a Bruker
AVANCE 3500 instrument (working frequencies 500 and
125 MHz, respectively; solvent DMSOꢀd6 + CCl4 (25 vol.%);
internal standard Me4Si). The starting compound I (sample 1)
was studied. The products of its thermal transformation were
studied at different stages: after 68 h of heating at 150 C (sample 2)
and after 24 h of heating at 182 C (sample 3). Signal assignment
was based on the values of chemical shifts, shapes of multiplets,
comparison of integral intensities of the signals, and HSQC 13Сꢀ
1H experiments.
The mass spectrometric analysis of the thermal transformaꢀ
tion products of compound I was conducted on an LCSM 2020
instrument (Shimadzu, Japan) using electrospray with ionizaꢀ
tion of a solution of the reaction products in acetonitrile.
The kinetics of thermal transformation of compound I was
studied on a DAK automated differential microcalorimeter6 in
glass sealed ampules (V  2 cm3) having no cold part, which
made it possible to keep all transformation products in the reacꢀ
tion zone. If not specially indicated, a weighed sample of the
studied substance was 50 mg at temperatures in a range of
148.8—201.6 С.
Manometric measurements were carried out on a setup deꢀ
veloped and made at the Institute of Problems of Chemical
Physics (Russian Academy of Sciences), which allows one to
work under the pressure in the reaction vessel from 0 to 40 atm
and at temperatures from 20 to 350 С.
Earlier,5 when studying the thermal stability of the
compound obtained in the reaction of benzodinitrile oxide
with ethylene and modeling the fragment of transversal
crossꢀlinking, we found that the N—O and С—С bonds
are cleaved in the isoxaline cycle and the final products,
viz., dinitrile (R)3—C6H—(CN)2 and acetaldehyde, are
formed. The kinetic parameters were determined, and the
mechanism for the reaction of crossꢀlinkage decomposiꢀ
tion was proposed.
The purpose of this work is to study the kinetics of the
thermal transformation of 3,3´ꢀ(2,4,6ꢀtriethylꢀ1,3ꢀphenylꢀ
ene)bis(5ꢀmethylꢀ1,2,4ꢀoxadiazole) (I). Compound I
models the fragment of transversal crossꢀlinkage formed
The absorbance of solutions of the reactions products in
chloroform was measured at the wavelength 330 nm on a Specꢀ
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 6, pp. 1434—1441, June, 2013.
1066ꢀ5285/13/6206ꢀ1434 © 2013 Springer Science+Business Media, Inc.

Oligomerization of oxadiazole derivatives
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
1435
ordꢀ40 UV/VIS instrument. The absorption regions of solutions
of compound I and products of its thermal transformation in
chloroform did not coincide. The initial regions of the kinetic
curves of product accumulation were plotted using the ampule
method by the results of determination of the absorbance of
solutions of the products, and the reaction rate constants were
calculated. In each measurements, the concentration of the soꢀ
lution based on initial compound I was 3•10–4 mol L–1. This
method does not allow one to study the reaction at large converꢀ
sions, since products with differed molar absorption coefficients
appeared with an increase in the reaction conversions.
Synthesis of 3,3´ꢀ(2,4,6ꢀtriethylꢀ1,3ꢀphenylene)bis(5ꢀmethꢀ
ylꢀ1,2,4ꢀoxadiazole) (I). Acetonitrile (TU 6ꢀ093534ꢀ82) (20 mL)
was poured to 1.5 g of 2,4,6ꢀtriethylbenzeneꢀ1,3ꢀdinitrile
oxide (II) (m.p. 96—98 С). The solution was kept for 55 days in
a dark container at 20 С. The reaction course was monitored by
thin layer chromatography on Silufol UVꢀ254 plates using
chloroform as an eluent. The yield of unpurified compound I
was 1.7 g (85%). Purification was carried out by the chromatoꢀ
graphic method (column l = 500 mm, d = 20 mm; silica gel
0.035—0.07 mm, 60 Å, chloroform as an eluent). A white crysꢀ
talline substance was isolated, m.p. 97—98 С. 1H NMR
(DMSOꢀd6 + CCl4), : 0.874 (t, 3 Н, СН3СН2, 3J = 7.5 Hz);
1.100 (t, 6 Н, СН3СН2, J = 7.5 Hz); 2.183 (q, 2 Н, CH3СН2,
3J = 7.5 Hz); 2.413 (q, 4 Н, CH3СН2, 3J = 7.5 Hz); 2.686
(s, 6 Н, Me); 7.230 (1 Н, Ph). 13С NMR: 12.52 (CH3); 15.96
(СН3СН2(2)); 16.28 (СН3СН2(1)); 24.76 (CH3СН2(1)); 26.82
(CH3СН2(2)); 124.73, 143.66, 146.40 (Ph); 127,25 (CH, Ph);
167,50 (NCN); 177.59 (NCO).
Xꢀray diffraction analysis of compound I. Colorless crystals
well cut as flattened tetragonal prisms were grown from a soluꢀ
tion of compound I in acetone. The crystallographic data for
compound I are given in Table 1. Xꢀray diffraction analysis was
carried out on a КМꢀ4 automated fourꢀcircle difractometer
(KUMA DIFFRACTION) with ꢀgeometry in the / scan
mode using the monochromatized МоꢀK radiation. The strucꢀ
ture was solved by a direct method using the SIR92 program7
followed by a series of calculations of the electron density maps.
The positions of hydrogen atoms were geometrically specified
and refined in the riding model. The positions and temperature
parameters of nonꢀhydrogen atoms in the anisotropic approxiꢀ
mation were refined by fullꢀmatrix least squares using the
SHELXLꢀ97 program.8
3
The CIF file containing the full information on the studied
structure was deposited with the Cambridge Crystallographic
Data Centre (CCDC 859039, www.ccdc.cam.ac.uk/
data_request/cif).
Table 1. Crystallographic data and parameters of Xꢀray difꢀ
fraction experiment for compound I
Results and Discussion
Parameter
I
Empirical formula
Molecular weight
Color, habitus
Crystal system
Space group
Z
a/Å
b/Å
c/Å
/deg
C18H22N4O2
326.40
Colorless prisms
According to the Xꢀray diffraction data, the reaction
of compound II with acetonitrile affords compound I conꢀ
taining 1,2,4ꢀoxadiazole fragments, while no alternative
1,2,5ꢀoxadiazole structures are formed. The general view
of molecule I is shown in Fig. 1. The crystal is built of two
crystallographically independent and similar in structure
molecules. The bond lengths in them are the same within
inaccuracies and range within values usual for the correꢀ
sponding values. The molecular packing in crystals is loosꢀ
ened, which is also indicated by the density of the crystals
(see Table 1). The ethyl group at the С(6) carbon atom of
one of the crystallographically independent molecules has
abnormally high thermal vibration coefficients, and the
С(17) and С(18) atoms are fixed in two randomly equiꢀ
probable positions. The crystal contains no hydrogen bonds
and shortened intermolecular contacts.
Triclinic
–
P1
4
8.293(2)
12.926(3)
17.467(4)
93.97(3)
102.18(3)
93.59(3)
1820.1(6)
1.191
/deg
/deg
V/Å
dcalc/g cm–1
Crystal size/mm
/deg
0.8×0.5×0.2
2.06—25.00
–7  h  1, –15  k  15,
–20  l  20
6992
Reflection indices
Number of measured
reflections
Number of independent
reflections
Number of refined parameters
Goodnessꢀofꢀfit
Number of reflections
with I  2(I)
Final R factors
(for reflections with I  2(I)
R1
5742
Rint = 0.0166
452
C(14)
O(2)
C(12)
O(1)
N(3)
C(10)
N(1)
C(13)
C(7)
1.066
2893
C(9)
C(8)
C(11)
N(4)
N(2)
C(2)
C(1)
C(6)
C(3)
C(4)
C(5)
0.0819
0.2436
C(17)
C(15)
C(18)
wR2
C(16)
Residual electron density
/e A–3
0.537
Fig. 1. Molecular structure of compound I.

1436
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
Goncharov et al.
Irel
186
80000
70000
60000
50000
40000
30000
20000
10000
471
120
715
707
521
621
838
368
940
236
405
563
331
781
80
100
200
300
400
500
600
700
800
900
m/z
Fig. 2. Fragment of the mass spectrum of the oligomerization products of compound I for 170 С, 48 h.
The study of the kinetics of the thermal transformation
of a melt of compound I by the manometric method
in vacuo in the temperature range from 120 to 180 С
at different degrees of filling of the vessels (m/v =
= 0.2—0.8 g cm–3) revealed that no gaseous reaction prodꢀ
ucts are formed in the thermal transformation of the linkꢀ
age, unlike the thermal decomposition of the linkage
formed in the reaction of benzodinitrile oxide with ethylꢀ
ene.5 The studied substance was colored in experiments,
indicating the reaction occurrence. As the reaction conꢀ
version increased, the transformation products gained
a resinꢀlike consistence.
The mass spectrometric analysis of the reaction prodꢀ
ucts showed that the reaction afforded products with the
molecular weight up to 3000 atomic units (Fig. 2). The
presence of these products can be explained by their therꢀ
mally initiated oligomerization of crossꢀlinking fragments
formed in the reaction of benzodinitrile oxide with nitriles.
The 1Н and 13С NMR spectra of transformation prodꢀ
ucts 2 and 3 exhibit signals coinciding with the signals in
the spectrum of the starting compound I and also weak
lines shifted relatively to the corresponding lines of the
initial monomer (Fig. 3). Several signals are distinctly seen
near each signal that coincides with the signal from comꢀ
pound I. Their number and the total intensity for sample 3
are higher than those for sample 2. The lines remain narꢀ
row, indicating a low degree of polymerization. In the
1H NMR spectrum of sample 3, normalization to the sigꢀ
nal 1 Н of the overall surface area of the signals in the
phenyl region (6.8—7.5 ppm) gives 21 H for the overall
surface area of the alkyl signals (0.6—3.1 ppm), which is
totally 22 Н according to the empirical formula of comꢀ
pound I (C18H22N4O2). It follows from this that the transꢀ
formation does not involve the benzene ring.
For the phenyl protons, the overall surface area of the
changed signals in sample 3 is 45% and that of the unꢀ
changed signals is 55%. This is approximately* consistent
with the estimate for the Me groups: about 40% for the
changed groups ( 1.55—2.14) and about 60% for the unꢀ
changed groups ( 2.70 0.1). In the spectrum of sample 2
with a lower conversion, the singlet at  1.76 ( 13С 23)
has the highest intensity (~3%). The singlet is retained in
the spectrum of sample 3, but now the Me group at  2.12
(15%) prevails. Two other sufficiently intense methyl sigꢀ
nals at 2.05 and 1.76 ppm are observed (10% each). Weakꢀ
er signals are also present. The estimation by the СН3
triplet and СН2 quartet of the single ethyl group gives
46 and 48% for the unchanged signals. Taking into acꢀ
count that the measurement inaccuracy is ~5% in this
case, the total estimate for the ratio of unchanged and
changed groups is 45 5/55 5%. Thus, no changes in
chemical shifts and, hence, in the chemical environment
occurred during the transformation for 50% of groups
in sample 3.
The kinetics of the thermal transformation of comꢀ
pound I was studied in detail using the method of isotherꢀ
mal calorimetry, which has no drawbacks characteristic of
manometric and spectrophotometric procedures of the reꢀ
* The estimate is approximate, because the sum of surface areas
(5.73) is smaller than 6 H, i.e., some signals do not fall to the
isolated integration regions.

Oligomerization of oxadiazole derivatives
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
1437
a
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0

b
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0

Fig. 3. 1Н NMR spectra for samples 2 (a) and 3 (b). For assignment, see the description of the spectrum of compound I in text. Signal
 2.5 is DMSOꢀd5, and  3.26 is a water admixture in the solvent.
action occurs with a heat effect that differs from zero.
Figure 4 shows the dependences of the heat release rates in
the oligomerization of I at different temperatures on the
current heat of the process, which is directly proportional
to the reaction conversion.
.
The total heat of the process Q0 determined by the
numerical integration of the time dependence of the heat
release rate is about 940 J g–1
.
To explain the value of the heat of the process, let us
assume that the main route of the thermal transformation of
compound I is its exothermic isomerization with the formaꢀ
tion of imidazole derivative III followed by its oligomerizaꢀ
tion with the cleavage of one of the N—C bonds in the
heterocycle to form a biradical. Further any end of the
biradical adds to the double bond of the heterocycle to form
a growing biradical oligomer, for example, with structure IV.
The pattern of oligomerization at one of two heteroꢀ
cycles of the molecule was exemplified. A similar process
of radical oligomerization involves the second imidazole
cycle of the compound formed by isomerization. If the
oligomerization proceeds via the proposed mechanism,
the quaternary C atoms are formed in the oligomer strucꢀ
ture of the products observed in the NMR spectrum.
The recombination of two ends of the oligomer biradiꢀ
cal results in chain termination and macrocycle formaꢀ
tion, whereas the recombination of ends of the adjacent
biradicals elongates the substance chain.
The heat of isomerization of two oxadiazole cycles to
the imidazole derivative calculated using the MOPAC
(PM3) program is approximately 330 kJ mol–1 for the iniꢀ
tial compound. The heat of the formation of the polyimidꢀ
azole derivative by the replacement of the C=N double
bond by two ordinary C—N bonds is –3 kJ mol–1. The
heat for the replacement of the С=С double bond by the
С—С ordinary bond and C—N ordinary bond is 41 kJ mol–1
Totally, the calculated heat effect with allowance of the
.

1438
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
Goncharov et al.
dQ/dt/mW g–1
the С(5)—О(1) bond in the second heterocycle and the
addition of the NO group by the nitrogen atom to the С(5)
carbon atom.
60
5
Possible heats of the reaction for the discussed strucꢀ
tures were estimated using the data on the energies of the
cleaved and formed bonds in the reactants and products.
Structures V and VI are formed through the cleavage of
one double C=N bond and one ordinary N—O bond and
formation of ordinary bonds C—N, N—N, and C—O.
50
40
4
30
3
The heat of the reaction will be Q = ЕC—N + ЕN—N
+
20
+ ЕC—O – ЕC=N – ЕN—O = 305 + 163 + 360 – 613 – 157 =
= 58 kJ mol–1, being approximately 20% of the value obꢀ
served in experiment. The formation of structure VII ocꢀ
curs due to the cleavage of one double C=N bond and
ordinary N—O and С—O bonds and the formation of orꢀ
dinary bonds C—С and С—N and one double bond N=O.
2
10
1
200
400
600
800
Q/J g–1
Fig. 4. Heat release rates dQ/dt in the oligomerization of comꢀ
pound I vs current heat of the process Q at 176.1 (1), 180.6 (2),
191.4 (3), 197.3 (4), and 201.6 С (5).
The heat of the reaction should be Q = ЕC—C + ЕC—N
+
+ ЕN=O – ЕC=N – ЕN—O – ЕC—O = 348 + 305 + 621 –
– 613 – 157 – 360 = 150 kJ mol–1, being approximately
50% of the experimental one. Based on the calculated
heats, one can assume a small contribution from the parꢀ
allel reactions corresponding to the formation of strucꢀ
tures V—VII to the overall rate of heat release.
calculation inaccuracy is close to that observed experiꢀ
mentally.
The formation of linkages of different types is proꢀ
posed as additional parallel routes of compound I transꢀ
formation also capable of increasing the weight of the
reaction products. The most probable route can be the
cleavage of the N(2)—O(1) bond and the addition of the
formed biradicals to the C(5)—N(4) or C(3)—N(2) douꢀ
ble bonds of the 1,2,4ꢀoxadiazole cycles, which should
result in the formation of a fragment of conjugated sevenꢀ
and fiveꢀmembered heterocycles. Of numerous possible
variants of addition, thermochemical concepts allow fragꢀ
ments V—VI and fragments VII to take place. Fragments
VII contain the sixꢀmembered heterocycle obtained from
two fiveꢀmembered heterocycles via the mechanism with
the formation of an intermediate carbene accompanying
by the С(3)—N(2) bond cleavage in the heterocycle folꢀ
lowed by the insertion of carbene with the С(3) atom at
The possibility of parallel processes is consistent with
the NMR spectral data. For the groups with the changed
chemical shift, the Me groups of the oxadiazole cycle exꢀ
hibit the highest effect (from  2.69 to 1.76). This is exꢀ
plained, most likely, by the transformation of the С(5)
carbon atom into the quaternary atom and indicates in
favor of the formation of crossꢀlinkage structures of types
V and VI. In these structures, one of the methyl groups
retains the initial chemical environment, while another
(Me*) changes it, which explains an approximately equal
ratio of the changed and unchanged groups in the polyꢀ
mer. The structures of types V, VI, and VII also agree with
the appearance of lines in the carbon spectrum that are
characteristic of quaternary carbon atoms with electroneꢀ
gative substituents ( 110—117).
The dependences of the heat release rate on time and
current heat of the processes are complicated. The obꢀ
served initial region of decreasing heat release rate, beginꢀ
ning from the release of about 8 J g–1 of heat, which correꢀ
sponds to ~1% reaction conversion, is exchanged first by
a region of rate increasing to the release of ~190 J g–1 (~20%
reaction conversion) and then by a region of decreasing
heat release rate to the complete cessation of the process.
To elucidate the problem about a relationship of the deꢀ
crease in the heat release rate in the initial region to the
thermal transformation of a more reactive admixture if
any in the starting sample I, we carried out experiment at
176.1 С interrupting the reaction at the minimum of the
rate, isolating the initial substance from the reaction mixꢀ
ture, and repeating the reaction. The curve of heat release
rate for the thermal transformation of compound I isolatꢀ
ed from the reaction mixture is identical to that of the heat

Oligomerization of oxadiazole derivatives
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
1439
release rate of the initial compound I. Therefore, the iniꢀ
tial region of heat release rate characterizes the reactivity
of intrinsic oxadiazole I. It also could be assumed that the
initial region is related to the equilibrium transformation
of one conformation of the substance into another. Howꢀ
ever, experiment at 170.5 С with interrupting of the reacꢀ
tion at the minimum of the heat release rate and cooling of
the sample to ~20 C with the purpose to shift the conforꢀ
mational equilibrium to the initial position followed by
the repeated placing the ampule into the calorimeter to
continue the reaction did not confirm this assumption.
After the ampule was heated, the same heat release rate
that was observed before interrupting experiment was deꢀ
tected. The regular decrease in the rate at the initial region
at the lowest boundary of the temperature range (176.1 C)
was observed within 75 min and at the upper boundary
(201.6 С) within 20 min. Additional studies are needed to
reveal a reason for such a decrease in the rate.
Assuming that the isomerization of oxadiazole I to the
imidazole derivative followed by oligomerization to form
the polyimidazole derivative as the major reaction that
explains the observed thermal effect of the reaction,
the kinetic scheme of the process will consist of the folꢀ
lowing stages.
(1) Chain generation:
— 1,2,4ꢀoxadiazole derivative transforms into the imidꢀ
azolol derivative;
plex of constants of oligomerization stages, the depenꢀ
dence of the specific reaction rate (d/dt)/(1 –)1.5 on
the reaction depth  was analyzed:
,
where d/dt = (dQ/dt)/Q0 is the process rate, and (dQ/dt)
is the heat release rate (Fig. 5).
The formation of oligomer already at small depths of
conversion increases the viscosity of the reaction medium
and, as a consequence, decreases the termination rate conꢀ
stant, which is limited by the diffusion rate of oligomer
radicals. As a result, the oligomerization rate increases in
the range of 1—20% reaction depth. At high depths of
conversion, an increase in the viscosity of the system reꢀ
sults in the situation where the rate of monomer diffusion
to the oligomer radical becomes the rateꢀdetermining step
for the chain propagation reaction, which formally leads
to a decrease in the chain propagation rate constant and,
finally, to a decrease in the ratio of the chain propagation
rate constant to the square root of the chain termination
rate constant. The oligomerization rate constant deterꢀ
mined by this ratio and the monomer concentration also
begins to decreases from a certain depth of conversion.
The experiment at 192.3 С in which an equal weight
of the oligomer reaction products was added to the startꢀ
ing compound I confirms the conclusion about the subꢀ
stantial influence of the viscosity of the reaction medium
on the polymerization rate (Fig. 6). The addition of the
product increases the initial heat release rate by 3.7 times.
Then the rate only decreases during the process. The heat
of the thermal transformation of oxadiazole I in the presꢀ
ence of the reaction product does not differ from that for
the thermal transformation of the pure compound.
— cleavage of the imidazole ring at the ordinary N—C
bond to form the biradical.
(2) Chain propagation:
— interaction of the biradical with the free valence on
the nitrogen atom with the C=N double bond of the imidꢀ
azole ring results in the formation of the free valence on
the C atom of the heterocycle and an increase in the biradꢀ
ical length by one structural unit;
— similar chain generation reactions occurs beginning
from another end of the biradical with the free valence at
the C atom.
The degree of acceleration remains almost unchanged
when the ratio of the weighed sample to the free volume of
(d/dt)/(1 – )1.5
(3) Chain termination:
— reactions of closure of macrobiradical end with free
valences at the nitrogen and carbon atoms to form macroꢀ
cycles with N—C, N—N, and C—С bonds. These reacꢀ
tions lead to the formation of a network (crossꢀlinked)
polymer. An increase in viscosity decreases the probability
of collision of the biradical ends to form the macrocycle,
which corresponds to a decrease in the chain termination
rate constant and results in the observed increase in the
rate of oligomerization of the starting compound.
The observed dependence of the heat release rate on
the reaction depth is determined by both the change in the
monomer concentration in the course of the process and
the change in the chain termination rate constants due to
an increase in the viscosity of the reaction mixture when
oligomer molecules are formed. To distinguish the influꢀ
ence of the increasing viscosity of the medium on a comꢀ
0.5
0.4
0.3
0.2
0.1
0.2
0.4
0.6
0.8
1.0

Fig. 5. Oligomerization rate of compound I (d/dt)/(1 –)1.5 vs
reaction conversion  at 192.3 С.

1440
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
Goncharov et al.
dQ/dt/mW g–1
crease in the specific reaction rate is observed when the
reaction conversions higher than 0.5 are achieved in exꢀ
periment with pure compound I. A similar increase in the
specific rate from the indicated conversion is also observed
when the reaction product is added to the initial comꢀ
pound I. It can be assumed that at these reaction conversions
an increase in the specific rate is determined by a signifiꢀ
cant decrease in the chain termination rate constant along
with a continuing increase in the viscosity of the reaction
medium. Second, the difference in specific rates between
the transformation of pure compound I and that with the
product additive at the 20% reaction conversion, for which
the specific rate for pure compound I has a maximum, is
only 1.2 times. Therefore, an additive of 25% reaction
product to the starting compound I should result in the
limiting increase in the initial rate of the process.
The temperature dependences of the initial and maxiꢀ
mum heat release rates in the coordinates of the Arrhenius
equation and the results on the initial oligomerization rates
obtained by the spectrophotometric procedure are preꢀ
sented in Fig. 8.
The scatter in converstions of achieving the maximum
rate is not too large (see Fig. 4). This is probably associatꢀ
ed with the temperature dependence of the viscosity of the
reaction medium: the higher the temperature, the lower
the viscosity of the medium and, hence, the maximum
rate is attained at greater conversions. The study of the
temperature dependence of the maximum rate of the process
is probably suitable for an analysis of the possibility to
retain the oligomerization mechanism at the initial stage
and during the development of the process.
40
30
20
10
1
2
3
200
400
600
800 Q/J g–1
Fig. 6. Heat release rate dQ/dt in the oligomerization of comꢀ
pound I vs current heat of the process Q at 192.3 С for the initial
oxadiazole I with an additive of the thermal transformation prodꢀ
uct (1) for pure oxadiazole I at weight samples m = 0.05 (2) and
0.1 g (3).
the ampule (m/V) changes twofold (see Fig. 6). Therefore,
the rate increase is related to the change in the properties
of the condensed phase only.
The dependences of the specific reaction rate (d/dt)/
(1–) on the reaction depth for the experiments considꢀ
ered above (see Fig. 6) are compared in Fig. 7. In both
experiments, the initial region of the rate drop is observed.
The distinction is that for the thermoinitiated oligomerꢀ
ization of pure compound I an increase in the viscosity of
the reaction mixture increases the specific rate already at
small depths of conversion. The subsequent rate drop after
the maximum of the transformation of compound I is
achieved is related to a decrease in its current concentraꢀ
tion. Two facts are noteworthy when comparing the deꢀ
pendences of the specific rates. First, the repeated inꢀ
The results on the initial reaction rate determined usꢀ
ing two procedures are well consistent. The value of
101.2 8.8 kJ mol–1 was obtained for the activation enerꢀ
gy, and the logarithm of the preꢀexponential factor was
6.6 1.0 (s–1). Within the experimental error, the initial
(d/dt)/(1 – )
log(k/h–1
–0.6
)
0.15
1
2
3
–1.0
–1.4
–1.8
–2.2
–2.6
–3.0
2
0.10
1
0.05
2.10
2.20
2.30
2.40
2.50 103/T/K–1
0.2
0.4
0.6
0.8
1.0

Fig. 7. Specific reaction rate (d/dt)/(1–) vs reaction converꢀ
sion  for the oligomerization of compound I (192.3 С) for the
initial oxadiazole I (1) and with an additive of the oligomerizaꢀ
tion product (2).
Fig. 8. Logarithm of the oligomerization rate of compound I vs
reciprocal temperature. The initial rate was determined by caloriꢀ
metry (1) and spectrophotometry (2); the maximum rate was
determined by calorimetry (3).

Oligomerization of oxadiazole derivatives
Russ.Chem.Bull., Int.Ed., Vol. 62, No. 6, June, 2013
1441
5. T. K. Goncharov, N. I. Golovina, V. V. Dubikhin, E. L.
Ignat´eva, A. V. Nabatova, G. M. Nazin, G. V. Shilov, Tez.
dokl. IV Mezhdunar. konf. po vysokoenergeticheskim materiꢀ
alam "HEMs 2008" [Proc. IV Intern. Conf. on HighꢀEnergy
Materials "HEMs 2008"] (Belokurikha, Altai Territory, Sepꢀ
tember 3—5, 2008), Biisk, 2008, 29 (in Russian).
6. O. S. Galyuk, Yu. I. Rubtsov, G. F. Malinovskaya, G. B.
Manelis, Zh. Fiz. Khim., 1965, 39, 2329 [Russ. J. Phys. Chem.
(Engl. Transl.), 1965, 39].
7. A. Altomare, G. Cascarno, C. Giacovazzo, A. Gualardi,
J. Appl. Crystallogr., 1993, 26, 343.
8. G. M. Sheldrick, SHELXꢀ97. Programs for Crystal Structure
Analysis, Gottingen (Germany), 1998, 2332.
and maximum rate have the same activation energy and
the preꢀexponential factor differs by 1.5 times. This can
indicate that the oligomerization proceeds via the single
mechanism in the course of the whole process.
References
1. US Pat. 3390204; Сhem. Аbstrs, 1968, 64, 35900.
2. M. A. Okhotnikov, S. I. Valuev, Zh. Prikl. Khim. [Russ. J.
Appl. Chem.], 2002, 75, 1555 (in Russian).
3. A. M. Belousov, E. A. Paznikov, G. Ya. Petrova, P. I.
Kalmykov, Zh. Prikl. Khim. [Russ. J. Appl. Chem.], 2003, 76,
1197 (in Russian).
4. A. I. Yakubov, D. V. Tsyganov, L. I. Belen´kii, M. M. Krayuꢀ
shkin, Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.),
1991, 40 [Izv. Akad. Nauk SSSR, Ser. Khim., 1991, 1201].
Received May 31, 2012;
in revised form February 22, 2013