The thermodynamic properties of four triphenylsilane acetylene peroxides

Article abstract of DOI:10.1134/S0036024406030046

The enthalpies of formation of four acetylene triphenylsilane peroxides were determined in the condensed and gaseous states. Three silicon-containing fragment increments for the Benson additive scheme and the enthalpies of formation of three radicals formed in the dissociation of the compounds at the O-O bond were calculated. Pleiades Publishing, Inc., 2006.

Full text of DOI:10.1134/S0036024406030046

ISSN 0036-0244, Russian Journal of Physical Chemistry, 2006, Vol. 80, No. 3, pp. 330–334. © Pleiades Publishing, Inc., 2006.
Original Russian Text © V.N. Dibrivnyi, G.V. Mel’nik, Yu.Ya. Van-Chin-Syan, A.P. Yuvchenko, 2006, published in Zhurnal Fizicheskoi Khimii, 2006, Vol. 80, No. 3, pp. 408–412.
CHEMICAL THERMODYNAMICS
AND THERMOCHEMISTRY
The Thermodynamic Properties
of Four Triphenylsilane Acetylene Peroxides
V. N. Dibrivnyi1, G. V. Mel’nik1, Yu. Ya. Van-Chin-Syan1, and A. P. Yuvchenko2
1 Lviv Politechnic National University, Lviv, Ukraine
2 Institute of Physicoorganic Chemistry, Belarussian Academy of Sciences, Minsk, 220072 Belarus
E-mail: vdibriv@polynet.lviv.ua
Received March 4, 2005
Abstract—The enthalpies of formation of four acetylene triphenylsilane peroxides were determined in the con-
densed and gaseous states. Three silicon-containing fragment increments for the Benson additive scheme and
the enthalpies of formation of three radicals formed in the dissociation of the compounds at the O–O bond were
calculated.
DOI: 10.1134/S0036024406030046
INTRODUCTION
3-methyl-3-tert-hexylperoxy-1-triphenylsilyl-1-butine
ëç3(ëç2)2(ëç3)2ëééë(ëç3)2ë≡ëSi(ë6ç5)3, (III)
Organosilicon compounds have been extensively
used in the production of modern materials and compo-
sitions. Thanks to their high thermal stability, hydro-
phobic properties, and stability during storage, silicon-
containing peroxides also find wide application as ini-
tiators of various free-radical processes, for instance, in
high-temperature (up to 600 K) initiation of polymer-
ization, polymer modification, elastomer vulcanization,
and lacquer composition solidification. At the same
time, data on the thermodynamic properties of silicon-
containing peroxides are virtually absent. The reason
for this is the complexity of preparing organosilicon
peroxides sufficiently pure (more than 99% major com-
ponent) for precision thermodynamic measurements
and the great complexity of such measurements them-
selves.
and 4-methyl-4-tert-butylperoxy-1-triphenylsiloxy-1-
pentine
(CH3)3COOC(CH3)2C≡CCH2OSi(C6H5)3. (IV)
EXPERIMENTAL
Compounds I–III were prepared in the reactions
between 3-methyl-3-tert-alkylperoxy-1-butines and
butyllithium. Next, lithium peroxyacetylenides formed
reacted with triphenylchlorosilane [2] to produce the
desired substances,
ROOCMe2C≡CH BuLi ROOCMe2C≡CLi
Ph3SiCl
ROOCMe2C≡CSiPh3.
In [1], we for the first time determined the enthalpies
of formation of seven liquid methylsilyl acetylene per-
oxides by burning the peroxides with an explosion in an
isothermic-shell calorimeter. The substances were
sealed in Terylene ampules for simultaneously igniting
the whole sample surface and increasing temperature at
the contact, which ensured maximum combustion com-
pleteness. We used double-walled ampules to prevent
the diffusion of substance molecules through the film.
Compound IV [3] was synthesized following the
scheme
ROOCMe2C≡CHBuLi ROOCMe2C≡CLi
+CH2O
ROOCMe2C≡CCH2OLi
Ph3SiCl
ROOCMe2C≡CCH2OSiPh3.
Crystals I were purified by low-temperature recrystalli-
zation from hexane. Liquids II and III were purified
by removing volatile impurities under evacuation for
3−4 h at 323 K and 0.3 kPa. Liquid IV was evacuated
for 2–3 h at 303–313 K and 0.4 kPa.
The products were characterized by elemental anal-
yses for carbon, hydrogen, silicon, and active oxygen;
their molecular weights were determined by cryoscopy,
and their IR, UV, and NMR spectra were recorded. The
individual character of the compounds was proved by
thin-layer chromatography on Silufol UV-254 plates.
We were the first to experimentally determine the
enthalpies of formation of four acetylene triphenylsi-
lane peroxides. These were 3-methyl-3-tert-butylper-
oxy-1-triphenylsilyl-1-butine
ëç3)3ëééë(ëç3)2ë≡ëSi(ë6ç5)3,
(I)
3-methyl-3-tert-amylperoxy-1-triphenylsilyl-1-butine
ëç3ëç2(ëç3)2ëééë(ëç3)2ë≡ëSi(ë6ç5)3, (II)
330

THE THERMODYNAMIC PROPERTIES OF FOUR TRIPHENYLSILANE ACETYLENE PEROXIDES
331
Table 1. Results obtained in experimentally determining the
temperature dependence of saturated vapor pressure over IV
The purity of the compounds was checked by high-
pressure liquid chromatography with the use of a sys-
tem consisting of a pump, UV and refractometric detec-
tors, and a Separon SGX CN column (length 150 mm
and diameter 3.3 mm, particle size 7 µm). The eluent
was a heptane–isopropanol mixture with a 98 : 2 vol-
ume ratio. The content of the major component was no
less than 99.2 wt % in all the compounds studied. We
were unable to determine the nature of impurities.
The temperature dependences of compound satu-
rated vapor pressures and the enthalpies of vaporization
were determined by the Knudsen integral effusion
method. The results obtained this way for I–III, the
design of the experimental unit, the reliability of its
operation, and the procedure for measurements were
reported in [4].
Saturated vapor pressure of IV was determined in a
series of 18 experiments by the extrapolation of the
vapor pressure in a cell (membrane orifice diameter was
0.5903 mm) to “zero” orifice area with the use of a cor-
rection factor of 1.14. The vacuum line ensured the
attainment of a 0.1 Pa pressure in 45 10 s. The mass
of the effused substance ∆m was determined by weigh-
ing the effusion cell with an accuracy of 5 × 10–6 g.
Errors in maintaining constant temperature and deter-
mining the temperature of measurements T and the total
effusion time τ were of 0.1 ä, 0.05 K, and 1 s,
respectively. The effective effusion time (calculated
time under stationary conditions during which the
effused substance weight equaled that under nonsta-
tionary conditions) was 105 s.
No.
τ, s
∆m × 103, g T, K
pc, Pa
p, Pa
1
2
4866
7266
7266
5466
5466
5466
5466
5467
5467
3668
3668
4566
4566
3666
3666
3666
3666
3666
0.400
1.320
1.350
1.565
1.550
1.515
1.600
1.875
1.900
1.870
1.950
2.610
2.700
3.025
3.150
2.905
3.050
3.745
353.5
362.7
362.7
367.8
367.8
367.8
367.8
371.6
371.6
376.0
376.0
379.4
379.4
382.5
382.5
382.9
382.9
386.0
0.066
0.149
0.152
0.236
0.234
0.228
0.241
0.284
0.288
0.425
0.443
0.479
0.495
0.693
0.722
0.666
0.699
0.862
0.075
0.170
0.173
0.269
0.267
0.260
0.275
0.324
0.328
0.485
0.505
0.546
0.564
0.790
0.823
0.759
0.797
0.983
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
combustion of benzoic acid corrected using the Jessup
factor was 26426.9 J/g. The initial pressure of oxygen
preliminarily purified from combustible impurities,
carbon dioxide, and water was 2.94 × 106 Pa. The initial
temperature of the main period was 298.15 K in all
experiments.
Solid peroxide I was tabletted prior to combustion.
Liquid substances II–IV were sealed in double-walled
ampules following the procedure described in [1].
Combustion was performed in platinum crucibles to
decrease carbon black formation. Insignificant carbon
black deposits on platinum crucible walls were
weighed with an accuracy of 5 × 10–6 g. Each combus-
The results of effusion experiments with IV, includ-
ing saturated vapor pressure p and pressure in the cell
pc, are listed in Table 1. The temperature dependence of
saturated vapor pressure is described by the equation
lnp = (27.3 1.2) – (105.3 4.5) × 102/T,
with the correlation coefficient ρ = 99.98. The enthalpy
of vaporization ∆vH averaged over the temperature
range of measurements was 87.6 3.7 kJ/mol.
The combustion of the compounds was performed tion experiment was followed by quantitative analyses
in a modernized V-08 MA isothermic-shell ( 0.003 K) of combustion products for carbon mono- and dioxides,
calorimeter. The calorimeter was modernized to carbon black, and nitric acid. The weight of the sub-
decrease heat loss and fluctuations. The orifices for stance burned was determined from the amount of car-
resistance thermometers were tightened with foam- bon dioxide [5] with an accuracy of 1 × 10–4 g. The
plastic sockets to a maximum degree possible. A resis- reliability of our gas analyses was substantiated in a
tance was connected in series to the heaters of the sys- series of calibration experiments with the combustion
tem for maintaining a constant shell temperature to of standard benzoic acid. The amounts of carbon diox-
decrease shell temperature fluctuations. The trans- ide formed in the combustion of 1 g of the film and 1 g
former for incandescing the wire that ignited samples of the cotton thread were experimentally found to be
was replaced by a set of capacitors. The energy of wire 2.2872 and 1.6284 g, respectively. The content of carbon
heating was as low as 1.3 0.3 J in calibration and mea- monoxide in combustion products was beyond the sensi-
surement experiments and was therefore excluded from tivity of our method for determining CO ( 1 × 10–6 g).
calculations. The energy equivalent of the calorimetric The amounts of nitric acid formed in combustion exper-
system (W = 14930.5 J/V) was determined with an iments were determined by titration with a 0.1 N solu-
accuracy of 0.056% in a series of ten experiments with tion of NaOH. When samples were prepared and exper-
the combustion of standard benzoic acid, K-1 brand iments conducted as described above, the completeness
(major component content 99.995 mol %). The heat of of combustion (k) was of 99.05–99.97%, which was
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY Vol. 80 No. 3 2006

332
DIBRIVNYI et al.
Table 2. Results obtained in experimentally determining the energies of combustion of the compounds
m, g
∆T, V
QΣ, J
Qt, J
Qa, J
Qn, J
Qc, J
k, %
–∆U, J/g
(CH3)3COOC(CH3)2C≡CSi(C6H5)3
0.127610
0.11690
0.3306
0.3022
0.5733
0.5365
0.5760
0.4463
0.4020
4936.2
4512.1
8559.9
8010.5
8600.3
6663.7
6002.3
49.36
46.90
49.44
47.52
51.20
37.58
37.58
965.95
971.34
948.63
921.09
930.27
1078.72
987.40
2.95
2.36
1.77
4.72
3.54
2.95
0.89
6.56
6.70
6.89
7.87
10.99
9.84
6.56
99.76
99.80
99.86
99.82
99.55
99.82
99.73
38413.3
38319.5
38486.4
38312.3
38476.9
38411.7
38390.8
0.221580
0.208400
0.223425
0.173020
0.155925
CH3CH2(CH3)2COOC(CH3)2C≡CSi(C6H5)3
0.128565
0.117135
0.123600
0.166170
0.127825
0.129925
0.128450
0.3998
0.3692
0.3856
0.4937
0.3945
0.4106
0.4110
5969.4
5512.5
5757.4
7371.4
5890.3
6130.7
6136.6
47.61
36.75
42.01
41.18
42.60
42.26
43.60
965.95
971.34
948.63
921.09
930.27
1078.72
987.40
2.95
2.36
1.77
4.72
3.54
2.95
0.89
11.97
10.50
15.58
20.17
18.20
12.30
9.68
99.48
99.59
99.84
99.95
99.46
99.63
99.65
38815.9
38675.8
38735.6
38677.4
38794.9
38774.0
38792.3
CH3(CH2)2(CH3)2COOC(CH3)2C≡CSi(C6H5)3
0.102105
0.107570
0.11290
0.3360
0.3503
0.3594
0.3820
0.3841
0.3870
0.3836
5016.8
5230.3
5366.2
5703.6
5735.0
5778.3
5727.5
50.11
49.61
49.95
51.28
48.02
37.42
45.69
1000.71
1012.07
1001.06
955.74
1.18
0.59
0.59
0.59
0.59
1.18
2.36
11.97
7.87
99.80
99.58
99.63
99.83
99.95
99.82
99.97
39018.5
38981.5
39032.7
39044.2
38897.9
38977.3
38951.2
13.45
9.68
0.120710
0.119675
0.122320
0.121305
1040.75
992.22
6.89
10.99
10.99
966.52
(CH3)3COOC(CH3)2C≡CCH2OSi(C6H5)3
0.107365
0.150230
0.152825
0.145165
0.144890
0.155300
0.146095
0.118120
0.141610
0.108865
0.120600
0.3301
0.4364
0.4441
0.4227
0.4280
0.4489
0.4283
0.3626
0.4156
0.3351
0.3622
4928.7
6515.9
6630.9
6311.3
6390.5
6702.5
6394.9
5414.0
6205.3
5003.4
5408.0
57.96
62.47
73.08
59.38
57.71
58.80
45.85
49.69
50.20
46.86
47.36
950.46
973.52
1017.80
963.77
1042.47
971.46
1012.18
1045.68
994.63
983.27
974.33
1.18
1.18
1.18
0.59
1.77
0.59
1.18
1.18
1.18
1.18
1.18
7.87
7.05
99.56
99.58
99.05
99.52
99.52
99.52
99.25
99.74
99.75
99.81
99.52
36731.9
36669.9
36655.1
36662.3
36720.3
36791.2
36844.5
36725.1
36603.0
36624.9
36622.9
10.82
9.51
6.23
14.92
7.71
10.33
11.97
8.36
10.00
quite acceptable for compounds with such complex solution of nitric acid (Qn = 59 kJ/mol [5]). The results
structures.
obtained in determining the energy of combustion of
I−IV are listed in Table 2. In addition to the notation
specified above, Table 2 contains the following nota-
tion: m is the burned substance weight, QΣ is the total
amount of energy released, ∆T is the corrected temper-
To take into account the energy (J/g) released in
experiments as a result of side processes, we introduced
corrections for the combustion of Terylene ampule
(Qa = 22944.2 [6]) and cotton thread (Qt = 16704.2 [7])
and the formation of carbon black (Qc = 32800) and a ature rise during measurements, and ∆U is the energy
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY Vol. 80 No. 3 2006

THE THERMODYNAMIC PROPERTIES OF FOUR TRIPHENYLSILANE ACETYLENE PEROXIDES
333
Table 3. Standard enthalpies of combustion and formation of acetylene peroxides, kJ/mol
∆fH° (g, 298.15 K)
∆fH° (cond.,
Compound –∆U° (298.15 K)
π
–∆nRT –∆cH° (298.15 K)
298.15 K)
experimental
calculated
I
II
III
IV
15914 26
16603 23
17250 21
16305 21
8
8
9
9
19
20
21
19
15933 26
16623 23
17271 21
16324 21
81 26
92 23
238 27
212 23
187 21
–119 21
233
212
61 21
192
–207 21
–119
of substance combustion under bomb conditions at the measurement errors. This is evidence that the enthalpy
initial temperature of measurements. The enthalpies of properties of the compounds under study are additive.
combustion of the compounds were taken to corre- Comparing the structures of compounds I–III with
spond to the reaction
those from [1] pairwise shows that the former can be
obtained from the latter and vice versa by the formal
replacement of the trimethylsilyl and triphenylsilyl
groups. The differences of the enthalpies of formation
for the pairs of substances that we compare are 492,
481, and 471 kJ/mol, respectively. These values coin-
cide to within the error of enthalpy determinations and
substantiate the additive character of enthalpy proper-
ties.
This circumstance allows us to calculate the contri-
butions of silicon-containing groups, which have been
unknown in additive schemes. We were able to deter-
mine such contributions only to the enthalpy of forma-
tion of compounds in the gas phase for the Benson
scheme, which contains the most complete list of the
contributions of groups present in the compounds stud-
ied. In addition, we used the increments (kJ/mol)
O−(C)(O), –19.32; C–(O)(C)2(Ct), –9.55; and
C−(Si)(H)3, –42.17 kJ/mol. The increments were calcu-
lated from the enthalpies of formation reliably deter-
mined for fairly large numbers of compounds [14–16].
Substituting the group contributions into the enthalp-
ies of formation of compounds I–III yielded an over-
determined system of three equations. Solving this
system allowed us to obtain the contribution of the
(ët)≡ëSi(ëb)3 fragment equal to 128 kJ/mol. The con-
tribution of the Si–(O)(Cb)3 group (–211 kJ/mol) was
found by substituting the corresponding group contri-
butions to the enthalpy of formation of IV.
CaHbOcSid(l, s) + (a + b/4 – c/2 + d)O2(g)
aCO2(g) + (b/2)H2O(l) + dSiO2(‡m. hydr.).
(1)
The standard energies of combustion ∆U°(298.15 ä)
and the enthalpies of combustion ∆ÒH°(298.15 ä) and
formation ∆fH°(298.15 ä) of the peroxides in the con-
densed state (Table 3) were calculated with the inclu-
sion of the Washburn correction π and correction for the
work of expansion ∆nRT (∆n is the change in the num-
ber of gas moles in reaction (1)). We used the follow-
ing reference ∆fH°(298.15 K) values (kJ/mol): CO2(g),
–393.514 0.046; H2O(l), –285.830 0.042 [9]; and
SiO2(am. hydr.), –939.39 0.52 [10]. The standard
deviation of the mean combustion energy value was
calculated using the Student test value for a 5% signif-
icance level. Table 3 also contains the standard enthal-
pies of formation of the peroxides in the gas phase
∆fH°(g, 298.15 K). These were determined using the
enthalpies of vaporization of IV and I–III; the latter
were 115.9 3.2, 120.3 5.8, and 126.3 3.0 kJ/mol,
respectively [1]. We also used the enthalpy of fusion of I:
41 1 kJ/mol [7].
RESULTS AND DISCUSSION
Compounds I–III make up a homologous series
each member of which is one ëç2 group longer than
the preceding one. The contribution of the ë–(ë)2(ç)2
group to the enthalpy of formation of compounds in the
The peroxide bond energy was calculated by con-
structing a system of peroxide homolytic decomposi-
tion reactions and the corresponding system of thermo-
chemical equations,
1
liquid and gaseous states and to the enthalpy of vapor-
ization is –24, –20, and 4.8 kJ/mol according to the
Benson [11], Kizin et al. [12], and Lebedev–Mirosh-
nichenko [13] schemes, respectively. Along the homol-
ogous series I–III, changes in the enthalpy of formation
of the liquid and gaseous compounds and in their
enthalpies of vaporization are –30 and –31, –26 and –
25, and 4.4 and 6.0 kJ/mol; they coincide with the cor-
responding contribution of the ëç2 group to within
.
.
R1OOR2
R1O + R2O ,
.
.
D(é–é) = ∆fH°(R1O ) + ∆fH°(R2O ) – ∆fH°(I);
.
.
R3OOR2
R3O + R2O ,
.
.
D(é–é) = ∆fH°(R3O ) + ∆fH°(R2O ) – ∆fH°(II);
1
.
.
Parenthesized atoms in the denotations of group contributions
R4OOR2
R4O + R2O ,
describe the environment of the central atom, and subscripts indicate
the number of atoms of a given kind in the environment; C is the
C atom at the triple bond, and C is the C atom in the benzene ring.
.
.
t
D(é–é) = ∆fH°(R4O ) + ∆fH°(R2O ) – ∆fH°(III);
b
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY Vol. 80 No. 3 2006

334
DIBRIVNYI et al.
4. V. N. Dibrivnyi, G. V. Lutsiv, Yu. Ya. Van-Chin-Syan,
.
.
R1OOR5
R1O + R5O ,
et al., Zh. Fiz. Khim. 73 (12), 2254 (1999) [Russ. J.
Phys. Chem. 73 (12), 2040 (1999)].
.
.
D(é–é) = ∆fH°(R1O ) + ∆fH°(R5O ) – ∆fH°(IV),
5. Experimental Thermochemistry, Ed. by F. D. Rossini
.
.
.
.
(Interscience, New York, 1956).
where R1O = Me3ëO , R2O = Ph3Sië≡CCMe2O ,
R3O = t-C5H11O , R4O = t-C6H13O , and R5O =
6. S. M. Pimenova, Candidate’s Dissertation in Chemistry
(Mosk. Gos. Univ., Moscow, 1981).
.
.
.
.
.
.
7. G. V. Mel’nik, Candidate’s Dissertation in Chemistry
(L’vovsk. Nats. Univ. im. I. Franko, Lvov, 2000).
(Ph)3SiOCH2C≡CCåÂ2O .
The known enthalpies of formation ∆fH° of the rad-
8. M. M. Popov, Thermometry and Calorimetry (Mosk.
Gos. Univ., Moscow, 1954) [in Russian].
icals produced in the dissociation of the compounds at
.
the peroxide bond are –96.2 kJ/mol for Me3CO [17]
.
9. “CODATA Recommended Key Values for Thermody-
namics 1977,” J. Chem. Thermodyn., No. 10, 903
(1978).
and –121.3 kJ/mol for t-C5H11O [18]. These data were
insufficient for solving the above system of equations.
However, considering the structural similarity of com-
pounds I–III and compounds studied in [1] and taking
into account that differences in structure between IV
and the other compounds begin with the atoms sepa-
rated by four bonds from the peroxide bond, we may
quite rigorously use the peroxide bond energy [1] equal
to 148 kJ/mol. Solving the system of equations then
yields the enthalpies of formation ∆fH° of the radicals
10. M. G. Voronkov, V. A. Klyuchnikov, T. F. Danilova,
A. N. Korchagina, et al., Izv. Akad. Nauk SSSR, Ser.
Khim., No. 9, 1970 (1986).
11. S. W. Benson, Thermochemical Kinetics: Methods for
the Estimation of Thermochemical Data and Rate
Parameters (Wiley, New York, 1968; Mir, Moscow,
1971).
12. A. N. Kizin, P. L. Dvorkin, G. L. Rigova, and Yu. A. Le-
bedev, Izv. Akad. Nauk SSSR, Ser. Khim., No. 2, 372
(1986).
13. Yu. A. Lebedev and E. A. Miroshnichenko, Thermo-
chemistry of Vaporization of Organic Compounds
(Nauka, Moscow, 1981) [in Russian].
.
.
Ph3Sië≡CCMe2O , (Ph)3SiOCH2C≡CC(CH3)2O , and
.
ëç3(ëç2)2(ëç3)2ëO (482, 125, and –147 kJ/mol,
respectively). The enthalpy of formation of the last rad-
ical obtained in [1] was –131 kJ/mol. We recommend
using the average value, –139 kJ/mol.
14. A. N. Mosin, Zh. Fiz. Khim. 50 (3), 866 (1976).
15. Yu. Ya. Van-Chin-Syan, Doctoral Dissertation in Chem-
istry (Mosk. Gos. Univ., Moscow, 1986).
16. I. B. Rabinovich, E. G. Kiparisova, and Yu. A. Aleksan-
drov, Dokl. Akad. Nauk SSSR 200, 1116 (1971).
REFERENCES
17. Yu. Ya. Van-Chin-Syan, N. S. Kachurina, Yu. P. Pav-
lovskii, et al., in Organic Peroxides and Homolytic
Reactions with Thei Participation: Interschool Collec-
tion of Scientific Papers (Volgograds. Politekhn. Inst.,
Volgograd, 1989), p. 69 [in Russian].
18. L. V. Gurvich, G. V. Karachevtsev, V. N. Kondrat’ev,
et al., Dissociation Energies of Chemical Bonds, Ioniza-
tion Potentials, and Electron Affinities (Nauka, Moscow,
1974) [in Russian].
1. V. N. Dibrivnyi, G. V. Mel’nik, V. V. Kochubei, et al.,
Zh. Fiz. Khim. 77 (2), 232 (2003) [Russ. J. Phys. Chem.
77 (2), (2003)].
2. A. P. Yuvchenko, E. A. Dikusar, N. A. Zhukovskaya,
et al., Zh. Org. Khim. 63 (1), 143 (1993).
3. A. P. Yuvchenko, E. A. Dikusar, E. S. Gurinova, et al.,
Zh. Org. Khim. 64 (8), 1692 (1994).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY Vol. 80 No. 3 2006