Borazine Thermal Decomposition in Unsaturated Vapor

Article abstract of DOI:10.1134/S1070363218120046

The temperature dependence of vapor pressure over liquid and solid borazine has been measured by static tensiometry method with a membrane null-manometer at 200–259 K. Gas-phase heat-induced decomposition of borazine in its unsaturated vapor has been studied over the 373–473 K range.

Full text of DOI:10.1134/S1070363218120046

ISSN 1070-3632, Russian Journal of General Chemistry, 2018, Vol. 88, No. 12, pp. 2476–2479. © Pleiades Publishing, Ltd., 2018.
Original Russian Text © A.S. Zavgorodnii, A.Yu. Timoshkin, 2018, published in Zhurnal Obshchei Khimii, 2018, Vol. 88, No. 12, pp. 1955–1959.
Borazine Thermal Decomposition in Unsaturated Vapor
a
a
A. S. Zavgorodnii and A. Yu. Timoshkin *
a
St. Petersburg State University, Universitetskii pr. 26, St. Petersburg, 198504 Russia
*
e-mail: a.y.timoshkin@spbu.ru
Received July 9, 2018
Revised July 9, 2018
Accepted July 12, 2018
Abstract—The temperature dependence of vapor pressure over liquid and solid borazine has been measured by
static tensiometry method with a membrane null-manometer at 200–259 K. Gas-phase heat-induced
decomposition of borazine in its unsaturated vapor has been studied over the 373–473 K range.
Keywords: borazine, tensiometry, vapor pressure, thermal decomposition
DOI: 10.1134/S1070363218120046
Borazine (B N H ), an inorganic heterocycle known
bromide as the only volatile product has yielded the
non-volatile polymer of the (Br1.26B N H ) composi-
tion [9].
3
3
6
since 1926 [1], has attracted attention of chemists as an
isoelectronic and isostructural analog of benzene. Gas-
phase decomposition of borazine and its derivatives is
of practical interest, as it can afford boron nitride and
boron-nitrogen inorganic polymers which have been
recognized as promising semiconductor coatings
isostructural to graphite. The boron nitride coatings are
advantageous over carbon due to better resistance
against oxidation. Thin films of hexagonal boron
nitride can be prepared via chemical vapor deposition
of volatile boron compounds (BCl or B H ) in the
3
3
1.26 n
This study aimed to investigate the kinetics of
borazine decomposition in the gas phase at 373–473 K
and to identify the reaction products.
Vapor pressure measurements. Three series of
measurements were performed. Due to small amount
of borazine, the vapor in the working volume of a null-
manometer was unsaturated even at room temperature.
Therefore, the working volume was frozen once with
liquid nitrogen, and then the pressure in the system
was measured during its slow heating owing to heat
exchange with the medium. Since borazine melting
point is 216 K [1], the pressure increase in the system
could be assigned to the processes of borazine
sublimation and evaporation. While cooling the system
to 77 K the presence of non-condensable admixture
gas (presumably hydrogen), in the system was detected.
The partial pressure of that gas was considered by
linearization of the dependence of the vapor pressure
logarithm on the reciprocal temperature in the low-
temperature range using the method of progressive
approximation [10]. The plot of the borazine vapor
pressure logarithm as a function of the reciprocal
temperature is shown in Fig. 1. Borazine evaporation
3
2
6
presence of ammonia [2, 3]. The major disadvantages
of this technique are high cost and low yield. In this
regard, the use of borazines as precursors of boron
nitride containing boron and nitrogen in the same
molecule seems more efficient way of the coatings
production [4].
Despite the practical interest, thermal stability and
kinetics of gas-phase decomposition of borazine have
been scarcely studied. Although the temperature
dependence of the heat capacity of liquid and solid
borazine has been determined experimentally over the
–310 K range [5], the high-temperature decomposi-
tion of gaseous borazine has remained unexplored.
Certain boron-substituted borazine derivatives
0
(
trichloro-, trialkyl-, trialkylamino-, and tribromo-
was described by the log (p/p ) = –3.50(±0.07)/T +
0
borazines) have been better studied [6-9]. Gas-phase
decomposition of tribromoborazine at temperatures up
to 443 K accompanied by the formation of hydrogen
10.5(±0.3) equation over the 217–259 K range; the
values of the evaporation enthalpy and entropy were
∆ H° = 29.1±0.6 kJ/mol and ∆ S° = 86.9±
vap 238
vap 238
2
476

BORAZINE THERMAL DECOMPOSITION IN UNSATURATED VAPOR
2477
–1
1
000/T, K
T, K
Fig. 1. Logarithm of the saturated vapor pressure above
1) liquid and (2) solid borazine as functions of reciprocal
temperature.
Fig. 2. Temperature dependence of the p/T ratio in the
system with borazine: (1) the first heating, (2) the second
heating.
(
–
1
–1
2
.5 J mol K . Conversion to the standard tempera-
The results of the measurements during the first
two constant-temperature stages are presented as the
temperature dependence of the p/T ratio (Fig. 2).
ture was carried out under approximation of constant
heat capacities of gaseous and liquid borazine in that
temperature range. The values of ∆ H° 27.0±
vap 298
–
1
–1
p(H2)
1
.8 kJ/mol and ∆ S° 79.1±6.7 J mol
K
are
vap 298
Δ1 = Δ
p(H2)
,
consistent with the published data (∆ H° = 29.3±
T
vap 298
–
1
–1
0
.4 kJ/mol and ∆ S°298 = 89.2±0.1 J mol
K
[11])
vap
p(BZ)
.
within the experimental errors.
Δ
2
= Δ
− Δ
T
T
Vapor pressure over solid borazine was described
The only gaseous component in the system at 77 K
was hydrogen. The exponential part of the plot
corresponding to temperatures below 310 K was
related to the borazine evaporation. At T 313±3 K,
borazine was completely transformed into the un-
saturated vapor. However, the components less volatile
than borazine and hydrogen passed into the vapor at
higher temperatures. It is worth notice that several
such products were present, and they were also
decomposed during the constant-temperature stage.
Thus, the 313–373 K range in the plot at the first
heating corresponded to vaporization of one of the
borazine decomposition products, whereas that part of
the curve was practically linear during the second and
third heating stages, evidencing the decomposition of
that product (Fig. 2). At the second heating stage,
another less volatile decomposition product was
evaporated at about 420 K; it was decomposed before
the third heating as well. Those products, in our
opinion, might be oligoborazines. After the third
constant-temperature stage and freezing of the system,
by the log (p/p ) = –3.8(±0.4)/T+12(±2) equation over
0
2
00–217 K temperature range. The high inaccuracy of
the fitting was caused by low measured vapor pressure,
narrow temperature range, and insufficient number of
low-temperature experimental points. The obtained
thermodynamic parameters of borazine sublimation
(
1
Δ H° = 31.9±3.5 kJ/mol and Δ S° = 102.7±
sub 209 sub 209
–
1
–1
6.4 J mol K ) are underestimated compared with
those calculated from the reference data as the sum of
the thermodynamic parameters of evaporation [11] and
melting [5]: Δ H° = 39.9±0.6 kJ/mol and Δ S° =
sub 209
–1
sub 209
–
1
1
38.4±0.1 J mol K .
Thermal decomposition in unsaturated vapor.
The experiment was condected as follows: a working
volume was frozen by liquid nitrogen to 77 K, then
heated at 2 K/min rate, and kept at constant
temperature of 373 and 473 K. After completion of the
each incubation stage, another freezing of the working
volume with liquid nitrogen followed by heating to
room temperature was performed.
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 88 No. 12 2018

2
478
ZAVGORODNII, TIMOSHKIN
a
Amounts of components in the system during borazine thermal decomposition in unsaturated vapor
Experimental stage
t, h
2 3
n, µmol n(BZ), µmol n(H ), µmol n(NH ), µmol n(olig), µmol
Start of the first constant-temperature stage
End of the first constant-temperature stage
7.76
0.967(5)
0.869(5)
0.803(7)
0.803(7)
0.765(6)
0.765(6)
0.013(2)
0.134(4)
0.134(4)
0.202(3)
0.202(3)
0.347(4)
0.0
0.084(4)
0.049(9)
0.122(9)
0.043(7)
0.043(7)
75.40 1.006(5)
0.020(4)
0.020(4)
0.106(3)
0.106(3)
0.068(5)
Start of the second constant-temperature stage 101.54 1.078(5)
End of the second constant-temperature stage 202.50 1.116(5)
Start of the third constant-temperature stage 292.26 1.118(5)
b
b
End of the third constant-temperature stage 418.47 1.152(6)
0.687
0.050
a
b
(
t) is the time from the experiment start. Estimated values.
an exponential curve segment appeared in the plot in
the 160–210 K range, which could be explained by the
presence of one more component in the system (likely
ammonia, the formation of which upon borazine
decomposition has been mentioned in [12]).
pressure and temperature at 373 and 473 K was
analyzed. The system volume remained constant during
the experiment, so the change in the monitored p/T
ratio might be taken as a measure of relative reaction
rate. The relative rate constant was calculated from the
experimental relative rate with regard for the decrease
in borazine amount in the system due to decom-
position. It was found that the relative reaction rate
remained practically constant upon the heating by 100 K.
The following model was suggested to calculate the
gas mixture composition in the tensimeter working
volume:
(
1) the gas phase contained hydrogen, ammonia,
borazine, and oligoborazines;
2) the mixture components were evaporated in
sequence;
The relative rate constants calculated from the
determined partial pressures of the components
(
–
1
–1
equaled k373 = 6.4±0.7 min and k473= 5.8±0.4 min
for the first and second constant-temperature stages,
respectively. The estimate of the activation energy
using Arrhenius equation gave –0.6±1.6 kJ/mol. The
closeness of the activation energy to zero indicated that
the borazine diffusion to the vessel wall was the rate-
limiting stage of the process. Thus, decomposition of
gaseous borazine occurred at active surface sites of the
system rather than in the gas volume and was limited
by the diffusion, otherwise the reaction rate would be
limited by the chemical reaction rate, and the activa-
tion energy would be nonzero. That conclusion agreed
with the experimental fact that liquid borazine is
(
3) decomposition of the mixture components
occurred only during the constant-temperature stages.
The Table contains both the total amount n of the
gaseous substances in the system determined directly
from the experimental data and the partial amounts of
the components in the mixture: borazine n(BZ), hyd-
rogen n(H ), ammonia n(NH ), and borazines n(olig),
2
3
calculated in scope of the suggested model. The changes
in the amounts of the mixture components after each
constant-temperature stage are also presented in the
table.
1
decomposed even at room temperature. The H NMR
spectrum of a borazine solution in deuterobenzene
contained a molecular hydrogen peak at 4.42 ppm.
It should be noted that the amount of borazine
calculated from the specimen weight was 0.970±
.002 mmol, thus a partial decomposition (≈10%) of
0
the compound occurred already when transferring the
sample into the working volume.
In summary, we demonstrated that gaseous borazine
was decomposed on active sites of the reaction vessel
surface rather than in its volume at 373–473 K. The
diffusion to the surface was the limiting stage. The
decomposition was accompanied by the formation of
volatile intermediates, probably oligoborazines. At
temperatures below 473 K, the solid decomposition
Basing on the obtained data, we calculated the ratio
of amounts of the evolved hydrogen and decomposed
borazine (1.8±0.2 at 373 and 473 K, consistent with
the value of 1.8 at 613 and 654 K obtained earlier [13]).
Study of the borazine decomposition kinetics.
The kinetics of the change of the ratio between total
product of [B N H ] composition was formed.
3 3 2.4±0.4 n
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 88 No. 12 2018

BORAZINE THERMAL DECOMPOSITION IN UNSATURATED VAPOR
2479
EXPERIMENTAL
ACKNOWLEDGMENTS
Ammonium sulfate (“chemical pure” grade) and
sodium borohydride (Sigma-Aldrich, 99%) were purified
by heating to 200°С under continuous evacuation.
Tetraglyme (Sigma-Aldrich, 99%) was dried over
The authors express their gratitude for the
measurements to the SPbGU Resource Center for
Magnetic Resonance Research.
CONFLICT OF INTEREST
No conflict of interest was declared by the authors.
REFERENCES
CaCl , distilled, and degassed under vacuum via a
2
single freezing–unfreezing.
1
The H NMR spectrum was obtained using a Bruker
Avance 400 instrument operating at 400.13 MHz.
1
2
3
4
. Stock, A. and Poland E., Ber. (D), 1926, vol. 59,
p. 2215.
. Sano, M. and Aoki, M., Thin Solid Films, 1981, vol. 83,
p. 247. doi:10.1016/0040-6090(81)90674-X
. Arya, S.P.S. and D’Amico, A., Thin Solid Films, 1988,
vol. 157, p. 267. doi 10.1016/0040-6090(88)90008-9
. Frey, J.E., Marchand, G.M., and Bolton, R.S., Eur. J.
Inorg. Chem., 1982, vol. 21, p. 3239. doi 10.1021/
ic00138a063
Synthesis of borazine was carried out according to
Wideman and Sneddon procedure [14] in a sealed
evacuated system under continuous evacuation via the
reaction:
3
(NH ) SO + 6NaBH
4 2 4 4
1
40−160°C, 3 h
2
B N H + 3Na SO + 18H .
3 3 6 2 4 2
tetraglyme
The synthesized product was purified by vacuum
1
distillation. H NMR spectrum (C D ), δ, ppm: 4.46 q
5. Lebedev, B.V. and Kulagina, T.G., Chem. Thermodyn.,
6
6
1
8
991, vol. 23, p. 1097. doi 10.1016/S0021-9614(05)
0140-5
(
[
3H, BH, J = 140 Hz), 5.08 t (3H, NH, J = 53.0 Hz)
15, 16]. It should be noted that the NMR spectra were
6
7
. Laubengayer, A.W. and Scaife, C.W.J., J. Chem. Eng.
Data, 1966, vol. 11, p. 172. doi 10.1021/je60029a012
. Volkov, V.V., Myakished, K.G, and Il’inchik, E.A.,
Khim. v Interesakh Ustoichivogo Razvitiya, 2009,
vol. 17, p. 233.
. Lei, Y., Wang, Y., Song, Y., and Deng, C., Ceram. Int.,
2011, vol. 37, p. 3005. doi 10.1016/j.ceramint.2011.04.021
. Kazakov, I.V. and Timoshkin, A.Yu., Russ. J. Inorg.
Chem., 2012, vol. 57, p. 557. doi 10.1134/
S0036023612040110
recorded in a sealed ampoule in this study, which
allowed us to detect the presence of molecular
hydrogen (δ 4.42 ppm).
Measurements in the saturated borazine vapor.
The membrane null-manometer [10] was used in the
experiment. A compartment calibrated by volume and
containing an ampoule with borazine was soldered to
the working volume, and the system was evacuated
with a 2NVR-5DM pump. Then the ampoule was
opened, and borazine was distilled into the working
volume (V = 58.43 mL including the calibrated
compartment) frozen with liquid nitrogen. After that,
the volume with ampoule pieces was unsealed, and the
device was placed in a thermostat. The pressure was
measured with an MRCh-3 mercury manometer.
Temperature of the working volume was controlled by
two chromel-alumel thermocouples and an ShCh-1516
digital millivoltmeter.
8
9
10. Davydova, E.I., Sevastianova, T.N., Suvorov, A.V., and
Timoshkin, A.Y., Coord. Chem. Rev., 2010, vol. 254,
p. 2031. doi 10.1016/j.ccr.2010.04.001
11. Wiberg, E. and Boltz, A., Ber. (D), 1940, vol. 73, p. 209.
12. Wilson, R.G., J. Appl. Phys., 1973, vol. 44, p. 5056. doi
10.1063/1.1662087
13. Laubengayer, A.W., Moews, P.C., Jr., and Porter, R.F.,
J. Am. Chem. Soc., 1961, vol. 83, p. 1337. doi 10.1021/
ja01467a020
1
4. Wideman, T. and Sneddon, L. G., Inorg. Chem., 1995,
Measurements in the unsaturated vapor. The
automatic tensiometer [17] was used for the measu-
rements. A weighed borazine sample (78 mg) was
distilled under vacuum into the working volume
cooled with liquid nitrogen (44.5 mL), and then it was
unsealed. Three series of measurements were carried
out. Each constant-temperature stage was carried out
for 3–5 days, and the working volume was cooled and
heated within several hours. Thus, heating and cooling
took 2–4% of the experiment time.
vol. 34, p. 1002. doi 10.1021/ic00108a039
5. Jeon, J.K., Nghiem, Q.D., Kim, D.P., and Lee, J.,
1
J. Organomet. Chem., 2004, vol. 689, p. 2311. doi
1
0.1016/j.jorganchem.2004.04.019
1
1
6. Yang, X., Zhao, L., Fox, T., Wang, Z.-X., and Berke, H.,
Angew. Chem. Int. Ed., 2010, vol. 49, p. 2058. doi
10.1002/anie.200906302
7. Doinikov, D.A., Kazakov, I.V., Krasnova, I.S., and
Timoshkin, A.Y., Russ. J. Phys. Chem. (A), 2017,
vol. 91, p. 1603. doi 10.1134/S0036024417080088
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 88 No. 12 2018