Vaporization and thermal decomposition of the AlI3Py adduct

Article abstract of DOI:10.1023/A:1023442708021

Vaporization of the AlI₃Py adduct was studied by mass-spectrometry and tensimetry. Thermodynamic characteristics of vaporization and gas-phase dissociation of the complex were determined.

Full text of DOI:10.1023/A:1023442708021

Russian Journal of General Chemistry, Vol. 72, No. 12, 2002, pp. 1874 1877. Translated from Zhurnal Obshchei Khimii, Vol. 72, No. 12, 2002,
pp. 1980 1984.
Original Russian Text Copyright
2002 by Timoshkin, Suvorov, Misharev.
Vaporization and Thermal Decomposition of the AlI3Py Adduct
A. Yu. Timoshkin, A. V. Suvorov, and A. D. Misharev
St. Petersburg State University, St. Petersburg, Russia
Received April 3, 2001
Abstract Vaporization of the AlI3Py adduct was studied by mass-spectrometry and tensimetry. Thermo-
dynamic characteristics of vaporization and gas-phase dissociation of the complex were determined.
This work is a continuation of our research into
donor acceptor complexes formed by Group III
halides with pyridine, which is one of the strongest
n-donors. Earlier we focused mainly on chloride and
bromide complexes.
[M2I5]+
[MI3Py]+
9.6
[MI3]+
0.9
54.0
AlI3Py
GaI3Py [3]
0.08
0.4
[MI2Py]
100
40.5
[MI2]
46.3
100
[MPy]+ [Py]+
+
+
Reliable experimental data on the thermodynamics
of gas-phase dissociation of iodide complexes are
inadequate to reveal the effect of the nature of halogen
atoms on the energy of donor acceptor bonds. It was
shown [1] that heating above 300 C produces irrever-
sible ammonolysis of the AlI3NH3 complex into
AlN(s.) and HI(gas), and thus thermodynamic charac-
teristics of its dissociation into components could not
be determined. Replacement of ammonia by a hetero-
cyclic aromatic donor (pyridine) makes the complexes
more stable pyrolytically [2]. This allowed thermo-
dynamic dissociation characteristics of a gallium
triiodide complex of pyridine to be determined in the
range 382 424 C [3]. However, as shown earlier [3],
vaporization of the GaI3Py adduct is accompanied by
its dissociation already in saturated vapor. For this
reason, the weight of the complex could not refined
under the experimental conditions, and the resulting
dissociation enthalpy involved a considerable inac-
curacy ( 10 kJ/mol).
AlI3Py
GaI3Py [3]
6.0
7.0
31.3
46.5
The most typical differences show that AlI3Py is
much more stable than GaI3Py. The peaks of ions
containing the Al N bond are very strong. The in-
tensity of the molecular ion peak is 10%, and the
base peak (100 %) is formed by an [AlI2Py]+ ion. This
Table 1. Mass-spectra of vapor over the AlI3Py complex
Ion
m/z
I, % (120 C) I, % (170 C)
[Al]+
[I]++
27
63.5
64
78
79
56.7
2.8
1.9
7.6
31.3
6.0
23.9
43.3
6.2
0.7
1.6
44.4
2.9
2.3
6.0
24.7
6.2
22.2
42.0
4.9
0.6
1.0
[HI]++
[Py H]+
[Py]+
[AlPy]+
[I]+
106
127
128
154
155
179
180
232
233
254
281
308
360
408
439
487
[HI]+
According to our ab initio and DFT quantum-
chemical calculations [4], the enthalpies of gas-phase
dissociation of aluminum iodide complexes of NH3,
PH3, AsH3 are higher than those of gallium iodide
complexes by 29.4, 15.9, and 13.3 kJ/mol. Therefore,
it was expected that AlI3Py would be more stable in
vapor than GaI3Py.
[AlI]+
[HAlI]+
1.9
1.9
0.7
0.7
1.4
1.9
0.6
0.6
[AlI(Py H)]+
[AlIPy]+
[I2]+
This work is devoted to an experimental study of
AlI3Py vaporization and its thermal dissociation in
vapor. Table 1 presents the mass spectra of the result-
ing species containing aluminum, iodine, and pyridine
(fragment ions of the latter are omitted). As seen, the
mass spectrum does not undergo essential changes on
heating. These data allow us to compare the mass
spectra of aluminum and gallium iodide adducts [3].
[AlI2]+
46.3
1.2
100.0
0.9
42.0
0.4
100.0
0.8
+
[Al2I2]
[AlI2Py]+
[AlI3]+
[AlI22Py]+
[AlI3Py]+
0.3
9.6
0.03
8.9
1070-3632/02/7212-1874$27.00 2002 MAIK Nauka/Interperiodica

VAPORIZATION AND THERMAL DECOMPOSITION OF THE AlI3Py ADDUCT
Table 2. Characteristics of AlI3Py saturated vapor
1875
logp (mm) = a/T + b
vap
Run Weight,
Temperature
range, K
Tav,
K
Hvap(Tav), Svap(Tav),
kJ/mol
Tb,
K
H
,
298
1
no.
mg
J mol 1 K
kJ/mol
a
b
1
2
3
>400
506 688 (96 points) 3464 38 7.503 0.062 598 66.3 0.7
88.5 1.2
92.4 1.5
78.2 1.6
749
747
780
78.7 4.5
81.3 4.7
71.6 4.7
57.55 516 658 (62 points) 3604 49 7.706 0.080 587 69.0 0.9
38.55 501 661 (52 points) 3187 49 6.964 0.082 581 61.0 0.9
ion contains the Al N bond and undoubtedly ori-
ginates from AlI3Py. At the same time, the base peak
in the mass spectrum of GaI3Py is formed by a [GaI2]+
ion which may originate from both GaI3Py and
GaI3the. The intensity of the [MI3]+ ion is insignifi-
cant (0.9%) in the system with Al and considerable
The calculated thermodynamic characteristics of
the vaporization are given in Table 2. Note that the
first and second experimental series fairly fit each
other, whereas the third series fails to fit the first two
series, apparently, because a much smaller sample was
used. The data were reduced to standard conditions
(54%) in the system with Ga; in the case of aluminum, by Watson’s formula [6], the normal boiling point
the dimeric ions [Al2I5]+ are entirely absent from the
mass spectrum.
was estimated from tensimetric data, and the critical
temperature (Tcr) was estimated by the formula Tav
=
1.41 Tb + 66(Tb is boiling point) [3]. The temperature
correction was 12.4 kJ/mol, and its inaccuracy was
taken to be 30%; thus the standard enthalpy of va-
The fragmentation with loss of iodine is much
more favorable than with loss of pyridine, and the
presence of very strong [GaIn]+ ions and very weak
[Ga2I5]+ ions definitely points to a dissociation of the
GaI3Py complex, and the absence of respective ions
in the case of AlI3Py, to a higher strength of the Al N
bond compared to Ga N. The resulting data suggest
that AlI3Py undergoes quite a very minor thermal
dissociation on vaporization, if any. These data agree
qualitatively with the bond-strength order Al N >
Ga N, as calculated in [4].
Note that the [AlI22Py]+ and [Al2I2]+ ions, which
have the lowest intensities at 170 C, can be formed
by ion-molecular reactions directly in the ion source,
because the concentrations of the parent ions are
extremely high (100, 46, and 56%) [schemes (1) (3)].
porization
H2v9a8p(AlI3Py)
was estimated at
80 5 kJ/mol.
Unsaturated vapor studies were performed using
samples of 20 80 mg. The temperature depen-
dences of the saturated and unsaturated vapor pres-
sures of the complex is given in Fig. 1. Repeated
heating of the complex up to 410 C (the upper opera-
tion limit of a molybdenum glass) showed that the
complex vaporizes without decomposition and does
not dissociate up to 410 C. Note that data obtained in
the first and subsequent heating cooling cycles nicely
fit each other. Experiments in the range 400 410 C
within 6 h showed that the concentration of the com-
pound in the membrane chamber (n/V = p/RT)
remains unchanged within the experimental error, i.e.
no processes occur in the system at these temperatures.
We note that the degree of GaI3Py dissociation into
components at these temperatures is 15% [3].
[AlI2Py]+ + Py = [AlI22Py]+,
[AlI2]+ + Al = [Al2I2]+,
AlI2 + [Al]+ = [Al2I2]+.
(1)
(2)
(3)
Thermodynamic dissociation characteristics of the
complex were determined in experiments in a quartz
device. These experiments showed that the reversible
thermal dissociation of the complex into components
becomes noticeable above 415 C. The process is
reversible up to 464 C, as follows from the fact that
heating and cooling data nicely fit each other .
This circumstance makes processes (1) (3) much
more probable than other possible ion-molecular
reactions.
Vaporization studies were performed by a static
method with a membrane zero gage [5]. To study the
saturated vapor of the AlI3Py adduct, we used a large
sample (400 500 mg). The vapor pressure of the
complex in the range 507 688 K is described
by Eq. (4).
The equilibrium dissociation constant of a 1:1
complex over the entire temperature range of un-
saturated vapor was calculated by Eq. (5).
log p = (3464 38)/T + (7.503 0.062), mm. (4)
Kdeis = ( p)2/(p0
2 p).
(5)
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 72 No. 12 2002

1876
TIMOSHKIN et al.
p (AlI3Py) values were determined by the initial points
100
p, mm
of thermal expansion of the complex, i.e. from the
point of transition into unsaturated vapor up to 410 C.
80
60
The temperature dependence of the dissociation
constant is rather intricate, and the entire set of points
cannot be used to calculate thermodynamic characte-
ristics. The upper temperature limit is 475 C, since at
higher temperatures noticeable pyrolysis occurs. As
that temperature is maintained constant, the pressure
does not remain constant and becomes to increase
appreciably (Fig. 1). For the remaining points, the
log Kdeis = f(1000/T) dependence is nonlinear. There-
fore, the dissociation enthalpy and entropy of the
complex were calculated by the step-by-step method
recommended in [7].
40
20
0
100 200 300 400 500 600
T, C
The resulting log Kdeis values (Fig. 2) were treated
by the least-squares procedure, and thus obtained
characteristics of gas-phase dissociation of the com-
plex are given in Table 3. The temperature range, in
which the process can be studied, is rather narrow
( 75 C). The considerable divergence of the cal-
culated Hdis values results from the fact that the
maximum p values in all the experiments do not
exceed 6 mm, and for run no. 1 it is less than 2 mm.
Therefore, for averaging we used data of run nos. 2
and 3. It was found that the dissociation enthalpy of
the AlI3Py complex is 156 7 kJ/mol, which is higher
than that of GaI3Py (140 7 kJ/mol) [3].
Fig. 1. Temperature dependence of vapor pressure for the
AlI Py adduct. For the field of unsaturated vapor, data
3
for the first four heating cooling cycles are given (run
no. 1, Table 3). (Crosses, triangles, circles, and plusses).
First, second, third, and fourth heatings, respectively;
(rhombs) saturated vapor pressure of AlI Py, and (straight
3
line) p (AlI Py).
3
log Kedis
0.4
0.6
Thus, this work showed that AlI3Py is much more
stable with respect to dissociation into components
than GaI3Py. However, the features of the system
make the gas-phase dissociation enthalpy and entropy
impossible to determine reliably.
0.8
1.0
1.2
1.4
Apart from dissociation, we also studied irrever-
sible pyrolysis of AlI3Py. The process starts at 475 C
(Fig. 1) and proceeds slowly. Complete decomposi-
tion of the complex with deposition of a carbon layer
on walls of the device was reached only upon keeping
the system above 800 C within 11.5 h. We estimated
the activation energy of the process at 210 kJ/mol in
the range 498 512 C, which appeared to be much
lower than the activation energy of pyridine pyrolysis
(439 8 kJ/mol) [8].
1.6
1.8
2.0
1.30
1.35
1.40
1.45
1000/T, K
AlI Py on reciprocal
1.50
1
dis
Fig. 2. Dependence of log K
On completion of the experiment, the pressure in
the system increased 3.5 times as compared to the
pressure of the complex, and the membrane chamber
was covered by a shiny layer of carbon. Bulk gas
e
3
temperature (run numbers are the same as in Table 3).
(Crosses, rhombs, and circles) Run no. 1, 2, and 3,
respectively
condensed at
100 C, and at a liquid nitrogen
Here p0 is the total pressure in the system (mm Hg),
p = p0 p is the difference between total pressure
p0 and hypothetical pressure p of thermal expansion
of vapor of undissociated complex (Fig. 1). The
temperature the pressure was only 4 0.5 mm. Che-
mical analysis of the frozen gas showed that it was
HI. Its pressure was 2.8 0.5 times higher than the
pressure of the uncondensed gas. We could not prove
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 72 No. 12 2002

VAPORIZATION AND THERMAL DECOMPOSITION OF THE AlI3Py ADDUCT
1877
Table 3. Characteristics of AlI3Py dissociation into components (unsaturated vapor)
Weight, Volume
of com- of devi-
plex, mg ce, ml
logKdeis (mm) = a/T + b
Run
no.
Temperature
range, K
Tav,
K
Hdis(Tav),
kJ/mol
Sdis(Tav),
Gdis
,
723
kJ/mol
1
J mol 1 K
a
b
1
2
3
28.45
85.35
68.1
54.32 697 745 (18 points) 8170 860 10.05 1.19
55.54 676 747 (57 points) 7413 408 9.498 0.573 707 141.8 7.8 126.6 11.0
54.22 671 735 (40 points) 9305 535 12.173 0.758 703 178.0 10.2 171.8 14.5
721 156.3 16.5 137.2 22.9
57.1
50.3
49.5
50.0
Average over run
nos. 2 and 3
671 747 (97 points) 8149 332 10.533 0.468 709 155.9 6.4 146.4 9.0
the presence of AlN in solid products, however, on
the assumption that the remaining gas is hydrogen,
we can represent the pyrolysis process by the follow-
ing scheme.
perature was controlled by two thermocouples
(chromel alumel), and the thermal emf was measured
with an ShCh-1516 digital voltmeter, the accuracy of
temperature measurements was 0.05 C. The volume
of the membrane chamber was found from the dif-
ference in the weights of water-filled and empty
systems. The accuracy of volume measurements was
0.02 ml, i.e. 0.07 0.04% for a 30 50-ml device.
AlI3NC5H5(gas) = AlN(s.) + 5C(s.) + 3HI(gas) + H2(gas).
This work showed that the adduct of aluminum tri-
iodide with pyridine vaporizes as monomeric AlI3Py
molecules which are stable in vapor up to 415 C. We
determined for the AlI3Py complex thermodynamic
characteristics of vaporization ( H5v9ap8 66.3 0.7 kJ/mol,
The composition of vapor over the adduct was
determined on an MKh-1321 mass spectrometer at an
ionizing energy of 70 eV and inlet batch temperatures
of 120 and 170 C.
vap
598
1
S
90.5 2 J mol 1 K ) and gas-phase dissociation
dis
dis
1
( H
156 7 kJ/mol,
S
146 9 J mol 1 K ).
709
709
REFERENCES
EXPERIMENTAL
1. Trusov, V.I., Cand. Sci. (Chem.) Dissertation, Lenin-
grad, 1974.
In view of the high hygroscopicity of the com-
pounds under study, all operations, beginning with
synthesis and purification of components up to filling
tensimetric devices, were carried out in sealed eva-
cuated all-glass systems completely eliminating
contact with atmospheric air and moisture. Pyridine
(chemically pure grade) was distilled on a 40 TT
column, degassed by continuous pumping, and deeply
dried over a mixture of preliminarily calcined NaX,
NaA, and CaA zeolites. The purity of pyridine was
controlled on a Tsvet-560 chromatograph.
2. Grigor’ev, A.A., Cand. Sci. (Chem.) Dissertation,
Leningrad, 1986.
3. Timoshkin, A.Yu., Grigor’ev, A.A., and Suvorov, A.V.,
Zh. Obshch. Khim., 1995, vol. 65, no. 10, p. 1634.
4. Timoshkin A.Y., Suvorov A.V., Bettinger H.F., and
Schaefer, H.F. III, J. Am. Chem. Soc., 1999, vol. 121,
no. 24, p. 5687.
5. Suvorov, A.V., Termodinamicheskaya khimiya paro-
obraznogo sostoyaniya (Thermodynamic Chemistry of
Vaporous State), Leningrad: Khimiya, 1970.
Aluminum triiodide was synthesized from elements
and additionally purified by repeated vacuum sublima-
tion. The AlI3Py adduct was synthesized from the com-
ponents in a vacuum and purified by double distilla-
tion at 180 C. The resulting complex is a colorless
crystalline substance (mp 110 112 C [9]), and its
melt tends to supercool. Before introducing into the
membrane chamber, the adduct was heated at 400 C
for no less than 1 h and distilled at 180 C.
6. Reid, R.C., Prausnitz, J.M., and Sherwood, T.K., The
Properties of Gases and Liquids, New York: McGraw-
Hill, 1977, 3rd ed.
7. Shubaev, V.A., Cand. Sci. (Chem.) Dissertation, Lenin-
grad, 1971.
8. Kiefer, J.H., Zhang, O., Kern, R.D., Yao, J., and Jur-
sic, B., J. Phys. Chem. A, 1997, vol. 101, no. 38,
p. 7061.
The pressure was measured with an MChR-3
mercury gage with an accuracy of 0.1 mm. The tem-
9. Wilson, J.W. and Worrall, I.J., J. Chem. Soc. A, 1968,
no. 2, p. 316.
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 72 No. 12 2002