Tensimetric investigation of the CrCl3-Cl2 system in the temperature range of 600-1200 K

Article abstract of DOI:10.1007/s11172-005-0006-0

The sublimation pressure of chromium trichloride was measured by the static method with a quartz membrane-gauge manometer in the temperature range of 875-1230 K. An approximating equation for the sublimation pressure vs. temperature was found. The enthalpy (259.4±4 kJ mol^-1) and the entropy (224.2±3.5 J mol^-1 K^-1) of sublimation at 298 K were calculated. For the process 2 CrCl3(g) + Cl2(g) = 2 CrCl4(g), the following values were obtained: Δ_r H°₂₉₈ = -207.1±11.6 kJ mol^-1 and Δ_r S°₂₉₈ = -173.6±10 5 J mol^-1 K^-1.

Full text of DOI:10.1007/s11172-005-0006-0

Russian Chemical Bulletin, International Edition, Vol. 53, No. 8, pp. 1621—1624, August, 2004
1621
Tensimetric investigation of the CrCl —Cl system
3
2
in the temperature range of 600—1200 K
ꢀ
L. N. Zelenina, Z. I. Semenova, V. A. Titov, and T. P. Chusova
A. V. Nikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russian Academy of Sciences,
3
prosp. Akad. Lavrent´eva, 630090 Novosibirsk, Russian Federation.
Eꢀmail: chu@che.nsk.su
The sublimation pressure of chromium trichloride was measured by the static method with
a quartz membraneꢀgauge manometer in the temperature range of 875—1230 K. An approxiꢀ
mating equation for the sublimation pressure vs. temperature was found. The enthalpy
259.4±4 ꢁk mol–1) and the entropy (224.2±3.5 k mol
–1
K
–1
) of sublimation at 298 K were
(
calculated. For the process 2 CrCl (g) + Cl (g) = 2 CrCl (g), the following values were
3
2
4
–
1
–1 –1
obtained: ∆ H° = –207.1±11.6 ꢁk mol and ∆ S° = –173.6±10 5 k mol
K .
r
298
r
298
Key words: chromium chlorides, vapor pressure, enthalpy, entropy, sublimation, dissoꢀ
ciation.
Anhydrous chromium trichloride is widely used as
starting material in inorganic and organoelement syntheꢀ
ses and to produce chromium coatings. The purity of
chromium is of prime importance for these purposes. Since
crystalline chromium trichloride is purified by sublimaꢀ
tion in a chlorine flow, the temperature dependence of
mium trichloride is very large (Fig. 1), and the composiꢀ
tions of the equilibrium gas and condensed phases are not
entirely clear as yet. The sublimation of chromium chloꢀ
ride at 900—1500 K is assumed to give rise to several
equilibria (Scheme 1).
CrCl (s) vaporization is needed for determining the optiꢀ
mal conditions of this process.
3
Scheme 1
Numerous publications have been devoted to the proꢀ
CrCl (s)
CrCl (g)
cesses that taꢁe place on heating CrCl in vacuo or under
3
3
3
an inert gas.1
—13
Nevertheless, the scatter of published
2
CrCl (s)
CrCl (s) + CrCl (g)
3
2
4
data on the saturated vapor pressure above solid chroꢀ
CrCl (s)
CrCl (g)
2
2
ln(p/Torr)
CrCl (g)
CrCl (g) + Cl (g)
4
2
2
1
6
The process taꢁing place on heating of chromium
trichloride in a chlorine stream
2
3
4
5
4
2
0
2
2
CrCl (g) + Cl (g)
2 CrCl (g),
4
3
2
is the subject of two inconsistent publications.7,11
This study is devoted to determination of the saturated
vapor pressure above crystalline chromium chloride in
the absence and in the presence of chlorine gas aimed at
eliminating the contradictions found in the literature.
–
.85 0.90 0.95 1.00 1.05 T –1•10 /K–1
3
0
Experimental Procedure
Fig. 1. Saturated vapor pressure (p) above solid chromium
trichloride vs. temperature (T ) according to various authors:
our data (Eq. (1)) (1), Ref. 1 (2), Ref. 3 (3), Refs 5, 6 (4),
Ref. 7 (5).
The temperature dependences of the vapor pressure above
pure CrCl3 and in the presence of Cl2 were measured by the
static method using a Noviꢁov—Suvorov quartz membraneꢀ
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 8, pp. 1561—1564, August, 2004.
066ꢀ5285/04/5308ꢀ1621 © 2004 Springer Science+Business Media, Inc.
1

1622
Russ.Chem.Bull., Int.Ed., Vol. 53, No. 8, August, 2004
Zelenina et al.
gauge manometer.1 The main characteristics of the setup deꢀ
4
ln(p/Torr)
run 1
1
5
scribed previously are as follows: the limiting error of temperaꢀ
ture measurements estimated by calibration against mercury,
naphthalene, and argon did not exceed ±1 K at 900 K and
increased to ±2 K at 1200 Kꢂ the accuracy of temperature mainꢀ
tenance was ±0.1 Kꢂ the maximum error in the pressure meaꢀ
surement caused by the diaphragm sensitivity and by the errors
in the corrections for the irreversible zero drift of the pressure
gauge varied from one experiment to another, being 0.2—5 Torr.
The P—T data were obtained by the temperatureꢀstep method
in the 573—1230 K range both with heating and with cooling.
The time of equilibration was varied from 50 h at low temperaꢀ
tures to 10 h at high temperatures. The pressures measured from
low to high temperatures and bacꢁwards were identical at the
same termperature. This guaranteed the attainment of equiꢀ
librium.
8
6
Eq. (1)
run 2
run 3
run 4
run 5
run 6
4
2
0
–2
Four samples of chromium trichloride prepared and purified
by different methods were used. According to preliminary
tensimetric experiments, all samples exhibit enhanced gas evoꢀ
lution on heating and require additional purification, which
was carried out in the following way. The sample (0.5—1 g)
was pacꢁed into 0.5—1 mL quartz tubes using an argonꢀfilled
dry box (with P O as the drying agent). The tubes were
T –1•103/K–1
0.85 0.90 0.95
1.00
1.05
Fig. 2. Saturated vapor pressure (p) above solid chromium
trichloride vs. temperature (T ) found in this worꢁ.
2
5
Results and Discussion
evacuated, sealed, and placed in a special 10—15ꢀmL quartz
reactor. The reactor with the tube was heated to 773—823 K
with continuous evacuation (rough exhaust, liquid nitrogenꢀ
cooled zeolite column) to remove the adsorbed gases and waꢀ
ter vapor from the reactor walls. Then the tube with the subꢀ
stance was broꢁen inside the reactor and the sample was heated
for ∼ 20 h with a continuously operating pump. During this peꢀ
riod, volatile fractions were deposited on the inner side of the
quartz tube connecting the worꢁing volume of the device to
the pump. Then the reactor was cooled and the worꢁing volume
of the reactor with the sample was sealed off in vacuo and heated
to the maximum temperature of the tensimetric experiment
Altogether six independent experiments on determiꢀ
nation of the saturated vapor pressure of CrCl3 in the
temperature range of 573—1230 K were carried out. The
ratio of the sample weight to the manometer worꢁing
–1
chamber was 0.9—10.5 g L . The total number of experiꢀ
mental points involved in the processing was 73. After
subtracting the residual pressure, the saturated vapor presꢀ
sure did not depend on the ratio of the sample weight to
the volume of the diaphragm chamber to within the error
of measurements (Fig. 2). This fact indicates that the
system under study is monovariant and allows combined
statistical processing of the data from all experiments.
(
1073—1230 K). After cooling, the worꢁing chamber was opened,
evacuated, and connected to a quartz membraneꢀgauge maꢀ
nometer. Then the valve separating the worꢁing chamber from
the diaphragm was broꢁen by a magnetic striꢁer, and the meaꢀ
surements were started. All of the authors of experimental studꢀ
ies noted the appearance of a residual pressure during the exꢀ
periments. We observed this effect as ∼ 8 Torr at 448 K. After
completion of the experiments, the substance in the worꢁing
chamber represented wellꢀformed druses of bright crimson crysꢀ
tals of CrCl3.
The calculations were carried out by a ꢁnown proceꢀ
dure17 using the objective function
,
According to atomic emission analysis, after the complete
static study including sample preparation and pressure measureꢀ
ment, the samples contained Ni (0.15—0.19% (w/w)), Fe
exp
where N is the number of the experimental pointsꢂ pi
the experimental pressureꢂ pi
is
calc
is the pressure calculated
(
(
0.02—0.09% (w/w)), and one sample was found to contain Si
0.54 % (w/w))ꢂ the contents of Ag, Be, Mn, In, Cd, Ga, Cu,
by the accepted physicochemical modelꢂ ∆pi and ∆Ti are
the maximum errors of pressure and temperature meaꢀ
surements, respectively.
Co, Bi, Pb, Tl, Mo, Au, Ge, Sn, Ti, Al, Mg, Zn, Pt, Ba, As,
and Te were below the detection limit.
Chlorine was prepared by thermal decomposition (833 K) of
platinum dichloride whose chemical analysis is given by (% w/w):
Pt, 73.42ꢂ Cl, 26.47ꢂ calculated (% w/w): Pt, 73.34ꢂ Cl, 26.66.
Gaseous Cl2 was condensed on cooling in a special quartz tube
The use of this objective function in the data processꢀ
ing provides reliable estimates for the desired parameters.
The search for the minimum of the function ϕ was based
on a program implementing the modified Newton—Gauss
algorithm with a step selected along a direction. The erꢀ
rors of the desired values were calculated taꢁing into acꢀ
count the Student´s coefficients for a 95% confidence
interval.
with a detachable tip. The used Cl was free from HCl, O2,
2
1
6
and H O. The weight of the PtCl sample was selected in such
2
2
a way that the chlorine pressure in the diaphragm zeroꢀpressure
manometer was ∼ 100 Torr at ∼ 20 °C.

Tensimetric investigation of the CrCl —Cl system
Russ.Chem.Bull., Int.Ed., Vol. 53, No. 8, August, 2004
1623
3
2
Table 1. Entropy (S° ) and temperature dependences of the heat capacity of chromium chloride as C °(T ) = a + bT +
2
98
p
+
c/T + d/T 2 + eT 2
Compound
S°298/k mol—1 K–1
a
b
c
d
e
CrCl (s)1
9
122.9±1.8
347.0±3
371.9±2.5
86.075
78.442
119.96
0.03162
1.1208•10
–1.327•10
–941.92
—
—
3
2
2
0
0
—3
5.8861•103
3.1199•102
6
—7
—6
CrCl (g)
–1.4458•10
–1.3479•10
1.4748•10
4.2224•10
3
—2
6
CrCl (g)
4
3
Since a residual pressure has been discovered in alꢀ
most each of the experiments, the temperature depenꢀ
dence of the total pressure was described by the formula
publication can apparently be explained by the presence
of volatile components in the samples. When processing
3
the results, the researchers did not subtract the residual
pressure, i.e., the pressure that appeared due to impuriꢀ
ties. The underestimated data obtained in one more study7
pcalc = exp(–∆vapH°/RT + ∆vapS°/R) + cRT,
are probably due to errors in the temperature measureꢀ
where ∆vapH° and ∆vapS° are the enthalpy and the entropy
of vaporization, respectively, at the temperature T and с is
the ballast gas density.
The processing of the experimental data furnished an
empirical equation for the variation of the equilibrium
pressure above solid chromium trichloride vs. temperature
ments (∼ 15—20 °C understatement). In addition, the meaꢀ
7
surements were carried out under permanent heating,
and, perhaps, the equilibrium did not have enough time
to be established.
A
series
of
measurements
(11
points,
850 K ≤ T ≤ 1150 K) were carried out to study the subliꢀ
ln(p/Torr) = (30.499 – 28978/T )±2δ,
(1)
mation of chromium trichloride in a Cl atmosphere. The
2
tensimetric data were processed assuming that three moꢀ
lecular species, CrCl , CrCl , and Cl , exist in the gas
δ2 = 0.022 – 50.2/T + 28500/T 2 (875 K ≤ T ≤ 1230 K),
4
3
2
phase.
The set of independent chemical reactions was written
as follows:
the values of vaporization enthalpy and entropy, and
the temperatures where these values are minimum:
1
8
–
1
∆vapH°1135.5 = 240.9± 4.0 ꢁk mol , ∆vapS°1141.0
=
–
1
–1
1
98.4±3.6 k mol
The vaporization entropy coincides, to within the
determination error, with the value ∆sublS°1141.0
K .
CrCl (s)
CrCl (g),
(2)
(3)
3
3
=
2 CrCl (g) + Cl (g)
2 CrCl4.
3
2
1
97.6 k mol–1
K , which we calculated from the third
–1
law of thermodynamics using published data (Table 1)
and assuming that chromium trichloride is the only chroꢀ
miumꢀcontaining component of the gas phase over the
whole temperature range.
Figure 3 shows the calculated partial pressures for the
temperature range under interest. The thermodynamic
characteristics under standard conditions and the temꢀ
perature dependence of the equilibrium constant (K ) for
eq
reaction (3) are shown in Table 2. The good agreement
between the results obtained by two processing proceꢀ
dures demonstrates the adequacy of the physicochemical
CrCl (s)
CrCl (g)
(2)
3
3
This fact indicates that sublimation of CrCl3 is the
major vaporization process in the temperature range studꢀ
iedꢂ hence, the synthesis of highꢀpurity chromium trichloꢀ
ride does not require additional purification in a chlorine
flow. The thermodynamic parameters for reaction (2) unꢀ
der standard temperatures are presented below.
p/Torr
6
4
2
00
00
00
1
Processing method
∆sublH°
/
259.4±4
∆sublS°
298
ꢁk mol–1
298
/k mol K–1
–1
2
Second law of
thermodynamics
Third law of
224.2±3.5
4
259.2±4
224.1±3.6
3
thermodynamics
9
00
950
1000
1050
1100
T/K
It can be seen from Fig. 1 that our results concerning
the vapor pressure above solid chromium trichloride coꢀ
incide, to within a confidence interval, with published
data.1 The essential divergence from the data of another
Fig. 3. Temperature dependences of the experimental total presꢀ
sure (1) and the calculated partial pressures (2—4) in the
CrCl (s) + Cl system: p (2), pCrCl3 (3), and pCrCl4 (4).
,5
3
2
Cl2

1624
Russ.Chem.Bull., Int.Ed., Vol. 53, No. 8, August, 2004
Zelenina et al.
Table 2. Thermodynamic characteristics of reaction (3)
7. H. Oppermann, Z. Anorg. Allg. Chem., 1968, 359, 51.
. K. Matsuzaꢁi, H. Morita, and Y. Sacꢁi, Koguo Kagaku Zasshi,
971, 74, 1592.
9. V. M. Kovba, Zh. Neorg. Khim., 1983, 28, 2689 [J. Inorg.
Chem. USSR, 1983, 28 (Engl. Transl.)].
8
1
Treatment procedure
–∆ H°
–∆ S°
Ref.
r
298
r
298
–
1
/
ꢁk
/k K
Second law
of thermodynamics
Third law
of thermodynamics
Second law
of thermodynamics
207.1±11.6
206.8±11.6
173.6±10
173.2±3.5
170.9±4**
—*
—*
10. k. S. Olgen and R. S. Wyatt, J. Chem. Soc., Dalton Trans.,
1987, 4, 859.
11. V. Plies, Z. Anorg. Allg. Chem., 1988, 556, 120.
12. A. N. Ryꢁov and Yu. M. Korenev, Zh. Neorgan. Khim.,
1990, 35, 3183 [J. Inorg. Chem. USSR, 1990, 35 (Engl.
Transl.)].
221.7±6,**
205.1±12**
7
11
1
3. R. k. M. Konings, High Temperature and Materials Science,
996, 105, 113.
14. G. I. Noviꢁov and A. V. Suvorov, Zav. Labor., 1959, 6, 750
Ind. Lab., 1959, 6 (Engl. Transl.)].
5. V. A. Titov, T. P. Chusova, and G. A. Koꢁovin, Izv. Sib. Otd.
Akad. Nauk SSSR. Ser. Khim. Nauk, 1979, 14, 3 [Bull. Sib.
Div. USSR Acad. Sci, Ser. Chem. Sci., 1979, 14 (Engl.
Transl.)].
Note. Temperature dependence of the equilibrium constant for
reaction (3) is described by the equation logKeq = –8.2848 +
1
+
*
10346.1/T.
Data of this worꢁ.
[
1
*
* Our calculation for standard conditions using the data of
Table 1.
model we chose and the absence of significant systematic
errors in both our results and published data.2 The Table
also gives the enthalpies and the entropies of reaction (3)
available from the literature.
16. Z. I. Zasorina, Yu. G. Stenin, and G. A. Koꢁovin, Izv. Sib.
Otd. Akad. Nauk SSSR. Ser. Khim. Nauk, 1979, 14, 16 [Bull.
Sib. Div. USSR Acad. Sci, Ser. Chem. Sci., 1979, 14 (Engl.
Transl.)].
0
1
7. V. A. Titov and G. A. Koꢁovin, in Matematicheskie problemy
khimii, Chast´ II [Mathematical Problems in Chemistry.
Part II], Izd. Sib. Otd. Aꢁad. Nauꢁ SSSR, Novosibirsꢁ,
The authors are grateful to E. S. Grinberg for ꢁindly
providing samples of chromium trichloride.
1
975, 25 (in Russian).
8. A. N. Kornilov and L. M. Vidavsꢁii, Zh. Fiz. Khim., 1969,
3, 2224 [Russ. J. Phys. Chem., 1969, 43 (Engl. Transl.)].
9. E. F. Titova, V. A. Titov, A. A. Trunov, G. A. Koꢁovin, L. I.
Chernyavsꢁii, and F. A. Kuznetsov, Bank dannykh po
svoistvam materialov elektronnoi tekhniki [Data Bank on The
Properties of Materials for Electronic Engineering], Preꢀ
print 90ꢀ16, Novosibirsꢁ, 1990, 44 (in Russian).
1
1
References
4
1
2
. F. Ephraim, Ber. Deutsch. Chem. Ges, 1917, 50, 1088.
. K. kellineꢁ and R. Koop, Z. Phys. Chem. (Leipzig), 1929,
1
45, 309.
3
4
. H. A. Doerner, Technol. Papers Bur. Mines., 1937, 577, 51.
. C. Wagner and V. Stein, Z. Phys. Chem., Abt. A., 1943,
2
0. B. Bartley, Ebbinghaus Combustion and Flame, 1995,
1
01, 3121.
1
92, 129.
. N. V. Galitsꢁii and V. M. Gus´ꢁov, Izv. Vuzov. Tsvetn. Met.,
965, 4, 75 [Izv. Vuz. Tsvetn. Met., 1965, 4 (Engl. Transl.)].
. G. I. Noviꢁov and N. V. Galitsꢁii, Zh. Neorg. Khim., 1965,
0, 576 [J. Inorg. Chem. USSR, 1965, 10 (Engl. Transl.)].
5
6
1
Received September 5, 2003;
in revised form February 3, 2004
1