Vaporisation of chromia in humid air

Article abstract of DOI:10.1016/j.jpcs.2004.06.092

Thermodynamic data for the computation of the vaporisation of Cr ₂O₃(s) in humid air show significant deviations between each other which lead to large differences in the computed partial pressures. The vaporisation of Cr₂O₃(s) in air with different humidities, p(H₂O)=0.0007 bar to p(H₂O)=0.3 bar, was, therefore, investigated at constant temperature of 1223 K by the vapour transpiration method. The results obtained were explained by the formation of CrO₃(g) as major vapour species at low p(H₂O) and the formation of CrO₂(OH)₂(g) as major vapour species at high p(H₂O). The pressures evaluated are compared with those obtained by thermodynamic computations using different data bases.

Full text of DOI:10.1016/j.jpcs.2004.06.092

Journal of Physics and Chemistry of Solids 66 (2005) 384–387
www.elsevier.com/locate/jpcs
Vaporisation of chromia in humid air
*
C. Gindorf , L. Singheiser, K. Hilpert
Institute for Materials and Processes in Energy Systems (IWV-2), Forschungszentrum J u¨ lich, 52425 J u¨ lich, Germany
Accepted 23 June 2004
Abstract
Thermodynamic data for the computation of the vaporisation of Cr
which lead to large differences in the computed partial pressures. The vaporisation of Cr
.0007 bar to p(H O)Z0.3 bar, was, therefore, investigated at constant temperature of 1223 K by the vapour transpiration method. The
2
O
3
(s) in humid air show significant deviations between each other
2
O
3
(s) in air with different humidities, p(H O)Z
2
0
2
results obtained were explained by the formation of CrO
3
2 2 2
(g) as major vapour species at low p(H O) and the formation of CrO (OH) (g) as
major vapour species at high p(H O). The pressures evaluated are compared with those obtained by thermodynamic computations using
2
different data bases.
q 2005 Published by Elsevier Ltd.
Keywords: A. Alloys, oxides; D. Thermodynamic properties
1
. Introduction
other except those for CrO(g) and CrO (g) (Table 1) and
2
lead then to very different partial pressures (Section 3).
Table 1 exemplifies some of the differences, for more
information see Refs. [2–4].
Chromium base alloys generally form a Cr O scale on
2
3
the alloy surface under oxidising conditions which protects
the alloy from corrosion attack. Chromia vaporisation in the
prevailing atmospheres is, therefore, of interest at the high
alloy service temperatures. This is for example the case in
the development of power plants with steam cycle and solid
oxide fuel cells, SOFC. Chromium base alloys are used in
SOFCs with metallic interconnnect and it has been shown
that chromium vaporisation can lead to a rapid degradation
of the electrical properties of SOFC [1].
In order to make a first step for the elucidation of the
differences, a transpiration apparatus was developed by us
and fundamental investigations were carried out using the
vapour transpiration method. Vaporisation measurements of
Cr O (s) with constant humidity and different temperatures
3
2
as well as constant temperature and different humidities
were carried out. The results of the mentioned first
measurements with different temperatures have been
published previously [5], the results of the measurements
mentioned last with different humidities are reported here.
Chromia vaporisation was, therefore, computed by
thermodynamic computations at temperatures of about
1
000 8C and in an atmosphere made of ambient air with
different amounts of humidity [1]. Gaseous chromium
oxides, hydroxides and oxihydroxides are formed under
these conditions. Thermodynamic data of the gaseous
species are available in the data bases of JANAF [2],
IVTANTHERMO [3], and Ebbinghaus [4]. Data of
hydroxides and oxihydroxides are only present in the
databases of IVTANTHERMO [3] and Ebbinghaus [4]. The
data from these databases differ significantly from each
2
. Experimental
The transpiration apparatus developed by us was
described in detail in Ref. [5]. The tubular vaporisation
reactor is completely made of quartz glass heated by a three
zone furnace. The vaporisation chamber is restricted by a
quartz frit (upstream) and a capillary (downstream) in order
to minimise diffusion effects. Synthetic air is used as carrier
gas. A well defined level of humidity in this gas is adjusted
by the use of a humidifier coupled with a condensor which
*
Corresponding author.
0022-3697/$ - see front matter q 2005 Published by Elsevier Ltd.
doi:10.1016/j.jpcs.2004.06.092

C. Gindorf et al. / Journal of Physics and Chemistry of Solids 66 (2005) 384–387
385
Table 1
Thermodynamic data of gaseous chromium species in the Cr
2
O
3
–H
2
O–O
2
system from the literature
H8298 (kJ mol
K1 K1
K
K1
Species
D
f
S8298 (J mol
)
D
f
)
JANAF
Ref. [2]
IVTANTHERMO
Ref. [3]
Ebbinghaus
Ref. [4]
JANAF
Ref. [2]
IVTANTHERMO
Ref. [3]
Ebbinghaus
Ref. [4]
Cr(g)
174.311
174.202
238.481
265.461
269.294
248.616
295.154
332.986
364.245
–
174.314
238.500
265.500
269.300
260.400
316.100
378.900
386.900
470.200
499.400
307.700
350.300
398.300
428.500
315.100
357.400
397.480
397.147
186.242
397.500
182.400
CrO(g)
239.3
188.280
CrO
CrO
2
(g)
(g)
269.200
K75.312
K292.880
K108.038
K322.033
60.801
K99.200
K323.500
78.9
3
266.178
CrOH(g)
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Cr(OH)
Cr(OH)
Cr(OH)
Cr(OH)
Cr(OH)
2
3
4
5
6
(g)
(g)
(g)
(g)
(g)
K330.715
K640.112
K859.313
–
K321.600
K631.400
K868.600
K1087.500
K1276.000
K246.800
K552.400
K792.000
K1012
–
–
CrOOH(g)
283.904
319.152
340.447
349.981
299.177
326.681
K199.264
K543.563
K793.199
K873.143
K441.781
K736.000
CrO(OH)
CrO(OH)
CrO(OH)
2
3
4
(g)
(g)
(g)
CrO
CrO
2
OH(g)
(OH) (g)
K438.000
K748.300
2
2
can be adjusted to different temperatures. The vaporising
chromium species condense on the condensing part of the
quartz tube, they are dissolved with conc. HNO3 and
determined by atomic absorption spectroscopy Fig. 1.
Transpiration experiments were carried out with
Cr O (s) powder (99% purum) placed in a quartz boat.
Cr O ðsÞ CO ðgÞ CH OðgÞ Z 2 CrO ðOHÞðgÞ
(4)
2
3
2
2
2
Fig. 2 shows a comparison of the experimental pressure
determined in this work (Table 2) with the partial pressures
obtained by thermochemical computations using the data
bases of IVTANTHERMO [3] and Ebbinghaus [4]. Only the
partial pressures of the most abundant species are shown in
Fig. 2. The gaseous species CrO (OH) (g) and CrO (g) are
2
3
Saturation of the carrier gas was adjusted by measuring the
chromium transport rate for different gas flow rates (Fig. 2).
Thermodynamic equilibrium is established in the gas flow
2
2
3
the most abundant chromium containing species in the
equilibrium vapour. It is interesting to note that the
CrO (OH) partial pressures predicted by the computations
K1
range between 100 and 350 ml min
2
2
using the different data bases deviate by more than two
orders of magnitude. This reflects the high uncertainty of the
thermochemical data for CrO (OH) (g) which were
3
. Results and discussion
2
2
estimated. In contrast to this, the thermochemical data of
CrO (g) from these two data bases [2,3] agree well which
The parameters and results of the experiments by the
vapour transportation method at the constant temperature of
223 K and different humidities are shown in Table 2. The
3
leads also to good agreement for their partial pressures. The
CrO partial pressure resulting from the data of the JANAF
1
3
time period of the experiments Dt, and the gas volume flow
rate n as well as the resulting mass transport rate of the
condensed chromium k(Cr) and the partial pressure p(Cr)
are given. The term p(Cr) means the partial pressure of a
gaseous species which contains only one Cr atom such as
CrO (OH) (g) or CrO (g). The partial pressure results from
data base [2] is by about a factor of 15 smaller than the CrO3
partial pressure by the data of IVTANTHERMO.
2
2
3
the gas volume flow rate n at the temperature T by the
equation
R T kðCrÞ
pðCrÞ Z
(1)
v MðCrÞ
where R is the gas constant and M(Cr) the atomic mass of
chromium. Eq. (1) is obtained on the basis of the ideal gas law.
Thermodynamic computations indicate that the follow-
ing vaporisation reactions are most abundant
Cr O ðsÞ C1:5O ðgÞ Z 2CrO ðgÞ
(2)
2
3
2
3
2 3
Fig. 1. Chromium vaporisation rate of Cr O (s) at 1223 K in humid air,
Cr O ðsÞ C1:5 O ðgÞ C2 H OðgÞ Z 2 CrO ðOHÞ ðgÞ (3)
2
3
2
2
2
2
p(H O)Z0.1 bar, for different flow rates.
2

386
C. Gindorf et al. / Journal of Physics and Chemistry of Solids 66 (2005) 384–387
Fig. 2. Vaporisation of Cr
2
O
3
(s) at 1223 K in air with different humidity and comparison with thermodynamic modelling using different data bases.
Table 2
Vaporisation of Cr
Section 3)
2
O
3
(s) in humid air, p(H
Dt (min)
2
O)Z0.02 bar, at 1223 K and parameters of the transpiration experiments (Meaning of the symbols see text in
K1
n (ml min
K1
k (kg s
Run no.
)
p(H
2
O) bar
m(Cr) (mg)
)
p(Cr) (Pa)
K13
K13
K13
K14
K14
K13
K14
K3
1
2
3
6
7
8
9
2834
4073
1351
2850
5610
371
200
150
150
200
150
150
150
0.1
42.4
28.5
31.0
3.9
2.49!10
1.17!10
3.82!10
2.28!10
1.92!10
4.58!10
3.50!10
3.46!10
2.16!10
7.07!10
3.16!10
3.57!10
8.47!10
6.47!10
K3
K2
K4
K4
K2
K4
0.05
0.2
0.0007
0.002
0.3
6.5
10.2
5.6
2668
0.007
Our experimental results (dashed red line in Fig. 2 in web
version) show that the Cr O (s) vaporisation is determined
by CrO (g) as most abundant species in the range of about
the partial pressure of CrO (OH) (g) at constant tempera-
2 2
2
3
ture where b is a constant (cf. Eq. (3)). Fig. 2 shows that the
experimental points can be fitted well by Eq. (5) and the
resulting values for p(CrO ) and b are given there.
3
p(H O)O0.005 bar, at higher values of p(H O) (p(H O)
2
2
2
3
!
0.005 bar) oxihydroxides, probably CrO (OH) (g), are
2 2
the dominant vapour species. Accordingly, chromium
vaporisation is practically independent of humidity at low
p(H O) (O0.005 bar) and it increases if the humidity is
2
raised above this value. Our experimental pressures clearly
support the accuracy of the data for CrO (g) by
3
IVTANTHERMO [3] and Ebbinghaus [4], they do not
support the data from these References for the species
CrO (OH) (g) (Fig. 2).
2
2
The total pressure, p(sum), can be represented by the
equation
pðsumÞ Z pðCrO Þ CpðCrO ðOHÞ Þ
(5)
3
2
2
if we assume that CrO (g) and CrO (OH) (g) are the only
2
3
2
chromium containing species in the vapour. The partial
pressure p(CrO ) does not depend from p(H O) and is
3
2
Fig. 3. Vaporisation of Cr
constant humidity, p(H O)Z0.02 bar, and comparison with thermodyn-
amic modelling using different data bases [5].
2 3
O (s) at different temperatures in air with
constant at constant temperature (cf. Eq. (2)). In contrast to
this, the relation p(CrO (OH) ) Z b p(H O) is valid for
2
2
2
2

C. Gindorf et al. / Journal of Physics and Chemistry of Solids 66 (2005) 384–387
387
Our own measurement of the temperature dependence of
the vaporisation of Cr O (s) in humid air, p(H O)Z
References
2
3
2
[
[
1] K. Hilpert, D. Das, M. Miller, D.H. Peck, R. Weiß, J. Electrochem. Soc.
43 (1996) 3642.
0
.02 bar, (see Fig. 3 and [5]) supports the results of this
work. Fig. 2 and the above discussion suggest that
CrO (OH) (g) is the most abundant vapour species under
1
2] M.W. Chase, C.A. Davies, J.R. Downey, D.J. Frurip, R.A. McDonald,
N. Syverud (Eds.), JANAF Thermochemical Tables, Part Ithird ed.,
American Institute of Physics, New York, 1986.
2
2
these conditions and determines the experimental pressures
shown in Fig. 3. The differences between the computed
partial pressure of CrO (OH) (g) and the experimental
[
3] V.S. Yungman, V.A. Medvedev, I.V. Veits, G.A. Bergman (Eds.),
IVTANTHERMO—A Thermodynamic Database and Software System
for the Computer, CRC Press/Begell House, Boca Raton, FL, 1993.
4] B.B. Ebbinghaus, Combust. Flame 93 (1993) 119.
2
2
pressure, assigned predominantly to this species in Fig. 3,
[
are consistent with those in Fig. 2 at high p(H O).
2
[5] C. Gindorf, L. Singheiser, K. Hilpert, Steel Res. 72 (2001) 528.