Determination of intrinsic kinetics parameters for MoO3 chlorination with Cl2 gas between 798 and 873 K

Article abstract of DOI:10.1016/j.tca.2012.05.026

In this work the kinetics of the chlorination of molybdenum trioxide has been studied by thermogravimetry between 798 and 873 K. The starting temperature for the reaction of MoO₃ with chlorine is determined at about 770 K. The influence of gaseous flow rate, sample mass, temperature, and chlorine partial pressure in the reaction rate is analyzed for two MoO₃ samples having different particle size and morphology. The experimental conditions for chemical control of the reaction rate were established for both types of samples. An average activation energy of 211 kJ mol^-1 and a reaction order of 1 with respect to chlorine partial pressure were determined for the chlorination of MoO₃ with gaseous chlorine. A complete rate equation was formulated that describes the reaction evolution of each type of solid.

Full text of DOI:10.1016/j.tca.2012.05.026

Thermochimica Acta 543 (2012) 211–217
Contents lists available at SciVerse ScienceDirect
Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca
Determination of intrinsic kinetics parameters for MoO3 chlorination
with Cl2 gas between 798 and 873 K
G. De Miccoa,b,c,∗, M. Carignana, C.A. Canavesiob, A.E. Bohéa,b,d
a Comisión Nacional de Energía Atómica (C.N.E.A.), Avenida Bustillo 9500, 8400 San Carlos de Bariloche, Argentina
b Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina
c Universidad Nacional de Cuyo, Instituto Balseiro, Avenida Bustillo 9500, 8400 San Carlos de Bariloche, Argentina
d Universidad Nacional del Comahue, Centro Regional Universitario Bariloche, 8400 San Carlos de Bariloche, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work the kinetics of the chlorination of molybdenum trioxide has been studied by thermogravime-
try between 798 and 873 K. The starting temperature for the reaction of MoO3 with chlorine is determined
at about 770 K. The influence of gaseous flow rate, sample mass, temperature, and chlorine partial pres-
sure in the reaction rate is analyzed for two MoO3 samples having different particle size and morphology.
The experimental conditions for chemical control of the reaction rate were established for both types of
samples. An average activation energy of 211 kJ mol−1 and a reaction order of 1 with respect to chlorine
partial pressure were determined for the chlorination of MoO3 with gaseous chlorine. A complete rate
equation was formulated that describes the reaction evolution of each type of solid.
Received 30 January 2012
Received in revised form 17 May 2012
Accepted 20 May 2012
Available online 29 May 2012
Keywords:
Chlorination
Molybdenum trioxide
Thermogravimtery
Intrinsic parameters
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
The MoO3 chlorination reaction has been investigated before
by several authors. A literature review of the articles published on
Chlorination reactions are applied in extractive metallurgy to
produce valuable metals such as titanium and zirconium. Simi-
larly, this process could be applied to low-grade mineral ores and
complex residues where a certain valuable metal is combined with
other elements in a multi-compound system, to extract the metal
of interest [1,2].
Molybdenum compounds have a widespread variety of tech-
nological applications including hydroprocessing catalysts used in
petroleum refineries, molybdenum-containing steel [3], polymer
solar cells [4], Mo containing alloys for biomedical applications [5],
among others, for this reason it is important to develop technolog-
ical strategies to recover this metal from different types of wastes
[6–9]. This is due not only to have economical sources of valuable
metals but also from an environmental point of view. As an exam-
ple, spent catalysts containing metals such as Mo, W and Ni are
hazardous toxic wastes that need to be correctly treated. Chlorina-
tion is one of the possible ways of metal reclamation and, for that
reason; studies on these reaction kinetics and mechanisms can be
applied in the development of suitable recovery technologies.
this subject until 1995 was presented by Djona et al. [10]. A vari-
ety of chlorinating agents were used in the works reviewed, such
as Cl2, SOCl2, or CCl4, PCl5, or different Cl2 mixtures: Cl2 + O2 + N2,
Cl2 + S2Cl2. Most of the studies are devoted to the study of the
carbochlorination reaction using different carbons, and others are
conducted in the presence of additional solids such as soot. The
literature review revealed that no systematic study about chlori-
nation of MoO3 by Cl2 was reported. The work of Djona et al. [10] is
a contribution on that subject. They determined a starting tem-
perature of 873 K (600 ◦C), an activation energy of 165 kJ mol−1
,
and a reaction order with respect to Cl2 of 0.75 for the chlori-
nation of MoO3 with Cl2–N2. They determined the function that
describes the solid evolution during the reaction X = kt (reaction
controlled by chemical reaction of grains having the form of slabs)
[10]. Another work was published by Ojeda et al. [11] who stud-
ied the effect of working variables on the chlorination of MoO3 in
the presence of carbon black, they identified that the only reaction
product was MoO2Cl2 by chemical analysis of the solid products.
They performed a thermodynamic analysis based in ꢀG◦ changes
and concluded that formation of MoO2Cl2 is the most likely product
for the chlorination of MoO3. Guethert et al. [12] also determined
that MoO2Cl2 is formed by the reaction of MoO3 and chlorine
(3.4% Cl2–0.1% O2–96.5% N2), and in presence of moisture, hydrol-
ysis of MoO2Cl2 leads to formation of MoO3. Finally, the X-ray
∗
Corresponding author at: Comisión Nacional de Energía Atómica (C.N.E.A.),
Avenida Bustillo 9500, 8400 San Carlos de Bariloche, Argentina.
E-mail address: demiccog@cab.cnea.gov.ar (G. De Micco).
0040-6031/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.tca.2012.05.026

212
G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
3. Experimental
List of symbols
The gases used in this work were Cl2 of 99.8% purity (Indupa,
Argentina) and Ar of 99.99% purity (AGA, Argentina). Two different
MoO3 were used for the study:
A
D
superficial area of sample layer, m2
gaseous diffusion coefficient for Ar–Cl2 mixture,
m2 s−1
E
g(X)
activation energy, kJ mol−1
•
MoO3(I): Aldrich 99.99% trace metal basis. SIGMA ALDRICH Co.
3050 Spruce Street, St. Louis, MO 63103, USA.
MoO3(II): Alfa Aesar Puratonic 99.998% (metals basis), Johnson
Matthey, 30 Bond Street, Ward Hill, MA 01835, USA.
solid reactant conversion function under the
absence of the effect of interstitial diffusion, dimen-
sionless
adsorption equilibrium constant of Langmuir
adsorption isotherm, dimensionless
pre-exponential factor of the intrinsic reaction con-
stant, s−1 kPa
•
K
The solids were well characterized by scanning electron
microscopy (SEM 515, Philips Electronics Instruments), the mor-
phology of each solid reactant is shown in Fig. 1. MoO3(I) is formed
by non-porous crystalline grains of 1–20 ␮m size, which form
agglomerates of different sizes up to about 300 ␮m. MoO3(II) is
composed by needle shaped non-porous crystalline particles of
about 200 ␮m length. In Fig. 1a and b the shape of the MoO3(I) and
(II) particles can be observed. In Fig. 1a (MoO3(I)) several agglom-
erates of different sizes can be seen as separate particles, while
in MoO3(II) (Fig. 1b) no agglomerates are formed and each par-
ticle is a needle shaped grain itself. In Fig. 1c the surface of one
agglomerate particle (MoO3(I)) is shown in detail and in Fig. 1d
the surface of a needle (MoO3(II)) can be observed. According to
SEM examinations, MoO3(II) will have two types of pores: small
pores formed among the spherical crystalline grains and large pores
formed among agglomerates of different sizes, both of them in
the range of macropores [15]. Meanwhile, MoO3(II) will only have
one type of pores formed among the needles, also in the range of
macropores. BET areas [16] were measured with Kr for both type
of samples, the values obtained were 1.2 × 104 3 × 102 cm2 g−1
and 1.64 × 103 30 cm2 g−1 for the initial powders of MoO3(I) and
MoO3(II) respectively. A theoretical calculation of the total surface
area assuming spherical and cylindrical particles of average radius
(5 and 25 ␮m) for MoO3(I) and MoO3(II) was made. The values
obtained for both samples were about 3 orders of magnitude lower
than the BET values, this is indicating that the solids have roughness
or porosity which is below the SEM resolution (about 1 ␮m).
Isothermal and non-isothermal chlorination reactions were
carried out in a thermogravimetric analyzer (TGA). This thermo-
gravimetric analyzer consists of an electrobalance (Model 2000,
Cahn Instruments, Inc.) suitable for working with corrosive atmo-
spheres, a gas line and a data acquisition system. This experimental
set-up has a sensitivity of 5 ␮g while operating at 1223 K under a
flow of 8 L h−1. Samples of 0.5–20 mg were placed on a flat quartz
crucible forming a loose packed bed, and inside the reactor an argon
flow of 1.3 L h−1 was maintained. To begin the non-isothermal
experiments, a chlorine flow of 0.8 L h−1 was introduced and, at
k*
kapp
f (pCl
n
apparent reaction rate constant, s−1
rate dependence on Cl2 partial pressure, kPa
particle shape factor in g(X), dimensionless
flux of chlorine, mol Cl2 s−1 m−2
experimental value of chlorine flux, mol Cl2 s−1 m−2
partial pressure of chlorine at the surface of solid
sample layer, kPa
partial pressure of chlorine in the bulk, kPa
gas constant, 8.314 × 103 kJ K−1 mol−1
initial radius of the spherical or cylindrical particles,
m
)
2
NCl
2
NCl
pCl
(exp)
2
,s
2
pCl
R
r0
,b
2
t
T
Vm
time, s
temperature, K
molar volume of the initial phase (MoO3), m3 mol−1
WMoO3 molecular weight of MoO3, g mol−1
X
ı
ꢁ
solid conversion, dimensionless
boundary layer thickness, m
Arrhenius pre-exponentional factor, mol m−2 s−1
diffraction pattern of MoO2Cl2 was reported for the first time in
2003 by Volkovicha et al. [13].
In this work the chlorination of molybdenum trioxide was stud-
ied. Two different reactive grade solids were measured to obtain the
kinetic parameters of the reaction (activation energy and reaction
order with respect to gaseous reactant). The results obtained can
contribute to the study of the chlorination reaction of complex sys-
tems containing other species such as waste catalysts as well as for
the design and development of molybdenum recovery processes
involving chlorination.
2. Thermodynamical considerations
The phase stability diagram for Mo–O–Cl system including
several molybdenum chlorides and oxychlorides was presented
previously [12] and is not repeated here. For the reaction conditions
used in our experiments (chlorine partial pressure 35 kPa used for
chlorination reactions and oxygen partial pressure below 10−4 kPa)
the stable solid state is MoO2Cl2 (at 773 K). Similarly, among all
possible reactions between MoO3 and Cl2 the only one that has a
negative ꢀG◦ value (above 836 K) is the following:
the same time, the heating started with a ramp of 260 K h−1
.
For the isothermal runs, samples of MoO3(I) and (II) powders
were heated at the working temperature in flowing Ar for 1 h after
which chlorine was injected into the reactor and mass change and
time were simultaneously recorded. This procedure was employed
to analyze the influenceof gaseousflow rate, initial mass of the sam-
ple, temperature, and chlorine partial pressure on the reaction rate.
Partially chlorinated samples were analyzed by SEM and no
morphological changes were observed, indicating that the overall
reaction, which forms only gaseous products, largely takes place
near the outer boundaries of the particles for both solids.
1
2
MoO3(s) + Cl2(g) → MoO2Cl2(g) + O2(g) ꢀG◦ (kJ)
= −0.1268 T(K) + 106.01
(1)
4. Results
Finally, equilibrium composition as a function of temperature
was calculated with HSC software [14]. It was obtained that starting
with chlorine and excess of molybdenum trioxide as input mate-
rials, the major gaseous products that will be formed are MoO2Cl2
and O2 for temperatures between 798 and 873 K.
4.1. Onset temperature of chlorination reaction
The starting temperature for the chlorination of MoO3 was
determined by non-isothermal thermogravimetric measurements.

G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
213
Fig. 1. SEM photographs of the initial sample (a) several agglomerates of different sizes of MoO3(I), (b) particles of MoO3(II), (c) the surface of an agglomerate of MoO3(I), (d)
the surface of a particle of MoO3(II).
In Fig. 2 the ratio between the mass change and the initial mass
of the sample as a function of temperature is shown for MoO3 (I)
and (II). The reaction starts at approximately 770 K, this value is
100 K lower than the previously reported [10]. The small difference
between both curves is attributed to a higher value of specific sur-
face area of MoO3(I), which allows to detect the mass change for this
sample slightly before the one corresponding to MoO3(II). In both
cases the sample starts to lose mass denoting the formation of a
gaseous product. To identify the gaseous product, the reaction was
carried out in a tubular reactor that allows to condense and collect
the vapor species formed. The formation of MoO2Cl2 was deter-
mined by XRD (Fig. 3) and EDS. In the XRD pattern shown in Fig. 3,
MoO3 was also detected, this was attributed to the decomposition
of MoO2Cl2 in presence of humidity [12].
The MoO2Cl2 has a high vapour pressure which indicates that it
is formed as a vapor by the chlorination reaction, the vapor partial
pressure of MoO2Cl2 is higher than 100 kPa at temperatures above
about 500 K [14].
4.2. Determination of kinetic parameters
To determine the kinetic parameters it is first necessary to find
the experimental conditions under which the reaction rate is not
influenced by mass transfer processes (i.e. gaseous starvation, dif-
fusion thorugh the gaseous boundary layer, diffusion throguh the
sample pores).
Starvation occurs when a sufficient portion of the reactant is
consumed by the solid, so that the particle is not in contact with
a gas of bulk composition, but the gaseous reactant concentration
Fig. 2. TG curve for non-isothermal chlorination of MoO3(I) and MoO3(II).
Fig. 3. XRD profile of the chlorination reaction product.

214
G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
Fig. 4. Influence of gaseous flow rate in the chlorination of (a) MoO3(I) and (b) MoO3(II), and influence of sample mass in the chlorination of (c) MoO3(I) and (d) MoO3(II).
is less than this value [17,18]. In this condition, the driving force
for the reaction will depend on the gas velocity, and an increase in
the flow rate will cause the same increase in the reaction rate. The
gaseous flow rate was raised from 4 L h−1 to 8 L h−1 in the chlorina-
tion of 1 mg of MoO3(I) at 848 K and MoO3(II) at 823 K to ensure that
the systems were not in gaseous reactant starvation. The thermo-
gravimetric (TG) curves obtained for the different flows are shown
in Fig. 4 a and b. It can be seen that not increase in the reaction rate
was obtained when duplicating the gaseous flow in both cases.
To analyze whether diffusion through the boundary layer affects
the overall rate, isohermal chlorination experiments were per-
fomed for temperatures between 798 and 873 K and calculations
were made to compare the experimental rate with the rate of chlo-
rine diffusion between the top of the crucible and the top surface
of the sample layer. The rate of diffusion through this space was
estimated considering ordinary diffusion with the Fick’s Law (Eq.
considering the reaction surface as the cross section area at the
top of the sample layer, we have:
1
NCl
= −r
(3)
(exp)
2
WMoO3 A
For both samples it is seen that mass transfer controlled rate
would be significantly faster than the observed rate. Consequently,
we conclude that gas phase diffusion had no effect in the overall
rate.
The effect of sample mass in the reaction rate was analyzed at
848 K for MoO3(I) and at 823 K for MoO3(II), as shown in Fig. 4c and
d. For MoO3(I) it can be seen that there is an increase in the rate of
conversion as sample mass decreases from 2 to 1 mg, but further
decrease of mass sample to 0.5 mg has no significant influence in
the reaction rate (Fig. 4c), since the difference in the rate values
obtained for 1 and 0.5 mg is within the range of the experimental
scatter. The same happens for MoO3(II) for sample masses of 2 and
1 mg (Fig. 4d).
(2)) the detalied description for the calculation of pCl ,s, is explained
2
elsewhere [19].
Table 1 shows the calculated and experimentally obtained molar
fluxes of chlorine for MoO3(I) and MoO3 (II) samples. The values of
For this reason, it can be concluded that for mass values lower
than 1 mg the overall rate is controlled by intrinsic particle kinetics
for both type of samples (MoO3(I) and MoO3(II)).
D and pCl used in the calculation are also presented in the table.
2
To determine the kinetic parameters, experiments were per-
formed at different temperatures between 798 and 873 K with
35 kPa partial pressure of Cl2 and 1 mg samples, for which the
gaseous diffusion had no effect on the overall rate. Fig. 5 shows
the TG curves obtained for different temperatures in terms of con-
version vs. time.
The experimental conversion vs. time curves were fitted with a
conversion function that describes a reaction proceeding accord-
ing to a topochemically contracting geometry [17]. This equation
may be used since the product of reaction being a gaseous species,
the rate-determining step of the mechanism of reaction falls to
be located at the surface of the reacting phase (i.e. the surface of
the particles of MoO3). Thus the kinetic curves may be described
D
NCl = −
(pCl ,s − pCl
)
(2)
,b
2
2
2
RTı
The values of D were calculated for an Ar–Cl2 mixture according
to the Chapman-Enskog correlation [20] and the corresponding
temperature, ı was considered to be the crucible heigth minus the
sample thickness, and pCl was calculated assuming equilibrium
,s
2
at the reaction surface.
The values of NCl
were calculated from the slope (r) of
(exp)
the mass changes vs2. time curves corresponding to the TG mea-
surments of the chlorination reactions at different temperatures.
The slope was calculated by linear fitting these curves up to about
0.7 of conversion. According to the reaction stoichiometry and

G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
215
Table 1
Comparison between rate of diffusion through crucible top and experimental rate of reaction.
Temperature [K]
(pCl
)
[kPa]
D [m2 s−1
]
Rate of diffusion between the
top of the crucible and sample
layer [mol Cl2/(m2 s)]
Experimental
rate MoO3(I)
[mol Cl2/(m2 s)]
Experimental
rate MoO3(II)
2
eq
[mol Cl2/(m2 s)]
798
823
833
848
873
8.6
6.1
5.3
4.2
3
6.4E−05
6.8E−05
6.9E−05
7.2E−05
7.6E−05
6.6E−02
7.5E−02
7.7E−02
8.2E−02
8.7E−02
1.4E−05
2.7E−05
3.7E−05
7.1E−05
–
5.7E−06
2.1E−05
–
3.4E−05
9.4E−05
by the same rate law than that of a “topochemically contracting
geometry”:
eliminated and mass transfer effects are minimized. However, the
effect of diffusion through sample pores may still be significant
since they worked with large sample masses (100 mg). For this
reason, the activation energy determined (165 kJ mol−1) may cor-
respond to a “mixed control regime” where diffusion within the
pores limits the overall rate of reaction instead of a “chemical con-
trol regime” where the overall rate is controlled by the intrinsic
chemical reaction.
g(X) = 1 − (1 − X)1/n = kapp
t
(4)
where X is the fractional conversion, and the value of n depends on
the solid geometry.
If the hypothesis of variable separation holds, the apparent reac-
ton constant (kapp) may be expressed by:
To analyze the influence of pCl2 , experiments were conducted
with different partial pressures of Cl2 between 10 and 55 kPa, and
with 1 mg samples at 823 K. The corresponding conversion vs. time
curves were fitted according to Eq. (4) with n = 3 for MoO3(I) and
n = 2 for MoO3(II). With the value of kapp obtained, the kapp vs. pCl
plots were calculated (Fig. 8a and b). From this plots, a linear rela2-
kapp = k∗ e−E/RT f (pCl
)
(5)
2
where f (pCl2 ) is a function of the reactant gas partial pressure and
k* an apparent pre-exponential factor.
This aproximation may not always be strictly valid. As an exam-
ple, if the adsorption of chlorine at the surface of the particles
is at equilibrium, and a subsequent step is rate-determining, the
resulting dependence of kapp could be of the form:
tion between kapp and pCl can be observed, which indicates that
2
the reacion order with respect to partial pressure of chlorine is 1.
Comparing this result with Eq. (6), the Langmuir model could still
apply if KpCl ꢀ 1 and K having a weak dependence on temperature.
K pCl
2
2
kapp = k0e−E/RT
(6)
However, to make a thorough study on the adsorption mechanisim,
much more experiments at different temperatures must be done,
and that is beyond the scope of the present paper. In Fig. 8c and d the
1 + K pCl
2
where k0 is the Arrhenius pre-exponential factor and K is the
adsorption equilibrium constant of Langmuir adsorption isotherm,
and is a function of temperature.
Our experimental results were well correlated with n = 3 for
MoO3(I) and n = 2 for MoO3(II), as is shown in Fig. 6. With the values
of kapp obtained at each temperature, the activation energy, E, can
be obtained from the slope of the ln kapp vs 1/T plot, according to:
tx (time requierd to achive a certain conversion value) vs. pCl plots
2
for x = 0.25, 0.5 and 0.75 are presented, together with the fittings
according to first order dependence on pCl2. The fitting equation is
tx = C P−1, being C = g(X(tx))/(k*e−E/RT), which is obtained combin-
Cl
2
ing and rearranging Eqs. (4) and (5) and calcualting g(X) for the fix
value of X corresponding to tx. It can be seen that this function fits
well the experimental data.
E
From Eq. (5), k* for each temperature can be calculated substi-
tuting f (pCl ) by pCl , (pCl = 35 kPa), and the activation energy
ln kapp = −
+ ln(k∗f (pCl2 ))
(7)
RT
2
2
2
This is shown in Fig. 7, where an activation energy of
216 6 kJ mol−1 was calculated for MoO3(I) and 207 12 kJ mol−1
for MoO3(II). This values are higher than the previously reported
value of 165 kJ mol−1 for MoO3 chlorination [10]. A possbile expla-
nation for this difference is that in their work Djona et al. studied the
effect of gas velocity on the reaction rate, and determined that it can
be considered negligible beyond about 18 cm min−1 (or 30 L h−1).
They conclude that for values above that flow, starvation is
previously obtained.
An average value of k* was considered to establish the following
relationship between conversion and time:
g(X) = 1 − (1 − X)1/3 = 1.91 × 10 e
p
2 t MoO3(I)
2 t MoO3(II)
(8)
(9)
8
−216/RT
Cl
g(X) = 1 − (1 − X)1/2 = 3.57 × 10 e
p
7
−207/RT
Cl
Fig. 5. Experimental conversion vs. time curves for temperatures between 798 and 873 K together with calculated conversion vs. time curves according to Eqs. (7) and (8).

216
G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
Fig. 6. (a, b) g(X) vs. time plot for the chlorination of 1 mg of MoO3(I) and MoO3(II) at temperatures between 798 and 873 K, 35 kPa of chlorine partial pressure and gas flow
rate of 4 L h−1, linear fitting also shown.
Fig. 7. ln kapp vs. T−1 plot for the calculation of activation energy (E): (a) MoO3(I) and (b) MoO3(II).
Fig. 8. (a, b) kapp vs. pCl plot for the determination of the reaction rate order with respect to pCl for MoO3(I) and MoO3(II). (c, d) Iso-conversional time vs. partial pressure
2
2
of chlorine plot and fitting according to first order reaction with respect to pCl
.
2

G. De Micco et al. / Thermochimica Acta 543 (2012) 211–217
217
The corresponding conversion vs. time curves according to Eqs.
(8) and (9) are plotted in Fig. 5 together with the experimental val-
ues. It shows that there is a good agreement between experimental
and calculated conversions for all temperatures.
pre-exponential factor and the solid evolution function due to the
difference in morphology and particle size of the samples studied.
Acknowledgments
By differentiating Eqs. (8) and (9) with respect to time and rear-
renging, the reaction rate expression can also be presented as:
The authors would like to thanks the Agencia Nacional de Pro-
moción Científica y Tecnológica (ANPCyT), Consejo Nacional de
Investigaciones Científicas y Técnicas (CONICET) and Universidad
Nacional del Comahue for the financial support of this work.
dX
dt
2 (1 − X)2/3 MoO3(I)
2 (1 − X)1/2 MoO3(II)
(10)
(11)
8
−216/RT
= 5.73 × 10 e
p
Cl
dX
dt
7
−207/RT
= 7.14 × 10 e
p
Cl
References
It is noticed that, since the pre-exponentional factor is not an
[1] B.G. Korshunov, Applications and potential uses of chlorination methods in
metallurgy of non common metals, Metall. Rev. MMIJ 8 (2) (1992) 1–29,
review.
[2] N. Kanari, E. Allain, R. Joussemet, J. Mochón, I. Ruiz-Bustinza, I. Gaballah, An
overview study of chlorination reactions applied to the primary extraction and
recycling of metals and to the synthesis of new reagents, Thermochim. Acta
495 (2009) 42–50.
intrinsic value, the difference in the value obtained for each type of
sample is due to the difference in morphology and particle size
of the two sample types. According to the grain model [17] for
chemical control of the reaction rate, and first order with respect
of gaseous reactant, k* can be expressed by:
[3] T.P. Hou, Y. Li, J.J. Zhang, K.M. Wu, Effect of magnetic field on the carbide pre-
cipitation during tempering of a molybdenum-containing steel, J. Magn. Magn.
Mater. 324 (2012) 857–861.
V
m ꢁ
r0
k∗
=
(12)
[4] L. Shen, S. Ruan, W. Guo, F. Meng, W. Chen, Semitransparent inverted polymer
solar cells using MoO3/Ag/V2O5 as transparent anodes, Sol. Energy Mater. Sol.
Cells 97 (2012) 59–63.
[5] C. Li, Y. Zhan, W. Jiang, b-Type Ti–Mo–Si ternary alloys designed for biomedical
applications, Mater. Des. 34 (2012) 479–482.
[6] B.W. Jong, S.C. Rhoads, A.M. Stubbs, T.R. Stoelting: Report of Investigations
9252—United States, Bureau of Mines, Moscow, ID, 1989, pp. 1–27.
[7] Z. Li, C.Y. Cheng, A literature review of the recovery of molybdenum and
vanadium from spent hydrodesulphurisation catalysts Part I.Metallurgical pro-
cesses, Hydrometallurgy 98 (2009) 1–9.
[8] M. Marafi, A. Stanislaus, Spent hydroprocessing catalyst management: areview
Part II. Advances in metal recovery and safe disposal methods, Resour. Conserv.
Recycl. 53 (2008) 1–26.
[9] B.B. Kar, B.V.R. Murthy, V.N. Misra, Extraction of molybdenum from spent cat-
alyst by salt-roasting, Int. J. Miner. Process. 76 (2005) 143–147.
[10] M. Djona, E. Allain, I. Gaballah:, Kinetics of chlorination and carbochlorination
of molybdenum trioxide, Metall. Mater. Trans. B 26B (1995) 703–710.
[11] M.W. Ojeda, J.B. Rivarola, O.D. Quiroga, Study on chlorination of molybdenum
trioxide mixed with carbon black, Miner. Eng. 15 (2002) 585–591.
[12] A. Guethert, R. Muenze, B. Eichler, Contribution to the thermodynamics of
molybdenum–oxygen–chlorine system, J. Radioanal. Chem. 62 (1–2) (1981)
91–101.
where ꢁ is the Arrhenius pre-exponential factor of the reaction rate
constant and does not depend of temperature and pressure condi-
tions, r0 is the initial radius of the spherical or cylindrical particles,
and Vm is the molar volume of the initial phase (MoO3).
The effect of particles size can be quantitativaly estimated from
the y-intercept of the ln kapp vs 1/T plot, substituting k* by Eq. (12) in
Eq. (7). Accoding to SEM examination an average radius of 5 ␮m for
MoO3(I) and 27 ␮m for MoO3(II) can be considered to calculate the
value of ꢁ (Vm(MoO3): 3.068 × 10−5 m3 mol−1). For both samples
a value of 3.1 × 107 mol m−2 s−1 was obtained. Even though there
is a good agreement between experimental results and theoretical
considerations, the values of Arrhenius pre-exponentional factor
(ꢁ)isaroughcalculation, sincer0 wasestimatedassumingspherical
and cylindrical geometries of the reacting particles.
5. Conclusions
[13] V.A. Volkovicha, T.R. Griffiths, R.C. Thied, B. Lewin, Behavior of molybdenum
in pyrochemical reprocessing: a spectroscopic study of the chlorination of
molybdenum and its oxides in chloride melts, J. Nucl. Mater. 323 (2003)
93–100.
[14] HSC Chemistry software version 6.12, Outotec Research Oy, Pori, Finland,
1974–2007.
The rates of molybdenum oxide chlorination have been mea-
sured at temperatures between 798 and 873 K and chlorine partial
pressures ranging from 10 to 55 kPa. The chlorination reaction
started at 770 K, and the formation of MoO2Cl2 as predicted by
thermodynamic considerations was confirmed by XRD.
[15] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids,
Academic Press, 1999.
For both type of samples used in this work (MoO3(I) and (II)),
the experimental conditions were achieved to measure the reaction
rate under intrinsic kinetic rate controlled regime. The depen-
dences of rate on gaseous reactant pressure, solid conversion and
temperature were determined. Two intrinsic parameters were
obtained: the activation energy of about 211 kJ mol−1 and the reac-
tion order with respect to chlorine partial pressure of 1, while the
Arrhenius pre-exponentional factor was estimated. Finally, a com-
plete rate equation was formulated for each solid which differ in the
[16] S. Brunaver, P.H. Emmete, E. Teller, Adsorption of gases in multimolecular lay-
ers, J. Am. Chem. Soc. 60 (1938) 309–319.
[17] J. Szekely, J.W. Evans, H.Y. Sohn:, Gas-Solid Reactions, Academic Press, New
York, NY, 1976, pp. 232 and 115–117.
[18] A.W. Hills, The importance of convective mass transfer in the reduction of
hematite, Metall. Trans. B 98B (1978) 121–128.
[19] G. De Micco, A.E. Bohé, H.Y. Sohn, Intrinsic kinetics of chlorination of WO3
particles with Cl2 gas between 973 K and 1223 K (700 ◦C and 950 ◦C), Metall.
Mater. Trans. B 42 (2) (2011) 316–323.
[20] J.O. Hirschfelder, C.F. Curtiss, R.B. Bird:, Molecular Theory of Gases and Liquids,
Wiley, New York, USA, 1956.