Kinetics of hematite chlorination with Cl2 and Cl2 + O2: Part I. Chlorination with Cl2

Article abstract of DOI:10.1016/j.tca.2009.08.007

Preliminary tests of the chlorination of two iron oxides (wüstite and hematite) in various chlorinating gas mixtures were performed by thermogravimetric analysis (TGA) under non-isothermal conditions. Wüstite started to react with chlorine from about 200 °C generating ferric chloride and hematite as the final reaction products. The presence of a reducing and oxidizing agent in the chlorinating gas mixtures influenced the chlorination reactions of both iron oxides, during non-isothermal treatment, only at temperatures higher than 500 °C. The chlorination kinetics of hematite with Cl₂ have been studied in details between 600 and 1025 °C under isothermal chlorination. The values of the apparent activation energy (E_a) were about 180 and 75 kJ/mol in the temperature ranges of 600-875 and 875-1025 °C, respectively. The apparent reaction order with respect to Cl₂ was found to be 0.67 at 750 °C. Mathematical model fitting of the kinetics data was carried out to determine the most probable reaction mechanisms.

Full text of DOI:10.1016/j.tca.2009.08.007

Thermochimica Acta 497 (2010) 52–59
Contents lists available at ScienceDirect
Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca
Kinetics of hematite chlorination with Cl and Cl + O : Part I.
2
2
2
Chlorination with Cl2
a,∗
a
a
a
b
a
N. Kanari , D. Mishra , L. Filippov , F. Diot , J. Mochón , E. Allain
a
Laboratoire Environnement et Minéralurgie, UMR 7569 CNRS, Nancy-University, ENSG, BP 40, 54501 Vandoeuvre-lès-Nancy Cedex, France
CENIM-CSIC, Av. Gregorio del Amo 8, Madrid 28040, Spain
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 May 2009
Received in revised form 19 August 2009
Accepted 25 August 2009
Available online 31 August 2009
Preliminary tests of the chlorination of two iron oxides (wüstite and hematite) in various chlorinating
gas mixtures were performed by thermogravimetric analysis (TGA) under non-isothermal conditions.
Wüstite started to react with chlorine from about 200 C generating ferric chloride and hematite as the
◦
final reaction products. The presence of a reducing and oxidizing agent in the chlorinating gas mix-
tures influenced the chlorination reactions of both iron oxides, during non-isothermal treatment, only at
◦
temperatures higher than 500 C.
The chlorination kinetics of hematite with Cl2 have been studied in details between 600 and 1025 C
Keywords:
Hematite
Thermogravimetric analysis
Chlorination
Kinetics
◦
under isothermal chlorination. The values of the apparent activation energy (Ea) were about 180 and
◦
7
5 kJ/mol in the temperature ranges of 600–875 and 875–1025 C, respectively. The apparent reaction
◦
order with respect to Cl2 was found to be 0.67 at 750 C. Mathematical model fitting of the kinetics data
was carried out to determine the most probable reaction mechanisms.
©
2009 Elsevier B.V. All rights reserved.
1
. Introduction
form of oxides or sulphides and it also constitutes an important
part of the primary ores of several valuable metals such as Cu, Zn,
Cr, Pb, Ti, Nb and Ta and of metallurgical wastes containing some
of these metals. The extraction of these metals and/or upgrading
of their ores by chlorination process depends on the amount and
nature of the iron compounds content. In some cases, such as from
laterites [4], jarosites [5], steel making waste dusts [6], and slags
from tin metallurgy [7], selective separation of nonferrous metals
are desirable, whilst in other cases, for example in the upgrading
of chromite [8], bauxite [9], ilmenite [10], etc., removal of Fe as
volatile chlorides has been proved to be highly beneficial. Selectiv-
ity in the processes, as mentioned above, is greatly dependent on
the nature of interaction between iron oxides and different chlori-
nating agents.
Chlorine technology is proved to be an efficient method for
the extraction of several non-common metals such as Zr, Hf, Nb,
Ta, Ti, rare-earth elements, the treatment of solid wastes, and the
upgrading of lean ores and minerals. The attractiveness of chlorine
metallurgy is essentially based on:
(
i) high reactivity of chlorine towards almost all metals, many
metal oxides and sulphides, etc.,
(
ii) appreciable differences in the boiling points of metal chlorides
and/or oxychlorides for a good number of metals, and
(
iii) low boiling points of metal chlorides and/or oxychlorides, etc.
The thermodynamic of Fe–O–Cl system are well known.
Furthermore, this item becomes easily available thanks to thermo-
chemical databases such as ‘HSC Chemistry’ [11] designed for many
different kinds of chemical reactions and equilibria calculations.
Whilst, the kinetics parameters are derived from experimental
studies. A literature survey on the chlorination of iron oxides
An overview concerning several chlorination methods devel-
oped by our laboratory in the past two decades was presented
recently [1]. Besides, various industrial practices of chlorine metal-
lurgy for several non-common metals, can be found from the review
work of Korshunov [2].
Iron is the fourth most abundant element, constituting about
.0% by weight of the earth crust [3]. It is found naturally in the
(wüstite, magnetite and hematite) in various chlorination medi-
5
ums was given in an early work [12]. Thus, only a summary of the
chlorination kinetics of hematite using the most usual chlorination
agents found in the literature is shown in Table 1.
∗
Corresponding author at: Laboratoire Environnement et Minéralurgie, UMR
569 CNRS, Nancy-University, ENSG, Institut National Polytechnique de Lorraine,
Rue du Doyen Marcel Roubault, BP 40, 54501 Vandoeuvre-lès-Nancy Cedex, France.
Tel.: +33 383 596 343; fax: +33 383 596 339.
The literature survey indicated that several reports are available
7
for the chlorination kinetics of hematite with Cl , but little infor-
2
mation is available for the oxychlorination behavior of hematite.
The values of the apparent activation energy of hematite chlorina-
E-mail address: ndue.kanari@ensg.inpl-nancy.fr (N. Kanari).
0
040-6031/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2009.08.007

N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
53
Table 1
Works devoted to the chlorination of hematite by different chlorinating agents.
◦
Chlorinating agent
C + Cl2
Temperature rangea ( C)
Observations
Reference
25–800
Formation of FeCl2 which volatilized as FeCl3.
[13]
◦
CO + Cl2
CO + Cl2
27–727
27–927
Reaction started at T > 577 C with formation of Fe2Cl6.
[14]
[15]
b
◦
Ea = 84 kJ/mol between 327 and 527 C.
◦
COCl2
COCl2
27–727
27–927
Reaction started at T > 427 C with formation of Fe2Cl6.
[14]
[15]
◦
◦
Ea = 88 kJ/mol between 237 and 297 C, Ea = 60 kJ/mol between 297 and 447 C.
◦
◦
c
◦
CCl4
CCl4
27–927
27–927
Ea = 125 kJ/mol between 397 and 597 C, n CCl4 = 0.5 at 527 C.
[16]
[15]
Ea = 130 kJ/mol between 452 and 552 C.
◦
Cl2
Cl2
Cl2
Cl2
Cl2
Cl2
Cl2
25–800
700–900
25–900
597 to >777
27–927
800–1200
600–950
Reaction started at 700 C generating FeCl3 and O2.
[13]
[17]
[18]
[19]
[15]
[20]
[21]
Ea = 136.8 kJ/mol, nCl2 = 0.85.
◦
Ea ≈ 96 kJ/mol between 700 and 900 C.
◦
◦
Ea = 188–191 kJ/mol between 597 and 777 C, Ea = 100 kJ/mol at T > 777 C.
◦
◦
Ea = 188 kJ/mol between 597 and 777 C, Ea = 100 kJ/mol between 777 and 807 C.
Ea = 167 kJ/mol, nCl2 = 1.
◦
◦
◦
Ea = 200 kJ/mol between 600 and 750 C, Ea = 125 kJ/mol between 750 C and 950 C
Cl2 + O2
Cl2 + O2
700–900
25–900
nCl2 = 2.75.
Oxygen had little effect from 0 to 40%.
[17]
[18]
◦
HCl
HCl
HCl
25–900
800–1000
25–750
Formation of FeOCl decomposing from 300 to 400 C.
[18]
[17]
[20]
Ea = 34 kJ/mol, nHCl = 1.
Formation of FeOCl, Ea = 5.9 kJ/mol.
a
When treatment starts from room temperature, non-isothermal conditions were also used.
Apparent activation energy.
Apparent reaction order.
b
c
tion varied from about 6–200 kJ/mol depending on the chlorinating
agent used and on the temperature range explored. Moreover, the
kinetics parameters obtained are generally dependent on the inves-
tigation methodology. For instance, research works regarding the
use of nominal flow rate of gases in order to minimize the star-
vation phenomena and mass transfers through the boundary layer
around the sample were seldom performed.
3. Results
3.1. Non-isothermal chlorination tests of iron oxides
Three experimental tests of wüstite sample chlorination were
performed under isothermal condition from 25 to 1000 C in
◦
Cl + CO, Cl + N and Cl + air gas mixtures having an equimolar pro-
2
2
2
2
The objective of the present study is to explore the chlorina-
tion and oxychlorination kinetics of hematite and to understand the
reaction mechanisms occurring between chlorine and iron oxides.
In this part, results of non-isothermal chlorination of wüstite and
hematite obtained under different atmospheres followed by the
portion of their constituents. The heating rate of the furnace was
about 25 C/min. The experimental results are grouped in Fig. 1 as
weight change (%) as function of temperature.
◦
A rough analysis of these plots suggests that the reaction of
wüstite with the three chlorinating gas mixtures started at tem-
◦
isothermal kinetics of hematite chlorination between 600 and
peratures as low than below 250 C and a maximum weight gain is
◦
◦
1
025 C are revealed and discussed.
reached at about 325 C. Since then, only losses of sample weight
◦
were observed up to temperatures approaching 500 C. The effect
of oxygen on the reaction of wüstite with chlorine seems to be lim-
2
. Materials and experimental procedures
The hematite sample of 99 pct purity used in this study was
obtained from Merck Chemicals (Eurolab, France). A sample of
wüstite was also used for some experimental tests. It was prepared
by hematite reduction using a controlled H –H O atmosphere. The
2
2
elemental composition and structure of both samples were checked
by scanning electron microscopy ‘SEM’ and X-ray diffraction ‘XRD’
analyses, respectively. The chlorination of iron oxides was mea-
sured by a thermogravimetric analysis (TGA) technique described
previously [22]. Its main unit is a CAHN 1000 microbalance hav-
ing a sensitivity of 10 ␮g. A sample of about 40 mg was uniformly
2
distributed in a quartz crucible having a section of about 0.5 cm .
This crucible is suspended to the balance by a quartz chain. During
the non-isothermal treatment, the sample was directly heated in
the chlorinating gas mixture by an electrical furnace. Data obtained
was plotted as evolution of % weight change as a function of tem-
perature. One may emphasize that all negative values represent %
weight losses (% WL), whilst positive ones express % weight gains.
For the isothermal runs, the sample was first heated in nitro-
gen atmosphere up to a given temperature before introducing the
reactive gases (Cl , or Cl + N ). The sample chlorination extent is
2
2
2
deduced from the percent weight change of the sample as a function
of time.
Fig. 1. Non-isothermal treatment of wüstite in different chlorinating gas mixtures.

5
4
N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
◦
Fig. 3. Vapor pressure evolution as a function of the temperature for ferrous and
Fig. 2. Phase stability diagram of the Fe–O–Cl system at 800 C.
ferric chlorides.
ited as the three curves almost overlap at temperatures lower than
of the furnace which was somewhat higher. If ferrous chloride
is formed (Eqs. (1) and (5)), it is probably transformed to ferric
chloride because of the high partial pressure of chlorine. Few boat
experimental tests performed earlier [12] about the chlorination of
◦
5
00 C. During the chlorination with Cl + CO, an additional con-
2
◦
tinuous weight loss was observed from 550 C and the full sample
had reacted at T < 1000 C. Whilst, a plateau in weight change was
observed up to about 750 C during chlorination with Cl + N and
◦
◦
2
2
wüstite in Cl + CO at 2 h showed that full reaction and volatiliza-
2
Cl + air, and about 50 and 42% WL of samples were observed at
2
◦
tion of the reaction products were achieved at 550 C indicating
◦
1
000 C, respectively.
that FeCl3 is the final chlorinated compound.
Tests of the chlorinating of wüstite for all chlorinating gas mix-
This study shows that the overall reaction of wüstite with chlo-
rine can be described by Eqs. (4) and (8). It seems that carbon
monoxide has little effect on the chlorination of wüstite at low tem-
peratures during non-isothermal treatment. Similar results were
reported in other investigations [14,15].
An hematite sample was also subjected to non-isothermal treat-
ments under conditions identical to those described above for
wüstite. Results obtained for the chlorination of wüstite allow to
◦
tures were stopped at 550 C and the partially reacted samples were
subjected to XRD analysis. Only hematite (Fe O ) was identified
2
3
as crystallized phase whichever was the chlorinating gas mixture
indicating the oxidation of Fe2 into Fe during chlorination.
+
3+
The reactions of ferrous oxide with Cl2 and Cl + CO to be
2
considered are shown by Eqs. (1)–(8), respectively. From a thermo-
dynamic point of view, all these envisaged reactions are favorable
◦
up to 1000 C [11]:
assume Eqs. (9) and (10) for the reaction of Fe O3 with Cl2 and
2
FeO + Cl → FeCl + 1/2 O
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Cl2 + CO, respectively. Data obtained for the non-isothermal chlo-
rination of hematite is charted in Fig. 4. Hematite seems to be
inert in chlorinating atmospheres up to about 550 C in spite of
2
2
2
2
8
2
/3 FeO + Cl → 2/3 FeCl + 1/3 O
◦
2
3
2
the presence of carbon monoxide. However, full carbochlorination
/3 FeO + Cl → 2/3 FeCl + 2/3 Fe O
2
3
3
4
◦
(
Cl + CO) of the sample was achieved at about 1000 C.
2
FeO + Cl → 2/3 FeCl + 2/3 Fe O
2
3
2
3
1
1
/3 Fe O + Cl → 2/3 FeCl + 1/2 O
(9)
2
3
2
3
2
FeO + Cl + CO → FeCl + 2/3 CO
2
2
2
/3 Fe O + Cl + CO → 2/3 FeCl + CO
(10)
2
3
2
3
2
2
5
4
/3 FeO + Cl + 2/3 CO → 2/3 FeCl + 2/3 CO
2
3
2
/3 FeO + Cl + 1/3 CO → 2/3 FeCl + 1/3 Fe O + 1/3 CO
2
3
3
4
2
/3 FeO + Cl + 1/3 CO → 2/3 FeCl + 1/3 Fe O + 1/3 CO
2
3
2
3
2
The phase stability diagram of the system Fe–O–Cl [11] is given
in Fig. 2. It suggests that at high chlorine partial pressure and low
partial pressure of oxygen the predominant stable phase is ferric
chloride. The ferrous chloride becomes a predominant specie only
at low partial pressure of chlorine. The evolution of vapor pressure
of ferrous and ferric chloride as a function of the temperature is
shown in Fig. 3 [23]. It is clear from this figure that FeCl3 possesses
◦
a high vapor pressure at temperatures close to 300 C. One may
mention that ferric chloride is volatilized as a dimer (Fe Cl ) at
2
6
low temperature and as a monomer (FeCl ) at high temperature.
3
◦
Ferrous chloride is in solid state up to about 700 C and it has a
vapor pressure nearing 1 atm only at 726 C.
◦
These thermodynamic considerations seem to be in good agree-
ment with the results of Fig. 1. In other words, wüstite reacted with
chlorine resulting to hematite and volatile ferric chloride as final
reaction products. The weight loss observed in Fig. 1 was shifted
◦
towards higher temperatures due to the heating rate (25 C/min)
Fig. 4. Non-isothermal treatment of hematite in different chlorinating gas mixtures.

N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
55
◦
Hematite starts to react with Cl + N from 700 C and only 30%
2
2
◦
of sample had reacted at 1000 C. Similarly, the reaction of Fe O
2
3
with Cl + air is obvious at high temperature and the reaction extent
2
◦
at 1000 C is somewhat lower than with Cl + N . This result may
2
2
be explained by the presence of oxygen shifting the reaction of
hematite chlorination (Eq. (9)) from the right to the left side of the
equation.
An examination of Figs. 1 and 4 indicates that the curves of the
chlorination of wüstite and hematite have same trend and shape
for a selected chlorinating gas mixture at temperatures higher than
◦
5
50 C. This is an indirect confirmation of Fe O formation during
2
3
the treatment of wüstite at low temperatures. Another information
from these non-isothermal tests is the formation of volatile ferric
chloride during the chlorination of iron oxides. With this assump-
tion, the isothermal chlorination of hematite at temperatures equal
◦
to and higher than 600 C leads to a complete gasification of the
reaction product. Consequently, the % WL of sample is directly
related to the reaction extent of hematite with chlorine.
3.2. Effect of chlorine flow rate
◦
Fig. 5. Isothermal treatment of hematite at 750 C for two chosen flow rates and
evolution of the reaction rate as a function of the reaction time.
In order to determine the intrinsic parameters of the gas–solid
reactions, it is suitable to minimize the effect of the external mass
transfer phenomena. This could be achieved by using a sufficient
flow rate of reactive gases. The effect of Cl2 gas flow rate on the
chlorination reaction rate of hematite is studied at 750 and 950 C.
The gas flow rate of chlorine was varied from 5 to 70 liter per hour
Fig. 6 summarizes the results of reaction rate evolution as a
function of chlorine velocity for both temperatures. As shown by
Fig. 6(a), the reaction rate at 750 C is almost independent from the
◦
◦
chlorine linear velocity at values higher than about 30 cm/min. As
◦
it could be expected, at 950 C, the chlorination rate of hematite
(
L/h).
Fig. 5 gives a typical example of evolution of weight change
is more dependent on the chlorine velocity (Fig. 6(b)). A chlorine
velocity higher than 50 cm/min will be required to minimize the
effect of mass transfer phenomena in the reaction. However, chlo-
rine with a Vg of 50 cm/min is used to investigate the effect of other
parameters on the chlorination of hematite.
(
right ordinate) of the sample as a function of time during hematite
◦
chlorination at 750 C at 5 and 70 L/h of chlorine. This figure con-
tains also the reaction rate evolution (left ordinate) versus time
for the two chosen flow rates of chlorine. The reaction rate is cal-
culated as −dWL/dt and expressed as −% WL/min. More than 80%
of sample has reacted for a reaction time of 80 min when a Cl2
flow rate of 70 L/h was used, whilst less than 48% of the reaction
were observed for the same time at 5 L/h of chlorine. As also shown
by Fig. 5, the reaction rate remains almost constant versus time
when a low flow rate was used, as illustrated by the linear aspect
of the weight change curve. This is a typical example of a reaction
governed by external mass transfer.
3.3. Effect of temperature
Series of isothermal tests were carried out at different temper-
◦
atures ranging from 600 to 1025 C using Cl2 at a constant chlorine
velocity of 50 cm/min. Fig. 7(a) and (b) shows the % weight change
◦
versus time curves obtained by the chlorination of Fe O . At 600 C
2
3
and a reaction time of 10 h, the weight loss is less than 25%. The reac-
tion rate increases steadily as the temperature rises between 600
Chlorine flow rates data were converted into chlorine linear
velocity ones ‘Vg’ taking into account the cross-section area of
◦
◦
and 1025 C and about 90 pct of sample are chlorinated at 950 C
for a reaction time of about 6 min.
−
1
the reactor and the obtained values are expressed as cm min
.
An average reaction rate is calculated for a reaction extent ‘X’:
The average reaction rate of the obtained isotherms is calcu-
lated using the initial almost linear stage for a reaction extent ‘X’
◦
0.05 ≤ X ≤ 0.40 for all isothermal data obtained at 750 and 950 C.
◦
Fig. 6. Effect of chlorine velocity on the reaction rate of Fe2O3 chlorination at 750 and 950 C.

5
6
N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
◦
Fig. 7. Isotherms of Fe2O3 chlorination between 600 and 1025 C.
(
0.05 < X < 0.40). The Arrhenius plot is given in Fig. 8. The apparent
where k = constant, t = chlorination time, X = extent of reaction
(ratio of weight of the reacted fraction to initial weight) and
Fp = particle shape factor (1 for infinite slabs, 2 for long cylinders
and 3 for spheres).
activation energy ‘Ea’ is 180 ± 4 kJ/mol for the temperature range
◦
of 600–875 C. Such a high value suggests that the chlorination
mechanism of Fe O3 is controlled by the chemical reaction in this
2
temperature range. As indicated by Fig. 8, the chlorination of Fe O
Eq. (11) is considered to describe a reaction controlled by the
chemical reaction in the case of shrinking nonporous particles
(with or without a solid porous product) and porous particles with
unchanged overall sizes. It also applies for a mechanism affected by
pore diffusion in the case of complete gasification of porous solids.
One may underline that the chlorination of hematite at the explored
temperature range belongs to the reactions of complete gasification
of solids. As mentioned in Section 3.1, iron tri-oxide (Fe O ) is con-
2
3
◦
beyond 875 C proceeds with an Ea value of about 74 kJ/mol. Sim-
ilar values of Ea, although in different temperature intervals, are
also reported in the literature. Bertoti et al. [15] found an apparent
activation energy of about 188 and 100 kJ/mol in the tempera-
◦
ture ranges of 597–777 and 777–807 C during the chlorination of
hematite by chlorine.
To correlate these activation energy values with a change of
reaction mechanism, a mathematical formulation of the experi-
mental data is attempted using the following equations [24]:
2
3
verted to iron tri-chloride (FeCl ) which is volatile at temperatures
3
◦
higher than 300 C.
Fig. 9(a) and (b) shows the mathematical fitting of the experi-
1
/Fp
1
− (1 − X)
= kt
(11)
(12)
(13)
(14)
mental data obtained during the chlorination of hematite between
◦
6
00 and 875 C by using Eq. (14). The correlation coefficient of
X = kt (for Fp = 1)
the data linearization is close to unity. This supports the hypoth-
esis that the chlorination reaction of Fe O3 particles is controlled
2
1
1
/2
/3
1
1
− (1 − X)
− (1 − X)
= kt (for Fp = 2)
= kt (for Fp = 3)
by the chemical reaction according to the shrinking sphere model
described by Eq. (14). Whilst, the best fitting data for temperatures
◦
higher than or equal to 900 C (Fig. 9(c)) is obtained by using Eq.
(
12). This result combined with the value of Ea ≈ 74 kJ/mol suggests
that the chlorination of hematite is probably diffusion controlled in
this temperature range. As mentioned in Section 3.2, the chlorine
velocity maintained at this temperature is lower than the opti-
mum one required for minimizing the mass transfer phenomena
(see Fig. 6(b)).
Typical examples of isotherms in two distinguished tempera-
ture regions are shown in Fig. 10 displaying the evolution of the
weight changes (right ordinate) and reaction rates (left ordinate)
as a function of the reaction time. Fig. 10(a) shows that the weight
change curve versus time follows mostly a one-third-power law
◦
function for the chlorination of hematite at 700 C. The deduced
reaction rate decreased rapidly with time, at least, for a weight loss
◦
up to 75 pct. As a contrast, plots of data obtained at 1000 C show
a linear dependency of weight change versus time with an almost
constant reaction rate during the first 3 min of chlorination. Again,
◦
these parameters’ variation at 1000 C must be indicative trends
only for a reaction controlled by a diffusion step.
3.4. Effect of partial pressure of chlorine
To determine this effect, a series of experimental tests was car-
◦
ried out at 750 C. Nitrogen was used as diluting gas and a total
gas velocity was kept at 50 cm/min. The chlorine partial pressure
◦
Fig. 8. Arrhenius diagram of Fe2O3 chlorination between 600 and 1025 C.

N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
57
Fig. 9. Mathematical fitting of data for the chlorination at: (a) and (b) T ≤ 875 ◦C using Eq. (14) and (c) T ≥ 900 ◦C using Eq. (12).
n
was varied from 0.20 to 1.00 atm. Fig. 11(a) gives the evolution of
weight change as a function of time for different chlorine par-
tion by chlorine. However, such value of ‘ Cl ’ seems to indicate
2
%
that the molecular chlorine could be dissociated before reacting
with hematite. Mathematical formulation of the experimental data
for different partial pressure of chlorine was established using Eqs.
(11)–(14). This data, shown by Fig. 12, matches Eq. (14) well with an
average correlation coefficient of 0.999 confirming that the rate of
the chemical reaction remains the rate controlling step of the over-
all process of chlorination for partial pressures of chlorine ranging
from 0.20 to 1.00 atm.
tial pressures. As it could be expected, the chlorination extent, for
a given reaction time, increased with partial pressure of chlorine
in the Cl + N gaseous mixtures. The evolution of reaction rate as
2
2
a function of Cl2 partial pressure is shown in Fig. 11(b) in natu-
ral logarithm scale. The apparent reaction order with respect to
n
chlorine ‘ Cl ’ derived from this figure is about 0.67. It appears
2
difficult to hypothesize about the paths of the hematite chlorina-
◦
Fig. 10. Reaction rate evolution versus time during hematite chlorination at 700 and 1000 C.

5
8
N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
◦
Fig. 11. Effect of chlorine partial pressure on the Fe2O3 chlorination at 750 C.
Acknowledgments
This work was performed in the frame of contract No. BRE2-
CT92-0173. Thanks to the financial support of the European Union
DG-XII).
(
References
[
1] 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
(
2009), in press.
[
[
[
[
2] B.G. Korshunov, Applications and potential use of chlorination methods in met-
allurgy of non-common metals, Metall. Rev. 8 (2) (1992) 1–33.
3] F.W. Clarke, The Relative Abundance of the Chemical Elements, vol. 11, Bull.
Phil. Soc., Washington, 1889, 131 pp.
4] S.B. Kanungo, S.K. Mishra, Kinetics of chloridization of nickel bearing lateritic
iron ore by hydrogen chloride gas, Metall. Mater. Trans. B 28B (1997) 389–399.
5] N. Kanari, A. Bonazebi, E. Allain, I. Gaballah, Decontamination of jarosite
through thermal treatment, in: Proceeding of the International Conference:
Recycling and Waste Treatment in Mineral and Metal Processing: Technical
and Economic Aspects, vol. 1, Luleå, Sweden, June, 2002, pp. 909–918.
6] L. Komorova, I. Imris, I. Novak, Chlorinating roasting of steel making waste
dusts, Zbornik Vedeckych Prac (Vysokej Skoly Technickej V. Kosiciach) 110
[
[
Fig. 12. Mathematical fitting of experimental data of Fig. 11(a).
(
1979) 97–109.
7] I. Gaballah, E. Allain, M. Djona, Extraction of tantalum and niobium compounds
from tin slags by chlorination and carbochlorination of pure oxides and con-
centrates, Metall. Mater. Trans. B 28B (1997) 359–369.
4
. Conclusions
[
[
8] N. Kanari, I. Gaballah, E. Allain, Use of chlorination for chromite upgrading,
Thermochim. Acta 351 (1–2) (2000) 109–117.
9] I. Szabo, A. Ujhidy, R. Jelinko, I. Vassanyi, Decrease of iron content of baux-
ite through high temperature chlorination, Hung. J. Ind. Chem. 17 (4) (1989)
1
. Wüstite reacted with chlorine at low temperatures generat-
ing hematite and ferric chloride as final reaction products. The
presence of carbon monoxide affected the chlorination rate for
hematite chlorination especially at high temperatures during
non-isothermal conditions.
465–475.
[10] F.C. Gennari, A.E. Bohe, D.M. Pasquevich, Effect of reaction temperature on the
chlorination of a Fe2O3–TiO2–C mixture, Thermochim. Acta 302 (1997) 53–61.
11] A. Roine, Outokumpu HSC Chemistry for Windows Version 3.0, Outokumpu
Research, Pori, Finland, 1997.
[12] N. Kanari, Extraction des métaux de valeur des concentrés de chalcopyrite et
de chromite par chloruration, Ph.D. Thesis, Institut National Polythechnique de
Lorraine, Laboratoire Environnement et Mineralurgie, Nancy, France, Novem-
ber 1995, 209 pp.
[13] Y. Okahara, I. Iwasaki, Chlorination of manganiferous iron ores, Trans. Soc.
Mining Eng. AIME 247 (1970) 73–80.
14] I.S. Pap, I. Bertoti, T. Szekely, I.Z. Babievskaya, L. Bottyan, TG studies on the reac-
tions of hematite, magnetite and wüstite with COCl2 and CO + Cl2, Thermochim.
Acta 92 (1985) 587–590.
[15] I. Bertoti, I.S. Pap, T. Szekely, I.Z. Babievskaya, Comparative thermogravimetric
study of chlorinations of hematite and wustite, J. Therm. Anal. 32 (1) (1987)
[
2
. Under isothermal conditions, hematite started to react with
◦
chlorine from at least 600 C and the reaction becomes very fast
◦
at temperatures approaching 1025 C.
3
. Two temperature regions have been distinguished for the chlo-
rination of hematite. The first one is comprised between 600
◦
and 875 C and is characterized by an apparent activation energy
[
of about 180 kJ/mol. Mathematical fitting of experimental data
indicates that the shrinking sphere model is the most ade-
quate to describe the reaction of hematite chlorination in this
temperature range. Beyond this temperature, there is a second
temperature dependency for which the overall reaction is mostly
controlled by diffusion phenomena with a value of Ea of about
281–292.
[16] I. Bertoti, I.S. Pap, T. Szekely, I.Z. Babievskaya, L. Bottyan, TG studies on the
reactions of hematite, magnetite and wüstite with CCl4, Thermochim. Acta 85
(
1985) 87–90.
7
4 kJ/mol.
[
17] G. Daradimos, U. Kuxmann, Chlorination of SnO2 and Fe2O3 with chlorine gas in
◦
4
. Chlorination of hematite at 750 C depended slightly on the par-
tial pressure of chlorine resulting to an apparent reaction order
◦
the temperature range from 700 to 1050 C, Erzmetall 24 (4) (1971) 163–172.
[18] R. Titi-Manyaka, I. Iwasaki, Thermogravimetric investigation of the chlorination
behaviors of some common metals and their oxides, Trans. Soc. Mining Eng.
AIME 252 (1972) 307–313.
19] V.I. Evdokimov, I.Z. Babievskaya, N.F. Drobot, I. Bertoti, I. Pap, T. Szekely, The
reaction of iron oxides with chlorine, Zh. Neorg. Khim. 30 (6) (1985) 1507–1512.
of 0.67. Decreasing the Cl partial pressure in the chlorinating gas
2
mixture from 1.00 to 0.20 atm does not affect the rate controlling
step of hematite chlorination.
[

N. Kanari et al. / Thermochimica Acta 497 (2010) 52–59
59
[
20] R.J. Fruehan, L.J. Martonik, Rate of chlorination of metals and oxides. Part III.
Annual Meeting ‘EPD Congress 1998’, TMS, San Antonio, TX, 1998, pp. 595–
609.
Rate of chlorination of Fe2O3 and NiO in Cl2 and HCl, Met. Trans. 4 (12) (1973)
2
793–2797.
[23] D.R. Lide, Ed., Handbook of Chemistry and Physics, 74th edition, CRC Press, Boca
Raton, New York, London, Tokyo, 1993–1994, 6–68 to 6–70.
[24] J. Szekely, J.W. Evans, H.Y. Sohn, Gas–Solid Reactions, Academic Press, New
York, NY, 1976, 68–70, 73–88, 109–131 and 232–235.
[
[
21] F.C. Gennari, D.M. Pasquevich, Kinetics of chlorination of hematite, Ther-
mochim. Acta 284 (1996) 325–339.
22] N. Kanari, I. Gaballah, Reactivity of magnesium oxide in different chlorina-
tion gas mixtures, in: B. Mishra (Ed.), Paper presented at the 127th TMS