Kinetics of salt roasting of chalcopyrite using KCl

Article abstract of DOI:10.1016/S0040-6031(00)00475-5

The process of salt roasting appears to be very promising because it lowers the reaction temperature and leads to water soluble chloride salt. This paper presents the kinetics of salt roasting of chalcopyrite using KCl as the chlorinating agent. Isothermal thermogravimetric studies were carded out in the temperature range of 523-773 K under oxygen, static air and argon atmospheres. It is observed that the weight gain is more in the presence of oxygen than in static air. Weight loss is observed when salt roasting is carried out under an inert atmosphere. It is deduced that the reaction is chemically controlled at temperatures below 600 K both under static air and oxygen atmosphere. At higher temperatures, the process is not thermally activated. The weight gain decreases with increase in temperature because of a change in the chemistry of the process. The roasting experiments carried out in a horizontal electric tube furnace under an oxygen atmosphere shows that up to 96% Cu can be recovered from the chalcopyrite by salt roasting with KCl and subsequent water leaching. (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S0040-6031(00)00475-5

Thermochimica Acta 362 (2000) 25±35
Kinetics of salt roasting of chalcopyrite using KCl
M. Chakravorttya, S. Srikanthb,*
aNon-Ferrous Process Division, National Metallurgical Laboratory, Jamshedpur, 831007 Bihar, India
bNML Centre, CSIR Madras Complex, Chennai, 600113 TN, India
Received 21 October 1999; accepted 17 February 2000
Abstract
The process of salt roasting appears to be very promising because it lowers the reaction temperature and leads to water
soluble chloride salt. This paper presents the kinetics of salt roasting of chalcopyrite using KCl as the chlorinating agent.
Isothermal thermogravimetric studies were carried out in the temperature range of 523 773 K under oxygen, static air and
argon atmospheres. It is observed that the weight gain is more in the presence of oxygen than in static air. Weight loss is
observed when salt roasting is carried out under an inert atmosphere. It is deduced that the reaction is chemically controlled at
temperatures below 600 K both under static air and oxygen atmosphere. At higher temperatures, the process is not thermally
activated. The weight gain decreases with increase in temperature because of a change in the chemistry of the process. The
roasting experiments carried out in a horizontal electric tube furnace under an oxygen atmosphere shows that up to 96% Cu
can be recovered from the chalcopyrite by salt roasting with KCl and subsequent water leaching. # 2000 Elsevier Science
B.V. All rights reserved.
Keywords: Kinetics; Chalcopyrite; Chloridising roasting
1. Introduction
1929 [1]. Kershner and Hoertel [2] reported that >95%
of Co, Ni and Cu could be extracted from cobalt and
nickel bearing complex sulphide concentrate in the
presence of sodium chloride after leaching the roasted
product with hydrochloric acid at unit pH. Dahlstedt
et al. [3] have carried out a thermodynamic analysis of
salt roasting of lean complex sulphide concentrate in
the presence of NaCl and O2 at 1000 K. They have
suggested that the condensed chlorides of valuable
metals can be separated out from the gangue materials.
Rajmohan and Jacob [4] suggest that the presence of
SO2 is essential in the salt roasting of lean oxide ores
to convert NaCl to Na2SO4 which is thermodynami-
cally more stable, whereas in case of sulphides, SO2 is
produced in situ by roasting. Bayer and Weidemann
[7] have carried out extensive studies on salt roasting
of common sulphides and chalcopyrite at low tem-
peratures (473±513 K) by thermoanalytical technique.
Various pyro- and hydro-metallurgical processes
for the extraction of metals from sulphides have been
practised for several decades. Chloridising roasting of
sulphides under an oxidising atmosphere with com-
mon salts for the extraction of metals has continued to
remain a subject of interest. The process is interesting,
since it produces soluble chloride of metals with
simultaneous rejection of iron.
Chloridising roasting of metal oxide and sulphide in
the presence of common salt has been reported by a
number of workers [1±7]. The processing of nickel
sulphide concentrate with MgCl2 was ®rst patented in
* Corresponding author. Tel.: ‡91-44-2350077;
fax: ‡91-44-2301027.
E-mail address: s_srikanth_99@yahoo.com (S. Srikanth).
0040-6031/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 0 - 6 0 3 1 ( 0 0 ) 0 0 4 7 5 - 5

26
M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
DG0 ˆ 711 064 ‡ 149:064T
They have observed that the mixture of KCl-CuFeS2
can be heated more rapidly to the reaction temperature
than other salts, such as NH4Cl, because KCl does not
volatilise during heating. Bayer and Weidemann [7]
have also observed that the salt-roasting reaction is
spontaneous for KCl, slower for NH4Cl and of vari-
able rate for NaCl. They have obtained a strong weight
gain in the presence of O2 (41%), whereas no reaction
was observed in nitrogen atmosphere. They also
reported that, at a higher temperature (>553 K), the
Fe-sulphate and Cu-chlorides decomposed with
weight loss. Ngoc et al. [6] have studied the salt
roasting of Sikkim complex sulphide concentrate
using NaCl at low temperatures (523 K). They have
concluded that the process is suitable for the extraction
of metal values from the off-grade concentrate and the
chemistry of the process is very complex since a
number of reactions involving oxidation, sulphation
and in situ chlorination may take place simultaneously
or sequentially. The principle reaction is:
CuO ‡ FeO ‡ 4KCl ‡ 2SO2 ‡ O2
ˆ CuCl2 ‡ FeCl2 ‡ 2K2SO4
(3)
DG0 ˆ 665 055 ‡ 538:7587T
CuFeS2 ‡ 4KCl ‡ 4O2 ˆ CuCl2 ‡ FeCl2 ‡ 2K2SO4
(4)
DG0 ˆ 1 376 119 ‡ 687:822T
4FeCl2 ‡ 32 O2 ˆ 2FeCl3 ‡ Fe2O3 ‡ Cl2
DG0 ˆ 72 865 ‡ 49:09T
(5)
(6)
2FeCl3 ‡ 32 O2 ˆ Fe2O3 ‡ 3Cl2
DG0 ˆ 121 736 ‡ 19:833T
All these reactions are exothermic and are feasible at
low temperatures. Reaction (4) is associated with a
weight gain of 26.6%, whereas, reactions (5) and (6)
are associated with weight loss. Bayer and Weide-
mann [7] also reported that various other species like
KFe(SO4)2, K2CuSO4Cl2 and K2CuCl4 may form
during the roasting of chalcopyrite with KCl in the
presence of oxygen. In addition, the phases CuSO4 and
(CuO)CuSO4 and other complex sulphate phases may
also form depending on the prevailing oxygen poten-
tial. In general, the overall stoichiometry of the salt
roasting reaction can be expressed as:
MS ‡ 2NaCl ‡ 2O2 ˆ MCl2 ‡ Na2SO4
(1)
On account of the complexity of sulphide concentrate
and because of the availability of oxygen during salt
roasting, a number of reactions other than the principle
reaction may take place sequentially or simulta-
neously. Dahlstedt et al. [3] have reported in their
study that there are several paths through which each
sulphide mineral may undergo reaction and the kinetic
factors in¯uence the roasting process. In this study,
the kinetics of salt roasting of chalcopyrite from
the Singhbum district in India using KCl as the
chlorinating agent were carried out by thermogravi-
metry as well as by analysing the leach recovery of
copper obtained from salt roasting ± water leaching
experiments.
CuFeS2‡2KCl‡ 15 O2
4
ˆ CuCl2‡ 12 Fe2O3‡K2SO4‡SO2
(7)
or
2CuFeS2‡8KCl‡ 17 O2
2
ˆ 2CuCl2‡Fe2O3‡4K2SO4‡2Cl2
(8)
The stoichiometric mass gain in reaction (7) is 16.8%,
and in reaction (8) it is 13.7%. It is dif®cult to describe
the salt-roasting reactions by simple chemical equa-
tions, because a number of reactions take place
simultaneously and/or sequentially. Therefore, the
equilibrium condensed and gas phase compositions
for the salt roasting of CuFeS2 with KCl (1.8 times the
stoichiometry as indicated by reaction 7 which corre-
sponds to 0.9 times the stoichiometric amount as
indicated by reaction (8) under an atmosphere of
static air at temperatures in the range of 523±773 K
were determined through free energy minimisation
2. The chemistry of salt roasting
The chemistry of salt roasting is very complex
because several processes such as oxidation, sulpha-
tion, in situ chlorination and evaporation of volatile
species occur either simultaneously or sequentially.
The various possible reactions that could occur during
the salt roasting of chalcopyrite using KCl are [3,5±7]:
CuFeS2 ‡ 3O2 ˆ CuO ‡ FeO ‡ 2SO2
(2)

M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
27
Table 1
calculations using the SOLGASMIX programme [8]. All
the relevant condensed and gas phases whose thermo-
dynamic data is available in the literature was included
for the free energy minimisation calculations. Ther-
modynamic data for all the species were taken from
standard compilation [9,10]. Since thermodynamic
data is not available for the species KFe(SO4)2,
K2CuSO4Cl2 and K2CuCl4, they could not be included
for the calculations.The results of the free energy
minimisation calculations indicate the formation of
the condensed phases K2SO4, CuCl2, Fe2O3 and
CuSO4 at 523 and 573 K, K2SO4, CuCl2, Fe2O3
and (CuO)CuSO4 at 573 K and K2SO4, CuCl,
Fe2O3 and (CuO)CuSO4 up to 773 K. The species
CuCl is liquid in the temperature range 696±823 K
and FeCl3 exists as a liquid between 577 and 603 K
and as a gas above 603 K. The gas phase comprises
predominantly of O2, Cl2, (CuCl)3 and SO3 and small
amounts of KCl, FeCl3 and SO2. The free-energy
minimisation calculations predict a maximum weight
gain of 17.95% up to 600 K and a decrease in weight
thereafter. Non-availability of thermodynamic data for
the complex species, such as KFe(SO4)2, K2CuSO4Cl2
and K2CuCl4, is a serious limitation towards the
analysis of the thermodynamics of the salt-roasting
process using free energy minimisation calculations.
Analysis of chalcopyrite concentrate used for salt-roasting experi-
ments
Constituents
Chemical analysis
(wt.%)
EDAX analysis
(wt.%)
Cu
31.5
35.33
31.43
33.24
Fe
31.09
35.65
1.61
S
SiO2
MgO
Al2O3
0.18
0.052
Kinetic studies were carried out by thermogravimetry
under isothermal conditions in the temperature range
523±773 K. The conditions adopted for the thermo-
analytical measurements are given in Table 2. About
50 milligrams of KCl and chalcopyrite of required
stoichiometry were mixed thoroughly in an agate
mortar prior to the experiment. The samples were
3. Experimental
The chalcopyrite received from Hindustan Copper,
Ghatsila, Bihar, India, was used for the salt-roasting
experiments. Re¯otation of the concentrate was done
to remove the impurities. The X-ray diffractogram of
the concentrate indicated predominantly the presence
of chalcopyrite and small amounts of pyrite. The
chemical analysis of the chalcopyrite concentrate used
is given in Table 1. The scanning electron micrograph
of the concentrate is shown in Fig. 1. The elemental
composition of the concentrate determined from
EDAX analysis is also given in Table 1. Pure analar
grade anhydrous KCl was used for the salt-roasting
experiments.
3.1. Thermoanalytical experiments
Thermoanalytical studies were carried out in a
simultaneous TG/DTA (SEIKO, model No. 320).
Fig. 1. SEM photograph of the chalcopyrite concentrate used for
salt-roasting experiments.

28
M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
Table 2
Experimental conditions of thermoanalytical measurements
Amount of sample material: 50 mg of CuFeS2 ‡ KCl
Reference material
Annealed a-Al2O3
0.104 mm
Mean particle size of the test material:
Situation of sample material
Furnace atmosphere
Thermocouples exactly in the centre of the sample with direct contact to the heated substance
O2 and Ar at a flow rate of 150 ml/min
Pt crucible, F 5 mm
Pt±Pt/Rh 13%
100oC/min, 523 723 K
Sample holder
Thermocouples
Heating rate and temperature
Packing density
Loose packed, no pressing
heated up to the reaction temperature at a heating rate
of 1008C/min for all the isothermal experiments.
Experiments were carried out in static air, oxygen
and argon atmosphere at a ¯ow rate of 150 ml/min.
Samples were held for 3 h at the speci®ed temperature
in each isothermal roasting experiment. To characte-
rise the salt-roasting process, in one of the non-iso-
thermal experiments, a heating rate of 38C/min was
adopted for better resolution and interpretation of the
TG/DTA plot.
out on the leach residue. XRD studies could not be
conducted on the roasted product because of its
extreme hygroscopic nature.
4. Results and discussion
4.1. Non-isothermal TG/DTA studies
The differential thermal analysis plot of the chal-
copyrite mixed with 80% excess KCl (excess of
stoichiometry as indicated by reaction (7) or 0.9 times
the stoichiometric amount as indicated by reaction (8))
up to 700 K under an oxygen atmosphere is shown in
Fig. 2. The DTA plot indicates that the reaction
initiates at around 550 K. Two major exothermic
peaks are observed at 580 and 603 K, two minor peaks
at 626 and 652 K and a small endothermic peak at
704 K. The two major exothermic peaks probably
3.2. Tube furnace roasting experiments
Salt-roasting experiments on chalcopyrite mixed
with KCl in the desired stoichiometry was also per-
formed in a horizontal electric Inconel tube furnace.
These roasting experiments were carried out in the 3±
4 g scale. KCl and CuFeS2 of required stoichiometry
were mixed thoroughly and taken in a pre-weighed
alumina boat. The weight of the alumina boat with
charge was recorded and the boat was introduced
inside the hot zone of the furnace after the temperature
had equilibrated at the set value. The temperature of
the sample was monitored using a Chromel±Alumel
thermocouple placed in contact with the alumina boat.
All the tube furnace roasting experiments were carried
out under a ¯owing oxygen atmosphere at a ¯ow rate
of 200 ml/min. After each roasting experiment, the
roasted mass was leached with distilled water at 708C
to bring all the copper into the solution. The leach
solution was ®ltered and the residue was collected and
kept in an oven for 4 h at 105 Æ 28C. The amounts of
Cu and Fe in the leach liquor were determined by
atomic absorption spectrophotometer and the amounts
‡
of K , Cl and SO4 were determined by conven-
tional analysis. X-ray diffraction analysis was carried
Fig. 2. The differential thermal analysis plot for the salt roasting of
chalcopyrite under an oxygen atmosphere.

M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
29
well with the theoretical weight gain computed for
reaction (4). However, the thermogravimetric results
do not support the occurrence of either the overall
reaction (7) which indicates a theoretical weight gain
of only 12.4% or the overall reaction (8) which is
associated with a weight gain of 18%. The differential
thermogravimetric plot has the same features as the
DTA plot.
The differential thermal analysis plot of the chal-
copyrite mixed with 80% excess KCl under an inert
atmosphere (Ar) indicates an endothermic peak proba-
bly corresponding to the dissociation of chalcopyrite
according to the reaction:
2CuFeS2 ! Cu2S‡2FeS ‡ Sꢀg† DH ˆ ‡169 803 J
(9)
Fig. 3. Thermogravimetric plot for the salt roasting of chalcopyrite
under an oxygen atmosphere.
4.2. Isothermal thermogravimetric studies
correspond to the oxidation of chalcopyrite as repre-
sented by reaction (2) and the chlorination reaction as
represented by reaction (3). The two minor peaks
probably correspond to the sulphation and other com-
plex phase formations. The small endothermic peak at
704 K corresponds to the melting of CuCl. These
observations are substantiated by the TG and DTG
results shown in Figs. 3 and 4, respectively. The TG
results indicate two distinct regions, the oxidation and
chlorination reactions (reactions (2±4)) between 550
and 620 K with a weight gain of 22.8%, and the
sulphation and other complex reactions between
650 and 700 K with a weight gain of 6.3%. The weight
gain of 22.8% for the ®rst stage of the reaction agrees
4.2.1. Effect of salt
The effect of the salt (KCl) quantity on the roasting
of chalcopyrite in a static air atmosphere at 623 K is
shown in Fig. 5. Under oxygen atmosphere, the weight
gain approaches the maximum even at low concentra-
tions of KCl and within a short time. In the case of
static air, it is observed that, at low amounts of KCl,
there is no reaction even up to 1 h of roasting. On the
contrary, a minor weight loss is observed for higher
periods of roasting probably due to the partial decom-
position of chalcopyrite or the evaporation of KCl.
The rate of weight gain increases with increase in salt
Fig. 4. Differential thermogravimetric plot for the salt roasting of
chalcopyrite under an oxygen atmosphere.
Fig. 5. Effect of salt (KCl) quantity on the roasting of chalcopyrite
under an oxygen atmosphere.

30
M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
30 min. The extent of weight gain increases with
temperature up to 598 K. A maximum weight gain
of 21.7% is observed, which is close to the value of
18% calculated for the formation of K2SO4, Fe2O3,
CuCl2 and some amount of CuSO4 from free energy
minimisation calculations. From Fig. 7 it is seen that,
as the temperature is increased further, the net weight
gain decreases. At 723 K, the net weight gain is only
8%. At temperatures above 623 K, a small weight loss
(up to 1.5%) is observed initially during the heating
period, after which a weight gain is recorded. The
initial weight loss may be attributed to the partial
decomposition of CuFeS2 and some evaporation of
KCl. The decrease in weight gain at temperatures
>600 K can be attributed to the change in the chemi-
stry of the process as predicted by the free-energy
minimisation calculations. Up to 590 K, the con-
densed phases predicted to form are K2SO4, CuCl2,
Fe2O3 and CuSO4. Above 598 K, (CuO)CuSO4 forms
instead of CuSO4 and at temperatures above 623 K,
CuCl forms instead of CuCl2. It may, however, be
mentioned that the formation of complex species such
as KFe(SO4)2, K2CuSO4Cl2 and K2CuCl4, which have
been reported to form during the salt roasting of
chalcopyrite with KCl [7], have not been taken into
account in the free-energy minimisation calculations.
At temperatures >700 K, the samples were sticky and
tend to ¯ow out of the crucible because of the melting
of CuCl.
The effect of temperature in the range of 513±593 K
and time on the salt roasting of chalcopyrite under a
¯ow of oxygen is shown in Fig. 8. At temperatures
>600 K, the sample melted and reacted with the Pt
crucible and, therefore, salt-roasting experiments
under a ¯ow of O2 could not be carried out >600 K.
Both, the total weight gain as well as the rate of weight
gain was more under a ¯ow of oxygen than under a
static air atmosphere. In this case also, there was
virtually no reaction at 523 K (weight gain of 2.1%)
even after two-and-a-half hours of roasting. However,
the rates of reaction were very high at 563 and 593 K.
The weight gain was slightly more at 563 K (22.2%)
than at 593 K (20.2%), probably because of the
decomposition of CuCl2 to CuCl and the oxidation
of FeCl2 at higher temperatures. Under oxygen atmo-
sphere, there was a strong weight loss during the initial
period (3.9% at 563 K and 15.5% at 593 K) before
weight gain was observed. Under a strong oxidising
Fig. 6. Effect of temperature and time on the salt roasting of
chalcopyrite under a static air atmosphere for low temperatures.
quantity and time. At higher concentrations of KCl,
the roasted mass tended to become sticky and effuse
out of the crucible.
4.2.2. Effect of temperature
The effect of temperature and time on the salt
roasting of chalcopyrite in a static air atmosphere is
shown in Figs. 6 and 7. At 523 K, virtually no reaction
or weight gain was observed even after two-and-a-half
hours of roasting. At 535 K, a small weight gain of up
to 4.5% was observed, probably corresponding to the
partial oxidation of chalcopyrite. However, at 573 K,
both the rate and extent of weight gain increases
strongly. Up to 18.3% weight gain is observed within
Fig. 7. Effect of temperature and time on the salt roasting of
chalcopyrite under a static air atmosphere for high temperatures.

M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
31
Fig. 8. Effect of temperature and time on the salt roasting of
chalcopyrite under an oxygen atmosphere.
Fig. 10. Effect of gas atmosphere on the salt roasting of
chalcopyrite.
atmosphere, the formation and evaporation of ferric
chloride may contribute to the weight loss in addition
to the other processes mentioned previously.
to form in the temperature range 523±723 K under an
inert atmosphere were CuCl, FeS2 and FeCl2 other
than the unreacted CuFeS2 and KCl. The gas phase
contains small amounts of S2, KCl, FeCl2 and Cu3Cl3
in addition to Ar. The predicted mass loss is in good
agreement with the observed values. The relative
effect of oxygen, static air and argon on the salt-
roasting process is demonstrated in Fig. 10. Both
the rate of weight gain and the net weight gain is
more with the use of a pure oxygen atmosphere.
4.2.3. Effect of gas atmosphere
Salt-roasting experiments were carried out under
argon, static air and oxygen atmospheres to determine
the effect of oxygen potential on the salt-roasting
process. The results of the thermogravimetry experi-
ments under a ¯owing Ar atmosphere at three different
temperatures are shown in Fig. 9. Under an argon
atmosphere, there is virtually no reaction and there is a
weight loss (up to 4%) which decreases exponentially
with time. The condensed phases that were predicted
4.3. Tube furnace roasting
4.3.1. Effect of particle size and bed depth
The effect of particle size on the recovery of copper
derived from the salt-roasting process is given in
Table 3. The mean particle size of chalcopyrite was
varied in the range of 0.07 to 0.152 mm. It is observed
that copper recovery improves with decrease in par-
ticle size. However, at particle size <0.1 mm, the
roasted mass was sticky and sintered to the alumina
boat and, consequently, leach recovery of Cu dropped.
Table 3
Effect of particle size on salt roasting of chalcopyrite
Mean particle size (mm)
Leach recovery of Cu (%)
0.152
0.104
0.070
21.00
57.00
44.00
Fig. 9. Results on the salt roasting of chalcopyrite under an inert
(Ar) atmosphere.

32
M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
Table 4
Effect of quantity of charge on the salt-roasting process
Time (min)
3-g scale
4-g scale
Cu recovery (%)
Fe recovery (%)
Cu recovery (%)
Fe recovery (%)
60
120
50.5
71.0
36
27
51
69
37
30
It is seen that the Cu recovery is maximum for
0.104 mm mean particle size.
recovery is 52% in 1 h of roasting at 573 K, whereas
it is 50% with 50% excess of KCl under the same
conditions of roasting.
In order to investigate the effect of bed depth on the
salt-roasting process, experiments were carried out
using different quantities of charge from 3 to 8 g.
Some of these results are presented in Table 4. It
appears that bed depth does not have a signi®cant
in¯uence on the salt-roasting process within the range
studied. Therefore, all the subsequent salt-roasting
experiments were carried out on the 3-g scale with
a mean particle size of 0.104 mm.
4.3.3. Effect of roasting time
The effect of time of salt roasting on the leach
recovery of Cu at several temperatures is delineated in
Fig. 12. It is seen that the leach recovery of copper
increases with time up to 180 min after which a steady
state is reached. One of the objectives of the salt-
roasting process under an oxidising atmosphere is to
convert all the copper to its chloride and the iron to
oxide form to minimise the amount of iron in the leach
liquor in the subsequent water leaching stage. The
amount of Cu and Fe in the leach solution for different
times of roasting is depicted in Fig. 13. It is seen that
the recovery of Fe in the leach solution is <10% for
roasting periods above 180 min.
4.3.2. Effect of salt quantity
The amount of salt required for the good recovery of
Cu is one of essential factors for the salt roasting of
chalcopyrite. Accordingly, a set of experiments were
carried out by varying the amount of KCl from 25 to
80% excess of stoichiometry (based on reaction (7)) at
573 K for 1 h at a ¯owing oxygen ¯ow rate of 200 ml/
min. The results are presented in Fig. 11. It is seen that
the leach recovery of copper increases with increase in
KCl content up to about 50% and thereafter remains at
a steady value. For 80% excess of KCl, the Cu
4.3.4. Effect of roasting temperature
The effect of roasting temperature in the range of
523 to 623 K on the leach recovery of Cu is depicted in
Fig. 12. In this temperature range, the Cu recovery
Fig. 11. Effect of salt (KCl) quantity on the leach recovery of
copper.
Fig. 12. Effect of temperature of salt roasting on the leach recovery
of Cu.

M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
33
gravimetric results indicate the formation of complex
sulphates with a strong weight gain. X-ray diffraction
studies could not be carried out on the roasted product
because of its extreme hygroscopic nature. The con-
‡
2
centrations of K , Cl and SO4 in the leach liquor
for various conditions of roasting are given in Table 5.
The high concentration of sulphate ion in the leach
liquor also indicates some copper sulphate and other
complex sulphate and chloride formation during salt
roasting. The X-ray diffraction analysis of the leach
residue indicates predominantly the presence of
K2CuSO4Cl2 and minor amounts of Fe2(SO4)3Á10H2O
and KFe(SO4)2ÁH2O (Table 5).
Fig. 13. Effect of time of roasting on the copper and iron
concentrations in the leach liquor.
4.4. Kinetic analysis of the salt-roasting process
Various kinetic models were examined to ascertain
the mechanism of the salt-roasting process of chalco-
pyrite. Because of the decomposition of the chalco-
pyrite and the several gaseous species evolving during
the salt-roasting process, it may be expected that the
reaction product would be porous and the process may
be controlled by chemical reaction at the interface. For
such a process, the rate of roasting is proportional to
the interfacial area:
increased with increase in temperature. Here, up to
96% Cu could be recovered by salt roasting at a
temperature of 623 K for 5 h. This trend is somewhat
different compared to the thermogravimetric results.
As mentioned earlier, in the TG experiments under a
¯ow of pure oxygen, the weight gain decreased
slightly at temperatures >575 K. For roasting tem-
peratures >623 K, the roasted mass had fused and
seeped out of the alumina boat resulting in low Cu
recovery. This may be attributed to the formation of
liquid CuCl and some FeCl3. For example, a roasting
experiment carried out at 673 K for 2 h under a
¯owing O2 atmosphere yielded Cu and Fe recoveries
of 15 and 26%, respectively.
1 dW
Z dt
dV
dt
ˆ
 kA
where Z is the density, W the weight of chalcopyrite, V
the volume of the particle at time t, A the interfacial
area and k the rate constant. If the particles are
assumed to be spherical, the above equation assumes
the form:
It is dif®cult to distinguish whether copper is pre-
sent as chloride or sulphate in the roasted product
since both these species are water soluble. Free-energy
minimisation calculations indicate the formation of
some copper sulphate at all temperatures. Thermo-
1=3
‰1 ꢀ1 a† Š ˆ kt
Table 5
Ionic concentration in the leach liquor
‡
2
SO4 (g/l)
Temperature of roasting (K)
Time of roasting (min)
K
(g/l)
Cl (g/l)
573
573
623
623
623
673
673
673
60
120
60
4.99
3.42
3.61
4.48
3.42
3.20
4.68
6.31
13.84
12.07
13.13
13.13
12.7
12.50
13.16
11.84
8.96
9.05
8.63
7.89
6.9
120
240
30
10.65
13.49
13.49
60
120

34
M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
Fig. 14. Plot of [1 (1 a00)1/3] vs. time from thermogravimetric
data for the salt roasting of chalcopyrite under a ¯owing oxygen
atmosphere.
Fig. 15. The Arrhenius plot for the salt roasting of chalcopyrite
under a ¯owing oxygen atmosphere.
which is the topochemical kinetic model, where a is
the fraction reacted, k the rate constant and t the time.
For the thermogravimetry measurements, it is dif-
®cult to deduce the fraction reacted because of the
complexity of the salt-roasting process and, therefore,
the fractional weight gain (a0) was used instead of the
a parameter to denote the fraction reacted. Fig. 14
shows a plot of [1 (1 a0)1/3] as a function of time in
the temperature range 513 to 593 K derived from the
thermogravimetric data for salt roasting with 80%
excess KCl in a ¯owing oxygen atmosphere. The plots
are linear, indicating that the salt-roasting reaction
follows the shrinking core model in the temperature
range of 523±600 K. The rate constant, k, obtained
from the slope of the lines shown in Fig. 14, was used
to draw an Arrhenius plot shown in Fig. 15 from which
an apparent activation energy of 77 kJ/mol was
obtained for the salt-roasting process under an oxygen
atmosphere.
the salt-roasting process is not thermally activated at
temperatures >600 K.
A similar kinetic analysis can be performed using
the leach recoveries of Cu derived from the salt-
roasting experiments carried out in the tube furnace.
Fig. 16 shows the plot of [1 (1 a)1/3] versus time in
the temperature range based on the results obtained in
the roasting experiments conducted in the tube fur-
nace. In this case, a is the fractional recovery of copper
in the leach solution. The plots are linear over the
entire temperature range.
A similar kinetic analysis was carried out on the
thermogravimetric data for salt roasting with 80%
excess KCl in a static air atmosphere. The activation
energy for the salt-roasting process under a static
atmosphere at low temperatures was deduced to be
102 kJ/mol, which is understandably more than that
derived under an oxygen atmosphere. As mentioned
earlier, at temperatures above 600 K in a static air
atmosphere, the weight gain decreases with increase in
temperature because of a change in the chemistry of
the process at higher temperatures. This indicates that
Fig. 16. Plot of [1 (1 a0)1/3] vs. time from Cu recoveries using
the salt roasting±water leaching experiments under a ¯owing
oxygen atmosphere.

M. Chakravortty, S. Srikanth / Thermochimica Acta 362 (2000) 25±35
35
5. Conclusion
when salt roasting is carried out under an inert
atmosphere (Ar).
Based on the thermogravimetric and salt-roasting
studies carried out in the present investigation, the
following conclusions can be drawn:
5. Salt roasting in the tube furnace followed by water
leaching experiments indicates an increase in Cu
recovery with temperature up to 623 K and 3 h of
roasting. Up to 96% Cu could be recovered by salt
roasting at a temperature of 623 K beyond 3 h.
The amount of Fe in the leach solution decreases
with increasing times and temperatures of roast-
ing. The recovery of Fe in the leach solution is
<10% for roasting periods above 180 min.
6. X-ray diffraction studies of the leach residue as
well as ionic concentration analysis in the leach
liquor indicate that copper is present predomi-
nantly as chloride, but complex sulphates and
chlorides also exist.
7. The kinetics of the salt-roasting process conforms
to the chemical control mechanism up to 600 K
both under static air and pure oxygen atmo-
spheres. However, the activation energy of the
process is higher under a static air atmosphere
(102 kJ/mol) than under a pure oxygen atmo-
sphere (77 kJ/mol).
1. The chemistry of the salt-roasting process is very
complex because several processes, such as
oxidation, sulphation, in situ chlorination and
evaporation of volatile species, occur either
simultaneously or sequentially. The free-energy
minimisation calculations indicate the formation
of the condensed phases K2SO4, CuCl2, Fe2O3 and
CuSO4 at 523 and 573 K, K2SO4, CuCl2, Fe2O3
and (CuO)CuSO4 at 573 K and K2SO4, CuCl,
Fe2O3 and (CuO)CuSO4 up to 773 K. However,
the non-inclusion of complex species, such as
KFe(SO4)2, K2CuSO4Cl2 and K2CuCl4 which
have been reported to form during the salt roasting
of chalcopyrite with KCl [7], is a serious
limitation in the analysis of the chemistry of the
process.
2. The non-isothermal thermal analysis results dis-
tinctly indicate four exothermic processes at 580,
603, 626 and 652 K with a weight gain and a
minor endothermic process at 704 K These have
been correlated to the oxidation, chlorination,
sulphation and liquefaction reactions occurring in
the salt-roasting process.
3. The isothermal thermogravimetric studies in a
static air atmosphere indicate increase in weight
gain with increase in salt quantity up to 80%,
increase in temperature up to 600 K. A maximum
weight gain of 21.7% is recorded at 598 K. At
temperatures >600 K, the extent of weight gain
decreases with increase in temperature. This is
attributed to the change in the chemistry of the
process. A similar trend is observed in the
thermogravimetric measurements carried out un-
der a ¯owing oxygen atmosphere. A maximum
weight gain of up to 22.2% could be achieved
under oxygen atmosphere.
References
[1] E.L. Duf®eld, Nickel, iron, copper, etc. from sulphide ores,
British Patent, 328, 698 (1929).
[2] K.K. Kershner, E.W. Hoertel, Recovery of cobalt and nickel
from complex sulphide ores of South Eastern Missouri, U.S.
Bureau of Mines Rep. Invest. 5740, 1961, 16 pp.
[3] A. Dahlstedt, S. Seetharaman, K.T. Jacob, Scand. J. Metall.
21 (1992) 242±245.
[4] N. Rajmohan, K.T. Jacob, Min. Eng. 5 (2) (1992) 235±243.
[5] T.K. Mukherjee, P.R. Menon, P.P. Shukla, C.K. Gupta, J.
Metals (1985) 28±33.
[6] N.V. Ngoc, M. Shamsuddin, P.M. Prasad, Hydrometallurgy 21
(1989) 359±372.
[7] G. Bayer, G.H. Weidemann, Thermochim. Acta 198 (1992)
303±312.
[8] H. Flynn, A.E. Morris, D. Carter, Using the UMR PC/SOLGASMIX
Software Package, University of Missouri-Rolla, 1986.
[9] I. Barin, Thermochemical Data of Pure Substances, third edn.,
VCH Publishers, Weinheim, 1995.
4. Both, the rate of weight gain as well as the net
weight gain are higher under a ¯owing oxygen
atmosphere than in a static air atmosphere.
However, there is a small weight loss (up to 4%)
[10] D.R. Stull, H. Prophet: JANAF Thermochemical Tables,
second edn., U.S. Department of Commerce, Washington,
1975.