68
P.S. Hlabela et al. / Thermochimica Acta 498 (2010) 67–70
temperature in a N2 atmosphere. If isothermal conditions were
Nomenclature
obtained, the CO/N2 mixture was introduced in such a way that
the total flow rate was 200 N mL/min. This was achieved by using 2
Brooks 5850 Mass Flow Controllers. All experiments were carried
out at a total pressure of 87.5 ( 1) kPa.
The isothermal runs were performed with different CO concen-
trations (2.4%, 4.8% and 9.6%), at different temperatures (850, 900,
950 and 1000 ◦C) for each concentration.
Eact
m
m0
mf
mt
n
pCO
R
t
activation energy (J mol−1
)
mass (mg)
initial mass (mg)
mass after reaction (mg)
mass at time t (mg)
order of carbon monoxide
partial pressure of CO (Pa)
universal gas constant (8.3144 J mol−1 K−1
time (min)
3. Results and discussion
)
T
temperature (K, ◦C)
3.1. Experimental data and processing of data
r0
X
initial reaction rate (s−1
conversion
)
An example of the raw data that is obtained from the TGA in the
conversion of BaSO4 to BaS at a final temperature of 850 ◦C in a 4.8%
CO in N2 mixture is given in Fig. 1.
Initially the gas atmosphere consisted of N2 only, and the tem-
perature was raised linear to the desired temperature, in this case
850 ◦C. During heating, the mass of the sample decreased slightly,
possible due to the release of moisture and other impurities. As the
temperature reached a constant value, the gas supply was shifted to
a CO/N2 mixture, in this case 4.8% CO. The shift in gas-supply caused
an abrupt but small change in both mass and temperature, which
can be seen in Fig. 1. Since the mass of the BaSO4 sample was not
exactly the same for all samples and some mass loss was observed
during heating, a comparison between different experiments was
done, by comparing the relative mass (m/m0), where m0 was taken
at the time of the change in gas atmosphere. A comparison of three
experiments, all carried out at 950 ◦C is given in Fig. 2.
solid carbon as a reducer and the effect that the partial pressure of
oxygen has on the kinetics of this reaction have been demonstrated
[7]. These studies have supported the assumption of gasification of
the solid carbon reducer to CO gas prior to BaSO4 reduction to BaS.
the reduction of BaSO4 using CO gas as a reducing agent, instead of
solid carbon. However, results on the decomposition of the cal-
cium analogue of BaSO4, i.e., CaSO4 in the CO atmosphere have
been previously reported [9]. This study concentrated on the effect
of the CO gas concentration, temperature and the heating rate on
the reduction of CaSO4. It was elucidated from this study that the
decomposition of CaSO4 to CaS and CO2, in the presence of CO,
begins at 780 ◦C, reaches a maximum at 865 ◦C and is complete at
955 ◦C. The DTA curves indicated that this reaction is exothermic
with a measured ꢀH value of 171( 33) kJ/mol.
It was concluded from these studies that by slow heating, in a
gaseous medium containing 20–100% of CO, CaSO4 is completely
reduced to CaS with maximal rate just below 865 ◦C. With CO con-
tent lower than 20% in the gas phase and at temperatures above
1050–1100 ◦C, the formation of CaO, SO2 and CO2 becomes domi-
nant and, correspondingly, the overall exothermic effect becomes
an endothermic effect [10]. It was further concluded that applica-
tion of inhibitors such as CO2 or O2 in the gas phase would prevent
the formation of CaS and hence promote the formation of CaO and
SO2.
The aim of this study was to investigate the reaction kinetics of
the reduction of barium sulphate with different concentrations of
CO at different temperatures, and to find a kinetic relation of the
reduction reaction. This information forms the basis on which the
optimisation of the full scale operation of the BaSO4 reduction to
BaS process will be based.
Fig. 1. Reduction of BaSO4 in a 4.8% CO/N2 mixture at 850 ◦C.
2. Experimental
2.1. Materials
A chemically pure BaSO4 (specially purified to >99.7%), from
BDH chemicals was used in all TGA experiments. The average
particle size of the BaSO4 range was 5 ( 1) m. A specialised pre-
pared CO/N2 gas mixture (9.6 ( 0.3)% CO in N2) and UHP N2-gas
(>99.999%), supplied by Afrox South Africa, were used.
2.2. Methods
The TGA measurements of the samples were performed by
recording the isothermal TG curves on a 2050 Du Pont TGA. The
BaSO4 samples (12–13 mg) were placed in a standard platinum
crucible (6 mm diameter and 5 mm depth). In all experiments the
sample was heated with a heating rate of 20 ◦C/min to the desired
Fig. 2. Reduction of BaSO4 with CO at 950 ◦C for different CO concentrations.