Abstract:
The Council for Scientific and Industrial Research (CSIR) is developing a novel process to
produce titanium metal at a lower cost than the current Kroll process used commercially. The
technology initiated by the CSIR will benefit South Africa in achieving the long-term goal of
establishing a competitive titanium metal industry.
A bubble column reactor is one of the suitable reactors that were considered for the
production of titanium metal. This reactor will be operated with a molten salt medium. Bubble
columns are widely used in various fields of process engineering, such as oxidation,
hydrogenation, fermentation, Fischer–Tropsch synthesis and waste water treatment. The
advantages of these reactors over other multiphase reactors are simple construction, good
mass and heat transfer, absence of moving parts and low operating costs.
High heat transfer is important in reactors when high thermal duties are required. An
appropriate measurement of the heat transfer coefficient is of primary importance for
designing reactors that are highly exothermic or endothermic.
An experimental test facility to measure wall heat transfer coefficients was constructed and
operated. The experimental setup was operated with tap water, heat transfer oil 32 and
lithium chloride–potassium chloride (LiCl–KCl) eutectic by bubbling argon gas through the
liquids. The column was operated at a temperature of 40 oC for the water experiments, at 75,
103 and 170 oC for the heat transfer oil experiments, and at 450 oC for the molten salt
experiments. All the experiments were run at superficial gas velocities in the range of 0.006
to 0.05 m/s. Three heating tapes, each connected to a corresponding variable AC voltage
controller, were used to heat the column media. Heat transfer coefficients were determined by inducing a known heat flux through the column
wall and measuring the temperature difference between the wall and the reactor contents. In
order to balance the system, heat was removed by cooling water flowing through a copper
tube on the inside of the column. Temperature differences between the column wall and the
liquid were measured at five axial locations.
A mechanistic model for estimating the kinematic turbulent viscosity and dispersion
coefficient was developed from a mechanism of momentum exchange between large
circulation cells. By analogy between heat and momentum transfer, these circulation cells
also transfer heat from the wall to the liquid.
There were some challenges when operating the bubble column with molten salt due to
leakages on the welds and aggressive corrosion of the column. The experimental results
were obtained when operating the column with water and heat transfer oil. It was found that
the heat transfer coefficient increases with superficial gas velocity. The values of the heat
transfer coefficient for the argon–water system were higher than those for the argon–heat
transfer oil system. The heat transfer coefficients were also found to increase with an
increase in temperature. Gas holdup increased with the superficial gas velocity. It was found
that the estimated axial dispersion coefficients are within the range of those reported in the
literature and the ratios of dispersion coefficients are in agreement with those in the
literature. The estimated kinematic turbulent viscosities were comparable with those in the
literature.