The in ﬂ uence of sulfur addition on the hazard-type reaction of ilmenite ores with sulfuric acid

The paper presents results of thermokinetic investigation of the hazard-type reaction of Norwegian and Australian ilmenite ores with sulfuric acid, modi ﬁ ed by the addition of elemental sulfur, to increase the process safety in industrial conditions. In the reactions of both ilmenite ores the addition of sulfur caused a reduction of the thermal power generated in the reaction and a decrease in the value of the thermokinetic parameter Δ T max / Δτ for almost the whole range of initial concentrations of sulfuric acid. It was also found that the addition of sulfur to the reaction did not negatively affect the degree of ilmenite leaching. The interpretation of the obtained thermokinetic curves allowed to determine safe process conditions for both types of titanium raw materials.


INTRODUCTION
In a number of chemical processes reaction kinetics should be controlled. It refers in particular to the reactions with a risk of uncontrolled run, called the hazard-type reactions. According to the defi nition, the hazard-type reactions are exothermic ones which can lead to thermal runaway 1, 2 . The reaction begins when the heat produced by the reaction exceeds the heat removed from the reactor. The excess heat raises the temperature of the reaction mass and increases the reaction rate, which increases the rate of heat production. An approximate criterion indicating the situation as dangerous is that the reaction rate -and hence the rate of heat generation -doubles with every ten degrees rise in temperature 3 . During the whole process, when the temperature rises, thermal runaway can occur. It happens because the rate at which heat is removed increases linearly but the rate at which heat is produced increases exponentially with temperature. Once the control of the reaction is lost, the temperature can rise rapidly leaving little time for correction. The reaction vessel may be at risk from over-pressurization due to violent boiling or rapid gas generation 4, 5 . One of the tools used to determine the process safety of hazard-type reactions is thermokinetics 6, 7 . It is a fi eld of science that in its research area deals with the change of thermal power over time, using calorimetry as a research method.
An example of a technology with a high risk of uncontrolled progress of the reaction is the production of titanium dioxide pigments 8- 10 . Titanium dioxide pigments are the most widely used white pigments 11 , and they are commercially produced by the sulfate and chloride processes 12, 13 . The main raw materials for sulfate process are ilmenite ores (mostly FeTiO 3 with TiO 2 content of 43-65%) and/or titaniferous slags (metallurgy-enriched ilmenite ores with TiO 2 content of 70-80%) 14 , whereas in the case of chloride process -synthetic rutile (with TiO 2 content of ~92%), produced from ilmenite ores in high-temperature hydrometallurgical leach process 15 .
In the sulfate process, described in detail elsewhere 14, 16-18 , ilmenites and/or slags are digested with highly concentrated sulfuric acid (80-95%) to produce liquor containing mainly titanyl and iron sulfates. After the separation of solid FeSO 4 · 7H 2 O, the liquor is hydrolyzed with water steam to initiate precipitation of hydrated TiO 2 which is then calcined at 800-900 o C to produce white solid TiO 2 . The fi rst stage of this process, the reaction of ilmenite with sulfuric acid, is classifi ed as a hazard--type reaction. The reaction is strongly exothermic and characterized by the possibility of a signifi cant amount of gas emission to the atmosphere, highly corrosive environment, high temperature of the process, and the possibility of real thermal explosion as a result of uncontrolled rate of reaction 9 .
This reaction is a crucial step in the production of titanium dioxide pigments, because it affects the efficiency of obtaining the fi nal product and its quality. Inadequately selected initial reaction conditions may result in incomplete conversion of the substrates (with large economic losses) and the risk of thermal explosion with potential fatalities 9, 19-21 . The priorities in titanium dioxide production, i.e. the effi ciency and safety of the process and the quality of the fi nal product, may be achieved by appropriate selection of the reaction conditions. However, improvement in the process effi ciency and safety can also be achieved by using appropriate additives 22 . Many attempts of using a variety of additives applied at different stages of titanium dioxide pigments production have been reported 23-25 . Nevertheless, there are no reports in the literature about the effect of any additives on the thermokinetics of this reaction.
The paper investigates the effects of the addition of sulfur on the reaction of ilmenites with sulfuric acid to increase the possibility of controlling the reaction rate. The main criterion for selecting an additive and the choice of sulfur for the reaction of ilmenite with sulfuric acid was to select an element that would not change the elemental composition of the substrates and would not adversely affect further stages of TiO 2 production and the properties of the fi nal product. The study aims to show that the addition of sulfur can reduce the thermokinetic parameters of the reaction, and thus reduce the risk associated with this reaction and increase the process safety.

EXPERIMENTAL
Two types of raw materials were selected for investigation of the effect of sulfur addition on the thermokinetics of reaction of ilmenite ores with sulfuric acid. The raw materials were derived from deposits located in Norway and Australia, and they were found to be signifi cantly different, both in phase and elemental composition 26, 27 . Moreover, the deposits of Norway ilmenite are among the largest in Europe, while those of Australian ilmenite are among the largest in the world. The contents of the main elements present in both ilmenite ores were determined using the X-ray fl uorescence method 26 and are presented in the form of oxides in Table 1 passing 40 μm sieves. The amount of sulfur added was at the level of 1%, 3% and 5% in relation to the mass of raw materials. The heat generated by the reaction mixture was recorded and then analyzed to determine the thermokinetic parameters of the reaction. After the completion of the reaction, the reaction mass was leached with deionized water, fi ltered, then the solution and the solid phase (leach residues) were analysed. No solidifi cation of the reaction mixture was observed. Iron and titanium in the solution were determined by classical titration analysis 29 . On this basis, the degree of Ti leaching in relation to the total Ti content in the raw material (wt%) was determined.

RESULTS AND DISCUSSION
The results of thermogravimetry measurements are presented in Fig. 1 and Fig. 2. The plots show the mass changes of the Norwegian and Australian ilmenite ores with and without the addition of sulfur, during heating up to 1000 o C. When heating both ilmenites without the addition of sulfur, an increase in mass is observed, initially, it is slow up to a temperature of about 400 o C, and then much faster to 800 o C. The increase in the samples' masses is due to the change in the level of iron oxidation that occurs in ilmenite ores 26 . The curves determined for both ilmenites with the addition of sulfur show a mass loss at 160-210 o C, followed by a slow increase in mass up to 400 o C, and then followed by a sharp mass increase up to 800 o C, as for the ilmenites without sulfur addition. Taking into account that the melting and boiling points of sulfur are 115°C and 444°C, respectively, and the fl ash point of sulfur lies between 165°C and 210°C, it can be concluded that the reduction in the sample mass, observed in the range of 160°C-210°C, can be caused by sulfur oxidation. The results indicate that the presence of sulfur does not affect further changes occurring in the samples. No signifi cant differences were found in the thermogravimetric curves of both materials. The shapes of experimental lines are similar and congruent changes occur in the same temperature ranges. The slightly larger weight loss observed for Norwegian ilmenite than for Australian one is a natural consequence of different composition of these materials.  Ilmenite ores were ground in a Pulverisette 5 planetary mill (Fritsch GmbH, Idar-Oberstein, Germany) to achieve about 92% of the product with a particle fraction below 40 mm. The ground product was mixed with 1-5 wt% of sublimed sulfur (Chempur, Poland, purity of 99.3%) and then the sample was stirred mechanically for about 5 minutes. Samples of the mixture were taken for the thermogravimetry experiments with the use of a 1500C derivatograph (MOM, Mateszalka, Hungary). The studies were performed at temperatures from the range 30 o -1000 o C, at a heating rate of 3 K/min. The rest of the mixture was placed in a special container and after reaching a thermal equilibrium the samples were transferred into the calorimeter containing sulfuric acid. Thermokinetics experiments were performed according to the procedure described in detail previously 28 . Because of the specifi c nature of the reaction of titanium raw materials with sulfuric acid, i.e. corrosive environment of the reaction, the possibility of thermal explosion and the emission of gases into the atmosphere, a non--isothermal and non-adiabatic calorimeter was used in the experiments. The capacity of the calorimeter was 0.6 dm 3 . The calorimeter was equipped with a heater, temperature sensor, feeder, and a safety valve 26 . The time constant and the heat transfer coeffi cient of the calorimeter were 257.5 1/min and 0.098 J/(K · s), respectively 26 . The study of the effect of sulfur addition on the reaction of ilmenite ores with sulfuric acid required the determination of the initial reaction conditions. Based on the previous experiments 27 the initial conditions of the studied reaction were taken as follows: the initial temperature of 80 o C and 90 o C for Norwegian and Australian ilmenites, respectively, the initial concentration of sulfuric acid 84% and the particle size distribution of the raw materials in the range of 88-92% particles addition of sulfur (1%) to the reaction of Norwegian and Australian ilmenite ores with sulfuric acid caused a signifi cant reduction in the maximum thermal power of the reaction. Moreover, the obtained thermokinetic lines show that after the addition of sulfur, the reaction proceeded more smoothly, and the time required to obtain the maximum thermal power was longer, especially for Norwegian ilmenite ores. Increasing the amount of sulfur in the reaction system to 3% and 5% did not signifi cantly affect the maximum thermal power or reaction time. Fig. 5 presents the results of studies of the effect of sulfur addition on the heat of reaction of ilmenite ores with sulfuric acid. For both materials, a signifi cant reduction in the heat of reaction (from 10% to 15%) is visible with the least amount of sulfur added (1%). Increasing the amount of sulfur added to 3% and 5% did not further reduce the heat of the reaction. The infl uence of sulfur addition on the thermal power of the reaction of Norwegian and Australian ilmenite ores with sulfuric acid is presented in Fig. 3 and Fig. 4, respectively.   As follows from the presented results, the maximum thermal power in the reaction of Norwegian ilmenite with sulfuric acid (Fig. 3) was about twice as high as that in the reaction of Australian ilmenite (Fig. 4). This difference can be explained by the different compositions of both materials. Australian ilmenite contains greater amounts of iron(III) oxides with smaller enthalpy of creation in comparison with iron(II) oxide 30 . A small A similar analysis was carried out for the kinetic parameter of the reaction, i.e. the apparent activation energy, which was assessed based on the kinetic model developed for this reaction in a liquid -solid heterogeneous system 31 . The following heat balance and the contracting core model equations 31 were used to estimate the apparent activation energy: where: W(t) -thermal power (W), Q -heat amount generated during the reaction, H r -heat of the reaction, α -conversion degree, C -heat capacity of the calorimetric system (J/K), G -coeffi cient of heat loss (J/K · s), S o -initial surface of the solid (m 2 ), k o -preexponential coeffi cient, E -apparent activation energy. Fig. 6 shows the dependence of the apparent activation energy on the amount of sulfur addition for the reaction of ilmenite ores with sulfuric acid. As in the previous case (Fig. 5), a substantial decrease in the apparent activation energy can be observed already with the addition of 1% sulfur to the reaction environment. When increasing sulfur addition to the reaction, further changes in the apparent activation energy are quite small. from the reaction products is also very important. The effect of sulfur addition on the content of iron at the second and third oxidation degree after leaching of the reaction products is illustrated in Fig. 8. The presented data show that the concentration of iron at the second degree of oxidation in the solution distinctly increased relative to that after the reaction without the addition of sulfur. For both raw materials the maximum concentration of Fe 2+ was observed after the smallest sulfur addition (1%). With a further increase in the amount of sulfur, the Fe 2+ concentration decreased in the case of Australian ilmenite.
An important parameter for evaluating the reaction thermokinetics is ΔT max /Δτ, which can also be used for safety assessment and reaction optimization 27 . This parameter is defi ned as the ratio of the temperature increase (from the initial to maximum temperature of the reaction) to the time in which this temperature increase occurred. A high value of this parameter indicates the probability of thermal explosion, whereas a low value may indicate inadequate conversion of the substrates. Optimal reaction conversion is achieved under conditions when the parameter ΔT max/ Δτ has high values, but in this case, there is a danger of thermal explosion. The effect of the amount of sulfur addition on the change in the thermokinetic parameter ΔT max/ Δτ is illustrated in Fig. 7. Similarly, as in the previous cases, there is a pronounced decrease in the value of this parameter already upon the addition of 1% sulfur, which could have a crucial impact on the level of the process safety. With an increasing amount of sulfur added, further changes in the value of ΔT max/ Δτ are insignifi cant.
The infl uence of the amount of sulfur added to the reaction environment on the level of leaching of elements Taking into account the results obtained, another series of experiments were conducted to examine the effect of changes in the initial concentration of sulfuric acid in reaction with ilmenite, with the addition of sulfur at the level of 1%. Fig. 9 and Fig. 10 show the effect of sulfuric acid concentration in the range from 82% to 90% on changes in the maximum thermal power for Norwegian and Australian ilmenite ores, respectively.
As indicated by the presented results ( Fig. 9 and Fig. 10), the addition of sulfur reduces the maximum thermal power of the reaction in the whole range of sulfuric acid concentrations. The estimated measurement error was from 1% up to 6%, for both ilmenites. A slightly different shape of the curves of the maximum heating power as a function of the concentration of sulfuric acid with and without the addition of sulfur can be also observed. The differences may indicate a signifi cant infl uence of the sulfur addition on the reaction mechanism. The observed decrease in the maximum of thermal power for Norwegian ilmenite was from 0.5-2.6 W/g to 0.2-1.7 W/g, and for Australian ilmenite from 0.4-0.79 W/g to 0.1-0.49 W/g.
The likely cause of the infl uence of sulfur addition on the reaction thermokinetics is the blocking of active centers of ilmenite ores by sulfur, which may change its oxidation degree in certain temperature ranges. Because heat energy is released during the reaction, part of this energy changes the level of sulfur oxidation, which may affect the level of iron oxidation from Fe 3+ to Fe 2+ . The confi rmation of this phenomenon is a higher concentration of Fe 2+ found in the solution after leaching, ilmenite. At sulfuric acid concentrations from 82% to 86%, the differences are slight and within the limit of the measurement error. Above 86%, an increase in the degree of TiO 2 leaching is visible in comparison to the reaction without the addition of sulfur. For Australian ilmenite, the situation is slightly different (Fig. 14). At sulfuric acid concentrations from 85% to 90%, the obtained results are at a similar level and are within the limit of the measurement error. It can be concluded that the addition of sulfur to the reaction environment does not reduce the degree of TiO 2 leaching, while in certain concentration ranges improvement of its leaching level is visible. compared to that in the solution after leaching from the products of the reaction carried out without the addition of sulfur. Moreover, despite the lower level of the thermal effect of the reaction and the value of the apparent activation energy, the degree of the substrates conversion is at the same level as without the addition of sulfur.
The infl uence of 1% sulfur addition on the change in the parameter ΔT max/ Δτ is presented in Fig. 11 and Fig. 12. For both ilmenites, the sulfur addition to the reaction vessel resulted in the reduction in the value of ΔT max /Δτ. In the reaction of sulfuric acid with Norwegian ilmenite (Fig. 11), the value of ΔT max /Δτ decreased from 1.5-3 K/min to 0.4-1.25 K/min, in the range of initial sulfuric acid concentrations of 82-88%. Only the value of ΔT max /Δτ for 90% sulfuric acid remained at the same level as in the reaction without the additive. Addition of sulfur to the reaction of Australian ilmenite (Fig. 12), resulted in a reduction by half of the parameter ΔT max /Δτ in all tested concentrations of sulfuric acid in comparison to the reaction without the additive. As mentioned earlier, a high value of ΔT max /Δτ indicates the probability of thermal explosion, while its low value may indicate incomplete conversion of the substrates.
In addition to the infl uence on the thermokinetics of the reaction, it is also important to investigate the effect of sulfur addition on the reaction effi ciency, in particular on the degree of TiO 2 leaching. Fig. 13 and Fig. 14 present a comparison of the levels of TiO 2 leaching in the reactions with and without sulfur addition, as a function of the sulfuric acid concentration. Fig. 13 shows the results obtained for the reaction with Norwegian

CONCLUSIONS
Based on the results of the conducted experiments, it was shown that a small 1% addition of elemental sulfur signifi cantly affects the thermokinetic parameters of ilmenites reaction with sulfuric acid. The addition of sulfur remarkably reduces the thermal power of the reaction and brings about a substantial decrease in the value of the kinetic parameter ΔT max /Δτ. At the same time, the addition of sulfur does not change the quality parameters of raw materials and does not adversely affect the conversion of the substrates. Moreover, the addition of sulfur does not negatively infl uence the degree of TiO 2 leaching; on the contrary, in a certain range of sulfuric acid concentrations, it improves this process. The observed infl uence of sulfur addition on thermokinetic parameters of the reaction of ilmenites with sulfuric acid is most probably related to the changes in the oxidation degree of sulfur and iron.
The presented results indicate that a small addition of sulfur to the reaction of ilmenites with sulfuric acid may signifi cantly reduce the risk of uncontrolled course of this hazard-type reaction, and thus increase the process safety of titanium dioxide pigment production.