Synthesis of Ti Powder from the Reduction of TiCl 4 with Metal Hydrides in the H 2 Atmosphere: Thermodynamic and Techno-Economic Analyses

: Titanium hydride (TiH 2 ) is one of the basic materials for titanium (Ti) powder metallurgy. A novel method was proposed to produce TiH 2 from the reduction of titanium tetrachloride (TiCl 4 ) with magnesium hydride (MgH 2 ) in the hydrogen (H 2 ) atmosphere. The primary approach of this process is to produce TiH 2 at a low-temperature range through an efﬁcient and energy-saving process for further titanium powder production. In this study, the thermodynamic assessment and technoeconomic analysis of the process were investigated. The results show that the formation of TiH 2 is feasible at low temperatures, and the molar ratio between TiCl 4 and metal hydride as a reductant material has a critical role in its formation. Moreover, it was found that the yield of TiH 2 is slightly higher when CaH 2 is used as a reductant agent. The calculated equilibrium composition diagrams show that when the molar ratio between TiCl 4 and metal hydrides is greater than the stoichiometric amount, the TiCl 3 phase also forms. With a further increase in this ratio to greater than 4, no TiH 2 was formed, and TiCl 3 was the dominant product. Furthermore, the technoeconomic study revealed that the highest return on investment was achieved for the production scale of 5 t/batch of Ti powder production, with a payback time of 2.54 years. The analysis shows that the application of metal hydrides for TiH 2 production from TiCl 4 is technically feasible and economically viable.


Introduction
In the past decade, the rapid growth in aerospace, power generation, and biomedical industries has resulted in significant attention on the industry of titanium (Ti) and its alloys. Titanium can be used in severe conditions such as high temperature and pressure and a corrosive environment. Moreover, it has high specific strength, toughness, and biocompatibility in the human body, making it a unique choice for medical implants [1,2]. Unfortunately, the growth in the application of titanium and its alloys (compared to similar metals such as aluminum and high-alloy steels) has been affected by its production costs [3,4].
Since the invention of the Kroll process for the production of titanium, it has been the primary practical production method in the industry [5]. However, this method has several drawbacks, such as long production time, elevated production temperature, and significant energy consumption [6]. Another major obstacle in producing low-cost titanium products is that the Kroll process final output is titanium sponge. It was reported that the total cost of traditional titanium components includes about 38% for the Ti sponge production via the traditional Kroll process, 15% related to Ti sponge refining, and the remaining 47% for the machining step, which produced almost 75% of the Ti sponge scrap [7]. Therefore, it was suggested to use powder metallurgy techniques such as additive manufacturing to produce near-net-shape products. For this purpose, the Ti sponge should be converted to Ti powder through other methods such as the hydride-dehydride (HDH) process, which is also an expensive process for the production of titanium powder [8,9].
For the mentioned reasons, many studies were conducted to introduce new techniques for Ti powder production. The two main precursors for Ti production are TiCl 4 and TiO 2 . It is noteworthy that TiCl 4 can be synthesized through the carbochlorination of TiO 2 , which is a fundamental concept of the TiRO method to produce Ti metal [10], or the ADMA process to synthesize TiH 2 powder [9]. Other potential methods are the Fray-Farthing-Chen Cambridge [11], Armstrong [12], Council for Scientific and Industrial Research Ti process [13], direct reduction of titanium slag process [14], and hydrogenassisted magnesiothermic reduction (HAMR) [15]. Moreover, a molten salt method using TiS 2 and TiO 2 via the Ono and Suzuki process led the way to cheaper Ti metal powder production [16][17][18]. It was reported that the recently reported HAMR process could reduce energy consumption to about 62% of the Kroll process [11]. It should be noted that Ti has a high chemical affinity with oxygen, and it is more difficult to remove the oxygen from TiO 2 compared to reducing chlorine from TiCl 4 . Furthermore, during the reduction of TiO 2 to Ti, the synthesized byproducts of oxides are challenging to remove due to their high melting point. On the other hand, the chloride species formed as byproducts during the reduction of TiCl 4 can be easily removed. Therefore, TiCl 4 is more favorable as a precursor for continuous production.
One of the primary mineral sources for the production of Ti is ilmenite, which is a titanium-iron oxide with the main chemical phase of FeTiO 3 . Ilmenite minerals represent the most available Ti resources, with about 90% of world consumption of Ti mineral [19]. It was reported that the earth's resources of ilmenite, rutile, and anatase are more than 2 billion tons. Traditionally, ilmenite is used to produce upgraded TiO 2 [19]. However, previous studies [20,21] have shown that high-quality TiO x C y N z could be produced from ilmenite through low-temperature chlorination. In this method, FeTiO 3 is reduced to TiO x C y N z under a mixture of H 2 -N 2 gas and is subsequently purified via aeration leaching (Becher process) with ammonium chloride (NH 4 Cl) lixiviant to remove metallic iron. The obtained TiO x C y N z is then chlorinated at 400 • C to produce TiCl 4 gas [22,23]. The author proposed directly reacting the obtained TiCl 4 gas with metal hydrides such as MgH 2 to produce TiH 2 powder at low-temperature ranges under the hydrogen (H 2 ) atmosphere [24][25][26]. The TiH 2 powder can then transform into Ti powder through the dehydrogenation process in a vacuum. The primary benefit of producing TiH 2 as an intermediate component is that the Ti powder can be produced with shallow oxygen content [27]. This phase is insoluble in water and diluted acid, which allows it to be separated from other products via acid leaching [28].
The feasibility of the proposed method was analyzed through chemical reactions involved in the production method and the obtained economic parameter. Technoeconomic analysis was also carried out using the SuperPro Designer software. The purpose of such analysis is to identify a production scale that has attractive economic performance. It is an important step in process development and commercialization. Data of laboratory-scale experimental may be used to perform process simulation study prior to pilot-scale testing. In the process industry, various technoeconomic works have been reported with the use of process simulation tools. For instance, Granjo et al. [29] made use of Aspen Plus and Monte Carlo simulation to evaluate different manufacturing methods of sodium methoxide. More recent works on technoeconomic analysis based on Aspen Plus simulation have been reported for supercritical CO 2 extraction of Eucalyptus globulus bark [30], chitin nanomaterials production [31], and anaerobic digestion of sugarcane vinasses for biogas production [32]. Moreover, there were authors who reported the use of SuperPro Designer for technoeconomic analysis of various biochemical, pharmaceutical, and food processes. Some of the early works were reported by Petrides and coworkers [33] for pharmaceutical processes. Later works on SuperPro Designer were reported for debottlenecking studies, for pharmaceuticals [34,35], and for fine chemicals [36]. More recently, SuperPro Designer was used for technoeconomic analysis for various manufacturing processes of chemicals, biochemicals, and food. These include the production of cocoa butter [37], monoclonal antibody [38], biodiesel [39], nanomaterials [40], catechin [41], biomass pyrolysis products [42], and thorium extraction from monazite [43]. It is also worth noting that Monte Carlo simulations were also reported for use in some of the above works where SuperPro Designer was utilized (e.g., [38,40]). This allows for the identification of important parameters that affect the economic characteristics of the process.
The current study attempts to address the thermodynamic, operational, and economic evaluation of the TiCl 4 reduction process with MgH 2 under H 2 atmosphere for the production of TiH 2 , which could be further dehydrogenated to Ti powder. The thermodynamic assessment was used to study the effect of temperature, partial pressure, and the molar ratio of the TiCl 4 to reductant on the formation efficiency of TiH 2 . Economic analysis was investigated for several production scales in order to identify critical economic parameters such as production revenue and gross margin. Moreover, sensitivity analyses were performed to evaluate the effect of different operating parameters on the economic performances of the selected production scale.

Materials and Methods
The processing cycle in the present study, developed for the production of Ti powder from ilmenite, is shown in Figure 1. In the first step, the ilmenite is reduced under H 2 -N 2 atmosphere to convert iron oxides present in the ilmenite to metallic iron. In the subsequent step, metallic iron is separated from the nitrided ilmenite through aeration leaching (Becher process). Following the filtration process, the leftover powder is TiO x C y N z which can be employed in the chlorination process. The product of the chlorination process is TiCl 4 gas. The complete explanation of the reduction and chlorination process of ilmenite is discussed elsewhere [20][21][22]. The produced TiCl 4 liquid is converted to gas above 150 • C and reacted with MgH 2 to synthesize TiH 2 powder via Equation (1). The TiH 2 can be separated from reaction byproducts through acid leaching and washing with distilled water to remove the MgCl 2 . Next, the filtered TiH 2 powder is dried for 12 h in an oven, and it can be converted to Ti powder via heating under Ar gas for dehydrogenation. Details of the latter process are listed elsewhere [44,45].

Thermodynamic Calculation
Previous studies [46,47] have discussed the thermodynamic principles of iron removal from nitride ilmenite. This study investigated the thermodynamic equilibrium of TiH 2 synthesis from TiCl 4 using MgH 2 as a reductant. This calculation was helpful for studying the possible species formation during the reaction. The HSC Chemistry v6.0.0 software (Outokumpu Research Oy, Tornio, Finland) was used for thermodynamic analysis and calculations. Thermochemical calculations are valuable tools for developing new chemical processes and mechanisms. This software provides powerful calculation methods to investigate different variables' effects on the chemical system at equilibrium [48].

Synthesis Process
In previous studies [21,22], the preparation of TiCl 4 gas from ilmenite was elaborated. In the current study, the TiH 2 was synthesized from TiCl 4 gas reacted with the MgH 2 powder at 500 • C for 12 h at the lab scale. Our previous attempts to synthesize TiH 2 at a temperature between 250 and 500 • C indicated that higher temperatures positively impacted the formation of TiH 2 [24][25][26][27][28]. The reduction products were then washed with water and diluted HCl to separate TiH 2 powder from the reaction byproducts. The X-ray diffraction (XRD) technique was employed for the study of phase evolution after the reduction process using Cu-Kα radiation (λ = 1.5404 Å, Bruker D8-advance, Billerica, MA, USA). The morphology and surface elemental analysis of the powder in each experimental step was studied with a field emission scanning electron microscopy equipped with electron dispersive X-ray spectroscopy (FESEM/EDX, Leo Supra 35VP, Oberkochen, Baden-Württemberg, Germany).

Economic Evaluation
In order to investigate the commercialization potential of the proposed method for the production of Ti metal powder, the technoeconomic assessment (TEA) was conducted using the SuperPro Designer software v8.5 [49]. The TEA calculates the production expenses, technology, and scaling-up issues of the product. The effect of individual economic parameters, i.e., costs of raw material, equipment costs, and energy consumption, on the capital cost investment of a pilot plant was evaluated. The economic performance was investigated for the three production scales, i.e., 1000 kg (1 t/batch), 5000 kg (5 t/batch), and 10,000 kg (10 t/batch) per batch of the final product (Ti powder).

Thermodynamic Assessment
The standard change in Gibbs free energy of formation (∆G • f ) of the chemical species that were used in this study is listed in Table 1. The thermodynamic data were extracted from Stull and Prophet (1971) [50] for the temperature range of 0-700 • C, except for MgH 2 , which was extracted from Brain (1997) [51] for the thermodynamic data up to 700 • C. The thermodynamics study on converting ilmenite to nitride ilmenite and subsequently TiCl 4 was published elsewhere [52,53]. The chemical potential diagrams (stability diagrams) of the M-Cl-H (M: Ti and Mg) systems at different partial pressures in Table 1 are illustrated in Figures 2 and 3. The phase stability diagrams demonstrated the stability boundaries of each phase in a ternary system as a function of temperature or in isothermal conditions. The stability diagrams were calculated based on the Gibbs free energy minimization. The partial pressure of gas species was calculated from the equilibrium constant and the activity of the gas species.
In Figures 2 and 3, the effect of temperature, the partial pressure of H 2 (P H 2 ), and partial pressure of chlorine (P Cl 2 ) on the stability of associated phases are illustrated. Figure 2 illustrates the Ti-Cl-H diagrams with the change in the partial pressure of H 2 gas. The values of the P Cl 2 were changed, starting from 1 atm until the appearance of TiH 2 phase at P H 2 = 1 atm below 500 • C. It was observed that for P Cl 2 ≤ 1 × 10 −30 atm, the TiH 2 phase was stable within the specified temperature range. It can be seen from Figure 2a that in the Ti-Cl-H system, the dominant solid phase at T ≤ 350 • C and P H 2 = 1 atm was TiCl 3 , and at the higher temperature range, the TiH 2 phase was more stable. Moreover, Figure 2b shows that with additional decreasing in the P Cl 2 to 10 −35 atm, there was no presence of TiCl 4 , and the TiH 2 stability area increased to T ≥ 250 • C. According to Figure 2a, the equilibrium interface between TiH 2 and TiCl 3 indicated that the stability area of TiH 2 started at a temperature above about 150 • C and P H 2 = 20 atm, and with an increase in temperature, it was expanded to P H 2 of 1 atm at a temperature around 300 • C. This particular distinction between the two diagrams in Figure 2a,b represents the importance of the effect of P H 2 and P Cl 2 on the stability diagrams. Hence, in the presence of H 2 atmosphere, it should be possible to reduce the TiCl 4 to TiH 2 .   Table 1. Standard Gibbs free energy of formation for the compounds in the investigated system from 0 to 700 • C. In order to further highlight the effect of P H 2 on the phase stability, the Ti-Cl-H predominance diagram was plotted for a constant amount of P H 2 at 1 atm in Figure 2a and 1 × 10 −30 atm in Figure 2b. As is illustrated in Figure 2a to obtain the TiH 2 predominance area at P H 2 = 1 atm, the P Cl 2 in the system must be as low as 1×10 −30 atm for T = 350 • C and 1 × 10 −35 atm for T = 250 • C. However, it can be observed from Figure 2b that at constant P H 2 of 1 × 10 −30 atm, there was no presence of TiH 2 . This is the main reason that there was a need to run the reduction process in the presence of H 2 gas instead of an inert atmosphere. Basically, in P H 2 ≥ 1 atm and very low amount of P Cl 2 , TiH 2 was a predominant phase. Previous work by Udayakumar et al. [54,55] in producing TiH 2 in an inert atmosphere showed a low yield and validating this thermodynamics analysis.

Component Standard Gibbs Energy of Formation
It should be emphasized that depending on the partial pressure of Cl 2 , the system may enter the predominance area of TiCl 3 and possibly TiCl 2 . This particular scenario indicates a possible thermochemical mechanism that TiH 2 may produce from reactions between TiCl 3 or TiCl 2 and MgH 2 rather than the reaction via Equation (1) at a fixed temperature and certain P H 2 and P Cl 2 . In this regard, the possible reactions in the system could be followed as in Equations (2)-(5) for the TiCl 4 -MgH 2 reaction system.
The predominance diagram for the Mg-Cl-H system at P Cl 2 = 1 × 10 −30 atm is shown in Figure 3. The interface between MgH 2 and MgCl 2 indicated the stability of metal chlorides at P H 2 = 1 atm and almost the whole temperature range, except for Mg, which is more stable at a temperature above 600 • C, according to Figure 3. From the latter figure, MgH 2 was a stable phase at a temperature above 350 • C and higher P H 2 . As a result, it is possible to conclude that reducing TiCl 4 with metal hydrides is possible, and the stable phases were TiH 2 and MgCl 2, depending on the partial pressure of chlorine gas. On a side note, this thermodynamics approach does not consider the possibility of simultaneous formation of different species in the same system, as only one species, for example, TiCl 3 or TiH 2 , is described as a feasible component at a particular condition. Given this, the Gibbs free-energy minimization method was carried out to evaluate the formation of various phases simultaneously.

Phase Composition Diagrams
The objective of this section is to present the equilibrium composition diagrams as in Table 1. Therefore, calculations were carried out for stoichiometric and nonstoichiometric amounts of reactants for the temperature range of 0-700 • C as given in Equation (1). The phase determination of the system at equilibrium condition was calculated with the help of the equilibrium module of HSC Chemistry v6.0.0 (Tampere, Southern Finland, Finland), which is based on the minimization of the Gibbs free energy [56]. The relevant elements (Ti, Cl, Mg, and H) with an ideal behavior were used to determine the stable phases of the system under the investigated conditions. It should be noted that a constant amount of H 2 equaling 2 mol was considered in all the calculations for the effect of excess H 2 in the system since the reduction experiment suggested proceeding under H 2 atmosphere. The molar ratio of TiCl 4 :MgH 2 was 1:2 for stochiometric conditions according to Equation (1). In this condition, the final equilibrium phases after reactions were TiH 2 , MgCl 2 , and H 2 in all temperature range according to Equation (1).
Furthermore, the nonstochiometric conditions with increasing the molar ratio of TiCl 4 :MgH 2 in Equation (1) were investigated with an increase in the latter molar ratio to 2:2 and 3:2, as shown in Figure 4. The changes in the molar ratio of the reactants affected the final products available in the reaction. From Figure 4, TiH 2 was formed at its highest amount of 0.6 mol below 100 • C. As the temperature raised to 700 • C, the TiH 2 molar amount gradually dropped to 0.1 mol when the TiCl 4 molar amount increased to 2 mol, as shown in Figure 4a. However, Figure 4b illustrates that with an additional increase in TiCl 4 to 3 mol, TiH 2 was not present above 100 • C. Furthermore, by increasing the TiCl 4 :MgH 2 molar ratio to 3:2, the TiCl 3 phase formed, in addition to TiH 2 . At a TiCl 4 amount of 2 mol, TiCl 3 decreased continuously from about 1.2 to 0.1 mol, with temperatures shown in Figure 4a. However, by increasing the TiCl 4 :MgH 2 molar ratio to 3:2, the change in the TiCl 3 amount was different, as shown in Figure 4b. It can be observed that the TiCl 3 amount dropped to about 2 mol at 100 • C, and there was no change up to 500 • C. Then, there was another decrease in the TiCl 3 amount from 2 to 1.3 mol above 550 • C. In addition, the amount of MgCl 2 as a final product in the system was constant in all the diagrams. According to Figure 4, it was clear that with increasing temperature, the molar amount of H 2 gas gradually increased up to about 4 mol at a TiCl 4 molar amount of 2 mol. It was also rapidly increased to about 4 mol when the TiCl 4 molar amount was 3 mol, and it was constant above 150 • C. Moreover, it shows the formation of H 2 gas in the system with an increase in a molar ratio of TiCl 4 :MgH 2 .
Another essential feature of diagrams in Figure 4 is the formation of the TiCl 2 phase as a product with the reverse chemical behavior of TiCl 3 and TiH 2 . While the molar amount of TiCl 3 and TiH 2 decreased, the molar amount of TiCl 2 began to increase. Since TiH 2 , TiCl 3 , and TiCl 2 were behaving comparably, there should be a reaction to consuming the first two phases to convert to a later one with increasing temperature. Moreover, at TiCl 4 molar amount of 3 and temperature range above 600 • C, interesting changes occurred as new gas phases, namely TiCl 3 (g), TiCl 4 (g), and Ti 2 Cl 6 (g), were formed, and the amount of TiCl 2 started to increase. The main reason for these behaviors is the possibility of the reaction between excess TiCl 4 and the produced TiH 2 in the system as Equation (6), which has negative ∆G • below 700 • C. In addition, the produced TiCl 3 can be reacted with TiH 2 to produce TiCl 2 as Equation (7). Moreover, the formation of TiCl 3 (g), TiCl 4 (g), and Ti 2 Cl 6 (g) are due to the transformation of TiCl 3 to the gas phase above 500 • C. Up to 500 • C, the TiCl 3 is in the solid state, and with an increase in temperature, it is completely decomposed to the gaseous state. Subsequently, the amount of TiCl 3 decreased above 500 • C. As a result, TiCl 2 and TiCl 4 (g) can be produced according to Equation (8). In addition, there is the possibility that TiCl 3 (g) transforms to Ti 2 Cl 6 (g) phase, whereas this reaction has negative ∆G • according to Equation (9).
To put it briefly, the formation of TiH 2 is possible through a reaction between MgH 2 or CaH 2 and TiCl 4 ; however, the ratio between TiCl 4 and metal hydrides was critically important to have TiH 2 following the completion of the reactions in the system.
The equilibrium molar amount and molar fraction of TiH 2 (X • TiH 2 ) as a function of the molar fraction of TiCl 4 (X • TiCl 4 ) and temperature for the TiCl 4 -MgH 2 reaction system are provided in Figure 5. X • P was defined as Equation (10).
where n • i is an output mole of component i in the system (i.e., TiCl 4 or TiH 2 ), and n • p is the molar amount of all the products after the reduction process based on the equilibrium composition calculation data from HSC software for Equation (2)- (9). For the sake of simplification, the amount of MgH 2 was set to 2 mol, similar to previous calculations for equilibrium compositions. It was evident from Figure 5a that TiH 2 formation started at X between almost 0.3 and 0.4 in all temperature range. With more TiCl 4 addition at a constant amount of MgH 2 , the formation of TiH 2 was impeded, while TiCl 3 or TiCl 2 phases began to produce, as discussed in Figure 4. Needless to say, all the calculations are based on thermodynamic data and minimization for the ∆G • in the system. Since TiCl 4 can be reduced to TiH 2 , TiCl 3 , and TiCl 2 via Equations (6)-(9) solely based on the molar ratio between TiCl 4 and MgH 2 , it is of interest to know the optimum molar fraction of TiCl 4 to MgH 2 , which can reach the highest yield of TiH 2 . If TiH 2 production efficiency was defined as the molar amount of produced TiH 2 as in Equations (2)-(9) divided by the stoichiometric molar amount of TiH 2 from Equation (1), then Figure 6 could be plotted for the TiCl 4 -MgH 2 reaction system. It can be observed from Figure 6 that the TiH 2 efficiency reached the highest level of 100% for the amount of X • TiCl 4 below 0.3 for all temperature ranges. In this condition, no TiCl 3 or TiCl 2 was produced, and TiCl 4 was entirely transformed to TiH 2 . In addition, an excess amount of MgH 2 could be retained in the system. Moreover, the calculated data showed that change in the temperature had no significant effect on the TiH 2 formation efficiency, except a slight increase in TiH 2 efficiency at a temperature below 250 • C and fraction between 0.4 < X • TiCl 4 < 0.5. It should be noted that these calculations were based on the equilibrium assumption, and the situation in the real system could be different.

Phase Analysis
The synthesis of TiH 2 was performed using MgH 2 powder as a reductant according to the process of Figure 1. The phase analysis of the final reduction products was employed to gauge the mechanism of the reaction according to Equations (1)-(9). Figure 7 represents the XRD diffraction patterns for raw ilmenite, reduced ilmenite before and after the Becher process, and the reduced powder with MgH 2 at 500 • C for 12 h. The ICSD reference number of the studied phases are 98-003-5383 (FeTiO 3 ), 98-001-7721 (Fe 2 TiO 5 ), 98-008-9933 (TiO x C y N z ), 98-008-7922 (Fe), 98-008-8801 (MgH 2 ), 98-008-9526 (Mg), 98-001-1144 (TiCl 3 ), 98-009-6474 (MgCl 2 ·(H 2 O) 4 ), 98-000-3817 (TiO 2 ), and 98-006-5530 (TiH 2 ). It is evident from Figure 7 that the raw ilmenite mainly consists of FeTiO 3 and pseudobrookite (Fe 2 TiO 5 ). All the ICSD# for phases detected were listed in Figure 7. Moreover, low-intensity peaks related to titanium dioxide (TiO 2 ) were also observed. It is shown in Equation (11) that ilmenite can be converted to the TiN through the carbothermal reduction and nitridation (CTRN) process. Previous studies by Rezan et al. [57,58] reported that the actual phase from the carbothermal reduction and nitridation of ilmenite was TiO x C y N z rather than TiN. On the other hand, the Fe 2 TiO 5 phase was transformed to FeTiO 3 via the reaction of Equation (12).   Figure 7 shows that the reduced powder after nitridation was a combination of the TiO x C y N z phase and metallic Fe as expected from Equation (11). This powder was later leached according to the Becher process [20], and the metallic Fe was removed entirely, as can be seen from the XRD data in Figure 7. The Fe particles can react with dissolved oxygen as Equation (13) Figure 7 it can be observed that in addition to the TiH 2 phase, other phases such as TiCl 3 , MgH 2 , and Mg were detected. Moreover, MgCl 2 (H 2 O) 4 was formed as a major byproduct of the reaction system. This phase was produced due to moisture adsorption on MgCl 2 . The formation of TiCl 3 can be explained through Equation (14). It was reported that in the Ti-Cl-H system, TiCl 4 has a high tendency to reduce to TiCl 3 in the presence of H 2 gas [59]. Simultaneously, TiCl 3 can react with MgH 2 powder to produce TiH 2 by Equation (15). Therefore, the possible chemical pathway for the production of TiH 2 from TiCl 4 reaction with MgH 2 could be from Equations (14) and (15). It should be noted that the presence of Mg in the XRD data results from the partial dehydrogenation of MgH 2 powder. Overall, the XRD patterns confirmed the formation of TiH 2 from the proposed method. Figure 8a-c illustrate the as-received ilmenite morphology together with the reduced ilmenite before and after the leaching process. Besides this, the morphology of the TiH 2 powder after washing the reduction products with distilled water and diluted HCl is illustrated in Figure 8d. Figure 8a-c indicates the formation of spherical-shaped particles, which could be related to the formation of metallic iron, and it can be easily removed from the system through the Becher process. The EDX analysis data of Figure 8c clearly shows the high percentage of Fe content from nitrided ilmenite with a spherical Fe particle forming due to the interaction of carbon saturated molten iron. Figure 8d demonstrates the morphology of the final powder after the reaction between TiCl 4 gas and MgH 2 powder. It can be observed that irregularly shaped particles were formed during the reduction process. The EDX analysis data of these particles demonstrated the presence of about 78 wt% Ti with the balance oxygen. The presence of oxygen in the EDX analysis data should be the result of the oxidation of TiH 2 surface, which was previously reported by   [60] that an oxide layer could be formed on the TiH 2 surface. The phase analysis and morphology study of the final powders indicated the formation of the TiH 2 powder with irregularly shaped particles from the reduction of TiCl 4 gas with MgH 2 powder.

Process Simulation and Economics Evaluation
In the previous section, the experimental results for the production of TiH 2 powder via thermodynamic assessment (Section 3.1) and phase and morphology characterization (Section 3.2) are presented. The results show that TiH 2 powder production from the proposed method was feasible technically. However, to examine the cost-effectiveness of the process, it is beneficial to investigate the economic behavior of the production process. Thus, in this section, the TEA was performed for the production of TiH 2 powder from ilmenite and subsequently Ti metal powder from TiH 2 as a final product. It is necessary to estimate the total production costs of the product before any financial investment is made for a pilot production [61].  (1) is EDX analysis micrograph for raw ilmenite, point (2) is EDX analysis micrograph for spherical particles in reduced ilmenite, and point (3) is EDX micrograph for obtained TiH 2 powder after water washing and acid leaching). Figure 9 shows the simulation flowsheet developed using SuperPro Designer [49] for Ti powder production from ilmenite ore. As shown, the ilmenite was grounded below 75 µm in procedure P-1 (Grinder GR-101), and it was mixed with coal in the procedure P-1/2. Next, the mixture was subjected to the CTRN process in reactor P-2 (vessel R-101) at 1250 • C for 5 h in the presence of H 2 and N 2 gases [62]. In the proposed method, plastic wastes such as polyethylene terephthalate (PET) can be used instead of coal since they are valuable carbon resources. It has been reported that PET stands as 8 wt% of the global solid waste [23]. In our previous studies [22,23], it was shown that the ilmenite could be reduced with PET via the CTRN process. Compared to the conventional method for the reduction of ilmenite, this is a sustainable method for recycling waste plastics from economic and environmental aspects. As it was shown in Equation (11), the reduction products were the mixture of TiN and Fe particles. After completing the reduction step, Fe particles were separated from TiN through aeration leaching using the Becher process (procedure P-3) [63]. For this purpose, air and ammonium chloride (NH 4 Cl 3.2 M) were injected into the leaching vessel AFR-101 for aqueous iron oxidation. The Fe particles can react with dissolved oxygen following the reaction stoichiometry in Equation (13) to form an iron oxide hydrate (Fe 2 O 3 ·xH 2 O). Later, the mixture was sent for filtration in the filter unit MF-101 (procedure P-4). The filtrate from the latter contains dissolved NH 4 Cl and aqueous Fe 2 O 3 ·xH 2 O, which can be separated from each other through filtration (procedure P-4/MF-102). Although the leaching process (P-3) was performed at 90 • C for 7 h to ensure complete oxidation of Fe, there is the possibility that a small portion of Fe particles remains in the TiN filter cake [63]. Therefore, the filter cake was washed with hydrochloric acid (HCl; 1 M) in procedure P-5 to remove all remaining Fe particles. The latter can be transformed into iron chloride (FeCl 2 ), following Equation (16). The FeCl 2 residue is collected in tank P-5/1 (vessel V101), which can be sold as a byproduct for the sewage treatment industry or recycled back to Fe 2 O 3 for the ironmaking industries [64,65].
A batch vessel reactor (P-6/R-102) is used to model the chlorination of TiN powder. The chlorination process was performed at 400 • C for 5 h, in the presence of Cl 2 gas, which converts TiN to TiCl 4 gas following Equation (17) [66]. It is worth noting that the product N 2 gas can be collected and reused in procedure P-2. The TiCl 4 gas can be reduced with MgH 2 to form a solid mixture of TiH 2 and MgCl 2 , following the reaction stoichiometry in Equation (1). This process was carried out in the reduction reactor P-7 (R-103), in which the MgH 2 powder was used as a reductant agent in the reduction of TiCl 4 to TiH 2 . This mixture is then subjected to water washing (P-8/WSH-101) for the removal of MgCl 2 , which is water soluble. The solution residue from water washing is then sent to the electrolysis (procedure P-8/1) for Mg metal production, which can later be reacted with H 2 gas to produce MgH 2 for the reduction process. On the other hand, the TiH 2 powder is used as a raw material for a dehydrogenation reactor (P-9) for Ti powder production [23,24]. It should be noted that the H 2 gas emission from procedures P-7 and P-9 can be collected and stored in vessel P-10/V-102 for further application in the ilmenite reduction process (P-2).
The main benefit of the current method compared to the traditional Kroll process is that the final product is TiH 2 or Ti metal powders rather than Ti sponge. Therefore, instead of machining the Ti sponge to produce final products that generate a large amount of scrap, the TiH 2 or Ti metal powder can be directly used in powder metallurgy techniques to generate the final products with less scrap and reduce energy consumption related to the machining step, resulting in an enhanced environmental impact. Moreover, the reaction temperature in the reduction reactor P-7 (R-103) is 500 • C, which is much lower than 900-950 • C in similar processes, where the TiCl 4 reduction is performed with molten Mg which requires a high temperature to melt Mg metal [9]. However, the current study proposes to employ MgH 2 powder at a lower temperature with less energy consumption, which is more environmentally friendly. Another major difference is that the reaction byproducts from reactor P-7 (R-103) separated via hydrometallurgical technique in process P-8 instead of using vacuum distillation at elevated temperature of 1000-1050 • C in similar processes [67].
The production process simulation was performed for the annual operation time of 7920 h/year, plant construction duration of 6 months and project lifetime of 15 years. The processing time and unit for the respective procedures of the model are summarized in Tables A1 and A2 in Appendix A. In addition, Table 2 summarizes the overall process parameters for different production scales. It can be observed that by increasing the production scale, the recipe batch time and recipe cycle time are increased, and the number of annual batches dropped. However, due to larger process throughput, the total amount of Ti product increases for the larger production capacity (e.g., 10 t/batch). Table 2. Overall process parameters for process simulation for different production scales of Ti powder.

Raw Material and Installed Equipment Costs
The required amount of raw materials was estimated according to the stoichiometric amounts given in Equation (1) and Equations (11)- (14). Table 3 indicates the amount of required materials per batch with their unit price for economic evaluation for different production scales. It should be noted that both HCl and NH 4 Cl streams contain water in the mixture (see its mass % in Table 3). Furthermore, the required equipment details were summarized in Table A3 in Appendix A. Moreover, Tables A4-A6 in Appendix B list the equipment unit price, annual required materials price, and utility consumption, respectively, for different production scales of the Ti powder plant. The annual materials cost can be calculated from each material unit price and the annual amount per batch by the number of batches per year in Table 2. Based on Tables A4-A6, it can be concluded that raw materials have the highest impact on production costs, regardless of the production scale. It should be noted that the SuperPro Designer software v8.5 (Scotch Plains, New Jersey, USA) has a built-in model for the calculation of equipment costs [49], based on the costs model in Peters et al. [68]. Table 3. The required amount of raw material with their unit costs for different production scale.  Table 4 summarizes the purchase cost of major equipment according to the studied production scales of Ti powder, calculated using the built-in model [49]. It should be noted that the equipment capacity is directly related to the production throughput. As can be seen, the highest equipment cost was associated with the TiCl 4 reduction furnace (R-103) and MgCl 2 electrolysis cell (R-104) for the Ti production scale of 1 t/batch, and ilmenite reduction reactor (R-101) for Ti production scale of 5 t/batch and 10 t/batch. Table 4. The total equipment cost per batch for different production scales of Ti powder in USD.

Economic Analysis
Based on the economic model of Peters et al. [68], the built-in model of SuperPro Designer v8.5 (Scotch Plains, NJ, USA) [49] determines the production costs as a percentage of the capital cost and operating costs, as summarized in Table A7 in Appendix C. It is evident from Table A7 that the equipment purchase cost is the highest factor in the direct plant costs, while the construction was a significant factor in indirect plant cost. Table 5 summarizes the capital fixed costs for Ti production scales of 1, 5, and 10 t/batch. It is noteworthy that the total equipment purchase cost, process piping, installation, insulation, instrumentation, buildings, electrical, yard improvement, and auxiliary facilities are considered direct plant costs. On the other hand, engineering and construction were categorized as indirect plant costs. Other parameters such as start-up and validation, insurance, maintenance, a local tax, and the factory expenses were included in calculating the capital investment in the economic modeling by the SuperPro Designer [49]. It is evident from Table 5 that direct and indirect plant expenses increase with an increase in production scale. The economic evaluation results for the investigated production scales were given in Table 6. It was considered that the Ti metal powder price is 250 USD/kg. The two main economic factors, i.e., return of investment (ROI) and internal rate of return (IRR) values, were compared for all the production scales as reported in Table 6. In general, a positive ROI value indicates that the net returns are positive since the total returns are higher than the process costs. Moreover, a project is economically feasible with a higher IRR than the minimum acceptable returns for investors [61,68]. The results in Table 6 demonstrated that maximum ROI and IRR were 39.36% and 63.20%, respectively, which was obtained for production scale of 5 t/batch. However, the ROI and IRR decreased to 35.18% and 53.83% for 10 t/batch production scale, respectively. This is mainly due to a significant increase in the total capital investment for 10 t/batch, with a marginal increase in its revenue. This leads to lower ROI and IRR as compared to the 5 t/batch production scale. Overall, the unit production cost for Ti powder ranges between 224.19 and 246.58 USD/kg Ti from this study. The Monte Carlo method was employed to quantify the uncertainty [69,70] for the sensitivity analysis, using Crystal Ball v11.1.2.4 add-in function in Microsoft Excel. This simulation determines which operational factors have the most significant impacts on the economic efficiency of the process. According to Table 6, the ROI (39.36%) and the unit production cost of the Ti powder ($226.69/kg) for the Ti powder production scale of 5 t/batch were investigated for the sensitivity analysis. The distribution of model variables was summarized in Table A8 in Appendix D. In general, triangular distribution was applied for the time, and normal distribution with 10% standard deviation was used for price fluctuations. The probability distribution of ROI and Ti unit production costs are illustrated in Figure 10. It is evident from Figure 10a that the change in ROI from 10 to 70% falls within the certainty of about 76.42%. Moreover, it was observed from Figure 10b that the change in Ti unit production cost amounts from $205 to $250 correlated with a certainty of occurrence of about 79.37%. The sensitivity analysis diagrams generated by the Crystal Ball ( Figure 11) reveals that the MgH 2 purchase cost was the most significant input variable on the unit production costs and ROI, which is followed by the ilmenite purchase price. Overall, the MgH 2 price has a negative impact on the economic analysis. Increasing the MgH 2 price results in lower ROI and increased unit production cost. Figure 11. Contribution of uncertain parameters to the variance of (a) ROI and (b) unit production cost.

Conclusions
Thermodynamic assessment of TiH 2 synthesized from TiCl 4 was investigated between MgH 2 with TiCl 4 in the presence of H 2 atmosphere at a temperature range from 0 to 700 • C. The thermodynamic assessment revealed that the formation of TiH 2 was feasible in the TiCl 4 -MgH 2 reaction system. However, the molar fraction of TiCl 4 has a critical role in the reduction process. With an increase in the molar fraction of TiCl 4 above 0.4, the TiH 2 formation was impeded, and the TiCl 3 phase started to produce significantly. According to the various amounts of TiCl 4 :MgH 2 molar ratio, different products were obtained. Experimental data confirmed the thermodynamic assessment on TiH 2 formation from TiCl 4 :MgH 2 were also extracted, and findings, and it was indicated that at the reaction temperature of 500 • C for 12 h, TiH 2 powder was formed. Moreover, the TEA of Ti metal powder production was conducted for 1, 5, and 10 t/batch production scales. The modeling results indicated that the highest ROI of 39.36% with 2.54 years of payback time was achieved at 5 t/batch production scale. Moreover, sensitivity analysis showed that the MgH 2 and ilmenite purchase prices had the highest negative impact on the ROI and unit production costs. The change in the raw materials by 10% resulted in the certainty of 76.42% for ROI changing between 10% and 70%. Similarly, the certainty for the unit product cost was achieved at about 79.37% for fluctuating between 205 and 250 $/kg. It has shown that the production of Ti powder via the TiH 2 route from ilmenite at the low-temperature range was economically feasible.

Conflicts of Interest:
The authors declare no conflict of interest.

Process Description
Reduction reactor at P-2 heating for 3 h at 1250 • C, extent of reaction 100% Aeration leaching at P-3 leaching at 90 • C for 7 h, extent of leaching 95% Microfiltration at P-4 filtration for 2 h with recovery yield of 90% Microfiltration at P-4/1 filtration for 4 h with recovery yield of 90% Acid washing in P-5 washing for 30 min and separation for 1 h with 100% recovery at room temperature Chlorination reaction at P-6 heating for 5 h at 350 • C, extent of chlorination 95% Reduction reaction in P-7 heating for 12 h at 500 • C, extent of reaction 95% Acid washing and separation in P-8 washing for 1 h with 100% recovery after filtration Electrolysis in P-8/1 reaction duration for 1 h at room temperature, extent of separation 100% Dehydrogenation in P-9 heating for 3 h at 700 • C, extent of reaction 100%