Modeling and Experimental Study of Ore-Carbon Briquette Reduction under CO–CO2 Atmosphere (2)

Keywords:

ore-carbon briquette,CO–CO2 atmosphere, reduction

Source: internal company

4. Solution Method

The FLUENT CFD package (v6.3, Fluent Inc., Lebanon, NH, USA) was used for conducting the numerical simulations. Equation (7) was spatially and temporally discretized using a fully implicit fifirst-order upwind scheme. The time step was 0.01 s, the under-relaxation factor was 0.1, and the convergence criterion used was 1.0 × 10–5. An explicit time integration method was adopted for solid-phase equations.

5. Results and Discussion

5.1 Determination of Model Parameters and Reaction Rates of Invovled Reactions

Before performing simulations, some parameters and rate expressions of the involved reactions must be determined. Lu and Sun reported that variation of the briquette porosity varied from Metals 2018, 8, 205 6 of 13 0.40 to 0.68 in the reduction process of the ore-carbon briquette, and the porosity of the briquette was thus assumed to be α = 0.50 in the present study.

Gaseous reduction of iron ore particles has been extensively studied in its thermodynamics and kinetics. Generally, shrink unreacted core model is consistent with the step-wise of iron oxides. As the hematite particles are very small, the reactions that are given by Equations (1)–(3) were considered to proceed independently. Their reduction rates were thus described using a one-interface unreacted shrinking core model, as Equation (19).

where K1 = exp(-1.445-6038/T) ，K1=exp(7.255+3720/T), and De,1= ∞ for R1; k2 = 1.70 exp(2.515 − 4811/T), K2=exp(5.289−4711/T) and De,2 =exp (−1.835− 7180/T)/Pg for R2; and, k3 = exp(0.805−7385/T), K3 = exp(−2.946+2744.63/T), andDe,3 = exp (0.485−8770/T)/Pg for R3 [35]. Defifinition of f i in Equation (19) is given in [32]. The internal gas diffusion resistance of R1 was not considered because the transformation of Fe2O3 to Fe3O4 (Equation (1)) proceeds very quickly in the initial stage of the briquette reduction.

The reaction rate of the reaction given in Equation (4) is Equation (20).

where f4 = 1.0 − ρC/ρC,0 and k4 = 1.8 × 103 exp(−139000/RT).

The rate of the reaction given in Equation (5) is Equation (21) [37].

where k5 = 0.011 exp(−42611/RT), and K5 = 1/K3.

In estimating NP in Equations (19)–(21), the true density of the hematite particles was assumed to be 5000 kg/m3 . γ in Equations (19)–(21) is a coeffificient for adjusting the specifific area of ore particles due to their irregular geometry, and it was determined to be 0.4 by a trial-and-error method.

5.2 Briquette Mass Change

Mass change in the briquette during reduction was caused by several reactions; the reactions given by Equations (1)–(4) led to a decrease in mass, whereas the reaction given by Equation (5) led to an increase in mass. Measured and model-predicted mass-loss fraction curves under different temperatures were compared, and the results are shown in Figure 3. In simulations, the briquette mass-loss fraction at time t was calculated using Equation (22).

The model predictions closely match the experimental measurements at 1273 K and 1373 K, as can be seen in Figure 3a,b; however, under 1473 K, some deviation occurs in the later reduction stage (Figure 3c). In view of the assumptions that are made in the model and errors in the measurements, the agreement between them is considered to be satisfying.

The shape of all of the mass-loss fraction curves presents some common features. The briquette reduction can be divided into three fairly distinguishable stages. The fifirst stage comprises the mass loss, the second stage corresponds to the mass loss reaching its maximum value, and the third stage includes the mass increase. The mass loss/gain characteristics became more evident with an increasing temperature.

Figure 3. Model-predicted and experimental mass-loss fraction curves under different temperatures: (a) 1273 K; (b) 1373 K; and, (c) 1473 K.

5.3 Briquette Reduction Degree and Briquette Carbon Conversion

Because the mass change of the briquette is affected by several reactions, it cannot adequately reflflect the briquette reduction behavior. However, the briquette reduction degree and carbon conversion are important parameters for evaluating the quality of the reduced briquette. In the simulations, the briquette reduction degree at time t was calculated using Equation (23), and the briquette carbon conversion at time t using Equation (24).

The model-predicted and measured briquette reduction degrees under three temperatures are shown in Figure 4. The average difference in reduction degree between model predictions and experimental measurements was less than 0.03; therefore, it can be concluded that the developed model is applicable to the reduction of the ore-carbon briquette. Model predictions at 1273 K present a successive increase of reduction degree throughout the reduction period. At 1373 K, the reduction degree increased in the early stage and reached its maximum reduction degree of 0.75 at approximately 12 min; thereafter, it decreased. Briquette reduction behavior under 1473 K was similar to that under 1373 K, except that it reached a maximum reduction degree of 0.80 at approximately 6 min. After 15 min, the reduction degrees of the briquette under 1473 K and under 1373 K are quite close, and by 20 min, the reduction degrees under the three temperatures are nearly the same. These fifindings indicate that, under the oxidative atmosphere, increasing the temperature does not increase the fifinal reduction degree of the briquette.

The model-predicted briquette carbon conversion is compared to the corresponding experimental measurements in Figure 5. The briquette carbon conversion increased with time at all three temperatures. Model-predicted curves at 1373 K and 1473 K show that, after reaching their maximum reduction degree (at 12 min under 1373 K and at 7 min under 1473 K), the carbon conversion increased much more slowly. By 20 min, the fifinal carbon conversion was lower at 1273 K than at 1373 K and 1473 K. Combing the results from Figures 4 and 5 indicates that more biochar could be consumed by the atmosphere with the increase of the temperature.

5.4 Briquette Reduction Progress

Briquette reduction under 1473 K was selected for further study of the briquette reduction progress because it displayed the major characteristics of hematite-biochar reduction under the oxidative atmosphere: fast reduction in the early stage and obvious metallic iron oxidation in the later stage.

The simulation results of the development of local PCO/PCO2, the local reduction degree profifile, and the local carbon conversion profifile on the briquette cross section are displayed in Figure 6. In Figure 6, the local reduction degree and carbon conversion at time t were calculated by 1.0 − (3ρFe2O3 /MFe2O3 + 4ρFe3O4 /MFe3O4 + ρFeO/MFeO)/(3ρFe2O3,0/MFe2O3 ) and 1.0 − ρC/ρC,0, respectively. These simulation results were then used to assume the reduction progress of the briquette. At the beginning stage, the dominant reactions were assumed as the reactions that are given by Equations (1), (2) and (4). Owing to the intense gas generation, the briquette reduction progress was assumed to not be inflfluenced by the atmosphere or gas diffusion. At 1 min, PCO/PCO2 was nearly uniformly profifiled at a level of approximately 4.0; both the reduction degree and the carbon conversion profifiles were uniform and at a degree of approximately 0.3. As time proceeded, the reactions that were given by Equations (3) and (4) became dominant. As the rate of generated product gas was reduced, the effect of gas diffusion increased. At 5 min, the atmosphere began to inflfluence the briquette reduction progress, and the PCO/PCO2 profifile in the briquette became uneven: at the core, it increased to 10.28, whereas at the surface, it only increased to 3.26. Correspondingly, the reduction degree increased to 0.88 at the core and to 0.60 at the surface, and carbon conversion increased to 0.76 at the core and to 0.90 at the surface. By 10 min, PCO/PCO2 remained higher than 2.95 in the internal part of the briquette (K5 = 2.95 under 1473 K), and reached 110 at the core, so that the reaction given by Equation (3) could still proceed there. However, PCO/PCO2 decreased to less than 2.95 at the surface of the briquette, and consequently, the reaction that was given by Equation (5) took place and became a dominant reaction. At 10 min, reduction degree increased in the internal area and reached 0.98 at the core; at the surface, it decreased to 0.50. By the end of the reduction, the dominant reactions changed to those given by Equations (4) and (5). The contour line of PCO/PCO2 = 2.95 became closer to the core, and PCO/PCO2 at the core decreased to 20.0, which reflflected that the atmosphere was extending its inflfluence toward the core. At 15 min, the reaction that was given by Equation (3) ceased. In the region with PCO/PCO2 > 2.95, the reduction degree remained higher than 0.90. However, the reaction given by Equation (5) became signifificant as the region with PCO/PCO2 < 2.95 was enlarged. The reduction degree at the surface further decreased to 0.40, and the weak gasifification of the remaining carbon particles was then attributed to CO2 from the atmosphere.

Figure 6. Simulation results on the development of the profifile of local PCO/PCO2, the profifile of local reduction degree, and the profifile of local carbon conversion.

Thus, the reduction of the briquette presented a mainly homogeneous reaction system in the period from 1 to 5 min, whereas the gas transfer and the oxidative atmosphere became considerable after 5 min. Both metallic re-oxidation and iron oxide reduction appeared as the briquette neared its maximum reduction degree.

SEM–EDS results on intermittent morphologies at the briquette side surface in the reduction course at 1473 K are shown in Figure 7. The ore particles were uniformly distributed at 5 min (Figure 7a), and a mixture of tiny metallic iron grains (Point 1 in Figure 7b), wustite grains (Point 2 in Figure 7b), and gangue grains (Point 3 in Figure 7b) was presented within ore particles (Figure 7b). The deformation and the sintering of ore particles had occurred by 10 min (Figure 7c), and iron grains became scarce in the particles (Figure 7d). The decrease of iron grains was due to the re-oxidation by the atmosphere, and the ore particle sintering and deforming were attributed to wustite reacting with gangue components (CaO, SiO2, and Al2O3) to form low-melting-point compounds (glass phase). By 15 min, the sintering degree of ore particles had increased (Figure 7e), and the size of some remaining iron grains in ore particles was enlarged (Figure 7f). Iron grain growth was attributed to an agglomeration of tiny iron grains that was facilitated by the glass phase. Overall, the SEM–EDS results indicate that the briquette side surface underwent oxidation after 5 min in the reduction course, which was in accordance with the simulated reduction process. The intense self-reduction in the early reduction stages lead to crack generation at the briquette surface so that iron oxidation near the surface was accelerated in the later stage as the porosity near the surface was increased. Therefore, crack formation near the surface could be the main reason for the deviation between model predictions and the experimental measurements in the later stage of reduction, as shown in Figure 3c.

Figure 7. Intermittent scanning electron microscopy–energy dispersive spectrometry (SEM–EDS) results at the briquette surface: (a,b) 5 min; (c,d) 10 min; (e,f) 15 min; and (g–i) EDS results of Points 1, 2 and 3 in Figure 7b, respectively.

6. Conclusions

1. A model to predict the reduction behavior of the ore-carbon briquette under CO–CO2 atmosphere was developed. The model included the kinetics of the stage-wise reduction of iron oxide, carbon gasifification and metallic iron oxidation, and it was with the assumptions of constant porosity and size of the briquette. The simulation results were validated by the experimental measurements and observations and the model was found to be reliable.

2. The CO–CO2 atmosphere can signifificantly inflfluence the fifinal reduction degree of the briquette, and the briquette cannot reach higher fifinal reduction degree by further increasing the temperature. Under higher temperatures, more carbon is consumed by the reactive atmosphere.

3. In the briquette reduction progress, the briquette reduction behavior is not initially inflfluenced by the CO–CO2 atmosphere; however, near the maximum reduction degree, both iron oxide reduction and metallic iron re-oxidation can occur in the briquette.

Tang, H., Yun, Z., Fu, X., & Du, S. Modeling and Experimental Study of Ore-Carbon Briquette Reduction under CO–CO2 Atmosphere. Metals, 8(4), 205. https://doi.org/10.3390/met8040205

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).