Dr Subhasish Mitra

Performance of ironmaking blast furnace systems critically depends on the permeability of the cohesive zone which involves flow of reducing gas through a packed bed of fused ferrous and coke particles at high temperature. This study systematically investigated the effect of operating temperature on bed permeability for two different ferrous burdens - pellets, and pellets-sinter mixture supported on coke particles. To quantify the structural changes in the bed, interrupted tests at various temperatures were conducted and bed porosity was estimated using synchrotron X-ray computed tomography (CT) technique. Bed porosity showed decreasing trends with increasing temperature. A CT image based computational flow dynamics (CFD) model was developed which showed linearly decreasing trends of bed permeability parameter with increasing temperature indicating a significant loss of bed permeability in the high temperature cases. This behaviour was more pronounced for the pure pellet case compared to the mixed burden case.

This study aimed to quantify the less explored complex multiphase hydrodynamics of a bubble swarm in the presence of surfactant which is the key to flotation process widely used in the resources and environmental engineering applications. Experiments were conducted in a rectangular column (cross-section: 100 mm × 100 mm) in batch mode by varying the gas flux (0.02 to and 0.08 cm/s) in the presence of an anionic surfactant sodium dodecyl sulphate (5 to 15 ppm). High-speed imaging was then used to visualise the unsteady bubble plume dispersion behaviour and estimate the bubble size distribution (BSD) which showed a reduction in the mean bubble size with the increasing gas flux and surfactant concentration. Next, particle image velocimetry (PIV) was utilised to measure the instantaneous velocity field which was used to determine the turbulence characteristics. It was shown that the bubble plume contributes to significant anisotropy in the flow field which increased in the higher surfactant concentration and gas flux cases. The energy containing turbulence length scale was characterized by the integral length scale, which was observed to increase linearly with both the gas flux and surfactant concentration. Also, the local turbulence energy dissipation rate exhibited a strong linear correlation with the bubble surface area flux parameter. In the presence of surfactant, the turbulence energy spectrum of the system exhibited a less steep slope in the inertial subrange regime compared to the Kolmogorov -5/3 slope. The spectrum also showed a leftward shift indicating energy addition to the larger turbulence length scales which was reflected in the formation of large recirculation zones around the bubble plume.

Abstract: With the movement toward hydrogen-enriched blast furnace operation to lower greenhouse gas emissions, ferrous burden design must be reconsidered to optimize furnace permeability. Increasing the ratio of direct charge lump ore in the ferrous burden also presents an opportunity to lessen the emissions associated with the production of sinter and pellets. Under traditional blast furnace conditions, lump ore usage is improved by mixing it with the sinter in the burden to promote their favorable high-temperature interactions (both chemical and physical). As such, mechanistic changes to the interaction must be understood to optimize burden design, including for future operations with hydrogen addition. In this study, liquid formation in both the metallic and oxide components of ferrous burdens is microscopically investigated. Oxide liquid and solid phase stability at the interfaces of dissimilar burdens are visualized using a novel mapping technique, and metallic iron is etched to reveal microstructures indicative of carbon. Results indicate that the inclusion of hydrogen promotes the gas carburization of metallic iron in sinter, but not lump. It was concluded that mixed burden softening and melting performance with hydrogen addition were improved through the addition of lump in two ways: the highly metallic lump particles provide structural support for the collapsing sinter bed and also suppress the formation of early liquid slag from the sinter. Graphical Abstract: (Figure presented.).

Hydrogen-enriched blast furnace (BF) operation is currently being assessed to mitigate greenhouse gas emissions while the steelmaking industry transitions to low carbon emission technologies. Increasing the usage of lump ore in the BF also presents opportunity to decrease carbon emissions, as it can be directly charged to the furnace without agglomeration. Use of lump ore in modern blast furnace operations is facilitated by high temperature interactions with sinter. With more emphasis on hydrogen enrichment in BF operations, the behaviour of lump and sinter mixed burdens must be characterised under new conditions. In this study, 15% hydrogen is added to the standard gas conditions of a Softening and Melting (S&M) apparatus (replacing nitrogen). Analysis of auxiliary reactions such as the Boudouard Reaction and the Water-Gas Shift Reaction is presented and their impact on burden reduction and performance assessed. Results indicate that with the inclusion of hydrogen, the performance of sinter burden deteriorates, while lump burden shows significant improvement. Interaction between sinter and lump still occurred with the inclusion of hydrogen in the gas, and the mixed burden behaviour of 20% lump and 80% sinter fell between that of the individual burdens. From interrupted experiments, it is noted at high degrees of reduction, the lump burden forms a solid metallic layer which maintains its interparticle voidage at high temperatures, supressing exudation of liquid slag.

In the present work, an experimental analysis was performed to characterise the flow field around a single bubble of different diameters ~ 2.77¿3.53 mm) rising in a quiescent medium aiming to determine the effect of bubble size on kinetic energy distribution. The velocity field was measured using a non-intrusive particle image velocimetry (PIV) technique and kinetic energy spectrum was determined in both transverse and longitudinal directions applying a Fast Fourier Transformation (FFT). Both small- and large-scale motions of the flow field were identified and separated using a discreate wavelet transformation (DWT) method. It was found that the energy spectrum of the large-scale motions depended on the bubble size while the small-scale energy spectrum was nearly independent of it. The slopes of the energy spectrum were found to be close to -5/3 and -3 for the large- and small-scale regimes, respectively and the transition of slope was observed to occur at the wavenumber corresponding to the bubble diameter. Using the measured velocity field data, a turbulence kinetic energy (TKE) budget analysis was performed involving five components namely kinetic energy production, turbulent transport, pressure diffusion, viscous diffusion, and energy dissipation. Overall, it was observed that in the vicinity of bubble surface, turbulence production term was not entirely balanced by the dissipation term; and turbulent transport and pressure diffusion term also had significant contributions.

In flotation, interactions of bubble-particle aggregates with turbulent flow structures in the liquid medium result in particle detachment. This study aims to simulate this phenomenon involving a bubble-particle aggregate (bubble diameter ~ 3 mm and particle diameter ~ 314 µm) interacting with a turbulent flow structure manifested as a confined vortex in a square cavity connected to a square cross-section channel. An interface resolved three dimensional (3D) computational fluid dynamics (CFD) model was developed to quantify the bubble-vortex interaction dynamics over a range of channel Reynolds numbers. The CFD model produced a good agreement with the experimentally measured vorticity magnitude, local energy dissipation rate, and bubble motion. It was shown that a bubble-particle aggregate could be captured within the vortex by suitably varying the channel Reynolds number, eventually leading to particle detachment. A separate force balance analysis was performed to determine a criterion for particle detachment utilising the CFD model predicted vorticity and local energy dissipation rate. It was shown that a critical local energy dissipation rate ~ 1.59 m2/s3 was required for particle detachment to occur, which was also verified experimentally.

In this study, the role of cavitation bubbles in the bubble-particle interaction dynamics was investigated using high-speed imaging. A lab-scale conceptual flotation cell was designed involving an ultrasonic field with adjustable power output and a continuous liquid recirculation stream. Coarse size range hydrophobic glass Ballotini particles were suspended in the cell by the combined action of a confined vortex and turbulence created by periodic ultrasonic pulses. Ultrasonic pulses in the liquid medium led to the generation of numerous fine size range cavitation bubbles which were observed to form particle clusters. These particle clusters were observed attaching to relatively larger size carrier bubbles forming stable bubble-particles aggregates, which subsequently floated. It was noted that a favourable state of bubble-particle interactions was achievable to produce such stable bubble-particles aggregates by suitably controlling the input power and period of the applied ultrasonic pulse.

In the iron ore sintering process, the resistance to air flow is a major factor in deciding the flame front speed, which influences the sinter productivity and quality. In this work, pressure drop during sintering and the resistance to air flow was investigated in milli-pot sintering for different coke rates. The sintering experiments were conducted in a milli-pot (diameter 53 mm, height 400 mm) and pressure and temperature were measured at the same locations in the bed by four taps located equidistant to each other. The yield of sinter product was measured following a modified drop test and the mineralogy of the sinter product was analysed. The results from milli-pot sintering were then compared to the reported results from standard pilot-scale sintering, and it was found that the lower half of the milli-pot bed gave a reasonable representation of the pilot-scale sintering process. The results of sinter mineralogy, yield and productivity of the lower half of milli-pot at 5.5-8.0% coke rate were found to be similar to pilot-scale sintering tests at a corresponding coke rate from 3.5 to 5.5%. The maximum resistance to air flow in the bed was found to be in the region between the leading edge of the flame front at ~100°C and the trailing edge of the flame front at ~1 200°C. This suggests that the maximum resistance to air flow includes the effect of de-humidification and combustion in addition to the high temperature "flame front" region usually defined at temperatures above 1 100°C or 1 200°C.

In this study, we report the rising behaviour of the millimetric size ellipsoidal shaped particle-laden bubbles (particle diameter dP ~ 114 µm, bubble diameter dB ~ 2.76 and 3.34 mm) in the range of bubble surface loading (BSL) from 0 to 0.6 both in absence and presence of a surfactant (Sodium Dodecyl Sulphate, 20% CMC). High-speed imaging was used to capture the trajectory of the particle-laden bubble and an image processing methodology was developed to quantify the bubble surface loading. Three different regimes were observed - bubble shape transition (nearly spherical to ellipsoidal), particle detachment (at bubble rear end), and steady (for high BSL) or expansion (for low BSL) of the particle surface covered zone. A threshold for bubble surface loading (BSL ~ 0.40) was determined which had reasonable agreement with the experimental observations. Bubble rise velocity was observed to decrease with bubble surface loading but this trend was less steep in presence of surfactant. It was noted that loss of bubble surface mobility was higher in presence of surfactant, however in absence of surfactant, bubble surface loading contributed significantly to surface immobility. Finally, a correction factor to Schiller-Naumann drag coefficient model was proposed accounting for the bubble surface loading both in presence and absence of surfactant.

Only when the process of particle detachment is well understood and modelled can minerals recovery using the flotation process be modulated to achieve a high efficiency by suitably changing the operating parameters. This is vitally necessary for the recovery of coarse particles in an energy efficient way, as detachment is the key limiting factor in the successful recovery of large particles. However, until the detachment mechanism is more fully understood, an upper limit on the floatable particle diameter still remains unidentified. To assess the current state of knowledge available in this area, a comprehensive literature review on the mechanisms and models of the bubble-particle detachment process in froth flotation is presented. In general, the detachment process is considered to be a stochastic process, and is usually attributed to the dynamic interactions with the turbulent flow structures (eddies) in the flotation environment which cause particles to detach because of dissipating energy. In this paper, previous studies on bubble-particle detachment have been critically analyzed with respect to the formulation of the models in predicting the detachment probability of particles. The models are classified into three different categories: force balance analysis; energy balance analysis and empirical analysis of particle size compared to maximum floatable particle size. Attention is also paid to an understanding of the mechanisms of bubble-particle detachment in quiescent and turbulent liquid flow fields. The predictions of all these models have been compared with the published experimental data and it was found that models which take an accurate consideration of the influence of eddies on a particle's detachment give the closest predictions. The generally held concept of bubble-particle detachment inside an eddy was experimentally validated, where a particle was observed to rotate on the surface of a bubble, resulting in a centrifugal acceleration 20 times that of gravitational acceleration. The aim of this paper is to review the developments and limitations of the existing models. The experimental work is reviewed so as to reveal the mechanisms of bubble-particle detachment. Therefore, the future development of models is identified in order to successfully predict particle detachment.

Vaporization of atomized feedstock is one of the critical processes in fluid catalytic cracking (FCC) risers; which is more often ignored in most of the FCC riser modelling studies. In this study, two different vaporization mechanisms of feedstock namely homogeneous mode and heterogeneous mode were studied. Different homogeneous models duly validated for various pure component droplets were applied to predict the vaporization time of the feed droplets typically expected in FCC feed vaporization zone. A new physical model for heterogeneous vaporization considering droplet-particle collision mechanics was also developed in the present study which compared well with the other existing heterogeneous modelling approaches. Comparison of the two vaporization modes indicates that under typical operating conditions of FCC riser, vaporization time of feed droplets predicted by heterogeneous mode is always lower than the homogeneous mode at least by an order of magnitude due to significant increase in heat transfer coefficient which accounts for droplet-particle contact. It is expected that actual vaporization time of feed droplets in an industrial FCC riser should lie in the range predicted by these two vaporization mechanisms which actually set the two limiting modes of vaporization. Obtained results predicted by the models could be used to aid design of the FCC feed vaporization zone.

It is critical to have an efficient energy budget in all the industrial process applications involving multiphase flow system where a significant amount of power is invested to achieve a desired outcome such as valuable particle collection and recovery in mineral flotation circuits. In order to achieve this aim there needs to be an ability to properly characterise the energy dissipation in the system; and from this knowledge to develop methodologies so that the supplied energy is distributed suitably among the eddies of different sizes which are responsible for enhancing different transport events such heat/mass transfer, mixing etc. The aim of the study was to obtain the 2D instantaneous velocity field in a homogeneous near isotropic turbulence field using particle image velocimetry (PIV) and then compute the space and time averaged specific energy dissipation rate from velocity field using four different methods, namely: (1) dimensional analysis, (2) velocity gradient, (3) structure function, and (4) energy spectrum. The system was studied in the Taylor Reynolds number range of 24-60, where it was found that the difference between the computed specific energy dissipation rates could be as much as 100 percent. Whilst it was found that there were uncertainties in all four methodologies, it is argued that the energy spectrum method is likely to give the most realistic quantification of the specific energy dissipation rate value since it was shown to satisfy the system energy balance which was not possible to do so for the other three methods. The energy spectrum method also had the added benefit of incorporating integral scale, Taylor microscale and Kolmogorov length scales in the quantification of the specific energy dissipation rate; whereas the other three methods are limited to either integral scale or Taylor microscale only. The limitation of the energy spectrum method, however, is the resolution of the energy spectrum down to the Kolmogorov length scale due to the noise in the measurement; and to resolve this problem a filter was applied to denoise in the dissipation range.

Turbulence plays a critical role in detachment process of bubble from a solid surface. To investigate this effect, detachment process of a stationary air bubble from a nozzle in both quiescent and turbulent liquid field was studied. A stationary vertical (flat-ended) needle of ID 1.24mm was used as a nozzle to generate a bubble which was anchored to the needle tip. Different sizes of bubbles were generated in quiescent liquid. Volume and contact angle for these bubbles were measured precisely using microscopic imaging technique and correlated. In the quiescent case experiments, a constant contact angle of 90° and bubble diameter of 3.05±0.004mm were obtained consistently. A simple force balance approach was proposed assuming bubble in equilibrium to determine this maximum bubble diameter during detachment. The detached bubble size calculated using this approach agreed fairly well with the experimental results. An oscillating grid device capable of operating at different frequencies was then applied to generate a homogeneous, near-isotropic turbulent velocity field around the anchored bubble. It was observed that for detachment of smaller bubbles, higher turbulence intensity was indeed necessary. The turbulent flow field was quantified using particle image velocimetry (PIV) technique and resolved into flow structures (eddies) of different length scales using a Gaussian filter. It was concluded that smaller eddies perturbed the bubble interface whilst the larger eddies contributed to weakening of the capillary force causing the bubble detachment. Energy dissipation profile obtained from the PIV images indicated significant energy dissipation near the bubble compared to the bulk fluid which supported the fact that strong interactions between bubble and eddies were indeed responsible for bubble detachment.

Accurate prediction of the interactions between evaporating liquid droplets and solids are critical for many industrially important processes. A model based on coupled Level Set-Volume of Fluid approach was developed to simulate the interaction of evaporating liquid droplets with hot solid surfaces. The model incorporates appropriate source terms in the multiphase calculations to account for the heat and mass transfer. Accurate and stable numerical procedure was developed and incorporated in open source solver OpenFOAM. A brief discussion on the model development along with several key issues that are associated with this process was presented. The resulting numerical model was validated through the experimental data of Chandra and Avedisian (Chandra, S., Avedisian, C.T., 1991. Proc. R. Soc. Lond., Ser. A 432, 13-41). Although some discrepancies were found between the numerical results and experimental data, the model was found to be capable of reproducing the reduced droplet spreading rate as the temperature of the surface is increased away from the saturation temperature. The decrease in rate of surface wetting results from the combined effects of surface tension, viscous forces and evaporation at the liquid-solid-vapour contact line. Further, the effects of increased pressure at the solid-liquid interface resulting from the rapid evaporation of the liquid, which in some cases can be quite severe such that the liquid gets lifted-off from the surface, were also captured, in good agreement with experimental observations. Finally, the effects of the solid temperature on the evaporation and heat transfer rates of the droplets were presented and analysed. © 2014 Elsevier Ltd.