CFD simulation of a gas–solid fluidized bed with two vertical jets

A computational fluid dynamics (CFD) model is used to investigate the hydrodynamics of a gas–solid fluidized bed with two vertical jets. Sand particles with a density of 2660 kg/m³ and a diameter of 5.0 × 10^?4 m are employed as the solid phase. Numerical computation is carried out in a 0.57 m × 1.00 m two-dimensional bed using a commercial CFD code, CFX 4.4, together with user-defined Fortran subroutines. The applicability of the CFD model is validated by predicting the bed pressure drop in a bubbling fluidized bed, and the jet detachment time and equivalent bubble diameter in a fluidized bed with a single jet. Subsequently, the model is used to explore the hydrodynamics of two vertical jets in a fluidized bed. The computational results reveal three flow patterns, isolated, merged and transitional jets, depending on the nozzle separation distance and jet gas velocity and influencing significantly the solid circulation pattern. The jet penetration depth is found to increase with increasing jet gas velocity, and can be predicted reasonably well by the correlations of Hong et al. (2003) for isolated jets and of Yang and Keairns (1979) for interacting jets.     

CFD study: Effect of pulsating flow on gas–solid hydrodynamics in FCC riser

Gas–solid flow in a fluid catalytic cracking (FCC) riser exhibits poor mixing in the form of a core–annulus flow pattern and a dense bottom/dilute top distribution of solids. To enhance gas–solid mixing, studies on dense fluidized beds have suggested using a pulsating flow of gas. The present study investigates the effect of pulsating flow on gas–solid hydrodynamics inside the FCC riser employing computational fluid dynamics. Two flow conditions are investigated: a cold flow of air-FCC catalyst in a pilot-scale riser and a reactive flow in an industrial-scale FCC riser. In the cold-flow riser, pulsating flows cause the slug flow of solids and thus increase the average solid accumulation in the flow domain and solid segregation towards the wall. In the industrial FCC riser, pulsating flows produce radial profiles that are more homogeneous. Pulsating flows further improve the conversion and yield in the initial few metres of height. At 7 m, the conversion from pulsating flow is 59%, compared with 44% in without pulsating flow. The results and analysis presented here will help optimize flow conditions in the circulating fluidized bed riser, in not only FCC but also applications such as fast pyrolysis and combustion.     

Using S-statistic for investigating the effect of temperature on hydrodynamics of gas–solid fluidization

The influence of temperature on fluidization was investigated by a statistical chaotic attractor comparison test known as S-statistic. After calibration of the variables used in this method, the S-test was applied to the radioactive particle tracking (RPT) data obtained from a lab-scale fluidized bed. Experiments were performed with sand as fluidized particles and in temperatures from ambient up to 600 °C with superficial gas velocities of 0.29, 0.38 and 0.52 m/s. Considering the behavior of bubbles and comparing with frequency domain analysis, it was concluded that S-statistic is a reliable method for characterization of fluidization process behavior at different temperatures.     

An investigation into flow mode transition and pressure fluctuations for fluidized dense-phase pneumatic conveying of fine powders

This paper presents the results of an ongoing investigation into the fluctuations of pressure signals due to solids–gas flows for dense-phase pneumatic conveying of fine powders. Pressure signals were obtained from pressure transducers installed along different locations of a pipeline for the fluidized dense-phase pneumatic conveying of fly ash (median particle diameter 30 μm; particle density 2300 kg/m³; loose-poured bulk density 700 kg/m³) and white powder (median particle diameter 55 μm; particle density 1600 kg/m³; loose-poured bulk density 620 kg/m³) from dilute to fluidized dense-phase. Standard deviation and Shannon entropy were employed to investigate the pressure signal fluctuations. It was found that there is an increase in the values of Shannon entropy and standard deviation for both of the products along the flow direction through the straight pipe sections. However, both the Shannon entropy and standard deviation values tend to decrease after the flow through bend(s). This result could be attributed to the deceleration of particles while flowing through the bends, resulting in dampened particle fluctuation and turbulence. Lower values of Shannon entropy in the early parts of the pipeline could be due to the non-suspension nature of flow (dense-phase), i.e., there is a higher probability that the particles are concentrated toward the bottom of pipe, compared with dilute-phase or suspension flow (high velocity), where the particles could be expected to be distributed homogenously throughout the pipe bore (as the flow is in suspension). Changes in straight-pipe pneumatic conveying characteristics along the flow direction also indicate a change in the flow regime along the flow.     

Hydrodynamics and energy consumption studies in a three-phase liquid circulating three-phase fluid bed contactor

The hydrodynamics and energy consumption have been studied in a cold flow, bubbling and turbulent, pressurized gas–liquid–solid three-phase fluidized bed (0.15 m ID × 1 m height) with concurrent gas–liquid up flow is proposed with the intention of increasing the gas hold up. The hydrodynamic behaviour is described and characterised by some specific gas and liquid velocities. Particles are easily fluidized and can be uniformly distributed over the whole height of the column. The effect of parameters like liquid flow rate, gas flow rate, particle loading, particle size, and solid density on gas hold up and effect of gas flow rate, solid density and particle size on solid hold up, energy consumption and minimum fluidization velocity has been studied. At the elevated pressures a superior method for better prediction of minimum fluidization velocity and terminal settling velocities has been adopted. The results have been interpreted with Bernoulli’s theorem and Richardson–Zaki equation. Based on the assumption of the gas and liquid as a pretend fluid, a simplification has been made to predict the particle terminal settling velocities. The Richardson–Zaki parameter n′ was compared with Renzo’s results. A correlation has been proposed with the experimental results for the three-phase fluidization.     

Rotational asymmetry of reactant concentration and its evolution in a circulating fluidized bed riser

Rotational asymmetric distribution of reactant (ozone) concentration and its evolution along with the gas-solid reactive flow were studied in a 76 mm i.d., 10.2 m high circulating fluidized bed (CFB) riser reactor. The superficial gas velocity ranged from 3 to 5 m/s and the solids circulation rates were 50 and 100 kg/(m² s). Experimental results show that the asymmetry of reactant distribution can extend to a height close to the length of flow developing zone of the CFB riser reactor and then disappears. Based on the hydrodynamics of the gas and solid phases in the solids entrance region, this asymmetry can be attributed to the effect of the solids entrance structure.     

Experimental investigation on flow asymmetry in solid entrance region of a square circulating fluidized bed

To study the influence of back feeding particles on gas-solid flow in the riser, this paper investigated the flow asymmetry in the solid entrance region of a fluidized bed by particle concentration/velocity measurements in a cold square circulating fluidized beds （CFB）. The pressure drop distribution along the riser and the saturation carrying capacity of gas for Geldart-B type particles were first analyzed. Under the condition of u0 = 4 m/s and Gs = 21 kg/（m^2 s）, the back feeding particles were found to penetrate the lean gas-solid flow near the entrance （rear） wall before reaching the opposite （front） wall, thus leading to a relatively denser region near the front wall in the bottom bed. Higher solid circulation rate （u0 =4 m/s, Gs = 33 kg/（m^2 s）） resulted in a higher particle concentration in the riser. However the back feeding particles with higher momentum increased the asymmetry of the particle concentration/velocity profile in the solid entrance region. Lower air velocity （u0 =3.2 m/s） and Gs =21 kg/（m2 s）, beyond the saturation carrying capacity of gas, induced an S-shaped axial solid distribution with a denser bottom zone. This limited the penetration of the back feeding particles and forced the flnidizing air to flow in the central region, thus leading to a higher solid holdup near the rear wall. Under the conditions of uo = 4 m/s and Gs = 21 kg/（m^2 s）, addition of coarse particles （dp= 1145 μm） into the bed made the radial distribution of solids more symmetrical.     

Lagrangian velocity and acceleration statistics of fluid and inertial particles measured in pipe flow with 3D particle tracking velocimetry

Three-dimensional particle tracking velocimetry (3D-PTV) has been applied to particle-laden pipe flow at Reynolds number 10,300, based on the bulk velocity and the pipe diameter. The volume fraction of the inertial particles was equal to 1.4 × 10⁻⁵. Lagrangian velocity and acceleration statistics were determined both for tracers and for inertial particles with Stokes number equal to 2.3, based on the particle relaxation time and the viscous time scale. The decay of Lagrangian velocity and acceleration correlation functions was measured both for the fluid and for the dispersed phase at various radial positions. The decay of Lagrangian velocity correlations is faster for inertial particles than for flow tracers, whereas the decay of Lagrangian acceleration correlations is about 25% slower for inertial particles than for flow tracers. Further differences between inertial and tracer particles are found in velocity fluctuations evaluated for both positive and negative time lags. The asymmetry in time of velocity cross-correlations is more pronounced for inertial particles. Quadrant analysis revealed another difference still near the wall: ejection and sweep events are less frequent for inertial particles than for tracers.     

Effects of various parameters on the attrition of bed material in a recirculating fluidized bed with a draft tube

We performed an experimental study to investigate the effects of various parameters on the attrition of bed material and its size distribution with increasing operation time in a recirculating fluidized bed (RCFB). The studied parameters included superficial velocity of fluidizing air, bed inventory, and spacing between the jet top and draft tube bottom (spacer height). The bed material was prepared from Indian Standard (IS) Grade I sand from sieves with a size range of 2.20–1.00 mm. Experiments were performed at ambient conditions, with the superficial air velocity ranging from 7.13–9.16 m/s, a bed inventory of 7–10 kg, spacing of 0.085 and 0.045 m between the jet top and draft tube bottom, and an operating time of 40 h. We investigated the influence of these parameters in terms of changes in the size distribution of particles, changes in the %-weight of particles of different size ranges, generation of particles with smaller diameters, the decrease of the downcomer bed height, variations in the coefficient of uniformity and coefficient of curvature, and material loss from entrainment of fines with increasing operation time. The mode of attrition was abrasion in all experiments. We found that with increasing operation time and other parameters (bed inventory, superficial air velocity, and spacer height) attrition of the bed material also increased. Generation and elutriation of fines were more pronounced at higher superficial air velocity, bed inventory, and spacer height.     

The influence of temperature on limestone sulfation and attrition under fluidized bed combustion conditions

The influence of temperature on attrition of two limestones during desulfurization in a fluidized bed reactor was investigated. Differences in the microstructure of the two limestones were reflected by a different thickness of the sulfate shell formed upon sulfation and by a different value of the ultimate calcium conversion degree. Particle attrition and fragmentation were fairly small under moderately bubbling fluidization conditions for both limestones. An increase of temperature from 850 °C to 900 °C led to an increase of the attrition rate, most likely because of a particle weakening effect caused by a faster CO₂ evolution during calcination. This weakening effect, however, was not sufficiently strong to enhance particle fragmentation in the bed. The progress of sulfation, associated to the build-up of a hard sulfate shell around the particles, led in any case to a decrease of the extent of attrition. Sulfation at 900 °C was less effective than at 850 °C, and this was shown to be related to the porosimetric features of the different samples.     

Flow fields and packing states in the discharge flow of noncircular particles—A SIPHPM simulation

Although the discharge flow of spherical materials has been extensively explored, the effect of particle shape on discharge is still poorly understood. The present work explores the two-dimensional discharge flow fields of noncircular particles using the soft-sphere-imbedded pseudo-hard particle model method. Rectangular particles having different aspect ratios (R_a = 1, 1.5, 2–5) and regular polygonal particles having different numbers of sides (N_s = 3–8, 10) are discharged through hopper beds having different orifice widths (D_i = 40, 70.77, 99.13, 125.74, 151.13 mm). The discharge rates of differently shaped particles in different beds are consistent with Beverloo’s relation. Moreover, the flow fields are computed and evaluated to study the effects of R_a, N_s, and D_i on particle discharge. The characteristics of particle–particle connections in the discharge process are evaluated according to the temporal evolution and spatial distribution of the contact points. Additionally, the effect of the initial packing on the discharge profile is investigated. The findings help clarify the discharge of noncircular particles.     

Direct reduction of hematite powders in a fluidized bed reactor

Ultrafine hematite powder was reduced to produce ultrafine iron powder in a 50%Ar–50%H₂ atmosphere at 450–550 °C in a fluidized bed reactor. The ultrafine hematite powder shows the typical agglomerating fluidization behavior with large agglomerates fluidized at the bottom of the bed and small agglomerates fluidized at the upper part of the bed. It was found that defluidization occurred even at the low temperature of 450 °C with low metallization rate. Defluidization was attributed mainly to the sintering of the newly formed iron particles. Granuation was employed to improve the fluidization quality and to tackle the defluidization problem, where granules fluidized like a Geldart's group A powder. Granulation was found to effectively reduce defluidization during reduction, without however sacrificing reduction speed. The as-reduced iron powders from both the ultrafine and the granulated hematite exhibited excellent sintering activity, that is, fast sintering at temperature of as low as ～580 °C, which is much superior as compared to that of nano/ultrafine iron powders made by other processes.     

Experimental investigation into transient pressure pulses during pneumatic conveying of fine powders using Shannon entropy

This paper presents the results of an ongoing investigation into transient pressure pulses using Shannon entropy. Pressure fluctuations (produced by gas–solid two-phase flow during fluidized dense-phase conveying) are recorded by pressure transducers installed at strategic locations along a pipeline. This work validates previous work on identifying the flow mode from pressure signals (Mittal, Mallick, & Wypych, 2014). Two different powders, namely fly ash (median particle diameter 45 μm, particle density 1950 kg/m³, loosely poured bulk density 950 kg/m³) and cement (median particle diameter 15 μm, particle density 3060 kg/m³, loosely poured bulk density 1070 kg/m³), are conveyed through different pipelines (51 mm I.D. × 70 m length and 63 mm I.D. × 24 m length). The transient nature of pressure fluctuations (instead of steady-state behavior) is considered in investigating flow characteristics. Shannon entropy is found to increase along straight pipe sections for both solids and both pipelines. However, Shannon entropy decreases after a bend. A comparison of Shannon entropy among different ranges of superficial air velocity reveals that high Shannon entropy corresponds to very low velocities (i.e. 3–5 m/s) and very high velocities (i.e. 11–14 m/s) while low Shannon entropy corresponds to mid-range velocities (i.e. 6–8 m/s).     

Evaluation of the dimensionless groups based scale-up of gas–solid spouted beds

In this work, the scale-up methodology of He et al. (1997) that is based on maintaining similar or close dimensionless groups in gas–solid spouted beds has been evaluated. Two geometrically similar spouted beds of 0.152 m and 0.076 m diameter have been used. It has been demonstrated experimentally, that there is non-similarity in the local hydrodynamic parameters such as solids holdup and dimensionless solids velocity, when all the dimensionless groups have been matched or close to each other in the two studied spouted beds. This confirms that the global hydrodynamic parameters should not be used to confirm the validity of or to evaluate the dimensionless groups scale-up based methodologies.     

Characterization of various structures in gas-solid fluidized beds by recurrence quantification analysis

Gas-solid fluidized beds are widely considered as nonlinear and chaotic dynamic systems. Pressure fluc- tuations were measured in a fluidized bed of 0.15 m in diameter and were analyzed using multiple approaches： discrete Fourier transform （DFT）, discrete wavelet transform （DWT）, and nonlinear recur- rence quantification analysis （RQA）. Three different methods proposed that the complex dynamics of a fluidized bed system can be presented as macro, meso and micro structures. It was found from DFT and DWT that a minimum in wide band energy with an increase in the velocity corresponds to the transition between macro structures and finer structures of the fluidization system. Corresponding transition veloc- ity occurs at gas velocities of 0.3, 0.5 and 0.6 m]s for sands with mean diameters of 150, 280 and 490/~m, respectively. DFT, DWT, and RQA could determine frequency range of0-3.125 Hz for macro, 3. ! 25-50 Hz for meso, and 50-200 Hz for micro structures. The RQA showed that the micro structures have the least periodicity and consequently their determinism and laminarity are the lowest. The results show that a combination of DFT, DWT, and RQA can be used as an effective approach to characterize multi-scale flow behavior in gas-solid fluidized beds.     

Constraint strength and axial/radial particle velocity profiles for an integrated riser outlet

To study axial/radial profiles of particle velocity in the affected region of an integrated riser outlet, a cold model was developed for the integrated riser reactor combining the gas–solid distributor with the fluidized bed. Constraints, related to the gas–solid distributor and the upper fluidized bed, imposed on the particle flow in the riser outlet region, were investigated experimentally. The experimental results showed that with increasing superficial gas velocity, these constraints have strong influences on particle flow behavior, the particle circulation flux in the riser, and the height of the static bed material of the upper fluidized bed. When the constraints have greater prominence, the axial profile of the cross-sectionally averaged particle velocity in the outlet region initially increases and then decreases, the rate of decrease being proportional to the constraint strength. Along the radial direction of the outlet section, the region where the local particle velocity profile tends to decrease appears near the dimensionless radius r/R = 0.30 initially and then, with increasing constraint strength, gradually extends to the whole section from the inner wall. Based on the experimental data, an empirical model describing the constraint strength was established. The average relative error of the model is within 7.69%.     

Study of material agglomeration during nozzle injection of multiviscosity liquid into a fluidized bed reactor by the electric conductance method

Fluidized bed agglomeration is an important and challenging problem for thermal cracking in fluid cokers. A low coker temperature can be problematic because the bitumen is injected into the fluidized bed with a different viscosity, resulting in formation of agglomerates of varying sizes, which slows the cracking reactions. In the present study, the bed material agglomeration process during nozzle injection of multiviscosity liquid was investigated in a fluidized bed operated at different mass ratios of the atomization gas to the liquid jets (GLR = 1%–3.5%) and gas velocities (3.9U_mf and 5.9U_mf) based on a conductance method using a water–sand system to simulate the hot bitumen–coke system at room temperature. During the tests of liquid-jet dispersion throughout the bed, different agglomeration stages are observed at both gas velocities. The critical amount of tert-butanol in the liquid jets that could lead to severe agglomeration of the bed materials (poor fluidization) at GLR = 1% is about 10 wt% at the low fluidizing gas velocity (3.9U_mf) and 18 wt% at the high gas velocity (5.9U_mf). This study provides a new approach for on-line monitoring of bed agglomeration during liquid injection to guarantee perfect contact between the atomized liquid and the bed particles.     

Vibration time series analysis of bubbling and turbulent fluidization

A non-intrusive vibration monitoring technique was used to study the hydrodynamics of a gas–solid fluidized bed. Experiments were carried out in a 15 cm diameter fluidized bed using 226, 470 and 700 μm sand particles at various gas velocities, covering both bubbling and turbulent regimes. Auto correlation function, mutual information function, Hurst exponent analysis and power spectral density function were used to analyze the fluidized bed hydrodynamics near the transition point from bubbling to turbulent fluidization regimes. The first pass of the autocorrelation function from one half and the time delay at which it becomes zero, and also the first minimum of the mutual information, occur at a higher time delay in comparison to stochastic systems, and the values of time delays were maximum at the bubbling to turbulent transition gas velocity. The maximum value of Hurst exponent of macro structure occurred at the onset of regime transition from bubbling to turbulent. Further increase in gas velocity after that regime transition velocity causes a decrease in the Hurst exponent of macro structure because of breakage of large bubbles to small ones. The results showed these methods are capable of detecting the regime transition from bubbling to turbulent fluidization conditions using vibration signals.     

Effect of coal particle swarm properties on the fluidization characteristics and coal beneficiation in a dense-phase gas–solid fluidized bed

This paper analyzes the influence of different coal mass fraction in an air dense medium fluidized bed (ADMFB). The effect of the low density particles layer on heavy sedimentation increased with increasing material layer thickness. The thickness of the low density particles layer also affected the final settling time of the high density particles. Increasing the thickness of the low density particles layer by Δh provoked an increase in the settling of high density particles that was related to their diameter (Δh/D). The pressure gradient across the bed was lower than that observed for the control experiment, which had only the dense material, owing to a decrease in the pressure gradient in Zones 1 and 5 (at the top and bottom of the bed, respectively). Introducing different coal sizes resulted in different fluidization environments, particle accumulation layers, and changes to the surrounding zone. However, the influence of the coal particles on the local bed characteristics was related to its concentration. The feeding mass fraction of 6–13 mm size and 13–25 mm size coal should be limited to10% and 13%, respectively. The ranges of possible deviation were found to be 0.08–0.15 and 0.07–0.10 for the respective samples.     

Two and three dimensional modeling of fluidized bed with multiple jets in a DEM–CFD framework

Fluidized beds with multiple jets have widespread industrial applications. The objective of this paper is to investigate the jet interactions and hydrodynamics of a fluidized bed with multiple jets. Discrete element modeling coupled with in-house CFD code GenIDLEST has been used to simulate a bed with nine jets. The results are compared with published experiments. Mono dispersed particles of size 550 μm are used with 1.4 times the minimum fluidization velocity of the particles. Both two and three dimensional computations have been performed. To the best of our knowledge, the results presented in this paper are the first full 3D simulations of a fluidized bed performed with multiple jets. Discrepancies between the experiment and simulations are discussed in the context of the dimensionality of the simulations. The 2D solid fraction profile compares well with the experiment close to the distributor plate. At higher heights, the 2D simulation over-predicts the solid fraction profiles near the walls. The 3D simulation on the other hand is better able to capture the solid fraction profile higher up in the bed compared to that near the distributor plate. Similarly, the normalized particle velocities and the particle fluxes compare well with the experiment closer to the distributor plate for the 2D simulation and the freeboard for the 3D simulation, respectively. A lower expanded bed height is predicted in the 2D simulation compared to the 3D simulation and the experiment. The results obtained from DEM computations show that a 2D simulation can be used to capture essential jetting trends near the distributor plate regions, whereas a full scale 3D simulation is needed to capture the bubbles near the freeboard regions. These serve as validations for the experiment and help us understand the complex jet interaction and solid circulation patterns in a multiple jet fluidized bed system.