CFD-DEM Study Research Articles

CFD-DEM study on the blocking conformation of pre-filled sand control screen by layering method

Preventing screen blockage due to sand poses a significant challenge. The article combines computational fluid dynamics (CFD) and discrete element method (DEM) to establish a fully coupled four-way interaction model. A sophisticated size particle injection model was established, taking into account the particle size distribution (PSD) characteristics of the formation. A layering method was used to study the sand control behavior of a high-performance pre-filled screen. The study also quantitatively investigated the impact of factors such as median sand size (D50), screen structure, and fluid velocity on both wellbore-screen annulus and internal screen plugging patterns. The findings reveal that the sand retention rate within the wellbore-screen annulus exceeds 93.8 %. The risk of pre-filled screen blockage is minimized when the ratio of gravel diameter to sand diameter falls within the range of 3.3–5. At a gravel filling rate of 97 %, fine sand predominantly moves radially, resulting in a low risk of blockage at the screen's entrance layer, thereby favoring oil and gas production. In cases with elevated fine sand content, it is advisable to maintain a production flow rate below 3 m/s to mitigate localized erosion damage caused by increased pressure resulting from screen blockage. This model holds significant importance in optimizing completion tools and production parameters.

Expanding the Controllable Range of the Mean Residence Time Ratio of Polydisperse Particles in Multiple-Chamber Fluidized Beds: A Coarse-Grained CFD-DEM Study

Expanding the Controllable Range of the Mean Residence Time Ratio of Polydisperse Particles in Multiple-Chamber Fluidized Beds: A Coarse-Grained CFD-DEM Study

Investigation of the Multi-particle Arch Formation on the Single Slot of a Sand Filter: CFD–DEM Study in Packed-Bed of Sand Particles

Investigation of the Multi-particle Arch Formation on the Single Slot of a Sand Filter: CFD–DEM Study in Packed-Bed of Sand Particles

CFD-DEM study on the raceway transport phenomena and thermo-chemical behaviors in a three-dimensional blast furnace: Effects of process parameters

An in-depth exploration of the kinetic behaviors of the raceway can provide practical insights for optimizing blast furnace (BF) operations. In this study, the transport phenomena and thermo-chemical behaviors in the raceway of a three-dimensional BF were simulated from particulate scale by using the discrete element method-computational fluid dynamics (DEM-CFD) approach. The effects of process parameters (blast velocity, oxygen concentration, blast temperature, coke size and distribution, tuyere insertion depth, and tuyere inclination) on coke combustion characteristics (mass fraction distributions of gas species and reaction kinetics rate) and thermo-chemical behaviors (particle volume fraction, raceway size, mass loss, and coke temperature) were investigated. Meanwhile, microstructures of coke bed were analyzed, and correlations were established for raceway size prediction. The results indicate that the mass fraction distributions of O2 and CO are similar, both showing balloon-like shapes. The distributions of both the CO2 mass fraction and the kinetic rates of reactions exhibit annular shapes. The raceway size increases or even overlaps with increasing blast velocity/oxygen concentration/PSD standard deviation of coke and decreasing blast temperature/coke size. Too large or too small tuyere insertion depth or tuyere inclination is not conducive to the formation of the raceway. Besides, the mass loss, temperature, and microstructures also vary with the process parameters. RSM model can well predict the raceway depth, width, and height. This study extends the particle-scale model by considering transport phenomena towards the global perspective of a real BF, and the obtained new findings on the complex reacting behaviors in the raceway will provide better insights into the optimization of the BF process.

Gas and powder flow characteristics of packed bed: A two-way coupled CFD-DEM study

A two-way coupled Computational Fluid Dynamics (CFD) and Discrete Element Method (DEM) calculation was performed to numerically investigate the gas and powder flow characteristics within a packed bed. The analysis and discussion focused on the effects of gas phase velocity, particle shape, and size on the flow behavior of the powders. The results revealed that the average absolute velocity of the powder increased with an increase in the inlet gas flow rate. Conversely, the average dimensionless velocity exhibited the opposite trend. Particle shape significantly impacted the flowability of the powders. A critical value of sphericity equal to 0.96 was observed. For powders with sphericity less than 0.96, the average velocity fluctuated around a progressively decreasing value over time, and the powders tended to accumulate in the inlet region. In contrast, powders with a sphericity greater than 0.96 exhibited average velocities fluctuating around a constant value throughout the simulation. Additionally, a smaller number of these particles remained within the fluid domain, indicating a higher flow rate through the packed bed and ultimately exiting the fluid domain. Regarding the influence of particle size, the flowability reached a minimum value at a radius of 0.14 mm, with the average powder velocity mirroring this trend. Powders with a radius exceeding 0.14 mm demonstrated increasing flowability and average velocity. The observed impact on flowability was likely due to a combination of mechanisms. The findings presented in this paper offer valuable insights into the dynamics of powders within a packed bed. This knowledge can be instrumental in the design and analysis of packed bed reactors.

Segregation and mixing of binary mixtures of spherical particles in a bubbling fluidized bed

Abstract This work reports a CFD-DEM study on the segregation and mixing of binary mixtures of particles in a bubbling fluidized bed. A simplified mixing index was applied to determine the instantaneous mixing state in the bed, with which the effects of superficial gas velocity and initial packing state on the fluidization behavior were further discussed. For the well-mixed initial conditions, the mixing index decreases with fluidization time until a dynamic equilibrium between segregation and mixing is achieved. In contrast, the mixing index first increases and then decreases with fluidization time for completely-segregated initial conditions. However, the final equilibrium between segregation and mixing will not be affected by initial packing states for a given superficial gas velocity. Moreover, the bubbling behavior shows a marked impact on segregation and mixing, i.e., mixing is enhanced during the formation and grow-up of bubbles, while segregation is strengthened during the eruption of bubbles. This makes it possible to improve the fluidization of binary mixtures of particles based on the bubbling behaviors in the fluidized bed.

Binary Packing of Spherical Particles with Moderate Size Ratios in Viscous Fluid: A CFD-DEM Study

Binary Packing of Spherical Particles with Moderate Size Ratios in Viscous Fluid: A CFD-DEM Study

CFD-DEM investigation of the dispersion of elongated particles in the Turbuhaler® aerosol device

Elongated drug particles have the potential to be superior to spherical drug particles for delivery to the deep lung due to their increased ability to align with the airflow. However, the performance of elongated particles within dry powder inhalers (DPIs) has been largely unexplored. This paper presented a CFD-DEM study to simulate the dispersion of elongated active pharmaceutical ingredient (API) particles with equivalent sizes of 1.55–5.53 μm in a commercial DPI device Turbuhaler®. The model, which adopted a multi-sphere approach to mimic the shape of elongated spherocylindrical particles, was validated by comparing experimental data of fluidized beds. The model was then utilized to simulate the dispersion of powders with different aspect ratios in Turbuhaler. Findings revealed that the depositions of elongated particles in the chamber and the mouthpiece were significantly higher than those of spherical particles, and the depositions increased with higher aspect ratios. The higher deposition of the elongated particles was due to much stronger inter-particle cohesion when particles were in flat contact with particles or wall surfaces. Similar to spherical particles, both emitted dose and fine particle fraction (FPF, % mass of particles below 5 μm) of elongated particles increased with flow rates due to intensified particle-wall collisions. On the other hand, the FPF of elongated particles in the device decreased more significantly with increasing cohesion than spherical particles. Results indicated that while elongated particles are preferred for drug delivery to deeper lungs, their poor dispersion in inhalers due to higher deposition needs to be considered.

CFD-DEM study on agglomeration and spout-assisted fluidization of cohesive particles

The agglomeration of cohesive particles can deteriorate fluidization quality and cause the defluidization of a bed, which is a common issue found in the applications of fluidized beds. This study aims to gain a better understanding of particle cohesion on agglomeration/fluidization behaviors and the effective methods for achieving a better fluidization quality, through numerical simulations based on the coupled approach of computational fluid dynamics and discrete element method (CFD-DEM). The effects of particle cohesion, gas velocities or flow conditions, and the bed geometry on the agglomeration and fluidization behaviors are analyzed. It is shown that the increase of particle cohesion can lead to deteriorated particle mixing, significant agglomeration of particles, and defluidization of the bed; the agglomeration-induced defluidization of highly cohesive particles is difficult to mitigate in a conventional flat-bottom fluidized bed. As large-sized agglomerates are more frequently found in the bottom of the bed, the spouted gas flow is then utilized and demonstrated to be effective in assisting the deagglomeration and fluidization of highly cohesive particles. Through the comparison of various spouted beds and spouted fluidized beds, the effective design of the bed bottom is identified for achieving a higher fluidization quality. Corresponding mechanisms underlying spout-assisted deagglomeration and fluidization are found to be much related to not only the enhanced particle-fluid but also particle-wall interactions in the confined space of a conical bed bottom, thus explaining the effectiveness and the importance of the bottom conical geometry of spouted beds. The obtained findings may help to understand the agglomeration-induced defluidization of fluidized beds and assist the fluidization of highly cohesive particles by the effective design of spouted beds.

Study the effect of cohesion on fine particle transportation through a pore doublet model: A CFD-DEM study

This research delves into the transportation of micro-sized particles within a doublet pore model featuring two distinct pore throat sizes. To simulate a two-phase particle-fluid suspension, the CFD-DEM method is employed, combining the Navier-Stokes equation with Newton's second law of particle motion. Cohesion forces between particles and pore throats are characterized using the Simplified Johnson-Kendall-Roberts (SJKR) contact model, which is applied in our simulation to identify locations of agglomerations. Results reveal that altering cohesive energy density, from 10,000 Jm3 to 100,000 Jm3, significantly increases the likelihood of agglomeration in the smaller pore throat. It becomes evident that areas in close proximity to the throats are the most prone to particle agglomerations, leading to pressure drops over time. The formation of particle clusters within the pore throats intensifies with both an increase in the number of particles and fluid velocity from 0.001ms to 0.01ms. Nevertheless, low flow rates (0.001ms) are insufficient to mitigate blockages.

A CFD-DEM study of a two-joint spouting flow pattern in a multiple rectangular spout fluidized bed

Rectangular spouted beds were expected to solve the scale-up problem of spouted beds by simply adding chambers. In practice, it is common to insert baffles between different chambers to prevent the interaction of different spouts and keep the symmetrical spouting. However, the interaction of multiple spouts on hydrodynamics is not well understood. Therefore, the present study systematically explored the optimal gas-solid mixing performance with the integrated effect of two spouts in the system. The results show that the two spouts would join together and then deflect alternatively when the distance is within a wide range, specifically, 0.04 m ∼ 0.12 m out of a 0.3 m column width. This flow pattern is termed as two-joint spouting. The two-joint spouting demonstrated a significant advantage in both efficiency and product homogeneity for gas-solid mixing. Compared to conventional spouting, the average particle temperature drop after 20 s was 3.5 °C lower, accompanied by a notable reduction in the standard error of particle temperature, which was only half that observed in conventional cases. These results imply that the two-joint spouting mechanism contributes to a more efficacious heat transfer process and fosters enhanced uniformity in temperature distribution across the bed. This work firstly unveils that the two-joint spouting is favorable for gas-solid mixing and provides a new approach for the scale-up of rectangular spouted beds.

Super-quadric CFD-DEM study of spout deflection behaviour of non-spherical particles in a spout fluidized bed

Non-spherical particles are commonly encountered in chemical engineering processes including fluidization processes, yet their instability phenomena still lack understanding. In this study, the spout deflection – a typical instability phenomenon of non-spherical particles in a pseudo-two-dimensional (pseudo-2D) spout fluidized bed is first studied by a super-quadric computational fluid dynamics-discrete element method (super-quadric CFD-DEM) model. The intensity of spout deflection is quantified by the spout deflection angle concept. The effect of particle shape (i.e., sphere, cubic, ellipsoid, and cylinder) on the flow in terms of flow pattern, spout deflection, pressure drop, particle orientation, and normal contact force is systematically studied and compared by the time- and frequency-domain analysis. It shows that the spout channel becomes narrower for cubic, ellipsoid and cylinder particles where the particle aspect ratio increases. With the increase of spouting gas velocity, the median value of the spout deflection angle decreases and increases for cubic and ellipsoid particles, respectively. Furthermore, the pressure drops in the vertical and horizontal directions of the cubic, ellipsoid, and cylinder particles are different from each other. At last, In the bed of cubic particles, the ordered arrangements appear close to the wall region and random arrangements occur in the other regions. This work reveals the influences of non-spherical particles on spout deflection and offers insights into practical applications through optimal operation.

Effect of feed-size segregation on energy consumption during jigging: A CFD-DEM study

Segregation of feed based on size is used to increase the efficiency of gravity concentration by jig devices. Due to the variety of size and density of the particles in the feed, choosing how to segregate it before jigging is still a challenge. This article segregates the feed (with a size range of 3 to 8 mm) into the different states and then, simulates the jigging performance with the two-way coupling method of computational fluid dynamics and the discrete element method in 3 dimensions. In different states, the energy consumption and the number of cycles required for gravity concentration were compared, qualitatively. The simulation shows that the segregation of the jig feed into two size classes (3 to 4) and (5 to 8) mm decreases the energy consumption and the number of cycles. Therefore, the segregation state where the fine-grained part has a close size range and the coarse-grained part has a wide size range will have the lowest energy consumption and the number of cycles.

Solid-liquid rotary kilns: An experimental and CFD-DEM study

Experiments for the validation of unresolved CFD-DEM software for solid–liquid flows are often expensive, time consuming and generally provide little insight into the local particles dynamics. Additionally, several DEM parameters such as the particle surface properties are often obtained from experiments in air and used for all CFD-DEM simulations even when the fluid is a liquid. We design and perform a simple and time efficient gas–solid and liquid–solid rotary kiln benchmark for the purpose of creating a validation case for unresolved CFD-DEM software which we use to validate the unresolved CFD-DEM model of the open-source software Lethe. This case, which contains dense solid–solid contacts and strong solid–fluid forces gives insight on the importance of proper calibration of DEM surface properties in solid–liquid flows as well as on the importance of the lift force. Furthermore, it is highly sensitive to the accuracy of the CFD discretization.

CFD-DEM study of impacts of the porous distributor medium on fluidization characteristics of a 2D-fluidized bed

Gas–solid bubbling fluidized beds are widely used in chemical, energy, construction, and other industrial fields due to their excellent fluidization performance. 2D-fluidized beds with planar rectangular columns of finite thickness are widely used in fluidized bed studies, and it is essential for understanding the hydrodynamics of gas-particle systems. Moreover, the distributor (porous medium) at the bottom of the bed has a crucial influence on the fluidization performance of the 2D-fluidized bed. In this work, the fluidization mechanism and gas–solid dynamic characteristics of the fluidized bed under three porous media are studied by CFD–DEM coupled with the porous medium model, and the accuracy of numerical simulation is verified by a high-speed photography experiment. Results show that with the increase in the flow resistance of the porous medium, the average and standard deviation of the bubble diameter decrease, and the time and position of particles in the dead zone to participate in the core–annular flow advance. At a low fluidization velocity, the dead zone in the bed can be considerably reduced by increasing the flow resistance.

CFD DEM Study of Gas Solid Fluidized Bed for Non Spherical Particles

CFD DEM Study of Gas Solid Fluidized Bed for Non Spherical Particles

CFD-DEM study of spout deflection behavior of cohesive particles in a spout fluidized bed

The spout deflection problem limits the operating range and poses a threat to the normal production of fluidized beds especially when the cohesive particles are practised, yet these processes were not well understood at the particle scale in the past. In this work, the spout deflection behavior of cohesive particles in a pseudo-2D spout fluidized bed is studied by Computational Fluid Dynamics - Discrete Element Method featuring a cohesion sub-model. The model is validated against experimental measurements of two cases. Then three types of particles are studied including dry particles and two types of cohesive particles covered by water and honey, respectively, for comparison in terms of flow pattern, spout deflection, and bed pressure drop. The simulation results show that compared with the case of dry particles, the cases of cohesive particles have smaller spout deflection, 16.3% smaller for water-covered particles and 68.1% smaller for honey-covered particles in their base cases. Further, the effects of key parameters are studied. The median value of the spout deflection angle for dry and cohesive particles all increases with increasing spouting gas velocity. The median values of the spout deflection angle of cohesive particles are significantly affected by initial particle surface liquid content. At last, the gas pressure drops in the vertical direction caused by the particle resistance are almost the same between the cases of dry and cohesive particles; however, the gas diffusion in the horizontal direction is highly correlated with the initial particle surface liquid content, liquid types, and spouting gas velocity. This work unveils the impacts of cohesive particles on spout deflection and sheds light on the optimal operation for practical applications.

CFD-DEM Study of heat and mass transfer of ellipsoidal particles in fluidized bed dryers

In this paper, a 2D computational fluid dynamics (CFD) approach combined with 3D discrete element method (DEM) is extended by incorporating heat, mass transfers, and drying kinetic models to study the drying of ellipsoidal particles with equivalent volumes in fluidized beds. The validity of the proposed model is tested in terms of hydrodynamic and mass transfer behaviors. It is shown that the simulation results are in well-agreement with existing correlation or data. Finally, the effects of aspect ratio on heat and mass transfer behaviors at the same superficial gas velocity are then analyzed to improve the fundamental understanding of the system. It is demonstrated that the spherical particle has the highest drying rate compared to other ellipsoidal particles due to its highest average Reynold number of individual particles. The obtained results should be useful to the development of the fundamental understanding and optimization of fluidized bed dryers.

Effect of operational and geometric parameters on the hydrodynamics of a Wurster coater: A CFD-DEM study

A coupled CFD-DEM method was used to study the hydrodynamics of a Wurster coater. Firstly, the CFD part of the model was validated by the accurate prediction of the pressure drops over a pseudo-2D fluidized bed under various gas velocities. The effect of gas velocity, gap height, tube length and batch volume of the particles on the cycle time and the residence time of the particles was thoroughly investigated. The central jet gas velocity u1 was found to speed up the particle cycle but undermine the coating efficiency. The gas velocity at the horizontal transport zone u2 was able to promote the horizontal transport of the particles but should not be too high, otherwise, it would obstruct the normal falling back of the particles in the downward zone. Big gap heights would decrease the coating efficiency but tube length had little impact on that. The increment of batch volume would commonly abase the cycle time and the working efficiency under a given u1. The de-fluidization problem arose when the batch volume increased to 550 mL. However, this problem could be swept out by the optimization of u1 and u2. In a mixture of different sizes, the coarse particles enjoyed higher coating efficiency and could travel closer to the nozzles. This may shield the fine particles from getting enough coating liquids, and thus coarse particles and fine particles were not recommended to get coated in the same batch.

CFD-DEM study of reactive gas-solid flows with cohesive particles in a high temperature polymerization fluidized bed

The cohesive particles are commonly encountered in high temperature polymerization fluidized bed reactors. However, detailed studies on the effect of cohesiveness on the hydrodynamics, heat transfer, and reaction characteristics in such reactors have been rarely reported. In this work, a CFD-DEM approach coupled with solid bridge force model and polymerization reaction kinetics model is established to simulate the reactive gas–solid flows with cohesive particles. The results demonstrate that cohesive particle agglomeration significantly increases the inhomogeneity of concentration and velocity fields, and hinders particle circulation. The hindered particle circulation further results in pronounced temperature inhomogeneity. Moreover, the local hot spots are readily formed in the agglomerates, which expand and deteriorate rapidly if not controlled promptly. Besides, how the cohesion model parameters, namely contact time and maximum cohesive force limitation, affect the onset temperature and the severity of particle agglomeration is revealed. The established model extends the particle-scale study of cohesive gas–solid flows to reactive systems.