Media


Enjoy! But please be patient as some of the files are large ...


GoMRI Poster
An electronic version of Abhijit's poster on the effect of surfactants during the BP oil spill will be presented at the GoMRI Conference.




Bed expansion of 725 μm resin particles
Bed expansion of 725 μm resin particles with respect to time (Eulerian-Eulerian simulation)




Pressure distribution around settling sphere and bifurcation at Re = 210 (DNS simulation)
Pressure distribution around settling sphere and bifurcation at Re = 210 (DNS simulation)




Particle mixing in a bubbling fluidized bed
Particle mixing in a bubbling fluidized bed




Granular Rayleigh-Taylor instability during sedimentation of about 2 millions of particles
Granular Rayleigh-Taylor instability during sedimentation of about 2 millions of particles




DPM simulation of a cylindrical spouted bed
DPM simulation of a cylindrical spouted bed




Numerical simulations of a multiphase rotodynamic pump
Numerical simulations of a multiphase rotodynamic pump




Falling cloud DPM
A particle cloud (Re=872, St=0.814, Fr=11.26 and rhop/rhof=2.4233)




Multiphase Flow

Multiphase reactors are widely used in chemical and allied industries and the development of their design procedures is an important area of chemical engineering research. For reliable and optimum design, it is desirable to have complete understanding of the fluid mechanics and its role on the design parameters. During the past 30 years, efforts have been made to understand the fluid dynamics by using Computational Fluid Dynamics (CFD).


Effect of Surfactant on Oil Spill

Abhijit Rao says:

During the Deepwater Horizon oil spill, nearly 2.1 million gallons of chemical dispersant were used to disperse oil. Of this, approximately 0.77 million gallons of dispersant was injected near the point of oil release. Surfactants are the active ingredients in dispersants which cause reduction in interfacial tension resulting in the production of smaller oil droplets. Consequently, the oil droplets are finely dispersed in the water column and are transported to a farther location by the existing ocean currents.

We conduct droplet experiments to study the change in the behaviour of an oil droplet released through a nozzle into a quiescent water column, on addition of a surfactant. We are also developing a numerical model based on Volume of Fluid method to emulate the above experimental observations.

Computational Fluid Dynamics (CFD) of Particulate Flows

Rupesh Reddy says:

Fluid flow past solid bodies is a common phenomenon in nature. The motion of solid particles and the interaction between these particles in multi-particle systems is of great importance in areas such as sedimentation and fluidization (adsorption, leaching, particle classification and backwashing of down flow granular filters), slurry transport (water lubricated transport of heavy crude and coal slurries), hydraulic fracturing (oil and natural gas production). The flow complexities in these multi-particle systems have so far prevented the detailed understanding of the transport phenomena (momentum, heat and mass transfer) in the interstices between the particles. This important subject has become amenable due to an increase in the computational power and the parallel developments in numerical techniques.

This research focus is on CFD moedling of the particulate flows using different approaches:

Eulerian-Eulerian approach: Both phases are viewed as interpenetrating continua, as of there is no true particle surface and hydrodynamic force on the particle is not computed but described by a drag law.

Direct Numerical Simulation (DNS): In this method, the fluid flow is governed by the continuity and momentum equations whereas the particles are governed by the equations of motion for a rigid body. Flow field around each individual particle is resolved and the hydrodynamic force between the particle and the fluid is obtained from the solution itself and is not modeled by any drag law. Our group has the in-house DNS code based on the fictitious-domain finite element method.

Discrete particle modeling of dense dispersed flows

Richard Wu says:

Dispersed flows as they take place in chemical reactors consist of a large number of finite solid particles, gaseous bubbles or liquid drops that are convected by a laminar or turbulent flow of a continuous liquid and/or gas phase. For such dense systems, the interaction among the constituents of the dispersed phase, the one between the dispersed and the continuous phases, and other physical or chemical processes take place at different length and time scales. Due to this temporal and spatial multi-scale character of the fluid dynamics in chemical reactors, a single fully detailed model is not feasible for industrial-scale chemical reactors. The multi-scale approach for these complex systems is described by a hierarchy of models at different length scales, where the interactions between phases are taken into account with different levels of detail. The results and insights obtained from the more fundamental microscopic models are used to develop closure laws to serve the macro- and mesoscopic models which can be used to compute the physical processes on a much larger scale.

The discrete particle model (DPM) is a mesoscopic modeling technique for dense particulate flows between direct numerical simulation (DNS) and the Eulerian two-fluid model (TFM). In this approach, the particle motion is described in a Lagrangian framework by directly solving the Newtonian kinetic equations of each individual particle. The constitutive relations for the dispersed phase are not required because the particle-particle interactions are modeled through a two-variant collision strategy; the hard-sphere variant or the soft-sphere variant. Although the flow details around particles are not well determined, the particle motion is resolved at such a particle scale that many important features related to the particle motion in gas-solid fluidized beds are reasonably well captured. We have developed a robust and efficient 3D discrete particle model (3D-DPM) with a coupling to commercial CFD software Fluent 6.3. It has been shown that the model can capture many important characteristics such as bubbling, spouting, particle clustering and core-annulus flow structures in fluidization systems. The model is featured by the accurate void fraction calculation, flexibility to complex domain meshed by unstructured grids, an O(1) complexity collision handling algorithm and strongly implicit two-phase coupling. Recently we are focusing on the development of a parallel computing version of the model.

Numerical strategy and mechanism of flow in a multiphase rotodynamic pump

Zhiyi Yu says:

The cost of a multiphase transporting facility is only about 70% of a conventional separate transporting facility by eliminating the separation equipment and extra pipelines. Among the multiphase pumps developed in the petroleum industry, rotodynamic pumps are now widely accepted.

The configuration of such pumps is similar to the hybrid of a pump and a compressor. Each impeller delivers a pressure boost with an interstage diffuser to homogenize and redirect flow into the next impeller. This interstage mixing prevents the separation of the gas-liquid mixture and enables stable pressure-flow characteristics. It can also provide a large flow rate. As the impeller tip clearance is sufficient to allow small sand particles to pass, this kind of pump is applicable to the multiphase flow containing small sand particles, and the manufacturing requirements are not so strict.

On account of the great advantage to multiphase pumping, the features and cause of two-phase distribution in rotodynamic pumps have become a subject of great concern and attempts have been made to develop the numerical methods for modeling the internal gas-liquid two-phase flow in turbo machinery. In this research, an extended two-fluid model (in Fortran) and corresponding numerical strategy based on the Navier-Stokes equations are developed. The additional source terms arising from fluctuations of gas volume fraction (GVF) are considered. The discrete equations are solved using a developed two-phase SIMPLEC algorithm in body-fitted coordinates with the staggered grid system. The results illustrate that this code can solve the gas-liquid two-phase flow in the rotodynamic pump with the inlet GVF up to 25%.

The phenomena of phase separation and systematical instability will occur in higher inlet GVF. Therefore, further research is being carried on aimed to develop a stable transient numerical strategy to investigate the mechanism of gas blocking phenomenon in different working conditions.

Hydrodynamic instability of falling clouds of particles

Oladapo Ayeni says:

The hydrodynamic instability of falling clouds of particles has been mentioned as a legitimate class of Rayleigh-Taylor instability. My research is the study of torus formation, particle leakage, breakup and other phenomena that occur as a result of the gravity driven flow of particle clouds. The key questions I seek to answer are how the initial particle distribution, number of particles and critical dimensionless numbers that govern such flows-Reynolds, Stokes and Froude numbers- affect the dominant wave numbers of the resulting instability. Volume fraction and particle-particle collision effects have been ignored in previous studies of such flows and my research also intends to introduce these in the ongoing discussion. In addition, I am investigating the mechanism behind particle leakage at low Reynolds numbers. All simulations are done with FLUENT and an in-house Discrete Particle Modeling code.