Most of my published work so far has been in the general area of computational fluid dynamics (CFD). More specifically, I was involved in simulating various fluid dynamics problems using the lattice Boltzmann method. I've dabbled in flow of molten polymers, transport of gases inside porous solid oxide fuel cell electrodes, suspension of solid particles in a fluid and simulation of turbulent flow.
The Google Scholar link to my work.
1 
Morphology Transitions in Multilayer Polymer Melts Due to Hole Growth and Layer Interaction Joshi, A. S. Clemson University, December 2004 
Note that articles may be subject to copyright. Download for personal use only.
16 
3D Porescale Resolved Model for Coupled Charge and Fluid Transport in a Vanadium Redox Battery Qiu, G., Joshi, A. S. , Dennison, C. R., Knehr, K. W., Kumbur, E. C. and Sun, Y. Electrochimica Acta, 64, 46  64 (2012) 

15 
Effect of relative humidity on contact angle and particle deposition morphology of an evaporating colloidal drop Chhasatia, V. H., Joshi, A. S. and Sun, Y. Applied Physics Letters , 97(23), 231909 (2010) 

14 
Direct Internal Reformation in the Solid Oxide Fuel Cell Anode: A PoreScale Lattice Boltzmann Study with Detailed Reaction Kinetics Grew K. N., Joshi, A. S. and Chiu, W. K. S. Fuel Cells, 10, 1143  1156 (2010) 

13 
Characterization and Analysis Methods for the Examination of the Heterogeneous Solid Oxide Fuel Cell Electrode Microstructure, Part 1: Volumetric Measurements of the Heterogeneous Structure Grew K. N., Peracchio A. A., Joshi, A. S. , Izzo J. R. Jr. and Chiu, W. K. S. Journal of Power Sources, 195(8), 2331  2345 (2010) 

12 
Wetting dynamics and particle deposition for an evaporating colloidal drop: a lattice Boltzmann study Joshi, A. S. and Sun, Y. Physical Review E, 82, 041401 (2010) 

11 
Numerical Simulation of Colloidal Drop Deposition Dynamics on Patterned Substrates for Printable Electronics Fabrication Joshi, A. S. and Sun, Y. IEEE Journal of Display Technology, 6, 579  585 (2010) 

10 
Erratum: Nondestructive Reconstruction and Analysis of SOFC Anodes using Xray Computed Tomography at sub50 nm Resolution [J. Electrochemical Soc., 155, B504 (2008)] Izzo, J. R. Jr., Joshi, A. S. , Grew, K. N., Chiu, W. K. S., Tkachuk, A., Wang, S. and Yun, W. Journal of the Electrochemical Society, 157(2), S5 (2010) 

9 
PoreScale Investigation of Mass Transport and Electrochemistry in a Solid Oxide Fuel Cell Anode Grew, K. N., Joshi, A. S., Peracchio, A. A. and Chiu, W. K. S. Journal of Power Sources195, 2331  2345 (2010) 

8 
Lattice Boltzmann Modeling of ThreeDimensional, Multicomponent Mass Diffusion in a Solid Oxide Fuel Cell Anode Joshi, A. S., Izzo J. R. Jr., Grew, K. N., Peracchio, A. A. and Chiu, W. K. S. Journal of Fuel Cell Science and Technology7(1), 011006 (2010) 

7 
Multiphase Lattice Boltzmann Method for Particle Suspensions Joshi, A. S. and Sun. Y. Physical Review E 79, 066703 

6 
Lattice Boltzmann method for Multicomponent Mass Transfer in a Solid Oxide Fuel Cell Anode with Heterogeneous Internal Reforming and Electrochemistry Chiu, W. K. S., Joshi, A. S. and Grew, K. N. European Physics Journal Special Topics 171, 159  165 (2009) 

5 
Nondestructive Reconstruction and Analysis of SOFC Anodes using Xray Computed Tomography at sub50 nm Resolution Izzo, J. R. Jr., Joshi, A. S. , Grew, K. N., Chiu, W. K. S., Tkachuk, A., Wang, S. and Yun, W. Journal of the Electrochemical Society 155(5), B504  B508 (2008) 

4 
Lattice Boltzmann Modeling of NonContinuum, MultiComponent Gas Transport Joshi, A. S., Peracchio, A. A., Grew, K. N. and Chiu, W. K. S. Journal of Physics D: Applied Physics 40, 7593  7600 (2007) 

3 
Lattice Boltzmann Method for Continuum, MultiComponent Mass Diffusion in Complex Geometries Joshi, A. S., Peracchio, A. A., Grew, K. N. and Chiu, W. K. S. Journal of Physics D: Applied Physics 40, 2961  2971 (2007) 

2 
Lattice Boltzmann Modeling of 2D Gas Transport in a Solid Oxide Fuel Cell Anode Joshi, A. S., Grew, K. N., Peracchio, A. A. and Chiu, W. K. S. Journal of Power Sources, 164, 631  638 (2007) 

1 
Computational Clarifications of Experimental Blend Morphology Transitions in Immiscible Polymer Melts Organized by Chaotic Advection Joshi, A. S. and Zumbrunnen, D. A. Chemical Engineering Communications, 193 (7), 765  781 (2006) 
20 
Development of a CADbased Cartesian Grid Generator for the Lattice Boltzmann Method Cantrell, J. N., Inclan, E. J., Joshi, A. S. , Popov, E. L. and Jain, P. K. American Nuclear Society Winter Meeting, 11  15 November, 2012, San Diego, CA 

19 
PRATHAM: Parallel Thermal Hydraulic Simulations using Advanced Mesoscopic Methods Joshi, A. S. , Mudrich, J. A., Popov, E. L. and Jain, P. K. American Nuclear Society Winter Meeting, 11  15 November, 2012, San Diego, CA 
A lava lamp is a decorative night lamp, invented by British accountant Edward CravenWalker in 1963. If you look at the picture above, you will see a blue colored glob of fluid rising up from the bottom because of buoyancy.
In some cases, the rising glob of fluid undergoes a beautiful transformation known as pinching. This involves the formation of a thin, necklike region and eventual breakup of the fluid fiber into a drop.
The physics of the necking and pinching process makes it impossible for the fluid to detach unless things are happening in 3D. In 2D, any such deformation would be opposed by interfacial forces. In 3D, there are two mutually perpendicular (opposing) curvatures and one of them can dominate over the other.
It is very easy to simulate this pinching process using a multiphase or multicomponent LBM. Several variations of this problem were used as important validations of the LBM model used in my thesis. One of the interesting ones was "end pinching" (see picture below):
Another useful problem for validation was the RayleighTomotika instability.
We now have the technology like xray tomography to accurately reconstruct porous media like rocks and battery electrodes. Features as small as micrometers can be explored and used in constructing mathematical models of the porous structure.
Based on an accurate representation of the pore structure, one can get a very accurate solution compared to volumeaveraged (macroscopic) approaches.
Using this idea, the LBM was applied to simulate multispecies gas transport in porous solid oxide fuel cell (SOFC) electrodes and flow of liquid electrolyte through a network of carbon fibers in a vanadium redox battery (VRB).
I'd like to thank Prof. Wilson K. S. Chiu (University of Connecticut) and Prof. Ying Sun (Drexel University) for their support in carrying out this work.
When a drop of coffee dries, a peculiar physical phenomenon occurs. The rate of evaporation of the solvent is very high at the perimeter of the drop, near the contact line. This causes flow of the solvent towards the perimeter, carrying with it the residual coffee particles. When all the solvent evaporates, what remains behind is a ring of coffee particles.
Although this is a very common observation, simulating this process at the scale of the individual solute particles has always been a challenge because of several poorly understood physical effects active near the contact line.
The lattice Boltzmann method (LBM) was extended to handle finitesized particles in a multiphase fluid to help improve our ability to analyze and study the effect of fluid and solid properties on the final particle deposition.
I'd like to thank Prof. Ying Sun (Drexel University) for her support in carrying out this work.
I was first introduced to CFD at IIT Bombay and was lucky to begin my journey with Prof. Kannan Iyer and Prof. Ghosh Moulic. My projects at IITB included numerical grid generation and code development for NavierStokes based CFD schemes.
It was a fantastic time and I worked on the SIMPLE algorithm and experimented with several versions of staggered and colocated schemes, including a version for general curvilinear coordinates.
After getting my Masters degree at IITB, I came to Clemson University to pursue doctoral studies. At Clemson, I joined a research group led by Prof. David Zumbrunnen. Dave introduced me to the lattice Boltzmann method (LBM) and encouraged me to use it in my research.
It is not easy to grasp the fundamentals of the LBM without having a firm background in statistical physics. I must admit that I do not fully understand several of its intricacies to this day. However, implementing the LBM is so seductively simple that I soon began to reproduce what I was doing before with NavierStokes based solvers.
Prof. Richard Miller, who served on my thesis committee, introduced me to the beauty of parallel computing using MPI. Combining LBM and MPI led to a very powerful and useful tool that I have applied to several problems since.
With the emerging trend of heterogeneous computing (CPU + GPU), I am convinced that LBM has an even brighter future.
LBM typically works with a uniform, Cartesian grid. Using it for flows in complex geometries requires the use of a preprocessor, which identifies the solid and fluid voxels in the 3D solution domain.
I participated in a recent initiative at Oak Ridge National Laboratory (ORNL) to simulate turbulent flow inside nuclear reactor geometries by developing a parallel LBM code called PRATHAM.
The flow geometry is often described by using STL files. We developed a light weight and highly parallel preprocessor called CARTGEN++, which can be used to obtain a voxelized representation of the complex geometry.
I would like to thank Dr. Prashant Jain and Dr. Emilian Popov for their support in carrying out this work.