About

I am an Assistant Professor of Physics at DigiPen Institute of Technology and my research focuses on the end stage of stellar evolution. I study the mass-loss process that occurs during the final stages of a star's life, simulating stellar winds in red supergiants (RSGs) and asymptotic giant branch (AGB) stars.

Marquis Who's Who Badge
The aim is to attempt to gain a better understanding of the complex mechanisms at work during this mass-loss process. In particular, I try to understand the interplay between stellar rotation, effects of stellar magnetic activity on the stellar atmosphere as well as dust formation and radiation pressure contributions to the stellar wind. This impacts the mass-loss of a variety of stars ranging from RSGs to AGBs to even Mira-type stars.

Another focus of my research has been in computational atomic structure in the strongest magnetic fields in the observable universe: those found in neutron stars and magnetized white dwarfs. My research here is aimed at developing fast and accurate atomic structure software, capable of execution in only a matter of seconds on run-of-the-mill computer architechture. I use such tools to take a peek at what atoms and molecules look like in intense magnetic fields.

A third and more recent focus of my research in understanding the process of star and planet formation. I am interested in understading the complex physical processes at work in the interiors of fragmented regions of molecular clouds that leads to ignition and stellar incipience, as it were. I am also interested similarly in the processes at work in the initial stages of the formation of rocky planets. Much of the early history of stars and planets eludes our understanding and my research in these fields aims to gain a better insight into these complex phenomena.

Available Projects

Atoms and Molecules in Strong Magnetic Fields

Theoretical, Computational and Observational • Suitable for graduate students

A large focus of my research deals with determining what happens to atoms and molecules in the extremely dense, hot and highly magnetised atmospheres of compact objects; these are magnetised white dwarfs and neutron stars. These are collectively the most highly magnetised objects in the known Universe - with magnetic fields that are billions of times greater than what we can ever achieve in the lab. Almost all the photons that we get from these unique astrophysical laboratories originate in their atmospheres, and in order to gain a good understanding of the nature of these objects, we need to be able to model their atmospheres with accuracy. A large part of this puzzle is interpreting their spectra properly. This requires a wealth of data for atomic and molecular electronic structure.

I have over half a dozen projects available for interested graduate students to work on in this area. There are projects available in both atomic as well as the more complicated molecular structure domains. A large portion of the work will also involve astronomical data from telescopes such as The Hubble Space Telescope. The aim is to explore this wholly uncharted area of computational astrophysics. I collaborate with researchers at the University of British Columbia, in Vancouver, Canada, and at the University of Oslo in Norway, as well as with researchers at the University of Mainz, in Germany, so there is the possibility for spawning further research directions with these international collaborators. If you are interested in working with me in this exciting area of physics, then get in touch with me.

Astrophysical Flows - Stellar Winds

Theoretical and Computational • Suitable for graduate students

I am interested in simulations of the prodigious mass-loss which occurs at the very last phase of evolution of the majority of the stars in the Universe, namely, as they end their lives on the Asymptotic Giant Branch (AGB). It has been observed that in the cool outer atmospheres of these dying stars, micron-sized dust grains can condense and grow. These form exiguous solar sails, and they absorb the momentum from the impinging stellar radiation from the interior. As a result of this, these dust grains move through the surrounding gas, and as they do so, they drag they gas along with them, and this results in a prodigious stellar wind. There are however several mysteries still. Approximately half the AGB stars in the Universe are oxygen-rich stars, and in these stars there is a huge mystery regarding how stellar material is transported from the stellar photosphere out to large distances where dust grains can form. I had proposed one such mechanism - a hybrid wind scenario, in which a magneto-centrifugal wind levitates material out to distances where dustr grains (silicates) can condense, and thereafter, the wind is a dual fluid wind, with the gas carrying the magnetic field, and the dust grains acting as solar sails and absorbing stellar radiation, and causing a large efflux. I am interested in developing 2D (or 3D) magnetohydrodynamic wind models for AGB stars, with dust grains in the flow, forming in-situ. This has never been done and would provide a great amount of insight regarding the question of whether magnetic fields in these stars (for which evidence is now starting to emerge) play a role in the mass-loss process, and if so, how does that ties in with observations of dust formation and a wind.

If you are interested in developing a dual-fluid hydrodynamics code and using it to look at the mysteries of how nearly half the stars in the Universe end their lives, then get in touch with me. I have several projects available for interested graduate students to work on in this area.

Rocky Planet Formation

Theoretical and Computational • Suitable for graduate and senior undergrad studentsp

I have recently become interested in exploring the question of how rocky planets form. This is of course tied to the question of how indeed planetisimals form in the first place - the precursors of planets. If we break this down even further, then it brings us to the mind-boggling question of how indeed do we take dust grains that can condense in protoplanetary disks, and convert them into planetesimals on timescales of about a hundred thousand years - a blink of an eye as far as a solar system is concerned. The mystery deepens when we think about the physics of dust coagulation and growth, and ask how do we go from millimetre-scale grains to kilometre-scale chunks relying only upon the physics of coagulation and sticking - which is an entire field of study unto itself. This is a complete mystery and an open question in astronomy. I am interested in a few related questions. First, carrying out simulations of in-situ dust grain formation and growth in protoplanetary disks. These can start out as semi-analytical models, and can get more advanced. I am also interested in asking the question - what size of objects can be made from simple particle-particle collisions and relying upon coagulation and sticking of dust grains, and third, what role does turbulence play in the protoplanetary disks, and do we need other mechanisms such as gravitational instabilities to aid in making the process rapid.

These exploratory questions are ideal for a graduate student or a senior undergraduate student to work on for their thesis projects. If this area of research and the big question of how rocky planets form excites you, then get in touch with me.

PSEUDASP

Code for intensely magnetised atoms

This is a 2D code in cylindrical coordinates for calculating the structure of atoms in strong and intense magnetic fields. It is based on a parallel pseudospectral implementation. The magnetic fields of strength that can be explored with this code are high B (> 100,000 T). It would be necessary to adaptively change the domain and mesh sizes for large fields as well as for mid-Z atoms.

The software is developed for 64-bit machines and utilises hyperthreading. It utilises the Armadillo C++ library and can also be compiled to utilise Intel MKL. It is highly recommended that you install the latest version of Armadillo C++ and include those libraries for compiling PSEUDASP. Minimum requirements for memory are about 8GB of RAM.

Please get in touch with me (see contact details below) for obtaining a suitable version of this code for your needs. It is requested that the author (Anand Thirumalai) be added as a co-author in any scientific articles that are the outcome of your research carried out with PSEUDASP.

PSEUDOGASP

GPU based code for intensely magnetised atoms

This is the GPU version of my 2D code in cylindrical coordinates for calculating the structure of atoms in strong and intense magnetic fields. It is written using CUDA to run on Nvidia GPUs. The purpose of the code is to explore atomic structure of mid- and high-Z atoms (e.g. S, P, Fe etc) in high fields without sacrificing accuracy and gaining a speed boost on the GPU.

The software is currently in active development and testing.

Please get in touch with me (see contact details below) if you would like to get involved in developing the software.

Recent Public Outreach

In The Moment Podcast (Seattle Town Hall, Ep. 118)

I recently had the privilege to read through an outstanding biography of the renowned scientist Vera Rubin, written by Ashley Jean Yeager.

I was invited by Seattle Town Hall to do a podcast to interview Ashley and talk about her debut book and about the astonishing and inspiring life of Vera Rubin.

See the following link for details: Podcast.

Contact Details

Anand Thirumalai
DigiPen Institute of Technology
9931 Willows Road, Redmond, WA, 98052
Ph: +1 (425) 629-4442
Email: