Doing a PhD in both computational and experimental condensed matter physics?

[member 666466]

I am torn between computational and experimental condensed matter physics for my PhD. My focus is on low dimensional systems (e.g. electron correlation/transport, broken symmetry at the boundaries). I'm currently in the process of applying for graduate schools, and so far, I've chosen all my desired advisors based on the experimental side. However, I greatly enjoy scientific computing and would like to combine computational physics into my PhD if possible, although I will choose experimental over computational if I was forced to make the choice.

As for my background, I'm located in United States + looking to do PhD in United States, my bachelors was in Chemical Physics (graduated this past May) with an equivalent of a minor in pure math (my school doesn't do minors), spent my freshman summer and senior thesis in experimental chemistry/physics and spent my sophomore/junior working on computational chemistry/physics research with a middle authorship in a supercomputing (comp sci) paper. My experimental background is mostly in fabrication and spectroscopy of nanomaterials/nanostructures, and my computational background is in supercomputing, kinetic theory simulations, quantum chemistry, and software development. Since graduation, I've been working as a programmer developing a high fidelity optics simulation software for government. As such, my background is neatly torn in half into pure computational and pure experimental.

Also, another reason why I can't discard computational so easily is that I would like the option of going back into scientific software development in the future, perhaps in a quantum computing company since much of quantum computation hardware is based on low D physics. Even though I have a work experience writing scientific software, I think the companies would not want me if my PhD was just purely in experimental physics.

Is it possible to do a combined experimental/computational PhD, perhaps with two advisors (one experimentalist, one computationalist) if I can't find one that does both? My ideal PhD is one where I get to experiment to validate results from my first-principles computational predictions. And even if it is possible, is it wise to even try or should I just focus on one? I would love to do both, but I don't want to become a master of none.

 

[ZapperZ]

First of all, you are putting the cart way before the horse. While you are certainly asked to state your intended area of specialization when you apply to grad schools, you are not tied to it! No one will look at your application later on and tells you "Oh no, you can't join our group because your application says that you want to do experimental condensed matter physics in low-dimensional system!" So if you change your mind later on, and AFTER you pass your qualifier (remember that hurdle?), you can certainly jump into another field entirely.

Secondly, while you may want to specialize in one area, it doesn't mean that by the time you are ready, that is the one might be available. I am talking about the availability of research funds, etc. to support your as an RA. Often times, student goes into an area of study that has research money. I did such a thing when I was deciding what to go into after I passed my qualifier. I don't even remember what I specified when I applied for grad school, but I certainly took the first available RA support that wanted me. Unless you already have NSF or other financial support that are not tied to any particular research group, in many cases, your choices are often limited if you do not want to be saddled with huge tuition bills.

Zz.

 

[Dr. Courtney]

My PhD was pretty much experimental/computational atomic physics. But the original intent and adviser were very strong in experiment. Computational stuff just happened as I realized as we went along that I could compute lots of interesting stuff related to our experimental work. Ended up publishing 5-6 papers from the computational work. Only 3 papers from the experimental work, but two were PRLs, so they got a lot more love.

 

[member 666466]

@ZapperZ I'm currently applying for NSFGRF, NDSEG, NPSC, and DOD's SMART fellowships. I would love to apply for DOE's CSGF fellowship, but that is specifically for people who will do computational research, and I know that I would give up computational if I was forced to choose, so I'm not applying for that fellowship. Once I'm done with my coursework, I'm planning on also applying for an IBM and Microsoft PhD fellowships to fund my thesis research years. So, fingers crossed, hopefully I will get the freedom to choose my research field/advisor. And worst comes to worst, I can go back to my undergraduate lab since my advisor offered full RA-ship, which isn't really terrible since my undergrad is a pretty decent school and my advisor is really awesome (honestly I'm doubtful if I will ever find a better advisor), but yeah I know what you mean about the funding.

@Dr. Courtney How were you able to do computational research with an advisor who does mostly experiment? My undergraduate advisor has some pretty strong connections with most of the professors that I want to work for, through academic tree or collaborations, but all of them are pure experimentalists as far as I am aware.

 

[Dr. Courtney]

@Dr. Courtney How were you able to do computational research with an advisor who does mostly experiment? My undergraduate advisor has some pretty strong connections with most of the professors that I want to work for, through academic tree or collaborations, but all of them are pure experimentalists as far as I am aware.

The grad students on the project before me had written code for computing the outcome of many of the experiments. The code was passed along and maintained and used. Results were mentioned in a couple of experimental papers by way of comparison, but nothing ground breaking, and no pure computation papers. I inherited the code along with the experimental project. The experiment stalled a time or two due to technical difficulties at about the same time I realized some questions might be addressed computationally since the code had been experimentally validated for a range of systems. Theorists expressed an interest in various meetings, so I cranked up the code and began exploring new regions of parameter space. Along the way, I became aware of a few computation tricks and techniques to improve the package and make the computations more efficient, so I improved the code quite a bit along the way. I was the first grad student in the group whose computational effort was more than 10-20%. My adviser gave me the go ahead for everything I wanted to try since I was good at it, enthusiastic, and theorists were eager for the results of my calculations.

 

[DEvens]

Questions of this nature are often decided based on what a graduate adviser is prepared to have you do when you work for them. Your PhD is going to be, to some extent, a negotiation with a prof. The prof has to be willing to see you make the attempt under his banner, so to speak. If you fail it is to some degree a reflection on him. If you do a good job that is also a reflection on him.

So, if you have a potential supervisor in mind, you should be talking to them as soon as possible. Maybe they will be telling you that you would be a good fit, and have suggestions as to what you might be doing. Or maybe they will be telling you that your ideas won't fit with his research program. Or maybe he will make other suggestions you suddenly find very attractive.

Profs are very likely to be OK with potential PhD candidates contacting them. The very worst possible response you will get is no, we have as many PhD students as we can handle right now. You can also ask them what other profs might be a good fit. Maybe there are profs you don't know about and had not even considered.

So, Google up a few possibles. Check out the preprint archives for papers in your general area of interest and see which ones you could see yourself working on. Then contact the authors and inquire about grad schools they think might be a good fit for you.