1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What mathematics do physics phd students learn?

  1. Jun 17, 2014 #1
    Hello again PF! I think I ask excessively often here so I tried my best to search in google. The result I got was this: http://superstringtheory.com/math/math1.html

    Well, that appears as all the math there is (no offense to mathematicians)! Likely the field of super string theory is the most math-intensive, but I would like to know what math courses a phd student would take in various fields of physics (particularly particle/nuclear physics, but I would also like to know others' as well). Would they compare to the maths an undergraduate math major learns?

    And few other questions regarding the same topic:
    1.) Although likely an extreme example, the amount of math needed as a physicist stated in the aforementioned website was astonishing. Of what classes is a phd student's semester typically consisted of?
    2.) If a student in his undergraduate years took/is educated in a math course one would take as a phd student, do you relearn it during your phd program?

    Thank you in advance and happy Tuesday
     
  2. jcsd
  3. Jun 17, 2014 #2
    That list seems ok for an undergraduate, but looking forward you may need to learn more mathematics depending on the field you choose. I'd add differential geometry, topology, functional analysis and abstract algebra (group theory and Lie algebra). That's what Roger Penrose puts in his book as requirements to learn theoretical physics. Particularly, learning differential manifolds will change your physical perspective in many ways, while giving you the background for general relativity.
     
  4. Jun 17, 2014 #3
    @Tosh5457: I think you forgot the next two pages which I was talking about. The list not only includes differential geometry, real and complex analysis, but also homology, fiber bundles, index theorems and bunch of other topics that seems rather overwhelming to even read their descriptions. Abstract algebra and topology are actually not in the list. May I ask their applications in physics?
     
  5. Jun 18, 2014 #4

    micromass

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    They are in the list, just not under that name. Abstract algebra is in the list under the guise of group theory. Although the group theory you need as a physicist is very different from what mathematicians see as abstract algebra.
    Topology is in the list under multiple items. For example, if you want to learn manifolds, homotopy, homology, then you need to know topology to some extent.
     
  6. Jun 18, 2014 #5

    WannabeNewton

    User Avatar
    Science Advisor


    If you want to do actual physics then focus on the physics. Youll pick up the math along the way. Time has to be spent wisely. Without stating your field of interest in physics one cannot make a remark as to how much math you would actually need to learn as theoretical physics can mean a lot of different things. The link you posted certainly does not apply to all areas of theoretical physics and one would be wasting their time trying to learn all that for any and all such areas.
     
  7. Jun 19, 2014 #6
    @WannabeNewton: Thank you for the reply, and as you said I will focus on physics. However as I'm still curious some of my questions have not been answered. If it helps, my primary field of interest is theoretical particle/nuclear physics. But I would like to know a general list of courses a phd student would take in a semester. For example, a phd student in his very first semester may take mechanics/electromagnetism, quantum field theory, abstract algebra, and spend other hours in research or teaching (then again this is a cluelessly hypothetical case).
     
  8. Jun 19, 2014 #7

    Matterwave

    User Avatar
    Science Advisor
    Gold Member

    A physics grad student would take whatever math classes he wished to take. But generally speaking, the only required one is probably a mathematical methods for physicists course that covers most of the basic necessary subjects of mathematics.

    For example, in my math methods course, we covered complex analysis (including complex integration, the Cauchy formula, method of residues, method of steepest descent, etc.), special functions and their generators (E.g. Hermite polynomials), Sturm-Liouville type differential equations, and Green's functions. More complex math can always be taken if the student so requires. And often a course that requires more advanced math will start off by teaching a little math first.
     
  9. Jun 19, 2014 #8

    AlephZero

    User Avatar
    Science Advisor
    Homework Helper

    Really? If I was astonished by that list, it would be because it was so little, not so much.

    There is nothing "advanced" on that web page. I work as an engineer, developing computational methods, and I use everything on that page (and a lot more) all the time.

    The answer to the question "what math does a PhD student need to learn?" is "whatever it takes to do the research". That's the basic difference between "research" and "solving the problems in a text book".
     
  10. Jun 19, 2014 #9

    WannabeNewton

    User Avatar
    Science Advisor

    You might want to take a look at the next page of the list.
     
  11. Jun 19, 2014 #10

    AlephZero

    User Avatar
    Science Advisor
    Homework Helper

    Fair comment. I didn't check that the OP had linked to the wrong page!
     
  12. Jun 19, 2014 #11
    My e&m professor is a nuclear physicist and he said he had never taken real or functional analysis.
     
  13. Jun 19, 2014 #12

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

    Other than a course in wavelet theory, my only math course during my graduate career was a two semester mathematical physics course out of Courant and Hilbert along with my teachers group theory notes (point and Lie groups)
     
  14. Jun 19, 2014 #13
    @Matterwave: Ok I'm starting to get a clear picture of how it works. Very useful information, thanks!

    @AlephZero: Well, it necessarily isn't the wrong page. But I owe you apologies for the confusion.

    @Hercuflea: Wow, I suppose nuclear physics isn't as math intensive as I thought. Thank you for the info!

    @Dr Transport: Then I suppose the need of math varies vastly upon different fields. May I ask what was your specialization?

    I'm starting to come to a conclusion that sophisticated math isn't required but useful if you know it. By sophisticated I mean anything above perhaps partial/ordinary differential equations and linear algebra. Any comment to this?
     
    Last edited: Jun 19, 2014
  15. Jun 20, 2014 #14
    It just depends on what you want to do, and even saying you need PDE's is ambiguous. You need to know separation of variables and a few other techniques, sure, but do you need to know the cauchy-kovalevskaya theorem? Most certainly don't.

    The bare minimum of what a PhD student needs to learn is likely to be covered in the departments mathematical methods course (or EM/mechanics if methods are taught there). Otherwise, it is too highly dependent on specialization to comment in meaningful detail.

    Crudely though... PDEs, ODEs, diff geo, group theory, asymptotic analysis but probably not anything that wouldn't be found in a math methods, GR or quantum text.
     
  16. Jun 20, 2014 #15
    That depends what you want to do. If you want to do experimental nuclear/particle physics, you will have to take QFT and particle physics classes, but that's it. It really won't matter if you understand the math behind them. But if you want to do theory, you will have to learn a lot of math to even master QFT, i.e, algebra, topology, etc. From personal experience, the amount of math you would need to understand those topics deeply would take you a couple of years (or maybe even more) to learn at the graduate level.
     
  17. Jun 20, 2014 #16

    WannabeNewton

    User Avatar
    Science Advisor

    What makes you think nuclear physics isnt math intensive? Just because it doesnt make use of marginalized esoteric math doesnt mean it isnt math intensive. You have a very wrong idea about how physics research or even physics grad school works. Seriously very few people have the time or motivation to learn all of that math. Not to mention unkess youre doing string theory or related subjects like topological field theory that math will be absolutely useless.

    The above poster is also mistaken. You certainly do not need algebra or topology to master qft and knowing these subjects wont help your understanding of qft at all beyond the basics of lie groups if what you care about is physics as opposed to mathematical physics. And if youre like me youll find math outside of a physics context very dry which makes the job of learning all that math even more unmotivated. There is a big difference between physics and mathematical physics. I can assure you that my past EM professor, who works in particle phenomenology, makes little use of pure math in his work and for good reason as it is often a hindrance.
     
  18. Jun 20, 2014 #17
    @WannabeNewton: Sorry if it sounded offensive in any way. I should have put the emphasis on "as," as I obviously had the wrong idea before that every physics discipline needed all the math there exists. I understand that physics and math are although to an extent related, very different matter. Although I may take a math course or two, my focus will always be physics.

    Thank you everyone for the reply!
     
  19. Jun 20, 2014 #18

    WannabeNewton

    User Avatar
    Science Advisor

    I didn't take any offense, sorry if it came off that way.
     
  20. Jun 20, 2014 #19
    There is a difference between theory and phenomenology. I stand by my argument that if you want to master QFT you MUST know topology and algebra beyond the usual group theory stuff, otherwise, you'd just be taking many things for granted, unlike quantum mechanics for example. To do the physics, you have to implement the math one way or another, and if you do not know the math behind QFT, then there is no way you are going to be able to do that.
     
  21. Jun 20, 2014 #20

    Dr Transport

    User Avatar
    Science Advisor
    Gold Member

    Semiconductor Theory, specifically Boltzmann Transport Theory and [itex] \vec{k} * \vec{p}
    [/itex] theory
     
    Last edited: Jun 20, 2014
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook