Physics 75: Thermodynamics


11/4/00 New homework added.

10/27/00 New homework added.

10/15/00 Added a few more papers to the list.

10/10/00 Added a few more papers to the list.

10/6/00 We've decided that the exam next Friday will be in-class, closed book. I'll provide a formulas sheet.

10/5/00 Some more papers added to the list.

10/4/00 To reiterate what we discussed at the beginning of class: I'd like you to decide what paper you'll discuss in your Wednesday talk by Monday in time for me to copy the paper and distribute it to the class on Monday (and earlier is even better!). Ideally, the presenter will work the details of the paper, delve into the references if necessary, and make explicit the ties to stat mech if they're not obvious. The rest of the class will read the paper with the aims of getting a basic understanding in order to be able to add discussion and ask questions. Hopefully our collective insights will add to our collective understanding.

There will be difficult points in all of these papers, and I don't expect that you'll be able to figure them out any of them entirely by yourself (although you should try as best you can). So, you should EXPECT to come to me to help elucidate the difficult points, which means you have to have looked at and thought about the paper in advance (and given me time to study it also). The 1-2 page outline is meant to force you to think in some detail in advance, and I want it by 5 on Tuesday to make sure that I've had a chance to read it and that you've had a chance to reflect on the material before you present it. Practically, , though, this deadline is too late to allow you to spend much time discussing difficult points, so I encourage you to get this out of the way much EARLIER.

You have (at least) two weeks between your talks; I suggest that you choose your next paper immediately after your talk read it (at least skim it) right away, and let me know what the paper is. In the next several days after that, write up your 1-2 page outline and collect together your questions on difficult points to discuss with me. We can resolve those, and once the physics is clear to you you can, at your leisure, think about which ideas to present to the class and how best to present them in the time available. As an assignment to do on Tuesday, reading and digesting a paper, writing two pages on it, and planning a presentation, is alot of work (even for me), and it probably won't be especially beneficial. However, it's not meant to be a one-day assignment; spread over two weeks, it's not bad, and in addition you have the opportunity to get help.

It might be helpful for me to explicitly state the goals of the exercise:

  • To allow the class to explore a wider variety of topics related to stat mech and thermodynamics that is convenient in the usual course format. Stat mech is more than just the mathematics, and as a discipline it's not very linear. Working statistical mechanicians often have a broad view of the applicability of their field to a range of problems, within and beyond traditional physics. This emphasis on the interdisciplinary aspect differentiates stat mech from many of the traditional subfields (within physics, much of dynamical systems theory also falls within the pale of stat mech). Workers spend considerable time looking looking at simple models, but also finding physical motivations/applications for their models.
  • To permit the reader to explore in greater depth some subject of interest related to stat mech and thermodynamics (including subjects relevant to your thesis). If you find a subject that interests you, you can do a series of related talks on the same subject. In the course of detailed study of even two or three papers (or book chapters, if you prefer) you can often penetrate pretty deeply into even a difficult subject. I'm happy to help you choose your readings to optimize your study or to help you on physics/technical material that goes beyond what we've covered in class.
  • To give the reader practice in making scientific presentations to colleagues, and to allow the rest of the class an explicit opportunity to reflect on what works and doesn't work in such presentations.
  • In the beginning I suggested a 10 minute time limit, but I've let you run as long as you chose. This was in part to let you see how it feels (and reduce the pressure of giving a talk a bit), and in part because a I realized 10 minutes was a miscalculation on my part: a typical conference talk is 10(-15) minutes, with several minutes at the end for questions. The audience usually saves its questions for the end. In our talks, the audience has asked more questions during the talk and fewer at the end. Since our setting is more informal than a conference talk and our goals more pedagogical, I'm happy to encourage the interactive discussion. Still, some structure is necessary, if only to ensure equitable distribution of time. So, let's say we cap the talks at a total of 20 minutes. You should plan on 10-15 minutes of talking (which by now you see isn't all that much), with 5 minutes factored in for questions throughout the talk. I'll time the talk and cut off at 20 minutes. If you're not used to giving a timed talk, practice in advance. I'll reserve a total of 10 minutes for myself, to make administrative remarks or closing comments for each talk. If you have the time, I encourage you to experiment a bit, with style or format (e.g. if you haven't given a talk with overheads, give it a try).

    This experiment might work well or poorly, depending on a variety of factors, and I'm still not certain that this is the best expenditure of time in an undergraduate class. A necessary condition for it to be useful at all is for us all to make a serious effort to understand the papers. Hopefully these changes in format will make the task easier, and as always I'll be grateful for your comments. I have yet to incorporate a formal, mandatory mechanism for explicit feedback on these talks, relying instead on the emails I've been asking you to send me evaluating your own and the other talks (which I receive only sporadically). A few of you have asked me what the rest of the class thinks of your talk, though, so I may try harder to squeeze more detailed explicit critiques from you. Your peers are certainly interested in what you think of their effort.

    10/2/00 Homework list updated. Exam #1 date changed to Friday, 10/13/00. Lecture list updated. Presentation schedule updated. Some new papers added.


    Course Information

    Course Description

    This course covers introductory thermodynamics and statistical mechanics, including:


    MWF 10-11, Th 11:30-12:30 in Rm. 209 Merrill Science Center


    Physics 35 or the equivalent, or my permission.

    Course requirements




    Required: Additional useful references (if the library doesn't have them, I'll try to get them):

    Lecture Schedule

    I'll try to post at least an outline of the lecture on the web, if I have time. This is more to motivate me to write them up that it is likely to be a useful resource for you... I'll also try to arrange some guest lectures from people that work in statistical mechanics or related field.

    A tentative lecture schedule will soon be available on the courseinfo website. However, since this is my first time teaching this course it will be very tentative, subject to change based on our interests and your background preparation.


    I'll keep scheduling information on this site primarily. I'm not yet used to Courseinfo, but as I get used to it I may post more stuff on there.

    Useful Links

    I'll post interesting or useful links pertinent to the course here as they I come across them. If you come across any others, please let me know. Most of the articles from American Journal of Physics are appropriate for class presentations. The ones from the Los Alamos archive are typically professional preprints and are sometimes quite long and/or heavy going.

    Presentation Schedule:

    I'll post the schedule below. If you can't talk on your designated day, please give a a little advanced warning so that we can reschedule. When you tell me what you'd like to speak on, I'll post it below (and a link to the paper, if possible). I'd suggest that the two that aren't speaking at least have a glance at the papers that the speakers will be presenting; it will make it easier for you to digest the talk and to take an interest in the presentation. Those that are presenting, I encourage you to discuss your paper with me, your classmates, other faculty, anyone else that might be interest. Again, you can choose your own paper (run it past me first) or I can suggest one for you. If it turns out that a paper you're interested in requires material beyond what we've covered in the course so far, I can give you a preview.

    Other Interesting talks in the Five-College area:

    Interesting and useful papers



    Applications of Thermodynamics

    Probability and Random Walks

    Statistical Mechanics

    Bose Einstein Condensation

    Other Applications of Statistical Mechanics

    Non-Equilibrium Statistical Mechanics

    Black holes

    Renormalization group

    Textbooks and Monographs: There are quite a few books on interesting stat-mech-related subjects that are at a level that's accessible to you. You could read a chapter or two and discuss. However, these are rarely self-contained, and it's difficult to get both a picture of what's interesting about the subject and some quantitative information by just reading a chapter. You'd probably want to choose a book of interest and then plan two or three related talks on subsequent chapters. You should check with me before pursuing these, though, since the substance, length, and difficulty of chapters can vary considerably within a text.

    Some longer research papers from the Los Alamos Archive (you can download a PDF version of the paper by clicking on Other Formats). These are too long to do as talks, although you could do a piece of one of them if one catches your eye. Not all of them will be accessible to you, but you can see what people do for research in these subfields.