Physics self teaching (curriculum and textbooks)

[atyy]

BTW, Schwartz does have a free and short 200+ page version :) There are two nice chapters on renormalization, especially the second one (chapter 23), although I find the presentation of canonical quantization in the early chapters a bit weird: http://isites.harvard.edu/fs/docs/icb.topic521209.files/QFT-Schwartz.pdf.

I very much like the presentation of Wilsonian renormalization in Kardar's notes, especially the overview in L7: http://www.mitocw.espol.edu.ec/cour...-physics-of-fields-spring-2008/lecture-notes/. A very important point mentioned by Kardar is that the coarse graining generates higher order terms, even if we originally left them out. There is a complementary discussion in Bilal's QFT notes (section 4.5): http://www.solvayinstitutes.be/pdf/doctoral/Adel_Bilal2014.pdf.

However, Kardar is talking about statistical field theory, not QFT. Fortunately, we can get to statistical field theory from canonical QFT via Feynman's path integral and imaginary time. But imaginary time is very formal trick and not obviously physical, so we need the Osterwalder-Schrader axioms that tell us it is actually ok, and we can recover a quantum theory from statistical field theories satisfying some conditions.

The most elementary site I know that mentions Osterwalder and Schrader's work is: http://www.einstein-online.info/spotlights/path_integrals.

Another popsci level mention of the Osterwalder-Schrader axioms is in the book "From perturbative to constructive Renormalization: http://www.rivasseau.com/3.html (it's actually a serious book, but it's so well written that even non-rigourous non-professional people like me get something out of it). Rivasseau's book is also very informative for seeing how well the rigourous constructive viewpoint goes together with the very physical picture of renormalization that Wilson gave us.

 

[almarpa]

Whoa!

All that sounds like chinese to me. I hope one day I will be able to have such discussions in this forum. In the meanwhile, I take note that Schwartz is one of the books to keep in mind when I get to QFT.

Regards.

 

[vanhees71]

I don't see so much controversy with atyy. The problem with QFT is that it is mathematically still not well defined. The rigorous branch is something more for mathematicians who like to work on the problem to establish a rigorous mathematical formalism for QFT. For sure it's not a good idea to start learning QFT from such books.

The best books for physicists are Weinbergs first 2 volumes and if you like to understand also SUSY also the 3rd volume. However, it's not for the beginner either. In my opinion, what you should learn first is to do perturbative calculations in QED, including the one-loop diagrams to learn renormalization. Then you are well prepared to understand Wilson's ideas about renormalization. This can also be enhanced by looking at the many-body QFT (for equilibrium in the Matsubara imaginary-time formlism first). For this purpose, up to now, I think Peskin-Schroeder was the best book, but with the mentioned caveats concerning the many typos and some conceptional glitches. Schwartz is a brand-new book, and I've only read some chapters yet to see, how well he does, and I think there are much less conceptional glitches in there than in Peskin-Schroeder but also providing the calculational tools to handle perturbative QFT calculations. After that you should be ready to read Weinberg's books or delve into the attempts of mathemical rigor, if you like.

 

[atyy]

Whoa!

All that sounds like chinese to me. I hope one day I will be able to have such discussions in this forum. In the meanwhile, I take note that Schwartz is one of the books to keep in mind when I get to QFT..

Or more importantly, keep in mind that you should learn the Wilsonian picture of renormalization, and that it came from the Kadanoff-Wilson picture of critical phenomena in classical statistical mechanics. Weinberg, Peskin and Schroeder, Srednicki, Schwartz and Zee all explain the Wilsonian viewpoint (I personally like Srednicki's). But they put it very far back in their books. The reason is that the Wilsonian viewpoint does not change any calculation, but it is a conceptual brealthrough, that tells us that renormalization is not a mysterious process of subtracting infinities from infinities. It also allows us to make sense of non-renormalizable theories like gravity, provided the we only look at low energies. However, it is very tedious to do the calculation the physical Wilsonian way, and the traditional renormalization procedure is much faster.

http://quantumfrontiers.com/2013/06/18/we-are-all-wilsonians-now/

 

[almarpa]

Nice discussion about QFT textbooks.

Any other discussion about other subjects books would be appreciated.

Regards.

 

[mettw]

Hi Almarpa,
I'm in a similar situation to yourself. I have a BSc in physics and after twenty years am now planning on going back and getting a Grad. Dip. So I have been reviewing all of my undergraduate courses.

1) Classical Mechanics: Taylor + Kleppner and Kolenkow + Fowles&Cassiday (this last one for slightly more advanced topics)

I have found Taylor's chattiness a little irritating, but it definitely covers everything clearly.

4) Quantum mechanics: I think I will start with Griffiths, as a gentle introductin to the subject, and then continue with Shankar or Zettili.

Eisberg, R., Resnick, R. "Quantum Physics of Atoms, Molecules, Solids, Nuclei and Particles", Wiley.

This is an excellent book that covers a lot of topics. The philosophy is also solid Copenhagen, unlike Griffiths which, in the little space given to the philosophy of QM, delivers the sort of ignorant tripe you find in popular science books. The only downside to Eisberg and Resnick is that it doesn't cover Bra-Ket notation which Griffiths does quite well, along with an explanation of Hilbert space etc.. I own both of them, but if I had access to a university library I only would have bought Eisberg and Resnick, simply because of the range of topics covered. Also, beware that Griffiths jumps straight in at the Schroedinger equation while Eisberg and Resnick starts with 4 chapters of foundations and the old quantum theory.

BTW, what maths texts are you looking at? I've just bought whatever the University of NSW, my Alma Mater, recommended.

 

[porcupine137]

I still think that Weinberg would be an awfully tricky place for someone start out learning QFT. I agree with that literature review on that. OTOH, I also think that Zee's QFT is a pretty tricky place to start from as well.

I still think a mix of Srednicki QFT, Schwartz QFT and the Standard Model, Student Friendly QFT, Hatfield QFT,Peskin QFT would be the best start.

The lit review guy doesn't seem to know about any of those books other than the last one. Srednicki's QFT can form a really good basis for a start.
He based it on his teaching QFT at U.C.S.B. I believe.

here is a reddit thread that I just discovered that brings up various books:
http://www.reddit.com/r/Physics/comments/1468eu/qft_text_for_selfstudy/

 

[almarpa]

BTW, what maths texts are you looking at? I've just bought whatever the University of NSW, my Alma Mater, recommended.

I am eventually using Mary Boas book as a mathematics resource whenever I need to refresh something that I have forgotten.

Regarding QM, I think I will use Griffiths book, but not from the beginning. I will use a different book to study topics such us blackbody radiation, photoelectric effect, and all those QM foundation topics. What I still do not know is what book will I use for that.

Regards.

 

[Shahi]

4) Quantum mechanics.
I have never studied QM, so Griffiths book on QM seems a good option for a introductory course (as I said, Iam really enjoyng his EM book). However, many people warn agaist this book. Other good choices should be Zettili's or Shankar's books, but I am not sure. Your opinion will be really wellcomed to take a decision.

5

Thanks all of you.

Best regards.

Alejandro.

PS: See you in the threads in this forum.

Hi
I think you missed a step, you should consider a good deal of Modern physics course , because without it you couldn't understand why the physics needs to change at the beginning of 20th century

So before any move forward ,, I think it's very necessary to read a Modern physics book , my advice to you is Robert Eisberg : quantum physics of atoms and molecules