Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. It's 100% free.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

QCD is the best-known example of theories with negtive beta function, i.e., coupling constant decreases when increasing energy scale. I have two questions about it:

(1) Are there other theories with this property? (non-Abelian gauge theory, principal chiral field, non-linear sigma model, Kondo effect, and ???)

(2) Are there any simple (maybe deep) reason why these theories are different from others? It seems that the non-linear constraint of the non-linear sigma model (and principal chiral model) is important, but I have no idea how to generalize this argument to other theories.

share|cite|improve this question
Do you want to consider (extended) SUSY or not? – Vibert May 25 '13 at 7:14

As gih correctly pointed out, there are a lot of QFTs in low dimensions which are asymptotically free.

In 3d, for example:

  1. Scalar $\phi^4$ field theory in 3d is asymptotically free. Renormalization flow increases the quartic coupling, taking you away from the free theory.
  2. Likewise, the 3d Gross-Neveu model (Dirac fermions with a 4-fermion interaction) is also asymptotically free. The 4-fermion interaction is relevant in 3d (rather than non-renormalizable, as it is in 4d).
  3. Plain old 3d Yang-Mills gauge theory in 3d is asymptotically free. (Even better, you can set things up so that the YM kinetic term itself is irrelevant, giving you a topological theory at low energies.)

It's not hard to combine these basic models to get more complicated but still asymptotically free models.

To address the 2nd part of your question: There's probably not anything model-specific going on here. It's just that the basic ingredients in these lower dimensional theories are better behaved.

share|cite|improve this answer
I disagree with you, there is a deep reason why one can find a RG trajectory in, for example, the $\phi^4$ theory in $d<4$ : it is because you assume that the flow start on a very specific trajectory that link the Gaussian fixed point and the Wilson-Fisher fixed point. If you pick an initial condition randomly, it won't be free in the UV. See also my answer here (and the reference) :… – Adam Jan 16 '14 at 17:04
@Adam I understood Tengen to be asking whether there were 'geometrical-ish' features of the models he mentioned which explained why the renormalization trajectories existed. You're right, however, that my second to last paragraph is confusing. – user1504 Jan 16 '14 at 20:13
Sure. I just wanted to stress that even the theory that we usually think as asymptotically free are like that only on very specific RG trajectories. – Adam Jan 16 '14 at 20:21

One example would be $\varphi^3$ theory, which is treated extensively in Srednicki.

share|cite|improve this answer
Is it artificial? Since there is no true ground state for $\phi^3$ thoery. The strong coupling property at low energy may indicates the collapse of perturbation theory. – Tengen Jan 24 '13 at 8:23
You are right, the theory is unstable. – Frederic Brünner Jan 24 '13 at 10:21
@Tengen: A purely $\phi^3$ theory is unstable, but one usually start with a stable theory (for instance, with both $\phi^3$ and $\phi^4$). But when one is using perturbation theory, the $\phi^4$ term is irrelevant compare to the $\phi^3$ and one can forget about it in the calculations. But of course, at the non-perturbative level (at the level of the functional integral), there are always terms that make the theory physical. – Adam Jan 16 '14 at 20:19

Yes, such examples are generic in 2+1 and 1+1D.

share|cite|improve this answer
Could you elaborate on this? You only answer the first part of the question. – Manishearth Dec 17 '12 at 8:50

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.