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

Question: What is the large gauge transformations for higher p-form gauge field on a spatial d-dimensional torus $T^d$ or a generic (compact) manifold $M$? for p=1,2,3, etc or any other integers. Is there a homotopy group to label distinct classes of large gauge transformations for p-form gauge field on $d$-dimensional torus $T^d$ or any $M$ manifold ? (shall we assume the theory is a topological field theory, or not necessary?) References are welcome.


Background: Large gauge transformation has been of certain interests. The Wiki introduces it as

Given a topological space M, a topological group G and a principal G-bundle over M, a global section of that principal bundle is a gauge fixing and the process of replacing one section by another is a gauge transformation. If a gauge transformation isn't homotopic to the identity, it is called a large gauge transformation. In theoretical physics, M often is a manifold and G is a Lie group.

1-form: The well-known example is a connection $A$ as Lie algebra value 1-form. We have the finite gauge transformation. $$ A \to g(A+d)g^{-1} $$ An example of a large gauge transformation of a Schwarz-type Chern-Simons theory, $\int A \wedge dA$, on 2-dimensional $T^2$ torus of the size $L_1 \times L_2$ with spatial coordinates $(x_1,x_2)$ can be $g=\exp[i 2\pi(\frac{n_1 x_1}{L_1}+\frac{n_2 x_2}{L_2})]$. This way, for the constant gauge profile $(a_1(t),a_2(t))$ (constant respect to the space, satisfying EOM $dA=0$), the large gauge transformation identifies: $$ (a_1,a_2)\to (a_1,a_2)+2\pi (\frac{n_1}{L_1},\frac{n_2 }{L_2}) $$

This seems the two $\mathbb{Z}^2$ integer indices $(n_1,n_2)$ remind me the homotopy group: $\pi_1(T^2)=\pi_1(S^1\times S^1)=\mathbb{Z}^2$.

2-form: If we consider a 2-form $B$ field for a Schwarz-type TQFT, do we have the identification by $\pi_2(M)$ on the $M$ as the based manifold? (Note that $\pi_2(T^d)=0$ - ps. from the math fact that $\pi_2(G)=0$ for any compact connected Lie group $G$.) Is this the correct homotopy group description? How does large gauge transformation work on $T^d$ or $M$?

3-form: is there a homotopy group description on large gauge transformation? How does its large gauge transform on $T^d$ or $M$?

share|cite|improve this question

The mathematical framework that I am familiar with for abelian p-form gauge theory (the one promoted by Freed, Moore and others) is that of Cheeger-Simons differential forms. In this framework, the space of topologically trivial p-form gauge fields over a manifold $X$ (the analogue of 1-form gauge fields on the trivial $U(1)$ bundle) are identified with the quotient of $\Omega^p(X)$ by the group of gauge transformation $\Omega^p_{\mathbb{Z}}(X)$. $\Omega^p_{\mathbb{Z}}(X)$ denotes the set of $p$-forms with integer values when integrated along any closed $p$-dimensional sub-manifold $\Sigma \subset M$. In the $p = 1$ case, $\Omega^1_{\mathbb{Z}}(X)$ is precisely the space of all closed $1$-forms of the form $dg g^{-1}$ for some gauge transformation $g:X \to U(1)$. For more details, I recommend this paper by Belov and Moore, particularly p. 17-19.

In this framework, there is a notion of large gauge transformations for p-form gauge fields. In the 1-form case, the gauge transformations $g(x)$ that are not connected to the identity are precisely those $g(x)$ such that $dg g^{-1}$ has non-zero value when integrated around some closed, non-contractible loop $\gamma$. In the above language, we can say that the connected components of the space $\Omega^p_{\mathbb{Z}}(X)$ correspond to $H^p(X;\mathbb{Z})/\text{torsion}$ (via the map sending the $p$-forms to their de-Rham class) and that the large gauge transformations lie in the components which are not sent to $0$ by this map.

share|cite|improve this answer

Your Answer

 
discard

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.