In Appendix B of QFT in a nutshell by Zee, a review of group theory is given. In the last paragraph of the appendix on page 533, he briefly discusses the helicity quantization of massless particles.

Firstly, he takes a rotation of $4\pi$ and considers possible helicities by using $exp(i4 \pi h) = 1$, where h is the helicity. Why he considers a rotation by $4\pi$ is not clear to me, and also why the path traced by this rotation can be shrunk to a point. Secondly, why is he not quantizing the helicity algebraically as conventionally done as in the case of angular momentum, but instead applying topological arguments. Is there a relation between the two?

  1. The path traced out by a rotation by $4\pi$ can be shrunk to a point in $\mathrm{SO}(3)$ because such a rotation corresponds to a closed curve in the double cover $\mathrm{SU}(2)\cong S^3$, which is simply connected and therefore any closed curve in it can be shrunk to a point. Curves which can be shrunk in the cover can also be shrunk in the base since covering maps are injective on the fundamental groups. Contrast this with a rotation by $2\pi$, which in $S^3$ is a path between two antipodal points, hence not closed and cannot be shrunk.

  2. Helicity is different from angular momentum because the angular momentum algebra is the Lie algebra of the massive little group $\mathrm{SO}(3)$, while helicity is for massless particles and hence must be related to the massless little group, which is a group variously denoted as $\mathrm{ISO}(2),\mathrm{SE}(2)$ or similar - it is the symmetry group of two-dimensional affine Euclidean space, hence consists essentially of $\mathrm{SO}(2)$ together with two-dimensional translations. Since the representations of translations are trivial, the question of the representations of the massless little group is reduced essentially to the representation theory of $\mathrm{SO}(2)\cong\mathrm{U}(1)$.

    This representation theory can be determined algebraically and one arrives at quantized values for helicity , but this cannot be done by the usual physicist's approach of representing the Lie algebra - the Lie algebra of $\mathrm{SO}(2)$ is simply the trivial Lie algebra $\mathbb{R}$, and there appear no quantization conditions in its representation theory - the quantization instead comes from the global structure of $\mathrm{SO}(2)\cong\mathrm{U}(1)\cong S^1$ and the fact that the possible unitary irreducible representations must be one-dimensional, hence completely classified by representation maps $\mathrm{U}(1)\to\mathrm{U}(1)$, since $\mathrm{U}(1)$ is the unitary group of a one-dimensional Hilbert space.

    It is straightforward to see that such maps are simply classified by how often they wind the $S^1$ around itself, i.e. by the integers. Zee's argument about the $4\pi$ rotation both sidesteps the need to do group representation theory instead of algebra representations and the need to determine that the representation classified by winding number 1 in this approach corresponds to the physical value of a helicity of 1/2.

  • $\begingroup$ A minor doubt, for massless particles where you consider the massless little group, why is Zee considering $4\pi$ rotations, instead of $2\pi$ rotations? Is it because $4\pi$ rotations can be shrunk to a point on $S_1$ while $2\pi$ won't? $\endgroup$ – Bruce Lee Nov 19 '17 at 4:33
  • $\begingroup$ @BruceLee The $4\pi$ rotation lives not in the little group, but still in the ordinary rotation group. From the physical definition of helicity, there is an axis such that a rotation by $\alpha$ acts on the state as $\mathrm{e}^{\mathrm{i}\alpha h}$, and we know that $\alpha = 4\pi$ needs to act as the identity. $\endgroup$ – ACuriousMind Nov 19 '17 at 11:41
  • $\begingroup$ That's still a little bit confusing. For massless particles, the little group is SO(2). Why do we still use the stuff (the 4pi rotation) from SO(3) which is the little group for massive particles? Shouldn't the discrete labels are completely determined by the little group? That is what Wigner tells us, right? $\endgroup$ – Wein Eld Apr 9 '18 at 9:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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