I heard that anyons can only be made in 2 dimensions because when you visualize the spacetime diagram of a 2-dimensional system with point particles, you can get braids, but if you do the same with a 3-dimensional system, you get only the un-braid since all braids can be unbraided in 4 dimensions.

But we can make braided surfaces in 4-dimensions by moving around rings in 3 dimensions! So does that mean we get 3 dimensional anyons if there are rings that behave like point-particle anyons?

I tried googling the keywords, but got no answers. I did find a video talking about braided surfaces using rings (Braids in Higher Dimensions - Numberphile).


2 Answers 2


Indeed -- in theory, at least -- anyonic statistics does not so much require the ambient space to be 2-dimensional, as it requires the anyonic defects to have co-dimension 2 (hence dimension 2 less than that of the ambient space).

To give an impression, the following is a list of a few articles that consider 1-dimensional anyons (anyonic strings) in 3(+1) ambient dimensions (all in theory -- there is nothing yet on this in experiment/engineering, as far as I am aware, even though it is quite conceivable that anyonic string defects can exist in realistic materials):

  • J. Baez, D. Wise, A. Crans: "Exotic Statistics for Strings in 4d BF Theory", Adv. Theor. Math. Phys. 11 (2007) 707-749 (arXiv:gr-qc/0603085)

  • A. Bullivant, J. F. Martins, P. Martin: "Representations of the Loop Braid Group and Aharonov-Bohm like effects in discrete (3+1)-dimensional higher gauge theory", Advances in Theoretical and Mathematical Physics 23 7 (2019) (arXiv:1807.09551)

  • L. Kong, Y. Tian, Z.-H Zhang: "Defects in the 3-dimensional toric code model form a braided fusion 2-category", J. High Energ. Phys. 2020 78 (2020) (arXiv:2009.06564)

In mathematical terms, passage from point-like anyons in 2(+1) dimensions to string-like anyons in 3(+1) dimensions corresponds to passing from the braid group (and its representations) to the "loop braid group" (pointer to an nLab page that I once wrote). This might also be called the "string braid group".

Going further up to any ambient dimension, a "$p$-brane" in $(p+2)(+1)$ dimensions -- often called a "defect brane" -- is expected to exhibit anyonic statistics, see for instance p. 65 of:

  • J. de Boer, M. Shigemori: "Exotic Branes in String Theory", Physics Reports 532 (2013) 65-118 (arXiv:1209.6056)

I am taking the liberty of closing by mentioning that just today we presented a proof of this expectation (assuming common hypotheses on the general mathematical rules of brane charges):

  • H. Sati, U. Schreiber: "Anyonic Defect Branes and Conformal Blocks in Twisted Equivariant Differential (TED) K-theory" (arXiv:2203.11838, nLab)

In 3 dimensions, you can have both point particles and loops/rings/strings (all mean the same thing). There are several known braiding processes involving these two kinds of objects:

  1. A particle can move around a string.

  2. If there are two loops, one can "thread" one through the other to braid them. But this is actually not different from 1: the process can be smoothly deformed to the process 1 by first shrinking the loop to a point.

  3. More interestingly, imagine there are three loops, two of them are linked to the third. Then we can do braiding 2 to the the two loops. Because of the third loop, now the deformation from 2 to 1 is no longer possible: we can not shrink any of the linked loop to a point without hitting another one. So this is in some sense a new braiding for loops in 3D.

The three-loop braiding process has been studied quite extensively in the context of topological phases in 3+1. All of above are basically taken from the paper https://arxiv.org/abs/1403.7437.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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