We derive a rigorous classification of topologically stable Fermi surfaces of non-interacting, discrete translation-invariant systems from electronic band theory, adiabatic evolution and their topological interpretations. For systems on an infinite crystal it is shown that there can only be topologically unstable Fermi surfaces. For systems on a half- space and with a gapped bulk, our derivation naturally yields a K -theory classification. Given the d − 1-dimensional surface Brillouin zone X s of a d-dimensional half-space, our result implies that different classes of globally stable Fermi surfaces belong in K −1 (Xs) for systems with only discrete translation-invariance. This result has a chiral anomaly inflow interpretation, as it reduces to the spectral flow for d = 2. Through equivariant homotopy methods we extend these results for symmetry classes AI, AII, C and D and discuss their corresponding anomaly inflow interpretation.
We propose that the Fermi surface anomaly of symmetry group
G in any dimension is universally classified by
G-symmetric interacting fermionic symmetry-protected topological (SPT) phases in
(0+1)-dimensional spacetime. The argument is based on the perspective that the gapless fermions on the Fermi surface can be viewed as the topological boundary modes of Chern insulators in the phase space (position-momentum space). Given the noncommutative nature of the phase-space coordinates, we show that the momentum space dimensions should be counted as negative dimensions for SPT classification purposes. Therefore, the classification of phase-space Chern insulators (or, more generally fermionic SPT phases) always reduces to a
(0+1)-dimensional problem, which can then be answered by the cobordism approach. In addition to the codimension-1 Fermi surface case, we also discuss the codimension-
p Fermi surface case briefly. We provide concrete examples to demonstrate the validity of our classification scheme, and make connections to the recent development of Fermi surface symmetric mass generation.
Effective gauge theories based on slave particle construction are widely used to describe quantum number fractionalization in strongly correlated electron systems. However, even setting aside the debates on the confinement issue of the slave particles, there are still significant conflicts between theory and experiment. In particular, a
T2/3 specific heat anomaly has been predicted to be the smoking-gun signature of low-lying gauge fluctuation in a
U(1) spin liquid with a large spinon Fermi surface, which has, however, never been observed. Here we show that such an anomaly is actually an artifact of a Gaussian approximation and is absent when the no-double-occupancy constraint on the slave particles is strictly enforced. We also show that projective construction based on slave particle representation provides a unified understanding of the mechanism of spin fractionalization in one- and two-dimensional spin liquids.