Constraints on the Dirac spectrum from chiral symmetry restoration and the fate of U ​ ( 1 ) A \mathrm{U}(1)_{A} symmetry

* · 2024

Read Constraints on the Dirac spectrum from chiral symmetry restoration and the fate of U ​ ( 1 ) A \mathrm{U}(1)_{A} symmetry by * (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-02517.txt.

Introduction

When quark matter heats up, a broken symmetry mends, and whether a second symmetry mends with it hangs on the spectrum's fine shape. Matteo Giordano works through that logic in a 2024 paper. The setting is quantum chromodynamics with two light quark flavors near zero mass. Scalar and pseudoscalar responses carry the clues. A sharp peak of near-zero modes in the Dirac spectrum is the pivot.

If that peak grows in a set way, the second symmetry stays broken even after the first mends, he shows. The peak's size must track the topological charge spread, marking it as topology-born. The argument ties spectrum shape to symmetry fate directly.

Catalog record

CREATOR*
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE28 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-02517

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Constraints on the Dirac spectrum from chiral symmetry restoration and the fate of U ​ ( 1 ) A \mathrm{U}(1)_{A} symmetry * Abstract I discuss chiral symmetry restoration in the chiral limit m → 0 m\to 0 of QCD with two light quark flavours of mass m m , focussing on its consequences for scalar and pseudoscalar susceptibilities, and on the resulting constraints on the Dirac spectrum. I show that U ​ ( 1 ) A \mathrm{U}(1)_{A} symmetry remains broken in the SU ​ ( 2 ) A \mathrm{SU}(2)_{A} symmetric phase if the spectral density ρ ⁡ ( λ , m ) \rho(\lambda;m) develops a singular near-zero peak, tending to O ⁡ ( m 4 ) / λ O(m^{4})/\lambda in the chiral limit. Moreover, SU ​ ( 2 ) A \mathrm{SU}(2)_{A} restoration requires that the number of modes in the peak be proportional to the topological susceptibility, indicating that such a peak must be of topological origin. Matteo Giordano 1 Introduction In the two-flavour chiral limit where the up and down quark masses are sent to zero, QCD has a \mathrm{SU}(2)_{L}\times\mathrm{SU}(2)_{R} chiral symmetry, spontaneously broken to its diagonal component \mathrm{SU}(2)_{V} at low temperatures, as well as an anomalous \mathrm{U}(1)_{A} axial symmetry. While it is known that chiral symmetry gets restored at higher temperatures [ 1 ] , two important and related questions remain open: what is the nature of the transition, and what happens to \mathrm{U}(1)_{A} in the symmetric phase. In fact, whether \mathrm{U}(1)_{A} remains broken or gets effectively restored can affect the order/class of the transition [ 2 , 3 , 4 , 5 , 6 ] , and is still a matter of debate. A strategy to gain insight into this issue is to study how the Dirac spectrum responds to chiral symmetry restoration [ 7 , 8 , 9 , 10 ] , an approach recently revived in Ref. [ 11 ] on which this contribution is based. Eigenvalues and eigenvectors of the Dirac operator fully encode the dynamics of quarks interacting with the gluon fields, and so their behaviour should reflect the status of the various symmetries. One should then be able to derive constraints on the spectrum following from the restoration of \mathrm{SU}(2)_{L}\times\mathrm{SU}(2)_{R} , which in turn could shed light on the fate of \mathrm{U}(1)_{A} in the symmetric phase. The best known relation between chiral symmetry and the Dirac spectrum is certainly the Banks-Casher relation [ 12 ] between the chiral condensate, \Sigma , and the spectral density, \rho(\lambda;m) , \rho(\lambda;m)=\lim_{\mathrm{V}\to\infty}\frac{\mathrm{T}}{\mathrm{V}}\left\langle\sum_{n,\lambda_{n}\neq 0}\delta(\lambda-\lambda_{n})\right\rangle\,, (1) where m is the fermion mass, \lambda_{n} the Dirac eigenvalues in a gauge-field background, and \mathrm{T} and \mathrm{V} are the temperature and the spatial volume of the system, respectively. In the chiral limit the Banks-Casher relation reads |\Sigma|=\pi\rho(0^{+};0) , implying that a finite density of near-zero modes in the chiral limit produces a symmetry-breaking chiral condensate. At low temperatures one then expects a nonzero density of near-zero modes at zero and small quark mass – an expectation supported by ample numerical evidence (e.g., the recent Ref. [ 13 ] ). 1 1 1 For m\neq 0 , partially quenched chiral perturbation theory actually predicts a logarithmic divergence in \rho(\lambda;m) at \lambda=0 , proportional to |m|\ln|\lambda| [ 14 , 15 , 16 ] . However, this divergence is visible only for |\lambda|\lesssim e^{-1/|m|} , and disappears in the chiral limit. One would similarly expect that the density of near-zero modes vanishes at zero and small quark mass in the symmetric phase at high temperature, but numerical results indicate a completely different behaviour – a singular spectral peak near zero at finite mass [ 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 ] . The fate of this peak in the chiral limit is still unclear: it certainly must vanish in order for chiral symmetry to be restored, but how fast it has to do so, and what this could imply for \mathrm{U}(1)_{A} , are questions not discussed in the literature before Ref. [ 11 ] . Even if one could ignore the peak entirely, opposite conclusions about \mathrm{U}(1)_{A} can be reached depending on one’s assumptions. If one assumes that (mass-independent) observables are analytic in m^{2} – a reasonable assumption to make in the symmetric phase, then under plausible assumptions on the spectral density one concludes that \mathrm{U}(1)_{A} must be effectively restored in the symmetric phase [ 9 , 10 ] . If one assumes instead commutativity of the thermodynamic and chiral limits – an equally reasonable assumption in the symmetric phase, one concludes that most likely \mathrm{U}(1)_{A} remains effectively broken [ 33 , 34 ] . Finally, assuming both m^{2} -analyticity and commutativity of limits, one shows that \mathrm{U}(1)_{A} can only be broken by a Dirac delta at zero appearing in the spectral density already at finite mass, \rho(\lambda;m)\sim m^{2}\Delta\,\delta(\lambda) , which seems quite unlikely [ 35 ] . It is then important to understand which (if any) of these assumptions follows directly from chiral symmetry restoration. 2 Chiral symmetry restoration in the scalar and pseudoscalar sector In a local quantum field theory, restoration of a symmetry in the limit in which one removes all sources of explicit breaking requires that correlators of local operators that are related by a symmetry transformation become equal. The same is expected to apply to susceptibilities, i.e., to the (suitably normalised) spacetime integrals of connected correlation functions, since in the symmetric phase one generally does not find massless excitations and the correlation length of the system is finite. This applies in a general setting, and I will refer to it as “level 1” restoration. In the specific case of chiral symmetry in gauge theories, since gauge fields are unaffected by chiral transformations, it is reasonable to assume that the conditions above extend also to correlators involving generic functionals of the gauge fields only, including nonlocal ones (such as the spectral density), on top of operators built out of fermion fields. I will refer to this additional assumption as “level 2” restoration. I now use these basic symmetry-restoration conditions to characterise the restoration of chiral symmetry in the scalar and pseudoscalar sector of a gauge theory with two light (eventually massless) fermions of equal mass, and a rather arbitrary content in gauge fields and additional massive fermions. To have both a mathematically sound framework and a good notion of chiral symmetry, I put the theory on the lattice using Ginsparg-Wilson fermions [ 36 ] . A Ginsparg-Wilson Dirac operator obeys the relation \{D,\gamma_{5}\}=2D\gamma_{5}RD , with R local. Ginsparg-Wilson fermions have an exact \mathrm{SU}(2)_{L}\times\mathrm{SU}(2)_{R} chiral symmetry [ 37 ] , that allows one to define suitable fermion bilinears with simple transformation properties under chiral transformations. I restrict here to scalar and pseudoscalar bilinears, since the corresponding susceptibilities can be expressed in terms of Dirac eigenvalues only, thus providing the sought-after constraints on the spectrum. The relevant bilinears are \displaystyle S \displaystyle\equiv\bar{\psi}(1-DR)\psi\,, \displaystyle P \displaystyle\equiv\bar{\psi}(1-DR)\gamma_{5}\psi\,, (2) \displaystyle\vec{P} \displaystyle\equiv\bar{\psi}(1-DR)\vec{\sigma}\gamma_{5}\psi\,, \displaystyle\vec{S} \displaystyle\equiv\bar{\psi}(1-DR)\vec{\sigma}\psi\,, where \psi , \bar{\psi} denote the light fermion fields, and carry spacetime (including Dirac), colour, and flavour indices, suppressed here and in the following for notational simplicity.

Read the complete text