Functional Renormalization Group meets Computational Fluid Dynamics: RG flows in a multi-dimensional field space

Niklas Zorbach, Adrian Koenigstein, Jens Braun · 2024

Read Functional Renormalization Group meets Computational Fluid Dynamics: RG flows in a multi-dimensional field space by Niklas Zorbach, Adrian Koenigstein, Jens Braun (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-2412-16053.txt.

Introduction

Niklas Zorbach, Adrian Koenigstein, and Jens Braun borrow fluid solvers for quantum field theory in a 2024 paper. The functional renormalization group tracks how a theory's effective potential flows across scales. In multi field models those flow equations grow hard to solve. The team recast them as flow equations of a fluid kind, with drift and spreading terms.

A modern finite volume scheme from fluid dynamics then does the solving. Benchmark tests on zero dimensional models with two field directions check the method. The trick ports decades of fluid solver craft into field theory. It may steady flow studies of models, like those for fermions, where older grids buckle.

Catalog record

CREATORNiklas Zorbach, Adrian Koenigstein, Jens Braun
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE143 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-16053

Browse the full library · About the Tower of Babel

Full text

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

Functional Renormalization Group meets Computational Fluid Dynamics: RG flows in a multi-dimensional field space Niklas Zorbach, Adrian Koenigstein, Jens Braun Abstract Within the Functional Renormalisation Group (FRG) approach, we present a fluid-dynamical approach to solving flow equations for models living in a multi-dimensional field space. To this end, the underlying exact flow equation of the effective potential is reformulated as a set of nonlinear advection-diffusion-type equations which can be solved using the Kurganov-Tadmor central scheme, a modern finite-volume discretization from computational fluid dynamics (CFD). We demonstrate the effectiveness of our approach by performing explicit benchmark tests using zero-dimensional models with two discretized field space directions or two symmetry invariants. Our techniques can be directly applied to flow equations of effective potentials of general (fermion-)boson systems with multiple invariants or condensates, as we also demonstrate for two concrete examples in three spacetime dimensions. I Introduction Within the last decades lattice Monte-Carlo simulations turned out to be one of the most successful methods to study nonperturbative problems in Quantum Field Theory (QFT) , such as Quantum Chromodynamics (QCD) . However, also Monte-Carlo simulations come with limitations. Especially in the context of critical phenomena, the study of phase transitions, and calculations at moderate and high densities, other nonperturbative approaches are required to complement lattice Monte-Carlo simulations. One of these methods is the Functional Renormalization Group (FRG) approach which is based on a flow equation for the quantum effective action, the Wetterich equation Wetterich:1992yh . The formal derivation of this equation immediately triggered further field-theoretic developments Ellwanger:1993mw ; Morris:1993qb ; Morris:1994ie . By now, this approach has been successfully applied to a wide range of problems in high-energy, condensed matter, statistical, and gravitational physics Wetterich:2001kra ; Berges:2000ew ; Pawlowski:2005xe ; Gies:2006wv ; Kopietz:2010zz ; Braun:2011pp ; Dupuis:2020fhh . One of the main advantages of the FRG is that it provides direct access to the effective action, which is directly linked to the vertex functions and thermodynamic observables of a system. A particularly important quantity is the effective potential, which represents the lowest-order contribution of an expansion of the effective action in derivatives of the fields and comprises all orders of local (point-like) interactions of the (effective) degrees of freedom of the system. Usually, within FRG calculations, the effective potential is calculated by taking advantage of the symmetries of the system in field space. Hence, it oftentimes suffices to solve the Renormalization Group (RG) flow equations for the effective potential as a function of a single field variable or field invariant. However, in many cases the effective potential exhibits a lower degree of symmetry and the RG flow equations have to be solved in a multi-dimensional field space. Typical situations arise in systems with multiple order parameters, condensates, and/or field invariants. Selected relevant examples from high-energy physics that have already been addressed with the FRG are mesonic models with chiral and diquark condensates Strodthoff:2011tz ; Strodthoff:2013cua ; Khan:2015puu ; Lakaschus2021 , models that comprise strange quarks and their condensation in addition to light quarks Schaefer:2008hk ; Mitter:2013fxa ; Otto:2019zjy ; Otto:2020hoz , models that allow for pion and sigma condensation Kamikado:2012bt and many more. Also in the context of condensed-matter physics, the effective potentials of many systems cannot be reduced to a single field-space direction. Some examples from FRG studies are systems that exhibit inverted phase diagrams and exhibit the Pomeranchuk effect Hawashin:2024dpp or (frustrated) magnetic systems Tissier:1999hv ; Tissier:2000tz ; Tissier:2001de ; Tissier:2001uk ; Delamotte:2003dw ; Delamotte:2003dw ; Delamotte:2015zgf ; Debelhoir:2016yfp ; Yabunaka:2024tmc ; Sanchez-Villalobos:2024vmd . Other examples are models for (2+1) -dimensional Dirac materials such as graphene Classen:2015mar ; Torres:2018jij . All of them can be approximately described via O(N)\times O(M) -symmetric models or in general models with two invariants. However, it should also be mentioned that probably the first FRG calculations with several invariants originated from the field of statistical physics, in particular studies of matrix models should be mentioned here Berges:1996ja , see also Refs. Eichhorn:2013zza ; Eichhorn:2016hdi ; Eichhorn:2018phj for a more recent analysis. I.1 Motivation and contextualization Given these many examples, the reader might object that the problem of solving FRG flow equations in multi-dimensional field space is already solved and it seems unnecessary to address this problem in a separate work again. Indeed, several methods have been developed to study the RG flow equations in multi-dimensional field space: The most common approach is to use a multi-dimensional Taylor expansion of the effective potential around a fixed or moving point in field space which is usually chosen to be the global infrared (IR) or flowing minimum of the effective potential, respectively. Here, one effectively reduces the problem to a set of coupled ordinary differential equations (ODEs) for the Taylor coefficients. Another approach is to use a grid-based method where the field space is discretized and the RG flow equations are solved on this grid. Often, “naive” multi-dimensional finite difference schemes have been used to solve the partial differential equation (PDE) for the effective potential in the past. However, it should also be mentioned that other approaches have been developed, such as a discretization of field space via spectral methods in terms of Chebyshev polynomials or by approximating the effective potential in terms of splines or other basis functions or a matching of Taylor expansions, see the references above for concrete examples. Mostly, these approaches are suited to address the respective problem at hand. Especially in the context of fixed-point searches and analyses or studies of second-order phase transitions, some of these techniques are very powerful. Though, in the context of first-order phase transitions and multiple competing minima in the effective potential, most of the methods are no longer unconditionally stable. For example, Taylor expansions certainly fail, if the correct expansion point jumps from one minimum to another during the flow or if the radius of convergence is limited by nonanalyticities in the effective potential. The latter can easily be caused by chemical potentials or other external fields. Also, grid-based methods can fail in such situations, if the discretization scheme is for example fundamentally based on the assumption of analyticity and smoothness of the effective potential. However, since all of these challenging situations are of high physical relevance, we decided to address the construction of a powerful and stable multi-purpose discretization scheme for FRG flow equations in multi-dimensional field space again. In this spirit, this paper addresses researchers who are interested in the challenges and the numerical techniques which are required to solve FRG flow equations of bosonic potentials with more than one background field. Furthermore, the manuscript addresses readers who are interested in the general connection between FRG flow equations and fluid dynamics. For a more detailed introduction to FRG via zero-dimensional QFT and recent developments on the connections between FRG and numerical fluid dynamics we refer the interested reader to Refs. Koenigstein:2021syz ; Koenigstein:2021rxj ; Steil:2021cbu and some directly related works Grossi:2019urj ; Grossi:2021ksl ; Stoll:2021ori ; Ihssen:2022xjv ; Ihssen:2022xkr ; Batini:2023nan ; Ihssen:2023dmx ; Ihssen:2023qaq ; Ihssen:2023nqd ; Ihssen:2023xlp ; Ihssen:2024miv ; Ihssen:2024ihp . I.2 Research objective Within this work, we propose to use modern numerical fluid-dynamic techniques to solve FRG flow equations in multi-dimensional field space. The idea of looking at flow equations from a fluid

Read the complete text