betaFeedback

WorksPhysics and Astronomy

Reentrant d=2+1d=2+1 Superfluidity at High Temperature

Abstract

We present a UV-complete model with a global U(1)U(1) symmetry in d=2+1d=2+1 dimensions that exhibits a Berezinskii-Kosterlitz-Thouless (BKT) ordered phase at arbitrarily high temperatures. A careful separation of scales yields a two-dimensional Euclidean effective field theory whose construction remains perturbatively controlled even when its soft modes become strongly coupled. We determine the BKT phase boundary by carefully matching this theory to existing lattice Monte Carlo results. The high-temperature phase supports gapless static modes and algebraically decaying correlations even as T→∞T\to\infty.

Introduction

Systems that undergo order–disorder phase transitions are usually expected to be ordered at low temperature and disordered at high temperature. In Quantum Field Theory (QFT), this expectation is reflected in the typically positive thermal corrections to scalar masses squared, which favor symmetry restoration at high temperature. However, this need not be the case. Indeed, in ref. [1] S. Weinberg showed that these corrections can be negative in theories with several scalar fields. A symmetry can then remain broken at high temperature, or become broken upon heating.

Recent constructions of local, UV-complete QFTs in d=3+1d = 3 + 1 realize symmetry nonrestoration at arbitrarily high temperature [2, 3, 4, 5, 6, 7]. High-temperature order has also been studied in lattice models through entropic mechanisms [8, 9, 10, 11]. In 2+12 + 1 dimensions, persistent breaking of discrete symmetries has been found in local models [12, 13, 14, 15, 16, 17, 18]. Persistent breaking of continuous symmetries is also possible in nonlocal QFTs with long-range interactions, including in 2+12 + 1 dimensions [19, 20]. Here we construct a local, UV-complete model in 2+12 + 1 dimensions with a small number of fields and quasi-long-range U(1)U(1) order at arbitrarily high temperatures.

Consider a complex scalar χ\chi in d=2+1d = 2 + 1 spacetime dimensions with negative mass squared and weak quartic self-interaction w.r.t. to the mass parameter. At zero temperature, its condensate spontaneously breaks U(1)U(1), producing a massless Goldstone mode.

At any nonzero temperature TT, the long-wavelength static modes are described by a two-dimensional Euclidean theory. The Coleman–Mermin–Wagner theorem then forbids spontaneous breaking of the continuous symmetry [21, 22, 23]. Nevertheless, at sufficiently low temperature the weakly coupled model lies deep in a Berezinskii–Kosterlitz–Thouless (BKT) phase. The zero-temperature Goldstone field survives as its massless compact phase field, supporting algebraic correlations and superfluidity, with vortices bound in vortex–antivortex pairs [24, 25, 26, 27]. Thus an arbitrarily small temperature restores the symmetry while preserving the massless phase field.

Further heating eventually drives the ordinary scalar theory into a disordered phase: positive thermal corrections to the mass parameter lead to a gapped phase, with correlators that decay exponentially rather than algebraically.

In this work we address how to realize a BKT phase at arbitrarily high TT in a local UV-complete model, with a small number of fields. To that scope, we couple χ\chi to a Conformal Field Theory (CFT) through a parametrically weak portal OΔ∣χ∣2\mathcal{O}_{\Delta}|\chi|^2, where OΔ\mathcal{O}_{\Delta} is a scalar operator of dimension Δ\Delta. The sign of the portal coupling is chosen so that the thermal one-point function ⟨OΔ⟩T∝TΔ\langle\mathcal{O}_{\Delta}\rangle_T \propto T^\Delta induces a negative scalar mass shift at leading order.

The mechanism can be summarised as follows. At high temperatures, the negative conformal contribution to the scalar mass squared dominates the positive scalar thermal correction for Δ>1\Delta> 1. The resulting dimensionless phase stiffness grows as K(T)∼(T/m)Δ−1K(T) \sim(T/m)^{\Delta-1} so that algebraic correlations decay more slowly as the temperature is increased. We show that with all the microscopic couplings held fixed and sufficiently weak, this entire mechanism is controlled in the window 1<Δ<3/21 < \Delta< 3/2. The upper bound ensures both stability of the vacuum effective potential and parametrically decreasing backreaction on the conformal sector at high temperature.

The energy operator of the three-dimensional Ising CFT has Δ≃1.4126\Delta\simeq1.4126 [28] and provides a concrete construction of high TT superfluidity with only a small number of fields, which in turn admits a simple lattice description.

Our analysis proceeds by integrating out the screened CFT sector and the nonzero scalar Matsubara modes to obtain a local two-dimensional complex scalar theory. Comparing its couplings with the nonperturbative BKT boundary [29] determines its phase. For a sufficiently weak portal, the disordered regime survives between the low- and high-temperature BKT phases, giving at least two BKT transitions and the reentrant sequence BKT–disordered–BKT.

The model

We consider the model

S=SU(1)+SCFT+h∫d3x m2−ΔOΔ(x)∣χ(x)∣2,(1)S = S_{U(1)} + S_{\mathrm{CFT}} + h \int d^3x\, m^{2-\Delta}O_\Delta(x)|\chi(x)|^2, \tag*{(1)}

where we denote by SU(1)S_{U(1)} the ordinary U(1)U(1) model,

SU(1)=∫d3x[∣∂μχ∣2+g2(∣χ∣2−m2g)2].(2)S_{U(1)} = \int d^3x \left[|\partial_\mu\chi|^2 + \frac{g}{2}\left(|\chi|^2-\frac{m^2}{g}\right)^2\right]. \tag*{(2)}

Here gg and mm both have mass dimension one, [g]=[m]=1[g]=[m]=1. Throughout this work, we take the dimensionless coupling λ≡g/m≪1\lambda\equiv g/m \ll1 to be small. The putative d=2+1d=2+1 CFT described by SCFTS_{\mathrm{CFT}} contains a scalar operator OΔO_\Delta of dimension Δ>1\Delta>1. We couple the two sectors through the portal interaction with hh.

We must ensure boundedness of the effective potential of χ\chi. After integrating out the CFT, it is given by

Veff=g2(∣χ∣2−m2g)2+ECFT(J)−ECFT(0).(3)V_{\mathrm{eff}}=\frac{g}{2}\left(|\chi|^2-\frac{m^2}{g}\right)^2+\mathcal{E}_{\mathrm{CFT}}(J)-\mathcal{E}_{\mathrm{CFT}}(0). \tag*{(3)}

to leading order in gg. We have defined J≡hm2−Δ∣χ∣2J\equiv hm^{2-\Delta}|\chi|^2 and ECFT(J)−ECFT(0)=A±∣J∣33−Δ\mathcal{E}_{\mathrm{CFT}}(J)-\mathcal{E}_{\mathrm{CFT}}(0)=A_\pm|J|^{\frac{3}{3-\Delta}}. Here ECFT(J)\mathcal{E}_{\mathrm{CFT}}(J) denotes the vacuum-energy density, defined from −log⁡ZCFT(J)/Vol-\log Z_{\mathrm{CFT}}(J)/\mathrm{Vol}, the constants A±A_\pm correspond to the two signs of JJ, and scaling dictates that ∣J∣|J| dependence. At large ∣χ∣|\chi|, the positive scalar quartic dominates the CFT contribution provided Δ<3/2\Delta<3/2. This ensures boundedness of the effective potential.

The deep-UV fixed point of (1) is the product of a free complex scalar and the CFT. The condensate sources OΔO_\Delta, introducing the scale MCFT∼m(∣h∣/λ)1/(3−Δ)≪mM_{\mathrm{CFT}}\sim m(|h|/\lambda)^{1/(3-\Delta)}\ll m. If this relevant deformation has a massive endpoint, MCFTM_{\mathrm{CFT}} sets the corresponding gap. In other words, depending on whether the source v2OΔUVv^2O_{\Delta_{\mathrm{UV}}} gaps or triggers a flow to another CFT, the leading irrelevant IR interaction for k≪MCFTk\ll M_{\mathrm{CFT}} will be (∂θ)4(\partial\theta)^4 or perhaps (∂θ)2OΔIR(\partial\theta)^2O_{\Delta_{\mathrm{IR}}}.3 In the window MCFT≪k≪mM_{\mathrm{CFT}}\ll k\ll m, the EFT is approximately a compact Goldstone boson times a weakly deformed CFT, with the two sectors coupled only through irrelevant operators.

We now turn to the theory at finite temperature TT, where the Euclidean theory lives on S1/T1×R2S^1_{1/T}\times\mathbb{R}^2. We assume that all degrees of freedom other than the scalar zero mode have static screening gaps of order 2πnT2\pi nT or larger. Integrating them out at spatial momenta

k/(2πT)≪1.(4)k/(2\pi T)\ll1. \tag*{(4)}

gives a local two-dimensional action. The condition (4) will be assumed throughout. With the normalization χn=0=T φ\chi_{n=0}=\sqrt{T}\,\varphi,

S2=∫d2x[∣∇φ∣2+r2(μ)∣φ∣2+u22∣φ∣4+δL2].(5)S_2=\int d^2x\left[|\nabla\varphi|^2+r_2(\mu)|\varphi|^2+\frac{u_2}{2}|\varphi|^4+\delta\mathcal{L}_2\right]. \tag*{(5)}

Here μ\mu is the two-dimensional MS‾\overline{\mathrm{MS}} subtraction scale, and δL2\delta\mathcal{L}_2 contains additional local interactions. The coefficients r2r_2 and u2u_2 are obtained by matching to the 2+12+1-dimensional theory; their expansions depend on the temperature regime. The matching equations are perturbative as long as hard thermal loops are small, i.e.e.

g2πT≪1and∣h∣(m2πT)2−Δ≪1(6)\frac{g}{2\pi T}\ll1\qquad\text{and}\qquad|h|\left(\frac{m}{2\pi T}\right)^{2-\Delta}\ll1 \tag*{(6)}

Furthermore, provided these conditions are satisfied, δL2\delta\mathcal{L}_2 in (5) is negligible.

At low temperature but within the regime of (4) and (6), the expansion is organised about the radial minimum ∣φ∣2≃m2/(gT)|\varphi|^2\simeq m^2/(gT). In the window MCFT≪T≪mM_{\mathrm{CFT}}\ll T\ll m, the leading dynamics is governed by the ordinary U(1)U(1) theory. 4 The leading parametric behavior of r2r_2 and u2u_2 take the form,

r2(T;μ)=−m2,u2(T)=gT.(7)r_2(T;\mu)=-m^2,\qquad u_2(T)=gT. \tag*{(7)}

The aforementioned corrections and the portal contributions are subleading in this analysis.

For T≫mT\gg m and the favorable sign bOh<0b_Oh<0, the leading matching gives

r2(T;μ)=r2,U(1)(T;μ)−∣bOh∣m2−ΔTΔ,(8)r_2(T;\mu)=r_{2,U(1)}(T;\mu)-|b_Oh|m^{2-\Delta}T^\Delta, \tag*{(8)}
r2,U(1)(T;μ)=−m2+gT2πln⁡Tμ,u2(T)=gT.(9)r_{2,U(1)}(T;\mu)=-m^2+\frac{gT}{2\pi}\ln\frac{T}{\mu},\qquad u_2(T)=gT. \tag*{(9)}

where we suppress further higher cumulant and higher loop corrections. At leading order in hh, the CFT one-point function shifts only the quadratic coefficient and drives it toward negative values at high temperature.

We address to which extent our approximation are valid. We analyse the regime in which the thermal mass is dominated by the CFT-induced term −∣bOh∣m2−ΔTΔ-|b_Oh|m^{2-\Delta}T^\Delta. In particular we identify the conditions under which the back-reaction of the scalar background on the thermal CFT state remains parametrically small, so that higher cumulants are sub-leading and the small-source expansion is controlled.

To determine the thermal background of χ\chi, we expand the CFT contribution around zero source:

ECFTT(J)−ECFTT(0)(10)\mathcal{E}_{\mathrm{CFT}}^T(J)-\mathcal{E}_{\mathrm{CFT}}^T(0) \tag*{(10)}
=J⟨O⟩T−J22∫01/Tdτ∫d2x⃗ ⟨O(τ,x⃗)O(0)⟩T,c+O(J3).(11)=J\langle O\rangle_T-\frac{J^2}{2}\int_0^{1/T}d\tau\int d^2\vec{x}\,\langle O(\tau,\vec{x})O(0)\rangle_{T,c}+O(J^3). \tag*{(11)}

where ECFTT\mathcal{E}_{\mathrm{CFT}}^{T} denotes the thermal free-energy and ⟨O(x)⟩T=bOTΔ\langle O(x)\rangle_{T}=b_{O}T^{\Delta}. The second cumulant is ultraviolet finite for Δ<3/2\Delta<3/2 and infrared finite if the connected correlator decays exponentially at large spatial separation.5 We assume a regular thermal small-source expansion also for the higher cumulants.

For T≫mT\gg m, the first cumulant correction to (3) dominates the quadratic part of the effective potential for Δ>1\Delta>1. The second and higher cumulants can be ignored as the dimensionless coupling ∣J(χT)∣/T3−Δ|J(\chi_{T})|/T^{3-\Delta} that suppresses the higher cumulants decreases for Δ<32\Delta<\frac{3}{2},

∣J(χT)∣T3−Δ≃bOh2mg(mT)3−2Δ→0,(12)\frac{|J(\chi_{T})|}{T^{3-\Delta}}\simeq\frac{b_{O}h^{2}m}{g}\left(\frac{m}{T}\right)^{3-2\Delta}\to0, \tag*{(12)}

where χT\chi_{T} denotes the thermal saddle and ⟨O⟩T=bOTΔ\langle O\rangle_{T}=b_{O}T^{\Delta}. At low TT, such that T/m≪1T/m\ll1, higher cumulants can likewise be neglected provided ∣h∣/λ≪(T/m)3−Δ|h|/\lambda\ll(T/m)^{3-\Delta}.

Thus the same bound Δ<3/2\Delta<3/2 ensures quartic domination of the vacuum effective potential at large field, ultraviolet finiteness of the second thermal cumulant, and parametrically decreasing CFT backreaction as T→∞T\to\infty.

A concrete candidate for this mechanism is the d=2+1d=2+1 Ising CFT. Its leading Z2\mathbb{Z}_{2}-even scalar, the energy operator ϵ\epsilon, has scaling dimension Δϵ≃1.4126\Delta_{\epsilon}\simeq1.4126 [28], within the required window 1<Δϵ<3/21<\Delta_{\epsilon}<3/2, and a nonzero thermal one-point function ⟨ϵ⟩T=bϵTΔϵ\langle\epsilon\rangle_{T}=b_{\epsilon}T^{\Delta_{\epsilon}}, with bϵ=0.63(7)b_{\epsilon}=0.63(7) [31]. Choosing OΔ=ϵO_{\Delta}=\epsilon and the portal sign so that hbϵ<0hb_{\epsilon}<0 then gives the negative thermal mass contribution needed for the mechanism.

The phases of the 2D theory

Here we determine whether the leading static theory (5) is gapped or exhibits quasi-long-range order. Throughout this section, we study the family of static theories parametrized by u2u_{2} and m02≡r2(μ0)m_{0}^{2}\equiv r_{2}(\mu_{0}), holding these parameters fixed in each theory. Here μ0>0\mu_{0}>0 is a fixed reference mass, u2>0u_{2}>0, and m02m_{0}^{2} can have either sign. The quadratic coefficient runs as r2(μ)=m02+u22πlog⁡(μ02/μ2)r_{2}(\mu)=m_{0}^{2}+\frac{u_{2}}{2\pi}\log(\mu_{0}^{2}/\mu^{2}). The phase diagram can be formulated without specifying their temperature dependence. Thermal matching will select a trajectory through this phase diagram, as discussed in Sec. IV.

In two-dimensions [φ]=0[\varphi]=0 and [r2]=[u2]=2[r_{2}]=[u_{2}]=2. The theory is superrenormalizable. The field and quartic coupling have no logarithmic ultraviolet renormalization, while the quadratic coefficient is additively renormalized. In the MS convention used below, the mass parameter runs as dr2(μ)/dlog⁡μ2=−u2/(2π)d r_{2}(\mu)/d\log\mu^{2}=-u_{2}/(2\pi) (see Appendix A). Consequently 2r2/u22r_{2}/u_{2} alone is not invariant under a change of subtraction scale. A convenient scale-independent coordinate on the two-dimensional theory space is R\mathcal{R}, defined by

R≡2r2(μ)u2+1πlog⁡2μ2u2.(13)\mathcal{R}\equiv\frac{2r_{2}(\mu)}{u_{2}}+\frac{1}{\pi}\log\frac{2\mu^{2}}{u_{2}}. \tag*{(13)}

At μ2=u2/2\mu^{2}=u_{2}/2, this reduces to R=2r2(μ)/u2\mathcal{R}=2r_{2}(\mu)/u_{2}. Equivalently, R=2m02u2+1πlog⁡2μ02u2\mathcal{R}=\frac{2m_{0}^{2}}{u_{2}}+\frac{1}{\pi}\log\frac{2\mu_{0}^{2}}{u_{2}}.

Deep massive regime: The deep massive regime is characterized by R≫1\mathcal{R}\gg1. Choosing the subtraction scale μ=μ0\mu=\mu_{0}, with μ02\mu_{0}^{2} of order m02>0m_{0}^{2}>0, we have r2(μ0)=m02r_{2}(\mu_{0})=m_{0}^{2} and no large perturbative logarithms. The gap is therefore Mgap2=m02[1+O(R−1)]M_{\mathrm{gap}}^{2}=m_{0}^{2}[1+O(\mathcal{R}^{-1})]. The two-point function ⟨φ(x)φ∗(0)⟩\langle\varphi(x)\varphi^{\ast}(0)\rangle decays exponentially as ∣x∣→∞|x|\to\infty, with correlation length Mgap−1M_{\mathrm{gap}}^{-1}. The finite mass gap and exponential decay of correlations therefore place the theory deep in the disordered regime.

Deep algebraic regime and the phase EFT: The deep algebraic regime is characterized by R≪−1\mathcal{R}\ll-1. Choosing the subtraction scale μ=μ0\mu=\mu_{0}, with μ02\mu_{0}^{2} of order ∣m02∣|m_{0}^{2}| and m02<0m_{0}^{2}<0, we have r2(μ0)=m02r_{2}(\mu_{0})=m_{0}^{2} and ∣m02∣/u2≫1|m_{0}^{2}|/u_{2}\gg1. The potential then has a well-developed Mexican-hat minimum. The classical saddle and radial mass are, to leading order, v2≃−Rv^{2}\simeq-\mathcal{R} and mρ2≃−u2Rm_{\rho}^{2}\simeq-u_{2}\mathcal{R}. The expansion in radial fluctuations is controlled by 1/v2∼−1/R≪11/v^{2}\sim-1/\mathcal{R}\ll1.

The compact phase of the complex field admits vortex defects. Writing φ=(v+ρ)eiθ/2\varphi=(v+\rho)e^{i\theta}/\sqrt{2} and integrating out the radial mode at momenta k≪mρk\ll m_{\rho}, we obtain the phase EFT

SEFT=K2∫d2x[(∂θ)2−1mρ2(∂θ)4+⋯ ],(14)S_{\mathrm{EFT}}=\frac{K}{2}\int d^{2}x\left[(\partial\theta)^{2}-\frac{1}{m_{\rho}^{2}}(\partial\theta)^{4}+\cdots\right], \tag*{(14)}

where θ∼θ+2π\theta\sim\theta+2\pi. Here KK is the stiffness matched at the radial scale, with K=v2[1+O(v−2)]K=v^{2}[1+O(v^{-2})]. Thus the spin-wave (sw) corrections are local derivative operators suppressed by mρm_{\rho}. Smooth configurations alone are not the complete EFT, because the phase is compact. An isolated vortex is a saddle in a fixed-winding sector; with neutral boundary conditions, vortices instead enter through interacting vortex–antivortex configurations weighted by their core actions. The organization of the compact phase EFT as a neutral vortex gas is standard [27, 32]. Let Zsw=∫Dθ e−SswZ_{\mathrm{sw}}=\int D\theta\,e^{-S_{\mathrm{sw}}}. With neutral boundary conditions, the finite-volume fugacity expansion begins with a vortex–antivortex pair and may be organized as Z=Zsw[1+O(y12)+⋯ ]Z=Z_{\mathrm{sw}}[1+O(y_{1}^{2})+\cdots], where y1=O(e−ccorev2)y_{1}=O(e^{-c_{\mathrm{core}}v^{2}}), with ccore>0c_{\mathrm{core}}>0. The infrared theory therefore approaches the compact-boson fixed line, where

⟨eiθ(x)e−iθ(0)⟩∼∣x∣−η,η=12πK.(15)\left\langle e^{i\theta(x)}e^{-i\theta(0)}\right\rangle\sim|x|^{-\eta},\qquad\eta=\frac{1}{2\pi K}. \tag*{(15)}

There is algebraic order (a.k.a. quasi long range order) but no nonzero infinite-volume order parameter.

The critical point: The determination of KK and y1y_{1} from (r2,u2)(r_{2},u_{2}) is semiclassically controlled only in the deep algebraic regime. Near the transition, radial and vortex-core fluctuations are both order one. In other words, for R=O(1)\mathcal{R} = O(1) the two-dimensional static theory is strongly coupled; from the 2+12+1-dimensional thermal perspective this is the soft sector, and locating its critical point requires a nonperturbative numerical treatment.

The asymptotic long-distance flow is nevertheless BKT. Writing KRK_{\mathrm{R}} for the renormalized infrared stiffness, the critical trajectory satisfies KR=2/πK_{\mathrm{R}} = 2/\pi, with Δv=πKR=2\Delta_v = \pi K_{\mathrm{R}} = 2. What cannot be obtained from the small-fugacity equations alone is the location of that trajectory in the static (r2,u2)(r_2,u_2) parameter space. Prokof’ev, Ruebenacker, and Svistunov (PRS) locate this trajectory by Monte Carlo simulation of a regulated ∣ψ∣4|\psi|^4 model and BKT finite-size scaling [29]. We translate the PRS conventions into the MS convention of (5) and find

Rc=−0.821(10).(16)\mathcal{R}_c=-0.821(10). \tag*{(16)}

The detailed lattice-to-dimensional-regularization conversion is given in Appendix A.

Eq. (15) is the nonperturbative position of the complex scalar theory on the BKT separatrix. The phase assignment is therefore R<Rc\mathcal{R}<\mathcal{R}_c for the BKT phase and R>Rc\mathcal{R}>\mathcal{R}_c for the massive phase. At the scale μ2=u2/2\mu^2=u_2/2, the negative value of Rc\mathcal{R}_c means that the transition occurs while r2(μ)<0r_2(\mu)<0. As R\mathcal{R} increases through Rc\mathcal{R}_c, the unit-vortex operator changes from irrelevant to relevant. The numerical value of Rc\mathcal{R}_c is fixed by the full nonperturbative radial and vortex-core dynamics.

From a static point to a thermal trajectory

A 2+12+1-dimensional UV completion supplies a one-parameter family of static theories of the form (5), {r2(T;μ),u2(T)}\{r_2(T;\mu),u_2(T)\}. The phase, either quasi-long-range order or disorder, is diagnosed by R(T)=2r2(T;μ)/u2(T)+π−1log⁡[2μ2/u2(T)]\mathcal{R}(T)=2r_2(T;\mu)/u_2(T)+\pi^{-1}\log[2\mu^2/u_2(T)], which is (12) evaluated along the one-parameter trajectory supplied by the UV completion and parametrized by TT. The PRS critical surface therefore appears as the horizontal line R=Rc\mathcal{R}=\mathcal{R}_c when R(T)\mathcal{R}(T) is plotted against temperature. We define δ(T)≡R(T)−Rc\delta(T)\equiv\mathcal{R}(T)-\mathcal{R}_c, so that

δ(T)<0: BKT,δ(T)>0: massive,(17)\delta(T)<0:\ \mathrm{BKT},\qquad\delta(T)>0:\ \mathrm{massive}, \tag*{(17)}

while the BKT transition occurs at δ(TBKT)=0\delta(T_{\mathrm{BKT}})=0.

For the ordinary U(1)U(1) model, i.e. the theory (2),

δU(1)={−2m2gT+O(log⁡[m2gT])→−∞,T→0,1πlog⁡(Tg)+O(1)→+∞,T→∞.(18)\delta_{U(1)}= \begin{cases} -\dfrac{2m^2}{gT}+O\left(\log\left[\dfrac{m^2}{gT}\right]\right)\to-\infty, & T\to0,\\ \dfrac{1}{\pi}\log\left(\dfrac{T}{g}\right)+O(1)\to+\infty, & T\to\infty. \end{cases} \tag*{(18)}

Indeed, at low temperature eq. (7) gives RU(1)(T)=−2m2/(gT)+O[log⁡(m2/(gT))]\mathcal{R}_{U(1)}(T)=-2m^2/(gT)+O[\log(m^2/(gT))]. At high temperature, hard matching in the MS scheme gives (9), therefore one obtains RU(1)(T)=π−1log⁡(T/g)+O(1)\mathcal{R}_{U(1)}(T)=\pi^{-1}\log(T/g)+O(1). By continuity, the trajectory therefore crosses the critical surface at least once.

For the favorable portal sign bϕh<0b\phi h<0 and 1<Δ<3/21<\Delta<3/2, the CFT-coupled trajectory shares the ordinary low-temperature endpoint but has the high-temperature asymptotics

δU(1)+CFT=−2∣bϕh∣λ(Tm)Δ−1+O(log⁡Tg)→−∞(19)\delta_{U(1)+\mathrm{CFT}}=-\frac{2|b\phi h|}{\lambda}\left(\frac{T}{m}\right)^{\Delta-1}+O\left(\log\frac{T}{g}\right)\to-\infty \tag*{(19)}

as T→∞T\to\infty.

Indeed, eqs. (8,9) give δU(1)+CFT(T)=δU(1)(T)−(2∣bϕh∣/λ)(T/m)Δ−1+…\delta_{U(1)+\mathrm{CFT}}(T)=\delta_{U(1)}(T)-(2|b\phi h|/\lambda)(T/m)^{\Delta-1}+\ldots. Since Δ>1\Delta>1, the portal contribution vanishes as T→0T\to0, so the low-temperature endpoint is unchanged. As argued above, the upper bound Δ<3/2\Delta<3/2 ensures, through (11), that the CFT backreaction decreases in this limit and the leading matching remains controlled.

Equation (18) places the coupled trajectory in the BKT region at arbitrary high TT. For a sufficiently small but nonzero favorable portal coupling, the intermediate massive region of the ordinary trajectory persists, so that max⁡TδU(1)+CFT(T)>0\max_T\delta_{U(1)+\mathrm{CFT}}(T)>0. By continuity, the trajectory then crosses the critical surface at least twice, yielding the reentrant phase sequence BKT-massive-BKT. The three deep regimes can be parametrically controlled, while the crossings themselves are fixed nonperturbatively by the PRS critical value.

For a sufficiently strong portal contribution relative to the scalar self-interaction, with the favorable sign, the intermediate massive interval can disappear, leaving a BKT phase at every nonzero temperature. Note that the asymptotic high TT suppression of CFT backreaction follows from Δ<3/2\Delta<3/2 and holds for any fixed couplings hh and λ>0\lambda>0, independently of their relative size.

To illustrate the endpoint asymptotics, we plot representative trajectories against x=log⁡(T/m)x = \log(T/m) in Figure 1. We remark that the couplings ensuring perturbative hard matching conditions (6) and the local expansion parameter (4) remain small throughout the plotted range, while the soft modes become strongly coupled near δ=0\delta= 0. In the plotted curves, t=T/mt = T/m and a=∣bOh∣a = |b_O h|; setting a=0a = 0 gives the ordinary trajectory. With δ=R−Rc\delta= \mathcal{R} - \mathcal{R}_c, the a=10−6a = 10^{-6} curve reaches R≃7.95\mathcal{R} \simeq7.95 at its maximum and R<−13\mathcal{R} < -13 at both ends of the plotted range. The curves are schematic and do not determine the transition temperatures.

Figure 1. We plot an analytic example, δU(1)+CFTrep(t)=−2λt+1πlog⁡(1+t)−2aλtΔ−1+c0\delta^{\mathrm{rep}}_{U(1)+\mathrm{CFT}}(t)=-\frac{2}{\lambda t}+\frac{1}{\pi}\log(1+t)-\frac{2a}{\lambda}t^{\Delta-1}+c_0, which reproduces the asymptotic endpoint behaviours. Here c0c_0 is a vertical offset. For illustration, λ=0.01\lambda=0.01, a=10−4,10−6a=10^{-4},10^{-6}, Δ=1.4\Delta=1.4, and c0=1.5c_0=1.5.

For sufficiently weak portal coupling, the coupled trajectory closely follows the ordinary U(1)U(1) asymptotics over a broad temperature range. A smaller portal coupling extends this range to higher temperatures. The reentrant BKT phase at high temperature is driven by the asymptotic dominance of the CFT contribution.

Discussion

We have established a local and UV-complete QFT, given by (1), that features a reentrant BKT–disordered–BKT phase sequence as the temperature increases.

A central feature of the mechanism is the simultaneous high-temperature limit

K(T)∼(T/m)Δ−1⟶∞,(20)K(T) \sim(T/m)^{\Delta-1} \longrightarrow\infty, \tag*{(20)}
∣J(χT)∣/T3−Δ∼(T/m)2Δ−3⟶0.(21)|J(\chi_T)|/T^{3-\Delta} \sim(T/m)^{2\Delta-3} \longrightarrow0 . \tag*{(21)}

For 1<Δ<3/21 < \Delta< 3/2, increasing temperature enhances the phase stiffness KK while suppressing the backreaction on the thermal CFT state.

The d=3d = 3 Ising CFT provides a concrete example through its energy operator, whose scaling dimension is Δϵ≃1.4\Delta_{\epsilon} \simeq1.4. Thus providing an example of a model with a small number of fields featuring BKT phase at arbitrarily high temperature.

Our model supports a BKT phase at arbitrarily high temperatures with only a few fields; the simplest realization involves just χ\chi and a critical Ising field. It would therefore be particularly interesting to construct a simple microscopic lattice model exhibiting the same reentrant BKT behaviour.

A few directions deserve further investigation. It would be interesting to investigate the real-time dynamics of the high-temperature BKT phase. Dimensional analysis suggests that the damping of phase fluctuations at long wavelength decreases with temperature. A careful computation of the hydrodynamic modes and their dissipation would allow us to test this expectation.

Finally, it would be interesting to analyse non-Abelian generalisations, in particular whether varying the temperature can drive transitions between WZW or their coset CFTs and massive phases. Gauged versions of these theories and of the Abelian model studied here would also be interesting.

Note added. This work was first presented at the University of Geneva on 8 October 2026 [33]. While this manuscript was being completed Ref. [34] appeared. There it is shown that, for sufficiently large NN, the bi-conical fixed point of an O(N)×U(1)O(N) \times U(1) model supports a BKT phase, and hence superfluidity, at all nonzero temperatures.

Acknowledgements

We are grateful to Olivier Delouche for useful discussions and comments on the draft. We thank Brando Bellazzini, James Ingoldby, Markus Luty, Francesco Riva, Junchen Rong, Giovanni Villadoro and Yuan Xin for useful discussions. S.D.C. and J.E.M. are supported by the European Research Council, grant agreement n. 101039756.

Lattice and dimensional regularisation matching

This appendix derives the critical value quoted in (16). PRS [29] study a classical complex field on a square lattice, with Hamiltonian

H=∑k∈BZ[E(k)−νlat]∣ψk∣2+U2∑i∣ψi∣4,E(k)=4−2cos⁡kx−2cos⁡ky,(22)H = \sum_{k\in\mathrm{BZ}} \left[E(\mathbf{k})-\nu_{\mathrm{lat}}\right]|\psi_k|^2+\frac{U}{2}\sum_i|\psi_i|^4,\qquad E(\mathbf{k})=4-2\cos k_x-2\cos k_y, \tag*{(22)}

and statistical weight e−He^{-H}. Here we adopt their simulation units T=a=1T=a=1, M=1/2M=1/2.

They locate the transition by varying νlat\nu_{\mathrm{lat}} until the measured superfluid density satisfies the Nelson–Kosterlitz criterion ns(L)=1πn_s(L)=\frac{1}{\pi}. In these units, the continuum phase stiffness is KR=2nsK_{\mathrm{R}}=2n_s,6 so this criterion corresponds to the universal critical value KR,c=2/πK_{\mathrm{R,c}}=2/\pi. The total density and chemical potential at this point define nc(L)n_c(L) and νc(L)\nu_c(L); Repeating this procedure at several lattice sizes, they extract the infinite-volume critical density and chemical potential using the finite-size dependence predicted by BKT theory. In the continuum limit of their lattice model, PRS obtain the lattice critical chemical potential

νlat,c=U2πlog⁡(4AμU),Aμ=211(6).(23)\nu_{\mathrm{lat,c}}=\frac{U}{2\pi}\log\left(\frac{4A_\mu}{U}\right),\qquad A_\mu=211(6). \tag*{(23)}

Introduce a physical lattice spacing aa and define the continuum complex field by Φ(xi)=ψi\Phi(x_i)=\psi_i, with xi=a ix_i=a\,i. No field rescaling is required in the PRS simulation units. Since E(k)=k2+O(k4)E(\mathbf{k})=\mathbf{k}^2+O(k^4), the continuum action and the PRS critical line take the form

S=∫d2x[∣∇Φ∣2+r0∣Φ∣2+u22∣Φ∣4],2r0,cu2=−1πlog⁡(4Aμu2a2).(24)S=\int d^2x\left[|\nabla\Phi|^2+r_0|\Phi|^2+\frac{u_2}{2}|\Phi|^4\right],\qquad\frac{2r_{0,\mathrm{c}}}{u_2}=-\frac{1}{\pi}\log\left(\frac{4A_\mu}{u_2a^2}\right). \tag*{(24)}

Here Aμ=211(6)A_\mu=211(6) is the constant quoted in (A2), and the bare parameters are r0=−νlat/a2r_0=-\nu_{\mathrm{lat}}/a^2 and u2=U/a2u_2=U/a^2, in the normalization used in the main text. To relate the bare mass r0r_0 to the renormalized parameter r2(μ)r_2(\mu), we compare the tadpole contribution in lattice and dimensional regularization.

The lattice and dimensionally regularized tadpoles are

Ilat(s)=∫BZd2p(2π)21p^ 2+s=14πlog⁡32a2s+O(a2slog⁡(a2s)),(25)I_{\mathrm{lat}}(s)=\int_{\mathrm{BZ}}\frac{d^2p}{(2\pi)^2}\frac{1}{\widehat{p}^{\,2}+s}=\frac{1}{4\pi}\log\frac{32}{a^2s}+O(a^2s\log(a^2s)), \tag*{(25)}
I_{\mathrm{DR}}(s)=\left(\frac{e^{\gamma_{\mathrm{E}}\mu^2}{4\pi}\right)^\epsilon}\int\frac{d^{2-2\epsilon}p}{(2\pi)^{2-2\epsilon}}\frac{1}{p^2+s}=\frac{1}{4\pi}\left[\frac{1}{\epsilon}+\log\frac{\mu^2}{s}\right]+O(\epsilon). \tag*{(26)}

Here s>0s>0 is a common auxiliary mass squared regulating the infrared, p^ 2=E(ap)/a2\widehat{p}^{\,2}=E(a\mathbf{p})/a^2 is the lattice kinetic kernel, and BZ=[−π/a,π/a]2\mathrm{BZ}=[-\pi/a,\pi/a]^2. The lattice result is expanded for a2s→0a^2s\to0. In dimensional regularization, d=2−2ϵd=2-2\epsilon and μ\mu is the MS‾\overline{\mathrm{MS}} subtraction scale; after subtraction, the finite part is IDRren(s)=(4π)−1log⁡(μ2/s)I_{\mathrm{DR}}^{\mathrm{ren}}(s)=(4\pi)^{-1}\log(\mu^2/s). With the normalization in (A3), the tadpole self-energy is 2u2I2u_2I. Matching the quadratic terms therefore gives, in the continuum limit,

r2(μ)=r0+2u2[Ilat(s)−IDRren(s)]=r0+u22πlog⁡32a2μ2.(27)r_2(\mu)=r_0+2u_2\left[I_{\mathrm{lat}}(s)-I_{\mathrm{DR}}^{\mathrm{ren}}(s)\right]=r_0+\frac{u_2}{2\pi}\log\frac{32}{a^2\mu^2}. \tag*{(27)}

The auxiliary mass ss cancels. Combining the PRS critical line (A3) with the definition (13) then yields

Rc=2r0,cu2+1πlog⁡64u2a2=1πlog⁡16Aμ=−0.821(10).(28)R_c=\frac{2r_{0,\mathrm{c}}}{u_2}+\frac{1}{\pi}\log\frac{64}{u_2a^2}=\frac{1}{\pi}\log\frac{16}{A_\mu}=-0.821(10). \tag*{(28)}

The cutoff aa and subtraction scale μ\mu cancel from RcR_c. In PRS notation, ξμ≡Aμ/16=13.2(4)\xi_\mu\equiv A_\mu/16=13.2(4), so the same result reads Rc=−log⁡ξμ/πR_c=-\log\xi_\mu/\pi.

References

  1. [1]S. Weinberg, “Gauge and Global Symmetries at High Temperature,” Phys. Rev. D 9 (1974) 3357–3378.DOI
  2. [2]S. Chaudhuri, C. Choi, and E. Rabinovici, “Thermal order in large N conformal gauge theories,” JHEP 04 (2021) 203, arXiv:2011.13981 [hep–th].DOI
  3. [3]S. Chaudhuri and E. Rabinovici, “Symmetry breaking at high temperatures in large N gauge theories,” JHEP 08 (2021) 148, arXiv:2106.11323 [hep–th].DOI
  4. [4]B. Bajc, A. Lugo, and F. Sannino, “Asymptotically free and safe fate of symmetry nonrestoration,” Phys. Rev. D 103 (2021) 096014, arXiv:2012.08428 [hep–th].
  5. [5]B. Bajc, G. Muča, F. Sannino, and S. Wagner, “Infinite heat order in 3+1 dimensions,” Phys. Rev. D 113 no. 12, (2026) 125029, arXiv:2604.01184 [hep–th].arxiv.org/abs/2604.01184
  6. [6]S. Harder, Z. Komargodski, F. K. Popov, A. Sharon, and Claude, “Resisting misalignment.” Bhaumik Institute blog post, Oct., 2026.
  7. [7]S. Harder and A. Sharon, “A Superfluid at All Temperatures,” arXiv:2610.10169 [hep–th].arxiv.org/abs/2610.10169
  8. [8]Y. Han, X. Huang, Z. Komargodski, A. Lucas, and F. K. Popov, “Entropic order,” Nature Commun. 17 no. 1, (2026) 87, arXiv:2503.22789 [cond–mat.stat–mech].arxiv.org/abs/2503.22789
  9. [9]T. Zhang, Z. Chen, and Z. Cai, “Infinite-temperature quantum phases and phase transitions,” 2025. https://arxiv.org/abs/2509.00855.
  10. [10]P.-S. Hsin and R. Kobayashi, “Exploring Entropic Orders: High Temperature Continuous Symmetry Breaking, Chiral Topological States and Local Commuting Projector Models,” arXiv:2604.18694 [cond–mat.str–el].arxiv.org/abs/2604.18694
  11. [11]Y.-L. Tsao and Z.-X. Luo, “Entropic Barriers and the Kinetic Suppression of Topological Defects,” arXiv:2602.16777 [quant-ph].
  12. [12]N. Chai, S. Chaudhuri, C. Choi, Z. Komargodski, E. Rabinovici, and M. Smolkin, “Thermal Order in Conformal Theories,” Phys. Rev. D 102 no. 6, (2020) 065014, arXiv:2005.03676 [hep-th].DOI
  13. [13]N. Chai, S. Chaudhuri, C. Choi, Z. Komargodski, E. Rabinovici, and M. Smolkin, “Symmetry Breaking at All Temperatures,” Phys. Rev. Lett. 125 no. 13, (2020) 131603.DOI
  14. [14]P. Liendo, J. Rong, and H. Zhang, “Spontaneous breaking of finite group symmetries at all temperatures,” SciPost Phys. 14 no. 6, (2023) 168, arXiv:2205.13964 [hep-th].DOI
  15. [15]B. Hawashin, J. Rong, and M. M. Scherer, “Ultraviolet-Complete Local Field Theory of Persistent Symmetry Breaking in 2+1 Dimensions,” Phys. Rev. Lett. 134 no. 4, (2025) 041602, arXiv:2409.10606 [hep-th].DOI
  16. [16]Z. Komargodski and F. K. Popov, “Temperature-Resistant Order in 2+1 Dimensions,” Phys. Rev. Lett. 135 no. 9, (2025) 091602, arXiv:2412.09459 [hep-th].arxiv.org/abs/2412.09459
  17. [17]M. Smolkin and L. Yung, “Thermal Order in the Biconical Model,” arXiv:2608.02720 [hep-th].
  18. [18]S. Chaudhuri, B. Hawashin, E. Rabinovici, and M. M. Scherer, “Exploring thermal order in conformal theories with multiple scalars coupled to an O(N) vector field,” arXiv:2609.00160 [hep-th].arxiv.org/abs/2609.00160
  19. [19]N. Chai, A. Dymarsky, and M. Smolkin, “Model of Persistent Breaking of Discrete Symmetry,” Phys. Rev. Lett. 128 no. 1, (2022) 011601, arXiv:2106.09723 [hep-th].DOI
  20. [20]N. Chai, A. Dymarsky, M. Goykhman, R. Sinha, and M. Smolkin, “A model of persistent breaking of continuous symmetry,” SciPost Phys. 12 no. 6, (2022) 181, arXiv:2111.02474 [hep-th].DOI
  21. [21]S. R. Coleman, “There are no Goldstone bosons in two-dimensions,” Commun. Math. Phys. 31 (1973) 259–264.DOI
  22. [22]N. D. Mermin and H. Wagner, “Absence of ferromagnetism or antiferromagnetism in one-dimensional or two-dimensional isotropic Heisenberg models,” Phys. Rev. Lett. 17 (1966) 1133–1136.
  23. [23]P. C. Hohenberg, “Existence of Long-Range Order in One and Two Dimensions,” Phys. Rev. 158 (1967) 383–386.DOI
  24. [24]V. L. Berezinskii, “Destruction of Long-range Order in One-dimensional and Two-dimensional Systems having a Continuous Symmetry Group I. Classical Systems,” Sov. Phys. JETP 32 (1971) 493–500.
  25. [25]V. L. Berezinskii, “Destruction of Long-range Order in One-dimensional and Two-dimensional Systems Possessing a Continuous Symmetry Group. II. Quantum Systems,” Sov. Phys. JETP 34 no. 3, (1972) 610–616.
  26. [26]J. M. Kosterlitz and D. J. Thouless, “Ordering, metastability and phase transitions in two-dimensional systems,” J. Phys. C 6 (1973) 1181–1203.
  27. [27]J. M. Kosterlitz, “The critical properties of the two-dimensional xy model,” J. Phys. C 7 no. 6, (1974) 1046.DOI
  28. [28]F. Kos, D. Poland, D. Simmons-Duffin, and A. Vichi, “Precision Islands in the Ising O(N) Models,” JHEP 08 (2016) 036, arXiv:1603.04436 [hep-th].DOI
  29. [29]N. Prokof’ev, O. Ruebenacker, and B. Svistunov, “Critical point of a weakly interacting two-dimensional bose gas,” Physical Review Letters 87 no. 27, (Dec., 2001). http://dx.doi.org/10.1103/PhysRevLett.87.270402.
  30. [30]T. Brauner, A. Esposito, and R. Penco, “Fractional Soft Limits of Scattering Amplitudes,” Phys. Rev. Lett. 128 no. 23, (2022) 231601, arXiv:2203.00022 [hep-th].
  31. [31]L. Iliesiu, M. Koloğlu, and D. Simmons-Duffin, “Bootstrapping the 3d Ising model at finite temperature,” JHEP 12 (2019) 072, arXiv:1811.05451 [hep-th].DOI
  32. [32]J. V. José, L. P. Kadanoff, S. Kirkpatrick, and D. R. Nelson, “Renormalization, vortices, and symmetry-breaking perturbations in the two-dimensional planar model,” Phys. Rev. B 16 (Aug, 1977) 1217–1241. https://link.aps.org/doi/10.1103/PhysRevB.16.1217.DOI
  33. [33]Joan Elias Miró. Reentrant Algebraic Order at High Temperature, 8 October 2026. Department of Theoretical Physics, Univ. of Geneva. https://agenda.unige.ch/events/view/46959.
  34. [34]B. Hawashin, S. Chaudhuri, I. F. Herbut, and M. M. Scherer, “All-temperature superfluidity in two dimensions,” arXiv:2610.20717 [cond-mat.str-el].arxiv.org/abs/2610.20717
  35. [35]Z. Hadzibabic and J. Dalibard, “Two-dimensional Bose fluids: An atomic physics perspective,” Riv. Nuovo Cim. 34 no. 6, (2011) 389–434, arXiv:0912.1490 [cond-mat.quant-gas].DOI

Paper details

vlnc:W_qe529u5p
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents