betaFeedback

WorksPhysics and Astronomy

Chiral Cherenkov radiation in the presence of a time-dependent chiral chemical potential

Abstract

The photon production process f→f+γf\to f+γ in the presence of a time-dependent chiral chemical potential μ5μ_5 is studied. The validity of the adiabatic approximation is demonstrated in the ultra-relativistic limit. Analytical expressions for the photon spectrum are derived for two models of the chiral magnetic conductivity b0∝μ5b_0\propto μ_5: (i) b0(t)=A1+B1tanh⁡(t/τ)b_0(t)= A_1+B_1\tanh(t/τ) and (ii) b0(t)=A2(t/τ)/(1+t2/τ2)b_0(t)= A_2(t/τ)/(1+t^2/τ^2). It is known that for constant b0b_0, photon emission may exhibit a resonance for one photon polarization depending on the sign of b0b_0. It is shown that in model (i), up to two resonances may occur, depending on the sign of the asymptotic values b0(∞)b_0(\infty) and b0(−∞)b_0(-\infty). No resonances are observed in model (ii). Numerical calculations are performed using parameters relevant to quark-gluon plasma, and the universality of the results is discussed.

Introduction

One of the most intriguing features of chiral media is the chiral magnetic effect whereby electric current is induced in the direction of the external magnetic field [1, 2, 3, 4, 5]. The corresponding transport coefficient b0b_0, known as the chiral magnetic conductivity, is proportional to the chiral chemical potential μ5\mu_5, which reflects the net chirality in the medium. The chiral imbalance can have various sources. In quark-gluon plasma it may be induced by the parallel chromoelectric and chromomagnetic fields in the initial stage following a heavy-ion collision [6] or by a sphaleron transition at a later stage [5]. The phenomenological manifestation of the chiral magnetic effect is electric charge separation with respect to the collision plane of the heavy ions [4]. Recent experimental measurements have produced encouraging signs of this charge separation [7]. Another manifestation of the chiral magnetic effect, is the chiral Cherenkov radiation, in which a circularly polarized photon is emitted by a charges fermion [8, 9]. This process, kinematically prohibited in the empty space, becomes possible owing to the momentum supplied by the chiral magnetic current. The chiral Cherenkov radiation contributes to the parton energy loss and the jet structure, which can be directly observed in relativistic heavy-ion experiments.

The time evolution of the chiral chemical potential is governed by three processes. The first is above-mentioned sphaleron transition, a stochastic process that generates the chiral charge. At very high temperatures, the transition time is estimated to be significantly shorter than the lifetime of the quark-gluon plasma [10, 11]. However, at phenomenologically relevant temperatures, these time scales are comparable. The second process is a graduate relaxation of the chiral chemical potential due to the inelastic processes breaking the chiral symmetry [12]. The third process is chirality transfer to the electromagnetic field via the inverse chiral cascade [13, 14, 15, 16, 17, 18, 19, 20]. These processes impart time-dependence to the chiral chemical potential, which, in turn, impacts the chiral magnetic effect, the charge separation [21], and the chiral Cherenkov radiation. The latter is the subject of this paper.

The chiral Cherenkov radiation is the excitation of the electromagnetic field in a chiral medium by a charged fermion. If the time-dependence of b0b_0 is neglected, the eigenstates of the electromagnetic field are circularly polarized monochromatic plane waves with a dispersion relation dependent on the photon polarization. The transition amplitude is then proportional to the delta function, which ensures energy conservation. However, when b0b_0 has a non-trivial time-dependence, the delta function gives way to a distribution proportional to the Fourier image of the photon amplitude. The relationship between the photon production rate and this amplitude is provided by perturbation theory.

The main goal of this paper is to compute the photon production rate, given the time-dependence of b0b_0. The precise form of b0b_0 is unknown. It therefore advantageous to employ models that both provide a qualitative description of its time-evolution, and are amenable to analytical methods. Two models are examined: (i) b0=A1+B1tanh⁡(t/τ)b_0 = A_1 + B_1 \tanh(t/\tau) and (ii) b0=A2(t/τ)/(1+t2/τ2)b_0 = A_2(t/\tau)/(1 + t^2/\tau^2), where A1A_1, B1B_1, A2A_2 and τ>0\tau> 0 are real numbers. Model (i) describes a transition between states characterized by different asymptotic values of b0b_0, whereas model (ii) describes a burst of chirality with a finite duration. It will be shown that the models exhibit qualitatively distinct spectra, differing from the spectrum associated with constant b0b_0.

The paper is structured as follows. In Sec. II the general formalism for computing the photon production rate at the leading order of perturbation theory in the ultra-relativistic approximation is reviewed. The main formula for the transition probability is (13). The photon amplitude a(t)a(t) satisfies a second order differential equation (3), which is solved for a given b0(t)b_0(t). Sec. III lays out the benchmark results for the photon rate at constant b0b_0. Figure 1 shows the corresponding photon spectrum. The exact solution to (3) is known only in a few relevant cases. Fortunately, all applications to relativistic heavy-ion collisions involve photons whose momenta kk are so large that one can use the adiabatic approximation, thereby obtaining a general expression for the field amplitude in § 4. The main results are derived in § 5 and Sec. VI where the photon spectra for the models (i) and (ii) are obtained and analyzed. The results are exhibited in Figure 2, Figure 3,Figure 4 and Figure 5. The final section summarizes and discusses the results.

Figure 1. The photon spectrum for a constant b0=10b_0=10 MeV. The incident fermion energy E=20E=20 GeV, mass m=0.3m=0.3 GeV and unit charge. The Lorentzian parameter ϵ=1\epsilon=1 MeV. Only the right-handed photons λ=1\lambda=1 are emitted.

Figure 2. The photon spectrum for b0(t)=A1+B1tanh⁡tτb_0(t)=A_1+B_1\tanh\frac{t}{\tau} with A1=B1=10 MeVA_1=B_1=10\ \mathrm{MeV} and τ=25 GeV−1≈5 fm/c\tau=25\ \mathrm{GeV}^{-1}\approx5\ \mathrm{fm}/c. The incident fermion energy E=20 GeVE=20\ \mathrm{GeV}, mass m=0.3 GeVm=0.3\ \mathrm{GeV}, and unit charge.

Figure 3. The photon spectrum for b0(t)=A1+B1tanh⁡tτb_0(t)=A_1+B_1\tanh\frac{t}{\tau} with A1=−B1=−10 MeVA_1=-B_1=-10\ \mathrm{MeV} (left), A1=0A_1=0, B1=10 MeVB_1=10\ \mathrm{MeV} (right) and τ=25 GeV−1≈5 fm/c\tau=25\ \mathrm{GeV}^{-1}\approx5\ \mathrm{fm}/c. The incident fermion energy E=20 GeVE=20\ \mathrm{GeV}, mass m=0.3 GeVm=0.3\ \mathrm{GeV}, and unit charge. The turquoise color indicates that the spectrum is independent of the photon’s polarization.

Figure 4. The photon spectrum for b0(t)=A1+B1tanh⁡t/τb_0(t)=A_1+B_1\tanh t/\tau with A1=15A_1=15 MeV, B1=5B_1=5 MeV τ=25\tau=25 GeV−1≈5^{-1}\approx5 fm/c. The incident fermion energy E=20E=20 GeV, mass m=0.3m=0.3 GeV, and unit charge.

Figure 5. The photon spectrum for b0(t)=A(t/τ)/(1+t2/τ2)b_0(t)=A(t/\tau)/(1+t^2/\tau^2) with A2=10A_2=10 MeV and τ=25\tau=25 GeV−1≈5^{-1}\approx5 fm/c. The incident fermion energy E=20E=20 GeV, mass m=0.3m=0.3 GeV, and unit charge. The spectrum does not depend on the photon’s polarization, which is reflected in turquoise color, and does not exhibit any resonances.

Photon Radiation Probability

The electromagnetic field, in the radiation gauge A0=0A^{0}=0, ∇⋅A=0\nabla\cdot\mathbf{A}=0, is described by the vector potential A\mathbf{A} obeying the equation:

∇2A−∂t2A+b0(t)∇×A=0.(1)\nabla^{2}\mathbf{A}-\partial_{t}^{2}\mathbf{A}+b_{0}(t)\nabla\times\mathbf{A}=0 . \tag*{(1)}

It is assumed that the chiral magnetic conductivity b0b_{0} is a given function of time unaffected by the radiation. The wave function of a photon of energy ω\omega and momentum k\mathbf{k} is a plane wave solution to (1):

A(t,r)=1Va(t)ϵλeik⋅r,(2)\mathbf{A}(t,\mathbf{r})=\frac{1}{\sqrt{V}}a(t)\boldsymbol{\epsilon}_{\lambda}e^{i\mathbf{k}\cdot\mathbf{r}}, \tag*{(2)}

where the polarization vector ϵλ\boldsymbol{\epsilon}_{\lambda}, λ=±1\lambda=\pm1, satisfies the condition ik^×ϵλ=λϵλi\hat{\mathbf{k}}\times\boldsymbol{\epsilon}_{\lambda}=\lambda\boldsymbol{\epsilon}_{\lambda} which implies that photons must be circularly polarized. Substituting (2) into (1) one finds that the amplitude a(t)a(t) is governed by the equation

a¨(t)+Ω2(t)a(t)=0,(3)\ddot{a}(t)+\Omega^{2}(t)a(t)=0 , \tag*{(3)}

where

Ω2(t)=k2−λb0(t)k.(4)\Omega^{2}(t)=k^{2}-\lambda b_{0}(t)k . \tag*{(4)}

The photon wave function will be normalized to one photon per unit volume as usual.

Denote the energy and momentum of the incident and final fermions as E,pE,\mathbf{p} and E′,p′E',\mathbf{p}' respectively. The wave function of an incident fermion with momentum p\mathbf{p} and spin ss reads

ψi(t,r)=12EVus(p)e−iEt+ip⋅r,(5)\psi_{i}(t,\mathbf{r})=\frac{1}{\sqrt{2EV}}u_{s}(p)e^{-iEt+i\mathbf{p}\cdot\mathbf{r}}, \tag*{(5)}

where E=p2+m2E=\sqrt{p^{2}+m^{2}}. The wave function of the final fermion ψf\psi_{f} is described by a similar expression. The photon radiation amplitude is given by

Sfi=−ie∫ψˉfγ⋅A∗ψid3xdt=−ieuˉs(p′)γ⋅ϵλ∗us(p)2V3/2EE′(2π)3δ(p−p′−k)a~∗(E−E′),(6)S_{fi}=-ie\int\bar{\psi}_{f}\gamma\cdot A^{*}\psi_{i}d^{3}xdt=-ie\frac{\bar{u}_{s}(p')\gamma\cdot\boldsymbol{\epsilon}_{\lambda}^{*}u_{s}(p)}{2V^{3/2}\sqrt{EE'}}(2\pi)^{3}\delta(\mathbf{p}-\mathbf{p}'-\mathbf{k})\widetilde{a}^{*}(E-E') , \tag*{(6)}

where I defined:

a~(E−E′)=∫−∞+∞e−i(E′−E)ta(t)dt.(7)\widetilde{a}(E-E')=\int_{-\infty}^{+\infty}e^{-i(E'-E)t}a(t)dt . \tag*{(7)}

Given a bounded function b0(t)b_{0}(t), the convergence of the integral in (7) as assured by inserting a positive infinitesimal parameter ϵ\epsilon:

a~(E−E′)=∫0∞e−i(E′−E−iϵ)ta(t)dt+∫−∞0e−i(E′−E+iϵ)ta(t)dt.(8)\widetilde{a}(E-E')=\int_{0}^{\infty}e^{-i(E'-E-i\epsilon)t}a(t)dt+\int_{-\infty}^{0}e^{-i(E'-E+i\epsilon)t}a(t)dt . \tag*{(8)}

The photon radiation probability is then calculated as

dw=12∑λ,s,s′∣Sfi∣2Vd3p′(2π)3Vd3k(2π)3=e28(2π)3E∑λ∣a~(E−E′)∣2d2k⊥dx1−xTr⁡[(p̸+m)ϵλ∗(̸p′+m)ϵλ],(9)dw=\frac{1}{2}\sum_{\lambda,s,s'}|S_{fi}|^{2}\frac{Vd^{3}p'}{(2\pi)^{3}}\frac{Vd^{3}k}{(2\pi)^{3}}=\frac{e^{2}}{8(2\pi)^{3}E}\sum_{\lambda}|\widetilde{a}(E-E')|^{2}d^{2}k_{\perp}\frac{dx}{1-x}\operatorname{Tr}[(\not p+m)\epsilon_{\lambda}^{*}(\not p'+m)\epsilon_{\lambda}] , \tag*{(9)}

where xx denotes the fraction of the incident fermion momentum carried away by the photon. Assuming that the incident fermion moves along the zz-axis, x=kz/pzx=k_{z}/p_{z} and k2=x2pz2+k⊥2k^{2}=x^{2}p_{z}^{2}+k_{\perp}^{2}, where k⊥k_{\perp} is the photon’s transverse momentum. The product of the photon polarization vectors can be written as:

ϵλμ∗ϵλν={12(δij−kikjk2)+iλ2ϵijℓkℓk,μ=i,ν=j,0,μν=0,(10)\epsilon_{\lambda}^{\mu*}\epsilon_{\lambda}^{\nu}= \begin{cases} \frac{1}{2}\left(\delta^{ij}-\frac{k^{i}k^{j}}{k^{2}}\right)+\frac{i\lambda}{2}\epsilon^{ij\ell}\frac{k^{\ell}}{k}, & \mu=i,\nu=j,\\ 0, & \mu\nu=0, \end{cases} \tag*{(10)}

where no summation over λ\lambda is implied.

In most applications the incident fermion can be regarded as ultra-relativistic meaning that pz≫k⊥,mp_z \gg k_\perp,m. In this approximation,

E=pz(1+m22pz2)(11)E=p_z\left(1+\frac{m^2}{2p_z^2}\right) \tag*{(11)}
E′=(1−x)pz(1+m2+k⊥22(1−x)2pz2)(12)E'=(1-x)p_z\left(1+\frac{m^2+k_\perp^2}{2(1-x)^2p_z^2}\right) \tag*{(12)}

and the differential probability (9) becomes:

dwd2k⊥dx=18(2π)3E∑λ∣a~(E−E′)∣22e2x2(1−x)2[k⊥2(2−2x+x2)+m2x4].(13)\frac{dw}{d^2k_\perp dx}=\frac{1}{8(2\pi)^3E}\sum_\lambda\left|\tilde a(E-E')\right|^2\frac{2e^2}{x^2(1-x)^2}\left[k_\perp^2(2-2x+x^2)+m^2x^4\right]. \tag*{(13)}

The problem of computing the photon spectrum thus reduces to determining the Fourier image (8) of the field amplitude a(t)a(t).

Photon spectrum at constant b0b_0

It is instructive to first consider the particular case of constant b0b_0 that was discussed before [8, 9, 22, 23]. In this case the positive frequency solution of (3) is

a=12ωe−iωt,(14)a=\frac{1}{\sqrt{2\omega}}e^{-i\omega t}, \tag*{(14)}

where ω\omega is the photon energy given by

ω=k2+M2≈xpz(1+M2+k⊥22x2pz2),(15)\omega=\sqrt{k^2+M^2}\approx xp_z\left(1+\frac{M^2+k_\perp^2}{2x^2p_z^2}\right), \tag*{(15)}

and I introduced a new parameter M2=−λb0kM^2=-\lambda b_0k. It follows then from (7) that

a~(E−E′)=2πδ(ω+E′−E).(16)\tilde a(E-E')=2\pi\delta(\omega+E'-E). \tag*{(16)}

Using (11) and (12), the argument of the delta-function can be written as

ω+E′−E=k⊥2+M2(1−x)+m2x22x(1−x)pz.(17)\omega+E'-E=\frac{k_\perp^2+M^2(1-x)+m^2x^2}{2x(1-x)p_z}. \tag*{(17)}

The inverse of this expression is the coherence length of the photon emission: ℓc=1/∣ω+E′−E∣\ell_c=1/|\omega+E'-E|. It signifies the distance traversed by the fermion as it radiates a photon carrying the fraction xx of its momentum, the transverse momentum k⊥k_\perp and the polarization λ\lambda. Substituting (17),(16) into (13) yields the differential photon radiation rate dw˙d\dot w at constant b0b_0 [8]:

dw˙dxd2k⊥=e28π2Ex[λEb0(1−x+x22)−m2x]δ(k⊥2−λb0Ex(1−x)+m2x2).(18)\frac{d\dot w}{dx d^2k_\perp}=\frac{e^2}{8\pi^2Ex}\left[\lambda Eb_0\left(1-x+\frac{x^2}{2}\right)-m^2x\right]\delta\left(k_\perp^2-\lambda b_0Ex(1-x)+m^2x^2\right). \tag*{(18)}

The photon spectrum (18) is plotted in Figure 1 for a set of parameters relevant in relativistic heavy-ion collisions. The delta-function is approximated by the Lorentzian: δ(x)≈ϵ/[π(ϵ2+x2)]\delta(x)\approx\epsilon/[\pi(\epsilon^2+x^2)]. The peaks of the spectrum correspond to the values of k⊥k_\perp and xx at which the argument of the delta function in (18) vanishes. At those values the coherence length ℓc\ell_c is divergent (in Figure 1 and the following figures it is regulated by ϵ\epsilon), as it does in the classical Cherenkov radiation. The radiated photons are right or left-handed circularly polarized if b0>0b_0>0 or b0<0b_0<0 respectively.

Throughout the paper, the following color coding is employed: right-handed photons are plotted in blue, left-handed photons are plotted in green, and their overlap is plotted in turquoise.

Adiabatic approximation

Eq. (3) cannot be analytically solved for an arbitrary function b0(t)b_0(t). However, for a sufficiently slow time-variation the solution can be obtained in the adiabatic approximation. In this section, a general formula for the amplitude a(t)a(t) based on this approximation is derived. First, introduce an auxiliary function R(t)R(t) defined as

a(t)=eiR(t).(19)a(t)=e^{iR(t)}. \tag*{(19)}

Substituting (19) into (3) one finds that it satisfies the equation

iR¨−R˙2+Ω2=0.(20)i\ddot{R}-\dot{R}^{2}+\Omega^{2}=0. \tag*{(20)}

The adiabatic approximation consists in treating R(t)R(t) as a slowly varying function such that

∣R¨∣≪R˙2.(21)|\ddot{R}|\ll\dot{R}^{2}. \tag*{(21)}

To develop the adiabatic expansion, introduce a bookkeeping parameter α\alpha into (20):

iαR¨−R˙2+Ω2=0,(22)i\alpha\ddot{R}-\dot{R}^{2}+\Omega^{2}=0, \tag*{(22)}

and substitute in it the expansion

R=∑j=0∞αjRj.(23)R=\sum_{j=0}^{\infty}\alpha^{j}R_j. \tag*{(23)}

By collecting the terms of the same order in α\alpha one obtains equations that govern the functions R0R_0 and R1R_1:

R˙02=Ω2,(24)\dot{R}_0^{2}=\Omega^{2}, \tag*{(24)}
iR˙0=2R˙0R˙1.(25)i\dot{R}_0=2\dot{R}_0\dot{R}_1. \tag*{(25)}

Solving these equations and plugging the result back into (19) yields the desired result, after setting α=1\alpha=1:

a(t)=12Ωe−i∫0tΩ(t′)dt′.(26)a(t)=\frac{1}{\sqrt{2\Omega}}e^{-i\int_0^t\Omega(t')dt'}. \tag*{(26)}

The sign in the exponent corresponds to the outgoing wave as appropriate for the emitted photon. Expanding Ω\Omega in the ultra-relativistic approximation k≫b0k\gg b_0:

Ω(t)≈k−λb0(t)2(27)\Omega(t)\approx k-\frac{\lambda b_0(t)}{2} \tag*{(27)}

leads to the final result for the amplitude:

a(t)=12ke−ikt+iλ2∫tb0(t′) dt′.(28)a(t)=\frac{1}{\sqrt{2k}}e^{-ikt+\frac{i\lambda}{2}\int^t b_0(t')\,dt'} . \tag*{(28)}

It is valid as long as (21) is satisfied which is equivalent to the condition:

∣b˙0∣≪kz2.(29)\left|\dot{b}_0\right|\ll k_z^2 . \tag*{(29)}

Given b0(t)b_0(t) one can compute the amplitude a(t)a(t) using (28) and its Fourier transform (8). The latter can be carried out analytically in a number of cases, but in general one has to rely on numerical methods. The next section deals with two examples which can be solved analytically.

Photon spectrum for b0(t)=A1+B1tanh⁡tτb_0(t)=A_1+B_1\tanh\frac{t}{\tau}

In this section I consider the following model for the time evolution of the chiral magnetic conductivity:

b0(t)=A1+B1tanh⁡tτ.(30)b_0(t)=A_1+B_1\tanh\frac{t}{\tau}. \tag*{(30)}

The initial and final values b0(t)b_0(t) are b0(−∞)=A1−B1b_0(-\infty)=A_1-B_1 and b0(∞)=A1+B1b_0(\infty)=A_1+B_1 respectively. The transition time between them is controlled by the parameter τ\tau.

Exact solution for a(t)a(t)

Substituting (30) into (4) one obtains the function Ω(t)\Omega(t):

Ω2(t)=kz2−λk(A1+B1tanh⁡tτ).(31)\Omega^2(t)=k_z^2-\lambda k(A_1+B_1\tanh\frac{t}{\tau}). \tag*{(31)}

The exact solution of (3) with Ω\Omega given by (31) was obtained in [24]. It is most readily expressed in terms of the asymptotic values of Ω\Omega as follows:

Ωin=kz2−λk(A1−B1),(32)\Omega_{\mathrm{in}}=\sqrt{k_z^2-\lambda k(A_1-B_1)}, \tag*{(32)}
Ωout=kz2−λk(A1+B1),(33)\Omega_{\mathrm{out}}=\sqrt{k_z^2-\lambda k(A_1+B_1)}, \tag*{(33)}
Ω±=12(Ωout±Ωin).(34)\Omega_{\pm}=\frac{1}{2}(\Omega_{\mathrm{out}}\pm\Omega_{\mathrm{in}}). \tag*{(34)}

In the ultra-relativistic approximation, expand (32),(33) and (34) at large kzk_z:

Ωin=kz−λ2(A1−B1)+O(kz−1),(35)\Omega_{\mathrm{in}}=k_z-\frac{\lambda}{2}(A_1-B_1)+O(k_z^{-1}), \tag*{(35)}
Ωout=kz−λ2(A1+B1)+O(kz−1),(36)\Omega_{\mathrm{out}}=k_z-\frac{\lambda}{2}(A_1+B_1)+O(k_z^{-1}), \tag*{(36)}
Ω+=kz−λA12,(37)\Omega_+=k_z-\frac{\lambda A_1}{2}, \tag*{(37)}
Ω−=−λB12.(38)\Omega_-=-\frac{\lambda B_1}{2}. \tag*{(38)}

The solution to (3) describing the emitted photon, namely the one that has positive frequency at the future infinity reads:

a(t)=12Ωoute−iΩ+t−iΩ−τln⁡[2cosh⁡tτ]2F1(1+iΩ−τ,iΩ−τ;1+iΩoutτ;12(1−tanh⁡tτ)),(39)a(t)=\frac{1}{\sqrt{2\Omega_{\mathrm{out}}}}e^{-i\Omega_+t-i\Omega_-\tau\ln[2\cosh\frac{t}{\tau}]}{}_2F_1\left(1+i\Omega_-\tau,i\Omega_-\tau;1+i\Omega_{\mathrm{out}}\tau;\frac{1}{2}\left(1-\tanh\frac{t}{\tau}\right)\right), \tag*{(39)}
∼12kze−i(kz−λA12)t+iλB12τln⁡[2cosh⁡tτ]2F1(1−iλB1τ2,iλB1τ2;1+ikzτ;12(1−tanh⁡tτ)).(40)\sim\frac{1}{\sqrt{2k_z}}e^{-i(k_z-\frac{\lambda A_1}{2})t+i\frac{\lambda B_1}{2}\tau\ln[2\cosh\frac{t}{\tau}]}{}_2F_1\left(1-\frac{i\lambda B_1\tau}{2},\frac{i\lambda B_1\tau}{2};1+ik_z\tau;\frac{1}{2}\left(1-\tanh\frac{t}{\tau}\right)\right). \tag*{(40)}

Using Euler’s hypergeometric transformation: 2F1(α,β;γ;z)=(1−z)γ−α−β2F1(γ−α,γ−β;γ;z){}_2F_1(\alpha,\beta;\gamma;z)=(1-z)^{\gamma-\alpha-\beta}{}_2F_1(\gamma-\alpha,\gamma-\beta;\gamma;z) and assuming, consistently with the ultra-relativistic approximation, that kzτ≫1k_z\tau\gg1 and kz≫∣B1∣k_z\gg\lvert B_1\rvert, so that γ≫α,β\gamma\gg\alpha,\beta, it emerges that the hypergeometric function in (40) reduces to unity.

Adiabatic approximation

The adiabatic approximation (28) precisely reproduces the exact solution (40) which can be seen by substituting (30) into (28) and replacing k≈kzk \approx k_z:

a(t)=12kze−i(kz−λA12)t+iλB12τln⁡[2cosh⁡tτ].(41)a(t)=\frac{1}{\sqrt{2k_z}}e^{-i\left(k_z-\frac{\lambda A_1}{2}\right)t+i\frac{\lambda B_1}{2}\tau\ln\left[2\cosh\frac{t}{\tau}\right]}. \tag*{(41)}

According to (29) this formula is valid when the parameters satisfy kz2τ≫∣B1∣k_z^2\tau\gg\lvert B_1\rvert. Thus the adiabatic approximation accurately describes the time-evolution of the electromagnetic field in the ultra-relativistic limit which spares one seeking exact solutions to (3).

To calculate the Fourier image of the amplitude using (8), cast it in the form:

a~(E−E′)=22kz∫0∞cos⁡[(E−E′−kz+λA12)t]e−(iλB12+ϵ)ln⁡[2cosh⁡tτ]dt.(42)\tilde{a}(E-E')=\frac{2}{\sqrt{2k_z}}\int_0^\infty\cos\left[\left(E-E'-k_z+\frac{\lambda A_1}{2}\right)t\right]e^{-\left(i\frac{\lambda B_1}{2}+\epsilon\right)\ln\left[2\cosh\frac{t}{\tau}\right]}dt. \tag*{(42)}

I used the fact that at t→∞t\to\infty, the infinitesimal term in the exponent becomes −ϵt/τ-\epsilon t/\tau as prescribed by (8) (up to the non-essential factor 1/τ1/\tau that simplify redefines ϵ\epsilon). Employing (35)–(38), (42) may be written as:

a~(E−E′)=22Ωout∫0∞cos⁡[(E−E′−Ω+)t]e−(iΩ−τ+ϵ)ln⁡[2cosh⁡tτ]dt.(43)\tilde{a}(E-E')=\frac{2}{\sqrt{2\Omega_{\mathrm{out}}}}\int_0^\infty\cos\left[(E-E'-\Omega_+)t\right]e^{-(i\Omega_-\tau+\epsilon)\ln\left[2\cosh\frac{t}{\tau}\right]}dt. \tag*{(43)}

Performing the Fourier transform using Eq. 1.9(5) of [25] yields:

a~(E−E′)=12Ωout2iΩ−τ−1τΓ(iΩ−τ)Γ(iτ2(E−E′−Ωin−iϵ))Γ(iτ2(−E+E′+Ωout−iϵ)).(44)\tilde{a}(E-E')=\frac{1}{\sqrt{2\Omega_{\mathrm{out}}}}\frac{2^{i\Omega_-\tau-1}\tau}{\Gamma(i\Omega_-\tau)}\Gamma\left(\frac{i\tau}{2}(E-E'-\Omega_{\mathrm{in}}-i\epsilon)\right)\Gamma\left(\frac{i\tau}{2}(-E+E'+\Omega_{\mathrm{out}}-i\epsilon)\right). \tag*{(44)}

Substituting (44) into (13) furnishes the photon spectrum:

dωdx d2k⊥=116(2π)3E2∑λ2e2x3(1−x)2[k⊥2(2−2x+x2)+m2x4]τ24∣Γ(iΩ−τ)∣2∣Γ(iτ2(E−E′−Ωin−iϵ))Γ(iτ2(−E+E′+Ωout−iϵ))∣2.(45)\frac{d\omega}{dx\,d^2k_\perp}=\frac{1}{16(2\pi)^3E^2}\sum_\lambda\frac{2e^2}{x^3(1-x)^2}\left[k_\perp^2(2-2x+x^2)+m^2x^4\right]\frac{\tau^2}{4\lvert\Gamma(i\Omega_-\tau)\rvert^2} \left\lvert\Gamma\left(\frac{i\tau}{2}(E-E'-\Omega_{\mathrm{in}}-i\epsilon)\right)\Gamma\left(\frac{i\tau}{2}(-E+E'+\Omega_{\mathrm{out}}-i\epsilon)\right)\right\rvert^2. \tag*{(45)}

The explicit dependence of the Γ\Gamma-function arguments on photon momentum is:

E′−E+Ωin=k⊥2−λx(1−x)pz(A1−B1)+m2x22x(1−x)pz,(46)E'-E+\Omega_{\mathrm{in}}=\frac{k_\perp^2-\lambda x(1-x)p_z(A_1-B_1)+m^2x^2}{2x(1-x)p_z}, \tag*{(46)}
E′−E+Ωout=k⊥2−λx(1−x)pz(A1+B1)+m2x22x(1−x)pz.(47)E'-E+\Omega_{\mathrm{out}}=\frac{k_\perp^2-\lambda x(1-x)p_z(A_1+B_1)+m^2x^2}{2x(1-x)p_z}. \tag*{(47)}

These expressions vanish at certain values of k⊥k_\perp and xx, which correspond to the resonances in the spectrum. The height and width of these resonances are regulated by the parameter ϵ\epsilon.

To plot the photon spectrum, select the values of the parameters relevant to the phenomenology of quark-gluon plasma produced in relativistic heavy-ion collisions. The chiral magnetic conductivity is estimated to be within the range of 10–100 MeV [26, 27, 28, 29]. Assuming that the time-dependence of b0b_0 is predominantly due to sphaleron transitions, the typical value of τ\tau is 1/αs2T1/\alpha_s^2T [10], in the range of 1–10 fm, shorter that the plasma’s lifetime.

Figs. 2, 3 and 4 present the photon spectrum for typical parameters. They describe the photon spectrum for various values of b0(∞)b_0(\infty) and b0(−∞)b_0(-\infty). The resonances correspond to vanishing of expressions (46) and (47) which occurs only if λ(A1−B1)=λb0(−∞)>0\lambda(A_1-B_1)=\lambda b_0(-\infty)>0 and λ(A1+B1)=λb0(∞)>0\lambda(A_1+B_1)=\lambda b_0(\infty)>0 respectively. The positions of the resonances satisfy the equations

k⊥2−λx(1−x)E(A1∓B1)+m2x2=0.(48)k_\perp^2-\lambda x(1-x)E(A_1\mp B_1)+m^2x^2=0. \tag*{(48)}

Clearly, a particular polarization is enhanced depending on the signs of asymptotic values b0(±∞)b_0(\pm\infty). When one of those values vanishes, only single resonance appears in the k⊥k_\perp spectrum, see Figure 2 and Figure 3 (left). The two polarizations overlap in Figure 3(right) because λb0(−∞)=−λb0(∞)\lambda b_0(-\infty)=-\lambda b_0(\infty). In Figure 4 one can see two distinct resonances in the right-handed photon spectrum because both asymptotic values are positive. There are twice as many resonances in the xx-spectrum as compared to the k⊥k_\perp-spectrum as is evident from (46) and (47). This is seen in Figure 2(right).

The resonant value of k⊥k_{\perp} at a given xx is

K⊥=λx(1−x)E(A1∓B1)−m2x2,(49)K_{\perp}=\sqrt{\lambda x(1-x)E(A_1\mp B_1)-m^2x^2}, \tag*{(49)}

provided that

0<x≤λE(A1∓B1)λE(A1∓B1)+m2,(50)0<x\le\frac{\lambda E(A_1\mp B_1)}{\lambda E(A_1\mp B_1)+m^2}, \tag*{(50)}

while the resonant values of xx at a given k⊥k_{\perp} are:

Xα=λE(A1∓B1)+(−1)αE2(A1∓B1)2−4[m2+λE(A1∓B1)]k⊥22[m2+λE(A1∓B1)],α=1,2,(51)X_{\alpha}=\frac{\lambda E(A_1\mp B_1)+(-1)^{\alpha}\sqrt{E^2(A_1\mp B_1)^2-4[m^2+\lambda E(A_1\mp B_1)]k_{\perp}^2}}{2[m^2+\lambda E(A_1\mp B_1)]},\qquad\alpha=1,2, \tag*{(51)}

provided that

0<k⊥2≤E2(A1∓B1)24[m2+λE(A1∓B1)].(52)0<k_{\perp}^2\le\frac{E^2(A_1\mp B_1)^2}{4[m^2+\lambda E(A_1\mp B_1)]}. \tag*{(52)}

Sudden transition

The adiabatic approximation does not require τ\tau to be very large, as was shown in the previous subsection. In fact, it can be rather small as long as the condition kz2τ≫∣B1∣k_z^2\tau\gg|B_1| is fulfilled. In particular, τ\tau can be much smaller than the coherence length ℓc\ell_c defined in § 3 for constant b0b_0. In our case, 1/τ1/\tau must be much larger than either of two expressions (46) and (47). This condition is certainly satisfied near the resonances. It also holds for not too large values of k⊥k_\perp. One can refer to such a scenario as a sudden transition. Taking the limit τ→0\tau\to0 in the exponent of (41) yields

a(t)=12Ωout(e−iΩouttθ(t)+e−iΩintθ(−t)).(53)a(t)=\frac{1}{\sqrt{2\Omega_{\mathrm{out}}}}\left(e^{-i\Omega_{\mathrm{out}}t}\theta(t)+e^{-i\Omega_{\mathrm{in}}t}\theta(-t)\right). \tag*{(53)}

Its Fourier-transform (7) is

a~(E−E′)=12Ωout{∫0∞ei(E−E′−Ωout+iϵ)t dt+∫−∞0ei(E−E′−Ωin−iϵ)t dt}(54)\widetilde{a}(E-E')=\frac{1}{\sqrt{2\Omega_{\mathrm{out}}}}\left\{\int_{0}^{\infty}e^{i(E-E'-\Omega_{\mathrm{out}}+i\epsilon)t}\,dt+\int_{-\infty}^{0}e^{i(E-E'-\Omega_{\mathrm{in}}-i\epsilon)t}\,dt\right\} \tag*{(54)}
=i2Ωout{1E−E′−Ωout+iϵ−1E−E′−Ωin−iϵ}.(55)=\frac{i}{\sqrt{2\Omega_{\mathrm{out}}}}\left\{\frac{1}{E-E'-\Omega_{\mathrm{out}}+i\epsilon}-\frac{1}{E-E'-\Omega_{\mathrm{in}}-i\epsilon}\right\}. \tag*{(55)}

In particular, if b0b_0 is constant, then B1=0B_1=0, Ωout=Ωin\Omega_{\mathrm{out}}=\Omega_{\mathrm{in}} and (55) reduces to (16) as can be seen by using the formula

1x±iϵ=∓iπδ(x)+P1x.(56)\frac{1}{x\pm i\epsilon}=\mp i\pi\delta(x)+P\frac{1}{x}. \tag*{(56)}

Photon spectrum for b(t)=A2(t/τ)/(1+t2/τ2)b(t)=A_2(t/\tau)/(1+t^2/\tau^2)

In this section, consider a scenario in which b0b_0 evolves in such a way that its value at distant past and future vanish. It can be described by the following model:

b(t)=A2tτ(1+t2τ2).(57)b(t)=\frac{A_2t}{\tau\left(1+\frac{t^2}{\tau^2}\right)}. \tag*{(57)}

Unlike the previous model, the exact solution to equation (3) with (57) is not known. I will therefore employ the adiabatic approximation that was developed in § 4 and successfully used in the previous section. Accordingly, plugging (57) into (28) yields

a(t)=12ke−ikt(1+t2τ2)iλτA24.(58)a(t)=\frac{1}{\sqrt{2k}}e^{-ikt}\left(1+\frac{t^2}{\tau^2}\right)^{\frac{i\lambda\tau A_2}{4}}. \tag*{(58)}

Its Fourier transform is given by

a~(E−E′)=12k∫0∞e−i(E′−E+k−iϵ)t(1+t2τ2)iλτA24dt+12k∫−∞0e−i(E′−E+k+iϵ)t(1+t2τ2)iλτA24dt(59)\tilde{a}(E-E')=\frac{1}{\sqrt{2k}}\int_{0}^{\infty}e^{-i(E'-E+k-i\epsilon)t}\left(1+\frac{t^2}{\tau^2}\right)^{\frac{i\lambda\tau A_2}{4}}dt+\frac{1}{\sqrt{2k}}\int_{-\infty}^{0}e^{-i(E'-E+k+i\epsilon)t}\left(1+\frac{t^2}{\tau^2}\right)^{\frac{i\lambda\tau A_2}{4}}dt \tag*{(59)}
=22k∫0∞(1+t2τ2)iλτA24cos⁡[(E′−E+k)t]e−ϵtdt.(60)=\frac{2}{\sqrt{2k}}\int_{0}^{\infty}\left(1+\frac{t^2}{\tau^2}\right)^{\frac{i\lambda\tau A_2}{4}}\cos\left[(E'-E+k)t\right]e^{-\epsilon t}dt . \tag*{(60)}

This integral can be taken using Mathematica 14.2 package and the result expressed in terms of the function:

Fν(x)=x12+iν[νJ−12−iν(x∗)Γ(−iν)−J12+iν(x∗)Γ(1−iν)sinh⁡(πν)+νcosh⁡(πν)Γ(−iν)H12+iν(x∗)],(61)F_{\nu}(x)=x^{\frac{1}{2}+i\nu}\left[\nu J_{-\frac{1}{2}-i\nu}(x^{*})\Gamma(-i\nu)-J_{\frac{1}{2}+i\nu}(x^{*})\Gamma(1-i\nu)\sinh(\pi\nu)+\nu\cosh(\pi\nu)\Gamma(-i\nu)H_{\frac{1}{2}+i\nu}(x^{*})\right], \tag*{(61)}

where ν=λτA24\nu=\frac{\lambda\tau A_2}{4} and HH is Struve’s function. Thus,

a~(E−E′)=π322−12+iν2kτ2iν(y2+ϵ2)−12−iνcosh⁡(πν)sinh⁡(πν)Γ(1−iν)Γ(−iν){Fν[(iy+ϵ)τ]+Fν[(−iy+ϵ)τ]},(62)\tilde{a}(E-E')=\frac{\pi^{\frac{3}{2}}2^{-\frac{1}{2}+i\nu}}{\sqrt{2k}\tau^{2i\nu}}\frac{(y^2+\epsilon^2)^{-\frac{1}{2}-i\nu}}{\cosh(\pi\nu)\sinh(\pi\nu)\Gamma(1-i\nu)\Gamma(-i\nu)}\left\{F_{\nu}\left[(iy+\epsilon)\tau\right]+F_{\nu}\left[(-iy+\epsilon)\tau\right]\right\}, \tag*{(62)}

where

y=E′−E+k=k⊥2+m2x22x(1−x)pz.(63)y=E'-E+k=\frac{k_{\perp}^{2}+m^{2}x^{2}}{2x(1-x)p_z}. \tag*{(63)}

In the sudden transition limit τ→0\tau\to0 (62) vanishes. Indeed,

lim⁡τ→0a~(E−E′)=i2k(1E′−E+k+iϵ−1E′−E+k−iϵ)=12k2πδ(y)=0,(64)\lim_{\tau\to0}\tilde{a}(E-E')=\frac{i}{\sqrt{2k}}\left(\frac{1}{E'-E+k+i\epsilon}-\frac{1}{E'-E+k-i\epsilon}\right)=\frac{1}{\sqrt{2k}}2\pi\delta(y)=0, \tag*{(64)}

because y>0y>0. Substituting (62) into (13) furnishes the photon spectrum:

dwdxd2k⊥=116(2π)3E2∑λ2e2x3(1−x)2[k⊥2(2−2x+x2)+m2x4]×πτ2cosh⁡2(πν)∣Fν[(iy+ϵ)τ]+Fν[(−iy+ϵ)τ]∣2y2+ϵ2.(65)\frac{d w}{d x d^{2}k_{\perp}}=\frac{1}{16(2\pi)^{3}E^{2}}\sum_{\lambda}\frac{2e^{2}}{x^{3}(1-x)^{2}}\left[k_{\perp}^{2}(2-2x+x^{2})+m^{2}x^{4}\right]\times\frac{\pi\tau}{2\cosh^{2}(\pi\nu)}\frac{\left|F_{\nu}\left[(iy+\epsilon)\tau\right]+F_{\nu}\left[(-iy+\epsilon)\tau\right]\right|^{2}}{y^{2}+\epsilon^{2}}. \tag*{(65)}

The numerical result representing the parameters of the quark-gluon plasma is shown in Figure 5. The most conspicuous feature of the spectrum is the equality of the photon polarizations indicating that the λ\lambda-dependence cancels out. This happens because ∣Fν+F−ν∣2\left|F_{\nu}+F_{-\nu}\right|^{2} is an even function of ν\nu and, consequently, of λ\lambda. Also, it is noteworthy that the resonances are absent, which aligns with the observations in the preceding section, where it was observed that the emergence of the resonances is contingent upon finite values of at least one of the asymptotics of b0b_0.

Summary

The chiral Cherenkov mechanism of photon radiation was investigated using two models of the time-evolution of the chiral magnetic conductivity b0(t)b_0(t). The first model b0(t)=A1+B1tanh⁡(t/τ)b_0(t)=A_1+B_1\tanh(t/\tau) describes a transition of b0b_0 between two asymptotically constant values b0(±∞)b_0(\pm\infty). This mechanism allows the production of only circularly polarized photons. In general, the cross sections for the two polarizations are significantly different. The main feature of the k⊥k_{\perp}-spectrum is the potential emergence of one or two resonances. The necessary conditions of the resonance emergence are the fulfillment of either or both of the following conditions: λb0(∞)>0\lambda b_0(\infty)>0 and λb0(−∞)>0\lambda b_0(-\infty)>0. The positions of the resonances are given by (49)–(52). The typical spectra are presented in Figure 2,Figure 3 and Figure 4. In contrast, only a single resonance may appear when b0b_0 is constant as illustrated in Figure 1.

The second model represents a pulse of chirality with a finite duration: b0(t)=A2(t/τ)/(1+t2/τ2)b_0(t)=A_2(t/\tau)/(1+t^2/\tau^2). In this case, there are no resonances, and the two polarizations have identical spectra.

The two models investigated in this paper suggest that the time-dependence of the chiral chemical potential diminishes the average polarization of the chiral Cherenkov radiation as compared to the constant b0b_0 case. The chiral Cherenkov resonance, appearing in the figures as a peak at a certain k⊥k_{\perp} (or two peaks at the corresponding xx’s), gives the largest contribution to the rate and emerges if b0(±∞)≠0b_0(\pm\infty)\ne0. It also determines the average polarization. The mathematical structure of the Fourier image of the amplitude a(t)a(t) suggests that these conclusions may hold universally. Indeed, the parameter ϵ\epsilon that controls the convergence of the time-integral at t→±∞t\to\pm\infty in (8) also sets the resonance height, which is proportional to 1/ϵ1/\epsilon. The resonance appears in any model in which b0b_0 is finite at either remote past or future.

Our calculations significantly relied on the adiabatic approximation developed in § 4. This approximation enabled us to analytically compute the photon amplitude and its Fourier transform. It holds when kz2≫∣b0∣/τk_z^2\gg\lvert b_0\rvert/\tau. For b0b_0 in the range of 10–100 MeV [26, 27, 28, 29] and τ\tau in the range of 1–10 fm, [10, 11] it implies conservatively that k≫100k\gg100 MeV. The validity of the adiabatic approximation in a wide range of photon energies further indicates that our main results—the qualitative structure of the photon spectrum—are not model-dependent. Nevertheless, further investigation of the model space, potentially employing numerical methods, would be of great interest.

I implicitly assumed that the medium lifetime and spatial dimensions are much larger than any time or distance scale associated with sphaleron transitions and the process f→f+γf\to f+\gamma. This assumption might be in tension with the relativistic heavy-ion phenomenology. It would be interesting to investigate how the finite dimensions of the realistic quark-gluon plasma impact the results presented in this paper.

Our analysis focused exclusively on applications to the quark-gluon plasma. Nevertheless, the developed formalism is equally applicable to describe photon radiation emanating from cosmic rays in the background of the axion field [30] or by electrons in Weyl semimetals [31].

Acknowledgments

This work was supported in part by the U.S. Department of Energy under Grant No. DE-SC0023692.

References

  1. [1]D. Kharzeev, Parity violation in hot QCD: Why it can happen, and how to look for it, Phys. Lett. B 633, 260 (2006), arXiv:hep-ph/0406125.DOI
  2. [2]D. E. Kharzeev, L. D. McLerran, and H. J. Warringa, The Effects of topological charge change in heavy ion collisions: ’Event by event P and CP violation’, Nucl. Phys. A 803, 227 (2008), arXiv:0711.0950 [hep-ph].DOI
  3. [3]D. E. Kharzeev, Topologically induced local P and CP violation in QCD x QED, Annals Phys. 325, 205 (2010), arXiv:0911.3715 [hep-ph].DOI
  4. [4]D. Kharzeev and A. Zhitnitsky, Charge separation induced by P-odd bubbles in QCD matter, Nucl. Phys. A 797, 67 (2007), arXiv:0706.1026 [hep-ph].DOI
  5. [5]K. Fukushima, D. E. Kharzeev, and H. J. Warringa, The Chiral Magnetic Effect, Phys. Rev. D 78, 074033 (2008), arXiv:0808.3382 [hep-ph].DOI
  6. [6]T. Lappi and S. Schlichting, Linearly polarized gluons and axial charge fluctuations in the Glasma, Phys. Rev. D 97, 034034 (2018), arXiv:1708.08625 [hep-ph].DOI
  7. [7]Charge Separation Measurements in Au+Au collisions at √sNN = 7.7–200 GeV in Search of the Chiral Magnetic Effect, (2025), arXiv:2506.00275 [nucl-ex].
  8. [8]K. Tuchin, Radiative instability of quantum electrodynamics in chiral matter, Phys. Lett. B 786, 249 (2018), arXiv:1806.07340 [hep-ph].DOI
  9. [9]X.-G. Huang and K. Tuchin, Transition Radiation as a Probe of the Chiral Anomaly, Phys. Rev. Lett. 121, 182301 (2018), arXiv:1808.00635 [hep-ph].DOI
  10. [10]P. B. Arnold, D. Son, and L. G. Yaffe, The Hot baryon violation rate is O (alpha-w**5 T**4), Phys. Rev. D 55, 6264 (1997), arXiv:hep-ph/9609481.DOI
  11. [11]G. D. Moore and M. Tassler, The Sphaleron Rate in SU(N) Gauge Theory, JHEP 02, 105, arXiv:1011.1167 [hep-ph].
  12. [12]M. Hongo, X.-G. Huang, M. Kaminski, M. Stephanov, and H.-U. Yee, Spin relaxation rate for heavy quarks in weakly coupled QCD plasma, JHEP 08, 263, arXiv:2201.12390 [hep-th].DOI
  13. [13]M. Joyce and M. E. Shaposhnikov, Primordial magnetic fields, right-handed electrons, and the Abelian anomaly, Phys. Rev. Lett. 79, 193 (1997), arXiv:astro-ph/9703005.
  14. [14]A. Boyarsky, J. Frohlich, and O. Ruchayskiy, Self-consistent evolution of magnetic fields and chiral asymmetry in the early Universe, Phys. Rev. Lett. 108, 031301 (2012), arXiv:1109.3350 [astro-ph.CO].DOI
  15. [15]Y. Hirono, D. Kharzeev, and Y. Yin, Self-similar inverse cascade of magnetic helicity driven by the chiral anomaly, Phys. Rev. D 92, 125031 (2015), arXiv:1509.07790 [hep-th].
  16. [16]Y. Akamatsu and N. Yamamoto, Chiral Plasma Instabilities, Phys. Rev. Lett. 111, 052002 (2013), arXiv:1302.2125 [nucl-th].DOI
  17. [17]K. Tuchin, Electromagnetic field and the chiral magnetic effect in the quark-gluon plasma, Phys. Rev. C 91, 064902 (2015), arXiv:1411.1363 [hep-ph].DOI
  18. [18]P. V. Buividovich and M. V. Ulybyshev, Numerical study of chiral plasma instability within the classical statistical field theory approach, Phys. Rev. D 94, 025009 (2016), arXiv:1509.02076 [hep-th].DOI
  19. [19]C. Manuel and J. M. Torres-Rincon, Dynamical evolution of the chiral magnetic effect: Applications to the quark-gluon plasma, Phys. Rev. D 92, 074018 (2015), arXiv:1501.07608 [hep-ph].DOI
  20. [20]V. P. Kirilin and A. V. Sadofyev, Anomalous Transport and Generalized Axial Charge, Phys. Rev. D 96, 016019 (2017), arXiv:1703.02483 [hep-th].DOI
  21. [21]K. Tuchin, Impact of domain walls on the chiral magnetic effect in hot QCD matter, Phys. Rev. C 97, 064914 (2018), arXiv:1802.09629 [hep-ph].DOI
  22. [22]K. Tuchin, Chiral Cherenkov and chiral transition radiation in anisotropic matter, Phys. Rev. D 98, 114026 (2018), arXiv:1809.08181 [hep-ph].DOI
  23. [23]E. Barredo-Alamilla, L. F. Urrutia, and M. M. Ferreira, Jr., Electromagnetic radiation in chiral matter: The Cherenkov case, Phys. Rev. D 107, 096024 (2023), arXiv:2305.07963 [hep-ph].
  24. [24]C. Bernard and A. Duncan, Regularization and renormalization of quantum field theory in curved space-time, Annals of physics 107, 201 (1977).DOI
  25. [25]H. Bateman, A. Erdélyi, W. Magnus, and F. Oberhettinger, Tables of integral transforms, Vol. 1 (McGraw-Hill New York, 1954).
  26. [26]D. E. Kharzeev and H. J. Warringa, Chiral Magnetic conductivity, Phys. Rev. D 80, 034028 (2009), arXiv:0907.5007 [hep-ph].DOI
  27. [27]A. Gynther, K. Landsteiner, F. Pena-Benitez, and A. Rebhan, Holographic Anomalous Conductivities and the Chiral Magnetic Effect, JHEP 02, 110, arXiv:1005.2587 [hep-th].DOI
  28. [28]H.-U. Yee, Holographic Chiral Magnetic Conductivity, JHEP 11, 085, arXiv:0908.4189 [hep-th].DOI
  29. [29]H. Li, X.-I. Sheng, and Q. Wang, Electromagnetic fields with electric and chiral magnetic conductivities in heavy ion collisions, Phys. Rev. C 94, 044903 (2016), arXiv:1602.02223 [nucl-th].DOI
  30. [30]A. Saveliev, R. Alves Batista, and F. Mishin, Spectra for reactions in astrophysical electromagnetic cascades with Lorentz invariance violation: The vacuum Cherenkov effect, Phys. Rev. D 111, 083001 (2025), arXiv:2412.17514 [astro-ph.HE].
  31. [31]J. Hansen, K. Ikeda, D. E. Kharzeev, Q. Li, and K. Tuchin, Magnetic Weyl semimetals as a source of circularly polarized THz radiation, Phys. Open 23, 100268 (2025), arXiv:2405.11076 [cond-mat.mtrl-sci].DOI

Paper details

vlnc:W_ngahjpc3
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