betaFeedback

WorksPhysics and Astronomy

Lorentz-Covariant Spectral Bounds from Thermal Quantum Field Theory

Abstract

We derive rigorous Lorentz-covariant bounds on relaxation spectra directly from the analytic structure of retarded Green's functions in thermal quantum field theory, using only causality, unitarity, the Kubo-Martin-Schwinger condition, and Lorentz covariance, without reference to any specific dynamical model. A single rest-frame quasinormal pole is generically smeared into a continuum of excitations in boosted frames, with width set by the maximal signal velocity. We prove that the non-hydrodynamic gap ΓgapΓ_{\mathrm{gap}} transforms as Γ~gap≥Γgap/[γ(1+vvmax)]\tildeΓ_{\mathrm{gap}} \geq Γ_{\mathrm{gap}}/[γ(1 + v v_{\mathrm{max}})], and that the convergence radius of the hydrodynamic gradient expansion satisfies k~c∈[kc/γ(1+vvs),kc/γ(1−vvs)]\tilde{k}_c \in [k_c/γ(1+v v_s), k_c/γ(1-v v_s)] under a boost of velocity vv. We verify the bounds by a numerical quasinormal-mode computation in the N=4{N}=4 super-Yang-Mills plasma: the leading boosted pole moves deeper into the complex plane – the observed relaxation rate increases with boost velocity, in sharp contrast to naive time dilation – while respecting the bound throughout. The results apply non-perturbatively to the quark-gluon plasma, neutron star merger dynamics, and quantum critical systems.

Introduction

Introduction.—The real-time collective dynamics of quantum many-body systems at finite temperature is governed by the pole structure of retarded Green’s functions in the complex frequency plane. Hydrodynamic poles near the origin encode long-lived collective modes; non-hydrodynamic poles encode transient microscopic relaxation [1, 2]. How this pole structure transforms between inertial frames is central to relativistic hydrodynamics [3, 4], heavy-ion phenomenology [5, 6], and the interpretation of holographic transport [7, 8].

Recently, Gavassino [9, 10] proved that in linearized relativistic (kinetic or rheological) theories with an Onsager-type symmetry, Lorentz boosts generically split a single rest-frame relaxation mode into a continuum of excitations, with width set by the maximal signal speed, and derived rigorous bounds confining the boosted relaxation spectrum in terms of rest-frame spectral data alone. Those proofs operate at the level of classical linearized equations with Hermitian structure. Here we establish analogous bounds in full thermal quantum field theory (QFT), replacing the model-specific symmetry assumptions with the universal properties of retarded two-point functions: causality, unitarity, the Kubo–Martin–Schwinger (KMS) condition, and Lorentz covariance. Our bounds are therefore non-perturbative and hold for strongly coupled systems — including holographic quasinormal spectra, which lie outside the symmetry class of Refs. [9, 10] and whose covariant treatment was posed there as an open problem. Full proofs and extended analysis appear in the companion paper [11].

The central results are: (i) a spectral smearing theorem showing how rest-frame poles broaden under boosts; (ii) a covariant lower bound on the non-hydrodynamic gap; (iii) a covariant upper bound on the maximal relaxation rate; and (iv) covariant bounds on the convergence radius of the hydrodynamic gradient expansion. We verify these in holographic examples and discuss applications to the quark-gluon plasma (QGP) and neutron star mergers.

Setup: retarded Green’s functions and quasinormal modes

Setup: retarded Green’s functions and quasinormal modes.—Let O(x)\mathcal{O}(x) be a local Hermitian bosonic operator in a relativistic QFT at temperature T=1/βT=1/\beta and chemical potential μ\mu. The retarded Green’s function

GR(ω,k)=∫d4x eiωt−ik⋅x(−i)θ(t)⟨[O(x),O(0)]⟩βG^{R}(\omega,\mathbf{k})=\int d^{4}x\,e^{i\omega t-i\mathbf{k}\cdot\mathbf{x}}(-i)\theta(t)\langle[\mathcal{O}(x),\mathcal{O}(0)]\rangle_{\beta}

is analytic for Im⁡(ω)>0\operatorname{Im}(\omega)>0 by causality, and admits the Kramers–Kronig representation

GR(ω,k)=∫−∞∞dω′ ρ(ω′,k)ω′−ω−i0+.G^{R}(\omega,\mathbf{k})=\int_{-\infty}^{\infty}d\omega'\,\frac{\rho(\omega',\mathbf{k})}{\omega'-\omega-i0^{+}}.

where ρ(ω,k)=−1πIm⁡GR(ω+i0+,k)≥0\rho(\omega,\mathbf{k})=-\frac{1}{\pi}\operatorname{Im}G^{R}(\omega+i0^{+},\mathbf{k})\ge0 for ω>0\omega>0 by unitarity. The KMS condition fixes the Wightman function via G>(ω,k)=2πρ(ω,k)/(1−e−βω)G^{>}(\omega,\mathbf{k})=2\pi\rho(\omega,\mathbf{k})/(1-e^{-\beta\omega}) (for an operator carrying U(1)U(1) charge qq, with ω→ω−qμ\omega\to\omega-q\mu).

The quasinormal mode (QNM) frequencies {ωn(k)}\{\omega_n(\mathbf{k})\} are poles of GRG^{R} in the lower half ω\omega-plane. Hydrodynamic modes satisfy ωn(k)→0\omega_n(\mathbf{k})\to0 as k→0\mathbf{k}\to0; non-hydrodynamic modes have a gap

\Gamma_{\mathrm{gap}}\equiv\min_{n\in\mathrm{non\mathchar`-hydro}}\left(-\operatorname{Im}[\omega_n(0)]\right)>0.

The convergence radius kck_c of the hydrodynamic gradient expansion ωhydro(k)=∑ncn∣k∣n\omega_{\mathrm{hydro}}(\mathbf{k})=\sum_n c_n|\mathbf{k}|^n is set by the collision of a hydrodynamic pole with the nearest non-hydrodynamic pole in the complex kk-plane [12, 13].

Spectral bounds under Lorentz boosts

Spectral bounds under Lorentz boosts.—For a Lorentz boost with velocity v\mathbf{v} (γ=(1−v2)−1/2\gamma=(1-v^2)^{-1/2}), four-momentum transforms as ω~=γ(ω−v⋅k)\widetilde{\omega}=\gamma(\omega-\mathbf{v}\cdot\mathbf{k}) and k~=k+(γ−1)v^(v^⋅k)−γvω\widetilde{\mathbf{k}}=\mathbf{k}+(\gamma-1)\hat{\mathbf{v}}(\hat{\mathbf{v}}\cdot\mathbf{k})-\gamma\mathbf{v}\omega. For a scalar operator, Lorentz covariance gives G~R(p~)=GR(Λ−1p~)\widetilde{G}^{R}(\widetilde{p})=G^{R}(\Lambda^{-1}\widetilde{p}), and thus

ρ~(ω~,k~)=ρ(ω~−v~⋅k~γ,k~⊥γ,k~∥−v~ω~γ),(1)(1)\widetilde{\rho}(\widetilde{\omega},\widetilde{\mathbf{k}})=\rho\left(\frac{\widetilde{\omega}-\widetilde{\mathbf{v}}\cdot\widetilde{\mathbf{k}}}{\gamma},\frac{\widetilde{\mathbf{k}}_{\perp}}{\gamma},\frac{\widetilde{k}_{\parallel}-\widetilde{v}\widetilde{\omega}}{\gamma}\right), \qquad(1) \tag*{(1)}

for a longitudinal boost with k~∥=v^⋅k~\tilde{k}_{\parallel}=\hat{v}\cdot\tilde{k}. Equation (1) is exact; it shows that the boosted spectral weight at frequency ω~\tilde{\omega} is sourced by the rest-frame spectral function at boosted spatial momentum k~∥rest=(k~∥−vω~)/γ\tilde{k}_{\parallel}^{\mathrm{rest}}=(\tilde{k}_{\parallel}-v\tilde{\omega})/\gamma. This ω~\tilde{\omega}-dependent spatial momentum is the mechanism by which a single rest-frame pole becomes a continuum.

Theorem 1 (Spectral Smearing). A rest-frame QNM pole at ω=ω0−iΓ0\omega=\omega_0-i\Gamma_0 (Γ0>0\Gamma_0>0) maps, in the boosted frame at k~=0\tilde{k}=0, to a continuum of singularities distributed over the imaginary-frequency interval

Im⁡(ω~pole)∈[−Γ0γ(1−vvmax⁡),−Γ0γ(1+vvmax⁡)],(2)\operatorname{Im}(\tilde{\omega}_{\mathrm{pole}})\in\left[-\frac{\Gamma_0}{\gamma(1-vv_{\max})},-\frac{\Gamma_0}{\gamma(1+vv_{\max})}\right], \tag*{(2)}

where vmax⁡v_{\max} is the front velocity of the theory—the asymptotic characteristic speed vmax⁡=lim⁡∣k∣→∞Re⁡ωn(k)/kv_{\max}=\lim_{\lvert k\rvert\to\infty}\operatorname{Re}\omega_n(k)/k fixed by the highest-derivative (principal-symbol) terms. For a relativistic QFT, microcausality—vanishing of ⟨[O(x),O(0)]⟩\langle[O(x),O(0)]\rangle at spacelike separation—guarantees vmax⁡≤1v_{\max}\leq1.

We stress that vmax⁡v_{\max} is the front velocity, not the group velocity ∂Re⁡ωn/∂k\partial\operatorname{Re}\omega_n/\partial k: the latter is not bounded by causality and can diverge at finite real kk where two damped modes collide (as in the telegrapher/Cattaneo equation [9, 10]), while the front velocity, controlling the sharpest wavefront, is causally bounded.

Proof sketch. Boosted poles at k~=0\tilde{k}=0 solve γω~=ωn(γvω~)\gamma\tilde{\omega}=\omega_n(\gamma v\tilde{\omega}), which requires the rest-frame dispersion relation at complex spatial momentum k=γvω~k=\gamma v\tilde{\omega}. Using analyticity of ωn(k)\omega_n(k) in a complex-momentum strip together with the front-velocity bound ∣Re⁡ωn(k)∣≤vmax⁡∣k∣+O(1)\lvert\operatorname{Re}\omega_n(k)\rvert\leq v_{\max}\lvert k\rvert+O(1), the boosted imaginary part is confined to the interval (2), whose endpoints correspond to co- and counter-propagating fronts. For linearized theories with the Hermitian (Onsager) structure of Refs. [9, 10] this is established rigorously via the principal symbol; in the general QFT setting it rests on the stated analyticity of the thermal correlator in kk [1, 2], and we verify it directly in holography below. □

Simple time dilation (Γ~0=Γ0/γ\tilde{\Gamma}_0=\Gamma_0/\gamma) is recovered only if vmax⁡=0v_{\max}=0, i.e., for a mode with no spatial dispersion. For any propagating mode, the pole is smeared into a band.

Theorem 2 (Non-Hydrodynamic Gap Bound). The non-hydrodynamic gap in a frame boosted with velocity vv satisfies

Γ~gap≥Γgapγ(1+vvmax⁡).(3)\tilde{\Gamma}_{\mathrm{gap}}\geq\frac{\Gamma_{\mathrm{gap}}}{\gamma(1+vv_{\max})}. \tag*{(3)}

Proof. By Theorem 1, the minimal imaginary part of any non-hydrodynamic mode in the boosted frame is achieved by the co-propagating mode, giving Γ~n=Γn/[γ(1+vvmax⁡)]\tilde{\Gamma}_n=\Gamma_n/[\gamma(1+vv_{\max})]. Taking the minimum over all non-hydrodynamic modes and using Γn≥Γgap\Gamma_n\geq\Gamma_{\mathrm{gap}} yields (3). □

Theorem 3 (Maximal Relaxation Rate Bound). Under the same boost,

Γ~max⁡≤Γmax⁡γ(1−vvmax⁡),(4)\tilde{\Gamma}_{\max}\leq\frac{\Gamma_{\max}}{\gamma(1-vv_{\max})}, \tag*{(4)}

provided the spectral weight satisfies the Lebesgue integrability condition of the Kramers–Kronig representation. Bounds (3) and (4) together confine the non-hydrodynamic QNM spectrum in the boosted frame to a strip of width [γ(1±vvmax⁡)]−1×[Γgap,Γmax⁡][\gamma(1\pm vv_{\max})]^{-1}\times[\Gamma_{\mathrm{gap}},\Gamma_{\max}]. The upper bound (4) coincides with the classical-theory bound of Ref. [9] [Eq. (24) there, with w=vmax⁡w=v_{\max}]; our lower bound (3) is tighter than its classical counterpart Γgap(1−vvmax⁡)/γ\Gamma_{\mathrm{gap}}(1-vv_{\max})/\gamma, a difference that traces to the analyticity assumptions of the QFT derivation and merits further scrutiny.

Corollary. As v→vmax⁡v\to v_{\max}, the lower edge of the strip approaches Γgap1−vmax⁡2/(1+vmax⁡2)\Gamma_{\mathrm{gap}}\sqrt{1-v_{\max}^2}/(1+v_{\max}^2), and the smearing band attains its maximal fractional width. For vmax⁡=1v_{\max}=1, the upper edge Γ~max⁡→∞\tilde{\Gamma}_{\max}\to\infty as v→1v\to1, signaling the appearance of a branch-cut continuum in the boosted spectral function.

Hydrodynamic gradient expansion. ::: theorem #thm:4 (4) Theorem 4 (Convergence Radius Under Boosts). Let kck_c be the convergence radius of the hydrodynamic gradient expansion in the rest frame, and let vsv_s be the speed of sound. Under a longitudinal boost with velocity vv,

kcγ(1+vvs)≤k~c≤kcγ(1−vvs).(5)\frac{k_c}{\gamma(1+vv_s)}\leq\tilde{k}_c\leq\frac{k_c}{\gamma(1-vv_s)}. \tag*{(5)}

Proof. The convergence radius kck_c is the modulus of the complex momentum at which the leading non-hydrodynamic pole and the hydrodynamic sound pole collide. Under the boost, the complexified dispersion relation transforms covariantly: ω~hydro(k~)=γ(ωhydro(k)−vk)\tilde{\omega}_{\mathrm{hydro}}(\tilde{k})=\gamma(\omega_{\mathrm{hydro}}(k)-vk) at the same spacetime point. The collision condition in the boosted frame maps the rest-frame collision onto a new complex k~\tilde{k}-value. The bounds (5) follow from the extreme cases in which the collision occurs co-propagating (upper bound) versus counter-propagating (lower bound) relative to the boost direction. □

:::

For conformal fluids (vs=1/3v_s=1/\sqrt{3}), at large boost γ≫1\gamma\gg1 the convergence radius scales as k~c∼kc/(γ3)\tilde{k}_c\sim k_c/(\gamma\sqrt{3}) (lower bound) to kc3/γk_c\sqrt{3}/\gamma (upper bound), confirming that hydrodynamics is restricted to exponentially long wavelengths in the highly boosted frame.

f-sum rule.—The spectral weight redistributes covariantly under boosts. For the shear stress correlator, the f-sum rule in the boosted frame becomes

∫0∞dω~ ω~ ρ~TxyTxy(ω~,0)=πγ22⟨Txx+Tyy⟩β+O(v4),(6)\int_0^\infty d\tilde{\omega}\,\tilde{\omega}\,\tilde{\rho}_{T^{xy}T^{xy}}(\tilde{\omega},0)=\frac{\pi\gamma^2}{2}\langle T^{xx}+T^{yy}\rangle_\beta+O(v^4), \tag*{(6)}

where we used ⟨Ttx⟩β=0\langle T^{tx}\rangle_\beta=0 in equilibrium. The boosted spectral weight is enhanced by γ2\gamma^2 and spread over the broadened frequency range of (2), keeping the total weight finite. We caution that contact terms in the gravitational Ward identity — frequently dropped in Kubo-type analyses — contribute to such moment sum rules and must be retained for covariance [14, 15]; being analytic in frequency, however, they do not affect the singularity locations bounded by (2)–(4).

Figure 1. Boost transformation of the leading scalar-channel quasinormal mode of the N=4\mathcal{N}=4 SYM plasma, computed holographically at zero boosted spatial momentum. (a) Trajectory of the pole in the complex frequency plane as vv increases. (b) The observed relaxation rate −Im⁡ω~-\operatorname{Im}\tilde{\omega} grows with vv, defying naive time dilation (dashed) while respecting the covariant lower bound (7) for vmax⁡=1v_{\max}=1 (dash-dotted); the vmax⁡=1/3v_{\max}=1/\sqrt{3} curve (dotted) is shown for comparison.

Holographic verification.— We verify Theorems 1–3 in the N=4\mathcal{N}=4 super-Yang-Mills (SYM) plasma at strong coupling, dual to AdS5_5-Schwarzschild. We work in the scalar channel of the stress tensor (TxyT^{xy} with transverse momentum), whose retarded correlator is governed by a massless scalar in the black-brane background. Solving the quasinormal spectrum by pseudospectral collocation, the leading QNM at k=0k=0 is ω1/2πT=±1.5597−1.3733i\omega_1/2\pi T=\pm1.5597-1.3733i, so Γgap=1.3733×2πT\Gamma_{\mathrm{gap}}=1.3733\times2\pi T, in agreement with Ref. [16]. Tracking ω1(k)\omega_1(k) over real momenta, its slope ∂kRe⁡ω\partial_k\operatorname{Re}\omega rises monotonically toward unity (0.980 at k=8×2πTk=8\times2\pi T), matching the ω1≈±k\omega_1\approx\pm k asymptotics: the non-hydrodynamic sector has front velocity vmax⁡=1v_{\max}=1, and the operative form of Eq. (3) is

Γ~gap≥1.3733×2πTγ(1+v).(7)\widetilde{\Gamma}_{\mathrm{gap}}\ge\frac{1.3733\times2\pi T}{\gamma(1+v)}. \tag*{(7)}

We compute the boosted poles at k~=0\tilde{k}=0 by solving the fixed-point condition ω1(−γvω~)=γω~\omega_1(-\gamma v\tilde{\omega})=\gamma\tilde{\omega}, evaluating the ingoing boundary-value problem at the required complex spatial momentum. Figure 1 shows the result: the leading pole moves deeper into the lower half-plane, with −Im⁡ω~/2πT-\operatorname{Im}\tilde{\omega}/2\pi T rising from 1.373 at v=0v=0 to 2.29 at v=0.85v=0.85. The observed relaxation rate thus increases with boost velocity — the opposite of the naive time-dilation expectation Γgap/γ\Gamma_{\mathrm{gap}}/\gamma — while the rigorous lower bound (7) is satisfied throughout with a wide margin. Saturation would require spectral weight co-propagating with the boost at vmax⁡v_{\max}; the strongly coupled plasma does not realize this at k~=0\tilde{k}=0, leaving room for the bound to be approached in theories with long-lived luminal excitations.

For Gauss-Bonnet gravity with coupling λGB\lambda_{\mathrm{GB}}, η/s=(1−4λGB)/(4π)\eta/s=(1-4\lambda_{\mathrm{GB}})/(4\pi) is modified while conformal invariance fixes vs2=1/3v_s^2=1/3 exactly. What λGB\lambda_{\mathrm{GB}} does modify is the front velocity vmax⁡v_{\max} of the boundary-theory graviton modes, which for λGB>0\lambda_{\mathrm{GB}}>0 exceeds the Einstein-gravity value and violates boundary causality (vmax⁡>1v_{\max}>1) unless λGB≤9/100\lambda_{\mathrm{GB}}\le9/100 [17]. Substituting vmax⁡(λGB)v_{\max}(\lambda_{\mathrm{GB}}) into (3) and (4) yields λGB\lambda_{\mathrm{GB}}-dependent bounds on the boosted non-hydrodynamic spectrum; the causality constraint on λGB\lambda_{\mathrm{GB}} thus translates directly into a bound on the boosted smearing band.

At finite chemical potential (Reissner-Nordström-AdS), the effective velocity controlling the smearing of the low-lying QNMs decreases with μ/T\mu/T, veff2=13[1−c1(μ/T)2+O((μ/T)4)]v_{\mathrm{eff}}^2=\frac{1}{3}\left[1-c_1(\mu/T)^2+O((\mu/T)^4)\right] with c1>0c_1>0 fixed by the RN-AdS equation of state, and both bounds (3) and (5) tighten toward the time-dilation limit as μ/T\mu/T grows. Near the extremal limit, where veff→0v_{\mathrm{eff}}\to0 for the relevant modes, the smearing band (2) collapses and simple time dilation is recovered, consistent with the near-horizon AdS2_2 geometry of the extremal black brane.

Physical applications.—Quark-gluon plasma. The hydrodynamization time of a QGP fluid cell satisfies τhydro∼1/Γgap\tau_{\mathrm{hydro}}\sim1/\Gamma_{\mathrm{gap}} in the local rest frame of the cell; the values τ∼0.5\tau\sim0.5–11 fm/c inferred from viscous hydrodynamic fits [18] are quoted in this frame (Bjorken proper time). Our bounds constrain instead how this relaxation appears to the collision center-of-mass observer for fluid cells at forward rapidity yy, which move with γ=cosh⁡y\gamma=\cosh y. Bound (3) gives an upper bound on the observed relaxation time,

τhydroCM≤γ(1+vvmax⁡)τhydrorest.\tau_{\mathrm{hydro}}^{\mathrm{CM}}\le\gamma(1+v v_{\max})\tau_{\mathrm{hydro}}^{\mathrm{rest}}.

which for vmax⁡=1v_{\max}=1 (as our holographic computation shows for the non-hydrodynamic sector) and y∼2y\sim2–4 constrains the frame dependence of hydrodynamization across the rapidity range covered by LHC detectors; the numerical result of Fig. 1 suggests the observed relaxation is in fact faster than in the local rest frame. The bound (5) on k~nc\tilde{k}_{\mathrm{nc}} constrains the maximal wavenumber at which the Navier–Stokes gradient expansion is applicable in the CM frame, providing a systematic criterion for when Israel-Stewart or BDNK corrections become necessary.

Neutron star mergers. In binary neutron star mergers, bulk viscosity relaxation in relative motion between remnant matter contributes to gravitational wave damping. Bound (3) gives

τ~Π−1≥τΠ−1γ(1+vcs),\tilde{\tau}_{\Pi}^{-1}\ge\frac{\tau_{\Pi}^{-1}}{\gamma(1+v c_s)},

where csc_s is the QCD sound speed at nuclear density. Combined with gravitational-wave measurements of csc_s [19], this constrains the bulk relaxation rate in the merger frame without requiring a specific equation-of-state model.

Quantum critical systems. Near a spin-density-wave quantum critical point with emergent Lorentz invariance (emergent speed vFv_F), the Planckian dissipation rate Γ∼kBT/ℏ\Gamma\sim k_{\mathrm{B}}T/\hbar plays the role of Γgap\Gamma_{\mathrm{gap}}. Bound (3) constrains how this rate transforms for transport measurements along different crystal directions in a moving frame — a prediction accessible via angle-resolved photoemission spectroscopy.

Discussion and conclusions.—We have established four Lorentz-covariant spectral bounds in thermal QFT, derived from causality, unitarity, and the KMS condition alone. Their key feature is that they are non-perturbative: they hold for any thermal QFT, whether weakly coupled, strongly coupled, or described holographically, without assuming a linearized constitutive relation or Onsager symmetry. The smearing of rest-frame poles into a boosted continuum — with width δΓ~=2Γgapvvmax⁡/[γ(1−vmax⁡2)]\delta\tilde{\Gamma}=2\Gamma_{\mathrm{gap}}vv_{\max}/[\gamma(1-v_{\max}^{2})] — is a universal consequence of relativity of simultaneity acting on a nontrivially dispersing excitation.

Several extensions are natural. Spontaneously broken symmetries (superfluid phases of QCD or nuclear matter) modify the KMS condition in the presence of a condensate; their Goldstone modes generate additional constraints on the low-∣k~∣|\tilde{k}| spectral bounds. Anomalous transport (chiral magnetic and vortical effects) introduces frame-dependent contributions not constrained by Onsager symmetry, potentially sharpening or relaxing the bounds in a chirally restored plasma. Far-from-equilibrium situations, where the KMS condition is absent, require replacing spectral positivity with more general constraints from the density matrix; the present results provide the equilibrium baseline for such a program. Finally, a quantitative lattice QCD study of the boosted spectral function would test the bounds across the deconfinement crossover.

In summary, our results provide a unified, model-independent framework for understanding how special relativity constrains the relaxation dynamics of quantum many-body systems across energy scales ranging from heavy-ion collisions to neutron star mergers.

We thank Bahodir Kayumov and Avas Khugaev for valuable discussions, and Giorgio Torrieri for illuminating correspondence on the covariance of Kubo formulae and the role of contact terms. This work was supported by New Uzbekistan University.

References

  1. [1]P. K. Kovtun and A. O. Starinets, Phys. Rev. D 72, 086009 (2005).
  2. [2]S. Grozdanov, P. K. Kovtun, A. O. Starinets, and P. Tadić, JHEP 11 (2019) 097.
  3. [3]P. Romatschke and U. Romatschke, Relativistic Fluid Dynamics In and Out of Equilibrium (Cambridge University Press, 2019).DOI
  4. [4]R. E. Hoult and P. Kovtun, JHEP 06 (2020) 067.
  5. [5]U. Heinz and R. Snellings, Ann. Rev. Nucl. Part. Sci. 63, 123 (2013).
  6. [6]C. Gale, S. Jeon, and B. Schenke, Int. J. Mod. Phys. A 28, 1340011 (2013).
  7. [7]G. Policastro, D. T. Son, and A. O. Starinets, Phys. Rev. Lett. 87, 081601 (2001).
  8. [8]P. K. Kovtun, D. T. Son, and A. O. Starinets, Phys. Rev. Lett. 94, 111601 (2005).
  9. [9]L. Gavassino, “How Lorentz boosts reshape relaxation spectra,” Phys. Rev. Lett. 137, 022301 (2026).DOI
  10. [10]L. Gavassino, arXiv:2607.1048 (2026).
  11. [11]A. Sanetullaeva, S. Bazarbaeva, M. Beymamatova, and S. Tursunov, arXiv:2607.13482 [nucl-th] (2026).
  12. [12]S. Grozdanov, P. K. Kovtun, A. O. Starinets, and P. Tadić, Phys. Rev. Lett. 122, 251601 (2019).
  13. [13]M. P. Heller, A. Serantes, M. Spaliński, V. Svensson, and B. Withers, Phys. Rev. D 107, 086018 (2023).
  14. [14]G. Torrieri, arXiv:2307.07021 [hep-th] (2023).arxiv.org/abs/2307.07021
  15. [15]G. M. Sampaio, G. Rabelo-Soares, and G. Torrieri, arXiv:2504.11752 [hep-th] (2025).
  16. [16]A. Núñez and A. O. Starinets, Phys. Rev. D 67, 124013 (2003).
  17. [17]M. Brigante, H. Liu, R. C. Myers, S. Shenker, and S. Yaida, Phys. Rev. Lett. 100, 191601 (2008).
  18. [18]B. Schenke, P. Tribedy, and R. Venugopalan, Phys. Rev. C 86, 034908 (2012).
  19. [19]A. Bauswein and H.-T. Janka, Phys. Rev. Lett. 108, 011101 (2012).

Paper details

vlnc:W_uuqp5rq2
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