---
title: Common Thermodynamic Vector Field
url: https://www.emergentmind.com/topics/common-thermodynamic-vector-field
type: topic
---

# Common Thermodynamic Vector Field

“Common thermodynamic vector field” is a non-uniform term used in several technical literatures for a vector-valued object that organizes thermodynamic structure. In relativistic equilibrium it denotes the four-temperature field $\beta^\mu=u^\mu/T$, whose Killing property characterizes global equilibrium [1606.06605, 1712.08031]. In mean-field QCD with vector interactions it denotes the flavor-independent isoscalar shift $-G_V^S\rho_q$ in the effective chemical potentials of light quarks [1003.4428]. In irreversible thermodynamics it can denote the generalized force vectors driving transport or the mesoscopic force field $F=D^{-1}b$ that generates cycle affinity and entropy production [1303.6809, 2004.08677]. Recent black-hole thermodynamic topology uses an auxiliary two-component vector field whose zeros locate Davies and Hawking–Page transitions [2404.02526, 2606.10680]. The expression therefore names a family of constructions rather than a single canonical object.

## 1. Principal meanings of the term

The term is used in distinct mathematical settings, but in each case a single vectorial structure is taken to control several thermodynamic observables, response functions, or phase-transition markers. The resulting usages are not interchangeable.

| Setting | Vector field or covector | Role |
|---|---|---|
| Relativistic equilibrium | $\beta^\mu=u^\mu/T$ | Encodes flow, temperature, and equilibrium kinematics |
| Mean-field QCD | Common isoscalar shift $-G_V^S\rho_q$ | Universally shifts $u$ and $d$ effective chemical potentials |
| Irreversible transport | $X_q=-\nabla T$, $X_A=F_A$ | Drives entropy-producing transport and relaxation |
| Mesoscopic NET | $F(x)=D^{-1}(x)b(x)$ | Generates cycle affinity and steady-state entropy production |
| Geometric thermodynamics | $\mathcal{E}_H=\sharp_\Lambda(dH)$ | Implements thermodynamic time evolution |
| Black-hole topology | $\boldsymbol{\varphi}=(\frac{1}{S}\partial_S F^2,-\cot\theta\,\csc\theta)$ | Unifies Davies and Hawking–Page critical points |

A recurring structural feature is universality within a chosen formalism. The object is “common” because it is shared by all flavors, by all observables transported along a flow, by all transport channels in an entropy production formula, or by multiple phase-transition loci. A persistent misconception is that the phrase denotes a unique spacetime vector; the literature instead uses it for spacetime vectors, auxiliary parameter-space vectors, generalized force fields, and even covector gradients, depending on context.

## 2. Relativistic thermodynamics: the four-temperature field

The most standardized usage is the relativistic four-temperature field
$$
\beta^\mu \equiv \frac{u^\mu}{T},
\qquad
u^\mu=\frac{\beta^\mu}{\sqrt{\beta^2}},
\qquad
T=\frac{1}{\sqrt{\beta^2}},
$$
with $u^\mu u_\mu=1$ and timelike, future-directed $\beta^\mu$ [1606.06605, 1712.08031]. In this formulation $\beta^\mu$ is primary: flow velocity and comoving temperature are derived from it, rather than specified independently. Global thermodynamic equilibrium requires $\beta^\mu$ to be a Killing vector,
$$
\nabla_\mu \beta_\nu+\nabla_\nu \beta_\mu=0,
$$
and in Minkowski space the general solution is affine,
$$
\beta_\mu(x)=b_\mu+\varpi_{\mu\nu}x^\nu,
$$
with constant antisymmetric thermal vorticity
$$
\varpi_{\mu\nu}=-\frac{1}{2}\big(\partial_\mu\beta_\nu-\partial_\nu\beta_\mu\big).
$$

This Killing property is not merely kinematic. It guarantees hypersurface independence of the covariant equilibrium density operator and implies that physical observables are stationary along the $\beta$-flow. A central result is that all Lie derivatives of all physical observables along $\beta$ vanish at equilibrium,
$$
\mathcal{L}_\beta O=0,
$$
including $\mathcal{L}_\beta T^{\mu\nu}=0$ and $\mathcal{L}_\beta j^\mu=0$ [1606.06605]. In this sense the $\beta$-frame is a preferred hydrodynamic frame: it simultaneously encodes thermal structure and the rigid transport of equilibrium observables.

The same field organizes accelerated equilibrium. In the pure-acceleration case one may choose
$$
\beta^\mu=\frac{1}{T_0}(1+az,0,0,at),
$$
or, after shifting the origin, $\beta^\mu=\frac{a}{T_0}(z',0,0,t)$, with flow lines given by Rindler hyperbolae [1712.08031]. Along these trajectories the proper temperature obeys Tolman’s law, and the acceleration satisfies
$$
a^\mu=\frac{1}{\beta^2}\,\varpi^\mu{}_\nu\,\beta^\nu.
$$
For a free scalar field, after Minkowski-vacuum subtraction, all local quadratic observables vanish when the global parameter equals the Unruh temperature $T_U=a/(2\pi)$, implying the local bound
$$
T\ge \frac{|A|}{2\pi}
$$
for accelerated equilibrium in the right Rindler wedge [1712.08031].

At local equilibrium the same four-temperature field governs spin polarization. For the free Proca field, the local-equilibrium density operator couples $\beta_\mu(x)$ to the canonical stress-energy tensor and an independent spin potential $\Omega_{\mu\nu}(x)$ to the canonical spin tensor,
$$
\widehat{\rho}_{\rm LE}
=
\frac{1}{Z_{\rm LE}}
\exp\!\left\{
-\int_\Sigma d\Sigma_\mu
\left[
\widehat{T}^{\mu\nu}\beta_\nu
-\frac{1}{2}\widehat{S}^{\mu,\lambda\nu}\Omega_{\lambda\nu}
\right]
\right\},
$$
with thermal vorticity and thermal shear derived from $\beta_\mu$ [2412.19416]. In this canonical pseudo-gauge, $\Omega_{\mu\nu}$ need not equal $\varpi_{\mu\nu}$ at local equilibrium. For spin-1 bosons, time-reversal symmetry forces the leading spin alignment to arise only at second order in gradients of the thermodynamic fields, so the tensor polarization is controlled by second-order structures built from $\beta_\mu$, $\Omega_{\mu\nu}$, and their derivatives [2412.19416]. This clarifies that even within the relativistic usage the common vector field is not the whole thermodynamic data set; it is the organizing field to which additional tensors may be coupled.

## 3. Mean-field QCD: the common isoscalar vector field

In the mean-field treatment of light-flavor susceptibilities, the common thermodynamic vector field is the vector–isoscalar mean field generated by net quark density [1003.4428]. The quark sector is coupled to an isoscalar vector field $V_\mu$ and an isovector vector field $B_\mu$, with medium-dependent couplings $G_V^S(T)$ and $G_V^V(T)$. Their mean-field effect is to shift the effective chemical potentials,
$$
\mu_u^*=\mu_u-G_V^S\rho_q-G_V^V\rho_I,
\qquad
\mu_d^*=\mu_d-G_V^S\rho_q+G_V^V\rho_I,
$$
where $\rho_q=n_u+n_d$ and $\rho_I=n_u-n_d$.

The isoscalar contribution $-G_V^S\rho_q$ is common to both flavors and depends only on total quark number. This is the paper’s “common thermodynamic vector field”: a universal repulsive mean field acting equally on $u$ and $d$ quarks [1003.4428]. By contrast, the isovector term changes sign between flavors and differentiates them through isospin density. At the level of the grand potential, vector interactions enter through shifted chemical potentials and quadratic density terms,
$$
\Omega(T,\mu_u,\mu_d;G_V^S,G_V^V)
=
\Omega_q(T,\mu_u^*,\mu_d^*)
-\frac{1}{2}G_V^S\rho_q^2
-\frac{1}{2}G_V^V\rho_I^2
-TR.
$$

The phenomenological importance of the common field is most transparent in susceptibilities at $\mu=0$. With $s(T)=G_V^S(T)/T$, $v(T)=G_V^V(T)/T$, and quark-sector potential $\Phi$, one obtains
$$
\chi_q=\frac{2\Phi}{T(1+2s\Phi)},
\qquad
\chi_I=\frac{2\Phi}{T(1+2v\Phi)},
$$
together with
$$
\chi_{uu}=\chi_{dd}=\frac{\chi_q+\chi_I}{4},
\qquad
\chi_{ud}=\chi_{du}=\frac{\chi_q-\chi_I}{4},
$$
or directly
$$
\chi_{ud}(T)=
\frac{\Phi^2(v-s)}{T(1+2s\Phi)(1+2v\Phi)}.
$$
The universal isoscalar repulsion suppresses quark-number fluctuations through the factor $(1+2s\Phi)^{-1}$, while the isovector interaction suppresses isospin fluctuations through $(1+2v\Phi)^{-1}$ [1003.4428]. The off-diagonal susceptibility vanishes if $s=v$ or if both couplings vanish.

This structure permits inversion from lattice data:
$$
s(T)=\frac{1}{T\chi_q}-\frac{1}{2\Phi},
\qquad
v(T)=\frac{1}{T\chi_I}-\frac{1}{2\Phi},
\qquad
s(T)-v(T)=\frac{1}{T}\left[\frac{1}{\chi_q}-\frac{1}{\chi_I}\right].
$$
Using two-flavor lattice data at $\mu=0$, the paper reports that $(s-v)T^3$ is large below $T_c$ and rapidly approaches zero above $T_c$, implying that the isoscalar coupling exceeds the isovector coupling in the hadronic phase, while $s\simeq v$ for $T\gtrsim T_c$ [1003.4428]. This is the precise sense in which a common thermodynamic vector field emerges in the deconfined phase: the net vector interaction felt by $u$ and $d$ becomes nearly identical. A rough comparison to typical NJL-model scalar couplings yields broad bounds of order $4\,{\rm GeV}^{-2}\lesssim G_V^S\lesssim 19\,{\rm GeV}^{-2}$, although the paper emphasizes that firmer conclusions require a more complete three-flavor analysis [1003.4428].

## 4. Irreversible transport and nonequilibrium cycle fields

In macroscopic irreversible thermodynamics for a charged multicomponent fluid with electric dipoles and magnetic moments, the “common thermodynamic vector fields” are the generalized force vectors appearing in entropy production [1303.6809]. With generalized chemical potentials $\bar{\mu}_A=\mu_A+q_AV-p_A\!\cdot\!E-m_A\!\cdot\!B$, the entropy production density is written as
$$
\sigma
=
\frac{1}{T}
\left\{
\sum_a \omega_a A_a
+j_s\!\cdot\!(-\nabla T)
+\sum_A j_A\!\cdot\!F_A
+\sum_A \Omega_A^m\!\cdot\!(p_A\times E)
+\sum_A \Omega_A^M\!\cdot\!(m_A\times B)
\right\},
$$
where the transport-driving vector fields are
$$
F_A
=
-\nabla\mu_A
-q_A\nabla V
-m_Av_A\nabla v
-I_A\omega_A\nabla\omega
+p_A\nabla E
+m_A\nabla B,
$$
along with the heat force $X_q=-\nabla T$ [1303.6809]. The same framework yields Onsager–Casimir transport laws, Debye relaxation, Landau–Lifshitz relaxation, the Lehmann and electric Lehmann effects, magnetization accumulation, and magnetization waves. Temperature gradients also induce effective fields,
$$
B_{\rm ind}=\epsilon_M\times\nabla T,
\qquad
E_{\rm ind}=\epsilon_P\nabla T,
$$
which feed back into the torque-like relaxation channels [1303.6809].

At mesoscopic scale, nonequilibrium thermodynamics of continuous diffusions identifies a different common field,
$$
F(x)=D^{-1}(x)b(x),
$$
constructed from the drift $b(x)$ and the diffusion metric $D^{-1}(x)$ [2004.08677]. Its wedge-curl,
$$
\nabla\wedge F(x),
$$
is the cycle-affinity bivector, while the divergence-free stationary flux admits a bivector potential $A(x)$ through
$$
J^*(x)=\nabla\times A(x).
$$
The local steady-state entropy production density is then
$$
\sigma(x)=A(x)\cdot[\nabla\wedge F(x)],
$$
and the Onsager operator
$$
\mathcal{O}=\nabla\wedge\big[e^\Phi D^{-1}\nabla\times\big]
$$
maps the vorticity potential to cycle affinity [2004.08677]. In this language the common field is independent of any particular macroscopic transport channel; all physical fluxes are encoded through a single probability flux in phase space.

A closely related decomposition writes the drift itself as
$$
b
=
-D\nabla\Phi
-Q\nabla\Phi
+\nabla\times Q,
$$
with $\Phi=-\ln\pi$ the mesoscopic potential and $Q=e^\Phi A$ the cycle velocity bivector [2007.08755]. The three terms are a generalized gradient field, a component perpendicular to $\nabla\Phi$, and a divergence-free field. The associated probability flux is
$$
J
=
-pD\nabla F
-pQ\nabla F
+\nabla\times(pQ),
$$
and the free-energy dissipation rate is purely gradient-driven,
$$
\dot{F}
=
\mathbb{E}\big[-(\nabla F)\cdot D\nabla F\big]
\le 0
$$
[2007.08755]. This suggests a systematic distinction between dissipation controlled by scalar potentials and housekeeping dissipation controlled by cyclic vector or bivector structures.

## 5. Geometric, manifold, and field-theoretic formulations

A separate group of works makes the thermodynamic vector field the generator of dynamics itself. On contact state spaces for simple thermo-mechanical systems, the common choice of evolution is
$$
\mathcal{E}_H=\sharp_\Lambda(dH),
$$
where $\Lambda$ is a skew-symmetric bivector and $H$ is the total energy [2012.07404]. This evolution vector field satisfies
$$
\mathcal{E}_H(H)=0,
\qquad
i_{\mathcal{E}_H}\eta=0,
$$
so energy is exactly preserved and the first law is enforced as horizontality with respect to the contact form. For Hamiltonians with positive semidefinite kinetic part one has
$$
\dot{S}=\Delta_Q(H)\ge 0,
$$
which realizes the second law through skew structure rather than through an explicit symmetric dissipation operator [2012.07404]. For two subsystems exchanging heat, the almost-Poisson tensor
$$
\Lambda_K
=
\Lambda_{Q_1}
+
\Lambda_{Q_2}
+
K\,\frac{\partial}{\partial S_1}\wedge\frac{\partial}{\partial S_2},
\qquad
K=k\left(\frac{1}{T_1}-\frac{1}{T_2}\right),
$$
yields
$$
\mathcal{E}_{H,K}(H)=0,
\qquad
\mathcal{E}_{H,K}(S_1+S_2)
=
k\,\frac{(T_2-T_1)^2}{T_1T_2}\ge 0
$$
[2012.07404].

In thermo-hydrodynamics as a field theory, the common thermodynamic object is the gradient of a material-time field $z^0(x)$ [1112.5842]. The material-time covector $\theta_\mu=\partial_\mu z^0$ acts as a temperature potential through
$$
T=\beta\,u^\mu\partial_\mu z^0,
$$
and the momentum conjugate to $z^0$ is
$$
\Theta^\mu{}_0=\beta \rho s u^\mu,
$$
whose conservation is precisely entropy conservation [1112.5842]. The construction ties temperature, entropy current, and canonical structure to one covector field. In equilibrium-oriented relativistic language this covector is closely related to the inverse-temperature four-vector, but the formalism is built from the material-time gradient rather than from a Killing field.

On equilibrium thermodynamic manifolds, exact differentials produce conservative thermodynamic vector fields. For simple compressible systems,
$$
dU=T\,dS-p\,dV,\qquad
dH=T\,dS+V\,dp,\qquad
dA=-S\,dT-p\,dV,\qquad
dG=-S\,dT+V\,dp,
$$
which induce gradient fields such as $v_U=(T,-p)$ in $(S,V)$ and $v_G=(-S,V)$ in $(T,p)$ [1808.03863]. The same perspective underlies algorithmic recovery of exact 1-forms from data. In the $(p,V)$ chart for one mole of monatomic ideal gas, the recovered entropy differential is
$$
dS=\frac{c_v}{p}\,dp+\frac{c_v+R}{V}\,dV,
$$
so the associated conservative vector field is $(c_v/p,(c_v+R)/V)$ [1808.03863].

A neighboring geometric formulation replaces the vector field by a canonical two-form on the equilibrium manifold,
$$
\Omega
=
-dT\wedge dS+dP\wedge dV
=
-U_{SV}\,dS\wedge dV,
$$
whose flux gives both mechanical work and reversible heat [2603.22559]. The Hodge dual of $\Omega$ yields a pseudovector only after choosing a metric and orientation, so this work is better interpreted as vector-adjacent rather than strictly vectorial. The local scalar field
$$
U_{SV}
=
\left(\frac{\partial T}{\partial V}\right)_S
=
-\left(\frac{\partial P}{\partial S}\right)_V
=
-\frac{T\alpha}{\kappa_T C_V}
$$
sets the signed work of infinitesimal cycles [2603.22559].

For externally driven systems, the Rayleigh–Bénard problem motivates a still different meaning. The Boussinesq dynamical vector field $F[X]$ is treated as the fixed point of a gradient flow in the space of mesoscopic vector fields, driven by a rate-entropy potential
$$
\Phi_{\rm SMT}(U,\nabla T)
=
-
S_{\rm SMT}(U)
+
\langle \nabla T,\dot{E}_{\rm SMT}(U)\rangle
$$
[2502.13984]. Near onset, the resulting rate-thermodynamic Landau theory uses the polynomial
$$
R(v;b)=a(b)v^2+w_3v^4,
$$
with the Boussinesq generator functioning as the common thermodynamic vector field for the driven system [2502.13984]. Here the field is not an observable state variable but the autonomous mesoscopic generator toward which more microscopic vector fields relax.

## 6. Black-hole thermodynamic topology and thermodynamic dipoles

In AdS black-hole thermodynamics the common thermodynamic vector field is an auxiliary two-component field on a parameter manifold rather than a spacetime field [2404.02526]. With one thermodynamic coordinate $\xi$—typically entropy $S$ or horizon radius $r_h$—and an auxiliary angle $\theta$, the unified field is defined by
$$
\phi^\xi
=
\frac{1}{\xi}\frac{\partial(F^2)}{\partial \xi},
\qquad
\phi^\theta
=
-\cot\theta\,\csc\theta,
$$
where $F$ is the ensemble-appropriate free energy. In the entropy representation,
$$
\phi^S
=
\frac{2F}{S}\frac{\partial F}{\partial S}
=
-2F\frac{\partial T}{\partial S}
=
-2F\frac{T}{C}.
$$
Hence $\phi^S=0$ if and only if either $F=0$ or $1/C=0$, so the zeros of the same vector field locate Hawking–Page and Davies-type transitions [2404.02526].

Duan’s $\phi$-mapping topological current theory assigns topological charge to these zeros. For simple isolated zeros, the sign is determined by the Jacobian, and the critical point corresponding to the Davies-type phase transition has charge $-1$ while the Hawking–Page point has charge $+1$ [2404.02526]. The construction was exhibited for the Schwarzschild AdS black hole, the Reissner–Nordström AdS black hole in the grand canonical ensemble, and Kerr AdS black holes in the grand canonical ensemble. In this setting the “common” aspect is literal: one vector field replaces the separate fields previously used for the two kinds of transition.

The 2026 extension recasts the same construction as a thermodynamic dipole [2606.10680]. The common thermodynamic vector field is
$$
\boldsymbol{\varphi}(S,\theta)
=
\left(
\frac{1}{S}\frac{\partial F(S)^2}{\partial S},
-\cot\theta\,\csc\theta
\right),
$$
and its zeros in the elementary AdS branch carry winding numbers
$$
w_{\rm D}=-1,
\qquad
w_{\rm HP}=+1.
$$
Because the total topological charge vanishes, the relevant invariant is the signed first moment. Defining
$$
P_S=S_{\rm HP}-S_{\rm D},
\qquad
P_T=T_{\rm HP}-T_{\rm D},
$$
and normalizing by Davies scales gives
$$
C_S=\frac{S_{\rm HP}-S_{\rm D}}{S_{\rm D}},
\qquad
C_T=\frac{T_{\rm HP}-T_{\rm D}}{T_{\rm D}}.
$$
For four-dimensional Schwarzschild–AdS one finds
$$
C_S=2,
\qquad
C_T=\frac{2}{\sqrt{3}}-1,
$$
while the normalized free-energy barrier is
$$
B=\frac{F(S_{\rm D})}{T_{\rm D}S_{\rm D}}=\frac{1}{3}
$$
[2606.10680]. In higher-dimensional charged non-rotating grand-canonical families the barrier becomes
$$
B(d)=\frac{1}{(d-1)(d-3)}.
$$

The same work emphasizes that universality depends on a common reduced geometry of the temperature profile, not merely on the existence of a vector field. Rotation in Kerr–AdS preserves the charges $w_{\rm D}=-1$ and $w_{\rm HP}=+1$ for $|\Omega|<1$, but deforms the normalized dipole lengths and introduces higher-order corrections to $C_S$ and $C_T$ [2606.10680]. This makes the black-hole usage conceptually distinct from the relativistic or QCD usages: the field is a local-topological diagnostic on an auxiliary thermodynamic manifold, not a transport or equilibrium flow field in spacetime.

Across these literatures, “common thermodynamic vector field” therefore designates a shared organizing structure rather than a universally fixed mathematical entity. It may be a Killing four-temperature, a flavor-independent mean field, a generalized transport force, a mesoscopic drift-derived force, an evolution generator on a contact manifold, or an auxiliary topological map. What is common is the function performed: one vectorial object encodes multiple thermodynamic constraints, response properties, or critical loci at once.

Source: https://www.emergentmind.com/topics/common-thermodynamic-vector-field