---
title: Thermodynamic Transport Bound (TB)
url: https://www.emergentmind.com/topics/thermodynamic-transport-bound-tb
type: topic
---

# Thermodynamic Transport Bound (TB)

Searching arXiv for the cited papers to ground the article in current literature.
{"query":"id:2507.01214 OR id:2502.07917 OR thermodynamic transport bound coherent transport", "max_results": 10}
I found recent arXiv records directly relevant to the topic, including "Thermodynamic bound on current fluctuations in coherent conductors" [2507.01214] and "Thermodynamic Uncertainty Relations for Coherent Transport" [2502.07917], along with related work on periodically driven coherent conductors [1906.04297], ballistic multi-terminal transport [1710.04928], generalized transport [2403.00048], and Markov jump processes [2508.00940].
The thermodynamic transport bound (TB) denotes a class of nonequilibrium constraints that relate transport observables to dissipation, entropy production, or closely allied thermodynamic quantities. In the most explicit recent usage for coherent quantum conductors, the TB is a universal bound on the full large-deviation function of a particle current, depending only on the mean current and the total steady-state entropy production rate; in related literatures, the same expression is used for bounds on current noise, diffusion coefficients, generalized observable transport, or thermoelectric performance [2507.01214][2502.07917].

## 1. Concept and scope

In current usage, a thermodynamic transport bound is a statement that transport cannot be made arbitrarily fast, precise, or improbable at fixed thermodynamic cost. The relevant “cost” is typically the steady-state entropy production rate \(\sigma\), and the relevant transport quantity may be a particle current, heat current, diffusion coefficient, or the finite-time change of an observable. This suggests that TB is best understood as a family of universal inequalities rather than a single formula.

A representative feature of this family is that the bound is expressed in terms of experimentally or operationally accessible transport data. In coherent fermionic transport, the bound is written in terms of the mean current \(J_\alpha\), the current noise \(S_\alpha\), or the full large-deviation function \(I_\alpha(j)\) [2507.01214][2502.07917]. In Markov jump processes, the same idea appears as
\[
\sigma \ge \frac{\bar J^2}{2D_J},
\]
with \(\bar J\) the long-time average current and \(D_J\) its diffusion coefficient [2508.00940]. In generalized Langevin transport, the bound constrains the finite-time change of any differentiable scalar observable \(z\) by the total entropy production \(\Delta S_{\rm tot}\) and a kinematic stretching factor \(\mathcal D^z(t)\) [2403.00048].

These bounds are not identical in derivation or regime of validity. Some are large-deviation statements for coherent conductors [2507.01214], some are thermodynamic uncertainty relations for current precision [2502.07917], some are finite-time speed limits [2403.00048], and some are performance bounds for thermoelectric devices or heat currents [1211.4737][2209.04666]. Their common content is that transport statistics are not independent of dissipation.

## 2. Full large-deviation bound for coherent conductors

The most developed recent formulation of the TB is the one for coherent, non-interacting fermionic conductors in the Landauer–Büttiker scattering framework [2507.01214]. For a particle current \(j\) associated with reservoir \(\alpha\), the long-time distribution obeys
\[
p_t(j)\sim e^{-I_\alpha(j)t},
\]
where \(I_\alpha(j)\) is the large-deviation function. The bound states that
\[
I_\alpha(j)\le \hat I_{\alpha,\mathrm{qu}}(j),
\]
with
\[
\hat I_{\alpha,\mathrm{qu}}(j)
=
(J_\alpha R_\alpha + j)\,\ln\frac{J_\alpha R_\alpha + j}{J_\alpha R_\alpha + J_\alpha}
+
(J_\alpha R_\alpha - j)\,\ln\frac{J_\alpha R_\alpha - j}{J_\alpha R_\alpha - J_\alpha},
\]
valid for
\[
-J_\alpha R_\alpha< j < J_\alpha R_\alpha.
\]
The dissipation factor is
\[
R_\alpha=\coth\!\left(\frac{\sigma}{4k_{\rm B}J_\alpha}\right),
\]
so the bound depends only on the mean current \(J_\alpha\) and the total steady-state entropy production rate \(\sigma\) [2507.01214].

The same result can be expressed at the level of the scaled cumulant generating function,
\[
\chi_\alpha(s)\ge \hat\chi_\alpha(s)
=
2J_\alpha R_\alpha
\ln\frac{R_\alpha\cosh(s/2)+\sinh(s/2)}{R_\alpha},
\]
from which the large-deviation bound follows by Legendre–Fenchel transform. This is the sense in which the TB constrains not only typical fluctuations but also rare current fluctuations [2507.01214].

Within the scattering formulation, the mean particle current is
\[
J_\alpha=\frac{1}{h}\sum_{\beta}\int dE\,T_{\alpha\beta}(E)\,[f_\alpha(E)-f_\beta(E)],
\]
and the total entropy production rate is
\[
\sigma
=
-\sum_\alpha \frac{J_\alpha^Q}{T_\alpha}
=
\frac{k_{\rm B}}{h}\sum_{\alpha,\beta}\int dE\,T_{\alpha\beta}(E)\,[f_\alpha-f_\beta]\,
\ln\frac{f_\alpha}{1-f_\alpha}\ge 0.
\]
The bound applies to arbitrary chemical-potential and temperature gradients in multi-terminal coherent conductors, provided the transmission coefficients satisfy
\[
T_{\alpha\beta}(E)=T_{\beta\alpha}(E).
\]
This condition is automatic in any two-terminal conductor and more generally follows from time-reversal symmetric dynamics inside the conductor [2507.01214].

## 3. Relation to thermodynamic uncertainty relations

The coherent-conductor TB contains the thermodynamic uncertainty relation as its small-fluctuation limit. Near the mean current,
\[
I_\alpha(j)\approx \frac{(j-J_\alpha)^2}{2S_\alpha},
\]
where
\[
S_\alpha=\lim_{t\to\infty}\frac{\langle \hat N_\alpha^2(t)\rangle-\langle \hat N_\alpha(t)\rangle^2}{t}
\]
is the zero-frequency current noise. Comparing the curvature of \(I_\alpha(j)\) with that of \(\hat I_{\alpha,\mathrm{qu}}(j)\) yields the coherent-transport thermodynamic uncertainty relation
\[
\mathcal Q_{\rm qu}
\equiv
\frac{S_\alpha}{J_\alpha}\,
\sinh\!\left(\frac{\sigma}{2k_{\rm B}J_\alpha}\right)\ge 1
\]
[2507.01214][2502.07917].

This quantum relation replaces the classical Markovian form
\[
\mathcal Q_{\rm cl}=\frac{S_\alpha \sigma}{2k_{\rm B}J_\alpha^2}\ge 1,
\]
and the corresponding classical large-deviation bound
\[
I_\alpha(j)\le \hat I_{\alpha,\mathrm{cl}}(j)
=
\frac{(j-J_\alpha)^2}{4k_{\rm B}J_\alpha^2}\,\sigma.
\]
For coherent fermionic transport, the quantum constraint is weaker in the precise sense that
\[
\hat I_{\alpha,\mathrm{cl}}(j)\le \hat I_{\alpha,\mathrm{qu}}(j),
\qquad
\mathcal Q_{\rm cl}\le \mathcal Q_{\rm qu},
\]
so Pauli blocking and energy filtering permit stronger suppression of fluctuations at fixed \(\sigma\) than classical Markovian dynamics allow [2507.01214].

The 2025 coherent-transport TUR was derived independently in a form that already displayed the characteristic hyperbolic-sine dependence,
\[
\frac{S_\alpha}{J_\alpha}\sinh\!\left[\frac{\sigma}{2B J_\alpha}\right]\ge 1,
\]
for non-interacting fermions in the Landauer–Büttiker framework, with arbitrary chemical and thermal biases and arbitrary multi-terminal geometry under time-reversal symmetry [2502.07917]. In that formulation, a modified bound with a numerical factor \(\psi_0\simeq 0.85246\) extends the TUR to broken time-reversal symmetry, but an analogous full large-deviation bound was not established there [2502.07917].

## 4. Saturation and optimal scatterers

The coherent TB is not merely formal. A two-terminal chain of quantum dots provides an explicit model approaching saturation [2507.01214][2502.07917]. The effective non-Hermitian Hamiltonian is
\[
H_{\rm eff}
=
\sum_{\ell=1}^{n-1} t_\ell (|\ell\rangle\langle \ell+1|+|\ell+1\rangle\langle \ell|)
-i\Gamma|1\rangle\langle 1|
-i\Gamma|n\rangle\langle n|
+E_0,
\]
and the transmission function is
\[
T(E)=4\Gamma\,\big|\langle n |(E-H_{\rm eff})^{-1}|1\rangle\big|^2.
\]
With a suitable choice of tunnel couplings, one obtains
\[
T(E)=\frac{1}{1+[2(E-E_0)/w]^{2n}},
\]
which converges to a boxcar transmission as \(n\to\infty\) [2507.01214].

For a narrow boxcar window, the current and entropy production are approximately
\[
J \approx \frac{w}{h}\tanh(\mathcal F/4),
\qquad
\sigma \approx \frac{k_{\rm B}w}{h}\,\mathcal F\tanh(\mathcal F/4),
\]
so that
\[
R=\coth\!\left(\frac{\sigma}{4k_{\rm B}J}\right)=\coth(\mathcal F/4).
\]
In this regime, the scaled cumulant-generating-function bound is saturated up to corrections of order \(w^2\), and the large-deviation bound and quantum TUR are saturated simultaneously [2507.01214].

Numerically, the single-dot case lies strictly below the classical bound and above the quantum TB, while increasing the chain length makes the exact large-deviation function approach the quantum bound from below; by \(n\sim 20\), the bound is nearly saturated [2507.01214]. This establishes the boxcar transmission as an optimal scatterer for the coherent TB, paralleling its role in coherent thermoelectric optimization [2502.07917].

## 5. Extensions across transport theories

The same thermodynamic logic appears in several neighboring frameworks.

| Framework | Representative bound | Transport object |
|---|---|---|
| Coherent conductors | \(I_\alpha(j)\le \hat I_{\alpha,\mathrm{qu}}(j)\) | Full current LDF |
| Markov jump processes | \(\sigma \ge \bar J^2/(2D_J)\) | Mean current and diffusion |
| Generalized transport | \(\Delta S_{\rm tot}\ge \frac{\langle \Delta z\rangle^2}{t\,\mathcal D^z(t)}\) | Observable transport |
| Ballistic multi-terminal transport | \(\sigma \varepsilon_\alpha \ge 2k_{\rm B}\) or \(\psi^\ast k_{\rm B}\) | Precision of extracted current |
| Periodically driven coherent conductors | \(\sigma \ge \frac{N}{4K^{xx}(N-1)}[J_\alpha^x]^2\) | Local matter or heat current |

For stationary Markov jump processes, the TB takes the form
\[
\sigma \ge \frac{\bar J^2}{2D_J},
\]
or in multidimensional form
\[
\sigma \ge \frac{1}{2}\,\mathbf{\bar A}^{\mathsf T}\mathbf D^{-1}\mathbf{\bar A},
\]
with \(\mathbf D\) the long-time covariance matrix of integrated currents. In that setting, the TB is the long-time limit of the thermodynamic uncertainty relation, with \(D_J=\lim_{T\to\infty}\operatorname{Var}(J_T)/(2T)\) identified as the diffusion coefficient of the current [2508.00940].

For generalized transport of any differentiable scalar observable \(z\) in underdamped, overdamped, or deterministic dynamics, the finite-time bound reads
\[
\Delta S_{\rm tot}
\ge
\frac{\Big\langle z_t-z_0-\int_0^t d\tau\,\partial_\tau z_\tau\Big\rangle^2}
{t\,\mathcal D^z(t)},
\]
with
\[
\mathcal D^z(t)=\frac{1}{t}\int_0^t d\tau\,
\left\langle
[\nabla_x z_\tau]^{\mathrm T}\gamma^{-1}(\tau)\nabla_x z_\tau
\right\rangle.
\]
This is explicitly described as a time-integrated generalized speed limit and extends thermodynamic transport bounds from single-molecule to bulk observables such as structure factors or radii of gyration [2403.00048].

For classical ballistic multi-terminal transport, a universal precision–dissipation trade-off was derived:
\[
\sigma \varepsilon_\alpha \ge 2k_{\rm B}
\]
under time-reversal symmetry, and
\[
\sigma \varepsilon_\alpha \ge \psi^\ast k_{\rm B},
\qquad
\psi^\ast \simeq 0.89612,
\]
when a magnetic field breaks time-reversal symmetry. An explicit chiral transport model saturates the weaker bound [1710.04928].

For periodically driven coherent conductors, the linear-response theory yields a family of bounds proving that any local matter or heat current imposes a non-trivial lower bound on the overall dissipation rate. A representative form is
\[
\sigma \ge \frac{N}{4K^{xx}(N-1)}[J_\alpha^x]^2,
\]
while far from equilibrium the theory produces quantum TUR-like inequalities and an operationally accessible lower bound \(\sigma\ge \sigma_\alpha^\ast\) in terms of the mean current, its zero-frequency noise, and the reservoir temperature [1906.04297].

Thermoelectric transport supplies a complementary usage. In nonlinear scattering theory, one finds the first-law constraint \(\sum_i(J_i+P_i)=0\), the second-law inequality \(\sum_i(-J_i/T_i)\ge 0\), and the quantum heat-current bound
\[
J_i \le \frac{\pi^2}{6h}N_i(k_{\rm B}T_i)^2,
\]
which limits refrigeration and power generation independently of Carnot efficiency [1211.4737]. In linear-response Boltzmann transport with a bounded transport distribution \(\Sigma(E)\le \Sigma_{\max}\), the optimal bounded transport distribution is a boxcar for \(ZT\) and a Heaviside function for the power factor, yielding practical upper limits that scale with \(\Sigma_{\max}T/\kappa_l\) [2209.04666].

## 6. Limitations, misconceptions, and distinct usages

The coherent-conductor TB is not a statement about arbitrary quantum transport. Its derivation assumes coherent, non-interacting fermionic transport, steady-state operation, and symmetric transmission coefficients \(T_{\alpha\beta}(E)=T_{\beta\alpha}(E)\). It is formulated for particle currents; heat or energy currents require different dimensional combinations and are not covered by the same bound [2507.01214][2502.07917].

A frequent misconception is that the TB simply restates the thermodynamic uncertainty relation. In the coherent-conductor setting, the TUR is only the quadratic, near-mean limit of a stronger statement about the entire large-deviation function [2507.01214]. Conversely, in jump-process theory the TB is explicitly the long-time transport form of the TUR rather than a separate finite-time principle [2508.00940]. Which interpretation is appropriate depends on the framework.

Another misconception is that the acronym TB has a uniform meaning across fields. In plasma physics, “TB” often means “transport barrier,” not “thermodynamic transport bound.” In the thermodynamic model of magnetically confined plasma boundary layers, the relevant bounds are threshold conditions for the existence of a high-gradient state,
\[
T_0>T_c,\qquad F>F_c(T_0),
\]
with \(F_c(T_0)\) non-monotonic and minimized at \(T_{\mathrm{opt}}=4T_c\) [2603.26919]. This is a thermodynamic threshold theory of transport barriers rather than a fluctuation or precision bound.

Across the literature, the strongest common conclusion is narrower and more robust: transport observables are thermodynamically constrained. In coherent conductors, dissipation bounds the full distribution of current fluctuations [2507.01214]. In stochastic transport, it limits the ratio of mean current to diffusion or precision [2508.00940][1710.04928]. In generalized nonequilibrium dynamics, it bounds the finite-time displacement of observables [2403.00048]. In thermoelectrics, it restricts the attainable combination of heat flow, power, and transport distribution [1211.4737][2209.04666]. This suggests that the thermodynamic transport bound is best regarded as a unifying principle: transport cannot be specified independently of entropy production.

Source: https://www.emergentmind.com/topics/thermodynamic-transport-bound-tb