Thermodynamic uncertainty relation for feedback cooling

Kousuke Kumasaki, Kaito Tojo, Takahiro Sagawa, Ken Funo · 2025

Read Thermodynamic uncertainty relation for feedback cooling by Kousuke Kumasaki, Kaito Tojo, Takahiro Sagawa, Ken Funo (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-06174.txt.

Introduction

Kousuke Kumasaki and co-authors published a research paper in 2025 on the thermodynamic price of cooling a system by measuring and steering it, known as feedback cooling. Such control can chill systems toward their limits, but the cost of doing so in finite time was unclear. The team set up an uncertainty relation for the process.

Their relation trades cooling efficiency against the rate at which disorder falls. Perfect efficiency at a finite disorder-reduction rate, they show, is reachable only in a limit where one fluctuation blows up. A Kalman-filter control scheme proves that limit can be approached in a real setup. The result pins down what feedback cooling must cost.

Catalog record

CREATORKousuke Kumasaki, Kaito Tojo, Takahiro Sagawa, Ken Funo
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE45 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-06174

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Thermodynamic uncertainty relation for feedback cooling Kousuke Kumasaki, Kaito Tojo, Takahiro Sagawa, Ken Funo Abstract Feedback cooling enables a system to achieve low temperatures through measurement-based control. Determining the thermodynamic cost required to achieve the ideal cooling efficiency within a finite time remains an important problem. In this work, we establish a thermodynamic uncertainty relation (TUR) for feedback cooling in classical underdamped Langevin systems, thereby deriving a trade-off between the cooling efficiency and the entropy reduction rate. The obtained TUR implies that simultaneous achievement of the ideal cooling efficiency and finite entropy reduction rate is asymptotically possible by letting the fluctuation of the reversible local mean velocity diverge. This is shown to be feasible by using a feedback control based on the Kalman filter. Our results clarify the thermodynamic costs of achieving the fundamental cooling limit of feedback control from the perspective of the TUR. Introduction.— Recent experimental advances have enabled a precise investigation of thermodynamics in microscopic systems, where thermal fluctuations play a crucial role. These systems require a theoretical framework beyond conventional thermodynamics, giving rise to stochastic thermodynamics Sekimoto 2010 ; Esposito et al. 2009 ; Campisi et al. 2011 ; Seifert 2012 ; Ciliberto 2017 ; Funo et al. 2018 ; Landi and Paternostro 2021 , which explicitly incorporates thermal fluctuations. In this framework, the second law constrains the entropy production to be nonnegative. This framework has been further extended to information processing systems, known as information thermodynamics Parrondo et al. 2015 ; Sagawa and Ueda 2008 ; Horowitz and Esposito 2014 ; Yada et al. 2022 ; Ptaszyński and Esposito 2019 ; Yada et al. 2024 ; Prech et al. 2025 . Although the second law represents a fundamental thermodynamic bound on physical processes, it does not give a tight bound on the entropy production in general finite-time processes. To this end, thermodynamic uncertainty relations (TURs) Barato and Seifert 2015 ; Gingrich et al. 2016 ; Horowitz and Gingrich 2017 ; Dechant and Sasa 2018 ; Dechant 2018 ; Van Vu and Hasegawa 2019 ; Lee et al. 2019 have been established as generalizations of the second law, providing bounds between entropy production and generalized currents, and revealing trade-offs between efficiency and power in thermal engines Shiraishi et al. 2016 ; Pietzonka and Seifert 2018 ; Tajima and Funo 2021 . TURs have also been extended to the setup of Maxwell’s demon Potts and Samuelsson 2019 ; Van Vu and Hasegawa 2020 ; Liu et al. 2020 ; Otsubo et al. 2020 ; Tanogami et al. 2023 , leading to the derivation of TURs involving partial entropy production Otsubo et al. 2020 ; Tanogami et al. 2023 , and the establishment of trade-off relations in information heat engines Tanogami et al. 2023 . Feedback cooling is a paradigmatic example of an information heat engine, where work is extracted by converting information into energy. Applying feedback based on measurement results enables the suppression of particle fluctuations. This technique has been realized in various experimental systems, including nanomechanical resonators Rossi et al. 2018 and optically trapped nanoparticles Gieseler et al. 2012 ; Magrini et al. 2021 ; Gonzalez-Ballestero et al. 2021 ; Kamba et al. 2023 ; Kamba et al. 2025 . While the standard formulation of the second law relates entropy production to the temperature of the heat bath, the primary objective in feedback cooling is to reduce the temperature of the system below the bath temperature. In the study of classical systems, second-law-like inequalities have been established for Langevin dynamics Horowitz and Sandberg 2014 ; Sandberg et al. 2014 , relating the kinetic temperature of the system to the rate of information acquisition via measurement, quantified by the transfer entropy Schreiber 2000 ; Horowitz and Sandberg 2014 ; Hartich et al. 2014 . In feedback cooling, the effective use of information is crucial for achieving both efficient cooling and high cooling rates. Although examples achieving the fundamental information thermodynamic bound of feedback cooling have been reported Horowitz and Sandberg 2014 ; Kumasaki et al. 2025 , there exists a trade-off between power and efficiency, in general. Therefore, it is also expected that to achieve the ideal efficiency, the cooling power should vanish. However, without a TUR specific to feedback cooling, it has remained unclear whether maximum efficiency can coexist with a finite cooling rate, and, if so, what additional cost would be required. In this Letter, we derive TUR for feedback cooling that describes a fundamental trade-off relation between the cooling efficiency and entropy reduction rate in underdamped bipartite Langevin systems. The key technique to derive the main result is to introduce a novel orthogonal decomposition of the local mean velocity and the partial entropy production, in analogy with the decomposition proposed in Dechant et al. 2022a ; Dechant et al. 2022b ; Kamijima et al. 2023 , while our decomposition is unique to bipartite systems. This decomposition allows us to derive the TUR for feedback cooling by involving the kinetic temperature of the system. As a byproduct, we show that one of the decomposed partial entropy production recovers the previously known second-law-like inequality Horowitz and Sandberg 2014 in a special case. From the obtained TUR, we identify that divergence of the fluctuation of the reversible local mean velocity is required to simultaneously achieve the maximum cooling efficiency and finite entropy reduction rate. We introduce a concrete example based on the Kalman filter and show that increasing the feedback gain to infinity allows for satisfying the above condition (see also Fig. 1 ). The obtained results provide design principles for achieving optimal feedback cooling, which is particularly relevant to nanoparticle experiments Gieseler et al. 2012 ; Magrini et al. 2021 ; Gonzalez-Ballestero et al. 2021 ; Kamba et al. 2023 ; Kamba et al. 2025 . Figure 1: Schematic of the TUR for feedback cooling, where \epsilon_{\rm I} is the cooling efficiency and \dot{W}_{\mathrm{ext}}/T_{\mathrm{kin}} is the entropy reduction rate. The region above the parabola is prohibited by the obtained TUR ( 15 ). In the limit of large feedback gain G\rightarrow\infty , the entropy reduction rate maintains a finite value while approaching the maximum cooling efficiency \epsilon_{\rm I}\rightarrow 1 (see also Fig. 3 ). Setup.— We analyze the setup of feedback cooling by considering a bipartite system that consists of a system described by a Brownian particle and a memory that registers measurement outcomes. The system is coupled to a thermal reservoir at temperature T and is subject to a damping force -\gamma v , where v is the velocity of the system. A feedback force -ay is applied to reduce the kinetic energy of the particle, where y denotes the outcome of continous measurement of the velocity. The time evolution of the joint probability distribution p(v,y) is governed by the following Fokker-Planck equation Munakata and Rosinberg 2013 ; Horowitz and Sandberg 2014 : \displaystyle\partial_{t}p(v,y)=-\partial_{v}J_{v}(v,y)-\partial_{y}J_{y}(v,y). (1) The probability currents in the v and y directions are given by \displaystyle J_{v}(v,y) \displaystyle=\underbrace{-\frac{ay}{m}p(v,y)}_{J_{v}^{\mathrm{rev}}(v,y)}\ \underbrace{-\frac{\gamma}{m}\Bigl(v+\frac{T}{m}\partial_{v}\Bigr)p(v,y)}_{J_{v}^{\mathrm{irr}}(v,y)}, (2) \displaystyle J_{y}(v,y) \displaystyle=-\frac{1}{\tau}(y-v)p(v,y)-\frac{\sigma^{2}}{2\tau^{2}}\partial_{y}p(v,y), (3) where we set the Boltzmann constant k_{\mathrm{B}} to unity. In Eq. ( 3 ), the parameter \tau defines the timescale of the low-pass filter applied to the measurement and \sigma^{2} characterizes the uncertainty in the measurement signal. We assume

Read the complete text