Concept · physics
Stochastic thermodynamics
3 studies1 discoveryEvidence last moved Sep 27, 2026
Stochastic thermodynamics applies work, heat and free energy to single small systems whose behaviour is dominated by thermal fluctuations. The evidence here comes from a colloidal experiment on first-passage times, a simulation of feedback-based work extraction, and molecular dynamics using the Jarzynski equality to compute nucleation free energies.
At the nanoscale, individual runs fluctuate wildly, yet averages over them obey exact relations that give equilibrium quantities from non-equilibrium data. This clears up the confusion that noisy, fast or imprecise processes can say nothing about free energies or efficiency.
Studies
3
Findings
4
5 supporting · 0 challenging · 3 qualifying citations
Open tensions
1
Latest change
Concept page published
Stochastic thermodynamics
Currently
What we know
- The timing of first arrivals can count hidden intermediate states.
- Imprecise measurements can still be used with full efficiency if the feedback is reversible.
- Exponential averaging of non-equilibrium work recovers an equilibrium free-energy barrier.
- Exact averages coexist with large run-to-run scatter, so sampling matters.
Largest unresolved question
The first-passage study is experimental and validated on real molecules, while the work-extraction and cavity results are simulations of idealised models; the latter show what the relations allow, not what is achieved in a lab.
Common misconceptions
A noisy measurement always wastes some of the work it could let you extract.
In the simulated protocol, reversible feedback extracted exactly kT times the mutual information, an efficiency of 1; the limit is the information gained, not the precision itself.
Free energies can only be computed from slow, equilibrium (quasistatic) processes.
The Jarzynski equality turns repeated fast, irreversible work measurements into a free-energy profile, as in the cavity-nucleation simulations, though it needs enough rare low-work runs.
First-passage times only tell you an average rate.
The shape of the short-time distribution revealed how many intermediate minima lay on the path, and its duration encoded their depth.
Related
Claim ledger
What the evidence shows
Drawn from 3 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
The timing of first arrivals can count hidden intermediate states.
For colloids crossing optical-trap landscapes, the short-time first-passage time distribution followed a power law whose integer slope equalled the number of intermediate minima; applied to a nanopore hopper and DNA hairpins, slopes of about 1 and 2 matched their known intermediate states.
- Can escape times reveal hidden steps in an energy landscape?— Recovers only the minimum number of states along the shortest path; needs minima clearly deeper than kT.
Imprecise measurements can still be used with full efficiency if the feedback is reversible.
In Langevin simulations of a trapped Brownian particle, reversible feedback after a noisy measurement extracted work equal to kT times the mutual information gained, the maximum allowed by the generalised second law; two measurements with double the error variance gave the same work as one precise measurement.
- Can sloppy measurements still power a perfect information engine?— Simulation only; requires idealised instantaneous control of the trap.
Exponential averaging of non-equilibrium work recovers an equilibrium free-energy barrier.
Averaging the work from 100 forced cavity-expansion runs with the Jarzynski equality gave free-energy curves with a nucleation barrier in a stretched Lennard-Jones liquid; barrier height and critical size both shrank with more negative pressure.
- How much energy does it take to open a bubble in stretched liquid?— Authors warn 100 samples may be too few; the cavity is field-forced, not a spontaneous bubble.
Exact averages coexist with large run-to-run scatter, so sampling matters.
In both the feedback simulation and the cavity simulations, individual runs varied widely even when averages matched theory: single feedback cycles fluctuated mostly in the confinement stage, and few cavity runs fell below the estimated free energy.
- Can sloppy measurements still power a perfect information engine?
- How much energy does it take to open a bubble in stretched liquid?
Study Role Design N Population Outcome Can sloppy measurements still power a perfect information engine? Supports Computational / modellingLangevin simulations of an overdamped Brownian particle in a harmonic trap with Gaussian-error measurements and reversible feedback, followed by quasistatic expansion. Simulated trajectories; the average cumulative work was computed over 200 cycles, each with 10 measurement steps. Model overdamped Brownian particle in a harmonic (optical-tweezer-like) potential Average extracted work per cycle versus information gained; work with differing measurement errors How much energy does it take to open a bubble in stretched liquid? Supports Computational / modellingMolecular dynamics of a Lennard-Jones liquid in which an expanding repulsive field carves out a cavity; the work done in 100 repeated runs is averaged with the Jarzynski equality to obtain free energy versus cavity radius at four negative pressures. No participants; each pressure condition used 100 independent simulations from different equilibrium starting configurations. Simulated Lennard-Jones (argon-like) liquid under negative pressure at reduced temperature 0.8 Free-energy change versus cavity radius; critical cavity size and activation energy; fitted surface tension and bulk free-energy difference
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
The first-passage study is experimental and validated on real molecules, while the work-extraction and cavity results are simulations of idealised models; the latter show what the relations allow, not what is achieved in a lab.
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
- A feedback engine turned every bit of noisy measurement into work, in simulation
Concept page published
Stochastic thermodynamics
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Rewordings are not listed.
- Concept page published
- A feedback engine turned every bit of noisy measurement into work, in simulationEvidence: Preliminary
Papers
3 studies in this library bear on Stochastic thermodynamics, ordered by citations.
- Can escape times reveal hidden steps in an energy landscape?
How quickly the fastest particles or molecules finish a journey reveals how many intermediate resting states they must pass through, and roughly how deep those states are.
- Can sloppy measurements still power a perfect information engine?
A simulated particle engine turns every bit of information from its measurements into work, and noisier measurements can match precise ones simply by measuring more often.
- How much energy does it take to open a bubble in stretched liquid?
Averaging the work from many fast, non-equilibrium simulations recovers the full energy barrier for opening a nanoscale cavity in a stretched liquid, even under mild conditions that ordinary simulations cannot reach.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
The first-passage study is experimental and validated on real molecules, while the work-extraction and cavity results are simulations of idealised models; the latter show what the relations allow, not what is achieved in a lab.
Ask PaperFren about Stochastic thermodynamics
Study this conceptflashcards and short-answer questions
State the Jarzynski equality in words and describe one practical difficulty in using it, with reference to evidence.
The Jarzynski equality says that the exponential average of the work done in many repeated non-equilibrium processes equals the exponential of minus the free-energy difference divided by kT. A molecular dynamics study used it to compute cavity nucleation barriers from 100 forced expansions of a stretched liquid. The difficulty is that the average is dominated by rare runs with unusually low work; the authors noted very few such runs, so more sampling might lower the barrier.
How can information be converted into work, and what sets the maximum?
Measuring a particle's position lets a controller adjust the trap to exploit fluctuations. The generalised second law caps the average extracted work at kT times the mutual information between measurement and state. Simulations of a trapped Brownian particle showed a reversible feedback protocol reaching this maximum even with Gaussian measurement error. This was a simulation with idealised control, so real devices would lose some efficiency.