Floquet Thermal Transistors Learn to Control Their Own Noise

A transistor is useful because a small signal can steer a much larger flow. Electronics built the modern world on that idea. Quantum thermodynamics is now asking a parallel question: can a small thermal signal steer a much larger heat current inside a nanoscale quantum device? A new paper in Physica Scripta, published online on July 30, 2026, moves that question from “can it amplify?” toward the more engineering-focused question “can it amplify without becoming too noisy?”

The paper, by Samir Das, Shishira Mahunta, Nikhil Gupt, Victor Mukherjee and Arnab Ghosh, is titled “Fluctuations and optimal control in a Floquet quantum thermal transistor.” It studies a three-terminal quantum thermal transistor made from three coupled qubits, labelled emitter, collector and base, each connected to its own heat bath. The base qubit is periodically driven. That time-periodic control is the Floquet ingredient: rather than changing a bath temperature directly, the device uses a modulation waveform to reshape heat transport through drive-assisted transition channels.

The important shift is from headline amplification to usable amplification. A thermal transistor that multiplies heat-current response but also multiplies fluctuations is not yet a practical thermal control element.

The result matters for Floquet.ca’s quantum-energy theme because heat routing, noise, and control cost are where elegant quantum thermodynamic ideas begin to look like real devices. Future quantum processors, sensors and cryogenic circuits will not only need to move information; they will need to move heat deliberately. A Floquet thermal transistor is one candidate building block for that thermal logic layer.

From quantum heat engine to quantum heat circuit

Much of quantum thermodynamics is introduced through heat engines: a quantum working substance contacts hot and cold reservoirs and converts part of the energy flow into work. Thermal transistors are different. Their job is not primarily to produce work. Their job is to regulate heat current. In a common-base electronic transistor, a small base signal controls a larger emitter-collector current. In a quantum thermal transistor, a small base heat current controls larger heat currents between other terminals.

The 2022 precursor paper, “Floquet Quantum Thermal Transistor” in Physical Review E, showed why periodic modulation is attractive. Traditional quantum thermal transistors often rely on changing the base-bath temperature. That can be energy-intensive, especially if the base environment has a large heat capacity. Floquet control offers another route: modulate the base qubit’s frequency and use sidebands to redirect heat flow. The authors reported that periodic driving can preserve transistor operation even in a cutoff region where static thermal transistors fail as viable heat modulators.

That earlier work established the device concept. The new 2026 paper asks what happens when the heat currents are treated statistically. At small scales, a “current” is not a smooth river. It is a stream of stochastic energy-exchange events. Two devices can have the same average heat current but very different noise. For thermal logic, cooling hardware or heat-management elements near quantum chips, the noisy version is much less useful.

Why full counting statistics changes the question

To quantify fluctuations, the authors use full counting statistics. Instead of calculating only the average heat current flowing into each bath, full counting statistics tracks the probability distribution of exchanged energy. From that distribution researchers can compute the variance, the Fano factor and thermodynamic uncertainty-relation ratios.

The paper examines two simple modulation patterns first: sinusoidal driving and pi-flip modulation of the base qubit frequency. The authors find that the variance of the base current can be much smaller than the variances of the emitter and collector currents. However, the Fano-factor story reverses: the base current can have a comparatively larger Fano factor because its mean current is intentionally small. That is exactly the transistor tension. A good transistor wants the base to do little energetic work while strongly steering other currents. But a tiny base current can make the signal look noisy when measured relative to its mean.

This is where the research becomes more than a toy model. The analysis gives designers a way to separate several goals that are often conflated:

• Amplification: a small change in base current produces a large change in emitter or collector current.
• Precision: heat-current fluctuations remain low enough for the device to be predictable.
• Low base load: the control terminal does not itself become a large heat-current channel.
• Thermodynamic accounting: the periodic drive, baths and fluctuations are all included rather than hidden.

Those goals can point in different directions. A waveform that maximizes gain may not minimize noise. A waveform that suppresses the Fano factor may increase the base current enough to weaken the transistor analogy. The paper’s value is that it makes those trade-offs calculable.

Optimal control enters the heat transistor

After testing standard waveforms, Das and colleagues apply the Chopped Random Basis, or CRAB, optimal-control protocol. CRAB is a practical strategy for searching over control pulses without specifying an impossibly detailed waveform in advance. The modulation is expanded in a truncated set of randomized basis functions, and the coefficients are optimized for a target objective.

Here the target can be chosen in two ways. One optimization seeks stronger dynamical amplification. Another seeks a lower Fano factor. The amplification optimization performs well: compared with sinusoidal and pi-flip modulation, the optimized waveform produces substantially higher amplification over a wide range of base-bath temperatures. The authors identify a particularly favorable window around base temperatures in the range 0.1 to 0.12 in their normalized units, where amplification peaks while the base current remains small compared with the emitter and collector currents.

The Fano-factor optimization is more sobering. It can push the transistor toward lower fluctuations and even toward analytical lower-bound behavior. But the paper reports a price: reducing the Fano factor is associated with a larger base current. In other words, the control terminal becomes more energetically active. That may be acceptable for some heat-control tasks, but it is not the cleanest transistor operation if the goal is tiny-control-signal amplification.

In nanoscale thermodynamics, control is never free. The question is not whether a waveform improves one figure of merit, but which thermodynamic bill gets moved elsewhere.

Why this belongs in the “beyond-Carnot” conversation

A thermal transistor is not a Carnot engine, and this paper does not claim to beat Carnot efficiency. That is precisely why it is useful for a mature beyond-Carnot discussion. The frontier is not magic efficiency. It is the discovery of quantum resources and time-dependent controls that alter what can be switched, amplified, stabilized or measured under thermodynamic constraints.

Floquet driving contributes a powerful resource: it creates energy sidebands. A periodically driven qubit can exchange heat quanta with baths at frequencies shifted by integer multiples of the drive frequency. That lets a device route energy in ways not available to a static Hamiltonian. But the same drive also injects structure, cost and potential noise. The 2026 transistor paper treats that reality directly by combining Floquet master-equation thinking with full counting statistics and optimal control.

For smart non-specialists, the clean analogy is traffic control. A static thermal device is like a road network with fixed lanes. Floquet modulation is like programmable traffic lights and reversible lanes operating on a schedule. Optimal control adjusts the schedule. Full counting statistics checks whether traffic flow is smooth or stop-and-go. A headline average flow is not enough if the jams are large, unpredictable or shifted onto the control lane.

Near-term relevance for quantum technologies

The authors explicitly connect their work to heat-modulation devices in near-term quantum technologies. That phrase should be read carefully. This is a theoretical model, not a packaged cryogenic component. But its ingredients are familiar to real platforms: qubits, engineered reservoirs, periodic microwave or control-field modulation, and open-system dynamics. Quantum computing labs already spend enormous effort managing dissipation, measurement backaction and thermal leakage. As devices scale, passive cooling will not be the whole story.

Potential application directions include:

• thermal routers that steer heat away from sensitive qubits or sensors;
• noise-aware thermal switches for mesoscopic circuits;
• diagnostics that infer bath coupling from heat-current fluctuations;
• building blocks for thermal logic, where heat currents carry control information;
• testbeds for thermodynamic uncertainty relations in driven quantum systems.

The strongest lesson is methodological. Quantum-energy devices should be evaluated not only by average power, efficiency or gain, but also by fluctuation metrics. A high-gain heat transistor with poor noise characteristics could destabilize the very circuit it is meant to regulate. Conversely, a low-noise device that requires a large base current may be better understood as an active heat pump or regulator than as a transistor-like amplifier.

The bottom line

“Fluctuations and optimal control in a Floquet quantum thermal transistor” advances the field by adding precision to the thermal-transistor playbook. The 2022 Floquet transistor paper showed that periodic modulation can produce large heat-current amplification, including regimes where static devices struggle. The 2026 follow-up asks the next engineering question: how noisy is that amplified heat current, and can the waveform be optimized?

The answer is encouraging but nuanced. CRAB optimal control can substantially enhance amplification. It can also reduce fluctuation measures. But optimizing one target can degrade another, especially by increasing the base current that a transistor ideally keeps small. For Floquet energy research, that trade-off is not a disappointment. It is a sign that the field is moving from proof-of-principle phenomena toward honest device design.