Entropy Transport in Programmable Quantum Junctions

Technical Summary: Entropy Transport in Programmable Quantum Junctions

Problem Statement
Traditional literature on open quantum systems has predominantly focused on the energy balance of central quantum systems. This work adopts a transport perspective, treating the quantum system as a junction mediating the flow of physical quantities between reservoirs. While charge, spin, and energy currents are well-understood, entropy transport presents a unique challenge. Unlike other currents, entropy flow is intrinsically nonlinear in the reduced density matrix ($\rho$) and cannot generally be expressed as the expectation value of a conventional quantum-mechanical operator. This nonlinearity places entropy transport outside standard linear transport theory, making its evaluation in macroscopic reservoirs a nontrivial problem. Furthermore, while recent frameworks have extended nonequilibrium quantum thermodynamics to nonlinear information measures, the specific behavior of entropy transport in programmable multi-qubit architectures operating far from equilibrium remains largely unexplored.

Methodology
The authors develop a general theoretical framework for an $N$-qubit quantum junction coupled to $M$ thermal reservoirs and a weakly coupled probe reservoir.

• Model: The system is described by a total Hamiltonian $H_{tot} = H_S + H_B + H_{SB}$, where $H_S$ includes qubit energies, pairwise interactions, and external periodic drives. The environment consists of harmonic oscillator baths.
• Dynamics: Under the Born-Markov and secular approximations, the system dynamics are governed by a Lindblad master equation. The probe reservoir is treated as weakly coupled ($\Gamma_p \ll \Gamma_A$) to ensure it does not perturb the junction's intrinsic dynamics.
• Entropy Definition: The authors utilize the information-theoretic definition of entropy flow. The rank-$\alpha$ Rényi entropy flow is defined as $I_\alpha \equiv -\frac{d}{dt} \ln \text{Tr}[\rho^\alpha]$. The physically relevant von Neumann entropy flow ($I$) is obtained via analytic continuation ($\lim_{\alpha \to 1} dI_\alpha/d\alpha$).
• Formalism: Using the Keldysh formalism, the entropy conductance is derived as the sum of an incoherent contribution ($I_{inc}$), arising from stochastic transitions (single-world processes), and a coherent contribution ($I_{coh}$), arising from cross-world correlations between replicated Keldysh contours.
• Comparative Analysis: The theory is applied to two specific architectures analogous to quantum computing gates:
  1. A driven single-qubit junction (X-gate analogue).
  2. A driven two-qubit junction where the control qubit is driven at the dressed frequency of the target qubit (Cross-Resonance/Cross-Resonance gate analogue).

Key Contributions and Results

1. Programmability of Entropy Transport: The study demonstrates that entropy conductance is not solely determined by reservoir parameters (temperature, coupling strength) but is a programmable quantity governed by the junction's microscopic Hamiltonian, driving fields, and quantum state. By tuning these parameters, one can actively engineer entropy transfer.

2. Thermodynamic Advantage of Multi-Qubit Architectures:

   • The authors introduce an enhancement factor $R = I_{2q}/I_{1q}$ comparing the stationary entropy conductance of the two-qubit junction to the single-qubit junction.
   • Result: In the weak-driving regime, the two-qubit junction achieves significantly higher entropy conductance than the single-qubit junction (enhancement > 40% at low temperatures) while requiring substantially lower driving power.
   • Mechanism: This advantage arises from the interplay between drive-induced excitation, inter-qubit hybridization, and reservoir-induced relaxation. The two-qubit system opens a conditional transport channel where the state of the driven qubit controls the excitation dynamics of the undriven qubit. However, this advantage is bounded; at strong driving amplitudes, the single-qubit junction eventually becomes more efficient due to direct energy injection.

3. Coherent Contribution to Entropy Current:

   • The analysis reveals a sizable coherent contribution ($I_{coh}$) to the entropy current that emerges only under resonant driving.
   • This term is proportional to the off-diagonal elements of the density matrix (quantum coherence). In the two-qubit case, this involves a collective superposition of multiple transition pathways, making entropy transport sensitive to the multi-qubit structure of the quantum state.
   • The coherent contribution is found to be negative, partially opposing the incoherent entropy current, and can reach magnitudes up to half of the incoherent contribution near its maximum.

4. Negative Differential Entropy Conductance (NDEC):

   • The paper identifies a regime of negative differential entropy conductance, where increasing the thermal bias (temperature of the main reservoir) suppresses the entropy flow into the probe reservoir ($\partial I / \partial T_1 < 0$).
   • This effect is analogous to negative differential thermal conductance but is driven by the control of the external drive amplitude and the internal spectral structure of the junction rather than reversing temperature gradients.
   • In the two-qubit junction, NDEC is controlled by the ratio of inter-qubit coupling to detuning ($J/\Delta$). It is robust in the detuning-dominated regime ($J/\Delta < 1$) but suppressed when hybridization is too strong ($J/\Delta > 1$).

5. Resonance Phenomena:

   • While the single-qubit junction exhibits a single entropy flow maximum at resonance, the two-qubit junction displays three distinct maxima corresponding to different anticrossings in the energy spectrum.
   • The largest entropy current occurs when driving the higher-frequency qubit near the transition frequency of the lower-frequency qubit (the operating condition for the Cross-Resonance gate).

Significance and Claims
The paper claims to establish entropy conductance as a controllable design parameter for quantum processors, directly linking quantum-device engineering to thermodynamic performance.

• Fundamental Insight: It reveals that quantum logic architectures (specifically multi-qubit gates) can function as programmable devices for entropy transport, capable of selectively accelerating, suppressing, or redirecting entropy flow.
• Practical Implications: The results suggest routes toward quantum feedback control, reservoir protection, and refrigeration in driven quantum circuits. Specifically, the ability to suppress entropy injection into a probe reservoir using lower driving power in multi-qubit systems offers a thermodynamic advantage for minimizing dissipation in quantum thermal machines.
• Experimental Relevance: The authors note that the theoretical predictions are attainable on superconducting-qubit platforms, where the relevant gate architectures (X-gate and CR-gate) and parameters are well-characterized.

The work concludes that entropy transport is governed by the spectral and coherent properties of the junction, not merely by thermal gradients, offering a new mechanism for engineering non-equilibrium entropy flow in quantum devices.