Strong Local Passivity in Unconventional Scenarios: A New Protocol for Amplified Quantum Energy Teleportation

TL;DR

The paper addresses strong local passivity (SLP) and the restrictive constraints traditionally placed on QET by showing SLP can persist beyond ground-state initialization, commutativity with the interaction, and entanglement. It introduces a local effective Hamiltonian and a flip-flop two-qubit protocol that yields energy extraction without requiring entanglement, achieving an energy extraction $E_{\\mathrm{extract}}^{\\text{new}} = 2\\sin^2(\\theta) \\sqrt{h^2+\\kappa^2}$ and a ratio of $7.2$ over the original protocol for selected parameters. Analytically, the protocol demonstrates that Bob’s conditional unitary, guided by the effective Hamiltonian, can maximize energy extraction, while remaining compatible with SLP post-measurement. The authors validate the approach experimentally on three IBM backends and an AerSimulator, confirming positive energy extraction and the feasibility of the workflow even with dynamic-circuit constraints. These results broaden QET's applicability to broader quantum technologies and motivate extensions to many-body systems and practical energy-access applications in quantum chemistry and beyond.

Abstract

Quantum energy teleportation (QET) has been proposed to overcome the restrictions of strong local passivity (SLP) and to facilitate energy transfer in quantum systems. Traditionally, QET has only been considered under strict constraints, including the requirements that the initial state be the ground state of an interacting Hamiltonian, that Alice's measurement commute with the interaction terms, and that entanglement be present. These constraints have significantly limited the broader applicability of QET protocols. In this work, we demonstrate that SLP can arise beyond these conventional constraints, establishing the necessity of QET in a wider range of scenarios for local energy extraction. This leads to a more flexible and generalized framework for QET. Furthermore, we introduce the concept of a ``local effective Hamiltonian,'' which eliminates the need for optimization techniques in determining Bob's optimal energy extraction in QET protocols. As an additional advantage, the amount of energy that can be extracted using our new protocol is amplified to be 7.2 times higher than that of the original protocol. These advancements enhance our understanding of QET and extend its broader applications to quantum technologies. To support our findings, we implement the protocol on quantum hardware, confirming its theoretical validity and experimental feasibility.

Figures (5)

• Figure 1: The ratio of Bob's extracted energy for the new protocol, $E_{\mathrm{extract}}^{\mathrm{new}}$, and Bob's extracted energy for the original protocol, $E_{\mathrm{extract}}^{\mathrm{original}}$, as a function of the coupling strength $\kappa$ in units of $h$.
• Figure 2: The quantum circuit implemented on IBM quantum hardware to verify the feasibility of our proposed QET protocol. Since the system begins in the first excited state of the Hamiltonian in Equation \ref{['flipflop']}, $|0\rangle \otimes |0\rangle$, no additional state preparation is required. Alice initiates the protocol by performing a measurement, which injects energy into the system. After receiving Alice’s measurement outcome, Bob applies a conditional operation to extract energy from the system. A final measurement is then performed to verify the energy distribution within the system.
• Figure 3: To address the current lack of support for dynamic circuits by IBM quantum hardware, we use an alternative but equivalent circuit. In this approach, Alice’s measurement is postponed until after Bob’s conditional operations, and Bob’s conditional operations are replaced with two-qubit controlled operations. It can be shown that the two circuits are equivalent, as they yield identical resulting density matrices.
• Figure 4: The qubit distribution map for the backends of ibm_brussels and ibm_kyiv is demonstrated. Qubit 0 is used for Alice's qubit, and qubit 1 is used for Bob's qubit.
• Figure 5: The qubit distribution map for the backend of ibm_torino is demonstrated. Qubit 0 is used for Alice's qubit, and qubit 1 is used for Bob's qubit.