arXiv:2610.38527
Harnessing Non-Markovian Memory for Quantum-Battery Capacity
Jun-Li Jiang, Xin-Zhu Liu, Xiu-Yong Ding, Xue Yang, Ming-Xing Luo
Abstract
Non-Markovianity is commonly characterized through information-theoretic signatures of memory. Here we take a thermodynamic perspective and ask whether such memory can be operationally harnessed to enhance quantum-battery capacity. Within the process-tensor framework, we compare memoryless and memory-enhanced strategies and define their performance gap as the memory-assisted capacity advantage of a multitime process. We prove that this advantage vanishes for Markovian process tensors and is bounded by a trace-distance measure of process non-Markovianity. In a complementary continuous-time setting, we show using random telegraph dephasing that memory accumulated during the evolution can enhance capacity retention relative to a matched Markovian reference under the same control class. These results identify non-Markovian memory as an operational resource for quantum-battery capacity.
AI Summary
The paper asks whether environmental memory can improve quantum-battery capacity, defined as the energy range accessible within a state's unitary orbit, after accounting for capacity supplied by control. It examines multitime process tensors and continuous-time random-telegraph-noise (RTN) dephasing. For a fixed process tensor, it compares a greedy, locally optimized sequential benchmark with coordinated input preparation, collective output readout, and restricted causal feed-forward. Controller costs are measured by each operation's maximum capacity increase; collective readout is adjusted to exclude activation available from uncorrelated outputs. The resulting advantage $\mathcal A_{\mathcal C}$ vanishes for Markovian processes and satisfies $0\leq\mathcal A_{\mathcal C}\leq2n(n+3)\Lambda N(\mathcal P)$, where $\Lambda$ is the local Hamiltonian's spectral width and $N$ is operational trace distance to the Markovian-process set. The proof combines capacity-balance inequalities with trace-norm continuity. An explicit three-step qubit construction gives gains of $0$ for greedy control, $(\sqrt3-1)E/2$ for joint input optimization, and $E/2$ for adjusted collective readout and restricted feed-forward. For RTN dephasing, a single rotation transfers transverse coherence into populations protected against subsequent dephasing. Exact augmented propagation retains system–noise correlations during pulse optimization. A numerical example achieves greater retained capacity than an independently optimized, fitted CP-divisible RTN reference under the same pulse family. The results concern specified control classes: persistent controller ancillas are excluded, and retained registers must be uncorrelated with subsequent inputs. The RTN comparison is a proof of principle using a fitted reference, not a general optimization over Markovian dynamics. Extensions to many-body batteries and realistic architectures remain open.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- Submitted to Physical Review A