Quantum control of the environment in open quantum systems enables rapid qubit reset

Abstract

Qubit reset is crucial in quantum technology and is typically achieved by coupling the qubit to a dissipative environment. However, the achievable speed and fidelity are limited by qubit-environment entanglement. We use exact tensor-network simulations and a time-dependent variational approach to investigate these effects for transmon qubits with a time-dependent system-environment coupling. We show that they are due to the formation of a polaron state and how this can be reversed using a time-dependent coupling. Coupling protocols are identified which achieve reset with an excited-state population of $10^{-6}$ in $10$ ns. A related paper [C. Ortega-Taberner, E. O'Neill and P. R. Eastham, arXiv:XXXX.XXXX] addresses the complementary case of control via a time-dependent Hamiltonian. Our work shows how the dynamics of the environment of an open quantum system can be controlled to design effective quantum processes in non-Markovian systems.

Figures (3)

• Figure 1: (a) Excited state population $P_+$ for reset in the spin-boson model (solid), showing exponential relaxation at early times and a non-zero residual population. This population corresponds to that predicted by the polaron grounnd state (dashed). (b) Spectral distribution of the bath displacements at $t=10\;\mathrm{ns}$ for the spin-boson model (solid) and corresponding displacements from the polaron ground state (circles).
• Figure 2: (a) Evolution of the residual population of the qubit during decoupling, obtained using TDVP for different values of the smoothness constant $\lambda$. For the linear case, $\lambda = 1$, the result of a TEMPO calculation is shown (dashed black). (b) Trajectory of the bath displacements during decoupling, for $\omega_k = \omega_C/2$ and different decoupling functions. (c) Displacements for $\lambda = 2$ and different bath frequencies.
• Figure 3: (a) Residual population of the qubit during decoupling, obtained for the optimal control given by the solution to the linear-quadratic regulator, given in (b). Computed for different values of the soft bound on the control $R$.