arXiv:2609.37296
Parametric DRAG for leakage-suppressed exchange gates in superconducting qubits
Yiwen Li, Xinsheng Tan, Yang Yu
Abstract
Programmable exchange interactions support quantum gates and many-body simulation in superconducting circuits. Fast excitation transfer, however, opens two leakage pathways in weakly anharmonic transmons, degrading gates and driving simulated dynamics outside the encoded state space. We introduce parametric derivative removal by adiabatic gate (PDRAG) to suppress both pathways while preserving the target exchange. Their conjugate structure determines first- and second-derivative corrections implemented through one real frequency command. At $g/2π=100$ MHz, calibrated PDRAG achieves a geometric-mean leakage reduction of $5.27\times10^5$ relative to the base pulse across 12.5--35 ns in Duffing simulations. At 13 ns, endpoint leakage reaches $1.98\times10^{-7}$ for the calibrated full-exchange pulse. Transferring the derivative coefficients to a charge Hamiltonian gives a mean 893-fold reduction over 12.5--22 ns after amplitude and carrier recalibration. Floquet return interference and charge-induced gap shifts explain the calibration and transfer gains, connecting compact pulse design to leakage-suppressed programmable interactions.
AI Summary
Fast parametric exchange in transmons also couples $|11\rangle$ to $|20\rangle$ and $|02\rangle$. This paper develops parametric DRAG (PDRAG) to suppress both leakage channels while retaining full excitation exchange. Their conjugate coupling structure places first-order spectral constraints at $\alpha_1$ and $-\alpha_2$. A complex envelope combining first- and second-derivative corrections satisfies these constraints; Bessel inversion converts it into one real frequency-modulation command. Numerical calibration adjusts the derivative coefficients, amplitude, and carrier. For a fixed two-transmon device with $g/2\pi=100$ MHz, carrier-resolved Duffing simulations cover 46 durations from 12.5–35 ns. Calibrated PDRAG reduces endpoint leakage by a geometric-mean factor of $5.27\times10^5$ relative to the base pulse and $1.92\times10^4$ relative to the better independently optimized single-derivative control. Leakage is $1.98\times10^{-7}$ at 13 ns. Transferring derivative coefficients to a charge Hamiltonian, with amplitude/carrier recalibration, gives a geometric-mean improvement of 372 over the base pulse across 12.5–22 ns; leakage is $2.17\times10^{-7}$ at 20 ns. Floquet return expansions explain calibrated cancellation, while dressed-gap shifts explain transfer reversals. Independent propagation and basis-convergence checks support the numerical results. The results concern coherent endpoint leakage in simulations, not experimental fidelities. Transient leakage remains, and transferred pulses lose their advantage at some longer durations. The operations have substantial conditional phase: at 20 ns, transferred best-fit fSim infidelity is $5.44\times10^{-8}$, but strict-iSWAP infidelity is approximately 0.227 even after local $Z$ corrections.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.