Crosstalk Suppression in Individually Addressed Two-Qubit Gates in a Trapped-Ion Quantum Computer

Crosstalk between target and neighboring spectator qubits due to spillover of control signals represents a major error source limiting the fidelity of two-qubit entangling gates in quantum computers. We show that in our laser-driven trapped-ion system coherent crosstalk error can be modeled as residual $X{\widehat{\sigma }}_{\varphi }$ interaction and can be actively canceled by single-qubit echoing pulses. We propose and demonstrate a crosstalk suppression scheme that eliminates all first-order crosstalk utilizing only local control of target qubits, as opposed to an existing scheme which requires control over all neighboring qubits. We report a two-qubit Bell state fidelity of 99.52(6)% with the echoing pulses applied after collective gates and 99.37(5)% with the echoing pulses applied to each gate in a five-ion chain. This scheme is widely applicable to other platforms with analogous interaction Hamiltonians.

Figure 1

(a) Schematic of a five-ion chain with two tightly focused beams addressing ion 1 and ion 2 and the rest of the ions considered to be the spectators impacted by intensity crosstalk. (b) Circuit model of the effect of crosstalk with one spectator qubit. The residual entanglement between each target qubit and the spectator qubit is expressed as the MS interaction $X{\widehat{\sigma }}_{\varphi }$. (c) Populations of ion 0 and ion 3 (red and blue, respectively) after applying 21 consecutive $XX(\pi /4)$ to ion 1 and ion 2. The population of ion 3 varies as the effective crosstalk depends on ${\varphi }_{\mathrm{beam}}$. The population of ion 0 sees a much smaller excitation, since ion 0’s coupling to the motional modes mostly involved in the MS gate is weaker. The population of ion 3 after crosstalk suppression using the echoing technique is also shown (green). The average population drops from 0.25 to 0.03.

Figure 2

Circuit diagrams for crosstalk suppression in the MS gate. (a) Neighbor suppression performed in the collective-gate echoing experiments where $Z(\pi )$ rotations are driven on the spectator ion. A single $XX(\pi /4)$ is added after the second $Z(\pi )$ pulse to generate the Bell state for fidelity measurement. (b) Local suppression performed in the collective-gate echoing experiments where $Y(\pi )$ pulses are applied to the target ions. (c) Local suppression performed in the individual-gate echoing experiments where the $XX(\pi /4)$ is split into two half MS evolutions and the echoing $Y(\pi )$ pulses are applied after each $XX(\pi /8)$.

Figure 3

Results for the two-qubit Bell state infidelity and spectator ion population (for the native gates without crosstalk suppression) plotted against the number of consecutive gates for (a) the collective-gate echo experiments comparing the native gates to the gates with neighbor and local suppression and (b) the individual-gate echo experiments comparing the native gates to the gates with local suppression. The simulated gate error range is shown by the shaded area. The infidelity contribution of a single MS gate with crosstalk suppression is postprocessed from the linear fit.