Enhancing the Coherence of a Spin Qubit by Operating it as a Feedback Loop That Controls its Nuclear Spin Bath

Abstract

In many realizations of electron spin qubits the dominant source of decoherence is the fluctuating nuclear spin bath of the host material. The slowness of this bath lends itself to a promising mitigation strategy where the nuclear spin bath is prepared in a narrowed state with suppressed fluctuations. Here, this approach is realized for a two-electron spin qubit in a GaAs double quantum dot and a nearly ten-fold increase in the inhomogeneous dephasing time T * 2 is demonstrated. Between subsequent measurements, the bath is prepared by using the qubit as a feedback loop that first measures its nuclear environment by coherent precession, and then polarizes it depending on the final state. This procedure results in a stable fixed point at a nonzero polarization gradient between the two dots, which enables fast universal qubit control.PACS numbers: 73.21. La, 03.67.Lx, 76.70.Fz Spins in semiconductors are attractive qubits because of their long coherence times [1][2][3], their electrical control and readout [4], and their potential for scalability [5]. Few-electron quantum dot devices have been used successfully in recent years to demonstrate universal control of electron spin qubits as well as single shot readout [6][7][8][9]. However, interaction of the qubit spin(s) with nearby nuclear spins is a significant source of decoherence in several systems [2-4, 10, 11]. It is therefore very attractive to prepare the spin environment of the electron in a way that mitigates this decoherence. One approach would be to polarize it [12], but the nearly complete polarization required for improved coherence[13] is difficult to achieve.Here, we present a method to narrow the distribution of the fluctuating nuclear hyperfine field while maintaining a weak polarization. In addition to the reduced decoherence, such a narrowed state is of interest for studying the long-time quantum dynamics arising from the spinbath interaction [14]. Our method relies on first letting the qubit evolve under the influence of the bath. The resulting final state of the qubit controls the effectiveness of a subsequent dynamic nuclear polarization step. The qubit thus acts as a complete feedback loop, and the outcome of its measurement of the controlled variable does not need to be known to the outside world.Electron-nuclear feedback mechanisms were previously observed in resonance locking experiments under microwave [15] and optical irradiation [16][17][18]. In Refs. [15,17], a narrowing of the hyperfine field was inferred from the observed bidirectional polarization keeping the system on resonance, but not experimentally verified. In Ref.[16], narrowing was detected spectroscopically, and Ref.[18] studied an ensemble of optically controlled quantum dots. Here, we directly measure the narrowed distribution of the hyperfine field and the dephasing time, T * 2 , of a single, electrically controlled qubit. T * 2 is enhanced by nearly an order of magnitude. In contrast to previous experiments, where the feedback mechanism is intrinsic to the polarization...

Nuclear spin effects in semiconductor quantum dots

“…Data from Ref. 44 and 4c) transfers up to one unit of angular momentum into the nuclear spin bath. Finally, one electron is pushed out of the double dot and the next cycle begins.…”

Collective nuclear stabilization in single quantum dots by noncollinear hyperfine interaction

“…The latter approach, referred to as nuclear stabilization (NS), has the merit that the prolongation of the electron spin coherence time persists for a long time, so that NS can be temporally separated from subsequent qubit operations. Intensive research efforts have led to successful NS in QD ensembles 4 and suppression of the fluctuation of the nuclear field difference between two coupled QDs 5 .…”

Dephasing time of GaAs electron-spin qubits coupled to a nuclear bath exceeding 200 μs

“…We have also verified that the Hahn-echo lifetime is not significantly affected by dynamic nuclear polarization, which can be used to increase T 2 * (see ref. 28 and Supplementary Information).…”