Three-stage decoherence dynamics of an electron spin qubit in an optically active quantum dot

Abstract

The control of discrete quantum states in solids and their use for quantum information processing is complicated by the lack of a detailed understanding of the mechanisms responsible for qubit decoherences [1]. For spin qubits in semiconductor quantum dots, phenomenological models of decoherence currently recognize two basic stages [2-4]; fast ensemble dephasing due to the coherent precession of spin qubits around nearly static but randomly distributed hyperfine fields (∼ 2 ns) [5-8] and a much slower process (> 1 µs) of irreversible relaxation of spin qubit polarization due to dynamics of the nuclear spin bath induced by complex many-body interaction effects [9]. We unambiguosly demonstrate that such a view on decoherence is greatly oversimplified; the relaxation of a spin qubit state is determined by three rather than two basic stages. The additional stage corresponds to the effect of coherent dephasing processes that occur in the nuclear spin bath that manifests itself by a relatively fast but incomplete non-monotonous relaxation of the central spin polarization at intermediate (∼ 750 ns) timescales. This observation changes our understanding of the electron spin qubit decoherence mechanisms in solid state systems.

Quantum Effects in Higher-Order Correlators of a Quantum-Dot Spin Qubit

“…This model disregards feedback of the central spin dynamics on the nuclear spin bath, which was proven to be a good approximation due to strong nuclear quadrupole coupling [5,20,26]. We also disregard transverse Overhauser field components since their effect is suppressed due to fast spin precession in the yz-plane.…”

“…indicates an averaging over many identical measurement sequences. In addition to direct evidence for the quantum nature of solid state qubits, we show that our method has practical importance since it provides a completely alternative route for measuring the coherence times of qubits which are typically measured through spin-echo techniques [20,21] or methods in the frequency domain, such as coherent population trapping [22]. Such application of higher order correlators has been theoretically anticipated previously [23] and can be applied to many other quantum systems in the solid state.…”

“…An electron spin is first optically prepared in the |↓ -state in an individual InGaAs QD by using a picosecond laser pulse [3,20], indicated by the label "Pump" in the figure, with a laser power corresponding to a Rabi-π-rotation that generates the neutral exciton state (|cgs → |↓⇑ ). Immediately after exciton generation, the hole escapes the QD within 4 ps [20] leaving behind the single electron spin (|↓⇑ → |↓ ). The implementation of an asymmetric tunnel barrier leads to electron lifetimes extending up to seconds whereas hole lifetimes are unaffected [1].…”

Computing with a single qubit faster than the computation quantum speed limit

“…So, finally, the quantum error correction may appear superior over the classical one because switches between nonorthogonal GHZ-like states do not slow calculations down for n q > 1 qubits in comparison to n q = 1 in N 1 limit. We note, however, that this quantum error correction protects only against specific random qubit-flip errors while operations with entangled states (9) can enhance the probability of the error with all-qubit flips, which cannot be treated with majority voting. We leave this problem for future studies.…”

“…The major two sources of errors that are expected for a spin qubit are the finite fidelity of gates and the uncontrolled environmental magnetic field fluctuations. The former problem is more important because there are already many single and few qubit platforms with long coherence times [8][9][10][11]. A non-perfect quantum gate can be characterized by the typical difference φ 0 between the rotation angle of a qubit and the desired angle π/n.…”