Quantum information (QI) has become a focus of research during the past two decades, with the goal of exploiting the potentialities offered by superposition and entanglement of quantum states (1). The first hardware implementations of QI relied on quantum systems hosting clean, well-isolated two-level systems such as atoms or ions. Despite the success of these systems, solid-state QI implementations promise robustness, miniaturization, established fabrication tools, scalability to large numbers of involved components, and easy connectivity to classical hardware. However, a major challenge is that the interaction of quantum states with the many-body environment in a crystal can compromise QI. In semiconductor nanostructures, the lattice nuclei offer sufficiently long QI storage times, but a truly coherent interface, which is needed to store QI faithfully in an ensemble of nuclear spins, has remained elusive. On page 62 of this issue, Gangloff et al. (2) report the realization of this goal for an electron with the nuclei in a quantum dot, which they achieved by exploiting hyperfine interactions.
A three-step optical protocol developed by Gangloff et al. allows an electron spin in a quantum dot to couple to the nuclear-spin bath.
GRAPHIC: C. BICKEL/SCIENCE
The most promising candidates for quantum bits in the solid state are defects (3), such as nitrogen vacancies in diamond, which are well isolated from their surroundings (4). Atom-like or, more precisely, defect-like localization can be mimicked with quantum dot structures with threedimensional carrier confinement (5). This approach has led to valuable QI hardware achievements, such as efficient and highquality single or entangled photon sources (6). As stationary qubits, quantum dots offer charge coherence times in the nanosecond range and spin coherence times in the microsecond range, but these values are still orders of magnitude shorter than other solid-state competitors (7). The nuclei of the quantum dot are responsible for this shortcoming. At cryogenic temperatures of a few kelvin, the lattice vibrations are frozen out, so that the main interaction that quantum dot carrier spins undergo is the hyperfine coupling with the lattice nuclei.
Long-lived spin coherence of lattice nuclei is well known from nuclear magnetic resonance studies, so nuclei have been considered as a resource into which QI from the carrier spins can be shuffled and stored with possible subsequent recovery through this coupling. However, the nuclear spin bath is highly complex. In the quantum dots studied by Gangloff et al., this bath is formed by tens of thousands of nuclei of indium, gallium, and arsenic atoms that form a very inhomogeneous ensemble. This inhomogeneity is the result of the spins being strongly perturbed by electric fields created by strain in these structures that are grown by molecular beam epitaxy. Further, even at low temperatures of a few kelvin, the nuclear system is “hot” and disordered. Thus, coupling of an electron spin to the nuclei is generally considered to be detrimental, as QI carried by carrier spins quickly dissipates, which limits the carrier spin coherence in quantum dots (7).
In order to use the nuclear bath as a quantum resource, a strong, nondissipative coupling would need to be established between carrier and nuclear spins, which until now has been considered almost impossible for quantum dots. However, hints that the hyperfine interaction does not necessarily have to act destructively could be found, such as in studies of the electron spin precession in a quantum dot ensemble in a magnetic field. Here, a nuclear magnetic field is established in each dot through electron-nuclei flip-flop processes so that the otherwise very inhomogeneous, broadly distributed precession frequencies are focused onto very few modes or even a single mode (8).
The success of the approach taken by Gangloff et al. depended on several breakthroughs. The authors initially cooled the nuclear spin systems by using optical pulses to trigger Raman transitions of the system. The electron spin in a single quantum dot was oriented, and then its polarization was transferred to the nuclei. For sufficiently long pumping under optimized conditions, a nuclear temperature as low as 200 µK was achieved, far below the 2 K temperature of the hot crystal.
In this regime, the authors found that the electron spin and nuclear ensemble formed a strongly coupled state. They could map out spectroscopically a change of total nuclear spin by a single unit of angular momentum. This change corresponded to a single nuclear magnon, the elementary quantum unit of a collective nuclear wave (see the figure).
Finally, Gangloff et al. coherently manipulated the coupled electron-nuclei entity in which a large fraction of quantum dot nuclei are involved, which launched a nuclear magnon by all-optical means. This process corresponds to a coherent, nondissipative exchange between the electron spin and the collective of nuclei—a prerequisite for a quantum memory.
These results could be the first step toward the development of a quantum memory interface with sufficiently long coherence time with quantum dots. The coherence benefits from the many-body nature of the nuclear system, which provides robustness. Moreover, every quantum bit would be associated with its dedicated inherent memory by default. This step has been the missing piece of the puzzle for a semiconductor nanostructure QI platform. As for carrier spin quantum bits, other key demonstrations (6) have been provided already, such as efficient initialization and manipulation on time scales of nanoseconds or even shorter as well as the efficient interconversion with photons for information transfer. Also, at a fundamental level, these results are highly interesting because a quantum many-body state for the nuclei has been established that can be coherently manipulated optically through the electron spin. It should be possible to create specific nonclassical nuclear states, such as Schrödinger cat states.
(원문: 여기를 클릭하세요~)