Qubits in isotopically enriched 28Si

同位素富集的28Si是一种令人着迷的自旋量子器件材料体系。在磷掺杂的28Si中使用浅供体的电子和核量子位提供了极长的自旋弛豫和消相时间，因此是半导体量子信息技术最有希望的候选者之一。量子逻辑在28Si:P中的主要挑战之一当然是电子和核自旋初始化，这是由Yang等人通过磷供体跃迁的选择性光泵浦实现的。这些光学跃迁的线宽是半导体中浅层给体系综中最窄的线宽之一，并且实验证明了光学自旋初始化程度相当高。然而，到目前为止，自旋极化的程度远低于理想的100%，这很可能与观察到的光谱洞燃烧到供体跃迁有关。本提案的第一个目标是深刻理解这种孔燃烧动力学的非平凡物理。Thewalt组在连续波实验中观察到最小的孔燃烧线宽比从供体束缚离子复合时间计算得到的要大4倍。原因尚不清楚，但可能归因于(a)与光激发的自由电子和空穴的相互作用，(b)由于不可避免的、无意的p型背景掺杂引起的内部电场波动，以及(c)电子-电子相互作用。我们将采用一种特殊的时间分辨泵浦-探针吸收装置来定量地确定不同的贡献，并利用这些信息(a)将光诱导自旋极化优化到99%以上，(b)将能量分离的自旋子系综写入供体束缚电子系综中。本建议的第二个目标是宏观供体自旋综的纠缠。我们将使用优化的自旋初始化，并将两个空间分离的宏观自旋系综写入高质量的28Si样品中。然后，我们用光脉冲对这些自旋综进行了纠缠，并用类自旋噪声技术验证了纠缠。供体束缚的电子自旋综与各自的核自旋综之间的纠缠已经被证明，但据我们所知，到目前为止，还没有在28Si中建立空间分离的宏观供体电子自旋综的纠缠。28Si中的磷供体是量子比特的优秀候选者，但在电信频段中孤立且明亮且以相同波长发射的单一光学缺陷将使可扩展硅量子光子学的实现成为可能。最近对碳注入硅的实验报道了这种单光子发射器，但具有强烈的波长变化。在高质量28Si中寻找具有相同波长的单光子发射器将是本提案的第三个任务。

Isotopically enriched 28Si is a fascinating material system for spin-based quantum devices. Electron and nuclear qubits using shallow donors in phosphorous doped 28Si offer extremely long spin relaxation and dephasing times and are thereby among the most promising candidates for a semiconductor quantum information technology. One of the major challenges of quantum logic in 28Si:P is of course the electron and nuclear spin initialization which has been achieved by Yang et al. by selective optical pumping of the phosphorous donor transitions. The linewidths of these optical transitions are among the narrowest linewidths of shallow donor ensembles in semiconductors and the experimentally demonstrated degree of optical spin initialization is rather high. However, the degree of spin polarization is well below the desirable 100 % so far which is most likely linked to the observation of spectral holes burned into the donor transitions.The first goal of this proposal is a profound understanding of the non-trivial physics of this hole burning dynamics. The group of Thewalt observed in continuous wave experiments minimal hole burning linewidths which are a factor of four larger than calculated from the donor bound trion recombination time. The cause is unknown but might be attributed to a combination of (a) interaction with optically excited free electrons and holes, (b) fluctuating internal electric fields due to unavoidable, unintentional, p-type background doping, and (c) electron-electron interaction. We will apply a special time-resolved pump-probe absorption setup to determine quantitatively the different contributions and use this information (a) to optimize the optically induced spin polarization to well above 99% and (b) to write energetically separated spin sub-ensembles into the donor bound electron ensemble. The second goal of this proposal is entanglement of macroscopic donor spin ensembles. We will use the optimized spin initialization and write two, spatially separated, macroscopic spin ensembles into a high quality 28Si sample. Afterwards, we entangle these spin ensembles by an optical pulse and verify the entanglement by a spin noise like technique. Entanglement of a donor-bound electron spin ensemble with the respective nuclear spin ensemble has been demonstrated before but entanglement of spatially separated macroscopic donor electron spin ensembles has not been established in 28Si so far, to the best of our knowledge.Phosphorous donors in 28Si are excellent candidates for qubits but single optical defects which are isolated and bright and emitting at identical wavelengths in the telecom bands would enable additionally the implementation of scalable silicon quantum photonics. Very recent experiments on carbon implanted Si report on such single photon emitter but with strongly varying wavelengths. The search for single photon emitters with identical wavelengths in high quality 28Si will be the third task of this proposal.