【作者】 居琛勇；

【导师】 杜江峰；

【作者基本信息】 中国科学技术大学 ， 核与粒子物理， 2010， 博士

【摘要】 量子计算研究的最终目标是建造实用的高性能量子计算机——一台以量子力学为基本原理对信息进行编码和计算的新型计算机。它在理论上被证明能够完全模拟当前的经典计算机,并且在一些特殊问题上更具有经典计算机无可比拟的优势。量子计算是近十多年来物理学研究中最热门的领域之一,已形成了集量子物理学、数学、材料科学和工程科学等多学科交叉研究的局面,吸引了大量的科研人员投身其中。然而,关于量子计算机最终以何种方式(模式)以及在何种物理体系中实现的问题,目前尚没有定论,仍处于多种途径并行研究的阶段。本文介绍了我们在量子计算新模式和新的物理实现体系上的工作。使用液体核磁共振技术,我们在核磁共振体系中第一次实验模拟了one-way量子计算模式的全过程,包括确定性地制备图态和实现one-way模式下的Deutsch-Josza算法。我们的工作验证了one-way量子计算模式的可行性,同时也将为其它物理体系开展one-way量子计算提供有益经验。在新的物理实现体系上,我们重点关注基于掺杂富勒烯的量子计算物理体系。我们提出了在掺杂富勒烯体系中构筑普适量子逻辑门集合的方案。通过电子自旋为媒介实现相邻核自旋之间的两量子比特逻辑门,这与核自旋上可进行的单量子比特逻辑门一起构成普适量子逻辑门集合。我们的方案将电子自旋3／2及其跃迁存在简并的特点和难点考虑在内,解决了前人方案中遗留的问题,为掺杂富勒烯量子计算的普适性提供了基础。进一步,我们提出了一个以电子自旋为辅助的基于掺杂富勒烯的普适和可扩展量子计算方案。充分利用掺杂富勒烯中电子自旋与核自旋并存的特点,使电子自旋在核自旋量子比特的寻址、初始化、量子逻辑门和读出上都起到重要的作用。与前人的方案相比,该方案在核自旋的寻址和量子逻辑门上具有明显的优势。此外,针对当前掺杂富勒烯仍属于系综实验研究的特点,我们给出了可在当前实验谱仪(含ENDOR功能的脉冲式电子自旋共振谱仪)上完成的自旋量子态重构技术,这为在掺杂富勒烯系综实验平台上开展少数量子比特的量子计算研究提供了方便。我们的工作力图展示将量子计算模式和物理实现这两个量子计算的不同层面紧密结合在一起的特点。我们提出的普适量子逻辑门集合实现方案,不仅是基于逻辑网络的量子计算模式的重要组成部分,也为在掺杂富勒烯体系中应用one-way量子计算模式提供了基础(建立图态)。以电子自旋为辅助的掺杂富勒烯量子计算方案中,我们通过综合使用基于逻辑网络和基于整体控制这两种量子计算模式,解决了核自旋的寻址问题,避开了原先存在的技术难点。我们预期最终量子计算机的实现将有赖于多种量子计算模式和物理实现体系的综合使用,而我们的工作初步展示了这一特征。更多还原

【Abstract】 The ultimate goal of quantum computation research is to build a powerful quantum computer, which is a new type computer based on quantum mechanics. It has been proved in theory that a quantum computer can fully simulate the classical computer, and further it can solve specific problems efficiently which classical computer cannot. Quantum computation research has the multidisciplinary character that combines quantum physics, mathematics, material science, and engineering etc. It has become one of the most active areas in physical research, and attracted lots of the most brilliant brains in the world. However, it is still not clear today that the ultimate quantum computer would be built in which mode and in which physical system. Many possible ways are under research.In this thesis we summarize our researches on the new quantum computation mode and new physical realization system. We experimentally simulated the one-way quantum computation mode in nuclear magnetic resonance (NMR) system for the first time, utilizing the liquid-state NMR technologies. We deterministically prepared a four-particle graph state, and demonstrated the Deutsch-Josza algorithm on it in the one-way manner. The results of our experiment verified the feasibility of the one-way mode, and would be helpful to other physical systems for future scalable one-way quantum computation. For the new physical realization system, we concentrate on the endohedral fullerenes. We proposed a scheme to realize the universal quantum gates in endohedral fullerenes. It solved the problem which due to the electron spin 3/2 and its transition degeneracy. Our scheme is a fundamental contribution to the endohedral fullerene based quantum computation. Further, we suggested a new quantum computation proposal based on endohedral fullerenes, in which the electron spins are used as auxiliary. Utilizing these auxiliary spins, it has more convenience in qubit (the nuclear spin) addressing, initialization, quantum gates, and read out, than previous proposals. Besides, we also provided a spin state tomography technology for the endohedral fullerene ensemble, which is feasible for the current electron spin resonance spectrometer with ENDOR (electron-nuclear double resonance) part.We anticipate that the ultimate realization of quantum computer would rely on the integrated system with different computation modes and even different physical realization systems. Our work have shown this distinguish feature in a primary step. The two-qubit gates which we proposed to construct the universal quantum gates in endohedral fullerenes, is not only an essential part of the logic circuit mode, but also could be used to prepare the graph state onto the nuclear spins which is the unique resource of the one-way mode. We also integrated the logic circuit mode and the global control mode in our electron spin assisted endohedral fullerene quantum computation scheme, according to which we solved the addressing problem of the nuclear spins. This reduces largely the technical requirements.更多还原