Quantum Gates, Algorithms, and Error Correction with a Neutral Atom Qubit Array

量子计算因其在解决经典计算机上难以解决的实际问题方面的潜力而引起人们的极大兴趣。应用领域包括密码学、数据库搜索、模式分类、求解大型耦合方程系统，以及设计新的功能材料或化合物。量子计算机是由量子比特或量子比特建造的，量子比特存储使用量子逻辑门处理的量子信息。该项目将继续发展量子计算，使用在单个原子的内态中编码的量子比特。虽然目前的技术水平涉及10-20个量子比特的实验，但人们认为需要拥有数千个量子比特的计算机才能实现量子计算的承诺。在这方面，本项目正在开发的基于原子的方法特别有吸引力，因为大量中性原子可以在很近的地方保持和控制，而不会相互干扰。该项目寻求提高几个关键性能指标的最新水平：量子比特长期保持量子状态的能力，逻辑门操作和量子比特测量的保真度，以及单个系统可以准备的量子比特数量。然后，这些进展将被用来演示用于数据库搜索的量子算法。该项目将通过培训学生和博士后研究人员，为科学的劳动力发展做出贡献。培训将是跨学科的，借鉴原子和激光物理、电子和计算机控制系统以及量子信息理论的方法和想法。研究成果将纳入大学教学课程。实验方法将使用捕获在由光定义的势井中的铯和Rb原子的二维阵列。势井将通过组合几种不同频率的激光来准备，这种方式可以保护存储的量子信息不受退相干的影响。一个可移动的光镊子系统将被用来安排俘获的原子，以100%占据多达50个量子比特位的阵列。产生纠缠的量子逻辑门将通过用激光脉冲激励原子到里德堡态来实现。具有定形时间分布的绝热脉冲将被用来提高纠缠操作的保真度。原子态的测量将不会与其他量子位产生串扰，也不会造成原子的损失，使用两种物种的方法，其中一种物种(铯)将用于记忆和量子逻辑，另一种物种(Rb)将用于测量。要测量的量子态将通过物种间里德堡门从铯原子转移到Rb原子。然后将利用这些能力来演示一种多量子比特量子算法，该算法为数据库搜索提供二次加速。该团队还将实施量子纠错，以保护量子比特免受比特翻转或相位翻转错误的影响。

Quantum computing is attracting great interest due to its potential for solving practical problems that are intractable on classical computers. Areas of application include cryptography, database searching, pattern classification, solving large systems of coupled equations, and design of new functional materials or chemical compounds. A quantum computer is built from quantum bits, or qubits, that store quantum information which is processed using quantum logic gates. This project will continue the development of quantum computing using qubits encoded in the internal states of individual atoms. While the current state of the art involves experiments with 10-20 qubits, it is believed that computers with many thousands of qubits will be needed to realize the promise of quantum computation. In this respect the atom based approach being developed in this project is particularly attractive since a large number of neutral atoms can be held and controlled in close proximity without undesired interference with each other. The project seeks to advance the state of the art of several key performance metrics: the ability of qubits to preserve their quantum states for a long time, the fidelity of logic gate operations and qubit measurements, and the number of qubits that can be prepared in a single system. These advances will then be used to demonstrate a quantum algorithm for database searching. The project will contribute to scientific workforce development through training of students and postdoctoral researchers. The training will be interdisciplinary, drawing on methods and ideas from atomic and laser physics, electronic and computer based control systems, and quantum information theory. Research results will be incorporated into the University teaching curriculum. The experimental approach will use a two-dimensional array of cesium and rubidium atoms trapped in potential wells defined by light. The potential wells will be prepared by combining laser light of several different frequencies in a way that protects stored quantum information from decoherence. A movable optical tweezer system will be implemented to arrange trapped atoms for 100% occupancy of an array with up to 50 qubit sites. Quantum logic gates to create entanglement will be performed by exciting atoms to Rydberg states with laser pulses. Adiabatic pulses with shaped temporal profiles will be used to improve the fidelity of the entangling operations. Atomic states will be measured without crosstalk to other qubits or loss of atoms using a two-species approach whereby one species (cesium) will be used for memory and quantum logic, and a second species (rubidium) will be used for measurements. The quantum states to be measured will be transferred from cesium to rubidium atoms using an interspecies Rydberg gate. These capabilities will then be leveraged to demonstrate a multi-qubit quantum algorithm that provides quadratic speedup for database searching. The team will also implement quantum error correction to protect qubits against bit flip or phase flip errors.