Bulletin of the American Physical Society
APS March Meeting 2021
Volume 66, Number 1
Monday–Friday, March 15–19, 2021; Virtual; Time Zone: Central Daylight Time, USA
Session X29: Calibration and Gates in Spin Qubit ArraysFocus Session Live
|
Hide Abstracts |
Sponsoring Units: DQI Chair: Aaron Weinstein, HRL Laboratories, LLC |
Friday, March 19, 2021 8:00AM - 8:36AM Live |
X29.00001: Quantum Computing with Spins in Silicon Invited Speaker: Jason Petta We have developed a device architecture that allows for the scalable fabrication of one-dimensional silicon spin qubit arrays [1,2]. Devices fabricated on isotopically enriched 28Si quantum wells allow for high fidelity control of four individually addressable spin qubits. Single qubit gate fidelities exceed 99.9% and we demonstrate ac-driven SWAP gates to transfer spin eigenstates with a fidelity of 98% [3]. The high degree of control offered by the device design allows for the transfer of a single electron across a linear array of nine quantum dots in ~50 ns. With more complex control sequences we perform parallel shuttling of two and three electrons through the array [4]. As a demonstration of automated tuning of dot arrays, we use an image analysis toolbox to automate the calibration of virtual gates in these devices [5]. |
Friday, March 19, 2021 8:36AM - 8:48AM Live |
X29.00002: Tunnel coupling measurement of Si quantum dots based on charge sensing Xinyu Zhao, Xuedong Hu Tunnel coupling in a double quantum dot (DQD) is one of the most important parameters. It determines the charge and spin dynamics, and is essential to many applications in quantum information processing. The most widely used approach to detect tunnel coupling of a GaAs DQDs is based on the charge sensing technique proposed by DiCarlo et. al. [PRL 92, 226801 (2004)]. However, this method can result in significant errors when it is applied directly to a Si DQD since the Si DQD has an extra valley degree of freedom. Here we propose an updated theory to account for the valley dynamics and provide a more accurate description of the tunnel coupling measurement in Si QDs. |
Friday, March 19, 2021 8:48AM - 9:00AM Live |
X29.00003: Systematically tuning a 2xN array of quantum dots with machine learning Giovanni Oakes, Jingyu Duan, John J. L. Morton, Alpha Lee, Charles G Smith, M Fernando Gonzalez-Zalba Spin qubits in quantum dots are a compelling platform for fault-tolerant quantum computing due to the potential to fabricate dense two-dimensional arrays with nearest neighbour couplings, a requirement to implement the sur-face code. However, due to the proximity of the surface gate electrodes cross-coupling capacitances can be substantial, making it difficult to control each quantum dot independently. By extending the number of quantum dots increases the complexity of the calibration process, which becomes impractical to do heuristically. Inspired by recent demonstrations of industry-fabricated silicon quantum dot bilinear arrays, we develop a theoretical framework to tune a 2×N array of quantum dots, based on the gradients in gate voltage space of different charge transitions that can be measured in multiple two-dimensional charge stability diagrams. To automate the process, we successfully train aneural network to extract the gradients from a Hough transformation from a stability diagram and test the algorithm on simulated and experimental data of a 2×2 quantum dot array. |
Friday, March 19, 2021 9:00AM - 9:12AM Live |
X29.00004: Deep Reinforcement Learning for Efficient Measurement of Quantum Devices Sebastian Orbell, Vu Nguyen, Dominic Lennon, Hyungil Moon, Florian Vigneau, Leon Camezind, Liuqi Yu, Dominik Zumbuhl, Andrew Briggs, Michael Osborne, Dino Sejdinovic, Natalia Ares Deep reinforcement learning is an emerging machine learning approach which can teach a computer to learn from their actions and rewards similar to the way humans learn from experience. It offers many advantages in automating decision processes to navigate large parameter spaces. This paper proposes a novel approach to the efficient measurement of quantum devices based on deep reinforcement learning. We focus on double quantum dot devices, demonstrating the fully automatic identification of specific transport features called bias triangles. Measurements targeting these features are difficult to automate, since bias triangles are found in otherwise featureless regions of the parameter space. Our algorithm identifies bias triangles in a mean time of less than 30 minutes, and sometimes as little as 1 minute. This approach, based on dueling deep Q-networks, can be adapted to a broad range of devices and target transport features. This is a crucial demonstration of the utility of deep reinforcement learning for decision making in the measurement and operation of quantum devices. |
Friday, March 19, 2021 9:12AM - 9:24AM Live |
X29.00005: Qubits in quantum dot arrays made with all-optical, 300mm wafer lithography Anne-Marije Zwerver, Tobias Stefan Krähenmann, Thomas Watson, Lester Lampert, Stephanie Bojarski, Hubert C George, Brennen Mueller, Jim Clarke, Lieven Vandersypen Spin qubits in gate-defined silicon quantum dots are promising qubit candidates due to their small size and relatively long coherence times. To pave the road towards large-scale quantum computing, making use of common CMOS fabrication techniques, like optical lithography and chemical-mechanical polishing is key. Spin qubit devices to date, however, still rely on the flexibility of e-beam lithography. |
Friday, March 19, 2021 9:24AM - 9:36AM Live |
X29.00006: 2x2 quantum dot arrays in Si/SiGe Marcel Meyer, Florian Unseld, Chien-An Wang, Luca Petit, Sergei Amitonov, Harmen Gerrit Johan Eenink, Nico Hendrickx, Kostas Tsoukalas, Anne-Marije Zwerver, Mateusz T Madzik, William Iain Lawrie, Delphine Brousse, Amir Sammak, Giordano Scappucci, Lieven Vandersypen, Menno Veldhorst Great progress has been made with silicon in scaling up the number of quantum dots in linear arrays, with realizations of up to nine quantum dots. However, a crucial requirement for quantum computation is the ability to engineer two-dimensional quantum dot arrays. This step will increase qubit interconnectivity and could enable to perform quantum error correction using the surface code. Here, we present the realization of two-dimensional silicon quantum dot arrays, defined in isotopically enriched Si/SiGe heterostructures using single layer and overlapping gate schemes. We show that quantum dots can be tuned to the single-electron regime, as confirmed by simultaneous charge sensing using two nearby quantum dot charge sensors. We concentrate on experimental results in gaining control over the quantum dot system focusing on quantum information processing with quantum dots realized in two-dimensional arrays. |
Friday, March 19, 2021 9:36AM - 9:48AM Live |
X29.00007: Adiabatic Quantum State Transfer in an Array of Spins Yadav Kandel, Haifeng Qiao, Saeed Fallahi, Geoffrey C. Gardner, Michael Manfra, John Nichol Any scalable quantum processor with many qubits requires interqubit connectivity for the efficient execution of useful quantum algorithms. In recent years, there has been significant progress in both theoretical and experimental efforts aimed at improving connectivity in spin-qubits, which are usually defined in one- or two-dimensional arrays. Here we demonstrate evidence of adiabatic quantum-state transfer (AQT) of single-spin eigenstates and two-spin entangled states in a chain of four spins in gate defined quantum-dots in a GaAs/AlGaAs heterostructure. We transfer spin states from one side of the array to the other side in tens of nanoseconds with simulated transfer probabilities exceeding 90% via adiabatic modulation of the nearest-neighbor exchange couplings between spins. We also demonstrate that this method is scalable for longer arrays of spin-qubits. AQT is robust to noise and pulse-timing errors and it will be useful for initialization, state distribution, and readout in large spin-qubit arrays. |
Friday, March 19, 2021 9:48AM - 10:00AM Live |
X29.00008: Quantum simulation of an antiferromagnetic Heisenberg chain with gate-defined quantum dots Cornelis van Diepen, Tzu-Kan Hsiao, Uditendu Mukhopadhyay, Christian Reichl, Werner Wegscheider, Lieven Vandersypen Emergent phases of strongly-correlated fermions are of central interest in condensed matter physics. Quantum systems with engineered Hamiltonians can be used as simulators of many-body systems to provide insights beyond the capabilities of classical computers. Magnetism naturally arises in the Mott-insulator regime of the Fermi-Hubbard model, where charges are localized and the spin degree of freedom remains. In this regime the occurrence of phenomena such as resonating valence bonds, frustrated magnetism, and spin liquids are predicted. Here we show that semiconductor quantum dots can be used to simulate quantum magnetism in the Mott-insulator regime. For this purpose we demonstrate several techniques including many-body spin-state preparation, singlet-triplet correlation measurements, and characterization of the quantum system with energy spectroscopy and global coherent oscillations. With these techniques we tune and probe a homogeneously coupled Heisenberg spin-chain in a linear array of four single-electron quantum dots, and find good agreement between experiment and numerical simulation. Our demonstrated control and techniques open new opportunities to simulate quantum magnetism, including spin liquid physics and quantum phase transitions. |
Friday, March 19, 2021 10:00AM - 10:12AM Live |
X29.00009: Orthogonal control of tunnel couplings and cascade-based readout in a quantum dot array Tzu-Kan Hsiao, Cornelis van Diepen, Uditendu Mukhopadhyay, Christian Reichl, Werner Wegscheider, Lieven Vandersypen Electrostatically-defined semiconductor quantum dot (QD) arrays offer a promising platform for quantum computation and quantum simulation. However, crosstalk of gate voltages to inter-dot tunnel couplings, and a limited sensitivity range of a charge sensor based on Coulomb repulsion, pose challenges for the control and readout of large-scale QD arrays. Here, we show two new techniques to overcome these issues. First, we present that the crosstalk on tunnel couplings can be efficiently characterized and compensated for, since the same exponential dependence applies to all gates. We demonstrate efficient calibration of crosstalk in a quadruple QD array and define a set of virtual barrier gates, with which we show orthogonal control of all inter-dot tunnel couplings. Next, we report on cascade-based remote, fast, and high-fidelity spin readout. The Coulomb repulsion allows an initial charge transition to induce subsequent charge transitions, inducing a cascade of electron hops. Combining the electron cascade with Pauli spin blockade, we demonstrate fast and high-fidelity readout of distant spins using a remote charge sensor in the QD array. Our work marks a key step forward in the control and readout of large-scale QD arrays. |
Friday, March 19, 2021 10:12AM - 10:24AM Live |
X29.00010: Machine learning enables completely automatic tuning of a quantum device faster than human experts Dominic Lennon, Hyungil Moon, James Kirkpatrick, Nina van Esbroeck, Leon Camenzind, Liuqi Yu, Florian Vigneau, Dominik Zumbuhl, Andrew Briggs, Michael Osborne, Dino Sejdinovic, Edward Laird, Natalia Ares An unavoidable obstacle to creating large circuits with spin qubits is device variability. Due to this variability, bringing a spin qubit into operation conditions requires a large parameter space to be explored. This process is becoming intractable for humans as the complexity of quantum circuits grows. We present a statistical algorithm that utilises machine learning to navigate the entire parameter space. We demonstrate fully automated tuning of a double quantum dot device in under 70 minutes, faster than human experts. This approach also provides a quantitative measurement of device variability, from one device to another and after a thermal cycle. This is a key demonstration of the use of machine learning techniques to explore and optimise the parameter space of quantum devices and overcome the challenge of device variability. |
Friday, March 19, 2021 10:24AM - 10:36AM Live |
X29.00011: Towards autonomous tuning of noisy quantum dots Josh Ziegler, Sandesh S Kalantre, Thomas McJunkin, Mark Eriksson, Jacob Taylor, Justyna Zwolak Gate-defined quantum dots (QDs), in which electrons are trapped in quantum wells defined by gate voltages, are a quantum computing platform that may scale effectively due to the maturity of semiconductor processing. However, initialization of these devices is not trivial and currently performed mostly manually or in a semi-scripted fashion guided by heuristics. There has been some progress towards autonomous tuning of these devices using machine learning (ML) methods, but the current best strategies are not robust to ever-present noise in the system or require human-labelled noisy data. We are working to overcome issues of imperfect devices while eliminating the labor of labelling data by incorporating noise into our QD simulator. With this approach, we broaden the applicability of autonomous tuning methods to less ideal devices while using a scalable simulation-based ML framework. |
Friday, March 19, 2021 10:36AM - 10:48AM Live |
X29.00012: Learning the states of quantum dot systems: The ray-based approach Justyna Zwolak, Thomas McJunkin, Sandesh S Kalantre, Samuel Neyens, Evan R MacQuarrie, Lisa F. Edge, Mark Eriksson, Jacob Taylor Given the progress in the construction of multi-quantum dot (QD) arrays in both 1D and 2D [1,2], it is imperative to replace the current practice of manual tuning to a desirable electronic configuration with a standardized and automated method. Recently, we have experimentally realized an auto-tuning paradigm proposed by Kalantre et al. [3] that combines machine learning (ML) and optimization routines, with ConvNets used to characterize the state and charge configuration of single and double QD states from measurements via the conductance of a nearby charge sensor [4]. Now we expand on this work and propose a novel approach where we use 1D traces (“rays”) measured in multiple directions in the gate voltage space to describe the position of the features characterizing each state (i.e., to “fingerprint” the state space). Using these “fingerprints” instead of 2D scans we train an ML algorithm to differentiate between various state configurations. Here, we report the performance of the ray-based learning on experimental data and compare it with our image-based approach. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2025 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700