Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

양자 컴퓨팅 소자의 상호작용 계수 추정과 자동화 제어를 위한 머신러닝 모델

정경훈

Published Jan 1, 2024• 0 citations
OpenAlex

Abstract

This dissertation introduces machine learning methodologies for the analysis and control of quantum systems, focusing on two distinct quantum platforms: nitrogen-vacancy (NV) centers in diamond and gate-defined semiconductor quantum dots (QDs). Firstly, we present a novel approach for identifying individual nuclear spins interacting with an NV defect center in diamond. The NV center, an optically active color center, consists of a substitutional nitrogen atom and an adjacent vacancy in the diamond lattice. The spin state of an electron in the NV center is influenced by hyperfine interactions with the nuclear spins of 13C isotopes. We developed a deep neural network-based algorithm to efficiently and automatically analyze these nuclear spins, leveraging CPMG-type dynamical decoupling spectroscopy, which is a foundational technique for single nuclear spin detection and control. Our deep learning approach facilitates the rapid and automatic detection of nuclear spins by decomposing complex spectral data, which has previously been challenging to analyze. This advancement enables efficient imaging of complex spin samples and automatic characterization of large spin- qubit registers. Secondly, we address the challenges associated with gate-defined semiconductor QDs, a leading platform for quantum computation due to their compatibility with semiconductor technology and tunability. As the number of QDs increases, their tuning complexity and susceptibility to time-varying noise necessitate frequent calibrations of device parameters. To mitigate these issues, we explore reinforcement learning for automating charge state tuning and device calibration in semiconductor quantum dot (QD) systems. We developed a simulated environment to model QD operations and trained a neural network using reinforcement learning. Our results demonstrate the automation of desired QD regime preparation and parameter calibration, optimizing the fidelity of spin qubits. Overall, this work demonstrates significant advancements in the automation and efficiency of quantum system characterization and control, paving the way for more practical and scalable quantum computing applications.

Related papers

Quantum control of single spins and single photons in diamond

Toeno van der Sar

This thesis describes a series of experiments on the control of the optical properties of the nitrogen-vacancy (NV) center in diamond, and on control of the electron and nuclear spin states associated with the NV center. The NV center is a fluorescing atomic defect center in diamond, consisting of a substitutional nitrogen atom adjacent to a vacancy in the diamond carbon lattice. The electron spin of the NV center can be initialized and read out optically, and coherently manipulated with high fidelity even at room temperature. This unique combination of properties has attracted much attention to the NV center in recent years for application in quantum information technologies. The experiments in this work focus on exploiting the intrinsic hybrid nature of the NV center. The NV center forms a hybrid spin register, as its electron spin is always coupled to the nuclear spin of its own nitrogen nucleus, and can furthermore be coupled to additional nuclear carbon-13 spins in the diamond. In addition, the electronic spin state can be mapped onto the state of a photon, in principle allowing long range transmission of quantum information and measurement-based entanglement of distant NV center registers. The long-term goal is to achieve a large scale quantum network where the different constituents each perform the task they are most well suited for: the stable nuclear spins store the quantum information, photons transfer quantum information over long distances, and the fast electron spin interfaces between spin and photon states. For measurement-based entanglement of distant NV centers it is of primary importance to maximize the emission and detection of coherent photons emitted by the NV center. Spontaneous emission can be controlled and enhanced by coupling an emitter to a photonic cavity. Chapters 3, 4, and 5 of this thesis describe experiments aimed at maximizing the coherent photon emission rate of an NV center by coupling the NV center to a photonic crystal cavity. One of the main challenges is to position a single NV center into the nanometer-sized cavity mode. Chapter 3 describes the development of a nanopositioning technique, which allows diamond nanocrystals ~50 nm in size containing single NV centers to be placed at arbitrary locations on a chip, with nanometer precision. By using a sharp etched tungsten tip, individual nanocrystals can be picked up, moved and placed under real-time imaging with an electron microscope. We explicitly demonstrate that the unique optical and spin properties of the NV center are conserved by the nanopositioning process. In chapter 4, we apply the nanopositioning technique to couple single NV centers contained in a diamond nanocrystals to gallium phosphide photonic crystal cavities. These cavities resonate in the visible with high quality factors, and are designed to have the lowest energy mode coincide with the zero-phonon line of the NV center. Efficient coupling is evidenced by a strong enhancement of NV center emission at the cavity wavelength. The high quality factor of a photonic crystal cavity is a result of careful design and fabrication. The introduction of a nanocrystal may therefore be expected to have a detrimental effect on the optical properties of a cavity. In chapter 5 we investigate the effect of a nanocrystal on the optical properties of photonic crystal cavities. Our simulations and measurements show that the effect of a nanocrystal is in fact only minor, a promising result for future work on cavity-QED systems in the solid state and with diamond defect centers in particular. Understanding and mitigating decoherence is a key challenge for quantum science and technology. The main source of decoherence for solid-state spin systems is the uncontrolled spin bath environment. Chapter 6 describes experiments that exploit quantum control of the NV center and recently developed dynamical decoupling techniques to study the properties of the surrounding bath of spins belonging to single substitutional nitrogen atoms. The coherence properties of a single NV center are affected by changes in the state of the bath spins, allowing a single NV center to be used as a probe to detect bath spin resonances and study the quantum dynamics of the spin bath. We use quantum control of the spin bath to extend the dephasing time of the NV center, important for the application of the NV center in DC magnetometry. In chapter 7 we describe the development and implementation of decoherence protected quantum gates for the hybrid spin register formed by the NV center electron spin that is coupled to the nuclear spin of its own nitrogen atom. Electron and nuclear spins evolve and decohere at vastly different rates, making it challenging to use such a hybrid system as a fully functional quantum register. We create a universal set of two-qubit quantum gates by integrating dynamical decoupling techniques into the gate operation. We demonstrate the power of our gate design by implementing for the first time Grover's quantum search algorithm on a solid-state spin register.

•Mar 27, 2012•1 citation

Towards a room temperature solid state quantum processor - the nitrogen-vacancy center in diamond

Philipp Neumann

This work investigates the applicability of the nitrogen-vacancy (NV) defect center in diamond for quantum computation and metrology purposes. The NV center is a molecule like defect in the diamond lattice which is optically active and possesses an electron spin in its ground state. The spin can be used as a quantum bit and as a field sensor. The NV center has three major properties enabling individual control of single defects. Its optical properties depend on the spin state, the spin can be polarized optically and the spin state has an exceptionally long lifetime. Single NV centers can be controlled by optical and microwave radiation. Generally, the research on individually controllable quantum systems is not only motivated by quantum computation and communication. With the ongoing miniaturization nowadays electronic and electro-optic devices will soon reach the limit of single quantum building blocks. While in these applications the full complexity of single quantum systems might not be needed, small sensors based on single atoms or molecules might very well take advantage of the susceptibility of quantum coherences to environmental conditions. Major challenges for all approaches towards single quantum system control are the individual addressing, the measurement of the quantum state and the protection of quantum coherence from environmental noise. Today, there is a variety of systems (including the NV center) that allow control on the single quantum level. These are for example single photons, single ions and atoms in traps, single spins in semiconductors or single molecules in solid state systems. Rather mesoscopic examples are superconducting devices, nanomechanical oscillators or surface plasmon polaritons. Regarding the NV center it has been shown that the magnetic moment of its electron spin allows individual sensing and coupling to other surrounding electron and nuclear spins. Especially single nuclear spins were shown to be exceptional quantum memories. In this work small quantum registers of up to 5 spins have been used to test major requirements for a feasible quantum processor. Therefore, mainly artificially created high purity diamond samples with controlled spin and impurity concentration were used. The NV centers were either grown in or created via ion implantation techniques in a controlled fashion. As a first requirement for quantum computation entanglement among up to three spins has been generated and was verified by quantum state tomography. This proves the access of the full Hilbert space. Especially, entanglement among nuclear spins was very long-living. In further experiments this processor was used for first quantum algorithms like super-dense coding and the Deutsch algorithm. In addition, this work demonstrates new methods to initialize nuclear spin qubits into a desired quantum state. This is particularly important as improper initialization inhibits quantum computation. One of these methods is related to dynamic nuclear spin polarization and the other to measurement based state preparation. Nuclear spins in the diamond lattice originate mainly from the C-13 isotope. By artificially altering its content the influence on the NV center spin was investigated. The coherence times of the NV center were increasing inversely with the nuclear spin density. Even increasing nuclear spin density to offer more nuclear spin qubits to an operational quantum register was possible. As an alternative route of scaling up the quantum register pairs of closely spaced NV centers were investigated. For the first time two qubit gates among two NV centers mediated by magnetic dipolar coupling have been realized. Furthermore the sensing ability was demonstrated by measuring the relative distance of the two centers (10 nm) with sub-nm precision. Eventually, this work contains new insight into the electronically excited state of the NV center which allowed a more advanced control of the electron and nuclear spins. A comparably high magnetic field was used to decouple electron and nuclear spins. Finally, this allowed quantum nondemolition measurements on single nuclear spins which manifests in the observation of quantum jumps. The new insight was used to drastically improve the performance of quantum state readout. Summarizing, it was shown that single NV centers and proximal nuclear spins form versatile small quantum registers which can be interconnected with neighboring such NV quantum registers. Although the number of qubits used in this work is still small the experimental results propose a way of how to proceed when building a quantum computer. Regarding quantum metrology, however, the use of such a small register which can sense fields and store and process the results would already outperform a single NV electron spin by orders of magnitude.

•Jan 1, 2012•11 citations

The nitrogen vacancy center in internal and external fields

Florian Dolde

This dissertation investigates applications of the Nitrogen vacancy in diamond (NV). The NV is a unique quantum system allowing for optical polarization and read out even at ambient conditions. Its electron spin (S=1) has coherence times in the order of milliseconds and is therefore an ideal candidate to investigate the foundations of quantum mechanics as well as exploiting quantum effects in metrology and information processing applications. To utilize the NV for quantum technologies, first the interaction of the NV with its surroundings has to be investigated and controlled. Here dynamical decoupling sequences (CPMG) could be utilized to extend the NV coherence time to a few milliseconds by decoupling the NV from its surrounding spin bath (given by the 13C isotope, natural abundance 1.1%). But not only decoupling sequences can be used to suppress the spin bath interaction, also eigen state tailoring can suppress the bath interaction. As long as strain is the dominant interaction, the NV eigen energies are not susceptible to small magnetic field changes. Therefore the magnetic dipole interaction with the spin bath is suppressed. But the spin bath is not only a nuisance limiting NV coherence times, but also a potential resource for quantum technologies. In order to harness the individual 13C nuclear spins, their interaction with the NV and among themselves has to be investigated. To not be limited by the NV coherence times, a new spectroscopy method, only limited by the NV lifetime, was developed. This can in principle be extended to the second range by using a nuclear spin memory. First advances in quantum technology were demonstrated in the field of metrology, where the Zeeman effect of the NV center allowed for precise nanoscale measurement of magnetic fields. Here this concept was extended to us the Stark effect to detect electric fields. A sensitivity of η=142.6±3.6 V/(cmHz1/2) was demonstrated. This is equivalent to the detection of a single elemental charge at a distance of 150 nm in one second. By using two NVs with a distance of less then 20 nm, a demonstration of nanoscale single fundamental charge detection was feasible. Here one NV was used as a controllable fundamental charge while the other NV was used as sensor NV, detecting the electric field of the fundamental charge. This allowed for the detection of a single electronic charge at ambient conditions. Another keystone of quantum technology is the reliable on demand or heralded creation of entanglement. In this dissertation, the successful entanglement of two electron spins in solid state at ambient conditions is demonstrated. The fundamental interaction used to create an entangled state was dipolar coupling between the two electron spins. In order to be able to harness weaker coupling strengths, an entanglement protocol based on dynamical decoupling was developed. A fidelity of F=0.67±0.04 was demonstrated. By using the intrinsic nuclear spins of the nitrogen it was possible to store the entangled state on the millisecond time scale. However, the theoretical limit for the fidelity given by the polarization and the coherence times is F=0.849. The discrepancy can be explained by pulse errors. These can be avoided by using optimal control yielding a fidelity of F=0.824±0.015. Also optimal control allowed for nuclear spin entanglement by storing the electron spin entanglement on the nitrogen nuclear spins. These experiments are first steps towards a room temperature quantum register, where the electron spins are used as bus to transport information and the nuclear spins act as a memory.

•Jan 1, 2014•2 citations

State Preparation and Metrology of Nitrogen Nuclear Spin in Diamond

Kilhyun Bang

A negatively-charged nitrogen-vacancy (NV) center in diamond is a promising system for quantum computation and quantum information. It has the diamond structure with a substitutional nitrogen atom and a neighboring vacancy. An electron spin in the NV center has an exceptionally long coherence time at room temperature. Thus the NV center has a potential to realize a room-temperature quantum computer which is more efficient than a classical computer. In this dissertation, we focus on the nitrogen nuclear spin as well as the electron spin in the NV center. Every NV center has the nitrogen nuclear spin. Because of the long coherence time of the nitrogen nuclear spin, it is a good candidate for a quantum memory. Thus it is important to prepare the nitrogen nuclear spin qubit in a given pure state for quantum computation. We provide a theoretical understanding of the popular nuclear spin initialization technique. Furthermore, we propose an optimal condition for the initialization of the nitrogen nuclear spin by including the local strain in the NV center. We expect that this optimal condition can improve the purity of the nuclear spin initialization. We also propose an efficient quantum measurement protocol for the hyperfine interaction between the electron spin and the ¹⁵N nuclear spin in the NV center. A precise knowledge of the hyperfine interaction is important to reduce an error in a coherent control of the ¹⁵N nuclear spin. In this protocol, a sequence of quantum operations with successively increasing duration is utilized to estimate the hyperfine interaction with successively higher precision approaching the quantum metrology limit. Unlike common quantum metrological methods, this protocol does not need the preparation of the nuclear spin in a pure state. In the presence of realistic operation errors and electron spin decoherence, we show the overall precision of our protocol still surpasses the standard quantum limit

•Jan 1, 2012•0 citations

Machine learning techniques for state recognition and auto-tuning in quantum dots

Sandesh S. Kalantre, Justyna P. Zwolak, Stephen Ragole +4 more

Abstract Recent progress in building large-scale quantum devices for exploring quantum computing and simulation has relied upon effective tools for achieving and maintaining good experimental parameters, i.e., tuning up devices. In many cases, including quantum dot-based architectures, the parameter space grows substantially with the number of qubits, and may become a limit to scalability. Fortunately, machine learning techniques for pattern recognition and image classification, using so-called deep neural networks, have shown surprising successes for computer-aided understanding of complex systems. We propose a new paradigm for fully automated experimental initialization through a closed-loop system relying on machine learning and optimization techniques. We use deep convolutional neural networks to characterize states and charge configurations of semiconductor quantum dot arrays when only measurements of a current−voltage characteristic of transport are available. For simplicity, we model a semiconductor nanowire connected to leads and capacitively coupled to depletion gates using the Thomas−Fermi approximation and Coulomb blockade physics. We then generate labeled training data for the neural networks, and find at least 90 % accuracy for charge and state identification for single and double dots. Using these characterization networks, we can then optimize the parameter space to achieve a desired configuration of the array, a technique we call “auto-tuning”. Finally, we show how such techniques can be implemented in an experimental setting by applying our approach to an experimental dataset, and outline further problems in this domain, from using charge sensing data to extensions to full one- and two-dimensional arrays, that can be tackled with machine learning.

arXiv:1712.04914•npj Quantum Information•Jan 14, 2019•122 citations

On Scalable Integrated Charge State Tuning for Semiconductor Quantum Dot Devices

Fabian Hader

Semiconductor quantum computing leverages quantum dots as qubits, where achieving high qubit fidelity is one essential requirement for scalability. The formation of qubits and associated sensor dots necessitates meticulous iterative voltage adjustments, a tuning process that is ordinarily executed manually by researchers. While various automated approaches have been proposed, only a few address scalability in the context of a system with a million qubits or consider shifting key data analysis tasks into the cryostat. The latter is necessary to alleviate wiring constraints and to minimize data transmission to room temperature electronics. This thesis explores scalable solutions for automated in-cryostat data analysis, identifying charge transition detection as a pivotal task. A simulation model is developed to generate realistic charge stability diagrams with known ground truth. The model’s output is benchmarked against experimental data to verify fidelity and diversity. Utilizing this model, a comprehensive evaluation of charge transition detection methods is conducted, comparing classical algorithms and a range of machine learning approaches. In consideration of energy-efficient integration, this study includes an initial investigation into reducing the complexity of convolutional neural networks. Subsequent to their optimization using simulated data, the approaches are benchmarked on both simulated and experimental data from diverse qubit samples. When applied to contemporary data qualities, evidence is presented that classical approaches are inadequate in terms of robustness and accuracy while numerous machine learning models demonstrate compelling performance. Among the latter, convolutional neural networks like U-Net are identified to be particularly eligible due to their advantageous structure for efficient hardware integration. Although the U-Net architecture already demonstrates a strong potential for network size reduction under current data quality constraints, a cryogenic hardware integration remains a significant challenge. Consequently, a subsequent study explores if enhanced sensor dot tuning and sensor compensation reduce the complexity of charge transition analysis and if they enable the utilization of ray-based detection methodologies. The findings indicate that particularly sensor compensation has a substantial impact on the data quality and allows a streamlining of the analytical process. However, achieving robust ray-based analysis remains feasible only if consistently reaching the highest observed measurement quality and utilizing machine learning. Even in this case, classical methods and a proposed averaging and thresholding circuit fall short of the requisite detection accuracy. Only with further improvements toward optimal data quality and additional enhancements to the proposed circuit, an integrated solution without machine learning appears attainable. Finally, the energy efficiency of scaled-down neural networks is assessed using state-of-the art hardware accelerators. The findings suggest that the cryogenic integration of machine learning-based charge transition detection is feasible, thus propelling the development of scalable and energy-efficient quantum control architectures.

•Jan 1, 2025•0 citations
This dissertation introduces machine learning methodologies for the analysis and control of quantum systems, focusing on two distinct quantum platforms: nitrogen-vacancy (NV) centers in diamond and gate-defined semiconductor quantum dots (QDs). Firstly, we present a novel approach for identifying individual nuclear spins interacting with an NV defect center in diamond. The NV center, an optically active color center, consists of a substitutional nitrogen atom and an adjacent vacancy in the diamond lattice. The spin state of an electron in the NV center is influenced by hyperfine interactions with the nuclear spins of 13C isotopes. We developed a deep neural network-based algorithm to efficiently and automatically analyze these nuclear spins, leveraging CPMG-type dynamical decoupling spectroscopy, which is a foundational technique for single nuclear spin detection and control. Our deep learning approach facilitates the rapid and automatic detection of nuclear spins by decomposing complex spectral data, which has previously been challenging to analyze. This advancement enables efficient imaging of complex spin samples and automatic characterization of large spin- qubit registers. Secondly, we address the challenges associated with gate-defined semiconductor QDs, a leading platform for quantum computation due to their compatibility with semiconductor technology and tunability. As the number of QDs increases, their tuning complexity and susceptibility to time-varying noise necessitate frequent calibrations of device parameters. To mitigate these issues, we explore reinforcement learning for automating charge state tuning and device calibration in semiconductor quantum dot (QD) systems. We developed a simulated environment to model QD operations and trained a neural network using reinforcement learning. Our results demonstrate the automation of desired QD regime preparation and parameter calibration, optimizing the fidelity of spin qubits. Overall, this work demonstrates significant advancements in the automation and efficiency of quantum system characterization and control, paving the way for more practical and scalable quantum computing applications.
This thesis describes a series of experiments on the control of the optical properties of the nitrogen-vacancy (NV) center in diamond, and on control of the electron and nuclear spin states associated with the NV center. The NV center is a fluorescing atomic defect center in diamond, consisting of a substitutional nitrogen atom adjacent to a vacancy in the diamond carbon lattice. The electron spin of the NV center can be initialized and read out optically, and coherently manipulated with high fidelity even at room temperature. This unique combination of properties has attracted much attention to the NV center in recent years for application in quantum information technologies. The experiments in this work focus on exploiting the intrinsic hybrid nature of the NV center. The NV center forms a hybrid spin register, as its electron spin is always coupled to the nuclear spin of its own nitrogen nucleus, and can furthermore be coupled to additional nuclear carbon-13 spins in the diamond. In addition, the electronic spin state can be mapped onto the state of a photon, in principle allowing long range transmission of quantum information and measurement-based entanglement of distant NV center registers. The long-term goal is to achieve a large scale quantum network where the different constituents each perform the task they are most well suited for: the stable nuclear spins store the quantum information, photons transfer quantum information over long distances, and the fast electron spin interfaces between spin and photon states. For measurement-based entanglement of distant NV centers it is of primary importance to maximize the emission and detection of coherent photons emitted by the NV center. Spontaneous emission can be controlled and enhanced by coupling an emitter to a photonic cavity. Chapters 3, 4, and 5 of this thesis describe experiments aimed at maximizing the coherent photon emission rate of an NV center by coupling the NV center to a photonic crystal cavity. One of the main challenges is to position a single NV center into the nanometer-sized cavity mode. Chapter 3 describes the development of a nanopositioning technique, which allows diamond nanocrystals ~50 nm in size containing single NV centers to be placed at arbitrary locations on a chip, with nanometer precision. By using a sharp etched tungsten tip, individual nanocrystals can be picked up, moved and placed under real-time imaging with an electron microscope. We explicitly demonstrate that the unique optical and spin properties of the NV center are conserved by the nanopositioning process. In chapter 4, we apply the nanopositioning technique to couple single NV centers contained in a diamond nanocrystals to gallium phosphide photonic crystal cavities. These cavities resonate in the visible with high quality factors, and are designed to have the lowest energy mode coincide with the zero-phonon line of the NV center. Efficient coupling is evidenced by a strong enhancement of NV center emission at the cavity wavelength. The high quality factor of a photonic crystal cavity is a result of careful design and fabrication. The introduction of a nanocrystal may therefore be expected to have a detrimental effect on the optical properties of a cavity. In chapter 5 we investigate the effect of a nanocrystal on the optical properties of photonic crystal cavities. Our simulations and measurements show that the effect of a nanocrystal is in fact only minor, a promising result for future work on cavity-QED systems in the solid state and with diamond defect centers in particular. Understanding and mitigating decoherence is a key challenge for quantum science and technology. The main source of decoherence for solid-state spin systems is the uncontrolled spin bath environment. Chapter 6 describes experiments that exploit quantum control of the NV center and recently developed dynamical decoupling techniques to study the properties of the surrounding bath of spins belonging to single substitutional nitrogen atoms. The coherence properties of a single NV center are affected by changes in the state of the bath spins, allowing a single NV center to be used as a probe to detect bath spin resonances and study the quantum dynamics of the spin bath. We use quantum control of the spin bath to extend the dephasing time of the NV center, important for the application of the NV center in DC magnetometry. In chapter 7 we describe the development and implementation of decoherence protected quantum gates for the hybrid spin register formed by the NV center electron spin that is coupled to the nuclear spin of its own nitrogen atom. Electron and nuclear spins evolve and decohere at vastly different rates, making it challenging to use such a hybrid system as a fully functional quantum register. We create a universal set of two-qubit quantum gates by integrating dynamical decoupling techniques into the gate operation. We demonstrate the power of our gate design by implementing for the first time Grover's quantum search algorithm on a solid-state spin register.
This work investigates the applicability of the nitrogen-vacancy (NV) defect center in diamond for quantum computation and metrology purposes. The NV center is a molecule like defect in the diamond lattice which is optically active and possesses an electron spin in its ground state. The spin can be used as a quantum bit and as a field sensor. The NV center has three major properties enabling individual control of single defects. Its optical properties depend on the spin state, the spin can be polarized optically and the spin state has an exceptionally long lifetime. Single NV centers can be controlled by optical and microwave radiation. Generally, the research on individually controllable quantum systems is not only motivated by quantum computation and communication. With the ongoing miniaturization nowadays electronic and electro-optic devices will soon reach the limit of single quantum building blocks. While in these applications the full complexity of single quantum systems might not be needed, small sensors based on single atoms or molecules might very well take advantage of the susceptibility of quantum coherences to environmental conditions. Major challenges for all approaches towards single quantum system control are the individual addressing, the measurement of the quantum state and the protection of quantum coherence from environmental noise. Today, there is a variety of systems (including the NV center) that allow control on the single quantum level. These are for example single photons, single ions and atoms in traps, single spins in semiconductors or single molecules in solid state systems. Rather mesoscopic examples are superconducting devices, nanomechanical oscillators or surface plasmon polaritons. Regarding the NV center it has been shown that the magnetic moment of its electron spin allows individual sensing and coupling to other surrounding electron and nuclear spins. Especially single nuclear spins were shown to be exceptional quantum memories. In this work small quantum registers of up to 5 spins have been used to test major requirements for a feasible quantum processor. Therefore, mainly artificially created high purity diamond samples with controlled spin and impurity concentration were used. The NV centers were either grown in or created via ion implantation techniques in a controlled fashion. As a first requirement for quantum computation entanglement among up to three spins has been generated and was verified by quantum state tomography. This proves the access of the full Hilbert space. Especially, entanglement among nuclear spins was very long-living. In further experiments this processor was used for first quantum algorithms like super-dense coding and the Deutsch algorithm. In addition, this work demonstrates new methods to initialize nuclear spin qubits into a desired quantum state. This is particularly important as improper initialization inhibits quantum computation. One of these methods is related to dynamic nuclear spin polarization and the other to measurement based state preparation. Nuclear spins in the diamond lattice originate mainly from the C-13 isotope. By artificially altering its content the influence on the NV center spin was investigated. The coherence times of the NV center were increasing inversely with the nuclear spin density. Even increasing nuclear spin density to offer more nuclear spin qubits to an operational quantum register was possible. As an alternative route of scaling up the quantum register pairs of closely spaced NV centers were investigated. For the first time two qubit gates among two NV centers mediated by magnetic dipolar coupling have been realized. Furthermore the sensing ability was demonstrated by measuring the relative distance of the two centers (10 nm) with sub-nm precision. Eventually, this work contains new insight into the electronically excited state of the NV center which allowed a more advanced control of the electron and nuclear spins. A comparably high magnetic field was used to decouple electron and nuclear spins. Finally, this allowed quantum nondemolition measurements on single nuclear spins which manifests in the observation of quantum jumps. The new insight was used to drastically improve the performance of quantum state readout. Summarizing, it was shown that single NV centers and proximal nuclear spins form versatile small quantum registers which can be interconnected with neighboring such NV quantum registers. Although the number of qubits used in this work is still small the experimental results propose a way of how to proceed when building a quantum computer. Regarding quantum metrology, however, the use of such a small register which can sense fields and store and process the results would already outperform a single NV electron spin by orders of magnitude.
This dissertation investigates applications of the Nitrogen vacancy in diamond (NV). The NV is a unique quantum system allowing for optical polarization and read out even at ambient conditions. Its electron spin (S=1) has coherence times in the order of milliseconds and is therefore an ideal candidate to investigate the foundations of quantum mechanics as well as exploiting quantum effects in metrology and information processing applications. To utilize the NV for quantum technologies, first the interaction of the NV with its surroundings has to be investigated and controlled. Here dynamical decoupling sequences (CPMG) could be utilized to extend the NV coherence time to a few milliseconds by decoupling the NV from its surrounding spin bath (given by the 13C isotope, natural abundance 1.1%). But not only decoupling sequences can be used to suppress the spin bath interaction, also eigen state tailoring can suppress the bath interaction. As long as strain is the dominant interaction, the NV eigen energies are not susceptible to small magnetic field changes. Therefore the magnetic dipole interaction with the spin bath is suppressed. But the spin bath is not only a nuisance limiting NV coherence times, but also a potential resource for quantum technologies. In order to harness the individual 13C nuclear spins, their interaction with the NV and among themselves has to be investigated. To not be limited by the NV coherence times, a new spectroscopy method, only limited by the NV lifetime, was developed. This can in principle be extended to the second range by using a nuclear spin memory. First advances in quantum technology were demonstrated in the field of metrology, where the Zeeman effect of the NV center allowed for precise nanoscale measurement of magnetic fields. Here this concept was extended to us the Stark effect to detect electric fields. A sensitivity of η=142.6±3.6 V/(cmHz1/2) was demonstrated. This is equivalent to the detection of a single elemental charge at a distance of 150 nm in one second. By using two NVs with a distance of less then 20 nm, a demonstration of nanoscale single fundamental charge detection was feasible. Here one NV was used as a controllable fundamental charge while the other NV was used as sensor NV, detecting the electric field of the fundamental charge. This allowed for the detection of a single electronic charge at ambient conditions. Another keystone of quantum technology is the reliable on demand or heralded creation of entanglement. In this dissertation, the successful entanglement of two electron spins in solid state at ambient conditions is demonstrated. The fundamental interaction used to create an entangled state was dipolar coupling between the two electron spins. In order to be able to harness weaker coupling strengths, an entanglement protocol based on dynamical decoupling was developed. A fidelity of F=0.67±0.04 was demonstrated. By using the intrinsic nuclear spins of the nitrogen it was possible to store the entangled state on the millisecond time scale. However, the theoretical limit for the fidelity given by the polarization and the coherence times is F=0.849. The discrepancy can be explained by pulse errors. These can be avoided by using optimal control yielding a fidelity of F=0.824±0.015. Also optimal control allowed for nuclear spin entanglement by storing the electron spin entanglement on the nitrogen nuclear spins. These experiments are first steps towards a room temperature quantum register, where the electron spins are used as bus to transport information and the nuclear spins act as a memory.
A negatively-charged nitrogen-vacancy (NV) center in diamond is a promising system for quantum computation and quantum information. It has the diamond structure with a substitutional nitrogen atom and a neighboring vacancy. An electron spin in the NV center has an exceptionally long coherence time at room temperature. Thus the NV center has a potential to realize a room-temperature quantum computer which is more efficient than a classical computer. In this dissertation, we focus on the nitrogen nuclear spin as well as the electron spin in the NV center. Every NV center has the nitrogen nuclear spin. Because of the long coherence time of the nitrogen nuclear spin, it is a good candidate for a quantum memory. Thus it is important to prepare the nitrogen nuclear spin qubit in a given pure state for quantum computation. We provide a theoretical understanding of the popular nuclear spin initialization technique. Furthermore, we propose an optimal condition for the initialization of the nitrogen nuclear spin by including the local strain in the NV center. We expect that this optimal condition can improve the purity of the nuclear spin initialization. We also propose an efficient quantum measurement protocol for the hyperfine interaction between the electron spin and the ¹⁵N nuclear spin in the NV center. A precise knowledge of the hyperfine interaction is important to reduce an error in a coherent control of the ¹⁵N nuclear spin. In this protocol, a sequence of quantum operations with successively increasing duration is utilized to estimate the hyperfine interaction with successively higher precision approaching the quantum metrology limit. Unlike common quantum metrological methods, this protocol does not need the preparation of the nuclear spin in a pure state. In the presence of realistic operation errors and electron spin decoherence, we show the overall precision of our protocol still surpasses the standard quantum limit
Abstract Recent progress in building large-scale quantum devices for exploring quantum computing and simulation has relied upon effective tools for achieving and maintaining good experimental parameters, i.e., tuning up devices. In many cases, including quantum dot-based architectures, the parameter space grows substantially with the number of qubits, and may become a limit to scalability. Fortunately, machine learning techniques for pattern recognition and image classification, using so-called deep neural networks, have shown surprising successes for computer-aided understanding of complex systems. We propose a new paradigm for fully automated experimental initialization through a closed-loop system relying on machine learning and optimization techniques. We use deep convolutional neural networks to characterize states and charge configurations of semiconductor quantum dot arrays when only measurements of a current−voltage characteristic of transport are available. For simplicity, we model a semiconductor nanowire connected to leads and capacitively coupled to depletion gates using the Thomas−Fermi approximation and Coulomb blockade physics. We then generate labeled training data for the neural networks, and find at least 90 % accuracy for charge and state identification for single and double dots. Using these characterization networks, we can then optimize the parameter space to achieve a desired configuration of the array, a technique we call “auto-tuning”. Finally, we show how such techniques can be implemented in an experimental setting by applying our approach to an experimental dataset, and outline further problems in this domain, from using charge sensing data to extensions to full one- and two-dimensional arrays, that can be tackled with machine learning.
Semiconductor quantum computing leverages quantum dots as qubits, where achieving high qubit fidelity is one essential requirement for scalability. The formation of qubits and associated sensor dots necessitates meticulous iterative voltage adjustments, a tuning process that is ordinarily executed manually by researchers. While various automated approaches have been proposed, only a few address scalability in the context of a system with a million qubits or consider shifting key data analysis tasks into the cryostat. The latter is necessary to alleviate wiring constraints and to minimize data transmission to room temperature electronics. This thesis explores scalable solutions for automated in-cryostat data analysis, identifying charge transition detection as a pivotal task. A simulation model is developed to generate realistic charge stability diagrams with known ground truth. The model’s output is benchmarked against experimental data to verify fidelity and diversity. Utilizing this model, a comprehensive evaluation of charge transition detection methods is conducted, comparing classical algorithms and a range of machine learning approaches. In consideration of energy-efficient integration, this study includes an initial investigation into reducing the complexity of convolutional neural networks. Subsequent to their optimization using simulated data, the approaches are benchmarked on both simulated and experimental data from diverse qubit samples. When applied to contemporary data qualities, evidence is presented that classical approaches are inadequate in terms of robustness and accuracy while numerous machine learning models demonstrate compelling performance. Among the latter, convolutional neural networks like U-Net are identified to be particularly eligible due to their advantageous structure for efficient hardware integration. Although the U-Net architecture already demonstrates a strong potential for network size reduction under current data quality constraints, a cryogenic hardware integration remains a significant challenge. Consequently, a subsequent study explores if enhanced sensor dot tuning and sensor compensation reduce the complexity of charge transition analysis and if they enable the utilization of ray-based detection methodologies. The findings indicate that particularly sensor compensation has a substantial impact on the data quality and allows a streamlining of the analytical process. However, achieving robust ray-based analysis remains feasible only if consistently reaching the highest observed measurement quality and utilizing machine learning. Even in this case, classical methods and a proposed averaging and thresholding circuit fall short of the requisite detection accuracy. Only with further improvements toward optimal data quality and additional enhancements to the proposed circuit, an integrated solution without machine learning appears attainable. Finally, the energy efficiency of scaled-down neural networks is assessed using state-of-the art hardware accelerators. The findings suggest that the cryogenic integration of machine learning-based charge transition detection is feasible, thus propelling the development of scalable and energy-efficient quantum control architectures.