Periodic Reporting for period 1 - QuESADILLA (Quantum Engineering of Superconducting Array Detectors In Low-Light Applications)
Okres sprawozdawczy: 2022-09-01 do 2025-02-28
Measuring light is one of the most fundamental tasks in experimental science. The range of optical intensities one may wish to measure is vast, but lower-bounded by the fundamental unit of light: the single photon. Understanding and optimising single-photon measurements is therefore of fundamental importance. Using ground-breaking quantum-enhanced design and state-of-the-art detector array technology, QuESADILLA aims to demonstrate how quantum engineering of single-photon detectors enables optimised optical measurements on multiple degrees of freedom simultaneously. The project is built on the close interaction of underlying measurement theory in combination with state-of-the-art technology for detector fabrication, operation, and analysis. Specifically, we design and build arrays of superconducting single-photon detectors, in combination with transformative technologies.
Superconducting nanowire single-photon detectors (SNSPDs) have revolutionised detection at low light levels, and they are finding new and exciting applications across many different scientific disciplines. Key to unlocking their full potential is a top-down understanding and engineering of their quantum mechanical response. This is achieved through a process called detector tomography, which describes the action of the detector through its "positive operator valued measures" (POVMs). Furthermore, immediately linking these quantum-defined figures of merit to detector fabrication allows optimised detectors to be constructed and tailored to specific measurement tasks. Analogously to engineering the quantum state of light to optimise quantum metrology, QuESADILLA engineers the quantum mechanical response of the detector to optimise measurements.
To achieve this goal, we pursue the following objectives:
1. Demonstrate large-scale quantum engineered coherent detection.
2. Demonstrate POVM-optimised single-photon spectrometry.
3. Demonstrate temporal mode selectivity for single-photon imaging.
All three objectives are underpinned by common toolbox of state-of-the-art SNSPD array fabrication technology, complementary optical and electronic device development, and scalable tomographic analysis.
Superconducting nanowire single-photon detectors (SNSPDs) have revolutionised detection at low light levels, and they are finding new and exciting applications across many different scientific disciplines. Key to unlocking their full potential is a top-down understanding and engineering of their quantum mechanical response. This is achieved through a process called detector tomography, which describes the action of the detector through its "positive operator valued measures" (POVMs). Furthermore, immediately linking these quantum-defined figures of merit to detector fabrication allows optimised detectors to be constructed and tailored to specific measurement tasks. Analogously to engineering the quantum state of light to optimise quantum metrology, QuESADILLA engineers the quantum mechanical response of the detector to optimise measurements.
To achieve this goal, we pursue the following objectives:
1. Demonstrate large-scale quantum engineered coherent detection.
2. Demonstrate POVM-optimised single-photon spectrometry.
3. Demonstrate temporal mode selectivity for single-photon imaging.
All three objectives are underpinned by common toolbox of state-of-the-art SNSPD array fabrication technology, complementary optical and electronic device development, and scalable tomographic analysis.
Objective 1A: Scalable Detector Tomography
In our study “Electrical trace analysis of superconducting nanowire photon-number-resolving detectors” (Schapeler et al., Physical Review Applied 22, 014024 (2024)), we utilized Principal Component Analysis (PCA) to evaluate the photon number resolving capability of SNSPDs. Our work revealed that almost all information about photon number is contained in the rising edge of the SNSPD trace. This definite answer will enable future work that can focus on the optimization of photon-number resolution with SNSPDs.
We have also successfully demonstrated scalable quantum detector tomography (Schapeler et al., “Scalable quantum detector tomography by high-performance computing”. Quantum Science and Technology 10, 015018 (2025) - in press) in collaboration with the Paderborn Center for Parallel Computing (PC2). In this work, we have reconstructed the POVMs of a temporally multiplexed detector capable of detecting millions of photons.
Objective 1B: Quantum enhanced coherent detection
Currently, a theoretical study on homodyne detection with squeezed local oscillator is in progress. Alongside, we are developing a spontaneous parametric down-conversion (SPDC) source to serve as a bright squeezed light source to investigate this approach experimentally.
We have also investigated balanced homodyne detection with SNSPDs (Protte et al., “Low-noise balanced homodyne detection with superconducting nanowire single-photon detectors.” Optica Quantum 2.1: 1-6 (2024)) and showed that SNSPDs can be used to measure continuous variables of optical quantum states. By interfering a continuous-wave local oscillator with the vacuum state, we could show that the variance of the difference in count rates of two SNSPDs is linearly proportional to the incident photon flux and that the resulting shot-noise clearance is suited to measure large amounts of squeezing.
Objective 2A: Single-photon microspectrometry with superconducting detector arrays
Work on this objective began in earnest with the acquisition of the superconducting thin-film sputtering system.
Objective 2B: Spectral decomposition & optimisation of detector response
We made first steps toward spectrally resolved POVMs by investigating the manual polarization control and optical attenuation in optical fibers for the telecom range. In a Bachelor project, we used wavelength ranges from 910 nm to 980 nm and from 1440 nm to 1640 nm together with manual fiber polarization controllers optimized for a wavelength of 1550 nm and investigated whether it is possible to access the full Poincaré-Sphere.
Objectives 3A & 3B: Temporal mode selectivity with ultrafast cryogenic bias electronics; High dynamic range measurement of continuous illumination
Our work on cryogenic electronic gating of SNSPDs began prior to the start of the project, with the first publication (Thomas Hummel, et al "Nanosecond gating of superconducting nanowire single-photon detectors using cryogenic bias circuitry," Opt. Express 31, 610-625 (2023)) submitted outside the reporting period. Nevertheless, we have continued this research line with a number of other cryogenic electronic components in other contexts, as well as optical approaches to SNSPD bias control (Thiele et al., APL Photonics 9, 076118 (2024)).
In our study “Electrical trace analysis of superconducting nanowire photon-number-resolving detectors” (Schapeler et al., Physical Review Applied 22, 014024 (2024)), we utilized Principal Component Analysis (PCA) to evaluate the photon number resolving capability of SNSPDs. Our work revealed that almost all information about photon number is contained in the rising edge of the SNSPD trace. This definite answer will enable future work that can focus on the optimization of photon-number resolution with SNSPDs.
We have also successfully demonstrated scalable quantum detector tomography (Schapeler et al., “Scalable quantum detector tomography by high-performance computing”. Quantum Science and Technology 10, 015018 (2025) - in press) in collaboration with the Paderborn Center for Parallel Computing (PC2). In this work, we have reconstructed the POVMs of a temporally multiplexed detector capable of detecting millions of photons.
Objective 1B: Quantum enhanced coherent detection
Currently, a theoretical study on homodyne detection with squeezed local oscillator is in progress. Alongside, we are developing a spontaneous parametric down-conversion (SPDC) source to serve as a bright squeezed light source to investigate this approach experimentally.
We have also investigated balanced homodyne detection with SNSPDs (Protte et al., “Low-noise balanced homodyne detection with superconducting nanowire single-photon detectors.” Optica Quantum 2.1: 1-6 (2024)) and showed that SNSPDs can be used to measure continuous variables of optical quantum states. By interfering a continuous-wave local oscillator with the vacuum state, we could show that the variance of the difference in count rates of two SNSPDs is linearly proportional to the incident photon flux and that the resulting shot-noise clearance is suited to measure large amounts of squeezing.
Objective 2A: Single-photon microspectrometry with superconducting detector arrays
Work on this objective began in earnest with the acquisition of the superconducting thin-film sputtering system.
Objective 2B: Spectral decomposition & optimisation of detector response
We made first steps toward spectrally resolved POVMs by investigating the manual polarization control and optical attenuation in optical fibers for the telecom range. In a Bachelor project, we used wavelength ranges from 910 nm to 980 nm and from 1440 nm to 1640 nm together with manual fiber polarization controllers optimized for a wavelength of 1550 nm and investigated whether it is possible to access the full Poincaré-Sphere.
Objectives 3A & 3B: Temporal mode selectivity with ultrafast cryogenic bias electronics; High dynamic range measurement of continuous illumination
Our work on cryogenic electronic gating of SNSPDs began prior to the start of the project, with the first publication (Thomas Hummel, et al "Nanosecond gating of superconducting nanowire single-photon detectors using cryogenic bias circuitry," Opt. Express 31, 610-625 (2023)) submitted outside the reporting period. Nevertheless, we have continued this research line with a number of other cryogenic electronic components in other contexts, as well as optical approaches to SNSPD bias control (Thiele et al., APL Photonics 9, 076118 (2024)).
Together with project partners from the High-Performance Computing Group in Paderborn and the Paderborn Center for Parallel Computing in Paderborn we have developed novel computational methods to perform large-scale quantum detector tomography. These methods utilize high-performance computing hardware and software and enable efficient parallelization to analyze problems in experimental quantum photonics. Concretely, we were able to reconstruct the quantum mechanical description of a quantum photonic detector, which can detect up to a million photons, and is described by a matrix with one hundred million elements. Building on this experimental result, we demonstrate the scalability of our computational method and reconstruct a matrix with one trillion elements (from simulated data) in just 66 minutes. This shows the unprecedented scale to which this method can be applied to. This work is published in Quantum Science and Technology “Scalable quantum detector tomography by high-performance computing” (Quantum Science and Technology 10, 015018 (2025) - in press). The solver is openly available online on GitHub (https://github.com/pc2/pqdts(odnośnik otworzy się w nowym oknie)) and can be cited via Zenodo (https://doi.org/10.5281/zenodo.10853650(odnośnik otworzy się w nowym oknie)).
We consider our work on scalable quantum detector tomography a significant advancement beyond the state-of-the-art. The utilization of high-performance computing hardware and software and the possible parallelization of our method will enable the investigation and quantum mechanical characterization of quantum photonic systems with unprecedented scales. To give a sense of the scale of the reconstructed matrix with one trillion elements, saving the matrix in dense form takes 33 terabytes of storage. Thus, our work demonstrates the first contribution in an emerging field of high-performance classical-computing-assisted large-scale experimental quantum photonics.
We consider our work on scalable quantum detector tomography a significant advancement beyond the state-of-the-art. The utilization of high-performance computing hardware and software and the possible parallelization of our method will enable the investigation and quantum mechanical characterization of quantum photonic systems with unprecedented scales. To give a sense of the scale of the reconstructed matrix with one trillion elements, saving the matrix in dense form takes 33 terabytes of storage. Thus, our work demonstrates the first contribution in an emerging field of high-performance classical-computing-assisted large-scale experimental quantum photonics.