arXiv subjects

Thomas Brunner

Publications and source records attributed to Thomas Brunner.

18 recordsLinked to original sources

arXiv2026

An optical cryostat for automated testing of silicon photomultipliers with vacuum-ultraviolet light

We present the design and performance of an optical cryostat for testing silicon photomultipliers (SiPMs) with vacuum-ultraviolet (VUV) light. The cryostat operates within a temperature range of 120 to 340 K, achieving a temperature stability of 1.2 mK, measured in a rolling 10-minute window once settled within a liquid nitrogen fill-cycle, and 13.8 mK over a complete fill-cycle including the refill transient. The cryostat includes a VUV-transparent CaF$_2$ viewport with a 142 mm diameter clear aperture, which limits the optically accessible sample area to 158 cm$^2$. The viewport is integrated into a sealed dark box and optical table to facilitate light delivery. VUV illumination is achieved using a custom xenon scintillation lamp or a xenon flash lamp, with a motorized two-axis linear rail system ensuring precise light positioning over the sample. This paper details the cryostat and optical system design, as well as commissioning current-voltage (IV) measurements using Hamamatsu VUV-sensitive SiPMs from 165 to 295 K and a measurement of the photodetection efficiency at 165 K.

physics.ins-det↗

arXiv2026

Generalised Perturbed Convective Wave Theory

The theory of the perturbed convective wave equation for compressible flows (cPCWE) is generalised to spatially varying mean-density fields. The resulting equation is an exact scalar reformulation of the acoustic perturbation equations and describes sound generation and propagation in moving inhomogeneous media using a single unknown. The intermediate variables of the associated workflow, in which a Helmholtz decomposition problem, a Poisson equation and the cPCWE are solved successively, are related to the vortical, entropy and acoustic modes of Kovasznay, providing a physical interpretation of each processing step. The quantitative accuracy is assessed against fully compressible direct numerical simulations (DNS) of two-dimensional isothermal mixing layer and Lighthill's analogy computed in the same framework at Mach numbers between M=0.2 and M=0.4, based on the velocity difference across the layer and the ambient speed of sound. Over this range of Mach numbers, the radiated power spans several orders of magnitude. For M>=0.25, the sound power levels obtained using the three methods agree within 0.9dB, and within 0.5dB for M>=0.3. At M=0.2, where the acoustic fluctuations are weakest relative to the hydrodynamic ones, Lighthill's analogy over-predicts the radiated power by 2.8dB. In contrast, the cPCWE deviates from the DNS reference by only -1.2dB. This closer agreement is because the cPCWE source term is confined to the vortex-pairing region, while convection and refraction are represented by its convective wave operator. Beyond reproducing the far-field sound, the cPCWE resolves the acoustic field within the shear zone itself, where the DNS' fields are masked by vortical fluctuations.

physics.flu-dyn↗

arXiv2025

Performance of FBK VUV-HD3 and HPK VUV4 SiPMs in the Light-only Liquid Xenon (LoLX) Detector

Silicon Photomultipliers (SiPMs) are a critical technology for the next generation of rare-event search experiments using liquid xenon (LXe). LXe scintillation is emitted primarily in the vacuum-ultraviolet (VUV), requiring photosensors with high VUV detection efficiency under cryogenic operating conditions. Direct characterization of VUV-sensitive SiPMs in LXe is therefore important for informing detector design and comparing device response. This work presents a direct comparison of Fondazione Bruno Kessler (FBK) VUV-HD3 and Hamamatsu (HPK) VUV4 SiPMs operated simultaneously within the Light-only Liquid Xenon (LoLX) detector. Using data collected with gamma sources placed outside the detector, we characterized the relative performance of these photosensors. Our analysis reveals that under these operating conditions, the HPK SiPMs observe 33-38% less light than the FBK devices, a larger difference than predicted by standard PDE models in vacuum measurements. We show that this discrepancy is resolved by an angular and wavelength dependent PDE model incorporating surface shadowing effects into our optical simulation, which then accurately reproduces the experimental data. This finding has significant implications for the selection and implementation of photosensors in future large-scale LXe detectors.

physics.ins-det↗

arXiv2025

Canadian Contributions to the Search for Neutrinoless Double Beta Decay

The search for neutrinoless double beta decay has internationally been recognized as the most promising approach to determine the Majorana nature of neutrinos. This hypothesized decay would, if observed, violate lepton number in weak interactions by two units, hence, prove the existence of physics beyond the Standard Model. Current experiments with sensitivity to neutrinoless double beta decay half lives of $10^{26}$ years did not observe such a decay and worldwide efforts are ongoing to deploy experiments with half-life sensitivities beyond $10^{28}$ years. Canadian groups have been involved in this search for more than four decades. This article summarizes the historical experimental efforts and describes current Canadian contributions to neutrinoless double beta decay searches and their theoretical interpretation.

nucl-ex↗

arXiv2025

Measuring dark currents in multiple cryogenic SiPMs with sub-pA sensitivity using an automated IV multiplexer

We present the design of an automated current-voltage (IV) multiplexer (MUX) that enables accurate measurement of the dark current in cryogenic silicon photomultipliers (SiPMs), achieving a sensitivity equivalent to detecting less than one avalanche per second. Dynamic pulse-counting measurements were used as a benchmark for reconstructing the dark current in static IV measurements. The IV-MUX features 15 channels on a single board and up to seven boards can be connected in parallel under the control of one Arduino microcontroller. To minimize leakage and enhance performance, the layout includes guard rings and high-isolation relays, enabling resolution of currents as small as 49 fA. The IV-MUX can be integrated into systems designed for IV or pulse-counting measurements, enabling seamless switching between IV and pulse-counting modes. Moreover, the IV-MUX is vacuum-compatible, validated by testing an SiPM array in a cryostat. This feature reduces the need for multiple feedthroughs when testing sensor arrays in vacuum. The design is open source and can be used to facilitate rapid and automated testing of SiPMs or similar low-current devices in one measurement cycle.

physics.ins-det↗

arXiv2024

Assessing Students' Understanding of Uncertainty in Undergraduate Physics Laboratory Courses at a Major Canadian University

Over the last five years, the McGill University Office of Science Education (OSE) has partnered with faculty members from the Department of Physics to form an education research group with the aim of charting the progression of student conceptual understanding of uncertainties across their undergraduate degree. The research conducted by this group seeks to provide further insight into the experimental skillset that students gain through undergraduate laboratory courses and noticeable gaps in student understanding that the department could address. In this paper, we evaluate the conceptual understanding of uncertainty using the Concise Data Processing Assessment (CDPA) test. First, we characterize the physics laboratory curriculum at McGill University by evaluating the evolution of CDPA scores across consecutive laboratory courses, and further propose the utilization of this tool for identifying gaps in student understanding. Following the analysis of student responses (N = 2023), we specifically investigate data collected in second-year courses to better understand what student errors reveal about common misconceptions in experimental physics. This more in-depth analysis focuses on data collected from students at the beginning and end of two consecutive experimental laboratory courses that build on each other. By the end of the second course, students have engaged with all the material covered in the CDPA test. Overall results show an upward shift in student understanding of uncertainties over time. Interestingly, we did not observe any change in CDPA scores comparing throughout and post-pandemic scores. Despite the upward shift, many students continue to struggle with uncertainties, basic data analysis, and curve fitting.

physics.ed-ph↗

arXiv2024

Ion manipulation from liquid Xe to vacuum: Ba-tagging for a nEXO upgrade and future $0 \nu \beta \beta$ experiments

Neutrinoless double beta decay {($0\nu\beta\beta$)} provides a way to probe physics beyond the Standard Model of particle physics. The upcoming nEXO experiment will search for $0\nu\beta\beta$ decay in $^{136}$Xe with a projected half-life sensitivity exceeding $10^{28}$ years at the 90\% confidence level using a liquid xenon (LXe) Time Projection Chamber (TPC) filled with 5 tonnes of Xe enriched to $\sim$90\% in the {$\beta \beta$}-decaying isotope $^{136}$Xe. In parallel, a potential future upgrade to nEXO is being investigated with the aim to further suppress radioactive backgrounds and to confirm $\beta \beta$-decay events. This technique, known as Ba-tagging, comprises extracting and identifying the $\beta \beta$-decay daughter $^{136}$Ba ion. One tagging approach being pursued involves extracting a small volume of LXe in the vicinity of a potential $\beta \beta$-decay using a capillary tube and facilitating a liquid-to-gas phase transition by heating the capillary exit. The Ba ion is then separated from the accompanying Xe gas using a radio-frequency (RF) carpet and RF funnel, conclusively identifying the ion as $^{136}$Ba via laser-fluorescence spectroscopy and mass spectrometry. Simultaneously, an accelerator-driven Ba ion source is being developed to validate and optimize this technique. The motivation for the project, the development of the different aspects, along with the current status and results, are discussed here.

physics.ins-det↗

arXiv2023

Simulation Study of Photon-to-Digital Converter (PDC) Timing Specifications for LoLX Experiment

The Light only Liquid Xenon (LoLX) experiment is a prototype detector aimed to study liquid xenon (LXe) light properties and various photodetection technologies. LoLX is also aimed to quantify LXe's time resolution as a potential scintillator for 10~ps time-of-flight (TOF) PET. Another key goal of LoLX is to perform a time-based separation of Cerenkov and scintillation photons for new background rejection methods in LXe experiments. To achieve this separation, LoLX is set to be equipped with photon-to-digital converters (PDCs), a photosensor type that provides a timestamp for each observed photon. To guide the PDC design, we explore requirements for time-based Cerenkov separation. We use a PDC simulator, whose input is the light information from the Geant4-based LoLX simulation model, and evaluate the separation quality against time-to-digital converter (TDC) parameters. Simulation results with TDC parameters offer possible configurations supporting a good separation. Compared with the current filter-based approach, simulations show Cerenkov separation level increases from 54% to 71% when using PDC and time-based separation. With the current photon time profile of LoLX simulation, the results also show 71% separation is achievable with just 4 TDCs per PDC. These simulation results will lead to a specification guide for the PDC as well as expected results to compare against future PDC-based experimental measurements. In the longer term, the overall LoLX results will assist large LXe-based experiments and motivate the assembly of a LXe-based TOF-PET demonstrator system.

physics.ins-det↗

arXiv2022

Evaluating the Adversarial Robustness of Adaptive Test-time Defenses

Adaptive defenses, which optimize at test time, promise to improve adversarial robustness. We categorize such adaptive test-time defenses, explain their potential benefits and drawbacks, and evaluate a representative variety of the latest adaptive defenses for image classification. Unfortunately, none significantly improve upon static defenses when subjected to our careful case study evaluation. Some even weaken the underlying static model while simultaneously increasing inference computation. While these results are disappointing, we still believe that adaptive test-time defenses are a promising avenue of research and, as such, we provide recommendations for their thorough evaluation. We extend the checklist of Carlini et al. (2019) by providing concrete steps specific to adaptive defenses.

cs.LG↗

arXiv2021

Status and Perspectives of Neutrino Physics

This review demonstrates the unique role of the neutrino by discussing in detail the physics of and with neutrinos. We deal with neutrino sources, neutrino oscillations, absolute masses, interactions, the possible existence of sterile neutrinos, and theoretical implications. In addition, synergies of neutrino physics with other research fields are found, and requirements to continue successful neutrino physics in the future, in terms of technological developments and adequate infrastructures, are stressed.

hep-ph↗

arXiv2019

Copy and Paste: A Simple But Effective Initialization Method for Black-Box Adversarial Attacks

Many optimization methods for generating black-box adversarial examples have been proposed, but the aspect of initializing said optimizers has not been considered in much detail. We show that the choice of starting points is indeed crucial, and that the performance of state-of-the-art attacks depends on it. First, we discuss desirable properties of starting points for attacking image classifiers, and how they can be chosen to increase query efficiency. Notably, we find that simply copying small patches from other images is a valid strategy. We then present an evaluation on ImageNet that clearly demonstrates the effectiveness of this method: Our initialization scheme reduces the number of queries required for a state-of-the-art Boundary Attack by 81%, significantly outperforming previous results reported for targeted black-box adversarial examples.

cs.CV↗

arXiv2019

Leveraging Semantic Embeddings for Safety-Critical Applications

Semantic Embeddings are a popular way to represent knowledge in the field of zero-shot learning. We observe their interpretability and discuss their potential utility in a safety-critical context. Concretely, we propose to use them to add introspection and error detection capabilities to neural network classifiers. First, we show how to create embeddings from symbolic domain knowledge. We discuss how to use them for interpreting mispredictions and propose a simple error detection scheme. We then introduce the concept of semantic distance: a real-valued score that measures confidence in the semantic space. We evaluate this score on a traffic sign classifier and find that it achieves near state-of-the-art performance, while being significantly faster to compute than other confidence scores. Our approach requires no changes to the original network and is thus applicable to any task for which domain knowledge is available.

stat.ML↗

arXiv2019

Bridging the Gap between Open Source Software and Vehicle Hardware for Autonomous Driving

Although many research vehicle platforms for autonomous driving have been built in the past, hardware design, source code and lessons learned have not been made available for the next generation of demonstrators. This raises the efforts for the research community to contribute results based on real-world evaluations as engineering knowledge of building and maintaining a research vehicle is lost. In this paper, we deliver an analysis of our approach to transferring an open source driving stack to a research vehicle. We put the hardware and software setup in context to other demonstrators and explain the criteria that led to our chosen hardware and software design. Specifically, we discuss the mapping of the Apollo driving stack to the system layout of our research vehicle, fortuna, including communication with the actuators by a controller running on a real-time hardware platform and the integration of the sensor setup. With our collection of the lessons learned, we encourage a faster setup of such systems by other research groups in the future.

cs.RO↗

arXiv2019

Towards Safety Verification of Direct Perception Neural Networks

We study the problem of safety verification of direct perception neural networks, where camera images are used as inputs to produce high-level features for autonomous vehicles to make control decisions. Formal verification of direct perception neural networks is extremely challenging, as it is difficult to formulate the specification that requires characterizing input as constraints, while the number of neurons in such a network can reach millions. We approach the specification problem by learning an input property characterizer which carefully extends a direct perception neural network at close-to-output layers, and address the scalability problem by a novel assume-guarantee based verification approach. The presented workflow is used to understand a direct perception neural network (developed by Audi) which computes the next waypoint and orientation for autonomous vehicles to follow.

cs.SE↗

arXiv2019

Graph Neural Networks for Modelling Traffic Participant Interaction

By interpreting a traffic scene as a graph of interacting vehicles, we gain a flexible abstract representation which allows us to apply Graph Neural Network (GNN) models for traffic prediction. These naturally take interaction between traffic participants into account while being computationally efficient and providing large model capacity. We evaluate two state-of-the art GNN architectures and introduce several adaptations for our specific scenario. We show that prediction error in scenarios with much interaction decreases by 30% compared to a model that does not take interactions into account. This suggests that interaction is important, and shows that we can model it using graphs. This makes GNNs a worthwhile addition to traffic prediction systems.

cs.LG↗

arXiv2018

Guessing Smart: Biased Sampling for Efficient Black-Box Adversarial Attacks

We consider adversarial examples for image classification in the black-box decision-based setting. Here, an attacker cannot access confidence scores, but only the final label. Most attacks for this scenario are either unreliable or inefficient. Focusing on the latter, we show that a specific class of attacks, Boundary Attacks, can be reinterpreted as a biased sampling framework that gains efficiency from domain knowledge. We identify three such biases, image frequency, regional masks and surrogate gradients, and evaluate their performance against an ImageNet classifier. We show that the combination of these biases outperforms the state of the art by a wide margin. We also showcase an efficient way to attack the Google Cloud Vision API, where we craft convincing perturbations with just a few hundred queries. Finally, the methods we propose have also been found to work very well against strong defenses: Our targeted attack won second place in the NeurIPS 2018 Adversarial Vision Challenge.

stat.ML↗

arXiv2017

Searching for $0\nu\beta\beta$ decay in $^{136}$Xe -- towards the tonne-scale and beyond

The quest for neutrinoless double-beta decay ($0\nu\beta\beta$) is a promising experimental approach to search for lepton number violation in weak interactions, a key ingredient in generating the matter-antimatter asymmetry through models of Leptogenesis. The $^{136}$Xe-based $0\nu\beta\beta$ experiments KamLAND-Zen and EXO-200 currently set the most stringent limits on this process using two very different techniques. Each are preparing the next generation experiment, which will search for $0\nu\beta\beta$ in the parameter space corresponding to the inverted hierarchy for neutrino mass. Both of these techniques scale well to larger volumes while incorporating interesting new techniques. We present the status of current and next generation experiments of these collaborations and present two developments with the potential to identify $\beta\beta$ decay events.

hep-ex↗

arXiv2014

An RF-only ion-funnel for extraction from high-pressure gases

An RF ion-funnel technique has been developed to extract ions from a high-pressure (10 bar) noble-gas environment into vacuum ($10^{-6}$ mbar). Detailed simulations have been performed and a prototype has been developed for the purpose of extracting $^{136}$Ba ions from Xe gas with high efficiency. With this prototype, ions have been extracted for the first time from high-pressure xenon gas and argon gas. Systematic studies have been carried out and compared to the simulations. This demonstration of extraction of ions with mass comparable to that of the gas generating the high-pressure into vacuum has applications to Ba tagging from a Xe-gas time-projection chamber (TPC) for double beta decay as well as to the general problem of recovering trace amounts of an ionized element in a heavy (m$>40$ u) carrier gas.

physics.ins-det↗