Swedish Defence Research Agency

Diethylamine (DEA) is a known precursor to some of the most toxic chemical warfare agents (CWA) as both V-agents and Fourth Generation Agents (FGAs) contain the dialkylamine functionality. DEA is also a readily available commercial substance with extensive use in the chemical industry for legitimate purposes. Because of this potential dual use, it is desirable to develop methods to trace the origin of dialkylamines if they are used for illicit purposes. We herein demonstrate that it is possible to differentiate six commercial batches of DEA using position-specific isotope analysis (PSIA) by both 2H and 13C NMR. Using a high-field NMR spectrometer together with a cryogenic 2H probe, we have produced 2H-{1H} NMR data with high accuracy and precision. The PSIA by NMR results show that the intramolecular 2H ratios of all six DEAs are significantly different while two of the six DEAs have unique 13C ratios. Further, the intramolecular isotopic variations can be used to link the DEAs to different suppliers. The two nuclei separately contribute to isotopic profiles of the DEAs. However, combining the two techniques provides a higher-resolved isotopic profile that can differentiate all DEAs and thus be useful in forensic investigations of illegal use of chemical weapons.

A coordinate independent technique of recursively choosing recalibration of a measurement system is presented. The approach is likelihood based and is formulated in terms of Information Geometry. A Riemannian metric is used for representing the importance of precision in different directions. The technique is illustrated in a localization scenario, where the error model is derived from the von Mises-Fisher family.

An experimental and numerical study has been performed, where initiation of a high explosive target by high velocity debris from a tungsten sphere was investigated. The target consisted of two spaced plates with the high explosive mounted directly behind the inner plate. Both the impact velocity and target obliquity have been varied to find impact conditions that lead to initiation of the high explosive. Numerical simulations were used to estimate critical impact velocities for a series of impact obliquities. The critical impact velocity was found to increase with increasing obliquity, presumably due to the reduction in residual velocity of the debris cloud, and with the reduction of the effective diameter of the debris cloud as the obliquity increases above 30°. • Presents experiments showing the critical impact conditions leading to initiation. • Complements experimental results with numerical simulations. • Suggests an expression relating the critical impact velocity to obliquity.

In radiological emergencies, rapid 90Sr quantification is vital. This study developed a method for 90Sr determination in milk and drinking water, using Sr-specific resin separation followed by ICP-MS/MS detection in MS/MS mode to eliminate 90Zr isobaric interference. Validation demonstrated high recoveries (95–97%) and detection limits (18–22 Bq kg−1) compliant with European regulation Euratom 2016/52. Inter-laboratory comparisons confirmed the method reproducibility and robustness across diverse setups. Contamination control was identified as essential for method success. The method enables rapid 90Sr measurement in milk and water, supporting radiological emergency response.

Aerogel is a name for a type of synthetic porous solid material originating from a gel, with high thermal insulation and ultra-low density. Today’s focus on sustainability has led to studies on renewable sources and production alternatives. With respect to this, aerogels of carboxymethylated nanofibrous cellulose (CMC) were studied, especially the impact of structural differences on the thermal and optical properties. Three types of CMC aerogels were investigated, 1) pure CMC, 2) CMC crosslinked with (3-glycidyoxypropyl) trimethoxysilane (GOPS), and 3) crosslinked CMC with the additive PEDOT:PSS. These were either frozen bidirectionally or without a specified direction of freezing for about 20 hours at low pressure and temperature. The resulting porous aerogels morphology were studied by scanning electron microscopy. Different structures and pore size distributions were obtained for the different types of aerogels. Directional hemispherical reflectance was measured in the visible-infrared (IR) wavelength range using a spectrophotometer and a Fourier Transform Infrared spectrometer. The thermal radiation from the samples were studied using a thermal IR camera, suggesting the thermal insulation level of the aerogels. The thermal and optical properties of the aerogels were assumed to depend more on the addition of GOPS crosslinkers and PEDOT:PSS rather than direction of freezing. This work shows the potential of using aerogels in military applications to reduce the thermal signature.

Terrain classification in unseen off-road terrain is one of the most challenging tasks within traversability analysis for autonomous ground vehicles (UGV). In known terrain, state of the art methods often utilize supervised learning of a neural network (NN) using labelled training data from said terrain to perform semantic segmentation and achieve good terrain classification results. However, annotating data is both expensive and time consuming, and most of these methods struggle with generalisation to unseen terrain. In order to avoid the process of annotating data for each new environment a UGV is intended to operate in, this paper investigates if it is possible to achieve good generalization performance in unseen terrain, without annotating any data, by utilizing public available datasets and combining them for training of a NN. The result is compared to a state-of-the-art method of weakly supervised learning fine-tuned on the unseen terrain. To enable a full comparison, we propose two novel evaluation metrics, Category IoU and Category Accuracy. The datasets that are combined for training of the NN are Rellis-3D, RUGD and YCOR, all collected in the United States. The unseen terrain is represented by the WildScenes dataset, collected in Australia. As well as being collected in a different country, the WildScenes dataset was collected during a period of more than six months at two different locations, ensuring a high degree of diversity compared to the datasets used for training. We show that, on the unseen WildScenes dataset, the performance of a model trained on the combined dataset achieves comparable performance to a weakly supervised model fine-tuned on the WildScenes dataset. Lastly, key aspects of combining off-road datasets with different types of classes for UGVs with different levels of off-road capabilities is discussed.

Abstract The digitalization of our societies makes them increasingly vulnerable to emerging cyberthreats. These cyberthreats can manifest themselves in the form of organized, sophisticated, and persistent threat actors, as well as nonadversarial mistakes. Staff involved in responding to cyberthreats and handling incidents in organizations need cyber situation awareness. This paper presents a case study on what challenges members of staff involved in cybersecurity in a large, complex organization experience when developing cyber situation awareness while handling a remote code execution vulnerability in the form of Log4Shell. Two types of qualitative empirical material were used for the case study, data collected through semi-structured interviews with ten informants, and internal documentation. The empirical material was analyzed to create a timeline of events in the organization. The results show how information about the threat spread throughout the organization, the types of artifacts that served as common operational pictures, and the role played by information sharing in maintaining staff cyber situation awareness. Three major challenges to the organization were found: (i) information sharing among staff was not effortless, (ii) there was no organization-wide common operational picture established, and (iii) inaccurate information was shared. This study adds a real-world contribution to the literature on organizational handling of cyberthreats.

This paper describes a study aimed at developing and applying a method for comparing the user experience (UX) of virtual reality (VR) systems, in order to identify aspects that differentiate VR systems from a user perspective as well as aspects that users consider important when selecting a VR system. A within-subjects study was conducted, in which twelve participants played a VR game using four different VR systems (i.e., VR headsets and hand controllers). The UX was evaluated through semi-structured interviews and questionnaires. Headset comfort, fit, and perceived image quality as well as unrestricted mobility (i.e., untethered headsets) were identified as important aspects for the UX. Hand controllers that allowed more natural interaction were generally preferred. Image sharpness and headset comfort were considered the most important aspects for selecting a VR system. The study demonstrated that the method can be applied and generate insights into the UX of VR systems.