Neural network enabled nanoplasmonic hydrogen sensors with 100 ppm limit of detection in humid air

Environmental humidity variations are ubiquitous and high humidity characterizes fuel cell and electrolyzer operation conditions. Since hydrogen-air mixtures are highly flammable, humidity tolerant H2 sensors are important from safety and process monitoring perspectives. Here, we report an optical nanoplasmonic hydrogen sensor operated at elevated temperature that combined with Deep Dense Neural Network or Transformer data treatment involving the entire spectral response of the sensor enables a 100 ppm H2 limit of detection in synthetic air at 80% relative humidity. This significantly exceeds the <1000 ppm US Department of Energy performance target. Furthermore, the sensors pass the ISO 26142:2010 stability requirement in 80% relative humidity in air down to 0.06% H2 and show no signs of performance loss after 140 h continuous operation. Our results thus demonstrate the potential of plasmonic hydrogen sensors for use in high humidity and how neural-network-based data treatment can significantly boost their performance.

The Swedish National Facility for Magnetoencephalography Parkinson's disease dataset

Parkinson's disease (PD) is characterised by a loss of dopamine and dopaminergic cells. The consequences hereof are widespread network disturbances in brain function. It is an ongoing topic of investigation how the disease-related changes in brain function manifest in PD relate to clinical symptoms. We present The Swedish National Facility for Magnetoencephalography Parkinson's Disease Dataset (NatMEG-PD) as an Open Science contribution to identify the functional neural signatures of Parkinson's disease and contribute to diagnosis and treatment. The dataset contains whole-head magnetoencephalographic (MEG) recordings from 66 well-characterised PD patients on their regular dose of dopamine replacement therapy and 68 age- and sex-matched healthy controls. NatMEG-PD contains three-minute eyes-closed resting-state MEG, MEG during an active movement task, and MEG during passive movements. The data includes anonymised MRI for source analysis and clinical scores. MEG data is rich in nature and can be used to explore numerous functional features. By sharing these data, we hope other researchers will contribute to advancing our understanding of the relationship between brain activity and disease state or symptoms.

Optical analysis of light-emitting electrochemical cells

The light-emitting electrochemical cell (LEC) is a contender for emerging applications of light, primarily because it offers low-cost solution fabrication of easily functionalized device architectures. The attractive properties originate in the in-situ formation of electrochemically doped transport regions that enclose an emissive intrinsic region, but the understanding of how this intricate doping structure affects the optical performance of the LEC is largely lacking. We combine angle- and doping-dependent measurements and simulations, and demonstrate that the emission zone in our high-performance LEC is centered at ~30% of the active-layer thickness (dal) from the anode. We further find that the emission intensity and efficiency are undulating with dal, and establish that the first emission maximum at dal ~ 100 nm is largely limited by the lossy coupling of excitons to the doping regions, whereas the most prominent loss channel at the second maximum at dal ~ 300 nm is wave-guided modes.

Multidimensional diffusion MRI with spectrally modulated gradients reveals unprecedented microstructural detail

Characterization of porous media is essential in a wide range of biomedical and industrial applications. Microstructural features can be probed non-invasively by diffusion magnetic resonance imaging (dMRI). However, diffusion encoding in conventional dMRI may yield similar signatures for very different microstructures, which represents a significant limitation for disentangling individual microstructural features in heterogeneous materials. To solve this problem, we propose an augmented multidimensional diffusion encoding (MDE) framework, which unlocks a novel encoding dimension to assess time-dependent diffusion specific to structures with different microscopic anisotropies. Our approach relies on spectral analysis of complex but experimentally efficient MDE waveforms. Two independent contrasts to differentiate features such as cell shape and size can be generated directly by signal subtraction from only three types of measurements. Analytical calculations and simulations support our experimental observations. Proof-of-concept experiments were applied on samples with known and distinctly different microstructures. We further demonstrate substantially different contrasts in different tissue types of a post mortem brain. Our simultaneous assessment of restriction size and shape may be instrumental in studies of a wide range of porous materials, enable new insights into the microstructure of biological tissues or be of great value in diagnostics.

On the Z1-dependence of electronic stopping in TiN

We present a thorough experimental study of electronic stopping of H, He, B, N, Ne and Al ions in TiN with the aim to learn about the energy loss mechanisms of slow ions. The energy loss was measured by means of time-of-flight medium-energy ion scattering. Thin films of TiN on silicon with a δ-layer of W at the TiN/Si interface were used as targets. We compare our results to non-linear density functional theory calculations, examining electron-hole pair excitations by screened ions in a free electron gas in the static limit, with a density equivalent to the expected value for TiN. These calculations predict oscillations in the electronic stopping power for increasing atomic number Z1 of the projectile. An increasing discrepancy between our experimental results and predictions by theory for increasing Z1 was observed. This observation can be attributed to contributions from energy loss channels different from electron-hole pair excitation in binary Coulomb collisions.

A novel ulvan lyase family with broad-spectrum activity from the ulvan utilisation loci of Formosa agariphila KMM 3901

Ulvan, which is one of the major structural polysaccharides of the cell walls of green macroalgae, is degraded by ulvan lyases via the β-elimination mechanism with the release of oligosaccharides that have unsaturated 4-deoxy-L-threo-hex-4-enopyranosiduronic acid (∆) at the non-reducing end. These ulvan lyases belong to the PL24 or PL25 or PL28 family in the CAZy database. In this study, we identify and biochemically characterise a periplasmic novel broad-spectrum ulvan lyase from Formosa agariphila KMM 3901. The lyase was overexpressed in Escherichia coli, and the purified recombinant enzyme depolymerised ulvan in an endolytic manner with a Km of 0.77 mg/ml, and displayed optimum activity at 40 °C and pH 8. This lyase also degraded heparan sulphate and chondroitin sulphate. Detailed analyses of the end-products of the enzymatic degradation of ulvan using 1H- and 13C-NMR and LC-MS revealed an unsaturated disaccharide (∆Rha3S) and a tetrasaccharide (∆Rha3S-Xyl-Rha) as the principal end-products. In contrast to the previously described ulvan lyases, this novel lyase is mostly composed of α-helices that form an (α/α)6 incomplete toroid domain and displays a remarkably broad-spectrum activity. This novel lyase is the first member of a new family of ulvan lyases.