Hermann Kohlstedt

We investigate charge tunneling in asymmetric junctions comprising a ferroelectric barrier BaTiO3 with a common La0.7Sr0.3MnO3 bottom electrode and various metallic top electrodes. Quantitative analysis of tunneling electroresistance within a direct tunneling framework reveals that the net change in average barrier potential upon polarization reversal scales with the depolarizing electrostatic field, governed by screening asymmetry at the ferroelectric–electrode interfaces. The observed electrode-dependent tunneling electroresistance is thus primarily controlled by the potential barrier height at the top interface. Our results, supported by theoretical considerations and experimentally determined lower bounds to the depolarization field, emphasize the key role of interface electrostatics in ferroelectric tunneling transport.

Haloclines – sharp salinity gradients frequently formed after heavy rainfalls in coastal habitats – can act as barriers for weakly swimming plankton, but their biomechanical constraints on relatively adept swimmers, such as cubozoan jellyfish, remain unexplored. We examined the vertical distribution of Tripedalia cystophora before and after establishing an artificial halocline (35→22 PSU) in an experimental swimming arena. After halocline formation, animals repeatedly entered the gradient layer but did not ascend beyond its upper boundary, despite repeated upward trajectories towards the gradient layer, indicating no obvious avoidance response. A hydrodynamic model supported these observations, demonstrating that stratification drag markedly increases thrust dissipation. Thus, centimetre-scale haloclines impose physical constraints that prevent T. cystophora from accessing surface waters through reduced effective upward momentum, rather than through behavioural change. Because the underlying hydrodynamic principles are general, we propose a stratification hypothesis to explain how sharp density gradients shape the vertical distribution of some aquatic animals.

The emerging nitride ferroelectrics, such as Al _1‐x Sc _x N promise to significantly advance our current information technology. In particular, two‐terminal memristive devices are ideal candidates for artificial intelligence accelerators and in‐memory computing due to their simplicity in design, non‐volatility and non‐destructive readout. The recent discovery of conductive domain walls in Al _1‐x Sc _x N is a promising enabler for such technology, offering several benefits compared to barrier height modulation‐ or tunneling‐based devices. First, domain walls can be highly conductive and feature high read currents (required for aggressive lateral scaling and fast access times), also in non‐epitaxial films without being restricted to the technologically challenging ultrathin thickness regime (10 nm). Second, nitride ferroelectrics are fully compatible with silicon and GaN technology on which the ferroelectric domain wall memory (FeDMEM) can be integrated with logic circuitry. Third, excellent scalability and temperature resistance of ferroelectric Al _1‐x Sc _x N were demonstrated, enabling scaled, low‐latency edge computing under extreme environmental conditions. In this study, a FeDMEM device consisting of a Pt/Al _0.72 Sc _0.28 N/Pt capacitor grown on Si substrates is electrically characterized in‐depth, revealing unique peculiarities in the memristive response. A read current density of 350 A/m ² and an ON/OFF ratio of 20 is achieved, allowing for consistent storing of up to eight levels of information.

This study examines the development of TiN/SiO_x/Cu/SiO_x/TiN memristive devices for neuromorphic applications using wedge‐type deposition and Monte Carlo simulations. Identifying critical parameters for the desired device characteristics can be challenging with conventional trial‐and‐error methods, which often obscure the effects of varying layer compositions. By employing an off‐center thermal evaporation method, a thickness gradient of SiO_x and Cu on a 4 inch wafer is created, facilitating detailed resistance map analysis through semiautomatic measurements. This approach allows for investigating the influence of layer composition and thickness while keeping other process conditions constant. Combining experimental data with simulations provides a precise understanding of layer thickness distribution and its impact on device performance. Optimizing the SiO_x layers to be below 12 nm, coupled with a discontinuous Cu layer with a nominal thickness under 0.6 nm, exhibits analog switching properties with an R_on/R_off ratio of >100, suitable for neuromorphic applications, while R × A and power exponent γ analysis show signs of multiple conduction mechanisms. The findings highlight the importance of SiO_x and Cu thickness in determining switching behavior, offering insights for developing high‐performance analog switching components for bioinspired computing systems.

This work presents the development and implementation of an interdisciplinary and intermedial science outreach programme designed for school students. The programme integrates biological systems and technological advancements to provide students with hands-on laboratory experiences and immersive media, including virtual reality videos and augmented reality posters. Through a co-creation process involving scientists and educators, the programme aims to enhance students understanding of bio-inspired information pathways and neurogenesis. Preliminary evaluations indicate high engagement and educational value, suggesting that such interdisciplinary approaches can significantly enrich science education. Graphical

As a result of a hundred million years of evolution, living animals have adapted extremely well to their ecological niche. Such adaptation implies species-specific interactions with their immediate environment by processing sensory cues and responding with appropriate behavior. Understanding how living creatures perform pattern recognition and cognitive tasks is of particular importance for computing architectures: by studying these information pathways refined over eons of evolution, researchers may be able to streamline the process of developing more highly advanced, energy efficient autonomous systems. With the advent of novel electronic and ionic components along with a deeper understanding of information pathways in living species, a plethora of opportunities to develop completely novel information processing avenues are within reach. Basal biological principles are highlighted, including phylogenies, ontogenesis, and homeostasis, with particular emphasis on network topology and dynamics. While in machine learning, system training is performed on virgin networks without any a priori knowledge, the approach proposed here distinguishes itself unambiguously by employing growth mechanisms as a guideline to design novel computing architectures. Within this framework, experiments on low-frequency relaxation type oscillators coupled via complex time-varying networks will be presents. The spatio-temporal development of the network results from a mutual interaction between the oscillator ensemble and the network structure. The blooming and pruning of suddenly appearing conductive bridges and their relation to the synchrony state of the oscillator ensemble will be discussed.

Wurtzite‐type Al_1−xSc_xN solid solutions grown by metal organic chemical vapor deposition are for the first time confirmed to be ferroelectric. The film with 230 nm thickness and x = 0.15 exhibits a coercive field of 5.5 MV cm⁻¹ at a measurement frequency of 1.5 kHz. The single crystal quality and homogeneous chemical composition of the film are confirmed by X‐ray diffraction and spectroscopic methods such as time of flight secondary ion mass spectrometry. Annular bright field scanning transmission electron microscopy serves to prove the ferroelectric polarization inversion at the unit cell level. The single crystal quality further allows to image the large‐scale domain pattern of a wurtzite‐type ferroelectric for the first time, revealing a predominantly cone‐like domain shape along the c‐axis of the material. As in previous work, this again implies the presence of strong polarization discontinuities along this crystallographic axis, which can be suitable for current transport. The domains are separated by narrow domain walls, for which an upper thickness limit of 3 nm is deduced but which can potentially be atomically sharp. The authors are confident that these results will advance the commencement of the integration of wurtzite‐type ferroelectrics to GaN as well as generally III‐N‐based heterostructures and devices.

Reservoir computing is an efficient and flexible framework for decision-making, control, and signal processing. It uses a network of interacting components varying from abstract nonlinear dynamical systems to physical substrates. Despite recent progress, the hardware implementation with inherent parameter variability and uncertainties, such as those mimicking the properties of living organisms’ nervous systems, remains an active research area. To address these challenges, we propose a constructive approach using a network of FitzHugh-Nagumo oscillators, exhibiting criticality across a broad range of resistive coupling strengths and robustness without specific parameter tuning. Additionally, the network’s activity demonstrates spatial invariance, offering freedom in choosing readout nodes. We introduce an alternative characterization of criticality by analyzing power dissipation, and demonstrate that criticality supports the robustness of the classification accuracy with respect to the readout shrinkage. Our results indicate criticality as a valuable property for classification problems, and provides design concepts for bio-inspired computational paradigms.

The emergence of digital computers has profoundly reshaped our interactions with technology and the processing of information. Despite excelling in data processing and arithmetics, these computers face limitations in tackling complex nondeterministic-polynomial (NP) problems. In response, researchers have started searching for new computational paradigms that possess the natural tendency of solving these problems. Oscillator-based optimizers are one such paradigm, where the idea is to exploit the parallelism of oscillators networks in order to efficiently solve NP problems. This involves a process of mapping a given optimization task to a quadratic unconstrained binary optimization program and then mapping the resulting program onto an inter-oscillator coupling circuit encoding its coefficients. This paper presents a comprehensive approach to constructing oscillator-based optimizers, offering both the rationale for employing oscillator networks and formulas for linking optimization coefficients to inter-oscillator coupling. Here, we cover most aspects of oscillator-based optimization starting from the design of the network up to its technical implementation. Moreover, we provide a platform-independent wave digital algorithm, which allows for emulating our network’s behavior in a highly parallel fashion. Graphical