Tech Xplore

A new computational design framework enables the creation of 3D-printed woven metamaterials with customizable stretch and failure properties. Unlike traditional metamaterials focused on stiffness, this approach supports soft, compliant structures suitable for applications in soft robotics, biomedical devices, and wearable technology. The open-source tool allows users to design, simulate, and print complex woven geometries, offering precise control over material behavior. By adjusting geometric parameters, users can tailor deformation and failure characteristics, significantly expanding the design space for functional textiles and flexible devices. This advancement streamlines the development of next-generation metamaterials for diverse engineering challenges.

A new AI tool leverages historical battery data and physics-based features to predict battery cycle life using only a few days of early-stage testing. By analyzing results from just 50 charge-discharge cycles, the tool can estimate how long a battery will maintain its capacity, reducing testing time and energy consumption by up to 95% compared to conventional methods. This approach, inspired by discovery learning, offers a scalable solution for accelerating battery development and may be applicable to other scientific and engineering domains where experimental bottlenecks slow innovation.

A recent study demonstrates a sustainable approach to recycling mattress waste by using fungi to transform shredded mattress foam into lightweight, fire-resistant building insulation. This innovative material remains stable at temperatures close to 1,000°C and offers heat-blocking performance comparable to commercial insulation products. With 1.8 million mattresses disposed of annually in Australia, this method presents a practical solution to reduce landfill waste. The process utilizes fungi related to those used in food and medicine, highlighting the potential for environmentally responsible materials in construction and future manufacturing applications.

Reinforcement learning has significant potential in fields such as autonomous vehicles and complex systems, but traditional methods often struggle to scale due to the complexity of large environments. Recent research introduces a mathematically rigorous and computationally efficient approach that transforms large-scale reinforcement learning problems into a more manageable form. By mapping these problems into a simplified domain, the method enables reliable approximations and efficient learning, even for infinite-dimensional systems. This advancement supports the development of scalable algorithms with strong convergence guarantees, opening new possibilities for applications in technology and medicine.

A new method for guiding light on silicon wafers with ultralow signal loss, comparable to optical fiber at visible wavelengths, has been demonstrated. This advancement enables the fabrication of photonic integrated circuits (PICs) with significantly improved coherence and efficiency. By adapting germano-silicate glass and advanced lithography, the approach achieves record-low loss, particularly in the visible spectrum. This technology has broad implications for precision measurement, data-center communications, and quantum computing, supporting the development of high-performance on-chip devices such as optical clocks, gyroscopes, and atomic sensors.

A new pilot facility in Korea has demonstrated the production of 100 kg of sustainable aviation fuel (SAF) per day using landfill gas from organic waste. This integrated process addresses key challenges in gas purification and conversion efficiency, utilizing advanced catalysts and a microchannel reactor to enhance liquid fuel selectivity and process stability. The approach leverages abundant, low-cost landfill gas, offering an alternative to limited and expensive feedstocks like used cooking oil. This development highlights the potential for decentralized SAF production at local waste sites, supporting carbon neutrality and advancing the circular economy in the aviation sector.

Recent research at Oak Ridge National Laboratory has advanced the understanding of neptunium, a key precursor in the production of plutonium-238 (Pu-238), which is essential for powering space exploration missions. By employing advanced thermal decomposition and characterization techniques, the study identified critical chemical and structural properties of neptunium. These insights are expected to improve the efficiency of Pu-238 production, supporting a secure domestic supply chain for radioisotope thermoelectric generators and enabling continued innovation in deep space exploration.

A new proof-of-concept material demonstrates programmable mechanical properties using Lego-like building blocks, enabling robots to adapt their stiffness and functionality in real time. By controlling the solidity of individual cells within each block, the material can be reconfigured repeatedly, offering significant flexibility. Initial tests showed that robotic fish equipped with this technology could alter their swimming paths without changing motor activity. Potential applications include adaptive medical devices and robotics capable of navigating complex environments. This innovation represents a step toward creating materials that mimic the dynamic adaptability of biological tissues. The research is published in Science Advances.

Recent advancements in metamaterials highlight the importance of unit cell size, number, and arrangement in determining mechanical properties such as elasticity and load sharing. These findings enable more precise predictions of structural behavior, which is particularly relevant for applications like bone implants, robotic grippers, and automotive bumpers. By tailoring the internal structure of metamaterials, it is possible to match the mechanical properties of implants to natural bone, promoting bone strength and longevity. Additionally, understanding how different forces affect metamaterials supports the development of safer, more durable engineered structures across various industries.

Recent advancements in 3D printing technology are enabling the use of recycled plastics to create construction-grade structural elements, such as beams and trusses. These 3D-printed components, produced from recycled PET polymers and glass fibers, have demonstrated the ability to meet and exceed key building standards, supporting loads over 4,000 pounds while remaining lighter than traditional wood-based alternatives. This approach offers a potential pathway to address both the global housing shortage and the environmental impact of single-use plastics, presenting a sustainable and scalable alternative to conventional construction materials.