Presenting the 2026 R&D 100 Awards Winners

Anaerobic Microbial Phenotyping Platform (AMP2), developed by PNNL with Ginkgo, is a first-of-its-kind automated platform that dramatically accelerates study of anaerobic microbes, essential to renewable fuels and industrial bioprocesses. AMP2 transforms a slow, manual process into a scalable, data-driven capability focused on advancing Earth modeling and the U.S. bioeconomy.

CybeRay is an intelligent, non-invasive radiosurgery platform that integrates real-time intrafraction image guidance, dynamic non-coplanar arc delivery, and intuitive workflow to deliver sub-millimeter precision ablative radiation to lesions anywhere in the body, offering patients a more precise, safer, and more accessible therapeutic option.

This multi-modal sensing technique determines the percentage of recycled content in plastic products by integrating various sensing modalities with AI/ML to quantify the percentage of recycled content in plastics. This method addresses a needed gap and offers a solution for regulatory compliance for recycled plastics.

“Globe Scan” is a non-contact laser probe that captures both sidewall and depth profiles of bores from φ6 to 100 mm, with ±5 µm accuracy. A single probe covers the full range and maintains performance under fluctuating factory temperatures (10–35 °C), enabling reliable high-speed in-line inspection at 10,000 points/s.

InFocus κ FDTR integrates diffraction-limited optical microscopy with frequency-domain thermoreflectance to enable submicron laser heating and microscale mapping of thermal properties, while retaining sensitivity to nanoscale-thick films and deep-interface thermal boundary conductance. Localized heating induces three-dimensional heat flow, enabling measurement of in-plane thermal conductivity, enhanced by aberration-corrected, normal-incidence beam scanning.

Tibial stress fractures are among the most common overuse injuries in athletes and military personnel, yet are difficult to diagnose. Researchers at the Johns Hopkins Applied Physics Laboratory (APL) have developed an ultrasound-based platform, OsteoSEE, that uses surface acoustic waves to detect early-stage stress fractures noninvasively and without ionizing radiation.

Photonic BMS ensures energy storage safety and enables rapid load management for AI data centers. Functioning as a “photonic nervous system” for any utility battery stack, it uses multi-photonic sensor nodes and State-of-Safety (SOS) algorithms to fuse multi-modal data, enabling preemptive diagnostics that surpass contemporary Battery Management Systems (BMS).

RADXtract is the most advanced failure analysis technology for next-generation semiconductors in harsh radiation environments offering a neutron irradiation test workflow, integrated particle simulations and electrical device modeling, and AI-driven predictive failure modeling to deliver spatially and temporally resolved failure data, thereby validating device reliability across multiple levels of detail.

The SPEAR UltraDetect BD-pTau 217 Assay enables highly sensitive and precise quantification of brain-derived pTau 217 in plasma, delivering 100% quantifiability across healthy and Alzheimer’s Disease samples using just 1 µL of diluted plasma, enabling robust early detection, stratification, and longitudinal analysis of this ultra-low abundance biomarker.

tSAGE is a genetic engineering method to rapidly accelerate insertion of DNA into thermophilic bacteria. It replaces a cumbersome 4 week, 10-step manual process with a single-step, automatable workflow, reducing engineering time by 90% for heat-loving microbes and unlocking high-throughput approaches to R&D for fuels and bioproducts production.

The Waters Xevo CDMS, the first market-ready CDMS, enables direct, accurate mass measurement of mega-mass biomolecules (up to 150+ MDa), including viral vectors, mRNA, and protein complexes. Powered by innovative ELIT technology, it delivers results in under 10 minutes, distinguishes capsid populations, and reduces sample needs by up to 100-fold.

Architector is an unmatched software for automatically generating 3D metal-containing molecules from simple chemical blueprints across the entire periodic table. The software enables high-throughput screening; conformational sampling of atomic arrangements around metal centers, including first and second layers; and unprecedented speed and chemical accuracy through an accessible and intuitive interface.

This new transmission planning solution enables fast, high-fidelity reliability and economic assessments for new generation and load interconnections. It combines physics-based modeling with advanced computational techniques to accelerate grid simulations, delivering accurate, scalable insights for planning and operational decision-making. *Hitachi Energy’s Optimization product and GLOW refer to the same product.

ICONS novel neutron detection technology accurately measures neutrons in both low- and high-flux environments, even in the presence of strong gamma radiation. ICONS responds rapidly, enabling real-time measurements of short neutron bursts or high-rate neutron sources. ICONS overcomes supply-chain constraints using naturally abundant, commercially available lithium-6 rather than scarce helium-3.

Networked microgrids are clusters of colocated, electrically connected microgrids that improve reliability, resilience, and economics by enabling resource sharing under coordinated control. The Microgrid Orchestrator delivers these benefits by applying decentralized optimization to coordinate power and energy transfers among microgrids, supported by protocols for stability, synchronization, and fault handling.

This 22-kW polyphase wireless charging system delivers twice the power of the SAE J2954 WPT3 reference design in a 61% smaller footprint; achieves 2.4× greater mutual inductance; reduces ferrite, copper, and aluminum usage by >55%; lowers field emissions by 50%; and reaches ~95% efficiency at ~200 mm ground clearance.

PaSTA is a foldable deployable truss structure that increases the structural stiffness of satellite solar arrays & booms. It replaces simple hinges with a preloaded truss, boosting the first natural frequency from below 0.3Hz to above 1.5Hz. A built-in synchronization system enables passive, kinematically determinate deployment controlled by a single damper.

Persistent DynAMICS is an intelligent sensing architecture that safeguards high-value assets by balancing cost and continuity inherent in large-scale monitoring. Rather than continuous, exhaustive data collection, Persistent DynAMICS adds a layer of decision-making on top of existing sensing systems delivering real-time activity tracking, improving detection sensitivity, and reducing false alarms.

APL engineers, with support from NASA’s SCaN program, developed a first-of-its-kind wideband polylingual terminal that enables satellites to communicate across diverse commercial and government networks and frequencies, supporting missions from low Earth orbit to cislunar space.

This radioactive airborne particulate monitor rapidly and accurately distinguishes and quantifies both naturally occurring and artificial radionuclides. It helps prevent unintended internal exposure among nuclear and resource industry workers. Compared to conventional methods, it improves the detection limit by over 100 times and reduces measurement time to less than one-sixth.

Smart Power Hub (SPH) is a compact three-port microgrid converter that integrates renewables, batteries and AC loads in a single unit. Field-tested systems demonstrate 47-56% lower grid-import and conversion losses, about 80% smaller system size, and kHz-order response using built-in energy management.

UNR (Ubiquitous Network-Native Radar Platform) enables Pre-6G sensing using existing 5G and Wi-Fi networks, transforming communication infrastructure into intelligent sensing systems through software-defined and AI technologies. By eliminating dedicated sensing hardware and additional site deployment, UNR reduces deployment cost and complexity while supporting scalable sensing applications across diverse industrial environments.

A team of scientists and engineers at ORNL, in partnership with Samsung Electronics America, developed an innovative dishwasher that uses thermoelectric modules to recover waste heat and enable efficient heat pumping. It reduces energy use by up to 30% and improves drying performance by 30–40% through simultaneous dehumidification and reheating.

A low-cost composite material (63% less expensive than commercially available medium-density fiberboard) that converts wastewater algae and agricultural residues into high-performance structural products. A novel bio-based binding mechanism enhances interfacial adhesion, yielding superior strength compared with conventional wood- and petroleum-based composites while reducing production time from days to seconds.

Purdue University’s Guinness World Record-setting CryoCharge-100 lithium-ion battery platform combines a pseudocapacitive niobium–tungsten oxide electrode with a low-freezing cyclopentyl methyl ether electrolyte for heater-free operation from −100 °C to +80 °C. Integrated with photovoltaics, it enables resilient terrestrial and deep-space energy storage for polar, aerospace, defense, and grid applications.

DuPont Liveo Pharma TPE Ultra-Low Temperature Tubing is a novel single-use bioprocessing tubing engineered to maintain flexibility, integrity, and purity at temperatures down to −86 °C, enabling reliable cold-chain handling of sensitive biologics while preserving weldability, pumpability, and low extractables performance.

Waterless dyeing is vital amid water scarcity and microplastic regulations. While supercritical CO₂ dyeing works for synthetic fibers, natural fibers fail due to incompatibility. DyeMate SC overcomes this barrier, enabling supercritical CO₂ dyeing of natural fibers with performance equal to conventional water dyeing, saving resources and greatly reducing environmental impact.

EcoFlux Anode redefines spherical natural graphite with hard carbon infilling and a fast-ion coating via an eco-friendly aqueous process. This innovation delivers <5% anode swelling, <20-minute charging capability, enhanced reversible capacity and nearly double cycle life, offering a cost-effective, high-performance solution for long-life and high-energy battery applications.

ENDURING is a low-cost, long-duration thermal energy storage technology that uses heated sand to provide hours to days of energy supply for power or heating, reducing energy costs and improving electric grid resilience. ENDURING surpasses competitors thanks to its versatile design, cost-effective scaling, and innovative system configuration.

These novel Cermets are high-density, extremely tough ceramic-metal composites that offer reduced storage footprints, operational flexibility, less complexity, and cost savings over current technologies used for nuclear waste immobilization, management and storage. Cermets and related manufacturing process are key innovations that will accelerate deployment of next-generation nuclear power plants.

The Embedded-Die PCB SiC Traction Inverter converts 800 V DC power into AC for EV motors. Using six embedded 1-kV SiC MOSFET dies with 1-nH commutation-loop inductance, it integrates PCB busbar, current/junction-temperature sensing, DC-link capacitors, and cooling to deliver 173 kW at 97% efficiency and 300-kW/L-class density.

Mesoporous optically clear heat insulators (MOCHIs) act as “frozen air,” delivering >99.5% visible-light transmission (outperforming 92% of glass) and ~10 mWm⁻¹K⁻¹ thermal conductivity (outperforming Mars Rover aerogels). Square-meter-large and inch-thick MOCHIs enable wall-grade insulated windows and skylights, glazing retrofits, unconcentrated solar thermal energy harvesting, and condensation-free super-insulated vehicle windshields.

First-of-its-kind laser-weldable NORYL resin engineered for the solar energy market uniquely combines robust thin-wall flame retardancy, improved weatherability, and superior sealing performance. It enables faster and cleaner manufacturing of PV microinverters while significantly enhancing their reliability and outdoor stability to accelerate global solar adoption and contribute to sustainable development.

The N₂O CAT system is an integrated abatement solution that sequentially combines thermal oxidation, high temperature filtration, and low temperature catalytic decomposition (≤500 °C) to remove nitrous oxide and flammable gases with >99% efficiency, while minimizing secondary pollutants, energy consumption, and retrofit complexity in semiconductor exhaust treatment.

The perforated bipolar membrane electrochemical conversion system efficiently turns waste CO2 into high-value chemicals using electricity. It’s the only electrochemical CO2 conversion technology to achieve commercial viability. Industries can use it today to significantly cut energy consumption and costs and turn waste into profits by tapping into multibillion-dollar chemical markets.

Pintura Autoenfriante is a water-based paint formulated with nanotechnology, capable of reflecting most solar radiation and efficiently emitting thermal energy toward the cold of space (high SRI). As a result, it can reduce indoor temperatures by up to 5°C compared to ambient conditions, lowering energy consumption and reducing carbon emissions.

REFLEX – Real-time Embedded Feedback and Layered Excitation – transforms passive electrochemical interconnects into intelligent, sensor-enabled components that provide real-time monitoring, active thermal control, and digital-twin capabilities for next-generation fuel cells, electrolyzers, and hydrogen energy systems.

GM’s UniSteel, which can be produced from 100% scrap using Electric Arc Furnace, combined with chain-die forming, enables the world’s first cold-formed 1.9 GPa door beam with varying cross section to offer mass efficiency at lower cost and reduced carbon footprint compared to a hot-stamped boron steel door beam.

The Argonne Rare Earth Magnet Electrorefining Process recovers rare earth metal directly from scrap NdFeB magnets and magnet swarf (manufacturing shavings) using molten salt electrorefining. This single-step process bypasses hydrometallurgical separation and conventional electrowinning, enabling a lower-carbon, domestic pathway to recycle critical rare earth materials for new magnet production.

Automated Craftsmanship is a breakthrough robotic system that transforms metallic vehicle panel customization from hours of costly manual work to minutes of automated processing. Using a 5-DOF power hammer, flexible robots, adaptive tooling, and AI-assisted control, it delivers precise, head-turning yet affordable designs, first piloted on 2025 Cadillac V-Series Blackwing editions.

Automated Single-station Manufacturing Process (ASMP) revolutionizes the footwear industry via 3D spinning, 6-axis robotics, and monomaterial design. Replacing intricate 8-step 3D knitting processes, it simplifies production into two stages, seamless upper formation and thermal fusion, reducing cost and impact while forming a closure-free upper with lightweight, adaptive fit.

AIMSS is the first autonomous, space-qualified mass spectrometer operational in orbit, delivering real-time spacecraft contamination monitoring and space plasma environment characterization. Its high mass resolution capability detects propellant leaks, thruster impingement, and out-gassing while validating space weather models, simultaneously protecting missions and advancing space situational awareness.

A production-ready One-Step Separation and Decolorization process that converts contaminated footwear waste into ≥99.5% pure recycled polyester (rPET, ISO 1833). Commercially deployed through footwear-manufacturing supply chains, the technology overcomes the PU/elastane contamination bottleneck and enables a circular-footwear architecture achieving 74.6% carbon reduction (ISO 14067-based assessment).

CarbonTrust is a blockchain-based dMRV platform by ASTRI that automates carbon tracing and verification. It features AI-powered fraud prevention to detect data anomalies and LLM-driven modules to prevent double-counting. The platform introduces new methodologies for urban and technology-driven mitigation projects in Asia.

Argonne’s cathode upcycling process upgrades directly recycled, low-nickel cathode powder into higher-energy materials that can deliver performance levels comparable to modern commercial cathodes. This scalable, manufacturing-compatible approach increases the value of recycled battery materials and gives battery makers flexibility to meet evolving market demands.

As the world races toward net-zero, CO2 is no longer waste but a resource. Supercritical CO2 can be used as a medium like water, but without producing waste. Combining CO2 with synthetic molecules produces a next-generation blood filter that does not use organic solvents and water.

Deimos is a reconfigurable system for evaluating novel nuclear reactor designs. Its modular design enables rapid testing of new reactor concepts and fuel materials under safe operational conditions. As the first modern proving ground for new US reactor designs, Deimos propels the nuclear energy renaissance forward.

The electrified plastic pyrolysis reactor (EPPR) converts hard-to-recycle, abundant polyolefin plastics into jet fuel and diesel-range chemicals with record-high yields in a continuous, scalable process. Operating without catalysts, it delivers exceptional durability, high throughput, lower CAPEX and OPEX, and low-emission fuel production from waste plastics for oil and gas markets.

Electro-Lith is a continuous electrochemical processing method for the precise pre-lithiation of silicon-based anode material particles, which then act as a drop-in option for later battery electrode manufacturing. Electro-Lith could revolutionize lithium-ion batteries by enabling safer, low-cost, low-carbon production of tailored electrode materials.

Electrochemical CO2 graphitization (ECOG) directly converts CO2 into high-value graphitic carbon using molten-salt electrochemistry at record-low temperatures (~550°C). The single-step process produces high-purity crystalline graphite, a critical material for energy storage, at low cost, offering a scalable pathway to transform CO2 into valuable resources for advanced energy technologies.

Flow-Ore is a transformative technology that operates by injecting precisely engineered fluids into low-grade ores and depleted oil and gas reservoirs to dissolve, mobilize, and recover critical minerals and materials (CMMs). Applied at scale, Flow-Ore will allow the United States to domestically meet and exceed the national demand for CMMs.

FlowVAM achieves high-production levels for tomographic Volumetric Additive Manufacturing (tVAM) by incorporating print material flow and eliminating time-consuming, manual replacement of material vials after each print. FlowVAM manifests the speed, high-resolution, material flexibility, and repeatability of tVAM to manufacture parts for optics, MEMS devices, tissue engineering, and other high-value industries.

FReTC is a revolutionary Rutherford-type HTS cable for high-current, high-field applications. A full-size FReTC, like a 60 kA cable, can be fabricated in a single process from tape-spools, ensuring high efficiency. Its robust flat-core architecture withstands massive electromagnetic-forces in fusion-reactors and accelerators, bridging the gap between laboratory and industrial scalability.

Grayscale Lithography via Annealed Resin Engineering (GLARE) is a method for producing the world’s smallest, continuously varying grayscale images on transparent substrates for use as photolithography masks in standard semiconductor processing.

High Velocity (HiVe) Joining is a breakthrough manufacturing solution that enables a cost-effective & reliable method to join difficult materials without welding, or fasteners. The high-speed impact process creates a metallurgical bond without melting, enabling clean, corrosion-resistant joints across advanced high-strength steel and lightweight multi-material designs.

IonTwin is a physics-informed AI digital twin platform for ion implantation that enables production-speed process prediction and optimization (<10 seconds). The system replaces Monte Carlo–centric workflows with a model-based approach that reduces reliance on repeated test wafers and measurements while supporting rapid, physics-consistent process evaluation for advanced semiconductor manufacturing.

A readily available molten salt system converts waste polyethylene and other polymers into gasoline- and diesel-like fuels under mild conditions. Using inexpensive, recyclable NaCl–AlCl3 salts as both solvent and catalyst eliminates the need for noble metals, organic solvents, or hydrogen, enabling scalable, high-value plastic upcycling from abundant waste streams.

MExS enables simultaneous multimaterial extrusion within a single printed bead during large-scale additive manufacturing. By synchronizing multiple pellet-fed extruders through a proprietary multiplexing nozzle architecture and control algorithms, MExS increases manufacturing throughput by up to 2x while enabling architected hybrid composite structures with tailored mechanical and multifunctional properties.

NARI’s technology utilizes recycled aluminum to create eco-friendly adsorption drying systems. Featuring system integration and low-temperature regeneration, these systems offer precise control and superior energy efficiency. This innovation aligns with UN SDG 7, 9, and 12, providing a sustainable, energy-efficient solution for global industrial dehumidification.

NEAREST is an alternative rare earth separation process for critical materials essential to the economy, national security, and clean energy. By using two neutral organic compounds with reverse lanthanide selectivity, it improves adjacent-ion separation and reduces the number of processing steps needed to isolate individual rare earth elements by 50%.

Obsidian is Path Robotics’ physical AI model for welding, transforming traditional robots from rigid, repeat-only machines into real-time perception and decision-making systems that can see, understand, and adapt in real time as a human would.

This process was developed to recover rare earths, phosphoric acid, uranium, and gypsum from phosphate byproducts. Utilizing a two-step modular membrane solvent-extraction system, coupled with a continuous-flow centrifugal decanter and controlled acid leaching, this intensified process transforms industrial waste streams into a profitable domestic source of critical minerals and materials.

Programmable tri-structural isotropic (TRISO) nuclear fuel enables the economic use of inherently safe TRISO fuel by replacing random particle packing with engineered architectures. Using additive manufacturing and automated particle placement, it increases fuel density, improves performance, and enhances safety, unlocking scalable, high-performance fuel for inherently safe nuclear reactors.

PROTECT is a bonded optical capping technology that prevents contamination-driven damage in high-power laser optics. By adding a transparent protective barrier, it blocks heat transfer from surface particles, increasing damage resistance by up to 20× and significantly improving reliability, reducing downtime, and improving performance across industrial, semiconductor, and defense systems.

The Pulsar-Based Timing System converts naturally occurring pulsar signals into a high-integrity, UTC-traceable timing source for critical infrastructure. By providing a resilient alternative to GNSS-based timing, it supports synchronization-sensitive applications with demonstrated holdover capability, distributed timing delivery, and compatibility with timing critical networks.

Quantum Microsystems combines AI decision-tree-optimized fabrics, a software-defined digital lithographic layer, superconducting flex interconnects, and reworkable packaging architectures to enable a fully functional wafer-scale computing platform. Ultradense, near-lossless interconnects integrate hundreds of heterogeneous chiplets to deliver speed-of-light, low-latency, ultrahigh-bandwidth chiplet-to-chiplet communication for scalable quantum computing, AI, and high-performance computing.

In collaboration with the Manufacturing Demonstration Facility and the Building Technologies Office, Oak Ridge National Laboratory has developed a high-strength and high-thermal-stability silicone composite by using fiberglass and recycled fiberglass and deployed this material in a full-size prototype residential oven developed with the Whirlpool Corporation (CRADA).

RoboVision Caliper is an AI-powered generalizable 3D vision system for quality inspection. Deployed on Cadillac Escalade EV doors at GM Factory Zero assembly plant, it inspects six weld-defect and two non-weld-defect types at tens-of-micron accuracy via multi-modal AI perception models, cutting production cycle time by 45%.

SATURN is a patented multispectral, high-speed sensor system that monitors thermal dynamics during metal additive manufacturing, enabling real-time process insight, defect detection, and digital calibration to ensure consistent, high-quality parts across machines and locations.

Polymer development requires characterization and testing of many formulas. However, polymers are viscous and thus difficult to mix and formulate by hand. SPOC technology combines active-mixing direct ink write (DIW) deposition methods with in-situ characterization to enable high-throughput screening and optimization of polymer formulas, greatly accelerating materials discovery.

The platform reconstructs the full multi-tier supplier network, for a given manufacturer’s product down to raw-material origin, without the need for time-consuming supplier surveys. Based on Graph AI technology, it returns confidence-scored deep-tier intelligence that powers supply chain risk, resilience, and compliance decisions that would not have otherwise been possible.

TeleSENSE is a low-cost distributed fiber-optic sensing platform that transforms standard telecommunications fiber into high-performance sensing fibers, scaled through agentic AI-enabled manufacturing for infrastructure, energy, telecommunications, medical device, and robotics applications.

Waste terpenes captured from the wood industry provide no/low-cost cyclic hydrocarbons that are otherwise released as toxic gases. These cyclic hydrocarbons are converted on metal and acid heterogeneous catalysts to aviation fuel aromatics and cycloalkanes. Aromatics and cycloalkanes enable alternative aviation fuel sources, improving energy independence of critical fuel.

This thermophilic one-pot biocatalytic platform transforms unrecycled polyethylene terephthalate plastic waste into a value-added aromatic, protocatechuic acid (PCA), in a single high-temperature step. The PCA produced has immediate commercial value: it can be sold as an antioxidant ingredient or a polymer precursor. This technology therefore supports sustainable, circular-economy plastic recycling.

UltraSep is a novel ultrasonic technique used to separate solids from liquids with broad applications. UltraSep is a compact, semi-autonomous system with few consumable parts, enhancing worker safety in industrial and nuclear settings and serving as an environmentally friendly alternative to conventional technologies.

VIBRANT is a surrogate nuclear reactor that replaces neutrons with photons to accelerate advanced reactor development. VIBRANT uses an LED-driven core, photodiodes, thermocouples and high-performance software interfaces to help mature advanced reactor instrumentation and control systems, validate models, and facilitate system integration prior to deployment in real-world reactor prototypes.

The 5G Interference Canceler is a first-of-its-kind prototyped algorithm that exploits features in the 5G New Radio downlink waveform to cancel the interference it imposes on co-channel radio frequency systems used by both government and commercial entities.

AIRC Prostate MR is, to date, the first and only prostate MRI AI solution with regulatory approval in the U.S., Europe, and China. It enables automated gland segmentation, lesion detection, and structured reporting, improving diagnostic accuracy, consistency, and efficiency. It supports scalable, high-quality prostate cancer diagnosis across diverse clinical settings.

The All Hazards Analysis tool employs machine learning and natural language processing to help analysts identify infrastructure interdependencies. By modeling how failures can propagate, it helps utilities and public entities plan for emergencies and mitigate their effects on essential services such as power, water and healthcare during crisis situations.

ALTRIOS is the first validated, open-source digital twin for the freight rail system. It represents a paradigm shift away from stovepiped legacy tools by creating unified, integrated simulations of train components, locomotives, fleets, rail networks, and freight terminals, plus energy use and costs over decades, to de-risk long-term, capital-intensive deployments.

The Autonomous Robot Control Hierarchy (ARCH) is a universal software framework for the programming and control of autonomous robotic systems. Its modular, object-oriented architecture provides platform-independent autonomous tools, including perception, planning, navigation, and control, that integrate seamlessly with any robot configuration, significantly reducing development time for complex autonomous behaviors.

Drugs and chemicals affect tissues very differently, and some tissues (e.g., brain and intestine) are particularly difficult to model for health/medicine. ATLAS-Tx, which is built on existing in vivo transcriptomics knowledge and data, translates the gene expression profile from one tissue (e.g., liver) to seven other tissues, including brain and intestine.

Multi-Event Grid Intelligence Platform for Wildfire Prevention and Resilience (AWARE) detects, classifies, and reports multiple power-grid anomalies within a single waveform, including low-current arcing. It provides utilities with early warning of hidden faults to help prevent wildfires, reduce outages, and improve grid resilience, reliability, and operational decision-making.

Fenix is a software fault-tolerance framework designed to enhance the performance and efficiency of large-scale computing. It utilizes existing resources and reduces the reliance on hardware reliability. This addresses the rising energy demands and environmental impacts associated with the modern computational power requirements of scientific computing and machine learning.

GeoWISE is a web-based, techno-economic, simulation platform that enables rapid design and evaluation of ground-source and district geothermal heating and cooling systems. It automatically generates building energy models, sizes geothermal systems, and estimates energy performance and costs using integrated simulation tools and a nationwide building database.

Argonne’s IMPACT model is a publicly available tool that evaluates the costs and environmental impacts of advanced materials manufacturing and processing technologies and benchmarks them against their conventional counterparts to guide technology research, development, and deployment. It has been used to inform lithium production and cathode material synthesis.

IntelliGlass enables the next generation of human-centric smart living by overcoming information overload in smart glasses. Through multimodal sensing and AI-driven adaptation, it optimizes visual, audio, and interaction experiences in real time, allowing smart glasses to seamlessly augment human intelligence, productivity, safety, and quality of life.

LAPIS (Linear Algebra Performance for Intermediate Subprograms) is a compiler that automatically turns a user’s high-level program into Kokkos C++ code. LAPIS supports both sparse and dense linear algebra, and the Kokkos code it generates can run on all supercomputer architectures.

LuGo is a novel quantum computing framework that drastically enhances quantum phase estimation (QPE). It utilizes parallelization to generate circuits independently, and thereby, eliminate exponential circuit duplication, which dramatically reduces generation time, gate counts, and circuit depth without sacrificing fidelity (solution accuracy).

AI is putting cyber exploitation on steroids. When embedded systems are hacked, missions fail and lives are lost. Magnetite is MIT Lincoln Laboratory’s secure-by-design, real-time operating system, built in Rust on verified seL4. It shrinks attack surface, limits compromise, and in 2025 transitioned into space and flight integrations.

OmniDrive is a high-reliability intelligent driving solution integrating 360° perception, AI monitoring, Level-3 autonomy, MRM, and secure OTA updating. Featuring a modular and retrofittable design, it enables rapid deployment across diverse vehicle platforms, significantly enhancing road safety, operational efficiency, and the scalability of intelligent transportation.

The Pele suite of reacting flow codes harnesses the world’s most powerful exascale computers to simulate the complex physics of turbulent reacting flows with ultra-high-fidelity. Its open-source architecture allows users to easily modify the code and integrate with artificial intelligence to advance scientific discovery and de-risk technologies.

RadSim is an open-source gamma-ray detector simulation platform that generates realistic spectra and gross counts from source to detector. By combining radiation transport, detector physics, and machine-learning response model, it enables researchers and industry partners to model gamma-ray detectors response to simulated flux from the transport tool of their choice.

Simulator of QUantum Network Communication (SeQUeNCe) is a modular, open-source discrete-event simulator that enables detailed, accurate, and highly customizable simulations of heterogeneous quantum networks. SeQUeNCe can be used to understand the trade-offs of alternative quantum network architectures, optimize quantum hardware, and study quantum networking protocols at scale.

TELEIOS-BATT, developed by Idaho National Laboratory, is an advanced suite of physics-based diagnostic and prognostic tools for batteries. It offers fast, accurate, and efficient monitoring, simulation, and prediction of battery health and aging, helping maximize safety, reliability, and cost-efficiency for modern battery-dependent industries.

POSSUM is the only reconstruction algorithm that implements the pioneering TOF-CGI technique that enables imaging and monitoring of radionuclide therapy treatments in PET/CT clinical scanners, resulting in superior efficiency, contrast, and signal-to-noise ratio when compared to state-of-the-art clinical options in nuclear medicine.

TMAP8 tackles fusion energy’s biggest hidden challenge: tritium management. It is the first open-source tritium modeling platform to meet the Nuclear Quality Assurance Level 1 (NQA-1) industry standard. The tool combines regulatory rigor; multiscale, multiphysics simulation; and broad industry accessibility and applicability to accelerate safe, commercially viable fusion power.

Vision Shepherd evolves manufacturing systems from capacity-driven models into ‘human-centric’ smart ecosystems. By utilizing AI Boxes and Embodied Robots, it proactively empowers frontline workers while institutionalizing knowledge assets. Vision Shepherd has helped Foxconn save USD 22.5 million annually and eliminate 1.85 million man-years of redundant rework.

Viz Subdural Plus uses AI to automatically identify, label, and quantify subdural collections on non-contrast head CT. Objective measurements of total volume, maximum thickness, and midline shift, combined with a color-coded overlay, support faster clinical evaluation and treatment decision-making.

CarbonLock Blocks turns concrete into a permanent carbon vault. By capturing CO₂ and mineralizing it within calcium-rich material, the concrete can achieve about 1% CO₂ uptake by mass. Each kilogram of added biochar delivers around 2 kilograms of carbon reduction, while AI optimization reduces cement use and improves material quality.