LLNL researchers have designed and developed a novel high-density, high-channel count 3D connector that enables hundreds or thousands of nonpermanent connections within a compact footprint. The connector addresses limitations of currently used conventional approaches that were described previously, which have an artificial ceiling on the number of recording sites of modern devices of no more…

LLNL’s researchers use physical vapor deposition (sputter deposition or electron beam deposition) to coat an inert gasket material (i.e. PTFE) with a conductive metal (i.e. copper). The gas diffusion electrode overlaps onto the copper coated gasket to allow for electrical conductivity between the catalyst surface and the flow field/current collector of a CO2 electrolyzer. The coated gasket…

Using their computational design optimization, LLNL researchers have developed copper-based dilute alloy catalysts (contains <10 at.% of the minority metal alloy component) and demonstrated these novel catalysts have improved energy efficiency and selectivity of the methane conversion reaction.  By alloying copper with a small amount of the electropositive minority metal element, the…

LLNL’s innovation offers an alternate synthetic route to graphite at lower cost using a molten salt mixture of CaCl2-CaCO3-CaO.  The synthetic production of graphite and other high-value carbon materials is accomplished in molten salt media via electrochemical reduction and transformation of the carbon from the carbonate ion. The broad electrochemical window of molten salts enables the…

The novel LLNL approach is to use projection microstereolithography (LAPµSL), starting with a photocurable methacrylate resin formulation consisting of a combination of a photoinitiator, photoabsorber, inhibitor, solvents, and other additives.  Prior to use, the resin is pretreated to control viscosity for easier handling.  The resin is fed to a LAPµSL printer which employs a near UV…

The inventors have developed a 3% Yttria partially-stabilized Zirconia (3YZ) ceramic ink that produces parts with both nano and microporosity and is compatible with two AM techniques: DIW and projection microstereolithography (PμSL). The 3YZ nano-porous ceramic printed parts had engineered macro cavities measuring several millimeters in length, wall thicknesses ranging from 200 to 540 μm, and…

This invention describes a multiple nozzle microfluidic unit that allows simultaneous generation streams of multiple layered coaxial liquid jets. Liquids are pumped into the device at a combined flow rate from 100 mL/hr to 10 L/hr. Droplets are created with diameters in the range of 1 µm to 5 mm and can be created with 1-2 shell layers encapsulating fluid. Droplets created from the system can…

Many of the disadvantages of current interface devices can be overcome with LLNL’s novel interface design, which relies on area array distribution where independent interface connector subassemblies are positioned in a planar grid.  Not only is the interface device expandable area-wise (without increasing contact force), but it could also be expanded height-wise, with multiple layers of…

Commercial fiber optic cables are the current standard for carrying optical signals in industries like communications or medical devices. However, the fibers are made of glass, which do not have favorable characteristics for applications that require flexibility and re-routing, e.g. typically brittle, limited selection of materials, dimension constraints.

Lawrence Livermore National Laboratory (LLNL) is offering the opportunity to collaborate in accelerating artificial intelligence (AI) for applied science, including research in key areas such as advanced material design, 3D printing, predictive biology, energy systems, “self-driving” lasers and fusion energy research.

Clinical images have a wealth of data that are currently untapped by physicians and machine learning (ML) methods alike. Most ML methods require more data than is available to sufficiently train them. In order to obtain all data contained in a clinical image, it is imperative to be able to utilize multimodal, or various types of, data such as tags or identifications, especially where spatial…

Some COVID-19 diagnoses are utilizing computed tomography (CT)-scans for triage. CT-scans produce immediate results with high sensitivity. The digital images produced by a CT-scan require physicians to identify objects within the image to determine the presence of disease. Object identification can be done using machine learning (ML) techniques such as deep learning (DL) to improve speed and…

MimicGAN represents a new generation of methods that can “self-correct” for unseen corruptions in the data out in the field. This is particularly useful for systems that need to be deployed autonomously without needing constant intervention such as Automated Driver Assistance Systems. MimicGAN achieves this by treating every test sample as “corrupt” by default. The goal is to determine (a) the…

LLNL has developed a new system, called the Segmentation Ensembles System, that provides a simple and general way to fuse high-level and low-level information and leads to a substantial increase in overall performance of digital image analysis. LLNL researchers have demonstrated the effectiveness of the approach on applications ranging from automatic threat detection for airport security, to…

LLNL has developed a brain-on-a-chip system with a removable cell-seeding funnel to simultaneously localize neurons from various brain regions in an anatomically relevant manner and over specific electrode regions of a MEA. LLNL’s novel, removable cell seeding funnel uses a combination of 3D printing and microfabrication that allows neurons from select brain regions to easily be seeded into…