North Carolina State University
Assembly Line
Researchers Demonstrate Self-Assembling Electronics
Researchers have demonstrated a new technique for self-assembling electronic devices. The proof-of-concept work was used to create diodes and transistors, and paves the way for self-assembling more complex electronic devices without relying on existing computer chip manufacturing techniques.
“Existing chip manufacturing techniques involve many steps and rely on extremely complex technologies, making the process costly and time consuming,” says Martin Thuo, corresponding author of a paper on the work and a professor of materials science and engineering at North Carolina State University. “Our self-assembling approach is significantly faster and less expensive. We’ve also demonstrated that we can use the process to tune the bandgap for semiconductor materials and to make the materials responsive to light – meaning this technique can be used to create optoelectronic devices.
Thuo calls the new, self-assembling technique a directed metal-ligand (D-Met) reaction. You start with liquid metal particles. For their proof-of-concept work, the researchers used Field’s metal, which is an alloy of indium, bismuth and tin. The liquid metal particles are placed next to a mold, which can be made to any size or pattern. A solution is then poured onto the liquid metal. The solution contains molecules called ligands that are made up of carbon and oxygen. These ligands harvest ions from the surface of the liquid metal and hold those ions in a specific geometric pattern. The solution flows across the liquid metal particles and is drawn into the mold. As the solution flows into the mold, the ion-bearing ligands begin assembling themselves into more complex, three-dimensional structures. Meanwhile, the liquid part of the solution begins to evaporate, which serves to pack the complex structures closer and closer together into an array.
“Without the mold, these structures can form somewhat chaotic patterns,” Thuo says. “But because the solution is constrained by the mold, the structures form in predictable, symmetrical arrays.”
Once a structure has reached the desired size, the mold is removed, and the array is heated. This heat breaks up the ligands, freeing the carbon and oxygen atoms. The metal ions interact with the oxygen to form semiconductor metal oxides, while the carbon atoms form graphene sheets. These ingredients assemble themselves into a well-ordered structure consisting of semiconductor metal oxide molecules wrapped in graphene sheets. The researchers used this technique to create nanoscale and microscale transistors and diodes.
“The graphene sheets can be used to tune the bandgap of the semiconductors, making the semiconductor more or less responsive, depending on the quality of the graphene,” says Julia Chang, first author of the paper and a postdoctoral researcher at NC State. In addition, because the researchers used bismuth in the proof-of-concept work, they were able to make structures that are photo-responsive. This allows the researchers to manipulate the properties of the semiconductors using light.
New Technique Improves Finishing Time for 3D Printed Machine Parts
North Carolina State University researchers have demonstrated a technique that allows people who manufacture metal machine parts with 3D printing technologies to conduct automated quality control of manufactured parts during the finishing process. The technique allows users to identify potential flaws without having to remove the parts from the manufacturing equipment, making production time more efficient. Specifically, the researchers have integrated 3D printing, automated machining, laser scanning and touch-sensitive measurement technologies with related software to create a largely automated system that produces metal machine components that meet critical tolerances.
When end users need a specific part, they pull up a software file that includes the measurements of the desired part. A 3D printer uses this file to print the part, which includes metal support structures. Users then take the printed piece and mount it in a finishing device using the support structure. At this point, lasers scan the mounted part to establish its dimensions. A software program then uses these dimensions and the desired critical tolerances to guide the finishing device, which effectively polishes out any irregularities in the part. As this process moves forward, the finishing device manipulates the orientation of the printed part so that it can be measured by a touch-sensitive robotic probe that ensures the part’s dimensions are within the necessary parameters.
Manufacturing service capability prediction with Graph Neural Networks
In the current landscape, the predominant methods for identifying manufacturing capabilities from manufacturers rely heavily on keyword matching and semantic matching. However, these methods often fall short by either overlooking valuable hidden information or misinterpreting critical data. Consequently, such approaches result in an incomplete identification of manufacturers’ capabilities. This underscores the pressing need for data-driven solutions to enhance the accuracy and completeness of manufacturing capability identification. To address the need, this study proposes a Graph Neural Network-based method for manufacturing service capability identification over a knowledge graph. To enhance the identification performance, this work introduces a novel approach that involves aggregating information from the graph nodes’ neighborhoods as well as oversampling the graph data, which can be effectively applied across a wide range of practical scenarios. Evaluations conducted on a Manufacturing Service Knowledge Graph and subsequent ablation studies demonstrate the efficacy and robustness of the proposed approach. This study not only contributes a innovative method for inferring manufacturing service capabilities but also significantly augments the quality of Manufacturing Service Knowledge Graphs.