Knowledge-integrated machine learning for materials: lessons from gameplaying and robotics

Self-driving laboratories (SDLs), which combine automated experimental hardware with computational experiment planning, have emerged as powerful tools for accelerating materials discovery. The intrinsic complexity created by their multitude of components requires an effective orchestration platform to ensure the correct operation of diverse experimental setups. Existing orchestration frameworks, however, are either tailored to specific setups or have not been implemented for real-world synthesis. To address these issues, we introduce ChemOS 2.0, an orchestration architecture that efficiently coordinates communication, data exchange, and instruction management among modular laboratory components. By treating the laboratory as an "operating system," ChemOS 2.0 combines ab initio calculations, experimental orchestration, and statistical algorithms to guide closed-loop operations. To demonstrate its capabilities, we showcase ChemOS 2.0 in a case study focused on discovering organic laser molecules. The results confirm ChemOS 2.0's prowess in accelerating materials research and demonstrate its potential as a valuable design for future SDL platforms. 

ChemOS 2.0: An orchestration architecture for chemical self-driving laboratories

Conventional silicon (Si)-based power devices face physical limitations—such as switching speed and energy efficiency—which can make it difficult to meet the increasing demand for high-power, low-loss, and fast-switching-frequency power devices in power electronic converter systems. Gallium nitride (GaN) is an excellent candidate for next-generation power devices, capable of improving the conversion efficiency of power systems owing to its wide band gap, high mobility, and high electric breakdown field. Apart from their cost effectiveness, GaN-based power high-electron-mobility transistors (HEMTs) on Si substrates exhibit excellent properties—such as low ON-resistance and fast switching—and are used primarily in power electronic applications in the fields of consumer electronics, new energy vehicles, and rail transit, amongst others. During the past decade, GaN-on-Si power HEMTs have made major breakthroughs in the development of GaN-based materials and device fabrication. However, the fabrication of GaN-based HEMTs on Si substrates faces various problems—for example, large lattice and thermal mismatches, as well as ‘melt-back etching’ at high temperatures between GaN and Si, and buffer/surface trapping induced leakage current and current collapse. These problems can lead to difficulties in both material growth and device fabrication. In this review, we focused on the current status and progress of GaN-on-Si power HEMTs in terms of both materials and devices. For the materials, we discuss the epitaxial growth of both a complete multilayer HEMT structure, and each functional layer of a HEMT structure on a Si substrate. For the devices, breakthroughs in critical fabrication technology and the related performances of GaN-based power HEMTs are discussed, and the latest development in GaN-based HEMTs are summarised. Based on recent progress, we speculate on the prospects for further development of GaN-based power HEMTs on Si. This review provides a comprehensive understanding of GaN-based HEMTs on Si, aiming to highlight its development in the fields of microelectronics and integrated circuit technology.

GaN-based power high-electron-mobility transistors on Si substrates: from materials to devices

Globus Automation Services: Research process automation across the space-time continuum

Abstract The development of automated high-throughput experimental platforms has enabled fast sampling of high-dimensional decision spaces. To reach target properties efficiently, these platforms are increasingly paired with intelligent experimental design. However, current optimizers show limitations in maintaining sufficient exploration/exploitation balance for problems dealing with multiple conflicting objectives and complex constraints. Here, we devise an Evolution-Guided Bayesian Optimization (EGBO) algorithm that integrates selection pressure in parallel with a q-Noisy Expected Hypervolume Improvement (qNEHVI) optimizer; this not only solves for the Pareto Front (PF) efficiently but also achieves better coverage of the PF while limiting sampling in the infeasible space. The algorithm is developed together with a custom self-driving lab for seed-mediated silver nanoparticle synthesis, targeting 3 objectives (1) optical properties, (2) fast reaction, and (3) minimal seed usage alongside complex constraints. We demonstrate that, with appropriate constraint handling, EGBO performance improves upon state-of-the-art qNEHVI. Furthermore, across various synthetic multi-objective problems, EGBO shows significative hypervolume improvement, revealing the synergy between selection pressure and the qNEHVI optimizer. We also demonstrate EGBO’s good coverage of the PF as well as comparatively better ability to propose feasible solutions. We thus propose EGBO as a general framework for efficiently solving constrained multi-objective problems in high-throughput experimentation platforms. 

Evolution-guided Bayesian optimization for constrained multi-objective optimization in self-driving labs

In this work, time-dependent threshold voltage (<inline-formula> <tex-math notation="LaTeX">${V}_{\text {TH}}$ </tex-math></inline-formula>) instability of p-GaN gate high electron mobility transistors (HEMTs) under OFF-state drain stress is systematically investigated. An anomalous <inline-formula> <tex-math notation="LaTeX">${V}_{\text {TH}}$ </tex-math></inline-formula> shift behavior is observed with an initial negative shift followed by a kink and a positive shift upon prolonged stress time. Through bias- and temperature-dependent stress/recovery experiments, it has been found that the process of electron releasing by donor states at the p-GaN/AlGaN and AlGaN/GaN interface is the primary reason for negative <inline-formula> <tex-math notation="LaTeX">${V}_{\text {TH}}$ </tex-math></inline-formula> shift in the initial stage. The hole deficiency in the p-GaN layer gradually dominates the change in <inline-formula> <tex-math notation="LaTeX">${V}_{\text {TH}}$ </tex-math></inline-formula> and finally leads to a positive <inline-formula> <tex-math notation="LaTeX">${V}_{\text {TH}}$ </tex-math></inline-formula> shift. Moreover, a continual leakage path between the gate and drain terminal during the stress stage is further revealed which is mainly ascribed to the trap-assisted electron tunneling across the AlGaN layer. In addition, the effect of OFF-state drain stress on static ON-resistance (<inline-formula> <tex-math notation="LaTeX">${R}_{\text{ON}}$ </tex-math></inline-formula>) is also studied, and it is found that the trapped electrons at the passivation/AlGaN interface and/or in the GaN buffer layer results in the <inline-formula> <tex-math notation="LaTeX">${R}_{\text{ON}}$ </tex-math></inline-formula> degradation.

Observation and Analysis of Anomalous VTH Shift of p-GaN Gate HEMTs Under off-State Drain Stress

In this work, the reliability degradation of a 650 V-class superjunction (SJ) MOSFET under continuous violent OFF-state avalanche-breakdown stress is studied. Decreased threshold voltage and increasing saturation current have been observed which are attributed to the injection of hot holes below the trench bottom into the gate oxide layer. High-energy electrons and holes are generated under OFF-state avalanche stress leading to accelerated and violent collision ionization. The body diode suffers from severe avalanche breakdown leading to heat generation by collision ionization. In addition, the injection of hot holes is further enhanced with the assistance of the electrical field at the bottom of the gate. These phenomena are further analyzed by experimental characterizations and theoretical simulation using TCAD tools. Our comprehensive study on the reliability degradation behaviors of SJ devices under OFF-state avalanche-breakdown stress can be instructive in future device optimization and failure analysis.

Impact Analysis of Off-State Avalanche-Breakdown Stress on 650 V-Class Superjunction MOSFET

In this work, multiple time-dependent gate breakdown (BD) experiments are performed on p-GaN gate high-electron-mobility transistors (HEMTs) under constant gate voltage stress. Progressive gate BD behaviors are observed during the time-dependent failure process. It is found that the time-to-BD (<inline-formula> <tex-math notation="LaTeX">${t}_{\text {BD}}$ </tex-math></inline-formula>) has a positive dependence on temperature but a weak relevance to p-GaN gate length. Moreover, it is confirmed that the Schottky junction first (1st) degenerates after the 1st stage jump of <inline-formula> <tex-math notation="LaTeX">${I}_{G}$ </tex-math></inline-formula> and the p-i-n junction fails in the subsequent process. The underlying failure mechanisms are revealed by further analyzing the degradation behaviors and the carrier transport mechanisms before/after the Schottky junction degrades. Hole Fowler–Nordheim (FN) tunneling at the metal gate/p-GaN interface guides the first degradation of the Schottky junction. Afterward, the trap-assisted electron tunneling across the AlGaN layer indicates the failure of the p-i-n junction. These two processes sequentially facilitate defect generation within the p-GaN and the AlGaN layer, resulting in a percolation path between the gate and the source terminal.

Investigation of the Progressive Gate Breakdown Behaviors in p-GaN Gate HEMTs

Trench-Gate Bipolar Transistor With Partially Buried Carrier Storage Layer for Enhanced Blocking and Switching Characteristics

An Object-Oriented Framework to Enable Workflow Evolution Across Materials Acceleration Platforms

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

J. Tan

Author Tools

Chat about Author