Tag Archives: Kevin Yager

Chatbot with expertise in nanomaterials

Leave a reply

This December 1, 2023 news item on phys.org starts with a story,

A researcher has just finished writing a scientific paper. She knows her work could benefit from another perspective. Did she overlook something? Or perhaps there’s an application of her research she hadn’t thought of. A second set of eyes would be great, but even the friendliest of collaborators might not be able to spare the time to read all the required background publications to catch up.

Kevin Yager—leader of the electronic nanomaterials group at the Center for Functional Nanomaterials (CFN), a U.S. Department of Energy (DOE) Office of Science User Facility at DOE’s Brookhaven National Laboratory—has imagined how recent advances in artificial intelligence (AI) and machine learning (ML) could aid scientific brainstorming and ideation. To accomplish this, he has developed a chatbot with knowledge in the kinds of science he’s been engaged in.

…

A December 1, 2023 DOE/Brookhaven National Laboratory news release by Denise Yazak (also on EurekAlert), which originated the news item, describes a research project with a chatbot that has nanomaterial-specific knowledge, Note: Links have been removed,

Rapid advances in AI and ML have given way to programs that can generate creative text and useful software code. These general-purpose chatbots have recently captured the public imagination. Existing chatbots—based on large, diverse language models—lack detailed knowledge of scientific sub-domains. By leveraging a document-retrieval method, Yager’s bot is knowledgeable in areas of nanomaterial science that other bots are not. The details of this project and how other scientists can leverage this AI colleague for their own work have recently been published in Digital Discovery.

Rise of the Robots

“CFN has been looking into new ways to leverage AI/ML to accelerate nanomaterial discovery for a long time. Currently, it’s helping us quickly identify, catalog, and choose samples, automate experiments, control equipment, and discover new materials. Esther Tsai, a scientist in the electronic nanomaterials group at CFN, is developing an AI companion to help speed up materials research experiments at the National Synchrotron Light Source II (NSLS-II).” NSLS-II is another DOE Office of Science User Facility at Brookhaven Lab.

At CFN, there has been a lot of work on AI/ML that can help drive experiments through the use of automation, controls, robotics, and analysis, but having a program that was adept with scientific text was something that researchers hadn’t explored as deeply. Being able to quickly document, understand, and convey information about an experiment can help in a number of ways—from breaking down language barriers to saving time by summarizing larger pieces of work.

Watching Your Language

To build a specialized chatbot, the program required domain-specific text—language taken from areas the bot is intended to focus on. In this case, the text is scientific publications. Domain-specific text helps the AI model understand new terminology and definitions and introduces it to frontier scientific concepts. Most importantly, this curated set of documents enables the AI model to ground its reasoning using trusted facts.

To emulate natural human language, AI models are trained on existing text, enabling them to learn the structure of language, memorize various facts, and develop a primitive sort of reasoning. Rather than laboriously retrain the AI model on nanoscience text, Yager gave it the ability to look up relevant information in a curated set of publications. Providing it with a library of relevant data was only half of the battle. To use this text accurately and effectively, the bot would need a way to decipher the correct context.

“A challenge that’s common with language models is that sometimes they ‘hallucinate’ plausible sounding but untrue things,” explained Yager. “This has been a core issue to resolve for a chatbot used in research as opposed to one doing something like writing poetry. We don’t want it to fabricate facts or citations. This needed to be addressed. The solution for this was something we call ‘embedding,’ a way of categorizing and linking information quickly behind the scenes.”

Embedding is a process that transforms words and phrases into numerical values. The resulting “embedding vector” quantifies the meaning of the text. When a user asks the chatbot a question, it’s also sent to the ML embedding model to calculate its vector value. This vector is used to search through a pre-computed database of text chunks from scientific papers that were similarly embedded. The bot then uses text snippets it finds that are semantically related to the question to get a more complete understanding of the context.

The user’s query and the text snippets are combined into a “prompt” that is sent to a large language model, an expansive program that creates text modeled on natural human language, that generates the final response. The embedding ensures that the text being pulled is relevant in the context of the user’s question. By providing text chunks from the body of trusted documents, the chatbot generates answers that are factual and sourced.

“The program needs to be like a reference librarian,” said Yager. “It needs to heavily rely on the documents to provide sourced answers. It needs to be able to accurately interpret what people are asking and be able to effectively piece together the context of those questions to retrieve the most relevant information. While the responses may not be perfect yet, it’s already able to answer challenging questions and trigger some interesting thoughts while planning new projects and research.”

Bots Empowering Humans

CFN is developing AI/ML systems as tools that can liberate human researchers to work on more challenging and interesting problems and to get more out of their limited time while computers automate repetitive tasks in the background. There are still many unknowns about this new way of working, but these questions are the start of important discussions scientists are having right now to ensure AI/ML use is safe and ethical.

“There are a number of tasks that a domain-specific chatbot like this could clear from a scientist’s workload. Classifying and organizing documents, summarizing publications, pointing out relevant info, and getting up to speed in a new topical area are just a few potential applications,” remarked Yager. “I’m excited to see where all of this will go, though. We never could have imagined where we are now three years ago, and I’m looking forward to where we’ll be three years from now.”

For researchers interested in trying this software out for themselves, the source code for CFN’s chatbot and associated tools can be found in this github repository.

Brookhaven National Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit science.energy.gov.

Here’s a link to and a citation for the paper,

Domain-specific chatbots for science using embeddings by Kevin G. Yager.
Digital Discovery, 2023,2, 1850-1861 DOI: https://doi.org/10.1039/D3DD00112A
First published 10 Oct 2023

This paper appears to be open access.

Directing self-assembly of multiple molecular patterns within a single material

Leave a reply

Self-assembly in this context references the notion of ‘bottom-up engineering’, that is, following nature’s engineering process where elements assemble themselves into a plant, animal, or something else. Humans have for centuries used an approach known as ‘top-down engineering’ where we take materials and reform them, e.g., trees into paper or houses.

Theoretically, bottom-up engineering (self-assembly) is more efficient than top-down engineering but we have yet to become as skilled as Nature at the process.

Scientists at the US Brookhaven National Laboratory believe they have taken a step in the right direction with regard to self-assembly. From an Aug. 8, 2016 Brookhaven National Laboratory news release (also on EurekAlert) by Justin Eure describes the research (Note: A link has been removed),

To continue advancing, next-generation electronic devices must fully exploit the nanoscale, where materials span just billionths of a meter. But balancing complexity, precision, and manufacturing scalability on such fantastically small scales is inevitably difficult. Fortunately, some nanomaterials can be coaxed into snapping themselves into desired formations-a process called self-assembly.

Scientists at the U.S. Department of Energy’s (DOE) Brookhaven National Laboratory have just developed a way to direct the self-assembly of multiple molecular patterns within a single material, producing new nanoscale architectures. The results were published in the journal Nature Communications.

“This is a significant conceptual leap in self-assembly,” said Brookhaven Lab physicist Aaron Stein, lead author on the study. “In the past, we were limited to a single emergent pattern, but this technique breaks that barrier with relative ease. This is significant for basic research, certainly, but it could also change the way we design and manufacture electronics.”

Microchips, for example, use meticulously patterned templates to produce the nanoscale structures that process and store information. Through self-assembly, however, these structures can spontaneously form without that exhaustive preliminary patterning. And now, self-assembly can generate multiple distinct patterns-greatly increasing the complexity of nanostructures that can be formed in a single step.

“This technique fits quite easily into existing microchip fabrication workflows,” said study coauthor Kevin Yager, also a Brookhaven physicist. “It’s exciting to make a fundamental discovery that could one day find its way into our computers.”

The experimental work was conducted entirely at Brookhaven Lab’s Center for Functional Nanomaterials (CFN), a DOE Office of Science User Facility, leveraging in-house expertise and instrumentation.

Cooking up organized complexity

The collaboration used block copolymers-chains of two distinct molecules linked together-because of their intrinsic ability to self-assemble.

“As powerful as self-assembly is, we suspected that guiding the process would enhance it to create truly ‘responsive’ self-assembly,” said study coauthor Greg Doerk of Brookhaven. “That’s exactly where we pushed it.”

To guide self-assembly, scientists create precise but simple substrate templates. Using a method called electron beam lithography-Stein’s specialty-they etch patterns thousands of times thinner than a human hair on the template surface. They then add a solution containing a set of block copolymers onto the template, spin the substrate to create a thin coating, and “bake” it all in an oven to kick the molecules into formation. Thermal energy drives interaction between the block copolymers and the template, setting the final configuration-in this instance, parallel lines or dots in a grid.

“In conventional self-assembly, the final nanostructures follow the template’s guiding lines, but are of a single pattern type,” Stein said. “But that all just changed.”

Lines and dots, living together

The collaboration had previously discovered that mixing together different block copolymers allowed multiple, co-existing line and dot nanostructures to form.

“We had discovered an exciting phenomenon, but couldn’t select which morphology would emerge,” Yager said. But then the team found that tweaking the substrate changed the structures that emerged. By simply adjusting the spacing and thickness of the lithographic line patterns-easy to fabricate using modern tools-the self-assembling blocks can be locally converted into ultra-thin lines, or high-density arrays of nano-dots.

“We realized that combining our self-assembling materials with nanofabricated guides gave us that elusive control. And, of course, these new geometries are achieved on an incredibly small scale,” said Yager.

“In essence,” said Stein, “we’ve created ‘smart’ templates for nanomaterial self-assembly. How far we can push the technique remains to be seen, but it opens some very promising pathways.”

Gwen Wright, another CFN coauthor, added, “Many nano-fabrication labs should be able to do this tomorrow with their in-house tools-the trick was discovering it was even possible.”

The scientists plan to increase the sophistication of the process, using more complex materials in order to move toward more device-like architectures.

“The ongoing and open collaboration within the CFN made this possible,” said Charles Black, director of the CFN. “We had experts in self-assembly, electron beam lithography, and even electron microscopy to characterize the materials, all under one roof, all pushing the limits of nanoscience.”

Here’s a link to and a citation for the paper,

Selective directed self-assembly of coexisting morphologies using block copolymer blends by A. Stein, G. Wright, K. G. Yager, G. S. Doerk, & C. T. Black. Nature Communications 7, Article number: 12366  doi:10.1038/ncomms12366 Published 02 August 2016

This paper is open access.

Cellulose nanocrystals and supercapacitors at McMaster University (Canada)

Leave a reply

Photos: Xuan Yang and Kevin Yager.

Photos: Xuan Yang and Kevin Yager. Courtesy McMaster University

I love that featherlike structure holding up a tiny block of something while balanced on what appears to be a series of medallions. What it has to do with supercapacitors (energy storage) and cellulose nanocrystals is a mystery but that’s one of the images you’ll find illustrating an Oct. 7, 2015 news item on Nanotechnology Now featuring research at McMaster University,

McMaster Engineering researchers Emily Cranston and Igor Zhitomirsky are turning trees into energy storage devices capable of powering everything from a smart watch to a hybrid car.

The scientists are using cellulose, an organic compound found in plants, bacteria, algae and trees, to build more efficient and longer-lasting energy storage devices or supercapacitors. This development paves the way toward the production of lightweight, flexible, and high-power electronics, such as wearable devices, portable power supplies and hybrid and electric vehicles.

A Sept. 10, 2015 McMaster University news release, which originated the news item, describes the research in more detail,

Cellulose offers the advantages of high strength and flexibility for many advanced applications; of particular interest are nanocellulose-based materials. The work by Cranston, an assistant chemical engineering professor, and Zhitomirsky, a materials science and engineering professor, demonstrates an improved three-dimensional energy storage device constructed by trapping functional nanoparticles within the walls of a nanocellulose foam.

The foam is made in a simplified and fast one-step process. The type of nanocellulose used is called cellulose nanocrystals and looks like uncooked long-grain rice but with nanometer-dimensions. In these new devices, the ‘rice grains’ have been glued together at random points forming a mesh-like structure with lots of open space, hence the extremely lightweight nature of the material. This can be used to produce more sustainable capacitor devices with higher power density and faster charging abilities compared to rechargeable batteries.

Lightweight and high-power density capacitors are of particular interest for the development of hybrid and electric vehicles. The fast-charging devices allow for significant energy saving, because they can accumulate energy during braking and release it during acceleration.

For anyone interested in a more detailed description of supercapacitors, there’s my favourite one which involves Captain America’s shield along with some serious science in my April 28, 2014 posting.

Getting back to the research at McMaster, here’s a link to and a citation for the paper,

Cellulose Nanocrystal Aerogels as Universal 3D Lightweight Substrates for Supercapacitor Materials by Xuan Yang, Kaiyuan Shi, Igor Zhitomirsky, and Emily D. Cranston. Advanced Materials DOI: 10.1002/adma.201502284View/save citation First published online 2 September 2015

This paper is behind a paywall.

One final bit, cellulose nanocrystals (CNC) are sometimes referred to as nanocrystalline cellulose (NCC).