Chemistry

Home » Archives for Jennifer Brown » Page 4
Author: Jennifer Brown
Vogiatzis Featured Image

Vogiatzis Group Publishes in Journal of Physical Chemistry Letters

by

Grier Jones, fifth year chemistry PhD student, and Associate Professor Konstantinos Vogiatzis recently published a new data-driven quantum chemistry method, based on the reduced-density matrix (RDM) formulation of quantum mechanics, in the Journal of Physical Chemistry Letters. This publication was developed in collaboration with University of Tennessee, Knoxville alumnus Professor A. Eugene DePrince (’05) and his research group at Florida State University. DePrince’s group specializes in the development of novel RDM methods for the treatment of strongly correlated electrons.

Strong electron correlation lies at the heart of molecular quantum mechanics and, in particular, at the heart of electronic structure theory. Configuration interaction (CI) theory provides an exact description of strong correlation, but it suffers from exponential scaling with respect to the number of correlated electrons and orbitals. As an alternative, variational two-electron RDM (v2RDM) methods have been introduced since the energy of a many-electron system can be formulated exactly using the two-electron RDMs (2RDMs). One interesting property is that the 2RDM can be formulated without explicit knowledge of the wave function. In practice, finding a wave function that maps explicitly to the 2RDM can be very tricky, and the resulting deviation between CI- and RDM-based methods can be very large.

To resolve this issue, a collaboration between the Vogiatzis and DePrince groups lead to the development of the data-driven v2RDM (DDv2RDM) method to learn CI-quality energies using data generated using the v2RDM-complete active space self-consistent field (CASSCF) method. Using proof-of-principle calculations, they found that the model learns the correction the v2RDM energy near-chemical accuracy (1 kcal/mol). They also introduced the use of SHapley Additive exPlanation (SHAP) values, a feature importance method based on cooperative game theory, to analyze the how their physics-based features affect model performance. The SHAP analysis confirmed that the features that impact the model performance the most (and least) correspond well to insights based on physical principles.

Read the full article here.

Smith diagram

Smith Breaks New Ground with Domain Wall Research

by

Kevin Smith, recent Ph.D. graduate from the department of chemistry, and Professor of Chemistry Janice Musfeldt have published the results of a collaborative investigation into the properties of ferroelectric domain walls. This research has generated a greater understanding of both a specific material, and domain walls in general, expanding the foundational knowledge critical to effectively using domain walls in future technologies.

Smith joined the chemistry department as a graduate student in 2015 and very quickly began investigating domain walls. Domain walls act as the boundaries between regions, or domains, of materials and have the potential to impact the properties and uses of that material.

Smith’s work specifically investigates the domain walls of ferroelectric materials, which have been a source of interest in the development of electronics. Efforts have been made to use domain walls as functional parts of devices as they could offer high speed memory reading and writing while requiring less energy to function.

Before ferroelectric domain walls can be successfully leveraged, researchers must develop a fundamental understanding of them and how they behave. It has long been hypothesized that these domain walls are atomically thin and conductive, but this had never been confirmed with a direct measurement at the wall. Smith and Musfeldt began investigating ferroelectric domain walls not with the intention of addressing this long-held belief, but with the goal of uncovering foundational information that could contribute to a greater understanding of these materials.

A collaboration with a group of physicists at Rutgers university, led by Henry Rutgers Professor Sang-Wook Cheong, provided Smith the material with which to begin his exploration.

“Our synthetic collaborators at Rutgers grew the material for us and provided some basic mapping on where to look for the domain walls,” said Smith. “We performed a line scan of the material with the near-field infrared microscope at Beamline 2.4 of the Advanced Light Source, or ALS, at Lawrence Berkeley National Lab. That’s when we started seeing these differences that we weren’t expecting to see.”

When thinking of a solid object, the expectation is often that the object is fairly uniform and that the components creating it are evenly distributed throughout that object. However, with the material Smith was investigating, the scan’s results were pointing toward different organizations of the material’s component parts in different regions of the material.

Smith and Musfeldt knew if they were going to uncover the source of these differences, they were going to need to investigate the material further, using the high-resolution infrared technique at the ALS to scan the material more thoroughly.

Beamline 2.4 of the ALS couples an atomic force microscope with synchrotron-generated infrared light to perform nanospectroscopy to examine materials on a much smaller scale than traditional microscopes. The microscope uses extremely sharply focused light delivered to an object at a very close distance. The response of the light as it interacts with the object is then collected and used to determine what is happening in that object.

“Using the ALS allowed us to examine these differences we were seeing in much greater detail. The material that we were studying was grown in such a way that it had two different types of metals in its A-site, scandium and lutetium. The ALS let us tease out three compositional arrangements for these materials that explained the differences. We found regions that were fairly evenly distributed, as well as both scandium-rich and lutetium-rich regions,” said Smith.

In addition to explaining the differences in domains with slightly different local composition, Smith and Musfeldt were able to determine the domain walls themselves were, in fact, much wider than traditionally believed. They also concluded that while they may have different conductivity than the surrounding regions, the domain walls were not metallic.

By successfully imaging ferroelectric domain walls, Smith and Musfeldt have accomplished something that has never been done before. As a result, they not only created a deeper understanding of these domain walls in a specific material, but also upended long-held beliefs about domain walls in general, paving the way for future innovation. Their work further highlights the importance of foundational and exploratory research in the development of future breakthroughs.

“This project really highlights the importance of curiosity in research,” said Musfeldt. “Kevin took an exploratory project and turned it into the most exciting thing in our lab with far-reaching implications.”

New materials are one potential path to improving existing technologies and generating new means of meeting the modern needs of people and society. Materials, however, are only useful insofar as they can be understood. Smith and Musfeldt’s work digs into the fundamental science behind a material’s properties, simultaneously creating a better understanding of that material and creating a roadmap for more effective uses for it in the future.

The full publication describing this research can be read here.

Vogiatzis named Bodossaki Distinguished Young Scientist

by

Konstantinos Vogiatzis, associate professor in the chemistry department, has been named a Bodossaki Distinguished Young Scientist Award winner. The award recognizes young Greek scientists for their work in a number of academic fields, including science, life sciences, applied science and technology, and the social sciences.

Vogiatzis’ work is centered on the development of computational methods based on electronic structure theory and artificial intelligence. He and his team apply this to chemical systems for clean, green technology.

“As an independent researcher, my work has focused on leveraging machine learning in computational chemistry, using modeling and simulation for the discovery of novel molecules and materials with enhanced properties,” said Vogiatzis. “The guiding objective of my research is to clarify the fundamental physical principles influencing the properties of molecules and materials through the interpretation of experimental data.”

Since 1993, the Bodossaki Foundation has distributed Distinguished Young Scientist Awards every two years. In that time, 57 Greek scientists have been recognized for outstanding research conducted across a global stage. Candidates for the Bodossaki Distinguished Young Scientist Award are nominated by peers, collaborators, and institutions in which they work. Vogiatzis was nominated by Vanda Glezakou, a colleague at Oak Ridge National Laboratory and fellow native of Greece.

Vogiatzis will attend a ceremony in Greece this summer where he will be presented with his award.As a Bodossaki honoree, Vogiatzis joins the ranks of Greek professors working at leading research institutions around the world, including Harvard University, the University of Oxford, and the University of Toronto.

“I would like to express my gratitude to the Bodossaki Foundation, both for recognizing my work and for the honor of being included among the outstanding scientists receiving these awards now and in years past,” said Vogiatzis. “This award is the result of a 17-year course of scientific study that began in the classrooms and research laboratories of Greek universities. This, however, is just the beginning and I look forward to many more years continuing the search for new discoveries in the field of chemistry.”

Vogiatzis joined the University of Tennessee, Knoxville in 2016. Since that time, he has authored more than 40 publications and mentored 15 graduate students. He is the recipient of the 2020 and 2022 Ffrancon Williams Endowed Faculty Award in Chemistry, the 2021 OpenEye Outstanding Junior Faculty Award presented by the American Chemical Society, and a 2021 NSF CAREER award.

Read more about the Bodossaki Foundation and the 2023 Distinguished Young Scientist awardees here.

 

Vogiatzis Group Publishes in npj Computational Materials

by

Associate Professor of Chemistry Konstantinos Vogiatzis, in collaboration with Professor of Mathematics Vasileios Maroulas and Eastman Chemical Company, has published a new machine learning model for predicting the properties of new polymeric materials.

Polymers are everywhere. From cookware to medical devices, polymers have become important to modern life due in part to a growing list of potential uses, and desirable properties like high durability and resistance to corrosion.

Creating new polymers can be an expensive, time-consuming process. Because of this, researchers attempt to predict the future properties of polymers using a variety of tools. Computational prediction methods allow researchers to screen polymer combinations for the desired properties before beginning experimentation. However, finding ways to represent polymers as machine-readable inputs can be difficult, creating a challenge for developing accurate prediction models.

Vogiatzis’ team is attempting to tackle these challenges by creating a deep learning method to predict polymer properties called PolymerGNN. PolymerGNN relies on state-of-the-art graph neural networks (GNN) and machine learning to predict the properties of new polymers using a database of complex polyesters.

“Polyesters offer a diverse material space formed by considering many different types of multifunctional acids and glycols, which are the building blocks of these materials,” said Vogiatzis. This, coupled with other complex properties of polyesters, creates a large materials design space Vogiatzis and his team were able to leverage in the development of PolymerGNN.

Vogiatzis worked with Vasileios Maroulas and students Owen Queen, Dr. Gavin McCarver and Sai Thatigotla to develop the general framework and GNN-based machine learning model for PolymerGNN. Collaborators from Eastman Chemical Company synthesized a set of more than 240 polymers and helped compile a database of properties which was used to train PolymerGNN.

Once trained, PolymerGNN accurately predicted both glass transition temperature and intrinsic viscosity. Glass transition temperature is the temperature at which a polymer shifts between a hard state and a softened state. Intrinsic viscosity is a measurement of a polymer’s molecular weight, which can indicate the polymer’s melting point, crystallinity, and tensile strength. These properties are fundamental to the ultimate physical traits of a given polymer and are critical to the development of adhesives, plastics, and more.

Vogiatzis’ team recently published this work in npj Computational Materials, an open access journal from Nature Research. They have also released PolymerGNN as an open-source codebase. Vogiatzis and Maroulas have collaborated on previous machine learning projects published by the American Chemical Society and Nature Communications. Read the most recent publication here.

Kevin Smith Featured by Berkeley Lab

by

Graduate student Kevin Smith and Professor Janice Musfeldt were recently featured by Lawrence Berkeley National Laboratory for their work with the Advanced Light Source (ALS). The highlight described the work in their paper entitled “Real-Space Infrared Spectroscopy of Ferroelectric Domain Walls in Multiferroic h-(Lu,Sc)FeO3” published in ACS Applied Matter Interfaces.

Smith and Musfeldt used infrared light from the ALS to investigate the properties of the domain walls that separate electrically polarized regions in a rare-earth ferrite material. Their findings open the door to broadband imaging of physical and chemical heterogeneity in ferroics, and improved understandings of the properties of flexible defect states. The complete highlight is available here. 

Chemistry Students Named Volunteers of Distinction

by

Two undergraduate chemistry students are included in the 2023 Volunteer of Distinction Award winners. Drake Robins and Clay West were nominated by faculty members and joined the ranks of students from across the university being honored.

Drake Robins is a fourth-year senior studying analytical chemistry. He is a member of the Air Force ROTC and has been working in Associate Professor Bhavya Sharma’s lab since his junior year. After graduation, Robins will join the United States Air Force and attend Undergraduate Pilot Training. Robins expressed his gratitude for the award and his time at the University of Tennessee.

“Academics and research have always been a top priority for me throughout my time at UT, and I feel extremely blessed to be recognized for it this close to graduation,” said Robins.

Clay West, also a fourth-year senior, is a student in the department’s American Chemical Society certified bachelor’s degree program. He plans to spend the year after graduation applying to graduate schools and preparing to pursue a PhD in organic chemistry. West stated he was grateful to receive the Volunteer of Distinction Award and considers it to be a reflection of the work he has put into earning his degree.

The Volunteer of Distinction Awards were created in 2021 by the university to recognize students across campus who exhibit extraordinary academic achievement, professional promise, or excellence in research. Previous award winners from the chemistry department include Maggie Eslinger, Hannah Hagewood, Elijah Hix, Galvin McCarver, and Wilson Wang.

Jones Wins NVIDIA GPU Poster Award

by

Grier Jones, fifth year chemistry PhD student, recently won a poster competition at the spring meeting of the American Chemical Society (ACS). His poster, entitled “Exploring the topology of electronic correlation with graph neural networks” earned the NVIDIA GPU Award for Best GPU Poster. The award targets excellent computational chemistry research using a graphical processing unit (GPU).

GPUs are most often associated with the high-quality images seen on gaming computers. However, the highly parallelized architecture of GPUs offers an acceleration platform that can outperform central processing units (CPUs) when processing large amounts of data in parallel. This has implications for scientific computing and machine learning applications, which have traditionally used CPUs.

Jones has developed a novel computational model that incorporates GPUs with graph neural networks (GNNs) and topological data analysis (TDA) to explore the topology of electron correlation. By incorporating two central motifs of the machine learning projects in the Vogiatzis lab, Data-Driven Quantum Chemistry (DDQC) and the application of persistent homology this study provides new perspectives on both the topological nature of electron correlation and the data-driven algorithms used to capture electron correlation.

For the purposes of this study, GPUs provided by the Infrastructure for Scientific Applications and Advanced Computing (ISAAC) cluster at the University of Tennessee were used. Training machine learning models on GPUs allows for the exploration of large datasets by reducing the computational time required to train the models. As a second step, persistent homology was used to characterize the transferability in the machine learning models between system size.

Jones expressed his gratitude to the Graduate Student Senate Travel Award and the Vogiatzis’ NSF-CAREER award for providing financial support for his participation in the ACS Spring 2023 National Meeting in Indianapolis. The award provides a professional workstation-level NVIDIA GPU, which Grier is excited to incorporate into his current and future projects.

The NVIDIA GPU Award for Best GPU Poster is a competitive biannual award sponsored by NVIDIA and the American Chemical Society’s Division of Computers in Chemistry.

chemistry data visualized

Jenkins Lab Published in Angewandte Chemie

by

The Jenkins Lab published their research “Giving Gold Wings: Ultrabright and Fragmentation Free Mass Spectrometry Reporters for Barcoding, Bioconjugation Monitoring, and Data Storage” in the international journal Angewandte Chemie. Graduate students Isabel Jensen and Gurkiran Kaur were co-authors on the piece. 

The widespread application of laser desorption/ionization mass spectrometry (LDI-MS) highlights the need for a bright and multiplexable labeling platform. While ligand-capped Au nanoparticles (AuNPs) have emerged as a promising LDI-MS contrast agent, the predominant thiol ligands suffer from low ion yields and extensive fragmentation.

In this work, they developed a N-heterocyclic carbene (NHC) ligand platform that enhances AuNP LDI-MS performance. NHC scaffolds are tuned to generate barcoded AuNPs which, when benchmarked against thiol-AuNPs, are bright mass tags and form unfragmented ions in high yield. To illustrate the transformative potential of NHC ligands, the mass tags were employed in three orthogonal applications: monitoring a bioconjugation reaction, performing multiplexed imaging, and storing and reading encoded information.

These results demonstrate that NHC-nanoparticle systems are an ideal platform for LDI-MS and greatly broaden the scope of nanoparticle contrast agents.

chemistry data visualized

Jenkins Group Published in Chemical Science

by

The Jenkins Lab published their research “Statistical copolymer metal organic nanotubes” in the journal Chemical Science. Graduate student Jacob Barrett co-authored the publication.

Metal–organic nanotubes (MONTs) are 1-dimensional crystalline porous materials that are formed from ligands and metals in a manner identical to more typical 3-dimensional metal–organic frameworks (MOFs). MONTs form anisotropically in one dimension making them excellent candidates for linker engineering for control of chemical composition and spacing.

A novel series of MONTs was synthesized utilizing a mixture of 1,2,4-ditriazole ligands containing both a fully protonated aryl moiety and its tetrafluorinated analog in ratios of, 0 : 1, 1 : 4, 1 : 1, 4 : 1, and 1 : 0, respectively. All MONTs were characterized by both bulk and nanoscale measurements, including SCXRD, PXRD, ssNMR and TEM, to determine the resulting co-polymer architecture (alternating, block, or statistical) and the ligand ratios in the solid materials.

All characterization methods point towards statistical copolymerization of the materials in a manner analogous to 3D MOFs, all of which notably could be achieved without destructive analytical methods.

Vogiatzis Publishes in Inorganic Chemistry Frontiers

by

The Vogiatzis group recently published a paper in Inorganic Chemistry Frontiers entitled “Data-driven ligand field exploration of Fe(iv)–oxo sites for C–H activation.”

Methane is the main component in natural gas and is expected to become more and more important to the development of fuels and chemicals for applications such as clean energy, light and heat production, and the development of organic chemicals. However, methane’s instability and flammability make storage and transportation difficult. It is possible to improve methane’s stability by converting it into methanol or light hydrocarbons.

One approach to this is the development of new catalysts that mimic naturally existing enzymes. The Vogiatzis group, led by Associate Professor Konstantinos Vogiatzis, focused their research on non-heme Fe(IV)-oxo model complexes.

“Computational studies provide a fundamental understanding of the electronic effects that control the reactivity of the Fe(IV)-oxo species, but also provide directions for the synthesis of the next generation of catalytic complexes and materials,” said Vogiatzis.

Vogiatzis and his team employed machine learning to more quickly and thoroughly investigate possible complexes that may be most effective. They developed machine learning models that use a novel molecular representation based on persistence homology, called persistence images.

“Our methodology uses a novel molecular fingerprinting method based on persistent homology, an applied branch of topology, that can encode the geometric and electronic structure together with molecular topology,” said Vogiatzis. “The new model is trained on accurate data from a few hundred Fe(IV)-oxo complexes and is capable of providing reliable information for thousands of complexes.”

Vogiatzis believes the insights uncovered in this research will aid in the construction of a theoretical framework for the design of novel catalysts for less energetically demanding industrial processes, including the conversion of methane and natural gas. This publication was co-authored by graduate students Grier Jones, Brett Smith, and Justin Kirkland, members of the Vogiatzis research group.