BNL@Yale Day Agenda

Location: Kline Tower, 14th Floor, Lounge

Welcome Address - Michael Crair, PhD, Vice Provost for Research, Yale University

Introduction to Brookhaven National Laboratory - James Misewich, Associate Laboratory Director, Energy and Photon Sciences Directorate, Brookhaven National Lab

Location: Kline Tower, 14th Floor, Lounge

Remarks on Research and Collaboration - Maurie McInnis, PhD, President, Yale University

Location: Kline Tower, 14th Floor, Lounge

11 - 11:10 am - Stacy Malaker, Associate Professor of Chemistry

• Title: Mucinomics as the next frontier of mass spectrometry 
• Abstract: Mucin-domain glycoproteins are densely O-glycosylated and play key roles in a host of biological functions. However, their dense O-glycosylation remains enigmatic both in glycoproteomic landscape and structural dynamics, primarily due to the challenges associated with studying mucin domains. Here, we present advances in the mass spectrometric analysis of mucins, including the characterization of mucinases, enrichment techniques, and complete mucinomic mapping of translationally relevant mucin proteins.

11:10 - 11:20 am - Fang Lu, Staff Scientist, Brookhaven National Lab Center for Functional Nanomaterials

• Title: Uncover Structure Complexity and Formation Mechanisms of Nanoparticle Synthesis and Self-Assembly through Coupled X-ray and Electron Microscopy Characterizations 
• Abstract: Advances in multicomponent nanomaterials depend on the ability to decode hidden structural complexity, which link nanoparticle growth kinetics with the symmetry of their assembled architectures. In this presentation, I describe how an integrated toolkit of X-ray scattering, 3D X-ray imaging, and electron microscopy tomography can reveal the mechanisms and pathways underlying shape-programmed synthesis and symmetry-guided superlattice formation. Two examples are highlighted: (i) the emergence of a low-density octo-diamond crystal assembled from Au tetrahedra—an overall achiral lattice composed of alternating chiral bilayers stabilized by particle–substrate interactions; and (ii) a room-temperature nanoparticle reshaping strategy that induces facet-selective atomic relocation at constant volume, enabling controlled facet exposure and associated functional tuning. By combining these complementary modalities, we resolve internal orientation fields, structural transformations, and surface evolution that would be inaccessible to any single technique. Overall, the focus is on mechanistic understanding and volumetric reconstruction to support predictive control of hybrid nanomaterial systems for functional photonics and catalytic architectures.

11:20 - 11:30 am - Juan Jimenez, Assistant Chemist, Brookhaven National Lab

• Title: From Methane to Methanol: Pd-iC-CeO₂ Catalysts Engineered for High Selectivity via Mechanochemical Synthesis
• Abstract: In the pursuit of selective conversion of methane directly to methanol in the liquid-phase, a common challenge is the concurrent formation of undesirable liquid oxygenates or combustion byproducts. However, we demonstrate that monometallic Pd-CeO₂ catalysts, modified by carbon, created by a simple mechanochemical synthesis method exhibit 100% selectivity toward methanol at 75°C, using hydrogen peroxide as oxidizing agent. The solvent free synthesis yields a distinctive Pd-iC-CeO₂ interface, where interfacial carbon (iC) modulates metaloxide interactions and facilitates tandem methane activation and peroxide decomposition, thus resulting in an exclusive methanol selectivity of 100% with a yield of 117 μmol/gcat at 75 °C. Notably, solvent interactions of H₂O₂ (aq) were found to be critical for methanol selectivity through a density functional theory (DFT)-simulated Eley−Rideal-like mechanism. This mechanism uniquely enables the direct conversion of methane into methanol via a solid−liquid−gas process.

11:30 - 11:40 am - Udo Schwarz, Professor of Mechanical Engineering

• Title: Quantitative Single-molecule Spatial Site Reactivity Analysis by Atomic-resolution Chemical Interaction Microscopy
• Abstract: The study of molecular interactions at surfaces is crucial for advancing catalytic technologies and addressing numerous challenges in materials science. Such interactions underpin vital processes such as CO2 reduction and ammonia synthesis but can also lead to undesirable phenomena like surface degradation. Quantitative single-molecule spatial site reactivity analysis has long been limited by experimental artifacts that obscure the intrinsic forces governing surface chemistry. Here we introduce an atomic force microscopy (AFM)-based methodology that enables direct, artifact-free measurement of intermolecular interaction landscapes with three-dimensional, sub-angstrom spatial and piconewton force precision. By combining CO-functionalized tips with a novel post-processing correction framework, we eliminate distortions introduced by the tip geometry and substrate influence—longstanding barriers in quantitative high-resolution scanning probe experiments. We apply this technique to map the interaction potentials between cobalt phthalocyanine (CoPc), a prototypical CO₂ reduction catalyst, and carbon monoxide, revealing site-specific reactivity patterns and their modulation through NH₂ substitution and molecular crowding. This advance transforms noncontact AFM into a chemically specific, quantitative tool for probing interaction forces between complex molecules, with accuracy previously accessible only in theoretical models. Our approach establishes a new standard for chemical force microscopy and opens the door to predictive structure–reactivity studies at the single-site level.

11:40 - 11:50 am - Lu Ma, Physicist, Brookhaven National Lab National Synchrotron Light Source II

• Title: Advancing Operando Materials Characterization through Combined X-ray Absorption and Scattering at QAS
• Abstract: Understanding the dynamic structural and chemical evolution of materials under realistic operating conditions is key to advancing battery and catalytic technologies. The Quick X-ray Absorption and Scattering beamline (QAS, 7-BM) at the National Synchrotron Light Source II (NSLS-II) provides a powerful platform for in situ and operando characterization, enabling real-time studies of functional materials across multiple modalities. QAS integrates high-throughput X-ray absorption spectroscopy (XAS) with complementary X-ray powder diffraction (XRD) and diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), establishing a versatile correlative framework for energy and catalysis research.

11:50 - 11:52 am - Abhishek Kundu, Postdoctoral Scholar, Hazari Group, Yale Department of Chemistry

• Title: Homogeneous Reductant Facilitated Cross-Electrophile Coupling of Aryl Bromides with NHP Esters
• Abstract: Alkyl-substituted N-hydroxyphthalimide (NHP) esters are valuable alternatives to alkyl halides in Ni-catalyzed cross-electrophile coupling reactions because they are typically more stable, can generate alkyl radicals under reductive conditions, and are readily prepared from carboxylic acids. However, the range of aryl halides that can be coupled with NHP esters in XEC has so far been largely limited to aryl iodides. Here, we describe a general method for coupling NHP esters bearing 1°, 2°, and strained ring 3° alkyl groups with aryl bromides. This method is compatible with a broad range of substrates, including drug-like aryl bromides, and operates effectively in various non-amide based solvents. The use of a homogeneous organic reductant is crucial for achieving high yields, as it enables precise control over alkyl radical generation from the NHP ester.

11:52 - 11:54 am - Andressa Vidal Muller, Research Associate, Brookhaven National Lab Artificial Photosynthesis Group

• Title: Reduction of CO to Methanol with Recyclable Organic Hydrides
• Abstract: The selective conversion of carbon monoxide (CO) into energy-dense liquid fuels such as methanol is a central challenge in sustainable chemistry. While CO is a key intermediate in both industrial Fischer-Tropsch processes and emerging carbon-recycling strategies, its controlled reduction under mild conditions remains difficult due to the high stability of the C≡O bond and the accumulation of unproductive intermediates. Molecular approaches offer unique opportunities to disentangle and control individual elementary steps, but many systems remain limited to stoichiometric reactivity. This talk will present recent advances in the reduction of CO to methanol using recyclable organic hydrides in combination with transition-metal catalysts. Benzimidazole-based organic hydrides provide a tunable platform for hydride transfer, enabling stepwise reduction of coordinated CO through metal-formyl and hydroxymethyl intermediates under mild conditions. However, key roadblocks such as slow kinetics, catalyst deactivation, and inefficient hydride turnover have historically prevented the transition from stoichiometric transformations to truly catalytic cycles. We have developed two complementary strategies to overcome these. First, electrochemical hydride regeneration enables the recycling of organic hydride donors, decoupling hydride transfer from sacrificial reagents and providing a direct link between molecular catalysis and renewable electricity. Second, cooperative reactivity with zinc halide additives plays a critical role in stabilizing key intermediates, accelerating hydride transfer, and suppressing deleterious decomposition pathways. Together, these advances create a general design strategy for catalytic CO-to-methanol conversion that integrates molecular catalyst control, recyclable hydride chemistry, and electrochemical regeneration. 

11:54 - 11:56 am - Steven L. Farrell, Goldhaber Fellow, Brookhaven National Lab National Synchrotron Light Source II

• Title: Probing In situ Structure-Activity Relationship in Methane-to-Methanol Catalysts Under High Pressure
• Abstract: Observing the structure and behavior of catalysts under reaction conditions is vital to understanding their function and improving their efficiency. Unlike gas-phase reactions at ambient pressure, which have simple relatively simple setups, probing catalysts in liquid phase reactions or under pressure is difficult. At the National Synchrotron Light Source II’s Inner Shell Spectroscopy beamline, we have recently commissioned a specially designed system for probing heterogeneous catalysts under high pressure in both liquid and gas environments. This system has helped elucidate critical mechanisms in catalyst behavior for applications like converting methane or carbon dioxide into methanol, with capabilities up to 500°C and 34 bar (500 psi). We probe catalysts across a variety of reaction systems, for example molybdenum disulfide (MoS2), a transition metal dichalcogenide using in situ synchrotron X-ray measurements, to study electronic behavior and local structure under pressure and in the presence of strong oxidizing reagents for methane-to-methanol. Using these methods, we observe unique electronic behavior in MoS2 that cannot be detected using ex situ methods, to uncover the origin of methane conversion activity in these catalysts, producing methane derivative oxygenates under mild conditions. By elucidating these mechanisms under bright synchrotron light through robust multimodal X-ray characterization and kinetic studies in real time, we can understand these catalysts work and how to better design and optimize them for peak methane conversion performance.

11:56 - 11:58 am - Justin Wedal, Postdoctoral Scholar, Hazari Group, Yale Department of Chemistry

• Title: Improved Productivity and Stability for Base-Metal Catalyzed CO2 Hydrogenation using Hemilabile Pincer Ligands
• Abstract: Pincer ligands are widely used to support transition metal catalysts, but first-row systems often deactivate too quickly for widespread use. Here, pincer ligands bearing an additional hemilabile donor were designed to stabilize Fe and Mn catalysts for CO2 hydrogenation to formate. Ligands of the type (RCH2CH2)N(CH2CH2PiPr2)2 (iPrPNRP; R = OMe, CH2OMe, NMe2, PiPr2) were synthesized and their associated Fe and Mn hydride complexes were prepared. Their performance in CO2 hydrogenation catalysis was compared to MeN(CH2CH2PiPr2)2-supported complexes, which lack a pendant donor. The pendant donor identity and arm length strongly influenced catalytic activity, but all pendant arm containing systems extended catalyst lifetime. The Fe catalysts achieve similar turnover numbers as the state of the art systems, while notably (iPrPNCH2OMeP)Mn(CO)2H achieved extremely high turnover frequency of 158,000 h-1 and turnover number of 838,000, exceeding most state-of-the-art catalysts including those based on precious metals. Mechanistic studies revealed that hemilabile coordination enhances stability by avoiding deactivation, although excessive binding can lead to inactive states. This work demonstrates hemilabile donors dramatically improve pincer-based catalysts and provides design principles for broader applications.

11:58 am - 12:00 pm - Peter Khalifah, Professor Stony Brook University and Joint Appointment Chemist Brookhaven National Lab Emerging Materials for Energy Technologies Group

• Title: Drinking from the firehose of synchrotron powder diffraction data – interdisciplinary opportunities and new frontiers
• Abstract: While powder diffraction methods are regularly used for the routine characterization of materials, the information content of powder diffraction patterns is far greater than is generally appreciated, especially when using synchrotron diffraction data. Since the time needed to collect high-quality synchrotron data in routine experiments is ~1 second, it is possible to acquire data sets with fine time resolution, with fine spatial resolution, or even to map samples in both space and time. Some examples of our recent work include characterizing vertical composition gradients in high-energy density battery cathodes, measuring lateral heterogeneity in pouch cell batteries to gain mechanistic insights into conversion chemistry reactions and into battery aging, measuring temperature gradients in cm-scale rods for crystal growth within sealed environmental chambers, characterizing reaction kinetics and “hidden” structural transitions that occur during ion exchange, and probing the reaction kinetics and synthesis pathways of nitride catalysts. For many of these systems, precise results from experiments have been used to build and validate predictive models. Additionally, we have been developing a variety of novel powder diffraction methods to characterize key material properties that affect or reflect the performance of functional materials. These include the determination of 3D nanoparticle morphology (including facet-specific surface areas), the absolute quantification of anisotropic strain and strain distributions, the resolution of composition gradients within particles, and the characterization of site defects with a sensitivity of one part in ten thousand. 

Location: Kline Tower, 14th Floor, Seminar Room

11 - 11:10 am - Jun Liu, Professor of Microbial Pathogenesis

11:10 - 11:20 am - Yang Yang, Associate Physicist Brookhaven National Lab National Synchrotron Light Source II

• Title: Quantitative Cellular Tomography (QCT) beamline, the future bio-imaging beamline at NSLS-II, BNL
• Abstract: Cryogenic soft X-ray transmission microscopy (cryo-SXT) is a high-resolution three-dimensional (3D) imaging technique that exploits soft X-ray energies within the so-called “water window” to visualize intact, frozen-hydrated cells in their near-native state. By eliminating the need for sectioning or chemical fixation, cryo-SXT preserves cellular architecture and enables quantitative 3D imaging of whole cells with subcellular spatial resolution. The technique offers high throughput and strong intrinsic contrast between carbon-rich biological structures and the surrounding aqueous environment, making it broadly applicable for studying cellular organization and ultrastructure. Its quantitative nature further enables correlative studies linking structural information to function. At NSLS-II, Brookhaven National Laboratory, we are designing and constructing the Quantitative Cellular Tomography (QCT) beamline to advance cryo-SXT capabilities. QCT will be part of a coordinated suite of NSLS-II instruments dedicated to multimodal and correlative cellular imaging, enabling robust and integrated workflows for biological discovery.

11:20 - 11:30 am - Jing Yan, Assistant Professor of Molecular, Cellular and Developmental Biology

• Title: Illuminating Biofilm Matrix Biochemistry with X-ray Crystallography and Scattering
• Abstract: Biofilms serve as a protective mechanism for many bacteria, including many pathogens. To form such communities, bacteria secrete macromolecules that form an extracellular matrix serving as a barrier against environmental threats, such as predation, antibiotics, and the host immune system. To be effective, a biofilm also must anchor to foreign surfaces, in the environment or in a host. However,  how this matrix self-organizes to support biofilm remain mysterious at molecular level, in most cases.  The human pathogen Vibrio cholerae produces biofilms primarily composed of an exopolysaccharide  called VPS (Vibrio polysaccharide), which consists of an unusually-modified repeating tetrasaccharide core unit. VPS engages with two secreted adhesion proteins, Bap1 and RbmC, which adhere the biofilm to abiotic and biotic surfaces and to strengthen the biofilm by interacting with VPS using a conserved beta propeller. To pinpoint the interaction between the adhesins and purified segments of VPS, we solved the ~1.6 Å X-ray crystal structure of Bap1 bound to fragmented VPS. The structure revealed a single binding site consisting of one tetrasaccharide unit involving an induced magnesium binding site. Unexpectedly, the tetrasaccharide adopted a roughly 90º   bent state caused by a rotation of the glycosidic bond between the central two monosaccharide units. Using a combination of mutagenesis, light scattering, and in situ fluorescent microscopy, we demonstrate that Bap1 not only facilitates biofilm adhesion, but is also required for proper VPS organization, through the identified binding pocket. Our structure reveals for the first time the interaction between a biofilm exopolysaccharide and matrix protein, as well as insights into confirmation changes of exopolysaccharide induced by this binding. Our findings provide a generic approach for studying the biophysical and biochemical properties of biofilm assembly, which may lead to new ways to treat disease caused by biofilm-forming bacterial pathogens by disrupting the exopolysaccharide-protein interactions.

11:30 - 11:40 am - Qun Liu, Distinguished Structural Biologist, Brookhaven National Lab Biology Department

• Title: RESCUE: Rare Earth Self-driving Cellular Uptake Engineering
• Abstract: Rare earths (REs) are vital for U.S. economic and national security, critical for a series of technologies including advanced energy systems, microelectronics, medical applications and defense, yet their domestic supply chain faces significant extraction challenges. The separation of individual REs remains a significant challenge in industrial extraction and refining due to their chemical similarity. The goal of the RESCUE project is the integration of artificial intelligence (AI), laboratory automation, and metabolic and protein engineering to drastically advance and accelerate foundational science for designing high-affinity metal uptake transporters and metabolic pathways for selective uptake and intracellular storage of critical REs from tailings, mine ash, and electronic waste. The long-term goal of RESCUE is to engineer metabolic pathways for enhanced extraction, storage, and uptake systems for critical minerals, contributing to U.S. critical minerals supply-chain resilience. For this pilot, RESCUE has three aims: (1) Develop AI models for protein engineering. (2) Establish an AI-driven automation framework. (3) Advance RE-extraction science via high-throughput characterization.

11:40 - 11:50 am - Ronald Breaker, Sterling Professor of Molecular, Cellular and Developmental Biology

• Title: Human Riboswitch Discovery and Analysis
• Abstract: RNA structures called riboswitches are known to selectively sense metabolites, toxic compounds, or elemental ions to regulate diverse biological processes. These molecular sensors and switches have primarily been found in bacteria, where they sense fundamental ligands likely to have been relevant to ancient organisms from the RNA World. The first riboswitch candidates have now been discovered in vertebrates, including humans, where they associate with mRNAs for genes relevant to major neurological functions. Key challenges regarding riboswitch discovery and analysis will be presented.

11:50 - 11:52 am - Liguo Wang, Director of Scientific Operations, Brookhaven National Lab Laboratory for BioMolecular Structure (LBMS), National Synchrotron Light Source II

• Title: Visualizing Biological Systems at the Molecular and Cellular Level at the Laboratory for BioMolecular Structure
• Abstract: Cryo-Electron Microscopy (cryo-EM) is a powerful technique for visualizing biological specimens and understanding molecular and cellular processes. Through the Laboratory for BioMolecular Structure (LBMS), Brookhaven National Laboratory provides peer-reviewed access, support, and training for cryo-EM. In FY2025, 1,076,806 images were collected on the Krios, leading to 17 publications (62.5% in high-impact journals) and 90 structures deposited in the EMDB. LBMS offers three levels of training, including an annual virtual course, semi-annual in-person workshops, and on-demand or remote training. Workshops are highly rated, with an average score of 4.85/5 and universal participant recommendation. In addition, LBMS operates a cryo-Electron Tomography (cryo-ET) user program supporting DOE-funded projects, with access to advanced sample preparation tools such as EM ICE and the Aquilos 2 cryo-FIB, enabling studies from molecular to cellular and tissue scales.

11:52 - 11:54 am Binyam Mogessie, Assistant Professor of Molecular, Cellular, and Developmental Biology 

• Title: Engineering Egg Aging to Mitigate Infertility
• Abstract: Female meiosis presents a significant challenge to cellular integrity. Mammalian oocytes stay arrested from before birth for years to decades, then perform chromosome segregation using a non-canonical spindle that is unusually large, lacks centrosomes, and is mechanically fragile. Failures in this process can lead to aneuploidy, infertility, and age-related reproductive decline, yet the cellular mechanisms that maintain spindle function over long periods remain poorly understood. My laboratory discovered and studies a previously underappreciated cytoskeletal system, spindle-associated actin, which operates within the meiotic spindle alongside microtubules. We find that this actin network contributes to spindle force balance and chromosome segregation fidelity, challenging the traditional view of the spindle as a purely microtubule-based machine. Disruption of this system with age sensitizes oocytes to segregation errors, suggesting that age-related aneuploidy reflects a failure of a broader cytoskeletal integrity program. To directly investigate how aging destabilizes this system, we have created a synthetic oocyte aging platform that allows controlled induction of aging-related molecular defects in young oocytes while maintaining a physiological context. This method enables us to separate chronological age from specific cellular failure states and identify the precise thresholds at which spindle structure and chromosome segregation break down. Addressing these questions requires multimodal, in situ structural and live-cell imaging, quantitative analysis of cytoskeletal dynamics across scales, and integrative physical and data-driven modeling. This approach positions national laboratory infrastructure and AI-enabled analysis as crucial tools for discovering previously inaccessible mesoscale cellular structures and understanding their failure with aging. Together, this work demonstrates how declining cellular integrity in eggs reveals key mechanisms of aging and fertility, defines biological limits affecting reproductive health across populations, and establishes the oocyte as a powerful model for studying structure-function relationships in complex cytoskeletal networks.

11:54 - 11:56 AM, Fred Sigworth, Professor Emeritus of Cellular and Molecular Physiology

• Title: CBMS access, training and outreach: Opportunities for training and Collaborations  
• Abstract: The Center for Biomolecular Structure (CBMS) supports a broad spectrum of scientific research, spanning structural biology to multimodal imaging of environmental and biologically relevant samples. Its mission is to provide researchers with state-of-the-art instrumentation, advanced analytical capabilities, and collaborative platforms to address major scientific challenges of the coming decade. The User, Training, and Outreach Core (UTOC) strengthens the missions of the National Institutes of Health and the Office of Biological and Environmental Research by fostering user engagement, education, and community-building initiatives. The CBMS training program emphasizes hands-on workshops and virtual tutorials focused on data analysis software and experimental methodologies. These efforts are complemented by beamline-specific manuals, comprehensive documentation, and direct guidance during beam time, ensuring that users are well supported from proposal development through data interpretation. CBMS staff also collaborate closely with Brookhaven National Laboratory Workforce Development and Science Education to mentor high school, undergraduate, graduate, and postdoctoral trainees. At the undergraduate level, students participate in U.S. Department of Energy programs such as the DOE Science Undergraduate Laboratory Internships, which provide immersive research experiences within the national laboratory environment. These internships introduce students to advanced instrumentation, interdisciplinary collaboration, and career pathways in science and engineering. For graduate students, the DOE Office of Science Graduate Student Research Program supports those who wish to apply one or more specialized techniques available at national laboratories to strengthen and expand research developed at their home institutions. This program enables doctoral candidates to complement their thesis work by accessing unique facilities, expertise, and methodologies that enhance the depth and impact of their graduate studies.

11:56 - 11:58 am Martin Fuchs, Beamline Scientist, Brookhaven National Lab National Synchrotron Light Source II

• Title: Dynamic macromolecular crystallography at NSLS-II
• Abstract: Protein dynamics underlie biological function - from ligand recognition and allostery to catalysis and transport. Because conformational landscapes can differ substantially between cryogenic and ambient conditions, room-temperature (RT) crystallography is increasingly used to reveal functionally relevant alternative states and transient binding sites that are often obscured by cryocooling [Ebrahim 2022; Fischer 2015]. Automated multi-crystal and serial micro-crystallography at the Frontier Macromolecular Crystallography beamline (FMX) at NSLS-II now make such measurements routine even for micron-sized crystals. Multi-temperature datasets are still scarce; fewer than 3% of PDB structures were collected at room temperature, and sufficient resolution [Fischer 2015]. A new time-resolved crystallography capability at FMX enables collection of structural snapshots of dynamic processes, for example of enzyme reaction intermediates along catalytic pathways. Importantly, serial crystallography and chemical triggering broaden the applicability of time-resolved crystallography beyond photoactivatable systems [Mehrabi 2019], enabling access to millisecond-to-minute dynamics of ligand binding, turnover, and enzymatic intermediate formation in a wide range of biomolecular reactions. We will present RT serial and time-resolved capabilities and first experiments at FMX and outline a beamline design for the NSLS-II storage-ring upgrade, NSLS-IIU, to exploit an approximately hundredfold increase in brightness. The planned beamline, DynMX, will resolve reaction time scales down to microseconds, covering the complete range of enzymatic reactions, and reaction initiation methods for all reaction types.

Lunch will have limited registration and will be first come first serve so register soon!

Enjoy lunch by Yale catering and network with Yale faculty & students and with representatives from Brookhaven National Lab.

Location: Peabody Museum Seminar Room 118

1:30 - 1:40 pm - Tianyu Zhu, Assistant Professor of Chemistry

• Title: Beyond conventional DFT: New electronic structure tools for surface chemistry
• Abstract: Understanding surface chemistry at metallic interfaces requires an accurate description of electronic correlation that often lies beyond the reach of conventional density functional theory (DFT). In this talk, I will describe a computational framework that advances beyond DFT by combining quantum embedding methods with data-driven machine learning (ML) models. On the accuracy side, quantum embedding approaches enable correlated quantum chemistry treatments of catalytically active sites embedded in extended metallic environments, providing direct access to adsorption energies, local densities of states, and spectroscopic observables. On the efficiency side, I will discuss Hamiltonian ML techniques that learn effective electronic structure models from first-principles calculations, enabling rapid prediction of spectral descriptors that inform molecular adsorption mechanisms. These developments will be illustrated using single-atom alloy surfaces as a representative class of catalytic systems, highlighting how embedding and ML can be integrated to achieve predictive and efficient simulations of surface chemistry.

1:40 - 1:50 pm - Justin Goodrich, Assistant Physicist, Brookhaven National Lab National Synchrotron Light Source II

• Title: Development of Quantum-Enhanced X-ray Microscopy at NSLS-II
• Abstract: Quantum imaging leverages photon–photon correlations to extract information beyond the limits imposed by classical shot noise. While such techniques are well established in the optical regime, their extension to hard X-ray energies has remained experimentally challenging due to extremely low nonlinear conversion efficiencies and the absence of suitable fast, energy-resolving detectors. At the Coherent Hard X-ray Scattering (CHX) beamline of NSLS-II, we are developing a platform for quantum-enhanced X-ray microscopy based on spontaneous parametric down-conversion in single-crystal diamond. We recently demonstrated X-ray quantum correlation imaging using a pixelated, time- and energy-resolving detector, achieving coincidence rates approaching 7.8 × 10³ photon pairs per hour, substantially exceeding previous synchrotron benchmarks. Using these correlated photon pairs, we performed twin imaging of structured test objects and a biological specimen, establishing coincidence-based X-ray imaging of a biological sample. Correlation-based detection intrinsically suppresses uncorrelated detector noise and provides a pathway toward sub-shot-noise transmission measurements in the photon-sparse regime. Our long-term objective, supported by the DOE Biological and Environmental program, is to enable low-dose X-ray microscopy for radiation-sensitive biological materials. Current efforts focus on improving photon-pair identification fidelity through enhanced detector energy resolution and background suppression, as well as engineering phase-matching conditions to optimize flux and spatial correlations. These developments lay the groundwork for future quantum-enhanced transmission imaging and phase-sensitive diffraction schemes that could extend X-ray imaging mechanisms beyond classical limits.

1:50 - 2:00 pm - Byung-Jun Yoon, Scientist, Brookhaven National Laboratory Computational Science Initiative

• Title: AI for Molecular Design Under Uncertainty
• Abstract: Molecular design and discovery have traditionally been labor-intensive and time-consuming endeavors. Although the development of quantitative structure–activity relationship (QSAR) models has significantly accelerated drug discovery by enabling predictive modeling of molecular properties, direct optimization within the vast and high-dimensional molecular/chemical space remains a formidable challenge. High-throughput virtual screening (HTVS) has proven to be a valuable tool, but its application to large-scale molecular libraries often incurs prohibitive computational costs. Furthermore, conventional HTVS pipelines frequently rely on expert-driven heuristics, which can limit their overall efficiency and predictive accuracy. In this talk, we will discuss how AI-driven strategies can be harnessed to address these challenges, enabling smarter design and faster discovery. We will explore recent advances in generative AI models for multi-objective molecular design and examine how AI-based methods can enhance the design and execution of HTVS campaigns. Special emphasis will be placed on the role of uncertainty-aware modeling, which facilitates robust and data-efficient decision-making in the presence of complex and noisy molecular landscapes. By integrating techniques such as uncertainty quantification, active learning, and optimal experimental design, these AI-driven strategies not only accelerate discovery but also improve the reliability of outcomes in real-world settings that are often resource-constrained and data-scarce.

2:00 - 2:10 pm - Shray Mathur, Scientific Associate, Brookhaven National Laboratory Center for Functional Nanomaterials (CFN)

• Title: AI-Powered Virtual Companions for Natural Human-Instrument Interaction at Synchrotron Beamlines
• Abstract: Scientific user facilities like synchrotron beamlines come with complex instrumentation and codebases that create a steep knowledge gap between researchers and the tools they need. Generative AI offers a way to bridge that gap — letting scientists communicate with their instruments naturally instead of through specialized software. We present VISION [1], a Virtual Scientific Companion that enables natural language interaction at synchrotron beamlines. Researchers can speak or type commands like “measure the sample for five seconds” or “align the sample,” and VISION translates these into executable code. Beyond instrument control, VISION handles data analysis, logs experimental events in a personalized notebook, and answers questions from beamline manuals and nanoscience literature — acting as an always-available companion throughout an experiment. The system assembles multiple AI-enabled cognitive blocks, each scaffolding foundation / large language models for a specialized task: speech recognition, command classification, code generation, data analysis, and scientific Q&A with retrieval-augmented generation. A key design choice is a dynamic prompting system where beamline-specific functions are stored in a simple JSON configuration file, allowing scientists to teach the assistant new capabilities or adapt it to entirely different beamlines without modifying the underlying code. Using VISION, we demonstrated the first voice-controlled experiment at an X-ray scattering beamline at the National Synchrotron Light Source II (NSLS-II) at Brookhaven National Laboratory, achieving low-latency LLM-based beamline operation. This work on natural language-based scientific experimentation is a building block toward a science exocortex [2] — a synthetic extension to the cognition of scientists that could transform how research is conducted at user facilities.

2:10 - 2:20 pm - Nikhil Malvankar, Associate Professor of Molecular Biophysics and Biochemistry

• Title: Single-Cell Imaging and Control of Microbial Metabolism and Colonization by Targeting Electron Transfer via Protein Nanowires
• Abstract: Every living cell needs to eliminate the surplus of electrons created by their metabolic processes. Most organisms achieve this by transferring electrons to soluble oxygen-like electron acceptors, which act as electron sinks. However, microbes that live in areas with limited or no oxygen, such as those residing in the deep ocean, in soil, or in the human body, have evolved strategies to export electrons to extracellular acceptors, including minerals or other bacteria. Geobacter uses long, thin, conductive filaments called “nanowires” to export electrons1,2. Nanowires are fundamental to global environmental processes1,2, including methane degradation, a major greenhouse gas3.
  Geobacter nanowires have fascinated the scientific community since their discovery in 2002. Until recently, nanowires were considered Type IV pili (T4P), polymers of the PilA-N pilin subunit, partly because T4P are required for electron transfer4. However, my lab showed that PilA-N pairs with a second protein, PilA-C, to form a T4P that are structurally inconsistent with electron transfer4. We further demonstrated that Geobacter produces additional filaments comprising outer membrane cytochrome (Omc) Z and S subunits, which can transfer electrons through a chain of heme groups3,5. I will present how we show that (i) these cytochrome filaments are the electron-conducting nanowires and (ii) the role of T4P in electron transfer is akin to a piston to secrete cytochrome nanowires on the bacterial surface. My team is exploring the structure, assembly, and electron transfer mechanism of nanowires, and evaluating their role in bacterial respiration, communication, and pathogenesis. By combining experimental and computational studies, our team is addressing three key questions: (1) How do microbes kickstart metabolism7 to build & use OmcS6 & OmcZ3  nanowires? (2) How are electrons transferred from the bacterial cytoplasm to surface-displayed nanowires. (3) How to tune nanowire conductivity using light9, pressure10, temperature11, electromagnetic fields10, humidity12, non-natural ‘click’ chemistry13 & coherence14 to control bacterial behavior?

2:20 - 2:30 pm - Chuntian Cao, Assistant Computational Scientist, Brookhaven National Laboratory Computational Science

• Title: Explainable Machine Learning of X-ray Absorption Spectra in Aqueous ZnCl2 Solutions Using Graph Neural Networks 
• Abstract: Machine learning (ML) provides powerful pathways for predicting spectroscopic observables from atomic structures, but its broader impact depends on making model predictions interpretable in terms of physical and chemical principles. Here, we introduce a physics-guided graph neural network (GNN) model that predicts Zn K-edge X-ray spectroscopy (XAS) spectra of aqueous ZnCl2 solutions. Training data are generated from ab initio XAS calculations on molecular dynamics snapshots obtained using a machine learning interatomic potential. The GNN reproduces experimental spectra across concentrations from dilute (<0.1 m) to highly concentrated (30 m, “water-in-salt”) regimes and scales efficiently to large, disordered liquid systems beyond the reach of conventional ab initio approaches. Gradient-based attribution analysis reveals that the model learns physically meaningful structure-spectrum relationships. Ligand-specific attributions reflect orbital hybridization patterns and the origin of the excitations derived from density functional theory. Bond-length attributions recover spectral shifts consistent with the multiple-scattering theory. This work bridges data-driven prediction with electronic-structure theory, establishing a general paradigm for interpretable ML that links atomic structure, electronic structure, and spectroscopic observables.

2:30 - 2:40 pm - Yugang Zhang, Staff Scientist, Brookhaven National Lab Center for Functional Nanomaterials (CFN)

• Title: AI-Guided Autonomous Nanoparticle Synthesis From Optimization to Chemical Insight
• Abstract: Autonomous experimentation platforms that tightly couple AI, high-throughput synthesis, and in situ characterization enable efficient exploration of vast chemical spaces while generating datasets rich enough to reveal underlying chemistry. We present a closed-loop platform that integrates a temperature-controlled droplet-flow microreactor with in situ synchrotron SAXS/WAXS and Bayesian optimization for autonomous colloidal nanoparticle synthesis. Using citrate-reduced gold nanoparticles as a model system, the platform navigated a design space of ~19,000 candidate recipes in ~150 experiments, achieving precise size targeting with narrow size distributions. Beyond optimization, the multi-length-scale scattering data accumulated across autonomous campaigns revealed a robust linear scaling between crystallite domain size and particle diameter, linking reaction conditions to internal structure. To accelerate decision-making in the loop, we deploy uncertainty-aware machine learning for real-time structural analysis, delivering ~1000× speedup over conventional fitting and improving convergence efficiency of the optimization. Finally, we demonstrate generality by extending the approach to shape-controlled Cu₂O nanoparticles. Together, these results show that AI-driven autonomy is not only an optimization engine, but a discovery framework that extracts mechanistic insight from strategically planned, information-rich experiments.

2:40 - 2:50 pm - Lin Yang, LiX Lead Beamline Scientist, Brookhaven National Lab National Synchrotron Light Source II

• Title: Multimodal X-ray imaging for plant science and biotechnology
• Abstract: In the last few years, the LiX beamline at NSLS-II has developed capabilities for scanning imaging and tomography based on both scattering and fluorescence contrasts. This is accomplished using fly-scanning with a typical beam size of 5 microns, and simultaneous data collection on a pair of pixel array detectors for small- and wide-angle X-ray scattering, as well as two two-channel silicon drift detectors located on either side of the sample for X-ray fluorescence. A software pipeline is provided to enable users to extract relevant features from the scattering data as the contrast mechanism for imaging. In addition to characteristic diffraction peaks for well-known materials (e.g. cellulose and starch), these features can also be based on components derived from machine learning algorithms. So far LiX users have published their studies on wood and growing plant stems. More active research is on-going to explore the application of scattering imaging to a broad range of biological tissues (e.g. plant seeds, leaves, roots, insect antenna, seashell), as well as biomaterials being processed. Scanning imaging is time-consuming, limiting the throughput of user experiments and sometimes resulting in obvious radiation damage to the samples. We have therefore implemented a micro-tomography detector for rapid, full-field imaging based on absorption contrast, to guide subsequent X-ray scattering and fluorescence data collection. We are looking for collaboration and science drivers to further develop the imaging method at LiX, including (1) correlative imaging on the same samples using chemical imaging to help interpret the scattering data so that it can be used as a proxy for chemical composition and nanoscale morphology, for instance in time-resolved measurements; (2) tomographic reconstruction based on sparse scanning data and the full-field micro-CT data to speed up scanning data collection.

2:50 - 2:52 pm - Swapnil Chandrakant Devarkar, Associate Research Scientist, Yale Department of Molecular Biophysics and Biochemistry

• Title: Structures reveal dual site targeting of the bacterial 70S ribosome by tetracyclines
• Abstract: The tetracycline class of antibiotics is widely used for treating bacterial diseases including Lyme disease, anthrax, acne vulgaris, and pneumonia. Using a series of high-resolution cryo-electron microscopy (cryo-EM) structures, we show that tetracyclines can simultaneously target the mRNA decoding center in the 30S subunit and the nascent peptide exit tunnel (NPET) in the 50S subunit of the bacterial ribosome. Among the tested tetracyclines, Doxycycline was unique in its ability to dimerize and bind the NPET at multiple locations. Structural comparison of Doxycycline, Minocycline, and Sarecycline bound to the Escherichia coli and Cutibacterium acnes 70S ribosome revealed species-specific differences affecting drug interaction and occupancy. Our results reveal a dual site mechanism of action for tetracyclines and provide a structural basis for rational design of narrow spectrum tetracyclines to overcome the rising threat of antibiotic resistance.

2:52 - 2:54 pm - Akhil Tayal, Beamline Scientist, Brookhaven National Lab National Synchrotron Light Source II

• Title: Operando High Energy-Resolution X-ray Spectroscopy at the NSLS-II Inner Shell Spectroscopy Beamline
• Abstract: Understanding how materials function under real operating conditions is essential for advancing catalysis, energy storage, and environmental technologies. In this talk, I will present how high energy-resolution X-ray spectroscopies (HERXS) provide detailed, element-specific insight into oxidation states, spin states, ligand environments, and metal–ligand interactions beyond what is accessible with conventional X-ray absorption spectroscopy (XAS). I will introduce the capabilities of the Inner Shell Spectroscopy (ISS) beamline at NSLS-II (Brookhaven National Laboratory, Upton, NY), highlighting its unique instrumentation for operando HERXS measurements. I will also discuss recent developments in AI/ML-enabled real-time data analysis that accelerate mechanistic understanding and guide materials design.
   

2:54 - 2:56 pm - Pranav Kantroo, Postdoctoral Scholar, Yale Department of Physics

• Title: Parkour in Protein Morphospace
• Abstract: We present a computational scheme to generate viable paths between homologous protein pairs through stepwise single residue mutations. These paths are composed of intermediate sequences with high fitness as predicted by the protein language model ESM2. To do this we need the means to generate sensible mutations for a given sequence, a computational proxy for fitness, a distance measure between protein pairs, along with a search strategy to navigate through the space. We use the One Fell Swoop (OFS) approach to calculate the mutation profiles of the intermediates, and use them as the proposal distribution to sample sensible candidate mutants. The fitness of the proposals is determined by their OFS pseudo-perplexity, while the proximity between two states is defined by their sequence alignment score as calculated through their ESM2 sequence embeddings. To navigate towards the target, we choose mutants with a high predicted fitness that are closest to the target state over iterative steps. We use this scheme to interpolate between progressively divergent protein pairs, some of which do not even acquire the same structural fold, and document the qualitative variation across the generated paths. The ease of interpolating between two sequences, as quantified by some cost function that depends on the path length and the functional plausibility of the intermediates, could potentially be used as a proxy for the likelihood of homology between them.

2:56 - 2:58 pm - Alexei Tkachenko, Physicist, Brookhaven National Laboratory Center for Functional Nanomaterials (CFN)

• Title: Evolutionary chemical learning in dimerization networks
• Abstract: Biochemical networks already process information inside living cells, but can we train a purely chemical system to perform a machine-learning task, without digital hardware or backpropagation? I will introduce chemical learning in Competitive Dimerization Networks (CDNs): mixtures of many molecular species (e.g., DNA or proteins) that reversibly bind into dimers and collectively compute through mass-action equilibrium. In this framework, each species acts like a “neuron,” while binding affinities and component concentrations play the role of nonnegative, bidirectional “synaptic weights.” I will then describe an experimentally realistic in-vitro training protocol based on directed evolution: mutation, selection, and amplification of DNA components, where variants are scored on batches of noisy “input cocktails” and iteratively enriched. As a proof of concept, we demonstrate multiclass classification with one-hot outputs: after evolutionary training, the CDN produces output fugacities with orders-of-magnitude on/off contrast and high mutual information between inputs and outputs, even under substantial input noise. A key technical point is a contrast-enhancing loss function that improves separability compared to standard MSE. Finally, I will compare evolutionary training with in-silico gradient descent (where applicable) and discuss why CDNs offer a promising route to adaptive, energy-efficient molecular computation for diagnostics, biosensing, and programmable soft matter.

2:58 - 3:00 pm - Gaoyuan Wang, Postdoctoral Associate, Molecular Biophysics and Biochemistry

• Title: Privacy-Preserving Quantum Analysis in Biomedicine
• Abstract: Quantum computing is gaining popularity due to its potential to detect complex patterns in data by leveraging unique quantum phenomena. It is particularly promising for complex data applications, such as those in biomedicine. However, in these settings, achieving sufficient statistical power often requires sharing and aggregating data from multiple participants or institutions. Sharing sensitive information—such as genomic data and treatment histories—raises significant privacy concerns. To address these challenges, we propose a quantum-native method for encoding entire biomedical data cohorts directly into special quantum states, which we refer to as composite states. These states provide a generic encoding suitable for a wide range of downstream analyses while preventing the inference of individual-level information. Quantum computations can be performed directly on composite states without access to the underlying raw data. Building on this approach, we introduce protocols that enable multi-party collaborative quantum neural network training, allowing multiple life science institutions to jointly perform analyses without revealing their private data. We validate both the privacy guarantees and the practical utility of composite states using two genomic studies.

Location: Peabody Museum RM 112

This session is intended for unstructured networking time between Yale faculty and BNL scientists. Some discussions will include access to BNL instrumentation, specific solicitation partnerships, and mutual areas of collaboration.

Location: Peabody Museum RM 118

Panelists will include:

• Andressa Vidal Muller
• Juan Jimenez
• Liguo Wang
• Lin Yang
• Peter Khalifah
• Sean McSweeney
• Yang Yang

Location Peabody Museum

Concluding Remarks - Scott Strobel, Ph.D. Provost, Yale University

Location: Peabody Museum Main Gallery