"From linear to circular: closing the loop Using AI/ML and multiscale modeling", Henry Chan, Department of Chemistry, University of Illinois at Chicago. Host: Subramanian Sankaranarayanan

Advancements in AI/ML algorithms and high-performance computing have empowered the use of data-driven approaches for studying a variety of rich and intriguing phenomena in nanoscience, as well as for accelerating the design and discovery of materials. The flexibility of AI/ML data models coupled with the predictive power of physics-based simulations pave the way for closing the loop on autonomous experimentation and possibly nanoscience discoveries that can help close the loop on circular economy. This talk summarizes my research experience on materials modeling, imaging, and AI/ML method development, which is focused on addressing open issues and challenges in big data, autonomous experimentation, and inverse design of materials for a circular economy. Finally, my talk will touch upon the future of scientific software development, user-friendly AI/ML tools, data management and their potential impact on the CNM user program.

"Understanding simple chemistries in complex environments via molecular simulations and machine learning", Mirza Galib, Department of Mechanical Engineering, University of Louisville. Host: Subramanian Sankaranarayanan

Solvation, charge transport and reactions in complex environments play a vital role in many atmospheric, material, and chemical processes. In addition to experiments, computational tools can significantly contribute to our fundamental understanding of these phenomena. For example, first-principles molecular simulations based on density functional theory (DFT) have significantly contributed to uncover molecular level picture of many complex chemical phenomena and new material design. However, the use of DFT (let alone more accurate quantum calculations) is limited to small system size (thousands of atoms) and time scales (few pico-seconds) due to high computational cost. Recent advancement in machine learning has opened a new window to overcome this problem. Modern machine learning techniques can be used to learn the interatomic forces generated by quantum level of theory and thereby extend the time scale and length scale of standard AIMD simulations significantly.

In this talk, I will discuss our recently developed protocol to learn quantum level interatomic forces efficiently via an artificial neural network. I will show two instances where machine learning accelerated molecular simulations to discover unprecedented molecular details. First, I will discuss in detail a case study of modeling chemical reactions in complex heterogeneous environment where solvation, polarization and reactions play a key role.

Secondly, I will discuss our ongoing work on modeling charge and defect transport in strongly correlated perovskites rare earth nickelates. Finally, I will discuss current limitations and prospective avenues to improve those so that machine learning models would be used on a routine basis in molecular simulations.

"Post synthetic materials design principles, plasmonic signature tracking, and computational fluid dynamics framework to build an artificial synthetic chemist", Dr. Progna Banerjee, McKetta Department of Chemical Engineering, The University of Texas at Austin. Host: Xiao Min Lin

What would the nanomaterials discovery landscape look like if we can utilize artificial synthetic chemistry platforms to handle permutations, formulation selection, nanoparticle assembly on chip, and subsequent analyses? Can we couple this system with machine learning based algorithmic materials selection to predict unconventional phases and properties in these nanomaterials? To explore the realization of these systems I shall speak on the materials chemistry, predictive modeling and guided design principles that inform our scientific perspectives. I shall then describe how the experimental and computational skills acquired through these studies can be translated to build our artificial synthetic chemist.

"Theoretical and Experimental Analysis of Liquid Transport in Fluidic Systems with Micro/Nanostructured Surfaces: Toward Application in Separation Processes", Dr. Dhiraj Nandyala, Micro-Thermo Fluid Dynamics Group, Dept. of Mechanical Engineering, Stony Brook University, Host: Xiao-Min Lin

This presentation will discuss the theoretical and experimental analyses of liquid transport in fluidic channels and small capillaries that have surfaces with either natural or synthetic micro/nanoscale structures that significantly affect interfacial processes such as wetting and spreading, imbibition and drainage, and liquid transport driven by pressure and/or capillary forces. In the first part of this presentation, I will discuss my work at the National Synchrotron Light Source (NSLS-II) of Brookhaven National Laboratory (BNL) where X-ray Photon Correlation Spectroscopy (XPCS) techniques are employed to characterize the flow velocity profiles and rheological properties of colloidal fluids. A Fourier decomposition technique and a multivariable optimization algorithm were developed and applied to determine the flow velocity profiles and mass diffusivity from the intensity autocorrelation function obtained from XPCS experiments. In the second part of this presentation, I will discuss the design, fabrication, and characterization of a fluidic diode at Stony Brook University (SBU) for potential application in water-oil separation and microfluidic handling. Theoretical and experimental results indicate that a simple capillary device with micro/nanopatterned glass surfaces can be employed as a fluidic diode due to the presence of a large surface energy barrier that prevents the transport of specific fluid pairs. The scientific and technical knowledge from this work can have a significant impact on the design of novel micro/nanofluidic devices. At the end, I will briefly discuss special projects in Machine Learning using optimization algorithms for object detection. The research work presented in this talk was supported by the NSF through award CBET-1605809 and the SBU-BNL Joint Photon Sciences Institute (JPSI) Fellowship

"Development of Low-Cost Rapid Screening Techniques of Iridium (III) Photocatalysts", Velabo Mdluli, Carnegie Mellon University. Host: Jie Xu

Several high-throughput optical screening methods for the photocatalytic activity of a structurally diverse library of cationic iridium(III) complexes ([Ir(C^N)₂N^N)]⁺), where C^N is a cyclometalating ligand and N^N an ancillary ligand, were developed. These low-cost colorimetric assays allow for real-time monitoring of photo-induced electron transfer processes involved in hydrogen evolution and organic photoredox reactions. The infra-structure we have built is useful for generating large standardized experimental datasets. The obtained rate constants of the [Ir(C^N)₂N^N)]⁺ photocatalysts are then correlated to calculated chemical descriptors such as density functional theory (DFT)-derived features to help establish structure-activity relationships. This data-driven approach to chemical research is necessary to accelerate discovery of faster and chemo-selective catalysts.