Millennium Café

Lost files, broken access, undocumented datasets, and simple handling errors can quickly turn into lost time, lost research, or even lost funding. At the same time, expectations from funders and regulators around data management, public access, and security are changing. This talk will explore what these changes mean for researchers, practical ways research groups can reduce risk, and how Penn State units are working together to make good data management easier. The MRI-to-ICDS automated data transfer will provide one example of how better infrastructure can make good practice the easy path.

Wolf Hey | Institute for Computational and Data Sciences (ICDS)

I will present two ideas driving my current research on food waste mitigation: food rescue and choice architecture. With approximately one-third of all food produced going to waste, the environmental and societal consequences are substantial, particularly given the connection between food waste and food insecurity. Addressing this challenge offers significant opportunities for interdisciplinary research, and I hope this talk will spark conversation and new collaborations.

Aydin Alptekinoglu | Smeal College of Business

Millennium Café will resume on November 10.
See you then!

A fundamental question in characterization is whether we have measured enough of a sample to trust the result. Using automated mineral mapping 9SEM-AM) as our test case, we are developing a Monte Carlo framework that relates sampling area, grain size, and abundance to uncertainty in measured quantities. The framework can also work backward to estimate how much material needs to be analyzed to reach a desired precision, potentially helping researchers design experiments more efficiently. I welcome feedback from the characterization community as we continue developing and validating the approach.

Jesse Reimink | Geosciences

What if experiments that normally take days could be done in much less time? Our transient reactor generates rich kinetic data on catalyst reactions in seconds to minutes, with AI-guided modeling used to uncover the underlying catalyst behavior. A new DOE Genesis Mission project brings together Penn State, Georgia Tech, Brookhaven National Laboratory, and ExxonMobil to advance this approach and explore how faster, data-rich experiments coupled with AI can accelerate experimental science.

Mike Janik | Chemical Engineering

This special edition of the Millennium Café will feature a moderated panel discussion exploring how artificial intelligence and automation are moving beyond data analysis and into the physical laboratory. Robotics can execute complex synthesis workflows, in situ characterization can provide real-time information as materials form, and increasingly sophisticated software can operate instruments, interpret results, and help determine what experiment should happen next.

The panel will explore what happens when AI and automation become active participants in experimentation. How can synthesis, characterization, data analysis, and decision-making be connected into closed-loop experimental workflows? What laboratory tasks are ready for automation today, and where does human expertise remain essential? As instruments become increasingly programmable and autonomous, how might the roles of scientists, laboratories, and scientific instrumentation itself change?

Joan Redwing | Materials Science & Engineering
Enrique Gomez | Chemical Engineering
Jim Kilcrease | Thermo-Fisher Scientific