Ruttan Hall B25
Cyber-Physical Advanced Biomanufacturing: Intercepting Physics-Informed Machine Learning with Electrified Process Intensification
Abstract: Legacy macroeconomic reliance on carbon-intensive, fossil-fueled petroleum refining has accelerated global climate volatility and sub-optimized industrial material lifecycles. To mitigate these grand challenges, my research program establishes a closed-loop Cyber-Physical Chemical Systems Engineering (CPSE) framework that bridges macro-scale techno-economic feasibility with micro-scale experimental transport phenomena, non-equilibrium thermodynamics, and scientific AI. This talk features three distinct pathways to demonstrate my group's current progress. First, an ambient-temperature, fully electrified ozone-cracking advanced biorefining platform is designed to precisely cleave unsaturated carbon-carbon double bonds in renewable lipid streams without thermal input, enabling precise molecular tailoring to synthesize high-utility bio-based drop-in energy carriers and advanced materials. Second, to eliminate destructive offline testing in these multi-phase reaction fields, I integrate in-house-engineered, inline dielectric edge-sensing networks that operate as a real-time Process Analytical Technology (PAT) infrastructure. By coupling these microelectronic edge devices with hybrid Physics-Informed Machine Learning (PIML) networks, our continuous process digital twins autonomously optimize mass-transfer limits, characterize transient multi-component kinetics, and completely suppress model divergence on sparse physical grids. Lastly, to overcome the severe latent-heat penalties associated with conventional thermal vaporization during feedstock pretreatment and downstream biorefining, I demonstrate a fully electrified, multi-frequency, multi-mode, modulated (MMM) ultrasonic dehydration and dewatering platform that intensifies non-thermal mass transfer. This comprehensive framework establishes a highly scalable template for fully autonomous, continuous advanced biomanufacturing and data-driven inverse process optimization.
Dr. Junli Liu is a Principal Investigator and Senior Research Scientist in the Department of Agricultural and Biological Engineering at Purdue University. He holds a Ph.D. in Biological Engineering from Purdue University. Dr. Liu specializes in Cyber-Physical Chemical Process Systems Engineering (CPSE), multi-scale transport phenomena, non-equilibrium thermodynamics, and scientific AI. Prior to his current appointment at Purdue, he conducted pioneering process-intensification research at the University of Illinois Urbana-Champaign (UIUC), where he engineered scalable, carbon-neutral ultrasonic biorefining systems with broad applications in chemical refining, advanced separation operations, and industrial food processing.
At Purdue University, Dr. Liu dedicated himself to accelerating industrial decarbonization and the discovery of green materials. His research program has pioneered fully electrified chemical-conversion infrastructure and ambient-temperature ozone-cracking pathways to synthesize next-generation, bio-based drop-in energy carriers and functional materials. His patented inventions include sustainable aviation fuels (SAF/biokerosene), advanced biolubricants, and bio-dielectric coolants designed for EV batteries and high-heat AI data-center immersion cooling. To enable fully autonomous process control, his group designs in-house microelectronic hardware—including wireless, inline microwave resonance spectroscopy (RS) edge sensors—and deploys physics-informed machine learning (PIML) frameworks to drive real-time Process Digital Twins without computational boundary divergence.