(2gj) Sustainable Product and Process Intensification through Molecular and Process Optimization

Sustainability, energy security and human healthcare requires innovative chemical products and process alternatives. For instance, biorefinery requires the design of solvents to ensure survival of microorganisms for biomass conversion and to separate bioproducts and biofuels into solvent-rich phase. Here, the solvents are intensified products which remove inhibitors of microorganisms and have higher affinity with bioproducts and biofuels. These also result in process intensification by combining fermentation and separation in situ with enhanced microbial conversion and product purification. Search of these innovative product and process alternatives is a complex system decision-making problem, which requires effective modeling and exploration of design space. My research in the field of molecular modeling and process system engineering offers solutions for these decision-making problems ranging from molecular to process scale. My group will focus on development of modeling and optimization methods for product design and process intensification with applications in sustainability, energy, and healthcare.

Research Interests

My group will integrate the methods of molecular dynamics simulation, process representation, optimization, and machine learning to identify and understand the synergy and trade-offs among system components. Specifically, my research interests will focus on two directions:

• Identification of optimal and innovative energy systems for production of platform chemicals, biofuels, and biodegradable plastics from renewable energy feedstocks. The graphic representation serves as basis of energy system modeling with key decisions on reaction pathway selections, feedstock types, separation methods and existence of intensification. I’m especially interested in graphic representation of innovative energy systems that enable structured modeling, optimization, and learning. Current challenges exist in representation and modeling of novel and intensified energy system without extensive exploration. My work addresses three fundamental research questions: 1) how to represent and model energy systems enabling fast optimization? 2) how to directly relate system graphic representation with design objectives? 3) how to achieve inverse design and retrofit of energy systems? The solutions to these questions help to answer three critical questions regarding sustainability and energy system design: 1) what are the desired unit operations to produce platform chemicals? 2) what are the thermodynamic hotspots for reducing energy and feedstock consumption in biorefinery? 3) how to generate heuristics of retrofitting industry-relevant energy systems and equipment? To address these questions, we develop and apply graph-theoretic, optimization-based methods, techno-economic analysis and machine learning for representation, exploration, and interpretation of energy system.
• Identification of optimal products with enhanced transport, reactivity, and separation. Chemical products can serve as solvation media for reaction (reaction solvents), decrease reaction barrier (catalyst), move certain compounds into different phases (separation solvents), and protect cells (drugs). Intensified chemical products with multi-functional properties are desired. Current challenges exist in systematic exploration and discovery of molecule and product intensification opportunities. My work addresses two fundamental research questions: 1) how to effectively explore molecular and mixture design space? 2) how to achieve fast and accurate property prediction and validation? Answers to these questions facilitate the optimal and rational design of solvents, catalysts, and drugs with applications in separation, reaction, and healthcare. For addressing these questions, we develop systematic framework consisting of molecular dynamics simulation, molecular geometry optimization, machine learning for discovering design opportunities and elucidate physicochemical insights.

Research Experience: My research has focused on the use of network representation, mathematical modeling, and optimization to identify innovative and intensified chemical flowsheets [1-2] and integrate property prediction, product design, and process design [3-4].

At the process level, my PhD work has focused on structured representation, modeling, and optimization of chemical flowsheets for innovative process design with energy, environment, and sustainability applications. This structured representation filled a research gap in systematic methods for process intensification. In contrast to conventional equipment-based process representation, we have adopted a bottom-up approach to represent chemical flowsheets using abstract phenomena building blocks. The combination of these building blocks represents equipment and collection of these building blocks represents chemical flowsheets. Based on this structured representation using building blocks, we have pioneered the development of a general optimization-based framework that combines separation methods selection, reaction selection, in situ reaction and separation, separation material selection and work-heat-mass integration. Hence this computational framework allows the automatic screening and design of optimal process flowsheets with identification of process intensification and selection of functional materials [1-2]. The computational framework I have developed is the first simultaneous optimization-based approach for process intensification. My research enabled a computer-aided platform which embeds a unified mathematical model for process synthesis, integration, and intensification. The platform automatically collects problem data, implement optimization, and generate process configurations.

At the molecular and process level, my postdoctoral work has focused on bridging the gap among property prediction, product, and process design for bioproduct and biofuel separation. Biomass-derived bioproduct and biofuel are often at low-concentration, which requires systematic decision making of separation methods based on feedstock properties. My research address four research questions: 1) how to explore temperature and solvent-dependent molecular structure space to improve property prediction? 2) how to explore solvent mixture space? 3) how ionic effects influence the self-assembly of aromatic compounds? 4) when extraction is more favored than distillation given system properties? I have integrated molecular modeling, surrogate modeling, process modeling, and collaborated with experimental researchers for computational validation to 1) screen ionic solution to facilitate self-assembly of bioproduct; 2) design green solvent mixtures for chromatography separation of depolymerized lignin-based aromatic compounds; 3) identify and interpretate the property domain where extraction is more economic than distillation. My research has enabled a multi-scale systematic and interpretable framework for selection of separation methods and products to purify bioproduct and biofuels [3-4].

Teaching Interests

As an Assistant Professor, I am eager to teach all undergraduate chemical engineering courses with special interests in Chemical Engineering Thermodynamics, Numerical Analysis for Chemical Engineers, Chemical Engineering Plant Design, Process Integration. I am interested in giving online courses that offer time and distance convenience for students from various background. I am also interested in introducing courses at undergraduate/graduate levels on Advanced Separations and Product Design and Process Intensification.

Teaching and Mentoring Experience: I have been a teaching fellow at Texas A&M Academy for Future faculty with professional training on curriculum development and teaching skills; I have been a teaching assistant for multiple undergraduate courses, e.g., Heat Transfer, Mass Transfer and Chemical Engineering Fluid Operations; I have substantially developed my supervising skills as I have mentored 8 undergraduate students from different engineering backgrounds through Aggie Research Leadership program.

Selected Publications:

[1] J. Li, S.E. Demirel, M.M.F. Hasan. Process Synthesis using Block Superstructure with Automated Flowsheet Generation and Optimization. AIChE Journal, 2018, 64(8), 3082.

[2] S.E. Demirel, J. Li, M.M.F. Hasan. Systematic Process Intensification using Building Blocks. Computers & Chemical Engineering, 2017, 150, 2–38.

[3] J. Li, C.T. Maravelias, R.C. Van Lehn. Adaptive Conformer Sampling for Property Prediction Using the Conductor-like Screening Model for Real Solvents. Industrial & Engineering Chemistry Research. Accepted, 2022. doi: 10.1021/acs.iecr.2c01163

[4] J. Li, R.C. Van Lehn, C.T. Maravelias. Data-driven Explainable Classification of Economic Extraction and Distillation for Bioproduct Separation. Separation and Purification Technology. In preparation.