Research

Sustainable Manufacturing of Portland Cement

Our research group is pursuing collaborative efforts to facilitate the production of Portland cement, along with other lime-based cements, via carbon-neutral and energy-efficient production routes. Our work encompasses several technological innovations. Firstly, we are developing novel electrolytic methods for low-temperature, carbon-neutral decomposition of limestone into lime. Secondly, we are spearheading high-temperature synthesis routes for ultrafast production of cements with physical and chemical properties akin to those of commercially-produced available counterparts. Thirdly, we are pioneering the use of renewable energy, instead of fossil fuels, to power all sub-processes of cement manufacturing. Finally, our work emphasizes the synergistic use of advanced experiments, thermodynamic and multiphysics simulations, coupled with artificial intelligence, to optimize the manufacturing process. This comprehensive approach enables us to produce a family of energy-efficient and sustainable, next-generation cements.

NeutraCEM: Reimagining Cement Manufacturing for Carbon Neutrality

EPIXC: Electrified Processes for Industry without Carbon

Design and Optimization of Performance of Alternative Cements

These projects are focused on fundamental materials research to enhance the utilization of low-calcium cements (LCCs) to formulate sustainable and durable binders for concrete. Machine learning and experiments that are guided by thermodynamics are employed to reveal mechanisms, and underlying composition-microstructure-performance links, that drive the physicochemical behavior of LCCs. In addition, synergistic interactions between LCCs and abundant waste materials such as calcined clay and fly ash are capitalized on to produce binders that exhibit compliance and constructibility metrics that are comparable to those of Portland cement concretes.

Sustainable and Durable Calcium Sulfoaluminate Binders Enabled by Multi-Physics Characterization and Theory-Guided Machine Learning

Center for Durable and Resilient Transportation Infrastructure

Enabling and Enhancing the Use of Solid Wastes in Concrete

These projects are focused on high-volume integration of solid waste into concrete production to provide a sustainable avenue for waste management, reducing environmental impact by repurposing materials that would otherwise accumulate in landfills. Certain solid wastes, such as fly ash, slag, or recycled concrete aggregates, can substitute or supplement traditional cement constituents, often enhancing the final product’s performance and durability. Machine learning is used to identify suitable waste materials for concrete production and predict the resulting mix’s performance characteristics. Supervised and reinforcement machine learning algorithms are used to process vast amounts of data from various waste types, understand complex interactions among different components, and predict the resultant concrete’s mechanical properties and durability. This aids in achieving optimal, eco-friendly concrete formulations, ensuring both high-quality construction and responsible waste management.

Physically Informed Data-Driven Methods for Greatly Enhancing the Use of Heterogeneous Supplementary Cementitious Materials in Transportation Infrastructure

Converting Off-Specification Coal Ash and Incinerator Ash into High-Value Blended Cementitious Materials through Enhanced CO2 Uptake

A Thermo-Kinetic Approach to Enhance the Use of Clays in Concrete

Artificial Intelligence-Based Smart Toolbox for 3D Printing of Concrete

Converting CO2 and Alkaline Solid Wastes into Carbon-Negative Supplementary Cementitious Materials for Co-decarbonization of Multiple Sectors

Immobilization of Hazardous Waste

As part of a collaborative project, our research group is employing a triad of laboratory experiments, computational models, and machine learning techniques to innovate the design of durable waste forms for nuclear waste immobilization. Building upon experimental insights, we’re using computational models and machine learning algorithms to effectively analyze complex data, identifying hidden patterns and correlations for enhanced insights. This integrative approach is expected to significantly enhance the safety, effectiveness, and efficiency of nuclear waste immobilization, propelling sustainable management of nuclear wastes.

Pioneering a Cermet Waste Form for Disposal of Waste Streams from Advanced Reactors (PACE-FORWARD)

Composition-Structure-Reactivity-Properties Correlations in Infrastructure Materials

Our research group is collaborating with external partners to revolutionize the understanding of siliceous materials such as cement and multicomponent glasses. We adopt an interdisciplinary approach that fuses traditional experimentation with multiphysics models and machine learning. This unique integration empowers us to investigate and establish intricate correlations between composition, structure, reactivity, microstructure, and properties in these materials. By unveiling these links, we can predict and tune the behavior of these materials under different conditions. This paves the way towards creating superior, tailored materials that meet diverse demands in areas ranging from construction to waste encapsulation, thereby contributing to sustainable development and technological advancement.

Decoding the Corrosion of Borate Glasses: From Fundamental Science to Quantitative Structure-Property Relationships

Low-Temperature Architected Cementation Agents (LAMINAE)

Physically Informed Data-Driven Methods for Greatly Enhancing the Use of Heterogeneous Supplementary Cementitious Materials in Transportation Infrastructure