DMREF: Turning Carbon Dioxide into 3D-Printed Concrete via Integrated Machine Learning, Simulations, and Experiments

点击翻译按钮获取中文摘要

Concrete, which is by far the most manufactured material in a world, faces many challenges, including a substantial embodied carbon intensity, slow strength-development rate, and poor strength-to-weight ratio. To overcome these challenges, alternative cements are being investigated that allow the uptake of carbon dioxide. This Designing Materials to Revolutionize and Engineer our Future (DMREF) award supports fundamental research to accelerate the synthesis and design of these alternative materials through an integrated computational and experimental approach. New knowledge of the mechanisms that control processing-structure-property relationships has the potential to enable the design and discovery of environmentally-friendly concrete materials that can be 3D printed into customizable forms. By allowing take of carbon dioxide in the processing, the carbon footprint associated with traditional concrete production can be greatly reduced. This research has the potential to redesign and reimagine concrete as a resource for the production of value-added products rather than as a waste. By seamlessly integrating experiments, simulations, and machine learning, this collaborative project will train students to be well-versed in both experimental and modeling approaches of relevance to materials science and engineering, resource management, and construction industries. A focus on the training of undergraduate students will engage the environmental conscience of the next generation of engineers.This objective of this research is to decipher the fundamental knowledge required to accelerate the design of a new 3D-printable portlandite-based cementitious binder that permits CO2 uptake. Toward this end, this research aims: (i) to understand, control, and optimize the rheology of concentrated portlandite suspensions to enable printability, (ii) to refine portlandite carbonation routes at ambient temperature to maximize CO2 uptake to accelerate the carbonation kinetics, and (iii) to discover new multi-material 3D-printed metastructures with high load-bearing capability and optimal strength-to-weight ratio. This research relies on an iterative closed-loop integration of simulation (i.e., from electrons to continua), experimental, and machine learning activities that mutually inform and advance each other. The synergy between experimental and computational approaches will shed new light on interfacial reaction processes of mineral sorbents. This project will also advance the state of the art in our understanding of the rheology of concentrated suspensions, and elucidate the molecular design principles behind the discovery of polymers that permit the printability of concentrated slurries. Finally, by pioneering machine-learning-informed multi-material 3D-printing, this research will develop new methods to optimize the geometry and spatial distribution of metastructures that are light, stiff, and strong. Overall, by marrying the benefits of CO2 mineralization and 3D-printing, this work will result in pioneering intellectual contributions to accelerate the design of transformative construction materials with desirable properties and low carbon impact.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.