With the advent of new crewed missions to extraterrestrial bodies such as the Moon and Mars, there is a need to prolong these missions to conduct science and experiments using habitats that are constructed on[1]site. Invariably, to realize such large-scale constructions it is envisioned that in-situ materials such as the lunar regolith will be used along with cement-like binders. However, the hydration and solidification of cement formed under the microgravity (10-6 g or µg) environment, and their impact on mechanical properties are not well understood. To reliably employ cement binders for the design of large structures under microgravity conditions, their mechanical properties such as the strength and Young’s modulus must be thoroughly characterized. Moreover, this insight will enable the design of reliable large-scale structures on extraterrestrial bodies as well as contribute to Earth-based cementitious science. In this talk, a micromechanics-based approach utilized for the mechanical evaluation of the cement samples that were cured onboard the International Space Station (ISS) will be presented. Due to sample size limitations and high porosity, experimental characterization of the space-returned samples via standard mechanical testing is not viable. Previously, Scanning Electron Microscope (SEM) images were employed to evaluate the microstructural development of the samples returned from space. This helped capture how gravity affects the microstructural formation in cement. A Convolutional Neural Networks (CNN) -based deep learning framework was used to reconstruct Representative Unit Cells (RUC) from these SEM images. The reconstructed RUCs captured the microstructure as well as the morphology of the space-returned cement samples. The high-fidelity RUCs are then directly utilized in the NASA Multiscale Analysis Tool (NASMAT) that employs multiscale recursive micromechanics. The proposed workflow estimates the mechanical properties of the µg cement sample and paves the way for investigating the process-property relationship of cementitious systems in the microgravity environment. The methodology presented here and demonstrated on space cement can be easily expanded and adopted within the recursive micromechanics framework to other material systems. This project is funded by NASA’s Physical Sciences Research Program (PI: Profs. Namiko Yamamoto, Aleksandra Radlińska, Penn State; Co-I: Dr. Vishnu Saseendran, Penn State; Collaborators: Drs. Evan Pineda, Brett Bednarcyk, NASA Glenn Research Center; Program Scientist: Dr. Enrique M. Jackson, NASA Marshall Space Flight Center).