The complexity of the hydration process of cementitious materials makes developing universal models that predict their mechanical behavior challenging. The current study used a novel molecular modeling method to develop a bottom-up model of the cement paste that incorporated different contents of the nano-red mud (nRM). A molecular dynamics (MD) simulation and dreiding force field were employed to build the ingredient-based models using the cement clinker phases and nRM oxides. Different cement-nRM mixtures were prepared and macroscopically tested under compression and flexure. In addition, a detailed microstructural analysis using XRD, SEM, EDS, and Raman was carried out on the prepared of nRM. The experimental results were used to validate the results of the developed molecular models. The simulation results demonstrated the capability of the atomistic models for predicting the strength and stiffness of the cement-nRM composites accurately. In addition, the developed hydrated cement model was found with the line of C–S–H-mineral structure (called tobermorite 11 Å). The molecular characterization, in terms of the radial distribution function and fractional free volume, also provided useful information on the local structures of the models. The molecular models of cement-nRM can be further used to predict the durability properties as well as gain deep insights into the mechanical performance of blended mixtures.

Original languageEnglish
Article number107902
JournalJournal of Building Engineering
StatePublished - 15 Nov 2023

Bibliographical note

Publisher Copyright:
© 2023 Elsevier Ltd


  • Cement
  • Mechanical properties
  • Microstructural analysis
  • Molecular dynamics simulation
  • Red mud

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Architecture
  • Building and Construction
  • Safety, Risk, Reliability and Quality
  • Mechanics of Materials


Dive into the research topics of 'Development of an atomistic model of cement-incorporated nano-red mud material'. Together they form a unique fingerprint.

Cite this