Skip to content
Open access

Prediction of the physical properties of cement paste at meso‐scale based on level‐set function and XFEM model

Oct 2026 · Vol 9 · 0 citations · 8 references

Abstract

Predicting how microstructural transformations scale up to macroscale engineering parameters is crucial for modern concrete design. This paper presents an integrated numerical framework coupling the Level Set Method (LSM) for cement hydration and the mechanical XFEM to evaluate the bulk material properties e.g., elastic modulus, mechanical strength, and permeability of hydrating C3S pastes developing from the initial state to 28 days. The LSM implicitly tracks the tortuous boundaries of evolving phases (C3S, C‐S‐H, portlandite, and voids) via continuous signed distance fields. Rather than relying on computationally expensive body‐conforming meshes, the XFEM framework utilizes these level set distributions to enrich uniform mesh elements with Heaviside jump functions, accurately capturing sharp stress and strain gradients across material interfaces. By assigning intrinsic eigen‐properties to each isolated domain, virtual mechanical loading and fluid intrusion tests are executed to perform numerical homogenization. The preliminary results demonstrate that the coupled LSM‐XFEM approach provides a mathematically precise, computationally efficient alternative for multiphase upscaling.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.