# Constitutive Modeling of Rheological Behavior of Cement Paste Based on Material Composition

## TL;DR

This paper introduces a new model to predict the flow behavior of cement paste based on its material composition, improving concrete design and performance.

## Contribution

A novel composition-based constitutive model using a virtual maximum packing fraction to predict cement paste rheology.

## Key findings

- The model accurately predicts yield stress and plastic viscosity with R2 > 0.98 for plain pastes.
- Superplasticizer effects are captured through modifications to the virtual maximum packing fraction.
- The model supports intelligent mix design for high-performance concrete applications.

## Abstract

The rheological behavior of cementitious paste plays a pivotal role in determining the workability, pumpability, and uniformity of fresh concrete. Classical rheological models often struggle to capture the complex flocculation and hydration effects inherent in cement-based systems, and they typically depend on parameters that are difficult to measure directly, limiting their practical utility. This study presents a novel composition-based constitutive model that introduces a virtual maximum packing fraction (ϕmax) to account for interparticle flocculation and entrapped water effects. By establishing quantitative relationships between powder characteristics—such as particle size and specific surface area—and rheological parameters, the model enables physically interpretable and measurable predictions of yield stress and plastic viscosity. Our validation against 65 paste formulations with varying water-to-binder ratios, mineral admixture types and dosages, and superplasticizer contents demonstrates strong predictive accuracy (R2 > 0.98 for plain pastes and >0.85 for blended systems). The influence of superplasticizers is effectively captured through modifications to ϕmax, allowing the model to remain both robust and parameter efficient. This framework supports forward prediction of paste rheology from raw material properties, offering a valuable tool for intelligent mix design in high-performance concrete applications such as self-consolidating and 3D-printed concrete.