MARATTO

article · Case Studies in Construction Materials

Predictive modeling for mechanical characteristics of ultra high-performance concrete blended with eggshell powder and nano silica utilizing traditional technique and machine learning algorithm

In plain language

This research evaluates the use of eggshell powder and nano-silica as partial replacements for cement in ultra-high-performance concrete. Eggshell powder was tested at replacement ratios ranging from 2.5 percent to 12.5 percent, both on its own and pre-mixed with nano-silica. Experimental testing assessed compressive, splitting, and flexural strengths, alongside sorptivity and microstructural properties. In parallel, the investigation developed predictive tools using response surface methodology and artificial neural networks to model these mechanical characteristics. The findings indicate that concrete containing a 10 percent replacement of cement with the pre-mixed eggshell powder and nano-silica composite delivers the best mechanical performance and a densely packed microstructure. Furthermore, machine learning via artificial neural networks predicted the properties of the concrete with greater accuracy than response surface methodology, offering a dependable way to design and optimise such concrete mixes.

Key takeaways

  • Pre-mixing eggshell powder with nano-silica produces superior mechanical performance in ultra-high-performance concrete compared to using eggshell powder alone.
  • A 10 percent replacement of cement with the eggshell powder and nano-silica composite yields the optimum mechanical strength and microstructural density.
  • Artificial neural network models predict the mechanical properties of this concrete mix more accurately than traditional response surface methodology.
  • Substituting cement with these materials lowers carbon dioxide emissions and reduces overall material costs.

Why it matters

Cement production is a major contributor to carbon emissions and construction costs. By validating locally available eggshell waste combined with nano-silica as an effective cement substitute, this work supports more sustainable building practices. Additionally, using accurate machine learning models allows concrete developers to simulate mix performance quickly, cutting down on the time, expense, and raw materials needed for physical laboratory testing.

Commercialisation angle

This work is relevant to concrete manufacturers, construction firms, and materials testing laboratories seeking lower-cost, lower-carbon concrete formulations. Because the findings rely on laboratory-scale batching and computational modelling, the technology sits at an applied and tested research stage. Commercial adoption would require industrial batch validation, standardisation of eggshell processing, and integration of the predictive artificial neural network tools into existing mix design workflows.

AI-generated from the published abstract. Always read the original work before citing.

Abstract

The construction industry currently utilizes nanomaterials to improve the mechanical performance of ultra-high-performance concrete (UHPC). Eggshell powder has recently gained popularity in the fields of construction and building materials owing to its local availability and better cementitious properties. In addition, the use of advanced mix design techniques has recently become popular for saving experimental cost and time. This study aims to assess the effects of new composite material using eggshell powder pre-mixed with nano-silica as a partial substitute for cement in the production of UHPC and develop models using Response Surface Methodology (RSM) and Artificial Neural Networks (ANN). Eggshell powder was substituted at different ratios, such as 2.5, 5, 7.5, 10, and 12.5 %, in plain UHPC for reference. Additionally, a new composite of eggshell powder premixed with nano silica in UHPC was developed for comparison. Mechanical properties such as compressive strength, splitting, and flexural strength were investigated in addition to sorptivity and microstructure analysis using thermogravimetric analysis (TGA) and scanning electron microscopy (SEM). Furthermore, the developed (ANN + RSM) models showed a high level of accuracy with the actual results and showed that these models can be employed to determine the mechanical characteristics of UHPC incorporating eggshell powder. The R 2 value, experimentally predicted variation, and error analysis showed that the ANN model was more accurate than the RSM model. The experimental results showed that the optimum results were obtained with better mechanical performance and a densely packed microstructure for a 10 % replacement ratio when the eggshell powder was pre-mixed with nano-silica. According to this study, ES and ES-PNS as partial cement replacements in UHPC reduce CO 2 emissions and material costs. ES and ES-PNS in ultra-high-performance concrete production have environmental and economic benefits. • This study shows the viability of eggshell powder in ultra-high-performance concrete (UHPC). • The investigation explores the partial replacement of cement with eggshell powder pre-mixed with nano silica. • UHPC with eggshell powder pre-mixed with nano silica shows enhanced mechanical performance, outperforming UHPC with eggshell powder alone. • The results show that eggshell powder pre-mixed with nano silica replacement of 10 % performs best. • The error analysis showed that the ANN model was more accurate than the RSM model.

Research topics

  • Innovative concrete reinforcement materials
  • Concrete and Cement Materials Research
  • Innovations in Concrete and Construction Materials

Sustainable Development Goals

Read the original research

This page summarises published work. The authoritative version sits with the publisher.

DOI: 10.1016/j.cscm.2024.e04025

Is something wrong with this record? Report it or request removal.

Discussion

Discuss this research

Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.

No discussion yet. Open the first thread.