Cemented paste backfill prepared with waste rock and unclassified tailings: Mechanical properties, micro-mesoscopic structure and WOA-XGBoost prediction model

To investigate the mechanical properties and flowability of waste rock-tailings composite cemented backfill, orthogonal tests evaluated the effects of the sand-to-cement ratio, waste rock particle gradation index, fiber content, and fiber length on strength and slump. The microstructure was characterized, and machine-learning models were used to assess their accuracy in predicting overall backfill performance. Results indicate that the sand-to-cement ratio is the dominant factor controlling mechanical performance. Increasing the sand-to-cement ratio and waste rock particle gradation index reduces both compressive and tensile strengths, with the 28 d compressive strength decreasing by 51.9% and 11.5%, respectively. With increasing fiber content and fiber length, both compressive and tensile strengths initially increase and then decrease, reaching their peak values at approximately 0.6% FC and 6 mm FL, respectively.The optimal mix identified by the efficacy coefficient method comprises a sand-to-cement ratio of 4, a waste rock particle gradation index of 0.4, a fiber content of 0.4%, and a fiber length of 6 mm. Increasing the sand-to-cement ratio reduces hydration products, promotes pore and microcrack development, and transforms the dense specimen microstructure into a loose state, thereby lowering strength. An appropriate polypropylene fiber dosage provides bridging, anchoring, and crack-arresting effects, suppressing crack propagation and preserving specimen integrity after failure. Furthermore, a whale optimization algorithm-extreme gradient boosting (WOA-XGBoost) model was developed. The test-set coefficients of determination (R²) for 28 d compressive strength, 28 d tensile strength, and slump were 0.956, 0.873, and 0.938, respectively, outperforming the standalone extreme gradient boosting (XGBoost), random forest (RF), and support vector regression (SVR) models. These findings provide theoretical support for mix-proportion optimization, micromechanism analysis, and intelligent performance prediction of waste rock-tailings composite cemented paste backfill.

Authors

Institutions

Publication Details

Journal
Materials Today Communications
Published
2026-11-17
DOI
https://doi.org/10.1016/j.mtcomm.2026.116667
Primary Topic
Tailings Management and Properties
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Cemented paste backfill prepared with waste rock and unclassified tailings: Mechanical properties, micro-mesoscopic structure and WOA-XGBoost prediction model

Ke Yang, Tong Wang, Shenghua Yin, Leiming Wang et al.
3 links
Materials Today Communications
Tailings Management and Properties
article

Cemented paste backfill prepared with waste rock and unclassified tailings: Mechanical properties, micro-mesoscopic structure and WOA-XGBoost prediction model

Ke Yang, Tong Wang, Shenghua Yin, Leiming Wang, Yongqiang Hou, Xinyue Zhang, Xinyuan Zhao, Xin Yu
article en

Abstract

To investigate the mechanical properties and flowability of waste rock-tailings composite cemented backfill, orthogonal tests evaluated the effects of the sand-to-cement ratio, waste rock particle gradation index, fiber content, and fiber length on strength and slump. The microstructure was characterized, and machine-learning models were used to assess their accuracy in predicting overall backfill performance. Results indicate that the sand-to-cement ratio is the dominant factor controlling mechanical performance. Increasing the sand-to-cement ratio and waste rock particle gradation index reduces both compressive and tensile strengths, with the 28 d compressive strength decreasing by 51.9% and 11.5%, respectively. With increasing fiber content and fiber length, both compressive and tensile strengths initially increase and then decrease, reaching their peak values at approximately 0.6% FC and 6 mm FL, respectively.The optimal mix identified by the efficacy coefficient method comprises a sand-to-cement ratio of 4, a waste rock particle gradation index of 0.4, a fiber content of 0.4%, and a fiber length of 6 mm. Increasing the sand-to-cement ratio reduces hydration products, promotes pore and microcrack development, and transforms the dense specimen microstructure into a loose state, thereby lowering strength. An appropriate polypropylene fiber dosage provides bridging, anchoring, and crack-arresting effects, suppressing crack propagation and preserving specimen integrity after failure. Furthermore, a whale optimization algorithm-extreme gradient boosting (WOA-XGBoost) model was developed. The test-set coefficients of determination (R²) for 28 d compressive strength, 28 d tensile strength, and slump were 0.956, 0.873, and 0.938, respectively, outperforming the standalone extreme gradient boosting (XGBoost), random forest (RF), and support vector regression (SVR) models. These findings provide theoretical support for mix-proportion optimization, micromechanism analysis, and intelligent performance prediction of waste rock-tailings composite cemented paste backfill.

Materials Today CommunicationsVol. 57
Anhui University of Science and Technology (CN), Chinese Academy of Sciences (CN), Xinjiang Institute of Engineering (CN), Institute of Geology and Geophysics (CN), University of Chinese Academy of Sciences (CN), University of Science and Technology Beijing (CN)
National Natural Science Foundation of China, Anhui University of Science and Technology
Responsible consumption and production, Industry, innovation and infrastructure
Tailings Management and Properties
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon