Homogenization mathematical model of the cemented filling slurry with crushing waste rock and whole tailings

Abstract

Aiming at the problems of pipeline transportation blockage and filling body stratification caused by waste rock–unclassified tailings high-concentration slurry, the effects of superplasticizer and stirring parameters on the slurry homogenization were experimented with, and the quantitative characterization of slurry homogenization was explored. Initially, the polycarboxylate (PC) superplasticizer with the best suitability was screened out based on the bleeding-slump test, and the mathematical correlations of the slump and bleeding rate with the optimal superplasticizer dosage range were obtained by regression. The rheological properties of the slurry and the filling body strength were then determined at different PC superplasticizer dosages, and separate mathematical models for correlations of slurry rheological parameters and mechanical properties with superplasticizer dosage were built. Next, the slurry surface images under different stirring conditions were acquired with the Nikon D350 camera, and their information entropies were calculated. Meanwhile, the OTSU algorithm was used to perform image segmentation thresholding, and the images were binarized via Matlab, followed by a calculation of the proportion of black pixels in the binarized images. Further, the variation trends of image information entropy and black pixels proportion with PC superplasticizer, rock/tailing ratio (mass ratio of waste rock to unclassified tailings), and stirring time were derived. Finally, the homogenization mechanism in the waste rock–unclassified tailings filling slurry was revealed based on the PC superplasticizer's regulatory role in fine particle absorption and dispersion, which was further validated by the relationship between the zeta potential of cement paste and the dosage of PC superplasticizer. On this basis, a quantitative model of slurry homogenization was developed based on the slump, bleeding rate, rheological properties, strength characteristics, and image information, and the optimal parameters of waste rock–unclassified tailings high-concentration filling slurry were obtained by multi-objective programming. The results show that the PC superplasticizer is highly suitable for the slurry, which can reduce its yield stress and plastic viscosity coefficient and improve its fluidity. When the dosage of PC superplasticizer is 0.50%, the yield stress and plastic viscosity of the slurry is reduced by 34.4% and 21.2%, respectively, in comparison to the case without superplasticizer. The slurry rheological properties conform to the Bingham plastic model. Increasing the PC superplasticizer dosage improves the early strength of the filling body and weakens the 28-d strength. Nonetheless, within the optimal dosage range, all the filling body strengths can meet the mine filling requirements. Slurry surface images with higher information entropy and a smaller proportion of black pixels indicate a higher degree of slurry homogenization. Moreover, the entropy value of slurry surface images tends to increase initially and then decrease with the prolonging of stirring time and the heightening of the rock/tailing ratio. When the rock/tailing ratio is constant, the proportion of black pixels is the largest at a stirring time of 3 min, followed by 5 min, and the smallest at 4 min. According to the quantitative model results of slurry homogenization, the reasonable dosage range of PC superplasticizer is 0.26%–0.5%, and the optimal stirring time is 4.3 min. The degree of homogenization is the best at a 0.5% dosage, at which point the slurry has a plastic viscosity μ of 0.79 Pa·s and a yield stress τ of 202.25 Pa.

Keywords

• slurry homogenization
• rheological properties
• strength properties
• image information
• polycarboxylate water reducer