The characterization of industrial process water properties: A methodology case study from the mining industry

Abstract

The dynamicity and variability in terms of quality of recycled mining process water (PW) brings challenges in the recovery and the quality of the final products. By turning water quality into operating parameters, its properties could be controlled and modified to hopefully provide added value to the process outputs.

This thesis aims to explore the challenges for transforming water quality into a set of operating parameters in mineral processing. This study focuses on three aspects for effective management of water towards optimizing process performance: (1) a database for linking water quality to flotation performance output and serving as a base to explore the PW characteristics, (2) tools for monitoring and controlling PW quality and (3) a laboratory protocol for creating and predicting the plant PW for testing and plant design. The traditional methods for those mentioned aspects developed for effluents (environmental purposes) have neglected the three major characteristics of recycled process water: dynamic, complex and mine specific, a shortcoming this thesis addresses.

Using two case studies, this thesis show that traditional, standardized methods that have been developed for sampling, preserving, and analyzing mining effluent for environmental purposes failed to apply for process water. Employing generalized standard methods without optimization to the prevailing conditions may cause high uncertainty of analysis results.This thesis also highlights that a weekly or daily sampling for monitoring effluents for environmental purposes is not suitable for operation purposes due to the high variation of process water quality. The sampling procedure should be customized. Moreover, controlling and monitoring complex water matrixes call for multivariate data analysis methods. Advance statistical parameters results of multivariate statistical models could allow monitoring both operating range of the inputs and the correlation between variables, i.e. the thermodynamic state of the water. Finally, the dissolution loop (D-loop) laboratory protocol described in this study is an important tool that permits the prediction of the water quality variation over time due to the dissolution of the ore when the plant is operated in closed water circulation. With the D-loop, recycled water that contains nearly the same impurities as the process waterat the particular plant could be produced to replace distilled water or synthetic water for laboratory testing. Such an approach is realistic and appropriate to solve the challenge of undefined and unspecific water matrix in laboratory testing.

The knowledge obtained through this dissertation increases the chance of closing the water loops with less impact on operating performance. Additionally, it challenges our theoretical understanding of mining water management by introducing a focus on the lack of tools and mindset for dealing with process water in the context of the operation.