Executive Summary

PROJECT HIGHLIGHTS

  1. Salinity and pH adjustments do not greatly increase fines agglomeration in a pumped system.

  2. Froth flotation grades generally increase at low salinities.

  3. Slime coatings on sulfides are less likely at low salinities.

  4. An unoxidised sulfide surface is less sensitive to changes in salinity than a gangue surface.


The aim of this project was to modify the surface charge (zeta potential) of fine pyrite, chalcopyrite and pentlandite particles in an attempt to selectively agglomerate them. The primary conditions that were adjusted were the salinity and pH. As part of this work an extensive review of the flotation of fine sulfide particles (less than 20 μm) was conducted. Several potential fines flotation methods were identified, including selective flocculation/aggregation, carrier flotation, use of smaller bubbles, high intensity conditioning (HIC)/shear aggregation, the use of inert grinding media, and separate flotation circuits for fines.

The splitting of fine material into a separate circuit and optimising the conditions for the flotation of valuable fines is currently the only proven method in practice. The grades and recoveries of this fine material can be high when conditions are optimised and inert grinding media are used. There is also some evidence suggesting HIC and some magnetic methods may aid in the recovery of valuable fine material. Fine gangue entrainment can be a problem with any fines flotation, so an efficient method of removal of fine gangue is required to prevent poor grades. The other methods such as flocculation/aggregation, carrier flotation and the use of smaller bubbles have yet to be successfully implemented for the specific purpose of increasing fines recovery.

The froth flotation results indicate that a positive zeta potential is typically associated with lower grades and potentially lower recoveries. The two factors which had the largest influence on the zeta potential were the sulfur content of the ore and the salinity of the process water. Ores with a high sulfur content had negatively charged particles that remained relatively unchanged with salinity. Ores with a low sulfur content tended to have particles that were less negatively charged at low salinity and positively charged at high salinity. A positive zeta potential for an ore is an indicator of a high amount of gangue material. The positively charged gangue material could form a slime coating on the negatively charged sulfide particles and reduce grades and recoveries.

Oxidation of sulfide minerals usually creates a superficial layer of metal hydroxides on particles due to hydrolysis of the cations. Metal oxide/hydroxides on the sulfide minerals result in the surfaces behaving more like metal oxide/hydroxide surfaces and this can lower the recovery by flotation. The zeta potentials shift from the unoxidised sulfide values towards those of the metal oxide/hydroxide products. These differences, between oxidised and unoxidised sulfide zeta potentials, increased with salinity. The zeta potential was affected to a much greater degree by oxidation of the sulfide than by precipitation of metal hydroxide species.

Regardless of the specific sulfide ore, salinity or pH, the zeta potentials were all less than 20 mV in magnitude. Particles with zeta potentials in this range tend to agglomerate but our particle sizing measurements did not show agglomeration. This may have been due to the break up of agglomerates during pumping in the sizing instrument, which would suggest that agglomerates formed by reducing the surface charge were weakly bound and easily broken up.

In general the viscosities, elasticities and yield stress values for the sulfide slurries did not change significantly between different ore samples, with pH or with changes in salinity. In some cases where the zeta potential magnitude was minimised these rheological values did increase. These values also increased at points where precipitation of a metal hydroxide was suspected. The small but significant changes in rheology which occurred when a metal hydroxide was forming were useful as a sensitive indicator of changes occurring within the slurry.

Multivalent ions, humate, sodium ethyl xanthate and potassium amyl xanthate did not significantly affect the zeta potential of the unoxidised sulfides. For the multivalent cations any effect was small and restricted to high pH where metal hydroxide species were likely to form. The only added species that had a large effect on the pure sulfide zeta potential was the quaternary ammonium compound cetyltrimethylammonium bromide (CTAB). The addition of CTAB had a significant effect on the zeta potentials of the sulfides shifting them to more positive values. The viscosity, elasticity and yield all increased as the zeta potential approached a value of zero. Although the zeta potential of the sulfides was minimised there was no evidence of any increase in agglomeration for the fine sulfides.


from: “Fine Particle Agglomeration in Process Slurries” MERIWA Report 295, Emmanuel Karakyriakos and Vincent A. Patrick, June 2012. ISBN 192098156X.