7. Conclusions and Recommendations

=230 microns, which will have a positive impact on the grinding energy requirements.

A QEMSCAN analysis revealed that only 85% of the Ni units are associated with sulphide minerals. The remaining Ni is included in serpentine and chlorite gangue minerals and, therefore, is considered nonrecoverable by means of flotation. The primary Ni sulphide mineral in the composites was millerite, which has the advantage of a higher Ni content compared to pentlandite (~65% Ni in millerite compared to ~35% Ni in pentlandite).

The rougher Ni recovery in the Nickel Zone composite ranged between 75-77% of total nickel at a saleable concentrate grade of 13-15% Ni. An overall Ni recovery of 77% in the rougher corresponds to a recovery of Ni in sulphides of more than 90%. A review of the QEMSCAN results suggests that the Ni in sulphide losses occur in form of very small grains of millerite and middlings/locked particles of millerite

and non-sulphide gangue minerals. Hence, even under optimized rougher flotation conditions the Ni rougher recovery is expected to remain below 80% of total nickel.

In order to take advantage of the very good initial rougher flotation response, a flowsheet was developed to recovery a high-grade Ni concentrate in the first rougher stage, before treating the remaining rougher concentrate in a regrind and cleaning circuit. This flowsheet was evaluated in test F9 and produced a concentrate grade of 20% Ni at a Ni recovery of 74% of total nickel.

The rougher tails of the Nickel Zone contained a considerable amount of carbonates, which render the tailings acid-neutralizing. If the high carbonate content is consistent throughout the mineralization, the tailings could be used as a cover material for an existing tailings facility.

The Nickel Zone composite contained a large amount of floatable non-sulphide gangue minerals that required the addition of CMC to depress the NSG minerals. A pre-float test was carried out, but yielded considerable Ni losses of 5%. Possibly a desliming stage prior to the rougher flotation may help to reduce the amount of CMC that is required in the flotation circuit.

The following recommendations are made for a future test program:

Review the end members within the deposit with a metallurgist and geologist to identify suitable composites that should be evaluated in the next phase of testing. The analysis would not only include the Ni head grade of the samples, but also the non-sulphide gangue minerals as they can have a significant impact on the flotation response of the mineralisation.

Carry out locked cycle tests to simulate a closed circuit operation. This would help to quantify the split of the Ni units in the intermediate process streams to the final concentrate and tailings;

Evaluate the stability of the proposed flowsheet on a number of variability composites to ensure that the first incremental rougher concentrate consistently produces a saleable concentrate;

Perform Bond ball mill grindability tests on selected composites at a coarser mesh of grind to take into account the relative coarse primary grind (recommended: 48 or 65 mesh). This will provide more accurate energy requirement values as samples are frequently sensitive to the mesh of grind;

Carry out preliminary solid-liquid separation tests on the tailings to quantify their settling and filtration properties;

Perform a detailed chemical analysis on the final concentrates and tailings to identify the