The shale horizon is of limited extent and has only been traced for a short distance to the south of Kgwakgwe Hill where it apparently terminates in the superficial deposits which constituted the most southerly of the historical mine workings. The shale also thins out rapidly to the west and to the north of the historical mine area where it appears to transgress over the quartzites and therefore, lies directly on the felsites.

The stratigraphy has been duplicated by thrusting in places. Small displacements have also been observed along steeply dipping east-northeast trending faults. The shales have been intensely folded and slumped in the vicinity of these dislocations and in addition, subparallel breccia zones and quartz veining may be evident.

The entire Transvaal package is cut out against the Waterberg sediments to the west of the Kgwakgwe Hill along what are thought to be northerly trending faults.

The manganese shale outcrops along the northerly scarp slope of the Kgwakgwe Hill and dips into the hill. The strata at K-Hill dip gently towards the northwest at an average of approximately 5 - 10° and small-scale S and Z folding is common. In plan, the manganese shale unit is more or less kidney shaped. The unit varies between approximately 2 m and 12 m thick with an average thickness in the order of 5 m and has a delineated extent of approximately 800 m N-S and 400 m E-W.

The outcrop defining the limit of the manganese shale is well defined, with the areas to the west being partially covered by alluvium and/or exposed by historical surface mining into the hillside.

The manganese mineralization occurs as a supergene enriched manganiferous shale (the Mn-Shale) within the Black Reef Quartzite Formation.

The Mn-Shale itself appears to represent primary manganese deposition (proto-ore) in a shallow marine basin as per the Canon and Force model. However, as is typical for most manganese deposits, there is clear evidence of upgrading by means of supergene enrichment.
This evidence includes:
• Only manganese oxide mineralogy has been noted to date;
• Observed manganese mineral textures are consistent with secondary precipitation; and
• The presence of fine manganese wad and the presence of cavities and vugs intersected in the drill holes in the Mn-Shale.

There are two possible time intervals during which weathering and supergene enrichment could have occurred, these being the recent period of exposure as well as the ancient period of exposure associated with the unconformity at the base of the Chert Breccia. If supergene enrichment is only related to the current exposure then the enrichment could be limited in extent by the current geomorphology and may only extend under the hill to the limit of weathering. However, if there was supergene enrichment during the period associated with the unconformity, then more extensive supergene enrichment may have occurred. The latter possibility appears to be supported by the drillhole intersections. Two mineralising processes are evident, namely an initial phase of mineralization by precipitation and diagenesis during sedimentation (forming the Mn-Shale or protoore) followed by one (and possibly two) phases of redistribution and concentration during weathering and supergene enrichment.

The envisaged mining method for the K-Hill Project is traditional open pit truck and shovel. Due to the low processing throughput, and reasonable strip ratio, the volume of total material moved (TMM) is easily manageable.

The overall inter-ramp slope angle used was 410. To accommodate small equipment and reduce dilution, mining is done on a 4-metre bench height, mined in 2 flitches of 2 metres. The face angle is 700 based on observations in the field. This gives a standard berm width of 3.2 m. No catch berms are considered necessary due to the relatively low height of the permanent walls. In most walls, the slope angle is considerably lower than the recommended inter-ramp angle, due to the placement of the ramps on the footwall, going into the hillside.

The pit optimisation shells indicate that the pits will have relatively a large amount of room on the pit floor with good access. Ramps are relatively short and in most instances two ramps can be accommodated without adding too much waste, thereby allowing for a drive-in ramp and a drive-out ramp. Considering the size of the equipment, such as the Caterpillar 725C2 articulated trucks (or equivalent), a ramp width of 10 m was chosen.

The clearing radius of the Caterpillar 725C2 is 8.05 m, and this has been taken into account in the pit designs.

The Waste Dump design has been done on industry standard parameters, using 370 face angles, and an overall slope of 230. Bench heights used were 20 m, and ramps were created at 30 m wide to allow easy access and reduce risk of collision. The Waste Rock Dump (WRD) was designed on the required waste dumping volumes, assuming a 30% swell factor.

Initial strategic schedules using the Milawa Balanced option on an annual basis indicated that a mining limit of 2 Mtpa should be enough to keep the plant full. However, closer analysis of mining on a yearly basis, as well as the increased waste of the practical pit designs, revealed that accessing RoM does require some opening-up of the pit floor and for this reason the final practical schedule mines more waste tonnes in the early years. Year 1 uses 3 months of prestripping and Year 2 has a peak total material movement of 2.02 Mt.

The schedule shows a stripping hurdle in the first couple of years of mining, but this can easily be overcome by adding an additional truck for the start up. An effort was made to smooth the stripping as much as possible, but mining into a hill made this somewhat difficult. It is anticipated that the next update to the resource model, which should see the two separate pits merge into a single larger pit, will make it easier to plan the stripping.

The comminution circuit will consist of several stages of crushing and grinding to achieve the target grind size, which is a P80 of 200 µm subject to further optimisation.

In order to preserve the water balance, a solid / liquid separation is likely to be required between the comminution and leaching stages. Given the relatively coarse target grind size, a vacuum belt filter is envisaged without prior thickening. Filtrate would be recycled internally within the grinding circuit.

Leaching
Leaching will be undertaken in a series of open topped tanks. The testwork indicates a total leach residence time of two hours. Filtered solids from the comminution circuit will be mixed with barren electrolyte returned from the EW stage, with reagent sulphuric acid added to meet the target acid strength (260 g/l based on the test work), as well as the reductant sucrose, which is consumed during the leach reaction. An on-site boiler will be used to generate the steam required to maintain the target leach reaction temperature of 90 °C.

Liquid/Solid Separation
As with the dewatering stage ahead of leaching, a vacuum belt filter is envisaged for the solid /liquid separation between leaching and SX. This stage will incorporate a cake washing stage, in order to both maximum soluble Mn recovery, and to minimise the residual acid content of the filter cake. The resulting washed solids will be suitable for conveying and dry stacking.

Solvent Extraction< ........