There can be no disagreement that S32's GEMCO operation is world-class. It is a significant contributor to S32's A$19b market cap. If you were to read Rocket's descriptions, you would get the impression that they have 44% ore just sitting in the ground and DSO ship it.
The extract below is GEMCO's ore processing flowsheet. After crushing and screening the ore, GEMCO's flowsheet involves a drum scrubber to remove clay (but less clay than E25!!). They then use DSM to separate the Mn from waste ore. Ferrosilicon is the material used by the DMS process to increase the water density to the level required so that Mn is sinking while the waste rock/unwashed clay just floats away. With DMS, E25 would basically have the same flowsheet as GEMCO. While GEMCO needs to deal with 4:1 and above strip ratios to get down to its ore, E25 has a strip ratio around 0.3:1 because they are extracting basically from surface with no significant overburden layer. While E25 has its ore extremely close to processing facilities, GEMCO is needing to use a fleet of Cat777's to get the ore from mine site to concentrator. So to mine gate, E25 (with DMS) and GEMCO would be running the same operation. GEMCO has more internal trucking to is production sites and has more waste rock to shift to get to their ore seams. They however benefit from dividing the end product by 44% rather than perhaps 32-33% in determining the cost per DMTU. This extra grade (and price for that grade) covers these additional costs that they have per unit of end rock product. There is a reason to believe E25 could have a low-cost operation to mine gate.
The lithium and Graphite JORC statements I've seen normally list a volume and grade. S32's annual report JORC commentary is a bit different and notes a deposit size, grade and yield. This yield is 47-48%. What I think is happening is S32 could have presented the Measured resource as being 21.5% grade. The indicated resource could have been 19.2% grade. They instead note the final grade expected to be achieved and what this represents as a yield from the rock/clay captured by the Mt's referenced. Everyone focuses on the 44.8% and 40.9%, not the 48% and 47% yield. Note - I'm not 100% certain on this interpretation but the underlying maths would be 44.8%*48%=21.5% and 40.9%*47%=19.2%.
A few annual reports ago, GEMCO was reporting its strip ratio. 2019 was 4.5:1. If this is circa 22% ore, GEMCO is needing to shift about 9T of rock for 1T of final product. E25 has a strip ratio of 0.3:1 and will need to shift around 3.2x re beneficiation. If operating properly E25 would need to shift about 4.2T of rock for 1T of final product (10% starting, 32% finishing). Needing to shift approximately half the rock GEMCO does is a factor in why the operations to mine gate can have a believable low price - if operating efficiently.
(20min delay)