This post is going to be quite simplistic so I apologise upfront.
@8horse in your post this comment is made:
"
A critical difference is the iron content. The iron content of a Talison spodumene concentrate is lower than the other concentrates which are expected to have iron contents in the range of 1.5% - 1.9%."
Firstly I don't believe the 1.5% - 1.9% Fe203 comment above, but it is clear the Fe203 content for the others is coming in around the 1.2% Fe203 mark (but recovery rates are lower than forecast for them). And I think there is a link here (see below)
By way of background, and taken from a post I wrote a while ago, in producing 6% grade spodumene concentrate in a simplistic form explanation you have essentially a floatation process in effect for producing spodumene concentrate and then, for example, a roasting process at 1050 degrees celsius to produce lithium carbonate (i.e. a bit more complicated than that but want to keep it simple). These processes themselves are about reducing the iron/impurities in your ore so that you have battery grade product for the EV market. Now the more you reduce the impurities in your spodumene concentrate process the better price you receive, i.e. just because a deposit say grades at head ore say 1% Fe203 doesn't mean your product sale will be that because the process flowsheet in spodumene is aimed at producing your concentrate but removing your impurities, an obvious point -
https://www.sgs.com/~/media/Global/Documents/Flyers%20and%20Leaflets/SGS-MIN-WA109-Hard-Rock-Lithium-Processing-EN-11.pdfThe more impurities you have in the spodumene concentrate in the roasting process, you will need to extract these impurities out but you also extract out lithium, a problem in itself. Impurities forms 'clinkers', and then you need to run a process to get the lithium out of the clinkers and you might only get 75% of that from the clinkers, and that affects the recovery of lithium overall in the roasting process and impacts the ability to produce battery grade carbonate. Best explained in this article:
https://www.linkedin.com/pulse/all-chemical-grade-spodumene-concentrates-same-harman-grant/Going back to my earlier post on this thread itself, what I implied is the MET tests used by AVZ for producing the 6% plus grade spodumene were essentially 'base level' or they came across as such (1st pass over the target), and that process got them to a decent value to start with, meaning it could be improved through optimisation techniques (so will be interesting how the new MET tests pan out). For the others,like say PLS, not clear what they did in their MET tests.
In terms of the upcoming MET tests for AVZ the key is understanding the process used and if that isn't explained then it is essentially a guessing game as to how the MET test process translates to costs and therefore an NPV in a 'real' producing project ontext. The only thing I will say is the simplified MET process in the previous Ann already produced a product with Fe203 content below 1.4% and a concentrate level above 6% which in effect you would expect with some more cost they would further reduce that Fe203 content and impurities whilst maintaining the concentrate level above 6%
without impacting the recovery rate assumption (an assumption by me). In my post above I said throw enough money in your MET tests you will achieve low Fe203 count and a good grade spodumene concentrate but that might not translate to what you get when you build your plant (and that would be reflective in the cost structures and recovery rates in your operational plant compared to your DFS). I am not saying that this is the issue you are seeing with say PLS/GXY/AJM, but the point is project ramp up for them has gone now for almost 1 year, and GXY is beyond project ramp up, but yet recovery rates still remain lower and opex costs higher than their respective DFS. Obviously PLS is putting some more capex spend in to increase recovery rates and hopefully reduce opex costs but the question is whether they will achieve their DFS goals around particularly recovery rates in the spodumene concentrate process. And the longer it takes the less sure I am they will but I suspect they will at some point in time - not a problem provided the spodumene price is right.
BTW, this post is not about the other but to explain a concept around ore grade, MET tests, spodumene process and recovery rates.Take two deposits, one deposit has a Li20 head grade of 1.6%, the other 1.0% Li20, but both have the same level of phosphorous, fluorine and Fe203 impurities in the ore. Just take Fe203 of 1%for now to illustrate a point I am rambling on but yet making because I haven't got a beer in my hand, for each deposit here to illustrate what grade means in a process. If your deposit grades 1.6% Li20 that means you need 4.7 tonnes of ore to produce 1 tonne 6% grade spodumene at an 80% recovery rate - but in doing that you have to get rid of an accumulated 4.7 units (1% * 4.7 and yes I am now converting percent to units to make a point) of Fe203 to get down to that one tonne spodumene concentrate, and obviously you can't do that in the spodumene concentrate process per se (i.e. get to zero as that will never happen in a spodumene making process) but you need to do enough to get Fe203 down to say 1%Fe203 (in this case remove 3.7 units) but still get to a spodumene concentrate of 6% at 1% Fe203 (thus allowing the offshore converters in the roasting processes to do their thing in say producing lithium carbonate and yes I know product likely destined for the hydroxide market etc etc).
If your deposit grades 1% Li20 you need 7.5 tonnes of ore to get to your 6% grade spodumene concentrate at a 80% recovery rate (but in a cummulatative total you now have to reduce the 7.5 units accumulated (7.5 tonnes ore * 1% Fe203) to produce that, but obviously you can't remove all of it but still need to remove in effect 6.5 units to ultimately get spodumene concentrate of 6% with Fe203 of 1.0%. If you can't do that try to target at least 1.5% Fe203.
In other words, using the same recovery rate of 80% you need to remove 43% more units of Fe203 in the example above for a deposit grading 1% Li20 compared to a deposit grading 1.6% Li20 (i.e. 3.7 units/6.5 units - 1), hence the key around grade, to produce one tonne 6% spodumene concentrate. It is a very very simplistic explanation I am giving above but this same basis applies to the other impurities, impurities that can be difficult to take out of a spodumene making process and carbonate making process (and obviously hydroxide making process) and obviously have a cost in removing them.
So if you have a lower grade deposit there is likelihood that you will not achieve your recovery rates ( 80% above) and/or you have a higher cost structure going from MET tests to operation (based on what we are seeing with some of them). The reason for a lower recovery rate is also in part IMO IMO IMO attributed to the fact at some point you cannot throw more money into the spodumene making process without actually impacting the quality of the spodumene itself and that is a balancing act as well between quality and grade (given the heating process in the carbonate making process is a key to iron removal per se as well, but above I am just talking spodumene making process). And in the carbonate process the more Fe203 you have there is a possibility of 'clinkers forming' and hence that can impact the prices paid to spodumene producers - see above link - hence the round robbin back to why spodumene producers need to provide a spodumene concentrate with as little rubbish in it.
Noting, in particular when someone says 6% grade spodumene, what they are saying is that 75% of that tonne is spodumene the rest needs to be removed and in this 'rest' want as little deleterious elements like iron as possible as the rest above the deleterious elements can be treated ok by offshore converters. If can produce a 6.5% spodumene concentrate in effect saying 81% of that tonne is spodumene hence the further benefit for offshore converters of selling product at above 6% grade.
The articles below provide some relative information as well.
http://www.meadegunnell.com/assets/PDF/17-003LithiumHardrock.pdfWrote a lot to say little - should have drunk a VB or two so as to be more succinct and less of a polly waffle. Need to beer.
All IMO IMO
