RichKen - that's a very nice summary and makes the point that grid battery storage is going to be needed even before you start factoring in the uptake in renewables.
To put some real numbers on this - the recent report by the US National Renewable Energy Laboratory ( https://www.nrel.gov/docs/fy19osti/74184.pdf ) shows that there's 28 GW of potential gas peaker plant replacement that would be possible right now, at an average US PV penetration of only 2.5%. This corresponds to your point number 2.
This is 28 GW of 4 Hour storage - i.e. 112 GWh of long duration energy storage.
If you consider building 6 hour and 8 hour storage then this number rises to 70 GW of gas peaker plants that can be displaced - this is something like 420 GWh of Energy Storage and this is the situation TODAY.
Once PV penetration rises to 10% and your point 3 kicks in, this figure rises to 100 GW - that would be something like 600 GWh of long duration energy storage.
The only type of energy storage that is currently specifically incentivised are the frequency regulation applications (your point 4) - as the report itself states "(28GW) - This about four times greater than the entire regulating reserve requirement in the United States (Denholm, Sun, and Mai 2019)." - thus the frequency regulation market, the only one currently entertaining our lithium-ion friends, is only about 7GW.
I'll keep my eyes peeled for some useful assessments of the size of the transmission deferment market (your point 1). I expect it to be of the same order as the 100's of GWh mentioned above. This of course is just for the United States.
Furthemore I believe that their analysis is flawed as I suspect that it does not take into account the key issue of curtailment - namely discarding excess energy that some sort of renewables generator has produced for you but is really too costly to store for later. This is ok because it hasn't cost you any fuel to make that energy and it hasn't produced any CO2.
If you read this paper you will see just how important it is to consider this possibility (and especially so as you get over 65% renewables penetration) as it massively reduces the total amount of grid storage that is required as part of a lowest cost overall system.
The cost threshold in that report are complete pie in the sky. It's like say yeah I'll buy an aston martin vantage DB5 if only they cut the price to less than 10 grand.
The rollout of energy storage has got absolutely nothing to do with hitting those 'targets' - it is happening already at prices 10 times higher than the numbers they've dreamt up and it's not charities that are doing the buying !
Furthermore that USGS report starts out:-
Vanadium is used primarily in the production of steel
alloys; as a catalyst for the chemical industry; in the making of
ceramics, glasses, and pigments; and in vanadium redox-flow
batteries (VRBs) for large-scale storage of electricity. World
vanadium resources in 2012 were estimated to be 63 million
metric tons, which include about 14 million metric tons of
reserves. The majority of the vanadium produced in 2012
was from China, Russia, and South Africa.
It also makes zero mention of either Vametco or Mokopane, probably because their data on South African Vanadium miners comes from this 1975 paper ( https://pubs.usgs.gov/pp/0926a/report.pdf )
I think we can conclude that we know a little more about South African Vanadium mining that this USGS report
Indeed - Largo's measured resource is 1.24% (total rock concentration) and BMN's Mokopane indicated resource is 1.32% (total rock concentration).
So the Speewah deposit at 0.3% total rock concentration of V2O5 may be large but it is very low grade. Which means it is staying in the ground.
Ninvestor - the more I heard of the interview the more it became clear that it was Mr Mickey Fulp that had missed the VRFB + Vanadium boat . He clearly has very little real idea of VRFB technology, claiming that they have a 'massive' footprint and can only be made a 'football' field (not 'soccer') scales.
numpty - V2O5 is sold by pound of material. FeV and Nitrovan are sold by the amount of contained Vanadium atoms.
Each Vanadium atom in Nitrovan costs c. 5% (maybe more now) more than a Vanadium atom in FeV. A Vanadium atom in FeV also cost slightly more than a Vanadium atom in V2O5 because you can chuck it straight into your steel furnace.
All you need to understand is that 4 years ago Ophidian and BigBiteNow had a long discussion here about which was the best brownfield processing plant in South Africa for BMN to buy - Vametco or Vanchem ?
Nobody ever imagined in their wildest dreams that we'd be able to snap up both of them. With very little dilution, and perhaps very little debt. At a stroke we put ourselves on the path to producing 10,000 tonnes of Vanadium per year (thats the actual amount of vanadium contained in our products, so if it was all in V2O5 form it would be 19,000 tonnes of V2O5) - that's equal to the total production of BOTH Glencore Rhovan and Largo combined , the only other two Vanadium producers in the world outside of China.