February 23, 2009

Could we run out of lithium for EV batteries?

Lithium is reasonably abundant, and reasonably widely distributed: it's mostly produced now in S. America, but China is expanding production, and there are substantial sources elsewhere. It can be recycled efficiently.

It's rather like uranium: in the short run there could be boom-bust cycles of supply expansion and shortfalls, but in the medium-term there aren't really resource limits.

There was a widely read analysis a couple of years ago that raised questions (The Trouble with Lithium: Implications of Future PHEV Production for Lithium Demand, William Tahil, Research Director, Meridian International Research, January 2007 http://www.meridian-int-res.com/Projects/Lithium_Problem_2.pdf ) but those questions have been answered pretty thoroughly. The amount used by each battery isn't that great: one estimate is that most lithium chemistries require around 3+lb/kWh of lithium carbonate, so for a 16KWH Volt type battery we would need about 50 lbs of lithium carbonate (or about 0.3kg of Lithium metal equivalent per kWh, per Tahil). At that level, there's more than enough lithium (see reference below). In the short term, battery producers are very experienced at this sort of thing - for instance GM is assembling the Chevy Volt battery from cells made by LG Chem, the largest li-ion cell producer in the world - I suspect LG is pretty good at getting long-term contracts for their supplies.

At $2.75/lb, that's only $137.50, or 3.4% of the likely Volt battery cost of $4k (wholesale in 2-4 years). A doubling in the price of lithium would only increase the cost of a $30K vehicle (after $7,500 credit) by $137.50.

Here's another good general discusion. If you want a more detailed discussion look here and here, and for some debate go here. A study by the Dept. of Energy's Argonne Lab here said "Known Lithium reserves could meet world demand to 2050".

What about recycling?

Well, according to this, lithium is so cheap currently that it hardly pays, but in a sign of unusual foresightedness, lithium recycling is being put in place. This site indicates that li-ion battery recycling is widely available.

Are lithium-ion batteries, like those in GM's Volt or the Nissan Leaf, unsafe?

No. They're using newer chemistries which are more stable than the the cobalt-based chemistry in laptops or the Tesla. A123systems iron-phosphate chemistry is very stable. Others, like the manganese-spinel LG chemistry is significantly different, and safer.

Toyota has questioned the supply of lithium. Do they know something?

No. The industry as a whole, including GM, Nissan, and Honda are banking on lithium batteries. They also are pretty good at sourcing supplies.

As far as Toyota is concerned - sadly, this is part of a pattern of dishonesty. Several years ago they committed to 1st gen li-ion (cobalt) from a supplier in their Keiretsu, and then Toyota had some Q/A PR problems, and 1st gen li-ion had thermal runaway (fire) problems, and Toyota became nervous. They decided to go w/2nd-gen li-ion, but were caught without a good supplier. They're "dissing" li-ion until they can get their act together, and on the road. This is similar to their dishonest "dissing" of competitors to the Prius, especially GM's Volt (here's an example).

You talked about short-term supply problems. Couldn't these be a problem?

Yes, and they already are. Prius and other hybrid production has been slowed down by NIMH battery shortages, and it looks to me like the same is true for li-ion batteries. OTOH, we shouldn't exaggerate the problem: the supplies are out there, and these large companies are very good at solving these problems over time. On the 3rd hand, in the case of a fuel emergency, it might be difficult to ramp up battery production overnight - we'll need contingency plans for the interim.

Uranium? You mentioned Uranium - is there a question about Uranium supply?

No, my reading doesn't support that. In the short run there certainly could be modest boom-bust cycles of supply expansion and shortfalls, but in the medium-term there aren't really resource limits.

I haven't really seen a definitive resolution of the question, but it looks to me like there are too many alternative sources of uranium, including weapons recycling, reprocessing and expansion of existing mines (including mines in the US, at substantially higher costs, of course), for us to have an absolute shortage.

Here's an example of the kind of change that might happen: "It is also relevant to note there that, as discussed above, enrichment capacity can to some extent be used to produce additional uranium supply, by operating enrichment plants with a lower U-235 assay in the tailings stream. This means that utilities can reduce their uranium demand by 10% or more provided they have access to sufficient enrichment capacity at a price which makes this economic (i.e. provided it is less expensive to buy more enrichment than to buy more uranium). Furthermore. so long as they have surplus capacity, enrichment Plant operators can physically operate their plants at lower tailings assays than that specified in contracts with utilities, effectively producing additional uranium (which they can then sell in the market). This is always likely to be an attractive option for enrichment plant operators. as their marginal costs of production will normally be less than the price paid by utilities for enrichment services." From Nuclear Development Market Competition in the Nuclear Industry: Nuclear DevelopmentBy OECD, Ad Hoc Expert Group on Market Competition in the Nuclear Industry page 60-61

It would be nice to have a really clearcut, definitive answer, but uranium supply looks nothing like oil to me. Oil flows into reservoirs. Where there are no reservoirs it's lost forever, and where there is a reservoir you have a pool with fairly defined edges, the edge of which you can hit relatively abruptly.

That's very different from uranium which is much more abundant relative to consumption, much more widely distributed, with ore-quality distribution that is much more uniform than oil.

There's a spirited discussion here.

No comments: