Fuel Cell Vehicles And Critical Metals: Supply And Demand

It is fuel-cell-vehicle (FCV) season again as many of the world’s premier car makers make their annual ritual announcement that they are ‘studying’ or putting into ‘limited production’ passenger-carrying vehicles for personal use (i.e. cars), propelled by electricity generated by ‘fuel cells.’

Once again, the perception of greeniosity is meant to trick us into thinking that the fundamental laws of economics have been suspended.

As far as I can determine, the electricity for FCVs will be generated when diatomic hydrogen molecules are split into hydrogen ions and free electrons, by the action of passing the hydrogen over a catalyst. This previous sentence is totally intelligible to a chemical engineer with the only undefined word in it being ‘catalyst.’

As far as I know the only such ‘practical’ catalysts known for such a reaction are the platinum-group metals (PGMs), primarily the metal palladium (Pd). There has been a lot of research over the last 20 years on trying to produce a fuel-cell chemistry based on a more readily available catalyst than a PGM but the results have not been economical. One such program backed by no less than Kleiner Perkins is for a Solid Oxide Fuel Cell (SOFC), which uses the extremely scarce rare-earth-element (REE) related metal scandium (Sc) in its catalyst.

The thing that all current fuel-cell technologies have in common, is that they rely for their operation on large amounts of very scarce materials such as PGMs or Sc, as in the discussion above.

There is another problem, the relative value to achieving the goal of reducing carbon emissions of a FCVm versus an internal combustion engine (ICE) vehicle, using a catalytic converter. This is the real issue of the most efficient use of strategic metals. Let’s say that a Pd-based fuel cell would use at least one ounce of Pd in order to be able to produce enough electricity to power a four-passenger car. That same amount of Pd could be used to manufacture 100 exhaust-emission catalytic converters, for hydrocarbon-fueled ICE-powered vehicles! Note well, that new global production of Pd is in the 200 tons per year range. This is twice what it was 10 years ago, but nearly impossible to increase as most of the world’s new Pd comes from its production as a byproduct of nickel mining in Russia and Canada, with a little more coming from South African platinum mining. North America produces some 10% in total of the world’s annual new Pd. It is difficult to see how green technologists could ask us to depend on either Russia or South Africa for an ‘assured supply’ of anything much less for an increased supply.

So, the best solution for constructing fuel cells is not to use environmentally precious Pd or any other PGM in such a horribly wasteful way. Unfortunately, the best SOFC, based on Sc, is an even worse solution. There simply is not enough Sc produced in the world. Currently just a few tons a year are produced, so it is believed, in the former Soviet Union.

So we can either rob Peter or mine an empty bank vault.

There is a real analogy here to the REE supply issue now facing the world, and even an interface, since Sc is only likely ever to be produced as a byproduct of REE production (which itself is ironically usually produced as a byproduct of iron mining).

PGMs used in automotive-exhaust emission control devices (catalytic converters) are so scarce as to be among the most recycled materials on the planet. In relative-percentage-recycled terms they are right up there with iron, copper, aluminum, lead, and gold. But it is in absolute terms that the comparison fails. An excellent example of this is the PGM rhodium (Rh), used to eliminate acid-forming nitrogen oxides from automotive ICE exhaust. The world production of new Rh as a byproduct of South African platinum production is 30 tons a year. Yet the apparent demand from the global OEM automotive industry is nearly 50 tons per year. This additional material must come from the extensive recycling of catalytic converters.

It is the same type of thing with the REEs with a notable exception geographically. In China extensive recycling of REE industrial process waste as well as of end-of-life waste, is one of three things that keeps the supply of the key heavy REEs terbium and dysprosium, nearly equal to the demand. The others are illegal production within China and purchase of heavy REE ore concentrates from outside of China. The three processes together provide a doubling of ‘official’ production of these key REEs.

Only now in 2014 is there even the beginning of a non-Chinese REE recycling industry. This is because with just one exception, there is no REE separation plant outside of China with the capability/capacity to separate the heavy REEs from ore concentrates or scrap; there are 38 such facilities in China.

What little Sc is produced in the world may be augmented by the three processes above, but officially there is no verifiable Sc production anywhere. So, if there is to be a fuel-cell-powered OEM automotive power-train revolution, it will have to be itself driven by a fuel-cell technology that as of now is unproven, and does not involve a need for large quantities of either PGMs or Sc.

At the moment, supplies of PGMs and Sc globally are either insufficient or unavailable. Thus fuel-cell-powered vehicles will be curiosities, or the toys of the elites, for the foreseeable future.