Thursday, March 15, 2018

The View from Les Houches: Of Rare Metals and Cute Kittens

Les Houches, March 2018. José Halloy of the Université Paris Diderot discusses mineral depletion in his presentation. Note how he utilizes Hubbert curves to estimate the trajectory of mineral extraction. He predicted that the dearth of very rare elements will negatively affect the electronics industry, perhaps killing it completely.

José Halloy's presentation at the Les Houches school of physics was focused on the availability of rare minerals for electronics. This is a problem that's rarely discussed outside the specialized world of the "catastrophists", that is of those who think that mineral supply may be strongly restricted by depletion in a non-remote future. In this field, Halloy seemed to side with the "hard" catastrophists, that is expressing the option that depletion will make certain things, perhaps even the whole electronics industry, impossible.

The problem, indeed, is there: modern electronics is based on the unrestricted use of very rare minerals - the term "very rare" indicates those elements which are present only in traces in the earth's crust and which, normally, do not form exploitable deposits of their own. If you pick up your smartphone, you probably know that it contains several of these very rare elements gallium (for the transistors), indium (for the screen), tantalum (for the condensers), gold (for the electric contacts) and more.

Most of these elements are "hitch-hikers" in the sense that they are produced as impurities extracted from the production of other elements: for instance, gallium is a byproduct of aluminum production. Whether we can continue to supply these elements to the electronic industry in the future depends on a host of factors, including whether we can continue to extract aluminum from its ores. In this sense, recycling is not a good thing since recycled aluminum, of course, does not contain gallium, because it has already been extracted during the refining phase. Note also that recycling tiny amount of very rare elements from electronic devices is extremely difficult and very costly. So, in the future, the supply of these elements is going to become problematic, to say the least.

Does it mean the end of electronics? José Halloy seemed to be very pessimistic in this sense, but I think the question was not posed in the correct way. If you ask whether current electronic devices can survive the future dearth or rare mineral, the answer is obvious: they can't. But the correct question is a different one: what kind of electronic devices can we build without these elements?

Here, I think we face a scarcely explored area. So far, the industry has been produced all kind of devices focusing solely on performance on the basis of the assumption that there aren't - and there won't ever be - mineral supply problems. Can we make a smartphone without gallium, indium and all the rest? That is, limiting the elements used to the basic ones, silicon, aluminum, and other common materials? It is a difficult question to answer because, really, it has never been addressed, so far.

Yet, I think there are excellent possibilities to develop a new generation of electronic devices which are both using very little (and perhaps zero) rare elements and which are designed for complete (or nearly complete) recycling. The basic element of all electronic circuits, transistors, can be made using silicon and, in general, there are alternatives to rare metals for most devices, even though in most cases not with the same performance. For instance, light emitting diodes (LEDs) are currently based on gallium nitride (GaN) and there seem to be no comparable substitutes. Without LED, we would have to go back to the old cathode ray tubes (CRTs) which we consider primitive today. But, after all,  CRTs performed well enough for us up to not many years ago. So, it would be an inconvenience, but not the end of the world.

So, it is clear that we'll have to settle on reduced performance if we want an electronics without rare elements, perhaps on a strongly reduced performance. But maybe we don't need the kind of performance we have been used to in order to keep going. Think about your smartphone: it is an incredibly complex and powerful device used mostly for trivial tasks such as looking at clips of cute kittens and sending likes and thumbs-up to other machines. Does "civilization" really need these devices? It is all to be seen.

For a fascinating discussion of an industrialized world running without rare metals, see the excellent book by Pierre Bihouix "L'age Des Low Tech" (in French - alas!)


  1. As always, thank you Ugo:

    I tend to think that electronics will have to revert to simpler systems. Being an old geezer, I can remember doing everything that I do now on an IBM PC back in 1985. I know for a fact that electronics can exist in a much simpler and less resource intensive manner.

    I feel that most of the ubiquitous gizmos that are being paraded out will not last beyond twenty or so years from now. I think that a regression to simpler, less flashy technologies will take place over that period.

    I would be more than happy reading your work on a smaller screen and block letters. You will have to lose the pictures, but it is the words that count. Broadband won't be required because the data coming at you will have to match the abilities of the hardware.

    What we have now is more marketing than anything else, bigger, better, faster cannot last. We will be fine, it will be different.

  2. China has almost monopoly on rare earth elements. There is possibility that it will use this monopoly against USA, after Tramp started trade war with China. More on this issue here:

  3. Ugo
    What are the restraints on optical fibre systems, including the undersea cables that span the world?

    1. Another subject which would need to be studied. There is nothing that an optical fiber system can do that a metal cable cannot do, but slower. And fibers do need LEDs, so it is all to be seen.

  4. It is worth mentioning that cathode ray tubes require some rare earth metals (which are not as rare as gallium and indium, but are environmentally destructive to extract), mainly yttrium, and small amounts of europium as a dopant. CRT displays themselves require considerably more power to operate than even fluorescent-backlit LCD. The circuitry itself is larger, and thus requires more material and embodied energy to manufacture. CRTs also produce X-rays, which requires the use of leaded glass to attenuate to safe levels. That lead can leach out of the glass once it's broken and in a landfill with acidic leachate. Fluorescent lamps with good color rendering generally also require rare earth phosphors. I believe white LEDs use considerably less phosphor per emitted lumen than fluorescent sources, but I would need to verify that.

    1. Bonjour Jon et bienvenue au club de ceux qui brisent les rêves de notre ami Ugo. Il nous restera donc le journal (uniquement avec de l'encre noire) les messages tambour et le télégraphe de Chappe (sémaphore). On va bien s'amuser.
      Jean-Luc Jourdain

  5. s/Pierre Bihouix/Philippe Bihouix/ :-)

  6. Non si può non essere d'accordo, anche senza avere la competenza tecnica e chimica di Ugo e di alcuni commentatori. La percezione che esagerare è sempre male fa parte della PRUDENZA, che non per caso è la prima delle virtù cardinali.

  7. Semiconductors really only need silicon, nitrogen, and phosphorus to be functional. One way to solve the LED problem is to get rid of visuals altogether and just rely on verbal input and output. Microphones and speakers are much simpler to build than displays.



Ugo Bardi is a member of the Club of Rome and the author of "Extracted: how the quest for mineral resources is plundering the Planet" (Chelsea Green 2014). His most recent book is "The Seneca Effect" (Springer 2017)