Peak aluminium?

The announcement that Rio Tinto is to close its alumina refinery at Gove struck me for a number of reasons, starting with the fact that members of my family are affected by it. First up it’s worth noticing what’s mentioned (the high dollar and low aluminium price, which flows through to bauxite and alumina) and what isn’t (the carbon tax and legislation for its removal). Having claimed that he was going to save industries like alumina and aluminium smelting from the carbon tax “wrecking ball”, Abbott is now shown up, once again, as a fraud[1].

In the short run, the obvious policy implication is that the RBA needs to be firmer in pushing the dollar down. It was, I think, a mistake to hose down talk of direct intervention, as was done recently. Given our declining terms of trade, we should be closer to $US0.80 than $US0.90 now, and heading down further.

The bigger question of interest, though, is the future of aluminium. The big story of the past 10-20 years has been the massive growth of production in China, driven by cheap coal-fired power and lots of subsidies. That’s driven prices down to historically low levels (inflation-adjusted, probably record lows). Production in Australia is now clearly uneconomic, but even the Chinese are losing billions.

Declining prices have driven steady growth in demand for aluminium. Since the supply of recycled aluminium is dependent on past production, there has been a multiplied effect on demand for primary aluminium, which is the big driver of greenhouse gas production in this industry.

The general assumption (as with most trends) has been that these trends will continue indefinitely. But it’s clear that prices have to rise just to cover costs, and will rise further as China starts to price the local and carbon costs of coal-fired electricity. Moreover, in technological terms, aluminium is definitely a 20th century commodity. Its inherent properties of lightness and strength gave it great advantages, but it is now being displaced in advanced uses by carbon fibre and in some basic uses by lightweight steels.

So, it seems to me quite plausible that aluminium demand could stabilise over the next decade or two, with the result that most demand can be met by recycling rather than energy-intensive production of primary aluminium from bauxite (via alumina).

Note: I topic-banned regular commenter Hermit from talking about aluminium smelters, as it become an idee fixe. The ban is lifted for this post.

fn1. Has any new PM ever been shown up so comprehensively in such a short time? Not in my memory, which goes back to Harold Holt, and includes some shockers.

76 thoughts on “Peak aluminium?

  1. Fran, an example: Congratulations! You have just inherited a coal power plant and attendant coal mine with an average output of 200 megawatts. The power plant is 33% efficient and like most Australian coal plants it only costs a few dollars a tonne to get the coal from the ground to the boilers. A tonne of coal when burned in an excess of oxygen produces 24 gigajoules of thermal energy which is then coverted into 8 gigajoules or 6,666 kilowatt-hours and a lot of waste heat. You can sell the electricity, after subtracting a pilsner weak carbon price, for an average of 5 cents a kilowatt-hour. The mine and coal plant are completely paid off and operations and maintenance come to about 1.5 cents a kilowatt-hour. This means that each year you run the plant, you personally, as the sole owner, will make over $60 million dollars from the sale of electricity. Now you might say that because the plant has an EROEI of less than 1:3 for producing electricity you will shut it down and leave it as a rusting monument to humanity’s folly, but would you really forgo all that money? Wouldn’t you be tempted to run it for at least a month or two just to save up a little nest egg and maybe buy a small chateau in France? And even if you’re ethical enough to close it down right away, how many other people who inherited this mine would be? Call me cynical but I think most people would keep the coal plant running, low EROEI be damned. Maybe they would give $10,000 to Philippine disaster victims or sponser a child or something to make themselves feel better.

  2. @Ronald Brak

    The “3” in the ratio is misleading.

    The energy in the coal is already in there. In the sense of EROEI it doesn’t come into play. The EROEI counts the energy required to excavate, process and move the coal into the furnace as against the useable energy that comes out the other end of the process (in this case, electricity. In the case of oil, the amount of work it does in ,say, a care engine).

  3. Fran, sorry, a tonne of black coal in a 33% efficient coal plant would produce about 2,200 kilowatt-hours, not over 6,600. Sorry about the mistake. It doesn’t change the point of my previous comment.

  4. Megan, are you happy with burning the burning of three tonnes of coal to heat a building with electrical resistance heating having the same EROEI as buring one tonne of coal to directly heat the building?

  5. @Ronald Brak

    It depends on how you define “happy”.

    I’m ‘happy’ to accept at face value the premise that a building could equally be heated by using the direct heat from burning 1 unit of coal or by the electricity produced from burning 3 units of coal.

    But in either case the coal doesn’t just turn up in the furnace all by itself. Energy is required to get to the point where some form of energy, in some amount, is useable.

  6. I certainly agree, Megan. Last time I checked pieces of coal didn’t have little arms to build coal plants with or little legs to walk inside once the job was done.

  7. @Ronald Brak

    You suggested that EROEI for coal is 1:3. It isn’t. It was pointed out that EROEI refers to the external energy inputs required to get an energy return from, in this case, coal. You moved the goal posts. I pointed out, again, that EROEI involves the energy inputs to get to the point of producing a useable energy return. You introduced the concept of bits of coal with little arms and legs. I, apparently incorrectly, took that as a concession wrapped in a churlish refusal to accept the point conceded. Hence: embarrassment.

    It doesn’t matter. The EROEI of Australian coal isn’t going to go away, unfortunately.

  8. I think I see the problem, Megan. You think we were arguing while I thought I was explaining something to you that you were having difficulty understanding. An argument is a connected series of statements intended to establish a proposition. If you tell me what your proposition is, and, very importantly, tell me what you think my proposition is, then provided we agree on what our propositions are and that the basis for an arguement exists, then I will gladly argue with you.

  9. Mixed feelings up here re Gove and alumina refnery.
    Some say good riddance the indigenous people never wanted it anyway.
    Unions say thay are looking at legal action as the AWU claims that, under a 2011 agreement, Rio Tinto is legally obligated to keep the refinery open if it is to mine bauxite at the site.
    ABC radio said yesterday that no, the agreement only states that the mine must remain open and working and does not refer to the refinery, therefore they can dig it up and ship it out holas bolas, news at 11.
    Most Gove locals are incensed with the mishandling by successive NT govts to lock up the gas supply/price. Doomsayers are out in force.
    Some more pro-active locals say good, now we can focus on some decent development i.e. sustainable eco-tourism.
    I have been to Gove a couple of times, the drive is absolutley awesome, some fantastic country, notably beautifull as you pass through arnhem land where no cattle have been allowed, lots of native grassess, and healthy timber. Couple of excellent river crossings, it is only about a 14 hr drive from Darwin, so pretty close really. When you get there, the best stuff is anywhere but the town really, great bush camping, great beach 4wd, great fishing, terrific people to meet, brilliant little outstations. And best of all you need to apply for a permit to go, so limited numbers of tourists really.

  10. @Martin W
    A Peak Oiler would say that oil fired alumina plants are simply the first to go in a long line. A couple of years later 14 hour drives for recreational purposes become out of the question. A couple of years after that bulldozers to dig bauxite and ships to take it away for processing get too expensive to run. That’s why they call aluminium frozen energy.

  11. The Bayer process generally used to process bauxite into alumina could use solar thermal as a source of heat. However, developing a method to use solar thermal for industrial processes like this would be very expensive, or at least it would be the first few times it is done. And solar thermal may well not turn out to be cheapest low emission option. Right now we are better off putting our efforts into building renewable capacity in Victoria and cleaning up one of the filthiest grids in the world. We can work out the best way to power a remote alumina refinery next week.

  12. @Hermit
    OK well the agreements between the NT govt and Rio Tinto go back a long way, the initial EIS stated very clearly that a transition from diesel to gas power generation was a requirement for future viability.
    Feb 2013 the then CLP leader, Terry Mills announced the deal had been done to committ and transpoert gas to Gove; “The Gove alumina refinery operated by Rio Tinto Ltd will stay open after the Northern Territory government pledged to release ten years’ worth of its contracted gas reserves to power the plant”, “The Commonwealth government has previously indicated it would be prepared to underwrite the $1.2 billion cost of building a new pipeline and other infrastructure to get the gas to Gove, which would deliver a boost to the Top End economy”.
    Fast forward to mid March 2013 and T Mills is dumped in favour of Adam Giles and then he decides to re-negotiate the entire deal; “Under the original March agreement with Rio Tinto subsidiary Pacific Aluminium, the Northern Territory government would have supplied some 300 petajoules (PJ) of gas over a 12-year period.
    However, Giles last week said that the gas deal, which was subject to a due diligence, contained “unacceptable risk” to the Northern Territory economy and its taxpayers, which would have faced a A$3.2-billion gas pricing risk. He also flagged the possibility of energy shortages in the long run.
    Instead, the state government is now proposing to sell Pacific Aluminium 13 PJ of gas a year, over a 15-year period, for a total of 195 PJ.
    “The alternative proposal gives longer-term certainty to the Gove community while protecting Territory taxpayers from billions of dollars in financial risk,” Giles said.
    Obviously the next complication with it all is the change of government federally, the NT simply does not have the tax and revenue generation base to do it alone, we continue to be dependent on financial support from the federal government and that is not going to change anytime soon.

    So it is pretty understandable why the community is unhappy with the CLP and Adam Giles, and of course Rio Tinto is protecting it’s own interests with little to no regard for the imapct on community.
    As per usual, absent from the debate is the Indigenous owners point of view, what they want and how they see it; par for the course really. Particularly in light of the fact that we are still suffering under the yoke of the intervention.

    A 14 hour drive into remarkable country with the best stuff and the end of the road is just one of the highlights of living up here, recreational yep, food for the soul yep. Come and check it out Mr hermit, be happy to show you around. 🙂

  13. RB I’m waiting for your calcs on how much solar it would take to run the night shift or wet season at Gove.
    MW I already live in the bush but 20km from a ski area. Dunno if I’ll ever get to the Top End.

    Way upthread I suggested unprocessed bauxite from Gove could be sent to China but for now that will be the alumina plants at Gladstone Qld which also process Weipa bauxite. However since Gladstone is getting into major LNG export their gas price will escalate as well. So China could get the bauxite after all. Those crazy Aussies with their carbon tax.

    For the 160C heat required for the Bayer process 24/7 I note some of the new mini-nukes on the drawing board are at aimed at heat rather than electricity. They will be too late to save jobs but note Ranger uranium is also up that way so the NT is not a nuclear virgin. As well as onsite diesel and fuel oil there are offsite CO2 factors such as caustic soda production. If that is powered by cheap grid electricity (chloralkali process) our good friend coal is never far away.

  14. Hermit, I started to work out how much solar power it would take to run Gove even though I said we could worry about it next week, but then I realised that you may not actually be asking for information. You may instead be arguing. So I thought for a while about what the proposition you might be arguing in favour of could be. You mentioned 160 degrees which is a temperature that is easily reached inside a solar cooker which is something that’s popular in a lot of developing countries and you mentioned nuclear power. So, if you are arguing, would the following be your position:

    Hermit’s proposition: Nuclear power can heat bauxite at a lower cost than large scale solar cooking in the Northern Territory.

    Is that correct?

  15. @Ronald Brak
    Correct but there is a catch. I’m not talking about just the middle of a cloudless day but a week with overcast rainy conditions as well as nights. Assume there is no oil or gas as a standby heat source. For comparison purposes assume a 200 MW thermal mini nuke costs a billion dollars or $5 per watt capex.

  16. Hermit, why assume that a 200 MW (electrical output) nuclear plant can be built for $1 billion when the estimated cost for Hinkley C which is to be built at an existing nuclear power site and in a country with an existing nuclear power infrastructure, is considerably more expensive per MW? If small reactors can be built at that cost why aren’t they building them in the UK and saving themselves over six billion dollars on the cost of Hinkley C?

  17. @Ronald Brak
    The first of the US approved small reactors won’t be ready til 2022 on the current pace. The Hinkley C units will be Areva EPRs of about 1600 Mwe I believe. Their average operating cost is expected to be low despite the high initial capex. Areva is heavily tied in with Electricite de France who own other British nukes and there may be some kickback to Chinese customers who are also putting up capital for the UK build. The Chinese seem to know how to build the same machines quicker. The capacity also had to be large to get the ‘strike price’ guarantee. Therefore in the medium term it was the big French units or nothing.

    The UK is also looking at US 4th generation reactors to burn up their plutonium stockpile. If they or the small US modular light water reactors become available ahead of time perhaps the Hinkley decision will seem slow and lumbering. Again the French have the leverage to slow it down. I don’t think small reactors will be used in ore processing in Australia we’ll just keep making excuses to burn coal or hand the work to China and India.

    It seems like that raw bauxite from Gove will be shipped to the Yarwun and QAL alumina refineries at Gladstone. Within a year or two they will do a dummy spit about the gas price. If Abbott is PM he’ll either quarantine some gas from LNG export or or hold the free market line. In the latter case I think a greatly reduced tonnage of raw bauxite will then go to China where the primary energy source will be trivially carbon taxed coal.

  18. Hermit, if you are going to assume what the cost of nuclear will be in the future, can I just assume that in the future solar thermal will cost less than half as much as nuclear? If not, why not? In fact, I think I’ll assume that solar PV will be installed for a dollar a watt by the time your small reactors are ready which means that using PV to generate electricity and then generating heat from that electricity would be cheaper than the cost you assumed for your small nuclear plant. (Passive thermal storage is both simple and cheap and not a problem.) Using PV to might seem an odd and inefficient way to generate heat compared to solar thermal, but at least it’s not as odd as building a nuclear reactor to process bauxite.

  19. @Ronald Brak
    You could assume we’ll all get around in jetpacks instead of cars. PV is down to $1/w as opposed to $19 (Ivanpah) or so for solar thermal with storage, some of which can (just) operate 24/7. However they will still need an oil or gas boost to cover rainy weeks such as in the NT wet season. BREE’s 2012 energy technology (AETA) report suggests (Table 4.25) solar thermal with storage could cost $200 per MWh by 2020. They have small nuclear (Table 4.38) costing $120 per Mwh.

    Yes it does seem strange to power mines with mini-nukes. If we can’t adjust our thinking we’ll be like meat eaters who don’t care to think of abattoirs. We’ll just get someone else to do the unseemly work for us. Assuming Gladstone gas gets too expensive to process Gove bauxite that means shipping raw bauxite to China. When we pay top dollar to import aluminium made from our ingredients our conscience will be clear.

  20. So Hermit, your proposition isn’t: Nuclear power can heat bauxite at a lower cost than large scale solar cooking in the Northern Territory.

    It is: Nuclear power can heat bauxite at a lower cost than large scale solar cooking in the Northern Territory in the future.

    Hermit, the operating life of many remote mining projects is often a decade or less. How can a nuclear plant be profitable when it will be abandonned in such a short period of time?

    Why would anyone build a nuclear reactor to provide heat for mining purposes at the cost you give when burning wood chips or other biomass would be a fraction of that amount?

  21. My impression is that the car companies still consider aluminium an important material to reduce the co2 emissions of cars. They are finally makeing a big effort to do so. In particular Audi mainly works with ever more sophisticated steal/aluminum combination. At least thats the story they tell in their public presentation at the Autostadt (ofc, most real life Audi are big fast gas guzzlers) . Only BMW is betting almost solely on huge fast progress with carbon. China btw, seems to do similar things in the carbon market, as in the aluminium market. It doesnt have to be that way. Carbon and aluminum could be produced only in places like Iceland, where cheap green electricity from water and geothermal is abundant and hard to transport. I hope we get there soon. Its one of the cheaper ways to avoid co2 emissions.

  22. One of the advantages of steel is it is currently easier to repair than aluminium or carbon fibre. But now we are shoving all sorts of collision avoidance stuff into cars which will make accidents less common, this is less of an issue. Also 3D printing may lower the cost of replacement parts. (Unless of course we let a large country bully us into making downloading the specs of generic spare parts a crime.)

    And an interesting thing with aluminium is eventually its production will peak, and since it is almost 100% recyclable this may happen reasonably soon, particularly if various improved carbon compounds start taking its place in many applications. And we last thing we want is to be left holding the bag when this happens and owning lots of new refining and smelting infrastructure that will never pay for itself.

    Now just when aluminium production will peak all depends on what happens in Nigeria and whether or not Indians go in for aluminium siding, but given how close China appears to be to living off its own recyled aluminium, it may not be long. (Sure they still need a lot of aluminium to complete and update their grid, but the bulk of the work appears to be done.)

  23. I did an unusual thing today and read a newspaper and found out that Indonesia is apparently going to ban the export of raw ore from mining operations. Apparently they think doing this will result in value added industries taking off. This wasn’t a sudden thing, it was in the pipeline for a long time, but apparently only recently it looked at though it was pretty sure to go ahead. This means China will be looking for additional unrefined bauxite to import and this may have been a major factor in the decision to close the Gove refinery.

  24. From what I can work out alumina eliminates 30-40% of the bulk of bauxite. As well as aluminium compounds bauxite also contains moisture, iron oxides, inert clay and silica. If ships carry Gove raw bauxite to China then they are burning 30% more fuel from the outset plus the later processes. Yet another greenhouse emissions source which comes off Australia’s account along with jobs and profits. We will become a nation that simply digs up things and sells them cheaply, the Nauru model.

    The bulk transport problem is also why Iceland can never be the world centre for electro smelting…we’ll burn fossil fuels regardless. Another suggestion has been DR Congo’s Grand Inga Dam if it is ever expanded to exceed the 30 GW from China’s Three Gorges Dam. Australia has more bauxite than anybody just not the low carbon electricity. Perhaps the deposit on a soft drink can should be $1 so we’ll value aluminium appropriately.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s