One of the best pieces of scientific news the last decades has been the spectacular improvements in solar energy generation. The current world price was set in 2017 when the Dubai government bought a large future solar contract for 7.3 US cents per Kilowatt Hour, a mere 1/6th of the price in 2010. Compared to the 1970s, solar cells now cost less than 1% of what they were then. Unless you own a coal mine, that counts as great news!
Let’s dream out loud a little as to what this revolution in solar might lead to this century. I expect solar to transform the deserts around the world, and I like the fantasy that solar power will be used to green Australia’s deserts by pumping desalinated water up to the top of the Dividing Range.
Before sharing such dreams, let us first discuss a few technological bottlenecks to wider-scale adoption. A continuing problem for solar is that it is intermittent, meaning that large-scale usage depends on technology to store surplus energy and transport it to and from the areas of generation to where it is used. Both long-distance electricity transport and large-scale storage remain very expensive and very limited in scope at the moment, despite technological advances in both.
As a rule of thumb, you lose 5% of the electricity for every 1,000 kilometres of electricity transport, and even that requires prohibitively expensive electricity lines. That rules out any fantasy wherein Australian solar farms supply New York!
Battery storage has come a long way since the 70s, with of course the big Tesla battery in South Australia showing that you can have large batteries that can turn on and off very quickly, which is important for solar applications because solar is very variable. Yet, even that battery is relatively small and not capable of storing whole days worth of population consumption, and it’s way too expensive as a storage device to allow solar to compete with fossil at the moment for large-scale supply to the grid. It’s current function is to smooth intermittent supply from fossil-fuel power stations, making fossil fuel more attractive!
In case you’re wondering: batteries in the form of ipads or electric cars are basically too small fry to make much of an impact on this equation.
You might think there is some clever combination that solves all problems. For instance, you might fantasise about storing surplus electricity by pumping up water to some high-mountain lake from which you later on draw electricity by having it fall down again. Think carefully about the main issues involved: you lose something like 20% of the energy pumping the water up at the mountain; you need very unusual mountainous terrain that allows you to have two large lakes from which the water tumbles and gets pumped up without much leakage at either end; and if the population is 2,000 kilometres away, you lose another 20% getting the electricity to and fro. All this is quite apart from the installation and running costs of the lakes, the pumps, the solar panels, and the electricity lines. From my reading, such a package is a long way off being commercially viable, and really only a longer-term dream for countries like China that have the requisite mountainous terrain.
The hope of course is that the bottleneck technologies continue to improve. They will have to for solar to replace fossil as the go-to source of energy for the main electricity grid. The same considerations, btw, also go for wind energy, which has seen similar reductions in price. Both technologies are now low-cost enough to be commercially interesting for many applications and in particular areas, but the package is still not quite there yet to knock fossil off its throne. That, btw, is partially because fossil fuel has become a lot cheaper too, particularly since the US and China found vast reserves of shale oil and gas. Hence the world’s carbon emissions are still increasing despite renewables hitting an all-time high of 25% of world energy consumption.
Relevant to this are the current developments in China. In the 00’s, the Chinese government wanted to be independent of fossil fuel rich regions, like the Gulf and the US. It invested massively in solar technology, which lead to huge cost reductions in solar technology, now copied elsewhere. Yet, the Chinese have recently found vast reserves of shale gas and as a result are scaling back their own solar projects, which tells you they don’t think solar is cheap enough to replace gas. When it comes to this sort of thing, the Chinese leaders are a very pragmatic bunch of engineers, so stories of evil conspiracies of the fossil fuel industry are unlikely to have mattered for this decision: costs will have dominated.
Supposing that the battery and transport costs of electricity indeed come down though, the future of solar seems immense. Let’s dream a little.
For one, cheap solar transforms deserts from places bereft of human activity to prised assets for electricity generation. You see, many deserts are near the equator where the sun is the brightest and land in the deserts is extremely cheap since there is nothing much else for humans to do there. So deserts are the logical places to house massive solar farms.
What is currently in the way of a place like Australia filling its deserts with solar farms is the transport costs of electricity and the big-battery issue. When those bottlenecks are gone, the dash for the deserts is on. Places like Australia, Saudi Arabia, but also California and Mexico, would be big beneficiaries. Plenty of cheap deserts in those places where nothing much else happens of high value.
Deserts in Africa and Central Asia could also be stacked with solar panels, but in those cases there is an additional bottleneck, which is that the main users of the electricity would be in other countries, which raises the issue of international politics. Like the oil pipelines that go through many countries, electricity cables that go through different countries would be prime targets for extortion and political in-fighting. The countries in Africa with lots of deserts are very politically volatile and any expensive electricity lines would undoubtedly get sucked into many conflicts, which essentially just increases the price. Central Asia is a bit more stable, but the same problem applies, so don’t expect the dash to happen there first.
The deserts might well be affected by another solar-related change, which is that intermittent availability is not a problem for desalination and water pumps. Hence one of the major processes that is not dependent on the big battery problem, nor even that of electricity transportation costs, is that of desalinating ocean water and pumping it to the deserts to make them greener. It is a prime thing to do with the excess electricity during sunny days, when the price would be close to zero. This too is highly relevant for Australia, Saudi Arabia, and other countries with deserts close to the sea.
In the case of Australia, the obvious scenario is for solar farms to supply the energy to desalinate huge volumes of water just East of the Dividing range, pumped up to the top of the Dividing Range, and then let loose to the West of that range, essentially desalinating parts of the desert and greening the interior. If you look at a geological map of Australia, the most suitable place to pump the water to would seem to be somewhere west of Lismore: from the top of the range there, one could let the water stream via a system of canals to the Great Basin to the North-West and into the Murray-Darling Basin to the South-West.
Admittedly, this is a mere pipe-dream at present, but hej, why not? Building these pipes, canals, and pumps could be one of the major infrastructure projects of the 21st century. Once the technology has been perfected in Australia, it is a good bet the companies would be commissioned to repeat the feat in many other places in the world, so it could become one of our comparative advantages.
Cheap solar unlocks the deserts in another way, which is that it provides the energy for lots of air conditioning, making them much more habitable, though obviously still confined to in-doors environments.
Finally, cheap solar makes the deserts more attractive places for extremely energy-intensive industries, like Bauxite-to-Aluminium processing or big chemical plants.
Then the issue of climate change. Would cheap solar (and wind), combined with cheaper big batteries and cheaper electricity transportation, on its own lead to such a reduction in fossil fuel usage so as to halt the warming of the earth? It would seem the answer at the moment is still ‘no’ for several reasons.
For one, one should always bear in mind that the increased greenhouse gas concentrations in the atmosphere are like a blanket over the earth that only very slowly gets reduced in its thickness if it is no longer added to. By very slowly, I mean that the natural processes that return atmospheric CO2 levels to pre-industrial values take centuries, if not thousands of years. So even if all human emissions were to stop abruptly today, the world would continue to warm for a long, long time yet.
It also remains the case that there is huge regional variation in just how cheap fossil fuel is and that this implies that the whole package containing solar would have to be extremely cheap to out-compete fossil for most applications nearly everywhere. You see, in some places, like the Gulf, the costs of pumping up the oil is almost zero and the only costs borne by the users is that of usage. Even if the solar panels were for free, the other elements (batteries and transport) will often be too expensive to compete with that. In areas with strong winds, abundant sunshine, and less easily available fossil fuels, the relative costs look different.
The convenience of fossil fuel should also not be under-estimated. Fossil fuels are very portable, pack a lot of energy punch for their weight, and of course they have the advantage of the huge existing infrastructure for its dissemination and usage (fuel stations, existing power plants, and lots of combustion engines). Even free solar energy would take a long time (decades?) to flush out those advantages.
You might think that cheap solar (and wind) could out-compete these mobility and energy-per-weight advantages of fossil fuels by, for instance, being used to create hydrogen that would also be mobile and an alternative to fossil fuel. Whilst there too, the technology is progressing, the key problem with hydrogen is that it is so bloody volatile and explosive. Just as you don’t put ammunition depots in the middle of the cities, so too do you really not want large stores of hydrogen anywhere near large population centers. The technology needs to progress a lot to overcome that bottleneck. Remember the Hindenburg disaster!
Finally, there remain applications for which solar (or batteries in general) is not energy-intensive enough. Key among those are aeroplanes and large ships. Those two applications alone would ensure the world continues to pump more new greenhouse gasses into the atmosphere than naturally gets taken out, which essentially means that the world as a whole would just slightly more slowly go through its fossil fuel reserves.
Doesn’t it help if the world burns its fossil fuel reserves more slowly? Basically, ‘no’. From a geological-time perspective, the difference between burning up the fossil fuel reserves in 50 years or 300 years is nigh-irrelevant for the eventual peak in global warming. You would thus still need to do something else than merely slow down the fossil burning if you want to halt global warming (like actively taking out CO2 or geo-engineering).
So technologies still need to improve spectacularly for solar (and wind) to seriously dent the climate change trajectories over the coming centuries. Yet, it would seem that the technologies are getting close to the point where we should expect major effects on our deserts already. Cheap solar should be expected to make them much more habitable and suitable for energy-intensive industries.
Paul
It’s technically possible ( has been done on a small scale)to use solar to split a oxygen atom of a carbon dioxide molecule and get carbon monoxide: the feedstock for synthetic hydrocarbon fuel.
Worth noting that the process of getting water down the other side of the range can also recover some of the energy as electricity. And as you mention, you can do this at a time that’s convenient rather than necessarily needing to use the water only when the sun is brightest.
Technically there are sites for pumped hydro on the east cost of Australia that would supply ~100x our current demand if we used them all. Instead we have an existing pumped hydro system that’s not used because the coal power company that owns it makes more money from demand-driven price fluctuations than they could by using it to reduce the fluctuations. That and most of the other problems are political, not technical (as they say “at time of writing Malcolm Turnbull is still Prime Munster”).
One huge problem for pumped hydro is that you’re taking land that white people can see, and that white people care about… that’s hard. But taking deserts off third world coloured people… that’s so easy it’s not even worth mentioning as an obstacle. Even if those third world coloured people are in Australia.
Also, one problem with aluminium production is that it’s very sensitive to changes in energy flow rates. Not just the brutal freezing of pot lines, but even relatively small, predictable changes require quite a lot of engineering effort to work with and there’s a minimum flow rate required just to keep the system “idle but molten”. That makes running it off solar quite challenging.
What can work really well is chemical-electrolytic refining, which is basically a chemical battery used in permanent “charge up” mode and you swap in new electrodes as it charges. That can easily be designed for intermittent operation. The problem is that the chemistry tends to be pretty toxic (the brutal option is usually “dissolve ore in hydrofluoric acid then use electricity to split that to fluorine gas and metal”) and not all metals can be refined that way. It’s a topic of ongoing research because it’s generally colder than smelting and relatively low-energy so it beats hot refining in a lot of ways (for example, aluminium was more expensive than gold before they worked out how to refine it electrolytically).
Moz what’s your view on the idea of using desert solar to make ammonia – a relatively safe transportable form of ‘hydrogen concentrate ‘?
I’m aware there’s research and fanboys :)
AFAIK the processes used to make it are amenable to intermittent operation, howls of outrage from the process people notwithstanding. As an industrial material it’s widely used and for that purpose it would be ideal. But it’s not hugely desirable as a fuel for a range of reasons (cleaning the NOx after combustion, for example) but if fuel cells can be made practical that would make it really good. At that point the significant energy required to make nitrogen bonds becomes a benefit.
I gather that there is some new process that it’s claimed would make converting the ammonia back into 3 hydrogen atoms and a nitrogen atom a practical proposition.
Would guess that using solar to make liquid fairly safe to transport concentrated energy would fit more easily into our existing systems.
There have been claims about new processes and demonstrations of various ways to use ammonia for some time now. A quick literature search turned up a review article from last century that claimed to give an overview of the various approaches being tested (I could only read the abstract without paying). Note that a lot of fuel cells are extremely sensitive to contaminants so the lab life can be three or four orders of magnitude longer than the working life using street-grade fuels.
From a random link
Half a percent contamination in fuel cell terms is garbage… even water filters in trucks would baulk at that much water in the diesel. If you had to strip 0.5% of anything out of the ammonia before you could make your fuel cell work you’re talking a very different material-handling process than “pour liquid into tank”. Viz, it could be done but it’s not going to be pretty.
You also have density issues – ammonia is about 4 kWh/l compared to diesel/petrol at about 10kWh/l meaning you need twice the volume (and thrice the mass) for the same energy, assuming similar conversion efficiency. Ideally a fuel cell would be twice as efficient as a diesel (90% rather than 45%) but it can’t be 3x :)
I’ll believe it when I can buy one.
Thanks Moz
Moz
http://www.abc.net.au/news/2017-05-11/hydrogen-breakthrough-could-fuel-renewable-energy-export-boom/8518916
Going of this report a membrane type system would be used to split the hydrogen of the nitrogen , before it is supplied to a fuel cell.
What do you think?
Fundamentally they’re turning 2NH3 -> N2 +3 H2 and I’m not sure what the energy balance is there, but I strongly suspect it’s positive (it takes energy to turn N2 + H2 -> NH3, especially splitting the nitrogen), so in theory a catalysed process using an osmotic membrane could work with little power input. That should also give you nice clean hydrogen, it’s just a question of what the dirt in the ammonia does to the membrane.
There are uncountable numbers of things like this that work astonishingly well right up until someone tries to build a factory making a million units a year, at which point you never hear about it again. Ruthenium based supercapacitors revolutionised the … scientific recordbooks… before quietly vanishing back into the “CSIRO inventions” file a few years ago.
Then there’s round-trip efficiency. It’s all very well having an energy-positive step at the mobile end of the process, but a: not too much energy because it likely appears as heat; and b: you have to shove that energy in at the front and that hurts your round-trip efficiency.
For example, NiFe batteries are awesome, tens of thousands of cycles with no maintenance, incredible temperature range, very stable, easy to manufacture from common elements… and 50% round-trip efficiency with 10% self-discharge per month, the latter rising as the batteries age/gain impurities. One company makes them for one very specialised application, but you can buy second hand ones fairly easily because a 50 year old cell is probably still fine.
You’ll know the CSIRO have cracked it when the researchers move to China and we start hearing about all the millions of dollars they’re getting for licensing it.
moz
Those NiFe batteries ,on a quick search ,sound interesting, are they available here?
we have plenty of space – our place is about half a hectare- what’s the cost of say 10kwh of storage ?
John, it’s less about the space and more about the efficiency. I did a bit of checking a couple of years ago but capacity for capacity they cost the same as lithium but were less efficient and peak discharge rate is much lower – C/40 or worse compared to C/2 or C/5 for lithium.
There’s an overview here:
https://batteryuniversity.com/index.php/learn/article/Nickel_based_batteries
Also, apparently a company in China makes them now too, and those are sold in the US.
As with everything, there are total fanboys around, I suspect if you search the ATA forums you will find someone who will help you. Whether you can find an installer to do the work is another question – it’s very much a DIY technology from what I can gather.
If you want to get experimental the SA flow batteries are much more interesting, but they’re commercially available so your opportunity to get in on the actual test phase is gone. They lost out to the PR steamroller that is Elon Musk in the SA big battery contest.
Moz thanks
For various reasons I don’t agree that maintaining continuous supply with renewables is as big a difficulty as Paul thinks, but for the sake of argument lets assume he’s right.
So lets think through a world where power is ultracheap at some times of day and relatively expensive at others. In the short term this is indeed a pretty useless state of affairs. But in the medium term your society and economy will change to fit – prices drive behaviour, remember? Current aluminium refineries are designed to operate continuously, for example, but there is no technical reason they can’t be built to operate only during the day (you just keep them hot overnight). You only need heating storage for a well insulated home for a few hours – molten salt banks for homes are an old technology that still works perfectly well, even if Tesla Powerwalls or their clones don’t keep falling in price. And so on.
It seems to me that Paul is showing surprisingly little faith in the ability of a capitalist economy to adapt – and adaptability is the one quality at which capitalism excels.
Hi DD,
I agree capitalism will adapt and that it’s rather good at adapting. I was just pointing to the current obstacles still in the way. I am not too excited about molten salt, but the market system will indeed test lots of things, including adapting to intermittent energy. That’s what markets do. They will probably surprise us in terms of the solutions they come up with.
I am mainly dreaming out loud what the advantages are of cheap solar for the deserts at close-to-current technologies.
The Kidston pumped hydro/solar/wind project is happening, so it’s not all pie in the sky.
http://www.genexpower.com.au/250mw-kidston-pumped-storage-hydro-project.html
ANU have produced a study into other sites suitable for pumped hydro and they seem to be everywhere.
http://www.anu.edu.au/news/all-news/anu-finds-22000-potential-pumped-hydro-sites-in-australia
Pumped hydro is very much an ancient technology, any claim that it’s pie in the sky is a work of very stable genius. It’s “water in the sky” but in a very concrete you-can-swim-in-it way. I draw your attention to the Tumut system in the Snowy Mountains for an example of real working pumped hydro in Australia.
https://en.wikipedia.org/wiki/Tumut_Hydroelectric_Power_Station
The problem is not technical, it’s political and economic. It will be interesting to see where the latest prosperity gospel happy-clappy takes the Liberal Party on this issue, but I’m not optimistic. At best I reckon we’ll get “do whatever you want, but I’m imposing an efficiency dividend of 30% on it”.
(Tumut even makes the list of “greater than 1GW” systems)
We are talking apples with oranges; the Kidston project is renewables with pumped hydro as being the “battery”.
The problem with the Snowy is that it’s a long way from places that are a long way away :-)
Smaller setups scattered around the country could be one way of overcoming the dilemma of distance.
I don’t understand how the source of power makes a difference. You’re still using a hydro dam to store electricity as gravitational potential energy. But by all means, if it makes you feel better, I’ll agree that the new improved PumpedHydro2000 is a revolutionary development in inner-city energy storage. Not like that silly remote Snowy Mountains system at all.
Oh, and the added joy of dealing with gold mine leachate in your hydro machinery. Which I’m sure will be no issue because we can rest assured that mining companies always clean up after themselves. And this is definitely not greenwashing by a company that finds itself with an unwanted hole they have to remediate.
More seriously, if this does turn out to be a useful way to deal with second hand mines that would be great.
Building a 270MW PV farm next to it is the useful part.
There’s some very careful wording on the front page “the first {whatever} in an old gold mine”… FFS, they could be building a childcare centre and that would still be true. “lake”… yeah, nah, it’s a mine bro. A mine pit full of water is still a mine pit. And hydro dams with daily cycles don’t make the sort of habitat that gets the local flora and fauna excited. If we’re lucky it’ll be unattractive enough that they don’t get flocks of endangered waterbirds going through the turbines.
And “the Snowy is too remote” … Kitson is in Far-North Queensland. 2GWh of storage at 0.25GW is not record-breaking, it’s useful but unremarkable for pumped hydro. Hopefully they will set up their extension cord to be able to draw 250MW (or whatever their pump capacity is) as well.
I like the idea, I just react really badly to overblown marketing. I’m more your hard-conservative greenie “convince me that this won’t screw things up even more”.
Rog,
smaller places are less efficient. Hydro-storage has been a source of many plans that have come to naught, with no shortage of reports and ‘identified sites’. One of the Netherlands’ most famous engineers (van Lely) dreamed of using a large in-land lake (Ijsselmeer) as a means of hydro-storage. It was deemed too expensive and subject to ground water leakage to ever get off the ground. The UK apparently stopped building these things 30 years ago, though it has no shortage of identified small sites.
China is apparently planning a few new big ones. Snowy is one of the most logical places in Oz, but as you say is a far way away. And don’t forget that you need two lakes (one up and one down), not one for the storage idea to work.
A danger in this debate is capture by special interests who chase huge subsidies for things that sound good but are useless from any economic or environmental point of view. Before you know it, we’d have another White Elephant, like the Bass Straight power line….
I would love all this to really work though. We are on the same side when it comes to what we hope for with respect to technological breakthroughs.
Our disagreements have been about what is politically feasible. I hope you’ll agree my predictions on that score have been pretty good, eg: https://economics.com.au/2013/11/04/australian-carbon-emission-politics-explained/
Gee, using the Isseljmeer was surely optimistic. You need a decent head for an efficient system – otherwise the pipes and turbines have to be absolutely massive (and hence expensive). But the proposal did have one thing right – given that most cities are near the coast, saltwater rather than freshwater pumped storage is likely to be much more available.
And as we’ve argued before, HVDC is not subject to the same distance losses as HVAC and unlike HVAC the capital costs are fairly invariant with distance (because you don’t need periodic transformers). Providing demand is large enough you don’t really have to have your major city too close to a desert or a major hydro scheme for large scale solar power to be a goer.
But of course supplementing and partly replacing current networks will take time and money and so absent a stiff international carbon price (which unfortunately is now unlikely) it will only happen as existing networks become old and inadequate. I’m afraid it is now too late to avoid serious damage to the planet – we can only try and avert a full-strategic-nuclear-exchange sized catastrophe.
Paul, I just listened to the most recent Energy Insiders podcast which focused on pumped hydro. Parkinson’s guests were Andrew Blakers of the ANU and Bill Armstrong from GE.
FWIW, they both seemed comfortable with the prospects for pumped hydro in Australia and saw 100% renewable electricity production backed by roughly 20 GW of pumped hydro as entirely realistic.
why am I not surprised that the director of a renewable energy research institute and a representative of the commercial company that builds a lot of this are appearing as good buddies on a podcast series run by a pro-renewable energy site, proclaiming the imminent dominance of what they make their money with?
I regard such stuff as infomercials.
I wondered if that would be your response.
Seems to me the question isn’t whether they’re keen on the idea but how technically and economically efficient pumped hydro is likely to be. I got the impression that although the idea has been around for a long time, like most things it’s undergoing constant technological improvement and that demand around the world is high and growing.
If you have links to good, substantive counterarguments I’d love to see them.
Curious about the size of storage needed ,for example:
if you had a fall of say 250 meters what’s storage volume you’d need to generate, X megawatts an hour for at least a few hours?
Hi John,
I hope you weren’t looking to me for an answer! No doubt the formulas are out there somewhere in Google land.
Hope you’re well.
Hi Ingolf
Peace be with you
I expect that the answers are somewhere :-)
Anne and I are well, and the National Gallery recently purchased a major work of mine that I made 17 years ago, so no complaints.
And with you John.
Sorry for popping up and then going AWOL. I’m delighted to hear about the acquisition . . . a deep and quiet pleasure I imagine.
Thanks
Hi Ingo,
I’ve given you the main counterarguments in the post, but you dont seem to like those :-)
Finding out what the real constraints are in these kinds of technologies is tricky because the internet is flooded with pieces of believers and particular interest groups who keep up a steady stream of up-beat infomercials and research papers. Despite that, the key figures I use above are easy to find (ie that 20% loss is a minimum of pumping up and the energy loss in long electricity lines, where again I have taken conservative estimates).
Where do you find sobering statements and information that tell you what is really going on? The best place is probably the international energy agency which tells you historical trends in hydro (eg http://www.iea.org/tcep/energyintegration/energystorage/ ) which tell you that despite continuous up-beat projections (eg this one in 2011 in ‘hydroworld’ about what was in store for Europe the next 10 years, little of which truly happened: https://www.hydroworld.com/articles/print/volume-21/issue-3/feature-article/what-drives-pumped-storage-development.html ) hydro capacity has hardly moved the last few years.
And why did the IEA think hydro was not growing very fast (not even keeping up with world energy consumption!)? It says “pumped hydro is concentrated in a few places – half of all installed capacity is found in just three countries – and growth will remain constrained by geography.”. If you dig enough in their website you’re bound to find further reports and papers on that point.
I wish you to reflect upon the many studies that supposedly identify thousands of suitable places in Australia and the UK alone and yet the reality that nothing has been built in the UK for the last 30 years, but lots has been shelved. What does that say about the institutes that produce such reports?
Hi Paul,
It wasn’t so much that I didn’t like them (nice one) but rather that I was unsure about the practical importance of pumped hydro’s energy efficiency (and any transmission losses due to remote locations). If it’s other advantages are sufficiently compelling, 75-80% efficiency might be more than good enough.
With my (very) limited knowledge of the industry, it doesn’t seem particularly surprising that there’s been no rush to build more pumped hydro capacity. The intermittency problem is pretty new and without it the case for investing in long-term storage can’t have been all that attractive. Now, however, with rapidly growing renewables generation the need seems to me to be somewhere between urgent and critical.
If it’s allowed to do so, the market will soon decide which forms of storage make most sense. As technologies evolve, that balance will probably constantly change, at least at the margin. Still, in terms of providing reliable, relatively low-cost utility scale long-term storage, pumped hydro presumably has to be on the shortlist.
I guess we’ll find out soon enough.