The Death Of Fossil Fuels

[IvanV]

Natural gas is gas which comes from gas reservoirs, and is a mix of various different components, not just methane. Methane is usually the most common component, but ethane, propane, butane (both isomers), even pentane can be present. Components with greater molecular masses generally get liquified and turned into other products.
But as these other components are also present, some natural gas mixtures are denser than air. So the venting arrangements for a propane-based heat exchange fluid need not be so very different to a standard venting arrangement (the point I was trying to make, obviously unsuccessfully!)

An extraordinary recent discovery is that there are extensive underground reservoirs of hydrogen. That can be called a natural gas too. We used to say, hydrogen, you can't just dig it out of the ground. But apparently it's looking like you can.

Since it occurs in quite different geologies from coal, oil and methane gas, it has only rarely been accidentally drilled into. Well, it is common to find small quantities of hydrogen in certain common locations. There have been a scant number of cases over the years of people finding hydrogen issuing in quantity from a hole in the ground. So hydrogen in useful quantity was thought likely a rare curiosity. But now geologists have been thinking about how it might form and collect, and have gone out looking for it in the places that suggests, they have all of a sudden been finding quite a lot of it. A single strike in France appears to contain several years worth of present global production of hydrogen. Though of course we need several orders of magnitude more than that for it to be material in decarbonisation. Though this is but one of a hundred or so of sources recently identified. It remains unclear quite how much there is in global terms, a lot more exploration is needed.

It might yet prove the serendipitous discovery that makes decarbonisation a lot easier, though of course from zero to a global scale hydrogen extraction program in completely different places from present gas reservoirs is a huge investment and conversion. And whilst some are suggesting that there is several centuries of global energy needs there, we don't know that is true yet.

An article on this in science.org

[bjn]

One of the difficult things to decarbonise is cement production. Portland cement is mostly calcium oxide with a few silicates and other minerals thrown in. The process to make it involves heating the limestone (calcium carbonate) up to high temperatures, generally using fossil fuels. The break down of limestone frees up the calcium, but the carbonate turns into carbon dioxide. So even if using an electric kiln as opposed to burning fossil fuels, you are always producing CO2.

Sublime Systems has developed an electrochemical method to create a drop in replacement cement that has no intrinsic emissions. It would require zero carbon electricity to completely avoid CO2 emissions. Their technique uses electrolysis to split calcium bearing non carbonate minerals to get CaO. Anything above 10% calcium will do, which generally means basalts. There is a lot of basalt out there. They can even use fly-ash as a feed stock. Once scaled up, they claim they would be cheaper than concrete with carbon capture and storage or a price for carbon, but not as cheap as traditional cement. However, cement is so cheap anyway, it may not matter for projects that are mandated to be carbon free for a variety of reasons.They may even be able to sell carbon credits to other polluters to offset the cost of their cement.

They currently produce 250 tonnes/year with their research plant, but are scaling up to a kilotonne/year commercial plant, aiming for 2026. They’ll use that to figure out how to optimise their production process before going for a megatonne plant, hoping for 2028.

I hope it works out. This could eventually displace 8% of all current carbon emissions in an otherwise hard to abate sector.

https://sublime-systems.com/

The Volts podcast has an episode on them.

https://www.volts.wtf/p/we-are-closing- ... ero-carbon

[IvanV]

They currently produce 250 tonnes/year with their research plant, but are scaling up to a kilotonne/year commercial plant, aiming for 2026. They’ll use that to figure out how to optimise their production process before going for a megatonne plant, hoping for 2028.

To put that in context, US production is around 100 megatonnes, and global 4,000 megatonnes. But given the high transport costs of cement, typically production is geographically distributed, and you'd want to replicate megatonne scale plants across the country, rather than make even bigger ones. It's not something there is very much international trade in, except where destination countries are small or lack raw materials.

[bjn]

One has to start somewhere. Sigmoid curves and all that.

It would also unlock cement production in places that have no limestone, eg: Iceland which has abundant low carbon electricity and is basically a giant chunk of basalt.

Also why a price on carbon would accelerate something like this.

[Grumble]

Natural gas is gas which comes from gas reservoirs, and is a mix of various different components, not just methane. Methane is usually the most common component, but ethane, propane, butane (both isomers), even pentane can be present. Components with greater molecular masses generally get liquified and turned into other products.
But as these other components are also present, some natural gas mixtures are denser than air. So the venting arrangements for a propane-based heat exchange fluid need not be so very different to a standard venting arrangement (the point I was trying to make, obviously unsuccessfully!)

An extraordinary recent discovery is that there are extensive underground reservoirs of hydrogen. That can be called a natural gas too. We used to say, hydrogen, you can't just dig it out of the ground. But apparently it's looking like you can.

Since it occurs in quite different geologies from coal, oil and methane gas, it has only rarely been accidentally drilled into. Well, it is common to find small quantities of hydrogen in certain common locations. There have been a scant number of cases over the years of people finding hydrogen issuing in quantity from a hole in the ground. So hydrogen in useful quantity was thought likely a rare curiosity. But now geologists have been thinking about how it might form and collect, and have gone out looking for it in the places that suggests, they have all of a sudden been finding quite a lot of it. A single strike in France appears to contain several years worth of present global production of hydrogen. Though of course we need several orders of magnitude more than that for it to be material in decarbonisation. Though this is but one of a hundred or so of sources recently identified. It remains unclear quite how much there is in global terms, a lot more exploration is needed.

It might yet prove the serendipitous discovery that makes decarbonisation a lot easier, though of course from zero to a global scale hydrogen extraction program in completely different places from present gas reservoirs is a huge investment and conversion. And whilst some are suggesting that there is several centuries of global energy needs there, we don't know that is true yet.

An article on this in science.org

It’s worth pointing out that transporting, storing and burning hydrogen do present challenges, so we should continue electrifying and ramping up renewable production, but if we can use hydrogen to address intermittency then we are a long way forward of where we were. (I haven’t read the article yet, maybe it addresses that)

[IvanV]

It’s worth pointing out that transporting, storing and burning hydrogen do present challenges, so we should continue electrifying and ramping up renewable production, but if we can use hydrogen to address intermittency then we are a long way forward of where we were. (I haven’t read the article yet, maybe it addresses that)

The reference is only about the availability of mined hydrogen, its possible cost, not how it might be used.

Clear advantages of it are for hard-to-decarbonise industrial processes, and addressing intermittency in electrical supply. Avoiding going down the carbon capture route, which is going nowhere slowly, is a potential large advantage. It may also be relevant for some mobile applications such as heavy vehicles, ships, planes, trains on lightly used routes, etc.

Space heating, supplied by piped hydrogen to your house, I would say still looks unlikely, but I'm not 100% certain about that. In principle, large amounts of mined hydrogen permit us to have piped hydrogen and heat our houses with it. A lot of people think, it's so much easier to achieve that than to get everyone to change to a heat pump, because heat pumps are large, expensive, etc. But piped hydrogen for heating is only likely if hydrogen is cheap enough, and, crucially, soon enough. If it isn't, enough people will transfer to heat pumps that the customers for piped hydrogen will not be dense enough for the grid to be maintainable. And we don't even know if mined hydrogen is that common yet. So my main guess is that widespread piped hydrogen still looks unlikely.

[bjn]

Piping H2 through existing pipe networks will need a whole range of upgrades. H2 loves to leak, embrittles metals, will need gasket and value upgrades, is very happy to ignite/explode at far lower concentrations than methane and has 1/3 the volumetric energy density of methane, requiring bigger pipes and/or triple the pressure. Upgrading every pipe and fitting from the H2 source to the H2 consumer will be a non trivial undertaking. It’s not just a case of tweaking your boiler.

Heat pumps have the advantage over electrolysed H2 for heating in that they have a COP over 3 now. If you diverted that electricity into H2 generation, you have losses at the electolyser, then losses for pumping it all around the network and losses even at the boiler. So (pulling figures out of the air slightly), for the same electricity you will get something like 4X the heat with a heat pump over burning H2 from the same electricity.

[FlammableFlower]

What IvanV is referring to is not electrolysis generated hydrogen, but direct mined/tapped - e.g. they've found underground hydrogen stores similar to natural gas, so it just needs tapping. 'Just' is in italics, as there's all the obvious chemical differences that make it not simply a question of doing the same thing as we do for natural gas but in a different place...

[Grumble]

Piping H2 through existing pipe networks will need a whole range of upgrades. H2 loves to leak, embrittles metals, will need gasket and value upgrades, is very happy to ignite/explode at far lower concentrations than methane and has 1/3 the volumetric energy density of methane, requiring bigger pipes and/or triple the pressure. Upgrading every pipe and fitting from the H2 source to the H2 consumer will be a non trivial undertaking. It’s not just a case of tweaking your boiler.

Heat pumps have the advantage over electrolysed H2 for heating in that they have a COP over 3 now. If you diverted that electricity into H2 generation, you have losses at the electolyser, then losses for pumping it all around the network and losses even at the boiler. So (pulling figures out of the air slightly), for the same electricity you will get something like 4X the heat with a heat pump over burning H2 from the same electricity.

If the H2 is mined then the problem of generating via electrolysis goes away, but all the other problems (adding generating NOx as a significant one) still apply.

[bjn]

What IvanV is referring to is not electrolysis generated hydrogen, but direct mined/tapped - e.g. they've found underground hydrogen stores similar to natural gas, so it just needs tapping. 'Just' is in italics, as there's all the obvious chemical differences that make it not simply a question of doing the same thing as we do for natural gas but in a different place...

That's why I was explicit about 'electrolysed H2'.

Mined H2 doesn't look like it's available in quantities to replace methane usage (though the article says it is present in orders of magnitude more than we currently consume H2), especially as it's a total pig to transport. If it does turn out to be a thing, I'd expect it to be used for industrial purposes near the site of extraction, eg: fertiliser production, which is way easier to ship and a value add process. If available near a grid, burning it in a combined cycle gas turbine and using the electricity heat pumps would be much more sensible for heating. CCGTs have an efficiency of ~50%, combined with a heat pump COP of 300% gets you 150%. That is hand waving away distribution lossses, but stack that against distributions losses of pumping H2 along with the capital cost to upgrade everything in the pipe network, and you probably come out evens at worst. Remember, the fungability of electricity is fantastic, so you can use that H2 generated electricity for a huge range of other purposes, including firming up renewables.

The interesting bit is they say naturally recuring H2 is 'renewable' as it mainly results from a reaction between iron bearing hot rocks and water. Coal is renewable as well, just that the time frames involved are tad too long to be useful. I'd be curious as to how fast naturally occuring H2 sources replenish themselves.

[Grumble]

What IvanV is referring to is not electrolysis generated hydrogen, but direct mined/tapped - e.g. they've found underground hydrogen stores similar to natural gas, so it just needs tapping. 'Just' is in italics, as there's all the obvious chemical differences that make it not simply a question of doing the same thing as we do for natural gas but in a different place...

That's why I was explicit about 'electrolysed H2'.

Mined H2 doesn't look like it's available in quantities to replace methane usage (though the article says it is present in orders of magnitude more than we currently consume H2), especially as it's a total pig to transport. If it does turn out to be a thing, I'd expect it to be used for industrial purposes near the site of extraction, eg: fertiliser production, which is way easier to ship and a value add process. If available near a grid, burning it in a combined cycle gas turbine and using the electricity heat pumps would be much more sensible for heating. CCGTs have an efficiency of ~50%, combined with a heat pump COP of 300% gets you 150%. That is hand waving away distribution lossses, but stack that against distributions losses of pumping H2 along with the capital cost to upgrade everything in the pipe network, and you probably come out evens at worst. Remember, the fungability of electricity is fantastic, so you can use that H2 generated electricity for a huge range of other purposes, including firming up renewables.

The interesting bit is they say naturally recuring H2 is 'renewable' as it mainly results from a reaction between iron bearing hot rocks and water. Coal is renewable as well, just that the time frames involved are tad too long to be useful. I'd be curious as to how fast naturally occuring H2 sources replenish themselves.

That’s where having a solid geological theory would help

[shpalman]

Maybe they're talking about this https://en.wikipedia.org/wiki/Serpentinization

[IvanV]

Mined H2 doesn't look like it's available in quantities to replace methane usage (though the article says it is present in orders of magnitude more than we currently consume H2), especially as it's a total pig to transport. If it does turn out to be a thing, I'd expect it to be used for industrial purposes near the site of extraction, eg: fertiliser production, which is way easier to ship and a value add process. If available near a grid, burning it in a combined cycle gas turbine and using the electricity heat pumps would be much more sensible for heating. CCGTs have an efficiency of ~50%, combined with a heat pump COP of 300% gets you 150%. That is hand waving away distribution lossses, but stack that against distributions losses of pumping H2 along with the capital cost to upgrade everything in the pipe network, and you probably come out evens at worst. Remember, the fungability of electricity is fantastic, so you can use that H2 generated electricity for a huge range of other purposes, including firming up renewables.

The interesting bit is they say naturally recuring H2 is 'renewable' as it mainly results from a reaction between iron bearing hot rocks and water. Coal is renewable as well, just that the time frames involved are tad too long to be useful. I'd be curious as to how fast naturally occuring H2 sources replenish themselves.

Ultimately what matters to the consumer is the price of H2 vs price of electricity. That assumes they even have the H2 available, which as you correctly point out is far from easy or cheap to achieve. But outcome varying with price comparison is what we see in the more reputable strategic reports on energy scenarios prepared for government: so the extent of space heating with heat pumps vs hydrogen varies with the H2 vs electricity price.

My previous judgments have been based on H2 from electrolysis. And H2 from electrolysis is almost certainly too expensive for to burn for heating in the house, vs heat pump, unless somehow that electrolysis takes place some place else where electricity is a lot cheaper than it is locally. People talk about electrolysis only when there is a surplus of wind/solar, so the electricity is free. But that means you get poor load factor from your electrolysis plants. H2 made that way only becomes cheap enough to burn if the capital costs of electrolysis plants get a lot, lot cheaper than is currently happening. And we can devise practical large scale electrolysis plants, which is currently a problem.

You make a strong case that electricity for heat pump should always be cheaper than H2 for heating, even if the H2 is mined. The market connection is that H2 is being burnt to smooth supply, ie when wind/solar is insufficient. So that kind of sets a ceiling on the price of peak electricity. Though we also have to pay for the capital cost of the H2 burning plants, and the less frequently they are switched on, the more premium over the marginal cost of running they will have to be paid. Then with both electricity and gas, for the consumer, the cost of distribution and other system costs are averaged into the price. That currently results in a larger premium over wholesale price for electricity than gas. We pay something like 3 times wholesale price for domestic electricity, but for gas the premium is only about 2 times. Part of that higher premium is a bunch of green taxes which are placed more heavily on electricity than gas, which is silly. But they have been talking about fixing that for years and they never actually take the step. But when it is hydrogen rather than methane that is being distributed, the gas distribution premium may well go up because of the difficulties of piping hydrogen. And if customers thin out, it goes up even more, as what is there has to have its costs shared over fewer customers. Though maybe we end up with hydrogen being piped only into areas of particularly dense population, and perhaps hydrogen is most useful in places with dense population where it is harder to have heat pumps everywhere.

So when we say that mined hydrogen implies only a 50% premium for hydrogen, taking into account efficiency of heating usage, that actually makes me worry whether by the time it gets to the consumer it actually becomes cheaper, or at least a close enough proposition to take the "easy" way out, even if energetically it is the wrong decision.

Then finally, for mined hydrogen to be present in sufficient quantity for it to be piped to people's homes. Well we now know that there is a lot of ground hydrogen that can be tapped, but as you correctly say, we don't know that there is that much ground hydrogen.

That adds a bit more detail into my thinking that it seems unlikely, but not with 100% certainty.

[Grumble]

Maybe they're talking about this https://en.wikipedia.org/wiki/Serpentinization

Yes, I think that well established as far as it goes, but I’m not sure we’re yet able to make useful predictions based on it.

[bjn]

One has to start somewhere. Sigmoid curves and all that.

It would also unlock cement production in places that have no limestone, eg: Iceland which has abundant low carbon electricity and is basically a giant chunk of basalt.

Also why a price on carbon would accelerate something like this.

I forgot, cement absorbs CO₂ over its lifetime, the linked article says up to 45% of the CO₂ produced when Portland cement is made. That means a zero/low carbon cement would be a carbon sink. If we replaced all cement production by zero carbon cement, we’d not only avoid 8% of global emissions, but absorb a further 3.5%, albeit of a period of years. Such a cement could be sold for carbon credits as well for its fundamental cementiness.

https://www.science.org/content/article ... ouse-gases

[bjn]

Excellent two episode podcast series from Michael Liebreich, "Net Zero Will Be Harder Than You Think – And Easier". He goes into what the difficulties are and how they can be overcome. If you prefer to read, he has links to his original essays on the pod cast web pages.

His five main reasons for optimism are...

• Exponential Growth which is down to the impressive learning curves of all renewable technologies. eg: 25% per doubling for solar, so in 1979 1W of solar panels cost $106, today it is $0.125, batteries are better than that, wind not quite so good. Rollout is still exponential and the kink in the logistic curve has yet to be seen. He makes the point that a fall in prices not only allows new technologies to replace uses of old one, but opens up a whole new uses. This allows growth of the tech to continue beyond most projections along with associated price drops. Most predictions for renewables rollouts have models with assumptions on floor prices and market saturation. Those assumptions have all proven bogus, thus the infamous graph of where each IEA projection of renewables growth is charted against actual rollouts (image below).
• Systems Solutions By creating a network of complementary components you can have an efficient and robust system. So a mix of solar, wind, batteries, interconnects, pumped hydro, long term H2 storage, nuclear, demand management and more.
• Great power competition Rivalry between the large powers who want to dominate the key technologies of the transition will drive it forward. eg: the Inflation Reduction Act in the USA.
• Disappearing Demand Achieving the transition will require alot less new stuff than we think. The expected demand for key minerals is overstated for a range of reasons, primarily the old adage of "the solution to high prices is high prices". This has kicked off substitution, efficiency, new technologies and especially recycling. None of which have been built into demand forecasts. In 2019 59% of batteries were sent to recycling, the current estimate is around 90%, 99% won't be far off (which is achieved with lead-acid batteries). Then you get many added benefits as FF demand reduces, eg: 40% of high seas shipping is fossil fuels, so we'll need less of that. FWIW, coal demand peaked in 2008.
• The Primary Energy fallacy If you just look at the energy going into economies, rather than energy being used to do useful things, you mis-estimate how much needs to change. This is because burning stuff is horribly inefficient, most of the energy released when stuff is burned is radiated as waste heat rather than doing the thing you want. His example is a petrol car that does 40mpg. It's radiator will be wasting 60-80% of the energy in the petrol, and in total the car will be consuming ~1.2kWh/mile when you factor in all upstream costs and ineffeciencies. An equivalent electric car will be consuming around ~0.3kWh/mile when you factor in upsteam inefficiences. If you only consider the primary energy consumption and assume like for like (which happens often), you will think you need 4X the amount of electricity to replace the equivalent car. Similar arguments for heat pumps in home heating and other FF based use cases.

His closing remarks are worth quoting...

In the same way that there came a point when discharging untreated sewage into the street, or smoking in public buildings, became unacceptable, it is becoming unacceptable to burn fossil fuels. The generation that regarded it as normal – irreplaceable, even a kind of birth-right – is losing its place at the head of the table and being replaced by a generation that is in no doubt about the need to “stop burning stuff”.

That may not make the technical challenges any easier, but it creates a virtuous circle between the inevitability of the transition, the attraction of talent, the tipping of the balance of risk in favor of net-zero solutions, and progress towards net zero.

That leaves only one question – particularly in the light of this past year’s deeply troubling temperature anomalies – will we get there in time?

https://www.cleaningup.live/audioblog-1 ... -harder-1/
https://www.cleaningup.live/audioblog-1 ... ii-easier/

[lpm]

The change in home solar in the last 2 years alone is amazing. Panels are now so cheap that north facing roofs are worth it.

But it takes time for hunan inertia to catch up - installers still think in terms of a handful of panels on the south roof.

And the strategy had reversed 180 - instead of minimising export during the day via high house/car usage, Octopus pricing now makes exporting the priority.

The UK is on track to becoming a net energy exporter and cheaper energy bills are better than a tax cut. If Labour invests properly there's a major standard of living improvement to come in 2030 to 2040.

[bjn]

Home solar is one of those use cases where the new tech moves into places older tech didn't. Apart from the edge cases of a few petrol generators and the odd hippy windmill, nobody was generating electricity at home. This is an new market that wasn't available to fossil fuels which will only add to the growth of renewables.

[Grumble]

Home solar is one of those use cases where the new tech moves into places older tech didn't. Apart from the edge cases of a few petrol generators and the odd hippy windmill, nobody was generating electricity at home. This is an new market that wasn't available to fossil fuels which will only add to the growth of renewables.

It seems to be the people who were making the odd hippy windmills who are now major movers and shakers in big wind companies.

[IvanV]

Excellent two episode podcast series from Michael Liebreich, "Net Zero Will Be Harder Than You Think – And Easier"....

A few years ago I found this paper, I don't seem to be able to relocate, which considered which technologies potentially can reduce cost and how much, based on historic experience, and the state of those technologies.

The key point was that while PVs, batteries and electrolysis tech can, in principle, continue to reduce in cost substantially - not that we can predict how fast or up until what point - wind can't.

And the evidence is that wind cost reduction is slowing down, to the extent that the cost has actually gone up just recently. That doesn't mean it can't start coming down again, but it's not going to do it very fast. The amounts that people bid in the past to be subsidised to build wind farms had some of the forward-looking cost reduction baked in - it didn't represent the current cost of wind farms, but rather the predicted cost of wind farms a handful of years into the future. Now it has got stuck, there has been construction cost inflation, and no one bid in the last round of sea-bed auctions, because there was a reserve price based on previous auction prices plus an assumption of continuing cost reduction.

It is not implausible that PV can be so cheap we can use it very wastefully. Whether and when, these are hard things to know, but it isn't something we can say will never happen. So it is PV over-production that is the best chance to be electrolysing hydrogen at scale, even in dim windy places like here. But to make it worthwhile using the over-production to electrolyse water, the electrolysers also have to be cheap enough we can use those very wastefully, because the surplus electricity from PV will come in narrow windows of time. When and if that might all add up, we can't say, but it might happen. It seems unlikely it happens soon enough that we can build it out at economy-sized scale in time for 2050. But by 2100, it might happen.

The fortunate serendipity here is the discovery of potentially large amounts of mined hydrogen. Because we have been doing nothing about the hard problems of decarbonisation, just kicking the can down the road, hoping something will turn up, to be frank. And maybe that's the thing that has turned up that will rescue us from our dilatoriness. Because we have been sufficiently dilatory that achieving net zero by 2050 has become very difficult, even if PV and batteries continue to reduce in cost fast. Because energy storage at scale with batteries is something truly massive.

The other important recent development that makes decarbonisation more feasible is heat pumps with 60C-70C water at acceptable COP. That will make decarbonisation of domestic space heating much more economically possible than it was a little while ago, when we were talking about the need to additionally spend multiple £10k amounts to insulate our old badly insulated housing stock to modern construction standards. It is still going to cost people more to buy and install a heat pump than replace a gas boiler, like £5k-10k more per installation. But that's a lot better than the many £30k and £70k installations that looked likely a little while ago.

[lpm]

Mustn't ignore the long lags. If the cost of offshore windfarms stops falling, it can still have 25 years of new building.

The old belief was the UK had great wind resources, useless solar. Now it has great wind, great solar in summer, terrible solar in winter. Which has the obvious implication of needing to become the world leader in seasonal storage.

[IvanV]

<IEA crap graph>

People gave up on the IEA being able to say anything sensible a long time ago. And so they set up the International Renewable Energy Agency (IRENA) in 2009. I can't say I'm totally impressed with them either, but it does produce some useful information. This is its 2023 renewable energy market analysis report. As you see, it only reports history and avoids forecasting anything. Because such forecasts are too politically charged, and as IEA shows you get your arm twisted into showing nonsense. But it does show falling costs of renewable technologies, and you are left to conclude for yourself what that might presage.

[bjn]

And the evidence is that wind cost reduction is slowing down, to the extent that the cost has actually gone up just recently. That doesn't mean it can't start coming down again, but it's not going to do it very fast. The amounts that people bid in the past to be subsidised to build wind farms had some of the forward-looking cost reduction baked in - it didn't represent the current cost of wind farms, but rather the predicted cost of wind farms a handful of years into the future. Now it has got stuck, there has been construction cost inflation, and no one bid in the last round of sea-bed auctions, because there was a reserve price based on previous auction prices plus an assumption of continuing cost reduction.

Is that before or after inflation is taken into account? My understanding that the price rise was linked to steel, concrete and other commodities needed to build turbines had surged. This hit most other construction projects as well, including Hinkley Point C which had another billion or two added it it.

That PV has dropped markedly in price before taking inflation into account is pretty astounding. Not everything scales quite that well.

[IvanV]

And the evidence is that wind cost reduction is slowing down, to the extent that the cost has actually gone up just recently. That doesn't mean it can't start coming down again, but it's not going to do it very fast. The amounts that people bid in the past to be subsidised to build wind farms had some of the forward-looking cost reduction baked in - it didn't represent the current cost of wind farms, but rather the predicted cost of wind farms a handful of years into the future. Now it has got stuck, there has been construction cost inflation, and no one bid in the last round of sea-bed auctions, because there was a reserve price based on previous auction prices plus an assumption of continuing cost reduction.

Is that before or after inflation is taken into account? My understanding that the price rise was linked to steel, concrete and other commodities needed to build turbines had surged. This hit most other construction projects as well, including Hinkley Point C which had another billion or two added it it.

That PV has dropped markedly in price before taking inflation into account is pretty astounding. Not everything scales quite that well.

Googling around, it seems that there was a construction cost increase of about 30% for onshore and maybe 40% for offshore suffered over roughly a 2-yer period when the worst of the Covid construction cost inflation was occurring. That would appear to be about double the general rate of construction cost inflation over the same period. Recent indications, at least for onshore, is that costs may be stabilising at the higher level.

These are clearly increases ahead of other kinds of inflation. Doubtless the trend of building ever bigger turbines to reduce cost per MW, that was the main driver of cost reduction leading up to that point, will continue. And maybe further innovation in the detailed construction of the things.