Power lines down during cold snap in US
Renewables have captured the public’s imagination,but can they actually be scaled up to power the entire nation, ask Mike Conley and Tim Maloney? In their new ebook, available at RoadmapToNowhere.com, they present their reasons why they are convinced 100% renewables is a myth – and why we should rely primarily on nuclear power.
Extraordinary claims requires extraordinary evidence.
The claim that the United States, much less the entire world, can be adequately powered by 100% renewable energy is extraordinary, indeed.
The claim that we can have an all-renewables grid with no backup from fueled power plants, and practically no energy storage, is even more extraordinary.
To confirm or dispel our doubts, we ran the numbers on the industry’s most highly regarded proposal, the Solutions Project’s 50-State Roadmap to 2050.
We unpack the 35-year strategy in our short, non-technical online book Roadmap to Nowhere. With all due respect to the Solutions Project, it’s not a solution.
Fuel-free systems
For $15.2 trillion, with no fueled backup and next to no energy storage, the Roadmap proposes a 1,591-gigawatt national grid of 100% renewables:
- A half-million ginormous 5-megawatt wind turbines, on land equal to New York, Pennsylvania, Vermont and New Hampshire, and in open sea regions equal to West Virginia
- 18 billion square meters of solar panels, on land equal to Maryland and Rhode Island
- Concentrated Solar Power (CSP) with thermal energy storage, on land equal to Connecticut
- Rooftop solar on 75 million homes, and nearly three million businesses
To compare, a 1,591-GW nuclear grid would cost less than $7 trillion, depending on the reactors used, on land equal to half of Long Island. With factory-built reactors and a streamlined regulatory process, it could be completed in ten years.
The essential difference between nuclear and renewables comes down to this: Fuel is energy storage. Renewables are fuel-free systems.
Mother Nature stores energy in substances we call fuel: stable, storable, portable stuff that we can use to generate power, when and where we want it.
Civilization has advanced by exploiting ever more energy-dense fuels: wood, coal, petroleum, gas and nuclear. The recent interest in renewables appears to be a reversal of this historical trend, in the sense that wind, water, and sunlight are typically regarded as less dense forms of fuel.
After sixty years of worldwide commercial nuclear power, the death rate per terawatt hour is lower than for solar or wind
Except they’re not really fuels at all. Renewables are fuel-free systems that exploit ambient natural phenomena. But the light and motion they exploit are not stable, storable, or transportable. They must either be utilized on the spot to make energy, or converted into something that can be stored for later use, such as the electricity in a battery, or the potential energy of water pumped uphill for hydroelectric power.
Renewable “fuel” may be free, but collecting and exploiting its energy is expensive. Converting it into smooth, dependable, on-demand power is even more expensive. Wind and sunshine ebb and flow, come and go. They should never be relied upon without substantial backup and storage.
Hyped-up fear
Since it’s an undisputed fact that fuel can power the nation, you would think the response to climate change would be a transition to carbon-free fuel, rather than a transition to fuel-free systems. In our view, the hyped-up fear of radiation and waste is the reason we don’t already have a nuclear grid.
No one died from the Fukushima meltdowns or from Three Mile Island, and no one will ever build a reactor like Chernobyl again. After sixty years of worldwide commercial nuclear power, the death rate per terawatt hour is lower than for solar or wind (and of course much much lower than for fossil fuels). And that’s with factoring in the total projected casualties from Chernobyl.
Nuclear’s energy density is millions of times greater than wind and solar, with less than 1% of the footprint. The technology is well proven, the fuel is abundant, and the energy is carbon-free.
Reactors could even power commercial shipping, and the factories to synthesize carbon-neutral fuel for air transport. Either one would make a serious dent in CO2 emissions.
When we run low on our finite endowment of natural gas (which should be any decade now; even sooner if we export the stuff), then what?
But instead of deploying a clean, compact and scalable technology to power the nation, a stupendous inventory of equipment is being subsidized and deployed on vast tracts of land, to collect the fitful energy of wind and sunlight.
And if it can’t be used at once, the energy must either be wasted or stored, if we can afford an adequate means of storage. But just one grid-day of storage for the bare-bones Roadmap raises the price to nearly $23 trillion.
It gets worse: Before the 35-year buildout is complete, we’ll also have to refurbish the 350,000 onshore wind turbines at least once, and the 150,000 offshore turbines three times or more.
And even when it is complete, it’ll never end: Solar panels only last about forty years. To maintain 18 billion square meters of panel, our renewable-industrial complex will have to fabricate, install, and recycle 1.23 million m2 every single day, without a break – forever.
Utility power plants should be compact sources of reliable power, free from the vagaries of weather, climate, season, or time of day, and under the operator’s control at all times. In a word, they should be decoupled from the environment.
Sprawling farms
The Roadmap would occupy over 130,000 square miles, plus the offshore region, and millions of rooftops. Utterly dependent on favorable weather, and without fueled backup or mass energy storage, its sprawling wind and solar farms will be utterly dependent on each other as well.
Fueled power plants are IN-dependent. Wind and solar farms are INTER-dependent.
Coal, gas, hydro and nuclear can operate on their own, in any weather. But since coal is verboten and nuclear is the work of the devil, the Roadmap’s wind and solar farms will rely on natural gas training wheels, until they get their collective act together and roll with the big boys.
Before that happy day arrives (if it ever arrives), the wind and solar farms that are up and running will actually be natural gas plants, supplemented with renewables. As Robert F. Kennedy, Jr. said, “The plants that we’re building, the wind plants and the solar plants, are gas plants.”
Indeed, his Ivanpah solar farm has been hit with a penalty for excessive CO2 emissions. They apparently used 62% more methane last year than predicted.
One day of storage for the Roadmap would cost more than an entire nationwide all-nuclear grid
Which raises an interesting question: When we run low on our finite endowment of natural gas (which should be any decade now; even sooner if we export the stuff), then what?
Existing battery technology is completely inadequate to back up the grid. All the lithium mined on earth in 2016 would give us a whopping eighteen minutes of all-grid battery storage. If all the vanadium mined in 2015 were devoted to flow batteries, it would provide one minute of storage.
Pumped hydro is the only existing storage technology that can adequately scale up. The problem is, we would need 156 billion cubic meters of water to generate one grid-day of power. That’s our national fresh water consumption (tap water, irrigation, the works) for more than four months.
Even at the bargain-basement price of $0.20 per installed watt-hour, building one grid-day of pumped hydro storage would cost $7.6 trillion, more than the price of an entire nuclear grid.
That bears repeating: One day of storage for the Roadmap would cost more than an entire nationwide all-nuclear grid.
It would be folly to commit to a project like the Roadmap, hoping that a breakthrough storage technology comes along. We don’t have the time, or the money, to explore the possibilities of a fuel-free lifestyle while embarking on a quest for the holy grail of cheap storage. It already exists. It’s called fuel.
In fact, uranium fuel is so energy dense that a nuclear grid would have more than 500 days of storage built right in: the fuel rods in the core of each reactor.
We only have this one chance at getting global de-carbonization right, so we have to build our energy future with proven and scalable technology. Because we’re not betting the farm, we’re betting the planet. And nuclear is the only carbon-free energy source that we absolutely know is up to the job.
Germany’s Energiewende is a cautionary case in point: Replacing their reactors with wind and solar has actually increased their CO2 emissions, from all the extra coal they’re burning to back up their renewables.
A fuel-free renewables grid would make every region utterly dependent on each other, whether they liked it or not. In a very real sense, a nationwide, interdependent grid would be the essence of Big Energy
And if the rest of the world follows their lead, we’ll have more to contend with than increased emissions: A global Roadmap’s solar panels would monopolize 90% of the world’s proven silver reserves, and one-third of proven copper reserves. (Transmission lines would be extra.)
In contrast, Generation-IV reactors would actually eliminate most long-distance transmission corridors. Since most of them won’t need water cooling, a Gen-4 can be placed wherever the power is needed, even in the harshest desert.
Green elephants
Aside from all the foregoing, the Roadmap has yet another drawback to consider. Once we start down that road, we’ll have to go all the way. Because the only possible chance to make the Roadmap work is to build all of it, or nearly all of it. That’s what a self-supporting, interdependent system is all about.
If we embark on a national buildout of fueled power plants and abandon the effort halfway through, we’d still have a collection of fully functioning, independent power plants. If we abandon the Roadmap halfway through, we’d have a herd of green elephants that will always need training wheels.
For the Roadmap to work, tens of thousands of wind and solar farms will have to be built in favorable weather locales. The problem is, we are nowhere near making long-term predictions about the weather. Climate yes, but weather no.
What if our wind-blown Northern Tier becomes the Northern Doldrums? What if Texas becomes the Monsoon State? A long-term weather shift could markedly degrade the productivity of wind and solar over a wide geographic area.
Go nuclear or go extinct
Supporters of renewables accept the science on climate change, and have great respect for Science Itself. And yet, they have embraced a multi-trillion-dollar scheme with a 35-year industrial mobilization, the success of which will ultimately depend upon accurate long-term weather forecasting, in a future of ever-growing climate disruption.
Political flare-ups are all but guaranteed, when a fiercely independent region finds itself exporting power to another region whose long-term weather luck has gone bad and stayed there.
A fuel-free renewables grid would make every region utterly dependent on each other, whether they liked it or not. In a very real sense, a nationwide, interdependent grid would be the essence of Big Energy.
And what happens if a region backs out of the Roadmap, and switches to, say, nuclear power? Could the Roadmap be re-drawn to work with the states that remain? Nobody knows, until we go down that road and see what transpires.
There is a school of thought that says we need to power down civilization. While it’s true that we as individuals should consume less energy, we as a global civilization actually need to power up.
Simply put, the world needs all the clean, carbon-free energy it can get. But there’s a catch: That energy source will have to be cheaper than coal, and just as reliable. Or the world will keep right on using coal.
Humans are like that. We’ve always had tribal minds, but now we have a global reach – big world, small planet. And there is no Planet B.
The good news is, the technology to cleanly power the planet already exists, without reinventing storage or hoping the weather cooperates. It can be deployed at the scale we need, precisely where we need it, in the time we have to act.
And if a nuclear-powered future seems too risky to consider, imagine growing old in a crowded, desperate, heavily-armed world of +4 degrees C, with rising seas and rolling blackouts.
The road ahead will be rough. But a reliable supply of cheap, clean and abundant energy will significantly improve our ability to adapt to climate change and mitigate its worst effects.
This is the challenge of our era, and will always be our legacy.
Go nuclear or go extinct.
Editor’s Note
Mike Conley is a writer living in Los Angeles who has been studying energy issues for several years. Tim Maloney is a retired community college professor of Electronics Technology and Machine Control, with an MS in Electrical Engineering and a PhD in Educational Psychology from the University of Toledo, and a BS in Engineering from Case Western Reserve University.
The authors are long-time members of the Thorium Energy Alliance, an advocacy group for the widespread acceptance and deployment of thorium-fueled Molten Salt Reactors.
They stress that they “don’t have a beef with renewables – other than the claim that they can be scaled up to power the entire national grid.”
See RoadmapToNowhere.com to read the free online book, or to download the pdf. An ebook will soon be available.
Check out the video at: https://www.youtube.com/watch?v=7O7bB1ghqvU
[adrotate banner=”78″]
Oh my god, not again. The nuclear mirage.
Well, thankfully it won’t happen. Way to expensive even now. By the time any new nuclear plant was finished the price difference with renewables would be even larger.
And Trump’s wish to subsidise coal and nuclear was rejected by the Federal Energy Regulatory Commission yesterday
A 100% renewables grid would be far more expensive than a nuclear grid. And far less dependable. We walk you through the numbers in the book. And, Mark Jacobson agrees with our price estimate for his bare-bones Roadmap ($15.2 Trillion.) An all-nuclear grid could be built in the US for less than half that, on a tiny sliver of the land. And have enough energy storage (in the form of fuel) for 18 months.
Funny, not that long ago utilities were convinced that grids would not be able to deal with more than 5% of renewables. Now we are arguing over the last 10 to 20% of renewables. By the time we will actually get to 80 or 90% of renewables the cost of solar power, wind power and storage will have come down even further and what seems cost-prohibitive now will happen without anybody noticing it as something special.
Which storage method do you think can be scaled sufficiently to stabilize a U.S. national renewables grid?
I would use power lines to Canada, to mexico and inside the US. including interconnecting Alaska to the rest of the US. Some swiming Offshore power some km out from th e coasts, some demand management with distributed thermal storages for heat and cold – not much need for electricity storage will remain then.
Changes for this to be done in the next decades- when things are replaced any way – in most cases.
All hand waving. How about some systems analysis, with costs?
The entire book is a systems analysis, with costs. And with hundreds of footnotes which contain the math that explains the analysis, and the source material we based our calcs on.
I don’t think there will be a single silver bullet. Demand side management, better connection of grids, lithium-ion batteries, flow batteries, supercapacitors, use of abandoned mines for pumped hydro, off-river pumped hydro, smarter use of existing hydro capacity, overbuilding of renewables and using the excess power for non-time critical services, and probably some technologies that currently only exist in the laboratory will all play a part.
We do refer to Rube Goldberg in Chapter 2.
But overbuilding of renewables is refused by the Roadmap.
When the share of wind+solar increases towards above 50%, wind+solar will produce more than needed during ~25% of the time.
That superfluous electricity will cost near zero, so it can be the base for competitive priced renewable gas; hydrogen (H²) and methane (CH⁴).
Nowadays the conversion to H² in unmanned Power-to-Gas plants (PtG) can be accomplished with yields of 80% – 90%. Even the conversion to CH⁴ is now at >75% and moving towards 80%. Check e.g:
https://goo.gl/dS23oF
The renewable H² can be stored in deep earth cavities such as empty gas fields, salt domes, rock cavities. NL and Germany use it to store huge amounts of energy (a winter supply). USA has far more of those cavities.
When needed the H² can be easily retrieved and converted into electricity in high yield fuel cells. Those are now gradually getting towards yields of 80%.
These processes require little human control & management, so they are cheap.
And storage in deep earth cavities is of course extreme cheap!
Far more dense populated Germany follows this route. They have many MW scale PtG pilot installations: http://www.powertogas.info/power-to-gas/pilotprojekte-im-ueberblick/?no_cache=1
They plan to start with regular rollout in 2024 (no need before wind+solar produce >50%).
If
a) The wind farms are gradually re-powered with tall towers and longer blades
b) tracking solar share increases
c) bi-facial solar increases winter solar output
d) energy efficiency reduces winter heating and lighting demand
e) conventional hydro is reconfigured for better load following, i.e. higher peaks lower minimums which will mean greater changes in reservoir and some stream levels,
Germany can probably reach 70%+ renewables before significant investments in P2G or large scale storage are required
Agree.
Note that wind+solar >50% implies a renewable share >61% due to the contribution of hydro + biomass.
https://www.energy-charts.de/energy_pie.htm?year=2018
As those are dispatchable, their production will concentrate on periods without wind & solar as then prices are higher.
The numbers still aren’t there, Peter. We calculated with the best turbines and panels we could find, and we even used Jacobson’s overinflated capacity factors.
Best case scenario is that a 100% RE grid is twice the price as a Gen-III 100% nuclear grid. And the RE grid cost only includes 5% storage (Jacobson’s idea, not ours.)
24 hrs pumped hydro storage for his national grid, at 20 cents per watt, would cost more than an entire national nuclear grid.
Again: the batteries for Jacobson’s grid alone would cost more than a nuclear grid.
See the book on the website for cost breakdowns.
I am actually not defending Jacobsen’s work so much as attacking yours.
a) The current dispatchable electricity system in the US FF, hydro and nuclear produces about 3,600 TWh per year. Capacity is about 8,700 GW i.e. the system runs at 49% capacity factor. I.e. for dozens of different reasons you need installed capacity to be around double the average load. You have not built that into your cost.
b) your costs for reactors are not supported by any existing project
c) you have neglected water use or if you intend to have most of the reactors by the sea, then you have vastly underestimated transmission cost
2. Re Renewable capacity Germany with old technology generates about 660 MWh per square km. If its old wind turbines were replaced on a 2 new for 3 old, and the same with solar it would be producing about 1,500 MWh per square km using no more land than it does now. The US has much better wind, solar and hydro resources than Germany and a much higher share of open land so it should be able to generate about 2,100–2,500 MWh per square km. The 48 states have an area of 8,080,000 square km so in total they can generate at least 17,000 TWh or 4 times the current electrical demand with the same density of wind and solar that Germany has today.
So space is not an issue, water is not an issue, transmission is less of an issue because about 25% of supply will be from rooftop solar (see NREL 2016) and more than 65% of the area of the 48 states is suitable for wind
If the US was to generate 40% its current electrical load from wind it would need another 75-95,000 3.5-5 MW class wind turbines. over 20 years that is about 4,100 turbines or 16 GW per year. That is only about 40% more than the GW rate the US will achieve in the next two years. In number of turbines it is actually about the same Put it another way Scotland whose economy is 1% of the size of the US installs one wind turbine a day. The US could therefore install 100 it actually needs to install 11/day. Again the Industrial capacity of the US is about 25 times that of Australia. Australia will commission 2,5 GW of wind this year with a hostile government. The US should therefore be able to install 60 GW+ per year or around 4 times the required rate.
In 2017 Germany installed 6.3 GW of wind. The US economy is 5 times as large so annual installation capacity should be 30 GW per year
If they need to be replaced every 25 years that is 3,700 units per year. Assuming complete replacement of the towers and turbines that is about 2,5-3 m tonnes of equipment. US industry produces 25-30 m tonnes of cars and light trucks. clearly not a big issue
In summary few of your assertions stand up to scrutiny
a) We limited our grid to 1,591-GW, which is Jacobson’s estimate for what we’ll need by 2050.
Present-day US reactors that run full-on to deliver baseload power do so at around 85% – 90% cap. factor. If their capacity is significantly lower than that, they’re probably being dialed back in favor of RE power.
The issue with space/land use for wind and solar isn’t so much the volume of land per se, but the far-flung dispersal of sun and wind-gathering equipment as a national energy solution. Logistics and labor rise as your equipment spreads out. There is no good reason to build a dispersed energy system if we don’t absolutely have to. And we don’t.
We cover the industrial buildout issues of the US quite thoroughly in the book. Yes, the US could probably do it, but why do it if we don’t have to, and we don’t. Building out Jacobson’s grid would amount to a 35-year mobilization on par with what the US home front did in WW II. That was 3.5 years. Jacobson’s plan would require 10X that much time and busyness. Which is nuts. Why go through all that labor and manufacturing to build a dispersed, intermittent, fuel-free grid, when we have the technology to do otherwise?
Wrong on b). The costs for our Gen III reactors are based on the existing and ongoing South Korean installations in the UAE, of around $4,500 an installed kW. ThorCon, Terrestrial, etc are projecting about half that, but we went with existing technology.
I could go on and on, but:
In summary, few of your rebuttals stand up to scrutiny. Most everything you write about here is addressed in the book.
I suggest you actually take the time to read the entire book before you spend any more time posting about how flawed it is.
You don’t know the costs of S. Korean build nuclear in UAE. Also since those experience major delay’s.
First unit would start in 2017 but it will probably become 2020…
It’s one of the reasons your estimates are unrealistic.
Thorcon made no progress in past years to realize its dream…
There are a number of reasons that US nuclear reactors run at 92% CF.
a) they can mostly be balanced quite well with nearby hydro, coal or gas plants. As the nuclear plants have zero or even negative short run marginal costs they will always win the auction and hydro, gas and coal will provide the balance. That works well as long as nuclear never supplies more than about 60% of the minimum load. In this case as nuclear only supplies 20% of US power generation then curtailment of nuclear through lack of demand rarely happens.
b) many US nuclear plants were designed quite conservatively for obvious reasons. New methods of modelling heat flux and finer control as well as plants upgrades allow many plants to run well above nameplate capacity for extended periods so the 92% figure is probably exaggerated any way. The true figure compared to actual capacity rather than original design capacity is probably 85%, still very good but it can only be achieved because the nuclear share is so low.
Whatever the stated costs of the Barrakah nuclear plant, they expect to deliver power (eventually?) for US$115/MWh and the Emirates have included no more nuclear in their new energy plan. Why have 23 proposed nuclear plants in the US dwindled to two even though the production tax credit $18/MWh was higher than that for wind, they got free cooling water, loan guarantees from the federal government and free catastrophe insurance but they still can’t make the sums add up. Why have both Hitachi and Toshiba cancelled plans in the UK after spending 100’s of millions of dollars with a potential price of 75pds per MWh
In almost all other fields of human endeavour decentralisation and local autonomy are increasing/ PCs vs mainframes etc. Even Amazon is building warehouses closer to customers. People want control and convenience as much as possible, that is why they are putting in rooftop solar and soon batteries even if the superficial economics are marginal.
According to NREL 35% of US demand can be supplied with 16% efficient rooftop solar. Now that 20% efficiency is available and 25% available well before the next nuclear plant, while thermal storage is becoming more viable, the optimum figure for rooftop PV for the US today is probably closer 45% of annual generation with no transmission losses. Then as more than 65% of the US land area is suitable for high CF wind and 85% is economical for ground mount solar, average transmission distances and volumes will actually fall with a high renewable grid so your argument is similar to that of railway companies that rail transport is cheaper than roads, build more trains and less roads. how has that worked in the US.
Re Civilian mobilisation. This year Australia will install 2.5-3 GW of wind and about 3 GW of utility solar. The renewable workforce is about 0.1% of the total workforce.
US industrial capacity is around 25 times that of Australia so you could install 60-70 GW of wind and 70-80 GW of utility solar per year. To supply all household energy, land transport and most industrial application you need about 6-7000 TWh/yr. or 5-6,000 TWh of new supply. At the above installation rates that will take about 16 years with 0.1% of your workforce. Even doubling the workforce for grid modernisation and storage means that you would need to employ about 320,000 people about 3/4s of the number of people employed as prison guards in the country. so your numbers again are out by a couple of orders of magnitude
The only estimate I have seen for SMR’s is NUscale which is $5,000/ kW. Even allowing double the annual AEP that is still higher than wind+solar. Thorcon haven’t a clue because neither they nor anyone else has solved the material science issue related to 700 C molten salt reactors.
a) 3600 TWh is correct for FF, hydro and nuclear. Counting biomass, dispatchable NRG was 3770 TWh in 2017.
US electric capacity is 1200 GW, not 8700 GW. Of that, about 1100 is dispatchable. See http://www.timothymaloney.net/Critique_of_100_WWS_Plan.html
Dispatchable capacity factor is 40%, not 49%. [3770 TWh actual / 9460 TWh potential = 40%]
I agree that we should have dispatchable capacity equal to at least double the US average load, which is 467 GW. That’s one of our objections to Jacobson’s Roadmap – not enough overbuild.
b) KEPCO’s Generation 3 project in UAE came in at $4.40 /Wavg. See FN 12 of our Ch 1.
Sanmen Generation 3+ units 1 and 2 apparently came in at about $5.50 /Wp, or $6 /Wavg.
China National Nuclear Corporation expects their learning curve to bring construction cost down from Sanmen’s 2018 cost of $5.50 to $3.00 /Wp, about $3.30 /Wavg. See http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
a) 3600 TWh is correct for FF, hydro and nuclear. Counting biomass, dispatchable NRG was 3770 TWh in 2017.
US electric capacity is 1200 GW, not 8700 GW. Of that, about 1100 is dispatchable. See my Critique of 100% WWS Plan.
Dispatchable capacity factor is 40%, not 49%. [3770 TWh actual / 9460 TWh potential = 40%]
I agree that we should have dispatchable capacity equal to at least double the US average load, which is 467 GW. That’s one of our objections to Jacobson’s Roadmap – not enough overbuild.
b) KEPCO’s Generation 3 project in UAE came in at $4.40 /Wavg. See FN 12 of our Ch 1.
Sanmen Generation 3+ units 1 and 2 apparently came in at about $5.50 /Wp, or $6 /Wavg.
China National Nuclear Corporation expects their learning curve to bring construction cost down from Sanmen’s 2018 cost of $5.50 to $3.00 /Wp, about $3.30 /Wavg. See http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
a) 3600 TWh is correct for FF, hydro and nuclear. Counting biomass, dispatchable NRG was 3770 TWh in 2017.
US electric capacity is 1200 GW, not 8700 GW. Of that, about 1100 is dispatchable. See my Critique of 100% WWS Plan.
Dispatchable capacity factor is 40%, not 49%. [3770 TWh actual / 9460 TWh potential = 40%]
I agree that we should have dispatchable capacity equal to at least double the US average load, which is 467 GW. That’s one of our objections to Jacobson’s Roadmap – not enough overbuild.
b) KEPCO’s Generation 3 project in UAE came in at $4.40 /Wavg. See FN 12 of our Ch 1.
Sanmen Generation 3+ units 1 and 2 apparently came in at about $5.50 /Wp, or $6 /Wavg.
Compare to utility scale PV solar, our FNs 7and 10 of Ch 11. http://www.nrel.gov/docs/fy16osti/67142.pdf
$1.75 /Wac / 21% CF = about $8 /Wavg.
China National Nuclear Corporation expects their learning curve to bring AP-1000 construction cost down from Sanmen’s $5.50 cost in 2018 to $3.00 /Wp, about $3.30 /Wavg. See http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
All valid and true, Bas. But the cost calculations in our book would still apply. From them, you will see that reactors are cheaper, more compact, and more dependable than any combination of wind or solar. By running reactors as baseload, their off-peak power can be used to run the processes you describe.
It is true that a reactor can be used to provide off peak power to run P2G, but the more discount you give to one user class the more you will have to charge another to recover the fixed costs and the fixed costs per MWh from either a Gen III or a Thorcon reactor are at least double those of wind and solar
But the calculations are wrong. You have not allowed anywhere near enough reserve capacity and have assumed a much higher availability than is achievable not to mention the costs of building and operating a vastly stronger grid that would be needed to minimise investment in generation
Please explain: ” the more discount you give to one user class the more you will have to charge another to recover the fixed costs and the fixed costs per MWh from either a Gen III or a Thorcon reactor are at least double those of wind and solar.”
We confined our hypothetical all-nuclear grid to Jacobson’s parameters for his national RE grid – the same reserve capacity, and the same assumptions about the strength of the national grid.
Most of the costs of a nuclear plant, capital repayments, interest, depreciation, security, even grid access charges licencing etc. are fixed some are slightly variable, staffing, some maintenance and fuel but about 80% of the cost is an annual cost, independent of production volumes. Therefore assuming the average fully absorbed cost is for example $80/MWh which is (2/3rds of the cost of Plant Vogtle). If you sell 20% of your production at $18/MWh to a P to G, Pumped hydro or battery plant then you have to charge everyone else $95/ MWh to break even. You can’t charge more than $18/MWh for these residual uses because the storage plant can build its own wind and solar for that price
Thanks Mike!
So we agree that cheaper PtG=>S=>GtP (with some biogas/-mass & (pumped) hydro) can cover periods without sun & wind.
Hence the role of batteries will be small => so those costs also.
Not really. P2G, like hydrogen production, has a high round trip cost. I think the best mass energy storage — other than the fuel rods in a reactor — is the pumped hydro / gravity piston. Small footprint, and 90% round trip efficiency.
Powering the entire country with renewables — primary energy, not just electricity — will require about 1,600 GWs by 2050, per Jacobson. 24 hrs of storage at current pumped hydro’s average price of 20 cents a watt, would cost more than an entire national buildout of reactors. And pumped has the best round-trip efficiency of any battery storage, and would last at least 100 years. Until batteries can substantially improve on those stats, wind and solar will never anchor the national grid.
Storage capacity of pumped hydro is low in most areas (few days?) and expensive. While that of gas is ultra cheap and can cover winters, even years of consumption!
Roundtrip of pumped is ~80%, but PtG allows for waiting until real low prices arrive (typical; store in summer for use in winter as done in NL).
The lower electricity purchase price, thanks to the waiting, more than compensates the lower roundtrip yield (~60%).
Even in dense populated NL (>10time more than USA) scientists are convinced that we can migrate towards 100% renewable for all energy without high costs…
A closed cycle, gravity piston, pumped hydro facility – the kind I favor – would not have water the supply issues of an open system.
Closed cycle, gravity piston, pumped hydro doesn’t have enough storage capacity.
Not sure that “expensive” is the way to present either proposal.
$15.2 trillion is a mere 10 years of U.S. fossil fuel expenditures, not counting subsidies. Considering the life-span of RE’s, and the quadrillions of dollars they can save us in adaptation costs, perhaps we should be using terms like “bargain”, “nominal”, “bought for a song”, “inexpensive”?
Roger – The optimistic lifespan of a PV panel is 40 years. Between major overhauls, onshore wind equipment lasts @ 25 years, and 10 years for offshore.
Jacobson recommends 18 billion square meters of 188-Wdc panels, 342,000 onshore 5-MW turbines, and 156,000 5-MW offshore turbines.
That would require the swapout of 1.23 million square meters of panel per day, and the initiation of 80 major overhauls, per day.
In contrast, 6,000 SMRs (small modular reactors) @ 250 MW, would generate the same energy Jacobson hopes to generate with the above RE.
With an average 7-year runtime between servicing and refueling, those SMRs would require 2.3 swapouts per day, with the old modular reactor taken back to a facility for refueling. And, those 6,000 SMRs would have 18 months of energy storage in their fuel rods, available 24 / 7.
Renewables are utterly dependent on favorable weather. Which is becoming less and less predictable by the day. Should we really invest $15 T on equipment that could easily become stranded due to an unforeseen shift in long-term weather patterns? Indeed, why should have to depend on favorable weather to gather the power we need?
So you assume a change of weather pattern would produce endless night or complete loss of wind?
And the mayor overhauls every 7 years of SMR are not significand different that those of the smaller wind turbines every 10-15 years, that’s standard, and in case of the wind turbines not a big job. Transporting highly radioactive components is something completely didfferent in cost and time.
18 billion squaremeters of today PV equipment would be 3.6 trillion W capacity – ptoducing about 5000 TWh per year off power. 500.0000 5MW turbines so 2,5 TW capacity at 3000 FLh would produce 7500 TWh of power. Yor 6000 SMA would produce less than half of the combined power of these two systems.
the recycling of 1,3 mio m² per day would mean 4,5 mio t of material each year, so about the same material moved for a singe coal plant so less than 1 promille of what is moved for fossil fuel supply today. I am quite sure uranium mining for the smr would move more material.
No, Helmut. A change of weather pattern could diminish wind or increase cloud patterns to where the wind or solar frarm is no longer cost-effective, and/or to where your storage and backup predictions no longer apply.
You’re citing peak capacities, not average. In the US, average PV is about 16.7% of that 3.6 trillion W capacity. Wind is optimistically around 40% capacity, so slice 60% off your turbine production numbers.
Working with average rather than peak capacities resolves the apparent discrepancy you’ve found in our SMR estimates.
,
I do not calculate with peak capaciies but with typical load factors of 1500 ful load hours per year, which show up when solar power is placed where it is in tendency a but more sunny. The best areas deliver much more power. 3000 full load hours is less than 40% capacity which you state, so I would have to increase the production numbers for wind when I follow your numbers.
Helmut – the capacity factors we used are 21% for PV, and ~38% for wind.
We derived these numbers from the Roadmap’s tables and footnotes, which are higher than in-the-field numbers in the US. Actual PV average in the US, for example, is about 16.7%.
Panel longevity of 40 years is from SunPower’s own estimate of their latest 188-peak wdc/m2 panel. We used that panel, even though it is more efficient than the 134-wdc PV used in the Roadmap’s calculations.
Plus, we calculated based on at least 70% of the PV panels being placed in the most optimum locales in the southwest deserts, on flat, clear land, with the most efficient packing factor.
I suggest you read some of the book, and the footnotes as well (all of which is quite readable.) Tim goes to great detail in our footnotes to explain exactly how we arrived at our figures.
Bottom line, we gave the Roadmap every advantage we could, and every benefit of the doubt, and the numbers still don’t add up.
6000 SMRs rated 250 MW (ThorCon) at 99% CF will provide 13,000 TWh /year. That’s about the same as the Roadmap’s PV panels plus wind turbines.
A molten-salt SMR with nearly 100% fission rate, not the paltry 3% for solid-fuel LWRs, will require negligible mining / material-moving. I calculate only one two-hundredth the amount of material as solid-fuel LWRs providing that same 13,000 TWh.
Even LWRs would require less than 200 million tonnes of uranium ore annually to provide that amount of electric energy. Check out slides 58 thru 61 at
http://www.dirkpublishing.com/dirkpublishing.com/Slideshow_downloads_files/ThoriumNuclearSlideshow.pdf
So call it one two-hundredth of 200 M tonnes, or about 1 M tonnes /year for molten salt SMRs. Compare to the 700 M tonnes of coal mined annually in the US. One seven-hundredth portion.
You will get no argument from me that SMR’s will be better than coal but your own analysis is also full of heroic assumptions. How are you going to achieve 99% CF.
a) the mechanical reliability is nowhere near that good. If you achieve 90% availability over the life, including refuelling cycles would be a major achievement. It has taken 40 years for the US nuclear fleet to reach 92% CF and although molten salt appears simpler, paper inventions always do it will also take 20-30 years to get the bugs out of the system. Remember the EPR and the AP 1000 were cheaper, safer, quicker to build, look who that turned out. and we know a lot more about erosion corrosion from steam than we do about molten salt
b) demand is not flat unless there is a huge amount of storage (that word again). France is a highly electrified economy, it can export surplus nuclear and imports wind coal, gas and nuclear, it has pumped hydro storage and yet its nuclear plants run at an average 72% capacity factor.
Demand on the Australian grid which I am more familiar with runs from 18 GW to 35 GW and our grid of all forms even before there were any renewables ran at best 55% CF.
The net result is that without massive storage and allowing for maintenance, refuelling, breakdowns, even of the grid connections as well as the actual generators, the maximum realistic CF you can use is around 60%, so straight up you have underestimated the investment by around 40%.
MSRs are being designed to refuel on the fly, and SMR shutdown time will be limited to the time it takes to swap out an old SMR for a new SMR – perhaps one entire day, maybe two, every seven years.
The advantages of a hypothetical all-nuclear grid over Jacobson’s Roadmap are so great, that even if we add 40% to our cost estimates for an all-nuclear grid, our main points and our conclusions would remain.
New PWR plants achieve >90% availability. The planned outage rate is around 4% and forced outage rate 4%. There is no reason why SMRs will not achieve similar levels, indeed their forced outage rate should be less than large complicated Gen III reactors. Large reactors need refuelling about every 24 months, whereas marine type reactors on which some SMRs are based are designed for 10 years of operation before needing refuelling. So SMRs should need refuelling far less frequently than large PWRs.
Planned outages for dispatchable gen. capacity is scheduled more than 12 months in advance, often during periods where system demand is low when the missing gen. capacity is not needed, so is quite manageable for the grid operator. Unlike intermittent renewables that cannot be relied on more than a few days in advance.
France’s PWR fleet comprises some plants that were commissioned in the 1970s so one would not expect >90% availability from them. Some older reactors are on long-term outages for life extension. Also some of their PWR plants load follow so reducing their CFs. Comparing EDF’s aggregate fleet availability with new PWRs that achieve >90% is meaningless.
There would be no need for massive storage either providing the UK doesn’t follow Germany down the path of over relying on intermittent renewables. The UK doesn’t have massive storage now. In future the UK’s grid will have greater interconnection capacity with mainland Europe, growing night time electric heating demand and EV charging demand so smoothing system demand troughs and peaks .
The more I read the more unrealistic you are. I am a mechanical engineer with 45 years of product development experience. No-one anywhere has achieved the availability you claim from a thermal plant, let alone a new design. Reciprocating steam and diesel engines have been around for over 100 years, they don’t have to to cope with radiation embritlement or erosion from dense molten fluids and they can’t achieve 95% availability over a 30 year life let alone 99% over 60 years. The steam turbines which your nuclear plants will have last 15 years max between major rebuilds.
I am sorry but you seem to be so naive about new product development that it seems your whole thesis just a dream.
Then in a grid to cope with transmission imbalances and outages as well as equipment downtime you need about 15-20% more capacity than the peak load. In modern grids where steady state industrial loads are falling that means peak capacity must be about twice average annual demand. i.e. my earlier figure was too kind, your investment was not 40% understated . the real investment is closer to 100% more than you claim. Even your base figure is optimistic as that appears to be based on a learning curve that is quite unsubstantiated.
Finally your assumption that early versions of the new plants will last 60 years is just fanciful. The oldest currently operating nuclear reactor is 48 years old. Many of the earlier reactors lasted anywhere from a year to 15 years and a few of the current fleet may at huge expense through long periods offline for almost complete rebuilds may reach 55 years and possibly 60 years. Some cars last 60 years to but most don’t and those that do don’t work very hard
In summary it appears to be a waste of time to to read your book, because the assumptions are so unrealistic.
That is a pity because I was hoping to learn about a promising new approach
As an engineer with 45 years experience, please explain how an energy-dense, weather-independent, always-on mass energy production system, loaded with 18 months of carbon-free fuel, would be a poor choice for powering the nation, compared to an intermittent, weather-dependent, fuel-free energy production system with no built-in backup or storage.
The 99% CF is for the ThorCon idea of two side-by-side Gen-4 SMRs at water’s edge, one fresh, the other aging toward its 7-year useful life. At 7-year maturity, it comes out onto the barge, and a new SMR is taken off the barge and dropped into its place.
92% is the fuel change-determined CF published by Westinghouse for Gen-3+ model AP-1000. Both CF values depend on constant demand for their power, of course.
I have done some reading recently on the Thorium/U233 molten salt reactors and fully agree. The materials problems are probably marginally understood for operation on a time-scale of several years, and barely understood on on a scale of many decades.
” The optimistic lifespan of a PV panel is 40 years. ”
Wow – Sorry, no. Current degradation rates of modern PV panels indicates that they will have 100 – 125 year lifespans. And the definition of lifespan is very conservative – it is when the panel can only out put 70% of its initial rating.
There are panels made in the 1970’s which are still putting out 90% + of their initial rated output.
I have NO idea where you pulled your number from, as it is close to the modern warranty for panels.
If this is the sort of expertise you bring to the table, I gotta ask – why should we believe what two amateurs with demonstrated bias for nuclear have to say, versus Jacobson and Delucchi, who are PhD professors in the field at Stanford, and who have been publishing their work in under peer-review?
Actually, Roger we pulled our numbers directly from Sunpower’s behind. And from NREL’s behind, also too.
As for experts vs amatuers — What we bring to the table is that we can do good old fashioned reading, writing, and ‘rithmetic, the three Rs from American grade school. Applying those basic skills, any objective observer can see that Jacobson’s numbers don’t add up.
Yes, we’re pro-nuclear. But if you really think our calculations are biased, then I’d strongly advise you to actually read Jacobson, and our book, and the footnotes.
If you find an actual error, please get back to us. If you can’t do that, or won’t, then you’re just pulling rank.
It doesn’t take a PhD to wade through the Roadmap. A high school kid with a calculator can do what we did, and if they actually take the time to read Jacobson’s stuff. It ain’t rocket science.
Well, I am pulling my estimates from published degradation rates – of real PV panels in the real world and how fast they go downhill. Degradation is asymptotic, much like how a Tesla battery degrades – more at the beginning, then less and less so.
So, again, I don’t know what your forty year figure represents. If it truly is expected usable lifespan, it’s wrong.
And I find your statements about which is going to be more expensive – RE or nuclear – pretty unbelievable. I think you are cherry-picking your numbers about nuclear costs at the very low end.
The proposed Hinkley Point reactor in Britain is chalked up to be $35 billion. They are calling it the most expensive object on Earth. That is about, what? 900% more expensive than the figure you use?
And another reason I am skeptical about your article. You claim nuclear will be half the price of RE. If so, how can it be that nuclear proposals are being withdrawn left and right because nuclear energy contracts from these proposed sites are running multiply more expensive than RE? How is that possible if nuclear is 1/2 the cost or RE?
If you think you have done such a good job on this article of yours, why not publish it under peer-review? Have you sent a copy to Jacobson, asked for an analysis or rebuttal?
“If so, how can it be that nuclear proposals are being withdrawn left and right because nuclear energy contracts from these proposed sites are running multiply more expensive than RE?”
Incorrect, new nuclear projects are being developed around the world. Nuclear is not multiple times more expensive than grid scale RE. If you are referring to the LCOE cost yardstick, that is a poor comparative measure of generation costs. LCOE figures for RE typically ignore system issues like the cost of dealing with RE’s intermittency problem and the need for huge transmission expansion and interconnector projects to move RE output to the load centres.
You also mentioned Lazard’s work. I think you will find the locations where RE is particularly low cost is the sun blessed US southern states that enjoy year round solar. Whereas the northern hemisphere has long winter dark periods.
On RE being “superabundant”, the problem of gluts followed by famine due to RE intermittency is why the world cannot be powered 100% by RE.
There are some interesting remarks in this post of Roger Lambert that I whish to address.
””The optimistic lifespan of a PV panel is 40 years. ”
Wow – Sorry, no. Current degradation rates of modern PV panels indicates that they will have 100 – 125 year lifespans. And the definition of lifespan is very conservative – it is when the panel can only out put 70% of its initial rating.”
Is there any accounting for this in the thesis of Jacobson? I believe he assumes an economical lifespan of 25 or 30 years (can’t recall exactly), after which they get replaced. Which is normal practice.
“There are panels made in the 1970’s which are still putting out 90% + of their initial rated output.”
That’s all fine and well, but has there been any accounting for any degradation in Jacobson’s thesis? I believe not. In fact, he assumes full replacement after 25 years.
“If this is the sort of expertise you bring to the table, I gotta ask – why should we believe what two amateurs with demonstrated bias for nuclear have to say, versus Jacobson and Delucchi, who are PhD professors in the field at Stanford, and who have been publishing their work in under peer-review?”
First element : Can you confidently state that jacobson has no bias against nuclear energy? From what I’ve read and heard from him so far, he has a very clear bias against nuclear energy. So bias cannot be a disqualifying factor here. It’s either that, or you’re intellectually dishonest.
Second element : If you assert that his opinion is more valid because he has a PhD, that’s an argument from authority and a logical fallacy.
On Topic: Jacobson has presented his thesis and the substantiation thereof. Anyone can consider his evidence, and see that there’s errors in there. No PhD is needed. Conley and Maloney have shown simple counter arguments that refute Jacobson’s thesis.
Aside from that, if you consider Jacobson’s thesis closely, you will find that he has omitted 3 trillion USD in Hydro Expansion. And that’s just accounting for the civil works, penstocks, generator buildings, generators, and switchyard extensions to the extant dam infrastructure. He modelled for 1300 GW of hydro, yet made no mention of it in his entire paper.
And that’s not taking into consideration whether it is even possible to achieve a collective discharge sufficient to generate a delta of 1300 GWh / h. Nor what it’s effects would be downstream.
I think these omissions alone are egregious enough to force a retraction.
As far as I could read the hydro expansion is included, it is “just” the addition of more turbines to existing dams, and the retrofit below for higher flows where neccesary (usually that’s not neccesary). Some people thought that expandind capacity of hydropower would need to add plenty new dams, and complaint that the new dams are not included. But Jacomson did not plan with more storage capacity, just with more turbine capacity at exisiting dams. Beside this he did not use other available options for the same tasks, maybe to keep the model simple.
But one can surely complain about the wording used by jacobeson sometimes when reating on critics – but the other side also uses a lot ob not so nice words too, which makes his reaction sometimes understandable.
I’m intimately familiar with his thesis and never does he write that we could add turbines.
Additionally, I’ve done the math. The 3 Billion USD is penstocks, turbine buildings and turbines only. Building the dam itself is less than 1/4th of the cost.
Additionally, addin more turbines to existing dams, sufficient enough to reach 1300 GW is impossible. You’d need to add 260~285 turbines to the Hoover Dam. Which is 15x as much as it has today.
Additionally, you cannot stack turbines, since after the initial drop you’ve extracted most of the energy from the water. And subsequent turbines would require equal height differences to make the same amount of energy.
In existing projects to add turbines I have seen about 500€/kW costs for this action, which would result in much lower costs than you have. And usually they are added in parallel to existing turbines.
Adding 260 Turbines to a 20 Turbine facility seems feasible to you?
And doing this to 80% of the dams present in the US?
@Mathijs Beckers – technically it will most likely be doable. There are enough dams in the world where expansion has happened, and the number will be much smaller today with >1GW turbines available. But I do not guess this will be the way to go, because grids stretch over the US border and also biomass residuals are available, while jacomseon excluded significant trade of power over the border and also biomass for residual load generation in his model, to show that it even works without them. In practice both will be used naturally. In reality turbine capacity will be expanded if and where it is economocal. When market rprovides a price premium for delivering residual loads due to high renewable contributions to the grid, additional turbines to some degree might be built if they are the most cost efficient way to deliver it.
“Is there any accounting for this in the thesis of Jacobson? I believe he assumes an economical lifespan of 25 or 30 years (can’t recall exactly), after which they get replaced.”
Thank you for making my case for me. This would make Jacobson’s numbers look even better.
“So bias cannot be a disqualifying factor here.”
My point was that self-avowed amateurs were making an amateur mistake. Jacobson, OTOH, is a pro who published under peer-review. I asked if the authors of this article had approached Jacobson for comment, or intended to submit their findings for peer-review, which would seem to be the proper venue, yeas?
Meanwhile, back at the farm, I called into question the unbelievable assertion that nuclear would be half the cost of RE. Because the pros in the world do not believe so. The nuclear industry appears to be withering on the vine because it is too expensive. Lazard LTD found that new RE, even with storage, has a cost less than the mere operating cost of nuclear (!) Only these two self-avowed amateurs appear to be making this claim.
” If you assert that his opinion is more valid because he has a PhD, that’s an argument from authority and a logical fallacy.”
I do not think you understand the fallacy of argument from authority. Jacobson’s studies have survived multiple peer-review. This article, by amateurs and without peer-review appears to be coming to conclusions which appear at first glance, to be contradictory to the understanding of the field, to put it in a nice way.
So, no, I do not think I will be making any “retraction”.
Mr Lambert, you are making an argument from authority and keep doing it. Can you please stop this? You say Jakobson published under peer review, but you omit to mention that his study has been severely attacked by a number of his colleagues – in the same scientific journal. The fact that something is published after peer review does not mean it is true. Far from it. The fact that something is not published under peer review does not prove it is false. Countless articles are published every day in journals that have not been peer reviewed. That in no way disqualifies them. So please confine yourself to real arguments. Thank you.
Thank you Karel. To your point , an amateur astronomer just last week found 5 new planets.
You assume to much Robert.
We stand on the shoulders of giants. So yes, I know the importance of peer-review.
Your attribution of professionality is an argument from authority. Calling the authors of this rebuttal “amateurs” is yet another inverse argument from authority. Who are you to say who can and cannot review an academic paper? What if the initial review missed critical flaws? And this happens. A lot of peer-reviewed papers get retracted because of it.
“So, no, I do not think I will be making any “retraction”.”
You’re basing this argument on the fact that we’re debating the “Roadmap to Nowhere”. However, you forget 2017 Clack et.al and Heard et.al. Additionally, you ignore the remarks of other academics like Hansen, Caldeira, Wigley, Emanuel, Maloney (a retired college professor…), and plenty more.
If you value peer-review, and expertise, you should acknowledge that the roadmaps that are being presented by Jacobson are failing scrutiny.
Having studied this entire matter closely for over seven years, I feel confident to state that Jacobson’s roadmap is a subsistence perpetuating proposition, which will not solve the problems of emerging economies and growing countries. And I think that it is highly immoral to accept it as gospel.
” the roadmaps that are being presented by Jacobson are failing scrutiny.”
Hardly the case.
Yes, we have approached Dr. Jacobson for comment.
Refer to Ch. 1, End Note #10. And to Ch. 12, End Note #2.
Far from the nuclear industry withering on the vine, China’s current construction activity will increase their nuclear capacity to 58 GWe by 2021. Their intent is to reach 150 GWe by 2030, and “much more” by 2050. For comparison, 150 GWe is approximately equal to US and France combined.
http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
https://us.sunpower.com/sites/sunpower/files/media-library/data-sheets/ds-e20-series-327-residential-solar-panels.pdf
Reference 4 on page 2.
That’s End Note #1 in Chapter 11.
Yes, and that’s just the time with warranty, which you do not get without significant additional payments for a thermal power plant. Warranty for older nuclear power stations were below 5 years, so should we assume a life expectancy like this for nuclear power? But even with replacement after 25-40years, reycling of the whole equipment into new panels is quite straightforward, and the amout of materials used is the same or lower than for uranium alone of nuclear power, given that not all material moved is ore, the main part is just dirt. (And other materials are needed beside uranimum as well)
#this is not the way to create a pro nuclear argunemt. Try to get the real world LCOE-Costs down to a fraction of what they are today.
Show a real world factory producing low costs SMR and not just hypothetcal concepts.
There’s nothing hypothetical about South Korea completing a generation 3 reactor on time and on money in the UAE. And as we stayed quite clearly and repeatedly in our book roadmap to nowhere, are argument does not depend on smrs or msrs. It holds entirely valid which Generation 3 reactors as well.
Yes, and the UAE decided to install solar and wind along with some gas power in the future due to lower costs. (Although the costs often stated for the reactors are the financing volume, not the costs as far as I can remember the result of a discussion about a year ago here)
Whatever carbon-free energy works in a particular locale is good stuff.
I’ll reiterate the note at the end of the article, which is what we say in the first paragraph of our book:
We don’t have a beef with renewables. We have a beef with the idea that renewables alone can reliably power the entire US, and therefore we don’t, and won’t, need any reactors.
To examine and expose the validity of this sweeping claim, we counterposed it with an equally hypothetical grid that could actually work – an all-nuclaear grid.
We don’t have a beef with renewables, Helmut. I think it’s great that the UAE, like China, is investing in nuclear AND renewables, where the US and most of western Europe is not.
The rest of the world, however, is pursuing nuclear, an invaluable carbon-free technology, especially for urban areas. More than half of humanity lives in cities. It’ll be 2/3rds of humanity by mid-century.
It would be an incredibly foolish thing, at this juncture of human history, to abandon ANY carbon-free source of reliable power.
The corollary of the 100% renewables meme is the equally absurd notion that we don’t need nuclear.
That is what we rebut in our book.
If renewables advocates really think the world can do without nuclear power, then they had better come up with some far better numbers and performance specs than what Jacobson has presented and proposed.
Helmut, your arguments that the world should abandon nuclear because RE is cheaper is a smokescreen. The world knows the true reason Germany has abandoned nuclear is not because nuclear has a bad safety record in Germany, or that it is too expensive, but that some people in Germany mistakenly fear nuclear power. The Green Party is responsible for this propaganda and delaying decarbonisation of Germany. Their political influence has forced Mrs Merkel to agree to prematurely close safe nukes in Germany. Coal capacity could have been closed sooner if nukes had not been closed prematurely and Germany’s 2020 carbon target likely not missed so spectacularly.
Even if RE costs were still very high and unaffordable at scale, the Greens and their supporters would still be anti-nuclear.
Germany has made a mistake abandoning the benefits of a reliable and consistent low carbon energy source. The UK and the US will not be making that mistake.
And the power cost for the UAE plant is 11c/kWh 3-4 times the power costs from PV at the same location and almost 70% more than solar thermal.
It is by no means certain that those plants will achieve 90%CF, because they like all thermal plants will be degraded in high ambient temperatures.
The reactors will last ate least twice as long, operate 24 / 7 regardless of weather (including sandstorms), and have 18 months of storage built right in.
Are you quoting peak or average? Have you factored in intermittency and storage? Degradation of performance for sand and dust? Have you factored in the lifespan of the solar equipment vs the reactors?
The 4 UAE reactors use sea water for cooling purposes drawn from the Gulf. The temperature is some 15C higher than the Korea reference plant design conditions. The higher cooling water temperature will reduce the APR1400 reactor output from about 1450MWe to 1360MWe.
The UAE reactor rating for CF purposes will hinge on the reactor rating for the Bakarah site conditions, not the rating in Korea. Hence higher ambient temperatures will not impact achieved CFs.
ASFAIK the power prices for Bakarah are not in the public domain so 11c/kWh is speculative. However, the cost of power from reactors 3&4 is likely to be less than 1&2.
Diablo Canyon sells electricity to the grid for 5 cents/kWh. Palo Verde: 4.3 cents/kWh. Columbia Nuclear Generating Station: 3 cents/kWh. Fully amortized plants: 2 cents/kWh — 1.5 cents for operations, 0.5 cents for fuel.
Sorry, but it’s opposite: An 100% renewable grid will be far (>2x) cheaper. In your ‘book’, more a pamphlet, you assume so many faulty and unrealistic numbers that your outcome has no value. A few:
– The real costs (LCOE) are important for a comparison.
But you ignore that because that would make the picture for nuclear less favorable because nuclear needs far more staff, maintenance, etc. So you consider investments only.
However even then you have to neglect interest and other (waste, risks, decommissioning) because otherwise nuclear becomes more expensive.
– “Germany … : Replacing their reactors with wind and solar has actually increased their CO2 emissions”
The real figures show opposite!
Year ; nuclear(% of production) ; emissions (from UBA)
2000 ; 171TWh(29%) ; 644 gCO²eq/KWh
2005 ; 163TWh(26%) ; 611 gCO²eq/KWh
2010 ; 141TWh(22%) ; 559 gCO²eq/KWh
2015 ; 92 TWh (14%) ; 528 gCO²eq/KWh
2017 ; 76 TWh (12%) ; 489 gCO²eq/KWh (=24% less)
– We discussed already the cheap PtG storage solution, so you can delete the battery costs and reduce other such as (pumped) hydro, etc. Especially since PtG is anyway needed because the chemical & steel industry need a.o. methane & hydrogen gas.
So your exceptional assumption:”One day of storage … cost more than an entire nationwide all-nuclear grid.” is really nonsense.
– “A global Roadmap’s solar panels would monopolize 90% of the world’s proven silver reserves …. (etc)”
Nonsense. Solar panels don’t need to use silver, etc
– “… the only possible chance to make the Roadmap work is to build all of it, or nearly all of it.”
Nonsense all kinds of mixes are possible and upcoming in the world. Check Germany; now 40% of its public electricity is generated by renewable and they still have nuclear.
– ” we are nowhere near making long-term predictions about the weather. ”
You make an issue of nothing. There is really no need for long term predictions. Renewable gas turbines need only a few hours to reach 100% of their capacity. With fuel cells it’s a matter of minutes…
– “What if our wind-blown Northern Tier becomes the Northern Doldrums … Monsoon State?”
Such changes take decades so enough time to adapt. Then the population will also move to better locations.
Apparently you try to create needless fear.
– “the world needs all the clean, carbon-free energy it can get.”
Nonsense. The world has more than enough carbon-free energy to harness, but the world is short of money!
So the world should invest in the most cost effective technologies. Nuclear is about the least cost effective technology, so investment in nuclear will delay the battle against climate change.
etc. etc.
We didn’t “discuss” P2G. It was mentioned and commented upon, but don’t get the idea that I think it’s a total storage solution.
Silver is used in existing panels. We based our calculations on existing technology. Until silver-free panels are tested, mass-marketed, and in widespread use, they’re as speculative as molten salt reactors.
The concern of unpredictable weather, which we explain in the book, is that it is folly to invest billions in weather-dependent systems, when the favorable weather they require is becoming less and less unpredictable.
Major changes in weather patterns may indeed take decades, but once you commit to building trillions of dollars of utility-scale RE farms, you better hope that unforeseen changes don’t come for several decades, or more.
If you disagree, then how long do you suggest we ride the downhill run of diminishing average capacities, before we abandon the failing wind or solar farm, and build in a new area? Have you calculated that into your LCOE for RE?
Staking our hopes on a nation-wide, fuel-free, weather-dependent grid is beyond folly. It’s insanity, especially for an advanced industrial nation. Most particularly in a world of increasing climate disruption, with ever less predictable long-term weather patterns.
Silver is used because it conducts electricity better, but no need. As long as silver stays relative cheap nobody will use other material.
Changing weather patterns take decades and the construction (or replacement) of wind & solar can gradually adapt. Note that Germany is now also installing wind turbines in areas with low wind against little extra costs, so they need less long power line capacity.
Biggest advanced European economy, Germany, is following the path towards 100% renewable. It now gets gradually more and more followers. Not only NL, etc. Even France is now gradually leaving nuclear behind and aggressively installing more wind and solar.
While the EU countries are far less suited for 100% renewable than USA (less dense populated, better wind and sum).
Note that in Germany population support for the Energiewende increased from 55% (~2003) towards >90% (2017). Population support for nuclear is everywhere <60%, mosty <50%…
Even in our less windy & sunny environment, the costs of new wind & sun are gradually coming down to €15/MWh. Levels lower than any fossil or nuclear generation method… It allows for over provisioning, the use of PtG, etc.
Have you factored changing weather, the consequent loss of production, abandoning a farm, or moving it, or building a new one in a more favorable locale, into wind and solar’s LCOE?
You are advocating a weather dependent, fuel-free method of energy gathering, with all gathering equipment exposed to the elements, and the whims of Mother Nature.
1. At this point in the RE buildout it is not possible to estimate their levelized costs. One reason is that no-one has yet attempted large-scale recycling of solar panels. If that is accomplished chemically using nitric or other acid, our ability to capture and sequester the resulting NOX gases at scale is unknown. NOX has GWP greater than 10.
Likewise for windmill blades made of fiberglass reinforced- or carbon reinforced-plastics.
Also, LCOE as commonly specified does not account for quality of electric energy – its dispatchability.
2. PtG is applicable to fuel cell transport and to high-temp chemical and steel processes, but it is not applicable to energy storage for re-injection into the electric grid at a later time. Because it has very poor turnaround efficiency of about 20%, unlike batteries and pumped hydro which are in the 90% range. [electrolyzer eff about 60% X H2 combustion turbine eff about 35% = about 20% ; fuel cell eff not much better than combustion eff]
3. PV solar now uses 31 mg silver per watt dc, anticipated to decline to 13 mg /W by 2026. [Intl. Tech. Roadmap for PV, http://www.itrpv.net ; FN 7 of our Ch 5] CSP now uses 13 mg silver per Wac. [FN 8, Ch 5]
Total silver demand for the US 100% RE Roadmap is about 51,000 tonnes [FN 10] , based on 2050 projections of 3390 GW PV and 550 GW CSP peak capacities. Global solar capacity (PV + CSP) is projected to be about 10 X US capacity, so global demand about 510,000 t, which is 89% of the world’s 570,000 t of proven silver reserves. [minerals.usgs.gov/minerals/]
1. “no-one has yet attempted large-scale recycling of solar panels”
In dense populated NL, recycling is obliged. Costs for larger amounts of solar panels (e.g. ~3,000 panels) are ~€2.50/panel.
So those costs become significant when PV-panel prices are at ~€25/panel.
Then LCOE will be <€15/MWh in NL, despite its poor insolation due to its high latitude.
2. In Germany there are many PtG installations that inject their production into NG pipelines.
The first, 2MW ‘Windgas Falkenhagen’ started in 2012.
That gas is a.o. used to produce electricity.
Your efficiency numbers are from previous century. Up-to-date PtG(H²) is now at 85%, H² fuel cells at 80%. So developments ongoing for fuel cells to power factories, cities, etc.
Note that Dutch scientists concluded that our NG grid can transport 100% H² with only small adaptations.
Of course the burners in the boilers etc. have to be replaced, but that is not a big issue (Belgium does it now converting from Dutch gas to high caloric Russian gas).
3. Silver is not a necessary component of PV-panels (as I stated above).
Why do you say that nuclear is the least cost effective energy technology?
Four Generation 3+ reactors came online in China in 2018, two at Sanmen and two at Haiyang, all of them versions of the Westinghouse model AP-1000.
Sanmen unit 1 is the FOIK in the world. It came in over budget at $7.3 billion. https://www.powermag.com/ap1000-reactor-set-for-commercial-operation-in-china/
We don’t have a firm cost statement on the other three but informed sources suggest Sanmen-2 was about two-thirds of Sanmen-1. Call it $5 billion, for a combined cost of $12.3 B for 2220 megawatts of electric capacity. That’s a unit-capacity cost of $5500 /kW.
Doing the arithmetic, with capable capacity factor of 92% and 60-year lifetime, gives 1.1 cents per kilowatt-hour, over the reactors’ lifetimes. [2220e6 W X 92% X 8760 h /yr X 60 yr = 1.07e15 Wh.
$12.3 billion / 1.07e12 kWh = 1.1 cents /kWh.
China National Nuclear Corporation expects their learning curve to bring construction cost down from $5500 to $3000 /kW, which would turn that 1.1 cents into 0.6 cents /kWh. See http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
and https://www.sciencedirect.com/science/article/pii/S1674927817301181?via%3Dihub
That’s why their National Development and Reform Commission is pricing new wholesale nuclear at 7 cents /kWh, at least 10% cheaper than wind’s 7.8 to 9.8 cents /kWh, depending on location. And at least 50% cheaper than solar in the western desert at 14.4 cents, or much cheaper than the eastern population centers at 20.8 cents /kWh. Check out the World Nuclear Assoc article.
Since the liberalization of the China’s electricity markets 2yrs ago, no construction starts of any regular nuclear power plant…
Your solar prices are way off. Demonstrated by China’s fast solar expansion in 2017.
China added 53GW of solar in 2017.
Astronomical levels of waste continue such as excess lighting of streets and buildings at night, using treated potable water for lawns, and designing buildings including houses that require huge amounts of energy to heat or cool the outdoors. Arguments against non-carbon forms of energy production clearly assume that society will never address the outrageous waste we call normal today. There are two ends to this calculus. Fix the waste, it is cheaper than finding energy to fuel the waste.
Amen to that Andrew! All energy production- and consumption- has environmental impacts. Half the problem, at least, is the way we waste energy when we use it. Unless it’s expensive, we won’t bother to spend the money or make any compromises to bother conserving it. The fact that we still allow the atmosphere to be used as if it were a free and limitless public sewer ensures that energy sources which benefit from this “free dump” will continue to be preferred.
I find it difficult to be concerned about energy usage when RE could be in such easy superabundance and at such a low cost as to render billing departments superfluous.
Fossil fuel energy use is bad, yes. But surely we will leave all that behind.
Roger – Superabundance? Low cost? I invite you to read the first chapter of Roadmap to Nowhere. In fact you should read the entire book.
I said it could be in superabundance – it is certainly there for the harvesting, about a million times or so more than we could ever use.
And cheap? Have you looked at wind, wind plus storage, solar, solar plus storage prices lately? We are talking 1.8 cents per kwh! And it is only going to get cheaper.
From Climateprogress [https://thinkprogress.org/colorado-wind-batteries-cheap-12e82b91a543/]:
“…the financial firm Lazard Ltd., which found that in many regions of North America, “the full-lifecycle costs of building and operating renewables-based projects have dropped below the operating costs alone of conventional generation technologies such as coal or nuclear.””
So, again, how is it that you two fellows are saying that NEW nuclear plants are going to be half the cost of RE, when evidently they are now under the -> operating <-cost of existing nuclear?
Our arguments and the costs we used to arrive at our conclusions are detailed quite clearly and in plain language in our book. I invite you to actually read the entire book since you seem to have so many questions regarding our fax our observations and our methodology.
You are challenging us because we are amateurs. I am challenging you to be a rational open-minded human being and actually read what we have to say instead of peppering us with loaded questions and disparaging remarks.
If you find a major factual error in our book,please get back to us.
You want me to read your entire book and get back to you? I do not need to – your conclusions about relative cost are contrary to consensus opinion.
Your work is not peer-reviewed yet contradicts and disparages peer-reviewed work.
Your claims about the costs, bidding success, and system availability of nuclear plants are outside of consensus opinions.
https://www.ecowatch.com/westinghouse-nuclear-2469123144.html
Union of Concerned Scientists:
https://www.ucsusa.org/nuclear-power/cost-nuclear-power#.WmTBtK6nFSE
http://progressive.org/dispatches/goodbye-nuclear-power-construction-of-two-of-four-remaining-/
Joe Romm: Building new renewables is now cheaper than just running old coal and nuclear plants:
https://thinkprogress.org/solar-wind-keep-getting-cheaper-33c38350fb95/
UK government’s own projections expect onshore wind power and large-scale solar to cost less per megawatt hour than new nuclear by 2025:
https://www.theguardian.com/environment/2016/aug/11/solar-and-wind-cheaper-than-new-nuclear-by-the-time-hinkley-is-built
List of cancelled nuclear reactors in the United States:
https://en.wikipedia.org/wiki/List_of_cancelled_nuclear_reactors_in_the_United_States
None of the articles you cite are peer reviewed so what are you complaining about? If you don’t want to confront the authors’ arguments, stop engaging in the discussion, or stop complaining about peer reviewing. Thank you.
Roger –
The costs and numbers we work with are directly from Sunpower, the panel manufacturer, and NREL, the national renewables energy lab. The numbers we cite from Jacobson’s paper are mainly on his Figure Two spreadsheet.
Our plain-language book explains the paper’s numerous shortcomings, including numbers from Jacobson that only make sense if a 100% packing factor is used to calculate solar acreage. (I think it’s an innocent typo. My co-author isn’t so forgiving.)
Every critical numerical assertion we make is backed up by footnotes to the source material, and many footnotes also contain our actual math, including explanations of how we worked with the numbers to arrive at the conclusions we did.
So yes, as a matter of fact, we do want you to read the whole book, including the footnotes, because quite frankly, you are wasting everybody’s time by trolling us with questions that are clearly and unambiguously answered in the book and its numerous footnotes.
1 GW of reserve is enough for a grid of the UAE’s size because the biggest single unit on the grid is less than 700 MW. With 1,400 MW units it will need at least 1,600 MW of spinning reserve. Even with that much if a grid connection to the plant goes down or two plants go off in close proximity the grid will crash
Assuming 85% CF this plant will generate about 40 TWh at an annual cost of US 11c/kWh including interest and depreciation. This does not include the gas spinning reserves
Based on recent Middle East prices 6 GW of Solar PV and 5 GW of solar thermal will generate the same amount of energy at a weighted average price of around 4.5 c. Even on the worst day it will provide more peak capacity than the best day for the nuclear plant. While it will need more secondary reserve it won’t need 1,600 MW spinning reserve and it won’t need 1,400 MW for 6 weeks during refuelling
UAE’s grid is being interconnected with Saudia Arabia’s Grid so the UAE will not need to hold 1400MW of spinning reserve to support their reactors as reserve needs will be shared with those of SA.
The LCOE of new nuclear v solar PV is not directly compatible as solar is intermittent which is an issue as the UAE has early evening demand peaks.
The reactors will be scheduled for refuelling away from peak demand periods so their capacity should not be missed by the grid.
That’s a good idea for us rich people in the first world. The problem is that there are about 7 times as many poor people as there are of us.
So your idea which will actually cost at least twice you projections, includes all sorts of assumptions about unsolved problems in material science, grid reliability, transmission efficiency and availability, water usage is a good idea.
A town of 110,000 people in Victoria Australia can go 90% renewable electricity with 25 wind turbines and 20% of their roofs covered with solar. They can do it bit by bit, the returns from the first installations paying for the second and so on and they have started already. By the time Thorcon’s first plant has demonstrated its reliability, the opportunity is passed. Whether the last 10% comes from solar thermal, batteries, pumped hydro in old mine sites, geothermal etc can be decided well down the track.
If it can happen in Southern Australia it can happen in all of Africa and most of the America’s
Plenty of fairy tales gain. The coal use in germany is not rising in the power sector it is falling. In 2017 id did fall almost 20 TWh brutto. And it will fall further in 2018. This year renewables and all kinds of coal did contribute almost the same to the public grid with a tiny advantage for coal – but it would not be a surprise if from next year on king renewable starts to reign in the german grid.
And not to forget Germany’s power export is increasing.
Seriously? Germany has opened many new coal power plants in 2017 some of whuch use Lignite! They have yet again wildly missed their own self imposed Co2 emissions targets and now have the most expensive electrical prices in the world!
Yeah – great rwsult!
Nut a single new coal plant was opened in germany in 2017, but several closed. Bt just tell which ones opened in germany according your opinion?
The many years delayed plant in Datteln might open in 2018, they discussed a long time after winning a long lawsuite in 2016 if they shold knock down the brand new plant before opening, or invest the last millions and open it altogh it will never earn its investment costs due to the very low wholesale prices in germany.
Consumer prices are high in germany, but electricity prices for large consumers, especially for very large consumer s are very low. Costs for aluminum smelters in germany are below 5ct/kWh in germany.
Emissions did fall again in the power sector, rising emissions in the traffic and heating ectors compensted this, so gouvernment will have to concentrate on these two sectors.
At the same time older less efficient coal and lignite power plants have been closed down (Frimmersdorf , Voere, Ensdorf). Overall there is a slight increase capacity, but because the new plants are more efficient, overall coal and lignite use has decreased. Nevertheless, the new lignite and coal power plants are not really needed. Renewable power production has more than compensated the closed down nuclear power plants. And Germany has become a power exporter, even on dark windless days like the 24th of January in 2017.
That new coal power plants were taken into service is thus not due to renewables not functioning, but to the complexities of politics.
Power plants take a long time to plan and build, so there is a natural tenedency in politics to protect the investors. Especially, since the government is the investor: The shares of lignite mining companies and coal and lignite power plants are in the hands of different levels of government like municipalities and states.
Furthermore, there is an historical intertwinement between the socialist party and the coal and lignite labour unions, and conservatives and liberals (in the European meaning) do not care so much about the environment and have an intuitive resistance against change and against things that originated from the progressive movement.
So overall the problem is politics, not renewables.
Early lignite phase-out could not succeed without first clarifying the means of defraying post-mining reclamation costs. No business case exists for alleviating the mining companies of this obligation. They in turn will require several more decades to accumulate the necessary capital to implement the comprehensive tasks involved.
In eastern Germany, the governments of Saxony and Saxony-Anhalt have now declared the continuation of lignite usage until mid-century. In Brandenburg, an application was recently submitted to extend the boundaries of the Welzow surface mine, thereby apparently sealing the fate of the village of Proschim (https://energypost.eu/divestment-will-not-block-german-lignite/).
Building more wind turbines and solar panels could not alleviate this stalemate, since renewable energy investors cannot be obligated to resolve a problem they did not create. The more competition that is created for the lignite industry, furthermore, the less able it will be to cover post-mining costs without public subsidies.
The costs to recultivate the open cast mines in gerany have always been colleced during operation and are already included in the allowence to start mining. So the capital is already accumulated. German law takes care about such points.
The expansion of Welzow-Süd II is canceled so far: https://www.rbb24.de/wirtschaft/thema/braunkohle/beitraege/brandenburg-lausitz-braunkohle-tagebau-jaenschwalde-wird-nicht-erweitert.html It’s not economic any more because the coal will not be needed as it looks like.
The applicable regulations and conventions do not seem to apply fully to this case.
In Capital magazine of December 13, 2017 (https://www.capital.de/wirtschaft-politik/braunkohle-eigentuemer-warnt-vor-kohle-ausstieg), it has been noted that EPH was not legally bound to cover LEAG debt. Therefore, post-mining obligations might not legally be enforceable if LEAG were to declare insolvency.
Furthermore, according to the state mining authority of Brandenburg (http://www.lbgr.brandenburg.de/sixcms/detail.php/845231), LEAG applied on November 28, 2017, to extend its mining activities to Proschim, Haidemühl, Jessen, and Welzow. If this application were refused, then LEAG might not be able to realize the turnover necessary for meeting these obligations.
EPA is not owner and operator of the mines , that is LEAG. Which would have to declare bankupt when money they have is so low that it only covers post mining obligations, but nothing else. The article tells that it is additionally under discussion that EPA will declare that it is also bound to these obligations.
The second refence declars that welzow Süd I, under operation since 1994 is allowed to continue operation on the same area as before till 2023. It is not about Welzow-Süd II.
German is not your native language?
You confirm my main point: coal/lignite mines and power plants are kept open because of politics, not because of renewables .
Or to formulate it more clearly:
Coal and lignite mining and burning continues not because it is needed, but because of politics.
I fell for Steve’s bluff, I could not find any info on new coal or lignite power plants coming online in 2017. to the contrary, netto 2.4 GW of coal and lignite was taken offline in 2017.
Sounds like Germany’s new coalition government will not be forcing the closure of more coal fired capacity soon.
http://www.dw.com/en/opinion-german-coalition-hopefuls-drop-climate-goals/a-42084682
A shambles of a political system having taken since September to form a workable Government. Horse trading on policies amongst political parties has watered down climate related policies too. That is not what German voters were expecting! Glad the UK doesn’t have a PR system that only delivers delay and compromise…
well the corridors for wind and solar deployment should be increased, a additional 4 GW onshore, 4GW solar and 1 GW offshore shall be tendered to be built in 2018 and 2019, the phase out of coal and lignite should get a shedule for each power plant and a end date for coal use, the share of renewables in the grid shall be rised to 65% in 2030, and the use of electric poer in th heating and traffic sector shall be increases, grid expansions accelerated. These are the points you missed as it seems.
Have a look at these statistics about the German power system: http://www.cigre.org/var/cigre/storage/original/application/85930102220abc8342ca5fe76c20adf2.pdf
Name them. And also name the closed ones.
According to the Energy-charts.de fell installed coal power in Germany by 2.5 GW between 2016 and 2017.
I agree that 100% renewables is at least for most countries not yet as simple and cheery a solution as suggested by Jacobson’s graphics. But getting to around 80% is not so hard; the problems come from the generation and storage required to fill in the last 10 or 20%, such as the massive hydropower mentioned or the electrolytic H2 used in 35% efficient P2P cycles. Using the problems of 100% renewable solutions to justify switching to 100% nuclear is not constructive. I would be more sympathetic to the authors if they tried to work out the role of nuclear in a mixed system. I don’t know what the world will look like in 2050 and I favor some continuing development work on improved reactors to keep our options open.
S. Herb – Our book is a rebuttal against the idea of a 100% renewables grid. To put such a grid’s shortcomings in sharp perspective, we counterposed it with an equally hypothetical all-nuclear grid. In the real world, we will of course have a mixed grid.
The World Nuclear Association is pushing for 25% nuclear worldwide by 2050 under its Harmony program so there is plenty of scope for renewable energy sources as we decarbonize. Take a look at http://world-nuclear.org/harmony
“…a much more important defining issue for 2017 is the very real start of a movement that recognises that powering the world with 100% renewables is a myth – and that chasing a myth will not get us to our global goal of meeting the world’s increasing energy needs while reducing carbon emissions and successfully combating climate change.”
http://mzconsultinginc.com/?p=944
Precisely 🙂
The article could have benefitted from far less hyperbole and some more data. Addressing one issue: discussions on storage tend to be one dimensional (I wish I had a $ for each time I have pointed this out). Storage can take many forms – e.g. google “Ice bear” – which could provide a useful (partial) solution wrt to A/C demand – when combined with roof-top PV. In the case of PV – the cells can be re-cycled in a variety of ways – thus further reducing costs. In the case of wind – re-powering can mean replacing the turbine but leaving the tower (steel) and the base (concrete – & for those that want to see a long lasting concrete structure – the Pantheon in Rome is one example). Arguments about 100% or not are for the moment moot, absent clear views as to where tech developments go. I’m sure Thorium nukes are wonderful at producing electricity and thus invite the authors to put their money & talent where their mouths are & invest in Thorium nukes. I hope they do well, time will tell..
It’s an intro / PR article. Read the book and its 150+ footnotes for more data. And, we cover overhauling turbines, as distinct from replacing the towers.
The supposed feasibility of 100% renewables was posited by Mark Jacobson. We counterposed it with an equally hypothetical 100% nuclear grid.
Lastly, why would you suppose that we don’t already invest our money and talent where our mouths are? Was there a point to your snark, other than snark for snark’s sake?
I’d don’t like PR hence my comment. No “snark” (whatever that is – did you mean sarcasm? in which case – why not say so?) was intended – you seem to be overly sensitive – I wonder why? I have no doubt thorium reactors can produce dispatchable electricity. If they can do so at a price able to compete with other systems – wonderful. If you are confident in this then as I said – put your money/talents where your mouths/words are. & for the record, I invest in renewables because they are one of the paths towards a de-carbonised electricity supply. They are not the only path but have some cost advantages over other systems.
Snark is an American slang meaning sarcasm. And you seem to be passive aggressive. I wonder why?
The cost advantages, or lack thereof, of renewables and nuclear are explored in the book.
The 10 year working life for offshore windpark seems quite a pessimistic assumption. The Vindeby windpark worked for 25 years.
Vindeby is in shallow sheltered Danish waters. Proper offshore North Sea located windfarms are more exposed to harsh weather so might not last as long as the small Vindeby machines.
OK, some other at the North and Irish Sea, involving larger wind turbines:
Blyth windpark, North Sea, operatonal since 2000
Horns Rev I, North Sea, operational since 2002
Arklow Bank, Irish Sea, operational since 2003
North Hoyle, Irish Sea, operational 2003
Check for more:
https://de.wikipedia.org/wiki/Liste_der_Offshore-Windparks
How much even if Open Sea wind turbines lasted as long as land turbines between overhauls, our thesis would be the same conclusions would be the same and major point will be the same.
The difference between 10 years or 25 years for overhaul may at first glance seem enormous, but in the grand scheme of Jacobson’s plan it is actually a quibbling minor detail. That’s because his plan is so egregiously out of whack that our rebuttal still holds water if you were to double the cost of reactors and double the lifespan of wind turbines and solar panels
The book makes the statement: “Wind and solar gear can last from 10–40 years: about 10 years for offshore wind turbines”.
In Denmark they recently decommissioned an off-shore wind farm – built in the 1990s – it was one of the first. “DONG Energy is preparing to decommission world’s first offshore wind farm, located off Danish island of Lolland” – 11 WTs running for 25 years. Thus a true statement would be “off-shore wind has only been running for +/-25 years & early turbines seem to last this long – it is not known at this stage how long 2nd & 3rd generation off-shore WTs will last although subsidies for these cover 16+ years”.
The facts suggest that the assertion in the book with respect to RES lifetimes is wrong, at least with respect to off-shore. That’s a pity because one then wonders what others elements in the book are incorrect – & thus calls into question other assertions made.
If offshore wind lasts 25 years, that would lower the national daily turbine overhaul rate to 54, down from 80. SMRs swap outs would still be far lower, at 2.3 / day.
Photovoltaic systems can easily be moved. In Germany people moving house, often take theri PV system with them. Moving windturbines is slightly more trouble but is already done one a regular basis: there is a lively market in second hand wind turbines, even across continents. Trough based concentrated solar power is completely modular and movable, for tower based CSP, everything but the tower is also modular and movable. So the fear that one is stuck with renewables at the wrong place if climate patterns would change drastically within a few decades is quite ridiculous.
So if the 100% renewables raod map is left halfway you have still 50% of clean power that can adapt to changing climates. The Danes are almost there.
The fear for green elephants.
Moving them is feasible, but with such low capacity factors a move would significantly eat into their lifetime EROEI.
With the investment required for an all-renewables country, the weather had better be predictable for a lot more more than a few decades.
Our point is, depending on favorable weather to keep the lights on is a quite ridiculous strategy to power a country.
In effect ,as a further comment, I think that both 100% renewable or nuclear are not the solutions.
I think that nuclear is the best way to accompany the reneable.
I agree, although which technology should predominate on the grid would still be open to debate.
Regarding the “impossibility” of producing the required RE-equipment:
Last year over 17 Million new vehicles were sold in the USA alone. That means 47 thousand pieces of highly complex hi-tec machinery with all kinds of individual adaptations are produced per day, just to satisfy US costumers. Yet, this posed no challenge to the car-industrial complex.
Re the UAE nukes.
1.It doesn’t matter if they last twice as long as solar the power cost includes amortisation and depreciation and hopefully an allowance for dismantling at the end, It does not include catastrophe insurance or long term waste storage so the published figure is less than the real cost,
2. They won’t run 24/7 at capacity unless they provide no more than about 30% of minimum load. Beyond that the gas plants required for spinning reserve would be overwhelmed in the case of a trip. The reason French plants run at 72%CF is not that they are any less reliable than US plants, in fact up until a few years ago they had higher availability than US plants. The problem is demand is not constant so they have to be rotated out of service and they still only supply 50% of peak demand. As for backup France has 53 plants, substantial hydro and large interconnectors so the failure of any one plant is relatively easily absorbed in their system. That is not the case in the UAE so their capacity factors and availability will be significantly lower than the US plants
3. No-one anywhere in the world has built a nuclear plant to work in such adverse conditions, particularly biological fouling of the cooling system. How can you blithely claim they will be as reliable as plants in completely different environments
The point we’re making with the UAE reactors is that it shows that Gen-III reactors can be built on-time and on-budget, to international standards, with a standardized design. And if they can perform in the UAE desert, even at less than full capacity, they should have no problem here in the US.
But the UAE shows that even if the Gen III reactors are built on time and on budget, they are not competitive.
As it looks like, UAE and Oman are interested to build substantial power lines over the sea to Iran and Pakistan, which will allow them to get better integrated in the Eurasian grid, which also allows to smooth out variability of renewable generation, as well as variability of demand.
New nukes have to build up a fund to cover decommissioning and waste costs at the end of life so that is not an excluded cost.
The UAE reactors will have been designed to integrate with their grid. The issue of reserve capacity needed to cover the unplanned loss of one reactor will have been covered so that the grid remains stable.
EDF’s reactors regularly meet more than 50% of peak demand – just check ‘grid watch’.
There are many conventional power stations with steam plant sited around the gulf. So the UAE nukes and their conventional island steam plants operating in Gulf conditions is nothing new. Power plant performance guarantees will be part of the contracts too.
One reason why the UAE has chosen nukes is because their output displaces burning indigenous fossil fuels that can be exported, or reduce the rate of depletion of their oil and gas resources. Renewables displace fossil fuels too, but their intermittency problem means not to the same extent as nuclear. Also, if demand is expected to grow in the UAE such that new generation capacity is needed to meet peak demand, any new renewables capacity will require near 100% backup. So for every 1GW of solar, 1GW of gas fired back-up is needed. Likely why on economic grounds renewables are not the no brainier option you might believe them to be in the Gulf area. Note that the Saudis too are now embarking on developing nuclear power with plans for up to 16 reactors.
I am not saying that the middle east plants won’t work, I am sure they will but the CF will be more like French plants than US ones.
Re French peak supply I should have been more specific, 50% of French annual maximum demand. And they manage the 60-70% of daily peak by exporting off-peak to Italy, Switzerland the Netherlands, the UK and even Germany.
Re the Saudi’s while they have plans and are in negotiations they have not yet placed final orders for more than two.
You confuse intermittency with backup needs. Peak demand in the middle east is when the sun is shining, if it is cloudy or dark demand falls so 1:1 backup is not necessarily, with the current power structure it is about 70-75%.
The emirates are also encouraging electric vehicles. On average an electric vehicle needs to be charged on a domestic charge point for 6 hours per week. That can easily be arranged with low power chargers in carparks or at residences between 8AM and 2PM when solar exceeds demand further reducing backup needs.
As for nuclear 5 of 5 Swiss nuclear power plants were out of action for months in 2015 40% of Belgian and 21 of 53 French plants for varying periods from the start of October 2016 so the ME plants will also have to have 100% backup.
In some ways it is worse because solar can be backed up by cold gas plants that can be ramped up as the sun goes down. A nuclear plant needs spinning reserves i.e. spare capacity in the system that can replace the entire output of the nuclear plant in less than 30 seconds if the plant trips.
As even gas plants running at a minimum stable level of about 30-35% of maximum take 5-10 minutes to reach full power you actually need about 3,500-to 5,000 MW of gas plants running to support a single 1400 MW generator thus the spinning reserves for a nuclear plant can cost more than the reserves for a solar plant. It is true that the gas plants supporting solar will accumulate more full load hours but the investment costs will be the same or higher for nuclear and depending on the price of gas the total cost of backing up nuclear can still be higher than backing up solar
The availability for a new nuclear plant is >92%. It has to be or the economics would not work. The low availability figures of 42% you are quoting are for old Swiss, Belgian and French plants.
The UAE peak demand is running at 27GW and four grid systems are being integrated. So there is no basis to support your claims that c. 5.6GW of new UAE nuclear capacity will not be able to operate at baseload in the 90% CF range.
Peak demand in the UAE is in the early evening. Solar depresses daytime demand, it does not contribute to meeting peak demand. In the ME sunset is very rapid. Too much solar capacity going off at once in the UAE would be difficult to manage in terms of plant ramp rates, like in California and the CAISO duck curve problem.
I assure you that the four UAE reactors will not require 100% conventional back-up gen. capacity. Those views are put about by antis. The UAE grid will need to have enough fast response reserve to deal with the unplanned loss of just one reactor, i.e. about 1.3GW, not all four simultaneously. The UAE grid is currently supplied by large CCGT plants which need a level of reserve anyway to cover the sudden loss of the exisiting largest generating unit.
C. 1GW of reserve is typical for a grid of the UAE’s size and is similar to the reserve needs of the UK grid during the summer. So you are wrong to assert in your final paragraph that 3500-5000 MW of back-up gas fired capacity is needed to support a UAE reactor. Also, running gas plants to back-up intermittent solar will not reduce UAE carbon emissions to the extent that their nuclear capacity will achieve. Back-up gas fired plants would also consume their indigenous gas reserves which they wish to export.
BTW I am a power engineer with over 30 years of experience.
1. I didn’t say they had 42% availability. What I said was that over the life of the plants that for all sorts of reasons there is a strong possibility all four plants might be out at once for hours, weeks or a remote possibility months. They can still achieve 95% availability and 1% of the time be all off together. i.e once every 10–15 years it is quite possible that they could be out for a month or more
2. You actually made my point for me they will get a high capacity factor because they supply 20% of peak demand. At that point they may work and by definition there will be plenty of backup in the system The question is whether they will still achieve US style availability in a harsh environment. As it took the US 40 years to reach 92% Capacity factors in a more benign place, 80-85% would be a very good result in my humble view.
3. They may be on time and on budget but they are not yet in commercial service so I would not be getting ahead of ourselves
1. With dispatchable plant it’s important to distinguish between capacity factor and availability. Some nukes in France are used to load follow. Consequently their availability to generate will be higher than their actual CFs. That affects fleet availability as does the fact EDF’s reactor fleet is ageing and some >30 year old plants are off for long-term refurbishment to life extend.
With must run renewables availability is not normally quoted, just CFs.
Planned outages for refuelling and maintenance of large reactors is scheduled to occur during periods of low system demand when there is spare gen. capacity. The probability of four reactors being off simultaneously, or three for that matter, and unplanned, is so small as not to be a credible risk so reserve would not be held to cover that eventually. Indeed all four reactors would likely only shut down together if the grid collapsed which would be a UAE wide problem managed by load shedding. Two reactors being off is credible based on one being off for maintenance during a period of low system demand and another tripping.
Note even then enough reserve capacity is only needed to cover the unexpected loss of just one reactor, not two.
2. Gen III reactors use mature technology. The UAE has invested $20bn and employed some of the best engineering advisors and companies around the world to make the project successful. Their grid is being reinforced and plans are in place to interconnect with surrounding Arab countries for security and economic reasons. The project is not a technology test bed, so I think your views are far too pessimistic.
Hmm, the sheduling for low demand periods only works when fuel changes alligns with a 12 Months rythm and if loads characteristic does not change over decades, and if fuel use is not shifted due to unplanned outages. In real world outages die to fuel change also happen during times of highest demand.
Also power plants with several blocks sometimes trip with all blocks due to a common cause, like e.g. the two BOA-Blocks of Neurath due to a error in the control software for all blocks, or in other cases due to failures in the grid connections.
EG in our neighbourhood a storm did blow a 80m metal roof onto a high voltage switchyard, causeing a shortcut in 11 circuits simultaniously, and overruling all redundancy considerations.
…..BTW Australia should be developing nuclear power stations too to deal with closing coal-fired capacity. China or Korea could supply cost effective plants. It’s not too late for the other states to avoid the problems S. Australia has experienced with intermittent gen. capacity.
why should they do such nonsense? It’s extraordinary expensive and slow. new wind solar PV &Thermal capacity is online and running before preplanning of nuclear really started, and delivers at much lower costs.
Nuclear power based on Korean, Russian and Chinese reactors is competitive. Renewables and solar alone can’t replace closing Aus. coal- fired capacity. Other states don’t want to repeat S. Australia’s error of relying too heavily on unreliable generation sources. Building gas fired plants to deal with renewables intermittency will not address Aus. carbon reduction targets.
Wind and solar PV has not delivered “at much lower costs” for Aus:
“A key criticism of nuclear has been price. But the renewables push has already seen Australia’s retail electricity prices soar. The wholesale cost of electricity has trebled in Victoria from $20 a megawatt hour to more than $60. In South Australia it has risen from $50 a MW/h to $110, leaving consumers paying the highest electricity costs in the world.”
https://www.theaustralian.com.au/news/inquirer/nuclear-the-energy-alternative-for-australia-none-dare-name/news-story/34bb25412a576fbb343ae02d4365b0c2
The Australian is a notoriously biased source. Australian blackouts were not caused by renewables, but by the fact that the regulatory system has not been adapted to the increase of renewables, as the Finkel Review makes clear. In addition coal power plants have been tripping all over the place, including just recently. You can check out any number of articles on Reneweconomy.com or on the Conversation or on Energy Post https://energypost.eu/south-australia-makes-fresh-power-play-bid-end-blackouts/ for a more nuanced view.
you forget that Australia now bulds solar thermal plants with storage at competitive prices in Port Augusta as (partial, in first step) replacement of the northern power station providing the dispatchable output you like to have.
Also grid expansions inside australia, and also to indonesia are under discussion- thich can change the game completely in the longer run (still less than the time to build a nuclear power station).
Starting in 2012 and finish-ing four 1400-MW nukes by 2020 doesn’t seem like a very long time. See
http://www.world-nuclear.org/information-library/country-profiles/countries-t-z/united-arab-emirates.aspx
Sec. Barakah power plant, 3rd paragraph
planning startet in 2008, and 12 years is a eternity, and in most places it is much longer, because most countries are not owned by a person which can decide on everything alone. I think it was in Zambia where they decided that it was too little rainfall in winter and spring, and so they need to have new solar capacity online and running in autumn to make up for the shortfall in hydropower. Different scale of time.
California has solar thermal plants already. It seems they have to burn large amounts of gas too to warm them in the morning. Only one has storage and its capacity factor seems to be less than 40%. See:
http://euanmearns.com/concentrated-solar-power-in-the-usa-a-performance-review/
Grid connections to Indonesia some 5000km distant from Aus. major load centres is not the answer either.
@ nigel west – the talk is about australia and the plant there, not some other plant in california operatin in combined gas-solar mode in California. Adding storage to solar thermal plant is becomming more and more usual. Also the plants are having some problems in the heat collecting circuit – accidentally also runningwith molten salt that is proposed in this thread for nuclear power generation, but in case of ST with much lower temperatures, no radiation and less dangers if something breaks.
Nigel. Some Australian coal power plants have long term low cost coal contracts at $27/tonne. This is quite often poor quality coal with 20%+ ash content so their short term marginal cost of operation is in the order of A$25-30/MWh.
Once these contracts expire and or the coal plant does, a new plant with export competing coal prices will have a break-even between A$90 and A$200 per MWh
Wind has recently been contracted at $48/MWh, solar PV at $70 and solar thermal at $78. who exactly is going to build nuclear when the lowest cost i the world is US$110 ($A 146)
Aus. should run a competition and invite proposals from Korean and Chinese reactor vendors who are building competitively. The Chinese company CGN plans to build a reactor in the UK.
The wind and solar prices you quote are only half the story for a renewables only solution. Aus. would need to massively extent its grid along the east cost and attempt to balance the system with storage. That would be very expensive, probably unworkable, and arguably more expensive than nuclear.
Alternatively, build gas fired CCGTs to back up wind/solar and accept the long-term carbon implications.
Coming online faster doesn’t mean much if the technology can’t be scaled up to do the job required.
Well, Longi plans to ramp up capacity to 45GWp/Year till 2020. That’s the capacity (TWh/year, not GWp) of roughly all nuclear manufacturers combined, but only one PV manufacturer of many. Locs it scales up much faster than anything else.
Nigel
I have recently done an analysis of adding one GW of firm capacity to the SA grid. It would need more storage and more backup than wind/solar/gas combination and cost seven times as much. A zero carbon system with wind, solar PV, 300 MW of solar thermal, 600 MW of pumped hydro and 100 MW of batteries would need no more than 40 GWh of storage. To back up a nuclear plant through a singe refuelling cycle needs 2,100 GWh
Peter, nuclear does not need back-up during refuelling providing it is scheduled to periods when system demand is low and the capacity is not needed.
Owners of nuclear plants are incentivised to avoid planned outages during periods of high system demand so they are generating when wholesale power prices are high to maximise revenues.
Nigel,
Nuclear needs a lot of back-up too. E.g.
Belgium generates ~50% of its power with 7 reactors. This autumn there were months in which only 2 could operate (during a few weeks even only one)…
I wouldn’t use Belgium’s reactors as a yardstick for global power reactor technology. Most present-day reactors have around 90% average capacity factor. Advanced reactors are designed to have more like 95% av. capacity.
One needs substantial backup and expensive spinning reserves as reactors can and do stop sometimes in a second. And they all may stop at ~same time as shown in Japan after Fukushima, when unacceptable system vulnerabilities showed.
In 2017, the global av. capacity factor was 80%…
Having said I would not waste my time, I did try to read the polemic and it seems to contain some valid criticisms of the Jacobsen model. However as I said elsewhere, your paper is full of heroic assumptions also.
For example you criticize flow batteries because there is a shortage of Vanadium, there are many flow battery technologies including zinc and bromine of which there is no shortage. Similarly there are vast untapped reserves of Lithium.
Then while your overall criticism of fuel cells has some merit, you say that they waste most of the energy as waste heat. Hydrogen fuel cells are about 60% efficient vs IC engines that are about 40% at peak but a little over 20% through the whole operating envelope. Nuclear plants will be even worse because the peak temperature is limited and the thermal efficiency of any nuclear plant is low so they will reject twice as much waste heat as a fuel cell.
As for water consumption, I agree a legitimate concern but existing coal and nuclear plants consume 40% of the fresh water withdrawals in the United States, far more than the most optimistic projections for water used for hydrogen generation.
Similarly you claim that SMR’s don’t need cooling water and Thorcon claim that they will operate at 700C. No steam plant in the world operates at 700C. Billions have been spent to increase steam temperatures with advanced materials etc and yet the current maximum is 600/620 C and progress beyond that is exponentially expensive. Even if they succeed it will still have to reject 1.2 MJ of heat for every MJ of electricity generated. It is possible to air cool steam plants but a) you can use an average of 5% of the power of the plant just to run the heat exchangers and b) at high ambients the efficiency and therefore the output of the plant is reduced by 8-10% or in some cases more, because the temperature differential across the steam cycle is lower
Then you rightly suggest that excess capacity is necessary for renewables but say it is not necessary with nuclear. The roadmap implicity acknowledges the excess capacity by using realistic capacity factors for wind and solar. While I can agree that more than Jacobsen suggests is needed, tracking solar farms today reach 31-32% CF in the southwest and some modern windfarms are achieving close to 50%, more in a few cases, and still blade lengths are growing and tower heights increasing so in effect the amount of backup required is falling.
Then you completely ignore whole classes of energy storage, the average water heater if grid connected can store 10-20 kWh of useful heat i.e heat above 100 F. a cubic metre of ice over 100 kWh
The criticism of the space requirements of windfarms is just silly. Farming and ranching and even most forms of commercial forestry continue virtually undisturbed and the windfarms use less than 0.3% of the area of the farm and an even smaller proportion of the useful area
Your favourite nuclear technology, the Thorcon reactor does not yet exist let alone been through multiple cycles of debugging and refinement and yet you claim 99% availability. a far more optimistic dream than the claim that batteries and wind turbines can’t be improved.
As I said earlier I have been involved in product development for many years and the life-cycle investment in R&D of successful technologies is tens if not hundreds of thousands of times the work that was required to demonstrate the first working commercial application.
Back of a truck? Thorcon 150-250 ton modules not including the steam turbines. Nuscale 700 tonnes, Transporting radioactive modules that big back to the reprocessing factory, simple easy ???
On balance we might find that nuclear will supply 10-20% of the energy in a zero carbon economy but you do significant damage to your case by spending 140 pages on such a biased analysis
Why mention the efficiency of a nuclear reactor and the associated steam plant? The cost of nuclear fuel is so small, compared to the capital costs of the plant, that efficiency is really not an issue.
Because the authors make a big point of energy from fuel cells being wasted as heat. While in overall cost terms it is less significant for nuclear power it is a significant constraint on their siting.
A significant fraction of the cost of the nuclear plant is related to it thermal efficiency. A mythical heat engine that could run at 1000C would be much more compact than current nuclear designs and need much less material even if some of that material is unobtainium
Yes, where possible large reactors are best sited on the coast or major rivers as they need vast amounts of cooling water. But the smaller SMR type reactors being planned will not be so constrained and could use air cooled condensors.
However, in the UK most civil reactors have been sited remotely so that should an emergency occur the surrounding population can be evacuated. It’s likely that in the early stages new SMR designs would need to be sited remotely too, not close to major conurbations.
The cooling requirements are still there if you have twenty 250 MW reactors or four 1,400 MW reactors, there is the same or more waste heat to be rejected. You can only dump the same amount of heat into a stream or lake whether in the 10 outfalls or one before you stuff up the ecology.
As I said air cooling in hot climates is very innefficient
Air cooling depends on the temperature difference between the machine and the ambient air. With a 700ºC reactor exit port, an MSR would find the hottest desert to be downright frigid.
I forgot to reply to this misinformed post. If the salt is 700 C using some as yet completely unknown metal for the heat exchanger, steam exit temperature from the boiler will be 680 C or less. No-one makes a steam turbine that can handle that temperature but assume again another unobtainium will be found at an economical price, the work done by the steam in the turbine is as a result of the expansion of the steam. As steam expands it cools down. It can only expand until the steam is slightly above the temperature of the cooling system. On a hot dry day the temperature of the cooling water is 23-30 C and its heat capacity is 4,100 MJ/k/cubic metre. Thus to a cubic metre of water to ambient 40 C from 25 C requires 60 GJ i.e. you already have 60 GJ/cubic metre of water before you even start to get any cooling from air. Then air density is approximately 1.2 kg/m3 and thermal capacity 1 kJ /kg/k.
In other words to remove the same amount of heat you need to move at least 4,000 times as much air. One could expand these calculations to account for the poor heat transfer capacity of air vs water etc but the net result is that air cooling in anywhere but very cold climates uses around 5% of the output of a nuclear plants and degrades summer performance even more.
The central argument is energy density, and the wisdom (or lack thereof) of avoiding carbon fuels by switching to a fuel-free paradigm, rather than switching to carbon-free fuel.
Whether the raw materials are abundant or not, a Rube Goldgerg contraption of flow batteries, wind and solar gear, fuel cells, water heaters, car batteries, etc. is what is truly silly.
In our view, nobody would seriously consider such a nationwide Goldbergian scheme — even if it could work — unless they have an outsize (and completely unrealistic) fear of radiation.
Our science and engineering skills, our time, money, effort, and materiel would all be far better utilized solving the problems of high-temp, radioactive, energy-dense power generation, than devising the mechanical equivalent of hamster wheels and rubber bands to power the nation.
Nuclear energy is incredibly safe. Working in a NPP subjects you to less radiation than eating one banana a week. More people fall off roofs and die installing solar every year in the US, than have ever been killed by 60 years of American commercial nuclear power (the death toll of which is a grand total of 8 — 5 of which were in non-nuclear construction and inspection accidents).
Placing 432,000 5-MW wind turbines on land equal to New York, Pennsylvania, Vermont and New Hampshire, on 35-story towers, with 47 of them requiring the start of a major overhaul every day, to (hopefully) generate the same power as 500 reactors, is what would really be silly. That the land can be used for farming and ranching is entirely beside the point. Why in the world should we use that much mechanical equipment, spread over that much acreage, if we don’t have to? And since we don’t have to, entertaining it as a rational approach to our future energy needs is indeed silly.
Reactors and coal plants don’t “consume” cooling water, they warm it up and it cools off again, but it’s still water that’s available for use. Hydrogen fuel cells, by contrast, disintigrate the water, and we have to wait for it to come back to earth as rain for that water to be available for use.
Yes, our favorite reactor technology is MSR, but we clearly show that our argument is just as valid using well-proven Gen-III APR technology.
Tracking solar achieving 32% CF, and wind achieving close to 50%, would change the overall numbers slightly, but our argument is entirely valid regardless.
The exit port temperature of an MSR can be used for supercritical CO2 (“Brayton”) turbines. Until such technology is in wide use, the generated steam can be expanded and thus cooled to the proper temperature to run multiple turbines. The claim that building a 700ºC reactor would be a waste of energy is specious. For one thing, the exit port molten salt could be used to generate space heating, or used for industrial process heat. Indeed, why would high-temp / dense / reliable heat energy be seen as problematic?
GE moves their HA-9 Harriet by truck and barge. No reason why NuScale can’t do the same with their modular reactors. ThorCon will employ ships, as their system is dockside. And, the ThorCon reactor will be water cooled. (We said most, not all, SMRs, can be air-cooled.)
The efficiency of an all-electric hydrogen fuel cell cycle is 26% — that is, only 26% of the volume of electricity that was originally used to split the water is eventually recovered as electricity when the water is reconsitituted.
1. You really don’t do your credibility any good by disparaging remarks like hamster wheels and Rube Goldberg.
2. You claim that an unproven immature technology can be 8 times as reliable over 60 years as any other known heat engine has been over 40 years even when some of the existing systems have been developed over 80-150 years
3. Then you reinforce your unreality by ignoring grid constraints demand variation etc to understate the capacity required even if your unimaginable availability could be achieved
4. You further stretch the boundaries of credibility by relying on a technology that has never been demonstrated. The highest temperature steam plants operate at 600/620 C not 700 C. They do not have to cope with radiation embrittlement or erosion from molten slurry on the reactor side. PWRs operate at steam temperatures from 275 C to 345C (AP1,000)
So why would you seriously propose a plant that operates at that temperature and claim life and reliability outside anything achieved by man.
5. The lower the temperature the larger share of the heat is rejected (look up Carnot efficiency) and the less effective air cooling becomes, so cooling water becomes a huge issue. Coal and nuclear plants together today produce about 50% of US electricity and account for 38% of all water withdrawals in the US (USGS). Nuclear plants use more water or are significantly less efficient if air cooled. Your plan would multiply water use by 6-8 times. i.e the US would have to find three times the total US water supply and you say Jacobsen is unrealistic
6. I am not a big fan of the hydrogen economy and I don’t disagree, splitting water and then running it back through a fuel cell or worse gas turbine is inefficient but the overall efficiency of drilling for oil refining it and transporting it and then running it through the average ICE is much worse than 26%.
7. I am also a fan of supercritical CO2 but it is far less risky to use that process with concentrated solar or even gas as a heat source than nuclear, so until it is proven for many years in a safe environment it won’t be a big part of any nuclear system.
8. Energy density is nice but not critical. Germany already generates 210TWh from renewables. The US has 28 times the area and far better solar wind and hydro resources so should be able to generate 6,000 TWh.
Over the next 25 years Germany will replace all its wind and solar with next generation technologies. Average output of its current wind turbines is 1.8GWh per year new models available today let alone in another 15-20 years produce 10-20 GWh. Similarly solar panels today are double current average rating of existing German installations and bi-facial designs produce 50% or more power in winter so by 2040 with the same number of wind turbines and solar panels as they have today with no advances in technology they could produce 700TWh of wind and 100 TWh of solar + 100 TWh from hydro biomass etc. 900TWh is around 60% of today’s primary energy consumption. Replacing most gas for heating with heat pumps and ICE engines for transport will reduce energy consumption in those sectors about 2/3rds so the only sectors where fossil fuel energy will be still the optimum choice is some high temperature process heating and long distance transport. Therefore Germany can easily electrify 90% of its economy by 2037 with renewables. That does not mean it will happen automatically, it will take another 15-20 years after that for all the old plants to transition so by the 2040 it will probably be around 30-35% renewable.
Total energy use of all types in the US is equivalent to about 30,000 TWh. Net of mining and self consumption of coal, petroleum products and gas it is about 28,000 of which about 80% comes from fossil fuels. Typical conversion efficiency of fossil fuels ranges from 15% to 80% an average somewhere around 40%. so the real useful energy delivered by fossil fuels is about 9,000 TWh. so total end user demand is around about 15,000 TWh. If Germany can supply 900 TWh from renewables the US could easily supply 25,000 due to its better resources and larger area .
I am sure you will find some errors in my brief analysis but none of them will be as serious as your huge leaps of faith in unproven technology. Anyway thankyou for the opportunity to further my knowledge
Well, if you support supercritical CO2 or any other high-temperature fluid, then the whole cooling water discussion becomes moot. There isn’t any water on the site.
Raising the turbine fluid temperature into the 700 deg C range with solar reflectors would probably be infeasible due to the amount of reflector mirror area required, some of which would necessarily have to be situated farther from the fluid-heating tower.
For trough technology, the increased length of heating pipe carrying the fluid to the turbine would likewise degrade its ability to retain its thermal energy.
The advantage of a nuclear reactor is that all its thermal energy is concentrated into a small volume, physically close to the turbine.
“Probably be infeasible”, just another guess on your part. The temperature has nothing to do with the area of the mirrors it is to do with the aiming concentration and the temperature capacity of the fluid heating tubes. If you accidentally burn a hole in a tube with a concentrating solar plant, the plant is down but no external damage and the heat exchanger can be accessed within 10-12 hours for repairs.
How long is it going to take to evacuate the salt, cool down and access the heat exchanger in a molten salt nuclear reactor when a molten salt tube cracks or erodes from the inside, then fix the pipe then X ray all the adjacent tubes, then clean up the mess of the now frozen salt leak.
It is at least 10 probably 15 years before supercritical CO2 will be approved for nuclear plants, as I said regulators will want at least 7 years commercial experience in low risk environments before approving it for nuclear. Then there is a large scale redesign and approval of the nuclear plant some of this might go on in parallel but as there are zero commercial supercritical CO2 plants in action and the first is 5 years+ away. In the meantime wind and solar are cheaper who is going to fund all the R&D.
Of course there are a lot of advantages of nuclear, just nowhere near enough to justify large scale deployment. In markets such as Korea, southeast China, the north east US and India, maybe 10-20% of total energy. In Texas and the Southwest, Australia, Brazil pretty much all of Africa etc zero-5%
You are very generous to Germany’s renewables statistic of 210 TWh /year. Less than two-thirds of that energy was produced by wind and solar. The rest came from burning biomass, 50 TWh, and from hydro, 21 TWh.
Hydro is tapped out. Combustion of biomass is “renewable” only in the sense that over two human generations, our grandchildren will eventually extract that carbon dioxide from the atmosphere by regrowing trees or whatever crop is used for fuel. In the meantime, the land is unusable as natural habitat and the CO2 has had lots of opportunity to capture infrared energy and to dissolve in the oceans.
[…] The 2040 figures I gave included the same amount of biomass and hydro as today. They did not fully include the vast capacity for offshore wind that is only recently being accessed and they assumed no real technological process beyond today’s technology
There will be more waste to energy, there will certainly be waste heat capture from industrial processes using low temperature organic fluids and may be some low temperature geothermal
My credibility?
You keep saying that I’m “relying” on Gen-IV MSR, when we state quite clearly in the book, and in several posts here, that our argument is entirely valid with Gen-III PWRs.
MSR, if it performs as predicted, will just make the argument even more solid than it already is. That is not taking huge leaps of faith in unproven technology.
You persist in your fresh water consumption argument, when seawater is commonly used to cool reactors. Municipal wastewater can be used as well (Palo Verde, AZ).
If fresh water is used, the water is withdrawn, but not consumed — it just gets warmed up. When it cools off, it’s available for other uses.
And yeah, as a matter of fact, a half-million ginormous wind turbines on 35-story towers, and 18 billion square meters of PV panels, all of it scattered hither and yon on 130,000 square miles, as a fuel-free scheme to power the nation, is the very essence of a Rube Goldberg contraption.
And somehow, the issue is my credibility?
then our road, power, water, gas and telephone networks, houses etc. are even more of a Rube Goldberg Contraption.
No, Helmut. A Rube Goldberg device is something that is far more complicated than what is needed to perform the task at hand, to the point of being laughable. That can’t be said for our roads, water, gas, telephones, etc.
But it can surely be said for a 100% fuel-free renewables scheme to power an advanced nation of 320 million people, that would require 500,000 5-MW wind turbines on 35-story towers, a third of them at sea, and 18 Billion square meters of solar panels, spread out over 130,000 square miles of land, plus a sea region the size of West Virginia. With virtually no backup or storage factored into the plan. Particularly and especially when the same work can be done with a few thousand reactors, with 18 months of built-in storage, on land totaling one half of Long Island.
Our road, water, gas, etc are not unnecessarily complicated
Roads gas supply, telecomunications etc are much more complex than a renewable power supply, with the sole difference that they are already built, and that you don’t have a imagination about their complexity because it’s not your business.
So if you think 500.000 simple wind turbines are complex, than it would be completely unimaginable to you that such complex systems like the ones mentioned before could ever be built by mankind, if they would not already exist so you do not need to bother with their complexity.
It is less complex than today’s ordinary parallel computer clusters and/or server farms according to both construction and operation.
But who cares, as a consumer, you have your Youtube, Netflix, iTunes, … whenever you want to, as much as you want to.
But they are significantly overbuilt just as the future power system will have to be. Your model assumes perfection which just isn’t realistic.
In any case the area required by solar panels is less than the area of paved roads in the US and at least half of it will be dual use such as grazing fields and in many cases improved use shade for car parks and even vegetable growing.
As for you windfarm area, ask the farmers who have them, how much productive land they have lost. about 0.3% in the highest density windfarms
Peter – Perfection? Hardly. If reactors were even half as productive as we state, an all-nuclear grid would still be a far better solution — cheaper, more compact, scalable and reliable — to our national energy needs.
A Gen III nuclear reactor needs to be accompanied by storage for ramp support, trip support, refuelling support and peak/trough shaving. Even though Japan only ever supplied 30% of its electricity from nuclear its 53 GW of nuclear was supported by 29 GW of pumped storage.
The problem for your argument that Gen III will work, is there isn’t enough water and it will cost more and need as much or more storage than a renewable grid. This best of breed plant is quoted at US 11 c/kWh if everything goes to plan. The new Rosatom plant in Turkey at US 12.5 c etc when solar PV in the middle East is 1.9 c solar thermal 7 c and they haven’t even seriously begun to explore wind
You cite the Emirates plant which is not yet turned on and yet their $160 bn future energy plan includes no more nuclear. The Koreans don’t expect to build any more nuclear than the plants currently under construction.
While I agree fears of nuclear plants are overblown and safety requirements excessive, they still can’t ramp quickly and even if someone figured that out, the economics get even worse because you are spreading fixed costs over less MWhr and thermal cycling reduces their life.
Japan is unique. The UK has not had to develop storage to support its nuclear plants.
At one stage Japan was aiming for an all nuclear future and saw storage as the main way of dealing with daily peaks and troughs in demand. They didn’t get there, and today demand management is far more sophisticated enabling demand curves to be smoothed. Early generation reactors were not designed to load follow either. Japan is isolated too from other grids so needs to be self sustaining.
Japan’s seismic issue means power stations are at risk of trip. That might have been a driver too for so much pumped storage to cope with the simultaneous loss of power gen. capacity during regular earthquakes.
The nice thing about renewables is that there is such a wide range of technologies and materials to choose from. Wind turbines can be built with and without gearbox, for the generator you can use permanent magnets or electro-magnets, towers can be build from steel, concrete and even wood. Solar cells can be made from silicon, Cadmium telluride, Cupper Indium Sulfate, and so on. Backplating of the cells can be done using silver, but also with copper. Copper cables can be replaced by aluminium cables, and so on.
Furthermore, recycling will also reduce the demand for new materials.
So overall the fear of material shortages or extreme high prices for materials due to growth in renewables seem rather far-fetched.
Depends on the price and availability of those alternative materials, and the number of panels you seek to fabricate.
there is enough material available of these alternatives, and for the mainstream panels the main parts of materials are among the most frequent in the earth crust.
Not for silver.
See Ch. 5, section “Low energy? You might have a mineral deficiency.”
End Notes #7, 8, 10 and 12.
Helmet – you are missing the point. The essence of a Rube Goldberg device is UNNECESSARY complexity perform the task required.
If you can generate power with 2-6,000 reactors (depending on their size) on 700 square miles, then why in the world would you bother to attempt to generate the same power with 500K turbines and 18 Billion sq meters of panels on 130,000 square miles?
Complexity is not measured in squaremiles nor in numbers of identical uints. Your SMR anrd their supply chains are much more complex as a sytem compared to the extremely simple Photovoltaic systems and the still simple wind turbines, interconnected by not too complex HVDC-GRIDS. Nuclear Power stations are notorious for their coplexity which is also cited as cause for the cost overruns of the AP 1000.
Helmet — The issue isn’t complexity. The issue is unnecessary complexity to perform the task at hand. A good example is mass energy storage.
A national renewables scheme will require a national web of Li-ion Batteries, flow batteries, pumped hydro storage, flywheels, ice, water heaters, car batteries, etc. when mass energy storage already exists. It’s called fuel.
reply to Nigel West comment on Jan 26, 13:43
Nigel, regarding http://euanmearns.com/concentrated-solar-power-in-the-usa-a-performance-review/ ,
those capacity factors graphed in Figures 2, 3 and 5 are bogus.
They are conjured by an accounting gimmick that sets aside a portion of the facility’s solar reflectors that are not counted when stating the peak capacity of the solar field. The industry’s term for this is “solar multiple”.
Refer to Ch. 11, End Note #33 for details.
The purpose of a solar multiple is not an accounting scam, it is to allow the plant to store energy while generating. So a plant that wants to offer 12 hours storage on average has to be able to capture 2.2 to 2.5 times its nominal generating capacity.
Having said that, at the same areal density as a solar PV farm the heliostats on a solar thermal farm are around 90% efficient in reflecting incoming solar back to the tower, the thermal system about 40% so notional 36% less operating energy for pumps, trackers etc etc probably 28-30% of the incoming radiation gets converted into electrical energy.
On the other hand a tacking solar PV farm is about 20% conversion efficiency so solar thermal is still more space efficient than PV
The ideal solution is probably a mix of solar thermal, solar PV and wind at the same site. By using the solar PV during the day the solar multiple of the solar thermal can be reduced and solar PV and wind will still generate when it is overcast. If prices go negative resistance heating from wind or PV can heat the the salt store. It is inefficient but it is better than zero price for your power
[…]
Reply to Peter Farley comment of Jan 11, 14:17
a) 3600 TWh is correct for FF, hydro and nuclear. Counting biomass, dispatchable NRG was 3770 TWh in 2017.
US electric capacity is 1200 GW, not 8700 GW. Of that, about 1100 is dispatchable. See my Critique of 100% WWS Plan.
Dispatchable capacity factor is 40%, not 49%. [3770 TWh actual / 9460 TWh potential = 40%]
I agree that we should have dispatchable capacity equal to at least double the US average load, which is 467 GW. That’s one of our objections to Jacobson’s Roadmap – not enough overbuild.
b) KEPCO’s Generation 3 project in UAE came in at $4.40 /Wavg. See FN 12 of our Ch 1.
Sanmen Generation 3+ units 1 and 2 apparently came in at about $5.50 /Wp, or $6 /Wavg.
Compare to utility scale PV solar, our FNs 7and 10 of Ch 11. http://www.nrel.gov/docs/fy16osti/67142.pdf
$1.75 /Wac / 21% CF = about $8 /Wavg.
China National Nuclear Corporation expects their learning curve to bring AP-1000 construction cost down from Sanmen’s $5.50 cost in 2018 to $3.00 /Wp, about $3.30 /Wavg. See http://www.world-nuclear.org/information-library/country-profiles/countries-a-f/china-nuclear-power.aspx
Comments on this now closed. thanks everyone for a (surely) record-breaking discussion!