Energy
blog.samaltman.com
blog.samaltman.com
Meanwhile, solar/wind is heading toward dirt cheap and trivial to set up. Environmental impact is minimal, too. It doesn't require giant corporations, government sponsorship, complex regulations, or exotic engineering skills to implement. With those incredible advantages, it doesn't need to be cheaper than nuclear - it just needs to be adequately cheap.
Solar with batteries is a lot more expensive than solar alone. Solar is doing well right now with natural gas backup, but a lot of us would like to avoid fossil entirely. Solar with nuclear backup might be a great combination, assuming demand correlates reasonably well with daylight hours.
Do you really want lakes full of residual waste for the next few tens of millennia?
The UK currently has no idea what to do with a lot of its waste. So it's simply left to rust and ferment in water - not a good outcome.
Sustainable intermittency turns out to be something of a myth anyway. Intermittency effects across Europe turn out to be negligible.
So instead of building nukes, you can spend the money on mixed-mode sustainables and a hugely improved distribution grid and get a cleaner and more reliable outcome overall.
You can also make sustainables distributed, and run them on a domestic scale as well as an industrial one.
PV roof installations have worked well in Germany and are starting to work well in the UK, even though neither location is known for being sunny.
You'd get much better results in the sunnier parts of the US.
Helion is a fusion reactor. Its "waste" is helium, and it uses a reaction that produces only 6% of its energy as neutron radiation.
In both cases the reactor itself may become somewhat radioactive, but that's another short-term problem.
Certainly, solar with batteries is a lot more expensive - but to my point, startup cost is low. Latency, or bandwidth? The long-term cost/kwh may be higher for solar, but the short-term startup expense will be much lower, unless you can get the factory-built reactors down into the thousands rather than millions of dollars.
If nuclear is cheaper than solar in a decade or two, but costs ten times as much in the short run, there's a tremendous advantage to solar, battery cost or not. Opportunity cost matters tremendously.
UPower's design doesn't require a large body of water: http://spectrum.ieee.org/energywise/energy/nuclear/startup-d...
For setup you deliver it with a truck and bury it. Other modular nuclear designs are a bit bigger but still in the range of small natural gas plants, which are competing quite well.
If Helion works out then fission and solar will both be mostly obsolete. They'd be 50MW plants, retailing power at four cents per kWh, with no significant safety concerns.
Too many reactors are designed without the market or financing in mind.
We decided on the simplest possible reactor optimized to a size useful to a market in dire need- just MW scale.
It has no pumps, no water in the reactor, and builds upon a legacy of data so that there will be minimal fuel and materials qualification, which adds up very quickly in both time and money.
Why hasn't it been done before? The key, as you bring up, is in manufacturing, simplicity, a relatively new and hugely growing microgrid market that didn't exist much before, and a business model that doesn't require the customer to buy the unit as opposed to power purchase.
[1] http://www.bloomberg.com/news/articles/2012-11-07/fukushima-...
(Nuclear reactors have the same "too big to fail" problem as the banking system: the expected value is positive, but the worst-case is a huge open ended liability and toxic asset)
It looks like UPower is currently targeting environments where traditional power is impractical and lots of power is needed, and plenty of budget is available - remote mines, military installations and such. Do you see a market for urban/residential power grid in the future, too? Or would that be too difficult a squeeze between distributed solar and traditional power plants?
That wouldn't work with current nuclear reactors, whose startup / shutdown sequences take hours. Typically, the grid operator runs nuclear plants at nearly 100% 24/7 (so called "baseload") due to this.
I'd be curious how truck-sized reactors work and whether or not you could operate them as load-followers rather than baseload. That would make them extremely attractive for replacing natural gas peakers, especially as more wind / solar get onto the grid.
But I suspect the small reactors will be more flexible. Molten salt reactors are supposed to load-follow automatically with a lag of thirty seconds or so. (And of course with small fusion reactors it wouldn't be a problem at all.)
http://ansnuclearcafe.org/2013/02/14/responding-to-system-de...
One of the main limitations is the stress it puts on the fuel.
Many advanced reactors overcome these limits, and if financially incentivized, they will definitely load follow. On top of that, the UPower reactor has a thermal transport time constant nearly 10 times that of other reactors, and its fuel is immune to the shocks that bother LWRs. In fact, the same type of fuel was used in a research reactor and would be ramped in power from 5 watts to 150 billion watts in less than 50 millionths of a second. That puts a lot of stress on fuel, yet this fuel kept its stride without breaking a sweat.
This reactor is built like a tank, and is designed to be quite resilient and flexible. It can definitely load follow to support a renewable heavy grid system. In fact it's been considered for use as a grid stabilizer at substations because of these abilities.
The only non-solar non-fusion energy I can think of is that which comes from the earth's core, geothermic. And even that is believed to be powered by nuclear fission.
For 1000 kV AC transmission line from Arizona to California, transmission losses should certainly be smaller than 5%, even less for DC. Arizona is pretty close to California when it comes to electricity transmission.
There are other concerns: elevated highly variable salt water ponds and wetlands, and general nastiness of saltwater engineering: corrosion and marine life growth especially.
But generally, pumped hydro is really hard to beat.
As an added bonus, they're based just down the road from me!
Transmission and distributed generation accounts for half the cost of electricity. The question for our electricity future is distributed versus centralized generation, and distributed will probably win.
Sure, UPower and others are working on reactors that are small scale, cheaper, safe, and hard to weaponize. But they're still a limited solution to the problems I'm bringing up.
Fusion is a fantasy. Maybe someday it will be real, but betting the world on it is foolish.
Solar is cheap, fine-grained, clean, sustainable, and not weaponizable. It solves all my core problems. Why should I care about difficult, expensive, dangerous nuclear?
I like this metaphor. I think what you might have implied but didn't explicitly mention was distributed vs centralized when thinking about these systems as well. Solar can be a distributed system whereas fission & fusion systems are centralized.
Even though you can make small fission systems and IIRC Skunkworks is working on a small fusion reactor (small = fits on freight truck...once the damn thing works) you still have issues with waste, heat, faults etc. so they need to remain isolated from living space. Thus they are better utilized in a centralized manner.
Sticking with the computer/internet metaphors, you could consider solar to be like a Solid State Drive while fission is very much the classic HDD, moving parts and all. The elimination of moving parts/complexity to generate electricity makes solar suitable for the home just as NAND is better for portable devices like phones.
Pros: speed vs capacity, Cons: cost vs complexity - pick one from each pile.
Very cool, but feels kinda not as solid if you know what I mean. Don't want to downplay how neat it is though.
That's what I mean in the context of centralized vs distributed - replacing a faulty solar panel is no biggie. Replacing a micro fusion reactor, while not as bad as replacing a tokamak reactor, is significant.
Neither is waste [...] unless you insist on crowbaring into the core
Reactors still need to be decommissioned at some point, though you're right - it's trivial comparatively. I will do a dance when fusion is a thing and the world will be a better place.
This is all Freeman's fault, isn't it.
Yet.
I'd love to see this change, but I'm not gambling humanity's future on the assumption that this is a solvable problem.
On storage: Rough calculations show, if there were just enough Powerwalls to backup US peak demand for one hour it would require 10x the global annual mining production of lithium. And that's just one hour. And that doesn't include the electricity production.
On panel material required for production: It's generally estimated that US power, with good transmission, would require enough solar panels to cover the entire state of Massachusetts. Most non solar advocates think it's this square footage that's important. But of course this can largely be put on built land or in deserts so that is a relatively moot point. In fact, I want to get solar panels on my roof. However the real concern is what does this look like in terms of material mining? In immense panel production factories? (which isn't the greenest mfg process ever, likely one of the reasons it is largely done in China)
On mining and transporting material required: Mining is almost entirely powered by fossils, it has to be. And so is most transport. And so is recycling of metals. So the energy density of an energy source really is a zero sum game. If it takes a millionth the material for one source versus the other, that adds up.
On maintenance Then in maintenance, solar farms are truly "farms"- they require a lot of water to wash away dust to operate optimally. A states' worth of water is significant.
On lifetime/end of life First of all the lifetime of a panel is very optimistically 30 years/for a nuclear plant 60-80 years, and for the UPower fuel in particular can be used and recycled repeatedly for about 70+ years.
afterlife/recycling Then in recycling at end of life, and this is why I got so excited about nuclear as a somewhat hippie child growing up around oil companies in Oklahoma, solar is going to require a lot of energy (and fossil fuels or nuclear) to recycle, while nuclear can produce energy in recycling its fuel.
The main import, to me, is: what is the energy density of this energy, and if emitting, how much pollution? Coal is far more energy dense than wind, which is why humans evolved from windmills and wood to coal. But it's so polluting which is why we are all working towards better sources, and the greater energy density (nuclear on order of 2M x any other source) that's roughly 2M less trucks transporting, 2M less mining to do, 2M less recycling, etc. Thats more on the environment than pure cost like you are saying but the costs add up if the full life cycle is taken into account on both sides.
What?!? I certainly appreciate the ambition, but humanity has spent seven decades and at least hundreds of billions of dollars on this very same project. What in the world is this tiny startup doing with a $5mm grant that is so easy and cheap that could possibly lead to that kind of breakthrough in fusion energy?
How is that a measure of doing something meaningful ? Take EADS making the Ariane rocket launchers, they have been spending billions and billions of Euros making rockets for years, yet they have no project like SpaceX and yet they will be rapidly obsolete once SpaceX works as expected.
It's not about the amount you spend, it's about experimenting in new areas not explored yet. Now, about fusion, I have no idea what they plan to do, but in principle there's always new things worth trying (even if they end up failing).
But to be honest, I think we need to encourage more effort, so I applaud the project anyway, and keep my fingers crossed.
I disagree on that point. SpaceX is not just making rockets cheaper, it's changing the paradigm by making rockets re-usable, which is very far from what EADS is planning to do currently. It's a game-changer and will divide the costs of launching something to space by 10 or more.
I cant comment on the Fusion startup, but we should not assume that companies operating with billions of cash are more effective at innovation than smaller ones with smaller budgets. They aren't, and that has been proven in many cases, SpaceX is just one good example coming to mind. In the pharma world, very small biotechs with minimal funding are responsible for the discovery of many new drug targets, and not the large pharma groups themselves.
YC has demonstrated they aren't just throwing darts here, so vague pessimism is not the best default.
http://www.nature.com/news/plasma-physics-the-fusion-upstart...
EDIT: This admittedly isn't very much information, but it's more than I could find on their official site.
They have had previous success[1], and the torus-shaped magnetic containment fields used in most fusion research is known to be a problem.
[0] https://en.wikipedia.org/wiki/Reversed_field_pinch
[1] http://www.alternative-energy-action-now.com/helion-energy.h... (yes, that domain sounds terrible, but the info seems ok)
Fusion advanced exponentially from 1970 to 2000, at about the pace of Moore's Law. Then governments decided to put most of their fusion money into a giant, poorly-managed construction project in France. But we're not that far off, and in those seven decades we've learned a lot about plasma physics, gotten much better computers for plasma simulations, and developed all sorts of other enabling technologies.
Sometimes it's not the size of the investment that dictates success. Often you need a good reset, with a smaller, better aligned team.
Everything that ever happened spent a long time not happening first. One must be super careful extrapolating inactivity.
From Wikipedia:
Of particular concern in nuclear waste management are two long-lived fission
products, Tc-99 (half-life 220,000 years) and I-129 (half-life 15.7 million
years), which dominate spent fuel radioactivity after a few thousand years.
The most troublesome transuranic elements in spent fuel are Np-237 (half-life
two million years) and Pu-239 (half-life 24,000 years).[39] Nuclear waste
requires sophisticated treatment and management to successfully isolate it
from interacting with the biosphere. This usually necessitates treatment,
followed by a long-term management strategy involving storage, disposal or
transformation of the waste into a non-toxic form.[40] Governments around the
world are considering a range of waste management and disposal options, though
there has been limited progress toward long-term waste management
solutions.[41]
Germany:
https://en.wikipedia.org/wiki/Greifswald_Nuclear_Power_Plant Decomissioned: 1990
Destruction started: 1995
Ongoing as of today....
To be honest I'm just asking the question that while we can all say "isn't it great how safe nuclear is" we still haven't 100% figured out the waste issue. Fusion... well I'll believe it when it's powering my DeLorean.
http://www.bbc.co.uk/news/business-31725365
Behind schedule and over budget is totally normal of UK big projects. But still, £58bn is a lot of money to fix something that's not a big problem.
I imagine there have been a lot of places in your country (as with any industrialized country) that have been horribly contaminated with non-nuclear pollution, then cleaned up. While these events can certainly be used as an argument for taking more care, they're almost never used as an argument to give up on the whole idea of industry.
Nuclear waste is bad, but it doesn't seem to be on the level of "we must lock this up so securely that God Himself cannot access it" as people seem to try for.
If something is infinite, meh, it's infinite, what you gonna do. But if it's a few million years ... whoa, we can reason about that!
I see your point. I think one needs to consider both the danger of the waste and the amount of it generated. Then try to find a good balance.
The long-lived fission products are a very small portion of the total waste, and if we only had fission products to worry about, the total waste would go back to the radioactivity of the original ore within three centuries or so, since most fission products are relatively short-lived. It's pretty easy to contain them for that long.
The great majority of the total waste, and almost all the long-lived waste, is transuranics. Those can also be considered unspent fuel, and fissioned for energy in fast reactors or with neutrons from D-T fusion reactors. The Russians have a couple fast reactors in commercial operation, one of which has provided 560MWe to their power grid since 1980. They're working now on using them to process transuranic waste.
[1] http://www.nrc.gov/waste/decommissioning/finan-assur.html
https://en.wikipedia.org/wiki/Nuclear_Decommissioning_Author...
Current estimates for one Nuclear Power Plant Site (Sellafield): £114bn which seems A LOT!
It's not going to be representative of the decommissioning costs of modern generation II or generation III reactors.
Regardless of how much one is for or against nuclear power, i think we can all come together and agree strongly that building an air cooled nuclear reactor that uses a flammable substance as a moderator and blows exhaust out a chimney is a catastrophically bad idea.
And I don't imagine that kind of activity results in easy cleanup.
The NRC estimates costs for decommissioning a nuclear power plant
range from $280-$612 million.http://www.nrc.gov/waste/decommissioning/finan-assur.html
Cost does seem quite high right?
They are supposed to. But here in germany the energy company lobbists basically said - in response to a few new proposed fees levied on coal plants - that anything threatening their revenues would cut into future revenue needed for decommissioning fees. And that's despite being required to have funds set aside for decommissioning.
I suspect some creative accounting going on, moving expected future profits into those funds or whatever.
So if their business models fail - e.g. due to renewables - it might turn out that in practice there never was any money set aside for decommissioning.
There are even talks about splitting off coal and fission plants into some sort of "bad bank" companies.
So I would be very wary of any such promises of things being priced in.
Corporations have again and again proven to be very good at privatizing profits while externalizing costs.
That's an important question, but climate change is a far larger bill underwritten by the tax payer (and millions or billions more who can't afford taxes), so I don't think the question is a deal-breaker.
- grey energy invested
- full lifetime costs from start of construction until fully recovered land
- inclusive recycling costs
- inclusive environmental impacts
Low level radion is dangerous. http://www.ianfairlie.org/news/recent-evidence-on-the-risks-...
"However the latest study, (Zablotska et al, 2013) is very large (over 110,000 workers) and succeeded in finding statistically significant leukemia increases, even at the relatively low doses experienced by most of these adult workers (average dose = 92 mSv)."
Both of these are critically important for existing waste but also having an emission free energy source with a closed fuel cycle. No other energy source is better than a hundred thousandth as energy dense and no other energy source could produce clean energy for its own recycling.
It's obviously just a cool technology but more than that it's amazing what that could mean for the environment and remote communities. Even for solar and wind materials mining, these remote mines generally have to burn tons and tons of diesel. That's what we are trying to fight.
Ours is designed to live in essentially a spent fuel cask. These things have been designed and tested to withstand being dropped from a thousand feet, being hit by missiles or airplanes. Seriously check out youtube there are crazy videos like this: https://www.youtube.com/watch?v=jBp1FNceTTA
So from the time the reactor leaves our factory, to when it's put underground, till a decade later when it's taken out and shipped back for refueling, the fuel is locked in the reactor which is locked in this uber robust "cask".
This design is a fast reactor which means it can recycle fuel. So beyond its first decade installation we can recycle the fuel approximately 6 cycles before there is any amount of leftover that must be removed. That would be about 70 years, and the volume would be about the size of a basketball (fully glassified) and the lifetime would be on the order of a hundred years. That could be stored at our central facility or another facility, as it would not be weaponizable.
Compare that to waste generated from burning dead dinosaurs - its storage is not a problem only because it's dumped into the atmosphere to poison everyone. You can say that nuclear plants have one less externality here.
It's silly to describe solar as atomic energy from sun’s fusion as a distinct category from carbon-based energy, because of course carbon-based energy was formed by capturing solar energy from the sun's fusion.
I agree this is silly. Harnessing the radiant energy of the sun is categorically different from converting wind or carbon into usable energy. Trying to group these into one category seems out of touch with the reality of developing the t technlogies.
Distinguishing between carbon and atomic is a pretty useful and meaningful distinction, although ultimately not perfectly correct.
Fine, call it a 'carbon-cycle' energy system. Which may make it more clear that we're doing a great job on the 'burning' part of the cycle, but not so good on the 'capturing' part of the cycle.
It's 2015, and we remain essentially stuck on 1950s B-Movie era stereotypes on how nuclear works, what it is, and how dangerous it is.
For what remains after that, well, there's nothing particularly wrong with sticking it in some extremely sturdy armored containers, sticking the containers in a shed, surrounding the shed with a razor-wire fence, and stationing a couple of guards at the gate.
The paranoia over nuclear waste is weird. What if civilization collapses and people in the far future discover the stuff and don't know it's dangerous? It lasts tens of thousands of years, after all! OK, but we leave dangerous stuff all over the place. There are tons of barrels of toxic chemicals, ponds full of poison, and heaps of awful stuff just sitting around, and more being produced all the time. Nuclear waste is bad because it lasts tens of thousands of years? How about arsenic or mercury, which lasts forever? And sure, precautions are taken with those toxins, but not on the "this must be placed in a geologically stable area, packaged to survive the fall of civilization, and signposted so that all future intelligence will know to avoid it" level of precaution.
We have no working waste solution as of today.
We have no working long term storage solution as of today.
We have not priced the cost of long term storage/plant decomissioning etc at the full cost into the price of nuclear power - that is for most countries.
https://en.wikipedia.org/wiki/Asse_II_mine
http://www.spiegel.de/international/germany/dealing-with-ass...
We have no working long term solution for storing arsenic, but people mostly seem fine with "just dispose of it in a way where it doesn't leech into the groundwater" and not the crazy restrictions everyone wants to put on nuclear waste storage.
Can't you protect yourself and go cleanup an arsenic spill "easily" compared to how you would cleanup a radioactive spent fuel rod container which tore open?
IIRC the latter is a real issue currently at Asse in Germany.
Most choices don't get made on the basis of "what would be a better idea?".
What do you mean? The waste is full of nuclear poisons (a technical term, isotopes that stop fission by neutron capture). Reusing it means reprocessing to extract the Plutonium, which is burnable in a reactor (particularly as MOX). The chemistry required for this is pretty messy, and the radioactive environment (fuel from a power station is very 'hot' radioactively speaking) so horrific that you need a fully robotic plant.
This hasn't gone great for the UK, https://en.wikipedia.org/wiki/Thermal_Oxide_Reprocessing_Pla...
But we do now have 112 tonnes of Plutonium as a result of our half a century of reprocessing. Which no one knows what to do with, and is an enormous liability.
What to do with the UK's plutonium seems obvious to me: use it as nuclear fuel. If this isn't being done it's presumably because of political opposition, not technical problems.
Not doing that route makes you less exposed to risk.
In an extremely complex world where risk needs to be minimized as much as possible, it's totally reasonable for society to decide not to extract plutonium at mass scale in a free market way.
Sure if the military controlled the whole thing (as they do via their own supply chain) that could perhaps feel more secure... but who controls outflows of waste between governments, etc?
Not doing that route makes you less exposed to risk.
In an extremely complex world where risk needs to be minimized as much as possible, it's totally reasonable for society to decide not to extract plutonium at mass scale in a free market way.
Sure if the military controlled the whole thing (as they do via their own supply chain) that could perhaps feel more secure... but who controls outflows of waste between governments, etc?
Oh and the mind-boggling cost. Who pays?
Nuclear is one of the most expensive forms of energy when you cost in the full price of making it safe, including dealing with waste and ensuring absolutely no proliferation of weapons.
Also, UPower can use the waste without putting it through a chemical separations process. In fact you can just take the SNF grind it up, and dump it into the UPower reactor alongside the rest of the fuel. It actually makes a pretty good fuel that way.
Even if they do figure out how to get engineering and manufacturing in place, then there are regulatory barriers to figure out. There will be politics involved, and legal challenges. And of course, you need to actually operate the sites on an ongoing basis.
All of these challenges can be overcome... but most people who know how to do so already work in the energy industry. What is really needed is a group of people who can bridge those gaps to take an innovate design from an engineers drawing board, and jump through all the hoops to make it a live production site. If such a group were to be built, real change could happen very quickly.
Frankly, most of the sketchy folk that I ran into during my projects were the investors, not the engineers.
I've seen a couple of both, but for the most part the engineers when they are wrong are deluding themselves as well, the investors that are sketchy seem to be more cynical and aware of what they are doing.
Kudos to Sam for putting not just his money, but his time where his mouth is. If only the rest of the internet could do the same!
Sam is in a fairly unique position to be able to do that...
On paper the waste issue is still not solved. And the risks are still large (yes I know, coal, gas and oil has it's drawbacks too).
It's also important to highlight that the UPower design can consume the entire actinide vector because it uses fast neutrons. A lot of the longer lived actinides cannot be fissioned or transmuted effectively by thermal neutrons so they just build up.
We like to say we are the ultimate disposal, and can take anything, including the waste from other waste consumers.
Fusion has no significant risk. With fission it depends on the design; I don't know much about UPower, but the IFR, another small fast reactor, demonstrated very impressive safety.
This makes a lot of sense.
Can anyone point out current research on this field? I don't seem to hear much about it.
Edit: Just found this after searching #photosynthesis in Twitter: http://www.huffingtonpost.com/2015/04/20/artificial-photosyn...
We should look to nature before designing things since it may have already solved the problem for us. Future technology should also be designed to gracefully degrade like nature, otherwise we're just replacing one problem with another.
I think Cosmos missed a really important lesson which is that the fuels at our disposal are all a function of time and distance. The longer a fuel source has been building, and the less distance it has to travel to be useful to us, the more valuable it may be. The sun is a result of billions of years of the shape-shifting games between mass and energy, all driven by gravity. The fusion energy produced in the sun then has to travel 93 million miles to us to be useful. The food chain harnesses this energy and accumulates it over time, and after hundreds of millions of years much of that energy has been sequestered into fossil fuels. While there is a tremendous amount of power emanating from the sun, it has to go a long way or accumulate for a long time to be useful to us. Nuclear fuel sources on the other hand bring the billions of years of nucleus building that previous generations of stars did for us to our door step. The parent stars of our sun produced heavy actinides like uranium or thorium, as well as the abundant light elements like deuterium, helium, and boron, and then scattering them across the cosmos along with leftover hydrogen in brilliant novae and supernovae. In our case, many of these elements were in the stardust that formed earth, and are here beneath our feet and above our heads.
Solar, wind, and nuclear will dominate the 22nd century, but we need both, and they do and can play well together. They just need to be treated and respected equally.
PS: https://what-if.xkcd.com/73/
Which of the following would be brighter, in terms of the amount of energy delivered to your retina: A supernova, seen from as far away as the Sun is from the Earth, or The detonation of a hydrogen bomb pressed against your eyeball?
A: Applying the physicist rule of thumb suggests that the supernova is brighter. And indeed, it is ... by nine orders of magnitude.
Use TCP/IP to synchronize the consumption of power with its production. Solar and wind are pretty awesome if you can schedule around them. And most of our power consumption is amenable to it.
The reason that China and India are the only countries going after fission is because of a singular element - Thorium. Both India and China have huge reserves of thorium that can be unlocked with molten salt reactors that are unviable anywhere else in the world (including the US, which gave up on this a long time ago[1]). Australia does have large reserves of thorium, but its projected energy needs are dwarfed by India and China's.
China is way ahead than India on this front with more than a billion dollars managed by Jiang Mianheng to conduct research into these new reactors. And which is why India is bending over backwards to sign the India-US Civil Nuclear Energy treaty.
Interestingly, a US company, Thorcon [2], has built a "hackable" MSR - though I dont know if it is any good.
[1] http://fortune.com/2015/02/02/doe-china-molten-salt-nuclear-... [2] http://fukushimaupdate.com/thorium-molten-salt-reactors-to-g...
I would love to work on a project like this.
On the grid front, nuclear is good for baseline, but not peak load. Most nuclear reactors can't just be flipped on and off with a switch, or even scale power quickly... they're more or less constant. So you still need a peak load system, which currently consists of gas plants - very expensive, since they're intermittent and offline most of the time.
It's worth pointing out that the problems enumerated above are partly the consequence of energy becoming cheap and available during the last century (coal, oil, etc) not lack of it. But the main reasons are the dominating philosophical and ethical standards of humanity during the energy boom. Cheap energy + wrong philosophy = bad application of energy = problems enumerated above.
So if you want to tackle any of those problems, you have to work on both variables in that equation, just increasing the availability of energy without raising awareness of how to apply it, will lead to unsatisfactory results in the long term.
Are we talking public cost? Because that's all that matters. So far the public cost of nuclear power has been extraordinary, due to accidents and waste.
I understand that some of these startups aim to process existing waste in relatively small distributed reactors but what is the public cost of spreading a bunch of "mini" toxic waste sites around the world that remain hazardous for 100 years, instead of centrally storing it?
Plus I've read that although these mini reactors are not directly producing material that could be used in a dirty bomb, that they could be converted to do so if they fell into the wrong hands. I may be oversimplifying here, but the question again is what is the public risk of distributing atomic fuels and reactors in a manner that makes them much less secure? Would this make them more susceptible to "war hacking" and could this be the mini-reactor equivalent of a nuclear disaster?
Nuclear costs have always been about the long term public costs, not the short term $/kWh.
This is before we even consider the taxpayer cost that's gone into nuclear tech development. I wonder if there will be more public money needed to take this tech to market, even if the test reactors bear fruit.
"In Finland the world’s first permanent repository is being hewn out of solid rock - a huge system of underground tunnels - that must last 100,000 years as this is how long the waste remains hazardous."
It was eye opening.
Fourth generation reactors (and thorium reactors) do improve the situation considerably (eg, 2 orders of magnitude less waste than a conventional reactor, and the waste stays around hundreds of years instead of tens of thousands[1]).
However there is an existing set of reactors that will be in use for decades, and their waste needs storage. There are also existing "temporary" storage facilities that need better long-term solutions.
[1] https://en.wikipedia.org/wiki/Thorium-based_nuclear_power#Po...
Why not just turn the waste storage sites into reactor farms?
That would sidestep the problems stemming from decentralization and probably achieve some economies of scale too. Maybe smelt some aluminum with the energy. But even if you essentially threw the energy away, it might be useful as purely a waste cleanup solution.
And isn't the fact that they'd be "spec ops" caliber guards reflect that there is in fact a security risk with these materials getting into the wrong hands? If so it seems like a committed enemy could attack one of the thousands of these sites successfully. The distributed model seems problematic.
However, my understanding is that Thorium-based fission reactors[1] reduce the risks somewhat. I haven't looked into this enough myself to decide either way, but I do have an open mind.
Fusion, OTOH is a totally different story. Maybe, one day we'll get that to work reliably and cheaply. I find it difficult to imagine a more significant change to the world we know.
[1] https://en.wikipedia.org/wiki/Thorium-based_nuclear_power
When it comes down to it, power generation is always risky. Reasonable efficiency demands that any power plant tries to maximize energy density, which is fundamentally in conflict with safety. Even solar and wind have risks, though obviously they are smaller than nuclear, coal or natural gas (though wind and solar are a different class of power generation and don't necessarily compete with those).
The main risks in nuclear power are loss of coolant, containment breach or a prompt criticality event. Thorium reactors are vulnerable to all of the above, but of course designs exploit various properties to mitigate the risk. The most famous thorium design is the liquid salt design, which is not as susceptible to loss of coolant flow accidents due to a reactivity feedback mechanism (it's subcritical if the liquid fuel stops being circulated), but it isn't inherently superior to other modern passively safe designs.
The real reason to be interested in thorium is that it has reduced proliferation concern and also thorium is available in places where uranium is rare (India). But it isn't fundamentally safer.
WHO estimates the increased cancer risk of people living inside the Fukushima Prefecture as being up to 70% higher (thyroid cancer for girls exposed as infants)[1]. Numerous other studies indicate increased cancer risk[2].
During the Fukushima meltdown, radiation levels of 3–170 μSv/h (= 17 mrem/hour) were measured within 30 km of the reactor[2]. Safe levels are 5000 mrem/year[3].
As I understand it, many argue that these of cancer risk estimates are high. I'll be happy to change my mind once medical scientist working in the field change theirs.
[1] http://www.reuters.com/article/2013/02/28/us-japan-nuclear-c...
[2] https://en.wikipedia.org/wiki/Radiation_effects_from_the_Fuk...
[3] http://www.nrc.gov/about-nrc/radiation/health-effects/info.h...
The reactors cannot be deviated for nefarious purposes. And the materials are not less secure. The materials are being consumed by the reactor, and they are not dangerous as they are. In fact these reactors could destroy weapons grade material that is slated to be destroyed for fractions of the cost of programs the US is pursuing. Plus the reactors are secured when deployed. They are also buried and completely cooled by natural forces so they always stay cool. No fuel overheating.
The reactors cannot be hacked, and if a bad actor commandeered one, all they could do is turn it off safely. Even if they tried to make it hotter it would just turn off and cool down. There just isn't enough fuel in the core to do anything else.
> The reactors cannot be deviated for nefarious purposes.
I'm no nuclear expert but a quick search on Thorium reactors brings up some controversy over its potential for weaponization:
Thorium, when being irradiated for use in reactors, will make uranium-232, which is very dangerous due to the gamma rays it emits. This irradiation process may be able to be altered slightly by removing protactinium-233. The irradiation would then make uranium-233 in lieu of uranium-232, which can be used in nuclear weapons to make thorium into a dual purpose fuel. [1]
[1] https://en.wikipedia.org/wiki/Thorium-based_nuclear_power#Po... which cites http://www.popularmechanics.com/science/energy/a11907/is-the... which cites Nature
For the rest, a great source is the book Plentiful Energy, by the chief scientists of another small fast-reactor project at Argonne. For that reactor, the fuel is a mix of plutonium isotopes which can't be used for bombs and are much more difficult to purify than natural uranium ore. The waste goes back to the radioactivity of the original ore in a couple centuries.
http://www.amazon.com/Plentiful-Energy-technology-scientific...
Does anyone know the current story? Can renewables scale up fast enough? Also, does the availability problem (i.e., renewables not being available when the sun/wind are not) prevent them from having a sufficient impact? I could imagine that, even if renewables weren't always available, their use still could reduce greenhouse gas emissions enough to mitigate climate change sufficiently.
Availability is a problem. Right now we're mainly backing up renewables with fossil plants.
Most people really aren't getting what climate scientists are saying these days, which is that we have to cut emissions drastically in the very near future to avoid disaster. If we exceed +2C, or possibly even +1.5C, positive feedbacks will take the planet several degrees further even with no more emissions from us. Right now we're at +0.8C. Every ton of CO2 we emit takes 30 years to have its full effect on the temperature, considering direct effects alone, so we've got another 30 years of warming locked in already.
+4C or so might not sound like much but judging by geological history, the effects would include an enormous reduction in the amount of food we're able to produce.
Batteries have not had the step change in cost vs deliverable, and the question of effective and not carbon intensive recycling of batteries and most e waste remains. If solar is x cents a kWh, must add cost of storage or fossil backup generation to make sense in both terms of cost as well as carbon.
For instance, the expected cost per kWh of the Powerwall is about 35c/kWh which must be added to cost of generation and the maximum amount of time for delivering that power back at peak discharge is only a few hours. Leaves some development still to do for power in the morning after the night, or even for a cloudy day. Solar thermal has run into similar reliability problems as solar photovoltaic.
Not sure what 'sama means by this, but I guess it's what I feel - government projects tend to go slow unless there's an actual, real security reason for them to go faster, in which case - like with Manhattan project - you get crazy amount of productivity and progress.
Deploying a bunch of black solar panels will also increase the Earth's energy budget, by absorbing more sunlight. But that's another insignificant effect.
Ironically enough, a donkey is 1/3 horsepower, which is 748 watts of power. A donkey at about 250 watts is therefore is roughly equivalent to a solar panel the same size. The donkey has the advantage over solar, in that it's dispatchable, while solar is not.
I wish we considered donkeys instead of Solar as a replacement for fossil fules, since the shortcomings might be more obvious to the innumerate.Nuclear energy is inherently centralized and difficult to decentralize. This creates all sorts of political and economic dynamics, some of which you (Sam, and YC) may benefit from, but some of which may be damaging to societies in various ways (think corruption, control, monopolies, etc.)
Obviously this isn't necessarily true for all possible as-yet-unimagined implementations of nuclear technology. But it's something to think about when comparing energy technologies.
Solar, on the other hand, while not necessarily inherently decentralized, is extremely decentralizable, leading to very different dynamics.
I'm not saying Nuclear is bad. I'm just saying this stuff should be factored in.
I don't see any future for nuclear if it doesn't fundamentally change the way it harvests the energy and when it solves the nuclear waste problem economically
Neither of the ycombinator companies produces long-term waste.
That's what you can't do with solar - with solar you already have a big footprint to power that first factory, and your footprint increases proportional to power use. 21 century calls for scaling roughly 20x = 2x (population growth) * 10x (rise in developing worlds livining standards). I don't want 20x solar footprint.
I couldn't agree more with Sam about the importance of energy for our civilization. Kudos to him for putting his efforts towards important stuff.
I have mixed, but mostly positive feelings about venture capital and energy startups. The fact is, it's a tough space. Large capital requirements, prototyping cycles often measured in years, and a low success rate. Everyone is still waiting for the energy unicorn to put Google, Uber, Yahoo, et al. to shame. And energy startups don't benefit from many of the things in SV that infotech startups do, such as ecosystem synergies and being co-located with all the new cool stuff in your industry. This is especially true with regards to one of SV's great strengths, the freedom to fail.
Where SV shines is in the short times from idea to testing. In most of the nuclear energy industry, going from idea to tested prototype can take decades. I think we all know the importance of short debugging and feedback cycles. Hirsch harped on this a few years back, and it's still a good point. Look at ITER, which were were talking about back in 1995. ITERative, it is not.
Some observations:
1) The teams and funding are a bit larger than they used to be. This is probably a good thing. The design turnaround time is a bit better, but not by much though. It's necessary to tweak a design once you have built it to learn from it and see what its ultimate performance can be. But it's all-too-easy to spend a year or two doing that. Do that a few times and then you're out (of money, time, your mind, what-have-you).
2) Location. There is no advantage to locating an energy company in SV except for proximity to funding (and Stanford, I suppose). We located by the NHMFL in Tallahassee. It's cheaper, and the magnet guys would moonlight for us. However, working with Tim over 3 time-zones had its challenges. I don't think we got the benefit of having a great VC as much as some of his other portfolio companies did (no complaints about him, just the distance). Some things are just hard to explain over the phone. But SV still isn't the right place. I think that there is a big opportunity for VCs to improve how they provide the value-added stuff that they do (beyond providing money) remotely, and energy is the space that needs it most. I don't know the VC job well enough to provide good suggestions, I just know there is an unmet need here.
3) Because the failure rate for startups is so high, it's important to have a decent failure path for the people involved. For software devs, SV jobs often provide a soft landing. Energy guys don't have that easily transferable skill set. So, fusion largely consists of old hands who are willing to spend 20 years ramming a single design through, and a bunch of young redshirts who are sure that they can beat the odds. When every design failure becomes a career failure, people aren't incentived to radically iterate designs quickly. Luckily for me, I learned radiation measurement and protection on-the-job (hey we have neutrons! How many neutrons? Woo hoo! Wait, oh shit!) so that skill transfered over into medical physics quite readily. But imagine what SV would look like if almost every software startup founder who failed once had left the software industry.
I wish good luck to Helion, UPower, and all the other teams fighting the good fight.
Lithium-ion batteries are still at least $100/kWh (probably more like $300/kWh) and assuming a useful life of 1000 charge/discharge cycles, you're talking between $0.10 and $0.30/kWh just for the STORAGE! That doesn't even take into account the cost of generating the energy to begin with, nor the losses that come with charging and discharging the battery, nor the capital cost on the inverter that absolutely isn't free and definitely doesn't last forever.
Until storage can be had for a few cents per kWh storage is an unsolved problem.
I suspect that it will continue to be an unsolved problem for quite some time. Not because it's impossibly hard, but because getting oil or gas or coal out of the ground is so easy and has such large energy gain (output energy / input energy) that you have to be very clever to beat it.
Electric power can probably be used to synthesize traditional fuels from carbon dioxide and water, to stay carbon-neutral. Plants can do that but in a quite inefficient manner. A (bio)chemistry breakthrough is badly wanted on this front.
Fusion: By '24/7' you mean the 0.5 s sustained burn at JET in the 90s. Humanity has not yet managed to sustain fusion for a full second on the Earth, let alone get any useful power out of it. ITER (the next generation Tokamak test reactor) isn't built yet. The intent is for it to achieve 1000 s of fusion burn time. Its over budget and behind time, and they've just had a management reshuffle to try and deal with this.
* Edited & extended to try and incorporate nbouscal's feedback from below
[1]: https://en.wikipedia.org/wiki/Illusion_of_transparency
Edit: +1, much clearer what you mean now.
The Helion design which Ycombinator funds is pulsed, so sustained burn time isn't applicable. It's also much smaller and cheaper than mainstream tokamaks.
The problem is that the sources like solar have both power density and availability problems in higher latitudes, and power density required to e.g. power a major factory may be a problem even in a sunny place.
I'm also not sure about price per kW of installed generating power; it looks like solar power is still way behind atomic power in this regard.
In general, everyone would love breakthroughs both in electricity generation, transmission, and accumulation. Looking at the amounts of coal still mined for burning, generation should not be overlooked.
(Of course it's far better for the environment than present coal/petrol/gas/wood, and their energy initially came from the sun anyway, but it will still have some environmental impact.)
Wind and hydroelectric, on the other hand, could very well have significant environmental impacts if deployed on a sufficiently-large scale, despite being technically "clean" energy (aside from the non-clean energy and materials used in their construction).
Further, it's all (99.99%) going to get released so net impact is going to be very local.
PS: In the end the real change would simply be changing the albino of the planet slightly. But, buildings and roads already cover far more land than solar is expected to anytime soon with minimal impact.
I can't find a source at the moment, my apologies. But suffice to say humans have a knack for considering their needs insignificant when in reality our Spaceship Earth looks ever more fragile the more we learn about it.
I'm a supporter of solar, and trying to get it on my house.
But I also did some back of the envelope calculations that showed, just if we had enough Powerwalls to backup US peak demand for one hour it would require 10x the global annual mining production of lithium. And that's just one hour. And that doesn't include the electricity production.
It's generally estimated that US power, with good transmission, would require enough solar panels to cover the entire state of Massachusetts. I think you're right that it isn't the land cover that would have much effect, after all buildings and roads already cover a lot of land. I think it's sheer material production.
Mining is almost entirely powered by fossils, it has to be. And so is most transport. And so is recycling of metals. So the energy density of an energy source really is a zero sum game. If it takes a millionth the material for one source versus the other, that adds up.
Then in maintenance, solar farms are truly "farms"- they require a lot of water to wash away dust to operate optimally. A states' worth of water is significant.
Then in recycling at end of life, and this is why I got so excited about nuclear as a somewhat hippie child growing up around oil companies in Oklahoma, solar is going to require a lot of energy to recycle, while nuclear can produce energy in recycling its fuel.
The main reality check, to me, is: what is the energy density of this energy, and if emitting, how much pollution? Coal is far more energy dense than wind, which is why humans evolved from windmills and wood to coal. But it's so polluting which is why we are all working towards better sources.
The US currently uses 40% of the 84 million acres or 340,000km2 crop for ethanol production which is arguably pointless. So, if we swapped just corn ethanol for solar cells we would have 6 times more land area than the entire state of Massachusetts covered in solar cells (135,000km2). And if we feel ethanol is really necessary we only need 8% of the total corn crop leaving 32% for corn ethanol.
If this still seems like a huge issue the soybean crop is far larger. http://www.epa.gov/oecaagct/ag101/cropmajor.html
PS: As to storage, we don't need 100% solar wind and hydro can make a huge difference. Pumped storage is also far cheaper at scale; it's just impractical when scaled down to home use.
There is a huge infrastructure cost involved in upgrading, building new or knocking down and rebuilding non-passive solar structures. Now, giving tax breaks to new construction that is passive would be great, but it wont put a dent in the energy consumption issue.
I say I agree with you, because I think that saying "cheap energy is the future" would have been 100% as valid a statement in 1965 as it is in 2015. Fifty years and a HUGE series of energy improvements later and we're still chasing the cheap energy dream. Guess what? Oil is cheap. And relative to cutting down wood to heat your house with a fireplace (a la most of human history) current solar is absurdly cheap.
What we need, and Musk knows this better than anyone, is better batteries to store energy when we're not using it more efficiently. Batteries will revolutionize the world, and Sam would do well to pay attention to what Musk is doing.
Wave power is even worse.
Geothermal from high-flux sites does work and works well, though its limited to a relatively small number of regions worldwide. Many of those have already developed much of the available potential. Iceland, Indinesia, Japan, Kenya, New Zealand, the Philippines, and the United States most notably. Kenya's Rift Valley is probably the biggest greenfield opportunity with significant impact opportunity (Kenya's existing electrical infrastructure is minuscule). In the US, resources generally exclude both the Yellowstone Caldera and Cascade range.
Large claims with no hard data comparisons? (eg. atomic power has major cost, density, and predictability advantages)
This(?):
>There will only be one cheapest source of energy
Really? There can't be two sources at or near equilibrium?
These systems are not deployed in isolated, designer environments, but instead are deployed in complex environments. Transportation and project logistics will prefer some sources over others.
The lack of any real metric for cost or "cheapness" is a red flag. Is cost being measured in nominal dollars? Will such a thing even exist in the 22nd century?
Cost: http://www.eia.gov/forecasts/aeo/electricity_generation.cfm -- on an LCOE basis, Advanced Nuclear is cheaper than everything but natural gas and that's entirely due to CapEx costs to build new plants -- which are what the modular nuclear companies that Sam's joining are working on fixing.
Density: https://en.wikipedia.org/wiki/Energy_density#Energy_densitie...
Predictability: Can be assessed by the capacity factor of plants that are running, the EIA publishes monthly data; http://www.eia.gov/todayinenergy/detail.cfm?id=14611
This seems to at least be partially taken into account in the "Total LCOE" numbers (sorry, but this report doesn't seem to have much as far as exactly how these numbers are calculated). Without subsidies, Solar PV is about 30% higher. However, when you look more closely at the actual ranges behind the averages that are the primarily cited data, you'll find that the Total LCOE estimate ranges for Advanced Nuclear and Solar PV actually overlap. The "minimum" Solar PV estimate is 97.8 and the Advanced Nuclear "maximum" is 101.0.
The claim that atomic energy has major advantages over solar is not borne out in this EIA report. In fact, the report seems to show that solar is highly competitive with almost every generation source and actually has a huge advantage in ongoing maintenance costs, being practically at the bottom.
I haven't investigated the other links yet, since the first one didn't seem to pan out.
Above I illustrated my calculations on just one hour of battery backup for US grid with lithium ion- would require 10x annual global production of lithium. Generally cost of battery backup looks like 35c/kWh and has limitations on duration so it must come back to fossils.
Solar makes a lot of sense if backup/storage/smoothing weren't needed since pv has gone down in cost. It still makes a lot of sense in many off grid applications, and for instance in my home we have solar concentrator lights and I would love to have solar water pre-heaters in the summer. But for on grid the capital requirements are much more complicated than generally illustrated, and ultimately the environmental impacts are greater than apparent at first glance.
For short term very high flux power conditioning they've got uses. For long-term storage not so much.
Seconds to minutes is likely their effective range. Not hours to days.
Solar has the advantage of trivial decommissioning, but a huge disadvantage in requiring things like storage, overproduction, long distance transmission, and smart grids, all of which raise its effective cost if you're not putting it on top of a fossil or nuclear-based power grid.
Individual solar panels and batteries, personal windmills, personal reactors, etc.
Imagine all the savings in infrastructure for energy transportation and reinvestment in other sectors. Imagine all the possibilities if people could switch their energy generation model as simple as buying a new product and installing it at home.
Individual energy independence, even if it will never be possible, that's where our dreams should be.
Then water, then food. That's disruption at seismic level. Post-scarcity world.
Solar has the advantage that it cannot be weaponized even in nonsense theory, greatly relaxing export regulation. This makes it a viable solution for developing countries and remote areas, not just the handful of nations with the wealth and sophistication to manage a reactor.
This isn't true because decentralization is easier, it's true because technologies which this isn't true of (require large capital expenditures or high degrees of expertise) can't be given to everyone and their mother.
By saying we're doing decentralized X we're moving the problem of requisite operating skill to the designers who must make their systems simple enough for laypeople. Likewise factories and engineers must find economies of scale in production of units which are distinct from the efficiencies we currently realize building relative few, much larger generators.
In short, while decentralization solves some of these problems, it also presupposes the solution of others. It may be a worthy goal, but there are challenges to get there.
Private moneys would compete for a slice of the pie bringing to market millions of ideas putting the pedal of progress to the metal.
Humanity needs to find a way to deal with high-energy technology if we're to move forward as a species.
The decentralization of production usually leads to a greater cost per unit due to economies of scale. This may not be true for solar, but it certainly is for nuclear, hydro, geothermal and probably wind. There are regions in our planet that don't get many hours of sunlight, how should they produce energy there?
Your post-scarcity world will surely not include Spain.
and then, as you indicated, with enough clean power, individuals can create desalinated water or even extract water from air. They could cleanly power greenhouses even in the arctic. We could remove carbon from the air.
People will install solar panels and enjoy not writing a check to the utility every month. But if their home battery breaks during a heat wave, they're not going to sit in an un-air-conditioned house and and think, "well at least I'm independent." No, they're going to expect the grid to send them power when they need it.
So while independent generation is a good idea, it's not likely to result in infrastructure savings. In fact the kind of smart grid that would be needed to deal with such widely varying local loads would almost certainly be more expensive than just maintaining what we've got.
It's worth remembering that electric generation started out as a very local and independent thing. Central generation won because it was cheaper and more reliable--despite the seemingly obvious losses and expenses of such a huge network.
By that time, I bet batteries would sell for less than a thousand, much less.
Tons of money and effort has been sunk in attempts to create more sources for energy and some of those have paid off and others are still under review. The 'fossil fuel hegemony' is actually quite a large investor in more than a few projects, their product is energy, not a particular kind of energy.
http://www.aps.org/publications/apsnews/200904/physicshistor...
https://en.wikipedia.org/wiki/Hafnium_controversy
(I'm sure you think muon-catalyzed fusion is right around the corner too)
Energy is a tiny component of modern civilization.
We are Hydrocarbon Man.
Yes, abundant energy is behind virtually all past development. What seems to be holding back the next 5 billion, however, is access to what that energy has made available.
Though the question of just how many billions can be provided for ought also be raised.