5 ms·
Of course intermittency, and time matching electrical load to generation. Fortunately we have great technology for that: https://www.greentechmedia.com/articles
by epistasis 7y ago
Of course intermittency, and time matching electrical load to generation. Fortunately we have great technology for that: https://www.greentechmedia.com/articles/read/pges-recording-breaking-battery-proposal-wins-loses https://www.greentechmedia.com/articles/read/pges-recording-...
Battery technology, be it lithium ion or one of the numerous flow chemistries, is on a continuously decreasing cost curve, while nuclear tech is on a continuously increasing cost curve, or at best stagnant.
Additionally, with nuclear, we are locking in todays prices for 60 years, without taking delivery for 10-15 years. Investing in storage tech with a 15-20 life span allows reinvesting in a cheaper tech after that.
Additionally, seasonal storage technologies, outside of the currently existing hydro dams, are likely to enter the stage, such as synthetic fuel.
And since a huge amount of our emissions come from sources that can't yet be electrified easily, we really we will likely need some sort of synthetic fuel strategy, which would also turn into a seasonal storage tech.
Electrical generation is just one part of the climate solution, and honestly renewables plus storage plus HVDC is, today, a 95% solution. Getting that last 5% also solves the other parts of the climate problem. We need to be looking beyond electrical grids to other regions of our economy, and we need to be doing it five years ago.
If the nuclear renaissance of 2008 had delivered a desirable product, it would be deployed in far more places. Instead, it was late, and also way too expensive. Nuclear may come back in some form, such as SMR, but as it is, it's neither necessary nor desirable to tackle climate change.
- nabla9 7y agoNone of the solutions you propose scale cost effectively as replacement of nuclear power. We can't even recycle lithium for batteries cost effectively yet.
- epistasis 7y agoI think that they're the only options that have a chance at scaling. We know how to build battery factories reliably, and we continually make them cheaper. We don't know how to build nuclear, much less on the grand scale that would be needed, probably 50-100 reactors for the US alone.
- nabla9 7y agoThe US already has 98 reactors. Gradually adding 50 more while wind (6.6%) and solar (1.5%) catch up would eventually provide 30% of US electricity from nuclear power. The optimal ratio depends on how alternative energy storage tech advances.
- epistasis 7y ago50 reactors is 50 bets, each at a cost of $5-$15B. And given our success rate, it's actually probably 75-100 bets if we want to end up with 50 reactors, each with a massive (25%-50%) chance of complete loss even before generating electricity. Even if you assume unchanging technology costs, it's unlikely that a successful build will recoup its costs. Once you account for advancing technology, it's virtually guaranteed that each of these nuclear reactors would never ever be able to compete on cost. So the upside to the bet is practically negligible. So clearly no rational economic actor is going to undertake a build on their own, after what we've learned from Vogtle and Summer. We wouldn't be able to find a contractor or a utility or anybody that wants to take on that risk. So what's left is a massive government subsidy, of somewhere between $500B - $1T to build reactors. What if, instead of government actors deploying $10B-$20B per year to build reactors, we spend half on well-proven technology like lithium-ion grid storage that the government owns and operates at cost, and spent the rest to buy the early products of industries that are developing, including nuclear SMR, in order to give them the chance to advance their learning curves and become more cost effective? Nuclear technology, as it exists today, is dead in the water. Even if we eliminate all regulation and all local NIMBY opposition, we are left with the problem of finding merchants willing to construct it. This is a serious problem for the technology, and it's innately tied to the technology itself, the engineering required to build these massive reactors, and the tight construction tolerances that are required to make them work. We need a nuclear tech revolution before it can compete with storage and renewables. Solar and storage are ultimately thin film technologies, which works great with our current manufacturing capabilities. Wind is lots of identical, smallish generators. Nuclear does not fit at all with our current construction capabilities or engineering practices and capabilities. Until it can be brought into the modern world (possibly through SMR), it is not a good technology.