Nuclear energy is a disruptively cheap and simple way to boil water, one of humanity's most important processes

16 Comments

1. Rod. Thanks for this post, which caused a surge of nostalgia!
   I still treasure the copy of “Steam” which I received from B and W, back in the day. B and W gave complementary copies of this book to EPRI project managers in coal and nuclear. (I was in nuclear.) It is not only a useful and beautiful book, but it was an introduction to issues outside of my own view of the world. Also, it made me feel proud to be part of this huge human endeavor called “raising steam.”
   Nice that they are new facilities in Maryland too. I visited them in Ohio, where they had a really top-notch research facility. They probably still have it. I think it would have been hard to move some of the really big equipment.

   1. Speaking of nostalgia, I was recently looking at the diagrams of some of the early nuclear plants, as published in Nuclear Engineering magazine. One of my favorites is Indian Point-1, linked to here:
      http://econtent.unm.edu/cdm4/item_viewer.php?CISOROOT=/nuceng&CISOPTR=29&CISOBOX=1&REC=19
      Click on Access this Item for a pdf image of the plant. IP-1 was an early B&W plant with a few interesting features, which included an oil-fired superheater for the main steam. The best part, the first core was a combination of U-235 and thorium! Too bad they were so quick to give up on thorium. We could have made so much progress by now if they had kept up designing seed and blanket cores. But that often means fuel reprocessing, which went out of favor in the 1970s.
      More diagrams can be found here.
      http://econtent.unm.edu/cdm4/browse.php?CISOROOT=/nuceng
      It is interesting to see how truly strange some of the early designs were, and to see how the industry evolved over time.

2. @Rod,
   I’ve been meaning to ask you a question about that 2.03 cents/kWh figure you’ve been using in a couple articles recently, including this one.
   You say that includes the cost of fuel, operations, and maintenance. Am I correct to assume that does not include the capital costs of building the plants, because most of those plants are now payed off?
   What is the average cost/kWh over the entire life of the plants, when you include all the costs of initially designing, licensing, and building the plants? It seems a bit disingenuous to me, to ignore those capital costs when discussing the cost of electricity from nuclear, since various sources I’ve seen indicate that those capital costs are the majority of the cost of nuclear power (something like 90 or 95%, I think)?

   1. Well, the 2.03 cents/kWh figure has one advantage: it is fairly well established, well known, and stable. Compare that to the (marginal) cost of operating a natural gas plant, which fluctuates wildly depending on the price of natural gas, since the fuel is the majority of the cost to run the plant.
      The levelized costs — i.e., average costs of producing electricity including capital, finance, fuel and operation over a plant’s lifetime (including decommissioning and waste disposal) — are more difficult to get a handle on, because the calculations depend heavily on the assumptions put into them. That is, fairly small changes in the assumed overnight costs, interest rate, and whether the grid infrastructure for the plant are included in the calculations can greatly influence the final value. Thus, you’ll find that there are some disreputable analysts who try to “game” the calculations, usually to show that nuclear is too expensive to be economical.
      I have seen many figures produced by various groups, and my best guess would be that the levelized cost of a plant that could be built today would come somewhere in the 4 to 5 cents/kWh range.

   2. The MIT study on the “Future of Nuclear Power” has revised the number to 8.4 cents per kWh. I’m thinking this is a pretty independent source

      1. I would view the MIT study as probably representing a reasonably credible source, who would not use any wildly bogus assumptions. So, at 8.4 cents/kWh, Nuclear, while still reasonably priced, is a lot more expensive than the 2.03 c/kWh being thrown out. I don’t think it does any service to the public debate to use numbers which can be easily argued against as not being really representative of the whole truth.
         Still, at 8.4 cents, the nuclear industry would probably be selling power at retail at a cost of around, what, maybe 12 cents? 12 cents is a little on the expensive side (right now), but by the time those plants are built and in operation, 12 cents might look like a bargain. I think the only other energy source which can really compete is conventional, carbon-belching coal, and if we really want to get off coal, we have to take that ‘cheapest option’ off the table.
         Nuclear definitely appears to be the next best deal after coal. Does that MIT future of nuclear power study assume any ‘economies of scale’ reducing the capital costs, or does it strictly base it’s projections on current cost estimates? I would think it’s the latter?
         The reason I ask is that, if we can get some nuclear regulatory reform, along with economies of scale (which Rod has talked about in past posts), that 8.4 cent figure could come down somewhat. If Nuclear costs could get down to around 6 c/kWh, so that they could sell at, say, 8 or 9 cents, it would be, I think, very competitive even with coal.

         1. “I would view the MIT study as probably representing a reasonably credible source, who would not use any wildly bogus assumptions. So, at 8.4 cents/kWh, Nuclear, while still reasonably priced, is a lot more expensive than the 2.03 c/kWh being thrown out …”
            Well, the MIT study is certainly one data point. I would caution against reading too much into their estimate, however. After all, this number is just the output of a model that was used to compare the levelized costs of nuclear to those of coal and gas.
            This model assumes an economic lifetime of nuclear plants of 40 years (or as little as 25 years) and an average lifetime capacity of 85% (or as low as 75%). Considering that many of today’s operating nuclear plants in the US have had their licenses extended to operate for 60 years (and all new reactor models being sold today have a design lifetime of 60+ years), part of the expected lifetime of the plant has been neglected. So after the first so many years at the levelized cost quoted above, the plant almost certainly will continue to run for an additional 20 years, 35 years, or more at the much lower cost of 2.03 cents/kWh.
            Also, as the 2009 update mentions, the average capacity factor for the US fleet of nuclear plants is now consistently above 90%, which indicates that the cost output by the model is likely an overestimate. Finally, the overnight cost of $4000/kW assumed in the update seems to be conservatively on the high side. These costs will vary quite a bit depending on the owner’s costs, which are largely site specific. For example, it should be cheaper to build a new reactor on an established site than on a greenfield site.
            “Does that MIT future of nuclear power study assume any ‘economies of scale’ reducing the capital costs, or does it strictly base it’s projections on current cost estimates?”
            The base case takes a pessimistic view and does not assume any reductions in capital costs. In the original 2003 study, however, several scenarios were also run to look at the effects of such improvements. Reducing the construction cost reduces the levelized cost of electricity from 6.7 cents/kWh to 5.5 cents/kWh. Reducing the construction time by 20% as well reduced the cost to 5.3 cents/kWh.

         2. You think 12 cents per kWh is expensive? In places like Hawaii where most of their electricity comes from burning oil, they are paying around 40 cents per kWh. It’s not uncommon for households there to have bills of $500+ a month. Hawaii has unfortunately banned nuclear power.

            1. @Jason: Well, compared to current coal electricity prices NOT COUNTING the ‘hidden costs’ you mentioned. In other words, the “price people see on their bills”. Here in Ohio, I’m paying something like 8 cents for generation (and statistically, most of Ohio’s generation is from Coal, and in *particular* my electricity is bought from Duke Energy which, although they have nukes in the Carolinas, doesn’t have any nuclear in Ohio, Indiana, or Ky, so I believe my power, and the cost of it, is driven by the economics of coal), and 5 cents for transmission. I can’t avoid paying the 5 cents transmission fee, regardless of the source of the power. The 12 cents for nuclear is more expensive than the current coal prices. However, coal prices have some volatility – it may be that a few years from now, coal costs could be 15 or 20 cents an hour, particularly if the Dems are ever successful in passing a carbon tax or cap and trade (probably unlikely).

               1. @ Jeff,
                  Most of the cost of a nuclear plant is in the capital costs, which ANY new coal plant has to share. So, when we are comparing plants – please remember that we are comparing the cost of electricity from both fully amortized coal and nuclear power plants. Thus the 2.03 cents IS a comparable cost.
                  Duke in Indiana is paying 2.8 billion or so (costs keep rising) for a coal plant near Edwardsport with a capacity of about 640MW. So the capital costs are about the same for a new build. What remains is the operating and fuel costs. These are quite competitive for Nuclear.

      2. El, you are omitting other hidden costs that affect the price we all pay for fossil fuels- the pollution, health detriments, military placements to secure supply lines, the ash ponds, oil spills, an occasional neighborhood blows up, trade deficits, messy political entanglements, mountain top removal – just to name a few. If you want total levelized costs, then you ought to compare all those factors. Once those are factored in, nuclear still looks pretty darn good even without cap-and-trade.

         1. The NRC requires nuclear power plant owners to have the funds to decommission the plants to green field status once they are shut down. This is included in the cost. Is there any similar requirement for all of the fossil fuel power plants, including the fly ash heaps, pipelines into the plants, etc? I’ve never heard of anything like that being required.

   3. @Jeff – I would be happy to begin including capital costs if the competition would begin including the costs that they like to gloss over or ignore altogether. When I do start including capital costs, however, I claim the right to include the very low costs of several million per unit that were achieved for about half of the US fleet.
      The MIT final numbers were produced by people who were competent spreadsheet manipulators. The key to deciding whether or not to accept their numbers is to recreate their model and understand the ASSUMPTIONS that they made. They include such assumptions as the ones that Brian lays out and they also include a significantly higher interest rate for nuclear than for the competitors.
      Levelized cost projections are no better than guesses; they can be manipulated at will because they are all based on the input numbers and those are simply guesses as well.

      1. I think 8.4 cents is pretty darn good price! Wiki has a helpful summary of seven studies that compare leveled costs for coal, natural gas, and nuclear. MIT does appear to be in the high range. Only one study suggests nuclear and gas are more competitive than coal (Royal Academy of Engineering, 2004). I agree with many comments here that this is probably pretty close to the true cost for nuclear (as opposed to the subsidized, off-book, and manipulated low cost of conventional sources like coal and NG). But the way assumptions work, if you apply the same assumptions in each of the cases you should get comparable numbers (and there will always be a margin or error or a range). And what you want to avoid is coming up with your own faulty, partial, and manipulated low number (but rather focus on real costs). There are some pretty talented visionaries in the research departments at MIT who do far more than manipulate spread sheets (and have a hand in this kind of report

3. Phrases like “nuclear energy is essentially a very complicated way to boil water,

   1. To be fair, nuclear energy *is* a slightly complicated (I don’t know about very) way to boil water. What people don’t realise is that being able to boil water is so absolutely critical to modern life.

Comments are closed.