I'm betting that nuclear fission will disprove Malthus once again 1

84 Comments

  1. What a disappointment. It appears Sami is yet another victim of the default-negative, default-no knowledge attitude to nuclear power. He fails to think critically and do research, which would quickly show that half the arguments against nuclear power that are put forward are simply wrong – re fuel supply, nonexistant waste problem EROEI, take too long to build (see France) – and the other half aren’t cut in stone (policy environment – we can change this, we have changed it before many times).

    The fact that Sami puts forward wind and solar as a solution is terribly predictably falling into character for those that have drunk the anti-research, anti-science Kool-Aid.

    I am so tired of reading these articles and books from people that claimed to have researched nuclear power, and then an amateur like me can poke giant holes in all their anti-arguments.

    1. Cyril, I’m not quite sure where to start responding to that, or even if I should. Clearly you have either not read my post or misinterpreted it in some fashion for you to claim I advocate wind and solar as the only solutions, am anti-science or anti-research. There is no need for that kind of polarization – nowhere, I believe, did I say I was “anti-nuclear”, a camp where you clearly put me in.

      And absolutely many things aren’t carved in stone. I never claimed they were. I’m simply saying that I do not believe, for a variety of reasons, that nuclear will end up accounting for the said portions of the global primary energy and electricity production in 20 years time.

      Nothing more than that, nothing less than that. Please don’t read into the prediction things that aren’t in it.

      1. Sami, there is no doubt that the business-as-usual scenario will come up short, on both nuclear and renewables. The business-as-usual is lots more growth, primarily via using more fossil fuels (even with all the energy efficiency we can muster).

        The business-as-usual scenario is not a low energy society. It is not powering down. We will in fact burn the furniture before we power down.

        That is your false bifurcation: you assume business as usual on nuclear, for reasons that mostly aren’t carved in stone, or are flat out wrong (uranium resource use, EROEI, etc, which you fail to adress and have clearly not researched well at all and likely have just taken the Storm and Smith poison). Yet you assume we can get our hands together to use less energy.

        That is simply being dishonest about your assumptions. Powering down will be so much harder to sell than even a nuclear powered world.

        We will perhaps have to agree to disagree on energy trajectory philosophy, but my criticism of your list of nuclear non-problems still stands.

        1. We indeed may need to agree to disagree.

          I however have to wonder how you can continue to read so much into my post / prediction that just isn’t there. For example, I do not assume business-as-usual, nor do I assume “business-as-usual” is a low-energy society. I did not claim renewables will be the silver bullet. I don’t assume we will somehow willingly use less energy.

          And I am not saying powering down will be an “easy sell” – what I am saying that we will be forced to do it (as energy becomes increasingly expensive) – and even that I am saying here, not in my post.

          Most importantly, perhaps, you are somehow assuming – simply because I don’t share your “nuclear optimism”, maybe, I don’t know – that I am “anti-nuclear”. Let me reiterate that I am not anti-nuclear. We should be on the same side – if sides must be taken – which is why I find the hostility odd. Or are differences of opinion not allowed in your camp?

          Because I’m fine with disagreement. I’m even fine with you saying all the grounds for my prediction are bullshit, that’s your right. There are multiple sources for facts and “facts”, and mine may be wrong in some – and so might yours.

          I’m predicting – predicting, as in making a prediction about the future, which is inherently uncertain, whereas you seem to “know” I will be wrong, which I find a rather curious position.

          So feel free to disagree with me as much as you want, but don’t assume to know for certainty what will happen over the next 20 years.

          1. I don’t know what will happen in the next 20 years. Most likely not enough on the side of renewables and nuclear, and too much on the side of fossil fuels. That is a likely business-as-usual case, and I think we agree on this.

            Differences of opinion are fine with me. We are all entitled to our own opinions, but not to our own facts. You bring up a list of “cons” against nuclear, and half of them are flat out wrong.

            I have explained you this 3x in this thread. You have not given me a single source that makes it clear that nuclear has a low EROEI (as just one example of where you are wrong), and you have not made a single attempt to convince me with any technical argument on your side. You simply leave me with a list of claims about nuclear which I know from my own reading are largely incorrect.

            No offense but I find such a writing style hollow and weak. If you want to write with authority on matters of energy, you may want to refrain from cooling water and EROEI claims against nuclear. They simply don’t exist (in fact power plant cooling is only a few percent of global freshwater consumption), and merely by putting them forward you are giving everyone the impression that you have the default-negative default-no-knowledge about nuclear power.

            Keep in mind this is a different subject that predicting what will happen. This is not about predictions, it is about getting your scientific facts in order.

            1. @Cyril R

              I don’t think you understand. Sami has not claimed to be an expert on energy and he has made his statement based on information he has gathered – perhaps casually – from sources that he generally trusts. His prediction accurately reflects that existing position.

              My bet was designed to start a conversation with someone that find interesting in hopes of sharing the view from my different perspective. This is not a fight; Sami has given no indication that he is actively trying to stop nuclear energy progress. There is no valid or useful reason for trying to turn it into a fight or even a judged debate. It is simply a conversation starter.

  2. I don’t know how this relates to your 20-year bet, but I found this report from a conference in India to be extremely encouraging:

    India: A hotbed of molten salt
    By David LeBlanc
    http://www.the-weinberg-foundation.org/2013/01/18/india-a-hotbed-of-molten-salt-2/

    LeBlanc describes strong interest in advanced designs among Indian scientists, and sessions that presented work underway around the world. LeBlanc also gave a presentation of his own, focusing on simplified designs that require as little new research as possible and are optimized for quick implementation.

    Here’s a video of LeBlanc introducing the concept (20 minutes):
    http://www.youtube.com/watch?v=370srr67Bnk

  3. With just the plants already under construction in China and India coming on line, how close will we get to your 25% prediction? What if they keep up their pace of new starts from the last 5 years (given China’s air pollution issues, I bet they are going to actually speed things up)?

  4. I won’t go too deep into details why here, but aspects like policy environments, long plant build lead times, limited skill base, waste fuel problems, susceptibility to climate change (primarily from water being used as a coolant), surprisingly low EROEI, high initial costs and energy expenditure as well as questions on the sufficiency of fuel supply all played a role in me coming to this conclusion.

    It is now clear to me, from both the article and Sami’s non-technical response to my critique, that the reason why Sami won’t go into technical details is that he doesn’t know them.

    I’m sorry Sami, but if you’re going to make flat-out lies like the water, waste problems, and EROEI, as arguments against nuclear power, it’s clear that you’re completely uninformed on these issues.

    And if you’re going to do this sort of standard laundry list of non-arguments that anit-nuclear propagandists have patented, then you should not be suprised if someone calls you unscientific, uninformed, and anti-nuclear.

    If you wish to write with authority on technical matters such as EROEI, you will have to go into details and you will have to do proper research.

    Putting forward the standard non-argument list of the anit-nukes isn’t furthering the debate, it is only embarrassing for you.

    1. @Cyril R

      I’ll have to jump in and defend my new friend, Sami. As he pointed out in his response to you, he is not opposed to the use of nuclear energy. As I mentioned in the initial post, he is actually hoping to lose the bet because he agrees that the world would be better off if we replaced more fossil fuel combustion with nuclear fission output.

      The issue that he has is that he just does not think it is going to happen that way. Many of my nuclear industry colleagues agree with his assessment – they believe that our regulatory system (here, in Europe and in most of Asia) is going to slow down progress enough to make my predictions wildly optimistic.

      My view is that the game can be changed, but that we need the help of everyone who really wants that change to occur. We cannot allow the hydrocarbon Establishment to divide and conquer us.

      Sami has the potential to be another strong advocate; he is an innovation professional in real life. He just needs to have some time and some good resources that will enable a journey of discovery about what really is possible with nuclear fission so that he can compare the potential to the rather dismal progress made so far. Then perhaps he will have a chance to reflect on the causes of the vast gulf between nuclear potential and our current technological reality to see who erected the barriers to progress.

      My hope is that sometime soon, he will start asking how he can help to remove the barriers that slow deployment of a power source so amazingly concentrated that a tiny pellet of U (or Th or Pu) oxide contains as much energy as a pickup truck load of coal – even if consumed in a very inefficient manner.

      1. Rod there are two different things at play here. The prediction – what will or will not happen – and the facts about nuclear power (re EROEI, cooling water claims).

        If Sami clearly doesn’t have his basic facts right, I am not interested in his predictions, and even if I were he still has to get the facts right about nuclear power.

        Sami clearly does not have the comprehensive knowledge about nuclear power that is required to espouse a dogmatic opinion.

        1. Sami clearly does not have the comprehensive knowledge about nuclear power that is required to espouse a dogmatic opinion.

          And he has not expressed one. Please take his advice and read his post and comments more carefully.

          1. Uh Rod, you quoted Sami as follows:

            I won’t go too deep into details why here, but aspects like policy environments, long plant build lead times, limited skill base, waste fuel problems, susceptibility to climate change (primarily from water being used as a coolant), surprisingly low EROEI, high initial costs and energy expenditure as well as questions on the sufficiency of fuel supply all played a role in me coming to this conclusion.

            I was simply pointing out that a number of these were flat out incorrect statements. Thus Sami has expressed a dogmatic opinion, based on incorrect facts.

            That is all I am pointing out. I have read Sami’s post carefully. Have you?

            1. @Cyril R

              This is the post I am talking about – http://www.groundswell.fi/sim/2013/01/25/nuclear-will-not-save-the-day/

              I simply quoted one paragraph to whet your appetite for learning more.

              Here is another quote from his post:

              “Still other people argue that nuclear power will play a key role in the future energy landscape – partly because it’s relatively clean, cheap, abundant, efficient, safe – and predictable in a way wind and solar may not be.”

              And here is another:

              “Let me go on record to say that if the rise of nuclear comes at the cost of (i.e. replacing) oil, gas and most of all coal, I am all for it and I hope I will be wrong with this prediction. I have no doubt nuclear energy will play an important role in the energy mix going forward; I simply do not believe it will be feasible to have nuclear energy go up that significantly in that “short” timeframe of 20 years.”

              Now do you see why I am asking you to read more carefully before you accuse him of expressing a dogmatic opinion?

          2. Now do you see why I am asking you to read more carefully before you accuse him of expressing a dogmatic opinion?

            We appear to be talking past one another, Rod. I see your point, it is valid. My point, that Sami has put forward a number of incorrect statements regarding cooling water use problems, uranium availability problems, and low EROEI.

            It is an important point, because Sami bases much of his pessimism on nuclear on these assumptions.

            I am happy to provide references to where Sami is wrong, if you want. But it isn’t hard to Google things like EROEI of nuclear. So I wonder a bit about our friend Sami.

            1. @Cyril R

              It would be a lot better to gently suggest better sources. There is incorrect, negative information available about EROEI of nuclear that appears right near the top of any Goggle search on the topic.

              My point is that it is not effective to attack someone and claim they don’t know their facts when the area of discussion is as fraught with misunderstanding and purposeful misinformation (often masquerading as peer reviewed literature for those who are not specialists in the field) as nuclear energy.

  5. Just one more thing about that “short” timeframe of 20 years.

    France switched to 80% nuclear power in 15 years.

    http://www.iea.org/stats/graphresults.asp?COUNTRY_CODE=FR

    Nuclear was a tiny portion of the electric supply up until 1977. In 1987 nuclear provided about 70% already. That’s 10 years to 70%. About 15 years to 80%.

    No one predicted this during the oil crisis in the 70’s. Yet that crisis directly led to a policy change.

    Interestingly, coal use was reduced around 60%, but oil only around 20%. This gap is only recently being adressed by increased use of electric railways, elimination of oil based heating in industry and homes, and the coming of age of electric car and battery technology.

    1. France switched to 80% nuclear power in 15 years.

      That’s very interesting.  I wonder if the USA could make a similar shift in energy use in that time frame?  Displace most coal and natural gas from electric generation, add new generation to replace NG for space heating and cooking, then use the NG for motor fuel.  Greenhouse emissions and oil imports would both drop like a rock and air quality would improve.

      1. It is clearly technically entirely feasible; the US currently gets a greater percentage from nuclear power than France did before it started off on a serious nuclear build.

        In fact, France just copied light water reactor technology from the USA (among others) and got started.

        1. @Cyril R

          The US also has an underused asset – at least 100,000 former sailors who are nuclear trained. Many of them do not work in the industry; in fact, many of them purposely decided to take their skills elsewhere because they did not like working in an industry full of people willing to impose so many exceptional rules on itself, inhibiting its growth.

          1. “an industry full of people willing to impose so many exceptional rules on itself,”
            You should use some of your resources and talk to some of the ex navy officers that have helped magnify this problem.

            In the early 70’s after leaving the Navy, I was amazed at the freedom we had on the operation of the reactor. When I asked why we did not need to do this common navy “safety” requirement, the most common answer was “We are dry land and can simply “scram” it or shut it down – we are not in a submarine under 100’s of feet of water.” I can remember doing a restart of the plant after confirming that the dispatcher had made an incorrect switching action and having the plant back online before the shift was over. After TMI INPO came to save the day. The common knowledge was “Don’t bother applying for a position at INPO unless you are ex-navy.” They brought with them all of the Navy/Rickover safety requirements. That same situation today would require two to three days to commence the startup, and then take a day or more because of the plant heatup rate limitations. Today, plants NRC/INPO ratings are hurt just as much by “un-planned” trips as by real NRC “safety” violations. Makes no difference that an eagle flew into your substation – it’s your fault!

            A few posts back there was a comment about the “need” for these excessive QA/QC requirements on a NPP because they would be safer. I suggest that they should compare the requirements/benefits against the hazards of the piping in a supercritical and ultrasupercritical pulverized coal power plant (pressures exceeding 3000PSI). You would be amazed at the number of welders that could not pass the welding test at a NPP that had just left a welding job at one of these supercritical plants. And over the last 40 years I have never heard of a problem at a coal plant.

            The nuclear power industry (in the USA) is killing itself with safety.

            1. @Rich

              I cannot explain your experience. No submariner I know would think it is a safe course of action to remain shutdown for days after an explainable trip (scram). We only had one reactor, staying shutdown under water is not safe. We learned with Thresher that a “safe” reactor on bottom of ocean is a contradiction based on poor prioritization.

        2. @ Cyril R.

          Your statement:
          In fact, France just copied light water reactor technology from the USA (among others) and got started.

          My response:
          Indeed. But what a different choice of industrial model and economies of scale across the pond! I love senator Domenici’s line: The US has 2 kinds of cheese and hundreds of nuclear plant designs. France has 2 nuclear plant designs and hundred of different kinds of cheese.

          1. Haha, that’s a good one!

            The French actually have PWR and BWRs just like all commercial plants in the USA. But they have standardized the sizes for their program, one of the keys to their success. Basically, the program rests on large (900 MWe) and extra large (1300 MWe) plants. More recently the newer xxl classes of 1500 and 1600 MWe have been included.

          1. Nice find! The undermoderated heavy water versions are almost fast reactors. It means a higher conversion ratio (PWR/CANDU can’t remove neutron poison FPs online).

            CANDUs are already on a quite tight coolant/fuel rod pitch. If they didn’t have moderator in the calandria, they would be potential fast reactors!

            Losses to U234 should be quite similar for moderated cores. It is only when you have a really fast spectrum core that the U234 essentially becomes a fissile fuel. For the same reason it doesn’t respond to delayed neutrons, so if you have a lot of this stuff in a solid fuel fast reactor, control can be difficult.

            A lower power density core would make less U234 from Pa233 capture in the first place, though. Which means lower losses to it indirectly. But CANDUs have quite a high flux on the actual fuel, probably not much away from 100 kWt/l. One advantage with the CANDU though is that the moderation occurs in the calandria mostly, away from the fuel. That reduces losses to resonant capture which will definately help with reducing losses to Pa233 (and thus U234 as well).

            I think the biggest difficulty with these designs is that the higher fuel efficient options involve low burnups and frequent reprocessing, and that’s not easy with ThO2 fuel. A faster version with the lower D2O to fuel ratio mentioned in the report could get a higher burnup, but the specific power suffers so you need more kg of fuel to start this reactor. One way to think of it is that the reactor produces a lot less power for a given physical size, which is a serious economic downside.

            1. One way to think of it is that the reactor produces a lot less power for a given physical size, which is a serious economic downside.

              @Cyril R

              Can you put any numbers on the words “a lot less” and “serious economic downside”?

              It is my experience that designers are really good at their particular specialty and understand the economics of their narrow scope of work, but they have less understanding of the impact of their decisions on the whole power plant.

              For example, I have been told that my idea of a large, low pressure pebble bed reactor is not economical because the core has a low power density. The critics have no understanding of the fact that though the cores I propose have five times the volume for the same heat output the decision shrinks the plant size and cost shrink dramatically due to the reduction in supporting system requirements and the change from a high pressure container to what is essentially a low pressure tank.

              Even the shield thickness changes due to the lower flux at the surface of a low power density core.

              Take a look at some of the artwork available at this link and imagine the cost implication of allowing the core to be five times larger if that allows every other component to take up half as much space – or be eliminated altogether.

              http://econtent.unm.edu/cdm4/browse.php?CISOROOT=/nuceng

          2. @Engineer-Poet

            I am going to read the paper with interest, but within the first couple of paragraphs of the introduction I realized that the authors had probably never read WAPD-TM-1612, Proof of Breeding in the Light Water Breeder Reactor (LWBR Development Program).

            That is not really surprising; the paper is not easy to find. However, I have a copy that I can share with interested colleagues.

            1. Things have changed since I last searched for WAPD-TM-1612. It is now available as a print on demand paper from NTIS (National Technical Information Service).

              http://www.ntis.gov/search/product.aspx?ABBR=DE88005093

              Though $60.00 is a little on the steep side, it is cheap for the information provided. It is about half of the price of a normal engineering text book, so there is no excuse for the authors of the INL paper to have failed to understand its availability or its implications.

              Here is the quoted abstract from that important 1987 vintage paper:

              WAPD-TM-1612
              ABSTRACT
              The light Water Breeder Reactor (LWBR) was developed to demonstrate practical breeding while producing electrical energy in a commercial pressurized water reactor generating station. After the 233U-thorium fueled core operated successfully for five years, expended fuel from the core was assayed to prove breeding — that is, production of more fissile fuel from thorium than was expended to produce energy. This demonstrates the existence of a vast source of electrical energy potentially capable of supplying the entire national need for many centuries using plentiful domestic thorium.

              The LWBR operated for more than 29,000 effective full power hours in the Shippingport Atomic Power Station and contained more fissile fuel at the end of its five-year operational lifetime than at the beginning. Five hundred twenty-four fuel rods were randomly selected from the spent core for nondestructive assay to determine their final fissile fuel content. Two different nondestructive assay measurements were performed on each rod. Seventeen of these rods were subsequently assayed destructively to validate the nondestructive results. Two estimation methods, which agreed closely, were used to determine the fissile fuel content of the spent core. The more accurate method gave a value of 507.98 kg for the end-of-life fuel content of the core, with a total standard deviation of 0.14%. The fuel content of the beginning-of-life core was determined during core manufacture to be 501.02 kg, with a standard deviation of less than 0.04%. The increase in fissile fuel during core life was 1.39% +/- 0.14%, in close agreement with prediction, thus demonstrating breeding in LWBR.

              By the way, just in case there are any thorium advocates who still pay attention to Atomic Insights after my attempts to discourage them from disparaging uranium to try to make thorium look better, the LWBR program was run by Naval Reactors; it was one of Rickover’s pet projects. Alvin Radkowsky was the project leader; he later started a company named Thorium Power that is now called Lightbridge.

              That should help you understand that the reason thorium was not more aggressively pursued has nothing to do with the false theory that thorium’s suppression was led by the uranium lobby because we wanted to make weapons from the byproducts of commercial reactors. It should even help you understand that Milton Shaw, a Rickover acolyte, was probably not making choices against thorium when he decided to concentrate resources on the fast breeder reactor. He was simply choosing a single path as having a higher probability of success than attempting to pursue two separate – both with advantages and disadvantages – at the same time.

          3. Can you put any numbers on the words “a lot less” and “serious economic downside”? It is my experience that designers are really good at their particular specialty and understand the economics of their narrow scope of work, but they have less understanding of the impact of their decisions on the whole power plant.

            For example, I have been told that my idea of a large, low pressure pebble bed reactor is not economical because the core has a low power density. The critics have no understanding of the fact that though the cores I propose have five times the volume for the same heat output the decision shrinks the plant size and cost shrink dramatically due to the reduction in supporting system requirements and the change from a high pressure container to what is essentially a low pressure tank.

            Yes, if you start off with a new design and change the entire system, you can get an economic advantage even with lower power densities or higher fuel loadings and such. But the reference wasn’t about a new reactor type, it was about a different fuel in an existing reactor type. The use of a different fuel does not change the balance of plant or even the reactor itself much at all.

            The reference considered a tighter lattice pitch that increased the incore heavy metal from 80 tonnes HM to 140 tonnes HM. Since it has less moderator (the heavy metal displaces water) it will need a higher enrichment as well. You could easily end up with 3x higher fissile loadings. Well that, or you go for a lower total power (but checking the reference, they actually assumed the power to be constant).

            Design changes in nuclear fuel are all about compromises. You can never do just one thing. Using more coolant reduces pressure drop, reduces fissile starting load, and improves transient and accident performance. But if you want to run on thorium with a high conversion ratio and solid fuel, you have to reprocess a lot. That means high reprocessing costs and low burnup. Getting higher burnup (and thus less frequent reprocessing) can clearly be done with less coolant. But then you’re back again…

            Once you are in the position of designing a nuclear plant from the ground up, a lot of those compromises aren’t so hard. The pebble bed advanced high temperature reactor being a good example, simply an unpressurized tub of salt where power density isn’t so important anymore. The molten salt reactor is a future development that will allow fewer compromises on the reprocessing side.

            1. The pebble bed advanced high temperature reactor being a good example, simply an unpressurized tub of salt where power density isn’t so important anymore. The molten salt reactor is a future development that will allow fewer compromises on the reprocessing side.

              I’m a fan of a pebble bed reactor that is simply a lightly pressurized (perhaps 100-200 PSI) tank full of hot nitrogen that goes directly to a turbine, gets cooled in a heat sink and gets compressed back into that same large tank. Standard air might work just as well if there is a thin coating on the outside of the graphite pebbles or if testing at high temperature proves that rapid oxidation is not an issue.

              See Adams Engines for more information.

            2. You could easily end up with 3x higher fissile loadings.

              Does that matter as much now – with centrifuge enrichment and the possibility of even cheaper laser enrichment – compared to the days when gaseous diffusion was the norm? When I visited Georges Besse II, I learned that it would be able to produce as many SWUs each year with a power input averaging 55 MWe as Georges Besse I did with 2700 MWe

          4. Does that matter as much now – with centrifuge enrichment and the possibility of even cheaper laser enrichment – compared to the days when gaseous diffusion was the norm?

            It matters less now, a tripling would add around half a cent per kWh for both fuel and enrichment. Still that adds up to about a cent per kWh, not terrible but definately noticeable if you want to compete with coal.

            More importantly is the effect on reprocessing costs. In the reprocessing cycle you wouldn’t have a big impact from either fuel or enrichment, but the cost of reprocessing is dependant on the amount of fissile to recover, so starting out with 3x as much fissile per GWe isn’t helping the economics of the reprocessing operations.

          5. I’m a fan of a pebble bed reactor that is simply a lightly pressurized (perhaps 100-200 PSI) tank full of hot nitrogen that goes directly to a turbine, gets cooled in a heat sink and gets compressed back into that same large tank.

            Wait, isn’t that going to give issues with nitrogen activation? Are you planning on using enriched nitrogen perhaps? (it’s not very expensive to enrich nitrogen, simply co-locate a centrifuge train next to an air seperation unit).

            Also 200 psi seems very low. Gasses are not good coolants and have high pressure drops through pebble beds. I doubt you could get away with anything under 500 psi.

            P.S. I don’t like PSI. Imperial units make my head swim.

            1. @Cyril

              Nitrogen will only get lightly activated in a neutron flux. N-14 undergoes an n-p reaction to become C-14, which is a low energy Beta emitter. It is easy to remove from N2 gas, and easy to shield and store. It has some commercial value. The cross-section for that reaction is quite modest. Natural nitrogen is 99.6% N-14 and the cross section is just 1.8 barns.

              I’m not sure what “enrichment” you think might be required.

              With regard to the pressure drop issue, the key is to properly configure the heat exchanger to have a large flow area and low velocity. The pressure drop is proportional to the cube of velocity (if I remember the relationship correctly). Just remember, the highest pressure in a combustion gas turbine is only 10-25 times atmospheric pressure, depending on the specific configuration. With the temperatures available in an HTR, the optimal power pressure ratio is less than 10:1.

              One more refinement worth considering is the fact that there is nothing magic about 6 cm diameter pebbles. I think a more optimal size is about 1 cm, or about the size of a shooter marble.

          6. Wow, did I ever start something with that link!  Mixed response to Rod and Cyril:

            I am going to read the paper with interest, but within the first couple of paragraphs of the introduction I realized that the authors had probably never read WAPD-TM-1612, Proof of Breeding in the Light Water Breeder Reactor (LWBR Development Program).

            They directly reference the Shippingport LWBR effort on page 9 and in reference 1.6.

            FWIW, I stumbled across that paper while searching for information on the quantity and cost of D2O for CANDUs.  Reading a bit further was an eye-opener.

            I think the biggest difficulty with these designs is that the higher fuel efficient options involve low burnups and frequent reprocessing, and that’s not easy with ThO2 fuel.

            I'm not too concerned with fuel efficiency.  I see D2O reactors as a stepping stone to molten salt technologies, which aren't something we're ready to build in bulk today.  But if a once-through ThO2 fuel cycle could still run self-sustaining in CANDUs, it would produce a spent-fuel stockpile ripe for LFTRs.  (It would also provide a market for thorium, which is currently going begging; that would make a lot of US rare-earth deposits economic to mine.  The Chinese lock on REEs is a national security issue.)

            A big part of the idea is public relations.  Starting a bunch of thorium-breeding CANDUs on DUPIC fuel immediately makes the case that SNF isn't waste, it's a resource.  DUPIC fuel could be produced on-site, and eventually returned to the same dry casks in which it's currently stored.  If the CANDUs don't need any makeup uranium after the start, it would allow the LWRs which produced the SNF to be decommissioned eventually while the site continues to produce energy.  Or perhaps the CANDUs could be sized so that the SNF inventory is sufficient to provide makeup DUPIC elements for the intended life of the CANDUs.

            Once it's established in the public mind that SNF is a resource, even used ThO2 fuel with too many fission products for CANDU use wouldn't be seen as a threat, but another energy resource in waiting.  The image needs to change, and that's a possible way to change it.

            Maybe the shutdown of Gentilly is a sign that someone needs to buy it to be the proving ground for concepts like this.

      2. If about 100 reactors currently supply 20% of the USA’s electricity, then less than 400 reactors (average reactor capacity has increased) should easily supply more than 80% of the USA’s electricity. So we would need to build about 300 reactors.

        Let’s assume that the nation could do that for $5 billion per reactor. It should be less. The current reality is more. It’s an easy number by which to multiply…

        So the total cost would be about $1.5 trillion or less than we squandered in Iraq in the last eleven years.

        Assuming that the economy is not already damaged to the point where it is less healthy than it was in 2003, we ought to be able to afford the investment, and we would get something back for our money.

        1. The AP-1000 is rated at 1100 GW(e), so 400 of them would have nameplate output of 440 GW; actual production, probably around 400-410 GW average.  Average US electric consumption is about 450 GW.

    2. I wonder if the WWII occupation of France had any influence on France’s eagerness to embrace nuclear energy? French industry virtually shut down during that period (because France had depended on British coal to fuel its power stations, which they obviously couldn’t get during the occupation) and vast amounts of milk went to waste (because there was no fuel to power the trucks to haul it to the dairies).

      Perhaps the 1970s oil crisis — a second demonstration of the danger of relying on imported fossil fuels — was just the straw that broke the camel’s back.

      1. France’s decision was explained by following:

        “No coal, no oil, no gas, no choice.” It also had a lot to do with the loss of colonial control over Algeria, which was the primary place where French based multinationals extracted petroleum.

        The US was not in a much better position, but our multinationals had excellent concessions in the Middle East. There is some good information out there about how petrodollars got recycled to US based firms, making lots of people happy with the results of the oil price shocks of the 1970s.

        Interestingly enough, one of the primary reasons that the first nuclear age came to an abrupt halt in the US in 1973-74 was the Arab Oil Embargo. Public Service of New Jersey had four new plants on order to service expected load growth, the load stopped growing and actually shrank rather rapidly when Jersey’s refineries shut down because their crude raw material was embargoed for a while. The company cancelled their orders, which included some of the Westinghouse Off Shore Power Systems and that helped start a wave of cancellations.

  6. Discuss the nuts and bolts of getting the lead out of nuclear promotion, public education and advertising. Far more important than spinning wheels debating nuclear types and rad dosages. It’s moot to discuss a train that’s not coming.

    James Greenidge
    Queens NY

  7. I want to believe we will fix things by switching to a bigger mix of nuclear energy. What we need to learn as citizens of our planet is that it is playing dice to think that anything we attempt to do to influence people’s thinking will actually be effective. So if collectively people can change enough to make a difference then maybe our wish will come true. It has been said here and other places that leadership can make a difference. It’s true there are people that have shown up at the right time to fill the void when times have gotten desperate.

    Conservation of energy is not a proactive way to solve our problems. We already know that stopping all pollution today would not allow the climate and oceans to spring back quickly. We could all start conserving like crazy but all that does is lower our energy consumption and our dependency on oil and maybe our energy bills.

    So I propose that we be more than just soothsayers. I say we play the dice and be one of the ping pong balls in the nuclear fission demonstration and maybe we will be one of the ones who hit a mousetrap. Perhaps that is a highly overly optimistic way of seeing things but I don’t think we have a choice if we want future generations to have a planet to enjoy.

    Just like those guys we knew in college wanted to get a girl. The successful ones made it a numbers game. They never let a “no” stop them from trying again.

    It’s not just convincing people that nuclear energy is right it’s also convincing the converted believers that they have the power to influence change. I suggest start at the local level. That is where you’ll most likely succeed.

  8. Rod, I hope you are right, and you might be if we can get enough public support to keep the professional anti-nuclear industry and their congregations from putting up artificial barriers.

    I personally think there is one ‘law’ of economics, which I think will play into nuclear’s favor – especially once SMR’s go into production. . . you can’t beat “cheaper”. “Cheaper” ALWAYS wins in reality. The professional antis keep trying to convince people that nuclear is “too expensive”. Well, if they are right, nuclear will fail no matter how much supporters try to boost it. Their own chosen power sources are so expensive, that they are really, really hoping, I think, to be able to make nuclear expensive enough that they are close.

    But, if they are wrong, and you are right, short of them putting artificial barriers in that inflate the cost of nuclear beyond the natural value, the power of “cheaper” will kick in and take over the world. At that point, no one will need to really advocate for nuclear power, it’ll just happen. . . and fast.

    The only other potential obstacle is that coal and natural gas, at least for the short term, have the power of “cheaper” on their side. . . mainly because they do not have to pay for their externalities. They shift most of the cost of their power source out onto everybody else. I just hope people can realize that when you take into account all those other costs from pollution and GHG emissions, fossil fuel power is *actually* terribly expensive.

    1. @Jeff S

      Your analysis is on the right track, but please understand that “the invisible hand” of markets that Adam Smith is only invisible to those who have not had enough experience to learn about how business decisions get made. Though they are not always as successful as they would like to be, there are thousands of extremely well paid people who spend most of their working hours in efforts to manipulate market conditions to favor their own success.

      They spend money on politics so they can influence the actual text of the law to make it cheaper for them to comply – and more expensive for their competitors to comply. They negotiate with their employees to keep wages and benefits low – and they sometimes do this with the tacit agreement of other businesses so that everyone keeps wages low. (That same tactic agreement mechanism is what keeps executive compensation at stratospheric levels.) They work to raise barriers to entry for competitors, to achieve “lock-in” with their customers, and to force suppliers to bear as many of the costs of production as possible.

      The incredible energy density associated with nuclear fission offers a fundamentally lower cost means of producing energy, but human decisions have chipped away at that advantage to the point where people marketing new nuclear power generation struggle to show investors and customers that they will be cheaper. That is made more difficult because many of nuclear energy’s cost advantages are “in the long run” while the cost disadvantages of fossil fuel are either born by people outside of the transaction or are far enough in the distance that they are ignored by the spreadsheet modelers.

      Winning the campaign to dislodge hydrocarbon dominance and enable nuclear energy development is NOT going to be easy. It will take some hard-nosed, competitive individuals who really WANT to win. We are making some progress, but also suffering some significant battle damage and need reinforcements.

  9. Nitrogen will only get lightly activated in a neutron flux. N-14 undergoes an n-p reaction to become C-14, which is a low energy Beta emitter. It is easy to remove from N2 gas, and easy to shield and store. It has some commercial value. The cross-section for that reaction is quite modest. Natural nitrogen is 99.6% N-14 and the cross section is just 1.8 barns.

    It’s 5x the cross section of light water! And when light water undergoes neutron capture it becomes heavy water, which is innocuous (heavy water has a tiny cross section as well).

    It’s not easy to remove radioactive carbon from a stream of nitrogen. Residual oxygen, even in ppm amounts, will oxidise it to CO or CO2. This is because the carbon is formed one atom at a time, and therefore maximally chemically reactive. I’m pretty sure that the amount of radioactive carbon formed is rather large with a natural nitrogen coolant.

    Even if the radiocarbon is not a problem, the high cross section is a practical safety problem (void) and results in poor fuel usage.

    I’m not sure what “enrichment” you think might be required.

    Enrichment in nitrogen-15. Add a centrifuge train to an existing air seperation unit. Nitrogen 15 has almost no absorption cross section to neutrons. Only big trouble is that you need gas whereas the ASU will make cryogenic liquid nitrogen. You’d need an evaporator unit that costs energy, but I’m thinking about a regenerative cryogenic scheme that limits the energy expenditure.

    With regard to the pressure drop issue, the key is to properly configure the heat exchanger to have a large flow area and low velocity. The pressure drop is proportional to the cube of velocity (if I remember the relationship correctly). Just remember, the highest pressure in a combustion gas turbine is only 10-25 times atmospheric pressure, depending on the specific configuration. With the temperatures available in an HTR, the optimal power pressure ratio is less than 10:1.

    Heat exchanger? I thought you were going to use a direct nitrogen cycle?

    In any case, gasses have very poor volumetric energy density and that means high flow rates. Velocity can be reduced with higher pressure, not with lower pressure.

    The pressure drop through an open air Brayton is limited. Pebble beds are something else.

    One more refinement worth considering is the fact that there is nothing magic about 6 cm diameter pebbles. I think a more optimal size is about 1 cm, or about the size of a shooter marble.

    There’s a limit to the physical size of the pebbles, primarily due to mechanical reasons. The bed weighs down on the pebbles and graphite has poor strength. Small pebbles would result in excessive breakage. Also smaller pebbles means even more pressure drop.

    I would like to read more on these nitrogen cooled cycles. I’m familiar with the helium cooled cycles, and the pressure drop is serious especially for medium-larger size reactors. And helium is a better coolant than nitrogen…

    Got anything I can digest?

    1. @Cyril R

      5 times a small number can still be a small number. Besides, the 1.8 barns is a measure of the microscopic cross section for absorption.

      Since nitrogen is a gas and water is not, there are a lot fewer atoms in the coolant spaces of a gas cooled reactor. The macroscopic cross section for absorption is, therefore a lot lower in a nitrogen cooled reactor than in a water cooled reactor. That mitigating factor is less when high pressure gas is used and can lead to the issue of a positive insertion of reactivity if the pressure is lost. When the pressure is already low and the only thing that can replace the coolant would be air, which is already 80% nitrogen, the amount of positive reactivity that can be inserted with a complete loss of pressure is quite small and manageable due to the high negative temperature coefficient of reactivity.

      In any case, gasses have very poor volumetric energy density and that means high flow rates. Velocity can be reduced with higher pressure, not with lower pressure.

      You may have missed the part where I talked about configuring the heat exchanger (which, by the way IS the reactor in this case) to increase the flow area to reduce the flow velocity.

      There are industries that use catalyst beds that look a lot like pebble bed reactors, only the catalysts are in the form of much smaller particles. In order to minimize the pressure drops through these beds, they use a radial flow design where the gas enters the external boundaries of the bed and flows inward to exit through an annulus in the center of the bed.

      Take a look, for example at a catalyst bed in the chemical process bed as illustrated by Johnson Screens:

      http://www.johnsonscreens.com/sites/default/files/6/705/Internals%20for%20Radial%20Flow%20Reactors.pdf

      Your comment about smaller pebbles being less resistant to damage does not make much sense to me. If they are the same material, they would be able to take the same kind of pressure as larger pebbles. As you noted, there is a limit to the physical size of the pebbles, which is why I suggest that something on the order of a centimeter would provide better performance than the 6 cm pebbles that were selected because of the design of the continuous refueling system chosen for the AVR. (My proposed system does not include circulation of the pebbles at all. Instead, I propose using control rods through guide tubes that can be programmed to provide good burn up characteristics paired with rotating drums in the reflector region.)

      There are more details available at http://adamsengines.blogspot.com/ though I never quite got around to finishing the documentation of other concept papers that are still buried in my library.

      1. Yes. The key take-away point is that it’s a bad idea to cool your reactor with liquid nitrogen. 😉

        A couple of historical tidbits:

        These catalyst beds have more to do with pebble bed nuclear reactors than most people think. Such chemical reactors are the origin of the name “pebble.”

        The typical size of the pebbles has more to do with tennis than continuous refueling. Ever notice that the pebbles are approximately the size of a tennis ball? That’s not just a coincidence.

        A radial flow design is nothing new for nuclear devices. Particle bed designs proposed for space propulsion use this type of flow. They are not large devices and yet the temperature gradient across the particle bed is incredible.

        1. The typical size of the pebbles has more to do with tennis than continuous refueling. Ever notice that the pebbles are approximately the size of a tennis ball? That’s not just a coincidence.

          Nuclear engineers like tennis, who knew?

      2. Since nitrogen is a gas and water is not, there are a lot fewer atoms in the coolant spaces of a gas cooled reactor. The macroscopic cross section for absorption is, therefore a lot lower in a nitrogen cooled reactor than in a water cooled reactor.

        Unfortunately, that’s not the entire story. For a given heat (and therefore neutron) flux the gas flow is many times higher than the flow of water. The latter is an excellent coolant, whereas gasses are not. The cP of nitrogen is ~ 1 kj/kg/K. Water has a cP of >2 kj/kg/K (depends on reactor operating temp&pressure).

        There are industries that use catalyst beds that look a lot like pebble bed reactors, only the catalysts are in the form of much smaller particles. In order to minimize the pressure drops through these beds, they use a radial flow design where the gas enters the external boundaries of the bed and flows inward to exit through an annulus in the center of the bed.

        That’s an excellent design remedy Rod, very good thinking here. Flow path length being a giant factor in pressure drop, arranging a “car radiator” design is very clever of you.

        Your comment about smaller pebbles being less resistant to damage does not make much sense to me. If they are the same material, they would be able to take the same kind of pressure as larger pebbles.

        The weight of the bed is basically insensitive to pebble size, but the failure strength of the pebbles is more or less proportional to the diameter (or was it the radius?). You also need to move pebbles about in the core, swap them out and in with new pebbles. That’s wear and tear. Interesting that you are not considering this in your design, that would solve a lot of the grinding related issues. Usually in nuclear reactor design, any piece of graphite that has to sustain extensive life in a reactor core must be above 20 mm.

    2. It’s not easy to remove radioactive carbon from a stream of nitrogen. Residual oxygen, even in ppm amounts, will oxidise it to CO or CO2. This is because the carbon is formed one atom at a time, and therefore maximally chemically reactive. I’m pretty sure that the amount of radioactive carbon formed is rather large with a natural nitrogen coolant.

      If the reactions that you describe occur, that’s okay as well. CO2 is a well established reactor coolant.

      Calculations I ran a long time ago with some pretty high powered advisors (including Dr. Madeline Feltus) indicated that the total activation in the N2 coolant was lower than the total activation in water cooled reactors when you include the effects of corrosion and chemical additives. In other words – N2 is not perfect, but it is good enough.

      The only N2 cooled machine ever produced only ran for a few hundred hours, but the issues that led to its demise had nothing to do with coolant activation. The HTRE air cooled machines also showed that there was some activation issues with using air as a coolant, but the element of concern there was argon, not nitrogen.

      CANDU reactors have had some issues with N2 activation, but the real problem was that their standard for C-14 was terribly low and had nothing to do with an amount that would pose risk to people and never compared the levels to those that would be produced by many other interactions associated with nuclear energy production. The regulators simply declared that C-14 was bad and worth many millions to avoid.

      http://www-pub.iaea.org/MTCD/publications/PDF/TRS421_web.pdf

      1. If the reactions that you describe occur, that’s okay as well. CO2 is a well established reactor coolant.

        The amounts are physically small enough to not interfere with thermalhydraulics or chemistry.

        Calculations I ran a long time ago with some pretty high powered advisors (including Dr. Madeline Feltus) indicated that the total activation in the N2 coolant was lower than the total activation in water cooled reactors when you include the effects of corrosion and chemical additives. In other words – N2 is not perfect, but it is good enough.

        Okay. If you’re assuming PWR then there’s the boric acid liquid radwaste. BWRs don’t have this though, and the liquid waste is easier to contain than gaseous CO or CO2 waste. Activated corrosion products are usually not a big deal as they aren’t volatile (with some exceptions). Recall that tritium in HWRs is a big pain to contain, even in the less diffusive T2O form, because it is volatile. The cross section for tritium is tiny, whereas the cross section with nitrogen 14 is large. You’re going to make a lot of C14.

    1. I don’t think the effort was overly concerned with efficiency.  It was a proof of concept, and nuclear heat is really cheap.  If I were buying something for the military I’d be concerned with how much output I got per dollar and per ton of stuff to be air-lifted.  Thermal efficiency wouldn’t figure at all.

      Thermal efficiency could probably have been increased considerably with higher reactor temperature (it was only 1200 F).  Would neon be better than nitrogen?  0.04 barns is a pretty small cross-section, and if it captures something it becomes stable sodium.

    2. @Cyril R

      The ML-1 test results report was published in 1963. Brayton cycle gas turbines were quite new at that time. In addition, as the abstract points out, and the paper describes in more detail, the turbine compressor unit built for the project was not properly matched and had known performance issues.

      We have 50 years worth of experience in building commercially successful, atmospheric air breathing Brayton cycle gas turbines to build on.

      As Engineer-Poet points out, the ML-1 reactor could only produce an output temperature of 630 C, which is just barely adequate for a Brayton cycle machine. It was a proof of concept, nothing more.

  10. I don’t think the effort was overly concerned with efficiency. It was a proof of concept, and nuclear heat is really cheap. If I were buying something for the military I’d be concerned with how much output I got per dollar and per ton of stuff to be air-lifted. Thermal efficiency wouldn’t figure at all.

    Nuclear heat is cheap, but reactors cost real money, as do the ancillary systems ranging from reactivity control to decay heat removal and containments. The lower the efficiency, the bigger those costs get. So I think it would definately show, just indirectly in the equipment cost.

    Efficiency can be improved with intercooling and reheat, and by increasing the pressure. But I think Rod doesn’t want any of those things.

    Thermal efficiency could probably have been increased considerably with higher reactor temperature (it was only 1200 F). Would neon be better than nitrogen? 0.04 barns is a pretty small cross-section, and if it captures something it becomes stable sodium.

    Good idea! I checked the NNDC database for cross sections. Interestingly the US, European and Asian databases don’t give neon at all… the Russian database has it, fortunately.

    It looks like you’re right; neon has almost no cross sections to neutron reactions at all, not n,a or n,d, n,t n,p, not anything!

    That’s amazing. It’s not too heavy either so should be a good coolant. Too bad we don’t have any neon turbomachinery (or do we?). Cost and availability would be much better than helium.

    In any case, operating hotter than 1200 F is a good idea. There’s not much efficiency in Braytons without high temperature.

    1. We should have a lot less difficulty building neon turbines than helium turbines.  The molecular weight is 5x as great, which means far higher density and lower blade speeds for a given pressure ratio.  Gamma is the same as helium, which actually hurts efficiency; intercooling between compression stages is expensive and inefficient.

    2. @Cyril R

      Part of the beauty of the system is that the required fuel has been proven through physical testing to be able to withstand a temperature of 1600 C without any damage at all. It will only slowly begin releasing fission products between 1600-2000 C and does not completely fail until it reaches 2500 C.

      Physical testing for reactors up to 10 MWth backs up models showing that for any core thermal power less than about 300 MW, it is impossible for the fuel to exceed 1600 C even if all coolant is lost and all flow stops.

      In other words, the need for emergency cooling, back up power, and many other supporting systems is eliminated. The chemically inert N2 gas coolant should reduce long term maintenance costs. The low pressure system should greatly simplify the containment design.

      I have no interest in exotic gases or fancy heat recovery or intercooling systems. My goal from the beginning of the design process (initiated in 1991) has been to combine the low capital cost of commercially available Brayton cycle machines with the high energy density of uranium, low fuel cost and zero emission characteristic of a nuclear reactor.

      1. Physical testing for reactors up to 10 MWth backs up models showing that for any core thermal power less than about 300 MW, it is impossible for the fuel to exceed 1600 C even if all coolant is lost and all flow stops.

        I’ll just add that this property is independent of the type gas that is used for the coolant. As Rod mentions, the limiting case is complete depressurization, when almost all of the coolant is lost. (The remaining gas at atmospheric pressure has a negligible effect on heat transfer.) The factors that determine the peak temperatures that occur in such a situation are the thermal conductivity, the thermal inertia, and the geometry of the core.

        The thermal conductivity depends on the materials that one uses, and in the case of a graphite pebble bed reactor, it is going to be largely dependent on the amount of irradiation that the graphite has experienced (the highly irradiated end-of-life situation being the most challenging). The thermal inertia and geometry are inherent, key parts of the design.

        1. True enough for gasses. For molten fluoride salt coolants, it gets a lot better, combining higher power density, higher power cores with low pressure operation.

      2. Given the growing uses of diffusion-welded heat exchangers, is it proper to call any of them “fancy” any more?  One of the things I took away from Dostal’s analysis of a supercritical CO2 cycle for nukes is that regenerators can handle more thermal power than the reactor output.

        1. Compact HXs are an amazing innovation, so much more compact, and no more flanges or seals or heat affected zones to produce leaks.

          Do you know if Inconel 718 is amenable to diffusion welding into a compact HX? I asked Heatric but they weren’t sure.

          1. Sorry, I’m not a metallurgist.  Maybe if you browsed some mfgr sites and looked at their list of materials you could get a hint.

          2. Compabloc claims just about any alloy that can be welded, including Inconel 600. I guess it depends how weldable 718 is.

            What’s special about Inconel 718?

            Thanks. Looks like all Inconel can be diffusion welded, but the temperature must be quite high to approach parent material strength.

            Inconel 718 is awesome, very good high temperature creep, tensile strength, and resistance to supercritical steam. It’s also decent under irradiation. It costs and arm and a leg but materials usage for a compact diffusion bonded HX is quite low.

      3. I have no interest in exotic gases or fancy heat recovery or intercooling systems.

        In that case, your choice is between helium and highly enriched nitrogen-15. Neither are exotic. I’m not convinced that natural nitrogen is acceptable, and will not be till someone comes up with a Ci/l or Bq/l figure for equilibrium C-14 activity in the coolant.

      4. In other words, the need for emergency cooling, back up power, and many other supporting systems is eliminated. The chemically inert N2 gas coolant should reduce long term maintenance costs. The low pressure system should greatly simplify the containment design.

        I realize all that. At the same time there are many downsides and it’s not clear to me that these aren’t bigger than the pro’s.

        Lower pressure will tend to increase reactor size (reduce fuel power density). Cost of TRISO fuel fabrication is considerable even for higher pressure helium pebble beds. It will be higher for a lower pressure nitrogen coolant as you can get less power out of the fuel for a given temperature limit.

        The passive conduction decay heat removal is very safe, but limits the reactor size to perhaps 300 MWt as you say. That means far less economy of scale. The argument in favor of this path is that there is more modularity and more economy of factory fabrication.

        Per Peterson did some work on materials inventory for different reactors. Gas cooled reactors came out very bad, with very high inventory per MWe. Lots of engineered materials means higher cost.

        I realize that the advocates are claiming under $2/Watt for these reactors, but all advocates of a specific design do that. None of it is reflected in any real project.

        I will note that the South African pebble bed project has gone bankrupt. This was partly due to high development cost of the reactor, and partly due to high development cost of the closed Brayton cycle.

        I have minor experience with gas coolants and it is not good. They are not good coolants. Supercritical CO2 is better because it’s not so gas-like.

        1. Cost of TRISO fuel fabrication is considerable even for higher pressure helium pebble beds. It will be higher for a lower pressure nitrogen coolant as you can get less power out of the fuel for a given temperature limit.

          The cost of TRISO coating isn’t that expensive. The economics of TRISO fuel is going to be governed by the amount of enrichment required and the fuel performance, which will determine the amount of burnup that the fuel will be able to achieve. These two factors are not independent of each other. The remaining factor is the thermal efficiency of the power conversion system, and the higher temperatures that are achievable by gas-cooled reactors can only help there.

          Per Peterson did some work on materials inventory for different reactors. Gas cooled reactors came out very bad, with very high inventory per MWe. Lots of engineered materials means higher cost.

          All materials are not the same. Just counting the tons of this and the tons of that doesn’t mean anything economically. The real cost comes from the QA process for safety-grade equipment that must be available and the redundant backups for this equipment.

          Compared to most other nuclear reactor designs, a modular gas cooled reactor is a very simple concept. The nuclear heat part consists of a big vessel, some cross vessels, and a blower (essentially a big fan); none of it is particularly complicated. For a typical pebble bed design, this “big” vessel is smaller than a BWR vessel, operates at a lower pressure than a BWR, and contains an inert gas, rather than high-temperature water and steam. The most complicated piece of equipment of a modular gas-cooled reactor is the refueling machine, and that’s not safety grade.

          In the nuclear world, less safety-grade material means less cost.

          I will note that the South African pebble bed project has gone bankrupt. This was partly due to high development cost of the reactor, and partly due to high development cost of the closed Brayton cycle.

          This was due to the major investor, the South African government, cutting off funding for the project.

          1. TRISO is quite expensive. It’s not just the multiple coatings (this is actually quite cheap) it’s the total cost including graphite combined with the poor HM density and poor specific power in this fuel type (compared to LWR fuel).

            On a per-kg basis it’s not so bad, but because of the poor HM density and the poor heat transfer with gas, you need a lot more kg of fuel.

            A big downside to integrating the fuel with a solid moderator. You get a lot of fuel, that needs to be swapped out every now and then.

            The poor HM density also means high enrichment, with some designs going up to 20% U235. That’s expensive, even with high burnup.

            For a typical pebble bed design, this “big” vessel is smaller than a BWR vessel, operates at a lower pressure than a BWR,

            I don’t think so. The specific vessel power (kW/l) is lower for gas cooled reactors than for BWRs. I once did an excercise of comparing the different vessel, fuel, and coolant power densities. The vessel of gas cooled reactors is usually smaller than BWRs because the power output is so much smaller. GE’s ESBWR may have a vessel of around 1000 m3 but it also produces 4500 MWt, 1550 MWe. So 4.5 MWt/m3. Gas cooled reactors have trouble getting to that level.

            Also, most gas cooled reactor proposals I’ve seen are operating at similar pressures as BWRs (7 MPa give or take).

            Gas cooled reactors operate typically in once through and single phase. In that sense they are simpler than a BWR. But the difference is pretty small compared to the recent designs such as ESBWR, which are greatly simplified.

            It’s true in gas cooled reactors you can have inert coolant, but in reality there will always be some moisture and other contaminants in the coolant, which are debilitating to graphite. It is not easy to keep moisture out permanently. Also high temperature materials cost is influenced by designing against high temperature creep, requiring quite costly alloys. For the vessel you can use standard nuclear grade steel. Perhaps there is a small cost advantage in not needing a stainless steel liner with inert coolant.

            If there is an economic gain to be had it is likely more in the power conversion side. A simplified Brayton can be much simpler and more compact than a saturated steam cycle. Keep in mind this is not true for a design that operates at low gas pressure that Rod wants to do.

  11. Wow, this radial flow reactor is amazing. Not only does it greatly reduce pressure drop via increased flow area and shorter heat exchange paths, it also allows the vessel wall to be cooled by the inlet gas stream (which is quite cold for a gas cooled reactor).

    With such a design you could use standard nuclear grade steel for the vessel yet get very high outlet gas temperatures…

  12. Here’s a nice summary of the PBMR work:

    http://www.academia.edu/2014701/Pebble_Bed_Modular_Reactors_Versus_Other_Generation_Technologies_Costs_and_Challenges_for_South_Africa

    Reactor operates at 9 MPa inlet pressure, considerably higher than a BWR. Vessel also operates at a considerably higher temperature than BWRs and PWRs. 480 Celsius versus 280 Celsius.

    Cost of fuel fabrication is quoted as $4100/kgU (base case). Actually no one knows what it will cost. But the base case cost is 20x the fuel fabrication cost of LWRs, that are typically around $200/kgU.

    Notice the quite high cost estimates, ranging typically 10-30 cents/kWh. 4 cents/kWh is the lowest case often quoted by advocates, but this clearly requires all aggressive targets to be met. The experience with pilot plants so far does not indicate this (eg THTR in Germany was really expensive).

    http://cybercemetery.unt.edu/archive/brc/20120621082115/http://brc.gov/sites/default/files/documents/1020659.pdf

    1. On a per-kg basis it’s not so bad, but because of the poor HM density and the poor heat transfer with gas, you need a lot more kg of fuel.

      OK. You’ve lost me here. First, you say that the TRISO is “quite expensive.” Then you say that … well … the TRISO is not so expensive, it’s the additional graphite in the core that’s so expensive. Then you somewhat bizarrely claim that it’s expensive because you need “a lot more kg of fuel.”

      Huh?

      What does the heat transfer rate with gas have to do with anything? The amount of electricity that you get out of a given amount of nuclear fuel depends on the burnup and the thermal efficiency of the power conversion system. Period. The heat transfer properties of the coolant will influence the rate at which you can get the energy out — in other words, the power level — but it does not determine how efficiently you are using the nuclear fuel, which is the most important factor that determines fuel costs.

      The poor HM density also means high enrichment, with some designs going up to 20% U235. That’s expensive, even with high burnup.

      These days, enrichment is cheaper than it has ever been. So you enrich your fuel to just under 20%. So what? The only reason that 20% is chosen is for political/regulatory concerns (i.e., it’s under the non-proliferation limits for LEU). If it weren’t for these weapons-related concerns, engineers would be pushing for even higher enrichments than that.

      Higher enrichments are not a net cost, because they buy you something, namely better fuel performance. Even advanced concepts for LWR fuel that are being studied by the DOE use higher enrichments.

      The vessel of gas cooled reactors is usually smaller than BWRs because the power output is so much smaller.

      Yes, that’s my point. There is nothing special about the “large” pressure vessels that are used in most modular high-temperature gas-cooled reactor (HGTR) designs. If you can provide them for BWR’s then there should be no problem providing them for gas-cooled reactors.

      GE’s ESBWR may have a vessel of around 1000 m3 but it also produces 4500 MWt, 1550 MWe.

      The designers of the ESBWR understand what I’m talking about. The size of the vessel just isn’t that important for the economics of the plant. It’s all of the other crap — redundant safety trains, additional safety-grade equipment, a large, robust containment structure — that cost so much. The reason why the ESBWR is labeled “economical” is that its designers tried to eliminate as much of that stuff as possible, and yet they made the vessel bigger.

      Also, most gas cooled reactor proposals I’ve seen are operating at similar pressures as BWRs (7 MPa give or take). … [PBMR] operates at 9 MPa inlet pressure, considerably higher than a BWR.

      When I think of “considerably higher,” I think of the pressures in a PWR, which are over twice the pressure in a BWR. A mere 30% more is not that impressive. The pressures in the primary loop of a modular gas-cooled reactor design are typically in the range of 5 to 7 MPa, with 7 being the high end. In the HTR-Modul, the premiere modular pebble-bed design developed in Germany, the pressure in the primary loop was chosen to be 6 MPa, slightly lower than the pressure inside a typical BWR.

      Here’s a nice summary of the PBMR work:

      Hmm … any summary that states that the “helium enters the reactor at 4,820 degrees Celsius and 9 megapascals (MPa), exits the reactor at 9,000 degrees Celsius …” won’t be described by me as “nice” anytime soon. I can only assume that the rest of their figures are equally as sloppy.

      Vessel also operates at a considerably higher temperature than BWRs and PWRs. 480 Celsius versus 280 Celsius.

      Well, you should realize that the operating parameters cited in the paper that you referenced, even after correcting for their order-of-magnitude mistakes, were taken from a time when PBMR was still competing to be the design to be chosen for the DOE’s NGNP project. Back then, the emphasis was on a nuclear reactor designs that could be used to generate hydrogen, possibly using the S-I chemical water-splitting process that requires temperatures on the order of 900 degrees Celsius. Thus, all of the candidate designs were claiming very high temperatures, which would necessitate using more expensive nickel-based alloys for the pressure vessel.

      The HTR-Modul used an inlet temperature of only 250 degrees Celsius, which means that the pressure vessel, during normal operation, would be at a temperature slightly lower than that.

      Gas cooled reactors operate typically in once through and single phase. In that sense they are simpler than a BWR. But the difference is pretty small compared to the recent designs such as ESBWR, which are greatly simplified.

      Greatly simplified? Yes, I suppose that ten safety release valves, eight depressurization valves, four independent high-pressure loops, and an additional four safety-related low-pressure loops must seem greatly simplified for a BWR, but it has a long way to go to match the simplicity of modular gas-cooled reactors.

      During an accident, an HTGR typically relies on only two redundant loops operating at atmospheric pressure to remove heat from the reactor cavity. These loops are entirely passive, and even if they fail, the decay heat can be conducted outward to the ground. Sure, the concrete is probably a complete loss, but even this scenario is not a serious issue when it comes to public safety.

      It’s true in gas cooled reactors you can have inert coolant, but in reality there will always be some moisture and other contaminants in the coolant, which are debilitating to graphite. It is not easy to keep moisture out permanently.

      Sure it is! You simply avoid using water-cooled bearings in your blower. This has been known for over 25 years. You seem to have a passing familiarity with the problems of Fort St. Vrain without having followed up to familiarize yourself with the lessons learned from that experience.

      Also high temperature materials cost is influenced by designing against high temperature creep, requiring quite costly alloys. For the vessel you can use standard nuclear grade steel.

      The internals of an HGTR are either in a relatively low temperature region (e.g., the bottom of the vessel) or are made of ceramics.

      Keep in mind this is not true for a design that operates at low gas pressure that Rod wants to do.

      I’m constantly baffled why, whenever I try to inject some useful information into a discussion, I’m always accused of defending something that I never wrote, said, or even alluded to. I’m not familiar enough with the design that Rod has in mind to think that I can comment intelligently on it. I’ll leave it to Rod to explain his own ideas.

      Notice the quite high cost estimates, ranging typically 10-30 cents/kWh. 4 cents/kWh is the lowest case often quoted by advocates, but this clearly requires all aggressive targets to be met.

      I noticed that these researchers used a rather high estimate for nuclear fuel costs in their calculations. More importantly, however, I noticed that they concluded, “if current assumptions about capital costs and construction time are valid, PBMR is actually cost competitive with renewables and other nuclear generation in most cases.”

      Imagine that.

      The experience with pilot plants so far does not indicate this (eg THTR in Germany was really expensive).

      The THTR-300 was not a modular pebble-bed reactor, and it used a different fuel cycle than the reactors that are being discussed here. Sorry, but you’re trying to compare apples to oranges.

      1. OK. You’ve lost me here. First, you say that the TRISO is “quite expensive.” Then you say that … well … the TRISO is not so expensive, it’s the additional graphite in the core that’s so expensive. Then you somewhat bizarrely claim that it’s expensive because you need “a lot more kg of fuel.”

        Huh?

        The other reference I gave show base case cost of $4100/kgU for TRISO pebbles, and $200/kgU for LWR fuel. That is a factor of 20 higher. A kg of triso fuel contains far less uranium, it contains mostly graphite. But the cost per kg of uranium is what matters. It’s true, the higher burnup and efficiency compensate somewhat, but they only add up to a factor of roughly 2, not 20.

        I like TRISO fuel. But I prefer that they be cooled with supercritical or at least boiling water, eliminating the graphite. A pool of water provides all the passive heat sink I need.

        What does the heat transfer rate with gas have to do with anything? The amount of electricity that you get out of a given amount of nuclear fuel depends on the burnup and the thermal efficiency of the power conversion system. Period. The heat transfer properties of the coolant will influence the rate at which you can get the energy out — in other words, the power level — but it does not determine how efficiently you are using the nuclear fuel, which is the most important factor that determines fuel costs.

        The heat transfer efficiency determines the maximum temperature at a given power density and thus (since the temperature is the limit for TRISO) the power density you can get from fuel for a given temperature. A good example is the fluoride salt cooled PB-AHTR. It has more economical fuel fabrication due to better heat transfer allowing higher power density. It also allows higher burnup because of the lower particle operating temperature (this determines internal pressure in the particles during normal operation, transients and accidents, an important factor in fuel failure). Power density of the core determines the cost and size of the vessel and building it’s in, and a large number of major cost items in fact in a nuclear plant.

        These days, enrichment is cheaper than it has ever been. So you enrich your fuel to just under 20%. So what? The only reason that 20% is chosen is for political/regulatory concerns (i.e., it’s under the non-proliferation limits for LEU). If it weren’t for these weapons-related concerns, engineers would be pushing for even higher enrichments than that.

        The engineers might, but the accountants wouldn’t. The current market price is around $140/SWU. That is not cheap.

        http://www.neimagazine.com/story.asp?storyCode=2063124

        Centrifuges have low energy costs, but they have considerable investment and operational costs (they are high rpm spinning machines made of expensive alloys).

        Cost of enrichment is a real factor in gas cooled TRISO fueled reactor design. The Berkeley work recently shifted to the lower enrichment, more frequent fuelling design, due to high enrichment costs. For LWRs going from 4% to 5% is often economic but going much beyond that wouldn’t.

        When I think of “considerably higher,” I think of the pressures in a PWR, which are over twice the pressure in a BWR. A mere 30% more is not that impressive. The pressures in the primary loop of a modular gas-cooled reactor design are typically in the range of 5 to 7 MPa, with 7 being the high end. In the HTR-Modul, the premiere modular pebble-bed design developed in Germany, the pressure in the primary loop was chosen to be 6 MPa, slightly lower than the pressure inside a typical BWR.

        These pressures are of a similar difficulty level. My point being that you have more vessel per kWe and higher temperatures than with BWRs.

        Yes, that’s my point. There is nothing special about the “large” pressure vessels that are used in most modular high-temperature gas-cooled reactor (HGTR) designs. If you can provide them for BWR’s then there should be no problem providing them for gas-cooled reactors.

        I doubt you understood the point. It’s not technical, it’s economic. If you have a similar cost vessel that produces say 4 or 5 times less electric output, the vessel cost per kWe is increased. High pressure vessels are expensive. Even for gas coolants. It’s a field apart.

        Hmm … any summary that states that the “helium enters the reactor at 4,820 degrees Celsius and 9 megapascals (MPa), exits the reactor at 9,000 degrees Celsius …” won’t be described by me as “nice” anytime soon. I can only assume that the rest of their figures are equally as sloppy.

        So they got a comma wrong! Happens to the best researchers. It’s easy to see what they meant. You are being unreasonable in dissing a good article over a misplaced comma.

        The HTR-Modul used an inlet temperature of only 250 degrees Celsius, which means that the pressure vessel, during normal operation, would be at a temperature slightly lower than that.

        The efficiency will suffer with lower inlet temperatures. Thermal stresses will also increase with a larger temp rise in the core, plus in the real world you have all sorts of high frequency cycling going on as well during normal operation and transients.

        Greatly simplified? Yes, I suppose that ten safety release valves, eight depressurization valves, four independent high-pressure loops, and an additional four safety-related low-pressure loops must seem greatly simplified for a BWR, but it has a long way to go to match the simplicity of modular gas-cooled reactors.

        If it’s a pressure vessel, it needs safety valves. Gas cooled vessel aren’t any different. You’re a smart guy, I’m sure you understand the ideal gas law, and what happens if there’s a transient to the temperature, and thus to the pressure.

        In any case, valves are not a major cost item. High throughput steam relief valves would cost less than 100k each.

        During an accident, an HTGR typically relies on only two redundant loops operating at atmospheric pressure to remove heat from the reactor cavity. These loops are entirely passive, and even if they fail, the decay heat can be conducted outward to the ground. Sure, the concrete is probably a complete loss, but even this scenario is not a serious issue when it comes to public safety.

        Alas, if only it were so simple!!! The ground is an insulator. I tried that idea and quickly learned it doesn’t work. You need an air cooling or water cooling DHR. Air cooling is possible, like RVACS, but the RVACS involves considerable safety grade chimneys, downcomers, risers, liners and insulation. All of that is safety grade and seismically qualified, of course. Can’t have the chimneys coming down in an earthquake.

        Sure it is! You simply avoid using water-cooled bearings in your blower. This has been known for over 25 years. You seem to have a passing familiarity with the problems of Fort St. Vrain without having followed up to familiarize yourself with the lessons learned from that experience.

        It helps to not have water cooled bearings and steam cycles and water cooled gas coolers. It doesn’t eliminate the problem. There’s no such thing as pure helium, this has been known for considerably longer than 25 years. Keeping it pure is not easy, especially because helium leaks out a a brisk pace, so new helium must be fed in. I think that a nitrogen-15 coolant powering a nitrogen closed Brayton with air cooling would solve most problems, as it would tend to not leak out like helium does.

        I’m constantly baffled why, whenever I try to inject some useful information into a discussion, I’m always accused of defending something that I never wrote, said, or even alluded to. I’m not familiar enough with the design that Rod has in mind to think that I can comment intelligently on it. I’ll leave it to Rod to explain his own ideas.

        I’m not accusing you of anything. I quite enjoy discussing with you and reading your comments on this website. I’m trying to figure out whether a low pressure, say 1 MPa design would work. At first glance it seems difficult because of the poor heat transfer at this low pressure on the one hand, and the large turbine size (and possibly lower efficiency) on the other hand.

        The THTR-300 was not a modular pebble-bed reactor, and it used a different fuel cycle than the reactors that are being discussed here. Sorry, but you’re trying to compare apples to oranges.

        It was a pebble bed reactor and it used a ceramic coated particle fuel. It was pretty big so should have gotten some economy of scale. Yet it was considerably more expensive than light water reactors of that time. Fuel cycle I don’t think is so important here, thorium or uranium. It did use a steam cycle so there’s added complexity there that may have explained some of the cost.

  13. By the way, here’s the reference from dr. Peterson from UCB that estimates metal, concrete and building volumes input or different nuclear plants:

    http://pb-ahtr.nuc.berkeley.edu/papers/05-001-A_Material_input.pdf

    Notice that the PBMR has several times the “nuclear” input as the ESBWR. This casts doubt on its economic viability.

    The GT-MHR however does much better, though again in nuclear input it’s about twice that of the ESBWR…

    One of the advantages of a BWR is that steam in condensable, whereas ideal gasses are not. This, combined with poor heat transfer from gasses (compared to boiling water which is excellent) results in very large buildings with these gas cooled reactors. Building size and concrete input are of course strongly related, so you suffer there as well.

    The idea of the modular gas cooled reactor folks is to not have a high pressure containment. The argument being that the fuel is very retentive and high temperature capable. I wonder if the NRC is going to buy it (I would).

    1. The question is what’s cheaper, N-15 or neon?

      Maybe it’s time to look at a sealed molten salt reactor with induction circulation pumps (no penetrations) and see how it compares for cost and bulk.

      1. A very good question! Enriching N-15 is likely a bit more expensive than neon. But it would allow nitrogen turbines that Rod likes for similarity to today’s open air cycles… quite likely, the cost of the coolant isn’t decisive in the total plant cost anyway. Even for expensive coolants like heavy water, this is the case. Enriched nitrogen should be much cheaper than heavy water.

        Regarding induction pumps, molten salts are not conductive to electricity, unfortunately. Canned pumps could work, but need some R&D for pushing salts and for the magnetic materials at elvated T. I do like the idea of molten salt coolants for TRISO fuelled reactors. It should be more economical, and makes the turbine cycle more attractive (you can do things like reheat much easier and more compactly with molten salt to gas HXs than gas-to-gas HXs). It will also put us on a more useful long term strategic trajectory, namely that of the molten salt reactor. Gas cooled reactors are interesting, but the only development advantage we’ll get for future reactors is the closed Brayton cycle…

        1. molten salts are not conductive to electricity, unfortunately.

          That will be a HUGE surprise to the groups doing electrorefining of SNF, as well as the folks refining aluminum for the last century or so.

          1. I meant that they aren’t conductive enough to be efficiently moved by an EM pump. They are certainly good enough for electrolytes. You could probably get some flow going with an EM pump, but it’s going to be mighty inefficient for the megawatts of pump work that a large MSR needs.

  14. Here’s a nice webpage where you can calculate the specific enthalphy of nitrogen gas.

    Using 10 bar pressure, 250 C inlet, 750 C outlet, allows to calculate a specific volumetric enthalphy gain (p-deltaH) of 146 kj/m3, generously assuming isobaric conditions.

    This is well over a thousand times less than the specific volumetric enthalpy gain of a BWR, and about 900 times less than that of a PWR.

  15. A nuclear gas turbine would be competing with natural gas turbines, but they can run at temperatures approaching ( or above ) the melting point of the turbine blades, since they can dump the exhaust gases into the atmosphere. Nuclear plants have to retain all their ‘exhaust’ and so have a couple of hundred degrees disadvantage to start with, if using the same fluid and similar pressures. Going to supercritical CO2 would need a lot more initial R&D, but should finish up with a much more compact and competitive product.

    1. @John

      Sure, but the heat from nuclear fuel costs about 65 cents per million BTU, even when all of the costs associated with fabrication, enrichment, storage, transportation and waste disposal are included. In contrast, the “cheap natural gas” available in North America today costs at least $3.50 per million BTU at the trading hubs, not delivered to the power plant. In Japan, LNG is landing at a cost of $16 per million BTU.

      Even more important, the very best natural gas fired turbines release 450 grams of CO2 for every kilowatt-hour of power that they produce. Nuclear heated gas turbines release Zero grams at the power plant and only about 8-20 grams when you include the entire fuel cycle.

      R&D is fun, but it does not produce any revenue and does not stop CO2 from entering the atmosphere.

Comments are closed.

Recent Comments from our Readers

  1. Avatar
  2. Avatar
  3. Avatar
  4. Avatar
  5. Avatar

Similar Posts