David LeBlanc – Molten Salt Reactor Designs, Options & Outlook @ TEAC4

David LeBlanc – Molten Salt Reactor Designs, Options & Outlook @ TEAC4


I’m Dr. David LeBlanc, and I was with Carlton university I’ve
left them to work full time on this. I’m kind of called the thorium expert a lot
but it’s molten salt reactors – that’s my field. So, talking about different aspects of
how we can design molten salt reactors – basically breeder or burner – one of the main
categories we can differentiate here. Breeder of course makes its own fuel after startup.
You need fuel to start them though. If we make just enough we call that breakeven.
and that’s kind of nice – you don’t have fuel coming in or going out, but that does require continuous processing
to remove fission products from the salt. A burner design or converter design or
what I’m gong to focus on – which is the DMSR or denatured
molten salt reactor, which is both a converter and denatured. Anyway, as you can see we already get into broader [categories]. That does need annual fissile makeup, but it can skip that fuel processing
and that is a surprisingly large advantage. And, just in general much, much, less
research and development needed. and you can really simplify your
core design. Molten Salt Reactor advantages – and I’ll stick the ones that are broad – to
all the different ways we can run these: Increaseed safety, reduced costs, resource
sustainability, greatly reduced long-lived waste. So, in my talks I’ll usually try to go through this
piece by piece but there’s no way I have time – I want to fit in a lot of different things. So, I think that most of the people in the
room that know a little bit about molten salts will agree about the increased safety and overall reduced costs. I want to focus a little
bit on resource sustainability and long lived waste. To differentiate between breeder and burner – because – the breeder design, once you
start that it is pretty amazing. Only about one ton of thorium but in most designs you at least have to work in the
chemical processing you might lose a little bit of thorium. So its actually up to about ten tons of
thorium per gigawatt-year, but that’s still basically free fuel. Maybe $30,000 worth of thorium
giving you $500,000,000 worth of electricity. But, you must when you’re talking fuel costs, you must
add the fuel processing cost and of course the cost of that starting
fissile material. Converter designs are simpler and
only require very modest amounts of uranium – when running them on low enriched uranium. Oak Ridge’s main design, the
MSR, typically about thirty five tons of uranium per gigawatt-year,

0:02:13.450,0:02:15.880
versus about two hundred tons for a light-water reactor and those are very specific known costs
that up only to about a tenth of a cent per kilowatt hour. Its really hard to
imagine and improvements. And, I will say uranium is not the enemy, okay? Only cheap uranium is unlimited
supply. Now, Ken showed us very important things about the potential
bottlenecks in the uranium supply but in general, if you allow the
price of uranium to increase there’s a rule of thumb – if the
real price of a a metal or mineral etc doubles, you basically get about ten times the reserves. So, we could have bottlenecks,
and I agree with that But, if you allow the price to rise,
let’s say [by] $500/kg that’s gonna really hurt light water
reactors – not really put them out of business, but that does next to nothing to
these more efficient designs – its only about $0.02 per kWh and we have basically unlimited supplies at that [price]. Uranium mining – it has a bad image but
it’s only a tiny fraction of world mining – less than a 0.1%, and it’s good employment. If uranium is used in these
designs, these converters, we could have all our electricity 2,500 GWe
without increasing current mining. And of course, we’re not going to get rid of
hydro and of course wind and solar should play their 10 to 20% percent part. So, if we do get rid of the old fleet
and introduce these, we beat the sustaining
thing with a lot less mining. Getting on to long-livee waste, fission
products are almost all benign after a few hundred years. We have a very small number of long
lived – and they’re really not that much of an issue. it’s really transuranics – everything above
uranium, neptunium, plutonium, etc, that’s the real reason for your “Yucca Mountains”, etc,
which are repositories – places you want to put things because you
might want to take them back later. All molten salt reactor designs produce
a lot less transuranics, so we can either keep recycling them
continuously back in the reactor or with the DMSR design – which I’ll show
more later – we basically keep them in [the reactor]. As long as we’re doing that we
can have up to a ten thousand fold improvement over transuranic waste, compared to conventional once through
designs and even including MOX use doesn’t improve things very well. Re-examining molten salt reactors, They are
often thought of as the thorium reactor, that’s that’s the label. But, by mandate they were developed
to be breeder reactors to compete with Sodium Fast Breeders the belief that time was there is almost
no uranium in the world, we’ll have a few years worth of these submarine reactors and then it will all [have to] be all either fast breeders or
Molten Salt Reactors (MSRs). We now know a lot better.
Molten salt reactors can be both burners or breeders but the choices really have to come down to pragmatic facts, not ideology or imposed
funding mandates that you have to be a breeder, etc… But no one can dispute the success of basically pitching things as thorium – you
can’t explain to the public that a reactor is better. They’re not going to
listen very long or or have any hope of understanding. So, I’m not saying to stop this, but I’m saying to
kind of realize, the message I want to try to get across is:
Come for the thorium but stay for the reactor. I didn’t have time to make a – I’ll get John to make t-shirts for the next one I guess. Back to breeders.
Researchers do with tend to focus on the breeder. I was in the same way, I didn’t
want to look at these converter designs, for the first few years i was into this. But, when you really look at the
R&D and operational costs of continuous processing it’s a lot higher than people assume. Salt processing should be
much cheaper than with solid fuels. But, you have to remember – anything nuclear
related – things do get very expensive. Solid fuel reprocessing conservatively is about $2000/kg, Now, how much cheaper would salt processing [need to] be –
[for] liquid fuels? A lot cheaper – 90% cheaper? 95%? 99%? When you look at the standard molten
salt breeder reactor, that’s where we have the most data, to match the fuel cycle costs of DMSR processing, All that processing within need to be less than $1/kg. I don’t think you can say you are going to have
a 2000-fold reduction. Now, other designs will need to process less, etc., but I think this issue gets
swept under the rug a little bit too much. So, when you remove that requirement to breed,
you open up all manners of design simplification. A burner has almost negligible fuel costs
assured resources, enhanced anti-proliferation features, and simpler R&D. So, it appears the obvious
choice, and of course at any point down the road and at the same time
we can be investigating breeder options or convert later. So, what is this DMSR converter
reactor I’m talking about? Oak Ridge developed this –
It was one of the last great advances had on the molten salt program. With very little funding, it was developed in the late 1970’s. it was designed to be a gigawatt output,
starting up with low-enriched uranium, as high of an enrichment as they could do
safely – for proliferation [concerns], so they could squeeze in as much thorium as
they can to basically make the neutronic budget a little better. But, there’s no salt processing.
Just add small amounts of low-enriched uranium annually, which we’d buy off the market, pretty low starting fissile and
it’s the same thing that other reactors use. They have better reactivity coefficients than MSBR –
people might want to ask me about that later – it is an interesting fact – and they only required about 1/6th
the annual uranium needs of a conventional reactor. Or- CANDUs are a little better but these
are still much, much, better than CANDU. And again, that fuel processing cost –
there is no fabrication, etc, or a lot less of it, very small, a light water reactor might be 0.6 cents to 1.0 cents per kWh. After that 30 year batch of salt, the
uranium can be removed and and reused, that’s fairly straightforward. Transuranics, that becomes kind of a national choice. If you want to bury them in the ground, that’s your choice, but I’d like to see them recycled –
this is a one-time job. There’s only going to be about one ton
of these transuranics in the entire batch of salt for a 1 GWe reactor. So, if you assume, we always have to
assume a small amount of processing loss – a tenth of a percent is a is a typical goal – and that’s the goal when we talk
about processing – the other end of the [spectrum] the pure [Th-U233] cycle – That really only means 1 kg of transuranics going to waste over 30 years, and that’s actually
is good or even better in most cases that are examined with
what we call the pure cycle – the pure thorium – U233 – because you’re processing a lot more
rapidly, more often, etc. And, this is the only real reactor –
I won’t get into the details – you can really say that this reactor – because we’re burning a
little bit of low-enriched uranium, we’re actually destroying or transmuting a fair amount of national radiotoxicity
from the ground. And, in this case, after 300 years, and if we do
these things and recycle the transuranics, we can make the claim that the planet, is less radiotoxic than when we started. After that 300 year waiting period, where we can trust an engineer to
vitrify, or to make things that aren’t going to leak in the water table, etc. The thorium reactor has – the pure cycle – has pretty much the same
output but they don’t transmute as much. How does a DMSR do so good?
Well, I do talks a lot in Canada and heavy water’s sort of king in Canada. But, isn’t heavy water the best moderator? The big thing is far less parasitic losses of neutrons. we don’t have any internal structure,
no burnable poisons, and a lot less neutron leakage. Light water reactor is, typically about 22% of parasitic losses, and that’s not even including fission products. CANDU is much better at 12% but the DMSR is
way down here at about 5%. So that’s really the real reason they’re so good. Plus, about half of your fission products actually leave as gasses.
The xenon, krypton, and a lot of things that started as xenon and krypton, come right out, and of course the most important, xenon-135 that just absorbs great amounts of neutrons. Extremely high proliferation resistance,
I’ll zip through this – we’re not really processing –
the salt stays in there for a long period of time. The uranium is always denatured meaning it’s
low enough in fissile content, you can’t use it as a weapon. Any plutonium present – it’s really
low quality, very dilute, in a very radioactive salt, and really hard to remove.
Like comparing to light-water Pu, a lot more spontaneous fission, a lot
more heat rate – so if it’s virtually impossible for light-water spent plutonium to be weaponized, it should be that much more impossible for these. These reactors, have no way to sort of
put things in and take them out to put in some kind of fertile material and take it out. And, some of the last people could say that – “Well, you could still have enrichment plants”.
Well we’re going to have a lot less of them. but if you don’t like enrichment, we could have a synergy with say, a natural uranium reactor like a CANDU. CANDU produces a lot of plutonium
actually, so on a single site a CANDU feeding it’s
waste plutonium as the makeup fuel for several DMSR’s. So, basically natural uranium in,
electricity, and fission products out. No- denatured designs – LFTRs – well
LFTRs are supposed to be a larger category but everyone has their own ideas of the best ways, there is interesting non-proliferation features, but It is true that likely, there’s no expert
you find on proliferation resistance that would that would tell you it’s an
improvement over existing reactors. It does have advantages, but these widespread claims of thorium being
a solution to proliferation – it’s only going to hurt us in the long run. And, we’ve heard that a bit before and it’s true. And John, I love him, and there’s no way
I could ever be angry at John, he’s such a great guy but – making these statements, of “thorium is non prolifering, etc”,
Well, if you use it in the reactor it can be. The claims of the effects of U-232 –
they’re greatly exaggerated. We had a sort of a public flogging
at dinner last night of some claims and he claims he got it from a reputable
source – talking about how a bomb pit would kill you a mile away within five minutes. The effects of U-232 are greatly exaggerated. They are important, they are great for detection, etc., but it’s not gonna kill you instantly It’s not going to kill you – you can sit next to it for weeks or
months before you ever get a lethal dose – and that’s after it’s built-up for many,
many years to build up the daughter products. so see Dr Ralph Moir’s paper or other
on the effects here. Yes, a country developing a graphite pile – that’s a lot easier, but not if you could buy a reactor right
off the market that’s got tons of U-233. So, proliferation dangers will always be
exaggerated by those who oppose nuclear power, but I don’t think the answer is making
similar exaggerations the other way, because we’re going to get caught on these in the end. So, there is a very good case for all these
reactors being not a proliferation worry, but please quit the great exaggerations we see. Okay, getting back off my soapbox here, Getting back to the Oak Ridge designs
which had very little time or funding on these, so there’s a lot of fertile ground for improvement – shorter batches of the salt – I don’t really
like the 30 year cycles, As long as you recycle the uranium in
these designs because there’s a fair amount of fissile trapped up there and it’s fairly easy, Transuranics can wait, you get a large improvement
in the uranium needs. 10 to 15 year batches are what I’d probably more like to see, And, you can pretty easily get
things down about 20 tons of uranium per GWe year. So, it’s really not that much more
than even the breeder cycle. And, that is just about 10% of a light-water
reactor’s [need for uranium]. So, again all the world’s electricity – but we’re not gonna get rid of hydro, etc., so we can actually get by with a lot less mining. What about no thorium? I’ll just duck the tomatoes coming in here. Running without thorium actually does
have some interesting advantages. We can start on much more common
5% or lower enrichment, the neutron economy is not as good, but
it’s still absolutely excellent compared to existing reactors. There’s no protactinium, you don’t want a high power density
if you have protactinium. There’s all kinds of things with removing that protactinium, and typically the melting points of these
salts are less when we don’t have thorium. I want to cover a lot of things so I’m not
really going to get into new options, not much ready for public disclosure and
I apologize as I said the same thing last time. But a very obvious thing is Oak Ridge’s work has been by force on solid fuel, TRISCO fueled, molten salt cooled. They’ve come up with a great deal of tricks for doing that better and it’s quite obvious
that these same tricks can be used in molten salt fuel. Just replace the TRISCO fuel, with just
graphite, put the fuel in the salt and you’ve got a pretty excellent design as well. Molten salt reactors in Canada – CANDU6 is
a is a good design, available now, but there’s no new R&D for the foreseeable future since it was sold to SNC-Lavalin. So, we have an enormous nuclear brain trust
basically going to waste. We went our own way before on the CANDU
we can do it again, and Canada also has unique
opportunitys in our oil sands. And again, our oil sands, most of
it is not going be mined – it’s all in situ – where you use steam assisted gravity drainage (SAGD), You make steam, pump it down, it basically helps heat up and dissolve the oil, and it gets sucked back up. They need pretty high-pressure steam, over a 1000 PSI, etc,. And, there’s a lot of things that molten salt reactors can
fit in – but I don’t really have time. The oil sands allure – it’s always been around in Canada, long viewed as an ideal proving ground, you don’t need a turbine and that’s
30 to 40% of your capital costs already, you don’t need R&D for a new turbine. Little joke here – but ask the South Africans – They had to develop for their pebble-bed [reactor]
work. They had to develop a whole new turbine, that was costing as much of more than the
entire nuclear program. And, of course these reactors would be
used in a remote situation. Many studies have shown that the nuclear produced steam is cost-effective
for the oil sands use. What scared me at first, is well –
these old studies had pretty low $/watt for the nuclear, but the the cost of the natural gas
systems had risen even faster. And, oil sands producers expect to
pay 200 billion dollars on carbon taxes over the next 35 years, and those funds are mandated to be
spent on clean-tech solutions. So, there is quite a great source [of funds for this development]. So why not conventional nuclear power? Basically, a study pointed out that the facilities are too large. We just can’t pump the steam around wide enough to use it, the pressures are too low and not flexible, and the steam just can’t be pushed around far enough. Ideal size is 300 to 400 MW thermal for a 30,000 barrel per day facility. Other potential problems [with conventional nuclear methods] – mainly [are all] about the the steam pressures are too low. We can talk about that later. The basic idea is using the (MSR) molten salt reactor combined with SAGD. We produce steam [at a] much higher temperature than needed, so
we either save money by doing it at lower temperatures, or you use the top end of that steam tor electricity generation, for
generating hydrogen by thermal chemical or high temperature electrolysis, etc., because is there is lot of money that needs to be
spent on upgrading of the bitumen on site. Bottom line, there’s there’s a massive
amount of oil available there. The molen salt reactors could help us get that out. Basically, We want to get off oil. Oil sands can help molten salt reactors come to being, And, with time molten salt reactors will bridge
not needing that oil in the first place. But, we could get north america off foreign oil pretty easily. Very quickly, on the Canadian pieces I’m working on –
trying to keep things simple as possible – I’ve got a big network of
connections around the world etc. [I’m] working with a group that were going
speak here but they couldn’t make it down – very bright guys, worked on how to
integrate this [technology] into oil sands. Biggest news though, is the interest of a
large Canadian based engineering firm which is – it’s not really a super-secret –
some people might even guess here – they don’t really want to
publicize, their they’re dotting their i’s crossing their t’s, probably
publicize soon, so please don’t push them that much. But, efforts led by ex-AECL member who
headed advanced reactor studies, they’re hiring a team and working
out collaboration agreements with me. We’ve been working towards a
consortium including the McMasters and University of Ontario institute of technology – our
two largest engineering schools – along whith Chalk River labs and most of those
talks have been going very good. It’s amazing the the level of interest
that we’re getting in the university system, and, of course the University of Saskatchewan,
Saskatchewan has their own great interest in small module reactors,
and of course, Oak Ridge. The CNS – I won’t really go into it – bottom line is, that’s our version of the NRC, talks of them have been very encouraging. And, they have changed their system
to be streamlined for small module reactors. And again, the government
of Saskatchewan is very interested. Conclusions, By any standard, molten salt reactors are
superior to all other offerings, not just by marginal improvements. They were mandated to be breaders but I really think the
simplified converter options appear to be an obvious route forward, at least for
the short term. It takes large and far-sighted investment
but potential returns are enormous. And, all factors do seem to be pointing
toward and ideal focal point of a broader North American effort to
realize this for the world. Because we’re not going to do this without the US,
without Oak Ridge expertise, etc., but I think we can do it a lot easier
a little north of the border.

9 thoughts on “David LeBlanc – Molten Salt Reactor Designs, Options & Outlook @ TEAC4

  1. @18:48: "No allusions that licensing a new reactor design will not be a huge challenge both for the vendor and CNSC."
    I'm still trying to work out the multi-layered negatives in this sentence. Plus it all hinges on whether he really meant "illusions" instead of "allusions"… Anyway, in general it sounds like the Canadian gov't is more interested in thorium and/or modular reactors than the USA. So that's promising.

  2. Go, Canada GO!
    DMSR is very interesting, not 'perfect' like the proposed LFTR, but its simplicity can't be underestimated. Anyway the most pragmatic first step in building MSRs

  3. Oil sands' $200 Billion carbon taxes over 35 years, mandated to be spent on cleantech is the most interesting part. siphoning just a fraction of that into MSR/LFTR should be a great start.

  4. Stop polluting the environment. Can't you just build some solar and wind powerplants? Im sure you can! Do so! 😉

  5. I don't know why this guy tried to speak as fast as possible to zip through all this. Would've been better to take his time to explain everything in more detail. And also he just read powerpoint slides.. We can all read the slides if we wanted, we don't need them read to us.

Leave a Reply

Your email address will not be published. Required fields are marked *