A critical analysis of future nuclear reactors designs

Once upon a time I used to be a fan of nuclear energy. As far as I saw it, nuclear energy was the silver bullet solution to all of our energy problems and more. However, the more I’ve learned about the industry the more critical I’ve become.

Notably the fact that most of the economic figures in support of nuclear power (a couple of typical delusions you’ll find here and here) come straight out of Hogwarts school of magic, wizardry….and economics (more realistic appraisals of nuclear economics can be found here and here). There is the question about the world’s limited stockpiles of fissile material, not helped by the fact that the LWR reactors that make up the bulk of our present capacity are ridiculously fuel inefficient, as in they only actually burn 2-3% of the fissile material present.

And what are we planning to do with all this waste? Various proposals have been made, but no nation on earth has yet to comprehensively solve this problem. Then there’s the glacially slow build rate of reactors, and of course, the nagging issue of nuclear safety.

But is there a better way?

Of course some supporters of nuclear energy would say that all of the problems I’ve just listed off relate to our choice of large light water reactors (as Richard Black at the BBC recently discussed). They claim that alternative designs would result in much safer reactors that are cheaper to build, easier to build and ultimately produce less nuclear waste. Various alternatives to the LWR have been proposed, these include:

High temperature gas Reactors , “modular” Pebble bed Reactors , the advanced CANDU reactor, so-called “fast” reactors and the Molten Salt reactor (MSR).

But could these reactors actually supply us with something better? In the following series of article below, I explored this question by subjecting these designs to a critical review.

The Mega LWR “death spiral”

But first of all what’s wrong with these large LWR’s? I explore some of these issues in part 4 of my little appraisal. Basically it all boils down to a fateful decision taken back in the 1950’s. The US government was in a race to get nuclear reactors up and running for military use, notably for the submarine fleet. A light water reactor was an obvious choice for a compact power source and one that could be developed reasonably quickly. When the civil nuclear industry in the US got going the corporations took these naval reactors, which in many cases they had themselves designed for the military, and simply scaled them up. There are a multitude of reasons why this decision to use mega-LWR’s in preference to anything else was taken (again I review them in part 4), but cost and ease of development were certainly key. But regardless of the “why’s?” the fact is that the nuclear industry did embarked on this plan and in the process of doing this the nuclear industry essentially laid a trap for themselves.

While the submarine reactors had outputs of between 15-60 MW_e the civil nuclear industry began building 500-1,600 MW_e behemoths. These large “megatron” LWR’s were scaled up to the point where they became inherently unsafe – if the cooling system for any reason failed, the reactor would go into meltdown. This meant the cooling systems and all backups related to it (including its backup power generators) HAD to work perfectly i.e. critical system components.

Unfortunately several accidents since then, notably TMI and Chernobyl, revealed flaws in the original design. The only way to correct these flaws was to include further safety systems, as well as by building a large concrete containment dome over the reactors to contain any radiation releases. The end result has been the size and scale of nuclear projects has ballooned in size, as has the costs of new nuclear build (the following video offers a humorous if foul mouthed appraisal of the situation regarding the Olkiluoto reactor in Finland, first of the new (don’t laugh) nuclear renaissance). All these safety critical components also need careful testing prior to commercial operation, meaning the pace of new nuclear construction has slowed to a crawl. Fukushima will now likely lead to another round of recriminations, further expensive upgrades, redesigns and a further round of reactor shutdowns.

Inevitably I therefore see the civil nuclear industry, so long as LWR are favoured as being caught in a never ending death spiral of further mishaps leading to redesigns and costs rises, which leads to reduced orders, which spreads the fixed cost of nuclear over a smaller number of reactors, which raises the cost yet further. All the while these design changes are slowing the pace of build down (leading to yet more cancelled plants), undermining the entire case for nuclear. Indeed its inevitable now that both the US and Britain will now see a major reduction in nuclear energy use in the next few decades (a recent Bulletin of Atomic Scientist’s article http://bos.sagepub.com/content/67/4/30.full discusses this), simply because there is no way they could now build reactors fast enough to cope with the rate they are about to go offline, nor indeed train the staff to run them (many in the nuke industry are getting old and will be looking for their bus passes pretty soon!) Inevitably, as has already happened in Germany, Italy and Canada recently, beyond a certain point cash strapped governments will just run out of patience, pull the plug and turn off the life support.

Criterion of Success…or failure!

In my analysis I established the following criteria with which to judge the relevant “fit for purpose” strengths of these reactor designs.

• Cost, Any alternative to the LWR must be cheaper. Nuclear energy is already more expensive than renewables at current prices, nevermind future prices. So if nuclear has a future its overall costs must be lower.
• Safety, As I said before, the LWR has numerous inherent safety flaws. The number one barrier to public acceptance of nuclear energy is safety. Argue all you want about it, but the LWR design amounts to an elaborate attempt at trying to make a silk purse out of a sow’s ear. Our preference would therefore be for a reactor that is not just safer, but inherently safer.
• Fuel efficiency, the global stockpiles of fissile material are limited. We could probably maintain the existing stock of reactors going for 50-80 years or so, but given that they only represent 5% of global energy output, that leaves us with the question of where does the other 95% of our energy come from and the obvious question as to whether nuclear energy is just more trouble than its worth. Better fuel economy would mean more reactors and greater market penetration.
• Reduced nuclear waste, the elephant in the room for nuclear energy is the ever growing waste mountain. We’ve yet to come up with a comprehensive solution to nuclear waste and until we do the argument of environmentalists is “if you’re in a hole, stop digging!”. So needless to say if the reactors we now review can generate a lot less waste that would make them a much more attractive proposition to the LWR. Obviously, if the opposite proves to be true, that’s a potential black mark against them.

In addition I also looked at the ability to use the Thorium cycle (given the limits of Uranium supplies), scalability of reactors (these “mega” LWR’s are just too big and unwieldy and can play havoc with the gird of many smaller nations) as smaller reactors might be more flexible, as well as the idea of modular design and mass production of reactors. This latter 2 points being discussed in part 10 of my little series.

If we can prove that any of the reactors we examine can tick all (or most) of these boxes then maybe the nuclear industry has some future, beyond its current Zombie walk to the grave routine with LWR’s.

The Verdicts

All in all my conclusion is that the case for future Generation IV nuclear reactors is much narrower than the supporters of nuclear energy would have you believe. While they do offer some advantages over LWR’s, notably in the area of safety, his comes with strings attached, notably higher capital costs. This is largely a result of the fact that many of these would need to be built from much more exotic materials, such as high temperature stainless steel alloys, Nickel alloys or Refractory materials, while the predominant material of choice in current reactors is steel (stainless and forged ferritic) and concrete. This materials requirement is itself an issue related to the high temperatures these alternative reactors would be required to operate at, not to mention the more aggressive and corrosive environment in some of them, notably the MSR proposals. Of course one to question whether these higher construction costs (and in some cases higher decommissioning costs) are justified.

But overall it is my conclusions that:

The CANDU does close off some of the safety loop holes associated with LWR’s, but it opens up a whole slew of new ones too and generally means higher rates of fuel consumption, lower thermal efficiency and increased amounts of nuclear waste being generated. Indeed, the Canadian government may well have exhausted its patience on this one, as they recently sold the CANDU reactor business for the bargain basement price of $15 million, as well as writing off several billion in outstanding debts. Not exactly a vote of confidence! To me it seems to be a case of the Fed’s picking up the CANDU and throwing it in at the deep end of the pool to see if it will sink or swim. I’ll leave you to guess what’s most likely to happen!

The High Temperature Gas Reactor (HTGR) offers an order of magnitude improvement in safety as well as potentially better fuel economy and high thermal efficiency. However, it will likely come at the expense of much higher construction costs (and probably a slower construction rate depending on material choices, which again depends on operating temperature), higher decommissioning costs and possibly higher volumes of nuclear waste (that last point I’ll admit is debatable, see the my post for more on that one). While the HTGR is fairly safe from meltdown scenarios, one would have likely weathered the Fukushima tsunami with minor damage, it also opens up a host of other safety issues, notably the potential fire risk associated with that graphite core (again a debateable point, see my full article here on this for more info).

The Gas cooled Fast Reactor (GcFR) offers the intriguing possibility of being able to transmute stockpiles of nuclear waste into less dangerous forms. However, it comes with a rather hefty price tag with a lot of R&D work still outstanding as the design is only in the early concept stage of development (read we don’t know if it even works yet!). In any event it will not eliminate the need for some geological storage facilities given the length of time it would take to develop and then build a sizeable number of said reactors, not to mention store the waste after its passed through the reactor. This, plus the hefty price tag associated with GcFR’s, could well make the whole idea uneconomic. Also the GcFR comes with some safety issues (it is not nearly as safe as the HTGR) and a severe proliferation risk.

The Molten Salt Reactor (MSR or LFTR) does offer a number of unique options in terms of safety improvements and improved fuel economy, plus reduced waste streams. However, its ability to achieve these goals is often heavily overstated by its supporters. Much like the GcFR above the design is at a very early stage in development, with much research into it abandoned back in the 1970’s. Any MSR reactor and its associated Chemical Processing Plant (CPP) would likely be expensive to build and slow to construct (again given the narrow and exotic nature of the materials choice the design enforces on us). Getting a decent thermal efficiency out of the plant might be problematic, which worsens the economic case for them. Also while certainly safer than a LWR in terms LOCA scenarios, the MSR comes with its own particular safety problems, notably that graphite core (fire!), the risks of a leak of radioactive material out of the CPP, or arguably worse a release of potential toxic and highly lethal fluorine gas. So all in all there may be a case for MSR’s, but its unproven at the moment and likely a much narrower case that its supporters would have you believe.

Indeed probably the biggest enemy of the MSR design is its own nutty cheerleaders who badly need to stay off the Kool-Aid. Casing point, without hours of my analysis article going online they were already running up vast blog strings of flaming trolls galore (see comments section of my page) or starting e-mailing me anonymously with various badly typed swear-word filled comments. I even picked up one or two stalkers trying to find out who I was and where I lived (yes really)! You also see the odd comment involving half baked megloamanic schemes (such as burning off the biosphere for uranium). While the best I can tell, the advocates of the other reactor designs I reviewed seem to have taken their punishment “like men”, the MSR fans reached for the tinfoil hat and the two-litre bottle of kool-aid. I shall leave it to the reader to decide who should be taken seriously!

Small to medium sized modular reactors do offer a good deal more flexibility in terms of how nuclear power could be used and yet a further improvement in safety. However, they also comes with lower economies of scale and thus higher construction costs and worse a slower rate of reactor roll out (at least in the early days). We could claw back on these two issues by mass producing said reactors in large volumes but as I point out (again see the full article), it is far from proven whether that would be economically viable and whether there is in fact a market for large numbers of small reactors.

Also as I outline, the case for small reactors would also require a major shift in public opinion, which post-Fukushima is unlikely to be forthcoming. Most of the reactor designs I’ve mentioned above would be wholly unsuitable for “mass” production, only a handful of PWR, BWR and HTGR designs would be feasible options. Worse still, by and large mass production means “dumbing down” our design, and that means accepting a reactor that’s much cheaper and easier to build but has a lower thermal efficiency, a higher rate of fuel consumption and ultimately produces larger volumes of nuclear waste compared to our “mega” reactors. With the exception of a small number of narrow cases, it’s difficult to envisage how this would offer an improvement on the current status quo.

Decommissioning costs, the Elephant’s still in the room!

Not only are the construction costs of many of these proposed reactors higher, but for some (but not all) the decommissioning costs would actually be higher and worse they will generate more nuclear waste from this process. This being a particular problem for graphite cored reactors such as the HTGR and the MSR. Other Graphite cored reactors are proving to be something of a nightmare to decommission, as I discuss in the section on HTGR’s.

As far as the spent fuel waste is concerned, some of these proposed reactors will indeed produce less, but others will actually produce more of it, thought it’s probably important to clarify what we mean by “more” or “less”. For example, CANDU as I point out, produces about 7 times (by mass) the amount of nuclear waste than a LWR. However, I’m quite sure the CANDU supporters will point out that because the waste from a CANDU is less radioactive it can be packed up much more tightly, reducing the size of any waste storage pen (but can it be packed sufficiently tightly to overcome that 7 times greater output?).

At the other end of the scale the HTGR’s have a very high rate of fuel burn up, and so would produce a lot less nuclear waste (pound for pound) than a LWR. However, the waste from a HTGR is contained within a graphite matrix which increases its volume to a much larger size than LWR waste. Hence one has to question which reactor we can claim “produces less waste”.

In a similar vein some of the waste output from a MSR will be mixed up with fluoride salts, from which it will have to be separated before going into long term storage. Disposal of said wastes have been described as “technically challenging”  although certainly doable. It’s estimated that it’s going to cost some $130 million to process the waste from one tiny 8 MW_th test reactor which ran for just over 5 years. Again it begs the question which reactor can truly claim to have the “smaller” waste footprint and the “cheaper” clean up bill.

Thorium….only for comic book heroes?

The Thorium cycle, as covered throughout my little study, does offer the option of solving some of the long term fuel supply issues surrounding nuclear energy. But the level to which it will do this is fairly narrow, as Thorium fuelled reactors still need fissile isotopes, drawn ultimately from Uranium, for startup purposes. Failing this they require the use of expensive (and generally uneconomic) fast reactors and reprocessing of spent fuel. So yes, while Thorium could help stretch things out, it can only help a little bit, but not nearly as much as the supporters of Thorium reactors would have you believe. Thorium fuelled reactors would still generate substantial quantities of nuclear waste and come with a number of potential proliferation risks attached. Even the UK National Nuclear Laboratories (NNL) pours cold water over the idea.

Brayton Cycle and Hydrogen Production….rumours of Rankine’s death have been greatly exaggerated

A proposal common to all Generation IV reactors, and some renewable power plant proposals (notably geothermal), is to use Brayton cycle instead of the Rankine cycle for power generation. This would offer a substantial improvement in terms of energy efficiency, and furthermore could bring down the costs of installation. However, there is still some work to do on this issue, so I won’t write off the Rankine cycle just yet! Similarly, the higher material limits required to raise reactor operating temperatures up to the level necessary to utilize the Sulfur-iodine process and make hydrogen directly (using the reactors heat) could well render the whole idea uneconomic. If we want hydrogen (from nuclear) that badly, build a reactor with a lower operating temperature out of cheaper materials, generate electricity and hook it up to an electrolyser! Less efficient yes, but likely cheaper. And if we really want hydrogen on the cheap, ditch the reactor and use CSP or wind energy!

Fusion?

Finally, I also had a look at Fusion power . This is the great white hope of nuclear energy and it has to be said we are making progress, but it’s a case of slow and steady progress. Indeed I would question whether we are in a position yet to even estimate how long it will take for fusion power to become commercial available…if indeed ever! Recent news from ITER is not positive, its now not due to go online till 2026, which would imply a completion of experiments in 2046. And it will take sometime beyond that before we wind up with a viable working commercial fusion reactor. As I speculate (here), it would likely be the latter half of this century (or the beginning of the next one) before we start to see Fusion play any sort of major role in mass global power generation. Also the first generation of Fusion reactors will be dependant on supplies of Lithium for fuel, of which there is only a limited global supply available, something that limits the amount of energy which can ultimately be generated from Fusion reactors, probably to between 8-20% of global energy use depending on whose figures you believe. Where does the other 92-80% come from?

And of course we have to contemplate the possibility that commercial Fusion energy never arrives. While speaking personally, I still have confidence that the necessary breakthroughs will be achieved according to a reasonable timetable, it would be foolish to blindly assume that they will. To build any nations energy strategy on the forlorn hope that fusion power will arrive on the scene by a certain date, makes about as much sense as selling your house and all your worldly goods because some preacher told you the world was going to end on a particular date.

Curb your enthusiasm!

All in all, my conclusions are that the case for future Generation IV nuclear reactors is much narrower than the supporters of nuclear energy would have you believe - even the case for Fusion doesn’t look that clear cut! And again I would note that this last point about Fusion is important, the way the nuclear energy supporters (and indeed many politicians and members of the public) go on you’d swear Fusion was already a slam dunk. Nothing could be further from the truth!

Nuclear energy supporters need to curb they’re enthusiasm for nuclear energy and accept that due to the high capital costs of reactor construction and the limited fuel supplies it will always only ever be a small bit player in a big energy market, at least as far as the current century is concerned. It currently generates about 1.9 – 5.1% of global energy (depending on how you do your maths) and I don’t see how it can be expanded beyond that level, indeed if they manage to maintain this level I suspect they’ll be doing well.

Even the most optimistic nuclear energy program we can draw up still has a substantial energy gap and something else will have to fill it. This of course means we’ll need to rely on renewables for substantially more energy than we currently get from it. Which means many nuclear energy supporters need to overcome their pathological hatred of renewables and if they are truly serious about combating climate change (as many claim to be) then they need to quit trying to throw the baby out with the bath water.