We've covered the argument that nuclear is not a climate-friendly energy source before, but here's new evidence -- a paper from the Irish sustainability thinktank Feasta, titled Why Nuclear Power Cannot Be a Major Energy Source:
The advantage of nuclear power in producing lower carbon emissions holds true only as long as supplies of rich uranium last. When the leaner ores are used - that is, ores consisting of less than 0.01 percent (for soft rocks such as sandstone) and 0.02 percent (for hard rocks such as granite), so much energy is required by the milling process that the total quantity of fossil fuels needed for nuclear fission is greater than would be needed if those fuels were used directly to generate electricity. In other words, when it is forced to use ore of around this quality or worse, nuclear power begins to slip into a negative energy balance: more energy goes in than comes out, and more carbon dioxide is produced by nuclear power than by the fossil-fuel alternatives.
Add to these concerns the carbon costs of building, maintaining and securing reactors, and shipping, processing and storing (for thousands of years) the waste created, and the carbon benefits of nuclear energy completely disappear, and, indeed nukes start to look not only dangerous but downright polluting. Or so, at least, the argument goes, and it does seem like the various folks making it are pretty credible.
more from the report:
A Lean Guide
Nuclear energy could sustain its present minor contribution of some 21/2 percent of global final energy demand for about 75 years, but only by postponing indefinitely the expenditure of energy that would be needed to deal with its waste.
Each stage in the nuclear life-cycle, other than fission itself, produces carbon dioxide.
The depletion problem facing nuclear power is as pressing as the depletion problem facing oil and gas.
The depletion of uranium becomes apparent when nuclear power is considered as a major source of energy. For instance, if required to provide all the electricity used worldwide - while clearing up the new waste it produced - it could (notionally) do so for about six years before it ran out of usable rich uranium ore.
Alternative systems of nuclear fission, such as fast-breeders and thorium reactors, do not offer solutions in the short/medium term.
The overall climate impact of the nuclear industry, including its use of halogenated compounds with a global warming potential many times that of carbon dioxide, needs to be researched urgently.
The option that a nation such as the United Kingdom has of building and fuelling a nuclear energy system on a substantial and useful scale is removed if many other nations attempt to do the same thing.
The response must be to develop a programme of "Lean Energy". Lean Energy consists of: (1) energy conservation and efficiency; (2) structural change to build local energy systems; and (3) renewable energy; all within (4) a framework, such as tradable energy quotas (TEQs), leading to deep reductions in energy demand.
That response should be developed at all speed, free of the false promise and distraction of nuclear energy.
Image: the recent implosion of the Trojan nuclear plant cooling tower. Video here.
(Nice grab, Dave!)
I would want to see some hard numbers before accepting such apparently preposterous assertions. Fission produces about two million times as much energy per unit mass as combustion of fossil fuels. The author, therefore, seems to be implying that it takes more than two million tons of fossil fuel to extract one ton of uranium from sandstone containing .01% uranium. Do you believe that?
Likewise, some numbers concerning known reserves of uranium ore and consumption rates would be handy, as would some backing for the claim that alternative systems of fission do not provide a short to medium term solution. The alternative systems require no new technology and can be built as quickly as standard nuclear reactors, so I know of no reason other than lack of political will that would prevent them from coming into use quickly.
What about thorium breeder reactors? Extraction of thorium is easier, as it is more plentiful.
Whether or not nuclear plants can be be operated and maintained indefinitely, with a significant positive net yield of energy, with safety, waste management and proliferation solved - there's still the crucial element of time.
According to the best climate science we have, we need to act very quickly. We need to make significant reductions in our greenhouse-gas emissions in the next ten years.
So let's think of this as a race. In one starting block, there's nuclear energy. In another block, there's improvements to energy efficiency, and in a third starting block, there's renewable energy such as wind and photovoltaic.
Ready set, GO!
The data I've looked at tell me that efficiency wins the race handily, renewable energy comes in second, and nuclear places a distant third.
That's not to rule out the potential for nuclear energy. My basic instinct is to be extremely skeptical, but I'm wary of making a knee-jerk judgment. No one really yet knows if we can run our civilization without "dense" sources of energy, yet also avoid a planet-wide Easter Island, a hideous destruction of natural biota and habitat. Climate disruption or destruction of significant biota look like even worse problems than nuclear. The greatest danger of nuclear energy may be in a false promise of salvation.
If you accept the best science on climate disruption, and you decide that nuclear offers no long-term solution; and if you ponder where that leaves us - then you better have exceptional bladder and bowel control.
Getting energy from uranium is like getting ethanol from marijuana. There are side uses to the product that might be disturbing to some.
Maybe they're considering the amount of time that the spend fuel rods will be monitored for the 100,000 years after they're used? ;P
I mean security lighting, cameras, and keeping those fences nice and shocking takes up a lot of energy!
You Know there is already a supply of highly enriched fissionable material that I would love to see used to generate electricity......
If we built nuclear power plants that used fuel from nuclear weapons I would change my opposition to nuclear power.
This avoids the mining and enriching steps and it would allow the US to start living up to its obligations under the NPT to reduce the number of warheads.
Solar power could easily meet the energy needs for security cameras, lights and electric fences. Not including nuclear energy in the energy mix would be like going deep sea diving with scuba gear and an oxygen tank for 1 hour and expect that you could make it last 4 hours with conservation. You can only get so far with conserving your oxygen.
In response to "Occam's comic": As of September 2005, under the Megatons to Megawatts program, the USA and Russia have already converted over 10,000 nuclear warheads into low enriched uranium for use in nuclear power plants. This is a well documented program, and is planned to continue until 2013.
I think I generally agree with the position Dave Foley has taken. He says that to reduce greenhouse emissions we should favor energy efficiency first, renewables second and nuclear third. I think that's a sensible position to take, without casting nuclear as irredemably evil.
I've been in the uranium biz for more than a decade and write often on energy subjects. A few points to ponder:
The energy required to mine uranium is small next to the energy extracted. Use some common sense: The uranium mining companies and nuclear fuel processors have to pay for the energy they consume. That cost is incorporated in the cost of the fuel. If the fuel costs more to produce than the value of the energy it yields in the form of electricity, no one would generate nuclear power. Nuclear fuel is not subsidized. It represents approximately 35% of the generating cost of nuclear power. The uranium component is a fraction of this 35%.
As to the energy inputs for mining when the ore grade drops, there are some dubious assumptions here. About a third of the world's uranium comes from Canada, where grades are running from 1.5% to 30% or more. Another 25% or so comes from Australia, where the grades are much lower, but a lot of that material is co-product with copper and gold. Energy costs are spread over the other metals. African uranium is also low-grade, but there it's more a question of labor than energy inputs. Finally, much of the world's low-grade product is produced by in-situ solution mining techniques (U.S., Uzbekistan, Kazakhstan, a little in Australia), where the ore is never pulled up and ground in a mill. Hence a relatively small energy input.
Occam's comic says that recycling bomb fuel means you don't have to enrich the product. Not necessarily so. For the Russians at least, the bomb fuel is highly contaminated with unwanted isotopes. It has to be blended with depleted uranium (which is isotopically cleaner than natural uranium), that itself must be re-enriched to about 1.5% U-235 before blending.
The Russians put in almost as much enrichment effort to make the blendstock as the blended fuel actually contains.
Another point: While uranium enrichment is energy-intensive in the U.S., which uses an obsolete gaseous diffusion technology, that's not true in Russia, which uses energy-efficient centrifuges (about 5% of the power required by gaseoous diffusion). Within a decade or so, everyone will be using centrifuges.
A final point: The economics of wind and solar are still discouraging. Without huge subsidies, neither has a chance in the market place, and even with subsidies, both lack the consistency of output to compete with conventional generating stations, nuclear or fossil.
In Germany, where wind is really big, 15% of grid capacity is windmills. Sounds good, but it produces only 3% of the power. In other words, you have an average capacity factor of 20% from your wind mill (versus about 90% with a nuclear power plant).
The only thing cheap about a windmill is the wind--everything else is appalling. The machines themselves cost $1,500 to $2,000 per kilowatt of capacity (about the same as a modern nuke plant, by the way), but you have to build four times as much capacity to have a comparable power output. Plus you have the added costs of transmisison lines (which can easily match the cost of the mills). Plus you don't really know what the output will be because you don't know in advance how hard the wind will blow, or for how long. So you can't dispatch it reliably in a power market system. Plus there are limited places where it makes sense to build them, and it's usually a long way from the load.
The economics of solar electricity generation are even worse. Only a portion of the spectrum can be captured by solar cells. There's a band gap problem. They are only good for a limited number of hours each day, and a limited number of days each year. The power is DC, and converting it to AC is a huge power drain. Plus the cells themselves are costly, and they only make sense when allied with a costly battery storage system.
Conservation is always the cheapest source of power, even if that sometimes means a lowered standard of living. However, when comparing renewables like wind and sun to nuclear, the fair question is whether you want to see electric rates stay about the same or be multiplied by a factor of at least three or four -- because that's what you'd be paying if you had to rely on renewables alone. Sobering, but there it is.
Robert, you're decrying subsidies for wind and solar. I don't see you decrying subsidies for nuclear energy.
Why is that?
Some of your assumptions about wind and solar seem way off, too.
If we pursue efficiency improvements to the extent economically justified and technically possible, then we're not necessarily too worried about electric rates because our electric bills will be low.
We of course don't want energy directly - we want the services that power sources provide. By focusing on obtaining those services in the most efficient and elegant manner possible, we can go further, faster, than with a widespread expansion of any power source, including nuclear.
In fact, pursuing efficiency (not lifestyle diminishment) with our best full measure might reduce our energy "footprint" to the point where much of our existing energy infrastructure: natural gas plants, cleaner-coal technology with carbon sequestration, existing hydropower - and some existing nuclear plants - could be a significant part of our energy mix. A biofuels infrastructure might not need to be much bigger than the current brewing and distilling industry. Wind energy is coming close to being competitive with fossil fuels, even without subsidy. (Fossil and nuclear fuels are of course also heavily subsidized - no nation makes energy policy on the basis of mere economics.)
Worldwide power consumption now exceeds 13 terrawatts. Try to come up with a new infrastructure to provide that without fossil fuels - one that's a sure bet, can be built with existing available investment capital, and can come on line fast. The single most-readily-available power source we have is embedded within our single largest industrial product: waste. Not addressing that with everything we've got is insane. You don't have to be pro or anti nuke to agree that no matter our power sources, it's criminal to use them wastefully. Let's keep an open mind and really listen to the data regarding nuclear; but let's have an open mind and take a good hard look at the potential to improve efficiency by at least a factor of 5 to 10.
An old german friend of mine worked on nuclear reactors and calls people who think of renewable energy as "greenies". Guess that meant that technically, nuclear energy isnt really green. I understand that the reason why nuclear energy is becoming unpopular is the high capital cost plus high operating cost plus high exposure to fuel risk. With all these in mind, the rest of the world cannot continue using it. What we need is cleaner energy solutions that come in sizes that every nation could afford because really, the problem isnt localized. Climate change is a global issue.
From this Hayes mining-engineering textbook chapter I get the electricity requirement to crush hard rock to pass 80 percent through a 100-micron screen as 25 kWh per tonne. 0.00005 percent uranium yielded from the rock as a result, and burned in a CANDU reactor, would give that much electricity. So this part of the money quote --
... When the leaner ores are used - that is, ores consisting of less than ... 0.02 percent (for hard rocks such as granite), so much energy is required by the milling process that the total quantity of fossil fuels needed for nuclear fission is greater than would be needed if those fuels were used directly to generate electricity ...
is incorrect. Milling can be done electrically, and at 0.02 percent uranium, a hard rock has enough uranium to produce electricity to mill ~400 other rocks of equal hardness and weight.
--- G. R. L. Cowan, former hydrogen fan
Boron: fire without exhaust gas
Joseph Willemsen: I didn't say anything nice about subsidies to the nuke industry either. I think it's mature enough to make it on its own merits--or not. Moreover, the goverment monies going into the civil nuke sector have little bearing on operating expenses/profitability for existing nuclear plants. The lion's share of government money going into the nuke area is R&D for advanced future reactors. Frankly, I see most of this as money down the john: The Department of Energy is worse than the military when it comes to boondoggle science projects.
A new subsidy program of several hundred million dollars has just been set up to help kick-start construction of new nuke plants. The problem being addressed here is that nobody wants to be first: The fifth or sixth plant in a series will almost certainly cost perhaps 10%-20% less than the first, due to learning curve effects. But if nobody's first, no one ever gets to be fifth. The subsidy is aimed at reducing the price difference -- and the planning risk -- between being first and being fifth or sixth.
I'm not saying I like it, but at least there's a defensible rationale. The subsidy would be limited in scope and limited in time frame.
Wind mill subsidies are a different story. Since 1992 there has been a production tax credit worth $15 per megawatt-hour generated. And that's just one of many overt and hidden subsidies that support the wind industry. (BTW, $15 per megawatt-hour just happens to be about the average of the total cost of operation, maintenance and fuel for U.S. nuclear electricity.) I submit that there is a whale of a difference between an operating subsidy that goes on substantially forever and a one-time kick-start of limited scope and duration.
You say you think my numbers are way off for wind and solar. Tell me: What do you think is the installed cost is per kilowatt of capacity for your renewables? And what do you think the capacity factors actually are? I think you'll find that to get a reliable output of signficiant size (say 1000 MWe), your capital costs will be a multiple of the cost of a new nuke plant.
Reden: I tend to agree that nuclear is not "green," but then again I wouldn't call it "black" either. It's somewhere in between. Its great advantage is that it completely contains its own waste stream. What we do with that waste stream is another matter, but we have decades to find the most acceptable solution.
In fifty years of operation, US nuke plants will produce about 60,000 tons of spent fuel. By contrast, the coal industry produces about one BILLION tons of waste EVERY YEAR, most of which is thrown into the atmosphere. That's the scale difference we're talking about. (And BTW, coal plants throw out more radiation than nuke plants and uranium mines combined. Why? Because there are small amounts of uranium (and radium) in much of the coal we burn. Some goes out the stack, some goes out with the ash. Nukeplants don't have stacks, or ash piles trucked out the door each day.)
Reden: You're right that nuke plants are capital-intensive to build. However, they are not costly to operate. Far cheaper than gas or oil, and about 10% cheaper than most coal-fired plants (depending on location). The only thing that's much cheaper and cleaner than nuke plants is large-scale hydro, and that's not an option that environmentalists have been backing lately.
Don't get me wrong--renewables make perfect sense in niche markets, and maybe larger markets in the years ahead if major techological breakthroughs occur. Meanwhile, to suggest that wind and solar are an alternative to large base-load generating stations--that's a reach, to put it politely.
Where I have a big problem is the way "renewables" are oversold as being a reliable source of large quantities of cheap energy. So far, renwables are not capable of operating on a large scale, they have not (and possibly can not) produce power predictably, and they most assuredly are not cheap.
A lot of people who deem themselves enviromentally responsible seem to be completely irresponsible when it comes to confronting the actual economic realities of wind and solar. When the few who seem acquainted with the numbers are compelled to concede that we're talking about an uneconomic proposition, it's always "Wait five years and it will be cheaper then." Well, I heard that five years ago, and five years before that, and five years before that.
Finally, I think it is disingenuous in the extreme to poll people about whether they'd rather have new nuclear plants or "renewbles" without revealing what the economics of available renewables actually are. As the Bush Administration has demonstrated so aptly, making policies without knowing the facts leads to very, very costly mistakes.
I wonder if the critics of this analysis have even looked at the research on which it is based (www.stormsmith.nl). The authors seem to take into account all of the objections that the critics raise here.
For instance, the authors assume that 70% of refiners worldwide use the energy-efficient centrifuge enrichment process, even though the current usage rate is nowhere near that. Their figures on the energy used to mine, mill, enrich and fabricate nuclear fuel are drawn from U.S. government sources.
Their estimates of world uranium reserves in the earth's crust are based on the known distribution of ore grades in the uranium reserviors of the world. They note that some sources, like the World Nuclear Association, say that abundant rich mines will be discovered in the future, even though that is a speculative view based on economic arguments rather than geological evidence.
One of the most striking points is the profitability and EROEI of conventional nuclear power is dependent on never properly disposing of wastes or decommissioning and disassembling power plants. In light of that, the continuing inaction on a permanent disposal site in the U.S. is a lot more understandable.
Laurence: You said: "For instance, the authors assume that 70% of refiners worldwide use the energy-efficient centrifuge enrichment process, even though the current usage rate is nowhere near that."
Who told you the current usage rate is "nowhere near that"? In fact, 70% is an excellent estimate. Current world-wide SWU deployment (the measurment unit for uranium enrichment activities) breaks down as follows: France: (GDP) 8 million SWU. U.S. (GDP) 5.5 million SWU. Urenco (UK, Germany, Holland) 8 million SWU (centrifuge). Russia: 22 million SWU (centrifuge). China (estimate) 1 million SWU (centrifuge). (Japan and Brazil have small programs, both centrifuge.)
In short, the only programs that are not centrifuge are in the U.S. and France. The French plant has a nominal capacity of about 10 MM SWU, and is running at about 80% of capacity. The U.S. plant has a nominal capacity of about 10.8 MM SWU, of which about 8 MM is deemed physically operable, and about 5.5 MM actually operates. Moreover, both the U.S. and the French GDPs are slated to be replaced by centrifuges over the next 7-8 years.
BTW, thanks for the link to stormsmith. It makes for interesting reading, but it is not well-informed.
Again, use your common sense: The mining companies have to buy their energy inputs. That cost is rolled into the cost of uranium. The fuel processors have to buy their energy inputs, too. That cost is also rolled into the cost of fabricated fuel. If the energy going in was greater than the energy coming out, the utilities would be paying more than they are receiving. Obviously, the reverse is true or they wouldn't be in business.
I don't care how good you think the authors' research is, or how impeccable their sources: If it leads to an absurd result, it is not good research (or it's good research and an absurd analysis).
Here are some real-world cost/production figures that (in terms of order of magnitudes) may provide a useful perspective.
A "generic" 1000 MWe reactor contains about 30 tons of enriched uranium fuel. With annual refueling--again, speaking generically--you'd change out 1/3 of the core, or about 10 metric tons of fuel. Thus, you may safely conclude that one year's worth of power output is derived from 10 metric tons of enriched uranium fuel.
The average capacity factor for a US nuclear plant is now about 90%. Therfore, in the span of a year, the plant would be expected to produce 24 times 365 times 0.9 times 1000 megawatt hours, or about 7.9 million megawatt hours of electricty.
10 metric tons of enriched fuel equals 10,000 kilograms. 7.8 million MWh deivided by 10,000 means that each kilogram of fuel kicked out enough heat to generate 790 megawatts of electricty. (The heat output is actually about 3 times that, but the thermal efficiency of conversion to electrcity is only about 33.3%.)
So, we now have 790 megawatts of electricity for each kg of fuel. It requires approximately 10 kilograms of natural uranium to make one kilogram of eneriched uranium fuel, so now we can conclude that each kg of natural (unenriched) uranium is responsible for 79 megawatt hours of produced electricity.
Even in the grossly inflated market conditions that prevail in the uranium market today, one kilogram of uranium (in a form ready for enrichment) costs $123. (Four years ago it was about $30, but that's another story.) At an average wholesale electricty rate of $45 per megawatt (which is pretty cheap these days), 79 megawatts will fetch the power company $3,555. (Its actual cash cost of production from its nuclear plant (which excludes amortization) would be on the order of $1,200 for those 79 megawatts.)
This should give you some idea of the "energy leverage" you get in uranium when turning it into electricity.
The huge increases in uranium prices, I can assure you, have nothing to do with the cost of energy inputs in the mining process.
By far most energy intensive step in producing nuclear fuel is enrichment. Even in the most energy-intensive process (gaseous diffusion, as used in the US and France), it takes about 2.2 megawatt hours for one separtive work unit (SWU), and it takes about 6 SWU to make one kilogram of enriched fuel. Thus, each kg of enriched fuel reflects 13.2 megawatt hours of investment -- but you're getting 790 megawatt hours back.
I don't know where those Dutch fellows are getting their information, but it sure as heck is not from anyone who has anything to do with actually producing nuclear fuel for a living.
Their report smacks of a long tradition in the antinuclear community of trying to prove that nuclear power makes no economic sense. The chronic failure of most of these activists to be able to discern real-world information, use basic arithmetic, and reach intellecutally defensible conclusions explains why they are not taken very seriously by people in the nuclear industry or government regulators.
Robert, those "Dutch fellows" are quite clear about their sources and provide a 7-page bibliography for their report. It is available on their website. As I said, they use U.S. government numbers for the energy required to mine, mill, enrich and fabricate nuclear fuel -- notably ERDA 76-1, A National Plan for Energy Reseach, Development and Demonstration: Creating Energy Choices for the Future (1976) by the U.S. Energy Research and Development Administration (ERDA).
If you know of errors in the ERDA numbers, or reasons why they should not be used, I think Jan Willem Storm van Leeuwen and Philip Smith would like to be informed. And so would those who have read the study and find it worth consideration, at the minimum.
Again, use your common sense: The mining companies have to buy their energy inputs. That cost is rolled into the cost of uranium. The fuel processors have to buy their energy inputs, too. That cost is also rolled into the cost of fabricated fuel. If the energy going in was greater than the energy coming out, the utilities would be paying more than they are receiving.
That's assuming all energy costs the same per BTU and no subsidies exist.
Robert's remark assumes correctly that non-uranium energy inputs are not positively subsidized; it does not assume that no negative subsidies exist, and in fact they very significantly do. Fossil fuel net income is a mainstay for Western governments.
Nor does it assume all thermal megawatt-hours, before tax, cost the same; it assumes only that the input ones aren't significantly cheaper than the outputs.
Making money by burning expensive fuel to produce very cheap fuel implies a very high energy output-to-input ratio, and that is indeed the situation of nuclear fuel miners and preparers who use fossil fuels. In current CANDU practice a kilogram of unenriched uranium yields 164 thermal MWh per kg, so the mining companies get 71 US cents per MWH. Such natural gas as they use has a pretax cost near US$6.80/mmBTU, that's US$23/MWh.
After getting their 71 cents, the miners pay tax; they get no subsidy. If government did give them a break, or even spot them a few pennies, it would in so doing be depriving itself of tens of dollars in fossil fuel tax revenue.
That would be socially responsible government behaviour; the reponsible officials would be selfless public servants.
--- G. R. L. Cowan, former hydrogen fan
I see - so now you're making the "government is addicted to oil taxes" argument to explain... what? And you're referencing ridiculous OPEC charts that mislead on how much money their industry makes to try and make a case for nukes?
Probably not the smartest rhetorical path to head down.
I believe the OPEC charts are not misleading. They seem to explain why speed limits are little enforced.
An interesting paper: http://arxiv.org/PS_cache/hep-ph/pdf/0501/0501111.pdf
I believe the OPEC charts are not misleading.
I'm sure you do. Why else would you link to them?
They seem to explain why speed limits are little enforced.
What percentage of government revenues in the US come from gasoline taxes? And how much of that revenue is dedicated to paying for transportation? Surely you must know if you're going to make that kind of statement.
Let me know when you get the numbers.
There's an article summarising a new, much more attractive technology for nuclear power based on thorium in the Australian science mag site, Cosmos (www.cosmosmagazine.com).
I'd not heard of this type of technology previously, and am particularly attracted to the prospect of it burning old nuclear waste as past of the electriticity generation process.
William Wood: In principle, thorium is a promising nuclear fuel, and there are several ways to deploy it.
In the early civil nuclear days, the US Govt developed several fuel designs that combined uranium and thorium. Indeed, the now shuttered Indian Point-1 unit, 50 miles from NYC, was orginally designed as a uranium-thorium reactor.
India is putting a lot of effort into building a thorium-fueled system. India is very short of uranium, but has thorium coming out if its ears. Rich stores of thorium in monazite sands are literally lying around on the surface. Miles and miles of it.
The Indian approach forsees construction of fast neutron reactors (cooled by liquid metal)and fueled largely by plutonium. (The plut comes from reprocessing spent fuel from their heavy water reactors.) Thorium fuel elements loaded into the fast reactors would then "breed" more fissile isotopes than the reactor consumes.
When thorium-232 is bombarded with netrons, some of it is converted into uranium-233 and uranium 232. The thorium fuel elements would then be reprocessed and the uranium isotopes separated chemically.
U-233 is a spendid fissile isotope -- and (IMHO) a huge proliferation risk. It's evem more missle than Pu-239, and you don't need an implosion device to detonate it. A compact U-233 warhead is simplicity itself. Scary -- and I'm a serious pro-nuke.
U-232 is a real nuclear ugly, however. It has a half life of 68.8 years, and decays into Thallium-208. Thallium-208 is a hard gamma emitter, which means any fuel containing U-232 requires rather elaborate handling precautions to keep it from poisoning the fuel workers, transporters and reactor operators. It can be done fairly easily, from an engineering standpoint, but it does add to costs.
Fast reactors present major engineering challenges as it is -- which is why the US., the British and the French have all abandoned their programs. (Russia is continuning in a half-hearted way, as is Japan.)
Moreover, the fuel reprocessing steps entailed in a thorium cycle genererally entail generating waste streams that are a major and costly challenge. While the fast reactors can help in coping with spent uranium reactor fuel (burning up or transmuting various spent fuel isotopes, reducing the heat, radiation load and longevity of waste that will be interred), bear in mind that other waste streams were gnerated to pull off this particular magic. Althoughy engineers love this stuff, commercial people take a very dim view indeed.