As a rough guide 1 gram of antimatter produces about 40 megatons of TNT, but costs presently around $60,000,000,000,000 to make, or $1.50 per kilogram of TNT power. It may become a source of power in decades or centuries time but is not viable at the moment.
So, we come down to the next source of power in its output compared to size, fusion power. So far, the record for a fusion reaction is about 100 seconds, most being 30 seconds or less, a bit short of the 50 million seconds of an average fission-based reactor by a factor of 5 hundred thousand. The published record for a fission reactor is 77 million seconds, not 100.
So far nobody has got a tokamak fusion reactor to actually produce anything except for very short periods, and always below the level of its inputs. The successes being counted when they can get them to work rather than running for more than a minute. Theoretically it has worked as long as you pick and choose what inputs and outputs you allow to count. 14 megawatts input over say half an hour, 30 seconds of 10 times higher output than input, 14-megawatt input after those 30 seconds for a few minutes. Net inputs 1000 times net outputs. Plus, usage of tritium that costs £22,000 per gram.
There was a report from the US department of Energy that in the 41 years between 1955 to 1996 225kg of tritium had been produced in the US, but it needs nuclear reactors to produce it, and 2/3rds of it had naturally decayed by the end of that time. If not used another 2/3rds would decay in another 40 years, leaving possibly 35kg. A typical start up for a fusion reactor uses up about 0.3kg of tritium or £300,000 each time. The hope is that the lithium containment for the reactor would produce the next amount of tritium after reprocessing. This of course mean having some way to remove the lithium while the system is still running, otherwise you need to start up again, using the tritium you produced.
We now have the latest papers that claim a fusion reactor is now a likely proposition as the theory has now been worked out, except that the key point of actually managing to make it a practical proposition and a working model being left out and totally unproven. The key is the inputs and sustaining the reaction. Considering that the first proven man-made fusion reaction other than using a thermonuclear bomb was in 1974, we are now at 46 years later and the record for the longest fusion reaction is 101.2 seconds. If you took the input compared to output while the reaction was produced then it produced more than its inputs, but over the whole course of the experiment this was not true, needing more inputs than outputs. A bit like clean power, everything before this point and after is discounted and irrelevant.
The next problem is that fusion is not the absolutely clean unlimited source of power that is claimed. You don’t want to be anywhere near one when it is running, and to keep the process sustainable on-site reprocessing of materials would have a similar problem to normal nuclear reprocessing plants to recover the tritium produced, usually in lithium for re-use. The technology is immensely costly, as is the raw materials.
The temperatures needed to run such systems need an internal level of 50-100 million C to be practical, so having that sort of containment, the level at the centre of a hydrogen bomb, is not the sort of thing you would want on your doorstep. Usually it’s only a small amount that is fused, so the reactions are limited, but a large sustainable reaction is still a major problem, especially if there is a catastrophic failure when the system is scaled up for commercial use.
All in all, I think it is likely that viable fusion power is still 10 years away, as it was 70 years ago, and quite possibly will be 10 years away in 70 years’ time. The old adage that ‘There is no such thing as a free lunch,’ meaning you can’t get something for nothing, there is always a cost, is still true, not mirroring the scientific belief by some that the equivalent of nuclear perpetual motion machine is possible.
It was the earlier belief with unlimited free power with nuclear fission, the actual end cost being higher than alternative fuels, and in some cases the massive cost forgotten or ignored in cases like Chernobyl and Fukushima, negative balances in an industry that will take decades to centuries to recover, so long as other examples of assumed due diligence and lack of respect for the processes don’t take place in the form of equating heightened risk and reduced safety in personal monetary terms. So far, no nuclear accident has been paid for by the people or companies who run the facilities that produced it, just by the people who live around it.
There is a lot of re-working the theory with fusion, much of it being new ways of thinking about it, claiming just one more push and we will get it. But his has been true for 70 years of just one more push, ‘this time we’ve got it right,’ or ’the new theory means it will work,’ when in truth the practical side of the theory always lets the theory down or proves it to be wildly inaccurate in the real world.
I would say that somewhere near $2 billion a year is spent on fusion research and has been for the past 50 years, so $100 billion spent for actually producing nothing so far. ITER, the latest system to be implemented will cost around $25 billion, but estimates by the US Department of Energy suggest the actual figure may be as high as $65 billion. The intended system is only thought of as an experiment, so real life running costs would be a fraction, but the test run is designed to produce a ten-fold advantage producing 500MWatt of energy for 20 minutes at a time from 50MWatt of input. At the current average market rate of $70 per MWh this is about $12,000 of electricity, a bit short of the $65 billion input by a factor of 5 million times smaller, so the efficiency of scale needs to be improved by 500 million percent to break even.
Fission works out to about 100 tonnes of material costing $2,770 per Kg, so $277 million that would run for 4 years based on complete renewal. During that time it would normally produce about 1GW for about 85% of the time over this period, which comes to 35,000GWh of value of $2.400 billion, so a proven 8.7 fold advantage.
Both systems have ancillary costs that bring down the true advantage, ignoring the real life use of them making them highly impressive, but mostly theoretical and impossible to achieve. If you added the whole cost of things like Chernobyl and Fukushima to the overall cost of nuclear fission then the average generation costs wouldn’t be about $80 per MWh but probably nearer $500 per MWh. Mainly because the plants generating the disasters that affected their populations, had people that could not afford or were given adequate restitution, also hit the rest of the world and were never held to account, but dismissed mainly in entirety by their respective and friendly governments. Basically, given the all clear to nuclear plants in the world for removing consequences of bad decisions, inadequate safeguards, or compromisable designs, ones where the only risks calculated are to return on capital, balancing theoretical safety/cost/return figures.
Sadly, because power production is a business, it’s undertaken for profit, the business is as open to the cowboys as any other business. And there are a lot of businesspeople who would take any amount of risks, as long as it’s not to themselves, only to other people. The ethos of business is provision at lowest cost, the ethos of government should be keeping its population safe. But, too many people in government see their role as provision at lowest cost, regulation and regulators being too in bed and open to political or commercial pressure to be even adequate for the purpose. 10 deaths will cost this amount, 100 this amount 1,000 this amount, which do we choose considering we can get away with 10,000 if it’s not close by or there won’t be any repercussions from over friendly local governments we can buy off. So the object of business is not the middle as in this chart, it’s how far along the blue line past the meeting point can we get away with? This they will try to push as far as possible, and even when there is a disaster the politicians form protective groups among their kind rather than demanding restitution for the havoc or damage created among their own countries.
Nuclear power, fission or fusion could give the most environmentally friendly source of power above all. But business and governments have proven beyond any doubt they will abuse the provision and incur totally unacceptable risks as there is no risks to themselves personally or financially. Get in, mess up, get out fast. The modern business school of take the money and run.
In the UK we have the chance to be power secure with resilience built in, but I’m afraid that political motivation, profit and poor management catering to naïve environmentalist viewpoints has rendered us up to catastrophic system failures, and economic and social disasters that could push us to third world, or as far as medieval standards of life and technology.
Covid-19 itself has shown creaks and cracks in the millions of interconnected and interrelated systems, itself being a relatively minor problem. The likelihood is that within the next 100 years we will see events that are major in comparison, when the systems will fail due to ‘resilience’ and ‘backup’ being seen as a swearwords among current businesspeople and government. ‘Somebody else will always cover it or sort it out.’ Example ‘We have contracts to supply ships, shame they don’t actually have any. But the contract is there so we’re safe.’ But there won’t be anybody. So it is with fusion, ‘this time it will work.’ And other areas, ‘This is clean energy.’ ‘Storage is not required.’ ‘Big expensive batteries will stop outages.’ ‘It’s cheaper if you amortise, normalise or discount figures.’ As long as you ignore everything that we say doesn’t count. Shame in the real world everything does and in the present.