Nuclear Fusion Research Performance Oversold

Technical Power Generation Is Overstated

Recent headlines have touted nuclear fusion research at ITER that will get ten times as much energy out as it puts into the process when completed. The problem is that this is the estimate for only the core of the process, not for energy generation of the power plant a whole.

ITER's new work is a great improvement. It is the highest efficiency nuclear fusion power generator ever built. The process as a whole, if it performs as expected when completed, would produce about 57% of the energy going into it (if we use the heat to generate electricity rather than raw heat), compared to 1-10% from the best previous contenders set back in 1997 or earlier. 

ITER would do just better than break even in generating raw heat, if it works as expected, but usually that isn't what you want a nuclear fusion power plant to do.

They need to get the 10:1 ratio they are talking about if it performs as promised, to about 18:1, to start generating net electrical power for the process as a whole.

Sabine Hossenfelder's blog, Backreaction, has more of the technical details.

It Also Has To Make Economic Sense

Also, in order for generating nuclear fusion power to make economic sense relative to alternatives like solar, wind, tidal power, and nuclear fusion, it needs to do much better than barely generating more power than breaking even. The net power produced has to be great enough to justify the immense cost of the facility and the research and development going into making it possible.

You need to get the cost of the power plant annualized by amortizing of the construction costs over its useful life, plus operating costs, including operating staff salaries and some minimal fuel expenditures, down low enough that the cost of generating electricity is less than about 12 cents per kilowatt hour, for nuclear fusion to make economic sense. This is $120 per megawatt hour.

People have penciled out the costs estimated for much a larger scale operation with a construction cost of about $3 billion. They estimate that operating costs would be  $25.20 per megawatt hours (about 21 percent of the going price of electricity). 

Assuming for sake of argument that this is correct, the depreciation and interest must be not more than $94.80 per megawatt hour, about $41.71 per megawatt hour on interest, and $53.09 per megawatt hour for construction costs.

Over 40 years, a power plant generates up to 350,400 megawatt hours per megawatt of power generation at any one time. So, the construction costs need to be less than $18.6 million per megawatt hour of electricity generation capacity. This would have to be $18.6 billion for a power plant that can generate 1000 megawatts of net electricity output.

We are getting close to getting to the point where this is technologically feasible, but what ITER is doing now won't get us there, even if it works.