Nuclear Power Won’t Save the World

It Won’t Even Help

San Onofre nuclear plant (awnisALAN, CC-BY-SA 2.0)

Originally published at Green Energy Times.

By George Harvey

An old paradigm

The electricity most of us use comes from a system that was designed mostly over a hundred years ago. It was built around concepts that benefited customers of that time. It started with baseload power plants with transmission lines carrying the electricity to towns and cities where customers lived.

A baseload power plant is designed for efficiency of scale and operation. In those days, that meant it had to be as big as possible. Since any ability to ramp power output up or down quickly would cost a lot extra, the plants were designed to have constant output. With constant output, a baseload plant had be sized to meet a demand that could be counted on always to be there. This is the base load, the lowest load that the grid would ever have over the course of time.

Since the baseload power plant was designed to cover the lowest load, any amount of electricity that would be in excess of that would have to come from other sources, all of which cost much more to run. They were load-following plants and peaker plants.

Baseload power plants were sited based on cost and access to resources they needed. Typically, they went up on inexpensive land at some distance from the market they served. They had to have access to fuel resources, which often meant that they needed their own docks or rail sidings. Also, they were often placed on bodies of water to take care of their cooling needs, which are great, because only about a third of the heat they produce could be used to generate electricity.

Originally, baseload plants mostly burned coal. When nuclear reactors were brought online, starting in mid-century, they fit right in with what was the current paradigm of the time. The difference was that they produced nuclear waste instead of air pollution and carbon dioxide.

We might note for reference here that when the state of Vermont was looking for a contract to replace electricity it had been getting from the Vermont Yankee (VY) nuclear plant, the owner of VY made an offer that they said the state could not refuse. It was the equivalent of 6.5¢ per kilowatt-hour (kWh). The state immediately found cheaper renewable electricity.

A new paradigm

By contrast, today the least expensive source of renewable power need not be large. Solar panels operate at the same efficiency whether they be in utility-scale arrays or on a residential roof-top. Significant amounts of electricity can be generated by solitary wind turbines.

Of course there is a statement, “The sun doesn’t always shine and the wind doesn’t always blow,” which happens to fall into a range of unintentionally disingenuous to simply deceptive. The amount of electricity coming from a given solar array is really rather predictable and tends to come best in periods of light winds. And wind turbines do best when the sun is not shining brightest, so they compliment each other. But more to the point, while a single wind turbine can be idled in calm weather, the wind never stops blowing over wider geographical areas.

We might ask whether the problem of variable output of wind and solar power is as big as the problem of inability of baseload power to follow loads. The answer to this can be seen in the relative costs of electricity from load following and peaking plants, on the one hand, and batteries, on the other. We could do a detailed analysis of this, but it is really not necessary because the utilities are showing the results of their own analyses.

A number of utilities are replacing plants powered by natural gas, which includes most load-following and peaking plants, with solar arrays and batteries. In one case, Entergy Mississippi is planning to replace older natural gas plants with solar and windpower. In the case of Entergy Arkansas, a combined-cycle (base-load) natural gas plant it had planned will not be built, and the company will build renewable resources instead. (KATV.com)(www.bit.ly/EAs-new-plan)

Comparing nuclear power with solar+storage

An article in PV Magazine in August compared the cost of two new nuclear reactors with a combination of solar photovoltaics (PVs) and battery storage that would replace them functionally, as dispatchable power sources running full time.(www.bit.ly/PV-Mag-compare) The article is titled, “Solar challenging nuclear as potential climate change solution.”

The author, who had some expertise in systems that include solar+storage (S+S), used actual costs for the Vogtle reactors that are being built in Georgia. The two reactors, which have been under construction since 2013, are expected to come online in 2022 and 2023, at a cost of roughly $30 billion, including $3 billion in finance costs. Their capacities will be 1,117 megawatts each.

The PV Magazine article calculates the cost of a solar array big enough to provide the same output as the nuclear reactors in the winter in Georgia. It assumes battery storage to supply the output of the nuclear plants for 16 hours, increased by 10% to be safe.

The author shows that the cost of the S+S system designed to replace the two new Vogtle reactors would cost a little less than $17 billion. That would represent a saving of about $10 billion, not counting finance costs.

While that sounds impressive, the article fails in a number of respects. Here are some:

Output of the S+S system is calculated to be the same as nuclear in the dead of winter. The nuclear plant’s output will be constant year round, but the S+S system will produce far more electricity nearly all year than in the dead of winter. The value of the extra electricity from S+S is not accounted for.

The cost of the nuclear plant does not include the backup systems it requires, but the price calculated for S+S does.

The load-following and peaker plants used to work with nuclear power, are slow to react to demand changes. By comparison, battery backup can respond nearly instantly, making it far more valuable.

Nuclear waste is an unsolved problem that the U.S. government guarantees, at taxpayer expense. The same is true for insurance, which is covered by the Price-Anderson act. S+S systems do not have comparable costs.

The author does not take into account Wright’s Law, a recognized law of economics referred to as “the learning curve.” It suggests that construction of a battery system of the size envisioned would be sufficient to drive the cost of storage down quickly enough to reduce the cost of the S+S system itself.

Electricity from new nuclear facilities is very expensive. It becomes far cheaper once the system is paid down. Please refer back to the bid from VY, of 6.5¢/kWh. By comparison the cost of electricity from S+S is very low. A report from February, 2020, which appeared at S&P Global, “Falling US solar-plus-storage prices start to level as batteries supersize,” says that power purchase agreements have dropped into the range of 3¢/kWh to 4¢/kWh. (www.bit.ly/Falling-SS-price) But the costs of solar, wind, and battery systems keep falling. According to the U.S. DOE’s National Renewable Energy Laboratory, in an article published at CleanTechnica, the costs of S+S systems declined by over 12% from the first quarter of 2020 to the same quarter in 2021 alone. (www.bit.ly/SS-dropping-costs)

Nuclear as an answer to climate change

There are some who feel that the nuclear industry may have a way to become relevant in the new “small modular reactors.” An article on this appeared in the October, 2021, issue of Green Energy Times, “When It Comes to Nuclear Power, ‘Advanced’ Isn’t Always Better.” (www.bit.ly/Advanced-nuclear) It explained that rhetoric around these reactors seemed to be unrealistic and achievable timetables were not able to help when we need most to address climate change, which is right now.

I would suggest that nuclear industry numbers about costs, timelines, and safety have historically been far off the mark, a problem that those promoting newer types of reactors have not addressed at all. In fact, it is almost as though the industry has three types of numbers.

There is one type that is simply correct, but it only relates to results of simple calculations.

A second type of number is one that relates to such things as the cost of a reactor or the time needed to build it. These seem very often to be off by a factor of 2. If a reactor is expected to take five years to build and cost $6 billion, it is probably best to bet that it will take ten years and cost $12 billion.

The third type is safety analysis calculations that can actually be checked have historically been off by an order of magnitude. Given the types of reactors that have operated commercially, the safety analysis made on them, and the time they have been running, we should probably have had one commercially operating reactor experience a partial or full melt down worldwide since commercial nuclear plants first started delivering energy. Instead, we have had eleven – that we know of.

All told, we might say that putting money into nuclear power goes beyond being a monumental waste. It detracts from the overarching issue of dealing with climate change by making that money unavailable for dealing with the problem using less expensive, more reliable energy that can be built far more quickly.