Seven Things You Thought You Knew about Nuclear Energy

But it might be time to re-think

By Matthew L. Wald

We are in the decade of nuclear renaissance, at least in terms of designing and planning. The 2030s could be a time of rapid deployment. But after years of coasting on the engineering and construction boom of the 1960s and 1970s, the industry has been out of mind for a while, and when it is in mind, the perception isn’t always accurate.

Here are some common misconceptions:

Nuclear Energy is a technology for making electricity

Reactors start out making heat. In today’s designs, heat boils water, which makes steam to spin a turbine, which turns a generator and makes electricity.

But one of the two flagship projects of the Advanced Reactor Demonstration Program is for a gas-graphite reactor that will make steam that could be used directly for chemical manufacturing. The new reactor is the Xe-100, and it is being built at a Dow plant in Texas that currently uses natural gas to make steam. If the goal is decarbonization, replacing methane burned to make steam for industrial use is as valid as replacing methane burned to make steam for electricity.

The Dow project isn’t a one-off, because only about a third of the global warming gas output is from the electric sector, which is where most nuclear energy is used today. Carbon-free electricity can address some of the rest, by substituting for motor fuel like gasoline and diesel, or by heating houses. But steam itself can run factories, and that can come from nuclear too. China is using steam from an AP-1000 nuclear reactor to heat a town, formerly heated with steam made by burning coal, and the Russians have been building reactors to do that for decades.

Many planners are also looking for ways to use nuclear energy to split water molecules to produce hydrogen. Nuclear-produced hydrogen would have direct industrial uses and would be cleaner than hydrogen made by splitting methane molecules. It can also be used to make electricity, turning it into a storage medium.

“Advanced Reactor” means “Small Modular Reactor”

Nuclear Generating stations of various types started out modestly sized, but grew in the 60s, 70s and 80s as operators saw economies of scale. Now, after a long gap in construction, proposed new plants are smaller. There are three motivations:

·       Smaller reactors can be built in a factory and shipped in just a few pieces, reducing cost and improving quality. That method would also reducing construction time by allowing reactor fabrication and building and infrastructure to happen simultaneously.

·       Smaller reactors can replace old coal plants of about the same size, re-using their water supplies and grid connections.

·       Smaller reactors can be added to meet smaller increments of demand growth and can be installed in countries where the power grid isn’t strong enough to handle the starts and stops of a bigger unit.

But there are also trends in the opposite direction. Owners of existing reactors have been on an aggressive campaign of “de-bottlenecking” their plants, looking for investments in turbines, generators, pumps and other equipment that will let them squeeze a few extra megawatts out of vintage plants.   The NRC says that it has approved applications for “uprates” equal to six gigawatts and has applications for another 2.5 gigawatts. (A large reactor is in the range of one gigawatt.)

And two recent designs are well outside the “small modular reactor” category. Two copies of the Westinghouse AP1000, which produces a little over one gigawatt, are now operating in Georgia. Four more are operating in China and more are likely to be built in the U.S. and abroad. The “AP” stands for “advanced passive,” with greater reliance on natural forces like gravity and natural heat circulation, and less reliance on pumps, valves, and operator actions. The reactor is not as big a departure from the 1960s and 1970s designs as some newer advanced designs, but it is a more advanced reactor.

And TerraPower’s Natrium project, in Wyoming, is intended to churn out a steady 345 megawatts. (The Energy Department’s definition of “small” is 300). It can turn out 500 megawatts at peak times by tapping into stored heat, and can reduce output during hours when other sources, like solar, are plentiful.

The nuclear industry can’t lose; it is backed by giant utilities.

Some giant utilities run nuclear plants, but they also operate solar farms (which they try to locate in places visible from the road), wind farms, and plants that run on gas or coal. To some extent, they are like parents who have lots of children and insist that they love each one equally. (They may not actually do this, but they still insist that they do.)

Some companies don’t think that nuclear energy fits in well with their corporate image. NextEra, which is the holding company that owns Florida Power & Light, runs ad campaigns filled with wind turbines. It is extremely good at building wind farms and collecting the investment tax credit. As part of its campaign to remake itself, NextEra dropped out of the nuclear industry’s biggest trade association.

And the connection of nuclear reactors to parent utilities has also been weakened by the trend of companies that owned one or two reactors to sell them off to fleet operators. Take the case of Indian Point, on the Hudson River north of New York City. Consolidated Edison, which serves New York City and Westchester County, built unit 2 and unit 3, and sold unit 3 to a state agency, the New York Power Authority when it ran into financial difficulties. Neither Con Ed nor the Power Authority was a very good operator, and the reactors suffered from frequent shutdowns.

Each entity sold its reactor to Entergy, which operated a fleet of reactors in the South and decided it wanted to make a profit from its expertise in reactor operations. Entergy also bought Vermont Yankee, which was owned by a group that included many small New England utilities.

Indian Point had been through many difficulties, including problems meeting the Nuclear Regulatory Commission’s emergency planning rules, imposed after the Three Mile Island accident of 1979. One of the counties within the 10-mile emergency planning zone said it would boycott the required emergency drills to force the reactors to close. But the State of New York, owner of one of the reactors, stepped in to substitute for the county in the drills, saving the reactors.

But later, the politics changed, and the then-governor, Andrew Cuomo, campaigned against the reactors when their initial 40-year licenses approached expiration. Entergy had combined the two reactors into a single plant, sharing resources, and improving their reliability. But it lacked the political muscle of Con Ed or the Power Authority, and eventually it gave up, retiring the plants.

In Vermont, the state legislature voted to force Vermont Yankee to close but lost that battle when a court ruled that only the Nuclear Regulatory Commission could do that. But Entergy, an entity foreign to New England, eventually gave up there too, and the reactor was closed. 

In addition, the utilities are looking for low risk and steady profit. That is something that existing reactors provide, but they are less certain about new ones.

The Utilities are Eager to Solve the Nuclear Waste Problem

Some are, some aren’t. Financially, it’s not their problem.

The reason is that the Federal government required the reactor operators to sign contracts with the Energy Department, under which the government promised to start accepting the fuel in January 1998 for permanent disposal. The utilities paid a fee, which was a tenth of a cent per kilowatt-hour that they generated. The money went into the Nuclear Waste Fund, established in 1982.

The Energy Department was supposed to adjust the fee to assure enough money to meet the program’s needs, but after the Yucca Mountain project stalled, the utilities sued, arguing that there was no program to fund, and the courts agreed.

The fund has a balance of about $48 billion.

The utilities also sued for their extra costs because the government hadn’t lived up to its contract obligations. They moved fuel to dry casks, which have low maintenance expenses but still require monitoring and guarding. The courts ruled that this was a contract dispute, and the government had to pay. The money comes out of a government account called the Judgment Fund, which is the same fund that would pay to fix your car if you were rear-ended by a government vehicle.

The catch is that Congress automatically replenishes the fund, without appropriation votes. The federal government has paid out about $7 billion to the utilities.

Companies that are developing advanced reactors would like the problem solved because the lack of a repository, even an interim one, are a perennial complaint by opponents. Where reactors are running, fields of dry casks are a minor add-on to operations, and don’t cost the owners any money. Where reactors are shut down, owners are eager for a solution. But for many companies, a de facto solution is in place already

“Baseload” is a virtue

In fact, it is better described as an attribute.

Electric generating stations divide into three categories, but nuclear is about to move to straddle two of them.

Traditionally, utilities have one category that is expensive to build but cheap to operate, and those run flat out, 24/7. They meet the lowest level of demand that is present around the clock, so are called base load. Today’s plants are in that category.

A second category is “mid-merit” plants, which start up in the morning as demand rises, and back off in evening as it falls. These have moderate operating costs and moderate construction costs.

The third is “peaking” plants, which are cheap to build but expensive to run, but which run only a few hundred hours a year, during high demand periods.

But the installation of variable renewable generators has scrambled this system. Wind production peaks at night, sometimes cutting into the “baseload” category so that coal and nuclear plants have had to reduce production. Solar arrives on sunny days and reduces the number of hours that mid-merit plants run. Peakers usually must run anyway, because in most cases, peak demand is at an hour when the sun is very low in the sky, or down.

The new push in reactors, though, is plants that can easily adjust their output. NuScale Power’s plants consist of a cluster of small reactors, some of which could be shut down for weeks at a time. NuScale’s design also allows all the steam to bypass the turbine and generator, going straight to the condenser, for instant reduction of power.

The Natrium project, in Kemmerer, Wyoming, features a giant tank of salt that can store energy as heat, and thus can vary its output with the turn of a dial. The plan is to build up the heat during the day, when solar is abundant, and tap it for electricity as the sun goes down.

Microreactors like Oklo’s Aurora Powerhouse, are meant to power a grid solo, meaning that they meet both baseload and peak demand on their own.

Advanced Reactors Are Better Because they are Safer than Legacy Reactors

Reactor owners don’t initiate conversations about safety, because doing so presumes, falsely, that there’s a problem. One nuclear advocate of my acquaintance argues that even making the argument is the equivalent of a restaurant that advertises that its kitchen is rat-free.

In fact, legacy reactors aren’t what new reactors should be compared to, because the choice of what to build next is not 1960s designs vs. today’s designs. And in any case, the risks calculated for either are exceptionally low.

The proper comparison is to competing alternative energy sources. Our World in Data calculated the number of deaths per terawatt-hour of generation. (For reference, the United States consumes about four terawatt-hours a year.)

Coal: 24.62 to 32.72, depending on coal type

Oil: 18.43

Biomass: 4.63

Gas: 2.82

Hydropower: 1.3

Wind: 0.04

Nuclear: 0.03

Solar: 0.02

As the Breakthrough Institute has previously noted, the Carbon Almanac puts coal deaths per 100 terawatt-hours at 2,462, vs 7 for nuclear.

Developers of advanced reactors make the point that the rather obscure metric used to measure safety, core damage frequency, is far better for the newer designs. But they are also better because, depending on precisely which design,  they can produce higher-temperature heat, which is useful in various industries; because they can gear up or down quickly, helping solve the intermittency problem introduced by solar and wind, and they can use a wider variety of fuels. 

Today’s Reactors are Going to Run for 80 Years, or Maybe 100.

The reactors were initially given 40-year operating licenses, which many people assumed was going to be their lifespans. Some didn’t last that long; some got a 20-year extension and many now have a second 20-year extension. A license isn’t a guarantee that the plant will continue to operate, any more than a car with a registration good for another two years is guaranteed to be on the road for that long.

To split atoms, the reactor must be in compliance with all requirements at all times. As with airplanes, that means that certain equipment can be out of service at any given time, but it must be fixed within a certain number of hours or days. For example, a plant with two emergency diesel generators can operate if only one is functional, but only for a limited time. Other equipment must be fully functional or the plant has to shut down.

At some point, the plant owners may determine that keeping the plant in compliance is not economically rational. The owners of Crystal River 3, in Florida, thought it was worthwhile to spend the money to replace the steam generators—giant heat exchangers that draw heat from mildly radioactive water and transfer it to clean water, which is boiled into steam and used to spin a turbine-generator. But the job required cutting a hole in the containment building to get the old equipment out, because the plant wasn’t designed with that repair in mind. And during that job, workers cracked the containment. With enough money, perhaps $1 billion, they could have fixed it, but utility managers placed a bet on the future price of natural gas, the alternative to nuclear in that part of Florida, and decided to retire the plant.

How long will today’s fleet of reactors from the 70s and 80s continue running? It depends on the estimated value of their electricity to the grid. Like my 2001 Honda Odyssey that simultaneously needed a new catalytic converter and a new transmission, the answer may be before the license expires. Nuclear is a relatively young technology and nobody is quite sure how long key components will last, although as a condition of license extension, operators have many procedures in place to determine their plants’ material condition.