Factual · Powerful · Original · Iconoclastic
In 2015, the United States Environmental Protection Agency (EPA) called for a reduction of 32 percent in the 2005 levels of carbon dioxide emissions.1 Administrators hoped to reach this particular El Dorado by 2030. The impact of their guidelines on anything beyond their satisfaction is apt be negligible. The guidelines affect only the generation of electricity, and by 2100, even full compliance would reduce global temperatures by a mere 0.018 of a degree.2
Not a very big deal.
The EPA guidelines were not written with global warming in mind. Nuclear power is already the greatest source of clean electricity in the United States and could largely eliminate fossil fuels by 2030.3 The guidelines seem designed to increase the use of solar and wind energy in the United States, though they will also increase the use of greenhouse gas (GHG)-emitting natural gas. But wind and solar energy do not provide completely clean electricity. They are variable renewable energy (VRE) sources, both low in concentration and intermittent. It takes resources to collect and concentrate them, and even more resources to make them available on demand.
In the United States, 62 nuclear plants with a total of 99 reactors already provide 19 percent of the nation’s electricity. Without them, annual CO2 emissions would be about 573 million tons higher than they are today.4
Wind and solar energy are not becoming competitive with other forms of electricity generation in the United States. Impressions to the contrary are based on flawed data. In its accounting, the U.S. Energy Information Administration (EIA) uses what it calls levelized costs. These represent the per-kilowatt-hour cost of financing, building, operating, and maintaining an electricity generation plant over its assumed financial life.
Levelized costs were designed to permit a comparison between all forms of electricity generation, but data for wind and solar energy, on the one hand, and other sources of fuel, on the other, do not necessarily correspond.
The nameplate capacity of a plant designates its production maximum under ideal circumstances. The capacity factor of a plant, by way of contrast, designates the percentage of its nameplate capacity that it realizes in day-to-day operations. Nuclear has an impressive capacity factor of about 92 percent in the United States, more than double what it was in 1972. This improvement is one of the unsung successes of nuclear power.5 The capacity factors for wind and solar energy are at the opposite end of the scale. According to the EIA, wind averages about 34 percent; the Global Wind Energy Council estimates it at only between 15 and 30 percent.6 The EIA estimates the capacity factor for solar energy at 28 percent; the industry itself gives a lower range of between 10 and 25 percent.7
These are interesting discrepancies between government and industry estimates.
Charles Frank of the Brookings Institution has ranked the various forms of energy generation in terms of the CO2 they displace. The clear winner was nuclear energy. It replaced almost six times the emissions of solar energy, four times that of wind, twice that of hydroelectric energy, and five times that of low-carbon gas.8 We can think of this another way. All energy takes energy to produce. The true cost of energy production is thus the difference between the energy invested in production and the energy returned from production. A recent German study found that nothing approached the returns of nuclear power.9 Hydroelectric power came second, then coal, then much further down, wind. Solar power actually consumed more energy than it produced.10
EIA comparisons do not look simply at current energy costs. All capital costs are amortized over a thirty-year period for all forms of energy. But this misrepresents the true costs of nuclear power. Nuclear power is by far the most expensive in upfront capital expenditures, in great part because of safety precautions. It is, however, extremely cheap in operational costs. Nuclear plants are meant to be permanent structures. Aside from routine upgrades, they require only maintenance. Fuel costs are extremely low.11
Nuclear power plants operate for longer than thirty years. An operating license for a nuclear plant in the United States runs for forty years. The Nuclear Regulatory Commission (NRC) has the authority to extend those licenses for another twenty years. It has already granted extensions to three quarters of U.S. plants and expects that eventually almost all will request them.12 The oldest U.S. commercial nuclear plant currently in use began operating in 1969 and is now licensed until 2029.13 The first nuclear aircraft carrier, the U.S.S. Enterprise, is only now being decommissioned after 55 years of service, its original twin reactors still running well.14
By comparison, wind and solar farms have far shorter lives. A wind turbine is a mass of moving parts subjected to constant friction. Go to a large wind farm on any given day, and invariably some turbines are not functioning.15 On average, they require maintenance every six months, and are not expected to last more than twenty years.16 Solar panels, usually equipped with a 25-year guarantee, do not simply quit or fall apart, but progressively produce less power.17
The reduction in the consumer price of wind and solar-generated electricity is not entirely a matter of technological breakthroughs. In the past, wind has benefited from the construction of larger and higher turbines, but at this point, no matter what alloy is used, metallic stress limits their size.18 Even the Global Wind Energy Council admits, “modern turbines are getting reasonably close to the theoretical limit of power one can extract from a stream of moving air.”19
And solar energy? It has gotten cheaper because the panels have gotten cheaper. But this is not from increased efficiency or improved technology. Banking on getting more energy from improved design, Solyndra infamously tanked.20 In the case of solar energy, what the lower prices really reflect is the declining cost of labor as production has been shifted to China and Taiwan.21 Those declines recently seem to have been halted by the imposition of tariffs by the United States and the restriction of imports by the European Union.22 In 2014, prices for solar panels began to rise.23 Solar installations in Europe dropped precipitously.24 Solar does have some potential to increase efficiency; higher conversion levels have been reached in the laboratory. But so far these have been far too expensive to market.25
A consideration of capacity factors is an exceptionally effective antidote to industry propaganda. The media and industry regale us with news of vast new wind and solar farms producing massive amounts of energy. And naturally they ignore the capacity factor issue, instead using the nameplate output. They assume bright sunshine and wind blowing at the perfect speed 24/7. The Berkshire Hathaway Topaz Solar Farm in southern California, which cost about US$2.5 billion, claims a rating of 550 MWh on its website and in press releases, but it has a 23 percent capacity factor and is actually generating on average only about 126 MWh.26
The largest wind farm in the United States, and second largest in the world, is Alta Wind Energy Center, which is also in southern California. Its website claims that, when completed, “[t]he project will supply 1,550MW [MWh] of clean renewable energy.”27
Its estimated capacity factor is thirty percent, meaning that its true output is just 465 MWh. The smallest U.S. commercial nuclear reactor generates far more electricity than Alta Wind Energy Center.28
Capacity factors for nuclear power plants approach 100 percent of nameplate output.29 Their generation costs are extremely low.30 The Palo Verde nuclear power plant in Arizona, with a capacity factor approaching 100 percent, produces about nine times that of the Alta Wind Energy Center—providing electricity for over 4 million people in four states.31 It does so at a production cost of only 1.33 cents per kWh, well below both fossil fuels and the cheapest fabricated prices for wind.32 The U.S. national average is 2.44 cents per hour, well below that of fossil fuels.33 Palo Verde also occupies six square miles while Alta spreads over 14 square miles. Despite its massive size, Alta claims to provide electricity to only 400,000 of the 115 million households in the U.S.34 Just 300 more Altas to go.
Rooftop solar panels are not economical either, even when compared to commercial farms.35 Energy from rooftop panels costs about twice that generated by solar farms, which enjoy an economy of scale. Nor does the planning for solar farms encompass issues such as the direction a roof might face, or sun-blocking objects such as trees or buildings.36
No technology can overcome the most significant drawbacks to variable renewable energy. To make up for their low capacity factors, wind and solar farms must be huge. They cannot be placed within cities. Instead, they must be located in the areas with the most wind and sun, regardless of where energy is needed.37 Land near major cities is inherently more expensive than land in the countryside. These two factors wildly increase real estate and transmission costs.
In the case of sunlight and wind, storage of energy becomes a crucial factor. Because sunlight and wind are intermittent, that energy must be stored, or other sources must always be available as backup. As the best locations are bought and developed, each subsequent site will generate less electricity per turbine or panel.38 Each new site will be further and further from customers, requiring ever-longer transmission lines.
Two of America’s three greatest metropolitan areas, New York and Chicago, receive little solar energy. They are located in some of the lowest states for solar capacity factors.39 The New York metropolitan area, with its 23 million inhabitants, is also outside the wind zone.
Chicago has lots of farmland fairly close by, but, windy as it is, it lacks for wind generation.40
Too pricy. Farmland close to Chicago, given its extraordinary fertility, costs about US$7,700 an acre.41 The land for a wind farm 60,000 acres in extent would cost about half a billion dollars. A small nuclear reactor producing an equivalent amount of energy, such as the Westinghouse AP1000 WPR Gen III+, the only type currently being installed in the United States, would require a mere five acres costing US$39,000.42
One significant but overlooked issue is the problem of transmitting electricity across enormously long high voltage lines. Part of the added cost is obvious: construction and maintenance. On average, transmission adds roughly 15 percent to the capital cost of building a wind project; delivering the power adds twice that.43 A thirty-percent surcharge is hardly inconsiderable. Take one real-world example: the SunZia planned wind/solar transmission line project.44 The line has been approved to extend as much as 550 miles, through part of New Mexico and Arizona, connecting solar and wind farms to yet more transmission lines, most ultimately feeding into California. Cost estimates are currently at US$2.2 billion.45
The price may yet rise further:
We may be talking US$2.8 billion for the transmission lines alone.
And more: the transmission line loss or bleeding means that the farther electricity is sent, the more of it is lost.46 Technology can only ameliorate the loss, even as those technologies themselves add cost. Iain McClatchie, in a blog posting with the revealing title, “The Terrible Cost of Moving Electricity,” provides a ballpark figure: “After 1000 miles, 8.71 percent is lost, and delivered power costs at least 19% extra.”47
Taking into account the cost of land for the panels or turbines, delivery costs of up to thirty percent for transmission line construction, and perhaps 19 percent more for a thousand miles of transmission lines, plus the cost of right-of-way under the pylons (normally four per mile), wind or solar power can turn out to be very expensive.48
None of this appears in EIA generation cost figures.
Any form of energy storage adds to capital costs. Energy is lost, first in converting to stored mode, and second in reconverting to electricity. The most favored system pumps water up into a reservoir during peak operating times, and later lets it fall back through turbines. Theoretically, superior systems may be possible, but, if the aim is not to be completely off the grid, storage has little purpose.
Researchers at Stanford University have concluded that it is better either to curtail production or to waste the excess and rely on other sources during intermittent periods.49 It turns out that, counting in the costs of even the cheapest storage, not only does solar power use more energy than it produces, but so does wind.50 Yes, it costs money to throw away energy. It just costs far more to store it.
Batteries are perhaps the worst in this respect. At a press conference in April 2015, Elon Musk unveiled two new battery storage units, called the Powerwall. Many in the technology media predicted the death of nuclear power.51 More sober analysts disagreed.52 Say that a conventional backup generator costs about US$1,000. Tesla’s 7kWh battery version, with additional required equipment and installation, costs about US$5,000; the 10kWh version, over US$7,000.53 America’s largest installer of solar panels, SolarCity, won’t even install the 7kWh battery, saying it “doesn’t really make financial sense.”54 The chairman and largest shareholder of SolarCity? Elon Musk. They will only install the largest unit as part of a package including solar panels. When told by a journalist that the batteries made no economic sense, Musk reportedly responded, “That doesn’t mean people won’t buy it.”55
As VREs make an ever-greater contribution to the power grid, their value must inevitably diminish. Explains Alex Trembath, a senior energy analyst at the Breakthrough Institute:
There are two forces at play. The first is the “merit-order effect,” in which VRE replaces high-cost power generation on the grid first, less-cost generation second, and on and on until the generation it’s replacing is very low-cost and so replacing it doesn’t create much or any value.56
The second, he says, is the capacity factor threshold. As soon as the percentage penetration of VRE is equal to its nominal capacity factor, then over-generation will occur more frequently, resulting in the need to dump, curtail, or store electricity instead of consuming it at the point of generation. It is simple oversupply and demand. When American corn farmers have a bumper crop, which to city people sounds like a good thing, they are in fact feeding too much maize into the world grid. Prices plummet and foreclosures begin.
That is not true of fossil-fuel or nuclear plants, which can be ramped up or down. Jesse Jenkins remarks that, “wind and solar depress the market price at exactly the times of day these VREs are generating the most power.”57 At low levels, this is not important, but at higher levels it becomes a vicious cycle. “Wind and solar costs,” Jenkins writes, “will have to keep falling to secure greater penetration levels and remain profitable at the ever lower and lower market prices caused by increasing VRE penetration.”58
This may be the only place where storage can play a significant role, delivering electricity when panels and turbines are under-generating. This is known as peak shaving.59 For this to work, storage prices must drop considerably.
For all the quixotic talk of wind and solar energy supplying most of our electricity, at best wind may eventually be able to provide on the order of 25–35 percent of a grid’s electricity, while solar energy may top out at 10–20 percent in most regions.60 Given that just a few regions supply most of America’s wind and solar power, some American grids may already be approaching this point. It has already been happening in Germany, a country with relatively high use of solar energy during the summer months.61 Indeed, in the United States, when too much energy is produced, the first generators to be shut down are VRE, because they are the most expensive.62
Consider a side-by-side comparison between Germany, the country most reliant on wind and solar, and France, the one most reliant on nuclear. The claim: “Germany Can Now Produce Half Its Energy from Solar.”63 The reality: that figure applied for a few hours on a single day. Germany got less than seven percent of its energy from solar that year.64 German weather is hardly ideal for solar power; it has about the same solar power potential per panel as Alaska.65 Nevertheless, Germany now leads the world in solar production.66 It is also number two in wind production.67 For these reason, it is a green energy poster child. Despite this image, Germany is an unfolding disaster both in terms of costs and GHG emissions.68
The reality is that Germany is very much a brown energy country. Wind and solar account for only about 16 percent of the total needed electricity; in a given week, they may generate less than five percent.69 About half of Germany’s electricity comes from coal, and most of that is lignite, the soft brown variety that produces the least energy, and the most air pollution and GHGs of any fossil fuel.70 Germany burns more lignite than any other country.71 It also mines forty-three percent of all European coal, and its exports go on to raise pollution and GHG emissions in other nations.72 Much of the natural gas it burns is imported from Russia.73
The explanation for burning brown lignite is simple: it is very cheap. The irony is that to afford wind and solar power, Germany must burn lignite, so much that German CO2 emissions have jumped since 2011.74 Germany is in fact the dirtiest EU country in terms of GHG emissions, both as a nation and per capita.75
Wind and solar energy require massive subsidies. To support them, the average German family has had to pay a surcharge of 47 percent on its electricity bill. This is about five times what some other European countries pay, and well over twice the average U.S. rate.76 The renewable energy surcharge alone is more than half the U.S. average for total electricity costs.77 In this, Germany is almost tied with Denmark, the world’s top user of wind power.78
The 2011 Fukushima incident evoked mass hysteria in Germany, leading the government to shutter some, and eventually all, of its nuclear plants.79 Before Fukushima, Germany generated a quarter of its electricity from nuclear power. As the country closes its remaining 17 plants, the demand for brown coal will increase, and its cost will climb. All of Germany’s reactors were installed between 1969 and 1989, and have been substantially or completely amortized. They are currently pumping out electricity at close to production cost. Germany’s Atomkraftangst will cost the country as much as US$1.9 trillion by 2030, or two thirds of the country’s entire GDP in 2011.80
Germany made a tragic mistake: it believed its own propaganda.
Let us now compare this situation with that of France. France derives 77 percent of its electricity from nuclear energy, making it the world leader.81 Nuclear energy output is the highest it has ever been, as it is in the United States, even without new plants.82 Fossil fuels contribute only about eight percent of electricity generation, about the same as wind and solar power.83 France is the world’s biggest electricity exporter.84
In both financial savings and reduction in GHG emissions, France has reaped the benefits of nuclear power. Electricity prices, for both household and industrial consumers, are the seventh cheapest in the EU. The speediest drop in GHGs on record occurred in France in the 1970s and 1980s, as the nation transitioned from fossil fuels to nuclear power, lowering emissions by roughly two percent per year.85 As of 2012, French CO2 emissions per capita were about half those of Germany.86
There it is. In terms of GHG emissions, Germany is filthy, France is pristine. Germans are paying a fortune, the French are getting—and exporting—cheap electricity.
This may not last. Under a European Commission directive, France might be forced to reduce nuclear output to just fifty percent.87 Or it may last. Compliance could take a while and ultimately be too expensive for the French to accept.88
The French model is workable for the U.S. France built three quarters of its reactors in just seven years.89 With current and new nuclear capacity, plus current hydroelectric and geothermal, the U.S. could easily generate one hundred percent carbon-free electricity by the 2030 EPA deadline—without the burden of wind or solar power.
But the United States, like Germany, is subsidizing and mandating solar and wind energy.90
Wind and solar energy become competitive in the United States only with government subsidies, and mandates at federal, state, and local levels.91 All forms of electricity generation receive subsidies, but nuclear power receives just ten percent of all federal support for providing a fifth of U.S. electricity. Wind and solar power, which provide only seven percent of electricity, get a stunning 64 percent of the subsidies.92 Solar energy alone took home about a fourth.
Why so much? Because it is that much more inefficient.
Subsidies for wind and solar energy amounted to over US$10 billion, or about US$80 per federal taxpayer. Solyndra’s solar tube plant received a US$535 million loan guarantee in 2009, before folding up shop two years later.93
Much of the federal support for variable renewable energy comes from the wind and solar production tax credits, or PTC.94 In 1992, these paid a wind operator 15 cents per MWh produced for the first ten years of operation. It is now 23 cents.95 Whenever it appears that Congress may not extend the subsidies, plans for new wind power sites plummet.96 The solar investment tax credit (ITC) provides thirty percent credit against the installation of solar panels.97 The ITC, plus one other federal subsidy, the IRS Modified Accelerated Cost Recovery System, “can reduce the cost of a commercial solar power system by 70-75% of the original installed price.”98
The United Kingdom, Spain, Italy, and even Australia, are now cutting subsidies for VRE energy.99 Thanks to solar subsidies, the Italians had some of the highest electricity prices in Europe.100 The government has now started cutting electricity prices, prompting complaints from solar energy companies.101 Even Germany has begun cutting back financial assistance to producers.102
A subhead from The Guardian declared, “Putting solar on the chopping block may damage the industry badly at a time when it just needs that last push to become independent.”103
Always that last push.
Perhaps the most damning indictment of variable renewable energy is from those who have invested in it. Berkshire Hathaway has about US$15 billion invested in clean energy and plans on doubling that in order to take advantage of tax credits.104 “I will do anything that is basically covered by the law to reduce Berkshire’s tax rate,” Warren Buffet stated last year, “For example, on wind energy, we get a tax credit if we build a lot of wind farms. That’s the only reason to build them. They don’t make sense without the tax credit.”105
Solar energy makes even less sense. Buffet is heavily invested in that as well.106
The U.S. Navy has been using nuclear reactors since 1955, with 27 different plant designs installed in 210 nuclear-powered ships, accumulating over 6,200 reactor-years of accident-free experience. Worldwide, including some civilian vessels, that comes to over 12,000 reactor-years.107 Naval nuclear plants are proof that reactors can be safe without massive concrete shielding.
Technological progress continues.108 Reactors derived from naval designs account for about 85 percent of the world’s nuclear electricity.109 Gen III and III+ reactors, first installed in 1996, have been improvements over the Gen II plants that currently provide all U.S. nuclear power. They benefit from improved fuel technology and superior thermal efficiency, as well as passive safety systems and standardized designs for reduced maintenance and capital costs.110 These small reactors also represent a move away from economies of scale; they require just a few acres and can be very close to where the customers are.111
Standardization means the major parts of Gen III and III+ plants can be made on assembly lines and then easily shipped. Gen II plants typically took between five and ten years to build; build time for Gen III and III+ is about five years.112 Uranium is used ever more efficiently.113 Units can be added as necessary; there is no need to try to predict future energy needs.
Compared to its predecessors, the Westinghouse AP1000, like the new naval reactors, has vastly fewer valves, pumps, and piping, along with about half the seismic building volume. Forget those huge containment domes; the AP1000 containment vessel—built off-site and dropped in with a crane—has a wall thickness of less than two inches, with a dome a mere 165 feet in diameter.114 The first Gen IV plants will come online around 2030 and will be a tremendous leap beyond even the Gen III+.
The amount of high-level nuclear waste from spent fuel generated commercially is about 75,000 imperial tons, making up less than one percent of total nuclear waste. The rest comes from weapons production. It has been handled and stored safely since the beginning of the nuclear age.115 Nevertheless, it remains dangerous for 100,000 years.
An American firm, Transatomic Power, at MIT, plans to have a demonstration molten salt reactor (MSR) operating within five years. A MSR uses fluid fuel in the form of very hot fluoride or chloride salt, and, in theory, can use up nearly all the potential energy in uranium. Traditional reactors only use about one percent. A MSR would generate up to 75 times more electricity per ton of mined uranium than the light-water reactors currently in use. Its waste would require storage on the order of only hundreds of years.116
Like most Gen IV reactor designs, a MSR does not require liquid cooling. Failure would leave the fuel in solid mode, unable to produce radioactive steam. MSRs can also run on thorium, seen by many researchers as the nuclear fuel of the future. It is vastly more abundant than uranium. Thorium reactors do not need to shut down every 18 months for fuel rod replacement, and do not need high-pressured water for cooling. They run at atmospheric pressure, allowing for a small size and reduced costs.117
The nuclear power industry has suffered three major accidents: Three Mile Island (U.S. 1979), Chernobyl (Ukraine, 1986), and Fukushima (Japan, 2011). More than anything else, these three accidents and their consequences have engendered a persistent sense of anxiety about the safety of nuclear power among both policy makers and the public at large.118 It would be foolish to diminish these events, but it would also be unwise to allow them to completely compromise the future of nuclear power.
We have learned a great deal.
U.S. licensing approval for new plants now requires that the effects of any accident be confined to the plant itself. Further, to avoid another Fukushima incident, water must now be stored inside a plant and natural circulation allowed along the containment walls to cool the reactors.119 One safety indicator is the calculated frequency of degraded core or core melt accidents. While the United Stated Nuclear Regulatory Commission specifies that reactor designs must meet a one-in-10,000-year core damage frequency, U.S. utility requirements are a more stringent one in 100,000 years. The best U.S. currently operating plants are about one in 1 million and the latest models are almost one in 10 million.120
It is not possible in the context of this essay to consider the full range of anxieties related to nuclear energy. No matter the arguments, a proportion of the public will always be afraid of it.