The Great Green Energy Transition Is Impossible

Prologue

First, my background: I have degrees in computer science, applied mathematics, and system engineering. I retired from the Caltech Jet Propulsion Laboratory, after 53 years, in 2020. During evenings for seventeen years I served as an adjunct associate professor of computer science at a small university. I am an associate editor of a professional journal. Before retirement I was a member of two international committees and one U.S. committee related to computer software, and contributed to two others without being an official member; I was instrumental in founding one of them. My professional responsibilities included producing mathematical models for engineers and scientists, implementing those models in software, operating the software to convert measurements into data, and helping to interpret the data. I learned a lot, in many fields, from my clients.

I do not have any explicitly-targeted education or experience with energy, but I have been studying it intensely for about twenty years, with extensive patient mentoring from leaders in the field, including the acting associate director of Argonne National Laboratory. My mentors have told me that my work easily exceeds the requirements for another master's degree.

Now for today’s article….

What I Requested

Several months ago, I asked California State Senator Anthony Portantino's office -- and California Assemblywoman Laura Friedman's office, and Los Angeles County Supervisor Lynn Barger's office, and Congressman Adam Schiff's office, and Senator Dianne Feinstein's office, and Senator Alex Padilla's office, and Energy Secretary Jennifer Granholm's office -- for a report of a comprehensive quantitative system-engineering life-cycle analysis of an all-renewable energy system. Nobody sent a report. I suspect it doesn't exist. But nobody was polite enough to reply “Sorry, we don’t have such a report.”

What I mean by comprehensive life-cycle analysis is one that includes minerals, metals, concrete, other materials, transportation, construction, operation, maintenance, safety, decommissioning, destruction, recycling, disposal, energy return on energy invested, energy payback period, financial payback period, and overall environmental effects. Operation requires methods for generators to synchronize voltage, frequency, and phase with the grid, and storage for when the weather doesn't cooperate with demand.

Because I received no such report I started doing some research.

I haven't put together a comprehensive analysis of my own, but I have found or developed a few pieces.

Materials

Professor Simon Michaux of Geologian Tutkimuskeskus (Geological Survey of Finland) and the University of Adelaide obtained the International Energy Agency (IEA) report concerning renewable energy. Therein, IEA specified a spectrum of technology units that included solar panels, wind turbines, transmission lines, distribution lines, transformers, circuit breakers, etc. Professor Michaux added up the materials requirements of these technology units and compared them to production rates and reserves (his Ph.D. is in mining engineering).

There is a 1,000-page report at https://tupa.gtk.fi/raportti/arkisto/42_2021.pdf. So as not to go far into the weeds, here's the result for copper alone: 4.5 billion tonnes (1,000 kilograms per tonne) of copper are needed. That's about six times the total amount that humans have so far extracted from the Earth. The rock-to-metal ratio for copper is more than 500, so it would be necessary to dig up and refine more than 2.25 trillion tonnes of ore. If mining were to continue worldwide at the same rate as in 2019, 189 years would be required to produce the copper needed to build the demanded units. The response might be just mine faster! But it takes twenty years for a new mine to be put into operation (largely due to permitting requirements), and fewer than one in ten discoveries turn out to be economic. And even if it were possible to mine faster, the amount of copper needed is five times more than the total amount known to exist.

Ten times more nickel is required than is known to exist. 26 times more cobalt is required than is known to exist.... That's for the first generation of units, so there are very few old ones to recycle. And most of that equipment is only 30% recyclable so succeeding generations would be almost equally impossible. Here are some weeds (MT = millions of tonnes, KT = thousands of tonnes):

[2] Rock-to-metal ratio is from Nedal T. Nassar et al, US Geological Survey.

Operation

If we pretend that Tinkerbell could sprinkle some magic pixie dust on the world's deserts and turn them to copper, or a herd of unicorns could tow an all-copper asteroid into Earth orbit, and we manage to build the demanded technology units, the next problem is “how do we operate an electrical supply system in which all generators use renewable sources?” In Burden of proof: A comprehensive review of the feasibility of 100% renewable-electricity systems, Renewable and Sustainable Energy Reviews 76, Elsevier (2017), pp 1122-1133, Ben Heard, Barry Brook, Tom Wigley, and Charles Bradshaw described the result of their analysis of 24 papers in refereed professional journals that claimed to explain how to do it. They evaluated them on seven criteria. No paper successfully addressed more than four.

One factor that Heard et al considered is that it is necessary for a generator to put power on the grid with voltage, frequency, and phase very accurately matching the grid. Otherwise, they risk damaging their equipment, the grid, and customers' equipment. If there's no “signal” on the grid, how do they do it? The way it's done now with hundreds of large generators is that each one starts up but doesn't connect its grid to the regional grid. Then the operators get on the telephone and agree which two will be synchronized and their grids connected. Then the ones with connections to the synchronized grid can synchronize and connect, .... That's why it takes a week to restart after a regional blackout. With millions of generators instead of hundreds, the grid might not get restarted before it fails again. This problem might be solvable, for example by converting the transmission grids from AC to DC, at significant additional expense, so there are no frequency or phase synchronization problems.

Storage

Assuming the system can be built, and we learn how to operate it, the next question is “what do we do when the sun isn't shining and the wind isn't blowing?” The glib answer is Storage! but that answer is never quantified by activists. Several people have tackled the quantity question. Their answer is 400-800 watt-hours of storage per watt of average demand, assuming average generation is equal to average demand. My calculations, using twelve years of California generation data, are much more pessimistic -- more than 3,000 watt hours per watt of average demand. Many people have looked at the technology of storage and concluded that the only physically feasible scheme uses batteries. Pumped storage, compressed air, flywheels, ultracapacitors, towing rocks up and down mountains or abandoned mineshafts, unicorn farts, etc. simply will not work at the necessary scale.

So how much would batteries cost? Using the most optimistic 400 watt hours requirement -- something a real engineer would never do -- and assuming installation is free -- another thing a real engineer would never do -- one might look in Tesla's catalogue and discover the price is $0.543 per watt hour -- before installation -- and the warranty period, roughly equal to the lifetime, is ten years. Activists insist that an all-electric American energy economy would have average demand of 1,700 gigawatts. If one evaluates the formula 1,700,000,000,000 * 400 * 0.543 / 10, the answer is $37 trillion, or about twice total USA 2020 GDP, every year, for batteries alone. Activists say build more generators so you don't need as much storage! Read the Materials section again.

Of course, even with 3,000 hours of storage, today's industrial economy will not survive with an all-renewable energy system the next time Mount Tambora erupts and gives the Earth another “year without a summer” as in 1816 -- and it will happen. 

Read the next article Adequate Storage for Renewable Energy is Not Possible.

Emission-free Vehicles

One proposition, endorsed by California Governor Newsom, is that no hydrocarbon-powered light vehicles will be sold in California after 2035. Of course, this edict is illegal without legislative action.

The argument is that the vehicles are emission free. But that's only at the point of operation, not a life-cycle analysis. They export emissions from rich neighborhoods where people can afford electric cars, to poor neighborhoods, such as around the coal-fired generators on the Navajo and Hopi reservations, so that poor people can enjoy the emissions instead, and tourists can't see the bottom of the Grand Canyon. Moreover, a study published by the Ifo Institute of Germany in 2019 found that an electric Tesla Model 3 emits 11% to 28% more CO2 over its lifespan than a diesel Mercedes C220D. Life-cycle analyses are subject to uncertainty, and no single study is an end-all, but this clearly proves that electric vehicles are far from emission-free.

Electric vehicles are 24% heavier than hydrocarbon-powered ones, on average, so they require more energy and produce more carcinogenic particulate emissions in the form of tire and brake dust, road damage, and stirred-up road dust -- and they don't pay into the highway maintenance fund that everybody else pays into at the gas pump. So they're another subsidy that rich people use to display their virtue. And when they catch fire, you can't put them out.

Why Do It?

At this point, one wonders “Why spend so much time, effort, and money on impossible schemes that damage the environment?” As Michael Shellenberger asks “Must we destroy the environment to save the planet?”

Another glib answer is because we have to do it! But why do we have to do it? To reduce carbon dioxide emissions, because they cause climate change. Notwithstanding what journalists and politicians and lawyers and architects and gynecologists and theologians and liberal arts professors at universities and ... insist, there is no cast-in-stone scientific consensus that human activity is in fact changing the climate, and that if it were that it would be harmful. Science is never “settled” if you're actually a scientist or understand science. The essence of science is skepticism.

The social reality is that journalists and other bullies have convinced the public of the “truth” of anthropogenic climate change and the immediate need to do something! about it.

One skeptical dissent is ironically from Stephen Schneider (with S. Icthtiaque Rasool), apparently before he got religion, from Atmospheric Carbon Dioxide and Aerosols: Effects of Large Increases on Global Climate, Science (July 9, 1971) pp 138-141. From page 139:

From our calculation, a doubling of CO2, produces a tropospheric [lower atmosphere] temperature change of 0.8°K. However, as more CO2 is added to the atmosphere, the rate of temperature increase is proportionally less and less, and the increase eventually levels off. Even for an increase in CO2 by a factor of 10, the temperature increase does not exceed 2.5°K.

Therefore, the runaway greenhouse effect does not occur because the 15-μm CO2 band, which is the main source of absorption, “saturates,” and the addition of more CO2 does not substantially increase the infrared opacity of the atmosphere.

This was written back when “scientists” had convinced journalists to worry about the coming ice age. Schneider was warning there was nothing we could do about it.

What does “saturate” mean? Here’s an easy way to visualize it. Imagine that you have a greenhouse and the glass blocks transmission of 90% of the infrared radiation (heat) that is re-emitted after visible light is absorbed inside. If you add an identical layer, the two of them together don’t block 180% of the re-emitted heat. The second layer blocks 90% of the 10% that got through the first layer, or 9% of the total re-emission. The two of them together block 99%, not 180%. Now read through this again and replace “glass” with “carbon dioxide.”

What Can Be Done

Assuming that something must be done brings the next question “Since we can't build the dreamt-of system, what can we do?”

Short Term

In the short term, the answer is to switch from coal to natural gas. Natural gas is four parts hydrogen and one part carbon, whereas coal is almost entirely carbon. And the thermal efficiency of a combined-cycle gas turbine-powered electric plant is almost twice that of a coal-fired plant. So burning natural gas emits far less carbon dioxide per kilowatt-hour of electricity produced.

In the longer term, fossil hydrocarbons will be depleted and we will come to realize that they're more valuable as chemical feedstocks than as fuel. Even now, 6,000 non-fuel products are made from petroleum, and almost as many from coal (read “Clean” Energy Exploitations by Ronald Stein and Todd Royal, ISBN 978-1-6657-0497-7).

What do we do when burning fossil hydrocarbon fuel is no longer possible?

Nuclear Power

The only physically feasible answer is the safest thing humanity has ever done, namely nuclear fission. A nuclear reactor requires far less material and far less land than solar panels and wind turbines with the same average capacity. An industrial solar installation requires 32 times more steel per megawatt of capacity than a nuclear power plant. Wind turbines require five times more steel and five times more concrete. The energy return from a nuclear power plant is 75 times more than the energy invested in it, instead of 2 or 4 times for solar panels or wind turbines. Without subsidies (which don't eliminate the cost but instead hide it in your tax bill), the minimum return required for economic viability is 7 times.  The service life of a nuclear reactor is at least fifty years, and with appropriate maintenance ought to be 100 years. The service life of a land-based wind turbine is about 25 years; offshore ones last about twelve years.

It’s Not Safe!

Activists will insist nuclear power is not safe, but if one examines reality, one finds that in the entire civilized world, nuclear power is safer than Teddy Kennedy's car. I suggest looking for safety reports concerning all parts of the energy sector in the ENSAD database at the Paul Scherrer Institut in Switzerland.

Three Mile Island and Fukushima

Nobody was injured, made ill, or killed by the Three Mile Island or Fukushima Daiichi incidents. No radioactive materials were released at Three Mile Island, and although the destruction of the antique Fukushima Daiichi reactors released radioactive materials, the average increase in radiotoxic dose expected for each resident of Fukushima Prefecture is 17 millisieverts over their entire lifetimes -- according to the report to the United Nations General Assembly from the United Nations Scientific Committee for the Effects of Atomic Radation (UNSCEAR). The fatal radiation dose is 38 Sieverts (not millisieverts) delivered all at once, not spread out over a lifetime (dose rate matters more than total dose, and dose is not cumulative, else I would be dead from 52 Sieverts of radiation spread out over eight weeks to treat prostate cancer).

UNSCEAR wrote that there was no evidence of increased mortality or morbidity or increase in cancer or non-malignant radiation-related disease in the Fukushima region. Japan over-cleaned. The dirt in Fukushima is now half as radioactive as the dirt in Denver.  The average radiation dose rate in Fukushima Prefecture is about one tenth that on the Tibetan Plateau, and about one seven hundredth that on Guarapiri Beach in Brazil.

People are not dropping like flies from radiation-related illnesses in Denver or Tibet or Brazil. The 15,000 Japanese who are living as refugees in their own country could safely return to their homes. 

Chernobyl

Of course, the perfect safety record of licensed reactors built in civilized countries that have a safety culture and licensing criteria is irrelevant to activists. They point to Chernobyl -- again without credible quantification. The Chernobyl reactor is irrelevant to anything that would be licensed and built in a country that has a safety culture and licensing criteria, just as the crash of the Hindenburg at Lakehurst, New Jersey on May 6, 1937 is irrelevant to air travel using Boeing of Airbus airplanes.

Nothing like the Chernobyl reactor will ever be built again, not even in Russia or China. If one reads the two reports that UNSCEAR provided to the UN General Assembly, and the report from the UN Chernobyl Forum -- composed of fourteen independent international agencies -- one finds that 134 people -- all plant workers and emergency first-responders -- were exposed to sufficient radiation to cause acute radiation syndrome, formerly called radiation poisoning. Of those, 28 died from that cause. Others have died from other causes, such as being run over by a garbage truck or beaten to death by an irate husband, but those don't count as radiation-related deaths.

Activists will say But there will be millions of cancers! Once again, read the UNSCEAR report. In the first report, they wrote that there is no evidence of radiation-related illness, including both cancer and non-malignant diseases. In the second report, they wrote “there is no scientific means to determine whether a particular cancer in a particular individual was or was not caused by radiation.” A few pages later they wrote that they statistically estimated that of the 6,000 thyroid cancers reported in the region during the succeeding fifteen years, there were probably fifteen excess cases of fatal juvenile thyroid cancer. Of course, it was inexcusable that those were fatal because thyroid cancer is easily treatable in a country that has a competent health care system.

OK, we'll allow those fifteen, making the entire sixty-year worldwide mortality caused by one inherently unsafe nuclear power plant 43, and zero for the hundreds of others that were actually designed with safety in mind. Has humanity ever done anything safer? Maybe only the space programs.

Inherently Safe Reactor Demonstrated in 1986

Even though the fleet of 1950's-design reactors currently in service in civilized countries has a perfect safety record, engineers and scientists at Argonne National Laboratory envisioned that they could build a system that would be inherently safe, solve all the world's energy problems, be economical to build and operate, and use uranium so efficiently that it would be an essentially inexhaustible energy supply. And they did it! The Experimental Breeder Reactor II (EBR-II) at Argonne West, now the Idaho National Laboratory, between Idaho Falls and Arco, operated perfectly for thirty years, until the Cliton administration terminated the research program, destroyed the reactor that Nobel Laureate Hans Bethe had described as “the best research reactor ever built,” and filled the building with concrete in 1994. When Mr. Cliton was told that cost more than to finish the research program, he pandered "I know; it's a symbol."

In 1986, an invited international audience watched while operators at EBR-II conducted the same experiment that operators at Three Mile Island had done in 1979: they turned off automatic emergency shutdown systems, turned off coolant circulation, and sat back and watched. The fuel pin cladding temperature spiked from about 1,000 degrees Fahrenheit to 1,430 degrees within thirty seconds. Then, unlike at Three Mile Island, the reactor cooled on its own and the core was below operating temperature within seven minutes. Coolant boils at 1,620 degrees, and fuel cladding melts at 3,360 degrees.

This was caused entirely by the physical characteristics of the fuel and structure, not by operators running around with their hair on fire, turning valves and flipping circuit breakers,  or complicated computer algorithms or fancy engineered emergency control systems. Operators restarted the reactor and tried the other method known to be able to damage a reactor: leave coolant circulation running, but turn off the water feed to the steam generator (so that circulating coolant heats up). The same outcome ensued, delayed by about thirty seconds.

This is the only kind of reactor that should be built.

Six weeks later, operators at Chernobyl tried the second experiment using an inherently unsafe reactor, ironically in a rush to get a safety test done, with tragically different results.

Nobody Knows What to Do With Nuclear Waste!

Activists never give up in the face of facts, so they'll trot out Nobody knows what to do about nuclear waste! But what is the substance we call “nuclear waste?” That's the wrong term for it. Spent nuclear fuel is actually valuable 5%-used fuel.  We’ve known the right thing to do for eighty years, but refuse to do it: separate the 5% of it that is fission products from the 95% of it that is unused fuel and consume the unused fuel to produce electricity and fission products. Other than reducing the amount by a factor of twenty, does that have beneficial effects? Indeed it does because unused fuel is radiotoxically dangerous for 300,000 years, while fission products need custody for 300 years.

It's daft to pretend one can hide anything for 300,000 years, even at the bottom of the Mariana Trench in the Pacific Ocean. The pyramids were plundered before 500 years! You definitely want to destroy spent fuel, not store it. So what do we do with that 5%? The first thing to know is its detailed composition. 9.26% of fission products -- caesium and strontium -- produce 99.4% of radiotoxicity and need custody for 300 years. 0.45% of fission products -- europium -- produce another 0.4% of radiotoxicity and need custody for 85 years. Half the rest are innocuous before thirty years, and the remainder aren't even radioactive -- and some such as palladium and rhodium are extremely valuable. So less than 10% of fission products are even a little bit of a problem.

How much fission products are produced? A simple easy-to-use rule of thumb is that producing a gigawatt-year (8,766,000,000 kilowatt hours) of electricity produces one tonne of fission products, of which 92.6 kilograms are caesium and strontium. An all-nuclear all-electric 1,700 GWe American economy would produce 156 tonnes of caesium and strontium per year. They have a density slightly more than two tonnes per cubic yard, so the amount is 78 cubic yards per year, or less than nine cement mixer trucks full -- for an entirely nuclear all-electric American energy economy! We can handle that.

Why don't we do it? Because the Carter Administration sabotaged it, in the naive belief that if the United States did not recycle spent fuel, nobody else would build nuclear weapons. I guess India and Pakistan and North Korea weren't paying sufficient attention.

So far, nobody in America has had the guts to start doing it again. Maybe that's just as well, because the system that was planned for Barnwell, SC, would have used a process called PUREX. This is a solvent-based system for which enormous amounts of land, plant, and equipment are needed, and it doesn't really do a complete job.

Engineers and scientists at Argonne National Laboratory have again filled the void. Part of  the project at EBR-II was fuel recycling. They developed a pyroelectric process, similar to aluminum refining, that is much simpler and less expensive than PUREX, and does a more complete job. They processed several tens of thousands of fuel pins using a machine about the size of a dishwasher (there is other equipment in the process that increased the size of the Fuel Processing Facility to a few thousand square feet).

The plant at Barnwell would have occupied almost 300 acres, cost about 0.08 cents per kilowatt hour to operate, and not done a complete job. The plant at Rokkasho in Japan has so far cost about $25 billion and occupies almost 1,000 acres. If it ever enters service, it will cost 0.5 cents per kilowatt hour to operate. A pyroelectric system, as designed by Argonne, with half the capacity of Rokkasho -- about 400 tonnes per year -- would occupy about forty acres (including parking, offices, labs, and storage), cost about $900  million to build, and cost 0.05 cents per kilowatt hour to operate (including capital amortization) -- about half what utilities paid into the Nuclear Waste Fund until courts decided they no longer needed to pay because the Department of Energy had reneged on their legally-required duty to take custody of spent fuel. The fund now stands at $43 billion, but the Nuclear Waste Act prohibits using it for reprocessing.

Weapons Proliferation!

After being confronted with this solution, activists will shift gears and shout Weapons proliferation! as if American nuclear power plant operators would be eager or allowed to sell or give spent fuel to terrorists in Iran or Afghanistan or North Korea. Assuming they managed to get their hands on spent fuel, and develop the technology to separate plutonium from it, they would find that it's not suitable to make weapons. Weapons-grade plutonium consists of 93% plutonium-239, and 7% other isotopes -- mostly plutonium-238 and plutonium-240. Plutonium that is produced in a commercial power plant contains 54% plutonium-239. The other isotopes produce enormously more heat, neutron emission, and gamma rays than plutonium-239. This would require remote fabrication, distort fine tolerances, and damage chemical explosives. The weapons might not work, or might pre-detonate -- not exactly what you want in your submarine or aircraft carrier.

Isotope separation is enormously more expensive than chemical separation, and is at least three times more difficult for plutonium than for uranium. Britain fiddled with this about sixty years ago using 63% plutonium-239 that had been prepared by operating a commercial-style reactor in an easily-detected nonstandard manner -- primarily by using a much shorter fuel residence time. They got a fizzle instead of a boom. They said “We will not bother to try that again.”

Nobody has ever deployed a weapon that was made using plutonium taken from a commercial nuclear power plant, and nobody ever will, because every other way to make one is easier and less expensive.

Not Enough Uranium!

Of course, that won't stop activists either. They know that if they succeed, they have to find or create a new problem, or become as relevant as the Women's Christian Temperance Union, which suffered from an excess  of success. So they'll shift gears again and shout There isn't enough uranium! But that's false too. The United States alone has 90,000 tonnes of uranium in spent fuel, and 900,000 tonnes of depleted uranium. In the right kind of reactor, all of that metal is fuel or future fuel -- already above ground, mined, milled, and refined. Using the above-cited rule of thumb that fissioning one tonne of heavy metal produces a gigawatt year of electricity, and dividing 990,000 tonnes by 1,700 tonnes per year (for 1,700 gigawatt years per year), shows that, by using the kind of reactors invented at Argonne and Idaho National Laboratories, the United States has enough fuel, above ground, mined, milled, and refined, to power an all-electric all-nuclear American energy economy for 582 years, and of course longer to the extent that other generators are still in service, without mining, milling, refining, enriching, or importing one new gram of uranium.

The Australian Mining Association estimates that there are enough uranium reserves that can be economically exploited at today's prices to power an all-electric all-nuclear world economy using today's conventional reactors for 1,200 years. Those reactors extract 0.6% of the energy from mined uranium. Using it 160 times more efficiently would make those reserves last for 192,000 years. If fuel were used 160 times more efficiently, increasing the price of uranium by that same factor of 160 would have no effect on the end-user price of electricity; this would make lower-quality reserves economically usable for much longer than 192,000 years. The oceans are known to contain about a million times more uranium than can be recovered on land -- and we already know how to separate it (but it's more expensive than mining so we don't do it).

The Earth has four times more accessible thorium than uranium. Thorium isn't fuel, but it is future fuel if used in the correct kind of reactor, which needs to be started with uranium or plutonium. One drawback of thorium is that it is converted to fuel -- uranium-233 -- about 1/5th as fast as depleted uranium -- future fuel -- is converted to plutonium.

Nuclear fission is an inexhaustible energy source.

Nuclear Fusion

A recent experiment at the National Ignition Facility at Livermore, California was applauded because three megajoules of energy were produced from two megajoules of laser light. What journalists didn’t tell you is that the power company had put 300 megajoules into the capacitors that were discharged into the lasers. And the equipment was damaged.

Nuclear fusion might someday be possible, but nobody knows yet how to do it, and if a real breakthrough were to occur today, it would take at least fifty years to bring the technology into commercial service.

Fusion produces much more energetic neutrons than does fission -- which makes it more thermodynamically attractive. But those neutrons make the entire reaction vessel radioactive, with an entirely different spectrum of metals, and more of them, than are present in fission products used to make the same amount of energy. Lewis Strauss famously said that nuclear power would be “too cheap to meter” because the contribution of the price of uranium, as it comes out of the ground, to the end-user price for electricity, is 0.001 cents per kilowatt hour.

Sunlight and wind are free, not 0.001 cents per kilowatt hour, but the power from solar panels and windmills isn't free. You have to pay the mortgage and the operators. The same would apply to a fusion device. So why do it, and especially why hang your hat on that now?

Epilogue

Yes, this is a long article, but I hope the education was worth your effort. There's much more at http://vandyke.mynetgear.com/Nuclear.html, including references for discussions above. I urge you especially to read Plentiful Energy (ISBN  9781466384606) by Charles E. Till and Yoon Il Chang. It's available on paper from Amazon, and Dr. Chang has generously given me permission to post a PDF on my page.

The question for the California legislature and governor now is “Because the current Don Quixote quest is physically and financially impossible, what is California going to do?”