Renewable electricity is booming. Wind electricity is now the cheapest kind there is. Solar power, even when combined with storage batteries to function when the sun doesn’t shine, can be had on the wholesale markets at a price competitive with natural gas, the cheapest fossil fuel.
Suddenly it looks like we really can get 100% of our electricity from renewables and do it economically. There are still plenty of naysayers who point to the intermittency of the wind and sun as insuperable obstacles to grid reliability, but there’s no risk of that anytime soon. The National Renewable Energy Laboratory says we can reach 80% renewable electricity on the grid without any major technological advances, and meanwhile another boom is going on in fields like energy storage that can maintain the necessary balance between supply and demand on the electrical network.
All this is good news, but the euphoria hides a question that is remarkably seldom asked: can we build all those millions of giant wind turbines and billions of solar panels using only renewable energy?
Making components like steel, cement and aluminum takes heat, and lots of it. Renewable energy can’t be burned (except for biomass in limited quantities), so we use it to make electricity. Electricity produces heat, sure, but we’re not talking about light-bulb heat. We need “industrial process heat” (IPH), often at temperatures above 1000˚C.
There are presently no renewable technologies capable of such high heat at anything near the necessary scale.
The necessary scale is vast. Replacing fossil fuels for the usual uses of electricity is only the start. If we’re going to rein in a worsening climate we’ll have to go off fossil fuels altogether, or at least to the point where they become so insignificant as to be irrelevant. Using only renewable energy we must, as the slogan goes, “electrify everything.”
We have to electrify transportation — not only supply the electricity but build the cars, trucks, tractors, trains and buses, the batteries to run them, the charging stations to replace all the gas stations, the electric distribution lines to get to the chargers.
We have to electrify the heating of buildings with old-fashioned resistance heating or the more efficient heat pumps, adding another huge demand for renewable electricity, or construct all new buildings and retrofit all the old ones with such heroic energy efficiency features that they can function almost without electricity, as “passive houses.” Residential and commercial buildings use 40% of total U.S. energy.[i] In the residential sector nearly half of energy goes to heating and cooling, in the commercial sector 41% including water heating, according to the Energy Information Administration, the statistical arm of the U.S. Department of Energy.
And, of course, this being the age of industrial, free market, globalized capitalism, we must go on making every product the market can stand, whether it’s good, bad or useless, at a rate that will ensure economic growth of at least 3% every year. At a 3% annual growth rate the size of the economy doubles every 23 years.
By all means we need to move to 100% renewable energy as rapidly as possible on a foundation of drastically increased energy efficiency. But before we do that we ought to consider whether we can’t drastically reduce our overall demand for energy. And if we’re going to electrify everything we must start with the industrial sector, where it looks like we’ll have to use much of our remaining allowance of fossil fuels to build the renewable manufacturing capacity we’ll need before we can build everything else.
Ask the experts
There is no shortage of plans for an all-renewable world, but few of them consider the IPH problem at all. The best known is the WWS plan by the team of researchers led by Mark Jacobson of Stanford University. WWS stands for Wind, Water and Sun. (It should be just Wind and Sun; most of the possible sites for hydroelectric dams are already taken.) Here’s their prescription for industry in a nutshell:
High-temperature industrial processes will be powered by electric arc furnaces, induction furnaces, dielectric heaters, and resistance heaters and some combusted electrolytic hydrogen.[ii]
More on these later, but for the moment note that not all these technologies are up to the job of large-scale, really high-temperature IPH. On the next page of their article Jacobson and his co-authors state that where electricity can’t be used directly,
remaining end uses (some heating, high-temperature industrial processes, some transportation) are assumed to use WWS power indirectly in the form of electrolytic hydrogen (hydrogen produced by splitting water with WWS electricity).[iii]
It comes down to hydrogen. They admit that fossil fuels will be needed during the transition to WWS.
By contrast, Amory Lovins and his Rocky Mountain Institute have little to say about heat sources in their book, Reinventing Fire, perhaps because their focus is efficiency and because they’re trying to make a business case. Their ultimate solution is the ultra-efficient design of factories, but they admit this is too unrealistic for now to be included in their “Reinventing Fire” scenario for 2050. In that year, “U.S. industry will still be using considerable fossil fuel, 77% of it natural gas.”[iv] They don’t think renewable electricity can do it.
Both Jacobson and Lovins assume a continually growing economy.
Coal, oil and natural gas can be packed into railcars, tankers and pipelines. Solar energy (including wind, as the sun imparts motion to the winds) is abundant but diffuse. It has to be harvested over large areas. It is not as good a source of concentrated heat on an industrial scale.
The main ingredients of wind turbines are steel and concrete. Solar panels have aluminum frames, glass casings and silicon cells to make the electricity. Blast furnaces for steel operate at 1700˚C, glass furnaces at 1500˚, aluminum smelters at 1000˚, cement kilns at 1450˚ and silicon cell manufacture at 1900˚.
The U.S. Environmental Protection Agency keeps (or used to keep) a web page on renewable IPH.[v] It says that 43% of industrial processes operate at temperatures above 750˚F (399˚C). The only renewable technology it shows as operating above 400˚F is concentrating solar.
Industry uses over 30% of total energy in the U.S. There are industrial processes that use electricity, especially steelmaking in small batches using scrap steel, and aluminum smelting. The U.S. Department of Energy sums up the constraints:
Lack of electrical systems for high-temperature processing: Industries that require energy-intensive, high-temperature processing (>1600°F [871˚C]), such as steel (except electric arc furnaces), petroleum refining, and chemicals, rarely use electric systems because conventional electric heating systems are very expensive or limited in these temperature regimes, with limited ability to provide other benefits. Overall, less than 5% of the total energy used by industrial process heating systems is from electricity, and electricity accounts for less than 13% of process heating in the iron and steel industry and less than 2% in the chemical industry. [vi]
There have been experiments in using solar energy in cement kilns and blast furnaces,[vii] but nothing has been brought up to scale, and scale is the problem.
To get that 5% of electrical IPH up to 100% is not a simple matter of adding 95% more generating capacity. Because wind and solar can’t be counted on to run around the clock, we’d have to either overbuild in geographically dispersed locations or back them up with some kind of energy storage that would kick in whenever they stop. The best wind turbines at the best wind locations are only beginning to reach 50% availability; the best solar panels barely achieve 20%. In tech-speak, they have lower “capacity factors” than fossil fuels. Using electricity to make a burnable fuel like pure hydrogen would take another enormous increment of generating capacity. This new renewable capacity will also require a massive increase in transmission lines, especially because the best wind is found in the Great Plains far from the population centers where the electricity is needed.
Let’s take a quick look at the three main candidates for renewable IPH — concentrating solar, electrified steel and aluminum, and hydrogen.
Unlike the familiar solar photovoltaic panels that produce electricity directly from sunlight, concentrating solar power plants use reflectors to focus the sun’s rays on tubes filled with water or oil. The fluid gets hot enough to make steam, which is used to spin a turbine just as in fossil or nuclear power plants. But these “linear focus collectors” can only achieve 500˚C and, if used strictly as a heat source, need to be located next to the factory with enough space — many acres — for all the mirrors.[viii]
Point-focus collectors can achieve up to 3500˚C but only by focusing the sun’s rays on a very small area. Just to do that takes a huge facility. How to transfer that heat evenly to a large kiln or furnace and how to maintain the desired heat if clouds cross the sun are problems. The world’s largest solar furnace, in Odeillo, France, is the size of an office building and produces only one megawatt of electricity 20% of the time.[ix]
Aluminum is smelted using an electrical process devised in the 1880s. “The most modern aluminum smelters need close to 13 kWh to produce 1 kg of aluminum”[x] (1 kilogram=2.2 pounds; the average American home uses 1,000 kilowatt-hours (kWh) of electricity per month). The Noranda Aluminum smelter in New Madrid, Missouri, which recently closed, drew 475 megawatts (MW) of electricity. That’s the equivalent of a sizable power plant all to itself.
Electric arc furnaces (EAFs), also known as mini-mills, make steel in small batches. They avoid the blast furnace stage of steelmaking, which reduces iron ore to iron, by using scrap steel. This is their biggest limitation: there’s not enough scrap to go around.[xi] EAFs are extremely electricity intensive, needing 35 MW to create the arc of electricity they send through the scrap[xii] or 440 kWh for each metric ton of steel produced[xiii] in a batch that takes 40–50 minutes[xiv], nearly half of what a household uses in an entire month.
Jacobson and his team named three other options for electric IPH: induction furnaces, dielectric heaters and resistance heaters.
Like EAFs, induction furnaces melt scrap steel and therefore suffer from the same limitation on supply. They can use 42 MW of electricity for a 65-tonne melt.[xv] “Small size” is listed as one of their advantages.[xvi]
Dielectric or capacitance heating works on nonconductive materials. Given the uses to which it is put (heating food, killing pests in harvested crops, textile drying, wood gluing, plastic welding) it is a low-temperature process.[xvii]
Resistance heaters are used to heat liquids and gases.[xviii] The heat can exceed 3600˚F. The furnaces are typically small, even small enough to fit on a countertop, but can be as large as a freight car. They seem not to be much used for melting metals but for “holding” molten iron and aluminum. Resistance heating is used extensively in the glass industry and for lower-temperature applications like heating food.[xix]
EAFs own the field of electric steelmaking. I have not discovered why induction and resistance heating are not used for this purpose. Perhaps they just need to be built bigger, or perhaps EAFs make them unnecessary, but the electricity demand of any combination of these three technologies will be very high.
Visions of a “hydrogen economy” have beckoned in recent years. Hydrogen is usually touted as a transportation fuel, either in fuel cells or in compressed form, but it could also make a fuel that can be burned as a replacement for natural gas, the primary industrial heat source.
Hydrogen is the most abundant element in the universe. Unfortunately, it does not occur on Earth in its pure form but must be separated from a compound. Right now this is usually natural gas, but it could also be done by electrolysis of water — passing an electric current through it to separate the hydrogen from the oxygen.
In a book called The Hype About Hydrogen Joe Romm said, “to replace all the gasoline sold in the U.S. today with hydrogen from electrolysis would require more electricity than is sold in the U.S. today.”[xx]
Romm’s book came out in 2004. Nothing has happened in the years since to prove him wrong. A critique of Jacobson’s WWS plan notes that Jacobson’s supporting documentation shows hydrogen “being produced at a peak rate consuming nearly 2,000 GW [gigawatts or billions of watts] of electricity, nearly twice the current [total] US electricity-generating capacity.”[xxi]
Moving and storing hydrogen are challenges. The smallest of atoms, it can leak through just about anything. It is also corrosive, so pipelines are expensive to build. Storing this ethereal element means compressing it into a liquid state. Hydrogen liquefies at –253˚C. “Liquefaction has a sky-high energy cost, some 40% of the usable energy in hydrogen.”[xxii] That’s a low return of energy for the amount of energy invested in the effort.
The renewable machine shop
Renewable electricity can’t make enough steel. It can’t make cement at all. It can’t power the big items like blast furnaces and cement kilns. Hydrogen as a gaseous fuel looks impracticable. The renewable machine shop, where machines make other machines using only renewable energy, is not in sight.
Assuming it can be done, electrifying everything — industry on top of transport, building energy, and electricity generation itself — will obviously take a huge output of wind towers, solar panels and transmission lines. Jacobson and his team think the footprint of wind and solar on the land will be minimal:
The new footprint over land required will be ~0.42% of U.S. land. The spacing area between wind turbines, which can be used for multiple purposes, will be ~1.6% of U.S. land [p. 2093].
That’s still a lot of land, and there’s already a growing backlash against wind and solar farms in areas that host a lot of them.[xxiii]
Jacobson argues that the WWS system will be much more energy efficient than the existing BAU (business as usual) fossil energy system. It’s often said that heating with electricity is 100% efficient because all of the electricity gets converted into heat. But the inefficiency is at the power plant where only one-third of the energy latent in coal is converted into electricity. A parallel inefficiency afflicts wind and solar due to their lower capacity factors, even though electric motors are far more efficient than internal combustion engines.
It’s difficult or impossible to calculate the difference in energy demand between the two systems:
In fact, it is not known whether the total life cycle energy required to manufacture the main components of the WWS energy system, mainly solar panels and wind turbines, will be much different from the total life cycle energy required to manufacture all of the components of the conventional BAU energy system, which includes power plants, refineries, mining equipment, oil and gas wells, pipelines, tanker ships, trucks, rail cars and more.[xxiv]
This is a false comparison because it assumes that WWS, but not BAU, only needs to account for power generation. To balance it with “all of the components of the conventional BAU energy system,” the comparison would have to count new electric arc furnaces, hydrogen infrastructure, batteries to propel vehicles and store intermittent energy to be released when wind and sun are down, electric vehicle charging stations, heat pumps, building retrofits, and more.
Revising Jacobson’s assumption, what he seems to be saying by implication is this: to remake the entire energy system, even if you don’t count the parts that would be made anyway (like auto bodies), would take multiples more energy than BAU — and it would take all the energy embedded in the BAU system just to make all the wind turbines and solar panels.
What to do?
I spend my professional and volunteer time advocating for energy efficiency and renewable electricity. The last thing I want to be is an apologist for fossil fuels. I undertook this task because I haven’t seen it done by anyone else in a systematic manner.[xxv] I’ve done an amateurish job. I hope someone proves me wrong. Otherwise we have to deal with the situation. Technological innovation may help, but we need to face the fact that burning fossil fuels, which contain millions of years’ worth of ancient solar energy, in the course of a century has been an energy pig-out we can’t maintain on a diet of incoming sunshine.
The new order should be conservation first, energy efficiency second, renewable energy third. Efficiency is doing the same things with less energy. Conservation is not using energy at all when we don’t have to. Renewable energy will be a lot easier and cheaper to do if we need less of it. The same is true of industrial process heat.
Since we’ll unavoidably be using fossil fuels for some time yet, IPH should be the top priority. As each year passes the world fails to make a serious dent in fossil fuel usage. We have a rapidly shrinking carbon budget — the amount of greenhouse gases we can continue to spew without reaching the danger zone where climate change slips out of our control. At best that’s the amount that keeps us below a 2˚C increase in average global temperature. Investment in renewables and efficiency would need to triple to $2.3 trillion dollars a year from 2016–2040 to do that; the world’s institutional investors currently invest $3.4 trillion a year on everything.[xxvi]
Using the medium estimate of the Intergovernmental Panel on Climate Change, we have an allowance of 760 billion tons of CO2 for the rest of the century, a little more than nine billion tons a year. The world is now dumping 40 billion tons of CO2 into the atmosphere each year.[xxvii]
As we cut back our fossil fuel allowance, we’d better put a major portion of it into IPH so that we can build the wind and solar farms we’ll need in order to make everything else. It’s time to stop ignoring the challenge of renewable industrial process heat.
Henry Robertson is an environmental lawyer and activist in St. Louis.
[ii] Jacobson et al., “100% clean and renewable wind, water, and sunlight (WWS) all-sector energy roadmaps for the 50 United States,” Energy Environ. Sci., 8:2093, 2094 (2015).
[iii][iii] Id. p. 2113.
[iv] Amory B. Lovins and Rocky Mountain Institute, Reinventing Fire: Big Bold Solutions for the New Energy Era, Chelsea Green Publishing (2011), p. 145.
[vi] U.S. Dept of Energy, https://energy.gov/sites/prod/files/2016/06/f32/QTR2015-6I-Process-Heating.pdf, p. 4.
[viii] Parthiv Kurup and Craig Turchi, Initial Investigation into the Potential for CSP Industrial Process Heat in the Southwest United States, National Renewable Energy Laboratory (2015), pp. 1–2.
[ix] Richard Heinberg and David Fridley, Our Renewable Future: Laying the Path for One Hundred Percent Clean Energy, Island Press 2016, pp. 98–100.
[xi] Alexander Otto et al., Power-to-Steel: Reducing CO2 through the Integration of Renewable Energy and Hydrogen in the German Steel Industry, Energies 2017, 10, 451, pp. 10, 17, http://www.mdpi.com/search?q=power-to-steel&authors=alexander+otto&article_type=&journal=§ion=&special_issue=&search=Search.
[xvi] CEATI, Electrotechnologies, https://www.ceati.com/freepublications/7020_guide_web.pdf, p.34.
[xx] Joseph J. Romm, The Hype About Hydrogen, Island Press 2004, p. 76.
[xxi] Christopher T. M. Clack et al., Evaluation of a proposal for reliable low-cost grid power with 100% wind, water, and solar, http://www.pnas.org/content/114/26/6722.full#ref-11.
[xxii] Romm, p. 100.
[xxiii] http://midwestenergynews.com/2017/09/27/university-compiles-lessons-learned-after-nine-years-of-wind-development-in-michigan/; https://www.usnews.com/news/best-states/virginia/articles/2017-06-13/100m-wind-project-suspended-following-tennessee-moratorium
[xxiv] Jacobson et al. 2015, p. 2113.
[xxv] The best single treatment I’ve seen is Heinberg and Fridley’s Our Renewable Future, pp. 95–102.
[xxvi] Dan Reicher et al., Derisking Decarbonization: Making Green Energy Investments Blue Chip, Stanford University, Oct. 27, 2017, pp. 3–4, https://energy.stanford.edu/sites/default/files/stanfordcleanenergyfinanceframingdoc10-27_final.pdf.