This piece was first published in October 2019 and has been lightly updated.
Climate activists are fond of saying that we have all the solutions we need to the climate crisis; all we lack is the political will.
While it’s true enough as policy goes — we certainly have enough solutions to get started and make big changes — as a technical matter, it is incorrect. Truly defeating climate change will mean getting to net-zero carbon emissions and eventually negative emissions. That means decarbonizing everything. Every economic sector. Every use of fossil fuels.
And actually, there are some sectors, some uses of fossil fuels, that we do not yet know how to decarbonize.
Take, for instance, industrial heat: the extremely high-temperature heat used to make steel and cement. It’s not sexy, but it matters.
Heavy industry is responsible for around 22 percent of global CO2 emissions. Forty-two percent of that — about 10 percent of global emissions — comes from combustion to produce large amounts of high-temperature heat for industrial products like cement, steel, and petrochemicals.
To put that in perspective, industrial heat’s 10 percent is greater than the CO2 emissions of all the world’s cars (6 percent) and planes (2 percent) combined. Yet, consider how much you hear about electric vehicles. Consider how much you hear about flying shame. Now consider how much you hear about ... industrial heat.
Not much, I’m guessing. But the fact is, today, virtually all of that combustion is fossil-fueled, and there are very few viable low-carbon alternatives. For all kinds of reasons, industrial heat is going to be one of the toughest nuts to crack, carbon-wise. And we haven’t even gotten started.
Some light has been cast into this blind spot with the release in late 2019 of two reports by Julio Friedmann, a researcher at the Center for Global Energy Policy (CGEP) at Columbia University (among many items on a long résumé).
The first report, co-authored with CGEP’s Zhiyuan Fan and Ke Tang, is about the current state of industrial heat technology: “Low-Carbon Heat Solutions for Heavy Industry: Sources, Options, and Costs Today.”
There’s a lot in these reports, but I’m going to guess your patience for industrial heat is limited, so I’ve boiled it down to three sections. First, I’ll offer a quick overview of why industrial heat is so infernally difficult to decarbonize; second, a review of the options available for decarbonizing it; and third, some recommendations for how to move forward.
Why industrial heat is such a vexing carbon dilemma
There’s a reason you don’t hear much about industrial heat: Consumers don’t buy it. It is a market dominated entirely by large, little-known industrial firms that operate outside the public eye. So unlike electricity, or cars, there is little prospect of moving the market through popular consumer demand. Policymakers will have to do this on their own. And it won’t be easy.
The biggest industrial emitters are cement, steel, and the chemical industries; also making a notable contribution are refining, fertilizer, and glass. As a group, these industries have three notable features.
First, almost all of them are globally traded commodities. Their prices are not set domestically. They compete with optimized supply chains around the world, with razor-thin margins. Domestic policies that raise their prices risk “carbon leakage” (i.e., companies simply moving overseas to find cheaper labor and operating environments).
What’s more, some of these industries, especially cement and steel, are especially prized by national governments for their jobs and their national security implications. Politicians are leery of any policy that might push those industries away. “As one indication, most cement, steel, aluminum, and petrochemicals have received environmental waivers or been politically exempted from carbon limits,” says the CGEP report, “even in countries with stringent carbon targets.
Second, they involve facilities and equipment meant to last between 20 and 50 years. Blast furnaces sometimes make it to 60. These are large, long-term capital investments, with relatively low stock turnover. “Few industrial facilities show signs of imminent closure, especially in developing countries,” the CGEP report says, “making deployment of replacement facilities and technologies problematic.” At the very least, solutions that can work with existing equipment will have a head start.
Third, their operational requirements are both stringent and varied. They all have in common that they require large amounts of high-temperature heat and high “heat flux,” the ability to deliver large amounts of heat steadily, reliably, and continuously. Downtime in these industries is incredibly expensive.
At the same time, the specific requirements and processes at work in these industries vary widely. To take one example, steel and iron are made using blast furnaces that burn coke (a form of “cooked” coal with high-carbon content). “Coke also provides carbon as a reductant, acts as structural support to hold the ore burden, and provides porosity for rising hot gas and sinking molten iron,” the CGEP report says. “Because of these multiple roles, directly replacing coke combustion with an alternative source of process heat is not practical.”
A cement kiln works somewhat differently, as do the reactors that power chemical conversions, as does a glassblower. The variety of specific operational characteristics makes across-the-board substitution for industrial heat difficult.
Each of these industries is going to require its own solution. And it’s going to have to be a solution that doesn’t raise their costs much or at least takes steps to protect them from international competition.
The options to date are not much to speak of.
The options for decarbonizing industrial heat are scarce
What are the alternatives that might provide high heat and high heat flux with less or no carbon emissions? The report is not sanguine: “The pathway toward net-zero carbon emission for industry is not clear, and only a few options appear viable today.”
Alternatives can be broken down into five basic categories:
- Biomass: Either biodiesel or woodchips can be combusted directly.
- Electricity: “Resistive” electricity can be used to, say, power an electric arc furnace.
- Hydrogen: This is technically a subcategory of electricity, since it is derived from processes powered by electricity; it is produced through steam reforming of methane (SMR) to make carbon-intensive “grey” hydrogen, SMR with carbon capture and storage to make “blue” hydrogen, or electrolysis, pulling hydrogen directly out of water, to make low-carbon “green” hydrogen.
- Nuclear: Nuclear power plants, either conventional reactors or new third-generation reactors, give off heat that can be carried as steam.
- Carbon capture and storage (CCS): Rather than decarbonizing the processes themselves, their CO2 emissions could be captured and buried, either the CO2 directly from the heat source (“heat CCS”) or the CO2 from the entire facility (“full facility CCS”).
All of these options have their difficulties and drawbacks. None of them is anywhere close to cost parity with existing processes.
Some are limited by the intensity of the heat they can produce. Here’s a breakdown:
Some options are limited by the specific requirements of particular industrial processes. Cement kilns work better with energy-dense internal fuel; resistive electricity on the outer surface doesn’t work as well.
But the biggest limitations are costs, where the news is somewhat disheartening, for two reasons.
First, even the most promising and viable options substantially raise operational costs. And second, the options that are currently the least expensive are not exactly the ones environmentalists might prefer.
There’s a lot in the report on the methodology of comparing costs across the technologies, but the main thing to keep in mind is that these cost estimates are provisional. They involve various contestable assumptions, and real performance data is often not available. So it’s all to be taken with a grain of salt, pending further research. That said, here’s a rough-and-ready cost comparison:
You might notice that most of the blue bars, the low-carbon options, are way over on the expensive right. The only ones that are reasonably affordable are nuclear and blue hydrogen.
Hydrogen is the most promising alternative
In terms of ability to generate high-temperature heat, availability, and suitability to multiple purposes, hydrogen is probably the leading candidate among industrial-heat alternatives. Unfortunately, the cost equation on hydrogen is not good: the cleaner it is, the more expensive it is.
The cheapest way to produce hydrogen, the way around 95 percent of it is now produced, is steam methane reforming (SMR), which reacts steam with methane in the presence of a catalyst at high temperatures and pressures. It is an extremely carbon-intensive process, thus “grey hydrogen.”
The carbon emissions from SMR can be captured and buried via CCS (though they rarely are today). As the chart above indicates, this kind of “blue hydrogen” is the cheapest low(er) carbon alternative for high-temperature industrial heat.
“Green hydrogen” is made via electrolysis, using electricity to separate hydrogen from water. If it is made with carbon-free energy, it too is carbon-free. There are a few different forms of electrolysis, which we don’t need to get into. The main thing to know is that they are expensive — the least expensive is more than twice as expensive as blue hydrogen.
Here’s a simplified cost chart, to make these comparisons clearer:
Note: These numbers reflect “what is actionable today within existing facilities.” The CGEP report stresses that “the authors do not discount future potential for lower cost systems.” More on that later.
For now, to a first approximation, all the available low-carbon alternatives substantially raise costs of industrial-heat processes against the baseline.
And here’s the real kicker: in most cases, it is cheaper to capture and bury CO2 from these processes than it is to switch out systems for low-carbon alternatives.
CCS is often cheaper than low-carbon alternatives
Take cement production. It requires temperatures of at least 1,450°C, so the only viable options are hydrogen, biomass, resistive electric, or CCS. Here’s how much they would increase cement (“clinker”) production costs:
As you can see, every low-carbon alternative raises costs more than 50 percent above baseline. The only ones that don’t raise it more than 100 percent are CCS (of the heat source only), blue hydrogen, or resistive electric in places with extremely cheap and plentiful carbon-free energy.
The alternative that climate hawks would most prefer, the carbon-free option that would work best for most applications, is green hydrogen. But that currently raises costs between 400 and 800 percent. Ouch.
The situation is much the same for steel:
And so on down the line, from chemicals to glass to ceramics; In almost every case, the cheapest near-term decarbonization solution is just to capture and bury the carbon emissions.
Of course, that’s just on average. The actual costs will depend on geography — whether there are suitable burial sites for CO2, whether natural gas is cheap, whether there’s a lot of hydro or wind nearby — but there’s no getting around the simple truth about today’s industrial-heat alternatives: What’s green isn’t very feasible, and what’s feasible isn’t very green.
Here’s a qualitative chart that tries to get at that relationship.
What’s most feasible is on the right. What’s most expensive is up top. There isn’t much in that lower-right feasible/cheap quadrant except blue hydrogen, for now.
The report emphasizes that these initial technology rankings are “temporary at best” and “highly speculative, uncertain, and contingent.” Much more needs to be understood about the costs and feasibility of these options. Their relative attractiveness may change quickly with technology development.
Which brings us to recommendations.
How to make green industrial heat cheaper and cheap industrial heat greener
One of the clearest results of all this research, Friedmann emphasizes several times, is that more research is needed. The data available on industrial heat alternatives is sparse and inconsistent, and there are few existing attempts to compare costs across categories. The most pressing need is for more analysis and research.
That said, there is a path forward. The ICEF paper reviews the information above and then offers a series of policy recommendations.
The first and most important is increased government support for research and development (R&D). This is where the “we have the solutions we need” message can be counter-productive. Yes, we need to get started immediately deploying available clean technologies at scale. But we also need to attend to those sectors of the economy we don’t yet know how to decarbonize. We need to identify promising technologies, as these reports do, and begin working consciously to bring them down the cost curve.
At the very least the US needs to grow its annual spending on clean energy research (about $15 billion) by about tenfold, establish some regional and sector-specific research centers, and draw in partners from industry to accelerate the process of commercialization.
Second, industrial heat is an area where government procurement could play a huge role — governments purchase large amounts of steel, concrete, and chemicals. “Procurement standards that give preferences to products with the lowest embedded carbon content could drive significant changes in industrial behavior,” says the ICEF report.
Third, government needs to help offset the increased costs of alternatives with fiscal subsidies, whether they are loan guarantees, direct grants, feed-in tariffs, or what have you. Public money is needed to move things along.
Fourth, many alternatives require new infrastructure (electrical lines or hydrogen pipelines, for example) and government can help provide it.
Fifth (note: not first, fifth): a price on carbon would push everything along faster. It’s not clear how high a price would have to be to fully offset the additional costs of alternatives — in certain applications, it will likely be impracticably high, so sector-specific policies will still be needed — but any price at all would help.
Sixth, in order to protect against international competition from countries with lower standards, carbon tariffs could be levied on imported industrial products with higher carbon content.
Seventh, there are always good old-fashioned mandates: Government could simply require declining use of fossil fuels in these sectors.
Finally, voluntary industry associations can help spread learning and best practices among companies, while a clean energy ministerial at the international level could do the same among countries.
As this list makes clear, there is a lot that needs to be done before “we have all the solutions we need” in heavy industrial sectors. And there are other sectors that remain difficult to decarbonize as well (shipping, heavy freight, airplanes), all of which would benefit from the same policies.
We need to be deploying what we know and learning more about what we don’t know at the same time.
A final note about electrification
These two reports might seem to be pro-CCS, but that is not the main thing to take away from them. Even if it is true that CCS is the cheapest current decarbonization option for some industries and sectors, it simply won’t be available in many areas. And in the long term, the goal must still be to eliminate fossil fuels to the extent practicably possible, by making alternatives more feasible and less expensive.
The only technology solution with a potential path down the cost curve to the point of being competitive with (properly priced) fossil fuels is electrification.
The charts above reveal two things about electrification of industrial heat. One, resistive electricity is the only low-carbon industrial-heat option competitive with CCS or blue hydrogen, and that’s only where clean electricity is extremely cheap and plentiful. And two, the only truly carbon-free, unlimited, all-purpose alternative available is green hydrogen, which requires plentiful renewable energy.
Both argue for the absolute imperative of making clean electricity cheaper.
At current prices and with current technologies, an all or mostly renewable grid would have difficulty with industrial heat, which requires enormous, intensive amounts of energy, reliably and continuously supplied. Some industrial applications could shift their demand around in time to accommodate renewables or make their processes intermittent, but most can’t. They need controllable, dispatchable power.
Building a renewable-based grid that could handle heavy industry would require much cheaper and more energy-dense storage, more and better transmission, smarter meters and appliances, and better demand response, but above all, it would require extremely cheap and abundant carbon-free electricity.
It all gets easier if clean power gets cheaper. It’s true of resistive electricity, it’s true of green hydrogen, and it’s true of almost all the difficult to decarbonize sectors. Cheap, abundant, clean electricity is the only road to a genuinely sustainable energy system unified around the grid and the free exchange of electrons.
While CCS may be the cheapest available option for some sectors today, that cannot be the final destination; it must be temporary. We need to maximize the amount of CO2 we pull from the air and bury it, drawing down atmospheric concentrations, and minimize the amount we emit. It may not be practicable any time soon, but on some time scale, electricity (and the hydrogen fuels it can create) must take over.
So it’s worth adding a final bit of policy recommendation, something the ICEF report does not explicitly call out: Make carbon-free electricity cheaper, by any means necessary. Extend tax credits for renewables and broaden them to other clean energy technologies; pass a national clean energy standard; pass sector-specific performance standards; build long-distance transmission lines; use government procurement; research and commercialize advanced nuclear and geothermal; explore new run-of-river hydro; and for god’s sake, put a price on carbon. On zero-carbon electricity, keep the pedal to the floor.
Even when it comes to industrial heat, which is so complicated, sensitive to prices, and varied in application — where CCS may be the best short-term expedient — it still remains true that falling clean power costs make everything easier in the long term.
“Give me a fulcrum and a place to stand,” the ancient mathematician Archimedes is reported to have said, “and I will move the world.” Cheap, abundant, carbon-free electricity is a fulcrum for Archimedes’ lever, a point of leverage that makes moving the rest of the world possible.