Volts podcast: Rebecca Dell on decarbonizing heavy industry

In this episode, Rebecca Dell, who runs the industry program at the ClimateWorks Foundation, offers a comprehensive overview of the problems of industrial decarbonization, the most promising technological solutions in steel, cement, and chemicals, and the kinds of policies that could accelerate progress. Incredibly informative.

Full transcript of Volts podcast featuring Rebecca Dell, February 11, 2022

(PDF version)

David Roberts:

For most of the carbon-intensive sectors of the economy — electricity, transportation, buildings — we have a pretty good sense of how to eliminate carbon emissions. None of those sectors will be easy to decarbonize. Every one is an enormous practical challenge. But in each case, the basic path to zero is clear, and it mostly involves switching out fossil-fueled machines with machines that generate or run on clean electricity.

Then there’s that other wedge on the pie chart, the one that gets less attention: industry. Manufacturing, mining, construction, and waste processing are responsible for about a third of global carbon emissions (about a quarter of US emissions).

The path to zero emissions in heavy industry is much murkier than it is for other sectors. Low-carbon alternatives are early in development and commercialization; in some cases, there are no alternatives except to capture and bury the carbon when it’s emitted.

In future pods, I might get deeper into some specific industries (like steel). But for this one, I wanted to attempt a broad overview: What You Need to Know About Decarbonizing Industry.

Nobody knows the sector and its challenges better than Rebecca Dell, who runs the industry program at the ClimateWorks Foundation. Dell previously worked at the Department of Energy, where she helped coordinate Obama’s climate action plan, and before that was a research scientist at Scripps Institution of Oceanography. She’s a researcher, author, and, as more attention turns to industry, an increasingly frequent podcast guest. (She was on Canary’s Catalyst pod last month.)

It takes a while — okay, almost two hours — but Dell and I manage to cover all the big industrial sectors, why they emit so much, prospects for reducing emissions, and the policies that could make it happen. If you’re looking for a one-stop-shopping primer on industry and climate, this is for you.

Without further ado, Rebecca Dell, welcome to Volts.

Rebecca Dell:  

Thanks so much for having me. I'm really happy to be here.

David Roberts:

I'm excited for this. We are going to attempt to cover a lot of ground. I want to try to give a 30,000-foot overview of industry and decarbonization; obviously any of the subtopics could be podcasts of their own. 

Among the Volts audience, people are probably basically familiar with the famous Energy Information Administration pie chart of where US greenhouse gases come from. There are wedges for transportation, electricity, buildings, agriculture — I think people mostly have their heads around how to decarbonize those. 

Then there's that big wedge that just says “industry.” My sense is that, to a lot of people, that is a bit of a black box — it’s not clear what's in it or how to approach decarbonizing it. Historically, that has been the neglected stepchild of the decarbonization conversation. But am I right in saying that attention on that little wedge has rapidly increased in recent years?

Rebecca Dell:  

Yes, and for people who work on this area, it's been exciting to see how much new interest has come in the last year or two. 

David Roberts:

Do you have an explanation for why?

Rebecca Dell:

The phenomenon that is more in need of explanation is why so few people were looking at this area until the last year or so, considering that the industrial sector globally, under the most parsimonious accounting, is responsible for a quarter of all greenhouse gas emissions, and under a broader definition, it's responsible for more than a third.

David Roberts:   

Does that roughly echo the US pie chart? Or is the US different because we have deindustrialized a little bit?

Rebecca Dell:  

The US is a little lower in terms of the portion of our emissions that come from the industrial sector. But if you add back in the greenhouse gas emissions that come from manufacturing products in other countries that will be consumed in the United States — you can think of those as our imported emissions — then you get back to something pretty close to the global average.

David Roberts:   

So let's say about a third — that's a lot of emissions to neglect for this long. When we say industry, what do we mean by that? What does that category inclue? What are the boundaries? And what, in terms of greenhouse gas emissions, are the top line items?

Rebecca Dell:  

That's a really important question, because when we talk about “industry” in the climate community, it’s a piece of stealth jargon. It’s the worst kind of jargon: it's a word that sounds like a normal word, but it actually is a jargon word. 

Basically, what we're talking about when we talk about industry is everything that's not agriculture or energy, which is to say, it's the material economy. It’s mining, manufacturing, construction, waste processing. It's physical stuff, as opposed to energy. 

As you might imagine, there are a lot of fields of human endeavor that are included in that very broad set of activities. It's a very heterogeneous sector. But for all of the millions of different types of activities that are included in the industrial sector, there's an astonishingly short list that are responsible for the overwhelming majority of the greenhouse gas emissions.

David Roberts:   

That's very useful for our podcast purposes.

Rebecca Dell:  

It is. It allows one to simplify one's focus considerably. There are three real standouts here: steel, cement, and commodity chemicals. 

The chemical industry itself is, again, varied and heterogeneous; they produce a lot of different products. But there are about 10 chemicals that are basically the precursors for two products — plastic and fertilizer — that dominate those emissions. You can think of this in four product categories: cement, steel, plastic, and fertilizer. Just making those materials is responsible for two-thirds of all the greenhouse gas emissions from the entire industrial sector.

David Roberts:   

Insofar as you figure out how to decarbonize those, will those lessons be transferable to all those other varied applications? Or are they so heterogenous that you have to do it one by one?

Rebecca Dell:  

The sources of greenhouse gas emissions are different in other areas. For example, a lot of the emissions in the waste processing area are what's called landfill gas: anaerobic digestion of poorly sorted solid waste trash leads to methane emissions. So that's in the one-third that's not accounted for. 

But a lot of it is manufacturing. It’s from lighter manufacturing activities: lower temperature processes, electric drive processes, cooling, motors, that sort of thing. A lot of that will be taken care of as the grid gets cleaner and as things that are relatively easy to electrify become more electrified.

David Roberts:   

If tomorrow, by magic, all electricity became clean, how much of that one-third of emissions would vanish?

Rebecca Dell:  

That's pretty much the difference between the one-quarter and the one-third numbers that I cited. For the one-quarter, the more parsimonious definition is “we are only looking at greenhouse gases that are coming out of smokestacks at factories,” what are called direct emissions. If you add in the greenhouse gas emissions from generating electricity that is consumed at industrial facilities, that gets you from a quarter up to over a third.

David Roberts:   

In terms of that quarter, how much of industry is devoted to fossil fuels themselves: mining, drilling, processing, transporting, refining, etc.? If we shift away from fossil fuels over time, how much of a chunk does that take out of the industry pie?

Rebecca Dell:  

None. 

The numbers I cited to you, the quarter and the third, those are global numbers. Here in the United States, we have a very unusual convention of including the fossil-fuel extraction industries as industrial activities. In the whole rest of the world, when people are doing their greenhouse gas inventories, they don't count that as an industrial activity; they count that as an energy transformation activity, so they lump those emissions in with power generation. 

If you look at that pie chart from EIA, or EPA — if you look at a strictly US source — that will include your refining and fossil-fuel extraction emissions, but global numbers don't include any.

David Roberts:   

That seems like a complication in comparing across countries, doesn't it? It's kind of a big chunk to have misfiled in one place or the other.

Rebecca Dell:  

Yeah, but we're America, and we like to do things our way.

David Roberts:   

Why do steel, cement, plastic, and fertilizer produce so many GHGs? 

Rebecca Dell:  

First, because we make them in larger quantities than we make anything else. These are the materials that we make other things out of. We make steel and cement in increments of billions of tons per year. We make commodity chemicals in increments of hundreds of millions of tons per year. These are the only products that we make in those volumes, so of course these are the products that have the biggest greenhouse gas impact. 

Second, all of these industries are a variation on the following theme: you dig something out of the ground and the first thing you do with it transforms a raw material into a useful molecule; everything that's downstream of that in your supply chain is arranging your useful molecules in different combinations and sizes and ratios. But all of that rearranging takes a lot less energy and emits a lot less greenhouse gas than making the useful molecule in the first place.

All of these industries are what we call primary commodity processing industries. In fact, if the big three that we talked about — steel, cement, chemicals — are the highest emitting industries, four through seven or eight are also primary commodity processing, just smaller ones. They're things like aluminum.

David Roberts:   

Let's look at those four: steel, cement, plastic, and fertilizer. Why does making steel specifically produce so much greenhouse gas? What is the traditional steel-making process?

Rebecca Dell:  

Steel emissions are so big because we make 2 billion tons of it per year. That's the best part of a thousand pounds of steel for every human being on Earth, every year. It sounds insane until you look at a suspension bridge, or a runway, or anything in our built environment. Then you have to think, oh yeah, I guess we do use an incredibly large amount of steel. We make everything out of it.

David Roberts:   

What is the raw material, and what is the processing that sends off so much greenhouse gas?

Rebecca Dell:  

With steel, we start with iron ore. Iron ore is iron oxide — iron atoms chemically bonded to oxygen atoms. Your audience may be more familiar with iron oxide by its common name, which is rust. 

Everybody knows that rust does not have the valuable material properties that steel has, so what we're doing when we make steel is stripping off those oxygen atoms and turning it into metallic iron, with a little bit of other elements mixed in to improve its properties. Steel is almost all iron by weight.

The main way we do that chemical reaction today is to use coal. We combine the iron ore and the coal together in a reactor called a blast furnace. We use metallurgical coal, which is also called coking coal. It's a special kind of coal, but it's still a lump of carbon. 

In the blast furnace, part of the coal gets burned for thermal energy to help the reaction go faster. All of those carbon atoms are a more attractive place for the oxygen atoms to go, so the oxygen atoms move from the iron oxide over to the carbon, and we get carbon dioxide. So we're getting carbon dioxide from two different sources. 

This is another theme that we'll see throughout the industrial sector: you have the energy emissions — the coal that you burn to get your reactor hot to make the chemical reaction go — but you also have a set of chemical reactions that are going on in there that are not combustion reactions. They're a different kind of chemical reaction that's also producing greenhouse gases. That’s what we call process emissions: any greenhouse gas emission that comes from doing anything except combustion.

David Roberts:   

My intuition tells me that energy emissions are going to be the easier ones to eliminate, because we have alternative sources of energy that don't emit greenhouse gases. Is that accurate?

Rebecca Dell:  

In many cases, yes.

It would be useful at this point to give a typology of solutions that applies across industries. There are a few buckets of decarbonization pathways that we can use across all of these industries. 

Bucket number one is material efficiency. We can just use less of this material in order to make the products and deliver the services that we want. 

David Roberts:   

Traditionally that's the cheapest, right? It's just changing your behavior, changing your processes, changing design.

Rebecca Dell:  

Yeah, that's a big one. The barriers there are typically not technical. They're barriers that have to do with incentives and social systems and cultural norms. That's very important, and we should definitely do it. 

Bucket number two, carbon capture and storage. You keep doing pretty much what you're doing now, but you figure out a way to collect all the carbon dioxide and put it underground. You don't have to like it, but you have to acknowledge that it exists and is a possibility.

David Roberts:   

I'm very familiar with capturing carbon dioxide off of combustion; that's the standard CCS model. Is capturing the carbon dioxide off of process emissions notably different or more difficult?

Rebecca Dell:  

There's a dumb version of carbon capture where you just take your flue gases at the end of the pipe and put them through some scrubbers and then put them through some amine sorbents, and you can do that on any flue gas. You could imagine doing that on the end of almost any pipe, but each industry has its own version of smarter carbon capture that is engineered to optimize for this industrial process. That varies a lot. 

Bucket three is hydrogen. As your recent guest Panama Bartholomy said, it is the answer to every question in energy before it has even been asked. 

Bucket number four is direct electrification. 

Bucket number five is bioenergy. 

Those are your five buckets across all of these industries.

David Roberts:   

Is there significance to the order you put them in?

Rebecca Dell:  

No. Well, I suppose I put bioenergy last because bioenergy cannot ever be more than a small part of the solution. There’s no way to provide enough biomass to do a large portion of the decarbonization in these industries. The IEA estimates that our current total biomass available for energy use on Earth is something like 55 or 60 exajoules of energy. The chemical industry today uses almost 50 exajoules of energy. The steel industry uses another 30 exajoules of energy. There’s just not enough to go around. Bioenergy might show up here or there, but it can't be the bulk solution, because there just aren’t enough joules there.

David Roberts:   

Bucket number one, material efficiency, applied to steel: I can imagine us using less steel. 

Rebecca Dell:  

One point on that: in the United States and in other high-income countries, we already use less steel. As countries get richer, their demand for steel tends to tail off. 

The reason for that is that as you become a middle-income country, that's when you build out an electric transmission and distribution system that actually gets to every house; you move your entire population into modern housing; people start having private vehicles for the first time in significant numbers; you build sanitation systems and aqueducts that bring clean water and public health to people. Once you've done that, your demand for steel is largely a function of population growth and replacing things as they wear out. 

We in high-income countries might have a lot of opportunities to reduce our demand for steel in order to be more material-efficient in our use of it, but there's this huge latent demand for more steel that is represented by the couple billion people on Earth who are still in poverty and have an entirely legitimate desire to have modern housing and sanitation and all of those things.

The other thing about steel, though, is that it's quite recyclable. As you have a lot of stuff for a long time, you develop a stock of steel that you can recycle. If current trends continue, all of the additional demand for steel that's going to come from countries emerging out of poverty can probably be met with recycled steel. The current volume of new steel production probably doesn't have to go up in order to continue to meet global needs, but it probably doesn't have to go down either.

David Roberts:   

Are we currently on a trajectory to hold it steady?

Rebecca Dell:  

If current trends continue. But I don't think I've actually said out loud yet in this interview how much greenhouse gas is emitted by the global steel industry. It’s 3.5 billion tons of carbon dioxide equivalent per year. It's more greenhouse gases than are emitted by any entire nation except the United States and China. Even if we're just holding current production constant, that's still an enormous problem to solve.

David Roberts:

By the same token, if you reduce it by even a small fraction, you are reducing a lot of tons.

Rebecca Dell:

Yes.

David Roberts:

What does smart carbon capture look like in steel? 

Rebecca Dell:

Frankly, there's not a lot of interest in steel CCS around the world right now. I'm happy to explain how it might work and why it would be hard. 

David Roberts:   

My attitude toward CCS is, don't do it unless you have to. If you're telling me you don't have to, I'm happy to put it out of my mind. 

Rebecca Dell:  

I think we can confidently walk past. Let’s move on to hydrogen. Hydrogen is where a lot of the excitement in the steel industry is right now. In my mind, the best argument for hydrogen is that making steel using hydrogen is the smallest increment of technology that we need to get to zero-greenhouse-gas steel.

David Roberts:   

Can you just substitute hydrogen into existing refining processes, or there's more to it than that?

Rebecca Dell:  

No. Today, more than 90 percent of what we call primary steel — steel that's made from iron ore, not recycled — is made with a blast furnace using coal. There's not a lot you can do about the blast furnace on the hydrogen front. 

But about 7 percent is made with an alternative process called direct reduction that uses methane instead of coal. We think that the hydrogen process might be quite similar to direct reduction and use quite similar equipment, so we don't have to start from zero.

This direct reduction process is only 7 percent of global production, but that still makes this a fully commercial, mature technology. You can call up the Midrex Corporation and say, “hello, I would like to buy a shaft furnace,” and they will make you one. 

Basically what we're talking about is reengineering this existing technology of the shaft furnace to use only hydrogen instead of what it currently uses, which is a mixture of hydrogen and carbon monoxide. If we're going to do that as our main route, we're going to have to build a lot more shaft furnaces. 

So the next option — direct electrification.

David Roberts:   

My favorite.

Rebecca Dell:  

There are a few different ways to do this. The one that's most advanced is something called molten oxide electrolysis, which is pretty much what it sounds like. You take your oxide (iron ore), you heat it up enough to melt it, you put a giant electric field across it, and the electric field is strong enough that it pulls apart the iron and the oxygen. 

This is pretty similar to how we currently make aluminum, so it's not crazy. It's a thing that should be able to be made to work. It's a pre-commercial technology, though — it's not ready for primetime yet. There are some companies that are working on it, it might be ready soon.

David Roberts:   

Are we to a demonstration project yet? Is it happening somewhere?

Rebecca Dell:  

Give it a year and there may be something exciting to announce.

David Roberts:   

Is that the cheapest form of direct electrification, or the most practical?

Rebecca Dell:  

That’s the one that’s closest to being ready to do. People are also thinking about low-temperature forms of electrolysis that you can do without having to melt the iron oxide first, but the one that I just described is the most mature. 

The biggest advantage of the hydrogen route is that it is the smallest increment of technology to get us to truly green steel. The biggest advantage of the direct electrification route is that it will require the least energy. If you are using green hydrogen — taking electricity and converting that into hydrogen — and then using that hydrogen to convert your iron ore into iron, you lose a lot of energy in that extra conversion. If you can just use the electricity directly, you get to keep another third of the energy.

The amount of energy that's involved is enormous. This industry uses a similar amount of global energy to its portion of global greenhouse gases, which is 7 or 8 percent.

David Roberts:   

None of these, except for carbon capture, seem to address existing blast furnace steel production facilities.

Rebecca Dell:  

Yeah, and this is incredibly important for the politics. In the steel industry, as in a lot of these industries, the reason why the facilities that we have today were built in the places that they were built is because they had the best access to raw materials and energy. Why did we put the steel mill there? Because we could get metallurgical coal to that place really cheaply. 

Also like a lot of these industries, most of the production is done at a relatively small number of very large facilities. I've visited a lot of steel mills over the years, and it's not unusual for a steel mill to have 10,000 employees. The biggest one I ever went to had 48,000 employees at one facility. It's the size of a city. 

David Roberts:   

So these are not things that you can move around easily. Switching geographies is not practical.

Rebecca Dell:  

It creates a lot of problems and a lot of social dislocation. The steel industry might not be a huge part of the total US economy today, but for the communities and even the states that are hosting these facilities, one facility is a really important part of your employment. People will fight really hard to keep these facilities, because they're so concentrated and the local community is so dependent on them. If you're moving from a coal-based process to an electricity-based process, frankly, the places that have the best access to metallurgical coal are not typically the same as the places that have the best access to cheap electricity.

David Roberts:   

So CCS is something you can offer these communities and these facilities to say, “you don't have to change anything fundamental, you don't have to move, you can continue to exist here and just bolt this thing onto your facility.”

Rebecca Dell:  

You can potentially maintain existing industrial economies, but it's not easy. 

One of the reasons why CCS is tough in the steel industry is that at what are called integrated steel mills, the traditional type of steel mills that have blast furnaces at them, typically you'll have three or four really big carbon dioxide sources — your blast furnace, some other major process furnaces, and things like that. Together, you have a lot of carbon dioxide coming out in one place, so you can see how it could be cost-effective to collect it all. 

But all of that together is often only half or maybe 60 percent of the carbon dioxide that's coming out of the facility overall. The rest of it is all these small sources — little process heaters here and there — that are distributed by the dozens all over a facility that's the size of a town. Thinking about how you would collect all of the carbon dioxide from all those distributed sources and do that cost effectively is really hard.

David Roberts:   

I came into this interview riding on a wave of green-steel hype, and nothing I'm hearing you say is justifying any of it.  What is all the excitement around steel? It sounds to me like we have no good options. 

Rebecca Dell:  

I didn't say anything nasty about hydrogen, did I?

David Roberts:   

I mean, it's going to substitute for that 7 percent with a special kind of furnace, right? Which is not nothing, but blast furnaces are most of the furnaces. 

Rebecca Dell:  

All of them are going to have to be retired.

David Roberts:   

That sounds like a brutal social and political process.

Rebecca Dell:  

I'm not going to claim that it will be straightforward, but we do have 30 years. Industrial equipment doesn't typically last longer than that. We don't have to do this all at once. But I have never seen a good idea for how we have a climate-safe blast furnace. 

We are in a process of closing most, if not all, of the coal-fired electricity stations around the world, and we all accept that this type of industrial equipment, this particular coal-fired type of furnace, is just not consistent with a climate-safe future. That is also true of blast furnaces.

David Roberts:   

So insofar as there's an elevator-pitch answer in steel, it is shutting down blast furnaces and building new facilities that either can work with hydrogen shaft furnaces or are some directly electrified process that we don't quite have worked out yet.

Rebecca Dell:  

We're getting close, though. That's probably how it's going to go. 

There are some interesting reports that came out in the last few months looking at pathways to steel decarbonization. Several different organizations have done this kind of analysis over the course of the last year, with different methodologies and approaches, and all of them basically come to the same conclusion: no new blast furnaces, and we're going to have to start shutting down the existing blast furnaces in pretty short order.

David Roberts:   

I'm guessing these new, less standardized and commodified processes are more expensive. What kind of delta are we talking about?

Rebecca Dell:  

This is a great question, and there are two answers. 

First, yes, we do expect that green steel and other green commodities will be more expensive than existing dirty means of producing them. Depending on who you ask, and depending on exactly which process you're talking about, those price premiums range from 20 percent up to 200 percent. 

That's okay — we pay for environmental performance all the time. Cars with catalytic converters are more expensive than cars without catalytic converters; we still think it's a good idea to put catalytic converters in our cars.

David Roberts:   

But if you're telling a country that's emerging out of poverty that it's going to be 200 percent more expensive for them to do so, that's not nothing.

Rebecca Dell:  

This leads me to the second point, which is that these industries — steel, cement, commodity chemicals — are incredibly valuable. The whole rest of the economy relies on the material that these industries produce.

However, from an economic perspective, they are extremely low-value-added industries. They have very tight margins. These are your classic commodity industries. The cost of these materials represents a very small part of the cost of the finished goods that are made out of them. 

From the perspective of the guy who owns the steel mill or chemical plant or cement kiln — sometimes it’s a girl, but usually a guy — he's like, “I have commodity-sized margins here. There is no room in my margin to pay for any kind of decarbonization.”

I would encourage you, however, to look at this through the other end of the telescope. Don't look from the perspective of the guy who owns the steel mill; look at it from the perspective of the person who's buying a car made out of steel. Even at a relatively high additional cost for decarbonization, that's only going to add a couple hundred bucks to the cost of your car. The average new car in the United States costs $37,000; $37,200 looks a lot more manageable.

David Roberts:   

This must have political economy consequences, too, right? If the steel mill owners can't get the car buyers on their side to rebel against this, how much power do they have on their own to politically resist these sorts of things? 

Rebecca Dell:  

The real political and economic problem here is not, “how do we afford to pay for decarbonization?” We can 100 percent afford to pay for decarbonization of steel and all of these other industries. The problem is, how do we pass the costs efficiently through the supply chain so that the place they land, the final consumer, is the person in the supply chain who can actually afford to pay.

David Roberts:   

They’re more dispersed the more you pass them down the chain too; less concentrated on any one constituency that might rebel against it.

Rebecca Dell:  

Yes. The policy challenge here is about how you pass those costs through. Ways that you can do that are things like product standards. Why do we have catalytic converters in our cars? In a practical sense, it's because you're not allowed to buy a car that doesn't have one. If you want to make sure that the costs of decarbonization get passed all the way through the supply chain, one way to do that is to have standards that require that products use clean materials.

David Roberts:   

Of course government procurement is always a huge piece of this too. Government can start that process. 

Rebecca Dell:  

Yes, the government can start by applying the standards to itself.

David Roberts:   

That's basically end users voluntarily taking on the cost, right? 

Rebecca Dell:  

I don’t think there's anything voluntary about my catalytic converter.

David Roberts:   

Well, policymakers deliberately choosing to put the costs on the final user so that it's less concentrated. The steel mill owners are all equally affected; none of them are being priced out relative to the others.

Rebecca Dell:  

To be philosophical for a second, there are only two pots of money in society: consumer dollars and taxpayer dollars. The question is, what ratio of consumer dollars to taxpayer dollars do we wish to use? That is going to change depending on circumstances, but that's the question.

David Roberts:   

This helps me have a more realistic view on steel, although it’s slightly dimmed my enthusiasm.

Rebecca Dell:  

I don't want to give you the impression that nothing's happening on steel. The Swedes have a project called HYBRIT, a hydrogen reduction steel project, which is the most advanced in the world. They recently announced that they made one of their mining vehicles entirely out of green steel — the first vehicle in the history of the world to be made out of green steel.

It’s only one vehicle, but the distance between one vehicle and two vehicles is a lot smaller than the distance between zero vehicles and one vehicle, and that distance keeps getting smaller over time. We're making real progress. We're not there yet — it's definitely at an earlier stage than our colleagues who are working on power or transportation — but we're making progress.

David Roberts: 

Let’s move on to cement. What is the raw material and what is the basic processing?

Rebecca Dell:  

The raw material is the main constituent of limestone. Limestone is a very common kind of rock; you can find it pretty much in any country. The main constituent of limestone is something called calcium carbonate. The main ingredient in cement is something called calcium oxide. You can hear right in the words — there is carbon in the limestone, there is no carbon in the cement. 

You dig up the rocks and cook them at 1,500℃, roughly 2,600℉. About 40 percent of your greenhouse gas emissions come from burning fuel to get your rocks that hot, and the other 60 percent, on average, is from burning off the carbon that was in the rock. The carbon coming out of the rock is your process emissions.

David Roberts:   

Once again, it's fairly easy to imagine the energy coming from a different low-carbon source, but the problem comes down to process emissions. When the carbon comes off the limestone and is released, is there some way of capturing it? Is there some way of doing this without releasing the carbon? What are the green cement options?

Rebecca Dell:  

Even if we decide that we don't want to use CCS in any other part of our economy, the place that we are most likely to end up relying on CCS as our primary decarbonization pathway is in the cement industry.

David Roberts:   

Are the emission sources more concentrated in cement than they are in steel?

Rebecca Dell:  

A cement kiln is a much simpler place than a steel mill. We were talking about steel mills with thousands of employees; if you go to a cement kiln, the typical number of guys on shift is maybe 25. You just have one big pipe in a cement kiln, so CCS is a lot more straightforward there. 

People do have ideas for alternative raw materials or alternative cement chemistries that might be able to address this process emissions problem without CCS, but it's probably going to be CCS. Part of that is my assessment of what the technical alternatives are, but an even more important reason is that cement and the thing we like to make out of it, concrete, are foundational to our buildings. It is literally the thing we make foundations out of. Almost every structure in our society relies on concrete, and the type of cement that we use, which is called ordinary portland cement, was first patented in 1824. We're coming up on its 200-year anniversary.

This is a material we feel very comfortable with. We know its material properties really well. And for obvious reasons, the construction industry is incredibly risk-averse about the structural properties of the things that it's building. Even if you have great ideas for alternative cement chemistries, the likelihood that the global construction industry would feel comfortable wholesale shifting over to them in 30 years time is a pretty tall order.

David Roberts:   

I can't imagine the process you would have to go through to demonstrate that your concrete would hold up buildings in every conceivable situation. 

Rebecca Dell:  

We do have performance-based standards for concretes and ways that we test different types. I don't want to say that there's nothing to be done here. 

The main ingredient in cement is something called clinker, and we already use a big range of different clinker factor — that's the percentage of clinker — in different cements around the world. Almost all the carbon dioxide comes from making the clinker. A lot of cement is 95 percent clinker, but it's also very common to use 65 percent clinker cements. You can cut 30 percent off your greenhouse gas emissions by using low-clinker cements, and those things are already technically mature and well-demonstrated — there are big structures made out of them that you can go and visit around the world. 

So there’s an opportunity to make at least 30 percent greenhouse gas emissions reductions, on average, just by going to the lowest clinker factor that's appropriate for whatever you're using. There's no technical barrier. It usually is cheaper. We should do it tomorrow — there's no reason not to. But a 30 percent emissions reduction still leaves you with 70 percent.

David Roberts:   

What about bucket one, using less? Is there a practical way to use a lot less cement? 

Rebecca Dell:  

Oh my god, we are so wasteful in the way that we use concrete. 

People have gone out to actual commercial and multifamily residential buildings and looked at how much structural material, primarily concrete, these buildings are using compared to how much structural material would be needed to support all the loads. They typically find that there is something in the neighborhood of twice as much structural material as is needed to comply with the very safety-protective building codes. 

Almost all of the studies that I've seen have been in Europe or the United States, so it's mostly in high-income countries that these numbers come from. We think the situation is probably even worse in developing countries. 

I live in the San Francisco Bay area. For a commercial or multifamily construction site in this area, certainly it's more expensive here than in other parts of the country, but it's not radically different. Depending on the size of the building, the typical payroll for one of these construction sites might be $5,000-$10,000 an hour. To get one of those mixer trucks full of concrete delivered to your construction site — we're not talking about fancy concrete here, just normal commodity concrete — that's about $1,000. So if you can save five or 10 minutes of time on your job site by wasting a truck full of concrete, you just saved money. 

This goes back to what I was saying about looking through the other end of the telescope. Why would you use materials efficiently when they are so cheap? For private construction here in the United States, the average amount of a construction project that is represented by the cost of the cement is less than 0.5 percent.

David Roberts:   

So cement is so cheap that people overuse it to save time, to save soft costs …

Rebecca Dell:  

To save anything. Everything is more expensive than cement. If you take an 18-wheeler and you fill it up to the statutory maximum weight for driving on an interstate highway in the United States, it will have approximately $2,600 worth of cement in it. It will have a one nice laptop worth of cement.

David Roberts:   

This does seem like an area where markets could work well. You want to put a higher price signal on it and then trust people to figure out how to eliminate some of it. Is that right?

Rebecca Dell:  

It's a good news / bad news situation. Because cement and all of these materials are so cheap, it is very hard to persuade people to use them efficiently. It is very hard to persuade people to value them in terms of the actual value that they provide to their lives. It's a little bit like water or electricity that way. 

However, because they're so cheap, the good news is that even if the green version is a little bit more expensive, or even a lot more expensive, than the dirty version, that doesn't actually make the products that these materials are made out of more expensive.

If we go back to this commercial building I was talking about, 0.5 percent of its costs are cement. Let's say we mandate dumb end-of-pipe CCS, the most expensive, worst engineering option that we can think of, for our cement decarbonization. We involve only lazy engineers in our project.

Even under that circumstance, maybe we'll double the cost of the cement — that only adds 0.5 percent to the cost of the building. In fact, it adds less than 0.5 percent to the cost of the building, because all of the construction costs that I've been citing don't include the cost of the land, which is often the most expensive thing.

David Roberts:   

So I was precisely wrong — it's probably going to be difficult to put a pure price signal high enough to make the market work. But you can get away with mandates, because it's not going to affect consumers much.

Rebecca Dell:  

Yes. This is part of what I was saying earlier — these industries are incredibly valuable, but because they're low-value-added, compared to the prices that consumers actually face, these materials are not typically an important line item.

David Roberts:   

Let's talk about chemicals. I know it's a varied category. Are there simple things to say about why chemical processing produces so much greenhouse gas, or does it vary a lot also?

Rebecca Dell:  

The chemical industry is very diverse. I remember talking to a colleague who was a senior sustainability person at BASF, the German chemical giant, and she told me that BASF has an 80/20 problem when it comes to their greenhouse gases. BASF makes approximately 100,000 products. Eighty percent of their greenhouse gases come from not 20 percent of their products, but from 20 products. 

Again, these are the basic materials that are the ingredients for all their other products — mostly fertilizer and plastic. 

For fertilizer, primarily we're talking about nitrogen fertilizer, so making that cleanly is mostly about making hydrogen cleanly. People do have some ideas for how to make nitrogen fertilizer that's not made out of ammonia, but the main idea is, if you want to make clean ammonia, you just need to start with clean hydrogen.

David Roberts:   

If you solve the green hydrogen problem, you solve the fertilizer problem downstream?

Rebecca Dell:  

Yep. There's not a whole lot to making fertilizer besides making hydrogen.

David Roberts:   

That's convenient. Plastics, I assume, are more difficult.

Rebecca Dell:  

Yes. With plastics, you have the same buckets of solutions we were talking about earlier. We could use less, and definitely we should. 

Plastics are interesting, because they’re carbon-based molecules; they're made out of a carbon-based material. When we make plastics out of fossil fuels, some of the fossil fuels are burned to provide energy, but for more than half of the fossil fuels that we use in plastics production, we're actually taking the carbon atoms and the hydrogen atoms that are in the fossil fuels and we're putting them into the plastic product. We're making the product out of the fossil fuels.

David Roberts:   

So every piece of plastic is, in some sense, carbon sequestration.

Rebecca Dell:  

You know, Shell says that all the time.

David Roberts:   

I take it back. Does it release the carbon when it breaks down?

Rebecca Dell:  

There is a scenario in which, if you collected all the plastic at the end of its life, and you made sure that it was clean and dry and well-segregated, and you put it in a nicely lined hole in the ground, it would be inert in that hole for a very long time and technically you could call that carbon storage. 

But that's not actually what we do with plastic at the end of its life, and the way that we actually treat plastic at the end of its life leads to a lot of greenhouse gas emissions.

David Roberts:   

What do we do? Do we burn it?

Rebecca Dell:  

Some of it we burn. That's like burning fossil fuels directly, and that's a very popular option around the world. 

Here in the United States we mostly put it into mixed garbage. When you have plastic and organic material mixed together in your garbage, the organic material, food waste, will decompose anaerobically. All the carbon atoms that were in that organic material will then leak out as methane instead of as carbon dioxide.

Methane, depending on the timeline you're looking at, is somewhere between 30 and 85 times more potent of a greenhouse gas than the carbon dioxide that you would get if you properly composted your organics. 

So even if the carbon atoms in the plastic are not decomposing quickly and turning into carbon dioxide, they are leading to methane emissions from trash, which are an important source of overall greenhouse gas emissions.

David Roberts:   

It seems like here, bucket one is the lowest-hanging fruit by far. We're so wasteful. Plastic is so gross and overused.

Rebecca Dell:  

In the United States, the EPA estimates that only 8 or 9 percent of plastic is even collected for recycling, and of that, only about half is actually recycled in any form at all. Almost always, the recycling process is that you have a wide variety of mixed plastic, you melt it down, and when you lump all of these different materials together you get very, very low-quality plastic, radically downcycling. 

Most of the plastics we use are, in theory, infinitely recyclable. If you have a high-purity waste stream, you can melt it down and get new, first-quality products that are just like the old ones. But we don't do that. 

We need to use less plastic, but we also need to have tight regulations on exactly what types of plastic can be used, so that there are only a few types out there and all plastic packaging is the same couple of types, so they can be easily segregated and meaningfully recycled. 

David Roberts:   

We can change the way we design plastic products and the types of plastic we make to encourage more recycling, but obviously you're never going to get to zero that way. Is there a way to avoid, bucket two, or are we stuck with CCS here? Are there real alternatives on the horizon to carbon plastic?

Rebecca Dell:  

Bioplastics are real. I occasionally will encounter a PLA fork or something like that. They're not a meaningful portion of current plastic production. And as we were talking about before, there's just not enough biomass to go around to make large quantities of plastic out of biomass, so that's going to be a niche item forever. 

We can take carbon atoms out of carbon dioxide and turn them into plastic; it requires an eye-watering amount of energy. This is important for carbon-utilization conversations generally: imagine we started with these big, exciting, highly energy-dense fossil fuel molecules; we had a combustion reaction, where we took out all of the energy that was stored in those molecules; and what we were left with was carbon dioxide, which was a combustion product. It's what's left over after you take out all of the energy.

If you want to turn it back into one of these big exciting molecules, you have to put more energy back in than you got out in the first place from burning it. The chemicals industry is the most energy-consuming industry of any industry in the world. It's only the third-most greenhouse gas emitting, because a lot of that energy is stored in the product and doesn't go directly into carbon dioxide.

A couple of years back, the big pan-European chemicals industry trade association published this fantastic report where they said, “okay, you guys want us to decarbonize, let's get serious about what that actually would be. Let's go through one process at a time and talk about the energy and feedstock requirements for the green alternatives in every case.”

What they found was that to produce the basket of chemicals that they were currently producing and to do it using carbon dioxide as their primary source of carbon would require something like 1,900 terawatt-hours per year of clean electricity. The IEA estimates that in their Paris compliance scenario in 2050, the total amount of electricity that is generated and used for all purposes on the continent of Europe is only about 3,400 terawatt hours per year.

More than half of all the electricity would have to go to the chemicals industry if you wanted to make all of your carbon-based chemicals out of carbon dioxide. 

So, it can be done, but we are really in too-cheap-to-meter territory with our electricity if we're doing that.

David Roberts:   

None of those sound like good options. What's your favorite for plastics? 

Rebecca Dell:  

It's got to be using less, material efficiency. I have seen no scenarios where you can get 1.5℃ or even 2℃-consistent reductions in emissions from the chemicals industry without dramatically reducing the amount of plastic that we use and dramatically increasing the quantity and quality of recycled plastic.

David Roberts:   

With steel, you mentioned that when you're developing as a country there are a lot of big one-time uses and then your usage tails off. Is there an arc for plastics?

Rebecca Dell:  

Not that we’ve been able to find. It’s just up and up and up. Since 2015, the rate at which total global plastic production is increasing has stopped accelerating.

David Roberts:   

I guess that's good news?

Rebecca Dell:  

The problem is not getting worse faster. It's just getting worse at the same very rapid rate that it was previously getting worse at. That's the nicest thing I can say about the trend for plastic production volumes.

David Roberts:   

I do want to touch on some policy options. It sounds like if we're looking big picture, at industry decarbonizing by 2030, the most difficult area is plastics. Is that roughly accurate, or they're all difficult?

Rebecca Dell:  

I don't like to think of any of them as difficult. I find that framing both unhelpful and inaccurate, because people just started noticing the importance of the industrial sector about a year ago. I often tell people that where we are in our decarbonization progress in these sectors is similar to maybe where the power sector was in the late 90s. 

I don't know, I was a kid then, but I'm assuming that in the 90s the concept of completely decarbonizing the power sector probably felt pretty hard to people who were out there trying to get solar panels installed and being called silly hippies. We have 20 or 25 years of progress that we've made since then.

It's not going to be easy, it's not going to happen by itself, but we have a line of sight to where we're going. We see how it's going to happen.

The situation in the industrial sector is not that it's somehow inherently harder. We're just at a much earlier stage in our decarbonization journey.

David Roberts:   

We waited a long time to get started, though. We do have to go faster in it than we did in electricity, arguably.

Rebecca Dell:  

That is true. We did take our sweet time to get started.

David Roberts:   

You're closely in touch with political and policy angles on this — do you see urgency around this commensurate with the scale and speed necessary to do it? 

Rebecca Dell:  

I mean, obviously not. Even the parts of the climate challenge that we're doing the best at we're not on track, and this is not one of the parts that we are doing the best at.

It is very clear to me that what we need to do to decarbonize these industries is entirely within our capacities here in the United States and also globally. Please don't interpret what I'm saying as any disrespect to the efforts of the Biden administration. The people who are doing this work in the Biden administration are very, very clear about what the scale of the challenge is, and they are attempting to move as fast as they possibly can. But they would probably be the first people to tell you, “what we're doing is not enough.”

David Roberts:   

Let's look at what we are doing, then. We had executive actions early on, we had the Recovery Act, then we got the Bipartisan Infrastructure Act. Are there big pieces of good policy on this that have already been passed? Secondly, are there good pieces of policy on this in Build Back Better that we, like everyone else, are sitting around waiting forever for action on? 

Rebecca Dell:  

The biggest thing that was in the bipartisan infrastructure law is a serious pile of money for commercializing and demonstrating clean industrial technologies. Most of that is going through the Department of Energy, and the way that the money was allocated is pretty flexible. The DOE currently has a lot of discretion about exactly how they spend that money, and there are a few different pots of it, so it's hard for me to give you a dollar amount that will go to the industrial sector, but it will be somewhere between half a billion and a few billion dollars. That's a serious amount of money.

David Roberts:   

It does sound like in some of these markets or sub-markets we are at that point where a visibly successful demonstration project could be triggering, could unleash things.

Rebecca Dell:  

It’s a thing we need really badly, and it's a thing that absolutely requires public money. There's a certain amount of technology risk that the private sector in these industries, in particular, is simply not going to pay for. 

David Roberts:   

What about Build Back Better? Is there some pot of gold at the end of that?

Rebecca Dell:  

It’s a much larger pot of money in Build Back Better. We go from a minimum amount of industrial decarbonization demonstration projects of $0.5 billion currently up to a minimum amount of $4 billion if Build Back Better gets passed. Then the upper limit, depending on how you count it, goes up commensurately. 

The other important thing is that Biden issued an executive order last month on federal sustainability which included for the first time direct instructions for the federal government to buy low-greenhouse gas building materials — read: steel and cement — when it builds stuff with federal money.

David Roberts:   

That’s not a small thing. That's a very big customer, right? 

Rebecca Dell:  

We call the family of policies where the government buys low-greenhouse gas building materials “Buy Clean.” If you look across all levels of government — federal, state, and local — almost half of all the cement in the United States is purchased with taxpayer dollars.

David Roberts:   

In other words, Buy Clean government policy could do a lot.

Rebecca Dell:  

If it's well-structured and aggressively implemented, it could make a huge difference. Build Back Better, in addition to demonstration, has a bunch of money in it to facilitate the implementation of Buy Clean.

David Roberts:   

Just at the federal level, or helping states or cities too? Presumably, government at any level could do a little bit of this.

Rebecca Dell:  

There are a lot of spillover benefits. If the federal government says, “we're going to do this,” that makes it much cheaper and easier for state and local governments to do it, even without direct federal subsidies.

For example, the federal government has to put in place the measuring and reporting frameworks for the greenhouse gas intensity of different products; they have to make sure that the low-carbon products are available wherever federal construction is happening. All of the fixed costs of getting the system up and running can be accepted by the federal government.

David Roberts:   

All of which makes it easier for the next person to do it.

Rebecca Dell:  

There's also a lot of exciting stuff happening at the state level. California was the first state to pass a Buy Clean law, but since then, five other states have passed Buy Clean laws of one type or another.

David Roberts:   

Mostly cement and steel?

Rebecca Dell:  

The specific set of materials that's covered varies from state to state. Some states it's cement only; some states have steel, cement, and other things; in California, unfortunately, it's everything except cement. The cement industry had good lobbyists.

David Roberts:   

Is this the sort of thing where if enough states get in on this, they're eventually going to force a sea change?

Rebecca Dell:  

These are concepts that need to be proved out. If you can have a state policy that leads to widespread use of comparatively very low-greenhouse gas building materials, it becomes a lot easier for the EPA to start regulating related issues. The goal here is to create a virtuous circle of greenhouse gas ambition.

David Roberts:   

On one side you have investment for demonstration projects and setting up these systems; on the other side, for demand-pull, you have Buy Clean. Are there other big-ticket policy items that have not yet been tackled?

Rebecca Dell:  

There are a lot of different ways to structure the investment side. You can do credit subsidies; you can do direct subsidies; you can also do direct federal investment, which we have done a lot of in years past and in fact, the Defense Production Act allows us to do an almost unlimited amount of, if we wanted to. There are a lot of good arguments to be made for direct federal investment in clean production. All of those things are really important. 

There are also some important governance issues. A lot of these industries and markets have had pretty poor enforcement of existing regulations, both around non-greenhouse gas pollution and around labor standards. We have some pretty good rules on the books that are very poorly enforced. If we want the energy transition and the clean transition across the economy to be sustainable politically, we have to be showing people real, direct benefits in their lives and their family's health. That has to be an important part of this conversation. 

Also in the governance bucket, the United States is very bad at industrial policy. It was not always true, but for the last 40 years or so, we've had this weird fantasy that we don't do industrial policy. We definitely do industrial policy, but it's incoherent and easy to be captured by the covered industries because we’re pretending that we're not doing it.

One of the consequences of this is that we have entirely hollowed out the expertise and the governance infrastructure of industrial policy, particularly at the federal level. There's hardly anybody whose job it is to think about these things in the federal government, compared to other countries whose manufacturing sectors we would like to emulate.

David Roberts:   

Look at Germany. It's all very explicit. It's right up front. They're very clear about what they want to do and how they're going to do it. It's so much more sensible.

Rebecca Dell:  

And they spend a lot of money on it. The main applied R&D in the manufacturing sector that's from the German government is a system of things called the Fraunhofer Institutes. They spend almost 3 billion euros a year on the Fraunhofer Institutes.

The analogous thing in the US government is the Advanced Manufacturing Office, which has an annual budget of $400 million. Our economy is five times larger than Germany's, so compared to the overall size of our economy, we are spending less than 5 percent of what they are spending on applied R&D. And that's the piece of industrial policy that we feel most comfortable with!

David Roberts:   

That cuts across every sector, right? We constantly talk about goals and targets, but the capacity to do things on purpose with our economy has been hollowed out. 

Rebecca Dell:  

This is particularly true in the manufacturing sectors.

David Roberts:   

Is that largely because we exported so much of it? 

Rebecca Dell:  

I might make the causal relationship go in the other direction. One of the reasons why we had a lot of deindustrialization was that we didn't have a concerted effort to maintain a vibrant industrial economy. For example, Germany still has most of its steel mills.

David Roberts:   

If anything, we deliberately accelerated the reverse process with trade deals and things like that. That seems like a long-term project, reversing that process. 

Rebecca Dell:  

And an important part of that is rebuilding our governance capacity. It blows my mind that the highest-ranking person in the federal government whose job it is to think about the future of the US manufacturing sector — the head of the Advanced Manufacturing Office at DOE — has the rank of Office Director.

There isn't a single Assistant Secretary anywhere, in any department, on this beat. That is wild. There are 20 million Americans employed in this sector.

David Roberts:   

The final piece I want to look at is international trade. Presumably either we or other countries are going to start using trade deals as an instrument of decarbonization in industry. Is that something we're trying to do, or that people are talking about?

Rebecca Dell:  

It is. You may recall that Donald Trump, when he was president, put tariffs on steel and aluminum, just because he felt like it. Late last year, around the same time as the big climate meeting that happened in Scotland, the US and EU made an announcement about how they are working together on a deal to transform those tariffs into something that is mutual and linked to greenhouse gases. 

There's definitely a lot of work happening in this space. We have not yet settled on what the best policy tools are to promote decarbonization. For solar panels, some people say “we should have free trade in solar panels so that there are cheap solar panels and everybody can have the cheapest possible clean electricity.” Other people say “no, if we want to decarbonize we should have high tariffs on solar panels so that countries can have employment and manufacturing and broader social benefits, which will make the whole country more supportive of solar power.”

We still have a lot of work to do to figure out what a truly climate-safe vision for trade policy is. There's a relatively narrow set of policies that are traditional trade policies. In most cases, for things like tariffs, that's often less important than, what are the international reverberations? What are the trade consequences of purely domestic policies like subsidies and procurement policies? 

David Roberts:   

In terms of stimulating global movement toward industrial decarbonization, our biggest tools are probably still domestic. Doing it as fast as we can and making those products cheaper, rationalizing the industry, etc., is probably going to be a bigger deal than any tariffs we put on.

Rebecca Dell:  

Our goal is usually not that the trade policy itself will promote decarbonization, but that we can put in place trade policies that will prevent international trade from undermining our purely domestic policies, so different countries can push their industries to decarbonize faster without having to worry that dirty production from overseas will flood into the market.

David Roberts:   

How do you prevent industry from just moving, or shifting production? 

Rebecca Dell:  

This is one of the reasons why things like procurement policy are so fantastic, because when the regulation is on the product, not on the facility, there's no incentive to move the facility. The market-creation policies allow you to sidestep some of these difficult questions. 

People talk a lot about how businesses hate regulations. Businesses don't hate regulations; businesses hate regulations that they have to comply with, but their competitors don't. 

David Roberts:   

Well, I can not thank you enough. I know 10 times more about this now than I did when we started, so I really appreciate you taking all the time. Maybe someday we'll drill down a little deeper into one of these many, many rabbit holes that we tripped so lightly over in this conversation.

Rebecca Dell:  

I would be more than happy to talk in greater detail in the future. As you can probably tell, talking about this is one of my favorite things to do.

David Roberts:   

It's so fun. Thanks so much, Rebecca.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.volts.wtf/subscribe