Daniel's been on a materials kick lately, and today he's taking us into the heavy stuff. Here's what he wrote.
"We talked in various episodes about the sustainability of different materials. We've spent more time on plastic than anything else — we covered the Eurobox and Dolav containers, questioned whether non-single-use plastics can have any role in a sustainable world. We looked at various wood species. Today, let's look at the heavy stuff. Namely, metals like stainless steel, aluminum, and others. Because when we looked at heavy-duty plastics used in industry and questioned what could replace plastic, the first thing people think about when we mention weight-bearing requirements is steel. So that begs the question: Where does stainless steel come from? We do not see it growing on trees. And what is the metallurgy that provides this metal? From a broad sustainability perspective, how does that compare in environmental harm to the production of plastics? As we mentioned in our episodes about plastics, when we talk about metal or use words like steel, we are often using linguistic shortcuts for a heterogeneous material with variants that might be very different in production. Let's talk about the sustainability question here."
So we've talked plastic to death on this show. Today Daniel wants us to look at what people reach for when plastic fails.
And he's right to point out the heterogeneity thing upfront. People say "steel" like it's one substance, and it's... no, it's a family. A big, argumentative family with very different carbon footprints.
The metallurgical mafia.
Something like that. So let's start with the obvious question — where does stainless steel actually come from?
Not trees, apparently.
Not trees. It starts with iron ore — hematite, magnetite — mined mostly in Australia, Brazil, China, India. You dig enormous holes in the ground, crush the rock, separate the iron oxide from everything else. Then you feed it into a blast furnace with coke, which is basically baked metallurgical coal, and limestone. The coke does two things — it provides heat and it strips the oxygen off the iron oxide. What comes out the bottom is pig iron, which is about four percent carbon and brittle as anything.
Pig iron. Because it sounds unappealing?
The name allegedly comes from the molds looking like a sow with piglets. But yes, also unappealing. Pig iron then goes into a basic oxygen furnace where you blast pure oxygen through it, burn off the excess carbon, and you get raw steel. That's the conventional route. Steel production accounts for about seven percent of global CO2 emissions. It's one of the single biggest industrial sources.
Seven percent. That's... more than I expected.
It's enormous. And that's just the iron and carbon part. Stainless steel adds a whole other layer. The defining ingredient is chromium — at least ten and a half percent by mass. Chromium is what makes it stainless. It forms this passive chromium oxide layer on the surface, invisible, a few atoms thick, and that layer prevents oxygen from reaching the iron underneath. No oxygen, no rust.
So the chromium is doing all the heavy lifting, chemically speaking.
And chromium comes from chromite ore, mined mainly in South Africa, Kazakhstan, India. South Africa alone has something like seventy percent of the world's known chromite reserves. So suddenly your stainless steel supply chain runs through a handful of countries, and the mining itself has its own environmental footprint — water pollution, tailings, energy for crushing and transporting.
And that's before we even get to the variants Daniel mentioned.
Right. So the most common stainless steel is type three-oh-four — eighteen percent chromium, eight percent nickel. It's what your kitchen sink is probably made of. Then there's three-sixteen, which adds molybdenum — about two percent — and bumps the nickel to ten percent. Three-sixteen is what you use in marine environments, chemical plants, anywhere you need serious corrosion resistance. And then there's four-thirty, which is seventeen percent chromium and zero nickel. It's cheaper, it's magnetic, but it's less corrosion-resistant.
And the sustainability differences between these?
Massive. Nickel is the big one. Nickel mining and refining is energy-intensive and environmentally messy — think acid mine drainage, sulfur dioxide emissions. Producing the nickel for a kilogram of three-oh-four stainless adds roughly five kilograms of CO2 equivalent per kilogram of steel. Molybdenum is even worse — it's a rare metal, mined in relatively few places, and the ore grades are low, so you're moving enormous amounts of rock for tiny yields. A kilogram of three-sixteen has a significantly higher carbon footprint than three-oh-four, and three-oh-four is higher than four-thirty.
So the corrosion resistance you're buying is also buying you a bigger carbon bill.
Literally. And the trade-off is real — if you use four-thirty in a marine application, it'll rust through in a few years and you'll replace it, which means you've just doubled or tripled your actual environmental impact. The longer-lasting alloy can be the better environmental choice even though it's dirtier to produce. That's the kind of second-order thinking that makes this whole comparison tricky.
Okay, so that's the steel side. What about aluminum? Because aluminum has this... halo around it. People think it's the green metal.
Aluminum's green reputation is one of the best marketing jobs in materials history. Here's the reality. Aluminum comes from bauxite ore, which is strip-mined — huge environmental impact right there, deforestation, soil loss. The bauxite goes through the Bayer process to produce alumina, which is aluminum oxide. Then the alumina goes into the Hall-Héroult process, which is electrolysis — you dissolve the alumina in molten cryolite and run a massive electric current through it to split the oxygen from the aluminum.
How massive?
About fifteen megawatt-hours per tonne of aluminum. Fifteen. That's roughly what a typical American household uses in a year and a half. Aluminum smelting accounts for about two percent of global CO2 emissions but about three percent of global electricity consumption. It's so energy-hungry that smelters are deliberately located near cheap electricity — Iceland, Quebec, the Pacific Northwest, places with abundant hydropower. Because if you're paying retail electricity rates, you simply cannot compete.
So the carbon footprint of aluminum depends entirely on what's powering the grid where the smelter sits.
Hugely. Virgin aluminum produced with Chinese coal-fired electricity can emit seventeen kilograms of CO2 equivalent per kilogram of aluminum. The same aluminum smelted with Icelandic hydropower might be closer to four or five. Same metal, same process, totally different carbon bill. That's the heterogeneity problem Daniel was getting at, and it applies to aluminum even more dramatically than to steel.
And that's before we even start comparing metals to plastics.
Right. So let's stack them up. Plastic production emits roughly one point eight to three and a half kilograms of CO2 equivalent per kilogram, depending on the polymer. Virgin steel from a blast furnace is about one point eight to two point five kilograms per kilogram. Virgin aluminum, global average grid, is twelve to seventeen kilograms per kilogram.
Wait. So on a per-kilogram basis, plastic is... roughly comparable to steel, and both are way better than aluminum?
On a per-kilogram basis, yes. But that's where most popular coverage stops, and that's where it goes wrong. Because a kilogram of steel does very different work than a kilogram of plastic. A steel I-beam in a building lasts fifty, sixty, a hundred years. A plastic Eurobox in industrial use might last ten to fifteen years before it cracks or degrades. A single-use plastic bottle lasts about twenty minutes.
So per year of service, the math flips.
It can flip dramatically. And then you add recycling. Steel is the most recycled material on Earth — about eighty-five percent of steel gets recycled at end of life. And here's the key metallurgical fact: steel can be recycled infinitely without any loss of quality. The iron atoms don't care how many times they've been through a furnace. You can take a nineteen-fifties car chassis, melt it down, and make a twenty-twenty-six bridge girder, and the properties are identical.
Because you're not changing the atomic structure, you're just... re-melting.
Same for aluminum — infinitely recyclable, and recycled aluminum uses ninety-five percent less energy than virgin. The aluminum can you recycle today could be back on a shelf as a new can in sixty days. The recycling rate for aluminum cans is about seventy-five percent, though it's lower for other aluminum products.
And plastic?
Plastic recycling globally is about nine percent. Nine. And unlike metals, plastic degrades with each recycling cycle — the polymer chains break, the material gets weaker, you have to downcycle it into lower-grade products. A plastic bottle doesn't become a new plastic bottle; it becomes park bench slats or carpet fiber, and then eventually it becomes landfill.
So the recycling infrastructure difference alone is enormous.
It's the difference between a mature, economically self-sustaining system and one that barely exists. Scrap steel has real economic value — there's a whole industry built around collecting it, sorting it, and feeding it into electric arc furnaces. Nobody's building a business around collecting used polypropylene food containers because the economics don't work.
And that's where the heterogeneity within steel gets really interesting, right? Because not all steelmaking uses the same process.
Right. The blast furnace route I described — iron ore, coke, basic oxygen furnace — that's the high-emissions path. But there's a completely different route: the electric arc furnace, or EAF. EAFs don't use iron ore at all — they melt scrap steel with enormous electric arcs. An EAF running on a hundred percent scrap can produce steel with about zero point five kilograms of CO2 equivalent per kilogram. That's a quarter of the blast furnace emissions.
So you go from two point five kilograms down to zero point five, just by changing the furnace type.
And the feedstock. But here's the catch — EAFs can't produce every grade of steel. High-quality automotive sheet steel, the kind used for car body panels that need perfect surface finish and precise formability, that still typically requires the blast furnace route or direct reduced iron. You can't get there with scrap alone because the tramp elements — copper, tin, other contaminants that accumulate in scrap — mess up the properties.
Tramp elements.
That's the actual metallurgical term. Tramp elements. They're the unwanted hitchhikers that come along with scrap steel, and they concentrate over multiple recycling cycles. For structural steel, rebar, things like that, it doesn't matter. For the exposed body panel of a car, it matters a lot.
So even within "recycled steel," there's a hierarchy of what you can actually make with it.
And the same kind of heterogeneity exists in plastics, but the difference is that steel's recycling system actually functions at scale. The EAF route accounts for about twenty-eight percent of global steel production, and that number is growing. Meanwhile, plastic recycling is stuck at single digits and the economics keep getting worse as oil prices drop and virgin plastic gets cheaper.
Let's talk about some concrete comparisons. Daniel mentioned the Eurobox context — heavy-duty industrial containers that need to bear weight. If you're choosing between a steel container and a plastic one for that application, how do you think about it?
For a heavy-duty, weight-bearing, long-lifetime industrial application, steel wins on environmental grounds almost every time — provided the container is designed for a long service life and will be recycled at end of life. A steel Dolav container might last thirty years in rough service. A plastic Eurobox might last ten to fifteen before UV degradation or mechanical fatigue kills it. Over thirty years, you're buying two or three plastic units for every one steel unit, and the steel unit gets recycled into new steel at the end.
And what about aluminum for that same application?
Aluminum is trickier. If it's virgin aluminum from a coal-powered smelter, the upfront carbon cost is so high that you'd need an extremely long service life to break even. But if it's recycled aluminum, or aluminum from a hydro-powered smelter, the calculus shifts. The problem is that aluminum is also softer than steel — for a weight-bearing industrial container, you'd need thicker walls, more material, which eats into whatever weight advantage you were hoping for.
There's a case study I want to throw at you. Stainless steel water bottle versus plastic water bottle. People buy the steel bottle thinking they're saving the planet. Are they?
They are, but the break-even point matters. A stainless steel water bottle needs to be used somewhere between fifty and a hundred times to break even on carbon compared to single-use plastic bottles. After that, every use is a net win. And most people who buy a reusable bottle use it hundreds or thousands of times. So yes, it's a genuine environmental improvement — but only if you actually use it. The person who buys three stainless steel bottles because they like the colors and leaves two in a cupboard has undone the benefit.
The consumption pattern matters as much as the material.
That's the thing I keep coming back to. We focus on the material because it's tangible — you can hold the steel bottle and feel virtuous. But the use pattern, the lifespan, the end-of-life pathway — those matter just as much, and they're harder to see.
What about automotive? Cars are switching to aluminum bodies for weight savings. Is that actually better?
It's a classic trade-off. Aluminum body panels reduce vehicle weight, which improves fuel economy or extends electric vehicle range. But the aluminum itself has higher production emissions than the steel it replaces. The break-even point is roughly fifty thousand kilometers of driving — after that, the fuel savings outweigh the higher production emissions. For a vehicle that'll be driven two hundred thousand kilometers, aluminum is a net win. For a low-mileage vehicle, steel might actually be better.
Fifty thousand kilometers. That's about three years of average driving.
Right. So for most cars, aluminum makes sense. But again — it depends on where the aluminum came from and where the steel came from. If you're comparing recycled aluminum to virgin blast-furnace steel, aluminum wins easily. Virgin aluminum from a coal grid versus recycled EAF steel, steel wins easily. The heterogeneity never stops mattering.
We've established that metals have high embodied energy but long lifespans and good recycling, while plastics have lower embodied energy but short lifespans and terrible recycling. Is there a simple heuristic here?
I think the simplest one is: match the material to the use case. For long-lived, weight-bearing, structural applications — buildings, bridges, industrial containers, heavy machinery — steel is almost certainly the better environmental choice, especially if you can source EAF steel with high recycled content. For lightweight, short-lived, disposable applications — packaging, single-use items — the problem isn't really the material choice between plastic and metal, it's the disposability itself.
Because nobody's making single-use steel sandwich wrappers.
Right. And if they did, the environmental impact would be absurd. The material has to fit the function. The problem with plastic isn't that it's always worse than metal — it's that plastic's properties make it too easy to use for disposable applications, and our waste management systems haven't kept up.
What about the mining side? We've been focused on production emissions, but extracting iron ore and bauxite and chromite has its own impacts.
Both are bad, just in different ways and different places. Metal mining means open pits, tailings dams, water pollution from acid mine drainage, habitat destruction. The Samarco dam disaster in Brazil in twenty-fifteen released millions of cubic meters of iron ore tailings and killed nineteen people. Oil extraction for plastics means spills, fracking-related methane leaks, groundwater contamination, and the whole geopolitics of petroleum. It's hard to say one is categorically worse — they're both extractive industries with serious impacts, just concentrated in different geographies and affecting different communities.
Different timelines. An oil spill is catastrophic in the moment and for years after. A tailings dam failure is catastrophic in the moment and the heavy metals persist essentially forever.
Forever is not an exaggeration. Heavy metal contamination doesn't biodegrade. It just... stays there, working its way up the food chain.
Where does this leave us? Let me try to pull together a practical framework.
Go for it.
First, there is no universal "better" material. The choice depends on use case, expected lifespan, recycling infrastructure, and the energy source used in production. Anyone who tells you "always choose metal" or "always choose plastic" is selling something.
Agreed.
Second, for heavy-duty industrial applications like the Eurobox replacement Daniel was thinking about, steel is likely better than plastic if the product is designed for long life and will be recycled. But aluminum might actually be worse than plastic if it's virgin and used for a short-lived application.
The aluminum one surprises people, but the numbers don't lie. Virgin aluminum's carbon footprint is so high that you need a very long service life or very high recycling rates to justify it.
Third, the single most impactful factor is the energy source. Steel made with green hydrogen through the direct reduced iron process, or aluminum smelted with hydroelectric power, can have dramatically lower footprints. The question to ask is: where was this made, and with what energy?
That's a question most consumers can't answer, which is a real problem. The supply chain is opaque.
Fourth, the practical heuristic: for long-lived, weight-bearing applications, go with steel — especially recycled or EAF steel. For lightweight, short-lived applications, consider whether the use pattern is the real problem, not the material. A reusable plastic container you keep for twenty years might be better than a steel one you replace every two years because it rusts.
The heterogeneity lesson. Never say "steel" or "plastic" as if they're single materials. Always ask: which variant, made how, recycled how?
That's the thing I think Daniel was really driving at. We use these linguistic shortcuts and then make sweeping environmental claims based on them, and the reality is so much messier.
Messy is the word. I've been reading about this for years and I still find new wrinkles. The molybdenum supply chain alone — most people have never heard of molybdenum, but if you're specifying three-sixteen stainless for a chemical plant, you're implicitly making a bet on molybdenum mining in China and Chile and the environmental practices there.
Molybdenum doesn't grow on trees either.
It does not.
The misconception beat. What's the single biggest thing people get wrong about this?
I think it's the idea that aluminum is always more sustainable than steel because it's lighter. The weight savings only pay off in specific applications like transportation, and only after a certain mileage threshold. In static applications — buildings, containers, infrastructure — the weight advantage is irrelevant, and virgin aluminum has five to ten times the carbon footprint per kilogram. You're paying a huge carbon premium for a property you're not using.
That's a clean way to put it. Paying a carbon premium for a property you're not using.
The other one is that recycling solves the problem equally for both. Metals have mature, economically self-sustaining recycling systems. Plastics don't. The chemistry of plastic degradation limits how many times you can recycle it, and the economics of collection and sorting are broken. Nine percent versus eighty-five percent — that's not a gap, that's a different universe.
Where do we go from here? The interesting question to me is what happens as the grid decarbonizes and green hydrogen becomes viable for steelmaking. Does the calculus shift entirely toward metals?
I think it shifts heavily in that direction, but not entirely. If you can make steel with green hydrogen — the direct reduced iron process using hydrogen instead of natural gas or coal — you eliminate most of the CO2 from ironmaking. Combine that with an electric arc furnace running on renewable electricity, and steel's carbon footprint drops to near zero. Aluminum smelted with renewables is already relatively low-carbon. So yes, metals could become the clear environmental winner for most applications.
But?
But plastics have their own potential pathways. Bio-based plastics from non-food feedstocks, chemical recycling that breaks polymers back down to monomers — these could close the gap. The question is whether they'll scale in time, and whether the economics will ever work without a carbon price. Right now, virgin fossil-based plastic is cheap, and that's hard to compete with.
The material we choose is ultimately a bet on the future. On energy prices, on recycling infrastructure, on regulation.
There's no perfect answer today, but there are better questions to ask. Instead of "is steel better than plastic," ask: what's the expected lifespan? Where was it made? What energy powered the smelter? Can it be recycled at end of life, and will it actually be recycled? Those questions get you closer to a real answer than any material label ever will.
If you've got a weird prompt about materials, sustainability, or anything else that doesn't fit the mainstream conversation, send it to prompts at my weird prompts dot com. We read every one.
Thanks to our producer Hilbert Flumingtop for making this show happen.
This has been My Weird Prompts. I'm Corn.
I'm Herman Poppleberry. We'll be back soon.
