Derek Muller makes an argument that's been strangely absent from most science content: the substance that built our world isn't steel or aluminum — it's cement, and we're drowning in it. His piece challenges what we think we know about concrete by revealing that every person on Earth uses 500 kg of cement annually, enough to make two cubic meters of concrete. But what's surprising is how little we understand about this fundamental material.
Muller opens with an immersive demonstration: "I am about to get buried in concrete and while that's happening I'm going to explain everything you need to know about this substance." The video hook — literally climbing into a concrete-filled fishbowl — makes something most people consider mundane suddenly feel viscerally dangerous. That's good science communication.
Cement vs. Concrete
The piece immediately corrects the most common confusion: "Cement is like the glue... concrete is cement plus aggregate so plus gravel and sand." This distinction matters because Muller argues cement is "the most important man-made substance on the planet" — we use more of it than any other substance apart from water.
The statistic he cites is staggering: every year, 500 kg of cement are created for every man, woman, and child on Earth. From this, Muller pivots to a broader observation about scale. "Every year we make a certain weight of goods out of copper, more out of aluminum, and then you have glass, asphalt, lime, iron is a big one for all the steel... but by far the solid product we make the most of is cementitious material essentially cement — we use as much of it as we do all other materials combined."
This framing effectively uses comparative economics to show just how dominant cement production is. The argument lands because it quantifies something abstract — human reliance on a single building material — in ways that reshape how we think about infrastructure.
Primitive Cement and Roman Innovation
Muller traces the history of cement-making with genuine narrative momentum, starting with primitive methods: "To make primitive cement the key ingredient is limestone which is basically calcium carbonate if you heat it up to around 1,000° that drives off CO2 out of the rock leaving calcium oxide."
The Romans discovered the solution to primitive cement's limitations by adding volcanic ash called pozzalana. Muller writes, "They used it to create the largest unreinforced concrete dome in the world — the Pantheon — which has stood for 2,000 years and they built concrete piers into the sea which hardened underwater."
This is where the piece gets genuinely interesting: Roman concrete could harden underwater because there's no CO2 available to penetrate thick slabs. The Romans solved this problem centuries before modern chemistry understood why.
Concrete is liquid rock — you can pour it into any shape you want, it's strong and durable and inexpensive and it is so easy to produce that people have been making a version of it for thousands of years.
Muller then reveals the key discovery: "Nearly 2,000 years later people discovered that adding clay or Shale to the Limestone before it was crushed and heated produced the same effect." The reason? Silica. "All of these materials contain silica and silica totally changes the chemistry."
Modern Testing and Compression Results
The piece pivots to concrete testing in a compression room: "We're required to maintain the concrete samples in 100% humidity so we've chose to submerge them in water in a lime bath." Muller then walks through standard industry practice — every batch gets tested at 7, 14, and 28 days.
The demonstration of actual hydraulic press testing is where Muller's visual storytelling pays off. The cylinders break under pressure around 10,000 PSI, with samples showing remarkable strength retention even as they approach failure thresholds. He notes: "This is going to be really strong — very strong."
But the most counterintuitive finding comes from comparing pure cement versus concrete mix: "You might think that since pure cement has the most glue it would be the strongest." The testing revealed something unexpected — all cylinders broke under about the same pressure.
Muller observes: "It fractured a lot as the load was applied... what 8,000 PSI now it failed now cement plus sand 9,163 PSI."
The practical implication is significant: if you can reduce cement content to 30% in the mix and still get the same strength characteristics, that's a major cost savings. Muller notes "the cement is the most expensive part of concrete so if you can get away with reducing it down to 30%... well then you should definitely do that."
Self-Healing Concrete
One of the piece's strongest sections reveals Roman concrete had a self-healing property. Muller explains: "There were some surprising advantages that Roman concrete had — for example, it was actually less well mixed than modern concrete so there were little blocks of undissolved calcium oxide or quick lime... when the concrete cracked and water got in there it would dissolve that calcium oxide forming calcium hydroxide and then you would get the new growth of calcium carbonate."
This is genuinely fascinating: Roman concrete was self-healing because unreacted lime remained inside cracks, absorbing water and precipitating new calcium carbonate to fill fractures. Modern concrete lacks this property entirely.
Critics might note that Muller frames Roman concrete as superior in some ways but acknowledges survivorship bias — we only see structures that survived 2,000 years, ignoring failures from the same period. The answer to whether Roman concrete was truly superior? "Well the answer is no in short." Muller adds: "By and large you know when we look back at the Roman structures we only see the ones that have survived to this day so there is a Survivor bias."
The Consistency Question
The final section addresses practical testing: "They have to make sure it's the right consistency for the customer not too dry and not too runny." Modern concrete uses super plasticizers — "It makes the concrete easier to work and spread around without really changing the water content very much."
The slump test measures concrete spread on a board, targeting 27 inches. This represents standard field testing that ensures structural integrity before placement.
Bottom Line
Muller's strongest argument is reframing cement as civilization's most important material — not just because of volume but because it literally shaped how we build. His vulnerability lies in the comparison between Roman and modern concrete being somewhat apples-to-oranges: ancient structures survived precisely because they're ancient, while modern buildings are designed for shorter lifespans.
The piece succeeds when it shows that chemistry discovered 2,000 years ago — adding volcanic ash or clay to limestone — still governs how we make today's skyscrapers. The Romans didn't know why their underwater concrete worked; now we do. That's the satisfying payoff: understanding what ancient builders accomplished through trial and error.