Quick version
Ascent, a tower in Milwaukee, was completed in July 2022 at 86.6 metres tall. Its builders used more than 80,000 cubic feet of glulam, which is timber glued up into thick beams, and 336,000 square feet of cross-laminated timber, or CLT. A CLT panel is made from layers of boards glued together, each layer turned crosswise to the one below it, so the finished slab is stiff in both directions and can work as a floor or a wall.
The Deep Harvest, Chapter 7: The Built Frontier
Ascent, a tower in Milwaukee, was completed in July 2022 at 86.6 metres tall. Its builders used more than 80,000 cubic feet of glulam, which is timber glued up into thick beams, and 336,000 square feet of cross-laminated timber, or CLT. A CLT panel is made from layers of boards glued together, each layer turned crosswise to the one below it, so the finished slab is stiff in both directions and can work as a floor or a wall. The wood in those panels holds carbon that the trees pulled out of the air while they grew, and it keeps holding it while it stays in the building.
A developer has a plainer reason to look at wood than carbon: the schedule. A study of a 12-story condominium in the Greater Toronto Area priced the same building three ways. The timber frame came to $3.01 million, steel to $3.11 million and concrete to $3.12 million. The timber schedule ran 113 days, against 157 for steel and 190 for concrete. The price gap between the three frames is about 3 to 4 percent. The gap in time is 77 days between timber and concrete.
What the wood stores
Embodied carbon is the CO2 released to make a building's materials and put them together, before anyone moves in. When the carbon stored in the wood is counted, the research behind this chapter finds CLT buildings come out 81 to 94 percent lower in climate impact than concrete buildings and 76 to 91 percent lower than steel ones. In one case study, Adohi Hall, the mass timber stored about 2,757 tonnes of CO2e. CO2e, or carbon dioxide equivalent, rolls the other greenhouse gases into a single CO2 figure. An equivalent steel-and-concrete building came out at about 3,026 tonnes of net emissions.
Under the EU Emissions Trading System, where heavy industry pays for each tonne of CO2 it releases, carbon sold for about 85 euros a tonne in 2025. At that price, the research puts the carbon cost of an eight-story concrete frame at 102,000 to 127,500 euros, and the cost for a CLT frame at about a quarter of that.
The panels have to come from factories. North America had 150,000 cubic metres of CLT production capacity in 2023, up 58 percent from 2020. Europe has more than 500,000 cubic metres a year from more than 40 plants in Austria and Germany.
Concrete that keeps the CO2
Three companies are working on concrete itself. Cement is made by heating limestone in a kiln. As the limestone breaks down it gives off CO2, a step called calcination, which accounts for 56 to 90 percent of cement's process emissions. The lumps that come out of the kiln are called clinker, and they are ground into Portland cement, the ordinary grey binder in most concrete.
Solidia changes both the clinker and the curing. Curing is the stage where fresh concrete hardens, normally with water and over weeks. Solidia hardens its concrete with CO2 instead, in hours, using no water. The company says making its clinker releases 570 kilograms of CO2 per tonne instead of 816, that curing then locks up 250 to 300 kilograms of CO2 per tonne, and that the total cut against Portland cement is about 70 percent.
CarbonCure works on ordinary ready-mix, the concrete batched at a plant and driven to the site in a mixer truck. It injects CO2 into the mix at about 0.15 percent of the cement's weight. That lets the producer use about 5 percent less cement and saves 12 to 13 kilograms of CO2 for every cubic metre poured.
Blue Planet goes after the aggregate, the stone and sand mixed through concrete. It makes synthetic carbonate rock from alkaline industrial waste, and each tonne of that aggregate holds 440 kilograms of mineralized CO2, meaning CO2 turned into solid stone.
Most of these figures come from the companies that sell the products, and none of the sources in the research behind this essay has yet been independently verified. The chapter's research summary projects that these materials together could displace 8 to 15 percent of cement by 2035, and links that to 400 to 600 million tonnes of CO2e. That projection rests partly on industry sources. It is a different kind of number from the project-level figures for Ascent, Adohi Hall and Solidia above, and the research does not reconcile the two.
Walls from hemp, boards from fungus
Hempcrete is the woody core of the hemp stalk mixed with a lime binder and packed into walls. The hemp stores 325 kilograms of CO2 per tonne of dried plant. After the lime and the trucking are counted, a 350 mm wall still comes out at about minus 35.5 kilograms of CO2 per square metre, so the wall holds more carbon than went into making it. A U-value measures how fast heat leaks through a wall, and lower is better; that wall's is about 0.17 W/m2K, and the research puts operational fuel savings at 40 to 50 percent. It scored zero for flame spread and smoke in the ASTM E84 fire test, and one-hour fire ratings under ASTM E119 are heading into the 2027 cycle of the International Residential Code. The IRC, the model rulebook for houses in the United States, added hempcrete as Appendix BL in 2024, and Austin, Texas and Minnesota are adopting it for their 2026 codes. The research puts construction cost at 4,800 to 6,500 pounds sterling per square metre and calls that competitive. What it does not have is service data: nobody in these sources can show how a hempcrete wall performs after 50 years.
Mycelium is the mass of fine root-like threads a fungus grows. Fed on wood waste or crop waste in a mould, it knits the loose material into a solid board. The United States produces 70 million tonnes of wood waste and 3 million tonnes of soy byproduct a year that could feed it. Ecovative grows a mycelium material it calls AirMycelium, and says it became margin-positive in 2025, cut unit costs by 65 percent in a year and raised gross sales 50 percent from the third quarter to the fourth quarter of 2025. No independent check of those numbers appears in the research.
The size of the mycelium business depends on which report you read. One describes 450 million square metres of mushroom-farm capacity already built worldwide that the material could unlock. The other gives Ecovative about three million square feet of annual capacity. The first is the world's existing farm space; the second is one company's output. The research gives both figures and does not connect them.
Mycelium has two hard limits. Compressive strength, the squeezing load a material takes before it crushes, sits around 0.1 to 0.3 MPa for the commercial boards, against 15 to 30 MPa for concrete. That makes mycelium infill between the parts that carry the building, never the frame. It also soaks up water: between 40 and 580 percent of its own weight, the main problem the research names. As with hempcrete, there is no 50-year record of it in a wall.
Robots on the site
The research behind the robots starts from the housing gap. Between 1.6 and 3 billion people live in inadequate housing. In 2022, 1.12 billion lived in informal settlements, up 130 million since 2015, and the world needs about 35 million new homes a year.
FBR's Hadrian is a bricklaying robot that lays 285 to 360 blocks an hour. A human bricklayer lays 40 to 60. 3D printers that squeeze out concrete layer by layer to form walls are the other machine, and the research credits printed homes with cutting costs by 20 to 45 percent against conventional building and shrinking build time from months to weeks. ICON has printed twelve two-story homes at Mueller in Austin, Texas. SQ4D got a permit for a 1,900-square-foot printed home in Calverton, New York. COBOD's BOD2 printer built a three-story building for the Saudi developer Dar Al Arkan for under 10,000 euros in supplies, using 99 percent local concrete from Cemex's D.fab mix. Two standards, UL 3401 from 2020 and the ICC's AC509, give builders a path to code-compliant 3D-printed construction.
The same research covers geopolymer, a cement substitute made by activating minerals such as industrial ash with alkaline chemicals instead of firing limestone. Its summary puts geopolymers at 26 to 45 percent lower CO2 than ordinary cement. Its detailed section gives a separate per-kilogram figure on a different footing, and the two are not reconciled, so this essay uses only the summary range.
Codes and first cost
The research names what holds these materials back, and it is not how they perform. The main obstacles are gaps in knowledge, building codes that differ from one place to the next, and first-cost bias, where a buyer takes the lowest upfront price even when another option costs less over the life of the building.
Part of the gap is paperwork. An Environmental Product Declaration, or EPD, is a standard sheet listing a product's embodied carbon. 83 percent of architects and engineers say EPDs influence what they specify, but 89 percent say manufacturers don't publish enough of them and 85 percent point to the lack of national rules. Sweden allows timber buildings up to 18 stories, and Oregon and Washington give timber first preference in public purchasing. The chapter also names skilled finishing crews as a constraint, but the research does not count how many of them are missing.
Waste-fed materials also depend on a waste supply that keeps coming. Blue Planet's aggregate needs alkaline industrial waste. At the Kalundborg industrial park, 17 companies have traded more than 20 waste streams among themselves for about 54 years, displacing 635,000 tonnes of CO2 a year.
A builder who cures concrete with CO2, buys aggregate made from industrial waste and frames in CLT is buying CO2, waste and timber from plants and industrial zones. That turns the building trade from a buyer of steel and cement into a steady customer for those plants and the power they run on. The CLT panels, CO2-cured pours and hempcrete walls already have test results behind them; how many plants get built to supply them depends on code adoption, EPD sheets, finishing crews and the upfront price on the bid.
Read more in The Deep Harvest, Chapter 7: The Built Frontier.
Comments
Reader notes
Comments aren't live yet. Send notes to @slop_dealer.