Concrete does reabsorb carbon dioxide from the air. The question has always been how much, and a UCLA-led team now puts the answer well below what the cement industry’s own decarbonisation arithmetic has been leaning on.
Using thermodynamic and diffusion-based modelling, researchers led by Gaurav Sant found that ambient carbonation offsets less than 10 percent of the cement industry’s annual CO2 emissions. Published estimates have run as high as 57 percent. The paper appeared in Communications Sustainability, a Nature Portfolio journal, with a publication date of July 25, 2026; UCLA announced the findings on August 26.
The 1,000-year number
The headline figure from the modelling is blunt. A typical beam, slab or pavement fully exposed to the atmosphere takes roughly 1,000 years to reach even 50 percent carbonation under normal outdoor conditions. Diffusion through a dense cementitious matrix is slow, and it gets slower as the carbonated front thickens. The team projects that concrete in service worldwide will passively absorb about 230 million tonnes of CO2 a year by 2030, against roughly 3 billion tonnes of cement-industry emissions in the same year. Cement production is already about 10 percent of global CO2, and global output is projected to approach 4.83 billion tonnes a year by 2030.
The end-of-life assumption is where it breaks
The sharper finding for contractors is about demolition, not chemistry. Optimistic carbonation estimates depend on crushed concrete continuing to absorb CO2 after the structure comes down, on the reasoning that crushing exposes vastly more surface area. The researchers point out that demolished concrete is typically landfilled, stockpiled or buried as road base, all conditions that restrict contact with air. That implicates construction and demolition waste practice directly, not just the mix design.
Ambient carbonation shows up as a line item in whole-building life-cycle assessments, in environmental product declarations and in industry roadmaps. It is the credit that lets a mix claim a lower net number without changing the mix. If the real offset is under 10 percent, that credit gets much thinner, and specifiers who have been accepting it should ask what number their consultant used.
What this doesn’t say
Sant’s own framing is calibrated: “Ambient carbonation cannot be relied upon as a meaningful tool for reducing atmospheric carbon dioxide accumulations.” That is a claim about scale, not about whether the effect exists. Every number here is a model output from one study under stated assumptions, and none of it is field measurement. The “as high as 57 percent” figure is UCLA’s characterisation of prior literature.
The practical read is that carbon reduction on a concrete job has to come from the binder, the mix and the structural quantity, because it isn’t going to arrive from the slab quietly breathing. Cincinnati’s Western Hills Viaduct replacement is a useful scale check: the city cut the main span from eight lanes to six partly to control cost, and less structure is also less cement, which is the only lever in that list that works on the timescale of a project. The release is at UCLA Newsroom; the paper is DOI 10.1038/s44458-026-00116-9.