Researchers from the Massachusetts Institute of Technology (MIT), CarbonCure Technologies, a concrete carbon mineralisation company, and the Indian Institute of Technology (IIT) Jodhpur have identified a previously unseen sequence of chemical reactions that takes place when CO₂ is introduced during cement mixing.
The peer-reviewed study, published in the Journal of the American Ceramic Society, used in situ Raman microspectroscopy to observe cement paste as it hydrated over its first 24 hours. The researchers found that dissolved CO₂ temporarily changes the normal hydration pathway, creating a silica gel network before conventional hydration resumes.
The resulting structure was associated with higher early-age compressive strength in the laboratory samples.
CO₂ changes the early hydration process
The study identified three stages in the reaction.
During the first four hours after CO₂ is introduced, the cement enters what the researchers call the ‘mineralisation’ stage. Clinker phases dissolve more rapidly in the presence of dissolved CO₂, producing calcium carbonate polymorphs and a temporary amorphous silica gel network.
Between four and eight hours, the system moves into the ‘transition’ stage. Once the available CO₂ has been consumed, hydration resumes. Portlandite precipitates through the pore network and reacts with the silica gel, producing a distinct form of calcium-silicate-hydrate, or C-S-H.
After eight hours, the material enters the ‘stabilisation’ stage and conventional hydration resumes. Ordinary C-S-H continues to form as the cement hardens.
The study found that the CO₂-treated sample reached a 24-hour compressive strength of 7.39 MPa, compared with 6.11 MPa for the reference sample. This represents an increase of about 13% under the tested conditions. The experiments used API Spec 10A Class G cement, with CO₂ additions of up to 1% by weight of cement.
Carbonate particles are not the source of the strength gain
The observations also change the picture of how CO₂ contributes to early strength.
Calcium carbonate particles formed during mineralisation were previously considered possible nucleation sites for C-S-H formation. The Raman observations did not support that explanation. Instead, the researchers found that C-S-H formed at the interface between silica gel and portlandite, while the carbonate particles became embedded within the developing silica gel network.
Silica gel therefore appears to play a central role in forming the more evenly distributed binder seen in the CO₂-treated samples.
The study describes the gel as a temporary phase. It forms during the early mineralisation stage and is subsequently consumed as C-S-H develops. The researchers concluded that this sequence provides a new framework for understanding how CO₂ affects cement hydration.
Implications for lower-carbon concrete
The findings are relevant to concrete production because changing the hydration process could affect how cement mixes are designed. CarbonCure said its carbon mineralisation systems are deployed at hundreds of concrete plants in more than two dozen countries.
According to the company, its technology has been used across more than 20,000 mix designs involving 350 distinct cements, 200 supplementary cementitious materials and thousands of admixture combinations. CarbonCure also said its producer partners reduce cement content by about 4–6% on average while maintaining equivalent performance and meeting project specifications.
The company said its systems have been used in more than 11 million concrete loads, including applications in residential construction, high-rise developments and infrastructure. Those figures refer to CarbonCure’s wider commercial deployment, not the experiments reported in the new study.
The study focused on cement paste and laboratory mortar samples rather than full-scale building projects. Its contribution is to show, in real time, how CO₂ alters cement chemistry during the first hours of hydration and how that altered sequence affects the developing binder structure.
