Cement production is one of the largest industrial sources of carbon dioxide, and researchers have long sought ways to make concrete itself part of the solution. A new study reports that calcium carbide residue, a high-calcium industrial by-product, can be blended into cement, cured with concentrated CO2 to lock away the greenhouse gas as solid calcium carbonate, and then finished with ordinary water curing that rebuilds the very binding gel that carbonation normally destroys. The work, combining laboratory spectroscopy with a full-cycle numerical model, tracks how the chemistry of these materials shifts through each stage of an unconventional curing sequence.

The central challenge the researchers tackled is a well-known paradox in carbonation curing. When CO2 diffuses into cement paste, it reacts with portlandite—calcium hydroxide produced during hydration—and with the calcium silicate hydrate (C-S-H) gel that gives concrete its strength. The reaction precipitates calcium carbonate, which fills pores and boosts compressive strength while permanently storing CO2. But the same reaction strips calcium out of the C-S-H gel, polymerising its silicate chains into a brittle, silica-rich skeleton and lowering the matrix alkalinity. The result is a material that sequesters carbon effectively yet suffers carbonation shrinkage, increased brittleness, and degraded flexural performance.

The team’s answer was a sequential three-stage protocol: a pre-curing period at moderate temperature and humidity, followed by accelerated carbonation in 20 percent CO2—concentrations representative of CO2-rich flue gas rather than natural air—and finally standard water curing at high humidity. Pre-curing proved critical. Without it, pores remained nearly water-saturated, blocking gas-phase CO2 transport. Four days of pre-curing tuned the internal moisture to a level that both opened diffusion channels and supplied enough water for CO2 to dissolve into carbonic acid. Quantitative X-ray diffraction showed that this simple adjustment raised the total calcium carbonate yield by 145.3 percent compared with specimens carbonated without pre-curing.

The spectroscopic evidence for what happened next is central to the study. Infrared analysis of the silicon–oxygen stretching band revealed that carbonation drove the polymerisation index of the silicate network from 0.14 in standard-cured paste up to 0.74, confirming severe decalcification of the C-S-H gel. Yet after the subsequent standard curing stage, the index fell back and the proportions of low-polymerisation Q1 and Q2 units recovered. The researchers attribute this to a net C-S-H gel regeneration mechanism: the highly polymerised, hydrated silica gel left behind by carbonation reacts in situ with residual calcium hydroxide—whether from unhydrated clinker or unreacted carbide residue particles—to reform dense C-S-H gel with a moderate calcium-to-silicon ratio of roughly 1.2 to 1.5.

This regeneration reaction resembles the classical pozzolanic reaction between silica fume and lime, but the authors argue it is chemically distinct and faster. Silica fume consists of dense, anhydrous glassy microspheres whose strong surface silicon–oxygen bonds must be broken before long-range ion migration can proceed. Carbonated silica gel, by contrast, is a highly hydrated, defective, porous skeleton rich in polar silanol groups, giving it far higher specific surface area and reactivity. Because it forms within the existing gel network, the secondary hydration can proceed in nanoscale micro-regions without long-range diffusion, replenishing the matrix with a flexible binding phase that offsets the rigid embrittlement imposed by the calcium carbonate skeleton.

To move beyond volume-averaged laboratory measurements, the team built a continuum-scale reactive-transport model covering the entire pre-curing, carbonation, and standard curing sequence. The model couples Fickian diffusion of CO2—modulated by evolving porosity, pore saturation, and an Arrhenius temperature factor—with hydration kinetics for the clinker minerals and first-order carbonation kinetics for portlandite, C-S-H, and the clinker phases themselves. A key feature is a hydration kinetic correction factor applied during standard curing: because carbonation strips away the hydrate layers encasing residual clinker, unreacted grains regain contact with water and hydrate faster. Calibrated against measured clinker consumption, the factor settled at 10, with a sensitivity scan from 1 to 20 showing that values below 5 badly underestimated re-hydration while values of 15 or above prematurely exhausted the clinker and inflated the predicted alkalinity recovery.

Model validation was deliberately separated from calibration. Predicted carbonation depth came in at 9.2 millimetres against an experimentally measured 10.3 millimetres, a relative error of 10.68 percent. For the major phases, the model reproduced the XRD-measured concentrations of calcium carbonate, portlandite, and C-S-H with relative errors of 0.91, 1.36, and 1.33 percent respectively—variables that played no role in calibrating the kinetic factor. The simulation also resolved the partially carbonated zone, the pH 9 to 11 band between fully carbonated and uncarbonated regions that the conventional phenolphthalein test, with its single colour-change threshold at pH 9, cannot visualise. This zone matters because steel reinforcement corrodes fastest precisely where pH sits between 9 and 11.5.

The simulated maps revealed a spatially heterogeneous chemistry during standard curing. Surface and core regions were dominated by clinker re-hydration, which restored portlandite concentrations and alkalinity once the external CO2 supply was cut off. The partially carbonated zone, however, hosted a three-way competition: residual carbonation driven by CO2 trapped in the pores, regeneration of C-S-H from polymerised silica gel, and continued clinker hydration. The peak increment of calcium carbonate during standard curing appeared not at the surface, where reactive calcium had been exhausted, but about 10 millimetres deep inside the partially carbonated band—the only region still holding both residual CO2 and unconsumed portlandite.

Sensitivity analyses across temperature, humidity, water-to-binder ratio, and carbide residue content exposed a fundamental trade-off. Conditions that maximise CO2 uptake—high temperature, low humidity, high water-to-binder ratio, high waste replacement—accelerate diffusion and deplete the active calcium reservoir so thoroughly that alkalinity cannot recover, dropping portlandite below the thermodynamic threshold of about 769 moles per cubic metre at which C-S-H remains stable. Under those conditions the regeneration rate collapses. Conversely, dense, low-porosity mixes preserve alkalinity but starve the carbonation reaction. The preferred window identified by the model sits at 20 to 25 degrees Celsius, 70 to 75 percent relative humidity, a water-to-binder ratio of 0.5, and a carbide residue replacement rate of 15 to 25 percent, where sequestration stays above the benchmark of roughly 6000 moles of calcium carbonate per cubic metre while portlandite remains high enough to sustain gel regeneration.

The authors are candid about the model’s limits: it treats the paste as a homogeneous continuum, omits explicit moisture conservation and ion-scale chemistry, was validated mainly at a single mix proportion, and approximates the three-dimensional cube with a mid-height two-dimensional section, which slightly underestimates total sequestration and corner carbonation. Even so, the study delivers an end-to-end quantitative picture of how waste-based cement can both bury carbon and recover part of its binding phase afterwards. By showing that the damage carbonation inflicts on C-S-H gel is at least partially reversible through a managed curing sequence, the work points to a route toward construction materials that recycle industrial waste, sequester flue-gas CO2 permanently, and still retain the flexible binding phase that concrete needs to hold together.

Subject of Research: Coupled hydration–carbonation phase evolution and C-S-H gel regeneration in calcium carbide residue–cement composites under multi-stage curing

Article Title: Spatiotemporal phase evolution and net C-S-H gel regeneration in calcium carbide residue–cement-based materials under coupled hydration-carbonation

Article References: Guan, H., Zhao, Y., Shi, J., Bai, J., & Li, L. (2026). Spatiotemporal phase evolution and net C-S-H gel regeneration in calcium carbide residue–cement-based materials under coupled hydration-carbonation. Case Studies in Construction Materials, 25, Article e06615. https://doi.org/10.1016/j.cscm.2026.e06615

Image Credits: AI Generated

DOI: 10.1016/j.cscm.2026.e06615

Keywords: carbonation curing, calcium carbide residue, C-S-H gel, carbon sequestration, cement chemistry, calcium carbonate, CO2 utilisation, reactive-transport modelling, partially carbonated zone, industrial waste recycling, low-carbon concrete, X-ray diffraction