Insights into the Role of H2O in the Carbonation of CaO Nanoparticle with CO2

Evidence and attribution¶

Authority of statements

Prose below summarizes the JPCC article identified by doi, title, and pdf_path.

Summary¶

C/H/O/Ca (and Al where applicable) ReaxFF RxMD simulates CO\(_2\) and H\(_2\)O co-adsorption and reaction on nanoparticulate CaO, distinguishing a fast surface/kinetic carbonation stage from a slow diffusion-limited stage in calcium looping. Results are cross-checked with TGA on sol–gel CaO (~31 nm crystallite size, ~11.85 m\(^2\)/g surface area). Steam is found to strongly enhance the diffusion-controlled stage with little change to the initial kinetic stage, traced to faster ion/gas transport, thicker carbonate product layers, and proton/OH-mediated disruption of the CaO interior. The motivation connects to CO\(_2\) capture cycles where carbonate product layers can choke gas access: if steam reshapes transport without accelerating the initial surface reaction, operating strategies can target humidity to extend utilization of CaO sorbent particles.

Methods¶

Reactive MD (ReaxFF in LAMMPS)¶

Six-processor parallel jobs use the ReaxFF implementation in LAMMPS to follow CO\(_2\)/H\(_2\)O adsorption and reaction on spherical CaO nanoparticles (~1.8 nm radius) centered in an 88 × 88 × 88 Å box with Packmol-generated gas placement. A representative setup contains 3591–2691 atoms (as printed in the article) with 300 CO\(_2\) molecules and H\(_2\)O:CO\(_2\) = 1:1; in-plane PBC apply to the full supercell. Simulations use the canonical (NVT) ensemble with a Berendsen thermostat (100 fs coupling), 0.25 fs timestep, 4 ns total trajectory length, and output every 1000 steps; additional runs scan 400–1000 K to probe temperature effects on carbonation.

Analysis and experiment¶

Reactive C/H/O/Ca/Al ReaxFF (parameterization cited in the article) drives Ca–O–C chemistry; carbonation degree vs radius, pair correlations, and diffusive flux diagnostics follow §3 of the JPCC paper. TGA uses ~5 mg sol–gel CaO after calcination (900 °C, N\(_2\), 10 min) and 650 °C carbonation with 15 vol% CO\(_2\) (60 min), with steam levels per Table 1 to separate kinetic vs diffusion-limited stages.

Barostat / pressure control during adsorption MD: N/A — NVT runs (no NPT barostat in the quoted gas-on-nanoparticle protocol).
Replica / metadynamics: N/A — not used.
External homogeneous electric field: N/A — not used (electrochemical driving forces enter through chemistry and composition, not an applied E-field term in the excerpted protocol).

Findings¶

Stage-specific steam effect: MD + TGA consistently show H\(_2\)O accelerates late-stage (diffusion-limited) carbonation while leaving the early kinetic stage largely unchanged.
Microscopic mechanism: OH from water dissociation promotes CO\(_3^{2-}\) formation and CaCO\(_3\) layer growth; protons migrate inward, hydroxylate the oxide, and open diffusion pathways—together improving CO\(_2\) penetration and conversion relative to dry conditions.
Practical reading: because steam mainly speeds the slow regime, process designers may see the largest gains when particles are already partially carbonated and diffusion limits uptake—matching the calcium-looping context emphasized in the introduction.
Corpus honesty: Quantitative TGA time constants and MD radial profiles should be taken from papers/Wang_CO2_H2O_CaO_NP_JPCC_2018.pdf (and SI references therein), not inferred from this summary alone.

Limitations¶

Nanoparticle models may omit full reactor-scale transport; ReaxFF uncertainty on multicomponent oxide carbonation kinetics remains. Sintering, particle packing, and interparticle diffusion in fixed beds are not captured in single-particle MD, so pilot reactor data remain necessary when scaling TGA-informed kinetics to process models. Particle polydispersity and intrapore diffusion in CaO pellets add further scale-up uncertainty.

reaxff-family