Enhanced rock weathering

Over geological time, atmospheric carbon dioxide is regulated by the weathering of silicate rock. Rain dissolves CO₂ to carbonic acid; the acid attacks calcium- and magnesium-bearing silicate minerals; the released cations travel with bicarbonate to the ocean. Enhanced rock weathering is the attempt to run that thermostat deliberately, by crushing reactive silicate rock — usually basalt — and spreading it where water and biological CO₂ are abundant, which in practice means farmland.

The reaction and its stoichiometry

For a calcium silicate the idealised reaction is

CaSiO3 + 2 CO2 + H2O  ->  Ca(2+) + 2 HCO3(-) + SiO2

Two moles of CO₂ are consumed per mole of divalent cation released. Pure olivine (Mg₂SiO₄) is the stoichiometric best case at roughly 1.25 tonnes of CO₂ per tonne of mineral. Basalt is a mixture, much of it unreactive glass and feldspar, and its practical potential is a few tenths of a tonne per tonne of rock — an order of magnitude that matters, because it sets the haulage and spreading tonnage for every tonne of removal claimed.

Rate is surface area, and surface area costs energy

Dissolution is a surface process, so rate scales with specific surface area, which scales roughly as the inverse of particle diameter. Grinding finer therefore speeds the reaction — and grinding energy rises steeply as particle size falls. There is consequently an optimum, not a maximum: past a certain fineness the CO₂ emitted to mill, haul and spread the rock eats the CO₂ the rock removes. Weathering rate also depends strongly on soil pH, temperature, rainfall and biological activity, so the same rock applied in a warm wet tropical soil and a cool temperate one are not the same intervention.

Two ways the carbon comes back

The first is carbonate precipitation. If the released Ca²⁺ meets bicarbonate and precipitates as CaCO₃ in the soil or river system, half the captured CO₂ is released again — the removal is halved, not lost, but the accounting must follow it. The second is cation retention: Ca²⁺ and Mg²⁺ held on soil exchange sites have not yet completed the journey that makes the removal durable, so measuring dissolution is not the same as measuring sequestration.

Why verification is the hard part

The reaction cannot be observed directly at field scale. Operators infer it from soil porewater alkalinity and cation chemistry, from immobile trace-element ratios that track how much rock has dissolved, and from models calibrated on those signals. The measured signal is small against large natural variability in agricultural soils.

This is where the field is genuinely unsettled, and the page says so plainly: field trials have reported removal well below model prediction, and some have not resolved a signal at all in early seasons. The disagreement is about method, not motive — how much of the released cation load actually reaches durable storage, over what timescale, and what an honest uncertainty band around a credit looks like. Anyone reading a per-hectare removal figure should ask whether it was measured or modelled.

A separate, non-carbon caution: basalts vary in nickel and chromium content, so trace-metal loading of soil is a real screening question, independent of whether the carbon accounting works.