The chemistry of enhanced weathering

Rock dissolving is not the same as CO2 being removed. What matters is which acid did the dissolving, which negative ions end up balancing the released calcium and magnesium, and whether that balance survives the long trip from soil to sea.

By the end of this module, you will be able to
  • Say where the acid that weathers rock really comes from, and why soil matters more than rain
  • Name the basalt minerals that dissolve, and what they release
  • Explain alkalinity from first principles: follow the electric charge, and you follow the carbon
  • Describe the two ways captured carbon is stored, and why one holds twice as much as the other
  • Name six places along the path from field to ocean where carbon can be delayed or lost
  • Explain how grain size and travel time set how long it all takes

In Weathering, erosion & soil you met the headline reaction:

CaSiO3 + 2CO2 + 3H2O → Ca²⁺ + 2HCO3⁻ + H4SiO4

It is true but hides the interesting part. It squeezes three separate steps into one line: CO2 becoming an acid, the acid attacking a mineral, and the products being carried away. Each step can go differently in a real field, and each one is a place where the CO2 accounting can go right or wrong. This module opens the equation up.

Where the acid comes from

CO2 dissolves in water, and a small part of it reacts to form carbonic acid (H2CO3), which splits into a hydrogen ion and a bicarbonate ion:

CO2 + H2O ⇌ H2CO3 ⇌ H⁺ + HCO3⁻

The double arrows mean the reactions run both ways and settle into a balance, an equilibrium. Left alone, very little of the CO2 ends up as bicarbonate, because the H⁺ and HCO3⁻ keep recombining. Here is the key insight: the bicarbonate is born in this first step, not when the rock dissolves. The rock's job is to use up the H⁺. Each H⁺ that a mineral consumes can no longer recombine, so the balance shifts right and more CO2 is pulled into the water to replace it. The mineral grain acts as a sink for acid, and that is what keeps CO2 flowing in.

Rain that has only met the atmosphere is weakly acidic, around pH 5.6. Soil water carries far more carbonic acid, because soil air holds 10–100 times more CO2 than the open air. That extra CO2 comes from respiration: roots and soil microbes "breathing out" as they burn sugars and decaying organic matter. Since plants made those sugars from atmospheric CO2 by photosynthesis, the carbon is still air carbon; the soil just concentrates it. So the real acid factory for weathering is not the sky but the soil.

root microbes pore air, rich in CO₂ CO₂ CO₂ CO₂ Basalt grain water film HCO₃⁻ HCO₃⁻ H⁺H⁺H⁺ Ca²⁺ Mg²⁺ silica to groundwater 1 2 3 4
  1. Roots and microbes breathe out CO₂, so the air in soil pores holds ~10–100× more CO2 than the sky.
  2. CO₂ dissolves in the water film: CO2 + H2O ⇌ H⁺ + HCO3⁻. The bicarbonate is born here, from air carbon.
  3. H⁺ attacks the grain (plagioclase, pyroxene, olivine), releasing Ca²⁺, Mg²⁺ and silica. Each H⁺ used up lets more CO2 dissolve.
  4. Water carries the ions away: the cations and their bicarbonate partners drain down towards groundwater.
Not to scale: real water films are far thinner than a grain. The rock supplies the calcium and magnesium; the carbon in the bicarbonate comes from CO2. The grain's job is to use up H⁺, which keeps more CO2 flowing into the water.

What basalt gives up

Basalt is a mafic rock (see Minerals & lithology), made mostly of three minerals plus, in some rocks, volcanic glass. The minerals crystallised hot, sit high in the Goldich series and are unstable at the surface; glass, with no crystal structure at all, weathers faster still. As acid eats into them they release base cations, the positively charged ions of calcium (Ca²⁺), magnesium (Mg²⁺), sodium (Na⁺) and potassium (K⁺), plus dissolved silica.

MineralRough recipeReleasesSpeed
Olivine(Mg,Fe)2SiO4Mg²⁺, iron, silicaFastest, but only a few per cent of most basalts
PyroxeneCa(Mg,Fe)Si2O6 (augite)Ca²⁺, Mg²⁺, iron, silicaModerate; a major mineral in basalt
Plagioclase feldsparCaAl2Si2O8 to NaAlSi3O8Ca²⁺, Na⁺, silica; aluminium stays behind in clayCalcium-rich kinds are faster; often about half of basalt
Volcanic glassNo fixed recipeA mix of everythingFast, where present, because it has no crystal structure

Olivine shows the chemistry most cleanly. Each unit of magnesium olivine (forsterite) can turn four CO2 molecules into bicarbonate:

Mg2SiO4 + 4CO2 + 4H2O → 2Mg²⁺ + 4HCO3⁻ + H4SiO4

Two things are worth noticing. First, iron does not help much. Iron released as Fe²⁺ quickly reacts with oxygen and falls out of solution as rusty iron oxide, and that reaction gives back the acid that dissolved it, so iron leaves little lasting effect. Calcium and magnesium (with a little sodium and potassium) are what count. Second, a typical basalt is roughly 10% CaO and 5–8% MgO by weight, which works out at roughly 3–4 kilomoles of calcium plus magnesium per tonne of rock. (A mole is a fixed count of atoms or molecules, about 6 × 10²³, which lets chemists compare numbers of particles rather than weights; a kilomole is a thousand moles. One kilomole of CO2 weighs 44 kg.) That calcium and magnesium content sets the ceiling on how much CO2 a rock could ever capture, a topic picked up in Choosing the rock.

Alkalinity: follow the charge

Here is the single most useful idea in this part of the course. Any volume of water must be electrically neutral: the total positive charge of its dissolved ions has to equal the total negative charge. Nature does not allow a bucket of water with a net charge.

So when a grain releases one Ca²⁺ ion, two units of negative charge must arrive with it, or two units of positive charge must leave. In practice, the negative partner is whatever anion came with the acid that did the dissolving. That gives a simple rule:

Every cation released must be balanced by a negative ion. If that partner is bicarbonate, CO2 has been captured. If it is nitrate, sulphate or chloride, from a strong acid, it has not.
Carbonic acid does the work · made from CO₂ in soil air Ca²⁺ = HCO₃⁻ HCO₃⁻ 2 CO₂ stored as bicarbonate both carbons came from the air +2 balanced by −1 and −1 Nitric acid does the work · from fertiliser nitrogen Ca²⁺ = NO₃⁻ NO₃⁻ No CO₂ stored yet the rock dissolved just the same +2 balanced by −1 and −1 Sulphuric acid does the work · from pyrite or acid rain Ca²⁺ = SO₄²⁻ No CO₂ stored one sulphate ion is enough +2 balanced by a single −2
The same calcium ion, three different partners. A soil sample would show the same amount of rock dissolved in every case. Only the water's chemistry reveals which story is true. Note the sulphate case: what has to balance is the charge, not the number of ions.

This is what alkalinity really measures. You met it as the water's capacity to neutralise acid. Thanks to charge balance, it can also be written as a simple subtraction:

alkalinity ≈ base-cation charge − strong-acid-anion charge

Here the base cations are Ca²⁺, Mg²⁺, Na⁺ and K⁺, and the strong-acid anions are nitrate (NO3⁻), sulphate (SO4²⁻) and chloride (Cl⁻). In words: take all the positive charge from base cations, subtract the part already balanced by strong-acid anions, and what is left must be balanced by bicarbonate (plus a little carbonate, CO3²⁻). In the near-neutral water of most soils and rivers, alkalinity is almost all bicarbonate, so the extra alkalinity above an untreated control tracks the CO2 captured, once any carbonate minerals dissolving in the soil or rock are accounted for (only about half the carbon they add comes from the air). It is also something a lab can measure, either by adding acid until the water's buffering is used up (a titration), or by measuring all the major ions and doing the subtraction.

Try it · which form is the dissolved carbon in?

Dissolved inorganic carbon comes in three forms: uncharged dissolved CO2, bicarbonate (HCO3⁻, one negative charge) and carbonate (CO3²⁻, two). Which one dominates depends on the pH. Drag the pH and watch the shares change.

100%50%0% Share of the dissolved carbon CO₂(aq) HCO₃⁻ CO₃²⁻ pH 4567891011 soil water ~4.5–7 rivers ~6.5–8.5 seawater ~8.1
Dissolved CO₂ (no charge)Bicarbonate HCO₃⁻Carbonate CO₃²⁻

Simplified. The curves use standard equilibrium constants at 25 °C (pK about 6.3 and 10.3 in fresh water; about 5.9 and 8.9 in seawater, where dissolved salts shift the balance). Real waters vary with temperature and salt content, so treat the shares as approximate.

Two ways to store the carbon

Once the rock has dissolved by carbonic acid, the captured carbon can end up in one of two places.

Pathway A · dissolved

Bicarbonate in water

The cations and bicarbonate stay dissolved and travel with the water, through soil and groundwater into rivers and finally the ocean. Each Ca²⁺ or Mg²⁺ balances two HCO3⁻, so up to two CO2 per cation.

In the ocean, bicarbonate stays for more than 10,000 years. One correction: seawater's carbonate chemistry re-balances as alkalinity is added, and some CO2 is given back to the air. Estimates vary, but storage ends up roughly 10–20% below the theoretical maximum.

Pathway B · solid

Carbonate mineral in soil

In drier soils, or where water evaporates, calcium and bicarbonate can recombine as solid calcium carbonate, called pedogenic carbonate ("made in the soil"):

Ca²⁺ + 2HCO3⁻ → CaCO3 + CO2 + H2O

One of the two carbons is locked in the solid; the other goes back to the air. So only one CO2 per Ca²⁺: half the efficiency, but stored in solid rock that can last for thousands of years or more, as long as it isn't later dissolved by acid.

A · Stays dissolved travels to the ocean Ca²⁺ + HCO₃⁻ HCO₃⁻ all dissolved in the sea 2 CO₂ stored per Ca²⁺ minus ~10–20% in seawater B · Precipitates forms solid carbonate in soil Ca²⁺ + HCO₃⁻ HCO₃⁻ CaCO₃ solid CO₂ to the air 1 CO₂ stored per Ca²⁺ the other goes back to the air
The same calcium ion and the same two captured carbons, two endings. Staying dissolved stores twice as much; precipitating stores half, but as solid rock. The same halving happens if carbonate precipitates later, in a river or the sea.

Follow the carbon

Between a grain dissolving and bicarbonate settling into the ocean, the water may travel for decades. Along the way, carbon can be delayed, re-routed or lost. The diagram follows one parcel of carbon from air to ocean, with the leak points numbered and, for each stage, what can actually be measured.

AirSoilPorewaterGroundwaterRiverOcean
Stage 1 of 6: Air
StageCarbon isCan leakMeasure
CO2 gas–too dilute to see one field's effect
in rock grains; cations on clay1 2 3soil cores: rock left, and exchangeable Ca
HCO3⁻ with Ca²⁺ and Mg²⁺4suction cups: alkalinity, pH, cations
HCO3⁻ in slow flow–wells, drains: alkalinity, flow, water age
HCO3⁻; some CO2 degasses5 6stream samples: alkalinity, discharge
HCO3⁻ for 10,000+ years6models only: far too dilute
Pick a stage, or press Follow the carbon, to move the parcel along its path. The circle above the porewater stage is a zoom on one grain in its water film. Measurement gets harder from left to right as the signal is diluted, which is why most ERW measurement focuses on the soil and the water just beneath it. The numbered leak points are explained in the cards below; the detail of how each stage is measured is in Measuring removal (MRV).
1 · Loss

Strong-acid weathering

Nitrogen fertiliser is converted by soil microbes to nitrate (nitrification), a process that releases H⁺: nitric acid, in effect. Oxidising pyrite (iron sulphide, FeS2) in rock, or acid rain, makes sulphuric acid. Rock dissolves, but the cations are balanced by nitrate or sulphate, so no CO2 is captured. Worse, strong acid meeting existing bicarbonate turns it back into CO2.

2 · Delay, or loss

Held on exchange sites

Clay and organic matter carry negative charge on their surfaces and hold cations like Ca²⁺ there. A soil's capacity to do this is its cation exchange capacity (CEC). Released cations may sit on these sites for years before being swapped off: a delay. But in acidic soils the sites are already loaded with H⁺ and aluminium (Al³⁺). Incoming Ca²⁺ pushes that acidity into the water, where it turns bicarbonate back into CO2, so exchange becomes a real loss. Many highly weathered tropical soils, including some in India, are acidic (diagram below). Others, such as many of the black soils on the Deccan basalts, are neutral to alkaline and hold solid calcium carbonate; there the bigger risk is that captured carbon re-precipitates as soil carbonate (point 6 and pathway B).

3 · Temporary

Taken up by plants

Crops absorb calcium, magnesium and potassium as nutrients. Most comes back to the soil when residues rot, but cations in a harvest leave the field. To keep charge balance, roots that take up cations release H⁺. That acts like added acid: it can speed dissolution, but it cancels alkalinity, so it works against the carbon capture.

4 · Loss (partial)

Locked in new clays

Some weathering products recombine into new secondary minerals, mainly clays. Cations built into them no longer balance bicarbonate. How much this matters varies with soil and climate, and is an active research question.

5 · Watch

Degassing in streams

Groundwater carries extra dissolved CO2 from the soil air. When it emerges into a stream, that excess bubbles out. This mostly releases CO2 that was never balanced by cations, so alkalinity is unchanged, but it raises the pH and can tip the water into precipitating carbonate (point 6). That is why accounting tracks alkalinity, not total dissolved carbon.

6 · Partial loss

Carbonate precipitation downstream

In rivers, lakes or the ocean, calcium and bicarbonate can precipitate as calcium carbonate. As in pathway B, that sends one of every two captured CO2 back to the air. Hard-water rivers, already rich in calcium, are most prone to it.

1 · Acidic soil: the clay already holds acid H⁺ and Al³⁺ sit on the negative sites −−−−−−−−−−−−−−− H⁺Al³⁺H⁺H⁺Al³⁺H⁺Al³⁺H⁺ Ca²⁺Ca²⁺ 2 · Ca²⁺ swaps on, pushing acid off: a loss −−−−−−−−−−−−−−− Ca²⁺Ca²⁺Ca²⁺H⁺Al³⁺Ca²⁺H⁺Al³⁺Ca²⁺ H⁺Al³⁺ H⁺ + HCO₃⁻ → CO₂ + H₂O CO₂ back to the air 3 · Later: Ca²⁺ moves on with bicarbonate a delay, once the sites are mostly Ca²⁺ −−−−−−−−−−−−−−− Ca²⁺Ca²⁺H⁺Ca²⁺Ca²⁺Ca²⁺H⁺Ca²⁺Ca²⁺ Ca²⁺ HCO₃⁻ HCO₃⁻ todrains
Cation exchange can be a loss or a delay. In acidic soils (step 2), the first calcium released by the basalt displaces acidity already held on the clay, much as farm lime does. That acidity turns bicarbonate back into CO2, so part of the rock's capacity is spent neutralising the soil instead of storing carbon. Once the sites hold mostly calcium (step 3), new calcium moves on with bicarbonate, but the carbon signal in drainage water can lag years behind the rock dissolving, and clay-rich soils with high CEC can hold back a lot.

How long it takes

Two clocks run one after the other. The first is dissolution: how long the rock takes to dissolve. For basalt powder in a farm field, that is years to decades, depending on the minerals, the grain size, the temperature, the rainfall and the soil's acidity (the controls you met in Weathering, erosion & soil). The second is transport: how long the dissolved products take to reach the river. Water moves through the root zone in months to years, groundwater can take years to decades to reach a stream, and rivers reach the sea in days to months.

Grain size matters most, because reaction only happens at surfaces. Halve the grain size and you double the surface area per tonne, so the rock dissolves roughly twice as fast. There is a neat consequence. If the surface of each grain retreats at a steady rate, a grain's diameter shrinks by the same amount each year, whatever its size. So the time to dissolve a grain completely is proportional to its starting size, and a little shrinking removes a lot of volume, because volume goes with the cube of the size.

Worked example: the shrinking grain

Assume each grain's surface retreats by 0.5 µm a year, so its diameter shrinks by 1 µm a year. (A number picked for easy arithmetic; real rates vary by a hundredfold or more between minerals and fields.)

A 10 µm grain disappears completely after 10 years. After 5 years its diameter has halved to 5 µm, so its volume is (½)³ = ⅛ of the start: about 88% has already dissolved.

A 100 µm grain takes 100 years to disappear. After 5 years it has shrunk to 95 µm, and (0.95)³ ≈ 0.86 of its volume remains: only about 14% dissolved.

So in a real powder, which always contains a range of sizes, the finest fraction does most of the early work. This is why the size distribution of the crushed rock matters so much, and why grinding finer is a trade-off against the energy and emissions it costs (see From quarry to field).

In real soils, rates also slow over time: surfaces can become coated with new clay and iron oxide, the most reactive minerals go first, and as the basalt raises the soil pH there is less acid left to drive the reaction. Removal therefore arrives as a gradually tailing curve over years to decades, not as a single event.

allhalfnone Share of the eventual removal, counted at each point 1 2 3 05101520 Years after spreading (schematic)
Rock dissolved in the fieldAlkalinity in drainage waterAlkalinity reaching the river
  1. Soil lag. Cation exchange and slow-moving soil water hold the signal back.
  2. Transport lag. Groundwater takes years to decades to reach a stream.
  3. The tail. Coated grains, the loss of the most reactive minerals and rising pH slow dissolution, so the last part arrives very slowly.
Two clocks, one after the other: the same removal shows up later the further downstream you measure it. The shapes are illustrative; real lags vary from months to decades with soil, climate and hydrology.
Try it · grain size and dissolution over time

One tonne of crushed rock contains a spread of grain sizes. Each grain's surface retreats at the same steady rate, so fine grains vanish quickly and coarse ones linger. Set the median size, the spread and the retreat rate, and see how much of the tonne has dissolved over 30 years.

100%75%50%25%0% Share of the tonne dissolved 10 µm grain 30 µm 100 µm 0% 051015202530 Years after spreading
Your powderSingle grains, for comparison

Faster in warm, wet, acidic soils and for reactive minerals; real rates are poorly known and vary a hundredfold.

Simplified. Grains are treated as spheres that shrink at a steady rate (the "shrinking grain" of the worked example above), with sizes spread in the bell-shaped-on-a-log-scale way typical of crushed rock. Real rates also slow over time, as described above, and only part of what dissolves becomes stored carbon. The default rate is not a measurement: it is tuned so that a typical powder dissolves roughly 20–30% in its first five years, one illustrative scenario. Real rates vary widely with mineralogy, soil and climate. The model in From quarry to field uses a different grain size and rate, tuned to the same illustrative 20–30%. Finer grinding costs energy and emissions; see From quarry to field.

Try it: the charge balance

The demo below does the accounting for one tonne of a simple basalt that releases about 3.5 kmol of calcium plus magnesium when it fully dissolves. If carbonic acid does all the work, each kilomole of Ca²⁺ or Mg²⁺ is balanced by 2 kmol of bicarbonate, so the maximum is about 3.5 × 2 × 44 ≈ 310 kg of CO2 per tonne of rock dissolved: about 0.31 t from calcium and magnesium, or about 0.35 t if the rock's sodium and potassium are counted too (this demo leaves them out). Change the share of weathering done by a strong acid, add strong acid beyond what the rock can neutralise, and choose whether carbonate forms in the soil or the bicarbonate reaches the ocean, to see what happens.

Try it · which acid dissolved the rock?

Illustrative numbers for one tonne of rock that has fully dissolved. Real fields dissolve only part of the rock, and every other loss in this module would reduce the figure further.

Positive charge released Ca²⁺ Mg²⁺ 7.04 kmol of charge Negative charge that balances it bicarbonate 7.04 nitrate 0.00 CO₂ stored per tonne dissolved 310 kg of a possible 310 kg −1000100200300 kg

Simplified. The rock releases 1.9 kmol of calcium and 1.6 kmol of magnesium per tonne. "Extra strong acid" is acid beyond what the rock neutralised: it turns bicarbonate already in the soil water (from earlier weathering or liming) back into CO2, one for one. Only calcium forms soil carbonate here, because magnesium carbonates rarely form in soils. The ocean option applies the ~15% seawater re-balancing to the carbon that stays dissolved.

Why this matters at Alt Earth

This is why measuring enhanced weathering properly means asking three questions, not one. Did the rock dissolve? Which acid dissolved it? And where did the released cations go: on to exchange sites, into plants and clays, or down into the water as bicarbonate? Charge balance turns those questions into things a lab can measure: cations, alkalinity and strong-acid anions such as nitrate and sulphate. On farmland that receives nitrogen fertiliser, the strong-acid share is not a theoretical worry, and an honest removal claim has to account for it.

The same question, which elements dissolve and travel and which stay put, is central to mineral exploration. Soil and stream-sediment surveys work because some metals move in water and others stay close to their source. Knowing an element's mobility tells an explorer how far a signal can have travelled from the deposit that made it.