Soda ash vs sodium bicarbonate for pH correction

The answer, up front

Soda ash — sodium carbonate, Na₂CO₃ — delivers 944 g of alkalinity as CaCO₃ per kilogram. Sodium bicarbonate, NaHCO₃, delivers 596 g. You therefore need about 1.59 kg of bicarbonate to do the work of 1 kg of soda ash, and bicarbonate costs more per kilogram to begin with. On cost per unit of neutralising capacity, soda ash wins and it is not close.

Buyers choose bicarbonate anyway, and the reason is that it is self-limiting. A bicarbonate solution equilibrates near pH 8.3 and stays close to it however much you add — dosing forty times as much moves it by three-tenths of a unit, downward. Soda ash has no such behaviour: pH climbs with concentration all the way to saturation, from 11.3 at 1% upward. Overdose soda ash into a lightly buffered stream and you can go straight through the 5.5–9.0 discharge window that Schedule VI of the Environment (Protection) Rules sets for effluent.

Two conditions attach to that, and they are not footnotes — they are the difference between a useful rule and a false promise. The bicarbonate ceiling holds only while the carbon dioxide stays in the water, and only while the material has not degraded to carbonate. Aerate the stream, cascade it over a weir or heat it, and the ceiling goes. Store the bags above 50 °C for a season and the ceiling goes. Both mechanisms are explained below, because a buyer who is sold an absolute ceiling and then discovers the conditions has been sold the wrong thing.

So the decision rule is short. Where overshoot is survivable — a large buffered stream, a well-instrumented controller with real residence time, a batch you can re-titrate — buy soda ash, in the dense grade, and save the money. Where overshoot is expensive, buy bicarbonate, keep it cool and turn it over, and understand that you have bought a reagent that is much harder to overshoot with, not one that makes overshoot impossible.

What a kilogram of each actually buys

Carbonate accepts two protons, bicarbonate one. That one fact generates the equivalent weights, and from them every alkalinity and dosing figure in this table. Alkalinity is conventionally reported as calcium carbonate equivalent, and the equivalent weight of CaCO₃ is 50.04 — so a kilogram of reagent delivers its number of equivalents multiplied by 50.04 grams of alkalinity.

Calculated from standard atomic weights on a 100% basis. Scale every dose by your own certificate of analysis — and read the next paragraph before you do, because the certificate figure is probably not on the basis you need.

The assay basis is where dosing calculations go wrong. The IS 251 requirements for soda ash — total alkalinity 98.5% minimum — are stated on a dry basis, and the standard separately permits up to 2% volatile matter at the time of manufacture and packing. A conforming consignment can therefore be 98.5% dry-basis and about 96.5% as received. That is a 2% dose error, the same order as the correction everyone remembers to make for assay. Dose on the as-received figure, not the dry-basis certificate number.

Two worked figures. To raise 1,000 m³ of water by 50 mg/L as CaCO₃ you need 53 kg of soda ash or 84 kg of bicarbonate on a 100% basis. To neutralise a tonne of 33% hydrochloric acid — 330 kg of HCl — you need about 0.48 t of soda ash or 0.76 t of bicarbonate, again on a 100% reagent basis.

One mechanism explains most of the gap between calculated and actual consumption. Soda ash only realises its second equivalent if the carbon dioxide leaves the water. Where the carbon stays dissolved as bicarbonate — a poorly stirred tank held at pH 7 or above, with no aeration and no vent path — the effective capacity falls towards one equivalent per mole, roughly doubling consumption against the textbook dose. If your usage runs well over calculation, look here before you suspect the assay.

A closed, unvented neutralisation vessel is a pressurisation and asphyxiation hazard, not a dosing inefficiency. Acid meeting a carbonate generates carbon dioxide in quantity; carbon dioxide is heavier than air and accumulates in pits, sumps and covered headspaces, which makes entry a confined-space job. Confirm the vent path is adequate before you investigate reagent consumption.

Why bicarbonate is self-limiting, and when it stops being

This is the property people buy, so it is worth stating precisely. The common formulation — “bicarbonate raises alkalinity without raising pH” — is not quite right, and the accurate version is more useful and more honest about its limits.

In a solution closed to the gas phase — no loss of carbon dioxide — the pH of a solution of a pure bicarbonate salt is approximately the mean of the two dissociation constants of carbonic acid: ½(6.35 + 10.33) = 8.34, independent of concentration. Measurement agrees. One producer’s published table for sodium bicarbonate at 20 °C:

A fortyfold increase in dose moves the pH three-tenths of a unit, and the wrong way. Within those conditions bicarbonate drives pH towards about 8.3 from either direction — it raises water below 8.3 and lowers water above it. That is a genuine and valuable property. Three things take it away.

1. Stripping the carbon dioxide. Two bicarbonate ions can give up a CO₂: 2 HCO₃⁻ → CO₃²⁻ + CO₂ + H₂O. Remove the gas — aeration, an aerated equalisation basin, a cascade or weir, air stripping, heating — and the balance shifts to carbonate and the pH rises. A bicarbonate-dosed stream that is subsequently aerated can exceed pH 9.0, and how far past depends on the dose: aeration with air can only strip down to equilibrium with atmospheric carbon dioxide, so the pH it settles at is set by the alkalinity you added. Heating, which drives the gas off far below that equilibrium, goes much higher. This is the same mechanism as the familiar complaint that pH drifts upward downstream of the neutralisation tank.
2. Carbonate in the material. Bicarbonate that has partly decomposed carries carbonate, which has no ceiling. This is what a solution-pH limit in the specification is there to control — and it is why the upper end of the ceiling is not a number we can give you in the abstract. It is whatever the solution-pH limit in the specification you buy against permits, which is one reason to specify it.
3. Heat, in storage or in the make-up tank. Decomposition to carbonate begins above 50 °C in the solid and at about 20 °C in solution. Degraded material behaves progressively more like soda ash.

None of that makes bicarbonate the wrong choice — it makes it a reagent with operating conditions. Keep the carbon dioxide in the water through the point where compliance is measured, specify and check the solution pH of the material, and store it cool and turn it over.

Soda ash behaves in the opposite way throughout. Carbonate hydrolyses, so pH climbs with concentration up to saturation: 11.3 for a 1% solution and 11.6 for a 5% solution, and the same drinking-water guideline that gives the 11.3 figure puts a 10% w/v liquid soda ash product at up to pH 12.5. Be careful with that last number rather than dismissive of it. Carbonate hydrolysis makes hydroxide concentration go as the square root of concentration, so a tenfold increase in strength should move pH by at most about half a unit — which does not get you from 11.3 to 12.5. The source does not explain the gap. The likeliest readings are that the two figures are not on the same basis, or that the 12.5 belongs to a more concentrated solution than its label suggests.

The 11.3 figure is for a 1% solution and the 11.6 figure for a 5% solution, both from producer safety data sheets; the 11.3 value is independently given in a national drinking-water guideline. Neither source states whether the concentration is w/w or w/v, which is a roughly 7% difference at these strengths — so treat them as indicative of the trend, and take the figure for the material you are buying from its own sheet. The 8.34 derivation uses standard textbook dissociation constants for carbonic acid at 25 °C.

The operating lesson is the same either way. Quote a concentration and a basis with every soda ash pH figure, and take the number for the material you are buying from its own sheet. A soda ash pH quoted with no concentration against it — and the word “saturated” is not a concentration — cannot be used for anything.

The sodium claim that does not survive the arithmetic

Bicarbonate is widely sold as the low-sodium option. It is 27.4% sodium against soda ash’s 43.4%, so on the face of it the claim is true. It does not survive the dose correction.

Each compound carries exactly one sodium per equivalent of alkalinity. So the sodium added per unit of alkalinity delivered is 22.99 ÷ 50.04 = 0.459 g of sodium per gram of alkalinity as CaCO₃, for both reagents, identically. You need 1.59 times as much bicarbonate and it carries 0.63 times the sodium; the two factors cancel, because they are the same factor inverted.

The practical consequence matters if you are discharging to land, blending for reuse, or watching sodium adsorption ratio for irrigation: switching from soda ash to bicarbonate to reduce sodium buys you nothing. What reduces sodium is a non-sodium alkali — lime or magnesium hydroxide — or dosing less of either by not overshooting. Bicarbonate’s advantage is that it reaches a given alkalinity at a much lower pH. That is real, and it is not a sodium advantage.

Which leads to the thing the self-limiting argument can make people forget. pH is not the only parameter on your consent. A bicarbonate overdose adds sodium, alkalinity and dissolved solids in direct proportion to dose even though the pH barely moves — so it is entirely possible to breach a total dissolved solids or sodium condition while the pH sits comfortably at 8.3. Those conditions come from your Consent to Operate or from an industry-specific entry — the dissolved-solids and percent-sodium rows were omitted from Schedule VI’s general standards in 1993, so do not go looking for them there. Dose to target, not to comfort.

In hard water, soda ash eats itself

This is the most expensive mistake on the page, because it damages equipment rather than merely wasting reagent. Carbonate and calcium precipitate calcium carbonate. That reaction is not a side effect — it is lime–soda softening, run deliberately to remove hardness. If removing hardness is your objective, soda ash is the correct reagent and this section is describing the process you want.

If it is not your objective, three things happen at once. The solution loses strength to precipitation, so you no longer know what you are dosing; scale forms in the make-up tank, the pump head and the injection quill; and the alkalinity you were trying to add leaves as a solid. An Australian national drinking-water guideline states this plainly in its sodium carbonate monograph and advises softening the service water.

• Make up soda ash solution with softened or RO water. Not raw borewell water, which across much of Gujarat and Rajasthan is hard enough to make this a daily problem rather than a theoretical one. Boiler condensate is often suggested and is sometimes suitable, but it can carry neutralising amines, iron oxides and oil carryover — so it is not an option for any potable, food or sensitive duty without checking what is in it.
• Where you need alkalinity and the hardness has to stay in the water — cooling water, potable blending, a stream going to biological treatment — bicarbonate avoids the bulk precipitation, because at pH 8.3 calcium carbonate supersaturation is far lower than at pH 11. Read the next paragraph before you treat that as a scale-free answer.
• Check the injection point, not just the tank. Scale forms where the concentrated reagent first meets the hard water, which is inside the quill.

The index usually quoted here is the Langelier Saturation Index, and it needs three corrections before it is safe to use.

1. Adding bicarbonate raises the index. The LSI has a term in titration alkalinity as well as terms in calcium and pH. Dosing bicarbonate raises the alkalinity term directly, so it pushes the index positive and increases calcium carbonate scaling tendency even though the pH hardly moves. In hard, high-cycle cooling water, adding bicarbonate alkalinity is a textbook cause of carbonate scale. It trades bulk precipitation for a scaling tendency to be managed with cycles of concentration and a scale inhibitor — it does not abolish the problem.
2. Know the validity range, and what replaces it. The LSI is standardised over roughly 10 to 10,000 mg/L total dissolved solids — the ASTM method is written for reverse-osmosis concentrate, so RO reject is the index’s intended application rather than an excluded one, provided it is under that limit. Above 10,000 mg/L the Stiff–Davis index is the correct tool. Check your TDS before quoting either.
3. Negative is not the safe side. An index below zero indicates undersaturation with respect to calcium carbonate and a tendency to dissolve the protective carbonate film, leaving metal and concrete exposed. The severity bands commonly quoted — −0.5, −1.0, −2.0 — are interpretive rather than standardised, so read them alongside a corrosion assessment for your actual materials. A pH correction that moves the index from −1 to +1 has traded corrosion for scale, not solved anything.

Dense or light soda ash is a handling decision, not a chemical one

IS 251:1998 classifies soda ash into dense, medium and light grades by bulk density, and states that all three have the same chemical composition — total alkalinity 98.5% minimum as Na₂CO₃ on a dry basis for each. Producer specifications commonly quote 99%. There is no strength difference to pay for, and a price premium justified on “dense is stronger” is being justified on something untrue.

The soda ash densities are the mid-points of the IS 251 ranges. Bicarbonate bulk density varies widely by grade and producer — the figure used here is one published grade, not a typical value — so take it from your own producer sheet before sizing anything. IS 251 defines a medium grade; we hold dense and light.

What you buy with dense grade is volumetric capacity and lower dusting. Neutralising a tonne of 100% sulphuric acid means storing and conveying about 0.98 m³ of dense soda ash against 1.73 m³ of light, on the densities above. Dense grade is roughly 1.8 times more compact per unit of neutralising capacity, which is why silos, pneumatic conveying and volumetric screw feeders are specified around it. Light grade is commonly said to dissolve faster; we have not found a producer figure for the dissolution rate of either grade, so test it in your own make-up tank rather than taking it on trust.

The safety gap is wider than most buyers assume

These two materials are often handled interchangeably by store staff. The international chemical safety cards do not treat them as equivalent. Sodium carbonate carries the GHS signal word Warning and the hazard statement ‘causes serious eye irritation’. Sodium bicarbonate carries no GHS hazard classification at all.

The wording difference on dust is the one to carry into a risk assessment: one card says harmful, the other says nuisance-causing. Eye protection and local exhaust are specified for soda ash handling, and nasal septum perforation on repeated exposure is on the card. Note the ingestion row too — soda ash is the worse of the two there as well. Neither material is combustible.

Both react with acids to produce heat and carbon dioxide — the sodium carbonate card uses the word “violently”. Never add soda ash or sodium bicarbonate, in any form, to concentrated acid. Adding a solid carbonate to concentrated acid froths violently and ejects hot acid, and a lid makes that worse rather than making it the only problem. The correct order is the reverse: make the reagent up as a dilute solution in a vented, stirred tank and feed the acid slowly into that.

Soda ash additionally has recorded explosion hazards with magnesium, phosphorus pentoxide, calcium hypochlorite and calcium, and a fire hazard with fluorine; producer sheets add aluminium powder and lithium. Bicarbonate’s card adds sodium-potassium alloy, some metals and ammonium compounds — worth knowing if your store holds ammonium salts. Keep both away from acid storage.

Bicarbonate releases twice the carbon dioxide

Per equivalent of acid neutralised, bicarbonate liberates 1.0 mol of CO₂ and carbonate 0.5 mol. The stoichiometry is forced, and it runs opposite to the intuition that the weaker base must be the gentler one:

• Na₂CO₃ + 2 HCl → 2 NaCl + H₂O + CO₂
• NaHCO₃ + HCl → NaCl + H₂O + CO₂

Neutralising a tonne of 100% sulphuric acid liberates up to about 249 m³ of carbon dioxide by the soda ash route and 499 m³ by the bicarbonate route, measured at 25 °C and one atmosphere, at full conversion. Those are mass-balance quantities. They tell you that ventilation, foaming and entrained carryover into the next stage are real design concerns for a covered tank, and that bicarbonate roughly doubles all three.

A total evolved volume is not a vent size. Vent and relief sizing is set by the peak volumetric generation rate at the worst credible addition rate, the allowable overpressure, and two-phase behaviour where foaming is credible — an API 2000 and, for foaming service, DIERS calculation by a competent engineer. Do not infer a vent from the figures above.

The words “up to” are doing real work. The gas only leaves to the extent the water is driven below about pH 7 and the gas is stripped. At pH 7 roughly 18% of the inorganic carbon is present as dissolved CO₂; at pH 8 about 2%. Neutralise to pH 7–8 without aeration and most of the carbon stays in the water — and comes out downstream, raising pH as it goes. That is the same mechanism that removes the bicarbonate ceiling, and it is why the reactor and the outfall can read differently.

The practical answer is to measure in both places. Add a process probe at the reactor, and leave the compliance probe where your consent puts it. The monitoring location for a consented parameter is fixed by your Consent to Operate and, for the highly polluting categories, by the Central Pollution Control Board’s online monitoring directions — so moving it is a consent variation, not a process decision.

Storage: bicarbonate starts to go off at 50 °C, not 270

The figure that circulates for sodium bicarbonate is 270 °C. That is complete decomposition. The international chemical safety card and a 2025 United States Department of Agriculture technical review both put the onset above 50 °C in the solid — and in aqueous solution at about 20 °C, complete on boiling. Soda ash by comparison is stable up to about 400 °C and only decomposes above that.

• A 50 °C shed roof space is not hypothetical in Ahmedabad in May. Bicarbonate wants covered, ventilated, cool storage and stock rotation rather than a consignment held through a summer. In a traded supply chain much of that time may be the supplier’s rather than yours — so ask any supplier, including us, for the date of manufacture or packing, how the lot has been stored, and a retest on anything held a long time. It is a fair question to put to whoever is holding the stock.
• Make bicarbonate solution up cold and use it. Do not hold it in a heated day tank, and do not heat the make-up tank to improve dissolution — you will convert it to carbonate and lose the self-limiting behaviour you paid for.
• Do not expect bicarbonate to stay bicarbonate in hot water or boiler service. It converts to carbonate and carbon dioxide, which is a condensate corrosion consideration rather than a dosing one.
• Soda ash make-up gains nothing above about 35 °C. Its solubility rises to a maximum near 35 °C at roughly a third by mass of solution and then falls as temperature rises further, because above that point the stable solid is the monohydrate. So a near-saturated tank that is heated past the maximum will deposit monohydrate; one cooled well below it will set hydrated crystals instead. Hold make-up below saturation at the coldest temperature the tank will ever see, and do not heat it.

On solubility itself, one caution worth more than a number. On a single consistent source — the international chemical safety cards — soda ash dissolves to about 30 g and bicarbonate to about 9.3 g per 100 mL of water at 20 °C, so soda ash is roughly three times the more soluble. But the published figures for soda ash disagree materially: other primary sources give about 21.6 g per 100 g of water, and a national drinking-water guideline gives 250 g/L. That is a spread of a third on the headline number. Do not do arithmetic with a solubility figure whose basis is unstated, and do not build a ratio out of two figures from different sources. Take the number for the material you are buying from its own producer sheet, with the basis named. Bicarbonate solubility also falls in the presence of sodium carbonate, so a contaminated or deliberately mixed solution will not reach the strength a single-salt figure suggests.

What the Indian standards and the discharge rules actually say

The limit you are dosing against

The general effluent discharge standard for pH is 5.5 to 9.0, and it is the same for all four discharge routes — inland surface water, public sewers, land for irrigation, and marine or coastal areas. It is one of the few parameters in Schedule VI of the Environment (Protection) Rules that does not vary by route, and online summaries routinely get this wrong by varying it.

Two qualifications. Where your industry has its own entry in the industry-specific standards, its parameters prevail; parameters that entry does not cover fall back to the general standards and to your consent. And your Consent to Operate may set a tighter figure than the rules do — the consent is the document you are actually judged against. For drinking water, IS 10500:2012 sets pH at 6.5 to 8.5 as the acceptable limit, and records no relaxation of it.

Soda ash

IS 251:1998, Soda Ash, Technical — fourth revision. On a dry basis: total alkalinity as Na₂CO₃ 98.5% minimum; matter insoluble in water 0.15% maximum; sulphate as Na₂SO₄ 0.08% maximum; chloride as NaCl 1.02% maximum; iron as Fe₂O₃ 0.007% maximum. Two of those carry process provisos that a purchase order can get wrong: the sulphate limit is 0.5% for material made by the modified Solvay process, and the chloride limit 0.4% for the sodium bichromate industry. Plus volatile matter 2% maximum at manufacture and packing, and three grades by bulk density with identical chemical requirements. Note what the standard does not contain: no heavy metal limit and no arsenic limit. If your application needs those controlled, they are yours to specify.

Sodium bicarbonate

The standard is IS 2124 — not IS 2251, which is frequently cited in error. Cite the edition. A third revision was published in 2023, superseding the 2000 edition, and we have not verified the 2023 grade schedule against the published text — so we are not going to tell you what grades it defines. That is exactly the position to take in a purchase order: name the current edition and the grade you mean, or name the producer specification you are buying against, and do not rely on a half-remembered grade name.

Bulk industrial bicarbonate is widely offered as “technical grade”, including by us. That is a trade descriptor, and the material behind it is sold against a producer specification. There is nothing wrong with buying it — it is what most water treatment runs on. There is something wrong with a purchase order whose only specification is the phrase itself, because it does not fix the assay basis, the solution pH or the heavy metal limits, and it will not help you in a dispute. Ask for the producer specification sheet and attach it.

Tariff and tax

Soda ash sits under HS heading 2836 20 and sodium bicarbonate under 2836 30. Both are inorganic chemicals of Chapter 28, and the applicable GST rate should be read off the current rate notification entry for heading 2836 — the schedules were re-notified with effect from 22 September 2025, so check rather than carry forward a remembered figure.

Two trade-remedy items matter to landed cost on imported soda ash, and both need checking rather than assuming. The Directorate General of Trade Remedies issued final findings on 29 September 2025 in an anti-dumping investigation into soda ash from Russia, Türkiye, the United States and Iran, recommending duty; as at 8 October 2026 we have not been able to trace the Ministry of Finance customs notification that would impose it. The Central Government normally has three months from publication of a final finding to act on it, so on that reading no duty was notified in time and none is in force — but confirm the current position with the Central Board of Indirect Taxes and Customs before you price an import rather than taking that from us. Separately, the minimum import price on soda ash was notified only up to 31 December 2025 and has therefore expired; we have not traced an extension, so do not assume one is running.

Regulatory references are current to our research on 8 October 2026 and are given so you can check them, not as a substitute for checking. Tariff, trade-remedy, standards and consent positions all change; verify against the current notification before committing a price or a specification.

Choosing, in one page

That last row is the one most often got wrong. IS 251 is titled Soda Ash, Technical; there is no food grade in it, and it sets no heavy metal or arsenic limit at all. A food, beverage or potable duty needs a material bought against a food specification with those limits written in, not a technical grade that happens to assay well. And a supplier carrying on a food business — which includes a trader supplying a food additive — requires an FSSAI licence for that activity under the Food Safety and Standards Act. That is the seller’s statutory duty, not your diligence exercise. Tell us the application is a food one and we will confirm in writing whether we can supply against it, and you should require the same of any supplier.

The honest summary is that soda ash is the cheaper reagent and bicarbonate the more forgiving one, and that the premium on bicarbonate is best understood as reducing the chance of a consequence — a breached consent, a ruined batch, a dead biological stage, a corroded line — rather than as a technical upgrade or a guarantee against one. Price both per kilogram of alkalinity delivered, decide what an overshoot would cost you, and the answer usually becomes obvious.