Hydrogen peroxide 50 percent: stabilisers, decomposition, venting, and why the drum has a breather cap

The answer, up front

1. Fifty percent of what? The Indian Standard defines the strength by mass. One litre of 50 percent by mass carries about 600 grams of hydrogen peroxide; one litre of a solution quoted as 50 percent on a weight-in-volume basis carries 500 grams. That is one sixth less chemical for the same word. Ask which basis before anything else.
2. The grade number is the strength, and it runs backwards. IS 2080 : 2021 has six grades, Grade 1 the strongest at a nominal 70 percent and Grade 6 the weakest at 3 percent. 50 percent is Grade 3. A quotation for “Grade 1” is a quotation for 70 percent material, not for a better 50 percent.
3. Stabilised is a performance test, not a declared ingredient. The standard’s title says stabilized hydrogen peroxide and the standard never names a stabiliser, never limits one and never requires one to be declared. What it requires is that the material pass an oxygen-evolution test. So a purchase order cannot specify the stabiliser, and a supplier cannot be held to one that was never written down.
4. The acidity row is a window, not a ceiling — for Grade 3 it has a minimum as well as a maximum, which is not how buyers expect a specification to read. A buyer who writes “free acid: maximum” into a purchase order has written a specification that under-acidified, less stable material passes.
5. The specified minimum strength applies on the date of packing. Not on delivery, not on opening. The standard requires the month and year of packing on the container, which is what lets a buyer audit it — and which is the single most useful thing to look at on an arriving drum.
6. The vent is not a defect. Hydrogen peroxide gives off oxygen continuously, so a sealed container pressurises. The standard sets ullage requirements that make a vented closure the only practical option on a 50 percent drum, and producers put the rule in four words: never confine hydrogen peroxide. Fitting a plain cap, or taping over the vent, is the single most dangerous thing anyone can do to a drum of this product.
7. And hydrogen peroxide has been under a Quality Control Order for years, not months. Mandatory certification took effect on 13 March 2022, after two deferrals of an order first notified in June 2020, and the Order now in force replaced that one in late November 2022. The goods are required to conform to IS 2080 : 2021 and to bear the Standard Mark — a prohibition drafted in the Bureau of Indian Standards Act against any person who manufactures, imports, distributes, sells, stores or exhibits for sale.

This is a buyer’s guide to bulk hydrogen peroxide, not legal advice, not a specification and not a design document. Velora Chemicals is a bulk supplier and trader sourcing from qualified manufacturers, not a manufacturer. Nothing here states or implies that Velora holds any licence, registration or consent. Hydrogen peroxide is an oxidiser, and nothing in this article is a substitute for the producer’s own safe-handling literature, the current code of safety, or a competent engineer. Every standard and instrument named below should be read in its current form before any of its numbers goes into a purchase order or a plant design.

The standard, the edition, and the grade that is a strength

The current Indian Standard is IS 2080 : 2021, Stabilized Hydrogen Peroxide — Specification (Second Revision), which supersedes IS 2080 : 1980 — an edition the Bureau of Indian Standards records as withdrawn. The standard was first published in 1962 and first revised in 1980. It carries Amendment No. 1 of August 2023, established on 2 August 2023 by a Bureau notification published in the Gazette of India on 8 August 2023, with the unamended standard remaining in force until 1 February 2024.

That matters more than usual here, because the 1980 edition is still the one most widely cited — including, as published, by the Government’s own list of Quality Control Orders, which names “IS 2080 : 1980 (Reaffirmed 2016)” against the hydrogen peroxide entries while the Order in force mandates the 2021 edition. A certificate or technical data sheet citing IS 2080 without a year, or citing the 1980 edition, is not evidence of conformity to the standard that is actually mandated.

Clause 3 of the 2021 edition sets out six grades, and in this standard the grade is the nominal strength. The numbering descends from strongest to weakest, which is the opposite of what most buyers assume a grade number means.

So 50 percent solution is Grade 3. A quotation offering “Grade 1 hydrogen peroxide” is offering 70 percent material, with a different transport classification, different packing rules and a different stability test temperature. And note the test temperature in the right-hand column: a stability figure of 0.3 millilitres per minute means nothing without the temperature it was measured at, because the standard uses 75 degrees Celsius for Grade 1 and 100 degrees Celsius for Grades 2, 3 and 4.

Clause 4.1 describes the material as a clear, colourless liquid free from suspended impurities and dust. Clause 4.2 is the one worth reading twice: the strength determined by the standard’s titration method, or by the alternative densitometer method, shall be not less than the nominal figure on the date of packing. The obligation is pegged to packing, not to delivery and not to the day the drum is opened.

What these figures are, and what we have not read. The clause numbers, grade table and requirements above are from the text of IS 2080 : 2021 and its Amendment No. 1 of August 2023; on the copy available to us, that amendment touches the packing clause and deletes a two-litre capacity limit for hermetically sealed containers of Grades 1 and 2. Two provenance warnings go with that. First, the copy we read was a scanned text conversion in which some table headings were garbled, so every figure in this article taken from the standard should be confirmed against the official Bureau text before any of it goes into a contract or a purchase order. Second, the Bureau’s own standards-portal entry for IS 2080 : 2021 records no amendment at all, against a gazetted notification that plainly establishes one — so even the amendment status of this standard is not cleanly readable from the Bureau’s public record, and the gazette is the instrument to go by. Where a standard has been made mandatory by a Quality Control Order, as this one has, the Bureau operates a view-only facility for reading it free of charge, so that is a cheap check rather than an expensive one. Separately, we have not read IS 14200 : 2025, Hydrogen Peroxide — Code of Safety (First Revision), which supersedes IS 14200 : 1994 — that is the Indian document for handling and storage, and it is a different standard from this one, written by a different committee.

Fifty percent of what: the three numbers that are all called fifty

This is the costliest ambiguity in the product, and unlike most basis traps it is not a question of rounding. Hydrogen peroxide strength is quoted three ways in commerce, and they are not close together.

The conversion to a volume basis is simply the density: at 1.20 grams per millilitre, 50 percent by mass is about 60 grams per 100 millilitres. So the two bases are about a fifth apart, and the direction is the one that costs a buyer money. Put it per litre, which is how a tank is filled: a litre of 50 percent by mass contains about 600 grams of hydrogen peroxide, while a litre of something described as 50 percent on a weight-in-volume basis contains 500 grams. That is one sixth less chemical, for the same two digits on the quotation.

The volumes figure is the old bleaching trade’s unit and it still appears. One producer publishes 50 percent by mass as 198 volumes, meaning 198 litres of oxygen obtainable from a litre of solution at zero degrees Celsius and one atmosphere. That is confirmable from the chemistry: a litre of 50 percent at 1.20 grams per millilitre holds 600 grams of hydrogen peroxide, which is 17.6 moles; two molecules give one of oxygen, so 8.8 moles of oxygen, which at 22.414 litres per mole is 197.7 litres. The producer’s published figure and the arithmetic agree to within a fifth of one percent, which is a useful thing to know, because it means a volumes figure can be converted rather than guessed at.

Do not confuse volume strength with volume expansion. The same producer publishes a second ratio for 50 percent solution, the volume expansion ratio of 1300, which is the ratio of the resulting gas volume — steam as well as oxygen — to the volume of liquid decomposed. It is not a concentration, it must never be used in an oxygen-yield calculation, and it is not a relief-sizing basis either: it is a ratio for complete decomposition, with no rate, no heat-release and no vessel term in it. The producer that publishes it states no use for it and refers sizing questions back to itself. Quoting 1300 as though it were a strength, and treating it as the input to a relief calculation, are both errors — and the second is the more expensive one.

One more unit, for completeness, because it is the one in which a tender is sometimes written: on a 100 percent basis. A tonne of 50 percent solution is half a tonne of hydrogen peroxide. That sounds obvious and it is routinely lost: a contract priced per kilogram of 50 percent solution and a contract priced per kilogram of active hydrogen peroxide are different contracts by a factor of two, and the second is the only one in which two different strengths can be compared at all.

The specification table, and the row with a floor

Table 1 of IS 2080 : 2021 is short — seven characteristics — and it is shorter than most buyers expect, for a reason given in the standard’s own foreword. Here it is for Grade 3, with Grade 4 alongside for contrast.

Two figures in that table we cannot vouch for, and one of them is the one buyers copy. The withdrawn 1980 edition set the acidity band for 50 percent material at 0.015 to 0.030 grams per 100 millilitres, the same band it set for 35 percent, where our scanned copy of the 2021 table reads 0.05 as the upper figure for Grade 3 — a value that also happens to sit in the adjacent row, which is how a garbled scan goes wrong. One Indian producer publishes its own 50 percent acidity as 0.015 to 0.030. So treat the lower figure and the existence of a floor as sound, and the upper figure as unconfirmed, and read it off the Bureau text before it goes into an order. The same caution applies to the units on the iron and copper rows, discussed below.

Read the acidity row first, because it is the only row in the standard that behaves differently from how buyers expect a specification to behave. For Grade 3 the limit is a band in grams per 100 millilitres as sulphuric acid, with a lower figure of 0.015: there is a minimum. The next section sets out what the producers do and do not say about acidity and stability; what matters here is that a buyer who copies the row across into a purchase order as free acidity with a maximum only has written a specification that under-acidified material passes — material outside the standard’s band, on the side where one producer puts the optimum.

Note the mixed basis inside a single table. The strength is percent by mass; the acidity and the two residue rows are grams per 100 millilitres, which is a volume basis. Converting a grams-per-100-millilitre figure into a mass percentage without dividing by the density inflates it by about a fifth. The metals rows are the ones to check hardest: our copy of the 2021 table gives iron and copper in parts per million, while the withdrawn 1980 edition expressed both on the grams-per-100-millilitre basis, and one Indian producer still publishes its iron limit that way. A limit copied across without its basis is not a limit.

What is not in the table is as important as what is. There is no chloride, sulphate, phosphate, nitrate or aluminium limit in IS 2080 : 2021; the foreword records that those impurity requirements, which the 1962 edition carried, were deleted at the 1980 revision, when residue on ignition, lead and arsenic were added instead. The 2021 revision added determination of strength by densitometer and of heavy metals by an instrumental method. So any table of chloride and sulphate limits attributed to IS 2080 is wrong, and where a supplier does publish such limits they are that supplier’s own tighter specification, not a requirement of the standard. If a process cannot tolerate chloride, that limit has to be a line in the purchase order, because the standard will not supply it.

Two further details in that table repay a second look. The lead limit is 10 parts per million for every grade, which is loose enough that a buyer coming from a food or electronic specification will assume a misprint; it is not one. And arsenic is expressed as the oxide, As2O3, not as arsenic, so re-basing it to elemental arsenic lowers the figure by about a quarter. Neither is a reason to be alarmed, and both are reasons to read the units column.

What stabilised means, and what it does not promise

The standard is titled Stabilized Hydrogen Peroxide. It never names a stabiliser. The word appears in the title and in the scope and nowhere else as a requirement: there is no stabiliser identity requirement, no concentration limit, and no obligation to declare one. In the Indian Standard, stabilised is defined entirely by passing clause 4.3 — for Grade 3, evolving not more than 0.3 millilitres of oxygen per minute per 25 millilitres at 100 degrees Celsius, in a test whose vessel is deliberately conditioned with nitric acid first. It is a performance definition, and the standard then passes the rest to the supplier: clause 5.1.6 requires the containers to be stored and handled as the manufacturer directs, in order to ensure stability.

What a stabiliser actually does is published by producers rather than by the standard. One producer’s safe-handling bulletin puts it as commercial grades containing stabilisers which chelate small amounts of impurities, providing protection against the effects of minor levels of contamination — and then adds the sentence that matters most: stabilisers are ineffective in dealing with gross contamination by decomposition catalysts. That is the sentence that disposes of the argument that stabilised material is forgiving of a dirty tank. It is not. It is forgiving of traces.

We found no primary source naming the stabiliser chemistry in any Indian 50 percent product, and no producer publishing a declared stabiliser content or range. Stannate, phosphate, nitrate and organic chelant systems are all discussed in the wider literature; we are not going to attribute any of them to a product on the strength of that. The honest buying position is that the stabiliser package is proprietary, is not specifiable through IS 2080, and is the single largest undocumented difference between two conforming materials. If a process has been qualified on one producer’s 50 percent, a change of producer is a change of stabiliser package, and that is a thing to test rather than to assume.

Decomposition: the arithmetic of a product that is always leaving

Hydrogen peroxide decomposes to water and oxygen and gives out heat doing it. The reaction is the whole of the product’s commercial and safety character, so it is worth having the numbers.

• The reaction. Two molecules of hydrogen peroxide give two of water and one of oxygen. One producer publishes the heat as 1240 British thermal units per pound of 100 percent hydrogen peroxide, and separately, in its property table for 50 percent solution, as 603 British thermal units per pound of solution and 22.8 kilocalories per mole, both at 25 degrees Celsius.
• The gas. The same producer’s volume strength figure of 198 means a litre of 50 percent solution can release about 198 litres of oxygen at zero degrees Celsius and one atmosphere. On a mass basis, which the producer does not publish and which is our arithmetic from the same stoichiometry, a kilogram of 50 percent solution yields about 165 litres of oxygen at zero degrees Celsius and about 180 litres at 25 degrees.
• The rate, in normal service. Two producers publish less than one percent of relative strength lost per year for commercial grades under normal or ambient conditions; a third publishes approximately one to two percent a year. So the figure is of that order rather than a single agreed number. Note what that is and is not — it is a producer statement about its own stabiliser package in normal storage, not a requirement of IS 2080, whose specified minimum applies on the date of packing.
• The temperature rule of thumb, and the published figures do not quite agree. Two producers state that the decomposition rate is roughly doubled by every 10 degrees Celsius increase; one producer’s page on stability and decomposition puts the rise at about 2.2 times for each 10 degrees Celsius. They are the same rule to different precision, and we attribute rather than average them. Either way, a drum in an unshaded Indian yard is not in the same storage condition as a drum in a shed, and the published annual loss figure assumes the shed.

What accelerates it is a short list, and almost all of it is avoidable. One producer names the homogeneous catalysts as alkalis, strong acids and salts of transition metals — nickel, chromium, copper, iron — and the surface catalysts as copper, brass, zinc and mild steel, with the warning that contamination can cause rapid decomposition at levels as low as 0.1 part per million; another names metals, metal salts, dust, rust and wood generally. Manganese belongs to a third list: the United States occupational pocket guide’s incompatibilities, which run iron, copper, brass, bronze, chromium, zinc, lead, silver and manganese. Beyond metals: rough vessel surfaces, dust, rust, wood, organic matter, ultraviolet light, and the enzyme catalase.

And then there is pH, which is where the acidity row in the specification table comes back. The asymmetry is sharp and it is the opposite way round from most chemical intuition. One producer states that stability is best near neutral, that the deterioration at lower pH is not normally large, but that at higher pH stability deteriorates very rapidly, and alkaline hydrogen peroxide may be very unstable. The other puts an optimum below pH 4.5 and states that above pH 5 the decomposition increases sharply, which is why commercial solutions are adjusted to below pH 5. So the two published positions agree on alkali and differ on acid: one has stability best near neutral with acid merely tolerated, the other has it best below pH 4.5. What is not in dispute is the direction of the danger. Alkali destabilises the material, sharply, on both accounts. Which makes the minimum in the acidity row look a good deal less like a drafting curiosity. The standard states the band and gives no reason for it, and we are not going to put words into the committee’s mouth; nor will we say more than the producers do, which on acid is less than it is often reported to be. Either way it is a reason not to specify the acidity away, and the buying point stands without any claim about why the clause is drafted as it is.

The practical shape of all of the above, for a buyer rather than a chemist: a caustic spill into a peroxide bund, a copper fitting in a transfer line, a galvanised bucket, a dirty hose shared with another product, or sweeping a spill into sawdust are not housekeeping lapses on this product. They are the documented mechanisms by which it decomposes. And decomposition in a confined space is a pressure event, not a yield loss.

The breather cap, and the ullage rule behind it

This is the part of the product that surprises people who have handled every other drummed chemical, so it is worth stating the principle before the detail. One producer’s five basic rules for storing hydrogen peroxide begin with four words: never confine hydrogen peroxide, followed by never contaminate it, never contact it, and always store it cool, shaded, clean and separate. The same producer puts it again more bluntly: never confine hydrogen peroxide in drums, tanks or pipes, because it always releases gas at a slow rate, which if confined results in overpressure.

IS 2080 : 2021 builds that into the packing clause as an ullage requirement, and the arithmetic is what forces the vent. Clause 5.1.2.1 permits containers with or without a valve or vent type closure, and then sets the empty space accordingly:

• With a valve or vent type closure: minimum ullage 10 percent.
• Without one, for Grades 2, 3 and 4: ullage not less than one half of the total capacity of the container.
• For Grades 5 and 6 without a vent, not less than one quarter.

So an unvented 200 litre drum of 50 percent could lawfully carry 100 litres. That is the standard making the commercial decision for the buyer: a vented closure is the only economic way to pack this product, and the vent is therefore not an optional refinement on an otherwise ordinary drum. The same clause records what is actually done in trade — Grades 2, 3 and 4 normally supplied in dedicated tankers, or packed in barrels and carboys wherein proper vents and seal rings are provided on the caps.

The producer rules for the drum itself are specific, and each of them exists because somebody once did the opposite:

• Each drum has a continuous vent. Do not block this vent.
• Store in the original vented container, upright, in a cool, ventilated area. That sentence comes from a safety standard published as the North American producers’ minimum recommendation, not from one company’s preference.
• Do not stack drums more than two high.
• Never return unused hydrogen peroxide to the storage container, and never return spilled material to the vessel. Material that has been out of the container has met something, and what it met is a catalyst until proven otherwise.
• Rolling a drum may cause leakage through the vent, which is a reason to handle upright rather than evidence of a faulty drum.
• Do not use wooden pallets, do not reuse the drums for other chemicals, and segregate the storage from other chemicals.

Nobody publishes a sentence reading do not fit a plain cap, so we will not put one in quotation marks. It follows from the rules that are published: the drum is required to be vented, the vent is required to be unobstructed, and the material is required never to be confined. Replacing a vented closure with a solid one, or taping over the vent to stop a smell, converts a drum into a pressure vessel with no relief.

For a tank rather than a drum, the same producers’ standard requires a continuous vent and a properly sized emergency relief valve on all hydrogen peroxide tanks, with the tank operated at atmospheric pressure and the design arranged to avoid entrapment, so that oxygen evolution cannot build up pressure. No published sizing basis was available to us — no required vent area, no design decomposition rate, no stated basis of any kind — and that standard refers the question back to the producer. We leave it there. Relief sizing for a decomposing oxidiser turns on a design decomposition rate and heat release against the vessel and the vent path, and we are not going to nominate an input for that calculation from a buying note: it belongs to an engineer with the producer’s own data in hand.

One threshold that does not bite at 50 percent, and should not be imported. That same producers’ standard requires temperature monitoring systems on tanks containing greater than 60 percent hydrogen peroxide. A 50 percent tank is outside that requirement on that source. We note it because the opposite error — presenting a 70 percent rule as a 50 percent rule — is the commonest way this product gets written about badly, and it runs through the safety literature in both directions.

Materials, and why the plastic drum is not a general permission

Hydrogen peroxide service is a short list of permitted materials and a longer list of prohibited ones, and several of the prohibitions are counterintuitive. From one producer’s published compatibility guidance:

• Recommended metals: aluminium alloys 1060, 1260, 5254, 5652 and 6063, at 99.5 percent minimum purity, and stainless steel types 304, 304L, 316 and 316L.
• Other acceptable materials: chemical glass, chemical ceramic, PTFE, and polyethylene — high density, cross-linked, unpigmented and ultraviolet stabilised — with PVC for temporary systems only. Polyethylene and PVC carry a footnote restricting them to 50 percent concentration or lower, which is exactly why the drum for this product is high density polyethylene and why that does not generalise to stronger grades.
• Unacceptable: brass, copper, nickel, iron and mild steel, bronze, synthetic rubbers, polypropylene and zinc.

Two of those deserve to be said out loud. Polypropylene is on the prohibited list while polyethylene is permitted, which defeats the common assumption that one industrial plastic can stand in for another. And aluminium is permitted only as those named high-purity alloys: a statement that aluminium is compatible with hydrogen peroxide, without the alloy numbers and the purity, is not a usable statement. Nobody we read names galvanised steel as such — the honest form of that warning is that zinc is prohibited by the producer and listed as incompatible by the occupational pocket guide, so a galvanised surface is excluded through its coating.

A second producer’s list is shorter and does not agree on the one limit that matters: it gives 304L or 316L stainless for tanks and piping, with pure aluminium, certain aluminium alloys, polyethylene and PVC permitted for solutions up to 60 percent by weight. Fifty percent and 60 percent are not the same permission, and a third producer’s manual puts the polyethylene ceiling at 50 percent as well. Take the lower figure: polyethylene and PVC are materials for this concentration and below, and nothing in this article should be read as clearing them for Grade 2 at 60 percent. And IS 2080 itself speaks to this, which is unusual for a product specification: clause 5.1.2 requires the container material of construction to be compatible with the properties of hydrogen peroxide, requires glass to be neutral glass and carboys to be virgin high density or high molecular high density polyethylene, and requires bulk tankers for Grades 1 to 4 to be SS 316L or SS 304L with a mirror finish surface, a minimum of 320 grit.

That grit specification is the standard quietly agreeing with the decomposition section. A rough surface is a catalytic surface. Which leads to the requirement that is most often skipped on a new installation:

Passivation is mandatory, and both producers say so in terms. The producers’ safety standard: the storage tanks must be properly passivated before putting into hydrogen peroxide service. The other producer: the proper preparation of the tank — cleaning, pickling and passivation — prior to first use is mandatory to ensure product stability. The payoff is published too: a properly passivated system can serve indefinitely, provided no internal surface damage or contamination has occurred. New, bright, unpassivated 316L is not yet a hydrogen peroxide material, and this is work for trained people with the producer’s procedure, not a rinse.

It is an oxidiser, not a fuel, and that changes the fire response

Hydrogen peroxide solutions are not flammable — that is a producer’s own words. What they are is a strong oxidising chemical that tends to react rapidly, sometimes violently, and that can ignite combustible material. The hazard at 50 percent is created mostly by what the material meets, not by the material as supplied.

• The detonability question has two conditions and quoting one of them is misleading. One producer states that vapours can explode where the hydrogen peroxide concentration in the vapour phase exceeds 26 mol percent, and that at normal pressure an explosive vapour composition requires liquid at 74 percent by weight or more and a liquid temperature above 100 degrees Celsius. Both conditions fail for 50 percent material at ambient temperature, and both have to be stated together: the concentration alone does not make it safe and the temperature alone does not make it dangerous.
• Mixtures are the real exposure. The same producer reports a general risk of detonation where hydrogen peroxide is 25 percent by weight or more of a mixture with organics; the other publishes an operating rule, never allow the hydrogen peroxide content to exceed 20 percent by weight in organic solutions. The United States emergency response data for the 20 to 60 percent band adds that solvents such as acetone, ethanol and glycerol will detonate on mixture with peroxide of over 30 percent concentration — a band that includes 50 percent.
• Concentrated peroxide may decompose violently in contact with iron, copper or chromium, on that same emergency response data, and containers may explode when heated.
• Evaporative concentration is the sleeper hazard. One producer notes that even at low concentrations ignition can occur under unfavourable conditions after a gradual concentration. A 50 percent spill left to dry does not become safer as the water goes; it becomes stronger.

Which is why the response to everything is water. The producers’ safety standard says to use water to respond to all incidents involving hydrogen peroxide — fires, spills and leaks alike; one producer says for firefighting use only water, and do not use other extinguishing agents, keeping containers cool by spraying, with bunker gear and self-contained breathing apparatus. Small spills are diluted with plenty of water to below one percent; large spills are approached from upwind and dammed with sand or earth, and combustible materials must not be used to absorb them. The reason is structural rather than a matter of preference: the material carries its own oxygen, so smothering agents such as foam, dry chemical or carbon dioxide cannot starve the reaction, while water both cools and moves the material down the concentration bands. Sawdust, rags or paper on a peroxide spill is the classic fatal error, and it is the same mechanism as returning spilled material to the drum.

Dilution water is not a free input. One producer warns that chloride above about 10 parts per million in the dilution water can cause severe corrosion, and the producers’ standard requires water used for on-site dilution to be analysed and approved by the supplier. A plant that dilutes 50 percent down to a working strength with whatever is in the line is making a decision about its own equipment, and it is a decision to settle with the producer rather than on the shop floor.

The exposure limit, and India has its own

The Second Schedule to the Factories Act, 1948, which section 41-F makes the schedule of maximum permissible threshold limit values of chemical and toxic substances in the work environment, lists hydrogen peroxide at serial number 64, at 1 part per million and 1.5 milligrams per cubic metre in the eight-hour time-weighted average columns. The short-term exposure limit columns against that entry are blank, and the entry carries no letter suffix — which is worth noting, because the neighbouring entries do: hydrogen chloride at serial 61 and hydrogen fluoride at 63 both carry the ceiling-limit marker. That is the figure that binds an Indian factory, and section 41-F applies it in any factory.

The international figures sit at the same level and carry the supporting detail: the United States recommended and permissible limits are both 1 part per million, given as 1.4 milligrams per cubic metre, the American industrial hygienists’ threshold limit value is 1 part per million as an eight-hour average with an animal carcinogen notation of unknown relevance to humans, and the level considered immediately dangerous to life or health is 75 parts per million.

Two honest observations about the Indian entry. First, 1.5 against 1.4 milligrams per cubic metre is not a difference of substance — both are 1 part per million, and the gap is the conversion basis. But it does not reconcile with the Schedule’s own stated method: the Schedule’s footnotes say the milligram figure is derived from the parts-per-million figure using the molecular weight divided by 24.45, and 34.01 divided by 24.45 is 1.39. We report the statutory figure as printed and flag the arithmetic rather than quietly correcting it, and we read it in two independent reproductions of the Schedule that agree on the serial number and both numeric columns. Second, the carcinogen notation carried by the international value has no counterpart in the Indian entry, which has no suffix at all. A compliance officer reading only the Second Schedule gets no carcinogenicity signal from it.

A sourcing warning that matters more here than on most products. The international chemical safety card for hydrogen peroxide, card 0164, is titled for solutions greater than 60 percent in water and tabulates properties only for 90 and 70 percent. It is the wrong card for 50 percent material, and a lookup by the chemical’s registry number serves it anyway. Its transport block gives packing group I, which is correct for its own band and wrong for 50 percent. The occupational pocket guide’s physical properties — relative density 1.39 and boiling point about 141 degrees Celsius — describe near-pure material rather than a working solution; they are close to the safety card’s figures for 90 percent and nowhere near a 50 percent product. We looked for a safety card covering the 20 to 60 percent band and did not locate one; we are not asserting that none exists. For 50 percent, the figures to use are the producers’ own property tables and the dedicated emergency response datasheet for the 20 to 60 percent band.

One of those properties is worth stating because the intuition is backwards: 50 percent solution freezes at about minus 52 degrees Celsius, on two producers’ agreeing figures — far below water, and far below the safety card’s melting point of about minus 11 degrees Celsius for 90 percent solution. The peroxide and water system has a eutectic. Unlike caustic soda lye, this is not a product with a cold-weather crystallisation problem in India. Its storage problem is heat, not cold. The same producers give the boiling point as 114 degrees Celsius, the density as 1.196 to 1.20 grams per millilitre at 20 degrees, and the vapour pressure as about 10 millimetres of mercury at 20 degrees.

The regulatory position, in the third person

Four regimes bear on hydrogen peroxide in India in ways that affect a purchase, and each attaches to a particular party. None of what follows is a statement about Velora.

Hydrogen peroxide has been under a Quality Control Order for years, and the dates matter

The instrument is the Hydrogen Peroxide (Quality Control) Order, 2022, notified as S.O. 4000(E) of 24 August 2022 by the Department of Chemicals and Petrochemicals under section 16 of the Bureau of Indian Standards Act, 2016, after consulting the Bureau. It supersedes the Hydrogen Peroxide (Quality Control) Order, 2020. It is four paragraphs long. It requires the goods in its table to conform to the corresponding Indian Standard and to bear the Standard Mark under a licence from the Bureau, under Scheme-I of Schedule-II of the conformity assessment regulations of 2018; the table entry reads hydrogen peroxide, IS 2080 : 2021. The Bureau is named as the certifying and enforcing authority, and contravention is punishable under the Act.

Its only carve-out is for goods meant for export. There is no small-producer exemption, no exemption by scale and no transitional proviso other than the commencement rule, which is the ninety-first day after publication in the Official Gazette. The Order states no calendar date; counting from 24 August 2022 puts its commencement in late November 2022, and that is our arithmetic rather than a gazetted date.

But the 2022 Order is not where the obligation starts, and a buyer who dates it from November 2022 will misjudge older stock and older contracts. The 2022 Order supersedes an earlier one, and the sequence published by the department runs:

So mandatory certification for hydrogen peroxide took effect on 13 March 2022, and the Bureau’s own standards-portal entry for IS 2080 : 2021 agrees, recording the Quality Control Order as implemented on that date and the certification as mandatory. The obligation was deferred twice before it bit, by a total of about fifteen months — so this product is an instance of the familiar Indian pattern of slipped commencement dates rather than an exception to it, and the second amendment is the one that ended the slipping. What we can say is narrower than a general negative: we read four departmental extension notifications issued between October 2022 and October 2023, covering fatty acids, morpholine, acetic acid, methanol, aniline, pyridine, picolines, potassium carbonate, acetone, sodium tripolyphosphate and methylene chloride, and hydrogen peroxide appears in none of them. Anyone treating IS 2080 conformity as a future requirement for this product is years late.

One documentary oddity that a buyer should know about, because it will turn up in a supplier’s paperwork. The department’s published list of chemical Quality Control Orders carries the three 2020-and-2021 hydrogen peroxide instruments and names IS 2080 : 1980 (Reaffirmed 2016) against them; the list shows no row for the 2022 Order at all, and it carries a last-updated stamp of February 2025. The 2022 Order itself names IS 2080 : 2021 in its table. The Order is the instrument and the department’s summary table is not, so the mandated standard is the 2021 edition — but a certificate citing IS 2080 : 1980 is a predictable consequence of that table, and it is not evidence of conformity to what the Order in force requires.

The prohibition itself is in section 17 of the Act, and its drafting is the thing to notice. Section 17(1) is written against no person, and the acts it prohibits for goods notified under section 16 are to manufacture, import, distribute, sell, hire, lease, store or exhibit for sale them without the Standard Mark under a valid licence, or to apply the Mark to goods that do not conform. Section 17(2) separately bars claiming conformity in public — in an advertisement, a sales promotion leaflet, a price list or a declaration on the goods — without a valid certificate or licence. So a technical data sheet or quotation reading conforms to IS 2080, issued by someone holding no licence, engages section 17(2) independently of whether any mark was ever applied to a drum. Section 29(3) carries imprisonment of up to two years or a fine, with a floor stated for a first offence and a higher one for repeats, extending to a multiple of the value of the goods; section 29(4) makes that offence cognizable.

Two opposite errors are common here and both should be avoided. One is to treat a Quality Control Order as the manufacturer’s problem alone: section 17(1) names distributing, selling and storing. The other is to read it as creating a licence for traders: the licence under the scheme is the manufacturer’s, and nothing in S.O. 4000(E) or in sections 16 and 17 creates a separate licence or registration for a reseller. The accurate statement is about the goods. Hydrogen peroxide sold or stored in India is required to conform to IS 2080 : 2021 and to carry a valid Standard Mark, and the prohibition reaches any person. And it has been so since 2022, not since last year.

The export carve-out cuts both ways, and this is the sharpest commercial line in the article. Goods meant for export are outside the Order, so an exporter’s unmarked material is lawfully unmarked for that purpose. Material diverted from that route into the domestic market is a different matter, and section 17(1) names the storer and the seller alongside the manufacturer. Unmarked drums offered cheaply on the explanation that they were made for export are worth declining on that ground alone. We state no position on how the Order interacts with imports under advance authorisation or by units in export zones, because we could not read the relevant trade policy provisions on an official source. We also note a customs instruction of September 2025 reported to exempt aerospace grade hydrogen peroxide imported for non-commercial space research from the Order, on the basis that the Indian Standard has no category for it; we identified the official page but could not open it, and that relief, if it reads as reported, is specific and must not be generalised to industrial material.

It is a hazardous chemical under the 1989 rules, and the duty sits on the site

Hydrogen peroxide is listed by name in Schedule 1, Part II of the Manufacture, Storage and Import of Hazardous Chemical Rules, 1989, at serial number 318, with no concentration qualifier. The absence of a qualifier appears to be deliberate drafting rather than an omission: other peroxide entries in the same list carry one, methyl ethyl ketone peroxide for instance being listed at a stated concentration. So the entry catches 50 percent material.

What it does not have is a threshold quantity. Hydrogen peroxide is in neither Schedule 2 nor Schedule 3, so the quantity-triggered machinery — advance notification for isolated storage, safety reports, on-site and off-site emergency plans — is not engaged by holding it, in any quantity. The organic peroxides and peracetic acid that do appear in Schedule 3 with tonnage thresholds are a different chemical class, and a reader skimming those schedules for the word peroxide will come away with thresholds that do not apply. The Gujarat Factories Rules carry a schedule that mirrors it, and on the reproduction available to us that schedule lists organic peroxides and peracetic acid and does not name hydrogen peroxide — which we give as the result of reading one commercial reproduction of a State instrument, not as a cleared position.

The duties that remain are the ones that do not depend on a threshold, and they point at the buyer. On the text available to us, two or three survive a Part II listing on its own. Two sit squarely on the occupier: a general duty, in an industrial activity under its control, to identify major accident hazards and take adequate steps to prevent them and to limit their consequences; and the duties on collection, development and dissemination of information, which require the occupier to obtain or develop a safety data sheet, to keep it accurate and to update it as significant new hazard information becomes available. A third is genuinely ambiguous: the rule requiring an occupier to notify a major accident to the inspectorate within forty-eight hours and then furnish a report reads as unconditional on its own terms, while the schedule headings list it among the rules the threshold quantities trigger. We cannot settle that, and a buyer should know that a reportable event may have a forty-eight hour clock on it. The advance site report, the safety report and the emergency plans are the threshold-triggered machinery named in the paragraph above, and they are not engaged for this chemical — a buyer should not commission them in the belief that this listing requires them. The safety data sheet duty is expressly the occupier’s. The container marking requirement is different: it has three heads — the contents, the name and address of the manufacturer or importer, and the physical, chemical and toxicological data — and on the face of the rule it is drafted against the container and allocated to no party at all, the manufacturer or importer appearing as label content rather than as duty-holder. So it is not safe to say the rules put nothing on the supplier: the marking duty names nobody, and the rules carry separate duties on an importer, which a buyer who imports directly carries itself. A supplier’s safety data sheet supports the occupier’s compliance; it does not discharge it.

We read those rules in a State pollution control board publication and a commercial consolidation rather than in the gazette, and the schedules rather than a clean text of every rule. So we give the duties in outline, quote none of them, and cite no sub-rule number. This allocation is the single most commonly inverted point in this area, and it was inverted once in drafting an earlier article in this series, so check it against the consolidated text published by the environment ministry before acting on this paragraph. Note also that an importing buyer is itself an importer under those rules and so wears both hats.

The tax entry splits, and most of the market is on the wrong side of the split

Hydrogen peroxide sits in customs heading 2847, and the rate notification of 17 September 2025, effective 22 September 2025, treats it in two places:

• Schedule I, at 5 percent, carries a named entry for heading 2847 reading medicinal grade hydrogen peroxide.
• Schedule II, at 18 percent, carries the residual Chapter 28 entry for all inorganic chemicals other than those specified elsewhere.

So bulk industrial or technical 50 percent solution is at 18 percent, through the residual entry, and there is no named entry for it. The 5 percent applies to medicinal grade. That split is the story rather than the number, because a code lookup against 2847 will cheerfully return one figure, and a buyer who asks a supplier which entry applies should be told that there is none — it is the residual.

The September 2025 rationalisation did not reach hydrogen peroxide. The Chapter 28 reductions from 18 percent to 5 percent announced then were sulphuric acid, nitric acid and ammonia for the fertiliser sector. And the published release of the 57th Council meeting of 8 October 2026 records nothing touching Chapter 28 or inorganic chemicals — its goods items were sublimation paper, toys, seaweed-extract biostimulants, the margin scheme on second-hand vehicles, reverse charge and tax deduction on certain scrap streams and used cooking oil, psyllium seeds, re-treaded tractor tyres and canteen stores exemptions. In any case a Council recommendation takes legal effect only through the notifications that follow it, which that release says itself.

On imports, we have to stop short, and it is worth saying exactly where. We state no basic customs duty rate, because the official tariff schedule could not be reached. And on anti-dumping duty we state neither that one is in force nor that none is. What is on the record is that the trade remedies directorate has run two hydrogen peroxide investigations — one covering China, Indonesia, Korea, Turkey and the European Union with final findings in 2008, and one covering Bangladesh, Taiwan, Korea, Indonesia, Pakistan and Thailand, initiated in 2016, with final findings in 2017, a remand finding in 2018, a sunset review final finding of 16 March 2022 and a new shipper review finding of September 2022. We could not read what the Department of Revenue did with the 2022 recommendation, because the findings are published as scanned images and the revenue notification file could not be opened. A recommendation is not a duty, and an import price built on the assumption either way would be the most expensive error available in this article. Check the chapter-wise anti-dumping notification index before relying on a landed cost.

Packaged commodity declarations, and weight against volume

Two points, and the first is simpler than the discussion it usually attracts. Chapter II of the Legal Metrology (Packaged Commodities) Rules, 2011 does not apply to packages containing more than 25 kilograms or 25 litres, so a 35 kilogram carboy, a 240 kilogram drum and a tanker are outside it on size alone, whoever buys them. There is a second, independent route out for packages meant for industrial or institutional consumers, and that route carries three conditions which are easy to half-remember:

1. The channel. The buyer must have bought directly from the manufacturer, from an importer or from a wholesale dealer. This is the condition that fails in practice: the same drum bought through a retail trader is not within the definition, and the package becomes a retail package.
2. Captive use. For use by that industry or institution — and for an institutional consumer, expressly not for commercial or trade purposes.
3. The declaration. The package must bear not for retail sale. No marking, no exemption.

And a trader reselling is neither an industrial nor an institutional consumer, which is why that route is not a general answer for anybody in the chain.

The second point is commercial rather than regulatory. The Fourth Schedule to those rules provides that liquid chemicals may be declared by weight or by volume, at the seller’s choice. Since 50 percent hydrogen peroxide is about 1.2 kilograms per litre, 200 litres and 200 kilograms are not the same quantity, and neither figure states the mass of hydrogen peroxide in the package. Three quantities can all be called 200: 200 litres of solution, 200 kilograms of solution, and the roughly 100 kilograms of active peroxide inside the second of them. A bleaching or effluent contract is buying the third.

Road and air movement

The transport classification is concentration-banded, which is the source of a great deal of published error. On the international road, rail, sea and United States regimes as set out by one producer:

So 50 percent is UN 2014, Class 5.1 oxidising substance with subsidiary risk 8 corrosive, packing group II, and the emergency response guide for it is guide 140. The same producer records that below 8 percent by weight the material is not subject to the transport rules, and that the United States regime splits UN 2014 into sub-bands at 40 percent while keeping the same number, class and packing group throughout.

And one hard commercial limit: air transport of hydrogen peroxide above 40 percent by weight is prohibited under the international air regulations. Fifty percent material cannot travel by air at all, which among other things is why there is no air option for a sample consignment.

In India the duties under the dangerous-goods provisions of the Central Motor Vehicles Rules, 1989 fall on named parties, and nothing we verified turns on the product being an oxidiser rather than a corrosive — the regime is keyed to the goods being of a dangerous or hazardous nature and to the applicable class label. In outline:

Separately, the driver of a goods carriage carrying dangerous or hazardous goods is required to hold a transport vehicle licence, to be able to read and write an Indian language and English, and to hold a certificate from a three-day course at an institute recognised by the State Government, endorsed on the licence.

Two limits on the paragraph above. We read those rules in a State transport department restatement rather than in the gazette, so we quote none of it verbatim and state no panel dimension or lettering size. And the transport emergency card is a document the driver must carry, but the information on it originates with the consignor — so a buyer checking a tanker on arrival is checking the consignor’s work, and asking the question does not transfer the duty. Who is the consignor on a given movement is a matter of the contract, and it belongs in the contract rather than being inferred from a delivery term.

A commercial aside that is pure basis error

Because the strength basis runs through everything in this product, it runs through the published capacity figures too, and the result is worth a paragraph because it is the sort of thing that ends up in a tender justification.

One Indian producer states an installed capacity of 150,000 tonnes per annum on a 50 percent by weight basis, at a plant in Maharashtra. Another states, in an investor presentation, an installed capacity of 54,120 tonnes per annum on a 100 percent basis across two Gujarat sites, as at 31 January 2023. Those two numbers cannot be compared as printed. Put them on one basis: 54,120 on a 100 percent basis is about 108,240 tonnes on a 50 percent basis, and 150,000 on a 50 percent basis is about 75,000 tonnes on a 100 percent basis. The first producer is the larger on either basis — so the basis error does not reverse the ranking here. What it does is change the size of the gap by a factor of two: about 39 percent larger, against the 2.8 times that the raw figures suggest. That is quite enough to change a sourcing conclusion, and it is the reason to put the basis next to the number every time.

What that paragraph is and is not. It is arithmetic on two companies’ own published figures, and it is offered as a demonstration of the basis problem rather than as a market analysis. We got the direction of it wrong once in drafting, by comparing one company’s 50 percent-basis figure against the other’s 100 percent-basis figure, which is exactly the mistake the paragraph is about. We publish no market size, no market share, no capacity ranking and no price. The second figure is nearly four years old on its own face. One of the two companies’ own pages describe it both as the largest producer in India and as one of India’s leading manufacturers, which is a further reason to quote capacity sentences rather than superlatives.

The short version

The current Indian Standard is IS 2080 : 2021, with an Amendment No. 1 established in August 2023 that the Bureau’s own portal entry does not record; the 1980 edition is withdrawn, and the Government’s own Quality Control Order list still names it, so cite the year — and confirm any figure from the standard against the Bureau text, because the copy we could read is an imperfect scan. The standard has six grades and the grade is the strength, descending, so 50 percent is Grade 3 and Grade 1 is 70 percent. The strength is by mass, and a litre of 50 percent by mass carries about 600 grams of hydrogen peroxide against 500 for a litre of something quoted as 50 percent by weight in volume — one sixth less for the same two digits; 50 percent by mass is about 60 percent by weight in volume, 198 volumes, and 23.5 percent active oxygen. The specification table has only seven rows, no chloride or sulphate limit at all, and one row that is a band rather than a ceiling: acidity, which has a minimum, in a product that both producers we read describe as destabilised sharply by alkali, though they differ on how far the acid side helps. Stabilised is a test result, not an ingredient — the standard names no stabiliser and requires none to be declared, so the stabiliser package is the largest undocumented difference between two conforming materials and a change of producer is something to test. The product decomposes continuously, typically under one percent of its strength a year in normal storage on two producers’ figures, faster with heat, with transition metals such as iron, copper, nickel and chromium, with dust, with rough surfaces and above all with alkali; and because that evolution never stops, a sealed container pressurises, which is why the standard demands 10 percent ullage with a vent and half the container without one, why the drum has a continuous vent that must not be blocked, why drums go upright and no more than two high, and why unused material never goes back in. Materials are high-purity aluminium of named alloys, 304 or 316 stainless passivated before service, high density polyethylene up to 50 percent — and not polypropylene, not copper, brass or bronze, not mild steel and not zinc. It is an oxidiser and not a fuel, so water is the only firefighting agent and combustible absorbents are the classic fatal mistake, while the detonation risk at this strength comes from what it is mixed with rather than from the product as supplied. India sets the airborne limit at serial 64 of the Second Schedule to the Factories Act, 1 part per million. And on paper: mandatory certification in force since 13 March 2022, after two deferrals, and now under the 2022 Order requiring conformity to IS 2080 : 2021 and the Standard Mark with an export-only carve-out; a hazardous chemical listing with no threshold quantity and the safety data sheet duty on the receiving site; 18 percent tax through the residual entry, with 5 percent reserved for medicinal grade; and UN 2014, class 5.1 with a corrosive subsidiary, packing group II, with air transport prohibited above 40 percent.