Get the Offline ArchiveFree lifetime updates

Making Bleach From Salt Water

11 min read

Passing direct current through salt water in a single undivided vessel produces sodium hypochlorite — liquid bleach — because the chlorine made at one electrode immediately reacts with the alkali made at the other instead of escaping. The electricity, not the salt, is the limiting input: about 53.6 amp-hours yields one mole of available chlorine at perfect efficiency, so a 10 amp supply running an hour gives roughly 13 grams, enough to dose several thousand litres of water. The product is weak, usually well under one percent, and it decays over weeks, so it is made as needed rather than stockpiled.

Part of Chlorination

Producing hypochlorite solution on site from brine and direct current, as an alternative to scavenging commercial bleach or handling chlorine gas. The deliberate opposite of the industrial chlor-alkali process, which exists to keep the two products apart.

The Reaction Industrial Chemistry Works Hard To Prevent Is The One You Want

Electrolyse brine and two things happen at once. At the anode, chloride ions give up electrons and become chlorine. At the cathode, water is split, releasing hydrogen and leaving hydroxide behind:

anode     2 Cl⁻          →  Cl₂ + 2 e⁻
cathode   2 H₂O + 2 e⁻   →  H₂  + 2 OH⁻

In a divided cell those products are kept apart, which is the whole point of the chlor-alkali process — it exists to harvest chlorine gas and caustic soda separately. Remove the divider and they find each other in solution:

Cl₂ + 2 OH⁻  →  ClO⁻ + Cl⁻ + H₂O

Net, the cell consumes salt and water and produces hypochlorite and hydrogen:

NaCl + H₂O  →  NaClO + H₂

That is the entire trick. The corpus’s own chlor-alkali article calls this reaction the thing that happens “without separation… rather than the pure chemicals you want” — and it is right, if what you want is chlorine and caustic. For a settlement that needs to make water safe, it is the opposite: the simplest cell gives the most useful product. No diaphragm to build or replace, no chlorine gas to contain, no caustic to evaporate.

Note what leaves the vessel, though. The hydrogen does not stay in solution, and it is why this process demands respect rather than care.

Hydrogen is the real hazard here, not the chlorine

Every molecule of hypochlorite you make comes with a molecule of hydrogen, and that hydrogen leaves the liquid immediately. It is 4 to 75 percent flammable in air — an enormous range — and at 2 g/mol it rises straight to the ceiling of whatever space you are working in, where it collects and waits.

Run the cell in the open air or under a roof open on at least two sides, never in a shed, a cellar or a closed room. Keep every ignition source away: no flame, no cigarette, no switch or relay that arcs, no motor brushes. Make and break the DC connections at the power source, away from the vessel, never by touching leads together over the brine — a spark at the surface is a spark in the exact place hydrogen is densest.

There is no smell and no colour to warn you. This is the mechanism that injures people doing this, and it does not announce itself.

Never let the solution turn acidic, and never store it near acid

Hypochlorite is stable in alkali and falls apart in acid, releasing chlorine gas: ClO⁻ + Cl⁻ + 2 H⁺ → Cl₂ + H₂O. The cell protects itself as long as it stays alkaline, which it does naturally — the cathode makes hydroxide continuously.

So: add no acid to the cell, ever, and keep the finished solution away from vinegar, descaler, battery acid and the mineral acids. Chlorine does severe respiratory damage at around 10 ppm within minutes, by the figure this corpus’s chlor-alkali article gives, and a bucket of bleach meeting a bucket of acid in a confined space is the classic way people are hurt by a chemical they made safely.

Also never mix it with ammonia or with urine — that gives chloramine vapours, a different poison by the same carelessness.

Your Anode Choice Decides Whether The Product Is Usable

This is the step most likely to be got wrong, because the obvious material is the wrong one.

The anode sits in chlorine as it is being born, which is about as aggressive an environment as a settlement can create. Do not use stainless steel for the anode. It will work, visibly, for a while — and it dissolves chromium and nickel into the solution as it does. You would be adding heavy metal to water you intend to drink, and hexavalent chromium is a carcinogen. Stainless is fine for the cathode, where nothing attacks it.

Anode materialVerdict
Graphite or hard carbonThe practical choice. Survives chlorine reasonably, erodes slowly, and sheds only carbon particles that settle out. Lantern-battery rods, arc-lamp carbons, brush stock, EDM graphite
Platinum-group coated titanium (MMO/DSA)What industry uses, and the best thing to salvage. Pool chlorinator cells are exactly this
Stainless steelNo. Leaches chromium and nickel into the product
Copper, brass, aluminium, lead, zincNo. All corrode and all contaminate
Iron or mild steelNo as an anode — it dissolves and the product turns rust-coloured. Fine as the cathode

Carbon rods erode and will eventually need replacing; that is normal and is not a sign of a problem. See Electrode Selection for the wider treatment.

What You Need

  • Salt — ordinary non-iodised salt if you have it. Around 30 to 60 g per litre of water is a reasonable working strength; this is not the saturated 300 g/L brine the industrial process wants, because you are not chasing yield per pass. Anti-caking agents are a cosmetic nuisance, not a safety problem
  • Water — the cleaner the better. Salt dissolved in muddy water makes bleach with mud in it
  • A DC supply — a battery, a solar panel through a controller, or a hand-cranked generator. A few volts and as many amps as you can spare. DC only: alternating current makes both electrodes do both jobs and produces nothing
  • Two electrodes — a carbon anode per the table above, and a steel or stainless cathode
  • A non-metallic vessel — glass, glazed ceramic, or HDPE. Not metal, which becomes part of the circuit, and not a container that ever held fuel or pesticide. See Container Selection
  • Open air, and a way to stir

Method: Run It Cool, Stirred, And Not For Too Long

Step 1: Mix the brine. Dissolve the salt fully in the water in your vessel. Undissolved salt on the bottom does nothing; it is not a reserve.

Step 2: Set the electrodes apart, both well submerged. A few centimetres between them. Closer means less resistance and more current, but also more mixing of fresh chlorine with the anode itself, which wastes it. Make sure neither can touch the other or the vessel wall if bumped.

Step 3: Connect the supply — anode to positive, cathode to negative. Reversing this ruins a steel cathode by making it the anode. If you are unsure which lead is which, the cathode is the one that fizzes much harder: hydrogen comes off in roughly twice the volume of anything at the other electrode.

Step 4: Stir, and keep it cool. Stirring is not optional. It carries chlorine away from the anode and into the alkali where it converts, instead of letting it bubble out of the surface and into your face. Heat is the enemy twice over — hypochlorite decomposes faster warm, and warmth pushes the reaction toward chlorate (below). If the vessel is noticeably warm to the hand, cut the current or stop and let it cool.

Step 5: Stop sooner than you think. The solution strengthens for a while and then starts going backwards. Hypochlorite that stays in the cell gets oxidised further to chlorate, which is not a disinfectant, is useless for your purpose, and is a strong oxidiser you did not intend to make. Long runs, warm cells and high current densities all push it that way. Run for a known time, stop, and judge the result by testing it rather than by how long you ran it.

Step 6: Decant and store cool, dark and closed. The product is pale yellow-green and smells of swimming pool. Opaque containers, out of sunlight, loosely capped — a sealed bottle of decomposing hypochlorite builds pressure.

The Electricity Is The Limiting Input, And You Can Size It

The chemistry fixes exactly how much current a given quantity of bleach costs, so this is worth working through once.

Making one ClO⁻ consumes one Cl₂ worth of electrons, which is two electrons. One mole of electrons is one faraday, 96,485 coulombs, so:

2 × 96,485 C  =  192,970 C per mole of hypochlorite
192,970 C ÷ 3,600 s/h  =  53.6 amp-hours per mole

One mole of hypochlorite carries the oxidising power of one mole of chlorine, 70.9 g, which is how chlorine dosing is conventionally expressed — “available chlorine”. So at perfect efficiency:

  • 53.6 Ah → 70.9 g of available chlorine
  • 10 Ah → about 13 g (10 ÷ 53.6 = 0.187 mol; 0.187 × 70.9 = 13.2 g)

Put that against a dose. Treating water at 2 mg/L of available chlorine:

13.2 g = 13,200 mg  ÷  2 mg/L  =  6,600 litres

So a 10 amp supply running for one hour makes, in principle, enough chlorine for several thousand litres of drinking water. That is the number that makes this worth building.

Now halve it, or worse, and do not trust the arithmetic over a test

Every figure above assumes 100 percent current efficiency, which no real cell achieves. Some current splits water instead of chloride, some hypochlorite is destroyed at the cathode as fast as it is made, some escapes as gas, and some goes on to chlorate. Published efficiencies for simple undivided cells vary widely and this guide has no sourced figure for a bucket cell, so treat the arithmetic as an upper bound and expect to need substantially more amp-hours than it predicts.

The arithmetic tells you the scale of what is possible — thousands of litres rather than dozens — and that is genuinely useful. It does not tell you the strength of what is in your bucket. Only a test does.

You Must Measure The Strength, Because Dosing Depends On It

A simple cell typically yields a fraction of a percent available chlorine, against 5 to 8 percent for commercial bleach — more than ten times weaker. Any dose instruction written for shop bleach will badly underdose water treated with this.

That matters because underdosed water is the dangerous failure: it looks and tastes treated while leaving pathogens alive. So:

  • Work out the dose from the strength you actually measured, using Dosing Calculation.
  • Confirm the result with Chlorine Residual testing, which is the only honest check that treatment worked.
  • Respect Contact Time. Chlorine needs time, and a weak solution needs more of it.
  • Remember chlorine is a poor choice against some organisms regardless of dose — Water Treatment at Scale sets out why disinfection is the last barrier and not the only one. Turbid water must be settled and filtered first, because suspended solids consume chlorine and shield what is hiding in them.

It Will Not Keep, So Make It On A Schedule

Hypochlorite solutions decompose on their own, faster when warm, in light, or contaminated with metals — and proportionally faster when weak, which is exactly what this process makes. A solution that tested adequately a month ago may be well down by now, with no visible change.

The practical consequence is a change in habit rather than technique: this is not a thing to stockpile, it is a thing to produce. A cell that runs for an hour a day from a solar panel is worth more to a settlement than a drum of bleach made once, because the drum is quietly losing strength while the cell is not. Make it close to when it is used, test it rather than trusting it, and size the batch to days rather than months.

What’s Next