Raise the chlorine and the problem goes away. That assumption sits behind a lot of disinfection programmes, and it is the one that quietly lets sites down. Legionella does not shrug off chlorine the way a hardy weed shrugs off a weak weedkiller. It survives mostly by hiding — inside the biofilm that coats pipe walls, and inside the amoebae that graze on that biofilm — so the disinfectant you dosed never reaches it at the strength the dosing chart promised [3].

That reframes the whole question. The useful thing to ask is not “how much chlorine kills Legionella?” It is “where in this system does the disinfectant actually arrive, at what concentration, and for how long?” Answer that and the limits of chemical control stop being mysterious.

Why “tolerance” mostly means shelter

Bacteria floating free in the water — planktonic cells — are relatively easy to kill. A measured disinfectant residual deals with them. The trouble is that most of the Legionella in a real system is not floating free. It is bound up in biofilm, the living slime on the inside of tanks, pipes and shower hoses, often sheltering inside the protozoa that feed there. Biofilm is both a physical and a chemical barrier: it consumes oxidising disinfectant at its surface and shields the cells deeper in.

So chlorine tolerance is less a property of the bacterium and more a property of where it lives. You can hold a textbook residual in the bulk water and still leave a thriving population in the biofilm a millimetre away. That is also why a single shock dose rarely fixes a biofilm problem on its own — the surviving film simply reseeds the water once dosing stops. For why that layer is so stubborn, see Legionella in biofilms.

The residual that matters is the one at the far tap

Chlorine is reactive by design, which means it gets used up. Every metre of pipe, every patch of scale and sediment, every warm tank draws down the residual. Dose to a sensible level at the plant and you can still measure almost nothing at the outlet on the top floor — and that outlet is exactly where the water has sat longest and warmest.

Dead legs and low-use branches make this worse. Water in a capped spur or a rarely-used shower exchanges slowly with the treated main, so whatever residual it once carried has long since decayed. Stagnation and loss of disinfectant tend to travel together, which is why no amount of clever chemistry substitutes for getting water to move. environmental factors affecting Legionella survival covers how temperature and stagnation drive survival alongside the chemistry.

It is also why banking on the residual in the incoming mains is a mistake. Mains water arrives treated, but that residual exists to protect the public network, not your storage and pipework. Once it sits in a cold tank for a day, the protection has gone.

Where the limit bites on a real site

Three situations come up again and again.

First, after remedial work a system gets a shock disinfection — hyperchlorination, an elevated free chlorine concentration held for a defined contact time — and a clean sample comes back. The clean sample is genuine, but it describes the water at one moment. If the biofilm survived, counts can climb again within weeks. Treat a post-disinfection result as a point to start monitoring from, not a finish line.

Second, where temperature control alone cannot be achieved, a continuous treatment programme — chlorine dioxide, or copper-silver ionisation, for example — may be used instead, and HSG274 sets these out as recognised options [2]. The catch is that they only work while a controlled residual genuinely reaches the whole system, including the awkward far ends. That means dosing control, residual monitoring at sentinel points, and managing by-products and material compatibility. A treatment rig left running without that monitoring is decoration.

Third, there is the choice to lean on chemistry as the primary control at all. In UK practice the traditional first line for hot and cold water systems is temperature — keep hot water hot, cold water cold, and water moving [1][5]. Chemical disinfection is a tool within that picture, not usually a replacement for it. Reaching for biocide because the temperatures are wrong treats the symptom and leaves the cause in place.

Before you bank a disinfection result

A short field check before you accept that a disinfection or treatment programme is actually controlling the risk:

  • Confirm what was dosed — disinfectant, concentration and contact time — and that it is recorded against the written scheme, not just remembered.
  • Measure the residual at sentinel and far outlets, not only at the plant, and note where it has decayed to little or nothing.
  • List the dead legs and low-use outlets the dose may never have reached, and flag each for removal or scheduled flushing.
  • Check whether biofilm-forming conditions remain — scale, sediment, lukewarm storage, slow turnover — because these will reseed the water [3][4].
  • Schedule follow-up sampling after the clean result, so regrowth is caught rather than assumed away.
  • Record the decision and name who owns the next check, with a review date set by risk.

A word on simply dosing higher

Pushing the residual up indefinitely is not a free safety margin. Higher oxidant levels accelerate corrosion, can attack seals, fittings and some pipe materials, raise disinfection by-products, and create taste, odour and discharge problems — all of it under COSHH for the people handling the chemicals. There is a practical ceiling, and chasing biofilm with brute-force chlorine usually costs more in damaged plant than it buys in control. Cleaning, descaling and temperature do the work chlorine cannot.

How sure can you be?

These are general principles, not a dosing specification. The right disinfectant, concentration, contact time and residual for your building depend on its materials, the supply water, and what your risk assessment and a competent water treatment provider determine together. Any figure you read — including the ranges described here as guidance — is a prompt to check the current HSE guidance and your own written scheme, never a setpoint to dial straight in. Sampling supports that judgement; it does not replace control of temperature, stagnation and cleanliness.

Common questions

Does a clear post-disinfection sample mean the chlorine worked?

It means the sampled water was clean at that moment. If biofilm survived in the pipework, the population can rebound after dosing stops, so the honest reading is “promising, now monitor” rather than “solved”.

Chlorine or temperature — which should be the primary control?

For most UK hot and cold water systems, temperature is the traditional primary control, with chemical treatment used where temperature alone cannot be maintained [1][2][5]. The two work together; they are not interchangeable.

Can I just hold a higher chlorine residual to stay safe?

Only up to a point. A higher residual decays faster against scale and long pipe runs, corrodes plant, and raises by-products. Reaching the biofilm and keeping water moving matter far more than the headline number.

What to do next

Pull your last disinfection or water-treatment report and check one thing: was the residual measured at the far outlets, or only at the plant? If it was only at the plant, you do not yet know whether the disinfectant reached the parts of the system where Legionella actually hides. Book a residual check at the sentinel and worst-case outlets, and decide what result would genuinely prove control before the next round of dosing.

Sources

[1] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [2] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm [3] CDC, “How Legionella Spreads”. https://www.cdc.gov/legionella/causes/index.html [4] CDC, “Controlling Legionella”. https://www.cdc.gov/control-legionella/index.html [5] HSE, “Hot and cold water systems”. https://www.hse.gov.uk/legionnaires/hot-and-cold.htm