Median pH 12.36. Every Single One Was Over the Corrosive Line.

Occupational safety rules put the irritant threshold for skin at pH 10.5 and the corrosive threshold at 11.5. A team in Cape Town bought 121 hair relaxers across 39 brands, measured each one three times on three consecutive days with a benchtop meter, and found a median pH of 12.36.

Not the worst one. The median.

Which means the ordinary product, the one in the middle of the shelf, is above the line at which a workplace regulator would classify a substance as corrosive to human skin — and it is sold to be applied to a scalp, by hand, at home, frequently by a family member, sometimes to a child.

Lye and no-lye are the same on the measure that matters

The distinction is real chemistry. Lye relaxers use sodium hydroxide. No-lye relaxers use guanidine hydroxide made on the spot by mixing calcium hydroxide with guanidine carbonate, or lithium hydroxide, or calcium hydroxide alone. No-lye has been marketed as the gentler option for decades on the strength of that difference.

In the Cape Town sample — 63 percent sodium hydroxide, 20 percent calcium hydroxide, 17 percent lithium hydroxide — there was no pH difference between lye and no-lye. P equals 0.274.

Nor between adult and children’s products. Sodium hydroxide formulas, p = 0.27. Lithium hydroxide, p = 0.68. Calcium hydroxide, p = 0.10. Six of the 76 sodium hydroxide products were marketed for children.

Whatever no-lye is doing, it is not lowering the pH. On this measure the distinction is a marketing claim rather than a chemical one.

There is a small study running the other way that is worth knowing about because it complicates rather than confirms. Five South African women, half-head design, lye against no-lye, with amino acid analysis. The no-lye product straightened better — and destroyed more cystine. Median cystine fell from 9.1 to 7.8 with the lye product and to 4.0 with the no-lye one. The authors then concluded the no-lye product was “more effective and safer to use,” which does not follow from their own fiber data. Five subjects. Note it and move on.

The part that should have been a recall

Alkaline hydrolysis does not stop when the hair looks straight. It stops when the pH comes down. That is what the neutralizing shampoo is for — it is not a conditioning step, it is the off switch.

Sixty-four of the 70 adult sodium hydroxide relaxers in that sample — 91 percent — were packaged without one. Four of the six children’s sodium hydroxide products were sold the same way.

A caustic at pH 12.4, sold for home use on a scalp, including on children, without the component that terminates the reaction.

The authors’ conclusion is measured and damning: the pH “may contribute to the high prevalence of alopecia in females with afro-textured hair. A review of permissible safe pH levels for cosmetic use is warranted.”

What the chemistry does to the fiber, which is not in dispute

Both kinds of relaxer work by breaking disulfide bonds between cystine residues and converting them to lanthionine. Lanthionization. The bond does not re-form. That is the whole point of the product — it is why the result is permanent until new hair grows — and it is also why the damage is permanent.

Khumalo’s amino acid work put numbers on it. Natural hair showed no weathering gradient: cystine at 14 mg per gram of amino acid nitrogen at the root, 14 at the tip. Relaxed hair from women with no complaints at all measured 7.5 at the root and 3.3 at the tip.

The line in that paper that ought to be on the box: cystine levels in the distal ends of asymptomatic relaxed hair, and in symptomatic relaxed hair, were similar to those found in trichothiodystrophy — a congenital hair-fragility disease.

Hair that its owner describes as fine, in good condition, no complaints, is biochemically comparable to a genetic disorder of brittle hair. That is not an exaggeration for effect. It is the comparison the authors drew.

What this does and does not prove

It does not prove relaxers cause scarring alopecia. That question is separately contested and the two largest datasets are negative — Olsen’s 529 women found no relaxer association, and Khumalo’s 1,042 schoolchildren, 78 percent of the girls already relaxing, produced zero cases of central centrifugal cicatricial alopecia. Both datasets, and the one that runs the other way, are gone through in the CCCA piece. Anyone telling you the causal chain is settled is ahead of the evidence in one direction or the other.

What it does establish: the product is corrosive by the standard a safety regulator would apply, the gentler version is not gentler on that measure, the fiber damage is permanent and comparable to a named disease, and the majority of these products are sold without the thing that stops the chemical reaction.

Acute scarring alopecia from relaxer burns exists as a published case series. Five patients. There is no population-level figure for how often relaxers burn a scalp, and I looked. Any percentage you have seen for that is unsourced.

If you relax anyway

The neutralizing shampoo is not optional and it is not the same as ordinary shampoo. If the box does not contain one, that is a reason to buy a different box, not a reason to skip the step.

Do not relax a scalp that is already broken — scratched, sore, recently washed with anything abrasive. Intact skin is the only barrier between pH 12.4 and the tissue underneath it.

And the children’s formulas are not a milder product. They are the same alkalinity in different packaging, and two thirds of the sodium hydroxide ones come without the off switch.

None of which is a verdict on the decision to use one. The reason that decision gets made is a professional norm that has been legislated against and never measured — a scoping review looking for evidence that the law changed anything found none, while the pH readings above are not in dispute.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.