An LD50 is the dose that killed half the animals in a test. It is a number about a specific experiment — a species, a route, a duration — and it stops being meaningful the moment those details are dropped. A great deal of bad chemical documentation consists of LD50 values with no species attached.
What the values mean
LD50 — median lethal dose. The dose, expressed per unit of body weight, at which half the test population died. Written as, for example, “LD50 oral rat 930 mg/kg”.
LC50 — median lethal concentration. Used where exposure is to a concentration in air or water rather than an administered dose, so it carries a duration as well: “LC50 inhalation rat 4 h 13,700 ppm”, or for aquatic toxicity “LC50 96 h fish 5.3 mg/L”.
EC50 — median effective concentration. The concentration producing a defined effect that is not death in half the population — immobilisation in daphnia, growth inhibition in algae. Ecotoxicology uses EC50 far more than LC50 because the endpoints of interest are usually sub-lethal.
What has to travel with the number
A value without these is not usable:
- Species — rat, mouse, rabbit; Oncorhynchus mykiss, Daphnia magna. Values differ between species by more than an order of magnitude.
- Route — oral, dermal, inhalation. The same substance can be nearly harmless by one route and lethal by another.
- Duration — for LC50 and EC50, mandatory. A 96-hour value and a 48-hour value are not comparable.
- Units — mg/kg body weight, mg/L, ppm, mg/m³. Converting between ppm and mg/m³ requires the molar mass and the conditions.
- Source — which study, and whether the value is measured or estimated.
“LD50 = 930” alone is not a fact. It is a fragment of one.
How they feed classification
Acute toxicity categories under CLP are set by thresholds on these values, per route. The categories run from 1, the most severe, to 4, and each has boundaries in mg/kg or in concentration. A value near a boundary is where classification decisions are argued, and where a document should show the value rather than only the resulting category.
For mixtures, CLP uses the acute toxicity estimate — the ATE — which allows a mixture to be classified from the values of its components without testing the mixture itself.
For aquatic classification, the M-factor works similarly: substances toxic far below 1 mg/L carry a multiplier so their contribution to a mixture is weighted accordingly.
What LD50 does not tell you
It does not describe chronic effects. A substance can have an unremarkable LD50 and be a carcinogen, a mutagen or a reproductive toxicant — those hazards are assessed on entirely different evidence and appear in different parts of a safety data sheet.
It does not give a safe level. The distance between a lethal dose and a level acceptable for repeated occupational exposure is not a fixed ratio. Derived no-effect levels and occupational exposure limits exist because that question needs its own answer.
It does not transfer between species without reasoning. Extrapolating from a rat study to a human is a defined step in a risk assessment, not an arithmetic one.
And it does not describe your material. An LD50 belongs to a substance; your container holds a preparation with a concentration, impurities and a physical form, all of which can change the outcome.
Reading them on a MolGod card
Where a MolGod substance page shows an acute toxicity value, it shows the species, the route, the duration where one applies, the units and the source that value came from. Where no value with those details is available, the page says so rather than showing a bare number. That rule — no data means no claim — is the same one that governs every other field on the page.
Related: classification categories and their thresholds are set in Regulation (EC) No 1272/2008 (CLP), Annex I, Part 3. Values quoted in MolGod documentation carry the source they were taken from.