Carbon-14 decays with a half-life of about 5,730 years.

A half-life describes the statistical behavior of a large collection of radioactive nuclei. After one half-life, about half the original carbon-14 remains undecayed on average. It does not mean that every individual nucleus survives for 5,730 years and then decays on schedule.

Living organisms exchange carbon with their surroundings through processes such as photosynthesis and feeding. After death, that exchange generally stops. The carbon-14 already present continues to decay, allowing its remaining abundance to provide information about the time since the material stopped exchanging carbon.

A radiocarbon measurement is not automatically a calendar date. The initial carbon-14 proportion has varied, and different carbon reservoirs can have different apparent ages. Researchers compare measurements with appropriate references and calibration data, while considering contamination and the history of the sample.

Carbon atoms always have six protons, but their isotopes differ in neutron number. Carbon-14 has eight neutrons, giving a total mass number of fourteen. It is radioactive and undergoes beta decay to nitrogen-14. In that process, a neutron changes into a proton, changing the element. The familiar name therefore identifies a specific unstable isotope, not a different chemical element.

Carbon-14 is produced in the atmosphere through reactions involving neutrons generated by cosmic rays and atmospheric nitrogen. It becomes incorporated into carbon dioxide and enters biological carbon cycles. While an organism is alive, these exchanges connect its carbon to environmental sources rather than allowing a completely isolated decay clock to run.

After exchange stops, radioactive decay progressively reduces the amount remaining. After two half-lives, the expected fraction is one quarter; after three, one eighth. The same fractional loss repeats over equal intervals, so the decline is exponential rather than a fixed number of atoms disappearing every year.

The method is most useful for once-living material from the last tens of thousands of years. As the remaining carbon-14 becomes very small, background signals and contamination make reliable measurement harder. A tiny amount of newer carbon can significantly affect a very old sample because it adds a comparatively carbon-14-rich component.

Radiocarbon dating cannot establish the age of Earth or directly date most ancient rocks. Those questions require other radiometric systems with suitable isotopes and geological contexts. Even for organic remains, a date concerns the sampled carbon: old wood reused in a younger structure, for example, need not record the construction date itself.

Radiocarbon dating measures a sample's carbon history, not the age of every object found beside it. The dated material must be associated convincingly with the event under investigation. A precise laboratory result can still answer the wrong archaeological question if the sample's origin or later contamination is misunderstood.