Radioactivity: Why Some Atoms Fall Apart

In the last week of February 1896, Henri Becquerel was stuck waiting for the sun. Working in his laboratory at the Muséum National d'Histoire Naturelle in Paris, he had been investigating phosphorescent uranium salts, expecting that crystals charged by sunlight might emit the mysterious new rays Wilhelm Röntgen had announced the previous November. His plan was simple: let the salts soak up daylight, set them on a photographic plate wrapped in thick black paper, and check whether anything penetrated the wrapping. But Paris turned overcast for days, and with no sunlight to charge his crystals, Becquerel gave up for the moment and shoved the whole arrangement, salt resting on wrapped plate, into a drawer.

When he finally developed that plate on the first of March, he found the silhouette of the uranium salt printed clearly onto it, sharp and unmistakable. No sunlight had ever reached the crystals. Whatever exposed the plate came not from outside the uranium but from inside it, an emission the salt produced entirely on its own, in the dark, with no apparent source of energy. Becquerel had stumbled onto a property of matter no one had suspected: certain atoms are not stable at all, and they fall apart on their own schedule, flinging out radiation as they go.

This article follows that thread from the drawer in Paris to the modern hospital and the dating of the Earth. What is actually coming out of an unstable atom, why do some decay in minutes and others over billions of years, and why is none of it touched by the ordinary chemistry of heat, pressure, and bonding?

An accident in a drawer becomes a new science

The strangeness of Becquerel's result is easy to undersell. Röntgen's X-rays, discovered just months earlier, needed an apparatus: a vacuum tube, a high voltage, a stream of electrons slamming into metal. Becquerel's uranium needed nothing, sitting inert in a closed drawer and radiating all the same. The cloudy weather, far from ruining the experiment, was what made the discovery possible, because it removed sunlight as an explanation and left the uranium alone as the only source.

The phenomenon found its champion in a young Polish-born physicist in Paris. Marie Skłodowska Curie, with her husband Pierre Curie, took up Becquerel's puzzle and pushed it much further. Starting in 1898 at the Paris School of Industrial Physics and Chemistry, the Curies processed tonnes of pitchblende, a dark uranium ore from the mines of Bohemia, chemically separating it fraction by fraction and measuring the activity of each. Some fractions were far more active than the uranium content alone could explain, which meant the ore held other, more intensely radiating elements in tiny amounts. From this painstaking work they isolated two new elements, polonium (named for Marie's homeland) and radium, the latter roughly a million times more radioactive than uranium itself.

The recognition that followed was historic. Marie Curie shared the 1903 Nobel Prize in Physics with Pierre and Becquerel, and in 1911 won the Nobel Prize in Chemistry for the isolation of radium. She remains the only person to hold Nobel Prizes in two separate sciences. The word radioactivity, the spontaneous emission of particles or radiation from unstable nuclei, was hers.

Three kinds of rays, sorted by a magnet

If unstable atoms emit something, the obvious next question is what. The answer turned out to be not one thing but three, and the person who untangled them was Ernest Rutherford. Working at McGill University in 1899, he passed the radiation through a magnetic field and watched it split into distinct components that bent in different directions and by different amounts. A magnetic field deflects moving charges, so the way each component bent revealed its charge and roughly its mass. Rutherford named the three by the first letters of the Greek alphabet: alpha, beta, and gamma.

An alpha particle is a helium-4 nucleus: two protons and two neutrons bound together, carrying a charge of +2 and a mass of about 4 atomic mass units. It is heavy and slow as radiation goes, and although energetic, it loses that energy almost immediately on contact, so a single sheet of paper or the dead outer layer of skin will stop it.

A beta-minus particle is an electron, but not one plucked from an atom's outer shells. It is created in the moment of decay, when a neutron inside the nucleus converts into a proton, an electron, and an antineutrino, ejecting the latter two. The charge is minus one and the mass is tiny. Being light and fast, beta particles penetrate further than alpha, but a few millimeters of aluminum will absorb them.

Gamma radiation is different in kind. It is not a particle of matter at all but a high-energy photon, a packet of electromagnetic radiation with zero mass and zero charge, which is why a magnet does not deflect it. With no charge to grab onto and no mass to stop, gamma rays pass through material readily, and attenuating them takes centimeters of dense lead or tens of centimeters of concrete. The property that makes them useful for sterilizing equipment and imaging the body also makes them the hardest of the three to shield.

The clock that nothing can slow down

Knowing what comes out of an unstable nucleus still leaves the deepest question: when. A given atom of uranium might sit unchanged for a billion years, while an atom of a short-lived isotope might decay in the next second, with no way to predict which. Decay is fundamentally statistical, and the law that governs it is the half-life, the time required for half of a radioactive sample to decay.

The arithmetic is clean. After one half-life, half the original nuclei remain. After two, a quarter. After three, an eighth. After n half-lives, the fraction left is one over two to the n. The sample never quite reaches zero; it just keeps halving. The remarkable thing is how rigid this clock is. Half-life is a property of the nucleus itself, fixed for each isotope, and does not depend on temperature, pressure, chemical bonding, or how much of the substance you have, which sets radioactivity apart from nearly everything else a chemist studies.

Each isotope keeps its own time, and the spread is staggering. Carbon-14 has a half-life of 5,730 years, uranium-238 sits at about 4.5 billion years, comparable to the age of the Earth, and iodine-131, a fission product, decays in just 8.02 days. Three isotopes, three clocks at wildly different speeds, all governed by the same simple halving law.

Reading time itself, from a mammoth to the Earth

Because each isotope decays at a fixed rate, a radioactive sample is a clock, and you can read elapsed time by measuring how much has decayed. The trick is matching the clock to the question, since an isotope is only useful for dating things on the order of its own half-life.

Carbon-14 is the clock for the recent past. Living things constantly take in carbon, including a steady trace of radioactive carbon-14, and stop when they die, after which the carbon-14 simply decays, so measuring how much remains gives the time since death. With its 5,730-year half-life, carbon-14 reliably dates organic material from a few hundred years old out to roughly 50,000 years, beyond which too little remains to measure. The technique was developed by Willard Libby at the University of Chicago in 1949, and it transformed archaeology.

For deep time you need a far slower clock. Uranium-238, halving only every 4.5 billion years, dates the oldest objects in the solar system: the most ancient meteorites come in at about 4.567 billion years, and the oldest surviving terrestrial minerals, tiny zircon crystals from Western Australia, date to roughly 4.4 billion years. Match the clock to the timescale: you would not date a Pleistocene mammoth with uranium, nor a Hadean zircon with carbon.

Inside the clinic, the 110-minute deadline

Radioactivity is not only a tool for looking backward; it is a workhorse of modern medicine, where the half-life dictates the logistics of an entire department. Consider positron emission tomography, the PET scan. It relies on fluorine-18, an isotope that decays by emitting a positron, the antimatter counterpart of the electron, with a half-life of 110 minutes. The fluorine-18 is built into fluorodeoxyglucose, or FDG, a glucose look-alike that the body's hungriest cells eagerly absorb.

After a patient is injected, tissues with high glucose demand, such as many tumors, the brain, and the heart, take up the FDG and concentrate it. Each emitted positron travels a tiny distance before meeting an electron, at which point the two annihilate and convert their mass into a pair of gamma photons, each carrying 511 kiloelectronvolts, flying off in exactly opposite directions. The scanner's ring of detectors catches both and traces the line between them, mapping where glucose is consumed.

That 110-minute clock rules everything around the procedure. Fluorine-18 cannot be made in advance and stored; within a few hours, most of it is gone. A PET center therefore needs either its own on-site cyclotron or same-day delivery timed to the minute, and any unused dose is simply lost to decay. The physics of half-life is not an abstraction here; it is a delivery schedule.

Putting exposure on a single ruler

Radiation makes people anxious, partly because it is invisible and partly because the numbers are unfamiliar. The dose absorbed by living tissue is measured in sieverts, but a more intuitive way to anchor the scale is the banana. An ordinary banana contains potassium, a small fraction of which is radioactive potassium-40, so eating one delivers about 0.1 microsievert. This gives an unofficial but genuinely useful yardstick, the banana-equivalent dose.

The comparisons are clarifying. A chest X-ray delivers roughly 100 microsieverts, about a thousand bananas. A transatlantic flight, where thinner atmosphere lets through more cosmic radiation, gives around 40. The annual public dose limit in the United States, above natural background, is 1,000 microsieverts, about ten thousand bananas. Acute radiation sickness does not begin until around 1,000,000 microsieverts, ten million bananas, tens of thousands of times beyond any routine encounter. Seeing them side by side does not make radiation harmless, but it locates each exposure honestly on a scale where a scan and a catastrophe sit nowhere near each other.

Why chemistry cannot touch the nucleus

Underlying all of this is one fact students consistently find counterintuitive: radioactive decay is a nuclear process, not a chemical one, and the ordinary levers of chemistry do not reach it. Chemical reaction rates depend sharply on temperature, pressure, concentration, and the bonds an atom forms, while radioactive decay depends on none of them. Cooling a uranium sample to liquid-helium temperatures does not slow its decay, heating it to its melting point does not speed it up, and dissolving it in acid does not move the clock at all, because the decay happens deep in the nucleus, far below the electron shells where all chemistry takes place.

The energy scales make the separation vivid. Breaking a chemical bond involves a few electronvolts, while a nuclear transition releases a few megaelectronvolts, about a million times larger. The nucleus operates in a different regime.

One last distinction guards against a common confusion: half-life is not lifetime. It tells you how long it takes for half a large population to decay, but no individual nucleus has a fixed lifespan, since decay is purely statistical. This explains an apparent paradox: a long half-life makes an isotope, gram for gram, less dangerous, because fewer of its nuclei decay each second. Uranium-238, halving over 4.5 billion years, is feebly active and safe to handle in modest amounts, while the same mass of iodine-131, halving in eight days, would be acutely hazardous.

Key Takeaways

Radioactivity is the spontaneous emission of particles or radiation from unstable nuclei, discovered by Henri Becquerel in March 1896 when uranium salts imaged themselves onto a wrapped photographic plate in the dark, with Marie and Pierre Curie isolating polonium and radium from Bohemian pitchblende in 1898 and Ernest Rutherford sorting the radiation by magnetic deflection in 1899 into alpha (a helium-4 nucleus of charge +2, stopped by paper), beta (an electron born when a neutron becomes a proton, stopped by millimeters of aluminum), and gamma (a chargeless, massless high-energy photon needing centimeters of lead). The quantitative backbone is the half-life, the fixed time for half a sample to decay, a property of the nucleus alone that is independent of temperature, pressure, chemistry, and amount; it ranges from carbon-14 at 5,730 years for radiocarbon dating, through iodine-131 at 8.02 days and fluorine-18 at 110 minutes for PET imaging, to uranium-238 at 4.5 billion years for dating the Earth, while the banana-equivalent dose (about 0.1 microsievert each) places every exposure on one honest ruler, and the fact that half-life is statistical, not a lifetime, explains why a long-lived isotope is, gram for gram, the safer one.