Where did the energy go? The puzzle that led to the neutrino

The same starting point, different endings

Imagine repeating a decay experiment many times with the same kind of radioactive nucleus. Each nucleus starts in the same state and decays to a specified state of the daughter nucleus. If the only products were that daughter and an electron, their masses and the conservation laws would fix the electron’s energy. It would be shared in the same way each time.

Beta-ray measurements did not show just one electron energy. They showed a continuous spread, with many electrons below the upper end of the distribution. A plot of how many electrons appear at each energy is called an energy spectrum. Here, it seemed that the same decay could deliver different amounts of visible energy.

That qualification about the daughter’s state matters: different decay branches can have different energy releases. The puzzle was the continuous spread within a branch. Merely adding more discrete nuclear levels did not explain it.

Sources: [1] · [2] · [3]

Could the electrons have lost energy on the way out?

An electron can collide with matter before it reaches an instrument. So one possible explanation was quite ordinary: perhaps every electron began with the same energy, but some surrendered more of it inside the source. The detector would then see a spread even though the decay itself produced a single energy.

Charles Ellis and William Wooster tested this possibility in 1927 using radium E, the historical name for bismuth-210. Alongside the spectrum, they measured heat. A calorimeter collects energy deposited in matter; it can count energy that no longer belongs to the outgoing electron.

If slowing inside the source explained the whole shortfall, the lost energy should reappear as heat. Their measurement instead supported an average energy associated with the continuous spectrum, rather than the assumption that every decay began by giving its electron the endpoint energy. The spread could no longer be dismissed as simple energy loss after emission.

Sources: [1] · [4]

Was a conservation law failing—or was the account incomplete?

Energy conservation works only when all the relevant products are included. An apparently unbalanced account can mean that the law has failed. It can also mean that something has escaped the measurement. The beta spectrum did not, by itself, choose between those explanations.

Niels Bohr considered whether energy conservation might have a restricted role in these processes. Wolfgang Pauli preferred to keep the conservation laws and add an electrically neutral particle. Such a particle could leave without the conspicuous ionization track of a charged electron.

Pauli’s move created a new obligation. The particle could not remain an adjustable amount of missing energy assigned to each event. It needed definite physical properties and, eventually, interactions that another experiment could identify.

Sources: [5] · [2] · [3]

Three products can share energy in many ways

In the modern description of ordinary beta-minus decay, the products include an electron, an electron antineutrino and a recoiling daughter nucleus. There are now more ways to divide the available energy while conserving both energy and momentum. The electron can take a large share in one event and a smaller share in another.

The equation below uses a simple approximation: the neutrino mass is neglected, and the available energy is defined for one nuclear decay branch. The daughter’s recoil is usually small, but it is included so that the accounting is clear.

This explains why a continuous electron spectrum is possible. It does not, on its own, calculate the shape of that spectrum. That requires the probabilities of different final states and the weak-interaction dynamics, as developed in beta-decay theory.

Q=T_e+E_{\bar\nu}+T_{\mathrm{recoil}}

For one decay branch, with neutrino mass neglected: the available energy Q is divided between electron kinetic energy, antineutrino energy and daughter recoil. This is an energy balance, not a formula for the event rate.

Sources: [5] · [6]

An explanation became something to look for

The calorimeter had helped rule out one mundane explanation. Pauli supplied a hypothesis for what the measurement was missing. A theory of beta decay then made the hypothesis useful for calculating other processes. Those were separate advances, carried out by different people.

The decisive next step was to seek an interaction caused by the emitted particles. Cowan, Reines and their colleagues eventually used a reactor source and a pair of closely timed signals to identify antineutrino interactions. That 1956 result did a different job from the earlier energy measurements.

The story begins with an unusually good question about an imperfect account of energy. Its strength comes from the changing tests: first ask whether the electron lost energy in the apparatus, then ask whether another particle can carry it, then build an experiment that can catch that particle interacting.

Sources: [7] · [5]

Primary sources & revision

First published and source-checked on 9 October 2026. Original explanatory text and diagrams by Sandbox Physics. Illustrations are schematic; no experimental event records are reproduced here. This is an independent educational publication, not an official Nobel or experiment collaboration publication.

Continue the story

Pauli’s letter: proposing a particle no one had seen Cowan and Reines: finding the neutrino in two flashes Neutrinos have mass. Why does oscillation not tell us how much?