Concept

Mossbauer effect — where it appears

The emission and absorption of gamma rays by nuclei bound in a crystal without any loss of energy to recoil, because the whole crystal takes up the momentum. The lines are so narrow that they resolve shifts of a part in 10¹⁵.

Named by 3 essays across one field — each of them below, with the objects they name alongside it.

How much slower a warm atom ticks. The fraction by which a free atom's clock runs slow on average because of its thermal motion, (3/2)kT/mc², against temperature on logarithmic axes, for five atoms used in clocks or tests of relativity: hydrogen 4.1 × 10⁻¹¹ at 300 K, aluminium ion 1.5 × 10⁻¹² at 300 K, strontium 4.8 × 10⁻¹³ at 300 K, caesium 3.1 × 10⁻¹³ at 300 K, mercury ion 2.1 × 10⁻¹³ at 300 K. Every line falls in proportion to the temperature and to one over the mass. A clock aiming at one part in 10¹⁸ must hold strontium below 0.63 mK and hydrogen below 7.3 µK — or know their temperature well enough to correct for it. Pound and Rebka's gravitational shift across 22.5 m, 2.46 × 10⁻¹⁵, is marked for scale.

The clock that runs slow because it is warm

Every atom in a gas is moving, and every moving clock runs slow. The first-order Doppler shifts of a hot gas cancel on average; the time dilation does not, because it has only one sign. A hydrogen maser at 40 °C runs slow by four parts in a hundred billion from this alone, a nucleus in a crystal runs slow even at absolute zero, and the first test of gravitational time dilation nearly drowned in a one-kelvin temperature difference.

relativity · Time dilation
An atom's line and a nucleus's, on one scale. Three energies, in electronvolts on a logarithmic scale, for an atomic transition, sodium's yellow line, and a nuclear one, the 14.4 keV gamma ray of iron-57: the amount by which emission and absorption by free atoms miss each other because of recoil, E²/Mc²; the natural width of the line, ħ over the excited state's lifetime; and the Doppler width from thermal motion at room temperature. For sodium the recoil miss is 2.1·10⁻¹⁰ eV, 197 times smaller than the natural width, so an atom absorbs its neighbour's light. For iron-57 it is 0.00392 eV, 839,015 times its natural width of 4.67·10⁻⁹ eV. The recoil grows as the square of the transition energy and the width does not, so a gamma ray emitted by a free nucleus misses an identical nucleus by nearly a million line widths.

The gamma ray its own kind cannot absorb

An atom can absorb the light another atom of its kind emits; that is how a sodium flame casts a shadow in sodium light. A nucleus cannot. When it emits a gamma ray it loses the photon's energy as mass, and the photon's momentum kicks it backward, taking a small share of the energy with it; to absorb the same gamma ray, an identical nucleus would need that share added rather than removed. For iron-57's 14.4 keV line the shortfall is nearly a million times the line's own width. In 1958 Rudolf Mössbauer found that a nucleus held in a crystal can hand its recoil to the whole crystal and lose almost nothing — and the resulting line is so sharp that it could weigh the fall of light down a twenty-two metre tower.

relativity · Mass-energy
A shift no larger than the line it is measured on. The fractional frequency shift between a gamma-ray source at the centre of a rotor and an absorber on it, ω²r²/2c², against rotation rate, for absorbers 4, 6.6 and 10 cm from the axis. At 500 turns a second an absorber 6.6 cm out is shifted by 2.4 × 10⁻¹³, comparable to the natural width of iron-57's 14.4 keV line, 3.3 × 10⁻¹³. Only the Mössbauer effect, discovered two years earlier, made a line that sharp available, and the rotor experiments followed at once.

The rim a rotor slows

Put a source of gamma rays at the centre of a spinning rotor and an absorber on its rim, and the absorber stops absorbing as the rotor speeds up. In the laboratory the reason is that the rim is moving and its clock runs slow. On the rotor, where the rim is at rest, the reason is that the rim sits lower in a centrifugal potential and its clock runs slow for the reason clocks run slow lower down in gravity. Both give ω²r²/2c², a few parts in ten trillion, and the experiment in 1960 measured it.

relativity · Accelerated frames

Named alongside it

The objects these essays reach for when they reach for this one.

Gravitational redshiftTime dilationAtomic clockCentrifugal potentialDebye modelDoppler effectEquivalence principleHeat capacityMass-energyMaxwell–Boltzmann distributionNatural linewidthRecoil

All concepts