Stellar Pulsation and the Instability Strip

Some stars vary in brightness because they pulsate: the whole star expands and contracts, brightening as it heats and dims as it cools. The oscillation is a standing sound wave, its period fixed by the mean density, and it is sustained against dissipation by an opacity valve in a partial-ionization layer of the envelope. Only stars occupying a narrow, nearly vertical strip of the Hertzsprung–Russell diagram pulsate this way, and their period–luminosity relation makes them the measuring rods of extragalactic distance.¹

The period–mean-density relation

A radial pulsation is a standing acoustic wave in the star. Its period is the time a sound wave takes to cross the star and return, so it depends on the sound speed and the radius. Model the star crudely as a sphere of uniform density $\bar{\rho }$ oscillating adiabatically. A displaced shell feels a restoring pressure whose adiabatic response is set by the first adiabatic exponent ${\Gamma }_1$, and the linearized equation of motion has the fundamental period

For ${\Gamma }_1=5/3$ the factor $3{\Gamma }_1-4=1$, and this reduces to $\Pi =\sqrt{3\pi /(G\bar{\rho })}$. The essential content is the scaling

the period–mean-density relation. The pulsation constant $Q$ is nearly the same for all stars pulsating in the fundamental mode, about $0.04\text{days}$ for classical Cepheids. Denser stars pulsate faster; the tenuous, luminous supergiants have the longest periods. Note the requirement ${\Gamma }_1>4/3$ for a real period: below it the restoring force vanishes and the star is dynamically unstable rather than oscillatory.

The fundamental mode has the whole envelope moving in phase, with the center fixed and the surface at maximum displacement. Overtones have one or more internal nodes where the gas stays still while layers on either side move oppositely, and they have shorter periods. Most Cepheids pulsate in the fundamental; many RR Lyrae stars pulsate in an overtone, giving the double-peaked period distribution seen in globular clusters.

The kappa mechanism

A pulsation left to itself would damp as sound waves dissipate. To persist, the star must do net work on the oscillation over each cycle, acting as a heat engine that absorbs heat when compressed and releases it when expanded. Ordinary stellar material does the opposite. Compression raises the temperature, and in the Kramers regime the opacity $\kappa \propto \rho T^{-7/2}$ falls with temperature, so a compressed layer becomes more transparent and lets heat leak out exactly when the engine would need to retain it. Across most of a star, pulsations are damped.

The exception is a partial-ionization zone. Where a species is half-ionized, compression does not raise the temperature much, because the energy goes into further ionization rather than into heating the gas. With the temperature nearly fixed, the density increase makes the opacity rise on compression. The layer dams the outward flux while compressed, storing heat, and releases it on expansion. This is the kappa mechanism (aided by the associated gamma mechanism, the same energy diversion into ionization). The dominant valve in classical pulsators is the second helium ionization zone, ${\text{He}}^{+}\to {\text{He}}^{2+}$, near $4\times 10^4\text{K}$.

The instability strip

The kappa mechanism drives a star only when the helium ionization zone sits at the right depth. Too hot, and the zone lies too near the surface, where it contains too little mass to store significant heat, so the driving is feeble. Too cool, and the outer envelope becomes convective; convection carries the energy and short-circuits the valve, quenching the pulsation. Between these limits the zone lies at a depth where its heat capacity is large and radiation still controls the flux, and the star pulsates. The favorable range is nearly independent of luminosity, so it defines an almost vertical band in the H-R diagram, the instability strip, tilted slightly because the cool edge depends weakly on luminosity.

The strip is populated by different stars at different luminosities:

• Classical Cepheids at the luminous top, massive stars ($3$–$20M_{\odot }$) crossing the strip on blue loops during core helium burning, with periods of one to a hundred days.
• RR Lyrae stars where the strip meets the horizontal branch, old low-mass stars of about $0.6M_{\odot }$ burning helium in the core, with periods near half a day.
• Mira variables, cool luminous stars on the asymptotic giant branch pulsating in the fundamental radial mode with periods of hundreds of days and large amplitudes.

Light and velocity curves

A pulsating star is observed through its light curve and its radial-velocity curve. The luminosity varies mainly with the surface temperature, which peaks slightly after the star has passed through minimum radius and is expanding. The radial velocity, measured from the Doppler shift of spectral lines, tracks the velocity of the surface and is nearly the time derivative of the radius. Maximum brightness therefore lags the epoch of maximum contraction, and the light and velocity curves are offset by roughly a quarter of a period. For Cepheids the light curve is characteristically asymmetric, rising fast and declining slowly.

The period–luminosity relation

More luminous Cepheids are larger and less dense, so by the period–mean-density relation they pulsate more slowly. This links period directly to luminosity: the period–luminosity relation, or Leavitt law,

with $a\approx -2.8$ and a small scatter set by the finite width of the instability strip. Because the period is measured from the light curve alone, independent of distance, the relation delivers the absolute magnitude, and comparison with the apparent magnitude gives the distance modulus. Cepheids are luminous enough to be resolved in galaxies tens of megaparsecs away, making them the rung that calibrates the extragalactic scale in the cosmic distance ladder.

Pulsation converts a star's structure into an observable clock. The period fixes the mean density and, through the instability-strip physics, the luminosity, and the brightness variation is visible across intergalactic distances. The same instability strip is crossed by massive stars on blue loops, treated in the evolution of massive stars, and by horizontal-branch stars after the helium flash, treated in post-main-sequence evolution of low-mass stars.