Helium Flash and Horizontal Branch

A star between roughly $0.8$ and $2\,M_\odot$ leaves the main sequence carrying a hydrogen-exhausted helium core that cannot yet burn. A hydrogen shell rains fresh ash onto that core, which contracts and heats — but before it reaches the $\sim10^8\,\mathrm{K}$ needed to ignite the triple-alpha reaction, its density climbs so high that the electrons go degenerate. In a degenerate gas the pressure is fixed by the electron momentum distribution and barely notices temperature, so the core is held up cold. This decouples the ordinary thermostat: normally an over-heated region expands, cools, and throttles its own burning, but a degenerate core cannot expand in response to heat. When helium finally ignites, the energy released raises the temperature without raising the pressure. A hotter core burns faster, which heats it further, which burns faster still — and because the triple-alpha rate near ignition scales as roughly $T^{40}$ , this feedback becomes a thermonuclear runaway: the helium flash. It self-terminates only when the temperature climbs high enough ( $T\gtrsim3\times10^8\,\mathrm{K}$ ) that thermal pressure overtakes degeneracy pressure; the gas "un-degenerates", expands, cools, and burning becomes stable. Remarkably, almost none of the flash luminosity escapes — it is spent lifting the core out of degeneracy against gravity. Because ignition happens at a nearly universal core mass, $M_c\approx0.45\,M_\odot$ , the tip of the red giant branch has a nearly fixed luminosity — one of the cleanest standard candles in astrophysics. After the flash the star settles onto the horizontal branch: quiescent core-helium burning at $L\approx50\,L_\odot$ , spread across a range of colours set by the surviving hydrogen envelope. Where the branch crosses the instability strip it produces the RR Lyrae variables; in metal-rich populations it collapses into the red clump.

Key takeaways

In stars below about $2.2\,M_\odot$ , the helium core becomes electron-degenerate before it is hot enough to ignite, so the eventual triple-alpha ignition is a thermal runaway — the helium flash — not a gentle turn-on.
Ignition occurs at a nearly universal core mass $M_c\approx0.45\,M_\odot$ and central temperature $T\approx10^8\,\mathrm{K}$ , which is why the tip of the red giant branch is a standard candle ( $M_{\rm bol}\approx-3.6$ ).
At peak the local energy generation reaches $\sim10^{11}\,L_\odot$ , yet the surface never brightens: the energy goes into lifting degeneracy, and the star actually dims as it leaves the RGB tip.
The runaway ends when thermal pressure overtakes degeneracy pressure at $T\gtrsim3\times10^8\,\mathrm{K}$ ; the core expands, cools, and settles into stable core-helium burning on the horizontal branch at $L\approx50\,L_\odot$ .
Colour along the horizontal branch is set mainly by the residual hydrogen envelope mass (RGB mass loss) and metallicity; the instability-strip crossing yields the RR Lyrae distance indicators, and the metal-rich analogue is the red clump.
Core-helium burning lasts $\sim10^8\,\mathrm{yr}$ , about $1\%$ of the main-sequence lifetime, ending when the core is helium-exhausted and the star ascends the asymptotic giant branch.
Key idea: Electron degeneracy breaks the thermostat that regulates nuclear burning — because $P$ is decoupled from $T$ , ignition of the steeply temperature-sensitive triple-alpha process becomes a runaway (the helium flash) rather than a gentle turn-on, self-terminating only when thermal pressure lifts degeneracy and places the star on the horizontal branch.

The core insight

Electron degeneracy breaks the thermostat that normally regulates nuclear burning: because $P$ is decoupled from $T$ , ignition of the strongly temperature-sensitive triple-alpha process becomes a runaway (the helium flash) rather than a gentle turn-on. The runaway self-terminates only when thermal pressure lifts degeneracy, after which stable core-helium burning places the star on the horizontal branch.

Where this lives today

The physics of the flash and the horizontal branch is working machinery across modern astrophysics. The tip-of-the-RGB (TRGB) method is now a leading route to the local Hubble constant: programs like the Carnegie-Chicago Hubble Program calibrate Type Ia supernovae off TRGB distances from Hubble and JWST, yielding an $H_0$ that sits between the Cepheid and cosmic-microwave-background values, central to the "Hubble tension" debate. Asteroseismology from Kepler and TESS has turned the red clump into a precision tool — oscillations in clump giants measure their masses, radii, and ages, letting Galactic archaeologists date the disk and bulge. RR Lyrae variables, catalogued in the millions by Gaia and OGLE, map the old halo and set the distance to the Galactic centre. And the second-parameter problem — why two clusters of identical metallicity can have utterly different horizontal-branch morphologies — remains an active puzzle, now entangled with the discovery of multiple populations and helium enrichment inside globular clusters.

The physical picture

A normal star is a self-correcting furnace: if a burning region overheats, the ideal gas expands, cools, and turns the burning down. This stellar thermostat works because pressure tracks temperature, $P\propto\rho T$ . Inside a degenerate helium core that link is severed — the pressure comes from electrons filling momentum states up to the Fermi level and depends on density alone, so adding heat pushes nothing back. When the triple-alpha reaction finally lights, the released energy cannot expand the core; it simply raises the temperature, which raises the burning rate, which raises the temperature again. The figure below traces where that runaway sends the star in the Hertzsprung–Russell diagram.

Figure

The evolutionary track through the tip of the red giant branch and onto the horizontal branch: a low-mass star climbs the RGB to L ≈ 2000 L☉ (M_bol ≈ −3.6), ignites helium in the flash, then settles on the horizontal branch at L ≈ 50 L☉ — blue at small envelope mass, red clump at large envelope mass, with RR Lyrae variables where the branch crosses the instability strip

The core mass at ignition is almost independent of the total stellar mass, because it is set by the physics of the degenerate core — a Chandrasekhar-like relation between core mass and maximum central temperature — not by the envelope. Nature ignites helium at the same $\sim0.45\,M_\odot$ core every time, which is why the RGB tip is a standard candle. The flash is really a phase transition of the core's support: it begins held up by cold, quantum electron pressure and ends held up by hot, classical gas pressure, with the nuclear energy the ticket price for that conversion — spent doing $P\,\mathrm{d}V$ work and mostly never leaving the core. The second figure plots pressure against temperature at fixed density, exposing the flat, temperature-blind degenerate branch on which the runaway lives.

Figure

Why ignition runs away: pressure versus temperature at fixed density. The degenerate electron pressure P ∝ ρ^(5/3) is flat — independent of T — so heating at ignition (T ≈ 1×10⁸ K) drives the temperature rightward at nearly constant pressure with no expansion to cool the gas. Only when the ideal-gas pressure P ∝ ρT overtakes it near T ≈ 3×10⁸ K does the core regain its thermostat, expand, and quench the flash

On the horizontal branch the luminosity is pinned by the nearly universal helium-core mass, but the surface temperature — the colour — is a lever controlled by how much hydrogen envelope survived RGB mass loss. A small envelope leaves a hot, blue HB star; a large envelope leaves a cool, red one, the red clump — the origin, with metallicity, of the whole zoo of horizontal-branch morphologies in globular clusters.

The core equations

The triple-alpha reaction fuses three helium nuclei into carbon, and near ignition its rate is ferociously temperature-sensitive — the fuse of the flash, since a tiny temperature excursion multiplies the energy release enormously.

\varepsilon_{3\alpha} \approx \varepsilon_0\,\rho^2 Y^3\,T_8^{40} \qquad (1)

where: $\varepsilon_{3\alpha}$ — energy generation rate per unit mass (erg g⁻¹ s⁻¹); $\rho$ — mass density; $Y$ — helium mass fraction; $T_8$ — temperature in units of $10^8\,\mathrm{K}$ ; $\varepsilon_0$ — constant collecting the nuclear cross-section factors. Reading it: the $T_8^{40}$ exponent — far steeper than the CNO cycle's $\sim T^{18}$ — is what turns degenerate ignition into a runaway. The core's support is set not by heat but by quantum mechanics: when electrons are packed densely enough that the exclusion principle forces them into high-momentum states, they exert a pressure that depends only on density.

P_e \approx \frac{(3\pi^2)^{2/3}}{5}\frac{\hbar^2}{m_e}\left(\frac{\rho}{\mu_e m_u}\right)^{5/3} \qquad (2)

where: $P_e$ — non-relativistic degenerate electron pressure; $\hbar$ — reduced Planck constant; $m_e$ — electron mass; $\mu_e$ — mean molecular weight per electron ( $\approx2$ for helium); $m_u$ — atomic mass unit; $\rho$ — mass density. Reading it: there is no $T$ on the right-hand side — add heat and $P_e$ does not budge, so the thermostat is gone. Whether the core is degenerate is decided by comparing the electron Fermi energy with the thermal energy scale. The Fermi energy is the kinetic energy of the most energetic filled electron state.

\varepsilon_F = \frac{\hbar^2}{2m_e}\left(3\pi^2 n_e\right)^{2/3}\gg k_B T \qquad (3)

where: $\varepsilon_F$ — electron Fermi energy; $n_e$ — electron number density ( $=\rho/\mu_e m_u$ ); $k_B T$ — thermal energy scale. Reading it: degeneracy dominates when $\varepsilon_F\gg k_B T$ — high density, modest temperature, exactly the RGB core — and the flash ends when rising $T$ pushes $k_B T$ up to $\varepsilon_F$ . The maximum central temperature a contracting degenerate core can reach grows with its mass, so demanding that it reach the triple-alpha threshold fixes the core mass at ignition — nearly the same for every low-mass star.

T_c^{\max}\ \text{increases with}\ M_c\ \Rightarrow\ T_c=10^8\,\mathrm{K}\ \text{at}\ M_c^{\rm ign}\approx 0.45\,M_\odot \qquad (4)

where: $T_c^{\max}$ — peak central temperature reached during contraction; $M_c$ — helium-core mass; $M_c^{\rm ign}$ — core mass at flash ignition. Reading it: a threshold fixed by nuclear physics plus a monotonic $T_c^{\max}(M_c)$ makes the ignition core mass nearly universal, independent of the envelope. The hydrogen-shell luminosity of a red giant is an extremely steep function of core mass, so a universal ignition mass produces a universal tip luminosity — the standard candle.

L \propto M_c^{\,7}\ \Rightarrow\ L_{\rm tip}\approx 2\times10^3\,L_\odot,\quad M_{\rm bol}^{\rm tip}\approx -3.6 \qquad (5)

where: $L_{\rm tip}$ — luminosity at the RGB tip / flash onset; $M_{\rm bol}^{\rm tip}$ — absolute bolometric magnitude at the tip. Reading it: fixing $M_c^{\rm ign}$ fixes $L_{\rm tip}$ , so the sharp cutoff atop a red-giant luminosity function is a distance indicator. After the flash the star burns helium quiescently in a non-degenerate convective core plus a hydrogen shell, sitting at a low, tightly clustered luminosity.

L_{\rm HB}\approx 50\,L_\odot,\qquad M_V\approx +0.6\ (\text{RR Lyrae}) \qquad (6)

where: $L_{\rm HB}$ — core-helium-burning luminosity on the horizontal branch; $M_V$ — HB/RR Lyrae absolute visual magnitude (weakly metallicity dependent). Reading it: luminosity is pinned by the fixed core mass; only the colour varies along the branch. The duration of core-helium burning follows from dividing the available nuclear energy reservoir by the burning luminosity.

t_{\rm He}\approx\frac{q\,f\,M_c\,c^2}{L_{\rm HB}}\sim 10^8\,\mathrm{yr} \qquad (7)

where: $t_{\rm He}$ — horizontal-branch lifetime; $q\approx7\times10^{-4}$ — fraction of rest mass released turning helium into carbon; $f$ — fraction of core helium actually burned; $M_c$ — core mass $\approx0.45\,M_\odot$ . Reading it: a small yield $q$ and a fixed low luminosity give a phase of $\sim10^8\,\mathrm{yr}$ , about $1\%$ of the main-sequence lifetime.

Derivation AThe thermal runaway condition under degeneracy

The goal is to show, by perturbing a burning region and asking whether a small temperature excursion damps or amplifies, that ignition in a degenerate gas is unstable and produces a flash, whereas the same ignition in an ideal gas is self-regulating. The whole difference turns on a single logarithmic derivative, $(\partial\ln P/\partial\ln T)_\rho$ — how strongly pressure responds to temperature at fixed density — which collapses to zero in a degenerate gas.

Step 1. Write the local energy balance for a blob of gas as $c_P\,\mathrm{d}T/\mathrm{d}t = \varepsilon_{3\alpha}(\rho,T) - (\text{losses})$ , so the blob heats whenever nuclear generation outruns radiative and expansion losses. Why: this isolates the competition between the runaway source term and the stabilizing feedback that acts through expansion.

Step 2. Consider an ideal gas, $P\propto\rho T$ . A temperature rise raises the pressure, the region expands against its surroundings, the density drops, and both the cooling from expansion and the reduced $\varepsilon_{3\alpha}\propto\rho^2$ throttle the burning. Why: this is the stellar thermostat — a positive temperature perturbation feeds back negatively, so ignition is stable and gentle.

Step 3. Now consider a degenerate gas, where $P\approx P_e(\rho)$ is nearly independent of $T$ so that $(\partial\ln P/\partial\ln T)_\rho\to0$ . A temperature rise produces essentially no pressure rise, hence no expansion, no density drop, and no compensating cooling. Why: the stabilizing branch of Step 2 is switched off entirely — the heat has nowhere to go except into more heat.

Step 4. Close the loop with the reaction's temperature sensitivity: $\delta T\uparrow\Rightarrow\varepsilon_{3\alpha}\propto T^{40}\uparrow\uparrow\Rightarrow\delta T\uparrow\uparrow$ . With a $T^{40}$ dependence, even a percent-level warming multiplies the energy release, and the temperature accelerates catastrophically. Why: this positive feedback with no damping is precisely the helium flash.

Step 5. The runaway self-terminates as $T$ climbs toward $\sim3\times10^8\,\mathrm{K}$ and $k_B T$ approaches $\varepsilon_F$ : the gas becomes non-degenerate, $P$ recovers its $T$ -dependence, the core finally expands and cools, and the ideal-gas thermostat is restored. Why: degeneracy both causes the flash and sets its own off-switch, which is why the event is violent but not disruptive.

Go deeper · Derivation B · 1 minWhy the tip of the red giant branch is a standard candle

Derivation BWhy the tip of the red giant branch is a standard candle

The goal is to derive the near-constancy of the ignition core mass, and hence of the tip luminosity, from the structure of a degenerate, self-gravitating helium core that must reach the triple-alpha threshold before it can ignite. The reasoning connects three fixed things — the polytropic mass-radius law of a degenerate sphere, the monotonic rise of its attainable central temperature with mass, and the nuclear ignition temperature set by laboratory physics — into one prediction: every low-mass star lights helium at almost the same $0.45\,M_\odot$ , and therefore at almost the same luminosity.

Step 1. Model the core as a non-relativistic degenerate polytrope with $P\propto\rho^{5/3}$ , for which hydrostatic equilibrium gives the inverted mass-radius relation $R_c\propto M_c^{-1/3}$ — a more massive core is smaller and denser. Why: this compactness is what allows a growing core to reach high central temperatures without help from the envelope.

Step 2. As the hydrogen shell dumps ash onto the core, $M_c$ grows, the core contracts, and virial heating raises the central temperature; the maximum $T_c$ attainable is a monotonically increasing function of $M_c$ alone — a degenerate analogue of the Chandrasekhar mass-temperature relation. Why: it makes ignition a question purely of core mass, decoupled from the total stellar mass.

Step 3. Impose the ignition condition $T_c\to10^8\,\mathrm{K}$ , the triple-alpha threshold fixed by nuclear physics, which the monotonic $T_c^{\max}(M_c)$ satisfies at $M_c\approx0.45\,M_\odot$ for every low-mass star. Why: a fixed threshold temperature plus a monotonic relation yields a fixed ignition mass — the heart of the standard candle.

Step 4. Convert to luminosity through the red-giant shell relation $L\propto M_c^{\,7}$ , an extremely steep dependence, so a nearly constant $M_c^{\rm ign}$ pins $L_{\rm tip}$ to a nearly constant value, $M_{\rm bol}^{\rm tip}\approx-3.6$ . Why: the tip appears at the same luminosity in every old population, calibrating the tip-of-the-RGB distance method, especially in the $I$ band where the tip is flattest.

Go deeper · Derivation C · 1 minEstimate of the horizontal-branch lifetime

Derivation CEstimate of the horizontal-branch lifetime

The goal is an order-of-magnitude timescale for the core-helium-burning phase, obtained by dividing the nuclear energy available in the helium core by the rate at which the star radiates it on the horizontal branch. The estimate needs only three inputs — the efficiency of helium-to-carbon fusion, the core mass, and the horizontal-branch luminosity — and it reproduces the canonical $\sim10^8\,\mathrm{yr}$ .

Step 1. Write the reservoir as $E_{\rm nuc}=q\,f\,M_c\,c^2$ with efficiency $q\approx7\times10^{-4}$ (the rest-mass fraction released when helium becomes carbon), burned fraction $f\sim1$ , and $M_c\approx0.45\,M_\odot$ . Why: only a tiny sliver of the rest energy is available, which is exactly why the phase is short despite the large mass.

Step 2. Insert numbers: $E_{\rm nuc}\approx(7\times10^{-4})(0.45)(2\times10^{33}\,\mathrm{g})(9\times10^{20}\,\mathrm{cm^2\,s^{-2}})\approx6\times10^{50}\,\mathrm{erg}$ . Why: this fixes the total energy the branch has to spend, in cgs units where an erg is a g cm² s⁻².

Step 3. Divide by the luminosity, $t_{\rm He}=E_{\rm nuc}/L_{\rm HB}\approx(6\times10^{50})/(50\times3.8\times10^{33}\,\mathrm{erg\,s^{-1}})\approx3\times10^{15}\,\mathrm{s}\approx10^8\,\mathrm{yr}$ . Why: dividing a reservoir by a rate gives a lifetime, and $10^8\,\mathrm{yr}$ is about $1\%$ of a solar-type main-sequence lifetime — a short, luminous, sharply defined stage.

Limiting cases

How the ignition mode and its outcome depend on stellar mass and core conditions:

Case	Result	Meaning
Low mass ( $M\lesssim2.2\,M_\odot$ )	$M_c\approx0.45\,M_\odot$ , degenerate ignition	Core is degenerate at ignition → helium flash. Near-universal core mass gives a standard-candle RGB tip.
Intermediate mass ( $2.2\lesssim M/M_\odot\lesssim8$ )	Non-degenerate core, quiescent ignition	Core stays hot enough to ignite helium gently before degeneracy sets in — no flash. These stars populate the blue loops and secondary clump.
Very low mass ( $M\lesssim0.5\,M_\odot$ )	$T_c^{\max}<10^8\,\mathrm{K}$	Core never reaches ignition; the star ends as a helium white dwarf (in isolation only over more than a Hubble time, or via binary stripping).
Degeneracy lifted	$T\gtrsim3\times10^8\,\mathrm{K}\Rightarrow P_{\rm gas}\gtrsim P_{\rm deg}$	Runaway self-terminates: thermal pressure restores expansion feedback and burning becomes stable on the HB.

Worked examples

1 · Is the pre-flash helium core actually degenerate?

Just before the flash the helium core has central density $\rho_c\approx1\times10^6\,\mathrm{g\,cm^{-3}}$ and temperature $T_c\approx1\times10^8\,\mathrm{K}$ , with $\mu_e=2$ . Show it is strongly degenerate by comparing the Fermi energy to the thermal energy.

Step 1. Electron number density: $n_e=\rho/(\mu_e m_u)=(10^6\,\mathrm{g\,cm^{-3}})/(2\times1.66\times10^{-24}\,\mathrm{g})=3.0\times10^{29}\,\mathrm{cm^{-3}}=\mathbf{3.0\times10^{35}\,m^{-3}}$ . This is the density of electrons whose momentum states must be filled.

Step 2. Fermi energy from equation (3): $\varepsilon_F=\dfrac{\hbar^2}{2m_e}(3\pi^2 n_e)^{2/3}$ . With $3\pi^2 n_e=8.9\times10^{36}\,\mathrm{m^{-3}}$ and $\hbar^2/2m_e=6.1\times10^{-39}\,\mathrm{J\,m^2}$ , this gives $\varepsilon_F\approx2.6\times10^{-14}\,\mathrm{J}=\mathbf{164\ keV}$ .

Step 3. Thermal scale: $k_B T=(1.38\times10^{-23})(10^8)=1.38\times10^{-15}\,\mathrm{J}=\mathbf{8.6\ keV}$ . The ratio is $\varepsilon_F/k_B T\approx\mathbf{19}$ , so $\varepsilon_F\gg k_B T$ and the core is strongly degenerate — the thermostat is disarmed and ignition will run away. (At $\varepsilon_F\approx164\,\mathrm{keV}$ the electrons are mildly relativistic, so the non-relativistic formula slightly overestimates $\varepsilon_F$ , but the conclusion holds.)

2 · The horizontal-branch lifetime from first principles

Estimate how long core-helium burning lasts, taking $M_c=0.45\,M_\odot$ , efficiency $q=7\times10^{-4}$ , burned fraction $f=1$ , and $L_{\rm HB}=50\,L_\odot$ .

Step 1. Nuclear reservoir: $E_{\rm nuc}=q\,f\,M_c c^2=(7\times10^{-4})(0.45\times1.99\times10^{33}\,\mathrm{g})(9\times10^{20}\,\mathrm{cm^2\,s^{-2}})\approx\mathbf{5.6\times10^{50}\ erg}$ — only $7\times10^{-4}$ of the burned mass emerges as energy.

Step 2. Luminosity in cgs: $L_{\rm HB}=50\times3.83\times10^{33}\,\mathrm{erg\,s^{-1}}=1.9\times10^{35}\,\mathrm{erg\,s^{-1}}$ .

Step 3. Lifetime: $t_{\rm He}=E_{\rm nuc}/L_{\rm HB}=(5.6\times10^{50})/(1.9\times10^{35})\approx3.0\times10^{15}\,\mathrm{s}\approx\mathbf{9\times10^{7}\ yr\approx10^{8}\ yr}$ . Compared to the $\sim10^{10}\,\mathrm{yr}$ main sequence, the horizontal branch is a brief $\sim1\%$ episode — which is why, for every horizontal-branch star, a cluster holds roughly a hundred main-sequence stars.

3 · A tip-of-the-red-giant-branch distance

The luminosity function of resolved red giants in a nearby galaxy shows its sharp bright cutoff at apparent bolometric magnitude $m_{\rm bol}=24.4$ . Using the standard-candle value $M_{\rm bol}^{\rm tip}=-3.6$ , find the distance.

Step 1. Distance modulus: $\mu=m_{\rm bol}-M_{\rm bol}^{\rm tip}=24.4-(-3.6)=\mathbf{28.0}$ , resting on the fixed tip magnitude that degenerate-ignition physics makes nearly universal.

Step 2. Convert to distance: $d=10^{(\mu+5)/5}\,\mathrm{pc}=10^{(28.0+5)/5}=10^{6.6}\,\mathrm{pc}\approx\mathbf{4.0\ Mpc}$ . The sharp TRGB cutoff yields distances to galaxies out to $\sim20\,\mathrm{Mpc}$ — an independent rung on the distance ladder.

Physics of the portrait

The helium flash is a place where nuclear physics, quantum statistics, and stellar structure collide, and it could only be understood once stars could be computed rather than merely observed. That step was taken by the man who turned stellar evolution into a numerical science.

Origins · 1 minMartin Schwarzschild and how this was discovered

Martin Schwarzschild

1912–1997 · first numerical models of the helium flash and red-giant evolution

Martin Schwarzschild — son of the physicist Karl Schwarzschild — converted stellar structure from an analytic art into a computational science. Working at Princeton in the 1950s and 1960s, often with Richard Härm, he used some of the earliest electronic computers to follow stars past the main sequence, where the equations are too stiff for pen and paper. It was in his and Härm's calculations of Population II red giants that the degenerate helium core, its runaway ignition, and the star's settling onto the horizontal branch first emerged from the numbers rather than from guesswork. The physical ingredients were assembled around him: Fowler had shown in 1926 that electron degeneracy pressure supports white dwarfs; Öpik and Salpeter had established the triple-alpha reaction, with Hoyle predicting the crucial carbon resonance; and Sandage's colour-magnitude diagrams had revealed the horizontal branch as a real, populated locus needing explanation. Schwarzschild's models tied these threads together — the picture that still underlies every reading of an old cluster's colour-magnitude diagram today.

History

1926

— Fowler explains white dwarfs by electron degeneracy pressure

1952

— Öpik and Salpeter establish the triple-alpha reaction rate

1953

— Sandage's colour-magnitude diagram of M3 delineates the horizontal branch

1955

— Hoyle and Schwarzschild model the ascent of the red giant branch

1962

— Schwarzschild and Härm compute the degenerate helium-core flash

1968

— Iben's evolutionary tracks pin down the horizontal branch and red clump

Going further

References. Carroll and Ostlie, An Introduction to Modern Astrophysics, 2nd ed., Ch. 13 (Post-Main-Sequence Stellar Evolution) · Kippenhahn, Weigert and Weiss, Stellar Structure and Evolution, 2nd ed., Ch. 33–34 · Hansen, Kawaler and Trimble, Stellar Interiors, 2nd ed. · Salaris and Cassisi, Evolution of Stars and Stellar Populations · Iben, I. 1974, ARA&A 12, 215 (post-main-sequence evolution).

Provenance. Numbers are standard textbook values: ignition core mass $\approx0.45\,M_\odot$ , $T_{\rm ign}\approx10^8\,\mathrm{K}$ , $\varepsilon_{3\alpha}\propto\rho^2 T^{40}$ , TRGB $M_{\rm bol}\approx-3.6$ , HB $L\approx50\,L_\odot$ , RR Lyrae $M_V\approx+0.6$ , HB lifetime $\sim10^8\,\mathrm{yr}$ — all from Carroll and Ostlie Ch. 13 and Kippenhahn, Weigert and Weiss Ch. 33–34. The worked examples are recomputed here: the pre-flash core has $\varepsilon_F/k_B T\approx19$ (strongly degenerate) at $\rho_c=10^6\,\mathrm{g\,cm^{-3}}$ , $T_c=10^8\,\mathrm{K}$ ; the HB reservoir $E_{\rm nuc}\approx5.6\times10^{50}\,\mathrm{erg}$ divided by $L_{\rm HB}=50\,L_\odot$ gives $t_{\rm He}\approx9\times10^7\,\mathrm{yr}$ ; and a TRGB cutoff at $m_{\rm bol}=24.4$ gives $\mu=28.0$ , $d=4.0\,\mathrm{Mpc}$ . A verifier should confirm: (i) the degenerate-pressure and degeneracy-criterion arithmetic; (ii) the runaway condition from the vanishing $(\partial\ln P/\partial\ln T)_\rho$ ; (iii) the HB-lifetime energy estimate; and (iv) that the $T^{40}$ exponent follows from stacked Gamow factors near $T_8\approx1$ .

Prerequisites. Helium Burning and the Triple-Alpha Process · White Dwarfs and Electron Degeneracy Pressure · Post-Main-Sequence Evolution and the Red Giant Branch · Nuclear Reaction Rates and the Gamow Peak. Leads to. Asymptotic Giant Branch and Thermal Pulses · The Chandrasekhar Mass · The Hertzsprung-Russell Diagram.

Ask the tutor: Ask me why a $0.45\,M_\odot$ degenerate core ignites in a runaway while a $5\,M_\odot$ star lights helium gently, or to work a TRGB distance together.

Is the helium flash an explosion that could destroy the star?

No. Although the local energy generation at peak reaches an astonishing $\sim10^{11}\,L_\odot$ — briefly rivalling a whole galaxy — the event is entirely contained. The energy is not radiated; it is spent doing work against gravity to expand the core and lift it out of degeneracy, and what little heat is left over must diffuse out through the enormous overlying envelope over a thermal timescale. The star's surface luminosity barely changes during the flash, and afterward the star settles quietly onto the horizontal branch, slightly fainter than at the RGB tip. "Flash" refers to the runaway of the internal burning rate, not to any external outburst.

Why is the ignition core mass nearly the same for every low-mass star?

Because it is set by the core's own physics, not the envelope's. A degenerate helium core is supported by electron pressure that depends only on density, and its maximum attainable central temperature is a monotonic function of its mass alone — heavier core, hotter centre. Helium ignites the instant that temperature reaches the triple-alpha threshold near $10^8\,\mathrm{K}$ , and since that threshold is fixed by nuclear physics, the core mass at ignition is fixed too, at $\approx0.45\,M_\odot$ , regardless of whether the surrounding star weighs $0.9$ or $2\,M_\odot$ . This near-universality is exactly what makes the RGB tip a standard candle.

What is the second-parameter problem?

Metallicity is the "first parameter" that broadly controls horizontal-branch colour: metal-poor clusters tend to have bluer branches. But astronomers found pairs of clusters with essentially identical metallicity yet strikingly different branch morphologies — one blue, one red. Some additional "second parameter" must be at work. The leading candidates are age, helium abundance, and, most directly, the amount of mass a star loses on the red giant branch, which sets how much hydrogen envelope survives and therefore whether the star lands blue or red. The problem is now tangled up with the discovery that globular clusters host multiple populations with different helium content, and it remains unresolved.

What are RR Lyrae stars and the red clump, and how do they relate?

Both are horizontal-branch stars — the same core-helium-burning phase — seen under different conditions. Where the horizontal branch crosses the classical instability strip (around $6000$ – $7500\,\mathrm{K}$ ), stars pulsate with periods near half a day: these are the RR Lyrae variables, superb standard candles ( $M_V\approx+0.6$ ) for old populations. The red clump is the metal-rich version of the branch: in metal-rich populations the branch collapses into a compact clump of cool stars with a tight absolute magnitude ( $M_K\approx-1.6$ ) that also serves as a distance indicator.

Does the Sun undergo a helium flash?

Yes. At $1\,M_\odot$ the Sun is comfortably in the low-mass regime that develops a degenerate helium core on the red giant branch, so in about $6$ – $7$ billion years it will climb to the RGB tip, ignite helium in a flash, and settle briefly onto the horizontal branch (as a red-clump star, given its metallicity) before ascending the asymptotic giant branch and ending as a carbon-oxygen white dwarf. The flash will be invisible from outside — the Sun will simply dim off the RGB tip, not brighten.

What ends the horizontal-branch phase?

Helium exhaustion in the core. After $\sim10^8\,\mathrm{yr}$ the central helium is used up, leaving an inert carbon-oxygen core that — like the earlier helium core — contracts and becomes degenerate. Helium burning then continues in a shell around it while hydrogen burns in a shell farther out, and the star expands and climbs the asymptotic giant branch. The horizontal branch is thus a temporary equilibrium: it lasts exactly as long as the core has helium to burn, and it ends by handing the star to the next, more luminous and more unstable, evolutionary stage.