Subgiant Star

What happens inside

The main sequence ends when central hydrogen runs out. For a star like the Sun, this happens after about 10 Gyr of steady fusion via the proton-proton chain. The transition is gradual: hydrogen abundance in the core drops, and to maintain pressure against gravity the core must contract and heat. By the time the central hydrogen mass fraction drops below ≈ 0.05, fusion in the core itself is fading, and the star begins to evolve out of the main-sequence band on the HR diagram.

Two things then happen in quick succession. First, a thin shell of hydrogen-rich material just outside the now-helium-rich core reaches ignition temperature (T ≈ 1.5 × 10⁷ K) and starts to fuse. This shell burning is qualitatively different from the previous core burning because it is geometrically thin and intrinsically unstable — even small perturbations in the burning rate translate into substantial luminosity changes. Second, the inert helium core, with no nuclear power source of its own, contracts under gravity. As it contracts it heats up, which feeds back into the shell burning rate and makes the shell burn faster.

The result is the subgiant phase: an inert helium core surrounded by a thin hydrogen-burning shell, surrounded by an expanding envelope.

The mirror principle and why the envelope expands

You might expect that since the core is shrinking, the whole star would shrink. The opposite happens. This is the mirror principle of stellar structure: when a thin burning shell is sandwiched between a contracting core and an envelope, the envelope expands.

The origin of this counter-intuitive behaviour is in the equation of hydrostatic equilibrium evaluated across the shell. Inside a thin shell with mass M_sh and radius R_sh ≪ R_*, you can show that any radius change in the shell, dR_sh, translates to opposite-signed radius changes for the core and envelope to maintain pressure continuity. As the shell narrows around the contracting core, the envelope above must accommodate the change with the opposite sign. Detailed numerical simulations (e.g. MESA, KEPLER) confirm this is a robust feature of all subgiants, not a coincidence.

So the surface of the star bloats outward. The envelope is so opaque that the luminosity escape time grows; the star's outer layers cool by adiabatic expansion. Effective temperature drops from ≈ 5800 K (Sun on the MS) to ≈ 4800 K (Sun at the base of the RGB). Luminosity, meanwhile, rises mildly — from 1 L_⊙ to ≈ 3-5 L_⊙ over the subgiant phase. The star's track on the HR diagram is therefore rightward (cooler) and gently upward (brighter): the subgiant branch.

Worked example: tracking a 1 M_⊙ star

Standard stellar models give the following snapshot of a 1 M_⊙ solar-metallicity star at three points:

                       Main-sequence    Subgiant      Base of RGB
Age (Gyr)              4.6              10.2           11.0
L (L_⊙)                1.00             2.5            3.8
Teff (K)               5778             5400           4900
log L (L_⊙)            0.00             0.40           0.58
log Teff (K)           3.762            3.732          3.690
R (R_⊙)                1.0              1.7            2.8
X_c (central H)        0.34             0.001          0
M_He core (M_⊙)        0.00             0.07           0.13

From "subgiant" to "base of RGB" the star moves about 0.18 dex (1.4×) in luminosity and about 500 K (a quarter of an A spectral subtype) in surface temperature. The phase takes about 0.8 Gyr. During this time the inert helium core grows by 0.06 M_⊙ as the shell deposits more helium ash, and the star's radius almost doubles. The track on the HR diagram appears almost flat in luminosity because L is rising slowly, but it sweeps a wide range in color because Teff is dropping faster.

The Hertzsprung gap

For young clusters whose main-sequence turnoff masses exceed about 1.5 M_⊙, the subgiant phase is so brief that essentially no stars are caught in it. The result is a "gap" on the HR diagram between the upper main sequence and the giant branch — the Hertzsprung gap, named after the Danish astronomer who first noticed it in cluster CMDs in 1909.

The gap's existence is one of the clearest pieces of evidence that post-MS evolution is fast for high-mass stars. A cluster with turnoff at 2 M_⊙ has stars crossing the subgiant region in only ~10 Myr versus a ~Gyr main-sequence lifetime — a 100:1 ratio that makes the gap a near-zero-occupation region. For older clusters with turnoffs near 0.9 M_⊙, the ratio shrinks to ~10:1 and the gap fills in with subgiants.

Massive stars cross the gap so fast that the entire region is observationally "empty" for clusters younger than about 200 Myr. For 5 M_⊙ stars the crossing is essentially instantaneous on the cluster timescale; the only stars seen are either firmly on the main sequence or already on the red giant branch.

Why subgiants are the best ages

Stellar ages from the main sequence are imprecise — a G2V star could be 100 Myr or 8 Gyr old at the same luminosity and temperature. Stellar ages from the red giant branch are degenerate — the RGB is nearly age-independent in luminosity. The subgiant branch is uniquely sensitive to age, because the luminosity at fixed Teff scales nearly linearly with cluster age. Each Gyr of additional age moves the subgiant locus ~0.15 mag fainter at the same color.

The technique has two modern applications:

• Globular cluster ages. Comparing observed subgiant locus to isochrones gives absolute cluster ages with precision of ±0.5 Gyr — accurate enough to resolve a 1 Gyr age difference between two halo clusters, which constrains the timescale of Galactic halo formation. The oldest clusters (M92, NGC 6397) are now dated to 12.5 ± 0.5 Gyr — within 1 Gyr of the age of the universe.
• Field-star ages from asteroseismology. Kepler and TESS have measured mode frequencies for thousands of subgiants. The "small frequency separation" δν between adjacent low-degree modes is set by the helium core mass and stellar age. Asteroseismic ages of subgiants are now routinely good to 5-7 percent. This enables Galactic Archaeology — reconstructing the star formation history of the Milky Way disk from the age distribution of millions of subgiants.

Comparison: subgiant vs neighbouring phases

Phase	Core state	Energy source	HR diagram region	Duration (1 M_⊙)	Mirror motion
Main sequence	H-burning core	p-p chain in core	MS band	~10 Gyr	Stationary
Turnoff	H nearly exhausted	Last p-p in core	Top of MS	~1 Gyr	Slow leftward then rightward
Subgiant	Inert He core contracting	H-burning thin shell	Subgiant branch	~10-50 Myr (strict SGB)	Rightward + mildly up
Base of RGB	Degenerate He core	H-burning shell, narrowing	Lower RGB	~50 Myr	Sharply upward
RGB tip	Degenerate He core, near ignition	H shell + gravity	Upper RGB	~50 Myr	Mostly upward at fixed Teff
Helium flash	Runaway 3α ignition	3α in degenerate core	(brief, mostly internal)	Minutes	—

Famous subgiants you can see

Procyon A (alpha Canis Minoris, F5IV-V, V = +0.34, 3.51 pc). The eighth-brightest star in the sky, a binary with a white-dwarf companion. Procyon's status as a subgiant is so marginal that classifications differ — some place it on the MS, others on the SGB. Asteroseismic measurements from MOST and the Mt Wilson 1.5 m show a stellar age of 1.87 ± 0.13 Gyr and a He core mass of 0.07 M_⊙, putting it just past the turnoff. Procyon is essentially what the Sun will look like in about 3 Gyr.

Beta Hydri (G2IV, V = +2.80, 7.46 pc). The brightest star in the Hydrus constellation and a near-clone of what the Sun will become in 4-5 Gyr. Mass 1.07 M_⊙, age 6.4 ± 0.6 Gyr (from asteroseismology), luminosity 3.5 L_⊙. Pulsates with 17 detected p-modes giving precise interior structure.

Eta Bootis (G0IV, V = +2.68, 11.41 pc). Mass 1.71 M_⊙ — a more massive subgiant. Its acoustic mode frequencies are some of the best-measured of any non-Sun star, used as a benchmark for subgiant models.

HD 209458 b's host star (G0IV, V = +7.65). A relatively normal subgiant, but notable as the first transiting exoplanet host. Knowing the host is a subgiant (not a main-sequence star) affects the inferred age of the planetary system to ~5 Gyr.

How the subgiant phase ends

The subgiant phase blends continuously into the red giant branch. Operationally, models flag the transition when (a) the H-shell thickness drops below ~10⁻³ M_⊙, (b) the convective envelope penetrates deep enough to mix CN-cycled material to the surface (first dredge-up), and (c) the luminosity starts rising sharply with little Teff change. At that point the star is firmly climbing the RGB, and the helium core continues to grow and become degenerate over the following ~500 Myr until the helium flash.

For a 1 M_⊙ star, the entire sequence MS → SGB → RGB tip → flash takes about 12 Gyr. The subgiant portion is a few percent of that, but it is the most informative few percent for ages and stellar structure tests. The Sun has about 5 Gyr left until it begins its own subgiant phase.

Common pitfalls

• Subgiant ≠ smaller-than-giant. The name suggests a "lesser giant" but subgiants are larger than MS stars and smaller than red giants — about 1.5-3 R_⊙ for solar-mass stars. The Morgan-Keenan luminosity class IV designation is what defines them, not a size cutoff.
• The HR track is not straight. Real subgiant tracks have small wiggles where shell burning rate changes faster than envelope contraction adjusts. These wiggles, sometimes called "Henyey hooks," are visible in detailed isochrone plots.
• The phase is not the same as "post-turnoff." The turnoff is the brightest point on the MS where stars are still burning H in the core. Subgiants are past the turnoff but before the RGB base. Strictly, the turnoff itself lasts ~100 Myr for a 1 M_⊙ star before the subgiant phase proper begins.
• Not every subgiant becomes a red giant immediately. Very low-mass subgiants (<0.5 M_⊙) move very slowly, and the universe is younger than their full evolution — these "subgiants" are theoretical extrapolations, not observed.
• Spectroscopic luminosity class IV is broad. Some IV stars are technically post-RGB or He-burning HB stars; the classification is photometric and degenerate. Modern catalogues use asteroseismology or precise Hipparcos parallaxes to distinguish bona fide subgiants from HB and AGB stars at similar Teff and magnitude.