Spotlight #32 – Astrophysics & Stellar Evolution: HR Diagram, Stellar Structure, Supernovae and Gravitational Waves

Every atom heavier than lithium was made inside a star, and most of the hydrogen and helium in the universe was forged in the first three minutes after the Big Bang. Stellar astrophysics sits at the intersection of nuclear physics, fluid dynamics, radiative transfer, and general relativity. The simplest stars — main-sequence objects in hydrostatic equilibrium — are well-described by a handful of equations. The complicated ones — asymptotic giant branch stars with multiple burning shells, rotating massive stars losing mass in winds, compact binary systems spiralling to merger — are among the most active frontiers in modern physics.

1. The Hertzsprung-Russell Diagram

When Ejnar Hertzsprung and Henry Norris Russell independently plotted stellar luminosity against surface temperature around 1910, they discovered that stars are not distributed randomly: the vast majority cluster along a diagonal band (the main sequence), with giant and supergiant branches above and white dwarfs below. The HR diagram is the most useful single diagram in stellar astrophysics because each evolutionary stage occupies a distinct region.

Stefan-Boltzmann relation:
  L = 4πR² σ T_eff&sup4;

Main-sequence mass-luminosity relation (approximate):
  L / L_sun ≈ (M / M_sun)^4      for 0.5 < M/M_sun < 50
  L / L_sun ≈ (M / M_sun)^2.5    for lower-mass stars

Main-sequence lifetime:
  t_MS ≈ t_sun · (M/M_sun)^{-2.5}    (solar t_MS ≈ 10 Gyr)
  1 M_sun: ~10 Gyr     5 M_sun: ~100 Myr     25 M_sun: ~7 Myr

HR diagram regions (L in L_sun, T in K):
  Main sequence:  0.001–10^6 L_sun, 2500–50 000 K
  Red giants:    10–1000 L_sun, 3500–5500 K
  Supergiants:   10^4–10^6 L_sun, 3500–30 000 K
  White dwarfs:  10^-4–0.1 L_sun, 5000–100 000 K

Spectral types (OBAFGKM, hottest to coolest):
  O: >30 000 K   B: 10 000–30 000 K   A: 7500–10 000 K
  F: 6000–7500 K   G: 5200–6000 K   K: 3700–5200 K
  M: <3700 K (cool red dwarf; most numerous type in the Galaxy)

A star’s position on the HR diagram changes as it evolves. The Sun will leave the main sequence in ~5 Gyr, expand to a red giant (R ~200 R_sun), lose its outer layers as a planetary nebula, and leave behind a white dwarf cooling track. Massive O-type stars evolve so rapidly that they are never in equilibrium with ambient gas pressure: they die in supernova explosions within 3–10 Myr of their birth. Star clusters are especially useful because all members share the same age; the turn-off point — the most luminous main-sequence star still burning hydrogen — gives the cluster age directly.

2. Stellar Interior Structure and Polytropes

A star is a self-gravitating ball of gas in hydrostatic equilibrium: gravity pulling inward and pressure gradient pushing outward balance at every layer. Three additional equations describe energy generation, energy transport, and how temperature changes with pressure. Together they form the stellar structure equations, which uniquely determine a star’s internal profile once its mass and composition are specified (Vogt-Russell theorem).

Hydrostatic equilibrium:
  dP/dr = −ρ G M(r) / r²

Mass continuity:
  dM/dr = 4πr² ρ

Energy transport (radiative zone):
  dT/dr = −3κρL / (64πσr²T³)
  κ = Rosseland mean opacity (cm²/g)

Convection criterion (Schwarzschild):
  |dT/dr|_rad > |dT/dr|_ad   → convective instability
  Mixing-length theory: convective cell scale ≈ α_MLT H_P  (α ~ 1.5-2)

Polytropic equation of state:
  P = K ρ^{(n+1)/n}   (polytropic index n)
  n = 3/2 → fully convective (low-mass stars, giant cores)
  n = 3   → radiation-pressure dominated (massive stars; Chandrasekhar)

Lane-Emden equation (dimensionless stellar structure):
  d/dξ[ξ² dθ/dξ] + ξ² θ^n = 0
  ξ = r / r_n (scaled radius),  θ = (ρ/ρ_c)^{1/n}

Solar interior (AGSS09 model):
  Core (r < 0.25 R_sun):  T_c = 1.57×10^7 K,  ρ_c = 151 g/cm³
  Radiation zone: 0.25–0.71 R_sun
  Convection zone: 0.71–1.00 R_sun  (depth ~200 000 km)

Helioseismology — measuring the speed of sound waves by observing surface oscillations — has verified the solar interior structure to sub-1% accuracy. The discrepancy between the 3D spectroscopic solar abundances (AGSS09) and helioseismic constraints remains an open problem in stellar physics: the “solar metallicity problem.”

3. Nuclear Burning Cycles

The energy source that allows stars to resist gravitational collapse is nuclear fusion. Low-mass stars (M < 1.3 M_sun) fuse hydrogen primarily via the proton-proton (pp) chain; more massive stars use the CNO cycle, whose rate is highly sensitive to temperature (T^17 vs T^4 for pp). As a star ages, it builds up a helium ash core, then ignites helium via the triple-alpha process, and — for massive stars — proceeds through carbon, neon, oxygen and silicon burning on progressively shorter timescales.

Proton-proton chain (dominant for M ≤ 1.3 M_sun):
  pp I:  4H → ⁴He + 2e’ + 2ν_e       (Q = 26.7 MeV)
  pp II: ⁷Be branch (important at T > 14 MK)
  pp III: ⁸B branch (produces high-energy neutrinos detected at SNO)
  Energy generation: ε_pp ∝ ρ X² T^4  (X = hydrogen mass fraction)

CNO cycle (dominant for M > 1.3 M_sun):
  ¹²C + p → ¹³N + γ → ¹³C + e’ + ν
  ¹³C + p → ¹⁴N + γ
  ¹⁴N + p → ¹⁵O + γ  ← rate-limiting step
  ¹⁵O → ¹⁵N + e’ + ν → + p → ¹²C + ⁴He (net)
  Energy generation: ε_CNO ∝ ρ X_CNO T^17

Triple-alpha process (helium burning, T > 10^8 K):
  2 ⁴He → ⁸Be* (resonance lifetime 2.6×10^-16 s)
  ⁸Be + ⁴He → ¹²C* (Hoyle state, 7.6644 MeV)  → ¹²C + 2γ
  ¹²C + ⁴He → ¹⁶O + γ  (competitive; determines C/O ratio)
  Hoyle resonance predicted by Hoyle (1954), confirmed by Cook et al.

Advanced burning stages (massive stars, T in units of 10^8 K):
  He burning:   T ~2,   lifetime ~10^6 yr   (>8 M_sun)
  C burning:    T ~5,   lifetime ~10^3 yr
  Ne burning:   T ~15,  lifetime ~1 yr
  O burning:    T ~20,  lifetime ~months
  Si burning:   T ~35,  lifetime ~days
  Iron core:    T ~70,  no energy release → core collapse
  Iron-group nuclei are the endpoint: binding energy peaks at ⁵⁶Fe/⁵⁸Ni

4. Core-Collapse Supernovae

When a massive star (>8 M_sun) accumulates an iron core more massive than the Chandrasekhar limit (~1.4 M_sun), electron degeneracy pressure can no longer support it. The core collapses from roughly Earth-size to a 10-km neutron star in under a second, releasing ~3 × 10^53 ergs — the gravitational binding energy of a neutron star. This is ≈ 100 times the total electromagnetic energy radiated by the Sun over its entire 10 Gyr lifetime, released in milliseconds.

Chandrasekhar mass (relativistic electron degeneracy limit):
  M_Ch = 5.83 Y_e² M_sun   (Y_e = electron fraction ≈ 0.5 for ⁵⁶Fe)
  M_Ch ≈ 1.44 M_sun    (actual value ~1.2-1.4 M_sun with corrections)

Collapse timeline:
  t = 0 ms:    Iron core exceeds M_Ch; free-fall begins
  t ~ 100 ms:  Core reaches nuclear density (ρ_nuc = 2.7×10^14 g/cm³)
  t ~ 110 ms:  Core bounces; shock wave launched at ~100-200 km/s
  t ~ 200 ms:  Shock stalls — photodissociation of iron costs 8.8 MeV/nucleon
  t ~ 500 ms:  Neutrino heating re-energises shock (BNNS mechanism)
  t ~ 1 s:     Successful explosion; shock breaks out ~15-20 hr later
  t ~ days:    Optical maximum

Neutrino burst (confirmed for SN 1987A):
  E_ν total: ~3 × 10^53 erg  (99% of collapse energy)
  Duration: ~10 s (neutron-star Kelvin-Helmholtz cooling)
  Kamiokande II + IMB + Baksan detected 24 neutrinos from SN 1987A (LMC, 160 kLy)

Nucleosynthesis in the explosive phase:
  r-process (rapid neutron capture): produces ~50% of A>100 nuclei
  Recent evidence: r-process in NS merger GW170817 kilonova (gold, platinum, europium)
  p-process: photodisintegration of seed nuclei; proton-rich isotopes
  ν p-process: neutrino-driven proton capture in early ejecta

Peak luminosity of core-collapse SNe:
  L_peak ~ 10^43 erg/s (optical)
  Decay powered by ⁵⁶Ni → ⁵⁶Co → ⁵⁶Fe (t_½ = 6.08 d, 77.2 d)
  Brightness ∝ M_Ni synthesised (~0.07 M_sun for SN 1987A)

5. Neutron Stars and Pulsars

The remnant core left behind by a core-collapse supernova is a neutron star — a sphere of ~1.4 M_sun compressed to ~10 km radius, supported by neutron degeneracy pressure and the repulsive hard core of the nuclear force. Mean density exceeds nuclear density; a teaspoon would weigh ≈ 5 × 10^8 tonnes. If the progenitor core exceeds ~2–3 M_sun, collapse continues past neutron star stability to a black hole.

Tolman-Oppenheimer-Volkoff (TOV) equation of hydrostatic equilibrium in GR:
  dP/dr = −G(M + 4πr³P/c²)(ρ + P/c²) / [r(r − 2GM/c²)]
  Reduces to Newtonian dP/dr = −ρGM/r² at low compactness

Maximum neutron star mass:
  Tolman-Oppenheimer-Volkoff limit ~0.7 M_sun (pure neutron matter, soft EOS)
  Real limit (stiff nuclear EOS): ~2.3–2.5 M_sun
  Measured: PSR J0952-0607 = 2.35 ± 0.17 M_sun (2022, heaviest confirmed NS)

Pulsar spin-down (magnetic dipole radiation):
  P_dot = −B²R^6Ω^3sin²α / (6Ic³)
  P = spin period (fastest known: J1748-2446ad, P = 1.396 ms)
  Characteristic age: τ = P / (2 P_dot)
  Surface B-field:    B ~ 3.2×10^19 √(P P_dot)  Gauss

Neutron star cooling:
  t < 100 yr:  Direct Urca (if M > M_DU), fast cooling, T_s ~ 10^6 K
  t ~ 10^3 yr: Modified Urca dominant, T_s ~ 5×10^5 K
  t ~ 10^5 yr: Photon cooling, T_s ~ 10^5 K
  Observed examples: Cassiopeia A (350 yr) showed cooling at ~2% per decade (Ho & Heinke 2009)

Equation of state (EOS) of dense matter:
  Hadronic: SLy, APR4 (nuclear+3-body forces)
  Exotic: hyperonic, kaon condensate, quark-gluon plasma
  Radius constraint (NICER): R = 12.35 ± 0.75 km for J0030+0451 (Riley 2019)

6. Gravitational Waves from Compact Binary Inspiral

General relativity predicts that accelerating masses radiate gravitational waves — ripples in spacetime curvature propagating at the speed of light. Binary systems of compact objects (neutron stars or black holes) lose energy to gravitational radiation, causing the orbit to shrink until the two bodies merge. LIGO/Virgo first detected this signal on 14 September 2015 (GW150914): two black holes of 36 and 29 M_sun merging at ~410 Mpc, producing a ring-down signal lasting ~0.2 seconds.

Quadrupole formula for GW power (leading-order PN):
  P_GW = 32G^4 m_1²m_2²(m_1+m_2) / (5c^5 a^5)
  a = orbital separation

Chirp mass ℳ (determines phase evolution):
  ℳ = (m_1 m_2)^{3/5} / (m_1+m_2)^{1/5}
  Directly measurable from frequency sweep f_GW(t)

Orbital decay (Peters 1964):
  da/dt = −64G³m_1m_2(m_1+m_2) / (5c^5 a³)
  Merger time: t_merge = 12a_0^4c^5 / (19G³ m_1m_2(m_1+m_2))
  PSR 1913+16 (Hulse-Taylor pulsar): t_merge ~ 300 Myr; orbital decay matches GR to 0.2%

Gravitational wave strain:
  h_+ = − (2Gℳ/c²r)(Gπfℳ/c^3)^{2/3} cos(2φ)
  h_× of same amplitude, 90° phase offset

GW150914 (first detection, Sept 2015):
  Component masses: 35.6 M_sun + 30.6 M_sun
  Merger mass: 63.1 M_sun  (radiated ΔE = 3.0 M_sun c² ~ 5.4×10^54 erg)
  Peak strain: h ~ 10^-21   (LIGO arm length 4 km; ΔL ~ 4×10^-18 m, 1/1000 of proton)
  Peak GW luminosity: ~3.6×10^56 erg/s (outshines all visible stars in observable universe)

GW170817 (first NS-NS merger, Aug 2017):
  Component masses: 1.17 + 1.36 M_sun
  Associated kilonova AT2017gfo: blue (Sr) + red (heavy r-process) component
  γ-ray burst GRB 170817A offset by 1.74 s → c_GW/c = 1 ± 10^-15 (tests Lorentz invariance)
  Hubble constant: H_0 = 70.0^{+12}_{-8} km/s/Mpc (multimessenger)

As of 2024, the LIGO-Virgo-KAGRA network has catalogued over 90 compact binary merger candidates across three observing runs (GWTC-3). The fourth observing run (O4, 2023–2025) is expected to double this sample. The next generation of detectors — Einstein Telescope (underground, 10 km arms) and Cosmic Explorer (40 km arms) — will detect binary black hole mergers throughout most of the observable universe, enabling direct measurement of the cosmic expansion history without electromagnetic observations.

Related Simulations