---
title: The Main Sequence and Its Structure
draft: false
module: Stellar Evolution
moduleNumber: 7
lessonNumber: 1
order: 701
summary: >
  A star settles onto the zero-age main sequence when core hydrogen ignition
  halts contraction. Homology scaling of the structure equations reproduces the
  mass–luminosity relation, and the burning mode splits the sequence into an
  upper branch with a convective core and a lower branch with a convective
  envelope. The main-sequence lifetime falls steeply with mass, and the turnoff
  of a coeval cluster serves as a clock.
topics: [Stellar Evolution]
sources:
  - book: Carroll & Ostlie
    ref: "Ch. 10 — The Interiors of Stars; §10.6 Stellar Models. Ch. 13 §13.1 The Main Sequence"
  - book: Maoz
    ref: "Ch. 4 — Stellar Evolution and Stellar Remnants"
---

A star spends most of its life fusing hydrogen into helium in its core. The locus
this burning traces in the luminosity–temperature plane is the main sequence, a
diagonal band that a randomly chosen star occupies about ninety percent of the
time simply because core hydrogen burning is the longest-lived stable phase. The
band is not a track a single star moves along; each point corresponds to a
different mass. The structure at a given mass, the run of luminosity with mass,
and the time spent burning all follow from the equations of stellar structure once
the energy source and the transport mechanism are fixed.[^co-ms]

## The zero-age main sequence

A contracting protostar radiates gravitational energy on the Kelvin–Helmholtz
timescale while its core heats. When the central temperature reaches
$\sim 10^7\ \text{K}$, the pp chain switches on and nuclear energy generation rises
until it balances the surface luminosity. Contraction stops: the star is now in
both hydrostatic and thermal equilibrium, supported by an ideal-gas pressure
maintained by nuclear burning. This chemically homogeneous, newly ignited
configuration is the **zero-age main sequence** (ZAMS). Its position in the
Hertzsprung–Russell diagram is fixed by one parameter, the mass, plus a weak
dependence on composition through the mean molecular weight $\mu$. The uniqueness
of the equilibrium for a given mass and composition is the content of the
Vogt–Russell result used when integrating stellar models.[^co-vr]

Arrival on the ZAMS is the endpoint of pre-main-sequence contraction along the
Hayashi and Henyey tracks, treated in [protostars and the pre-main
sequence](/astrophysics-cosmology/ism-and-star-formation/protostars-and-the-pre-main-sequence).
The lower end of the sequence terminates near $0.08\,M_\odot$, below which the core
never reaches hydrogen ignition and the object becomes a brown dwarf; the upper end
runs to $\sim 100\,M_\odot$, where radiation pressure and instabilities limit
further mass.

## Homology and the mass–luminosity relation

The dependence of luminosity on mass follows from a scaling argument. Suppose stars
of different mass have the same dimensionless structure, so that any two are related
by a uniform rescaling of radius and density. Such stars are called **homologous**,
and each physical variable at fractional radius $x = r/R$ scales as a power of the
total mass and radius. The governing relations are hydrostatic equilibrium, the
ideal-gas equation of state, and radiative diffusion.

Hydrostatic equilibrium sets the pressure scale. From $\d P/\d r = -Gm\rho/r^2$,
dimensional replacement of each factor gives

$$
P_c \sim \frac{G M^2}{R^4},
$$

since $\rho \sim M/R^3$ and $m \sim M$, $r \sim R$. The ideal-gas law
$P = \rho k T/(\mu m_{\rm H})$ then fixes the central temperature,

$$
T_c \sim \frac{\mu m_{\rm H}}{k}\frac{P_c}{\rho_c}
\sim \frac{\mu m_{\rm H} G}{k}\frac{M}{R}.
$$

The central temperature rises with mass and falls with radius: more massive stars
have hotter interiors. Radiative diffusion carries the luminosity outward,

$$
L \sim \frac{4\pi r^2 c}{3\kappa\rho}\frac{\d(a T^4)}{\d r}
\sim \frac{c\,a\,T_c^4\,R^4}{\kappa M},
$$

using $\rho \sim M/R^3$ and $\d T^4/\d r \sim T_c^4/R$. Substituting $T_c \propto \mu M/R$
cancels the radius entirely when the opacity $\kappa$ is constant:

$$
L \propto \frac{\mu^4 M^4}{\kappa M} = \frac{\mu^4}{\kappa}\,M^3 .
$$

Constant opacity is the electron-scattering regime, $\kappa = 0.2(1+X)\ \text{cm}^2\,
\text{g}^{-1}$, appropriate to the hot interiors of massive stars. The result
$L \propto M^3$ is the **mass–luminosity relation**. When the dominant opacity is
instead Kramers, $\kappa \propto \rho T^{-7/2}$, the same procedure gives a steeper
$L \propto M^{5}$ with a mild radius dependence, so the observed exponent runs from
about $4$ near a solar mass to about $3$ for the most massive stars. A single power
law $L \propto M^{3.5}$ fits the ensemble to within the scatter set by composition
and evolutionary state.

$$
% caption: Luminosity against mass on log axes; the homology slope of about 3.5
% (electron scattering steepening toward 4 at low mass through Kramers opacity)
% tracks the observed sequence, with points scattered by composition and age.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (0,0) -- (9.0,0) node[right, black!70] {log mass};
\draw[->, black] (0,0) -- (0,5.2) node[above, black!70] {log luminosity};
% homology line slope ~3.5
\draw[acc, very thick] (0.6,0.5) -- (7.8,4.9);
\node[acc, anchor=north west] at (5.6,3.6) {slope near 3.5};
% scattered data points about the line
\fill[black] (1.2,0.75) circle (1.5pt);
\fill[black] (1.9,1.35) circle (1.5pt);
\fill[black] (2.6,1.6) circle (1.5pt);
\fill[black] (3.3,2.35) circle (1.5pt);
\fill[black] (4.0,2.5) circle (1.5pt);
\fill[black] (4.7,3.15) circle (1.5pt);
\fill[black] (5.4,3.25) circle (1.5pt);
\fill[black] (6.1,3.95) circle (1.5pt);
\fill[black] (6.8,4.1) circle (1.5pt);
% solar reference marker
\draw[black, densely dotted] (2.6,0) -- (2.6,1.6);
\node[black, anchor=north] at (2.6,-0.05) {Sun};
\end{tikzpicture}
$$

The mass–radius relation follows from a second constraint. The luminosity generated
by nuclear burning is $L \sim M\,\epsilon$ with $\epsilon \propto \rho T_c^\nu$, and
$\nu \approx 4$ for the pp chain, $\nu \approx 18$ for CNO. Equating this to the
radiative luminosity above and eliminating $L$ gives $R \propto M^{(\nu-1)/(\nu+3)}$.
For pp burning this predicts $R \propto M^{0.4\text{–}0.8}$, and for CNO burning
$R \propto M^{0.6\text{–}0.8}$; the sequence is nearly $R \propto M^{0.7}$ on
average. Radius grows more slowly than mass, so the mean density
$\bar\rho \propto M/R^3 \propto M^{-1.1}$ drops toward higher mass. Massive stars are
large, hot, luminous, and tenuous; low-mass stars are small, cool, faint, and dense.

## Upper and lower main sequence

The mode of energy transport changes across the sequence because the temperature
sensitivity of the energy source changes. Two regimes divide at roughly
$1.2\,M_\odot$.

- **Upper main sequence** ($M \gtrsim 1.2\,M_\odot$). The central temperature
  exceeds $1.8\times 10^7\ \text{K}$, so the CNO cycle dominates. Its rate
  $\epsilon \propto T^{18}$ concentrates the entire luminosity in a small central
  volume. The radiative gradient there is far too steep for radiation to carry the
  flux, so the core is **convective**, mixing fresh fuel inward. The envelope,
  hotter and less opaque, stays **radiative**.
- **Lower main sequence** ($M \lesssim 1.2\,M_\odot$). The pp chain
  ($\epsilon \propto T^4$) spreads energy generation over a larger central region,
  and the core is **radiative**. The cool outer layers have high opacity from the
  $\text{H}^-$ ion, which steepens the radiative gradient near the surface and drives
  an outer **convective** envelope. The Sun sits here: a radiative interior out to
  $0.71\,R_\odot$ with a convective zone above it.

The crossover in the burning mechanism is the same $\nu = (\tau - 2)/3$ scaling
derived from the Gamow peak in [thermonuclear reaction rates and the Gamow
peak](/astrophysics-cosmology/nuclear-astrophysics/thermonuclear-reaction-rates-and-the-gamow-peak);
the two hydrogen networks and their crossover temperature are worked out in
[hydrogen burning: pp chains and the CNO
cycle](/astrophysics-cosmology/nuclear-astrophysics/hydrogen-burning-pp-chains-and-cno).

$$
% caption: A one-solar-mass star (radiative core, convective envelope) beside a
% ten-solar-mass star (convective core, radiative envelope); the transport mode
% inverts because CNO burning concentrates the flux at the center.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
% left: 1 Msun
\draw[very thick] (2.3,2.4) circle (1.9);
\draw[densely dashed, thick] (2.3,2.4) circle (1.35);
\draw[black] (2.3,2.4) circle (0.65);
\node[black!75] at (2.3,2.4) {pp core};
\node[black!70] at (2.3,3.55) {radiative};
\node[black!70] at (2.3,1.15) {convective};
\node[black!70] at (2.3,4.6) {1 solar mass};
% right: 10 Msun
\draw[very thick] (7.4,2.4) circle (2.2);
\draw[densely dashed, thick] (7.4,2.4) circle (0.95);
\draw[black] (7.4,2.4) circle (0.95);
\node[black!75] at (7.4,2.4) {CNO core};
\node[black!70] at (7.4,3.55) {radiative};
\node[black!70] at (7.4,1.25) {convective};
\node[black!70] at (7.4,4.9) {10 solar masses};
\end{tikzpicture}
$$

## The main-sequence lifetime

The time a star burns hydrogen is the fuel available divided by the rate of
consumption. A fraction $f \approx 0.1$ of the mass passes through the core and
releases $0.007\,c^2$ per unit mass of hydrogen converted to helium, so the
available energy is $E \approx 0.007\,f\,M c^2$. The burning rate is the luminosity,
giving

$$
t_{\rm MS} = \frac{E}{L} = 0.007\,f\,\frac{M c^2}{L}.
$$

For the Sun this evaluates to about $10^{10}\ \text{yr}$. Inserting the
mass–luminosity relation $L \propto M^{3.5}$,

$$
t_{\rm MS} \propto \frac{M}{L} \propto M^{-2.5},
\qquad
t_{\rm MS} \approx 10^{10}\ \text{yr}\left(\frac{M}{M_\odot}\right)^{-2.5}.
$$

The lifetime falls steeply with mass: a $10\,M_\odot$ star burns out in about
$3\times 10^7\ \text{yr}$, a $30\,M_\odot$ star in a few million years, while a
$0.5\,M_\odot$ star would outlast the present age of the universe many times over.
Massive stars are rare, luminous, and short-lived; low-mass stars are common, faint,
and effectively eternal on cosmic timescales. This inverse scaling underlies the use
of the main-sequence turnoff as an age indicator.

## The band and the turnoff

The main sequence has a finite width for two reasons. Stars differ in composition,
and higher metallicity shifts a star toward lower temperature at fixed mass. The
larger effect is that a star evolves slightly during core hydrogen burning: as helium
accumulates, the mean molecular weight rises, and $L \propto \mu^4$ forces the
luminosity up. The star brightens and expands, drifting upward and to the right of
the ZAMS until core hydrogen is exhausted at the **terminal-age main sequence**. The
occupied region between ZAMS and terminal-age main sequence is a band roughly one
magnitude wide, not a line.

$$
% caption: The zero-age main sequence as a band in the H-R diagram, hottest and most
% luminous at high mass (upper left) and cool and faint at low mass (lower right);
% temperature increases to the left in the conventional orientation.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (9.0,0) -- (0,0);
\node[black!70, anchor=north] at (4.5,-0.15) {temperature increases to the left};
\draw[->, black] (9.0,0) -- (9.0,5.2) node[above, black!70] {luminosity};
% ZAMS band (two parallel curves)
\draw[very thick] (1.0,4.9) .. controls (3.0,3.6) and (5.0,2.0) .. (8.2,0.7);
\draw[thick] (1.3,5.2) .. controls (3.3,3.9) and (5.3,2.3) .. (8.5,1.0);
\node[black!75, anchor=west] at (1.5,4.6) {30 solar masses};
\node[black!75] at (4.6,2.9) {1 solar mass};
\node[black!75, anchor=east] at (8.2,1.0) {0.2 solar mass};
\node[anchor=south, rotate=-30] at (5.6,2.15) {main sequence};
\end{tikzpicture}
$$

A star cluster forms its members at nearly one time from one cloud, so they share an
age and composition and differ only in mass. Plotting the cluster in the H-R diagram
shows every member of mass above the turnoff already evolved off the main sequence,
because for those masses $t_{\rm MS}$ is shorter than the cluster age. The
**main-sequence turnoff** is the point where stars are just now exhausting core
hydrogen: its mass satisfies $t_{\rm MS}(M) = t_{\rm cluster}$. Reading the turnoff
luminosity and inverting the lifetime relation dates the cluster. As a cluster ages
the turnoff slides down the sequence to lower mass and luminosity, so a young cluster
turns off high on the diagram and an old globular cluster turns off just above the
Sun's position.

$$
% caption: Isochrones of a coeval cluster at three ages; the turnoff where stars
% leave the main sequence migrates to lower luminosity as the cluster ages, and its
% position dates the cluster through the lifetime relation.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (9.0,0) -- (0,0) node[left, black!70] {hotter};
\draw[->, black] (9.0,0) -- (9.0,5.2) node[above, black!70] {luminosity};
% shared lower main sequence
\draw[black, thick] (8.5,0.7) .. controls (6.0,1.8) and (4.5,2.6) .. (3.2,3.4);
% young cluster: high turnoff, giant branch peeling up-left early
\draw[very thick] (3.2,3.4) .. controls (2.7,3.7) and (2.4,4.2) .. (2.6,4.8);
\fill (3.2,3.4) circle (2pt);
\node[anchor=west] at (3.3,3.5) {young};
% intermediate cluster
\draw[very thick, densely dashed] (4.6,2.55) .. controls (4.1,2.9) and (3.8,3.5) .. (4.0,4.2);
\fill (4.6,2.55) circle (2pt);
\node[black!70, anchor=west] at (4.7,2.65) {intermediate};
% old cluster: low turnoff near solar
\draw[black, very thick] (6.2,1.7) .. controls (5.7,2.05) and (5.4,2.6) .. (5.6,3.4);
\fill[black] (6.2,1.7) circle (2pt);
\node[black!70, anchor=west] at (6.3,1.8) {old};
\end{tikzpicture}
$$

The mass ordering established here propagates through everything that follows. Low-
and intermediate-mass stars leave the main sequence to become red giants and end as
white dwarfs, worked out in [post-main-sequence evolution of low-mass
stars](/astrophysics-cosmology/stellar-evolution/post-main-sequence-low-mass-evolution).
Massive stars run through the heavier burning stages in [the evolution of massive
stars](/astrophysics-cosmology/stellar-evolution/the-evolution-of-massive-stars) and
end in core collapse. The dividing mass, near $8\,M_\odot$, is set by whether the
core ever reaches the temperatures required to burn carbon.

[^co-ms]: Carroll & Ostlie, §13.1 — The Main Sequence: the ZAMS, the division into upper and lower main sequence by burning mode and convection, and the mass dependence of structure and lifetime.
[^co-vr]: Carroll & Ostlie, §10.6 — Stellar Models: homology relations, the mass–luminosity relation, and the Vogt–Russell theorem fixing the equilibrium structure from mass and composition.