---
title: Thermonuclear Supernovae
draft: false
module: Stellar Death and Compact Remnants
moduleNumber: 8
lessonNumber: 3
order: 803
summary: >
  A carbon-oxygen white dwarf driven toward the Chandrasekhar mass ignites its
  degenerate fuel and unbinds itself in a thermonuclear runaway, the Type Ia
  supernova. The light curve is powered by the radioactive decay of nickel-56 to
  cobalt-56 to iron-56, and the Phillips relation between peak brightness and decline
  rate makes these events standardizable candles. Their near-uniform luminosity turns
  them into the distance indicators that revealed cosmic acceleration.
topics: [Stellar Death and Compact Remnants]
sources:
  - book: Carroll & Ostlie
    ref: "Ch. 18 — Close Binary Star Systems; §18.6 White Dwarfs in Binaries; Ch. 15"
  - book: Ryden
    ref: "Ch. 6 — Measuring Cosmological Parameters"
  - book: Maoz
    ref: "Ch. 4 — Stellar Death and Remnants"
---

A Type Ia supernova is the thermonuclear disruption of a white dwarf. Unlike a
core-collapse event, it involves no massive star and no neutron star; it is a
detonation of degenerate carbon and oxygen that leaves nothing behind. The defining
observational facts are a spectrum with no hydrogen but a strong silicon absorption
line, a light curve powered by radioactivity, and a peak luminosity uniform enough
that the small residual scatter can be calibrated away. This lesson sets out the
progenitor channels that bring a white dwarf to ignition, the physics of the burning
front, the radioactive decay chain that lights the ejecta, and the width-luminosity
relation that makes Type Ia supernovae the standardizable candles behind the
discovery of cosmic acceleration.

## Progenitor channels

A [white
dwarf](/astrophysics-cosmology/stellar-death-and-compact-remnants/white-dwarfs-and-the-chandrasekhar-limit)
in isolation simply cools forever. To explode it must gain mass or merge, and it must
be carbon-oxygen: a helium white dwarf is too light to reach ignition, and an
oxygen-neon-magnesium white dwarf tends to collapse by electron capture rather than
detonate. Two channels dominate the discussion.

- **Single-degenerate.** The white dwarf accretes hydrogen or helium from a
  non-degenerate companion — a main-sequence star, subgiant, or red giant — through
  [Roche-lobe overflow](/astrophysics-cosmology/binaries-and-gravitational-waves/binary-systems-and-mass-transfer).
  If the accretion rate lies in a narrow window, the accreted hydrogen burns steadily
  to carbon-oxygen on the surface and the white dwarf grows toward the Chandrasekhar
  mass. Ignition occurs near $M_{\mathrm{Ch}}$, which naturally explains the
  uniformity of the explosions: they all detonate at nearly the same mass.
- **Double-degenerate.** Two white dwarfs in a close binary spiral together through
  [gravitational-wave](/astrophysics-cosmology/binaries-and-gravitational-waves/gravitational-waves-from-inspiraling-binaries)
  emission and merge. The combined mass can exceed $M_{\mathrm{Ch}}$, and the
  disruption of the lighter dwarf onto the heavier one triggers ignition. This channel
  allows super-Chandrasekhar and sub-Chandrasekhar events and may account for the
  observed diversity.

The relative importance of the two channels remains unsettled; the near-absence of
detected companions and hydrogen in Ia remnants favors double-degenerate mergers for
many events, while the tight standardizability points to a common near-Chandrasekhar
mass scale.[^co-ia]

$$
% caption: The two progenitor channels: a white dwarf accreting from a
% non-degenerate companion toward the Chandrasekhar mass, or two white dwarfs merging
% after a gravitational-wave inspiral.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
% single-degenerate
\draw[black] (1.5,3.6) circle (0.85);
\node[black, font=\scriptsize] at (1.5,3.6) {donor};
\draw[acc, very thick] (3.2,2.9) circle (0.3);
\fill[acc!16] (3.2,2.9) circle (0.3);
\draw[->, thick] (2.2,3.2) .. controls (2.7,3.1) and (2.9,3.0) .. (3.0,2.95);
\node[font=\scriptsize, anchor=west] at (3.5,2.9) {white dwarf};
\node[black, anchor=north] at (2.6,1.9) {single-degenerate};
% double-degenerate
\draw[acc, very thick] (6.4,3.5) circle (0.35);
\fill[acc!16] (6.4,3.5) circle (0.35);
\draw[acc, very thick] (7.7,3.0) circle (0.28);
\fill[acc!16] (7.7,3.0) circle (0.28);
\draw[->, thick] (6.7,3.3) .. controls (7.0,3.15) and (7.2,3.1) .. (7.4,3.05);
\draw[black, densely dashed] (7.05,3.25) ellipse (1.15 and 0.75);
\node[black, anchor=north] at (7.0,1.9) {double-degenerate};
\end{tikzpicture}
$$

## Carbon ignition and the burning front

The white-dwarf interior is degenerate, and degenerate matter is a thermal
runaway waiting to happen: its pressure does not depend on temperature, so a
rise in temperature raises the reaction rate without raising the pressure that would
expand and cool the gas. Compressional heating near the center, or the deep heating of
a merger, lifts the temperature until carbon ignites at
$T \approx 7\times10^8\ \mathrm{K}$. The runaway is explosive because there is no
pressure relief valve until the temperature climbs high enough,
$\sim 10^9\ \mathrm{K}$, to lift the electron degeneracy.

The burning front propagates in one of two modes.

> **Definition (Deflagration and detonation).** A **deflagration** is a subsonic
> burning front that propagates by thermal conduction and turbulent mixing; the fuel
> ahead of it is warned by pressure waves and pre-expands. A **detonation** is a
> supersonic front driven by a shock; the fuel is burned before it can expand. A pure
> detonation of a Chandrasekhar-mass white dwarf overproduces iron-peak elements; a
> pure deflagration leaves too much unburned carbon.

The favored models begin as a deflagration and transition to a detonation partway
through — the **delayed-detonation** picture. The initial subsonic burn lets the star
expand, lowering the density, so that when the detonation sweeps the outer layers it
produces the intermediate-mass elements (silicon, sulfur, calcium) seen in the
spectrum alongside the iron-peak core. Roughly half a solar mass of the white dwarf
is burned to radioactive nickel-56, and the total energy released,
$\sim 10^{44}\ \mathrm{J}$, exceeds the gravitational binding energy and unbinds the
star completely at ejecta speeds of $\sim 10^4\ \mathrm{km\,s^{-1}}$.

$$
% caption: The burning front in the delayed-detonation model: a subsonic deflagration
% starts near the center and lets the star pre-expand, then transitions to a
% supersonic detonation that consumes the outer layers.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
% panel A: deflagration
\draw[very thick] (1.8,2.2) circle (1.4);
\draw[black, densely dotted] (1.8,2.2) circle (0.6);
\foreach \a in {20,80,140,200,260,320}
  \draw[->, black] ({1.8+0.65*cos(\a)},{2.2+0.65*sin(\a)}) -- ({1.8+0.95*cos(\a)},{2.2+0.95*sin(\a)});
\node[black, anchor=north] at (1.8,0.55) {subsonic front};
% panel B: detonation
\draw[very thick] (6.0,2.2) circle (1.6);
\draw[acc, very thick] (6.0,2.2) circle (1.05);
\foreach \a in {20,80,140,200,260,320}
  \draw[->, acc, thick] ({6.0+1.1*cos(\a)},{2.2+1.1*sin(\a)}) -- ({6.0+1.55*cos(\a)},{2.2+1.55*sin(\a)});
\node[black, anchor=north] at (6.0,0.4) {detonation (supersonic)};
\end{tikzpicture}
$$

## The radioactive light curve

A Type Ia supernova has no central engine after the explosion. The ejecta expand and
would cool adiabatically to invisibility, but they are kept hot by the radioactive
decay of the nickel-56 synthesized in the burn. The chain is

$$
{}^{56}\mathrm{Ni} \xrightarrow{\;6.1\ \mathrm{d}\;} {}^{56}\mathrm{Co}
\xrightarrow{\;77.2\ \mathrm{d}\;} {}^{56}\mathrm{Fe},
$$

both steps electron captures that emit gamma rays. The gamma rays thermalize in the
expanding ejecta and are reradiated as the optical display. The **peak luminosity is
set by the mass of nickel-56 synthesized**, through Arnett's rule: the luminosity at
maximum equals the instantaneous decay power, because the light-curve rise time
matches the effective diffusion time. Since a Chandrasekhar-mass explosion produces a
characteristic $\sim 0.6\,M_\odot$ of nickel-56, the peak luminosity is nearly
universal, $M_V \approx -19.3$.

> **Worked example.** The decay power per unit nickel mass at $t = 0$ is
> $\varepsilon_{\mathrm{Ni}} \approx 6.5\times10^{9}\ \mathrm{J\,s^{-1}\,g^{-1}}$.
> For $M_{\mathrm{Ni}} = 0.6\,M_\odot = 1.2\times10^{33}\ \mathrm{g}$ the initial
> power is
> $L_0 \approx \varepsilon_{\mathrm{Ni}} M_{\mathrm{Ni}} \approx
> 8\times10^{42}\ \mathrm{W}$, or about $2\times10^{9}\,L_\odot$. By Arnett's rule the
> peak luminosity, reached after a rise time of $\sim 18\ \mathrm{days}$, approaches
> this radioactive power, matching the observed $M_V \approx -19$.

After maximum the light curve declines, first on the nickel timescale and then, after
a few weeks, on the $77$-day cobalt timescale, which sets the slope of the exponential
tail. The two decay constants are visible directly as two slopes in the light curve.

$$
% caption: The radioactive decay chain powering the light curve: nickel-56 decays to
% cobalt-56 in days and cobalt-56 to stable iron-56 over months, and the two lifetimes
% appear as the two decline slopes.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\node[draw] (ni) at (1.3,3.0) {Ni-56};
\node[draw] (co) at (4.3,3.0) {Co-56};
\node[draw, black] (fe) at (7.3,3.0) {Fe-56 stable};
\draw[->, very thick] (ni) -- (co) node[midway, above, black, font=\scriptsize] {6.1 d};
\draw[->, very thick] (co) -- (fe) node[midway, above, black, font=\scriptsize] {77.2 d};
% mini light curve below
\draw[->, black] (0.6,0.3) -- (8.4,0.3) node[right, black!70] {time};
\draw[->, black] (0.6,0.3) -- (0.6,2.1) node[above, black!70] {log L};
\draw[acc, very thick] (0.9,0.6) .. controls (1.6,1.9) and (2.0,1.9) .. (2.5,1.75)
  -- (3.6,1.25) -- (7.8,0.5);
\node[black, anchor=south west, font=\scriptsize] at (2.4,1.7) {peak};
\node[black, anchor=south, font=\scriptsize] at (5.6,0.75) {cobalt tail};
\end{tikzpicture}
$$

## The width-luminosity relation and standardization

Type Ia supernovae are not perfectly identical, but their deviations are correlated.
Intrinsically brighter events have **broader** light curves — they rise and fall more
slowly — and fainter events decline faster. This is the **Phillips relation**, usually
parametrized by $\Delta m_{15}$, the magnitude decline in the $15$ days after peak: a
larger $\Delta m_{15}$ (faster decline) goes with a fainter peak. Physically, more
nickel-56 both brightens the peak and, by raising the ejecta opacity and temperature,
lengthens the diffusion time and broadens the curve.

Correcting each light curve by its width — the **stretch** correction — collapses the
family onto a single template. The residual scatter after correction is only about
$0.1$–$0.15\ \mathrm{mag}$, or $\sim 7\%$ in distance. This makes Type Ia supernovae
**standardizable candles**: not standard out of the box, but reducible to a standard
by a measured one-parameter correction.

> **Definition (Standardizable candle).** A source whose intrinsic luminosity is not
> fixed but is a known function of an observable — here the light-curve decline rate.
> Measuring that observable predicts the absolute magnitude, and comparison with the
> apparent magnitude yields the [distance
> modulus](/astrophysics-cosmology/observational-foundations/magnitudes-fluxes-and-the-distance-modulus).

$$
% caption: Stretch correction: the raw Type Ia light curves scatter in peak
% brightness and width, but scaling each by its decline rate collapses them onto one
% template, the basis of standardization.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
% left: raw family
\draw[->, black] (0.4,0.3) -- (4.2,0.3) node[right, black!70] {time};
\draw[->, black] (0.4,0.3) -- (0.4,4.0) node[above, black!70] {brightness};
\draw[very thick] (0.7,0.6) .. controls (1.4,3.6) and (1.7,3.6) .. (2.2,2.6) -- (3.9,0.7);
\draw[black, very thick, densely dashed] (0.7,0.6) .. controls (1.2,2.7) and (1.5,2.7) .. (1.9,2.1) -- (3.0,0.6);
\draw[black, very thick, densely dotted] (0.7,0.6) .. controls (1.6,3.1) and (2.1,3.1) .. (2.8,2.2) -- (4.0,0.7);
\node[black, anchor=south] at (2.2,-0.15) {raw};
% right: standardized
\draw[->, black] (5.0,0.3) -- (8.6,0.3) node[right, black!70] {time};
\draw[->, black] (5.0,0.3) -- (5.0,4.0);
\draw[acc, very thick] (5.3,0.6) .. controls (6.0,3.3) and (6.4,3.3) .. (7.0,2.4) -- (8.3,0.7);
\node[black, anchor=south] at (6.8,-0.15) {stretch-corrected};
\end{tikzpicture}
$$

## Type Ia as cosmological distance indicators

Because their corrected peak luminosity is known, Type Ia supernovae are visible as
standard candles to redshifts beyond $z \approx 1$, far enough to probe the expansion
history. Plotting distance modulus against redshift builds the **Hubble diagram** at
cosmological distances, an extension of the local
[distance ladder](/astrophysics-cosmology/observational-foundations/the-cosmic-distance-ladder).
At low redshift the diagram gives the Hubble constant; at high redshift its curvature
measures the deceleration or acceleration of the expansion.

In 1998 two teams found that distant Type Ia supernovae were **fainter**, hence more
distant, than a decelerating universe predicts. The observed residual in the Hubble
diagram implied that the expansion has been accelerating, driven by a dark-energy
component. Type Ia supernovae thus supplied the first direct evidence for cosmic
acceleration, the subject of the [dark-energy
lesson](/astrophysics-cosmology/cosmology-expansion-and-dynamics/dark-energy-and-the-accelerating-universe).
Their reliability rests on the physics of this lesson: a near-Chandrasekhar mass scale
that fixes the nickel yield, and the width-luminosity relation that removes the
remaining scatter.

$$
% caption: The Hubble diagram of distant Type Ia supernovae lies above the
% decelerating prediction, the upward residual in faintness that revealed an
% accelerating expansion.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (0,0) -- (8.6,0) node[right, black!70] {redshift};
\draw[->, black] (0,0) -- (0,4.4) node[above, black!70] {distance modulus};
% decelerating model (lower)
\draw[black, very thick, densely dashed] (0.4,0.5) .. controls (3.0,2.2) and (5.5,3.0) .. (8.2,3.4);
\node[black, anchor=north west] at (5.2,2.7) {decelerating model};
% accelerating (upper, data)
\draw[acc, very thick] (0.4,0.5) .. controls (3.0,2.6) and (5.5,3.7) .. (8.2,4.2);
\node[acc, anchor=south east] at (5.6,3.9) {accelerating (data)};
% a few data points
\foreach \x/\y in {2.2/2.15,4.0/3.05,5.6/3.55,7.0/3.95}
  \fill[acc] (\x,\y) circle (1.6pt);
\end{tikzpicture}
$$

## Summary

A Type Ia supernova is the thermonuclear disruption of a carbon-oxygen white dwarf
brought to carbon ignition either by accretion toward the Chandrasekhar mass
(single-degenerate) or by merger (double-degenerate). Degenerate fuel burns in a
runaway because pressure is temperature-independent; a deflagration that transitions
to a detonation burns $\sim 0.6\,M_\odot$ to nickel-56 and unbinds the star. The light
curve is powered by ${}^{56}\mathrm{Ni}\to{}^{56}\mathrm{Co}\to{}^{56}\mathrm{Fe}$,
with the peak luminosity fixed by the nickel mass through Arnett's rule. The Phillips
width-luminosity relation reduces the residual scatter to $\sim 7\%$, making these
events standardizable candles. Their Hubble diagram, extended past $z \approx 1$,
revealed the accelerating expansion and remains a primary probe of dark energy.

[^co-ia]: Carroll & Ostlie, §18.6 and Ch. 15 — White dwarfs in binaries, the accretion and merger channels, carbon deflagration/detonation, and the radioactive light curve.