Core-Collapse Supernovae

A star born with more than about $8\,M_\odot$ spends its life fusing its way up the periodic table, and the reward for that labour is its own destruction. Each burning stage — hydrogen, helium, carbon, neon, oxygen, silicon — leaves an ash that becomes the next stage's fuel, building an onion-shell structure with the most tightly bound nuclei at the centre. That centre is an iron core, and iron is where the road ends: no fusion of iron releases energy, so the core has no way to generate the pressure that held the star up for millions of years. Supported only by the same relativistic electron degeneracy pressure that holds up a white dwarf, the iron core carries a Chandrasekhar-like maximum mass. Silicon burning in the overlying shell keeps raining fresh iron onto it, and the moment the core crosses that threshold, pressure loses to gravity everywhere at once. What follows is not a gentle contraction but a catastrophe measured in milliseconds. The core implodes in near free-fall, accelerated by two runaway processes — photodisintegration of iron by thermal gamma rays, which is endothermic, and electron capture, which removes the very electrons providing support. Collapse halts abruptly only when the inner core reaches nuclear density and the strong force stiffens matter into incompressibility; the inner core overshoots, rebounds, and launches a shock. The gravitational binding energy liberated, $\sim 3\times10^{53}\,\mathrm{erg}$ , is staggering — but $99\%$ of it leaves invisibly as neutrinos, and the entire drama of whether the star explodes hinges on depositing just a percent of that flux back into the overlying matter. SN 1987A, whose two dozen detected neutrinos arrived hours ahead of the light, proved the picture right.

Key takeaways

A core-collapse supernova ends the life of a star with initial mass $\gtrsim 8\,M_\odot$ when its inert iron core, supported by electron degeneracy pressure, exceeds an effective Chandrasekhar mass and collapses in under a second.
Iron is the endpoint of exothermic fusion — the highest binding energy per nucleon, $\approx 8.79\,\mathrm{MeV}$ for $^{56}\mathrm{Fe}$ — so the core has no nuclear energy left and collapse is unavoidable.
Two microphysical runaways drive it: photodisintegration $\gamma + {}^{56}\mathrm{Fe}\to 13\,{}^{4}\mathrm{He}+4n$ ( $Q=-124.4\,\mathrm{MeV}$ ) and electron capture $e^-+p\to n+\nu_e$ , both stripping pressure support.
Collapse halts at nuclear saturation density $\rho_{\rm nuc}\approx 2.7\times10^{14}\,\mathrm{g\,cm^{-3}}$ ; the inner core bounces and launches a shock that stalls, then is revived by delayed neutrino heating aided by convection and the SASI.
Roughly $99\%$ of the $\sim 3\times10^{53}\,\mathrm{erg}$ escapes as neutrinos, $\sim 1\%$ ( $\sim 10^{51}\,\mathrm{erg}$ ) becomes ejecta kinetic energy, and only $\sim 0.01\%$ becomes light — yet the peak brightness rivals a galaxy.
The remnant is a neutron star (or a black hole for the most massive progenitors); SN 1987A confirmed the neutrino-driven picture with 24 detected neutrinos.
Key idea: The gravitational binding energy released when a $\sim 1.5\,M_\odot$ core collapses from Earth-sized to $\sim 10\,\mathrm{km}$ is $\sim 3\times10^{53}\,\mathrm{erg}$ — a hundred times the entire radiated luminosity of the Sun over its lifetime. Almost all of it leaves as neutrinos; the supernova is what happens when just $\sim 1\%$ is deposited in the overlying matter.

Where this lives today

Core collapse is one of the most computationally demanding problems in astrophysics, and it is being solved now. State-of-the-art three-dimensional radiation-hydrodynamics codes — Fornax, CHIMERA, Aenus-Alcar, and others — evolve the full neutrino transport through the collapsing star and, for the first time in the past decade, routinely produce successful neutrino-driven explosions for a range of progenitors, reproducing explosion energies of $\sim 10^{51}\,\mathrm{erg}$ and $^{56}\mathrm{Ni}$ yields that match observations. The next Galactic supernova is the prize target of multi-messenger astronomy: Super-Kamiokande, IceCube, and the upcoming DUNE experiment would together record thousands to tens of thousands of neutrinos, resolving the collapse, bounce, and cooling in real time, while LIGO-Virgo-KAGRA listen for the $\sim 100$ – $1000\,\mathrm{Hz}$ gravitational-wave signature of the asymmetric bounce and turbulent convection. The SNEWS network stands ready to broadcast a neutrino alert hours before the optical brightening — exactly the head start SN 1987A demonstrated by accident. Meanwhile JWST dissects the ejecta and dust of SN 1987A and other remnants, the newly identified population of "failed" supernovae (stars that simply vanish) tests the black-hole-formation branch, and the neutrino-driven wind off the cooling proto-neutron star remains a live candidate site for forging some of the heaviest neutron-rich elements.

The physical picture

Picture the iron core as a white dwarf trapped inside a burning star. Around it sit concentric shells, each still fusing its fuel into the ash of the shell within — a stratified onion whose layers were laid down in sequence as the star aged. The stages run faster and faster as they climb the binding-energy curve, because each releases less energy per gram and the core must radiate ever harder to stay hot: hydrogen burning lasts millions of years, but silicon burning, the last stage, is over in about a day. The first figure lays out this structure and the timescale collapse it encodes.

Figure

The onion-shell structure of a massive star just before collapse. Successive burning stages ran faster and faster — hydrogen over ~10⁷ yr down to silicon in ~1 day — leaving an inert iron core supported by electron degeneracy pressure whose mass climbs toward the Chandrasekhar limit ≈ 1.44 M_⊙

The energy accounting is the heart of the physics, and the second figure is its punchline. When the core collapses from Earth-sized to a $\sim 10\,\mathrm{km}$ neutron star, it liberates $E_B\sim GM^2/R\sim 3\times10^{53}\,\mathrm{erg}$ . Set that against the $\sim 10^{51}\,\mathrm{erg}$ needed to unbind the overlying envelope and the $\sim 10^{49}\,\mathrm{erg}$ that ever emerges as visible light: energetically, the spectacular optical supernova is a rounding error. The neutrinos carry essentially the whole budget, and the entire theoretical difficulty is the coupling — how does a fraction of a percent of that neutrino luminosity get deposited efficiently enough to blow the star apart?

Figure

The energy budget of collapse. Liberating E_B ≈ (3/5)GM²/R ≈ 3×10⁵³ erg from a 1.5 M_⊙ core falling to R ≈ 10 km, roughly 99% escapes as neutrinos, ~1% (≈10⁵¹ erg) becomes ejecta kinetic energy, and only ~0.01% (≈10⁴⁹ erg) is radiated as light — the supernova we see is the least energetic channel

Neutrinos, ordinarily the most weakly interacting particles known, become the protagonists precisely because the density is so extreme. Once $\rho \gtrsim 10^{12}\,\mathrm{g\,cm^{-3}}$ even neutrinos are trapped — their diffusion time out of the core exceeds the collapse time — so the nascent proto-neutron star radiates them slowly over seconds from a "neutrinosphere," exactly as a hot body radiates photons from its photosphere. That is why the collapse takes milliseconds but the neutrino burst lasts seconds. And the explosion is irreducibly three-dimensional: in strict spherical symmetry, simulations of ordinary progenitors generically fail to explode. Turbulent convection and the standing-accretion-shock instability (SASI) let matter dwell longer in the heating region and boost the effective energy deposition, tipping many models over into a successful — and characteristically lopsided — explosion that kicks the newborn neutron star across the galaxy.

The core equations

The single most important quantity is the gravitational binding energy liberated when the core collapses. Treating the proto-neutron star as a uniform sphere, this energy sets the entire scale of the event — the neutrino luminosity, the explosion energy, and the light curve are all fractions of it.

E_B \approx \frac{3}{5}\,\frac{G M^2}{R} \;\sim\; 3\times10^{53}\,\mathrm{erg} \qquad (1)

where: $E_B$ — gravitational binding energy liberated; $G$ — Newton's constant, $6.674\times10^{-8}\,\mathrm{cm^3\,g^{-1}\,s^{-2}}$ ; $M$ — proto-neutron-star mass, $\approx 1.5\,M_\odot$ ; $R$ — final radius, $\approx 10\,\mathrm{km}$ . Reading it: shrink a solar-mass-and-a-half from planetary to city scale and gravity hands you $\sim 10^{53}\,\mathrm{erg}$ — the master energy of the whole phenomenon. The core collapses because it lives at a fixed maximum mass. Like a white dwarf, the iron core is held up by relativistic electron degeneracy pressure, and that pressure allows only a Chandrasekhar mass set by fundamental constants and the number of electrons per nucleon.

M_{\rm Ch} \approx 1.457\left(\frac{2}{\mu_e}\right)^2 M_\odot \qquad (2)

where: $M_{\rm Ch}$ — maximum mass supportable by relativistic electron degeneracy pressure; $\mu_e$ — mean molecular weight per electron, $\approx 2$ for iron-peak matter. Reading it: the effective threshold is raised a little by finite-temperature entropy and lowered by electron capture, landing near $1.3$ – $1.5\,M_\odot$ in realistic cores; silicon burning drives the core across it. Once support fails, nothing resists gravity and the core falls essentially in free-fall. The relevant clock is the gravitational free-fall time, which depends only on density and shortens dramatically as the collapse proceeds.

\tau_{\rm ff} = \sqrt{\frac{3\pi}{32\, G\rho}} \;\approx\; \frac{2\times10^{3}}{\sqrt{\rho/(\mathrm{g\,cm^{-3}})}}\ \mathrm{s} \qquad (3)

where: $\tau_{\rm ff}$ — gravitational free-fall time; $\rho$ — mean density of the collapsing core. Reading it: at $\rho\sim 10^{9}\,\mathrm{g\,cm^{-3}}$ (the iron core) $\tau_{\rm ff}\sim 0.06\,\mathrm{s}$ , but by $\rho\sim 10^{12}\,\mathrm{g\,cm^{-3}}$ it is $\sim 2\,\mathrm{ms}$ — the centre runs away far faster than pressure waves can respond. Two microphysical processes actively worsen the collapse. The first is photodisintegration: the same thermal gamma rays that fill a $10^{10}\,\mathrm{K}$ core carry enough energy to shatter iron back into alpha particles and neutrons, undoing the star's entire nuclear history and stealing thermal pressure.

\gamma + {}^{56}\mathrm{Fe} \;\longrightarrow\; 13\,{}^{4}\mathrm{He} + 4n \qquad (Q = -124.4\ \mathrm{MeV}) \qquad (4)

where: $\gamma$ — thermal photon at $T\sim 10^{10}\,\mathrm{K}$ ; $Q<0$ — the reaction absorbs energy; the alpha particles disintegrate further, $\gamma+{}^{4}\mathrm{He}\to 2p+2n$ at $Q=-28.3\,\mathrm{MeV}$ . Reading it: every erg spent breaking nuclei is an erg not supporting the core, so photodisintegration acts as a thermal sink that softens the equation of state. The second is electron capture, which removes the degenerate electrons that provide almost all of the pressure and converts protons to neutrons, releasing a flood of electron neutrinos.

e^- + p \;\longrightarrow\; n + \nu_e, \qquad e^- + (Z,A)\;\longrightarrow\;(Z\!-\!1,A)+\nu_e \qquad (5)

where: $e^-$ — degenerate electron providing support; $\nu_e$ — electron neutrino carrying off energy and lepton number; $(Z,A)$ — heavy nucleus capturing an electron. Reading it: the electron fraction $Y_e=1/\mu_e$ falls from $\sim 0.5$ toward $\sim 0.3$ ; fewer electrons means less pressure, tightening the noose. The collapse stops only when the strong nuclear force asserts itself. At nuclear saturation density the repulsive short-range core of the nucleon-nucleon interaction makes matter suddenly stiff, the adiabatic index jumps above $4/3$ , and the inner core rebounds.

\rho_{\rm nuc} \approx 2.7\times10^{14}\,\mathrm{g\,cm^{-3}} \;\Longrightarrow\; \text{core bounce} \qquad (6)

where: $\rho_{\rm nuc}$ — nuclear saturation density where the equation of state stiffens. Reading it: the inner $\sim 0.5\,M_\odot$ homologous core overshoots this density, springs back, and drives a shock wave outward into the still-infalling outer core — the origin of the "bounce shock." The proto-neutron star then cools by neutrino emission, and its luminosity follows directly from the binding-energy budget spread over the diffusion time.

L_\nu \sim \frac{E_B}{\Delta t} \sim \frac{3\times10^{53}\,\mathrm{erg}}{10\,\mathrm{s}} \sim 3\times10^{52}\,\mathrm{erg\,s^{-1}} \qquad (7)

where: $L_\nu$ — total neutrino luminosity summed over all six species; $\Delta t$ — Kelvin-Helmholtz cooling time of the proto-neutron star, $\sim 10\,\mathrm{s}$ . Reading it: for those few seconds the object outshines, in neutrinos, the combined photon output of every star in the observable universe. Finally, the light we actually see is not powered by the explosion at all but by radioactivity. The shock synthesizes $^{56}\mathrm{Ni}$ , whose decay chain deposits gamma rays that thermalize in the expanding ejecta and light the late-time curve.

^{56}\mathrm{Ni} \xrightarrow{6.1\,\mathrm{d}} {}^{56}\mathrm{Co} \xrightarrow{77\,\mathrm{d}} {}^{56}\mathrm{Fe} \qquad (8)

where: $^{56}\mathrm{Ni}$ — nickel-56 freshly synthesized, $\sim 0.01$ – $0.1\,M_\odot$ ; the labels are the decay half-lives (Ni, $6.1\,\mathrm{d}$ ; Co, $77.2\,\mathrm{d}$ ). Reading it: after shock breakout the light curve declines on the $^{56}\mathrm{Co}$ slope of $\sim 0.01\,\mathrm{mag\,day^{-1}}$ — the same isotope that powers Type Ia supernovae, but here in far smaller quantity.

Derivation AWhy the iron core must collapse

The goal is to show that an iron core supported by relativistic electron degeneracy sits exactly at the knife-edge of dynamical stability, and that the two microphysical processes operating inside it — photodisintegration and electron capture — each push it over that edge. The universal criterion is that a self-gravitating gas is stable only if its effective adiabatic index $\Gamma$ exceeds $4/3$ ; we will derive that threshold from how internal and gravitational energy respond to compression, locate the relativistic degenerate gas precisely at $\Gamma=4/3$ , and then argue that both runaways lower $\Gamma_{\rm eff}$ below it, guaranteeing collapse.

Step 1. Model the core as a polytrope and track its two energies under a uniform compression that raises the density $\rho$ . The internal (pressure) energy scales as $U\propto PV\propto\rho^{\Gamma-1}$ , while the gravitational energy scales as $\Omega\propto -GM^2/R\propto -\rho^{1/3}$ since $R\propto\rho^{-1/3}$ at fixed mass. Why: reducing the whole problem to how each energy responds to a squeeze is what makes the stability question a simple comparison of exponents.

Step 2. Ask whether a small compression is resisted or amplified. Stability requires the internal energy to rise at least as fast as the gravitational energy falls, i.e. the exponent inequality $\Gamma-1>1/3$ , which is exactly $\Gamma>4/3$ . If $\Gamma<4/3$ , compression releases net energy and runs away catastrophically. Why: this is the origin of the famous $4/3$ threshold — it is nothing but the statement that pressure must stiffen faster than gravity deepens the well.

Step 3. Evaluate $\Gamma$ for the actual supporting gas. An ultra-relativistic degenerate electron gas has $P\propto\rho^{4/3}$ , so $\Gamma=4/3$ exactly. The iron core therefore sits at the margin of stability with no safety factor whatsoever. Why: this is why massive stars are doomed — their support mechanism is neutrally stable, so any destabilizing perturbation, however small, tips it into collapse rather than being damped.

Step 4. Add photodisintegration. When gamma rays break iron into alphas and nucleons, energy delivered by a compression goes into raising the internal energy less than it should, because part of it is diverted into dissociation. The effective index $\Gamma_{\rm eff}$ therefore drops below $4/3$ . Why: an energy sink that opens up under compression is precisely what softens the equation of state below the stability threshold.

Step 5. Add electron capture. Removing degenerate electrons lowers $Y_e$ and directly reduces the degeneracy pressure that supplies $\Gamma$ , again pushing $\Gamma_{\rm eff}$ below $4/3$ and reinforcing the instability begun by photodisintegration. Why: both processes attack the same margin from different directions, which is why the collapse, once it begins, is so violently unavoidable. The standard $4/3$ result follows from the virial theorem for a polytrope and is derived in Carroll and Ostlie, ch. 15.

Go deeper · Derivation B · 1 minThe Chandrasekhar mass from dimensional analysis

Derivation BThe Chandrasekhar mass from dimensional analysis

The aim is to obtain the maximum mass an iron core can carry — the Chandrasekhar mass — from nothing but the balance between relativistic electron degeneracy pressure and gravity, and to expose the remarkable fact that this balance fixes a mass while leaving the radius entirely undetermined. We will write both pressures as powers of density, discover that both scale as $\rho^{4/3}$ in the relativistic limit, watch the density and radius cancel, and read off a mass built purely from $\hbar$ , $c$ , $G$ , and the mass per electron.

Step 1. Write the relativistic degeneracy pressure. The electrons fill momentum space to a Fermi momentum $p_F\sim\hbar\,n_e^{1/3}$ , and in the ultra-relativistic limit each has energy $\sim p_F c$ , giving $P_{\rm deg}\sim\hbar c\,n_e^{4/3}\sim\hbar c\,(\rho/\mu_e m_H)^{4/3}$ . Why: it is the $4/3$ power specifically — the relativistic softening of the electron gas — that makes a maximum mass exist at all.

Step 2. Write the gravitational pressure needed for equilibrium. From hydrostatic balance $P_{\rm grav}\sim GM^2/R^4$ , and substituting $\rho\sim M/R^3$ to eliminate $R$ gives $P_{\rm grav}\sim G\,M^{2/3}\rho^{4/3}$ . Why: casting gravity in the same $\rho^{4/3}$ form as the pressure is what lets us compare them directly and see the radius drop out.

Step 3. Balance the two. Setting $P_{\rm deg}\sim P_{\rm grav}$ gives $\hbar c\,(\rho/\mu_e m_H)^{4/3}\sim G\,M^{2/3}\rho^{4/3}$ , and the common factor $\rho^{4/3}$ cancels entirely. Why: the cancellation is the whole punchline — because both sides carry the same density dependence, equilibrium cannot fix a density or a radius, only a mass.

Step 4. Solve for the mass. Rearranging yields $M\sim(\hbar c/G)^{3/2}(\mu_e m_H)^{-2}\approx 1.4\,(2/\mu_e)^2\,M_\odot$ , a pure combination of fundamental constants. Why: the mass limit is not an accident of iron chemistry but a deep consequence of quantum statistics and gravity; the exact coefficient $5.83/\mu_e^2$ requires the full Lane-Emden solution, giving $1.44\,M_\odot$ for $\mu_e=2$ .

Go deeper · Derivation C · 1 minWhen neutrinos become trapped

Derivation CWhen neutrinos become trapped

The purpose here is to explain the single most important timescale mismatch of the whole event — why the collapse lasts milliseconds but the neutrino burst lasts seconds — by finding the density at which neutrinos stop streaming freely out of the core and start diffusing. We will estimate the neutrino-nucleus cross section, which is anomalously large because coherent scattering off heavy nuclei enhances it, compute the mean free path, and impose the condition that it shrink below the core radius, arriving at a trapping density of $\sim 10^{12}\,\mathrm{g\,cm^{-3}}$ .

Step 1. Estimate the cross section. Coherent neutral-current scattering off a nucleus of mass number $A$ scales as $\sigma_\nu\sim\sigma_0\,(E_\nu/m_e c^2)^2 A^2$ , enhanced by both the $E_\nu^2$ weak-interaction factor and the coherent $A^2$ factor. Why: it is this $A^2$ coherent enhancement that makes neutrinos — famously able to pass through light-years of lead — nonetheless opaque inside a collapsing core.

Step 2. Form the mean free path from the number density of scatterers, $n_{\rm nuc}=\rho/(A m_H)$ , giving $\lambda_\nu=1/(n_{\rm nuc}\sigma_\nu)$ . As the density climbs during collapse, $\lambda_\nu$ falls steeply. Why: the mean free path is the physical link between the growing density and the moment neutrinos can no longer escape in a straight line.

Step 3. Impose the trapping condition $\lambda_\nu\lesssim R_{\rm core}\sim 10^{7}\,\mathrm{cm}$ . Evaluating the cross section and density together, this is met at $\rho_{\rm trap}\sim 10^{11}$ – $10^{12}\,\mathrm{g\,cm^{-3}}$ . Below this the neutrinos stream out freely, deleptonizing the core; above it they are advected inward with the flow. Why: fixing the exact density at which streaming turns to diffusion is what pins down when lepton number gets locked into the core.

Step 4. Follow the consequence. Trapped neutrinos leak out only on the Kelvin-Helmholtz diffusion time of the hot proto-neutron star, $\sim 10\,\mathrm{s}$ , far longer than the millisecond collapse. Why: this diffusion bottleneck is exactly why the observed SN 1987A burst lasted about 13 seconds rather than a millisecond — the neutrinos had to random-walk out. The trapping density is the standard value quoted by Bethe (1990).

Limiting cases

The outcome of core collapse is not one story but a family of them, selected by the progenitor's initial mass, metallicity, rotation, and binary history:

Case	Result	Meaning
Electron-capture SN	$M_{\rm ZAMS}\sim 8$ – $10\,M_\odot$	An O-Ne-Mg core (not iron) collapses via electron capture onto $^{24}\mathrm{Mg}$ and $^{20}\mathrm{Ne}$ before silicon ignites, giving a low-energy explosion and a low-mass neutron star.
Standard iron-core collapse	$M_{\rm ZAMS}\sim 10$ – $25\,M_\odot$	Classic neutrino-driven explosion leaving a neutron star of gravitational mass $\sim 1.2$ – $1.6\,M_\odot$ ; the canonical Type II-P supernova.
Black hole / fallback	$M_{\rm ZAMS}\gtrsim 25\,M_\odot$	Weak or failed explosion; substantial fallback pushes the remnant past the neutron-star maximum mass, forming a black hole, sometimes with a faint or absent optical transient.
Pair-instability	$M_{\rm core}\sim 64$ – $133\,M_\odot$	Electron-positron pair production softens the equation of state ( $\Gamma<4/3$ ), igniting explosive oxygen burning that disrupts the whole star with no remnant — a distinct mechanism, not iron-core collapse.

Common trapA core-collapse supernova is a giant thermonuclear bomb

It is not. The collapse is a gravitational implosion, and fusing iron would absorb energy, not release it — there is no runaway burning that unbinds the star. (Type Ia supernovae, by contrast, genuinely are thermonuclear, from a white dwarf detonating carbon and oxygen.) The energy source here is gravity: $\sim 3\times10^{53}\,\mathrm{erg}$ of binding energy released as the core shrinks to nuclear density, of which the visible explosion claims barely a percent. A second, subtler trap is imagining the bounce shock alone blows up the star. In realistic simulations the prompt shock always stalls within milliseconds, drained by dissociating the infalling iron it plows through and by neutrino losses; reviving it demands delayed neutrino heating plus the multidimensional hydrodynamic instabilities (convection, SASI) that a spherical model cannot capture.

Worked examples

1 · The energy budget of collapse (fully worked)

A $1.5\,M_\odot$ iron core collapses to a neutron star of radius $R=10\,\mathrm{km}$ . Estimate the gravitational binding energy released, then partition it into neutrinos, kinetic energy, and light, and compare the total to the Sun's entire radiative output.

Step 1. Insert numbers into equation (1). With $M=1.5\,M_\odot=3.0\times10^{33}\,\mathrm{g}$ and $R=10^{6}\,\mathrm{cm}$ , $E_B\approx 0.6\times(6.674\times10^{-8})\times(3.0\times10^{33})^2/(10^{6})$ . Evaluate the pieces: $(3.0\times10^{33})^2=9.0\times10^{66}$ , times $6.674\times10^{-8}$ gives $6.0\times10^{59}$ , divided by $10^{6}$ gives $6.0\times10^{53}$ , times $0.6$ gives $E_B\approx 3.6\times10^{53}\,\mathrm{erg}$ — the canonical few $\times10^{53}\,\mathrm{erg}$ .

Step 2. Partition it. Roughly $99\%$ leaves as neutrinos, so $E_\nu\approx 3.6\times10^{53}\,\mathrm{erg}$ ; about $1\%$ becomes ejecta kinetic energy, $E_{\rm kin}\approx 3.6\times10^{51}\,\mathrm{erg}$ (rounded to the standard $\sim 10^{51}\,\mathrm{erg}$ , one "bethe" or "foe"); and only $\sim 0.01\%$ ever emerges as light, $E_{\rm light}\approx 3.6\times10^{49}\,\mathrm{erg}\approx$ $10^{49}\,\mathrm{erg}$ .

Step 3. Compare to the Sun. The Sun radiates $L_\odot=3.8\times10^{33}\,\mathrm{erg\,s^{-1}}$ for $\sim 10^{10}\,\mathrm{yr}=3.2\times10^{17}\,\mathrm{s}$ , a lifetime output of $\sim 1.2\times10^{51}\,\mathrm{erg}$ . So $E_B$ exceeds the Sun's entire life's luminous work by a factor of a few hundred — yet the visible supernova, $\sim 10^{49}\,\mathrm{erg}$ , is only about $1\%$ of even that solar total. The neutrinos carry $\sim 300$ solar lifetimes of energy in ten seconds; the light we see is an afterthought.

2 · Can photodisintegration really drain the shock?

A stalled shock carries roughly the explosion energy, $\sim 10^{51}\,\mathrm{erg}$ . Estimate the energy cost of photodisintegrating a full $1.4\,M_\odot$ iron core back into free nucleons, and compare.

Step 1. The disintegration $\gamma+{}^{56}\mathrm{Fe}\to 13\,{}^{4}\mathrm{He}+4n$ costs $124.4\,\mathrm{MeV}$ per iron nucleus of $56$ nucleons, i.e. $124.4/56\approx 2.22\,\mathrm{MeV}$ per nucleon (complete breakup to nucleons costs a bit more, but this captures the scale).

Step 2. Count the nucleons in $1.4\,M_\odot$ : $N=1.4\times1.989\times10^{33}\,\mathrm{g}/(1.66\times10^{-24}\,\mathrm{g})\approx 1.68\times10^{57}$ nucleons.

Step 3. Multiply and convert. Energy $=1.68\times10^{57}\times 2.22\,\mathrm{MeV}\times 1.602\times10^{-6}\,\mathrm{erg/MeV}\approx$ $6\times10^{51}\,\mathrm{erg}$ . That is several times the entire explosion energy — so a shock that has to fight its way out through even a fraction of the iron core burns its whole energy budget dissociating nuclei. This is quantitatively why the prompt shock stalls and why neutrino re-heating is indispensable.

3 · How fast does the core fall?

Evaluate the free-fall time, equation (3), at the density of the iron core and near neutrino trapping.

Step 1. At the iron core, $\rho\sim 10^{9}\,\mathrm{g\,cm^{-3}}$ : $\tau_{\rm ff}\approx (2\times10^{3})/\sqrt{10^{9}}=(2\times10^{3})/(3.16\times10^{4})\approx$ $0.06\,\mathrm{s}$ .

Step 2. Near trapping, $\rho\sim 10^{12}\,\mathrm{g\,cm^{-3}}$ : $\tau_{\rm ff}\approx (2\times10^{3})/\sqrt{10^{12}}=(2\times10^{3})/(10^{6})\approx$ $2\,\mathrm{ms}$ . The collapse accelerates as it deepens: by the time the centre is dense enough to trap neutrinos, the local dynamical time is milliseconds, and no pressure wave can cross the core fast enough to arrest it. Collapse is essentially free-fall — a genuine implosion, not a slow settling.

Physics of the portrait

The core-collapse supernova is where a dozen strands of twentieth-century physics converge — nuclear binding, quantum degeneracy, the weak interaction, general relativity, and computational hydrodynamics — and no single person owns it. The portrait below traces the idea itself, from a reckless prophetic guess in 1934 to the multi-messenger science of today.

Origins · 1 mincore-collapse supernova and how this was discovered

core-collapse supernova

the gravitational collapse of an iron core into a neutron star or black hole · nuclear astrophysics and neutrino-driven explosions

The concept was born in a leap of astonishing prescience. In 1934, barely two years after the neutron's discovery, Walter Baade and Fritz Zwicky coined the word "supernova" and proposed — in a single paragraph — that these events mark "the transition of an ordinary star into a neutron star," powered by the gravitational binding energy released, and that they might be the source of cosmic rays. Every element of that guess proved correct decades before it could be tested. The energetics were sharpened once Subrahmanyan Chandrasekhar's maximum mass (1931) made clear that a degenerate iron core could not be held up indefinitely, and once J. Robert Oppenheimer and George Volkoff (1939) showed a neutron star has its own mass limit. The mechanism took longer. Stirling Colgate and Richard White argued in 1966 that neutrinos, not the bounce shock, must deposit the energy that drives the explosion — a radical claim when neutrino cross sections were barely known. Hans Bethe and James Wilson formalized the modern delayed neutrino-heating mechanism in 1985, distinguishing the failed prompt shock from a shock revived seconds later by neutrino absorption in the gain region. The decisive confirmation came in February 1987, when detectors on three continents caught two dozen neutrinos from SN 1987A, matching the predicted $\sim 3\times10^{53}\,\mathrm{erg}$ and seconds-long duration almost exactly. Since then the frontier has been computational: the realization, led by Hans-Thomas Janka and others, that the explosion is intrinsically multidimensional — that convection and the standing-accretion-shock instability are not corrections but essential ingredients — has turned a spherical thought experiment into a three-dimensional simulation problem still being solved on the world's largest supercomputers.

History

1934

— Baade and Zwicky coin "supernova" and propose the neutron-star remnant powered by gravitational collapse

1939

— Oppenheimer and Volkoff compute the maximum mass of a neutron star

1966

— Colgate and White propose that neutrinos drive the explosion

1985

— Bethe and Wilson formulate the delayed neutrino-heating mechanism

1987

— Neutrinos from SN 1987A detected by Kamiokande-II, IMB, and Baksan

Going further

Texts. Carroll and Ostlie, An Introduction to Modern Astrophysics, 2nd ed., ch. 15 (The Fate of Massive Stars) · Binney and Tremaine, Galactic Dynamics, 2nd ed. (dynamical timescales and virial energetics) · Ryden, Introduction to Cosmology (supernovae as standard candles and metal enrichment) · Dodelson, Modern Cosmology (neutrino physics and thermal-history context) · Bethe, Rev. Mod. Phys. 62, 801 (1990), Supernova mechanisms · Janka, Ann. Rev. Nucl. Part. Sci. 62, 407 (2012), Explosion mechanisms of core-collapse supernovae.

Provenance. A verifier should recompute: the binding energy $E_B\approx\tfrac{3}{5}GM^2/R\approx 3.6\times10^{53}\,\mathrm{erg}$ for $M=1.5\,M_\odot$ , $R=10\,\mathrm{km}$ ; the Chandrasekhar mass $M_{\rm Ch}=5.83/\mu_e^2\,M_\odot=1.44\,M_\odot$ for $\mu_e=2$ ; the free-fall time $\tau_{\rm ff}=\sqrt{3\pi/(32G\rho)}$ , giving $0.06\,\mathrm{s}$ at $10^{9}\,\mathrm{g\,cm^{-3}}$ and $2\,\mathrm{ms}$ at $10^{12}\,\mathrm{g\,cm^{-3}}$ ; the photodisintegration $Q$ -values ( $-124.4\,\mathrm{MeV}$ for $^{56}\mathrm{Fe}\to 13\alpha+4n$ , from $13\times28.3-492.3$ ; and $-28.3\,\mathrm{MeV}$ for $^4\mathrm{He}\to 2p+2n$ ); the full-core dissociation cost $\sim 6\times10^{51}\,\mathrm{erg}$ for a $1.4\,M_\odot$ iron core; nuclear saturation density $2.7\times10^{14}\,\mathrm{g\,cm^{-3}}$ ; and the $^{56}\mathrm{Ni}$ ( $6.1\,\mathrm{d}$ ) / $^{56}\mathrm{Co}$ ( $77.2\,\mathrm{d}$ ) half-lives. Standard derivations are in Carroll and Ostlie ch. 15; mechanism details in Bethe RMP 62, 801 (1990) and Janka ARNPS 62, 407 (2012). SN 1987A neutrino counts (Kamiokande-II 11, IMB 8, Baksan 5) are historical detections.

Prerequisites. Advanced Burning Stages in Massive Stars · White Dwarfs and Electron Degeneracy Pressure · Nuclear Reaction Rates and the Gamow Peak · Helium Burning and the Triple-Alpha Process. Leads to. Neutron Stars and Pulsars · Stellar-Mass Black Holes · Nucleosynthesis and the r-process · Supernova Remnants and Cosmic Rays.

Ask the tutor: Ask me to walk through why the shock stalls and exactly how delayed neutrino heating and the SASI conspire to revive it.

If iron cannot release fusion energy, why does the core keep getting hotter and denser instead of just sitting there?

Because the surrounding silicon shell keeps burning and dumping fresh iron onto the core, and the core has no way to radiate away the resulting compression. Every gram of iron added deepens the gravitational well, and with no nuclear energy source to generate offsetting pressure, the core contracts and heats. That heating is self-defeating: hotter photons begin photodisintegrating the iron, which absorbs energy, and the rising density favours electron capture, which removes pressure. So the core does not "sit there" — it is driven relentlessly toward the Chandrasekhar limit, at which point contraction becomes catastrophic collapse.

If 99% of the energy comes out as neutrinos, why is it so hard to make the star explode?

Precisely because neutrinos interact so weakly. The core releases a colossal $\sim 3\times10^{53}\,\mathrm{erg}$ , but the overlying matter is nearly transparent to it — the vast majority of neutrinos stream straight out without depositing anything. To unbind the envelope needs only $\sim 10^{51}\,\mathrm{erg}$ , about one percent of the flux, yet capturing even that percent is delicate: it requires the neutrinos to be absorbed in a specific "gain region" behind the stalled shock, where heating exceeds cooling. In spherical symmetry this often fails narrowly. The multidimensional turbulence that lets matter linger longer in the heating zone is what usually tips the balance to success.

What is the difference between Type II, Type Ib, and Type Ic supernovae?

All three are core-collapse events; the labels are spectroscopic, reflecting what envelope the progenitor kept. Type II supernovae show hydrogen lines — the star retained its hydrogen envelope, typically a red supergiant. Type Ib show no hydrogen but do show helium; the progenitor lost its hydrogen layer to strong winds or a binary companion. Type Ic show neither hydrogen nor helium, from a star stripped of both outer layers. The underlying explosion is the same iron-core collapse in every case; only the outermost skin differs. (Type Ia, confusingly grouped by the old "I means no hydrogen" naming, is a completely different, thermonuclear, white-dwarf event.)

What decides whether the remnant is a neutron star or a black hole?

Chiefly the progenitor's mass, structure, and how much matter falls back after the shock is launched. Stars of roughly $10$ – $25\,M_\odot$ typically explode successfully and leave a neutron star of $\sim 1.2$ – $1.6\,M_\odot$ . Above $\sim 25\,M_\odot$ the explosion is often weak or fails outright; the shock cannot clear the deep, dense envelope, matter rains back onto the proto-neutron star, and once it exceeds the neutron-star maximum mass (a few solar masses) it collapses to a black hole. But the mapping is not simply monotonic in mass — the detailed density structure of the core, the "compactness," matters, and some stars in this range may collapse quietly to black holes with little or no visible transient.

Why did the neutrinos from SN 1987A arrive before the light?

Because they are made at the very moment of collapse, deep in the core, and escape almost immediately, whereas the light comes from the shock reaching the star's surface much later. The neutrino burst is emitted as the core deleptonizes and the proto-neutron star cools — right at bounce. Those neutrinos, once free of the neutrinosphere, travel to us at essentially light speed. The shock wave, by contrast, must climb outward through the entire stellar envelope, taking hours to break out and finally release photons. For SN 1987A the neutrinos preceded the optical brightening by about three hours — not because they travel faster than light, but because they had a head start of the whole envelope-crossing time.

Does the core actually bounce, and is that bounce the explosion?

The bounce is real but is not, by itself, the explosion. When the inner core overshoots nuclear density, the stiffened matter rebounds like a compressed spring and launches a genuine shock wave outward. But that prompt shock does not make it out: as it plows into the still-infalling outer core it spends its energy dissociating iron into nucleons and radiating neutrinos, and within milliseconds it stalls into a stationary "standing accretion shock." What we call the explosion is the later revival of this stalled shock by neutrino heating over the following tenths of a second. So the bounce sets the stage and creates the shock, but the star only explodes if neutrinos, helped by multidimensional turbulence, re-energize it.

Are core-collapse supernovae important for the chemical makeup of the universe?

Enormously. As the shock passes through the silicon and oxygen shells it burns them explosively to iron-peak elements near nuclear statistical equilibrium, producing much of the cosmic $^{56}\mathrm{Ni}$ that decays to $^{56}\mathrm{Fe}$ , along with most of the oxygen, neon, and magnesium that dominate the galaxy's metal budget. Because their massive progenitors are short-lived, core-collapse supernovae enrich the interstellar medium quickly, seeding each new generation of stars and planets with heavy elements. The neutrino-driven wind off the proto-neutron star may also help forge some of the heaviest neutron-rich isotopes. In a real sense the oxygen you breathe and the calcium in your bones were made in these explosions.