Case Study 18.2 — From Resonances to Stars: How Compound Nuclear Reactions Build the Elements

The Problem

Look at the periodic table. Every element heavier than iron — copper, silver, gold, platinum, uranium — was produced by neutron capture reactions in extreme astrophysical environments. The nuclear physics that governs these reactions is the same compound nucleus theory developed in this chapter: Breit-Wigner resonances, level densities, Hauser-Feshbach cross sections. But the astrophysical context adds urgency: the precise values of these nuclear quantities, for thousands of nuclei, many of them unstable and never measured in a laboratory, determine the chemical composition of the universe.

This case study traces the connection from individual nuclear resonances to the observed abundances of the heaviest elements.

The Abundance Pattern: Evidence for Two Processes

The Solar System abundances of heavy elements, determined from meteoritic and spectroscopic data, show a distinctive pattern: two peaks appear at each magic neutron number ($N = 50, 82, 126$), offset from each other by a few mass units. The heavier peak in each pair lines up with the stable isotopes at the magic numbers; the lighter peak corresponds to the neutron-rich progenitors that beta-decay to stability.

This double-peaked structure was explained by Burbidge, Burbidge, Fowler, and Hoyle (1957, "B$^2$FH") and independently by Cameron (1957): two distinct neutron capture processes operate in different astrophysical environments.

The $s$-Process (Slow Neutron Capture)

Site: Thermally pulsing asymptotic giant branch (AGB) stars — red giants with masses $\sim 1$–$8 M_\odot$ in the late stages of evolution.

Neutron source: ${}^{13}$C($\alpha$, n)${}^{16}$O (main component, $kT \approx 8$ keV) and ${}^{22}$Ne($\alpha$, n)${}^{25}$Mg (marginal component, $kT \approx 23$–$26$ keV).

Neutron density: $n_n \sim 10^7$–$10^{11}$ cm$^{-3}$ — low enough that the mean time between captures ($\tau_{\text{cap}} \sim$ months to thousands of years) is typically much longer than beta-decay half-lives of unstable isotopes on the path. The $s$-process path therefore follows the valley of stability.

Key nuclear physics input: The Maxwellian-averaged neutron capture cross sections $\langle\sigma\rangle$ at stellar temperatures ($kT = 5$–$90$ keV). These are computed from the resolved resonance parameters (Breit-Wigner) and the statistical model (Hauser-Feshbach) for each isotope on the $s$-process path.

The $\sigma N_s$ Systematics:

In the steady-flow approximation, the $s$-process abundance $N_s$ of isotope $i$ on the main $s$-process path satisfies:

$$\langle\sigma\rangle_i N_{s,i} \approx \text{const}$$

This means isotopes with small neutron capture cross sections (such as magic-number nuclei) accumulate in high abundances. The $s$-process abundance peaks at:

The cross section of ${}^{208}$Pb (doubly magic, $Z = 82$, $N = 126$) is only 0.36 mb at $kT = 30$ keV — two orders of magnitude smaller than typical non-magic nuclei. This nucleus is the terminus of the $s$-process; almost all $s$-process material piles up at ${}^{208}$Pb, which captures a neutron, forms ${}^{209}$Pb, which beta-decays to ${}^{209}$Bi. Further capture leads to alpha decay (via ${}^{210}$Po), recycling material back to Pb.

The $r$-Process (Rapid Neutron Capture)

Site: Neutron star mergers (confirmed by GW170817 in 2017) and possibly certain core-collapse supernovae (collapsar disk winds, magnetically driven jets).

Neutron density: $n_n \sim 10^{20}$–$10^{28}$ cm$^{-3}$ — so high that neutron capture is much faster than beta decay ($\tau_{\text{cap}} \ll \tau_\beta$). The $r$-process path runs far to the neutron-rich side of stability, where nuclei have never been observed in the laboratory.

Key nuclear physics input: Neutron capture cross sections (Hauser-Feshbach, since no experimental data exist for most $r$-process nuclei), nuclear masses (which determine neutron separation energies and hence the balance between capture and photodisintegration), beta-decay half-lives, and fission properties.

Level densities are critical: For the exotic nuclei on the $r$-process path, the Hauser-Feshbach cross sections depend sensitively on the level density at the neutron separation energy. The Fermi gas model with $a \approx A/8$ is the starting point, but corrections for shell effects, deformation, and pairing — all poorly constrained far from stability — introduce large uncertainties.

The GW170817 Connection

On August 17, 2017, the LIGO-Virgo gravitational wave detectors observed the inspiral and merger of two neutron stars, designated GW170817. Within 1.7 seconds, the Fermi Gamma-ray Burst Monitor detected a short gamma-ray burst (GRB 170817A). Over the following days and weeks, ground- and space-based telescopes observed an electromagnetic counterpart — a kilonova — whose infrared brightness and spectral evolution matched theoretical predictions for the radioactive decay of freshly synthesized $r$-process elements.

The kilonova light curve analysis indicated that $\sim 0.05 M_\odot$ of $r$-process material was ejected, with a composition rich in lanthanides (elements 57–71, which have large atomic opacities and produce the characteristic red/infrared kilonova emission).

The connection to this chapter is direct: the nuclear physics that determines the kilonova light curve includes:

1. Neutron capture cross sections — computed from the Hauser-Feshbach model using level densities (Fermi gas model plus corrections) and optical model transmission coefficients.
2. Beta-decay rates — which determine when nuclei on the $r$-process path beta-decay toward stability.
3. Fission properties — particularly for the heaviest nuclei ($A > 250$), where fission recycling redistributes material to lighter masses.

The theoretical modeling of GW170817 pushed the compound nucleus model to its limits: predictions for thousands of unmeasured nuclei, using level densities extrapolated far from the experimentally constrained region.

The Role of Nuclear Data Facilities

Measuring $s$-Process Cross Sections

The $s$-process demands precise neutron capture cross sections at stellar energies ($kT = 5$–$90$ keV, corresponding to neutron energies of roughly 1 keV to 300 keV). Two experimental approaches dominate:

1. Time-of-flight measurements at facilities like n_TOF (CERN) and GELINA (Belgium): A pulsed neutron source provides a broad energy spectrum, and individual resonances are resolved and fitted with R-matrix codes. The Maxwellian average is then computed from the resonance parameters.

2. Activation measurements at quasi-Maxwellian neutron sources: Facilities like Karlsruhe (Germany) produce a neutron spectrum approximating a Maxwellian at $kT = 25$ keV using the ${}^{7}$Li(p,n) reaction just above threshold. The sample is irradiated, and the capture cross section is inferred from the induced activity. This method is particularly useful for radioactive targets.

Constraining $r$-Process Inputs at FRIB

The Facility for Rare Isotope Beams (FRIB) at Michigan State University, which began operations in 2022, produces rare isotopes by fragmentation of heavy beams. Many of the neutron-rich nuclei on the $r$-process path will be produced at FRIB for the first time, enabling measurements of:

• Nuclear masses (Penning trap mass measurements) — these determine neutron separation energies, which set the $r$-process path.
• Beta-decay half-lives — which determine the flow timescale at each waiting point.
• Excited-state properties (gamma-ray spectroscopy) — which constrain level density models.

Direct measurement of neutron capture cross sections on short-lived $r$-process nuclei remains extremely challenging (the targets are radioactive and produced in tiny quantities). Instead, indirect methods — such as $(d,p)$ transfer reactions that populate the same compound nuclear states as $(n,\gamma)$ — are used to constrain the relevant nuclear structure input.

A Quantitative Example: ${}^{138}$Ba and the $N = 82$ Peak

Barium-138 ($Z = 56$, $N = 82$) is a magic nucleus. Its neutron capture cross section at $kT = 30$ keV is $\langle\sigma\rangle_{30} = 4.0$ mb, compared to $\sim 200$ mb for non-magic neighbors. This small cross section arises from two compound-nuclear effects:

1. Large level spacing: The $N = 82$ shell closure gives ${}^{139}$Ba (the compound nucleus of ${}^{138}$Ba + n) a very low level density at the neutron separation energy. Fewer resonances means less capture. The level density parameter $a$ is effectively reduced near magic numbers (the Ignatyuk shell correction is large and negative).

2. Small neutron widths: The first excited state of ${}^{139}$Ba above $S_n$ has a small overlap with the s-wave neutron continuum, giving small reduced neutron widths. This is because the wave function of the captured neutron must match onto the continuum through the closed $N = 82$ shell, which creates a mismatch.

The net effect: ${}^{138}$Ba sits at the bottom of a neutron capture "valley," and $s$-process material accumulates there, producing the observed abundance peak at $A \approx 138$.

The quantitative comparison is striking. The measured MACS values across the chart of nuclides, combined with the $\sigma N_s \approx$ const relation, predict abundance ratios that agree with the observed Solar System values to within 10–20% for most isotopes on the main $s$-process path. This agreement — constructed from decades of careful cross section measurements at facilities like the Karlsruhe 3.7 MV Van de Graaff, the n_TOF facility at CERN, and the DANCE detector at LANSCE — is one of the great quantitative triumphs of nuclear astrophysics.

The comparison also reveals which isotopes are not produced by the $s$-process. Certain proton-rich isotopes (the "p-nuclei," such as ${}^{92}$Mo, ${}^{96}$Ru, ${}^{144}$Sm) are shielded from both $s$-process and $r$-process production by stable isobars. They must be produced by the $p$-process (photodisintegration of pre-existing heavier nuclei in supernova shock waves) or the $\nu p$-process.

Sensitivity Studies: How Nuclear Uncertainties Affect Abundances

Modern $r$-process calculations use Monte Carlo sensitivity studies to identify which nuclear properties most strongly affect the predicted abundances. For each nucleus on the $r$-process path, the mass, beta-decay rate, and neutron capture rate are varied within their uncertainties, and the impact on the final abundance pattern is assessed.

Key findings from such studies:

• Nuclear masses (especially neutron separation energies) are the most important input. A change of 500 keV in $S_n$ can shift the equilibrium $r$-process path by one or two neutrons, dramatically altering the abundance pattern near the $A \approx 130$ peak.
• Beta-decay half-lives at the waiting points ($N = 50, 82, 126$) determine the flow timescale and hence the position and shape of the abundance peaks.
• Neutron capture rates (from Hauser-Feshbach calculations using Fermi gas level densities) become critical during the "freeze-out" phase when the neutron density drops and the path moves back toward stability.
• Fission properties (barriers, fragment distributions) of the heaviest nuclei ($A > 250$) determine the upper end of the $r$-process and the recycling of material to lower masses.

The interplay between all these nuclear inputs makes the $r$-process one of the most complex multi-physics problems in astrophysics.

The Frontier: Nuclear Physics Meets Multi-Messenger Astronomy

The next decade will see convergence of several frontiers:

• FRIB will produce $\sim 80$% of the isotopes on the main $r$-process path for the first time.
• LIGO/Virgo/KAGRA (in the O5 observing run, expected late 2020s) will detect many more neutron star mergers, providing a statistical sample of kilonova light curves.
• JWST and next-generation ground telescopes will obtain high-quality spectra of kilonovae, potentially identifying individual $r$-process elements.
• Nuclear theory (ab initio methods, density functional theory) will improve predictions of level densities and reaction rates for unmeasured nuclei.

The compound nucleus model — Born in 1936 from Bohr's billiard-ball analogy — remains the theoretical engine connecting the nuclear properties measured at FRIB to the elemental abundances observed in kilonovae. It is a remarkable span: from a single resonance at 6.67 eV in ${}^{238}$U to the origin of gold, platinum, and uranium in the collision of neutron stars.

Discussion Questions

1. The $s$-process abundance pattern directly reflects the neutron capture cross sections, which in turn reflect the nuclear level density at the neutron separation energy. Explain the causal chain: shell model $\to$ magic numbers $\to$ low level density $\to$ small cross section $\to$ high abundance.

2. The $r$-process path runs through nuclei with $S_n \sim 2$–$3$ MeV. Using the Fermi gas model, estimate the level density at $S_n$ for a typical $r$-process nucleus with $A \sim 130$ and compare to the level density of ${}^{138}$Ba + n (stable, $S_n \approx 4.7$ MeV). What does this comparison tell you about $r$-process neutron capture rates?

3. GW170817 was detected simultaneously in gravitational waves and across the electromagnetic spectrum. Why was the detection of the kilonova (the optical/IR counterpart) particularly important for nuclear physics?

4. The $s$-process terminates at ${}^{209}$Bi (the heaviest stable nucleus against beta decay). How are elements heavier than bismuth (polonium, radium, thorium, uranium) produced? Which process is responsible, and why?