Nuclear reactions are orbit programmes

Plain-language explainer for doi:10.5281/zenodo.21249152 (#548)


The central idea in one sentence

Every elementary nuclear reaction — fusion, fission, beta-decay, capture, isomerism — is one of five ISA opcodes acting on a nucleon orbit-occupancy register, and the three most important structural facts in nuclear astrophysics (magic numbers, the CNO bottleneck, r-process abundance peaks) follow from orbit-occupancy bookkeeping alone, with zero free parameters.


What the standard picture gives you

The nuclear shell model (Mayer and Jensen, 1949) organises nucleons into single-particle energy levels split by spin-orbit coupling. Fill the levels from the bottom up; wherever there is a large gap, the nucleus is especially stable. The nucleon numbers at those gaps are the magic numbers 2, 8, 20, 28, 50, 82, 126 — explaining why ${}^{4}$He, ${}^{16}$O, ${}^{40}$Ca, ${}^{208}$Pb are unusually stable, why they are especially abundant in the universe, and why the r-process pauses at them during neutron-star mergers.

The shell model needs two fitted parameters (the spin-orbit coupling constants) to get the level ordering right. It does not by itself explain why those particular numbers are magic — it reads them off a fitted level diagram. Lattice QCD, the only systematic first-principles approach to nuclear structure, cannot currently predict magic numbers at all.


The orbit-occupancy description

Represent the state of a nucleus as an orbit-occupancy vector $\mathbf{v} = (\nu_1, \nu_2, \ldots, \nu_k)$, where $\nu_i$ is the number of nucleons in the $i$-th shell-model level. Elementary nuclear reactions become ISA opcodes on $\mathbf{v}$:

Opcode Nuclear reaction Example
FLIP Neutron/proton capture $(Z,N) + n \to (Z, N{+}1) + \gamma$
FLOP Beta-decay ${}^{13}$N $\to$ ${}^{13}$C $+ e^+ + \nu_e$
SPLAT Fusion ${}^2$H $+$ ${}^3$H $\to$ ${}^4$He $+ n$
SPLIT Fission / alpha-emission ${}^{235}$U $+ n \to {}^{141}$Ba $+ {}^{92}$Kr $+ 3n$
TWIST Nuclear isomerism ${}^{99m}$Tc $\to$ ${}^{99}$Tc $+ \gamma$
ORBIT Closed reaction cycle CNO cycle; pp-chain

A nuclear reaction sequence is an ISA programme. The CNO cycle that powers the Sun is: SPLAT, FLOP, SPLAT, SPLAT, FLOP, SPLAT·SPLIT — a six-opcode ORBIT programme that regenerates ${}^{12}$C and releases net energy.


Why magic numbers are tropical vertices

In the Maslov–Gibbs Einsum framework, the Gibbs distribution $\pi_k(\beta) = e^{-\beta E_k}/Z(\beta)$ at $\beta \to \infty$ (zero temperature) concentrates entirely on the ground state. For an orbit-occupancy register, the ground state is the configuration where every level up to the lowest large gap is filled — an orbit-complete configuration at a large energy gap. That is precisely the definition of a magic number.

The magic numbers are therefore tropical vertices of the orbit-occupancy register: the configurations that minimise the free energy in the $\beta \to \infty$ limit. No fitted parameters — just the gap structure, which follows from the Pauli exclusion principle and the ordering of single-particle levels.

All seven magic numbers (2, 8, 20, 28, 50, 82, 126) are correctly identified this way (experiment x548a, 7/7).


The CNO bottleneck from the log-rate criterion

The CNO cycle has six steps. Four are proton-capture (SPLAT) reactions with different S-factors $S_0$ (nuclear matrix elements at stellar energies) and Coulomb barriers. The rate of each step is:

\[\log \langle \sigma v \rangle \propto \log S_0 - 3\left(\frac{E_G}{4(kT)^2}\right)^{1/3}\]

where $E_G = (\pi \alpha Z_1 Z_2)^2 \cdot 2\mu c^2$ is the Gamow energy. The bottleneck step is whichever SPLAT has the smallest log-rate.

The ${}^{14}$N$(p,\gamma){}^{15}$O step (Step 4) has $Z_2 = 7$ rather than $Z_2 = 6$, giving a Gamow exponent about 45 units more negative than the other steps. This Coulomb suppression — invisible if you only compare S-factors — makes Step 4 the bottleneck by a factor of $\sim 10^3$. The orbit-occupancy description identifies this correctly (experiment x548b) without any nuclear structure calculation, using only $Z$, $A$, and the measured S-factor.


r-process peaks as tropical fixed points

The rapid neutron-capture process (r-process) builds heavy elements in neutron star mergers by repeated FLIP (neutron capture) followed by FLOP (beta-decay). At magic neutron numbers $N = 50, 82, 126$, the next FLIP requires opening a new shell — a large energy gap — so neutron-capture cross-sections drop by $10$–$10^3\times$. Material piles up at the magic-$N$ waiting point until FLOP can proceed.

The pile-up produces the observed r-process abundance peaks at $A \approx 80, 130, 195$. In the MGE framework, the peak location is a tropical fixed point of the FLIP/FLOP dynamics — determined entirely by orbit-occupancy, zero parameters. The peak width is set by the inverse temperature $\beta$ at r-process conditions.

Predictions: $N=50 \to A=80$ ($\Delta A = 0$), $N=82 \to A=130$ ($\Delta A = 0$), $N=126 \to A=196$ ($\Delta A = 1$). All three correct (experiment x548c, 3/3).

The lanthanide-rich ejecta that made the kilonova AT2017gfo (the optical counterpart of GW170817) glow red for days is the fingerprint of the $N=82$ tropical fixed point playing out in real time in a neutron star merger, 130 million light-years away, confirmed by gravitational-wave astronomy in 2017.


Is the ISA description more fundamental than the shell model?

In one sense, no: the shell model is the microscopic theory, and the ISA orbit description is derived from it. In another — more important — sense, yes.

The shell model predicts magic numbers by reading off the gaps in a fitted level diagram. The ISA description explains why those configurations are special: they are tropical vertices of the orbit-occupancy register. The explanation requires no parameters and no wavefunctions — only the existence of a gap and the Pauli exclusion principle that forbids two nucleons from occupying the same orbit.

This is the same relationship as between G-walk chemistry and DFT:

Level Microscopic theory ISA / orbit description More fundamental for
Chemistry DFT (Kohn-Sham equations, fitted functional) G-walk chemistry (orbit occupancy) Spin-state transitions, catalytic cycles
Nuclear Shell model (fitted spin-orbit coupling) G-walk nucleonics (orbit occupancy) Magic numbers, nucleosynthesis peaks
Particle QCD / Feynman diagrams Amplituhedron / ISA programme Scattering amplitude structure

At each level, the orbit description strips away the wavefunction machinery and identifies the structural reason for the phenomenon. It is exact for the class of properties that orbit-occupancy determines — and for those properties, it is more predictive than the microscopic theory because it has fewer parameters.

The quark model tells you what nucleons are made of. G-walk nucleonics tells you why nuclei with certain nucleon numbers are magic, why the CNO cycle has a bottleneck, and why the universe has a peak of heavy elements at $A=130$. Both are right. They answer different questions.


What this paper does not claim

Not a replacement for nuclear structure theory. Binding energies, scattering lengths, electromagnetic moments, and sub-shell structure require the shell model, ab initio nuclear theory, or lattice QCD. G-walk nucleonics describes which orbit is occupied, not the wavefunction within the orbit.

Not a source of reaction cross-sections. The astrophysical S-factor must be measured (at LUNA or similar underground laboratories) or computed from nuclear structure theory. The log-rate criterion uses $S_0$ as an input.

Not a predictor of absolute r-process yields. Peak locations are zero-parameter predictions; peak heights (absolute abundances) require r-process network calculations with astrophysical conditions ($Y_e$, entropy, expansion timescale).


See also: doi:10.5281/zenodo.21224107 (G-Walk Chemistry — the same framework one level up, for electrons) · doi:10.5281/zenodo.21224111 (G-Walk Protein Design — orbit occupancy for metalloprotein active sites) · doi:10.5281/zenodo.20721743 (The $H^k$ Complexity Ladder — the cohomological framework underlying the tropical fixed-point classification)