Saturation of Nuclear Binding from Lattice Hamiltonians

This paper resolves a conundrum regarding nuclear binding by demonstrating through Hartree-Fock variational upper bounds that while lattice Hamiltonians with only two-nucleon potentials fail to accurately describe nuclear saturation compared to previous Monte Carlo results, those including three-nucleon potentials achieve constant binding energy per nucleon primarily due to the dense packing inherent to the lattice discretization rather than repulsive interactions.

The Gist

Imagine trying to build a stable house out of Lego bricks. In the world of atomic nuclei, the "bricks" are protons and neutrons, and the "glue" holding them together is the nuclear force. Physicists have long been trying to figure out exactly how this glue works to create a perfect balance: the nucleus shouldn't fall apart, but it also shouldn't collapse into a tiny, super-dense ball. This perfect balance is called saturation.

Recently, a group of researchers proposed a new way to simulate these Lego bricks using a digital grid (a "lattice"). They claimed that if you use a specific type of "attractive glue" (forces that only pull things together, never push them apart), you could perfectly recreate how real nuclei behave.

However, the authors of this paper, Rothman, Hagen, Heinz, and Papenbrock, decided to double-check that claim. They found that the previous simulations were missing a crucial piece of the puzzle.

Here is the breakdown of their findings using simple analogies:

1. The Two Competing Stories

• Story A (The Previous Claim): Some scientists ran computer simulations on a grid and said, "Hey! If we just use attractive glue (pulling forces) for our Lego bricks, we get the perfect house every time. The bricks stick together just right, and the house doesn't collapse."
• Story B (The Reality Check): Other scientists, using different methods (simulating in "continuous space" rather than a grid), said, "That doesn't make sense. If you only have pulling glue, the house should collapse into a tiny ball. You need some pushing glue (repulsion) to keep it from getting too dense."

2. The Investigation: The "Hartree-Fock" Test

The authors of this paper acted like detectives. They took the exact same "Lego instructions" (the Hamiltonians) used in the previous grid simulations and ran their own, more rigorous check using a method called Hartree-Fock.

Think of the Hartree-Fock method as a "best-case scenario" test. It calculates the absolute lowest energy a system could possibly have. If the system is unstable in this best-case scenario, it's definitely unstable in reality.

What they found:

• The "Pulling Only" Glue Failed: When they tested the instructions that used only attractive forces (no pushing), the "houses" (nuclei) collapsed. They were way too heavy and dense. The previous simulations that claimed these worked were actually solving the wrong math problem.
• The "Three-Brick" Glue Worked (But for a weird reason): When they added a special "three-brick" force (where three bricks interact at once), the nuclei did stabilize. The energy levels looked correct.

3. The Big Twist: It Was a "Grid Glitch"

Here is the most surprising part. The authors discovered that the reason the "three-brick" glue worked wasn't because of some deep physical law of nature. It was an artifact of the grid itself.

The Analogy:
Imagine you are trying to pack people into a room.

• In the real world (Continuous Space): If you keep adding people, they eventually push back against each other because they can't occupy the same space. You need a "repulsive force" to stop the room from becoming a crushing crowd.
• On the Grid (The Simulation): The researchers were packing people into a grid of squares. As the grid got full, the "glue" (the attractive force) tried to pull people to their neighbors. But because the grid was so full, the "people" (nucleons) couldn't move to the next square—they were blocked by the other people already there.

The authors realized that the saturation (the perfect balance) wasn't caused by a repulsive force pushing back. Instead, it was caused by traffic jams. The attractive force tried to pull everyone together, but the grid was so packed that they physically couldn't move closer. The "glue" ran out of room to work.

4. The Conclusion

The paper concludes that:

1. The previous claims that "attractive forces alone" create perfect nuclei were incorrect because the simulations didn't solve the equations accurately.
2. The "saturation" seen in the successful grid simulations was a lattice artifact—a side effect of the digital grid being too crowded, not a fundamental property of nuclear physics.
3. Therefore, we still don't have a simple, perfect explanation for how alpha particles (helium nuclei) stick together in a way that matches reality. The mystery of nuclear binding remains an open challenge.

In short: The authors showed that a popular digital simulation was tricked by its own grid. The "perfect balance" it found wasn't real physics; it was just the digital equivalent of a traffic jam where cars can't move closer because the road is full.

Technical Summary

Problem Statement
The paper addresses a significant conundrum in nuclear physics regarding the binding of $\alpha$ particles and the saturation of nuclear matter. Previous auxiliary-field Monte Carlo (AFMC) simulations of specific lattice Hamiltonians (Refs. [19–21]) suggested that simple attractive two-nucleon potentials, alone or combined with attractive three-nucleon potentials, could accurately reproduce nuclear binding energies and radii for light nuclei up to oxygen. This finding contradicts established continuum-space approaches, where Hamiltonians containing only attractive two-body forces typically fail to saturate at correct densities or overbind medium-mass nuclei. The authors note that while lattice computations have been difficult to scrutinize due to the reliance on different basis sets (harmonic oscillator vs. continuum) in other ab initio methods, the recent release of the publicly available NuLattice package now allows for a direct comparison and re-evaluation of these lattice Hamiltonians.

Methodology
The authors perform Hartree-Fock (HF) computations to establish variational upper bounds for the ground-state energies of finite nuclei and nuclear matter. They utilize established lattice Hamiltonians from Refs. [19, 20] (chiral effective field theory at leading order) and Ref. [21] (pion-less effective field theory).

• Finite Nuclei: Calculations are performed for $^4$He, $^8$Be, $^{12}$C, and $^{16}$O using a discrete three-dimensional spatial lattice with extent $L=6$. Initial states are constructed as compact clusters of nucleons, and the HF equations are solved self-consistently.
• Nuclear Matter: The authors compute the equation of state for symmetric nuclear matter and neutron matter on lattices of varying extents ($L=5, 6, 7$). They utilize closed-shell configurations to construct one-body density matrices and evaluate energy expectation values for one-, two-, and three-body terms.
• Verification: The study benchmarks the two-nucleon interaction, validates the NuLattice HF code against coupled-cluster reference states, and provides analytical derivations for expectation values in fully occupied lattices to verify numerical results.

Key Results

1. Failure of Two-Body Only Hamiltonians: For the Hamiltonian from Ref. [20] (containing only attractive two-body potentials), the HF results show that while the $^4$He energy is a plausible upper bound, heavier nuclei ($^8$Be, $^{12}$C, $^{16}$O) are significantly overbound. The binding energy per nucleon exceeds the empirical value of $\approx 8$ MeV, and no saturation is observed. This contradicts the AFMC results of Ref. [20], suggesting those simulations did not accurately solve the presented Hamiltonian.
2. Issues with Mixed Potentials: For the Hamiltonian from Ref. [19] (containing local and nonlocal short-range potentials), the HF results also show overbinding and a lack of proper saturation compared to AFMC results. The authors conclude that the AFMC simulations in Ref. [19] did not accurately solve the Hamiltonian presented in that work.
3. Saturation Mechanism in Three-Body Hamiltonians: For the Hamiltonian from Ref. [21] (containing attractive two- and three-body potentials), the HF results yield a saturation point close to the empirical value. However, the authors identify the mechanism for this saturation as a lattice artifact. In the Thomas-Fermi approximation, attractive potentials scale as $\rho^2$ and $\rho^3$, which would normally overwhelm the kinetic energy ($\rho^{5/3}$) and cause collapse. On the lattice, however, as density increases and nucleons densely pack the lattice sites, nonlocal interactions are suppressed by Pauli blocking (nucleons cannot move to neighboring sites). This reduces the attractive potential energy relative to the kinetic energy, artificially stabilizing the system.
4. Continuum Limit: The authors demonstrate that in the continuum limit ($a \to 0$), the attractive two-body potential leads to a system collapse, confirming that the saturation observed in Ref. [21] is specific to the discrete lattice formulation and not a feature of the continuum Hamiltonian.

Significance and Claims
The paper claims to resolve the discrepancy between lattice AFMC results and continuum expectations by demonstrating that the previously reported accurate binding from simple lattice Hamiltonians was likely an artifact of the numerical solution method (specifically, the use of a large temporal lattice spacing $a_t = 1/(150 \text{ MeV})$ in the transfer matrix formalism) rather than a physical property of the Hamiltonians.

• The authors assert that Hamiltonians with only attractive two-body forces do not yield accurate binding, consistent with continuum theory.
• They conclude that the saturation observed in Ref. [21] is a consequence of dense packing on the lattice (a lattice artifact) rather than repulsive three-nucleon forces, which are typically required for saturation in continuum formulations.
• Consequently, the paper states that an ab initio understanding of $\alpha$-particle binding and clustering using these specific simple lattice Hamiltonians remains an open challenge, as the previously claimed successes appear to be numerical artifacts.