Few-nucleon matrix elements in pionless effective field theory in a finite volume

The Big Picture

Imagine trying to measure the weight of a fish, but you can only do it inside a tiny aquarium. The fish is cramped, its behavior constrained by the walls, and the measurement you get isn’t quite what you’d find in the open ocean.

Now imagine the fish is a nucleus, the aquarium is a finite computational grid, and the “weight” is a fundamental nuclear property. That’s the problem facing physicists who use Lattice Quantum Chromodynamics (LQCD), one of the most direct tools for computing nuclear properties from first principles.

LQCD places quarks and gluons on a discrete spacetime grid and simulates their interactions. But every LQCD calculation lives in a finite box, and extracting real-world physics from that box requires careful theoretical bridges. For simple two-particle scattering, those bridges exist. For heavier nuclei and their matrix elements? The path has been far murkier.

This paper by William Detmold and Phiala Shanahan at MIT’s Center for Theoretical Physics and IAIFI extends finite-volume pionless effective field theory (FVEFT_π/) to nuclear matrix elements for the first time. Nuclear matrix elements describe how nuclei respond to external forces, from magnetic fields to the weak nuclear force, and are essential for connecting QCD to experiment. The extension opens a systematic route from lattice calculations to real nuclear physics.

Key Insight: By combining finite-volume effective field theory with a variational optimization technique, the researchers can take finite-box LQCD calculations of nuclear properties and reliably extrapolate them to the infinite-volume limit that nature actually lives in.

How It Works

The starting point is pionless effective field theory (EFT_π/), a framework for nuclear physics at energies too low to produce pions. Rather than tracking every quark and gluon, EFT_π/ describes nucleons interacting through contact forces parameterized by low-energy constants (LECs), numbers that encode all the short-distance physics we don’t need to resolve explicitly.

The core idea: run this EFT inside the same finite box as the LQCD calculation. Finite-volume effects are long-range phenomena, arising from long-wavelength physics being cut off by the box walls, placing them squarely within EFT_π/‘s domain of validity. Match the EFT to LQCD results in the box, fix the LECs, then run the EFT without a box to get infinite-volume predictions.

The computational engine is the stochastic variational method (SVM), a technique for building approximate wavefunctions of few-body systems. The SVM constructs trial quantum states as combinations of Gaussian basis functions, then iteratively adjusts them to minimize the system’s energy. For two- and three-nucleon systems, this yields highly accurate variational wavefunctions both inside the box and in infinite volume.

The procedure has two matched stages:

1. Energy matching: Two- and three-body LECs are tuned to reproduce finite-volume binding energies from NPLQCD collaboration LQCD calculations, computed across three lattice volumes (L ∈ {3.4, 4.5, 6.7} fm) at a pion mass of 806 MeV. That’s roughly six times the physical value, which makes the simulations computationally tractable. With LECs fixed, infinite-volume binding energies follow directly.

2. Matrix element matching: Using wavefunctions from stage one, finite-volume matrix elements of external currents are computed within FVEFT_π/ and matched to LQCD calculations of the same operators. This fixes additional two-body counterterms, coupling constants that encode how nucleon pairs respond collectively to external probes beyond what single-nucleon responses predict.

The framework covers scalar matrix elements (sensitive to how nucleon mass arises from quark condensates), tensor matrix elements (related to quark spin distributions), axial matrix elements (tied to weak interactions and beta decay), magnetic moments, and the isovector longitudinal momentum fraction. The axial matrix element is a particularly high-value target: it governs proton-proton fusion in the sun and neutrino-nucleus scattering, and its two-body LEC, L₁,A, has been notoriously difficult to pin down.

Why It Matters

Extrapolating LQCD matrix elements to infinite volume is not a technical footnote. It’s a prerequisite for comparing LQCD predictions with experiment. Every finite-volume LQCD result carries contamination from the box walls, and without a systematic way to remove it, the computational effort behind LQCD cannot be fully translated into measurable physics.

The FVEFT_π/ framework developed here provides that translation for few-nucleon systems across a wide range of properties simultaneously. It is also more flexible than Lüscher’s method, an established model-independent approach formally limited to two- and three-particle systems that doesn’t naturally extend to matrix elements of heavier nuclei.

As LQCD calculations gradually approach physical quark masses and larger nuclei, this framework will be ready to interpret them. The immediate next step: applying this machinery at lighter quark masses, closer to physical reality, and extending it to larger nuclei and properties like electromagnetic polarizabilities.

Bottom Line: Detmold and Shanahan have built the theoretical infrastructure to systematically extract infinite-volume nuclear matrix elements from finite-volume lattice QCD, connecting first-principles computation to measurable nuclear physics for scalar, tensor, axial, magnetic, and momentum-fraction properties of two- and three-nucleon systems.