PREX, CREX, and MREX: Probing Neutron Skins

Updated 11 January 2026

The experiments measure neutron skin thickness via helicity-dependent cross-section asymmetry, achieving sub-0.05 fm precision in heavy nuclei.
They employ parity-violating electron scattering and Bayesian inference to extract the weak charge form factor and neutron density profile with minimal model dependence.
Results from these initiatives provide crucial constraints on the nuclear symmetry energy, impacting neutron star modeling and energy density functional calibrations.

Parity-violating electron scattering (PVES) has become the preeminent experimental tool for model-independent measurements of the neutron density distribution in heavy nuclei. The PREX (Lead Radius Experiment), CREX (Calcium Radius Experiment), and the planned MREX (Mainz Radius EXperiment) form a sequence of high-precision initiatives that collectively address fundamental issues in nuclear structure, the density dependence of the symmetry energy, and the consistency between nuclear and astrophysical constraints.

1. Experimental Objectives, Technique, and Infrastructure

The principal aim of the PREX, CREX, and MREX programs is to extract the neutron skin thickness, defined as $\Delta r_{np} = \langle r_n^2 \rangle^{1/2} - \langle r_p^2 \rangle^{1/2}$ , with sub-0.05 fm precision for heavy neutron-rich nuclei such as $^{208}$ Pb and $^{48}$ Ca. The significance of $\Delta r_{np}$ lies in its sensitivity to the nuclear symmetry energy and its slope parameter $L$ at nuclear saturation density $\rho_0$ —a central quantity for both nuclear structure and neutron star modeling (Xu et al., 2020).

These experiments employ parity-violating elastic electron scattering at forward angles, using high-current, highly polarized, longitudinal electron beams incident on isotopically pure targets. The observable is the helicity-dependent cross-section asymmetry:

$A_{PV} = \frac{\sigma_R - \sigma_L}{\sigma_R + \sigma_L}\ ,$

which—dominated by weak neutral current (Z $^0$ ) exchange with the neutron distribution—directly probes the weak charge form factor $F_W(q)$ . By comparing $F_W(q)$ to the precisely known charge form factor $F_{ch}(q)$ (from conventional electron scattering), the neutron distribution can be extracted with minimal model dependence (Zhao et al., 2024, Davoudiasl et al., 17 Dec 2025).

PREX targets $^{208}$ Pb at JLab, measuring $A_{PV}$ at $q \simeq 0.398$ fm ${}^{-1}$ .
CREX focuses on $^{48}$ Ca at JLab, at $q \simeq 0.873$ fm ${}^{-1}$ . The lighter mass and amenability to ab initio calculations provide a critical cross-check for nuclear models.
MREX, planned at Mainz/MESA, aims for sub-0.03 fm precision in both nuclei and possibly additional targets.

Current PVES facilities achieve statistical and systematic error control at the permille to few-percent level, with experimental precision continually improving due to advances in beam stability, detector technology, and background suppression (Reed et al., 2020).

2. Theoretical Framework, Model Independence, and Extraction Formalism

PVES observables are rigorously interpretable via the Standard Model. The leading-order Born formula for the parity-violating asymmetry relates $A_{PV}$ to nuclear structure quantities as:

$A_{PV}(q) = - \frac{G_F q^2}{4\pi \alpha \sqrt{2}} \frac{F_W(q)}{F_{ch}(q)}\ ,$

where $G_F$ is the Fermi constant, $\alpha$ the fine-structure constant, and $F_W(q)$ the weak charge form factor (Davoudiasl et al., 17 Dec 2025).

The extraction of $F_W(q)$ uses distorted wave Born approximation (DWBA) calculations that incorporate strong coulomb distortions, but the method remains exceptionally model-independent compared to hadronic probes. Single-parameter (rms radius) or two-parameter Fermi (half-density radius $c$ , surface diffuseness $a$ ) shapes for $\rho_W(r)$ are fit to the data. Statistical consistency and systematic robustness are ensured by exploiting Markov-Chain Monte Carlo methods and Bayesian parameter estimation frameworks (Xu et al., 2020, Zhao et al., 2024, Zhang et al., 2022).

A recent refinement is the use of the difference $\Delta F = F_{ch}(q) - F_W(q)$ , which is measured directly and constrains both the neutron skin and the surface structure. In particular, with measurements at multiple momentum transfers (as foreseen for MREX), both $c$ and $a$ can be determined, fully specifying the neutron density profile (Reed et al., 2020).

3. Bayesian Inference, Symmetry Energy Constraints, and Posterior Structure

PREX, CREX, and MREX data are interpreted within Bayesian statistical frameworks using Skyrme-Hartree-Fock, relativistic mean field (RMF), and energy density functional (EDF) approaches. The focus is on the symmetry energy $S(\rho)$ and particularly its slope $L$ at saturation density:

$L \equiv 3\rho_0 \left. \frac{dS}{d\rho} \right|_{\rho_0}\ .$

Experimental likelihoods incorporate the measured $\Delta r_{np}$ (or, equivalently, $\Delta F$ ), with uncertainties dominated by beam statistics and detector systematics. Priors on nuclear matter parameters are typically chosen broad and uniform, reflecting the goal of unbiased inference. Posterior distributions for $L$ and related parameters are obtained by marginalization over nuisance quantities (Xu et al., 2020, Zhao et al., 2024).

Numerically, the PREX and CREX data pull the inferred posteriors in different directions:

PREX favors $\Delta r_{np}(^{208}\mathrm{Pb}) \simeq 0.283 \pm 0.071$ fm, implying $L \sim 70-90$ MeV within most functionals.
CREX yields $\Delta r_{np}(^{48}\mathrm{Ca}) \simeq 0.121 \pm 0.026$ fm, implying $L \sim 30-40$ MeV (Reed et al., 2023, Kumar et al., 2023, Zhang et al., 2022).

The conflicting inferences highlight the need for model extensions, notably in handling the surface symmetry energy and higher-order empirical parameters that can decouple low- and high-density regimes (Mondal et al., 2022).

4. Model Tensions, Isovector Spin-Orbit Effects, and Resolution Strategies

A dominant theoretical challenge is the so-called CREX–PREX dilemma: canonical EDFs predict a strong correlation between $\Delta r_{np}$ in $^{208}$ Pb and $^{48}$ Ca, yet experiment finds a large skin in Pb and a much smaller one in Ca. This suggests missing physics in the standard isovector channel.

Recent studies identify the isovector spin-orbit interaction, parameterized (in Skyrme or RMF models) by $b'_4$ , as a powerful lever arm. Enhancement of the isovector spin-orbit term can selectively reduce $\Delta r_{np}$ in $^{48}$ Ca without strongly affecting $^{208}$ Pb, matching the reported data. However, the magnitude of the required enhancement (e.g., $\beta \sim -300$ in RMF) is unphysically large and destroys agreement with empirical single-particle spectra and shell closures (Kunjipurayil et al., 10 Mar 2025, Zhao et al., 2024).

A more physically consistent approach is simultaneous refitting of central, spin-orbit, and tensor terms under combined experimental constraints. Linear "pocket formulas" express $S_V$ and $L$ as separate functions of $\Delta F_{Ca}$ and $\Delta F_{Pb}$ , facilitating the efficient planning of future experiments such as MREX (Zhao et al., 2024).

5. MREX and the Next Generation of Parity-Violating Measurements

MREX at MESA will extend the capacity of PVES programs by enabling a second measurement at larger $q$ for both $^{208}$ Pb ( $q \sim 0.76$ fm ${}^{-1}$ ) and $^{48}$ Ca ( $q \sim 1.28$ fm ${}^{-1}$ ). This provides heightened sensitivity to the surface diffuseness parameter $a$ of the weak charge distribution. Simulation studies indicate that modest beam times ( $\sim$ 50–100 days) suffice to extract $a$ to a few percent precision, provided detector and beam systematic controls (Reed et al., 2020).

By jointly fitting data at two $q$ -values, both the half-density radius $c$ and diffuseness $a$ are constrained, yielding an essentially model-independent neutron density profile. This fully determines $\Delta r_{np}$ and $S(\rho)$ around and below $\rho_0$ , with minimal sensitivity to modeling assumptions. MREX's results will permit rigorous global Bayesian updating, integrating nuclear, astrophysical, and ab initio data to map the symmetry energy (and its density derivatives) (Xu et al., 2020, Davoudiasl et al., 17 Dec 2025).

6. Impact on Nuclear Matter Properties, Astrophysical EOS, and Future Outlook

The results from PREX, CREX, and anticipated MREX analyses feed directly into calibrations of nuclear density functionals, with broad consequences for neutron-rich nuclei and neutron star structure. Posterior distributions on $L$ , $K_{sym}$ , and other EOS parameters impact predictions for neutron star radii $R_{1.4}$ , tidal deformabilities $\Lambda_{1.4}$ , and core–crust transition densities (Kumar et al., 2023, Reed et al., 2023, Mondal et al., 2022).

A key finding is that models accommodating both PREX and CREX generally require either (a) extreme surface or higher-order couplings, or (b) a very stiff EOS at supra-nuclear densities—leading to neutron star radii and $\Lambda_{1.4}$ values exceeding current multimessenger bounds. This suggests that either the standard nuclear structure paradigm requires revision or the data reflect a confluence of "soft" and "stiff" symmetry energy behavior at different densities, or both (Reed et al., 2023, Zhao et al., 2024).

Planned and proposed future experiments, including multi- $q$ PVES (e.g., double-q CREX or extended MREX) and continuous- $Q^2$ mapping at an Electron-Ion Collider, will further decouple and diagnose the various empirical parameters underlying the isovector sector and the symmetry energy (Davoudiasl et al., 17 Dec 2025, Reed et al., 2020). Systematic inclusion of QED radiative corrections, especially photon vacuum polarization, will be necessary as experimental precision approaches the sub-percent level (Reed et al., 4 Jan 2026).

Summary Table: Key Targets and Precision Goals for PREX, CREX, and MREX

Experiment	Target	$q$ (fm $^{-1}$ )	Reported/Goal $\sigma_{\Delta r_{np}}$ (fm)
PREX-II	$^{208}$ Pb	0.398	$0.06$ (goal), $0.071$ (reported)
CREX	$^{48}$ Ca	0.873	$0.02$ (goal), $0.026$ (reported)
MREX	$^{208}$ Pb	0.76	$0.03$ (goal, projected)
MREX	$^{48}$ Ca	1.28	$0.02$ (goal, projected)

The PREX, CREX, and MREX program, interpreted using Bayesian theory with advanced EDFs, is driving a paradigm shift in the precision mapping of the nuclear symmetry energy and confronting long-standing theoretical assumptions about the density dependence of the isovector channel in nuclear matter. The results inform both terrestrial and astrophysical observables and promise sharper convergence between nuclear structure and neutron star modeling as experimental, theoretical, and computational methodology advances further (Xu et al., 2020, Zhao et al., 2024, Kunjipurayil et al., 10 Mar 2025, Mondal et al., 2022, Reed et al., 2023).