Invisible neutron decay and light BSM particles

TL;DR

The paper investigates invisible neutron decay within the SMEFT framework, showing that SM-only operators require $d=12$ for decays like $n\to 3\nu$, while introducing light BSM states (RH neutrinos, ALPs, a light scalar, or a light $Z'$) can realize $d=(7-9)$ operators that yield observable invisible decays without charged-lepton final states. It systematically builds four UV scenarios, derives the relevant effective operators and decay rates, and enumerates all tree-level UV completions, highlighting how symmetry assignments suppress competing $d=6$ channels. The work maps the full set of possible mediators and provides LHC reinterpretations for current and future constraints, emphasizing how collider searches for vector-like quarks, leptoquarks, and diquarks can probe these invisible-neutron-decay channels. If an invisible neutron decay were observed, it would point to new neutral BSM degrees of freedom and potentially connect to neutrino mass generation, with JUNO, Hyper-Kamiokande, and future colliders offering complementary tests of these scenarios.

Abstract

In Standard Model Effective Field Theory (SMEFT) invisible neutron decay arises from $d=12$ operators. Adding new, light particles to the field content of the SM, such as right-handed neutrinos, allows to construct operators for invisible neutron decay at much lower dimensions. Observing invisible neutron decay, if nucleon decays with charged leptons remain absent, would therefore point towards the existence of new neutral degrees of freedom. Here, we discuss four cases: (i) Adding right-handed neutrinos to the SM; (ii) a right-handed neutrino and an axion-like particle; (iii) a right-handed neutrino and a (nearly) massless singlet scalar; and (iv) a right-handed neutrino and a light $Z'$. We give the general tree-level decomposition for the resulting $d=(7-9)$ operators for invisible neutron decay. We also briefly discuss LHC searches related to the exotic states found in these UV completions.

Figures (4)

• Figure 1: $p(n) \rightarrow \pi^{+}(\pi^{0})+\text{missing}$ induced by the $d=9$$u_{R}d_{R}d_{R}N_{R}N_{R}N_{R}$ operator.
• Figure 2: Topologies for tree-level realizations of the $d=9$ neutron decay operators, see Eqs. \ref{['eq:opsNR1']} and \ref{['eq:opsNR2']}. Depending on the chiralities of the outer fermion fields, the dashed lines can be either a scalar, $S$, or vector, $V$. The mediator $F$ in Topology B must be introduced as a vector-like fermion, but when it is a SM singlet, it can also be a Majorana fermion.
• Figure 3: Example Feynman diagrams for single and pair production of VLQs and their decays.
• Figure 4: Topology for the operators with four fermions and 1 scalar, such as $u_{R} d_{R} d_{R} N_{R} \phi$ and $u_{R} d_{R} d_{R} \overline{N_{R}} a$, i.e., One of the outer legs should be a scalar, $\phi$ or $a$. The directions of arrows on the outer legs depends on the decomposition. Although the mediators are given with solid lines in this topology, they can be scalars $S$, vectors $V$ or fermions $F$, depending on the distribution of outer fields.