Scattering off unstable states
Francesco Giacosa, Vanamali Shastry · 2025
Read Scattering off unstable states by Francesco Giacosa, Vanamali Shastry (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-03145.txt.
Introduction
Francesco Giacosa and Vanamali Shastry tackled scattering that involves unstable particles in 2025, a gap in the standard quantum field theory toolkit. A particle that lives long enough can enter or leave a scattering event like a stable one. Yet the standard theory strictly covers only fully stable end states, and a specific singularity appears at certain angles when unstable particles take part.
Their finite-time formalism gives analytic results with no such singularity, even when the time window grows without bound. That solves the problem at the level experiments need. It also justifies the everyday practice of treating long-lived particles, such as weakly decaying pions, as stable during strong interactions. Readers get firmer ground under a shortcut every particle physicist uses.
Catalog record
| CREATOR | Francesco Giacosa, Vanamali Shastry |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-03145
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Scattering off unstable states Preprint: APS/123-QED Francesco Giacosa, Vanamali Shastry Abstract Unstable states that live long enough may appear as in(out)going particles in scattering experiments. Yet, the standard QFT approach strictly applies only to fully stable asymptotic states. This is evident when scattering involving unstable particles develops a t t -channel singularity at specific angles. We employ a finite-time formalism leading to analytic results without the singularity (even in the infinite-time limit), thus solving the problem at a fundamental level. In turn, the approach also justifies treating long-lived particles, like weakly decaying pions, as stable during strong interactions. The vast majority of particles listed in the Particle Data Group (PDG) compilation are unstable ParticleDataGroup:2024cfk . Among leptons, the muon and the tau particles decay weakly, and the neutrinos oscillate into each other. Among hadrons, only the proton is currently known to be stable, while all other hadrons present in the PDG decay via weak, electromagnetic, or strong interactions. If an unstable particle, elementary or not, lives long enough, it may be implemented as an ingoing/outgoing state in scattering experiments. Typical examples are the \mu^{+}\mu^{-} , pion-pion, pion-nucleon, and nucleon-nucleon processes. With the advances in femtoscopy techniques, the scattering of additional unstable hadrons (including those that strongly decay) can be studied experimentally Fabbietti:2020bfg . Nevertheless, the usually applied quantum field theoretical (QFT) formalism is rigorously defined only for asymptotically stable states. One may at first ignore this aspect and proceed with the textbook scattering of stable particles, e.g. Ref. Peskin:1995ev , arguing that it is applicable also for unstable states that live long enough. However, there are some cases where the standard approach fails badly because a divergence appears in the differential cross-section. This divergence is referred to as the ‘ t -channel singularity’ and was originally discussed in 1961 by Peierls Peierls:1961zz in the framework of the hadronic N^{*}\pi scattering via an exchange of a (stable) nucleon N . This singularity also affects, e.g. W^{+}e^{+} , Z^{0}e^{-} , and \mu^{+}\mu^{-} scattering Grzadkowski:2021kgi . For simplicity, but with no loss of generality, let us consider a theory with 3 scalar fields, denoted with A , B , and C with masses m_{A,B,C} . We choose m_{C}\geq m_{B}\geq m_{C} . In particular, we assume that m_{C}>m_{A}+m_{B} , rendering the decay C\rightarrow AB kinematically possible. The corresponding interaction Lagrangian \mathcal{L}=gABC leads to \Gamma_{C\rightarrow AB}=\frac{g^{2}|{\vec{k}_{A}}|}{8\pi m_{C}^{2}} , where \vec{k}_{A} is the three-vector of the outgoing particle A. Consider the scattering AC\rightarrow AC via the exchange of B as shown in Fig. 1 . The tree-level amplitude for this diagram is, \displaystyle i\mathcal{M}=i\mathcal{M}_{t}+i\mathcal{M}_{s} \displaystyle=\frac{i(ig)^{2}}{t-m_{B}^{2}+i\varepsilon}+\frac{i(ig)^{2}}{s-m_{B}^{2}+i\varepsilon}\text{ ,} (1) where t=(p_{C}-k_{A})^{2} and s=(p_{A}+p_{C})^{2} are the usual Mandelstam variables 1 1 1 Strictly speaking, the exchange channel is of the u -type, but the literature deals with t -channel singularity, see below. The scattering angle \theta is chosen as the angle between \vec{p}_{C} and \vec{k}_{A} , in agreement with the t -channel. . For the in(out)going momenta, see Fig. 1. Using the standard QFT (sQFT) approach for asymptotic states and neglecting the fact that C is not stable, the differential and the total cross-section reads in the center-of-mass (c.o.m.) frame Peskin:1995ev : \frac{d\sigma}{d\Omega}=\frac{1}{16E_{in}^{2}}\frac{\left|i\mathcal{M}\right|^{2}}{(2\pi)^{2}}\text{ , }\sigma=\int\mathrm{d\Omega}\frac{d\sigma}{d\Omega}=\int_{0}^{\pi}\mathrm{d\theta}\frac{d\sigma}{d\Omega}2\pi\sin\theta\text{ ,} (2) where E_{in}=\omega_{A}(\mathbf{p}_{c})+\omega_{B}(\mathbf{p}_{c}) is the total initial energy with \omega_{A,C}(\mathbf{p})=\sqrt{\mathbf{p}^{2}+m_{A,C}^{2}} being the energy of A(C) . A C C A B p_{A} p_{C} k_{C} k_{A} q_{B} (a) A C C A B (b) Figure 1 : (a) The t -channel diagram that can potentially contain a divergence. (b) The s -channel diagram representing the same scattering process (the momentum assignments are as the t -channel). Quite interestingly, if certain conditions are met, t=m_{B}^{2} is possible, then a singularity in the t -channel amplitude of Eq. ( 1 ) appears. More specifically, this is the case if three-momentum’s length \left|\mathbf{p}_{C}\right| takes the value between \left|\mathbf{p}_{C}\right|_{\min}=\frac{\sqrt{m_{A}^{4}+m_{B}^{4}+m_{C}^{4}-2m_{A}^{2}m_{C}^{2}-2m_{B}^{2}m_{C}^{2}}}{2\sqrt{2m_{A}^{2}-m_{B}^{2}+2m_{C}^{2}}} (3) and \left|\mathbf{p}_{C}\right|_{\max}=\frac{\sqrt{m_{A}^{4}+m_{B}^{4}+m_{C}^{4}-2m_{B}^{2}m_{C}^{2}-2m_{A}^{2}m_{C}^{2}}}{2m_{B}} (4) in the center-of-mass frame. The value of the singular scattering angle \theta_{sing} for which t=m_{B}^{2} is given by \cos\theta_{sing}=\frac{-(\omega_{C}(\mathbf{p}_{C})-\omega_{A}(\mathbf{p}_{C}))^{2}+m_{B}^{2}+2\left|\mathbf{p}_{C}\right|^{2}}{2\left|\mathbf{p}_{C}\right|^{2}}\text{ .} (5) In particular, \left|\mathbf{p}_{C}\right|_{\min} corresponds to \theta_{sing}=\pi and \left|\mathbf{p}_{C}\right|_{\max} to \theta_{sing}=0 . In Fig. 2 we show the behavior of \left|\mathbf{p}_{C}\right|_{\min} , \left|\mathbf{p}_{C}\right|_{\max} as functions of m_{B} , the singular angle \theta as function of \left|\mathbf{p}_{C}\right| for a given choice of m_{A,B,C} , and the ‘naive’ differential and total cross-section in the standard QFT approach. Whenever the singularity is present, the total cross-section diverges. (a) (a) p_{C} vs. m_{B} . (b) (c) d\sigma/d\Omega vs. \theta . (c) (b) \theta_{\text{sing}} vs. |\mathbf{p}_{C}| . (d) (d) \sigma vs |\mathbf{p}_{C}| . Figure 2 : Summary of plots for the ‘naive’ standard QFT (sQFT) approach: (a) Minimum and maximum values |\mathbf{p}_{c}|_{\text{min/max}} as a function of m_{B} for m_{C}=1 and m_{A}=1/10 [a.u.]. (b) Singular angle \theta_{\text{sing}} as a function of |\mathbf{p}_{C}| for m_{B}=0.5 [a.u.]. (c) Differential cross-section as a function of \theta for |\mathbf{p}_{C}|=0.4 and \Gamma_{C}=0.1 [a.u.]. The singular angle is \theta_{sing}=1.15 rad. (d) Total cross-section as a function of p_{C} : the band extends from |\mathbf{p}_{C}|_{min}=0.276 to |\mathbf{p}_{C}|_{max}=0.733 [a.u.], where the cross-section is not defined because of the singularity. Note, for the case m_{C}<m_{A}+m_{B} , the particle C is stable, thus, no t -singularity appears. However, if m_{B}>m_{A}+m_{C} , a singularity appears at tree-level in the s channel because s=m_{B}^{2} is kinematically possible. This singularity is easily overcome by noticing that B\rightarrow AC is now open; then, resumming the propagator of the B particle amounts to the replacement s-m_{B}^{2}\rightarrow s-m_{B}^{2}+im_{B}\Gamma_{B} . Can a similar resummation solve the problem of the t -channel singularity occurring when m_{C}>m_{A}+m_{B} ? This is not the case because the particle B is truly stable. Hence, the pole of its propagator is physical and persists at any order. In order to understand better the singularity, it is useful to resort to an idealized case with a flux j_{C} of C -particle and a single particle A at rest. The number of scattered particles per second in the solid angle d\Omega is dN_{\text{scattered}}=j_{C}(d\sigma/d\Omega)d\Omega . Thus, dN_{\text{scattered}} diverges when \theta tends to the singular value. This result is troubling and differs from Coulomb scattering, where the divergence takes place for \theta\rightarrow 0 , meaning an arbitrarily imperceptible deviation of the scattered particles. Various solutions to the t -channel singularity puzzle have been proposed: • A FAPP solution proposed in Ref. Melnikov:1996iu and further elaborated in Ref. Dams:2002uy takes into account the finite size of the beam, which cures the divergences. Following the heuristic discussion above, if the flux of C -particles has a limited section, the number of incident particles per unit time is finite and the cross-section cannot diverge. Interestingly, the reduction of the cross-section due to beam-size effect has been measured in Bremsstrahlung processes Blinov:1982vp . • The exchanged particle B has a finite width because the particle B is not really stable: m_{B}^{2}\rightarrow m_{B}^{2}-im_{B}\Gamma_{B} . In the case of N^{*}\pi scattering, when the neutron is considered as the exchanged particle, the infinity is cured by the fact that the neutron has a nonzero width. Yet, since it is extremely small, the cross-section is very large. Moreover, this solution does not apply when the particle is stable, as in the case with an exchanged proton. There can be, however, other mechanisms that confer to the exchanged particle a finite width: these can be for instance