Exploiting dijet resonance searches for flavor physics

Published for SISSA by Springer Received: April 1, 2021 Revised: July 10, 2021 Accepted: July 22, 2021 Published: August 9, 2021 Exploiting dijet resonance searches for flavor physics a Dipartimento di Fisica, Università di Torino & INFN — Sezione di Torino, I-10125 Torino, Italy b Theoretische Physik 1, Naturwissenschaftlich-Technische Fakultät, Universität Siegen, Walter-Flex-Straße 3, D-57068 Siegen, Germany c Albert Einstein Center for Fundamental Physics, Institut für Theoretische Physik, Universität Bern, Sidlerstrasse 5, CH-3012 Bern, Switzerland d CERN, Theoretical Physics Department, CH-1211 Geneva 23, Switzerland e INFN — Sezione di Trieste, SISSA, Via Bonomea 265, 34136, Trieste, Italy E-mail: , , Abstract: In this work, we reinterpret ATLAS and CMS dijet resonance searches to set robust constraints on all hypothetical tree-level scalar and vector mediators with masses up to 5 TeV, assuming a diquark or a quark-antiquark coupling with an arbitrary flavor composition. To illustrate the application of these general results, we quantify the per(∗)+ missible size of new physics in B̄q → Dq {π, K} consistent with the absence of signal in dijet resonance searches. Along the way, we perform a full SMEFT analysis of the aforementioned non-leptonic B meson decays at leading-order in αs . Our findings uncover a pressing tension between the new physics explanations of recently reported anomalies in these decays and the dijet resonant searches. The high-pT constraints are crucial to drain the parameter space consistent with the low-pT flavor physics data. Keywords: Beyond Standard Model, Heavy Quark Physics ArXiv ePrint: 2103.10332 Open Access, c The Authors. Article funded by SCOAP3 . https://doi.org/10.1007/JHEP08(2021)036 JHEP08(2021)036 Marzia Bordone,a,b Admir Greljoc,d and David Marzoccae Contents 1 2 Dijet searches at high-pT 2.1 Pair production of dijet resonances 2.2 Dijet resonance 3 4 6 3 General EFT analysis of b → cudi 11 4 Simplified models 4.1 Colored mediators 4.2 Colorless scalar doublet model 14 15 19 5 Conclusions 22 A Details of the EFT analysis 24 B Details on the tree-level mediators 29 C Flavor constraints on the scalar sextet Φ6 34 D Flavor constraints on the colorless scalar Φ1 37 1 Introduction Non-leptonic decays are a challenging playground for the Standard Model (SM), particularly for non-perturbative approaches to quantum chromodynamics (QCD). A question which is often neglected in the literature is how much new physics can hide in these decays? In particular, given the present constraints from complementary new physics searches at low and high energies, what is the allowed deviation from the SM predictions? To address this question, we exploit dijet resonance searches at high-pT colliders as a complementary probe of the hypothetical new physics entering non-leptonic decays. We imagine a bosonic mediator X coupled to quarks which, on the one hand, modifies the low-energy meson decays, while on the other hand, can be directly produced and (or) decayed at the LHC by the same interactions. Several challenges are stemming from the lack of knowledge of the underlying microscopic theory beyond the SM. Firstly, there is a broad range of relevant masses that needs to be covered. Here we consider mX up to 5 TeV relying crucially on the latest experimental progress on dijet resonance searches by ATLAS and CMS collaborations [1–4]. Secondly, there is a variety of possible representations and flavor couplings of the mediator X, for which the expected signal rates in pp collisions –1– JHEP08(2021)036 1 Introduction –2– JHEP08(2021)036 differ. One of the main outcomes of our study is the reinterpretation of the existing dijet resonance searches in a general form applicable to flavor physics. Note that, in the narrow-width approximation, the presence of several couplings can only add up in the total pp → X → jj rate. This is a very useful feature for flavor physics. Non-leptonic meson decays depend on the product of two couplings when the resonance X is integrated out at tree level. The absence of the signal in the dijet resonance searches sets limits on all Xq i q j couplings simultaneously, thus also on their product. In order to exemplify the importance of these results, we investigate potential new (∗)+ physics (NP) effects in the branching ratio of B̄s → Ds π and B̄ → D(∗)+ K decays. These decays are mediated by the underlying b → cūdi quark-level transitions, where di = d, s, rendering their SM theory predictions amongst the most reliable in the sector of nonleptonic decays and are obtained in the framework of QCD factorisation (QCDF) [5]. Since the quarks entering in these decays are distinguishable, topologies like penguin contribution or weak annihilation do not contribute, rendering the description of these decays rather clean. The most up-to-date predictions for the branching ratios have been presented in ref. [6]. Next-to-leading power corrections, arising at order O(ΛQCD /mb ), are found to be subleading compared to the leading-power ones, strengthening the predictive power of QCDF for these channels. The possibility of NP effects in four-quarks operators has been already entertained, with the focus on low-energy inclusive observables, see for example refs. [7–12]. Interestingly, the aforementioned update uncovered an intriguing tension with the data, not yet thoroughly analyzed in the NP context. A fit to all the available experimental information (∗)+ concerning B̄s → Ds π and B̄ → D(∗)+ K decays is performed, and the current combination of the experimental measurements for their branching ratios is extracted [6]. The comparison with the respective theory predictions shows that the latter always overestimate the former, with a combined discrepancy of about 4.4σ [6]. This trend has already been observed in the literature (see for example refs. [5, 12–14]) but has become more apparent due to the updated theory results in ref. [6]. A satisfactory explanation of this puzzle is not yet articulated. On the theory side, the hypothesis of a big deviation due to the missing subleading contributions in QCDF seems to be unlikely, since they overshoot the current estimates by at least one order of magnitude. Hence, it seems motivated to entertain the possibility of this deviation being due to NP. In this paper, we try to understand to which extend the NP solution is viable, especially in connection with bounds from related processes, most notably dijet searches at high-pT . The bounds that we obtain from dijet searches can be applied to a broader class of four quark operators, beyond the ones mediating b → cūd(s) transitions [10, 15–34]. The paper is organised as follows. In section 2 we study dijet searches at the LHC to set generic constraints on hypothetical new resonances. In section 2.1 we discuss the constraints on the pair production pp → XX → (jj)(jj) from gauge interactions and subsequent decay into je (...truncated)