Discovering hidden sectors with mono-photon Z' searches

TL;DR

This work addresses the challenge of detecting hidden-sector states coupled to a Z' boson by studying invisible Z' decays in the mono-photon channel at the LHC. The authors show that the signal can be captured with a simple description based on the Z' mass $M_{Z'}$ and two effective charges, and they exploit on-peak Drell-Yan measurements to predict those charges if decays to SM neutrinos are the only invisible channel. A thorough background analysis and consideration of PDF-related uncertainties demonstrate that a 1 TeV Z' could be discovered with about 50 fb^{-1}, while a 1.5 TeV state would require about 1 ab^{-1}, with the reach limited by the normalization of the dominant $Z\gamma$ background. The study provides a path to separating hidden-sector decays from SM neutrino decays and highlights the importance of controlling systematic errors in Z gamma predictions to maximize the LHC sensitivity to hidden sectors.

Abstract

In many theories of physics beyond the Standard Model, from extra dimensions to Hidden Valleys and models of dark matter, Z' bosons mediate between Standard Model particles and hidden sector states. We study the feasibility of observing such hidden states through an invisibly decaying Z' at the LHC. We focus on the process pp -> γZ' -> γX X*, where X is any neutral, (quasi-) stable particle, whether a Standard Model (SM) neutrino or a new state. This complements a previous study using pp -> Z Z' -> l+ l- X X*. Only the Z' mass and two effective charges are needed to describe this process. If the Z' decays invisibly only to Standard Model neutrinos, then these charges are predicted by observation of the Z' through the Drell-Yan process, allowing discrimination between Z' decays to SM neutrinos and invisible decays to new states. We carefully discuss all backgrounds and systematic errors that affect this search. We find that hidden sector decays of a 1 TeV Z' can be observed at 5 sigma significance with 50 fb^{-1} at the LHC. Observation of a 1.5 TeV state requires super-LHC statistics of 1 ab^{-1}. Control of the systematic errors, in particular the parton distribution function uncertainty of the dominant Z γbackground, is crucial to maximize the LHC search

Figures (4)

• Figure 1: Example diagram giving rise to the signal process $pp \to \gamma Z' \to \nu\bar{\nu}+\not\!\!\!E_T$. The particle labeled $\nu$ can denote either a SM neutrino or hidden sector state.
• Figure 2: Plot of signal and various backgrounds as a function of the photon $p_T$. For the signal, a 1 TeV $U(1)_{\chi}$ model $Z'$ with an additional hidden state has been assumed. The last bin includes all photons with $p_T > 400$ GeV. We note that the various histogrammed backgrounds are stacked atop each other.
• Figure 3: Effects of "realistic" systematic and statistical errors in each $p_T^{\gamma}$ bin. Last bin includes overflows. The $U(1)_{\chi}$ model $Z'$ with an additional hidden state has been assumed.
• Figure 4: Required integrated luminosity for $3\sigma$ evidence of hidden sector $Z'$ decays for a given cross section $\sigma$ into hidden states, including the cut $p_T^{\gamma}>100$ GeV. We have assumed the realistic errors described in the text.