Probing Ultra-Compressed Scotogenic Dark Matter at the HL-LHC via 4D Spacetime Tracking

Technical Summary: Probing Ultra-Compressed Scotogenic Dark Matter at the HL-LHC via 4D Spacetime Tracking

Problem Statement
The minimal scotogenic model offers a compelling framework for simultaneously generating neutrino masses and a dark matter (DM) candidate. A specific regime of interest involves the fermionic DM branch where the thermal relic density is achieved via co-annihilation between the dark matter fermion ($N_1$) and a charged inert scalar ($\tilde{\eta}^\pm$). This mechanism requires a compressed mass spectrum with a splitting $\Delta M \equiv m_{\tilde{\eta}^\pm} - m_{N_1} \sim \mathcal{O}(\text{few GeV})$.

The paper identifies a "blind spot" in current Large Hadron Collider (LHC) searches for the ultra-compressed regime where $\Delta M \lesssim 2$ GeV. In this scenario, the decay $\tilde{\eta}^\pm \to \ell^\pm N_1$ produces a soft charged lepton with a momentum in the parent rest frame of $p^* \simeq \Delta M \lesssim 2$ GeV.

• Limitations of Existing Searches: Standard displaced-lepton searches typically require $p_T \gtrsim 30\text{--}60$ GeV. While some recent searches reach lower thresholds (e.g., $p_T > 15$ GeV), they target topologies with two displaced tracks or vertices, which do not match the single soft daughter of the scotogenic decay.
• Reconstruction Challenges: For $\Delta M \lesssim 2$ GeV, the laboratory-frame $p_T$ of the daughter lepton sits at the edge of standard 3D track reconstruction thresholds ($p_T \gtrsim 2$ GeV). Furthermore, the gyration radius of such low-momentum particles in the 3.8 T magnetic field ($r \approx 1.8$ m) often exceeds the tracker radius, leading to marginal reconstruction efficiency. Existing disappearing-track searches retain only partial, unoptimised acceptance for this specific topology.

Methodology
The author proposes a dedicated "4D spacetime tracking" strategy for the High-Luminosity LHC (HL-LHC), utilizing the precision timing capabilities of the CMS MIP Timing Detector (MTD) as a fourth observable alongside spatial coordinates.

• Signal Topology: The signal arises from electroweak associated production $pp \to W^* \to \tilde{\eta}^\pm \tilde{\eta}^0$. The neutral scalar $\tilde{\eta}^0$ decays invisibly, while the charged scalar $\tilde{\eta}^\pm$ travels macroscopically ($d_{\text{vtx}} \sim 10\text{--}1000$ mm) before decaying into a soft lepton and invisible $N_1$.
• The 4D Strategy: The core discriminant is the arrival-time delay ($\Delta t$) of the daughter lepton at the MTD barrel ($R = 1.17$ m).
  • The parent $\tilde{\eta}^\pm$ is slow ($\beta \approx 0.35$ for $m=200$ GeV), while the daughter lepton is ultra-relativistic ($\beta \approx 1$).
  • This velocity difference creates a macroscopic time delay: $\Delta t \approx d_{\text{vtx}}/c \cdot (1/\beta_{\tilde{\eta}} - 1)$. For a benchmark of $m_{\tilde{\eta}^\pm}=200$ GeV and $c\tau=300$ mm, $\Delta t \approx 693$ ps.
  • This delay is orders of magnitude larger than the sub-picosecond delays of Standard Model (SM) backgrounds ($\Delta t_{\text{bkg}} \lesssim 0.2$ ps) and the detector resolution ($\sigma_t \approx 30$ ps).
• Selection Criteria (SR-4DT): The analysis defines a signal region requiring:
  1. High missing transverse energy ($E_T^{\text{miss}} > 105$ GeV trigger, $>150$ GeV signal).
  2. A leading jet ($p_T > 100$ GeV).
  3. A low-$p_T$ disappearing-track stub ($p_T > 0.7$ GeV, $|d_0| \in [1, 300]$ mm) that terminates inside the tracker.
  4. A delayed timing hit in the MTD barrel ($\Delta t > 200$ ps) spatiotemporally matched to the stub ($\Delta R < 0.1$).
  5. A veto on prompt leptons ($p_T > 25$ GeV).
• Simulation: Signal events were generated at LO with MadGraph5_aMC@NLO and processed through Pythia 8 and Delphes 3.5 with a custom MTD card ($\sigma_t = 30$ ps). Backgrounds ($W/Z/t\bar{t} + j$) were generated with tightened phase-space cuts to oversample kinematic tails. A total of $4.20 \times 10^6$ background events were simulated.

Key Contributions and Results

• Background Suppression: The strategy achieves zero surviving background events across $4.20 \times 10^6$ simulated prompt SM events. The $\Delta t > 200$ ps requirement suppresses SM backgrounds by a factor of $\sim 10^{-11}$ per event (a $6.7\sigma_t$ fluctuation). The mandatory association of the timing hit with a low-$p_T$ disappearing stub further suppresses accidental pile-up coincidences.
• Signal Efficiency: The SR-4DT selection achieves signal efficiencies of 12–36% across the probed mass range. For the benchmark ($m_{\tilde{\eta}^\pm} = 200$ GeV, $c\tau = 300$ mm), the efficiency is 14.05%, yielding approximately 2,331 signal events at 3000 fb$^{-1}$.
• Exclusion Reach:
  • Under conservative background assumptions (accounting for the statistical upper limit of the zero-background MC sample, $B_{95} < 438$ at 3000 fb$^{-1}$), the analysis projects a 95% C.L. exclusion reach up to scalar masses of $\simeq 670$ GeV for $c\tau = 300$ mm.
  • Under nominal zero-background assumptions, the reach extends to $\simeq 1100$ GeV.
  • The strategy remains sensitive across a wide lifetime range ($c\tau \in [10, 3000]$ mm), though efficiency drops at very short lifetimes (insufficient $\Delta t$) and very long lifetimes (decay outside the tracker volume).
• Comparison with Existing Searches: Truth-level estimates indicate that existing searches (ATLAS/CMS disappearing tracks, soft displaced tracks, displaced vertices) have partial or negligible acceptance for this topology. For the benchmark point, existing searches show efficiencies ranging from 0% to 8.5%, significantly lower than the 14% achieved by the proposed 4D strategy.

Significance and Claims
The paper claims that the 4D spacetime tracking strategy successfully opens a discovery window for the "ultra-compressed" scotogenic regime ($\Delta M \simeq 2$ GeV), which is conventionally the most elusive region for LHC searches.

• Motivation: This regime is motivated by the co-annihilation mechanism required for the correct thermal relic density.
• Compatibility: The small Yukawa couplings ($y \sim 10^{-7}$) required to produce the optimal decay lengths ($c\tau \sim 300$ mm) are compatible with sub-eV neutrino masses. The paper clarifies that while neutrino data constrain the product $y^2\lambda_5$, they do not uniquely fix $y$ alone, allowing the charged scalar lifetime to be a free phenomenological parameter within the experimentally accessible range.
• Robustness: The author emphasizes that while the zero-background result is derived from Delphes-level simulations (which do not model instrumental or combinatorial backgrounds), the physics arguments regarding the separation of timescales and the geometric pointing requirements provide strong motivation that the background remains negligible. The strategy relies on the unique combination of a macroscopic decay length and a precision timing layer to convert a spatial displacement into a temporal delay, bypassing the $p_T$ thresholds that limit conventional 3D tracking.

The study concludes that this dedicated timing-assisted approach makes the co-annihilation corridor an accessible target for the HL-LHC, extending the coverage of the scotogenic parameter space beyond the reach of current 3D displaced-track reconstruction techniques.