⚛️ phenomenology

Toward a Comprehensive Exploration of Flavored Dark Matter Models

This paper introduces a comprehensive, publicly available framework for analyzing flavored dark matter models that integrates relic density calculations, detection limits, collider constraints, and flavor observables, demonstrating its application to Majorana dark matter scenarios coupled to leptons and quarks.

Original authors: Benedetta Belfatto, Monika Blanke, Jan Heisig, Michael Krämer, Lena Rathmann, Felix Wilsch

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Benedetta Belfatto, Monika Blanke, Jan Heisig, Michael Krämer, Lena Rathmann, Felix Wilsch

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe is a giant, bustling city where most of the citizens are invisible. We call them Dark Matter. For a long time, scientists thought these invisible neighbors were all the same—boring, identical clones that just floated around and occasionally bumped into each other. But what if they aren't clones? What if they have personalities, families, and even different "flavors," just like humans have different tastes in music or food?

This paper is like a massive, high-tech detective kit built to investigate these "flavored" Dark Matter citizens. The authors, a team of theoretical physicists, didn't just guess; they built a digital simulation engine (a "toolchain") to test how these flavor-filled dark worlds would behave. They wanted to see if these models could survive the harsh reality of our universe, which is full of strict rules and busy police officers (experiments) looking for clues.

The Two Main Characters

To test their detective kit, the authors focused on two specific types of these flavored Dark Matter characters. Think of them as two different gangs trying to blend into the city:

The "Leptophilic" Gang: These guys only hang out with the "lepton" crowd (electrons, muons, and taus). They are shy and avoid the heavy "quark" neighborhoods.
The "Quark-Philic" Gang: These guys are the opposite; they love to mingle with the down-type quarks (the building blocks of protons and neutrons).

The authors simulated millions of scenarios for these gangs, checking if they could explain how much Dark Matter exists in the universe today (the "relic density") without getting caught by the universe's most sensitive security cameras.

The Great Filter: Who Survives?

The simulation revealed that surviving in this universe is incredibly hard. It's like trying to sneak into a VIP party where the bouncers are extremely strict.

For the Leptophilic Gang (Leptons):
The biggest hurdle for this group is a rule called "Flavor Violation." Imagine if a muon (a heavy cousin of an electron) suddenly decided to turn into an electron and shoot out a flash of light (a photon). The universe has a strict "No Muon-to-Electron-Flash" policy.

The Result: The simulation showed that if these Dark Matter particles have strong connections to both muons and electrons, they get caught immediately. The "Flavor Police" (experiments looking for that forbidden flash) rule out huge chunks of their territory.
The Survivors: Only the gangs with very specific, "hierarchical" structures survive. This means they must be very picky, mostly hanging out with one type of lepton and ignoring the others. Even then, the LHC (the world's biggest particle collider) hasn't seen them yet, but it hasn't ruled them out either. There's still a lot of space for them to hide, especially if they are heavy.

For the Quark-Philic Gang (Quarks):
This group faces an even tougher set of bouncers. Because they interact with the stuff inside our atoms, they are under constant surveillance by "Direct Detection" experiments (massive tanks of liquid xenon waiting for a bump) and "Flavor" experiments (watching how particles mix and change).

The Result: The simulation was brutal here. About 99% of the random points they tested were thrown out immediately because they didn't have the right amount of Dark Matter. Of the few that survived that, the flavor constraints (like the mixing of neutral mesons, which are like unstable particle couples) wiped out another 85% in the general scenario.
The Survivors: The only ones who made it through the gauntlet were the ones with a very specific "Bottom-Philic" structure. This means they almost exclusively hang out with the "bottom" quark (the third generation) and ignore the lighter ones. If they tried to be too friendly with the lighter quarks, the "Direct Detection" tanks would have spotted them, and the "Flavor Police" would have arrested them for causing too much mixing trouble.

The Verdict: Hierarchy is Key

The main takeaway from this digital investigation is that if Dark Matter has flavors, it can't be a chaotic free-for-all. It needs a strict hierarchy.

What is ruled out? The paper explicitly rules out models where Dark Matter couples equally to all flavors or where the couplings are messy and random. If you try to make a model where Dark Matter talks to electrons, muons, and taus with equal strength, the simulation says it's impossible—it gets crushed by the experimental limits.
What is suggested? The authors suggest that the only viable models are those where the Dark Matter has a "favorite" generation (like the bottom quark or a specific lepton) and treats the others with extreme distance. This "hierarchical" structure is the only way to sneak past the flavor constraints.

How Sure Are We?

It's important to remember that this is a simulation. The authors didn't find a new particle in a lab; they built a computer model to see what would happen if these particles existed.

They used known physics rules (like the Standard Model) and plugged in millions of different numbers for mass and strength of interaction.
They checked these numbers against real-world data from the LHC, the Planck satellite, and various underground detectors.
The results show that while these flavored models could exist, they are extremely constrained. They aren't "proven" to be the answer, but the paper proves that if they are the answer, they must follow very specific, hierarchical rules.

In short, the universe seems to be saying: "If you want to be Dark Matter with a personality, you'd better be very specific about who you hang out with, or you'll get caught." The authors have handed the scientific community a new, powerful toolkit to keep hunting for these elusive, flavor-filled ghosts.

Technical Summary: Toward a Comprehensive Exploration of Flavored Dark Matter Models

Problem Statement
The Standard Model (SM) successfully describes laboratory experiments but fails to explain fundamental issues such as the origin of flavor hierarchies, the matter-antimatter asymmetry, and the nature of Dark Matter (DM). While simplified models of DM often introduce a single DM particle and a mediator, these frameworks frequently neglect the potential for a non-trivial flavor structure within the dark sector. Flavored DM models, where the DM candidate is the lightest state of a flavor multiplet, offer phenomenological advantages, such as relieving tensions between relic density and direct detection limits. However, significant portions of the flavored DM parameter space remain unexplored, particularly regarding the interplay between thermal freeze-out, flavor-violating constraints, and collider searches. Previous studies have often been limited to specific sectors (e.g., Minimal Flavor Violation) or specific particle types, lacking a unified framework to simultaneously constrain these models across all experimental frontiers.

Methodology
The authors present a comprehensive numerical toolchain designed to facilitate the phenomenological study of flavored DM models. This framework integrates several publicly available tools to perform a global analysis:

Model Generation: FeynRules is used to derive Feynman rules and generate CalcHEP and UFO model files for 20 distinct scenarios involving scalar or fermionic DM (real/complex, Dirac/Majorana) coupled to various SM fermions via a t-channel mediator.
Relic Density and Detection: micrOMEGAs is employed to calculate the thermal relic abundance (assuming canonical freeze-out), direct detection cross-sections (spin-independent and spin-dependent), and indirect detection annihilation cross-sections.
Collider Constraints: SModelS v3 decomposes the full model into simplified topologies to test against ATLAS and CMS results, returning exclusion $r$ -values.
Flavor Physics: The framework utilizes a Standard Model Effective Field Theory (SMEFT) approach. Matchete performs one-loop matching of the DM models onto the SMEFT (Warsaw basis). The Wilson package then evolves these coefficients via Renormalization Group (RG) equations down to the electroweak scale and subsequently to the Low Energy Effective Theory (LEFT). Finally, smelli (based on flavio) computes a global likelihood for flavor observables, including meson mixing, CP violation, and lepton flavor violation (LFV).

The toolchain is validated and demonstrated through a proof-of-principle analysis of two specific Majorana DM scenarios:

Scenario A: DM coupling to right-handed charged leptons (leptophilic).
Scenario B: DM coupling to right-handed down-type quarks (quarkphilic).

In both cases, the authors assume Dark Minimal Flavor Violation (DMFV), where the coupling matrix $\lambda$ is the sole source of new flavor and CP violation.

Key Contributions and Results

Leptophilic Scenario (Right-handed Charged Leptons):
- Constraints: The relic density requirement is the most restrictive, excluding approximately 97% of the scanned parameter space. Among the surviving points, flavor constraints (specifically $\mu \to e\gamma$ and $\mu \to eee$ ) exclude an additional ~25% of the parameter space.
- Collider Limits: LHC searches are relatively weak for this scenario, excluding only 0.3% of the relic-density-compliant points, primarily in the low-mass region ( $M_{\tilde{\chi}_1} < 200$ GeV, $M_\phi < 500$ GeV).
- Direct/Indirect Detection: Direct detection constraints are negligible due to the absence of tree-level interactions with nuclei. Indirect detection constraints are also weak because the leading annihilation channel is p-wave suppressed.
- Viable Structure: The surviving parameter space favors hierarchical coupling structures where off-diagonal elements in the coupling matrix are suppressed to satisfy LFV bounds.
Quarkphilic Scenario (Right-handed Down-type Quarks):
- Constraints: This scenario faces significantly tighter constraints. Relic density excludes ~99% of points. Direct detection (LZ, XENON1T) excludes an additional 32.5% in a general hierarchical scan (due to tree-level quark couplings) but only ~5% in a "bottom-philic" scan.
- Flavor Physics: Flavor constraints are the dominant limitation, excluding an additional 85.3% of the remaining points in the general hierarchical scenario. The most stringent bound arises from the CP-violating parameter $\epsilon_K$ in $K^0-\bar{K}^0$ mixing.
- Scan Strategies: To efficiently probe viable regions, the authors implement two specific scan setups:
  1. General Hierarchical: Samples coupling eigenvalues to allow various hierarchies.
  2. Bottom-Philic: Assumes a strong hierarchy ( $D_3 \gg D_2 \gg D_1$ ) and small mixing angles, mimicking a dominant coupling to the third generation.
- Results: The bottom-philic scenario is significantly more viable, with 975 allowed points compared to 168 in the general hierarchical case. This confirms that hierarchical flavor structures are essential for evading simultaneous constraints from meson mixing and direct detection.
- Collider Limits: LHC constraints are stronger than in the leptophilic case, excluding ~30% of relic-compliant points in the general scan and ~24% in the bottom-philic scan, primarily via di-jet + missing energy signatures.

Significance
The paper claims to provide a generic and flexible framework for the study of flavored DM models, addressing a gap in the literature where large portions of the model space remain unexplored. By combining relic density, detection limits, collider data, and a rigorous SMEFT-based flavor analysis, the authors demonstrate that flavored DM models are heavily constrained by flavor observables, particularly meson mixing and LFV. The results quantitatively support the necessity of hierarchical flavor structures in the coupling matrix to remain phenomenologically viable. The authors emphasize that their toolchain, publicly available on GitHub, enables the extension of these results to other scenarios, including those with left-handed mediators or scalar DM, which were not the focus of this specific proof-of-principle study. The work serves as a methodological foundation for future model-building efforts aimed at naturally hierarchical couplings in the dark sector.

The Two Main Characters

The Great Filter: Who Survives?

The Verdict: Hierarchy is Key

How Sure Are We?

More like this