First broadband search for fixed-polarization dark photon dark matter with GigaBREAD
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: High Energy Physics: Experiment. The day in high-energy experiment: detectors, searches and limits.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "First broadband search for fixed-polarization dark photon dark matter with GigaBREAD".
Mira: Here's the gist of the paper. First broadband search for fixed-polarization dark photon dark matter with GigaBREAD is presented to test whether vector dark matter produced by a misalignment-like mechanism may retain an approximately fixed polarization, leading to sidereal modulation of the detector response as the Earth rotates The authors reanalyze the GigaBREAD data using a polarization-matched likelihood that accounts for the time-dependent projection of a fixed DP polarization onto the detector axis They find no evidence for a fixed-polarization DP signal and set upper limits on the kinetic mixing parameter at the level of χ ∼ 10−12 These limits are about 20% better than a naive limit obtained by rescaling the previous GigaBREAD limit under the fixed-polarization assumption.
Kai: Signal Modeling and Polarization Dependence The expected discrete signal Skj in the j-th frequency bin of the k-th time bin is modeled as Skj = P0(χ, mDP) ηkj Cj (mDP) cos2ϑ(tk), (2) This model incorporates conversion power, polarization modulation, and frequency-bin response.
Mira: The polarization factor cos 2ϑ(t) arises from the relative geometry between the DP polarization and the detector orientation The instantaneous polarization factor can be written as cos 2ϑ(θX, ϕX, t) = (Xˆ · Zˆ(t) 2, zenith-aligned, Xˆ · Vˆ (ϕd, t) 2, horizontal This factor is expanded into three distinct time dependencies: a constant term Cconst, a daily modulation Cω⊕ (t) related to the sidereal frequency ω⊕, and a twice-daily modulation C2ω⊕ (t) related to 2ω⊕.
Kai: Statistical Framework for Likelihood Analysis The probability of obtaining the observed data is described by the likelihood function as L(DP0, mDP, θX, ϕX) = exp
− 1/2 Xk,j ykj − Skj σkj 2: The log-likelihood is further decomposed into data-only contribution Λdata, noise-weighted overlap Λcross, and template selfoverlap Λtemp To accelerate the likelihood evaluation, the authors exploit the periodicity of geometric factors by expanding cos 2ϑ and cos 4ϑ into a finite series of sidereal harmonics up to n = 2 and n = 4, respectively.
Paper summary: Kai: So, thinking about the title of this paper, "First broadband search for fixed-polarization dark photon dark matter with GigaBREAD", it really tells you what they were trying to do. It's about using a specific kind of detector, GigaBREAD, to look for a particular type of particle interaction.
Mira: And the core idea is testing the assumption that vector dark matter might maintain its polarization over time, which introduces this sidereal modulation we discussed earlier. So they are really probing the physics of how this dark matter interacts with light and spacetime geometry on Earth.
Lev: From my side, it’s interesting because it shows how you can use the full bandwidth of a dish antenna experiment for extended periods to look for these subtle time variations, even when you have to deal with the complexity of polarization factors.
Kai: Right. And what they actually did is they found no signal in that mass range, leading them to set an upper limit on kinetic mixing at ten to the power of negative twelve. That's a concrete number we can use for planning future searches.
Mira: So, the implication is that if this kind of fixed-polarization dark matter exists, it should show up as a modulation in our data when we look at different parts of the sky over several days.
Lev: It means if we build future experiments, they need to be sensitive enough to catch these time-dependent signals instead of just looking for a steady one.
Kai: Exactly. And this methodology they developed here, this way of handling the polarization dependence through harmonic expansion, can be directly applied to other broadband dark photon searches and similar experiments involving time-dependent or polarization-dependent signals.
Mira: So we’ve seen that the approach works for this specific search, and it gives us a way forward for future experiments looking at these kinds of signals.
Conclusion: Kai: So, this paper is about doing a broadband search for dark photons using this specific detector called GigaBREAD and testing if they can keep their polarization fixed while being dark matter.
Mira: Right, so the authors are looking at whether vector dark matter produced in a specific way might actually have a steady polarization, which would cause a pattern to show up when the Earth rotates.
Lev: It's interesting because it’s not just looking for any dark photon signal; they’re specifically modeling how that polarization changes with the detector’s orientation on Earth.
Kai: Exactly, and what they found is no statistically significant evidence for this fixed-polarization signal in that mass range of forty-four to fifty two microelectronvolts.
Mira: So, they set an upper limit on the kinetic mixing parameter at ten to the power of negative twelve. That means if it’s there, it’s really weak.
Lev: And what they did is they reanalyzed the GigaBREAD data using a likelihood that accounts for this time-dependent projection of the polarization onto the detector axis.
Kai: That makes sense, because assuming a fixed polarization and just averaging everything out would give you a much looser limit, which is why this new analysis helps.
Mira: The authors used both Bayesian and frequentist methods to treat those unknown polarization directions as nuisance parameters during their search for the signal.
Lev: From an engineering standpoint, that harmonic expansion they did to speed up the likelihood evaluation is smart because it lets them precompute things that don't depend on the specific angles.
Kai: So, they are telling us that this type of dark matter interaction, if it has a fixed polarization, should be incredibly subtle and is currently constrained to be very weak.
Mira: It points toward the fact that vector dark matter produced by a misalignment mechanism might indeed retain that steady polarization over time.
Lev: And this result at ten to the power of negative twelve sets a strong constraint for future experiments trying to find these specific types of signals.
Jialin Yu, *Stefan Knirck, ^Gabe Hoshino, ^Rakshya Khatiwada, Alex Lapuente, David W. Miller, Andrew Sonnenschein
Illinois Institute of Technology Department of Physics Department of Physics Laboratory for Particle Physics and Cosmology Department of Physics University of Chicago Enrico Fermi Institute Kavli Institute for Cosmological Physics
hep-ex, astro-ph.IM, hep-ph, physics.ins-det
Submitted: 2026-11-05
Updated: 2026-11-05
Comments: 14 pages, 6 figures
Code: https://github.com/cajohare/AxionLimits
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 80/100
The gist: First broadband search for fixed-polarization dark photon dark matter with GigaBREAD is presented to test whether vector dark matter produced by a misalignment-like mechanism may retain an…
Key concepts
- Dark Photon (DP)
- A hypothetical particle that mixes with photons and could be a candidate for dark matter. This paper specifically looks for a dark photon with a fixed polarization, testing if its production mechanism leaves this property intact.
- Sidereal Modulation
- A time-dependent signal caused by the Earth's rotation. If the dark photon has a fixed polarization, its interaction strength with a detector will change predictably as the detector rotates relative to the fixed polarization direction in space.
- Kinetic Mixing Parameter ($\chi$)
- A measure of how strongly a dark photon mixes with ordinary photons. The search sets upper limits on this parameter, which quantifies the potential coupling strength between dark matter and light particles like photons.
- Likelihood Analysis
- A statistical method used to determine if the observed data matches a theoretical signal. The authors use complex likelihood functions that account for how the signal's polarization changes over time due to Earth's rotation.
Terminology
Summary
First broadband search for fixed-polarization dark photon dark matter with GigaBREAD is presented to test whether vector dark matter produced by a misalignment-like mechanism may retain an approximately fixed polarization, leading to sidereal modulation of the detector response as the Earth rotates The authors reanalyze the GigaBREAD data using a polarization-matched likelihood that accounts for the time-dependent projection of a fixed DP polarization onto the detector axis They find no evidence for a fixed-polarization DP signal and set upper limits on the kinetic mixing parameter at the level of χ ∼ 10−12 These limits are about 20% better than a naive limit obtained by rescaling the previous GigaBREAD limit under the fixed-polarization assumption. p. 0 However, vector dark matter produced by a misalignment-like mechanism may retain an approximately
fixed polarization, leading to sidereal modulation of the detector response as the Earth rotates. We reanalyze the GigaBREAD data using a polarization-matched likelihood that accounts for the
time-dependent projection of a fixed DP polarization onto the detector axis.Where the paper says it
Signal Modeling and Polarization Dependence
The expected discrete signal Skj in the j-th frequency bin of the k-th time bin is modeled as Skj = P0(χ, mDP) ηkj Cj (mDP) cos2ϑ(tk), (2) This model incorporates conversion power, polarization modulation, and frequency-bin response. The polarization factor cos 2ϑ(t) arises from the relative geometry between the DP polarization and the detector orientation The instantaneous polarization factor can be written as cos 2ϑ(θX, ϕX, t) = (Xˆ · Zˆ(t) 2, zenith-aligned, Xˆ · Vˆ (ϕd, t) 2, horizontal This factor is expanded into three distinct time dependencies: a constant term Cconst, a daily modulation Cω⊕ (t) related to the sidereal frequency ω⊕, and a twice-daily modulation C2ω⊕ (t) related to 2ω⊕. p. 2 The instantaneous polarization factor can then be written as
cos2ϑ(θX, ϕX, t) = (
|Xˆ · Zˆ(t)|
2
, zenith-aligned ,
|Xˆ · Vˆ (ϕd, t)|
2
, horizontal
(7)
We can expand |Xˆ ·Zˆ(t)|
2 as
|Xˆ ·Zˆ(t)|
2 = Cconst + Cdaily(t) + C2−daily(t)
revealing three distinct time dependencies [4]:
• Constant term (time-independent):
Cconst =
1
2
cos2λ cos2
θX + sin2λ sin2
θX. • Daily modulation (first harmonic of sidereal
frequency, ω⊕):
Cω⊕ (t) = 2 cos λ sin λ cos θX sin θX cos(ω⊕t − ϕX).Where the paper says it
Statistical Framework for Likelihood Analysis
The probability of obtaining the observed data is described by the likelihood function as L(DP0, mDP, θX, ϕX) = exp [− 1/2 Xk,j ykj − Skj σkj 2] The log-likelihood is further decomposed into data-only contribution Λdata, noise-weighted overlap Λcross, and template selfoverlap Λtemp To accelerate the likelihood evaluation, the authors exploit the periodicity of geometric factors by expanding cos 2ϑ and cos 4ϑ into a finite series of sidereal harmonics up to n = 2 and n = 4, respectively. This harmonic decomposition allows for the factorization of cross and template terms into precomputable harmonic components independent of polarization angles. p. 4 This harmonic decomposition allows us to
factorize the cross and template terms. For instance,
Λcross(θX, ϕX, mDP) can be rewritten as a linear combination of frequency-dependent, precomputable data sums
and analytically derived angular coefficients:
Λcross =
X
j
Cj (mDP)
X
2
n=0
h
b
(2)
n,c(θX, ϕX) C
cross
n,j
+ b
(2)
n,s(θX, ϕX) S
cross
n,j i
(15)
where the geometric coefficients b
(2)
n,c and b
(2)
n,s are analytical functions of the polarization angles (θX, ϕX), independent of time and frequency.Where the paper says it
Parameter Treatment and Inference
The likelihood depends on the full parameter vector ψ ≡ (P0, mDP, ν), where P0 is the parameter of interest and ν ≡ (θX, ϕX) denotes nuisance parameters describing the unknown polarization direction Both Bayesian and frequentist approaches are adopted to treat these nuisance parameters. The Bayesian approach marginalizes over ν with a prior π(ν), while the frequentist approach profiles over them by maximizing the likelihood LFreq(DP0, mDP) = max ν L(DP0, ν, mDP). p. 5 To obtain
an inference that depends only on P0 (for each mDP), we
marginalize over ν with a prior π(ν), ν = (θX, ϕX):
LBayes(D|P0, mDP) = Z
L(D|P0, ν, mDP) π(ν) dν. (22)
Frequentist approach: In our frequentist analysis,
we treat the polarization angles as fixed but unknown
nuisance parameters and profile over them by maximizing
the likelihood [22, 25]:
LFreq(D|P0, mDP) = max
ν
L(D|P0, ν, mDP). Consequently, the azimuthal polarization angle ϕX was treated as an independent nuisance parameter
for each run: it was marginalized (separately summed)
within each run in the Bayesian combination and profiled within each run in the frequentist case. The overall product across runs is therefore taken after the inner
summation or maximization over ϕX.Where the paper says it
Where the paper says it
Search Results and Limits
The analysis finds no statistically significant evidence for a dark photon signal within the surveyed mass range of 44–52 µeV Consequently, upper limits are set on the DP conversion power P0 as a function of DP mass mDP. The isotropic result represents the original GigaBREAD2023 limit reported in Ref. p. 8 Figure 5 compares the updated GigaBREAD limits with previously
published results. The isotropic result represents the original GigaBREAD2023 limit reported in Ref. 254 on time averaged data over the observation duration T, following the implementation in
AxionLimits [27]. The two lower curves show the fixedpolarization limits obtained by applying our Bayesian
and frequentist methods to the same GigaBREAD 2023
dataset. The two lower curves show the fixedpolarization limits obtained by applying our Bayesian
and frequentist methods to the same GigaBREAD 2023
dataset. These limits are about 20% lower in χ than the
previously rescaled limit and are comparable to the original isotropic limit.Where the paper says it
Where the paper says it
Where the paper says it
Implications for Experimental Design
The optimal detector geometries for observations spanning an integer number of sidereal days, assuming uniform time weighting, are summarized in Table II For a zenith-pointing detector, this occurs at a half-opening angle of 54.7◦ (λ = 35.3◦) and gives ⟨cos 2ϑ(t)⟩T = 1/3, consistent with Ref. p. 8 7
◦
(|λ| = 35. 3
◦
)
and gives ⟨cos2 ϑ(t)⟩T = 1/3, consistent with Ref. This gives more weight to times
when the detector is closely aligned with the DP polarization, improving sensitivity compared with time averaging
alone. As a result, the optimal half-opening angle decreases from 54. Previous searches, including GigaBREAD
and several other dish antenna searches, have assumed randomly polarized DP dark matter. However, vector dark matter produced by a misalignment-like mechanism may retain an approximately
fixed polarization, leading to sidereal modulation of the detector response as the Earth rotates.Where the paper says it
Where the paper says it
Where the paper says it
Conclusion
Vector dark matter produced by a misalignment mechanism may retain an approximately fixed polarization, leading to a time-dependent signal modulation due to Earth’s rotation In this work, they find no evidence for a dark photon signal and set fixed-polarization limits on the kinetic mixing parameter χ ∼ 10−12 over 44–52 µeV mass range Broadband dish antennas can observe the full bandwidth of the experiment for the duration of the experimental run, making them particularly suitable for this type of analysis. The methodology developed here can be directly applied to other broadband dark-photon searches and similar experiments involving time-dependent or polarization-dependent signals. p. 9 Future extensions of this analysis will explore additional data sets and further refinements of the statistical
framework. The methodology developed here can be directly applied to other broadband dark-photon searches
and similar experiments involving time-dependent or
polarization-dependent signals.Where the paper says it
Acknowledgments
The authors acknowledge Abigail Vieregg and her group for providing access to the RF-shielded room and RF absorber materials They thank Argonne National Laboratory, operated under Contract No. DE-AC02-06CH11357, for use of the solenoid magnet facility which enabled the search for axion-like particles This work was funded by the U. p. 9 We thank Argonne National Laboratory, operated under Contract No. DE-AC02-06CH11357, for
use of the solenoid magnet facility, which enabled the
search for axion-like particles. This work was funded by
the U. Department of Energy through the resources of
the Fermi National Accelerator Laboratory (Fermilab),
a DOE Office of Science HEP User Facility. First broadband search for fixed-polarization dark photon dark matter with
GigaBREAD
Jialin Yu,1, 2, ∗ Stefan Knirck,3, † Gabe Hoshino,4, 5, 6, ‡ Rakshya Khatiwada,1, 2
Alex Lapuente,4, § David W. Miller,4, 5, 6 and Andrew Sonnenschein2, 5
1Department of Physics, Illinois Institute of Technology, Chicago, Illinois 60616, USA
2Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA
3Laboratory for Particle Physics and Cosmology, Department of Physics,
Harvard University, Cambridge, Massachusetts 02138, USA
4Department of Physics, University of Chicago, Chicago, Illinois 60637, USA
5Enrico Fermi Institute, University of Chicago, Chicago, Illinois 60637, USA
6Kavli Institute for Cosmological Physics, University of Chicago, Chicago, Illinois 60637, USA
(Dated: November 6, 2026)
We present the first search for fixed-polarization dark photon (DP) dark matter in the 44–52 µeV
mass range, corresponding to 10.Where the paper says it
Where the paper says it
Where the paper says it
Appendix Details
To generate background-only toy datasets (H0), the authors employ an Artifact-Isolation Monte Carlo method This approach models the correlations between neighboring frequency bins and the reduction in residual variance introduced by Savitzky–Golay baseline subtraction The final simulated H0 dataset is then assembled by the measured noise standard deviation.
Likelihood Calculation Simplification
The authors decouple the time-dependent modulations from the polarization parameters through an analytical harmonic factorization optimized for rapid evaluation This allows for the precomputation of polarization-independent time-domain harmonic sums Ccross and Scross. The modulation operators cos 2ϑk and cos 4ϑk depend on the sidereal rotation ω⊕tk and the dark-photon polarization angles (θX, ϕX). p. 2 • Twice-daily modulation (second harmonic of
sidereal frequency, 2ω⊕):
C2ω⊕ (t) = 1
2
cos2λ cos2
θX cos(2ω⊕t − 2ϕX). The relative strengths of these components depend on
the polarization latitude θX and the experimental latitude λ.Where the paper says it
Geometric Coefficients
The angular amplitudes b(m)n,c (θX, ϕX) and b(m)n,s (θX, ϕX) are uniquely determined by the detector’s geographical orientation The non-zero coefficients for the geometries analyzed in this work are given below.
Test Statistic Calibration
The authors generated three hundred independent noise toy ensembles to validate results and accurately characterize the background The resulting k0 and q0 histograms were averaged to form a smooth empirical null distribution f(k0H0) and f(q0H0) with improved statistical stability. A global 90% threshold of k global 0,th = −2. p. 6 The injected-signal point shown here is a leakage-induced excess in a relatively distant neighboring bin, with both k0 and q0
exceeding 100 near the signal center, beyond the plotted ranges. frequentist global 90% CL threshold is calibrated from the
Monte Carlo trials;
covery significance:
q0 = 2 ln
sup
P0≥0
L(P0)
L(0) (27)
Here sup
P0≥0
L(P0) is the maximum likelihood in a nonnegative power range. We use the background-only MC distribution to define
a global candidate threshold for this statistic. With 300
background-only MC toy datasets, we obtain a global
90% threshold of k
global
0,th = −2. Neff ≃ 5 × 105
,
which lies between the number of frequency bins (Nj =
230, 392)and the total number of scanned templates
(NmDP = 1,151,965), reflecting correlations between
nearby frequencies and sub-bin offsets. Given Neff, the relation between local and global probabilities is
pglobal = 1 − (1 − plocal)
Neff
,
The global 90% candidate threshold is obtained from
the local null-MC distribution by using Neff to convert a
10% global background-only exceedance probability to
the corresponding local probability (Fig.Where the paper says it
Where the paper says it
Where the paper says it
Power Limit Estimation
The Bayesian approach derives the 90% credible upper limit PBayes 90% by integrating the marginalized posterior probability distribution over physically allowed signal powers P0 ≥ 0 The frequentist approach uses a Monte Carlocalibrated Neyman construction to obtain a local frequentist upper limit. The 90% confidence-level upper limit PFreq 90% is determined by the Neyman inversion boundary qobs = qcut(Ptest), with test powers satisfying qobs < qcut(Ptest) excluded. p. 7 To prevent unphysical negative limits arising from downward background fluctuations, we floor the inverted upper limit at the background-only hypothesis, P0 = 0. Finally, the 90% confidence-level upper limit P
Freq
90% is
determined by the Neyman inversion boundary qobs =
qcut(Ptest), with test powers satisfying qobs < qcut(Ptest)
excluded.Where the paper says it
Comparison with Previous Experiments
Figure 6 shows the updated fixed-polarization limits from the combined GigaBREAD 2023 and 2024 datasets, together with previous constraints The orange region shows the fixed-polarization exclusion limit derived in this work, taking the larger of the Bayesian and frequentist upper limits at each mass. This result at χ ∼ 10−12 is slightly stronger than the previous limit assuming random DP polarization [5]. p. 8 For comparison, the
dark blue region denotes the previous rescaled fixedpolarization reinterpretation of the GigaBREAD exclusion limit, and the orange region shows fixed-polarization
exclusion limit derived in this work, taking the larger of
the Bayesian and frequentist upper limits at each mass. This result at χ ∼ 10−12 is slightly stronger than the previous limit assuming random DP polarization [5], benefiting from additional data collected with a different detector orientation.Where the paper says it
Future Extensions
Future extensions of this analysis will explore additional data sets and further refinements of the statistical framework The methodology developed here can be directly applied to other broadband dark-photon searches and similar experiments involving time-dependent or polarization-dependent signals.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Polarization-aware likelihood analysis for dark photon searches allows AI systems to detect signals that retain a fixed polarization direction, which is crucial for vector dark matter produced by misalignment-like mechanisms. This enables the system to set upper limits on kinetic mixing parameters with improved precision compared to analyses assuming randomly polarized dark matter.
-
The use of time-dependent modulation factors, such as cos2 theta(t) and cos4 theta(t), in likelihood calculations allows AI systems to account for sidereal modulation of detector response as the Earth rotates. This capability enables the system to distinguish between signals that are time-independent and those exhibiting specific periodic variations tied to Earth's rotation.
-
The application of Bayesian and frequentist statistical frameworks, including marginalization over nuisance parameters like polarization angles (theta X and phi X), allows AI systems to derive robust upper limits on the signal power P0 without assuming a fixed polarization angle beforehand. This provides more reliable constraints on the kinetic mixing parameter chi at a given dark photon mass mDP.
-
The harmonic factorization of likelihood terms allows AI systems to precompute temporal harmonics, which significantly reduces the computational bottleneck of summing over all time spectra during likelihood evaluation. This enables faster and more efficient processing of large datasets, such as those from GigaBREAD runs.
-
The development of a Monte Carlo method for generating background-only toy datasets (Artifact-Isolation Monte Carlo) allows AI systems to accurately characterize the noise and correlations introduced by instrumental effects like Savitzky–Golay baseline subtraction. This provides a stable and statistically sound null distribution for setting detection thresholds.
-
The analysis demonstrates that incorporating polarization modulation and likelihood-based approaches can improve the sensitivity of broadband experiments to polarized dark photons. This suggests that future AI-driven search strategies should prioritize analyzing time-dependent or polarization-dependent signals over simpler, time-averaged methods.
Abstract
We present the first search for fixed-polarization dark photon (DP) dark matter in the 44-52 μ eV mass range, corresponding to 10.7-12.5GHz. GigaBREAD is a room-temperature broadband dish antenna experiment in which DP convert to ordinary photons on a cylindrical metallic emission surface and are focused onto a radio-frequency receiver. Previous searches, including GigaBREAD and several other dish antenna searches, have assumed randomly polarized DP dark matter. However, vector dark matter produced by a misalignment-like mechanism may retain an approximately fixed polarization, leading to sidereal modulation of the detector response as the Earth rotates. We reanalyze the GigaBREAD data using a polarization-matched likelihood that accounts for the time-dependent projection of a fixed DP polarization onto the detector axis. We find no evidence for a fixed-polarization DP signal and set upper limits on the kinetic mixing parameter at the level of χ about10-12. These limits are about 20% better than a naive limit obtained by rescaling the previous GigaBREAD limit under the fixed-polarization assumption.
Related papers
- DrSai: an agentic AI system toward autonomous physics analysis workflows at BESIII
- CANTON- mu Proposal: A Next-Generation Muon g-2 Measurement at Sub-0.1 ppm Precision
- Search for dark photon dark matter near 1,meV using a superconducting mixer
- ANUBIS detector and its sensitivity to neutral long-lived particles
- How to scale your HEP ML models: A recipe for robust architecture comparisons at scale
- Search for the 16 O(ppp) to 13 C pi+ pi+ e+ Decay Mode in Super-Kamiokande Using Machine Learning Techniques