betaFeedback

WorksPhysics and Astronomy

An improved synthetic signal injection routine for the Haloscope At Yale Sensitive To Axion Cold dark matter (HAYSTAC)

Abstract

Microwave cavity haloscopes are among the most sensitive direct detection experiments searching for dark matter axions via their coupling to photons. When the power of the expected microwave signal due to axion-photon conversion is on the order of 10−2410^{-24}~W, having the ability to validate the detector response and analysis procedure by injecting realistic synthetic axion signals becomes helpful. Here we present a method based on frequency hopping spread spectrum for synthesizing axion signals in a microwave cavity haloscope experiment. It allows us to generate a narrow and asymmetric shape in frequency space that mimics an axion's spectral distribution, which is derived from a Maxwell-Boltzmann distribution. In addition, we show that the synthetic axion's power can be calibrated with reference to the system noise. Compared to the synthetic axion injection in HAYSTAC phase I, we demonstrated synthetic signal injection with a more realistic lineshape and calibrated power.

Introduction

The axion is a well-motivated solution to the strong charge–parity problem in quantum chromodynamics and is also a dark matter candidate[1, 2, 3, 4]. The most sensitive direct detection experiments by far are based on the microwave cavity haloscope technique[5]. In a haloscope experiment, the axions would convert into photons of equal energy inside a resonant microwave cavity permeated by a strong magnetic field. The energy of each axion-converted photon would be equal to the axion’s total energy—the sum of its rest-mass energy mam_a and kinetic energy. Therefore, the lineshape of the resulting microwave signal appearing in a haloscope experiment would inherit the axion’s kinetic energy distribution. Suppose νa=ma/(2π)\nu_a = m_a/(2\pi) is the axion mass in SI-frequency units, we denote the spectral distribution of an axion with mass mam_a as fνa(ν)f_{\nu_a}(\nu). The distribution fνa(ν)f_{\nu_a}(\nu) encodes axion’s properties derived from the pseudo-isothermal halo model[6], and accounts for the modulation due to the Earth’s rotation around the center of the galaxy[7, 8]. Here we describe a method to synthesize a microwave signal whose lineshape resembles that from an expected axion signal. Injecting and detecting such synthesized signals into the detector allows us to characterize the detector’s response, validate the analysis procedure[8, 9, 10], and perform a blind analysis[11].

Our method for synthesizing axion signals is inspired by a patent for radio frequency (rf) hopping by Markey and Antheil[12]; it is related to the frequency hopping spread spectrum (FHSS) technique, which has been applied in military and wireless communication to prevent interception and reduce interference. In our experiment, this technique allows us to generate an axion’s spectrum by hopping between a large number of random rf tones sampled from the axion’s spectral distribution. This way we can produce the spectral spread associated with the axion’s kinetic energy distribution. As will become more apparent later, the axion’s lineshape function is asymmetric and narrow. The frequency hopping method allows us to overcome the technical difficulties associated with signal shaping in the frequency domain. In principle, this technique can be used to synthesize any generic spectral shapes. It is also a realistic simulation: each axion’s energy is a random variable that follows fνa(ν)f_{\nu_a}(\nu) and the observed lineshape is an ensemble-averaged result.

In this note, we describe the methods involved in signal injection in § 2 and provide a summary and outlook in § 2.1.

Methods

The synthetic axion signal is injected as a transmission (tx) input through the weak port of the cavity and subsequently coupled out from the receiver’s port. Figure 1 shows the schematic of synthetic signal injection. The main device that synthesizes axion signals is an analog signal generator (model: Agilent E8257D), labeled as PSG2. In between PSG2 and the cavity’s weak port, an rf switch (model: ZFSWA2-63DR+) and a power splitter/combiner (model: ZX10-2-71-S+) are added in this order. The rf switch allows us to interrupt signal injection from PC1 via a USB DAC (model: USB-6009-NI). For instance, the injection signals are blocked during calibration measurements between axion datasets, as they would bias the noise measurement. The power splitter allows us to simultaneously send in a weak tone near the JPA resonance as a way to monitor and stabilize the JPA gain[9]. The variable temperature stage (VTS) allows us to carry out in situ noise calibration[8] as part of the power calibration for the axion signal injected.

Schematic of the hardware injection signal path

Figure 1. Schematic. New components added for hardware injection are shown as gray nodes; the rest already exists as part of the experiment. Blue arrows and black dashed arrows indicate analog and digital signals respectively. PC2 controls PSG2 to inject a number of rf tones whose frequencies are sampled from fνa(ν)f_{\nu_a}(\nu) to simulate axions with mass νa\nu_a. The synthetic axion signals are injected through the weak port of the cavity and then measured and amplified by the receiver chain consisting of two amplifiers (AMPs): a JPA followed by a High-electron-mobility transistor. Subsequently, the amplified signals are down-converted, mixed with the local oscillator (LO) signal provided by PSG1, and then filtered by the low-pass filter (LPF). The time-series signal collected by the Analog/Digital Converter (ADC) is then Fourier transformed into an intermediate-frequency (IF) spectrum. The VTS helps us characterize the system’s total noise by comparing the cavity’s spectrum to blackbody spectra at controlled temperatures. The signal from the VTS is only coupled in during a calibration measurement, indicated by the dotted blue arrows.

The Haloscope At Yale Sensitive To Axion Cold dark matter (HAYSTAC) has been using squeezed-state receivers since 2021[13, 14]: the squeezing JPA (SQZ) is also included in this diagram, as it affects the total noise, even though the injected signal does not go through it; the full scheme with additional details for the squeezed-state receiver chain can be found in Refs.[13, 14, 15]

To synthesize the microwave signal for injection, we use the “Step (digital) sweep” mode of PSG2 by providing it with a list of frequencies, amplitudes, and dwell times. In aggregate, the list of rf tones would reproduce the axion’s lineshape. In this experiment, by default, we fix the amplitude and dwell time and only change the frequency of each point. The frequency precision is set to 0.1 kHz or 8 significant digits for a GHz frequency, chosen based on the precision of data acquisition. Due to the device’s output-byte limit during transmission, the maximum length for such a list is about n8=35n_8 = 35. For sub-kHz frequency precision, we empirically found that

N≫2,000N \gg2{,}000 samples are needed to reveal the shape of the sampling distribution. Therefore, to iterate through a list with N≫n8N \gg n_8 rf tones, it is necessary to transmit ⌈N/n8⌉\lceil N/n_8 \rceil sublists. To begin, we establish the connection between PSG2 and PC2 via virtual instrument software architecture (VISA) and enable the rf output on PSG2. The steps for sending NN rf tones are as follows:

  1. Send n8n_8 frequencies, dwell time τd\tau_d, and amplitude PsynP_{\mathrm{syn}} to PSG2 from PC2 in one command.

  1. Wait for τw\tau_w before sending another command to PSG2. Here τw=n8τd+τr\tau_w = n_8\tau_d + \tau_r includes the time required for the sweep n8τdn_8\tau_d and an additional uniformly random delay between 2 and 5 seconds τr∼U(2,5)\tau_r \sim U(2,5) required for VISA communication.

  1. Check PSG2’s status to confirm that the sweep completed. If not, generate an error message reporting PSG2’s status (settling or sweeping) and wait 2 more seconds before checking the status again.

  1. Repeat steps 1-3 until all NN items are iterated over.

Based on the timing parameters, we can derive the average duty cycle of this signal injection routine as ηˉ=n8τd/τˉw\bar{\eta} = n_8\tau_d/\bar{\tau}_w, after averaging over ⌈N/n8⌉≫1\lceil N/n_8 \rceil\gg1 repetitions.

By default, the dwell time is τa=5\tau_a = 5 ms, resulting in η=4.8%±1%N8\eta= 4.8\% \pm\frac{1\%}{\sqrt{N_{8}}}. In step 2, τr\tau_r is randomized to avoid introducing any additional frequency patterns into the signal.

Rejection sampling

To generate a list of random frequencies for signal injection, the probability density function of the sampling distribution is the axion’s spectral function

fνa(ν)=2π(321r1νa⟨β2⟩)sinh⁡(3r2(ν−νa)νa⟨β2⟩)e−3(ν−νa)νa⟨β2⟩−3r2/2,(1)f_{\nu_a}(\nu) = \frac{2}{\sqrt{\pi}}\left(\sqrt{\frac{3}{2}}\frac{1}{r}\frac{1}{\nu_a\langle\beta^2\rangle}\right)\sinh\left(3r\sqrt{\frac{2(\nu-\nu_a)}{\nu_a\langle\beta^2\rangle}}\right)e^{-\frac{3(\nu-\nu_a)}{\nu_a\langle\beta^2\rangle}-3r^2/2}, \tag*{(1)}

where ⟨v2⟩≈270\sqrt{\langle v^2\rangle}\approx270 km/s is the virial velocity, ⟨β2⟩=⟨v2⟩/c2≈8×10−7\langle\beta^2\rangle=\langle v^2\rangle/c^2\approx8\times10^{-7} (cc is the speed of light), vs≈220v_s\approx220 km/s the orbital velocity of the solar system about the center of the galaxy, and r=vs/⟨v2⟩≈2/38r=v_s/\sqrt{\langle v^2\rangle}\approx\sqrt{2/3^8}. Given the function form of fνaf_{\nu_a}, it can be seen that applying inverse transform sampling to sample from it is not obviously trivial. Here we use rejection sampling[16] as an alternative. To apply it, we find another distribution whose probability density function y(ν)y(\nu) satisfies ξ⋅y(ν)>fνa(ν) ∀ν\xi\cdot y(\nu)>f_{\nu_a}(\nu)\ \forall\nu for some ξ>0\xi>0. In our case, the Cauchy distribution

y(ν)=1σπ11+((ν−μ)/σ)2(2)y(\nu)=\frac{1}{\sigma\pi}\frac{1}{1+((\nu-\mu)/\sigma)^2} \tag*{(2)}

with μ=νa+1\mu=\nu_a+1 kHz and σ=3\sigma=3 kHz, is sufficient because as ν→∞\nu\to\infty, fνa/y→0f_{\nu_a}/y\to0. Furthermore, the cumulative function of a Cauchy distribution is invertible; therefore it is straightforward to apply inverse transform sampling to sample from yy. Let us consider rejection sampling as an iterative process. In each iteration, we sample a random number from yy, noting it as ν′\nu', and then decide whether to reject or accept it based on another random number, uu, chosen from the uniform random distribution between 0 and 1. If u≤fνa(ν′)/(ξ⋅y(ν′))u\leq f_{\nu_a}(\nu')/(\xi\cdot y(\nu')), we accept ν′\nu'; otherwise reject. This iteration continues until we have NN samples. Figure 2 shows the outcome of this method when the sampling distribution is fνaf_{\nu_a} with νa=4.7216300\nu_a=4.7216300 GHz as an example. It can be seen that the overall acceptance rate depends on the ratio of ∫dνfνa\int d\nu f_{\nu_a} to ∫dν ξ⋅y\int d\nu\,\xi\cdot y, i.e. the ratio of areas under fνaf_{\nu_a} and ξ⋅y\xi\cdot y; with ξ=3\xi=3, ∫dνfνa/∫dν ξ⋅y≈0.3\int d\nu f_{\nu_a}/\int d\nu\,\xi\cdot y\approx0.3.

Applying rejection sampling to create a list of random frequencies

Figure 2. Applying rejection sampling to create a list of N=40,000N=40,000 random frequencies from an axion’s spectral distribution with νa=4.7216300\nu_a=4.7216300 GHz. The envelope distribution (dash-dotted gray line) for the sampling distribution fνaf_{\nu_a} (solid gray line) is obtained by scaling the Cauchy distribution by ξ=3\xi=3. The histogram (blue) shows the samples obtained using this method conform to the sampling distribution fνaf_{\nu_a}.

Validation of Spectral Shape

Validation of this procedure is performed with a ∼10\sim10 hour signal injection run to verify the signal has the correct spectral shape. The injection is done at relatively high power such that the signal is clearly visible above the noise even without the standard processing[8]. This decouples any deviations caused by the injection routine from shape changes introduced by the processing and filtering scheme used to analyze the data. The observed signal is fit to the expected shape given by Eq. (1) in a 30 kHz window around the injected frequency. The fit model includes a linear background component to approximately capture the spectral shape of the cavity in this range. Results from the fit in which only the amplitude of the expected signal per Eq. (1) and the background model parameters are floating are shown in Figure 3. This results in a χ2/ndf\chi^2/\mathrm{ndf} of 1.57 showing good agreement between the observed and injected signal shapes. The χ2/ndf\chi^2/\mathrm{ndf} can be improved minimally to 1.56 when allowing the width of the lineshape to float by also varying the virial velocity but remains within 1% of the true velocity used in the injection.

Image not extracted

Figure 3. (top) Observed power in a 30 kHz window near an injected signal from a high power injection run. To validate the signal shape, the spectrum is minimally processed with no filtering applied. A fit to the lineshape in Eq. (1) is shown as a solid line. This fit includes a linear (mν+b)(m\nu+b) background component to capture the cavity’s spectral variation over this range. The residuals between the fit and data are shown in the bottom panel and give a χ2/ndf=1.57\chi^2/\mathrm{ndf}=1.57 showing good agreement between the observed and injected spectral shapes.

Power calibration

Though the power of each rf tone injected can be programmed on PSG2, neither the weak port’s coupling efficiency nor the transmission line loss is known, so it is necessary to carry out a power calibration measurement to determine the time-averaged power of a synthetic axion signal. At fixed cavity frequency, suppose we inject axions, each with power PsynP_{\mathrm{syn}}, near the cavity resonance for duration τint\tau_{\mathrm{int}}. The observed signal-to-noise ratio (SNR) depends on the system noise power at the injected axion’s detuning with respect to the cavity frequency νc\nu_c, hνcNsys(∣νa−νc∣)h\nu_cN_{\mathrm{sys}}(|\nu_a-\nu_c|) (hh is the Planck constant), PsynP_{\mathrm{syn}}, τint\tau_{\mathrm{int}}, η\eta, and the measurement bandwidth which is chosen to match the axion linewidth Δνa∼νa/106\Delta\nu_a\sim\nu_a/10^6. This then allows us to estimate PsynP_{\mathrm{syn}} from a standard axion measurement

Psyn=SNR⋅(hνcNsys(∣νa−νc∣))⋅Δνaητint.(3)P_{\mathrm{syn}}=\mathrm{SNR}\cdot\left(h\nu_cN_{\mathrm{sys}}(|\nu_a-\nu_c|)\right)\cdot\sqrt{\frac{\Delta\nu_a}{\eta\tau_{\mathrm{int}}}}. \tag*{(3)}

As an example, Figure 4 shows some results from a 12-hour measurement with the cavity and JPA at fixed frequencies. This data is taken at a lower injected power to produce a more realistic signal which is closer to the detection threshold of the experiment From this dataset, SNR and NsysN_{\mathrm{sys}} are derived following the analysis routine outlined in Ref.[8]. Each raw spectrum is first normalized by dividing out the average baseline. This step allows us to identify bins contaminated by IF noise. After removing contaminated bins from the first normalized spectrum, the spectrum is normalized again by dividing out the Savitzky-Golay fit to itself and subtracting 1. Then the resulting spectrum are rescaled to account for the detuning-dependent sensitivity and summed to produce a combined spectrum. This spectrum is normalized to the expected standard deviation in each bin as shown in Figure 4a. To derive the SNR of the injected signal, we find the total power of the injected signal by convolving the spectrum in Figure 4a with the axion lineshape function fνaf_{\nu_a}. In this case, the resulting SNR is about 17.6. The scale factor used in producing the rescaled spectrum is the ratio between the noise power per analysis bin Δνb=100\Delta\nu_b=100 Hz and expected signal power, both of which are plotted in Figure 4b. At detuning ∣νa−νc∣∼390|\nu_a-\nu_c|\sim390 kHz, the power of an axion-converted microwave signal is expected to be PKSVZ=5.6(1)×10−25 WP_{\mathrm{KSVZ}} = 5.6(1) \times10^{-25}\ \mathrm{W}, assuming the axion-photon coupling strength gaγγg_{a\gamma\gamma} (appearing in the axion-photon Lagrangian) in the Kim-Shifman-Vainshtein-Zakharov (KSVZ) model[17, 18], gaγγKSVZ=(−3.70×10−7/MeV2)⋅mag_{a\gamma\gamma}^{\mathrm{KSVZ}} = (-3.70 \times10^{-7}/\mathrm{MeV}^{2}) \cdot m_a nominally[9]. This is calculated from various experimental parameters, including the magnetic field strength and cavity volume[13, 19]. The system noise in units of photon numbers Nsys(∣νa−νc∣)=0.26(2)N_{\mathrm{sys}}(\lvert\nu_a-\nu_c\rvert)=0.26(2) is derived from in situ calibration measurements[13, 14, 15]. Evaluating Eq. (3) with these values, we find Psyn=2.2(2)×10−23 WP_{\mathrm{syn}} = 2.2(2) \times10^{-23}\ \mathrm{W}, and that the injected axions have a coupling strength at 6.2(5) gaγγKSVZ6.2(5)\,g_{a\gamma\gamma}^{\mathrm{KSVZ}} level.

Image not extracted

Figure 4. (a) Power excess (points) in the combined spectrum normalized to the expected standard deviation (error bars) and overlaid with fνaf_{\nu_a} (dashed line) in a 50 kHz window around the synthetic axion signal. (b) Expected signal power PKSVZP_{\mathrm{KSVZ}} (black) and system noise hνcNsysh\nu_cN_{\mathrm{sys}} integrated over an analysis binwidth Δνb\Delta\nu_b (red) at fixed cavity frequency νc\nu_c as functions of detuning ∣ν−νc∣|\nu-\nu_c|. The green shaded region indicates the frequencies within Δνa\Delta\nu_a of the injected signal. The cavity presents a noise source to the receiver that is frequency dependent with ∼1.1\sim1.1 MHz bandwidth and varies with ∣ν−νc∣|\nu-\nu_c|. The response to an axion signal has a different frequency dependence that is more sharply peaked at the cavity resonance. The plot shows the detuning-dependent variation of the axion signal and system noise.

Summary

We demonstrated synthetic axion injection using the rf hopping method[20]. The injected axion has the spectral distribution of a cold dark matter axion and its power is calibrated using the total noise power as a reference. This is a more realistic synthetic signal, in terms of spectral shape and power, as compared to the prior demonstration in HAYSTAC Phase I (cf. Appendix F in Ref.[8]). Furthermore, as HAYSTAC and other axion haloscope experiments, including CAPP and ADMX, enter the data production phase, mitigating bias will become increasingly important[21]. For instance, salting and blinding have become an integral part of WIMP (weakly interacting massive particle) dark matter searches[22, 23]. The method we developed can serve as the basis for implementing blinding and salting for axion dark matter searches.

Acknowledgments

We acknowledge the full HAYSTAC collaboration for the apparatus we used in this experiment, including the microwave cavity and JPAs. Y.Z. thanks Joseph Howlett for helpful discussion and references on mitigating bias in dark matter searches. HAYSTAC is supported by the National Science Foundation under grant numbers PHY-1701396, PHY-1607223, PHY-1734006 and PHY-1914199 and the Heising-Simons Foundation under grants 2014-0904 and 2016-044. Y. Z., S. G., E. G., S. C., M. J., and R. M. are supported in part by the Department of Energy under Grant No. DE-AC02-07CH11359. S. G. is supported in part by the National Science Foundation under Grant No. DMR-1747426.

AUTHOR DECLARATIONS

Conflict of Interest

The authors have no conflicts to disclose.

Author Contributions

Yuqi Zhu: Investigation (equal); Data Curation (equal); Formal analysis (equal); Writing–original draft (lead). M. J. Jewell: Investigation (equal); Data Curation (equal); Formal analysis (equal); Writing—Review & Editing (equal). Claire Laffan: Data Curation (equal); Formal analysis (equal). Xiran Bai: Writing—Review & Editing (equal). Sumita Ghosh: Writing—Review & Editing (equal). Eleanor Graham: Writing—Review & Editing (equal). S. B. Cahn: Writing—Review & Editing (equal). Reina H. Maruyama: Project Administration (equal); Funding acquisition (equal); Writing—Review & Editing (equal). S. K. Lamoreaux: Project Administration (equal); Funding acquisition (equal); Conceptualization (lead); Writing—Review & Editing (equal).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. [1]¹R. D. Peccei and H. R. Quinn, “CP conservation in the presence of pseudoparticles,” Phys. Rev. Lett. 38, 1440–1443 (1977).
  2. [2]²R. D. Peccei and H. R. Quinn, “Constraints imposed by CP conservation in the presence of pseudoparticles,” Phys. Rev. D 16, 1791–1797 (1977).
  3. [3]³S. Weinberg, “A new light boson?” Phys. Rev. Lett. 40, 223–226 (1978).DOI
  4. [4]⁴F. Wilczek, “Problem of strong P and T invariance in the presence of instantons,” Phys. Rev. Lett. 40, 279–282 (1978).DOI
  5. [5]⁵P. Sikivie, “Experimental tests of the “invisible” axion,” Phys. Rev. Lett. 51, 1415–1417 (1983).
  6. [6]⁶R. Jimenez, L. Verde, and S. P. Oh, “Dark halo properties from rotation curves,” Mon. Notices Royal Astron. Soc. 339, 243–259 (2003).DOI
  7. [7]⁷M. S. Turner, “Periodic signatures for the detection of cosmic axions,” Phys. Rev. D 42, 3572–3575 (1990).pubmed.ncbi.nlm.nih.gov/10012760
  8. [8]⁸B. M. Brubaker, L. Zhong, S. K. Lamoreaux, K. W. Lehnert, and K. A. van Bibber, “HAYSTAC axion search analysis procedure,” Phys. Rev. D 96, 123008 (2017).
  9. [9]⁹R. Brubaker, L. Zhong, Y. Gurevich, S. Cahn, S. Lamoreaux, M. Simanovskaia, J. Root, S. Lewis, S. Al Kenany, K. Backes, et al., “First results from a microwave cavity axion search at 24 μeV,” Phys. Rev. Lett. 118, 061302 (2017).DOI
  10. [10]¹⁰S. Al Kenany, M. Anil, K. Backes, B. Brubaker, S. Cahn, G. Carosi, Y. Gurevich, W. Kindel, S. Lamoreaux, K. W. Lehnert, et al., “Design and operational experience of a microwave cavity axion detector for the 20–100 μeV range,” Nucl. Instrum. Meth. Phys. Res. A: Accel. Spectrom. Detec. Assoc. Equip. 854, 11–24 (2017).
  11. [11]¹¹C. Bartram, T. Braine, E. Burns, R. Cervantes, N. Crisosto, N. Du, H. Karandza, G. Leum, P. Mohapatra, T. Nitta, et al. (ADMX Collaboration), “Search for ”invisible” axion dark matter in the 3.3-4.2 μeV mass range,” Phys. Rev. Lett. 127, 261803 (2021).DOI
  12. [12]¹²H. K. Markey and G. Antheil, “Secret communication system,” (1941).
  13. [13]¹³K. M. Backes, D. A. Palken, S. A. Kenany, B. M. Brubaker, S. B. Cahn, A. Droste, G. C. Hilton, S. Ghosh, H. Jackson, S. K. Lamoreaux, et al., “A quantum enhanced search for dark matter axions,” Nature 590, 238–242 (2021).
  14. [14]M. J. Jewell, A. F. Leder, K. M. Backes, X. Bai, K. van Bimber, B. M. Brubaker, S. B. Cahn, A. Droster, M. H. Esmat, et al., “New results from HAYSTAC’s Phase II operation with a squeezed state receiver,” (2023).
  15. [15]K. M. Backes, A Quantum-Enhanced Search for Dark Matter Axions, Ph.D. thesis (2021).
  16. [16]B. Flury, “Acceptance–rejection sampling made easy,” SIAM Review 32, 474–476 (1990).DOI
  17. [17]M. Dine, W. Fischler, and M. Srednicki, “A simple solution to the strong CP problem with a harmless axion,” Physics Letters B 104, 199–202 (1981).
  18. [18]J. E. Kim, “Weak-interaction singlet and strong CP invariance,” Phys. Rev. Lett. 43, 103–107 (1979).DOI
  19. [19]P. Sikivie, “Detection rates for “invisible”-axion searches,” Phys. Rev. D 32, 2988–2991 (1985).pubmed.ncbi.nlm.nih.gov/9956084
  20. [20]The code for implementing synthetic axion injection is made available as an open-source project at https://github.com/yuqizhyuzy/syntheticaxioninjection.git
  21. [21]D. Baxter, I. M. Bloch, E. Bodnia, X. Chen, J. Conrad, P. D. Gangi, J. E. Y. Dobson, D. Durnford, S. J. Haselschwardt, A. Kachoth, et al., “Recommended conventions for reporting results from direct dark matter searches,” Eur. Phys. J. C 81 (2021), 10.1140/epjc/s10052-021-09655-y.DOI
  22. [22]D. S. Akerib, S. A. Alsum, H. M. Araújo, X. Bai, A. J. Bailey, J. Balajthy, P. Beltrame, E. P. Bernard, A. Bernstein, Besiadszinski, et al. (LUX Collaboration), “Results from supernovae for dark matter in the complete lux exposure,” Phys. Rev. Lett. 118, 021303 (2017).DOI
  23. [23]E. Aprile, J. Aalbers, F. Agostini, M. Alfonsi, L. Althueser, F. D. Amaro, M. Anthony, F. Arneodo, L. Baudis, B. Bauermeister, et al. (XENON Collaboration 7), “Dark matter search results from a one-ton-year exposure of xenon1t,” Phys. Rev. Lett. 121, 111302 (2018).DOI

Paper details

vlnc:W_ud4ycdzc
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents