betaFeedback

WorksPhysics and Astronomy

n−n′n-n' Oscillations: Sensitivity of a first UCN beam experiment

Abstract

Oscillations of the neutron into a hidden sector particle are processes predicted in various Standard Model extensions. This extra channel for neutron disappearance has not been tested experimentally in large portions of the oscillation parameter space. Several efforts have been recently made on revising the oscillation time limits at low mass-splitting in ultra-cold neutron (UCN) storage experiments, and at larger mass-splitting in passing-through-wall experiments. In this work, we present the expected sensitivity of an experiment searching for neutron hidden neutron oscillations at intermediate mass-splitting via the application of magnetic fields in the range B0=30−1100B_0=30-1100 μμT. This experiment was performed at the Institut-Laue-Langevin using a novel UCN counter to monitor the beam flux. The measured UCN rate and the data collection technique predict a sensitivity on the oscillation time at the level of a couple of seconds.

Introduction

Models predicting the existence of hidden matter have gained special interest since they address several yet unsolved problems in particle physics and cosmology simultaneously [1]. One of these models presents hidden matter as an exact copy of all known particles and their interactions, named mirror matter. Particles in this extra sector, which were originally proposed as a solution to the P-violation observed in the weak interaction [2], could mix with ordinary particles through non-standard model interactions [3]. Such processes could lead to new channels of CP and baryon symmetries violation, and therefore shed light on the baryogenesis problem [4]. In addition, it has been argued that both sectors also interact through gravity, allowing to present mirror matter as a natural candidate for dark matter.

The mixing of the neutron (nn) with its hidden counterpart (n′n') is described1 by the Hamiltonian [5]

H^=(mn+ΔEϵnn′ϵnn′mn+δm),(1)\hat{H} = \begin{pmatrix} m_n + \Delta E & \epsilon_{nn'} \\ \epsilon_{nn'} & m_n + \delta m \end{pmatrix}, \tag*{(1)}

with ΔE\Delta E the neutron energy due to its interaction with environment, δm=mn′−mn\delta m = m_{n'} - m_n the mass-splitting between the neutron and the hidden neutron masses, and ϵnn′=1/τnn′\epsilon_{nn'} = 1/\tau_{nn'} the mixing parameter accounting for the briefness of the oscillations. The general purpose of n−n′n-n' probing experiments is to trigger the n−n′n-n' oscillations by canceling the energy degeneracy, i.e. by matching ΔE\Delta E and δm\delta m. This has been done via small magnetic fields (ΔE=μnB0∼10−3 neV\Delta E = \mu_n B_0 \sim10^{-3}\,\mathrm{neV}) in UCN storage experiments [6], and through material Fermi potentials (ΔE=VF∼102\Delta E = V_F \sim10^{2} neV) in regeneration experiments [7]. Apart from the 5σ5\sigma anomaly reported by Berezhiani [8] in UCN storage experiments, no confirmation of n−n′n-n' oscillations has been reported so far. The present work targets scanning n−n′n-n' oscillations on the yet unexplored energy range ΔE=10−3−10−1\Delta E = 10^{-3}-10^{-1} neV via magnetic fields in the range B0=30−1100B_0 = 30-1100 μ\muT. Moreover, inspection of this energy range permits the reexamination of the intervals in the mass-splitting favored by the 5σ5\sigma anomaly.

Experimental setup

At the EDM port of the PF2 instrument at the ILL we studied n−n′n-n' oscillations by direct measurement of the UCN flux. Unlike past UCN experiments, the setup did not include a storage chamber. Instead, we searched for oscillations during the UCN passage across a 6-m-long guide connecting the EDM beam port to an UCN counter. In order to scan the δm\delta m parameter in the targeted range, a static magnetic field generated by a 5-m-long solenoid is applied along the main guide (Figure 1). If the degeneracy cancels for a given magnetic field (∣δm∣=μnB0|\delta m| = \mu_n B_0), a signal of n−n′n-n' oscillations would be observed as a drop of the detected beam flux. Two correction coils were added at the solenoid edges to extend the magnetic-field coverage and increase its uniformity. In addition, a 4.5-m-long cylindrical mu-metal shield was placed around the solenoid to improve the magnetic-field uniformity.

Figure 1. UCN beam setup for testing n−n′n-n' oscillations at the PF2/EDM beam port at ILL reactor.

UCN counting and background analysis

For the measurement of the UCN beam flux we used a novel gaseous UCN detector (GADGET). This fast counter, specially developed by us for the n2EDM project [9], was essential at the ILL EDM beam port where UCN fluxes can reach up to 10610^{6} s−1^{-1}. The fast performance of GADGET is based on its detection principle. After crossing a 30-μ\mum-thick entrance window, UCNs enter the detector sensitive volume filled with a gas-mixture of 3He{}^{3}\mathrm{He} and CF4\mathrm{CF}_4 at 15 and 500 mbar, respectively. Following the UCN absorption by the 3He{}^{3}\mathrm{He} nuclei, the emitted reaction products, 1H{}^{1}\mathrm{H} and 3H{}^{3}\mathrm{H}, induce scintillation on the CF4\mathrm{CF}_4 molecules whose decay times are of 6 ns [10]. A set of three photo-multiplier tubes (PMT) optically coupled to the gas chamber convert the scintillated light into voltage pulses lasting up to 30 ns. These signals were read and filtered with a triple coincidence window by the FASTER digital acquisition system, capable of sampling at 500 MHz with 12 bits.

Background discrimination was based on pulse shape analysis (PSA) constructed with the total integrated charge (QtotQ_{\mathrm{tot}}) and amplitude (AtotA_{\mathrm{tot}}) over the three PMT signals. Figure 2 shows the PSA maps measured for two detector configurations during UCN delivery: 200 s of ordinary detection (left side) and 3 hours recording with only CF4\mathrm{CF}_4 gas filling at 500 mbar, i.e. no 3He{}^{3}\mathrm{He} (right side). This last configuration allows the identification of three light-emitting processes different from the main UCN detection. First, events produced from γ\gamma and β\beta environmental backgrounds interacting with the CF4_4 molecules are mostly contained in category ‘γ+β\gamma+ \beta’. Second, events created from the γ\gamma interaction with electrons of the chamber’s quartz windows are enclosed by the ‘Cherenkov’ category. Finally, ambiguous events also including UCN capture on nuclei different from 3^{3}He build up a region mainly contained by category ‘UCN2’. This partitioning allows defining the ‘UCN1’ category essentially free of background. Taking into account the counting rates of both measured configurations, background contamination in ‘UCN1’ during ordinary detection is estimated to be 0.04%, which is 5 times smaller than the statistical fluctuations.

Figure 2. Pulse shape analysis from GADGET counter measuring in normal conditions (left) and with no 3^{3}He gas absorber (right). The total counting rates are 280 kHz and 1 kHz, respectively.

Data taking and expected sensitivity

We formulate a self-normalized measurement of the UCN by counting the number of neutrons (NN) at three magnetic fields AA, BB and CC. Performing a double counting at BB, the normalized ratio sensitive to n−n′n - n' oscillations was defined as

RABC=NB+NB′NA+NC{=1,if no oscillations<1,if oscillations at field B>1,if oscillations at field A or C.(2)R_{ABC} = \frac{N_B + N'_B}{N_A + N_C} \begin{cases} = 1, & \text{if no oscillations}\\ < 1, & \text{if oscillations at field } B\\ > 1, & \text{if oscillations at field } A \text{ or } C. \end{cases} \tag*{(2)}

Single measurements of RABCR_{ABC} were obtained from the counting of UCN passing through the experimental setup in cycles lasting 200 s. Few seconds of every cycle were discarded to ramp the magnetic field and the rest was divided into 4 batches of 44 s where the magnetic field values followed the sequence

{A,B,B,C}={B0−20 μT,B0,B0,B0+20 μT}.(3)\{A, B, B, C\} = \{B_0 - 20\,\mu\mathrm{T}, B_0, B_0, B_0 + 20\,\mu\mathrm{T}\}. \tag*{(3)}

The 20 μT\mu\mathrm{T} magnetic field step inside a cycle was chosen larger than the resonance width (∼2.4 μT\sim2.4\,\mu\mathrm{T}) so that oscillations would occur at only one of the field configurations. Correspondingly, the magnetic field shifting step (ΔB0\Delta B_0) in between cycles was set in ΔB0=3 μT\Delta B_0 = 3\,\mu\mathrm{T} to guarantee that all the intermediate values were covered by the scanning process.

According to Eq. (2), a n−n′n - n' oscillation signal is identified if ∣RABC−1∣>ζΔRABC\lvert R_{ABC} - 1\rvert> \zeta\Delta R_{ABC}, with ζ\zeta a coefficient depending on the confidence level (CL), and ΔRABC\Delta R_{ABC} the uncertainty of RABCR_{ABC}. In contrast, if all measurements of ∣RABC−1∣\lvert R_{ABC} - 1\rvert are contained within their uncertainties, the model parameters τnn′\tau_{n n'} and δm\delta m can be bounded. Figure 3 shows the exclusion regions expected from the ‘no-signal’ scenario assuming RABC=1R_{ABC} = 1 and an UCN flux ΦUCN∼280\Phi_{\mathrm{UCN}} \sim280 kHz, with uncertainties purely derived from statistical fluctuations, i.e. ΔRABC∼4/(ΦUCN⋅44 s)∼2.7×10−4\Delta R_{ABC} \sim\sqrt{4/(\Phi_{\mathrm{UCN}} \cdot44\,\mathrm{s})} \sim2.7 \times10^{-4}. Every cycle measurement determines an exclusion region characterized by three maxima located at ∣δm∣/μn=A,B\lvert\delta m\rvert/\mu_n = A, B and CC. The sensitivity at BB is two times larger than the one at AA or CC due to the larger counting time at the former (2×442 \times44 s). After scanning the triplet ABCABC every 3 μT3\,\mu\mathrm{T} and overlapping the resulting exclusion regions, the upper envelope shows that τnn′>4\tau_{nn'} > 4 s for any ∣δm∣/μn|\delta m|/\mu_n contained in the targeted interval [30−1100] μT[30 - 1100]\,\mu\mathrm{T}.

Figure 3. Expected sensitivity of the UCN beam experiment. The upper envelope is built with the 95% C.L. exclusion regions between 125 and 215 μT\mu\mathrm{T}. Individual contributions of three cycles are displayed with colored regions to better illustrate the upper envelope shape. The same pattern extends over the full ∣δm∣/μn|\delta m|/\mu_n range.

Outlook

Given that no measurement has constrained δm\delta m in the targeted interval at the scale of few seconds, this sensitivity analysis shows that the chosen scanning technique can result on a new experimental limit of τnn′\tau_{nn'}. A later publication will be dedicated to the actual exclusion obtained from the measured RABCR_{ABC} points while considering extra model refinements such as the field uniformities, UCN velocity distributions, and fluctuations of the neutron flux.

Acknowledgments

We would like to acknowledge the technical support by T. Brenner during the experiment.

References

  1. [1]Foot, R. (2014). Int. J. Mod. Phys. A, 29(11n12), 1430013.
  2. [2]Lee, T. D., Yang, C. N. (1956). Phy. Rev., 104(1), 254–258.
  3. [3]Sarrazin, M., Petit, F. (2012). Eur. Phys. J. C 72, 2230.
  4. [4]Berezhiani, Z. (2016). Eur. Phys. J. C 76, 705.
  5. [5]Hostert, M., McKeen, D., Pospelov, M., Raj N. (2022) arXiv:2201.02603 [hep-ph].DOI
  6. [6]Abel, C., et al. (2021). Phy. Let. B, 812, 135993.
  7. [7]Almazán, H., et al. (2022). Phys. Rev. Lett. 128, 061801.
  8. [8]Berezhiani, Z., Nesti, F. (2012). Eur. Phys. J. C 72, 1974.
  9. [9]Ayres, N. J., et al. (2021). Eur. Phys. J. C 81, 512.
  10. [10]Lehaut, G., et al. (2015). Nucl. Instrum. Methods. Phys. Res. A, 797, 57–63.

Paper details

vlnc:W_wt6jf268
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