A Cosmic Radiation Modular Telescope on the Moon: The MoonRay Concept

Abstract

The MoonRay project is carrying out a concept study of a permanent lunar cosmic-ray (CR) and gamma-ray observatory, in view of the implementation of habitats on our satellite. The idea is to build a modular telescope that will be able to overcome the limitations, in available power and weight, of the present generation of CR instruments in Low Earth Orbit, while carrying out high-energy gamma-ray observations from a vantage point at the South Pole of the Moon. An array of fully independent modules (towers), with limited individual size and mass, can provide an acceptance more than one order of magnitude larger than instruments in flight at present. The modular telescope is designed to be deployed progressively, during a series of lunar missions, while collecting meaningful scientific data at the intermediate stages of its implementation. The operational power will be made available by the facilities maintaining the lunar habitats. With a geometric factor close to 15 m²sr and about 8 times larger sensitive area than FERMI-LAT, MoonRay will be able to carry out a very rich observational program over a time span of a few decades with an energy reach of 10 PeV allowing the exploration of the CR “knee” and the observation of the Southern Sky with gamma rays well into the TeV scale. Each tower (of approximate size 20 cm × 20 cm ×100 cm) is equipped with three instruments. A combined Charge and Time-of-Flight detector (CD-ToF) can identify individual cosmic elements, leveraging on an innovative two-layered array of pixelated Low-Gain Avalanche Diode (LGAD) sensors, with sub-ns time resolution. The latter can achieve an unprecedented rejection power against backscattered radiation from the calorimeter. It is followed by a tracker, providing also photon conversion, and by a thick crystal calorimeter (55 radiation lengths, 3 proton interaction lengths at normal incidence) with an energy resolution of 30–40% (1–2%) for protons (electrons) and a proton/electron rejection in excess of 10⁵. A time resolution close to 100 ps has been obtained, with prototypal arrays of 3 mm × 3 mm LGAD pixels, in a recent test campaign carried out at CERN with Pb beam fragments.

1. Introduction

A new generation of space instruments have carried out direct measurements of charged cosmic rays (CR) with unprecedented precision during the last two decades. Above a few hundred GV, unexpected deviations of the fluxes from a pure power-law spectrum have been observed by balloon and space-borne experiments, including ATIC [1], CREAM [2], PAMELA [3], AMS-02 [4], CALET [5], DAMPE [6], and NUCLEON [7]. The proton and helium spectra have been found to undergo a smooth and progressive increase of the spectral index, starting above a few hundred GV and continuing up to ∼10 TV. At even higher energies, this trend changes and both spectra show a “softening”, as observed by DAMPE [6] and CALET [8]. The presence of a “spectral bump” came as a surprise and triggered a number of theoretical speculations invoking different scenarios, as the dominance of one (or more) nearby supernova remnant(s) (SNR), the confinement and gradual release of CRs from the source, or the possibility of an anomalous diffusive regime near the acceleration sites.

2. High-Energy Cosmic Rays

The experimental challenge for the years to come is to reach even higher energies and explore the “knee” with direct measurements (approximately one decade around 3–4 PeV), where the spectral index increases from ∼2.7 below 10¹⁴ eV to ∼3.1 above 10¹⁶ eV. This ambitious goal calls for an experimental apparatus with a geometric acceptance of at least 3 m²sr, i.e., more than one order of magnitude larger than the space instruments in flight at present. This objective is being pursued by a large-acceptance payload HERD [9], designed to be installed aboard the Chinese Space Station. On a longer time scale, large missions are under study for a possible launch around the mid of the century. They include AMS100 [10] and ALADinO [11], two large instruments that have been proposed for long-term observations at the Lagrangian point L2. However, the maximum weight and power available on a single payload represent the main limitations to the deployment of a large area instrument with a sufficient collecting power to accumulate, in a reasonable mission timescale (10–15 years), the statistics needed to study the “knee”.

As an alternative to a single payload, one could consider a constellation of many satellites. While this approach might look appealing for the study of high-energy gamma rays, it would be completely inadequate for TeV charged particles. Their detection requires a calorimetric depth 5–6 times larger than FERMI-LAT [12] and similar to HERD with a seamless calorimetric volume minimizing energy leakage, while a cluster of independent calorimetric instruments could be prone to large systematic errors.

3. The MoonRay Concept

An alternative is a large “ground telescope” on the Moon for the study of high-energy gamma rays and charged CR. The key idea of MoonRay [13] is modularity, with the design of a telescope segmented into relatively lightweight modules (“towers”), each working as an independent functional unit. The sensitive area of the telescope would be extended during consecutive missions with the deployment of extra towers, while keeping the instrument operational in between two subsequent phases of its upgrade.

The MoonRay concept has been formulated assuming an hypothetical scenario where lunar habitats had become a reality. In this case, the lunar power infrastructures, in charge of life support for the astronauts, should be able to sustain a power budget by far much larger than the amount required for the telescope (a few kW).

The low lunar gravity (about 1/6 than on Earth), facilitates the installation of modules of reasonable size and weight. The study of a modular apparatus (Figure 1) operating on the Moon, and designed to carry out direct observations of charged cosmic radiation and high-energy gamma rays, is briefly described in the following.

4. Outline of the MoonRay Cosmic Radiation Telescope

The design of the MoonRay telescope relies on recent advances in high-energy particle instrumentation. Each tower includes three main instruments:

• A particle identifier (CD-ToF), integrating high resolution Time-of-Flight (ToF) capabilities into a charge detector (CD), for the simultaneous measurement of the electric charge and of the time of arrival of the incident particle. This is a newly developed instrument based on Low-Gain Avalanche Diode (LGAD) sensors with sub-ns time resolution. It allows a clean charge identification of the CR incident particle thanks to an efficient rejection of the copious backscattering background generated by its interaction with the calorimeter;
• A modular tracking system (MT) based on semiconductor position detectors (interspaced with thin photon absorbers) or, alternatively, on a scintillating fiber tracker with silicon photodetectors (SiPM) readout;
• A homogeneous calorimeter (MCAL) segmented in layers of reasonable size and weight, each consisting of a 3D array of scintillating crystals readout by photodiodes.

The “core” of the telescope consists of a bidimensional array of 10 × 10 towers (MoonRay-100), as baseline configuration, or up to 16 × 16 towers (MoonRay-256), as shown in Figure 1. Its effective acceptance (

Table 1. Effective acceptance for MoonRay-256 at 100 TeV.

Tiles	m2	GFeff (m2sr) Proton (55${\mathit{X}}_0$)	GFeff (m2sr) Electron (30${\mathit{X}}_0$)
Top-Down	3.2 × 3.2	11.3	12.1
Lateral	3.2 × 0.7	3.5	3.8
Total		14.8	15.9

) can reach 14.8 (15.9) m²sr for protons (electrons) at 100 TeV.

“Lateral walls” extend the coverage to the four lateral sides. This is achieved with horizontal modules that are very similar in design to the upper part of a tower (Figure 2). In fact they implement only the CD-ToF and Tracker sub-systems, but not the calorimeter, whose functionality is already provided, at all incident angles, by the “core” of the telescope. The horizontal modules are equipped with a longer tracker, obtained by increasing the number of silicon strip detector (SSD) layers. Hermetic coverage of the intermediate angles is provided by “short towers”, equipped with extra CD-ToF layers. With this geometry, charge measurement and backscattering rejection is extended to all sides of the telescope, increasing the coverage of gamma-ray detection to angles a few degrees above the horizon.

5. High-Energy Gamma Rays

The present experimental scenario calls for new observations of gamma rays, including the identification of FERMI unassociated sources, the emission from the Galactic Center, studies of Galactic winds and new gamma-ray pulsars, a detailed study of Vela SNR, investigation on the origin of the FERMI bubbles, identification of new sources in the Southern Hemisphere, improved modeling of galactic $\gamma$-ray diffuse background, search for narrow spectral lines as signatures of DM annihilation, study of transients (GRBs), and of the association of Gravitational Waves (GW) with their electromagnetic counterparts.

The spectrum of gamma rays spans almost 7 decades in energy and about 14 in flux. The latter rapidly decreases at higher energies, therefore the instrument effective area has to grow with increasing energy. As a first-order comparison with FERMI-LAT effective area of 0.9 m² (on axis), MoonRay-256 would represent an eightfold increase in the effective area with an on source duty cycle ∼ six times longer. With a geometric factor close to 15 m²sr (Table 1) and about eight times larger sensitive area than FERMI-LAT (

Table 2. Effective area for MoonRay-256.

Tiles	SSD Tiles	Areaeff (m2)
Top-Down	1024	3.3
Lateral	4 × 128	2.6
Corner	4 × 128	1.6
Total	2048	7.5

), MoonRay will be able to carry out a very rich observational program with the observation of the Southern Sky in gamma rays, from GeV to multi-TeV, from the lunar South Pole.

Assuming a candidate site, possibly located at Mount Mouton (84.6° S latitude, 130.0° E longitude), close to the Moon South Pole, the visibility above the horizon of important sources in the Southern Sky is summarized in Figure 3, showing the expected modulation due to the lunar day, and a duty cycle of 100% for the sources considered. A preliminary evaluation of the expected PSF of MoonRay as a function of energy indicates an expected angular resolution of ∼0.05° at 1 TeV, to be compared with 0.08° for FERMI and CALET, an expected 0.05° for HERD and CTAO South, and ∼0.2° for LHAASO at the same energy, the last two being ground experiments.

6. The CD-ToF Particle Identifier

Calorimetric instruments are affected by the presence of backscattered (BSC) radiation from the calorimeter. It includes charged particles, but also photons and neutrons, back-propagating and interacting with the upstream detector material where they generate secondary ionization. The identification of the incident cosmic ray is usually carried out by a precise measurement of its electric charge via a single (or multiple) dE/dx measurement(s) provided by segmented scintillators or by multi-strip/pixel silicon detectors placed at the top of the experimental apparatus. Backscattered radiation degrades the charge resolution whenever it generates background ionization falling onto the same element hit by the incident particle. It also makes the association of hits to tracks more difficult and can spoil the angular resolution. The problem gets worse as the energy of the primary increases, the amount of BSC albedo growing approximately as the square root of the energy.

A traditional approach to the design of charge identifiers is placing them at a reasonable stand-off distance from the calorimeter, optimizing their granularity and performing redundant charge measurements. However, a high granularity implies a large number of readout channels, which is detrimental to the available power budget. A different approach exploits the difference in the time of arrival between the incident particle and the backscattered radiation hitting the detector at a later time. For example, with a flight path L = 30 cm to the calorimeter, a relativistic particle takes about 1 ns to reach the calorimeter and about as much is taken by the BSC radiation to hit the charge detector. An efficient rejection of the BSC background can be achieved with a “5D sensor”, i.e., a pixelated instrument that can simultaneously measure the position (x, y, z), charge (q), and timing (t) of the incident particle. A smaller flight path (down to ∼10 cm) between the charge detector and the calorimeter might be chosen to increase the acceptance. In this case, the ToF should achieve a sub-ns time resolution. Avalanche detectors with an internal gain layer can achieve a better timing resolution than traditional silicon strip detectors due the large slew rate dV/dt ot the signal generated by the avalanche process. The granularity of the CD-ToF instrument results from a tradoff between a minimization of the number of events, with a BSC hit falling on the same pixel crossed by the incident CR particle, and the power budget. The proposed implementation for MoonRay is based on pixelated LGADs with a relatively modest internal gain < 10 and with 3 mm × 3 mm granularity. The CD-ToF design is based on two staggered layers of LGAD pixels, ensuring the availability of at least one measurement for particles crossing dead areas in one layer (or at the tower boundaries), while providing a redundant charge measurement over most of the instrumented area.

7. Development of LGAD Arrays

The expected charge resolution of the Cd-ToF was simulated with GEANT4. The energy deposits in the LGAD sensor, generated by incident ultra-relativistic ions of atomic number 1 ≤ Z ≤ 28, is convolved with the parameterized response of the LGAD sensor and with the simulated noise and linearity of the front-end electronics. At low Z values, the charge resolution is dominated by energy straggling, while the contribution from the electronics noise becomes significant (but sub-dominant) only at Z > 20.

Preliminary results on the expected performance of MoonRay, obtained with GEANT4 simulations, indicate that the charge identification capability provided by the CHD-ToF allows for a clean separation of individual elements from proton to Z=28 (see Figure 3 in [13]).

The timing resolution of LGAD sensors depends on several independent contributions including jitter, time-walk, Landau noise, signal distortion, and digitization error. It has been modeled by several authors (including [14,15]). The target timing resolution for MoonRay is 100–150 ps.

During a recent campaign at CERN (November 2024) with a beam of fragmented Pb ions, five layers of 3 mm × 3 mm LGAD pixels, with either 150 or 275 $\mathsf{\mu }$m thickness, (see Figure 4) were interfaced with large dynamic range front-end electronics based on the VA32HDR14 chip, digitized, and read out. Each layer hosted 16 LGAD devices with 2 × 2 pixels each, with several configurations of guard-ring geometry and inter-pixel distance. Preliminary results on the charge resolution obtained at the beam test are shown in Figure 5 where the charge resolution for helium is better than 0.1 charge units.

The LGAD timing performance was also measured at the beam test with a dedicated double layer of LGAD pixels interfaced to a digital front-end circuitry connected to a high bandwidth digital oscilloscope. The preliminary results presented in [16] are consistent with a time resolution close to 100 ps.

8. Conclusions

While, at the moment, it is difficult to predict whether cosmic radiation observatories on the Moon will receive attention and substantial funding in the near future, the assumption of a possible presence of human habitats on the Moon is not unrealistic on a longer time scale. Therefore, we believe that a concept study of MoonRay should be pursued.

A conceptual approach towards the design of a modular lunar telescope, based on relatively small and lightweight independent modules, was presented in this paper. An active detector development of the LGAD arrays for the CD-ToF sub-detector and of the associated front-end electronics is being pursued in Italy in the framework of the R&D activities of the Istituto Nazionale di Fisica Nucleare (INFN). Preliminary beam test results of a prototype array have been reported.