Gamma-ray Astrophysics in the MeV Range: the ASTROGAM Concept and Beyond
Alessandro De Angelis, Vincent Tatischeff, Andrea Argan, Soren Brandt, Andrea Bulgarelli, Andrei Bykov, Elisa Costantini, Rui Curado da Silva, Isabelle A. Grenier, Lorraine Hanlon, Dieter Hartmann, Margarida Hernanz, Gottfried Kanbach, Irfan Kuvvetli, Philippe Laurent, Mario N. Mazziotta, Julie McEnery, Aldo Morselli, Kazuhiro Nakazawa, Uwe Oberlack, Mark Pearce, Javier Rico, Marco Tavani, Peter von Ballmoos, Roland Walter, Xin Wu, Silvia Zane, Andrzej Zdziarski, Andreas Zoglauer
NNoname manuscript No. (will be inserted by the editor)
Gamma-ray Astrophysics in the MeV Range
The ASTROGAM Concept and Beyond
Alessandro De Angelis · VincentTatischeff · Andrea Argan · SørenBrandt · Andrea Bulgarelli · AndreiBykov · Elisa Costantini · Rui Curadoda Silva · Isabelle A. Grenier · LorraineHanlon · Dieter Hartmann · MargaridaHernanz · Gottfried Kanbach · IrfanKuvvetli · Philippe Laurent · MarioN. Mazziotta · Julie McEnery · AldoMorselli · Kazuhiro Nakazawa · UweOberlack · Mark Pearce · Javier Rico · Marco Tavani · Peter von Ballmoos · Roland Walter · Xin Wu · Silvia Zane · Andrzej Zdziarski · Andreas Zoglauer
Received: date / Accepted: date
Abstract
The energy range between about 100 keV and 1 GeV is of interestfor a vast class of astrophysical topics. In particular, (1) it is the missingingredient for understanding extreme processes in the multi-messenger era;(2) it allows localizing cosmic-ray interactions with background material andradiation in the Universe, and spotting the reprocessing of these particles; (3)last but not least, gamma-ray emission lines trace the formation of elements inthe Galaxy and beyond. In addition, studying the still largely unexplored MeVdomain of astronomy would provide for a rich observatory science, includingthe study of compact objects, solar- and Earth-science, as well as fundamentalphysics. The technological development of silicon microstrip detectors makesit possible now to detect MeV photons in space with high efficiency and lowbackground. During the last decade, a concept of detector (“ASTROGAM”)has been proposed to fulfil these goals, based on a silicon hodoscope, a 3Dposition-sensitive calorimeter, and an anticoincidence detector. In this paper
A. De AngelisDipartimento di Fisica e Astronomia “Galileo Galilei”via Marzolo 8, Padova I-35131, ItalyTel.: +39-049-827-5942E-mail: [email protected]. TatischeffUniversit´e Paris-Saclay, CNRS/IN2P3, IJCLab,F-91405, Orsay, France a r X i v : . [ a s t r o - ph . I M ] F e b Alessandro De Angelis et al. we stress the importance of a medium size (M-class) space mission, dubbed“ASTROMEV”, to fulfil these objectives.
Keywords
Gamma-Ray Astronomy · Multi-Messenger Astronomy · High-Energy Astrophysics
PACS · · Gamma-ray astronomy has experienced a period of impressive scientific ad-vances and successes during the last decade. In the high-energy range studiedwith space instruments, above 100 MeV, the
AGILE and
Fermi missions ledto important discoveries. In particular, the Large Area Telescope (LAT) of the
Fermi satellite has established an inventory of over 5000 sources of variouskinds (blazars, pulsars, supernova remnants, high-mass binaries, gamma-raybursts (GRBs), etc.) showing a variety of gamma-ray emission processes [120].Similarly, in the hard X-ray/low-energy gamma-ray band, the latest catalogof sources detected with the Burst Alert Telescope (BAT) of the
Neil GehrelsSwift observatory contains 1632 sources in the range 14–195 keV [114]. But atintermediate photon energies, between 0.2 MeV and 100 MeV, only a few tensof steady sources have been detected so far, mostly by the COMPTEL instru-ment on board the
Compton Gamma-Ray Observatory (CGRO; see Ref.[107]),such that this particular field of astronomy has remained largely unexplored.Many of the most spectacular objects in the Universe have their peakemissivity at photon energies between 0.2 MeV and 100 MeV (e.g. gamma-raybursts, blazars, pulsars, etc.), so it is in this energy band that essential physicalproperties of these objects can be studied most directly. This energy range isalso known to feature spectral characteristics associated with gamma-ray emis-sion from pion decay, thus indicating hadronic acceleration. This fact makesthe MeV energy region of paramount importance for the study of radiating,nonthermal particles and for distinguishing leptonic from hadronic processes.Moreover, this energy domain covers the crucial range of nuclear gamma-raylines produced by radioactive decay, nuclear collision, positron annihilation,or neutron capture, which makes it as special for high-energy astronomy asoptical spectroscopy is for phenomena related to atomic physics.The fact that the MeV domain lags far behind compared to its neighbors(X-rays and gamma rays of high or very high energy) in terms of detectionsensitivity (Fig. 1) is due to instrumental difficulties specific to this domain.In particular, below a few MeV, the lack of signature of the creation of anelectron-positron pair is a limitation to the capability to separate gamma raysfrom charged particles and to evaluate the incoming direction of the photons.Moreover, while this is the domain of nuclear gamma-ray lines, which makesit extremely interesting for astrophysics, it also gives a strong instrumentalbackground due to the deactivation of irradiated materials in space. amma-ray Astrophysics in the MeV Range 3 -14 -13 -12 -11 -10 -9 -2 -1 ASTROMEV
SPIIBIS-ISGRIIBIS-PICsITJEM-X COMPTELEGRET CTA SouthFermi-LAT LHAASO HiSCOREMAGIC HESS/VERITASHAWC
Energy (MeV) S e n s i t i v i t y ( er g c m - s - ) Fig. 1: Point source continuum differential sensitivity of different X- and γ -ray instruments (see [37]). The hatched area indicates the targeted level ofsensitivity of the next generation gamma-ray observatory for a source effectiveexposure of 1 year.However, recent progress in silicon detectors and readout microelectron-ics can allow the development of a new space instrument reaching a gain insensitivity of about two orders of magnitude compared to CGRO/COMPTEL[36]. In addition, the instrument can achieve excellent spectral and spatialresolution by measuring the energy and three-dimensional (3D) position ofeach interaction with the detectors. Such a mission, dubbed “ASTROMEV”in this White Paper, has the potential to answer key questions in astrophysicsthrough a dedicated core science program: – Processes at the heart of the extreme Universe in the era ofmulti-messenger astronomy
Observations of relativistic jet and outflow sources (both in our Galaxyand in active galactic nuclei, AGNs) in the X-ray and GeV–TeV energyranges have shown that the MeV–GeV band holds the key to understandingthe transition from the low-energy continuum to a spectral range shapedby very poorly understood particle acceleration processes. ASTROMEVwill: (1) determine the composition (hadronic or leptonic) of the outflowsand jets, which strongly influences the environment – breakthrough po-larimetric capability and spectroscopy providing the keys to unlocking thislong-standing question; (2) identify the physical acceleration processes inthese outflows and jets (e.g. diffusive shocks, magnetic field reconnection,plasma effects), that may lead to dramatically different particle energydistributions; (3) clarify the role of the magnetic field in powering ultrarel-ativistic jets in gamma-ray bursts, through time-resolved polarimetry andspectroscopy.In addition, measurements in the ASTROMEV energy band between 100keV and 1 GeV will have a big impact on multi-messenger astronomy inthe 2030s. In particular, MeV energies are expected to be the characteristiccutoffs in Neutron Star - Neutron Star (NS-NS) and Black Hole - Neutron
Alessandro De Angelis et al.
Star (BH-NS) mergers, giving a decisive input to the study of the energeticsof these processes. Moreover, a detector sensitive in the MeV region willallow the detection of the π peak, disentangling hadronic accelerationmechanisms from leptonic mechanisms, and thus providing an independentinput to neutrino astronomy. – The origin and impact of high-energy cosmic-ray particles onGalaxy evolution
ASTROMEV will resolve the outstanding issue of the origin and propa-gation of low-energy cosmic rays affecting star formation. It will measurecosmic-ray diffusion in interstellar clouds and their impact on gas dynam-ics and state; it will provide crucial diagnostics about the wind outflowsand their feedback on the Galactic environment (e.g., Fermi bubbles [113],Cygnus cocoon [6]). ASTROMEV will have optimal sensitivity and energyresolution to detect line emissions from 511 keV up to 10 MeV, and a va-riety of issues will be resolved, in particular: (1) origin of the gamma-rayand positron excesses toward the Galactic inner regions; (2) determina-tion of the astrophysical sources of the local positron population from avery sensitive observation of pulsars and supernova remnants (SNRs). Asa consequence ASTROMEV will be able to discriminate the backgroundsto dark matter (DM) signals. – Nucleosynthesis and the chemical enrichment of our Galaxy
The ASTROMEV line sensitivity is more than an order of magnitude bet-ter than previous instruments. The deep exposure of the Galactic planeregion will determine how different isotopes are created in stars and dis-tributed in the interstellar medium; it will also unveil the recent historyof supernova explosions in the Milky Way. Furthermore, ASTROMEV willdetect a significant number of Galactic novae and supernovae in nearbygalaxies, thus addressing fundamental issues in the explosion mechanismsof both core-collapse and thermonuclear supernovae. The γ -ray data willprovide a much better understanding of Type Ia supernovae and their evo-lution with look-back time and metallicity, which is a pre-requisite for theiruse as standard candles for precision cosmology. The photon energy range from 100 keV to 1 GeV is crucial in different sectorsof astrophysics (for a review, see [37]).2.1 The extreme extragalactic UniverseThe Universe contains objects with extreme properties that can be studiedby measuring emission from particles that are accelerated near them. Theemission is very intense, permitting measurements at very large distance, or amma-ray Astrophysics in the MeV Range 5
Table 1:
Required instrument performance to achieve the core science objectives.
Parameter Value
Spectral range 100 keV – 1 GeVField of view ≥ . < × − MeV cm − s − at 1 MeV (any source)for 10 s observation time < × − MeV cm − s − at 10 MeV (high-latitude source)(3 σ confidence level) < × − MeV cm − s − at 500 MeV (high-latitude source)Line flux sensitivity < × − ph cm − s − for the 511 keV linefor 10 s observation time < × − ph cm − s − for the 847 keV SN Ia line(3 σ confidence level) < × − ph cm − s − for the 4.44 MeV line from LECRs ≤ . ◦ at 1 MeV (FWHM of the angular resolution measure)Angular resolution ≤ . ◦ at 100 MeV (68% containment radius) ≤ . ◦ at 1 GeV (68% containment radius)Polarization sensitivity Minimum Detectable Polarization <
20% (99% confidence level)for a 10 mCrab source in T obs = 10 s ( ∆E = 0 . − ∆E/E = 3% at 1 MeV ∆E/E = 30% at 100 MeVTime tagging accuracy 1 µ s (at 3 σ ) redshift, when the Universe was young and many galaxies still forming. Inmany cases, a substantial fraction of the radiated power appears in the MeVband, and so a new gamma-ray observatory sensitive in the energy domainwould offer an ideal view of the violent processes operating close by supermas-sive BHs, inside the powerful explosions that we see as gamma-ray bursts, andduring the merger of binary neutron stars (NS). By deciphering many aspectsof particle acceleration in the Universe, we address the question of why the en-ergy distribution is so unbalanced: only a few particles carry an extreme shareof the available energy, and by their feedback they shape numerous cosmicobjects.GRBs are explosive events with peak emission in the MeV band. Accuratepolarimetry in this energy domain would permit measuring the structure andamplitude of the magnetic field that shapes the acceleration and transportof particles [117]. Lorentz-invariance violation can be searched for [77], andtogether with future gravitational wave detectors the relation between GRBsand the mergers of compact objects can be determined.Clusters of galaxies are the largest gravitationally-bound structures in theUniverse. In fact, they are still forming, leading to particle acceleration atstructure formation shocks. Measuring their emission in the MeV band inconjunction with radio-band data lifts degeneracies in the interpretation andpermits a precise study of the energy redistribution into magnetic field andaccelerated particles, together with the feedback they impose on the clusterstructure [24].The MeV gamma-ray background (see Fig. 2) contains invaluable collec-tive information about nucleosynthesis in distant SNe, DM annihilation, andsupermassive BHs. The latter are often visible also visible as AGN, and theyare the most luminous persistent sources in the Universe, many of which emit Alessandro De Angelis et al.
Seyfert galaxies (Gilli 2007) All blazars (Giommi & Padovani 2015) Star-forming galaxies (Lacki et al. 2014) Radio galaxies (Inoue 2011)
Adapted from Ackermann et al. (2015)
ASTROMEV
Fig. 2: Compilation of the measurements of the total extragalactic gamma-ray intensity between 1 keV and 820 GeV [8], with different components fromcurrent models; the contribution from MeV blazars is largely unknown. Thesemi-transparent band indicates the energy region in which a new gamma-rayobservatory could dramatically improve on present knowledge.their bulk power in the MeV band. A sensitive observatory in this energydomain can use these unique beacons to study the formation history and evo-lution of supermassive BHs at times when the Universe had only a fractionof its current age. MeV-band observations address the energy limit to whichelectrons may be accelerated, the location where this happens. By studyingthe spectral response to changes in the activity of these objects, we can dis-tinguish the emission from electrons from that of energetic ions. The MeVband is ideally suited for this inquiry, because emission at higher gamma-rayenergies may be absorbed by Extragalactic Background Light (EBL), and thespecific contribution from photo-pair-production by high-energy cosmic nucleiis a critical discriminant in the soft gamma-ray band, as an analysis of the re-cent detection of a statistical association of a 300-TeV neutrino event [2] withan extended gamma-ray flare of the Active Galactic Nucleus TXS0506+056shows [67]. Finally, the MeV band carries the cascade emission of all the ab-sorbed Very-High-Energy (VHE) gamma-ray emission that is emitted in theUniverse, and so its study provides a unique view of its extreme particle ac-celeration history, including the feedback on the intergalactic medium and themagnetic-field genesis therein.Last but not least, the MeV range is the perfect companion for multi-messenger astronomy. On top of the Spectral Energy Distribution (SED) ofthe electromagnetic (EM) emission by TXS0506+056, mentioned before, therecent NS-NS merger generating the GW170817 event and the correspondinggamma-ray signal detected by
Fermi
GBM and
INTEGRAL has shown thatthe EM cutoff of this class of mergers is in the MeV range [81]. Furthermore,for a sufficiently close event, ASTROMEV could detect the continuum andnuclear line emissions expected from a kilonova (KN) following a merger event amma-ray Astrophysics in the MeV Range 7 like GW170817. This remarkable event highlighted the importance of KNs inthe nucleosynthesis of heavy elements by the r process [101,3]. The predictedline emission in the MeV band [65,80] could be detected with ASTROMEVup to a maximum distance of ∼
10 Mpc. The expected rate of KNs is stillquite uncertain: 200–400 KN Gpc − yr − if associated with GRB emission[38] and ∼ − yr − as Gravitational Wave (GW) detection only[3]. However, if the prompt emission of the KN is associated with substantialgamma-ray emission, absorption edges, possibly variable in time, from freshly-formed elements from the immediate environment of the source may also bedetected (e.g., [13]).ASTROMEV could allow to detect a sub-class of short GRBs having apeculiar origin, i.e., a transition from a Neutron star (NS) to a more compactstellar object called a Quark Star (QS) [126,97]. Distinctive signatures com-pared to the binary (NSNS or NSBH) merger scenario are the shortness of theprompt gamma-ray emission, estimated to be ∼ . ∼
100 keV with possibly a thermal spectrum and spectral features dueto the heavy composition of ejecta. In addition the associated GW emissionwould be different from the quoted binary merger (NS-NS, NS-BH) events.2.2 Cosmic ray interactionsA clear understanding of the origin and evolution of cosmic rays (CRs) is stillmissing despite one century of impressive observational discoveries and theo-retical progress [21]. Understanding their origin is an interdisciplinary probleminvolving fundamental plasma physics, to describe the diffusive shock accel-eration process, as well as astrophysical and particle-physics diagnostics, tocharacterize the particle properties and the local conditions in the accelera-tion zones. While we still lack a reliable explanation for the existence of CRsnear and beyond PeV energies in the Milky Way and beyond EeV energies inextragalactic space, we also know very little about the Galactic population oflow-energy CRs, with energies below a few GeV per nucleon. We still need in-formation on their sources and injection spectra into the interstellar medium,on their transport properties and flux distribution at all interstellar scales inthe Galaxy, and on their impact on the overall evolution of the interstellarmedium and on the dynamics of Galactic outflows and winds. The perfor-mance of a new mission such as ASTROMEV would provide unique results ina number of important CR issues.Sensitive observations of a set of CR sources, such as young SNRs, acrossthe MeV–GeV bandwidth, would allow for the first time to distinguish theemission produced by the interactions of CR nuclei with the ambient gas andthe non-thermal emission from CR electrons [52,9,29,73,84]. Combined withhigh-resolution radio and X-ray observations of the remnants (e.g., [104]),gamma-ray data in the MeV–GeV region would provide information on CRinjection into the acceleration process, on the structure of magnetic fields inside
Alessandro De Angelis et al.
Fig. 3: An example of the capability of ASTROMEV to transform our knowl-edge of the MeV-GeV sky. Upper panel: The upper left figure shows the 1-30MeV sky as observed by COMPTEL in the 1990s; the upper right figure showsthe simulated Cygnus region in the 1-30 MeV energy region from ASTROMEV.Lower panel: comparison between the view of the Cygnus region by
Fermi in 8years (left) and that by ASTROMEV in one year of effective exposure (right)between 400 MeV and 800 MeV.the remnants, and on the spectrum of CRs freshly released into surroundingclouds.
Fermi
LAT could resolve only one case of CR activity in a Galactic super-bubble to study the collective effects of multiple supernovae and powerfulwinds of young massive stars [6]. An improved angular resolution in the sub-GeV range would provide more case studies (see Fig. 3 for an illustration withthe Cygnus region), individually as well as collectively in the inner Galaxy,which would help to probe the interplay between CRs and the turbulentmedium of star-forming regions during the early steps of their Galactic voyage[28,59]. Individual massive binary stars like η Carinae, which is one of themost luminous massive binary systems in the Galaxy and the likely progeni-tor of the next Galactic supernova, are promising candidates to study particle amma-ray Astrophysics in the MeV Range 9 acceleration by their powerful winds [16]. Following their time variability fromradio to gamma-ray energies can provide key diagnostics on the accelerationefficiency.The
Fermi
Bubbles are one of the most spectacular and unexpected dis-coveries based on the
Fermi -LAT data [113,7]. However, the origin of thesegigantic lobes above and below the Galactic Center (GC) is still unknown:possible sources are outflows from the supermassive black hole Sgr A* (AGNscenario) or combined wind from massive star activities and supernova explo-sions in the central molecular zone (starburst scenario). Improving the angularresolution relative to the
Fermi -LAT PSF will be essential in the derivationof the shape of the
Fermi
Bubbles at energies below 1 GeV, and to detectthe expected spectral differences between the AGN (leptonic) and starburst(hadronic) models.CR nuclei of energies below a few GeV per nucleon contain the bulk en-ergy density of the Galactic CRs. They are the main source of ionizationand heating in the highly obscured star-forming clouds that are well screenedfrom UV radiation. At the same time they are the source of free energy andpressure gradients to support large-scale magnetohydrodynamic (MHD) out-flows and Galactic winds that control the overall evolution of a galaxy [59,94].MeV gamma-ray observations of the inner Galaxy with the targeted sensitivity(Fig. 1) would provide the first nuclear spectroscopic data on the low-energyCR population [18]. The energy coverage of the telescope would also allow aprecise separation of the CR nuclei and electron/positron populations (andspectra) across the Galaxy. The higher-resolution images (see Fig. 3) wouldshed light on the degree of correlation between the CR distributions and stel-lar activity, at the scale of cloud complexes up to that of spiral arms, in orderto better constrain the diffusion properties of CRs in a galaxy (e.g., [59,90]).Last, but not least, maps of the total interstellar gas mass inferred fromCRs and the GeV data from a new mission with a targeted resolution < (cid:48) at1 GeV (see Table 1) would serve a broad community wishing to improve thecalibration of gas tracers (radio and dust tracers) in a large variety of cloudstates [102].2.3 Explosive nucleosynthesis and chemical evolution of the GalaxyExploding stars play a very important role in astrophysics since they injectimportant amounts of kinetic energy and newly synthesized chemical elementsinto the interstellar medium in such a way that they completely shape thechemical evolution of galaxies. Furthermore, the “pyrotechnical” effects as-sociated with such outbursts can be so bright and regular that they can beused to measure distances at the cosmological scale. For instance, Type Ia SNe(SNIa) led to the discovery that the Universe was expanding in an acceleratedway [106,98].The majority of outbursts are associated with instabilities of electron de-generate structures in single stars (core collapse and electron capture super- Table 2:
Star-produced radioisotopes relevant to gamma-ray line astronomy.
Isotope Prod. site a Decay chain b Half-life c γ ray energy (keV)and intensity d7 Be Nova Be (cid:15) −→ Li* 53.2 d 478 (0.10) Ni SNIa, CCSN Ni (cid:15) −→ Co* 6.075 d 158 (0.99), 812 (0.86) Co (cid:15) (0 . −→ Fe* 77.2 d 847 (1), 1238 (0.66) Ni SNIa, CCSN Ni (cid:15) (0 . −→ Co* 1.48 d 1378 (0.82) Co (cid:15) −→ Fe* 272 d 122 (0.86), 136 (0.11) Na Nova Na β + (0 . −→ Ne* 2.60 y 1275 (1) Ti CCSN, SNIa Ti (cid:15) −→ Sc* 60.0 y 68 (0.93), 78 (0.96) Sc β + (0 . −→ Ca* 3.97 h 1157 (1) Al CCSN, WR Al β + (0 . −→ Mg* 7.2 · y 1809 (1)AGB, Nova Fe CCSN Fe β − −→ Co* 2.6 · y 59 (0.02) Co β − −→ Ni* 5.27 y 1173 (1), 1332 (1) a Sites which are believed to produce observable gamma-ray line emission. Nova:classical nova; SNIa: thermonuclear SN (type Ia); CCSN: core-collapse SN; WR:Wolf-Rayet star; AGB: asymptotic giant branch star. b (cid:15) : orbital electron capture. When an isotope decays by a combination of (cid:15) and β + emission, only the most probable decay mode is given, with the correspondingfraction in parenthesis. c Half-lives of the isotopes decaying by (cid:15) are for the neutral atoms. d The values in brackets correspond to the number of photons emitted in the gamma-ray line per radioactive decay. novae) or when they accrete matter from a companion in a close binary sys-tem (SNIa and classical novae, for instance). Systematic research on transientevents have revealed a surprising variety of outbursts that goes from “Ca-rich” transients, placed in the gap between Type Ia SNe and novae, Type Iax,“02es-like” SNe, “super-Chandrasekhar” SNe in the domain of the so-calledthermonuclear SNe [63,64], to, e.g., Type IIn, Type In, and so-called “impos-tors” in the domain of core collapse of massive stars [127,72,27].Many of these events, if not all, imply the activation of thermonuclearburning shells that synthesize new isotopes, some of them radioactive. As theejecta expand, more and more photons avoid thermalization and escape, suchthat they can be used as a diagnostic tool. Each of the different explosionscenarios leads to differences in the intrinsic properties of the ejecta, like thedensity and velocity profiles, and the nature and distribution of the radioactivematerial synthesized. This translates into differences in the light curves andline widths of the expected gamma-ray emission. Therefore, the observationwith gamma rays becomes a privileged diagnostic tool with respect to othermeasurements thanks to the penetration power of high energy photons andthe association of gamma-ray lines to specific isotopes created by the explosion[71]. amma-ray Astrophysics in the MeV Range 11 C o k e V li ne f l u x [ − ph c m − s − ] W7 (Chandrasekhar − Deflagration)He − DetonationMerger DetonationPulsating Delayed DetonationSuperluminous He − DetonationSPI DataSPI Exposure e-ASTROGAM SN 2014J
Fig. 4:
Light curve of the 847 keV line from Co decay in SN 2014J.
INTEGRAL data(adapted from Fig. 4 in Ref. [46], red data points) are compared to various models of TypeIa SN [116]. A simulation of the e-ASTROGAM response [36] to a time evolution of the 847keV line such as in the W7 model [91] shows that the sensitivity improvement by a newgamma-ray space mission (blue points) can lead to a much better understanding of the SNprogenitor system and explosion mechanism.
Table 2 displays the main detectable gamma-ray line emissions expected inseveral nucleosynthesis events (see Ref. [44] and references therein). The radio-isotopes with a relatively short lifetime can be used to directly characterize theindividual explosion events or the first stages of the remnant, while the long-lived ones, i.e., with lifetimes much longer than the characteristic time betweenevents, will produce a diffuse emission resulting from the superposition of manysources that can provide information on stellar nucleosynthesis [43,40], but alsoon the physical conditions and dynamics of the Galactic interstellar medium(see, e.g., [69]).It is important to distinguish here between guaranteed and opportunity ob-servations. By guaranteed, we understand observations that can be predictedwith enough anticipation and with the certitude that they can be included intothe ordinary mission scheduling. Three examples of guaranteed observationswould be:1. Measurement of the total mass of Ni/ Co ejected by SNIa. This valueis fundamental to calibrate the Phillips [100] relation and the yield of syn-thesized Fe. The explosion time and location are not known a priori, butthanks to the sensitivity of a new gamma-ray observatory [36], it is expectedthat about a dozen SNIa will occur at a distance smaller than 35 Mpc inthree years of mission. The observations will have to be performed around50–100 days after the explosion, when all the SN properties (subtype, lu-minosity,...) will already be known [30,31,46]. ASTROMEV will achieve amajor gain in sensitivity compared to
INTEGRAL for the main gamma-ray lines arising from Ni and Co decays (Fig. 4) allowing, for eventslike SN 2014J, the exquisitely accurate (at percent level) measurements of Al Radioac,vity: Special Messengers • Radioac,vity provides a clock • Al radioac,vity gamma rays trace nucleosynthesis ejecta over ~few Myrs • Radioac,ve emission is independent of density, ionisa,on states, … electrons in the ISM (free free radio emission, WMAP) (Benne&+2003 ) starlight (2 μm IR emission, (Skrutskie+2006 ) positrons in the ISM (511 keV γ -‐ray emission, INTEGRAL/SPI) (Siegert+2016 ) nucleosynthesis ejecta in the ISM (1809 keV Al γ -‐ray emission, CGRO/COMPTEL) (Diehl+1995 ) cosmic rays exci,ng ISM (GeV gamma-‐ray emission, Fermi-‐LAT) (Acero+2015 ) Fig. 5:
The diffuse emissions of our Galaxy across several astronomical bands: ASTROMEVwill explore the link between starlight (second image from top) and CRs (top and bottom).The current-best images of positron annihilation (3rd from top) and Al radioactivity (4thfrom top) gamma rays illustrate that this link is not straightforward, and e-ASTROGAMwill uncover more detail about the astrophysical links and processes. (Image composed byR. Diehl, from observations with WMAP, 2MASS,
INTEGRAL , CGRO, and
Fermi ; Refs.[5,19,109,108,41]) the Ni mass, the mass of the progenitor and the expansion velocity, easilydifferentiating between major astrophysical scenarios.2. Clumping degree of core-collapse SNRs as a diagnostic of internal asymme-tries [83,122]. This property can be obtained from the radioactive emissionof the Ti/ Sc chain [58,57]. The targeted sensitivity would allow thedetection of this emission in all young Galactic SNRs and in the remnantof SN1987A.3. Mapping of the positron annihilation radiation [74,79,68,125,108] and thelong-lived isotopes Al and Fe [42,123,124,85,39]. The expected hugeincrease in sensitivity compared to current gamma-ray missions shouldallow the building of detailed maps of these Galactic diffuse emissions (seeFig. 5), which will shed a new light on nucleosynthesis in massive stars,SNe and novae, as well as on the structure and dynamics of the Galaxy[69]. Individual objects (e.g., SNRs) should also be detected in these lines.Given the explosive nature of the events considered here, the majority of theobservations will belong to the category of Targets of Opportunity (ToO). The amma-ray Astrophysics in the MeV Range 13 information and the relevance of the observation will depend on the distanceof the events. Two examples would be:1. Novae. The targeted sensitivity would allow the detection of the Na (1275keV) line to a distance large enough to observe about one nova per year,but that of the Be (478 keV) line demands a shorter distance and is thusuncertain during the three years of nominal mission duration. Thereforethe results that can be obtained from every individual event will dependnot only on the nature of the event, but also on the distance [32,55,62].2. Type Ia and Core-collapse SNe. The detection of the early gamma-rayemission before the maximum optical light in the SNIa case [45,71] andthe determination of the amount of Ni ejected by CCSN [86] would befundamental to understanding the nature of the progenitor in the firstcase and of the explosion mechanism in both cases. Given the expectedsensitivity, it is foreseen to detect these details to a distance of about tenMpc, which ensures the detection of several events and opens the possibilityof comparing SN subtypes.The observation of ToOs is unpredictable, but extremely rewarding if success-ful, and exploding stars and related phenomena are within this category. It isimportant to realize that the targeted increase of sensitivity would guaranteethat a significant number of events will be observed in an effective way.2.4 Observatory science in the MeV domainSince the MeV domain is largely unexplored, the observatory science could beparticularly interesting. We summarize here some of the topics, referring thereader to [37] for a more complete treatment.
Neutron stars (NSs) and black holes (BHs) are the most compact objects inthe Universe, capable of distorting the structure of the space-time aroundthem. They are observed to manifest themselves in a great variety of ways:pulsating and bursting, accreting from a binary companion, interacting withits wind, or even merging with it. NSs are found both in binary systems,often with other compact stars such as white dwarfs or NSs, or as isolatedsources. The most extreme neutron stars, the explosive magnetars, are foundwith the same outward characteristics, such as spin period and in some casesalso surface magnetic field, as those of more placid rotation-powered pulsars,but they show a spectacular bursting and flaring activity in the gamma-rayband. Understanding the evolutionary link between different NSs classes andtheir inter-relation is one of the holy grails of compact object astrophysics.Observations at MeV energies can uniquely address this issue.The magnetar-like phenomenology is likely caused by a twisted toroidalmagnetic field structure capable of releasing a power larger than that of dipole spin-down and of causing instabilities, magnetic field reconnection, and crustalfractures that ultimately result in their spectacular flaring emission. Manymagnetars have hard non-thermal components extending to at least 100 keVwith no observed cutoffs, although one is expected in the MeV band fromCOMPTEL upper limits. Particle acceleration along closed field lines and itshighly non-linear competition with attenuation from photon splitting and pairproduction are not yet understood, nor is the geometry of the magnetic linesbundle where currents flow. The exploration of such hard tails and cutoffs aswell as their phase-resolved behavior and polarization with a sensitive, largethroughput gamma-ray space mission is key to resolving these issues.Another NS puzzle is the nature of pulsar gamma-ray emission. Fermi revo-lutionized gamma-ray pulsar studies increasing the number of pulsars detectedabove 100 MeV from 7 with CGRO/EGRET to about 200 today [4]. However,in the soft gamma-ray region there are only 18 detections above 20 keV andonly four have been detected with pulsed emission in the range 1 – 10 MeV.Such MeV pulsars appear to have the peaks of their spectral energy distri-butions at MeV energies, so the clues to their nature lie in measurements bymore sensitive detectors like ASTROMEV. Thus, as Fermi did at higher ener-gies, ASTROMEV can revolutionize the number of detected sources and ourunderstanding of pulsar physics at this energy.The application of pulsar emission models to current data is plagued by thepoor knowledge of pulsar inclination and viewing geometry for most sources.The expected polarization signature, in fact, depends significantly on the geo-metry of the system and the location of the emitting zones. Gamma-ray po-larization measurements with ASTROMEV will be crucial to nail down thesystem inclination (magnetic and spin axis with respect to the proper mo-tion), reveal the magnetic field topology, locate the emission region(s) in themagnetosphere and identify the emission mechanism. Particularly importantwill be ASTROMEV information on the misalignment between spin and propermotion axis, which is still highly debated and is linked to the way in whichthe kick is imparted to a proto-neutron star during its formation and to theduration of the physics of the acceleration phase (see, e.g., [110,92,87]). An-other key information is the misalignment between spin and magnetic fieldaxis, which is crucial to quantify the contribution of pulsars to GW emissionsince orthogonal rotators will be efficient sources of gravitational radiation [34,112,35,48].A number of pulsars in binary systems are thought to have intra-binaryshocks between the pulsar and companion star that can accelerate particlesof the pulsar wind to greater than TeV energies. Gamma-ray binaries, witha young rotation-powered pulsar in orbit around a massive Be star, showorbitally modulated emission at radio, X-ray, GeV, and TeV energies. Modelswith either inverse-Compton or synchrotron radiation can fit the X-ray toGeV spectrum and better measurements at MeV energies would constrainthe mechanism. Observations of accreting X-ray binaries, that contain eitherNS-NS or NS-BHs, at MeV energies can uncover the emission mechanismsthat are operating as well as the role of the jets in these sources. An exciting amma-ray Astrophysics in the MeV Range 15 possibility is the detection of a 2.2 MeV neutron-capture line coming fromthe inner parts of the accretion disc or from the NS atmosphere, which wouldbe a major discovery and give new constraints on accretion physics and thegravitational redshift at the NS surface, respectively.Binaries containing millisecond pulsars and low mass companions also showorbitally-modulated X-ray emission from intra-binary shocks and three of theseare observed to transition between rotation-powered and accretion-poweredstates [95,111,105]. Observations with the proposed ASTROMEV observatorywill fill in the spectral gap from 0.1 – 100 MeV to help us understand the natureof these transitions and the limits to acceleration in the pulsar wind shock.Many millisecond pulsars are found in globular clusters.
Fermi has discov-ered both gamma-ray emission from many clusters and also pulsations frompulsars within some clusters. The nature of the diffuse X-ray and TeV emissiondetected from several clusters is presently a mystery and could come from mag-netospheric emission or from electron-positron pairs ejected from the pulsarsin the cluster. ASTROMEV will map the extent of the diffuse X-ray compo-nent in the MeV range, settling the crucial question of its origin which is stillunanswered by current data.
The same gamma-ray emission mechanisms at play in celestial sources can bestudied in more detail, even if in different environmental conditions, in localgamma-ray sources such as those present in the Solar System. In particularthe interactions of CRs with radiation fields and matter, at the Sun and withother Solar System bodies, such as the Moon, the acceleration of particles andtheir emission in the upper atmosphere, the physics of magnetic reconnectionand particle acceleration in solar flares are examples of science objectives thatcan be explored by observing gamma rays coming from the Sun, the Moon,the Earth, and other bodies in the Solar System.Solar flares are the most energetic phenomena in the Solar System. Theseevents are often associated with explosive Coronal Mass Ejections (CMEs).The frequency of both flares and CMEs follows the 11-year solar activitycycle, the most intense ones usually occurring during the maximum. Whattriggers the flares is presently not completely understood. Flare energy maybe considered to result from reconnecting magnetic fields in the corona. Phe-nomena similar to solar flares and CMEs are believed to occur at larger scaleselsewhere in the Universe, such as in AGNs [47]. These energetic phenomenafrom the Sun are therefore the most accessible laboratories for the study ofthe fundamental physics of transient energy release and efficient particle ac-celeration in cosmic magnetized plasmas. The gamma-ray emission from solarflares results from the acceleration of charged particles which then interactwith the ambient solar matter in the regions near the magnetic field lines.Accelerated electrons mainly produce soft and hard X-rays via non-thermalbremsstrahlung. Accelerated protons and ions emit at higher energies: nuclearinteractions produce excited and radioactive nuclei, neutrons and pi-mesons.
All of these products subsequently are responsible for the gamma-ray emis-sion via secondary processes, consisting of nuclear gamma-ray lines in the1-10 MeV range and a continuum spectrum above 100 MeV. The high-energygamma-ray emission light curve can be similar to the one observed in X-rays,lasting for 10–100 s and indicating the acceleration of both ions and electronsfrom the same solar ambient. This is referred to as the “impulsive” phaseof the flare. However, some events have been found to have a long-durationgamma-ray emission, lasting for several hours after the impulsive phase. Anew gamma-ray mission covering a very broad energy range, from about 100keV to 3 GeV, will have the opportunity to study the evolution in time ofthe hard-X and gamma-radiation from each solar flare event, helping to con-strain models of acceleration and propagation. It will detect the de-excitationslines from accelerated ions, which will be fundamental to gain insight intothe chemical abundances and about the physical conditions where acceleratedions propagate and interact. Spectral analysis at higher energies will also al-low disentangling the electron bremsstrahlung and pion-decay components. Apolarized bremsstrahlung emission in hard X-ray from solar flares is expectedif the phase-space distribution of the emitting electrons is anisotropic withimportant implications for particle acceleration models.The Moon is one of the brightest sources of high-energy gamma rays in theSolar System. Gamma rays from the Moon originate in the shower cascadesproduced by the interactions of Galactic CR nuclei with the lunar surface.The lunar gamma-ray emission depends on the fluxes of the primary cosmic-ray nuclei impinging on the Moon and on the mechanisms of their hadronicinteractions with the rock composing the lunar surface. In addition to pro-viding a new accurate measurement of the lunar gamma-ray spectrum in thesub-GeV band, the proposed observatory will extend the energy range ob-served by previous missions towards lower energies. This feature will providethe unique opportunity to explore possible gamma-ray lines in the hundredsof keV to MeV region, originating from the decays of excited states producedin the interactions of CR nuclei with the lunar rock. Measurements of thegamma-ray flux from the Moon also provide a useful tool to study the prop-erties of CRs and to monitor the solar cycle, since it depends on the primaryCR nuclei fluxes, which change with the solar activity. The lunar gamma-raydata at low energies will also represent a powerful tool to monitor the solarmodulation and to study the CR spectra impinging on the Moon surface.Terrestrial gamma-ray flashes (TGFs) are very intense gamma-ray emissionepisodes coming from the upper atmosphere and strongly correlated with light-ning activity. They are generally interpreted as bremsstrahlung high-energyradiation emitted by free electrons in the air, accelerated to relativistic ener-gies by intense electric fields presents in the atmosphere under thunderstormconditions. The importance of gamma-ray observations from space satellitesflying in Low Earth equatorial orbit is based on the possibility of detectingTGFs in the tropical regions where the frequency of thunderstorms is higher.Gamma-ray observations should also confirm the possible presence of a high-energy population of TGFs emitting at energies greater than 40 MeV. amma-ray Astrophysics in the MeV Range 17
The topic of fundamental physics in the context of high-energy astrophysics isoften related to fundamental symmetries of nature which can be studied overcosmological distances, at high energies and in extreme environments.Gamma rays as a probe have been used for a variety of subjects in fun-damental physics, the most studied question for gamma-ray observations ingeneral being the quest for dark matter (DM). The exploration of topics infundamental physics that can be addressed with a new observatory in thegamma-ray MeV range is gaining momentum: axion-like particles and pri-mordial black-holes as well as possible observations elucidating the questionof matter-antimatter asymmetry and, last but not least, different aspects ofsearches for DM particles with some focus on small masses.The existence of DM is by now established beyond reasonable doubt, seee.g. [20,10], however its nature is one of the most pressing questions in sciencetoday. Among the most popular DM candidates are weakly interacting massiveparticles (WIMPs), with masses and coupling strengths at the electroweakscale. Besides the fact that many of these are theoretically very well motivated,such as the supersymmetric neutralino [75], an attractive feature of this classof candidates is that the observed DM abundance today can straightforwardlybe explained by the thermal production of WIMPs in the early Universe.WIMPs are searched for by a variety of techniques: directly by placing sensitivedetectors in underground locations with the aim to detect WIMP-inducednuclear recoils and indirectly by detecting the secondary products of WIMPannihilation or decay.WIMP candidates can also be produced at the Large Hadron Collider(LHC) by proton-proton collisions, which then would need to be confirmedby astrophysical observations. The latest LHC results, based on almost 40fb − of data at √ s = 13 TeV (e.g. [1]) did not reveal any sign of WIMP DM.In indirect detection the Fermi
Large Area Telescope managed to push thesensitivity below the canonical thermal WIMP cross-section for WIMPs in themass range from about 5 GeV to 100 GeV without firmly confirmed detection.There is, however, significant remaining uncertainty, e.g., on DM distribution,which motivates further searches. Direct detection, mainly led by deep under-ground liquid xenon time projection chambers, has improved sensitivity by twoorders of magnitude in the last decade without any DM evidence, see e.g. [82,14]. While clearly it is too early to abandon the WIMP paradigm, especiallyin the view of experimental programs in the next five years, the communityhas started to shift focus to alternative models for DM.A particularly interesting, and experimentally largely unexplored region isDM masses at or below the GeV scale. For example, thermal production mayalso be an attractive option for smaller DM masses [50]. Other relevant DMmodels with (sub-)GeV masses include light gravitino DM [115], inelastic DM[121], light scalar DM [22], or secluded DM [103]. Recently, an anomaly inthe absorption profile at 78 MHz in the sky-averaged spectrum [23] has been interpreted as an excess cooling of the cosmic gas induced by its interactionwith DM particles lighter than a few GeV [17].Targets for searches for DM are commonly those of enhanced DM density:the Milky Way galaxy, including the GC, dwarf galaxies and groups of galax-ies, as well as galaxy clusters. The GC is by orders of magnitude the largestpotential source of signal from DM annihilation. Dwarf spheroidal galaxiesprovide the cleanest target with the potential to derive the DM distributionfrom spectral velocities and are (unlike the GC) essentially free from con-ventional sources or diffuse backgrounds that could hamper an identificationof DM induced signal. Galaxy clusters are potential targets if a substantialfraction of DM is in substructures. Diffuse backgrounds, such as the Galac-tic and extragalactic backgrounds, are promising targets, especially exploitingangular autocorrelation or in cross-correlation with other wavelengths, using,for example, galaxy catalogues. For a more detailed review of challenges andopportunities of different gamma-ray signatures and techniques, see e.g. [33,54].
The instrument performance required to achieve the core science objectives,such as the angular and energy resolution, the field of view, the continuumand line sensitivity, the polarization sensitivity, and the timing accuracy, aresummarized in Table 1. – The very large spectral band is required to give a complete view of themain nonthermal processes at work in a given astrophysical object. The100 keV – 1 GeV energy band includes, in particular, the 511 keV linefrom e + e − annihilation, the nuclear de-excitation lines, the characteristicspectral bump from pion decay, the typical domains of nonthermal electronbremsstrahlung and Inverse Compton emission, as well as the high-energyrange of synchrotron radiation in sources with high magnetic field ( B ≥ – The wide field of view of the telescope is especially important to enable themeasurement of source flux variability over a wide range of timescales bothfor a-priori chosen sources and in serendipitous observations. Coupled witha sky-scanning mode of operation, this capability enables continuous mon-itoring of source fluxes that will greatly increase the chances of detectingcorrelated flux variability with other wavelengths. The wide field of viewis particularly important for the study of blazars, GRBs, Galactic com-pact objects, supernovae, novae, and extended emissions in the Milky Way(CRs, radioactivity). It will also enable, for example, searches of periodicityand orbital modulation in binary systems. – One of the main scientific requirements is to improve dramatically the de-tection sensitivity in a region of the electromagnetic spectrum, the so-called amma-ray Astrophysics in the MeV Range 19
MeV domain, which is still largely unknown. The sensitivity requirementis relevant to all science drivers discussed above. Thus, the goal of detect-ing a significant number (
N >
5) of SN Ia in gamma rays after 3 yearsrequires a sensitivity in the 847 keV line < × − ph cm − s − in 1 Msof integration time (Table 1). – Another major requirement for a future gamma-ray observatory is to im-prove significantly the angular resolution over past and current missions,which have been severely affected by a spatial confusion issue. The requiredangular resolution will improve
CGRO /COMPTEL and
F ermi -LAT by al-most a factor of 4 at 1 MeV and 1 GeV, respectively. The targeted angularresolution given in Table 1 is close to the physical limits: for Comptonscattering, the limit is given by the Doppler broadening induced by thevelocity of the atomic electrons, while for low-energy pair production, thelimit is provided by the nuclear recoil. Such an angular resolution will allowa number of currently unidentified gamma-ray sources (e.g. 1323 sourcesin the 3FGL catalog [120]) to be associated with objects identified at otherwavelengths. – The required polarization sensitivity will enable measurements of a gamma-ray polarization fraction >
20% in about 40 GRBs per year, and a polar-ization fraction >
50% in about 100 GRBs per year. Such measurementswill provide important information on the magnetization and content (lep-tons, hadrons, Poynting flux) of the relativistic outflows, and, in the caseof GRBs at cosmological distance, will address fundamental questions ofphysics related to vacuum birefringence and Lorentz invariance violation(e.g., [56]). The polarization sensitivity will also enable the study of thepolarimetric properties of more than 50 blazars, pulsars, magnetars, andblack hole systems in the Galaxy. – The required spectral resolution for the main science drivers of the missionis largely within the reach of current technologies (Sect. 4). Thus, themain gamma-ray lines produced in SN explosions or by low-energy cosmicray (LECR) interactions in the ISM are significantly broadened by theDoppler effect, and a FWHM resolution of 3% at 1 MeV is adequate. Inthe pair production domain, an energy resolution of 30% will be morethan enough to measure accurately putative spectral breaks and cutoffs invarious sources and distinguish the characteristic pion-decay bump fromleptonic emissions. – The required timing performance is mainly driven by the physics of magne-tars and rotation-powered pulsars (Sect. 2.4.1), as well as by the propertiesof TGFs (Sect. 2.4.2). The targeted microsecond timing accuracy is alreadyachieved in, e.g., the AGILE mission [119].Requirements for the Ground Segment are standard for an observatory-class mission. Target of Opportunity observations (ToOs) are required to followparticularly important transient events that need a satellite repointing.Table 3 summarizes our estimates of the number of sources detectable in3 years by a mission having the performance summarized in Table 1. It is
Table 3:
Estimated number of sources of various classes detectable in 3 years by a gamma-ray mission with the performance shown in Table 1. The last column gives the expectednumber of sources not known before in any wavelength.
Source type Number in 3 yr New sources
Galactic ∼ ∼ ∼ ∼ ∼ <
10 MeV) 70 – 100 35 – 50Supernovae 10 – 15 10 – 15Novae 4 – 6 4 – 6GRBs ∼ ∼ Total 3000 – 4000 ∼ based on current knowledge and log N − log S determinations of Galactic andextragalactic sources, including GRBs. It takes information from the latest Swift -BAT Hard X-ray survey catalog [114], the
INTEGRAL -IBIS catalog[70], and the 4th
Fermi -LAT catalog [120]. It is noteworthy that the lattercatalog contains more than 1300 unidentified sources in the 100 MeV – 300GeV range with no counterparts at other wavelength, and most of them willbe detected by the new gamma-ray mission, in addition to a number of newunidentified sources. The discovery space of such a mission for new sourcesand source classes would be very large.
Tracker of secondary particles in which the cosmic gamma rays undergo aCompton scattering or a pair conversion and a
Calorimeter to absorb andmeasure the energy of the scattered gamma rays and electron-positron pairs. Inaddition, an
Anticoincidence system covering the main detectors is neededto veto the prompt-reaction background induced by charged particles in space.Silicon represents the best choice of detector material for the Tracker be-cause of its low atomic number, which favors Compton interactions compared amma-ray Astrophysics in the MeV Range 21 (a) (b) (c)
Fig. 6: (a)
Assembly of 16 Si microstrip detectors in one layer of a
Fermi
LATtower [15]. (b)
Detail of the
Fermi
LAT Tracker showing the wire bondingstrip to strip of two Si detectors. (c)
Detail of the
AGILE
Tracker showingthe Si sensor bonding with the front-end electronics ASIC through a pitchadaptor (see [119] and references therein).to photoelectric absorption, as well as recent technological advances made onDouble-sided Silicon Strip Detectors (DSSDs) and readout microelectronics(see Fig. 6). In addition, the use of silicon as the scatterer makes it possibleto minimize the effect of Doppler broadening, which constitutes an essentialphysical limit to the angular resolution of a Compton telescope. To increasethe detection surface, Si detectors should be daisy-chained with wire bondingstrip to strip, each layer comprising typically 4 × × (a) (b) Fig. 7: (a)
INTEGRAL /PICsIT modular detection unit with the inset showingone of the CsI(Tl) bars [76]. (b)
ASIM /MXGS detector module comprisingfour 20 × × CZT detectors each having 64 pixels (2.5 mm pixel pitch)[93]. MXGS comprises 64 CZT modules providing sensitive area of 1000 cm .ASIM was launched on April 2018 and is mounted on the Columbus moduleon the International Space Station.The basic detector element of the Calorimeter should be made of a ma-terial with a high atomic number for an efficient absorption of the scatteredgamma rays and the electron-positron pairs. The Calorimeter needs to be a 3-D position-sensitive detector with good energy resolution to capture both Comp-ton and pair interactions, and also contribute efficiently to the backgroundrejection. A pixelated array of high- Z scintillation crystals, such as Thalliumactivated Cesium Iodine (CsI(Tl)) or Cerium Bromide (CeBr ), readout bysilicon drift detectors (SDD) can offer a high stopping power together withgood spectral and spatial resolutions. An array of state-of-the-art semicon-ductors such as CdZnTe (CZT), see Fig. 7, can also provide a very accuratemeasurement of the interaction location (sub-mm 3D position determinationfor E γ >
200 keV), energy determination ( <
1% FWHM @ 662 keV) of thescattered Compton photons and improved polarimetric performance [25,66].
Fig. 8:
Left panel AGILE
Anticoincidence detector flight unit [99].
Right panel
Scintillator tile detector assembly (shown unwrapped) of the
Fermi /LAT Anti-coincidence [89]. The green wavelength-shifting fibers carry light to the opticalconnector in the foreground.The Anticoincidence detector should achieve a charged particle backgroundrejection efficiency > . Fermi -LAT and
AGILE (Fig. 8). It is classicallydesigned with thin plastic scintillators covering the top and four sides of theinstrument. The scintillator tiles can be coupled to silicon photomultipliers(SiPM) by optical fibers, which should provide the best solution to collect thescintillation optical light.4.2 Measurement principleFigure 9 shows representative topologies for Compton and pair events. ForCompton events, point interactions of the gamma ray in the Tracker andCalorimeter produce spatially-resolved energy deposits, which have to be re-constructed in sequence using the redundant kinematic information from mul-tiple interactions. Once the sequence is established, two sets of informationare used for imaging: the total energy and the energy deposit in the first in-teraction measure the first Compton scatter angle. The combination with thedirection of the scattered photon from the vertices of the first and second in-teractions generates a ring on the sky containing the source direction. Multiple amma-ray Astrophysics in the MeV Range 23
Fig. 9: Representative topologies for a Compton event and for a pair event.Photon tracks are shown in pale blue, dashed, and electron and/or positrontracks are in red, solid.photons from the same source enable a full deconvolution of the image, us-ing probabilistic techniques. For energetic Compton scatters (above ∼ an e-ASTROGAM-like detector on a low inclination LEO is approximately65 kHz. After trigger level, this rate reduces to a total rate of around 4 kHz[36]. In the next data reduction step, smart and fast event selection is required.Data processing on board a satellite is constrained by computational resourcesand communication bandwidth. The former limits complexity of a processingchain, the latter sets an upper limit on the amount of data that can be trans-mitted to Earth and thereby sets a lower bound for the amount of selectionrequired to extract the relevant data. One straightforward way to raise theefficiency of on-board processing is to ensure the correct categorization of anevent at the beginning of the processing chain. This is a task where MachineLearning can play a major role. Since the tracker data has the largest dis-crimination power concerning the event type (Compton- or Pair-event) anyclassification attempt should start there. The x- and y-strips in each trackerlayer provide a natural way to generate images from the event using raw data(i.e., x-z- and y-z-maps in ADC channels). This makes the application of imagerecognition techniques feasible. Convolutional Neural Nets (CNN) [78] are theleading technique for this task. However the computational effort for process-ing an image via CNNs rises with the size of the image. The capability to runa complex CNN in an FPGA-accelerated System-on-Chip environments wasdemonstrated by e.g. [60]. There are also efforts to develop ASIC solutions forCNNs (e.g. [96]) that provide an increase in speed. Such a network could tageach event for further processing or disposal.In a further step, pair events are processed by an on-board Kalman filter(e.g. [53]) to check the event for viability. A natural representation of a Kalmanfilter in the context of Deep Learning is a Recurrent Neural Net (RNN) [88].In a small toy example, Gu et al. [61] show that the use of an RNN leads toa model that behaves equivalently to a Kalman filter but possesses a betterresistance to noisy input than the conventional approach. As above, furtherdevelopments are required to make such systems operate on space-grade pro-cessors.
Previous studies, in particular for e-ASTROGAM (a proposal for ESA’s M5opportunity), have shown that the scientific requirements presented in Sects. 2and 3 could be met by a M-size mission. The typical envelopes of the missionare: – Payload mass: about 1 ton – Satellite dry mass: about 2.5 tons – Satellite power: about 2 kW in nominal science operation – Telemetry budget: about 1.5 MbpsThe detection sensitivity requirement (Sect. 3) would be consistent withthe launch of the mission to an equatorial low-Earth orbit (LEO) (typicalinclination i < . ◦ and eccentricity e < .
01) of altitude in the range 550 – 600 amma-ray Astrophysics in the MeV Range 25 km. Such an orbit is preferred for a variety of reasons. It has been demonstratedto be only marginally affected by the South Atlantic Anomaly and is thereforea low-particle background orbit, ideal for high-energy observations. The orbitis practically unaffected by precipitating particles originating from solar flares,a virtue for background rejection. Finally, ESA has satellite communicationbases near the equator that can be efficiently used as mission ground stations.Extensive simulations of the detection performance using state-of-the-artnumerical tools [129,26] and a detailed numerical mass model of the satellitetogether with a thorough model for the background environment have shownthat a mission like e-ASTROGAM would achieve [118]: – Broad energy coverage ( ∼ – Excellent sensitivity for the detection of key gamma-ray lines e.g. sensitivityfor the 847 keV line from thermonuclear supernovae 70 times better thanthat of the
INTEGRAL spectrometer (SPI); – Unprecedented angular resolution both in the MeV domain and above a fewhundreds of MeV, i.e., improving the angular resolution of the COMPTELtelescope on board the CGRO and that of the
Fermi /LAT instrument bya factor of ∼ – Large field of view ( > – Pioneering polarimetric capability for both steady and transient sources[117].
Considering the ASTROGAM concept, the detector technology, silicon tracker,plastic scintillator-based anticoincidence, and crystal calorimeter have beenalready successfully used in space, and the payload would be based on a veryhigh (TRL) for all crucial detectors and associated electronics. However, amoderate R&D effort should be considered.For the silicon tracker, the 2D bonding of 4x4 (or even 5x5) DSSDs needssome R&D activities to implement mechanical jigs to guarantee the alignmentof silicon tiles during the bonding on the two sides of silicon planes. In addition,a bonding machine, able to work on a large area of such silicon planes, shouldbe identified on the market. The 2D bonding procedure has already beenestablished for the PAMELA [11] and AMS [12] space missions and it is wellestablished. The current fabrication technology of large silicon wafers up to 300mm in diameter could be also investigated to reduce the number of bondingsby using larger area of DSSDs.For the calorimeter, the technology and fabrication process of the SiliconDrift Detectors is the same as the one that was the subject of an extensive de-velopment activity within the assessment phase of the LOFT ESA M3 mission [49] and more recently for the eXTP project [128]. The low-noise front-endelectronics would require some efforts to optimize the signal to noise perfor-mance and to reduce the power consumption.A beam test campaign would also need to study the performance of thesingle detector and the whole system as well. In addition, environmental spacetesting would be also required to space qualify the assembling of the detectors.Finally, it is not unlikely that by the mid-XXI century new technologies areavailable and ready for space missions, from monolithic Si tracking elementsto completely new concepts.
The e-ASTROGAM concept for a gamma-ray space observatory can revolu-tionize the astronomy of medium/high-energy gamma rays by increasing thenumber of known sources in this field by more than an order of magnitude andproviding polarization information for many of these sources.The technology is ready but foreseen improvements in the next decade,in a framework called ASTROMEV, can further enhance the performance.ASTROMEV will be a protagonist of multi-messenger astronomy and playa major role in the development of time-domain astronomy. New windowsof opportunity (sources of gravitational waves, neutrinos, ultra high-energycosmic rays) will be fully and uniquely explored.
References
1. Aaboud, M., Aad, G., Abbott, B., et al. (ATLAS collaboration) 2017,Phys. Rev. Lett., 119, 1818042. Aartsen, M.G., et al. (IceCube collaboration), 2018, Science, 779, 1323. Abbott, B. P., Abbott, R., Abbott, T. D., et al. 2017, ApJ, 848, L124. Abdo, A. A., Ajello, M., Allafort, A., et al. 2013, ApJS, 208, 175. Acero, F., Ackermann, M., Ajello, M. et al., 2015, ApJS, 218, 236. Ackermann, M., et al., 2011, Science, 334, 11037. Ackermann, M., et al. (Fermi-LAT collaboration), 2014, ApJ, 793, 1, 648. Ackermann, M., Ajello, M., Albert, A., et al. 2015, ApJ, 799, 869. Ackermann, M., Ajello, M., Allafort, A. et al., 2013, Science, 339, 807(A13)10. Ade, P. A. R., et al. (Planck collaboration), 2016, A&A, 594, A1311. Adriani, O., Bonechi, L., Bongi, M., et al. 2003, Nuclear Instruments andMethods in Physics Research A, 511, 7212. Alcaraz, J., Alpat, B., Ambrosi, G., et al. 2008, Nuclear Instruments andMethods in Physics Research A, 593, 37613. Amati, L., Frontera, F., Vietri, M., et al. 2000, Science, 290, 95314. Aprile, E., et al. (XENON collaboration), 2017, Phys. Rev. Lett., 119,181301 amma-ray Astrophysics in the MeV Range 27
15. Atwood, W. B., Bagagli, R., Baldini, L., et al. 2007, Astroparticle Physics,28, 42216. Balbo, M. and Walter, R., 2017, A&A, 603, A11117. Barkana, et al., 2018, Nature, 555, 7118. Benhabiles-Mezhoud, H., Kiener, J., Tatischeff, V., & Strong, A. W., 2013,ApJ, 763, 9819. Bennett, C.L., et al., 2003, ApJS, 148, 120. Bergstrom, L., 2012, Annalen der Physik, 524, 9–10, 47921. Blasi, P. 2013, A&A Rev., 21, 7022. Boehm, C., & Fayet, P., 2004, Nucl. Phys. B, 683, 21923. Bowman, et al., 2018, Nature, 555, 6724. Brunetti, G. & Jones, T. W. 2014, International Journal of Modern PhysicsD, 23, 1430007-9825. Budtz-Jørgensen, C. and Kuvvetli, I., in IEEE Transactions on NuclearScience, vol. 64, no. 6, pp. 1611-1618, 201726. Bulgarelli, A., Fioretti, V., et al., 2012, Proc. SPIE, 8453, 84533527. Burrows, A., 2013, Rev. Mod. Phys., 85, 24528. Bykov, A. M., 2014, A&A Rev., 22, 7729. Cardillo, M., Tavani, M., Giuliani, A. et al., 2014, A&A, 565, 7430. Churazov, E., Sunyaev R., Isern J., et al., 2014, Nature, 512, 40631. Churazov, E., Sunyaev R., Isern J., et al., 2015, ApJ, 812, 6232. Clayton, D.D., & Hoyle, F., 1974, ApJ, 187, L10133. Conrad, J., Cohen-Tanugi, J., & Strigari, L. E., 2015, J. Exp. Theor. Phys.,121, 6, 110434. Cutler, C. 2002, Phys. Rev. D, 66, 08402535. Dall’Osso, S., Shore, S. N., & Stella, L. 2009, MNRAS, 398, 186936. De Angelis, A., Tatischeff, V., Tavani, M. et al. 2017, Exp. Astronomy, 44,25-8237. De Angelis, A., Tatischeff, V., Grenier, I.A. et al. 2018, J. of H. EnergyAstrophysics, 19, 1-10638. Della Valle, M., Guetta, D., Cappellaro, E., et al. 2018, MNRAS, 481,435539. Diehl R., 2013, RPPH, 76, 02630140. Diehl, R., 2016, Journal of Physics Conference Series, 703, 01200141. Diehl R., et al., 1995, A&A, 298, 44542. Diehl, R., Halloin, H., Kretschmer, K., et al. 2006, A&A, 449, 102543. Diehl, R., Halloin, H., Kretschmer, K., et al., 2006, Nature, 439, 4544. Diehl, R., Hartmann, D. H., & Prantzos, N., 2011, “Astronomy with Ra-dioactivities”, Lecture Notes in Physics, Berlin Springer Verlag, Vol. 81245. Diehl R., Siegert T., Hillebrandt W., et al., 2014, Science, 345, 116246. Diehl R., Siegert T., Hillebrandt W., et al., 2015, A&A, 574, A7247. Di Matteo, T. 1998, MNRAS, 299, L1548. Lasky, P. D. 2015, PASA, 32, e034. doi:10.1017/pasa.2015.3549. European Space Agency, https://sci.esa.int/documents/33900/35959/1567259880108-ESA_SRE_2013_3_LOFT.pdf
50. Feng, J. L., & Kumar, J., 2008, Phys. Rev. Lett., 101, 231301
51. Forot, M., Laurent, P., Grenier, I. A., Gouiff`es, C., & Lebrun, F. 2008,ApJ, 688, L2952. Giuliani, A., Cardillo, M., Tavani, M., et al., 2011, ApJ, 742, 30–3453. Giuliani, A., et al. 2006, Nucl. Instr. and Meth. A, 568, 2, 69254. Gaskins, J. M., 2016, Contemp. Phys. , 57, 4, 49655. G´omez-Gomar, J., Hernanz, M., Jos´e, J., & Isern, J., 2004, MNRAS, 296,91356. G¨otz, D., Laurent, P., Antier, S., et al. 2014, MNRAS, 444, 277657. Grefenstette, B. W., Fryer, C. L., Harrison, F. A., et al., 2017, ApJ, 834,1958. Grefenstette, B. W., Harrison, F. A., Boggs, S. E., et al., 2014, Nature,506, 33959. Grenier, I. A., Black, J. H., and Strong, A. W., 2015, ARA&A, 53, 19960. Gschwend, D., 2016, Zynqnet: An FPGA-Accelerated Embedded CNN,M.Th. ETH Zurich61. Gu, J., et al. 2017, Dynamic Facial Analysis: From Bayesian Filtering toRecurrent Neural Network, in: IEEE CVPR 201762. Hernanz, M. 2008, in Classical Novae, Second Edition, eds. M.F. Bode andA. Evans, Cambridge Astrophysics Series 43, CUP, Cambridge, 25263. Hillebrandt, W., & Niemeyer, J. C., 2000, ARA&A, 38, 19164. Hillebrandt, W., Kromer, M., R¨opke, F., & Ruiter, A., 2013, Front. Phys.,8, 11665. Hotokezaka, K., Wanajo, S., Tanaka, M., et al. 2016, MNRAS, 459, 3566. Howalt Owe, S., Kuvvetli, I., C. Budtz-Jørgensen, C., and A. Zoglauer,A., 2019, Journal of Instrumentation, vol. 14, no. 0167. The IceCube Collaboration et al., 2018, Science 361, eaat137868. Kn¨odlseder, J., Jean, P., Lonjou, V., et al., 2005, A&A, 441, 51369. Krause, M. G. H., Diehl, R., Bagetakos, Y., et al., 2015, A&A, 578, A11370. Krivonos, R., Tsygankov, S., Lutovinov, A., et al. 2012, A&A, 545,A27; see also
71. Isern J., Jean P., Bravo E., et al., 2016, A&A, 588, A6772. Janka, H.-T., 2012, Annual Review of Nuclear and Particle Science, 62,40773. Jogler, T. & Funk, S., 2016, ApJ, 816, 10074. Johnson, W. N., III, Harnden, F. R., Jr., & Haymes, R. C., 1972, ApJ,172, L175. Jungman, G., Kamionkowski, M., & Griest, K., 1996, Phys. Report, 267,19576. Labanti, C., Di Cocco, G., Ferro, G., et al. 2003, A&A, 411, L14977. Lang, R., et al., 2019, Phys. Rev. D 99, 04301578. LeCun, Y., et al. 1989, Neural Computation, 1, 4, 54179. Leventhal, M., MacCallum, C. J., & Stang, P. D., 1978, ApJ, 225, L1180. Li, L.-X. 2019, ApJ, 872, 1981. The LIGO and Virgo Collaborations, et al., 2017m ApJL, 848:L1282. Liu, J., Chen, X., & Ji, X., 2017, Nature Phys., 13, 3, 212 amma-ray Astrophysics in the MeV Range 29
83. Mahoney, W. A., Varnell, L. S., Jacobson, A. S., et al., 1988, ApJ, 334,L8184. Malkov, M. A., Diamond, P. H., & Sagdeev, R. Z., 2011, Nature Commu-nications, 2, 19485. Martin, P., Kn¨odlseder, J., Diehl, R., et al. 2009, A&A, 506, 70386. Matz, S. M., Share, G. H., Leising, M. D., Chupp, E. L., & Vestrand,W. T., 1988, Nature, 331, 41687. Mignani, R. P., Testa, V., Gonz´alez Caniulef, D., et al. 2017, MNRAS,465, 49288. Mirowski, P,, LeCun, Y., 2009, Dynamic Factor Graphs for Time SeriesModeling in: Machine Learning and Knowledge Discovery in Databases(ECML/PKDD’09)89. Moiseev, A. A., Hartman, R. C., Ormes, J. F., et al. 2007, AstroparticlePhysics, 27, 33990. Nava, L., Benyamin, D., Piran, T., & Shaviv, N. J., 2017, MNRAS, 466,367491. Nomoto K., Thielemann F.-K., Yokoi K., 1984, ApJ, 286, 64492. Noutsos, A., Kramer, M., Carr, P., et al. 2012, MNRAS, 423, 273693. Østgaard, N., Balling, J. E., Bjørnsen, T., et al. 2019, Space Science Re-views, 215, 2394. Pakmor, R., Pfrommer, C., Simpson, C. M., & Springel, V., 2016, ApJ,824, L3095. Papitto, A., Ferrigno, C., Bozzo, E., et al. 2013, Nature, 501, 51796. Parashar, A., et al. 2017, arXiv:1708.0448597. P´erez-Garc´ıa, M.A., Daigne, F., Silk, J., 2013, ApJ 768, 14598. Perlmutter, S., Aldering, G., Goldhaber, G., et al. 1999, ApJ, 517, 56599. Perotti, F., Fiorini, M., Incorvaia, S., Mattaini, E., & Sant’Ambrogio, E.2006, Nuclear Instruments and Methods in Physics Research A, 556, 228100. Phillips, M. M. 1993, ApJ, 413, L105101. Pian, E., D’Avanzo, P., Benetti, S., et al. 2017, Nature, 551, 67102. Planck collaboration,
Fermi collaboration, 2015, A&A, 582, A31103. Pospelov, M., Ritz, A., & Voloshin, M. B., 2008, Phys. Lett. B, 662, 53104. Reynolds, S. P., 2008, Ann. Rev. A&A, 46, 89105. Roy, J., Ray, P. S., Bhattacharyya, B., et al. 2015, ApJ, 800, L12106. Riess, A. G., Filippenko, A. V., Challis, P., et al. 1998, AJ, 116, 1009107. Sch¨onfelder, V., Bennett, K., Blom, J. J., et al. 2000, A&A Suppl., 143,145108. Siegert, T., Diehl, R., Khachatryan, G., et al., 2016, A&A, 586, A84109. Skrutskie M. F., et al., 2006, ApJ, 131, 1163110. Spruit, H. & Phinney, E. S. 1998, Nature, 393, 139111. Stappers, B. W., Archibald, A. M., Hessels, J. W. T., et al. 2014, ApJ,790, 39112. Stella, L., Dall’Osso, S., Israel, G. L., et al. 2005, ApJ, 634, L165113. Su, M., Slatyer, T.R., & Finkbeiner, D.P., 2010, ApJ, 724, 1044114. Swift BAT 105-Month Hard X-ray Survey Catalog at https://swift.gsfc.nasa.gov/results/bs105mon/ https://fermi.gsfc.nasa.gov/ssc/data/access/lat/8yr_catalog/https://fermi.gsfc.nasa.gov/ssc/data/access/lat/8yr_catalog/