The ASTRI Project: prototype status and future plans for a Cherenkov dual-mirror small-telescope array
S. Vercellone, O. Catalano, M.C. Maccarone, F. Di Pierro, P. Vallania, G. Bonnoli, R. Canestrari, G. Pareschi, P. Caraveo, N. La Palombara, M. Fiorini, L. Stringhetti, E. Giro, G. Tosti
aa r X i v : . [ a s t r o - ph . I M ] A p r th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012 The ASTRI Project: prototype status and future plans for a Cherenkovdual-mirror small-telescope array
S. Vercellone, O. Catalano, M.C. Maccarone
INAF – IASF Palermo, Via U. La Malfa 153, I–90146 Palermo, Italy
F. Di Pierro, P. Vallania
INAF – Osservatorio Astrofisico di Torino, Via Osservatorio 20, I–10025 Pino Torinese, Italy
G. Bonnoli, R. Canestrari, G. Pareschi
INAF – Osservatorio Astronomico di Brera, Via E. Bianchi 46, I–23807 Merate, Italy
P. Caraveo, N. La Palombara, M. Fiorini, L. Stringhetti
INAF – IASF Milano, Via E. Bassini 15, I–20133 Milano, Italy
E. Giro
INAF – Osservatorio Astronomico di Padova, Vicolo dell’Osservatorio 5, I–35122 Padova, Italy
G. Tosti
Dip. di Fisica, Universit `a degli Studi di Perugia, I–06123 Perugia, Italy on behalf of the ASTRI Collaboration
ASTRI (“Astrofisica con Specchi a Tecnologia Replicante Italiana”) is a flagship project of the Italian Ministryof Education, University and Research. Within this framework, INAF is currently developing a wide field of view(9.6 ◦ in diameter) end-to-end prototype of the CTA small-size telescope (SST), devoted to the investigationof the energy range from a fraction of TeV up to tens of TeVs, and scheduled to start data acquisition in2014. For the first time, a dual-mirror Schwarzschild-Couder optical design will be adopted on a Cherenkovtelescope, in order to obtain a compact optical configuration. A second challenging, but innovative technicalsolution consists of a modular focal surface camera based on Silicon photo-multipliers with a logical pixel size of6.2 mm ×
1. THE ASTRI PROJECT ANDPROTOTYPE
ASTRI (“Astrofisica con Specchi a TecnologiaReplicante Italiana”) is a flagship project of the Ital-ian Ministry of Education, University and Researchstrictly linked to the development of the ambitiousCherenkov Telescope Array (CTA, Actis et al. [2011]).CTA plans the construction of many tens of telescopesdivided in three kinds of configurations, in order tocover the energy range from a tens of GeV (Large SizeTelescope, LST), to a tens of TeV (Medium Size Tele-scope, MST), and up to 100 TeV and beyond (SmallSize Telescope, SST). Within this framework, INAFis currently developing an end-to-end prototype of theCTA small-size telescope in a dual-mirror configura-tion (SST-2M) to be tested under field conditions, andscheduled to start data acquisition in 2014.For the first time, a wide field of view (FoV =9 . ◦ in diameter) dual-mirror Schwarzschild-Couder(SC, Vassiliev et al. [2007]) optical design will beadopted on a Cherenkov telescope, in order to ob-tain a compact (f-number f/0.5) optical configurationand equipped with a light and compact camera basedon Silicon photo-multipliers with a logical pixel sizeof 6.2 mm × . ◦ (obtained by means of an optimization pro- cess among the commercially-available detectors, theoptics performance, and the overall costs of the proto-type). Figure 1 (left panel) shows the proposed tele-scope layout, whose mount exploits the classical alt-azimuthal configuration, and which is fully compliantwith the CTA requirements for the SST array. TheASTRI SST–2M prototype will be placed at Serra LaNave, 1735 m a.s.l. on the Etna Mountain near Cata-nia, at the INAF “M.G. Fracastoro” observing station,and will begin data acquisition in 2014 (Maccarone[2011]).
2. THE ASTRI DUAL-MIRRORSMALL-SIZE PROTOTYPE2.1. The Optical Design
The proposed layout (Canestrari et al. [2011]) isfully compliant with the CTA requirements for theSST array. Moreover, our design has been optimizedin order to ensure a light concentration higher than80% within the dimension of the pixels over the entirefield of view (Figure 1, right panel) and taking intoaccount the segmentation of the primary mirror (M1)and the dimension and position of the camera. Thetelescope design is compact having a 4.3 m-diameter eConf C121028 th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012
Figure 1:
Left panel:
Right panel:
Fraction of the light concentration into half pixel for different incidence angles.
M1, a 1.8 m-diameter secondary mirror (M2) and aprimary-to-secondary distance of 3 m. The SC op-tical design has an f-number f/0.5, a plate scale of37.5 mm/ ◦ , a logical pixel size of approximately 0 . ◦ and an equivalent focal length of 2150 mm. Consid-ering 1984 pixels, this setup delivers a FoV of 9 . ◦ indiameter and a mean value of the active area of about6.5 m , taking into account: the segmentation of M1,the obscuration of M2, the obscuration of the camera,the reflectivity of the optical surfaces as a function ofthe wavelength and incident angle, the losses due tothe camera’s protection window and the efficiency ofthe silicon detectors as function of the incident angles(ranging from 25 ◦ to 72 ◦ ). The primary mirror M1 is segmented into 18 tiles;the central one is not used because completely ob-structed by the secondary mirror M2. The segmenta-tion requires three types of segments having differentsurface profiles. Figure 2 shows an image of a pro-totype of one M1 mirror segment manufactured withthe glass cold-shaping technology. The segments havehexagonal shape with an aperture of 849 mm face-to-face. Each segment will be equipped with two actua-tors plus one fixed point for alignment. Only tilt mis-placements will be corrected. The secondary mirror ismonolithic, and has a radius of curvature of 2200 mmand diameter of 1800 mm. M2 will be equipped withthree actuators. The third actuator also makes thepiston/focus adjustment for the entire optical systemavailable.
The optical design will be implemented by meansof a telescope structure composed of primary and sec-
Figure 2: Photographic image of a prototype of one M1mirror segment. ondary mirrors cells, a pillar and counterweights, thedrives systems and a focal surface interface. The tele-scope mount exploits the classical alt-azimuthal con-figuration (see Figure 1, left panel).
The SC optical configuration allows us to de-sign a compact and light camera. Currently,the ASTRI camera has a dimension of about500 mm ×
500 mm ×
500 mm, including the mechanicsand the interface with the telescope structure, for atotal weight of about 50 kg (see Figure 3 for a camerasystem breakdown). Such small detection surface, inturn, requires a spatial segmentation of a few squaremillimeters to be compliant with the imaging resolvingangular size. Among the available light sensors thatoffer photon detection sensitivity in the 300–700 nm eConf C121028 th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012 Figure 3: Camera system breakdown at component level.
Photon Detection Module1 PDM = 4x4 Units ASTRI Focal Surface37 PDMsFoV diameter = 9.6 o HamamatsuS11828-3344Size imposed by the manufacturer1 Unit = 4x4 px1px = 3x3mm Figure 4: Modular approach adopted to cover with activedetectors the entire ASTRI FoV. band, a fast temporal response and a suitable pixelsize, we selected the Hamamatsu Silicon Photomulti-plier (SiPM) S11828-3344M (Hamamatsu [2011]). Inorder to cover the full 9.6 ◦ FoV, we used a modularapproach, as shown in Figure 4. We call
Unit thephysical aggregation (imposed by the manufacturer)of 4 × × × logical pixel , whichturns out to be of 6.2 mm × ◦ ), while the Photon Detection Module (PDM) is composed of 4 × Figure 5: On-axis simulated event on the ASTRICamera. See Section 2.4 for details. others, allowing maintenance of small portions of thecamera. To fit the curvature of the focal surface, eachPDM is appropriately tilted with respect to the opti-cal axis. Figure 5 shows an on-axis simulated event fora primary gamma-ray with E=10 TeV, a core distanceof 142.77 m and including a night-sky background of1 . × ph m − s − sr − , (about 3 p.e. pixel − ). Thecolor-bar shows the number of photo-electrons (p.e.)in each pixel. Although ASTRI SST–2M will mainly be a tech-nological prototype, it will perform scientific observa-tions on the Crab Nebula, MRK 421, and MRK 501.Preliminary calculations (Vallania et al. [2012]) showthat in the maximum sensitivity range ( ≥ σ in a fewhours, while in the energy range ≥
10 TeV a fluxlevel of 1 Crab at 5 σ can be reached in a few tensof hours. Figure 6 shows a comparison among the ex-pected ASTRI prototype sensitivity as a function ofthe energy (yellow stars, computed at 5 σ and 50 hrof observation) and those of a few Image AtmosphericCherenkov Telescope (IACT) ones (Whipple, MAGIC,H.E.S.S., CTA) and of large field of view detectorsfor one-year integration (Fermi-LAT) Because of theirstrong flux and spectral variations in the two Markar-ian sources, estimates of exposures are more uncer-tain. In case of large flares, with fluxes up to 5–10Crab Units, detection could be reached on a muchshorter time-scale (Bonnoli and Vercellone [2012]), al-lowing intra-night variability studies. eConf C121028 th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012
Figure 6: Expected ASTRI SST–2M prototypesensitivity as a function of the energy (yellow stars,computed at 5 σ and 50 hr of observation). Adaptedfrom Vallania et al. [2012] and reference therein.
3. THE ASTRI SST-2M MINI-ARRAY
A remarkable improvement in terms of performancecould come from the operation, in 2016, of a mini-array, composed by a few SST-2M telescopes and tobe placed at final CTA Southern Site. PreliminaryMonte Carlo simulations (Di Pierro et al. [2012]) yieldan improvement in sensitivity that for 7 telescopescould be a factor 1.5 at 10 TeV w.r.t. H.E.S.S., asshown in Figure 7. The ASTRI SST–2M mini-arraywill be able to study in great detail relatively bright (afew × − erg cm − s − at 10 TeV) sources with anangular resolution of a few arcmin and an energy res-olution of about 10–15 %. The ASTRI SST–2M mini-array sensitivity were calculated taking into account5 energy bins per decade, a 5 σ significance, a numberof event/energy-bin ≥
10, a signal rate >
5% w.r.t.the background rate, an integration time of 50 hr, aminimum number of images used in the event recon-struction of 3 and 5, respectively, and for an arrayconfiguration as shown in Figure 8.Moreover, thanks to the array approach, it will bepossible to verify the wide FoV performance to detectvery high energy showers with the core located at adistance up to 500 m, to compare the mini-array per-formance with the Monte Carlo expectations by meansof deep observations of few selected targets, and toperform the first CTA science, with its first solid de-tections during the first year of operation. Prominentsources such as extreme blazars (1ES 0229+200),nearby well-known BL Lac objects (MKN 501) andradio-galaxies, galactic pulsar wind nebulae (CrabNebula, Vela-X), supernovae remnants (Vela-junior,RX J1713.7 − Figure 7: Expected ASTRI SST–2M mini-arraysensitivity as a function of the energy (computed at 5 σ and 50 hr of observation), compared with H.E.S.S. andCTA ones (see Section 3 for details).From Di Pierro et al. [2012].Figure 8: A possible layout for a 7-telescope mini-arrayconfiguration (see Section 3 for details).From Di Pierro et al. [2012]. well as the Galactic Center can be observed in a previ-ously unexplored energy range, in order to investigatethe electron acceleration and cooling, relativistic andnon relativistic shocks, the search for cosmic-ray (CR)Pevatrons, the study of the CR propagation, and theimpact of the extragalactic background light on thespectra of the sources. The large field of view of theASTRI mini-array will allow us to monitor, duringa single pointing, a few TeV sources simultaneously.Figure 9 shows the current TeV sources as listed inthe TeVCat compilation. Red, green and cyan cir-cles represent the 9 . ◦ (optical) field of view diam-eter for three possible pointings along the Galactic http://tevcat.uchicago.edu/ eConf C121028 th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012 -180+180 Figure 9: Blue dots are the known TeV sources as listed in the TeVCat Catalogue. The grey line represents theCelestial Equator. The red, green and cyan circles are the ASTRI mini-array (optical) field of view. The left panels arezooms centered on the ASTRI mini-array pointings.Figure 10: Pulsar wind nebula Vela-X spectral energydistribution. See Hinton et al. [2011] for details.
Plane. The grey line represents the Celestial equa-tor. Although the actual sensitivity will substantiallydrop for off-axis sources, a few targets can be moni-tored simultaneously, as shown in the three panels onthe left. Simultaneous detection of hard and intenseGalactic sources could be feasible, e.g. in the caseof Vela–X and Vela–Jr. Several scientific cases canbe addressed by the ASTRI mini-array. For the firsttime, the energy range above a few tens of TeV canbe explored with an improved sensitivity comparedto the current IACTs. The nearby and powerful pul-sar wind nebula (PWN) Vela-X is a typical sourcewhich can be considered as a primary target for theASTRI mini-array. Figure 10 shows its spectral en-ergy distribution (SED), as reported in Hinton et al.[2011], where a clear peak is visible in the H.E.S.S.data (Aharonian et al. [2006a]) at about 10 TeV, anda cut-off at about 70 TeV, making the ASTRI mini-array crucial to explore this portion of the SED.
Figure 11: Supernova remnant RX J1713.7 − Supernova remnants (SNR) are typical GalacticTeV emitters. RX J1713.7 − Fermi (Abdo et al. [2011]) and the combined study withH.E.S.S. (see Figure 11), show that the high-energyand very high-energy (VHE) emission could be inter-preted in the framework of a leptonic scenario. Nev-ertheless, the good energy resolution of the ASTRImini-array above 10 TeV and its improved sensitivitybeyond a few tens of TeV, will improve our knowledgeon the main emission mechanism acting in this sourcein the GeV and TeV energy bands.The ASTRI mini-array will be extremely important eConf C121028 th Fermi Symposium : Monterey, CA : 28 Oct-2 Nov 2012 -14 -13 -12 -11 HESS VERITAS E20,low IR E19,low IR E19, best fit E14, low IR 5 hr CTA sensitivity 50 hr ν F ν ( e r g c m - s - ) E (GeV)
Figure 12: SED of the extreme blazar 1ES 0229+200(see Murase et al. [2012] for details). to investigate the VHE emission from extragalacticsources as well. Figure 12 shows the SED of the ex-treme blazar 1ES 0229+200 (Murase et al. [2012]). Aclear detection of VHE emission above a few tens ofTeV from such a blazar could provide fundamentalinformation on the long-standing debate on the emis-sion mechanisms in this energy band. In particular,since the cosmic-ray-induced cascade displays a sig-nificantly harder spectrum above 10–20 TeV, a detec-tion above ∼
30 TeV would be only compatible withan hadronic origin of the gamma-rays (Murase et al.[2012]).
4. SUMMMARY
The ASTRI SST–2M end-to-end prototype will beinstalled and operated during Spring 2014 at theINAF Observing Station in Serra La Nave, Sicily. TheASTRI prototype performance will provide crucial in-formation on several topics, such as the dual-mirrorSchwarzschild-Couder optical design, the SiPM-basedfocal surface and the software/data-handling architec-ture, all of them innovative with respect to the cur-rent IACT design. Moreover, the prototype site willallow us to obtain a direct measurement of promi-nent gamma-ray sources, such as the Crab nebula,MRK 421 and MRK 501. The planned ASTRI mini-array, operated starting from 2016, will constitute thefirst seed of the future CTA Project, and will be opento the CTA Consortium for both technological andscientific exploitation.
Acknowledgments
This work was partially supported by the ASTRIFlagship Project financed by the Italian Ministry of Education, University, and Research (MIUR) andlead by the Italian National Institute of Astrophysics(INAF).
References
M. Actis, G. Agnetta, F. Aharonian, A. Akhperjanian,J. Aleksi´c, E. Aliu, D. Allan, I. Allekotte, F. Antico,L. A. Antonelli, et al., Experimental Astronomy ,193 (2011), 1008.3703.V. Vassiliev, S. Fegan, and P. Brousseau, AstroparticlePhysics , 10 (2007), arXiv:astro-ph/0612718.M. C. Maccarone, Tech. Rep. ASTRI-TN-IASFPA-3300-009, INAF/IASFPA (2011).R. Canestrari, P. Conconi, and O. Catalano,Tech. Rep. ASTRI-IR-OAB-3100-009, INAF/OAB(2011).Hamamatsu, Tech. Rep. KSX-0046 B-2, HamamatsuPhotonics K.K., Solid State Division (2011).P. Vallania, F. Di Pierro, and C. Morello, Tech.Rep. ASTRI-MC-IFSITO-5000-001, INAF/OATo(2012).G. Bonnoli and S. Vercellone, Tech. Rep. ASTRI-TN-IASFPA-3000-018, INAF/IASFPA (2012).F. Di Pierro, C. Morello, and P. Vallania, Tech.Rep. ASTRI-MC-IFSITO-5000-005, INAF/OATo(2012).J. A. Hinton, S. Funk, R. D. Parsons, and S. Ohm,Astrophysical Journal Letters , L7 (2011),1111.2036.F. Aharonian, A. G. Akhperjanian, A. R. Bazer-Bachi, M. Beilicke, W. Benbow, D. Berge,K. Bernl¨ohr, C. Boisson, O. Bolz, V. Borrel,et al., Astronomy & Astrophysics , L43 (2006a),arXiv:astro-ph/0601575.H. Muraishi, T. Tanimori, S. Yanagita, T. Yoshida,M. Moriya, T. Kifune, S. A. Dazeley, P. G. Edwards,S. Gunji, S. Hara, et al., Astronomy & Astrophysics , L57 (2000), arXiv:astro-ph/0001047.F. Aharonian, A. G. Akhperjanian, A. R. Bazer-Bachi, M. Beilicke, W. Benbow, D. Berge,K. Bernl¨ohr, C. Boisson, O. Bolz, V. Borrel,et al., Astronomy & Astrophysics , 223 (2006b),arXiv:astro-ph/0511678.A. A. Abdo, M. Ackermann, M. Ajello, A. Allafort,L. Baldini, J. Ballet, G. Barbiellini, M. G. Baring,D. Bastieri, R. Bellazzini, et al., Astrophysical Jour-nal , 28 (2011), 1103.5727.K. Murase, C. D. Dermer, H. Takami, and G. Migliori,Astrophysical Journal , 63 (2012), 1107.5576., 63 (2012), 1107.5576.