The detector

The main challenge for the REDTOP experiment is represented by the rejection of the background since, at the energy where the experiment will run, the ratio of η-production to inelastic proton scattering is about 1:200.  Of course, the background events will look very different from the interesting events, both from the kinematic and topological point of view. In order to make the signal-to-noise more favorable, the experiment has been designed with the following two peculiarities:

  • The target is sparsified, with multiple elements along the beam axis. This will help to reduce the pile-up since the signal event and the background event will likely occur in different elements
  • Since the final events of interest in general contain fast leptons, the apparatus will use the Cerenkov process to detect those particles. By tuning appropriately the refractive index of the radiators, the background events, with only hadrons in the final states, will be mostly invisible to the experiment.



The Target Systems

The target systems (the red circle in the picture above) for REDTOP is composed of ten round foils of beryllium or lithium, each about 1/3 mm (3/4 mm for Li)  thick. The diameter of the foil is about 1cm. The target systems are held inside a beam pipe made of either carbon-fiber or beryllium: the pipe will also help in maintaining the vacuum and support the aerogel on its external wall.  A proton with 1.8 GeV of kinetic energy has a 2.5% probability to make an inelastic scattering in any of the foils. The probability that one such scattering would produce an η meson is about 0.47%. Since the average intensity of the beam is 1011 protons/sec, the expected number of η mesons produced in one year in the target systems is close to 1×1014


The Vertex Detector

The vertex detector has three main tasks:a)  participating in the reconstruction of charged tracks originating in the target region; b) rejecting photon converting in the target materials; c) reconstructing track with very low transverse momentum. Two detector technology options are currently being investigated for the vertex detector: a wafer-scale silicon sensor and a mat of thin scintillating fibers.

Option I – Wafer-scale silicon sensor

This option is based on the current R&D performed by Alice collaboration for their ITS3 detector. It consists of a three-layer of flexible Silicon Genesis: 20 micron thick wafer, with CMOS sensors. The wafer-scale sensors are 900 mm long, fabricated using stitching. A total of six such sensors is needed to cover the innermost region of REDTOP. extremely low material budget. Key benefits of this option are: a) a material budget corresponding to 0.02-0.04% X0, b)  homogeneous material distribution resulting in an almost-negligible systematic error from the material; c) 3-dimensional hit information. A picture of the ITS detector proposed by Alice is shown below.


Option II – Mats of scintillating fibers

The second option for the vertex detector is based on three layers of scintillating fibers, with a diameter of 0.25 mm. The technology chosen in this case is identical to that developed by the LHCb Collaboration for their planar tracker. However, the active surface of the tracker in the case of REDTOP is about 0.24 m2 vs 360 m2 for LHCb. The total readout channels is about 18,000. The expected vertex resolution is of the order of 70 um.

The Central Tracker.

Two options are currently being investigated for the central tracker: an Optical-TPC and a LGAD silicon tracker.

Option I – LGAD detector

A silicon tracker based on the LGAD technology is being considered a better alternative to the OTPC when a very low-Z material is considered for the target (like lithium or deuterium). In such cases, in fact, simulations with GenieHad have indicated that the multiplicity of a typical background event is less than 10 particles, easily reconstructable thanks to the high granularity of the detector. The tracker is better suited for the PIP-II run, where the particles have lower momentum and many muons of interest would be below the threshold of detection of the OTPC. Furthermore, the 3He ion needed for tagging the eta mesons at PIP-II, could be easily reconstructed in the LGAD detector, but they would be completely invisible in the OTPC. The structure of such device, a scaled version of SID vertex detector,  is shown in the picture below.


The tracker occupies the same spatial region as  the OTPC. It consists of five polyhedral layers for the barrel and four planar layers for the endcaps. The total area instrumented is about 11 m2. The pixels have a pitch of 3 mm. The total channels are about 17×106 and require more than 850 readout asics.

Option II – Optical TPC (OTPC)

The OTPC (the yellow dodecagon) works on the same principles as a conventional TPC. Namely, it will measure the momentum and the position in space of a charged track by using a deflecting solenoidal field. However, rather than using the ionization process to detect those particles, one uses the Cerenkov effect. The momentum of that particle is, then, reconstructed by the photons emitted inside the volume of the OTPC and detected by an array of photo-sensors surrounding the radiator. About 500,000 SiPM’s will be needed to cover the inner walls of the device. The pattern of detected photons will provide also the position in space of the initial particle. The advantage of an OTPC over a conventional TPC is that it will be sensitive only to those fast particles (leptons and fast pions ) that are generated in the decays of the η meson we are interested in (fully described in the REDTOP Physics page). On the other side, hadrons and slower pions will be under the Cerenkov threshold and, therefore, will not contribute to recorded hits. A prototype of an OTPC has been built and tested by the T1059 Collaboration at Fermilab.

The Cerenkov TOF

A Cerenkov radiator is located immediately inside the calorimeter for the Central Tracker option 2 (LGAD detector). It is made of small cubes of borosilicate glass with a side of 3 cm. The Cerenkov light produced in each cube by the particles escaping the Central Tracker are read-out by a SiPM. The detector has two purposes:

  • select the particles fast enough to make Cerenkov light (Cerenkov threshold detector);
  • measure the TOF of such particles.

That information is used in the Level-0 trigger to reject background events with slow particles.

The Cerenkov Radiators of the OTPC

Two Cerenkov radiators are present in the OTPC (option I above):

  • A double aerogel cylinder, about 3 cm thick at the inner wall, supported by the beam pipe (the dark green ring in the picture above) . The innermost aerogel has nD =1.22 while the outermost has nD = 1.3;
  •  low-pressure nitrogen gas filling the rest of the volume of the OTPC. The pressure is adjusted in order to have nD = 1.000145.

Muon particles with momentum larger than 160 MeV will radiate only in the aerogel. The radius, the center and the skewness of the ring will help to determine its velocity. The dual refractive systems will help in discriminating the pions from the muons. On the other side, electrons and positrons will radiate in aerogel as well as in the gas. The ring produced by such light particles has a radius mostly independent from their velocity and much larger than that of the muons. That feature will provide a particle identification for electrons vs. muons. Furthermore, the Cerenkov photons radiated from the gas will generate a characteristic pattern in the photo-sensors. From the analysis of that pattern, one can estimate the momentum of the electron and its position in space. The picture below shows what happens in REDTOP’s OTPC when a 100 MeV electron (red track) travels trough the gas in a 0.6 T magnetic field. Several Cerenkov photons (cyan tracks) are generated and detected by the optical sensors surrounding the gas.

REDTOP 100MeV e-


The ADRIANO2 Dual-Readout Calorimeter

REDTOP calorimeter plays an important role in searches for physics BSM. It has two main roles: a) measuring the energy of neutral particles with sufficient precision so that the rare decays of the f η and η’ mesons could be disentangled from the backgrounds; b) contributing to the PID systems to separate EM particles from proton and neutrons. It also participates in the Level-0 trigger, therefore, it needs to be very fast.

The main requirements for REDTOP Calorimeter are summarized below:

  • Energy resolutionσ(E)/E=3%/√E
  • Particle Identification (PID) between electromagnetic and hadronic showers with an efficiency better than 99%;
  • Time resolutions <80psec;
  • Detector response within ~100nsec

The first requirement is due to the necessity to reconstruct π0‘s generated in the target and the decaying into two photons and γe+eas their decay products contribute to the combinatoric background feeding in the signal. Furthermore, several decays of the η/η’ mesons contain γ’s or π0‘s. Therefore, an higher energy resolution of the latter will allow a more efficient reconstruction of events associated to η/η’ decays.

Particle Identification is important to reduce the hadronic background. Some form of PID needs to be implemented already in the level-0 trigger, to reduce the number of non- η/η’‘ events to a more manageable level. The requirement on the time resolution is part of PID, since baryonic particles (which constitute the largest fraction of particles generated in the primary interaction), have a much slower time-of-Flight (TOF) compared to the η/η’ decay products. Finally, a fast detector response is necessary considering the expected event rate in the calorimeter from inelastic p-Li events is of the order of 0.5 GHz. Assuming a pipeline with a depth of fifty, the event needs to be processed by the Level-0 trigger in less than 100 nsec.

The technology chosen for fulfilling the above requirements is a particular implementation of dual-readout calorimetry called: ADRIANO2, briefly described below. ADRIANO2 calorimeter is divided into two sections. The innermost section (about 40 cm in depth) is designed as a high-granularity, integrally active dual-readout calorimeter,  constituted by alternated tiles of lead-glass and scintillating plastic optically separated and read-out individually. Its main task is to reconstruct EM showers and measure the energy of photons and electrons.  The outermost section has a similar layout as the innermost section but includes tiles of passive material to reduce the nuclear interaction length of the detector so that showers generated by the hadrons are fully contained in the overall depth of the calorimeter (about 80 cm). Both sections have dual-readout capabilities and participate in the PID systems.

Dual-readout calorimetry is a novel detector technique that has received good attention in the last few years. One version of the latter, named ADRIANO (A Dual-Readout Integrally Active Non-segmented Option), has been under development at Fermilab by the T1015 Collaboration for many years. The picture below shows three ADRIANO prototypes tested at FTBF in 2015.


It is based on the two simultaneous measurements of the energy deposited by a hadronic or electromagnetic shower into two media with different properties. The first medium is usually a plastic scintillator (or any substance producing scintillating light). Consequently, any charged particle depositing energy in that medium will produce scintillation and will make a signal in the corresponding electronics.  The second medium is usually a heavy glass with high refractive index (nD >1.8) and high density (ρ >1.8). This medium will only be sensitive to charged particles with large velocity, such that light is emitted via the Cerenkov effect. Consequently, only the charged electromagnetic component of the shower (namely: electrons and positrons) or photons producing pairs will produce a signal in the corresponding electronics. Furthermore, the high density of the medium will make it an ideal integrally active absorber for all particles impinging on the detector. Summing the Cerenkov (C) and the scintillation (S) signal one has the total energy of the detected particle. Furthermore, by comparing the scintillation (S) vs. the Cerenkov (C) signals one also has information about the ID of the particle which generated the shower. This is shown, for example, in the plot below for simulated particles with Ekin=100 MeV.


The main advantage of using an ADRIANO calorimeter for detecting the photons from the η decay is that the nasty background from most neutrons entering the detector could be easily rejected. Furthermore, the different behavior in terms of S vs. C of muons and pions will help the double aerogel systems in disentangling the signals from the two different types of particles. The ADRIANO calorimeter corresponds to the outermost blue dodecagon in the sketch of the detector at the beginning of this page.

A novel implementation of the above technique, ADRIANO2 has been proposed for REDTOP. The building block of ADRIANO2 is a pair of, optically separated, plastic, and lead glass tiles each with a side of 3 cm. The lights generated by particles going through is collected with SiPM directly coupled to the tiles. The advantages vs the original technique are several:

  • The high granularity in 3-D of the detector helps in disentangling complex events, with high multiplicity showers;
  • Particle-flow  algorithms can be applied to improve the energy resolution of the calorimeter;
  • The prompt Cerenkov light generated inside the glass tiles is read-out immediately, without wavelength shifting, providing fast timing information to the trigger systems and complementing the dual-readout based, PID with TOF measurements.

In summary, ADRIANO2 merges the benefits of a dual-readout and of a CALICE-type calorimeter,  creating the base for a new generation of high-performance detectors.

The Muon Polarimeter (optional)

The muon polarimeter (the green bars in the detector sketch) is an array of plastic scintillators inserted between the inner and the outer shells of ADRIANO. They have the task of counting the number of electrons and positrons emitted when a muon is stopped inside the ADRIANO calorimeter. This happens in a short range of depths after the muon has lost all of its energy. The initial polarization of the muon is not lost in this process since the ADRIANO lead-glass is homogenous and isoscalar. If the muon carries a non-null polarization, the electrons (or positrons) are emitted non-isotopically in the solenoidal magnetic field. The muon polarimeter will count those electrons: any unbalance in the left-right or front-backward counting is associated with a non-null transverse or longitudinal polarization of the initial muon.

The Photon Polarimeter (optional)

The photon polarimeter (the purple ring in the sketch of the detector) works on the same principles as the muon polarimeter. A 5 mm thick scintillating plate is inserted inside the volume of the OTPC. When a photon hits that material it has a finite probability of converting into a e+e- pair. If the circular polarization of the photon is zero, then the decaying pair has an isotropic distribution inside the 0.6 T solenoidal magnetic field. Otherwise, the asymmetry of the direction of the e+e- pair will provide an estimate of the polarization of the initial photon.