The Silicon Charge Detector of the High Energy Cosmic Radiation Detection experiment

The High Energy Cosmic Radiation Detector (HERD) is an experiment designed for direct measurement of cosmic rays on the Chinese Space Station in 2027. Its goals include precise measurements of cosmic ray energy spectra, mass composition, electron/positron spectra, cosmic rays anisotropy, gamma ray astronomy and indirect searches for Dark Matter. HERD features a 55X0, homogeneous, 3D segmented imaging calorimeter and can detect particles from the top and four lateral sides, providing precise energy measurements and electron/proton separation for a wide field of view. A key detector in HERD is the Silicon Charge Detector (SCD). SCD measures the charge of particles before interaction with other materials, minimizing cosmic rays nuclei fragmentation and reducing systematics on nuclei flux measurement. Thorough studies, TCAD and/or SPICE simulations, and accelerator tests on prototypes have been conducted to evaluate the tracking and charge resolution capabilities of the SCD. Further testing with 300 μm detectors is planned in the coming months to fully characterize the SCD's performance. This paper presents the results of the simulation studies and of the measured performance with particles.


Introduction
The High Energy cosmic-Radiation Detection (HERD) facility [1] is a forthcoming experiment that is scheduled to be installed on board China's Space Station (CSS) in approximately 2027.The experiment is based on a novel design that leverages the advantages of a homogeneous, isotropic, and finely segmented 3D calorimeter.
The space mission will significantly expand the direct measurements of cosmic rays and gamma rays by at least one order of magnitude in energy, surpassing current limits.As a result, the experiment will achieve important and pioneering objectives in various research fields, including dark matter exploration, cosmic ray observations, and gamma ray astronomy, using a single instrument.
The HERD experiment aims to expand the direct measurements of proton and helium fluxes up to a few PeV, and other nuclear species up to tens or hundreds of TeV/n.It will provide the first direct measurements of the knee structure for proton and helium, and enhance the measurements of the spectral features of all nuclear species.
Figure 1 displays the current design of the detector, with an exploded view of the primary subsystems.-1 -The primary subsystem of HERD is the homogeneous, electromagnetic calorimeter (as visible in figure 1, where only the external support structures are shown).It comprises 7497 scintillating crystals composed of lutetium-yttrium oxyorthosilicate (Lu2(1-x)Y2xSiO5, commonly referred to as LYSO).The cubes are arranged in a spherical-like 3D mesh, enabling the acceptance of particles not only from the zenith direction but also from the four lateral sides.The bottom side is reserved for mechanical supports and the platform.This configuration allows for a rough increase in instrument acceptance by a factor of 5 compared to a conventional design, similar to the current generation of in-orbit detectors, while maintaining the same mass and power budget.

The Silicon Charge Detector
The Silicon Charge Detector (SCD) is the outermost detector of the experiment, consisting of five thin detector units covering 5 of the six available faces of the experiment assembly (figure 2).These five units are composed of silicon microstrip detectors mounted on an aluminum honeycomb and carbon fiber structure to minimize the passive material density.The purpose of the SCD is to measure the charge of incident particles up to Z = 26.Together with the data obtained from PSD, the SCD will reconstruct the charge with an excellent resolution of the order of 0.1 charge units.A further function of SCD is the incident particle track reconstruction, with a spatial resolution on the single measurement better than 40 μm.
By measuring particle charge before interaction with other materials, the SCD will minimize the cosmic rays nuclei fragmentation, avoiding early charge-change interactions in PSD.This will reduce the systematic uncertainty on the reconstructed charge due to the nuclei fragmentation.Being highly segmented it will also allow to minimize the effects of backscattered secondary particles coming from the calorimeter by means of its tracking capabilities.
-2 -To evaluate the SCD's charge measurement performance, a full HERD detector GEANT4 simulation was conducted, considering the charge collection processes in the silicon sensor.The charge resolution was estimated using nuclei with energies ranging from 10 GeV/n to 1 TeV/n.
Two key parameters were examined: Charge Collection Efficiency (CCE) the amount of signal collected after summing all the readout strips, and Charge Sharing Ratio (CSR), the ratio between signal on the readout strips next to the injection with respect to the one on the injection.Different sensor configurations were studied using TCAD and/or SPICE simulations, including variations in implantation pitch, implantation width, readout pitch, and decoupling capacitance, to choose the final detector configuration to be used.

Flight configuration design
The flight configuration of the SCD will consist of five thin detector units.One unit, placed on the TOP side, is a 1.6 × 1.6 m 2 square, while the others, on the side faces, measure at 1.4 × 0.9 m 2 .
Each of the unit contains 8 layers silicon microstrip detectors, mounted alternately in orthogonal directions.Each XY pair is composed of a supporting structure made of a woven carbon fiber frame with high modulus unidirectional carbon fiber skins and an aluminum honeycomb core (figure 3).Silicon detector "ladders" will be used, where several silicon sensors are daisy chained and read out from one side by dedicated electronics.Each of the side units will contain 1024 silicon tiles, while the top unit will have 2048 of them, for a total of 6144 tiles.With the use of the designed 97.5 × 97.5 mm 2 silicon sensors this gives rise to a total silicon area of 58.41 m 2 [6].
To reduce the number of the readout channels while maintaining a satisfactory performance in terms of spatial and charge resolution the readout is performed with the so called floating strip approach, where a certain number of the implanted strips are not read out.
Several options are being evaluated to readout the sensor.These include the more expensive commercial option IDEAS IDE1140 [7], as previously used for space experiment silicon trackers, and the in-house developed INFN-To ASTRA ASIC by INFN (Istituto Nazionale di Fisica Nucleare) [8], which is suitable for the readout of long microstrip sensors with a high dynamic range.

The prototype
A first prototype version of the detectors to be used for the SCD was constructed in 2022.The silicon sensors, made by Hamamatsu Photonics, are 150 μm thick, and have a 96 × 96 mm 2 active area segmented in 1920 strips with a 50 μm implantation pitch.
Of the 1920 strips implanted on the silicon, only 640 of them are read out, for a final readout pitch of 150 μm.The capacitive coupling between readout strips and floating strips, combined with the analog readout of the channels give raise to a substantial gain for spatial resolution, while also helping with charge collection for particles hitting far from the readout region.The behaviour of this readout scheme implementation is well known from past experiences [9].

CERN SPS 2022 data campaign
The detectors that compose the prototype were tested at CERN SPS on the ion beam at the H4 experimental area in the Autumn of 2022, where a total of 4 measuring stations have been used.The ion beam employed was generated through the collision of a 150 GeV/n Lead primary beam extracted from the CERN Super Proton Synchrotron with a 4 cm Beryllium target.Following this, the secondary fragment beam was carefully selected based on its rigidity using magnetic optics, enabling the precise selection of the beam's A/Z ratio.This secondary beam consisted of a wide range of elements with atomic numbers (Z) spanning from 1 to 82.
Next, the ion beam was directed to the experimental area, where prototypes of SCD were installed, as depicted in figure 4(a), togheter with prototypes of the other HERD subdetectors.
Each SCD station is equipped with two single-sided silicon strip detectors placed front-to-front to measure the two orthogonal views perpendicular to the beam direction of the apparatus, as showed in figure 4(b).Each detector is made of one AC-coupled silicon microstrip sensor, read out by ten IDE1140 ASIC chips.-4 -Data acquired has been processed with dedicated software to reconstruct hit clusters from the signal of neighboring strips after a cut on signal-to-noise ratio.The clusters were subsequently associated to a 3D track employing a combinatorial approach.A spatial alignment procedure has been applied, employing an iterative method focused on minimizing track residuals.
To correctly reconstruct the impinging particle charge, two sets of corrections have been computed and applied to cluster data, to account for the variation in gain of the different IDE1140 ASICs in each detector and the signal response of the detector with respect to the impact position of the particle.To correct the gain of the different ASICs, the peak position for different Z values is evaluated from clusters in the readout region.The gain of each IDE1140 ASIC is then evaluated from the offset of the reconstructed peaks from those of the reference chip.
In order to correct for the dependence from the impact position, a second correction is required.The non linear charge division across readout strips is a well-known feature of silicon microstrip detectors using floating strip readout strategies.In order to study the behaviour of the charge collection as a function of the impact point between the readout strips we define the parameter  =  1 /( 1 +  2 ), where  1 and  2 are the two highest signals in the cluster.The effect of the capacitive coupling gives rise to the presence of peaks in the  distribution, with two extra peaks at about 1/3 and 2/3 as expected from the presence of two floating strips (figure 5(a)).Both the  distribution and the  2 / 1 ratio (figure 5(b)) show a charge coupling between the readout strips of about 0.8%.While this dependence means that the signal for a given charge depends on the impact position, the interstrip region effectively acts as a "low gain" region.Defining the  parameter of each cluster it is possible to select the particles impinging near a readout strip ( values near 0 or 1) or inbetween two readout strips (intermediate  values).Selecting only clusters in the readout region it is thus possible to reconstruct up to Z = 8, while for particles in the floating strip region it is possible to reconstruct up to Z = 14.The use of the strips of the cluster not used to define the  should give capabilities in reconstructing Z > 14 particles.
To get from the raw cluster ADC content to a reconstructed charge value an handwritten peak finding procedure is applied to characterize the signal variation with , with the ASICs of the same sensor put together after gain correction (figure 6).
-5 - After all the corrections have been applied, it is finally possible to combine the info from all sensors to estimate the fragment reconstructed charge.The setup allows to clearly distinguish charges up to at least Z = 10 (figure 7(a)): the charge distribution was obtained by averaging the measurement of the clusters reconstructed by 8 sensors on the beam that make up a track.In magenta it is also showed the obtained resolution restricting to measurements with a limited spread.The resolutions achieved for Z up to 10 are then compared to Monte Carlo simulations and a current setup in orbit: figure 7  -6 -

Conclusions
The High Energy cosmic-Radiation Detection (HERD) facility will be installed on the Chinese Space station in 2027 and will search for signatures of Dark Matter particles and study the cosmic rays knee up to the PeV.The Silicon Charge Detector (SCD) will be used for charge identification and particle tracking, with large area coverage (approx.60 m 2 ) using silicon strip detectors arranged on 5 of the 6 faces of the experiment.Charge measurement up to Z = 26 and tracking resolution better than 40 μm are foreseen for the final detector.
The first prototype with a 150 μm thick sensor has been tested at CERN SPS in October 2022 with a fragmented ion beam.The corrections to be applied to the charge reconstruction algorithm have been evaluated and the combined info of several detectors allowed the reconstruction of charges up to Z = 16 with good charge and spatial resolution.The use of external detectors is needed to study spatial resolution and charge reconstruction for Z > 16.Finally, a second prototype with a 300 μm thick sensor will be tested in a forthcoming data campaing.

Figure 1 .
Figure 1.Exploded view of the HERD detector.

Figure 3 .
Figure 3. Exploded view of one of the SCD layers.

Figure 4 .
Figure 4. Setup used during the 2022 campaign (a) and exploded view of one of the XY stations (b).

Figure 6 .
Figure 6.Correction of the cluster signal w.r.t. parameter.
(b) shows the resolution from the SCD setup (blue dots) compared with two different MC predictions (solid lines) and the performance of the AMS-02 Inner tracker (black dots).

Figure 7 .
Figure 7. Charge reconstructed using all 8 sensors (a) and comparison of obtained charge resolutions with MonteCarlo simulations (b).