31st RD50 Workshop on Radiation hard semiconductor devices for very high luminosity colliders
The photoconductivity and mobility of carriers was investigated in highly irradiated Si. The measurements were performed in microstrip samples at different temperature and different bias, up to high electric field regime. It was observed decrease the mobility with increase of fluence. The photoconductivity spectra demonstrated the main defects and its filling, and an increase of surface recombination with the increase of bias.
Phosphorus and Boron doped profiles of silcion wafers have been investigated to reconstruct active doping profile from Transmission line measurements. It is known that total doping profiles is accessible through Secondary Ion Mass spectrometry analysis. Here this methods enable us to determine electrical active doping profiles and see the change of this concentration after irradiation . Un-irradiaded and irradiated samples have been analysed and preliminary results will be presented.
Non-ionising energy loss of radiation produces point defects and defect cluster in silicon, which result in a significant performance degradation of detectors. In this contribution we present results of defect spectroscopy using TSC (Thermally Stimulated Current) measurements on silicon pad diodes irradiated by electrons of different energies, where significantly different ratios of point to cluster defects are expected. A new method based on SRH (Shockley-Read-Hall) statistics is introduced, which takes into account point and cluster defects and allows determining their dependence on particle energy and annealing conditions.
We present the ionizing energy depositions in a 300 $\mu$m thick silicon layer after fast neutron impact. With the Time-of-Flight (ToF) technique, the ionizing energy deposition spectra of recoil silicons and secondary charged particles were assigned to (quasi-) monoenergetic neutron energies in the range from 180 keV to hundreds of MeV. We show and interpret representative measured energy spectra and, study the competition of ionizing energy losses (IEL) and non-ionizing energy losses (NIEL) by separating the IEL of the recoil silicon from energy depositions by products of nuclear reactions.
The data give supplementary information to the results of a previous measurement by Sattler and are compared with different theoretical predictions (by Norgett-Torrens and Robinson, and Akkerman and Barak).
We present a study of electrically active radiation-induced defects formed in 4H-SiC epitaxial layers following irradiation with fast neutrons, 600 keV H+ and 2 MeV He++ ion implantations. The Schottky barrier diodes (SBD) were formed by evaporation of nickel through a metal mask with patterned quadratic apertures of 1 mm × 1 mm, while Ohmic contacts were formed by nickel sintering at 950 °C in Ar atmosphere on the back side of the silicon carbide substrate. The low leakage current of the order of pA (1 mm2 active area) have been measured even for the highest applied reverse biases of up to 500 V.
We have combined Laplace-Deep Level Transient Spectroscopy (Laplace-DLTS) measurements and density functional theory (DFT) calculations to study the carbon vacancy (VC) in n-type 4H-SiC. The Vc is known as the main “lifetime killer” in 4H-SiC. Using Laplace-DLTS we were able to provide sub-lattice resolved spectra of the broad DLTS peak at 290 K commonly observed in n-type 4H-SiC. This peak, previously ascribed to two closely related traps referenced as Z1/2, has two components with activation energies for electron emission of 0.58 eV and 0.66 eV. We compared these results with the acceptor levels of VC obtained by means of hybrid density functional supercell calculations. The calculations support the assignment of the Z1/2 signal to a superposition of emission peaks from double negatively charged VC defects. Taking into account the measured and calculated energy levels, the calculated relative stability of VC in hexagonal (h) and cubic (k) lattice sites, as well as the observed relative amplitude of the Laplace-DLTS peaks, we assign Z1 and Z2 to VC(h) and VC(k), respectively.
Nitrogen enriched wafers have shown some improvement after irradiation. NitroStrip is an RD50 project that has fabricated p-in-n wafers with four different materials, FZ, DOFZ, nitrogen enriched and MCz. Here we will show measurements of unirradiated and some irradiated detectors.
Acceptor removal has been studied on a set of p-type sensors irradiated with protons up to 7E15 neq/cm2. Two sets of diodes were used: thin epitaxial diodes with different resistivities (10, 50, 250 and 1000 Ohm cm) and high resistivity float zone diodes with different thicknesses (100, 150, 200 and 285 um).
CV, IV and TCT measurements were performed to extract the effective doping concentration of these devices. TCT collected charge versus voltage was used to evaluate the sensor's bulk space charge.
In an annealing study on a subset of the epitaxial sensors irradiated to 7.32E+14 neq/cm2, evidence of type inversion was observed on some of the devices.
Defect spectroscopy was conducted using TSC technique in order to study the correlation between BiOi concentration and acceptor removal.
All collected data is used to revise the fitting of the Neff vs fluence plots and extract acceptor removal rate parameters.
In the framework of monolithic active silicon sensors (MAPS), a new fabrication process that consists in bonding a CMOS wafer on a high resistivity silicon bulk which is used as the active sensor is investigated. In this way, it is possible to optimize both the read-out electronics and the sensing element. The electrical properties of the bonding interface are therefore important, since the charge generated by the passage of a particle must cross the interface. To characterize this charge transfer between the ‘‘CMOS wafer’’ and the high resistive bulk, Transient Current Technique (TCT) has been proposed. The characterization method consists first in the bonding of two silicon wafers with the same doping concentration. Then, Schottky diodes are fabricated on the bonded substrate and measured using TCT with laser injection. To avoid the use of complex setups for TCT analysis, a different charge injection method has been proposed. This new electrical injection TCT, where a voltage pulse is applied to an injection contact of a PN diode, enables the transient current to be generated ‘on chip’, further allowing online characterization of silicon detectors during irradiation tests and high energy physics experiments.
This presentation will report on latest Two-Photon Absorption-TCT measurement campaigns carried out within 2 on-going RD50-projects, namely measurements of diodes and HVCMOS devices. The presentation will then focus on recent measurements of proton irradiated LGAD detectors, showing a clear double junction after 1e14 neq/cm2. Some other systematics of TPA measurements encountered so far will be briefed.
We present a theoretical model to predict the plasma delay time in the signal generated by the Two Photon Absorption technique and a confrontation with real measurements.
https://indico.cern.ch/event/681859/
The worldwide fusion community has been developing understanding of plasma physics and
fusion plasmas due to the development of theoretical models, rapid advances in computer
simulation techniques and pioneering work in plasma diagnostics. A large fraction of such know-
how is derived from a complete suite of spectroscopic and particle diagnostics. It is thus crucial
to maintain the utilization of well-established plasma diagnostics techniques for basic diagnosis
and operation of burning plasma experiments in the near future. The main issues constraining or
even eliminating many conventional measurements are lack of port access, long-pulse
operations, high D-D and D-T neutron fluxes, gamma-induced noise and possibly, the presence
of high- magnetic fields. Conventional silicon detectors are used due to the availability of good
quality homogeneous material, and high charge carrier transport properties. Unfortunately, these
detectors can only withstand maximum neutron fluences in the range of 1013 to 1014
neutrons/cm2. The main concern in future uses of Si-detectors is, therefore, that their lifetimes
could be severely shortened by neutron damage since future sensors will have to withstand
fluences of 1015 up to 1017 neutrons/cm2. The fusion community is thus forced to invest in new
solutions that are compatible with CERN’s very-high-luminosity experiments, using new kinds
of radiation-hardened silicon sensors or semiconductor materials other than pure silicon like
Diamond, silicon-carbide, semi-insulating GaN, CdTe among others. These novel radiation-
hardened detectors and associated electronics have not been tested in a fusion experiment. In this
talk we will review key concepts of magnetically confined fusion plasmas, introduce detector
challenges for ITER, and discuss possible synergies with the high-energy physics community
(e.g. RD50), like testing detectors in present fusion experiments with the aim of providing
radiation-hardened detectors for ITER-DT phase as well as DEMO-like fusion reactors.
Avalanche Photo Diodes (APDs) produced by Radiation Monitoring Devices are examined as candidate timing detectors for HL-LHC applications.
These APDs are operated at 1.8 kV, resulting in a gain of up to 500.
The timing performance of the detectors is evaluated using a pulsed laser.
The effects of radiation damage on current, signal amplitude, noise, and timing of the APDs are evaluated using detectors irradiated with neutrons up to $\Phi_{eq} = 10^{15}$ cm$^{-2}$.
In view of the future HL-LHC upgrade, a variety of technologies are being considered for particle tracking. One of these technologies is Deep Diffused APDs (DD-APDs). Several DD-APDs were characterised through CV/IV and TCT measurements before and after neutron irradiation. The irradiation took place at the Jožef Stefan Institute (Ljubljana, Slovenia). The fluences to which the devices were exposed are 3E13, 6E13, 3E14 and 1E15 n/cm$^2$. The results obtained from these studies will be shown in this presentation.
We will discuss mainly gain and timing resolution of thin LGAD both before and after neutron irradiation.
In the past year, the CMS and ATLAS collaborations have defined the detector geometry of their respective timing layers. Even though both collaborations have selected UFSD in their baseline design, the requirements for the two experiments differ in key aspects such as with pixel size, radiation hardness, number of layers. In this contribution we review the requirements and challenges in the design and production of the sensors for CMS and ATLAS, outlining similarities and differences.
Over the last few years, Fondazione Bruno Kessler, in collaboration with the universities of Trento and Turin, has been involved in developing of Silicon sensors with low internal gain, the so-called Ultra-FAST Silicon Detectors (UFSD). Such a detector is based on the concept of Low-Gain Avalanche Detectors (LGAD), which are silicon detectors with an internal, low multiplication mechanism (gain ~ 10).
The first production batch (completed in 2016) was fabricated on 275 $\mu$m thick Silicon substrates. It was aimed at testing both the functionality and the reliability of the new proposed fabrication technology, showing excellent results in terms of gain and timing resolution.
In this contribution, we report on the last developments carried out at FBK on UFSD technology in the last year. A second pilot batch (completed in late spring 2017) has been produced on Silicon-to-Silicon wafers with a thickness of 50 $\mu$m, in order to improve the timing performance. In this production, we tested also new techniques to improve the radiation hardness of the devices. Two different dopant elements (Boron and Gallium) have been used to realize the multiplication junction, as well as carbon co-implant has been tested on some wafers.
We report on the preliminary electrical characterization of these devices, discussing the effect of doping and carbon co-implant on the detector noise and gain.
AC-LGAD are LGAD with simplified structure.
Basically all segmentation in the P+, N+ and oxide are eliminated and only the AC-coupled metal contact is segmented.
We have tested with laser, beta and alpha the first prototypes from CNM Run 10478 which were produced without optimization of the doping profile and simulated the response with TCAD and SPICE. A program going forward is described.
An in-deep study of a p-in-p LGAD prototype (dubbed as I-LGAD) is presented. Contrary to the conventional LGAD devices, currently proposed for the HL-LHC mip timing detectors, the I-LGAD has a non-segmented deep p-well ( the multiplication layer). Therefore, I-LGADs should idealy present a constant gain value over all the sensitive region of the device without gain drops between the signal collecting electrodes; in other words, I-LGADs should have a 100% fill factor by design. We have experimentally confirmed this feature on a strip-like segmented i-LGAD and compare it agaist a conventional strip-like LGAD and PIN devices.
The advantages TCAD based studies offer during the development of semiconductor sensors are multiple: they are predictive, they provide insight
and they capture and visualise theoretical knowledge.
In this talk I will report on Silvaco TCAD Atlas Device simulator, in particular about the modelisation of the most fundamental semiconductor
physics parameters in that tool, the band gap energy Eg.
The effect of its value on the reverse current and on the diode current scaling with temperature in forward and reverse bias will be discussed,
for boht unirradiated and irradiated bulks.
The goal of the report is to provide a solid basis for discussion on Silvaco TCAD tools and its predictions.
TCAD simulation has become a valuable tool in the design and understanding process of silicon sensors for high energy physics applications. The predictive power is limited by the uncertainties arising from different available semiconductor physics models but also from the simulation software itself.
In this talk we will present a comparison between Silvaco and Synopsys predictions for not-irradiated as well as irradiated structures. The irradiation models known as New Delhi and the latest version of the Perugia model will be used and compared in two dimensional simulations regarding observables like charge collection efficiency, current-voltage and capacity-voltage curves.
The impact of the active base with a low electric field on the bulk current in Si detectors irradiated to F ≥ 1x10^15 neq/cm2, i.e. to fluences of interest for the experiments at HL-LHC was studied. The simulated profiles of the electric field E(x) and of the bulk current densities j(x) showed that active base gives different contribution to the detector current operating as electrically neutral conductive base or electrically neutral depleted region, which depends on bias voltage and F. A comparison between the simulated and experimental j vs. F data at fluences upto ~10^17 neq/cm2 showed both j(F) dependences to be converted from linear to the square-root, which leads to saturation of the detector current associated with the impact of several mechanisms.
Due to internal charge multiplication, Low Gain Avalanche Detectors (LGAD) enable to produce rather thin (~ 50 µm) silicon sensors with a relatively low operating voltage. A typical P-type LGAD consists of n+-p-p–p+ layers with p-well formed by deep diffusion of boron into p– layer. Another way of p-well formation by using an epitaxial process may lead to lower sensitivity of the breakdown voltage on n+ layer thickness (M.Carulla at al., 29th RD50 Workshop). Possible boron compensation by radiation defects would cause gain degradation in P-type LGAD structures. This effect may be reduced by using N-type (p+n-n– n+) structure with a phosphorus doped n-well epitaxial layer. In this investigation we compared complementary P-type and N-type LGAD sensors.
The static and dynamic characteristics of the silicon LGAD structures have been simulated using a drift-diffusion approach implemented in the software package Synopsys TCAD Sentaurus. The total width of the diode was 50 µm. Only a 5 µm wide central part of the diode was simulated. The equal thickness of p-well and n-well layers (3.2 µm) leads to quite different breakdown voltage of complementary diodes due to the difference of the electron and hole ionization coefficients in silicon. The breakdown voltage is about two times lower in N-type diodes at the same current level. As a result, the electric field in the n–-layer is much lower than the carrier velocity saturation field. TCT calculations show that in such a case the carrier extraction is too long and the charge collection is much lower due to the recombination via deep levels in the irradiated diode. It is necessary to reduce the n-well thickness by 70 nm in order to obtain an equal charge collection in both types of diodes.
Particle detection was emulated through photo-excitation of a 2x50 µm^2 excess carrier domain at the center of the active volume in a diode. Shockley-Read-Hall (SRH), Auger and recombination through two deep levels was taken into account. The detailed investigation of the carrier and the electric field distribution dynamics during transient processes has been performed.
This talk aims to present extensive comparisons between measurements and numerical simulations about the UFSD2 recent production of Low-Gain Avalanche Detectors (LGAD) fabricated by FBK in Trento.
In particular, the methodological study on avalanche models and their calibration we introduced at the previous RD50 meeting, in Kraków, will be applied to the new batch of 50 micron detectors, which include four gain layer configurations: Boron, Boron with Carbon, Gallium and Gallium with Carbon.
By discussing the results we will show that an accurate simulation procedure, coupled to the use of modeling parameters coming from the experimental analysis, can provide a robust tool able to predict the behavior of irradiated/non-irradiated LGAD sensors in terms of gain and leakage current. This simulation is at the basis of designing next UFSD3 production at FBK.
TRACS is a C++11 based software that carries out an effective calculation of the induced current over irradiated and non-irradiated silicon microstrip and pad detectors. Its new achievements and developments will be presented.
Allpix Squared is a generic open-source simulation framework for pixel detectors. Its goal is to ease the implementation of detailed simulations for both single detectors and more complex setups such as beam telescopes. Predefined detector types can be automatically constructed from simple model files describing the detector parameters.
The simulation chain is arranged with the help of intuitive configuration files and an extensible system of modules, which implement the separate simulation steps. Currently available modules include realistic charge carrier deposition using the Geant4 toolkit, propagation of charge carriers in silicon either using a drift-diffusion model or a projection onto the sensor implants, and a simulation of the detector front-end electronics including noise, threshold, and digitization. Detailed electric field maps imported from TCAD simulations can be used to precisely model the drift behavior of the charge carriers, bringing a new level of realism to the simulation of particle detectors.
The history of every simulated object, including the Monte Carlo truth information of the original ionizing radiation, is preserved and can be stored to file, allowing for a direct comparison with reconstructed position information.
The framework is written in modern C++ and comes with fully documented source code as well as an extensive user manual. Its modular approach allows for a flexible set-up of the simulation and facilitates the reuse of independent, well-tested algorithms.
This contribution provides an overview of the framework and its different simulation modules, and presents first comparisons with test beam data.
A study of Silicon pixel sensors of size 50 um x 50 um fabricated at CNM using double-sided 3D technology is presented. Sensors are bump-bonded to the pixelated ROC4Sens read-out chip which exactly matches the sensor geometry. We analyze the response of two hybrid assemblies in a test beam of 5.6 GeV electrons. Results of charge collection, charge sharing, spatial resolution and efficiency are shown for non-irradiated sensors for different track incidence angles.
A new generation of CNM 3D pixel sensors with small pixel sizes of 50x50 and 25x100 $\mu$m$^{2}$ and reduced electrode distances are developed for the HL-LHC upgrade of the ATLAS pixel detector. For the first time, pixel detectors are irradiated and studied up to the unprecedented fluence of 2.5$\times10^{16} n_{eq}$/cm$^2$, i.e. for the full expected HL-LHC life time to explore the limits of the 3D technology. Since a readout chip with the desired pixel size is still under development by the RD53 collaboration, first prototype small-pitch pixel sensors were designed to be matched to the existing ATLAS IBL FE-I4 readout chip for testing. Irradiation campaigns with such pixel devices have been carried out at KIT (Karlsruhe) with a uniform irradiation of 23 MeV protons up to a fluence of 1$\times10^{16} n_{eq}$/cm$^2$, as well as at CERN-PS with a non-uniform irradiation of 23 GeV protons in several steps up to a peak fluence of 2.5$\times10^{16} n_{eq}$/cm$^2$. The hit efficiency has been measured in several beam tests at the CERN-SPS. The performance of these devices is significantly better than for the previous generation of 3D detectors or the current generation of planar silicon pixel detectors, demonstrating the excellent radiation hardness of the new 3D technology.
The talk will present the last batch of 3D n-on-p detectors produced by FBK-CMM designed for LHC Phase-2 Upgrade. The process was performed both on Si-Si and SOI wafers with 130 μm device thickness. Several kinds of pixel detectors, e.g., compatible with read-out chips such as FE-I4, CMS and RD53A, have been realized on wafers. We will report on the sensor technology and offer an overview of the first electrical characterization of the devices we have produced. This activity was supported by AIDA 2020 and INFN (ATLAS - CMS) - FBK Pixel R&D Collaboration.
P-type sensors have been selected as a base material for many future detectors due to the superior radiation hardness. The properties and performance of irradiated sensors has been intensively measured, but some aspects like the annealing behavior are still under evaluation.
In this contribution long term annealing studies at room temperature and 60°C on ATLAS12 p-type mini strip sensors, irradiated up to a fluence of 3e14 $\frac{n_{\mathrm{eq}}}{cm^2}$, will be presented. The charge collection, leakage current and effective doping concentration behavior have been interpreted with existing annealing models.
The effect of the annealing temperature has been evaluated in order to conclude some properties of the mechanisms behind the annealing of p-type silicon. The behavior of these sensors will also be compared to previous measurements with irradiated up to 2e15 $\frac{n_{\mathrm{eq}}}{cm^2}$, in which charge multiplication was observed. The results presented in this work contribute to the evaluation of sensors which are candidates to be used within HL-LHC Upgrade.
The innermost tracking detector of the ATLAS experiment consists of planar n-in-n silicon pixel sensors. Closest to the beam pipe lays the insertable b-layer (IBL). Its pixels are arranged in a pitch of 250um x 50um, with a rectangular shaped n-implant.
Based on this design six modified pixel designs have been developed in Dortmund.
The new pixel designs are arranged in structures of ten columns and have been placed besides structures with the standard design on one sensor. Because of a special guard ring design, each structure can be powered and investigated separately. Several of these sensors have been bump bonded to FE-I4 read-out chips. One of these modules has been irradiated with reactor neutrons up to a fluence of $5 \times 10^{15}\, \text{n}_{\text{eq}}\text{cm}^{-2}$, another has been irradiated with protons at CERN-PS IRRAD to a mean fluence of $6 \times 10^{15}\, \text{n}_{\text{eq}}\text{cm}^{-2}$.
This contribution presents the results of these irradiated devices, including important sensor characteristics, charge collection determined with radioactive sources and hit efficiency measurements, performed in laboratory and test beam. They are also compared with the results of non-irradiated devices.
Facing the high luminosity phase of the LHC (HL-LHC) to start operation around 2026, a major upgrade of the tracker system is in preparation for the ATLAS experiment. Thanks to the small material budget and their high charge collection efficiency after irradiation, thin planar pixel modules are the baseline option to instrument all layers of the pixel system beginning from the second layer.
To optimise the sensor layout towards the decreased pixel cell size of $50\times50\;\mu$m$^2$, TCAD device simulations are being performed before and after irradiation to investigate charge collection efficiency and electrical field properties. Two different common irradiation models will be compared regarding these observables.
In addition, sensors of 100-150$\;\mu$m thickness, interconnected to FE-I4 read-out chips featuring the previous generation pixel cell size of 50x250$\;\mu$m$^2$, are characterised at testbeams at the CERN-SPS. The performance of sensors irradiated up to a fluence of 1e16$\;$n$_\text{eq}$/cm$^2$ is compared in terms of charge collection and hit efficiency before and after annealing. This study aims to reproduce the effects of storage time at room temperature of high energy physics pixel detectors during maintenance periods.
The increased LHC luminosity as well as maximal particle energy require more precise data on the intensity of radiation field at the magnet coils of collider. For that the radiation sensors must be placed in the proximity of the coils, which minimizes the fraction of debris in beam loss monitoring. Silicon sensors were chosen as one of the candidates and the related R&D was started six years ago jointly by CERN BE-BI-BL group and RD39 collaboration.
The sensors developed at the Ioffe Institute and in situ tested at temperatures 1.9 and 4.3 K showed reproducible characteristics and gave new findings in the detector physics. Selected data will be presented to illustrate the differences of silicon detector radiation degradation at RT and the temperature of liquid He.
In 2016 PSI developed a pixel readout chip dedicated to sensor studies. It provides a full alanalogue signal without zero suppression, but needs a fast (~100ns) external trigger and has some speed limitations. The array is 160 x 155 pixels with a pitch of 50 Micrometer in both directions. A simple readout system based on the CMS pixel digital test board has been designed which was used in test beams at DESY. Another, non CMS specific and more powerful readout systems are in developement. The talk explaines the chip and its operation.
Please schedule Tue after 10:00 or Wed if possible.
The CiS research institute is engaged in developments of radiation detector technologies on several different fields. Current projects are dealing e.g. with active edge sensors, large area thinned sensors, sensor-chip packaging technologies and defect engineering.
An active edge sensor run is finished and evaluated. Three side wall doping methods have been tested in combination with two wafer thicknesses as well as with n- and p-substrates. The electrical measurements show the functionality of sensors with inactive edge widths down to 50µm.
For large area sensors, the need for smaller thicknesses can be approached by etching cavities to the sensors back side while guaranteeing stability on wafer level by thick frames at the edges. A single sided n-in-p pixel technology is currently transferred to 6” wafer size with membranes up to 4x4cm² and target thicknesses of 50, 100 and 150µm. The technology is furthermore extended to a double sided process which will enable double sided thin sensors.
I will present test results of OVERMOS1, a MAPS CMOS detector based on high resistivity substrate. Following a short description of the main features of OVERMOS1 and differences with respect to previous OVERMOS incarnation, I will describe experimental results of test structures, both for standard and neutron irradiated devices, and comparisons with TCAD simulation results.
Active silicon sensors produced in CMOS technology are commonly manufactured on substrates of intermediate resistivity and are usually operated under partial depletion. Irradiation and consequent effective acceptor removal changes the depletion depth and therefore the amount of collected charge.
We will present a study of proton irradiation effects in the fluence range of 4e14 - 4e15 neq/cm2 on passive test structures on samples produced by AMS. Wafers of different initial resistivity ranging between 20 and 1000 Ohm-cm were tested. The depletion depth was estimated by Edge-TCT and MIPs from a Sr90 source were used to measure the collected charge.
The measurements will be compared to an earlier neutron irradiation study and the effects on charge collection will be discussed.
High Energy Particle Physics experiments at the LHC use hybrid silicon detectors, in both pixel and strip geometry, for their inner trackers. These detectors have proven to be very reliable and performant. Nevertheless, there is great interest in the development of depleted CMOS silicon detectors, which
could achieve similar performances at lower cost of production and complexity.
Studies of novel depleted CMOS prototypes, denoted as LF-CPIX, fabricated by Lfoundry are presented. The devices were measured before and after irradiation to a fluence of 10^15 neq/cm2 at the Birmingham MC40 Cyclotron facility.
The ATLAS experiment is planning a major upgrade of its tracking detectors during the Phase-II LHC shut down, to better take advantage of the increased luminosity that will be provided by the HL-LHC. New depleted CMOS sensors are being developed for this upgrade. Preliminary results of the evaluation of TowerJazz 180nm CMOS sensors using the Edge-Transient Current Technique (Edge-TCT) are presented. The suitability of various pixel sizes is assessed including 50um and 20um-sized structures. Charge sharing is quantified using multi-pixel measurements also at various pixel sizes. Radiation damage effects due to neutrons are also investigated.
Depleted Monolithic Active Pixel Sensors (DMAPS) built with High Voltage CMOS (HV-CMOS) technology are investigated as an option to cover large areas in the outermost layers of the future pixel detector of the ATLAS Inner Tracker (ITk) at HL-LHC.
The H35Demo is a large area HV-CMOS demonstrator prototype chip developed for the ITk which features a large fill factor layout with 25x250 um^2 pixel cells.
It was designed by KIT, IFAE and University of Liverpool, and produced in AMS 350 nm CMOS technology with different resistivities. The chip consists of four pixel matrices: two matrices including digital electronics in the periphery and thus designed to be operated as monolithic detectors, and two matrices meant to be capacitive coupled to ATLAS FE-I4 chips.
H35Demo chips have been irradiated with reactor neutrons at JSI up to 2e15 n_{eq}/cm^2 and with 23 MeV protons at KIT up to 1e15 n_{eq}/cm^2. The performance in terms of hit efficiency of the monolithic matrices have been investigated in two beam test campaigns, at Fermilab with a 120 GeV proton beam and at CERN SpS with 180 GeV pions. For particle tracking the FE-I4 telescope provided by University of Geneva was used. Results obtained for not irradiated and irradiated chips will be presented.
Fluorescence measurements are one option to investigate silicon sensor properties [1]. The idea is to deploy an X-ray source pointing on a target material and leading to the emission of monochromatic X-rays. This allows to measure charge spectra and the calibration of sensors. In the presentation details on a setup recently installed will be shown and details on the commissioning given. Second, measurement results of monolithic active pixel sensors will be presented. The TowerJazz investigator chip [2] has been developed in a technology which is one option for the Phase-2 upgrade of the pixel detector of the ATLAS experiment. Measurements to calibrate it and determine its gain and energy resolution will be shown for different sensor types.
[1] L.-D. Pohl et al., Obtaining spectroscopic information with the ATLAS FE-I4 pixel readout chip, Nuclear Instrum. Meth. A788 (2015), 49.
[2] H. Pernegger et al., First tests of a novel radiation hard cmos sensor process for depleted monolithic active pixel sensors, Journal of Instrumentation 12 (2017) P06008.
As the HV-CMOS technology is emerging as a prime candidate for many future experiments in particle physics, it is a priority for the RD50 collaboration to develop and study particle detectors in this technology. In this context, the collaboration has started a new effort to design and manufacture dedicated test structures and matrices of HV-CMOS and HV-MAPS pixels. Two HV-CMOS submissions are foreseen at the moment with the collaboration. The first one is an MPW to test the technical design aspects and basic functionality of the sensing diodes and readout electronics. In contrast, the second submission is a large area demonstrator to probe into more advanced features, such as improving the sensor time resolution, implementing novel sensor cross-sections and studying pre-stitching options. The MPW is mandatory to guarantee the success of the large area and thus more expensive demonstrator. Present measurement plans with both submissions include irradiations with a wide range of fluences. Both submissions are in the 150 nm HV-CMOS technology node from LFoundry, as this offers several attractive features that can be beneficial for particle detectors.
This contribution describes the status of the two HV-CMOS submissions mentioned above, paying special attention to the very recent MPW. In particular, I will review the main design aspects of the prototype as well as the TCAD simulations done in parallel to optimize the sensor layout. I will also give details about the timeline of this project and how we will proceed with measurements. More information will be given at the workshop.