DMAPS Monopix developments in large and small electrode designs
Christian Bespin, Marlon Barbero, Pierre Barrillon, Ivan Berdalovic, Siddharth Bhat, Patrick Breugnon, Ivan Caicedo, Roberto Cardella, Zongde Chen, Yavuz Degerli, Jochen Dingfelder, Leyre Flores Sanz de Acedo, Stephanie Godiot, Fabrice Guilloux, Toko Hirono, Tomasz Hemperek, Fabian Hügging, Hans Krüger, Thanushan Kugathasan, Cesar Augusto Marin Tobon, Konstantinos Moustakas, Patrick Pangaud, Heinz Pernegger, Francesco Piro, Petra Riedler, Alexandre Rozanov, Piotr Rymaszewski, Philippe Schwemling, Walter Snoeys, Maxence Vandenbroucke, Tianyang Wang, Norbert Wermes, Sinuo Zhang
DDMAPS Monopix developments in large and small electrode designs
C. Bespin a, ∗ , M. Barbero b , P. Barrillon b , I. Berdalovic d , S. Bhat b , P. Breugnon b , I. Caicedo a , R. Cardella d , Z. Chen b ,Y. Degerli c , J. Dingfelder a , L. Flores Sanz de Acedo d , S. Godiot b , F. Guilloux c , T. Hirono a , T. Hemperek a , F.H¨ugging a , H. Kr¨uger a , T. Kugathasan d , C. Marin Tobon d , K. Moustakas a , P. Pangaud b , H. Pernegger d , F. Piro d , P.Riedler d , A. Rozanov b , P. Rymaszewski a , P. Schwemling c , W. Snoeys d , M. Vandenbroucke c , T. Wang a , N. Wermes a ,S. Zhang a a Universit¨at Bonn, Physikalisches Institut, Nußallee 12, 53115 Bonn, Germany b Aix Marseille University, CNRS / IN2P3, CPPM, 163 Avenue de Luminy, 13009 Marseille, France c IRFU, CEA-Saclay, 91191 Gif-sur-Yvette, France d CERN, Espl. des Particules 1, 1211 Meyrin, Switzerland
Abstract
LF-Monopix1 and TJ-Monopix1 are depleted monolithic active pixel sensors (DMAPS) in 150 nm LFoundryand 180 nm TowerJazz CMOS technologies respectively. They are designed for usage in high-rate and high-radiationenvironments such as the ATLAS Inner Tracker at the High-Luminosity Large Hadron Collider (HL-LHC). Both chipsare read out using a column-drain readout architecture. LF-Monopix1 follows a design with large charge collectionelectrode where readout electronics are placed inside. Generally, this o ff ers a homogeneous electrical field in thesensor and short drift distances. TJ-Monopix1 employs a small charge collection electrode with readout electronicsseparated from the electrode and an additional n-type implant to achieve full depletion of the sensitive volume. Thisapproach o ff ers a low sensor capacitance and therefore low noise and is typically implemented with small pixel size.Both detectors have been characterized before and after irradiation using lab tests and particle beams. Keywords:
Depleted monolithic active pixel sensor, pixel detectors, monolithic pixels, MAPS, DMAPS, radiationhardness
1. Introduction
Monolithic pixel sensors designed in commercialCMOS technologies were first proposed in the 1990sfor charged particle tracking. They o ff er low materialbudget, low cost and easy module assembly comparedto well established hybrid pixel detectors [1]. How-ever, monolithic active pixel sensors, for instance theALPIDE chip for the ALICE experiment at LHC [2]are not suitable for high-radiation and high-rate environ-ments as their charge collection mechanism is mainly bydi ff usion. In order to cope with conditions such as thoseexpected at the future HL-LHC, charge collection hasto be done by drift in a depleted sensor volume, leadingto depleted monolithic active pixel sensors (DMAPS).Large-scale prototype chips employing high resistivitysubstrates and high bias voltages have been designed invarious CMOS technologies with integrated fast readoutelectronics on the sensor substrate [3, 4, 5]. ∗ Corresponding author
Email address: [email protected] (C. Bespin)
Two of these devices, LF-Monopix1 and TJ-Monopix1, fabricated in 150 nm LFoundry and 180 nmTowerJazz technology, respectively, have been char-acterized with respect to the requirements of the AT-LAS ITk outer pixel layer. Detectors in this environ-ment have to withstand up to 10 n eq / cm non-ionizingenergy loss (NIEL) and 50 Mrad total ionizing dose(TID) radiation damage while maintaining timing re-quirements of 25 ns.
2. Design
For a planar arrangement the depletion depth d of thesensor volume depends on the resistivity ρ of the mate-rial and the applied bias voltage V : d ∝ (cid:112) ρ V Thus, to achieve full depletion of the sensing volume thedesign of DMAPS employs both highly resistive ma-terials as well as high voltage features o ff ered by the Preprint submitted to Elsevier July 10, 2020 a r X i v : . [ phy s i c s . i n s - d e t ] J u l oundries. In the following, the di ff erent sensor ap-proaches of the reported pixel sensors are described. LF-Monopix1 is designed in 150 nm CMOS tech-nology and uses a so-called large collection electrodethat encapsulates the pixel electronics. The substratematerial has a resistivity larger than 2 k Ω cm and canwithstand bias voltages larger than 200 V while o ff eringthicknesses down to 100 µ m. Due to the large n-typecollection electrode, the sensing volume can be fullyand homogeneously depleted while keeping drift pathsin the volume short, and therefore creating a radiation-hard design. This comes at a cost of the comparablylarge sensor capacitance of LF-Monopix1 of an esti-mated ∼
400 fF based on measurements of test struc-tures [6] and a power consumption of ∼ µ W / pixel fora 250 µ m × µ m pixel with full digital logic inside. As PWELL PWELLNWELLNWELL NWELLDEEP PWELLPWELL PWELL
DEEPNWELL DEEPNWELLVERY DEEP NWELLP SUBSTRATE
Figure 1: Schematic cross section of the LFoundry 150 nm process.Charge collection is done by an n-type implantation that encapsulatesthe readout electronics. depicted in figure 1, the in-pixel electronics make use ofthe nested well structure o ff ered by the foundry. In par-ticular, the deep p-well implant allows for full CMOSlogic, isolating the n-wells of PMOS transistors fromthe charge collection node. TJ-Monopix1 is designed in 180 nm TowerJazz tech-nology and makes use of a small collection electrodewhich is separated from the pixel electronics. It is basedon the ALPIDE chip [2] designed for the ALICE exper-iment at LHC. Figure 2 shows a schematic cross sec-tion of this design, where charges are created in a 25 µ mthick p-type epitaxial layer with a resistivity of morethan 1 k Ω cm.This approach o ff ers a low sensor capacitanceof ∼ µ W / pixel [7], but is typically not as radiationhard as a large collection electrode design due to longerdrift distances and inhomogeneous drifting fields. Inorder to achieve full depletion and radiation hardness P + SUBSTRATEP - EPITAXIAL LAYERCOLLECTION N-WELLLOW DOSE N-TYPE IMPLANTDEEP PWELLPWELL PWELLNWELL DEEP PWELLPWELL PWELLNWELL (a) Sensor design with full deep p-well coverage. P + SUBSTRATEP - EPITAXIAL LAYERCOLLECTION N-WELLLOW DOSE N-TYPE IMPLANTDEEP PWELLPWELL PWELLNWELL DEEP PWELLPWELL PWELLNWELL (b) Sensor design with reduced deep p-well coverage.
Figure 2: Schematic cross section of the TowerJazz 180 nm process.Charge collection is done by an n-type implantation which is sepa-rated from the readout electronics. CMOS transistors are shielded bya deep p-type implant to prevent the existence of a competing collec-tion electrode from the n-wells of the PMOS transistors. an additional low dose n-type layer has been implantedthat extends the depletion region under the electronicson the pixel edge [8]. A deep p-well shields the n-wellof PMOS transistors and prevent the existence of a com-peting collection electrode and allows for full CMOSlogic of the electronics. There are two variants of thisdeep p-well implant integrated in the chip: one that fullycovers the in-pixel electronics and one that covers themonly partially, which changes the shape of the electricalfield in the sensor close to the charge collection elec-trode. Results presented in this paper are obtained usingpixels with reduced deep p-well coverage.Three of the four available pixel flavors use slightlyvarying versions of a PMOS reset circuit that is DCcoupled to the front-end. One exemplary schematic isshown in figure 3a. The fourth flavor utilizes an addi-tional high voltage supply on a diode to reset the chargecollection node and uses AC coupling to the front-endcircuit (3b).
Both LF-Monopix1 and TJ-Monopix1 make use ofa column drain readout architecture, similar to the oneused in the FE-I3 ATLAS pixel readout chip [9]. Thedesign is fully monolithic with all the dedicated readoutelectronics integrated into the pixel cell. Charge createdin the sensing volume and collected by the collection2 resetIreset p-well front end (a) Front-end design with PMOS reset circuit and DCcoupling. p-well front endHV V baseline I baseline (b) Front-end design with additional high voltage supplyand AC coupling Figure 3: Schematics of the pixel reset electronics and AC and DCcoupling options for TJ-Monopix1. electrode is amplified and converted into a voltage pulsethat is compared to an adjustable threshold. The dig-ital output of the discriminator defines a leading edgeand trailing edge pulse determining the time of arrivaland the time over threshold (ToT) – the total length ofthe pulse. Both are measured in units of clock cyclesof 40 MHz.
3. Measurements
Various measurements on both detectors were per-formed before and after di ff erent types of irradiation.NIEL damage studies were performed with chips irradi-ated with neutrons at the TRIGA Mark II Research Re-actor of the JoÅef Stefan Institute in Ljubljana [10] upto 10 n eq / cm . The TID tolerance has been tested withX-ray irradiated chips up to 100 Mrad. E ffi ciency re-sults were obtained using a 2 . LF-Monopix1 has been characterized in regard to thepixel electronics’ performance such as gain and noise. These measurements were conducted with untuned sen-sors cooled down to approximately − ◦ C. Additionalresults are given on the hit detection e ffi ciency measuredduring a test beam campaign. Results for irradiated sen-sors have been obtained using the same chip as in [3].The gain of LF-Monopix1 has been measured asvoltage response of the amplifier to a Fe signal.Gain distributions before and after neutron irradiationto 10 n eq / cm are depicted in figure 4 (top) with noobservable change due to radiation. Injecting an exter- Figure 4: Gain and noise distribution of LF-Monopix1 before andafter neutron irradiation to 10 n eq / cm . nal voltage pulse into each pixel allows for the studyof the equivalent noise charge (ENC) of the chip. Theprobability that a pixel responds to a charge can be de-scribed by an S-curve function with increasing probabil-ity at higher charge. Noise behaviour can be extractedfrom the steepness of the curve and the resulting dis-tribution is shown in figure 4 (bottom) before and afterirradiation. The threshold for the two untuned samplesare 7200 e − and 6900 e − respectively. An increase ofENC of about 150 e − can be seen after 10 n eq / cm ,but the chip can still be operated at a reasonably lowthreshold to achieve high detection e ffi ciency.Hit detection e ffi ciency for LF-Monopix1 has beenreported in [3] with average values of 99 . . n eq / cm neutron irradiatedchip at thresholds of 1800 e − and 1600 e − , respectively.Corresponding e ffi ciency maps are shown in figure 5,where ine ffi cient regions correspond to masked pixelsthat were excluded from the calculation.3 .0 0.5 0.0 0.5 1.03210123 V e r t i c a l p o s i t i o n ( mm ) (a) (b) H i t e ff i c i e n c y ( % ) Horizontal position (mm)
Figure 5: Hit detection e ffi ciency of an (a) unirradiated and(b) 10 n eq / cm neutron irradiated LF-Monopix1 chip. From [3]. Similar measurements as for the large collectionelectrode prototype have been performed with the TJ-Monopix1 chip, including measurements of thresholdand noise, hit detection e ffi ciency as well as TID toler-ance. Results have been obtained using a sensor geom-etry with reduced deep p-well coverage (RDPW) andAC-coupling with additional high voltage applied to thecollection electrode to enhance the electrical field in thesensor. Injecting fixed charge pulses into the pixel electron-ics allows for the measurement of threshold and noise ofthe detector. The mean equivalent noise charge for TJ-Monopix has been reported to be 16 e − before and 23 e − after neutron irradiation [12]. A visual representationis shown in figure 6a. It can be seen that both distri-butions show an asymmetric tail towards larger ENCvalues which is an indication for random telegraph sig-nal (RTS) noise [13], which will be improved in a fu-ture design. While the noise can be extracted from thesteepness of the function describing the pixel response,the threshold value is defined as the amount of charge,where the probability is 50 %. The resulting distributionfor the threshold after converting the result into elec-trons is shown in figure 6b. The observed threshold in-creases from 348 e − to 569 e − after neutron irradiationto 10 n eq / cm while the spread of the distribution in-creases by 100 % from 33 e − to 66 e − [12].Due to the increased noise and large threshold disper-sion after irradiation, operation at low threshold levelsis not possible. This can influence measurements on thehit detection e ffi ciency in case signal charge is lost dueto trapping and the remaining detected charge is lowerthan the threshold. Changes on the front-end electron-ics have been studied in a small test chip called Mini- (a)(b) Figure 6: (a) ENC and (b) threshold distributions of TJ-Monopix1before and after neutron irradiation. From [12].
MALTA [14] that allow for lower operational thresholdsas in TJ-Monopix1. ffi ciency The chips have been operated at the threshold settingsreported in section 3.2.1. As for LF-Monopix1, noisypixels were masked to keep the noise occupancy belowthe ATLAS ITk requirements. Figure 7 shows the re-sulting e ffi ciency maps for an unirradiated and a neu-tron irradiated (10 n eq / cm ) chip with reduced deepp-well coverage. While the e ffi ciency before irradiationis 97 . . ffi ciency of an unirradiated TJ-Monopix1 chip with reduced deep p-well coverage, aloss of e ffi ciency in the corners and on vertical edges ofa 2 x 2 pixel submatrix can be observed. Studying thedesign layout of such a submatrix shows a coincidenceof ine ffi cient regions with large active areas of the ma-trix. Due to a limited spatial resolution of the setup fora cooled irradiated detector during the test beam cam-4 a)(b) Figure 7: Hit detection e ffi ciency of TJ-Monopix1 (RDPW) (a) beforeand (b) after neutron irradiation to 10 n eq / cm . Ine ffi cient regionscorrespond to masked pixels that were excluded from the e ffi ciencycalculation. paign it was not possible to investigate in-pixel e ff ectsfor the e ffi ciency loss after irradiation with this detector.Further studies of in-pixel e ffi ciencies on devices imple-menting the same sensor geometry [5] as well as TCADsimulations [15] have shown a weak lateral electricalfield under the deep p-well areas where charge is lost,especially after irradiation. There are already two mod-ifications successfully tested to mitigate this e ff ect [14]. The TID tolerance of TJ-Monopix1 has been mea-sured in an X-ray irradiation campaign at Bonn Univer-sity. Chips have been irradiated using an X-ray tubeat a dose rate of 0 . / h while being cooled downto approximately − ◦ C. Measurements were conducteddirectly after every irradiation step. Figure 9a shows thenormalized gain of TJ-Monopix1 in dependence of TID.After initially stable behavior a large drop of about 80 %can be observed between 0 . ff er-ent front-end design with AC coupling and additionalhigh voltage applied to the collection electrode indi-cates a potentially better radiation tolerance up to thetested 0 . (a) (b) Figure 8: In-pixel e ffi ciency of an unirradiated 2 x 2 pixel submatrixof TJ-Monopix1 and the layout of such a submatrix showing the rele-vant areas of deep p-well, n-well and active areas. Ine ffi cient regionscorrelate with large active areas. in figure 9a. Due to time constraints a measurement tohigher doses was not possible and further studies areneeded to investigate this observation.
4. Conclusions
Large-scale DMAPS prototypes have been charac-terized for usage at high-rate and high radiation envi-ronments like the ATLAS experiment at HL-LHC withpromising results. While LF-Monopix1 shows radi-ation hardness up to at least 10 n eq / cm and con-sistently high e ffi ciency, the small electrode prototypeTJ-Monopix1 shows room for improvements concern-ing its radiation hardness. Following the results fromboth demonstrator chips, these will be improved withlarger matrices in both technologies. The matrix ofLF-Monopix2 is increased to 1 cm × µ m × µ m and optimizingthe analog front-end. It was succesfully submitted inspring 2020. TJ-Monopix2 will be produced as full sizechip of 2 cm × µ m × µ m pixels and iscurrently under design. Process modifications that weretested in a dedicated test chip MiniMALTA will be im-plemented. These modifications shows enhanced chargecollection, especially after radiation damage and loweroperational threshold [14]. The analog front end of TJ-Monopix2 will be improved to reduce noise, increaseTID tolerance and allow for lower threshold values aswell as adding a threshold tuning on pixel level to re-duce threshold dispersion between pixels. Acknowledgments
This project has received funding from the DeutscheForschungsgemeinschaft DFG, grant WE 976 / Dose / krad0.00.20.40.60.81.01.2 F e P e a k ( n o r m a li z e d ) (a) Normalized gain of TJ-Monopix1 depending on TIDusing a DC coupled frontend with high voltage appliedto the p-well and substrate. Dose / krad0.00.20.40.60.81.01.2 F e P e a k ( n o r m a li z e d ) (b) Normalized gain of TJ-Monopix1 depending on TIDusing an AC coupled front end with additional high volt-age on the n-well. Figure 9: Normalized gain of TJ-Monopix1 versus total ionizing dosemeasured with an Fe source. Due to time constraints the sampleshown in figure 9b was only irradiated to 0 . References [1] R. Turchetta, J. Berst, B. Casadei, G. Claus, C. Colledani,W. Dulinski, Y. Hu, D. Husson, J. L. Normand, J. Riester,G. Deptuch, U. Goerlach, S. Higueret, M. Winter, A mono-lithic active pixel sensor for charged particle tracking and imag-ing using standard VLSI CMOS technology, Nuclear Instru-ments and Methods in Physics Research Section A: Accel-erators, Spectrometers, Detectors and Associated Equipment458 (3) (2001) 677 – 689. doi:https://doi.org/10.1016/S0168-9002(00)00893-7 .[2] M. Mager, ALPIDE, the Monolithic Active Pixel Sensor forthe ALICE ITS upgrade, Nuclear Instruments and Methods inPhysics Research Section A: Accelerators, Spectrometers, De-tectors and Associated Equipment 824 (2016) 434 – 438. doi:https://doi.org/10.1016/j.nima.2015.09.057 .[3] M. Barbero, P. Barrillon, C. Bespin, S. Bhat, P. Breugnon,I. Caicedo, Z. Chen, Y. Degerli, J. Dingfelder, S. Godiot,F. Guilloux, T. Hemperek, T. Hirono, F. Hgging, H. Krger,K. Moustakas, A. Ouraou, P. Pangaud, I. Peric, D.-L. Pohl,P. Rymaszewski, P. Schwemling, M. Vandenbroucke, T. Wang,N. Wermes, Radiation hard DMAPS pixel sensors in 150 nm CMOS technology for operation at LHC (2019). arXiv:1911.01119 .[4] M. Kiehn, A. Herkert, A. Weber, A. Schning, A. Miucci,A. Fehr, C. Blattgerste, C. Grzesik, D. Sultan, D. Immig, D. For-shaw, D. Ferrere, D. Wiedner, E. Za ff aroni, E. V. Figueras,F. Ehrler, F. Lanni, G. Iacobucci, H. Augustin, H. Liu, H. Chen,I. Peric, J. Anders, J. Hammerich, J. Krger, J. Vossebeld,K. Chen, L. Xu, L. Noehte, L. Huth, M. Vicente, M. Benoit,M. Weber, M. Prathapan, N. Berger, S. Dittmeier, S. Sevilla,S. Tang, T. Golling, W. Wu, W. Wong, Performance of the AT-LASPix1 pixel sensor prototype in ams aH18 CMOS technologyfor the ATLAS ITk upgrade, Journal of Instrumentation 14 (08)(2019) C08013–C08013. doi:10.1088/1748-0221/14/08/c08013 .[5] R. Cardella, I. A. Tortajada, I. Berdalovic, C. Bespin, F. Dachs,V. Dao, L. F. S. de Acedo, F. Piro, T. Hemperek, T. Hirono,B. Hiti, T. Kugathasan, C. M. Tobon, K. Moustakas, H. Perneg-ger, P. Riedler, E. J. Schioppa, A. Sharma, L. S. Argemi,W. Snoeys, C. S. Sanchez, T. Wang, N. Wermes, MALTA: anasynchronous readout CMOS monolithic pixel detector for theATLAS High-Luminosity upgrade, Journal of Instrumentation14 (06) (2019) C06019–C06019. doi:10.1088/1748-0221/14/06/c06019 .[6] M. Loepke, Kapazittsmessung von CMOS-Pixelstrukturen,Bachelor’s thesis, University of Bonn (August 2017).[7] K. Moustakas, M. Barbero, I. Berdalovic, C. Bespin,P. Breugnon, I. Caicedo, R. Cardella, Y. Degerli, N. E. Plaja,S. Godiot, F. Guilloux, T. Hemperek, T. Hirono, H. Krger, T. Ku-gathasan, C. M. Tobon, P. Pangaud, H. Pernegger, P. Riedler,P. Rymaszewski, E. Schioppa, W. Snoeys, M. Vandenbroucke,T. Wang, N. Wermes, CMOS monolithic pixel sensors based onthe column-drain architecture for the HL-LHC upgrade, NuclearInstruments and Methods in Physics Research Section A: Ac-celerators, Spectrometers, Detectors and Associated Equipment936 (2019) 604 – 607. doi:https://doi.org/10.1016/j.nima.2018.09.100 .[8] W. Snoeys, G. A. Rinella, H. Hillemanns, T. Kugathasan,M. Mager, L. Musa, P. Riedler, F. Reidt, J. V. Hoorne, A. Fenig-stein, T. Leitner, A process modification for CMOS mono-lithic active pixel sensors for enhanced depletion, timing perfor-mance and radiation tolerance, Nuclear Instruments and Meth-ods in Physics Research Section A: Accelerators, Spectrome-ters, Detectors and Associated Equipment 871 (2017) 90 – 96. doi:https://doi.org/10.1016/j.nima.2017.07.046 .[9] I. Peri, L. Blanquart, G. Comes, P. Denes, K. Einsweiler, P. Fis-cher, E. Mandelli, G. Meddeler, The FEI3 readout chip forthe ATLAS pixel detector, Nuclear Instruments and Methods inPhysics Research Section A: Accelerators, Spectrometers, De-tectors and Associated Equipment 565 (1) (2006) 178 – 187,proceedings of the International Workshop on SemiconductorPixel Detectors for Particles and Imaging. doi:https://doi.org/10.1016/j.nima.2006.05.032 .[10] L. Snoj, G. Åerovnik, A. Trkov, Computational analysis of ir-radiation facilities at the JSI TRIGA reactor, Applied Radi-ation and Isotopes 70 (3) (2012) 483 – 488. doi:https://doi.org/10.1016/j.apradiso.2011.11.042 .[11] N. Heurich, F. Frommberger, W. Hillert, P. Hnisch, The NewExternal Beamline for Detector Tests at ELSA, in: Proc. of In-ternational Particle Accelerator Conference (IPAC’16), Busan,Korea, May 8-13, 2016, no. 7 in International Particle Accelera-tor Conference, JACoW, Geneva, Switzerland, 2016, pp. 4088–4090. doi:doi:10.18429/JACoW-IPAC2016-THPOY002 .[12] I. Caicedo, M. Barbero, P. Barrillon, I. Berdalovic, S. Bhat,C. Bespin, P. Breugnon, R. Cardella, Z. Chen, Y. Degerli,J. Dingfelder, S. Godiot, F. Guilloux, T. Hirono, T. Hemperek, . Hgging, H. Krger, T. Kugathasan, K. Moustakas, P. Pan-gaud, H. Pernegger, D.-L. Pohl, P. Riedler, A. Rozanov, P. Ry-maszewski, P. Schwemling, W. Snoeys, M. Vandenbroucke,T. Wang, N. Wermes, The monopix chips: depleted monolithicactive pixel sensors with a column-drain read-out architecturefor the ATLAS inner tracker upgrade, Journal of Instrumentation14 (06) (2019) C06006–C06006. doi:10.1088/1748-0221/14/06/c06006 .[13] K. S. Ralls, W. J. Skocpol, L. D. Jackel, R. E. Howard, L. A. Fet-ter, R. W. Epworth, D. M. Tennant, Discrete resistance switchingin submicrometer silicon inversion layers: Individual interfacetraps and low-frequency ( f ?) noise, Phys. Rev. Lett. 52 (1984)228–231. doi:10.1103/PhysRevLett.52.228 .[14] M. Dyndal, V. Dao, P. Allport, I. A. Tortajada, M. Barbero,S. Bhat, D. Bortoletto, I. Berdalovic, C. Bespin, C. Buttar,I. Caicedo, R. Cardella, F. Dachs, Y. Degerli, H. Denizli, L. F. S.de Acedo, P. Freeman, L. Gonella, A. Habib, T. Hemperek,T. Hirono, B. Hiti, T. Kugathasan, I. Mandi´c, D. Maneuski,M. Mikuˇz, K. Moustakas, M. Munker, K. Oyulmaz, P. Pangaud,H. Pernegger, F. Piro, P. Riedler, H. Sandaker, E. Schioppa,P. Schwemling, A. Sharma, L. S. Argemi, C. S. Sanchez,W. Snoeys, T. Suligoj, T. Wang, N. Wermes, S. Worm, Mini-MALTA: radiation hard pixel designs for small-electrode mono-lithic CMOS sensors for the high luminosity LHC, Journal of In-strumentation 15 (02) (2020) P02005–P02005. doi:10.1088/1748-0221/15/02/p02005 .[15] M. Munker, M. Benoit, D. Dannheim, A. Fenigstein, T. Kugath-asan, T. Leitner, H. Pernegger, P. Riedler, W. Snoeys, Simula-tions of CMOS pixel sensors with a small collection electrode,improved for a faster charge collection and increased radiationtolerance, Journal of Instrumentation 14 (05) (2019) C05013–C05013. doi:10.1088/1748-0221/14/05/c05013 ..