Reliability and quality assurance methods for silicon detectors and their application to the development of detectors with alumina thin films

Reliability and quality assurance methods for silicon detectors and their application

to the development of detectors with alumina thin films

Tatyana Arsenovich

Division of Particle Physics and Astrophysics Department of Physics

Faculty of Science University of Helsinki

and

Helsinki Institute of Physics P.O. Box 64, FI- 00014, Helsinki, Finland

ACADEMIC DISSERTATION

To be presented for public criticism,

with the permission of the Faculty of Science of the University of Helsinki, on 11.11.2020 at 12:00 o’clock via remote connection.

Helsinki 2020

i Internal Report HIP-2020-03

ISBN 978-951-51-6742-2 (printed version) ISBN 978-951-51-6743-9 (electronic version) http://ethesis.helsinki.fi

Unigrafia Helsinki 2020

ii Custos:

Prof. Kenneth Österberg Helsinki Institute of Physics Finland

Supervisors:

Prof. Paula Eerola Prof. Kenneth Österberg Dr. Jaakko Härkönen Prof. Panja Luukka

Department of Physics, University of Helsinki and Helsinki Institute of Physics

Finland

Reviewers:

Prof. Bjarne Stugu

Department of Physics and Technology University of Bergen

Norway

Prof. Ari Jokinen Department of Physics University of Jyväskylä Finland

Opponent:

Prof. Richard Brenner

Department of Physics and Astronomy Uppsala University

Sweden

iii

Abstract

The Large Hadron Collider (LHC) accelerator at CERN will be updated into High- Luminosity Large Hadron Collider (HL-LHC) between 2025-2027, and as a result, all of the LHC experiments have to be upgraded to meet the goals set for high-quality physics data taking. Also the Compact Muon Solenoid (CMS) Tracker will undergo several planned upgrades aimed to improve the characteristics of its detectors without negative impact on physics potential. In the HL-LHC, the level of radiation will increase significantly, thus, the radiation hardness of the detectors should be improved while at the same time also improving their capability of handling the higher amounts of data. The candidate materials and technologies for the development of the detectors need to be reviewed taking the requirements of the HL-LHC data taking into account. The quality assurance of these detectors is of utmost importance to identify possible failures as soon as possible in the design phase and later during the production of the devices. The quality assurance methods should be verified to be able to reliably provide the needed characterisation parameters.

In this thesis, the processing of the samples and their characterisation are described from the reliability point of view. Descriptions of processing steps, theoretical models and measurements methods are accompanied by the discussion of possible failures and suggestions how to prevent them. Special attention is put on the impact of the so-called human factor and the importance of knowledge transfer.

The purpose of this work is to study the long-term stability of silicon detectors with Al2O3

thin films grown with the Atomic Layer Deposition (ALD) method and implemented as an insulation and surface passivation layer. The test samples for these studies were processed during 2011-2015 in the cleanroom facilities of the Micronova centre for Micro and Nanotechnology by Dr. E. Tuovinen. Electrical characterisation and characteristic measurements with source were performed in 2014-2018 in the Detector laboratory and cleanroom facilities of Helsinki Institute of Physics (HIP). In addition, Highly Accelerated Temperature and Humidity Stress Tests (HAST) like those performed in the HIP Detector laboratory facilities is suggested as a new approach for studies of the long-term stability of the detectors.

Electrical characterisation demonstrated good long-term stability of capacitance, depletion voltage and leakage current characteristics of the test samples with Al2O3 insulation and surface passivation layer. The result of the characterisation demonstrates that samples with surface passivation are able to withstand higher bias voltage than samples without such passivation. Characteristic measurements with a Cs-137 source confirmed that the surface passivation with Al2O3 does not affect the general detector performance. Characterisation of single pixel sensors coated with the Al2O3 after the flip-chip bonding demonstrated that the additional ALD run is not harmful for the structure of the detector and does not affect its behaviour. Thus, ALD-deposited alumina coating can be recommended as a material for additional protection of silicon detector structures.

iv

Acknowledgements

The research for this doctoral thesis was done between 2014-2018 while I worked as doctoral student at CMS Upgrade project at Helsinki Institute of Physics (HIP) in Helsinki, Finland. I am thankful to the heads of the HIP and the Detector Laboratory, Department of Physics at the University of Helsinki and Micronova Center for micro and nanotechnology for providing the facilities and equipment for the research presented in this thesis.

I am grateful for the patient supervision of my doctoral studies and would like to thank Dr. Jaakko Härkönen and Prof. Paula Eerola for the possibility to start this challenge, and Prof. Panja Luuka and Prof.Kenneth Österberg for the chance to finalize it.

I would like to thank Prof. Bjarne Stugu and Prof. Ari Jokinen for the accurate pre- examination of the manuscript of my thesis and their kind comments and suggestions to improve it.

I am pleased to acknowledge many people from the HIP who helped to materialize this thesis. I am thankful to my colleagues from the CMS Upgrade project: Akiko Gädda, Vladislav Litichevskyi, Alexander Winkler, Aneliya Karadzhinova-Ferrer, Jennifer Ott, Esa Tuovinen, Erik Brücken for the fruitful work together. I am thankful to my colleagues Tiina Naaranoja, Jouni Heino and Pirkitta Koponen for their practical help with my research in HIP Detector Laboratory. I am thankful to HIP coordinators Taina Hardén and Taina Onnela for their care and support with everyday issues.

I am especially grateful to Eija Tuominen and her family for the all-embracing support particularly during the difficult moments.

Finally, I thank my parents and friends for keeping me motivated during all these years.

St.Petersburg, 15.04.2020 Tatyana Arsenovich

v

List of abbreviations

2S Strip-Strip modules AC alternating current

ALD Atomic Layer Deposition

ASIC Application-specific integrated circuit BB Bump bonding

CCE Charge Collection Efficiency

CERN European Organization for Nuclear Research CMS Compact Muon Solenoid

CV capacitance-voltage measurement CVD Chemical Vapor deposition DAC digital-to-analog converter DC direct current

DTB Digital Test Board DUT Device Under Test ESD Electrostatic discharge ESR Equivalent Series Resistance

EYETS extended Year-End Technical Stop FIT Failure In Time

FZ Float Zone (silicon)

GPIB General Purpose Interface Bus

HAST Highly Accelerated Temperature and Humidity Stress Test HCUR high current terminal

HCUR low current terminal HDI High Density Interconnect HIP Helsinki Institute of Physics

HL-LHC High-Luminosity Large Hadron Collider HPOT high potential terminals

IV current–voltage measurement

ISO International Organization for Standardization LHC Large Hadron Collider

LPOT low potential terminals LS long shutdowns

MCz Magnetic Czochralski (silicon) MIP Minimum Ionizing Particles

MOS metal-oxide-semiconductor structure PLM Product Lifecycle management PS Pixel and Strip modules ROC readout chip

SCM Single Pixel Chip Module (SCM) SMU Source-Measure Unit

SNR signal to noise ratio SST Silicon Strip Tracker

TCT Transient Current Technique TEC Tracker End Cap

vi TEDD Tracker End-Cap Double Disks

TIB Tracker Inner Barrel TID Tracker Inner Disk TOB Tracker Outer Barrel

vii

List of symbols

B susceptance

C capacitance

d depth of the depletion region

Ք deviation

D dissipation factor E electric field

Ei intrinsic Fermi level

Et energy level of the center of the bandgap Ew weighting field

f physical frequency f(t) failure density function F(t) number of failed devices

G conductance

ॳ rate of electron-hole generation

I current

Ig current caused by generation processes Iind induced current

Ileak leakage current Imax current amplitude k Boltsmann’s constant

L inductance

n electrons (charge carriers)

NA acceptor concentration (in p-type bulk Ndop=NA) ND donor concentration (in n-type bulk Ndop=ND) Ndop doping concentration

Neff effective space charge density ni intrinsic carrier concentration

Nt concentration of carriers in semiconductor p holes (charge carriers)

q elementary charge Q total charge Qc quality factor

R resistance

Rpl radius of contact plate Թ reliability

r volumetric ratio

S square

T period

T temperature

t time

U potential difference (voltage) Uw weighting potential

ॼ recombination rate

V voltage

viii Vdrift velocity of charge carriers

Vfd full depletion voltage

Vth thermal velocity of the carriers

w thickness of the detector (distance between contact pads)

šത mean of the data set

Y admittance

Z complex impedance

α standard error

εr relative dielectric constant of the material ε0 vacuum permittivity, ε0=8,85∙10¹² F/m

θ phase angle

λ(t) failure rate

μ mobility

ρ charge density

σ distribution of charge over the surface σn capture cross sections for electrons σp capture cross sections for holes ɐ_୒ିଵ standard deviation

τg generation lifetime

ϕ potential

χc capacitive reactance

χ reactance

ω angular frequency

ix

List of figures

Figure 1 CMS detector

Figure 2 Projected LHC performance through 2038

Figure 3 Schematic cross-section through the original CMS Strip Tracker Figure 4 Schematic view of the CMS Pixel detector

Figure 5 TOB silicon strip detector module

Figure 6. LHC roadmap of long shutdowns, runs, and upgrade phases

Figure 7 Comparison of the Phase-1 pixel detector and the initial detector layout Figure 8 Scheme of one quarter of upgraded tracker layout.

Figure 9 Illustration of operational principle of silicon detector

Figure 10 Propagation of signal in p-in-n type and n-in-p silicon detectors Figure 11 P-stop and p-spray technique

Figure 12 Dissected parts of the pixel modules for barrel layers of pixel tracker Figure 13 Schematics of the pixel array

Figure 14 Layout of the pixels on the sensors for the forward and barrel detector.

Figure 15 Cross-sectional view of basic principle of strip sensor for SST

Figure 16 Schematic of Czochralski and float-zone silicon crystal growth techniques Figure 17 Cross-sectional view to the vacuum chamber of magnetron sputtering system Figure 18 Schematic illustration of one ALD cycle

Figure 19 Optical lithography steps.

Figure 20 Fairchild wafer dicing station

Figure 21 Microscope view of wire bonds connecting sensor to fan-out circuit, and electron micrograph of bonded wire

Figure 22 Schematic view of flip-chip bonding of pixel sensor Figure 23 3D reconstruction of defective solder bump

Figure 24 Ground wire pad of the readout chip after destructive sparking

Figure 25 Layout of the wafer for the CMS endcap silicon strip detector and standard half-moon set

Figure 26 HIP standard diode structure

Figure 27 Schematic cross-sectional structure of HIP standard diode Figure 28 Set of HIP standard diodes for studies of Al2O3 thin films Figure 29 Wafer layouts containing samples for groups A, C, and group B Figure 30 Cross-sectional structure of HIP standard diode with surface passivation Figure 31 Layout of the pixel sensor wafer

Figure 32 Single sensor (SCM) with the 20nm layer of Al2O3 as the surface passivation Figure 33 Process of the research and development.

Figure 34 Flat capacitor composed from different dielectric materials Figure 35 Electric field of flat capacitor with equal contact pads

Figure 36 Contact pads of flat capacitor and consequent potential, and two surface areas with different potentials

Figure 37 Cross-sectional view of typical capacitor structures used in integrated circuits Figure 38 Microscopic picture of HIP standard diode guard rings

Figure 39 Equipotentials and flux lines at the edge of a thick electrode (near field) Figure 40 Parameters of sine function

Figure 41 Vector diagram for current and voltage through the capacitor

x Figure 42 Complex impedance plane and impedance components

Figure 43 Capacitor in AC electric scheme and vector diagram of capacitor in AC electric scheme

Figure 44 Resonance circuit for capacitance measurements Figure 45 Schematic configuration of simple bridge.

Figure 46 Agilent E4980 LCR meter

Figure 47 Connection of Agilent E4980 to DUT with 4-point Kelvin connection Figure 48 Scheme of parallel and series resistances in LCR meter

Figure 49 Probe station in Karlsruhe for measurements in dark and cold environments Figure 50 CV configuration of Karlsruhe setup

Figure 51 Connections at CERN probe station for CV and IV measurements (2017) Figure 52 Overview of the HIP first probe station at cleanroom

Figure 53 Older HIP CV probe station

Figure 54 HIP probe station for characterisation of the CMS pixel modules Figure 55 Overview HIP probe station in 2017

Figure 56 CV configuration (connections) of HIP2017 probe station

Figure 57 Positioning of the probe needles for standard CV-IV tests of HIP standard diode

Figure 58 General connection rules for LCR meter E4980

Figure 59 Cable extension and prober connection in four-terminal pair configuration for measurements of structures on the wafer

Figure 60 Gain and offset errors in a capacitance measurements

Figure 61 Histogram of values of background capacitance and its deviations of HIP probe station

Figure 62 Typical CV plot obtained by HIP probe station

Figure 63 Depletion region, and depletion region and applied voltage as simple electrical scheme

Figure 64 Capacitance as a function of applied potential, and values of standard deviation for capacitance

Figure 65 Determination of depletion voltage

Figure 66 Depletion voltage for all HIP standard diodes Figure 67 Setup for IV measurements

Figure 68 Determining the leakage current at full depletion for HIP standard diode Figure 69 Behaviour of leakage current

Figure 70 Configuration of HIP probe station used for IV measurements Figure 71 Keithley 6487 input

Figure 72 Contact pad of pixel sensor damaged with incorrect application of improper needle

Figure 73 Result of pixel alive test for the ROC sensor

Figure 74 IV characterisation of pixel sensors performed at Micronova facilities Figure 75 Visualization of the IV criteria.

Figure 76 Internal calibration mechanism of the ROC Figure 77 Result of Pretest procedure

Figure 78 Result of PixelAlive procedure Figure 79 Result of the Bump Bonding test Figure 80 Results of Bump Bonding test

xi Figure 81 Result of bump-bonding test

Figure 82 Leakage current of HIP standard diodes

Figure 83 Efficiency maps for the single sensors before and after the additional ALD Figure 84 Energy loss of electrons in silicon

Figure 85 Typical signal processing chain

Figure 86 Typical setup for characterisation measurements with the source

Figure 87 Cremat CR-160-R7 evaluation board with installed CR-200-100ns shaping amplifier

Figure 88 Amptek DP5 digital pulse processor

Figure 89 Block diagram of the DP5 installation in a characterisation setup Figure 90 Caesium-137 spectrum

Figure 91 Characterisation spectra of Cs-137 obtained by four groups of HIP standard diodes.

Figure 92 Fitted curves for characterisation data for the four groups of HIP standard diodes.

Figure 93 Scheme of the TCT measurement, form of electric field and comparison between TCAD simulations and measurements

Figure 94 Schematic layout of HIP TCT measurement setup (red laser) Figure 95 Schematic view of charge carriers movement during TCT operation Figure 96 Current pulse shapes measured after electron and hole injection

Figure 97 Determination of depletion voltage using linear fits to the charge collection efficiency (CCE) plot

Figure 98 Signal waveforms for HIP standard diodes.

Figure 99 Collected charge (normalised) for all groups of samples Figure 100 Bathtub curve

Figure 101 General product lifecycle vision common for commercial product lifecycle management

Figure 102 Lifecycle in science

Figure 103 Growing crystals on Al2O3 passivated sample

Figure 104 Microscopic granular structures on surfaces of p-type and n-type HIP standard diodes

Figure 105 Evolution of blisters

Figure 106 Dust accumulation after dicing

Figure 107 Temperature and humidity chamber, and samples placed on plastic bench inside the chamber

Figure 108 Leakage current before and after HAST Figure 109 Depletion voltage before and after HAST Figure 110 Presence of oxide over contact openings

xii

List of tables

Table 1 Properties of Al2O3, SiO2 and HfO2

Table 2 Sets of HIP standard diodes for studies of Al2O3 thin films Table 3 Parallel and series models

Table 4 Equipment for Karlsruhe and HIP probe stations Table 5 Capacitance and depletion voltage for all samples

Table 6 Theoretical calculations of HIP standard diode capacitance Table 7 Criteria for grading bare modules

Table 8 Pixel and bump bonding problems of single sensors before and after additional ALD of Al2O3

Table 9 Summary of the characterisation data for all groups of HIP standard diodes Table 10 Failure mechanisms related to processing chain

Table 11 Reliability of HIP standard diodes with Al2O3 surface passivation during 2014-2018

xiii

List of equations

(1) † ൌ ට^ଶக_୯୒^బக౨୚

ౚ౥౦ Depth of the depletion

region ^p.16

(2) _୤ୢ ൌ^୯୒_ଶக^{ౚ౥౦୵మ}

బக_౨ Depletion voltage p.16

(3) ഥ_{ୢ୰୧୤୲}ൌ Ɋ

Velocity of charge carriers in depletion region

p.16 (4) ൌ ɔ̴ͳ െ ɔ̴ʹ Potential difference p.43

(5) ൌ_஦ ^୕

భି஦_మൌ^୕_୙ Capacitance p.43

(6) ൌ ɐ

ʹɂ_଴൅ ɐ ʹɂ_଴ൌ ɐ

ɂ_଴

Electric field of two parallel conductive plates

p.43

(7) ൌ_க^஢

౨ήக_బ

Electric field of two parallel conductive plates filled with the material with dielectric constant ɂ_୰

p.43

(8) ൌ ή ™ ൌ_க^஢

౨ήக_బή ™ Potential difference expressed with electric field

p.43

(9) ൌ ɐ ή Total charge p.44

(10) ൌ ^୕_୙ൌ^ୗήக_୵^౨ήகబ Capacitance p.44

(11) _ଵൌ_க^஢

౨ήக_భ, _ଶൌ െ_க^஢

౨ήக_మ

Electric field for layers with different permittivity

p.45

(12) ɔ_ଵെ ɔ_ଶൌ_க^஢

౨ήக_భή ™_ଵ൅_க^஢

౨ήக_మή ™_ଶ Potential difference between two parallel plates

p.45

(13) _୲୭୲ൌ_౭భ ^கబήୗ

಍భା^౭మ_಍మାڮା^౭౤_಍౤ Total capacitance for n dielectric layers in capacitor

p.45

(14) _୲୭୲ൌ భ ^ଵ

ిభା^భ

ిమାڮା^భ

ి౤

Capacitance of n

consequent capacitors ^p.45

(15) ܥ ൌ^{ሺఌభή௥ାఌమሻఌబήௌ}

ሺଵା௥ሻ௪

Capacitance of n consequent capacitors filled with two constituent dielectrics

p.45

(16) ൌ ට൬ቀ^௟ή୙_ୢቁ^ଶ൅ ቀ^௠ή୙_ୢ ቁ^ଶ൅ ቀ^௡ή୙_ୢ ቁ^ଶ൰ ൌ^୙_ୢ Electric intensity p.45 (17) _{ୡ୰୷ୱ୲ୟ୪} ൌ^{௟మήக౮ା௠మ}_ୢ^{ήக౯ା௡మήக౰}

Capacitance of anisotropic crystal structure

p.46

xiv

(18) ɐ^כ ൌ ɂ_଴ή _୬ൌ^க_ଶ஠ή୶^బή஦బ Redundant charge

density ^p.47

(19) ^כ ൌ ׬ ɐ^כ†š ൌ^கబή஦_ଶ஠^బή୐Ž^ξୗ_୵ Total redundant

charge ^p.47

(20) ^כ ൌ ൅ ή ɂ_଴ή Ž^ξୗ_୵ Capacitance with

fringe field correction ^p.47

(21) ^כ ൌ ൅ ή ɂ_଴ή ɂ_୰ή Ž^ξୗ_୵

Capacitance with fringe field

correction, capacitor filled with a medium with relative

permittivity ɂ_୰

p.47

(22) ^כ ൌ ൅^ୖ_ସ஠^౦ౢቀŽ^{ଵ଺஠ήୖ}_୵^౦ౢെ ͳቁ Kirchhoff’s formula

(Gaussian units) ^p.47

(23) ‹ ൌ _୫ୟ୶ή •‹ ቀ^ଶ஠ή୲_୘ ൅ Ȳቁ ൌ _୫ୟ୶ή

•‹ሺɘ ή – ൅ Ȳሻ

Alternating current defined by sine function

p.49

(24) @_௠௔௫ ൌ _୫ୟ୶ή ‡^୨ஏ Amplitude of

alternating current ^p.50

(25)

ൌ ට_୘^ଵ׬ ‹_଴^{୘ ଶ}†–ൌ

ට^ଵ_୘׬ _଴^୘ ୫ୟ୶ଶ ή •‹^ଶɘ– ή †–ൌ^୍ౣ౗౮

ξଶ

Effective value for

current ^p.50

(26) ൌ^୙ౣ౗౮_ξଶ Effective value for

voltage ^p.50

(27) ൌ^୉ౣ౗౮

ξଶ Effective value for

field ^p.50

(28) _ୟ୴ൌ_୘^ଵ

ൗଶ׬_଴^୘ൗଶ_୫ୟ୶ή •‹ ɘ– ή †– ൌ_஠^ଶ_୫ୟ୶ Average value for

current ^p.50

(29) _ୟ୴ൌ^ଶ_஠୫ୟ୶ Average value for

voltage ^p.50

(30) _ୟ୴ ൌ_஠^ଶ_୫ୟ୶ Average value for

electric field ^p.50

(31) ൌ _୫ୟ୶ή •‹ ɘ– Potential defined by

sine function ^p.50

(32) “ ൌ ൌ _୫ୟ୶•‹ ɘ– Charge of the

capacitor in case of sine potential

p.50

(33) ‹ ൌ^ୢ୯

ୢ୲ൌ ɘ ή ή _୫ୟ୶ή …‘• ɘ– ൌ

୙_ౣ౗౮

ଵ னൗ ήେή •‹ ቀɘ– ൅^஠_ଶቁ Flowing current p.51 (34) _୫ୟ୶ ൌ^୙ౣ౗౮_஧

ౙ Peak current p.51

(35) ‹ ൌ^ୢ୯_ୢ୲ ൌ ^ୢ୙_ୢ୲ Flowing current p.51

xv

(36) ൌ^ଵ_େ׬ ‹†– Potential on the

capacitor and current through the capacitor

p.51

(37) ൌ ൅ Œ Complex impedance p.51

(38) Ʌ ൌ ƒ”…–ƒ ቀ^ଡ଼_ୖቁ Phase angle p.52

(39) ȁȁ ൌ ඥሺ^ଶ൅ ^ଶሻ Admittance p.52

(40) _ୡൌ ^ଵ

୧னେൌ ^ଵ

னେ‡^{୨ቀିಘమቁ} Reactive impedance

of the capacitor ^p.52

(41) _ୡൌ െ_னήେ^ଵ Variable resistance p.52

(42) –ƒ Ɂ ൌ_୕^ଵ

ౙ Loss tangent δ p.53

(43) ൌ^ȁଡ଼ȁ

ୖ ൌ ^ୋ

ȁ୆ȁ Dissipation factor p.53

(44) _୫ୟ୶ൌ ή _୫ୟ୶ Ohm’s law p.53

(45) ܴ_௖ ൌ_ఠή஼^ଵ Capacitance defined

with variable resistance

p.53

(46) ൌ ^୙

ටୖ^మାቀ୐னି_ిಡ^భቁ^మ

Current defined with

variable resistance ^p.54

(47) ή ɘ ൌ_େήன^ଵ Condition for

constant potential ^p.54

(48) _୶ ൌ_୐ήன^ଵ

బమ Capacitance defined

with inductance ^p.54

(49) ൜ ‹_ଵଵെ ‹_ଶଶ൅ ‹_୴୴ൌ Ͳ

െ‹_୶ୡ୶൅ ‹_଴ୡ଴൅ ‹_୴୴ൌ Ͳ Equations for branches of mentioned circuit

p.55

(50) ൜‹_ଵൌ ‹_୶൅ ‹_୴

‹_଴ൌ ‹_ଶ൅ ‹_୴ Equations for points

of mentioned circuit ^p.55

(51) ൜‹_ଵൌ ‹_୶

‹_ଶൌ ‹_଴ Equations for points

of mentioned circuit, resistivity inserted

p.55

(52)

ەۖ

۔

ۖۓ ^ୡ୶ൌ^ୖ_ୖ^భ

మ_ୡ଴

ଵ

େ_౮னൌ^ୖ_ୖ^భ

మή_େ^ଵ

బήன

_୶ ൌ^ୖ_ୖ^మ

భή _଴

Balance conditions p.55

(53) െ^ୢ_ୢ୶^మɔሺšሻ ൌ_ఌ^஡

బఌ_ೝൌ_ఌ^୯

బఌ_ೝή _ୣ୤୤ Poisson equation for

2D case ^p.73

(54) ൌ ට^{கబήக౨}_ଶ୚^{ή୯ή୒౛౜౜}

Capacitance defined using Poisson equation

p.73

(55) ’ ് _୧^ଶ Equilibrium

concentrations of charge carriers

p.79

xvi

(56)

ॳ ൌ െॼ ൌ ቈ ^{஢౦ή஢౤ή୚౪౞ή୒౪}

஢౤ήୣ୶୮ቀ^{ు౪షు౟}_ౡ܂ ቁା஢౦ୣ୶୮ቀ^{ు౟షు౪}_ౡ܂ ቁ቉ _୧ؠ_த^୬౟

ౝ Rate of electron-hole

pair generation ^p.79

(57) _୥ ൌ ׬ “ॳ†š ൎ “ॳ† ൌ^୯୬_த^౟ୢ

ౝ

ୢ

଴ Current caused by

generation processes in depletion layer

p.79

(58) ^{୍ౢ౛౗ౡ൫୚౥౦൯}

୍_ౢ౛౗ౡሺ୚_౜ౚሻ ൏ ʹ Ratio of values of the

leakage current ^p.87

(59) ൌ_୕^୕

బ Charge Collection

Efficiency ^p.100

(60) _୧୬ୢሺ–ሻ ൌ െ“ ή ˜തതതതതത ή ഥ_{ୢ୰న୤୲ ୵}

Current induced on a collecting electrode by the motion of charge carriers

p.105

(61) _୵ሺ†ሻ ൌ^ଵ_ୢ Weighting field p.106

(62) ˜ത_{ୢ୰୧୤୲} ൌ Ɋ ή ഥ

Velocity of charge carriers in the depletion region

p.106

(63) _୧୬ୢൌ^{୬ή୯ήஜή୉}_ୢ Induced current

generated by n charge carriers

p.106

(64) Քൌ š_୧െ šത Deviation p.112

(65) ɐ_୒ିଵ ൌ ට_୒ିଵ^ଵ σ^୒_୧ୀଵ†_୧^ଶ Standard deviation p.112

(66) ߙ ൌ^ఙಿషభ_ξே Standard error p.112

(67) Թሺ–ሻ ൌ_୒^{୒౩౫౨౬}

౟౤౟౪౟౗ౢ Reliability of a

population of devices ^p.115

(68) ሺ–ሻ ൌ ͳ െ Թሺ–ሻ Number of failed

devices ^p.115

(69) ɉሺ–ሻ ൌ_Թሺ୲ሻ^୤ሺ୲ሻ ൌ_{ଵି୊ሺ୲ሻ}^୤ሺ୲ሻ Failure rate p.115

xvii

Content

ABSTRACT ... III ACKNOWLEDGEMENTS ... IV LIST OF ABBREVIATIONS ... V LIST OF SYMBOLS ... VII LIST OF FIGURES ... IX LIST OF TABLES ... XII LIST OF EQUATIONS ... XIII CONTENT ... XVII

AUTHOR’S CONTRIBUTION ... 1

INTRODUCTION AND MOTIVATION ... 3

STRUCTURE OF THIS WORK ... 7

I. BACKGROUND OF THE RESEARCH ... 8

1. CMS experiment at Large Hadron Collider (LHC) ... 8

2. CMS Tracker ... 9

2.1 Initial design ... 10

2.2. Upgrade of CMS silicon Tracker ... 12

II. SILICON DETECTORS... 15

1. Operational principle of silicon detector ... 15

2. Detectors based on n-type and p-type silicon ... 16

3. Silicon detectors in CMS Tracker ... 18

3.1 Pixel sensors ... 18

3.2 Silicon strip detector ... 21

4. Processing of the detectors ... 21

4.1. Silicon wafers ... 22

4.2 Doping ... 23

4.3 Thin film processing ... 24

4.4 Patterning ... 26

4.5 Dicing and storing ... 27

4.6 Connections and assembly ... 28

4.7 Environment of processing facilities ... 30

xviii

5. Development of silicon detectors in HIP ... 31

5.1 Material choice ... 31

5.2. Test samples ... 34

III. CHARACTERISATION ... 40

1. Capacitance measurements... 41

1.1 Capacitance theory ... 42

1.2 Capacitance measurement methods ... 52

1.3. Instrumentation for capacitance measurements ... 54

1.4. Errors in CV measurements and possible corrections ... 66

1.5 Applications of capacitance measurements for quality assurance of silicon detectors ... 71

1.6 Results of capacitance measurements of HIP test samples ... 75

2. Leakage current measurements ... 78

2.1 Leakage current theory... 78

2.2 Leakage current measurements ... 79

2.3 Instrumentation for leakage current measurements ... 82

2.4 Errors in leakage current measurements and possible corrections ... 83

2.5 Applications of leakage current measurements for quality assurance of silicon detectors ... 85

2.6 Results of leakage current measurements of HIP test samples ... 92

3. Characteristic measurements with the source (Caesium-137) ... 98

3.1 Readout of the signal ... 98

3.2 Setup for characteristic measurements ... 100

3.3 Errors in characteristic measurements ... 102

3.4 Results of characteristic measurements of HIP test samples ... 102

4. Transient Current Technique (TCT) measurements ... 104

4.1 Transient Current Technique theory ... 104

4.2 Instrumentation for TCT measurements ... 106

4.3. Errors in TCT measurements ... 107

4.4 Applications of the TCT measurements ... 107

4.5 Results of TCT measurements of HIP test samples... 108

5. Evaluation of measured data ... 110

IV. RELIABILITY ... 112

1. Reliability terms and definitions ... 113

1.1 Statistical models and definitions ... 113

2. Physics of failure ... 114

2.1 Classification of failures ... 115

3. Commercial and scientific approaches ... 120

3.1 Product lifecycle management ... 120

3.2. Standards and tests ... 121

xix

4. Reliability of HIP test samples ... 122

4.1. Physics of failure and defects ... 123

4.2. Long-term stability studies with the HAST ... 126

V. RESULTS AND DISCUSSION ... 131

1. Summary ... 131

1.1 Test samples ... 131

1.2. Characterisation results ... 132

1.3 Reliability of HIP test samples ... 142

2. Discussion and outlook ... 143

VI. CONCLUSIONS ... 146

BIBLIOGRAPHY ... 148

APPENDIXES ... 156

A. RD50 Recommendations towards a standardisation IV and CV measurements in Si diodes ... 156

B. ISOBOX ... 158

C. Relay system for Karlsruhe probe station ... 159

D. Probe needles ... 160

1

Author’s contribution

The research for this thesis was carried out within the Compact Muon Solenoid (CMS) Upgrade project at the Helsinki Institute of Physics during the years 2014–2018. During this time, the author was in charge of the electrical characterisation of detector structures and participated in the characterisation of various test structures within the framework of the project. The author also participated in the development of the characterisation methodology and processing, and was in charge of the reliability studies presented in this thesis, as well as the development of approaches for these studies. This thesis also refers to the preliminary research carried out by the author during the 2011–2012 within the framework of the author’s Master’s thesis.

The publications listed below are a result of a group effort. The author participated in the electrical characterisation (CV and IV) of the samples presented in these publications.

The author presented the results of the studies on the long-term stability of samples described in this thesis at international conferences.

List of publications

1. Processing of n+/p-/p+ strip detectors with atomic layer deposition (ALD) grown Al2O3 field insulator on magnetic Czochralski silicon (MCz-si) substrates

J. Härkönen, E.Tuovinen, P.Luukka, A.Gädda, T.Mäenpää, E.Tuominen, T.Arsenovich, A.Junkes, X.Wu, Z.Li

Nuclear Instruments and Methods in Physics Research A: Volume 828 (2016), p 46–51 Abstract. Detectors manufactured on p-type silicon material are known to have significant advantages in very harsh radiation environment over n-type detectors, traditionally used in High Energy Physics experiments for particle tracking. In p-type (n+ segmentation on p substrate) position-sensitive strip detectors, however, the fixed oxide charge in the silicon dioxide is positive and, thus, causes electron accumulation at the Si/SiO2 interface. As a result, unless appropriate interstrip isolation is applied, the n-type strips are short-circuited. Widely adopted methods to terminate surface electron accumulation are segmented p-stop or p-spray field implantations. A different approach to overcome the near-surface electron accumulation at the interface of silicon dioxide and p-type silicon is to deposit a thin film field insulator with negative oxide charge. We have processed silicon strip detectors on p-type Magnetic Czochralski silicon (MCz-Si) substrates with aluminum oxide (Al2O3) thin film insulator, grown with Atomic Layer Deposition (ALD) method. The electrical characterization by current–voltage and capacitance−voltage measurement shows reliable performance of the aluminum oxide. The final proof of concept was obtained at the test beam with 200 GeV/c muons. For the non-irradiated detector the charge collection efficiency (CCE) was nearly 100% with a signal-to-noise ratio (S/N) of about 40, whereas for the 2×10¹⁵ neq/cm² proton irradiated detector the CCE was 35%, when the sensor was biased at 500 V. These results are comparable with the results from p-type detectors with the p-spray and p-stop interstrip

2 isolation techniques. In addition, interestingly, when the aluminum oxide was irradiated with Co-60 gamma-rays, an accumulation of negative fixed oxide charge in the oxide was observed.

2. Atomic Layer Deposition (ALD) grown thin films for ultra-fine pitch pixel detectors

J.Härkönen, J.Otta, M.Mäkelä, T.Arsenovich, A.Gädda, T.Peltola, E.Tuovinen, P.Luukka, E.Tuominen, A.Junkes, J.Niinistö, M.Ritala

Nuclear Instruments and Methods in Physics Research A: Volume 831 (2016), p 2-6 Abstract. In this report we cover two special applications of Atomic Layer Deposition (ALD) thin films to solve these challenges of the very small size pixel detectors. First, we propose to passivate the p-type pixel detector with ALD grown Al2O3 field insulator with a negative oxide charge instead of using the commonly adopted p-stop or p-spray technologies with SiO2, and second, to use plasma-enhanced ALD grown titanium nitride (TiN) bias resistors instead of the punch through biasing structures. Surface passivation properties of Al2O3 field insulator was studied by Photoconductive Decay (PCD) method and our results indicate that after appropriate annealing Al2O3 provides equally low effective surface recombination velocity as thermally oxidized Si/SiO2 interface. Furthermore, with properly designed annealing steps, the TiN thin film resistors can be tuned to have up to several MΩ resistances with a few μm of physical size required in ultra-fine pitch pixel detectors.

3. Processing and characterization of epitaxial GaAs radiation detectors

X.Wu, T.Peltola, T.Arsenovich, A.Gädda, J.Härkönen, A.Junkes, A.Karadzhinova, P.Kostamo, H.Lipsanen, P.Luukka, M.Mattila, S.Nenonen, T.Riekkinen, E.Tuominen, A.Winkler

Nuclear Instruments and Methods in Physics Research A: Volume 796 (2015), p 51-55 Abstract. GaAs devices have relatively high atomic numbers (Z=31, 33) and thus extend the X-ray absorption edge beyond that of Si (Z=14) devices. In this study, radiation detectors were processed on GaAs substrates with 110-130 μm thick epitaxial absorption volume. Thick undoped and heavily doped p⁺ epitaxial layers were grown using a custom-made horizontal Chloride Vapor Phase Epitaxy (CVPE) reactor, the growth rate of which was about 10 μm/h.

The GaAs p⁺/i/n⁺ detectors were characterized by Capacitance Voltage (CV), Current Voltage (IV), Transient Current Technique (TCT) and Deep Level Transient Spectroscopy (DLTS) measurements. The full depletion voltage (Vfd) of the detectors with 110 μm epi-layer thickness is in the range of 8–15 V and the leakage current density is about 10 nA/cm². The signal transit time determined by TCT is about 5 ns when the bias voltage is well above the value that produces the peak saturation drift velocity of electrons in GaAs at a given thickness.

Numerical simulations with an appropriate defect model agree with the experimental results.

3

Introduction and motivation

The Large Hadron Collider (LHC) at CERN [1] will be updated to the High-Luminosity LHC (HL-LHC) in 2025–2027, and as a result all of the LHC experiments must be upgraded to meet the goals set for high-quality physics data taking. One of the experiments is Compact Muon Solenoid (CMS) [2], whose Tracker will undergo several planned upgrades aimed to improve the characteristics of its detectors without compromising its’ physics performance.

With the start of HL-LHC, the level of radiation will increase significantly, thus, the radiation hardness of the detectors should be improved. The number of sensors will be increased, the amount of readout channels will respectively be increased, and all connections and supporting structures will also be renewed to cope with the more demanding operational conditions. The larger pileup and associated increase in particle density requires higher detector granularity to reduce the occupancy to ensure efficient charged particle tracking and increased bandwidth to accommodate the increased data rates. The trigger capability will also be improved to allow efficient triggering of the interesting physics event in the higher pileup conditions of HL- LHC. The candidate materials and technologies for the development of the detectors need to be reviewed taking the requirements for the Tracker upgrade [3] into account. The quality assurance of these detectors is of utmost importance to identify possible failures as early as possible.

The CMS upgrade project at the Helsinki Institute of Physics promoted studies of new materials when I joined as a Master student in 2011. From the materials with high permittivity Al2O3, HfO2, ZrO2, and Ta2O5 were considered as promising candidates to be included into the layouts of future detector structures. The focus was on a processing approach aiming to deposit a thin film with uniform properties. Further development should result in detector structures with stable properties — and to ensure this, reliable characterisation approaches should be applied.

I participated in the first attempts to achieve Hafnium dioxide (HfO2) deposition, which encountered significant obstacles due to problems with the equipment available at that time.

It was impossible to derive a reliable combination of parameters leading to the successful deposition of the conformal thin film of HfO2. I learnt how the reliability of the components and equipment might affect the result of processing. In the first instance, I focused my attention on the impact of factors typically underestimated, such as human factors and processing facilities policies. I studied the quality assurance methods for the detector characterisation as well, but the importance of their reliability I discovered later in 2014, when I started my work as a doctoral student in the CMS Upgrade project.

During 2011–2012, under the supervision of Dr. E. Tuovinen, I processed test structures with Al2O3 thin films to study their capacitive properties. Some of the first test detector structures with Al2O3 surface passivation and insulation layers were processed also at that time, and I participated in their characterisation performed at HIP and CERN facilities. In 2014, I continued to study the properties of these structures and discovered that newly produced similar structures had significantly different characteristics. My attempts to find the reason for this difference resulted in a deep review of our approaches and theoretical aspects, and finally led to changes in the laboratory practices.

When experimental samples fail to show the expected results, it might seem trivial. Such samples should simply be considered as failed, and studies should concentrate on the reason for its failure in order to prevent this kind of problem in the future. But there was something

4 intriguing in the collected measurement results. I witnessed the failure of certain samples that later demonstrated characteristics ideally corresponding to the preliminary calculated values.

Different measurement results were obtained from the same sample with the same equipment but by different operators, or by same operator, but in different days; or new test samples processed using already successfully tested receipt demonstrated weird characteristics. The fundamental scientific principle of repeatability cannot be contravened, thus some hidden factors should be unveiled. Many of those were found during my doctoral studies and I would like to share my experience in this thesis to help identify more such issues in the future.

For example, human factors play a significant role in the electrical characterisation of the detector structures. Manually operated probe needles are used to establish contact with the samples during these tests. Different operators have different hands and vision, thus the force applied with the probe needle to the surface of the sample is different. It might result in the penetration of the probe needle through the conductive layer and the sample gets damaged.

Proper training and information exchange should help to minimize this problem and other possible issues related to the operator’s action, such as harmful hold and movement of the sample. The importance of knowledge and skill transfer is another underestimated factor. In research institutions, a high rotation of personnel is natural: it is expected that scientists participate in multiple projects during their career. Proper knowledge transfer is necessary to organise smooth and effective research projects, despite the rotation of personnel.

Electrical characterisation of the detector structures with Al2O3 processed in 2014–2015 revealed differences in their properties from preliminary calculated values and properties of structures that had been previously processed. Such conclusions can only be reached in cases of accurate sample identification. I found that labelling and tracking of the samples are also underestimated factors that can be critical for the research flow in case of the rotation of personnel: important information can be completely lost if not properly documented before a responsible person leaves the project. In most cases, the layout for processing of the experimental samples on silicon wafer includes numbers or letters to identify structures after dicing. Unfortunately, cases where any kind of identification mark is missing are not so rare.

The layout of the wafer might include several types of test structures developed by experts from different institutions. These kinds of layouts are widely used to reduce production costs, since full processing lines are not always available for research projects. I suggested to use coordinates to identify structures in the arrays in such cases and, together with my colleagues, we established an abbreviation system to sort and track our test structures; a more advanced version of this system could be useful also for collaborative research.

At that time, the range of our test structures was extended for detailed studies of the capacitive properties of Al2O3, and the size of the structures was decreased. For reliable measurement results, the instrumentation was reviewed and upgraded since the order of expected capacitance values was the same as the order of the background capacitance of the previous configuration of the probe station. Preliminary calculated values started to match with the measured values of the capacitance after several iterations of probe station upgrades.

Documentation of this process is an important factor, together with the need to review the reliability of the instrumentation for every new stage in the project.

The instrumentation providing the data should be reliable. The reliability should be sufficient to remove the probability that mistakes and artefacts will be adopted as actual results. Thus, a procedure for verification of reliability should exist for all levels of development, characterisation, and operation of the instrumentation, namely particle

5 detectors, for every part of these detectors and for the whole system of detectors in the experiment. The theory used for preliminary calculations and modelling should also be reviewed. Minor issues previously neglected due their minimal effect might become notable with the further miniaturisation of the detector structures. In the case of capacitance measurements, for instance, fringe field or different dielectric constants may be used for the components of multilayer structures. This thesis considers the review of capacitance calculations and measurements due to the possible changes in detector geometry.

Silicon-based particle detectors can become non-operational because of degradation due to the radiation induced defects and other long-term effects. Severe problems can occur if our estimations of the detector lifetime in harsh radiation conditions were inaccurate enough and thus the detectors died much earlier than expected. This is why it is of utmost importance that the calculations and models used for the preliminary estimations are correct and reliable.

During 2014–2018, I organised several sessions of characterisation of the detector structures with the Al2O3 layers produced in 2012. The data collected provided the possibility to consider the long-term stability of these structures and create a reference for further studies of radiation effects on such structures. However, it is not always possible to organise such long-term studies, thus an approach to study aging effects should be invented for further research of other types of materials and structures. I suggest to adopt industrial accelerated temperature and humidity stress test (HAST) generally used for reliability studies of the semiconductor devices in industry.

Characterisation sessions of the structures with Al2O3 layers also aim to examine the possibilities to include alumina thin films to the structure of the detector. I studied the behaviour of the standard test diode structures to prove the ability of the suggested usage of aluminium oxide as an insulator and surface passivation material for the p-type silicon detectors. It was shown that additional surface passivation improves the stability of the diodes, which did not demonstrate breakdown until 500 V (limitation of the setup), whereas similar structures without surface passivation broke down at 150–200 V.

I examined the possibility to add surface passivation with Al2O3 to the processing chain of the pixel sensors. The approach was confirmed to be non-harmful for the flip-chip bonding between the sensor and the readout chip. I performed characterisation of the pixel sensors with surface passivation, which revealed no significant difference in their behaviour before and after the additional deposition run.

I summarize my work by listing my experience in this thesis. I sought to create a text that I would have liked to have at the beginning of my doctoral studies, and I hope it will help my colleagues in the future. The aim is to list the studies of long-term stability of the detector structures with alumina thin films and make a conclusion about them, as well as to explain the applied methods and possible reasons for failure by illustrations from these studies. I extensively described related theoretical aspects and collected important points from the practice to create a text helpful for further knowledge transfer, because the necessity of continuity of the information related to research practices is the key factor of proper quality management in research practices. I described quality assurance methods used for the characterisation of silicon detectors from the reliability point of view. I mentioned the effect of human factors on reliability in scientific research as a subject of special attention. I collected descriptions of various test setups to ensure further continuity of the characterisation methods for the project. I reviewed possible errors and approaches to prevent the most common errors or minimize their effects focusing on the methods of quality assurance based

6 on electrical characterisation. I introduced basic reliability concepts and suggested to adopt reliability approaches for systematic studies of the silicon detectors. More detailed description of the structure of this thesis is provided below.

Tatyana Arsenovich Helsinki – St.Petersburg 2018–2019

7

Structure of this work

This thesis is organised into six chapters. Chapters II–IV consist of the overview of subject theory accompanied by actual experimental results relevant to its topic.

Chapter I is dedicated to an overview of the CMS project at LHC and sets the background of the research. The aim of this chapter is to introduce one of the biggest projects involving silicon detectors and to mention the challenges associated with their upgrades.

In Chapter II, facts about silicon detectors are collected. The operational principle of silicon detectors is defined generally and typical structures of the detectors used for the CMS project are provided. The main processing techniques are reviewed, and possible reliability issues associated with them are mentioned. The description of selected test samples processed in the Helsinki Institute of Physics is provided with the motivation of material choice for further research. The aim of this chapter is to introduce the main subject of this study, test samples with Al2O3 layers. One group of test samples consists of standard test diodes, which can be found in the layouts of most wafers developed experimentally. It includes samples with an Al2O3 insulation layer and samples with an Al2O3 insulation layer and additional surface passivation layer of Al2O3. A second group of test samples consists of single sensors (SCM) with alumina surface passivation.

Chapter III describes techniques used for the characterisation of silicon detectors on the stage of development. Attention is focused on the challenges associated with electrical measurements. The results of capacitance and leakage current measurements (CV–IV) and Transient Current Technique (TCT) measurements of the test samples are provided in order to examine the behaviour of the samples with Al2O3 insulation layer and samples with Al2O3

insulation and an additional surface passivation layer. The behaviour of test samples as detectors is examined during the characterisation measurements with the radioactive source (Cesium-137). The possibility of including surface passivation with Al2O3 into the fabrication of the detectors is examined by studies of the effect of ALD processing on the behaviour of a flip-chip bonded single pixel sensor. The aim of this chapter is to suggest algorithms of characterisation and illustrate them with collected data.

Chapter IV introduces: the main concepts of reliability as a discipline; the specificity of the applications for scientific projects; approaches for testing the long-term stability of detector test structures; and the need for background regarding changes in the characteristics of the samples over time. The aim of this chapter is to demonstrate the possibility of adopting industrial reliability tests to the development process of silicon detectors

Chapter V summarizes the results mentioned in Chapters 2–4. The conclusion that additional surface passivation with Al2O3 can be introduced into the processing of silicon detectors for the improvement of their characteristics and long-term stability is discussed in Chapter V and summarised in Chapter VI. The aim of this chapter is to define the conclusion and discuss it.

8

I. Background of the research

Development of particle detectors starts from the analysis of their future applications, destination, and project needs. The description of the detector structures studied in this thesis starts from the overview of their dedicated destination, the CMS Tracker at the Large Hadron Collider (LHC) [1]. This chapter provides an overview of the CMS experiment at LHC [2].

Changes in the Tracker design are shown to illustrate challenges associated with different stages of the upgrades.

1. CMS experiment at Large Hadron Collider (LHC)

The LHC is the largest and most powerful particle accelerator in the world. It consists of a 27-kilometre ring of superconducting magnets with a number of accelerating structures to boost the energy of the particles along the way. Two high-energy particle beams are accelerated close to the speed of light before they will collide at the experiments and produce an array of particles. Produced particles are analysed by several detecting systems arranged into multilayer constructions at the experiments [1, 2].

The Compact Muon Solenoid (CMS) is one of the experiments at LHC situated at one of the four collision points. It is a general-purpose detector with a broad physics programme ranging from studying the Standard Model (including the Higgs boson) to searching for extra dimensions and particles that could make up dark matter.

Figure 1 CMS detector [2]

9 The most significant part of the CMS detector is a huge solenoid magnet generating a uniform magnetic field of 3.8 tesla confined by a steel yoke (Figure 1). The beam collision point in the centre is surrounded by several layers of detecting systems based on different detecting principles to enable the measurement of particles with different properties, and, hence, with different mechanisms of interaction with matter.

The silicon tracker situated in the heart of the CMS closest to the beam collision point is exposed to the highest level of radiation, and with the planned updates to LHC its operational conditions will become even harsher. Thus, the requirements for the upgraded silicon tracker are defined by operational conditions. Research in fundamental particle physics is based on information provided by the detectors at powerful accelerators, and with every new physics discovery the requirements for the detecting systems, as well as for the colliding machines, are becoming increasingly advanced. Complicated systems require more complex reliability analysis on different levels from single sensors to the complete detector system during all R&D and development stages.

The update of the LHC will turn it into the High-Luminosity Large Hadron Collider (HL- LHC) with extended potential for discoveries [3]. The luminosity of HL-LHC is planned to exceed the luminosity of the LHC, and the experiments will have to adjust their capabilities to process the respectively increased amount of data [4]. Figure 2 shows the timeline of projected changes in luminosity and planned maintenance. In addition, many critical components of the accelerator should be replaced as they will reach the end of their lifetime due to radiation damage [5].

Figure 2 Projected LHC performance through 2038, showing preliminary dates for long shutdowns (LS) of the LHC and projected luminosities [6]

2. CMS Tracker

CMS Tracker is a complicated array of silicon particle sensors and readout electronics. It is the heart of the CMS experiment and its reliability requirements are at the highest possible level, with the aim to collect data with 100% efficiency. The initial design of the silicon tracker was developed with requirements derived from the expected performance of the LHC.