Study of straw tube trackers for the comet experiment

29 2.11 The drift line of electrons in the straw tube tracker... We can see that theyhave nice agreement between measurement and simulation for muon decay spectrum.They can get only two

Study of Straw Tube Trackers for The COMET Experiment

Master of Physics Nguyen Minh Truong

Department of Physics, Graduate School of Science

Osaka University, Japan

July 31st 2012

“ First of all, I would like to express my sincere thanks to people who supported

me in the development of my thesis I’m grateful for all the support and motivation

of my supervisor from the Osaka University, Prof Y.Kuno He allowed my attend

to his group and left me to myself with his proper advice Thank you for providing

me the brain power to study in physics I also thank Prof S Mihara and Prof

H Nishiguchi for their education of experimental basis for me I appreciate theirinterest in my work, as well as their advice and patience as this thesis evolved Theywas always available when I had questions and I especially would like to thank themfor their clear, efficient and fast help whenever it was needed I also was supported

by many friends in Kuno group I am grateful to them At last, I would like to thank

my family for their unlimited support ”

1.1 Introduction 1

1.2 Theoretical Motivation 2

1.3 µ− → e− conversion 6

1.4 Overview of The COMET Experiment 12

1.5 Pulsed Proton Beam 14

1.6 Muon Stopping Target 15

2 Straw Chamber 17 2.1 Requirement for the COMET tracker 17

2.2 Gas study 20

2.3 Principal operation of the straws tube tracker 27

3 Study of Prototype Straw Tube Tracker 35 3.1 Overview of Prototype Straw Tube Tracker 35

3.2 Study the sag of the wire 40

3.3 Gas gain study 43

3.4 Efficiency of straw tube tracker 46

3.5 Study the drift velocity of straw tube tracker 52

3.5.1 r t calibration 52

3.5.2 Drift velocity of straw tube tracker 55

3.6 Position resolution 58

List of Tables

1.1 Requirement of the pulsed proton beam [2] 141.2 Lifetimes and relative strength of µ−- e−conversion process in differentmuon stopping material [2] 161.3 Parameter of muon stopping target [2] 162.1 The drift times of various gas mixtures at a nominal operational voltage 21

List of Figures

1.1 History of charge LFV searches[9] 41.2 Diagram of neutrino’s mass contribute to µ− → e− transition[2] 51.3 Electron spectrum, normalized to the free-muon decay rate Γ0 Thesolid blue line is for carbon, the black dotted line for aluminum, thegreen dot-dashed line for silicon and the red dashed line for titanium[11] 81.4 End-point region of the electron spectrum for aluminum The squarescorrespond to the spectrum with recoil effects The triangles is thespectrum neglecting recoil The right plot is a zoom for Ee > 100MeV, the solid (dashed) line on this plot corresponds to the Taylorexpansion around the end-point with (without) recoil [10] 91.5 Cross section through SINDRUM II connected to the PMC magnet [12] 10

1.6 Recent results by SINDRUM-II Momentum distributions for three dif-ferent beam momenta and polarities: (i) 53 MeV/c negative, optimized for µ− stops, (ii) 63 MeV/c negative, optimized for π− stops, and (iii)

48 MeV/c positive, optimized for µ+

stops The 63 MeV/c data were scaled to the different measuring times The µ+

data were taken using

a reduced spectrometer field [12] 11

1.7 The COMET experiment schematic layout of the muon beam line and the detector [2] 13

1.8 Bunched proton beam in a slow extraction mode [2] 15

2.1 Straw tube tracker perpendicular with charge particle in the COMET experiment [1] 18

2.2 Layout of straw tube layer [1] 19

2.3 Drift velocity in different Argon and Ethan mixture [19] 22

2.4 Drift velocity of Argon and Ethan was simulate by Garfield 23

2.5 Diffusion of Argon and Ethan was simulate by Garfield 25

2.6 Townsend of Argon and Ethan was simulate by Garfield 26

2.7 Two layer of Kapton over-woven straw 27

2.8 Construction of straw tube tracker [1] 27

2.9 Electric field distribution in straw [21] 28

2.10 Contours electric field in straw is simulated by Garfield 29

2.11 The drift line of electrons in the straw tube tracker 30

2.12 Cross section as the function of electron energy in Ar gas [22] 32

2.13 The different drift line of electron in case no magnetic field (left) and

magnetic file (right) 33

2.14 electron drift line and cluster in straw tube with magnetic field 34

3.1 The design of prototype 36

3.2 Straw tracker (left) and 7 straws are placed in the two layer as show in the right 37

3.3 Design of the prototype end plate of the prototype 38

3.4 The gas line to flow the gas mixture to the straw chamber [23] 39

3.5 Electric read out the signal from wire anode 40

3.6 Principle of the sag of the wire 41

3.7 Garfield simulation the sag of the wire 41

3.8 Compare the sag of the wire in Garfield simulation and calculation 42

3.9 The setup experiment for the gas gain study 43

3.10 Logic trigger for gas gain study 43

3.11 QDC histogram for the gas gain study 44

3.12 Mean QDC value as a function of the high voltage of straw 44

3.13 The amplification gain of the preamplifier and the ASD buffer 45

3.14 The gas gain of straw tube tracker 45

3.15 Image of fiber [24] 46

3.16 8 fibers are put close together 47

3.17 Experiment setup for hit efficiency 47

3.20 β energy spectrum of Sr90 source 49

3.21 Energy deposit in straw tube tracker by simulation 50

3.22 Energy deposit spectrum on straw tube with events passed through straw tube and fiber 51

3.23 Energy deposit spectrum on straw tube with events passed through straw tube and fiber 51

3.24 Setup experiment for drift velocity study 53

3.25 Time distribution of straw tube tracker 54

3.26 r t calibration for the upper straw tube tracker 55

3.27 Setup experiment for drift velocity study 56

3.28 rt relation of straw tube trackers 57

3.29 Fit r-t relation with linear function order 6th 57

3.30 Residual distribution of straw tube tracker 59

3.31 Effect of multiple scattering on straw tube tracker 59

Chapter 1

COMET EXPERIMENT AT

J-PARC

1.1 Introduction

In the Standard Model of elementary particle physics, we assume that neutrino mass

is zero and Lepton-Flavour Conservation is built However, from the fact that the trinos can change its flavours, neutrino oscillation was observed at Super-Kamiokande

neu-in 1998 [1], we know that neutrneu-inos with different flavours are mixed and have zero mass Therefore, Lepton-Flavour Conservation is known to be violated at least inthe neutral lepton sector However, Charged-Lepton-Flavour violation has not beenobserved yet experimentally

non-There have been some experiments on the world trying to observe the Charged

Muon to Electron Transition) at J-PARC also tries to explorer the flavour Violation process The main purpose of the COMET experiment is to searchfor coherent neutrino-less conversion of muons to electrons (µ− - e− conversion), inthe reaction µ− + N (A, Z) → νµ + N (A, Z 1), with the sensitivity of 10−16 TheCOMET will use a bunched proton beam, slowly-extracted from J-PARC Main Ring.

proper-• CLFV can have sizable contributions from new physics and thus can manifestthemselves in future experiments

• CLFV gives no sizable contribution in the Standard Model unlike the FCNCprocess of the quarks; such contributions give serious background events and

limit the sensitivity to new physics.

When the muon was discovered the first time in 1937 by Neddermeyer and Anderson[3], people has a big question ”Can a muon convert into neutrino and electron?” It

is know that muon’s mass is about 200 times heavier than the electron’s and the lifetime is about 2.2 µs, but this fact was not consistent to Lthhypothesis that the muon

is responsible to β decay Then, it was considered that muon is an excited state ofelectron and it could be decay into an electron and photon The first experimentalsearch for charge lepton flavor violation was done by Hincks and Pontecorvo in 1947[4] But they can not get the positive result and set the upper bound for the brachingratio of muon decay to electron process less than 10%

Another hypothesis is that muon can decay into three particles was suggest bystudying the continuous electron spectrum in 1948 [5] One of them is electron andthe others are neutrinos However, at that time, neutrino was assumed zero massand the idea of lepton flavor was not predicted And then, the experiment searchfor muon lepton flavor violation have be carried out such as neutrinoless µ− - e−

conversion process and µ− → e− + γ But all of the results were negative and theyset the strong limit on these branching ratio This lead to the idea that there aretwo kinds of neutrino [6, 7] This idea was conform by experiment at BrookhavenNational Laboratory (BNL) [8] It leads to there are three lepton flavor, muon flavor,electron flavor and tau flavor This idea finally was added to the Standard Model.From late of 19th century, the searches for LFV with various elementary particles

can see that the experimental upper limits have been continuously improved during

50 year since the first LFV experiment by Hincks and Pontecorvo and the upper limitnow is 10−13

Figure 1.1: History of charge LFV searches[9]

Although LFV searches carried out the negative result, the neutrino experimentwith the discovery of neutrino oscillation has show that neutrino has mass and leptonflavor violated with neutrino species [1] Thus, Standard Model have to modifiedneutrino mass in their calculation and it also conclude that LFV can be occur This

lead that new physics beyond the Standard Model can be find at TeV scale Thisscale is expected cLFV experiments and the COMET experiment is one of them.Although CLFV has never been observed by experimental It is know that a littleextended SM, which already include small mass of neutrino in their calculation, ispredict that cLFV is very small to observed One example is that the the predictionfor Br(µ− → e−γ) < 10−54 is given by the graph in figure 1.2

Figure 1.2: Diagram of neutrino’s mass contribute to µ− → e− transition[2]

Thus, the discovery of LFV, the search for µ− → e− conversion would imply newphysics beyond the SM Now, all of new physics hypothesis beyond the StandardModel predict LFV at some level For example, supersymmetric (SUSY) models,extra dimension models, little Higgs models, models with new gauge Z bosons, modelswith new heavy leptons, lepto-quark models, etc Each of them gives a prediction forflavor changing neutral currents (FCNC), including cLFV Thus, the search for µ−

1.3 µ− → e− conversion

There are some methods to find proposed cLFV such as µ−- e− conversion, µ− → e−

γ and µ− → eee, but the most spectacular is coherent neutrino-less conservation ofmuons to electrons (µ−- e−conversion), µ−+ N (A, Z) → e−+ N (A, Z), because thisprocess has better limits expected with current technology than other process When

a negative muon is stopped in a target, it is caught by an atom and a muonic atom

is formed After muon cascade down to lowest energy level in the muonic atom, it isbound in muonic 1s ground state The destiny of muon is then decay in orbit (µ− - e−

νµ νe) or capture by a nucleus, namely µ− + N(A,Z) → νµ + N(A,Z-1) In case wesuppose the physics beyond the Standard Model, there is another possibility process,which is the neutrino-less muon capture process µ− + N(A,Z) → e− + N(A,Z) Thisprocess is name µ− - e− conversion in a muonic atom In this process, lepton flavornumbers Lµ and Le are not conserved, while the total lepton number L is conserved.The final state of the nucleus N(A,Z) can be either the excited state or the groundstate If the final nucleus is the excited state, it will transit to the ground state, which

is called the coherent capture Since all of the nucleons participate in the process,the rate of the coherent capture over non coherent capture is increasing by the factor

of the number of nucleons in the nucleus

The coherent µ− - e− conversion event signature in a muonic atom is a energy single electron, which is emitted from the conversion with an energy Eµe ∼

mono-mµ - Bµ Here, mµ is the mass of muon and Bµ is the binging energy of the 1smuonic atom

There are some reason why µ− - e− conversion are attracting First of all, the

energy of electron is about 105 MeV, it is so far the end-point energy of the muondecay spectrum Secondly, because of mono-energy of electron signal, no coincidencemethod required Finally, there is a potential to improve sensitivity by using a highrate muon beam for µ− - e− conversion process, it will be an advantage to otherprocess, such as µ+

→ e+ + γ and µ+

→ e+e+e+ process

Although energy of the signal electron is 105 MeV and far from end-point of themuon decay spectrum, there are several other potential sources of electron backgroundevents in the energy region around 100 MeV originating either from beam particles

or cosmic rays was be used As for beam related background, the background eventsmay originate from muons, pions and electrons in the beam; background also maycome from muon decay-in-orbit (DIO) or radiative muon capture (RMC) Pions alsocan produce background events through radiative pion capture (RPC) Gamma raysfrom RMC and RPC produce electrons mostly through pair production inside thetarget

The muon decay-in-orbit is studied the first time by C E Porter and H Primakofe

in 1951 [13] On 60 years continuous, several theory is given to study this problem In

1974, R D Violier et al [14] gave expressions to calculate the electron spectrum whichallow including relativistic effects in the muon wavefunction, the Coulomb interactionbetween the electron, and the nucleus, and a finite nuclear size Late, in 1982 and

1997, Oshanker [15, 16] did study the high-energy end of the electron spectrum, andhis results allow us quickly estimate of the muon decay-in-orbit contribution to the

the background for the COMET experiment The figure 1.3 presents the end-pointregion of the electron spectrum for some materials target and the figure 1.4 is theirresult for the an aluminum target (the intended target in the COMET experiment).

Figure 1.3: Electron spectrum, normalized to the free-muon decay rate Γ0 The solidblue line is for carbon, the black dotted line for aluminum, the green dot-dashed linefor silicon and the red dashed line for titanium [11]

The latest search for µ− - e− conversion was done by SINDRUM II at PSI Inthis experiment, they used the beam momentum 50 MeV/c which contains the samenumber of µ− and π− bombard to the gold target The figure 1.5 show the crosssection of SINDRUM II [12]

Figure 1.4: End-point region of the electron spectrum for aluminum The squarescorrespond to the spectrum with recoil effects The triangles is the spectrum neglect-ing recoil The right plot is a zoom for Ee >100 MeV, the solid (dashed) line on thisplot corresponds to the Taylor expansion around the end-point with (without) recoil[10].

The SINDRUM II ’s result is showed in the figure 1.6 We can see that theyhave nice agreement between measurement and simulation for muon decay spectrum.They can get only two events outside of background region, but they do not stay inthe interest region, the region of signal for µ− - e− conversion Maybe these eventscome from cosmic rays or RPC Finally, they set the current upper limit on B( µ− +

Au → e− + Au) < 7 x 10−13 [17]

Figure 1.5: Cross section through SINDRUM II connected to the PMC magnet [12]

Figure 1.6: Recent results by SINDRUM-II Momentum distributions for three ent beam momenta and polarities: (i) 53 MeV/c negative, optimized for µ−stops, (ii)

differ-63 MeV/c negative, optimized for π− stops, and (iii) 48 MeV/c positive, optimizedfor µ+

stops The 63 MeV/c data were scaled to the different measuring times The

µ+

data were taken using a reduced spectrometer field [12]

1.4 Overview of The COMET Experiment

The COMET experiment aim to search for coherent neutrino-less conservation ofmuon to electron in a muonic atom The COMET experiment will use the new

µ− source, transport and detector design, all of them was described in the CDR ofCOMET [2]

To increase the sensitivity by a facctor 10.000 over the current limit, there are someimportant feature was insert on the COMET experiment [2] First of all is the highlyintense muon source: To achieve the 1016

experiment sensitivity, the total number ofmuons needed is the order of 1011

Therefore, we need to produce the highly intensemuon beam line There are two method was applied in the COMET experiment

to increase the muon beam intensity One of them is to use the high proton beampower Another one is use the pion collecting system with high efficiency With thepion capture system, 8.5 1020

protons at 8 GeV will be used to obtain the number ofmuon in order 1011

Secondly is the pulsed proton beam: Around 105 MeV region of electron signalfrom µ−-e− conversion, there are many background events One of them comes frombeam-related background events To suppress that background events, the COMETexperiment will use the pulse proton beam with beam extinction system Because ofthe life time of muonic atoms in order 1 micro-second, a pulsed beam with beam backbucket, which are short compare with the muonic life time, will remove the beambackground events

Finally is curved solenoid for charge and momentum selection: The capture pionsdecay to muons, which are transported with high efficiency through a superconducting

solenoid magnet system Because of producing electron background events in the 100MeV energy region of beam particles with high momentum, the curved solenoid wasused to eliminated these background.

Figure 1.7: The COMET experiment schematic layout of the muon beam line andthe detector [2]

1.5 Pulsed Proton Beam

In order to reduce the background from the beam-related background events, pulseproton beam with beam extinction system is used in the COMET experiment More-over, the time of separation of the pulsed proton beam need to be around 1 µs, thistime corresponds to the lifetime of the muon in the muonic atom Therefore, thepulse muons beam with 1 µs time separation is produce After muon bombard to themuon stopping target, the electrons signal will be emitted and come to the detectorduring the interval time of the proton pulse Furthermore, to achieve the expectedsensitivity of the COMET experiment, the residual protons between pluses needs to

be smaller than the number of protons in the main pulse around 109 times That isthe reason why the COMET experiment decide to use the beam extinction system.The table 1.1 give some requirement of the pulse proton beam and the the figure show

a typical time structure for the pulse proton beam in the COMET experiment

Table 1.1: Requirement of the pulsed proton beam [2]

Average Current 7 µABeam emittance 10 πmm.mradProton per Bunch <1011

Bunch Separation 1 ∼ 2 µsBunch Length 100 ns

Figure 1.8: Bunched proton beam in a slow extraction mode [2]

1.6 Muon Stopping Target

According to the COMET experiment design, the muon stopping target have mize muon stopping efficiency and the minimize energy loss of the µ− - e−conversionelectrons, the material of target is Aluminum and it is placed in the Muon TargetSolenoid, it also connect to the pion capture solenoid by curved muon transportsolenoid Why Aluminum is choose to be the material of muon stopping target in theCOMET experiment? As describe above, the time window of the experiment will beopened about 700ns after the primary proton pulse Since the high Z target will haveshort lifetime of muonic atomic, the Z target should not be too high The muoniclifetime of Aluminum is 880 ns [12] is satisfies with the time window experiment.Moreover, the braching ratio of µ− - e− conversion process, B(µ− + N → e− + N)

maxi-for muon stopping target

Table 1.2: Lifetimes and relative strength of µ− - e− conversion process in differentmuon stopping material [2]

Aluminum Titanium Lead

Table 1.3: Parameter of muon stopping target [2]

Item SpecificationMaterial Aluminum

Disk radius 100 mmDisk thickness 200 mNumber of disks 17Disk spacing 50 mm

Chapter 2

Straw Chamber

2.1 Requirement for the COMET tracker

The purpose of the electron tracker for the COMET experiment is to measure thehelical trajectory of electron in a solenoid magnetic field Muons stopping in a targetproduce electrons or positrons, it will depend on the physical process These chargedparticles will move in a uniform magnetic field (1T) [1] which follow helical trajecto-ries A straw tube tracking detector is used to measure the particle trajectory andfrom these information we can determine the charge particles momentum The trackerconsists of straw tubes, arranged transverse to a uniform magnetic field, as described

in the figure 2.1 The straw tube will measure the drift time which correspond to theeach straw tube radial distance at the closest approach of a charged particle from thesense wire To suppress the multiple scattering, the tracker is operated in vacuum

Figure 2.1: Straw tube tracker perpendicular with charge particle in the COMETexperiment [1]

experiment sensitivity First of all, the efficiency of straw tube tracker to obtainthe position information might be reduced since the detector was operated in highoccupancy rate environment Secondly, accidental overlap of electron backgrounddecay in orbit would mimic the event with an energy of the signal electron (105MeV) Finally, there is a contribution of the superposition of lower energy signals

in the trigger detector [1] To solve these problems, the straw tube tracker shouldhave diameter small enough and time resolution high enough Moreover, we need toincrease segmentation, optimizing the geometric design and have pulse shaping andtiming appropriate of straw tube signal

In the design of the COMET experiment, there are 5 super layers of a simplestraw tube tracker and each straw tube tracker consists of X, X’, Y and Y’ planes

to provide X and Y information as in the figure 2.2 [1] According to the conceptualdesign report of the COMET experiment, there are some requirement for the strawtube tracker:

• Straw tube tracker should withstand for the total rate of charged particle up to

800 kHz

• Straw tube tracker should withstand for the total flux of gamma up to 8 MHz

• The momentum resolution of straw tube tracker much less than 1 MeV/c

Figure 2.2: Layout of straw tube layer [1]

2.2 Gas study

Since charge particles are produced in very high rate after muons bombard to thetarget, one of the requirements for the straw tube tracker is ability separation betweenadjacent tracks However, the drift distance is relatively long, the diffusion of gas instraw tube tracker can limit the separation of detector Therefore, we should considerthe gas with small diffusion constant

In order to detect two events separately, another requirement for straw tubetracker is that the drift velocity should be around 5 cm/s Therefore the candi-date gas for straw tube tracker do not only have drift velocity around 5 cm/s butalso have low diffusion in order to operate in both a high rate and high multiplicityenvironment The table 2.1 below show the drift time of various gas mixtures at anominal operational voltage [1]

According the COMET experiment design, the straw tube tracker is necessary towork in the magnetic field 1(T) along the axis of the solenoid It is known that theradial drift velocity will be reduced in the magnetic field, for the case of Ar and C2H6

mixture the velocity decreases by only 15% In addition, the COMET experimentintend to limit the gas gain approximately 5 x 104

, which can be easily achieved with

Ar and C2H6 mixture Therefore, Ar and C2H6 mixture is one of candidate gas forthe experiment And the drift velocity of Argon and Ethan mixture was studied from

1980 [19] Figure 2.3 shows the drift velocity of Argon and Ethan in various mixture

Table 2.1: The drift times of various gas mixtures at a nominal operational voltage

Gas Velocity (cm/µs) Drift time (ns)

According to the result of Garfield simulation, we will have a drift velocity ofAr/C2H6 of 5 cm/s at electric field from 500 to 1000 V/cm region, which is consis-tent with measurement data show in Figure 2.3 Moreover, one another importantrequirement for the drift velocity gas of straw tube tracker is the stable operationcondition surely at high particle rates The excited atoms of an Ar gas will returns

to the ground state by emitting photons which have energy corresponding to the

ion-Figure 2.3: Drift velocity in different Argon and Ethan mixture [19]

Figure 2.4: Drift velocity of Argon and Ethan was simulate by Garfieldhappening, usually we will use the Ar and C2H6 mixture to absorb the emitted pho-tons and produce the stable region of drift gas According to the result in figure 2.4,

we can select the stable operation region of drift gas That is electric field from 30 to

50 kV/cm region Thus the voltage operate for straw tube tracker should around 1.5kV

Furthermore, as discussing above, the diffusion of gas in straw tube tracker isone of the limit separation parameter of detector and we should consider the gaswith small diffusion constant The result from GARFIELD simulation show us theproperty of Ar and C2H6 gas mixture One of the property of gas mixture is diffusion

Another problem which we should consider when selecting the gas for the strawtube tracker is the Townsend avalanche effect It is a gas ionization process, atthe beginning a very small amount of free electrons are accelerated by a sufficientlystrong electric field, give rise to electrical conduction through a gas by avalanchemultiplication This effect will damage the straw tracker, so we should not operate toomuch the high voltage for anode wire According the result of GARFIELD simulation

in figure 2.6, the Townsend avalanche effect will appear when we apply the highvoltage at 1x105

V/cm Inclusion, the COMET experiment will choose the Ar andC2H6 gas mixture for their requirement and the apply voltage for the anode wire isaround 1.5 kV

Figure 2.5: Diffusion of Argon and Ethan was simulate by Garfield

Figure 2.6: Townsend of Argon and Ethan was simulate by Garfield

2.3 Principal operation of the straws tube tracker

The basic of the gas detector is explained by F Sauli [21] In the COMET experiment,

a straw tube tracker is a conducting straw tube with assuming 5 mm diameter whichbuilds the cathode of the system It is made from a polyimide film (Kapton), twolayer of Kapton are over-woven straw as the figure 2.7 Each of layer is approximate1.2 µm thickness and the glue thickness between two Kapton layer will make theKapton layers approximate 3 µm The construction of straw is show schematically infigure 2.8

Figure 2.7: Two layer of Kapton over-woven straw