Transcript Higher gain can be obtained in Argon Isobutane Gas

Performance Studies of BULK Micromegas
with Different Amplification Gaps
Purba Bhattacharya1, Sudeb Bhattacharya1, Nayana Majumdar1,
Supratik Mukhopadhyay1, Sandip Sarkar1, Paul Colas2, David Attie2
1
Applied Nuclear Physics Division, Saha Institute of Nuclear Physics, Kolkata, India
2
DSM/IRFU, CEA/Saclay, Gif-sur-Yvette CEDEX, France
Motivation

Micromegas – promising candidate for TPCs including ILD main tracker

Bulk Micromegas – built using printed circuit board fabrication techniques




Important parameters that determine choice of a particular bulk over another are
detector gain, gain uniformity, energy and space point resolution, comfortable
operating regime (in terms of voltage settings, signal strength etc), stability and
ageing characteristics (ion back-flow), capability to efficiently pave large readout
surfaces with minimized dead zone (due to spacers) …
These parameters are known to depend on geometry of the detector (amplification
gap, mesh hole pitch, wire radius etc), electrostatic configuration within the detector,
gas composition, pressure …
Systematic comparison of different bulk Micromegas has been carried out to weigh
out various possibilities and options and guide our choice for specific applications
Comparison with numerical simulations using Garfield has been performed to verify
the mathematical models and confirm our understanding of the device physics
BULK MICROMEGAS
Details of BULK Micromegas:
 10x10 cm2 active area
 Amplification gap: 64 m, 128
m, 192 m and 220 m
 Stainless steel mesh, wire
diameter 18 m, pitch 63 m/
78 m
 Dielectric Spacer, diameter 400
m, pitch 2 mm
Microscopic view of Bulk Micromegas
Mesh Hole ~ 45m
Pitch ~ 63m
Spacer Diameter ~ 400m
Spacing between two spacer ~ 2 mm
Experimental Set Up
Gas
Cylinder
Purification
System
Residual
Gas Analyzer
RGA Spectrum for fixed Argon –
Isobutane Gas Mixture
Pressure
Gauge
Gas Flow Out
Gas Mixing
System
Gas Flow In
Gas Chamber
& Detector
Pre-Amp
(Model No. –
142IH)
Oscilloscope
Typical MCA Spectrum of 55Fe
Amplifier
(ORTEC 672)
Power Supply
(High Voltage)
(N471A)
Filter
MultiChannel
Analyzer
Numerical Simulation
Simulation tools
Radiation Source
Garfield framework: to model and
Ionization
Drift Volume
simulate two and three dimensional
drift chambers
Drift and Diffusion
of Electrons
Ionization: HEED
Drift and Diffusion: MAGBOLTZ
Amplification: MAGBOLTZ
Amplification Gap
Transfer Gap
Amplification and
further Diffusion
Potential, Field: neBEM
(nearly exact Boundary Element
Method)
Readout Pads
Pad Response
Signal
Garfield + neBEM + Magboltz + Heed
Variation of Electric Field
(a) With Mesh Hole Pitch
(Wire Diameter: 18 m)
(b) With Amplification Gap
(Please note, Y-Axis is in log scale)
In each case detector characteristics
(gain, resolution…) changes
accordingly
Gain : G = Nt / Np = kP/ Np , where
Nt Total number of electrons
Np  Primary electrons
k  Constant, depends on Preamplifier, Amplifier, MCA specification
P  Peak Position
(Maximum allowable
voltage: Sparking limit)
Variation of gain with amplification field in different argon-based gas mixture (drift field 200 V/cm)
Higher gain can be obtained in Argon Isobutane Gas Mixture
Variation of gain with amplification
field for three different amplification
gap – higher gain can be obtained
with larger amplification gap leading
to a comfortable operating regime
(Maximum allowable
voltage: Sparking limit)
Variation of gain with amplification
field for two different pitch – for
larger pitch, sparks start at higher
field and so a higher value of gain
can be obtained
Comparison with Simulation Results
Trend similar in case of both detectors→ Simulated results considerably lower without
Penning
Roles of different parameters : Penning Transfer Mechanism → Increase of gain, Needs
further investigation on transfer rates
The simulated gain in other gas and other gap also agrees quite well with experimental data
Energy Resolution : R = P/P, where
p  r.m.s. of the pulse height distribution
P  peak position
Variation of energy resolution at 5.9 keV with gain in different argon-based gas mixture
At this drift field, at higher gain, Argon Isobutane gas shows better energy resolution
Variation of energy resolution with
gain for three different amplification
gap – 128 m shows better resolution
Variation of energy resolution
with gain for two different pitch
– 63 m shows better resolution
Comparison with Simulation
Numerical estimates follow trend of measured data
Gain variation and electron transparency needs further investigation
Similar trend observed in other cases also
Estimation of Electron Transparency
 Fraction of electrons arriving in
amplification region
 Depends on field ratio, drift voltage
 Depends on hole-pitch
Every electron collision is connected with red
lines,
 inelastic collisions  excitations  ionizations.
Experiment :
Ratio of signal amplitude at a given Edrift over signal amplitude at Edrift where gain is maximum
Simulation:
 Microscopic tracking of electrons from randomly distributed points (100 m above mesh)
 Two different models for mesh modelling: one dimensional thin wire segments for Edrift < 100
V/cm and three-dimensional polygonal approximation of cylinders for Edrift > 100 V/cm
Experiment:
Variation with electron transparency with field ratio for three different amplification gap
At this pitch value, the electron transparency reaches maximum value at much higher drift field
The larger gap detector reaches maximum value at lower drift field in comparison with smaller
gap
Comparison of Experimental Data with Simulation Results
(Amplification Gap: 64 m and 192 m; Pitch: 63 m)
Simulation Results agree quite well with Experimental Data
Calculation with higher pitch (78 m) is in progress
Ion Backflow
 Secondary ions from amplification region drift to drift region
 Distortion of electric field; degrades stable operation of detector
 Micromegas micromesh stops a large fraction of these ions
 Backflow fraction : Nb/Nt  (1/FR)(p/t)2 where
Nb  average number of backflowing ions
Nt  average total number of ions
FR  field ratio,
p  pitch of the mesh,
t  diffusion
Avalanche of Electrons (2D picture)
Drift of Secondary Ions (2D picture)
Simulation of IBF: Variation with IBF with Field Ratio
a) for different argon based gas mixture
(Amplification gap: 128 m)
b) for three different amplification gap
(Ar:Isobutane 90:10)
 Preliminary simulation results
show expected trends
 Need further investigation and
experimental verification
Experimental Set Up:
Preliminary data was taken at
CEA, Saclay
We are trying to build up a
similar set up at SINP
Value of IBF follows the theoretical
prediction
Besides the contribution of ions
from avalanche, additional
contribution from ions between
drift plane and test box window
affect the data – implementation of
2nd drift mesh – improvement of
results
Effect of Spacer (Diameter 400 m, Pitch 2 mm, Amplification Gap 128 m)
Electric field in axial direction through different holes
Without Spacer
With Spacer
 Spacers cause significant
perturbation resulting in
Drift lines
increased field values,
and
particularly in the regions where
Avalanche
cylinders touch the mesh
 Electron drift lines get distorted
near the dielectric spacer
Electron Transparency and Gain (Without and With Spacer)
Without Spacer
With Spacer
Position of track
above mesh
25 m
50 m
100 m
25 m
50 m
100 m
Electrons crossing
mesh
97.794
97.304
97.549
97.549
95.343
95.833
Electrons reaching
middle of
amplification area
97.794
97.304
97.549
54.902
92.892
95.343
Gain
600
594
596
338
570
584
 Electrons are lost on the spacer
resulting in reduced gain
 Signal strength reduces and it has a
longer tail
Signal
Summary
Experiments and numerical simulations carried out using different bulk Micromegas
(amplification gaps 64 m, 128 m, 192 m, 220 m; Pitch 63 m, 78 m) in several
argon based gas mixtures
Important detector parameters such as gain, energy resolution, transparency estimated
Observed conflicting advantages of different parameters, e.g., configuration that leads
to higher gain and more stable operation (amplification gap 220 m) provides less
attractive energy resolution
Smaller pitch (63 m ) found to be generally more useful
Preliminary calculation of ion back flow compare favorably with measurements
Effects of spacers on gain and signal indicated significant changes occurring around
the spacer
Successful comparisons with simulation indicate that the device physics is quite well
understood and suitably modeled mathematically
Acknowledgement
1. We acknowledge CEFIPRA for partial financial support
2. We thank our collaborators from ILC-TPC collaboration for their help and suggestions
3. We acknowledge Rui de Oliveira and the CERN MPGD workshop for technical support
4. We happily acknowledge the help and suggestions of the members of the RD51
collaboration
5. We are thankful to Abhik Jash, Deb Sankar Bhattacharya, Wenxin Wang for their help in
some measurement and Pradipta Kumar Das, Amal Ghoshal for their techinal help
6. We thank our respective Institutions for providing us with necessary facilities