Development of Advanced Gaseous Detectors for Muon Tracking and Triggering in Collider Experiments Liang Guan 1,2 16-10-2014 Dissertation Defense Supervised by Prof. Xiaolian Wang 1 , Prof. Zhengguo Zhao 1 and Prof. Junjie Zhu 2 1 University of Science and Technology of China 2 University of Michigan
81
Embed
Development of Advanced Gaseous Detectors for Muon ... · layer of high resistive sheet directly on the ... • Parallel ionization multiplier (PIM): multiple mesh layers and multi
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Development of Advanced Gaseous Detectors for Muon
Tracking and Triggering in Collider Experiments
Liang Guan 1,2
16-10-2014 Dissertation Defense
Supervised byProf. Xiaolian Wang1, Prof. Zhengguo Zhao1 and Prof. Junjie Zhu2
1 University of Science and Technology of China2 University of Michigan
Standard model Dark matter Super symmetryLiang Guan ([email protected]) Dissertation Defense 16 October 2014 2
• Collider experiments, utilizing high energy accelerators and large spectrometers, are unique to discover new particles, resonances, phenomena … Changing our understanding of fundamental building blocks of the nature and their interactions
• Very broad physics topics: Standard Model, SUSY, Extra dimension etc…
• Hunting for heavy new particles relies on capturing high momentum secondary particles from their decays via different channels. (decay to muons is one of the important channel. e.g. H ZZ* 4µ)
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 5
• In the context of ATLAS muon upgrade program, we performed extensive studies on three advanced gaseous detectors for muon tracking and trigger in future collider experiments.
• Studies are carried out on Micro-mesh gaseous structure (Micromegas) (Part I), Resistive plate chamber (RPC) (Part II) and Thin gap multi-wire chamber (TGC) (Part III).
• Researches are focused on addressing critical issues of applying these detectors for muon detection in harsh high rate environment and understanding their basic characteristics which affect the timing and tracking.
• Research approaches: simulation, calculation, lab and beam tests …
Garfield, Magboltz, HEED: electron transportation ...Ansys Maxwell 3D, neBEM: electric field
X-ray source• Attempting an alternative ways to make Micromegas spark-tolerant: attaching a thick layer of high resistive sheet directly on the anode
• Standard Micromegas is vulnerable to discharge when highly ionizing particles are present. Coating a thin resistive layer on the metal anode (separated by thin insulating layer and grounded at its periphery).
• High resistive materials studied
» High gain» Lateral charge spread in thick
resistive layer» Spark-protection
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 12
• “Spark” signals recorded directly using 50 Ω terminated oscilloscope:
Gain
• Spark amplitudes: mostly < 100 mV; up to 0.5 V at high gain, rate less than ~10-4
per detected photon count • Spark signal duration : < 200 ns• Charge released: less than few nC; mostly at few tens of pC effective to protect front-end
electronics
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 14
• Parallel ionization multiplier (PIM): multiple mesh layers and multi-stage of avalanches. Originally intended to be used for tracking low energy beta rays.
• Attempt to design a structure and operate PIM at GEM-mode: only electrons extracted to the bottom induction gap. fast signal, fast timing!
• Prompt electron signal component:
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 15
Development of a novel method to fabrication Micromegas: thermal bonding. Good energy resolution, basic performances parameters comparable to bulk Micromegas. Reasonable gain uniformity (< 20%).
Many simulation studies for optimizing thermo-bonded Micromegas with woven wire mesh.
Attempts made to develop high resistivity anode Micromegas for spark tolerance High resistivity material significantly reduce discharge amplitude.
Attempts made to develop fast timing parallel ionization multiplier (PIM) Analysis of prompt charge concentration; experimentally assesses the viability to operate PIM at “GEM-mode”.
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 18
Introduction- Proposed fast tracking trigger system based on RPC
• Thin gap RPCs for excellent timing, fine pitch readout strips for precision coordinate measurements, dual end readout to make RPCs as mean-timers and also for second coordinate measurement
• Ideas need to be assessed Beam tests
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 22
• Charge centroid (1.27-mm-pitch):- Resolutions: ~ 200 µm (Best) and ~ 220 µm (Average)- Limited by the nonlinear charge representation and small cluster size (~2.3)
• Using timing information only (1.27-mm-pitch): - Hit position determined to be the average center of strips in the cluster- Resolutions < 290 µm Useful for fast tracking at trigger level. (ATLAS NSW requires 300 µm online resolution per detector layer)
Precision Coordinate Spatial Resolution
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 28
• 2nd Coordinate measurements• Hit position determine from signal arrival time difference from two ends of a strip• Resolution: ~1 cm (w/ 100 ps resolution TDCs) and ~ 7 mm if averaging the
reconstructed positions from multiple strips.
Mean-time and 2nd Coordinate
v ~15 cm/ns
• Mean-time: Average signal arrival time from two ends of a strip measured to be independent of hit position
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 32
We proposed a fast tracking trigger scheme based on thin gap RPCs.
Several beam tests to study thin gap RPC on-line/off-line spatial resolution, timing etc. possibility to construct (sub-ns x sub-mm x sub-cm) trigger logic cells. It will be very powerful to handle muon triggering in an unexpected high rate environment.
Rate capability tests: critical to reduce delivered charge per count. RPC with 1 mm gaps and 1010 Ω∙cm resistive electrodes can be fully efficient to muons at > 18 kHz/cm2.
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 35
Similar structure as TGC in the present end-cap muon system except:
• Strip readout (3.2 mm x 1-2 m): precision measurement of track position in η• Pad segmentation (8 cm x 8 cm): fast pattern recognition for selecting readout strips• Lower cathode resistivity (~1 MΩ/ 100 kΩ/)
Introduction – sTGC for NSW
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 38
• Full simulation of single-layer time spectrum: take into account the gain fluctuation, electronics and reference detector time jitters > 95% events within 25 ns
• Multiplayer timing performance improves by shifting wire positions wrt. adjacent layers minimize the probability of passing low eclectic field region
0 degree
(With arbitrary offset)
2.85 kV
Single layer time spectrum Time spectrum from 3/4 coincident
with shifted wires
Timing – Fully Simulated Time Spectrum
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 45
• Integration over time gives the collected charge on each strip
Important parameters to determine the charge spread:Resistive paint (cathode) resistivityStrip-cathode coupling capacitor (distance, dielectric constant)Electronics integration timeStrip width and pitch
sTGC Charge Sharing (cont.)
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 48
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 50
Part III summary
Various simulations of sTGC electric field, electron transportation etc.
sTGC timing capability full-fills ATLAS NSW LV-1 trigger requirement (LHC 25 ns BX discrimination). Does not degrade with reduced HV, mag. field in NSW.
We have built an analytic model to describe charge dispersion in resistive layer and charge sharing among strips (still in developing):
• Could be implemented in ATLAS main software framework for sTGC digitization
• Very crucial to understand the impact of resistive paint non-uniformity to off-line muon reconstruction accuracy
Liang Guan ([email protected]) Dissertation Defense 16 October 2014 52
Summary
• Extensive R&D on three gaseous detectors are carried out to explore theircapabilities for muon tracking and triggering in collider experiments.
• We developed a novel way of constructing Micromegas. Detailed simulation andexperimental studies are performed. Attempts are made to construct highresistivity anode Micromegas and parallel ionization multiplier to address sparkissues and to improve parallel mesh structure detector timing capability.
• Our studies indicate RPCs are capable of providing (sub-ns x sub-mm x sub-cm) high granularity logic trigger cells Powerful for rejecting backgrounds and improving muon selectively
• Simulation study suggest sTGC is capable to preform LHC BX discrimination inNSW environment. We developed an analytical charge sharing model for betterunderstanding the detector characteristics.
• “Studies on Fast Trigger and High Precision Tracking with Resistive Plate Chamber”, 2013 CPAD and Instrumentation Frontier Community meeting, Argonne, IL, USA (口头报告)
• “Development of 1 mm low resistivity Bakelite Plate for thin-gap Resistive Plate Chamber”, RPC 2012 Conference, Frascati, Italy (会议海报)
• “Simulation Studies of Characteristics and Performances of sTGC for ATLAS Muon New Small Wheel”, 2013 US ATLAS Workshop Upgrade Session, Argonne, IL, USA (口头报告)
8 oral talks in total in national and international conferences
Backup slides
hardware based trigger that searches for high transverse momentum leptons, photons, jets and large missing and total transverse energy. Reduce 40MHz rate to 75kHz.
ATLAS Level -1 Trigger
dN_mu vs PT
Hot roll and hot press
Hot roll and hot press
Trans. diffu. coefficient
• Typical Fe-55 spectrum
Characteristics of Mmegas in Argon based mixture Ar/CO2 93:7
• Gas Gain
• Electron transparency • (Fe-55) Energy resolution vs. Gain
Gain ~104
Vm [V]
Gai
nFW
HM[%
]
Ea/Ed Vm [V]
“knee” @80
Uniformity correction
PIM fast signal calculation
PIM bottom mesh electron extraction coefficient
Prompt charge signal from 1.15 m gap
Beam test (w/ NINO) trigger logic
RPC time resolution (w. MRPC as reference)
Magnetic field around the SW
Earliest Cluster Arrival Time Distributions
Degree 0 Degree 5 Degree 10 Degree 15
Degree 20 Degree 25 Degree 30 Degree 35
Degree 40 Degree 45 Degree 50
Earliest Cluster Arrival Time vs. Track Hit Position
Degree 0 Degree 5 Degree 10 Degree 15
Degree 20 Degree 25 Degree 30 Degree 35
Degree 40 Degree 45 Degree 50
Algorithm for timing determination
• Trigger: n out of 4 coincidence
• Timing tag is given after nth latest response of a layer
Layer 1
Layer 2
Layer 3
Layer 4
Time Trigger!
Layer 1
Layer 2
Layer 3
Layer 4
Time Trigger!
3/4 coincidence time spectrum – single layer efficiency effect
• For instance, wire displacement = 0.5 mm
Tail disappears as efficiency approaches 100%
From J. Dubbert’s Talk @ ATLAS muon week 27th,March,2013
sTGC big sector layout
• sTGC big sector is subdivided into 4detection areas in azimuthal direction.
• Maximum wire length ~1 m
• Signal propagation velocity: 27ns/cm 3.7 ns arrival time difference for 1m
3/4 coincidence time spectrum – including additional jitters
• Signal propagation jitter
• Electronics jitter
• Single layer time spectrum measured with SonyASD/VMM+TDC well reproduced assuming a total external time jitter of 3 ns
• The subtracting ~2ns jitter from largereference scintillator : ~2.3 ns
𝜌𝜌(𝑥𝑥, 𝑡𝑡) = 𝑅𝑅𝑅𝑅/4𝜋𝜋𝑡𝑡𝑒𝑒(−𝑅𝑅𝑅𝑅4𝑡𝑡 𝑥𝑥2)𝐷𝐷 𝑥𝑥 =
𝑄𝑄2𝜋𝜋𝜎𝜎
𝑒𝑒(− 𝑥𝑥22𝜎𝜎2)
𝜌𝜌′ 𝑥𝑥, 𝑡𝑡 = 𝐷𝐷(𝑥𝑥) ⊗𝜌𝜌(𝑥𝑥, 𝑡𝑡) =𝑄𝑄
2𝜋𝜋(2 𝑡𝑡𝑅𝑅𝑅𝑅 + 𝜎𝜎2)
𝑒𝑒(− 𝑥𝑥2
4 𝑡𝑡𝑅𝑅𝑅𝑅+2𝜎𝜎
2)
⇒ Charge density on readout strip
Initial charge spread Dispersion on resistive layer
Induced charge distribution on strip
Charge sharing
𝜌𝜌 𝑡𝑡 = 𝑤𝑤1
𝑤𝑤2𝜌𝜌′ 𝑥𝑥, 𝑡𝑡 𝑑𝑑𝑥𝑥 =
𝑄𝑄2
[𝐸𝐸𝐸𝐸𝐸𝐸𝑅𝑅𝑅𝑅
4𝑡𝑡 + 2𝑅𝑅𝑅𝑅𝜎𝜎2𝑤𝑤2 − 𝐸𝐸𝐸𝐸𝐸𝐸(
𝑅𝑅𝑅𝑅4𝑡𝑡 + 2𝑅𝑅𝑅𝑅𝜎𝜎2
𝑤𝑤1)]
Charge density of strip (w1,w2) vs. time
Strip width/pitch: 2.7/3.2mm
RC: 25 ns
σ: 1.12 mm
The charge sharing among strips depends on:• Dispersion time constant RC
• Charge integration time
Charge sharing
Validation: Garfield + Psipice
Initial charge spread:• Built wire chamber model and segmented
cathode to 1.5 mm pitch strips. Charge on cathode is the superimposition from each strip
Charge diffusion through resistive layer:• Built equielent1D RC network in Pspice