Download Document

Detectors
1. Accelerators
2. Particle detectors overview
3. Tracking detectors
Why do we accelerate particles ?
(1) To take existing objects apart
1803 J. Dalton’s indivisible atom
atoms of one element can combine with atoms of other
element to make compounds, e.g. water is made of
oxygen and hydrogen (OH)
1896 M. & P. Curie find atoms decay
1897 J. J. Thomson discovers electron
1906 E. Rutherford: gold foil experiment
Physicists break particles by shooting other
particles on them
Why do we accelerate particles ?
(2) To create new particles
1905 A. Einstein: energy is matter
E=mc2
1930 P. Dirac: math problem predicts antimatter
1930 C. Anderson: discovers positron
1935 H. Yukawa: nuclear forces (forces between
protons and neutrons in nuclei) require pion
1936 C. Anderson: discovers pion muon
First experiments used cosmic rays that are
accelerated for us by the Universe
are still of interest as a source of extremely energetic
particles not available in laboratories
Generating particles
Before accelerating particles, one has to create
them
electrons: cathode ray tube
(think your TV)
protons: cathode ray tube
filled with hydrogen
It’s more complicated for other particles (e.g.
antiprotons), but the main principle remains
the same
Basic accelerator physics
Lorentz Force:
F = qE + q(vB)
magnetic force: perpendicular to velocity, no
acceleration (changes direction)
electric force: acceleration
Accelerators: Cockroft-Walton
A (series of) voltage gap(s)
Maximum energy of a single gap is 200 kV,
limited by discharge
CW accelerator at Fermilab: 750 kV
Accelerators: Van de Graaf
Van de Graaf generator: an electrostatic
machine which uses a moving belt to
accumulate very high voltages on a hollow
metal globe
1: metallic sphere
2: electrode connected to 1
3: upper roller
4: belt (positive side)
5: belt (negative side)
6: lower roller
7: lower electrode (ground)
8: spherical device, used to
discharge the main sphere
9: spark
Surfing the electromagnetic wave
Charged particles ride the EM wave
create standing wave
use a radio frequency cavity
make particles arrive on time
Self-regulating:
slow particle  larger push
fast particle  small push
Surfing the electromagnetic wave
How to create a standing wave ?
Klystron (S. & R. Varian)
electrons flow into cavity, excite eigen modes
creates standing electromagnetic waves
A similar device (magnetron) found in your
microwave oven
325 MHz Klystron for Proton Driver Linac (Fermilab)
Cyclotron
1929 E.O. Lawrence
The physics: centripetal force
mv2/r = Bqv
Particles follow a spiral in a constant magnetic field
A high frequency alternating voltage applied between Delectrodes causes acceleration as particles cross the gap
Advantages: compact design (compared to linear
accelerators), continuous stream of particles
Limitations: synchronization lost as particle velocity
approaches the speed of light
the world largest cyclotron
at TRIUMF (520 MeV protons)
Synchrotron
The idea: both magnetic field strength and
electric field frequency are synchronized with
the traveling particle beam
particle trajectories confined to a thin vacuum
beamline  no large magnets, expandable
synchrotron radiation limits its use for electrons
Currently, accelerators of this type provide
highest particle energies in the world
Summary on accelerator types
Electrostatic accelerators
acceleration tube: breakdown at 200 keV
Cockroft-Walton: improves to 800 keV
AC driven accelerators
linear: cavity design and length critical
circular accelerators:
cyclotron: big magnet, non-relativistic
synchrotron: vacuum beamline, expandable, small
magnets and cavities
synchrotron radiation large for light particles
Hadron vs electron colliders
electron
proton
Point-like particle
yes
no
Uses full beam energy
yes
no
Transverse energy sum
zero
zero
Longitudinal energy sum
zero
non-zero
Synchrotron radiation
large
small
Large Electron-Positron collider
Location: CERN (Geneva, Switzerland)
accelerated particles: electrons and positrons
beam energy: 45104 GeV, beam current: 8 mA
the ring radius: 4.5 km
years of operation: 19892000
Tevatron
Location: Fermilab (Batavia, IL)
accelerated particles: protons and anti-protons
beam energy: 1 TeV, beam current: 1 mA
the ring radius: 1 km
in operation since 1983
Large Hadron Collider
Location: CERN (Geneva, Switzerland)
accelerated particles: protons
beam energy: 7 TeV, beam current: 0.5 A
the ring radius: 4.5 km
scheduled start: 2007
Future of accelerators
International Linear Collider: 0.53 TeV
awaiting directions from LHC findings
political decision of location
Very Large Hadron Collider (magnet
development ?): 40200 TeV
Muon Collider (source ?) 0.54 TeV
lepton collider without synchrotron radiation
capable of producing many more Higgs particles
compared to an e+e collider
Conclusions
Motivation for particle acceleration
understand matter around us
create new particles
Particle accelerator types
electrostatic: limited energy
AC driven: linear or circular
Modern accelerators
TeVatron, LHC
accelerators to come: ILC, VLHC, muon collider…
Detectors
1. Accelerators
2. Particle detectors overview
3. Tracking detectors
Detectors and particle physics
detectors allow one to detect particles 
experimentalists study their behavior
new particles are found by direct observation or by
analyzing their decay products
theorists predict behavior of (new) particles
experimentalists design the particle detectors
Overview of particle detectors
What do particle detectors measure ?
spatial location
trajectory in an EM field  momentum
distance between production and decay point  lifetime
energy
momentum + energy  mass
flight times
momentum/energy + flight time  mass
Natural particle detectors
A very common particle detector: the eye
detected particles: photons
sensitivity: high (single photons)
spatial resolution: decent
dynamic range: excellent (11014)
energy range: limited (visible light)
energy discrimination: good
speed: modest (~10 Hz, including processing)
Photographic paper
1895 W. C. Röntgen: sensitivity to high energy
photons (X-rays) invisible to the eye
working medium: emulsion
Properties:
detected particles: photons
sensitivity: good
spatial resolution: very good
dynamic range: good
no online recording
no speed resolution
The Geiger counter
1908 H. Geiger
passing charge particles ionize the gas
ions (electrons) drift towards cathode (anode)
cause an electric pulse, can be heard in a speaker
Properties:
detected particles: charged particles (electrons, ,…)
sensitivity: single particles
spatial resolution: none (detector size) – can be fixed
dynamic range: none – can be fixed
speed: high (determined by charge drift velocity)
The cloud chamber
1911 C. T. R. Wilson (1927 Nobel Prize)
the first tracking detector (tracking=many spatial
measurements per particle)
Principle of operation:
an air volume is saturated with water vapor
pressure lowered to generate super-saturated air
charge particles cause saturation of vapor into small
droplets  can be observed as a “track”
photographs allow longer inspection
The cloud chamber
Properties:
detected particles: charged particles (electrons, ,…)
sensitivity: single particles
spatial resolution: excellent
dynamic range: good
as particle slows down, droplets occur closer to each
other
if placed inside a magnet, can observe curled trajectories
speed: limited (need time to recover the supersaturated state)
Photographic emulsions
Rarely used in modern experiments due to
principal restrictions:
cannot be read out electronically
used to need a lot of technicians looking at photographs
by eye – inefficient, boring, and error prone
today using pattern recognition software (think OCR)
cannot be used online
One advantage is excellent spatial resolution
(<1 m)
Were used in the -neutrino discovery
(DONUT, 2000)
Modern detector types
 Tracking detectors
 detect charged particles
 principle of operation: ionization
 two basic types: gas and solid
 Scintillators
 sensitive to single particles
 very fast, useful for online applications
 Calorimeters
 measure particle energy
 usually measure energy of a bunch of particles (“jet”)
 modest spatial resolution
 Particle identification systems
 recognize electrons, charged pions, charged kaons, protons
Tracking detectors
 A charged track ionizes the gas
 10—40 primary ion-electron paris
 multiplication 3—4 due to secondary ionization
 typical amplifier noise 1000 e—
 the initial signal is too weak to be effectively detected !
 as electrons travel towards cathode, their velocity increases
 electrons cause an avalanche of ionization (exponential increase)
 The same principle (ionization + avalanche) works for
solid state tracking detectors
 dense medium  large ionization
 more compact  put closer to the interaction point
 very good spatial resolution
Calorimetry
The idea: measure energy by total absorption
also measure location
the method is destructive: particle is stopped
detector response proportional to particle energy
As particles traverse material, they interact
producing a bunch of secondary particles
(“shower”)
the shower particles undergo ionization (same
principle as for tracking detectors)
It works for all particles: charged and neutral
Electromagnetic calorimeters
Electromagnetic showers occur due to
Bremsstrahlung: similar to synchrotron radiation,
particles deflected by atomic EM fields
pair production: in the presence of atomic field, a
photon can produce an electron-positron pair
excitation of electrons in atoms
Typical materials for EM calorimeters: large
charge atoms, organic materials
important parameter: radiation length
Hadronic calorimeters
In addition to EM showers, hadrons (pions,
protons, kaons) produce hadronic showers due
to strong interaction with nuclei
Typical materials: dense, large atomic weight
(uranium, lead)
important parameter: nuclear interaction length
In hadron shower, also creating non detectable
particles (neutrinos, soft photons)
large fluctuation and limited energy resolution
Muon detection
Muons are charged particles, so using tracking
detectors to detect them
Calorimetry does not work – muons only leave
small energy in the calorimeter (said to be
“minimum ionization particles”)
Muons are detected outside calorimeters and
additional shielding, where all other particles
(except neutrinos) have already been stopped
As this is far away from the interaction point, use
gas detectors
Detection of neutrinos
In dedicated neutrino experiments, rely on
their interaction with material
interaction probability extremely low  need huge
volumes of working medium
In accelerator experiments, detecting neutrinos
is impractical – rely on momentum
conservation
electron colliders: all three momentum components
are conserved
hadron colliders: the initial momentum component
along the (anti)proton beam direction is unknown
Multipurpose detectors
 Today people usually combine several types of
various detectors in a single apparatus
 goal: provide measurement of a variety of particle
characteristics (energy, momentum, flight time) for a variety
of particle types (electrons, photons, pions, protons) in
(almost) all possible directions
 also include “triggering system” (fast recognition of
interesting events) and “data acquisition” (collection and
recording of selected measurements)
 Confusingly enough, these setups are also called
detectors (and groups of individual detecting elements
of the same type are called “detector subsystems”)
Generic HEP detector
D detector at Fermilab
D detector is one of two large multipurpose
detectors at Fermilab (another one is CDF)
name = one of six intersection points
D: fairly typical HEP detector
D: tracking system (1)
Vertex detector: Silicon Microstrip Tracker
four layers of silicon detectors intercepted with
twelve disks + (recent addition) Layer 0
D: tracking system (2)
Outer tracking detector: Central Fiber Tracker
sixteen double layers of scintillating fibers
D: calorimeter
Liquid argon / uranium calorimeter, consisting
of central and two end calorimeters
D: outer muon system
The outermost part of the detector, surrounds
the whole thing
Proportional Drift Tubes, Mini Drift Tubes
Central (Forward) muon SCintillators
D: other elements
Magnet: a central solenoid magnet (2 T) and
outer toroid magnet
Luminosity scintillating counters
Central and forward preshower
Forward proton detector (Roman pots)
Data acquisition, trigger system, …
Conclusions
Particle detectors follow simple principles
detectors interact with particles
most interactions are electromagnetic
imperfect by definition but have gotten pretty good
crucial to figure out which detector goes where
Three main ideas
track charged particles and then stop them
stop neutral particles
finally find the muons which are left
Detectors
1. Accelerators
2. Particle detectors overview
3. Tracking detectors
Gas detectors
As a charged particle crosses a gas volume, it
creates ionization
electrons get kicked out of atoms
the rest of atom becomes electrically charged (ion)
In absence of external field, ions and electrons
recombine back to neutral atoms
electrons drift to anode
ions drift to cathode
E = V/r ln(b/a)
Ionization
Affected by many factors
gas temperature
gas pressure
electric field
gas composition
Important parameters:
ionization potential
mean free path
Some gases eat up electrons (“quenchers”)
Ionization as a function of energy
Ionization probability gas dependant
General features:
threshold (~20 eV)
fast turn on
maximum (~100 eV)
soft decline
eV
Mean free path
Average distance an electron travels before it
hits an atom – determined by gas density
At ambient pressure (1013 hPa), air density is
2.71019 molecules/ccm, and mean free path is
68 m
At high vacuum (10—3…10—7 hPa), mean free
path is 0.1…1000 m
What happens after ionization ?
After collision, ions (electrons) thermalize and
travel until neutralized through electron (ion),
wall, negative ion (other molecule)
Mean free path for electrons ~4 times longer
than for ions
Ions diffuse slowly, electrons diffuse quickly
Diffusion velocity depends on gas
Avalanche
 Steps of an avalanche:
 a primary electron proceeds towards the anode,
experiencing ionizing collisions
 due to the lateral diffusion, a drop-like avalanche,
surrounding the wire, develops
 electrons are collected during ~1 ns
 a cloud of positive ions slowly migrates towards the cathode
Ionization chamber
Low voltage, no secondary ionization – just
collect ions
example: smoke detector
radiation source (Am-241) emits -particles
they pass through ionization chamber, creating current
smoke absorbs -particles and interrupts current
Proportional counter
Higher voltage, tuned to provide proportional
regime:
each avalanche is created independently from
others  total amount of charge created remains
proportional to the amount of charge liberated in
the original event, which in turn is proportional to
the particle’s kinetic energy
Spark chamber
Device similar to Geiger counter
Ionizing particles produce sparks along its way
that can be photographed and used later for
reconstruction of tracks
My diploma work was done on the ITEP’s 3m
magnet spectrometer equipped with spark
chambers
Regimes in a tracking chamber
Gas tracking detectors: summary
detector
voltage
avalanches
regime
ionization
chamber
low
no
single ion
collection
proportional
counter
medium
isolated
proportional
Geiger-Müller
counter
high
maximal
saturated
Multi Wire Proportional Chamber
1968 G. Charpak (1992 Noble Prize)
the idea: make a proportional counter with a lot of
anodes placed between two cathode planes
by looking at which wires were fired, can determine
position of the particle
if the proportional mode is used, can determine
particle’s energy + improve position resolution (by
interpolation)
drift chambers: measure time of arrival of the
electron avalanche  improve position resolution +
provide a timing reference point
MWPC electric field
Homogeneous field away from anode wires
Field near wires very sensitive to their position
from G. Charpak’s Noble lecture
MWPC design
Constraints
precise position measurements require precise and
small wire spacing
homogeneous fields require small wire spacing
large fields require thin wires
geometric tolerances cause gain variations
Geometry and problems
required precision: sub millimeter
long chambers need strong wires (W/Au plated)
and high tension to minimize sagging
Choice of gas
It’s a magic
low working voltage
high gain operation
good proportionality
high rate capability
long lifetime
fast recovery
price
…
Operation conditions
Pressure: slightly above atmospheric
avoid incoming gas “pollution”
a large tracker is not really air tight
not too high (difficult to maintain)
Temperature: slightly lower than room t.
avoid large temperature gradients
affected by environment (e.g. cooling of nearby
systems)
Limitations of chambers
High occupancy: OK
used in Alice (heavy ion collisions at LHC)
Radiation hardness
tough but manageable (need gas flow)
Speed
is a problem for LHC applications (25 ns bunch
crossing)
ion drift is limiting factor
can be addressed with special technologies (GEM)
Time Projection Chamber (RHIC)
Brookhaven Nat’l Lab, Relativistic Heavy Ion
Collider
Shown: Gold-Gold collision
Solid state detectors
Basic operation principle same as gas detectors
gas
liquid
solid
Density
low
moderate
high
Atomic number
low
moderate
moderate
Ionization energy
moderate
moderate
low
Signal speed
moderate
moderate
high
Silicon detectors
Solid state tracking detectors: semiconductor
diodes with reverse bias
normally there is no current (except very low “dark
current”)
a charged particle creates a track of carriers
(electron-hole pairs) along its way  charge pulse
Why silicon ?
Low band gap width: 1.12 eV (large number of
charge carriers / unit energy loss)
Energy to create an e/h pair: 3.6 eV (an order of
magnitude smaller than ionization energy for
gases)
high carrier yield
low Poisson noise
no gain stage required
better energy resolution and high signal
Why silicon ? (cont’d)
High density and atomic number
reduced range of secondary particles
can build thin detectors
better spatial resolution
High carrier mobility
typical charge collection times <30 ns
no slow component (ions)
Excellent mechanical rigidity
Industrial fabrication techniques
Detector and electronics can be integrated
Problems
Cost
proportional to area covered
most of the cost is moving to read out channels
Material budget
for complex detectors can be as large as ~1—2
radiation lengths
affects calorimeters behind the detector
affects tracking accuracy (multiple scattering)
Typically need cooling to reduce leakage
current (thermal energy = 1/40 eV)
Radiation hardness
What is it ?
particles damage silicon crystal structure
band gap decreases
leakage currents increase
gain drops
detector looses efficiency and precision
What to do ?
exchange detectors
ATLAS: replace inner detector after 3 yrs of operation
switch to radiation hard technology (e.g. diamonds)
Diode strip detectors
Idea (1980’s): divide the large-area diode into
many small strip-like regions and read them
out separately
Typical strip pitch p = 20—few hundred m
Position measurement precision:
digital readout:  = p/12
analog readout:  = p/(S/N) (S = signal, N = noise)
-function
Let a particle pass the detector between two
strips (i) and (i+1) at coordinate x = xi…xi+p
If strip (i) collects charge qi, and strip (i+1)
collects charge qi+1,
(x) = qi/(qi+qi+1)
ideally, (x) = 1, x<xi+p/2, and (x) = 0, x>xi+p/2
in practice, it’s not true:
finite charge cloud size (~5 m)
charge capacitance between strips
non-uniform electric field
Lorentz shift
If a detector is placed in magnetic field (parallel
to its strips), charge careers are deflected as
they drift towards the strips
introduces systematic shift of the measured position
signal gets spread between several strips
increases cluster sharing (bad)
with analog readout, improves spatial resolution (good)
Double sided readout detectors
Idea: use both types of carriers to make two
position measurements for the same amount of
material
n-side charge
Usually cross strips  2-dim measurement
From charge correlation can resolve ambiguities
p-side charge
Pixel detectors
Provide 3-dim points with very high precision
main issue is readout
can read out individual pixels or entire
rows/columns
Electrodes are close !
low full bias
low collection distance
no charge spreading
fast charge sweep out
Pixel vs strip detector operation
SiO2 +ve
-ve
+ve
-ve
-ve
-ve
p+
h+
h+
e-
n
E
W3D
pixel detector
E
e-
n+
+ve
strip detector
W2D
Pixel detector at ATLAS
Conclusions
Tracking detectors
detect charged particles
measure arrival time and charge deposition
derive 3 dimensional location and energy
Design
inner detectors: silicon (strip/pixel), highest track
density resolution (tens of m)
outer detectors: gas detectors, lower resolution
(hundreds of m)