Transcript 252b Lecture 7: Tracking: Proportional and Drift Chambers

Tracking Detectors:
Proportional and Drift Chambers
How we measure particle momenta & ID
(in some cases): 2 types of tracking detectors
Prof. Robin D. Erbacher
University of California, Davis
References: R. Fernow, Introduction to Experimental Particle Physics, Ch. 9, 10
D. Green, The Physics of Particle Detectors, Ch. 8
K. Kleinknecht, Ch. 3
http://pdg.lbl.gov/2004/reviews/pardetrpp.pdf
Radiative  Energy Loss
Optical behavior of medium:
characterized by complex dielectric constant
Sometimes instead of ionizing an atom or exciting matter,
the photon can escape the medium: (transition, C, etc).
Nuclear Interactions
We have seen that the electromagnetic interaction is
responsible for:
• Energy loss of charged particles
• Small angle scattering of charged particles
• Production and interaction of photons
In some processes, the nuclear interaction dominates:
• Particle creation reactions
• Interactions at high energy / large momentum transfer
• Interactions of neutral particles other than photons
Nuclear cross section:
N ~ A
2
3
2
p
Goes as A2/3 times Compton
wavelength2 of proton
Brem. cross section:
 B ~ Z 
2
2
e
Goes as Z2 due to coherence, 3
penalty due to 3 vertices
For A~3(Li), they’re comparable,
for A>3, X0(Brem) dominates


Strong Interactions
Hadrons are subject to the strong interaction in matter.
The neutron: ideal probe since no real em interactions.
For attenuation in matter, the absorption cross section is:
abs = T - el - q
Here el is coherent elastic scattering off a nucleus, and
q is quasielastic scattering from individual nucleons.
Absorption lengths of particles:
N 
A
N A abs
For particle production, T(pp, pn)~ln(s), T(n-nucl)~A2/3.
The e+e- or e-nucleon initial state can couple to hadronic
final state via a virtual , and
 apart from resonances, the
e+e- hadronic production cross section falls off as s-1.
Strong Interactions
• What is λn? The mean distance (in cm, or g/cm2)
a particle travels in a material before
encountering a nucleus
• Nuclei act as hard-sphere scattering centers
• Cross section is proportional to nuclear radius
squared, or Z2/3
Nuclear interaction
length:
lead ~ steel: 17 cm
material
X0 (g/cm2)
λn (g/cm2)
H2
63
52.4
Al
24
106
Fe
13.8
132
Pb
6.3
193
Weak Interactions
The 2nd fundamental nuclear interaction: weak interaction.
Important for  interactions, decays of quasistable particles.
Neutrinos are direct probes of the weak interaction, but do
not interact with matter very easily.
Weak interaction can play role in charged lepton
interactions: does not preserve parity. Thus, signatures
separating weak from em interaction are parity-violating.
Asymmetries: polarized e-N scattering (SLAC) and muon
pair production in e+e- collisions at PETRA.
Two major classes of weak interaction known:
Charged current: e-nucleon via W e (lepton # cons.)
Neutral current: e-nucleon via Z0, final state lepton neutral: e
Tracking:
Proportional Counters
and Drift Chambers
Charged Particle Tracking
• Two main types: gas wire chambers, silicon
• (We heard about silicon last time)
• Innermost detectors: want precise trackingSi!
• Outer detectors: silicon too expensive!
• Basic design: ionization chamber with HV sense
wire:
amplification of
103 - 105 in high
field near wire
Ionization Wire Chambers
Wire Chambers: Probably the most commonly used
detection devices in high energy physics experiments.
The Basics of Wire Chambers:
• Charged particles travels through a volume of gas
• Gas is ionized by the passage of the particle
• Ionization drifts & diffuses in an electric (and magnetic)
field toward an electrode
• Collection and amplification of anode signal charge, and
charge induced on cathode creates detectable signals
• Measurement of points on trajectory determines p
Processes in Gases
When a charged particle passes
through gases subject to an E field, it
loses energy by:
• Elastic scattering (small)
• Excitation: gas atoms/molecules
excite then deexcite by  emission
• Ionization (most important)
Ionization: One or more electrons are
liberated from atoms of the medium,
leaving positive ions and electrons.
Energy imparted to atom exceeds
ionization potential of gas.
Ionization Potentials of Vapors
-Clifton G. Found, G.E.
Phys. Rev. 16, 41-53 (1920)
©1920 The American Physical Society
Ionization
potential
i  A(V  V0 )
3
2
Langmuir's Eqn

Primary Ionization Potential
Ionization Potentials of Vapors -Clifton G. Found, G.E.
Phys. Rev. 16, 41-53 (1920)
©1920 The American Physical Society
Secondary and Total Specific Ionization
Average energy lost in creating ion pair ~30 eV.
Primary ionization: number of ionizing collisions per unit
length for the incident particle. (Poisson distribution)
Secondary ionization: some of the electrons ( rays) have
energy to cause more ionization.
Total specific ionization: total # ions actually created per
unit length. (Landau distribution: long tails)
Deexcitation: Further ionization from deexcitation of metastable atoms (Penning effect). Example:
Ne* + Ar  Ne + Ar+ + ePeak fraction of E lost can double by adding 0.1% Ar gas.
Number of Ions v. Voltage
Simplest case:
Parallel plate
capacitor
Ionization chamber: Voltage increased such that
the charge arriving on plates = charge formed
Proportional region: Initial electrons accelerated
enough to ionize more; avalanche pulse
proportional to primary ionization; reaches ~108
After Ionization, What Next?
Once ion pairs are created, many processes can occur:
Recombination, charge exchange, attachment, absorption.
Want electrons/ions to recover after signals have been read out.
Recombination of positive/negative ions to neutrals:
X+ + Y-  X + Y + 
Electrons can be removed by
recombination as well:
X+ + e-  X + 
Add gas w/ large electron affinity:
energy difference of lowest (ground)
state of neutral and lowest state of
corresponding negative ion.
X + e-  X- + 
Water vapor, O2, ethanol, SF6, CCl4, freon
Charge Exchange
Another way to eliminate positive ions is through
Charge Exchange.
Ionization potential of the ion is greater than some molecule
mixed with the gas, usually polyatomic gas like ethanol or
methylal.
Gas quenches the ion multiplication by neutralizing ions of
the main chamber gas: dissipates ionization energy by
dissociating into smaller fragments and absorbs s emitted
in radiative deexcitation process.
This is called a quenching gas.
Proportional Chambers
Consider a simple cylindrical proportional
tube of outer radius b at voltage V0 and
inner (wire) of radius a at voltage zero.
a
V0  2 ln(b / a), V(r) = V0
b
V=0
Electric field inside the chamber:
E  2 /r,
V0
ln(r / a)
V0
, E(r)=
ln(b / a)
r lnb / a
Charged particle -> ionization. e- move toward anode. High fields near
wire->mulitiplication of e-s by collisions: at small r the energy gain can
exceed ionization potential. Runaway process, like avalanche in PMTs.
Gas gain: Change dN(r) in # electrons at r in multiplication process, &
depends on first Townsend coefficient : inverse distance b/w collisions
dN(r)  N(r)dr, N(r)  N0er ,  ~ 1/ L
ion
Mean free path
for ionization:
L
ion
Typical Gas gain~105. For > r~20 pairs, or 108, Geiger region: runaway!
Pulse Formation
In the ionization chamber region, we have a gain of unity. For these field
strengths, the dominant signals are due to electrons.
For proportional chambers, primary ionization electrons are multiplied
due to high fields near wire, E(r)~1/r. Electrons collect on anode, ions
remain, ‘sheathing” wire, moving slowly toward cathode.
Moving chargesenergy changescapacitively-induced signal, dV.
Suppose multiplication takes place at N radii away r~Na. Voltage on
anode due to e- and ions: V- and V+. Let qs be source charge.
The induced voltage due to e- and ion motion is:
Na
b
qs
+
dV 
E(r)dr,
V   dV , V   dV 
a
Na
CV0
q ln(N )
q ln(b / Na)
V-  s
,
V+  s
 V C ln(b / a)
C ln(b / a)
Signal due to ions dominates, as they travel all the way to the cathode.
Drift Velocity and Mobility
A gas with molecules of mass M has a thermal velocity vT:
vT  3kT M
For N2, vT~0.1 cm/ (Earth escape v~1.1 cm/s). For electrons, vT~4-40
cm/, below the binding energy of atoms.
Wire chambersoperate in an electric field: consider the drift velocity for
electrons and ions in a uniform electric field. Let the mean time
between collisions, , be L /vT , and the acceleration be a~eE/M. Then
vd
eE L
~ a ~
M vT
 free path is proportional to the collision cross
Using that the mean
1
section: L  N0 / Awe express the drift velocity in terms of mobility.
Mobility: drift velocity per unit reduced electric field E/.
vd

eE 
A /N 0  E
~

M
vT

Here
vd

E

Multiwire Proportional Chambers
The MWPC was invented by Charpak
at CERN. Principle of proportional
counter is extended to large areas.
Stack several wire planes up in
different direction to get position
location.
1992 Prize
Avalanche developing
Precision Drift Tracking
• Next idea: stack up
proportional wire drift tubes,
measure time of arrival of the
ionization pulse
• Find track from tangents to
circles
• Can get about 150 μm
position resolution
• But: too much material!
Large Area Drift Chambers
The “open cell” drift chamber uses field and sense wires:
field wires create shape of electric field, sense wires detect
time of arrival of pulse.
This is the design of the CDF drift
chamber - the cells are tilted to take
into account ExB drift!
Drift Chambers
There is an unequivocal correlation of the time difference t
between the particle passage and the rising edge of the
anode signal with the distance between the point of primary
ionization and the anode wire in a proportional chamber.
t0=initial ionization,
t1=avalanche near wire begins
z

t1
t0
v D (t)dt (where z is drift pat h)
If the drift velocity is vD a constant, it becomes linear.
Example: With constant vD~55 mm/, a time measurement of 4 ns can
 m! (Typical resolutions for MWPC~0.5 mm)
give spatial resolution ~200
To keep vD constant, keep field along direction of path constant.
Impossible to do with MWPC layout! Zero field between anode wires.
Change geometry: add field wire -HV between two anode wires +HV:
+HV2
-HV1
Drift region
Negative potential
Drift region
Negative potential
-HV1
Chamber Geometries
Planar: Large planar detectors resolution limited by spatial
positioning of wires, and gravity sagging. ~100-300 m.
Smaller limited by TDC resolution (>40 m) and e- diffusion.
Cylindrical: Four main types are proportional, cylindrical, jet,
and TPC (time projection chamber). Used in colliding beam
experiments, in conjunction with solenoidal fields.
The regular cylindrical is open celled, whereas the jet
chamber has radial partitions. The TPC is unique in that the
electrons drift along the E and B field lines, and z coordinate
is more accurate than the rest.
Proportional: These are the familiar (mainly rectangular)
counters with anode wires sandwiched between cathode
sheets. Simple, but not as good resolution without cathode
strip readout.
Chamber Geometries
Proportional: wires between planes
Jet Drift Chamber
+HV
+
+
+
+
+
+
Cylindrical Drift Chamber
-HV
E
E
B
Time Projection Chamber
+
+
+
+
+
+
+HV