In order to measure the longitudinal polarization of the 3-6 GeV high intensity TJNAF electron beam, a
Compton Polarimeter was built by CEA Saclay,
The Compton polarimeter is now running since february 1999 and has been used by severals HALL A
experiments. Two collaborators from Saclay have been awarded for this apparatus :
Christian Cavata awarded the Prix Joliot-Curie 2000 of the Société Française de
Physique and Maud Baylac awarded the SURA 2000 Thesis Prize.
| Compton Polarimetry principles|
The Compton effect, light scattering
off electrons, discovered by Arthur Holly Compton (1892-1962), Nobel
prize in Physics, 1927, is one of the cornerstone of the wave-particle
duality. Compton scattering is a basic process of Quantum Electro-Dynamic (QED),
the theory of electromagnetic (EM) interactions.
During 50's and 60's, the QED theoretical developments allow Klein and Nishina to compute accuratly
the so-called Compton interaction cross section. Experimental physicists performed serveral experiments which
are in good agerement with the predictions. This is now a well established
theory, and is thus natural to use the EM interaction, such as Compton
scattering, to measure experimental quantities such as polarization of an electron beam .
Many of the Hall A experiments of Jefferson Laboratory using a polarized electrons beam require a
measurement of this polarisation as fast and accurate as possible. Unfortunately the standard
polarimeters, like Möller or Mott,
require the installation of a target in the beam. Therefore, the polarisation
measurement can not to be performed at the same time than the data taking because the beam, after the
interaction with the target, is misdefined in terms of polarization, momentum and position. Another
physical solution has to be found in order to permit a non-invasive polarisation measurement of the
beam. This is the principle of the Compton Polarimetry.
The Jefferson Lab electron beam, which polarisation is flipped 30 times per second, is interacting with
a laser beam of measured circular polarisation.
| ||From theoretical point of view|
This physical process is described by QED which allows to calculate the cross sections of the polarized
electrons scattering off polarized photons as a function of their energies and scattering angle.
The cross sections are different not equal if the incident helicity of the electron are in opposite
directions. One define the theoretical cross sections asymmetry Ath
by the ratio of the difference
over the sum of these two quantities. With the kinematical parameters used at JLab, the mean value of
these asymmetry is of order of few %.
|From experimental point of view|
Using a specific sdetup, the number of Compton interactions can be measured for each incident
electrons helicity state (aligned or antialigned with the propagation direction). These numbers are
dependant of process cross sections, luminosity at the interaction point and time of the experiment.
At first order, assuming the time and luminosity are equal for the both electron helicity states, the
counting rates asymmetry is directly proportionnal to the theoretical cross section asymmetry.
|From one to the other|
The proportionnality factor is equal to the values of the photon circular polarization Pphoton
multiplied by the
electron polarization Pelectron, so that :
Aexp = Pelectron P
Measuring the photons polarization and experimental asymmetry, calculating theoretical asymmetry,
one can deduce the electron beam polarization. One electron over a billion is interacting with
the photon beam which means 100000 interactions per second. So as only few incident electrons
are interacting, these polarization measurements are completly non-invasive for the electron beam
in term of positions, the orientations and the physical
characterictics of the beam at the exit of the polarimeter.
|Compton polarimeter principles at JLab|
The backward scattering angle of the Compton
photons being very small, the first priority is to separate these particles from the beam using a
magnetic chicane. The energy of the backward photons will be measured by an electromagnetic calorimeter,
the so-called PbWO4 coming from the LHC's R & D. The third dipole of the chicane, coupled to the electrons
detector, will be used as a spectrometer in order to measure the scattered electron momentum. To perform
a quick polarization measurement, the photon flux has to be as high as possible. A Fabry-Pérot Cavity,
made of 2 multi-layers concave mirrors with very high reflectivity, will amplify this flux to a factor
greater than 7000. The 15 meters long Compton Polarimeter has been installed in the last linear section
of the arc tunnel, at the entrance of the Hall A at spring 98. The complete setup, including the optical
cavity was installed in February 99 and is running successfully since then.
|The description of the Compton Polarimeter|
|The Compton polarimeter consists on :|
|The optical setup|
This is our photon target ! the optical setup is made off 4 parts :
- a 300mW infra-red Laser,
- the first optical path to make in form the laser beam in terms of size and polarization,
- the resonant cavity which delivers more than 1kW of circularly polarized infra-red light
- optical devices to measure the circularly polarization of the photons at the exit of the cavity
Basics of resonant Fabry-Perot cavity for
Compton polarimetry can be found in Nuclear Instruments And Methods In Physics Research
Section A412 1 (1998) pp. 1-18
|The magnetic chicane|
The Compton magnetic chicane consists of 4 dipoles (1.5 T maximum field,
1 meter magnetic length) here after called D1,2,3,4.
(D1,D2) deflect the electrons vertically down
the beam through the Compton interaction point (CIP) located at the center of
optical cavity. After the CIP, the electron are vertically up deflected (D3,D4) to reach
the Hall A target. The scattered electron are momentum analyzed by the third dipole and
detected thanks to 4 planes of silicon strips.
The magnetic field is scaled with the beam energy, insuring the same vertical deflection at the CIP,
up to 8 GeV electrons for 1.5 T field.
The Chicane parameters
- The distance between the geometrical axis of the
dipoles (D1,MMC1P01) and (D2,MMC1P02) in the longitudinal plane is 5400 mm
- The distance between the beam entry axis in (D1,MMC1P01) anfd the
beam exit axis in (D2,MMC1P02) in the bending plane (vertical axis) is 304 mm
- The longitudinal magnetic length on the axis of (D1,MMC1P01) and
(D2,MMC1P02) is 1000 mm.
Under these conditions :
- the bending angle is 3.22261o
- The radius of curvature is r=17.7887 m
- B * r (T.m) = 3.33564 p (GeV/c)
- At the centre of (D2,MMC1P02) : B(T) = 0.1875145 p (GeV/c).
| ||The photon detector|
To detect Compton backscattered photons, an electromagnetic
calorimeter is used. It consists of 25 PBWO4 cristals (2cmx2cmx23cm) read by XP1911
Philips photomultiplier tubes and is located in the line of sight of the optical cavity,
just behind the third dipole of the chicane. Details on this calorimeter can be found
in Nuclear Instruments And Methods In Physics Research Section A443 2-3 (2000) pp. 231-237
|The electron detector|
It is made of 4 planes of silicon strips composed
of 48 strips each of width 650 (600 + 50) microns and 500 microns
thick. The planes are staggered by 200microns to allow for better
resolution and the first strip of the first plane is about 8 mm away from the
Distance between the CIP and the first strip is 5750 mm. We recall that
between the CIP and the end of the Dipole 3 is 2150 mm.
For a beam of 3.362 GeV the Compton
edge is at 3.170 GeV. This corresponds to a deviation of 17 mm. Thus at
this energy, only one half of the Compton spectrum is covered and it extends
to the 13th strip of the first plane.
The trigger logic looks for a coincidence
between a given number of plane in a "road" of 2 strips. For each trigger
it outputs a signal check by the Polarimeter DAQ.
|The fast acquisition system|
The goal of this system is to acquire
for each electron helicity state the energie of the scattered photons at a
rate up to 100 kHz. The energy of each Compton event can be
reconstructed from the signals of the 25 PMT of the photon calorimeter
with front-end electronics and ADCs. Each helicity state, given
by the accelerator, is also numbered.
Further information is given for each
event (type of event, status of the polarimeter at event's time) and for
each polarization period (duration, dead time, counting rate,...).
A specific tool, the so-called spy_acq, has been developped in Tcl/Tk
to manage all acquisition system parameters. Finally, a web-based logbook
is available on this site at