Measurement of Electron Beam Polarization at Stanford Linear Accelerator Center

1
Darcy Lambert Laura Peterson and Professor Piotr
Decowski
How accurately was electron beam polarization
measured in the E158 experiment at SLAC?
Measurements of Performance of the Detector Used
in a Precise High energy Electron-Electron
Scattering Experiment at SLAC
Measurement of Electron Beam Polarization at
Stanford Linear Accelerator Center
Set Up and Procedure
Stanford Linear Accelerator
We then placed a light filter over one of the
LEDs, which was to remain there for the duration
of the experiment. We called this LED1 (the other
was LED2) With the PMT hooked into a high
voltage of -600 V, we took measurements by
digitizing PMT signal in a CAMAC ADC connected
to a computer, with light from LED1, then LED2,
and then LED(12), using the shutter to control
which light shone through. We then placed a
filter over both LED 1 and 2 and took
measurements again. We repeated this process
with a number of filters providing different
light attenuations. The measurements were
repeated with the HV at -700 V, -800V, and
-900V.
We began by placing the photomultiplier inside of
a long black tube and then into a light resistant
box so that external light could not interfere
with our measurements or damage the delicate
PMT. We directed two Light Emitting Diodes
(LEDs) through collimators at the PMT, and
placed a remote controlled shutter between the
LEDs and the PMT so that we could block the
light from one or the other in order to observe
just one at a time (or have them both open).
Background and Introduction
The ultimate goal is a measurement of parity
violation in electron-electron scattering Parity
violation (or invariance with repspec to reversal
of coordinate axes) is an extremely small effect
(100 parts per billion) that stems from
interference between electromagnetic and weak
interactions between electrons (exchange of a
photon vs. exchange of the Zo boson). The
simplified manifestation of parity violation,
predicted by the Standard Model, is that high
energy left handed electrons would have a greater
probability of collision with randomly polarized
electrons (in liquid hydrogen, for instance) than
high energy right handed electrons.
Results
In order to measure polarization of the beam,
polarized electrons that have been accelerated to
48 GeV are scattered from polarized atomic
electrons in a magnetized iron foil. The
probability of scattering--a process very well
described by quantum electrodynamics -- depends
on the relative orientation of bombarding and
target electron spins (a.ka. helicity) thus,
measuring the amount of electrons scattered from
the iron foil will give us information about the
polarization of the beam-- if the spins of the
electron beam and the foil are more or less
aligned, the scattering will be different than if
they are antiparallel. This phenomenon is quite
intuitive. It makes sense that if you collide the
south pole of one magnet with the south pole of
the other it would have a different effect than
if you collided the north of one and the south of
the other, and electron spins make tiny magnets.
The scattered electrons pass through first a
block of tungsten that amplifies the electron
signal, and then through a quartz block (faster
than the speed of light!) connected to a
photomulitplier, and as they pass they produce
photons, which then travel to the PMT cathode and
induce a signal. The output signal of the PMT is
proportional to the number of photons produced in
the quartz, and thus the number of electrons
scattered from the foil, and finally also the
polarization of the electron beam.
Looking at the actual data points, we can see
that the linearity of our photomultiplier is not
quite exact (bottom two panels). The ratio of
the two LEDs (top right) is more or less
constant, as we would hope, but there still
exists a discrepancy. The top left panel shows
deviation at -800V of the ratios of the ADC
channels with both LEDs illuminating the PMT to
the sum of signals produced by each LED
individually (as a function of the optical filter
transmission). Ideally, this graph would just be
a straight line at 0, but as you can see the
nonlinearity increases at lower light
intensities, we found that this was equally true
for all HVs.
The purpose of the E158 experiment at the
Stanford Linear Accelerator Center was to make
the first measurement of parity violation in
electron-electron scattering. A precise
measurement of parity violation in
electron-electron scattering would serve as a
test of the Standard Model -- the most exact
model of subatomic particles known today. How we
measured parity violation In the E158
experiment, we accelerated longitudinally
polarized electrons to an energy of 48 GeV, which
were then scattered at a very small angle (.03o)
off electrons in 1.5 m long liquid hydrogen
target. The amount of parity violation was
determined from the asymmetry of the number of
scattered electrons with bombarding electrons
polarized in the direction parallel (helicity 1)
and antiparallel (helicity -1) to the direction
of their motion-- i.e. right/left handed. What
we did here at Smith In order to make an
accurate measurement of parity violation, we need
to know very accurately the polarization of the
high energy beam of electrons. This poster
describes our test of the linearity of the
photomultiplier that was used in the detector at
SLAC to measure that polarization. The necessity
for excellent linearity of the photomultiplier is
important for precise determination of
polarization, and hence also of parity violation.

Goal
Finally, the graph on the right is something of a
synthesis of the above information and it shows
the deviation from linearity in percent as a
function of the ADC channel. In the E158
experiment, i.e. in the polarimetry measurement
runs, the PMT was most usually at HV-700V with
signals measuring between channel numbers
800-1000. In this region the deviation from
linearity is less than 1 percent, which is
accurate enough to facilitate the precision that
the test for parity violation demands.
In our final measurement of parity violation, we
would like our error to be less than 10, which
demands that the beam polarization should be
measured with accuracy not worse than 1. The
overall goal of our work at Smith is to check the
linearity of our photomultiplier (PMT) used in
the polarimetry measurement at SLAC. In order to
do this, we need to make sure that as we adjusted
the light intensity, the ratio of the PMT signal
of LED1, with the permanent filter, to LED2
remains constant. If the ratio remains constant,
then it is a sign of good linearity which will
help us to achieve maximum precision in the
measurement of polarization and then in the final
test of parity violation. Combining the data
from both LED1 and 2, we graphed the relative
light intensity versus the signal from our PMT
(manifested as an ADC channel readout). We can
see that at all voltage levels, the data appears
to be very linear, but we needed to see just how
straight these lines actually are.