nsepol20000418

**Figure 1:** (top) Scheme for the measurement of the polarization using the Pi192 Co/Ti test piece. (bottom) Calculation of the magnetic field (in T) of the face of the supermirror test piece (solid rectangle) and a quadrant of a proposed supermirror plate for the full analyzer. The contour lines indicate the absolute magnitude of the field for a 10A current, while the arrows indicate the direction of the field in the plane of the paper.
$\resizebox {\textwidth}{!}{\includegraphics{fielddist.eps}}$

I replaced the monolithic Fe/Si analyzer with the Pi192 Co/Ti supermirror test piece in the analyzer solenoid of the NSE. The test piece was positioned with its long axis (200 mm) along the field (and beam) direction and its short axis (64 mm) vertical. The long axis of the test piece was reported by Ian Anderson to be the easy axis for magnetization. The leading edge of the test piece was located at the center of the coil (4056 mm from the sample position). A boroflex/Cd mask with a 4mm gap was placed against the leading edge of the test piece (see Fig. 1). For a 10A current through the polarizer coil, the field varied over the test piece from 62G down to 49G (see Fig. 1).

The current in the analyzer solenoid was raised momentarily to 48A and then reduced to 9.6A. A measurement in the direct beam gave a polarization P = 0.826(3) (see Fig. 2), which after 90 minutes was not significantly different P = 0.820(2). With a current of 45A in the analyzer coil (minimum field on sample is then 220 G), the measured polarization was 0.848(3).

**Figure 2:** (top) Detector images of the center of the direct beam both above and at the position of the test supermirror piece, for both $\pi$ -flipper on and off. (bottom) Integration of image along the vertical axis, as a function of the horizontal axis distance, for $\pi$ -flipper on and off, and the resulting polarization. The integration along y is between pixels 43 and 94. (1 pixel = 0.85mm)
$\resizebox {0.8\textwidth}{!}{\includegraphics{centeredanal.eps}}$

The polarization was also measured to be P = 0.901(1) at the edge of the main beam by moving the second arm of the detector by 3cm from the beam center position (see Fig. 3). As well, the beam polarization measured from a coherently scattering sample was 0.863(3), confirming my earlier analysis that the center of the main beam is contaminated by a low-divergence unpolarized component transmitted through the polarizer.

The performance observed for this Co/Ti polarizer is acceptable for the operation of the NSE spectrometer.

**Figure 3:** (top) Detector images of the edge of the direct beam (3cm from the center) both above and at the position of the test supermirror piece, for $\pi$ -flipper both on and off. (bottom) Integration of image along the vertical axis, as a function of the horizontal axis distance, for $\pi$ -flipper on and off, and the resulting polarization. The integration along y is between pixels 43 and 94. (1 pixel = 0.85mm)
$\resizebox {0.8\textwidth}{!}{\includegraphics{offsetanal.eps}}$

About this document ...

This document was generated using the LaTeX2HTML translator Version 99.1 release (March 30, 1999)

The command line arguments were:
latex2html -html_version 3.2,math -no_math -local_icons -split 0 nsepol20000418

Polarization of ILL Co/Ti supermirror Pi192

About this document ...