Precision Electron-Beam Polarimetry at 1 GeV Using Diamond Microstrip Detectors

We report on the highest precision yet achieved in the measurement of the polarization of a low-energy, $\mathcal{O}(1\mathrm{GeV})$, continuous-wave (CW) electron beam, accomplished using a new polarimeter based on electron-photon scattering, in Hall C at Jefferson Lab. A number of technical innovations were necessary, including a novel method for precise control of the laser polarization in a cavity and a novel diamond microstrip detector that was able to capture most of the spectrum of scattered electrons. The data analysis technique exploited track finding, the high granularity of the detector, and its large acceptance. The polarization of the $180\text{−}\mu \mathrm{A}$, 1.16-GeV electron beam was measured with a statistical precision of $<1\%$ per hour and a systematic uncertainty of 0.59%. This exceeds the level of precision required by the $Q_{\text{weak}}$ experiment, a measurement of the weak vector charge of the proton. Proposed future low-energy experiments require polarization uncertainty $<0.4\%$, and this result represents an important demonstration of that possibility. This measurement is the first use of diamond detectors for particle tracking in an experiment. It demonstrates the stable operation of a diamond-based tracking detector in a high radiation environment, for two years.

Popular Summary

Electron beams can be produced with the electron spins aligned along a fixed direction, and experiments that use these kinds of beams rely on the precise knowledge of the fraction of electrons that are aligned (i.e., the “polarization” of the beam). One way of measuring beam polarization is Compton scattering, the collision of the beam electrons with photons from a laser, where the scattered electrons and/or the backscattered light are detected. The difference in the Compton scattering rates for electrons with spins oriented the same way or opposite to the spins of the laser photons is very well known. For a known laser polarization, the electron-beam polarization can be deduced by comparing the expected with the measured rate difference. Here, we share results of a new Compton polarimeter at our laboratory that has achieved the highest precision ever for low-energy (about 1 GeV) electron beams.

Our polarimeter incorporates key innovations such as ensuring that the laser beam is 100% polarized and a novel diamond-based electron detector that images the paths of the scattered electrons and can withstand very high radiation doses given the experimental duration of 200 days. Furthermore, this polarimeter can be operated without disrupting the electron beam. We have used this new polarimeter for the $Q_{\text{weak}}$ experiment, a high-precision test of the standard model of particle physics that requires an electron-beam polarization precision better than 1%. Our results significantly surpass this requirement.

We anticipate that our results will pave the way for upcoming standard-model tests with even more stringent ($<0.4\%$) requirements on the precision of the beam polarization.

Figure 1

Schematic diagram of the JLab Hall C Compton polarimeter. Four identical dipole magnets form a magnetic chicane that displaces the 1.16-GeV electron beam vertically downward by 0.57 m. An external low-gain Fabry-Pérot laser cavity provides a high-intensity (about 1.7 kW) beam of circularly polarized green (532-nm) photons. The laser light is focused at the interaction region (${\sigma }_{\text{waist}}\sim 90\mu \mathrm{m}$), and it is larger than the electron-beam envelope (${\sigma }_{\mathrm{x}/\mathrm{y}}\sim 40\mu \mathrm{m}$). The photon detector was not used for these results.

Figure 2

A schematic diagram of the CVD diamond plate mounted on an alumina frame, which forms a single detector plane. There were 96 Ti-Pt-Au strips deposited on the front face of the diamond plate, which was attached to the frame using a silver epoxy. The strips were connected to Au traces on the alumina frame with aluminum wire bonds. The traces terminated on two 50-pin connectors. A high voltage (HV) bias of about $-300\mathrm{V}$ was applied to the back side of the diamond plate via a miniature HV connector.

Figure 3

Yield and asymmetry data from a single detector plane plotted versus detector strip number, for a typical hour-long run. Statistical uncertainties only. (Top panel) The charge normalized yield at a beam current of $180\mu \mathrm{A}$ and laser intensity of 1.7 kW. The laser-on yield is shown in red, and the laser-off (background) yield is shown in shaded blue. (Middle panel) The measured Compton asymmetry (background-subtracted). The solid red line is a fit to Eq. (2). (Bottom panel) The background asymmetry from the laser-off period. The solid red line is a fit to a constant value.

Figure 4

Scheme for maximizing the circular polarization at the cavity. Laser light entering the system passes through a PBS, half-wave plate ($\lambda /2$), quarter-wave plate ($\lambda /4$), and vacuum window (VW) before it is either reflected off the cavity entrance mirror (CM1) or becomes resonant in the cavity. Note that in this figure, the element VW also includes three steering mirrors, which are incorporated in the model but left out of the figure for simplicity. Depending on the polarization state at CM1, reflected light will either arrive in a reflected photodiode (RPD), used for frequency-locking feedback, or be sampled by the polarization signal (PS) photodiode behind a steering mirror. Light arriving at the cavity entrance mirror is fully circular when there is no signal in PS. Before the experiment, with part of the beam-line vacuum pipe removed, it was possible to do a direct measurement (DM) of the circular polarization in the cavity.

Figure 5

Direct test of the laser-polarization maximization technique. The correlation between the laser DOCP directly measured after the cavity mirror and the polarization signal extracted from the reflected light. The left panel shows the full range, while the right panel is a zoomed-in version of the region of maximum DOCP.

Figure 6

(Top panel) Typical Monte Carlo simulated Compton spectra for a single detector plane; ideal (black open circles), with noise (red), and with detector efficiency (blue, shaded). (Bottom panel) The Compton asymmetry extracted from the simulated spectrum including detector efficiency (blue circles), and a two-parameter fit to the calculated asymmetry (red line). The input asymmetry was 85%.

Figure 7

The extracted beam polarization for the 1.16-GeV, $180\text{−}\mu \mathrm{A}$ electron beam, as a function of run number and averaged over 30-hour-long periods, during the second-run period of the $Q_{\text{weak}}$ experiment (blue, solid circle). Also shown are the results from the intermittent measurements with the Møller polarimeter [2] (red, open square). The inner error bars show the statistical uncertainty, while the outer error bar is the quadrature sum of the statistical and point-to-point systematic uncertainties. The solid bands show the additional normalization or scale-type systematic uncertainty (0.42% Compton and 0.65% Møller). The dashed and solid (green) vertical lines indicate changes at the electron source.