Load-locked dc high voltage GaAs photogun with an inverted-geometry ceramic insulator

A new dc high voltage spin-polarized photoelectron gun has been constructed that employs a compact inverted-geometry ceramic insulator. Photogun performance at 100 kV bias voltage is summarized.

Figure 1

Diagrams of the two recent gun designs: (a) conventional large-bore cylindrical insulator and (b) inverted gun with compact tapered insulator that extends into the gun vacuum vessel. Both drawings have the same scale with top flanges ${10}^{\prime \prime }$ diameter. The cathode/anode gap is $\sim 6\mathrm{cm}$ for both designs.

Figure 2

Electrostatic field gradient map of the electrode inside the CEBAF inverted gun. The maximum field gradient within the cathode/anode gap is $4.5\mathrm{MV}/\mathrm{m}$. The gradient at the joint between the insulator and electrode is higher, $\sim 7\mathrm{MV}/\mathrm{m}$, but does not appear to be problematic. The photograph shows the actual electrode attached to the insulator and ${10}^{\prime \prime }$ CF flange, with photocathode puck being installed from behind.

Figure 3

Rate-of-rise plot showing pressure inside the unpumped gun high voltage chamber as a function of time, following the $400°\mathrm{C}$ vacuum bakeout. Pressure was measured with a spinning rotor gauge. The outgassing rate of the vacuum vessel can be calculated from the slope of the line using Eq. (1).

Figure 4

A false-color quantum efficiency (QE) map of the strained-superlattice GaAs photocathode inside the CEBAF inverted gun after production beam delivery from four locations (labeled). The map was obtained by scanning the drive laser beam ($\sim 0.5\mathrm{mm}$ FWHM) across the active area of the photocathode (diameter 5 mm), and measuring photocurrent at low bias voltage. $X$ and $Y$ units are arbitrary stepper motor intervals.

Figure 5

A representative plot of daily QE measurements for each drive laser (one per experiment hall) vs extracted charge from the new inverted gun. The data represent QE measurements from one photocathode location, after having already extracted $\sim 170\mathrm{C}$ from previous locations (i.e., this explains the suppressed zero of the x axis). A simple exponential fit to the data indicates $1/e$ charge lifetime of $\sim 70\mathrm{C}$.