Effects of ion bombardment on bulk GaAs photocathodes with different surface-cleavage planes

Bulk GaAs samples with different surface cleave planes were implanted with 100 and 10 000 V hydrogen ions inside an ultrahigh vacuum test apparatus to simulate ion back-bombardment of the photocathode inside a DC high voltage photogun. The photocathode yield, or quantum efficiency, could easily be recovered following implantation with 100 V hydrogen ions but not for 10 000 V ions. Moreover, the implantation damage with 10 000 V hydrogen ions was more pronounced for GaAs photocathode samples with (100) and (111A) cleave planes, compared to the photocathode with (110) cleave plane. This result is consistent with enhanced ion channeling for the (110) cleave plane compared to the other cleave planes, with ions penetrating deeper into the photocathode material beyond the absorption depth of the laser light and beyond the region of the photocathode where the photoemitted electrons originate.

Figure 1

Electron impact ionization cross section for hydrogen molecules—the dominant gas species inside modern dc high voltage photoguns—as a function of electron kinetic energy.

Figure 2

Schematic representations of the GaAs crystal for three surface cleave planes, looking normal to the surface—from left to right, (100), (110) and (111A). The red boxes denote the recurring geometric cells used for calculating the atomic number density for each cleave plane and $a$ is the GaAs lattice constant.

Figure 3

(top) Hydrogen ion distribution and (bottom) vacancy distribution within GaAs for different hydrogen ion energies. All simulations performed using the same hydrogen ion dose of $8.1\times 10^5\mathrm{nA}\mathrm{s}$. Plots were obtained using SRIM [18].

Figure 4

Schematic representation of the ultrahigh vacuum test apparatus (left) and a photograph of the apparatus (right).

Figure 5

Ion trajectory from the hydrogen ion source to the photocathode sample when the photocathode sample is biased at 100 V (left) and 10 000 V (right). Ions exit the dissociator glassware from a 1 mm diameter hole located approximately 10 cm beneath the photocathode sample.

Figure 6

Photocathode QE versus laser wavelength for commercial bulk GaAs samples with different surface cleave planes, prior to ion implantation.

Figure 7

Typical QE maps of the photocathode obtained by scanning a focused low power green laser beam across the photocathode following activation. These maps show the bulk GaAs (111A) sample. Top: After the third activation following implantation with 100 V hydrogen ions. Bottom: After the third activation following implantation with 10 000 V hydrogen ions. The two-dimensional plots help to illustrate ion implantation uniformity: 100 V hydrogen ions were uniformly distributed across the entire sample but 10 000 V hydrogen ions were focused to a region denoted by the red circle, consistent with the particle tracking code simulations.

Figure 8

QE versus heat cycles for bulk GaAs with three different surface cleave planes. Black square: no implantation; red circle: implantation with 100 V hydrogen ions; blue triangle: implantation with 10 000 V hydrogen ions. Errors for no implantation and 100 V results are small because QE was uniform across sample; errors for 10 000 V results are large because ion damage was limited to small region of photocathode. Heat cycle numbers: 1: $250°\mathrm{C}$ for 20 min; 2: $370°\mathrm{C}$ for 30 min; 3: $490°\mathrm{C}$ for 1 h; 4 and 5: $550°\mathrm{C}$ for 1 h.

Figure 9

QE ratio—samples implanted with 10 000 V hydrogen ions to samples that were not exposed to hydrogen ions—as a function of heat cycle, for bulk GaAs with three different surface cleave planes. Heat cycle numbers: 1: $250°\mathrm{C}$ for 20 min; 2: $370°\mathrm{C}$ for 30 min; 3: $490°\mathrm{C}$ for 1 h; 4 and 5: $550°\mathrm{C}$ for 1 h.