The influence of CO, CO2 and H2O gases on the stability of negative-electron-affinity (NEA) GaAs photocathodes during operation is investigated in the present work. We have found that exposure both to H2O and CO2 decreases the photocurrent of the photocathode. However, exposure to CO, which is known as a harmful gas to various photocathodes, has little effect on the photocathode stability. Furthermore, the effects of these gases on the restoration of the photocurrent by additional cesium deposition are investigated. These results are discussed with regard to the Cs/O activation layer which plays an important role in NEA GaAs photocathodes.

No abstract

The present paper describes a three-dimensional (3D) thick-photoresist microstructuring technique that exploits the effect of exposure wavelength on dissolution rate distributions in a thick-film diazonaphthoquinone (DNQ) photoresist. In fabricating 3D microstructure with specific applications, it is important to control the spatial dissolution rate distribution in the photoresist layer, since the lithographic performance for 3D microstructuring is largely determined by the details of the dissolution property. To achieve this goal, the effect of exposure wavelength on dissolution rate distributions was applied for 3D microstructuring. The parametric experimental results demonstrated (1) the advantages of the fabrication technique for 3D microstructuring and (2) the necessity of a dedicated simulation approach based on the measured thick-photoresist property for further verification. Thus, a simple and practical photolithography simulation model that makes use of the Fresnel diffraction theory and an empirically characterized DNQ photoresist property was adopted. Simulations revealed good quantitative agreement between the photoresist development profiles of the standard photolithography and the moving-mask UV lithography process. The simulation and experimental results conclude that the g-line (λ = 436 nm) process can reduce the dimensional limitation or complexity of the photolithography process for the 3D microstructuring which leads to nanoscale microstructuring.

To investigate the accommodation of a large lattice mismatch, the misfit strain of a GaP layer grown on a GaAs(100) substrate and a GaAs layer grown on a GaP(100) substrate were studied by X-ray diffraction and Raman spectroscopy as a function of the layer thickness. Epilayers were grown by MBE. The misfit strain began to relaxed at a thickness of 50 nm for GaP and 80 nm for GaAs. The magnitude of the saturated strain was 0.48% for GaP and 0.13% for GaAs. These magnitudes are insufficient to make a coherent interface. The tendency of our results are different from Matthews' prediction. His theory deals with layer-by-layer growth. Whereas the RHEED pattern showed an island formation during the initial stage of growth, subsequent islands coalesced to form a flat surface. Thus, the difference between the prediction and the experimental results was probably due to the difference in the growing mechanism.

No abstract