The photocurrent attenuation of GaAs photocathode within one hour after activation under three different vacuum pressure (5×10-9Pa, 5×10-8Pa, 5×10-7Pa) were recorded by automatically activated monitor. The results show that: the photocurrent quickly descend in the beginning and then descend linearly at a low slope; the amplitude of the quickly descending area were 10%, 14.74% and 36%separately, with the respective slope of the linear descending area were -0.00653, -0.01132and -0.02. Three samples’ gas components of H2, CH4, CO, H2O, O2, CO2 etc under the same vacuum pressure (5×10-8Pa)during photocurrent attenuation were collected by quadrupole mass spectrometer. By comparing the gas components content and the attenuation law of the photocurrent, it has been found that H2O and H2 had a greater impact on the stability of GaAs photocathode in the ultra-high vacuum environment and H2O was the predominant effect. This paper has important guiding significance and reference value in studying the stability of GaAs photocathode and the improvement of semiconductor photocathode process.
The gallium arsenide (GaAs) photocathode generally requires a high temperature thermal cleaning before (Cs, O) activation in order to obtain an atomic level clean surface. The process is useful to adsorb and deposit cesium and oxygen atoms. Generally considered, the photocathode needs to be cooled to 60℃ to activate for achieving better results. People usually keep the annealing time for at least 1.5 hours in practical production. In order to explore the effect of annealing time on cesium atoms which were adsorbed on GaAs photocathode, the experiment monitored the activation curves of three GaAs photocathodes samples which annealed for 0.5 hour, 1.0 hour, 1.5 hours respectively, and then compared the occurrence moment of the photocurrent and the first cesium peak by different annealing waiting time. The difference of the activation curves reflects indirectly that the photocathode surface temperature had an influence on the adsorption of cesium atoms during activation process. This phenomenon could explain from two aspects about atoms adsorption and electronic transport. The work has referential significance for experimental research and industrial production.
Ar+ ion etching and X-ray photoelectron spectroscopy (XPS)depth profile analysis have been performed on the native oxide layerof GaAs(100) surface. The composition of the native oxide layer,that isthe oxide phases of gallium and arsenic, was characterized precisely. It is indicated that native oxide phases on extreme surface of GaAs(100) consist of a mixture of Ga2O3, As2O3 and As2O5. Furthermore, the respective distribution of oxide phases of gallium and arsenic along the depthwere compared and analyzed.A seemingly contradictory phenomenon was found, that is As enrichment exist in total oxide layer, but the content of Ga oxide was greater than that of As oxide in the oxide layer except for the outmost surface layer.Based on the comprehensive influence of oxidation process, etching, segregation and growth process, the intrinsic mechanism of the change of oxides along etching depth was discussed. According to the analyzed results, the oxide layer of GaAs (100) surface should be divided to two layers,that is the outmost layer containing oxides of Ga and As and the intermediate layer including only oxide of Ga.The concentration of As oxides in the outmost layer and the enrichment of As in total oxide layer are derived from surface structure inhomogeneity. The throughout total oxide layer of Ga oxide is attributed to its stronger oxidability.In the present work, the system study for native oxide layer of GaAs surface provides the powerful foundation for understanding surface state of GaAs and surface treatment.
In order to improve the electronic gain and luminance gain of low-light-level image intensifiers, microchannel plates(MCP) are adopted as the electron multiplier mechanism. According to the relevant experimental analysis, the resistance between channels is a limited value. Due to there are resistive coupling between any two adjacent channel of MCP, the electron transmission and the electron multiplication in a certain channel will be interfered by its adjacent channels, This phenomenon would affect the quality of image transmission and field of view of image intensifier. In low-light condition, the input current of MCP is small, the current gain of each channel is same, MCP has the area of linear current amplification and distortion-free image transmission. But when input current is large and close to saturation, lower current in channels has more current gain, leading to the contrast change of the image. This paper analyzes the transmission properties of electrons in the channels. It is proved that there is an electrical relationship between adjacent channels,throuht the circuit equations with relevant circuit parameters such as the resistance of secondary electron emission layer, resistance of resistive layer, the resistance between two adjacent channels, and so on. The analysis method and research results provide technical guidance for the improvement of electronic gain, luminance uniformity and preparation process of MCP.
The 3rd generation low-light-level image intensifiers should be aged for 100 hours before its normal use. In order to know the influence of ageing processing on GaAs photocathodes, five 3rd generation low-light-level image intensifiers were aged with the life testing instrument of low-light-level image intensifier in an experiment. With the spectral response testing instrument, the intensifiers were measured for totally 8 times to get their spectral response respectively before they were aged and in a half year after aged, and to calculate the integral sensitivity according to the spectral response curves. Based on the fluctuating spectral response curves and the varying integral sensitivity, it was indicated that the aged intensifiers up to standard had more stable photocathode sensitivity, smaller decrease in their spectral response curves, while those not up to standard had more obvious decline as a whole in their spectral response curves. Additionally, the threshold wavelength of all intensifiers was moving toward shortwave. The degeneration of GaAs photocathode resulted from the instability of the Cs-O layer on GaAs photocathode surface. During the ageing processing, the lack of a longtime light radiation on Cs-O layer, the widening surface barrier and the decreasing escape probability led to less photoelectronic emission and lower sensitivity. Moreover, the destruction of dipole layer resulted in smaller bending of surface band and higher vacuum level, so that the electrons in impurity level could not escape and the threshold wavelength moved toward shortwave. Thus the ageing processing played a role of picking out the 3rd generation low-light-level image intensifiers to get rid of the products not up to standard and to put the photocathodes of products up to standard into a relatively stable random failure period.
The gallium arsenide (GaAs) photocathode was generally cleaned by radiant heating, direct heating, ion bombardment annealing, and so on. In this paper, the radiant heating method, namely thermal cleaning method, was adopted for GaAs photocathode surface purification. Using this method could obtain an atomic clean surface, ensure the integrity of the GaAs surface lattice, and guarantee the uniformity of surface cleaning effect at the same time. But because the accurate measurement of the GaAs photocathode surface temperature in the vacuum system was very difficult, the residual gas analyzer (RGA) was used in this experiment to monitor the residual gas composition in ultrahigh vacuum during the thermal cleaning process and determine the thermal cleaning temperature by the partial pressure curves of As and Ga. It was found that the first peaks of As and Ga elements both appeared after heating about one hour, accompanied with H2O, N2/CO, CO2 and other common gas. According to partial pressure curves of H2O, N2/CO, CO2 and the heating time, it could be judged that the temperature at that time was not high, which should be under 150°C.After thermal cleaning experiment of three GaAs photocathodes, it was found that the peak value of As partial pressure at low temperature was generally within 10-11mbar~10-10mbar, and the peak value was at 10-10mbar at high temperature. Sometimes it was appeared that the peak value of As partial pressure at low temperature was even higher than the peak value at high temperature. The As volatilization phenomenon occurred at low temperature indicated that the elemental As exist on the GaAs photocathode surface or near surface after the chemical etching process, and the As could volatilize from GaAs photocathode at low temperature in the beginning of thermal cleaning. This research has guiding significance for further understanding the thermal cleaning mechanism of GaAs photocathode and improving the thermal cleaning technology.
Using the projected augmented wave potential by the density functional theory based upon gradual gradient approach method and the slab model, from the calculated surface, we identify the relaxed atoms sites of GaAs(110) surface, the electronic structure of elements K and O adsorpted on binding sites of ideal GaAs(110) surface have also been calculated, especially the total energy of the adsorption system. The comparison results of calculated total energy showed: for K and O elements at highest coverage of Θ=1ML on GaAs(110) surface, they were not formed to local domain of competitive chemical adsorption, while they were formed to a compound uniformity phase of cooperative chemical adsorption. Our calculated results providing theoretical basis and reference for the application of alkali oxidation adsorpted on GaAs surface to form a negative electron affinity photocathode.
Photocurrent of GaAs photocathode activated with Cs and O was tested by auto-activation monitor, the fitting curves of photocurrent showed that the photocurrent of the photocathode after the first activation declines exponentially, and then declines linearly with very small slope |k1|; the photocurrent after the second activation rises exponentially, and then declines linearly with a slope|k2| which is a bit larger than |k1|.Based on the mechanism difference between twice annealing of the photocathode, the degeneration behavior of the photocathode was analyzed by three-dipoles model and XPS test after the first activation and succedent thermal cleaning. It is indicated that Cs2O dipoles on the surface are saturated after the photocathode was activated for the first time, the remained Cs and Cs2O in the ultra-high vacuum chamber which deposited on the photocathode surface will prevent the emission of photoelectrons. The photocathode surface with Cs and O reconstructed when it was annealing for the second time, a lot of Cs2O dipoles changed into more stable GaAs-O-Cs dipoles, and this phenomenon would happened immediately as soon as the photocathode was activating for the second time. After the residual Cs and Cs2O dipoles depleted, the neutral gas CO2, H2O, O2, damaging the surface dipoles layer, are the main factors resulted in the decline of photocurrent. Due to the instable Cs2O dipoles on the surface of photocathode have greater chances of converting into stable GaAs-O-Cs dipoles when photocathode was activated for the first time, the photocurrent declines more slowly compared with the second activation. The discussion for the phenomenon is of great significance for exploring the photoemission mechanism of Ⅲ-Ⅴ semiconductors.
Single-MCP Cs2Te solar blind ultraviolet image intensifier couldn’t detect weaker ultraviolet radiation, such as the
ultraviolet radiation near high-voltage wire insulating column. To increase the ultraviolet radiation gain, the double-MCP,
tri-MCP or multi-MCP units are introduced into ultraviolet image intensifiers. In this paper, two pieces of MCP are
cascaded in "V" shape as the electron multiplier of ultraviolet image intensifiers. Processed and scrubbed by the
single-MCP electron scrubbing traditional technology, the desired effect after electron scrubbing still could not be
achieved and flicker phenomenon appeared in field of view. The flicker noise appeared when the image intensifier was
working because the second MCP was not efficiently scrubbed. In order to completely scrub the two pieces of MCP
simultaneously, eliminate the flicker noise, reduce the dark current and achieve a stable MCP gain, the double-MCP
electron scrubbing method should be optimized without changing the assembly process. Combined with MCP
pre-treatment and pre-electron-scrubbing before assembled and scrubbed in “V” shape, flicker noise could be eliminated
effectively, and dark current could be lowered, which could increase the gain and get a clear field of view. Comparing
with two different methods of double-MCP electron scrubbing, either method has its own advantages and disadvantages.
Ultraviolet radiation gain can be increased from 103~104 cd·m-2/W·m-2 to 1.0×105 cd·m-2/W·m-2 by using method of pre-treatment and pre-electron-scrubbing. With prospective ultraviolet radiation gain achieved, double-MCP Cs2Te solar blind ultraviolet image intensifier is manufactured.
In order to know more about the surface state of GaAs(100) epitaxial wafer during a storage period of two years, the
XPS analysis was carried out four times on the surface, respectively polished by chemical etching, stored in desiccator
for half a year, one year and two years. The results indicated that even after cleaned by proper etchant solutions, the fresh
surface was slightly oxidized with Ga2O3, As2O3 and organic contaminant. The epi-wafer was always exposed to air during the storage period, so more and more oxides turned out. The mixed oxide layer comprised of C-OR, COOR, Ga2O3, As2O3 and As2O5 appeared after only half a year. In the ageing process of two years, the oxide types of gallium or arsenic did not change with stable content of Ga2O3 and remarkably fluctuating relative contents of As2O3 and As2O5. Based on the intensity ratio of Ga 3d-Ga2O3 to Ga 3d-GaAs, the thickness of oxide layer was estimated. The oxide layer generated after chemical polishing was very thin, just only 0.435nm thick, and then it grew rapidly, approximately 1.822nm after one year while almost no change any more subsequently. It was indicated that after the epi-wafer was
stored for one year, because of volatile As2O3 or As2O5, there remained a large amount of Ga2O3 in oxide layer, which prevented the reactions between bulk material and oxide layer with oxygen. So native oxide layer plays a role as passive film to protect epi-wafer against the environment during a long storage period.