PULSE LASER DEPOSITION METHOD FOR UV-PHOTOCATHODE PRODUCTION

Production method of alkali halides UVphotocathodes durable to environment exposure was proposed. Photocathodes with high homogeneous and adhesion photoemission layer were made. Quantum yield measurements results of manufactured photocathodes are presented. Developed method is promising for creation of high aperture electron injectors and detectors.


Introduction
UV-photocathodes have wide application as UVdetectors, electron injectors for acceleration facilities an etc. Especially actual problem is creating wide aperture UV-photocathodes with homogenous photoemission layer, which can withstand exposure to air without degradation of quantum yield.
Pure metals have low quantum efficiency even after laser cleaning of emitting surface [1].Common method to improve quantum efficiency is to cover metal surface by alkali halides.This method uses complex and expensive vacuum equipment, but still to get good homogenous film with good adhesion and quantum yield one should to have enormous experience in such technologies.
In paper [2] pulse laser deposition (PLD) method was used to cover metal substrate by Mg thin film.Pure film with high homogeneity and adhesion was achieved.It is known that alkali halides films considerably improve quantum efficiency of pure metal [3].Also such coatings can withstand exposure to air, that allows creating wide aperture UV-photocathodes without requirement oversized high vacuum chamber and sophisticated soldering techniques.

Production
PLD has long been known and used for various applications.This technique particularly effective for production of films with complex stoichiometry, growth speed is high, mechanical parameters also are very good.The disadvantages of PLD are: the availability of large-sized particles on the surface at the wrong laser regime and poor controllability of the process.Fig. 1 shows schematic of the film growth.To implement the PLD film growth of halide compounds we developed and manufactured vacuum chamber and pulsed solid state laser.The chamber had a few quartz windows to enter laser radiation, gauge Edwards WRG-S, fore line pump, turbo molecular pump Varian S.P.A, noncontact target heater (808nm 7W laser diode with collimator), focusing lens with F=50mm.We made lamp pumped Nd:YAG solid-state laser with passive Q-switching based on LiF saturable absorber.The laser parameters are: pulse duration -100ns, wavelength -1064nm, frequency < 20Hz, pulse energy < 1J, TEM00 beam quality.Fig. 2 shows schematic of designed laser.Laser power and frequency controlled by power source.It's important that pulsed laser should work in threshold regime, when plasma plume exist, but not much power putted into each pulse.Otherwise film can be damaged by big fragments of target and droplets.Deposition time was picked experimentally to get optimal 0.5um thin film [3].
Deposition was lasted for 15 frequency) depending on the material.T was controlled by interferometer.
As substrate we used Cu and Ni flat disks 1.5mm thickness and 20mm diameter.Our preliminary tests showed that quality of produced films greatly depend on target and substrate preparation.We had pure CsBr compound in powder form, so we pressed it into dense tablet, CsI compound was in crystal form so any additional preparation required.Substrate preparation was: mechanical cleaning by fine grinding paper, laser cleaning, chemical cleaning in hydrochloric acid solution, washing in distilled water, drying at temperature.
After all preparations target and placed into vacuum chamber, fore line pump and turbo molecular pump produced vacuum about 10 torr and then pulsed laser turned on.The substrate and the target should be set normal with respect to each other, incidence angle of laser radia about 45 degree to normal of target When PLD process was over, substrate was annealed for 3 hours in 60deg C and then got out from vacuum chamber to environment.

Quantum yield measurements
Experimental studies of quantum yield of produced photocathodes were made by measuring the total charge accumulated from measuring capacity.Photocurrent flow between anode mesh and photocathode after UV laser short pulse exposition cause the charge.Fig, 4 represent measuring set up.
We applied DC high voltage 0-20kV between mesh and photocathode (SRS PS375 power source used), distance was set 0-5mm.Measuring capacity C=10nF also protects oscilloscope from high voltage gap breakdown.Total charge was gained from area under voltage pulse of cable capacity recharge to measuring capacity through measuring resistor (oscilloscope input resistance, TDS 2022B or LeCroy WaveRunner 62Xi R=1MOmh.Cable capacity was much lower measuring capacity so recharge pulse form d lasted for 15-16min (20Hz n the material.Thickness As substrate we used Cu and Ni flat disks 1.5mm mm diameter.Our preliminary ality of produced films greatly depend on target and substrate preparation.We had pure CsBr compound in powder form, so we it into dense tablet, CsI compound was in eparation was not required.Substrate preparation was: mechanical cleaning by fine grinding paper, laser cleaning, chemical cleaning in hydrochloric acid solution, , drying at room target and substrate were placed into vacuum chamber, fore line pump and turbo molecular pump produced vacuum about 10 -5 on.The substrate and the target should be set normal with respect to each other, incidence angle of laser radiation is to normal of target, see fig. 3. over, substrate was C and then got out from vacuum chamber to environment.

Quantum yield measurements
studies of quantum yield of produced made by measuring the total charge accumulated from measuring capacity.Photocurrent flow between anode mesh and photocathode after UV laser short pulse exposition measuring set up.
20kV between mesh power source was 5mm.Measuring capacity C=10nF also protects oscilloscope from high voltage gap breakdown.Total charge was gained area under voltage pulse of cable capacity recharge to measuring capacity through measuring resistor (oscilloscope input resistance, Tektronix LeCroy WaveRunner 62Xi) much lower than ulse form did not distort respect to case then measuring capacity shorted circuit.To make sure all emitted electrons reach anode mesh we adjusted laser pulse energy and applied DC voltage.By achieving voltage at which a linear relationship emitted charge practically independent of the applied voltage, we can indirectly d quantum efficiency.But this is true only if the of flight of the electrons from the cathode to the grid is much greater than the laser pulse du This condition is satisfied for picosecond laser sources, but not satisfied for UV eximer lasers with pulse duration 5-7ns.For longer laser pulses measuring result provides lower QY relatively direct QY measurement techniques.After 10minutes exposure to air, we mount photocathode and tungsten 520um step) into Teflon holder, distance from cathode to mesh was set to 2mm.Mesh was soldered to positive electrode of high voltage power source.Assembled unit was put into measurement vacuum chamber (10 -7 torr).photodiode made from developed photocathode and mesh.As UV sources we used ArF and KrF excimer lasers with wavelength 193 and 248nm respectively, pulse duration was 5ns, 100Hz, average pulse energy (a) Fig. 5. Photocathode testing assembly UV laser beam was focused by CaF2 lens, on cathode surface was 1-3mm In our experiments we scan with UV emitting surface, to study whole surface.Response signal distort respect to case then measuring capacity was shorted circuit.To make sure all emitted electrons reach anode mesh we adjusted laser pulse energy By achieving saturation voltage at which a linear relationship ceases and practically independent of the applied voltage, we can indirectly determine the quantum efficiency.But this is true only if the time of flight of the electrons from the cathode to the grid is much greater than the laser pulse duration.This condition is satisfied for picosecond laser sources, but not satisfied for UV eximer lasers with 7ns.For longer laser pulses measuring result provides lower QY relatively direct QY measurement techniques.yield measurement circuit.A -anode mesh, cable capacity, K1 -discharge isolate capacity, R= 1МΩ -measurement 20kV HV.
After 10minutes exposure to air, we mounted mesh (35um diameter, 520um step) into Teflon holder, distance from cathode to mesh was set to 2mm.Mesh was soldered to positive electrode of high voltage power source.Assembled unit was put into measurement .Fig. 5 shows vacuum photodiode made from developed photocathode and mesh.As UV sources we used ArF and KrF excimer lasers with wavelength 193 and 248nm respectively, pulse duration was 5ns, frequency -100Hz, average pulse energy was less than 1mJ.UV laser beam was focused by CaF2 lens, spot size mm diameter, regulated.our experiments we scan with UV spot on QY homogeneity of whole surface.Response signal was broadened because of RC constant of measuring circuit (about 100us in our case).By measuring incident light energy and calculating electrical charge from area under the response signal curve (1) we got quantum yield (2).
where Q -electrical charge, I(t) -current, U(t)voltage, R -measuring resistance, Y -quantum yield, N el -number of emitted electrons, N phnumber of incident photons, W -UV light energy, hω -single photon energy.

Results and discussions
Better result was obtained with Cu substrates, achived quantum yield presented in Tab 1.To compare PLD technique we managed production of CsI photocathode on Cu substrate with more traditional thermal evaporation technique.PLD showed worse maximum yield, but films were very homogenous and quantum yield in different spots of photoemission area did not differ more than 20%.Vapor sample showed very poor homogeneity, yield even disappeared in some areas.Also vapor sample demonstrated poor adhesion, photoemission film got away by slight rubbing by finger, PLD samples show very good adhesion, film endure high pressure friction by rubber material.Achieved quantum yield results are lower, than known in literature [4].We attribute this to the fact that the time spent in the air was too long so the coating had absorbed too much moisture.Annealing in the measurement vacuum chamber we did not performed in this work.

Table 1 .
Quantum yield measurement results.