[en] In a series of experiments several single cell and multi-cell niobium cavities made from reactor grade and high RRR niobium (frequencies were 700 MHz, 1300 MHz and 1497 MHz) have been baked--after initial testing--in-situ around 145 C for up to 90 hours prior to being recooled. Surprisingly, all cavities showed significant improvements in Q-values between 4.2 and 1.6K. The BCS surface resistance was lowered by nearly a factor of two. This cannot be explained by solely a reduction of dielectric losses caused by adsorbates at the surface or by a decrease of the mean free path due to possibly diffusion of oxygen into the surface layer. In several experiments also the high field behavior of the cavity improved after the in-situ baking procedure. The observed effect opens the possibility for the CEBAF upgrade cavities, which in turn will permit to run the cavities at higher gradients if field emission loading can be prevented. Utilizing this effect can possibly translate into sizeable cost savings since fewer modules are needed for the upgrade program

[en] Superconducting niobium cavities for application in particle accelerators are usually fabricated by standard techniques such as forming of subcomponents by deep drawing and joining by electron beam welding. Even though these techniques are being used successfully in many larger-scale accelerator projects and improvements in accelerating gradients have been achieved over the last several years, there are often still problems with making defect-free electron beam welds. In addition, the manufacturing costs for such devices are significant and a drastic reduction in production costs is a necessary condition for future very large scale applications in, e.g., linear colliders. Seamless cavities made by spinning from a single sheet of material will dramatically reduce the fabrication costs and eliminate any problems associated with electron beam welding. The fabrication technique for seamless niobium cavities has been developed over the last few years at INFN LNL and several proto type single-cells of different material thickness and purity have been manufactured as well as a 5-cell cavity. Results from tests on these cavities after application of surface treatment techniques, such as buffered chemical polishing, ''barrel polishing'' and high temperature heat treatments, are discussed in this contribution. Q-values as high as 10¹¹ and accelerating gradients up to E_acc approximately 30 MV/m have been measured

[en] The performance of a single-cell cavity made of RRR > 200 large-grain niobium has been investigated as a function of material removal by buffered chemical polishing. Temperature maps of the cavity surface at 1.7 and 2.0 K were taken for each step of chemical etching and revealed several ''hot-spots'', which contribute to the degradation of the cavity quality factor as a function of the RF surface field, mostly at high field levels. It was found that the number of ''hot-spots'' decreased for larger material removal. Interestingly, the losses of the ''hot-spots'' at different locations evolved differently for successive material removal. The cavity achieved peak surface magnetic fields of about of 130 mT and was limited mostly by thermal quench. By measuring the temperature dependence of the surface resistance at low field between 4.2 K and 1.7 K, the variation of niobium material parameters as a function of material removal could also be investigated. This contribution shows the results of the RF tests along with the temperature maps and the analysis of the losses caused by the ''hot-spots''

[en] This paper reports on the photoemission observed during the tests on a single-cell 1500 MHz niobium cavity with a polycrystalline alumina rf interface window operated at 2K. The light emission from the window at 2K was detected using a multi-channel fiber optics/PMT system and recorded with a 100 Ms/sec acquisition system. The spectral composition of two types of discharges has been measured. The results indicate that both broadband and narrow-band emissions are present, depending on the discharge type

[en] Four single cell niobium cavities fabricated from Tokyo-Denkai material of RRR=200 have been tested repeatedly with the purpose to evaluate different fabrication and processing techniques used at KEK and Jefferson Lab, respectively. Two cavities--K-15 and K-16--have been manufactured completely at KEK prior to shipment to Jefferson Lab. In addition, K-16 had received a barrel polishing treatment at KEK, resulting in the removal of 40 μm of material from the surface. Cavity K-17 was electron-beam welded at Jefferson Lab; the deep drawing of the half cells and the trimming of the cups for electron-beam welding were done at KEK, however. Cavity JL-1 was completely fabricated at Jefferson Lab. Often, some processing field levels related to electronic activity in the cavities, possibly multipacting, have been seen at KEK and the purpose of this investigation is a verification of such observations. In addition, a comparison of different fabrication procedures and surface treatments are of interest for optimizing cost and performance for larger scale application. In several cavities, accelerating gradients between 20 MV/m ≤ E_acc ≤ 27 MV/m have been measured with only little field emission loading. In one of the cavities, resonant electron loading was ''provoked'' by rinsing it with oil contaminated acetone. The observed multipacting levels at E_acc=13 MV/m and 25 MV/m could be identified with the help of simulation calculations as 1-point and 2-point multipacting across the equator of the cavity. There is - as previously reported - a rather strong dependence of the quench field levels on the amount of material removed from the surface, confirming a picture of a surface damage layer which becomes depleted of defects as more and more of it is removed

[en] In a recent contribution to PAC '99, we reported about results from experiments with single-cell and one five-cell seamless cavities, which had undergone various surface treatments such as buffered chemical polishing, tumbling, grinding and high temperature heat treatments. Q-values as high as 10¹¹ and accelerating gradients up to E_acc = ∼ 30 MV/m had been measured. In the last several months we have carried out several more experiments on additional cavities made from seamless tubes and the results are discussed in this contribution. Also, one of the cavities mentioned in the PAC contribution was electropolished and was tested after only high pressure rinsing and in-situ baking. Results from these experiments are reported in this contribution

[en] The SNS linac contains superconducting cavities with phase velocities .61 and .81 for the acceleration of H- ions. Each of the 6-cell cavities, resonant at 805 MHz at 2 degrees K, is powered by a 550 kW pulsed klystron via a coaxial coupler. The specifications for the coupler require that it withstand the full klystron power in full reflection for the duration of the RF pulse (1.3+1.5(decay) ms) at a repetition rate of 60 pps, with an average power of 48 kW. The coupler must provide a Qext of 7.3 x 105 for the medium beta and high beta cavities, respectively. A design derived from proven experience at other laboratories (primarily at KEK, for Tristan and KEK-B cavities at 508 MHz) has been adopted for the production of the 50 omega prototype couplers. The couplers include a planar coaxial window with matching compensation elements, cooled by conduction via a water system from the air side of the inner conductor. The coupler's outer conductor between the window and the cavity is actively cooled by a helium circuit with input at 5 degrees K and output at about 165 degrees K. Vacuum, arcing, infrared, and electron current monitoring and interlocking is done through several ports located near the window. The coaxial coupler is matched to a WR975 waveguide through a door-knob transition, also derived from the KEK design. Numerical calculations have been performed on the electromagnetic properties of the coupler and on the multipacting behavior of the coaxial line and window. A capacitor for DC bias will be used to control multipacting. The first coupler prototypes have been produced by industry and the results of the RF testing on them at room temperature will be presented

[en] Secondary Ion Mass Spectrometry (SIMS) can characterize the surface and near surface of Nb used in accelerator cavities. Results show Nb oxide in the 2-3 nm range, a depleted H concentration in the oxide compared with the bulk, and N, C, O lower in an annealed single crystal sample than several polycrystalline samples. Other metallic contaminants are primarily at the surface, but tantalum is distributed uniformly through the material

[en] In a previous paper [1] we have reported about initial tests of single cell 1500 MHz cavities made from high purity niobium with three different Ta contents of 160 ppm , ∼600 ppm and ∼1400 ppm. These cavities had been treated by buffered chemical polishing several times and 100 (micro)m, 200 (micro)m and 300 (micro)m of material had been removed from the surfaces. This contribution reports about subsequent tests following post purification heat treatments with Ti and ''in situ'' baking. As a result, all cavities exhibited increased quench fields due to the improved thermal conductivity after the heat treatment. After the ''in situ'' baking at 120 C for ∼ 40 hrs the always present Q-drop at high fields disappeared and further improvements in accelerating gradient could be realized. Gradients as high as E_acc = 35 MV/m were achieved and there were no clear indications that the cavity performance was influenced by the Ta content in the material. A multi-cell cavity from the high Ta content material has been fabricated and initial results are reported

[en] The multipacting phenomena in accelerating structures and coaxial lines are well documented and methods of mitigating or suppressing it are understood. The multipacting that occurs in a quarter wave choke joint designed to mount a cathode insertion stalk into a superconducting RF photoinjector has been analyzed via calculations and experimental measurements and the effect of introducing multipacting suppression grooves into the structure is analyzed. Several alternative choke joint designs are analyzed and suggestions made regarding future choke joint development. Furthermore, the problems encountered in cleaning the choke joint surfaces, factors important in changes to the secondary electron yield, are discussed and evaluated. This design is being implemented on the BNL 1.3 GHz photoinjector, previously used for measurement of the quantum efficiency of bare Nb, to allow for the introduction of other cathode materials for study, and to verify the design functions properly prior to constructing our 703 MHz photoinjector with a similar choke joint design