Search

CuCrZr alloy is candidate heat sink material for the ITER blanket, first wall, and divertor. During ITER operation it will be exposed to a combination of elevated temperatures, heat flux, and intense fast neutron radiation. This environment will challenge the performance of components and joints based on CuCrZr. To address this issue, mechanical tests were performed with irradiated and reference specimens of CuCrZr and its joints with 316 L(N)-IG (ITER Grade) stainless steel made by hot isostatic pressing. The reactor exposure up to ∼0.7 dpa was performed in the BR2 reactor at SCK•CEN, in water at a temperature of 257 °C A special design was used to allow irradiation of specimens axially pre-stressed at different strain levels. The post-irradiation examination included: (i) tensile test, (ii) measurements of plastic deformation of samples axially loaded during irradiation (in situ creep test); (iii) thermal creep tests on irradiated samples. The fracture surfaces were examined in a hot cell using a Scanning Electron Microscope (SEM). The results were compared with data obtained from mechanical tests and SEM/EDX fracture surface analysis on non-irradiated reference samples. The level of a possible creep under irradiation is below the experimental uncertainty.

In order to keep the overall length of the compact linear collider (CLIC), currently being studied at the European Organization for Nuclear Research (CERN), within reasonable limits, i.e. less than 50 km, an accelerating gradient above 100 MV/m is required. This imposes considerable demands on the materials of the accelerating structures. The internal surfaces of these core components of a linear accelerator are exposed to pulsed radio frequency (RF) currents resulting in cyclic thermal stresses expected to cause surface damage by fatigue. The designed lifetime of CLIC is 20 years, which results in a number of thermal stress cycles of the order of 2.33•1010. Since no fatigue data existed in the literature for CLIC parameter space, a set of three complementary experiments were initiated: ultra high cycle mechanical fatigue by ultrasound, low cycle fatigue by pulsed laser irradiation and low cycle thermal fatigue by high power microwaves, each test representing a subset of the original problem. High conductivity copper alloys in different temper states and several techniques to improve their fatigue life were investigated. The results obtained by the three techniques are presented and the relations between them are determined. The RF fatigue experiments had conditions similar to the CLIC application, but the achievable number of cycles was limited. The data obtained by RF is extrapolated with the ultrasonic fatigue experiments which show a similar relative merit for the candidate alloys. Based on the results, a precipitation hardened copper zirconium alloy and an aluminum oxide dispersion strengthened copper alloy fulfill the CLIC requirements, the former being slightly better in terms of fatigue, but more sensitive to the temper state. The ultra-high cycle ultrasound data showed surface roughening, which appeared at stress amplitudes lower than the fatigue strength. Different crack growth rates were observed between the precipitation strengthened and the dispersion hardened copper alloys. Compressive mean stresses were studied at ultra high number of cycles regime and they were not found to have an effect on fatigue performance under mechanical loading. Surface damage, due to RF induced fatigue, was observed to be anisotropic.

The CLIC study group at CERN has built two X-band accelerating structures to be tested at SLAC in NLCTA. The structures consist of copper cells with insert irises made out of molybdenum and tungsten, clamped together and installed in a vacuum tank. These structures are exactly scaled versions from structures tested previously at 30 GHz and with short pulses (16 ns) in the CLIC Test Facility at CERN. At 30 GHz these structures reached gradients of 150 MV/m for tungsten and 195 MV/m for molybdenum. These experiments were designed to provide data on the dependence of rf breakdown on pulse length and frequency. This paper reports in particular on the high-gradient test of the tungsten-iris structure. At the shortest possible pulse length of 22 ns a gradient of 125 MV/m was reached at X-band, 20 % lower than the 150 MV/m measured at 30 GHz in the CLIC Test Facility. The pulse length dependence and the dependence of the break down rate as a function of gradient were measured in detail. The results are compared to data obtained from the molybdenum-iris experiment at X-band which took place earlier as well as to 30 GHz data.

The CLIC (Compact LInear Collider) is being studied at CERN as a potential multi-TeV e+e- collider. The acceleration of the particles is done by RF (Radio Frequency). The surfaces of the RF (radio frequency) accelerating cavities are exposed to high pulsed RF currents which induce cyclic thermal stresses. These cyclic stresses are crucial for the fatigue lifetime of the cavities. To study the fatigue phenomenon properly the induced stresses must be well known. ANSYS FEM simulations were made to study the thermo-structural behaviour of the CLIC accelerating structure in copper zirconium, bimetallic and diamond coated constructions. The simulations showed the existence of high thermal stresses and low stress level shockwaves. It was also shown that the bimetallic structure increases stress values due to the differences in material properties. Diamond coating was found to reduce the thermal stresses.

The need for high accelerating gradients for the future multi-TeV e+e- Compact Linear Collider (CLIC) imposes considerable constraints on the materials of the accelerating structures. The surfaces exposed to high pulsed RF (Radio Frequency) currents are subject to cyclic thermal stresses which are expected to induce surface break up by fatigue. Since no fatigue data exists in the literature up to very large numbers of cycles and for the particular stress pattern present in RF cavities, a comprehensive study of copper alloys in this parameter range has been initiated. Fatigue data for selected copper alloys in different states are presented

This note presents the CLIC parameter set as of mid 2005 and describes the different sub-systems, stressing how the design of the different components is driven. This design emerged from a better understanding of limitations for normal conducting accelerating structures, which led to a new optimised design for the CLIC 30 GHz accelerating structure. The structure parameters and improvements in other sub-systems have resulted in a major revision of the parameters. The overall layout and efciencies for CLIC with this updated parameter-set are presented.

The rf and mechanical designs, and cold measurement results of CTF3 30 GHz rf components are presented. The power levels and pulse lengths to which a number of these devices have been operated is reviewed.

The CLIC (Compact LInear Collider) is being studied at CERN (European Organization for Nuclear Research) as a possible future high-energy (0.5-5 TeV centre-of-mass) physics facility. The current aim of the CLIC Study Team is to demonstrate the key feasibility issues before 2010. CLIC will be about 33 kilometers long and will be buried 100 meters underground. The main linac of CLIC consists of accelerating structures with the following demanding performance requirements: accelerating gradients of about 150 MV/m, power flows of about 200 MW, 1-2 ìm dimensional tolerances, an optical-quality surface finish and ultimately a low mass production cost. About 80% of CLIC's 33 kilometer length will be filled with main beam accelerating structures, which will require of the order of ten thousand tons of raw material and millions of individual parts. One of the main limiting factors of the main linac accelerating structures is the thermal fatigue due to the pulsed surface heating. The accelerating RFpower pulses heat up repeatedly the surface of the structures by RF currents. These thermal stress cycles are expected to cause a surface break-up by fatigue.

Inspired by the very high gradients (150-195 MV/m) achieved at CERN in 30 GHz accelerator structures made with tungsten and molybdenum irises and operated with short (16 ns) rf pulses [1], an X-band (11.4 GHz) version of this structure design was built at CERN and tested at SLAC. The goals of this experiment were to provide frequency scaling data on high gradient phenomena at similar pulse lengths, and to measure the structure performance at the longer pulse lengths available at SLAC (the CLIC test facility, CTF II, could provide only 16 ns pulses for high power operation and 32 ns pulses for medium power operation). Earlier high gradient tests of 21 GHz to 39 GHz standing-wave, single cells indicated no significant frequency dependence of the maximum obtainable surface field [2]. The X-band scaling test would check if this was true for travelling-wave, multi-cell structures as well. For the experiment, the CLIC group at CERN built a 30 cell accelerating structure that consisted of copper cells and molybdenum irises that were clamped together. The structure was mounted in a vacuum tank and installed in the Next Linear Collider Test Accelerator (NLCTA) beam line at SLAC where it was operated at high power for more than 700 hours.

In order to achieve accelerating gradients above 150 MV/m, alternative materials to copper are being investigated by the CLIC study. The potential of refractory metals has already been demonstrated in tests in which a tungsten-iris and a molybdenum-iris structure reached 150 and 193 MV/m respectively (30 GHz and a pulse length of 15 ns). In order to extend the investigation to the pulse lengths required for a linear collider, a molybdenum-iris structure scaled to X-band was tested at the Next Linear Collider Test Accelerator (NLCTA). The structure conditioned to only 65 MV/m (100 ns pulse length) in the available testing time and much more slowly than is typical of a copper structure. However the structure showed no sign of saturation and a microscopic inspection of the rf surfaces corroborated that the structure was still at an early stage of conditioning. The X-band and 30 GHz results are compared and what has been learned about material quality, surface preparation and conditioning strategy is discussed.