Gianluigi Ciovati

Effect of low-temperature baking on the radio-frequency properties of niobium superconducting cavities for particle accelerators

Radio-frequency superconducting (SRF) cavities are widely used to accelerate a charged particle beam in particle accelerators. The performance of SRF cavities made of bulk niobium has significantly improved over the last ten years and is approaching the theoretical limit for niobium. Nevertheless, RF tests of niobium cavities are still showing some “anomalous” losses that require a better understanding in order to reliably obtain better performance. These losses are characterized by a marked dependence of the surface resistance on the surface electromagnetic field and can be detected by measuring the quality factor of the resonator as a function of the peak surface field. A low-temperature (100–150°C) “in situ” bake under ultrahigh vacuum has been successfully applied as final preparation of niobium RF cavities by several laboratories over the last few years. The benefits reported consist mainly of an improvement of the cavity quality factor at low field and a recovery from “anomalous” losses (so-called “...

Preliminary Results From Single Crystal and Very Large Crystal Niobium Cavities

We have fabricated and tested several single cell cavities using material from very large grain niobium ingots. In one case the central grain exceeded 7“in diameter and this was used to fabricate two 2.2 GHz cavities. This activity had a dual purpose: to investigate the influence of grain boundaries on the often observed Q-drop at gradients Eacc> 20 MV/m in the absence of field emission, and to study the possibility of using ingot material for cavity fabrication without going through the expensive rolling process. The sheets for these cavities were cut from the ingot by wire electro-discharge machining (EDM) and subsequently formed into half–cells by deep drawing. The following fabrication steps were standard: machining of weld recesses, electron beam welding of beam pipes onto the half cells and final equator weld to join both half cell/beam pipe subunits. The cavities showed heavy Q–disease caused by the EDM. After hydrogen degassing at 800 ° C for 3 hrs in UHV and about 200 μm total removals from the inner surface by BCP 1: 1: 1, the cavities showed promising results, however, the Q-drop was still present. In the two cavities made from large grain material accelerating gradients of 30 MV/m have been reached. After “in-situ” baking the Q-drop disappeared. The smaller cavities made from single crystal material showed very low residual resistances and accelerating gradients up to Eacc= 45 MV/m were reached (one of the highest ever achieved), corresponding to a peak surface magnetic fields (Bp) of 160 mT.

Cavities for JLab's 12 GeV upgrade

The future 12 GeV upgrade of CEBAF requires new cryomodules in both linacs to increase the energy gain per pass to 1090 MeV. Until recently, the design of new cryomodules, which should deliver on average operational voltage of 70 MV each, was based on 7-cell superconducting cavities that are an extended version of the 5-cell structures currently used in the machine. The 5-cell cavities were constructed 20 years ago at Cornell University (Original Cornell-shape) for the Cornell Electron Storage Ring (CESR). The geometry of these structures met specifications at the time CESR was constructed but is not optimized for the future operation of CEBAF. Two improved cavity shapes have been proposed. This contribution presents the RF features of both new shapes and discusses advantages for the machine operation resulting from the improvement. In addition, we comment on the measurements on copper models of both new cavities and present results of the multipacting calculations.

Decrease of the surface resistance in superconducting niobium resonator cavities by the microwave field

Measurements of the quality factor, Q, of Nb superconducting microwave resonators often show that Q increases by ≃10%–30% with increasing radio-frequency (rf) field, H, up to ∼15–20 mT. Recent high temperature heat treatments can amplify this rf field-induced increase of Q up to ≃50%–100% and extend it to much higher fields ≃100 mT, but the mechanisms of the enhancement of Q(H) remain unclear. Here, we suggest a method to reveal these mechanisms by measuring temperature dependencies of Q at different rf field amplitudes. We show that the increase of Q(H) does not come from a field dependent quasi-particles activation energy or residual resistance, but rather results from the smearing of the density of state by the rf field.

Improved oxygen diffusion model to explain the effect of low-temperature baking on high field losses in niobium superconducting cavities

Radio-frequency (rf) superconducting cavities made of high purity niobium are widely used to accelerate charged particle beams in particle accelerators. The major limitation to achieve rf field values approaching the theoretical limit for niobium is represented by “anomalous” losses which degrade the quality factor of the cavities starting at peak surface magnetic fields of about 100mT, in the absence of field emission. These high field losses are often referred to as Q drop. It has been observed that the Q drop is drastically reduced by baking the cavities at 120°C for about 48h under ultrahigh vacuum. An improved oxygen diffusion model for the niobium-oxide system is proposed to explain the benefit of the low-temperature baking on the Q drop in niobium superconducting rf cavities. The model shows that baking at 120°C for 48h allows oxygen to diffuse away from the surface, and therefore increasing the lower critical field towards the value for pure niobium.

Development of Large Grain/Single Crystal Niobium Cavity Technology at Jefferson Lab

Approximately two years ago we started to develop high performance niobium accelerating cavities based on large grain or single crystal high purity niobium. We have fabricated and tested 15 single cell cavities of various shapes and frequencies between 1300 MHz and 2300 MHz using material from a total of 9 different very large grain niobium ingots from four niobium suppliers. The materials differed not only in grain sizes, but also in RRR — value and in the amount of Ta contained in the material. In one ingot supplied by CBMM the central grain exceeded 7 inches in diameter and this was used to fabricate two 2.2 GHz cavities. A single crystal 1300 MHz mono‐cell cavity was also produced at DESY by rolling out a single crystal to the size required for this cavity. It was sent to Jlab for surface treatment and testing. In addition, we have fabricated three 7‐cell cavities: two of the Jlab high gradient (HG) shape and one of the ILC Low Loss shape. Two 9‐cell TESLA shape cavities are presently in fabrication a...

Testing of HOM Coupler Designs on a Single Cell Niobium Cavity

Coaxial higher order mode (HOM) couplers were developed initially for HERA cavities and subsequently for TESLA cavities. They were adopted later for SNS and Jlab upgrade cavities. The principle of operation is the rejection of the fundamental mode by the tunable filter and the transmission of the HOMs. It has been recognized recently that for continuous wave or high duty factor applications of the TESLA coupler the output pick-up probe must stay superconducting in order to avoid its heating by the fundamental mode residual magnetic field leading to deterioration of the cavity quality factor. In addition, the thermal conduction of existing rf feedthrough designs is only marginally sufficient to keep even the niobium probe tip superconducting in cw operation. We have equipped a single-cell niobium cavity with the modified HOM couplers and tested the new designs by measuring Q vs Eaccbehavior at 2 K for different feedthroughs and probe tip materials.

Status report on multi-cell superconducting cavity development for medium-velocity beams

The Rare Isotope Accelerator (RIA) is being designed to supply an intense beam of exotic isotopes for nuclear physics research. Superconducting cavities are to be used to accelerate the CW beam of heavy ions to 400 MeV per nucleon, with a beam power of up to 400 kW. Because of the varying beam velocity, several types of superconducting structures are needed. This paper covers the fabrication of three prototype RIA 6-cell /spl beta//sub g/ = 0.47 cavities and the RF tests on the first and second of these cavities.

Superconducting Radio-Frequency Technology R&D for Future Accelerator Applications

Superconducting rf (SRF) technology is evolving rapidly, as are its applications. While there is active exploitation of what one may call the current state-of-the-practice, there is also rapid progress in expanding in several dimensions the accessible and useful parameter space. While state-of-the-art performance sometimes outpaces thorough understanding, the improving scientific understanding from active SRF research is clarifying routes to obtain optimum performance from present materials and opening avenues beyond the standard bulk niobium. The improving technical basis understanding is enabling process engineering to improve both performance confidence and reliability and also unit implementation costs. Increasing confidence in the technology enables the engineering of new creative application designs. We attempt to survey this landscape to highlight the potential for future accelerator applications.

Residual Resistance Data From Cavity Production Projects at Jefferson Lab

A fundamental limitation towards achieving high quality factors in superconducting radio-frequency cavities is the so-called residual resistance. Understanding and controlling the residual resistance has important implications towards improving the efficiency and reduce the operating cost of continuous wave superconducting linear accelerators. In this contribution we will report on the residual resistance values obtained from measurements of the quality factor of a large set of cavities, with resonant frequency between 805 MHz and 1.5 GHz, all of them processed and tested at Jefferson Lab. Surface treatments included both buffered chemical polishing and electropolishing. The results indicate an approximate value of the residual resistance of about 7-10 nΩ.