H. Hseuh

The Brookhaven muon storage ring magnet

Abstract The muon g-2 experiment at Brookhaven National Laboratory has the goal of determining the muon anomalous g-value a μ (=(g−2)/2) to the very high precision of 0.35 parts per million and thus requires a storage ring magnet with great stability and homogeniety. A superferric storage ring with a radius of 7.11 m and a magnetic field of 1.45 T has been constructed in which the field quality is largely determined by the iron, and the excitation is provided by superconducting coils operating at a current of 5200 A. The storage ring has been constructed with maximum attention to azimuthal symmetry and to tight mechanical tolerances and with many features to allow obtaining a homogenous magnetic field. The fabrication of the storage ring, its cryogenics and quench protection systems, and its initial testing and operation are described.

The superconducting inflector for the BNL g-2 experiment

The muon g-2 experiment at Brookhaven National Laboratory (BNL) has the goal of determining the muon anomalous magnetic moment, a(mu) (= (g-2)/2), to the very high precision of 0.35 parts per million and thus requires a storage ring magnet with great stability and homogeneity. A super-ferric storage ring has been constructed in which the field is to be known to 0.1 ppm. In addition, a new type of air core superconducting inflector has been developed and constructed, which successfully serves as the injection magnet. The injection magnet cancels the storage ring field, 1.5 T, seen by the entering muon beam very close to the storage ring aperture. At the same time, it gives negligible influence to the knowledge of the uniform main magnetic field in the muon storage region located at just 23 rum away from the beam channel. This was accomplished using a new double cosine theta design for the magnetic field which traps most of the return field, and then surrounding the magnet with a special superconducting sheet which traps the remaining return field. The magnet is operated using a warm-to-cold cryogenic cycle which avoids affecting the precision field of the storage ring. This article describes the design, research development, fabrication process, and final performance of this new type of superconducting magnet

Outgassing and surface properties of TiN coated SNS ring vacuum chambers

The stainless steel vacuum chambers of the 248 m Spallation Neutron Source (SNS) accumulator ring are being coated with /spl sim/ 100 nm of titanium nitride (TiN) to reduce the secondary electron yield (SEY). The coating is produced by DC magnetron sputtering using a long cathode imbedded with permanent magnets. The outgassing rates of several SNS half-cell chambers, with and without TiN coating, and before and after in-situ bake, were measured. The SEY of the coated chamber coupons were also measured with primary electron energy of 50-3000 eV. By varying the coating parameters, films of different surface roughness were produced and analyzed by scanning electron microscopy. It was found that the outgassing rate varies as a function of surface roughness of the TiN layer, and chambers with a rougher surface have a lower SEY.

Hydrogen Outgassing and Surface Properties of TiN‐Coated Stainless Steel Chambers

The stainless steel vacuum chambers of the 248m accumulator ring of Spallation Neutron Source (SNS) are coated with ∼ 100 nm of titanium nitride (TiN) to reduce the secondary electron yield. The coating is produced by DC magnetron sputtering using a long cathode imbedded with permanent magnets. The outgassing rates of several SNS half‐cell chambers were measured with and without TiN coating, and before and after in‐situ bake. One potential benefit of a TiN coating is to serve as hydrogen permeation barrier that reduces the ultimate outgassing rate. By varying the coating parameters, films of different surface roughness were produced and analyzed by Auger electron spectroscopy, scanning electron microscopy and atomic force microscopy to illustrate the dependence of the outgassing on the film structure.

Commissioning of RHIC vacuum systems

The Relativistic Heavy Ion Collider (RHIC) has two concentric rings 3.8 km in circumference, There are three vacuum systems in RHIC; the insulating vacuum vessels housing the superconducting magnets, the cold beam tubes surrounded by the superconducting magnets, and the warm beam tube sections at the insertion regions and experimental regions. The vacuum requirements and the design of three vacuum systems are described. The experience gained during the commissioning of these vacuum systems is presented with emphasis on locating helium leaks in the long arc insulating vacuum system.

Construction and Power Test of the Extraction Kicker Magnet for Spallation Neutron Source Accumulator Ring

Two extraction kicker magnet assemblies that contain seven individual pulsed magnet modules each will kick the proton beam vertically out of the SNS accumulator ring into the aperture of the extraction Lambertson septum magnet. The proton beam then travels to the 1.4 MW SNS target assembly. The 14 kicker magnets and major components of the kicker assembly have been fabricated by BNL. The inner surfaces of the kicker magnets were coated with TiN to reduce the secondary electron yield. All 14 power supplies have been built, tested and delivered to ORNL. Before final installation, a partial assembly of the kicker system with three kicker magnets was assembled to test the functions of each critical component in the system. In this paper we report the progress of the construction of the kicker components, the TiN coating of the magnets, the installation procedure of the magnets and the full power test of a kicker magnet with the power supply.

Design and implementation of SNS ring vacuum system with suppression of electron cloud instability

The Spallation Neutron Source (SNS) ring is designed to accumulate, via H injection, 1.6x10 protons per pulse at 60 Hz and 1 GeV energy [1]. At such beam intensity, electron cloud is expected to be one of the intensity-limiting mechanisms that will complicate ring operation. This paper presents the design of the ring vacuum system and the mitigation strategy adopted to suppress the electron cloud instability. These measures include the titanium nitride (TiN) coating of the chamber walls to reduce the secondary electron yield (SEY), the tapered magnetic field for the collection of the stripped electrons at injection, clearing electrode dedicated for the injection region and parasitic ones using the BPMs around the ring, solenoid windings in the field free regions, and the possibility of beam scrubbing at high pressure. SNS RING VACUUM SYSTEM The SNS ring, with a four-fold symmetry and a circumference of 248 m, consists of 4 arc sections of 34 m each with FODO lattice and 4 straight sections of 28 m each with doublet lattice [1]. Each ring arc section has 8 halfcells and one quartercell. The halfcell magnets and vacuum chambers [2] are symmetrically grouped to the middle quartercell, such that they are mirror images to each other with respect to the center quartercell. This strategy reduces the individual chamber components into two types, lowering the cost of fabrication and assembly. Each halfcell chamber, as shown in Fig. 1, consists of a dipole chamber section, pump ports, quadrupole section, BPM and bellows. The 2 m long dipole chamber has a very large elliptical cross section of 23 cm (H) x 17 cm (V), curved with a bending angle of 11.25. The halfcell chambers are ~ 4 m long and made of stainless steel 316L or 316LN. To avoid radiation induced stress corrosion, the bellows are fabricated with inconel 625. In all, there are four types of halfcell chambers; three with 21 cm quadrupole pipe and one with 26 cm quadrupole pipe. * SNS is managed by UT-Battelle, LLC, under contract DEAC05-00OR22725 for the U.S. Department of Energy. SNS is a partnership of six national laboratories: Argonne, Brookhaven, Jefferson, Lawrence Berkeley, Los Alamos, and Oak Ridge. Corresponding author email: hseuh@bnl.gov . Fig.1 Schematics of SNS ring arc halfcell chambers. Each chamber is ~4 m long and fabricated with 316L stainless steel and inconel bellows. The four straight sections are dedicated for injection, collimation, extraction, and RF and diagnostics. Each straight section has two sets of doublet magnets. The two doublet chambers in each section are identical, except at the extraction section, which reduces the number of spares for the future. Each doublet chamber consists of a 30 cm OD straight pipe welded to the 30cm bellows and BPM. The length of the doublet chambers ranges from 3 m to 5 m, as shown in Fig. 2. The balance of the straight sections is occupied by special chambers housing the injection, collimation, rf, extraction and diagnostic equipment. Fig. 2. Schematics of typical SNS ring straight section doublet chambers for 30 cm quadrupole doublets. The electron-cloud effect in the SNS ring is expected to be one of the intensity-limiting mechanisms and a potential threat to the high-intensity operations [3]. Main sources of electrons are: electrons generated at the injection stripping foil, from proton grazing losses at the collimator surface and other chamber wall, beam induced multipacting, and from beam residual gas ionization. The electron-cloud effects include neutralization tune shift and resonance crossing, electron-cloud instability, emittance growth and beam loss, pressure increase, heating of the vacuum pipe and interference with beam diagnostics. To combat the electron-cloud effects, several measures have been adopted at the design stage of the vacuum systems and implemented. Detailed description of these measures is given in the following sections. TITANIUM NITRIDE COATING One major contributing factor to the electron multipacting is the secondary electron emission (SEY) of the vacuum surface facing the beam. Most SNS ring chambers are fabricated from stainless steel, which has a peak SEY of ~ 2.5. The SEY can be reduced to < 2 if the surface is coated with titanium nitride (TiN) [4]. TiN coating has been routinely applied to high power RF windows and tuners to reduce multipacting. TiN coating of regular accelerator beam tubes was done for PEPII LER [5] using DC sputtering. Due to the large cross sections of the SNS chambers, magnetron DC (MDC) sputtering was developed for TiN coating for its high deposition rate, low operating voltage and pressure [6]. This is a result of the increased plasma density formed by the electrons confined within the magnetic field, which adds to the sputtering rate. Improved stoichiometry and uniformity were also achieved with MDC. NEG coating was considered but not adopted since the ring vacuum system is not designed for in-situ bake, which is needed to activate NEG and achieve low SEY for the coated surface. In-situ bake of the SNS ring vacuum system posts high risk to the large aperture flanges which tend to leak due to relative thermal motion. Reliability of the bakeout systems over long period of time in a high radiation environment is also not proven. Fig. 3 A long titanium cathode for magnetron DC sputtering of TiN coating for SNS vacuum chambers. The permanent magnets/spaces enhance the discharge plasma density thus the deposition rate. Much work has been done on the formation of TiN by magnetron sputtering as an industrial hard coating using planar electrodes and magnets. Due to the SNS chamber geometry, a linear titanium cathode with a suitable magnetic field was developed as shown in Fig. 3. Commercially available Alnico magnets are inserted in a 1.5” diameter titanium tube used as a cathode. The magnets are stacked with 0.5” spacers resulting in a looping magnetic field of several hundred gauss projected from the cathode surface. The 0.5” diameter hole in the center of the magnets allows for water cooling of the cathode. This “low cost” cathode works in conjunction with a 10 KW DC power supply to produce a satisfactory field and discharge plasma as shown in Fig. 4. Fig. 4 The discharge plasma during TiN coating of the SNS ring vacuum chambers. The brighter rings are the locations of the spacers between the permanent magnets. Sample coupons from the coated chambers were analyzed by Auger Electron Spectroscopy (AES). Typical results of AES analysis are shown in Fig. 5. with little oxygen contamination and the correct stoichiometry ratio of Ti and N. The SEY of the coupons were measured by colleagues at CERN in as-received condition, i.e. without in-situ bake and with very low accumulative dosage. The coated samples have significantly lower SEY values when compared with bare stainless as shown in Fig. 6. AES Survey Spectra 0 200 400 600 80

Improvement of RHIC warm beam vacuum for high intensity operation

With increasing ion beam intensity during recent RHIC operations, pressure rises of several decades were observed at a few warm vacuum sections. Improvement of the warm sections has been carried out in last years shutdown. Extensive in-situ bakes, additional UHV pumping, electron detectors and beam tube solenoids have been implemented. Vacuum monitoring and interlock were enhanced to reduce premature beam aborts. The effectiveness of these measures in reducing the beam induced pressure bumps and in increasing the vacuum system reliability are discussed and summarized.

Status of the BNL muon (g−2) experiment

The muon (g−2) experiment at Brookhaven has just completed a 3-month run for checkout and initial data-taking. In the first two months beam was taken in a parasitic mode where one out of ten AGS pulses was delivered for commissioning of the beam line, quadrupoles, detectors, and data acquisition system. This was followed by four weeks of dedicated data collection. The main components of the experiment, which include the pion/muon beam line, the superconducting inflector, the superferric storage ring with its pulsed electric quadrupoles and magnetic field measurement system, and the detector system based on lead-scintillating fiber electron calorimeters, have been satisfactorily commissioned. The muon (g−2) precession frequency is clearly seen as a large signal. It is estimaed that over 25×106 decay positrons with energies greater than 1.5 GeV have been detected.

Beam vacuum chambers for Brookhaven's muon storage ring

An experiment is being built at Brookhaven to measure the g-2 value of the muons to an accuracy of 0.35 ppm. The muon storage ring of this experiment is designed to produce a dipole field with homogeneity to 1 ppm using a continuous superconducting magnet. The beam vacuum system in the storage ring will operate at 10/sup -7/ Torr and consists of twelve sector chambers. The chambers are constructed of aluminum and are approximately 3.5 m in length with a rectangular cross-section of 16.5 cm high by 45 cm at the widest point. The design features, fabrication techniques and cleaning methods for these chambers are described. Monte Carlo simulation of the pressure distribution and finite element analysis of the chamber deflection are summarized with good correlation shown to measured values obtained during tests of the prototype chamber.