Linear Injector Research Articles

H- beam formation simulation in negative ion source for CERN's Linac4 accelerator

The caesiated surface negative ion (H-) source is the first element of CERN's LINAC4 a linear injector designed to accelerate negative hydrogen ions to 160 MeV. The IS03 ion source is operated at 35 mA beam intensity and reliably feeds CERN's accelerator chain, H- ions are generated via plasma volume and caesiated molybdenum (Cs-Mo) plasma electrode surface mechanisms. Studying the beam extraction region of this H- ion source is essential for optimizing the H- production. The 3D Particle-in-cell (PIC) Monte Carlo (MC) code ONIX (Orsay Negative Ion eXtraction [1]), written to study H- beam formation processes in neutral-beam injectors for fusion, has been adapted to single aperture accelerator H- sources. The code was modified to match the conditions of the beam formation and extraction regions of the Linac4 H- source [2]. A set of parameters was chosen to characterize the plasma and to match the specific volume and surface production modes. Simulated results of the extraction regions are presented and benchmarked with experimental results obtained at the Linac4 test stand [3].

Correlation of H- beam properties to Cs-coverage

A caesiated RF driven source delivers H- ions that, after stripping at the end of the 160 MeV H- linear injector, provides protons to CERN's accelerator complex including LHC, where the protons reached a record energy of 6.8 TeV. In Caesiated RF sources, H- ions are produced via dissociative attachment of electrons onto roto-vibrationally excited H2-molecules (volume) and re-emission as negative ions of protons or hydrogen atoms colliding on a low work function caesiated molybdenum plasma electrode (surface). During initial caesiation, the production mechanism evolves from the initial Cs-free volume production to a predominant surface production mode; the observed stunning reduction of co-extracted electrons is concomitant to an increase of the H- ion current to RF-power yield. This paper describes the evolution of the beam-profile at today's operational beam intensities of 35 mA for various ratios of volume and surface ion-origin. The presence of surface produced ions occurring on a conical plane is characterized by the electron to ion ratio and by measurement of the Cs-coverage of the molybdenum plasma electrode down to a fraction of a monolayer. Angular distributions are extracted from beam profile and Beam Emission Spectroscopy (BES) measurements. These experimental results provide an initial comparison to beam formation simulation that, at a later stage, could be coupled to beam transport software packages.The paper focuses on the caaesiation transient to present experimental evidence for 3D beam formation studies, it provides insight into the mixing of volume and surface production modes, reduction of co-extracted electrons and Cs-coverage. The paper also establishes magnetic field induced asymmetries in the beam's current density.

Physical design of a compact injector for synchrotron-based proton therapy

A compact room-temperature linear injector has been purposed to accelerate an 18.0 mA proton beam to 7.0 MeV for synchrotron-based proton therapy. The total length is appropriately 5 m. It mainly consists of a 3.01 m radio frequency quadrupole (RFQ) and a 0.82 m compact interdigital H-mode drift tube linac (IH-DTL) structure. Based on a fast-bunching strategy, the RFQ, operated at 325 MHz, accelerates protons to 3.0 MeV. The phase advances have been taken into consideration, and parametric resonance has been carefully avoided by adjusting the vane parameters. After the modulation of the transverse and longitudinal phase advances and a compact external quadrupole triplet, the proton beam is injected into the subsequent IH-DTL. Based on modified Kombinierte Null Grad Strukter (KONUS) beam dynamics, it accelerates protons up to 7.0 MeV, which is composed of a re-bunching section and an accelerating section. The accelerating gradient reaches 4.88 MV/m. The overall dynamic simulation results show that the whole accelerating gradient reaches up to 1.62 MV/m with a transmission efficiency above 95%. The transverse and longitudinal normalized RMS emittances at the exit of the DTL are 0.23 π mm·mrad and 2.216 π keV/u·ns, which meet the synchrotron injection requirements. The details of the specific design of this injector are presented in this paper.

Beam Formation Studies on the CERN IS03b H Source

An H- ion source is being operated at the new 160 MeV linear injector (Linac4) of the CERN accelerator complex. The source’s plasma is of the Radio Frequency Inductively Coupled Plasma type (RF-ICP), without magnetic cusp and runs with Cs-loss compensation [1]. Vertical downward oriented filter- and electron dump-dipolar magnetic fields expand over the plasma chamber, beam-formation, beam-extraction and electron dump regions and generate horizontal asymmetry and beam angular deflection partially compensated by mechanical alignment of the front-end. The H- beam is generated via volume and caesiated plasma surface modes, the latter inducing a radial asymmetry characterized by an increased current density close to the plasma electrode surface [2]. Asymmetries affecting the meniscus shape, or its current density have to be simulated via 3D Particle In Cell Monte Carlo (PIC-MC) solvers, such as the Orsay Negative Ion eXtraction code (ONIX) [3]. Validation of these simulations require dedicated measurements. This contribution describes experimental methods newly implemented at CERN’s ion source test stand and initial results for Optical and Beam Emission Spectroscopy (OES, BES), emittance and beam profile measurements. In a later stage, the gathered data sets can be compared to source plasma parameters and extracted beam parameters from PIC-MC simulations, once coupled to the Ion Beam Simulation (IBSimu) [4] beam transport code. The experimental parameter space includes RF-power, density of neutrals, position of the RF coil and extraction field. Beams of H-, D- and protons were produced; examples of measured data are presented in this contribution.

A CMOS Timer-Injector Integrated Circuit for Self-Powered Sensing of Time-of-Occurrence

Self-powered sensing of the time-of-occurrence of an event is challenging, because it requires access to a reliable time reference or a synchronized clock. In this paper, we propose for the first time a self-powered integrated circuit that is capable of time-stamping asynchronous mechanical events of interest. The core of the proposed design is the integration of two self-powered modules: 1) a chip-scale Fowler–Nordheim tunneling-based timer array, for generating a precision, relative time reference; and 2) a linear piezoelectricity-driven hot-electron injector acting as a floating-gate memory to record the onset of mechanical events. This paper presents measured results from a $4 \times 4$ fully programmable timer array system-on-chip (SoC) and a linear injector array SoC, both of which have been prototyped in a standard double-poly CMOS process. The synchronization error of the timer array with respect to an external software clock was measured to be less than 1% over a duration of 100 h, and the average accuracy in sensing the time-of-occurrence of the event was measured to be 6.9%. The minimum activation energy of the self-powered system was measured to be 840 nJ (measured for event durations of 1 s), which is significantly lower than the energy that can be harvested from typical mechanical impacts.

Parametric Evaluation of Swirl Injector Dynamics in the High-Frequency Range

This work augments the classical linear swirl injector dynamics theory to evaluate the hydrodynamic frequency response characteristics of a single element as an analysis aid for experimental results from cold flow testing. Specific focus is placed on how surface waves within the swirl injector are modeled. A modified formulation of wave interactions within the vortex chamber includes disturbance origins displaced from the posterior wall of the injector. Differing phase delays in both downstream and upstream directions are accounted for by dispersive calculations valid at all frequencies rather than by the traditional long wave approximation. The model is semiempirically tailored to hydraulic conditions for a study element, and a response function analysis is presented for an example case. Distinct features in the range of 2000–4000 Hz are found to be predominately related to surface wave patterns within the injector’s vortex chamber. Comparisons are made to experimental data, showing qualitative agreement with injector response and detection of computed high-frequency surface wave responses. Calculations are then mapped over steady mass flow conditions that range from 39 to 89% of the injector’s nominal design point. The results show that response features associated with surface wave dynamics consistently manifest across the entire parameter space investigated for the study injector element.

High power microwaves generation from intense relativistic electron beam using Kali-1000 pulse power system

Intense relativistic electron beam (IREB) is generated using Kali-1000 (Kilo Ampere Linear Injector) pulse power system, providing input electric power of 4 GW and stored energy of 1000 J. IREB is used for generating Non Ionizing (High Power Microwaves, HPM) and Ionizing (Flash X-rays) radiation. The present manuscript discusses generation of HPM by Kali-1000 system using an axial Virtual cathode oscillator. Kali-1000 pulse power system is a low impedance (~4 Ω) high voltage generator, consisting of Radial Tesla transformer to step up the voltage from 28 to 300 kV in 5 μs, which energizes the line, pulse discharged by a self triggered pressurized SF6 closing switch across a Relativistic Electron Beam (REB) diode load. The typical electron beam parameters are 300 kV, 15 kA and 100 ns with a current density of 3–4 hundreds of amperes per square centimeter. Experiments are carried with a REB diode consisting of graphite cathode and SS anode mesh. Measured pulse duration of HPM pulse (FWHM) varies from 70 to 110 ns on shot-to-shot basis. The major operating output frequencies have been measured to be 5.8 and 6.2 GHz within a range of 4.8–6.7 GHz using B-dot magnetic field sensor coupled to high frequency digital oscilloscope. Microwave output power from the virtual cathode oscillator has been measured to be 50 MW using double ridged waveguide horn antenna coupled with Schottky detector diode. The estimated beam to microwave power conversion efficiency is thereby found to be 1.1%. Particle-in-Cell (PIC) simulation has been performed for a set of Vircator experiments being driven by the Kali-1000 system. Experimentally measured cathode current, dominant output frequency, radiation pattern, etc., have been compared with simulation results and are in good agreement. The dominant emission mode from the virtual cathode oscillator is shown to be the TE01 mode based on the emission pattern of microwave induced air breakdown in this experiment.

The development of a room temperature electron cyclotron resonance ion source (Lanzhou electron cyclotron resonance ion source No. 4) with evaporative cooling technology at Institute of Modern Physics.

LECR4 (Lanzhou electron cyclotron resonance ion source No. 4) has been successfully constructed at IMP and has also been connected with the Low Energy Beam Transport (LEBT) and Radio Frequency Quadrupole (RFQ) systems. These source magnet coils are cooled through evaporative cooling technology, which is the first attempt with an ECR ion source in the world. The maximum mirror field is 2.5 T (with iron plug) and the effective plasma chamber volume is 1.2 l. It was designed to be operated at 18 GHz and aimed to produce intense multiple charge state heavy ion beams for the linear injector project SSC-Linac at IMP. In February 2014, the first analyzed beam at 18 GHz was extracted. During about three months' commissioning, some outstanding results have been achieved, such as 1.97 emA of O(6+), 1.7 emA of Ar(8+), 1.07 emA of Ar(9+), and 118 euA of Bi(28+). The source has also successfully delivered O(5+) and Ar(8+) ion beams for RFQ commissioning in April 2014. This paper will give a brief overview of the design of LECR4. Then, the latest results of this source at 18 GHz will be presented.

In-water gas combustion in linear and annular gas bubbles

A new pulsed-cyclic method of in-water gas combustion was developed with separate feed of fuel gas and oxygen with the focus on development of new technologies for heat generators and submerged propellers. The results of calorimetric and hydrodynamic measurements are presented. In-water combustion of acetylene, hydrogen, and propane was tested with the operation frequency of 2–2.5 Hz and with a linear injector. The combustion dynamics of combustion of stoichiometric mixture with propane (C3H8+5O2) was studied for a bubble near a solid wall; the produced gas bubble continues expansion and oscillations (for the case of linear and annular bubbles). It was demonstrated that gas combustion in annular bubbles produces two same-magnitude pulses of force acting on the wall. The first pulse is produced due to expansion of combustion products, and the second pulse is produced due to axial cumulative processes after bubble collapse. This process shapes an annular vortex which facilitates high-speed convective processes between combustion products and liquid; and this convection produces small-size bubbles.

Design of Normal-Conducting Quadrupole Magnets for Linac4 at CERN

Linac4 is the first element of the Large Hadron Collider Injectors Upgrade Project and will replace the existing Linac2 as linear injector of protons for the CERN accelerators. A new transfer line will link Linac4 to the Proton Synchrotron Booster. Approximately a hundred normal-conducting electro-magnets are required for beam steering and focusing along the linac and the transfer line. This text concentrates on the design of the Linac4 quadrupole magnets. The design and the first magnetic measurements of the inter-tank quadrupole electro-magnets are discussed. In addition, the design and optimization of the transfer line quadrupole magnets are highlighted. The compatibility of the magnetic requirements and the power converter constraints for the latter magnet type is assessed.

Compressive Self-Powering of Piezo-Floating-Gate Mechanical Impact Detectors

This paper describes a novel compressive self-powering technique that significantly extends the powering and sensing range of our previously reported piezo-floating-gate (PFG) sensors for applications in mechanical impact monitoring. At the core of the proposed technique is a nonlinear impedance circuit that dynamically loads the output of a piezoelectric transducer in a manner such that the sensor can be self-powered at low-levels of mechanical strain and yet is able to sense and detect large variations in strain-levels. The compressive approach requires precise programming of event detection thresholds and requires precise nonvolatile event counting, both of which are achieved using variants of a linear floating-gate injector circuit. Measured results obtained from prototypes fabricated in a 0.5- μm standard CMOS process validate the proposed compressive powering and the proposed programming technique.

An Ultra-Linear Piezo-Floating-Gate Strain-Gauge for Self-Powered Measurement of Quasi-Static-Strain

In this paper we describe a self-powered sensor that can be used for in-vivo measurement of the quasi-static-strain and also for in-vivo measurement of the L1 norm of the strain signal. At the core of the proposed design is a linear floating-gate injector that can achieve more than 13 bits of precision in sensing, signal integration and non-volatile storage. The injectors are self-powered by the piezoelectric transducers that convert mechanical energy from strain-variations into electrical energy. A differential injector topology is used to measure the quasi-static strain by integrating the difference between the L1 norm of the piezoelectric signal generated during the positive and negative strain-cycles. The linear floating-gate injectors are integrated with charge-pumps, digital calibration circuits and digital programming circuits to form a system-on-chip solution that can interface with a standard bio-telemetry platform. We demonstrate the proof-of-concept self-powered measurement of quasi-static strain and L1 norm of the strain signal using sensor prototypes fabricated in a 0.5- μm standard CMOS process and validated using a bench-top biomechanical test setup.

Design of BEPCII bunch current monitor system

BEPCII is an electron-positron collider designed to run under multi-bunches and high beam current condition. The accelerator consists of an electron ring, a positron ring and a linear injector. In order to achieve the target luminosity and implement the equal bunch charge injection, the Bunch Current Monitor (BCM) system is built on BEPCII. The BCM system consists of three parts: the front-end circuit, the bunch current acquisition system and the bucket selection system. The control software of BCM is based on VxWorks and EPICS. With the help of BCM system, the bunch current in each bucket can be monitored in the Central Control Room. The BEPCII timing system can also use the bunch current database to decide which bucket needs to refill to implement ``top-off'' injection.

The general RF tuning for IH-DTL linear accelerators

The RF tuning is the most important research for achieving the resonant frequency and the flatness of electric field distributions along the axis of RF accelerating structures. The six different tuning concepts and that impacts on the longitudinal field distributions have been discussed in detail combining the RF tuning process of a 1:2 modeled 20.85 MV compact IH-DTL cavity, which was designed to accelerate proton, helium, oxygen or C 4+ from 400 keV/ u to 7 MeV/u and used as the linear injector of 430 MeV/ u synchrotron [Y.R. Lu, S. Minaev, U. Ratzinger, B. Schlitt, R.Tiede, The Compact 20MV IH-DTL for the Heidelberg Therapy Facility, in: Proceedings of the LINAC Conference, Luebeck, Germany, 2004 [1]; Y.R. Lu, Frankfurt University Dissertation, 2005. [2]] in Heidelberg Heavy Ion Cancer Therapy (HICAT). Some of tuning concepts are also suitable and effective for the tuning of RFQ and/or other RF accelerating structures. Finally good field flatness in IH-DTL cavity has been realized successfully. The experience got from the model cavity tuning benefits real power cavity tuning, which is only needed to be tuned by the plungers. The cavity had a beam commissioning successfully for the initial beam acceleration at the end of 2006.

Acceleration of Carbon Nuclei up to Relativistic Energy in the U-10 Synchrotron

The reconstruction of a system of ring-shaped accelerators has been completed at the Institute of Theoretical and Experimental Physics: carbon nuclei have been accelerated in the U-10 synchrotron up to maximum energy 4 GeV/amu C4+ ions from a laser ion source are preaccelerated in the I-3 linear injector up to 1.5 MeV/amu and in the UK booster synchrotron up to 50 MeV/amu, injected via the charge-transfer scheme C4+ → C6+ into the U-10 synchrotron on an increasing magnetic field and accelerated to maximum energy. The basic technological features of the acceleration scheme employed and the results of the first physical experiment with relativistic carbon nuclei are described.

Status of the Terawatt Accumulator Accelerator project

The construction of the Terawatt Accumulator (TWAC) facility is nearly completed at the ITEP in Moscow. All the major milestones have been successfully passed with a beam of carbon ions, except for the final result (the high power beam accumulation), which is on the way. The beam of C4+ ions delivered by the laser ion source is accelerated up to the energy of 300 MeV/amu by two steps—in the linear injector I3 and in the booster synchrotron UK. The accelerated beam is extracted from the UK ring and transferred to the U10 accumulator ring. Non-Liouvillian stripping technique (C4+ ⇒ C6+) is applied for stacking of C6+ batches into the accumulator ring U10. First experiments with extracted beam of ions have started in 2002. Status of the TWAC components, current results of activities aiming at mastering the ion beam stacking technique, and outlook for the TWAC advance are presented.

Enhanced ion acceleration by focusing an intense electron beam inside a dielectric collimator

Enhancements in the acceleration of positive ions were observed by converging an intense electron beam (250 keV, 6 kA, 80 ns, cm from the Kilo Ampere Linear Injector (KALI-200)), into a short dielectric collimator having a radius of about the beam radius and a length around the potential-well characteristic length . These were in comparison to the results of a beam injected from a planar diode into an collimator where the ion current, , recorded was only to 5 A of 150-250 ns pulse width which included - particles of , , of peak energy of , 10.8 and 14.4 MeV respectively. The ion energy increased with the electron beam propagation distance and resulted in the elongation of the ion pulse in total agreement with the theory. The magnitude of the ion current followed the trend of the propagated electron current which decreased with distance. When the electron beam was focused by a convex anode having , the ion current increased to 9-16 A and to MeV as a result of the field enhanced potential well. was further enhanced to 20-30 A and the energy to 1.5 MeV by increasing the collimator thickness to 4-8 mm. Peak energies of , and generated were of the order of 18, 24 and 28.5 MeV. In terms of the energy invested, these results are an improvement over other work.

Optimization of SiO 2 film conformality in TEOS/O 3 APCVD

Conformality of TEOS/O 3 SiO 2 thin films has been found to be deposition region dependent when the films were deposited using a standard atmospheric pressure chemical vapor deposition (APCVD) linear injector system at ∼ 400 °C. In previous work, flow-like step coverage was only observed away from the central gas injection area where ozone was delivered from the outer port. In this work, we employed a new injector port configuration where ozone was delivered from the central port and TEOS was delivered from the outer port. In this case, flow-like step coverage was observed directly under the gas injection, yielding a gap filling capability of a 0.2 μm space between aluminum lines with a 3:1 aspect ratio and a 0.35 μm space between PolySi-coated thermal oxide structures with a 2:1 aspect ratio. All these results suggest strongly that the actual concentration distribution of gas species at the wafer surface including O 3, TEOS, and reaction intermediates generated upon gas mixing in the chamber can vary substantially with position depending upon the reactor geometry and flow configuration, leading to diverse surface characteristics such as conformality.

The high peak current polarized electron source of the Stanford Linear Collider

The Stanford Linear Collider injector requires two 2 ns pulses of 4.5–5.5 × 10 10 electrons, separated by 61 ns at 120 Hz, from its source. Since 1992, these currents have been provided by a polarized electron source based on GaAs photocathodes. A beam polarization of 76 ± 4% has been measured at the end of the 50 GeV linac. At low photocathode quantum efficiencies, and for excitation near threshold, the maximum current delivered by the source is constrained, not by the space charge limit of the gun, but by a “charge limit” of the photocathode. The charge limited current is proportional to the photocathode quantum efficiency, but the proportionality varies for different photocathode types. Experience with high polarization strained GaAs photocathodes on a test beamline and on the SLC is presented.

Beam-foil spectroscopy of nickel and copper at 1.16 MeV/amu

The beam-foil spectra of nickel and copper obtained at Orsay on the linear injector of the heavy ion accelerator with beams of 1.16 MeV/amu have been analysed in the 110–380 Å range. Numerous lines attributed to Ni XVI to XX and Cu XVII to XXI transitions have been identified. The lifetimes of the upper levels of the strongest copper transitions have been measured and are compared with theory and with previous experiments.