Beam Induced Damage Studies of the IFMIF/EVEDA 125 mA CW 9 MeV D⁺ Linear Accelerator

IFMIF (International Fusion Material Irradiation Facility) will be a Li(d,xn) neutron source providing equivalent neutron spectrum of DT fusion reactions and comparable neutron flux of future commercial reactors. Such a facility is an essential step in world fusion roadmaps to qualify suitable structural materials capable of holding the unrivalled neutron irradiation inside the nuclear vessel of a fusion reactor. IFMIF, presently in its EVEDA (Engineering Validation and Engineering Design Activities) phase, is installing LIPAc (Linear IFMIF Prototype Accelerator) in Rokkasho (Japan), a 125mA CW 9 MeV deuteron beam as validating prototype of IFMIF accelerators. The MPS of LIPAc manages the interlocks for a fast beam stop during anomalous beam losses or other hazardous situations. High speed processing is essential to achieve MPS goals driven by investment protection principles. Beam losses may lead to severe damages by excessive thermal stresses, annealing or even burn/melting of materials. The assumptions to estimate the practical safe times for a fast beam shutdown during the accelerator operational life are here described. IFMIF AND LIPAC, ITS ACCELERATOR PROTOTYPE Fusion materials research has fuelled for decades the world endeavours towards high current linacs [1]. The required neutron flux >10 m·s with a broad peak at 14 MeV to simulate the irradiation conditions of the plasma facing components in a fusion reactor is obtainable through Li(d,xn) stripping reactions; however those fluxes demand deuteron currents in the 10 mA range in CW mode. The first world attempt of such conditions was framed by the Fusion Materials Irradiation Test Facility, FMIT, in the early 80s; with unexpected difficulties and lessons learnt in operating in CW mode [2]. The International Fusion Materials Irradiation Facility, IFMIF, consists of two deuteron accelerators at 125 mA in CW and 40 MeV impacting on a flowing lithium screen. It is since 2007, in its Engineering Validation and Engineering Design Activity phase, EVEDA, where the only remaining activity of its broad mandate (that has provided an engineering design [3] of the plant and, among many other technical challenges, validated the stable operation of its lithium loop [4] and its irradiation modules capable of housing above 1000 specimens and characterize structural materials simultaneously in twelve different irradiation capsules independently cooled [5]) is its Linear IFMIF Prototype Accelerator, LIPAc, presently under installation and commissioning in the International Fusion Energy Research Center (IFERC) in Rokkasho (Japan), by European and Japanese laboratories [6]. A full account of the validation activities under IFMIF/EVEDA has already been provided [7]. Collective phenomena driven by space charge forces become the main limitation on achieving high intensity beams. In low β regions, the beam outward radial Coulomb forces prevail over the inward radial Ampere ones, but they mutually cancel in the relativistic domain. Thus, space charge repulsive forces are stronger the lower the beam energy is. The successful operation of LIPAc, with its deuteron beam current of 125 mA in CW at 9 MeV as the output of the first planned cryomodule of IFMIF will validate the 40 MeV required for the Li(d,xn) source [1,8]. Figure 1: Above Comparison of IFMIF accelerators and LIPAc, their 1.125 MW beam average power prototype accelerator, matched up to the 1st SRF linac at 9 MeV. Below breakdown of the contribution for LIPAc. IONS INTERACTION WITH MATTER The physics of heavy ions with matter was first unravelled semi-classically by Bohr in 1913 based on his atomistic model (making use of the impact parameter between target nuclei and impacting particle) [9], and through relativistic quantum mechanics by Bethe in 1932 [10] (making use of the momentum transfer by the particle to the cloud of electrons) being both expressions for the stopping power − of the absorber identical for nonrelativistic ions ( ≪ 1), and with the ion only dependent variables its kinetic energy and charge. Among the different possible types of radiation, only ions show a fixed range; a mono-energetic beam of ions traversing matter loses its energy without any change in the number of particles, and eventually all are stopped reaching practically the same depth. The combination of the logarithmical dependence of the stopping power on the ions speed and their slowing Proceedings of IPAC2016, Busan, Korea WEPMR044 07 Accelerator Technology T33 Subsystems, Technology and Components, Other ISBN 978-3-95450-147-2 2373 C op yr ig ht © 20 16 C C -B Y3. 0 an d by th e re sp ec tiv e au th or s down by ionization and excitation of the atoms of the medium (density effects being marginal for light ions) as it penetrates matter, leads to a maximum of the rate of energy loss in proximity to its range. This maximum is called the Bragg peak, however its frequent use as either range in the material or the point of release of ions energy is misleading. The range R(EK) of a charged ion at a given kinetic energy, EK, in a given material is the distance it penetrates before coming to rest and it is obtained integrating the inverse of the stopping power. It can be easily shown that for two given ions moving at same velocity the ratio of their ranges follows = (1) where Mi is the ion mass and zi its charge [11]. This equation allows the straightforward estimation for any ion from the well-known range of protons. The Stopping and Range of Ions in Matter (SRIM) code made available by J.F. Ziegler (http://www.srim.org/) has become the worldwide friendly tool for the calculation of the stopping power and range of ions while flying through matter. BEAM INDUCED DAMAGES High current hadron linacs have traversed in recent years the 1 MW beam average power frontier in SNS; next decade beam average powers of 5 MW will be reached in IFMIF-DONES, ESS and MYRRHA [12] and even possibly 15 MW with CADS thanks to its planned 1.5 GeV proton beam at 10 mA CW current. In turn, within few years a beam circulating in HL-LHC will store energies close to 700 MJ, like a Japanese shinkanzen travelling at full speed along the 27 km ring of the LHC. With typically beam radiuses in the order of mm, the power densities handled in case of an accidental missteering at these beam powers may cause catastrophic damages in shorter times than any possible fast beam shutdown signal can be executed. Fortunately, this scenario is improbable thanks to the magnets inductive time constants, which are longer than beam fast shutdown times, typically within ~10 μs; however, during commissioning phases and beam injection stages, operational thresholds are to be set before a stable beam is in place. Beam losses are to be carefully controlled; obviously the higher the current, the more severe potentially becomes their impact. The hands-on maintenance criterion for a proton beam of <1 W/m, or/and <10 total beam loss remains valid if currents are ~100 mA CW, even with deuterons like in LIPAc despite their substantial stronger activating capability. Though hands-on maintenance should not be impacted by an occasional possible beam mis-handling, in high current accelerators beam halo becomes the main source of beam losses and measures for its minimization and monitoring are to be devised. The diffusivity, = , is the only parameter of the second-order partial differential heat equation that describes the variation of temperature in a given region over time. It is as important for a transient heat conduction scenario as conductivity is for the steady state described by Fourier law. The diffusivity measures how fast a material can carry heat away from a heat source. In processes as fast as a beam mis-steered, what matters is not only the energy content, but the time involved in its diffusion. Diffusivity presents units of LT and thus, regardless the accurate difficult solution of the heat equation [13], a characteristic thermal diffusion time of the absorbed heat can be easily estimated. Physics related with a high energy, high current beams accidentally interacting with matter are complex, involving several disciplines. In addition, there is limited practical experience and poor understanding of the behaviour of material exposed to such extreme conditions, where phase transitions involving melting, vaporization and even plasma generation can occur. At the same time, even in absence of changes in material phase, if times constant of heat diffusion are substantially larger than beam heat deposition times, impacted regions can be exposed to high local strains with propagation of severe thermo-elastic stress waves [14] leading potentially to sudden vacuum leaks or damage of structurally weak elements. Prevention of beam induced accidents drive the design of the Machine Protection System (MPS); however their scarce occurrence makes them be one of the less explored events in accelerators technology. The gap in specialized literature is notorious, with divergence in the analysis approach and occasionally with an unnecessary sophistication. An accurate calculation of beam induced damages, following accelerator structures accidental exposures to the extreme conditions of the planned high intensity hadron facilities is not indispensable given the inherent uncertainties linked both to the exact conditions taking place and to the irradiated materials response. Thus, conservative worst case scenarios defining operational boundaries are suitable as a design operational basis for commissioning and beam injection stages, when MPS is occasionally by-passed and fatal errors can occur. LIPAC CASE BRIEF STUDY LIPAc, with its 9 MeV deuteron beam at 125 mA in CW, will traverse the 1 MW beam average power frontier in 2019. Its 1.125 MW beam average power is equivalent to a full shuttle moving through Japanese national roads. Its beam will fly through the full length of the accelerator in 2.7 μs time. Four differe