RMATE: A device to test radiation-induced effects under controlled magnetic field and temperature
I. Garcia-Cortes, S. Cabrera, M. Medrano, A. Moroño, P. Muñoz, A. Soleto, I. Bugallo, A. Nieto, R. Altimira, A. Bollero, J. Camarero, J. L. F. Cuñado
RRMATE: A device to test radiation-induced effectsunder controlled magnetic field and temperature
I. Garc´ıa-Cort´es a , S. Cabrera a , M. Medrano a , A. Moro˜no a , P. Mu˜noz a ,A. Soleto a , I. Bugallo b , A. Nieto c , R. Altimira c , A. Bollero d , J. Camarero e,d ,J. L. F. Cu˜nado d a Laboratorio Nacional de Fusi´on, CIEMAT. Avda. Complutense 40, 28040 Madrid, Spain b Servicios de Apoyo a la Investigaci´on, SEGAINVEX-UAM,Universidad Aut´onoma deMadrid, 28049 Madrid, Spain c Ingenier´ıa Magn´etica Aplicada, Avda. Catalunya, 5, 08291 Ripollet Barcelona, Spain d Instituto Madrile˜no de Estudios Avanzados en Nanociencia IMDEA-Nanociencia, CampusUniversidad Aut´onoma de Madrid, 28049 Madrid, Spain e Departamento de F´ısica de la Materia Condensada and Instituto ”Nicol´as Cabrera”,Universidad Aut´onoma de Madrid, 28049 Madrid, Spain
Abstract
This study shows the development and performance assessment of a novel set-upthat enables the research of structural materials for fusion reactors, by makingpossible simultaneous application of temperature (up to 450 ◦ C) and magneticfield (close to 0.6 T) during irradiation experiments. These aspects become crit-ical as structural materials in fusion reactors are exposed to intense radiationlevels under the presence of strong magnetic fields. Moreover, material micro-structural could be modified by radiation-induce propagating defects, which arethought to be sensitive to magnetic fields. The device has three main compo-nents: magnetic closure, sample holder with integrated heater, and radiationshield. It is provided with a thermal shield to prevent other elements of thedevice to heat up and fail. A mapping of the magnetic flux in the region wheresample and heater are located has been modeled by finite elements simulationsoftware and correlated with magnetic measurements.
Keywords:
PACS numbers:, 28.52.s, 28.52.Av, 39.10.+j, 75.50.Bb, 07.55.w
Email address: [email protected] (J. L. F. Cu˜nado)
Preprint submitted to Fusion Engineering and Design May 30, 2019 a r X i v : . [ phy s i c s . i n s - d e t ] M a y . Introduction Fusion reactors based on magnetic confinement need special materials forstructural and vacuum vessel, capable of resisting high level of radiation inpresence of the strong magnetic fields required for plasma confinement [1, 2].In consequence, the selection and design of the materials that can face the ex-treme conditions of fusion power plant have been described as one of the greatmaterial science challenges [3, 4]. High-chromium ferritic/martensitic steels aregood candidates as structural materials for fusion reactors because of their highresistance to irradiation [5, 6]. However, it is well-known that micro-structuraland mechanical properties of these materials are modified by radiation-inducepropagating defects [7, 8]. In addition, since they are based on Fe-Cr alloys,which are ferromagnetic materials, the presence of high magnetic fields couldalso play a role in the performance of these steels. Concerning this subject,several theoretical studies point to magnetism as being a critical factor in radi-ation induced damage in structural materials [9, 10]. These predictions suggestthat magnetism can be a non-negligible factor in defining the defect propertiesinduced by ion irradiation or in determining the atomic distribution. However,a few experimental and theoretical efforts have been done in order to elucidatethe role played by magnetism in damage. It has been found that Fe + irradia-tion on ferromagnetic films induced changes in the magnetic state for variousirradiation doses [11, 12].In this context, recent irradiation experiments in Fe-Cr alloys were carriedout to study the influence of external magnetic fields in damage under fieldstrengths of around 0.4 T at room temperature [7, 13]. The results of theseworks point to the external magnetic field as a non-negligible parameter invacancy profile and chromium mobility due to the defect creation. Howeverpermanent magnets used in these works have magnetic field values well belowof those expected in future reactors (several Tesla). On another hand, theapproach used thus far to set the external field on the sample was to fit itin contact with the permanent magnet, where the magnetic field is maximum,2ut this means a drawback in order to heat the sample, as this would alterthe characteristics, or even damage the permanent magnet by going above itsmaximum operation temperature. Accordingly the experimental set up of theseworks does not allow the realization of experiments at high temperature (above200 ◦ C) with application of a maximum magnetic field. A direct consequence ofthis limitation is the scarcity of results on the behavior of structural materialsunder relevant conditions for fusion reactors applications, i.e., high temperatureand high magnetic fields.In order to overcome these limitations, a new device, named RMATE (for
Radiation-induced effects under controlled MAgnetic field and TEmperature ),has been designed and construct from scratch. It consists of a closed chamberin whose interior the sample can be located, exposed to the ion beam throughan axial entrance, at controlled conditions of temperature and magnetic field,solving the aforementioned problems. The chamber itself constitutes a mag-netic closure that drives the magnetic flux and concentrates it at the sampleposition. Flux lines are kept parallel to the ion beam, so that deflection due toLorentz force is avoided. RMATE has been designed to be mounted inside oneof the vacuum chambers of the Danfysik implanter line, at the research centerCIEMAT (Madrid, Spain) [14, 15]. For this reason, its shape has constrains im-posed by the implanter chamber in which is located, and under this restrictions,the shape has been optimized to get the maximum field values at the sampleposition. This manuscript is organized as follows: a description of the deviceis introduced first, followed by simulation of the magnetic flux in the magneticclosure based on optimized shape; each element of the device, as finally builtin our workshops, is described accompanied by magnetic characterization. Thestudy concludes with a discussion of the results as well as discussion of futureimprovement perspectives. 3 . Design of the irradiation device prototype
The RMATE device is designed to keep the sample at temperature of upto 450 ◦ C under magnetic fields of 0.3-0.6 T (with a single permanent magnetand depending of the ferromagnetic nature of the sample under study), allowingat the same time to expose the sample to ion beam irradiation. The magneticfield can be created either by a permanent magnet or a coil inside the magneticclosure (in this paper we focus in the simplest case of a permanent magnet),while the temperature is controlled by using an oven or a filament heater. Threemain components compose the device: the magnetic closure, the sample holderwith an integrated heater and a tandem radiation shield. The sample is setinside the magnetic closure, at its axis, facing an entrance through which it isexposed to the ion beam. The magnetic closure has been conceived as a closedcavity generated by a revolution profile, as show in the 3D scheme of Figure 1,where most of the inner parts have been removed for clarity. The closure splitsinto two parts, (which are joint with screws at their equator’s slabs): a base part ending in a flat face (left part of Figure 1), and the so-called dome , due toits particular shape, (right part of Figure 1). The beam entrance is an inwardperforated cone located in the dome, at its axis, as indicated in the figure.A NdFeB permanent magnet (N35 type, 30 mm diameter and 10 mm thick)is axially located in the base, sitting on its inner side. The particular closed ‘pot-like’ shape of the magnetic closure, made of ferromagnetic material, forcesmagnetic lines to concentrate in the axis of the chamber, closing magnetic fluxthrough its perimeter, as indicated by the dashed red closed line depicted inthe upper cut (visual schematic illustration). The final profile of the revolutionfigure is the result of an optimization process carried out by simulation, whichwill be detailed in the following sections.
In order to create the profile shape of the magnetic closure, we first startedwith an approximate initial shape and carried out sequential simulations until4 igure 1: 3D view, with quarter cutoff, of the magnetic closure. The two parts of therevolution-shaped chamber can be identified (indicated as BASE and DOME). The dome isprovided with the beam access (dashed blue arrow). Magnetic flux lines follow a closed circuit,as depicted by the dashed red line (visual guide), running through the axis and the perimeter.The magnet can be seen here sitting on the base, and the sample is represented as a diskin between the magnet and the entrance cone. Two oblique optical accesses have been setto permit in-situ vectorial magneto-optical Kerr effect magnetometry (v-MOKE), using laserlight, represented here by a red line reflection on the sample surface. we got to the final (optimized) shape. The initial shape consisted of a roughrevolution figure with the magnet in the inner side of its base, and the beamentrance at its opposite side (early-stage pot-shaped closure). This initial shapewas designed to fit inside the Danfysik ion implanter chamber. Finite elementssimulation software FEMM [16] and MAXWELL [17] where used to create thefinal optimized magnetic closure profile, as can be seen in Figure 2. The op-timization pursues to get the maximum magnetic field in the sample position,located at the axis inside of the magnetic closure, while keeping flux lines paral-lel to it, so that the ion beam is not deflected by Lorentz force. This ensures the5eam reaching the sample without any loss of intensity and/or energy. Figure 2shows the magnetic field intensity simulation as a color pattern superimposedto the profile of the closure cross section. The material used in the simulation isAISI 420 steel. The magnetic field intensity color pattern goes from red for thehighest field values to blue for the lowest (dark blue being zero, according tocolor code displayed in the legend of the figure). The magnet can be seen sittingon the inner base side, on the left, while the perforated entrance cone can be seenin the dome at the right. Different tests where made at Ingenier´ıa Magn´eticaAplicada (IMA) [18] for several shapes, in order to reach an optimized one. Thesample position is schematically indicated in Figure 2 as a small disk with yellowand black blades, located between the magnet and the beam entrance. In thispoint, simulations show a field value around 0.4 T. In Figure 3 the magneticflux lines can be seen, together with a plot of the magnetic field as a functionof the distance from the magnet, in direction of the entrance cone.
3. Description of the device elements
For this first prototype of the magnetic closure we used AISI 420 RE steel,easy to acquire and mechanize. It consists in an axis-symmetric hollow potdivided at its equator into the already described base and dome parts, fixedtogether by screws located in slabs at the joint. 3D schematics of the designcan be seen in Figure 1, where the revolution shape has been cutoff in a quarterview. The two parts are clearly identified:1. Base: this is the left part of the closure shown in Figure 1, where thewhole device is supported. The axial magnet is located in the inner sideof the base, represented with a cylinder in the figure. Since the base is incontact with the cooler plate, it acts as thermal sink for the rest of theclosure. More details are explained later with the help of Figures 4 and 5.2. Dome: this is the part on the right side of 1. It has the ion beamentrance, that can be seen as an inward perforated cone. Two additional6 igure 2: Simulation of the prototype magnetic closure, using a single NdFeB magnet. Colorpattern is used to identify the magnetic field intensity (in Tesla), from highest (red) to lowestintensities (dark blue). These simulations have been done on the basis of AISI 420 steel. Thesample is located at the position indicated by the yellow and black bladed disk. Magnetic fluxat such position is about to 0.4 T. oblique entrances are located symmetrically besides the beam entrance,at a certain angle. These are conceived for future optical access suitablefor in-situ vectorial resolved magneto-optical Kerr effect magnetometry(v-MOKE) [19, 20], so that a laser beam can go through one of theseoblique accesses to reach easily the sample, and go out after reflecting onit through the other (symmetrically located) oblique access. These obliqueaccesses can be also seen in Figure 4. Between the magnet and the entrancecone, there is room enough to allocate the radiation shield and the sample7older with the heater, in such a way that the sample is decoupled fromthe magnet and the radiation shield inner walls. This decoupling of thesample holder and heater is carried out by using a support rod directlyfixed to the support flange (explained later, see also Figure 5).Schematic illustration in Figure 1 has been simplified to allow for a betterview of the magnetic closure profile, by showing only the magnet inside and thesample disk (a complete cross section can be seen in Figure 6). The magnet,radiation shield and heater-sample holder set-up are collinear with the closureaxis. A lateral window (cut by the sectioning line) can be seen at the lower partof the figure. This window is used to allow the sample support rod going outto the flange, where it is fixed, and also allows the wires and sensors to get in.Figures 4 and 5 provide more details.
Since the research of materials for fusion reactor applications requires settingthe samples at controlled high temperatures, a heater is necessary. RMATE isprovided with a sample holder that has an integrated heater on it. In Figure 5the sample holder can be observed hanging from the flange by a rod fixed to itwith a non-conducting ceramic waffle, not seen in this figure (refer to Figure 7a).In this way, we prevent any contact with the rest of the device elements, thusavoiding heat conduction through the structure. The sample is mounted in theinner side of the sample holder, so that it is as close as possible to the magnet.The heat generated in the sample holder is dissipated by the thermal shield.An additional braid (not shown) connects the heater-sample holder rod to thecooler plate. This is important because the rod is fixed to the flange through aceramic piece and hence specific cooling must be set on it.
Radiation shield becomes necessary in order to prevent overheating of themagnet (NdFeB-based magnets have a maximum working temperature of around200 ◦ C) or magnetic closure. A tandem radiation shield has been designed so8hat first, the magnet is thermally separated from the heater-sample holder, andsecond, the heater does not affect the closure, preventing the magnet from beingindirectly heated up through it. The tandem radiation shield can be identifiedin yellow color in both Figures 4 and 5. The left part of the tandem correspondsto the radiation shield of the magnet itself, indicated as
Magnet section in Fig-ure 5, and the right part corresponds to the heater-sample holder, indicated as heater-sample holder section in the same figure.In order to keep the radiation shield at low temperatures, six copper screws(M6) connects it thermally to a cooler plate, traversing the magnetic closurebase. This design allows that both the radiation shield and the base are inthermal contact with the cooler plate. In Figure 4 the magnetic closure appearswith a 1/4 cutoff, allowing to see the cooler plate and the radiation shield. InFigure 5 the magnetic closure has been removed so that the connection betweenthe tandem radiation shield and the cooler plate can be observed (screws appearalmost hidden). In Figure 6, the whole device cross section is depicted, showinghow the cooler plate, the base and the radiation shield are fixed by the sixaforementioned screws, ensuring a total thermal contact between them, whilethe heater-sample holder remains isolated.The cold finger supports the cooler plate, and it also acts as structuralelement to support the whole device to the Danfysik chamber flange (Figure 7).
4. Assembly and characterization of the prototype
The final prototype of the magnetic closure device was built at UAM-SEGAINVEXand CIEMAT workshops. The magnetic closure was made of AISI 420 RE steel,with a base diameter of 80 mm and a total length of 62 mm. The cooler platehas the same diameter as the base, and it is made of 7 mm thick copper. It isprovided with an additional copper thin foil in between, in order to ensure anoptimal thermal contact. The cooler plate is fixed to the cold finger, made ofcopper and also in good thermal contact.In Figure 7 three pictures of the set-up are shown. The set-up is upside9own, with the flange here acting as a support platform. The cold finger andcold plate can be seen supporting the device from its base through the coolerplate (better seen in Figure 7(b)). In Figure 7(a), the dome and the cap of theradiation shield have been removed, allowing to see the heater-sample holderset-up, and the rod that supports it to the flange. This rod ends in a mecha-nizable Makor ceramics to isolate electric contacts. Notice the lateral windowof the radiation shield through which the heater-sample holder rod and anyrequired electrical contacts can traverse the closure reaching the ceramic sup-port. In Figure 7(b), the cap of the radiation shield has already been put on(the screws that fix the cup are independent of the copper screws that fix theradiation shield, the base and the cooler plate). The beam entrance through thecap of the radiation shield is visible here. Notice that the heater-sample holderis not in contact with any part of the radiation shield. The cap of the radiationshield has a hole to allow the ion beam to reach the sample. In Figure 7(c),the dome of the magnetic closure has been set back, allowing to see the set-upas it is mounted inside the final vacuum chamber of the CIEMAT Danfysik ionimplanter.Magnetic field characterization of the RMATE device was carried by testingfield values at three points inside the device, always in the axis, indicated withnumbers 1 through 3 in Figures 3 and 6. At position 1, where the sample islocated, a value of 298 mT was measured. At position 2, the field drops to162 mT, and at point 3 value measured is 93 mT. As can be seen in Figure 3,top graph, the values are in agreement with those expected by the simulations.Notice that because of size and shape restrictions, the shape of the dome makesflux lines to scape towards the dome walls, rather than towards the entrancecone. This is due to proximity and reduces concentration in the axis.In order to overcome this issue and reach higher field values at sample posi-tion, there are two possibilities (that could also be combined): on the one hand,a stacked-magnet configuration could be used to concentrate the flux lines closerto the axis; on the other hand, a ferromagnetic sample holder with particular10esign can be used to attract magnetic flux to the sample position. Stacked-Magnet configuration has been simulated, and can be seen in Figure 8. In thisconfiguration, the field value at the expected sample position (red dashed crossin the figure) will be around 0.55 T. This configuration can be mounted by sim-ply reconfiguring the second section of the radiation shield, provided that theheater is slightly reduced in length. Simulations where carried out also with asample holder consisting on a ferromagnetic AISI 420 disk (not shown). Thefield thus obtained could reach values as high as 0.8 to 1 T. Notice that, asthe samples under research for fusion applications itself is ferromagnetic, thefinal field that it sees inside the magnetic closer will be higher than the mea-sured one for our prototype. Also it must be taken into account that the steelused in the prototype is not the ideal one, and further simulations with bettersteels are now underway, together with more tests to improve the closure pro-file. We expect to get field values close to 1 T in near future. In any case, thepresented prototype already allows for reasonable magnetic field values at sam-ple position, under controlled temperatures and with the possibility of ins-situv-MOKE magnetometry.
5. Conclusions
This is the first time that on purpose magnetic closure/sample holder de-vice is built for the investigation of temperature and field-dependent irradiationdamage of structural materials for fusion reactors. It has been developed to befitted in the ion implanter (Danfysik) irradiation facility currently in operationat CIEMAT. This compact irradiation device is very easy to adapt to biggerimplanter. The magnetic closure profile has been optimized to concentrate mag-netic flux in the sample position, reaching values of the order of 0.3 T or higherwhen ferromagnetic sample is mounted. It can also keep the sample undercontrolled temperatures of up to 450 ◦ C, and, remarkably, its design permitsto carry out in-situ vectorial resolve magneto-optical Kerr effect magnetometry.The magnetic closure can be further optimized by improving constituent materi-11ls, or even more, modifying the shape in case space restrictions are not present(as is the case in other ion implanter facilities), and opens new perspectives inthe research of materials for fusion reactor applications.
6. Acknowledges
This work was mainly financed by the following projects: Spanish MINECO(Ministerio de Econom´ıa y Competitividad) under project ENE2016-76755-R(AEI/FEDER, UE), Comunidad de Madrid through projects NANOMAGCOST-CM, project n. P2018/NMT4321, and NANOFRONTMAG, project n. S2013/MIT-2850. P. M. acknowledges a pre-PhD contract of the Spanish MINECO. IMDEAnanociencia acknowledges support from the Severo Ochoa Program (MINECO,Grant SEV-2016-0686). We also acknowledge support from UAM-SEGAINVEX,and in particular, we acknowledge to J. R. Marijuan and J.M. Corts.
7. ReferencesReferences [1] O. Motojima, Nucl. Fusion. , 104023 (2015).[2] D. Stork, P. Agostini, et al., J. Nucl. Mater. , 277291 (2014).[3] D. Stork, S. J. Zinkle, Nucl. Fusion , 092001 (2017).[4] S. J. Zinkle, Phys. Plasmas , 058101 (2005).[5] E. A. Little, D. A. Stow, J. Nucl. Mater. , 11-24, 25-39 (1979).[6] F. A. Garner, M. B. Toloczko, J. Nucl. Mater. , 123-142 (2000).[7] F. J. S´anchez, I. Garc´ıa-Cort´es, et al., Nucl. Mater. Energy. , 476479(2016).[8] B. G´omez-Ferrer, I. Garc´ıa-Cort´es, et al., Phys. Rev. B , 220102(R)(2014). 129] T. Seletskaia, Y. Osetsky et al., Phys. Rev. Lett. , 046403 (2005).[10] L. Malerba, A. Caro et al., J. Nucl. Mater. , 112125 (2008).[11] K. Papamihail, K. Mergia, F. Ott, Yves Serruys, Th. Speliotis, G. Apos-tolopoulos, and S. Messoloras Nuc. Mater. end Energy , 459-464 (2016).[12] K. Papamihail, K. Mergia, F. Ott, Yves Serruys, Th. Speliotis, G. Apos-tolopoulos, and S. Messoloras Phys. Rev. B , 100404-1-5 (2016).[13] I. Garc´ıa-Cort´es, T. Leguey et al., J. Nucl. Mater. [17] MAXWELL by ANSYS , 046109 (2015)[20] Ibit . Rev. Sci. Instrum. , 053904 (2014)13 igure 3: Magnetic flux simulation for the prototype can be seen in color line scheme. Amagnetic field profile taken at the axis is displayed on top. This profile corresponds to theregion between the magnet surface (set as 0 mm in the horizontal axis) and the entrancecone (30 mm away). Three points where measured experimentally, at positions marked inred circles (see the text and Figure 6). The corresponding experimental data can be seensuperimposed on the field profile, with red symbols (yellow filled). Reference dashed red lineshave been depicted for clarity. igure 4: 3D scheme of the magnetic closure as mounted on the flange for the Danfysikirradiation chamber. The closure appears with a quarter cutoff that allows to see the tandemradiation shield. The cooler plate can be seen behind the closure in orange color, attached tothe cold finger, which is the vertical orange cylinder. The cold finger acts both as supportingstructure and feedthrough. The aperture in the radiation shield can be seen, as well as thelaser, used for in-situ v-MOKE magnetometry (indicated by the red line) and the lateraloptical accesses. A rod emerges from inside the radiation shield through a window, traversingalso the magnetic closure through another window, ending at the flange, where is fixed. Thisrod supports the heater-sample holder (not seen here, refer to Figure 5) inside the radiationshield, without touching any part of it. igure 5: Details of the heater-sample holder, its supporting rod, radiation shield, coolerplate and cold finger. The magnetic closure and the radiation shield cap have been remove inthis 3D scheme, so that the sample holder and heater can be seen. Thermal contact with thecooler plate can be seen in gray color (although they are copper screws). These have doublefunctionality: they keep magnetic closure and radiation shield fixed to the cooler plate, andallow heat sink from radiation shield and closure into the cooler plate and cold finger. igure 6: Schematic diagram of the cross-section of the magnetic closure with all the innerelements on its actual position. Notice that the sample position is as close as possible tothe point where the maximum of magnetic field is obtained. Points 1,2 and 3 indicate thepositions where the magnetic field has been experimentally measured in the prototype. igure 7: Pictures of magnetic closure (explanations in the text). igure 8: Simulation of the magnetic closure with stacked magnet can be seen here. At theindicated sample position, fields of about 0.6 T can be reached. When a ferromagnetic sampleis located in that position, re-conducted magnetic flux lines allows reaching higher fields, closeto 1 T.igure 8: Simulation of the magnetic closure with stacked magnet can be seen here. At theindicated sample position, fields of about 0.6 T can be reached. When a ferromagnetic sampleis located in that position, re-conducted magnetic flux lines allows reaching higher fields, closeto 1 T.