Atomistic Simulations of Divertor-Plasma Interactions in Fusion Reactors

Show full item record



Permalink

http://urn.fi/URN:ISBN:978-952-10-8957-2
Title: Atomistic Simulations of Divertor-Plasma Interactions in Fusion Reactors
Author: Lasa Esquisabel, Ane
Contributor: University of Helsinki, Faculty of Science, Department of Physics
Publisher: Helsingin yliopisto
Date: 2014-04-26
URI: http://urn.fi/URN:ISBN:978-952-10-8957-2
http://hdl.handle.net/10138/44713
Thesis level: Doctoral dissertation (article-based)
Abstract: The world's energy demand and harmful green-house gas emissions are continuously increasing, while the fossil fuel reservoir may soon end. Currently, there is no clear alternative to the traditional energy production methods for a safe and clean future. Fusion could be part of the solution offering a green-house gas free, virtually endless, safe and large scale energy production. A major challenge for fusion is, however, to produce more energy than needed to achieve and maintain the fusion reaction. The most feasible fusion reaction is based on two hydrogen isotopes: deuterium and tritium, which fuse to produce a helium atom and a neutron. For these atoms to fuse, they must overcome the repulsive interaction between them, requiring extreme temperatures. Thus, the particles ionize forming gas plasma. On Earth, this condition can only be met by isolating the plasma from its environment, for instance, by using closed magnetic fields to form a torus-like shaped plasma, also known as tokamak. However, the plasma particles will interact with the reactor walls as their confinement is never perfect, the exhausted plasma must leave the reactor and impurities are introduced in the plasma boundary to control its characteristics. The plasma-wall interactions are especially intense at the divertor, where the plasma is designed to touch the wall. Understanding these processes is essential to develop safe, long-lasting materials and to avoid contaminating the plasma fuel. The main candidates as first wall materials in future fusion reactors are beryllium for the main wall, and tungsten and carbon for the divertor. Also, the materials may mix due to wall erosion, transport of the eroded particles and their deposition in a new location. Plasma-wall interactions can be studied in current experimental reactors or in linear plasma devices. However, this work is often insufficient to understand the underlying mechanisms. Further, the effects of plasma-wall interactions in materials develop in a wide range of time and length scales. Multi scale modelling is a tool that allows to overcome these challenges, improving the predictions for future fusion reactors. In this thesis, the plasma wall interactions taking place in a fusion reactors divertor have been studied by computational means. The interaction of pure and mixed divertor materials, with plasma and impurity particles were modelled. The work was mainly based on atomistic scale calculations, and a Kinetic Monte Carlo algorithm has also been developed to extend the results to macroscopic scales, enabling a direct comparison with experiments. First, deuterium irradiation of various W-C composites has been modelled, focusing on deuterium implantation, variations of the substrate composition and C erosion mechanisms. Carbon was preferentially eroded, varying the substrate's composition throughout the irradiation. The presence of carbon also affected the D implantation characteristics. As carbon became less likely to be an ITER first wall material, the present work focused on the tungsten-beryllium-deuterium system. The tungsten-beryllium mixing showed a strong dependence on irradiation energy and angle. Further, the presence of Be led to higher fuel implantation and W erosion was suppressed by mixed layer formation. The obtained yields were compared to Binary Collision Approximation results, in order to improve the description of the latter method. Furthermore, an unexpected and possibly harmful phenomenon has been addressed in this thesis: porous nano-morphology formation in tungsten by helium plasma exposure. First, the main characteristics and active mechanisms in the system were identified by atomistic simulations. Then, the porous morphology growth was modelled by implementing these processes in a Kinetic Monte Carlo code, resulting in rates that agreed with experimental findings. A morphology growth model was derived where the time dependence is driven by the evolution of the surface roughness, which is a stochastic process and thus evolves as the square root of time.Världens energibehov och utsläppen av skadliga växthusgaser stiger hela tiden, medan reserverna av fossila bränslen sinar. För tillfället finns det inget klart alternativ till de traditionella energiproduktionsmetoderna för en säker och ren framtid. Fusion kan vara en del av lösningen, tack vare en växthusgasfri, i praktiken outsinlig, säker och storskalig energiproduktion. En stor utmaning för fusion är att producera mer än den energi som krävs för att skapa och upprätthålla fusionsreaktionen. Den mest genomförbara fusionsreaktion bygger på två väteisotoper: deuterium och tritium, som genom fusion blir en helium atom och en neutron. För att dess atomer skall fusionera är de tvungna att överkomma den frånstötande kraften mellan dem, vilket kräver extrema temperaturer. Därmed joniseras partiklarna och det bildas ett gasplasma. På jorden kan detta ske endast genom att separera plasmat från omgivningen, t.ex. genom att använda slutna magnetiska fält för att skapa ett torusformat plasma. Separationen är dock aldrig perfekt och partiklarna i plasmat kommer att växelverka med reaktorväggarna. Plasma-vägg växelverkan är speciellt kraftig vid divertorn, somn är designad att möta plasmat. Det är kritiskt att förstå dessa processer för att utveckla säkra och långlivade material och undvika förorening av bränslet i plasmat. Plasma-vägg växelverkan kan undersökas i befintliga experimentella reaktorer eller i linjära plasmamaskiner. Det experimentella arbetet är dock ofta otillräckligt då det gäller att förstå de underliggande mekanismerna. Vidare uppstår effekterna i material av plasma-vägg växelverkan på en bred tids- och längdskala. Multiskalsmodellering är ett verktyg som överkommer dessa utmaningar och förbättrar antagandena för framtida fusionsreaktorer. I denna avhandling undersöks plasma-vägg växelverkningarna som tar plats i divertorn i en fusionsreaktor genom datorberäkningar. Växelverkan mellan rena och blandade divertormaterial och plasma, samt orenheter, har modellerats. Studien är främst baserad på beräkningar på atomskala. En kinetisk Monte Carlo algoritm har utvecklats för att förlänga resultaten till makroskopiska skalor, vilket tillåter en direkt jämförelse med experiment.
Subject: physics
Rights: This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.


Files in this item

Total number of downloads: Loading...

Files Size Format View
lasa_dissertation.pdf 4.583Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record