Plasma–material interaction (PMI) is one of the major challenges in the development of future fusion reactors, as it critically affects plasma–facing components and, in partic- ular, the performance of diagnostic first mirrors (FMs). FMs are the first elements of the optical diagnostics chain and are responsible for directing the light emitted by the thermonuclear plasma to the detectors of the diagnostics apparatus. These optical com- ponents are directly exposed to the harsh reactor environment and suffer erosion from plasma particles (especially neutrals from charge–exchange reactions) as well as contam- ination by sputtered material from the first wall. When plasma particles interact with the reactor wall, sputtering of wall species may occur; these impurities can then enter the plasma, be transported across the vacuum chamber, and redeposit on various surfaces, including FMs. Both erosion and redeposition degrade mirror reflectance and thereby compromise the reliability of the entire diagnostics chain. For this reason, contaminants must be periodically removed to restore mirror optical properties, making effective in–situ cleaning strategies essential. Among the different possibilities, plasma cleaning offers a promising solution for the removal of mirror contaminants. This thesis investigates PMI phenomena in the framework of plasma cleaning of diagnostic first mirrors, focusing on rhodium mirrors contaminated with fusion–relevant boron (B) and boron–tungsten (BW) deposits. Indeed, ITER, the largest fusion experiment ever designed, will feature both the divertor and the first wall made of tungsten. This choice implies the need to develop an effective strategy for removing residual oxygen from the vacuum vessel prior to plasma operation. The adopted solution is boronization, which consists in depositing a thin boron layer on all plasma–facing surfaces. Owing to its high chemical affinity with oxygen, boron efficiently acts as a oxygen getter and removes the residual traces remaining after vacuum pumping. This consideration motivates the selec- tion of the contaminants investigated in this work and highlights the necessity of studying their interaction with plasma. Rhodium thin films were fabricated via nanosecond pulsed laser deposition (ns-PLD) and subsequently contaminated with nanostructured layers produced by femtosecond pulsed laser deposition (fs-PLD), a novel approach in this context. The contaminated samples were then exposed to deuterium plasmas, representative of reactor operation, and argon plasmas, relevant for in-situ cleaning. Plasma exposures were performed in the GyM linear device at the Istituto per la Scienza e Tecnologia dei Plasmi (ISTP–Milan). This machine enables the generation of plasma with ion energies and fluxes comparable to those of charge-exchange neutrals (CXNs) impacting ITER’s first wall (Γ ≈ 3.7 × 1020 ions/m2s, E ≈ 200 eV), and allows fine tuning of ion impact energies to closely reproduce ITER in-situ cleaning conditions. These features enabled the investigation of different scenarios by varying exposure times and ion energies. Post–exposure characterization was carried out using Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDXS) to assess morphological and compositional changes. The results reveal a strong dependence of erosion and cleaning efficiency on plasma type and ion energy. Argon plasma exposures induced pronounced nanostructuring of boron films together with the creation of surface patterns very different from that of the pris- tine state. However, Ar proved most effective for tungsten removal, while causing limited damage to the rhodium mirror. Deuterium plasma exposures, on the other hand, nearly completely removed boron nanoparticles without macroscopic damage to the mirrors, al- though at 227 eV (the characteristic energy of CXNs) nanostructuring of the rhodium mirror itself was observed. These findings suggest that CXNs could contribute to a self– cleaning mechanism for boron deposits during reactor operation, when mirrors are sub- jected to CXNs flux, provided that mirror reflectance is preserved. Overall, this work highlights the complementarity of argon and deuterium plasmas for cleaning purposes and establishes a reproducible laboratory–scale framework to inves- tigate plasma erosion phenomena. The novelty lies in the use of fs–PLD to produce nanostructured boron–based contaminants and in the experimental assessment of their interaction with fusion–relevant plasmas. These results provide a valuable contribution to the optimization of in-situ cleaning strategies for future studies, particularly in view of their application to ITER and next–generation fusion devices.
L’interazione plasma–materia (PMI) rappresenta una delle principali sfide per lo sviluppo dei futuri reattori a fusione, poiché influisce in maniera critica sui componenti diretta- mente esposti al plasma e, in particolare, sulle prestazioni dei primi specchi diagnostici (First Mirrors, FM). Questi ultimi costituiscono il primo elemento della catena diagnostica ottica e hanno il compito di convogliare la luce emessa dal plasma termonucleare verso i rivelatori degli apparati diagnostici. Essendo direttamente esposti all’ambiente estremo del reattore, i primi specchi sono soggetti a erosione da parte delle particelle di plasma (in particolare da parte dei neutri prodotti da reazioni di scambio carica) e contaminazione da parte di materiale sputterato proveniente dalla prima parete. Quando le particelle di plasma interagiscono con la prima parete, infatti, possono causarne l’erosione e gli atomi sputterati possono entrare nel plasma come impurità, essere trasportati in tutta la camera da vuoto e ridepositarsi su diverse superfici, inclusi i primi specchi. Erosione e rideposizione compromettono la riflettanza degli specchi e, di conseguenza, l’affidabilità dell’intera catena diagnostica. Per questo motivo, la rimozione periodica dei contaminanti è fondamentale per ripristinare le proprietà ottiche dei primi specchi, rendendo indispens- abile lo sviluppo di efficaci strategie di pulizia in–situ. Tra le varie possibilità, la pulizia al plasma (plasma cleaning) rappresenta una soluzione promettente. Questo lavoro di tesi affronta i fenomeni di PMI nel contesto del plasma cleaning degli specchi diagnostici, con particolare attenzione a specchi in rodio contaminati con depositi di boro (B) e boro–tungsteno (BW), rilevanti per la fusione. Infatti, ITER, il più grande esperimento di fusione mai progettato, avrà sia il divertore sia la prima parete realizzati in tungsteno. Questa scelta comporta la necessità di individuare una strategia efficace per rimuovere l’ossigeno residuo presente nella camera da vuoto prima dell’iniezione di plasma. A questo scopo viene adottata la boronizzazione, che consiste nel depositare un sottilissimo strato di boro su tutte le superfici esposte al plasma. Il boro, grazie alla sua elevata affinità chimica con l’ossigeno, consente infatti di legare e rimuovere in maniera efficiente le tracce residue rimaste dopo il pompaggio. Da qui deriva la scelta dei contaminanti considerati in questo lavoro di tesi e la conseguente necessità di studiarne l’interazione con il plasma. I film sottili di rodio sono stati fabbricati tramite nanosecond pulsed laser deposition (ns-PLD) e successivamente contaminati con strati nanostrutturati prodotti tramite fem- tosecond pulsed laser deposition (fs-PLD). L’utilizzo di questa tecnica rappresente un approccio innovativo alla produzione di contaminanti. I campioni così contaminati sono stati quindi esposti a plasmi di deuterio, rappresentativi delle condizioni operative del reattore, e a plasmi di argon, di interesse per cleaning in-situ. Le esposizioni sono state condotte nella macchina lineare GyM dell’Istituto per la Scienza e Tecnologia dei Plasmi (ISTP-Milano), in grado di generare plasmi con energie e flussi ionici paragonabili a quelli dei neutri di scambio carica (CXN) incidenti sulla prima parete di ITER (Γ ≈ 3.7 × 1020 ions/m2s, E ≈ 200 eV), e di modulare l’energia degli ioni per riprodurre in maniera fedele le condizioni attese nei sistemi di cleaning in-situ di ITER. La caratterizzazione post–esposizione è stata condotta tramite Scanning Electron Microscopy (SEM) ed En- ergy Dispersive X-ray Spectroscopy (EDXS), al fine di valutare le modifiche morfologiche e composizionali dei campioni. I risultati hanno evidenziato una forte dipendenza dei processi di erosione e dell’efficienza di pulizia dal tipo di plasma e dall’energia ionica. Le esposizioni in plasma di argon hanno indotto una marcata nanostrutturazione dei film di boro assieme alla creazioni di pattern superficiali completamente differenti da quelli del film di partenza. Ad ogni modo, le espo- sizioni con Ar esi sono dimostrate le più efficaci per la rimozione del tungsteno, causando modificazioni contenute al substrato di rodio. Le esposizioni in plasma di deuterio hanno invece portato a una quasi completa rimozione delle nanoparticelle di boro senza danni macroscopici agli specchi, sebbene a 227 eV (ener- gia caratteristica dei CXN) sia stata osservata nanostrutturazione del rodio stesso. Questi risultati suggeriscono che i CXN possano contribuire a un meccanismo di auto–pulizia dei depositi di boro durante l’operatività del reattore, proprio quando vengono bombardati da tali neutri di deuterio, a condizione che la riflettanza speculare degli specchi venga preservata. Nel complesso, questo lavoro mette in evidenza la complementarità delle caratteristiche dei plasmi di argon e deuterio in termini di pulizia dei contaminanti in B e BW e propone uno schema riproducibile su scala di laboratorio per lo studio dei fenomeni di erosione indotti dal plasma. La novità del lavoro risiede nell’impiego della fs-PLD per la pro- duzione di contaminanti nanostrutturati a base di boro e nella valutazione sperimentale della loro interazione con plasmi rilevanti per la fusione. Questi risultati rappresentano un contributo prezioso all’ottimizzazione delle strategie di pulizia in-situ per studi futuri, che serviranno a completare la base teorica per ITER e per i futuri dispositivi a fusione.
Plasma cleaning of boron-based contaminants for fusion diagnostic first mirrors
LABERINTI, PAOLA
2024/2025
Abstract
Plasma–material interaction (PMI) is one of the major challenges in the development of future fusion reactors, as it critically affects plasma–facing components and, in partic- ular, the performance of diagnostic first mirrors (FMs). FMs are the first elements of the optical diagnostics chain and are responsible for directing the light emitted by the thermonuclear plasma to the detectors of the diagnostics apparatus. These optical com- ponents are directly exposed to the harsh reactor environment and suffer erosion from plasma particles (especially neutrals from charge–exchange reactions) as well as contam- ination by sputtered material from the first wall. When plasma particles interact with the reactor wall, sputtering of wall species may occur; these impurities can then enter the plasma, be transported across the vacuum chamber, and redeposit on various surfaces, including FMs. Both erosion and redeposition degrade mirror reflectance and thereby compromise the reliability of the entire diagnostics chain. For this reason, contaminants must be periodically removed to restore mirror optical properties, making effective in–situ cleaning strategies essential. Among the different possibilities, plasma cleaning offers a promising solution for the removal of mirror contaminants. This thesis investigates PMI phenomena in the framework of plasma cleaning of diagnostic first mirrors, focusing on rhodium mirrors contaminated with fusion–relevant boron (B) and boron–tungsten (BW) deposits. Indeed, ITER, the largest fusion experiment ever designed, will feature both the divertor and the first wall made of tungsten. This choice implies the need to develop an effective strategy for removing residual oxygen from the vacuum vessel prior to plasma operation. The adopted solution is boronization, which consists in depositing a thin boron layer on all plasma–facing surfaces. Owing to its high chemical affinity with oxygen, boron efficiently acts as a oxygen getter and removes the residual traces remaining after vacuum pumping. This consideration motivates the selec- tion of the contaminants investigated in this work and highlights the necessity of studying their interaction with plasma. Rhodium thin films were fabricated via nanosecond pulsed laser deposition (ns-PLD) and subsequently contaminated with nanostructured layers produced by femtosecond pulsed laser deposition (fs-PLD), a novel approach in this context. The contaminated samples were then exposed to deuterium plasmas, representative of reactor operation, and argon plasmas, relevant for in-situ cleaning. Plasma exposures were performed in the GyM linear device at the Istituto per la Scienza e Tecnologia dei Plasmi (ISTP–Milan). This machine enables the generation of plasma with ion energies and fluxes comparable to those of charge-exchange neutrals (CXNs) impacting ITER’s first wall (Γ ≈ 3.7 × 1020 ions/m2s, E ≈ 200 eV), and allows fine tuning of ion impact energies to closely reproduce ITER in-situ cleaning conditions. These features enabled the investigation of different scenarios by varying exposure times and ion energies. Post–exposure characterization was carried out using Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray Spectroscopy (EDXS) to assess morphological and compositional changes. The results reveal a strong dependence of erosion and cleaning efficiency on plasma type and ion energy. Argon plasma exposures induced pronounced nanostructuring of boron films together with the creation of surface patterns very different from that of the pris- tine state. However, Ar proved most effective for tungsten removal, while causing limited damage to the rhodium mirror. Deuterium plasma exposures, on the other hand, nearly completely removed boron nanoparticles without macroscopic damage to the mirrors, al- though at 227 eV (the characteristic energy of CXNs) nanostructuring of the rhodium mirror itself was observed. These findings suggest that CXNs could contribute to a self– cleaning mechanism for boron deposits during reactor operation, when mirrors are sub- jected to CXNs flux, provided that mirror reflectance is preserved. Overall, this work highlights the complementarity of argon and deuterium plasmas for cleaning purposes and establishes a reproducible laboratory–scale framework to inves- tigate plasma erosion phenomena. The novelty lies in the use of fs–PLD to produce nanostructured boron–based contaminants and in the experimental assessment of their interaction with fusion–relevant plasmas. These results provide a valuable contribution to the optimization of in-situ cleaning strategies for future studies, particularly in view of their application to ITER and next–generation fusion devices.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_Laberinti_Tesi_01.pdf
accessibile in internet per tutti
Dimensione
54.43 MB
Formato
Adobe PDF
|
54.43 MB | Adobe PDF | Visualizza/Apri |
|
2025_10_Laberinti_ExecutiveSummary_02.pdf
accessibile in internet per tutti
Dimensione
7.11 MB
Formato
Adobe PDF
|
7.11 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/243947