Charge collection by perfectly absorbing bodies (probes, dust particles, satellites) immersed in magnetized plasmas is usually addressed in a simplified manner by approximating particle transport as 2D and employing the cross-section of the body as the collecting area. One of the most rigorous analytical solutions, for fully ionized plasmas in the regime of equal ion and electron temperatures (directly relevant for fusion plasmas), has been developed back in 1970 by J. R. Sanmartin. His kinetic theory solution revealed that when the body is close to the plasma potential, the potential distribution along the B-field becomes non-monotonic and a so-called potential ‘overshoot’ emerges. Fluid models, both for weakly ionized and for fully ionized plasmas, were later proposed whose solutions for the electron current at repulsive potential were constructed by relating particle fluxes to the nonmonotonic potential structure of Sanmartin’s model, i.e. using his potential solution as an a-priori postulation. However, up to now, no numerical tests have succeeded in simulating this particle collection regime due to the prohibitively high computational cost associated with the length scale separation of the problem. Namely, for electrons, the collection perpendicular to B-field is strongly suppressed, and flux conservation results in the formation of an extremely extended collection area along it. Simultaneously, it is necessary to resolve length scales of the order of the Debye length. Capitalizing on the state-of-the-art tool, Curvilinear-Particle-In-Cell (CPIC) and frontier computational facilities at Los Alamos National Laboratory, in this work we have simulated this previously inaccessible collection regime. For the first time, the emergence of the non-monotonic potential profile has been confirmed. Moreover, the electron current was shown to be reasonably close to the model predictions and drastically suppressed compared to the typically employed unmagnetized Orbital Motion Limited model. The results of this work have direct implications for applications to probes in magnetized plasmas as well as modelling of dust and droplet survival in fusion environments.
La raccolta di carica da parte di corpi perfettamente assorbenti (sonde, particelle di polvere, satelliti) immersi in plasma magnetizzato è solitamente studiata approssimando il trasporto delle particelle come 2D e assumendo l’area proiettata del corpo come area di raccolta. Nel 1970 J. R. Sanmartin sviluppò una delle teorie analitiche più rigorose per questo regime, valida per plasmi completamente ionizzati con elettroni e ioni alla stessa temperatura (regime direttamente rilevante per plasmi da fusione). La sua teoria cinetica ha rivelato che nel caso in cui il potenziale del corpo è vicino al potenziale del plasma, il profilo di potenziale lungo il campo B diventa non monotono ed appare un cosiddetto ‘overshoot’ di potenziale. Successivamente sono stati proposti modelli fluidodinamici, sia per plasmi debolmente che completamente ionizzati, in cui la corrente elettronica nel caso di potenziale repulsivo viene calcolata assumendo il profilo non monotono di potenziale derivato da Sanmartin come un’ipotesi a priori. Le differenze di svariati ordini di grandezza nelle lunghezze caratteristiche di questo regime implicano un enorme costo computazionale che ha reso finora impossibile simulare numericamente questo problema. Infatti, perpendicolarmente al campo B il moto degli elettroni è fortemente soppresso, e la conservazione del flusso di particelle impone un’area di raccolta molto estesa lungo le linee di campo. Al contempo, è necessario risolvere dimensioni nell’ordine della lunghezza di Debye in tutto il dominio. Grazie al codice Curvilinear-Particle-In-Cell (CPIC) e alle infrastrutture di Los Alamos National Laboratory, è stato possibile simulare questo regime di raccolta prima inaccessibile. Per la prima volta è stata confermata la presenza del profilo non monotono di potenziale. Inoltre, la corrente elettronica derivata da CPIC è risultata abbastanza vicina alle previsioni del modello analitico e fortemente ridotta rispetto al modello ‘Orbital Motion Limited’, comunemente impiegato nonostante sia sviluppato per plasma non magnetizzato. I risultati hanno implicazioni dirette sia per quanto riguarda l’impiego di sonde in plasma magnetizzato che per la modellazione della polvere in plasma da fusione.
Particle-In-Cell modelling of magnetized plasma flux collection by spherical bodies
Pennati, Luca
2022/2023
Abstract
Charge collection by perfectly absorbing bodies (probes, dust particles, satellites) immersed in magnetized plasmas is usually addressed in a simplified manner by approximating particle transport as 2D and employing the cross-section of the body as the collecting area. One of the most rigorous analytical solutions, for fully ionized plasmas in the regime of equal ion and electron temperatures (directly relevant for fusion plasmas), has been developed back in 1970 by J. R. Sanmartin. His kinetic theory solution revealed that when the body is close to the plasma potential, the potential distribution along the B-field becomes non-monotonic and a so-called potential ‘overshoot’ emerges. Fluid models, both for weakly ionized and for fully ionized plasmas, were later proposed whose solutions for the electron current at repulsive potential were constructed by relating particle fluxes to the nonmonotonic potential structure of Sanmartin’s model, i.e. using his potential solution as an a-priori postulation. However, up to now, no numerical tests have succeeded in simulating this particle collection regime due to the prohibitively high computational cost associated with the length scale separation of the problem. Namely, for electrons, the collection perpendicular to B-field is strongly suppressed, and flux conservation results in the formation of an extremely extended collection area along it. Simultaneously, it is necessary to resolve length scales of the order of the Debye length. Capitalizing on the state-of-the-art tool, Curvilinear-Particle-In-Cell (CPIC) and frontier computational facilities at Los Alamos National Laboratory, in this work we have simulated this previously inaccessible collection regime. For the first time, the emergence of the non-monotonic potential profile has been confirmed. Moreover, the electron current was shown to be reasonably close to the model predictions and drastically suppressed compared to the typically employed unmagnetized Orbital Motion Limited model. The results of this work have direct implications for applications to probes in magnetized plasmas as well as modelling of dust and droplet survival in fusion environments.File | Dimensione | Formato | |
---|---|---|---|
2024_04_Pennati_Tesi_01.pdf
accessibile in internet per tutti
Descrizione: testo tesi
Dimensione
32.44 MB
Formato
Adobe PDF
|
32.44 MB | Adobe PDF | Visualizza/Apri |
2024_04_Pennati_Executive_Summary_02.pdf
accessibile in internet per tutti
Descrizione: executive summary
Dimensione
1.51 MB
Formato
Adobe PDF
|
1.51 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/10589/219376