This thesis aims to verify the capabilities of a new algorithm for computing energetic levels. The algorithm is based on the many body perturbation theory through the GW approximation. This allows for ab-initio simulations through a two-step process. First, a calculation based on density functional theory is performed, followed by a GW elaboration. These methods are available in the software package Quantum ESPRESSO. The student will apply the GW method to a series of small molecules known as GW 100, for which the HOMO energies (which correspond to the vertical ionization energies) computed through various codes and approximations are reported. First of all, the student will learn the basic theory of the DFT and GW methods and he will learn to use the codes in the Quantum ESPRESSO package. The student will then obtain the parameters required for the calculations in order to guarantee the convergence of the results. Then, the student will interface with the GW code through AiiDA, which allows the automatic execution of a computation, including the gathering of the data. This will lead to the development of python code. This will allow the student to obtain the energy levels for the reference set GW 100. The student will also have the opportunity to assemble and use a small cluster based on Raspberry Pi 4B+, to assess the scaling capabilities of the new method.
Il lavoro di tesi mira a verificare le potenzialità di un nuovo algoritmo per il calcolo dei livelli energetici. L’algoritmo è basato sulla teoria delle perturbazioni a molti corpi tramite l’approssimazione GW. Questo permette calcoli a principi-primi tramite un procedimento a due passi. Prima viene effettuato un calcolo tramite il metodo del funzionale della densità a cui segue l’elaborazione GW. Tali metodi sono disponibili nel pacchetto software Quantum-Espresso. Lo studente applicherà il metodo GW a una serie di piccole molecole, conosciuta come GW100, per le quali sono state riportate le energie dell’HOMO, corrispondenti alle energie di ionizzazione verticale, calcolate con diversi codici e diverse approssimazioni. Per prima cosa lo studente apprenderà la teoria di base dei metodi DFT e GW e imparerà l’uso dei codici del pacchetto Quantum-Espresso. Lo studente ricaverà quindi i parametri da usare nei successivi calcoli in maniera da garantire la convergenza dei risultati. Poi lo studente dovrà interfacciare il codice GW del Quantum-Espresso con il codice Aiida che permette l'esecuzione automatizzata dei calcoli compresa la raccolta dati. Questo comporterà la scrittura di una parte di codice in python. Questo permetterà allo studente di ottenere i livelli energetici per tutta la serie di riferimento GW100. Lo studente avrà anche l’opportunità di assemblare ed utilizzare un piccolo cluster di calcolo basato su schede Raspberry 4b+ in maniera da verificare le capacità di scaling del nuovo metodo.
Systematic study with the GW approach of the ionization energy and electronic affinity of small molecules
CIGAGNA, SIMONE
2021/2022
Abstract
This thesis aims to verify the capabilities of a new algorithm for computing energetic levels. The algorithm is based on the many body perturbation theory through the GW approximation. This allows for ab-initio simulations through a two-step process. First, a calculation based on density functional theory is performed, followed by a GW elaboration. These methods are available in the software package Quantum ESPRESSO. The student will apply the GW method to a series of small molecules known as GW 100, for which the HOMO energies (which correspond to the vertical ionization energies) computed through various codes and approximations are reported. First of all, the student will learn the basic theory of the DFT and GW methods and he will learn to use the codes in the Quantum ESPRESSO package. The student will then obtain the parameters required for the calculations in order to guarantee the convergence of the results. Then, the student will interface with the GW code through AiiDA, which allows the automatic execution of a computation, including the gathering of the data. This will lead to the development of python code. This will allow the student to obtain the energy levels for the reference set GW 100. The student will also have the opportunity to assemble and use a small cluster based on Raspberry Pi 4B+, to assess the scaling capabilities of the new method.File | Dimensione | Formato | |
---|---|---|---|
Cigagna_Simone.pdf
accesso aperto
Dimensione
567.59 kB
Formato
Adobe PDF
|
567.59 kB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/32195