The rapid development of the global economy has prompted the crisis of fast depletion of fossil fuels and increased environmental pollution. Therefore, great efforts are now spent towards the development of renewable energy sources. Main limit of these technologies is their strong dependence on atmospheric and environmental factors, which means that they are able to provide energy only in an intermittent way. To help for their market growth, efficient and cheap energy storage devices must be implemented. Together with renewable energy sources, the green oil-free future passes also through electric transportations. In this particular case, the main limit of electric vehicles (EVs) regards the small autonomy and the costs, mainly linked to the materials present in the energy storage systems. Nowadays, the most widespread energy storage device is the Li-ion battery: it is a robust technology that is able to guarantee reliability and high performance. Nevertheless, the high flammability and the organic electrolytes, in parallel with the high costs of raw materials, are still unsolved problems. New generation of energy storage systems composed of low-costs, abundant and recyclable materials is necessary for a sustainable future. In this work, the Na0.44MnO2 phase is presented as a promising cathode to be used in these devices. In the first chapter, there is a brief technical introduction about batteries, capacitors and supercapacitors while the main electrochemical behaviours of materials are explained. The second chapter starts with the presentation of Li-ion batteries, the most widespread energy storage system, used in most portable and domestic electric appliances. Properties and advantages are explored, though great attention is given about the unsolved issued (flammability, toxicity, costs). In this chapter, is also shown how Li-ion technology won’t be able to be sustainable in the future because of the low abundance of Li in earth’s crust and the uneven distribution of Li reserves in the world. A valid alternative is the Na-ion technology because of the abundance of this element (that makes it cheaper) and its non-toxicity. Moreover, since Na and Li are one below the other in periodic table, their chemical properties are very similar and some materials already proven to be efficient in Li-ion systems can be used in Na-ion ones as well. In the third chapter is given an overview about the most promising materials for Na-ion electrodes. The introduction of Na0.44MnO2 (NMO) phase starts in chapter 4, with the description of the manganese-oxide family, characterized by a wide variety of crystals with 1D, 2D and 3D channels where sodium ions can easily diffuse inside/outside. Crystal structure and electrochemical properties of the various allotropic forms of the oxide are presented. The NMO phase, here described in detail, is particularly interesting because of its very good performance in organic electrolytes and its ability of being electrochemically active even in aqueous electrolytes. Very few literature is present about this material, though some research groups have already proven the ability of NMO to be cycled in a reversible way, showing a very good stability and long life-time. The technique of Cyclic Voltammetry, very common for the evaluation of materials electrochemical behaviour, is explained. The definition of ‘capacity’ and ‘capacitance’ is also given, to understand their differences and when they should be used. Experimental procedures and obtained results are reported in chapter 6 and 7, respectively. After the Solid-State synthesis of NMO phase, the obtained powder was characterized in terms of morphology and crystal structure, to compare with results reported in literature. Always for comparison with other papers, electrochemical behaviour of NMO was studied by Cyclic Voltammetry technique. The followed recipe was found to be effective as it allowed to obtain pure NMO phase while behaviour was found to be Faradaic, though strongly limited by kinetics. For the study of NMO through cyclic voltammetry, an electrode was prepared composed by a graphite sheet as the current collector, over which a paste electrode was applied. The paste composition includes three main ingredients: the active material (NMO in this case), a binder (PTFE) to give free-standing mechanical properties and a conductive additive (Carbon Black). The third component was necessary to increase electronic conductivity of the electrode, in order to appreciate faradaic behaviour at acceptable scan rates. Main focus of this work was not about performance of such electrode, already studied by other laboratories, but that of exploring side aspects, though not less important. First of all, the relation between performance (here reported only in terms of capacitance) and electrode composition (in terms of weight percentage of NMO). Values of capacitance were derived by Cyclic Voltammetry tests using a three-electrode setup of the cell. In the end, the best composition was found to be that with the highest percentage of active material, namely 70% w/w, though the real advantage in capacitance was observable only at the very low rate of 0,1 mV/s. Stability or degradation at specific potentials was the second scope of this research, to verify the possibility of enlarging the working potential window, important for the energy density of final device. In order to do that, many cyclic voltammetry tests were run using many different voltage intervals. The occasion was taken to explore oxygen overpotential and its effect on electrode. The Pourbaix diagram of manganese-oxide was taken as reference for the possible regions of degradation. According to the observations, when cycling the electrode in the low potential region below about 0.15 V SHE, some negative effect arose, leading to some irreversible phenomena, thus reducing the cycle-life of electrode itself. Oxygen evolution appeared at a quite high potential, namely 1,45 V vs SHE, and it resulted extremely dangerous for NMO, such that even after few cycles any faradaic reaction was not observable on CV plot. Anyway, also between 1,1-1,4V vs SHE was found a sort of wave in oxidation, whose counterpart wasn’t observable in reduction, indicating the irreversible formation of permanganate, though not clear signs of degradation were detected, maybe because of the very small number of cycles run in the test. The third part of the work was about a detailed study on sodium deintercalation and lattice evolution of NMO. A series of polarization tests were run, trying to freeze the active material at specific potential values. These values were selected looking at the cyclic voltammetry plot, choosing the more significant points (namely, current peaks and valleys). Only the points between 0,2-1 V vs SHE were taken in consideration, as they fell within the voltage window for reversible cycling. All polarized samples were studied by X-ray diffraction and lattice parameters were derived by Rietveld refinement. Pristine NMO powder was also studied with XRD, to have a reference. What has been found was a sensible reduction in b parameter length, with the a and c parameters remaining unaltered. This reduction was more evident in correspondence of current peaks, while was quite negligible at valleys, indicating that the origin of such shrinkage had to be found in faradaic mechanisms, namely sodium-ion deintercalation and manganese oxidation. A possible explanation involving Jahn-Teller distortion of MnO6 octahedra was given: when oxidation of manganese at Mn2 site of unit cell occurred, the tetragonal distortion of the complex disappeared, and a regular octahedron was obtained. The Mn2 octahedra are directed along b direction, such that the shrinkage of their vertical axis produced automatically the shrinkage in b parameter. Accordingly, it is believed that sodium deintercalation has a poor effect on lattice parameters, maybe due to the very large size of channels compared to the ionic radius of Na+. Anyway, the same mechanism is fundamental to define the electrochemical behaviour of NMO phase. The exact sodium diffusion mechanism was not identified experimentally. The only way to explain it was considering the Coulombic repulsions between sodium ions at specific site of the same channels: the larger the overall repulsion, the lower the potential necessary to have sodium diffusion. According to this hypothesis, an energy ranking for the different site is suggested and then their possible association with the voltammetric features.
Il rapido sviluppo dell’economia globale ha portato in auge il problema del consumo dei combustibili fossili e l’aumento dell’inquinamento ambientale. Questa situazione ha fatto da trampolino di lancio per le energie rinnovabili che negli ultimi vent’anni hanno conquistato porzioni sempre maggiori di mercato. I principali limiti di queste tecnologie risiedono principalmente nella loro forte dipendenza da fattori ambientali e atmosferici che li rendono capaci di fornire energia solo in modo intermittente. Il fattore chiave per garantire la definitiva affermazione delle rinnovabili risiede nella possibilità di implementare grossi impianti di stoccaggio per l’energia prodotta, in modo da garantire una fornitura continua di energia al consumatore. Ovviamente, per essere competitivi sul mercato, tali impianti dovranno essere affidabili, efficienti e poco costosi. Il cammino verso un futuro sostenibile passa anche per l’elettrificazione dei trasporti, sebbene i veicoli elettrici attualmente disponibili siano penalizzati di costi elevati e scarsa autonomia. Oggigiorno, il dispositivo di stoccaggio più diffuso è la batteria agli ioni di litio: è una tecnologia efficiente e affidabile, capace di garantire grandi performance. Nonostante ciò, il rischio di infiammabilità, l’uso di elettroliti organici, insieme ai costi elevati delle materie prime, rimangono problemi irrisolti. La nuova generazione di sistemi energetici di stoccaggio dovrà essere composta da materiali economici, abbondanti e riciclabili. In questo lavoro, la fase Na0.44MnO2 è presentata come un valido catodo da essere impiegato in tali sistemi. Il primo capitolo è un’introduzione tecnica circa batterie, condensatori e supercondensatori insieme alla definizione dei principali modelli di comportamento elettrochimico. Il secondo capitolo inizia con la presentazione delle batterie al litio, con le loro proprietà e i loro vantaggi. Grande attenzione viene data in seguito su problematiche specifiche per questo tipo di dispositivi: infiammabilità, tossicità, costi. Nello stesso capitolo viene argomentata la tesi che predice la non sostenibilità di questa tecnologia nell’immediato futuro, principalmente a causa del ridotto contenuto di minerali di litio nella crosta terrestre e la distribuzione altamente disomogenea delle risorse. Una valida alternativa è rappresentata dai dispositivi di stoccaggio agli ioni di sodio, uno degli elementi in assoluto più abbondanti sulla Terra, cosa che ne riduce il costo, e che non presentano problemi di inquinamento o tossicità. Inoltre, siccome sodio e litio sono uno sotto l’altro nella tavola periodica degli elementi, le loro proprietà chimiche sono molto simili, tanto che alcuni materiali già impiegati nelle batterie al litio risultano efficienti anche in quelle al sodio. Nel terzo capitolo viene data una panoramica riguardo i materiali più promettenti da essere impiegati come elettrodi nei dispositivi di stoccaggio agli ioni di sodio. L’introduzione della fase Na0.44MnO2 (NMO) inizia con il capitolo 4, con la descrizione della famiglia degli ossidi di manganese, caratterizzati da una grande varietà di cristalli con canali estesi in una, due o tre dimensioni, attraverso i quali gli ioni di sodio possono facilmente diffondere. Si illustrano più in dettaglio struttura cristallina e relative proprietà elettrochimiche delle diverse forme allotropiche di MnO2. La fase NMO, che qui è descritta in maggior dettaglio, è particolarmente interessante per la sua ottima performance non solo in elettroliti organici ma anche in quelli a base acquosa. Poca letteratura è disponibile al riguardo, sebbene qualche gruppo di ricerca sia stato in grado di mostrarne il comportamento elettrochimico e l’ottima stabilità a ciclatura. Nel capitolo successivo, la voltammetria ciclica, tecnica ampiamente usata in campo elettrochimico per la valutazione di nuovi elettrodi, è descritta nei concetti e nella configurazione di cella. È stata colta l’occasione per discutere in dettaglio la fondamentale differenza tra ‘capacità’ e ‘capacitanza’, due grandezze fisiche molto diverse ma che troppo spesso vengono confuse in seno persino a molti gruppi di ricerca. Le procedure sperimentali e i risultati ottenuti sono descritti rispettivamente nei capitoli 6 e 7. Dopo la sintesi allo stato solido della fase NMO, la polvere ottenuta è stata caratterizzata nella morfologia e nella struttura cristallina e infine comparata al materiale ottenuto in letteratura. Sempre al fine di comparare il materiale con i risultati di altri lavori, il comportamento elettrochimico è stato definito osservando la curva ottenuta mediante voltammetria ciclica. La procedura di sintesi utilizzata è stata trovata corretta e ha permesso di ottenere una fase NMO pura, caratterizzata da un comportamento faradico (come previsto), sebbene fortemente limitato da fattori cinetici. Per lo studio dell’NMO mediante voltammetria ciclica non è stato possibile utilizzare il materiale puro nella forma ottenuta: per ogni test sono stati preparati elettrodi il più possibile simili tra di loro, costituiti da un substrato di grafite sul quale è stato depositato un sottile film di pasta, composta dal materiale attivo (NMO), da un legante (PTFE) e da un additivo per incrementarne la conducibilità elettrica (in questo caso è stato usato del Conductive Carbon o Carbon Black). L’obbiettivo principale di questo lavoro non è stato quello di dimostrare le performance elettrochimiche dell’NMO in ambiente acquoso quanto quello di esplorare aspetti laterali, sebbene non meno importanti nel definirne la reale applicabilità in futuri dispositivi. Prima di tutto, è stata analizzata la relazione tra performance (qui presentate solo in termini di ‘capacitanza’) e composizione chimica (in termini di rapporto in peso tra il materiale attivo e l’additivo carbonioso). I valori di capacitanza sono stati derivati mediante voltammetria ciclica usando una configurazione di cella a tre elettrodi. Al termine di questo iter, la composizione migliore è stata individuata in quella con la più alta percentuale di materiale attivo tra quelle proposte, 70% w/w, sebbene un vantaggio sensibile si è avuto solo per valori estremamente ridotti di velocità di scansione (0,1 mV/s). Lo studio della degradazione del campione a differenti potenziali è stato il punto focale della seconda parte di questo lavoro, per capire la larghezza massima dell’intervallo di potenziale da poter usare garantendo la piena reversibilità del processo elettrochimico. A questo scopo, numerosi test sono stati condotti mediante voltammetria ciclica, adottando intervalli di potenziale sempre diversi fra loro. Nello stesso momento, si è voluto analizzare in dettaglio il fenomeno dell’elettrolisi dell’acqua, per capire il valore di sovratensione legato allo sviluppo di ossigeno sulla superficie del materiale. Secondo quanto osservato, un’importante degradazione prende atto ogni volta che il limite inferiore dell’intervallo di potenziale permette la riduzione di Mn(III) a Mn(II), a potenziale inferiore a circa 0,15 V SHE. Gli effetti dell’elettrolisi sono osservabili direttamente dalla voltammetria ciclica per valori di potenziale sopra 1,45 V, e a seguito dello sviluppo di ossigeno si ha una degradazione estremamente rapida dell’elettrodo stesso. L’elettrolisi è preceduta dall’ossidazione del manganese(IV) a formare permanganato, MnO4-; tale processo è osservabile durante l’ossidazione con una sorta di onda nel grafico della voltammetria, tra 1,1-1,4 V, e sembra essere irreversibile dato che non è stato possibile identificarne la controparte durante la riduzione del campione. In ogni caso, non si sono rilevati effetti di degradazione dell’elettrodo, probabilmente a causa dei pochi cicli effettuati. La terza e ultima parte del progetto si è incentrata sulla diffusione del sodio lungo i canali del cristallo durante l’ossidazione, e la conseguente evoluzione del reticolo cristallino. Per ‘congelare’ i campioni a specifici livelli di potenziale si è ricorso ad una serie di polarizzazioni. I valori di potenziale selezionati erano relativi a punti di particolare interesse (picchi di corrente e valli) nella curva di voltammetria ciclica dell’elettrodo. Tutti i suddetti punti sono stati scelti nell’intervallo 0,2-1 V vs SHE, cioè l’intervallo nel quale l’elettrodo manifesta un comportamento reversibile. I campioni polarizzati sono stati analizzati mediante diffrazione a raggi X e i parametri reticolari sono stati calcolati degli spettri ottenuti mediante l’affinamento di Rietveld. Con loro, anche il materiale puro è stato analizzato come ulteriori riferimento per l’analisi dello sviluppo reticolare. Una sensibile riduzione del parametro b della cella è stata sperimentalmente provata, mentre gli altri due parametri, a e c, sono risultati pressoché costanti lungo tutto il processo di ossidazione. La riduzione del parametro b non è stata lineare con il potenziale, ma era più concentrata in corrispondenza dei picchi, indicando che la causa principale di tale modificazione fosse da ricercare nel meccanismo di carica / scarica dell’elettrodo, attraverso i processi di intercalazione /de-intercalazione dello ione sodio e di riduzione / ossidazione della coppia Mn(III)/Mn(IV). Una possibile spiegazione di questa evoluzione asimmetrica ha incluso la distorsione di Jahn-Teller applicata agli ottaedri dei complessi MnO6: l’ossidazione del materiale avviene solo in corrispondenza del sito Mn2, nella quale Mn(III) si trasforma in Mn(IV), rendendo non più stabile la distorsione tetragonale del complesso e riducendo al tempo stesso la lunghezza del suo asse verticale. Siccome gli ottaedri del sito Mn2 sono orientati lungo il parametro b della cella elementare, una riduzione dell’asse verticale si traduce automaticamente nella riduzione del parametro b. La de-intercalazione di sodio sembra aver avuto un ruolo scarso nell’evoluzione reticolare, probabilmente a causa della grande larghezza dei canali ionici, sebbene lo stesso meccanismo sia comunque fondamentale nella definizione del comportamento elettrochimico. Non è stato possibile identificare sperimentalmente l’esatto meccanismo di diffusione ionica. Si è definita invece, in modo qualitativo, una gerarchia dei siti di intercalazione del sodio, considerando la repulsione elettrostatica tra ioni di sodio vicini in siti specifici dei canali: maggiore era la repulsione alla quale erano soggetti, minore era il potenziale necessario per generare la diffusione.
Sodium intercalation and structural modifications in Na0.44MnO2 positive electrode for aqueous electrolyte energy storage devices
ROMANENGHI, RICCARDO
2018/2019
Abstract
The rapid development of the global economy has prompted the crisis of fast depletion of fossil fuels and increased environmental pollution. Therefore, great efforts are now spent towards the development of renewable energy sources. Main limit of these technologies is their strong dependence on atmospheric and environmental factors, which means that they are able to provide energy only in an intermittent way. To help for their market growth, efficient and cheap energy storage devices must be implemented. Together with renewable energy sources, the green oil-free future passes also through electric transportations. In this particular case, the main limit of electric vehicles (EVs) regards the small autonomy and the costs, mainly linked to the materials present in the energy storage systems. Nowadays, the most widespread energy storage device is the Li-ion battery: it is a robust technology that is able to guarantee reliability and high performance. Nevertheless, the high flammability and the organic electrolytes, in parallel with the high costs of raw materials, are still unsolved problems. New generation of energy storage systems composed of low-costs, abundant and recyclable materials is necessary for a sustainable future. In this work, the Na0.44MnO2 phase is presented as a promising cathode to be used in these devices. In the first chapter, there is a brief technical introduction about batteries, capacitors and supercapacitors while the main electrochemical behaviours of materials are explained. The second chapter starts with the presentation of Li-ion batteries, the most widespread energy storage system, used in most portable and domestic electric appliances. Properties and advantages are explored, though great attention is given about the unsolved issued (flammability, toxicity, costs). In this chapter, is also shown how Li-ion technology won’t be able to be sustainable in the future because of the low abundance of Li in earth’s crust and the uneven distribution of Li reserves in the world. A valid alternative is the Na-ion technology because of the abundance of this element (that makes it cheaper) and its non-toxicity. Moreover, since Na and Li are one below the other in periodic table, their chemical properties are very similar and some materials already proven to be efficient in Li-ion systems can be used in Na-ion ones as well. In the third chapter is given an overview about the most promising materials for Na-ion electrodes. The introduction of Na0.44MnO2 (NMO) phase starts in chapter 4, with the description of the manganese-oxide family, characterized by a wide variety of crystals with 1D, 2D and 3D channels where sodium ions can easily diffuse inside/outside. Crystal structure and electrochemical properties of the various allotropic forms of the oxide are presented. The NMO phase, here described in detail, is particularly interesting because of its very good performance in organic electrolytes and its ability of being electrochemically active even in aqueous electrolytes. Very few literature is present about this material, though some research groups have already proven the ability of NMO to be cycled in a reversible way, showing a very good stability and long life-time. The technique of Cyclic Voltammetry, very common for the evaluation of materials electrochemical behaviour, is explained. The definition of ‘capacity’ and ‘capacitance’ is also given, to understand their differences and when they should be used. Experimental procedures and obtained results are reported in chapter 6 and 7, respectively. After the Solid-State synthesis of NMO phase, the obtained powder was characterized in terms of morphology and crystal structure, to compare with results reported in literature. Always for comparison with other papers, electrochemical behaviour of NMO was studied by Cyclic Voltammetry technique. The followed recipe was found to be effective as it allowed to obtain pure NMO phase while behaviour was found to be Faradaic, though strongly limited by kinetics. For the study of NMO through cyclic voltammetry, an electrode was prepared composed by a graphite sheet as the current collector, over which a paste electrode was applied. The paste composition includes three main ingredients: the active material (NMO in this case), a binder (PTFE) to give free-standing mechanical properties and a conductive additive (Carbon Black). The third component was necessary to increase electronic conductivity of the electrode, in order to appreciate faradaic behaviour at acceptable scan rates. Main focus of this work was not about performance of such electrode, already studied by other laboratories, but that of exploring side aspects, though not less important. First of all, the relation between performance (here reported only in terms of capacitance) and electrode composition (in terms of weight percentage of NMO). Values of capacitance were derived by Cyclic Voltammetry tests using a three-electrode setup of the cell. In the end, the best composition was found to be that with the highest percentage of active material, namely 70% w/w, though the real advantage in capacitance was observable only at the very low rate of 0,1 mV/s. Stability or degradation at specific potentials was the second scope of this research, to verify the possibility of enlarging the working potential window, important for the energy density of final device. In order to do that, many cyclic voltammetry tests were run using many different voltage intervals. The occasion was taken to explore oxygen overpotential and its effect on electrode. The Pourbaix diagram of manganese-oxide was taken as reference for the possible regions of degradation. According to the observations, when cycling the electrode in the low potential region below about 0.15 V SHE, some negative effect arose, leading to some irreversible phenomena, thus reducing the cycle-life of electrode itself. Oxygen evolution appeared at a quite high potential, namely 1,45 V vs SHE, and it resulted extremely dangerous for NMO, such that even after few cycles any faradaic reaction was not observable on CV plot. Anyway, also between 1,1-1,4V vs SHE was found a sort of wave in oxidation, whose counterpart wasn’t observable in reduction, indicating the irreversible formation of permanganate, though not clear signs of degradation were detected, maybe because of the very small number of cycles run in the test. The third part of the work was about a detailed study on sodium deintercalation and lattice evolution of NMO. A series of polarization tests were run, trying to freeze the active material at specific potential values. These values were selected looking at the cyclic voltammetry plot, choosing the more significant points (namely, current peaks and valleys). Only the points between 0,2-1 V vs SHE were taken in consideration, as they fell within the voltage window for reversible cycling. All polarized samples were studied by X-ray diffraction and lattice parameters were derived by Rietveld refinement. Pristine NMO powder was also studied with XRD, to have a reference. What has been found was a sensible reduction in b parameter length, with the a and c parameters remaining unaltered. This reduction was more evident in correspondence of current peaks, while was quite negligible at valleys, indicating that the origin of such shrinkage had to be found in faradaic mechanisms, namely sodium-ion deintercalation and manganese oxidation. A possible explanation involving Jahn-Teller distortion of MnO6 octahedra was given: when oxidation of manganese at Mn2 site of unit cell occurred, the tetragonal distortion of the complex disappeared, and a regular octahedron was obtained. The Mn2 octahedra are directed along b direction, such that the shrinkage of their vertical axis produced automatically the shrinkage in b parameter. Accordingly, it is believed that sodium deintercalation has a poor effect on lattice parameters, maybe due to the very large size of channels compared to the ionic radius of Na+. Anyway, the same mechanism is fundamental to define the electrochemical behaviour of NMO phase. The exact sodium diffusion mechanism was not identified experimentally. The only way to explain it was considering the Coulombic repulsions between sodium ions at specific site of the same channels: the larger the overall repulsion, the lower the potential necessary to have sodium diffusion. According to this hypothesis, an energy ranking for the different site is suggested and then their possible association with the voltammetric features.File | Dimensione | Formato | |
---|---|---|---|
2019_4_Romanenghi.pdf
solo utenti autorizzati dal 01/04/2020
Descrizione: Testo della tesi
Dimensione
2.91 MB
Formato
Adobe PDF
|
2.91 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/10589/148008