Increasing population and industrialization are continuously oppressing the existing energy resources and depleting the global fuel reservoirs. The elevated pollution from the continuous consumption of non-renewable fossil fuels also seriously contaminates the surrounding environment. The use of alternate energy sources and chemical raw materials can be an environment-friendly solution to cope with these challenges. Industrial chemistry is in the midst of changing its fossil nature by using renewable biomass and agro-industrial/urban wastes as feedstocks, thus reducing, at the same time, its energy and environmental impact. Lignocellulosic biomass, mainly composed of cellulose, hemicellulose and lignin, ranges from softwood to agriculture and forestry wastes and represents the most abundant resource for modern biorefinery. A possibility is to utilize the lignocellulosic agro-industrial biomass in the fermentation into bioethanol. The conversion of lignocellulosic biomass into higher value-added products like fine chemicals or bio-fuel normally requires a multi-step processing that includes pretreatment (mechanical, chemical, or biological), enzymatic hydrolysis and fermentation process. In this work, the focus has been posed on the pretreatment step. Pretreatment performs the function of breaking down cell wall physical barriers as well as cellulose crystallinity and association with lignin so that enzymes can access the biomass macrostructure to perform hydrolysis. Between existing methods, organosolv pretreatment is especially interesting because it also allows recovery of pure lignin as a solid material and carbohydrates as syrup. This fractionation technique involves the use of aqueous organic solvents with or without catalysts. Recently, the bio-derived γ-valerolactone has received significant attention as a solvent for organosolv pretreatment because it can achieve high yields at low acid concentrations. Moreover, lignin can be easily precipitated from γ-valerolactone/water mixtures through addition of water. This thesis aims to experimentally evaluate, starting from results reported in literature, the optimal conditions for a small-scale batch pretreatment of lignocellulosic biomass in γ-valerolactone/water and to propose a kinetic law for the removal of lignin and hemicellulose. The potential benefits of a ternary solvent system that includes water, γ-valerolactone and the cheaper and more widely used acetone were also investigated. After the determination of an analytical method through the help of extensive practical testing and of the notions and information found in literature, the biomass raw material was analyzed for its composition. It was found to be very rich in lignin (35,6%) and have a comparatively low cellulose content (26,6%), while most of the remaining mass was composed by hemicellulose (31,8%). The bulk of the experiments performed used a solvent mixture of 80:20 w/w% γ-valerolactone/water, which literature reports to be the best solvent ratio for an organosolv fractionation that uses this innovative, biomass derived, stable and biodegradable organic solvent. After checking the repetitiveness of the experimental results, the influence of several operative parameters was studied. With most reactions, where only those performed with the mildest conditions are excluded, the lignocellulosic biomass seemed to reach an almost constant decomposition after a certain reaction time. The experiments were thus carried out for 1-2h at the longest. Very high recovery rates of all the components were registered, reinforcing the potential of this system to be used as a pretreatment step in a modern biorefinery. The concentration of acid catalyst, varied between 25 mM and 75 mM, significantly influenced the biomass decomposition. Hemicellulose and lignin removal from the biomass greatly improved when the acid concentration was increased, but only if the reaction duration was short. With long reactions, the degradation only increased between 25 mM and 35 mM, while higher sulfuric acid concentrations brought little to no change. Cellulose, instead, appeared to suffer a degradation whose severity directly depended on the catalyst concentration chosen, without being limited at 35 mM. The temperature tested varied between 120 ºC and 160 ºC, to avoid excessive formation of hydrolysis inhibitor products, which is favored by high temperatures. Glucan decomposition resulted almost unchanged when a different operative temperature was chosen. Lignin and hemicellulose removal were clearly favored by higher temperatures, especially in the rapidity with which a certain initial amount was degraded. The curves of lignin and hemicellulose decomposition at different temperature, however, seemed to converge to a common trend of lower steepness after long reaction durations. To compare the effectiveness of the use of γ-valerolactone, the much more technologically mature acetone/water organosolv mixture was tested, using operating conditions that had been reported to be optimal. The final yield in terms of delignification and xylan removal was rather similar to that achieved with the same temperature and acid concentration in γ-valerolactone/water, and the glucose decomposition was lower. The degradation, however, appeared slower and the product recovery less optimal. The combination of acetone, γ-valerolactone and water resulted rather unproductive. While it appeared that the substitution of even a small amount of the acetone of an 80:20 w/w% acetone/water mixture with γ-valerolactone could already improve the pretreatment performance, the result was too similar to that obtained with 50:50 acetone/water to constitute an economic improvement. It should be considered that using two solvents could bring higher separation costs and that acetone is currently cheaper than γ-valerolactone. It could be interesting, still, to further investigate comparing the presented data with the results from fractionation in mixtures of acetone, γ-valerolactone and water with a higher water fraction. Empiric kinetic models were also proposed and evaluated for lignin and xylan removal. In both cases, the power law model provides an acceptable fit, but with unrealistic parameters. A double exponential model, which uses the assumption that a part of the polymer is degraded with a different kinetic that the other, was also considered, together with possible modifications. This second choice appeared in the two cases a better fit, but the high number of minimization parameters implies that a higher quantity of experimental data would be needed for a more reliable result. The implementation of the investigated reaction in the context of an integrated biorefinery could be highly interesting because of the optimal use of the biomass source for the production of high-value products such as ethanol, high-quality lignin and xylan derivatives with a high recovery rate. The solvent used also presents the important advantage of being directly derived from cellulose or hemicellulose, so its production could even be integrated in the process.
L'aumento della popolazione e l'industrializzazione stanno continuamente opprimendo le risorse energetiche esistenti e esaurendo le riserve globali di petrolio. L'elevato inquinamento derivante dal consumo continuo di combustibili fossili non rinnovabili, inoltre, contamina gravemente l'ambiente circostante. L'uso di fonti energetiche e materie prime alternative può essere una soluzione sostenibile per far fronte a queste sfide. La chimica industriale sta cambiando la sua natura fossile usando biomassa rinnovabile e rifiuti agroindustriali ed urbani come materie prime, riducendo così, allo stesso tempo, il suo impatto energetico ed ambientale. La biomassa lignocellulosica, composta principalmente da cellulosa, emicellulosa e lignina, spazia dal legno tenero all'agricoltura e ai rifiuti forestali e rappresenta la risorsa più abbondante per la moderna bioraffineria. Una possibilità è quella di utilizzare la biomassa agroindustriale lignocellulosica nella fermentazione in bioetanolo. La conversione della biomassa lignocellulosica in prodotti a più alto valore aggiunto come prodotti chimici fini o biocarburante richiede normalmente una lavorazione a più fasi che include pretrattamento (meccanico, chimico o biologico), idrolisi enzimatica e fermentazione. In questo lavoro, l'attenzione è stata posta sulla fase di pretrattamento. Il pretrattamento svolge la funzione di abbattere le barriere fisiche della parete cellulare, ridurre la cristallinità della cellulosa e la sua associazione con la lignina in modo che gli enzimi possano accedere alla macrostruttura della biomassa per portare a termine l'idrolisi. Tra i metodi esistenti, il pretrattamento con organosolv è particolarmente interessante perché consente anche il recupero della lignina in forma pura come solido e dei carboidrati come sciroppo. Questa tecnica di frazionamento prevede l'uso di solventi organici acquosi con o senza catalizzatori. Di recente, il γ-valerolattone, solvente di derivazione biologica, stabile e biodegradabile, ha ricevuto un'attenzione significativa come solvente per il pretrattamento organosolv perché può raggiungere rese elevate a basse concentrazioni di acido. Inoltre, la lignina può essere facilmente precipitata dalle miscele γ-valerolattone/acqua attraverso l'aggiunta di acqua. Questa tesi si propone di valutare sperimentalmente, a partire dai risultati riportati in letteratura, le condizioni ottimali per un pretrattamento in batch su piccola scala della biomassa lignocellulosica in γ-valerolattone/acqua e di proporre una legge cinetica per la rimozione di lignina ed emicellulosa. Sono stati anche studiati i potenziali benefici di un sistema di solvente ternario che include acqua, γ-valerolattone e acetone più economico e più ampiamente usato. Dopo la determinazione di un metodo analitico attraverso l'aiuto di approfonditi test pratici e delle nozioni e informazioni trovate in letteratura, la materia prima è stata analizzata per la sua composizione. Si è scoperto che la biomassa utilizzata, polvere di gusci di mandorle, era molto ricca di lignina (35,6%) e aveva un contenuto di cellulosa relativamente basso (26,6%), mentre la maggior parte della massa rimanente era composta da emicellulosa (31,8%). La gran parte degli esperimenti condotti ha utilizzato una miscela solvente di 80:20 w/w% γ-valerolattone/acqua, che la letteratura riporta come il miglior rapporto tra solventi per un frazionamento organosolv che utilizza γ-valerolattone. Dopo aver verificato la ripetitività dei risultati sperimentali, è stata studiata l'influenza di numerosi parametri operativi. Con la maggior parte delle reazioni, dove sono escluse solo quelle eseguite con le condizioni più blande, la biomassa lignocellulosica sembra raggiungere una decomposizione quasi costante dopo un certo tempo di reazione. Gli esperimenti sono stati quindi fatti durare 1-2 ore al massimo. Sono stati registrati tassi di recupero molto elevati di tutti i componenti, confermando il potenziale di questo sistema come fase di pretrattamento in una moderna bioraffineria. La concentrazione del catalizzatore, acido solforico, variata tra 25 mM e 75 mM, ha influenzato significativamente la decomposizione della biomassa. La rimozione dell'emicellulosa e della lignina dalla biomassa migliorava notevolmente quando la concentrazione di acido era aumentata, ma solo se la durata della reazione era breve. Con reazioni lunghe, la degradazione è aumentata solo tra 25 mM e 35 mM, mentre concentrazioni più elevate di acido solforico hanno apportato cambiamenti minimi o nulli. La cellulosa, invece, sembrava soffrire di una degradazione la cui severità dipendeva direttamente dalla concentrazione del catalizzatore utilizzata, senza essere limitata a 35 mM. La temperatura testata variava tra 120 ºC e 160 ºC, per evitare un'eccessiva formazione di prodotti inibitori dell'idrolisi, favorita dalle alte temperature. La decomposizione del glucano è risultata pressoché invariata quando è stata scelta una diversa temperatura operativa. La rimozione di lignina ed emicellulosa era chiaramente favorita da temperature più elevate, specialmente nella rapidità con cui una certa quantità iniziale veniva degradata. Le curve di decomposizione della lignina e dell'emicellulosa a diverse temperature, tuttavia, sembravano convergere in un andamento comune di bassa pendenza con tempi di reazione lunghi. Per confrontare l'efficacia dell'uso del γ-valerolattone, è stata testata l’efficienza della miscela solvente acetone/acqua per il trattamento organosolv, tecnologia molto più matura, utilizzando condizioni operative riportate dalla letteratura come ottimali. La resa finale in termini di delignificazione e rimozione dello xilano è risultata piuttosto simile a quella ottenuta con la stessa temperatura e concentrazione di acido in γ-valerolattone/acqua, e la decomposizione del glucosio è stata inferiore. La decomposizione, tuttavia, è apparsa più lenta e il recupero del prodotto meno ottimale. La combinazione di acetone, γ-valerolattone e acqua è risultata piuttosto improduttiva. Nonostante sia apparso che la sostituzione anche di una piccola quantità di acetone in una miscela 80:20 w/w% di acetone/acqua con γ-valerolattone potesse già migliorare le prestazioni del pretrattamento, il risultato raggiunto con il sistema di tre solventi è stato eccessivamente simile a quello ottenuto con 50:50 w/w% di acetone/acqua per costituire un miglioramento economico. Va considerato che l'uso contemporaneo di due solventi organici potrebbe comportare costi di separazione più elevati e che l'acetone è attualmente più economico del γ-valerolattone. Potrebbe essere interessante, tuttavia, approfondire lo studio con un confronto dei dati presentati con i risultati del frazionamento in miscele di acetone, γ-valerolattone e acqua con una frazione di acqua più elevata. In questa tesi sono anche proposti e valutati modelli empirici cinetici per la rimozione di lignina e xilano. In entrambi i casi, il modello della legge di potenza fornisce una misura accettabile, ma con parametri non realistici. È stato anche considerato un modello con doppio esponenziale, che utilizza il presupposto che una parte del polimero sia degradata con una cinetica diversa rispetto alla restante, insieme a possibili variazioni del modello stesso. Questa seconda scelta è apparsa nei due casi più adatta, ma l'elevato numero di parametri di minimizzazione implica che sarebbe necessaria una maggiore quantità di dati sperimentali per un risultato più affidabile. L'implementazione della reazione studiata nel contesto di una bioraffineria potrebbe essere molto interessante grazie all'uso ottimale della biomassa per la produzione di prodotti di valore come etanolo, lignina di alta qualità e derivati dello xilano con un alto tasso di recupero. Il solvente utilizzato presenta anche l'importante vantaggio di essere direttamente derivato dalla cellulosa o dall'emicellulosa, quindi la sua produzione potrebbe persino essere integrata nel processo.
Acid-catalysed fractionation of lignocellulosic biomass in polar aprotic solvent/water mixtures
CORTI, ARIANNA
2018/2019
Abstract
Increasing population and industrialization are continuously oppressing the existing energy resources and depleting the global fuel reservoirs. The elevated pollution from the continuous consumption of non-renewable fossil fuels also seriously contaminates the surrounding environment. The use of alternate energy sources and chemical raw materials can be an environment-friendly solution to cope with these challenges. Industrial chemistry is in the midst of changing its fossil nature by using renewable biomass and agro-industrial/urban wastes as feedstocks, thus reducing, at the same time, its energy and environmental impact. Lignocellulosic biomass, mainly composed of cellulose, hemicellulose and lignin, ranges from softwood to agriculture and forestry wastes and represents the most abundant resource for modern biorefinery. A possibility is to utilize the lignocellulosic agro-industrial biomass in the fermentation into bioethanol. The conversion of lignocellulosic biomass into higher value-added products like fine chemicals or bio-fuel normally requires a multi-step processing that includes pretreatment (mechanical, chemical, or biological), enzymatic hydrolysis and fermentation process. In this work, the focus has been posed on the pretreatment step. Pretreatment performs the function of breaking down cell wall physical barriers as well as cellulose crystallinity and association with lignin so that enzymes can access the biomass macrostructure to perform hydrolysis. Between existing methods, organosolv pretreatment is especially interesting because it also allows recovery of pure lignin as a solid material and carbohydrates as syrup. This fractionation technique involves the use of aqueous organic solvents with or without catalysts. Recently, the bio-derived γ-valerolactone has received significant attention as a solvent for organosolv pretreatment because it can achieve high yields at low acid concentrations. Moreover, lignin can be easily precipitated from γ-valerolactone/water mixtures through addition of water. This thesis aims to experimentally evaluate, starting from results reported in literature, the optimal conditions for a small-scale batch pretreatment of lignocellulosic biomass in γ-valerolactone/water and to propose a kinetic law for the removal of lignin and hemicellulose. The potential benefits of a ternary solvent system that includes water, γ-valerolactone and the cheaper and more widely used acetone were also investigated. After the determination of an analytical method through the help of extensive practical testing and of the notions and information found in literature, the biomass raw material was analyzed for its composition. It was found to be very rich in lignin (35,6%) and have a comparatively low cellulose content (26,6%), while most of the remaining mass was composed by hemicellulose (31,8%). The bulk of the experiments performed used a solvent mixture of 80:20 w/w% γ-valerolactone/water, which literature reports to be the best solvent ratio for an organosolv fractionation that uses this innovative, biomass derived, stable and biodegradable organic solvent. After checking the repetitiveness of the experimental results, the influence of several operative parameters was studied. With most reactions, where only those performed with the mildest conditions are excluded, the lignocellulosic biomass seemed to reach an almost constant decomposition after a certain reaction time. The experiments were thus carried out for 1-2h at the longest. Very high recovery rates of all the components were registered, reinforcing the potential of this system to be used as a pretreatment step in a modern biorefinery. The concentration of acid catalyst, varied between 25 mM and 75 mM, significantly influenced the biomass decomposition. Hemicellulose and lignin removal from the biomass greatly improved when the acid concentration was increased, but only if the reaction duration was short. With long reactions, the degradation only increased between 25 mM and 35 mM, while higher sulfuric acid concentrations brought little to no change. Cellulose, instead, appeared to suffer a degradation whose severity directly depended on the catalyst concentration chosen, without being limited at 35 mM. The temperature tested varied between 120 ºC and 160 ºC, to avoid excessive formation of hydrolysis inhibitor products, which is favored by high temperatures. Glucan decomposition resulted almost unchanged when a different operative temperature was chosen. Lignin and hemicellulose removal were clearly favored by higher temperatures, especially in the rapidity with which a certain initial amount was degraded. The curves of lignin and hemicellulose decomposition at different temperature, however, seemed to converge to a common trend of lower steepness after long reaction durations. To compare the effectiveness of the use of γ-valerolactone, the much more technologically mature acetone/water organosolv mixture was tested, using operating conditions that had been reported to be optimal. The final yield in terms of delignification and xylan removal was rather similar to that achieved with the same temperature and acid concentration in γ-valerolactone/water, and the glucose decomposition was lower. The degradation, however, appeared slower and the product recovery less optimal. The combination of acetone, γ-valerolactone and water resulted rather unproductive. While it appeared that the substitution of even a small amount of the acetone of an 80:20 w/w% acetone/water mixture with γ-valerolactone could already improve the pretreatment performance, the result was too similar to that obtained with 50:50 acetone/water to constitute an economic improvement. It should be considered that using two solvents could bring higher separation costs and that acetone is currently cheaper than γ-valerolactone. It could be interesting, still, to further investigate comparing the presented data with the results from fractionation in mixtures of acetone, γ-valerolactone and water with a higher water fraction. Empiric kinetic models were also proposed and evaluated for lignin and xylan removal. In both cases, the power law model provides an acceptable fit, but with unrealistic parameters. A double exponential model, which uses the assumption that a part of the polymer is degraded with a different kinetic that the other, was also considered, together with possible modifications. This second choice appeared in the two cases a better fit, but the high number of minimization parameters implies that a higher quantity of experimental data would be needed for a more reliable result. The implementation of the investigated reaction in the context of an integrated biorefinery could be highly interesting because of the optimal use of the biomass source for the production of high-value products such as ethanol, high-quality lignin and xylan derivatives with a high recovery rate. The solvent used also presents the important advantage of being directly derived from cellulose or hemicellulose, so its production could even be integrated in the process.File | Dimensione | Formato | |
---|---|---|---|
Tesi - Arianna Corti.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
5.93 MB
Formato
Adobe PDF
|
5.93 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/10589/164644