From bio-oils to chemicals: study of hydrodeoxygenation reactions of model compounds

In the last decade the growing dread of fossil fuel depletion as well as the increasing change in climate due to anthropogenic emissions have provided a compelling motivation for exploring alternative source of fuel, chemicals and energy. First generation bio fuels (i.e Bioethanol and Biodiesel) developed in the last 30 years from food grade feedstock and edible crops have pointed out the potentiality of biomass conversion into chemicals and fuels; however, these technologies have a major disadvantage in the competition with land usage for food production. In order to prevent this problem and avoid unjustified land usage researchers and industries have focused their attention on second generation biofuels, which are based on the conversion of biomass obtained from agricultural waste. The variable chemical composition of lignocellulosic biomass and the possibility of converting this feedstock in different high value products has prompted an important analogy with the refinery infrastructure, and from this comparison the concept of “Bio-refinery”, an integrated facility that could produce power, fuels and chemicals, has been introduced. Nowadays the carbohydrate fractions (i.e. hemicellulose and cellulose) are converted respectively by fermentation and chemo catalysis in fuels (i.e. bio-ethanol) and chemicals (i.e. diols), while a unique and competitive conversion pathway for lignin has not yet achieved a commercial level, even if some lignin conversion processes have been tested on pilot scale. The progressive diffusion of second generation bioethanol plant will make available in the next future massive amount of lignin and therefore the study of its conversion in high value chemicals is key step that will enhance the biorefinery sustainability. Moreover in the Kraft Pulp Industry new technologies to separate and extracts lignin from black liquor have been developed, and some industrial plant are already existing. This will put on the market additional volumes of lignin for possible bio-based chemical production. Currently lignins are cracked into liquid products by different processes. The liquid product obtained from these technologies are generally named “bio oils” and is a complex mixtures of different chemical compounds such as phenols, ethers and carboxylic acids. Due to their high oxygen content (15 - 40%wt), this mixture presents some detrimental properties like high viscosity, thermal instability and immiscibility with the most common liquid fuel; therefore, overcoming these deleterious properties by implementing an efficient deoxygenation process is an essential processing goal. Catalytic hydrodeoxygenation offers a suitable way to convert oxygen rich bio oils into hydrocarbons: an optimal tuning of the operating conditions (i.e. hydrogen pressure and temperature) allows the modulation of the deoxygenation degree from a simple stabilization (elimination of reactive functions such as carbonyl and olefin) to a complete conversion toward aromatic or aliphatic hydrocarbons. Most of the works published in recent years have demonstrated that HDO of bio oil is feasible, although it requires a significant effort in overcoming operating problems such as reactor plugging, catalyst deactivation and yields optimization. The complex composition and the variability of bio oils enhance the difficulty of understanding the main variables involved in HDO reactions and increases the effort of developing an efficient analytical background. A solution which allows to overcome this problems by simplifying both the reactivity of the system and products analysis is the study of a model compounds. The model compounds of choice are mono- or di-aromatic molecules which keep the linkage and the functional groups contained in bio-oils (i.e ether linkage, hydroxyl groups); the most important model used to mimic the reactivity of real mixtures are phenol, guaiacol, siringol and dimeric models such as dibenzyl etherl/biphenol. A stable and active catalyst is necessary for the conversion of bio-oils through HDO reaction; several works have been focused on conventional hydro-treating catalysts based on nickel or cobalt promoted by molybdenum, typically used in hydro-desulfuration (HDS) of crude oil. A critical review of the state of art has pointed out how pre-sulfide catalyst such as CoMo/Al2O3 or NiMo/Al2O3 are very active in HDO reactions, nevertheless in order to maintain constant their activity H2S has to be fed, leading to sulfur contamination in the products. The limitations of sulfide catalysts have prompted the use of more stable catalysts based on metal oxides such as cobalt and nickel. Initially, the performance of four commercial catalysts have been evaluated in the batch HDO of GUA, an aromatic molecule which carries the typical functional groups found in several bio-oil type. The catalysts were selected considering the state of the art and criteria such as cost and availability. The best performances were obtained using nickel based catalysts, as the cobalt ones have shown a lower activity towards deoxygenation reactions The aromatic ring saturation is a side reaction that always occurs during HDO, causing a lack of selectivity towards aromatic compounds such as benzene, toluene and xylene. This work has pointed out how an optimal tuning of hydrogen pressure could limit this reaction and reduce waste of hydrogen Temperature and residence time were also evaluated through a series of reactions carried out at 25 bar. The optimal processing temperature was found to be 300°C; at low temperature hydrogenation reaction are thermodynamically favored, and a higher selectivity toward naphtenics has been registered. Conversely, at high temperature the increase of both the aromatic components and the hydro-deoxygenation degree does not economically justify the choice of the highest temperature as an industrial plant operating parameters. In all the configurations tested, GUA reached a full conversion in one hour of reaction, while at lower residence times the main products has always been oxygenated compound such as cyclohexanol and phenol, this investigation has pointed out the predominance of hydro-deoxygenation reaction at long contact time The study of other monomeric and dimeric compounds has also carried out. Monomeric molecules such as cresol, anisole, and eugenol have shown results consistent with the results found for GUA. Conversely, dimeric model compounds exhibit a very complex reactivity, directly correlated with the presence of aromatic carbon-oxygen or carbon – carbon bonds. The final part of this work has been focused on the development of a continuous process carried out at low pressure, aimed at converting GUA in methylated phenol thought trans-methylation reactions. In this case cobalt/alumina catalyst at different temperatures and residence times has been used, and the experimental data have been collected to fit a good agreement with a first order kinetic model.

Nell’ultimo decennio il contesto macroeconomico e l’attenzione a problematiche ambientali ha portato ad un sempre maggiore interesse della comunità scientifica nei confronti delle fonti di energia alternative come strumento necessario per l’elaborazione di un modello di crescita sostenibile. La lignina, per le sue caratteristiche quali, fonte rinnovabile, elevata disponibilità ed alto tenore di carbonio può rappresentare un’alternativa al petrolio. Attualmente questo polimero poli-aromatico viene convertito per via termochimica o per depolimerizzazione catalitica in miscele di composti generalmente chiamate “bio-olii”. L’elevato tenore di ossigeno che caratterizza i bio-olii rende necessario lo sviluppo di un efficace processo di deossigenazione per la conversione di bio-olii in compositi chimici di base. Lo scopo di questo lavoro è lo studio dei processi catalitici di idro-deossigenazione per la conversione di bio-olii in cicloalcani ed aromatici deossigenati. Data la complessità chimica di queste miscele, per una modellazione dalla loro reattività è stata selezionata una molecola modello, il 2 metossi-fenolo, che rappresenta lo scheletro dei uno dei tre principali monomeri costituenti la lignina. Tale composto è infatti caratterizzato sia dalla presenza di un gruppo fenolo che di un etere, e fornisce quindi un importante punto di partenza per comprendere la reattività chimica di queste funzionalità tipicamente presenti sia nella lignina che nei suoi prodotti di depolimerizzazione. La reazione di idro-deossigenazione della molecola modello è stata condotta con differenti catalizzatori eterogenei a base di Nichel e Cobalto al fine di investigare gli effetti del sistema catalitico sulla selettività dei prodotti. Infine, è stato valutato l’effetto sulla selettività di alcuni parametri di reazione quali temperatura e pressione. Lo studio condotto ha permesso di evidenziare il ruolo delle principali variabili nel processo, permettendo inoltre di ottimizzare la conversione verso determinate classi di prodotti.