This dissertation deals with the characterization of several biomass materials under thermal conversion conditions using small--scale equipment. The fuels are tested under the conditions of slow and fast heating rate pyrolysis and combustion, with the main goal of investigating the chemistry of fuel--bound nitrogen. New energy policies put forward in almost every country in the world, and especially in the European Union, are strongly promoting the use of renewable energy sources. Decreasing the use of imported fossil fuels in favour of locally available renewables is the answer to many energy--related problems of the 21st century: global warming, security of supply and high energy prices among some. Among renewable sources, biomass materials hold a special position because they can, in the short term, substitute or integrate fossil fuels in all of their applications applying comparatively few changes to the existing equipment. Biomass wastes, from agriculture or other processes, are convenient in more respects since their use would not only substitute fossil fuels but it would also valorize waste streams. These materials, however, present several issues that are highly delaying their deployment on a large scale. Three of the most important problems are dealt with in this thesis: the heterogeneous nature of the materials, high amount of ash forming matter containing troublesome compounds such as K, Cl and P, and finally, high content of nitrogen. First of all, many biomass residues contain a higher amount of nitrogen compared with woody biomass or even coal. This high content of fuel--N could directly translate into high NOx emissions in combustion conditions or into a high content of nitrogen containing gases such as NH3 and HCN in the syngas from gasification. Primary measures, such as air staging, can be applied directly in the reactor in order to promote the reduction of NOx and NOx--precursors to molecular nitrogen. However, in order to apply such measures and optimize the syngas composition or minimize emissions without relying on expensive catalysts, a detailed knowledge of the mechanisms of fuel--N conversion is required. This thesis has as its main purpose to study the release of volatile nitrogen compounds under pyrolysis conditions and the analysis of the emissions of NO under combustion conditions from high--N fuels. Secondly, as explained in the first two chapters of this dissertation, the definition of biomass is very broad and it includes materials with extremely different composition and characteristics. Additionally, the interest in exploiting some of these materials, such as manures, for energy conversion has never been high enough to trigger substantial research. As a consequence, fundamental data such as reactivity and products distribution are almost completely lacking for many biowastes. One of the purposes of this thesis is, therefore, to gather extensive fundamental data for potential fuels, which have not yet fully characterized. Finally, some elements such as K, Cl, P and S, contained in biomass materials, are known to cause several problems during boiler operation. At high temperatures alkali silicates with melting temperatures lower than the operating one are formed; these partly molten particles can then create issues like slagging, fouling, loss of fluidization and, when Cl is present, corrosion of the boiler surfaces. Together with specific research on boiler materials and optimization of operating conditions, possible pre--treatments used to remove these compounds from the fuel before entering the reactor could greatly enhance the overall process. In this thesis, the effects of a water--leaching pre--treatment on the fuels' reactivity and product yields during pyrolysis are explored. After a general introduction, Chapter 2 has the purpose of providing the reader with an overview of definitions and concepts that are used in the rest of the dissertation. The main components of biomass are listed as well as their behavior under thermal conversion conditions. A special focus is given to nitrogen structures, such as proteins and heterocyclic compounds, and the mechanisms of their decomposition under pyrolysis conditions as reported in available literature studies. Finally, an overview of NOx formation mechanisms is given. The materials studied in this work and the setups used are introduced in Chapter 3. The materials are divided into two main categories: agricultural residues, including wheat straw, olive residues and peach stones, and biomass waste streams, including dry distiller's grains, palm kernel cake and chicken manure. All these materials have a high nitrogen content and are not yet fully exploited because of the issues mentioned above. Several different small--scale setups have been used in this work depending on the conditions that needed to be study. Chapter 4, then, presents the results of measurements performed on agricultural residues under slow pyrolysis conditions. The measurements were carried out using a thermogravimetric analyzer (TGA) connected with a Fourier Transform InfraRed spectrometer (FTIR). All the fuels presented peaks of reactivity at approximately 330 - 360°C associated with the decomposition of cellulose. A shoulder was also found at lower temperatures, approximately 290 - 310°C, and was associated with the pyrolysis of hemicellulose. The main volatile species released resulted to be CO2 for all the fuels followed by CO and methane. No nitrogen compounds were detected because of too little concentrations in the pyrolysis gases which could not be accurately detected by the FTIR. The kinetic parameters for the pyrolysis reactions were found using a Distributed Activation Energy Model (DAEM). This analysis revealed a common reaction path for the main structures (cellulose and hemicellulose) among the various fuels. The water--leaching pre--treatment seemed very effective in removing the troublesome inorganic elements from these fuels. The removed elements, furthermore, had a catalytic effect on the pyrolysis of the fuels; once removed, the samples reacted at higher temperatures and with less reaction paths. Only the peach stones sample, which already had a very low ash content, did not seem to be affected by the treatment. Chapter 5 reports the results of a similar analysis to the previous one, that was performed on different biomass residues: dry distiller's grains with solubles (DDGS) and chicken manure. Also these fuels were found to react with two main peaks at about 280°C and 330°C. Accordingly, also the kinetic parameters resulted very similar to the ones of the agricultural residues, clearly indicating a similar decomposition path for common structures. For DDGS and chicken manure, an additional component was found to react at approximately 400 - 430°C and it was assigned to proteinic structures decomposing. The main volatile nitrogen compounds were found to be NH3 and HCN for DDGS with traces of HNCO. The manure released more of its fuel--bound nitrogen in volatile form and the main compound was found to be HNCO, followed by HCN and ammonia. The water leaching pre--treatment, similarly to what was found for the peach stones sample, did not seem to affect substantially the reactivity of the fuels nor their ash composition. The share of fuel--N released as light volatiles, however, increased for the washed fuels. Building up on the results of the previous two chapters, Chapter 6 describes the results of fast pyrolysis measurements of DDGS and palm kernel cake. These tests were carried out on a heated foil reactor integrated with an FTIR using much higher heating rates, closer to industrial applications. A numerical model of the reactor has been developed in order to have a better insight into the temperature and velocity profiles in the reactor chamber. The simulations, in combination with non--contact temperature measurements, have indicated a significant difference between the expected foil temperature and the actual one. This has been corrected in the experimental campaign, granting a more precise knowledge of the actual conditions. The fast pyrolysis measurements have shown an increased weight loss compared to slow heating rates. CO2 resulted to be still the main light volatile at temperatures below 900°C while CO became more relevant at high temperatures due to tar cracking in the hot area around the foil. Compared to low heating rates measurements of the DDGS sample, HCN was the main volatile nitrogen compound while the yield of NH3 was much lower. The palm kernel cake sample only released detectable yields of HCN, nor ammonia nor HNCO were found. Even at high temperatures, approximately 10% of the initial N was retained in the char of DDGS. The water leaching pre--treatment again did not affect the weight loss behaviour neither the main gaseous product distribution but the yields of light volatile nitrogen compounds increased for the washed samples. After the pyrolysis behaviour of different nitrogen compounds has been analyzed, Chapter 7 presents the results of measurements under combustion conditions for several biomass residues. It is shown that the devolatilization profiles for carbon to CO + CO2 and fuel--N to NO are very similar among very different fuels, implying that a common approach could be taken for their modeling and it could very well be acceptable as a first approximation. Moreover, the conversion of fuel--N to NO appeared to follow a decaying trend where the fuels with lower initial N content presented a higher conversion than fuels with higher nitrogen content. Fuels with high nitrogen content, in fact, are likely to release a larger amount of it in volatile form with a consequent larger amount of NHi radicals available for thermal De--NOx reactions. Finally, the release of carbon as CO2 and nitrogen as NO during devolatilization was found to be proportional for temperatures around 800°C while at 1000°C almost all of the NO was released during devolatilization. Finally, in Chapter 8 the main conclusions of the work described in thesis are drawn and recommendations for future research are expressed.