In this master thesis, the potential of ternary diagrams as prediction method for biomass slagging tendency was investigated. Ternary diagrams are generally powerful thermody-namic tools because they enable to make decisions without doing any experiment. How-ever, a preliminary literature research showed that ternary diagrams are mostly used for classication purpose and not as an established prediction method. The limits of ternary diagrams are rst the availability of data and secondly, the prediction accuracy. Indeed, in these ternary systems, the ash composition is approximated to the sum of the main oxides. By neglecting the other components, the melting temperature predicted is generally way higher than the one measured by Ash Fusion Test (AFT) for example be-cause these components contribute to decrease the melting temperature. Thus, further developments are needed to extract the most out of this potential prediction method. In the practical part of this thesis, the combustion of olive pellet was performed in a lab-scale uidized bed reactor provided by the Institute of Chemical, Environmental and Bioscience Engineering at TU Wien. The experiment was performed two times in order to check results consistency. The bed agglomeration was detected at 950 C approx-imatively when bed temperature and pressure under the bed decrease. The pressure drop can be explained by agglomerates formation which oer a bigger counter-pressure to the air pressure. The bed temperature decreases because the combustion eciency goes down due to agglomerates formation. The bed temperature (950 C) measured at the agglomeration time is between the deformation temperature (840 C) and the ow temperature measured by AFT (1400 C) which is what was expected.