A physicochemical methodology for blending biomass to solve ash agglomeration in combustion processes has been used. The method is based on prediction of high melting point compounds by phase diagram and FactSage calculations coupled with experiments at laboratory and pilot scales. Experimental validation of the calculations was carried out by annealing at 1000 °C followed or not by air quenching at the laboratory scale, using either compressed ash or biomass pellets. Samples were characterized using scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy and powder X-ray diffraction. Two biomasses, oak bark, a sawmill residue, and wheat straw, an agricultural residue, were selected on the basis of their ash composition (rich in Ca, K, or Si). Several mixtures were selected with a minimum of the liquid phase calculated by thermodynamic equilibrium. The results show that the chemical reaction was obtained after annealing because new phases were obtained, whereas they were absent from the ashes of single biomasses. The mixture is not simply a dilution. A slightly smaller chemical reaction was observed in biomass pellets than in compressed ash. These results agree with the predictions, except for the K2Ca6Si4O15 ternary compound, absent from the thermodynamic databases but clearly identified in the ash of the 50:50 wheat straw/oak bark mixture. To our knowledge, it is the first time that this phase has been observed in biomass ashes. It has been recently observed in phase equilibria and crystallographic studies. This underlines the interest of the predictions but also the need to improve existing thermodynamic databases, especially on the CaO–K2O–SiO2 system.

No abstract