In these models, a feedstock first reacts to yield volatiles and intermediate solid products (Fig. 11.2c). The following steps then depend on the feedstock and the
model. In most models, the volatiles undergo thermal cracking while the intermediate solid products further decompose into other final and/or intermediate products and so on. For highly heterogeneous materials, where the notion of ‘pseudo-component’ loses its physical meaning, this conceptualization appears more realistic. However, the main disadvantage of this type of model is the high number of parameters and the difficulty to segregate the different reactions experimentally to clearly estimate their kinetics parameters. Considering all the possibilities for the different reactions under various conditions, this family of models will lead to a myriad of different models.

Table 11.2 lists the kinetics parameters of several models available in the scientific literature along with their range of validity. Note that the parameters reported in Ta­ble 11.2 were all obtained from static thermogravimetric experiments. Furthermore, some of these models assumed an order of reaction, while pyrolysis reactions are in reality characterized by multiple elemental reactions with their specific reaction order. Moreover, it has been demonstrated that the product yield during biomass pyrolysis is strongly dependent on the temperature, while all the models listed in Ta­ble 11.2 assume that the product yields follow the non-isothermal or isothermal TGA temperature profile used to evaluate the kinetics. Therefore, none of these models can confidently reproduce the variability of products yields with respect to temperature. Also, there is significant uncertainty as to whether these kinetic expressions will be accurate when extrapolated to industrial pyrolysis conditions. To model industrial scale pyrolysis processes, it is critical to develop reliable and robust pyrolysis models based on experimental data obtained at representative conditions.