Scientific research produces a vast volume of information and knowledge about natural phenomena, typically published in papers. This is particularly striking for the enzymatic hydrolysis of lignocellulose, a critical bioprocess for the production of second-generation biofuel. Our objective is to build Qualitative Reasoning (QR) models that capture the knowledge reported in scientific papers and implement putative explanations for concrete observations. QR is an Artificial Intelligence modelling technique that captures knowledge as causal relations to simulate the system behaviour over time from its structure. The rationale for using a qualitative over a quantitative technique is mainly the incomplete understanding of a system, in this case the cellulose degradation mechanism. When developing a QR model of this kind, we first create a base-model, which is then extended to included more features, and explain additional observations. The model presented in this paper captures the interpretations described in three different scientific papers related to the target system and its behaviour. The base-model implements an interpretation based on the accumulation of inactive enzymes. The extension contains model fragments that capture knowledge about the substrate conditions over the process. Both the capacity to represent results of each paper and the target behaviour are examined and discussed.