Catalytic cracking of biomass tar for hydrogen-rich gas production: Parameter optimization using response surface methodology combined with deterministic finite automaton

In a two-stage fixed-bed reaction system, with Ni-Ca-Co/HZSM-5 as the catalyst and toluene as the tar model compound, a Box-Behnken experiment was designed using the response surface module in Design-Expert.V8.0.6.1, investigating the influences of cracking time, injection rate of model compound, and catalyst dosage. The process conditions of catalytic cracking of tar were determined using the external standard method of areas with hydrogen-rich gas yield and H2 concentration as two response values. Subsequently, these conditions were optimized and validated using the deterministic finite automaton (DFA). Further exploration was conducted on the pathways for catalytic cracking of different categories of tar model compounds to produce hydrogen-enriched gas. The findings demonstrate that the quadratic polynomial mathematical model established from Box-Behnken experimental data is highly significant at P < 0.01. The adjusted coefficient of determination R2 (adj) is 0.9964, with a prediction coefficient R2 (pred) (0.9785) differing by only 0.02, indicating strong agreement between experimental and predicted values. The model exhibits high reliability and precision. Based on the DFA, the optimal conditions for the catalytic cracking of tar were identified, including a catalyst dosage of 5.00 g, cracking time of 70.93 min, and injection rate of model compound of 0.84 mL/min. An appropriate injection rate of model compound and extended residence time can enhance hydrogen yield, achieving high-quality hydrogen-rich gas. Under these conditions, the catalytic cracking of toluene yielded hydrogen-rich gas at a rate of 213.90 mL/g-toluene, and H2 comprised 60.40 % of total gas production. The predicted satisfaction value of 0.9788 approximates the actual satisfaction of 0.9873, validating good model fit and repeatability. This verifies that combining the response surface methodology (RSM) with the DFA to optimize the conditions for the catalytic cracking of tar to prepare high-purity hydrogen-enrich gas is highly reliable. Significant differences in gas release rates and component concentrations were observed among representative tar model compounds in the Ni-Ca-Co-modified catalytic system with HZSM-5 molecular sieve. Through investigating the cracking pathways of tar model compounds, the possible reactions, and synergies among components during real tar cracking were identified. This research lays a solid groundwork for understanding the mechanisms of cracking of real tar, thereby advancing biomass conversion efficiency, and the purification and utilization of high-temperature gases.