Using genetic algorithms to systematically improve the synthesis conditions of Al-PMOF

The synthesis of metal-organic frameworks (MOFs) is often complex and the desired structure is not always obtained. In this work, we report a methodology that uses a joint machine learning and experimental approach to obtain the optimal synthesis of a MOF. A synthetic conditions finder was used to derive the experimental protocols and a microwave based high-throughput robotic platform was used for the synthesis of Al-PMOF ([H2TCPP[AlOH]2(DMF3(H2O)2)]). Al-PMOF was previously synthesized using a hydrothermal reaction, which gave a low throughput yield due to its relatively long reaction time (16 hours). In this work, we carried out a systematic search for the optimal reaction conditions using a microwave assisted reaction synthesis. For this search we used a genetic algorithm and we show that already in the 2nd generation we obtained conditions that give excellent crystallinity and yield close to 80% in much shorter reaction time (50 minutes). In addition, by analysing the failed and partly successful experiments, we could identify the most important experimental variables that determine the crystallinity and yield.