Metal organic frameworks (MOFs) are microporous materials composed of inorganic building units held together by organic molecules. As a result of their construction, there is an enormous variety of compositions, structures, properties and applications. Due to their almost infinite chemical space containing different substances, it is advantageous to use computer-based methods instead of experimental methods for the discovery of MOFs. In this work, Bayesian optimization was applied based on a given dataset to optimize the MOF discovery process and find the optimal synthesis of MOF nanoparticles. Bayesian optimization is a global optimization strategy for hard to evaluate black-box functions. However, the given data set used to train the model was small and therefore, it was difficult to obtain useful predictions. As a solution, Gaussian process regression was applied to obtain more datapoints. The overall goal of this work was to find a suitable kernel for the Gaussian Process and compare different models of Bayesian optimization. After evaluating the single-objective Bayesian optimization, a multi-objective Bayesian optimization was applied and the Pareto front was calculated.