Comprehensive Mapping of Compositional Dynamics in the Formose Reaction Network

A chemical reaction network (CRN) can generate complex molecules from simple starting materials through multiple interconnected reactions. The formose reaction, which produces monosaccharides from formaldehyde (HCHO), is a prototypical non-enzymatic CRN. Although more than half of the feedstock in this reaction is converted into non-target products (products other than the target monosaccharides), traditionally referred to as "tar," the compositional evolution of these products has remained unexplored. In this study, we visualized the formose reaction's compositional landscape (target and non-target products) by applying van Krevelen (VK) diagrams and Kendrick mass defect (KMD) plots using data from high-resolution mass spectrometry, enabling comprehensive visualization of the products' composition without structural identification of individual compounds. Under NaOH catalysis, various side reactions proceeded with low selectivity, leading to a wide variety of non-target products, whereas under Ca(OH)₂ catalysis, dehydration-type side reactions became predominant in the later stages. These findings reveal that the product distribution in the formose reaction is highly catalyst- and stage-dependent. The developed visualization framework provides a powerful tool for understanding complex CRNs and suggests rational strategies for reaction control, such as suppressing dehydration to improve the yield of target products in the presence of Ca(OH)₂ catalyst.