The chemistry, properties and performance of flame-retardant rubber composites: Collecting, analyzing, categorizing, machine learning modeling, and visualizing.

Rubbers combine the flexibility with mechanical strength, supporting myriad applications, but suffer from inherent flammability. Formulation and production of flame-retardant rubber composites (FRRCs) have intensively been practiced over years, but not comprehensively reviewed. This necessity has outlined collecting, analyzing, screening, classifying, and interpreting the literature with the aim of classifying the FRRCs. We quantified and visualized the flame retardancy of natural and synthetic FRRCs by grouping FRRCs based on their flame retardants (FRs) among green, mineral, phosphorus-based, nitrogen-based, carbonaceous, and hybrids of two or more types. Available data on cone calorimetry, limited oxygen index (LOI), and UL-94 of FRRCs were carefully extracted and plotted. Flame Retardancy Index (FRI) was used to specify the Poor , Good , or Excellent classes of flame retardancy in association with the chemistry and concentration of FRs to broaden the future innovation avenues. Machine Learning (ML) modeling enabled visualization of flame retardancy landscapes of natural and synthetic rubbers in terms of the chemistry and concentration of FRs. Overall, a downward trend in mechanical properties of FRRCSs against FRs amount was oxplored. This study proposed a general guideline for recognizing gaps in previous investigations and mechanistic interpretations. In conclusion, we highlight that the future FRRCs should take advantage of hybridizing FRs in order to meet the fire safety requirements, which would be possible by the innovative colorful checkered flame retardancy chart presented in this survey. [Display omitted] • Reviewed and classified flame-retardant natural and synthetic rubber composites (FRRCs). • Visualized the effect of chemistry of rubbers and flame retardants (type, amount and number). • Took chemical fingerprint of FRRCs by Machine Learning (ML) and discussed mechanistically. • Proposed an inventory map to systematically select the flame retardants for rubber composites. [ABSTRACT FROM AUTHOR]