Continuous telescopic synthesis and optimization of sudan yellow-3G in AmarFLO reactors

GOAL

Developing a safer, more controlled, and scalable synthesis method for the Azo Dye Sudan Yellow-3G

Azo dyes represent one of the most significant classes of synthetic colorants, extensively used across industries such as textiles, plastics, inks, and food. However, traditional batch manufacturing processes often face challenges related to safety risks, inefficient heat and mass transfer, and inconsistent product quality.

Close-up of an organized wooden grid holding an assortment of vibrant powdered pigments in various colors including yellow, orange, brown, red, green, and blue.

Flow diagram of a chemical process with four reactors in sequence: Micro Reactor Step I receives phenyl hydrazine and methyl acetoacetate in AcOH with utility I at temperature T1°C; output flows to Micro Reactor Step II with NaOH solution and utility II at T2°C; parallel inputs of aniline in HCl and water plus NaNO2 solution in water enter Micro Reactor Step III with utility II at T2°C; outputs from Steps II and III feed into a Slurry Reactor Step IV producing the reaction crude outlet.

METHOD

Continuous telescopic synthesis via MicroFLO™ and SlurryFLO™ reactors

A continuous telescopic workflow was implemented using Amar-made MicroFLO™ reactors for the preparation of the coupler and diazonium salt, followed by the azo coupling step in a SlurryFLO™ reactor. This setup allowed hazardous and highly exothermic reactions to be handled in a safer, continuous manner while maintaining precise control over reaction parameters. The process was further optimized through the ReactWise software platform, which employed machine learning to identify optimal conditions, reduce experimental workload, and generate advanced process analytics.

Key parameter identification

A parameter importance plot revealed that coupling temperature (CTemp) and the molar ratio of sodium nitrite to aniline were the dominant factors, contributing 27.0% and 25.3% of model variance, respectively. Identifying these critical variables is essential, as it allows us to carry these learnings forward to refine and streamline future campaigns.

Bar chart titled 'Parameter Importance Analysis' showing relative importance percentages: CTemp about 28%, Eq of NaNO2:An about 26%, Eq of MAA:PhH about 15%, CTime about 10%, DTime about 8%, DTemp about 7%, and Eq of NaOH:PhH about 5%.

OUTCOME

ReactWise reduced the experimental workload by 97%

We reduced the number of experiments to 71, compared to a full factorial design (2,187), while still uncovering non-linear dependencies and narrow operating windows that would be difficult to identify using traditional methods.

Bar chart titled Optimization Efficiency comparing number of experiments between Full Factorial (3-level) with 2,187 experiments and ReactWise BO with 71 experiments, highlighting 97% fewer experiments for ReactWise BO.