Reducing agrochemical emulsifier complexity through dataguided experimentation
Estefania Ramirez, formulation scientist, and Kelly Buchek, R & D Manager at Stepan Company’ s Crop Productivity R & D department, describe developing flexible emulsifier solutions for agrochemical oil systems using a cheminformatics-based modelling platform.
A
recent formulation development programme at Stepan explored whether data-guided experimentation, supported by a cheminformatics platform from Citrine Informatics, could reduce the complexity of agrochemical emulsifier development, while maintaining performance standards required for modern agrochemical products. In this context, the Stepan – Citrine project addresses the growing formulation challenges created by increasingly diverse oil chemistries and the limitations of traditional portfolio expansion approaches.
Agrochemical formulations increasingly incorporate a diverse range of oil solvents, including both petroleum and vegetable-based oils. While this supports performance, sourcing, and sustainability objectives, it also increases formulation complexity, particularly for emulsifier systems that must function reliably across different oil chemistries.
Traditionally, this complexity has been managed by expanding product portfolios, with different emulsifiers tailored to relatively narrow solvent windows. Although it is effective, this approach increases development time, formulation effort and complexity for both manufacturers and customers.
Defining a target
The initial objective of the project was to develop a universal emulsifier solution capable of stabilising both petroleum and vegetable-based oil formulations for use as tank mix adjuvants( TMAs). Within agrochemical formulations, such an‘ all-in-one’
Figure 1 – Interactive visualisation of AI model
emulsifier is a longstanding ambition, offering potential benefits in product simplification and formulation flexibility, reducing costs all round.
Early experimental results, supported by modelling insights from Citrine, highlighted the fundamental differences between the two oil classes. Without this modellingled insight, the development effort would probably have converged on a single, compromised formulation that masked critical differences between oil systems and failed to fully meet performance needs. Rather than pursuing a single compromised formulation, the team reframed the challenge to focus on maximising flexibility while preserving robust emulsification performance.
This led to the development of two complementary emulsifier solutions: one optimised for vegetable-based oil solvents, including both unmodified oils and methylated seed oils, and the other for petroleum-derived oils, including mineral and paraffinic oils. Together, these formulations cover a broad range of agrochemical applications and represent a substantial improvement over previous approaches that required a larger number of more narrowly defined emulsifiers.
Applying cheminformatics
The collaboration leveraged Stepan’ s formulation knowledge alongside the Citrine Informatics platform, utilising historical formulation and performance data to develop an initial modelling framework centred on critical emulsification properties.
Following initial onboarding and training, the team engaged in regular working sessions to support data
36 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981