Automatic Optimization of Gradient Conditions by AI Algorithm Using Integrated LC System

Significance of the Topic

High performance liquid chromatography method development is traditionally labor intensive and requires specialist expertise to design gradient programs and interpret results. Automating this process with artificial intelligence accelerates workflows, reduces manual intervention and enhances reproducibility, benefiting pharmaceutical, environmental and industrial laboratories.

Objectives and Study Overview

This study demonstrates automatic optimization of gradient conditions for the separation of seven small-molecule analytes using LabSolutions MD AI software integrated with the Shimadzu i-Series LC system. The goals were to achieve a minimum resolution greater than 3.0 and limit the elution time of the last peak to under 10 minutes.

Methodology

An initial set of five gradient profiles was defined (acidified water versus acetonitrile) with ranges of organic modifier from 20 % to 95 %. A Shim-pack Scepter C18 column (100 mm × 3.0 mm, 1.9 µm) at 40 °C, a flow rate of 0.7 mL/min and UV detection at 254 nm were employed. LabSolutions MD alternated between AI-driven condition searches and correction analyses to refine the gradient according to the set resolution and runtime criteria.

Used Instrumentation

• Shimadzu i-Series integrated LC system (LC-2080C 3D)
• LabSolutions MD method development software
• Shim-pack Scepter C18-120 column (100 mm × 3.0 mm I.D., 1.9 µm)
• UV detector at 254 nm

Main Results and Discussion

Initial gradient runs revealed insufficient separation between two target peaks. After four AI-guided correction cycles, an isocratic hold and slope adjustments yielded baseline resolution above 3.0 for all seven compounds and reduced the last peak elution time below 10 minutes. The system effectively explored the multidimensional gradient space without human intervention.

Benefits and Practical Applications

• Significant reduction of manual workload in method development
• Ability to optimize both new and existing LC methods
• Consistent achievement of user-defined performance criteria
• Accessibility for analysts with limited chromatographic experience

Future Trends and Potential Applications

Advances may include expansion to multiple detection modes, integration of real-time feedback for continuous method refinement, predictive maintenance linked to AI diagnostics and cloud-based libraries for rapid method transfer across laboratories.

Conclusion

The AI algorithm in LabSolutions MD, combined with the Shimadzu i-Series LC platform, successfully automated gradient optimization for a seven-component mixture, meeting stringent resolution and runtime requirements while minimizing human input.

References

• Fujisaki S Application News Automatic Optimization of Gradient Conditions by AI Algorithm Using Integrated LC System LabSolutions MD Technical Report C190-E309 Shimadzu Corporation 2025