Gradient Condition Search for Target Peak Separation in Chromatography

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing gradient elution methods in liquid chromatography struggle to completely separate all peaks in chromatograms when the number of components is large, especially when the smallest resolution is less than a threshold, making it difficult to isolate a specific component of interest.

Innovation Solution

A method and system that utilize a model to predict the resolution of a target component under various gradient conditions, allowing for the selection of an optimal gradient condition that maximizes the separation of the target component from other components by creating a model based on chromatogram data and calculating predicted resolutions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the number of components in a sample is large, then the number of peaks in the chromatogram increases, but the smallest resolution becomes smaller than the threshold making it difficult to completely separate all peaks

Engineering Contradiction:

Improvenumber of componentsVSAvoidresolution

Core Design Contradiction:

Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent divides the problem of separating all peaks into segments by allowing the user to specify a target component. Instead of requiring complete separation of all peaks, the system focuses on optimizing separation for the specific component of interest, thereby resolving the contradiction when the number of components is large

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by creating a dedicated model for the target component that predicts its resolution under different gradient conditions. This localized approach optimizes the separation quality for the specific component rather than uniformly optimizing for all components

Inventive Principle:

Principle #3Local quality

2Manufacturing precision

If a regression model is used to calculate the smallest resolution in a chromatogram, then the user can recognize a gradient condition that increases the smallest resolution, but this does not address the user's desire to separate a specific peak when not all peaks can be separated

Engineering Contradiction:

Improvesmallest resolutionVSAvoiduser's desire for specific peak separation

Core Design Contradiction:

Manufacturing precisionVSAdaptability or versatility

Solution Approach 1:

The patent inverts the conventional approach by shifting the focus from optimizing the smallest resolution among all peaks to optimizing the resolution of a user-specified target component. This inversion allows the system to adapt to user-specific needs even when complete separation of all peaks is not achievable

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent changes the parameter being optimized from 'smallest resolution' to 'resolution of target component'. By creating a model that predicts the resolution of the specific target component under different gradient conditions, the system provides adaptability for user-specific separation needs

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260063604A1Method for searching gradient condition, non- transitory computer-readable recording medium, information processing apparatus, and analysis system

Publication Date: 2026.03.05 SHIMADZU CORP

AI summary

A method for searching a gradient condition according to the present disclosure includes a step of receiving information specifying a target component, a step of acquiring first and second chromatogram data obtained by analyzing a sample containing two or more components under mutually different first and second gradient conditions, a step of creating, based on the first and second chromatogram data, a model for the target component that indicates a relationship between a gradient condition and a resolution, a step of using the model to calculate a predicted resolution of the target component corresponding to each of a plurality of gradient conditions, and a step of selecting, from the plurality of gradient conditions, a gradient condition for which the predicted resolution is the largest or for which the predicted resolution is equal to or greater than a threshold value.