Fast Data Sorting with Modified Principal Component Analysis to Distinguish Unique Single Molecular Break Junction...

Area of Science:

• Multivariate data analysis
• Molecular electronics
• Surface science

Background:

• Single-molecule break junction experiments generate large, complex datasets.
• Current analysis methods can be subjective and time-consuming.
• Objective sorting of molecular conductance data is crucial for understanding molecular behavior.

Purpose of the Study:

• To introduce a novel, automated data analysis method for sorting large datasets.
• To apply this method to single molecular break junction conductance data.
• To demonstrate the method's objectivity and efficiency compared to existing techniques.

Main Methods:

• The study employs principal component analysis (PCA) on the correlation matrix of the data.
• Data is sorted based on projections onto the first or second principal component.
• No prior hypotheses about data features are required.

Main Results:

• The PCA-based method successfully sorted mixed datasets of two distinct molecules.
• The method automatically distinguished between high (π-bridge) and low (σ-bridge) conductance molecules.
• Analysis of a single molecule dataset revealed two distinct groups, likely representing different molecular geometries.

Conclusions:

• The developed PCA method offers an objective, automated, and time-efficient approach for analyzing complex multivariate data.
• This technique significantly enhances the analysis of single-molecule break junction experiments.
• The method is broadly applicable to various scientific fields dealing with large datasets.