Development of Machine-Learning Techniques for Time-of-Flight Secondary Ion Mass Spectrometry Spectral Analysis: Application for the Identification of Silane Coupling Agents in Multicomponent Films

Abstract

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) is an important analysis technique that can gather vast amounts of information from surfaces. Recently, machine learning was combined with ToF-SIMS to successfully extract useful information from mass spectra. However, the descriptor generation required for ToF-SIMS analysis using machine learning remains challenging because it requires a lot of effort, is time-consuming, and significantly limits the versatility and practicality of the machine learning approach for ToF-SIMS analysis. Herein, we proposed a new approach to avoid the descriptor generation: to regard ToF-SIMS spectra as images and apply the convolutional neural network (CNN) to analyze these spectral images. We applied and assessed this approach for the identification of silane coupling agents in multicomponent films. Furthermore, the CNN showed higher accuracy than descriptor-based approaches, suggesting its usefulness in achieving the automation and standardization of the ToF-SIMS analysis.

Recent Developments in Machine Learning for Mass Spectrometry

“…For the determination of aerosol composition, a pipeline was developed where multinomial logistic regression was used to classify aerosols based on mass spectra, and subsequent fractional composition was predicted using boosted trees on the probabilities generated by the logistic model . Notably, to combat potential information loss during ion m / z binning arising from the combination of neighboring m / z intensities (Figure ), Imamura and co-workers reported the classification of silane films using a CNN on literal image (.jpg’s not MSI’s) representations of TOF mass spectra . In an additional example, the Musah group utilized direct analysis in real time (DART) HRMS to identify maggot-derived compounds by employing a combination of hierarchical classification trees and an MLP .…”

“… 40 Notably, to combat potential information loss during ion m / z binning arising from the combination of neighboring m / z intensities ( Figure 4 ), Imamura and co-workers reported the classification of silane films using a CNN on literal image (.jpg’s not MSI’s) representations of TOF mass spectra. 41 In an additional example, the Musah group utilized direct analysis in real time (DART) HRMS to identify maggot-derived compounds by employing a combination of hierarchical classification trees and an MLP. 42 We note that this report contains excellent descriptions of model selection and post training validation.…”

Machine learning-assisted structure annotation of natural products based on MS and NMR data

“…Using this method, they successfully identified the silane coupling agent in multi-component film. 78…”

“…Using this method, they successfully identi-ed the silane coupling agent in multi-component lm. 78 Although molecular ngerprint based methods are not as widely used as direct MS/MS comparisons in the annotation of tandem mass spectrometry, their importance cannot be overlooked. The transformation of tandem mass spectrometry data into ngerprint using ML algorithms enables the extraction of crucial information required for constructing molecular structures.…”

Atomic spectrometry update: review of advances in the analysis of metals, chemicals and materials

“…Lang et al developed a method whereby the generation of descriptors was not required for TOF-SIMS spectra to be processed into images by the convolutional neural network . 207 The authors applied the methodology to the analysis of thin films of different silane coupling agents, sometimes in admixture. The convolutional neural network out-performed the descriptor-based approach.…”