Publications

Abstract Blind Source Separation (BSS) techniques aim to extract a set of source signals from a measured mixture in an unsupervised manner. In the chemical instrumentation domain source signals typically refer to time-varying analyte concentrations, while the measured mixture is the set of observed spectra. Several techniques exist to perform BSS on Ion Mobility Spectrometry, being Simple-to-use interactive self-modeling mixture analysis (SIMPLISMA) and Multivariate Curve Resolution (MCR) the most commonly used. The addition of a multi-capillary gas chromatography column using the ion mobility spectrometer as detector has been proposed in the past to increase chemical resolution. Short chromatography times lead to high levels of co-elution, and ion mobility spectra are key to resolve them. For the first time, BSS techniques are used to deconvolve samples of the gas chromatography - ion mobility spectrometry tandem. We propose a method to extract spectra and concentration profiles based on the application of MCR in a sliding window. Our results provide clear concentration profiles and pure spectra, resolving peaks that were not detected by the conventional use of MCR. The proposed technique could also be applied to other hyphenated instruments with similar strong co-elutions.

JTD Keywords: Blind Source Separation, Multivariate Curve Resolution, Ion Mobility Spectrometry, Gas Chromatography, Hyphenated instrumentation, SIMPLISMA, co-elution

Multivariate Curve Resolution (MCR) aims to blindly recover the concentration profile and the source spectra without any prior supervised calibration step. It is well known that imposing additional constraints like positiveness, closure and others may improve the quality of the solution. When a physico-chemical model of the process is known, this can be also introduced constraining even more the solution. In this paper, we apply MCR to Ion Mobility Spectra. Since instrumental models suggest that peaks are of Gaussian shape with a width depending on the instrument resolution, we introduce that each source is characterized by a linear superposition of Gaussian peaks of fixed spread. We also prove that this model is able to fit wider peaks departing from pure Gaussian shape. Instead of introducing a non-linear Gaussian peak fitting, we use a very dense model and rely on a least square solver with L1-norm regularization to obtain a sparse solution. This is accomplished via Least Absolute Shrinkage and Selection Operator (LASSO). Results provide nicely resolved concentration profiles and spectra improving the results of the basic MCR solution.

JTD Keywords: Blind source separation, Ion mobility spectrometry, Multivariate curve resolution, Sparse solution, Non negative matrix factorization

Miniaturization is a powerful trend for smart chemical instrumentation in a diversity of applications.. It is know that miniaturization in IMS leads to a degradation of the system characteristics. For the present work, we are interested in signal processing solutions to mitigate limitations introduced by limited drift tube length that basically involve a loss of chemical selectivity. While blind source separation techniques (BSS) are popular in other domains, their application for smart chemical instrumentation is limited. However, in some conditions, basically linearity, BSS may fully recover the concentration time evolution and the pure spectra with few underlying hypothesis. This is extremely helpful in conditions where non-expected chemical interferents may appear, or unwanted perturbations may pollute the spectra. SIMPLISMA has been advocated by Harrington et al. in several papers. However, more modem methods of BSS for bilinear decomposition with the restriction of positiveness have appeared in the last decade. In order to explore and compare the performances of those methods a series of experiments were performed.

JTD Keywords: Ion Mobility Spectrometry (IMS), Blind Source Separation (BSS), Multivariate Analysis, SIMPLISMA, MCR, Non-Negative Matrix Factorization (NMF)