Rapid Classification of Coffee Products by Data Mining Models from Direct Electrospray and Plasma-Based Mass Spectrometry Analyses
| Autor: | Roberto Gamboa-Becerra, Josaphat Miguel Montero-Vargas, Abigail Moreno-Pedraza, Sandra Martínez-Jarquín, Eligio Gálvez-Ponce, Robert Winkler |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Chemical process
Electrospray Chromatography Chemistry Electrospray mass spectrometry 010401 analytical chemistry Sample processing 04 agricultural and veterinary sciences Cost efficacy computer.software_genre Weight range Mass spectrometry 040401 food science 01 natural sciences Applied Microbiology and Biotechnology 0104 chemical sciences Analytical Chemistry 0404 agricultural biotechnology Yield (chemistry) Data mining Safety Risk Reliability and Quality Safety Research computer Food Science |
| Zdroj: | Food Analytical Methods. 10:1359-1368 |
| ISSN: | 1936-976X 1936-9751 |
| DOI: | 10.1007/s12161-016-0696-y |
| Popis: | In coffee manufacture, analytical methods with high-throughput and cost efficacy are required for process development and quality control. Thus, we investigated the applicability of direct mass spectrometry methods to distinguish coffee products according to species, geographic origin and processing. We tested the performance of the established method direct-injection electrospray mass spectrometry (DIESI-MS) and the emerging method low-temperature plasma ionization mass spectrometry (LTP-MS). Both methods are capable of classifying coffee products, but DIESI-MS and LTP-MS yield complementary information about the chemical composition of the samples. DIESI-MS shows a broad molecular weight range of compounds. In contrast, LTP-MS detects mainly low molecular weight compounds, which correspond to quality-related ingredients, such as caffeine and purines. LTP-MS displays a high potential for rapid quality control measurements and online monitoring, because no sample processing is required. Data mining methods support the discovery of ‘important’ compounds, which are responsible for the discrimination between sample groups, and reveal associated chemical processes. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink