Date of Publication
2024
Document Type
Bachelor's Thesis
Degree Name
Bachelor of Science in Biochemistry
Subject Categories
Biochemistry
College
College of Science
Department/Unit
Chemistry
Thesis Advisor
Emmanuel V. Garcia
Defense Panel Chair
Marissa G. Noel
Defense Panel Member
Joan Candice V. Ondevilla
Abstract/Summary
Coffee has been one of the most popularly consumed beverages and it is also among the most traded agricultural products in the world. The varietal species and the environment in which the coffee is grown affect the taste and quality of the coffee beans. Due to high demand and competition in the market, coffee has been reported to have high incidence of fraud and adulteration. This emphasizes the need for developing an objective method to trace the origin of coffee beans and also to authenticate the quality of local Philippine coffee. This study utilized Random Forest models that can classify the regional origin and the coffee variety of Philippine green coffee bean samples from regions, IV-A (QUE), VI (NOC), IX (ZDS), and CAR based on the multi-elemental fingerprints of the samples through portable X-ray fluorescence (p-XRF) and Energy-dispersive X-ray fluorescence (ED-XRF) analysis. Moreover, the Random Forest models constructed from p-XRF and ED-XRF analysis were compared in its capacity to classify the beans according to varietal species and regional origin. The coffee bean samples collected were oven-dried, grounded and then pulverized in preparation for p-XRF and ED-XRF analysis. The multi-elemental fingerprint of the coffee beans were obtained through p-XRF and ED-XRF spectroscopy. Random Forest models for regional origin and varietal classification of the green coffee beans were able to achieve 100% accuracy after refinement of models that exhibited errors in classification. Minor increase in accuracy was observed in the Random Forest models constructed from the ED-XRF multi-elemental data, however, the Random Forest models constructed from p-XRF were able to perform equally at regional and varietal classification after model refinement.
Abstract Format
html
Language
English
Format
Electronic
Keywords
Coffee—Varieties --Philippines; X-ray spectroscopy
Recommended Citation
Salviejo, G. C., & Cabial, R. L. (2024). Varietal and regional traceability of Philippine green coffee beans through multi-elemental analysis using portable and energy-dispersive x-ray fluorescence spectroscopy. Retrieved from https://animorepository.dlsu.edu.ph/etdb_chem/58
Upload Full Text
wf_yes
Embargo Period
4-18-2024
