Kromatografi Lapis Tipis Menggunakan Perangkat Interpretasi ImageJ untuk Penentuan Kafein dalam Biji Kopi Robusta

Abstract

This research aimed to evaluate the use of ImageJ software as an alternative method for interpreting Thin Layer Chromatography (TLC) chromatograms. This TLC was used to analyse caffeine from robusta coffee. The experimental procedure involved caffeine extraction using a combination of ultrasonic-assisted extraction and liquid–liquid extraction with chloroform, and spotting the extract onto silica gel F254 TCL plate. Mobile phase optimization yielded the best separation at an ethyl acetate : methanol ratio of 5:1, with an average Rf value of 0.74. Quantitative analysis was conducted by photographing TLC plates using a smartphone camera Realme 5i, followed by image processing in ImageJ by converting intensity pixel into optical density. Method validation was carried out according to ICH guidelines, including tests of linearity, LOD, LOQ, precision, and accuracy. The results demonstrated excellent linearity (R² = 0.9979), limit of detection (63,746 ppm), and limit of quantification (193,171 ppm), precision with %RSD of 3,103% (200 ppm), 1,940% (800 ppm), and 1,439% (1200 ppm), and accuracy within 91–103%. The caffeine content of robusta coffee samples was determined to be 4,101% ± 0,030 w/w. These findings confirm that ImageJ provides a feasible, simple, and cost-effective alternative method for quantitative TLC analysis of caffeine.