Estimation of foliar glucose content of areca palm by a smartphone app and Fourier transform infrared spectroscopy based multivariate modeling

Abstract

Carbohydrates are essential molecules in the metabolism of plant systems whose quantification is crucial. The study aims to estimate foliar glucose content using the Smartphone-based Color Grab app by color change upon reaction with a 3,5-dinitrosalicylic acid reagent and mid-infrared spectra. The hue showed a negative correlation of − 0.959 with glucose content with sensitivity, detection limit and precision of 13.46 μg/mL,μg/mL,0.035 μg/mL, and 0.229% respectively. The glucose concentration to color coordinates displayed a linear response between 50 to 600 µg/mL. The linear regression equation with hue of standards was used to predict spectrophotometrically measured glucose concentration of leaf extracts with R² = 0.934 and sensitivity of 13.46 μg/mL. Multivariate analysis of infrared spectrum (650–4000 cm^‐1−1) of powdered arecanut leaves indicated elastic net and partial least square regression as the best models with R² of 0.99. The study has practical implications in smartphone or infrared spectra-based glucose measurements for low glucose (< 1 mg/mL) samples.