Impact of coffee-cherry fermentation time on physicochemical, spectral and sensory characteristics of Arabica coffee beans in two agroecological zones of Huila, Colombia

This study investigated how coffee-cherry holding (pre-pulping in-fruit fermentation) time (0–96 h) modulates the chemical composition (caffeine, trigonelline, chlorogenic acids quantified by HPLC) and the infrared spectral fingerprints (FTIR/FTNIR) of Arabica coffee produced in two contrasting agroecological zones (ZAE 1 and ZAE 5) in Huila, Colombia. Physical quality indicators (defects, yield-related parameters) and sensory scores (SCA protocol) were assessed as complementary evidence to contextualize the chemical–spectral changes and their processing implications This study evaluated the impact of in-fruit fermentation time (0–96 h) on the physical, chemical, spectral, and sensory quality of Arabica coffee in two agroecological zones (ZAE 1 and ZAE 5) of Huila, Colombia. A randomized block design was used, testing five fermentation durations across the two zones. Physical quality parameters (defect rate, yield), sensory attributes (SCA protocol), chemical composition (caffeine, trigonelline, chlorogenic acids via HPLC), and spectral profiles (FTIR/FTNIR) were assessed. Results showed that extended fermentation times (≥72 h) increased physical defects and decreased the proportion of sound beans. Caffeine and trigonelline levels remained stable, while chlorogenic acids varied significantly with fermentation time and agroecological zone. FTNIR effectively captured process kinetics (PC1=70%), and FTIR provided functional chemical specificity. Sensory scores ranged from 78 to 86, with ZAE 1 displaying more consistent profiles over time, whereas ZAE 5 showed greater fluctuation. The findings indicate that there is no universal optimal fermentation time; however, a 24–48 h window is considered a safe operational range, adjustable to local conditions and under basic process control (temperature, pH, °Brix). The study recommends the development of predictive multivariate models to adapt fermentation strategies based on microenvironmental conditions and improve postharvest decision-making. • Cherry holding (0–96 h) interacts with agroecological zone to affect quality. • Holding ≥72 h increases defects (pasilla) and reduces sound beans. • Chlorogenic acids vary with time and zone; caffeine and trigonelline stay stable. • FT-NIR orders process kinetics; FTIR fingerprint bands add functional specificity. • A practical 24–48 h window preserves sensory scores while limiting physical losses.