Gas chromatography-mass spectrometry determination of bioactive compounds and mineral composition of optimised oven dried Eichhornia Crassipes using Artificial Neural Network and Response surface method

In this study, the oven drying process of Eichhornia crassipes ( E. crassipes ) was optimized and modelled using Response surface method (RSM) and Artificial Neural Network (ANN) and the influence of the drying conditions on its mineral and bioactive compounds composition was assessed. The optimization was achieved using Rotatable Central Composite Design using RSM and the optimized parameters were temperature (40-70°C), time (80-180 min) and air velocity (1-3.5m/s) with the response as moisture content (MC) (%). The Atomic Absorption Spectrophotometer (AAS) was used to determine the mineral composition and Gas Chromatography–Mass Spectrometry (GC-MS) analysis for the bioactive compounds. The optimal predicted conditions for the drying process were temperature (51°C), air velocity (3.4m/s) and time (146 min) with a corresponding MC of 49.3%. The mineral composition of E. crassipes roots and leaves was Fe at 557.28 ± 7.92mg/l; 21.84 ± 2.084mg/l, Na 14.08 ± 2.12mg/l; 16.27 ± 1.180mg/l, Zn at 0.17 ± 0.010 mg/l; 0.094 ± 0.007mg/l, Ca at 31.11 ± 6.040mg/l 58.94 ± 5.033mg/l and K at 162 ± 5.540mg/l; 373.56 ± 9.04m/l respectively. The GC-MS results revealed piperazine, octadecanoic acid, 2,3-dihydroxypropyl ester, beta-sitosterol and n-Hexadecanoic acid as the main compounds. The optimal conditions of E. crassipes oven drying were successfully identified by RSM while holding the favourable minerals and bioactive compounds. The ANN model had a better predictive performance with R 2 at 0.9897 then the RSM model. Future research is recommended focusing on investigating the influence of other drying techniques on E. crassipes and optimizing their drying conditions.