Pectin extraction from Helianthus annuus (sunflower) heads using RSM and ANN modelling by a genetic algorithm...

Area of Science:

• Agricultural Chemistry
• Biotechnology
• Food Science

Background:

• Pectin extraction from plant sources is crucial for food and pharmaceutical industries.
• Sunflower heads represent an underutilized agricultural byproduct with potential for pectin recovery.

Purpose of the Study:

• To optimize pectin extraction from sunflower heads using advanced modeling techniques.
• To compare the efficacy of Artificial Neural Network coupled with genetic algorithm (ANN-GA) against Response Surface Methodology (RSM).

Main Methods:

• Pectin extraction parameters (time, temperature, frequency, solid/liquid ratio) were optimized.
• Response Surface Methodology (RSM) and Artificial Neural Network coupled with genetic algorithm (ANN-GA) were employed for modeling and optimization.
• Fourier-transform infrared spectroscopy (FTIR) and 13C Nuclear Magnetic Resonance (NMR) were used for characterization.

Main Results:

• ANN-GA exhibited higher prediction efficiency than RSM for pectin yield.
• A maximum pectin yield of 29.1 ± 0.07% was predicted by ANN-GA under optimized conditions (10 min, 59.9°C, 30 Hz, S/L ratio of 1:29.9).
• Experimental validation closely matched the predicted yield (29.5 ± 0.7%).

Conclusions:

• ANN-GA is an effective method for optimizing pectin extraction process parameters from sunflower heads.
• The optimized conditions provide a high yield of pectin suitable for further applications.
• Characterization confirmed the identity of the extracted pectin.