Evaluation of the prediction of biogas production from maize silages with Near Infrared Spectroscopy (NIRS)

Abstract The biogas potential of the input substrate used in biomethanation plants is an important parameter that strongly influences the choice of biogas stakeholders when establishing feeding rations of digester. But a major limitation to make use of this parameter is the time required to perform the analysis based on bioassays. In this paper, Near InfraRed Spectroscopy (NIRS) was investigated to predict the biogas potential and the degradation kinectics of various maize silages. Very encouraging results were obtained for the calibration of the biogas potential (R 2 =0.97), whereas the NIRS prediction was less accurate for kinetic parameters. In this context, NIRS appears to be a promising analytical technique to predict biogas potential and to guide biogas stakeholders in their quest for valuable substrates to be converted into methane through the anaerobic digestion process. Introduction Acquiring information about the biomethane potential (BMP) of the input substrates is particularly recommended before their use in an anaerobic digester. Unfortunately, the BMP determination is very time-consuming and requires dedicated laboratory equipments. The most common way to get BMP is to incubate the substrate in a lab-scale anaerobic digester with adequate bacterial inoculum and to monitor over time the degradation, with measurements of the produced biogas and analysis of its composition, mainly methane and carbon dioxide (Angelidaki et al. , 2009). This procedure requires up to 45 days to reach the complete biodegradation. In this way, alternative methods to determine BMP have been investigated. Near InfraRed Spectroscopy (NIRS) is one of the most promising among them. NIRS is a fast, non-destructive and inexpensive analytical technique (Burns and Ciurczak). The figure of the absorption at each NIR wavelength (typically 1000-2500nm) constitutes the absorption spectra of the sample. Substrates made of complex organic matter (carbohydrates, fats and proteins among main compounds) have their own spectrum allowing the quantification of the biochemical composition. Consequently, NIRS appears fully