Ability of NIR Spectroscopy to Determine the Biomethane Potential from Organic Wastes and Agricultural Products - BIONIR

Coordinating Institution: CRP Gabriel Lippmann
Contracting Partner(s): Administration des Services Techniques de l'Agriculture (ASTA) , Centre Wallon de Recherches Agronomiques (B) , Centre Indépendant de Promotion Fourragère (B)
From: 01/05/2009
To: 30/04/2012
Budget: 350,000.00€
Contact(s): Delfosse Philippe

Progress Summary 2009

Biomethanation plans commonly use agricultural commodities and animal effluents to feed anaerobic digesters and produce biogas consisting mainly in methane and carbon dioxide. Farmers involved in biomethanation get organic substrates analysed in order to reduce ration costs, maximise methane production and profit.

The true objective of a substrate analysis is therefore to predict the biomethane potential (BMP) and indirectly the digester performance. The determination of BMP is tedious and typically requires a digestion test elapsing over a minimum of 30 days. This long period delays drastically the decision of the farmer to incorporate a newly proposed substrate and therefore, the farmer can miss a lucrative opportunity.

Considering the wide and successful use of Near Infrared Spectroscopy, a non-destructive and fast analytical method, in analysing forages and food, the BIONIR project has the objective (1) to assess its ability to predict the BMP, the digestion kinetics, and the H2S emission for agricultural commodities and residues, and (2) to evaluate various energy crops, with an emphasis on elucidating which part of the maize plant (corn versus stem and leaves), is more adequate for biomethanation. To achieve these objectives, a partnership has been set up that involves various specific expertises; (1) the CRP-GL for BMP determination and the substrate composition, (2) the CRA-W for the NIRS calibration and evaluation, (3) the ASTA and CIPF for providing agricultural substrates. Four maize varieties showing a range of precocity, some sorghum and sunflower varieties were cropped in Belgium (CIPF) and Luxembourg (ASTA) during the years 2008 and 2009.

The yield was determined and in the case of maize 3 dates of harvest were evaluated. The harvests were preserved as silage. All the ASTA samples and about half of the CIPF samples were analysed for their content in dry matter and organic dry matter. The infrared spectra were recorded for all the 2009 samples produced by the ASTA. A preliminary calibration realized on a limited number of maize, sorghum and sunflower samples provided promising results if the samples are grouped according to the type of organic matrix analysed (maize: R2=0.95, sorghum: R2=0.99, sunflower: R2=0.98). When all the samples are include in a unique calibration, the correlation is less attractive (R2=0.83).

The results indicate that specific calibrations will have to be elaborated for each organic matrix and that it is highly improbable that a universal model will suffice to predict BMP from any type of organic substrate.

Figure: Tropical sorghum accessions (front) and locally adapted maize varieties (back) grown in Rédange, Grand-Duchy of Luxembourg, 2007.