A validated near-infrared spectroscopic method for methanol detection in biodiesel. (25th April 2016)

• Record Type: Journal Article
• Authors: Paul, Andrea; Bräuer, Bastian; Nieuwenkamp, Gerard; Ent, Hugo; Bremser, Wolfram
• Is Part Of: Measurement science & technology. Volume 27:Number 6(2016:Jun.)
• Publication Date: 2016-04-25
• Subjects: FAME -- headspace GC -- online methods -- PLS -- SVM -- EMRP -- NIRS; Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/0957-0233/27/6/065002
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: Biodiesel quality control is a relevant issue as biodiesel properties influence diesel engine performance and integrity. Within the European metrology research program (EMRP) ENG09 project 'Metrology for Biofuels', an on-line/at-site suitable near-infrared spectroscopy (NIRS) method has been developed in parallel with an improved EN14110 headspace gas chromatography (GC) analysis method for methanol in biodiesel. Both methods have been optimized for a methanol content of 0.2 mass% as this represents the maximum limit of methanol content in FAME according to EN 14214:2009. The NIRS method is based on a mobile NIR spectrometer equipped with a fiber-optic coupled probe. Due to the high volatility of methanol, a tailored air-tight adaptor was constructed to prevent methanol evaporation during measurement. The methanol content of biodiesel was determined from evaluation of NIRS spectra by partial least squares regression (PLS). Both GC analysis and NIRS exhibited a significant dependence on biodiesel feedstock. The NIRS method is applicable to a content range of 0.1% (m/m) to 0.4% (m/m) of methanol with uncertainties at around 6% relative for the different feedstocks. A direct comparison of headspace GC and NIRS for samples of FAMEs yielded that the results of both methods are fully compatible within their stated uncertainties.