Assessing near-infrared reflectance spectroscopy for the rapid detection of lipid and biomass in microalgae cultures

Abstract

With intensification of interest in microalgae as a source of biomass for biofuel production, rapid methods are needed for lipid screening of cultures. In this study, near-infrared reflectance spectroscopy (NIRS) was assessed as a method for analysing lipid (specifically, total fatty acid methyl esters (FAME) obtainable from processing) and biomass in late logarithmic and stationary phase cultures of the green alga Kirchneriella sp. and the eustigmatophyte Nannochloropsis sp. Culture samples were filtered, scanned by NIRS and chemically analysed; by combining these sets of information, models were developed to predict total biomass, FAME content and FAME as a percentage of dry weight in samples. Chemically derived (actual) and NIRS-predicted data were compared using the coefficient of determination (R ²) and the ratio of the standard deviation (SD) of actual data to the SD of NIRS prediction (RPD). For Kirchneriella sp. samples, models gave excellent prediction (R ² ≥ 0.96; RPD ≥ 4.8) for all parameters. For Nannochloropsis sp., the model metrics were less favourable (R ² = 0.84–0.94; RPD = 2.5–4.2), though sufficient to provide estimations that could be useful for screening purposes. This technique may require further validation and comparison with other species, but this study shows the potential of the NIRS as a rapid screening method (e.g. up to 200 sample analyses per day) for estimating FAME or other microalgal constituents and encourages further investigation.