Microalgal fatty acid composition: Rapid assessment using near-infrared spectroscopy

The feasibility of assessing microalgal fatty acid composition using near-infrared spectroscopy (NIRS) is described. The chlamydomonad microalga, Rhopalosolen saccatus (previously known as Characium saccatum), was isolated from the Fitzroy River, Central Queensland, Australia. R. saccatus was grown in batch culture with varying phosphorus nutrition and assessed for dry matter, total lipid and fatty acid composition using gas chromatography (GC). Transmission spectra (1100–2500 nm) were acquired of liquid culture, and reflectance spectra were acquired of wet and dry filtrates of cultures and of methyl esters. Partial least square (PLS) regression models were built on biomass, total lipid and a number of fatty acids. All sample presentation models supported PLS regression model with a cross validation correlation coefficient (Rcv) >0.87 for biomass and Rcv >0.68 for total lipid; however, the use of dry filtrates of culture is recommended as the sample presentation mode of choice. Models for fatty acids based on culture transmission spectra, reflectance spectra of wet and dry culture filtrates, or reflectance spectra of methyl esters in solvent were not acceptable. Dry extracts of methyl esters supported adequate models for fatty acids from C8:0 to C22:0, with the exception of capric and behenic acids, with an Rcv of 0.89–0.94; however, in practice, samples processed to this stage can be easily analyzed by GC. Near-infrared spectroscopy can be a potential choice for rapid estimation of biomass (dry matter) and lipid content and composition in microalgae, with further work required to demonstrate oping robustness of the calibration model in prediction of unknown samples. © 2015, Springer Science+Business Media Dordrecht.