期刊
JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY
卷 61, 期 50, 页码 12307-12314出版社
AMER CHEMICAL SOC
DOI: 10.1021/jf403086f
关键词
microalgae; biomass; lipids; infrared spectroscopy; multivariate calibration; chemometrics
资金
- U.S. Department of Energy [DE-AC36-08-GO28308]
- National Renewable Energy jointly as part of the BioEnergy Technology Office (BETO) [9.6.1.8]
- Sustainable Algal Biofuels Consortium project
- DOE [DE-EE0003372]
One of the challenges associated with microalgal biomass characterization and the comparison of microalgal strains and conversion processes is the rapid determination of the composition of algae. We have developed and applied a high-throughput screening technology based on near-infrared (NIR) spectroscopy for the rapid and accurate determination of algal biomass composition. We show that NIR spectroscopy can accurately predict the full composition using multivariate linear regression analysis of varying lipid, protein, and carbohydrate content of algal biomass samples from three strains. We also demonstrate a high quality of predictions of an independent validation set. A high-throughput 96-well configuration for spectroscopy gives equally good prediction relative to a ring-cup configuration, and thus, spectra can be obtained from as little as 10-20 mg of material. We found that lipids exhibit a dominant, distinct, and unique fingerprint in the NIR spectrum that allows for the use of single and multiple linear regression of respective wavelengths for the prediction of the biomass lipid content. This is not the case for carbohydrate and protein content, and thus, the use of multivariate statistical modeling approaches remains necessary.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据