期刊
PROCESSES
卷 10, 期 12, 页码 -出版社
MDPI
DOI: 10.3390/pr10122614
关键词
UV-Vis spectrometry; partial least squares; antibiotics; online monitoring; hospital wastewater
Real-time monitoring of antibiotics in hospital and pharmaceutical wastewater using UV-Vis spectroscopy combined with chemometric methods is considered a promising method. This study explored the relationship between absorption spectra and antibiotics and investigated the influence of optical path length on the limit of detection (LOD). The study also investigated multiple antibiotics in wastewater and selected wavelengths with better predictability for each antibiotic using chemometric methods.
Real-time monitoring of antibiotics in hospital and pharmaceutical wastewater using ultraviolet-visible (UV-Vis) spectroscopy is considered a promising method. Although gas chromatography-mass spectrometry (GC-MS) and other methods can detect antibiotics with quite low limits of detection (LOD), they possess various limitations. UV-Vis spectroscopy combined with chemometric methods is a promising choice for monitoring antibiotics. In this study, two immersed in situ UV-Vis sensors were used to explore the relationship between absorption spectra and antibiotics and study the influence of the optical path length on the LOD. The LODs of sensor 2 using a 10 cm optical path is up to 300 times lower than that of sensor 1 using a 0.5 mm optical path. Moreover, multiple antibiotics in the wastewater were investigated in real-time manner. The absorption spectra of 70 groups of wastewater samples containing different concentrations of tetracycline, ofloxacin, and chloramphenicol were measured. The results indicate that the nine wavelengths selected by interval partial least squares (iPLS) after the second derivative pretreatment have better predictability for ofloxacin and the six wavelengths selected by competitive adaptive reweighted sampling (CARS) after the first derivative. The multi-fold cross-validation results indicate that the model has a good predictive ability.
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