Polybrominated diphenyl ethers (PBDEs) levels in environmental media have increased over the last 20-25 years in the world. In aquatic environments PBDEs were found to be accumulated along food chain and Endocrine disruptors toxicity. In this study PBDEs were investigated in sediment and fish tissues from Lake Chaohu in central eastern China. There were 10 PBDEs congeners detected out of all 41 PBDEs. BDE-47 was of the highest with 5.17 ng/g in sediment and 58.47 ng/g in fish. PBDEs were evenly distributed across the surface sediment in the whole lake. It implied that the main source of PBDEs may not be an inflow river like Nanfei. Tissue distribution patterns of PBDEs in four fish species were in the order of BDE-47 > BDE-99 > BDE-100 > BDE-66 > BDE-138 > BDE-183 > BDE-154 > BDE-153. Octa- and deca-BDEs were below the detection limit. The concentrations of all PBDE congeners were higher in gills, livers, and kidneys than those in muscles and adipose tissue. Furthermore, PBDEs in different tissues had some different distribution patterns with fish size. Those discrepancies appeared to be correlated with the PBDEs pollution fluxes varying with the change of the year and their metabolism divergences in fish tissues.

Online quantitative analysis of reaction gases or exhaust in industrial production is of great significance to improve the production capacity and process.Anovel method is developed for the online quantitative analysis of reaction gases or exhaust using quantitative mathematical models combined with the linear regression algorithm of machine learning. After accurately estimating the component gases and their contents in the reaction gases or exhaust, a ratio matrix is constructed to separate the relevant overlapping peaks. The ratio and calibration standard gases are detected, filtered, normalized, and linearly regressed with an online process mass spectrometer to correct the ratio matrices and obtain the relative sensitivity matrices. A quantitative mathematical model can be established to obtain the content of each component of the reaction gases or exhaust in real time. The maximum quantification error and relative standard deviation of the method are within 0.3% and 1%, after online quantification of the representative yeast fermenter tail gas.