Chlorinated coumarins, which are as cytotoxic as highly toxic halobenzoquinones toward CHO-K1 cells, have recently been identified as disinfection byproducts in drinking water disinfection processes. Therefore, detecting coumarins in water samples collected at various stages from drinking water treatment plants helps assess the formation of chlorinated coumarins in drinking water. Hence, a simple, rapid, accurate, and sensitive method for quantifying coumarins in water samples is required. In this study, a method was developed based on solid-phase extraction coupled with high performance liquid chromatography-triple-quadrupole mass spectrometry for analyzing seven coumarins in water samples from drinking water treatment plants, including 6,7-dihydroxycoumarin, 7-hydroxycoumarin, 6-hydroxy-4-methylcoumarin, 8-chloro-7-hydroxycoumarin, coumarin, 7-chloro-6-hydroxy-4-methylcoumarin, and 3,8-dichloro-7-hydroxycoumarin. Sample pretreatment involved solid-phase extraction using HLB columns, followed by elution with water and methanol, each containing 0.25% formic acid. The extracted solution was separated via gradient elution using a Phenomenex Luna C18 column (100 mm×2.0 mm, 3 mm) with 0.1% formic acid aqueous solution and methanol as the mobile phases, with analytes detected by triple-quadrupole mass spectrometry equipped with electrospray ionization source in multiple reaction monitoring mode. The matrix effect, precision, and accuracy of the developed method were investigated using raw and treated water as matrices. Matrix effects of 0.84-1.12 were recorded for the detection of 6,7-dihydroxycoumarin, 7-hydroxycoumarin, 6-hydroxy-4-methylcoumarin, and coumarin in raw water, while values of 0.67-0.70 were recorded for 8-chloro-7-hydroxycoumarin, 7-chloro-6-hydroxy-4-methylcoumarin, and 3,8-dichloro-7-hydroxycoumarin in finished water. The three chlorinated coumarins exhibited matrix effects above 0.80 after the finished water matrix had been diluted four times. These results suggest that only simple solid-phase extraction or sample dilution is required to accurately determine the seven coumarins in drinking water from treatment plants. In addition, these coumarins exhibited good linear relationships at their respective mass concentrations. The precision and accuracy of the method were evaluated using raw and treated water as matrices. The seven coumarins exhibited good linearities by triple-quadrupole mass spectrometry in a certain range, with correlation coefficients (r) greater than 0.99 and method detection limits (MDLs) of 0.67-1.12 ng/L. The seven coumarins exhibited recoveries of 61.4%-91.5% at three spiked levels (20, 50, 100 ng/L) with relative standard deviations (RSDs, n=6)≤11.2%. The developed method can be used to analyze water samples from various treatment stages of a drinking-water treatment plant. 7-Hydroxycoumarin, 6,7-dihydroxycoumarin, and coumarin were detected at levels of 0.21-27.9 ng/L in 100% of the samples, while 6-hydroxy-4-methylcoumarin was not detected in raw water, post-coagulated water, and post-carbon-filtered water, but was found in sand-filtered water and finished water, with higher levels recorded for the latter (4.69 ng/L) than the former (1.79 ng/L). 8-Chloro-7-hydroxycoumarin was only detected in treated water (0.07 ng/L). This method is highly precise and accurate, provides results in short analysis times, and can be used to effectively monitor coumarins in real water samples and assess their removal during drinking-water treatment.
Keywords: coumarins; disinfection byproducts; drinking water; high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS); solid-phase extraction (SPE).