Ultra-high-pressure liquid chromatography tandem mass spectrometry method for the determination of 9 organophosphate flame retardants in water samples

Graphical abstract

Few methods are available for comprehensive organophosphate flame retardants (PFRs) detection in water and wastewater. Gas chromatography has been employed previously, but this approach is less selective, not amenable for use with deuterated standards and can suffer unfavorable fragmentation. Ultra-high-pressure liquid chromatography tandem mass spectrometry (UHPLC-QqQ-MS/MS) has become the most promising platform, already applied successfully for analysis of selected PFRs in some environmental matrices like water and wastewater. However, the presence of some interferences from the dissolvent, the equipment and the used materials should be taken into account. The procedure involves: The first determination of PFRs by UHPLC-QqQ-MS/MS using a trap column to distinguish the interferences coming from the instrument and mobile phases. The optimization of the LC separation to distinguish all target compounds and their interferences. This method coupled to a solid-phase extraction (SPE) improve the detection and quantification of PFRs. Method details

Interference analysis
Ultra-high-pressure liquid chromatography tandem mass spectrometry (UHPLC-QqQ-MS/MS) is likely the most promising platform for the determination of PFRs in water and wastewater matrices [1][2][3][4][5]. However, the presence of some interferences from the dissolvent, the equipment and the laboratory material can disturb their analytical determination. We made, up to our knowledge, the first determination of PFRs using a trap column to distinguish these interferences. Furthermore, we prove two different solvents, acetonitrile (ACN) and methanol (MeOH), observing the presence of analyte's contamination in both cases. The system is applied to the determination of PFR in water.
Figs. 1 and 2 present chromatograms of a non-spiked river sample and the same sample spiked with the analytes.

Reagents and samples
(a) Compounds: the nine PFRs and the four deuterated PFRs used as internal standard (IS) ( Table 1) were purchased from LGC Standards (Germany).    (f) Flow rate: 0.2 mL/min and the sample volume injected was 5 mL.
(g) Analysis: performed in positive ionization mode.
(h) Data acquisition: carried out in selected reaction monitoring (SRM) to identify and quantify using two precursor-product ion transitions, retention times, and the ratio of intensities between the two product ions.
Fragmentor and collision energies were optimized for each compound individually ( Table 2).

Validation of the method
The validation of the instrumental parameters (Table 3) was performed by determining recoveries, limit of detection (LOD), limit of quantification (LOQ) and linearity over the range to obtain suitable R 2 . The quantification was performed using the internal standard method. The mixture of the internal standards was added to water to a concentration of 200 ng/L to get a final concentration in the injected extract of 50 ng/mL. The reported parameters were calculated from distilled, river and wastewater samples spiked with standard solutions prepared in methanol at the appropriate concentrations, in a way that the volume of organic solvent added were never higher than 250 mL. The samples were processed with plastic materials as much as possible to avoid adsorption to glass. There were not river or wastewater samples free of PFRs. Then, to perform these experiments several non-spiked samples were analyzed and the peak area of those compounds were subtracted to that found in the spiked samples.
The LOD was calculated as the mass of analyte required to produce a signal-to-noise (S/N) ratio of 3:1 and the LOQ of 10:1. S/N ratios were calculated using MassHunter Workstation Software (GL Sciences, Tokyo, Japan). LODs and LOQs were estimated from water samples spiked at 4 ng/L for river water and at 100 ng/L in wastewater. Values reported in Table 3 are LODs and LOQs in the injected extracts that would correspond to LODs ranging from 0.12 to 1 ng/L and LOQs from 1.2 to 10 ng/L in water samples. The results obtained for the LODs and LOQs show that the proposed method is sensitive enough for the determination of the PFRs in water samples. The precision studies were carried out from the evaluation of intra-day and inter-day variations of the areas ratio between the analyte and its IS using water extracts suitably stored. The intra-day precision of peak areas ratio (five replicates at 400 ng/L), expressed by means of the percentage of relative standard deviation (%RSD (n = 5)) were lower than 11.4% and the inter-day precision (five replicates) were less than 20%. Recoveries were calculated from samples spiked at 400 ng/L analyzed in quintuplicate.
The calibration of the LC-MS/MS was conducted using seven different concentrations (from the LOQ to 300 ng/mL) of PFR standard solutions in methanol with 50 ng/mL of each deuterated compound used as IS. The use of IS prevent the matrix effects The robustness of the method was clearly ascertained during the optimization procedure, by establishing the consequences of the deliberate introduction of minor reasonable variations (mostly different water volume analyzed, sample flow-rates during the extraction step, cartridge drying times In italics, the compounds used as IS. and eluent composition) and by the similar results obtained checking different types of water. The results obtained from the variations of AE20% in the different parameters were not significantly different from those achieved by the validated method. These results proved that the proposed method was robust.
Each 10 samples, one instrumental and one procedural blank were analyzed to serve as quality control. Samples analyzed shown clearly two peaks, the first one corresponding to the TClPP found in the river water and the second to the instrumental background, indicating background contamination from the injection system and tubing of the LC-MS/MS can be successfully separated from the sample contamination. Table 4 shows the performance of the system in water samples (river, influent and effluent wastewater).