Imma Ferrer and Michael Thurman discuss a novel method for the determination of contaminants in drinking water. They explain that by using direct aqueous injection by liquid chromatography/mass spectrometry/mass spectrometry sample preparation can be simplified and analysis time significantly reduced
Pesticides and other organic contaminants in drinking water pose potential human health risks. Agricultural and industrial uses of these chemicals are major sources of such contamination. To ensure the quality of drinking water in the US, the Environmental Protection Agency (EPA) has a number of monitoring requirements. EPA Method 538 has been developed and implemented for the determination of selected organic contaminants in drinking water, most of which are organophosphate pesticides.
EPA Method 538 involves analysis of water by direct aqueous injection and liquid chromatography/mass spectrometry/mass spectrometry (LC/MS/MS). The method measures the presence of 11 target analytes, using five deuterated internal standards. The analytes are separated and identified by comparing the acquired transition ions and retention times to calibration standards obtained under identical LC/MS/MS conditions. The concentration of each analyte is determined by internal standard calibration following standard procedures. Because the method requires no sample extraction, it is rapid and inexpensive relative to other LC/MS/MS methods.
This article describes a specific implementation of the EPA Method 538, using the Agilent 1290 Infinity LC System and an Agilent 6460 Triple Qudrupole LC/MS System using Jet Stream technology. The 10-minute ultra-high performance liquid chromatography (UHPLC) chromatographic analysis is more than twice as fast as the original EPA Method 538, saving time and solvent costs. The method was modified by adding a second transition for all analyte ions for confirmation, which satisfies the European Union (EU) specifications for unequivocal identification by mass spectrometry, specified requirements according to the EU drinking water directive (Council Directive 98/83/EC, 1998). This gives an even greater assurance of correct identification than prescribed by the EPA. The utility of the method was demonstrated using local water samples.
EPA Method 538
Method 538 requires a 40mL water sample preserved with sodium omadine and ammonium acetate. A 950µL aliquot is removed and placed in a vial, along with a 50µL aliquot of five deuterium-labelled internal standards. The organic solvent content of the sample should not exceed 5%. The samples are collected in baked amber glass bottles and stored at 4 °C until analysed. The water sample is passed through a polytetrafluoroethylene (PFTE) filter (0.2µm) before addition of internal standards, in order to prevent plugging of the analytical column. The sample is then ready for direct injection into the LC/MS/MS system. Blanks are also passed through the filter to check for interferences.
Table 1 shows the 10 organophosphate analytes included in EPA Method 538 along with the polynuclear aromatic heterocycle quinolone, which are important possible drinking water contaminants1. Five deuterated standards that are part of the method are also shown in Table 1. One advantage of the EPA method is that solid phase extraction (SPE) is no longer needed for sample preparation, which means that total analysis time is cut at least in half. In addition, suppression from the sample matrix is reduced because the matrix is not concentrated as may occur with SPE. Although concentration of the sample may enable lower detection limits, this advantage will be mitigated by suppression effects. In addition, the sensitivity of the instrument negates the need for concentration of the sample. The method is quite simple, requiring only the addition of the internal standard mixture to the water sample.
Limits of Detection and Linearity
The EPA Method 538 calls for one multiple reaction monitoring (MRM) transition per compound1. The adaptation of the method described includes a second transition in order to provide a confirmation ion for each detected compound. This change also conforms to standard analytical procedures that call for a second confirming transition for analysis by LC/MS/MS using triple quadrupole methods, as well as ion-ratio percentages.
The extracted ion chromatogram (EIC) for the 11 compounds of EPA Method 538 is shown in Figure 1, using a 10-minute rapid gradient with UHPLC. The 11 compounds elute in approximately six minutes. The more polar compounds, such as methamidophos, acephate, and aldicarb sulphoxide, elute in the first minute of the chromatogram. The more hydrophobic compounds, such as diisopropyl methylphosphonate (DIMP), aldicarb, fenamiphos sulphoxide and sulphone, along with thiofanox, elute at the end of the chromatographic run. Good peak shape, which improves sensitivity and increases the limit of detection, was accomplished with this gradient.
The limits of detection (LODs) for the 11 analytes varied from 1ng/L for aldicarb sulphoxide, which was the most sensitive compound, to 500ng/L for acephate, which was the least sensitive compound (Table 2). The wide variation in LODs reflects the ability of each analyte to form ions in electrospray. The most polar analytes such as acephate and methamidophos were the least sensitive, while fenamiphos sulphone and thiofanox were some of the most sensitive compounds and also the most hydrophobic. Quinoline in particular is much more sensitive using the Agilent 6460 Triple Quadrupole LC/MS System with Jet Stream Technology because it is a stable compound with a nitrogen heteroatom. Thus, this adaptation of Method 538 meets the criteria for a sensitive method for organophosphate pesticides in drinking water. The extra MRM transition used in this adaptation of Method 538 is an important component of a valid method for water quality analysis of pesticides in water samples. The EU specifications for unequivocal identification by mass spectrometry require two transitions, and this procedure has become an unofficial standard worldwide.
Excellent linearity was achieved with the direct injection method for all of the analytes and the R2 values were =0.9999 for all compounds in this method.
Testing of Drinking Water Samples
EPA Method 538 was tested on several samples from river, reservoir and drinking water sources. The method detected only DIMP in one of the samples taken from a reservoir prior to drinking water treatment (Figure 2). The treated drinking water contained no detectable organophosphate pesticides. Thus, the method was found to be reliable and useful for the analysis of drinking water contaminants, as well as rapid and sensitive.
Running EPA Method 538 on the Agilent 1290 Infinity LC System and the Agilent 6460 Triple Quadrupole LC/MS System with Jet Stream Technology shortens time-to results by almost a factor of three, and increases reliability of the method by adding a second transition. In addition, the detection limits and adaptations conform to the requirements of this method1.
1. Shoemaker JA. 2009. EPA Method 538: Determination of selected organic constituents in drinking water by direct aqueous injection-liquid chromatography/tandem mass spectrometry. EPA/600/R-09/149, 40p http://www.epa.gov/microbes/Method%20538_Final.pdf
2. Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31998L0083:EN:NOT
3. Ferrer I and Thurman M. 2012. Agilent Technologies, publication number: 5990-9670EN. EPA Method 538: Determination of Selected Organic Contaminants in Drinking Water by Direct Aqueous Injection with the Agilent 6460 Triple Quadrupole LC/MS System . http://www.chem.agilent.com/Library/applications/5990-9670EN.pdf