The use of GC/MS to Determine Water Quality

Sue Godwin1, Rex Lewis1 and Scott J Campbell2

1Bristol Water, PO Box 218, Bridgewater Road, Bristol, UK; 2SpectralWorks Ltd, The Heath Business & Technical Park, Runcorn, UK

First Published: BMSS 2005

Introduction

The determination of the quality of water is an important public health and environmental issue; therefore it is important that the analysis be conducted in a reliable and reproducible manner. Traditionally samples are analysed against a list of target compounds. This approach means that unexpected compounds will be missed. It is therefore common place to re-analyse the data files manually to determine missed compounds.

In this paper we examine the use of a novel software application which by using its data processing algorithms can carry out both a qualitative and quantitative analysis. This double analysis means that all the compounds within a data file can be found in a reliable non biased manner. The spectra that are produced by the software are automatically refined and the background is subtracted, thus enabling reliable and accurate identification of non target compounds.

Figure 1. TIC of Water Sample

Figure 2. Found Compounds

Figure 3. Phenol XICs

Figure 4. Phenol

Figure 5. NIST Result for Phenol

Figure 6. Subsection of TIC

Figure 7. Unknown Compound

Figure 8. Unknown Spectrum

Figure 9. NIST Result for Unknown

Figure 10. 2nd Unknown Spectrum

Figure 11. NIST Result for 2nd Unknown

Results and Discussion

Figure 1 shows the total ion chromatogram (TIC) from the analysis of a water sample contaminated with phenols. Table 1 shows a list of target compounds manually detected when the file was analysed using the native instrument software. These target compounds were then entered into AnalyserPro and the file was analysed.

In figure 2 we can see where the target compounds were found as shown by the green dots. Figure 3 shows the extracted ions for one of thoses target compounds, Phenol, its spectrum is shown in figure 4 and the NIST library search result is shown in figure 5. The match factors returned by NIST were over 900 for both the forward and the reverse searching factors. Figure 6 shows an expanded region of the chromatogram showing a further two compounds that were not in the original target compound list, compounds that were found but were not on the target compound list are shown in blue.

The compound displayed in figure 7 shows the extracted ion chromatograms of an unknown, the spectrum for this compound is shown in figure 8 and the NIST library search result is shown in figure 9. The spectrum in figure 10 is a second unidentified compound with its NIST library search result shown in figure 11. The 2 compounds were analogs of each other.

Conclusion

We have shown that the use of AnalyserPro enables the rapid analysis of data files for both targeted and qualitative analysis.