Analysis of Musty Odorous Compounds in Drinking Water by SPME GC-TOFMS

Significance of the Topic

Complaints about taste and odor in drinking water are a common operational and public-relations issue for water suppliers. Odorous compounds such as haloanisoles have extremely low odor threshold concentrations (sub-ng/L), so analytical methods must combine high sensitivity with reliable identification to support complaint investigation, treatment optimization and regulatory compliance.

Objectives and Study Overview

This application note evaluated an automated SPME–GC–TOFMS workflow for the trace analysis of thirteen musty/earthy odorous organic compounds (haloanisoles and related species) in drinking water. The aims were to achieve quantification at sub-nanogram-per-liter levels, demonstrate robust identification at these low concentrations, and assess calibration performance and accuracy versus certified standards.

Methodology

Key methodological points:

• Sample preparation: 15 mL water aliquots spiked with an internal standard (2,4,6-TCA-d5). Incubation at 40 °C for 1 min and SPME extraction for 30 min using an automated L-PAL3 autosampler with agitation and a conditioning station.
• SPME fiber: 3-phase DVB/C-WR/PDMS fiber, 80 µm (Restek), selected to improve extraction of more polar haloanisoles relative to 100% PDMS fibers.
• Thermal desorption: 3 min desorption at 250 °C in splitless mode.
• GC conditions: Agilent 8890, Rxi-5Sil MS column (30 m × 0.25 mm × 0.25 µm), He carrier at 1.4 mL/min, oven program 50 °C (2 min) → 200 °C at 10 °C/min → 300 °C at 25 °C/min (2 min), transfer line 280 °C.
• MS detection: LECO Pegasus BTX time-of-flight mass spectrometer, ion source 250 °C, mass range 40–500 m/z, acquisition at 10 spectra/s.
• Standards and calibration: working standards prepared in methanol; calibration range typically 0.067–13.333 ng/L with an internal standard (2,4,6-TCA-d5). Lowest calibration levels around 0.067 ng/L for most analytes.

Used Instrumentation

Instrumentation used in the study:

• LECO Pegasus BTX GC-TOFMS system
• Agilent 8890 gas chromatograph
• LECO L-PAL3 autosampler with SPME conditioning station and agitator
• Rxi-5Sil MS GC column (30 m × 0.25 mm × 0.25 µm)
• SPME fiber: Restek DVB/C-WR/PDMS 80 µm

Results and Discussion

Analytical performance and key findings:

• Sensitivity: Most target haloanisoles achieved limits of quantification (LOQ) of 0.067 ng/L (67 pg/L). Exceptions: 2,3,4,6-tetrachloroanisole LOQ = 1 ng/L and pentachloroanisole LOQ = 0.5 ng/L.
• Linearity: Calibration curves were linear across the reported ranges, with R² values >0.99 for nearly all compounds, demonstrating reliable quantification down to the low-ng/L and sub-ng/L domain.
• Identification: Deconvolution (ChromaTOF) produced cleaner mass spectra at trace levels, improving library match scores (example: 2,4,6-TCA similarity improved from ~752 raw to ~893 after deconvolution). Deconvolution plus high spectral acquisition rate enabled extraction of pure component spectra from complex backgrounds.
• Mass accuracy and selectivity: Reported mass accuracy of 50–100 pm was used to reduce background interference around quantifier ions and improve signal-to-noise at trace levels (illustrated for 2,4,6-tribromoanisole at 0.1 ng/L).
• Chromatographic separation: The chosen GC program and column provided baseline separation of the target mixture; use of the 3-phase SPME fiber enhanced responses for relatively polar haloanisoles.
• Accuracy: Calculated concentrations across the calibration range deviated from certified values by well under 20% for all compounds, meeting the acceptance criteria of the referenced French validation protocol.

Benefits and Practical Applications of the Method

Practical advantages and applications include:

• Capability to detect and quantify odorous haloanisoles at or below their sensory odor thresholds, supporting complaint investigations and sensory risk assessment.
• High confidence in compound identification through full-spectrum TOFMS acquisition and advanced deconvolution, reducing false positives from matrix interferences.
• Automation of SPME steps (conditioning, extraction, desorption) improves reproducibility and throughput for laboratory workflows.
• Applicability to water utilities, environmental monitoring laboratories, and research on formation pathways (e.g., methylation of halophenols during disinfection or in distribution systems).

Future Trends and Potential Uses

Relevant directions and developments to extend the method's impact:

• Further improvements in fiber coatings and passive sampling strategies to increase sensitivity and time-weighted average monitoring in distribution systems.
• Integration with high-resolution MS and non-target screening workflows to detect emerging odorous contaminants and unknown haloanisole derivatives.
• Higher automation and multiplexing for routine monitoring programs, including on-line SPME interfaces and rapid sample throughput.
• Application of chemometrics and machine learning to deconvolution, spectral matching and pattern recognition for faster source attribution and trend analysis.

Conclusion

The combined SPME–GC–TOFMS approach on the LECO Pegasus BTX platform delivers the sensitivity, selectivity and spectral quality required for reliable identification and quantification of trace haloanisoles in drinking water. LOQs near 0.067 ng/L for most targets, robust linearity and satisfactory accuracy demonstrate that the method is suitable for investigating taste-and-odor complaints and for routine monitoring where sub-ng/L detection is needed.

References

1. Mallevialle J., Suffet I. H. Identification and treatment of tastes and odors in drinking water - Chapter 5: Classification of tastes and odors. AWWA Research Foundation, Denver, 1987, p. 291.
2. Jensen J. Chlorophenols in the terrestrial environment. Rev. Environ. Contam. Toxicol., 1996, (146), p. 25-51.
3. Bruchet A. Solved and unsolved cases of taste and odor episodes in the files of Inspector Cluzeau. Water Sci. Technol., 1999, (40), n°6, p. 15-21.
4. Malleret L., Bruchet A. A taste and odor episode caused by 2,4,6-tribromoanisole. J. Am. Water Works Assoc., 2002, (94), n°7, p. 84-95.
5. Nyström A., Grimvall A., Krantz-Rülcker C., Sävenhed R., Akerstrand K. Drinking water off-flavour caused by 2,4,6-trichloroanisole. Water Sci. Technol., 1992, (25), n°2, p. 241-249.
6. Montiel A., Ouvrard J., Rigal S., Bousquet G. Étude de l'origine et du mécanisme de formation de composés sapides responsables du goût de moisi dans les eaux distribuées. Tech. Sci. Méthodes, génie urbain, génie rural, 1987, (82), n°2, p. 73-83.
7. Corbi E., Benanou D., Cantet J., Tabet J. C. Identification of new odorous compounds in Swedish water: mixed haloanisoles and phenolic precursors. Water Sci. Technol., 2007, (55), n°5, p. 185-193.
8. Anselme C., N'Guyen K., Bruchet A., Mallevialle J. Can polyethylene pipes impart odors in drinking water? Environ. Tech. Letters, 1985, (6), p. 477-488.
9. Roche P., Tondelier C., Benanou D. Influence of NOM chlorination on halophenols: appearance and control on a water treatment plant. TECHNEAU EU Project, 2009, final report, p. 303-325.
10. Benanou D., Acobas F., de Roubin M. R., David F., Sandra P. Analysis of off-flavors in the aquatic environment by stir bar sorptive extraction-thermal desorption-capillary GC/MS/olfactometry. Anal. Bioanal. Chem., 2003, (376), n°1, p. 69-77.
11. Benanou D., Acobas F., de Roubin M. R. Optimization of stir bar sorptive extraction applied to the determination of odorous compounds in drinking water. Water Sci. Technol., 2004, (49), n°9, p. 161-170.
12. AFNOR. NF 90-210 - Water quality - Protocol for the evaluation of a quantitative physicochemical analysis method compared to a reference method. 1999.