High Throughput Multi-Residue Screening of Drinking Water using the SPE-DEX and Pegasus GC-TOFMS

High-throughput multi-residue screening of drinking water using automated SPE-DEX and Pegasus GC-TOFMS

Importance of the topic

The rapid and reliable screening of organic contaminants in drinking water is critical for public health, regulatory compliance and operational decision-making in water utilities. Traditional workflows based on liquid–liquid extraction (LLE) followed by GC-MS are labor intensive, use large solvent volumes, and require long sample preparation and analysis times. Automated solid-phase extraction (SPE) combined with solvent drying by a hydrophobic membrane and high-speed GC time-of-flight mass spectrometry (GC-TOFMS) can substantially reduce turnaround time and increase sample throughput while enabling broad-spectrum, multi-residue screening.

Objectives and overview of the study

The study evaluated an automated workflow for qualitative multi-class screening of drinking water contaminants. Key aims were to: demonstrate automated SPE with physical solvent drying (DryDisk/Gore‑Tex membrane), couple extracts to fast GC-TOFMS for rapid analysis, assess performance across diverse analyte classes (phenols, acids, triazines, organochlorine and organophosphorus pesticides, THMs, etc.), and illustrate qualitative identification capability at low microgram-per-liter spike levels (1–10 µg/L). Quantitation was not performed because the final extract volume was not controlled in this exploratory work.

Methods and sample preparation

• Sample matrix and spiking: One liter of dechlorinated tap water spiked with target pesticides/contaminants at 1–10 µg/L. Acidification to pH 2 was applied for initial extraction; in one workflow a second extraction at pH 12 was also performed to capture additional classes after methylation.
• Automated SPE and solvent drying: Horizon Technology SPE-DEX 4790 equipped with 50 mm hydrophilic DryDisk. Pre-conditioning solvents (dichloromethane, acetone, water) were delivered automatically, samples were processed directly from original bottles, analytes eluted with DCM, and residual water removed by the DryDisk Gore‑Tex membrane (physical drying without chemical desiccants).
• Derivatization/methylation: Selected extracts were methylated prior to GC analysis to improve volatility and detectability for certain acidic compounds.
• Injection and GC method: 1 µL cold splitless injection. Fast GC protocol using a narrow-bore DB-5 column (20 m × 0.18 mm × 0.18 µm) with a rapid oven program (example: 40°C hold 3 min; 80°C/min to 280°C hold 3.5 min) yielding a total run time ~9.5 min. High column flow (≈1.8 mL/min) and rapid optic temperature ramping in the injector were employed to shorten chromatographic time.
• TOFMS acquisition: LECO Pegasus III TOFMS with high spectral acquisition rates (configured here at 30 spectra/s for the method). Mass range 30–500 m/z, solvent delay ~200 s, ion source ~230°C, transfer line ~260°C. Automated peak finding and spectral deconvolution (ChromaTOF) used for rapid identification of coeluting and low-level analytes.

Used instrumentation

• Horizon Technology SPE-DEX 4790 automated SPE workstation
• Horizon Technology DryDisk (Gore‑Tex hydrophobic membrane drying)
• ATAS GL Optic 2 programmable injector and Focus robotic sample handler
• Agilent 6890 gas chromatograph with DB-5 narrow bore column (20 m × 0.18 mm × 0.18 µm)
• LECO Pegasus III Time‑of‑Flight Mass Spectrometer
• LECO ChromaTOF software for peak finding and mass spectral deconvolution

Main results and discussion

• Broad applicability: The automated SPE-DEX/DryDisk workflow successfully transferred analytes across a wide polarity and volatility range—from very volatile trihalomethanes (e.g., dibromochloromethane, tribromomethane) to high-boiling compounds such as squalane.
• Rapid analysis: Combined GC runtime (~9.5 min) and automated TOF processing (≤2 min) provided a substantial time reduction versus conventional GC–quadrupole/ion‑trap MS methods (an order of magnitude faster in this report).
• Resolving power and deconvolution: The Pegasus III TOFMS and ChromaTOF algorithms resolved closely eluting pesticides; the authors report instances where seven pesticides were resolved and identified within a 2-second chromatographic window.
• Identification quality: Representative spectral matches included dibromochloromethane (~80% similarity), dichlorvos (~90% reference match), phenol and d6‑phenol (>87% matches), and azinphos compounds (matches around 75% in some cases), demonstrating useful library identification even at 10 µg/L spike levels.
• Qualitative but not quantitative: Because extracts were not solvent-exchanged to a controlled final volume or further concentrated, the study focused on qualitative screening and identification rather than quantitative performance (recoveries, limits of detection/quantitation not fully established).
• Stability and recovery benefits: Short GC analysis times likely reduced on-column degradation of labile pesticides (e.g., azinphos methyl/ethyl), contributing to good peak shapes and apparent recoveries in this qualitative assessment.

Benefits and practical applications of the method

• High throughput screening: Automated extraction and fast GC-TOFMS allow rapid screening of many samples per day, suited to monitoring programs and incident response.
• Reduced manual labor and solvent use: SPE automation and membrane drying eliminate large-volume LLE and chemical drying agents, lowering consumables, operator time, and potential human error.
• Improved detectability of coeluting compounds: TOFMS spectral deconvolution facilitates detection and identification of coeluting analytes and low-level unknowns in complex water matrices.
• Versatile multi-class coverage: The workflow handled diverse compound classes (phenols, triazines, organochlorines, organophosphates, acids, THMs), making it suitable as a broad screening platform prior to targeted quantitation methods.

Limitations

• No quantitative validation in this study: Final extract volume was uncontrolled and no recovery/LOD/LOQ data were reported, so method requires thorough validation before routine quantitative use.
• Library-dependent identification: Quality of identifications depends on spectral library coverage and match thresholds; some compounds returned moderate matches (≈75–80%), necessitating confirmatory steps for regulatory reporting.
• Matrix effects and method robustness: Additional work is needed to characterize matrix interferences, membrane compatibility for diverse chemistries, and long-term reproducibility of automated SPE hardware.

Future trends and potential applications

• Quantitative method development: Implementing controlled concentration/solvent-exchange steps, internal-standard strategies and full validation (recoveries, precision, LOD/LOQ, linearity) will enable routine quantitative monitoring.
• Shorter columns and faster separations: Use of shorter GC columns (e.g., 10 m) and optimized temperature/flow programs could further reduce runtimes while maintaining resolution with high-rate TOF acquisition.
• Expansion to suspect and non-target screening: High-resolution, fast-acquisition TOF instrumentation combined with deconvolution supports suspect screening and discovery of unknown contaminants in water supplies.
• Integration with automation and LIMS: End-to-end automation (robotic sample handling, automated SPE, and software-driven reporting) will increase sample throughput and reduce turnaround time in routine water quality laboratories.
• Regulatory adoption and field deployment: After robust validation, the approach could be incorporated into regulatory monitoring schemes, emergency response workflows, and screening tiers that trigger targeted confirmatory analyses.

Conclusion

The combined automated SPE-DEX/DryDisk extraction and Pegasus III GC‑TOFMS workflow provides a rapid, broad‑scope qualitative screening capability for drinking water contaminants. The approach demonstrably reduces analysis time and handles a wide range of compound classes with effective spectral deconvolution and identification. To transition from promising qualitative screening to routine quantitative monitoring, targeted method optimization and formal validation (controlled extract volumes, recovery studies, limits of detection, matrix effects) are required. With these developments, the technique has clear potential to increase laboratory throughput and improve early detection of water contaminants.

References

• Scott S., Turner D., High Throughput Multi-Residue Screening of Drinking Water using the SPE-DEX® and Pegasus® GC-TOFMS, LECO Corporation application note, 2008.