Monitoring of 12 DBPs in drinking water using a microextraction TD-GC-MS method
Advances in Sample Preparation, Volume 14, 2025, 100168: Graphical abstract
Disinfection of drinking water is vital for public health but leads to the formation of harmful disinfection by-products (DBPs). To comply with EU Directive 2020/2184, a sensitive and selective HS-HiSorb-TD-GC-MS method was developed and validated for quantifying 12 DBPs in drinking water. The method showed excellent linearity (1–120 ppb), low detection limits (0.33–3.33 ppb), and good precision and recovery.
Its performance was verified across concentrations, with intra-day and inter-day variability ranging from 1.3–10% and 3.3–15%, respectively. Recovery ranged from 80–120%. The method was tested for sample stability at multiple storage temperatures and applied to real tap water samples. Its sorbent-based approach supports green analytical practices, offering a sustainable solution for routine DBP monitoring.
The original article
Monitoring of 12 DBPs in drinking water using a microextraction TD-GC-MS method
Photini Papaioakeim, Efstathios A. Elia, Agapios Agapiou
Advances in Sample Preparation, Volume 14, 2025, 100168
https://doi.org/10.1016/j.sampre.2025.100168
licensed under CC-BY 4.0
Selected sections from the article follow. Formats and hyperlinks were adapted from the original.
The disinfection of drinking water is considered a monumental achievement for public health. The World Health Organization (WHO) emphasizes on the critical importance of water safety for public health, including various aspects such as drinking, domestic activities, food production, and recreation [1]. Various chemical disinfectants have been used to eliminate harmful pathogens (e.g., cholera, dysentery, etc.) that can be present in untreated water, these include the use of chlorine, ozone, chloramine, and chlorine dioxide. However, a number of risks to human health appeared following the use of these means of water disinfection, such as cancer, genetic damage and developmental effects, particularly after long-term exposure [2]. Therefore, public water supply systems are regularly monitored to ensure that treated water is within the established disinfection by products (DBPs) regulatory limits [3].
Green sample preparation microextraction techniques are considered efficient and cost-effective alternative tools for the extraction of water DBPs, offering improved detection via preconcentration. In particular, solid-phase microextraction (SPME) is a widely used sorbent-based technique, where a coated fiber is used for the adsorption of analytes [[19], [20], [21], [22], [23]]. An enhanced version of this technique is the SPME arrow, designed with a larger surface area and capacity compared to standard SPME fibers, which increases extraction efficiency. A step forward, is HiSorb™ that offers much higher extraction capacity and is suitable for the analysis of higher molecular weight compounds. HiSorb™ extraction tool consists of a stainless steel rod, part of which is coated with an adsorbent material. It is a passive extraction technique (equilibrium extraction) based on the same principle as the SPME technique. HiSorb features a larger phase volume of 30 µL compared to the 0.6 µL of SPME, allowing for the adsorption of more analytes, thereby enhancing the sensitivity of the method. Additionally, HiSorb™ is more durable than SPME fibers [24]. These techniques are particularly effective for extracting DBPs at low concentrations, ensuring reliable analytical results [25,26].
SPME and SPME arrow analyses require only a standard GC system with a minor modification on the inlet, whereas HiSorb needs an extra thermal desorption (TD) unit directly connected to the GC column hence acting as the inlet for the system. The abundant nature of THMs, in combination with their regulated levels and negative health-effects, render them as potential candidate molecules for water quality studies. Although THMs are primarily associated with water treatment processes, their presence is extended to other water-containing products. Using various microextraction techniques, THMs were detected in a wide array of samples e.g., whey [27], beer [23], soft drinks [12], and swimming pools [22].
The aim of the present work was to develop a headspace (HS)-HiSorb-TD-GC-MS analytical method for the simultaneous analysis of DBPs (Table 1) in line with the EU 2020/2184 directive. A limited number of green and sustainable analytical methods exist for monitoring tap water DBPs [28,29]. Given the urgent need to transition to more sustainable and environmentally friendly analytical chemistry techniques, this method incorporates these essential characteristics for effectively monitoring DBPs in tap water.
2. Materials and methods
2.4. HiSorb-TD-GC-MS
TD was performed using an Unity-xr thermal desorber (Markes International Ltd.) connected to an Agilent GC-7890B gas chromatograph and an Agilent MSD-5977B mass spectrometer (TD-GC-MS). The analytical details of the TD-GC-MS method are presented in the supplementary material (Table SM1). Helium (99.999 %, Linde) was used as the inert carrier gas at a flow rate of 3.45 mL/min. Chromatographic separation was achieved using a DB-5MS capillary column that was heated as follows: initial temperature 35°C, maintained for 2 min, reached 250°C at a rate of 15°C min−1, and maintained for 0.5 min. The total analysis time was 16.83 min that included a 1 min solvent delay to protect the electron ionization (EI) filament. EI (70 eV) mass spectra were collected in full scan mode in the range 35–350 m/z. The quadrupole, interface and ion source temperatures were set at 150°C, 250°C, and 230°C, respectively. Chromatograms were processed using MSD ChemStation (Agilent), and the NIST20 library (National Institute of Standards and Technology) was used to identify the detected analytes. Three replicates were performed per sample, and the results were expressed as the area ratios of the analyte to IS2 (1,2-Dibromoethane) area.
3. Results and discussion
3.1. Optimization of HiSorb extraction
3.1.8. Selection of chromatographic column
After investigating the optimal sampling conditions using SPB-624 column, two chromatographic columns were examined for the analysis of DBPs. Initially, a SPB-624 column was tested but although it separated the four THMs, it resulted in co-elution of the two HKs. To address this issue, an alternative stationary phase was investigated. According to the literature, a potentially suitable column is the DB-5MS column [44]. Upon changing the column, the co-elution issue was resolved, and the total analysis time significantly improved providing several advantages, both for the analyst in terms of time, and for the environment, as less energy was required for the analysis. Additionally, as shown in the respective chromatograms (Figs. 2a and b), the peak intensity for all THMs at 50 ppb was significantly higher using the DB-5MS column. Finally, due to the higher temperature limits of the DB-5MS compared to the SPB-624 column (350 vs. 250°C), it could be possible to also analyze DBPs with higher boiling points using a DB-5MS column. It was anticipated that the DB-5MS column would deliver better results and require shorter analysis times compared to the SPB-624 column due to its shorter length and thinner film thickness. It is important to note that both columns displayed additional peaks unrelated to the samples, such as siloxanes. After the DB-5MS column was selected as the most suitable chromatographic column for all the compounds analyzed, the necessary checks were performed to ensure that the results from the optimization experiments were validated. Therefore, method validation and all subsequent water sample analyses were performed using the optimized HS-HiSorb-TD-GC-MS method with the DB-5MS column.
Advances in Sample Preparation, Volume 14, 2025, 100168: Fig. 2. HS-HiSorb-TD-GC-MS chromatogram of the THMs using (a) the SPB-624 chromatographic column at 50 ppb, and (b) the DB-5MS chromatographic column at 50 ppb.
3.4. Monitoring DBPs in tap water samples
The suitability of the developed method was tested by analyzing tap water samples from Cyprus. Note that, in Cyprus, water disinfection is carried out using NaOCl. As anticipated, the four THMs (TCM, BDCM, DBCM, and TBM), along with CH and DBAN, were detected and quantified, with measured concentrations of 2.22, 18.31, 29.97, 20.27, 9.49, and 10.75 ppb, respectively. The total THMs concentration was 70.78 ppb, within the regulatory guidelines. An exemplary HS-HiSorb-TD-GC-MS chromatogram of a tap water sample is shown in Fig. 4, presenting the six detected DBPs. Additional compounds were identified in the water samples that are not of significance to this study but could be used for other tap water quality studies. These include various organic compounds, such as bromo‑chloro‑methane, dibromo-methane, 1‑bromo-2‑chloro-ethane, nonanal, 2,4-di‑tert-butylphenol, o-hydroxybiphenyl, etc. Some of these compounds are not necessarily associated with the water disinfection process but have exogenous origins. For example, o-hydroxybiphenyl is a biocide commonly used as a preservative and a fungicide, particularly for waxing citrus fruits. Additionally, the tallest peaks in the chromatogram primarily correspond to siloxanes likely originating from the thermal degradation of the PDMS adsorbent material used in the analysis. This highlights an advantage of full-scan analysis over SIM mode analysis, as it enables the detection of ionizable compounds within the sample, providing a comprehensive overview of its composition [52].
Advances in Sample Preparation, Volume 14, 2025, 100168: Fig. 4. Typical HS-HiSorb-TD-GC-MS– chromatogram of a tap water sample
4. Conclusions
A green sample preparation microextraction method was developed and optimized for the simultaneous monitoring of 12 low molecular weight DBPs in water samples using HiSorb combined with TD-GC-MS. The optimized conditions - headspace PDMS sampling, 2 g NaCl addition, a 24-hour equilibrium time, 90 min HiSorb sampling, simultaneous heating in a water bath at 60°C, and stirring at 1000 rpm - maximized extraction efficiency, particularly for THMs.
The method demonstrated excellent linearity (R² = 0.984–0.999), low detection limits (0.33–3.33 ppb), and good precision (1.3–10 % intra-day, 3.3–15 % inter-day), with recoveries ranging from 80 to 120 %. Finally, the stability of THMs in the water samples was examined, revealing that they remained stable for up to 3 days when stored at 4°C.
However, the high initial cost of thermal desorption systems may limit accessibility, and extended equilibrium times may be a drawback for time-sensitive analyses. Despite these limitations, the method represents a step toward more sustainable and efficient DBP monitoring in water samples.
