Analyzing Brominated Flame Retardants in Food Using GC–APCI-MS–MS

Gas chromatography–atmospheric pressure chemical ionization (GC–APCI) offers increased limits of sensitivity in food analysis. Carlos Sales Martinez from the Research Institute of Pesticides and Water in the University Jaume I, Castellón, Spain, has been exploring the novelty of this technique for the analysis of food samples. He recently spoke to LCGC about this work.

Q. In 2015, your group published a study presenting a novel analytical approach using gas chromatography–atmospheric pressure chemical ionization-tandem mass spectrometry (GC–APCI-MS–MS) to analyze brominated flame retardants (BFRs) in food (1). What led your group to investigate this approach?

A:

Our group previously collaborated with other groups to improve our understanding of the methods used to analyze BFRs. This scheme brought us in contact with Mª José González and Belen Gómara from the Institute of General Organic Chemistry at CSIC in Madrid, Spain. This group had previously applied gas chromatography–electron ionization-tandem mass spectrometry (GC–EI-MS–MS) to polybrominated diphenyl ethers (PBDEs) and novel brominated flame retardants (BFRs) 1,2-Bis(2,4,6-tribromophenoxy)ethane (BTBPE) and decabromodiphenyl ethane (DBDPE) to investigate a wide variety of complex matrices, but observed sensitivity limitations, particularly for the highly brominated congeners (2). Our study was derived from their research: We wanted to explore the potential of the new atmospheric pressure chemical ionization (APCI) source designed for GC to analyze these compounds.   

Q. What are the benefits of GC–APCI compared to traditional GC–EI for persistent contaminants and volatile organic compounds (VOCs) in food samples?

A:

The novel APCI source is a soft ionization source that enhances the formation of the molecular ion and/or protonated molecule with a significantly reduced in-source fragmentation. In most cases this is the base peak of the spectrum. Electron ionization sources provide a very low intensity (or absence) of the molecular ion when analyzing persistent organic pollutants (POPs) such as PBDEs, some novel BFRs, hexabromocyclododecane (HBCD), as well as some pesticide families like pyretroids or endosulphans, among many others. Generally the MS–MS transitions use a fragment ion as a precursor. However, APCI promotes the formation of the (quasi)molecular ion, which becomes suitable as a precursor ion for MS–MS experiments, resulting in the enhancement of both sensitivity and specificity, and decreasing interferences. In non-target analysis, the high fragmentation observed in EI sources could provide similar spectra for different components, so having the molecular ion or the protonated molecule is a great advantage when elucidating unknown compounds as well as for “suspect” analysis approaches.  

Q. Can you comment on the sensitivity of the new method?

A:

For most of the BFRs studied, sensitivity was increased by around 10 times using GC–APCI-MS–MS compared to the previous methods using GC–EI-MS–MS. This improvement was even better for highly brominated congeners. As a result of their particular structure, we decided to add several modifiers to the source to promote the formation of protonated molecules. Although the addition of water provided cleaner MS spectra and improved the repeatability, working without modifiers was found to be more sensitive and was therefore selected for further experiments (1). In addition, working in charge-transfer conditions allowed us to develop a screening method for trace levels of HBCD. Normally this would require analysis with liquid chromatography (LC), and several additional sample treatment steps (3). The use of high carrier gas flow rates, up to 4 mL/min for BFRs, also provided a noticeable increase in detectability and resolution for later-eluting compounds like BDE-209 or DBDPE. Using high carrier gas flow is extremely important for developing rapid chromatographic methods.   

Q. What were the main analytical challenges you encountered and how did you overcome them?

A:

In the case of BFRs we wanted to develop a considerably faster method for a relatively high number of compounds, which would normally require separate injections or large chromatographic analyses. Finding the most suitable column and conditions took a while. When working with HBCD we noticed a huge signal decrease with a few analyses on regular 5% phenyl/95% dimethylpolysiloxane phase columns, so the use of a 100% dimethylpolysiloxane phase column ready for high temperatures was required. In the case of BDE-209 and DBDPE, we struggled significantly in terms of reproducibility until we raised the MS interface temperature up to 350 °C, though BDE-209 and DBDPE elute at an oven temperature of 325 °C. We have noticed in several other applications that having a considerably long MS interface in some GC–APCI systems leads to a decline in sensitivity for later-eluting compounds because the columns degrade faster. To avoid this kind of problem we have implemented inert post-columns with press-fit connections in most of our experiments. In addition, the use of deactivated post-columns reduces the already short column-change time in APCI, so we strongly recommend their use for POPs and metabolomics-based analyses.  

Q. What advice would you give to anyone thinking of using GC–APCI for the first time?

A:

While APCI has been shown to render enhanced sensitivity in most tested applications, the lack of spectral libraries can be a drawback for some experiments, especially regarding non-target analysis and unknowns. In such cases, the advantage of an increased sensitivity may be less important than having instant tentative identification by searching in spectral EI libraries. However, the expected presence of the (quasi)molecular ion in the APCI spectrum makes the suspect analysis of, in principle, an unlimited list of compounds from a sample feasible. With regards to the use of modifiers, from our experience, one should always try with several of them depending on the type of ionization you are working with (positive or negative) and study the analyte response in terms of sensitivity and repeatability. 
   

Q. Are there any other applications where you think GC–APCI could offer the analyst improved results?

A:

GC–APCI-MS–MS has already demonstrated impressive capabilities to enhance the sensitivity and specificity of a broad range of POPs such as PBDEs, dioxins, furans, and polycyclic aromatic hydrocarbons (PAHs) as well as pesticides and other contaminants (1,3,4,5). The quantification at low levels of MS–MS driven analytes can be improved by the use of APCI, especially in matrices for which EI extensive fragmentation produces interference. Additionally, GC–APCI combined with high-resolution quadrupole time-of-flight MS (HR–QTOF-MS) opens a new line of research in the “suspect” screening field, where large lists of potential compounds can be searched based solely on the expected presence of the (quasi)molecular ion in the APCI spectrum (6,7).  

Q. What is your group working on next?

A:

We are currently investigating the behaviour of halogenated flame retardants (HFRs) when using APCI, and studying the potential of APCI-MS–MS applied to HFRs further. Another interesting topic we plan to focus on is applying APCI–QTOF (MSE) to samples that have already been analyzed in a retrospective screening of unknown metabolites of PBDEs and other polybrominated compounds of interest. We believe that extremely sensitive analytical tools are still needed for the monitoring of both the metabolites and the original contaminants at low levels to evaluate the removal and bioaccumulation of these compounds in the environment. Finally, we are also planning to explore the potential of GC–APCI-HRMS in food quality and biomedical fields by making use of metabolomics-based approaches, in which the aforementioned benefits of this novel ionization source could potentially be useful.    

Acknowledgement

  The Research Institute for Pesticides and Water [IUPA] is directed by Félix Hernández and is where the GC–APCI research is conducted. Tania Portolés has been working with this novel technique for seven years. Joaquim Beltrán and Juan Vicente Sancho have also closely collaborated in this line.  

References

  1. T. Portolés

et al

.,

Anal. Chem.

87

(19), 9892–9899 (2015). 2. Belén Gómara

et al

.,

Anal. Chim. Acta

597

(1), 121–128 (2007). 3. C. Sales

et al

.,

Anal. Bioanal. Chem.

408

(2), 449–459 (2016). 4. B. Van Bavel

et al

.,

Anal. Chem.

87

(17), 9047–9053 (2015). 5. T. Portolés

et al

.,

J. Chromatogr. A

1260

, 183–192 (2012). 6. T. Portolés

et al

.,

J. Chromatogr. A

1339

, 145–53 (2014). 7. T. Portolés

et al

.,

Anal. Chim. Acta.

838

, 76–85 (2014).  

Carlos Sales

is a PhD student in sciences at the Research Institute of Pesticides and Water in the University Jaume I, Castellón, Spain. His research is focused on the use of gas chromatography coupled to mass spectrometry techniques (GC–MS) with special emphasis on the use of the novel atmospheric pressure chemical ionization (APCI) source for the identification, confirmation, and quantification of persistent organic pollutants and volatile organic compounds in complex environmental and food samples. His interests also include the study of the potential of GC–MS-based metabolomics in food and biological fields.                    

Tania Portolés

is a member of the Research Institute of Pesticides and Water in the University Jaume I, Castellón, Spain. Her research is focused on the use of gas chromatography–mass spectrometry (GC–MS) in last generation instruments, with special emphasis on high-resolution MS (HRMS) for the identification, confirmation, and quantification of organic contaminants in the environment, food, and biological fields. Special emphasis has been given in recent years to research performed with the atmospheric pressure chemical ionization source, in which her work has been a pioneer at the international level. She is a teacher of analytical chemistry for the Master’s degree in applied chromatographic techniques at University Jaume I.