Using GC-MS Workflow to Test for Phthalates

The first issue concerns the risk of sample contamination during GC-MS analysis, which affects the reliability of the resulting data. Phthalates are ubiquitous in the environment so they can easily contaminate samples during preparation and analysis, and potentially be carried over from injection to injection. The use of clean glassware, correct GC consumables, high purity standards, and solvents are crucial for producing data that analysts can have confidence in. Moreover, poorly optimized experimental conditions can cause phthalates to persist in instrument inlets, transfer lines, and ion sources, causing contamination over extended analyses and resulting in false positive results.

A second key challenge is related to the complexity of fatty matrices, such as cooking oils, which are difficult to analyze directly using GC-MS and often require extensive sample clean up procedures prior to injection. Additionally, heavier fractions, like triacylglycerols, can be difficult to elute from the chromatographic column due to their high boiling points. Given these challenges, more robust GC-MS workflows are required to ensure the reliable detection of phthalate contaminants in complex fatty foods.

Lastly, phthalates are characterized by similar molecular structures and physical properties. Many of them produce similar fragment ions and can co-elute if the chromatographic separation is not optimized. The correct choice of capillary column and MS quantification ions are important for the reliable identification of phthalates.

An Advanced GC-MS Workflow

A novel approach has recently been developed that overcomes the challenges of detecting phthalates in fatty foods. The new workflow, which makes use of the Thermo Scientific ISQ 7000 GC-MS system configured with the sensitive Advanced Electron Ionization (AEI) source, has been successfully used for the detection of 13 phthalates in vegetable oil, offering a fast, sensitive, and robust method for phthalates quantification.

To assess the linearity, limit of detection (LOD), and limit of quantification (LOQ) of the new method, vegetable oil samples were spiked with phthalates at three concentration levels (5, 25, and 50 µg/kg). The spiked vegetable oil samples were added to acetonitrile, vortexed, and sonicated before being centrifuged. The supernatant was collected and extracted to dryness, reconstituted into hexane, and subsequently analyzed for phthalates by GC-MS. To minimize the risk of contamination and to handle the high boiling nature of the analytes and the matrix, low bleed and highly inert consumables combined with optimized instrument conditions were used. The method employed polytetrafluoroethylene and siloxane vial closures, bleed-temperature-optimized inlet septa, and used optimized syringe washes, inlet, and MS temperature conditions. The results showed no heavier compound carryover, highlighting the robustness of the method.

Using timed selective ion monitoring (timed-SIM) mode enabled a significant improvement in analytical selectivity and sensitivity over full-scan acquisition (Figure 2), as only data on masses of interest were collected, rather than the full mass range. Thermo Scientific Chromeleon Chromatography Data System software was used to automatically optimize scan rate and dwell time for faster experimental setup and analysis. The system demonstrated selective and sensitive detection of phthalates in complex vegetable oil matrices.

A Better Analytical Approach for Phthalate Testing

The new timed-SIM GC-MS workflow achieved estimated LOQs ranging from 5 to 25 µg/kg, and all 13 phthalates showed excellent linear responses, with an average R²=0.999. An assessment of the recoveries of the pre- and post-spiked vegetable oil samples (across the three 5, 25, and 50 µg/kg spiking levels) returned average recovery values between 80 and 102 percent, well within the required method performance limits. These results highlight the ideal limits of detection achieved by the method, even when studying challenging food samples.

The robustness of the AEI ion source over time was demonstrated by the ion ratio stability being within ± 10 percent over 100 repeated injections of the 50 ng/mL spiked vegetable oil extract. The improved geometry of the AEI source enhanced the ionization efficiency while generating a highly focused ion beam, reducing the risk of source contamination. Thanks to the enhanced sensitivity of the AEI source, the oil extract can be further diluted before injection, or a higher split ratio can be used, maintaining sub-ppb limits of detection, giving more flexibility in sample preparation and lowering the risk of contamination to the GC flow path.