Comparison between Different Stationary Phases for the Separation of Phthalates using Gas Chromatography-Mass Spectrometry (GC-MS)

Dan Li, Rebecca Stevens, and Chris English

Phthalates are widely used as plasticizers in a variety of industry products. However, some phthalates are considered as endocrine disruptors[1] associated with a number of problems, including birth defects[2], high blood pressure in children[3], pregnancy-induced hypertensive heart diseases[4], respiratory problems[5], and obesity[6]. The European Union (EU) and United States Environmental Protection Agency (US EPA) have restricted the use of the most harmful phthalates (see Table 1).

Table 1. Elution times for Phthalates on various Restek GC columns.

A commonly used technique for phthalate analysis is gas chromatography-mass spectrometry (GC-MS), which is simple, fast and inexpensive. GC-MS also provides additional mass spectral information and thus serves as an ideal instrumental platform for phthalate identification and quantification. Selection of GC columns is critical, because each stationary phase has a unique selectivity, which affects the relative elution order and resolution. Phthalate separation is challenging due to structural similarity. For instance, many phthalates share a common base peak ion (m/z 149) which makes identification of coeluting phthalates difficult. Technical grade mixtures and isomers further complicate identification of target phthalates.

Researchers have evaluated different stationary phases for phthalate analysis. A recently published review summarized the most used GC and liquid chromatography (LC) columns[7]. According to the literature, GC-MS has higher sensitivity compared to LC-MS for phthalate determination, and the most commonly employed GC columns in descending order of popularity are 5-type, XLB-type, 35-type, 17-type, 50-type, and 1-type.

Separation specific to a stationary phase is achieved by adjusting conditions. The Pro EZGC program is a fast modeling software which can optimize GC parameters (e.g., carrier gas type, flow rate, temperature program, column dimensions, and guard column) to produce the shortest run time on a given type of stationary phase. In this study, libraries of 37 phthalates (see Table 2 for phthalate names) were built into the Pro EZGC program for seven of the most frequently used column phases: Rtx-440, Rxi-XLB, Rxi-5ms, Rtx-50, Rxi-35sil, Rtx-CLPesticides, and Rtx-CLPesticides2. These stationary phases were evaluated for the analysis of both regulated and non-regulated phthalates.

Table 2. Elution times for Phthalates on various Restek GC columns.

The standard EPA Method 8061A Phthalates Mixture (cat. #: 33227) consists of 15 components, each at a concentration of 1,000 μg/mL. Benzyl benzoate (cat. #: 31847) was used as the internal standard. All other phthalate standards were purchased from Chem Service (West Chester, PA). The GC-MS analysis was performed on a Shimadzu GC-MS QP2010 Plus instrument. The GC-MS was equipped with one of seven Restek columns using 30 m×0.25 mm×0.25 µm dimensions (0.20 µm for Rtx-CLPesticides2 column) (see Table 1 for the column types). The GC-MS experimental parameters were listed in Table 3.

Table 3. GC-MS parameters meters

Standards and tested samples were dissolved and diluted in methylene chloride. Standard solutions were prepared at 50 μg/mL (80 μg/mL for benzyl benzoate). During sample preparation, plastics were strictly avoided; all preparation work was performed using glassware (volumetric flask, syringe, vial, etc).

A direct comparison between the columns for the separation of EPA and EU regulated phthalates was performed. The elution times of seven different phases were predicted by the Pro EZGC program under the same optimized GC conditions (Table 1). Coelutions were counted as compound pairs with a resolution less than 1.5. The total analysis time is less than 6 min under these conditions. In order to confirm the prediction, chromatograms of each stationary phase were collected under the same optimized condition (Figure 1). Because the column length is not exactly 30 meters long as in the simulation, the retention times are slightly different from the prediction. The elution orders and coeluting pairs were exactly the same as predicted. Among the seven phases, Rtx-440, Rxi-XLB and Rtx-CLP and Rxi-35 sil provided baseline separation for all EPA and EU listed phthalates. The two isomers of bis(4-methyl-2-pentyl) phthalate were not resolved on the seven phases. The elution order was comparable on the Rtx-440, Rxi-XLB, Rtx-CLP and Rxi-5ms columns. Interestingly, differences in the elution orders were observed on Rxi-35 sil and Rtx-50 phases. Most notably, the elution orders of 4 pairs of phthalates changed on Rxi-35sil phase, including bis(2-methoxyethyl) phthalate / bis(4-methyl-2pentyl) phthalate (peak 6 and 7/8), bis(2-ethoxyethyl) phthalate / diamyl phthalate (peak 9 and 10), butyl benzyl phthalate / hexyl-2-ethylhexyl phthalate (peak 12 and 13), and bis(2-n-butoxyethyl) phthalate / bis(2-ethylhexyl) phthalate (peak 14 and 15). The Rtx-440 and Rxi-35sil columns are ideal as a parallel dual column set for electron capture detector (ECD) analysis, where Rxi-35sil column serves as a good confirmation column. Rtx-440 and Rtx-XLB columns showed the highest resolution in this fast analysis. Peaks that coeluted on other phases were well resolved on Rtx-440 and Rtx-XLB columns. For instance pairs that are not resolved on other phases include: bis(2-ethylhexyl) phthalate and dicyclohexyl phthalate (peak 15 and 16) on Rxi-5ms; bis(2-ethylhexyl) phthalate and butyl benzyl phthalate (peak 15 and peak 12) on Rtx-50 column; and bis(2-methoxyethyl) phthalate and bis(4-methyl-2-pentyl) phthalate (peak 6 and peak 7,8) on Rtx-CLP2 column. It is challenging to separate technical isomer mixtures, such as diisononyl phthalate and diisodecyl phthalate (peak 18 and peak 19). Fortunately, unique extracted ions are available for identification and quantification, i.e., m/z 293 for diisononyl phthalate, and m/z 307 for diisodecyl phthalate (see Figure 1).

A comprehensive comparison between the seven stationary phases for the separation of 37 phthalates (a total number of 40 peaks including 3 isomers) was performed using retention times predicted by the Pro EZGC program (see Table 2). The GC parameters, specified in Table 3, provided separation of 34 out of 40 peaks on both Rtx-440 and Rxi-XLB columns in less than 40 min. The two phases have different coelutions. The chromatogram on the Rtx-440 column was collected and shown in Figure 2. For some pairs that were not baseline-resolved, the resolution is still adequate for qualitative analysis.

There is no single condition set optimal for all phases. The fast program with the most peaks resolved was selected for the column comparison. Further optimization for each phase can be achieved using the Pro EZGC program.

The most commonly used seven GC columns were compared for phthalates analysis. The Pro EZGC program provides flexibility in GC optimization. The superior selectivity and efficiency of Rtx-440 and Rxi-XLB columns resulted in a fast runtime in both regulated phthalates and the extended list. With good resolution, higher maximum operating temperature (340 ºC for Rtx-440 and 360 ºC for Rxi-XLB), and minimum phase bleed, the Rtx-440 and Rxi-XLB columns are the preferred choices for phthalate analysis. A dual column set of Rtx-440 and Rxi-35sil is an alternative method for analyte confirmation.

 

Acknowledgements

The authors would like to thank Shimadzu Corporation for their consultation with the operation of the QP2010 Plus GC-MS Instrument

References

[1] Choi, H.; Kim, J.; Im, Y.; Lee, S.; Kim, Y., The association between some endocrine disruptors and hypospadias in biological samples. Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering 2012, 47 (13), 2173-9. [2] Nassar, N.; Abeywardana, P.; Barker, A.; Bower, C., Parental occupational exposure to potential endocrine disrupting chemicals and risk of hypospadias in infants. Occupational and environmental medicine 2010, 67 (9), 585-9. [3] Trasande, L.; Sathyanarayana, S.; Spanier, A. J.; Trachtman, H.; Attina, T. M.; Urbina, E. M., Urinary phthalates are associated with higher blood pressure in childhood. The Journal of pediatrics 2013, 163 (3), 747-53 e1. [4] Werner, E. F.; Braun, J. M.; Yolton, K.; Khoury, J. C.; Lanphear, B. P., The association between maternal urinary phthalate concentrations and blood pressure in pregnancy: The HOME Study. Environ. Health 2015, 14, 75. [5] Jaakkola, J. J.; Knight, T. L., The role of exposure to phthalates from polyvinyl chloride products in the development of asthma and allergies: a systematic review and meta-analysis. Environ. Health Perspect. 2008, 116 (7), 845-53. [6] Hatch, E. E.; Nelson, J. W.; Stahlhut, R. W.; Webster, T. F., Association of endocrine disruptors and obesity: perspectives from epidemiological studies. Int. J. Androl. 2010, 33 (2), 324-32. [7] Net, S.; Delmont, A.; Sempere, R.; Paluselli, A.; Ouddane, B., Reliable quantification of phthalates in environmental matrices (air, water, sludge, sediment and soil): a review. Sci. Total Environ. 2015, 515-516, 162-80.