Technical Note pubs.acs.org/ac
Characterization of Phenol and Alkyl Phenols in Organic Matrixes with Monoethylene Glycol Extraction and Multidimensional Gas Chromatography/Mass Spectrometry J. Luong,†,‡ R. Gras,‡ H. J. Cortes,†,§ and R. A. Shellie*,† †
Australian Centre for Research on Separation Science (ACROSS), University of Tasmania, Private Bag 75, Hobart, Tasmania 7001, Australia ‡ Dow Chemical Canada ULC, Highway 15, Fort Saskatchewan, Alberta, T8L 2P4, Canada § HJ Cortes Consulting LLC, Midland, Michigan 48642, United States S Supporting Information *
ABSTRACT: The use of monoethylene glycol as an extraction medium for removing phenol and alkyl phenols in organic matrixes such as hydrocarbons is introduced and combined with a practical analytical multidimensional gas chromatography approach. The analytical approach has been successfully developed for the characterization of phenol, cresols, xylenols, and alkyl phenols like 4-ethylphenol and 2,3,5-trimethylphenol. The technique employs a single-step extraction of the analytes with monoethylene glycol and sonication, followed by multidimensional gas chromatography with mass spectrometry in selected ion monitoring mode for the detection and quantitation. Extraction efficiency of phenol approached 100% while cresols, xylenols, and 4-ethylphenol were 97% or higher and 2,3,5-trimethylphenol was better than 91% under the analytical conditions used. With the technique described, a complete analysis can be conducted in less than 16 min. Reproducibility of area counts at two levels, namely, 5 ppmw and 50 ppmw over a period of 2 days were found to be less than 4% (n = 20). The analytes of interest was found to be linear over a range from 100 ppbw to 250 ppmw with correlation coefficient of at least 0.999 and detection limit of 50 ppbw. Spike recoveries from 500 ppbw to 250 ppmw for all analytes range from 96 to 102%.
P
polyglycols to extract phenolic compounds from coal tar has been patented.13,14 Polyglycols, however, are specialty chemicals, can be difficult to obtain, and due to the high molecular weight involved, not suitable for gas chromatography (GC) analysis. In contrast, monoethylene glycol (MEG) possesses many chemical properties similar to polyglycols owing to the fact that it is the building block for polyglycols. MEG is amenable for GC analysis with a boiling point of 197 °C at 760 mm Hg. With a flash point of 116 °C, MEG is a stable extract with no issue(s) related to evaporation after the extraction is completed. Further, MEG is a commodity chemical and is readily available worldwide. In the present approach, we introduce a single-step extraction of phenol, cresols, xylenols and other alkyl phenols such as ethyl phenols and trimethyl phenols with MEG as an extraction medium from organic matrixes such as hydrocarbons, coal tar, and industrial solvents to name a few.
henol, cresols, and xylenols (dimethyl phenols) are widely applied in the pharmacological and chemical industries. They are involved in the production of polymers, drugs, dyes, explosives, pesticides, disinfectants, antiseptics, medicinal preparations like mouthwash, throat sprays/lozenges, stabilizers, and antioxidants.1−5 The chemical properties of phenol are rather unique, due to the presence of a hydroxyl group and an aromatic ring that are complementary to each other in facilitating both electrophilic and nucleophilic type of reactions. Phenol has high reactivity of its ring toward electrophilic substitution. A number of analytical approaches have been reported for the determination of phenolic compounds, especially for aqueous matrixes like wastewater or potable water.6−12 Among sample preparation reported in literature, the most common one for phenols involves the use of a multistep liquid−liquid extraction.1,2,11 Phenol is a weak acid and readily forms sodium phenate. Thus phenolic compounds can be extracted with a 10% sodium hydroxide solution to convert phenols to their respective sodium phenates. Acidifying the caustic fraction converts the phenates back to phenols, which can finally be extracted with an organic solvent for analysis. This approach, while adequate, can be rather complicated and time-consuming. A novel industrial chemical process involving the use of high molecular weight © XXXX American Chemical Society
Received: April 3, 2013 Accepted: June 6, 2013
A
dx.doi.org/10.1021/ac400981z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Technical Note
Figure 1. Schematic diagram of the MDGC/MS analytical system.
■
EXPERIMENTAL SECTION An Agilent 6890N (Agilent Technologies, Wilmington, Delaware), equipped with an Agilent G-4512A autosampler, two split/splitless inlets, a flame ionization detector, an Agilent 5975B VL Mass Selective Detector, and an auxiliary pressure control module was used. The GC was also equipped with a SilFlow five-port planar microfluidic device, PN 123726 (SGE Analytical Science, Ringwood, Australia) for Deans switching purpose. A three-port, 24 V 5 W dc switching valve (Agilent G2399-60610) and support kit (Agilent G-2399-67610) was employed for the Deans switching task. The valve was connected to the planar microfluidic device with a custommade stainless steel tubing having a 1.1 mm outer diameter for the SilFlow device and a 1.59 mm (1/16 in.) outer diameter for the switching valve. A schematic diagram of the analytical system is illustrated in Figure 1. Gas chromatographic conditions used are as follows: a 20 m × 0.18 mm i.d. × 0.4 μm DB-1 (Agilent Technologies, Sacramento) was used in the first dimension and a 20 m × 0.25 mm i.d. × 1 μm VF-200 ms (Agilent Technologies, Middelburg, The Netherlands) was used in the second dimension. Deactivated but uncoated fused silica tubing (1.9 m × 0.15 mm i.d.) was also employed to balance the flows of the Deans switch. The column flow in the first dimension was 0.6 mL/min helium while the column flow in the second dimension was 1.5 mL/min helium, both in constant flow mode. The inlet pressure was 21.5 psig at 50 °C, while the auxiliary pressure was 8.0 psig at 50 °C to deliver the flows required. The inlet temperature was 250 °C, operating in split mode at a ratio of 40:1 and equipped with an Ultrainert liner PN 5190-2294 (Agilent Technologies, Folsom). The rationale behind using multidimensional gas chromatography for said application will be discussed later in the technical note. The temperature was programmed from 50 °C (0.5 min) to 250 °C at 15 °C/min and maintained at 250 °C for 2 min. The MSD detector ion source temperature was at 250 °C, quadrupole temperature was at 200 °C, and the transfer line temperature was at 300 °C. The gain for the electron multiplier gain was set at 5. The flame ionization detector (FID) temperature was at 300 °C with a hydrogen flow rate at 30 mL/ min, air flow rate at 350 mL/min, and nitrogen flow rate at 30 mL/min. Chromatographic data obtained were collected using MSD ChemStation version B.02.02.SP1 (Agilent Technologies, Waldbronn, Germany). Table 1 lists the parameters for monitoring the compounds mentioned under SIM mode.
Table 1. Parameters for Monitoring Target Compounds by MDGC/MS retention time (min)
compound phenol o-cresol 2,5-dimethylphenol 4-ethylphenol 3,5-dimethylphenol 2,3,5-trimethylphenol
1
D
8.4 9.8 11.5 11.7 11.7 13.3
2
D
10.1 11.4 13.1 13.3 13.4 15.0
cut window (min)
D
target ion (Da)
qualifier ion (Da)
8.1−8.7 9.5−10.1 11.2−12.0 11.2−12.0 11.2−12.0 13.0−13.6
94 108 107 107 107 121
66 77 122 122 122 136
1
Carrier and fuel gases such as helium, air, and nitrogen used for system performance studies were acquired from Air Liquide (Edmonton, Canada). Phenol and alkyl phenols such as ocresol, 2,5-dimethylphenol, 3,5-dimethylphenol, 4-ethylphenol, and 2,3,5-trimethylphenol were obtained from Sigma-Aldrich (Oakville, Canada). ACS grade solvents like cyclohexane, MEG were obtained from Fisher Scientific (Edmonton, Canada). In terms of sample preparation, if there were visible suspended particulates in the sample, the sample was first filtered with a 0.2 μm Whatman cellulose filter with a disposable B&D plastic syringe. In total, 1 g of the sample was then transferred to an Agilent 2 mL autosampler vial with a disposable polyethylene plastic pipet and 1 g of MEG was added to the sample. The sample vial was then sealed with a Teflon lined cap and put into a 50 mL Pyrex beaker that was filled with 25 mL of water. The level of the water in the beaker was maintained close to but below the cap of the sample vial to prevent any possibility of water leaking into the sample. With this arrangement, seven samples can be sonicated all at once. Each sample was sonicated for 10 min at 50 °C with a Branson 3510 Ultrasonic Sonicator (Branson Ultrasonics, Danbury). After sonication was completed, an aliquot of the glycol fraction (bottom fraction) was transferred to another Agilent 2 mL autosampler vial with a new disposable polyethylene pipet, capped with a Teflon lined cap, and analyzed per the method described.
■
RESULTS AND DISCUSSION In this investigation, we introduce an analytical approach for the characterization of phenol, cresol, xylenols (2,5-dimethylB
dx.doi.org/10.1021/ac400981z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Technical Note
Figure 2. (a) Chromatogram of phenol and alkyl phenols on first dimension column (FID). (b) Overlay of chromatograms (n = 2) of a mixed standard of 20 ppmw each phenol and alkyl phenols on second dimension column (MS-SIM).
MDGC was considered as a potential solution to resolve the issue encountered. With MDGC, only analytes of interest were directed to the MS, the remaining extract was quarantined in the first dimension; either being diverted to the FID or backflushed to vent. The column used in the first dimension was a 20 m × 0.18 mm i.d. × 0.4 μm polydimethylsiloxane (PDMS) based stationary phase to provide a coarse separation for the analytes against the matrix and as well as other chromatographic interferences. However, 4-ethylphenol and 3,5-dimethyl phenol are coeluted as shown in Figure 2a. The column used in the second dimension was a 20 m × 0.25 mm i.d. × 1 μm trifluoropropyl methyl polysiloxane based stationary phase to provide enhanced selectivity with a stronger π−π solutestationary phase interaction and having a compatible maximum operating temperature of 325 °C with the column employed in the first dimension. With the separation of the analytes of interest occurring at an elevated temperature (first component at 142.5 °C), a low phase ratio column was required for the second dimension separation to improve solute retention/
phenol, 3,5-dimethylphenol), and other alkyl phenols like 4ethylphenol and 2,3,5-trimethylphenol in nonpolar organic matrixes like hydrocarbons, industrial solvents, or diluents. A single-step extraction was employed with MEG and sonication followed by multidimensional gas chromatography and mass spectrometry (MDGC/MS) in selected ion monitoring mode (SIM) for the detection and quantification of the analytes cited. The use of MEG as an extraction medium was found to be highly effective in removing phenols and alkyl phenols from the organic matrix. Initially, the MEG extracts were analyzed using single dimension GC/MS with a 20 m × 0.25 mm i.d. × 1.0 μm trifluoropropyl methyl polysiloxane based stationary phase column. However, this is not a viable option as both the ion source and quadrupole of the mass spectrometer required cleaning to restore detector performance after about only 1 week of operation (150 analyses). The major source of the contamination was attributed to large amount of MEG injected into the system. C
dx.doi.org/10.1021/ac400981z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Technical Note
Figure 3. Overlay of chromatograms of a 250 ppmw each mixed phenol standard before and after extraction: black trace, before extraction; blue trace, after extraction.
Figure 4. Chromatogram of a coal tar extract diluted with recycled solvent.
vent to improve sample throughput and overall analytical system cleanliness. With MDGC, the ion source of the mass spectrometer only needs to be cleaned on average once every 9 months instead of a weekly event with GC/MS. The use of a planar microfluidic device for Deans switch offers key advantages such as low dead volume, inert, and leak free under the course of normal usage. The effectiveness of planar microfluidic devices and MDGC for flow switching has been addressed in details elsewhere.15−17 Key factors influencing extraction efficiency were evaluated and optimized including the duration of extraction, extraction temperature, and the use of sonication to enhance the overall extraction efficiency. The impact of temperature on average extraction efficiencies of the targeted compounds (n = 3) at a constant time of 20 min was determined (see Figure S1 in the Supporting Information). A mixed standard of 250 ppmw each
solute-stationary phase interaction. The dimensions of the column were chosen to ensure proper operation and balancing of the Deans switch. This was critical because one outlet was at atmospheric pressure (FID), while the other was at subambient pressure (MSD). Unlike in the first dimension, with this column, 4-ethylphenol and 3,5-dimethylphenol were adequately resolved (Rs = 1.3) while the rest of the analytes of interest were well separated from one another as shown in Figure 2b. MDGC offers a number of tangible analytical advantages such as the capability to leverage the difference in selectivity between the column set used to achieve improved separation as demonstrated with the critical pair mentioned, the capability to quarantine the majority of the sample matrix in the first dimension, and with the advent of an extra pressure source provided by the Deans switch, the option of back-flushing the heavier contaminants/matrixes trapped in the first dimension to D
dx.doi.org/10.1021/ac400981z | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Technical Note
in cyclohexane was used. At 20 °C, the lowest extraction efficiency was 79.0% for 2,3,5-trimethylphenol while the highest efficiency was 89.0% for phenol. At 60 °C, approximately 20 °C lower than the lightest component (cyclohexane) in a typical industrial recycled solvent, the average extraction efficiencies for 2,3,5-trimethylphenol and phenol were 82.0% and 92.0%, respectively. It can be concluded that an increase in temperature improved extraction efficiencies of the analytes of interest but not by a wide margin. The impact of sonication on the average extraction efficiencies of the targeted compounds (n = 3) at a constant temperature of 50 °C was also determined (see Figure S2 in the Supporting Information). As in the previous experiment, the same mixed standard was used. An average extraction efficiency of greater than 90% for all analytes was achieved in less than 5 min extraction. For the key target compounds such as phenol, cresol, and xylenols, an extraction efficiency of greater than 95% can be achieved after 5 min of sonication at 50 °C with phenol reaching 99.0%. The lowest extraction efficiency was 92.5% for 2,3,5-trimethylphenol, which is still quite respectable for a substituted alkyl phenol. Extending the extraction time to 30 min at 50 °C marginally improved the average extraction efficiency for all compounds with 2,3,5-trimethyl phenol reaching 93.4% while phenol at 99.8%. Clearly, the use of sonication aided in improving overall extraction efficiency for the targeted compounds. On the basis of the data collected, for practical purpose and to maintain adequate sample throughput, an extraction time of 10 min with sonication at 50 °C was selected for the extraction of phenol, cresol, xylenols (2,5dimethylphenol and 3,5-dimethylphenol), 4-ethylphenol, and 2,3,5-trimethyl phenol. Figure 3 shows an overlay of chromatograms of a mixed standard of 250 ppmw each of the compounds mentioned before and after extraction with MEG, demonstrating the effectiveness of the method in removing phenol and alkyl phenols. Reproducibility of area counts at two levels, namely, 5 ppmw and 50 ppmw over a period of 2 days were found to be less than 4% (n = 20). The analyte of interest was found to be linear over a range from 100 ppbw to 250 ppmw with a correlation coefficient of 0.999 and detection limit of 50 ppbw. Spike recoveries from 500 ppbw to 250 ppmw for all analytes range from 96 to 102%. As an illustration of performance, an industrial coal tar sample, diluted with recycled solvent was analyzed with the method described. Figure 4 shows the chromatogram obtained. Phenol, cresol, xylenols, and trimethylphenol were identified in the sample.
hydrocarbons. Extraction efficiency of phenol approached 100% while cresols and xylenols were at least 95% or higher under the established conditions.
■
ASSOCIATED CONTENT
S Supporting Information *
Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +61-3-6226-7656. Fax +61-3-6226-2858. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Special thanks to Dr. Taylor Hayward for his help in preparing standards. Jeff Mason, Vicki Carter, and Andy Szigety of The Dow Chemical Company, Analytical Technology Center were acknowledged for their encouragement and support. Robert Shellie is the recipient of an Australian Research Council Australian Research Fellowship (Project Number DP110104923).
■
REFERENCES
(1) Gardziella, A.; Pilato, L.; Knop, A. Phenolic Resins−Chemistry, Applications, Standardization, Safety, and Ecology, 2nd ed.; Springer Publishing: New York, 2001. (2) Pilato, L. Phenolic Resins: A Century of Progress, Springer Publishing: New York, 2010. (3) Wright, W.; Kumar, A. Hydrocarb. Process. 2005, 45, 71−76. (4) Busca, G.; Berardinelli, S.; Resini, C.; Arrighi, L. J. Hazard. Mater. 2008, 160, 265−288. (5) Babich, H.; Davis, D. Regul. Toxicol. Pharmacol. 1981, 1, 90−109. (6) Nollett, L. Handbook of Water Analysis; CRC Press: Boca Raton, FL, 2007. (7) Rappoport, Z. The Chemistry of Phenols; John Wiley & Sons Ltd.: West Sussex, U.K., 2003. (8) Rodriguez, I.; Llompart, M.; Cela, R. J. Chromatogr., A 2000, 885, 291−304. (9) Jiang, H.; Fang, Y.; Fu, Y.; Guo, Q. J. Hazard. Mater. 2003, 101, 179−190. (10) Matar, S.; Hatch, L. Chemistry of Petrochemical Processes; Butterworth-Heinemann: Woburn, MA, 2001. (11) Grimmett, H.; Anderson, E. U.S. Patent 2,807,605, September 24, 1957 (12) Munch, J. Determination of Phenols in Drinking Water by Solid Phase Extraction and Capillary Column Gas Chromatography/Mass Spectrometry (GC/MS), Method 528, U.S. Environmental Protection Agency: Cincinnati, OH, 2000. (13) Pavlin, M. Extraction of phenolics from hydrocarbons. U.S. Patent 4,503,267, March 5, 1985. (14) Kimberlin, C.; Mattox, W. U.S. Patent 2,902,428, September 1, 1959. (15) Dawes, P.; Barnett, B.; Hibbert, R. Proceedings of the 34th International Symposium on Capillary Chromatography, Riva Del Garda, Italy, May 30−June 4, 2010, Sandra, P., Ed.; 2010; poster. (16) Freeman, R.; Dawes, P.; Barnett, B.; Hillbert, R. Proceedings of the 35th International Symposium on Capillary Chromatography, San Diego, CA, May 1−5, 2011; CASSS: Emeryville, CA, 2011; poster. (17) Luong, J.; Gras, R.; Shellie, R. A.; Cortes, H. J. Sep. Sci. 2013, 36, 182−191.
■
CONCLUSIONS A practical analytical approach has been successfully developed for the characterization of phenol, cresol, xylenols like 2,5dimethylphenol and 3,5-dimethylphenol, alkyl phenols like 4ethylphenol, and 2,3,5-trimethylphenol in organic matrixes. The technique employs a single-step extraction of phenolic compounds with MEG and sonication followed by MDGC/ MS in SIM mode for the detection of the compounds of interest. Neither GC/MS in SIM mode following MEG extraction nor MDGC/MS in SIM mode without MEG extraction provided sufficient sample cleanup for this challenging analysis. The former suffered from system cleanliness problems, while the latter was affected by excessive matrix interference (Figure S3 in the Supporting Information). The use of MEG as an extraction medium was found to be highly effective in removing phenol, cresols, and xylenols from E
dx.doi.org/10.1021/ac400981z | Anal. Chem. XXXX, XXX, XXX−XXX
