A P P L I C AT I O N N O T E Liquid Chromatography Author: Wilhad M. Reuter PerkinElmer, Inc. Shelton, CT Analysis of Phenolic Antioxidants in Edible Oil/Shortening Using the PerkinElmer Altus UPLC System with PDA Detection Introduction Phenolic antioxidants are commonly used in food to prevent the oxidation of oils. Oxidized oil and fats cause foul odor and rancidity in food products, which is a major cause for concern to the food industry. Globally, regulations vary, but current maximum allowable levels are as low as 100 µg/g (100 ppm). This application note presents a UHPLC method for the analysis of the ten most common phenolic antioxidants that may be found in such products. The application was carried out with minor modifications to the AOAC Official Method 983.15 (1). This method applies to the analysis of finished food products. A 2.7-μm SPP (superficially porous particle) C18 column was used, allowing one to achieve very high throughput at a back-pressure considerably lower than that for UHPLC columns. This method was then applied to a commercial vegetable shortening product, which per label claim, was reported to contain at least one of the antioxidants being analyzed. Method conditions and performance data, including linearity and repeatability, are presented. Solvents, Standards and Samples All solvents and diluents used were HPLC grade and filtered via 0.45-µm filters. Experimental Hardware/Software For all chromatographic separations, a PerkinElmer® Altus™ UPLC® System was used, including the Altus A-30 Solvent delivery Module, Sampling Module, A-30h Column Module and PDA (photodiode array) Detector with a 10-mm path-length flow cell. All instrument control, analysis and data processing was performed using the Waters® Empower® 3 Chromatography Data Software (CDS) platform. Method parameters Using a 100-mL volumetric flask, a 100-ppm stock standard was made up by dissolving 10 mg of each of the ten antioxidant standards in methanol and then bringing the flask up to the mark with methanol. Individual calibrant standards were prepared using the 100-ppm stock solution. The HPLC method parameters are shown in Table 1. Table 1. UHPLC Method Parameters HPLC Conditions Column: PerkinElmer Brownlee™ 2.7 µm 2.1 x 100 mm C18 (Part# N9308404) Mobile Phase: Solvent A: Water; Solvent B: Acetonitrile Solvent program: Flow Rate 0%A (mL/min) Time (min) The phenolic antioxidant standard kit #2 (catalog# 40048U) was obtained from Supelco® (Irvine, CA). This included nordihydroguaiaretic acid (NDGA), propyl gallate (PG), octyl gallate (OG), lauryl gallate (dodecyl gallate (DG)), 2-tert-butyl4-hydroxyanisole (BHA), 2,6-di-t-butyl-4-hydroxymethylphenol (Ionox 100), tert-butylhydroquinone (TBHQ), 3,5-di-t-butyl4-hydroxytoluene (BHT) and ethoxyquin. In addition, a 2,4,5-trihydroxybutyrophenone standard (THBP; catalog# 26201-X9) was obtained from SynQuest® (Alachua, FL). %B %C %D Curve 1 Initial 0.600 60.0 40.0 0.0 0.0 Initial 2 4.50 0.600 45.0 55.0 0.0 0.0 6 3 7.00 0.600 18.0 82.0 0.0 0.0 6 4 10.00 0.600 18.0 82.0 0.0 0.0 6 5 10.10 0.600 60.0 40.0 0.0 0.0 11 The sample (“Sample X”) was a commercially available vegetable shortening purchased at a local food market. The sample was prepared by dissolving 3 grams of Sample X in 15 mL of hexane in a 50-mL centrifuge tube and vortexing for 5 minutes. The resulting solution was then extracted with three 30-mL portions of acetonitrile, combining the three extracts into a 250-mL evaporation dish. The combined extract was evaporated down to 1-2 mL and reconstituted to 6 mL with methanol. ChromatogramOnly Report Prior to injection, all calibrants and samples were filtered through 0.22-µm filters to remove small particles. Equil. Time (“Next inj. Delay Time”): 3 minutes Analysis Time: 10 min. Flow Rate: 0.6 mL/min. (maximum pressure during run: 6600 psi) Oven Temp.: 35 ºC Detection: Altus A-30 PDA; wavelength channels: 280 and 220 nm Injection Volume: 1 µL Results and Discussion Figure 1 shows the chromatographic separation of the 10 phenolic antioxidants in under nine minutes. Figure 2 shows the overlay of 10 replicate 50-ppm standard injections, THB P 0.30 PG 0.35 0.25 AU 0.20 DG B HT Ionox 100 B HA 0.05 E thoxyquin TB HQ 0.10 NDGA OG 0.15 0.00 0.00 1.00 2.00 3.00 4.00 5.00 Minutes Figure 1. Chromatogram of 50-ppm phenolic antioxidant standard; wavelength = 280 nm. 2 6.00 7.00 8.00 9.00 10.00 demonstrating exceptional reproducibility. Retention time % RSDs ranged from 0.10 (early eluters) to 0.03 (later eluters). In a previous application note (2), it has been noted that ethoxyquin may not be well detected at 280 nm. However, we did not observe this, and we could easily detect the analyte at 5-ppm levels. The same injection was also captured on a separate channel, set to 220 nm, as shown in Figure 3. At this wavelength, it is evident that the ethoxyquin has approximately two times the signal intensity. However, this additional signal intensity was not really required here, as current maximum allowable concentrations for phenolic antioxidants only go down to 100 ppm, which was easily handled at 280 nm. 0.30 0.25 0.20 AU 0.15 0.10 0.05 ChromatogramOnly Report 0.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 Minutes Figure 2. Overlay of 10 replicates of 50-ppm check standard; wavelength = 280 nm. 0.80 PG 0.90 0.70 P roject Name: E thoxyquin Reported by User: S ystem Report Method: Overlay Report 0.60 Report Method ID: 1803 Page: 1 of 1 0.50 B HT DG Ionox 100 B HA 0.10 NDGA 0.20 THB P TB HQ 0.30 OG AU 0.40 P henolic Antioxidants Date P rinted: 1/22/2015 11:28:36 AM US /E astern 0.00 -0.10 -0.20 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 Minutes Figure 3. Chromatogram of 50-ppm phenolic antioxidant standard; wavelength = 220 nm. 3 P rocessing Method: P henolicAntoxidant S ample280nm P roject Name: P henolic Antioxidants P rocessing Method ID: 1827 S ystem: Wilhads First S ystem Channel: P DA Ch1 [email protected] Calibration ID: 1839 P roc. Chnl. Descr.: **** Figure 4Date shows three representative calibration resultsAM overE Sa Tconcentration range of 5 to 100 ppm. All ten components had linearity Calibrated: 1/22/ 2015 10:21:24 coefficients > 0.999 (n = 3 at each level). C a libra tion P lot 250000 TBHQ X Value Level Name 1 E thoxyquin Level 1 3 E thoxyquin Level 1 2200000 E thoxyquin 4 E thoxyquin 5150000 E thoxyquin 6 Area 7 E thoxyquin E thoxyquin Level 1 Level 2 Level 2 Level 2 Level 3 5.000000 5.000000 5.000000 10.000000 10.000000 10.000000 25.000000 Peak: Ethoxyquin R esponse Calc. Value % Deviation Manual Ignore 4180.550000 4414.930001 4278.369980 10363.410017 10201.120056 10215.014999 28429.939914 4.318036 -13.639 Yes No 4.404122 -11.918 Yes No 9.616364 -3.836 4.524300 9.759185 9.628592 25.658410 -9.514 Yes -2.408 Yes Yes -3.714 Yes 2.634 Yes 8100000 E thoxyquin Level 3 25.000000 28483.275113 25.705347 2.821 Yes P rocessing Method: P henolicAntoxidant S ample280nm P roject Name: 9 thoxyquin Level 3 25.000000 28456.070181 25.681406 2.726 Yes P rocessing Method EID: 1827 S ystem: 10 E thoxyquin Level 4 50.000000 56468.419900 50.333325 0.667 Yes Channel: P DA Ch1 [email protected] Calibration ID: 11 50000 E thoxyquin Level 4 50.000000 56591.655240 50.441777 0.884 Yes P roc. Chnl. Descr.: **** Level 4 50.000000 57545.764468 12 E thoxyquin 51.281429 2.563 Yes Date Calibrated: 13 E thoxyquin 1/22/ 2015 10:21:24 AM E S T 99.936509 Level 5 100.000000 112833.269883 -0.063 Yes 14 15 1 0 E thoxyquin E thoxyquin Level 5 Level 5 100.000000 100.000000 X Value Ionox 100 Level 1 5.000000 Ionox 100 Level 1 4 Ionox 100 Level 2 3 Ionox 100 Level Ethoxyquin 1 5.000000 120000.0 Reported by User: S100000.0 5ystem Ionox 100 Level 2 6 Ionox Report Method: WMR Cal100 Report Level 2 80000.0 7 Ionox 100 Level 3 Report Method ID: 1860 1860 8 Ionox 100 Level 3 Page: 1 of 1 60000.0 Area 99.586066 -0.875 Yes -0.414 Yes No No No No No No No WMR Cal Report P henolic Antioxidants Wilhads First S ystem 1839 No No No No R2=0.999834 4364.180067 C a libra tion P lot-11.676 4.416189 Yes No -7.682 Yes No 100.00 Name 5.000000 TB HQ; E4539.660024 quation Y = 2.20e+003-8.457 X -1.05e+003; R^2 0.999834 4.577142 Yes No 2 9 112435.055810 99.125131 No 100 20.00 Peak: Ionox40.00 60.00 80.00 CONC. R esponse Calc. Value % (PPM) Deviation Manual Ignore 0.00 Level Name 111911.288762 No Ionox 100 Level 3 10 Ionox 100 Level 4 40000.0 11 Ionox 100 Level 4 12 Ionox 100 Level 4 20000.0 10.000000 10.000000 10.000000 25.000000 25.000000 25.000000 50.000000 50.000000 50.000000 4581.929966 10318.269980 10297.460106 10161.870040 27359.725085 27233.360184 27073.160146 54778.930195 54755.159915 54701.550188 13 Ionox 100 Level 5 100.000000 108693.381248 14 Ionox 100 Level 5 100.000000 107803.295527 0.0 15 Ionox 100 Level 5 100.000000 108025.861196 4.615913 9.877366 9.858279 -1.226 Yes No P rojectNoName: -1.417 Yes 9.733914 -2.661 Yes No 25.392130 1.569 Yes No 25.508034 25.245193 50.657324 50.635521 50.586350 100.108435 99.292035 99.496176 2.032 Yes 0.981 Yes 1.315 Yes 1.271 Yes 1.173 Yes 0.108 Yes -0.708 Yes -0.504 Yes No P henolic Antioxidants Date P rinted: 2/9/2015 3:01:31 P M US /E astern No No No No No No No R2=0.999686 -20000.0 0.00 20.00 40.00 60.00 CONC. (PPM) 80.00 100.00 Name E thoxyquin; E quation Y =tion 1.14e+003 C a libra P lot X -7.26e+002; R^2 0.999686 120000.0 Ionox 100 Area Reported by User: S ystem Report Method: WMR Cal Report 100000.0 1880 Report Method ID: 1880 Page: 1 of 1 80000.0 P roject Name: P henolic Antioxidants Date P rinted: 2/9/2015 3:03:16 P M US /E astern 60000.0 40000.0 20000.0 0.0 R2=0.999824 -20000.0 0.00 20.00 40.00 60.00 80.00 CONC. (PPM) 100.00 Name Ionox 100; E quation Y = 1.09e+003 X -4.51e+002; R^2 0.999824 Figure 4. Three representative results of 5-level calibration sets for the phenolic antoxidants; wavelength = 280 nm. 4 Reported by User: S ystem Report Method: WMR Cal Report P roject Name: P henolic Antioxidants Date P rinted: 0.18 0.16 AU Figure 5 shows the chromatographic results of Sample X overlaid with the 50-ppm standard. A peak eluting at exactly the time 0.14 of TBHQ (tert-butylhydroquinone) was observed. This was consistent with the product label claim. By back-calculating the 0.12 concentration in the original sample, it was determined that Sample X contained approximately 12-ppm of TBHQ. The actual concentration could not be verified as it was not provided in the 0.10 product’s label claim. Per Figure 6, upon closer examination of the chromatogram of Sample X, a small peak at about 8.23 minutes was also observed. This matched the elution time for DG (dodecyl gallate) in the standard mix. If this was indeed DG, its concentration was Overlay R eport below the calibration curve, estimated to be <0.5 ppm. Further verification of the identity of this peak was not pursued. 0.080.26 0.24 0.060.22 0.20 0.04 TBHQ 0.18 0.02 0.16 0.14 AU 0.000.12 0.10 -0.020.08 0.06 0.00 1.00 0.04 2.00 3.00 4.00 5.00 Minutes 0.02 6.00 0.00 -0.02 0.00 1.00 2.00 3.00 4.00 5.00 Minutes 6.00 7.00 8.00 9.00 10.00 ChromatogramOnly R eport Figure 5. Chromatogram of Sample X (blue) overlaid with 50-ppm standard (black); wavelength = 280 nm. R eported by User: S ystem 0.035 R eport Method: Overla y R eport R eported by User: S ystem R eport Method: Overla y R epor R epor t Method ID:t 1803 R eport Method ID: 1803 P age: 1 P age: 1 of 1 of 1 0.030 P roject Na me: P henolic Antioxidants Da te P rinted: 1/22/2015 ChromatogramOnly R eport TB HQ 5:08:17 P M US /E astern 0.025 0.020 AU 8.10 8.15 8.20 8.25 8.30 8.35 Minutes 8.40 8.45 8.55 8.50 8.60 8.65 0.015 0.010 R e ported by User: S ystem R e port Me thod: Chroma togra mOnly R eport R e port Me thod ID: 1851 1851 P age: 1 of 1 P roject Na me: P henolic Antioxida nts Da te P rinted: 1/22/ 2015 10:55:55 AM US /E a stern DG 0.005 0.000 0.00 1.00 2.00 3.00 4.00 5.00 Minutes 6.00 Figure 6. Chromatogram of Sample X with zoomed in area just after 8 minutes; wavelength = 280 nm. 7.00 8.00 9.00 10.00 5 Conclusion References This work has demonstrated the effective chromatographic separation of ten phenolic antioxidants using a PerkinElmer Altus UPLC® with a PDA detector and the Empower® 3 CDS system. The results exhibited excellent retention time repeatability as well as exceptional linearity over the tested concentration ranges. At an analytical wavelength of 280 nm, the sensitivity for all 10 phenolic antioxidants was found to be more than adequate to accommodate the current maximum allowable concentration limit of 100 ppm. 1. Official Methods of Analysis, Method 983.15, Association of Official Analytical Chemists (AOAC), Arlington, VA USA We were able to identify and quantitate the phenolic antioxidant content in a commercial vegetable shortening product and the results matched the label claim of the manufacturer. 2. Monitor Antioxidant Additives in Foods, Using HPLC (AOAC Method 983.15) or Capillary GC and a Supelco Reference Standards Kit, Application Note 78, Sigma-Aldrich/ Supelco, 2004 3. Analysis of Common Antioxidants in Edible Oil with the PerkinElmer Flexar™ FX-15 System Equipped with a PDA, Application Note, PerkinElmer, Inc. From a food quality perspective, considering the ever growing emphasis on food monitoring, this application is intended to serve as a valuable guide for the monitoring of edible oils/shortening. It should be noted that in the U.S., per label claims, only some of the vegetable shortenings reported any amount of phenolic antioxidant. None of the edible oils that were found in stores reported any phenolic antioxidants. However, although only edible vegetable shortening was tested for this study, the provided sample preparation procedure and chromatographic application easily lend themselves to the analysis of edible oils as well. PerkinElmer, Inc. 940 Winter Street Waltham, MA 02451 USA P: (800) 762-4000 or (+1) 203-925-4602 www.perkinelmer.com For a complete listing of our global offices, visit www.perkinelmer.com/ContactUs Copyright ©2015, PerkinElmer, Inc. All rights reserved. PerkinElmer® is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners. 012086_01PKI