TN-89 Fast and Robust Analysis of Organic s from Wine using HPLC-UV Brian Rivera Product Manager In addition to chromatography, Brian also has a passion for ice cream-making, and enjoys experimenting with bold, new flavors. Laura Snow, Matt Trass, Michael Klein, Sean Orlowicz, and Brian Rivera Phenomenex, Inc., Madrid Ave., Torrance, CA 9 USA Introduction Organic acids are a major contributor to the stability, flavor, color, and balance of wine. Organic acids are found in both wine and the grapes from which they are derived. Of particular importance are Tartaric and Malic acids. Tartaric acid contributes largely to a wine s acidity and can form crystalline deposits, which affect red wines more than white wines. This is the preferred acid addition to wine because it is not as easily degraded by microbes. Another critical acid is Malic acid, which is metabolized through the ripening stage and drops in concentration after harvest. Acetic, Citric, and Succinic acids are also critical acids in wine and contribute to the overall acidity and flavor profile for both red and white wines. Previous studies have shown that HPLC can be used as a method for the quantitation of organic acids, which can be used for both fermentation monitoring and quality control,. Typical separation modes by HPLC include ion-exclusion and reversed phase. Ion-exclusion, though simple and straightforward, is limited in experimental design space and can result in run times exceeding minutes. Reversed phase allows for more flexibility in method development and relatively faster analysis times, but can present challenges since organic acids are relatively polar compounds. As such, column selectivity plays a critical role in development of a quantitative reversed phase method for organic acids. In this study, we present two orthogonal methods for the testing of organic acids in red and white wines. The first method uses a Kinetex F coreshell pentafluorophenyl phase HPLC column and is an appropriate screening method for the identification of some critical acids, such as Tartaric and Lactic acid. However, for full separation of all relevant organic acids, a Synergi Hydro-RP column with enhanced polar selectivity is needed. Reasonable run times are below minutes and the following organic acids can be accurately quantitated: Tartaric acid, Malic acid, Lactic acid, Citric acid, Acetic acid and Succinic acid. Additionally, Shikimic acid, a known interference, can also be separated. Materials and Methods Standards were purchased from Sigma Aldrich. Red and white wines were purchased from a local grocer. Standards were prepared in water, then diluted : in mobile phase A prior to injection. Wine samples were also diluted : in mobile phase, then filtered with a Phenex. µm syringe filter (P/N AF--) prior to injection onto the HPLC system. Results and Discussion The initial screening method used a Kinetex. µm F core-shell column. Although the column performed well with neat standards, co-elution of interference peaks in red and white wine samples complex indicated this column might not be optimal for quantitative organic acid sample analysis. Specifically, the known interferent Shikimic acid coeluted with Malic acid (Figure ). Nonetheless, the column did show a good proof of concept for a simple chromatographic screen that could be used for the quantitation of certain organic acids, such as Tartaric acid, or if the analyst is not particularly interested in quantitation of Malic acid. Screening Method using Kinetex. µm F Column: Kinetex. µm F, Å Dimensions: x. mm Part No.: F-7-E Mobile Phase: mm Potassium Phosphate, ph. Flow Rate:. ml/min Temperature: Ambient Detection: UV @ nm Injection: µl Sample:. Tartaric. Shikimic. Malic. Lactic. Acetic. Citric 7. Succinic Figure. Standards Screening Method using Kinetex. µm F Core-Shell Column Norm. 7 7 HPLC analysis was performed on an Agilent system with a DAD detector and data was collected using ChemStation software (Agilent, Santa Clara, CA. USA). 7 App ID 7.... min For additional technical notes, visit www.phenomenex.com Page of 8
TN-89 Further method development was then performed on a Synergi µm Hydro-RP column. This column was selected due to its stationary phase hydrogen bond donating capabilities. To prevent carryover of late eluting matrix interferences, a gradient clean up step was implemented. Although the total run time was extended to minutes, the analysis was still a shorter run time than previously reported methods. An acidic mobile phase with a ph of. was required to ensure protonation of the organic acids, thus helping in retention by reversed phase mechanisms. Again, this is primarily based upon the hydrogen bond donating capabilities of the polar endcapped Synergi Hydro-RP column. Finally, temperature was another parameter that was investigated during method development and helped in the improved resolution of Shikimic acid, and thus the higher range of the column stability ( C) was used. Figure. Organic Standard (Calibrator Level - Low) min 7 App ID Final Optimized HPLC-UV Method using Synergi µm Hydro-RP Column Column: Synergi µm Hydro-RP 8Å Dimensions: x. mm Part No.: G-7-E Mobile Phase: A: mm Potassium Phosphate, ph. B: Methanol Gradient: Time (min) % B...9. Flow Rate: ml/min Temperature: C Detection: UV @ nm (ambient) Injection: µl Sample:. Tartaric. Malic. Shikimic. Lactic. Acetic. Citric 7. Succinic 7 7 Figure. Organic Standard (Calibrator Level - High) 7 App ID 9 min After method optimization, a five point calibration curve was generated and a linear regression was used to determine the organic acid content for red and white wines. Figures and show calibrator levels at lowest and highest concentration. All sample components are baseline resolved, allowing for the easy quantitation of organic acids. Data for organic acid standards are summarized in Table. Using peak height of the standards run in triplicate, linearity was determined for each organic acid. A correlation coefficient of >.99 for all standards was observed and % RSD for all standards were below % with the exception of one data point for Citric acid (7.%). Succinic acid was included in the standard mix. Although linearity for the standards was determined to be. to. mg/ml, matrix interferences did not allow for quantitative measurements when samples were evaluated. Page of 8
TN-89 Table. Organic Standards (Collected data) Sample Number Standard Rep Rep Rep Mean % RSD Calculated Conc. (mg/ml) Expected Conc. (mg/ml) % Error for Standards 7 Succinic Citric Acetic Lactic Shikimic Malic Tartaric Standard 8.8 8. 8.7 8.....7 Standard 7. 7 7. 7.7...8 Standard 9.9...... -.7 Standard... 8..8.77.7.8 Standard 8 8. 8. 8.9... Standard..8...7...9 Standard....97.9.. Standard..... -.7 Standard..9.7.9.79.7.98 Standard....77... Standard.....9...79 Standard 7. 7 7 7..... Standard..9.7.9.8.. -.9 Standard..8...7.7. Standard.9.....8 Standard 9. 9. 9. 9.....7 Standard 8.9 9 9. 9..... Standard 9. 9. 9. 9..7.. -. Standard 8. 7.8 9. 8.7..88.87.8 Standard 99. 97. 99. 98.7....8 Standard.8.7.9.8.... Standard....8... Standard 79. 79. 79. 79.7.9.. -. Standard. 9.9 9.9...8.87 -.8 Standard.7.7..7.... Standard..7..9.8.9. 7. Standard.7.8..7..8.. Standard...7..9. -.8 Standard 99 99. 99.7..89.87.79 Standard.9..7.7.... Standard 7. 7. 7. 7.7....7 Standard..9...7... Standard 8. 8. 87 8.... -.9 Standard 8.7 9.7.7 9.7.77.8.87 -.7 Standard 7. 7.7 7.8 7..... For additional technical notes, visit www.phenomenex.com Page of 8
TN-89 The eight samples of red and white wines were run in triplicate. Organic acids were identified based upon retention time. Concentrations for organic acids were then determined for each organic acid of interest. Table indicates the calculated organic acid concentration in each tested wine. Organic acid concentrations calculated as below their respective standard concentration is indicated with a. Matrix interferences also contributed to the inability to quantitate some organic acids. This is especially true with the late eluting succinic acid. Figures - illustrate the chromatogram for each of the wine samples. All confirmed peaks are labeled, based on known retention times. Table : Calculated Concentrations for Organic Standards (mg/ml) Wine Tartaric Malic Shikimic Lactic Acetic Citric Succinic Mixed Red...8.7.78 Cabernet Sauvignon C.7.8.. Cabernet Sauvignon P.8.8.9.7 Bordeaux.9.7..8 Pinot Noir.8.77.7. Chardonnay.7.8.7 Sauvignon Blanc...79... Chablis.8.9.88.79. excluded because concentration is below standard level. Figure. Mixed Red Figure. Cabernet Sauvignon C 8 8 7 7 App ID App ID min min Page of 8
TN-89 Figure. Cabernet Sauvignon P Figure. Bordeaux 8 App ID App ID 7 min min Figure 7. Pinot Noir Figure 8. Chardonnay 8 8 App ID 7 App ID min min Figure 9. Sauvignon Blanc Figure. Chablis 8 8 7 App ID 7 App ID min min For additional technical notes, visit www.phenomenex.com Page of 8
TN-89 Conclusion Previously reported HPLC methods for the quantitation of organic acids from wine are either limited in selectivity or time prohibitive. In this study, we developed a fast and robust method using the Synergi µm Hydro-RP column to resolve the organic acids of interest in minutes. Including a gradient cleanup step, the total runtime is minutes. If small matrix interferences are not a concern, the method could easily be shortened by simply increasing the flow rate. Finally, should a fast, simple screen be needed, the screening method developed on a Kinetex F core-shell column might be of particular interest, especially if Malic acid is not a concern. Further studies could include identification and quantitation of additional interferences in the complex wine sample matrices. Specificity could then be better assessed. Additionally, other sample preparation methods such as solid phase extraction might be investigated to further improve quantitation and robustness of the assay across other wine samples. References. Kliewer, W. M., L. Howarth, and M. Omori. 97. Concentrations of Tartaric and Malic s and Their Salts in Vitis Vinifera Grapes. American Journal of Enology and Viticulture 8:-.. Jackson, Ron S. Wine Science Principles and Applications. rd ed. Amsterdam: Elsevier/Academic, 8. 7-9.. Lopez, E. F., and E. F. Gomez. Simultaneous Determination of the Major Organic s, Sugars, Glycerol, and Ethanol by HPLC in Grape Musts and White Wines. Journal of Chromatographic Science (99): -7.. Kordis-Krapez, M., Abram, V., Kac, M. et al., Determination of organic acids in white wines by RP-HPLC. Food Tech. Biotechnol., 9(), 9 99. We Develop Methods WITH You It s OK if you don t have time to develop or run a method yourself. Our full-service analytical support laboratory will do it for you. www.phenomenex.com/phenologix Page of 8
TN-89 Ordering Information Kinetex SecurityGuard. μm Analytical Columns (mm) ULTRA Cartridges Phases x. x. 7 x. x. x. /pk EVO C8 B-7-E D-7-E F-7-E AJ-99 F B-7-E D-7-E F-7-E AJ-9 Biphenyl B--E D--E F--E AJ-97 XB-C8 B-9-E C-9-E D-9-E F-9-E AJ-878 C8 A--E B--E C--E D--E F--E AJ-878 C8 B-97-E C-97-E D-97-E F-97-E AJ-877 HILIC B--E C--E D--E F--E AJ-877 Phenyl-Hexyl B-9-E C-9-E D-9-E F-9-E AJ-877 for. mm ID SecurityGuard μm Analytical Columns (mm) ULTRA Cartridges Phases x. x. x. x. /pk EVO C8 B--E D--E F--E G--E AJ-99 Biphenyl B-7-E D-7-E F-7-E G-7-E AJ-97 XB-C8 B--E D--E F--E G--E AJ-878 C8 B--E D--E F--E G--E AJ-878 C8 B-8-E D-8-E F-8-E G-8-E AJ-877 Phenyl-Hexyl B--E D--E F--E G--E AJ-877 for. mm ID Core-Shell Technology If Kinetex core-shell columns do not provide at least an equivalent separation as compared to a competing column of the same phase, return the column with the comparative data within days for a FULL REFUND. µm.. µm.7 µm µm Ordering Information Synergi SecurityGuard µm MidBore Columns (mm) Cartridges (mm) Phases x. x. x. x. x.* Max-RP B-7-Y F-7-Y G-7-Y AJ-7 Hydro-RP B-7-Y F-7-Y G-7-Y AJ-7 Polar-RP A--Y B--Y F--Y G--Y AJ-7 Fusion-RP B--Y F--Y G--Y AJ-7 for ID:.-. mm SecurityGuard µm Analytical Columns (mm) Cartridges (mm) Phases x. x. 7 x. x. x. x.* Max-RP A-7-E B-7-E C-7-E F-7-E G-7-E AJ-7 Hydro-RP A-7-E B-7-E C-7-E F-7-E G-7-E AJ-7 Polar-RP A--E B--E C--E F--E G--E AJ-7 Fusion-RP B--E C--E F--E G--E AJ-77 for ID:.-8. mm * SecurityGuard Analytical Cartridges require holder, Part No.: KJ-8 Full Range Selectivity LC Columns If Synergi analytical columns do not provide at least an equivalent separation as compared to a competing column of similar particle size, similar phase and dimensions, return the column with comparative data within days for a FULL REFUND. Additional dimensions available. Contact your Phenomenex HPLC Specialist for more information or visit www.phenomenex.com/synergi For additional technical notes, visit www.phenomenex.com Page 7 of 8
TN-89 Terms and Conditions Subject to Phenomenex Standard Terms and Conditions, which may be viewed at http://www.phenomenex.com/termsandconditions. Trademarks Kinetex is a registered trademark, and Synergim Midbore, SecurityGuard, and Phenex are trademarks of Phenomenex. Agilent is a registered trademark of Agilent Technologies, Inc. Australia t: + ()-98- f: + ()-98- auinfo@phenomenex.com Austria t: + ()-9- f: + ()-9- anfrage@phenomenex.com Belgium t: + () (French) t: + () 8 (Dutch) f: + ()-879 beinfo@phenomenex.com Canada t: + (8) -8 f: + () 8-778 info@phenomenex.com China t: +8 () 8-8 f: +8 () 89-8 chinainfo@phenomenex.com Denmark t: + 8 88 f: + 8 nordicinfo@phenomenex.com Finland t: +8 ()9 789 f: + 8 nordicinfo@phenomenex.com France t: + () 9 f: + () 9 franceinfo@phenomenex.com Germany t: +9 ()-88- f: +9 ()-88- anfrage@phenomenex.com India t: +9 ()- f: +9 ()- indiainfo@phenomenex.com Ireland t: + () 7 f: + -79 eireinfo@phenomenex.com Italy t: +9 7 f: +9 7 italiainfo@phenomenex.com Luxembourg t: + ()-87 f: + ()-879 nlinfo@phenomenex.com Mexico t: -8-8- f: --8-778 tecnicomx@phenomenex.com The Netherlands t: + ()-87 f: + ()-879 nlinfo@phenomenex.com New Zealand t: + ()9-789 f: + ()9-789 nzinfo@phenomenex.com Norway t: +7 8 f: + 8 nordicinfo@phenomenex.com Puerto Rico t: + (8) -HPLC f: + () 8-778 info@phenomenex.com Spain t: + 9--8 f: + 9--9 espinfo@phenomenex.com Sweden t: + ()8 9 f: + 8 nordicinfo@phenomenex.com United Kingdom t: + ()-7 f: + ()-79 ukinfo@phenomenex.com USA t: + () - f: + () 8-778 info@phenomenex.com All other countries Corporate Office USA t: + () - f: + () 8-778 info@phenomenex.com Disclaimer Dimensions and chromatographic conditions are the same for all columns unless otherwise noted. Phenomenex is not affilitated with Agilent. SecurityGuard is patented by Phenomenex. U.S. Patent No.,, CAUTION: this patent only applies to the analytical-sized guard cartridge holder, and does not apply to SemiPrep, Prep, or ULTRA holders, or to any cartridges Phenomenex. All rights reserved. www.phenomenex.com Phenomenex products are available worldwide. For the distributor in your country, contact Phenomenex USA, International Department at international@phenomenex.com TN78_W Page 8 of 8