APPLICATION NOTE Gas Chromatography/ Mass Spectrometry Authors: Sharanya Reddy Thomas Dillon PerkinElmer, Inc. Shelton, CT Profiling of Aroma Components in Wine Using a Novel Hybrid GC/MS/MS System Introduction The global wine industry is an established market with more than one million wine makers globally. The present market size is approximately three billion U.S. dollars with an annual growth rate of ~3%. Aroma profiling of wine is important for wine formulation and quality control because aromas are known to contribute to the organoleptic/sensory properties of wine. Wine aroma is determined by hundreds of volatile compounds present at a wide range of concentrations. Characterization of these aroma contributing compounds can help with wine characterization and therefore with quality control of wine. The AxION iqt GC/MS/MS is a novel hybrid mass spectrometer system that offers full scan mass spectral information with excellent sensitivity at fast data acquisition speeds. It is ideal for detection of components present at both trace levels and at high concentrations compared to scanning instruments such as quadrupoles.
The ecipher data analysis package, provided with the instrument, includes a deconvolution algorithm with automatic "spectral library" database searching for identification of components in complex matrices including wine. The results obtained can be imported into the PerkinElmer TIBCO Spotfire Software for Principle Component Analysis (PCA) to identify the chemical components that are unique to the wine varieties. Experimental Sample Preparation Wine (Merlot, Chardonnay or Rosé, 20 ml each) was mixed with dichloromethane (10 ml) and stirred at room temperature for 30 min. The samples were centrifuged at 3500 RPM. The dichloromethane layer was carefully removed into a centrifuge tube containing anhydrous sodium sulfate (2 g), vortexed for 0.5 secs and centrifuged at 3500 RPM for 10 min. The organic layer was removed, dried under nitrogen gas to 1 ml and injected (1 µl). Results and Discussion A full scan chromatogram of Chardonnay wine analyzed by the AxION iqt system is shown in Fig. 1. Deconvolution of data by ecipher resulted in 107 peaks being detected, including peaks representing <1% of the total ion current. The user set thresholds for peak area and signal-to-noise (S/N) determine Table 1. GC conditions. Gas Chromatograph PerkinElmer Clarus 680 Injector Type: Injector Temperature: 250 C Injection: Oven Program: Analytical Column: Carrier Gas: Table 2. MS conditions. Mass Spectrometer Programmable Split/Splitless 1 μl, splitless 40 C, hold for 2 min., ramp to 180 C at 12.5. C/min., ramp to 240 C/min. and hold for 5 min. PerkinElmer ELITE- 5 MS, 30 m X 0.25 mm X 0.25 μm 1 ml/min. Helium PerkinElmer AxION iqt MS/MS GC Transfer Line Temperature: 250 C Source Temperature: 250 C Acquisition Time: 0.2 sec Mass Range: 40-400 Ion Source: Solvent Delay: the number of peaks detected by the deconvolution software. The spectra of peaks detected were matched against the NIST 1 library database. Table 3 shows representative components tentatively identified in white wine. EI 3 min. 160000000 140000000 white wine-chard-1_r1[ms], Time 13.4013 mins, Scan# 2766, 8.94e+005, Relative 0.52% All Channels 1000000 120000000 900000 100000000 80000000 800000 700000 600000 60000000 500000 400000 [75] 13.37 40000000 13.32 13.34 13.36 13.38 13.40 13.42 13.44 13.46 13.48 13.50 20000000 [18] 4.29 [29] 5.47 [35] 6.72 [47] 9.61 [1] [42] [99] [112] 3.10 7.88 15.88 18.74 0 ^ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Figure 1. Peaks identified in Chardonnay (in blue) with thresholds settings for S/N at 10 and area of 500. Inset shows peaks identified at <1% of the total ion current. [60] 10.78 [66] 12.07 [75] 13.37 Reten on me (min) [85] 14.59 [110] 17.17 2
Table 3. Representative components tentatively identified in Chardonnay by matching against NIST library. RT Found % Area Match Reverse Match Name 4.364 0.31 928 928 1-Propanol 4.480 0.30 840 840 Butanoic Acid, ethyl ester 5.031 2.99 912 912 1-Propanol, 2-methyl- 5.531 1.51 918 918 1-Butanol, 3-methyl-, acetate 6.526 8.03 855 855 1-Butanol, 3-methyl- 8.093 1.71 800 828 ETHYL (S)-(-)LACTATE 8.186 0.33 878 878 1-Hexanol 8.440 0.08 907 907 1-Propanol, 3-ethoxy- 9.457 0.03 819 819 2H-Pyran, 3,4-dihydro-6-methyl- 10.080 0.06 752 952 1,2-Ethanediol, monoacetate 10.099 2.10 810 810 2,3-Butanediol, [R-(R*,R*)]- 11.087 0.50 917 917 Butanoic Acid, 4-hydroxy- 12.964 1.68 884 884 Hexanoic acid 13.292 0.16 894 894 L-Arginine, N2-[(phenylmethoxy)carbonyl]- 14.028 0.07 772 911 4H-Imidazol-4-one, 2-amino-1,5-dihydro- 14.211 0.37 945 945 2H-Pyran-2,6(3H)-dione 14.637 2.02 847 847 Butanedioic acid, hydroxy-, diethyl ester 11.523 0.49 833 850 Butanedioic acid, diethyl ester 17.903 0.12 813 813 S-Hydroxymethylfurfural 18.967 0.12 914 914 Ethanone, 1-(3-hydroxy-4-methoxyphenyl)- 19.106 0.39 799 814 Propan-2-one, I-(4-isopropoxy-3-methoxyphenyl)- 22.719 2.07 810 815 Homovanillic acid 22.892 0.13 897 970 1-Cyclooctene-1-acetic acid, à, à-dimethyl- Peaks detected in Chardonnay were saved as a target library and the peaks detected in Merlot were matched against this list. The compounds common to both wines and unique to each wine can be sorted by the software with color coding (Fig. 2). 28000000 26000000 24000000 = Common to both wines = Chardonnay, not Merlot = Merlot, not Chardonnay 22000000 20000000 18000000 16000000 14000000 12000000 10000000 8000000 6000000 4000000 2000000 0 [98] [97] 6.53 [103] [108] [117] [138] 9.08 11.06 5.00 [100] [129] [131] [107] [113] 12.93 16.59 [158] [171] [174] 7.95 13.94 15.55 10.06 12.00 17.57 18.63 20.86 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Reten on me (min) [176] 22.68 Figure 2. Peaks identified in Merlot were matched against compounds in Chardonnay. The compounds common to both wines are shown in green. The compounds exclusive to Merlot and Chardonnay are marked in blue and red color respectively. 3
A reference target list was generated to include peaks detected and identified in replicate samples of Merlot, Chardonnay and Rosé wine. Each of the replicate injections (n=6) was compared against this reference list and the data matrix generated was exported into Excel for PCA analysis by TIBCO Spotfire software (Fig. 3). The PCA plot shows good separation of the grape varieties based on the six replicates analyzed for each sample. The loadings plot (Fig. 3) helped identify the chemical compounds that are unique to each of the wines. For example, it was observed that diethyl malate, 2, 4-hexadienoic acid, and tryptophol are present at higher concentrations in Rosé compared to the other two wine types. The compounds hydroxy cinnamic acid ethyl ester, diethyl succinate and 3,5-dimethoxy-4- hydroxy benzoic acid contribute to the uniqueness of Merlot while ethyl lactate and 2-(2-Hydroxyethyl)phenol is higher in Chardonnay compared to the other two wines. Compounds mapped out in the center of the loadings graph are present in all three wine varieties at similar concentrations. The data of the loading graph for the compounds phenylethyl alcohol, ethyl lactate and dimethyl malate is displayed as bar charts in Fig. 4 further emphasizing the chemical difference between the wine varieties. The bar charts confirm the excellent reproducibility between replicate injections. Rosé Merlot Chardonnay 2,4-hexadienoic acid diethyl malate tryptophol 3,5-dimethoxy-4- hydroxy benzoic acid ethyl lactate hydroxy cinnamic acid ethyl ester diethyl succinate 5-(3-hydroxypropyl)- 2,3-dimethoxyphenol 2,3-butanediol 2-(2-Hydroxyethyl)phenol Figure 3. PCA analysis of wine samples using TIBCO Spotfire software. The loadings graph on the right hand side maps the chemical space to the wine variety. phenylethylalcohol Rosé Merlot Chardonnay ethyl lactate diethyl malate 4 Figure 4. Intensities of phenylethyl alcohol, ethyl lactate and diethyl malate plotted for multiple injections of Chardonnay, Merlo and Rosé are shown as bar charts.
Discussion The AxION iqt GC/MS/MS system in full scan mode can identify trace level analytes (low ppb) in a complex matrix such as wine. The ecipher peak deconvolution software in conjunction with the PerkinElmer PCA TIBCO Spotfire software can be used to profile wine aromas and to identify differences between the wine varieties. References 1. NIST/EPA/NIH Mass Spectral Library, NIST Standard Reference Database 1A, 2014. 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. 012443_01 PKI