Analysis of odourant compounds in wine : - With headspace solid-phase microextraction and gas chromatography- mass spectrometry

(1)

Analysis of odourant compounds in wine

-

With headspace solid-phase microextraction and gas chromatography-

mass spectrometry

By Emma Ödmar

Analytical science program in chemistry with focus on forensics

Candidate for Degree of Bachelor of Science School of Science and Technology

Örebro university Spring term 2018

(2)

Table of contents Abstract ... 3 1.Introduction... 4 1.1Aim ... 4 1.2Scope ... 4 1.3Background ... 4 1.3.1 Wine ... 4

1.3.2 Headspace solid-phase microextraction (HS-SPME)... 5

1.3.3 Gas chromatography-mass spectrometry (GC-MS) ... 6

2.Materials and methods... 7

2.1Materials... 7 2.1.1 Chemicals ... 7 2.1.2 Samples ... 8 2.1.3 Sample preparation ... 9 2.1.4 Standard compounds ...10 2.2Methods ...11

2.2.1 Headspace solid-phase microextraction ...11

2.2.2 Gas chromatography-mass spectrometry...12

2.2.3 Method issues ...13

3.Results ...14

3.1Artificial wine standard ...14

3.2Wine analysis result ...15

3.2.1 White wines ...15 3.2.2 Orange wines ...18 3.2.3 Red wines ...19 3.2.4 Sparkling wines ...24 3.3Quantification ...25 3.4Identified compounds ...25 3.5Statistics ...26 4.Discussion ...27 4.1Limitations ...27 4.2 Quantitative analysis ...27

(3)

2 4.3Aroma compounds ...28 4.5Comparison of wines ...29 4.5.1White wines ...29 4.5.2Orange wines ...30 4.5.3 Red wines ...30 4.5.4Sparkling wines ...30 5.Conclusion ...30 6.References ...32 7. Appendix ...34

(4)

Abstract

Wine is a drink that can enhance the flavour experience of food, which is why it is important that the wine’s sensory profile is explained correctly to the consumers. In this study,

headspace solid-phase microextraction gas chromatography mass spectrometry was used to characterise odourant compounds in wine to find chemical markers to explain wine sensory profiles instead of sensory analysis.

The study included 16 different wines, red, white, orange and sparkling, where the nine most abundant peaks in each wine sample were evaluated. Homologue patterns based on areas were used to compare profiles between different wines. When studying homologue patterns for each wine and comparing within wine groups, differences and similarities can be seen. All wine samples contained isoamyl alcohol and the majority of them also contained ethyl decanoate, octanoic acid and decanoic acid. Six out of eight red wines contained ethyl succinate and five of them also contained ethyl hexanoate. All white wine samples showed presence of ethyl octanoate and ethyl hexanoate. The orange wines also contained ethyl octanoate and ethyl hexanoate, along with pentanoic acid. Both sparkling wines contained ethyl octanoate and ethyl hexanoate in addition with phenylethyl alcohol.

However, a more thorough study covering more compounds to identify the less obvious differences of wine would have to be performed for a more precise explanation of the wine’s characterisation and sensory profile. It should be noted that the method of this study does leave room for improvements to improve the quality of the results. For example, since the most abundant compounds are not necessarily the ones with the most powerful odours, quantification based on response of an internal standard would strengthen the study. Additional compounds in the samples could also be further investigated. Statistically the method would also need improvement for satisfactory results regarding reproducibility of the samples.

(5)

1. Introduction

1.1 Aim

A current problem in Sweden is the large volume of destruction of wine that has a deviant sensory profile than what was intended. By combining chemical characterisation and sensory analysis it would be possible to re-label these wines with an appropriate label. Combining this with recommendations of suitable food for the new correct profile the wines could be sold in the stores instead of being sent to destruction.

The aim in this project is to find chemical markers that can be connected to certain sensory profiles, so that chemical profiling could be used instead of sensory analysis for new profiling of wines, with the goal to reduce the large volumes of wine sent to destruction.

1.2 Scope

This project is a part of a larger study and covers the gas chromatography analysis of volatile odourant compounds in wine samples. Liquid chromatography of the same samples will also be performed but will not be included in this project. This and sensory analysis could be used for comparison of results within the study.

1.3 Background

1.3.1 Wine

Wine is a very popular drink that has been produced for thousands of years (1). Wine is obtained when fermenting grape must and reflects a distinguishable flavour. Originally, wine was only produced in Europe, by the Romans and the Greeks, but now the production has spread all over the world where the climate is suitable. Wine is usually appreciated combined with food as it enhances the flavour experience and is also a social phenomenon.

Chemically, wine consists of non-volatile tastants, such as phenolic compounds, and volatile aroma compound, such as phenolic acids (1, 2). The aroma is complex because the aroma compounds have different origins (1). The compounds can be found in different places such as the grape itself, produced during grape processing, alcoholic fermentation and during maturation of the wine.

Consumers need correct description of the product that they are purchasing, but traditional descriptive methods are time demanding (3). There is also issues with vocabulary use to really explain characterisation of products. Multiple studies, using methods such as

headspace solid-phase microextraction gas chromatography mass spectrometry (HS-SPME- GC-MS) and solid phase extraction high performance liquid chromatography (SPE-HPLC),

(6)

to characterise wine by chemical analysis instead of sensory analysis have been performed to create a more time effective and more easily understandable characterisation of the available products (3, 4, 5).

1.3.2 Headspace solid-phase microextraction (HS-SPME)

Solid-phase microextraction is a solvent-free, rapid extraction technique that can be used within different fields such as food analysis, environmental analysis and drug analysis (6). The technique is popular because of its speed, no solvents, extraction and concentration of compounds occur simultaneously, and it is inexpensive (7). SPME is a popular technique for flavour analysis since it is sensitive enough for the volatile compounds with varying boiling points (8). It can either be performed to extract compounds directly out of liquid or from the headspace of the sample, which can be either liquid or solid.

The device consists of a fiber optic rod of fused silica coated with a polymer film, shown in Figure 1 (6, 7). Depending on what fiber and extraction conditions are used, the sensitivity and accuracy can be affected (8). Headspace-SPME (HS-SPME) is depending on the equilibrium of experimental parameters such as sample volume, extraction time, extraction temperature, sample matrix and concentration of volatiles. The headspace mechanism is based on equilibrium of analytes between fiber coating, headspace and sample. This means that the technique is highly dependent on the vapour pressure of the targeted volatile compounds.

Figure 1: The composition of a solid-phase microextraction device (7).

What fiber to use depends on the analyte’s attributes, such as molecular weight and polarity

(7)

Table 1: Analyte’s attributes regarding molecular weight, polarity and volatility and what solid-phase microextraction, SPME, fiber is recommended. Adopted from Sigma-Aldrich (9).

Analyte type Recommended fiber

Gases and low molecular weight compounds (MW 30-225)

75 µm/85 µm Carboxen/polydimethylsiloxane

Volatiles (MW 60-275) 100 µm polydimethylsiloxane

Volatiles, amines and nitro-aromatic compounds (MW 50-300)

65 µm polydimethylsiloxane/divinylbenzene

Polar semi-volatiles (MW 80-300) 85 µm polyacrylate Non-polar high molecular weight compounds (MW

125-600)

7 µm polydimethylsiloxane

Non-polar semi-volatiles (MW 80-500) 30 µm polydimethylsiloxane Alcohols and polar compounds (MW 40-275) 60 µm Carbowax (PEG) Flavor compounds.: volatiles and semi-volatiles, C3-

C20 (MW 40-275)

50/30 µm divinylbenzene/Carboxen on polydimethylsiloxane on a StableFlex fiber Trace compound analysis (MW 40-275) 50/30 µm divinylbenzene/Carboxen on

polydimethylsiloxane on a 2cm StableFlex fiber Amines and polar compounds (HPLC use only) 60 µm polydimethylsiloxane/divinylbenzene

1.3.3 Gas chromatography-mass spectrometry (GC-MS)

In gas chromatography the mobile phase is an inert carrier gas such as nitrogen or helium

(10). The stationary phase consists of a thin layer of polymer or liquid on an inert solid

support inside a column made out of metal or glass. Samples to be analysed are carried through the column by the carrier gas and the analytes are separated from each other due to different interaction with the stationary phase that makes different components stay longer in the column than others.

Gas chromatography can be coupled with mass spectrometry detection (10). When the analytes exit the GC column they are ionized by either positive or negative ionization with subsequent fragmentation. Ions of different fragments are sorted by mass to charge ratios, m/z, to form a fragmentation pattern. Each analyte has a unique fragmentation pattern which makes it possible to identify the compounds with software with mass spectral libraries such as National Institute of Standard and Technology, NIST. GC alone can only separate semi-volatile and semi-volatile compounds with great resolution but no further information is achieved, while MS alone can only provide detailed structural information and fragmentation patterns. Together these two methods can give molecular fingerprints for different compounds to be

(8)

identified in samples.

2. Materials and methods

2.1 Materials

2.1.1 Chemicals

Table 2: All the chemicals used in the study, purity and where they were bought.

Chemical Purity Distributo

r

Chemical Purity Distributor

Methanol 98/100% Analytical reagent grade Fisher scientific UK

Tartaric acid Kebo

AB, stockhol m

Ethanol 99,7% Solveco,

Sweden

Decanoic acid >98%

Sigma-Aldrich (St.Louis MO) L-carvone 99% Sigma-Aldrich (St.Louis MO)

Octanoic acid >99%

Sigma-Aldrich (St.Louis MO) Gamma- Terpinen e 97% Sigma-Aldrich (St.Louis MO) alpha-Pinene 98% Sigma-Aldrich (St.Louis MO) p-Cymene analytica l standard Sigma-Aldrich (St.Louis MO) Citral analytica l standard Sigma-Aldrich (St.Louis MO) alpha- Phellandren e >85% Sigma-Aldrich (St.Louis MO) 3-Carene analytica l standard Sigma-Aldrich (St.Louis MO) Limonene analytica l standard Sigma-Aldrich (St.Louis MO) Isoamy l acetate >97% Sigma-Aldrich (St.Louis MO) Eucalyptol 99% Sigma-Aldrich (St.Louis MO) 4- isopropylbenz y l alcohol >97% Sigma-Aldrich (St.Louis MO) β- Damascenone analytica l standard Sigma-Aldrich (St.Louis Farnesene analytica l standard Sigma-Aldrich (St.Louis

(9)

MO) MO) Estragole 98.8% Sigma-Aldrich (St.Louis MO) D-Camphor >97% Sigma-Aldrich (St.Louis MO) alpha- Terpineo l analytica l standard Sigma-Aldrich (St.Louis MO) Syringol 99% Sigma-Aldrich (St.Louis MO) Furfury l alcohol 99% Sigma-Aldrich (St.Louis MO) 5-methyl furfural 99% Sigma-Aldrich (St.Louis MO) Guaiacol >99% Sigma-Aldrich (St.Louis MO) Eugenol 99% Sigma-Aldrich (St.Louis MO) Vanillin 99% Sigma-Aldrich (St.Louis MO) Methyl guaiacol 99% Sigma-Aldrich (St.Louis MO) Furfural 99% Sigma-Aldrich (St.Louis MO) Hydroxymeth y lfurfural 99% Sigma-Aldrich (St.Louis MO) Ethyl dodecanoate >98% Sigma-Aldrich (St.Louis MO) β-Citronellol analytica l standard Sigma-Aldrich (St.Louis MO) 1-heptanol >99.5% Sigma-Aldrich (St.Louis MO) Ethyl decanoat e >98% Sigma-Aldrich (St.Louis MO) 1-octanol >99% Sigma-Aldrich (St.Louis MO)

Ethyl lactate >98%

Sigma-Aldrich (St.Louis MO) 1-propanol >99% Sigma-Aldrich (St.Louis MO)

2.1.2 Samples

The study comprises a total of 16 wines including four white wines, two orange wines, eight red wines and two sparkling wines as shown in Table 3. Capped eight ml vials were filled with 0.9g NaCl and three ml of each sample and then filled with nitrogen gas until analysis.

(10)

Table 3: Shows the type of wine, producer, name, country, region, grape and year of production for each wine included in the study.

Type of wine Producer and

name

Country/region Grape Year

White Pellerin

Chardonnay

France/Bugey Chardonnay 2014 (Sample 1) and 2016 (Sample 2)

White Causse-Marines

Greilles

France/Gaillac Mauzac, Loin the L’oeil, Muscadelle 2014 (Sample 3) and 2015 (Sample 4) Orange Causse-Marines Zacm’orange

France/Gaillac Mauzac 2015 (Sample 5) and 2016 (Sample 6)

Red Pellerin Gamay France/Bugey Gamay 2014 (Sample 7)

and 2015 (Sample 8) Red Causse-Marines Peyrouzelles France/Gaillac Braucol, Syrah, Duras 2014 (Sample 9) and 2016 (Sample 10)

Red Karim Vionnet

Beaujolais Nouveau

France/Beaujolais Gamay 2016 /Sample 11) and 2017 (Sample 12) Red Domaine la Rabidote France/Saint Chinian Grenache, Carignan, Syrah 2015 (Sample 13) and 2016 (Sample 14)

Sparkling Balviet Cordon du

Bugey

France/Bugey Gamay, Poulsard 2013 (Sample 15)

Sparkling Labarthe

Ancestrale

France/Gaillac Mauzac 2014 (Sample 16)

All samples were analysed in duplicate, and one sample of each type of wine was analysed in quadruplicate to evaluate the method statistically. The samples were chosen at random and ended up being samples 1, 5, 9 and 16.

2.1.3 Sample preparation

To perform qualitative and quantitative analysis of the chosen compounds (Table 4) an artificial wine was prepared. A standard solution of each compound with a concentration around 1000 µg/ml was made. The exact achieved concentrations are presented in table 3. The artificial wine was prepared to be 13% ethanol and pH ~3.5 (11). An alcohol content of 13% was chosen instead of 11% since most of the samples are red wines which generally has higher percentage of ethanol. This was prepared by mixing ethanol with water and then adjusting the pH with tartaric acid. This solution was used to make an artificial wine blank and an artificial wine standard. The artificial wine standard is the artificial wine mixed with

(11)

all compounds and since the analytes of interest were dissolved in methanol, the artificial wine blank is the artificial wine mixed with methanol instead of the compound mastermix. All compounds were prepared in a mastermix to be added into the artificial wine. The final concentration of the mastermix was 29 ppm. The target concentration for the artificial standard was 10 ppm in 20 ml artificial wine. The artificial wine standard had 34.5% of master mix in the final volume which was compensated for in the blank by preparing 34.5% methanol with the artificial wine.

In all samples, artificial wine blank and artificial wine standard, 10 ppm internal standard was added. The internal standard used was hexanoic acid labeled with deuterium in three places.

2.1.4 Standard compounds

Compounds included in the standard solution were chosen based on other studies (4, 5) and what chemicals were available in the MTM laboratory at Örebro university.

Table 4: Compounds included in the study, solvents used for dilution and final concentration of the prepared standard.

Compound Solvent Molecular

weight Concentration (µg/ml) 3-Carene Methanol 136.23 1149 4-isopropylbenzyl alcohol Methanol 150.22 985

alpha Terpineol Methanol 154.25 1037

alpha-Phellandrene Methanol 136.23 1054 alpha-Pinene Methanol 136.23 1121 Citral Methanol 152.23 1156 D-Camphor Methanol 152.23 1063 Estragole Methanol 148.20 985 Eucalyptol Methanol 154.25 1019 Farnesene Methanol 204.35 995 gamma-Terpinene Methanol 136.23 1118 L-Carvone Methanol 150.22 1092 Limonene Methanol 136.238 1130 p-Cymene Methanol 134.22 1057 Eugenol Methanol 164.2 1077 Furfural Methanol 96.09 1028

Furfuryl alcohol Methanol 98.101 996

Guaiacol Methanol 124.14 1051

5-

HyrdoxymethylFurf

(12)

ural

Methyl Furfural Methanol 110.112 1135

Methyl Guaiacol Methanol 138.166 1040

Syringol Methanol 154.162 1059 Vanillin Methanol 152.15 1064 3-Methylbutyl acetate/isoamyl acetate Methanol 130.18 1066 β-Citronellol Methanol 156.27 1069 β-damascenone Methanol 190.28 1041

Ethyl decanoate Methanol 200.32 1063

Ethyl dodecanoate Methanol 228.37 1096

Ethyl lactate Methanol 118.13 1040

1-heptanol Methanol 116.20 1025

1-propanol Methanol 60.10 1163

1-octanol Methanol 130.23 1022

Octanoic acid Methanol 144.21 1013

Decanoic acid Methanol 172.26 1070

2.2 Methods

2.2.1 Headspace solid-phase microextraction

The fiber chosen for the HS-SPME was 50/30µm DVB/CAR/PDMS StableFlex by Supelco, USA, which is a combination of three stationary phases: divinylbenzene, carboxen and polydimethylsiloxane (11). This fiber allows extraction of volatile and semivolatile compounds with molecular weights between 40 and 275 Da. The fiber was assembled, conditioned and cleaned according to instructions by the manufacturer, Supelco/Sigma- Aldrich (12).

A previous study using the same fiber showed that analytes in white wines reach equilibrium after 20 minutes and analytes in red wines after 30 minutes (13). Different sample volumes were tested, 70 ml and 3 ml, and it showed no significant change in results. The extraction temperature had no crucial effect when extracting volatile analytes, but the condensation of ethanol that occurs above 30°C causes non-repeatable extraction as experienced by

Hyötyläinen et al. (13) which is why the extraction was kept at a temperature below 30°C. For this study, the sample volume chosen was 3 ml mixed with 0.9g NaCl and the extraction time was 30 minutes for all types of wines. Extraction was done at ~25°C while stirring with a magnetic stirrer. The SPME fiber was injected manually into the GC-MS and left in the injector for 5 minutes to make sure that it is clean. This was tested with the artificial wine standard and it showed that 5 minutes was enough to clean the fiber and use again directly

(13)

afterwards.

At the start of each day of HS-SPME extractions, the artificial wine blank was analysed, followed by the artificial wine standard and then the samples. Tests were performed with the artificial wine standard to confirm that the same vial could be analysed multiple times with similar result.

2.2.2 Gas chromatography-mass spectrometry

For gas chromatographic separation a 30 m long, 0.250 mm diameter and 0.25 µm film thickness DB-WAXETR column from Agilent Technologies, USA, was used. The

instruments used were a HP 6890 series GC system and a HP 5973 mass selective detector from Agilent Technologies, USA. Three different GC-MS methods were tested to evaluate which one would be the best to use. Scan was mode was used between 30 and 300 m/z and NIST 11, database was used to identify the compounds in the standard. The tested methods were the following:

Method 1 and 3, was adopted from (7), with changed quadrupole temperature for method 3.

Ion source, quadrupole and injector temperatures were set to 230°C, 150°C and 250°C, respectively, and the injector was equipped with a SPME liner. The flow rate of the carrier gas (helium) was set to 1.1 ml/min. Splitless mode and positive electron ionization was used and the scan range went from 30 to 300 m/z. The oven program was run according to Table 5 and gave a total run time of 33 minutes. The same settings and oven program, with a change of the quadrupole temperature to 106°C was used for method 3.

Table 5: The Gas chromatography-mass spectrometry, GC-MS, oven program settings for method 1 and 3.

Step °C/min Temperature (°C) Hold time (min)

1 0 50 2

2 7 110 10

3 10 230 0

Method 2, adopted from (4).

Ion source, quadrupole and injector temperatures were set to 230°C, 150°C and 250°C, respectively, and the injector was equipped with a SPME liner. The flow rate of the carrier gas (helium) was set to 1.0 ml/min. Splitless mode and positive electron ionization was used and the scan range went from 30 to 300 m/z. The oven program was run according to Table 6 and gave a total run time of 76 minutes.

(14)

Table 6: The GC-MS oven program settings for method 2.

Step °C/min Temperature (°C) Hold time (min)

1 0 40 1

2 2 135 0

3 5 212 0

4 15 250 10

The chosen method was method 3 since that method resulted in the best chromatographic separation in relation to the time needed. The chromatograms obtained from each method are presented in Figures 2-4 in the Appendix The only difference between method 1 and 3 is the quadrupole temperature. The lower temperature gave an improved chromatogram with increased number of peaks present, which is why method 3 was used instead of method 1.

2.2.3 Method issues

Initially an artificial wine blank with no internal standard was run, followed by a blank with internal standard. A new peak with a retention time around 23.23 was obtained and expected to be the internal standard. The same thing occurred in the artificial wine standard (Figure 5). However, after all samples were run and the peaks were integrated, this peak was identified by NIST 11 as pentanoic acid instead of labeled hexanoic acid, as was added. In the wine samples, both pentanoic acid and hexanoic acid could be identified by NIST 11, but with very similar retention times and in some samples they were co-eluted. For this reason, the internal standard could not be used to quantify compounds due to the uncertainty with possible co- elution with pentanoic acid in the wine samples.

Figure 5: Chromatograms the artificial wine standard with internal standard at the top, artificial wine blank without internal standard in the middle and artificial wine blank with internal standard at the bottom. The obtained peaks were identified by NIST 11 as pentanoic acid instead of the

(15)

actual mass-labeled hexanoic acid, which lead do the internal standard not being usable for quantification.

3. Results

3.1 Artificial wine standard

In the artificial wine standard only 15 compounds could be identified with mass spectra comparison and NIST 11, which are presented in Table 7. The remaining compounds that could not be found could possibly be lost in background noise, not being volatile enough or being too volatile hence too low concentrations to be seen. The chromatographic result is shown in Figure 6.

Figure 6: Chromatogram from the artificial standard containing 10 ppm of the 34 compounds presented in Table 4.

Table 7: The compounds with their corresponding retention time and area in the artificial wine standard that were identified by NIST 11.

Compound Retention time Area

3-Carene 4.709 1111712 4-isopropylbenzyl alcohol 27.694 3579 alpha-Phellandrene 4.989 525771 alpha-Pinene 4.014 7660 D-Camphor 11.292 18351 Estragole 15.363 167964

(16)

Gamma-Terpinene 6.421 2345116 Limonene 5.567 2437386 p-Cymene 6.842 1624548 Eugenol 28.460 12768 3-Methylbutyl acetate/isoamyl acetate 4.330 28177 Ethyl decanoate 15.050 1674775 Ethyl dodecanoate 23.906 4546498 Octanoic acid 27.276 37497 Decanoic acid 29.746 13686

3.2 Wine analysis result

All wines were analysed in duplicate and wine sample number 1, 5, 9 and 16 were analysed in quadruplicate. Since the internal standard could not be used for calculations, the overall results in tables are presented in area units. The ten peaks with highest intensity were chosen to be included in the study. However, one of these ten were the internal standard that was excluded which means that the figures presented in the results shows the nine most abundant peaks according to their areas. The reason for choosing these peaks was due to time

limitations and not being able to do proper quantitative calculations. What sample corresponds to which wine is presented in Table 3.

(17)

Figure 7: Homologue patterns based on area of the nine most abundant peaks of the wine Pellerin from 2014. 1-4 are replicates of the same wine sample.

Figure 8: Homologue patterns based on area of the nine most abundant peaks of the wine Pellerin from 2016. 1-2 are replicates of the same wine sample.

Figure 9: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Greilles from 2014. 1-2 are replicates of the same wine sample.

(18)

Figure 10: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Greilles from 2015. 1-2 are replicates of the same wine sample. In 4.2 the pentanoic acid area was put as zero due to co-elution with the peak for hexanoic acid.

Figure 11: Homologue patterns based on area of the nine most abundant peaks of the white wines Pellerin from 2014 (number 1-4), Pellerin from 2016 (number 5-6), Causse-Marines Greilles from 2014 (number 7-8) and Causse-Marines Greilles from 2015 (number 9-10).

(19)

3.2.2 Orange wines

Figure 12: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Zacm’ orange from 2015. 1-4 are replicates of the same wine sample.

Figure 13: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Zacm’ orange from 2016. 1-2 are replicates of the same wine sample.

(20)

Figure 14: Homologue patterns based on area of the nine most abundant peaks of the orange wines Causse-Marines Zacm’ orange from 2015 (number 1-4) and Causse-Marined Zacm’ orange from 2016 (number 5-6).

3.2.3 Red wines

Figure 15: Homologue patterns based on area of the nine most abundant peaks of the wine Pellerin Gamay from 2014. 1-2 are replicates of the same wine sample.

(21)

Figure 16: Homologue patterns based on area of the nine most abundant peaks of the wine Pellerin Gamay from 2015. 1-2 are replicates of the same wine sample.

Figure 17: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Peyrouzelles from 2014. 1-4 are replicates of the same wine sample.

(22)

Figure 18: Homologue patterns based on area of the nine most abundant peaks of the wine Causse-Marines Peyrouzelles from 2016. 1-2 are replicates of the same wine sample.

Figure 19: Homologue patterns based on area of the nine most abundant peaks of the wine Karim Vionnet Beaujolais Nouveau from 2016. 1-2 are replicates of the same wine sample.

(23)

Figure 20: Homologue patterns based on area of the nine most abundant peaks of the wine Karim Vionnet Beaujolais Nouveau from 2016. 1-2 are replicates of the same wine sample. In 12.2 the pentanoic acid area was put as zero due to co-elution with the peak for hexanoic acid.

Figure 21: Homologue patterns based on area of the nine most abundant peaks of the wine Domaine la Rabidote from 2015. 1-2 are replicates of the same wine sample.

(24)

Figure 22: Homologue patterns based on area of the nine most abundant peaks of the wine Domaine la rabidote. 1-2 are replicates of the same wine sample.

Figure 23: Homologue patterns based on area of the nine most abundant peaks of the red wines Pellerin Gamay from 2014 (number 1-4), Pellerin Gamay from 2016 (number 5-6), Causse-Marines Peyrouzelles from 2014 (7-8), Causse-Marines Peyrouzelles from 2016

(25)

(number 9-10), Karim Vionnet Beaujolais Nouveau from 2016 (number 11-12), Karim Vionnet Beaujolais Nouveau from 2017 (number 13-14), Domaine la Rabidote from 2015 (number 15-16) and Domaine la Rabidote from 2016 (number 17-18).

3.2.4 Sparkling wines

Figure 23: Homologue patterns based on area of the nine most abundant peaks of the wine Balviet Cordon du Bugey from 2013. 1-2 are replicates of the same wine sample.

Figure 25: Homologue patterns based on area of the nine most abundant peaks of the wine Labarthe Ancestrale from 2014. 1-4 are replicates of the same wine sample.

(26)

Figure 26: Homologue patterns based on area of the nine most abundant peaks of the sparkling wines Balviet Cordon du Bugey from 2013 (number 1-2) and Labarthe Ancestrale from 2014 (number 3-6).

3.3 Quantification

Tentative concentrations were calculated for the compounds found in the wine samples that also were found in the artificial wine standards. The compounds were ethyl decanoate, ethyl dodecanoate, octanoic acid and decanoic acid. The calculated concentrations are presented in Table 8 in the Appendix.

Out of the four compounds, the most abundant being found in 15 out of 16 wines was decanoic acid with concentrations ranging from 16 to 265 ppm. The second highest abundance was octanoic acid, detected in 15 out of 16 wines with concentrations ranging from 13 to 124 ppm. Ethyl decanoate was also found in 15 out of 16 wines but at much lower concentrations ranging from 0.3 to 7 ppm. Ethyl decanoate could only be found in 5 out of 16 wines at very low concentrations ranging from 0.1 to 0.3 ppm.

3.4 Identified compounds

Common for all wine samples was the presence of isoamyl alcohol, and as previously mentioned the majority of the samples also contained ethyl decanoate, octanoic acid and decanoic acid. Some different trends were seen between the different colours of wines. Both sparkling wines contained ethyl hexanoate, ethyl octanoate and phenylethyl alcohol. Six out

(27)

of eight red wines contained ethyl succinate and five of them also contained ethyl hexanoate. Wine sample number 14 was the most different out of the red wine samples with high

abundance of tetradecamethylhexasiloxane, 2,2,4,4,5,5,7,7-Octamethyl-3,6-dioxa-2,4,5,7- tetrasilaoctane and tetrakis(trimethylsiloxy)silane.

Common compounds in the orange wines apart from the previously mentioned were ethyl hexanoate, ethyl octanoate and pentanoic acid. All white wine samples also contained ethyl hexanoate and ethyl octanoate.

3.5 Statistics

Statistical calculations for relative standard deviation was performed and results are shown in Table 9. Each sample was analysed in quadruplicate from four different sample vials. All replicates of sample 1 was analysed on day one, all of sample 5 on day two, one of sample 9 on day two, three of sample 9 on day three and all of sample 16 on day 4. The statistics are also not as adequate as initially planned due to not being able to quantify more compounds. The relative standard deviation ranges from 6-34% for ethyl decanoate, 35-48% for ethyl dodecanoate, 13-21% for octanoic acid and 12-59% for decanoic acid. This indicates that none of the extractions are repeatable with good results due to the high deviation. Reasons for non-repeatable extractions could be the HS-SPME fiber performance, reactions with air despite filling the vials with nitrogen gas or not homogenous samples.

Table 9: The relative standard deviation percentage for ethyl decanoate, ethyl dodecanoate, octanoic acid and decanoic acid for samples 1, 5, 9 and 16 that were analysed in

quadruplicate.

Limit of detection, LOD, and limit of quantification, LOQ, for the samples has been calculated separately according to the artificial wine standard that was analysed the same day as the samples. This resulted in four different LOD and LOQ for each compound that is presented in Table 10. The LOD was calculated as (concentration/signal to noise)*3 and the LOQ as (concentration/signal to noise)*10. The concentration was 10 ppm and the exact signal to noise ratios are presented in Table 12 in appendix.

(28)

Table 10: Limit of detection, LOD, for ethyl decanoate, ethyl dodecanoate, octanoic acid and decanoic acid in the different samples depending on the day they were analysed.

Table 11: Limit of quantification, LOQ, for ethyl decanoate, ethyl dodecanoate, octanoic acid and decanoic acid in the different samples depending on the day they were analysed.

The concentration for these four compounds in the wine samples were higher than LOD and LOQ which lead to possible quantification of the four compounds in Table 8 in the Appendix. The concentrations were calculated by using the areas presented in Table 13 in the Appendix.

4. Discussion

4.1 Limitations

Normalised homologue area patterns to characterise wine is not the best approach, since the compounds with the largest area is not necessarily the compound with the strongest odour. In databases like PubChem, one can find the different concentrations needed for each compound to give a distinct odour, which means that quantitative analysis would be needed to properly characterise the wines. Since quantification was not possible in this study due to issues with identifying the internal standard, the homologue area patterns are used as a method to perform characterisation of the wines. However, as stated previously, this approach is not the most optimal due to the compounds differences in concentration for detectable odour and therefore, the results might not correspond to the actual odour differences of the wines.

4.2 Quantitative analysis

Quantitative analysis was performed for those compounds that were present in both of the samples and the standard, presented in Table 8 in the Appendix. Normalisation against an internal standard was not possible since the molecular ion could not be seen in the mass spectrum, which made correct identification by NIST impossible. A comparison of the mass spectra of pentanoic acid and hexanoic acid from NIST is shown in Figure 27 and 28 in the Appendix. Calculations of concentrations for other compounds identified in the samples was

(29)

not performed. Correct calculations for these compounds would require the injection of these compounds of a known concentration for the calculation of their respective response factors. Because of this, only four compounds could be quantified and for the remaining compounds no quantification was made. The nine most intense peaks in each sample was integrated and identified as shown in Figures 7-26. However, in these figures pentanoic acid is included which might not be the correct area to pentanoic acid alone due to the possible mix-up and/or co-elution with the internal standard hexanoic acid. Pentanoic acid is still included in the figures since in most samples, apart from 4.2 and 12.2 it was possible to see two peaks, and the software could also identify the two peaks as different compounds.

As shown in Figures 7-26, when looking at the relation between different compounds in samples, some had high reproducibility within the duplicates/quadruplicates while some did not. The obtained areas in each replicate of the same sample did differ significantly in some samples. Areas are shown in Table 13 in the Appendix. However, the varying intensities do not affect the homologue patterns since they are normalised, but quantification may not give an accurate description.

4.3 Aroma compounds

A previous study performed by Capone et al. (4) and Li et al. (5) has presented characteristics for volatiles determined by HS-SPME-GC-MS. The analytes of interest were the following presented in Table 14.

Table 14: The volatile analytes with a characterised odour presented in studies by Capone et al. (4) and Li et al. (5).

Compound Odour quality

Ethyl hexanoate Apple peel, fruit

Ethyl octanoate Melon, wood, pineapple

Ethyl decanoate Floral, soap, fruity

Ethyl succinate Wine, fruit

Ethyl dodecanoate Fruity, floral, sweet

Octanoic acid Butter, almond, fatty

Decanoic acid Rancid, fat

Ethyl 9-decanoate Fruity, fatty

Isoamyl alcohol (3-methyl-1-butanol) Harsh, nail polish, fusel Phenethyl alcohol (2-Phenylethanol) Floral, rose

(30)

Ethyl hexadecanoate Fruity

According to the odour quality presented in Table 14, the most significant odours from volatile compounds in all wines included in the study were melon, wood, pineapple, floral, soap, fruity, butter, almond, fatty, wine, rancid, harsh and nail polish. In this study, all wine samples contained relatively large amounts of isoamyl alcohol and ethyl octanoate which would indicate that all the wines have a shared odour. While the odour of isoamyl alcohol is regarded as unpleasant, ethyl octanoate has a pleasant odour of melon, wood and pineapple. The majority of the wines, apart from sample 12 and 13, contained ethyl hexanoate which would indicate a fruity apple peel odour.

Compounds without any previously studied odour quality, such as different siloxanes, silanes and butylated hydroxytoluene, were also found in some samples. Butylated hydroxytoluene is, according to the Pubchem database, a compound that is used in foods to prevent oxidation and free-radical formation, which would explain its presence in some of the wines. However, no explanation to the presence of siloxanes and silanes has been found.

The wines Domaine la Rabidote from 2015 and 2016 were used for taste testing for the sensory analysis of the study. The wine from 2015 corresponds to sample 13 and the wine from 2016 corresponds to sample 14. According to the results, sample 13 was experienced as spicier and possessed a greater flavour of minerals and greens. Sample 14 was experienced as more fruity and floral. Comparing with Table 14, this can be explained by Sample 14 containing ethyl dodecanoate that has a fruity and floral aroma while this compound was not one of the nine most abundant in sample 13. Sample 13 contains slightly less ethyl decanoate which also indicates that it will not taste as floral as sample 14. None of the nine most

abundant peaks could explain the mineral and greens flavour being stronger in sample 13 compared to sample 14.

4.5 Comparison of wines

4.5.1 White wines

The homologue patterns of the white wines presented in Figure 11 shows that the four most abundant compounds, isoamyl alcohol, ethyl hexanoate, ethyl octanoate and ethyl decanoate are very similar. The most obvious difference between the patterns would be the wine Causse-Marines Greilles from 2015 containing ethyl 9-decanoate while the others do not, which would indicate that this wine is more fruity and fatty. The wine Causse-Marines Greilles from 2014 is the only wine without ethyl dodecanoate which indicates it being less fruity, floral and sweet compared to the other wine samples.

(31)

4.5.2 Orange wines

The homologue patterns of the two orange wines presented in Figure 14 are generally alike. The only identified compound to separate them that has a known aroma according to Table 13 is ethyl dodecanoate. With the abundance of ethyl dodecanoate being the only difference of the analytes with known aroma in the homologue pattern, it would indicate that the wine Causse-Marines Zacm’ orange from 2015 is more fruity, floral and sweet compared with the same wine from 2016.

4.5.3 Red wines

The homologue patterns for the red wines presented in Figure 23 shows that the distribution between the three most abundant compounds, isoamyl alcohol, ethyl octanoate and ethyl decanoate, are quite similar. The wines Pellerin Gamay from 2014 and 2015, Causse-Marines Peyrouzelles from 2015 and 2016 and Karim Vionnet Beaujolais Nouveau from 2016 could have more of an apple peel and fruity aroma due to higher abundance of ethyl hexanoate. The higher abundance of phenethyl alcohol in most of the samples indicate a rose aroma. The wine Domaine la Rabidote from 2016 is the only sample of red wine where ethyl

dodecanoate was identified among the most abundant ones, which implies that this wine is more sweet, fruity and floral than the others.

4.5.4 Sparkling wines

There is a significant difference between the two sparkling wine’s homologue patterns presented in Figure 26. This is most likely due to the Balviet Cordon du Bugey wine being rosé and Labarthe Ancestrale being white. The nine most abundant compounds are similar but the distribution of them are not. The white sparkling wine has ethyl decanoate as one of the nine most abundant compounds which could indicate that this wine is slightly more fruity and floral than the rosé sample. When comparing the white sparkling wine with the normal white wines, there is also a significant difference in the distribution of compounds, mostly regarding the pentanoic acid. The high intensity peaks of pentanoic acid in both of the sparkling wine samples compared to the other samples, could indicate co-elution with the internal standard and therefore not accurate homologue patterns.

5. Conclusion

When identifying and making homologue patterns of the nine most abundant compounds in wine samples, some differences can be seen. However, to get a more thorough understanding of the chemical markers that separates each wine’s sensory profile, more compounds would

(32)

have to be included in the study. The chromatograms of the wine samples contained a high number of peaks that were not included in this study, but could be used for a more detailed characterisation. Quantification of all analytes included in the study could help with further characterisation of the samples, which was not possible due to the possible co-elution of pentanoic acid and the internal standard, mass-labeled hexanoic acid. To get a more exact characterisation of wine, a method with better statistical values regarding reproducibilit y would be needed.

(33)

6. References

1. Agerlin Petersen, M., Helgesdotter Rogsnå, G., Hersleth, M., Misje. K.E., Rathe, M. et al. (2017). From wine to wine reduction: Sensory and chemical aspects. International Journal of

Gastronomy and Food science, 9, pp. 62-74. doi:https://doi.org/10.1016/j.ijgfs.2017.06.006

2. Ánfeles Pozo-Bayon, M., Esteban-Fernándes, A., Jiménes-Girón, A., Muños-González, C. et al. (2018). Aroma release in the oral cavity after wine intake is influences by wine matrix composition. Food chemistry, 243, pp.125-133. doi:https://doi.org/10.1016/j.foodchem.2017.09.101 3. Alegre, Y., Ferreira, V., García, D., Hernández-Orte, P., Razquin, I. et al. (2017). Rapid strategies for the determination of sensory and chemical differences between a wealth of similar wines. European Food Research and Technology, 243, pp. 1295-1309. doi: 10.1007/s00217-017-2857-7

4. Capone, D., Jeffery, D., Wang, J. and Wilkinson, K. (2016). Chemical and sensory profiles of rosé wines from Australia. Food Chemistry, 196, pp. 682-693. doi:

https://doi.org/10.1016/j.foodchem.2015.09.111

5. Li, B., Lu, J., Sun, J., Tian, T., Wu, D. et al. (2018). Optimization of fermentation

conditions and comparison of flavour compounds for three fermented greengage wines. LWT - Food Science and Technology, 89, pp.542-550. doi:https://doi.org/10.1016/j.lwt.2017.11.006 6. Coelho, G., Elias, A. and Santos, B. (2016). Use of HS-SPME for analysis of the influence of salt concentration and temperature on the activity coefficient at infinite dilution of ethanol- water-salt systems. Fluid Phase Equilibria, 429, pp. 21-26. doi:

https://doi.org/10.1016/j.fluid.2016.08.030

7. Salihovic, S. (2009). Analysis and identification of volatile oak related compounds by HS- SPME-GC-MS and MEPS-GC-MS.

8. Bekhit, A.E.D., Hamid, N., Law, T.F., Ma, Q.L. and Robertson, J. (2013). Optimization of headspace solid phase microextraction (HS-SPME) for gas chromatography mass

spectrometry (GC-MS) analysis of aroma compounds in cooked beef using response surface methodology. Microchemical Journal, 111, pp. 16-24. doi:

https://doi.org/10.1016/j.microc.2012.10.007

9.https://www.sigmaaldrich.com/technical-documents/articles/analytical/selecting-spme- fibers.html#fiber

(34)

10. Maqbool, K., Zameer Hussain, S. (2014). GC-MS: Principle, Technique and its

application in Food Science. International Journal of Current Science, 13, pp. 116-126. ISSN 2250-1770.

11. Caliari, V., Grützmann Arcari, S., Sganzerla, M., Tiexeira, G. (2017). Volatile

composition of Merlot red wine and its contribution to the aroma: optimization and validation of analytical method. Talanta, 174, pp. 752-766. doi:https://doi.org/10.1016/j.talanta.2017.06.074 12.https://www.sigmaaldrich.com/content/dam/sigma-

aldrich/docs/Sigma/General_Information/1/t794123.pdf

13. Hyötyläinen, T., Kallio, M., Kallonen, R., Lehtonen, P., Patrikainen, E. et al. (2014). Characterisation of Wines by Comprehensive Two-Dimensional Gas Chromatography and Chemometric Methods.

(35)

7. Appendix

Figure 2: The chromatogram obtained from method 1.

(36)

Figure 4: The chromatogram obtained from method 3

Table 8: The tentative concentrations of each compound that could be identified in the artificial wine standard and in the wine samples. The concentrations were calculated by using the average area of the compound in all replicas of the same wine and then comparing to the area of the same compound in the artificial wine standard, where each compound was 10 ppm.

(37)

Figure 27: Mass spectrum of pentanoic acid from NIST.

(38)

Table 12: The signal to noise ratios for ethyl decanoate, ethyl dodecanoate, octanoic acid and decanoic acid in the artificial wine standard analysed different days. 20180503 applies to samples 1.1-4.1, 20180507 applies to samples 4.2-9.1, 20180508 applies to

samples 9.2-13.2 and 20180509 applies to samples 14.1-16.4.

Table 13: The ten most abundant peaks in every sample including retention time and compound.

Sample Peak1 Peak2 Peak 3 Peak 4 Peak 5

1.1 Rt: 5.680 Name: Isoamyl alcohol Area: 105709 Rt: 6.251 Name: Ethyl hexanoate Area: 223869 Rt: 10.028 Name: Ethyl octanoate Area: 1628309 Rt: 14.970 Name: Ethyl decanoate Area: 609569 Rt: 23.772 Name: Ethyl dodecanoate Area: 137337

Peak 6 Peak 7 Peak 8 Peak 9 Peak 10 IS

Rt: 27.275 Name: Octanoic acid Area: 93471 Rt:29.073 Name: Hexasiloxan Area: 417222 Rt: 29.741 Name: Decanoic acid Area :53454 Rt: 31.754 Name:1- Monolinoleoylg lycerol trimethylsilyl ether Area: 38272 Rt: 23.328 Name: Hexanoic acid Area: 9818

Peak1 Peak2 Peak 3 Peak 4 Peak 5

1.2 Rt: 5.668 Name: Isoamyl alcohol Area: 147800 Rt: 6.262 Name: Ethyl hexanoate Area: 282423 Rt: 10.036 Name: Ethyl octanoate Area: 1812129 Rt: 14.975 Name: Ethyl decanoate Area: 611358 Rt: 23.770 Name: Ethyl dodecanoate Area: 82224

Rt: 27.274 Name: Octanoic acid Area: 85512 Rt: 29.074 Name: Hexasiloxane Area: 33875 Rt: 29.740 Name: Decanoic acid Area: 55018 Rt: 31.756 Name:1- Monolinoleoylg lycerol trimethylsilyl ether Area: 30342 Rt: 23.330 Name: Hexanoic acid Area: 12644

1.3 Peak1 Peak2 Peak 3 Peak 4 Peak 5

Rt: 5.667 Name: Isoamyl alcohol Area: 121416 Rt: 6.251 Name: Ethyl hexanoate Area: 277255 Rt: 10.028 Name: Ethyl octanoate Area: 1686534 Rt: 14.967 Name: Ethyl decanoate Area: 566863 Rt: 23.772 Name: Ethyl dodecanoate Area: 62086

(39)

Peak 6 Peak 7 Peak 8 Peak 9 Peak 10 IS Rt: 27.272 Name: Octanoic acid Area: 118601 Rt: 29.082 Name: Hexasiloxane Area: 21773 Rt: 29.741 Name: Decanoic acid Area: 6648 Rt: 31.746 Name: 1- Monolinoleoylg lycerol trimethylsilyl ether Area: 15200 Rt: 23.325 Name: Hexanoic acid Area: 12187

Rt: 5.684 Name: Isoamyl alcohol Area: 116566 Rt: 6.225 Name: Ethyl hexanoate Area: 259423 Rt: 10.032 Name: Ethyl octanoate Area: 1748399 Rt: 14.965 Name: Ethyl decanoate Area: 526642 Rt: 23.766 Name: Ethyl dodecanoate Area: 46412

Rt: 27.273 Name: Octanoic acid Area: 136152 Rt: 29.073 Name: Hexasiloxane Area: 20141 Rt: 29.739 Name: Decanoic acid Area: 70641 Rt: 31.752 Name: 1- Monolinoleoylg lycerol trimethylsilyl ether Area: 25193 Rt: 23.322 Name: Hexanoic acid Area: 19439

Rt: 5.638 Name: Isoamyl alcohol Area: 113089 Rt: 6.257 Name: Ethyl hexanoate Area: 303409 Rt: 10.044 Name: Ethyl octanoate Area: 2437741 Rt: 14.999 Name: Ethyl decanoate Area: 1261610 Rt:23.771 Name: Ethyl dodecanoate Area: 90366

Rt: 27.272 Name: Octanoic acid Area: 181616 Rt: 29.741 Name: Decanoic acid Area: 98937 Rt:29.839 Name: Ethyl hexadecanoate Area: 40423 Rt: 30.478 Name: Hexasiloxane Area: 11231 Rt :23.324 Name: Hexanoic acid Area: 16742

Rt: 5.687 Name: Isoamyl alcohol Area: 174739 Rt: 6.254 Name: Ethyl hexanoate Area: 298876 Rt: 10.042 Name: Ethyl octanoate Area:2368781 Rt: 15.001 Name: Ethyl decanoate Area: 1308387 Rt: 23.766 Name: Ethyl dodecanoate Area: 83251

Rt: 27.273 Name: Octanoic acid Area: 133930 Rt: 27.739 Name: Decanoic acid Area: 67553 Rt: 29.837 Name: Ethyl hexadecanoate Area: 32559 Rt: 30.476 Name: Hexasiloxane Area: 10455 Rt: 23.232 Name: Hexanoic acid Area: 16287

Rt: 5.684 Name: Isoamyl alcohol Rt: 6.258 Name: Ethyl hexanoate Rt: 10.000 Name: Ethyl octanoate Rt: 14.949 Name: Ethyl decanoate Rt: 23.231 Name: Pentanoic acid

(40)

Area: 120079 Area: 26507 Area: 252121 Area: 238438 Area: 54857

Rt: 24.546 Name: Phenethyl alcohol Area: 22444 Rt: 25.022 Name: Butylated Hydroxytoluene Area: 23257 Rt: 27.270 Name: Octanoic acid Area: 90799 Rt: 29.793 Name: Decanoic acid Area: 86011 Rt: 23.239 Name: Hexanoic acid Area: 9736

Rt: 5.681 Name: Isoamyl alcohol Area: 186239 Rt: 6.252 Name: Ethyl hexanoate Area: 92794 Rt: 10.017 Name: Ethyl octanoate Area: 1039160 Rt: 14.972 Name: Ethyl decanoate Area: 670092 Rt: 23.228 Name: 49355 Pentanoic acid Area: 49355

Rt: 24.543 Name: Penethyl alcohol Area: 34070 Rt: 25.023 Name: Butylated Hydroxytoluene Area: 49062 Rt: 27.270 Name: Octanoic acid Area: 76613 Rt: 29.740 Name: Decanoic acid Area: 70923 Rt: 23.320 Name: Hexanoic acid Area: 10764

Rt: 5.676 Name: Isoamyl alcohol Area: 146410 Rt: 6.254 Name: Ethyl hexanoate Area: 165514 Rt: 10.028 Name: Ethyl octanoate Area: 1642664 Rt: 14.983 Name: Ethyl decanoate Area: 955959 Rt: 16.875 Name: Ethyl 9- decenoate Area: 48483

Rt: 23.223 Name: Pentanoic acid Area: 40513 Rt: 23.765 Name: Ethyl dodecanoate Area: 58824 Rt: 27.268 Name: Octanoic acid Area: 111712 Rt: 29.738 Name: Decanoic acid Area: 90138 Rt: 23.324 Name: Hexanoic acid Area: 12645

Rt: 5.686 Name: Isoamyl alcohol Area: 184207 Rt: 6.257 Name: Ethyl hexanoate Area: 95539 Rt: 10.015 Name: Ethyl octanoate Area: 1084114 Rt: 14.970 Name: Ethyl decanoate Area: 850334 Rt: 16.866 Name: Ethyl 9- decanoate Area: 43769

Rt: 23.224 Name: Pentanoic acid Area: *Co-elution with IS Rt: 23.768 Name: Ethyl dodecanoate Area: 233915 Rt: 27.267 Name: Octanoic acid Area: 217827 Rt:29.738 Name: Decanoic acid Area: 180615 Rt: 23.312 Name: Hexanoic acid Area: *Co-elution with pentanoic acid

Rt: 5.682 Name: Isoamyl Rt: 6.265 Name: Ethyl Rt: 10.002 Name: Ethyl Rt: 14.940 Name: Ethyl Rt: 23.220 Name:

(41)

alcohol Area: 267498 hexanoate Area: 58049 octanoate Area: 520950 decanoate Area: 295861 Pentanoic acid Area: 149556

Rt: 23.758 Name: Ethyl dodecanoate Area: 72101 Rt: 25.017 Name: Butylated Hydroxytoluene Area: 40840 Rt: 27.268 Name: Octanoic acid Area: 86847 Rt: 29.737 Name: Decanoic acid Area: 67859 Rt: 23.314 Name: Hexanoic acid Area: 14988

Rt:5.681 Name:Isoamyl alcohol Area: 227208 Rt: 6.252 Name: Ethyl hexanoate Area: 59463 Rt: 10.000 Name: Ethyl octanoate Area: 531694 Rt: 14.943 Name: Ethyl decanoate Area: 317139 Rt: 23.222 Name: Pentanoic acid Area: 93778

Rt: 5.678 Name:Isoamyl alcohol Area: 238145 Rt: 6.252 Name: Ethyl hexanoate Area: 61612 Rt: 10.000 Name: Ethyl octanoate Area: 471614 Rt: 14.936 Name: Ethyl decanoate Area: 224560 Rt: 23.222 Name: Pentanoic acid Area: 129117

Rt:5.682 Name:Isoamyl alcohol Area: 276275 Rt: 6.252 Name: Ethyl hexanoate Area: 67132 Rt: 10.001 Name: Ethyl octanoate Area: 549364 Rt: 14.936 Name: Ethyl decanoate Area: 287891 Rt: 23.216 Name: Pentanoic acid Area: 91359

Rt: 5.685 Name:Isoamyl Rt: 6.256 Name: Ethyl Rt: 10.007 Name: Ethyl Rt: 14.946 Name: Ethyl Rt: 23.219 Name:

(42)

alcohol Area: 273187 hexanoate Area: 132631 octanoate Area: 865854 decanoate Area: 422536 Pentanoic acid Area: 65517

Rt: 27.267 Name: Octanoic acid Area: 87425 Rt: 27.450 Name: Tetrakis(trimeth ylsiloxy)silane Area: 29632 Rt: 29.048 Name: Tetradecamethy lhexasiloxane Area: 24547 Rt: 29.736 Name: Decanoic acid Area: 50004 Rt: 23.320 Name: Hexanoic acid Area: 14765

Rt: 5.685 Name:Isoamyl alcohol Area: 219217 Rt: 6.256 Name: Ethyl hexanoate Area: 56236 Rt: 10.000 Name: Ethyl octanoate Area: 546257 Rt: 14.939 Name: Ethyl decanoate Area: 332823 Rt: 23.222 Name: Pentanoic acid Area: 73536

Rt: 27.267 Name: Octanoic acid Area: 114584 Rt: 27.459 Name: Tetrakis(trimeth ylsiloxy)silane Area: 9852 Rt: 29.055 Name: Tetradecamethy lhexasiloxane Area: 9591 Rt: 29.736 Name: Decanoic acid Area: 71130 Rt: 23.316 Name: Hexanoic acid Area: 14508

Rt: 5.682 Name: Isoamyl alcohol Area: 306752 Rt: 6.253 Name: Ethyl hexanoate Area: 51989 Rt: 9.997 Name: Ethyl octanoate Area: 473818 Rt: 14.936 Name: Ethyl decanoate Area: 246898 Rt: 15.892 Name: Ethyl succinate Area: 70328

Rt: 23.222 Name: Pentanoic acid Area: 89229 Rt: 27.267 Name: Octanoic acid Area: 57471 Rt: 29.733 Name: Decanoic acid Area: 29448 Rt: 29.831 Name: Ethyl hexadecanoate Area: 28919 Rt: 23.320 Name: Hexanoic acid Area: 9208

Rt: 23.224 Name: Pentanoic acid Area: 87608 Rt: 27.269 Name: Octanoic acid Area: 41357 Rt: 29.735 Name: Decanoic acid Area: 25895 Rt: 29.830 Name: Ethyl hexadecanoate Area: 29617 Rt: 23.315 Name: Hexanoic acid Area: 8334

(43)

Peak 6 Peak 7 Peak 8 Peak 9 Peak 10 IS Rt: 23.215 Name: Pentanoic acid Area: 280608 Rt: 24.533 Name: Phenylethyl Alcohol Area: 47780 Rt: 27.267 Name: Octanoic acid Area: 60731 Rt: 29.733 Name: Decanoic acid Area: 20866 Rt: 23.316 Name: Hexanoic acid Area: 14270

Rt: 23.216 Name: Pentanoic acid Area: 116318 Rt: 24.534 Name: Phenylethyl Alcohol Area: 62212 Rt: 27.265 Name: Octanoic acid Area: 70516 Rt: 29.734 Name: Decanoic acid Area: 25656 Rt: 23.311 Name: Hexanoic acid Area: 14843

Rt: 5.685 Name: Isoamyl alcohol Area: 285673 Rt: 6.253 Name: Ethyl hexanoate Area: 44596 Rt: 9.997 Name: Ethyl octanoate Area: 331984 Rt: 12.389 Name: Dodecamethylp entasiloxane Area: 31517 Rt: 14.923 Name: Ethyl decanoate Area: 103315

Rt: 15.892 Name: Ethyl succinate Area: 98189 Rt: 23.216 Name: Pentanoic acid Area: 90588 Rt: 27.267 Name: Octanoic acid Area: 116460 Rt: 29.737 Name: Decanoic acid Area: 56667 Rt: 23.314 Name: Hexanoic acid Area: 15216

(44)

Peak 6 Peak 7 Peak 8 Peak 9 Peak 10 IS Rt: 15.892 Name: Ethyl succinate Area: 113628 Rt: 23.216 Name: Pentanoic acid Area: 262874 Rt: 27.264 Name: Octanoic acid Area: 116118 Rt: 29.734 Name: Decanoic acid Area: 42885 Rt: 23.310 Name: Hexanoic acid Area: 60144

Rt: 5.685 Name: Isoamyl alcohol Area: 321310 Rt: 6.255 Name: Ethyl hexanoate Area: 79950 Rt: 10.004 Name: Ethyl octanoate Area: 682833 Rt: 14.933 Name: Ethyl decanoate Area: 344185 Rt: 23.219 Name: Pentanoic acid Area: 68507

(45)

Rt: 24.533 Name: Phenylethyl Alcohol Area: 61447 Rt: 27.267 Name: Octanoic acid Area: 165183 Rt: 27.430 Name: Tetrakis(trimeth ylsiloxy)silane Area: 45873 Rt: 29.733 Name: Decanoic acid Area: 135691 Rt: 23.313 Name: Hexanoic acid Area: 20010

Rt: 24.536 Name: Phenylethyl Alcohol Area: 87659 Rt: 27.267 Name: Octanoic acid Area: 212642 Rt: 27.426 Name: Tetrakis(trimeth ylsiloxy)silane Area: 7756 Rt: 29.736 Name: Decanoic acid Area: 102759 Rt: 23.310 Name: Hexanoic acid Area: 31897

Rt: 5.683 Name: Isoamyl alcohol Area: 332162 Rt: 9.995 Name: Ethyl octanoate Area: 413365 Rt: 14.931 Name: Ethyl decanoate Area: 281671 Rt: 23.213 Name: Pentanoic acid Area: *Co-elution with IS Rt: 24.531 Name: Phenylethyl Alcohol Area: 79465

Rt: 25.363 Name: Hexadecamethy lheptasiloxane Area: 35312 Rt: 27.265 Name: Octanoic acid Area: 131774 Rt: 27.428 Name: Tetrakis(trimeth ylsiloxy)silane Area: 36838 Rt: 29.734 Name: Decanoic acid Area: 95066 Rt: 23.308 Name: Hexanoic acid Area: *Co-elution with pentanoic acid

Rt: 5.677 Name: Isoamyl alcohol Area: 345020 Rt: 10.003 Name:Ethyl octanoate Area: 710859 Rt: 14.942 Name: Ethyl decanoate Area: 581398 Rt: 23.211 Name: Pentanoic acid Area: 205433 Rt: 24.536 Name: Phenylethyl Alcohol Area: 76490

Rt: 25.367 Name: Hexadecamethy lheptasiloxane Area: 20396 Rt: 27.263 Name: Octanoic acid Area: 85301 Rt: 27.432 Name: Tetrakis(trimeth ylsiloxy)silane Area: 23069 Rt: 29.732 Name: Decanoic acid Area: 71286 Rt: 23.312 Name: Hexanoic acid Area: 12636

Rt: 5.687 Name: Isoamyl Rt: 9.993 Name: Ethyl Rt: 12.362 Name: Rt: 14.926 Name: Ethyl Rt: 15.895 Name: Ethyl

(46)

alcohol Area: 476014 octanoate Area: 318451 Dodecamethylp entasiloxane Area: 47419 decanoate Area: 204098 succinate Area: 156539

Rt: 5.685 Name: Isoamyl alcohol Area: 413249 Rt: 9.991 Name: Ethyl octanoate Area: 283484 Rt: 12.363 Name: Dodecamethylp entasiloxane Area: 46508 Rt: 14.927 Name: Ethyl decanoate Area: 238996 Rt: 15.892 Name: Ethyl succinate Area: 132457

Rt: 5.685 Name: Isoamyl alcohol Area: 284374 Rt: 9.988 Name: Ethyl octanoate Area: 126614 Rt: 14.920 Name: Ethyl decanoate Area: 211547 Rt: 15.886 Name: Ethyl succinate Area: 72142 Rt: 23.754 Name: Ethyl dodecanoate Area: 172023

Rt: 24.531 Name: Phenylethyl Alcohol Area: 99331 Rt: 25.353 Name: .2,2,4,4,5,5,7,7- Octamethyl- 3,6-dioxa- 2,4,5,7- tetrasilaoctane Area: 33269 Rt: 27.424 Name: Tetrakis(trimeth ylsiloxy)silane Area: 36596 Rt: 29.019 Name: Tetradecamethy lhexasiloxane Area: 28299 Rt: 23.311 Name: Hexanoic acid Area: 8021

Rt: 5.680 Name: Isoamyl alcohol Area: 270764 Rt: 9.986 Name: Ethyl octanoate Area: 122176 Rt: 14.915 Name: Ethyl decanoate Area: 133113 Rt: 15.887 Name: Ethyl succinate Area: 75383 Rt: 23.746 Name: Ethyl dodecanoate Area: 80492

Rt: 24.529 Name: Phenylethyl Alcohol Area: 120059 Rt: 25.348 Name: .2,2,4,4,5,5,7,7- Octamethyl- 3,6-dioxa- Rt: 27.416 Name: Tetrakis(trimeth ylsiloxy)silane Area: 47557 Rt: 29.014 Name: Tetradecamethy lhexasiloxane Area: 38806 Rt: 23.309 Name: Hexanoic acid Area: 9161

(47)

2,4,5,7- tetrasilaoctane Area: 42375

Rt: 5.680 Name: Isoamyl alcohol Area: 227115 Rt: 6.247 Name: Ethyl hexanoate Area: 19395 Rt: 9.986 Name: Ethyl octanoate Area: 44016 Rt: 12.361 Name: Dodecamethylp entasiloxane Area: 31211 Rt: 15.890 Name: Ethyl succinate Area: 68091

Rt: 5.683 Name: Isoamyl alcohol Area: 244239 Rt: 6.251 Name: Ethyl hexanoate Area: 29756 Rt: 9.986 Name: Ethyl octanoate Area: 54985 Rt: 12.351 Name: Dodecamethylp entasiloxane Area: 47427 Rt: 15.887 Name: Ethyl succinate Area: 70399

Rt: 23.211 Name: Pentanoic acid Area: 235918 Rt: 24.532 Name: Phenylethyl Alcoho Area: 367526 Rt: 27.262 Name: Octanoic acid Area: 137948 Rt: 29.735 Name: Decanoic acid Area: 322025 Rt: 23.309 Name: Hexanoic acid Area: 25950

Rt: 5.682 Name: Isoamyl alcohol Area: 193634 Rt: 6.250 Name: Ethyl hexanoate Area: 58923 Rt:9.991 Name: Ethyl octanoate Area: 270454 Rt: 14.911 Name: Ethyl decanoate Area: 56212 Rt: 23.210 Name: Pentanoic acid Area: 534590

Rt: 24.531 Name: Phenylethyl Alcohol Area: 57305 Rt: 25.007 Name: Butylated Hydroxytoluene Area: 22193 Rt: 27.264 Name: Octanoic acid Area: 463571 Rt: 29.734 Name: Decanoic acid Area: 334390 Rt: 23.304 Name: Hexanoic acid Area: 46899