The Pear Aroma in the Austrian Pinot Blanc Wine Variety : Evaluation by Means of Sensorial-Analytical-Typograms with regard to Vintage , Wine Styles , and Origin of Wines

e perceived pear aroma in wines was associated with various shortand medium-chain ester compounds. A consumer study conrmed this assumption. In total, eight ester compounds from a series of 100 aromatic substances were associated with the pear aroma. In this study, a valid stable isotope dilution assay headspace solid-phase microextraction gas chromatography mass spectroscopy (SIDA-HS-SPME-GC-MS) method was developed for the analysis of these compounds, and 102 Austrian Pinot blanc wines of vintages 2013–2016 were analysed. ey were assessed with regard to vintage and origin inuences as well as wine styles. However, an attempt was made to capture the synergies of these compounds for the pear aroma.With the detection of ethyl (E,Z)-2,4-decadienoate and methyl (E)-geranoate, two volatile compounds were measured which had not previously been detected in Austrian wines. e eight analysed esters were perceived very dierently. erefore, specic odour activity values for the various sensations could be calculated with a mathematical combination of the analyses and the results of the sensory studies. A vintage inuence on the sensorial descriptor “overripe pear” and a relationship between wine style and total pear aroma were determined.

Modern sensory studies are generally conducted with the help of trained panels because they are usually more reliable than individual people.Accordingly, the quality of the tasting results depends on the selection and training of the tasters [15].On the other hand, extensive consumer studies have been conducted in order to assess overall attractiveness [15][16][17].A preliminary by Philipp et al. [18] with 81 consumers (3 age groups, 32 females) and by six experts has shown that ethyl (E,Z)-2,4-decadienoate, methyl (E)-geranoate, isoamyl acetate, ethyl dodecanoate, and ethyl decanoate can be clearly identi ed as pear-like aroma (>1/3 of positive answers).According to the study, an aroma is considered a potential pear-like aroma if the substance is perceived by at least two consumers in each age group or by at least nine consumers in total (>10%).Ethyl octanoate, ethyl hexanoate, and isoamyl octanoate are thus also included in the substances described as pear aroma by the consumers.
In the interpretation of GC-MS data, some authors [8,[19][20][21] only considered those aromatic substances, whose odour activity values (OAV) were over a value specified by the author (e.g., in Vilanova et al. [20] OAV > 0.2).In contrast, other researchers suggest that the significance of the synergies of similar smelling compounds should be considered [17,22,23]. is study attempts to capture the synergies of compounds which smell like pear.To our knowledge, there is no exemplary work on this topic.A valid stable isotope dilution assay headspace solid-phase microextraction gas chromatography mass spectroscopy (SIDA-HS-SPME-GC-MS) method should be developed to quantify the aromas largely responsible for the pear aroma of the wines.Applying this method, a selection of Austrian Pinot blanc wines of different vintages will be analysed, and the determined substances will be interpreted with respect to perception and wine style [24].

Samples.
A total of 102 Austrian samples of Pinot blanc were produced for the Austrian market and marketed as Austrian quality wine.All wines were described by the producers as varieties typical to the region.101 samples are from the provinces of Burgenland, Lower Austria, Vienna, and Styria and 1 sample from Tirol.60 wines from the vintage 2016, 34 wines from the vintage 2015, 5 wines from the vintage 2014, and 3 wines from the vintage 2013 were used for the gas chromatographic analysis.Following volatile aromatic substances were quantified: ethyl (E,Z)-2,4decadienoate, methyl (E)-geranoate, isoamyl acetate, ethyl hexanoate, ethyl octanoate, isoamyl octanoate, ethyl decanoate, and ethyl dodecanoate.ese substances were selected for analysis because a previous sensory study showed that these compounds have potential pear-like flavors [18].Two bottles of each wine were procured from the producers in an 8-to 13-month period after the harvest and stored at +4 °C until analysis.Only samples from vintage 2015 and 2016 were used for the study of the influence of the cultivation region and the wine style.

Device Parameters.
For the analysis, a gas chromatograph from Agilent Technologies (6890 N series II (Santa Clara, USA)) was used.e injector temperature was 250 °C.
e column installed was a ZB-WAXplus (60 m × 0.25 mm ID × 0.25 μm d f ) from Phenomenex (Torrance, USA).Helium (grade 5.6 (Linde GmbH (Stadl-Paura, Austria))) was used as the carrier gas at a constant flow of 1.2 ml/min.e total flow was set at 19.04 psi 26.4 ml/min.e unit was operated in the splitless mode.e temperature program began with a holding time of 8 minutes at 40 °C injection temperature and was then heated up to 2 Journal of Food Quality a temperature of 168 °C by an increase of 3 °C/min.en, the oven was heated by 20 °C/min up to 250 °C and kept constant at this temperature for another 5 minutes.e total running time was 75 minutes.A mass-selective detector from Agilent Technologies (Type 5975 (Santa Clara, USA)) was used as the detector.e temperature of the transfer line was 250 °C.e electron multiplier voltage was 1400 V. Samples were measured in single-ion monitoring (SIM) mode.e evaluation was made using the relative ratio of the peak area of the sample to the peak area of the internal standard.

Calibration and Validation of the
Method.An external calibration was carried out in a synthetic wine by means of six calibration points and the use of an internal standard.For the validation, the limit of detection (LOD) and the limit of quantitation (LOQ) were determined in accordance with the German Standard DIN 32645 [27,28].To attain the reproducibility, the coefficient of variation was calculated from ten replicate analyses.e average recovery was identified as quintuple determination for a level of concentration.

Calculations of the Results and Statistics.
In order to describe the pear aroma in wines, the concept of pearspecific odour activity value has been introduced as a mathematical merging of the analytical and sensory aroma analyses.It was calculated according to (1) taking Table 1 into account.e data in Table 1 were collected by two different expert panels using descriptive tests and as per the "best estimated threshold" procedure (threshold value tests) [18,29].e 18 experts for determination of the detection threshold were members of the official sensory wine control.However, the aromas were classified by six other experts into five defined pear groups on a scale of 0-100 mm, analogous to their typicity (overripe pear (T1), fresh pear (T2), cooked, processed pear (T3), exotic pear candy (T4), and oily waxy pearlike (T5)).ese six experts were experienced in the field of descriptive studies and were trained on the different sensations.
e terms (T1 to T5) arose as consensus of these experts [18]: Odour activity value (OAV) is defined as the ratio of each compound to its detection threshold concentration [30].
e different pear-specific odour activity values (OAV overripe pear , OAV fresh pear , OAV cooked, processed pear , OAV exotic pear candy , and OAV oily waxy pear-like ) result from the sum of the products of the defined typicity and the odour activity value of each compound.e overall pear-specific odour activity value (OAV pear ) is the sum of the five OAVgroups [18].
e statistical analysis of the data was performed using SPSS Statistics 22.0 (IBM) regarding differences from the vintage (2013-2016), the origin (Burgenland, Lower Austria, Styria, and Vienna for the vintages 2015 and 2016), and wine style (Leithaberg DAC, wines (>13.0%vol) and wines (≤13.0%vol) for the vintages 2015 and 2016).For the categorisation of the alcohol, the information on the bottle label was used.
First, the data set was tested for normal distribution using an exploratory data analysis.No outliers were eliminated.In the case of a normal distribution, the mean values of the independent samples were compared using a onefactorial analysis of variance and tested for variance homogeneity.If variance homogeneity was proved, an evaluation was done using the Tukey B test at the significance level of 0.05.If there was no variance homogeneity, the Dunnett-T3 test was also used at the significance level of 0.05.If the respective data set was not distributed normally, a Kruskal-Wallis test was conducted at the significance level of 0.05 and evaluated in the form of pairwise comparisons.e influence of the vintage, the origin, and the wine style on the concentration of the individual measured aromatic substances and the calculated pear-specific OAV could be examined using this test [31][32][33].

Results of the HS-SPME-GC-MS-Method Development.
In addition to the relevant aroma compounds-isoamyl acetate, ethyl hexanoate, ethyl octanoate, isoamyl octanoate, ethyl decanoate, methyl (E)-geranoate, ethyl dodecanoate, and ethyl (E,Z)-2,4-decadienoate-other methyl and ethyl esters of cis-trans isomers of decadienoic acid were also determined according to Brandes et al. [25,26].On the one hand, all of these analytes belong to the same group of chemicals, and on the other hand, they have different polarities which makes the analysis difficult [9]. e expected concentration ranges of the different analytes span from Journal of Food Quality is fact also represents a large challenge to the method [33].
For the method development, three di erent bres (100 µm PDMS, 95 µm carbon wide range (WR)/PDMSbipolar, and 85 µm polyacrylate) were tested.A satisfying response could only be achieved with the carbon wide range bipolar bre in all compounds.e polar compound isoamyl acetate produced no response with respect to PDMS bre as well as ethyl (E,Z)-2,4-decadienoate with respect to polyacrylate.
e response of ethyl decanoate and ethyl dodecanoate was also extremely low with polyacrylate.is is consistent with the results of Antalik et al. [9].
As described by various authors [9,[34][35][36][37], there is a correlation between the ionic strength of the solution and the adsorption of the analytes on the bre.e concentration of the salt as well as the choice of the salt is important.Trials were conducted with three di erent salts and di erent amounts (NaCl, NaH 2 PO 4 , and MgSO 4 ).Ethyl (E,Z)-2,4decadienoate in particular showed the highest response with NaH 2 PO 4 .e salt amount for this salt is increased above 1 g/5 ml; however, there were adverse e ects on the adsorption of isoamyl acetate.erefore, 1 g/5 ml NaH 2 PO 4 was chosen for these experiments.
e adsorption temperature is also crucial for the SPME process.In theory, a higher temperature leads to an increase in the partial vapour pressure of the analytes, but at the same time to a reduction of the adsorption onto the bre [9].As shown in Figures 1 and 2, the responses of ethyl (E,Z)-2,4decadienoate and ethyl dodecanoate were poor at 30 °C and reached the optimum at 50 °C.On the other hand, the response of isoamyl acetate decreased already between 30 °C and 40 °C but was still satisfactory at 45 °C.An adsorption temperature of 45 °C and an adsorption time of 30 min were chosen for the method.Several studies [9,24] have shown that wine esters have di erent SPME extraction pro les.Quanti cation problems can only be avoided if the internal standard is structurally as similar as the analyte [9,24,38].erefore, three deuterated standards (d 5 )-ethyl hexanoate (used for isoamyl acetate, ethyl hexanoate), (d 5 )-ethyl octanoate (used for ethyl octanoate), and (d 5 )-ethyl decanoate (used for ethyl decanoate, isoamyl octanoate, methyl (E)-geranoate, ethyl dodecanoate, and decadienoic esters) were synthesised and used.ey behaved similarly to the analytes, and therefore, their choice for the quanti cation of the di erent analytes was useful (Figures 1 and 2).

Stability of the Deuterated Standards. (d 5 )
-Ethyl hexanote, (d 5 )-ethyl octanoate, and (d 5 )-ethyl decanoate were synthesised from (d 6 )-ethanol and the corresponding acid chloride.Since these internal standards show more than one deuterium, their retention times and mass spectra di er slightly from their protonated counterparts.e stability of the deuterated esters should be checked, since transesteri cation reactions can take place in an acidic solution such as wine in the presence of ethanol and other alcohols [9].Stability was measured over 18 hours in wine and in synthetic wine.It has been found that (d 5 )-ethyl hexanoate, (d 5 )-ethyl octanoate, and (d 5 )-ethyl decanoate were suciently stable over a period of at least 8 hours.e sample series were adapted to these conditions, and all samples were analysed in less than 8 hours.ese statements are consistent with the ndings of Antalick et al. [9].

Quanti cation of Decadienoic Acid Esters.
For the qualitative detection of the di erent methyl and ethyl ester of the decadienoic acid, a self-synthesised mixed standard was available [25,26].e quanti cation of the ethyl (E,Z)-2,4decadienoate equivalents was done using a procured pure standard.For the quanti cation of the methyl esters and ethyl esters, the quanti er mass of 182 and 196, respectively, was used to get a good separation of the peaks (Figure 3).e usually used quanti er mass of 151 was not suitable because ethyl dodecanoate showed a response with a similar retention time (Table 2) [25,26].

Calibration and Validation Results of the Analytical
Method.An external calibration was carried out in the synthetic wine by means of six calibration points and the use of an internal standard mixture.e functions were linear with regard to the concentration range of the analytes common in wine (correlation coe cient of all analytes ≥0.989) (Table 2).
e limit of detection (LOD) and limit of quanti cation (LOQ) were determined in accordance with the German Standard DIN 32645 [27,28] and can be found in Table 3.
is is particularly necessary because the LOD and LOQ may be outside the linearity range at high calibration ranges (isoamyl acetate, ethyl hexanoate, ethyl octanoate, and ethyl decanoate) if a signal-noise validation procedure would be performed [40].
e detection limits varied depending on the compound, especially in the concentration range between 0.119 µg/L and 80.580 µg/L and were below the expected concentrations in wine.For the determination of the repeatability (Srt), a wine was measured 10 times.

In uence of Vintage and Origin on the Pear Aroma of
Austrian Pinot Blanc Wines.102 Austrian Pinot blanc wines of di erent vintages and origins were measured for the pear aroma's primary substances using the SIDA-HS-SPME-GC-MS.e concentrations of isoamyl acetate (Table 4) uctuated from 211.1 to 5816.7 µg/L.According to the literature [6], this compound ranges in wine from 30 and 8100 µg/L.Journal of Food Quality  4) also corresponded to the average data from the literature [9,33,43].ese aromas are counted among the classical fermentation aromas (tertiary aromas) and are due to the amino acid or fatty acid metabolism of the yeast [6].As expected, there was no significant difference between the vintages (Table 4 left side) of the wines and the contents of these four aromatic substances.Interestingly, there were determined significant differences between the wines from Vienna (2015 vintage) and Burgenland (2016 vintage) regarding the ethyl hexanoate and between the wines from Burgenland (2016 vintage) and from Styria (2016 vintage) regarding the ethyl decanoate (Table 4 right side).ese differences are likely the result of the choice of wines (sample size and age).Rather, it can be assumed that the largest influence on the content of isoamyl acetate, ethyl hexanoate, ethyl octanoate, ethyl decanoate, and ethyl dodecanoate derived from the fermentation conditions, in particular, the choice of yeast, fermentation temperature, yeast-available nitrogen, pH value, concentrations of unsaturated fatty acids, sterol, and the oxygen content [6,44,45].
Concentrations of 1.89-11.20 µg/L were measured from the isoamyl octanoate compound (Table 4).In the literature, concentrations for red wine are specified in the range from 1.2 to 4.0 µg/L.Antalik et al. [9] assumed that the concentration of this ester is higher in rose wine, white wine, and red wine vinified by carbonic maceration than in classical red wines.41 of the 102 wines, however, showed concentrations below 4.00 µg/L.e average of all 102 wines was 5.01 µg/L, which was slightly higher than the above mentioned range for red wines.ere was no significant impact from the vintage and the origin of wines with this aroma compound.
e Austrian Pinot blanc wines had calculated concentrations from 0.15 to 1.83 µg/L, calculated as ethyl (E,Z)-2,4decadienoate (Table 4).Because these compounds were already verified in grapes of the vine genus Vitis labrusca and their hybrids, it would be possible to assign them to the primary aromas.On the other hand, the results showed that there were no significant differences in vintage and origin.
Philipp et al. [18] found a sensorial detection threshold of 2 μg/L in a 10% alcoholic solution; however, the specific odour threshold for perceived pear aroma was significantly higher.It was determined with an individual high dispersion and low reproducibility using the BET single method and resulted in 20-400 µg/L.However, statistical calculation models have shown that these compounds are indeed relevant for the pear aroma in interaction with other aromatic substances (isoamyl acetate, ethyl dodecanoate, and methyl (E)-geranoate) determined with 24 volunteers [18].
Methyl (E)-geranoate was detected for the first time in Austrian wines and in general, firstly in Pinot blanc with a concentration up to 4.73 μg/L. is value corresponds to the data in the literature.Antalik et al. [9] have found 0.83 µg/L in French red wines.However, this substance was found in high concentrations especially in Gewürztraminer and some sweet wines, for example, Coteaux du Layon from the Chenin blanc grape variety [46].
Table 5 shows the calculated pear-specific odour activity values in relation to the vintage.e total pear aromatic of the wines decreased with increasing age.
is effect was primarily due to the decrease in isoamyl acetate concentrations (Table 4).With the decomposition of this aromatic substance, the sensorial perceived fruitiness of the wines decreased [42]. is is reflected in the significantly lower OAV of the "exotic pear candy" (OAV exotic pear candy ) in the 2013 and 2014 vintages.On the other hand, the OAV of the "overripe pear" (OAV overripe pear) greatly depended on the vintage.2014 and 2016 vintages have been described as rather cool vintages and thus have lower OAV overripe pear than the very warm 2015 vintage.
Table 5 shows the different pear-specific-OAV associated to the origin of the wines.In Austria, well-known Pinot blancs are mainly produced in Styria (Südsteiermark, Vulkanland), Lower Austria ( ermenregion, Weinviertel, Klosterneuburg), Vienna, and Burgenland (Leithaberg DAC).However, the Pinot blanc is represented in all wine regions in Austria [1].It is interesting to note that the wines of Styria (2015 vintage) and Burgenland (2016 vintage) showed lower total pear-specific odour activity value (OAV pear ) than the wines from Vienna (2015 vintage) and Styria (2016 vintage).is is mainly due to the already described lower isoamyl acetate content of wines from Styria (2015 vintage) and the very high isoamyl acetate content of wines from Styria (2016 vintage) and the resulting significantly lower OAV exotic pear candy .Significant differences between the wines from Styria (2016 vintage) and Burgenland (2016 vintage) were found regarding the OAV fresh pear .Since these differences have not been observed in the 2015 vintage, it could be due to the sampling (sample size and age).Similarly, a vintage or origin correlation with OAV cooked, processed pear and OAV oily waxy pear-like also is unlikely.

Influence of the Wine Style and the Alcohol Content on the
Pear Aroma.Since there are no defined Pinot blanc styles in Austria, with the exception of the Leithaberg DAC Journal of Food Quality With regard to the aromatic substances (Table 6), significant differences in the concentrations of isoamyl acetate, ethyl hexanoate, and ethyl octanoate could be observed.e wines from the Leithaberg DAC (2016 vintage) showed the lowest concentrations, while wines with higher alcohol content tended to have higher values than the wines of the category up to 13% vol.To our knowledge, a direct correlation between higher alcohol content and higher ester content is not described in the literature yet.However, it has been shown that the perception of aroma compounds in fact depends on the alcohol content [6].e fact that Leithaberg DAC wines have lower concentrations of isoamyl acetate (2016 significant, 2015 tendency) than other Pinot blanc wines is understandable due to the defined wine style [47].
e sensorial-analytical-typograms of the pear aroma can be seen in Table 7, with regard to wine styles and alcohol content.
e study revealed that wines belonging to the category "above 13.0 vol.% alcohol" tended to have higher total OAV pear . is is due to various aspects.e alcoholrich wines were characterised more by the overripe pear aromas than the wines with lower alcohol levels (2015 significant, 2016 tendency).Similar observations, some of them significant and some of which with tendency, could be made for the other categories: cooked pear, exotic pear candy, and oily waxy pear-like, while the impression exotic pear candy dominated the 2016 vintage.e wines of the 2015 vintage (alcohol content above 13% vol and Leithaberg DAC), however, showed significantly higher contents of overripe pears impression than the wines of the 2016 vintage.
e sensorial-analytical-typograms depict what is well known about wines from these vintages: the 2015 vintage was an overripe vintage, while 2016 was more fresh-fruity one.Wines with higher alcohol content were from more mature grape material than wines with lower alcohol content if not enriched or concentrated.e wines produced in this way showed more mature aromas than light-fruity wines.erefore the division of the wine into two categories according to the alcohol content made sense either for the traditional wine description as well as for the discussion of the presented sensorial-analytical-typograms.

Conclusion
In the course of this present work, it is attempted to describe the synergies of eight short-and middle-chain esters as the main contribution to the pear aroma of Pinot blanc wines.
e pear-specific odour activity values presented as sensorialanalytical-typograms were calculated from the analysed concentration of these esters in 102 Pinot blanc wines from Austria and the results of sensory studies [18].e overall pear-specific odour activity value (OAV pear ) of the 102 wines ranged from 32.4 to 147.7.e largest contributors to the OAV pear were the impressions for "exotic pear candy" (primarily characterised by isoamyl acetate) and "oily waxy pear-like" (characterised by ethyl hexanoate, ethyl decanoate).
e sum of OAV "fresh pear" and "overripe pear" ranged from 8.6 to 24.5; however, they were important for the individual expression.
As cited by Philipp et al. [18], these sensorial-analyticaltypograms serve the representation of subjective sensorial perceptions and highly specific chemical analyses of volatile substances.Since these are simplifications, it must be mentioned that the primary value of the sensorial-analyticaltypograms lies in the consideration of the differences between the individual wines and not in an attempt to obtain a correlation between sensorial and chemical data.
In this paper, the influence of the vintage, the wine styles, and the origin on the pear aromatic of the Pinot blanc wines was examined.Differences in the concentration levels of volatile substances were clear, easy, and understandable, presented in the proposed sensory-analytical-typograms.It was found that there was a vintage influence on the OAV overripe pear .In mature years like 2015, the values are higher than in immature years like 2014 and 2016.In addition, the sensation of "exotic pear candy" drops with increasing age of the wines.
is is accompanied by the reduction of isoamyl acetate [42].On the other hand, an influence of the wine style on the pear aromatic wine was observed.It was determined that the Pinot blanc wines with higher alcohol content (category > 13.0 vol.%) tend to have a higher total pear-specific odour activity value than wines with a lower alcohol content.
ese results are in good accordance with the standard description of Austrian wines from these vintages and wine styles [48].
If it would be possible to measure a larger number of samples over several years, the question of regional disparities could be clarified better.In any case, further studies, especially on the technological and viticultural influence on the pear aromatic, are recommended.
e pear-specific odour activity values model and the presentation in the form of sensorial-analytical-typograms seem well suited for the interpretation of such questions in the future.
Figure 2:e response progression at di erent adsorption temperatures-compounds with (d 5 )-ethyl decanoate as internal standard.

Table 1 :
[18]lts of the expert survey/perception threshold as per the BET process[18].

Table 4 :
Concentrations of esters regarding vintage and wine-growing region.
All the parameters are given with their mean, standard deviation (SD), lowest amount, and highest amount.For each parameter, different letters in the same row indicate significant differences between variants (p ≤ 0.05).

Table 5 :
Pear-specific odour activity value in relation to vintage and wine-growing region (sensorial-analytical-typograms).All the parameters are given with their mean and standard deviation (SD): for each parameter, different letters in the same row indicate significant differences between variants (p ≤ 0.05).

Table 6 :
Concentrations of esters in relation to the wine stylistics and alcohol content.

Table 7 :
Pear-specific odour activity value in relation to wine stylistics and alcohol content (sensorial-analytical-typograms).Water Management.Most of our science projects are directly financed by the ministry.A big thanks to the organizing team of the "In Vino Analytica Scientia 2017," who honoured this sensorial-analytical-typograms concept with the 2nd Poster Award at the Poster Contest and the OIV (Organisation for Vine and Wine), which gave us the opportunity to present the results of the sensory analysis at the OIV Congress in Bulgaria in 2017.e authors would also like to thank the undergraduate students Leona Tönnies, Karoline Taferner, Melis Yaltir, Lukas Plöckinger, Michael Bayer, Stefan Schmiedel, Marcus Bülow, Pirmin Winklhofer, Michael Lutschounig, Josef Stumvoll, Alexander Weinberger, Sezer Sari, and Daniel Gammert who actively helped in collecting wines.