Olfactory Impact of Higher Alcohols on Red Wine Fruity Ester Aroma

Nov 3, 2015 - *(J.-C.B.) Phone: +33 5 57 57 58 63. .... 2-Methylbutyl acetate in wines: Enantiomeric distribution and sensory impact on red wine fruit...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/JAFC

Olfactory Impact of Higher Alcohols on Red Wine Fruity Ester Aroma Expression in Model Solution Margaux Cameleyre,†,‡ Georgia Lytra,†,‡ Sophie Tempere,†,‡ and Jean-Christophe Barbe*,†,‡ †

Univ. Bordeaux, ISVV, EA 4577 Œnologie, F 33140 Villenave d’Ornon, France INRA, ISVV, USC 1366 Œnologie, F 33140 Villenave d’Ornon, France

‡

S Supporting Information *

ABSTRACT: This study focused on the impact of five higher alcohols on the perception of fruity aroma in red wines. Various aromatic reconstitutions were prepared, consisting of 13 ethyl esters and acetates and 5 higher alcohols, all at the average concentrations found in red wine. These aromatic reconstitutions were prepared in several matrices. Sensory analysis revealed the interesting behavior of certain compounds among the five higher alcohols following their individual addition or omission. The “olfactory threshold” of the fruity pool was evaluated in several matrices: dilute alcohol solution, dilute alcohol solution containing 3-methylbutan-1-ol or butan-1-ol individually, and dilute alcohol solution containing the mixture of five higher alcohols, blended together at various concentrations. The presence of 3-methylbutan-1-ol or butan-1-ol alone led to a significant decrease in the “olfactory threshold” of the fruity reconstitution, whereas the mixture of alcohols raised the olfactory threshold. Sensory profiles highlighted changes in the perception of fruity nuances in the presence of the mixture of higher alcohols, with specific perceptive interactions, including a relevant masking effect on fresh- and jammy-fruit notes of the fruity mixture in both dilute alcohol solution and dearomatized red wine matrices. When either 3-methylbutan-1-ol or butan-1-ol was added to the fruity reconstitution in dilute alcohol solution, an enhancement of butyric notes was reported with 3-methylbutan-1-ol and freshand jammy-fruit with butan-1-ol. This study, the first to focus on the impact of higher alcohols on fruity aromatic expression, revealed that these compounds participate, both quantitatively and qualitatively, in masking fruity aroma perception in a model fruity wine mixture. KEYWORDS: red wine, perceptive interaction, aromatic reconstitution, higher alcohols, ethyl esters and acetates, masking effect

■

varietal compounds, such as furanones15 (Furaneol and homofuraneol, for example), and C13-norisoprenoids, such as β-damascenone, in the expression of fruity aromas.9,15,16 Moreover, several molecules, including, for example, dimethyl sulfide12,17,18 (truffle notes), whiskey lactone10 (woody aromas), methoxypyrazines19 (vegetal notes), diacetyl20 (buttery notes), acetic acid20 (vinegar notes), and phenylacetaldehyde21 (rose nuances), may modulate the perception of fruity aromas and, in some cases, mask the perception of fruity notes. From a quantitative standpoint, higher alcohols (HA) represent the main group of volatile compounds in many alcoholic beverages3,22 and, although their presence has long been known,23 their sensory impact has not been exhaustively studied. HA are formed by yeast during the fermentation process, either from grape amino acids via the Ehrlich pathway or directly from sugars. The use of radioactive markers revealed that, under enological conditions, 35% of HA derived from sugars and 65% from amino acids.24 Formation of these compounds depends on various factors: yeast strain, fermentation temperature, must pH, or even aeration.25 Also, the variety and maturity of grape berries affect concentrations

INTRODUCTION Over a thousand aromatic compounds have been identified in wine, including alcohols, esters, organic acids, and ketones, among others. Although it is highly probable that not all of them contribute to wine aroma, this variety of chemical families gives wine its tremendous aromatic complexity.1−3 In the literature, red wines are often described by fruity aromas, such as red-berry- and black-berry-fruit.4,5 Many authors have contributed to the characterization of aromatic compounds involved in the fruity notes of red wines, but no “key” aromatic molecules, responsible for fruity aroma typicality, have yet been identified.6−8 Pineau et al. 9 demonstrated the existence of a typical fruity character of red wines, characterized by black- and red-fruit aromas, and identified a fruity pool consisting of 12 compounds that contributed to these aromas. These studies showed that at least part of the fruity aroma of red wines results from perceptual interactions between various aromatic compounds, particularly ethyl esters and acetates, even if these compounds are present at concentrations below their olfactory thresholds. Many examples of the enhancing effects of aromatic molecules have been studied.10−13 Lytra et al.14 showed that adding certain ethyl esters and acetates increased the perception of fruity aromas, thanks to synergistic effects. Whereas interactions among fruity compounds have been exhaustively studied, interactions between fruity and non-fruity compounds have received little attention. Indeed, some previous studies have described the indirect role of some © 2015 American Chemical Society

Received: Revised: Accepted: Published: 9777

July 16, 2015 October 12, 2015 October 18, 2015 November 3, 2015 DOI: 10.1021/acs.jafc.5b03489 J. Agric. Food Chem. 2015, 63, 9777−9788

Article

Journal of Agricultural and Food Chemistry Table 1. Ethyl Ester, Acetate, and Higher Alcohol Concentrations Used for Sensory Analysesa ethyl esters and acetates (μg/L)

C3C2

C4C2

C6C2

C8C2

2MeC3C2

(2S)2MeC4C2

(2S)- and (2R)2OH4MeC5C2 (95:5, m/m)

150

200

200

200

250

50

400

higher alcohols (mg/L)

C2C4

C2C6

C2iC4

C2iC5

3OHC4C2

3MeC4C2

2MB

3MB

2MP

P

B

10

2

50

250

300

50

50

200

100

30

4

a

C3C2, ethyl propanoate; C4C2, ethyl butanoate; C6C2, ethyl hexanoate; C8C2, ethyl octanoate; 2MeC3C2, ethyl 2-methylpropanoate; (2S)2MeC4C2, (S)-ethyl ethyl (2S)-2-methylbutanoate; (2S)- and (2R)-2OH4MeC5C2, ethyl (2S)- and (2R)-2-hydroxy-4-methylpentanoate; C2C4, butyl acetate; C2C6, hexyl acetate; C2iC4, 2-methylpropyl acetate; C2iC5, 3-methylbutyl acetate; 3OHC4C2, ethyl 3-hydroxybutanoate; 3MeC4C2, ethyl 3methylbutanoate; 2MB, 2-methylbutan-1-ol; 3MB, 3-methylbutan-1-ol; 2MP, 2-methylpropan-1-ol; P, propan-1-ol; B, butan-1-ol. not contain any trace of the compounds included in this study and had a very low-intensity neutral aroma. The fruity reconstitution (FR) was prepared with 13 ethyl esters and acetates in dilute alcohol solution or DRW at the average concentrations found in red wines.13 Complete reconstitution (CR) was achieved by the mixture of these 13 ethyl esters and acetates (at the same concentrations) in dilute alcohol solution or DRW, with the 5 HA at levels representative of those found in wine according to Ribereau-Gayon et al.27 and as described in Table 1. Aromatic reconstitutions were also established in a Vin de Pays d’Oc (Cabernet Sauvignon), chosen for its low HA content (Table 2) and fruity aroma quality, as evaluated by four experts of the laboratory staff (higher alcohol levels were quantitated in duplicate in five red wines (results not shown)).

of these alcohols, due to quantitative and qualitative differences in the amino acid composition of the must.26 HA are present in total concentrations ranging between 0.2 and 1.2 g/L in white wines and between 0.4 and 1.4 g/L in red wines.27 HA include aliphatic alcohols, such as propan-1-ol, 2methylpropan-1-ol, 2-methylbutan-1-ol, and 3-methylbutan-1ol, and aromatic alcohols, 2-phenylethan-1-ol being one of the most important for wine flavor.28 With the exception of 2phenylethan-1-ol, reminiscent of roses, they are described as smelling of fusel oil, solvent, or malt.29 Research has suggested that HA may contribute to the aromatic complexity of wine or, in other cases, mask certain flavors, depending on their concentrations.2 Below 300 mg/L, HA are usually considered to contribute to the desirable complexity of wine, whereas at concentrations exceeding 400 mg/L, they are regarded as having a negative impact on wine quality.2 3-Methylbutan-1-ol, 2-methylbutan-1-ol, 2-methylpropan-1-ol, propan-1-ol, and butan-1-ol have similar chemical structures and odors in wine. Their sensory impact in red wine has not been studied in depth, especially their capacity to affect the characteristic fruity aroma expression. The goal of this work was to study the qualitative and quantitative impact of these HA on a fruity mixture composed of 13 ethyl esters and acetates, corresponding to the red wine fruity pool.

■

Table 2. Amount of Higher Alcohols Added to Wine for Sensory Profilesa concentration in wine (mg/L) amount added to wine to produce a low concentration (mg/L) amount added to wine to produce a medium concentration (mg/L) amount added to wine to produce a high concentration (mg/L)

2MB

3MB

2MP

P

B

44.5 16

202 0

37 13

18 0

0 1.3

41

88

43

2

2

56

148

78

22

2.7

a 2MB, 2-methylbutan-1-ol; 3MB, 3-methylbutan-1-ol; 2MP, 2methylpropan-1-ol; P, propan-1-ol; B, butan-1-ol.

MATERIALS AND METHODS

Chemicals and Odorant Stimuli. Absolute ethanol (analytical grade, 99.97%) and sodium sulfate (99%) were provided by Scharlau Chemie S.A, Barcelona, Spain. Microfiltered water was obtained using a Milli-Q Plus water system (resistivity = 18.2 MΩ cm; Millipore, Saint-Quentin-en-Yvelines, France). Tartaric acid and sodium hydroxide were acquired from VWR-Prolabo, Fontenay-sous-Bois, France. Standard grade purity compounds were obtained from commercial sources as follows: ethyl propanoate, ethyl 2-methylpropanoate, ethyl butanoate, ethyl hexanoate, ethyl octanoate, ethyl 3hydroxybutanoate, 2-methylpropyl acetate, butyl acetate, ethyl 3methylbutanoate, hexyl acetate, 2-methylbutan-1-ol, 3-methylbutan-1ol, 2-methylpropan-1-ol, propan-1-ol, and butan-1-ol from SigmaAldrich, Saint-Quentin-Fallavier, France; 3-methylbutyl acetate from VWR-Prolabo, Fontenay-sous-Bois, France; ethyl (2S)-2-methylbutanoate and ethyl (2R)- and (2S)-2-hydroxy-4-methylpentanoate (95:5, m/m) were synthesized by Hangzhou Imaginechem Co., Ltd. (Hangzhou, China). Aromatic Reconstitution. Dilute alcohol solution was prepared with ethanol and microfiltered water to obtain an ethanol level of 12% vol (v/v) and 5 g/L of tartaric acid (pH adjusted to 3.5 with sodium hydroxide). Dearomatized red wine (DRW) was prepared according to the method described by Lytra et al.,20 by evaporating red wine to twothirds of original volume using a Rotavapor (Laborota 4010 digital Rotary Evaporator, Heidolph, Germany) with a 20 °C bath temperature. The liquid was then mixed with ethanol and microfiltered water to reproduce the alcohol concentration and volume of the original wine. The DRW was then supplemented with 5 g/L LiChrolut EN resin (40−120 μm) and stirred for 12 h. The resulting DRW did

Gas Chromatography−Olfactometry (GC-O) Analyses. GC-O analyses were carried out to ensure that the high-purity reference compounds did not contain any odoriferous impurities and to ascertain that the compound considered was responsible for the odor properties identified. Olfactometry analyses were carried out using an HP-6890 gas chromatograph (HP, Wilmington, DE, USA), equipped with a flame ionization detector (FID) and a sniffing port (ODO-I SGE, Ringbow, Australia), and connected to the column exit by a flow-splitter. GC effluent was combined with humidified nitrogen (Air Liquide, France) at the bottom of the glass-sniffing nose (SGE, Victoria, Australia) to avoid nasal dehydration. A microvolume of each pure odorant was directly injected in splitless−split mode (injector temperature, 240 °C; splitless time, 30 s; split flow, 50 mL/min). The column was a BP21 (SGE, Ringwood, Australia), 50 m × 0.32 mm i.d., and film thickness was 0.25 μm. The oven was programmed at 40 °C for the first 4 min and the temperature increased at a rate of 10 °C/ min to a final isotherm at 230 °C for 10 min. The carrier gas was hydrogen 5.5 (Air Liquide, France) with a column head pressure of 15 psi. Ethyl Ester and Acetate Analysis. Chromatographic conditions and sample preparation were as optimized by Antalick et al.30 The fiber (Supelco, Bellefonte, PA, USA) was coated with 100 μm stationary-phase, polydimethylsiloxane film (PDMS-100). A 10 mL sample was placed in a 20 mL headspace vial, 3.5 g of sodium chloride were added, and the vial was tightly sealed with a PTFE-lined cap. The solution was homogenized in a vortex shaker and then loaded onto a Gerstel autosampling device (Mülheim an der Ruhr, Germany). The program consisted of swirling the vial at 500 rpm at 40 °C for 2 min, 9778

DOI: 10.1021/acs.jafc.5b03489 J. Agric. Food Chem. 2015, 63, 9777−9788

Article

Journal of Agricultural and Food Chemistry then inserting the fiber into the headspace at 40 °C for 30 min as the solution was swirled again, and then transferring the fiber to the injector for desorption at 250 °C for 15 min. Gas chromatography analyses were carried out on an HP 5890 GC system coupled to an HP 5972 quadrupole mass spectrometer (Hewlett-Packard), equipped with a Gerstel MPS2 autosampler. Injections were in splitless mode for 0.75 min, using a 2 mm i.d. non-deactivated direct linear transfer (injector temperature, 250 °C; interface temperature, 280 °C) and a BP21 capillary column (50 m × 0.32 mm, film thickness, 0.25 μm, SGE). The oven temperature was programmed at 40 °C for 5 min, then raised to 220 °C at 3 °C/min, and held at that temperature for 30 min. The carrier gas was helium N5.5 (Air Liquide, France) with a column-head pressure of 8 psi. The mass spectrometer was operated in electron ionization mode at 70 eV with selected-ion-monitoring (SIM) mode. Higher Alcohol Analyses. Fifty microliters of internal standard (4-methylpentan-2-ol 50 g/L in pure alcohol) were added to a 5 mL sample. The solution was homogenized in a vortex shaker, and a microvolume was injected in split mode into an HP-6890 gas chromatograph, coupled to a flame ionization detector (injector temperature, 200 °C), using a CP-Wax 57 CB column (50 m × 0.32 mm i.d.; film thickness, 0.25 μm; Varian). The oven was programmed at 40 °C for the first minute and raised to 200 °C at 8 °C/min, the final isotherm lasting 20 min. The carrier gas was hydrogen 5.5 (Air Liquide, France). Sensory Analyses. General Conditions. Sensory analyses were performed as described by Martin and de Revel.31 Samples were evaluated at controlled room temperature, in individual booths,32 using covered, black ISO glasses,33 containing about 50 mL of liquid, coded with three-digit random numbers. Sessions lasted approximately 5 min. Sensory Panels. Panel 1 consisted of 18 judges, 8 males and 10 females, aged 30.9 ± 7.7 (mean ± SD) years. Panel 2 consisted of 16 judges, 7 males and 9 females, aged 30.4 ± 6.7 (mean ± SD) years. All panelists were research laboratory staff at ISVV, Bordeaux University, selected for their experience. Discriminative Testing Methods. Triangular tests were performed, by panel 1 for various aromatic reconstitution samples31,34 (Tables 3

and 4). A first set of triangular tests (Table 3; tests 1−11) consisted of evaluating the individual perception of each higher alcohol in the fruity

Table 4. Olfactory Impact of the Omission of Higher Alcohols from the Complete Reconstitutiona

complete reconstitution in dilute alcohol solution test 12 test 13 test 14 test 15 test 16 test 17 test 18 test 19 test 20 test 21 test 22 test 23 test 24 test 25 test 26 test 27 test 28 test 29 test 30 test 31 test 32 test 33 test 34 test 35 test 36 test 37 test 38 test 39 test 40 test 41

Table 3. Olfactory Impact of the Individual Addition or the Mixture of Higher Alcohols in Different Matricesa

fruity reconstitution in dilute alcohol solution test 1 test 2 test 3 test 4 test 5 test 6 dilute alcohol solution alone test 7 test 8 test 9 test 10 test 11

13 ethyl esters and acetates

2MB

3MB

2MP

P

B

x

−

−

−

−

−

x x x x x x

x − − − − x

− x − − − x

− − x − − x

− − − x − x

− − − − x x

−

−

−

−

−

−

− − − − −

x − − − −

− x − − −

− − x − −

− − − x −

− − − − x

difference observed

13 ethyl esters and acetates

2MB

3MB

2MP

P

B

x

x

x

x

x

x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

− x x x x − − − − x x x x x x − − − − − − x x x x x − − − −

x − x x x − x x x − − − x x x − − − x x x − − − x − x − − −

x x − x x x − x x − x x − − x − x x − − x − − x − − − x − −

x x x − x x x − x x − x − x − x − x − x − − x − − − − − x −

x x x x − x x x − x x − x − − x x − x − − x − − − − − − − x

difference observed

= *** = = = *** ** = = *** *** *** = = = *** *** *** = ** ** *** *** *** = * * *** *** ***

a ***, 0.1% significance level; **, 1% significance; *, 5% significance level; =, no significant difference; x, presence of listed compounds; −, absence of listed compounds; 2MB, 2-methylbutan-1-ol; 3MB, 3methylbutan-1-ol; 2MP, 2-methylpropan-1-ol; P, propan-1-ol; B, butan-1-ol

= *** = = * ***

reconstitution (FR) and in dilute alcohol solution. Each higher alcohol, present at the concentrations listed in Table 1, was compared to FR or dilute alcohol solution alone. In the second phase, the same panel was subjected to triangular tests (Table 4; tests 12−41): first, the omission of one higher alcohol (Table 4; tests 12−16) and then two (Table 4; tests 17−26), three (Table 4; tests 27−36), and four alcohols (Table 4; tests 37−41). For each triangular test, three numbered samples were presented in random order: two identical and one different. Each judge used direct olfaction to identify the sample perceived as different in each test and gave an answer, even if s/he was not sure. The results of all of the triangular tests were statistically analyzed, according to the tables given in the literature, on the basis of the binomial law corresponding to the distribution of answers in this type of test.31 Olfactory thresholds were determined by panel 1 for 3methylbutan-1-ol and butan-1-ol, using a three-alternative forcedchoice presentation (3-AFC) in two different matrices: dilute alcohol solution alone and FR in dilute alcohol solution.35 Each session

*** *** = = =

a ***, 0.1% significance level; **, 1% significance; *, 5% significance level; =, no significant difference; x, presence of listed compounds; −, absence of listed compounds; 2MB, 2-methylbutan-1-ol; 3MB, 3methylbutan-1-ol; 2MP, 2-methylpropan-1-ol; P, propan-1-ol; B, butan-1-ol

9779

DOI: 10.1021/acs.jafc.5b03489 J. Agric. Food Chem. 2015, 63, 9777−9788

Article

Journal of Agricultural and Food Chemistry

Figure 1. Evolution of higher alcohol concentrations during a sensory analysis session in (a) dilute alcohol solution, (b) fruity reconstitution, and (c) dearomatized red wine. consisted of 10 forced-choice tests. Each test contained one positive sample supplemented with various concentrations of the compound to be evaluated, increasing by a dilution factor of 2 in each step (concentrations from 1.95 to 1000 mg/L for 3-methylbutan-1-ol and from 3.9 to 2000 mg/L for butan-1-ol) (Table S1 in the Supporting Information). The “olfactory threshold” of specific mixtures were also established. A range of concentrations of FR were tested (from 0.1 to 50 mL), with a dilution factor of 2. The “olfactory threshold” of the FR was thus measured in different matrices: dilute alcohol solution, dilute alcohol solution containing a compound of interest, and dilute alcohol solution containing HA at various concentrations (Table S2 in the Supporting Information). The detection threshold was defined as the concentration at which the probability of detection was 50%. This statistical value was determined using an adaptation of the ASTM-E1432 method.35 The concentration/response function is a psychometric function and fits a sigmoid curve (y = 1/ (1 + e(−λx))). Detection probability was corrected using the chance factor (P = (3·p − 1)/2, where p = proportion of correct responses for each concentration and P = proportion corrected by the chance effect, 1/3 for 3-AFC). Sigma Plot

8 (SYSTAT) software was used for graphic resolution and ANOVA transform for nonlinear regression.36 The significance of the observed difference between olfactory thresholds was statistically tested by calculating the 95% confidence interval on the detection probabilities. In addition, interaction effects for certain mixtures were evaluated using Feller’s additive model,37 adapted by Miyazawa et al.38 The probability of detecting the mixture, p(AB), was defined as follows: p(AB) = p(A) + p(B) − p(A)p(B), where p(A) represents the probability of detecting component A and p(B) that of detecting component B. If the panel’s detection performance for the mixture was below the sum of probabilities, some degree of suppression had occurred relative to statistical independence. A performance above the sum of probabilities indicated that some form of mutual enhancement or synergy had occurred. Moreover, if detection performance matched the sum of probabilities, no mixture interaction had occurred. Descriptive Testing Methods. Descriptive analyses of 3-methylbutan-1-ol and butan-1-ol were carried out by panel 2 in dilute alcohol solution, using 500 and 2000 mg/L concentrations, respectively. Descriptive analyses of a 5 HA mixture containing 200 mg/L of 2methylbutan-1-ol and 2-methylpropan-1-ol, 500 mg/L of 3-methyl9780

DOI: 10.1021/acs.jafc.5b03489 J. Agric. Food Chem. 2015, 63, 9777−9788

Article

Journal of Agricultural and Food Chemistry

Figure 2. Detection probability of fruity reconstitution in different matrices: aromatic impact of higher alcohols at (a) low, (b) medium, and (c) high concentrations added to fruity reconstitution in dilute alcohol solution. ∗, expressed in milliliters of total fruity reconstitution (FR) diluted in 50 mL of matrix. DAS, dilute alcohol solution; HA, higher alcohols. The curves are drawn according to a sigmoid function. the level of interaction: σ = Imix/(IFR + IHA), where Imix is the perceived odor intensity of the mixture. τ and σ were calculated for the intensity of fresh- and jammy-fruit aroma. The mean experimental results for the panel were presented using the synthetic representation σ = f(τ). The graph was divided into several parts, according to the interaction level. The position of experimental points reflects the interaction level. Cain and Drexler40 referred to mixture interactions in terms of the overall perceived intensity of a mixture compared to the intensity of each separate component. These authors specified several cases of perceived strength of a mixture. First, the intensity may be as strong as the sum of the perceived intensities of the unmixed components, exemplifying complete addition (σ = 1). The intensity may be also more intense than the sum of its components, exemplifying hyperaddition (σ > 1), or less intense than the sum of its components, exemplifying hypoaddition (σ < 1). In addition, Frijters41 differentiated three cases of hypoaddition: the terms “partial addition”, “compromise”, and “subtraction” are used if the quality intensity of the mixture is greater than, intermediate to, or smaller than that of the individual compounds. For each sample, the significance of the observed perceptual interaction was statistically tested by calculating the 95% confidence interval on the mean intensity of the 16 subjects for both σ and τ.

butan-1-ol, 100 mg/L of propan-1-ol, and 50 mg/L of butan-1-ol were also carried out by the same panel. Judges were asked to complete a free listing task, choosing a maximum of five descriptors. Various reconstitutions in dilute alcohol solution or DRW were presented to panel 2 to evaluate sensory profiles for overall aroma, fresh- and jammy-fruit, solvent, and butyric notes. The first glass consisted of dilute alcohol solution or DRW supplemented with the 13 ethyl esters and acetates at the concentrations indicated in Table 1. The second contained dilute alcohol solution or DRW supplemented with the 5 HA (individually or mixed) at the concentrations indicated in Table S3 (Supporting Information). The third consisted of dilute alcohol solution or DRW supplemented with both the 13 ethyl esters and acetates and the 5 HA (individually or mixed) at the concentrations indicated in Table S3 (Supporting Information). Four samples of aromatic reconstitutions in red wine were presented. The first consisted of red wine alone. The other three glasses contained wine spiked with HA to achieve their low, medium, and high concentrations in wines (Table 2; levels of HA have been assayed in >60 red wines). Each sample was presented twice during each session. For each sample, the subject rated the intensity of five descriptors (overall aroma, fresh- and jammy-fruit, solvent, and butyric) on a 10 cm scale printed on paper, labeled “no odor perceived” on the left and “very intense” on the right. Statistical data were analyzed using the Wilcoxon signed-rank statistical nonparametric test (XLSTAT software). All descriptors are mean-centered per panelist and scaled to unit variance. The statistically significant level was 5% (p < 0.05). Experimental data were also reported on a graph based on two parameters [σ = f(τ)] for binary mixtures, as proposed by Patte and Laffort.39 τ refers to the ratio of perceived intensity of the FR made from the 13 ethyl esters and acetates, to the sum of perceived intensities of mixture’s components individually, prior to mixing: τ = IFR/(IFR + IHA), where IFR and IHA are the perceived odor intensities of a reconstitution containing 13 ethyl esters and acetates and HA at different concentrations, respectively, prior to mixing. σ reflects the ratio between the perceived intensity of the mixture and the sum of the perceived intensities of its components, prior to mixing, and reflects

■

RESULTS AND DISCUSSION Preliminary Verification. Odorant Stimulus Purity. All compounds used were olfactively pure, and any olfactory impurities were detected by the three judges who performed this analysis. Moreover, FID analysis confirmed the products’ very high purity (>99%). Evolution of Sample Composition during Sensory Analysis. The evolution kinetic of the HA in dilute alcohol solution, fruity reconstitution (FR), and dearomatized red wine (DRW) was evaluated to assess the stability of the composition of the samples submitted to the panel. This kinetic evaluation demonstrated that HA remained stable in dilute alcohol solution for 40 min (Figure 1a). Afterward, the concentrations 9781

DOI: 10.1021/acs.jafc.5b03489 J. Agric. Food Chem. 2015, 63, 9777−9788

Article

Journal of Agricultural and Food Chemistry

HA in the mixture and that the effects observed were not due to the composition of the matrix. Organoleptic Impact of Higher Alcohols on Quantitative Perception. Perception of Compounds, Individually and in Mixtures. As indicated in Table 3, in dilute alcohol solution, the addition of 3-methylbutan-1-ol or butan-1-ol or the five HA in FR was perceived by the panel (tests 2, 5, and 6). However, when each high alcohol was added to dilute alcohol solution separately, only 2- and 3-methylbutan-1-ol were perceived (Table 3, tests 7 and 8). These results revealed that, in this matrix, 2- and 3-methylbutan-1-ol were present in suprathreshold concentrations. These findings are in agreement with olfactory thresholds previously established for 2- and 3methylbutan-1-ol in dilute alcohol solution (10% ethanol): 40 and 30 mg/L, respectively.43,44 These results also highlight the interesting role of butan-1-ol, which affected the aroma of the mixture, even when added at subthreshold concentrations (the olfactory threshold of butan-1-ol was evaluated at 150 mg/L2). The result for 2-methylpropan-1-ol (test 9) was not consistent with the threshold established by Guth43 in a similar matrix (40 mg/L in water/ethanol, 90:10 w/w), as no olfactory effect was observed at a concentration 2.5 times higher. The difference between these data may be due to the fact that the olfactory threshold is extremely variable. Indeed, many factors may influence the results of sensory evaluations, such as the size and genetic background of the panel, their familiarity with the proposed sample, the environment of the tasting session (room temperature, type of glasses used, etc.), sample temperature, and the methodology used (coding, presentation order of samples, etc.).16 The findings concerning propan-1-ol and butan-1-ol (tests 10 and 11) are in agreement with the olfactory thresholds reported in the bibliography: 306 and 150 mg/L, respectively.45 Olfactory thresholds for 3-methylbutan-1-ol and butan-1-ol were also determined in our study. The olfactory threshold of 3-methylbutan-1-ol was evaluated at 29.6 mg/L in dilute alcohol solution and at 30.6 mg/L in FR. These values highlighted that this compound had a direct olfactory impact at the concentration tested in both matrices, that is, about 6 times higher than its threshold. The olfactory threshold for butan-1-ol was evaluated at 368.6 mg/L in dilute alcohol solution and at 458.5 mg/L in FR, indicating that this compound was tested at a concentration considerably lower, about 1%, than its olfactory threshold determined under the same experimental conditions, particularly considering the matrix and panel. Sensory effects at such low relative concentrations were reported for some red wine esters for the first time by Pineau et al.9 Hierarchization of the Impact of HA. Omission tests results are presented in Table 4. All omissions involving 3methylbutan-1-ol were significantly perceived by the panel, highlighting the importance of this compound in aromatic reconstitutions (tests 13, 17, 21−23, 27−29, and 33−35). When 3-methylbutan-1-ol and 2-methylbutan-1-ol were presented individually in FR, the judges had more difficulty differentiating the samples, compared to the complete reconstitution (CR) (Table 4, tests 37 and 38). Considering also the impact of the individual omissions of these compounds, these results tend to highlight the important role of 3methylbutan-1-ol among the mixture of HA. This fact is not totally surprising, as this compound was present in the HA mixture at the highest concentration, well above its olfactory threshold. Similar results were observed in some studies using mixtures involving pyridine and linalool, linalyl acetate, or

Figure 3. Aromatic impact of (a) 3-methylbutan-1-ol and (b) butan-1ol added to fruity reconstitution in dilute alcohol solution. ∗, p < 0.05 on the centered and scaled values. FR, fruity reconstitution; 3MB, 3methylbutan-1-ol; B, butan-1-ol. Histograms represent the noncentered and nonscaled values. Error bars indicate standard error deviation.

of the compounds in the solution decreased, with the exception of propan-1-ol. In FR, the HA were more stable over time than in dilute alcohol solution, with