Article pubs.acs.org/JAFC
Characterization of the Key Aroma Compounds in Turkish Olive Oils from Different Geographic Origins by Application of Aroma Extract Dilution Analysis (AEDA) Songul Kesen,† Hasim Kelebek,§ and Serkan Selli*,# †
Department of Food Technology, Naci Topcuoglu Vocational High School, Gaziantep University, 27600 Gaziantep, Turkey Department of Food Engineering, Faculty of Engineering and Natural Sciences, Adana Science and Technology University, 01110 Adana, Turkey # Department of Food Engineering, Faculty of Agriculture, Cukurova University, 01330 Adana, Turkey §
ABSTRACT: The aroma and aroma-active compounds of olive oils obtained from Nizip Yaglik (NY) and Kilis Yaglik (KY) cultivars and the effect of the geographical area (southern Anatolian and Aegean regions) on these compounds were analyzed by gas chromatography−mass spectrometry−olfactometry (GC-MS-O). For this purpose, two oil samples were obtained from their native geographical area including NY from Nizip province and KY from Kilis province (southern Anatolian region of Turkey). Another two oils of the same cultivar, NY-Bornova (NY-B) and KY-Bornova (KY-B), were obtained from the Olive Oil Research Center-Bornova, Izmir province (Aegean region of Turkey) to compare geographical effect on aroma and aroma-active compounds. Simultaneous distillation and extraction (SDE) with dichloromethane was used for extraction of volatile components. SDE gave a highly representative aromatic extract of the studied olive oil based on the sensory analysis. Totals of 61, 48, 59, and 48 aroma compounds were identified and quantified in olive oils obtained from NY, NY-B, KY, and KY-B cultivars, respectively. The results of principal component analysis (PCA) showed that the aroma profile of native region oils was discriminately different from those of Bornova region oils. Aldehydes and alcohols were qualitatively and quantitatively the most dominant volatiles in the oil samples. Aroma extract dilution analysis (AEDA) was used for the determination of aroma-active compounds of olive oils. The number of aroma-active compounds in native region oils was higher than in Bornova region oils. Within the compounds, aldehydes and alcohols were the largest aroma-active compounds in all olive oils. KEYWORDS: olive oil, olfactometry, aroma, aroma-active compounds, geographical effect
■
INTRODUCTION Olive oil is obtained from the olive (Olea europaea), a traditional tree crop of the Mediterranean area. It is almost unique among oils as it can be consumed in crude form without refining, possessing increased stability as well as nutritional and healthy features with respect to other vegetable oils.1 According to olive oil production data released in 2012, Spain (992,000 tonnes), Italy (570,000 tonnes), Greece (351,800 tonnes), Turkey (206,300 tonnes), Syria (200,000 tonnes), and Tunisia (192,600 tonnes) are the leading production countries.2 Aroma compounds stimulate much more qualities and therefore are mainly responsible for the characteristic flavor of foods.3 Olive oil aroma consists of a complex mixture of volatile compounds, which includes mainly aldehydes, alcohol, ketones, and esters. Enzymatic reactions and autoxidation play crucial roles in the occurrence of aroma substances. The formation of these compounds starts at the moment of cell disruption during crushing of olives and continues during the extraction process. Synthesis of the aroma compounds of virgin olive oil (VOO) occurs through the lipoxygenase (LOX) pathway comprising mainly the actuation of LOX and hydroperoxide-lyase enzymes. The volatile compounds found in virgin olive oil are mainly C-6 and C-5 compounds biogenerated from polyunsaturated fatty acids through the lipoxygenase pathway. These compounds are mainly responsible for the characteristic aroma of olive oil, which is of prime © 2013 American Chemical Society
importance in the food industry, because it plays a significant role in consumer choice.4−6 More than 180 different aroma compounds have been found in olive oils.6 However, it is well-accepted that only a small fraction on the large number of volatile compounds occurring in food actually contribute to the overall aroma. The technology of gas chromatography−olfactometry (GC-O) made it possible to divide identified volatiles into odor-active and non-odor-active compounds with regard to their existing concentration in the studied sample.7,8 Limited numbers of researchers have analyzed the aroma-active compounds of olive oils: Guth and Grosch9 determined potent odorants of different virgin olive oils using aroma extract dilution analysis (AEDA). They indicated that (Z)-3 hexenol, (E)-2-hexenal, and (Z)-3hexenal were mainly responsible for the green odor notes of four different olive oils. Reiners and Grosch10 studied the aroma-active compounds of virgin olive oils from Italy, Spain, and Morocco by AEDA and GC-O of headspace samples. Morales and Aparicio7 investigated the best extraction conditions (temperature and time of malaxing) on the basis of odor-active compounds that are responsible for the unique Received: Revised: Accepted: Published: 391
October 8, 2013 December 24, 2013 December 27, 2013 December 27, 2013 dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
attributes of virgin olive oil. Collin et al.11 determined the aroma-active compounds in extracts obtained by simultaneous extraction/distillation (SDE) of Moroccan black and green olives. In this study, odor-active compounds were identified by using the AEDA method and GC-O. Cultivar is one of the most important factors that significantly influences volatile compound composition and sensorial characteristics of virgin olive oil. The same olive cultivar grown in different locations produces oils with different volatile profiles.6 In addition to cultivar, the degree of fruit ripening, environment, growing season, traditional growing habits of various cultivars, extraction methods, processing techniques, and storage conditions affect the aroma profiles of olive oils.12,13 It is well-known that VOO obtained from different cultivars, under identical growth conditions, harvested at roughly equal ripeness degree, and processed in the same manner, is characterized by more or less different composition and concentration of minor components.8,14 Up to now, there is no information on the identification of the main odorants in Turkish olive oils. NY and KY are economically important cultivars to olive production of the southern Anatolian region and used mainly for producing oil. Therefore, the aim of the present study was to investigate aroma and aroma-active compounds in two of the most important monovarietal olive oils (cvs. Nizip Yaglik (NY) and Kilis Yaglik (KY)) produced in two geographical areas of Turkey. Before the GC-MS-O analysis, the aromatic extract obtained from SDE was analyzed by sensory evaluation to assess its representativeness.
■
extraction. Cold-press has been a widely used method for many years to produce high-quality olive oils.16 Olives in the mill after grinding for 30 min underwent malaxation (kneading). Then fractions were divided into decanters. These fractions were oil and wastewater plus oil cake. Fresh olive oils were put into glass bottles and were kept in a dark and cool place until analysis. Standard Chemical Analysis. Moisture, oil content, percentage of free fatty acidity, and peroxide values of the samples were calculated following the methods described in the regulation of the European Economic Commission.17 The color of the olive oil was measured in a colorimeter using a HunterLab Color QuestXE (Reston, VA, USA). Results for L*, a*, and b* color system profile were recorded using CIE. Measurements were made at room temperature. Olive oils were placed in a sample cell (6 cm diameter), 2 cm deep, and the data were read. Extraction of the Volatile Compounds. The volatile compounds of olive oils were extracted by SDE in a Likens-Nickerson apparatus (Neubert-Glas, Geschwenda, Germany). This method has already shown its reliability for the extraction of volatile components of olive oils.18,19 For the extraction, 40 mL of olive oil, 100 mL of pure water, 25 mL of 30% NaCl, and 40 μg of 4-nonanol as the internal standard were put into a 500 mL distillation flask on one hand, and on the other 40 mL of dichloromethane solvent was pipetted into another (a 100 mL distillation flask). Both flasks were placed in a heater, and then extraction was performed for approximately 3 h. The temperatures of the sample mixture and the dichloromethane flasks were maintained by a water bath at 70 and 50 °C, respectively. After the dehydration by anhydrous sodium sulfate, the resulting organic extract was condensed to 5 mL in a Kuderna Danish concentrator (SigmaAldrich, St. Louis, MO, USA) and then to 200 μL under a gentle stream of pure nitrogen.18 Olive oils were extracted in triplicate, and three independent extracts were obtained for each sample. The extracts were subsequently kept at −20 °C in a 2 mL glass vial before the analysis. Representativeness Test of the Extract. Sample Preparation and Presentation. The panel was composed of seven assessors (two females and five males between 25 and 46 years old) from the Department of Food Engineering, University of Cukurova. The assessors were previously trained in odor recognition and sensory evaluation techniques and had experience in gas chromatography− olfactometry. In the present study, we used a cardboard smelling strip (reference 7140 BPSI, Granger-Veyron, Lyas, France) for checking the representativeness of the aromatic extract obtained by SDE. Smelling strips have already given good results for the representativeness test of orange juice20 and Dwarf Cavendish banana extracts.21 Two solvents (dichloromethane and pentane plus dichloromethane) were evaluated for the representativeness using SDE. First, 5 mL of olive oil was placed in a 25 mL brown coded flask as a reference for representativeness tests. Olive oil aromatic extracts obtained from two solvents were adsorbed onto a cardboard smelling strip. After 1 min (the time necessary for solvent evaporation), the extremities of the strips were cut off, and then they were placed in dark coded flasks (25 mL) and presented to the panel after 15 min. Dichloromethane and pentane/dichloromethane (2:1) are very volatile solvents. After evaporation, no panelists detected the odor of the solvents. Samples were assessed at room temperature (20 °C) and in neutral conditions. Similarity and Intensity Tests. Similarity and intensity tests were performed to demonstrate the closeness between the odor of extracts and the olive oil. These test procedures were well documented in our previous studies.20 Results were analyzed with an analysis of variance with Statgraphics Plus software (Manugistic, Inc., Rockville, MD, USA). GC-FID, GC-MS, and GC-MS-O Analyses of Aroma Compounds. The GC system consisted of an Agilent 6890 chromatograph equipped with a flame ionization detector (FID) (Wilmington, DE, USA), an Agilent 5973-network-mass selective detector (MSD), and a Gerstel ODP-2 (Linthicum, MD, USA) sniffing port using deactivated capillary column (30 cm × 0.3 mm) heated at 240 °C and supplied with humidified air at 40 °C. This system allowed us to simultaneously obtain a FID signal for the quantification, an MS signal for the
MATERIALS AND METHODS
Chemicals. The water used in the study was purified by a Millipore-Q system (Millipore Corp., Saint-Quentin, France). The reference aroma compounds were obtained from the following sources: Hexanal, (E)-2-pentenal, (E)-2-hexenal, (Z)-3-hexenal, nonanal, nonanol, (E)-2-heptenal, (E)-2-octenal, and (E)-2-decenal were obtained from Fluka (Buchs, Switzerland). 2-Methyl-1-propanol, 1-penten-3-ol, pentanol, 3-penten-2-ol, hexanol, (E)-3-hexenol, (Z)-3hexenol, (E)-2-hexenol, 2-hexanol, heptanol, 2-ethyl-1-hexanol, (Z)-2pentenol, phenol, hexyl acetate, (Z)-3-hexenyl acetate, ethyl salicylate, methyl palmitate, ethyl palmitate, limonene, linalool, (Z)-4-heptenal, 1-octen-3-ol, (E)-2-nonenal, benzaldehyde, (E,E)-2,4-heptadienal, (E,E)-2,4-decadienal, δ-octalactone, γ-nonalactone, γ-decalactone, δdecalactone, δ-dodecalactone, and γ-dodecalactone were purchased from Sigma-Aldrich (Steinheim, Germany). Octanal, octanol, (E)-2octenol, heptanal, benzyl alcohol, 2-phenylethanol, 6-methyl-5-hepten2-one, β-pinene, α-copaene, α-farnesene, farnesol, acetic acid, octanoic acid, nonanoic acid, decanoic acid, hexanoic acid, dodecanoic acid, and guaiacol came from Merck (Darmstadt, Germany). Pentane, dichloromethane, 4-nonanol, sodium chloride, and sodium sulfate were also obtained from Merck (Darmstadt, Germany). Dichloromethane was freshly distilled prior to use. All chemicals used as standards and solvents were of GC grade having purity up to 99.0%. Olives. Olive varieties (NY, KY, NY-B, and KY-B) averaging 50−70 kg were used in this study. Two olive samples were obtained from their native geographical area including NY from Nizip province and KY from Kilis province (southern Anatolian region of Turkey). To investigate geographical effect, the same olive cultivars were bought from the Olive Research Center located in Bornova-Izmir province (NY-B and KY-B; Aegean region of Turkey). Sound olives were harvested at the optimum maturity stage in 2011 seasons. The maturity indices were calculated according to the International Olive Oil Council method.15 Extraction of the Olive Oils. First, leaves were removed and then the healthy olives washed with water. Then, the cold-press with dualphase centrifuge (Olio Mio Mini, Italy) was used for olive oil 392
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Table 1. Olive Fruits and Olive Oils General Propertiesa olive/olive oil property
a
NY
NY-B
KY
KY-B
maturity index oil content (%) moisture content (%)
4.1 ± 0.00a 26.30 ± 0.12b 50.33 ± 0.62a
3.9 ± 0.00b 21.80 ± 0.10d 47.46 ± 0.75b
4.2 ± 0.00a 30.50 ± 0.17a 47.04 ± 1.07b
3.9 ± 0.00b 24.50 ± 0.14c 43.62 ± 0.42c
free fatty acidity (oleic acid %) peroxide value (mequiv oxygen/kg oil) L* a* b*
0.68 ± 0.08a 6.40 ± 0.16c 60.64 ± 0.06c −1.87 ± 0.03c 73.05 ± 1.69b
0.77 ± 0.18a 8.02 ± 0.13a 64.86 ± 0.04b 2.37 ± 0.02a 102.21 ± 0.13a
0.67 ± 0.14a 7.40 ± 0.18b 51.70 ± 0.05d −2.11 ± 0.02d 74.38 ± 0.10b
0.69 ± 0.08a 8.30 ± 0.16a 68.60 ± 0.03a 1.17 ± 0.01b 101.13 ± 0.10a
Results are mean of three replications ± standard deviation. Values with different letters in same row are significant statistically (p < 0.05).
■
identification, and the odor characteristics of each compound detected by sniffing port. GC effluent was split 1:1:1 among the FID, MSD, and sniffing mode via a Dean’s switch. Aroma compounds were separated on the DB-WAX (30 m length × 0.25 mm i.d. × 0.5 μm thickness; J&W Scientific, Folsom, CA, USA) column. A total of 3 μL of extract was injected each time in pulsed splitless (40 psi; 0.5 min) mode. The injector and FID detectors were set at 270 and 280 °C, respectively. The flow rate of carrier gas (helium) was 1.5 mL/min. The oven temperature of the DB-WAX column was first increased from 50 to 200 °C at a rate of 5 °C/min and then to 260 °C at 8 °C/min with a final hold at 260 °C for 5 min. The same oven temperature programs were used for the MSD. The mass detector was operated in scan mode, with an electronic impact ionization energy of 70 eV. The GCMS interface and ionization source temperatures were set at 250 and 180 °C, respectively. Identification and quantification were performed in full scan mode with a mass/charge range of 30−300 amu at 2.0 scan/s scan rate. The compounds were identified by comparing their retention index and Wiley 6 and NIST 98 mass spectral data libraries installed in the GC-MS and the instrument’s internal library created from previous laboratory studies. Some of the identifications were confirmed by the injection of the chemical standards into the GC-MS system. Retention indices of the compounds were calculated by using the retention data of the linear alkane series. After identification, the concentrations of aroma compounds were calculated according to the internal standard.20 Aroma Extract Dilution Analysis. The original aroma extracts were analyzed by GC-MS-O using three experienced sniffers. For AEDA, the concentrated aromatic extract (200 μL) of olive oils was stepwise diluted 1:1 using dichloromethane as the solvent to obtain dilutions of 1:1, 1:2, 1:4, 1:8, 1:16, and up to 1:1024 of the original extracts.22 Sniffing of dilutions was continued until no odor could be detected by GC-MS-O. Each odor was thus assigned a flavor dilution factor (FD factor) representing the last dilution in which the odor was still detectable. As the FD factor of aroma compounds increases, also the degree of flavor activity increases.23,24 This method has been used in different foods such as green olives25 and orange juices.26 Sensory Analysis of Olive Oils. Sensory analysis of olive oil samples was made by a group of 10 trained panelists according to the published methods of the International Olive Oil Council determining the terms of organoleptic assessment.27 Olive oil samples (15 mL) were placed in tasting glasses coded with different four-digit numbers. The temperatures of samples were kept at 28 ± 2 °C. A profile sheet was used by tasters. Each panelist smelled and then tasted the oil in the tasting glass and detected the intensity of the positive (fruity, bitter, pungent) and negative attributes (fusty/muddy, musty/humid, wineyvinegar, metallic, and rancid) on a 10 cm scale. Statistical Analyses. The findings of this study were subjected to analysis of variance using the SPSS 17 software package, and Duncan’s multiple-comparison test was used to find significant differences.28 On the other hand, principal component analysis (PCA) was carried out using XLStat-Pro7.5 (2007) for Windows (Addinsoft, New York, NY, USA).
RESULTS AND DISCUSSION
Chemical Composition of Olive Oils. General properties of the assayed samples are shown in Table 1. As can be seen in Table 1, maturity indices were found between 3.9 and 4.2. This is an important factor in determining the harvest time of olives and must be between 2.5 and 4.5 for processed olive oil.29 Oil and moisture contents of olives were found between 21.80 and 30.50% and between 43.62 and 50.33%, respectively. Oil contents of olive samples grown in the native region were higher than in olives grown in the Bornova geographical area. In addition, a similar situation was observed in the moisture content of the samples. This finding agrees with previous studies30,31 reporting that a significant difference of olive oil composition exists among the different growing regions. The percentage of free fatty acidity, peroxide values, and color values are usually used for quality characteristics and depend on the quality of the olives and on possible enzymatic alteration of olive fruits.32 The percentage of free fatty acidity values of all olive oils were not significantly different (p < 0.05) and under the limit of 0.8% established for extra virgin olive oil. Peroxide values also were below the limit of 20 mequiv oxygen/ kg oil, which is accepted as the limit for extra quality of VOO (Table 1). Peroxide values of native area oils were lower than in oils obtained from the Bornova area. In a previous study, acidity and peroxide values of olive oils obtained from 12 olive samples grown on islands of Italy and France were measured within the ranges of 0.20−1.34% and 3.28−16.15 mequiv oxygen/kg of oil, respectively.33 Color properties of olive oils influence consumer acceptance. Color values varied depending on the cultivar and growing region. The L*, a*, and b* values of olive oils obtained from olives grown in the Bornova area were higher than those of olives grown in native areas. Representativeness of the Extract. Intensity and Similarity Evaluation. The intensity scores of the aromatic extract obtained from dichloromethane and pentane/dichloromethane solvents on smelling strips were found to be 63.5 and 64.2 mm on a 100 mm unstructured scale, respectively. The intensity scores of both extracts were close to each other. The differences of intensity scores for the two tests were not found as statistically significant. With regard to similarity evaluation, this score of dichloromethane extract (75.7 mm on a 100 mm unstructured scale) was found to be better than that of pentane/dichloromethane (50.2 mm on a 100 mm unstructured scale). The similarity scores of two solvents were found to be significantly different (p < 0.05). The similarity score of the dichloromethane extract was found to be acceptable and quite high and also scored relatively more closely to the reference sample. On the basis of these results, we selected the dichloromethane solvent for extracting olive oil aroma 393
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Table 2. Volatile Compounds of Olive Oil Samples concentrationa (μg/kg) no.
LRIb
aroma compound
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
1074 1077 1104 1108 1121 1135 1140 1157 1177 1180 1182 1186 1190 1197 1248 1254 1285 1292 1295 1307 1309 1311 1312 1313 1325 1328 1332 1350 1373 1378 1384 1386 1388 1413 1434 1485 1492 1498 1505 1508 1527 1530 1536 1537 1554 1590 1616 1665 1666 1685 1692 1711 1719 1745 1763 1792 1840 1849 1861
hexanal (E)-hexyl-2-butenoate 2-methyl-1-propanol β-pinene (E)-2-pentenal (Z)-3-hexenal 2-ethyl-(E)-2-butenal 1-penten-3-ol 3-penten-2-ol heptanal 3-methyl-2-butenal limonene (E)-2-hexenal 3-hdroxy-3-methyl-2-butanone pentanol 2-ethenyl-2-butenal hexyl acetate octanal 4-heptanol 2-hexanol ethyl chloroacetate (E)-2-heptenal (Z)-3-hexenyl acetate (Z)-4-heptenal 3-methyl-2-butenol (Z)-2-pentenol 6-methyl-5-hepten-2-one hexanol (E)-3-hexenol (Z)-3-hexenol nonanal 3-octanol (E)-2-hexenol acetic acid (E)-2-octenal (E,E)-2,4-heptadienal 1-octen-3-ol heptanol α-copaene 2-ethyl-1-hexanol benzaldehyde (E)-2-nonenol (E)-2-nonenal linalool octanol (E)-2-octenol (E)-2-decenal nonanol (E)-β-farnesene α-terpinolene α-muurolene 3-dodecenal (E,E)-α-farnesene α-farnesene (E,E)-2,4-decadienal ethyl salicylate hexanoic acid guaiacol benzyl alcohol
NY
NY-B
3163.3a 46.7b 243.2a 0.0 70.2b 372.2a 391.2a 708.8a 655.5b 0.0 307.7a 707.7a 9772.9a 63.3b 0.0 0.0 0.0 135.6c 22.9b 923.5a 0.0 0.0 0.0 269.5a 115.6b 997.9a 87.3a 1971.6a 521.6a 681.7c 163.8b 47.9b 2658.1a 20.7b 211.7a 97.5c 59.8b 44.0c 663.6a 70.9b 49.0c 0.0 118.3a 103.7a 91.6b 25.2a 223.7b 55.8b 32.5a 0.0 0.0 0.0 985.1a 0.0 331.4b 0.0 62.7b 110.4a 91.5a
1485.7d 42.1b 59.0b 0.0 37.8c 244.0c 0.0 188.1b 487.2d 48.3c 0.0 179.6c 5754.0c 86.8a 127.0b 663.1a 211.6a 128.0c 0.0 231.0c 52.2a 0.0 846.3a 0.0 0.0 267.7c 35.1c 849.6c 0.0 779.5c 244.5a 0.0 1158.6c 0.0 0.0 73.0c 0.0 60.0b 0.0 81.1b 55.6c 77.2a 0.0 0.0 50.3c 0.0 202.2c 33.9c 0.0 87.6a 101.9b 69.0b 0.0 142.9c 152.1d 15.1a 27.8d 71.2b 78.0b
394
KY 2124.9b 192.0a 223.0a 140.7a 108.2a 143.6d 141.4b 140.8c 803.6a 241.0a 133.0b 582.7b 9247.0b 0.0 199.8a 406.6b 0.0 360.8a 53.6a 439.3b 0.0 96.4b 0.0 42.2b 0.0 564.8b 88.5a 1877.4a 0.0 1426.1a 237.4a 103.2a 2410.3b 50.7a 150.4b 302.3a 70.2a 151.9a 0.0 475.0a 159.6b 22.0b 94.9b 99.4a 233.8a 15.8b 737.8a 74.5a 0.0 18.5c 301.7a 282.0a 20.1b 2104.7a 365.4a 0.0 93.8a 0.0 0.0
KY-B
identificationc
1966.6c 0.0 68.8b 73.4b 75.1b 317.5b 0.0 108.6d 544.1c 170.9b 0.0 93.0d 4995.2d 83.8a 0.0 0.0 142.2b 204.2b 0.0 431.8b 45.9b 319.2a 160.1b 0.0 145.2a 0.0 68.5b 1346.6b 0.0 1066.2b 224.6a 0.0 657.7d 23.6b 104.4c 222.8b 0.0 25.1d 451.1b 0.0 215.3a 80.8a 0.0 54.4b 18.4d 0.0 163.4d 0.0 0.0 37.9b 75.0c 0.0 0.0 514.5b 203.2c 22.5a 48.9c 0.0 0.0
LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI,
MS, std MS tent. MS, std MS, std MS, std MS, std MS tent. MS, std MS, std MS, std MS, std MS, std MS, std MS, tent MS, std MS tent. MS, std MS, std MS, std MS, std MS tent. MS, std MS, std MS, std MS tent. MS, std MS, std MS, std MS, std MS, std MS, std MS tent. MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS tent. MS tent. MS tent. MS tent. MS tent. MS, std MS, std MS, std MS, std MS, std MS, std
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Table 2. continued concentrationa (μg/kg) no.
LRIb
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
1870 1881 1995 2042 2049 2091 2103 2157 2192 2216 2220 2233 2270 2293 2384 2391
aroma compound 2-phenylethanol δ-octalactone phenol γ-nonalactone farnesol octanoic acid γ-decalactone nonanoic acid 3-ethylphenol δ-decalactone δ-dodecalactone methyl palmitate ethyl palmitate decanoic acid γ-dodecalactone dodecanoic acid general total
NY
NY-B
546.9a 90.6a 20.8b 127.4a 289.2a 52.8b 240.9a 212.2b 74.3b 150.0a 85.6a 93.9b 96.3b 95.3c 255.2a 110.2a
208.9c 0.0 0.0 62.5c 119.9c 42.5b 0.0 137.9c 0.0 0.0 0.0 0.0 0.0 74.8d 0.0 52.9b
31090a
16285b
KY 48.5d 42.8b 0.0 0.0 0.0 77.3a 49.6b 328.6a 45.2 82.6b 0.0 421.2a 168.9a 185.6a 33.7b 0.0 29837a
KY-B 258.0b 0.0 43.3a 87.1b 143.7b 95.7a 54.5b 199.0b 229.2a 0.0 0.0 0.0 0.0 120.6b 0.0 0.0
identificationc LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI, LRI,
MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS, std MS tent. MS, std MS, std MS, std MS, std MS, std MS, std MS, std
16802b
Results are the means of three repetitions as μg/kg. Values with different letters in the same row are significant statistically (p < 0.05) bLinear retention index calculated on DB-WAX capillary column. cMethods of identification: LRI (linear retention index); MS tent., (tentatively identified by MS); std (chemical standard). When only MS or LRI is available for the identification of a compounds, it must be considered as an attempt of identification. Standard deviation of all aroma compounds was below 10%. a
(factor 1, 54.32%; factor 2, 36.50%). Figure 1A represents the projection of the variables with regard to the single factor (PC1 or PC2) on the factor plane (PC1 × PC2). The application of the PCA algorithm showed two distinct groups (Figure 1B). The first group represents the results of NY and KY oils. The second one was characterized by KY-B and NY-B. The first group was positively associated with PC1 (lactones, alcohols, aldehydes, terpenes, and carboxylic acids), whereas the second group was negatively associated with PC1 (ketones and esters). It seems that each sample results in a different quarter of the PCA space: for example, KY-B has a higher quantity of ketones, NY-B is richer in esters, NY is characterized by a higher amount of lactones, and finally KY is characterized by terpenes and carboxylic acids. As previously stated, growing areas affected the aroma profiles of olive oils by different researchers. Issaoui et al.36 studied the aroma profiles of virgin olive oils from the Tunisian Chemlali and Chétoui cultivars, grown in two different locations. They found significant differences of aroma profiles between the oils from both cultivars when grown in the different locations. In the results of their study, they stated that aroma profiles were influenced by the pedoclimatic conditions. Our data were also in accordance with the results obtained by Tura et al.37 The authors investigated the geographical origin and cultivar effects on the typical volatile profile of extra virgin olive oils. They noted that the growing area and cultivars significantly affected the aroma profiling of oils. Most of the volatile compounds identified in the present study have already been identified in Italian, Spanish, and Greek olive oils.16,18,31 Aldehydes and alcohols quantitatively and qualitatively represent the main group of the volatile fraction in all olive oils compared to other volatile compounds. The main aldehyde compounds in oil samples were (E)-2-hexenal and hexanal (Table 2). The total concentration of these compounds in cv. NY (15678 μg/kg) was greater than those of the KY (15375 μg/kg), KY-B (9182.5 μg/kg), and NY-B (9157.4 μg/kg) oils.
compounds. As previously stated, it is of great importance to assess the representativeness of the aromatic extracts in a matrix with characteristics similar to those of the original product.34 When we compared other studies, the similarity score of the banana extracts was found to be 66.7 mm,21 and that of rainbow trout (Oncorhynchus mykiss) was found to be 51.1 mm.34 Briefly, the similarity and intensity results indicated that the representative aromatic extracts were achieved with the SDE method to determine the aroma-active compounds of olive oils. Similarly, the SDE technique has shown to be a reliable method for the determination of Moroccan black olive odor active compounds by Collin et al.11 and hazelnut oils odorants by Pfnuer et al.35 Volatile Composition of Olive Oils. The volatile compounds identified in olive oils and linear retention index values on the DB-WAX column for these compounds are presented in Table 2. Mean values (μg/kg) of the GC analyses of triplicate extractions are reported. As can be seen in Table 2, a wide variety of volatile compounds was found in the VOO samples. Differences were observed in terms of the number of volatile compounds depending on growing regions. Totals of 61, 48, 59, and 48 compounds, including aldehydes, alcohols, ketones, esters, terpenes, lactones, carboxylic acids, and phenols, were identified and quantified in NY, NY-B, KY, and KY-B olive oils, respectively. The highest amount of volatiles was found in NY oil (31090 μg/kg), followed by KY (29837 μg/kg), KY-B (16802 μg/kg), and NY-B (16285 μg/kg). When the geographical effect was compared, the major difference observed was that the total amount of volatile compounds in the native region oils was approximately 2 times higher than that found in the Bornova region oils (Table 2). To evaluate the possibility of differentiating the oil samples taking into account the aromatic profile, we applied a multivariate statistical analysis using the total concentration of volatile in each of the eight chemical groups from Table 2. The variables were selected for the PCA, and the explained variance was 90.81% 395
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Figure 1. (A) Projection of the variables on the factor plane (PC1 × PC2) considering the aroma compounds quantified. (B) Score plot for the two principal components showing the two geographical origins studied.
Significant difference between the two growing areas was found for aldehydes, which were found in higher concentrations of the native region oils than the Bornova region oils. Kiralan et al.38 also pointed out a significant difference in the composition of volatile components between the oils from the same cultivar and from different geographic regions. (E)-2-Hexenal and hexanal were present at the highest concentration in all oils, which accounted for between 13.82 and 21.42% and between 54.39 and 62.83% of total aldehyde compounds, respectively.
The amounts of (E)-2-hexenal and hexanal in the native region oils reached higher levels than the Bornova region oils. Similarly, Aparicio et al.39 reported that the major aroma compounds identified in most virgin olive oils from European countries are (E)-2-hexenal, hexanal, hexanol, and 3-methylbutan-1-ol. (E)-2-Hexenal is produced by the enzymatic breakdown of the 13-hydroperoxide of linolenic acid in leaf homogenates, whereas hexanal is formed by the action of the enzyme aldehyde-lyase.4,5 These aldehydes are widespread, as 396
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Table 3. Aroma-Active Compounds of Olive Oil Samples (FD ≥ 8) FD factora b
no.
LRI
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
1074 1108 1121 1135 1171 1176 1178 1180 1182 1186 1190 1243 1254 1269 1285 1292 1300 1326 1350 1378 1403 1408 1412 1485 1498 1521 1544 1547 1554 1563 1592 1616 1665 1708 1711 1719 1745 1755 1763 1824 1874 1922 1980 2103 2216 2233 2270 2384
c
aroma compound
odor description
hexanal β-pinene (E)-2-pentenal (Z)-3-hexenal 2-ethyl-(E)-2-butenal 1-penten-3-ol 3-penten-2-ol heptanal 3-methyl-2-butenal limonene (E)-2-hexenal unknown 2-ethenyl-2-butenal unknown hexyl acetate octanal (Z)-3-hexenyl acetate (Z)-2-pentenol hexanol (Z)-3-hexenol unknown (E)-2-octenal (E)-2-hexenol (E,E)-2,4-heptadienal heptanol α-copaene unknown linalool octanol unknown α-terpinolene (E)-2-decenal nonanol (E)-2-nonenol 3-dodecenal (E,E)-α-farnesene α-farnesene unknown (E,E)-2,4-decadienal guaiacol unknown 2-phenylethanol unknown γ-decalactone δ-decalactone methyl palmitate ethyl palmitate γ-dodecalactone
cut grass green plant, floral green plant, grassy vegetables, grassy, herbal grassy, floral grassy, green plant green, grassy green plant, oily floral, sweet light floral cut grass, green citrus fruity, green pickled olive fruity citrus, lemon fruity, green fatty floral, grass cut grass, herbal floral green plant, fatty grassy, green, fruity fatty margarine, fatty sweet, fruity sweet, floral lilac, lavender floral, grassy sweet, fruity floral, light soapy, fatty citrus fruity, waxy fatty floral, herb floral, green plant cocoa powder, chocolate fatty, solvent olive paste, soapy green plant floral, rose floral fruity fruity fruity fruity fruity
NY
NY-B
KY
KY-B
128
64
64 32 16 64
64 16 8 128
128 128 128
64 16 64
8 128 256 1024 64
128 16
16 128 512
256
64
512
64 512 512 16
32 16
256 128
128 128 128 16 64
64 32 128
32 128 16 32
64 64 256
64 32 16
256 16 16 16 64 32 64 256
16 16 32
16 128 16 16 16
32 16
16 8
8 32
32 16
32 64 8 32 128 8 32 64 8 32 512
8 16 128 32
8
16 16 8 32 32
8 8 8 128 256
8
a
FD factor is the highest dilution of the extract at which an odorant is determined by aroma extract dilution analysis. bLinear retention index calculated on DB-WAX capillary column. cOdor description as perceived by panelists during olfactometry.
nonanal was found to be an appropriate way to detect the beginning of oxidation and its evolution,19 and Morales et al.43 proposed that, when the ratio of hexanal to nonanal was below 2, the oil should be considered as rancid. The ratios of these two aldehydes in NY, NY-B, KY, and KY-B olive oils were found to be 19.3, 6.1, 9.0, and 8.8, respectively. Alcohols are produced by the action of ADH enzyme. It is widespread in the plant kingdom and is responsible for the
they have already been found in many other olive oils, such as Spanish cvs. Arbequina, Cornicabra, Morisca, Picolimon, Picudo, and Picual40 and Tunisian cvs. Jemri Ben Gerdane, Chemlali Zarzis, and Zalmati.41 Vichi et al.42 studied the effect of different growing regions on the aroma profile of same cultivars. The authors noted that there were significant differences in the mean values of hexanal of the oils in relation to different regions. Among the aldehydes, the ratio hexanal/ 397
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
(citrus-like), (E)-2-nonenal (paper-like, fatty), and (E,E)-2,4decadienal (fatty, solvent) were potent aldehyde compounds in olive oils. Similarly, hexanal (fresh cut grass), (E)-2-hexenal (green), and heptanal (fatty, soapy) have been detected as aroma-active aldehyde compounds in Portuguese extra virgin olive oils from Galega Vulgar and Cobrançosa cultivars by Peres et al.47 Alcohols were the second largest class of odor-active compounds in the oil samples. A total of 11 aroma-active alcohol compounds were found in the samples. These were 1penten-3-ol (grassy, green plant), 3-penten-2-ol (green, grassy), (Z)-2-pentenol (fatty), hexanol (floral, grassy), (Z)-3-hexenol (cut grass, herbal), (E)-2-hexenol (grassy, green plant, fruity), heptanol (margarine, fatty), octanol (floral, grassy), nonanol (citrus), (E)-2-nonenol (fruity, waxy), and 2-phenylethanol (rose). Within these, 1-penten-3-ol, 3-penten-2-ol, hexanol, (Z)-3-hexenol, and (E)-2-hexenol were common aroma-active alcohol compounds in all oil samples. The FD factors of these compounds varied depending on cultivar. Alcohols have less sensorily significant than aldehydes due to their higher odor threshold values. The most powerful aroma-active alcohol compound was 1-penten-3-ol to contribute to the aroma profile of the olive oils. The FD factor of these compounds was 1024 for NY. This compound was generated from polyunsaturated fatty acids (PUFA) by the influence of an enzyme such as lipoxygenase or hydroperoxidase.48 (Z)-3-Hexenol (FD = 256 for KY and KY-B) was the second most important aroma-active alcohol compound in the oils. Most of the potent alcohols found in our samples have already been identified in olive oils by Luna et al.8 and Dierkes et al.46 and in green olives by Iraqi et al.25 Among the alcohols, 2-phenylethanol was detected in NY, NY-B, and KY-B oils with FD factors of 128, 32, and 32, respectively, and described as a floral-rose odor. As stated previously, 2-phenylethanol has been cited as a major aromaactive compound in Moroccan black olives by Collin et al.11 and in olive oils by Dierkes et al.,46 providing a floral-fruity and sweet-winey odor, respectively. β-Pinene (green plant, floral), limonene (light floral), αcopaene (sweet, fruity), linalool (lilac, lavender), α-terpinolene (floral, light), (E,E)-α-farnesene (floral, herb), and α-farnesene (floral, green plant) were detected as aroma-active terpene compounds in the olive oil extracts. The samples NY, KY, and KY-B contained four aroma-active terpene compounds, but NY-B only one. On the basis of GC-O, the most potent aromaactive terpenes were limonene and linalool, which have 128 FD factors. β-Pinene and α-farnesene were detected only in KY and KY-B oils. In addition, limonene was not sensed in oil obtained from the Bornova region. Among the terpenes, limonene, linalool, and α-farnesene were identified as being the key odorants in Moroccan green olives.25 Hexyl acetate (fruity, green), (Z)-3-hexenyl acetate (fruity, green), methyl palmitate (fruity), and ethyl palmitate (fruity) were detected in oil aromatic extracts as aroma-active compounds. Esters are the main volatile compounds found in most fruits and are responsible for the fruity notes.20 Increasing the alcohol acetyl transferases (AAT) activity can enhance the production of volatile esters in olive oils.42 Within the esters, ethyl palmitate had the highest FD factor (256) value detected by GC-O in KY oil. Lactones contribute to the characteristic fruity odors of the olive oil samples. γ-Decalactone (fruity), δ-decalactone (fruity), and γ-dodecalactone (fruity) were identified as aroma-active lactone compounds. As summarized in Table 3, oil produced
formation of volatile alcohols that contribute to the aroma of olive oils. The native region olive oils contained more alcohols than the Bornova region olive oils. The total amount of alcohols in the native region oils was approximately 2 times higher than the that of the Bornova region oils. The highest amount of alcohols was observed in NY oil (10534 μg/kg), probably related to higher activity of the alcohol dehydrogenase (ADH) enzyme, followed by KY (9333.6 μg/kg), KY-B (4751.4 μg/kg), and NY-B (4737.2 μg/kg) oils. Within the alcohols, C6 alcohols such as hexanol, (E)-2-hexenol, and (Z)-3-hexenol were observed as high amounts in oil samples. When compared to other studies, a similar pattern of the three mentioned C-6 alcohols was identified in virgin olive oils from Spain,40 Italy,42 and Tunisia.44 (E)-2-Hexenol predominated in NY (2658.1 μg/ kg) and KY (2410.3 μg/kg) oil samples. Similarly, hexanol was observed in highest amount in the native region (NY = 1971.6 μg/kg and KY = 1877.4 μg/kg) oil samples. In previous studies, higher amounts of C-6 alcohols were determined in Tunisian and Italian olive oils, and the amount of hexanol and (E)-2hexenol ranged between 50 and 2970 μg/kg and between 410 and 6920 μg/kg, respectively.44 GC-MS-O Results. The results of olfactometric analysis are given in Table 3. Application of the AEDA on the olive oil extracts revealed 29, 20, 30, and 27 aroma-active compounds in the NY, NY-B, KY, and KY-B oils, respectively. The differences in the number of aroma-active compounds of the oils are mainly caused by concentration differences of these compounds. As can be seen in Table 3, the native region oils contained more aroma-active compounds than the Bornova region oils. The FD factors of the compounds were detected to fall within the range of ≥8−1024 (Table 3). The compound having the highest FD factor (1024) was 1-penten-3-ol for NY oil, 2-ethenyl-2-butenal (FD = 512) for NY-B oil, (E)-2-hexenal and 2-ethenyl-2-butenal (FD = 512) for KY oil, and (E)-2hexenal and (Z)-3-hexenol for KY-B oil. Aldehydes were the largest class of aroma-active components of the olive oil extracts. Odor threshold values of these compounds are generally lower than those of volatile compounds; thus, they have important potential effect of the overall aroma of olive oils. A total of 14 aldehydes were identified in olive oils as key compounds (Table 3). Aldehydes are generally characterized by intense sensory description by panelists and are associated with green, cut grass, green plant, citrusy, fatty, and sweet notes. The total numbers of aromaactive aldehydes were 9, 5, 11, and 8 in NY, NY-B, KY, and KYB oils, respectively. Among these compounds, hexanal, (Z)-3hexenal, (E)-2-hexenal, and (E,E)-2,4-decadienal were common to all varieties. These aldehydes are widespread, as they have already been found in many other olive oils.9,10 Within the aroma-active aldehydes, (E)-2-hexenal (FD = 512) for NY and KY oils and 2-ethenyl-2-butenal (FD = 512) for NY-B and KY oils were the most powerful aroma-active compounds to contribute to the aroma profile of oils. The first has a cut grass note, and the latter has a fruity, green note. The FD factor of (E)-2-hexenal was 256 for NY-B and KY-B. Solinas et al.45 suggest that monovarietal virgin olive oils could be distinguished by the (E)-2-hexenal compound. This aldehyde is a well-known powerful odorant that contributes to the characteristic aromas of olive oils.9 The other most important aldehydes, which have FD factors of 256, were 2-ethyl-(E)-2butenal (grassy, floral) and octanal (citrus, lemon) for NY and KY, respectively. In a previous study, Dierkes et al.46 stated that hexanal (green), (E)-2-hexenal (green, apple-like), nonanal 398
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
Figure 2. Sensory properties of olive oils.
previous study, sensory properties of olive oils were evaluated by using the medians of defects and positive attributes.50 In conclusion, this first study revealed key aroma compounds of olive oils obtained from NY and KY cultivars and the effect of growing regions on these compounds. The results suggest that besides cultivars, growing area influences volatile formation in olive oil. The native region oils were distinguished from the Bornova region samples by their number and quantity of volatile compounds. Totals of 61, 48, 59, and 48 aroma compounds were identified and quantified in olive oils obtained from NY, NY-B, KY, and KY-B cultivars, respectively. Aldehydes were found as, qualitatively and quantitatively, the majority of aroma-active compounds in the oil samples. On the basis of sensory analysis, the olive oils studied in the present work were defined as extra virgin olive oils, and the highest fruity median was found in NY oil followed by KY, NY-B, and KY-B oils, respectively.
from NY olives had a higher number and FD factors than the others. On the basis of the FD factor values, the most powerful aroma-active lactone compound was γ-dodecalactone in NY oil, qualified as having a fruity odor. The FD factor of this compound was 512. All of the potent lactones found in our samples have already been identified in black olives by Collin et al.11 The other aroma-active compound identified in the olive oils was guaiacol (olive paste, soapy). This compound was detected only in NY and NY-B oils. The FD factors of this volatile phenol were 32 and 16, respectively. This compound was also determined in Italian, Spanish, and Moroccan olive oils as the aroma-active compound. The highest FD value (128) of this compound was determined in Moroccan olive oils.10 A total of eight unknown compounds may contribute to the overall aroma of the olive oil samples. Among the unknowns, 14 (LRI = 1269) was found to be the pickled olive odor in KY and KY-B oils. Unknown 27 (LRI = 1544), with a sweet, floral note, was detected in only NY and NY-B oils. Unknowns 30 and 41 were found in oils obtained from the Bornova region, whereas unknown 43 was found in oils obtained from the native region. Sensory Analysis of Olive Oils. The results of the positive and negative sensory attributes of the olive oils are given in Figure 2. Olive oils are classified as extra virgin olive oils if the median of defects is equal to zero and the median of fruity is more than zero.27 The studied oils showed no defects in the sensory analysis, so they were considered extra virgin olive oils. The highest fruity median was found in NY oil (6.50). This value was quoted at 6.40, 6.10, and 6.0 for KY, NY-B, and KY-B oils, respectively. The bitterness values of olive oils were detected between 3.90 and 4.60. According to median value of positive attributes (fruity, bitterness, and pungent), oils from the native region had the higher values than those from the Bornova region. The highest value (5.70) of pungent property was detected in olive oil obtained from cv. KY. According to Regulation EEC/640/2008,49 the term “medium” may be used when the median of the positive attribute concerned is between 3 and 6. In our research olive oil samples were characterized as medium for all criteria (fruity, bitter, and pungent). In a
■
AUTHOR INFORMATION
Corresponding Author
*(S.S.) E-mail:
[email protected]. Phone: + (90)-322-3386173. Fax: + (90) 322 338 66 14. Funding
We thank the Scientific and Technical Research Council of Turkey (TUBITAK, Project 110O602) and Cukurova University Research Fund (ZF-2010-D24) for financial support. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We thank all of the members of our sensory panel, who devoted much of their precious time to the sensory assessment of olive oil extracts.
■
REFERENCES
(1) De la Torre, F.; Bautista, R.; Canovas, F. M.; Claros, G. Isolation of DNA from olive oil and oil sediments: application in oil fingerprinting. J. Food Agric. Environ. 2004, 2, 84−89. (2) FAO. Agricultural Statistical Database, 2013 (http://www.faostat. fao.org).
399
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
(3) Grosch, W. Handbook of Food Analysis: Physical Characterization and Nutrient Analysis − Aroma Compounds; Nollet, L. M. L., Ed.; CRC Press: Boca Raton, FL, USA, 2004; p 717. (4) Kiritsakis, A. K. Flavor components of olive oil − a review. J. Am. Oil Chem. Soc. 1998, 75, 673−681. (5) Ridolfi, M.; Terenziani, S.; Patumi, M.; Fontanazza, G. Characterization of the lipoxygenases in some olive cultivars and determination of their role in volatile compounds formation. J. Agric. Food Chem. 2002, 50, 835−839. (6) Kalua, C. M.; Allen, M. S.; Bedgood, D. R.; Bishop, A. G.; Prenzler, P. D.; Robards, K. Olive oil volatile compounds, flavor development and quality: a critical review. Food Chem. 2007, 100, 273−286. (7) Morales, M. T.; Aparicio, R. Effect of extraction conditions on sensory quality of virgin olive oil. J. Agric. Food Chem. 1999, 76, 295− 300. (8) Luna, G.; Morales, M. T.; Aparicio, R. Characterisation of 39 varietal virgin olive oils by their volatile compositions. Food Chem. 2006, 98, 243−252. (9) Guth, G.; Grosch, W. A comparative-study of the potent odorants of different virgin olive oils. Fat Sci. Technol. 1991, 93, 335−339. (10) Reiners, J.; Grosch, W. Odorants of virgin olive oils with different flavor profiles. J. Agric. Food Chem. 1998, 46, 2754−2763. (11) Collin, S.; Nizet, S.; Muls, S.; Iraqi, R.; Bouseta, A. Characterization of odor-active compounds in extracts obtained by simultaneous extraction/distillation from Moroccan black olives. J. Agric. Food Chem. 2008, 56, 3273−3278. (12) Gomez-Rico, A.; Salvador, M. D.; La Greca, M.; Fregapane, G. Phenolic and volatile compounds of extra virgin olive oil (Olea europaea L. cv. Cornicabra) with regards to fruit ripening and irrigation management. J. Agric. Food Chem. 2006, 54, 7130−7136. (13) Rodriguez, P. R.; Barreiro, C. G.; Grande, B. C.; Gandara, J. S. Dynamic headspace/GC-MS to control the aroma fingerprint of extravirgin olive oil from the same and different olive varieties. Food Control. 2012, 25, 684−695. (14) Koprivnjak, O.; Majetic, V.; Bubola, K. B.; Kosic, U. Variability of phenolic and volatile compounds in virgin olive oil from Leccino and Istarska Bjelica cultivars in relation to their fruit mixtures. Food Technol. Biotechnol. 2012, 50, 216−221. (15) IOOC, International Olive Council. Guide for the Determination of the Characteristics of Oil-Olives, COI/OH/Doc. No 1, Nov 2011 (http://www.internationaloliveoil.org). (16) Ranalli, A.; Contento, S.; Schiavone, C.; Simone, N. Malaxing temperature affects volatile and phenol composition as well as other analytical features of virgin olive oil. Eur. J. Lipid Sci. Technol. 2001, 103, 228−238. (17) EEC, European Economic Commission. Regulation EEC/1429/ 92 Amending Regulation EEC2568/91 on the characteristics of olive oil and olive residue oil and on the relevant methods of analysis. Off. J. Eur. Communities 1992, 150, 17. (18) Vichi, S.; Guadayol, J. M.; Caixach, J.; Tamames, E. L.; Buxaderas, S. Comparative study of different extraction techniques for the analysis of virgin olive oil aroma. Food Chem. 2007, 105, 1171− 1178. (19) Kesen, S.; Kelebek, H.; Selli, S. Characterization of the volatile, phenolic and antioxidant properties of monovarietal olive oil obtained from cv. Halhali. J. Am. Oil Chem. Soc. 2013, 90, 1685−1696. (20) Selli, S.; Kelebek, H. Aromatic profile and odor-activity value of blood orange juices obtained from Moro and Sanguinello (Citrus sinensis L. Osbeck). Ind. Crop Prod. 2011, 33, 727−733. (21) Selli, S.; Gubbuk, H.; Kafkas, E.; Gunes, E. Comparison of aroma compounds in Dwarf Cavendish banana (Musa spp. AAA) grown from open-field and protected cultivation area. Sci. Hortic. 2012, 141, 76−82. (22) Schieberle, P.; Grosch, W. Evaluation of the flavor of wheat and rye crust by aroma extract dilution analysis. Z. Lebensm. Unters. Forsch. 1987, 185, 111−113.
(23) Fickert, B.; Schieberle, P. Identification of the key odorants in barly malt (caramalt) using GC/MS tecniques and odor dilution analyses. Nahrung 1998, 42, 371−375. (24) Tairu, A. O.; Hofmann, T.; Schieberle, P. Studies on tke key odorants formed by roasting of wild mango seeds (Irvingia gabonensis). J. Agric. Food Chem. 2000, 48, 2391−2394. (25) Iraqi, R.; Vermeulen, C.; Benzekri, A.; Bouseta, A.; Collin, S. Screening for key odorants in Moroccan green olives by gas chromatography-olfactometry/aroma extract dilution analysis. J. Agric. Food Chem. 2005, 53, 1179−1184. (26) Averbeck, M.; Schieberle, P. Characterization of the key aroma compounds in a freshly reconstituted orange juice from concentrate. Eur. J. Lipid Sci. Technol. 2009, 229, 611−622. (27) IOOC, International Olive Council. Sensory Analysis of Olive Oil Method for the Organoleptic Assessment of Virgin Olive Oil, COI/T.20/Doc. No 15/Rev. 4, 2011 (http://www. internationaloliveoil.org). (28) Benincasa, C.; De Nino, A.; Lombardo, N.; Perri, E.; Giovanni, S.; Tagarelli, A. Assay of aroma active components of virgin olive oils from southern Italian regions by SPME-GC/ion trap mass spectrometry. J. Agric. Food Chem. 2003, 51, 733−741. (29) Ferguson, L.; Sibbett, G. S. Olive Production Manual, 2nd ed.; 180s; University of California, Agriculture & Natural Resources Publishing: Los Angeles, CA, USA, 2004. (30) Ogutcu, M.; Yilmaz, E. Comparison of the virgin olive oils produced in different regions of Turkey. J. Sensory Stud. 2009, 24, 332−353. (31) Longobardi, F.; Ventrella, A.; Casiello, G.; Sacco, D.; TasioulaMargari, M.; Kiritsakis, A. K. Characterization of the geographical origin of western Greek virgin olive oils based on instrumental and multivariate statistical analysis. Food Chem. 2012, 133, 169−175. (32) Di Giovacchino, L.; Solinas, M.; Miccoli, M. Effect of extraction systems on the quality of virgin olive oil. J. Am. Oil Chem. Soc. 1994, 71, 1189−1194. (33) Cerretani, L.; Bendini, A.; Del Caro, A.; Piga, A.; Vacca, V.; Caboni, M. F.; Toschi, T. G. Preliminary characterisation of virgin olive oils obtained from different cultivars in Sardinia. Eur. Food Res. Technol. 2006, 222, 354−361. (34) Selli, S.; Rannou, C.; Prost, C.; Robin, J.; Serot, T. Characterization of aroma-active compounds in rainbow trout (Oncorhynchus mykiss) eliciting an off-odor. J. Agric. Food Chem. 2006, 54, 9496−9502. (35) Pfnuer, P.; Matsui, T.; Grosch, W.; Guth, H.; Hofmann, T.; Schieberle, P. Development of a stable isotope dilution assay for the quantification of 5-methyl-(E)-2-hepten-4-one: application to hazelnut oils and hazelnuts. J. Agric. Food Chem. 1999, 47, 2044−2047. (36) Issaoui, M.; Flamini, G.; Brahmi, F.; Dabbou, S.; Hassine, K. B.; Taamali, A.; Chehab, H.; Ellouz, M.; Zarrouk, M.; Hammami, M. Effect of the growing area conditions on differentiation between Chemlali and Chétoui olive oils. Food Chem. 2010, 119, 220−225. (37) Tura, D.; Failla, O.; Bassi, D.; Attilio, C.; Serraiocco, A. Regional and cultivar comparison of Italian single cultivar olive oils according to flavor profiling. Eur. J. Lipid Sci. Technol. 2013, 115, 196−210. (38) Kiralan, M.; Ozkan, G.; Koyluoglu, F.; Ugurlu, H. A.; Bayrak, A.; Kiritsakis, A. Effect of cultivation area and climatic conditions on volatiles of virgin olive oil. Eur. J. Lipid Sci. Technol. 2012, 114, 552− 557. (39) Aparicio, R.; Morales, M. T.; Alonso, V. Authentication of European extra-virgin olive oils by their chemical compounds, sensory attributes and consumers’ attitudes. J. Agric. Food Chem. 1997, 45, 1076−1083. (40) Gomez-Rico, A.; Fregapane, G.; Salvador, M. D. Effect of cultivar and ripening on minor components in Spanish olive fruits and their corresponding virgin olive oils. Food Res. Int. 2008, 41, 433−440. (41) Zarrouk, W.; Haddada, F. M.; Baccouri, B.; Oueslati, I.; Taamalli, X. F.; Lizzani-Cuvelier, L.; Daoud, D.; Zarrouk, M. Characterization of virgin olive oil from southern Tunisia. Eur. J. Lipid Sci. Technol. 2008, 110, 81−88. 400
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
Journal of Agricultural and Food Chemistry
Article
(42) Vichi, S.; Pizzale, L.; Conte, L.; Buxaderas, S.; Tamames, E. L. Solid phase microextraction in the analysis of virgin olive oil volatile fraction: characterization of virgin olive oils from two distinct geographical areas of northern Italy. J. Agric. Food Chem. 2003, 51, 6572−6577. (43) Morales, M. T.; Rios, J. J.; Aparicio, R. Changes in the volatile composition of virgin olive oil during oxidation: flavors and off-flavors. J. Agric. Food Chem. 1997, 45, 2666−2673. (44) Baccouri, O.; Bendini, A.; Cerretani, L.; Guerfel, M.; Baccouri, B.; Lercker, G.; Zarrouk, M.; Ben Miled, D. D. Comparative study on volatile compounds from Tunisian and Sicilian monovarietal virgin olive oils. Food Chem. 2008, 111, 322−328. (45) Solinas, M. F.; Angerosa, F.; Marsili, V. Research of some flavor components of virgin olive oil in relation to olive varieties. Riv. Ital. Sostanze Grasse 1988, 65, 361−368. (46) Dierkes, G.; Bongartz, A.; Guth, H.; Hayen, H. Quality evaluation of olive oil by statistical analysis of multicomponent stable isotope dilution assay data of aroma active compounds. J. Agric. Food Chem. 2012, 60, 394−401. (47) Peres, F.; Jelen, H. H.; Majcher, M. M.; Arraias, M.; Martins, L. L.; Ferreira-Dias, S. Characterization of aroma compounds in Portuguese extra virgin olive oils from Galega Vulgar and Cobrançosa cultivars using GC-O and GC X GC −ToFMS. Food Res. Int. 2013, 54, 1979−1986. (48) Ganeko, N.; Shoda, M.; Hirohara, I.; Bhadra, A.; Ishida, T.; Matsuda, H.; Takamura, H.; Matoba, T. Analysis of volatile flavor compounds of sardine (Sardinops melanostica) by solid phase microextraction. J. Food Sci. 2008, 73, 83−88. (49) EEC, Commission Regulation (EC) No. 640. Off. J. Eur. Union 2008, 178, 11−16. (50) Escuderos, M. E.; Uceda, M.; Sanchez, S.; Jimenez, A. Instrumental technique evolution for olive oil sensory analysis. Eur. J. Lipid Sci. Technol. 2007, 109, 536−546.
401
dx.doi.org/10.1021/jf4045167 | J. Agric. Food Chem. 2014, 62, 391−401
