SHORT COMMUNICATION HS–GC–MS–O analysis and sensory acceptance of passion fruit during maturation Natália S. Janzantti1 • Magali Monteiro2 Revised: 17 April 2017 / Accepted: 26 April 2017 / Published online: 8 May 2017 � Association of Food Scientists & Technologists (India) 2017 Abstract The odor-active compounds of the conventional yellow passion fruit influence the aroma during ripeness and the acceptance of the juice. HS–GC–MS and GC– OSME analysis and sensory acceptance of the conventional passion fruit from different stages of ripeness were studied to characterize the aroma of the fruit and, aroma and flavor of the juice. Ethyl butanoate, ethyl hexanoate and propyl acetate showed high odoriferous importance in the passion fruit from the 1/3 yellow skin color. Cis-3-hexen-1-ol and diethyl carbonate plus the odor-active compounds from the 1/3 yellow skin color showed high odoriferous importance in the 2/3 yellow skin color, and butyl acetate and alpha- terpineol plus the same odor-active compounds from 2/3 were the most important for the 3/3 yellow skin color. There was difference in the aroma and flavor of the juices, with higher acceptance means for the passion fruit from the 3/3 yellow skin color. The passion fruit volatile compounds peak area, odoriferous intensity and sensory acceptance of the juices increased during ripeness, indicating that the conventional passion fruit characteristic aroma is com- pletely expressed when the fruit reaches the whole matu- ration, at the 3/3 yellow skin color. Keywords HS–GC–MS–OSME � Passion fruit (Passiflora edulis Sims f. flavicarpa Deg.) � Stage of ripeness � Olfactometry � Sensory acceptance Introduction Brazil is the greatest producer and consumer of passion fruit in the world. 838 thousand ton of passion fruit were produced in 2013, mostly Passiflora edulis Sims f. flavi- carpa Deg., which accounts for 95% of the cultivation area (IBGE 2014). Yellow passion fruit is used as in natura fruit, mainly to produce industrialized juice. Yellow passion fruit should be harvested when still connected to the plant to improve shelf life and commercial value. In Brazil, passion fruit is regularly harvested after falling to the ground favoring microbial contamination and dehydration. The skin color is a simple criterion to identify the ideal point of harvest, making it possible to obtain ripe, more uniform fruits, with better nutritional and commercial quality, and phytosanitary conditions. Although the entire yellow skin color of ripe fruits, fruits with 2/3 yellow skin color that are not completely ripe may show physico- chemical and sensorial characteristics close to the ripe fruits (De Marchi et al. 2000; Amaro and Monteiro 2001; Macoris et al. 2012). The passion fruit aroma is described as predominantly fruity, floral and slightly sulfurous (Winterhalter 1991; Pino 1997). The esters ethyl butanoate, hexyl butanoate, ethyl hexanoate and hexyl hexanoate are the major con- tributing to the characteristic fruity and sweety aroma of passion fruit, while the terpenes limonene, myrcene, trans- ocimene, terpinolene, linalool and alpha-terpineol con- tribute to the floral and fruity aroma, the aldehydes hex- anal, octanal and benzaldehyde contribute to the green and & Magali Monteiro monteiro@fcfar.unesp.br 1 Department of Food Engineering and Technology, São Paulo State University - UNESP, São José Do Rio Preto, SP 15054-000, Brazil 2 Department of Food and Nutrition, School of Pharmaceutical Science, São Paulo State University - UNESP, Araraquara, SP 14801-902, Brazil 123 J Food Sci Technol (July 2017) 54(8):2594–2601 DOI 10.1007/s13197-017-2671-z http://crossmark.crossref.org/dialog/?doi=10.1007/s13197-017-2671-z&domain=pdf http://crossmark.crossref.org/dialog/?doi=10.1007/s13197-017-2671-z&domain=pdf citrus aroma, and the alcohol hexanol is important for green aroma of passion fruit (Pino 1997; Werkhoff et al. 1998). There is little information about the aroma of pas- sion fruit at different stages of ripeness (Casimir et al. 1981; Janzantti and Monteiro 2014). The volatile compo- sition of yellow passion fruit depends on the cultivation system, cultivar and edaphoclimatic conditions, among others factors. The stage of ripeness could influence the aroma profile of passion fruit and consequently its accep- tance. Organic yellow passion fruit (Feltrin selection) showed the same odoriferous profile at the 2/3 and 3/3 yellow skin color, pointing out its consumption in earlier stages of ripeness (Janzantti and Monteiro 2014). The aim of this work was to study the odor-active compounds of the conventional yellow passion fruit from different stages of ripeness corresponding to the skin color to identify which of them characterize the aroma during ripeness and influence acceptance of aroma and flavor of the juice. Materials and methods Chemicals The volatile compounds standards (purity [99.0%) were from Sigma-Aldrich (St. Louis, USA) and Fluka (Stein- heim, Germany), all listed in Table 1. Dichloromethane was from J. T. Baker (Philipsburg, USA) and n-alkanes (C8–C30) from Supelco (Bellefonte, USA), both GC grade. NaCl analytical grade was from Merck (Darmstadt, Germany). Yellow passion fruit samples Conventional yellow passion fruits (Passiflora edulis Sims f. flavicarpa Deg.) from Feltrin selection were cultivated in Sumaré, SP, Brazil (22�4901900 S and 47�1600100 W, 625 m altitude). Fruits were harvested in the stages of ripeness corresponding to the 1/3, 2/3 and 3/3 yellow skin color (De Marchi et al. 2000; Amaro and Monteiro 2001). Passion fruits (15 kg) of each stage of ripeness were harvested at random, choosing fruits of similar size. After harvesting fruits were selected, classified according to the skin color, and washed. The pulp was separated from the seeds and peel, packed into hermetically closed glass flasks and analyzed. Isolation of volatile compounds from pulp In each stage of ripeness volatile compounds were isolated in triplicate using dynamic headspace (HS) by vacuum suction (79.99 mm Hg) at room temperature (25 �C) and a Porapak Q trap (100 mg, 150–180 lm, Waters Associates, Milford, U.S.A). Passion fruit pulp (300 g) and NaCl (30/ 100 g) were put in the volatiles capture system flask, iso- lated for 2 h and eluted with 300 lL dichloromethane (Macoris et al. 2011). High resolution gas chromatography (HSGC–FID) A 2010 Shimadzu (Kyoto, Japan) gas chromatograph with a flame ionization detector (GC-FID) and a DB-Wax (30 m 9 0.25 mm 9 0.25 lm) column (J&W, Folsom, USA) at 40 �C for 10 min, then 200 �C at 3 �C/min, held for 10 min and hydrogen at 1.3 mL/min were used. Injection (2 lL) was in splitless mode. Injector temperature was 230 �C and detector 250 �C. A pentadecane internal standard solution (10 lL, 4 lL/mL) was added into the isolate and the volatile compounds were quantified. The relative peak area was obtained. Gas chromatography–mass spectrometry (GC–MS) A 5975C Agilent (Wilimington, USA) gas chromatograph– mass spectrometer, with electron impact ionization source (70 eV) in scan mode and mass range of 35–350 m/z, and a DB-Wax column (30 m 9 0.25 mm 9 0.25 lm) at 40 �C for 10 min, then to 200 �C at 3 �C/min, held for 10 min and helium at 1.3 mL/min were used. Injector temperature was 230 �C and detector 240 �C. The DB-5 (J&W, Fol- som, USA) column (60 m 9 0.25 mm 9 0.25 lm) at 50 �C, then to 250 �C at 3 �C/min, held for 10 min, and helium at 1.0 mL/min were also used. Injector and detector temperatures were 250 �C. The volatile compounds were identified according to the criteria reported by Janzantti and Monteiro (2014). A solution of alkanes in dichloromethane was injected in the DB-Wax and DB-5 columns, under the same chromato- graphic conditions, to calculate the retention indices. Gas chromatography–Olfactometry (GC–O) The odor-active compounds were analyzed by OSME (Da Silva et al. 1994) using the SCDTI (time-intensity data collection system) data collection program (Da Silva 1999). The GC–FID was modified for the GC–O–OSME analysis according to Janzantti et al. (2012). Four trained and selected judges, 23–37 years old, were requested to appoint the intensity using the SCDTI scale and describe the odor of the passion fruit pulp from each stage of ripeness in triplicate. Data collected from the SCDTI software for each GC–O analysis session of each judge and each stage of ripeness were used to construct the individual aromagram, considering that peaks should be detected in at least two of three repetitions. Then, a consensual J Food Sci Technol (July 2017) 54(8):2594–2601 2595 123 Table 1 Volatile and odor-active compounds of the conventional passion fruit in the 1/3, 2/3 and 3/3 yellow skin color stages of ripeness Peak RI1 RI2 Compound3 Aroma description GC–FID4 GC–O5 1/3 2/3 3/3 1/3 2/3 3/3 1 922 \800 ethyl propanoatea,b,c Not detected by OSME 0.92 1.73 3.32 – – – 2 932 \800 propyl acetatea,b,c Skin, passion fruit nd–0.20 0.26 6.40 4.62 6.65 7.15 4 939 \800 methyl butanoatea,b,c Passion fruit/fruity 12.47 8.31 2.34 3.40 3.47 2.11 6 974 \800 2-methylpropyl acetatea,b,c Passion fruit 9.69 16.12 11.41 – 0.93 0.93 9 1019 803 ethyl butanoatea,b,c Sweet/strawberry 435.71 621.08 1047.15 8.77 8.85 8.49 10 1027 \800 2-methyl-3-buten-2-olb,c Not detected by OSME 11.24 18.23 18.01 – – – 12 1066 815 butyl acetatea,b,c Green, sweet 5.98 11.60 15.44 – 2.77 4.72 14 1073 \800 hexanala,b,c Not detected by OSME 51.93 19.60 13.70 – – – 15 1096 \800 diethyl carbonateb,c Plastic nd–0.32 0.72 0.79 3.37 5.23 6.60 17 1125 886 o-xyleneb,c Plastic tr–0.15 nd–0.72 0.94 – – – 18 1131 \800 methyl 2-methylbutanoateb,c Not detected by OSME 1.19 4.31 3.40 – – – 19 1155 \800 1-butanola,b,c Sweet/fruity 1.20 2.72 6.02 2.32 2.08 3.26 20 1172 993 beta-myrcenea,b,c Citric/fruity 13.99 38.03 28.49 – – 1.57 22 1175 934 methyl hexanoatea,b,c Strawberry, earthy 0.84 1.52 5.87 – 0.91 1.55 23 1184 1029 D-limonenea,b,c Not detected by OSME 22.60 41.87 22.56 – – – 24 1210 972 butyl butanoatea,b,c Sweet 1.55 2.02 4.33 – – 1.15 25 1228 1001 ethyl hexanoatea,b,c Syrup, guarana 54.38 70.88 160.33 6.42 6.13 6.61 26 1244 1052 cis-beta-ocimeneb,c Not detected by OSME 8.75 15.31 17.18 – – – 28 1268 1019 hexyl acetatea,b,c Floral 15.03 38.10 62.22 – – 0.94 29 1274 1006 octanala,b,c Sweet, acid 1.09 2.09 2.05 – 1.54 1.61 30 1275 990 ethyl cis-3-hexenoateb,c Fruity/floral nd–0.45 nd–tr tr–0.35 2.00 2.06 1.52 32 1300 1009 trans-3-hexenyl acetateb,c Fruity, citrus 0.51 1.42 2.17 – – 1.24 33 1309 991 cis-3-hexenyl acetateb,c Fruity, green 15.37 34.02 42.77 2.33 2.53 3.04 34 1328 1021 ethyl trans-2-hexenoateb,c Floral 1.18 1.55 1.32 – – – 36 1355 870 1-hexanola,b,c Passion fruit, skin 13.41 25.66 52.21 – – 2.58 37 1362 1039 trans-3-hexen-1-olb,c Not detected by OSME 1.18 1.84 3.40 – – – 38 1382 855 cis-3-hexen-1-ola,b,c Passion fruit, grass 4.77 9.39 11.95 2.31 4.26 4.35 39 1402 1193 butyl hexanoatea,b,c Unripe fruit 0.51 2.38 4.25 – – 1.72 40 1406 1193 hexyl butanoatea,b,c Not detected by OSME 5.38 18.12 43.99 – – – 42 1449 1198 ethyl octanoatea,b,c Earthy 0.23 0.34 1.25 3.88 3.43 3.26 43 1452 1185 cis-3-hexenyl butanoateb,c Not detected by OSME 2.25 5.84 10.94 – – – 46 1495 1385 alpha-copaeneb,c Green, fruity tr–0.18 tr–0.45 0.97 1.52 1.12 0.98 47 1515 971 benzaldehydeb,c Lavander 0.77 1.29 1.38 – 1.14 0.87 A 1537 – ni Citric nd nd nd 1.96 2.09 1.32 B 1570 – ni Citric, lavander nd nd nd – – 1.59 48 1575 1100 beta-linaloola,b,c Sweet, citrus 0.73 2.27 2.36 1.63 2.02 2.42 51 1598 1387 hexyl hexanoatea,b,c Synthetic, rubber 1.95 7.22 18.11 1.11 1.12 1.11 52 1646 1381 cis-3-hexenyl hexanoateb,c Green, citric 0.70 2.31 4.55 – 1.44 1.40 53 1689 1487 germacrene Db,c Passion fruit 0.36 0.23 0.28 0.94 1.75 1.85 C 1705 – ni Passion fruit nd nd nd 1.17 1.12 0.50 54 1715 1195 alpha-terpineola,b,c Passion fruit/leafy nd -0.92 0.72 2.59 1.44 2.87 4.48 56 1753 – ni Green, passion fruit nd–0.30 nd nd – 1.80 2.54 57 1771 – ni Green, leafy nd nd–0.52 nd – 0.98 2.11 D 1785 – ni Green, sweet nd nd nd – 1.09 1.57 E 1813 – ni Passion fruit nd nd nd 1.73 1.51 1.31 58 1840 – ni Sweet, molasses nd nd–0.34 nd 4.27 3.83 3.57 F 1913 ni Rubber, passion fruit nd nd nd – 0.87 1.92 2596 J Food Sci Technol (July 2017) 54(8):2594–2601 123 aromagram was constructed based on individual aroma- grams, in which peaks should be detected by at least two of four judges. Description of each odor was established combining descriptions of all judges, and checked against literature (Acree and Arn 2004; Jordán et al. 2002; Jan- zantti et al. 2012; Janzantti and Monteiro 2014). Retention indices in the consensus aromagram were calculated to confirm that the chromatographic data and compounds identity were correctly related to the olfactometric data. The odoriferous intensity 4.0 or more, between terms ‘‘moderate’’ and ‘‘strong’’, was of high importance, and those between 3.0 and 3.9, of moderate importance and between 0.1 and 2.9 of weak importance. Sensory evaluation 102 consumers (students and servers of UNESP) recruited from both gender, between 18 and 50 years old, who liked at least slightly and consumed passion fruit pulp and juice at least once in 15 days, participated in the acceptance test. Pulps from the 1/3, 2/3 and 3/3 yellow skin color stages of ripeness were used to prepare juices in a proportion of pulp:water of 1:4 (v/v). Juices were served (30 mL) at 12 �C in plastic glasses, coded with random three-digit numbers in monadic and randomized order (Macfie et al. 1989). Consumers evaluated appearance and aroma of the juices in standardized sensory booths, then they sweetened the juice with sugar or sweetener according to their taste, and evaluated overall impression, flavor, characteristic passion fruit flavor and sourness. A nine-point hedonic scale (9 = like extremely, 5 = neither like nor dislike, 1 = dislike extremely) was applied. Statistical analysis The olfactometric data were analyzed using Excel (Mi- crosoft Office, 2007). Sensory data were submitted to ANOVA and Tukey test (P B 0.05). The principal com- ponent analysis (PCA) was performed based on the cor- relation matrix, using relative peak area of volatile compounds and odoriferous intensity, as well as acceptance means. The volatile compounds were represented by name and/or chemical class and aroma description of passion fruit in each stage of ripeness. PCA was carried out using the Statistica 7.0 software (StatSoft, Tulsa, USA). Results and discussion Sixty volatile compounds were detected in the conven- tional yellow passion fruit at the 1/3 yellow skin color stage of ripeness, 62 at 2/3 and 60 at 3/3 (Table 1 and Fig. 1a). Table 1 continued Peak RI1 RI2 Compound3 Aroma description GC–FID4 GC–O5 1/3 2/3 3/3 1/3 2/3 3/3 59 1941 – ni Sweet 0.65 1.11 0.90 – – – G 1960 – ni Fruity, peach nd nd nd 2.44 2.15 2.31 60 1962 – ni Passion fruit, sweet nd 0.33 nd – – 0.10 61 1963 1487 dodecanolb,c Sweet 0.93 1.31 1.07 1.36 1.33 0.78 63 2072 – ni Candy floss nd–0.21 0.30 0.54 2.33 4.88 4.49 64 2166 – ni Candy floss/caramel nd nd tr – – 2.22 65 2188 – methyl hexadecanoateb,c Metalic/solvent nd–0.22 nd–0.25 0.38 – – – 66 2217 – ni Solvent nd nd tr 4.72 1.89 3.74 1 RI = retention index of peak in the DB-Wax column 2 RI = retention index of peak in the DB-5 column; - not calculated 3 Compounds have identical number in the chromatogram and aromagram. Letters were attributed to the odor active compounds not correlated to the volatile compounds detected by GC-FID 4 GC-FID: relative area of peak (n = 3) in the GC-FID, multiplied by 100, nd compound not detected, ni compound not identified, tr trace, relative area of peak\ 0.1 5 GC-O: maximum odoriferous intensity, C 4.0: between moderate and strong, 3.0 to 3.9: moderate, 0.1 to 2.9: weak, – none (not detected by OSME) a compound identified by pure standards b compound identified by mass spectrometry c compound identified by calculating the retention index J Food Sci Technol (July 2017) 54(8):2594–2601 2597 123 Esters (26) was the main chemical class, plus alcohols (7), terpenes (6), aldehydes (3), acid (1) and hydrocarbon (1). Ethyl butanoate, ethyl hexanoate, hexanal, D-limonene and a non-identified compound (peak 8, RI 1016) were major in the 1/3 yellow skin color. Exception of hexanal, the 2/3 yellow skin color had the same compounds as the 1/3, plus hexyl acetate, beta-myrcene, cis-3-hexenyl acetate and 1-hexanol. The 3/3 yellow skin color showed the same compounds as the 2/3, plus hexyl butanoate (Table 1). All the compounds had relative peak area increased during maturation, exception of methyl butanoate and hexanal. Hexanal was major only in the 1/3 yellow skin color and had peak area reduced during ripeness, probably because its conversion into hexanol (Damodaran et al. 2007; El Hadi et al. 2013; Janzantti and Monteiro 2014). Passion fruit volatile compounds are formed from chemical and enzymatic reactions of non-volatile precursors, degradation products of carotenoids, free fatty acids and sulphur-con- taining components (Engel and Tressl 1991; Winterhalter 1991; Pino 1997). No sulphur compounds were identified in this work (Table 1), as reported in other studies (Jordán et al. 2002; Pontes et al. 2009). In the organic passion fruit (Feltrin selection) from the same stages of ripeness, ethyl butanoate, ethyl hexanoate and cis-beta-ocimene were major, exception of ethyl hex- anoate at the 1/3 yellow skin color (Janzantti and Monteiro Fig. 1 Chromatogram (a) and consensual aromagram (b) of the volatile compounds from the conventional passion fruit in the stages of ripeness corresponding to the 1/3, 2/3 and 3/3 yellow skin color 2598 J Food Sci Technol (July 2017) 54(8):2594–2601 123 2014). In the conventional passion fruit from Afruvec material, ethyl butanoate and hexanoate, hexyl acetate, butanoate and hexanoate, and hexanol were also major compounds in the 3/3 yellow skin color (Macoris et al. 2011; Janzantti et al. 2012), differently from the present work. Hexyl hexanoate, 1-hexanol, hexadecanoic acid, linalool and alpha-terpineol were reported as major com- pounds in the ripe passion fruit from Golden Yellow variety (Pino 1997), different from the present work. Dif- ferences in passion fruit volatile composition are probably due to the material/cultivar, edaphoclimatic conditions and extraction technique. Twenty-three odor-active compounds were perceived in the conventional passion fruit at the 1/3 yellow skin color, and 34 and 42 in 2/3 and 3/3, respectively (Table 1 and Fig. 1b). Ethyl butanoate (sweet, strawberry) and hex- anoate (guarana syrup), and propyl acetate (skin, passion fruit) showed the highest odoriferous importance for the passion fruit in the 1/3 yellow skin color. Cis-3-hexen-1-ol (passion fruit, grass) and diethyl carbonate (plastic) plus Fig. 2 PCA of volatiles and odor-active compounds from the conventional passion fruit in the 1/3, 2/3 and 3/3 yellow skin color stage of ripeness and sensory acceptance of the juices. 1/3 = conventional passion fruit from the 1/3 yellow skin color stage of ripeness, 2/3 = conventional passion fruit from the 2/3 yellow skin color stage of ripeness, 3/3 = conventional passion fruit from the 3/3 yellow skin color stage of ripeness J Food Sci Technol (July 2017) 54(8):2594–2601 2599 123 the compounds from the 1/3 yellow skin color showed high odoriferous importance in the 2/3 yellow skin color, and butyl acetate (sweet, green) and alpha-terpineol (passion fruit/leafy) plus the same compounds from 2/3 were the most important for the 3/3 yellow skin color. The volatile compounds improved aroma intensity dur- ing ripeness. Butanol (sweet/fruity), cis-3-hexenyl acetate (fruity, green), and alpha-terpineol (passion fruit/leafy) showed weak odoriferous intensity in the 1/3 and 2/3 yel- low skin color and moderate in 3/3. Butyl acetate (green, sweet) and a non-identified compound (RI 1339, plastic, sweet) without odoriferous importance in the 1/3, showed weak odoriferous importance in 2/3 and moderate in 3/3. Beta-myrcene (citric/fruity), butyl butanoate (sweet), hexyl acetate (floral), trans-3-hexenyl acetate (fruity, citrus), 1-hexanol (passion fruit, skin), butyl hexanoate (unripe fruit), and two non-identified compounds (RI 1570, citric, lavender and IR 2166, candy floss/caramel) were weakly perceived only in the 3/3. Ethyl butanoate and hexanoate had high odoriferous importance in the organic passion fruit (Feltrin selection) in the 1/3 and 2/3 yellow skin color, plus propyl acetate in the last one (Janzantti and Monteiro 2014). Ethyl hexanoate, diethyl carbonate and propyl acetate showed high odorif- erous importance in the conventional and organic passion fruit from the Afruvec material, with a more expressive intensity in the organic fruit (Janzantti et al. 2012). Acceptance analysis Acceptance of the juices increased during maturation. Means of appearance ranged from 7.1 to 7.4, aroma 6.2 to 7.4, overall impression 5.6 to 6.9, flavor 5.3 to 6.9, char- acteristic passion fruit flavor 5.4 to 7.0 and sourness 5.0 to 6.0, from the 1/3 to 3/3 yellow skin color. No difference (P[ 0.05) was found between the 2/3 and 3/3 yellow skin color juice appearance and 1/3 and 2/3 for overall impression, flavor, passion fruit characteristic flavor and sourness. Juices from all stages of ripeness differed (P B 0.05) for aroma. PCA analysis The 1/3 yellow skin color, loaded negatively on PC 1 and positively on PC 2, was characterized by hexanal and methyl butanoate, and aroma ‘‘earthy’’ and ‘‘sweety’’. They are important for the unripe passion fruit aroma. The 2/3 yellow skin color, loaded positively on PC 1 and negatively on PC 2, was characterized mainly by D-limonene, beta- myrcene and esters group 2 (2-methylpropyl acetate, methyl 2-methylbutanoate and ethyl trans-2-hexenoate). The 3/3 yellow skin color, loaded positively on PC 1 and PC 2, was characterized by propyl acetate, esters group 1 (ethyl propanoate; diethyl carbonate; amyl isobutanoate; butyl, 3-methyl-2-butenyl, hexyl, trans-3-hexenyl and cis- 3-hexenyl acetate; ethyl, butyl, hexyl, trans-3-hexenyl and cis-3-hexenyl butanoate; methyl, ethyl, butyl, hexyl and cis-3-hexenyl hexanoate; ethyl cis-3-hexenoate; ethyl octanoate and methyl hexadecanoate), terpenes, aldehydes and alcohols, and aroma ‘‘passion fruit’’, ‘‘passion fruit, fruity’’, ‘‘passion fruit, sweety’’, ‘‘candy’’, ‘‘citrus’’ and ‘‘plastic’’, important for the ripe passion fruit aroma (Fig. 2). Sensory acceptance in the ripe fruit was high- lighted. Characteristic and natural aroma of passion fruit were completely developed in the ripe fruit as perceived by the judges using GC–O–OSME and acceptance test. Conclusion The CG–O–MS discriminated quite well the conventional passion fruit during maturation. The passion fruit volatile compounds peak area, odoriferous intensity and sensory acceptance of the juices increased during ripeness, indi- cating that the conventional passion fruit characteristic aroma is completely expressed when the fruit reaches the whole maturation, at the 3/3 yellow skin color. These outcomes will guide passion fruit producers to harvest ripe fruit in order to yield juice of better quality and flavor to meet consumer demand. Acknowledgements The authors thank to Mariana Macoris for the GC–O analysis, the Brazilian Federal Agency for Support and Eval- uation of Graduate Education-CAPES (Prodoc No. 06/58967-0), Cia das Ervas and Fazenda Gralha Branca-Sumaré for the financial support. References Acree TE, Arn H (2004) Flavornet. http://www.flavornet.org/ flavornet.html. Accessed 26 June 2016 Amaro AP, Monteiro M (2001) Rendimento de extração da polpa e caracterı́sticas fı́sico-quı́micas do maracujá amarelo (Passiflora edulis f. Flavicarpa Sims. Deg.) produzido por cultivo orgânico e convencional em relação à cor da casca. Alim Nutr 12:171–184 Casimir DJ, Kefford JF, Whitfield FB (1981) Technology and flavour chemistry of passion fruit juice and concentrate. Adv Food Res 27:243–293 Da Silva MAAP (1999) Avaliação de atributos sensoriais por técnicas tempo-intensidade. In: Almeida TCA, Hough G, Damásio MH, Da Silva MAAP (eds) Avanços em Análise Sensorial. Livraria Varela, São Paulo, pp 49–61 Da Silva MAAP, Lundhal DS, McDaniel MR (1994) The capability and psychophysics of Osme: a new GC-olfactometry technique. In: Maarse H, Van Der Heij DG (eds) Trends in Flavour Research. Elsevier, Amsterdam, pp 191–209 Damodaran S, Parkin KL, Fennema OR (2007) Fennema’s food chemistry. CRC Press, Boca Raton De Marchi R, Monteiro M, Benato EA, Silva CAR (2000) Uso da cor da casca como indicador de qualidade do maracujá amarelo 2600 J Food Sci Technol (July 2017) 54(8):2594–2601 123 http://www.flavornet.org/flavornet.html http://www.flavornet.org/flavornet.html (Passiflora edulis Sims. f. flavicarpa Deg.) destinado à indus- trialização. Food Sci Technol 20:381–387 El Hadi MAM, Zhang F-J, Wu F-F, Zhou C-H, Tao J (2013) Advances in fruit aroma volatile research. Molecules 18:8200–8229 Engel KH, Tressl R (1991) Identification of new sulfur-containing volatiles in yellow passion fruits (Passiflora edulis F. flavi- carpa). J Agric Food Chem 39:2249–2252 IBGE. Instituto Brasileiro de Geografia e Estatı́stica. Banco de dados agregados. (2014) Quantidade produzida, valor da produção, área plantada e área colhida da lavoura permanente. ftp://ftp. ibge.gov.br/Producao_Agricola/Producao_Agricola_Munici pal_%5Banual%5D/2013/tabelas_pdf/tabela03.pdf. Accessed 26 June 2016 Janzantti NS, Monteiro M (2014) Changes in the aroma of organic passion fruit (Passiflora edulis Sims f. flavicarpa Deg.) during ripeness. Food Sci Technol LEB 59:612–620 Janzantti NS, Macoris MS, Garruti DS, Monteiro M (2012) Influence of the cultivation system in the aroma of the volatile compounds and total antioxidant activity of passion fruit. Food Sci Technol LEB 46:511–518 Jordán MJ, Goodner KL, Shaw PE (2002) Characterization of the aromatic profile in aqueous essence and fruit juice of yellow passion fruit (Passiflora edulis Sims F. Flavicarpa degner) by GC-MS and GC/O. J Agr Food Chem 50:1523–1528 Macfie HJ, Bratchell N, Greenhoff K, Vallis CV (1989) Designs to balance the effect of order of presentation and first-order carry- over in hall tests. J Sens Stud 4:129–148 Macoris MS, Janzantti NS, Garruti DS, Monteiro M (2011) Volatile compounds from organic and conventional passion fruit (Pas- siflora edulis F. flavicarpa) pulp. Food Sci Technol 31:430–435 Macoris MS, De Marchi R, Janzantti NJ, Monteiro M (2012) The influence of ripening stage and cultivation system on the total antioxidant activity and total phenolic compounds of yellow passion fruit pulp. J Sci Food Agr 92:1886–1891 Pino JA (1997) Los constituyentes volatiles de la fruta de la passion. Alimentaria 280:73–81 Pontes M, Marques JC, Câmara JS (2009) Headspace solid-phase microextraction-gas chromatography-quadrupole mass spectro- metric methodology for the establishment of the volatile composition of Passiflora fruit species. Microchem J 93:1–11 Werkhoff P, Guentert M, Krammer G, Sommer H, Kaulen J (1998) Vacuum headspace method in aroma research: flavor chemistry of yellow passion fruits. J Agric Food Chem 46:1076–1093 Winterhalter P (1991) Fruits IV. In: Maarse H (ed) Volatile compounds in foods and beverages. Marcel Dekker Inc., New York, pp 389–409 J Food Sci Technol (July 2017) 54(8):2594–2601 2601 123 ftp://ftp.ibge.gov.br/Producao_Agricola/Producao_Agricola_Municipal_%255Banual%255D/2013/tabelas_pdf/tabela03.pdf ftp://ftp.ibge.gov.br/Producao_Agricola/Producao_Agricola_Municipal_%255Banual%255D/2013/tabelas_pdf/tabela03.pdf ftp://ftp.ibge.gov.br/Producao_Agricola/Producao_Agricola_Municipal_%255Banual%255D/2013/tabelas_pdf/tabela03.pdf HS--GC--MS--O analysis and sensory acceptance of passion fruit during maturation Abstract Introduction Materials and methods Chemicals Yellow passion fruit samples Isolation of volatile compounds from pulp High resolution gas chromatography (HSGC--FID) Gas chromatography--mass spectrometry (GC--MS) Gas chromatography--Olfactometry (GC--O) Sensory evaluation Statistical analysis Results and discussion Acceptance analysis PCA analysis Conclusion Acknowledgements References