Frostbitten Olives - ACS Publications - American Chemical Society

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Study of Volatile Compounds of Virgin Olive Oils with “Frostbitten Olives” Sensory Defect Inmaculada Romero, Diego Luis Garcia-Gonzalez, Ramon Aparicio-Ruiz, and Maria Teresa Morales J. Agric. Food Chem., Just Accepted Manuscript • Publication Date (Web): 10 May 2017 Downloaded from http://pubs.acs.org on May 10, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

Journal of Agricultural and Food Chemistry

Study of Volatile Compounds of Virgin Olive Oils with “Frostbitten Olives” Sensory Defect

Inmaculada Romero†, Diego L. García-González†*, Ramón Aparicio-Ruiz†, María T. Morales‡

†

, Instituto de la Grasa (CSIC), Campus Universidad Pablo de Olavide - Edificio 46, Ctra. de Utrera, km. 1 - 41013 – Sevilla, Spain.

‡

, Department of Analytical Chemistry, University of Seville, C/ Prof. García González, 2, 41012 - Sevilla, Spain.

*Corresponding author (Tel: +34954611550; E-mail: [email protected])

1 ACS Paragon Plus Environment


1

ABSTRACT

2

The freeze injuries in olives are responsible for the ‘frostbitten olives’ sensory

3

defect that is sometimes detected in virgin olive oil. This defect is becoming one of the

4

most common negative attributes today because the climate change has modified the

5

weather patterns. The temperature changes can take place abruptly, with rapid freeze-

6

thaw cycles, or gradually. These changes produce significant alterations in the quality of

7

the oils. This study analyzed the volatile composition of virgin olive oils characterized

8

with ‘frostbitten olives’ defect. The volatile information allowed grouping these oils

9

into two types characterized with two different profiles. One of them is characterized by

10

“soapy” and “strawberry-like” perceptions and the presence of two volatile compounds

11

(ethyl 2-methyl butanoate and ethyl propanoate). The second profile is characterized by

12

“wood” and “humidity” descriptors, and a high concentration of two volatiles (pentanal,

13

and octanal). These results on volatiles explain the existence of two sensory profiles

14

associated to the “frostbitten olives” defect.

15 16

Keywords: virgin olive oil, volatiles, ‘frostbitten olives’ defect, quality.

17 18 19 20 21 22 23 24 25


Page 2 of 27

Page 3 of 27


26

INTRODUCTION

27

According to different regulations1,2 virgin olive oils are classified into several

28

categories. Extra virgin oils (EVOO) are characterized by the absence of sensory defects

29

while the other categories -virgin (VOO), ordinary (OVOO) and lampante (LVOO) -

30

could have sensory defects with a perception intensity that ranges from low to very

31

high.1 Several malpractices related to olive oil processing and inadequate olive oil

32

storage are responsible for olive oil sensory defects though severe climatic conditions

33

and olive infestations also produce them.

34

The four most known sensory defects (fusty, rancid, winey-vinegary and

35

mustiness-humidity) have been widely studied3 and their volatile markers are well

36

described as well as the origin and causes of the defects. However, other sensory

37

defects, which were unusual only a few decades ago, are becoming much more common

38

today and little is known about their chemical characteristics. The change on the

39

weather conditions in the last years is leading to a higher incidence of sensory defects

40

produced by a non-adequate temperature during ripening of olives. Thus, low

41

temperatures can damage the olives before harvesting so modifying the chemical

42

composition and sensory quality of the final product. This undesirable phenomenon

43

results in the olive oil off-flavor called ‘frostbitten olives’, which is one of the sensory

44

defects described in the current regulation.4 This sensory defect is becoming one of the

45

most common negative attributes today because the climate change has leaded to new

46

weather patterns characterized with several freeze-thaw cycles in warm autumns and in

47

the beginning of winters. Thus, olive groves are often affected by frost within the

48

Mediterranean basin5 and in new producing countries such as Australia, Argentina and

49

USA.6-8 Previous studies have revealed that frost damage caused changes in the

50

composition of virgin olive oils such as slight decreases in chlorophyll and carotenoid



51

contents.9,10 These study also showed that virgin olive oils from olives affected by

52

freeze injured are characterized with low stability, pungency and bitterness.

53

Numerous studies have been carried out on the characterization of virgin olive

54

oils of different qualities and categories by quantitative analysis of the volatile

55

compounds,3,11-13 but no study has been carried out on the composition of volatile

56

compounds of ‘frostbitten olives’ sensory defect. Since the frost of the olives produce

57

chemical and biochemical changes in the virgin olive oil14-16, it is expected to determine

58

a characteristic volatile profile associated to oils with ‘frostbitten olives’ defect. Thus,

59

the aim of this paper is the characterization of the volatile composition that explain the

60

occurrence of ‘frostbitten olives’ sensory defect and to select the most characteristic

61

compounds associated to the identified sensory profiles.

62 63

MATERIALS AND METHODS

64

Reagents

65

Octane (99%), pentanal (98%), hexanal (98%), heptanal (95%), octanal (98%),

66

nonanal (95%), E-2-hexenal (97%), E-2-heptenal (98%), ethanol (96%), butan-1-ol

67

(99%), butan-2-ol (99.5%), 3-methyl-butan-1-ol (98%), hexan-1-ol (98%), E-3-hexen-1-

68

ol (98%), heptan-2-ol (99%), 4-methyl-pentan-2-one (99%), pentan-3-one (99.5%),

69

heptan-2-one (98%), 1-penten-3-one (98%), 6-methyl-5-hepten-2-one (99%), octan-3-

70

one (97%), acetic acid (99%), propanoic acid (99.5%), butanoic acid (99%), pentanoic

71

acid (99%), ethyl acetate (99.5%), ethyl propanoate (99%), ethyl butanoate (98%) and

72

ethyl 2-methylbutanoate (94%) were purchased from Sigma-Aldrich (St. Louis, MO),

73

The standards of methyl heptadecanoate (C17:0 methyl ester), silica gel 60-200

74

µm mesh, Sudan I (1-phenylazo-2-naphtol), phenolphthalein, starch soluble, potassium

75

iodide, potassium iodate, sodium thiosulfate pentahydrate, n-hexane and n-heptane were


Page 4 of 27

Page 5 of 27


76

purchased from Sigma-Aldrich (St. Louis, MO). Ethyl ether was from Romil Ltd.

77

(Cambridge, U.K.).

78 79

Samples

80

Twenty-two samples of olives were collected from orchards placed in the

81

Spanish main producer regions - Andalusia and Castilla-La Mancha - (Table 1). The

82

samples were made at industrial scale from olives affected by frost during the crop

83

2013/2014 that were processed in two-phase centrifugation systems to obtain filtered

84

virgin olive oils (VOO). The samples were produced by the same condition. All VOO

85

samples were described with the ‘frostbitten olives’ sensory defect by the panelists of a

86

panel test under the procedure of IOC4. Discussions with agronomists at the olive mills

87

accredited the drop of the temperature before collecting the olives (from -2 ºC and -4 ºC

88

as minimum temperature). No other sensory defect was detected in those samples,

89

excepting in samples 15 and 17 where a slight rancidity was also detected as a

90

secondary defect. A set of seven extra virgin olive oils (EVOO), provided by the same

91

producers, was also analyzed.

92 93

Quality parameters

94

Free acidity, peroxide value (PV) and specific UV absorption at 232 nm (K232)

95

and 270 nm (K270) were determined according to the methods of the International

96

Organization for Standardization (ISO), ISO 66017, ISO 396018 and ISO 365619,

97

respectively, as they are quality parameters included in the trade standard regulation of

98

the International Oil Council (IOC).1 For each parameter, two replicates of samples

99

were analyzed.



100

The quantitation of fatty acid ethyl esters (FAEEs) was carried out according to

101

the IOC method for the determination of the content of waxes, fatty acid methyl esters

102

(FAMEs) and fatty acid ethyl esters (FAEEs) by capillary gas chromatography using 3

103

grams of silica.20

104 105

Sensory assessment of olive oils

106

Sensory assessment was carried out by 19 trained panelists (16 females and 3

107

males between 25 and 50 years old) of an accredited panel who were trained for

108

recognition of VOO selected attributes according to the methodology proposed by the

109

International Olive Council.4 Each sample (15 mL) were tasted in normalized blue-

110

coloured glass containers to mask the colour differences. The temperature of the oils

111

was kept at 28±2 °C. Samples were labeled with a digit code and served following a

112

balanced rotation plan.

113 114

Odor threshold of volatile compounds

115

Odor detection threshold for each volatile compound was determined by means of the

116

procedure described by Luna et al., 2006.12 The volatile compounds were diluted in a

117

refined olive oil, which was previously tested by SPME-GC (described below) to

118

guarantee the absence of oxidation compounds. All the testing sessions were carried out

119

by 8 assessors (4 males and 4 females) on the same weekday (Tuesday morning) to

120

improve the consistency of results. Three samples were presented to the assessors,

121

following the triangle test method, and the results were statistically analyzed. A result

122

was considered consistent when at least 75% of the assessors provided the same odor

123

threshold. Before the triangle tests, an exploratory analysis was carried out by

124

presenting a series of dilutions to the assessors to estimate the concentration range


Page 6 of 27

Page 7 of 27


125

where the odor threshold may be placed. Fifteen milliliters of each sample were kept in

126

standardized tasting glasses at 29 ± 2ºC for 15 min and then tested. The sensory

127

description of the compounds shown in Table 2 was provided by assessors who smelled

128

the dilutions at a concentration above the odor threshold.

129 130

Determination of volatiles

131

The sample preparation was carried out according to the method applied in

132

previous works.21 The concentration step was carried out on a Combipal (CTC

133

Analytics AG, Zwingen, Switzerland). An olive oil sample (2 g) was placed in a 20 mL

134

glass vial, tightly capped with polytetrafluoroethylene (PTFE) septum, and left for 10

135

min at 40 ºC to allow for the equilibration of the volatiles in the headspace. After the

136

equilibration time, the septum covering each vial was pierced with a SPME needle and

137

the fiber was exposed to the headspace for 40 min. The SPME fiber (1 cm length and

138

50/30 µm film thickness) was purchased from Supelco (Bellefonte, PA), and it was

139

endowed

140

divinylbenzene/carboxen/polydimethylsiloxane (DVB/CAR/PDMS). The fiber was

141

previously conditioned following the instructions of the supplier.

with

the

Stable

Flex

stationary

phase

of

142

The volatiles adsorbed by the fiber were thermally desorbed in the hot injection

143

port of a GC for 5 min at 260 ºC with the purge valve off (splitless mode) and injected

144

in a TR-WAX capillary column (60 m × 0.25 mm i.d., 0.25 µm; Teknokroma, Spain) of

145

a Varian 3900 gas chromatograph (Palo Alto, CA) with flame ionization detector (FID).

146

The carrier gas was hydrogen, at a flow rate of 1.5 mL/min. The oven temperature was

147

held at 40 ºC for 10 min and then programmed to rise at 3 ºC/min to a final temperature

148

of 200 ºC. The signal was recorded and processed with the WorkStation (v6.41)

149

software. Each sample was analyzed in duplicate.



150

The identification of the volatile compounds was first carried out by mass

151

spectrometry21 and later checked with standards3,22, which were purchased from Sigma-

152

Aldrich (St. Louis, MO). Table 2 shows the volatile compounds identified in this study.

153

The quantification of volatile compounds was carried out with 4-methyl-2-pentanol as

154

an internal standard.

155 156

Gas chromatography-olfactometry

157

Gas chromatography-olfactometry (GC-O) was applied to VOO samples to

158

assess the aroma notes produced by volatile compounds. GC-O analyses were

159

performed on a Varian 3900 GC (Palo Alto, CA) equipped with an ATAS olfactory port

160

OP275 with a glass nasal cone (Veldhoven, Netherlands). The 40% of the flow was

161

directed to the FID, while 60% was directed into the heated (40 ºC) sniffing port.

162

The GC-O analysis was performed by four non-smoking and non-anosmic

163

assessors (3 females and 1 male). The perceived odor intensity was evaluated and

164

recorded using the scale of 1 (weak odor intensity), 2 (moderate odor intensity) and 3

165

(strong odor intensity). Only when the four assessors smelled the compound, the aroma

166

was taken into account. Assessors were suggested to limit the description of sensory

167

perceptions to the descriptors defined by the International Olive Council4 when possible

168

although some perceptions were named with different semantic terms. A consensus-

169

building discussion was held with assessors to decide the final sensory descriptors

170

qualifying the samples.

171

These sensory descriptors were processed in terms of their modified frequency

172

(MF), which is the result of combining intensity and frequency of detection of

173

descriptors, according with the formula proposed by Dravnieks (1985): MFሺ%ሻ = ඥFሺ%ሻ × Iሺ%ሻ


Page 8 of 27

Page 9 of 27


174

where F(%) is the detection of frequency of an aroma attribute expressed as percentage,

175

and I(%) is the average intensity expressed as percentage of maximum intensity. The

176

odors detected with a MF (%) higher than 50 correspond to the most relevant

177

compounds present in each sample from a sensory viewpoint.23-24

178 179

Statistical analysis

180

Univariate and multivariate algorithms were applied by means of Statistica 8.0

181

(Statsoft, Tulsa, OK). Brown-Forsythe25 tests was used to select the compounds that

182

showed significant differences (p1 in “H”

252

samples although only three (ethyl propanoate, ethyl 2-methylbutanoate and octanal)

253

were characterized with “soapy” and “strawberry-like” sensory perceptions. From the

254

perspective of OAV, ethyl 2-methylbutanoate (OAV=83.57) is the volatile with the

255

highest significance in “H” samples. Other relevant volatiles that were associated to

256

undesirable sensory perceptions of “H” samples were E-2-heptenal (OAV=16.40) and

257

heptan-2-ol (OAV=3.00). In addition to the compounds that contribute to negative

258

attributes, hexanal (OAV=7.70) and hexan-1-ol (OAV=5.60) were associated to the

259

usual pleasant and green sensory perceptions of these VOOs.

260

The average profile of “D” samples does not differ from “H” samples in a

261

qualitative but in a quantitative aspect. Thus, these results show that the olives

262

underwent similar chemical processes of degradation due to the low temperatures,

263

although with some variation in the volatile composition. The Brown-Forsythe test

264

showed that “D” samples were characterized with higher concentrations of pentanal and

265

octanal compared with “H” samples (Table 2). In the case of pentanal, the concentration

266

was three times higher in “D” samples. This compound provided attributes qualified as

267

“wood, oily” (OAV=2.38) while octanal was qualified as “soapy” (OAV=6.16). These

268

attributes corresponded to the sensory descriptors identified by panelists. E-2-heptenal

269

also contributed with a “woody” aroma with a high OAV (20.40) but this compound

270

was not selected by Brown-Forsythe tests (Table 2). Some studies carried out with

271

peppers submitted to a slow frozen process also showed high concentration of

272

aldehydes.32 Although “D” samples show a profile of volatiles responsible for

273

undesirable descriptors higher than “H” samples, their profiles also have volatiles


Page 12 of 27

Page 13 of 27


274

(hexanal, E-2-hexenal and hexan-1-ol) responsible for positive attributes with OAVs

275

>1.

276

In order to facilitate the sensory interpretation of the volatile composition, a GC-

277

O method was applied to evaluate the sensory characteristics and intensity of the

278

odorants (odor impact zones) of volatiles. In particular, the study was focused on

279

identifying the odour impact zones associated to the cherished strawberry-like

280

perception of “H” samples since this attribute has not been studied before. Table 2

281

shows the odorants responsible for the flavor as well as the odour intensity and the

282

modified frequency (MF%), a mixture of intensity and frequency of detection.33 The

283

study revealed that there were four odor impact zones that were basically named

284

“sweet”, “fruity”, “green” and “rancid”. In terms of sensory impact, and considering

285

these 4 odour impact zones, ten volatile compounds were the most descriptive

286

(MF%>50) (Table 2). Ethyl propanoate, hexanal and E-2-hexenal showed the highest

287

percentage of MF (MF%=95.7) followed by ethyl 2-methylbutanoate (MF%=79.0) and

288

hexan-1-ol (MF%=70.7). Only six of these compounds - ethyl propanoate, ethyl 2-

289

methylbutanoate, hexanal, hexan-1-ol, nonanal and acetic acid - have OAVs >1.0. In

290

general terms, the most relevant impact zone (higher MF%) was “fruity” and two of the

291

compounds associated to this zone were those esters responsible for the strawberry-like

292

perception (e.g. ethyl propanoate and ethyl 2-methylbutanoate). This result agrees with

293

the volatiles selected as markers of ‘frostbitten olives’ sensory defect in “H” samples.

294

Thus, the concluding information is that the relevant perceptions concern strawberry-

295

like, and fresh green sensory perceptions combined with others resulting from

296

fermentation, which is expected to occur in frostbitten olives where the fruit tissues are

297

damaged. The interpretation of the results was complemented with statistical analysis of

298

the data. The univariate algorithms - ANOVA and Brown-Forsythe test - were applied



299

to distinguish between EVOOs and the both VOO groups (“D” and “H”), and between

300

these two kinds of VOOs from frosted olives as well. ANOVA test pointed out that ten

301

volatiles with OAV>1 in at least one of the three groups (“H” or “D” or “E”) (ethyl

302

propanoate, pentanal, butan-2-ol, ethyl butanoate, hexanal, octan-3-one, octanal, hexan-

303

1-ol, E-3-hexen-1-ol and nonanal) showed significant differences (p

Frostbitten Olives - ACS Publications - American Chemical Society

Recommend Documents