Enantioselective analysis of secondary alcohols and their esters in purple and yellow passion fruits

Article abstract of DOI:10.1021/jf072464n

The enantiomeric compositions of the acetates, butanoates, hexanoates, and octanoates of the secondary alcohols 2-pentanol, 2-heptanol, and 2-nonanol were determined in yellow (Passiflora edulis f. flavicarpa) and purple (Passiflora edulis Sims) passion fruits. The compounds were isolated by means of simultaneous distillation-extraction. Enantiodifferentiation was performed via multidimensional gas chromatography using heptakis(2,3-di-O-methyl-6-O-tert- butyldimethylsilyl)-β-cyclodextrin as chiral stationary phase. The series of homologous 2-alkyl esters, which are typical constituents of purple passion fruits, were shown to be present as nearly optically pure (R)-enantiomers. The proportions of the (S)-enantiomers varied in different batches and were dependent on the alcohol moieties of the esters. For minor amounts of esters detected in yellow fruits, the (R)-enantiomers were also dominating. However, the enantiomeric excesses were significantly lower than in the purple variety. Enantioselective analysis of the free alcohols revealed that 2-heptanol exhibited opposite configurations in purple and yellow passion fruits. A similar phenomenon was observed for 2-pentanol, which was present in the yellow fruits as a nearly racemic mixture. Data determined in extracts obtained by other techniques (liquid-liquid extraction, vacuum headspace technique) showed that the isolation procedure had no significant impact on the enantiomeric ratios.

Full text of DOI:10.1021/jf072464n

J. Agric. Food Chem. 2007, 55, 10339–10344 10339
Enantioselective Analysis of Secondary Alcohols
and Their Esters in Purple and Yellow Passion Fruits
HEDWIG STROHALM, MÁRTA DREGUS, ASTRID WAHL, AND KARL-HEINZ ENGEL*
Lehrstuhl für Allgemeine Lebensmitteltechnologie, Technische Universität München, Am Forum 2,
D-85350 Freising-Weihenstephan, Germany
The enantiomeric compositions of the acetates, butanoates, hexanoates, and octanoates of the
secondary alcohols 2-pentanol, 2-heptanol, and 2-nonanol were determined in yellow (Passiflora edulis
f. flavicarpa) and purple (Passiflora edulis Sims) passion fruits. The compounds were isolated by
means of simultaneous distillation-extraction. Enantiodifferentiation was performed via multidimen-
sional gas chromatography using heptakis(2,3-di-O-methyl-6-O-tert-butyldimethylsilyl)-ꢀ-cyclodextrin
as chiral stationary phase. The series of homologous 2-alkyl esters, which are typical constituents of
purple passion fruits, were shown to be present as nearly optically pure (R)-enantiomers. The
proportions of the (S)-enantiomers varied in different batches and were dependent on the alcohol
moieties of the esters. For minor amounts of esters detected in yellow fruits, the (R)-enantiomers
were also dominating. However, the enantiomeric excesses were significantly lower than in the purple
variety. Enantioselective analysis of the free alcohols revealed that 2-heptanol exhibited opposite
configurations in purple and yellow passion fruits. A similar phenomenon was observed for 2-pentanol,
which was present in the yellow fruits as a nearly racemic mixture. Data determined in extracts obtained
by other techniques (liquid–liquid extraction, vacuum headspace technique) showed that the isolation
procedure had no significant impact on the enantiomeric ratios.
KEYWORDS: Passion fruits; Passiflora edulis f. flavicarpa; Passiflora edulis Sims; secondary alcohols;
2-alkyl esters; enantioselective analysis; MDGC
INTRODUCTION
differentiation of the two passion fruit varieties. Purple passion
fruits contain acetates, butanoates, hexanoates, and octanoates
of 2-pentanol, 2-heptanol, and 2-nonanol as prominent con-
stituents; in the yellow variety these esters were not detected
(9) or were present at only trace levels (5, 10). The nonesterified
alcohols 2-pentanol and 2-heptanol were detected as minor
constituents in both varieties.
Among the 400 species of Passiflora, yellow (Passiflora
edulis f. flaVicarpa) and purple (Passiflora edulis Sims) passion
fruits are of major commercial importance owing to their
attractive exotic flavor. The volatile constituents of these tropical
fruits have been investigated by several research groups over
the past decades. The broad spectrum of constituents identified
has been described in comprehensive reviews (1, 2). The
presence of chiral sulfur-containing aroma compounds is a
typical feature of yellow passion fruits (3–5). Some unusual
norterpenoids, such as edulans (6) and megastigmatrienes (7)
have been discussed as essential contributors to the floral
character of purple passion fruits. Esters (aliphatic, aromatic,
and terpenoid) constitute the predominant class of compounds
in both varieties. A characterization of the aromatic profile
of an aqueous essence of yellow passion fruit by gas
chromatography-olfactometry revealed 2-methylbutyl hex-
anoate and hexyl hexanoate to be among the most intense
odorants (8).
Investigations of naturally occurring enantiomeric composi-
tions by capillary gas chromatographic (GC) separation of
diastereoisomeric esters of (R)-(+)-R-methoxy-R-trifluoro-
methylphenylacetic acid (MTPA) revealed that the free second-
ary alcohols 2-pentanol and 2-heptanol present in yellow fruits
mainly consisted of the (S)-enantiomer (11). In contrast, free
2-heptanol contained in purple passion fruits was almost
exclusively present as the (R)-enantiomer. 2-Heptanol esterified
in the 2-heptyl esters of purple passion fruits was shown to be
optically pure and possessed the (R)-configuration.
The indirect GC separation applied at that time was based
on the derivatization of enantiomers to diastereoisomeric esters.
For 2-heptyl esters this required the following workup sequence:
(i) liquid–solid chromatography of a passion fruit aroma extract
on silica gel to obtain an ester fraction; (ii) alkaline saponifica-
tion of the ester fraction; (iii) isolation of the liberated 2-heptanol
by preparative GC; and (iv) derivatization of the alcohol with
(R)-(+)-MTPA and subsequent GC separation. In addition to a
A subgroup of esters, short-chain fatty acid esters of odd-
numbered secondary alcohols, turned out to be suitable for a
* Author to whom correspondence should be addressed (telephone
+49 (0)8161 71 4250; fax +49 (0)8161 71 4259; e-mail k.h.engel@
wzw.tum.de).
10.1021/jf072464n CCC: $37.00
2007 American Chemical Society
Published on Web 11/16/2007

10340 J. Agric. Food Chem., Vol. 55, No. 25, 2007
Strohalm et al.
solution (pH 6.8) for 1 min. After addition of the internal standard
(2-hexanol, 300 µg) and pressing of the homogenate through a sieve,
the obtained juice was divided into three aliquots of 1000 mL.
Vacuum Headspace (VHS) Technique. One aliquot was transferred
into a 2 L round-bottom flask and connected to a vacuum headspace
apparatus (5). The flask was brought to a temperature of ∼35 °C in a
water bath (Gerhardt, type SV 24). Vacuum was applied for 3 h (1–10
mbar; Leybold-Heraus vacuum pump, type D4A). The volatiles were
condensed in three cooling traps, which were cooled with ice–water (I
and II) and liquid nitrogen (III), respectively. After 3 h, the aqueous
condensates were allowed to thaw, combined, and extracted three times
with 50 mL of a mixture of n-pentane and diethyl ether (1:1, v/v). The
combined extracts were dried over anhydrous sodium sulfate and
concentrated at ∼40 °C to a final volume of 1 mL using a Vigreux
column.
reduced sensitivity, this laborious procedure did not allow the
determination of the enantiomeric compositions of individual
ester homologues.
In the meantime, there has been tremendous progress regard-
ing the direct separation of enantiomers using chiral stationary
phases (12). The aim of this study was to exploit these analytical
techniques and to determine the naturally occurring enantiomeric
distributions of chiral secondary alcohols and their individual
esters in purple and yellow passion fruits via direct enantio-
differentiation using multidimensional gas chromatography and
a modified ꢀ-cyclodextrin as chiral stationary phase.
MATERIALS AND METHODS
Liquid–Liquid Extraction (LLE). One aliquot was transferred into a
Kutscher-Steudel liquid–liquid extractor (14). Extraction was performed
for 24 h using 200 mL of a mixture of n-pentane and diethyl ether
(1:1, v/v) as solvent.
SDE. One aliquot was subjected to SDE as described above (Isolation
of volatiles).
Materials. Purple (P. edulis Sims) and yellow passion fruits (P.
edulis f. flaVicarpa) from Colombia were purchased from a local market.
Yellow passion fruits were also directly obtained from Thailand by air
freight. The fruits were stored at 5 °C before analysis.
Chemicals. Authentic reference chemicals were purchased from
commercial sources (Aldrich, Steinheim, Germany; Merck, Darmstadt,
Germany). (S)-2-Pentanol, (R)-2-heptanol, and (S)-2-nonanol were
purchased from Aldrich, Steinheim, Germany. (S)-2-Heptanol was
obtained from Fluka, Steinheim, Germany. (R)-2-Pentanol and (R)-2-
nonanol were obtained from Wako Pure Chemical Industries, Ltd.,
Tokyo, Japan. Esters were synthesized from the corresponding acyl
chlorides using 4-dimethylaminopyridine as catalyst. The secondary
alcohol (20 mmol) was dissolved in pyridine (15 mL), and 4-dimethyl-
aminopyridine (250 mg) was added at room temperature. The mixture
was cooled in an ice–water bath, and the acyl chloride (20 mmol) was
added dropwise. After 3 h of stirring at room temperature, methanol
was added. The reaction mixture was poured into a stirred suspension
of MTBE and water. The organic layer was separated and washed with
HCl (2 N), water, saturated sodium bicarbonate solution, and saturated
sodium chloride solution, dried over anhydrous magnesium sulfate,
filtered, and concentrated. The obtained pale yellow oil was distilled
under reduced pressure to yield the product (yields 65–91%). The
identities of the compounds were confirmed by mass spectrometric
analyses: 2-pentyl acetate [m/e (%)], 43 (100), 87 (33), 70 (21), 55
(14), 42 (8), 41 (6), 71 (4), 45 (4), 115 (3), 61 (3); 2-pentyl butanoate,
71 (100), 43 (46), 70 (23), 89 (15), 115 (12), 55 (11), 41 (10), 42 (8),
88 (8), 72 (5); 2-pentyl hexanoate, 99 (100), 43 (57), 71 (41), 70 (40),
117 (27), 55 (18), 60 (16), 41 (14), 42 (13), 87 (13); 2-pentyl octanoate,
127 (100), 70 (56), 43 (50), 57 (49), 145 (38), 55 (29), 41 (21), 71
(21), 144 (19), 60 (18); 2-heptyl acetate, 43 (100), 87 (37), 56 (19), 98
(13), 41 (11), 55 (11), 57 (9), 70 (9), 69 (8), 58 (7); 2-heptyl butanoate,
71 (100), 43 (40), 57 (28), 115 (22), 56 (20), 98 (17), 41 (16), 89 (13),
70 (12), 55 (11); 2-heptyl hexanoate, 99 (100), 43 (37), 57 (33), 56
(27), 98 (24), 71 (22), 117 (19), 41 (16), 55 (15), 70 (14); 2-heptyl
octanoate, 127 (100), 57 (74), 98 (34), 56 (26), 43 (25), 145 (25), 55
(22), 41 (19), 70 (16), 144 (16); 2-nonyl acetate, 43 (100), 87 (41), 55
(18), 56 (17), 41 (13), 70 (12), 69 (12), 97 (12), 84 (11), 126 (10);
2-nonyl butanoate, 71 (100), 43 (38), 115 (19), 41 (14), 55 (14), 56
(14), 57 (12), 70 (12), 89 (12), 126 (11); 2-nonyl hexanoate, 99 (100),
43 (42), 71 (34), 56 (21), 117 (20), 55 (19), 126 (18), 41 (16), 57 (16),
97 (15); 2-nonyl octanoate, 127 (100), 57 (53), 43 (31), 126 (25), 84
(25), 55 (24), 145 (23), 41 (18), 97 (17), 71 (16).
Capillary Gas Chromatography (HRGC-FID). The separations
were performed on a Carlo Erba Mega II 8575 series gas chromatograph
(Thermo Fisher Scientific, Dreieich, Germany) equipped with a split/
splitless injector (215 °C, split ratio 1:10) and a flame ionization detector
(FID) operating at 230 °C. The column used was a 60 m × 0.25 mm
(i.d.) fused silica capillary column coated with DB-Wax (0.25 µm film
thickness; J&W Scientific). The oven temperature was programmed
from 40 °C (5 min hold) at 4 °C/min to 230 °C (25 min hold). Carrier
gas used was hydrogen at a constant inlet pressure of 105 kPa. Data
acquisition was done via Chromcard software (Thermo Fisher
Scientific).
Quantification. Quantifications were carried out using 2-hexanol
as internal standard (100 µg, stock solution in diethyl ether/ethanol,
4:1, v/v) and taking into account extraction recoveries and FID
responses. The recoveries after SDE from buffer solution were
determined in model experiments with authentic compounds (100 µg,
stock solution in diethyl ether/ethanol, 4:1, v/v). The following average
values were obtained: 2-pentyl esters (68%), 2-heptyl esters (75%),
2-nonyl esters (99%). For 2-heptyl hexanoate (2.5 mg) spiked to the
fruit homogenate (300 g) the recovery via SDE was 84%. FID response
factors were determined with solutions of authentic compounds relative
to the internal standard (0.1 µg/µL diethyl ether). The limits of detection
and the limits of determination were calculated according to described
procedures (15, 16), using a series of five dilutions of reference
compounds ranging from 0.5 to 10.0 µg/kg.
Gas Chromatography-Mass Spectrometry (GC-MS). Mass spec-
tral data were obtained on a gas chromatograph–mass spectrometer (GC
8000TOP with a Voyager GC-MS, Thermo Fisher Scientific) equipped
with a split/splitless injector (220 °C, split ratio 1:10). The separation
was performed on a 30 m × 0.25 mm (i.d.) fused silica capillary column
coated with DB-WaxEtr (0.5 µm film thickness; J&W Scientific). The
oven temperature was programmed from 40 °C (5 min hold) at 4 °C/
min to 240 °C (25 min hold). Carrier gas used was helium at a constant
inlet pressure of 75 kPa. Ionization energy was set at 70 eV, source
temperature at 200 °C, and interface temperature at 240 °C. Data
acquisition was done via MassLab software (Thermo Fisher
Scientific).
Isolation of Volatiles by Simultaneous Distillation-Extraction
(SDE). Three hundred grams of purple passion fruit pulp or 400 g of
yellow passion fruit pulp was homogenized with 750 mL of phosphate
buffer solution (pH 6.8) for 1 min. After addition of the internal standard
(2-hexanol, 100 µg), the homogenate was pressed through a sieve. The
juice was transferred into a 2 L round-bottom flask connected to an
SDE apparatus (13). Distillation-extraction was performed for 2 h,
using 200 mL of a mixture of n-pentane and diethyl ether (1:1, v/v) as
solvent. The aroma extract was dried over anhydrous sodium sulfate
and concentrated at ∼40 °C to a final volume of 1 mL using a Vigreux
column (30 cm × 2 cm i.d.).
Multidimensional Gas Chromatography. For enantioselective gas
chromatographic analysis an instrumentation of two coupled GC 8000
series gas chromatographs (Thermo Fisher Scientific) with two
independent temperature controls and with FID on each GC system
was used. The columns were coupled via a moving column stream
switching device (MCSS) (17, 18). An achiral column for preseparation
(precolumn) was installed in the first GC oven and was connected via
the MCSS device and a deactivated fused silica transfer capillary (1 m
× 0.25 mm i.d.) with the chiral separation column (main column) in
the second GC oven. In the first oven two types of achiral precolumns
were used: (I) DB-Wax (60 × 0.32 mm i.d., 0.25 µm film thickness,
J&W Scientific) and (II) DB-5 (60 × 0.32 mm i.d., 0.25 µm film
thickness, J&W Scientific). For both columns the temperature program
Comparison of Isolation Techniques. Nine hundred grams of purple
passion fruit pulp was homogenized with 2250 mL of phosphate buffer

Secondary Alcohols in Passion Fruit
J. Agric. Food Chem., Vol. 55, No. 25, 2007 10341
Figure 1. Typical gas chromatographic separation of secondary alcohols and their esters isolated from purple passion fruits by SDE [peak numbers
correspond to Table 1; for GC conditions see Capillary Gas Chromatography (HRGC-FID) under Materials and Methods; IS ) internal standard, 2-hexanol].
Table 1. Secondary Alcohols and Their Esters Isolated from Purple and
started at 40 °C (5 min hold) and was programmed at 4 °C/min to 230
Yellow Passion Fruits by Means of SDE
°C. In the second GC oven a chiral column was installed coated with
25% heptakis(2,3-di-O-methyl-6-O-tert-butyldimethylsilyl)-ꢀ-cyclodex-
concentrationsa (µg/kg)
trin in SE54 (30 m × 0.32 mm i.d., 0.25 µm film thickness). Synthesis
purple passion fruits, yellow passion
KIc (DB-Wax) Colombia (batch 1) fruits, Thailand
of the cyclodextrin derivative and column preparation were carried out
peakb
compound
alcohols
2-pentanol
2-heptanol
2-nonanol
in-house (19). The temperature of the chiral main column was
programmed from 37 °C (10 min hold) to 200 at 2 °C/min. Injection
into the precolumn was done in the split mode (215 °C; split ratio 1:15).
Carrier gas used was hydrogen at a constant inlet pressure of 165 kPa.
The outlet pressure of the column was 98 kPa (measured within the
“dome” of the MCSS device via an additional manometer), which also
corresponded to the actual inlet pressure for the chiral column in the
second GC oven. The FID of the first oven was set at 230 °C and the
FID of the second oven at 200 °C. Data acquisition and control of
the MCSS device were done via Chromcard software (Thermo Fisher
Scientific). MDGC analysis of the volatile compounds was performed
via four cut combinations. The cut intervals using DB-Wax as
precolumn were as follows: [I] 2-pentanol (12.18–12.35 min), 2-hep-
tanol (20.08–20.21 min); [II] 2-heptyl esters (17.81–17.95 min;
22.93–23.08 min; 29.38–29.57 min; 35.37–35.51 min), 2-nonyl hex-
anoate (35.37–35.51 min); [III] 2-pentyl butanoate (15.81–15.99 min),
2-pentyl hexanoate (23.14–23.22 min), 2-nonyl acetate (24.86–25.00
min); [IV] 2-pentyl acetate (10.34–10.49 min), 2-pentyl octanoate
(29.66–30.16 min), 2-nonyl butanoate (29.66–30.16 min). The limits
of detection and determination of the MDGC transfer were calculated
according to described procedures (15, 16), using a series of five
dilutions of reference compounds ranging from 0.5 to 3.0 µg/kg per
enantiomer.
2
5
9
1125
1328
1525
98 ( 19
68 ( 13
<2
22 ( 4
15 ( 11
ndd
esters
1
3
7
12
4
2-pentyl acetate
2-pentyl butanoate
2-pentyl hexanoate
2-pentyl octanoate
2-heptyl acetate
2-heptyl butanoate
2-heptyl hexanoate
2-heptyl octanoatee
2-nonyl acetate
1074
1216
1407
1605
1266
1401
1591
1788
1466
1600
1787
1982
138 ( 37
459 ( 171
438 ( 122
70 ( 30
215 ( 38
3800 ( 628
2403 ( 215
72 ( 6
nd
nd
<2
nd
nd
6
31 ( 13
36 ( 16
nd
nd
nd
10
14
8
11
13
15
<2
2-nonyl butanoate
2-nonyl hexanoatee
2-nonyl octanoate
187 ( 20
17 ( 3
<2
nd
nd
a Data from triplicate experiments for each batch; mean ( standard error.
b Numbers correspond to chromatogram peaks shown in Figure 1. c Kovats retention
indices. d Below limit of detection (<0.7 µg/kg). e Not separated; concentrations
estimated on the basis of the ratios of characteristic MS fragments: m/e 127 (2-
heptyl octanoate) and 99 (2-nonyl hexanoate).
RESULTS AND DISCUSSION
kg, respectively) of purple passion fruits. In contrast, in the
yellow variety these two esters were contained only as minor
volatiles. Apart from traces of 2-pentyl hexanoate, no other
esters of secondary alcohols could be detected. The free
secondary alcohols 2-pentanol and 2-heptanol were present in
both varieties; in the purple fruits also traces of 2-nonanol were
detected.
These quantitative patterns were in agreement with those
obtained after isolation of volatiles via LLE and were confirmed
in the other batches of purple and yellow passion fruits listed
in Table 2 (data not shown). They are in agreement with
previously reported data (5, 10) and confirm that esters of
secondary alcohols are suitable for a differentiation of the two
passion fruit varieties (9).
Quantification. Volatile constituents were isolated from
purple and yellow passion fruits by means of SDE. The extracts
obtained were analyzed by means of GC-FID and GC-MS.
Figure 1 shows the typical gas chromatographic separation of
volatiles isolated from purple passion fruits; the target com-
pounds of this study (secondary alcohols and their esters) are
indicated by numbers.
Quantitative data obtained from triplicate analyses of purple
and yellow passion fruits are presented in Table 1. In purple
fruits the complete series of the acetates, butanoates, hexanoates,
and octanoates of 2-pentanol, 2-heptanol, and 2-nonanol could
be detected. The 2-heptyl esters were dominating with regard
to the alcohol moieties, and the butanoates and hexanoates were
the major representatives with regard to the fatty acyl moieties.
Accordingly, 2-heptyl butanoate and 2-heptyl hexanoate were
shown to be prominent ester constituents (3800 and 2400 µg/
Enantioselective MDGC. Several stationary phases were
screened regarding their suitability for enantioselective analysis
of the target compounds of this study. The enantiomers of the

10342 J. Agric. Food Chem., Vol. 55, No. 25, 2007
Strohalm et al.
Table 2. Enantiomeric Distributions of Secondary Alcohols and Their Esters from Purple and Yellow Passion Fruits
enantiomeric distributions (%)
purple passion fruits
yellow passion fruits
Thailanda Colombiaa
Colombia (batch 1)a
Colombia (batch 2)b
Colombia (batch 3)b
compound
alcohols
R
S
R
S
R
S
R
S
R
S
2-pentanol
2-heptanol
2-nonanol
70.8
85.7
–
29.2
14.3
–
85.5
92.5
–
14.5
7.5
–
84.3
72.0
–
15.7
28.0
–
49.7
29.1
ndd
50.3
70.9
nd
51.7
42.5
nd
48.3
57.5
nd
c
esters
2-pentyl acetate
2-pentyl butanoate
2-pentyl hexanoate
2-pentyl octanoate
2-heptyl acetate
2-heptyl butanoate
2-heptyl hexanoate
2-heptyl octanoatef
2-nonyl acetate
>98.0e
99.2
>98.7
>97.9
>98.4
99.4
99.8
>98.8
–
>95.3
>89.9
93.5
98.8
>99.5
99.2
99.8
99.1
>96.3
97.5
99.1
–
>98.2
92.9
96.4
>98.9
>98.6
97.7
99.3
98.6
>87.7
96.9
98.6
–
nd
nd
–
nd
nd
–
nd
nd
89.1
nd
nd
nd
10.9
nd
0.8
7.1
3.6
6.5
1.2
nd
nd
87.2
93.1
nd
nd
nd
nd
nd
nd
nd
12.8
6.9
nd
nd
nd
nd
nd
nd
nd
0.6
0.2
0.8
0.2
0.9
2.3
0.7
1.4
>84.6
94.4
nd
nd
nd
5.6
nd
nd
nd
nd
nd
–
1.8
2-nonyl butanoate
2-nonyl hexanoatef
2-nonyl octanoate
98.2
>98.8
–
2.5
0.9
–
3.1
1.4
–
nd
nd
–
a Triplicate isolation of volatiles and triplicate analyses of enantiomers. b Single isolation of volatiles and triplicate analyses of enantiomers. c Below limit of determination
(for alcohols, <0.7 µg/kg; for esters, <1.2 µg/kg). d Below limit of detection (<0.4 µg/kg). e Ratios “greater than (>)” were calculated using the peak area corresponding to
the limit of determination for the second enantiomer. f Coelution of the enantiomers.
homologous series (C5-C10) of secondary alcohols had been
separated on octakis(2,3-di-O-butyl-6-O-tert-butyldimethylsilyl)-
γ-cyclodextrin (20). Short-chain esters of 2-pentanol and
2-heptanol could be enantiodifferentiated using heptakis(2,3,6-
tri-O-methyl)-ꢀ-cyclodextrin and heptakis(2,3,6-tri-O-pentyl)-
ꢀ-cyclodextrin (Lipodex C) as chiral stationary phases (21, 22).
The recently developed modified cyclodextrins bearing acetal
functions as side chains also showed unusually high separation
factors for the short-chain 2-alkyl esters. However, the separation
factors decreased significantly with increasing lengths of the
acyl and alcohol moieties (23). Finally, heptakis(2,3-di-O-
methyl-6-O-tert-butyldimethylsilyl)-ꢀ-cyclodextrin was selected
as chiral stationary phase. Using this modified cyclodextrin,
known, for example, from the enantioseparation of chiral sulfur-
containing passion fruit volatiles (24), baseline separations could
be achieved for the enantiomers of the complete series of 2-alkyl
esters. Figure 2a shows exemplarily the enantioseparation of
the 2-heptyl esters. The order of elution of the enantiomers was
assigned by co-injection of optically pure reference compounds.
The enantiomeric distributions of secondary alcohols and their
esters determined in extracts obtained by SDE from purple and
yellow passion fruits are listed in Table 2. Data originate from
investigations of batches of different origins and from triplicate
analyses of enantiomers of each extract. In purple passion fruits
esters of secondary alcohols are present as nearly optically pure
(R)-enantiomers. For the quantitatively dominating 2-heptyl
butanoate and 2-heptyl hexanoate, the average proportion of
the (R)-enantiomer determined in two batches from Colombia
was 99.6%. The maximum proportions of (S)-configured esters
were 2.3 and 0.7%, respectively. For the ester homologues
occurring at lower concentrations, in most cases the amounts
detected for the (S)-enantiomers were below the limits of
determination. In these cases, quantification of the minor
enantiomer and accurate calculation of an enantiomeric ratio
were not possible (25). Data obtained for the 2-pentyl esters
indicate that for esters containing this alcohol moiety the excess
of the (R)-enantiomer was slightly less pronounced.
For 2-heptyl butanoate, 2-heptyl hexanoate, and 2-pentyl hexa-
noate detected as minor constituents in yellow passion fruits, also
the (R)-enantiomers were dominating. However, the enantiomeric
excesses were significantly lower than in the purple variety.
MDGC was employed to transfer the secondary alcohols and
their esters from the achiral precolumn to the chiral main
column. Taking into account the elution behavior of the target
compounds, four cut combinations (see Materials and Methods)
were designed, which enabled the transfer and enantiodiffer-
entiation of the complete set of alcohol and ester homologues.
The only compounds for which overlapping on both precolumns
and on the chiral stationary phase could not be avoided were
2-heptyl octanoate and 2-nonyl hexanoate.
In purple passion fruits the free secondary alcohols 2-pentanol
and 2-heptanol were mainly present as (R)-enantiomers. How-
ever, the enantiomeric excess was less pronounced than for the
esters. On the other hand, in yellow passion fruits the (S)-
enantiomer dominated for 2-heptanol, and for 2-pentanol nearly
racemic mixtures were found.
Exemplarily, Figure 2b shows the enantiodifferentiation of
2-heptyl esters in purple passion fruits obtained after transfer
from a DB-Wax as achiral precolumn. To confirm the results
and to rule out the transfer of coeluting compounds, MDGC
analyses were repeated for the same extract using DB-5 as a
precolumn of opposite polarity (Figure 2c). Comparison of the
two chromatograms demonstrates that both approaches delivered
the same results and revealed the four 2-heptyl ester homologues
to be present in purple passion fruits nearly exclusively as (R)-
enantiomers.
Comparison of Isolation Techniques. Various techniques
have been shown to be suitable for the isolation of volatiles
from passion fruits (26). In this study SDE was chosen because
preliminary tests demonstrated that this technique provided high
recoveries of the target compounds (on average 30% more than
with LLE). To rule out changes of the naturally occurring
enantiomeric distributions due to the thermal treatment involved
in the SDE approach, the results were compared with those
obtained by isolation of the volatiles by the gentler VHS
technique and by LLE. As shown in Table 3 for the esters

Secondary Alcohols in Passion Fruit
J. Agric. Food Chem., Vol. 55, No. 25, 2007 10343
Figure 2. Stereodifferentiation of 2-heptyl esters on heptakis(2,3-di-O-methyl-6-O-tert-butyldimethylsilyl)-ꢀ-cyclodextrin as chiral stationary phase; for GC
conditions see Multidimensional Gas Chromatography under Materials and Methods: (a) racemic reference compounds; (b) compounds from SDE extract
of purple passion fruits after transfer from DB-Wax; (c) compounds from SDE extract of purple passion fruits after transfer from DB-5.
Table 3. Enantiomeric Distributions of Secondary Alcohols and Their
ents in purple passion fruits possess (R)-configuration. The
Esters Isolated from Purple Passion Fruits by Different Isolation
Techniques
possibility to investigate the homologous series of individual
esters revealed that this result is also valid for esters containing
2-pentanol and 2-nonanol as alcohol moieties. The sensitivity
of the methodology also revealed that the esters are not optically
enantiomeric distributionsa (%)
SDEb
LLEc
VHSd
pure. The proportions of (S)-enantiomers varied in different
batches and were dependent on the alcohol moieties of the esters.
The opposite configurations of free 2-heptanol in purple and
yellow passion fruits could be confirmed. In addition, a similar
phenomenon could be demonstrated for 2-pentanol.
compound
2-pentanol
2-heptanol
2-heptyl butanoate
2-heptyl hexanoate
R
S
R
S
R
S
60.2
84.6
99.4
99.8
39.8
15.4
0.6
58.6
85.1
99.4
99.8
41.4
14.9
0.6
59.2
84.2
99.4
99.8
40.8
15.8
0.6
0.2
0.2
0.2
The 2-alkyl esters may be biosynthesized from the corre-
sponding alkan-2-ols by the action of alcohol acyltransferases
(27). At first glance, the high concentrations of nearly optically
pure (R)-2-alkyl esters in purple passion fruits suggest pro-
nounced (R)-enantioselectivities of these enzymes. On the other
hand, the only low amounts of (S)-configured free alkan-2-ols
indicate that such esterifications may not proceed via “in vivo
kinetic resolutions”, that is, nearly selective esterification of one
enantiomer and accumulation of the antipode as nonused
substrate. The data suggest that the alkan-2-ols may already be
present mainly as (R)-enantiomers. Stereospecific reductions of
the corresponding alkan-2-ones, which have also been described
a Triplicate analyses of enantiomers. b Simultaneous distillation-extraction.
c Liquid–liquid extraction. d Vacuum headspace.
2-heptyl butanoate and 2-heptyl hexanoate and for the secondary
alcohols 2-pentanol and 2-heptanol, the isolation procedure
applied had no significant impact on the enantiomeric ratios.
Biogenetic Aspects. Investigations of the enantiomeric
compositions via MDGC confirmed the predictions previously
made on the basis of GC separations of diastereoisomeric
derivatives of 2-heptanol (11): 2-Heptyl acetate, butanoate,
hexanoate, and octanoate contained as typical volatile constitu-

10344 J. Agric. Food Chem., Vol. 55, No. 25, 2007
Strohalm et al.
(13) Schultz, T. H.; Flath, R. A.; Mon, T. R.; Eggling, S. B.; Teranishi,
R. Isolation of volatile components from a model system. J. Agric.
Food Chem. 1977, 25, 446–449.
(14) Gattermann-Wieland Die Praxis des organischen Chemikers Teil
1. Allgemeine Arbeitsanweisungen, 42nd ed.; Wieland, T., Mayer,
C., Eds.; de Gruyter Verlag: Berlin, Germany, 1972; pp 71–73.
(15) Hädrich, J.; Vogelgesang, J. Konzept 2000 - Ein statistischer
Ansatz für die analytische Praxis. Teil 1: Nachweisgrenze,
Erfassungsgrenze und Bestimmungsgrenze (1. Mitteilung). Dtsch.
Lebensm. Rundsch. 1999, 95, 428–436.
(16) Hädrich, J.; Vogelgesang, J. Konzept 2000 - Ein statistischer
Ansatz für die analytische Praxis. 2. Mitteilung: Berechnungs-
beispiele. Dtsch. Lebensm. Rundsch. 1999, 95, 495–499.
(17) Sulzbach, H. Controlling Chromatograph Gas Stream by Adjusting
RelatiVe Positioning off Supply Columns in Connector Leading
to Detector; Carlo Erba Strumentazione GmbH: Germany, 1991;DE
4017909.
(18) Sulzbach, H. Multidimensionale Kapillarsäulen-Gaschromato-
graphie. GIT Fachz. Lab. 1996, 40, 131.
(19) Schmarr, H.-G. Beiträge zur On-line LC-GC Kopplung und
modifizierte Cyclodextrine als chirale stationäre Phasen in der
Kapillar-GC. Dissertation, Johann Wolfgang Goethe Universität,
Frankfurt/Main, Germany, 1992.
(20) Maas, B.; Dietrich, A.; Bartschat, D.; Mosandl, A. tert-Butyldi-
methylsilylated cyclodextrins: versatile chiral stationary phases
in capillary gas chromatography. J. Chromatogr. Sci. 1995, 33,
223–228.
as passion fruit constituents (1, 2), might be one pathway leading
to optically enriched secondary alcohols (11).
In yellow passion fruits the obtained data sets on quantities
and enantiomeric compositions of secondary alcohols and the
esters detected as minor constituents suggest low activities of
acyl transferases exhibiting preference for (R)-alkan-2-ols and
starting from nearly racemic mixtures of the secondary alcohols
as substrates.
The data elaborated indicate significant metabolic differences
between passion fruits and bananas, another tropical fruit that
contains secondary alcohols and the corresponding esters as
typical volatile constituents. In bananas, consistently high
excesses of the (S)-enantiomers of 2-pentanol and 2-heptanol
and their acetates and butanoates have been described (21, 22,
28, 29). According to the reported sensory evaluations, the (S)-
esters of secondary alcohols exhibit characteristic fruity, estery
notes (30).
At this point assignments of the actual roles and the
stereospecificities of enzymes involved in the biosynthesis of
secondary alcohols and their esters remain speculative. However,
the data elaborated in this study provide additional valuable
details, and the analytical approach presented may serve as a
useful tool assisting in the elucidation of pathways by molecular
biological investigations.
LITERATURE CITED
(21) Mosandl, A.; Fischer, K.; Hener, U.; Kreis, P.; Rettinger, K.;
Schubert, V.; Schmarr, H.-G. Stereoisomeric flavor compounds.
48. Chirospecific analysis of natural flavors and essential oils using
multidimensional gas chromatography. J. Agric. Food Chem.
1991, 39, 1131–1134.
(22) Bernreuther, A. Analytik chiraler Aromastoffe anhand multidi-
mensionaler Gaschromatographie mit chiralen Phasen. Disserta-
tion, Julius-Maximilians-Universität, Würzburg, Germany, 1992.
(23) Takahisa, E.; Engel, K.-H. 2,3-Di-O-methoxymethyl-6-O-tert-
butyldimethylsilyl-ꢀ-cyclodextrin, a useful stationary phase for
gas chromatographic separation of enantiomers. J. Chromatogr.,
A 2005, 1076, 148–154.
(24) Weber, B.; Dietrich, A.; Maas, B.; Marx, A.; Olk, J.; Mosandl,
A. Stereoisomeric flavour compounds. LXVI. Enantiomeric
distribution of the chiral sulphur-containing alcohols in yellow
and purple passion fruits. Z. Lebensm. Unters. Forsch. 1994, 199,
48–50.
(25) Lebensmittelchemische Gesellschaft (GDCh). Enantioselektive
Analyse zur Natürlichkeitsbewertung von Aromastoffen. Lebens-
mittelchemie 2001, 55, 24–26.
(26) Brat, P.; Brillouet, J.-M.; Reynes, M.; Cogat, P.-O.; Ollé, D. Free
volatile components of passion fruit puree obtained by flash
vacuum-expansion. J. Agric. Food Chem. 2000, 48, 6210–6214.
(27) Sanz, C.; Olias, J. M.; Perez, A. G. Aroma biochemistry of fruits
and vegetables. In Phytochemistry of Fruit and Vegetables;
Tomás-Barberán, F. A., Robins, R. J., Eds.; Oxford University
Press: New York, 1997; pp 125–155.
(28) Fröhlich, O.; Huffer, M.; Schreier, P. Chirality evaluation of
2-heptanol and Z-4-hepten-2-ol from banana fruits using multi-
dimensional gas chromatography. Z. Naturforsch. 1989, 44C, 555–
558.
(29) Schubert, V.; Diener, R.; Mosandl, A. Enantioselective multidi-
mensional gas chromatography of some secondary alcohols and
their acetates from banana. Z. Naturforsch. 1991, 46C, 33–36.
(30) Mosandl, A.; Deger, W. Stereoisomeric flavor compounds. XVII.
Chiral carboxylic esters - synthesis and properties. Z. Lebensm.
Unters. Forsch. 1987, 185, 379–382.
(1) Whitfield, F. B.; Last, J. H. The flavour of the passionfruit - a
review. In Progress in Essential Oil Research; Brunke, E.-J., Ed.;
de Gruyter Verlag: Berlin, Germany, 1986; pp 3–48.
(2) Shibamoto, T.; Tang, C. S. Minor tropical fruits - mango, papaya,
passion fruit, and guava. In Food FlaVours. Part C; Morton, I. D.,
MacLeod, A. J., Eds.; Elsevier: Amsterdam, The Netherlands,
1990; pp 221–280.
(3) Winter, M.; Furrer, A.; Willhalm, B.; Thommen, W. Identification
and synthesis of two new organic sulfur compounds from the
yellow passion fruit (Passiflora edulis f. flaVicarpa). HelV. Chim.
Acta 1976, 59 (5), 1613–1620.
(4) Engel, K.-H.; Tressl, R. Identification of new sulfur-containing
volatiles in yellow passion fruits (Passiflora edulis f. flaVicarpa).
J. Agric. Food Chem. 1991, 39, 2249–2252.
(5) Werkhoff, P.; Güntert, M.; Krammer, G.; Sommer, H.; Kaulen,
J. Vacuum headspace method in aroma research: flavor chemistry
of yellow passion fruits. J. Agric. Food Chem. 1998, 46, 1076–
1093.
(6) Murray, K. E.; Shipton, J.; Whitfield, F. B. The chemistry of food
flavour; I. Volatile constituents of passionfruit, Passiflora edulis.
Aust. J. Chem. 1972, 25, 1921–1933.
(7) Whitfield, F. B.; Sugowdz, G.; Casimir, D. J. (6E,2′E) and
(6Z,2′E)-6-(but-2′-enylidene)-1,5,5-trimethyl-cyclohex-1-ene: im-
portant volatile constituents of the juice of the purple passion fruit
(Passiflora edulis Sims). Chem. Ind. (London) 1977, 12, 502–
503.
(8) Jorda´n, M. J.; Goodner, K. L.; Shaw, P. E. 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. Agric. Food Chem. 2002, 50, 15.
(9) Engel, K.-H.; Tressl, R. Differentiation of yellow and purple
passion fruits by investigation of their ester composition. Chem.
Mikrobiol. Technol. Lebensm. 1983, 8, 33–39.
(10) Chen, C.-C.; Kuo, M.-C.; Hwang, L. S.; Wu, J. S.-B.; Wu, C.-M.
Headspace components of passion fruit juice. J. Agric. Food
Chem. 1982, 30, 1211–1215.
(11) Tressl, R.; Engel, K.-H. Biogenesis of chiral aroma compounds
and their analytical characterization at trace level. In Progress in
FlaVour Research; Adda, J., Ed.; Elsevier: Amsterdam, The
Netherlands, 1984; pp 441–454.
(12) Schurig, V. Separation of enantiomers by gas chromatography.
J. Chromatogr, A 2001, 906, 275–299.
Received for review August 16, 2007. Revised manuscript received
October 10, 2007. Accepted October 14, 2007.
JF072464N