Enantiomeric purity and odor characteristics of 2- and 3-acetoxy-1,8- cineoles in the rhizomes of Alpinia galanga Willd.

Article abstract of DOI:10.1021/jf9807465

(S)-(+)-O-methylmandelate esters of trans- and cis-1,3,3-trimethyl-2- oxabicyclo[2.2.2]octan-5- and 6-ols (2- and 3-hydroxy-1,8-cineoles) were prepared, and eight diastereomers were separated. The absolute configuration of the asymmetric carbons of the cineole moiety of each diastereomer was determined by ¹H NMR data according to the Mosher theory. Each mandelate was reduced with LiAlH₄ to obtain optically pure hydroxy-1,8-cineoles, this being followed by acetylation to afford optically pure acetoxy-1,8-cineoles. These acetates were subjected to chiral GC, using a cyclodextrin column, and the enantiomeric purity of trans- and cis-1,3,3-trimethyl-2- oxabicyclo[2.2.2]octan-5-and 6-yl acetates in the aroma concentrate from the rhizomes of Alpinia galanga was determined as 93.9 (5S), 19.4 (5R), 63.5 (6R), and 100 (6R) % ee, respectively. The aroma character of each enantiomer was also evaluated by GC-sniffing.

Full text of DOI:10.1021/jf9807465

J. Agric. Food Chem. 1999, 47, 685−689
685
En a n tiom er ic P u r ity a n d Od or Ch a r a cter istics of 2- a n d
3-Acetoxy-1,8-cin eoles in th e Rh izom es of Alpin ia ga la n ga Willd .
Kikue Kubota,*,† Yuki Someya,† Reiko Yoshida,† Akio Kobayashi,† Tetsu-ichiro Morita,‡ and
Hiroyuki Koshino‡
Laboratory of Food Chemistry, Ochanomizu University, 2-1-1 Ohtsuka, Bunkyo-ku, Tokyo 112-8610, J apan,
and The Institute of Physical and Chemical Research (RIKEN), Wako, Saitama 351-0198, J apan
(S)-(+)-O-methylmandelate esters of trans- and cis-1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-5- and
6-ols (2- and 3-hydroxy-1,8-cineoles) were prepared, and eight diastereomers were separated. The
absolute configuration of the asymmetric carbons of the cineole moiety of each diastereomer was
1
determined by H NMR data according to the Mosher theory. Each mandelate was reduced with
LiAlH4 to obtain optically pure hydroxy-1,8-cineoles, this being followed by acetylation to afford
optically pure acetoxy-1,8-cineoles. These acetates were subjected to chiral GC, using a cyclodextrin
column, and the enantiomeric purity of trans- and cis-1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-5-
and 6-yl acetates in the aroma concentrate from the rhizomes of Alpinia galanga was determined
as 93.9 (5S), 19.4 (5R), 63.5 (6R), and 100 (6R) % ee, respectively. The aroma character of each
enantiomer was also evaluated by GC-sniffing.
Keyw or d s: Alpinia galanga W.; rhizome; enantiomeric purity; odor; acetoxy-1,8-cineole; 1,3,3-
trimethyl-2-oxabicyclo[2.2.2]octan-6-yl acetate; 1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-5-yl acetate;
O-methylmandelate
INTRODUCTION
aroma constituents (Ko¨nig et al., 1992; Wang et al.,
1996). In this study, we prepared the optically pure
acetoxy-1,8-cineoles and the enantiomeric purity of
trans- and cis-2- and 3-acetoxy-1,8-cineoles (1a -4a ) in
the rhizomes of A. galanga was examined by chiral GC-
MS analyses. The odor characteristics of each enanti-
omer were also evaluated by sniffing the GC eluate.
The rhizomes of greater galangal (Alpinia galanga
Willd.) exhibit a woody, floral, and spicy note which is
different from ginger, and they are widely used as a
spice throughout southeast Asia. We have described in
our previous paper that four isomers, trans- and cis-2-
acetoxy-1,8-cineoles ((1SR,4RS,6SR)- and (1SR,4RS,6RS)-
1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-6-yl acetate, 1a
and 2a ) and trans- and cis-3-acetoxy-1,8-cineoles ((1SR,
4RS,5RS)- and (1SR,4RS,5SR)-1,3,3-trimethyl-2-oxa-
bicyclo[2.2.2]octan-5-yl acetate, 3a and 4a ), were newly
identified positively as natural aroma components from
the rhizomes, and they presented individually interest-
ing odors (Kubota et al., 1998). Although they incorpo-
rated three asymmetric carbons in the molecule, two of
them were bridgehead carbons; that is, each isomer of
the acetoxycineoles had two optical isomers associated
with the carbon bearing the acetoxy group. It was
observed by gas chromatographic analysis with a chiral
column that each acetoxy-1,8-cineole in A. galanga could
be separated into the optical isomers in a specific ratio,
although the absolute configuration of each peak was
not determined. It is important to investigate the
enantiomeric purity of chiral aroma components, be-
cause the flavor character or biological activity of some
compounds can often be different between enantiomers
(Haring et al., 1972; Acree et al., 1985; Mosandl et al.,
1986). Recently, the direct enantiomeric separation
technique by capillary gas chromatography using some
modified cyclodextrins as the chiral stationary phase
has been noted to examine the enantiomeric purity of
EXPERIMENTAL PROCEDURES
Ma ter ia ls. The oxygenated volatile fraction of the rhizomes
of A. galanga was prepared in the same way as that described
in the previous paper (Kubota et al., 1998). trans- and cis-2-
hydroxy-1,8-cineoles (1 and 2) and trans- and cis-3-hydroxy-
1,8-cineoles (3 and 4) were also synthesized as described in
the previous paper (Kubota et al., 1998).
Rea gen ts. (S)-(+)-O-Methylmandelic acid (Sigma-Aldrich,
Tokyo, J apan) and 4-(dimethylamino)-pyridine (DMAP; Merck
KGaA, Darmstadt, Germany) were used.
P r oced u r e for P r ep a r in g th e O-Meth ylm a n d ela te Es-
ter s. Oxalyl dichloride (39.7 mmol) was added to a solution of
(S)-(+)-O-methylmandelic acid (1.00 g, 6.02 mmol) in dry
benzene (10 mL), and the resulting mixture was stirred at
room temperature for 4 h. The benzene and excess oxalyl
chloride were removed in vacuo, and the obtained acid chloride
was dissolved in 20 mL of dried dichloromethane. A solution
of hydroxy-1,8-cineole (13.2 mmol) in dry pyridine was added,
followed by the addition of DMAP (5.0 g) to give a white
suspension. After being stirred overnight, the reaction mixture
was diluted with 40 mL of diethyl ether, and the white
precipitate was filtered off. The organic phase was successively
washed with saturated aqueous cupric sulfate and water and
dried over sodium sulfate, and the solvent was removed in
vacuo to give a yellow oil. The oil was subjected to HPLC
analysis to separate each diastereomer. cis-2- and trans-3-
hydroxy-1,8-cineole were esterified as a mixture before separa-
tion into the four kinds of O-methylmandelate by HPLC.
P u r ifica tion by HP LC. The reaction mixture of the
diastereomers of (S)-O-methylmandelate from the foregoing
* Author to whom correspondence should be addressed (fax
+81-3-5978-5759; e-mail kubota@comfort.eng.ocha.ac.jp).
† Ochanomizu University.
‡ The Institute of Physical and Chemical Research.
10.1021/jf9807465 CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/05/1999

686 J. Agric. Food Chem., Vol. 47, No. 2, 1999
Kubota et al.
Ta ble 1. 1H NMR Da ta for P r oton s of th e Hyd r oxy-1,8-cin eole Moiety of (S)-O-Meth ylm a n d ela te Ester s 1S,R-4S,R (600
MHz, CDCl3, TMS, δ Va lu e)a
H
1S
1R
2S
2R
4
5a
5b
6a
1.51 (m)
2.59 (m)
1.31 (m)
4.72 (m)
1.41 (m)
2.48 (m)
0.93 (m)
4.71 (m)
1.49 (m)
1.88 (m)
2.12 (m)
1.38 (m)
1.48 (m)
1.93 (m)
6b
7a
7b
8a
4.61 (dd, 2.9, 10.2)
1.69 (m)
1.41 (m)
1.98 (m)
1.41 (m)
0.73 (s)
1.26 (s)
1.25 (s)
4.73 (dd, 2.9, 10.0)
1.73 (m)
1.48 (m)
1.96 (m)
1.34 (m)
1.06 (s)
0.95 (s)
1.20 (s)
1.65 (m)
1.47 (m)
1.95 (m)
1.47 (m)
0.65 (s)
1.18 (s)
1.25 (s)
1.75 (m)
1.54 (m)
1.88 (m)
1.21 (m)
1.00 (s)
1.16 (s)
1.24 (s)
8b
9Me
10Me
11Me
H
3S
3R
4S
4R
4
5a
1.59 (m)
5.36 (m)
1.73 (m)
5.34 (m)
1.53 (m)
1.71 (ddd, 2.0, 2.0, 3.9)
5b
6a
6b
7a
7b
8a
8b
5.05 (ddd, 2.4, 6.0, 10.8)
1.71 (m)
2.09 (dd, 10.7, 14.2)
1.57 (m)
1.39 (m)
2.00 (m)
1.46 (m)
1.09 (s)
0.96 (s)
1.12 (s)
5.01 (ddd, 2.0, 5.9, 10.7)
1.48 (ddd, 3.4, 5.9, 14.2)
2.03 (dd, 10.7, 14.2)
1.58 (m)
1.38 (m)
2.06 (m)
1.47 (m)
1.05 (s)
1.18 (s)
1.20 (s)
2.16 (m)
1.37 (m)
1.59 (m)
1.37 (m)
1.59 (m)
1.47 (m)
1.05 (s)
1.21 (s)
1.22 (s)
2.10 (m)
1.12 (dd, 2.7, 14.9)
1.59 (m)
1.29 (m)
1.78 (m)
1.78 (m)
0.99 (s)
1.26 (m)
1.24 (s)
9Me
10Me
11Me
a
1S (1R,4R,6S) isomer; 1R (1R,4S,6R) isomer; 2S (1R,4S,6S) isomer; 2R (1S,4R,6R) isomer; 3S (1R,4S,5S) isomer; 3R (1S,4R,5R) isomer;
4S (1S,4R,5S) isomer; 4R (1R,4S,5R) isomer. The structures should be referred to Figure 2.
esterification was subjected to preparative HPLC to separate
and trap each diastereomer. The HPLC conditions were as
follows: instrument, J asco 880-PU pump equipped with a
Senshupak Pegasil silica 120-5 column (20 φ × 250 mm) and
a UV-970 UV/vis detector (230 nm); eluting solvent, 1:99
2-propanol/hexane. After the solvent had been removed in
vacuo, each mandelate was subjected to NMR analyses.
into the GC instrument equipped with a DB-WAX column (0.25
mm × 60 m, J &W Scientific) held at 60 °C for 4 min and then
increased at 2 °C/min to 200 °C, because the amount of sample
injected was limited in cyclodextrin column. The elution time
was confirmed in coincidence with that of synthesized authen-
tic compound. The temperature of injection and detector port
was 200 °C. The end of the column divided to feed the FID
and ODO-1 sniffing adapter (GL Sciences, Tokyo, J apan) at
1:2.
1
NMR Sp ectr a l Mea su r em en ts. H and 13C NMR spectra
of the O-methylmandelates of hydroxycineoles were recorded
with a J EOL J NM A-600 spectrometer. Chemical shift data
are expressed as δ values in relation to tetramethylsilane
(TMS) as an internal standard. Chemical shift assignments
were established through HMBC, HMQC, and NOEDF experi-
ments.
P r ep a r a tion of Acetoxy-1,8-cin eoles. An optically pure
alcohol was obtained by reducing the ester with LiAlH4 in dry
ether. Without purification, the obtained alcohol was acety-
lated with acetic anhydride in dry pyridine and subjected to
GC-MS analysis.
Ch ir a l GC-MS a n d GC-Sn iffin g. Using synthesized trans-
and cis-2- and 3-acetoxy-1,8-cineoles (racemate), the gas
chromatographic conditions were examined to separate each
enantiomer clearly. The optical isomers of trans- and cis-2-
and cis-3-acetoxy-1,8-cineoles were separated using a CP-
Cyclodextrin-â-236-M-19 column (0.25 mm × 50 m; Chrompack,
Middelburg, The Netherlands), and Chiraldex G-TA column
(0.25 mm × 30 m; Tokyo Kasei Kogyo Co., Tokyo, J apan) gave
a good separation for trans-3-acetoxy-1,8-cineole on GC-MS.
GC-MS was recorded on a Hewlett-Packard 5972 instrument
under an ionization voltage of 70 eV, and the gas chromatic
conditions for both columns were as follows: Hewlett-Packard
5890 series II gas chromatograph instrument; He as the carrier
gas at 0.7 mL/min; oven temperature at 130 °C; injection port
and detector temperatures at 140 and 150 °C, respectively.
The structure was confirmed by comparing the mass spectrum
and the retention time with racemic acetoxy-1,8-cineoles
synthesized. The elution order of each enantiomer on GC-MS
was determined by cochromatography with synthesized opti-
cally pure acetoxycineole and the racemate. The aroma
characteristics of each isomer were evaluated by sniffing the
eluate from the GC column. In this case, the reaction product
of each optically pure acetate already mentioned was injected
RESULTS AND DISCUSSION
P u r ifica tion a n d NMR An a lysis of th e Ester s of
2- a n d 3-Hyd r oxy-1,8-cin eole w ith (S)-(+)-O-Meth -
ylm a n d elic a cid . The reaction mixtures of the (S)-O-
methylmandelate esters from 2- and 3-hydroxy-1,8-
cineoles were subjected to preparative HPLC, respect-
ively. The ester fraction from trans-2-hydroxycineole
gave two peaks at 17.3 and 19.0 min, and 1S and 1R
were eluted in that order. The esters from the mixture
of cis-2- and trans-3-hydroxy-1,8-cineoles were sepa-
rated to four peaks at 21.0, 22.1, 23.8, and 27.7 min on
HPLC. The elution order was 2S, 3R, 3S, and 2R. The
esters from cis-3-hydroxy-1,8-cineole also gave two
peaks at 18.9 and 21.0 min, and 4R and 4S were eluted
in that order. Each peak was trapped and purified by
rechromatography, and each O-methylmandelate ester
separated was subjected to NMR analyses. The signals
were assigned by referring to the data of HMBC,
HMQC, and NOEDF and are summarized in Tables 1
and 2. In this paper, the common names and the
systematic names for hydroxy- and acetoxy-1,8-cineoles
were used and the numbering of the position in the
structure corresponds to the systematic name. The
relationship to each other is shown in the legend of
Figure 1. Trost et al. (1986) have confirmed that the
model proposed by Dale and Mosher (1973) for correlat-
ing NMR shifts with absolute stereochemistry does
extend to the O-methylmandelates. It is known that the
substituent which eclipses the phenyl ring of the

Purity of Acetoxy-1,8-cineoles in A. galanga
J. Agric. Food Chem., Vol. 47, No. 2, 1999 687
Ta ble 2. 13C NMR Da ta for Ca r bon s of th e
Hyd r oxy-1,8-cin eole Moiety of (S)-O-Meth ylm a n d ela te
Ester s 1S,R-4S,R (150 MHz, CDCl3, TMS, δ Va lu e)
C
1S
1R
2S
2R
3S
3R
4S
4R
1
3
4
5
6
7
8
9
10
11
70.8
73.7
33.7
32.4
73.3
25.8
21.9
23.6
28.8
28.5
70.7
73.7
33.6
32.2
73.5
26.0
21.7
24.2
28.8
28.5
70.8
73.7
33.2
32.3
73.9
29.6
21.7
22.5
27.9
28.8
70.6
73.6
33.0
31.7
73.5
29.5
21.7
23.0
27.5
28.8
70.3
73.2
36.9
70.4
39.7
30.7
14.4
26.9
28.6
28.4
70.4
73.2
37.1
70.4
39.7
30.6
14.8
26.8
28.7
28.4
69.9
72.9
37.4
73.5
40.1
30.0
21.0
26.7
30.1
30.0
69.9
72.9
37.5
73.7
39.9
30.0
21.0
26.6
30.4
30.1
a
The structures should be referred to Figure 2.
F igu r e 2. Structures of (S)-(+)-O-methylmandelate esters of
2- and 3-hydroxy-1,8-cineoles. For the numbering, the legend
of Figure 1 should be referred to. Absolute configurations of
each compound are as follows: 1S (1S,4R,6S); 1R (1R,4S,6R),
2S (1R,4S,6S); 2R (1S,4R,6R); 3S (1R,4S,5S); 3R (1S,4R,5R);
4S (1S,4R,5S); 4R (1R,4S,5R).
F igu r e 1. Structures of 2- and 3-acetoxy-1,8-cineoles. The
common names and the systematic names are as follows: 1a ,
trans-2-acetoxy-1,8-cineole (trans-1,3,3-trimethyl-2-oxabicyclo-
[2.2.2]oct-6-yl acetate); 2a , cis-2-acetoxy-1,8-cineole (cis-1,3,3-
trimethyl-2-oxabicyclo[2.2.2]oct-6-yl acetate); 3a , trans-3-
acetoxy-1,8-cineole (trans-1,3,3-trimethyl-2-oxabicyclo[2.2.2]oct-
5-yl acetate); 4a , cis-3-acetoxy-1,8-cineole (cis-1,3,3-trimethyl-
2-oxabicyclo[2.2.2]oct-5-yl acetate). Each acetoxy-1,8-cineole
has two optical isomers, and one of them is shown here. The
absolute configuration is indicated in parentheses. The num-
bering corresponds to the systematic name.
of C-10 and -11 enhanced Ha of C-8 and CH of C-4 and
C-5 by NOEDF, which showed that Hb was eclipsed by
phenyl ring more than Ha in C-8. Actually in NMR, the
shielding by the phenyl ring at CH of C-4 and at Ha
and Hb of C-8 was observed and the shifted value of
Hb of C-8 was highest in 3S. On the other hand, in 3R,
CH3 of C-9 and Hb of C-6 and C-7 were shielded. Among
them Hb of C-6 was shifted most to upfield, and it
agreed with the result of NOESY, in which Hb was not
enhanced with the irradiation of the methyne proton of
C-5. The structures of 3S and 3R were thus established
to be those of the (S)-O-methylmandelate of (1R,4S,5S)-
and (1S,4R,5R)-1,3,3-trimethyl-2-oxabicyclo[2.2.2]octan-
5-ol, respectively, as shown in Figure 2. Similarly, it was
observed that CH of C-4 and the methyl protons of both
of C-10 and -11 in 4S and Ha of C-6 in 4R were shielded
by the phenyl ring. The NOEDF experiments supported
these results well, where CH of C-4 and Ha of C-6 were
enhanced by the irradiation of CH3 of C-10 in 4S and
Ha of C-6 and CH of C-4 were enhanced by the
irradiation of methyl protons of C-10 and -11 in 4R.
Therefore, 4S and 4R were determined to be the (S)-O-
methylmandelate of (1S,4R,5S)- and (1R,4S,5R)-1,3,3-
trimethyl-2-oxabicyclo[2.2.2]octan-5-ol, respectively. In
the 13C NMR spectra shown in Table 2, there was little
variation in chemical shift between the esters of each
enantiomeric alcohol, but some difference of the profile
was observed between cis-trans isomers or positional
isomers of the alcohol moiety.
extended Newman projection in which the intervening
ester linkage is omitted is always upfield as a result of
its shielding by the phenyl ring. As shown in Figure 2,
with compound 1S, this group involves the methyl
protons of C-9 and methylene protons of C-7, and in the
enantiomeric alcohol which corresponds to diastereomer
1R, this group incorporates Hb of C-5, with Hb of C-8
also being shielded. In the NOEDF experiment of 1R,
the enhancement of CH of C-4 and -6 and Ha of C-5
was observed by the irradiation of methyl proton of
C-10. As is shown in Table 1, the δ values of the
corresponding protons are shifted to upfield which are
in good agreement with this theoretical model. From
these results, the absolute structures of 1S and 1R were
determined to be those of the (S)-O-methylmandelate
of (1S,4R,6S)- and (1R,4S,6R)-1,3,3-trimethyl-2-oxa-
bicyclo[2.2.2]octan-6-ol, respectively. Similarly in 2S, the
CH3 protons of C-9 were shielded. In 2R, the CH3
protons of C-10, CH of C-4, and Ha and Hb of C-5 were
shielded clearly. The enhancement of Ha of C-5 and -8
and the methyne proton of C-4 in 2R was observed by
the irradiation of methyl protons of C-10 by NOEDF.
These results explained that Ha of C-5 and the methyl
protons of C-10 were in the same direction which lay
close to phenyl ring in space and their δ values were
shifted more than others in 2R. The structures of 2S and
2R were thus established to be those of the (S)-O-
methylmandelate of (1R,4S,6S)- and (1S,4R,6R)-1,3,3-
trimethyl-2-oxabicyclo[2.2.2]octan-6-ol, respectively. The
Mosher theory was also applied to the 3-hydroxy-1,8-
cineole group. In 3S, the irradiation of methyl protons
Deter m in a tion of th e En a n tiom er ic P u r ity. Each
mandelate ester was reduced with LiAlH4 to obtain the
corresponding optically pure alcohol. After extraction of
the alcohol without additional purification, the alcohol
was acetylated with acetic anhydride in dry pyridine
and submitted to GC-MS analysis. The enantiomers

688 J. Agric. Food Chem., Vol. 47, No. 2, 1999
Kubota et al.
Ta ble 3. Con cen tr a tion of th e En a n tiom er s of Acetoxy-1,8-cin eoles in th e Rh izom es of A. ga la n ga a n d Th eir Od or
Ch a r a cter istics
GCb
tR (min)
peakc
area (%)
compda
configuration
% ee
63.5
odor impressiond
1a
1S,4R,6S
1R,4S,6R
1R,4S,6S
1S,4R,6R
1R,4S,5S
1S,4R,5R
1S,4R,5S
1R,4S,5R
21.58
22.63
26.26
25.00
14.10e
13.80e
24.14
23.78
18
82
0
100
97
3
woody (weak)
woody, A. galanga-like
fruity, sweet
2a
3a
4a
100.0
93.9
weak
sweet floral (weak)
sweet floral (weak)
camphoraceous
mild woody
40
60
19.4
a
The structures should be referred to Figure 1. The numbering of the configurations corresponds to the systematic name. b,eRetention
times by GC-MS when using the CP-Cyclodextrin-â-236-M-19 and Chiraldex G-TA columns, respectively. The conditions are defined in
d
the Experimental Procedures section. c Peak area was calculated by total ion count chromatography. Evaluated by GC-sniffing.
present in a much larger amount than the enantiomer
at 63.5% ee. On the other hand, cis-3 (4a ) was present
as almost a racemate at 19.4% ee.
It was noted that 1a and 3a were the main compo-
nents of the aroma concentrate and that one of their
enantiomers was in large excess in the rhizomes,
whereas 2a and 4a were each present in very small
amounts.
Od or Ch a r a cter istics a n d Con tr ibu tion of Ac-
etoxy-1,8-cin eoles to th e Od or of A. ga la n ga . Since
there was so little of the acetate isolated from the
reaction mixture, the odor was evaluated by sniffing the
GC column eluate. As shown in Table 3, it was observed
that each enantiomer presented a different odor from
the other except for the 3a pair. Mariani et al. (1995)
have synthesized a series of alkyl esters of hydroxy-1,8-
cineoles and described the odorous note of trans-2-
acetoxy-1,8-cineole to be woody, pine oillike, and vio-
letlike. However, they did not mention anything about
the optical isomers. In the case of 1a , the (1R,4S,6R)
isomer presented a woody and A. galanga-like odor,
which was stronger than that of the enantiomer. In
F igu r e 3. TIC chromatograms of synthesized 2- and 3-ac-
addition, the amount of the stronger enantiomer in the
etoxy-1,8-cineoles (A) and acetoxy-1,8-cineoles in oxygenated
rhizomes was about four times more than the other
volatile fraction from the rhizomes of A. galanga (B) using CP-
(1S,4R,6S) isomer as already described. In the case of
2a , the (1R,4S,6S) isomer, which was not found in the
rhizomes, showed a fruity and sweet odorous note, while
the odor of the enantiomer was weak. So 2a seems not
to have had a large effect on the odor of greater
galangal. 3a was almost optically active in the rhizomes,
but the odors of both enantiomers were too weak to
evaluate. Although it has been selected from an aroma
extract dilution analysis as one of the potent odorants
because its concentration in the aroma concentrate was
highest among 1a -4a (Mori et al., 1995), 3a seems not
to have had much of a qualitative effect on the charac-
teristic flavor. 4a presented a different odor between
each enantiomer, although it might have very little
effect on the odor of A. galanga because of the low
concentration. It was thus concluded from these results
that, among the acetoxy-1,8-cineoles, trans-2-acetoxy-
1,8-cineole (1a ) was the most potent odorant in the
rhizomes of A. galanga both quantitatively and quali-
tatively.
Cyclodextrin-â-236-M-19 and Chiraldex G-TA* columns. See
Experimental Procedures for the conditions. Key: 1, 3a
(racemate); 2, 1a (1S,4R,6S); 3, 1a (1R,4S,6R); 4, 4a (1R,4S,5R);
5, 4a (1S,4R,5S); 6, 2a (1S,4R,6R); 7, 2a (1R,4S,6S); 8, 3a
(1S,4R,5R); 9, 3a (1R,4S,5S). The structures of each compound
should be referred to Figure 1.
could be clearly separated using the modified â-cyclo-
dextrin column (CP-Cyclodextrin-â-236-M-19), except
for trans-3-acetoxy-1,8-cineole (3a ) which was separated
using another cyclodextrin column (γ-cyclodextrin col-
umn, Chiraldex G-TA) as shown in Figure 3. An
optically pure acetate was used to determine the rela-
tionship between the stereostructure and elution order
by GC-MS as shown in Table 3. Similarly, the enantio-
meric ratio of the acetoxy-1,8-cineoles in an oxygenated
compound fraction of the volatile concentrate from the
rhizomes of A. galanga (Kubota et al., 1998) was
investigated under the same conditions by using the two
kinds of cyclodextrin columns just described. The total
ion current (TIC) chromatograms by GC-MS are shown
in Figure 3, and the results are summarized in Table
3. It was found in this research that cis-2-acetoxy-1,8-
cineole (2a ) was optically active for the (1S,4R,6R)
isomer, while trans-3 (3a ), which produced the largest
peak by GC-MS among the four acetates, was also
almost optically active for the (1R,4S,5S) isomer in the
rhizomes. In trans-2 (1a ), the (1R,4S,6R) isomer was
ACKNOWLEDGMENT
We thank Ms. Y. Kurobayashi of the Technical
Research Center of T. Hasegawa Co. Ltd. for technical
assistance in evaluating the odor characteristics.

Purity of Acetoxy-1,8-cineoles in A. galanga
J. Agric. Food Chem., Vol. 47, No. 2, 1999 689
LITERATURE CITED
octan-6-ols: Synthesis, odour evaluation and in vitro pre-
liminary toxicity assays. Int. J . Cosmetic Sci. 1995, 17, 187-
196.
Mori, H.; Kubota, K.; Kobayashi, A. Potent aroma components
of rhizomes from Alpinia galanga Willd. L. Nippon Shoku-
hin Kagaku Kogaku Kaishi 1995, 42, 989-995.
Mosandl, A.; Heusinger, G.; Gessner, M. Analytical and
sensory differentiation of 1-octen-3-ol enantiomers. J . Agric.
Food Chem. 1986, 34, 119-122.
Trost, B. M.; Belletire, J . L.; Godleski, S.; McDougal, P. G.;
Balkovec, J . M.; Baldwin, J .; Chrity, M. E.; Ponticello, G.
S.; Varga, S. L.; Springer, J . P. On the use of the O-
methylmandelate ester for establishment of absolute con-
figuration of secondary alcohols. J . Org. Chem. 1986, 51,
2370-2374.
Wang, D.; Kubota, K.; Kobayashi, A. Optical isomers of methyl
jasmonate in tea aroma. Biosci., Biotechnol. Biochem. 1996,
60, 508-510.
Acree, T. E.; Nishida, R.; Fukami, H. Odor thresholds of the
stereoisomers of methyl jasmonate. J . Agric. Food Chem.
1985, 33, 425-427.
Dale, J . A.; Mosher, H. S. Nuclear magnetic resonance enan-
tiomer reagents. Configurational correlations via nuclear
magnetic resonance chemical shifts of diastereomeric man-
delate, O-methylmandelate, and R-methoxy-R-trifluorom-
ethylphenyl acetate (MTPA) esters. J . Am. Chem. Soc. 1973,
95, 512-519.
Haring, H. G.; Rijkens, F.; Boelens, H.; van der Gen, A.
Olfactory studies on enantiomeric eremophilane sesquiter-
penoids, J . Agric. Food Chem. 1972, 20, 1018-1021.
Ko¨nig, W. A.; Gehrcke, B.; Icheln, D.; Evers, P.; Do¨nnecke, J .;
Wang, W. New, selectively substituted cyclodextrins as
stationary phase for the analysis of chiral constituents of
essential oils. J . High Resolut. Chromatogr. 1992, 15, 367-
372.
Kubota, K.; Nakamura, K.; Kobayashi, A.; Amaike, M. Ac-
etoxy-1,8-cineoles as aroma constituents of Alpinia galanga
Willd. J . Agric. Food Chem. 1998, in press.
Mariani, M.; Neuhoff, C.; Bargagna, A.; Longobardi, M.; Ferro,
M.; Gelardi, A. Esters of 1,3,3-trimethyl-2-oxabicyclo[2.2.2]-
Received for review J uly 8, 1998. Revised manuscript received
November 3, 1998. Accepted November 5, 1998. This work was
partly supported by a grant from the Urakami Foundation.
J F9807465