Identification and formation of volatile components responsible for the characteristic aroma of mat rush (Igusa)

Article abstract of DOI:10.1271/bbb.100053

An aroma concentrate of the mat rush (igusa) was prepared by combining solvent extraction with the solvent-assisted flavor evaporation (SAFE) technique. An aroma extract dilution analysis (AEDA) applied to the volatile fraction revealed 51 odor-active peaks with FD factors between 4³ and 4⁷. Among the perceived odorants, twelve peaks with the higher FD factors (≥4⁶) were proved to be the most important components of the characteristic aroma in mat rush. Eleven odorants were identified or tentatively identified from the twelve peaks as methional, (E, Z)-2, 6-nonadienal, (E)-2-nonenal, (E,E)-2, 4-nonadienal, (E,E,Z)-2,4,6-nonatrienal, trans-4,5-epoxy-(E)-2-decenal, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, 3-hydroxy-4,5-dimethyl-2(5H)-furanone, isovaleric acid, methyl anthranirate, and vanillin. The FD factors of the odor-active peaks in dried mat rush were observed to be much higher than those in raw mat rush. This finding suggests that the drying process during manufacturing of the mat rush Is one of the most important factors for the formation of the characteristic mat rush aroma.

Full text of DOI:10.1271/bbb.100053

Biosci. Biotechnol. Biochem., 74 (6), 1231–1236, 2010
Identification and Formation of Volatile Components Responsible
for the Characteristic Aroma of Mat Rush (Igusa)
Kenji KUMAZAWA,y Nao SAKAI, Hiroko AMMA, Satoshi SAKAMOTO, Masaki KODAMA,
Yoshiyuki WADA, and Osamu NISHIMURA
Ogawa & Company Ltd., 15-7 Chidori, Urayasu, Chiba 279-0032, Japan
Received January 21, 2010; Accepted March 23, 2010; Online Publication, June 7, 2010
[doi:10.1271/bbb.100053]
An aroma concentrate of the mat rush (igusa) was
prepared by combining solvent extraction with the
solvent-assisted flavor evaporation (SAFE) technique.
An aroma extract dilution analysis (AEDA) applied to
the volatile fraction revealed 51 odor-active peaks with
FD factors between 43 and 47. Among the perceived
odorants, twelve peaks with the higher FD factors (ꢀ46)
were proved to be the most important components of the
characteristic aroma in mat rush. Eleven odorants were
identified or tentatively identified from the twelve peaks
as methional, (E,Z)-2,6-nonadienal, (E)-2-nonenal,
(E,E)-2,4-nonadienal, (E,E,Z)-2,4,6-nonatrienal, trans-
4,5-epoxy-(E)-2-decenal, 4-hydroxy-2,5-dimethyl-3(2H)-
furanone, 3-hydroxy-4,5-dimethyl-2(5H)-furanone, iso-
valeric acid, methyl anthranirate, and vanillin. The FD
factors of the odor-active peaks in dried mat rush were
observed to be much higher than those in raw mat rush.
This finding suggests that the drying process during
manufacturing of the mat rush is one of the most
important factors for the formation of the characteristic
mat rush aroma.
authors gave no details about the importance of the
recognized compounds, because the GC analysis was not
coupled with a GC-olfactometry (GC-O) technique such
as an aroma extract dilution analysis (AEDA).6) Many of
the previous investigations chose steam distillation to
prepare the analytical samples, and it is known that this
method makes it difficult to obtain the original aroma of
the analytical samples.7) The actual significance of the
odorants in the mat rush aroma is therefore still mostly
unclear.
The harvested raw mat rush is generally processed
into tatami products, in which the processing steps
involved include soaking in clay mud, which is called
sendo, drying, and weaving by a machine. The raw mat
rush aroma is quite different from that of the dried mat
rush that is used in the tatami products. It can therefore
be expected that the manufacturing process is closely
connected with the formation of the characteristic aroma
of the mat rush.
The objective of the present investigation was to
elucidate the potent odorants in the original mat rush
aroma by AEDA. Furthermore, in order to clarify the
generation of the characteristic aroma of the mat rush
during manufacture, a comparison was made of the dried
(muddied and non-muddied) mat rush and raw mat rush.
Key words: mat rush; igusa; odorant; aroma; aroma
extract dilution analysis (AEDA)
Tatami, which is enshrined in Japanese culture and
grows naturally in Japan’s climate, is a significant part of
such traditions as flower arrangement and the tea
ceremony, and is closely involved with the Japanese
lifestyle. Tatami, which is prepared from igusa (mat rush)
as the main raw material, is the original Japanese flooring
material. Many Japanese consider a room floored with
the tatami as a space providing peace of mind. The tatami
has such characteristics as purification of the air, thermal
insulation, humidity adjustment, acoustic absorption,
elasticity, and aroma,1) and it is said that these character-
istics influence the peace of mind feeling. The character-
istic and comfortable aroma in particular makes the mat
rush extremely popular with the Japanese, and it has
recently been suggested that the mat rush aroma has a
relaxation effect.2) Interest in the odorants producing the
mat rush aroma is therefore high.3–5)
Materials and Methods
Mat rush (igusa). The mad rush (Juncus effusus L. var. decipiems
Buchen.) samples were dried (muddied and non-muddied) and raw.
These samples were produced in Kumamoto Prefecture (Japan) in 2008
and 2009, and presented by the Kumamoto Prefectural Agricultural
Research Center (Kumamoto, Japan).
Chemicals. The following compounds were synthesized according
to the literature procedures: 2-acetyl-1-pyrroline,8) (Z)-1,5-octadien-3-
one,9) 3-mercapto-1-hexanol,10) and cis- and trans-4,5-epoxy-(E)-2-
decenals.11) Compound nos. 1, 7–11, 13, 15, 19–23, 25–28, 31, 33, 37,
43, 45, 46, 49, and 51 were obtained from Tokyo Chemical Industry
Co., (Tokyo, Japan); nos. 2–4, 12, 16, and 41 were obtained from
Sigma-Aldrich Japan (Tokyo, Japan); nos. 18 and 50 were obtained
from Wako Pure Chemical Industries (Osaka, Japan); no. 29 was
obtained from Nihon Firmenich (Tokyo, Japan); and no. 36 was
obtained from Acro¯s Organics (Geel, Belgium) (Table 1).
Investigations to elucidate the odorants in the mat
rush have already been reported,3–5) and many volatile
components have been identified, mainly by gas chro-
matographic (GC) and GC-mass spectrometric (GC-MS)
measurements. However, there were some problems
with these investigations. For instance, many of the
Synthesis of (E,E,E)-2,4,6-nonatrienal. The target compound was
prepared from (E,E)-2,4-heptadienal as the starting material by Horner-
Wadsworth-Emmons olefination, DIBAL reduction and final Dess-
Martin oxidation. Ethyl (E,E,E)-2,4,6-nonatrienoate was first synthe-
sized by using diethylphosphonoacetate. To a solution of ethyl
diethylphosphonoacetate (8.97 g, 40.0 mmol) in THF (100 ml) in an
y
To whom correspondence should be addressed. Fax: +81-47-305-1423; E-mail: kumazawa.kenji@ogawa.net

1232
K. KUMAZAWA et al.
argon atmosphere was added sodium hydride (a 60% suspension in oil,
1.60 g, 40.0 mmol) at ꢀ70 ꢁC, and 30 min later, to this mixture was
added a solution of (E,E)-2,4-heptadienal (purity >95%, 2.53 g,
23.0 mmol; Tokyo Chemical Industry Co., Tokyo, Japan). The mixture
was stirred for 1 h at ꢀ35 ꢁC and then quenched with a saturated
NaHCO3 solution. The organic layer was separated, and the aqueous
layer was extracted with a mixture of n-hexane and ethyl acetate (5/1,
v/v). The combined extracts were washed with brine, dried (Na2SO4),
and concentrated under vacuum. The resulting residue was purified by
column chromatography on silica gel (silica gel 60N; Kanto Chemical
Co., Tokyo, Japan), using 9–3% ethyl acetate in n-hexane, into a
yellow oil of ethyl (E,E,E)-2,4,6-nonatrienoate (3.91 g, 94% yield).
This ethyl (E,E,E)-2,4,6-nonatrienoate was reduced to the correspond-
ing alcohol. To a solution of ethyl (E,E,E)-2,4,6-nonatrienoate (3.78 g,
21.0 mmol) in THF (140 ml) was added DIBAL (63.0 ml of a 1.0 mol/l
n-hexane solution) at 0 ꢁC in an argon atmosphere. After stirring for 1 h
at the same temperature, the reaction was quenched with water. To the
mixture was added a saturated potassium sodium tartrate solution, and
stirring was continued for an additional 30 min. The organic layer was
separated, and the aqueous layer was extracted with diethyl ether. The
extract was washed with brine and dried (Na2SO4), and then the
solvent was removed under vacuum. The residue was purified by
column chromatography (silica gel, 9–50% ethyl acetate in n-hexane)
to give 2.78 g (96% yield) of (E,E,E)-2,4,6-nonatrienol as an
amorphous solid. In the final step, the target compound, (E,E,E)-
2,4,6-nonatrienal, was synthesized from the corresponding alcohol by
selective oxidation with Dess-Martin periodinane. To a solution of
(E,E,E)-2,4,6-nonatrienol (1.37 g, 9.88 mmol) in methylene chloride
(66 ml) was added Dess-Martin periodinane (37 ml of a 15 wt %
methylene chloride solution, 18 mmol) at 0 ꢁC in an argon atmosphere.
After stirring for 1 h, the reaction mixture was quenched with a
saturated Na2CO3 solution and extracted with a mixture of n-hexane
and ethyl acetate (1/1, v/v). The combined extracts were successively
washed with a saturated Na2S2O3 solution and brine, dried (Na2SO4),
and concentrated under vacuum. The residue was purified by column
chromatography (silica gel, 17–50% ethyl acetate in n-hexane) to give
0.92 g (68% yield) of (E,E,E)-2,4,6-nonatrienal as a yellow oil (purity
>98%). This structure was confirmed by mass spectrometry (electron
impact [EI] mode) and nuclear magnetic resonance (1H and 13C). 1H-
NMR characterization of (E,E,E)-2,4,6-nonatrienal afforded data for ꢀ
(multiplicity, coupling constant [in hertz], and relevant H at carbon
[numbering refers to Fig. 1]). MS/EI m=z (%): 136 (10, [Mþ]), 121
(4), 107 (6), 91 (7), 79 (12), 66 (3), 53 (3), 39 (4); 1H-NMR (400 MHz,
CDCl3) ꢀ ppm: 1.05 (3H, t, J ¼ 7:6 Hz, C9), 2.21 (2H, dt, J ¼ 7:6,
7.6 Hz, C8), 6.04–6.22 (3H, m, C2, 6, 7), 6.35 (1H, dd, J ¼ 15:2,
11.2 Hz, C4), 6.65 (1H, dd, J ¼ 15:2, 10.4 Hz, C5), 7.12 (1H, dd,
J ¼ 15:2, 11.2 Hz, C3), 9.57 (1H, d, J ¼ 7:6 Hz, C1); 13C-NMR
(100 MHz, CDCl3) ꢀ ppm: 13.7, 26.7, 128.4, 129.4, 131.3, 143.9,
144.5, 153.0, 194.2.
of mat rush) and concentrated to approximately 2 ml. Part of the
concentrated volatile fraction (150 ml) was applied to a glass column
(10 ꢂ 0:7 cm i.d.) filled with silica gel in n-pentane. Elution was
performed with the following solvents: n-pentane (20 ml, fraction I),
n-pentane/diethyl ether (20 ml, 10 þ 1, v/v, fraction II), n-pentane/
diethyl ether (20 ml, 1 þ 1, v/v, fraction III), and diethyl ether (20 ml,
fraction IV). The solution was concentrated to approximately 150 ml as
already described. The basic and weakly acidic volatiles were isolated
by treating the other part of the concentrated volatile fraction (200 ml)
dissolved in about 10 ml of diethyl ether. The basic volatiles were
extracted with 1 M hydrochloric acid (2 ꢂ 5 ml). The combined acid
extract was then washed with diethyl ether (1 ꢂ 5 ml), and the washed
acid extract was neutralized with aqueous sodium hydroxide and then
extracted with diethyl ether (2 ꢂ 5 ml). This extract was washed with
brine (2 ꢂ 5 ml), dried over anhydrous Na2SO4 and finally concen-
trated to 50 ml (B, the basic fraction). The organic phase, which
removed the basic volatiles, was extracted with 1 M-NaOH (2 ꢂ 5 ml),
after extracting with a saturated solution of NaHCO3 (2 ꢂ 5 ml). The
combined extract with 1 M-NaOH was then washed with diethyl ether
(1 ꢂ 5 ml). The washed 1 M-NaOH extract was neutralized with
hydrochloric acid and then extracted with diethyl ether (2 ꢂ 5 ml).
This extract was washed with brine (2 ꢂ 5 ml), dried over anhydrous
Na2SO4 and finally concentrated to 50 ml (WA, the weakly acidic
fraction).
Gas chromatography-olfactometry (GC-O). An Agilent 6850 series
gas chromatograph equipped with a thermal conductivity detector
(TCD) (Agilent Technologies, Palo Alto, USA) and fused silica
column (30 m ꢂ 0:25 mm i.d., coated with a 0.25 25-mm film of DB-
Wax; J & W Scientific, Folsom, USA) were was used in the splitless
injection mode (splitless time: 1 min). The column temperature was
programmed from 40 ꢁC to 210 ꢁC at the rate of 5 ꢁC/min for all runs.
The respective injector and detector temperatures were 250 ꢁC and
230 ꢁC, helium being used as the carrier gas at the flow rate of 1 ml/
min. A glass sniffing port was connected to the outlet of the TCD and
heated by a ribbon heater, with moist air being pumped into the sniffing
port at about 100 ml/min to quickly remove the odorant from the
sniffing port that had been eluted from the TCD.
Aroma extract dilution analysis (AEDA). The original odor
concentrate of the mat rush was stepwise diluted with methylene
chloride to 4n (n ¼ 3{8), and aliquots (1 ml) of each fraction were
analyzed by capillary GC in a DB-Wax column. The odorants were
then detected by GC eluate sniffing (GC-O). The flavor dilution (FD)
factors of the odorants were determined by AEDA.6) The FD factor of
the original odor concentrate of the raw mat rush was converted into
dry weight and is defined as 41. Before measuring the FD factor, two
panelists repeatedly checked the retention time and odor quality of the
odorants by using each diluted sample (1:16), and then the FD factor of
each odorant was determined by being detected at the dilution step by
both panelists.
Isolation of the volatiles from mat rush. The crushed mat rush
(dried, 25 g; raw, 20.8 g) was added to 1% water that contained diethyl
ether (500 ml), the mixture then being stirred and extracted at room
temperature. The water content of the raw mat rush is generally about
70%,12) so the dry weight of this raw mat rush was equivalent to 1/4 of
the weight of the dried mat rush. After standing for 2 h, the residual
substances were removed by passing through filter paper. The filtrate
(the aroma extract, approximately 350 ml) was dried over anhydrous
Na2SO4, and then concentrated to a volume of approximately 50 ml by
rotary evaporation (35 ꢁC at 550 mm of Hg). To remove the non-
volatile material, the aroma extract from the mat rush was distilled
under reduced pressure (40 ꢁC at 5 ꢂ 10ꢀ3 Pa) by the solvent assisted
flavor evaporation (SAFE) method.13) The distillate was dried over
anhydrous Na2SO4, and the solvent was then removed by rotary
evaporation (35 ꢁC at 550 mm of Hg) to approximately 5 ml. Further
concentration to approximately 150 ml was achieved in a nitrogen
stream. The resulting concentrate was used as the sample for the
AEDA and GC-MS analyses.
Gas chromatography-mass spectrometry (GC-MS). An Agilent
7890 N gas chromatograph coupled to an Agilent 5975C inert XL
series mass spectrometer (Agilent Technologies, Palo Alto, USA) was
used. The column was a 60 m ꢂ 0:25 mm i.d. DB-Wax fused silica
capillary type (J & W Scientific, Folsom, USA) with a film thickness of
0.25 mm. The column temperature was programmed from 80 ꢁC to
210 ꢁC or from 40 ꢁC to 210 ꢁC at the rate of 3 ꢁC/min. The injector
temperature was 250 ꢁC, and the flow rate of the helium carrier gas was
1 ml/min. An injection volume of 1 ml or 0.2 ml was applied, using the
split (the split ratio was 1:30) or splitless technique. The mass
spectrometer was used with an ionization voltage of 70 eV (EI) and ion
source temperature of 150 ꢁC.
Identification of the components. Each component was identified by
comparing its Kovats GC retention index (RI), mass spectrum and odor
quality with those of the authentic compound.
Enrichment of the odorants for identification. The identification
experiments were conducted on the mat rush volatiles that had been
isolated from the crushed mat rush by combining solvent extraction
with the SAFE technique as already described. These procedures were
repeated, and all the volatile fractions were combined (total 500 g
Proton and carbon magnetic resonance spectrometry (1H- and 13C-
NMR). The 1H- and 13C-NMR spectra were recorded in a CDCl3
solution with a Bruker AVANCE 400 spectrometer operated at
400 MHz or 100 MHz. Tetramethylsilane was the internal standard for
the 1H-NMR measurements.

Identification of the Potent Odorants in Mat Rush
1233
Table 1. Potent Odorants with FD factors (ꢃ43) in the Mat Rush (igusa)
No.
RIa
Fractionb
Compoundc
Odor qualityh
log4 (FD factor)
1
2
1090
1137
1237
1298
1338
1370
1378
1378
1427
1439
1451
1464
1493
1501
1535
1565
1578
1582
1621
1648
1665
1701
1719
1724
1733
1766
1775
1813
1823
1832
1842
1847
1857
1876
1895
1936
1951
1966
1986
2001
2028
2075
2167
2178
2195
2236
2253
2271
2456
2553
2565
hexanale
green
green
5
3
3
4
4
5
5
4
3
4
6
4
3
4
6
5
6
7
4
3
7
6
3
4
4
3
3
5
3
3
3
3
5
6
4
3
3
3
3
6
7
3
4
3
7
7
3
4
3
4
7
(Z)-3-hexenalde
(Z)-4-heptenale
1-octen-3-onee
3
II
II
B
metallic, hay-like
metallic, mushroom-like
roasty
4
5
2-acetyl-1-pyrrolinee
(Z)-1,5-octadien-3-onede
dimethyl trisulfidee
(Z)-3-hexenole
6
metallic
roasty, putrid
green
7
8
9
II
(E)-2-octenal
acetic acid
methionale
2,6-dimethyl-3-ethylpyrazinee
2,3-diethyl-5-methylpyrazinee
(Z)-2-nonenalef
green, fatty
sour
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
green, potato-like
roasty, nutty
roasty, nutty
green, earthy
sweet, fatty
green, earthy
green, earthy
green
B
B
II
II
II
(E)-2-nonenal
(E,E)-2,6-nonadienale
unknown
(E,Z)-2,6-nonadienale
butyric acid
phenylacetaldehyde
sour
sweet, honey-like
sour
III
III
isovaleric acid
(E,E)-2,4-nonadienale
(Z)-6-nonenole
unknown
sweet, fatty
green
sweet
pentanoic acid
sour
green
III
II
(E,Z)-2,6-nonadienole
methyl salicylatee
(E,E)-2,4-decadienal
ꢁ-damascenone
green, minty
sweet, fatty
sweet, honey-like
spicy, sweet
green
unknown
hexanoic acid
3-mercapto-1-hexanolde
2-methoxyphenol
(E,E,Z)-2,4,6-nonatrienalef
(E,E,E)-2,4,6-nonatrienale
(Z)-3-hexenoic acide
(E)-2-hexenoic acide
unknown
fruity, grapefruit-like
spicy, burnt
sweet
II
II
II
II
sweet
green
green
sweet
III
III
cis-4,5-epoxy-(E)-2-decenaleg
trans-4,5-epoxy-(E)-2-decenaleg
4-hydroxy-2,5-dimethyl-3(2H)-furanonede
unknown
sweet
sweet
sweet, caramel-like
spicy, phenolic
spicy
WA
B
eugenol
unknown
spicy
3-hydroxy-4,5-dimethyl-2(5H)-furanonede
methyl anthranilatee
unknown
sweet, burnt sugar-like
fruity, grape-like
spicy, phenolic
spicy, metallic
sweet
unknown
coumarine
phenylacetic acide
vanillin
III
sweet
sweet, vanilla-like
aRetention index in the DB-Wax column (30 m ꢂ 0:25 mm i.d.; coated with a 0.25-mm film) observed for GC-O.
bFraction in which most of the compound appeared after column chromatography on silica gel of the aroma concentrate (I–III) or after separating in to the basic (B)
and weakly acidic (WA) fractions.
cThe compound was identified by comparing with the reference substance on the basis of the following criteria: retention index (RI) of stationary phases given in the
DB-Wax column, mass spectrum, and odor quality.
dThe MS signals were too weak for unequivocal interpretation. The compound was tentatively identified by comparing with the reference substance on the basis of the
following criteria: retention index (RI) of stationary phases given in the DB-Wax column and odor quality.
eNewly identified compounds in mat rush.
f The cis-trans isomer was tentatively identified by matching its mass spectrum, and retention index of the polar stationary phase with those of literature data.14,15)
gThe stereochemistry was not determined.
hOdor quality assigned during AEDA.
Results and Discussion
crushed mat rush had a favorable polarity range, and
diethyl ether was comparatively better for extraction than
n-hexane or ethyl acetate. However, the aroma intensity
of the diethyl ether extract was weak, and it could not
reproduce the original aroma quality of the mat rush. A
small amount of distilled water was then added to the
Potent odorants in mat rush
The aroma concentrate of the mat rush was prepared
by combining solvent extraction with the SAFE tech-
nique. The solvents used for aroma extraction from the

1234
K. KUMAZAWA et al.
8
6
4
2
CHO
CHO
CHO
9
7
3
1
5
Abundance
400
Abundance
1200
79
136
Abundance
3500
79
79
136
360
320
280
240
200
160
120
80
A
RI = 1867
B(34)
RI = 1878
C (35)
136
1000
800
600
400
200
0
3000
2500
2000
1500
1000
500
91
RI = 1896
107
91
107
107
91
66
121
121
121
66
39
39
66
39
53
53
55
40
0
0
30 40 50 60 70
80 90 100 110 120 130 140
30 40 50 60 70 80 90 100 110 120 130 140
30 40 50 60 70 80 90 100 110 120 130 140
(m/z)
(m/z)
(m/z)
(m/z:136)
31.0
31.2
31.4
31.6
31.8
32.0
32.2
32.4
32.6
32.8
Time (min)
Fig. 1. Mass Chromatogram of the Volatile Concentrate of the Mat Rush Showing the Extracted Ion (m=z: 136), and Mass Spectra of Geometric
Isomers of 2,4,6-Nonatrienal in the Mat Rush.
diethyl ether, and the intensity and quality of the aroma
extract were markedly improved. In particular, the best
aroma extract was provided when distilled water was
added to diethyl ether in the range from 1% to 5%. Based
on this result, the preparation method of for the volatile
fraction was determined for to identification of the potent
odorants in the mat rush was determined. This involved
extracting the aroma extract from the crushed mat rush
with 1% water-containing diethyl ether, and then the
volatile fraction was prepared by a SAFE treatment to
remove the non-volatiles from the aroma extract. The
obtained volatile concentrate well-reproduced the char-
acteristic aroma of the mat rush, and was regarded as the
most preferable analytical sample for screening the
potent odorants by the GC-O technique.
The AEDA technique that was applied to the volatile
concentrate, which had been prepared from the crushed
mat rush, revealed 51 odor-active peaks with FD factors
between 43 and 47, and it was confirmed that the aroma
of the mat rush consisted of many potent odorants
(Table 1). After enriching the volatile fraction by
column chromatography or solvent separation, thirty
eight of these odorants could be identified by the GC-
MS and GC-O analysis. Among the perceived odorants,
two unknown compounds (nos. 34 and 35), which had
similar mass spectra, were assumed to be geometric
isomers of 2,4,6-nonatrienal by comparing the mass
spectra to the literature data.14) The mass chromatogram
of the volatile concentrate of the mat rush in Fig. 1
shows the extracted ion with m=z 136, and at least three
kinds of compound (A, B and C), which gave similar
mass spectra, were contained in the mat rush. When
considering the retention indices, nos. 34 and 35 may be
considered to respectively correspond to B and C.
(E,E,E)-2,4,6-nonatrienal was synthesized to confirm the
structures of nos. 34 and 35, and its retention index was
compared to the peaks of the volatile concentrate of the
mat rush. The result showed that the retention index of
synthesized (E,E,E)-2,4,6-nonatrienal closely agreed
with that of compound C, enabling no. 35 in the mat
rush to be confirmed as (E,E,E)-2,4,6-nonatrienal. The
three kinds of 2,4,6-nonatorienals in the mat rush have
also been found14) in oat flakes, and their retention
indices on the polar stationary phase were in good
agreement with the observed values for the mat rush.
Accordingly, based on the retention index of (E,E,E)-
2,4,6-nonatrienal, compounds A and B were assigned as
the E,Z,E and E,E,Z isomers, while the structure of
no. 34, which gave a high FD factor for the mat rush
aroma, was tentatively identified as (E,E,Z)-2,4,6-non-
atrienal. The mass spectral signals of the remaining
odorants were too weak for any unequivocal interpre-
tation; these odorants were therefore tentatively identi-
fied by matching their retention indices and odor quality
to those of the standard compounds from GC-O (DB-
Wax). The result showed that, in addition to the 38
previously identified odorants, another five compounds
could be added as tentatively identified odorants.
These findings identified forty-three potent odorants
involved in the characteristic aroma of mat rash. Since
thirty were newly identify compounds in the mat rush,
the major portion of the potent odorants of mat rush can
be assumed to be too low and thus difficult to clarify.
Formation of the potent odorants in mat rush
Raw mat rushes, which are generally harvested in the
early summer (from June to the end of July), are
processed into igusa seat products (so-called tatami
omote). The steps involved in processing tatami omote
involve soaking in clay mud, which is called sendo,
drying at a comparatively low temperature for a long time
(approximately 50–70 ꢁC for more than 10 h), and finally
machine weaving. The muddied and dried mat rush was
found to have the characteristic aroma, despite the raw

Identification of the Potent Odorants in Mat Rush
1235
Table 2. Comparison of the Potent Odorants Showing High FD-Factors (ꢃ44) in the Dried (muddied and non-muddied) and Raw Mat Rush Samples
log4 (FD factor)
No.
RIa
Compound
Odor qualityb
green
Dried (muddied)
Dried (non-muddied)
Rawc
1
4
1090
1298
1338
1370
1378
1378
1439
1451
1464
1501
1535
1565
1578
1582
1621
1665
1701
1724
1733
1813
1857
1876
1895
2001
2028
2167
2195
2236
2271
2553
2565
hexanal
1-octen-3-one
5
4
4
5
5
4
4
6
4
4
6
5
6
7
4
7
6
4
4
5
5
6
4
6
7
4
7
7
4
4
7
4
4
<4
<4
<4
<4
<4
<4
<4
<4
<4
<4
4
mushroom-like
roasty
metallic
5
2-acetyl-1-pyrroline
(Z)-1,5-octadien-3-one
dimethyl trisulfide
(Z)-3-hexenol
<4
5
6
7
roasty, putrid
green
sour
4
8
<4
4
10
11
12
14
15
16
17
18
19
21
22
24
25
28
33
34
35
40
41
43
45
46
48
50
51
acetic acid
methional
green, potato-like
roasty, nutty
green, earthy
sweet, fatty
green, earthy
green, earthy
green
5
2,6-dimethyl-3-ethylpyrazine
(Z)-2-nonenal
(E)-2-nonenal
4
<4
5
(E,E)-2,6-nonadienal
unknown
5
<4
<4
5
5
(E,Z)-2,6-nonadienal
butyric acid
isovaleric acid
6
sour
sour
4
<4
<4
<4
<4
<4
<4
<4
<4
<4
4
6
(E,E)-2,4-nonadienal
unknown
pentanoic acid
sweet, fatty
sweet
sour
5
4
4
(E,E)-2,4-decadienal
2-methoxyphenol
(E,E,Z)-2,4,6-nonatrienal
(E,E,E)-2,4,6-nonatrienal
trans-4,5-epoxy-(E)-2-decenal
4-hydroxy-2,5-dimethyl-3(2H)-furanone
eugenol
sweet, fatty
spicy, burnt
sweet
4
4
5
sweet
sweet
<4
6
sweet, caramel-like
spicy
6
<4
<4
<4
<4
<4
<4
5
<4
6
3-hydroxy-4,5-dimethyl-2(5H)-furanone
methyl anthranilate
unknown
sweet, burnt sugar-like
fruity, grape-like
spicy, metallic
sweet
6
3
phenylacetic acid
vanillin
4
sweet, vanilla-like
7
aRetention index in the DB-Wax column (30 m ꢂ 0:25 mm i.d.; coated with a 0.25-mm film) observed for GC-O.
bOdor quality assigned during AEDA.
cThe FD factor of the original odor concentrate of the raw mat rush was converted into the dry weight equivalent and is defined as 41 (See the Materials and Methods
section).
I Thermally generated odorants from amino acids and/or carbohydrates
HO
O
OH
S
CHO
COOH
S
S
S
O
O
O
(no. 41, FD 7)
(no. 45, FD 7)
(no. 21, FD 7)
(no. 11, FD 6)
(no. 7, FD 5)
II Lipid degradation odorants
O
CHO
CHO
CHO
CHO
(no. 18, FD 7)
(no. 22, FD 6)
(no. 35, FD 6)
(no. 40, FD 6)
CHO
CHO
CHO
CHO
O
(no. 15, FD 6)
(no. 6, FD 5)
(no 16, FD 5)
(no. 1, FD 5)
(no. 28, FD 5)
III Thermal degradation odorants from ferulic acids
Other
OH
O
OH
O
O
O
NH
2
CHO
(no. 33, FD 7)
(no. 46, FD 7)
(no. 51, FD 7)
Fig. 2. Classification of Most Contributed Odorants (FD ꢃ 45) by Hypothetical Precursors.

1236
K. KUMAZAWA et al.
mat rush just after harvesting hardly presenting any
aroma. The sendo used in the muddying process is clay
powder, and it is believed to be important for enhancing
homogeneous drying and producing the characteristic
color, luster, and aroma of the mat rush.12) The potent
odorants contained in the dried mat rush (muddied and
non-muddied) were compared to those of the raw mat
rush (Table 2). On the basis of this result, a marked
difference in the FD factors was observed between the
dried and raw mat rush samples, despite the FD factors
and composition of the potent odorants in the dried mat
rush (between the muddied and non-muddied) being
similar. It is therefore assumed that the drying process is
the most important factor for forming the characteristic
aroma of the mat rush, and that the mud had little effect
on the formation of this characteristic aroma.
Acknowledgment
We are grateful to Mr. Y. Tanaka, Mr. K. Fuchigami,
and Mr. S. Nishida of the Kumamoto Prefectural
Agricultural Research Center for presenting the mat
rush samples and for their useful suggestions.
References
1) Morita H, ‘‘All of Igusa’’ (in Japanese), ed. Shimizu A,
Shin-Me, Yokohama, pp. 106–117 (2008).
2) Morita H, ‘‘All of Igusa’’ (in Japanese), ed. Shimizu A,
Shin-Me, Yokohama, pp. 18–23 (2008).
ˆ
H and Goto S, Nippon Nogeikagaku Kaishi
3) Kameoka
(in Japanese), 52, 323–327 (1978).
4) Shibuya S, Toyota T, and Muraki S, Proceeding of the 28th
Symposium on the Chemistry of Terpenes, Essential Oils, and
Aromatics (in Japanese), pp. 221–223 (1984).
Since the mat rush aroma was mainly generated
during the drying process, it is considered that the
formation of the potent odorants would be strongly
affected by the heating. It therefore seems that many of
the most contributive odorants (FD ꢃ 45) could be
classified by the hypothetical precursors into the three
groups (I, II and III) (Fig. 2). Group I contains the so-
called thermally generated odorants,7) which were
generated by the reaction of amino acids with carbohy-
drates or themselves, the furanones (nos. 41 and 45),
isovaleric acid (no. 21), methional (no. 11) and dimethyl
trisulfide (no. 7) also falling into group I. Group II
contains the lipid degradation odorants,11,14,16) which are
known to be generated by both heating and by light and
enzyme reactions, with hexanal (no. 1), the unsaturated
aldehydes (nos. 15, 16, 18, 22, 28, 35 and 40) and a
ketone (no. 6) also falling into group II. Finally, the
odorants in group III are derived from ferulic acid,7,17)
with 2-methoxyphenol (no. 33) and vanillin (no. 51) that
are formed by thermal degradation accompanied by
decarboxylation and oxidation also falling into group III.
It is therefore considered that the characteristic odorants
in mat rush have the corresponding precursors. These
hypothetical precursors are generally non-volatile com-
pounds included in a variety of foodstuffs, although
there have been few detailed studies concerning the
chemical composition of the mat rush. It will be
important to understand the formation mechanism of
the mat rush odorants and their precursors in the future
to improve the aroma quality of tatami products.
5) Takatani Y, Kawata J, and Miyazawa M, Proceeding of the 51st
Symposium on the Chemistry of Terpenes, Essential Oils, and
Aromatics (in Japanese), pp. 91–92 (2007).
6) Schieberle P, ‘‘Characterization of Food: Emerging Methods,’’
ed. Goankar A, Elsevier, Amsterdam, pp. 403–431 (1995).
7) Belitz HD, Grosch W, and Schieberle P, ‘‘Food Chemistry’’ 3rd
revised edition, Springer-Verlag, Berlin, Heidelberg, pp. 342–
408 (2004).
8) Buttery RG, Ling LC, Juliano BO, and Turnbaugh JG, J. Agric.
Food Chem., 31, 823–826 (1983).
9) Swoboda PAT and Peers KE, J. Sci. Food Agric., 28, 1019–
1024 (1977).
10) Winter M, Furrer A, Willhalm B, and Thommen W, Helv. Chim.
Acta, 59, 1613–1620 (1976).
11) Kumazawa K, Wada Y, and Masuda H, J. Agric. Food Chem.,
54, 4795–4801 (2006).
12) Morita H, ‘‘All of Igusa’’ (in Japanese), ed. Shimizu A, Shin-
Me, Yokohama, pp. 126–127 (2008).
13) Engel W, Bahr W, and Schieberle P, Eur. Food Res. Technol.,
209, 237–241 (1999).
14) Schuh C and Schieberle P, J. Agric. Food Chem., 53, 8699–
8705 (2005).
15) Chetschik I, Granvogl M, and Schieberle P, J. Agric. Food
Chem., 56, 10237–10243 (2008).
16) Belitz HD, Grosch W, and Schieberle P, ‘‘Food Chemistry’’ 3rd
revised edition, Springer-Verlag, Berlin, Heidelberg, pp. 157–
244 (2004).
17) Fiddler W, Parker WE, Wasserman AE, and Doerr RC, J. Agric.
Food Chem., 15, 757–761 (1967).