Characterization of typical potent odorants in raw and cooked Toona sinensis (A. Juss.) M. Roem. by instrumental-sensory analysis techniques

Article abstract of DOI:10.1016/j.foodchem.2018.12.112

Toona sinensis (A. Juss.) M. Roem. (TS) possesses a unique and pleasant flavor and is consumed as a popular seasonal vegetable in certain parts of eastern and southeastern Asia. The potent odorants in raw and cooked TS were identified by combined sensory and instrumental analysis techniques, including sensory descriptive aroma profiling and two complimentary volatile isolation methods combined with gas chromatography-olfactometry (GC-O) techniques. Highly volatile odorants were determined by static headspace dilution analysis (SHDA)-GC-O, while those of intermediate- and semi-volatility were determined by solvent-assisted flavor evaporation-aroma extract dilution analysis (SAFE-AEDA). Among the numerous odorants identified by SHDA and SAFE-AEDA, (E,E)-bis-(1-propenyl) disulfide was found to be predominant in both raw and cooked TS. In agreement with results of sensory descriptive analysis, hexanal, (Z)-3-hexenal, (E)-2-hexenal and (Z)-3-hexen-1-ol contributed green, grassy and leafy aroma notes; while hydrogen sulfide, methyl thiirane, (E,E)-bis-(1-propenyl) disulfide and (E,Z)-bis-(1-propenyl) disulfide contributed pungent, sulfurous and alliaceous notes in TS.

Full text of DOI:10.1016/j.foodchem.2018.12.112

Accepted Manuscript
Characterization of Typical Potent Odorants in Raw and Cooked Toona sinen-
sis (A. Juss.) M. Roem. by Instrumental-Sensory Analysis Techniques
Wenxi Yang, Keith R. Cadwallader, Yuping Liu, Mingquan Huang, Baoguo
Sun
PII:
S0308-8146(19)30021-4
DOI:
https://doi.org/10.1016/j.foodchem.2018.12.112
FOCH 24071
Reference:
Food Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
5 September 2018
22 December 2018
29 December 2018
Please cite this article as: Yang, W., Cadwallader, K.R., Liu, Y., Huang, M., Sun, B., Characterization of Typical
Potent Odorants in Raw and Cooked Toona sinensis (A. Juss.) M. Roem. by Instrumental-Sensory Analysis
Techniques, Food Chemistry (2019), doi: https://doi.org/10.1016/j.foodchem.2018.12.112
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Characterization of Typical Potent Odorants in Raw and Cooked
Toona sinensis (A. Juss.) M. Roem. by Instrumental-Sensory
Analysis Techniques
Wenxi Yang a,b, Keith R. Cadwallader c, Yuping Liu a,*, Mingquan Huang
b, Baoguo Sun a,b
a
Beijing Advanced Innovation Center for Food Nutrition and Human Health, Beijing Technology
and Business University, Beijing 100048, China;
b
Beijing Key Laboratory of Flavor Chemistry, Beijing Technology and Business University,
Beijing 100048, China;
c
Department of Food Science and Human Nutrition, University of Illinois at Urbana-Champaign,
1302 West Pennsylvania Avenue, Urbana, IL 61801, USA
ABSTRACT
Toona sinensis (A. Juss.) M. Roem. (TS) possesses a unique and pleasant flavor and is consumed as
a popular seasonal vegetable in certain parts of eastern and southeastern Asia. The potent odorants in
raw and cooked TS were identified by combined sensory and instrumental analysis techniques,
including sensory descriptive aroma profiling and two complimentary volatile isolation methods
combined with gas chromatography-olfactometry (GC-O) techniques. Highly volatile odorants were
determined by static headspace dilution analysis (SHDA)-GC-O, while those of intermediate- and
semi-volatility were determined by solvent-assisted flavor evaporation-aroma extract dilution
analysis (SAFE-AEDA). Among the numerous odorants identified by SHDA and SAFE-AEDA,
(E,E)-bis-(1-propenyl) disulfide was found to be predominant in both raw and cooked TS. In
agreement with results of sensory descriptive analysis, hexanal, (Z)-3-hexenal, (E)-2-hexenal and (Z)-
3-hexen-1-ol contributed green, grassy and leafy aroma notes; while hydrogen sulfide, methyl
thiirane, (E,E)-bis-(1-propenyl) disulfide and (E,Z)-bis-(1-propenyl) disulfide contributed pungent,
sulfurous and alliaceous notes in TS.
1

ARTICLE INFO
Keywords:
Toona sinensis, flavor, odorants, gas chromatography-olfactometry, gas chromatography-mass
spectrometry, sensory descriptive analysis
2

1. Introduction
Toona sinensis (A. Juss.) M. Roem. (TS, also known as Xiangchun in Chinese or Chinese
mahogany) is a member of family Meliaceae. It is a deciduous tree, and is cultivated in eastern and
southeastern Asia, e.g., in North Korea, eastern, central and southwestern China, Nepal, northeastern
India, Myanmar, Thailand, Malaysia and western Indonesia (Hsu, Huang, Chen, Mao, & Chaw, 2012).
Because the young leaflets and tender shoots of TS have a pleasant aroma and are nutrient-dense,
they have been consumed as vegetables. When the leaves are young, they are dark-red or red-brown
and possess a strong and charactersitc aroma. However, when the leaves become older, they turn
green, and their aroma intensity decreases. Raw TS shoots are seldom used directly as food materials,
since they contain some toxic compounds (nitrites). After cooking (e.g. blanching), the leaves are
served in salads, stir-fried with egg, pickled with salt, or roasted, dried and used as a healthy tea.
Recent studies on TS have mainly focused on the extraction and analysis of its nutrient substances,
such as quercetin, gallic acid (Yu, Chang, Kuo, Tsai, & Chang, 2012), limonoids (Hu, Song, Mao,
Wang, & Zhao, 2016) and flavonoid glycosides (Zhang et al., 2014).
Aroma is an important sensory attribute to TS. Nevertheless, there are very few reports on the
volatile composition of TS. In these studies, nearly all of extraction methods used were based on
headspace solid-phase microextraction (HS-SPME), and the columns used in gas chromatography-
mass spectrometry (GC-MS) were nonpolar, such as HP-5, DB-5, or PE-5. Furthermore, the volatile
substances were identified mainly by matching mass spectra with literature or databases. Only three
previous studies have analyzed the aroma-active compounds in TS by gas chromatography-
olfactometry (GC-O). Li and others analyzed the HS-SPME isolates of TS (from Beijing, China) by
GC-O, and reported ten aroma-active regions, two of which were thought to possess the characteristic
aroma of TS and believed to be sulfur-containing compounds based on GC/flame photometric
detector (FPD) analysis. But, these two sulfur-containing compounds were not identified (Li, Wang,
Dai, Su, & Chen, 2011). Liu et al. (2013) investigated the aroma-active compounds of TS (Shanxi,
China) by SPME-GC-O. Twenty-six odorants were tentatively identified by their mass spectra and

GC retention indices (RI). Two compounds (Z)- and (E)- isomers of 2-mercapto-3,4-dimethyl-2,3-
dihydrothiophene were thought to be major contributors to the characteristic aroma of TS, but their
identities were not confirmed by analysis of authentic reference compounds (Liu et al., 2013). Using
the same approach, Li et al. (2017) also studied the aroma-active compounds of TS (Tianjin, China).
Fifteen aroma-active regions were detected, and 11 odorants were tentatively identified by only their
mass spectra (Li, Wang, Zhao, & Liu, 2017). In the above studies, the compounds which contributed
to the overall odor profiles of TS were not determined. In a separate study, (S,S)‐γ-glutamyl‐(Z-S‐1-
propenyl)thioglycine, (S,S)-γ-glutamyl-(E-S-1-propenyl)thioglycine and γ-glutamyl-(Z-S-1-
propenyl)-cysteine were identified as potential nonvolatile precursors to volatile sulfur compounds
in TS; however, the odor contribution of these volatile sulfur compounds were not determined (Li,
Eidman, Gan, Haefliger, Carroll, & Pika, 2013).
A systematic and comprehensive study on the characteristic aroma-active components of TS is
clearly warranted. Herein, we report on the combined use of various sensory- and instrumental-based
analysis techniques to accomplish this task. The objectives of the present study were (1) to compare
the sensory descriptive aroma profiles of raw and cooked TS; (2) to identify the aroma-active
compounds in raw and cooked TS; (3) and to compare raw and cooked TS with respect to their
predominant odorants.
2. Materials and Methods
2. 1. Samples
Three batches of fresh and tender TS shoots were purchased from a local supermarket (Yonghui
Superstore, Beijing, China) in April and May 2017 (purchased within two weeks of each other). TS
shoots were cultivated in Shandong Province, China. Sample preparation and preservation procedures
were as follows: raw TS shoots were frozen by liquid nitrogen, then vacuum packaged in food grade
polyethylene plastic storage bags and stored at -60 °C until analysis. Cooked TS shoots were blanched
4

in boiling water (100 °C) for 60 s, cooled to room temperature, and then frozen and stored as
described above.
2. 2. Chemicals
Authentic reference aroma compounds were supplied by the companies given in parentheses:
acetaldehyde, ethanethiol, dimethyl sulfide, methyl thiirane, 2-methylpropanal, 3-methylbutanal,
thiophene, 2,3-butanedione, methyl allyl sulfide, α-pinene, dimethyl disulfide, propyl sulfide,
hexanal, 1-pentanethiol, 2-methyl-2-pentenal, 2,5-dimethylthiophene, α-terpinene, (E)-2-hexenal,
butyl sulfide, 2-isopropyl-4-methylthiazole, 2,4,5-trimethylthiazole, (E,E)-2,4-hexadienal, methional,
diallyl disulfide, (Z)-3-hexen-1-ol, linalool, (E,Z)-2,6-nonadien-1-al, -caryophyllene, 3-
methylbutanoic acid, 2-methyl-3-(methylthio)furan, borneol, 4-hydroxy-2,5-dimethyl-3(2H)-
furanone and eugenol, mixture of (Z)-isoeugenol and (E)-isoeugenol, phenylacetic acid, vanillin and
n-alkanes (C5−C28) (Sigma-Aldrich, St. Louis, MO); 2-methylbutanal, (Z)-3-hexenal and 1-octen-
3-one (Bedoukian, Danbury, CT); (Z)-6-nonenal (Alfa Aesar, Lancashire, U.K.); allyl mercaptan
(Fluka, Buchs, Switzerland); ethyl acetate (J.T. Baker Chemical Co, Phillipsburg, NJ ); 1-hexanol
(Polysciences Inc, Warrington, PA); and dimethyl trisulfide (Columbia Chemical Corporation Inc,
Brunswick, Ohio).
Unless otherwise indicated, all reagent grade chemicals were obtained from Fisher Scientific (Fair
Lawn, NJ), including anhydrous sodium sulfate (analytical grade), dichloromethane [high
performance liquid chromatography (HPLC) grade] and sodium chloride (analytical grade).
Anhydrous tetrahydrofuran (analytical grade) was purchased from Aladdin (Shanghai, China), sulfur
powder (analytical grade, purified by precipitation) was obtained from Beijing Yili Fine Chemicals
Co., Ltd (Beijing, China), magnesium chips, iodine beads (anhydrous) (Z)-1-bromo-1-propene (0.97)
and (E)-1-bromo-1-propene (0.99) were purchased from Sigma-Aldrich and 1-propenylmagnesium
bromide (0.5M solution in THF) was supplied by Acros Organics (United Kingdom).
5

Ultra-high-purity (UHP) nitrogen and UHP helium were purchased from S.J. Smith (Davenport,
IA). Odorless distilled water was prepared by boiling glass-distilled water in an open flask until its
volume was reduced by one-third of the original volume. Diethyl ether was freshly distilled prior to
use.
2. 3. Synthesis of references
2. 3. 1. Synthesis of (Z or E)-1-propenylmagnesium bromide
A mixture of magnesium turnings (1.079 g, 450mmol), anhydrous tetrahydrofuran (40 mL) and
iodine beads (0.091 g) was placed in a 100 mL three-neck flask fitted with a magnetic stir bar, an
addition funnel, a condenser with drying tube and a rubber stopper for introducing N2. A portion (1
mL) of a solution of (Z)-1-bromo-1-propene (0.97, 5 g, 41 mmol) or (E)-1-bromo-1-propene (0.99, 5
g, 41 mmol) in anhydrous tetrahydrofuran (10 mL) which was put into the addition funnel was added
to the flask. The reaction mixture was heated by oil bath under stirring. The heating was not stopped
until the color of iodine disappeared, and then the remaining 1-bromo-1-propene solution was added
into the flask. Finally the flask was warmed for 30 min in oil bath at 80 °C to finish the preparation
of the (Z/E)-1-propenylmagnesium bromide which was used for the next step.
2. 3. 2. Synthesis of bis-(1-propenyl) mono- and di-sulfides
The mixture from 2. 3. 1. was combined with sulfur powder (1.312 g, 41 mmol, in 40 mL anhydrous
tetrahydrofuran ) for synthesize bis-(1-propenyl) mono- and di-sulfides.
The same apparatus used in 2. 3. 1 was applied in this part, except that prior to the reaction the flask
was cooled in an ice-water bath. 1-Propenylmagnesium bromide solution (30 mL, 15 mmol) was
injected by syringe into the flask by piercing the rubber stopper. After addition of sulfur powder (544
mg, 20 mmol, in 20 mL anhydrous tetrahydrofuran) the reaction mixture was first stirred in ice bath
for 0.5 h, and then for 0.5 h at room temperature. The reaction mixture was poured into a beaker
containing a 30 g mixture of ice and water. The organic layer was separated, and the aqueous layer
6

was extracted with diethyl ether (3 × 100 mL). The organic layer and the diethyl ether extracts were
combined, washed with saturated sodium chloride solution (2 × 100 mL), dried over anhydrous
sodium sulfate and then filtered to remove any debris. The filtrate was concentrated under a gentle
stream of nitrogen gas. Characterization of bis-(1-propenyl) mono- and di-sulfides was carried out
by GC-MS and GC-O-FPD.
(Z,Z)-bis-(1-propenyl) sulfide:
MS-EI m/z (%): 99 (100), 114 (95, M+), 45 (43), 39 (39), 85 (32), 41 (29), 71 (23), 72 (18), 73
(18), 59 (11); RI: 1200 (DB-WAX), 914 (HP-5MS).
(E,Z)-bis-(1-propenyl) sulfide:
MS-EI m/z (%): 99 (100), 114 (99, M+), 45 (44), 39 (38), 85 (32), 41 (29), 71 (23), 72 (19), 73
(18), 59 (12); RI: 1209 (DB-WAX), 923 (HP-5MS).
(E,E)-bis-(1-propenyl) sulfide:
MS-EI m/z (%): 114 (100, M+), 99 (97), 45 (44), 39 (35), 85 (33), 41 (29), 71 (23), 72 (18), 73
(18), 59 (12); RI: 1220 (DB-WAX), 926 (HP-5MS).
(E,E)-bis-(1-propenyl) disulfide:
MS-EI m/z (%): 146 (100, M+), 45 (92), 41 (60), 74 (58), 39 (55), 73 (46), 71 (34), 113 (34), 82
(24), 72 (10); RI: 1447 (DB-WAX), 1115 (HP-5MS).
(E,Z)-bis-(1-propenyl) disulfide:
MS-EI m/z (%): 45 (100), 146 (98, M+), 74 (72), 41 (71), 113 (60), 39 (58), 73 (40), 71 (38), 82
(22), 67 (20); RI: 1468 (DB-WAX), 1120 (HP-5MS).
(Z,Z)-bis-(1-propenyl) disulfide:
MS-EI m/z (%): 71 (100), 146 (75, M+), 45 (56), 113 (43), 41 (26), 39 (26), 73 (24), 82 (16), 74
(15), 72 (14); RI: 1491 (DB-WAX), 1126 (HP-5MS).
The relative proportions of isomers in the synthesis products were determined by GC-O-FPD.
From (Z)-1-bromo-1-propene
Bis-(1-propenyl) sulfide (Z,Z/E,Z/E,E): 77.7%/22.3%/0
7

Bis-(1-propenyl) disulfide (Z,Z/E,Z/E,E): 54.7%/23.6%/21.4%
From (E)-1-bromo-1-propene
Bis-(1-propenyl) sulfide (Z,Z/E,Z/E,E): 0/21.4%/78.6%
Bis-(1-propenyl) disulfide (Z,Z/E,Z/E,E): 0/0/100%
From 1-propenylmagnesium bromide (Z and E mixture)
Bis-(1-propenyl) sulfide (Z,Z/E,Z/E,E): 25.3%/42.2%/32.5%
Bis-(1-propenyl) disulfide (Z,Z/E,Z/E,E): 54.5%/25.7%/19.8%
2. 4. Sensory Aroma Profiling
Aroma profiles were determined following closely the method previously descibed by Zhou,
Wintersteen, & Cadwallader (2002). Panelists (4 male and 7 female, consisting of university
students/staff, 23−48 years of age) had previously received extensive training in descriptive sensory
analysis (>20 h), specifically in flavor profiling of various food samples. The panel underwent an
additional 4 h of training to become familiar with the evaluation of TS aroma. Raw or cooked TS
(chopped, 3.5 g) was placed in a 500 mL low density polyethylene (LDPE) wash bottle with siphon
tubes removed from the caps. Bottles were covered with aluminum foil to minimize visual bias and
labeled with random 3 digit codes. Samples were equilibrated at room temperature (approx. 27 °C)
for 30 min before presenting to panelists. Panelists evaluated each sample by gently squeezing the
bottle and taking short sniffs of the air emitted from the nozzle. Aroma intensities were scored on 15
cm universal scales anchored on the left with “none” and on the right with “very strong” which
corresponded to intensity ratings of 0 and 15, respectively (Meilgaard, Civille, & Carr, 1999). Each
term was anchored with a standard reference with an intensity rating determined by the panel, as
follows: “green/grassy” represented by 1.0 g of cut-grass (intensity score of 12); “cooked meat”
represented by 0.2 g of a ground vitamin B pill (intensity score of 10); “alliaceous” represented by
0.9 g of chopped raw garlic (intensity score of 14); “earthy” represented by 3.2 g of chopped
Hershey’s dark chocolate (intensity score of 9); “pungent” represented by 1.4 g of chopped raw onion
(intensity score of 13). Individual panelist rating results were revealed at the end of each sensory
8

analysis session, and final aroma profiles of the samples were reported on the basis of discussion and
by consensus of the panel.
2. 5. Isolation of Volatile Compounds.
2. 5. 1. Static Headspace
The procedure was based on the method described by Cadwallader, Potts, Brisske-BeVier, &
Mirarefi (2011) with some modifications. Chopped TS (5 g) was placed in a 60 mL glass serum vial.
The vial was sealed with a polytetrafluoroethylene (PTFE) faced silicon septum and then incubated
in a water bath at 35 °C for 25 minutes. Following incubation, a headspace volume was withdrawn
for GC-O. A fresh sample was used for each headspace volume tested.
2. 5. 2. Direct Solvent Extraction and Solvent-Assisted Flavor Evaporation (DSE-SAFE)
Thirty grams of raw or cooked TS chopped shoots plus 30 g of sodium chloride were added to in a
250 mL glass bottle. The bottle was frozen (-60 °C) and then dichloromethane (120 mL, at -60 °C),
avoiding to produce thermal artifacts during homogenizing, was added, and the mixture homogenized
by Ultra Turrax (IKE, Germany) at the speed of 10000 rpm for 2 min. The blended sample was passed
through a Whatman No. 4 filter paper and the filter cake was washed twice with 10 mL of
dichloromethane.
The solvent extract was subjected to SAFE in order to remove any nonvolatile material. SAFE
(ACE Glassware, Vineland, NJ) was conducted for for 2.5 h at 40 °C as described previously
(Rotsatchakul, Chaiseri, & Cadwallader, 2008). The resulting aroma extract was concentrated to 5
mL using a 15 cm Vigreux column at 45 °C water bath. Final extract was dried over 1 g of anhydrous
sodium sulfate and then further concentrated under a gentle stream of nitrogen to 400 μL. Extracts
were transferred to 2 mL amber glass vials equipped with PTFE-lined closures and stored at -60 °C
until analysis.
2. 5. 3. HS-SPME
9

HS-SPME-GC-MS was used to aid in the identification of odorants detected by static headspace
and SAFE. A 2-cm SPME fiber, coated with 50/30 μm divinylbenzene/carboxen/polydimethyl
siloxane (Supelco, Bellefonte, PA, U.S.) was preconditioned at 270 °C for 1 h prior to analysis. For
analysis, a sample (8 g, chopped raw or cooked TS) was transferred to a 40 mL static headspace
amber glass bottle and the bottle sealed with an aluminum cap (with PTFE /silicon septum, thickness
1.3 mm; Supelco). The operating conditions for HS-SPME were as follows: pre-incubation and
absorption temperature, 35 °C; pre-incubation time, 20 min; absorption time, 20 min; GC desorption
time, 10 min; GC desorption temperature, 250 °C. Analyses were performed in duplicate.
2. 6. Analysis of Volatile Compounds
2. 6. 1. GC-O Conditions for Static Headspace Dilution Analysis (SHDA)
SHDA was performed using an Agilent 6890N GC (Agilent Technologies, Inc.; Palo Alto, CA,
U.S.) equipped with an flame ionization detector (FID) and olfactory detector port (ODP2, Gerstel).
Separations were preformed using either an RTX-WAX (15 m × 0.53 mm × 1.0 μm; Restek,
Bellefonte, PA) or RXI-5MS (15 m × 0.53 mm × 0.5 μm; Restek) GC column. Following incubation,
a headspace volume (25, 5, 1, 0.2 or 0.04 mL) was withdrawn by means of a heated (45 °C) gas-tight
syringe which injected the same volume of headspace they would withdraw from glass vials, then
withdraw the headspace as follows: 25 mL syringe for 25 mL headspace, 5 mL syringe for 5 mL
headspace, 1 mL syringe for 1 mL headspace, 250 μL syringe for 200 μL headspace, 50 μL syringe
for 40 μL headspace, and 10 μL syringe for 5 μL headspace. Then the syringe injected into a CIS-4
cooled injection system (Gerstel GmbH & Co. KG, Germany) operating in the solvent vent mode
(vent pressure, 8.5 psi; vent flow, 25.0 mL/min; splitless time, 1.5 min); initial temperature, -120 °C
(0.1 min hold); ramp rate at 12 °C/s to 220 °C (10 min hold); to final temperature at the same ramp
rate, 260 °C (10 min hold).Column effluent was split (1:5) between FID and ODP using two
deactivated fused silica capillaries. FID and olfactory transfer line temperatures were set to 250 °C.
The GC oven temperature was programmed from 35 °C (hold for 5 min) to 225 °C (hold for 30 min)
10

at a rate of 6 °C/min (for both RTX-WAX and RXI-5MS columns). The carrier gas was helium at a
constant flow rate of 5 mL/min. The makeup gases of the ODP consisted of nitrogen (sheath gas) and
humidified air (nose cone). Odor qualities (by GC-O) were determined by three panelists (1 female
and 2 males, 24−54 years old). The details of the procedure have been previously described
(Lapsongphon, Yongsawatdigul, & Cadwallader, 2015; Zhou et al., 2002).
2. 6. 2. GC-O Conditions for Aroma Extract Dilution Analysis (AEDA)
Each aroma extract from DSE-SAFE was stepwise diluted (1:3,v/v) with dichloromethane
according to the general AEDA technique (Grosch, 1994). The diluted samples were transfered to
PTFE-lined screw caps and stored at -60 °C prior to analysis. GC-O was conducted using a 6890N
GC (Agilent Technologies Inc.) equipped with a FID and an ODP2. The aroma extract (2 μL) was
injected in the cool on-column mode (+3 °C oven tracking mode) to avoid injection bias and reduce
the chance for thermal degradation of any labile compounds. The conditions for GC analysis were
the same as those used for GC-O in 2. 6. 1, except the ramp rate was 10 °C/min for the RTX-WAX
column and 6 °C/min for the RTX-5MS column. GC-O was conducted by four panelists (2 female
and 2 males, 20−32 years old).
2. 6. 3. GC-O Conditions for SPME
GC-O for SPME was carried out using an Agilent 7890 GC equipped with a FID or a FPD (Agilent
Technologies Inc.) and an ODP (ODP3, Gerstel). The effluent was split between the FID/FPD and
ODP in a 1:2 ratio by volume using Y-type splitter and two deactivated fused silica capillaries. The
FID and FPD were held at 250 °C and 200 °C, respectively. The temperature of the ODP transfer line
was 200 °C. SPME samples were analyzed using a DB-WAX column (30 m × 250 μm × 0.25 μm;
Agilent Technologies Inc.) or an HP-5MS column (30 m × 0.25 mm × 0.25 μm; Agilent Technologies
Inc.). Initial oven temperature was 35 °C (held for 5 min), then increased to 225 °C at 10 °C/min
(DB-WAX) or 6 °C/min (HP-5MS), with a final hold time of 30 min. Injection was done in the hot
splitless mode (4 min valve-delay) at 250 °C (for FID) or 150 °C (for FPD). Helium was used as the
11

carrier gas at a constant flow rate of 1 mL/min. The effluent sent to the ODP was combined with a
stream of humidified air.
2. 6. 4. GC-MS
Analyses were conducted using an Agilent 7890B GC coupled to a 5977A mass-selective detector
(MSD) (Agilent Technologies Inc.). The SAFE isolates (2 μL) or SPME extracts were analyzed on
both a DB-WAX capillary column (30 m × 0.25 mm × 0.25 μm; Agilent Technologies Inc.) and an
HP-5MS column (30 m × 0.25 mm × 0.25 μm; Agilent Technologies Inc..). Initial oven temperature
was 35 °C (held for 5 min), then increased to 225 °C at 10 °C/min (DB-WAX) or 6 °C/min (HP-
5MS), with a final hold time of 30 min.. Injections were made in the hot splitless mode (250 C; 4
min valve-delay). The carrier gas was helium at a constant flow rate of 1 mL/min.. The MSD
conditions were as follows: capillary direct interface temperature, 250 °C; mass quadrupole
temperature, 150 °C; ionization energy, 70 eV; ion source temperature, 230 °C; mass range, 33–300
amu; EM voltage, autotune, +200 V; scan rate, 5.3 sans/s. The chromatograms were analyzed using
MassHunter Qualitative Analysis Software B.06.00 (Agilent Technologies, Inc.).
2. 7. Compound Identification
The retention index (RI) of each compound was calculated using the retention time (RT) of that
compound compared against the RTs of a series of standard n-alkanes (C5-18 for non-polar column:
RXI-5MS and HP-5MS, C5-28 for polar columns: RTX-WAX and DB-WAX), according to Van
Den Dool, & Kratz (1963). The RI values, in combination with aroma descriptions, FPD and mass
spectral library results (2014 National Institute of Standards and Technology), were compared with
literature values and online databases (Flavornet and NIST Chemistry WebBook) to provide tentative
compound identifications. Authentic standards of tentatively identified compounds were analyzed by
GC-O and GC-MS to confirm their RIs and aroma attributes for the purpose of positive identification.
12

3. Results and discussion
3. 1. Sensory Aroma Profiles of Raw and Cooked TS
To obtain an idea of the overall aroma profiles of raw and cooked TS, sensory descriptive analysis
was performed by rating six main aroma attributes including alliaceous, cooked meat, earthy,
green/grassy, pungent and overall intensity (Figure 1). Except for the green/grassy note, the
intensities of the all aroma attributes were higher in cooked TS than in raw TS. In contrast the
green/grassy attribute had the strongest intensity in raw TS, followed by pungent, alliaceous, cooked
meat and earthy notes. In cooked TS, the alliaceous attribute had the strongest intensity, followed by
pungent, green/grassy, cooked meat and earthy notes. The following sections describe the results of
sensory-instrumental based methods for the identification of the potent odorants responsible for the
characteristic aroma of raw and cooked TS.
3. 2. Potent Headspace Odorants Detected by SHDA
In the case when aroma isolation is done by extraction techiques, some highly volatile compounds
might be overlooked because they can be lost during the various preparation and workup steps
employed. For this reason the highly volatile aroma compounds were evaluated in this study by the
SHDA. Prior to the analysis, the static headspace above TS was collected using a gas-tight syringe,
then expelled out toward to the noses of panelists to make sure the captured headspace possessed the
characteristic odor expected for TS. Similar odors descriptions were used by panelists to describe the
odors compared with those for the sensory aroma profiles for TS. Based on these results, SHDA was
thought to be a good way to analysize high volatile aroma compounds in TS, and the most potent
compounds could be screened by analysis of decreasing static headspace volumes.
A total of 32 and 36 aroma-active regions were detected in the the static headspace of raw and
cooked TS, respectively, on both RTX-5MS and RTX-WAX columns (Table 1). Twenty-five
odorants were common constituents of raw and cooked TS. Most of the compounds identified in TS
(nos. 1, 2, 4, 5, 10, 11, 13, 19, 20, 22, 26, 29, 32, 33, 37, 41, 42, 45, 47-49, 65, 71 and 75) were sulfur-
13

containing compounds. The results of SHDA also supported the differences between raw and cooked
TS as indicated by the sensory aroma profiles. For example, aldehydes and alchohol containing six
carbons (nos. 21, 24, 30 and 50) contributed green, grassy and leafy notes. Certain sulfur compounds
(nos. 32, 33, 37, 42, 46 and 48) contributed cooked meat notes, while some other sulfur-containing
odorants and two unknown compounds (nos. 9, 10, 11, 13, 26 and 51) contributed alliaceous notes.
These compounds were detected at higher flavor dilution (FD) factors in cooked TS than in raw TS.
This is in agreement with the sensory results which showed that cooked TS had higher intensities for
cooked meat and alliaceous notes. Some aldehydes and sulfur compounds (nos. 1, 3, 4, 5 and 7 ) were
described as pungent, and some other compounds, including aldehydes and terpenes (nos. 12, 18, 27
and 61) contributed earthy notes.
Among the compounds identified in raw TS, acetaldehyde, dimethyl sulfide, 2-/3-methylbutanal,
(Z)-3-hexenal, α-terpinene and (E,E)-bis-(1-propenyl) disulfide were detected at a relatively high FD
factor of 25, followed by 2-methylpropanal, allyl mercaptan, methyl thiirane, 2,3-butanedione,
hexanal, diisopropyl disulfide and (Z)-3-hexen-1-ol detected, all of which had an FD factor of 5.
Among the odorants identified in cooked TS, hydrogen sulfide had the highest FD factor of 625,
followed by dimethyl sulfide, methyl thiirane, hexanal, (Z)-3-hexenal, 2,5-dimethylthiophene,
dibutyl sulfide, 2-isopropyl-4-methylthiazole, dimethyl trisulfide, (E,E)-bis-(1-propenyl) disulfide,
linalool and β-caryophyllene with FD factors of 25. Acetaldehyde, ethanethiol, thiophene, -pinene,
-terpinene and (E,E)-bis-(1-propenyl) disulfide were detected at an FD factor of 5.
With respect to odorants found in both raw and cooked TS, some compounds (nos. 3, 7, 10, 12, 15,
27, 32, 50) which had high FD factors (≥ 5) in raw TS had lower FD factors in cooked TS, except for
dimethyl sulfide, (Z)-3-hexenal and (E,E)-bis-(1-propenyl) disulfide which had the same FD factor
in both raw and cooked TS. Some compounds (nos. 2, 12, 30, 32, 35, 50, 59) were only detected in
raw TS. There were two possible reasons for this. The first was that these compounds might react
with other compounds to form new substances during blanching. The second was that they possibly
leached into the water during blanching so that these odorants had lower FD factors or couldn’t be
14

detected in cooked TS. However, some compounds (nos. 11, 18, 21, 33, 37, 42, 48 and 61) had higher
FD factors in cooked TS than in raw TS. Additionally, some odorants (nos. 1, 4, 9, 13, 26, 40, 51, 55,
66, 75 and 87) were only detected in cooked TS.
3. 3. Potent Odorants Detected by AEDA
A total of 59 odorants were detected in TS by AEDA (Table 2). Results of AEDA were in good
general agreement with those of SHDA, with 29 of the compounds detected by AEDA also found by
SHDA. Seven highly volatile compounds (nos. 1, 2, 3, 5, 7, 9 and 10) detected by SHDA were not
detected by AEDA. In contrast, 12 odorants with intermediate and low volatilities (nos. 74, 76-86
and 88) were only detected by AEDA. Odorants identified by AEDA included those with green,
grassy and leafy (nos.21, 24, 25, 30, 40 and 50), cooked meat (nos. 28, 29, 32, 33, 37, 41, 42, 47, 48
and 65), alliaceous (garlic-like, cooked onion-like) (nos. 8, 11, 19, 49 and 81), pungent (nos. 4, 22
and 80), earthy (nos. 12, 18, 27, 58, 61 and 78) and sweet notes (nos. 66, 72, 79, 82, 83, 85-88).
Fifty-five odorants were detected in raw TS, with (E,E)-bis-(1-propenyl) disulfide having the
highest FD factor of 19683, followed by an unknown compound (no. 58; woody, pine), β-
caryophyllene, (Z)-isoeugenol, phenylacetic acid and vanillin detected at an FD of 729. Hexanal, (Z)-
3-hexenal, methional, linalool, unknown (no. 80; pungent, sweaty) and cinnamyl acetate were all
detected at a somewhat lower FD factor of 243. An additional 5 odorants, all with FD factors of 81,
were identified as 2-methyl-2-pentenal, 2,4,5-trimethylthiazole, 3-methylbutanoic acid, geraniol and
eugenol.
Forty-eight odorants were detected in cooked TS. (E,E)-Bis-(1-propenyl) disulfide had the highest
FD factor of 59049, followed by (E,Z)-bis-(1-propenyl) disulfide (FD factor of 6561) and 2,4-
dimethylthiophene, 2,4,5-trimethylthiazole, β-caryophyllene, borneol, (Z)-isoeugenol, phenylacetic
acid and vanillin (All with an FD factor of 2187). Odorants having an FD factor of 729 were identified
as methyl thiirane, (Z)-3-hexen-1-ol, unknown (no. 72; sweet, woody, fruity, mint), cinnamyl acetate
and p-vinyl guaiacol. An FD factor of 243 was assigned to dimethyl disulfide, (E)-4,5-expoxy-(E)-
2-decenal and eugenol.
15

Forty-four volatile compounds were commonly detected in raw and cooked TS. Those with
relatively high FD factors were (E,E)-bis-(1-propenyl) disulfide, β-caryophyllene, (Z)-isoeugenol,
phenylacetic acid, vanillin and cinnamyl acetate. Hexanal, (Z)-3-hexenal, two unknowns (no. 58;
woody, pine; and no. 80; pungent, sweaty) and linalool had higher FD factors in raw than in cooked
TS, as blanching in boiling water might decrease the contents of these compounds. This could explain
why people blanch TS to remove green and grassy notes. Compared with raw TS, sulfur compounds
had much higher FD-factors in cooked TS and contributed cooked meat and alliaceous notes. These
included methyl thiirane, dimethyl disulfide, 2,5-dimethylthiophene, bis-(1-propenyl) sulfide,
isopropyl disulfide, 2,4,5-trimethylthiazole, (E,E)-bis-(1-propenyl) disulfide and (E,Z)-bis-(1-
propenyl) disulfide. In addition, compounds with high FD factors in cooked TS that contributed sweet
and woody odor characteristics were β-caryophyllene, borneol, (E)-4,5-expoxy-(E)-2-decenal,
cinnamyl acetate, eugenol, p-vinyl guaiacol, (Z)-isoeugenol, (E)-isoeugenol, phenylacetic acid and
vanillin.
3. 4. Aroma Chemistry of TS
The green/grassy attribute was predominant in the aroma profile of raw TS (Figure 1); whereas,
cooked TS possessed high intensities of alliaceous and pungent notes. Based on the results of SHDA
and AEDA, several sulfur-containing compounds were identified as potent odorants in TS, and they
contributed cooked meat, alliaceous and pungent notes. In addition, phenolic compounds imparted
sweet notes in TS, especially in cooked TS.
3.4.1 Potent Sulfur-Containing Odorants in TS
A total of 51 odorants were detected by SPME in the static headspace of TS (Table 3). Twenty-
two of volatile compounds detected in SPME extracts were also found in static headspace and SAFE
isolates. In agreement with the results of SHDA, hydrogen sulfide, acetaldehyde, 2-methylpropanal,
allyl mercaptan, 2,5-dimethylthiophene, an unknown (no. 51; sulfurous, alliaceous) and (E,E)-bis-1-
(1-propenyl) trisulfide were also detected in SPME extracts. Whereas, 2,4-dimethylthiophene, 2-
16

isobutyl-3-methoxypyrazine, -farnesene, eugenol, (Z)-isoeugenol and (E)-isoeugenol were detected
in both SPME extracts and SAFE isolates. Therefore, results of SPME were in good general
agreement with the SHDA and SAFE results, since SPME mainly measured volatiles with high and
intermediate volatility; however, some odorants with low volatilities were also detected.
Sixteen odorants were detected only by the SPME. The included 10 sulfur-containing compounds.
Noteworthy among these were 3,4-dimethylthiophene, (Z)-2-mercapto-3,4-dimethyl-2,3-
dihydrothiophene and (E)-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene. Liu et al (2013) had
previously reported the above four compounds in TS by SPME-GC analysis.
In the present study (E,E)-bis-(1-propenyl) disulfide, (E,Z)-bis-(1-propenyl) disulfide and (Z,Z)-
bis-(1-propenyl) disulfide were positively identified by their RIs, odor descriptions, mass spectra and
by comparison with synthesized standard compounds. They were further comfirmed by GC-FPD
analysis (by both SAFE and SPME). Previously, the bis-(1-propenyl) disulfides were only tentatively
identified in TS seedlings and tender shoots by comparison of their mass spectra to those published
in a mass spectral library database (Li et al, 2017).
The bis-(1-propenyl) disulfides (nos. 46-48), which are common components of Allium distilled
oils, have been reported to rearrange at 85 °C into 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene,
which could then futher degraded to form 3,4-dimethylthiophene or 3,4-dimethyl-2-thienyl disulfides
(Figure 2; Block, Bayer, Naganathan, & Zhao, 1996). These researchers found that significant
qualities of 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene and 3,4-dimethylthiophene were formed
when GC injector temperatures of 250-280 °C were used, but were not detected when a lower injector
temperature of 120 °C was used (Block, Putman, & Zhao, 1992). In the above reaction, a dithio-
Claisen rearrangement was proposed as the key step (Block, & Shu, 1990). Elsewhere, 2,4-
dimethylthiophene, 3,4-dimethylthiophene (no. 31), 2,5-dimethylthiophene, (Z)-mercapto-3,4-
dimethyl-2,3-dihydrothiophene (no. 63) and (E)-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (no.
67) were identified in distilled oils of Welsh onions and scallions (Kuo, & Ho, 1992).
17

We performed additional experiments to better understand the effect of GC injection port
temperature on the stability and formation of sulfur-containing compounds found in TS. For this
comparison the temperature of MSD transfer line was held constant at 200 °C. It was found that the
abundance of compounds nos. 31, 63 and 67 increased significantly as a function of increasing
injection port temperature from 175 °C to 250 °C (Table 4); whereas, compounds nos. 46, 47 and 48
all decreased. At all injection temperatures, (E,E)-bis-(1-propenyl) disulfide and (E,Z)-bis-(1-
propenyl) were always detected at higher abundance than (Z,Z)-bis-(1-propenyl) disulfide.
Furthermore, (Z,Z)-bis-(1-propenyl) disulfide was not detected when an inlet temperature above
225 °C was used.
These above three compounds (nos. 31, 63 and 67) could be detected when inlet injector
temperature was 175 °C, which meant that the bis-(1-propenyl) disulfides (nos. 46-48) had degraded.
When an inlet temperature at 150 °C was applied for GC-O/FPD (SAFE and SPME), peaks for these
compounds (nos. 31, 63 and 67) were still detected in raw and cooked TS, and the corresponding
odors could be recognized by GC-O.
That fact that the disulfides in TS are thermally labile makes procedures that rely on thermal
desorption, such as SPME which requires hot split/splitless injection, unrealiable. A better
alterntative to avoid thermal decomposition of the disulfides would be to use technique that utilizes
low temperature injection, such as the analysis of aroma extracts by cool on-column injection or static
headspace analysis using cyrogenic solvent-vent mode injection. This explains why compounds nos.
31, 63 and 67 were not detected by SHDA and AEDA, while instead bis-(1-propenyl) disulfides (nos.
46-48) were identified with the highest FD-factors in raw and cooked TS. Therefore, results of SHDA
and AEDA better represent the true aroma profiles of raw and cooked TS.
Volatile sulfur compounds clearly made the greatest contribution to the characteristic aroma of TS.
In particular, (E,E)-bis-(1-propenyl) disulfide, (E,Z)-bis-(1-propenyl) disulfide and (Z,Z)-bis-(1-
propenyl) disulfide appear to make the greatest contribution to the overall aroma profile of TS,
especially cooked TS. Bis-(1-propenyl) sulfides and (E,E)-bis-(1-propenyl) trisulfide were also
18

identified with low FD factors. The nonvolatile precursors for these compounds were previously
identified as (S,S)-γ-glutamyl-(Z-S-1-propenyl)thioglycine, (S,S)-γ-glutamyl-(E-S-1-propenyl)-
thioglycine, γ-glutamyl-(Z-S-1-propenyl)-cysteine, γ-glutamyl-(E-S-1-propenyl)-cysteine, Z-S-1-
propenyl-L-cycsteine and E-S-1-propenyl-L-cycsteine (Li et al., 2013). Hydrogen sulfide formed by
Strecker degradation from cysteine, was detected in cooked TS at a high FD factor by SHDA
Hydrogen sulfide is highly volatile compound that can easily be recognized even at trace
concentrations, which might be the one reason why cooked TS has an unique odor after cooking.
Allyl mercaptan and methyl thiirane were identified in raw and cooked TS, and methyl thiirane had
a higher FD-factor in cooked TS as it could be formed during blanching. The (Z)- and (E)-1-propenyl
mercaptans are isomers of allyl mercaptan and methyl thiirane, but they weren’t detected in raw and
cooked TS. These two compounds are believed to be very unstable precursors of methyl thiirane and
bis-(1-propenyl) sulfides (Li et al., 2013).
During the Strecker degradation, 2-methylbutanal, 3-methylbutanal and methional can be formed
from leucine, isoleucine and methionine, respectively (Parker, 2014). Methional can undergo
subsequent degradation to acrolein and methanthiol. Methanthiol easily oxidises to form dimethyl
disulfide (Granvogl, Beksan, & Schieberle, 2012). Dimethyl sulfide can also be formed from a
thermal degradation of S-methylmethionine (vitamin U), which is the S-methylated form of
methionine and a non-proteinogenic amino acid that occurs in various vegetables (Scherb, Kreissl,
Haupt, & Schieberle, 2009). Methional and dimethyl sulfide were also identified in tomato paste
(Buttery, Teranishi, Ling, & Turnbaugh, 1990) and cooked asparagus (Ulrich, Hoberg, Bittner,
Engewald, & Meilchen, 2001).
In general, thiazoles are mainly formed by non-enzymatic browning reactions between reducing
sugars and amino acids in the presence of hydrogen sulfide. 2,4,5-Trimethylthiazole was detected in
TS and contributed a sulfurous, vegetable odor note to overall aroma profile. This compound can be
generated form the reaction between ascorbic acid and cysteine (Yu, & Zhang, 2010).
3.4.2 Potent Green, Grassy and Leafy Smelling Compounds in TS
19

Several C6-aldehydes and C6-alcohols contributed green, grassy, and leafy aroma notes to TS,
especially to raw TS. These included hexanal (grassy, leafy), (Z)-3-hexenal (grassy, leafy), (E)-2-
hexenal (green, leafy), 1-hexanol (green) and (Z)-3-hexen-1-ol (green, leafy). These volatile
compounds are synthesized in green leaves from -linolenic and linoleic acids bia biosynthetic
pathways (lipoxygenase and hydroperoxide lyase) (Hatanaka, 1993). The metabolism of
polyunsaturated fatty acids through the lipoxygenase pathway is known to generate of the “green”
aroma notes in many vegetables and fruits (Raffo, Masci, Moneta, Nicoli, Pulgar, & Paoletti, 2018).
3.4.3 Potent Sweet Compounds in TS
Several phenolic compounds with herbal, sweet, cloves and vanilla-like aroma notes were
identified as potent odorants in TS. These included cinnamyl acetate (sweet, spicy), eugenol (sweet,
woody, cloves), p-vinylguaiacol (sweet, woody, cloves), (Z)- and (E)-isoeugenol (sweet, spicy,
cloves), phenylacetic acid (floral, rose) and vanillin (sweet, caramel). These compounds had high FD
factors in both raw and cooked TS. None of these compounds have been previously identified in TS,
especially in cooked TS. Some of these compounds might originate from degradation of ferulic acid
(Fiddler, Parker, Wasserman, & Doerr, 2002). Phenylacetic acid was possibly formed via the Strecker
degradation of phenylalanine (Hofmann, Münch, & Schieberle, 2000). Previously, some phenolic
antioxidants were identified in TS (fresh young leaves and shoots), including gallic acid and its
derivatives, gallotannins and flavonoids (Wang, Yang, & Zhang, 2007; Yang, Wang, Xing, Dai, &
Chen, 2011). However, the odorants with sweet notes and high FD-factors couldn’t be perceived
during sensory aroma analysis. It could be these compounds have high boiling points, such as,
cinnamyl acetate (265 °C), eugenol (254 °C), p-vinylguaiacol (224 °C), isoeugenol (266 °C),
phenylacetic acid (265.5 °C ) and vanillin (285 °C), or they are at low concentrations such that their
odors could be suppressed by other high abundance volatile compounds.
Linalool (floral, sweet, lemon) had high odor potency in raw TS. This odorant is reported as one
of the most aroma-active compounds contributing to the overall flavor of orange peel oil (Dharmawan,
20

Kasapis, Sriramula, Lear, & Curran, 2009; Fischer, & Schieberle, 2009). It is also a major component
of the beverage prepared from Darjeeling black tea (Schuh, & Schieberle, 2006). β-Caryophyllene
(sweet, woody) had high FD factors in raw TS (FD = 729) and cooked TS (FD = 2187), and this
compound was also identified in TS by microwave-assisted extraction-(MAE)-HS-SPME followed
by GC–MS and was in highest abundance among 45 compounds reported (Mu, Wang, Liu,
Yuan,Wang, & Fan, 2007).
4. Conclusions
The study represents the first comprehensive determination of the potent aroma components of raw
and cooked TS, and clearly demonstrates the importance of sulfur-containing odorants, especially,
(E,E)-bis-(1-propenyl) disulfide and (E,Z)-bis-(1-propenyl) disulfide, in the overall aroma of cooked
TS. These and other sulfur-containing compounds contribute cooked meat, alliaceous and pungent
aroma notes, while other compounds provide green/grassy, woody and sweet aroma notes.
Collectively, these potent odorants produce the unique and characteristic aroma profile of TS.
Additional studies should be considered, including quantitative analysis and determinination of
odor-activity values of potent odorants along with aroma recombination studies. Meanwhile, the
thermal degradation for some sulfur compounds in TS should be further investigated during cooking
as well as during the aroma isolation procedures and instrumental analyses.
Conflict of interest statement
The authors declare no conflict of interest.
Acknowledgement
The authors gratefully acknowledge the Open Project Program of Beijing Key Laboratory of Flavor
Chemistry, Beijing Technology and Business University (BTBU), and the National Key Research
and Development Program (2017YFD0400501) for their financial support.
21

Appendix A. Supplementary data
Supplementary data to this article can be found online at
Figures showing standard mass spectras for (Z,Z)/(E,Z)/(E,E)-bis (1-propenyl) disulfide isomers
(Figure S1), mass spectras comparision between samples’ and standard databases’ in NIST 14 (Figure
S2); table containing data for all compounds listed in Table 1, 2, 3 and 4 (Table S1).
References
Block, E., & Shu, H. Z. (1990). Onion essential oil chemistry. Cis-and trans-2-mercapto-3,4-
dimethyl 2,3-dihydrothiophene from pyrolysis of bis(1-propenyl) disulfide. Tetrahedron
Letters, 31(35), 4999–5002. http://doi.org/10.1016/s0040-4039(00)97788-8
Block, E., Putman, D., & Zhao, S.-H. (1992). Allium chemistry: GC-MS analysis of thiosulfinates
and related compounds from onion, leek, scallion, shallot, chive, and Chinese chive. Journal of
Agricultural and Food Chemistry, 40(12), 2431–2438. http://doi.org/10.1021/jf00024a018
Block, E., Bayer, T., Naganathan, S., & Zhao, S.-H. (1996). AlliumChemistry: Synthesis and
sigmatropic rearrangements of alk(en)yl 1-propenyl disulfide S-oxides from cut onion and
garlic, Journal of the American Chemical Society, 118(12), 2799–2810.
http://doi.org/10.1021/ja953444h
Buttery, R. G., Teranishi, R., Ling, L. C., & Turnbaugh, J. G. (1990). Quantitative and sensory
studies on tomato paste volatiles. Journal of Agricultural and Food Chemistry, 38(1), 336–
340. http://doi.org/10.1021/jf00091a074
Cadwallader, K., Potts, D., Brisske-BeVier, L., & Mirarefi, S. (2011). Contribution of volatile
sulfur compounds to the characteristic aroma of roasted garlic. Volatile Sulfur Compounds in
Food (Ed. Vol. 1068, pp. 137–151). Washington, DC: American Chemical Society. (Chapter
7.) http://doi.org/10.1021/bk-2011-1068.ch007
Dharmawan, J., Kasapis, S., Sriramula, P., Lear, M. J., & Curran, P. (2009). Evaluation of aroma-
active compounds in pontianak orange peel oil (Citrus nobilis Lour. Var. microcarpaHassk.)
by gas chromatography−olfactometry, aroma reconstitution, and omission test. Journal of
Agricultural and Food Chemistry, 57(1), 239–244. http://doi.org/10.1021/jf801070r
Fiddler, W., Parker, W. E., Wasserman, A. E., & Doerr, R. C. (2002). Thermal decomposition of
ferulic acid. Journal of Agricultural and Food Chemistry, 15(5), 757–761.
http://doi.org/10.1021/jf60153a003
Fischer, A., & Schieberle, P. (2009). Characterisation of the key aroma compounds in the peel oil
of Pontianak oranges (Citrus nobilis Lour. var. microcarpa Hassk.) by aroma reconstitution
experiments. European Food Research and Technology, 229(2), 319–328.
http://doi.org/10.1007/s00217-009-1058-4
Granvogl, M., Beksan, E., & Schieberle, P. (2012). New insights into the formation of aroma-active
strecker aldehydes from 3-oxazolines as transient intermediates. Journal of Agricultural and
Food Chemistry, 60(25), 6312–6322. http://doi.org/10.1021/jf301489j
Grosch, W. (1994). Determination of potent odourants in foods by aroma extract dilution analysis
(AEDA) and calculation of odour activity values (OAVs). Flavour and Fragrance Journal, 9,
147–158.
Hatanaka, A. (1993). The biogeneration of green odour by green leaves. Phytochemistry, 34(5),
1201–1218. http://doi.org/10.1016/0031-9422(91)80003-J
22

Hofmann, T., Münch, P., & Schieberle, P. (2000). Quantitative model studies on the formation of
aroma-active aldehydes and acids by strecker-type reactions. Journal of Agricultural and Food
Chemistry, 48(2), 434–440. http://doi.org/10.1021/jf990954c
Hsu, C.-Y., Huang, P.-L., Chen, C.-M., Mao, C.-T., & Chaw, S.-M. (2012). Tangy scent in Toona
sinensis (Meliaceae) leaflets: Isolation, functional characterization, and regulation of TsTPS1
and TsTPS2, two key terpene synthase genes in the biosynthesis of the scent compound.
Current Pharmaceutical Biotechnology, 13(15), 2721–2732.
http://doi.org/10.2174/138920112804724864
Hu, J., Song, Y., Mao, X., Wang, Z.-J., & Zhao, Q.-J. (2016). Limonoids isolated from Toonaand
their radical scavenging, anti-inflammatory and cytotoxic activities. Journal of Functional
Foods, 20(C), 1–9. http://doi.org/10.1016/j.jff.2015.10.009
Kuo, M. C., & Ho, C. T. (1992). Volatile constituents of the distilled oils of Welsh onions (Allium
fistulosum L. variety maichuon) and scallions (Allium fistulosum L. variety caespitosum).
Journal of Agricultural and Food Chemistry, 40(1), 111–117.
http://doi.org/10.1021/jf00013a021
Lapsongphon, N., Yongsawatdigul, J., & Cadwallader, K. R. (2015). Identification and
Characterization of the Aroma-Impact Components of Thai Fish Sauce. Journal of
Agricultural and Food Chemistry, 63(10), 2628–2638. http://doi.org/10.1021/jf5061248
Li, J., Wang, J., Dai, Y., Su, C., & Chen, M. (2011). Composition of characteristic aroma of Toona
sinensis and its change during storage. Journal of Beijing Forestry University, 33(3), 127–131.
http://doi.org/10.13332/j.1000-1522.2011.03.013
Li, J.-X., Eidman, K., Gan, X.-W., Haefliger, O. P., Carroll, P. J., & Pika, J. (2013). Identification
of (S,S)-γ-glutamyl-(cis-S-1-propenyl)thioglycine, a naturally occurring norcysteine
derivative, from the Chinese vegetable Toona sinensis. Journal of Agricultural and Food
Chemistry, 61(31), 7470–7476. http://doi.org/10.1021/jf401946d
Li, N., Wang, H., Zhao, F., & Liu, C. (2017). Effects of sulfur compounds of Toona sinensis
seedlings on the aroma compounds. Food Research and Development, 38(13), 35–39.
http://doi.org/10.3969/j.issn.1005-6521.2017.13.008
Liu, C., Zhang, J., Zhou, Z., Hua, Z., Wan, H., Xie, Y., et al. (2013). Analysis of volatile
compounds and identification of characteristic aroma components of Toona sinensis (A. Juss.)
Roem. Using GC-MS and GC-O. Food and Nutrition Sciences, 04(03), 305–314.
http://doi.org/10.4236/fns.2013.43041
Meilgaard, M., Civille, G. V., & Carr, B. T. (1999). Sensory Evaluation Techniques,
(3rd ed.). CRC Press LLC: Boca Raton, FL.
Mu, R., Wang, X., Liu, S., Yuan, X., Wang, S., & Fan, Z. (2007). Rapid determination of volatile
compounds in Toona sinensis (A. Juss.) Roem. by MAE-HS-SPME followed by GC–MS.
Chromatographia, 65(7-8), 463–467. http://doi.org/10.1365/s10337-007-0183-0
Parker, J. K. (2014). Thermal generation or aroma. Flavour Development, Analysis and Perception
in Food and Beverages (Ed. pp. 151–185). Elsevier Ltd. (Chapter 8.)
http://doi.org/10.1016/B978-1-78242-103-0.00008-4
Raffo, A., Masci, M., Moneta, E., Nicoli, S., Pulgar, J. S., & Paoletti, F. (2018). Characterization of
volatiles and identification of odor-active compounds of rocket leaves. Food Chemistry, 240,
1161–1170. http://doi.org/10.1016/j.foodchem.2017.08.009
Rotsatchakul, P., Chaiseri, S., & Cadwallader, K. R. (2008). Identification of Characteristic Aroma
Components of Thai Fried Chili Paste. Journal of Agricultural and Food Chemistry, 56(2),
528–536. http://doi.org/10.1021/jf072499n
Scherb, J., Kreissl, J., Haupt, S., & Schieberle, P. (2009). Quantitation of S-methylmethionine in
raw vegetables and green malt by a stable isotope dilution assay using LC-MS/MS:
comparison with dimethyl sulfide formation after heat treatment. Journal of Agricultural and
Food Chemistry, 57(19), 9091–9096. http://doi.org/10.1021/jf901765q
Schuh, C., & Schieberle, P. (2006). Characterization of the Key Aroma Compounds in the
Beverage Prepared from Darjeeling Black Tea: Quantitative Differences between Tea Leaves
23

and Infusion. Journal of Agricultural and Food Chemistry, 54(3), 916–924.
http://doi.org/10.1021/jf052495n
Ulrich, D., Hoberg, E., Bittner, T., Engewald, W., & Meilchen, K. (2001). Contribution of volatile
compounds to the flavor of cooked asparagus. European Food Research and Technology,
213(3), 200–204. http://doi.org/10.1007/s002170100349
Van Den Dool, H., & Kratz, P. D. (1963). A generalization of the retention index system including
linear temperature programmed gas-liquid partition chromatography. Journal of
Chromatography, 11, 463–471.
Wang, K.-J., Yang, C.-R., & Zhang, Y.-J. (2007). Phenolic antioxidants from Chinese toon (fresh
young leaves and shoots of Toona sinensis). Food Chemistry, 101(1), 365–371.
http://doi.org/10.1016/j.foodchem.2006.01.044
Yang, Y., Wang, J., Xing, Z.-E., Dai, Y.-Q., & Chen, M. (2011). Identification of phenolics in
Chinese toon and analysis of their content changes during storage. Food Chemistry, 128(4),
831–838. http://doi.org/10.1016/j.foodchem.2011.03.071
Yu, A.-N., & Zhang, A.-D. (2010). Aroma compounds generated from thermal reaction of l-
ascorbic acid with l-cysteine. Food Chemistry, 121(4), 1060–1065.
http://doi.org/10.1016/j.foodchem.2010.01.049
Yu, W.-J., Chang, C.-C., Kuo, T.-F., Tsai, T.-C., & Chang, S.-J. (2012). Toona sinensis Roem leaf
extracts improve antioxidant activity in the liver of rats under oxidative stress. Food and
Chemical Toxicology, 50(6), 1860–1865. http://doi.org/10.1016/j.fct.2012.03.068
Zhang, W., Li, C., You, L.-J., Fu, X., Chen, Y.-S., & Luo, Y.-Q. (2014). Structural identification of
compounds from Toona sinensis leaves with antioxidant and anticancer activities. Journal of
Functional Foods, 10(C), 427–435. http://doi.org/10.1016/j.jff.2014.07.015
Zhou, Q., Wintersteen, C. L., & Cadwallader, K. R. (2002). Zhou, Q., Wintersteen, C. L., &
Cadwallader, K. R. (2002). Identification and quantification of aroma-active components that
contribute to the distinct malty flavor of buckwheat honey. Journal of Agricultural and Food
Chemistry, 50(7), 2016–2021. http://doi.org/10.1021/jf011436g
overall intensity
15
raw TS
12
cooked TS
9
pungent
green, grassy
6
3
0
earthy
cooked meat
alliaceous
24

Figure 1. Sensory aroma profile comparison of raw and cooked Toona sinensis.
[Intensity from 0 to 15, where 0 = none, 15 = very strong.]
Figure 2. Thermal degradation of bis-(1-propenyl) disulfides (nos. 46-48 ) to form
compounds nos. 31, 63 and 67.
25

Table 1. Odorants Detected by GCO-Static Headspace Dilution Analysis of Toona sinensis
FD-factor d
raw Toona sinensis cooked Toona sinensis
retention index b
no. a
compound
odor description c
RTX-WAX
<500
752
RXI-5MS
<500
517
RTX-WAX
RXI-5MS
RTX-WAX
RXI-5MS
1
hydrogen sulfide f
rotten egg
625
25
25
25
25
25
25
25
25
25
25
25
5
125
25
25
25
25
25
1
5
46
11
37
42
26
33
61
9
dimethyl sulfide i
(E,E)-bis-(1-propenyl) disulfide h
methyl thiirane i
2-isopropyl-4-methylthiazole i
dimethyl trisulfide i
2,5-dimethylthiophene i
dibutyl sulfide i
β-caryophyllene i
unknown
pungent, sweet, corn
sulfurous, cooked onion
alliaceous, pungent, garlic
sulfurous
25
25
5
25
25
1
1450
916
1108
587
1343
1376
1158
1263
1586
854
1017
986
1
1
sulfurous, cooked cabbage
sulfurous, garlic
sulfurous, green
1
1
869
1
1
sweet, woody
alliaceous, garlic
sulfurous, alliaceous
floral, sweet, lemon
pungent, yogurt
51
55
3
1490
1532
617
unknown
linalool i
<500
801
acetaldehyde i
5
1
25
5
-pinene i
18
4
1011
716
pine
5
ethanethiol i
sulfurous, pungent
sulfurous, stink, garlic
grassy, leafy
5
13
24
924
thiophene i
5
1141
(Z)-3-hexenal i
25
5
1
25

21
7
1083
823
801
547
hexanal i
grassy, leafy
pungent
5
5
5
5
1
1
25
1
2-methylpropanal i
22
1090
813
1-pentanethiol i
sulfurous, pungent, baked onion
1
1
1
41
45
15
10
49
29
40
66
75
27
48
19
20
65
71
87
12
32
50
1363
1447
971
991
908
570
2,4,5-trimethylthiazole m
methional i
sulfurous, vegetable
cooked potato
1
1
5
1
1
1
1
1
1
1
1
1
1
1
2,3-butanedione i
buttery, creamy
5
5
1
1
879
allyl mercaptan i
sulfurous, alliaceous
sulfurous, alliaceous, garlic
sulfurous, savory, baked onion
green
1480
1183
1356
1082
1692
diallyl disulfide i
1
bis-(1-propenyl) sulfide h
1-hexanol i
borneol i
woody, pine, sweet
raw garlic
1854
1075
2-vinyl-4H-1,3-dithiine h
-terpinene i
1002
1120
738
earthy, woody
25
1
5
5
1
1
1
1
1
(Z,Z)-bis-(1-propenyl) disulfide h
dimethyl disulfide i
dipropyl sulfide i
sulfurous, alliaceous, meaty
sulfurous, alliaceous, cabbage
sulfurous, garlic
1
1
895
1
1173
1332
1283
633
2-methyl-3-(methylthio) furan i
(E,E)-bis-(1-propenyl) trisulfide h
phenylacetic acid i
2/3-methylbutanal i
diisopropyl disulfide i
(Z)-3-hexen-1-ol i
sulfurous, vitamin B
sweet, pungent
1
1
floral, rose
915
1241
1492
malty, dark chocolate
sulfurous, meaty
green, leafy
25
5
1
5
27

30
35
59
2
1212
1299
1577
<500
854
979
(E)-2-hexenal i
green, leafy
1
1
1
1
1
1
1
1-octen-3-one i
mushroom
1154
(E,Z)-2,6-nonadien-1-al i
sweet, cucumber
methanethiol f
sulfurous, rotten
b
c
a Numbers correspond to those in Tables 2, 3 and 4. Retention indices determined by GCO on two different stationary phases (RTX-Wax and RXI-5MS). Odor
quality as perceived during GCO. d Flavor dilution (FD) factor=highest headspace volume tested (25 mL) divided by the lowest headspace volume required to detect a
e
compound (i.e., 5, 1, 0.2, or 0.04 mL). Compound was positively identified, based on retention index (RI); odor quality (O); mass spectra (MS); reference standard
compound (S) and sulfur detector (SD). Compound was tentatively identified by O. g Compound was tentatively identified by O, SD. h Compound was tentatively
f
identified by RI, O. i Compound was tentatively identified by RI, O, S. j Compound was tentatively identified by RI, O, MS. k Compound was tentatively identified by
RI, O, SD. l Compound was tentatively identified by RI, O, S, SD. m Compound was tentatively identified by RI, O, S, MS. Compound was tentatively identified by
n
RI, O, SD, MS.
Table 2. Aroma-Active Compounds in Toona sinensis Determined by Aroma Extract Dilution Analysis
FD-factor d
retention index b
no. a
compound
odor description c
raw Toona sinensis
RTX-WAX RXI-5MS
19683 243
cooked Toona sinensis
RTX-WAX
RXI-5MS
RTX-WAX
RXI-5MS
46
47
1450
1104
(E,E)-bis-(1-propenyl) disulfide k
(E,Z)-bis-(1-propenyl) disulfide k
sulfurous, cooked onion
59049
729
sulfurous, cooked onion, roasted
coffee
1473
1117
9
9
6561
81
61
85
87
88
41
28
1582
2235
2524
2524
1369
1187
1410
1411
1277
1384
989
β-caryophyllene m
(Z)-isoeugenol i
sweet, woody
729
729
729
729
81
1
1
2187
2187
2187
2187
2187
2187
81
81
27
sweet, spicy, cloves
floral, rose
phenylacetic acid i
vanillin i
9
sweet, caramel
27
3
2,4,5-trimethylthiazole m
2,4-dimethylthiophene h
sulfurous, vegetable
sulfurous, pungent, meaty
81
27
28

66
82
50
72
11
83
78
45
86
44
1688
2120
1495
1774
916
1170
962
borneol i
woody, pine, sweet
sweet, spicy
3
243
27
27
1
27
3
2187
729
729
729
729
243
243
81
243
cinnamyl acetate h
(Z)-3-hexen-1-ol i
unknown
green, leafy
sweet, woody, fruity, minty
alliaceous, pungent, garlic
sweet, woody, cloves
metallic, woody
methyl thiirane l
eugenol i
2149
1990
1443
2331
1407
1348
1373
904
81
27
243
27
27
9
(E)-4,5-expoxy-(E)-2-decenal h
methional l
3
1
cooked potato
27
9
1442
(E)-isoeugenol i
unknown
sweet, spicy, cloves
sulfurous, alliaceous, cooked
81
81
79
2005
4-hydroxy-2,5-dimethyl-3(2H)-furanone i
pungent, sweet, caramel
27
9
81
81
29
69
35
18
25
53
33
58
55
80
24
2066
1211
1745
1296
1032
1150
1521
1278
1558
1543
2024
1150
unknown
sulfurous, garlic
sulfurous
81
81
27
27
27
9
bis- (1-propenyl) sulfide h
-farnesene j
sulfurous, bell pepper
mushroom
27
3
986
919
1-octen-3-one i
-pinene m
1
1
1
1
pine
2-methyl-2-pentenal i
2-isobutyl-3-methoxypyrazine h
dibutyl sulfide i
unknown
green, juicy
81
27
1191
1081
sulfurous, bell pepper
sulfurous, green
woody, pine
9
9
9
3
1
9
729
243
243
81
3
1096
801
linalool i
floral, sweet, lemon
pungent, sweaty
grassy, leafy
9
3
9
3
unknown
3
(Z)-3-hexenal m
243
3
29