Abstract
Changes in volatile compounds during early storage of an aseptic-cup (AC) coffee beverage without milk and sugar were examined. AC coffee samples were stored at 10°C for 0, 1, and 2 weeks. Retronasal aroma (RA) compounds of the AC samples were collected using a retronasal aroma simulator (RAS) coupled to a solidphase microextraction (SPME) fiber and then analyzed using gas chromatography-olfactometry (GC-O) and gas chromatography-mass spectrometry (GC-MS). The GC-O analysis detected 17 odor compounds. Changes in the compounds during 2-week storage were quantitatively analyzed using selected ion monitoring GC/MS. Statistical analyses of the peak areas showed that eight compounds decreased significantly (p < 0.05) during early storage, and that trends of the decreases differed among the compounds. Furthermore, a duo-trio test using an analytical sensory panel confirmed that the aroma of the 2-week sample was similar to that of the 0-week sample upon addition of the model flavor (composed of the eight compounds) to the 2-week sample.
Introduction
Carton packs such as Tetra Brik Aseptic (TBA) and SIG Combibloc are used worldwide for the production of long-life aseptic coffee beverages. On the other hand, a chilled-cup coffee beverage that is aseptically filled in a plastic cup with a straw is commercially available on the Japanese coffee beverage market. The products have a shelf life of more than 60 days at temperatures under 10°C. In addition to its pleasant flavor, owing to desirable manufacturing and distribution conditions such as ultra-high- temperature (UHT) sterilization, the new package style is one of the reasons for the popularity of chilled-cup coffee beverages.
A chilled-cup coffee beverage is usually consumed using a straw. Therefore, retronasal aroma (RA), which is caused by flavor compounds traveling from the mouth to the nasal cavity via the nasopharynx and the lungs, is important for flavor perception. A retronasal aroma simulator (RAS) was developed by Roberts and Acree (1995) for use as one of various mouth models (van Ruth et al., 1994; Roberts and Acree, 1995; Elmore and Langley, 1996). These are useful tools for the study of RAs, in addition to other methods such as breath-by-breath analysis (Graus et al., 2001) and oral breath sampling (OBS) (Roozen and Legger-Huysman, 1995). The RAS has been used for analyses of coffee aroma (Akiyama et al., 2009; Michishita et al., 2010).
Some studies on quality changes of TBA citrus juice and juice drink products during long-term storage were reported (Kacem et al., 1987; Roig et al., 1999). Pérez-Martmez et al. (2008) investigated changes in the volatile fraction of coffee brews stored at 4 and 25°C for 30 days. However, changes in the quality of aseptic-cup (AC) coffee beverages during storage remain to be clarified. Moreover, changes in flavor during storage of AC coffee beverages are an important issue, as these products are often stored for a long period of time before consumption. Changes in flavor components of AC coffee beverages without milk (black coffee) during storage are considered to be greater than those of coffee beverages with milk; however, the quantitative determination of flavor changes of black coffee alone is insufficient. Therefore, quantitative sensory analysis, gas chromatography (GC) analysis, taste sensing analysis, and electronic nose (E-nose) analysis were carried out in order to characterize the quality changes of AC black coffee beverages during long-term storage of 2 months. Gas chromatography-mass spectrometry (GC-MS) analysis revealed that 12 RAS volatile compounds significantly (p < 0.05) decreased during storage (0 2 months). Consequently, the analysis showed that the quality change between 0- and 1-month samples was greater than that between samples stored for 1 and 2 months, and also suggested that the quality change between 0- and 2-week samples was remarkably large (Watanabe et al., 2012).
This study was conducted to reveal the changes in RAS compounds of a chilled-cup coffee beverage during early-stage storage, and to provide essential information for preserving the freshness of AC coffee beverages. First, RAS odor compounds of the stored samples were detected using gas chromatography- olfactometry (GC-O). Secondly, changes in the odor compounds during early storage were quantitatively analyzed using GC-MS. Finally, the effects of compounds that showed significant changes during storage on the aroma characteristics of an AC coffee beverage were confirmed by sensory evaluation.
Materials and Methods
Aseptic-cup coffee beverage Using an industrial-scale column extractor at a plant (Morinaga Milk Industry Co., Ltd., Tokyo, Japan), a coffee extract of 2.2°Bx was brewed from roasted arabica (Coffea arabica) coffee beans (L value = 16) with reverse osmosis (RO) hot water (90C). The extract was immediately cooled to 10°C and then diluted to a solute concentration of 1.2° Bx with RO water. Additionally, coffee flavors were added to the diluted extracts. After sterilization under ultra-high temperature processing (at 120°C for 2 seconds), the coffee sample (pH 5.5), which contained more than 5.0% coffee component in terms of raw beans, was aseptically filled to a plastic cup with a plastic straw, an aluminum (AL) inner lid, and a plastic overcap at room temperature.
The specific dimensions of the cups containing 240 mL of the sample were as follows: bottom diameter, 54 mm; top diameter, 68 mm; and height, 118 mm. The headspace in the cups was replaced with N2 gas immediately after filling with the coffee sample. The cup had gas-barrier and shading properties, the latter provided by a paper-based label wound around the cup. The layered components of the cup and lid are shown as follows:
Cup layer: (outside) PS/adhesive agent/EVOH/adhesive agent/ PS/adhesive agent/PET (inside)
Lid layer: (outside) AL/adhesive agent/sealant film (inside)
The above abbreviations represent the following: PS, polystyrene; EVOH, ethylene vinyl alcohol copolymer; PET, polyethylene terephthalate.
The samples were then cooled to 10°C. Samples of 0-week storage were the samples that were prepared on a plant scale 3 days before analysis. Samples of 1- and 2-week storage were stored at 10°C for 1 or 2 weeks after the analysis of the 0-week samples.
SPME device For sampling RAS volatile compounds of the coffee beverage, SUPELCO DVB/Carboxen/PDMS fiber (50/30 μm thickness) was selected from six types of SPME fibers for use with the SPME device (Sigma-Aldrich Co., St. Louis, MO, USA) as described in a previous study on freshly brewed drip coffee aroma (Akiyama et al., 2007). Before RAS sampling, the SPME fiber was reconditioned in fiber conditioner (GL Sciences Inc., Tokyo, Japan) according to the manufacturer's instructions (data sheet T7941231; Sigma-Aldrich Co.).
RAS parameters and SPME sampling of RAS volatiles A method using an RAS (Roberts and Acree, 1995) was used to sample retronasal aroma. The RAS consisted of a 1-L stainless steel blending container with a water jacket for controlling the temperature (38°C), a voltage controller and high torque-speed motor to precisely control the rotational speed of the shear blade (650 rpm), simulating a model mouth, a controlled nitrogen gas supply (1000 mL/min) as a carrier gas to sweep over the stored samples (200 mL), and artificial saliva (40 mL) consisting of 20.0 mmol/L NaHCO3, 2.75 mmol/L K2 HPO4, 12.2 mmol/L KH2 PO4, and 15.0 mmol/L NaCl. RA compounds of AC samples stored for 0, 1, and 2 weeks were collected by exposing a SPME fiber to the effluent gas, using a RAS coupled to a SPME fiber of 2 cm length. After sampling for 2 min, the SPME fiber was injected into the GC-O or GC-MS inlet port and the analyte was thermally desorbed for 10 min at 225°C or 250°C, respectively (Akiyama et al., 2009). Each SPME sample was conducted in triplicate.
GC-O analysisThe GC-O analysis by a dilution method which involves varying the split ratio of GC injection was developed by Deibler et al. (2004), and has been successfully applied to GC-O analysis of the aromas of freshly brewed espresso coffee (Akiyama et al., 2009). Volatile compounds of the stored sample in the effluent gas of the RAS were thus collected using SPME fibers. After sampling, the fibers were placed into the injection port of the GC-O, and thermally desorbed for 10 min at 225°C at different dilution ratios (1, 1/3, and 1/9) using three different settings (1:0, 1:3.2, and 1:16.3). GC-O analysis was conducted in triplicate using CharmAnalysisTM on a 6890 GC (Agilent Technologies, Santa Clara, CA, USA) modified by DATU Inc. (Geneva, NY, USA) (Acree et al., 1984). A fused silica capillary column DB-WAX (15 m x 0.32 mm i.d., 0.25 μm film thickness; Agilent Technologies) was used, with a helium carrier gas flow of 3.2 mL/min. The oven temperature was set to an initial temperature of 40°C, increased at 6°C/min to 230°C, and held for 30 min. The injection and detector ports were maintained at 225°C, and the injection purge on the GC was off for the initial 1 min. The retention time of each compound was converted to Kovats indices using C6-C28 n-alkanes (Tokyo Kasei Kogyo Co., Ltd., Tokyo, Japan).
GC-O evaluation During GC-O analysis using CharmAnalysisTM, a trained sniffer sniffs the air emitting from the olfactometer and holds down the mouse button when he or she detects an odor in the air, and the computer records the perception, time, and duration of the sensation (Acree et al., 1984). Acidic, buttery-oily, green-blackcurrant, green-earthy, nutty-roast, phenolic, smoke- roast, soy sauce, sweet-caramel, and sweet-fruity were the aroma descriptors used in all GC-O experiments to describe potent odor- ants. These descriptors were chosen from the results of the previous study (Akiyama et al., 2009). The sensory significance of each odorant was evaluated and expressed as flavor dilution (FD) factor (Schieberle and Grosch, 1987). The FD factor for a compound is the ratio of its concentration in the initial extract to its concentration in the most dilute extract in which the odor was detected by the GC-O.
GC-MS analysis After RAS sampling for 2 min, the SPME fibers were placed into the injection port of the GC-MS. GC-MS analysis was performed on a 7890A gas chromatograph equipped with a 5975C inert XL mass spectrometer (Agilent Technologies). The capillary column was DB-WAX (60 m x 0.25 mm i.d., 0.25 μm film thickness, Agilent Technologies), and the flow rate of the helium carrier gas was 1.6 mL/min at 50°C. The oven temperature was programmed to an initial 50°C for 2 min, increased at 3°C/min to 220°C, and next held at 220°C for 75 min. The injection port was maintained at 250°C. The inlet was operated in the split mode (split rate 10:1). GC-MS analysis was carried out in triplicate with selected ion monitoring (SIM). The analytes were detected within time windows and identified on the basis of their retention indices and their fragment ions by comparison with standard compounds. GC-MS analysis was conducted under the same conditions where quantitative comparisons of data were necessary.
Identification of volatile compoundsThe standard compounds of guaiacol, 4-ethylguaiacol, 2,3-diethyl-5-methylpyrazine, 2-methylbutanal, 3-methylbutanal, 2-methoxy-3-isopropylpyrazine, 2-isobutyl-3-methoxypyrazine, furfurylthiol, methyl 2-methyl-3- furyl disulfide, methyl furfuryl disulfide, 4-hydroxy-2,5-dimethyl- 3(2H)-furanone, dimethyl trisulfide and 4-octanol were obtained from Tokyo Kasei Kogyo Co., Ltd., 4-vinylguaiacol, 2-ethyl-3,5- dimethylpyrazine and ?-damascenone were from Sigma-Aldrich Co., methyl 2-methylbutyrate was from CTC Organics (Atlanta, GA, USA), and 2-acetylpyridine was from Frutarom (UK) Ltd. (Hartlepool, England). Volatile compounds were identified by comparing their mass spectra and Kovats indices to those of standard compounds. Also, tentative identification of some potent odorants found only by the GC-O analyses was made by comparing their Kovats indices and aroma properties to those of standard compounds.
Quantification of potent odorants showing significant changes during storage An aliquot (200 g) of the 0- or 2-week sample was transferred to a 500 mL conical flask, 4-octanol (1 μg/mL in methanol, 2 mL) was added as the internal standard, and dichloromethane (DCM, 200 mL) was added slowly. The mixture was stirred gently with a magnetic stirrer for 1 hour at room temperature (10 _ 25°C), then separated in a separating funnel and dried over anhydrous sodium sulfate. The extract was carefully concentrated in a rotary evaporator to a final volume of about 1 mL. The GC-MS analysis was performed on an HP 5975C inert XL (Agilent Technologies) with a fused silica capillary column DB-WAX (60 m x 0.25 mm, 0.25-μm film thickness, Agilent Technologies). The flow of the helium carrier gas was 1.6 mL/min. The oven temperature was programmed to an initial 50°C for 2 min, then increased at 3°C/min to 220°C, and held at 230°C for 75 min. The injection port was maintained at 250°C. The inlet was operated in the split mode (split rate 20:1). Ions were monitored in SIM mode: m/z 41, 57, 58 for 2-methylbutanal and 3-methylbutanal, m/z 57, 88, 101 for methyl 2-methylbutyrate, m/z 126 for dimethyl trisulfide, m/z 122, 149, 150 for 2,3-diethyl-5-methylpyrazine, m/z 94, 124, 151 for 3-isobu- tyl-2-methoxypyrazine, m/z 53, 81, 160 for furfuryl methyl disulfide, m/z 77, 135, 150 for 4-vinylguaiacol, and m/z 69, 73, 87 for 4-octanol, the internal standard. A calibration curve was constructed for the eight compounds using standard solutions in dichloro- methane. GC-MS analysis was carried out in triplicate.
Flavor modeling for sensory evaluation For sensory evaluation, the aroma models were prepared on the basis of the quantitative data obtained from the AC coffee beverages. The eight compounds showing decreases during 2-week storage were mixed as follows: 4-vinylguaiacol, 21.3 mg; 3-methylbutanal, 1.8 mg; 2-methylbutanal, 1.2 mg; furfuryl methyl disulfide, 0.20 mg; 2,3-diethyl-5-methylpyrazine, 0.040 mg; dimethyl trisulfide, 0.012 mg; methyl 2-methylbutyrate, 0.006 mg; 2-isobutyl-3- methoxypyrazine, 0.004 mg. The mixture was made up to 100 g with an aqueous solution of ethanol (50% w/w) to yield the 1000-fold concentrations as determined in the AC coffee beverage.
Sensory evaluation
Panel. Assessors were 15 experienced staff (4 female and 11 male) of 22 to 40 years of age from San-Ei Gen F.F.I. Inc. The assessors had been working in the field of flavor and beverage development for more than 3 years.
Test area. Sensory evaluation was carried out in an odor free test room equipped with air conditioning (room temperature 24 to 25°C; humidity 50 to 53%). Each assessor evaluated the samples in a private booth separated by white partitions.
Sample preparation. In preparation of samples for sensory evaluation, 80 mL of the 0-week sample was used as a reference sample, and to a second 80-mL 2-week sample, a 0.1% solution of the model flavor, composed of the eight odorants, was added to make the composition of the added 2-week sample (2-week+ sample) equal to that of the 0-week sample. Each 80-mL model solution was placed in a 180-mL white plain plastic cup coded with a random 3-digit number, which was covered with a white plain plastic lid and offered to the expert sensory panel.
Duo-trio test. A duo-trio test was performed as follows. The assessors received a set of three samples, one (0-week sample) of which was labeled as a reference and the other two samples (2-week and 2-week+ samples) each had a different code. Each assessor tasted the samples using a white plain straw, comparing the reference sample (0-week sample) to the other coded samples (2-week and 2-week+ samples). The assessors pointed out which of the coded samples they believed to be similar to the reference in terms of “flavor intensity” and “overall flavor characteristics”. The number of similar responses was counted and the significance was determined by a binomial test.
Statistical analysis Analysis of variance (ANOVA), the Tukey-Kramer honestly significant difference (HSD) test, and a Student's t-test were conducted using the JMP8 software package (SAS Institute Inc., Cary, NC, USA).
Results and Discussion
GC-O analysis GC-O analysis using CharmAnalysisTM was conducted on 0-, 1-, and 2-week samples to detect potent odorants. Table 1 shows 17 components as potent odorants detected by GC-O analysis of the volatiles in the RAS exhaust. 2-Acetylpyri- dine was not detected in the 0-week sample, but was detected in the 1- and 2-week samples (FD factor = 3 in both samples). The components were classified into eight groups according to odor descriptors such as phenolic, nutty-roast, buttery-oily, green-earthy, sweet-fruity, smoke-roast, sweet-caramel, and other. The ions selected for monitoring these odorants and FD factor (0-week sample) are also shown in Table 1. FD factors are used to indicate the most intense odorants in the flavor extract.
Table 1
Potent odorants in the retronasal aroma simulator (RAS) exhaust of the coffee beverage and selected ions for the GC-MS analysis.
| No. |
|
Descriptor |
Component |
Selected ion (m/z) |
Time (min) |
Retention index |
FDa factor (0 week) |
| 1 |
-1 |
Phenolic |
Guaiacol |
81, 109, 124 |
41 . 71 |
1860 |
27 |
|
-2 |
|
4-Ethylguaiacol |
122, 137, 152 |
47.86 |
2032 |
27 |
|
-3 |
|
4-Vinylguaiacol |
77, 135, 150 |
53.39 |
2197 |
9 |
| 2 |
-1 |
Nutty-roast |
2-Ethyl-3,5-dimethylpyrazine |
108, 135, 136 |
25.66 |
1465 |
27 |
|
-2 |
|
2,3-Diethyl-5-methylpyrazine |
122, 149, 150 |
27.08 |
1497 |
9 |
|
-3 |
|
2-Acetylpyridine6 |
79, 93, 121 |
31 . 61 |
1603 |
- |
| 3 |
-1 |
Buttery-oily |
2-Methylbutanal |
41, 57, 58 |
5.56 |
911 |
3 |
|
-2 |
|
3-Methylbutanal |
41, 57, 58 |
5.62 |
914 |
27 |
| 4 |
-1 |
Green-earthy |
2-Methoxy-3-isopropylpyrazine |
137, 152 |
24.62 |
1438 |
3 |
|
-2 |
|
2-Isobutyl-3-methoxypyrazine |
94, 124, 151 |
28.38 |
1527 |
27 |
| 5 |
-1 |
Sweet-fruity |
Methyl 2-methylbutyrate |
57, 88, 101 |
7.49 |
1008 |
3 |
|
-2 |
|
P-Damascenone |
69, 121, 190 |
40.45 |
1826 |
27 |
| 6 |
-1 |
Smoke-roast |
Furfurylthiol |
81, 114 |
24.31 |
1432 |
9 |
|
-2 |
|
Methyl 2-methyl-3-furyl disulfide |
112, 113, 160 |
34.47 |
1670 |
9 |
|
-3 |
|
Methyl furfuryl disulfide |
53, 81, 160 |
39.93 |
1810 |
9 |
| 7 |
-1 |
Sweet-caramel |
4-Hydroxy-2,5-dimethyl-3(2H)-furanone |
128 |
48.07 |
2036 |
3 |
| 8 |
-1 |
Other |
Dimethyl tiisulfide |
126 |
21.94 |
1381 |
9 |
b Detected in 1- and 2-week samples (FD factor = 3 in both samples).
The FD factors revealed that guaiacol and 4-ethylguaiacol (phenolic odor), 2-ethyl-3,5-dimethylpyrazine (nutty-roast odor), 3-methylbutanal (buttery-oily odor), 2-isobutyl-3-methoxypyrazine (green-earthy odor), and β-damascenone (sweet-fruity odor) had a large FD factor (27). Also, the FD factors for each odor descriptor showed that phenolic odor had the largest intensity among the odors in the 0-week sample.
GC-MS analysis GC-MS (SIM) analysis was performed to compare the released amount of the potent odorants from the RAS. Table 2 shows the peak areas of each selected ion of the coffee aromas at the various storage periods and relative standard deviation (RSD) of each peak area. Reproducible peak areas (RSD < 10%) were obtained in this study. Among the 17 potent odorants detected by GC-O, 2-methoxy-3-isopropylpyrazine (green-earthy odor), furfurylthiol (smoke-roast odor), methyl 2-methyl-3-furyl disulfide (smoke-roast odor), and 4-hydroxy-2,5- dimethyl-3(2H)-furanone (sweet-caramel odor) were not detected by GC-MS (SIM) analysis. It was thought that the collected amounts were too small, or the odor thresholds were too low for these components to be detected by the analysis method in this study. The FD factors of these components showed small values of 3 or 9, as in Table 1.
Table 2
Residual ratios of the potent odorants in the retronasal aroma simulator (RAS) exhaust during 0- to 2-week storage.
|
|
|
|
|
Peak area |
|
|
|
|
|
|
0 week (0W) |
1 week (1W) |
2 weeks (2W) |
|
| No. |
|
Descriptor |
Component |
Retention index |
Meatf |
Residual ratio (%) |
RSD6 (%) |
Mean |
Residual ratio (%) |
RSD (%) |
Mean |
Residual ratio (%) |
RSD (%) |
ANOVA" (Tukey-Kramer HSD test0' ) |
| 1 |
-1 |
Phenolic |
Guaiacol |
1860 |
2.11 × 105 |
100.0 |
9.8 |
2.05 × 105 |
97.5 |
7.9 |
2.01 × 105 |
95.5 |
7.9 |
|
|
-2 |
|
4-Ethylguaiacol |
2032 |
6.85 x104 |
100.0 |
8.6 |
6.58 × 104 |
96.0 |
9.6 |
6.56x 104 |
95.7 |
8.3 |
|
|
-3 |
|
4-Vinylguaiacol |
2197 |
2.10 × 104 |
100.0 |
6.4 |
1.23 × 104 |
58.6 |
9.9 |
8.82x 103 |
42.1 |
8.4 |
*(0-1W, 1-2W, 0-2W) |
| 2 |
-1 |
Nutty-roast |
2-Ethyl-3,5-dimethylpyrazine |
1465 |
2.48 × 104 |
100.0 |
9.2 |
2.29 xlO4 |
92.0 |
6.6 |
2.26 × 104 |
91.2 |
9.6 |
|
|
-2 |
|
2,3-Diethyl-5-methylpyrazine |
1497 |
7.97 × 103 |
100.0 |
8.5 |
6.68 × 103 |
83.8 |
7.5 |
6.64 x103 |
83.3 |
8.0 |
|
|
-3 |
|
2-Acetylpyridine |
1603 |
1.29 × 104 |
100.0 |
7.5 |
1.21 × 104 |
93.7 |
8.6 |
1.14 × 104 |
88.2 |
9.5 |
|
| 3 |
-1 |
Buttery-oily |
2-Methylbutanal |
911 |
3.61x106 |
100.0 |
2.5 |
3 .08 × 106 |
85.3 |
9.3 |
2.65 × 106 |
73.4 |
2.7 |
*(0–1W, 0–2W) |
|
-2 |
|
3-Methylbutanal |
914 |
1.74 × 106 |
100.0 |
4.7 |
1.56 × 106 |
90.0 |
7.1 |
1.36 × 106 |
78.0 |
3.3 |
*(0–2W) |
| 4 |
-1 |
Green-earthy |
2-Methoxy-3-isopropylpyrazine |
1438 |
- |
- |
- |
- |
- |
- |
|
|
|
|
|
-2 |
|
2-Isobutyl-3-methoxypyrazine |
1527 |
2.60 × 103 |
100.0 |
6.2 |
2.13 × 103 |
81.7 |
7.4 |
1.89 × 103 |
72.7 |
9.4 |
*(0–1W, 0–2W) |
| 5 |
-1 |
Sweet-fruity |
Methyl 2-methylbutyrate |
1008 |
1.71 xlO4 |
100.0 |
3.5 |
1.44 × 104 |
84.4 |
9.0 |
1.28 × 104 |
74.7 |
9.9 |
*(0–2W) |
|
-2 |
|
P-Damascenone |
1826 |
1.49 × 103 |
100.0 |
9.3 |
1.26 × 103 |
84.5 |
9.6 |
1.22 × 103 |
81.9 |
8.3 |
|
| 6 |
-1 |
Smoke-roast |
Furfurylthiol |
1432 |
- |
|
- |
- |
|
- |
- |
|
- |
|
|
-2 |
|
Methyl 2-methyl-3-furyl disulfide |
1670 |
- |
|
- |
- |
|
- |
- |
|
- |
|
|
-3 |
|
Methyl furfuryl disulfide |
1810 |
2.27 × 104 |
100.0 |
9.7 |
1.90 × 104 |
83.5 |
9.3 |
1.53 × 104 |
67.5 |
9.1 |
*(0–2W) |
| 7 |
-1 |
Sweet-caramel |
4-Hydroxy-2,5-dimethyl-3(2H)- furanone |
2036 |
- |
|
- |
- |
|
- |
- |
|
- |
|
| 8 |
-1 |
Other |
Dimethyl trisulfide |
1381 |
6.67 × 103 |
100.0 |
7.3 |
6.08 × 103 |
91.2 |
8.4 |
4.93 × 103 |
73.9 |
7.0 |
*(1–2W, 0–2W) |
|
|
Total |
|
|
5.74 xlO6 |
100.0 |
|
5.01 xlO6 |
87.2 |
|
4.36x 106 |
75.8 |
|
|
a Peak areas are the mean values of three measurements.
b Relative standard deviation.
d Significant differences (*p < 0.05) are found between the samples shown within parentheses.
Residual ratios in Table 2 show the percentage of the peak areas of each odorant based on the peak areas of the 0-week sample. Integrated peak areas of the selected ions decreased to about 87% at the 1-week sample, and to about 76% at the 2-week sample, based on those of the 0-week sample. Nutty-roast odorants of pyrazines remained stable during the 2-week storage period, and phenolic odorants of guaiacols also showed some stability, except for 4-vinylguaiacol, which showed the largest decrease among the odorants. The difference in stability among guaiacols appeared to originate from the instability of a vinyl group in 4-vinylguaiacol. Among the potent odorants in the RAS exhaust, 4-vinylguaiacol, 2-methylbutanal and 2-isobutyl-3-methoxypyrazine showed significant decreases after 1-week storage, and 3-methylbutanal, methyl 2-methylbutyrate, methyl furfuryl disulfide, and dimethyl trisulfide decreased significantly during 2-week storage. Eight of 13 potent odorants detected by GC-MS showed significant decreases during 2-week storage. The difference in the decreasing pattern of each odorant suggested that the coffee aroma quality changed during the early storage period.
Among the eight volatiles, seven (excluding methyl furfuryl disulfide) were contained in the 12 volatiles (guaiacol, 4-ethylguaiacol, 4-vinylguaiacol, 2-ethyl-3,5-dimethylpyrazine, 2,3-diethyl-5-methylpyrazine, 2-acetylpyridine, 2-methylbutanal, 3-methylbutanal, 2-isobutyl-3-methoxypyrazine, methyl 2-methylbutyrate, β-damascenone, and dimethyl trisulfide) that decreased significantly and showed differences in the decreasing pattern of each odorant during 2-month storage (Watanabe et al., 2012). Methyl furfuryl disulfide showed a significant decrease in the 2-week sample, but not in the 2-month sample.
Sensory evaluation To confirm whether changes in the aroma characteristics of the AC coffee beverage during storage are caused by the eight compounds that showed significant changes, a duo-trio test was carried out by an expert panel composed of 15 assessors using 0-week, 2-week, and 2-week + samples. The 2-week+ sample was prepared by adding a model flavor to the 2-week sample. The model flavor was composed of the eight compounds on the basis of the decrement of each compound during 2-week storage. The results of quantitative analysis by GC-MS (SIM) are shown in Table 3. Four (4-vinylguaiacol, furfuryl methyl disulfide, dimethyl trisulfide, and 2-isobutyl-3-methoxypyrazine) of the eight quantified compounds exhibited a significant decrease during 2-week storage. For the other four compounds, the significant decreases during 2-week storage were detected only in the RAS- SPME procedure. These differences suggested that the quantitative decrease of flavor compounds in coffee brew might not be the only factor influencing flavor release.
Table 3
Quantification of potent odorants in a dichloromethane (DCM) extract of the coffee beverage.
|
μg/100 g of coffee beverage |
|
|
| Compound |
0 week |
2 weeks |
0 week-2 weeks |
Student's t-test |
Residual ratio (%) |
| 4-Vinylguaiacol |
53.1 ± 3.5a |
31.7± 2.3 |
21.3 |
** |
59.7 |
| 3-Methylbutanal |
32.8± 2.7 |
31.0± 1.7 |
1.8 |
|
94.5 |
| 2-Methylbutanal |
66.2± 2.6 |
65.0 ± 1.7 |
1.2 |
|
98.2 |
| Furfuryl methyl disulfide |
0.86± 0.07 |
0.66± 0.02 |
0.20 |
** |
76.7 |
| 2,3-Diethyl-5-methylpyrazine |
1.52± 0.04 |
1.48± 0.02 |
0.04 |
|
97.4 |
| Dimethyl trisulfide |
0.067± 0.006 |
0.055± 0.003 |
0.012 |
* |
82.1 |
| Methyl 2-methylbutyrate |
0.099± 0.002 |
0.093 ± 0.006 |
0.006 |
|
93.9 |
| 2-Isobutyl-3-methoxypyrazine |
0.033 ± 0.001 |
0.029± 0.001 |
0.004 |
* |
87.9 |
a Mean ± standard deviation (n = 3)
The 0-week sample was labeled as reference and the other two samples were labeled with a different 3-digit code. The two coded samples were randomly presented to the 15 assessors. Fifteen evaluation scores were provided for the group. As a result, 12 out of 15 scores in the group indicated that the 2-week+ sample significantly (p < 0.05) resembled the reference sample (0-week sample) in terms of aroma quality by a binomial test. Also, 11 out of 15 scores in the group indicated that the 2-week+ sample tended to resemble the reference sample (0-week sample) in terms of aroma intensity (Figure 1.). Consequently, a similarity between 0-week sample and 2-week+ sample was significantly recognized in terms of aroma quality. This demonstrated that the decrease of the eight compounds largely affects changes in aroma characteristics during early stage storage.
Conclusion
Changes in RA profiles of an aseptic cup coffee beverage during 0-week to 2-week storage were studied using an RAS. The potent odorants in the retronasal coffee aroma were identified by GC-O analysis of the volatiles in the RAS exhaust. Out of 17 odorants detected by GC-O, 13 odorants were detected by GC-MS analysis. The peak areas of the odorants in the RAS exhaust were compared among the coffee samples stored for 0 week to 2 weeks. Among the 13 detected compounds, eight odorants decreased significantly during the 2-week storage period. The phenolic odor compound 4-vinylguaiacol showed the largest decrease, while pyrazines with a nutty-roast odor showed little decrease during the storage period. Differences in the pattern of each odorant suggested that RA character changes during the early storage stage. The addition of potent odorants, equal in amount to the observed decreases, to the 2-week coffee samples was detected by the sensory panel as similar to the 0-week coffee sample in terms of the intensity and quality of the coffee aroma. The results indicated that the decreases of the eight potent odorants in the coffee were a contributing factor to the change in coffee character during the early stage of storage.
Abbreviations
AC; aseptic-cup, AL; aluminum, ANOVA; analysis of variance, DCM; dichloromethane, E-nose; electronic nose, EVOH; ethylene vinyl alcohol copolymer, FD factor; flavor dilution factor, GC-MS; gas chromatography-mass spectrometry, GC-O; gas chromatography-olfactometry, HSD; honestly significant difference, OBS; oral breath sampling, PET; polyethylene terephthalate, PS; polystyrene, RA; retronasal aroma, RAS; retronasal aroma simulator, RO; reverse osmosis, RSD; relative standard deviation, SIM; selected ion monitoring, SPME; solid-phase microextraction, TBA; tetra brick aseptic, UHT; ultra-high-temperature
References
- Acree, T.E., Barnard, J., and Cunningham, D.G. (1984). A procedure for the sensory analysis of gas chromatographic effluents. Food Chem., 14, 273–286.
- Akiyama, M., Murakami, K., Ikeda, M., Iwatsuki, K., Wada, A., Tokuno, K., Onishi, M., and Iwabuchi, H. (2007). Analysis of the headspace volatiles of freshly brewed arabica coffee using solid-phase microextraction. J. Food Sci., 72, C388–396.
- Akiyama, M., Murakami, K., Ikeda, M., Iwatsuki, K., Wada, A., Tokuno, K., Onishi, M., Iwabuchi, H., and Sagara, Y. (2009). Analysis of freshly brewed espresso using a retronasal aroma simulator and influence of milk addition. Food Sci. Technol. Res., 15, 233–244.
- Deibler, K.D., Llesca, F.M., Lavin, E.H., and Acree, T.E. (2004). Calibration of gas chromatography inlet splitting for gas chromatography olfactometry dilution analysis. Flavour Fragr. J., 19, 518–521.
- Elmore, J.S. and Langley, K.R. (1996). Novel vessel for the measurement of dynamic flavor release in real time from liquid foods. J. Agric. Food Chem., 44, 3560-3563.
- Graus, M., Yeretzian, C., Jordan, A., and Lindinger, W. (2001). In-mouth coffee aroma: breath-by-breath analysis of nose-space while drinking coffee. In “Proceedings of the 19th ASIC Colloquium”, Trieste. Italy.
- Kacem, B., Matthews, R.F., Crandall, P.G., and Cornell, J.A. (1987). Nonenzymatic browning in aseptically packaged orange juice and orange drinks. Effect of amino acids, deaeration, and anaerobic storage. J. Food Sci., 52, 1665–1667.
- Michishita, T., Akiyama, M., Hirano, Y., Ikeda, M., Sagara, Y., and Araki, T. (2010). Gas chromatography/olfactometry and electronic nose analyses of retronasal aroma of espresso and correlation with sensory evaluation by an artificial neural network. J. Food Sci., 75, S477–489.
- Pérez-Martínez, M., Sopelana, P., de Peña, M.P., and Cid, C. (2008). Changes in volatile compounds and overall aroma profile during storage of coffee brews at 4 and 25°C. J. Agric. Food Chem., 56, 3145–3154.
- Roberts, D.D. and Acree, T.E. (1995). Simulation of retronasal aroma using a modified headspace technique: investigating the effects of saliva, temperature, shearing, and oil on flavor release. J. Agric. Food Chem., 43, 2179–2186.
- Roig, M.G., Bello, J.F., Rivera, Z.S., and Kennedy, J.F. (1999). Studies on the occurrence of non-enzymatic browning during storage of citrus juice. Food Res. Int., 32, 609–619.
- Roozen, J.P. and Legger-Huysman, A. (1995). Sensory analysis and oral vapour gas chromatography of chocolate flakes. In “Proceedings of the 4th Wartburg aroma symposium: aroma: perception, formation, evaluation”, Ed. by Rothe, M. and Kruse, H.P., Eisenach, pp. 627–632.
- Schieberle, P. and Grosch, W. (1987). Evaluation of the flavour of wheat and rye bread crusts by aroma extract dilution analysis. Z. Lebensm. Unters. Forsch., 185, 111–113.
- van Ruth, S.M., Roozen, J.P., and Cozijnsen, J.L. (1994). Comparison of dynamic headspace mouth model systems for flavour release from rehydrated bell pepper cuttings. In “Trends in Flavour Research”, Ed. by Maarse, H. and van der Heij, D.G.., Elsevier Science, Amsterdam, pp. 59–64.
- Watanabe, R., Akiyama, M., Miyai, T., Ohata, M., Imayoshi, Y., Ikeda, M., Ichihashi, N., and Iwabuchi, H. (2012). Changes in flavor characteristics of a chilled-cup coffee beverage during storage. In “Proceedings of the 24th ASIC Colloquium”, San Jose, Costa Rica, pp. 316–321.
