Flavor Release from a Banana Soft Drink Complex Model System: Evaluation of the Efficiency of Different Adsorbents for Trapping the Released Volatiles During Storage

Abstract

The main objective of the present study was to evaluate the release of aroma compounds from a banana soft drink complex model system in comparison with their release from simple model systems, each contains individual food ingredients. The effect of different food ingredients (sweeteners and thickeners at different concentrations), used in formulation of the soft drinks, on flavor release from simple model systems containing banana flavor was evaluated separately. The optimum release of isoamyl acetate (the most potent odorant of banana aroma) was observed at a concentration of 10% of each investigated sweetener (sucrose, glucose, and corn syrup). Pectin and xanthane showed the highest release at a concentration of 2.5 and 0.8% w/w, respectively. The level of each ingredient that showed the optimum release of isoamyl acetate was selected and used in formulation of a banana soft drink complex model system. The released volatiles were trapped by tenax and activated carbon. The gas chromatographic-mass spectrometric analysis revealed a gradual decrease (p < 0.05) in the volatile compounds release from the complex banana soft drink model system during storage for 90 days. However, the total content of the volatiles adsorbed by activated carbon trap was higher than that trapped by tenax. The percentage of isoamyl acetate to total volatiles (isoamyl acetate/total volatiles %) was calculated for each sample during storage. A distinct linear correlation was found between the calculated values and storage time (r = 0.97 and 0.92 for volatiles trapped by tenax and activated carbon, respectively). Sample stored for 60 days showed the highest value. These findings confirmed the results of aroma sensory evaluation.

Keywords:

• Aroma release
• Soft drink
• Banana flavor
• Model system solution
• Sensory evaluation

INTRODUCTION

Flavor release of a particular food is a major factor determining consumer acceptance. The modification of volatiles release can be influenced by the interactions between flavor compounds and a variety of non-flavor matrix components.^[1,^2] Variations in food matrix composition highly affect the binding and release of volatile flavor compounds.^[3−5]

Flavor is perhaps the most important characteristic of a soft drink, so it is not surprising that development in the flavor and soft drinks industries have closely linked. The technology of flavor for soft drinks has progressed considerably during last years. Considerable research has been conducted to better understand the interaction mechanisms between the flavor compounds and the ingredients used in formulation of the soft drinks.^[6−8] However, most of these studies used single ingredient or simple aqueous model systems, which don’t closely resemble real soft drinks.

In fact, most food products are complex mixtures consisting of various types of ingredients, which further complicate flavor-matrix interactions and hence, the disposition of individual flavor compounds in the food system. Therefore, results obtained from the simple model systems may not be directly applied to real foods. In case of soft drinks reformulation, modification of the flavoring system is usually required a method that can directly measure flavor binding in real soft drinks instead of an individual ingredient. This certainly can aid in the efficient design of a flavoring system and improve management of flavor in the new production development to ensure optimum flavor quality of the product.

The impact of a food component on the retention or release of a volatile compound usually implies the headspace analysis. Trapping involving headspace concentration, using porous polymer absorbents, has been widely used for the analysis of aroma release.^[9,^10]Activated carbon has been used for the adsorption and recovery of volatile compounds.^[11,^12]

The main objective of the present study was to evaluate the aroma release from a banana soft drink related complex model system in comparison with their release from simple model systems, each contains individual food ingredient. To achieve this aim the effect of different ingredients used widely in formulation of the soft drinks (sweeteners and thickeners at variable levels) on aroma release from simple models containing banana flavor was evaluated separately. A banana soft drink complex model system was then formulated in view of the obtained results. The efficiency of activated carbon as a cheap adsorbent in trapping the released volatiles was evaluated in comparison with tenax. The study was extended to follow the changes in release of the most potent volatiles and sensory characteristics of the banana soft drink complex model system during storage for 90 days.

MATERIALS AND METHODS

Flavor and Chemicals

Non-commercial banana flavor dissolved in propylene glycol was kindly donated by Greatco Company (essential oils, fragrances, flavors, extracts), Giza, Egypt. Sucrose, glucose, pectin, xanthan, carboxymethyl cellulose (CMC), authentic volatile compounds, standard n-paraffins (C₆–C₂₂), and Tenax TA (poly [2, 6-diphenyl-p-phenylene oxide])were purchased from Sigma (St. Louis, MO, USA). Corn syrup DE 60 was purchased from Roquette frere, Lestrim, France. Citric acid was purchased from Panreac Quimica Sa (E-08110 Montcada i Reixac [Barcelona] Espana). Activated carbon (prepared from pistachio shells) was kindly donated by Department of Surface Chemistry, National Research Center, Egypt. All other chemicals were at least of analytical grade.

Preparation of Simple Model System Solutions

The effect of variable levels of different sweeteners and thickeners on release of banana aroma was investigated, separately, by using simple model system solutions. Sweetened model solutions containing variable concentrations (5, 10, 20, 40, and 60% w/w) of sucrose, glucose, or corn syrup were prepared separately.^[13,^14] Thickened model solutions containing different concentrations (0.05, 0.1, 0.25, 0.7, and 2.5% w/w) of pectin,^[13] (0.02, 0.1, 0.4, and 0.8% w/w) xanthan,^[7] and (0.5 and 1.84% w/w) CMC^[14] were prepared separately. Banana flavor was added (0.25 % w/v) to each model solution, under agitation in closed system at 60°C for 3 min. All model solutions were prepared at least 24 h before evaluation and stored at 4°C.^[15] The samples were brought to room temperature before isolation of the headspace volatiles.

Preparation of a Complex Banana Soft Drink Model System Solution

The complex banana soft drink model solution was formulated in view of the obtained results concerning the optimum release of isoamyl acetate (most potent odorant of banana aroma) from each simple model solution as follows: 5 g sucrose, 5 g corn syrup, 1.25 g pectin, 0.4 g xanthan, and 0.4 sodium-citrate. All were dissolved in 100 mL distilled water and 0.25 g banana flavor was added and mixed thoroughly with magnetic stirrer for 3 min at 60°C.The pH was adjusted as required (pH 5) by using (50% w/w) citric acid solution. To investigate the storage effect on aroma release seven samples were prepared and packaged in glass bottles, sterilized, and stored immediately in refrigerator at 4°C. CMC was excluded because it gave rise to a highly viscous solution. A sample was withdrawn at time intervals of 15 days and subjected to isolation and analysis of headspace volatiles and sensory evaluation.

Isolation of Headspace Volatiles

The headspace volatiles released from different simple model system solutions and the complex banana soft drink model system were isolated by using the so-called mouth model^[15] consisting of sample flask (100 mL) maintained at a temperature of 37°C (water bath). A purified nitrogen gas flow (20 mL/min) passed through the stirred solution for 10 min to trap the volatile compounds in 0.25g of tenax TA or 6.0 g of activated carbon positioned in a glass tube. The experiments were conducted in triplicate. The volatiles were eluted from the tenax and activated carbon traps by using diethyl ether. For accurate determination of the volatiles released from the complex model system, 1 mL of internal standard (IS; 2- methyl valerate [10 μg/mL in methanol]) was added to the samples before extraction. The solvents containing the volatiles were dried over anhydrous sodium sulfate for 1 h. The volatiles were obtained by evaporation of the solvents under reduced pressure to final volume of 100 μl.

Gas Chromatographic (GC) Analysis

GC analysis was performed by using a Hewlett-Packard model (5890) equipped with flame ionization detector (FID). A fused silica capillary column DB5 (60 m × 0.32 mm i.d.) was used. The oven temperature was programmed from 50 to 240°C at a rate of 3°C/min. Two microliters of each sample was injected in the split ratio of 1:10. Helium was used as the carrier gas, at flow rate 1.1 mL/min. The injector and detector temperatures were 220 and 250°C, respectively. The retention indices (Kovats index) of the separated volatile components were calculated with hydrocarbons (C6–C22, Aldrich Chemical Co.) as references.

Gas Chromatographic-Mass Spectrometric (GC-MS) Analysis

The analysis was carried out by using a coupled gas chromatography Hewlett-Packard (5890)/mass spectrometry Hewlett-Packard-MS (5970). The ionization voltage was 70 eV, mass range m/z 39–400 amu. The GC conditions were carried out as mentioned earlier. The isolated peaks were identified by matching with data from the library of mass spectra (Natural Institute of Standard & Technology) and comparison with those of authentic compounds and published data.^[16] The quantitative determination was carried out based on peak area integration. For the complex model system the relative volatile concentrations in the seven stored samples were determined by comparison with the concentrations of IS (methyl valerate), assuming a response factor of one.

Sensory Evaluation

The changes in the sensory characteristics of the formulated banana soft drink during storage for 90 days were followed. The different sensory qualities (aroma, taste, mouthfeel, color, and overall acceptability) were estimated^[17] and scored by 20 panelists (Food Technology & Nutrition Division, National Research Centre, Cairo, Egypt) who were non-smokers. A nine-point hedonic scale was used anchored on the extremes “worst” (score 1) and “excellent” (score 9). Samples were evaluated at room temperature. Each panelist received the samples separately in colorless glass cups and provided with water to drink between evaluations. A sensory score 4.5 was taken as the cut off for acceptability.

Statistical Analysis

Data were analyzed using one-way analysis of variance (ANOVA) and least significant difference (LSD) was performed to determine any significant difference among various treatments that were used to compare the means. Differences were considered to be significant at p < 0.05.^[18]

RESULTS AND DISCUSSIONS

Composition of Banana Flavor

More than 15 volatile compounds were separated on GC column, however only seven esters, accounting for 97.53% w/w of the total volatiles, were investigated in the present study (Table 1) since they were the best separated and therefore, most suitable for quantitation. Description of the volatile components and method of identification are cited in Table 1. The predominant compound was isoamyl acetate (37.63%) followed by butyl acetat (27.87%), and isoamyl butanoate (17.69%). As reported in previous studies^[19−21] esters are the main contributors to banana aroma.

TABLE 1 Volatile compounds identified in banana flavor, methods of their identification, and odor description

RI	Volatile compounds	Relative area %	Methods of identification	Description	References
617	Ethyl acetate	4.48	MS,KI,ST	Banana, ethereal, pineapple	Mayr et al.^[^19^]
774	Isobutyl acetate	6.85	MS,KI	Banana, rum	Mayr et al.^[^19^]
801	Ethyl butanoate	0.96	MS,KI,ST	Green fruity	Mahattanatwawee et al.^[^20^]
818	Butyl acetate	27.87	MS,KI,ST	Fruity	Mahattanatwawee et al.^[^20^]
861	Ethylmethyl butanoate	2.05	MS,KI	Fruity, sweet green	Mahattanatwawee et al.^[^20^]
884	Isoamyl acetate	37.63	MS,KI,ST	Banana, pear	Mayr et al.^[^19^]
1064	Isoamyl butanoate	17.69	MS,KI	Apricot, banana, pineapple	Mayr et al.^[^19^]
	Total identified (%)	97.53

RI: Retention indices (RI) on DB5 column; Compounds identified by GC-MS (MS) and or retention indices (RI) on DB5 and or by comparison of MS and RI of standard compounds (ST) run under similar GC-MS conditions.

Aroma Release from Sweetened Model System Solutions

The volatile components of banana aroma released from model solutions with increasing concentration of sucrose, glucose, or corn syrup are cited in Table 2 with their relative area percentages. Increasing the concentration of sucrose up to 60% revealed a significant (p < 0.05) increase in release of ethyl acetate, butyl acetate, and ethylmethyl butanoate. Release of isobutyl acetate and isoamyl butanoate showed an opposite trend, this result is in agreement with that of Rabe et al.^[22] These two components may be entrapped in amorphous micro regions of hydrogen bonded sucrose molecules^[14] formed at high concentration. The increase in release of isoamyl acetate at 10–40% sucrose may be attributed to the “salting out” effect of sucrose, whereby sucrose interacts with water, increasing the concentration of flavor compounds in remaining “free” water.^[9] The variability in release of the volatile compounds (Table 2) at high sucrose concentration may indicate a competition between sucrose and aroma molecules in water hydration. Furthermore, sucrose/sucrose interactions via hydrogen bonding^[23] were shown to influence molecular mobility.^[24] The mobility of sucrose/water systems decreased in particular at high sucrose concentration >40%. Additionally, flavor/sucrose interaction may occur at high sucrose concentration.^[22]

TABLE 2 Volatile compounds of banana flavor released from simple model system solutions containing sweeteners at different concentrations. (Values expressed as relative area percentages to total volatiles)

		Concentration % (w/w)
		Sucrose					Glucose					Corn syrup
Volatile compounds	Control	5	10	20	40	60	5	10	20	40	60	5	10	20	40	60
Ethyl acetate	4.48^a	5.13^b	4.23^b	4.95^b	4.73^ab	4.89^ab	1.02^b	1.01^b	1.25^b	1.76^b	1.38^b	0.65^b	0.94^c	1.06^c	0.84^bc	0.93^c
Isobutyl acetate	6.85^a	7.36^a	4.64^b	4.04^b	3.42^bc	3.25^c	3.73^b	4.05^b	3.20^b	3.72^b	3.77^b	4.27^b	3.69^b	4.08^b	4.22^b	3.59^c
Ethyl butanoate	0.96^a	0.92^a	0.99^a	0.20^b	0.39^b	0.36^b	0.79^b	0.36^c	0.27^c	−	−	0.11^b	0.13^b	0.14^b	0.14^b	−
Butyl acetate	27.87^a	25.45^b	26.14^a	27.49^a	31.03^c	31.24^c	31.99^b	30.82^b	34.57^c	33.69^c	35.13^d	33.69^b	34.11^b	34.53^b	33.94^b	33.63^b
Ethylmethyl butanoate	2.05^a	2.14^a	2.96^b	3.04^b	2.18^a	2.84^b	0.18^b	0.14^b	0.15^b	−	1.16^c	0.72^b	0.61^b	1.09^c	1.60^c	2.04^d
Isoamyl acetate	37.63^a	37.69^a	45.27^b	42.78^c	44.92^b	41.19^c	41.49^b	42.74^c	42.44^bc	42.46^bc	40.65^b	40.34^b	42.05^b	41.64^b	39.95^b	40.61^b
Isoamyl butanoate	17.69^a	18.63^a	14.28^b	14.73^b	11.13^c	13.73^b	17.74^a	17.16^a	15.42^b	15.18^b	14.96^b	16.84^a	15.47^b	14.90^c	15.35^c	15.45^bc
Total identified (%)	97.53	97.32	98.51	97.23	97.80	97.50	96.94	96.28	97.30	96.81	97.05	96.62	97.00	97.44	96.04	96.25

Control: Aroma released from banana flavor in distilled water with no added sweeteners; Values followed by different superscript letters in the same row for each sweetener are significantly different (p < 0.05).

Model solutions containing increased concentration of glucose showed a significant (p < 0.05) decrease in release of all volatile components except for butyl acetate and isoamyl acetate that showed opposite trend. However, the model solution containing 10% glucose exhibited the highest release of isoamyl acetate. The influence of corn syrup on release of banana volatiles showed the same trend as that with glucose (Table 2). Corn syrup is formed by hydrolysis of starch and contains molecules larger than sucrose. Therefore, it has lower binding sites for water when used at same concentration as sucrose. Therefore, more free water is available, decreasing the concentration of flavor molecules in the water, reducing the release in gas phase.

The results mentioned above revealed that the optimum release of isoamyl acetate was at concentration of 10% w/w for all investigated sweeteners. The synergistic effect of mixing the sweeteners on release of the volatile compounds was studied. Different model solutions containing blends; sucrose/glucose (S/G, 1:1), sucrose/corn syrup (S/C, 1:1), glucose/corn syrup (G/C, 1:1), and sucrose/glucose/corn syrup (S/G/C, 1:1:1) were prepared at fixed concentration 10% w/v. The results revealed that sample (S/C) registered the highest release (data not shown) of isoamyl acetate (the most potent odorant of banana aroma). So it was selected as a sweetener in preparation of the banana soft drink complex model system.

Aroma Release from Thickened Model System Solutions

Release of ethyl acetate, isobutyl acetate, ethyl butanoate, and isoamyl butanoate showed significant (p < 0.05) decrease by increasing the concentration of each tested thickener (Table 3). Butyl acetate and isoamyl acetate showed opposite trend. In previous studies,^[7] it was observed that the ester retention increased with increasing aroma compound volatility and xanthan concentration and depends on their physicochemical properties. Boland et al.^[25] stated that, the effect of hydrocolloids on flavor release may be due to two mechanisms; one is the physical entrapment of flavor molecules within the food matrix. The second mechanism involves interactions between the flavor molecules and the gel components, e.g., gelatin.^[26] Bakker et al.^[27] found that binding occurred between diacetyl and gelatin, while Baek et al.^[28] concluded that gelatin did not bind furfuryl acetate.

TABLE 3 Volatile compounds of banana flavor released from simple model system solutions containing thickeners at different concentrations. (Values expressed as relative area percentages to total volatiles)

		Concentration % (w/w)
		Pectin					Xanthan				CMC
Volatile compounds	Control	0.05	0.10	0.25	0.70	2.50	0.02	0.10	0.40	0.80	0.50
Ethyl acetate	4.48^a	4.88^b	4.80^b	1.17^c	0.25^d	0.24^d	0.34^b	0.35^b	0.25^c	0.38^b	0.42^b
Isobutyl acetate	6.85^a	3.70^b	3.30^b	3.83^b	3.79^b	3.92^b	4.39^b	4.50^b	3.66^c	3.26^c	3.07^b
Ethyl butanoate	0.96^a	0.13^b	0.12^b	−	−	−	−	−	−	−	−
Butyl acetate	27.87^a	30.29^b	31.06^b	32.68^c	31.29^c	33.80^d	33.48^b	32.67^b	35.62^c	32.83^b	40.61^b
Ethylmethyl butanoate	2.05^a	4.29^b	4.08^b	3.35^c	−	−	−	−	−	−	−
Isoamyl acetate	37.63^a	37.56^a	41.02^b	41.27^b	46.62^c	47.60^c	42.81^b	41.34^c	42.63^b	44.39^d	41.23^b
Isoamyl butanoate	17.69^a	15.24^b	12.78^c	15.21^b	14.26^b	10.24^d	16.98^ad	18.65^b	14.66^c	16.17^d	11.37^b
Total identified (%)	97.53	96.09	97.16	97.51	96.21	95.80	98.00	97.51	96.82	97.03	96.70

Control: Aroma released from banana flavor in distilled water with no added thickeners; Values followed by different superscript letters in the same row for each thickener are significantly different (p < 0.05).

TABLE 4 Effect of storage on release of the volatile compounds from a banana soft drink complex model system (volatile compounds recovered by activated carbon trap)

		Time of storage
RI	Volatile compounds	Zero	15-day	30-day	45-day	60-day	75-day	90-day
616	Ethyl acetate	211.7^a ± 33.7	243.9^b ± 5.7	202.7^a ± 18.1	153.9^c ± 1.9	51^d ± 2.0	36.5^e ± 1.1	18.6^f ± 1.7
774	Isobutyl acetate	−	−	−	−	5.3 ± 0.8	−	−
800	Ethyl butanoate	−	−	−	3.1 ± 0.1	−	−	−
817	Butyl acetate	62.4^a ± 4.3	20.6^b ± 1.0	24.2^c ± 0.7	15.3^d ± 1.8	39.8^e ± 1.1	5.9^f ± 0.5	8.1^g ± 0.9
861	Ethylmethyl butanoate	497.1^a ± 9.2	425.5^b ± 6.5	197.5^c ± 6.5	22.9^d ± 0.9	−	3.0^e ± 1.0	−
884	Isoamyl acetate	72.6^a ± 3.1	26.1^b ± 0.7	31.1^c ± 0.5	18.7^d ± 0.5	46.8^e ± 0.7	11.4^f ± 0.1	10.5^f ± 1.2
1064	Iosamyl butanoate	21.6^a ± 1.4	8.0^b ± 0.3	8.2^b ± 1.5	3.9^c ± 0.8	9.8^d ± 0.7	6.3^e ± 0.1	3.1^c ± 0.4
Total yield of volatiles		865.4^a ± 15.0	724.1^b ± 10.2	463.7^c ± 11.0	217.8^d ± 9.9	152.7^e ± 12.4	63.1^f ± 8.5	40.3^g ± 4.4

RI: Retention indices (RI) on DB5 column; Values expressed as μg/100 ml soft drink model system; Values followed by different superscript letters in the same row are significantly different (n = 3, p < 0.05).

TABLE 5 Effect of storage on release of the volatile compounds from a banana soft drink complex model system (volatile adsorbed on tenax trap)

		Time of storage
RI	Volatile compounds	Zero	15 day	30 day	45 day	60 day	75 day	90 day
616	Ethyl acetate	218.3^a ± 13.0	220.9^a ± 3.7	84.7^b ± 1.6	72.6^c ± 0.2	72.3^c ± 0.4	49.01^d ± 0.5	27.7^e ± 1.0
774	Isobutyl acetate	−	−	−	−	7.3 ± 0.0	−	−
800	Ethyl butanoate	−	−	−	−	−	−	−
817	Butyl acetate	16.3^a ± 0.0	20.9^b ± 0.5	15.8^a ± 1.0	12.6^c ± 0.8	58.5^d ± 0.3	10.99^c ± 0.0	6.9^e ± 0.9
861	Ethylmethyl butanoate	439.0^a ± 32.5	241.2^b ± 15.7	59.6^c ± 6.4	43.1^d ± 1.1	−	0.71^e ± 0.5	6.2^e ± 1.2
884	Isoamyl acetate	39.5^a ± 1.1	27.1^b ± 0.1	18.85^a ± 0.6	18.9^a ± 0.9	35.8^c ± 0.4	7. 1^d ± 0.3	8.9^e ± 0.5
1064	Isoamyl butanoate	6.4^a ± 0.4	6.5^a ± 0.2	5.1^b ± 0.3	8.4^c ± 0.2	16.4^d ± 0.2	2.66^e ± 0.2	2.3^e ± 0.4
Total yield of volatiles		719.5^a ± 20.1	516.6^b ± 15.5	184.1^c ± 10.8	155.6^d ± 11.3	190.3^c ± 12.2	70.47^e ± 7.7	52.0^f ± 8.3

RI: Retention indices (RI) on DB5 column; Values expressed as μg/100 ml soft drink model system; Values followed by different superscript letters in the same row are significantly different (n = 3, p < 0.05).

TABLE 6 Sensory evaluation of the banana soft drink complex model system (fresh and stored for three months)

Storage time (days)	Mouthfeel	Aroma	Taste	Color	Overall acceptability
0.0	7.53 ± 0.99	7.45^a ± 0.09	7.66 ± 0.61	7.41 ± 0.51	7.58^a ± 0.56
15	7.45 ± 0.09	7.45^a ± 0.09	7.40 ± 0.09	7.45 ± 0.14	7.50^a ± 0.10
30	7.42 ± 0.05	7.49^a ± 0.09	7.40 ± 0.09	7.45 ± 0.14	7.51^a ± 0.09
45	7.38 ± 0.06	7.63^a ± 0.04	7.61 ± 0.10	7.63 ± 0.07	7.62^a ± 0.07
60	7.36 ± 0.05	8.47^b ± 0.12	7.63 ± 0.09	7.61 ± 0.06	8.02^b ± 0.08
75	7.31 ± 0.14	8.35^b ± 0.25	7.67 ± 0.11	7.65 ± 0.21	7.73^b ± 0.18
90	7.32 ± 0.98	7.96^c ± 0.51	7.77 ± 0.75	7.58 ± 0.04	7.70^ab ± 0.82
R^2	0.929	0.581	0.366	0.683	0.329

Values followed by different superscript letters in the same column are significantly different (n = 20; p < 0.05).

Aroma Release from a Banana Soft Drink Complex Model System during Storage

As shown in Tables 4 and 5 except for ethyl acetate, activated carbon was more efficient in trapping the released aroma compounds than tenax. The total content of released volatiles trapped with activated carbon was higher than that with tenax for the samples stored up to 45 days, then opposite trend was observed during the last time of storage. Despite isoamyl acetate was the major released compound at the selected levels (Tables 2 and 3) of the ingredients used in formulation of the complex model system, yet it showed low release from the fresh sample trapped either with activated carbon or tenax. Isobutyl acetate and methyl butanoate could not be detected in the complex banana soft drink model system (Tables 4 and 5). Whereas, ethyl acetate and ethyl methyl butanoate were the major compounds trapped from the fresh samples either with activated carbon or tenax.

The release pattern of all volatile compounds was followed during storage for 90 days. In general all compounds trapped by activated carbon or tenax exhibited significant (p < 0.05) decrease at the end of storage time. Ethyl acetate content adsorbed by tenax trap showed insignificant (p > 0.05) increase after storage for 15 days followed by significant (p < 0.05) decrease with increasing the storage period. Whereas it showed significant increase in sample adsorbed by activated carbon trap followed by significant (p < 0.05) decrease during storage from 30–90 days. In a previous study,^[4^] the effect of pectin on flavor release from orange beverage emulsion during storage was followed. The authors observed an increase in the release content of ethyl acetate at the beginning of storage. Conversely, the release of this compound was significantly (p < 0.05) decreased during the last two months of storage. In the same study, ethyl butanoate similarly decreased (p < 0.05) during the last two months of storage.

Isoamyl acetate, the potent odorant of banana flavor, exhibited the highest release from the sample stored for 60 days compared with the other stored samples (Tables 4 and 5). It showed a remarkable decrease (p < 0.05) during the rest time of storage.

Sensory Evaluation

The effect of different thickeners such as, pectin, xanthan, CMC, and Arabic gum on the sensory qualities of some food products have been investigated in recent studies.^[29,^30] On the other hand the effect of different pH values and temperatures on the storage stability of aspartame that used as sweetener in orange flavored soft drink has been evaluated.^[31] However, to best of the authors’ knowledge, no study could be found concerning the relation between the flavor acceptability and composition of the released volatiles.

In the present study, effect of storage on the sensory characteristics of the banana soft drink complex model system was followed during 90 days (Table 6). The observed results revealed insignificant (p > 0.05) variation for mouthfeel, taste, and color. Whereas, the aroma and overall acceptability attributes showed gradual (p < 0.05) increase up to 60 days followed by a significant decrease at the end of storage time. However, the correlation coefficient R² between the slight variations in the mouthfeel as a function of storage time was higher than that of each tested sensory quality (Table 6).

As shown in Tables 4 and 5, the total yield of the released volatile compounds showed significant (p < 0.05) decrease during storage. Therefore, to confirm the results of sensory evaluation, the percentage of isoamyl acetate content in total identified volatiles (isoamyl acetate/total volatiles %) of each stored sample was calculated as a function of storage time. The results revealed a distinct correlation (r = 0.97 and 0.92) between the calculated values and time of storage for volatiles trapped by tenax and activated carbon, respectively. However, sample stored for 60 days exhibited the highest value. These results are consistent with those of the odor sensory analysis (Table 6).

CONCLUSION

The results of this study clearly illustrate that the release of flavor compounds from the simple model systems, each containing a single food ingredient, do not give an accurate indication for their release from a real food system. This finding may be due to the interaction effects of independent variable on the release variation of the volatile compounds. In general flavor release is complex and several mechanisms can take place such as the competition between the independent ingredients and aroma molecules in water hydration, in addition to the ingredient/ingredient interactions via hydrogen bonding which influence molecular mobility. Furthermore flavor/ingredient interaction may occur. The results revealed the high efficiency of activated carbon a convenient and easily handling adsorbent as a cheap adsorbent for trapping the volatiles of banana flavor. Storage for 90 days revealed significant (p < 0.05) decreases in release of all volatile compounds. However, the aroma and overall acceptability attributes were improved during storage, in particular sample stored for 60 days which showed the highest scores. A linear correlation was found between the percentage of isoamyl acetate in total volatiles (isoamyl acetate/total volatiles %) and storage time. This finding confirmed the results of sensory evaluation.

ACKNOWLEDGMENT

The authors wish to thank Professor Dr. Mohamed Selime at the Department of Surface Chemistry, National Research Center, for providing the activated carbon.