African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 6, Num. 4, 1998, pp. 397-405

A. Namutebi

Department of Food Science & Technology, Makerere University, P. O. Box 7062, Kampala, Uganda

Code Number: CS98042

ABSTRACT

Sensory and headspace gas chromatography analyses were done to evaluate the changes in the flavour of pasteurised and chemically preserved passion fruit (Passiflora edulis Sims.) juices stored at -18, 4 - 8 and 23ºC over a 12 week period. The colour of the juices was monitored by a Minolta colorimeter. The flavour attributes, except acidity and sweetness intensity of pasteurised juice were inferior to the chemically preserved juice. Headspace chromatograms of pasteurised juice showed slight changes in the volatile compound profile. Pasteurised juice was darker, less red and yellow (L: 24.66, a: -3.88, b: 8.43) compared to the chemically preserved (L: 33.22, a: -4.95, b: 15.47) juice. With the exception of acidity and sweetness intensity, other flavour attributes were also found to be dependant on storage temperature and time. Hunter L value, was relatively stable with increased storage temperature and time, but the red and yellow hue, and Hunter a and b values respectively, were altered. Esters were positively correlated to flavour intensity, persistence of fruit flavour and overall impression of passion fruit juices. However, ethanol and ethyl acetate were positively correlated to off-flavour and presence of an after taste in the juices. These results show that passion fruit flavour attributes and colour are dependant on method of preservation, storage temperature and time. Passion fruit juice stored at -18ºC had a superior flavour and colour irrespective of the preservation method and storage time.

Key Words: Chromatography, flavour attributes, Passiflora edulis Sims., preservation method, storage temperature and time, volatile compounds

Des analyses chromatographiques du gaz sensoriel et de l'espace ont été faites pour évaluer les changements de la saveur du jus de fruit de la passion (Passi flora edudis Sims) pasteurisé et chimiquement conservé à 18,4-8 et 23ºC pendant une période de douze semaines. La couleur du jus était contrôlée au moyen d'un colorimètre Minolta de vérification des couleurs. Les attributs de la saveur, à l'exception de l'intensité d'acidité et de la douceur du jus pasteurisé étaient inférieurs au jus conservé chimiquement. Les chromatogrammes headspace de jus pasteurisé ont montré des légers changements dans le profil d'une composition volatile. Le jus pasteurisé était très foncé, moins rouge et moins jaune (L:24,66; a:-3,88; b:8,43) comparé au jus conservé chimiquement (L:33,22; a:-4,95; b: 15,47). Les attributs de la saveur, sans compter l'intensité d'acidité et de la douceur, dépendent de la température et du temps de conservation. La valeur L, était relativement stable avec l'augmentation de la température et du temps de conservation, mais les teintes jaune et rouge, avaient respectivement des valeurs de chasseurs a et b altérées. Les esters ont été positivement corrélés avec l'intensité de la saveur, la persistance de la saveur du fruit et l'impression générale du jus de fruit de la passion. Cependant, l'éthanol et l'acétate d'éthyle ont été positivement correlés au manque de saveur et à la présence d'un arrière-goût dans les jus. Ces résultats montrent que les attributs de la saveur du fruit de la passion et la couleur dépendent du temps et de la température de conservation. Le jus de fruit de la passion conservé à -18ºC avait eu une saveur et une couleur supérieures sans tenir compte de la méthode de préservation et du temps de conservation.

Mots Clés: chromatographie, attributs de la saveur, Passiflora edulis, méthode de préservation, le temps et la température de conservation, composés volatiles

Passion fruit, Passiflora edulis Sims., is grown in most tropical and subtropical countries. In Uganda, it is grown mainly for fresh juice production.

Flavour and colour are critical factors in evaluation of fruit products (Shewfelt, 1986; Askar et al. 1993). According to Shewfelt (1986) flavour is constituted of taste and aroma components. Chen et al. (1982, 1987) stressed that passion fruit flavour is a vital attribute which makes the juice a desirable ingredient for many formulated beverages and products (Yu & Chiang, 1986; Lue & Chiang, 1989; Arjona & Matta, 1991), and hence serves as a natural concentrate due to its unique flavour and high acidity. The unique flavour is attributed to several volatile compounds (Shimbamoto and Tang, 1990).

Passion fruit juice is preserved by both heat treatment and chemical preservation, although greater flavour deterioration is observed with heat preservation (Fang et al. 1986; Jagtiani et al. 1988). Frozen storage is superior to the two methods of preservation (Pruthi, 1963).

Uganda produces the purple, yellow and hybrid passion fruits, but the purple and hybrid fruits are most common. The purple and hybrid fruits are freshly extracted to passion fruit juice usually for immediate consumption, and output is far below the market demand. Accordingly, there is a strong interest in the commercial sector for large-scale production of passion fruit juice. However, continual supply of passion fruit juice in the market dictates employment of both heat and chemical preservation. This study evaluated the effect of preservation method, storage temperature and time on the flavour attributes and colour of passion fruit juice.

Extraction and preparation of passion fruit juice. Mature purple passion fruit (P. edulis Sims.) were purchased from Eugen AS, Oslo, Norway. The fruits were washed and stored in a cold room at 7ºC, a day prior to extraction. Pulp was extracted manually from three batches of 20 kg fruit each. The pulp was macerated and sieved using two nylon cheese cloths to separate pulp and seeds from juice. Percentage yield of juice based on the weight of fruit was between 27.6 - 30.5%. Sugar (45%, w/w) was added to juice, thereafter heated to 40 - 50ºC, and homogenised (Rannie homogeniser No. 4580/71, Rannie, Denmark) at approximately 178 kgm-2.

Juice treatments. Homogenised juice was divided into three portions. Portion A was pasteurised at 750C for 30 - 50 seconds by a plate heat exchanger (PHE Alpha- Laval P20-HB, Sweden); potassium sorbate (800 ppm) was added to portion B; and portion C received no treatment. Each of these portions were further subdivided into three portions and stored at -18, 4 - 8 and 23ºC. These portions were further subdivided into four portions for analysis at the beginning of the experiment, and after four, eight and twelve weeks. All juices were filled into 250 ml sterile glass bottles and tightly capped.

Physicochemical analysis of raw juice. Total titratable acidity (% anhydrous citric acid) was determined by titrating with 0.1 M NaOH, pH measured by a PHM 61 laboratory pH meter (Radiometer, Copenhagen) and total soluble solids (degrees Brix) by a hand refractometer (K-32, Kebo-Grave, Stockholm).

Sensory analysis. A panel of six judges familiar with passion fruit juice were trained prior to the start of the experiment by use of the triangular test method (ISO 4120, 1983). Flavour profiling (ISO 6564, 1985) was used to evaluate the juices. Flavour descriptive attributes of passion fruit juice unanimously agreed upon were flavour intensity, acidity (sourness), sweetness, persistence of passion fruit flavour, off-flavour, presence of an after taste and overall impression of the juice. The attributes were rated on a 7-point scale (1 = imperceptible to 7 = very pronounced). A score of 4.0 for flavour intensity, persistence of passion fruit flavour and overall impression was the optimum level below which attributes were considered undesirable while it was the desired level for acidity and sweetness intensity. A score of 1.0 was the desired level that indicated absence of an off-flavour and after taste in the juice. Judges were served with coded juice samples (10.8 - 12.0ºBrix) in white polypropylene cups at ambient temperature under normal light.

Headspace gas chromatography analysis. Qualitative changes of the volatile compounds of the juices were analysed by static headspace chromatography (HSGC) using the DANI HSS 3950 headspace sampler (DANI S.p.A-20052 Monza MI, Italy). Gas was sampled above the headspace of 10 g of test sample after equilibration (50ºC; 105 min) and injected at 180ºC into an HRGC 5300 Mega Series gas chromatograph (Carlo Erba Strumemtazione, 20090 Rodano MI, Italy). Volatile compounds were separated along a 25 x 0.53 mm i.d.; 5Fm film wall coated tubular column (CP-SIL 5CB, Chromapak) with a temperature programme of 53ºC, 1 min; 15ºC min to 70ºC, 2 min; 20ºC min-1 to 130ºC, 2 min; 30ºC min-1 to 160ºC, 12 min. Volatile compounds were detected by a flame ionisation detector at 200ºC.

Volatile compounds of passion fruit juice were identified by a dynamic headspace sampling procedure using a Hewlett-Packard gas chromatography mass spectrometer. Volatile compounds were obtained from the headspace of 25 ml of the test sample (at 50ºC; with nitrogen gas bubbled into sample at 50 ml^-1 min for 30 min) and adsorbed onto a Tenax GR tube (350 mg, 60 - 80 mesh, Alltech Associates). Volatile compounds were desorbed through two stages by an automatic thermal desorption Ainjector@ system (Perkin Elmer ATD 400, Beaconsfield Bucks , England). Tenax-GR tube was desorbed at 225ºC by helium gas (flow rate: 30 ml^-1 min; for 5 min) and compounds concentrated onto Tenax-TA (60 - 80 mesh, Chromapak) at -30ºC. Thereafter, rapid heating of the Tenax-TA from -30ºC to 250ºC (40ºC^-1s; maintained at 250ºC for 7 min) followed. Volatile compounds were automatically transferred to an HP 5890 Series II gas chromatograph and separated by a high resolution gas chromatograph column DB5 MS, 30 m x 0.25 mm i.d.: 1 Fm film, coated with 5% phenyl methyl siloxane (J and W Scientific INC., Folson, CA), using the same temperature programme. Tentative identification was achieved by matching mass spectra of the volatile compounds with published mass spectra (Wiley database) and by a comparison made between the GC-MS and the HSGC chromatograms.

Colour measurement of Passion fruit juice. Colour changes of passion fruit juice were measured by a tristimulus colorimeter (Minolta Chroma Meter CR-200) which measures the reflective colours of surfaces. The Judd Hunter L a b scale, a widely used colour scale for foods (Gnanasekharan et al. 1992) which corresponds most closely to the visual signals, was selected. L value was a measure of lightness or brightness, a value a measure of the red - green and b value a measure of the yellow - blue character of passion fruit juice. The colorimeter was calibrated against a standard white plate (L = 97.79, a = -0.43 and b = +2.25) and the lighting conditions selected were the CIE illuminant C. Five millimeters of passion fruit juice diluted in a ratio of 1:3 were sampled for analysis.

Statistical analysis. The experiment was a three-factor factorial design replicated thrice. The data were subjected to analysis of variance (General Linear Model, version 6.10, SAS Institute, NC) to determine significant differences among preservation methods and storage conditions. Treatment means were compared by Tukey T-test at 0.05%. Principal component analysis (SIRIUS version 2.3, Pattern Recognition Systems, Bergen, Norway) was used to determine the correlation between the volatile compounds and the sensory analysis data.

Some Physicochemical characteristics of raw passion fruit juice. The pH and total soluble solids ranges of passion fruit juice, 2.9 to 3.1 and 14.2 to 15.4ºBrix, respectively, were similar to those obtained in studies by Pruthi (1963), Casimir et al. (1981) and Jagtiani et al. (1988). However, total titratable acidity, 0.92 - 1.37% anhydrous citric acid, was unusually low .

Effect of preservation method on flavour attributes and colour of passion fruit juice. Pasteurisation altered (P<0.05) flavour attributes of passion fruit juice except for the acidity and sweetness intensity attributes (Table 1).

TABLE 1. Mean sensory scores for the flavour attributes of passion fruit juice preserved by different methods

Mean score values of the attributes in a column followed by the same letter are not significantly different (p£ 0.05). The scales used for scoring the flavour attributes were: Flavour intensity: 1 = imperceptible to 7 = very pronounced; Acidity: 1 = not acidic to 7 = very acidic; sweetness intensity: 1 = not sweet taste to 7 = very sweet: Off-flavour: 1 = none to 7 = very pronounced; presence of an after taste: 1 = none to 7 = very pronounced; persistence of fruit flavour: 1 = none to 7 = very pronounced; overall impression: 1 = very low to 7 = very high

The alteration was attributed to the changes in volatile compounds brought about by heat treatment (Pruthi, 1963; Casimir et al., 1981). Heat preservation is most detrimental to the flavour of passion fruit juice due to the high volatility of the aroma compounds (Casimir et al., 1981). Pasteurised juice had a higher score for off-flavour and after taste compared to chemically preserved and non-treated juices, probably due to the heat treatment. This change in flavour attributes of pasteurised juice (Table 1) coincided with the slight differences in peak heights of the volatile compounds of the pasteurised juice (Fig. 1). Casimir et al. (1981) also reported significant differences in peak heights of volatile compounds in passion fruit juice compared to fresh juice, and that, in addition to the reduction in concentration of the volatile compounds, hydrolysis of esters and acid-catalysed oxidation of the megastig-matrienes (subsitituted cyclohexenes) occurs.

The colour of passion fruit juice was likewise altered by method of preservation (Table 2). Pasteurised juice was darker, (low Hunter L value more red and less yellow than the chemically preserved and non-treated juices (Table 2). Fang et al. (1986), Isaacs et al. (1988) and Saenz et al. (1991) also reported that Hunter L value of fruit juice decreased on heating. The red and yellow hue of pasteurised juice also changed as indicated by the low Hunter a and b values as compared to the chemically preserved and non-treated juices (Table 2).

TABLE 2. Hunter Lab values of passion fruit juice preserved by different methods

Turkey comparison test showed that Hunter L a b values in a column followed by the same letter are not significantly different (p£ 0.05)

Colour differences that occurred were probably due to changes in the carotenoid pigments and development of furfural and hydroxy-methylfurfural compounds (Fang et al. 1986). Simpson (1985) attributed the loss of the orange-yellow colour of passion fruit juice to degradative reactions, such as oxidative, cis-trans isomerisation and formation of epoxides, which carotenoids are prone to under various processing conditions.

Effect of storage conditions on the flavour attributes and colour of passion fruit juice. Storage temperature significantly (P<0.05) affected the flavour attributes of passion fruit juices, except for acidity and sweetness intensity (Table 3).

TABLE 3. Effect of storage temperature on the mean scores for flayour attdbutes of passion fruit juice

Passion fruit juices stored at -18ºC had pronounced flavour intensity, fruit flavour, high overall impression, and had no off-flavour and after-taste compared to juices stored at refrigeration (4 - 8ºC) and ambient (23ºC) temperatures. The relatively pronounced off-flavour and after taste for juices stored at 4 - 8ºC and 23ºC were attributed to the fermented off-flavour of the non-treated juices detected by the judges.

Storage time likewise affected (P<0.05) the flavour attributes of passion fruit juices, except for acidity, sweetness intensity and presence of an after taste (Table 4).

Passion fruit juice at the start of the experiment had a pronounced flavour intensity compared to juices stored after 4, 8 and 12 weeks. The relatively pronounced off-flavour attribute for juices stored after 4, 8 and 12 weeks coincided with loss in flavour intensity. Despite the alteration in flavour intensity of these juices, scores for persistence of fruit flavour and overall impression were almost similar (Table 4).

Hunter L values did not change significantly (P>0.05) during storage, suggesting that the lightness parameter of passion fruit juice was relatively stable (Tables 5 and 6).

TABLE 5. Effect of storage temperature on Hunter L a b values of Passion fruit juice

Hunter L a b values in column followed the same letter are not significantly different (p£ 0.05)

TABLE 6. Effect of storage time on Hunter L a b values of passion fruit juice

Hunter L a b values in a column followed by the same letter are not significantly different (p£ 0.05)

However, increased storage temperature and time significantly altered (P<0.05) the red and yellow hue of passion fruit juice as reflected by the decrease in Hunter a and b values, respectively (Table 5). Although the yellow hue of passion fruit juice decreased with higher storage temperature, no significant difference (P>0.05) in Hunter b values for juice stored at -18ºC and 4 - 8ºC were noted. However, Francis (1985) and Rodriguez-Amaya (1993) attributed oxidative reactions as the main cause of carotenoid degradation leading to colour changes.

Relationship between volatile compounds and flavour attributes of passion fruit juice. Most identified passion fruit juice volatile compounds included esters, alcohols and ketones. Peaks containing the Edulans (I and II) and b-ionone volatile compounds, significant contributors of the rose-like aroma (Casimir et al. 1981, Tressl and Engel, 1983 and Jagtiani et al. 1988) of passion fruit juice, were found although their identification could not be confirmed. The trace quantities of megastigmatrienes, also responsible for the rose like aroma (Casimir et al. 1981), were probabaly hidden in the larger peak of hexyl hexanoate (Chen et al. 1982).

Correlation coefficients of selected volatile compounds and flavour attributes of passion fruit juice are shown in Table 7.

TABLE 7. Correlation coefficients of selected volatile compounds and flavour attributes of passion fruit juice

a- Fl, flavour intensity; b - PPF, persistence of passion fruit flavour; c - Ol overall impression; * -OF, off flavour; ** - AT, after taste

Most volatile compounds, particularly the esters, except ethyl acetate, were positively correlated to flavour intensity, persistence of fruit flavour and overall impression attributes. Nisperos-Carriedo and Shaw (1990) reported that the desirable flavour of juices is associated with mainly esters. Ethanol and ethyl acetate were however positively correlated to off-flavour and presence of an after taste, suggesting that these volatile compounds imparted an off-flavour and after taste to passion fruit juice. Judges noted fermented off-flavours in the non-treated juices. Though ethyl acetate was positively correlated to off-flavour and presence of an after taste, it contributes to fruit flavour and aroma (Nisperos-Carriedo and Shaw, 1990). Shamaila et al. (1992) noted that such correlations of flavour compounds with sensory attributes are indicative and important for the flavour quality changes during storage.

Pasteurisation was found to deteriorate passion fruit juice flavour and colour compared to chemical preservation. Passion fruit flavour attributes, except for acidity and sweetness intensity, were found to be dependant on the preservation method employed. Passion fruit flavour attributes, with the exception of acidity and sweetness intensity, were also found to be storage temperature and time dependant. However, -18ºC storage temperature insignificantly preserved flavour attributes and colour of passion fruit juice. The lightness parameter, Hunter L value, was relatively stable during storage compared to the red and yellow hue, the Hunter a and b values respectively.

The author greatly acknowledges the Norwegian Council of Universities= committee for Development of Research and Education (NUFU) for the financial support provided to do the study. Sincere gratitude is extended to Professor J. Apeland, Professor J. Narvhus and Dr. M. E. Magnus for their constructive criticism. Due thanks are extended to Professor K. Steinsholt for the valuable guidance in the statistical analysis of the data.

1. Association of Analytical Chemists. 1995. Fruits and fruit products. In: Official Methods of Analysis of AOAC International. 16th ed. Cuniff, P. (Ed.). Vol. II.
2. Arjona, H. E. and Matta, F. B. 1991. Post-harvest quality of passion fruit as influenced by harvest time and ethylene treatment. HortScience 26: 1297-1298.
3. Askar, A., El-Ashwah, F. A., Omran, H. T. and Labib, A. A. S. 1993. Colour stability of tropical nectars and a simple method for its determination. XXI Symposium of the International Federation of Fruit Juice Producers. Budapest. pp. 219-234.
4. Casimir, D. J., Kefford, J. F. and Whitefield, F. B. 1981. Technology and flavour chemistry of passion fruit juices and concentrates. Advances in Food Research 27:243-295.
5. Chen, C. C., Kuo, M. C., Hwang, L. S., Wu, J. S. B. and Wu, C. M. 1982. Headspace components of passion fruit juice. Journal of Agriculture and Food Chemistry 30:1211-1215.
6. Chen, S. L., Wu, J. S. B. and Wu, C. M. 1987. The preparation of concentrated passion fruit juice by methods of flavour recovery and enzyme treatment. Journal of Chinese Agricultural Chemistry Society 25:177-187.
7. Fang, T. T. Chen, H. E. and Chiou, L. M. J. 1986. Effects of heat treatment and subsequent storage on the quality of passion fruit (Passiflora edulis) juice. XIX Symposium of the International Federation of Fruit Juice Producers. Fruit juices for Europe: Advances in fruit and vegetable, and beverage technology. Den Haag. pp. 105-123.
8. Francis, J. F. 1985. Pigments and other colorants. In: Food Chemistry. 2nd ed. Fennema, O.R. (Ed). Marcel Dekker, Inc. New York and Basel. pp. 573-574
9. Gnanasekharan, V., Shwefelt, R. L. and Chinnan. 1992. Detection of colour changes in green vegetables. Journal of Food Science 57:149-154.
10. International Standard, ISO 4120. 1983. Sensory Analysis - Methodology - Triangular Test. 1st Ed.
11. International Standard, ISO 6564. 1985. Sensory Analysis - Methodology - Flavour Profile Methods. 1st Ed.
12. Isaacs, A. R., Bradley, B. F. and Nottingham, S. M. 1988. The storage stability of passion fruit concentrate. Food Technology in Australia 40:318-323.
13. Jagtiani, J., Chan, Jr, H. T. and Sakai, W. S. 1988. Passion Fruit. In: Tropical Fruit Processing. Academic Press Inc. San Diego. pp. 157 - 167
14. Lue, S. J. and Chiang, B. H. 1989. Deadcification of passion fruit juice by ultrafiltration and ion-exchange processes. International Journal of Food Science and Technology 24:395- 401.
15. Nisperos-Carriedo, M. O. and Shaw, P. E. 1990. Comparison of volatile flavour components in fresh and processed orange juices. Journal of Agriculture and Food Chemistry 38:1048-1052.
16. Pruthi, J. S. 1963. Physiology, Chemistry and Technology of Passion Fruit. Advances in Food Research. 2. Chichester, C. O., Mrak, E. M. and Steward, G. F. (Eds.), pp. 245-257. Academic Press. New York.
17. Rodriguez-Amaya, D. B. 1993. Stability of carotenoids during the storage of foods. In Shelf Life Studies of Foods and Beverages: Chemical, Biological, Physical and Nutritional Aspects. Chara-lambous, G. (Ed.). Elsevier Science Publishers.
18. Saenz, C., Costell, E. and Calvo, C. 1991. Effect of thermal treatment on the rheological behaviour and colour parameters of passion fruit juice. XI International Congress of Fruit Juice. Report of Congress. Sao Paulo.
19. pp. 495-500.
20. Shamaila, M. M., Powrie, W. D. and Skura, B. J. 1992. Analysis of volatile compounds from strawberry fruit stored under modified atmosphere packaging (MAP). Journal of Food Science 57:1173-1176.
21. Shewfelt, T. R. L. 1986. Flavour and colour of fruits affected by processing. In: Commercial Fruit Processing. 2nd ed. Woodroof, J. G. and Luh, B. S. (Eds.). AVI Publishing Co., INC. Westport. Connecticut.
22. Shimbamoto, T. and Tang, C. S. 1990. Minor tropical fruits - mango, papaya, passion fruit and guava. In: Food Flavours. Part C. The Flavour of Fruits. Morton, I. D. and Macleod, A. J. (Eds.), pp. 252-267. Elsevier Science Publishers. Amesterdam.
23. Simpson, K. L. 1985. Chemical changes in natural food pigments. In: Chemical Changes in Food Processing. AVI Publishing Co., INC. Westport. Connecticut.
24. Tressl, R. and Engel, K. H. 1983. Formation of esters and terpenoids in passion fruits and their importance to quality evaluation. In: Instrumental Analysis of Foods. Vol. 1. Charambolous, G. and Inglett, G. (Eds.), pp. 153-182. Academic Press. New York.
25. Yu, Z. R. and Chiang, B. H. 1986. Passion fruit juice concentration by ultrafilteration and evaporation. Journal of Food Science 51:1501-1505.

Copyright 1998, African Crop Science Society

The following images related to this document are available:

Line drawing images