INFLUENCE OF OIL SUBSTITUTION BY INULIN ON THE HEADSPACE CONCENTRATIONS OF CARVONE AND LIMONENE RELEASED FROM CARAWAY PREPARATIONS

Grażyna Bortnowska
Department of Food Technology, West Pomeranian University of Technology, Szczecin, Poland

ABSTRACT

Effects of fat replacement by inulin on the release of carvone and limonene from a model mixture were studied in relation to the fat component concentration and the flavouring form used (natural caraway flavouring or crushed seeds of caraway, Carum carvi (L.)). The static headspace analysis applied to the model mixture showed the amounts of carvone and limonene, the major caraway seed volatile flavour compounds released from the model mixture, to be rapidly diminishing with increased oil content in the fat component, the effect being particularly noticeable within the 1 to 10%, concentration range, regardless of the flavouring form used. Replacement of the maximum of 30 and 50% of oil with inulin, i.e., 3 and 5% inulin concentration, respectively, in an oil – water mixture resulted in an almost unchanged amount of caraway flavour released to the model mixture headspace, compared to the inulin-free sample. It was only when 80% of oil were replaced by inulin that volatility of carvone and limonene was observed to increase, the increase averaging about 30% in those mixtures with 10% equivalent of fat component concentration, and showed a slight dependence on the flavouring form used. As shown by the sensory analysis, inulin replacing 80 and 100% of oil shifted the model mixture flavour, rendering the flavour of a substance inulin was derived from more perceptible, particularly in samples with 20 and 40% equivalent of fat component concentrations.

Key words: caraway flavouring, inulin, static headspace, odour intensity.

INTRODUCTION

Flavour release from food has been acknowledged as an important factor in determining the perceived flavour quality of many foods [25]. The dynamics of flavour release from a food product has been described by a number of mathematical models [1, 10, 13]. However, the complexity of the problem is evidenced by the fact that results of new studies are continuously appearing, also because flavour measuring techniques are being continuously improved [14, 16, 19, 28].

Low-energy food manufacture calls for the reduction of fat content. Proteins and carbohydrates being the major materials to manufacture fat substitutes from [5, 22, 27]. The group of carbohydrate-derived substitutes includes inulin, a natural product hot water-extracted from endive roots. Inulin can be used as a dietary fibre-enriching component of numerous food products. It is also regarded as a prebiotic on account of its promoting the growth of desirable bacteria controlling intestinal function [26].

Lipids are the food components that are most efficient in inhibiting volatile flavour compounds [6, 10,12, 20]. In addition, they greatly affect food taste and improve flavour stability [5]. It is difficult to predict how volatile flavour compounds will react with lipid substitutes. Most probably, the reaction will differ from that with a lipid because the physico-chemical structure of the system will change. In their studies on interactions between flavour compounds and protein macromolecules, [21] used lipid substitutes manufactured from whey, maltodextrin, egg whites, and microcrystalline cellulose. They observed the substitutes to only slightly affect the composition of volatile compounds of flavouring substances in the model system headspaces. The release of volatile flavour compounds from inulin-containing food products was treated only by [8] who, in their study on the behaviour of selected volatile flavour compounds in systems involving fresh cheese, olein, and inulin, concluded that inuline behaviour did not change in the systems studied.

On the other hand, the available literature contains no papers dealing with the release of volatile flavour compounds from inulin-enriched food products.

Analysis of flavour compounds above the surface of a system in thermodynamic equilibrium, or the so-called static headspace analysis, produces an objective and measurable indication of flavour release from foods [9, 15, 18, 29]. Volatile flavours enter the nose orthonasally (i. e. through the nostrils) before eating and ,, sniffing” of food provides our first flavour impression. This signal is derived from the volatiles in the air above a food (the headspace) and can be measured relatively easily [25].

The study was aimed at following effects of gradual substitution of oil with inulin on the release of carvone and limonene from a model mixture flavoured with caraway applied as crushed caraway seeds and a natural caraway flavouring.

MATERIALS AND METHODS

Raw materials

The study involved:

– two forms of caraway: dried seeds of the plant Carum carvi (L.), marketed by Kamis and purchased in a retail outlet, and a natural caraway seed flavouring manufactured by the Pollena-Aroma Aromatic Substances Factory in Warsaw, purchased directly from the manufacturer. Dry caraway seeds were crushed in an NMK 110 food processor, equipped with an MKK-100 spice grinder (Spomasz). According to manufacturer’s specification, the natural caraway flavouring consisted of volatile oil and rape seed oil. The caraway oil concentration in the natural caraway flavouring was 7.5% [24], the crushed caraway seeds containing usually 4-7% [23]. During the experiment, the flavourings were stored at 4±2°C in the dark, in tight containers, the amounts to be used being retrieved as needed for the analyses;

– inulin, a carbohydrate lipid substitute, manufactured in the Netherlands and marketed as Frutafit^® TEX, purchased from Hortimex in Konin;

– vegetable rapeseed oil manufactured by the Oil Plant at Kruszwica and purchased in a retail outlet.

Model mixture

The model mixtures were prepared by dispersing fat components (oil and/or lipid substitute – inulin) in water, using an MPW–302 laboratory homogeniser (Mechanika Precyzyjna, Warszawa) for 30 s at 3.500 rpm; one of the flavourings (0.4% relative to the model mixture weight) was added, gradually substituting inulin for 30, 50, 80, and 100% of oil, 1 g inulin being equivalent to 4 g oil [7]. The final fat component concentration in the model mixture was equivalent to 1, 3, 5, 10, 20, and 40% of oil w/w.

The homogenised samples were assayed immediately. The model mixtures were prepared stage-wise, from identical raw materials. To account for any changes that could occur in the raw materials during storage, a blank sample, consisting of the caraway flavouring homogenised with water, was prepared each time.

Static headspace analysis

The composition of headspace volatile flavour compounds was determined in a Perkin Elmer gas chromatograph coupled with mass spectrometer and equipped with a headspace autosampler (Perkin Elmer International C.V. Switzerland).

A mass detector with electron ionisation was applied. Phase separation was performed in a PE-SMS capillary column (30 m × 0.25 mm × 0.25 ×m) at a temperature varying from 50 to 150°C and using 10°C×min^-1 gradient and helium as a carrier gas; the column flow rate was 0.7 ml×min^-1. The samples were incubated for 15 min at 70°C, once a thermodynamic equilibrium in the model mixture headspace had been reached; caraway and inulin concentrations used were those selected for this study. Peaks were identified from mass spectra of the compounds separated, by comparing them with Wiley’s spectrum library and by comparing the spectra obtained with those of the standards. The assays were run in duplicate, two measurements for each sample being taken.

Sensory analysis

The model mixtures were evaluated orthonasally for the caraway odour intensity, using the odour profiling method [2] under conditions specified by the Polish standard PN-ISO 6658, by a 4-strong panel the members of which fulfilled the basic criteria for odour detection and identification (PN-ISO 5496). Following trial evaluations and discussions with the panel members, characteristic features of a model mixture odour profile were specified. It was decided that important was the evaluation of intensity of caraway oil and other (foreign) odours. Should the other odour profile be selected, the panel members were asked to determine its nature. Each panel member evaluated the intensity of the odour profiles selected by smelling the model mixtures and expressed the perceived odour intensity on a scale of 0 to 4: an intensity scored 0 when the odour was imperceptible, 4 reflecting intensive odour [2].

Statistical treatment

Significance of differences between selected samples was tested with a 1-way analysis of variance at the error probability level of p £ 0.001 or p £ 0.05. The analyses were run with the Microsoft Excel 2000 software.

RESULTS AND DISCUSSION

The static headspace method was used to quantify lipid or inulin – aroma compound interaction in oil-water mixtures.

Changes in carvone and limonene (the two major volatile substances producing odour in caraway) release to the model mixture headspace were assessed by comparing chromatographic peak areas of the substances relative to carvonene and limonene chromatographic peak areas in the control sample (without oil and/or inulin) headspace, similarly to [4].

The relationship between the measured relative chromatographic peak areas of each flavour and the equivalent of fat concentration was plotted for each inulin ratio (Figs. 1, 2). Those samples the fat concentration of which contained oil only were denoted by A; the samples with 30, 50 and 80% of oil replaced by inulin were denoted by B, C and D respectively; the inulin – only samples were denoted by E. An increased oil concentration was observed to be accompanied by clearly decreasing contents of carvone and limonene in the model mixture headspaces (Figs. 1-2, Sample A). Similar lipid effects on the release of volatile flavour compounds from model emulsions were observed [6, 10, 20]. [3] explained that the effects were related to hydrophobic properties of the majority of volatile flavour compounds which are better soluble in the lipid phase than in water. This is why lipids are food components that are most efficient in reducing partial pressure of volatile flavour compounds in the headspace, thus reducing human ability to detect a flavour [17]. Furthermore, a 1 to 10% addition of oil to the mixture was observed to rapidly diminish carvone and limonene volatility in the headspace, while a further increase of oil concentration (20 and 40%) did not result in such pronounced changes, both in those samples flavoured with caraway seeds and those enriched with natural caraway flavouring (Figs. 1-2, Sample A). [8] who studied the release of selected esters from model systems using the dynamic headspace method and found too, a 10% olein addition to be sufficient to significantly inhibit flavour perception.

Substituting 30% and 50% of oil with inulin, i.e., adding 3% or 5% inulin, at the maximum, to an oil-water mixture did almost nothing to change the volatility of carvone and limonene released to the headspace of the model mixture, relative to an inulin-free sample (Figs. 1-2, Samples A, B, C). [8] found addition of inuline at a concentration as high as 10% to do nothing to modify the release of certain volatile flavour compounds. From the technological standpoint it is however, important to find an inulin concentration at which the amount of carvone and limonene in the headspace of a model mixture is changed and to see if the change is acceptable to consumers. At this stage, it can be contended that inulin, used as a fat substitute at the amount of 5% at the maximum, replacing thus 50% of oil in a mixture containing 40% equivalent of fat component concentration, can reduce the energy content of a food product without changing its flavour, regardless of whether the product is caraway seed – or natural caraway flavouring – enriched.

When 80% of oil were replaced by inulin, a clear increase in the amount of carvone and limonene in the headspace of the model mixture was observed and depended on the flavouring type (Figs. 1, 2, Sample D). The amount of volatiles released to the headspace of the flavoured mixtures with 80% of oil replaced by inulin was higher than in those samples containing less fat component and decreased with increasing concentration of the volatiles, regardless of the flavouring type used (Figs. 3, 4). In addition, the amounts of carvone and limonene released from those model mixtures were higher in the caraway seed-flavoured samples than in the natural caraway flavouring-enriched ones, the difference being particularly pronounced at equivalent of fat component concentration lower than 5%, i.e., inulin concentrations of 1% at the maximum. Most probably, the small amount of vegetable oil, a flavouring solvent, added to the model mixture can also inhibit the release of carvone and limonene, which is particularly well visible at low equivalent of fat component concentrations. In the mixtures with 10% equivalent of fat component concentration, in which 80% of oil were substituted by inulin, the amount of carvone and limonene in the headspace was observed to increase by an average of about 30%, the increase amounting to about 10% - for both types of flavouring – in the mixtures with 20% and 40% equivalent of fat component concentrations (Figs. 3, 4, Sample D).

Legends:

Legends:

The headspace of the model mixture with 100% of oil being replaced by inulin showed a clearly highest increase of carvone and limonene, regardless of the flavouring type used and fat component concentration (Figs. 1-2, Sample E). [11] contended that carbohydrates, particularly polysaccharides applied in small amounts, reduce volatility of flavour compounds in relation to water as a result of molecular interactions, while mono- and bisaccharides cause the so-called salting effect, thus increasing the flavour release, relative to water. A similar process was noticeable in the headspace of the caraway seed-flavoured model mixture with 100% of oil replaced by inulin, particularly during carvone release (Figs. 1–4, Sample E). Thus, it follows that inulin inhibits the release of carvone and limonene in the headspace much less readily than oil, but only when inulin replaces at least 80% of oil in a model mixture.

Legends:

Legends:

Odour intensity was evaluated in three selected model mixtures which differed in their equivalent of fat component concentration (from 1% to 40%). Mixtures containing oil only, those with inulin as a surrogate of 80% of oil, and those with inulin only were evaluated. The sensory analysis was applied to evaluate the intensity of caraway, oil, and other (foreign) odours. Results of the sensory analyses are presented in Figures 5 and 6, significance of differences being summarised in Tables 1 and 2. Intensity of the caraway seed odour in the model mixtures was observed to rapidly decrease with increasing oil concentration; simultaneously, the intensity of oil odour was increasing as well. At higher equivalent of fat component concentrations, an additional odour, described as bland, was detected (Figs. 5 and 6). Addition of as little as 3% of oil to the model mixture reduced caraway seed odour perception by an average of about 50%, relative to the control, regardless of the flavouring type used. On the other hand, caraway seed odour was not detected at all in the caraway seeds-flavoured samples containing 20% and 40% of oil. The sensory evaluation of the model mixtures containing high concentrations of equivalent of fat component is consistent with the static headspace analysis results, as no carvone peak could be identified in the headspace chromatograms (Figs. 4 and 6, Samples A). On the other hand, the intensity of oil odour increased so rapidly as to render the changes between the 1% oil samples and the control statistically very highly significant, both in the natural caraway flavouring-enriched and caraway seeds-flavoured mixtures (Table 1). Replacement of 80% of oil with inulin increased the caraway seed flavour perception many times, relative to the control, the pronounced increase beginning from the model mixture with 3% equivalent of fat component concentration; the intensity of oil odour was at the same time clearly reduced, the differences being significant or very highly significant (Figs. 5, 6 and Table 2). When testing the samples in which 80% of oil were replaced by inulin, the evaluators noted an increase, with increasing equivalent of fat component concentration, of a delicate odour close to that of starch. The caraway seed odour of the model mixture in which 100% of oil were replaced by inuline was comparable to the control and did not depend on the equivalent of fat component concentration, which was also supported by statistical analysis (Table 1). The oil odour was perceived only in those samples containing 1 and 3% fat component concentration that had been enriched with natural caraway flavouring (Table 1). Moreover, replacement of 100% of oil with inulin changed the nature of the model mixture odour. In addition to the intense caraway seed odour, the evaluators reported perceiving a pleasant fruit aroma the intensity of which increased with increasing equivalent of fat component concentration.

It follows then that inulin added to a mixture at an amount replacing 80% of oil facilitates perception of caraway seed odour, while substitution of 100% of oil does not alter the odour perception, relative to the control. It has to be pointed out, however, that the inulin odour profile, which is typical of a raw material inulin is obtained from, may alter the model mixture odour. In addition, the nature of mixture odour depends on the inulin concentration in aqueous solution or in oil-water mixtures.

CONCLUSIONS

1. The release of carvone and limonene from a model mixture decreases with increasing oil concentration, regardless of flavouring type used.

2. Substitution of 30 and 50% of oil with inulin does not change the amounts of carvone and limonene in the model mixture headspace, relative to an inulin-free sample.

3. The amounts of carvone and limonene in the headspace of model mixtures containing 3% and 5% equivalent of fat component concentrations, in which 80% of oil were replaced by inulin, were almost twice as high when the mixtures were flavoured with caraway seeds, compared to adding natural caraway flavouring.

4. Replacement of 80% of oil by inulin in the mixtures with 10% equivalent of fat component concentration increased the amount carvone and limonene in the headspace by an average of about 30% and to some extent depended on the flavouring type applied.

5. Inulin replacing 80% and 100% of oil changed the nature of the model mixture oduor towards that of the raw material inulin was derived from, particularly in samples with 20 and 40% equivalent of fat component concentrations.

REFERENCES

1. Andriot I., Harrison M., Fournier N., Guichard E., 2000. Interactions between methyl ketones and beta-lactoglobulin: sensory analysis, headspace analysis and mathematical modeling. J. Agric. Food Chem., 48, 4246-4251.

2. Baryłko–Pikielna N., 1975. Zarys analizy sensorycznej żywnosci [Sensory analysis of food; an outline]. WNT, Warsaw [in Polish].

3. Buttery R. G., Bomden J. L., Guadagni D. G., Ling L. C., 1973. Flavor compounds: Volatilities in vegetable oil and oil – water mixtures. Estimation of odour thresholds. J. Agric. Food Chem., 21, 198-201.

4. Charles M., Lambert S., Brondeur P., Courthaudon J-L., Guichard E., 2000. Influence of formulation and structure of an oil-in-water emulsion on flavor release. in: Flavor release. D. D. Roberts, A. J. Taylor (eds.) American Chemical Society, Washington, DC.

5. Duda Z., 1998. Zamienniki tłuszczu stosowane w przetwórstwie mięsnym [Fat substitutes in meat industry]. Gosp. Mięs., 2, 22-26 [in Polish].

6. Ebeler S. E., Pangborn R. M., Jennings W. G., 1988. Influence of dispersion medium on aroma intensity and headspace concentration of menthone and isoamyl acetate. J. Agric. Food Chem., 36, 791-796.

7. Encyklopedia dodatków do żywnosci, 2000. [Enclyclopaedia of food additives] www.hortimex.com.pl [in Polish].

8. Gijs L., Piraprez G., Perpete P., Spinnler E., Collin S., 2000. Retention of sulfur flavours by food matrix and determination of sensorial data independent of the medium composition. Food Chem., 69, 319-330.

9. Girard B., Nakai S., 1991. Static headspace gas chromatographic method for volatils in canned salmon. J. Food Sci., 55, 1271-1274.

10. Harrison M., Hills B. P., Bakker J., Clothier T., 1997. Mathematical models of flavor release from liquid emulsions. J. Food Sci., 62, 4, 653-658.

11. Godshall M. A., 1997. How carbohydrates influence food flavor. Food Technol. 51, 1, 63-67.

12. Haahr A. M., Bredie W. L. P., Stahnke L. H., Jensen B., Refsgaard H. H. F., 2000. Flavour release of aldehydes and diacetyl in oil/water system. Food Chem., 71, 355-362.

13. Harrison M., Hills B. P., 1996. A mathematical model to describe flavour release from gelatine gels. Int. J. Food Sci. Technol., 31, 167-176.

14. Juteau A., Cayot N., Chabanet C., Doublier J. L., Guichard E., 2004. Flavour release from polysaccharide gels. Trends Food Sci. Technol.,15, 7-8, 394-402.

15. Kolb B., Ettre L. S., 1997. Static Headspace-Gas Chromatography. Theory and Practice. Wiley-Vch, New York.

16. Lian G., Malone M. E., Homan J. E., Norton I. T., 2004. A mathematical model of volatile release in mouth from the dispersion of gelled emulsion particles. J. Controll. Rel., 98, 139-155.

17. Leland J. V., 1997. Flavor interactions: the greater whole. Food Technol., 51, 1, 75-80.

18. Maeztu L., Sanz C., Andueza S., De Pena M., Bello J., Cid C., 2001. Characterization of espresso coffee aroma by static headspace GC-MS and sensory flavor profile. J. Agric. Food Chem., 49, 5437-5444.

19. Nakamura A., Yoshida R., Maeda H., Furuta H., Corredig M., 2004. Study of the role of the carbohydrate and protein moieties of soy soluble polysaccharides in their emulsifying properties. J. Agric. Food Chem., 52, 5506-5512.

20. Rooks K. B. de, 1997. How lipids infulence food flavor. Food Technol., 51, 1, 60-62.

21. Schirle-Keller J. P., Chang H. H, Reineccius G. A., 1992. Interaction of flavor compounds with microparticulated proteins. J. Food Sci., 57, 6, 1448-1451.

22. Słomińska L., 1999. Węglowodanowe zamienniki tłuszczu [Carbohydrate lipid substitutes]. Przem. Spoż., 7, 12-15 [in Polish].

23. Strzelecka H., Kamińska J., Kowalski J., Walewska E., 1982. Chemiczne metody badań roœlinnych surowców leczniczych [Chemical methods of analysis of plant pharmaceuticals]. PZWL., Warsaw [in Polish].

24. Sokołowska B., 2004. Personal information of the Head of Aroma Laboratory. Pollena-Aroma Flavouring Substances Factory in Warsaw (Poland).

25. Taylor A. J., 2000. Technique for measuring volatile release in vivo during consumption of food. in: Flavor release. D. D. Roberts, A. J. Taylor (eds.) American Chemical Society, Washington, DC.

26. Tomomatsu H., 1994. Health effects of oligosaccharides. Food Technol., 10, 61- 65.

27. Tyszkiewicz I., 1992. Zamienniki tłuszczu w technologii żywnoœci o obniżonej energetycznoœci [Fat substitutes in low-energy food technology]. Przem. Spoż., 5-6, 132-134 [in Polish].

28. Ye A., Hemar Y., Singh H., 2004. Influence of polysaccharides on the rate of coalescence in oil-in-water emulsions formed with highly hydrolyzed whey proteins. J. Agric. Food Chem., 52, 17, 5491-5498.

29. Zhon M., Robards K., Glennie-Holmes M. Helliwell S., 1999. Analysis of volatile compounds and their contribution to flavor in cereals. J. Agric. Food Chem., 47, 10, 3941- 3953.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.