37

E. T. Contis et al. (Editors) Food Flavors: Formation, Analysis and Packaging Influences © 1998 Elsevier Science B.V. All rights reserved

393

Comparison study of UHT milk aroma L. Hashim and H. Chaveron Laboratoire Biophysicochimie et Technologie Alimentaires, University de Technologie de Compi^gne, B.P. 20529, 60205 Compiegne-France. Abstract The flavor of milk may change when the product is submitted to thermal treatments. The heating of milk causes the formation of volatile compounds from milk components. Volatile compounds of UHT whole milk, UHT semiskimmed milk, and UHT skimmed milk were analysed u s i n g gas chromatography. The odor intensity of different molecules was realized by a sniffing test of the volatile compoimds of different milks at the outlet of the gas chromatography-capillary column. Sensory analysis of different milks was conducted by a trained taste panel. It showed t h a t milk flavor was affected by thermal t r e a t m e n t and milk composition. According to the different analytical techniques described, relevant differences were found among the studied milks. 1. EWRODUCTION The consumer acceptance and preference for milk as a beverage is influenced by its flavor more than any other attribute (1). Flavor is a property detected by the senses, in particular taste and smell, and thereby requires taste panel work for its evaluation. Milk has a pleasant mouth-feel, determined by its physical nature, i.e. an emulsion of fat globules in a colloidal aqueous solution, and a slightly salty and sweet taste due to the presence of salts and lactose (2). The flavor of fresh milk, although characteristic, is normally of a low intensity. Heated milk should have acceptable flavor characteristics, the milk from which it is processed m u s t meet appropriate physical, chemical, microbiological and sensory quality standards. When milk is heated, changes in flavor occur, the kind and intensity of the flavor depending on the time and temperature of the treatment (3). The term heated flavors is used to include all the flavors which are produced by thermal processing of fluid milk. At least 400 volatile compounds have been detected in milk processed in different ways (4). The flavor of ultra-high-temperature (UHT) processed milk has been described as "cooked, cabbagey, and sulfur". The intensity of this flavor and

394

associated odor has been correlated with the degree of free sulfhydryls (SH) and volatile sulfides liberated via the heat denatioration of the whey protein (5, 6). The lipid components of milk are important contributors to milk flavor. Compounds such as volatile fatty acids, dicarbonyls, and monocarbonyls impart flavor to milk though present only in trace amounts. Various ketones, saturated aldehydes , and unsaturated aldehydes affect flavor though present in concentrations of parts/million or parts/billion (7). Flavor changes in milk arise because of changes in its chemical constituents. The various types of flavor defect in milk have been reviewed (8). Therefore, researchers studying flavors use a combination of taste panel work and chemical analysis (9). The purpose of the this research was to study the differences in flavor profiles of UHT whole milk, UHT semi-skimmed milk, and UHT skimmed milk by gas chromatography-sniflfing tests and sensory analysis. 2. EXPERIMENTAL DATA 2.1 Samples UHT whole milk, UHT semi-skimmed milk, and UHT skimmed milk were purchased in France. These milks were processed by UHT systems and filled in Tetra Brik 1 Liter containers. 2.2 Analjiical analysis Steam distillation-microextraction was used to extract the volatile compounds of milks (10). Samples of 100 mL of each milk were analyzed to obtain the volatile extract. Volatile compounds were separated by capillary column gas chromatography (GC) using a Girdel-30 equipped with a flame ionization detector (FID). A CP Wax 52B (polyethyleneglycol, 50m X 0.32mm) fused silica capillary column (Chrompack) operated with helium as the carrier gas was employed. The column temperature was programmed from 50°C to 220°C at a rate of 5°C/min. Injector and detector temperature were set at 250°C. Chromatographic data were processed with a computing integrator (Shimadzu C-R6A). The odor intensity of different molecules was realized by sniffing different milk volatiles at the outlet of the gas chromatography-capillary column. 2.3 Sensory evaluation Sensory analysis of different milks was done by a trained taste panel using a flavor profile test with different descriptos (10 students from the Technical University of Compiegne were trained to do the tests). A 10-point scale was used. Representative samples were independently and randomly presented for evaluation.

395

3. CHROMATOGRAPHIC ANALYSIS Figure 1 shows the gas chromatogram of UHT whole milk aroma.

Figure 1. Capillary gas chromatogram of UHT whole milk volatiles

The aroma chromatograms of different milks were, in general, very similar. The most important differences between them were the intensities and the areas of certain peaks. Using sniff-test, the odor intensity and odor description was determined for 16 peaks. These peaks show a typical milk aroma. The results are presented in Table 1 and Figures 2, 3, and 4. It should be noted that throughout this study the only property evaluated was the milky aroma. The data shown in Figures 2, 3, and 4 clearly demonstrate that UHT whole milk has the most interesting milky odor. It can be seen from these figures that peaks 1, 2, 5, 7, and 10 in UHT whole milk have odor intensities superior to those in the other milks. These peaks have shown the most interesting and typical milky aromas. Table 1 shows the odor descriptions of various peaks. It can be noticed that the odors vary from buttery, creamy, fruity, chemical, milky, sweet, roasted, and burnt. The two latter aromas have low odor intensities and were detected only at the end of the chromatogram.

396

V3

u o

IS

O

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16

Peak number Figure 2. Odor intensity of UHT Skimmed milk (Peak numbers are as indicated in Figure 1)

0^

o

o

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16

Peak number

Figure 3. Odor intensity of UHT Semi skimmed milk (Peak numbers are as indicated in Figure 1)

397

en

o O

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16

Peak number Figure 4. Odor intensity of UHT Whole milk (Peak numbers are as indicated in Figure 1) Table 1 Odor description of volatile compounds from UHT milks Peak number

Retention timeCmin)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

4.4 5.5 7.1 9.6 10.4 13.4 14.3 17.4 20.5 22.4 24.5 26.3 31.5 34.9 38.2 40.6

L 16

Odor description Buttery 1 creamy 1 creamy creamy | milky milky milky-fruity milky-slightly milky-chemical milky-sweet milky milky milky-chemical milky milky-roasted milky-btirnt

398

4. SENSORY ANALYSIS The flavor of particular interest in heated milk is the "cooked" flavor. This flavor changes rapidly during the early days of storage. The vocabulary used for description is also not straightforward, and terms, such as "cooked", "cabbagey", "sulphury" and "caramellised", are all frequently used (5). One of the best accounts of the flavor changes in milk on heating and during storage is given by Ashton (11). Sensory evaluation provides the most practical method for monitoring the type and intensity of heated flavors. Figures 5, 6, and 7 show the intensity of the milk descriptors studied. The descriptors sweet, creamy, thickness, milky and hedonic (most pleasant) have higher scores and intensity in whole milk compared to other milks. It can be seen that these differences are in linear relation with the composition of different milks. It should be noted that differences between whole milk and semiskimmed milk are lower than differences between semi-skimmed milk and skimmed milk. The flavor profile of whole milk is more acceptable than that of the other milk, even with a few off-flavors.

a 0^

u o C/5

Descriptors

Figure 5. Intensity of different descriptors of UHT skimmed milk

399

V3

a

o

Descriptors Figure 6. Intensity of different descriptors of UHT semi-skimmed milk

0)

o C/5

1 CO

o U

1 u

4>

1

1 <

'3 o

M

^

P^

H

Descriptors

Figure 7. Intensity of different descriptors of UHT whole milk

400

5. CONCLUSIONS Additional research is needed to identify the nature and the intensity of different volatile compounds. These results could be compared to those obtained from dried milks, and other dairy products to establish a method for monitoring the type and intensity of heated flavors. This study shows the use of chromatographic techniques for monitoring the concentration of volatile compounds associated with heated flavors and the sensory evaluation of these compounds using olfactory techniques with trained panelists.

a REFERENCES 1. E.L. Thomas, J. Dairy Sci., 64 (1981) 1023. 2. D.J. Manning and H.E. Nursten, Developments in Dairy Chemistry-3, Fox (ed), London and New York, 1985. 3. M.M. Calvo and L. Hoz, Int. Dairy J., 2 (1992) 69. 4. H.T Badings and R. Neeter, Neth. Milk Dairy J., 34 (1980) 9. 5. R.S. Mehta, J. Food Protection, 43 (1980) 212. 6. A.P. Hansen, K.R. Swartzel and Giesbrecht, J. Dairy Sci., 63 (1980) 187. 7. J.E. Kinsella, Chemistry and Industry, 2 (1969) 36. 8. H.T. Badings, Dairy Chemistry and analysis, Walstra&Jennes (eds), 1984. 9. M.J. Lewis, Modern Dairy Technology-1, Robinson (ed), London and New York, 1986. 10. L. Hashim and H. Chaveron, Food Research International, 6 (1994) 537. 11. T.R. Ashton, J. Soc. Dairy Technol., 18 (1965) 65.