From chocolates to cakes and from soft drinks to champagne_ by vivi07


									JFS E: Food Engineering and Physical Properties

Bubble-Included Chocolate: Relating Structure with Sensory Response
ABSTRACT: Bubbles impart a very unique texture, chew, and mouth-feel to foods. However, little is known about the relationship between structure of such products and consumer response in terms of mouth-feel and eating experience. The objective of this article is to investigate the sensory properties of 4 types of bubble-containing chocolates, produced by using different gases: carbon dioxide, nitrogen, nitrous oxide, and argon. The structure of these chocolates were characterized in terms of (1) gas hold-up values determined by density measurements and (2) bubble size distribution which was measured by undertaking an image analysis of X-ray microtomograph sections. Bubble size distributions were obtained by measuring bubble volumes after reconstructing 3D images from the tomographic sections. A sensory study was undertaken by a nonexpert panel of 20 panelists and their responses were analyzed using qualitative descriptive analysis (QDA). The results show that chocolates made from the 4 gases could be divided into 2 groups on the basis of bubble volume and gas hold-up: the samples produced using carbon dioxide and nitrous oxide had a distinctly higher gas hold-up containing larger bubbles in comparison with those produced using argon and nitrogen. The sensory study also demonstrated that chocolates made with the latter were perceived to be harder, less aerated, slow to melt in the mouth, and having a higher overall flavor intensity. These products were further found to be creamier than the chocolates made by using carbon dioxide and nitrous oxide; the latter sample also showed a higher intensity of cocoa flavor. Keywords: aerated chocolate, bubbles, sensory evaluation, X-ray tomography, 3D analysis



rom chocolates to cakes and from soft drinks to champagne, bubbles are found in a variety of food products. While they do not add any nutritional value to foods, they do change the texture and mouth-feel characteristics: for example, brittleness or creaminess of an aerated chocolate bar, light mouth-feel of whipped cream, sponginess of cake or bread, fizziness of beer or champagne. A stable foam on top of beer or cappuccino also adds visual appeal to the product (Campbell and Mougeot 1999). Thus, in recent years, new aerated products are appearing in our supermarkets with novelty value and increased consumer acceptance. Aerated products are also perceived to be lighter in terms of calories, thereby gaining a positive market image. Despite widespread industrial practice of bubble inclusion into foods, the mechanisms of bubble formation and behavior are far from being well understood. In this context, bubble mechanisms have recently been studied by various authors (Massey 2002; Haedelt and others 2005; Jakubczyk and Niranjan 2006). Recent studies have attempted to develop an understanding of the relationship between process parameters and dispersion characteristics. Haedelt and others (2005) investigated bubble formation in liquid-tempered chocolate induced by applying vacuum. This article also considered the role played by the following ingredients in the development of the dispersion characteristics: milk fat, cocoa butter, vegetable fats, and various emulsifiers. Jakubczyk and Niranjan (2006) reviewed the formation and stability of bubbles in whipped cream. Massey (2002) studied the formation of bubbles in cake batter and determined how

MS 20060606 Submitted 11/8/2006, Accepted 1/26/2007. Authors Haedelt and Beckett are with Nestl´ Product Technology Centre, Nestec York, YO91 e 1XY, U.K. Author Niranjan is with School of Food Biosciences, The Univ. of Reading, Whiteknights PO Box 226, Reading RG6 6AP U.K. Direct inquiries , to author Haedelt (E-mail:

the operating conditions changed the aeration profile. Campbell and Mougeot (1999) reviewed the literature on aerated food products based on the type of food, the methods of including bubbles, the stabilization mechanism involved, and bubble life-time and content. Despite the availability of literature on the devices used to aerate food systems and techniques to study the effect of operating parameters, little is known about the relationship between structure of such products and consumer response in terms of mouth-feel and eating experience. The quality and acceptance of bubble-containing products are very much dependent on the dispersion characteristics. In the context of aerated chocolates, consumer perception can be related to either the smooth mouth-feel of microaerated chocolate (characterized by the presence of smaller bubbles hardly visible to the unaided eye) or the crispness and brittleness of macroaerated chocolate (characterized by the presence of larger bubbles) (Haedelt 2005). Thus, a difference in bubble size can result in very different mouth-feel responses. In practice, differences in bubble sizes and gas hold-up can be created in chocolates by using different gases in the process. The specific aim of this work is to gain an understanding into the relationship between the measured dispersion characteristics and the sensory response of bubble-containing chocolates produced by using 4 different gases: carbon dioxide, nitrogen, nitrous oxide, and argon. Solidified chocolate dispersions were characterized by measuring gas hold-up (that is, the volume fraction of gas based on the total dispersion volume) and bubble size distribution. For the first time, 3D image analysis of bubble-containing chocolates was undertaken after reconstructing 3D structure from a series of X-ray tomographs. Barigou and Lim (2004) studied a number of cellular food products such as strawberry mousse and chocolate muffins, quantifying 2D analysis of bubble sections and determining 3D characteristics by using a stereological modeling technique which essentially assumed
Vol. 00, Nr. 0, 2007—JOURNAL OF FOOD SCIENCE E1

C 2007 Institute of Food Technologists doi: 10.1111/j.1750-3841.2007.00313.x

Further reproduction without permission is prohibited

E: Food Engineering & Physical Properties

Bubble-included chocolate . . .
that bubbles were spherical (Saltykov area analysis method). A 3D system pressure of 4.5 bar, chocolate temperature of 30 ◦ C, 28 rpm reconstruction software used in the present work allows the volumes mixing head speed, and pump speed at 38 rpm. of bubbles to be determined accurately. As mentioned earlier, aerated chocolates were produced using 4 different gases: carbon dioxide, nitrogen, nitrous oxide, and argon. All gases were food grade and supplied by BOC (Guildford, Surrey, Materials and Methods U.K.). In all cases, the chocolate was saturated with the gas and Chocolate resulted in a good expansion after pressure release. A typical milk chocolate recipe, supplied by Nestl´ PTC York, U.K., e was used which contained sugar, dried whole milk, cocoa mass, hard Cooling and setting of the chocolate nonlauric vegetable fat, emulsifier (lecithin), and vanilla flavor. The After pressure release the samples were deposited into plastic total fat content varied between 28.7% and 29% (w/w). The chocolate molds (33 × 100 × 15 mm, 50 mL capacity), kept at room temperacontained a minimum of 25% cocoa solids and 14% milk solids. ture, and transferred immediately into a refrigerator kept at 10 ◦ C to allow setting.

Tempering procedure

Before chocolate can satisfactorily be processed from liquid to solid it must be tempered. Chocolate mixes were tempered in a Sollich Turbo temper (Solltemper U, 40 kg) to form appropriate crystals (Nelson 1999). Tempering is important, as it promotes controlled crystallization of the desired triglycerides, which critically influence setting characteristics, foam stability, and demolding properties. The kettle in which the chocolate was held was a stirred tank, which could be heated or cooled. From this tank, the chocolate was pumped into the tempering portion of the machine where it was seeded with a small amount of tempered chocolate. The temper was carried out by cooling the chocolate from an initial temperature of 41.5 to 28 ◦ C to cause reasonably rapid crystal growth, and then the chocolate was gradually heated to 29.5 to 30 ◦ C in a heat exchanger. The extent of temper was measured using a tempermeter (Tricor System, Model 505A, Elgin, Ill., U.S.A.) to ensure that the chocolate set in the correct crystal form (β crystals). A built-in algorithm in this instrument gave the extent of temper in chocolate temper units (CTUs). The principle of this method has been described by Nelson (1999). The tempered chocolate at 29.5 ± 0.5 ◦ C was directly transferred to the batch holding tank for external gas injection.

Density and gas hold-up measurement
The gas hold-up (ε) was calculated by comparing the density of the aerated chocolate (ρ f ) with the gas-free density of solid chocolate (ρ i ). ε = 1− ρf ρi × 100 (1)

The method used to determine densities (ρ f and ρ i ) has been reported in an earlier paper (Haedelt and others 2005). An average of 5 values was taken following each gas hold-up determination and the results are expressed together with the corresponding standard deviation values.

E: Food Engineering & Physical Properties

Structure analysis by X-ray tomography
The method used to obtain X-ray images of cross-sections of aerated chocolate has been reported in an earlier paper (Haedelt and others 2005). A 3rd-generation cone-beam X-ray CT scanner, described in detail by Jenneson and others (2002), was used. X-rays were passed through the chocolate sample in many different directions, and images of a number of contiguous slices, spaced 100 µm apart, were generated. These slices were used to visualize bubble sections using Image Pro Plus software (Media Cybernetics, Silver Spring, Md., U.S.A.) in order to determine the distribution of bubble section areas and diameters. 3D images were also reconstructed from the above planer images by using Image Pro Plus software, 3D Constructor Version 5.1. Given the complexity involved in reconstructing the entire chocolate cylinder, it was divided into 10 volume sections each of approximately 330 mm3 . Bubbles of volume less than 0.001 mm3 and greater than 4 mm3 were ignored in the analysis, since they were deemed not to be typical within the dispersion. For each chocolate, the mean bubble volume, the standard deviation, and the number of bubbles (in the volume considered) of 5 replicated measurements were determined (see Table 2).

Batch process with external gas injection (Figure 1)
The tempered chocolate was poured into a holding tank (1) where pressure was applied. The chocolate was then pumped toward the mixing head (4) by a dosing pump (2). Just prior to entering the mixing head, provision is made to inject air or other gases into the chocolate under controlled pressure (3). The gas was finely dispersed into the chocolate mass by the action of the stator and rotor arrangement in the mixing head. A needle valve (5) enabled sample withdrawal when the mixture was returned to atmospheric pressure and the dissolved gas came out of solution (6). The temperature was controlled by heated water jackets around the tank and all the pipes. As it is a total recycle system, the chocolate is recycled into the batch tank and then gets pumped to the external gas injection again, without being degassed. The operating conditions were chosen as follows:

Figure 1 --- Batch process with external gas injection


JOURNAL OF FOOD SCIENCE—Vol. 00, Nr. 0, 2007

Bubble-included chocolate . . .
Sensory analysis
In order to relate the dispersion characteristics with sensory properties an organoleptic study on the 4 different chocolate samples was undertaken. An attempt was made to determine the differences between the intensities of selected sensory attributes. The panelists for this study were 20 members of the School of Food Biosciences, Univ. of Reading (U.K.), who were trained prior to the sensory study. The 1st part of the training involved exposing the panelists to samples of 3 commercial aerated chocolates: Nestl´ Aero Japan, Nestl´ Cailler e e Air Switzerland, and Nestl´ Bros Holland, all of which have different e texture and flavor. Panelists were encouraged to generate as many terms or descriptors to describe the sensory attributes. The whole panel then discussed all the terms generated and formed a standardized set of descriptors. The commercial chocolates also served as reference descriptors; for example, Aero Japan represented an extremely strong cocoa flavor, whereas Bros Holland represented an extremely aerated product with a lower flavor profile. Furthermore, a vocabulary development training was also carried out to ensure that the panelists were familiar with the terms and comfortable using them. In addition, terms were grouped into 3 categories: texture, flavor, and basic taste, in which the terms reflecting a common sensation were combined and ambiguous terms could be more easily identified. Texture descriptors were further subdivided into biting, chewing, sucking, and after swallowing. The panelists agreed on 14 terms out of the 30, which are summarized in Table 1. The panel was further screened for taste impairment by undertaking thresholds tests on sweetness, acidity, and bitterness. A series of solutions of increasing concentration of sodium chloride, sucrose, and quinine sulphate were presented to panelists to determine whether they could perceive and name the basic tastes correctly. As the need for quantitative assessment of attributes is especially important in descriptive analysis, a line scale was selected to provide the panelists with an infinite number of places in which to indicate the relative intensity for an attribute. The line scale was anchored at both ends. The panelists were asked to make a vertical line across the horizontal line at the point that best reflected the relative intensity of the particular term. The panelists were further encouraged to make full use of the line scale by clearly marking differences between the samples. By measuring the distance along the line to the mark, a numerical value was obtained. A statistical package specially developed for sensory evaluation (FIZZ, Biosystemes, France) was used for this analysis. At the time of the analysis, the chocolate samples were 2 weeks old. They had been stored in a refrigerator and were presented at room temperature (20 to 21 ◦ C) in individual booths. Samples were presented in random order with coded identification. All 4 samples were evaluated twice, henceforth referred to as Replicate 1 and 2, to check for reliability of the responses.

Results and Discussion
Effect of gas type on structural properties
Table 2 shows 4 chocolate samples produced by bubbling nitrogen, argon, nitrous oxide, and carbon dioxide. Table 2 summarizes the results on 2D and 3D X-ray image analysis and gas hold-up values for the different gases but using the same base recipe. From Table 2, it is obvious that nitrogen results in the formation of much smaller bubbles (d mean = 0.13 mm) than other gases, and also produces very low gas hold-up values (ε = 29%). Argon also produces small bubbles (d mean = 0.19 mm) but the hold-up values are slightly higher (ε = 34%). As both gases result in the formation of small bubbles that are not readily detected by the human eye, their aeration characteristics will henceforth be referred to as microaeration. Carbon dioxide and nitrous oxide, on the other hand, produce larger 2D bubble sections (mean diameter of 0.51 mm and 0.41 mm, respectively, which are equivalent to mean bubble volume of 0.18 mm3 and 0.15 mm3 ). Moreover, the gas hold-up values are also much greater (68% and 66%, respectively) (see Table 2). Larger bubbles and greater values of voidage result from the higher solubilities of these gases in cocoa butter (Haedelt 2005). The higher solubility also results in the formation of a greater number density of nuclei initially, which promote coalescence to form larger bubbles upon pressure release. Haedelt (2005) has described a simple technique that can be used to compare the solubility of gases in a given liquid under otherwise similar conditions. Using this method, the solubility of the 4 gases in chocolate can be arranged in the following order that represents progressively decreasing solubility: nitrous oxide, carbon dioxide, argon, and nitrogen. Table 2 also compares image analysis across a 2D plane and a reconstructed 3D section. It is immediately obvious that the 2 techniques essentially yield similar values of mean bubble diameter and are therefore comparable. It may be noted that some gas will escape on setting but this is consistent for all samples.

Table 1 --- Glossary of terms used to describe sensory properties of bubble-containing chocolate Term Texture: Biting Hardness Definition Limits of the scale soft–hard nil–extreme dense–aerated moist–dry smooth–gritty nil–extreme short-long nil–extreme nil–extreme nil–extreme nil–extreme nil–extreme nil–extreme nil–extreme E3

Resistance to pressure from the front teeth when biting through the sample (bite resistance) Crumbliness Degree to which the bar breaks into many and small pieces Aeration Perceivable aeration when biting through the bar Chewing Dryness Product that breaks into bits when chewed (dry) as opposed to a product that forms a smooth paste (evaluated over the first chews) Grittiness Perceived particle size as evaluated between tongue and palate Stickiness Measurement of the stick-power between the tongue and palate Sucking Melting time Time necessary for the product to melt in the mouth (by chewing the product 3 times and then sucking it) After swallowing Mouth coating Sensation of a fat film perceived in the mouth after swallowing the product Flavor: Overall flavor intensity Intensity of the combination of all the flavors perceived Cocoa Intensity of the cocoa flavor Creaminess Intensity of creamy (milky/buttery) flavor Basic tastes: Sweet Intensity of the sweet taste Acidic Intensity of the acidic taste Bitter Intensity of the bitter taste

Vol. 00, Nr. 0, 2007—JOURNAL OF FOOD SCIENCE

E: Food Engineering & Physical Properties

Bubble-included chocolate . . .
Effect of gas type on measured responses
The 4 aerated chocolate samples were also evaluated by a sensory panel and 2-way analysis of variance (ANOVA) was carried out to show statistically which attributes differed between the samples. Separate ANOVAs were carried out for each attribute with samples and panelist response as variables. The results are presented in Table 3 which shows the mean score of the 2 samples for each attribute and shows which attributes differ between the samples and whether the response between the panelists was significantly different. Furthermore, the significance of the interaction between the samples and panelist response was investigated. As evident from Table 3 (row nr 1, 3, and 7), the 4 different chocolate samples were perceived to be significantly different in terms of hardness, aeration, and melting time (P < 0.001). They were also found to differ in terms of overall flavor intensity, cocoa flavor, and creaminess (P < 0.05) (row nr 9, 10, and 13 in Table 3). For the other attributes, the differences perceived were nonsignificant. This does not imply that products are the same with respect to such attributes. It simply means that a difference cannot be detected for the attributes concerned. The P-values for the panelists also indicate that there are differences between panelists. Ideally the interaction between panelists must not be significant. But the 20 panelists made different use of the line scale for each intensity. Variation between panelists was highly significant for all attributes except crumbliness, melting time, overall flavor intensity, cocoa flavor, and bitter taste. The P-values also show interactions between samples and panelists but this was only observed for 2 attributes, hardness, and dryness. Overall, there were, however, significant differences between the 4 bubble-containing chocolate samples, indicating that they are significantly different, even when low interactions between panelists are taken into account. By calculating the least significant difference (LSD), we were able to indicate which samples differed significantly at a 95% level. This can also be seen by plotting the different attributes and the assessors responses to the different chocolates in terms of intensities, as shown in Figure 2. The diagram aids in grouping the chocolates on the basis of different sensory intensities.

Effect of gas type on measured responses and structural properties
It is obvious from Figure 2 that in terms of the extent of aeration and hardness, nitrogen and argon gave very similar results while carbon dioxide and nitrous oxide samples were perceived to be more aerated and softer. The differences between these 2 sets of gases were significantly high (P < 0.001). This is consistent with the experimental data on dispersion structure, where carbon dioxide and nitrous oxide result in the formation of larger gas voids and

E: Food Engineering & Physical Properties

Table 2 --- 2D and 3D bubble dimensions and gas hold-up values for 4 different gases Chocolate type 2D image analysis of X-ray images CO 2 N2O Ar N2

2D reconstructed X-ray image 2D d mean Nr of bubbles 3D image analysis of X-ray images Chocolate type

0.51 ± 0.32 227 ± 3 CO 2

0.41 ± 0.21 222 ± 27 N2O

0.19 ± 0.1 234 ± 5 Ar

0.13 ± 0.09 218 ± 8 N2

3D reconstructed X-ray image 3D d mean 0.58 ± 0.26 0.50 ± 0.25 Bubble volume 0.18 ± 0.03 0.15 ± 0.03 Nr of bubbles 234 ± 37 240 ± 37 Gas hold-up values as determined from density measurement by the flotation method ε (%) ± St. Dev. 68 ± 0.45 66 ± 0.17
St. Dev. are given, in case of d mean they resemble the bubble section diameter spread.

0.18 ± 0.17 0.02 ± 0.01 175 ± 23 34 ± 0.22

0.16 ± 0.12 0.014 ± 0.01 149 ± 16 29 ± 0

Table 3 --- Mean panel scores for sensory attributes for the 4 chocolates Attributes Texture Hardness Crumbliness Aeration Dryness Grittiness Stickiness Melting time Mouth coating Overall intensity Cocoa Sweet Acidic Creaminess Bitter CO 2 26.7 46.6 65.7A 35.6 33.9 35.4 24.6B 36.0 47.5B 44.3AB 55.1 21.0 38.5BC 21.4

N2O 22.7 43.0 68.1A 33.8 33.1 39.8 27.4B 39.4 56.2A 50.3A 56.1 26.1 33.3C 28.6

Ar 64.2 34.4 23.5B 44.5 34.9 34.9 44.2A 42.8 57.4A 44.2AB 61.5 20.4 43.8AB 20.9

N2 70.6 34.8 21.5B 41.8 32.2 32.2 42.7A 41.4 48.9B 40.5B 61.6 19.6 47.8A 19.8




LSD 11.76 8.67


∗∗∗ ∗ ∗ ∗∗∗


∗∗∗ ∗∗ ∗




∗∗∗ ∗∗ ∗∗




10.12 6.59 6.81 8.22

NS = not significant. ∗ P < 0.05. ∗∗ P < 0.01. ∗∗∗ P < 0.001. Means with different letters are significantly different at P < 0.05. (LSD = least significant difference of samples = t distribution ∗ SE).


JOURNAL OF FOOD SCIENCE—Vol. 00, Nr. 0, 2007

Bubble-included chocolate . . .
Figure 2 --- Sensory perception of aerated chocolate using 4 different gases

larger bubble volumes, hence thinner lamellae that will melt faster. Thus, a chocolate that is less aerated (that is, lower gas hold-up) is denser, and therefore perceived to be harder. A difference between the chocolates was also observed in relation to melting time where the nitrogen and argon samples scored higher, suggesting that they offered greater resistance to melting in the mouth than the samples made with nitrous oxide and carbon dioxide. This is expected since the bubbles are smaller in size and fewer in number in the case of argon and nitrogen, resulting in a lower gas hold-up. Hence these chocolates take longer to melt. For the attribute “overall intensity,” argon and nitrous oxide samples scored highly in terms of flavor intensity and differed significantly (P < 0.01) from the carbon dioxide and nitrogen samples. This perceived difference is very interesting; since the base recipe used was the same for all the samples, the only difference was the gas used to create the structure. A difference in texture can be explained by noting the difference in gas solubilities in chocolate, which consequently results in a different bubble size and hold-up. However, a difference in perceived flavor is more difficult to explain. One possibility is to relate this observation to the melting time, which, in turn, can be related to the structure. Based on this hypothesis slower melting (as observed in the case of argon and nitrogen) should result in a higher overall flavor intensity. The relationship between melting and flavor release has also been addressed by Ziegler and others (2001). However, the sample made using nitrous oxide does not fit with this hypothesis. One possible explanation is that the chocolate gassed with nitrous oxide melts rather quickly in the mouth and releases the flavor more readily. It is interesting to note that the nitrous oxide sample was also judged to be strongest in terms of cocoa flavor, and differed significantly (P < 0.05) from the nitrogen sample. Argon and carbon dioxide samples, on the other hand, were perceived to be comparable in terms of cocoa flavor. Overall, the difference in intensity between the 4 bubble-containing chocolate samples is, however, much lower for the flavor attributes, than for the texture attributes. In the case of creaminess, the nitrogen sample differed significantly from the nitrous oxide sample (P < 0.01), whereas the argon sample was similar to the nitrogen sample. Carbon dioxide was perceived to give similar creaminess to nitrous oxide, with a lower but insignificantly different value to argon. Creaminess can also be related to the structure and melting time of the aerated chocolate. Nitrogen and argon—perceived by the panel to be more creamy—caused the formation of small bubble nuclei and lower gas hold-up values. As the chocolate was less aerated, it was perceived to be harder, and took longer to melt, but gave the chocolate a “silky” mouth-feel, which was judged to be highly creamy. Samples made from carbon dioxide and nitrous oxide, on the other hand, melted quickly in the mouth as a result of the larger volume fraction of gas present.

his study concluded that gases such as nitrous oxide and carbon dioxide, which result in the formation of larger bubbles and greater gas hold-up in chocolates, are perceived to be significantly different from nitrogen and argon in terms of textural attributes, such as extent of hardness, creaminess, and melting time in the mouth. Interestingly, carbon dioxide and nitrous oxide were found to be more soluble in chocolate. Hence they resulted in larger voids, which led to the chocolate melting very fast in the mouth and being perceived to be less hard, more aerated, and less creamy. Argon and nitrogen samples, on the other hand, caused the formation of smaller bubbles and a lower gas hold-up. These samples were perceived to be creamier and harder. The nature of the gas used clearly influenced flavor perception even though the chocolate base recipe was essentially the same. Chocolates made by using nitrous oxide were perceived to possess more flavor in general, with a particularly strong cocoa flavor. While this may be advantageous in some situations, the relatively high cost of nitrous oxide may limit its use.


Special acknowledgments to BBSRC and Nestl´ PTC York for their e financial support. Thanks are also due to Peter Cooke, Nicolas Galley, ¨ Martina Suß, Karin Gartenmann, Julia Sarhy, and David Coleman, who, at the time the article was written, were employed at Nestl´ PTC e York, as well as to Caroline Hargreaves at the Dept. of Soil Science and Prof. Don Mottram at the School of Food Biosciences at the Univ. of Reading.

Barigou M, Lim K. 2004. X-ray micro-computed tomography of aerated cellular food products. Intl. Conference on Engineering and Food, Montpellier, France. Campbell GM, Mougeot E. 1999. Creation and characterisation of aerated food products. Trends Food Sci Technol 10(9): 283–96. Haedelt J. 2005. An investigation into bubble inclusion into liquid chocolate [PhD thesis]. School of Food Biosciences. U.K.: Univ. of Reading Haedelt J, Pyle DL, Beckett ST, Niranjan K. 2005. Vacuum induced bubble formation into liquid tempered chocolate. J Food Sci 70(2): 159–64. Jakubczyk E, Niranjan K. 2006. Transient development of whipped cream properties. J Food Engr 77(1):79–83. Jenneson PM, Gilboy WB, Morton EJ, Gregory PJ. 2002. An X-ray micro-tomography system optimised for the low-dose study of living organisms. Appl Radiat Isot 58:177– 81. Massey AH. 2002. Air inclusion mechanisms and bubble dynamics in intermediate viscosity food systems [PhD thesis]. School of Food Biosciences. U.K.: Univ. of Reading. Nelson RB. 1999. Tempering. In: Beckett ST, editor. Industrial chocolate manufacture and use. 3rd ed. Oxford: Blackwell Science. p 231–58. Ziegler GR, Mongia G, Hollender R. 2001. The role of particle size distribution of suspended solids in defining the sensory properties of milk chocolate. Intl J Food Prop 4(2):175–92.

Vol. 00, Nr. 0, 2007—JOURNAL OF FOOD SCIENCE


E: Food Engineering & Physical Properties

To top