Dry mixture in particulate form for preparation of liquid foods with dispersed gas bubbles

Abstract

The present invention provides a dry mixture in particulate form containing a gas release agent, a flavour component and a hydrocolloid. Upon dissolution in water, gas bubbles are released into the continuous liquid phase, and these bubbles remain dispersed in the continuous liquid phase. These dry mixes can be used for preparation of liquid food products or beverages.

Claims

1. A composition in the form of a dry mixture in particulate form for preparation of a beverage or liquid food composition containing dispersed gas bubbles in a continuous liquid phase, the dry mixture in particulate form comprising: an instant flavour component in particulate form; a water-soluble gas release agent in particulate form that releases gas bubbles upon reconstitution in water; and xanthan gum in particulate form, wherein the xanthan gum is obtained from the fermentation of Xanthomonas campestris pathover campestris, deposited with the American Type Culture Collection (ATCC) under the accession no. PTA-11272, wherein the xanthan gum is present in the composition at a weight fraction of less than 4.0 wt % and provides an apparent yield stress of at least 0.3 Pa within a period of 30 seconds after addition of water to reconstitute the xanthan gum; and wherein the instant flavour component is suitable to prepare a beverage or liquid food composition selected from the group consisting of: soups, bouillons, sauces, gravies, and/or seasonings; other savoury food products; tea and tea-based beverages, containing an extract from the plant Camellia sinensis; herbal infusions, preferably containing an extract selected from mint, camomile, rooibos, rosehip, hibiscus, raspberry, or any combination of these; ice cream and/or desserts and/or milk shakes, which are intended for serving at a temperature below 0 C.; soy-based beverages, wherein these beverages in reconstituted form contain at least 0.3% by weight of ingredients originating from soybean, wherein the ingredients comprise a soy protein; dressings; and spreads.

2. A composition according to claim 1, wherein the composition comprises one or more native starches.

3. A composition according to claim 1, wherein the dry mixture in particulate form comprises pregelatinised starch or pregelatinised modified starch at a concentration of less than 0.5 wt %, based on dry weight.

4. The composition according to claim 1, wherein the weight fraction of xanthan gum is from 1.0 wt % to less than 4.0 wt %.

5. The composition according to claim 1, wherein the water-soluble gas release agent further comprises a free flowing aid.

6. The composition according to claim 5, wherein the free flowing agent is selected from the group consisting of silicon dioxide, tricalcium phosphate, and combinations thereof.

7. A composition in the form of a beverage or liquid food product containing gas bubbles in the continuous liquid phase, obtainable by bringing a composition according to claim 1 into contact with water.

8. A method for preparation of a beverage or liquid food product, comprising bringing a composition according to claim 1 into contact with water.

9. A method according to claim 8, wherein the weight ratio between dry mixture in particulate form and water ranges from 1:100 to 1:1.

10. A method according to claim 8, wherein the temperature of the water ranges from 40 C. to 100 C.

Description

DESCRIPTION OF FIGURES

(1) FIG. 1 Pictures of aerated instant mushroom soup samples, from Example 1. All pictures taken about 1 minute after adding water to the dry mix.

(2) FIG. 2 Pictures of aerated instant mushroom soup samples, from Example 2. All pictures taken 30 seconds after adding water to the dry mix.

(3) FIG. 3 Pictures of aerated instant mushroom soup samples, from Example 2. All pictures taken 40 seconds after adding water to the dry mix.

(4) FIG. 4 Graphs showing translation of precision spheres as function of time, experiments from Example 3. 3-1: .circle-solid. HDPE 5.69 mm, .Math. HDPE 3.17 mm; .square-solid. HDPE 3.17 mm; .diamond-solid. PS 4.76 mm; .box-tangle-solidup. PS 4.76 mm. 3-2: .circle-solid. PS 4.76 mm; .Math. HDPE 5.69 mm; .square-solid. HDPE 3.17 mm. 3-3: .circle-solid. HDPE 3.17 mm (upper curve); .circle-solid. PS 4.76 mm (lower curve).

(5) FIG. 5 Graphs showing translation of precision spheres as function of time, experiments from Example 3. 3-4: HDPE 3.17 mm; .Math. PS 4.76 mm; .square-solid. HDPE 5.69 mm. 3-5: .circle-solid. PS 4.76 mm. 3-6: .circle-solid. HDPE 3.17 mm, .Math. HDPE 3.17 mm; .square-solid. HDPE 5.69 mm; .diamond-solid. PS 4.76 mm; .box-tangle-solidup. HDPE 5.69 mm; PS 4.76 mm; .circle-solid. PS 4.76 mm.

(6) FIG. 6: Graphs showing translation of precision spheres as function of time, experiments from Example 3.

(7) FIG. 7 Pictures of aerated milk tea sample, from Example 4. Pictures taken 1, 5, 15 and 30 minutes, respectively, after adding water to the dry mix.

(8) FIG. 8 Pictures of aerated soy milk sample, from Example 5. Pictures taken 1.5, 5, 15, and 30 minutes, respectively, after adding water to the dry mix.

EXAMPLES

(9) The following non-limiting examples illustrate the present invention.

(10) Raw Materials

(11) Composition instant mushroom cream soup powder as used (ex Unilever Germany, Heilbronn, Germany).

(12) TABLE-US-00001 TABLE 1 Composition of dry instant mushroom soup mixes Ingredient Mix 1 Mix 2 Native potato starch, 8% moisture 11.1% Native potato starch, Granulated 22.2% Salt 5.9% 8.8% Yeast extract (18% NaCl) 5.8% 8.7% Creamer 32.9% 57.9% Flavours, spices, herbs 5.7% 8.6% Mushroom powder 10.7% 16.0% Native potative starch, 8% moisture: ex Sdstrke GmbH (Schrobenhausen, Germany). Native Potato Starch Granulated: contains 87% native potato starch and 13% glucose syrup (maize), ex Avebe (Veendam, The Netherlands). Creamer: contains palm oil and palm oil stearin (76.8%), lactose (6.6%), Na, Ca caseinate (7.8%), potassium phosphate dibasic (1.0%), glucose syrup (7.8%). Xanthan gum: Keltrol AP, Keltrol AP-F, and Keltrol RD ex CP Kelco (Nijmegen, The Netherlands).

(13) Keltrol AP and AP-F are described and claimed in WO 2012/030651 A1.

(14) The particle size distribution of Keltrol AP, Keltrol AP-F powders was determined in house, using a Mastersizer 2000 (Malvern Instruments Ltd., Malvern, Worcestershire, UK), equipped with a Sirocco powder accessory. The average sizes were:

(15) TABLE-US-00002 D3,2 D4,3 [micrometer] [micrometer] Keltrol AP 53 109 Keltrol AP-F 30 78 Modified starches: Agglomerated Prejel VA70, and Eliane SC160 from Avebe (Veendam, The Netherlands). Prejel is a pregelatinised hydroxypropyl distarch phosphate of potato origin; and Eliane is a pre-gelatinised waxy potato starch containing more than 99% amylopectin. Gas release agent: Vana-Cappa B01 ex FrieslandCampina Kievit (Meppel, The Netherlands). Ingredients of the powder are maltodextrin, modified starch (starch sodium octenyl succinate), and silica free flowing agent. Contains nitrogen gas under pressure, and the gas release is about 22 mL per gram dry agent upon dissolution in water.

Example 1 Suspending Air Bubbles in Instant Mushroom Soup with Xanthan Gum

(16) To show the ability of hydrocolloids combined with gas release agent and flavour compound to suspend gas bubbles, the following experiments were carried out.

(17) A powdered premix was made containing 10.0 g mushroom soup mix (mix 1 as in Table 1), 3.0 g gas release agent, and the required amount of hydrocolloid as indicated in Table 2, thoroughly mixed, and put into a 300 mL tall form glass beaker. 150.0 g of hot water, just after boiling, was added to the dry premix in one swift motion and the contents were vigorously stirred with a spoon for 30 sec. The volume of soup mix and water was 160 mL. Timing was started at the moment the water was poured. After stirring, the beaker was put on a stand and pictures were taken at preset time intervals.

(18) TABLE-US-00003 TABLE 2 Hydrocolloids used and description of results, in soup mix 1 as in Table 1. Aerated Foam Initial Aerated Foam bulk layer gas bulk layer liquid volume bubble liquid volume Hydro- volume on top volume volume on top Hydro- colloid after 40 after in bulk after 5 after 5 colloid amount sec 40 sec liquid min min Exp. type [g] [mL] [mL] [mL] [mL] [mL] 1-1 Keltrol 0.2 218 <10 58 208 <10 AP-F 1-2 Keltrol 0.4 212 5 52 202 5 AP-F 1-3 Keltrol AP 0.2 203 5 43 150 50 1-4 Keltrol AP 0.4 203 5 43 200 6 1-5 Keltrol RD 0.2 153 53 0 153 44 1-6 Keltrol RD 0.4 152 49 0 152 45

(19) An amount of 0.2 g hydrocolloid means that the concentration of the hydrocolloid is about 0.12% by weight of the prepared instant soups. The weight of the gas is not taken into account here. An amount of 0.4 g hydrocolloid means that the concentration of the hydrocolloid is about 0.24% by weight of the prepared instant soups. The concentration of native starch that is present in the dry soup mix 1 leads to a concentration of about 2.0% by weight of the of the prepared instant soups.

(20) The required apparent yield stress for a spherical bubble having a diameter of 100 micrometer in the soup mix containing xanthan gum would be 0.35 Pa (calculated with equation 1), in order to keep the bubble dispersed. For a 200 micrometer bubble the required apparent yield stress would be 0.70 Pa.

(21) In experiments 1-1, 1-2, and 1-4 there was no clear interface visible between a little foamy layer on top of the continuous liquid phase (bulk liquid) and the continuous liquid phase. In these experiments the used hydrocolloids provided the necessary yield stress sufficiently rapid in order to suspend the gas bubbles in the continuous liquid phase. The amount of foam layer on the top was negligible as compared to the total volume of the bulk liquid. In experiment 1-3 an interface was visible, showing a foam layer on top of the continuous liquid phase.

(22) In experiments 1-5 and 1-6 a regular xanthan gum is used, as described in the prior art. This xanthan gum does not provide the yield stress as required to keep gas bubbles in the continuous liquid phase. A sharp cut interface between a foamy top layer and the continuous liquid phase was visible, and most gas bubbles were present in the foam layer on top.

(23) The size of the gas bubbles in these samples was estimated to range from about 150 to 180 micrometer, as determined by bright field optical light microscope (Malvern Morphology G3).

(24) FIG. 1 shows pictures of the instant soups of the experiments as listed in Table 2. All pictures were taken 58 or 60 seconds after adding water to the soup mix. These pictures show the difference in the amount of foam layer on top. Experiments 1-1, 1-2, and 1-4 do not show an interface at all, and the gas bubbles remain in the continuous liquid phase. Experiment 1-3 shows an interface which is not very sharp, with more foam on top of the continuous liquid phase than experiments 1-1, 1-2, and 1-4. Experiments 1-5 and 1-6 show a sharp interface, with a clear thick foam layer on top of the continuous liquid phase.

Example 2 Suspending Air Bubbles in Instant Mushroom Soup with Modified and Native Starches

(25) Similar experiments as in Example 1 were done in order to determine the effect of various starch types on the dispersion of gas bubbles.

(26) A powdered premix was made containing 10.0 g mushroom soup mix (mix 1 as in Table 1), 3.0 g gas release agent, and the required amount of hydrocolloid as indicated in Table 3, thoroughly mixed, and put into a 300 mL tall form glass beaker. 150.0 g of hot water, just after boiling, was added to the dry premix in one swift motion and the contents was vigorously stirred with a spoon for 30 sec. The volume of soup mix and water was 170 mL. Timing started at the moment the water was poured. After stirring, the beaker was put on a stand and pictures are taken at preset time intervals.

(27) TABLE-US-00004 TABLE 3 Starches used and description of results, in soup mix 1 as in Table 1. Aerated Foam Initial Aerated Foam bulk layer gas bulk layer liquid volume bubble liquid volume Hydro- volume on top volume volume on top Hydro- colloid after 40 after in bulk after 5 after 5 colloid amount sec 40 sec liquid min min Exp. type [g] [mL] [mL] [mL] [mL] [mL] 2-1 Prejel 4.0 209 5 39 201 8 VA70 2-2 Eliane 4.0 212 7 42 200 8 SC160 2-3 Native 4.0 205 7 35 199 7 starch* *native starch: Native Potato Starch Granulated as in Table 1.

(28) A concentration of 4 g of hydrocolloid means that the concentration of the hydrocolloid is about 2.4% by weight of the prepared instant soups. The weight of the gas is not taken into account here. The concentration of native starch that is present in the dry soup mix 1 leads to a concentration of about 2.0% by weight of the of the prepared instant soups. So in experiment 2-3 the total concentration of native starch is about 4.4% by weight.

(29) Also in these experiments the size of the gas bubbles in these samples was estimated to range from about 150 to 180 micrometer, as determined by bright field optical light microscope (Malvern Morphology G3).

(30) The required apparent yield stress for a spherical bubble having a diameter of 100 micrometer in the soup mix containing xanthan gum would be 0.35 Pa (calculated with equation 1), in order to keep the bubble dispersed. For a 200 micrometer bubble the required apparent yield stress would be 0.71 Pa.

(31) Experiments 2-1 and 2-2 showed that the effect of 4.0 g of Prejel or Eliane modified starches had a similar effect as xanthan gums Keltrol AP and AP-F, although in much higher amounts (4.0 g starch vs. 0.2 g Keltrol AP-F). The amount of modified starch that is required is 20 times higher than the amount of Keltrol AP-F, in order to obtain the same effect. Gas bubbles were dispersed in the liquid soup.

(32) Experiment 2-3 with native starch shows the same behaviour as experiments 2-1 and 2-2. Gas bubbles were suspended in the liquid, and the effect of the addition of 4.0 g native starch in these compositions is the same as 4.0 g of the modified starches. Therefore the native starches do not provide a benefit over the native starch.

(33) Subsequently another set of experiments was done with a soup mix which did not contain native starch (mix 2 in Table 1). The dry soup mix of mix 2 is the same as mix 1, with the difference that mix 2 does not contain the native starches. A powdered premix is made containing 7.0 g mushroom soup mix (mix 2 as in Table 1), 3.0 g gas release agent, and the required amount of hydrocolloid as indicated in Table 4, thoroughly mixed, and put into a 300 mL tall form glass beaker. 150.0 g of hot water, just after boiling, is added to the dry premix in one swift motion and the content is vigorously stirred with a spoon for 30 sec. The volume of soup mix and water was about 166 mL. Time is started at the moment the water is poured. After stirring, the beaker is put on a stand and pictures are taken at preset time intervals.

(34) TABLE-US-00005 TABLE 4 Starches used and description of results, in soup mix 2 as in Table 1 (without native starch). Aerated Foam Initial Aerated Foam bulk layer gas bulk layer liquid volume bubble liquid volume Hydro- volume on top volume volume on top Hydro- colloid after 40 after in bulk after 5 after 5 colloid amount sec 40 sec liquid min min Exp. type [g] [mL] [mL] [mL] [mL] [mL] 2-4 Prejel 4.0 170 25 4 168 19 VA70 2-5 Eliane 4.0 167 35 1 167 20 SC160 2-6 Keltrol 0.2 155 37 0 152 33 AP-F 2-7 Keltrol 0.3 195 10 35 155 41 AP-F 2-8 Keltrol 0.4 195 6 35 166 24 AP-F 2-9 Keltrol 0.5 200 0 40 188 3 AP-F

(35) In experiment 2-4, the total volume of the liquid was lower than in experiment 2-1. After 30 seconds a clear interface between a foam layer on top of the liquid soup and the continuous liquid phase (bulk liquid) was observed. This shows that using 4.0 g of Prejel without native starch in the dry soup mix was not sufficient to keep the gas bubbles in the continuous liquid phase. Similarly in experiment 2-5 a relatively thick foam layer was formed using 4.0 g of Eliane without native starch. Therefore also using 4.0 g of Prejel without native starch in the dry soup mix was not sufficient to keep the gas bubbles in the continuous liquid phase. The differences between the experiments 2-1, 2-2, and 2-3 on the one hand and 2-4 and 2-5 on the other hand are shown in FIG. 2. The pictures in this figure are made about 30 seconds after addition of hot water to the soup mix.

(36) The experiments 2-6, 2-7, 2-8, and 2-9 with Keltrol AP-F show that this type of xanthan gum does not require the presence of native starch, in order to keep the gas bubbles in the continuous liquid phase. Experiment 2-6 shows that within 40 seconds a foam layer on top of the liquid was formed. Experiment 2-7 shows a behaviour which is similar, at a longer time scale though. The gas bubbles cream to the top within 2 minutes after the preparation of the soup. In experiment 2-8 a foam layer slowly developed. In experiment 2-9 0.5 g of Keltrol AP-F was added, and in that case bubbles did not cream to the top of the instant soup, even 30 minutes after the soup preparation. FIG. 3 shows pictures of soup samples, taken about 40 seconds after the addition of water to the dry mixes.

(37) When comparing experiments 2-9 and 1-1, the following can be observed. In experiment 1-1 the dry soup mix as added to the glass beaker contains about 3 g of native starch. In that case 0.2 g Keltrol AP-F is sufficient to keep the gas bubbles dispersed in the continuous liquid phase. In case no native starch is present, then 0.5 g Keltrol AP-F is required to obtain the same effect. Therefore it appears that 0.3 g Keltrol AP-F has the same functionality of 3 g of native starch, in order to keep the gas bubbles dispersed. Therefore the amount of hydrocolloid that is required is much lower than the hydriocolloids of the prior art.

Example 3 Qualitative Determination of Yield Stress Under Dynamic Conditions

(38) In this experiment model solutions have been prepared containing the relevant hydrocolloid (either 0.2 g, 0.4 g, or 4 g), together with icing sugar (sucrose, 5.0 gram) and erythritol (2.0 g) to prevent lumping of the dry hydrocolloid. The premix was dry mixed well, and subsequently put into a tall form 300 ml glass beaker.

(39) Three different types of precision plastic spheres (The Precision Plastic Ball Company Ltd., UK) are added to the premix in the beaker. These spheres are: high density polyethylene (HDPE) spheres, diameter of 3.17 mm coloured green, density of 0.952 g.Math.cm.sup.3, high density polyethylene (HDPE) spheres, diameter of 5.69 mm coloured bright red, density of 0.952 g.Math.cm.sup.3, polystyrene (PS) sphere, diameter of 4.76 mm, coloured dark red, density 1.04 g.Math.cm.sup.3.

(40) The size and density of the spheres was chosen in such a way that they would behave like gas bubbles of approximately 0.1 mm (4.76 mm PS sphere), 0.2 mm (3.17 mm HDPE sphere), and 0.3 mm (5.69 mm HDPE sphere). The differences to bubbles are that the terminal velocity of the probe spheres will be an order of magnitude bigger in a Newtonian fluid and that the PS sphere is going to sediment instead of cream.

(41) For the experiments with xanthan gums, 150 g of water at ambient temperature was poured on top of a dry premix and was vigorously manually stirred with a metal spoon for 30 seconds. The density of the final solutions was (1.0140.001) g.Math.cm.sup.3 at 20 C. Xanthan gum's behaviour is independent of the water temperature.

(42) Equation 1 can be written for these spheres as:

(43) $\begin{array}{cc}\frac{\left({}_l-{}_{\mathrm{pp}}\right){\mathrm{gD}}_{\mathrm{pp}}}{2\sqrt{2}},& \left(2\right)\end{array}$

(44) Where .sub.pp is the density of the probe particle, and D.sub.pp is the probe particle diameter.

(45) In the following table the critical yield stress for the three probe particles used in these dynamic yield stress measurements is given together with the equivalent bubble diameter in the respective model solutions, calculated with densities at 20 C. The probed yield stress depends on the particle size, particle density and the density of the model solution. With the three probe particles we cover more or less the range of apparent yield stress that would immobilize bubbles with diameters ranging from about 100 to 400 micrometer.

(46) TABLE-US-00006 TABLE 5 Critical yield stress for the three probe particles, together with the equivalent bubble diameter in the respective model solutions. [Pa] in Equivalent Equivalent xanthan [Pa] D.sub.b [mm] D.sub.b [mm] Particle D.sub.pp .sub.pp gum in starch in xanthan in starch material [mm] [kg .Math. m.sup.3] solution solution gum solution solution HDPE 3.17 952.0 0.68 0.78 0.19 0.22 HDPE 5.69 952.0 1.22 1.40 0.35 0.40 PS 4.76 1040.0 0.43 0.28 0.12 0.08

(47) For the experiments with modified starches, 150 g of hot water (just after boiling) was poured on top of the premix and is vigorously stirred by hand with a metal spoon for 30 seconds. Here hot water is used, in order to gelatinise the starch and make it functional. The density of the final solutions was (1.0230.001) g.Math.cm.sup.3 at 20 C.

(48) After the stirring the spheres were suspended at a certain height in the liquid, and depending of the yield stress generated by the hydrocolloid, they would slowly move upward, or downward, or they would remain at its place. The higher the yield stress, the slower the spheres would move. The beaker was positioned on a stand and pictures were taken at fixed time intervals for 5 minutes. This way the movement of the spheres could be followed in time. The translation of the spheres relative to its starting position can be plotted as function of time in a graph. In case the processes are too fast to be captured on pictures, a video record was made instead.

(49) If there is no yield stress in the system, the spheres will move with a constant velocity through the liquid. If sufficient yield stress is developed by the time the picture taking will have commenced the spheres will stay motionless. If yield stress is developing during the time of the experiments, the spheres' motion is going to be declarative, i.e. they will slow down and eventually stop moving. The trajectories of the spheres in the experiments described above are measure using video imaging software ImageJ. As a result we get the translation of each type of sphere with time in the studied system.

(50) The following experiments were performed.

(51) TABLE-US-00007 TABLE 6 Description of experiments with precision spheres. Hydrocolloid amount Hydrocolloid Exp. Hydrocolloid type [g] concentration* [wt %] 3-1 Keltrol AP-F 0.2 0.13 3-2 Keltrol AP-F 0.4 0.25 3-3 Keltrol AP 0.2 0.13 3-4 Keltrol AP 0.4 0.25 3-5 Keltrol RD 0.2 0.13 3-6 Keltrol RD 0.4 0.25 3-7 Prejel VA70 4.0 2.5 3-8 Eliane SC160 4.0 2.5 *corrected for the icing sugar and erythritol

(52) The movement of the spheres in each experiment has been plotted in various graphs in FIG. 4 and FIG. 5. In some cases duplicate measurements are shown, wherein two similar spheres are followed. In general reproducibility is very good, as the trajectories of these two spheres almost coincide. In experiment 3-1 the largest sphere translates the most from its initial position, as compared to the other spheres. The smaller spheres only have a small translation. In experiment 3-2 the concentration of hydrocolloid has doubled, and the spheres nearly do not move. The maximum measured translation is about 0.25 cm. This shows that the yield stress in this system is high enough to suspend the spheres. In experiment 3-3 the yield stress did not develop rapidly enough to keep the largest sphere suspended, this sphere floated to the surface. The smaller spheres initially show a relatively rapid movement, which then decelerates because of the development of sufficient yield stress to keep the small spheres suspended. In experiment 3-4 the translation was very small, like in experiment 3-2. The yield stress that develops This shows that the yield stress in this system is high enough to suspend the spheres. In experiment 3-5 the behaviour of the spheres is different than in the previous experiments. The HDPE spheres rapidly moved to the surface of the liquid, and the PS sphere sedimented within 2 seconds. This is shown in FIG. 5, where the translation of the particle lies on a straight line with a constant slope. This is indicative of typical Newtonian fluid rheology. Keltrol RD does not have any effect on dissolution or yield stress development. In experiment 3-6 the spheres show similar behaviour as in experiment 3-5, although the time scale is different. the HDPE particles initially accelerate, and after that move with constant velocities until they surface. This is a typical behaviour of probe particles in Newtonian fluid, and this shows that the presence of Keltrol RD in the solution does not lead to the development of yield stress large enough to oppose the buoyancy force acting on the HDPE particles. The PS particles show different behaviour: they initially decelerate and then move at constant velocities. The initial deceleration might be due to the nature of the experiment. In this case the PS particle were thrown into the solution after the video recording had started, i.e. they had some initial non zero velocity when they contacted the solution. Therefore, they decelerated due to the viscous drag of the solution. After the initial period of time all three PS particles moved with the same constant velocity during the time of the measurement, showing the same Newtonian behaviour of the surrounding solution. In experiment 3-7 the HDPE spheres rapidly moved to the surface of the liquid, while the PS spheres only showed limited movement, as shown in FIG. 6 (duplicate measurement). The yield stress was sufficient to suspend the PS spheres. Also in experiment 3-8 similar behaviour of the spheres was observed. The HDPE spheres rapidly moved to the surface, while the PS spheres remained suspended during the experiment, see FIG. 6 (duplicate measurement).

(53) Therefore the amount of modified starch used to keep spheres suspended in the continuous liquid phase, is much higher concentration than the xanthan gums Keltrol AP and Keltrol AP-F. The amounts of Keltrol AP or Keltrol AP-F are 10 to 20 times lower than the amounts of modified starches to obtain the same effect.

Example 4 Preparation of Aerated Milk Tea

(54) Lipton 5 Bean Milk Tea (ex Unilever China, Shanghai, China) instant milk tea powder was used to prepare milk tea. The dry powder is individually packed in sachets, each containing in total 21.7 g of tea extract and milk powder. The amount of milk protein in a prepared milk tea in a cup is more than 0.5%, when following the instructions on pack.

(55) A sachet was taken and added to an empty cup; this was mixed with 0.3 g xanthan gum Keltrol AP-F, and 2.0 g of gas release agent Vana Cappa B01. 150 g of water just after boiling was added, and the preparation was manually stirred. This resulted in gas bubbles dispersed in the continuous aqueous phase, as is shown in FIG. 7. No foam layer on the top formed, and the total volume of the aerated milk tea decreased only very slowly during a period of 30 minutes.

Example 5 Preparation of Aerated Soy Beverage

(56) An aerated soy drink was prepared, by blending 2.0 g of gas release agent Vana Cappa B01, 0.3 g xanthan gum Keltrol AP-F, and 14.0 g of a spray dried soy milk. This aerated soy beverage was prepared as a proof of principle. Therefore the spray dried soy milk drink that was used, had the same composition and the same spray drying process was applied as described in O. Syll et al. (Dairy Sci. & Technol. (2013) 93:431-442).

(57) Soy supreme fiber reduced with 45% w/w total protein (ex SunOpta Grains and Food Group, St. Hope, Minn., USA) was used to prepare the soy milk to be spray dried. This soy powder was combined with maltodextrin (dextrose equivalent 17, ex Glucidex Roquette, France). The soy protein amount in this mixture was 30% of the total amount of solids. The soy milk was prepared by dissolving the mixture of soy powder and maltodextrin. The total solids concentration in the soy milk was 20 wt %. After spray drying, the dry soy powder-maltodextrin mixture was used to blend with the gas release agent and the xanthan gum.

(58) 150 g water at ambient temperature was added to this dry mixture, and the preparation was manually stirred. This resulted in gas bubbles dispersed in the continuous aqueous phase, as is shown in FIG. 8. No foam layer on the top formed, and the total volume of the aerated soy drink decreased only very slowly during a period of 30 minutes.