FIELD OF THE INVENTION

The present invention relates to an powdered instant whipping agent providing an enhanced whipped food texture, and to a method of preparing such an instant whipping agent. The invention further relates to the use of such an instant whipping agent for preparing a whipped food product.

BACKGROUND TO THE INVENTION

Whipping agents are widely used in the food industry and at home. Whipping agents serve to incorporate gas such as air into food products by whipping them with an edible liquid, for instance milk or water. The obtained whipped products can then be used as, for instance, instant desserts, cake fillings and decorations, ice cream fillings and the like.

Whipping agents are typically spray-dried emulsions, comprising fat, one or more emulsifiers, carbohydrates and protein. A typical composition of such a whipping agent contains about 30 - 40 wt.% (vegetable) fat, 9 - 18 wt. % emulsifier, 30 - 50 % carbohydrates, and about 3 - 8 wt.% protein.

An important function of whipping agents is the ability to be easy to whip, and produce a high enough overrun and firmness to keep the whipped product stable (i.e. shape retention) for a longer period of time. This can for instance be hours and even up to days.

Further, whipped or aerated food products made with conventional whipping agents should have an acceptable sensory profile.

The typical sensory profile of conventional whipping agents can be described as giving a sticky and long lingering mouthfeel and lacks a dairy experience in both taste and mouthfeel. This is unlike whipped cream, which is characterized by a sensory profile with a creamy taste and mouthfeel and is dissolving more quickly in the mouth.

The need therefore exists to provide an improved powdered whipping agent providing to the whipped product sensory properties that are more in line with whipped dairy cream than the conventional whipping agents, and which also has a good overrun, good firmness, and good stability over time.

SUMMARY OF THE INVENTION It has now been found that by adapting the composition of conventional powdered whipping agents with respect to the amounts of fat and emulsifier, an enhanced sensory effect could be obtained compared to traditional whipping agents. Specifically, it has been found that by raising the amount of fat, and at the same time lowering the emulsifier content as compared to conventional powdered whipping agents, a surprisingly good texture and mouthfeel, which are all important sensory aspects, could be obtained and at the same time, good whipping properties, such as good overrun and foam stability, were achieved as well.

Accordingly, in a first aspect, the invention relates to a powdered whipping agent suitable for preparing a whipped topping, the whipping agent comprising, by weight of the whipping agent: a. 57 - 70 wt. % fat or oil; b. 1 - 6.5 wt.% of an alpha-tending emulsifier, selected from the group consisting of lactic acid esters of mono and/or diglycerides of fatty acids, acetic acid esters of mono and/or diglycerides of fatty acids, propylene glycol monostearate, and mixtures thereof; c. 0.2 - 2.0 wt. % of nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms; d. 1.0 - 5 wt. % proteinaceous emulsifier e. 18 - 35 wt. % carbohydrates f. 0.1 - 4 wt. % moisture; and wherein the weight/weight ratio of the fat or oil to the alpha-tending emulsifier is 10 or higher and lower than 30 In a second aspect, the invention relates to a method for preparing a powdered whipping agents according to the invention, comprising the steps of: a. Preparing a mixture of water and the fat or oil, the alpha-tending emulsifier, the nonionic mono and/or diglyceride of fatty acids having an acyl chain of 14 - 20 carbon atoms; the proteinaceous emulsifier, the carbohydrates, and optionally other components; and wherein the weight/weight ratio of the fat or oil to the alpha-tending emulsifier is 10 or higher and lower than 30; b. Homogenizing the mixture to obtain an emulsion; c. Spray drying the emulsion to obtain a powder.

In a third aspect, the invention relates to a method for preparing a whipped food product having an overrun of 100- 300 % comprising whipping the composition of the invention in a food acceptable liquid.

In a fourth aspect, the invention further relates to the use of the powdered whipping agent according to the invention for the preparation of a whipped food product, in particular a dessert, cake or pie filling or decoration, ice cream filling, creme Chantilly, or mousse.

DETAILED DESCRIPTION Definitions.

The term “powdered whipping agent” us used herein has its conventional meaning and refers to a dry composition which is preferably spray-dried, and may be added to a edible hquid such as water or milk, after which it may be whipped to form a stable and firm foam.

The term “fat or oil” us used herein has its conventional meaning and refers to triglycerides, being fatty acid esters of glycerol, the fatty acids may be saturated or unsaturated.

The term “alpha-tending emulsifier” has its conventional meaning and refers to organic acid esters of mono- and/or diglycerides wherein most of said esters are present, or are able, to crystallize in their alpha crystalline form (see e.g. Gerard L. Hasenhuettl, Food emulsifiers and their applications, 2008 p. 305.

The term “nonionic mono- and/or diclycerides” has its conventional meaning and refers to mono- and/or di-esters of fatty acids and glycerol, which contain substantially no ionic or ionizable residues. The term “Overrun” is defined by means of the following formula: d / (b-a) x 100 - 100 = % overrun wherein: a = weight of empty cup b =weight of cup filled with whipped product d = cup volume.

Conventional powdered whipping agents normally have fat or oil content of around 30 - 40 wt.% (see f.i. Hasenhuttl 208, page 230), and an alpha-tending emulsifier content of around 9 - 18 wt.%, which makes them suitable for providing a whipped topping that shows good overrun and foam stability. However, as stated before, the sensory aspect can be qualified as a sticky, long lingering mouthfeel and lack of dairy experience in both taste and mouthfeel, giving a sensation far away from dairy cream and therefor the “creamy” sensation is absent.

EP 3 639 672 relates to a whipping agent which is free from palm-based bulk edible oil and fat and free from palm-based emulsifiers. It comprises on a dry weight bases 25-75 % of a non-palm based edible oil or fat; 5 - 30 % of a non-palm based alpha-tending emulsifier; 1 - 15 % of a proteinaceous emulsifier, and 25 - 60 % carbohydrates. It was observed that the products according to EP 3 639 672 have a sticky, long hngering mouthfeel, and its working examples teach the skilled person towards a low fat or oil to alpha-tending emulsifier weight ratio of about 2.

By using the composition according to the invention, major progress has now been made in improving the sensory attributes of whipped toppings.

In the art, it has never been contemplated or suggested to deviate from this relatively low fat or oil content and relatively high alpha-tending emulsifier content (which may be characterized by the fat or oil to alpha-tending emulsifier wt./wt. ratio) in powdered whipping agents. Rather, a solution to solve the sensory aspect was sought by using for instance different fat or oils, or fractions thereof, or adding flavor ingredients.

It now appeared that deviating from this traditional view to apply low fat and high alpha-tending emulsifier amounts, resulted unexpectedly in enhanced sensory aspects whilst still excellent whipping properties (overrun and foam stability) were achieved.

As stated hereinbefore, the whipping agent has a higher fat or oil content compared to conventional powdered whipping agents.

The whipping agent according to the invention preferably comprises more than 59 wt. %, more preferably more than 60 wt.%, even more preferably more than 60.5 wt. %, most preferably more than 61 wt. % fat or oil. Preferably, the whipping agent comprises less than 70 wt.%, more preferably less than 69 wt.%, even more preferably less than 68 wt.%, most preferably less than 67 wt. % of oil or fat.

The whipping agent preferably comprises 60 - 69 wt. %, more preferably 61 - 68 wt. % most preferably 62 - 67 wt. % fat or oil.

The fat or oil in the whipping agent may comprise one or more vegetable oils. Preferably, the fat or oil is selected from the group consisting of palm oil, palm kernel oil, coconut oil, palm kernel stearin, sunflower oil, rapeseed oil, soybean oil, corn oil, shea oil, and/or mixtures thereof.

The fat or oil may be natural, non-hardened, or fully hardened, or mixtures thereof. Hardened fat or oil is defined as fat or oil in which substantially all the double carbon-carbon bonds is the fatty acid residues of the oil or fat have been saturated by hydrogenation.

The skilled person will understand that a combination of fat and oil can also be used.

It is preferred that the oil of fat in the whipping agent comprises more than 80 wt.% of vegetable oil or fat, more preferably more than 90 wt.%, most preferably more than 95 wt. %, based on the weight of the fat or oil present. In a most preferred embodiment, all the fat or oil is vegetable fat.

The fat or oil may optionally comprise some milk fat and/or butter oil. However, it was found that a too high amount of milk fat or butter oil resulted in a too soft and less firm whipped foam. Accordingly, preferably, the fat or oil comprises less than 20 wt. % of milk fat and/or butter oil, preferably less than 15 wt.%, most preferably less than 10 wt. %, based on the weight of the fat or oil present in the whipping agent. For the sake of clarity, the amount of milk fat and butter oil together in the fat or oil present in the whipping agent, is at most 20 wt. %, preferably at most 15 wt. %, most preferably at most 10 wt. %, based on the weight of the fat or oil present in the whipping agent. Preferably, the fat or oil comprises more than 1%, more preferably more than 3 %, most preferably more than 5 % milk fat and/or butter oil, based on the weight of the fat or oil in the whipping agent.

The powdered whipping agent according to the invention preferably comprises 1.5 - 6.0 wt. % alpha-tending emulsifier selected from the group consisting of lactic acid esters of mono and/or diglycerides of fatty acids, acetic acid esters of mono and or diglycerides of fatty acids, propyleneglycol monostearate, and mixtures thereof, more preferably 1.7 - 4.7 wt. %, even more preferably 1.9 - 4.5 wt. %, and most preferably between 2.0 and 4.0 wt. % based on the weight of the whipping agent.

Preferably, the alpha-tending emulsifier in the whipping agent is a lactic acid ester of mono and or diglycerides of fatty acids or an acetic acid ester of mono and or diglycerides of fatty acids, or mixtures thereof. Most preferably, the alpha tending emulsifier is a lactic acid ester of mono and/or diglycerides of fatty acids.

The fatty acids in the alpha-tending emulsifiers are preferably saturated, more preferably fully saturated.

It appeared that the presence of a second emulsifier next to the alpha-tending emulsifier further improved the whipping ability in overrun, foam stability and mouthfeel of whipped food products using the whipping agent according to the invention.

Accordingly, the whipping agent according to the invention preferably comprises 0.3 - 1.9 wt. %, more preferably 0.4 - 1. 7 wt. %, most preferably 0.5 - 1.5 wt. % of nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms, based on the weight of the powdered whipping agent.

Preferably, the nonionic emulsifier is a monoglyceride of fatty acids having an acyl chain of 14 - 20 carbon atoms. In one embodiment, the nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms comprise an unsaturated acyl chain. Preferably, the unsaturated acyl chain is a C 16 or C18 acyl chain, more preferably a C 18 chain, even more preferably an oleic acid chain. An oleic acid chain is a C 18: 1 chain. Most preferably, the nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms comprising an unsaturated acyl chain is a glycerol mono-oleate (GMO).

In another embodiment, the nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms comprise a saturated acyl chain. Preferably, the nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms comprising a saturated acyl chain comprise a C18 chain.

More preferably, the nonionic mono- and or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms comprising a saturated acyl chain comprising a C18 chain is glycerol monostearate (GMS).

It has now been found that the wt./wt. ratio of fat or oil to alpha-tending emulsifier in the whipping agent according to the invention is important. Where traditional whipping agent have for instance around 35 wt. % fat or oil and e.g. 16 wt. % alpha tending emulsifier, forming a wt./wt. ratio of fat or oil to alpha tending emulsifier of about 2, the whipping agent of the invention has a much higher wt./wt. ratio. Accordingly, the weight/weight ratio oil or fat to alpha-tending emulsifier in the powdered whipping agent is 10 or higher. The weight/weight ratio oil or fat to alpha-tending emulsifier in the powdered whipping is lower than 30, preferably lower than 25.

The weight/weight ratio oil or fat to alpha-tending emulsifier in the powdered whipping lies between 10 and 30, preferably between 13 and 25, most preferably between 15 and 22.

The weight/weight ratio of fat or oil to nonionic mono- and or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms may preferably be higher than 29, more preferably more than 38, most preferably more than 57. Preferably, the weight/weight ratio of fat or oil to nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms is lower than 240, more preferably lower than 140, most preferably lower than 100.

In an embodiment, in the powdered whipping agent, the weight/weight ratio of alpha-tending emulsifier, selected from the group consisting of lactic acid esters of mono and/or diglycerides of fatty acids, acetic acid esters of mono and/or diglycerides of fatty acids and propyleneglycol monostearate to nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms, is preferably between 1.5 and 10, more preferably between 1.7 and 6.0, most preferably between 2 and 4.

The powdered whipping agent according to the invention preferably comprises 0.1

- 1.5, more preferably 0.2 - 1.2, most preferably 0.4 - 1.1 wt.% of additional components selected from the group of stabihzers, hydrocolloids, flavoring agents and coloring agents, and combinations thereof.

Suitable stabilizers may be selected from group of phosphates, diphosphates, polyphosphates, hexametaphosphates, their sodium or potassium salts, and mixtures thereof. Preferably, stabilizers may be present in an amount of 0.7 - 1.5 wt. %. based on the weight of the whipping agent. Preferred is dipotassium phosphate.

Hydrocolloids may be selected by the skilled person as required, and can suitably comprise e.g. locust bean gum, gelatin, gellan gum, xanthan gum, guar gum, gum Arabic, starch, modified starch, gum karaya, gum tragacanth, cellulose gum, carrageenan, agar, alginate, pectin, carboxymethylcellulose, hydroxymethylcellulose, and mixtures thereof. For the purpose of the invention, gelatin is not comprised in the definition of proteinaceous emulsifier.

The powdered whipping agent preferably comprises 1.5 - 4.5, more preferably 2.0

- 4.0 wt. % proteinaceous emulsifier.

In the powdered whipping agent according to the invention, at least 50 wt. % of the proteinaceous emulsifier is a milk protein or a plant protein or combinations thereof.

Suitable milk proteins preferably comprise milk powder, skim milk powder, casein, casein salts such as sodium, potassium, and or calcium caseinate, whey protein, whey protein concentrate, and mixtures thereof. More preferably, the milk protein comprises caseinate, most preferably sodium caseinate.

Suitable plant proteins preferably comprise soy protein, pea protein, faba protein, oat protein, wheat protein, and potato protein, their hydrolysates, and mixtures thereof.

Preferably more than 50 wt. %, more preferably more than 75 wt. %, even more preferably more than 90 wt.% of the of the proteinaceous emulsifier in the whipping agent is a milk protein. Most preferably, all the proteinaceous emulsifier is milk protein.

The whipping agent according to the invention preferably comprises 19 - 33, more preferably 20 - 31 wt. % carbohydrates.

Carbohydrates used in the whipping agent according to the invention preferably comprise hydrolyzed starches, glucose syrup, maltodextrins, lactose and mixtures thereof.

In a preferred embodiment, the carbohydrates comprise glycose syrup.

In another preferred embodiment, the carbohydrates comprise a mixture of glucose syrup and lactose.

Preferably, the carbohydrates in the whipping agent according to the invention comprise lactose. Addition of lactose appeared to support the sensory profile in terms of dairy taste of the whipped product.

More preferably, 30 - 70 wt. % of the carbohydrates comprise lactose, most preferably 31 - 60 wt.%

In a preferred embodiment, the carbohydrates in the whipping agent comprise a mixture of glucose syrup and lactose, wherein 30 - 70 wt. %, more preferably 31 — 60 wt. % of the total weight of glucose syrup and lactose, is lactose.

In a second aspect, the invention relates to a method for preparing a powdered whipping agent according to the invention, comprising the steps of: a. Preparing a mixture of water and the fat or oil, the alpha-tending emulsifier, the nonionic mono and/or diglyceride of fatty acids having an acyl chain of 14 - 20 carbon atoms; the proteinaceous emulsifier, the carbohydrates, and optionally other components; and wherein the weight/weight ratio of the fat or oil to the alpha-tending emulsifier is 10 or higher and lower than 30; b. Homogenizing the mixture to obtain an emulsion; c. Spray drying the emulsion to obtain a powder.

Thus, preferably, in step (a) the amount of the components fat or oil, the alpha tending emulsifier, the nonionic mono and/or diglyceride of fatty acids having an acyl chain of 14 - 20 carbon atoms; the proteinaceous emulsifier, and the carbohydrates are chosen such that one arrives at the composition according to claim 1.

Preferably, the mixture in step a. comprises 57 - 73 wt. % fat or oil; 1 - 6.7 wt.% an alpha-tending emulsifier, selected from the group consisting of lactic acid esters of mono and/or diglycerides of fatty acids, acetic acid esters of mono and/or diglycerides of fatty acids, propyleneglycol monostearate, and mixtures thereof; 0.2

- 2.1 wt. % of nonionic mono- and/or diglycerides of fatty acids having an acyl chain of 14 - 20 carbon atoms; 1.0 - 5.2 wt. % proteinaceous emulsifier; and 18 - 36.5 wt % carbohydrates, based on dry weight of the components in the mixture. Preferably, step a. is carried out a temperature of between 65° C - 75° C.

The emulsifying step is preferably a high pressure homogenization, more preferably a two-step high pressure homogenization. The homogenization pressure preferably comprises a range of 80 - 200 bar, more preferably 100 - 150 bar for the first homogenization step. The second homogenization step preferably comprises a range of 20 - 60 bar, more preferably 25 - 55 bar.

The spray-drying step is carried out according to industrial practice known in the art.

It is preferred that in step a, the carbohydrates comprise a source of lactose. It is further preferred that the source of lactose comprises 60 - 100 wt. %, more preferably 65 - 95 wt.%, most preferably 70 - 90 wt.% of lactose by weight of the source of lactose. In a preferred embodiment, the source of lactose comprises or is whey permeate, more preferably sweet whey permeate. It is preferred that the whey permeate is in powder-form.

Whey permeate, as the skilled person is aware, is a by-stream from ultrafiltration of whey, a process in which whey proteins are separated from whey, resulting in a whey protein-enriched stream, and a lactose-rich stream. As stated hereinbefore, addition of lactose supports the sensory profile in terms of dairy taste of the whipped product.

In a third aspect, the invention relates to a method for preparing a whipped food product having an overrun of 100- 300 % comprising whipping the composition according to the invention in a food acceptable liquid.

The food acceptable hquid may comprise milk or water, or mixtures thereof.

The food acceptable liquid may further comprise, sweeteners, such a sucrose and/or dextrose, thickeners such as locust bean gum, gelatin, gellan gum, xanthan gum, guar gum, gum Arabic, starch, modified starch, gum karaya, gum tragacanth, cellulose gum, carrageenan, agar, alginate, pectin, carboxymethylcellulose, hydroxymethylcellulose and mixtures thereof, flavors, coloring agents etc. Preferably, the whipping is done at a temperature between 0 - 15 °C.

The whipped food product can suitably be, or used in, for instance, a cake or pie filhng, dessert, mousse, creme Chantilly, or ice-cream.

The whipped food product preferably comprises at least 8 wt. %, more preferably at least 15 wt. %, most preferably 20 - 35 wt. % of the whipping agent according to the invention, based on the weight of the whipped food product.

In a fourth aspect, the invention relates to the use of the powdered whipping agent according to the invention for the preparation of a whipped food product, in particular a dessert, cake or pie filhng or decoration, ice cream filling, creme Chantilly, or mousse.

The invention will now be illustrated by the following non-limiting examples. DESCRIPTION OF FIGURES

Figure 1: Results of the PCA on the sensory profiles obtained by the CAR method. In this plot, the dots represent the products whereas the words represent direction of intensity (starting from the center of the graphic). Legend:

Sticky_mf = sticky mouthfeel Firm_mf = firm mouthfeel Soft/Silk_mf = Soft/silk mouthfeel Creamy_mf = creamy mouthfeel Cream_t = creamy taste Milk_t = milky taste Cold_mf = Cold mouthfeel

Dissolves quick_mf = dissolves quick in mouth. Examples Example 1

Preparation of a whipping agent according to the invention.

First, an aqueous mixture at 70 °C was prepared by mixing 41.2 kg of water, 22.3 kg of glucose syrup DE28 (Roquette), 2.6 kg of sodium caseinate (EM 7, FrieslandCampina Ingredients), and 0.8 kg of dipotassium phosphate.

Then, 18.6 kg of the fractionated palm kernel oil (PKO, Cargill), 27.3 kg of fully refined coconut oil (CNO, (Cargill), 4.0 kg of Anhydrous Milk Fat (AMF, FrieslandCampina Butter), 2.4 kg of alpha tending emulsifier (Lactem P22 SG, DuPont), and 0.8 kg of unsaturated, non-ionic mono/diglyceride (GMO [glycerol mono-oleate], Myverol 18-35 PL SG, Kerry) were mixed and added to the aqueous phase and stirred. Subsequently, the aqueous/oily phase was homogenized in a two-step homogenization using a pressure of 100/50 Bar to form an emulsion and was pasteurized at a temperature of 78 °C. After homogenization, the emulsion was spray-dried at approximately (T(in) 150°C and T(out) 85°C. The whipping agent had the following wt. % composition (table 1):

Table 1: Composition of whipping agent nr. 1.

Wt./wt. ratio [fat] : [ alpha-tending emulsifier] 21.0:1.

Examples 2 -7 and 9: Preparation of further whipping agents according to the invention. Example 8 is a comparative example.

Additional whipping agents according to the invention were prepared, along with several comparative examples. The samples were made in a similar way as in example 1, except for Monte DP 570, a commercial whipping agent, which is readily available from FrieslandCampina Professional. This sample DP 570 is a reference sample against which the invention is tested. The product from EP 3 639 672 Al was evaluated as well. The other reference product against which was tested, was whipped dairy cream. This was obtained by whipping DEBIC 32% fat cream (FrieslandCampina Professional) together with 10 wt. % sugar: 460 grams of liquid cream was whipped for 7 minutes with 40 grams icing sucrose at 6 °C in a Hobart mixer, set at level 2 speed, followed by level 3 speed (about 30 seconds) until the cream had reached the desired structure, and an overrun of about 80 % was obtained.

Overrun and firmness were determined according to the same method as for the whipping agents. Table 2: Composition of samples 2 - 9 (in wt. %).

In examples 3 -9, the fat had the same wt./wt./wt. ratio as in example 1.

The coconut type fat used in example 3 was hydrogenated coconut fat (Cargill). In example 2, the wt./wt. ratio of the fats used was PKO/CNO = 1: 1.5

Table 2.

(*) total emulsifier = alpha-tending emulsifier + mono/diglyceride taken together. (**) sweet whey permeate contains about 85% lactose, based on dry solids (Modern approaches to lactose production, Durham R.J. , in Dairy-Derived Ingredients, pp 103-144, 2009).

Therefore, in example 2, the total amount of carbohydrate (lactose) is 26 % by wt., based on the weight of the whipping agent.

Table 3A: Composition of reference sample Monte DP 570 (wt. %):

Table 3B: prior art composition from EP 3 639 672 Al. In table 3 of EP 3 639 672 , comparative experiment 2 has the following composition:

(*) The emulsifier is a lactic acid ester of mono/diglycerides of fatty acids based on palm oil (LACTEM). All working examples in EP 3 639 672 show a fat or oil to alpha-tending emulsifier weight ratio of lower than 10.

Functionality testing. The above mentioned whipping agents were used for the preparation of whipped products, and were tested for whippability (overrun) and foam firmness as follows: In a mixer (KitchenAid), 150 g of the above-mentioned whipping agents were blended with 40 g of icing sugar and 350 ml water of 5-7°C. Mixing continued for three minutes at the maximum speed (level 10). After whipping, the overrun and firmness were determined.

Overrun is determined as follows:

The foam obtained was brought into an overrun cup, with a pre-defined volume and weight. The % overrun was calculated by means of the following formula: d / (b-a) x 100 - 100 =% overrun wherein: a = weight of empty cup b =weight of cup filled with whipped product d = cup volume.

For determining the foam firmness (force required to penetrate the foam), use was made of a texture analyzer TA-CT2Ϊ of Stable Micro Systems using a 16.7 gram weight with the following settings: speed 1 mm/s, time 15 seconds, measured value: force (in grams) after 10 seconds penetration. The measurement was done in a cup filled with foam and measured 5 minutes after the product had been whipped. A second firmness measurement was done after 2 hrs. storage in the refrigerator. Results of the measurements:

Table 4. Overrun and firmness of the whipped products. From the overrun and stabihty test it can be concluded that a good overrun and stability is obtained for the samples according to the present invention. The skilled person will judge an overrun of more than 100% in combination with a firmness of > 100 g directly after whipping, and a firmness of > 150g after 2 hrs. at 4°C to be very satisfactory. Comparative example DP570 has a good overrun and firmness, however, as stated before, it has a sticky mouthfeel. Comparative example from table 3B has no good firmness, and moreover has a sticky mouthfeel.

Comparative example 8 has improved taste and good overrun, however, the foam firmness is unacceptably low. Sensory test.

Although whipping agents can be characterized by their physical and/or chemical parameters, the sensory aspect is even more important. Sensory attributes of food products however can only be assessed by human beings since machine testing is, even if it would be available, highly inadequate to distinguish between subtle nuances in food products.

Trained sensory panelist are able to distinguish between sensory aspects, and such a panel is used here to evaluate the tested whipped products.

14 Trained panehsts were enrolled in the sensory test.

The sensory test was carried out using the “CAR” - methodology (Comparison Against a Reference), which is a sensory methodology in which assessors score samples in comparison to a reference sample. It is carried out as follows:

After being familiarized with the reference itself, each assessor receives each sample in a monadic sequence together with the reference samples, and are asked to score the sample on a pre-defined list of attributes using an uneven category scale (usually a 7-point scale, as depicted below) by stating whether the sample is less intense (lower part of the scale), equal (middle of the scale), or more intense (higher part of the scale) than the reference.

Way less Less than A little less Equal to A little more More than Way more than the ref the ref than the ref the ref than the ref the ref than the ref o o o o o o o

Description of the 7-point scale used in the CAR methodology.

On this scale, the assessors are trained to use consistently the scale as following: Equal to the reference means no difference between reference and the test sample on that specific attribute;

A httle more/less than the reference: assessors have noted a hint of a difference between the reference and the test sample, but this difference is still hard to detect on that specific attribute;

More/Less than the reference: assessors have consistently noted a difference on that specific attribute, also after re-tasting/smelling. - Way more/Less than the reference: very clear difference, easy to detect on that specific attribute

In practice, a hidden/blind reference is added to the test to check for the quality of the assessors’ responses. Additionally, one of the sample could also be duplicated to ensure that it receives very similar scores.

The application of the CAR methodology is two-fold: Mainly, when the interest is in comparing samples to the reference (e.g. recipe change, shelf-life studies, etc.), it provides a direct quantitative measurement of the samples against the reference. However, such methodology can also be used to compare the samples between each other. In such case, the reference sample is used as an anchor to ensure that the scale is used consistently, especially if the test is done over multiple sessions/days. In the literature, methodologies that relate to CAR include the difference from control test and the A/Not A with level of sureness (Meilgaard, M. C., Civille, G. V., and Carr, B. T. (2007). Sensory Evaluation Techniques (4 ^th ed.), CRC Press).

Thus, a CAR test was performed and results were discussed, and 8 attributes were selected as key attributes.

For each product, the average score provided across all assessors during the CAR test was computed for each of the 8 key attributes. Such table (rows x attributes) - also known as “product profile” - was submitted to Principal Component Analysis (PCA) in order to visualize graphically the relationship between samples and attributes.

The central ideal of PCA is to reduce the dimensionality of a data set (here the product profiles) consisting of a large number of interrelated variables (here sensory attributes), while retaining as much as possible of the variation present in the data set. This is achieved by transforming to a new set of variables (called the Principal Components or PCs) which are uncorrelated, and which are ordered so that the first few retain most of the variation present in all of the original variables. See Jolliffe, I. T. (2002). Principal Component Analysis, Springer (ISBN 978-0-387- 22440-4)

The results of the PCA performed on the table of sensory profiles is shown in Figure 1. In this Figure, the dots represent the products, and the words represent the sensory attributes in such a way that:

- Two products are close together if they have similar sensory profiles, and are distant if their profiles differ.

- The cosinus of the angle (starting from the centre of the graphic) between two attributes reflects the correlation between the two attributes: o two attributes situated next to each other (e.g. Cream_t and Milk_t) have a correlation close to 1, meaning that products with strong cream taste also have a strong milk taste; o two attributes situated in opposite direction (e.g. Dissolves quick_mf and Firm_mf) have a correlation close to - 1 meaning that products which dissolve quicker (mf) are less firm (mf); o two attributes that are orthogonal (e.g. Sticky _mf and Soft/Silk_mf) have a correlation close to 0 meaning that the stickiness (mf) does not affect the perception of soft/silk (mf).

Figure 1 shows that the first PC (horizontal hne) represents by itself approx. 53% of the total information present in the sensory profile table and opposes product DP570 (described as intense in Sticky _mf and Firm_mf) to product Dairy Cream (described as (Cream_t, Milk_t, Cold_mf, and Dissolves quick_mf). The second PC represents approx. 27% of the total information, and is particularly due to sample 6 and 7 which are described as low in Cream y_mf and Soft/Silk_mf compared to the rest.

The results show clearly the effect of the invention: the compositions of the invention have their selected attributes closer to those of whipped dairy cream than reference topping DP 570.

Summarizing the physical and sensory test results, it can be concluded that only the whipping agents according to the invention fulfill the requirements as stated in the background. Such products have a sensory profile much closer to whipped cream, have a good overrun and good foam firmness directly after whipping and after cold storage for several hours. Conventional whipping agents like DP 570 or products described in the prior art always have a sticky mouthfeel and sometimes a too low firmness. The comparative composition example 8 has an improved mouthfeel, but has an unacceptable foam firmness.