COMPOSITION FOR COATING FROZEN CONFECTIONERY AND A PROCESS FOR MANUFACTURING SAME

Field of the invention

The present invention relates to a composition for coating a frozen confection, in particular to a coating composition having low sugar content. The invention also relates to a method for manufacturing the same. Background

Coated frozen confections are products which are highly appreciated by consumers. Sweetness is a major driver for consumer preference. Another important feature is the texture of the coating.

With the increasing concern for health and wellness there is an increasing need for reducing calories, sugars and fats also in frozen confections. Nutritionists recommend consumers to decrease added sugar intake and favour consumption of unrefined carbohydrates, especially in children diets.

Many frozen confections have been put on the market that claim low or no sugar by containing polyols and/or intense sweeteners. However, polyols are suspected to have laxative effects when consumed at high levels and use of intense sweeteners does not have a good image for consumers and in particular for products aimed at young children.

It is common knowledge that sugars play an essential role in the sensory properties of frozen confections. Sugars have at least a dual function in frozen confections. They provide sweetness and flavour enhancement and depress the freezing point making the frozen confection palatable, they also contribute to the texture of the coating.

It is generally known that low quantities of sugars (mono- and di- saccharides) in frozen confections will lead to a reduction in sweetness, flavour and increased hardness, with a risk of a mouth "burning" sensation. For ice cream scooping of the product becomes harder with lower levels of sugar. All sugars do not have the same sweetening power and freezing point depression factor. Usually, mono-saccharides such as glucose, galactose or fructose depress the freezing point more than di-saccharides like maltose, lactose, sucrose.

In the present context the term "sugars" in this document will be defined as a mixture of mono- and di- saccharides. For example, sucrose, glucose, fructose, maltose are sugars according to this definition. Moreover, the term "sugar" will be defined as sucrose, or common sugar.

Chocolate-like or compound coatings based on fat are commonly used for coating frozen confection. The physical properties of the coating, in particular its bite properties and setting time, are determined by the crystallization of the fat. Traditionally compound coatings for frozen confection have been manufactured with coconut oil which has a saturated fat acid (SFA) level of 91%. With high amounts of fats in the coatings the SFA levels are typically above 50%.

Also regarding fats the consumers are looking for products which are healthier but provide the same properties to the product. Solutions to this problem exist in the form of coatings blends comprising particular liquid oils which are lower in SFA and fractions of palm oil. The viscosity of these blends is key to achieving the SFA reduction because too thick a coating will result in more fat in the coating and consequently a bigger quantity of SFA. It is therefore desirable to have certain limit of added sugar, fat and saturated fat acids to frozen confection. Examples of such limits per portion are e.g.: 150 kcal , fat: 9g, sat. fat: 7g. Particular targets for kids products are 110 kcal or less per portion.

Most straightforward solution to reduce the amount of sugar is to simply reduce the amount of sugar mixed in the coating preparation. However this would also reduce the amount of total solid and affect properties of the coating. Due to the high fat content such a coating would have a very low, almost water like, viscosity causing only very thin layer of coating. Common solution to balance this reduction in total solid is to use a bulking agent, for example sugars (fructose), polyols (erythritol, sorbitol), fibers (PromitorTM), and salts. Unfortunately these bulking agents can be undesirable for several reasons: too expensive, bad perception from the consumer (clean label), undesirable digestive trouble, metallic or bitter aftertaste.

Common bulking agents like fibres would reduce the sensory properties of the coating (less "sugary" sweet or metallic aftertaste). Finally, sugar replacement by a bulking agent will influence the application properties of the coating solution. When liquid (coating bath) the melting temperature and viscosity could be modified; when solid (hard coating layer around the frozen product) the snap could be modified. There is no teaching in the prior art of what would be an optimal solution for including fibers in the coating. There is therefore a need to reduce the amount of sugar in a frozen confection coating while keeping the physical characteristics of said coating, e.g. snap, mouth feel, melting behavior, no grittiness, sweetness perception, and avoid undesirable aftertastes.

Furthermore there is a need to have confectionary coatings where the physical attributes of the coating meet the requirements of the operational parameters, e.g. dripping and setting time, pick-up weight, plastic viscosity, yield value without impact on coating breakage or cracks.

Furthermore, there is a need for a reduced amount of both sugar and SFA in a frozen confectionery coating while maintaining the properties discussed above.

Object of the invention

It is the object of present invention to provide a coating for frozen confection which is reduced in sugar, fat, SFA or a combination thereof.

Summary of the invention

In a first aspect the invention relates to a composition for coating frozen confection comprising, expressed in weight % based on the total weight of the coating,

5 - 65 wt. % particles comprising an amorphous continuous phase, the amorphous continuous phase having a glass transition temperature of at least 40°C, the particles having a closed porosity of 10 - 60%, and

25 - 85 wt. % added fat, and

0 - 55 wt. % non-fat particles.

In a second aspect the invention relates to a frozen confection comprising a coating composition according to the invention.

It has been surprisingly found that by preparing a composition for coating frozen confection particles having an amorphous continuous phase with a particular closed porosity and comprising sweetener, bulking agent and optional surfactant, the overall level of sweetener (e.g. sucrose) can be reduced without having a detrimental effect on the sweetness of the coating. At equivalent volumes, the amorphous porous particles have at least equivalent sweetness compared to conventional crystalline sucrose. Thus, the porous particles according to the composition of the present invention may provide a reduction of sugar content without the need to use artificial sweeteners (for example high-intensity sweeteners) and/or without the need to use materials such as silica or cellulose.

Without wishing to be bound by theory it is believed that porous particles comprising sweetener (for example sucrose) in the amorphous state and having porosity (particularly internal closed porosity) provide a material which dissolves more rapidly than crystalline sugar particles of a similar size. This rapid dissolution in the oral cavity when consumed leads to an enhanced sweetness perception and ensures that more of the sugar is dissolved and reaches the tongue rather than being swallowed untasted.

The porous nature of the particles serves to aerate the coating composition reducing its density. The aeration from the amorphous porous particles has surprisingly been found to be stable against heat shock damage of frozen confection.

Porous particles comprising an amorphous continuous phase according to the composition of the present invention overcome the problems normally associated with amorphous sugar based powder materials and can, contrary to known amorphous sugar based materials, be used in coating compositions. So for example, because of the hygroscopic nature and so its water content amorphous sugar is not typically used in such compositions. It undesirably absorbs water from the environment and other ingredients present leading to undesirable increases in viscosity. Furthermore the amorphous state can be unstable, and amorphous sugars, such as sucrose or dextrose, tend to rapidly crystallise in the presence of moisture and/or release moisture from crystallisation. Amorphous porous particles comprising sweetener, bulking agent and surfactant have been found to be more moisture-stable than simple amorphous sugars.

It has also been found that a coating composition which comprises approximately or fully spherical particles having an amorphous continuous phase according to the invention provides coatings which have a viscosity which is reduced compare to coating compositions with conventional crystalline sucrose as the main sweetener. This has the surprising effect that coatings with low fat content which are otherwise to viscous to process can be used for coating frozen confection while still allowing a thin coating layer be added to the frozen confection. As a result healthy frozen confection coatings can be obtained which are low in sugar, of thin thickness and at the same time low in fat. Particles with closed porosity serve to aerate the coating composition reducing its density. The aeration due to particles with closed porosity is stable against heat damage. In contrast, conventional air bubbles directly in the fat phase of a coating composition are very susceptible to the fat melting. Aeration volume is often lost if the coating composition is subjected to one or more heat cycles.

Brief Description of the Drawings

Figure 1 shows scanning electron micrograph of a sample of the skimmed milk and sucrose amorphous porous particles formed in example 1. The particle has been fractured during preparation.

Figure 2 shows the evolution of pick-up weight as a function of fat content (wt. %). Filled symbols correspond to low SFA coatings with fat low in SFA and with crystalline sugar, empty symbol correspond to low SFA coatings with amorphous porous sugar. Φ corresponds to the porosity of the coating, Δ corresponds to the pick-up weight reduction between coatings with crystalline sugar and coatings with amorphous porous sugar.

Figure 3 shows the evolution of coating volume as a function of fat content (wt. %). Filled symbols correspond to low SFA coatings with crystalline sugar, empty symbol correspond to low SFA coatings with amorphous porous sugar. Δ corresponds to the volume reduction between coatings with crystalline sugar and coating with amorphous porous sugar.

Figure 4 illustrates the reduction of fat content in coatings with amorphous porous sugar while maintain similar coating volume than with reference coating (with crystalline sugar). Filled symbols correspond to low SFA coatings with crystalline sugar, empty symbol correspond to low SFA coatings with amorphous porous sugar. Δ corresponds to the fat reduction (wt. %) between coatings with crystalline sugar and coating with amorphous porous sugar.

Figure 5 shows the impact of lecithin addition to ice-cream coating on pick-up weight of coatings with crystalline sugar as a function of fat content (wt. %). Filled symbols correspond to low SFA coatings without lecithin, empty symbol correspond to low SFA coatings with lecithin. Δ corresponds to the pick-up weight reduction.

Figure 6 shows the impact of lecithin addition to ice-cream coating on pick-weight of coatings with amorphous porous sugar as a function of fat content (wt. %). Filled symbols correspond to low SFA coatings without lecithin, empty symbol correspond to low SFA coatings with lecithin. Δ corresponds to the pick-up weight reduction.

Figure 7 shows example of (A) non heat-shock and (B) heat-shock coated frozen confections used for sensory evaluation.

Figure 8 is a plot of glass transition temperature (Tg/ °C) versus sucrose content for amorphous porous particles of sucrose and skimmed milk powder at 25 °C and a water activity of 0.1.

Figure 9 is a plot of dissolution (%) (vertical axis) versus time (s) (horizontal axis) for porous amorphous powders with different compositions. Figure 10 is a plot of dissolution (%) (vertical axis) versus time (s) (horizontal axis) for amorphous powders with different levels of closed porosity. Figure 11a, l ib, 11c, l id are synchrotron radiation X-ray tomographic microscopy images for amorphous powders.

Detailed description of the invention

The coating composition according to the invention comprises 5 - 65 wt. % particles comprising an amorphous continuous phase having a glass transition temperature of at least 40°C, the particles having a closed porosity of 10 - 60%. Preferably the composition comprises 10 - 50 wt. % of the particles having an amorphous continuous phase. The amount of the particles may be selected depending on the amount of sugar reduction desired in the product.

According to the present invention the term 'amorphous' as used herein is defined as being a glassy solid, essentially free of crystalline material.

In the present context the term glass transition temperature (Tg) as used herein is to be interpreted as is commonly understood, as the temperature at which an amorphous solid becomes soft upon heating. The glass transition temperature is always lower than the melting temperature (Tm) of the crystalline state of the material. An amorphous material can therefore be conventionally characterised by a glass transition temperature, denoted Tg. A material is in the form of an amorphous solid (a glass) when it is below its glass transition temperature.

Several techniques can be used to measure the glass transition temperature and any available or appropriate technique can be used, including differential scanning calorimetry (DSC) and dynamic mechanical thermal analysis (DMTA).

The glass transition temperatures (Tg) may be measured by Differential Scanning Calorimetry (TA Instrument Q2000). A double scan procedure may be used to erase the enthalpy of relaxation and get a better view on the glass transition. Details of the scanning procedure that may be used are indicated in the Examples.

According to the present invention the amorphous continuous phase of the porous particles has a glass transition temperature of at least 40°C or higher, preferably at least 50°C or higher and more preferably at least 60°C or higher.

According to the present invention the term porous as used herein is defined as having multiple small pores, voids or interstices, for example of such a size to allow air or liquid to pass through. In the context of the present invention porous is also used to describe the aerated nature of the particles according to the present invention.

In the present invention the term porosity as used herein is defined as a measure of the empty spaces (or voids or pores) in a material and is a ratio of the volume of voids to total volume of the mass of the material between 0 and 1 , or as a percentage between 0 and 100%.

Porosity can be measured by means known in the art. For instance, the particle porosity can be measured by the following equation:

Porosity= Vp-Vcm/Vp x 100 wherein Vp is the Volume of the particle and Vcm is the volume of the matrix or bulk material.

In the present context the term closed or internal porosity refers in general terms to the total amount of void or space that is trapped within the solid. As can be seen in figure 1, porous particles according to the present invention show an internal micro structure wherein the voids or pores are not connected to the outside surface of the said particles. In the present invention the term closed porosity is further defined as the ratio of the volume of closed voids or pores to the particle volume. It is preferred that the particles comprising an amorphous continuous phase has a closed porosity of 20 - 60%, preferably 40 - 60%>, more preferably 45 - 55%.

Increasing the porosity of the particles increases their dissolution speed in water (see Example 10). This increased dissolution speed enhances the sweetness impact of the particles. However, increasing the porosity of the particles also increases their fragility. It is advantageous that the porous particles according to the present invention exhibit closed porosity. Particles with closed porosity, especially those with many small spherical pores, are more robust than particles with open pores, as the spherical shapes with complete walls distribute any applied load evenly. When added to a coating composition according to the invention with a fat-continuous phase, closed porosity has 5 a further advantage over open porosity in that fat does not penetrate inside the particle.

This penetration for the open pores inside the particles would reduce the "free" fat available to coat all the particles in the composition and lead to an increase in viscosity.

The porous particles according to the invention may have a normalized specific surface 10 of between 0.10 and 0.18 m-1, for example between 0.12 and 0.17 m-1. The porous particles according to the invention may have a normalized or of between 0.10 and 0.18 m-1 (for example between 0.12 and 0.17 m-1) and a particle size distribution D90 of between 30 and 140 microns (for example between 40 and 90 microns).

Normalized specific surface

^ ^ interstitial surface area of pores + external surface area of material solid volume of material

In the present context non-fat particles are preferably selected from the group consisting of: sugar, fibres, cocoa powder, nuts, milk powder, emulsifier, flavour or a combination thereof. The non-fat solids provide structure, flavour and colour to the coating. These

20 particles may inherently comprise some fat in the ingredient e.g. cocoa powder may comprise 10 to 12 % fat. For the purpose of this application such ingredients are considered non-fat. However, for the calculations of the total fat and SFA the fat inside these ingredients are included. The non-fat particles may be present in the composition in an amount of 0.05 - 55 wt. %, preferably in an amount of 0.1 - 55 wt. %, more

25 preferably 1.0 to 50 wt. %.

The non-fat particles may be refined to a particle size of 20 to 60 μιη, for example by roll refining.

30 In a particular preferred embodiment of the invention the particles comprising the amorphous continuous phase are approximately spherical, for example they may have a sphericity of between 0.8 and 1. It has surprisingly been found that with the approximately or fully spherical particles comprising the amorphous continuous phase the viscosity of the coating is reduced. This has the benefit that at thinner coatings may be obtained compared to coatings of similar formulation in terms of fat content and with conventional sucrose.

In the present context sphericity may be measure by standard techniques for doing so. For examples, sphericity may be measured by the Camsizer XT. It is an opto-electronic instrument, allowing the measurement of the size and shape parameters of powders, emulsions and suspensions. The technique of digital image analysis is based on the computer processing of a large number of sample's pictures taken at a frame rate of 277 images/seconds by two different cameras, simultaneously. The sample is lightened by two pulsed LED light sources during the measurement. Particle size and particle shape (including sphericity) are analyzed with a user-friendly software which calculates the respective distribution curves in real time. The perimeter of a particle projection and the covered area were measured to obtain the sphericity. Alternatively, the particles may be non-spherical, for example they may have been refined, for example by milling or roll refining.

In the context of the present invention, the term fat refers to triglycerides. Fats are the chief component of animal adipose tissue and many plant seeds. Fats which are generally encountered in their liquid form are commonly referred to as oils. In the present invention the terms oils and fats are interchangeable.

In a preferred embodiment of the invention the composition for standard coating frozen confection comprises 30 - 70 wt. %, preferably 35 - 60 wt. % added fat.

In an alternative embodiment a low fat coating composition comprises 25 - 50 wt. %, preferably from 25 to 40 wt. % added fat.

A particular preferred embodiment of the coating composition comprises the fat amounts according to the invention with approximately or fully spherical particles having an amorphous continuous phase. It has been found that with these spherical particles the viscosity of the coatings are reduced compare to coating compositions with conventional sucrose as the main sweetener. This has the effect that coatings with low fat content which are otherwise to viscous to process can be used for ice cream coatings while still allowing a thin coating layer be added to the frozen confection. As a result healthy frozen confection coatings can be obtained which is both thin and low in fat. For examples, coating compositions comprising 25 - 50 wt. %, preferably from 25 to 40 wt. % fat may be obtained.

Advantageously, the coating composition has a plastic viscosity in the range of 0.05 to 7 Pa.s and yield stress of 0.05 to 5 Pa at 40°C. This allows thin coatings to be obtained. In the present context flow properties of the coatings are determined using a Physica MCR 501 -Anton Paar (Germany) Rheometer equipped with a CC27/S geometry (Serial Number: 20689). Detail of the measurements may be as indicated in the Examples.

Advantageously, fat having a solid fat content of the fat within the range of 20 % to 80%, preferably between 20 to 50% at -15°C within 2 minutes of crystallization may be used in the composition for the frozen confection.

It has surprisingly been found that fat with a low SFA content may be used in the coating compositions. In a preferred embodiment fat has a saturated fatty acid content of less than 80 wt.%, preferably less than 50 wt. %, more preferably between 35-45 wt. % based on a total fatty acid basis.

An example of fats for high SFA coating composition comprises coconut oil and palm olein.

Examples of a low SFA coating composition comprises a combination of hard palm fraction and oil or combination of medium soft palm mid fraction and oil, the oil preferably being high oleic oil.

The oil in the low SFA coating may be selected from the group consisting of: high oleic sunflower oil, high stearic high oleic sunflower oil, high oleic safflower oil, high oleic soybean oil, high oleic rapeseed oil such as high oleic canola oil, high oleic algal oil, high oleic palm oil, high oleic peanut oil or a combination thereof.

Other medium soft fat which may be used in the coating is selected from the group consisting of: palm oil medium soft fractions including soft stearin and mid fractions, shea, cocoa butter, cocoa butter equivalents, cocoa butter replacers, or a combination thereof.

In a preferred embodiment of the invention the composition comprises particles comprising an amorphous continuous phase comprising a sweetener, a bulking agent and optionally a surfactant.

According to the present invention the term sweetener as used herein refers to substance which provides a sweet taste. The sweetener may be a sugar, for example a mono, di or oligo-saccharide. The sweetener may be selected from the group consisting of sucrose, fructose, glucose, dextrose, galactose, allulose, maltose, high dextrose equivalent hydro lysed starch syrup, xylose, and combinations thereof. Accordingly, the sweetener comprised within the amorphous porous particles according to the invention may be selected from the group consisting of sucrose, fructose, glucose, dextrose, galactose, allulose, maltose, high dextrose equivalent hydrolysed starch syrup xylose, and any combinations thereof. The sweetener may be sucrose.

In a preferred embodiment the porous particles according to the present invention comprise sweetener (for example sucrose) in the amount of 5 to 70%, preferably 10 to 50%, even more preferably 20 to 40%.

In one preferred embodiment the porous particles according to this invention comprise at least 70%> sweetener (for example sucrose). Sucrose and skimmed milk provide an amorphous porous particle which has good stability against recrystallization without necessarily requiring the addition of reducing sugars or polymers. In an embodiment, the particles according to the invention may be free from reducing sugars (for example fructose, glucose or other saccharides with a dextrose equivalent value. The dextrose equivalent value may for example be measured by the Lane-Eynon method). In a further embodiment of the coating composition, the particles according to the invention may be free from oligo- or polysaccharides having a three or more saccharide units, for example maltodextrin or starch.

According to the present invention the term bulking agent refers to a food ingredient that increases food volume or weight without significantly impacting flavour. The bulking agent according to the present invention may be a material which increases food volume or weight without impacting the utility or functionality of a food. In an embodiment of the present invention, the bulking agents of the present invention are low or non-calorific additives which impart bulk and provide advantageously healthier alternatives to for example sucrose. The bulking agent may be a biopolymer, for example a sugar alcohol, saccharide oligomer or polysaccharide. In an embodiment, the bulking agent may be a sugar alcohol, saccharide oligomer or polysaccharide which is less sweet than crystalline sucrose on a weight basis.

In an embodiment, the porous particles of the present invention comprise a bulking agent in the amount of 5 to 70%, for example 10 to 40%, for further example 10 to 30%, for still further example 40 to 70%. In one embodiment, the porous particles of the present invention comprise 10 to 25% of the bulking agent.

According to the present invention the bulking agent may be selected from the group consisting of sugar alcohols (for example isomalt, sorbitol, maltitol, mannitol, xylitol, erythritol and hydrogenated starch hydrolysates), lactose, maltose, fructo- oligosaccharides, alpha glucans, beta glucans, starch (including modified starch), natural gums, dietary fibres (including both insoluble and soluble fibres), polydextrose, methylcellulose, maltodextrins, inulin, dextrins such as soluble wheat or corn dextrin (for example Nutriose®), soluble fibre such as Promitor® and any combination thereof. In an embodiment of the present invention the bulking agent may be selected from the group consisting of lactose, maltose, maltodextrins, soluble wheat or corn dextrin (for example Nutriose®), polydextrose, soluble fibre such as Promitor® and any combinations thereof. The surfactant comprised within the particles according to the composition of the invention aids the formation of porosity, in particular closed porosity. In an embodiment, the amorphous porous particles of the present invention comprise a surfactant in the amount of 0.5 to 15 wt.%, for example 1 to 10 wt.%, for further example 1 to 5 wt.%, for further example 1 to 3 wt.%. According to the present invention the surfactant may be selected from the group consisting of lecithin, whey proteins, milk proteins, non-dairy proteins, sodium caseinate, lysolecithin, fatty acid salts, lysozyme, sodium stearoyl lactylate, calcium stearoyl lactylate, lauroyl arginate, sucrose monooleate, sucrose monostearate, sucrose monopalmitate, sucrose monolaurate, sucrose distearate, sorbitan monooleate, sorbitan monostearate, sorbitan monopalmitate, sorbitan monolaurate, sorbitan tristearate, PGPR, PGE and any combinations thereof. In embodiments according to the present invention wherein the bulking agent is derived from milk powder such as skimmed milk powder, sodium caseinate is inherently present. In embodiments according to the present invention wherein the bulking agent is whey powder, whey protein is inherently present. The surfactant according to the composition of the invention may be a non-dairy protein. In the context of the present invention the term "non-dairy proteins" refers to proteins that are not found in bovine milk.

The particles having an amorphous continuous phase may have a particle size (D90) below 150 μιη. The particle size values in the present context may be measured by a Coulter LS230 Particle Size Analyzer (laser diffraction) or any other similar machine as known to those skilled in the art. In present invention the term particle size as used herein is defined as D90. The D90 value is a common method of describing a particle size distribution. The D90 is the diameter where 90 % of the mass of the particles in the sample have a diameter below that value. The D90 value may be measured for example by a laser light scattering particle size analyzer. For example, the particle size of particles comprised within fat based confectionery materials such as chocolate compound coatings may be measured by laser light scattering. The coating particle size distribution given in the present context have been measured using static light scattering technique (Malvern Mastersizer 2000, Malvern Instruments Ltd., and Worcestershire, UK) after dilution in MCT (Medium Chain Triglycerides). The particle size values of powders have been measured by digital image analysis such as by using a Camsizer XT (Retsch Technology GmbH, Germany). The composition for coating the frozen confection according to the invention is preferably free from emulsifier, selected from sunflower lecithin, soya lecithin polyglycerol polyricinoleate (PGPR; E476), ammonium phosphatide (YN; E442) or a combination thereof. In particular it has been found that adding lecithin to coating compositions according to the invention may have little or no effect. Consequently, coating compositions without lecithin may be formulated in accordance with the invention.

The porous particles comprised within the coating composition of the invention may be obtained by foam drying, freeze drying, tray drying, fluid bed drying and the like. Preferably the porous particles comprised within the coating composition of the invention are obtained by spray drying with pressurized gas injection.

The spray in a spray drier produces droplets that are approximately spherical and can be dried to form approximately spherical particles. However, spray driers are typically set to produce agglomerated particles, as agglomerated powders provide advantages as ingredients in terms of flowability and lower dustiness, for example an open top spray drier with secondary air recirculation will trigger particle agglomeration. The agglomerated particles may have a particle size distribution D90 of between 120 and 450 μιη. The size of spray-dried particles with or without agglomeration may be increased by increasing the aperture size of the spray-drying nozzle (assuming the spray- drier is of sufficient size to remove the moisture from the larger particles). The porous particles comprised within the coating composition of the invention may comprise un- agglomerated particles, for example at least 80 wt.% of the amorphous porous particles comprised within the composition of the invention may be un-agglomerated particles. The porous particles comprised within the coating composition of the invention may be agglomerated particles which have been refined.

In a further aspect the invention relates to the use of the coating composition according to the invention for coating frozen confection.

A preferred thickness of the coating is from 0.5 to 5 mm, more preferably from 0.5 to 1.5 mm. In an additional aspect, the invention relates to a frozen confection coating material having the same sweetness as a control coating material, the control having a sucrose content between 40 and 55 %, but wherein the sucrose content has been reduced by at least 20 % compared to the control by means of particles comprising an amorphous continuous phase, the amorphous continuous phase having a glass transition temperature of at least 40°C, and the particles having a closed porosity of 10 - 60%, and a glass transition temperature at least 40°C, and wherein the coating material contains no mono, di or tri -saccharides apart from sucrose or lactose and contains no sugar alcohols or high intensity sweeteners. Examples

By way of example and not limitation, the following examples are illustrative of embodiments of the present disclosure. Determination of Glass transition temperature

Glass transition temperatures (Tg) were measured by Differential Scanning Calorimetry (TA Instrument Q2000). A double scan procedure was used to erase the enthalpy of relaxation and get a better view on the glass transition. The scanning rate was 5 °C/min. The first scan was stopped at approximately 30 °C above Tg. The system was then cooled at 20 °C/min. The glass transition was detected during the second scan and defined as the onset of the step change of the heat capacity.

Amorphous porous particles survival rate

The porosity (φ) provided by amorphous porous sugar was calculated by using the volume density (p) of the reference coating and the one of amorphous porous coating. As an example, the porosity of recipe containing amorphous porous sugar was calculated as shown in the equation below:

* ⁽ " ^{2) = 1} - ¾ i) ⁽¹⁾

The survival rate corresponds to the ratio between the measured porosity and the expected porosity, calculated by considering the initial closed porosity of amorphous porous sugar. Volume density measurement

Volume density was calculated by measuring the weight and the dimensions of sample volumes by sliding caliper. The volume density is the average of triplicates with a standard deviation of 2%.

Cryo-SEM

Cryo-Scanning Electron Microscopy (Cryo-SEM) was used to investigate the microstructure of the amorphous porous particles of the present invention within a icecream coating matrix.

A lcm3 piece of sample was glued into a Cryo-SEM sample holder using TissueTek. It was rapidly frozen in slushy nitrogen prior to its transfer into the cryo-preparation unit Gatan Alto 2500 at -170 °C. The frozen sample was fractured using a cooled knife, making its internal structure accessible. A slight etching of superficial water was performed in the preparation unit for 15 min at -95 °C, followed by sample stabilization at -120 °C. A final metallic coating was done by an application of a 5 nm platinum layer onto the surface in order to ensure good conduction of the electron for improved visualization. For visualization a FEI Quanta 200 FEG at 8 kV in high vacuum mode was used. Determination of sphericity

Sphericity was measured by the Camsizer XT as indicated above.

Particle size

The particle size values given herein may be measured by a Coulter LS230 Particle Size Analyzer (laser diffraction) as indicated above. The coating particle size distribution given herein have been measured using static light scattering technique (Malvern Mastersizer 2000, Malvern Instruments Ltd., and Worcestershire, UK) after dilution in MCT (Medium Chain Triglycerides). The particle size values of powders are measured by digital image analysis by using a Camsizer XT (Retsch Technology GmbH, Germany).

Rheology measurements

Flow properties of the coatings have been evaluated using a Physica MCR 501 -Anton Paar (Germany) Rheometer equipped with a CC27/S geometry (Serial Number: 20689). Measurements have been performed at 40°C, applying shear rates within the range 2 to 50 s ^"1 . Viscosity data is calculated from shear stresses measured throughout the shear rate range. Yield stress value was calculated dividing value of the stress at 5 s-1 (ramp up) by 10 expressed in Pascals (Pa). Plastic Viscosity Value was calculated by multiplying of the viscosity at 40 s-1 (ramp up) by 0.74 expressed in Pascals second [Pa.s]

Dripping/Setting time/Pick-up weight measurements

The coatings were completely melted and equilibrated at dipping temperature of 40° C. Temperature of the coatings were repeatedly monitored before dipping each commercial uncoated ice-cream sticks. The surface temperature of the ice-cream sticks were between -13°C and -15°C. The time taken for the dripping of the coatings to stop was noted as drip/dry time for each coating recipes. After dripping of the excess coatings, setting time of the coatings were calculated by touching coated surface of the frozen confections wearing nitrile hand gloves. Inspection was carried until no traces of the compound coatings were observed to adhere on the gloves. These holding times were recorded as the setting time for particular coating recipes. The pick-up weight of the coatings were recorded via the decrease in weight of the total coating mass after dipping of each ice-cream stick.

Heat shock method

After dipping, coated frozen confections were stabilized at -20°C for at least for one week before subjecting to heat-shock test. During the heat-shock test, coated frozen confections were maintained for 12 hours at -20°C and consecutively another 12 hours at -8°C. Such temperature cycling were continued for a period of two weeks before final evaluation. Similar coated frozen confections maintained at -20°C for two weeks without any temperature cycling were used as reference samples for comparison.

Example 1:

Preparation of amorphous porous particles

Table 1: Ingredients Ingredients Amount (wt. %)

Particles A B

Water 50 50

Sucrose 35 30

Skimmed milk powder 15 20

All ingredients were weighed separately and then mixed with a polytron PT3000D mixer until full dissolution at room temperature with a speed rate between 6000 and 12000 rpm. The inlet solution is transferred in a vessel at controlled temperature of 55°C and is then pumped at 100-130 bar. High pressure nitrogen is injected at 0.5-2 NL/min for at least 10 mins or a least until full dissolution of the gas in the solution is achieved. After a pre-heating at 60°C, the solution is spray-dried using a one-stream closed-top spray drier according to the parameters listed in the table below. Table 2: Spray-drying parameters

Amorphous porous particles were obtained having an internal structure with closed porosity, see micrograph Figure 1. The powder A contained 2.05 wt. % moisture, had a closed porosity of 51.3 %, a D90 of 51.8 microns and a Tg of 51.1 °C. The powder B contained 2.17 wt.% moisture, had a closed porosity of 50.3 %, a D90 of 46.3 microns and a Tg of 52.1 °C. Similar amorphous porous particles were produced from a mixture containing 50 wt. % water, 33.95 wt. % sucrose, 14.55 wt. % Promitor® soluble fibre (Tate & Lyle) and 1.5 wt. % sodium caseinate. Measured sphericity values were between 0.85 and 0.89.

Example 2:

High-SFA compound coating manufacture It should be mentioned that as amorphous porous sugar is aerated and as the crystalline sugar is replaced by aerated sugar on a volume basis, recipes comprising amorphous sugar are expected to show lower volume density due to porosity. Thus, for the sake of comparison, in the examples reported herein, all the ingredient compositions for recipes comprising amorphous porous sugar are expressed in wt. % based on the reference coating weight (with crystalline sugar) for a similar volume.

In addition, as amorphous porous sugar particles contain already skimmed milk powder in their composition, the quantity of skimmed milk powder added in the coating recipe has been adjusted in order to produce comparable coatings with crystalline sugar and with amorphous porous sugar having exactly the same composition in term of skimmed milk, fat, cocoa, vanillin and lecithin contents, considering identical coating volumes.

Frozen confection coating recipe composition, reference sample containing crystalline sugar (Recipe A) and variant with amorphous porous sugar (Recipe B) are reported in Table 3. The coatings are traditional recipes with high SFA and 50% fat content.

At first, the dry ingredients were mixed with part of the fat blend, followed by refining (approximately to 40 microns particle size distribution). The amorphous porous sugar was not refined and was directly added to the pre-refined mass along with the residual fat and the lecithin. The final coating mixture was blended in a Stephan mixer at 50°C before application.

Recipe B is expected to show lower volume densities (which are expressed by the decrease of the total weight and the increase of the expected porosity values in Table 3). Hence, considering identical coating volumes, Recipe A and B have exactly the same skimmed milk, fat, cocoa, vanillin and lecithin contents.

Table 3. Composition of high SFA ice-cream compound coating recipes with crystalline or aerated structured sugar for a fixed volume (volume of the coating with crystalline taken as reference).

Coconut oil 44.1 43.7

Palm olein 5.4 5.4

Cocoa Powder 4.9 4.9

Lecithin 0.5 0.5

Amorphous porous sugar 0.0 18.8

Total weight (g) 100.0 80.2

Expected porosity (%) (calculated) 0.0 19.8

Density (g/mL) 1.13 0.91

Effective porosity (%) (measured) 0.0 19.5

Survival rate (%) 0.0 98.1

Total fat (wt. %)* 50.7 50.7

% SFA in the recipe* 41 41

Particle size (D90) (μηι) 43 39**

* based on the reference coating weight for a fixed volume

** pre-refined mass before addition of amorphous porous sugar particles

Rheology of Ice-cream coatings

The rheological behaviors of coatings, Recipe A and Recipe B respectively at 40°C are displayed in Table 4. The measurements confirmed that Recipe B containing amorphous porous sugar displayed significant reduction of plastic viscosity and yield stress value. Looking at the particle size distribution of the coating samples (Table 3), it is obvious that the larger size of amorphous porous sugar than that of regular refined sugar and their spherical shape will provide lower surface area and impart less steric hindrance. It is known that in concentrated suspensions (for e.g. coatings for frozen confections), particles are in a closer contact; therefore even modest differences in particle size will affect the relative motion appreciably during flow.

Table 4. Plastic viscosity (Pa.s) and yield stress (Pa) of the coatings

Physical characteristics of coatings

Ice-cream sticks with surface temperature -13°C to -15°C was coated with the different coating recipes (Table 3) by dipping. The coatings were maintained at a constant temperature of 35°C before dipping. The coating characteristics during the ice-cream dipping process are shown in Table 5. Significant differences were observed, where the pick-up weight and setting time were found to decrease for Recipe B containing amorphous porous sugar.

Table 5. Comparison of physical characteristics of coatings

Application of amorphous porous sugar particles in coatings with approximately 50% closed porosity reduced overall density of coatings (Table 3). Decrease in density of Recipe B coating was found to decrease pick-up weight but very interestingly not significantly the volume and thickness of the coatings (Table 5). It should be noted that in the following examples, differences in coating volume or pick-up weight below 10% are not being considered as significant. Also, no major difference in dripping and setting time was observed between the coating recipes (Table 5)

Example 3:

High-SFA compound coating manufacture containing no lecithin

Frozen confection coating recipes with higher SFA content prepared at pilot plant scale have been elaborated and compositions are reported in Table 6. The coating manufacturing steps were the same as described in Example 3. No lecithin was added to the coating recipes described in this example.

Table 6. Composition of high SFA ice-cream compound coating recipes without addition of lecithin containing crystalline or aerated structured sugar for a fixed volume (volume of the coating with crystalline sugar taken as reference).

Ingredients Recipe Al Recipe Bl Crystalline sugar 32.6 0.0

Skimmed milk powder 12.8 6.9

Coconut oil 44.3 43.9

Palm olein 5.4 5.4

Cocoa Powder 4.9 4.9

Lecithin 0.0 0.0

Amorphous porous sugar 0.0 18.9

Total weight (g) 100.0 80.1

Expected porosity (%) (measured) 0.0 19.9

Density (g/mL) 1.13 0.91

Effective porosity (%) 0.0 19.5

Survival rate (%) 0.0 97.7

Total fat (wt. %)* 50.5 50.5

% SFA in the recipe* 41 41

Particle size (D90) 43 39**

* based on the reference coating weight for a fixed volume

** pre-refined mass before addition of amorphous porous sugar particles

Rheology of Ice-cream coatings

The rheological behaviors of coatings, Recipe Al and Recipe Bl respectively at 40°C are displayed in Table 7. The measurements confirmed that Recipe Bl containing amorphous porous sugar displayed significant reduction of plastic viscosity and yield stress value even without the presence of lecithin in the coating compared to Recipe Al . It is surprising to find that the plastic viscosity and yield stress value of the Recipe B (see Table 4) is comparable to that of Recipe Bl, which contained same fat type and content i.e. 50%. This implies that the structural morphology and size of the amorphous porous sugar particles contributes considerably further to the coating rheological properties than that of addition of lecithin.

However, in comparison, Recipe Al containing crystalline sugar showed much higher plastic viscosity and yield stress value than that of Recipe A (see Table 4). The effect of lecithin in minimizing the flow properties of coatings for frozen confections are vividly evident and higher in case of crystalline sugar particles. Table 7. Plastic viscosity (Pa.s) and yield stress (Pa) of the coatings

Physical characteristics of coatings

Ice-cream sticks with surface temperature -13°C to -15°C were coated with the different coating recipes (Table 6) by dipping. The coatings were maintained at a constant temperature of 35°C before dipping. The coating characteristics during the ice-cream dipping process are shown in Table 8. Significant differences were observed, where the pick-up weight and setting time were found to decrease for Recipe Bl containing amorphous porous sugar and no lecithin. This is in agreement with the rheological behavior (plastic viscosity and yield stress values) observed for the coatings as shown in Table 7. Absence of lecithin in coating Recipe Al containing crystalline sugar displayed an increase in pick-up weight and setting time compared to Recipe A. Table 8. Comparison of physical characteristics of coatings

Example 4:

Low-SFA compound coating manufacture with varying amount of amorphous porous sugar

Frozen confection coating recipes with low SFA content prepared at pilot plant scale have been elaborated and compositions are reported in Table 9. The coating manufacturing steps were the same as described in Example 3. Recipe D and E coating variants were produced to substitute in volume a part of sucrose at different levels: 50 vol. % and 100 vol. % respectively compared to Recipe C. The fat use in the examples is a hard palm fraction commercially available from suppliers Cargill, AAK, or Wilmar.

Table 9. Composition of compound coating recipes for frozen confections

Rheology of Ice-cream coatings

The rheological behaviors of different coatings at 40°C are displayed in Table 10. The measurements confirmed that with an increasing amount of amorphous porous sugar in the coatings from Recipe C to E samples, plastic viscosity and yield stress value were reduced significantly. Looking at the particle size distribution of the coating samples (Table 9), it is obvious that the larger size of amorphous porous sugar than that of regular refined crystalline sugar and their spherical shape will provide lower surface area and impart less steric hindrance. It is known that in concentrated suspensions (for e.g. coatings for frozen confections), particles are in a closer contact; therefore even modest differences in particle size will affect the relative motion appreciably during flow.

Table 10. Plastic viscosity (Pa.s) and yield stress (Pa) of the coatings

Physical characteristics of coatings

The coating characteristics during the ice-cream dipping process are shown in Table 11. Significant differences were observed: with an increasing concentration of the amorphous porous sugar: the pick-up weight and setting time were found to decrease (up to 23%).

Application of aerated sugar particles in coatings with approximately 50% closed porosity reduced overall density of coatings (Table 3). Decrease in density of Recipe D and Recipe E coatings were found to decrease pick-up weight but not majorly the volume and thickness of the coatings (Table 11). It should be noted that difference in coating volume below 10%> is not being considered as significant.

Here, at 50% fat content in the recipes for a fixed volume the diminution of pick-up weight can be majorly attributed to aeration of the coatings due to amorphous porous sugar (-20%) and not to the, even though important, changes in term of rheological properties. Also, no major difference in dripping and setting time was observed between the coating recipes (Table 5)

Table 11. Comparison of physical characteristics of low SFA coatings (ml) weight (g) time (s) time (s)

Recipe C 13.2 15 12 35

Recipe D 12.6 13.0 14 28

Recipe E 12.4 11.5 12 26

Reduction (%)* 6 23 - -

* Calculated between Recipe C and Recipe E

Example 5:

Low-SFA compound coating manufacture containing crystalline and amorphous porous sugar with varying fat content

Frozen confection coating recipes with low SFA and varied fat content (30-60%) prepared at pilot plant scale containing only crystalline sugar and amorphous porous sugar have been elaborated and compositions are reported in Table 12 and Table 13 respectively. The coating manufacturing steps were the same as described in Example 3.

The added fat use in the examples is a medium soft palm mid fraction commercially available from suppliers Cargill, AAK, or Wilmar.

Table 12. Composition of low SFA ice-cream compound coating recipes with crystalline sugar

Table 13. Composition of low SFA ice-cream compound coating recipes with amorphous porous

Rheology of Ice-cream coatings

The rheological behaviors of different coatings at 40°C are displayed in Table 14. The measurements confirmed that with an increasing amount of fat content in the coatings from Recipe F to I samples, plastic viscosity and yield stress value were reduced significantly. The particle size (D90) of the coating samples as well as lecithin content (Table 9) were similar (as the recipes were prepared using same pre-refined mass) and had little effect on the rheological properties. Table 14. Plastic viscosity (Pa.s) and yield stress (Pa) of the coatings

Physical characteristics of coatings

Figure 2 shows the evolution of pick-up weight achieved during the ice-cream dipping process for both crystalline sugar coating recipes (Recipe F-I) and amorphous porous sugar coating recipes (Recipe J-M) as a function of fat content (30-60 %).

In general, with an increasing concentration of fat content, the pick-up weight of the compound coatings, irrespective of the type of sugar present, displays significant reduction of pick-up weight. However, it is surprising to observe the reduction of pickup weights with the presence of amorphous porous sugar for similar fat content compared to the crystalline sugar variants, especially at low fat contents, i.e. 30 and 40%. The 30 vol. % fat content recipe (Recipe F) with crystalline sugar was not processable (i.e. cannot be dipped) due to its extreme rheological properties (Table 14), hence the pick-up weight for this particular recipe is not recorded in Figure 2. Interestingly, the alternative coating recipe (Recipe J containing 30% fat content) with amorphous porous sugar particles was processable and displayed fair amount of coating pick up weight (~23g). The pick-up weight was even lower than that of Recipe G (~26g) containing crystalline sugar and more fat content than Recipe J. This demonstrates the opportunity of using amorphous porous sugar particles in coating for frozen confection to reduce the overall fat content and have even lower pick-up weight at similar volume (Figure 3). Significant differences in pick-up weight (reduction of 41%) were also observed between Recipe G and Recipe K containing 40 % of fat. However, the extent of reduction of pick-up weight tends to decrease with increasing fat content in coating recipes (Recipe L and M). The pick-up weight reduction between coatings i.e. Recipe H and Recipe L; Recipe I and Recipe M can be attributed majorly to the porosity or density difference due to amorphous porous sugar and not to the, even though important, changes in term of rheo logical properties.

Figure 3 shows the evolution of coating volume of different coating recipes as a function of fat content. Major difference in volume and thickness of the coatings were observed for lower fat content recipes containing amorphous porous sugar i.e. Recipe J (30 %) and Recipe K (40 %). It should be noted that difference in coating volume below 10% is not being considered as significant, which is in the case of Recipe L (50 %>) and Recipe M (60 %).

Also, no major difference in dripping and setting time was observed between the coating recipes for both crystalline sugar coating recipes (Recipe F-I) and amorphous porous sugar coating recipes (Recipe J-M) (data not shown). Example 6:

Modulation of fat content in coatings for frozen confections with amorphous porous sugar while maintaining similar coating volume as coatings containing crystalline sugar

Figure 4 shows potential fat reduction between coatings with crystalline sugar (Recipe F-I; see table 12) and coating with amorphous porous sugar while maintaining similar coating volume. For example, similar coating volume as Recipe G containing 40 % fat can be achieved using the amorphous porous sugar while reducing the fat content of the recipe by 18%, i.e. approximately 33 % fat content is required in the coating recipe. Similarly for Recipe H and Recipe I (see Table 12), 9% and 7% reduction in fat content can be achieved using the amorphous porous sugar while maintaining similar coating volume.

Example 7:

Low-SFA compound coating manufacture containing crystalline and amorphous porous sugar with varying amount fat content containing no lecithin

Frozen confection coating recipes with low SFA and varied fat content (30-60%) prepared at pilot plant scale containing only crystalline sugar and amorphous porous sugar have been elaborated in Table 15 and Table 16 respectively. The coating manufacturing steps were the same as described in Example 3. No lecithin was added to the coating recipes described in this example.

Table 15. Composition of low SFA ice-cream compound coating recipes without lecithin and with crystalline sugar

Table 16: Composition of low SFA ice-cream compound coating recipes without lecithin and with aerated structured sugar for a fixed volume (volume of the coatings with crystalline sugar in Table 15 taken as reference)

Cocoa Powder 6.8 5.9 4.9 3.9

Lecithin 0.0 0.0 0.0 0.0

Amorphous porous sugar 26.6 22.8 18.9 15.3

Total weight (g) 72.0 76.0 80.1 83.9

Expected porosity (%) 28.0 24.0 19.9 16.1

Density (g/mL) 0.93 0.91 0.92 0.91

Effective porosity (%) 25.0 23.5 18.6 15.7

Survival rate (%) 89.3 98.1 93.2 98.0

Total fat (wt. %)* 30.5 40.8 50.7 60.5

% SFA in the recipe* 12.3 18.2 22.7 27.1

Particle size (D90) 45.7 45.7 45.7 45.7

* based on the reference coating weight for a fixed volume

** pre-refined mass before addition of amorphous porous sugar particles

Rheology of Ice-cream coatings

The rheological behaviors of different coatings containing no lecithin at 40°C are displayed in Table 17. The measurements confirmed that with an increasing amount of fat content in the coatings from Recipe Fl to Ml samples, plastic viscosity and yield stress value were reduced significantly, irrespective of the type of sugar in the coating recipes. Removal of lecithin has a significant impact on the rheological properties of coatings containing crystalline sugar (see Table 14 and 17), whereas no major change is observed for the coating recipes containing amorphous porous sugar (see Table 14 and 17).

Table 17. Plastic viscosity (Pa.s) and yield stress (Pa) of the coatings

Recipe Kl 0.64 0.52

Recipe LI 0.19 0.14

Recipe Ml 0.10 0.07

Figure 5 and 6 show the impact of lecithin addition to ice-cream coating on pick- weight of coatings with crystalline sugar and amorphous sugar respectively as a function of fat content. The effect of addition of lecithin (0.5 wt. %) in coating recipes containing crystalline sugar (Recipe Fl-Il; see Table 15), irrespective of fat content in reducing pick-up weight was found to be significant, especially at lower fat content (40 %). As previously described in Example 5, coating recipes at 30 % fat content containing crystalline sugar with (Recipe F) or without lecithin (Recipe Fl) was not processable and hence not shown in Figure 5.

However, the effect of addition of lecithin (0.5 wt. %) in coating recipes containing amorphous porous sugar (Recipe Jl-Ml; see Table 16), irrespective of fat content in reducing pick-up weight was found to be negligible, except at very low fat content of 30 %. It should be noted that difference in coating pick-up weight below 10% is not being considered as significant. It is in agreement with the comparison of rheo logical properties of the coatings (with or without lecithin content) as shown in Table 14 and 17. This shows that no lecithin is required in formulation of coatings for frozen confections containing amorphous porous sugar described in this invention, at fat contents above 30 %.

Example 8:

Sensorial comparison of coating containing amorphous porous sugar for frozen confection under heat-shock and non-heat shock conditions Frozen confection coating recipes with high SFA prepared containing only amorphous porous sugar have been elaborated in Table 18. The coating manufacturing steps were the same as described in Example 3. No lecithin was added to the coating recipe described in this example. Several frozen confections were dipped and coated with the coating recipe. After dipping, coated frozen confections were stabilized at -20°C for at least for one week before subjecting to heat-shock test (see experimental section for method details). Coated frozen confections maintained at -20°C for two weeks without any heat-shock cycling were used as reference samples for comparison. Table 18. Composition of high SFA ice-cream compound coating recipes with aerated structured sugar for a fixed volume (volume of the coating with crystalline sugar taken as reference).

Different key attributes in regards to the organoleptic properties of the non heat-shock and heat-shock coated frozen confections were evaluated during blind testing by a test panel consisting 6 participants. Table 19 summerizes the number of panelists having noticed a difference between the two samples. In general, no major differences were observed between the non heat-shock and heat-shock coated frozen confections. Comparatively both the samples were perceived similar by the test panel. Figure 7 displays the (A) non heat-shock and (B) heat-shock coated frozen confections.

Table 19. Organoleptic evaluation of Non heat-shock and heat-shock coated frozen confections Attributes Score

Heat-shock vs non heat-shock samples

General taste/Off-flavour 0

Sweetness perception 0

Grittiness 0

Mouth coating 0

Melting rate -1

Pitch 0

Hardness (up to 7 minutes of 0

consumption)

Example 9

The effect of altering the composition of the amorphous matrix of the amorphous particles was examined for different ratios of milk powder (SMP) and sucrose. The amorphous matrix should be stable against crystallization. In the case of confectionery filling manufacture for example, the matrix should remain amorphous under the temperature and humidity conditions experienced during mixing and refining. If processing or storage conditions approach those at which the amorphous material passes through the glass transition then there is a possibility that crystallization will occur leading to a collapse of the particles, for example the lactose present in amorphous porous particles of skimmed milk powder and sucrose may crystallize.

Amorphous porous particles with different ratios of sucrose:SMP were produced; 40:60, 50:50, 60:40, 70:30 and compared to pure amorphous sucrose and SMP. The amorphous SMP was spray dried. The amorphous sucrose was obtained by freeze drying (Millrock, US). A solution containing 10% (weight basis) of sucrose was prepared. It was frozen at -40 °C for 6 hours allowing the formation of ice crystals. Primary drying is performed at 150 mTorr. Ice crystals sublimate and leave voids behind leading to a highly porous structure. Secondary drying consists of a temperature ramp from -40 °C to 40 °C at 1 °C/hour. During that stage residual water bound to the matrix is removed by desorption leading very low moisture content, typically 1-2% as measured by ThermoGravimetric Analysis.

As the samples initially have different water activity (aw) values the sorption isotherms were drawn to calculated Tg at the same aw.

1) Sorption isotherms were built by collecting samples during short periods of time (i.e. typically over 48h) stored in two types of desiccators (one for partial drying and one for humidification). The Tg of each sample was obtained by using the second scan of DSC experiment at 5 °C/min heating ramp. The first scan should stop at about 30 °C above the Tg in order to avoid relaxation enthalpy interference with Tg measurement. Onset Tg of the product is then determined using a second scan. After 2h heating at Tg+5°C aw is measured at 25°C.

2) BET fitting is performed over the data of moisture content as a function of aw (0.08-0.35) and the Gordon Taylor over the data of Tg as a function of aw

(corresponding range).

a. Brunauer-Emmett-Teller equation (BET):

M fa ^M ^ ^{C a} w

d ^{b ( w ) "} (l - a _w ) [l _{+ (} C - l) aj where C is a constant and Mm is the BET monolayer moisture content (on dry basis) a. Gordon-Taylor equation (Gordon and Taylor, 1952):

_ kWTg _Water + (l-W Tg _dry

⁹ few+fe(l-w) where w is water content on a weight basis, Tg,water is the glass-transition temperature of water estimated at -135 °C, Tg,dry is the glass-transition temperature of sucrose and k is a curvature constant.

The glass transition temperature (Tg) is plotted against sucrose content in Figure 8 for amorphous particles at a water activity of 0.1 and 25 °C. It can be seen that there is a much more pronounced decrease in glass transition temperature for increasing sucrose content at or above 40 % (a ratio of 0.66 : 1). This means that there is a significant decrease in stability (against crystallization) when the level of sucrose in an amorphous matrix containing sucrose and skimmed milk powder exceeds 40 %. Therefore, when seeking to reduce the sucrose content in a food product by replacing crystalline sucrose with amorphous porous particles of the invention containing sucrose and skimmed milk an optimum proportion to use is around 40 % sucrose and 60 % skimmed milk powder.

Example 10 The effect of altering porosity and composition on dissolution speed and sweetness impact was investigated. Amorphous porous particles were prepared as in Example 1 , with the inlet solution containing 50 wt.% water and 50 wt.% of sucrose + SMP (skimmed milk powder) at the appropriate ratio. No sodium caseinate was added as this is already present in SMP. Particle size distribution was measured using a Camsizer XT (Retsch Technology GmbH, Germany).

Samples with different levels of porosity, but with similar particle size distributions and the same composition were prepared. Sample G was prepared with no gas injection. This produced a very low level of closed porosity (6 %>). Varying the gas flow up to 2 normal litres per minute allowed increasing levels of closed porosity to be generated.

The closed porosity was obtained by measuring the matrix and apparent densities.

The matrix density was determined by DMA 4500 M (Anton Paar, Switzerland AG). The sample was introduced into a U-shaped borosilicate glass tube that is excited to vibrate at its characteristic frequency which depends on the density of the sample. The accuracy of the instrument is 0.00005 g/cm ³ for density and 0.03 °C for temperature. The apparent density of powders was measured by Accupyc 1330 Pycnometer (Micrometrics Instrument Corporation, US). The instrument determines density and volume by measuring the pressure change of helium in a calibrated volume with an accuracy to within 0.03 % of reading plus 0.03 % of nominal full-scale cell chamber volume.

Closed porosity is calculated from the matrix density and the apparent density, according to the following equation: „, , . „ /„ Papparent \

Closed porosity = 100. 1 1 )

^ Pmatrix '

The dissolution test was performed as follows. 30.0 g ±0.1 g of water (milliQ grade) was placed in a 100 mL beaker (h = 85 mm 0 = 44 mm) with a magnetic stirrer (L = 30 mm 0 = 6 mm). The stirring rate was adjusted to 350 rpm and 1.000 g ±0.002 g of powder was added in the solution. During the dissolution, the refractive index of the solution was registered each second until a plateau corresponding to complete dissolution was reached. Refractive index was measured using a FISO FTI-10 Fiber Optic Conditioner These experiments were performed at room temperature (23-25 °C).

The result of varying composition is shown in Figure 9. Powders with a lower proportion of sucrose dissolve more slowly. The result of varying the porosity is shown in Figure 10. The powders with significant porosity (A and F) dissolved much more rapidly than the un-gassed sample (G).

Example 11

The porous structure of amorphous particles was examined using synchrotron radiation X-ray tomographic microscopy (SRXTM), at the TOMCAT beamline of the Swiss Light Source (SLS), Paul Scherrer Institut, Switzerland. The acquisition followed a standard approach with the rotation axis located in the middle of the field of view. Exposure time at 15 keV was 300 ms and 1,501 projections equi-angulary distributed over 180° were acquired. Projections were post-processed and rearranged into corrected sinograms. Stacks of 2161 16 bits Tiff images (2560 X 2560 pixel) were generated with a resolution of 0.1625 μιη per pixel.

Slice data were analysed and manipulated using Avizo 9.0.0 (https://www.fei.com/software/amira-avizo/) software for computed tomography.

The routine used for the measurement was the following. For each sample, 3 stacks of 500 images were analysed. After sub volume extraction, stacks of images were thresholded using an automatic routine to specifically select the matrix material and calculate its volume. Then the surface of each sample was estimated using the surface generation module of the software and the surface values were extracted. Normalized specific surface was calculated as the ratio of the matrix volume by the total material surface (external and pores). Powders with different levels of closed porosity (A, F and G from Example 10) were imaged, together with a powder (H) as a comparative example which did not contain protein. Powder H was prepared in a similar manner to that described in Example 1 , except that the inlet solution contained 50 % water, 25 % sucrose and 25 % of a 21 DE maltodextrin (Roquette) and carbon dioxide was used instead of nitrogen. Powder H had a closed porosity of 31 % and a particle size D90 of 184 μιη. The images are shown in Figure 11a (A), Figure l ib (F), Figure 11c (G) and Figure l id (H). The calculated normalized specific surfaces (mean of three sets of 500 slices) were as follows:

As can be seen from the images, the porous structure of powders A and F comprise multiple small pores. The internal surface of these pores leads to a high normalized specific surface value. The normalized specific surface for sample F is lower than sample A, consistent with the measured lower closed porosity value. Sample G, where no gassing was applied, has a low porosity and a low normalized specific surface value. For sample H it can be seen that although it has a similar closed porosity value to sample F, the structure is very different, with large voids within the particles. Such a structure is physically weaker than multiple small pores, and if the outer walls of the particles are broken, no (or very little) porosity remains. Sample H has a correspondingly lower normalized specific surface value.

It should be understood that various changes and modifications to the presently preferred embodiments described herein will be apparent to those skilled in the art. Such changes and modifications can be made without departing from the spirit and scope of the present subject matter and without diminishing its intended advantages. It is therefore intended that such changes and modifications be covered by the appended claims.