TECHNICAL BACKGROUND

The present disclosure generally relates to the manufacture of cereal flakes with specific texture properties based on an adjustment of the physical properties of the cereal flakes during the manufacturing process.

Cereal flakes are a popular breakfast food and have been manufactured for several years. Cereal flakes norma lly include cereal grains such as corn, rice, wheat and oats and can also include sugars, salts, oils and other flavorings, coloring and preservation agents, vitamins and mineral fortifiers.

Cereal flakes can be made by an extrusion process which combines mixing, cooking, kneading, shearing and forming. Typically, raw materials are fed into an extruder barrel and a screw conveys the raw materials along the barrel. The raw materials undergo heating, compression and mixing to form a thermoplastic mass. The raw materials are then extruded out of the extruder barrel into strands or ropes, which are cut into pellets. The pellets are then dried, flaked with a flaking roller, and toasted. The toasting process sometimes causes a slight expansion of the flakes or a curling of the flakes.

Alternatively, cereal flakes are made by a roller-drying process which combines preparing a slurry, cooking, drying, and milling. Typically, raw materials are mixed into a container, fed onto the surface of a cooking and drying roller, the cooked sheet is detached from the roller, dried and milled to the desired particle size.

US 4044159 discloses a ready-to-eat extruded cereal product. The only example from this document proposes extruded flakes of about 0.015 inches thickness (about 0.038 mm) which a re then conveyed into a mill. This two-step technology is known for the manufacture of milled cereal flakes. Neither the structural properties of the flakes nor their reconstitution properties are disclosed.

One problem with current cereal flakes is that manufacturers must provide additional processing steps after extrusion in order to adjust the physica l properties of extruded cereal flakes in order to obtain specific texture profiles after reconstitution. For example, texture profiles could range from flakes that immediately become soggy when wetted to fla kes that remain crispy when wetted. SUMMARY OF THE INVENTION

The present disclosure provides improved cereal flakes and a method of manufacturing the same. In an embodiment, the disclosure allows for the correlation of the structure of the flakes (e.g., mean cell size, mean wall thickness and porosity) with physical (e.g., absorbed water or milk) and sensory (e.g., texture) properties upon reconstitution.

For example, the disclosure provides a food-grade flake having a porosity of between 40% and 70%, a mean wall thickness of between 100 and 180 microns (μιη), a thickness of between 0.5 and 2 mm, a length of between 2 and 8 mm, a liquid absorption index of between 80 and 200 g liquid/100 g of flakes.

In another embodiment, the flake is based on carbohydrate material.

In another embodiment, the flake is obtainable by a cooking-extrusion-direct expansion process. In another embodiment, the flake is obtainable by a process that does not comprise a flattening step. In another embodiment, the flake is flat or curved.

In another embodiment, one minute after rehydration without agitation, the flake retains its overall shape. In another embodiment, three minutes after rehydration, the flake provides a textured piece. In another embodiment, the flake is reconstitutable into water or milk, at 65°C, in less than 5 minutes. In another embodiment, the flake has an overall diameter of between 2 mm and 8 mm.

The disclosure also provides a dry food composition comprising flakes as described above, wherein the food product is selected from an infant cereal product, an adult cereal product, a dehydrated sauce, a dehydrated soup, a dehydrated dish, a reconstitutable beverage, or a dehydrated pet food.

The disclosure also provides a process for manufacturing food-grade flakes which comprises: mixing a carbohydrate material in a single-screw or twin-screw extruder to form a dough; cooking the dough in the extruder at a temperature comprised between 100°C and 180°C and at a water content comprised between 5% and 30%; extruding the dough through an extrusion die; cutting the dough into flakes directly at the outlet of the die, at a cutting frequency ranging from 200 Hz to 3000 Hz ; and drying the flakes to a water content between 2% and 8%.

In another embodiment, the process does not contain a flattening step.

In another embodiment, the process further includes mixing the carbohydrate material with one or more of a non-cereal starchy component, a protein component, a lipid component, a fortification ingredient, and a flavor ingredient. Additional features and advantages are described herein, and will be apparent from, the following detailed description and the figures.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows X-ray tomography images of rolled flakes as described in the comparative example. Scale: A: 8 x 8 mm; B: 8 x 4 mm

Figures 2, and 3 are X-ray tomography images of direct flakes as described in example 1. Scale: A: 8 x 8 mm; B: 8 x 4 mm

Figure 4 is a graph showing the mean wall thickness (in microns μιη) and absorbed liquid (or liquid a bsorption index) (g water / lOOg dry flakes) of direct flakes having different recipes, as described in example 2.

DETAILED DESCRIPTION

Before the disclosure is described, it is to be understood that this disclosure is not limited to the particular apparatuses, methods, compositions, and experimental conditions described. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present disclosure will be limited only to the appended claims.

As used in this disclosure and the appended claims, the singular forms "a," "an" and "the" include plural referents unless the context clearly dictates otherwise. As used herein, "about" is understood to refer to numbers in a range of numerals. Moreover, all numerical ranges herein should be understood to include all integers, whole or fractions, within the range. The food compositions disclosed herein may lack any element that is not specifically disclosed herein. Thus, "comprising," as used herein, includes "consisting essentially of" a nd "consisting of."

The present disclosure relates to cereal flakes and the manufacture of cereal flakes with specific texture properties based on an adjustment of the physical properties of the cereal flakes during the manufacturing process. As used herein, "cereal flake" refers to a carbohydrate-based product having a porous structure and a delicate crisp texture due to the cutting and drying of the flake. It is understood that the apparatuses and methods of the present disclosure can also be used to produce simila r foods from a direct extrusion process, such as other carbohydrate-based edible products. Although the present disclosure is described by referring to cereal flakes, the present disclosure is not intended to be limited to cereal flakes. For example, the flakes do not have to be used for a cereal product but can be used for other food products. "Pellets" refers to extruded balls that are not flaked with flaking rolls, as discussed below. "Rolled flakes" refers to extruded pellets that are flaked with flaking rolls, as discussed below. "Direct flakes" refers to extruded products that are cut directly as they exit an extruder, as discussed below, meaning that the flakes are produced in a single operation without downstream processing such as flattening or further slicing of extruded pellets. "Reconstitution" refers to the addition of a liquid to cereal flakes. "Reconstitutable" refers to the time when the maximum water absorption is reached by a cereal flake. "Porridge" refers to a smooth, heterogeneous mixture with a grainy texture. "Porosity" refers to the ratio of the volume of empty spaces (cells) versus the full volume of a product, as explained in the Methods section. "Mean wall thickness" refers to the average thickness of the pores of the pellets, rolled flakes and/or direct flakes as measured using x- ray tomography, and as explained in the Methods section. "Liquid absorption index" refers to the amount of liquid that can be absorbed by flakes, as explained in the Methods section.

The food-grade flakes according to the invention are based on carbohydrate material. For instance, a dry or moist carbohydrate-based ingredient mix is prepared. For instance, the carbohydrate material comprises starchy material, maltodextrin, monosaccharides, disaccharides and fibers. In a preferred embodiment, the carbohydrate material includes a whole-grain cereal. For instance, the carbohydrate cereal includes maize, oat, rice, barley, rye, wheat, sorghum, buckwheat, sweet potato, cassava, bean, pea, amaranth, and mixtures thereof, preferably as flour. The ingredient mix typically includes 50- 99% cereal flour, 0-50% sugar, 0.05-1.8% salt, 0-6% oil or fat and 0 to 25% added water. The cereal flour may be maize, oat, rice, barley, rye, wheat, sorghum, buckwheat, sweet potato, cassava, bean, pea, amaranth flour or combinations thereof. The sugar may be sucrose, invert syrup, fructose syrup, glucose syrup with various dextrose equivalents, maltodextrins with various dextrose equivalents and combinations thereof. The mix may also include other potential ingredients such as milk, milk powder, fruit powders, whole grain flours, cocoa powder, malt extract, bran (flour and/or bits), flavoring and/or coloring agents, leavening agents (typically in an amount of 0-1%), flour-improvers such as enzymes (typically in an amount of 0-0.02%). The ingredient mix may further include pieces of edible material such as parts of nuts, nut paste, almonds, sugar, chocolate, , as well as seed husks which can be found in plain flour. The mix can be prepared by first mixing powdery components to obtain a dry mix. The dry mix can be fed to the cooker-extruder as such, or it can be mixed together with liquid or fluid components, prior to entering the extruder. After feeding the ingredient mix into the extruder, it can be further mixed in a first mixing section of a traditional food extruder, especially a twin screw extruder. Water (and/or steam) and/or a sugar solution and/or a fat solution can be injected in the extruder. This is preferably carried out at a low feed rate.

A direct flaking unit includes an extruder barrel, an extrusion die, a circular extrusion knife, a cutter cage and a motor. The ingredient mix is cooked in the extruder. Cooker extruders are continuous machines gathering several unit operations (conveying, mixing, melting/cooking, expanding, shaping) into one machine. The ingredient mix can be fed and cooked in a twin or single screw extruder with a specific screw configuration. Heating elements are regulated to ensure a certain temperature profile. Cooking the mixture may be carried out at 100°C to 180°C, with a water content from 5% to 30%, in subsequent sections of the extruder where the mixture is heated, compressed and sheared so that it forms a cooked thermoplastic mass. Under these conditions, the material melts due to the combination of mechanical friction between the screw and the thermal energy given through the barrel. The melt is then conveyed to the die where it is subjected to pressure.

The thermoplastic mass may be extruded by having it pushed by an extruder screw or twin screw through the openings of a die at the end of the extruder. As the die constitutes the final restriction at the outlet of the extruder, it has a chosen geometry which gives a defined shape to the product. As the thermoplastic mass is forced through the narrow die, it can be followed by a sudden release of pressure that leads to a rapid evaporation of steam. This causes the resulting product to have an expanded, puffed appearance.

As the thermoplastic mass exits the extruder barrel through the extrusion die, the mass is cut into flakes by an extrusion knife. In a preferred embodiment, the extrusion knife includes a plurality of blades which are spaced at equal distances around the perimeter of the extrusion knife. In use, the extrusion knife is positioned to ensure minimum gap between the knife and the contact surface of the extrusion die at the end of the extruder barrel. As the thermoplastic mass is pushed through the extrusion die, the extrusion knife is rotated by a motor in a plane of rotation parallel to the contact surface of the extrusion die and the blades on the extrusion knife cut a plurality of direct flakes off of the extruded mass protruding through the extrusion die.

The thermoplastic mass, or dough, is cut at a cutting frequency that depends on the rotation speed of the extrusion knife and on the number of blades on the extrusion knife. In embodiments of the invention, a cutting frequency of 200 Hz to 3000 Hz can be achieved. The inventors have observed that the cutting frequency has a direct impact on the structural properties and reconstitution properties of the extruded product. The specific structure of the flakes of the present invention is obtained in part thanks to a cutting frequency of 200 Hz to 3000 Hz, preferably of 250 Hz to 1000 Hz, most preferably of 300 Hz to 1000 Hz or from 500 Hz to 1000 Hz. n another embodiment, the cutting frequency ranges from 200 Hz to 800 Hz, preferably of 250 Hz to 800 Hz, most preferably of 300 Hz to 800 Hz or from 500 Hz to 800 Hz. The impact of the cutting frequency was observed on many difference recipes, as shown in the Examples.

The knife and extrusion die are surrounded by a cutter cage that is clamped to the direct flaking unit. As the knife rotates and cuts the direct flakes off of the die, the flakes fall down the cutter cage and are collected at the bottom of the cage.

The manufacturing conditions for direct flakes made from a direct flaking process differ from the manufacturing conditions for producing rolled flakes. In a preferred embodiment, the direct flakes can be made of a carbohydrate material mixed together with one or more of a non-cereal starchy component, a protein component, a lipid component, fortification ingredients and flavour ingredients. The mixed raw materials are then cooked in an extruder at a temperature of 100-180°C with a water content of 5-30%. The mixed raw materials can then be directly cut from an extruder die at a cutting frequency range of 200- 3000 Hz and dried to a water content of 2-8%. The cutting frequency can be selected as explained above.

The extruded product is dried. The drying step can be carried out using an infrared (IR) heater or by hot-air drying. Typically, the product is placed on a mesh wire belt which goes through an IR drying machine, such that IR radiations occur from above and below the product. The drying step typically reduces the water content of the product from about 15% to a moisture content of about 1 to 6.5%. Typically, the final flakes are dried to a final residual water content of from 2% to 8%. The cutting frequency can be selected as explained above.

As set forth in the test results described below, it has been determined that desired physical (e.g., absorbed milk) and sensory (texture) properties of cereal flakes upon reconstitution can be achieved by controlling the structure of the flakes produced by direct flaking (e.g., mean wall thickness, porosity). The main structure attributes driving absorbed liquid were discovered to be mean wall thickness and porosity. Absorbed liquid is negatively correlated with mean wall thickness. For a set porosity, absorbed liquid increases when mean wall thickness decreases.

The preferred direct flakes have a porosity of 40-70%, more preferably 50-70%, a mean wall thickness of 100-180 μιτι, more preferably 100-140 μιτι, a thickness of 0.5-2 mm, more preferably 0.8-1.6 mm, a length of 2-8 mm, and a liquid absorption index of 80-200 g of liquid/lOOg of flakes, more preferably 80-150 g of liquid/lOOg of flakes. Such direct flakes are particularly suitable for preparing textured drinks and beverages.

In a preferred embodiment, the major component of the ingredient mix for the direct flakes is a carbohydrate-based material.

The following non-limiting examples present scientific data developing and supporting the concept of correlating the structure of cereal flakes (e.g., mean wall thickness, porosity) with the physical (e.g., absorbed milk) and sensory (texture) properties of the cereal flakes upon reconstitution. Specifically, these examples correlate the structure of cereal flakes produced by direct flaking with the physical properties, for instance liquid absorption index, and sensory properties of the cereal flakes upon reconstitution.

Upon reconstitution in water or milk, without agitation, most of the flakes retain their overall shapes. Because the structure of the flakes is not broken due to flattening, it is believed that water migrates into the structure of the flake, which provokes swelling of the flakes, without altering their overall shape substantially. The matrix of the flakes does not disintegrate in water. In other words, it is still possible to recognize the shape of the flakes after reconstitution. As a consequence, the flakes retaining their overall shape upon reconstitution provide a specific texture and appearance to the reconstituted product which clearly differs from either a smooth porridge or from distinct flakes such as breakfast cereal flakes.

In another embodiment, three minutes after rehydration, the flake provides a textured piece. A textured piece is a piece that has swollen due to rehydration, but does not melt or disintegrate between tongue and palate. It retains some grittiness and chewiness, which shows they still have a residual structure. In another embodiment, the flake is reconstitutable into water or milk, at 65°C, in less than 5 minutes. METHODS

Determination of the internal structure of flakes by microcomputed X-ray tomography and

3D image analysis

Images acquisition offtakes

X-ray tomography scans are performed with a 1172 Skyscan MCT (Bruker MicroCT,

Kontich, Belgium) with a X-ray beam of 40kV and ΙΟΟυμΑ. Scans are performed with the Skyscan software (version 1.5 (build 9)A (Hamamatsu lOMp camera), reconstruction with the Skyscan recon software (version 1.6.3.0) and 3D image analysis with CTAn software (version 1.10.1.3, 64-bit).

Each flake was placed in the X-ray tomography chamber. For a pixel size of 5um, the camera was set up at 2000x1048 pixels and samples were placed in the Near position. Exposure time was 295 ms. Scans were performed over 180^, the rotation step was 0.4^ and the frame averaging was 10.

The reconstruction of the dataset was performed over 900 slices in average, with the settings contrast at 0.008-0.35. Smoothing and ring artefact reduction were set up at 1 and 5, respectively.

3D analysis of the images

3D image analysis was performed on 10 μιη per pixel reduced datasets over 450 slices. The analysis was performed in two steps: (i) a first step to select the flakes to be analysed by excluding the outer area surrounding the flakes, (ii) the second step to obtain the porosity of the flakes.

(i) Selection of the particles, i.e. volume of interest:

The images in grey levels were segmented. The segmentation was performed at a grey level of 30, and then images were cleaned by removing any single spots smaller than 2 pixels, and then dilated by mathematical morphology. The selection of the volume of interest was performed through the shrink-wrap function, and then this volume was eroded by mathematical morphology to adjust to the surface of the flakes. Dilation and erosion parameters were chosen in order to obtain the best adjustment to the surface of the particles.

(ii) 3D image analysis:

The images were reloaded and segmented at a grey level of 30. The porosity was then calculated as the ratio of the volume of voids in the flakes out of the flakes volume, the flakes volume being equal to the volume of interest defined above (i). The structure separation gave the pores size distribution of the flakes. The structure thickness gave the thickness of walls distribution. Determination of the absorption of liquid of flakes when soaked in a liquid

Determination of the liquid absorption index is performed in the presence of excess water, to ensure full rehydration of the flakes and ensure that water is not a limiting factor in this measurement. 40 grams of product are weighted accurately in a 600mL beaker. 160mL of distilled water are heated up at 65°C in a microwave equipped with a temperature sensor. The hot water is then poured on to the flakes in the beaker. The flakes are maintained under water with a plunger for one minute. The complete soaking time is thus one minute, then the hydrated flakes or porridge are poured on the sieve (2mm mesh size) to drain any leftover water. The draining time is two minutes and 30 seconds; the wet flakes are then weighted. The absorbed liquid in grams per 100 grams of dry product is obtained as such: ((wet weight of flakes - dry weight of flakes) / (dry weight of flakes)) x 100. The experiment is performed in duplicates and the average standard deviation obtained is less than 3%.

EXAMPLES

Extrusion conditions

A flour composition as shown in the examples below is fed into a twin screw extruder (Clextral Evolum 88) with the following configuration: L/D = 16 (length of the processing zone/external diameter of screws), and an overall screw and barrel lengths of 1408mm. The screw configuration is shown in Table 1.

Water is added into the second barrel of the extruder to reach a water content in the dough of from 15 to 22% inside the extruder. The temperature profile is 20°C/20 ^o C/80 ^o C/100 ^o C/120 ^o C/140°C in the successive barrels, to cook the dough. The dough reaches a temperature of 140°C to 150°C at the extrusion die and a pressure of 30 bar to 120 bar, leading to an instant moisture flash-off and expansion of the extruded product. The extruded product is cut, to give expanded extruded products with a humidity of between 9 and 12%. Length (mm) Pitch (mm) Flight

110 110 C2F

352 88 C2F

198 66 C2F

396 44 C2F

88 -44 CF2C

264 44 C2F

Table 1

Comparative Example: Rolled flakes

Flakes are prepared using the recipes A, or B, with the extrusion and cutting parameters, described in Table 1 and Table 2.1.

The extruded product is cut into expanded balls. The expanded balls are transported with a pneumatic conveyor toward the flaking roller, where they are flattened under a pressure of 90 bar to 100 bar, roller gap of 0.9 mm and a roll temperature of 60°C.

Porosity, mean wall thickness and liquid absorption were measured as described in the methods section. The flattened pieces have a porosity from about 30% to 36%, and thick walls with a mean wall thickness from 85 to 95 microns. They hydrate up to 380g/100g of product, as shown in Table 2.2.

Figure 1 A is a transversal view of three flattened extruded pieces by X-ray tomography, and figure 1 B is a coronal view by X-ray tomography of one of these pieces. As can be seen from these views, the flattened pieces have cracks in their structure, allowing liquid to penetrate. They have a very different internal morphology compared with direct flakes shown for instance on Figure 2 A and 2 B, as will be seen with the examples below.

RECIPE Unit A B

Corn % wt 25 25

Soya % wt 25 25

Millet % wt 47 47

Sugar % wt 3 3

EXTRUSION

product temperature °C 140 140

screw speed rpm 320 320

moisture into extruder % 21 22

pressure Bar 33 60

CUTTING

Cutting frequency Hz 340 240

Die diameter mm 3 3

Table 2.1

Table 2.2

Example 1: Direct extruded flakes

Direct flakes may be prepared using the recipes E, or F, with the extrusion and cutting parameters, described in Table 1 and Table 3.1.

The extruded product is cut into flakes. Figures 2, and 3 are transversal and coronal views of direct extruded flakes by X-ray tomography. Porosity, mean wall thickness and liquid absorption were measured as explained in the methods section. The direct extruded flakes have a porosity from about 47% to 65%, and walls with a mean wall thickness from 105 to 180 microns. They absorb from 95 to up to 120g water per lOOg of product.

The porosity of direct flakes is greater than that of rolled flakes. The mean wall thickness of direct flakes is greater than that of rolled flakes. Overall, the liquid absorption of the direct flakes is much smaller than that of rolled flakes. However, upon reconstitution in water or milk, the direct flakes retain their overall shape and they provide some texture to a porridge. They do not disintegrate in water or milk as rolled flakes do.

A serving (e.g. from 30 g to 50 g) of flakes E, or F, can be reconstituted in warm milk or water (e.g. from 90 mL to 120 mL at, e.g., from 50°C to 65°C). After a few minutes without agitation, the direct flakes retain their overall shape and they provide texture to a porridge. Some flakes may disintegrate while most of the flakes are still visible as swollen flakes.

Table 3.1

Table 3.2

Example 2: Direct extruded flakes

Figure 4 shows the mean wall thickness (μιτι) and the liquid absorption of direct flakes made with difference carbohydrate ingredients. Recipes E and F of Example 1 are identified in this graph. Diamonds correspond to recipe where corn, wheat and rice are the main carbohydrate ingredients (recipe 1). Triangles correspond to recipes where corn is the main carbohydrate ingredients (recipe 2).

Rolled flakes have lower porosity than direct flakes and also thicker walls. They absorb up to 380g/100g of product. Due to flattening, they have cracks in their structure, allowing liquid to penetrate. They have a very different interna l morphology compared with direct flakes, as can be seen when comparing the X-ray tomography of rolled flakes (Figure 1) and of direct flakes (Figures 2, and 3) .

Direct flakes do not have cracks as there is no flattening step for their production. Their lower porosity and thicker walls in comparison to rolled flakes, provide a different texture than rolled flakes upon reconstitution in water or milk.

Table 4

Example 3: Direct Flaking Sensory Tests

The direct flakes are assessed in replicate by a panel of 12 people trained for sensory texture description. During a monadic test, they were asked to assess sensory attributes with a score from 0 to 10. The correlation between physical and structural parameters, and sensory texture attributes was determined using the Pearson correlation coefficients with the minitab software.

The table below shows the statistical positive and negative correlations of physical and structural parameters of direct flakes, with sensory attributes of the same direct flakes after reconstitution.

As can be understood from Table 5, the mean wall thickness of the direct flakes correlates negatively with liquid absorption. Liquid absorption also correlates with sensory texture attributes of the reconstituted product.

Direct flakes having a high liquid absorption have thin walls, high porosity and provide smooth and mushy products such as smooth porridges.

Direct flakes having a low liquid absorption have thicker walls, lower porosity and provide more textured products. Physical and structural Sensory texture attributes

parameters Positive correlation Negative correlation

Porosity Cream, soggy Hard points, crispy, semolina, gritty, chewy, floury,

toothpacking

Mean walls thickness Hard points, crispy, Soggy, creamy, smooth, viscosity, toothpacking, semolina, liquid absorption

gritty, chewy, floury,

textured

Liquid absorption Smooth, mushy Textured, hard points, crispy, toothpacking, chewy, floury

Table 5