Conventional fruit jams are known. Fruit jams (e. g. strawberry jams) are generally prepared by mixing fruit pulp with sugars, i. e., sucrose, and have a high moisture content.

Water activity or Aw reflects moisture content of a food product. For conventional fruit jam, Aw ranges from 0.7 to 0. 8. Due to this high Aw, the moisture transfer from the jam to the baked flour based products (e. g. biscuits) is very high. The driving force for this moisture transfer is the difference in water activity between the jam and the initially drier biscuit. Therefore, water migrates rapidly from the jam to the biscuit which has a lower water activity (Aw of about 0.2) until the water activity equilibrium is attained. The time to reach this water activity equilibrium depends, among other factors, of the Aw of the jam, percentage of jam in the biscuit, etc. However, equilibrium is usually observed after few days of storage. The conventional fruit jams cannot be used properly to produce a crispy biscuit since they rapidly "wet"the biscuit. As a consequence of the rapid loss of crisp texture, the filled product is not attractive for the consumer and it also has a poor shelf life having a stale taste.

Therefore, fruit jam-filled biscuits on the market are either"soft", or they have non-aqueous fillings comprising vegetable fat plus colourings and flavourings.

Usually, in order to avoid sugar crystallisation in the aqueous phase, the water activity of flavoured fillings is lowered by the addition of non-sugar ingredients such as sorbitol, glycerol or other polyhydric alcohols. However, the polyhydric alcohols-based products suffer from flavour and

texture problems. Polyhydric-alcohols cause undesirable tastes in the filling although they are needed to drastically reduce the water activity. However the polyhydric alcohol based products suffer from flavour and taste problems. In some countries, the use of polyhydric alcohol in fillings prevents the possibility to label the product as a"jam". Therefore, there is also a clearly perceivable demand for"clean labels" in the food packages.

EP 372 596 A2 discloses cookies made with low Aw fibre- containing fillings. The filling has a very low water activity (0.2-0.35) so that crispness of the cookies outer dough is preserved. However, the filling comprises of a significant amount of glycerine or propylene glycol (at least 16 wt. %) as a humectant which acts to reduce Aw. Such humectants are considered as non-natural that would be an obstacle to a "clean labelling". The humectants also impart an artificial aftertaste, in particular bitterness when propylene glycol is used. An undesirable fibrous texture is also imparted to the filling. Furthermore, at the given water activity level, the filling is not soft enough for still providing a pleasant mouthfeel and processability is likely to be difficult.

US 4 774 095 relates to a flavoured filling for dough- based products comprising an aqueous phase, a sugar dissolved in aqueous phase, a thixotropic cohesive network of cellulosic fibrils or microfibrils, edible polyols as humectant (at least 5 wt. %) and, high methoxy pectin in the form of pectin gel lumps. This invention causes the same disadvantages on the texture due to the fibre and to the taste due to the use of non natural sugars in the filling as in the case of EP 372 596 A2. Labelling may also be constraining.

EP 515 864 relates to low fat or no fat sugar based confections such as caramels which maintain micro-biological stability over shelf life. In order to replace fat, prevent sugar crystallisation and to form a gel the formulation employs a complex mixture of hydrocolloids which are both cationic and thermosensitive such as carrageenan, furcellarin and gellan.

Therefore, there remains a need for a low water activity aqueous filling for baked flour based products having the characteristics suitable to maintain crispness of baked product and softness of filling using only natural sugars to reduce Aw.

An object of the invention is to propose a low water activity flavoured filling with an excellent mouthfeel and palatability with no artificial aftertaste while exhibiting a stable soft, non-crystallised and non-fibrous texture.

Another object of the invention is to propose a low water activity flavoured filling which can experience a long shelf life, preferably of 3 to 6 months, and which can preserve crispness of flour based products during this shelf life.

Another object of the invention is to propose a low water activity flavoured filling which is capable to be easily processed in standard available depositing and sandwiching machinery.

For that, the invention is a soft and pumpable polyhydric alcohol-free flavoured filling for a flour-based product comprising of: (a) an aqueous phase, (b) a mixture of sugars dissolved in the aqueous phase wherein the sugars comprises fructose, glucose and sucrose in relative proportions which prevent crystallisation of the filling and set water activity value of the filling from 0.38 to 0.47, and (c) a flavouring and acidifying component in amount effective to lower the pH of the filling below 5.

In another aspect, the invention also relates to a baked flour based product comprising a flavoured filling wherein the filling comprises: (a) an aqueous phase, (b) a non-crystallised mixture of sugars dissolved in the aqueous phase wherein the mixture of sugars essentially consists of fructose, glucose and sucrose, (c) a flavouring and acidifying component in an amount effective to lower the pH of the filling below 5 and,

wherein the filling is set at a water activity value which prevents the baked flour based product loosing its crispness during at least 4 months.

It has been surprisingly found that a controlled combination of the three sugars with no significant addition of chemicals, as above-identified humectants, could produce a supersaturated solution of suitable water activity level in the range desired for significantly reducing moisture transfers while preventing sugars from crystallising during an extensive period of time. The absence of crystallisation has mainly two advantages: firstly, it gives a soft mouthfeel desired for an attractive product even after an extensive period of storage and secondly, it favours processability of the filling.

Reducing Aw to such a low range means that relatively large amounts of soluble substance (i. e., sugars) must be added to the aqueous phase. Normal sugars such as glucose (or glucose syrups) and sucrose cannot be used alone or in combination due to a limited solubility which causes crystallisation and also poor Aw-lowering properties. It must be noted that sugar concentration in the water phase of the filling must be well above solubility limits (i. e., supersaturated solutions) for obtaining the suitable water activity level. An important benefit of the invention is so to avoid the use of non-sugar ingredients in amounts significant to have efficient Aw-lowering properties and high solubility of the group of polyhydric alcohols; i. e., glycerol, sorbitol or other polyols.

For that, it has been determined that the flavoured filling advantageously comprises.

5 to 15 wt % of sucrose, 15 to 40 wt % of glucose, 10 to 50 wt % of fructose.

More preferably, it comprises 8 to 10 wt of sucrose, 20 to 37 wt % of glucose, 15 to 45 wt % of fructose.

The given sugar percentages are to be considered as based on sugar dry equivalent.

Various sources of natural sugars to meet given proportions can be used. Crystalline fructose can be used or other sources of fructose such as high fructose corn syrup.

Honey or equimolecular mixture of glucose and fructose obtained by enzymatic hydrolysis of sucrose (invert syrup) may also be used which have proved to be advantageous as giving highly soluble and produce supersaturated solutions easily.

It has also been found that fructose avoids or, at least delays, crystallisation when mixed to the two other sugars in the aqueous phase. Fructose alone has a very high solubility limit of about 380.8 g/lOOg of water. However, fructose would be too sweet if used alone at the Aw values required. Relative sweetening power is usually of 124 (at 20C) for D-fructose, 100 for sucrose and 69 for D-glucose (Comparison for 10 g of sugar in 100 ml of water solution). Fructose is also more expensive than the two other sugars. Therefore, a balance brought by the two other sugars is advantageously proposed to compensate the deficiencies of fructose.

For that, the use of sucrose, also know as saccharose, has proved to be essential to adjust the sweetness of the filling in the given range using sensory evaluations. Sucrose has also a fairly good solubility limit of about 208.7 g/lOOg of water (at 25°C). However, sucrose has poor Aw lowering properties and could not be used alone. Sucrose has also the advantage to be a less expensive sugar compared to fructose.

A higher amount of glucose than proposed would easily form crystals during storage because of its low solubility. On the other hand, glucose has proved to be essential to compensate and maintain a more stable balance of sweetness due to the high and varying sweetening power of fructose at different temperatures. Glucose is also cheaper than fructose. Glucose under powder form is preferred to glucose syrup as some glucose syrups (of low D. E.) may contain too high proportions of saccharides with high molecular weight, which do not properly contribute to lowering the Aw.

The combination of the three sugars as proposed prevents crystallisation of the filling, at least during the intended shelf life. Crystallisation causes hardening of the sugar and is undesirable, as the plastic filling would turn into a hard product which would not make it suitable for use as a filling in a flour based product.

The reduction of water activity requires a relatively high proportion of sugar in water with a tendency to crystallise which is very high. In particular, the ratio by weight of (sugar): (water) ranges from about (4.5): (1) to (6.3): (1), preferably (6.0): (1).

Therefore, only the proposed controlled mixture of the three sugars of invention has proved to produce a filling with such a low water activity which will avoid loss of crispness in the flour based product but at the same time which will remain consistently plastic to provide an acceptable mouthfeel and to ensure processability on existing equipment.

As aforementioned, invert sugar which is an equimolecular mixture of glucose and fructose may be used as replacement of some or all of the glucose and fructose in the given proportions. However, whenever invert sugar is used, a proportion of sucrose also remains necessary.

However, the addition of other types of sugars such as maltose or lactose in significant amounts (at least 3W by weight) is not suitable. These sugars would lead to a risk of crystallisation due to their very limited solubility and relatively poor Aw lowering.

Baked flour-based products of the invention, such as sweet biscuits, may have'per se'a water activity of 0.3 and below before they are filled with the filling of the invention.

However, it has been found that in baked flour-based products, loss of desirable crispness occurs when the water activity surpasses a critical equilibrium value. The value depends to some extent to the formulation of the dough, but for most biscuits we found that it may be considered between 0.35 to 0.5, more preferably 0.39 to 0.43. If the equilibrium water activity of the biscuit increases above that critical maximum

value due to moisture transfer from the filling to the biscuits, the biscuit will loss its crispness. The invention resolves this problem by reducing drastically the water activity of the aqueous filling in such a way that at equilibrium between the filling and biscuit, the equilibrium water activity of the dough-based product is at or below the critical range, i. e., 0.35 to 0.5, preferably, 0.38 to 0.47 and, even more preferably 0.39 to 0.43. Therefore, according to an important aspect of the invention, a certain moisture transfer is allowed between the biscuit and the filling due to the low water activity level of the biscuit itself but at the equilibrium, the biscuit is kept crispy and the filling is kept soft. It must be understood that if the filling would transfer too much water to the biscuit, the result would lead to both a hard and non-palatable filling and a soggy biscuit.

In the context of the invention, the water activity or Aw refers to the ratio between the partial vapour pressure of water in the sample and the vapour pressure of pure water at the same temperature. It must be noted that the water activity values in the present description have been measured by the "Aqua Lab"water activity meter model CX-3 by Decagon Devices Inc.

We also found that the equilibrium water activity was also dependent on the amount of filling with respect to the dough part. If the amount of the filling is small, a higher water activity is tolerated and vicecersa. We found in our work that the optimum ratio of filling part to dough part in the dough- based product was between 25 : 75 to 35 : 65, preferably 30: 70.

The flavoured filling of the invention preferably comprises an amount of fruit pulp for the flavoured component.

The amount of fruit pulp is preferably at least of 15 to 28 wt% of the filling. It must be noted that the fruit pulp contributes in part or entirely to the constitution of the aqueous phase. Sources of fruit pulp are not limited. They can be strawberry, apple, orange, raspberry, prune, rhubarb, etc.

The addition of fruit pulp in the given amount has the consequence to lower the pH of the filling below 5 and,

preferably in a range of 3 to 3.4. In general, a small amount of pectin is preferred which confers after setting, a firmer texture to the filling. An amount of less than 1 % by weight of pectin may be used. For the pecting to start, an acidic medium must be introduced otherwise pectin will not react and the filling will remain a little liquid as compared to the sticky and firm texture of the usual fruit jam which is more desirable.

After mixing the various ingredients for making the filling, the moisture is evaporated until the final moisture content is achieved. Suitable final moisture content is considered to be within 8 to 12 %. We have found that under 9 % moisture content, the filling formulation would be too hard after equilibrium has occurred with the biscuit. Low moisture content also results in difficulties for pumping the filling into the deposit machines.

The use of non-natural Aw lowering agents such as humectants will be avoided at least in effective amounts. As effective amounts of humectants such as polyols, it is meant amounts sufficient to lower the Aw from at least 0.02 points.

Effective amounts would be amounts of glycerol from about 1% wt of humectant in the final filling.

Although the low Aw flavoured filling is soft, depending upon specific filling formulations, it has been found difficult to deposit using conventional sandwich stencil depositing equipment used in biscuit industry. By conventional stencil depositing equipment, we mean deposit machines such as standard"high speed creaming"machines commercialised by APV Baker Limited, SASIB UK LTD, PETERS USA LTD or others of various types and configurations such as 4-lane, 2-lane or 6- lane, etc., high speed creamers. The problem has been solved by adding a certain amount of hydrogenated vegetable fat to the filling formulation. The fat participates to the reduction of stickiness and also imparts lubrication for machining and cutting the deposit process onto the baked product. The hydrogenated vegetable fat also improves the finished appearance of the jam in that it gives a better gloss to the

surface. The use of animal fat, even if possible in theory, would be less desirable because of its taste.

Of course, other depositing machines may be used as well for which there are less depositing problems such as a PSD (Pacing Sandwiching Depositor) commercialised by SASIB UK LTD or others. In that case, the use of hydrogenated fat may be avoided.

The hydrogenated vegetable fat may have a melting point ranging from 28 to 35°C. Preferably, the hydrogenated vegetable fat is a vegetable fat from Soya which has a melting point comprised between 30 to 34 °C. It is preferably used in an emulsion with the aqueous phase in an amount from 6 to 9 % by weight of the total filling. The importance of the melting point is related to the final product mouthfeel and its ability to be emulsified and to stay emulsified during the whole process including filling production and depositing.

Suitable emulsifiers are glycerol-based lipids such as glycerol monostearate having a mono form between 55 to 90.

Monostearate is preferred as the mono form is the most active form. However, glycerol-based lipids with different levels of glycerol monosterate, disterate and tristerate may also : used. The concentration of the emulsifier has proved to be important since above a given value, air incorporation may cause undesirable colour changes of certain fruit-flavoured fillings. For instance, strawberry filling may turn from a red colour to an undesirable pink colour in the presence of higher levels of emulsifier. Therefore, the concentration of emulsifier is preferably in the range of 0.15 to 0.45%, preferably 0. 15-0.25% by weight of the fruit pulp content in the filling. The consistency of the filling is changed sufficiently to be allowed to flow more easily in standard deposit machines.

It has been found critical to achieve the correct amount of emulsification. Too little would not ensure that the fat would stay in emulsion whereas too much will start to create the possibility of aerating a fruit based filling during the process. As stated above, aerating may change the colour of

the fruit based filling from red to pink because of the incorporation of small air bubbles into the filling. The emulsifier is added in proportion to the fruit pulp as the fruit pulp is the main contributor of water to the mixture which needs to be emulsified.

Additional ingredients such as natural or artificial flavouring agents and colouring agents can be added to the filling as desired. Similarly, acids such as maleic or citric acids may be added in suitable amount to enhance acidic taste and assist in the setting of the gel. Pectin may or may not also be added to modify the fluidity of the filling depending on the type of application. When pectin is used, the pH is in a range of 3 to 3.8 pH.

The baked flour-based product may be chosen among short biscuits, cookies, or wafers or a combination thereof, or other similar crispy dough-based products or combinations thereof.

The invention encompasses a baked flour-based product comprising the flavoured filling of the invention. Typically preferred baked flour-based products are short dough biscuits comprising flour, sugar, fat, salt, water, milk and baking powder in suitable amount.

According to the present invention, a short dough biscuit is biscuit typically having to 100 parts of flour 30-45 parts of sugar and 20-45 parts of fat.

The baked flour-based products of the invention may have a water activity ranging from 0.15 to 0.20 before it is combined to the flavoured filling. The baked flour based products of the invention remain crispy after equilibrium is reached with the filling wherein Aw equilibrium is obtained at values at or below 0.35-0.5, preferably at or below 0.38-0.43. For instance, we found that the critical water activity was 0.40 for cookies and 0.44 for wafers.

The flavoured sugar filling preferably represents between 25 to 35% by weight of the filled product. The filled flour based product exhibits a shelf life of about 6 months or more at ambient temperature.

Example 1-Filling preparation : A strawberry. jam type filling having Aw in the desired range (0.39-0.43) is prepared according to the following formulation of Table 1.

Table 1: INGREDIENTS % BY WEIGHT Sucrose 7. 9 Pectin (if required) 0. 3 Citric acid 0. 2 Glucose (crystalline) 22.0 Fructose 44.< Strawberry pulp 22. 1 Glycerol Monostearate 0. 1 Vegetablefat 7. 0 Colour,Flavour Csp In a second embodiment, the filling is prepared according to the ingredients of Table 2 as follows.

Table 2 INGREDIENTS % BY WEIGHT Sucrose 7. 9 Pectin (if required) 0. 35 Citric acid 0. 2 Glucose (crystalline) 20.5 Invert sugar 42. 0 Strawberry pulp 22. 1 Glycerol Monostearate 0. 1 Vegetablefat 7. 0 Colour,Flavour Csp

After mixing the ingredients, moisture is evaporated until a final soluble solids content in a range of 84 to 86 % is achieved. Despite the low moisture content the fruit filling is still soft and elastic and performs very well as a filling for biscuits thanks to the specific choice and proportions of sugars in the filling.

Example 2-Dough preparation and final depositing The short dough biscuit is prepared separately according to the formulation of Table 3 as follows.

Table 3: INGREDIENTS % BY WEIGHT Sugar 20 Vegetablefat 13 Salt 0. 3 Water 7. 0 Flour 59 Salt 0. 5 Lecithin 0. 5 Milkpowder 2 Flavouring 0. 2 Bakingpowder 0*5 The biscuit is made by mixing the ingredients in a mixer in two stages; Fat, sugar, water, salt, lecithin and flavouring are mixed together for 4 minutes. Flour, milk powder, and baking powder are then added and mixed for a further 3 minutes. The dough is then put into a rotary moulder

to form the required shape and weight. The shaped dough pieces are the baked in an oven for approximately 5 minutes to the desired colour and moisture (typically between 2 and % moisture).

Depositing is carried out whilst the flavoured filling is maintained at a temperature of 40 to 45 °C in the depositing mode so as to maintain a pumpable consistency to the filling.

The amount of pectin can be optionally lowered, increased or removed to change the consistency of the filling and make it easier to pump. In particular, the level of acids and pectin may vary according to the nature of the fruit pulp.

In a"Peters"depositor, the filling passes under a stencil of the depositor and is wire cut onto a first biscuit portion whilst the biscuit portion rests in horizontal position. A second biscuit is placed on top of the deposited filling. A few seconds after, the biscuits is placed in vertical position Example 3-Comparative sensory rating tests-Standard shelf lifestudy Sweet biscuits were filled with 25 % weight of low Aw- strawberry jam (formula given on Table 1) adjusted to three different water activities, namely Aw = 0.42; Aw = 0.52 and Aw = 0. 60. The filled cookies were packed and heat sealed in moisture-proof metallized polypropylene pouches and stored at 25 °C at 80 % relative humidity. After two-months storage they were opened and tested for crispness (sensory) according to the following ratings: Very crispy Crispy Slightly crispy Not crispy The water activity of the biscuits was also measured after the two-months storage. Results were as follows:

Jam Aw Sensory rating Final Aw of biscuit 0.42 Very crispy 0.35 0.52 Slightly crispy 0.44 0.61 Not crispy 0.47 This example demonstrates that using a jam with Aw above the target value (Aw = 0.42) does not adequately preserve crispness during storage.

It also shows that when the water activity of the sweet biscuit increase above about 0.40 crispness is not preserved.

Example 4-Accelerated Shelf-life study (high relative humidity) Sweet biscuits were filled (25 % weight) with strawberry jam of Aw = 0. 42 (formula according to Table 1), packed in metallized polypropylene pouches and stored at 25 °C and a relative humidity of 80 %.

Samples were evaluated (sensory) for crispness at 1,2,3, and 4 months. ( 4 months being equivalent to 6 months) Results were as follows: Month Sensory rating 1 Very crispy 2 Crispy 3 Crispy 4 Crispy As is clear from the foregoing, various modifications of the present invention may be made without departure from the spirit and scope of the disclosure and description, and the invention may be embodied and/or practised suitably in the absence of and/or to the exclusion of physical structure and/or process steps and/or manipulations, conditions, substances employed, present and/or manipulated, and/or limitations not specifically disclosed therein.