The present invention relates to a frozen dessert product, including and in particular ice confection products, usually non-aerated, although aerated products are not excluded, and preferably to water ice products. When quiescently freezing these products relatively large ice crystals are formed, starting from the surface where the fluid mix is in thermal contact with a cold surface. This results in a dendritic crystal structure (needles) . Smaller ice crystals may be obtained by either continuously or intermittently whipping or agitating the mix when at least some ice crystals are present, resulting in a different eating experience, such as a softer texture and a creamier, smoother and less icy mouth feel. Obtaining these properties either requires an additional process step and more complicated production machines or additional additives.

An object of the present invention is to provide a system for easily controlling the ice crystal growth in quiescently frozen dessert products and especially in frozen confection products such as in particular fat free ice confection products like sorbet ice products and specifically water ice products where otherwise large dendritic ice crystals may be experienced.

Briefly in accordance with the present invention the growth of dendritic ice crystals in quiescently frozen dessert products can be controlled and the eating experiences of these quiescently frozen products may be improved by controlling ice crystal dendrite thickness not to exceed 100 μm and more preferred not to exceed 80 μm and even more preferred not to exceed 60 μm.

The controlled growth of the dendritic ice crystals is in particular important with water ices, which normally freeze

very hard and crunchy when quiescently frozen.

Determination of the thermal transition temperature may be carried out as follows by differential scanning calorimetry. About 10 mg of solution is sealed in a pan. An empty pan is used as a reference and these pans are crash-cooled and thereafter warmed at a rate of 5°C per minute. The thermal transition temperature is seen as a change in the heat capacity before the ice melting or dissolving curve. This change may be a step or a peak. There may be other small changes prior to this change, but the thermal transition is defined as the last change before the ice melting/dissolving curve.

The frozen products meeting the crystal size requirement of this invention, although quiescently frozen, showed a softer, chewier texture than the usual quiescently frozen water ice products. A smaller ice crystal size is achieved by altering the formulation to give a higher average molecular weight. For example conventional sweeteners such as sucrose and dextrose are replaced by corn syrups to achieve this purpose. Maltodextrins can also be used. Also the lactose content in any milk solids used may be lowered or even be absent. The use of higher molecular weight materials causes the water to solidify as an amorphous solid. This has negligible diffusion and is thus relatively stable. The glass transition temperature (Tg) of the products of the invention will usually be at least - 18°C and, in particular, at least -13°C.

The invention will be exemplified in the following example.

A fruit ice mix of composition A in % by weight as given in Table I was prepared in a usual manner and quiescently frozen by putting it in a metal mould for an ice lolly and placing the mould in a brine bath at -40°C. Simultaneously four comparative samples were prepared having the

compositions K, L, , N in Table I (wt %) .

After removal from the brine bath and demoulding the samples were tested for ice crystal dendrite thickness in μm at -20°C.

Table I

MD 40 ** 26, 2.0 water 69 76.0 76.92 77.1 75.1 ice crystal dendrite thickness (urn) 57.1 134.5 177.0 156.3 125.5

* LBG Locus bean gum ** MD 40 40 DE corn syrup solids

From the above results may be concluded that by monitoring the composition the crystal size is controlled.

A further experiment was made by preparing water ice mixes B, C and a comparative one P, the compositions in % by weight being given in Table II. These mixes were quiescently frozen (Q) by placing in metal moulds and putting these results in a brine bath at about -40°C. In addition the control mix P was slush frozen (S) in a usual manner by passing it through a scraped surface heat exchanger where it is agitated whilst freezing so that a slush, ie a dispersion of ice crystals in an aqueous fluid is obtained. This slush is filled into metal moulds which are put into the brine bath. The four different products were subjected to sensory tests by a trained taste panel.

The mean score on a lO-sca ^' le (1 = low, small; 10 = high, large) for some sensory attributes are shown in Table II.

From Table II may be concluded, that the improvement in texture which is brought about by slush-freezing can be matched or even bettered by controlling the crystal size using the composition. Thus products similar to slush frozen water ice can be produced without the need for extra equipment and processing steps as associated with slush freezing.