WO2011099876A1

WO2011099876A1 - Dairy product and process

Info

Publication number: WO2011099876A1
Application number: PCT/NZ2011/000023
Authority: WO
Inventors: Julita Maria Manski; Peter Gilbert Wiles
Original assignee: Fonterra Co-Operative Group Limited
Priority date: 2010-02-15
Filing date: 2011-02-15
Publication date: 2011-08-18

Abstract

A process is provided for preparing a yoghurt comprising combining (1) a heat treated daily composition comprising casein and (ii) a heat-treated mineral-depleted whey protein stream, wherein the mineral depleted whey protein stream is supplemented with a calcium-depleted source of casein that is included in the heat treatment step.

Description

DAIRY PRODUCT AND PROCESS

Technical field

The invention relates to a process for preparing a yoghurt and the product. Background

The texture of cultured foods and yoghurt (yogurt) and yoghurt-like foods has been the subject of intensive research and innovation for many years. One direction that processors have taken to modify the texture of yoghurt is to add polysaccharides, and non-dairy proteins e.g. gelatine (gelatin) etc. However, many customers prefer all-natural products, specifically products that comprise only dairy ingredients and starter cultures. Such products can be categorised as 'all dairy'.

To achieve this outcome, innovation has focused on means to modify the properties of the dairy proteins. One simple method of altering the properties of die milk proteins in yoghurt is the use of heat treatments to provide controlled denaturation of the proteins and thereby improve the texture and reduce syneresis - the physical and natural phenomenon of acidic milk protein gels to exude whey or serum from the coagulum. More sophisticated methods to modify the properties of the milk proteins have been directed towards modifying then ionic environment during the denaturation process.

WO 2008/063089 (Bhaskar and Valentim) discloses a process for preparing a yoghurt by including within a starting milk composition a calcium-depleted milk ingredient. The calcium-depleted milk ingredient may be at least 50% depleted in calcium and replaced with Na or K ions. The calcium- depleted yoghurt milk composition may also be modified in terms of raising the proportion of whey protein present relative to the proportion of casein. The fortification may be as high as 10 part casein to 90 parts whey protein. The whey protein may be sourced from whey protein permeate, whey protein concentrate or whey protein isolate. Preferably the yoghurt milk composition is heat treated (70 - 95°C and 5-20 minutes).

Bhaskar and Valentim disclose that the calcium-depletion of the milk ingredients may be carried out by methods disclosed in WO 01 /41579 or WO 01 /41578. In WO 2008/026940 (Bhaskar) discloses a process where "The calcium depleted milk protein concentrate can be used to prepare protein stabilised food products. Without being bound to particular theory, where oil or fat is dispersed in an aqueous medium, or water is dispersed in the lipid phase, the protein stabilised food product can be described as emulsified. In systems with Litde fat, stabilisation can surprisingly take the form of benefits to texture or reduced syneresis".

The food composition prepared using the calcium-depleted milk protein concentrate preferably contains from 0.01 % to 10% w/w of the ingredient (expressed on a dry basis [DB]), more preferably from 0.1 % to 5% DB of the calcium-depleted MPC.

The invention is also useful for stabilising suspensions of proteins, for example casein micelles and insoluble proteins.

The term "milk protein concentrate" (MPC) refers to a milk protein product in which greater than 40%, preferably greater than 50%, more preferably greater than 55%, most preferably greater than 70% of the solids-not-fat (SNF) is milk protein (by weight) and the weight ratio of casein to whey proteins is between about 95:5 and about 50:50, preferably between 90: 10 and 70:30, most preferably between 90:10 and 80:20. Such concentrates are known in the art. MPCs are frequently described widi the % diy matter as milk protein being appended to "MPC". For example MPC70 is an MPC with 70% of the dry matter as milk protein. Generally MPCs are prepared by processes involving ultr filtration eidier to prepare a stream enriched in casern or a stream enriched in whey protein. The streams may be blended to attain desired ratios of casein to whey protein. In another embodiment, the milk protein concentrate may be prepared by blending a stream of skim milk with a stream of whey protein concentrate prepared by ultrafiltration, treating either the skim rnilk stream or the combined stream by cadon exchange and optionally ^'concentrating or drying.

The term "texture" refers broadly to a rheological property of a food composition containing the ingredient of this invention. Rheological properties include gel and foam strengdis, viscosity and stress-strain characterisdcs when subject to either static or dynamic deformation; The texture of foodstuffs is important in terms of ease of handling, stability during keeping and defining shelf- Life and most importandy as a part of the product's sensory characteristics - namely die consumers' percepdons during mastication. A protein dispersion is a food product where the protein is in a particulate or micellar form suspended or dispersed amongst a continuous phase.

Calcium-depleted MPCs for use in the invention may be prepared according to the methods of WOOl /41 578.

Calcium-depleted MPC may be heat treated. WO2004/057971 describes a heat treated and decalcified milk protein concentrate (HY-MPC) that is a calcium-depleted MPC having whey proteins denatured. The denaturation is carried out by headng at a temperaaire above 65°C for sufficient time to allow denaturauon of whey proteins. The heating is generally carried out at a pH of 6.0-7.0, preferably 6.5-7.0. Preferably, heating is for at least 4 minutes in this embodiment.

Preferably the calcium-depleted MPC is dried to a moisture content of less than 5%, or a water activity level that facilitates storage of the dry ingredient for several months without undue deterioration.

The application of the ingredient of this invention is useful in facilitating fat emulsion stability in a wide variety of applications that involve fat droplet dispersions in an aqueous-based continuous phase. Non-limiting applications include, whole milk, buttermilk, filled and imitation milks, milk powders and filled milk powders, fat containing retentate powders, reconsdtuted milks, retentates and creams, coffee creamer and coffee whitener, ice-cream, infant formula, yoghurt (including set, stirred and drinking), mousse, soups, sauces, liqueurs, meat products, pet foods, mayonnaise, snack products, chocolate, confectionary, fat containing gels and the Like.

WO 01 /41578 (Bhaskar, Singh & Blazey) discloses: In another aspect, the invention provides a method of cheese manufacture which includes the step of adding a 30-100%, preferably 40-100% more preferably 50- 100% calcium depleted MPC or MPI powder to the milk used as the starting material. In particular the invention provides a method of cheese manufacture comprising: (a) dispersing in milk a dried MPC or MPI having at least 70% dry matter as milk protein; (b) treadng the resulung mixture with one or more coaguladng enzymes to produce a curd, and (c) processing the curd to make cheese; wherein the dried MPC or MPI has calcium depledon of 30-100%. WO 2004/057971 (Bhaskar, Havea & Elston) discloses the preparation of a calcium depleted MPC ingredient useful in cheese making.

The denatura ion of whey proteins can be achieved by either, or, combinations of any treatments that can induce whey protein denaturation including these: direct steam injection, indirect heating using for plate heat exchangers, ohmic heating, microwave heating, ultra high pressure treatment, alkali treatment followed by neutralisation (see, for example, WO 01 /52665). Heating is the preferred option, particularly heating the solution at pH 6.0-7. 0 (preferably pH 6.5-7. 0) at a temperature, preferably >65 °C, and for a time, preferably > 4 nun, sufficient to allow denaturation of whey proteins.

The heat treatment is . . . sufficient to allow denaturation of whey proteins and interaction with casein, (a) drying to prepare a dried product; wherein after step (ID) and before step (c) the pH of the solution is adjusted if necessary so that the heating at step (c) is carried out on a solution having a pH of 6.0-7. 0, preferably 6.5- 7.0.

The whey protein content of the product is about that of skim milk. The whey protein content is in a denatured state.

WO 2006/034857 (Bovetto et al.) disclose that a texturised yoghurt-like product may be prepared by fermenting a heat treated demineralised and pH adjusted whey protein stream (supplemented with lactose).

In NZ526878 (Anema and Lee) disclose a process to prepare a texturised product where a mixture of casein and whey protein are heated together at a selected pH in the range 5.0 to 8.0. After the heating step, the product is acidified to a lower pH. The process is directed towards producing a cheese-like product and there is no teaching of the use of cation modified sources of casein.

In WO2007/026053, Manner et al. disclose a process for the preparation of a texturised yoghurt using a calcium depleted source of casein. There is no teaching of use of a calcium depleted casein source with a purified whey protein source. US20080199567 (Aymard et al.) discloses a process for preparing yoghurt-like products enriched in protein, especially whey protein. The product may include caseinate. The product contains fibre. The whey protein may be denatured prior to the yoghurt making process (e.g. Simplesse 100 E, CP Kelco) and there is no teaching of separate heating of native whey protein and sodium caseinate.

US20090130258 (Aymard et al.) discloses a process for preparing yoghurt-like products enriched in protein, especially whey protein that includes the use of guar gum.

WO2005041677 (Burling et al.) discloses the preparation of a food stabilising ingredient involving the thermal denaturat on of low mineral whey protein streams.

It is an object of the invention to provide a method for preparing a yoghurt with improved texture using dairy ingredients, or to at least provide the public with a useful choice.

Disclosure of the invention

The applicants have found surprisingly that by modifying the process disclosed in WO

2008/063089, a yoghurt product of superior texture can be produced. The applicants have found that prior to acidification, all or part of the calcium depleted casein source may be co-heated with a whey protein stream low in minerals and then combined with the separately heat-treated yoghurt milk stream, which may contain the remaining portion of the calcium depleted casein source to provide a yoghurt product with a better texture. The mediod also allows maintenance of texture at a lower protein content.

In one aspect, the invention comprises a process for preparing a yoghurt comprising

a) combining (l) a heat treated d iry composition comprising casein and (ϋ) a heat treated mineral- depleted whey protein stream, wherein the mineral depleted whey protein stream is supplemented with a calcium depleted source of casein that is included in the heat treatment step,

b) acidifying the combined heat treated streams until a pH of approximately 4.5 is reached (typical range pH 3.8-5.5, preferably 4.2-4.7),

c) optionally smoothing the acidified material; wherein the pH of the whey-rich stream is adjusted with acid or base to a pH in the range of 6.0-8.0, preferably between 6.7 and 7.5 prior to heating with the calcium depleted casern source. Preferably the heat treated dauy composition comprising casein and the heat treated mineral depleted whey protein stream are both liquids. But one or both may be dried powders or a blend. Where necessary, liquid may be added to form the material to be acidified.

Generally the product is packaged alter the acidification step, or after the smoothing step when this is used.

The acidification may be carried out using fermentation.

Optional ingredients may be added at any convenient stage in the process.

Figure 1 shows a general scheme of the invention.

The dairy composition comprising casein may be any source of casein that is dispersed in an aqueous medium. Fresh or pasteurised milk or skim milk would most commonly suffice.

Reconstituted milks may be used. Retentates or milk protein concentrates (MPC) dispersed or dissolved in water or milk may be used also. The casein content is usually in the range 2-12%

The yoghurt may be a fermented dairy product as defined by CODEX STANDARD FOR

FERMENTED MILKS CODEX STAN 243-2003 1 , a daily product that involves the addition of an acidifying agent, a yoghurt drink, Petit Suisse, Greek yoghurt, a smoothie, a dessert, a spreadable product or a mayonnaise-like product.

The calcium depleted source of casein may be sodium casemate, potassium caseinate or ammonium casemate. Alternatively it may be a calcium depleted skim milk stream, a calcium depleted milk powder, a calcium depleted retentate or a calcium depleted MPC as taught bv WO 1 /41578, WO 01 /41 579, WO 2004/057971 , etc. These non-casemate ingredients mav be prepared in sodium, potassium or ammonium versions with preferably at least 50% or the calcium replaced with monovalent cations. More preferably more than 80% of the calcium is replaced with mono-valent cations and most preferably 90% or more is replaced with mono-valent cations.

Whey protein ingredients or streams that are low in minerals are known in the art. Whey protein isolates such as WPI895 from Fonterra Co-operative Group Limited, Auckland, New Zealand, or Bipro™ from Davisco, Le Suer, MN, USA are illustrative examples. The mineral content (ionic strength) of the whey protein stream is adjusted to allow interaction between die whey proteins and the calcium depleted casein source during heating. Generally, the low mineral whey protein should have an ash content of less than 2.5g per l OOg of solids and more preferably less than 2g per l OOg of solids. Such ash contents are known in the art to be achievable, for example, using ultrafiltration, dia filtration or dialysis. In another aspect, the calcium content of the low mineral whey protein stream may be reduced in calcium ions to preferably less than 80mg/ l 00g whey protein. The ratio calcium / beta-lactoglobulin of the whey protein ingredient or streams should be lower than 10 mg/g, and preferably lower than 5 mg/g, and even more preferably lower than 2 mg/g.

Alternatively, low mineral content whey streams may be prepared using electrodialysis or ion exchange processes or combinations of both and these may be additionally used in conjunction with ultrafiltration processes. A whey protein stream, with a denaturation level of less than 20%, more preferably less than 10%, may be sourced from any convenient supply of whey or microfiltered milk permeate.

The casein stream can be combined with a portion of the calcium depleted source of casein as taught by WO 2008/063089 prior to its own heat treatment, wherein between 10% and 100% of the calcium depleted casein source is combined with the low mineral whey protein stream to provide a yoghurt milk. More preferably, between 20% and 100% of the calcium depleted casein source is combined with the low mineral whey protein stream.

The low mineral whey protein stream may be combined with a calcium depleted source of casein in the range of proportions by weight of 99 parts whey protein to 1 part casein, to 40 parts whey protein to 60 parts casein. More preferably the proportions of low mineral whey protein and calcium depleted casein are in the range 95:5 to 45:55.

For convenience, the low mineral whey protein stream and a calcium depleted source of casein can be pre-prepared by mixing the dried ingredients in the desired proportions mentioned-above.

The casein and whey protein streams are subjected to separate heat treatment steps. The heat treatment given to each stream may be similar or different and is directed towards inducing protein interactions that ultimately result in a yoghurt product with superior texture. The heat treatment can range from about 60°C for some tens of minutes, e.g. 30, to about 140°C for a few seconds, preferably at least 2 seconds. All combinations within tins range are technically and commercially feasible wherein higher temperatures generally require shorter treatment times. Treatments at 75°C for an hour to 130°C for 5 seconds are preferred. Treatments at 80 to 95°C for 5 to 20 minutes are more preferred. Treatments at 95°C-130°C for 2 seconds-10 minutes may also be used with the shortest treatments being preferred at the higher temperatures in the range.

The heat treated casein rich stream and the heat U'eated whey protein rich stream may be mixed together to provide a yoghurt milk with casein to whey protein ratios by weight in the range 95:5 to 30:70, alternatively 75:25 to 10:90, more preferably 75:25 to 30:70, most preferably 70:30 to 40:60. This means that the heat U'eated casein rich stream and the heat treated whey protein rich stteam may be mixed together in the range of proportions by weight of 60 parts heat treated whey protein rich s eam to 40 parts heat U'eated casein rich sueam, to 5 parts heat ueated whey protein rich sueam to 95 parts heat ueated casein rich stream, to provide a yoghurt milk. The overall protein content of the mixture may be between 1.5% and 15%, more preferably between 2% and 8%. The mixture may optionally contain fat in the range about 0.1 % to about 15%.

Before or following the mixing of the streams various optional ingredients may be added as desired. Opuonal ingredients may include simple and complex saccharides, natural and synthetic

polysaccharides, native and modified pectin, native and modified starch, gelatine, edible fibre, gums, yeast, vitamins and the like.

The temperature of the mixture can be adjusted to facilitate the growth of food approved lactic acid producing bacteria. Powdered lactic cultures may be added directly to the yoghurt milk, or a separate starter culture may be prepared and mixed into the yoghurt milk.

In an alternative embodiment, the acidification may be achieved all or in part by indirect means. The formation of lactic acid and the reduction in pH may be obtained, all or in part, using a slow release acidulant such as glucono-delta-lactone (GDL). The optimum pH required by the formation of acid is close to the isoelectric point of the casein proteins. Alternatively, the required pH is close to 4.5, but pH values in the range 3.8 to 5.5 are generally satisfactory. Small reductions in pH to about 5.5 may be achieved by the direct addition of a food approved acid.

Once the required pH has been obtained, the preparation of the yoghurt product may be conducted by processes well known in the art. Set, stirred (smoothed) or drinking yoghurt styles may be produced. Fruits, prepared fruit preparations, colouring, minerals and flavourings may be added at any convenient stage up to the final packaging. Smoothing of yoghurt fermentation may be carried by methods known in the art for smoothing out or breaking up any yoghurt lumps or curds present. Such smoothing may be carried out passing the fermentate through a mesh screen or a valve. Alternatively, the fermentate may be sheared using a blender.

Unless otherwise specified percentage compositions are ratios are on a weight basis. Brief Description of the Figure Figure 1 shows a general scheme of the invention. EXAMPLES

The following examples further illustrate the invention.

1. MATERIALS

Low-heat skim milk powder (SMP), cream, lactose, calcium-depleted milk protein concentrate (MPC 4862), sodium casemate and whey protein isolate (WPI 895) are all from Fonterra Co-operative Group Ltd, Auckland, New Zealand.

A prototype of low mineral cheese whey protein concentrate (WPC) was manufactured at the Pilot Plant of Fonterra Research Centi'e, Palmerston North, New Zealand usmg ultrafiltration, diafilitration and cation-exchange processes to remove die minerals from the feed stream. The product was then spray dried. The low mineral cheese WPC powder had the following composition: 78.3% true protein ((TN-NPN)*6.38), 5.0% lactose, 1.2% ash, 35 mg calcium per 100 g powder, 4.9% denatured whey protein (by HPLC analysis).

Glucono-delta-lactone (GDL) was obtained from Sigma.

HC1, NaOH and sodium azide were obtained from Merck.

Potassium sorbate was obtained from Hawkins Watts Ltd (New Zealand).

Sugar (extra fine grade) was obtained from Chelsea Sugar Refinery (Auckland, New Zealand).

Yoghurts were fermented with starter culture YF-L702 {Streptococcus thermophilus and Lactoba /h/s delbruecki ssp. Bulgarian), Christian Hansen, Denmark). 2. METHODS 2.1 PREPARATION OF STIRRED AND SET ACID GDL GELS

2.1.1 Stirred acid GDL gels manufacture

For stirred acid gel manufacture, ingredients were recombined with tap water (with typical batch size between 150 - 750 g) at ambient temperature and hydra ted (stirred) for at least 2 hours. All milks contained 0.01 % w/w sodium azide to prevent microbial growth during preparation and storage. After recombining, the pH of the milks was measured and adjusted where required using 1 M HC1 or 1 M NaOH. In general, all control milks and all whey protein-rich streams containing calcium- depleted casein were adjusted to pH 6.8 before heating, unless stated differendy in the specific examples. A day after recombining, milks were given a heat treatment of 85°C for 1 5 min holding time (come-up time was typically 5 min) in a water bath, except for Example 6 where the milk and various milk streams were heated at 1 10°C for 2 mill in an oil bath (come-up time was typically 45 seconds). After heating, the milks were cooled in ice water to below 50°C. The various milk streams (base milks and whey protein-rich streams) were mixed for about 30 min as required to provide a yoghurt milk. For each yoghurt milk that was going to be acidified with GDL, the appropriate GDL concentration was determined to reach pH 4.0-4.3 after 5.5 h of incubation at 42°C. The second day after recombining, the GDL was added to the yoghurt milks and incubated for 5.5 h at 42°C. After the incubation time, the acid gels were cooled in ice water while gentlv breaking the curd. The acid gels were smoothed when they reached a temperature of 18-22°C using an Ultra-Turrax 0anke & Kunkel, IKA Labtechnik, N8 head) at the lowest speed (8000 rpm) for typically 1 -5 min until the stirred product looked visually smooth. After smoothing, each stirred acid gel product was packed in 00 g cups and transferred into a chiller (4-8°C) until further analysis at day 7.

2.1.2 Analysis of set acid gels (measurement of G' at 10°C and 42°C)

Sub-samples of the yoghurt milks that were used for incubation and to make stirred acid gels (as above) were acidified in situ in the rheometer (Physica MCR301 , cup and bob CC27, Anton Paar, Austria) to follow the gelation profile at 42°C. A 25 g sub sample of heated milk was warmed to 42°C in a water bath and dien the appropriate amount of GDL was added and the gently stirred sample was transferred as quickly as possible to the pre-warmed rheometer cup. The G' and G" were followed for 5.5 h (strain 0.1 %, frequency 0.1 Hz) and after 5.5 h the gel was cooled in situ to 10°C with l °C/min and held for 30 min at 10°C. A frequency sweep (strain 0.5%, frequency 0.01 -10 Hz) and a strain sweep (strain 0.1 -300%, frequency 0.1 Hz) was performed to characterize the set gel at 10°C. G' at 42°C after 5.5. h of incubation and at 10°C is reported here.

2.1.3 Analysis of stirred acid gels (measurement of pH, viscosity) The pH of stirred acid gels was measured at 10°C using Radiometer Copenhagen pH meter, Model pHM209. At least duplicate measurements were carried out. The pH meter was calibrated before use with standardised buffer soluuons at pH 7.0 and 4.0.

After 7 days of storage, the viscosity of stirred acid gels was measured at 10°C using the same rheometer and geometry set-up as above. Upward and downward flow curves (shear rate 0.1 -500 s ' , 30 data points collected, with each shear rate applied for 10 s) were measured in duplicate from 2 cups from each gel sample. The measured viscosity at 53 s ¹ from the upward flow curve is reported.

2.2 PREPARATION OF YOGHURTS

2.2.1 Set and stirred yoghurt manufacture

All ingredient powders were recombined with reverse-osmosis water at 50 ± 5°C until dissolved completely. All milks contained 0.02% w/w potassium sorbate to prevent growth of moulds. The milk streams (typical batch size of 23 kg) were processed in-line using a process comprising warming of the milk to 60°C, homogenising (1 50/50 bar, Rannie Lab, APV, Denmark), heat treating (95°C for 8 min) and collecting die milks at 42°C. This heat treatment was used for all yoghurt examples. . The various milk streams were mixed prior to fermentation according to the required recipes to provide a yoghurt milk. Then the milks were inoculated with starter culture (YF-L702, Christian Hansen, Denmark) and incubated at 42°C. Incubation was stopped when pH 4.6 was reached, and the curd of the set yoghurts was gently broken and passed through a plate heat exchanger (APV, Denmark) to cool the curds to 20°C. All yoghurts were smoothed using a back pressure valve (Mono Pump Baureihe 25) until the yoghurts looked visually smooth. After smoothing, the stirred yoghurts were packed in 100 g cups and stored in a chiller (4-10°C) until furdier analysis at day 7.

2.2.2 Analysis of set yoghurts (area force-distance curve)

The texture profile of set yoghurt was measured using a Stable Micro Systems TA-XT2 Texture Analyser with a real tune graphics and data acquisition software package (Texture Expert) from Stable Micro Systems, Godalming, Surrey GU7 1YL UK. A penetration test was carried out for set yoghurt samples (at 5-10°C ex fridge) using a 1 3 mm (0.5 inch) diameter cylindrical probe at a constant rate (1 mm/s) for a set distance (20 mm), then withdrawing the probe at 5 mm/s. The response was recorded as force (g) versus distance. The initial peak force generated during the initial penetration of the yoghurt (the first peak - fracture force) and the work, i.e. positive area (gxmm) under the force/distance curve (consistency) were recorded. The test was performed in triplicate from different yoghurt cups. 2.2.3 Analysis of stirred yoghurts (pH, drained syneresis, viscosity)

The pH of both set and stirred yoghurt samples was measured at 10°C using Radiometer

Copenhagen pH meters, Models pHM210, or pHM82. At least duplicate measurements were carried out. The pH meter was calibrated before use with standardised buffer solutions at pH 7.0 and 4.0.

The drained syneresis of stirred yoghurt (after 1 week of storage at 4°C) was measured in duplicate from different yoghurt cups using mesh screens that were wetted with water prior to placing 38-40 g of sdi'ied yoghurt on the screens. The screens were placed on pre-weighed cups and the samples were left for 2 hours standing at 4°C. After 2 h, the screen and yoghurt was removed from the cup and the cup with the drained whey was weighed. The drained syneresis was the percentage of whey drained form the yoghurt sample (weight basis).

The viscosity of stirred yoghurts was measured in duplicate from different yoghurt cups using a Haake VT500 Viscometer (Haake Mess-Technik, GmbH u. Co., Karlsruhe) fitted with the M cup and rotor sensor system at 10°C (yoghurt sample straight from the fridge). The shear rate was increased from 0 to 120 s^{' 1} over a period of 180 s, collecting 100 data points, then the shear rate was reduced to 0 s^"' over 10 s collecting only 2 data points. The response was measured as shear stress and viscosity. The apparent viscosity value at 50 s^{' 1} is reported.

3. Experimental Scheme

The examples illustrate the invention, i.e. the effect of mixing a heated casein protein-rich stream that may or may not contain a calcium depleted source of casein at the moment of heating and a heated whey protein-rich stream that contains a calcium depleted source of casein at the moment of heating, compared to a control milk that has been processed in the traditional way where all ingredients/components are heat treated simultaneously. Several variadons of the invendve method to provide a superior yoghurt texture are demonstrated in the examples. The variadons include:

1. different propordons of mixing the heated casein protein-rich and whey protein-rich streams after heating to provide the yoghurt milk;

2. different proportions of the calcium depleted source of casein present at the moment of heaung the casein protein-rich and whey protein-rich streams;

3. different protein and fat contents and final caseimwhey rauos in the yoghurts and acid gels.

Table 1 gives an outline of die inventive method variants and the coding used in the examples. Besides a superior texture formed when using the invendve method (e.g. via increase in viscosity), the examples also demonstrate that this texture benefit can be applied to reduce protein in a yoghurt formulation. Finally the examples show the effect of the pH of the whey protein-rich stream before heating and the wide heating regime that can be used.

The effect of the novel process and its variations is elucidated with examples of stirred and set acid gels acidified with GDL and set and stirred yoghuits containing cultures. The effect of the novel process is demonstrated with some key properties that are relevant for yoghurt functionality, texture and sensory properties. Strong correlations exist between theological parameters (such as viscosity and G') and sensory properties of yoghurts (e.g. Van Oosten-Manski et al. , Conference Proceedings, pp 388-391 , Fifth International Symposium on Food Rheology and Structure - ISFRS 2009, Zurich, Switzerland, June 1 5- 1 8, 2009). An increase in viscosity is generally accepted to be reflected in the texture or sensory properties of stirred yoghurts as desirable and an increase in G' or the area under the force-distance curve is related to an increase in in-mouth firmness of set yoghurts. Acid gels have proven to be a good model system for yoghurt, and thus properties and trends observed in acid gels generally correlate well with yoghurt properties.

Table 1 - Outline of the inventive method variants used in the examples and the yoghurt / acid gel compositions of the samples prepared with the inventive method variants

invention)

79.75 20.25 0 75 25 E 4.2 0.1 % #

85 15 0 0 100 F 4.94 0.1 % #

EXAMPLE 4 79 1 5 6 0 100 G 4 65 0.1 % #

(demonstrating 71 29 0 66 34 H 4.94 0.1 % # protein

reduction using 65 29 6 66 34 I 4.65 0.1 % # the invention) 71 29 0 66 34 J 4.94 0.1 % #

65 29 6 66 34 K 4.65 0.1% #

EXAMPLE 5

(demonstrating

superior texture

using the L, M, N,

50 50 0 0 100 3.8 0, 1 % # invention and O

effect of pH of

whey protein- rich stream)

EXAMPLE

6(demonstratin

g protein

reduction using

86.4 1 3.6 0 71 29

the invention P, Q 4.64 0, 1% # when heating

at temperatures

above 100°C)

EXAMPLE 7

(demonstrating

superior texture 75 25 0 0 100 R 4.5 1 % using the

invention)

EXAMPLE 1: MPC 4862 and WPI 895 are used in 2 inventive method variants and

compared to a non-inventive control (acid gels with 4.2% protein and 65:35 casein:whey ratio)

Stirred acid gels with SMP, MPC 4862 and WPI 895 were prepared with final caseiniwhe rauo of 65:35 using the procedure of Section 2.1. The control sample was made by heat treating all

ingredients together (Table 2). Two inventive samples were made using method variants A and B (Table 2), which illustrate different divisions of decalcified caseins over the casein protein-rich and whey protein-rich streams at the moment of heating (see Table 1 and 2). Surprisingly, both the set gel properties G^? and the stirred gel viscosities of the samples made using inventive method variants A and B were higher than those of the control sample (Table 3). All the stirred acid gels had the same gross composition. The example illustrates that separately^" heating whey proteins with (part of) the calcium-depleted caseins and separately heating the base skim nulk (casein rotein -rich, stream) with or without part of the calcium-depleted caseins leads to a greater viscosity than the traditional heating process (about 39% viscosity increase). Method variants A and B gave similar results. Table 2 - Formulations of milks and gross compositions of stirred acid gels (control and samples obtained with inventive method variants A and B) using MPC 4862 and WPI 895, with final casem:whey ratio 65: 35 and 4, 2% protein

Table 3 - Physical properties of stirred acid gels (control and samples obtained with inventive method variants A and B) using MPC 4862 and WPI 895, with final casein:whey ratio 65:35 and 4 2% protein

EXAMPLE 2: Sodium caseinate and WPI 895 are used in one inventive method variant and compared to a non-inventive control (acid gels with 4,2% protein and 70:30 casein:whey ratio) Stared acid gels with SMP, sodium caseinate and WPI 895 were prepared with final casein:whey ratio of 70:30 using the procedure of Secuon 2.1. The control sample was made by heat treating the ingredients together and one inventive sample with method variant C was made (Table 4) where all decalcified caseins were added to the whey protein-rich stream prior to heating. In the current example, the caseimwhey ratio of the whey protein-rich stream with calcium-depleted casein is higher (higher casein concentration) than in the first example. Surprisingly, the stirred gel viscosity of the sample made using method variant C was higher than the control sample (Table 5). The two stirred acid gels had the same gross composition. This sample illustrates that sodium caseinate, essentially a decalcified casein ingredient, was used in the inventive blend to increase the viscosity of a stirred acid gel by about 20% over the control.

Table 4 - Formulations of milks and gross compositions of stirred acid gels (control and sample obtained with inventive method variant C) using sodium caseinate and WPI 895, with final casein:whey ratio 70:30 and 4.2% protein

Table 5 - Physical properties of stirred acid gels (control and sample obtained with inventive method variant C) using sodium caseinate and WPI 895, with final casein:whey ratio 70:30 and 4 2% protein

EXAMPLE 3: MPC 4862 and low mineral cheese WPC are used in 2 inventive method variants and compared to a non-inventive control (acid gels with 4.2% protein and 65:35 casein:whey ratio)

Stirred acid gels with SMP, MPC 4862 and low mineral cheese WPC were prepared with a final casein:whe ratio of 65:35 using the procedure of Section 2.1. The control sample was made by heat treating the ingredients together and two inventive samples using method variants D and E were made (Table 6) , to illustrate different divisions of decalcified caseins over the casein protein -rich and whey protein-rich streams at the moment of heating. In process variant E, 75% of the decalcified casein was added to the casein-nch stream and 25% of the decalcified casein was added to die whey protein-rich stream prior to heating (see Table 1 and 6). Surprisingly, the stirred gel viscosities of the samples made using method variants D and E were higher than those of the control sample, leading to about 20% viscosity increase (Table 7). The increased shearing time required with the Ultra- Turrax to obtain a smooth product was consistent with the increases in viscosity. Note that all stirred acid gels had the same gross composition. Again, process variants D and E gave similar results, which revealed surprisingly that varying proportions of the calcium-depleted source of casein may be

split between the base milk and whey protein-rich stream with a benefit over the control.

Table 6 - Formulations of milks and gross compositions of stirred acid gels (control and samples obtained with process variants D and E) using MPC 4862 and low mineral cheese WPC, with final casein whey ratio 65:35 and 4.2% protein

Milk sample codes FCX-51 FCX-57 FCX-512 FCX-55 FCX-58 FCX-513 FCX-54

Table 7 - Physical properties of stirred acid gels (control and samples obtained with process variants D and E) using MPC 4862 and low mineral cheese WPC, with final casein:whey ratio 65: 35 and 4.2% protein

EXAMPLE 4: MPC 4862 and WPI 895 used in 6 inventive method variants with 3 different final casein:whey ratios and 2 final protein levels ('yoghurts with 4,94% and 4.65% protein to demonstrate protein reduction) and compared to a non-inventive control (yoghurts with 4.94% protein and 75:25 casein whey ratio)

Set and stirred yoghurts with SMP, MPC 4S62 and WPI 895 were prepared with three different final ratios (74:26, 7 1 :29 and 68:32) using the procedure described in Section 2.2. The control sample was made by heat treating all ingredients together with a final

ratio of 75:25 in the yoghurt. Samples were made by using inventive method variants F, G, H, I, ] and K where different proportions of decalcified caseins were added to the whey protein-rich streams prior to heating (100% of decalcified caseins were added to die whey protem-nch streams in method variants F and G, whereas 34% of decalcified caseins were added to the whey protein-rich streams in all the other method variants; see Table 1 and 8). This example illustrated a viscosity benefit of using the various inventive mediod variants and also the potential to use the novel process to reduce the protein in d e final yoghurt (0.3% protein reduction . in final protein content (4.65% compared to 4.94%), or 6% protein reduction in general (0.3% / 4.94%)). All yoghurt milks were processed using the yoghurt preparation procedure detailed in Section 2.2. All of the separately heated base milks with or without decalcified casein were pH adjusted to pH 6.65 whereas the whey protein-rich streams (pH 6.93) required no adjustment. The yoghurts contained a final carbohydrate (from sugar and lactose) concentration of 10.3 g per 100 g of yoghurt.

At equal protein content method variant F (but also method variants H and J, although the final casein:whey ratio of these yoghurts was different compared to the control) gave a higher viscosity and clearly increased set gel strength (area under force-distance curve) than the control yoghurt (Table 9) . Also the syneresis was slightly reduced for samples obtained with the method variants F, H and } compared to the control at equal protein content.

All the protein reduced yoghurts showed a very similar or unproved viscosity and set gel strength compared to the control, which indicated that the target properties of the control could be matched by using the inventive process to allow for protein reduction in the yoghurts.

Table 8 - Formulations of milks and gross compositions of yoghurts (control and samples obtained with inventive method variants F, G, H , I , and K) using MPC 4862 and WPI 895, with two final protein levels (4.94% and 4.65%) and different casein:whey ratios

Table 8 Continued

65% of

71 % of

casein-rich casein-rich

Whey Whey stream and

Casein- stream and Casein- protein- 29% of protein- 29% of rich rich ^' whey rich whey rich

stream stream protein-rich stream protein-rich

stream stream and stream by

6% of water weight

by weight

Low-heat SMP (g) 1 260. 1 1 0 1260 1 1 1 152.84 0 1 1 52 84

M PC 4862 (g) 0 104 76 1 04.76 0 104.76 104.76

WPI 895 (g) 0 92.90 92.90 0 92.90 92.90

Sugar (g) 41 7.06 0 41 7, 06 462.71 0 462.71

Lactose (g) 144.00 0 144.00 157.31 0 157.31

Potassium sorbate (g) 1 .70 0.70 2,40 1 .56 0.70 2.26

Culture Christian Hansen YFL-702 (g) na na 2.40 na na 2.40

Water (g) 6697.-13 3281 .64 9978.77 6025.58 3281 .64 10025 22

Total (g) 8520.00 3480.00 12000.00 7800.00 3480.00 12000.40

Final protein target (%) 4.94 4.94 4.94 4.94 4.94 4.65

Final casein:whey ratio target 80 : 20 40 : 60 68 ; 32 80 : 20 40 : 60 68 : 32

Final ash target (%) 1 .17 0.26 0.91 1 . 17 0.26 0.84 pH before heating (-) 6.65 6.93 * 6.65 6.93 *

Milk sample codes FC -18 FC -19 FCM-12 FCM-22 FCM-19 FCM-13

Low-heat SMP (g) 1260. 1 1 0 1260.1 1 152.84 0 1 1 52.84

MPC 4862 (g) 0 130 29 1 30.29 0 130.29 130.29

WPI 895 (g) 0 70.16 70.16 0 70.16 70.16

Sugar (g) 417,06 0 417.06 462.71 0 462.71

Lactose (g) 144.00 0 144.00 157.31 0 157.31

Potassium sorbate (g) 1 .70 0.70 2.40 1 .56 0.70 2.26

Culture Christian Hansen YFL-702 (g) na na 2.40 na p. a 2.40

Water (g) 6697. 13 3278.86 9975.98 6025.58 3278.86 10022.43

Total (g) 8520.00 3480.00 12000.00 7800.00 3480.00 12000.40

Final protein target (%) 4.94 4.94 4.94 4.94 4.94 4.65

Final casein:whey ratio target 80 : 20 50 : 50 71 : 29 80 : 20 50 : 50 71 : 29

Final ash target (%) 1 .17 0.31 0.92 1.17 0.31 0.85 pH before heating (-) 6.65 6.93 * 6.65 6.93 *

Milk sample codes FCM-18 FCM-20 FC -14 FCM-22 FCM-20 FCM-15 Table 9 - Physical properties of yoghurts (control and samples obtained with inventive method variants F, G, H, I, and K) using MPC 4862 and WPI 895, with two final protein levels (4.94% and 4 65%) and

different casein whey ratios

EXAMPLE 5: Sodium caseinate and WPI 895 are used in 4 inventive method variants (to demonstrate the effect of the pH of the whey protein-rich stream before heating¹) and compared to a non-inventive control (acid gels with 3.8% protein and 60:40 caseimwhey ratio)

Stirred acid gels with SMP, sodium caseinate and WPI 895 were prepared with final casetn:whey ratio of 60:40 using the procedure of Section 2.1. The control sample was made by heat treating the ingredients together and four inventive samples using method variants L, M, N and O were made (see Table 1 and 10) where all decalcified caserns were added to the whey protein-nch stream prior to heating. In addition, prior to heating, the pH of the whey protein-rich stream was adjusted to values in the range of 5.8 - 8.5 to illustrate the desired operating range for the novel process. The pH ot the casein protein -rich stream prior to heating was kept constant at a value ot 6.8. The inventive samples were prepared by mixing 50% of the casern protein-rich stream and 50% of the whey protein-rich stream. The examples in Table 1 1 indicate that the best effect is around pH 6.8 for the pH of the whey protein-rich stream before heating, as this sample had the highest viscosity increase compared to the control (34%). At lower pH (5.81) the viscosity was much lower compared to the control and at higher pH values (7.49 and 8.52) the viscosities are slightly reduced. Table 10 - Formulations of milks and gross compositions of stirred acid gels (control and samples obtained with inventive method variants L, M, N and O where the whey protein-rich streams was pH- adjusted to different pH values before heating) using sodium caseinate and WPI 895, with final ^' casein:whey ratio 60:40 and 3.8% protein

Table 1 1 - Physical properties of stirred acid gels (control and samples obtained with inventive method variants L, M, N and O where the whey protein-rich streams was pH-adjusted to different pH values before heating) using sodium caseinate and WPI 895, with final casein:whey ratio 60:40 and 3.8% protein

EXAMPLE 6: MPC 4862 and WPI 895 are used in 2 inventive method variants using heating temperatures above 100°C (acid gels at 4.95% and 4.65% protein and 74:26 casein:\vhey ratio), and compared to a non-inventive control (acid gels with 4.95% protein, to

demonstrate protein reduction, and 75:25 casein:whey ratio)

Set acid gels with SMP, MPC 4862 and WPI 895 were prepared with milks that were heat treated above 100°C. A control sample with a final protein content of 4.95% protein and final caseimwhey ratio of 75:25 in the acid gel was made by heat treating all ingredients together at 1 1 0°C for 2 minutes to provide evidence of the firmness (G') achievable for set acid gels when heated at higher temperatures. The samples that were prepared with the inventive medrod variants P and Q,

(referring to high temperature heating and different protein contents; see Table 12) had a final protein content of 4.95% or 4.65% protein and final casein:whey ratio of 74:26 in the set gel to show the process can be applied to reduce protein in yoghurts. In this variant 29% of the decalcified caseins were added to the whey protein-rich stream prior to headng. The pH of the whey protein-rich stream prior to heating was adjusted to approximately 7.0. The casein protein-rich and whey protein-rich streams were given a heat treatment of 1 10°C for 2 minutes (Table 1 1 ) . The acid gel milks were prepared by mixing the heated casein protein-rich stream and the heated whey protein-rich stream to the casein rich stream before acidification. The examples in Table 13 show that the protein reduced sample with the inventive method variant Q, matched the firmness of the control at 4.95% protein in the high heating regime of 1 10°C for 2 minutes.

Table 12 - Formulations of milks and gross compositions of set acid milk gels (control at 4.95% protein and samples with 4.95% and 4.65% protein obtained with inventive method variants P and Q, where the casein protein-rich and whey protein-rich streams were heat treated at a temperature of 1 10°C with a holding time of 2 minutes using MPC 4862 and WPI 895

Table 12 continued

Table 13 - Physical properties of acid milk gels (control at 4.95% protein and samples with 4.95% and 4.65% protein obtained with inventive method variants P and Q where the casein protein-rich and whey protein-rich streams were heat treated at a temperature of 1 10°C with a holding time of 2 mm using MPC 4862 and WPI 895

PHYSICAL DATA

Inventive method Inventive method EXAMPLE 6 Control

variant P variant Q

Set acid gels

Protein content (%) 4 95 4.95 4 65

G' of set gel at 10°C,(Pa) 2933 3723 3272

G' increase against control na 27% 11 %

pH of set gel at day 1 4.3 4.3 4.3

EXAMPLE 7: MPC 4862 and low mineral cheese WPC are used in one inventive method variant and compared to a non-inventive control (yoghurts with 4.5% protein, 1% fat and 67:33 casein:whey ratio)

Stirred and set yoghurts with S P, MPC 4862 and low mineral cheese WPC were prepared with a final

ratio of 67:33 using the procedure of Section 2.2. The control sample was made b heat treating the ingredients together and one inventive sample using method variant R was made (Table 14), illustrating that the novel process provides a viscosity benefit in a yoghurt formulation containing 1% fat. In method variant R, 100% of the MPC 4862 casein proteins were added to the whey protein-rich stream prior to heating. After heating the casein protein-rich and whey protein- rich streams were mixed in a 75:25 ratio before fermenting the yoghurt milk. Table 15 shows that a viscosity benefit of 36% and a reduction in syneresis was found when using the novel process (method variant R).

Table 14 - Formulations of milks and gross compositions of yoghurts containing 1 % fat (control and sample obtained with inventive method variant R) using MPC 4862 and low mineral cheese WPC

Inventive method variant R

75% of casein-rich

RECIPE EXAMPLE 7 Control Whey

Casein-rich stream + 15% protein-rich of whey stream

stream protein-rich stream by weight

Low-heat SMP (g) 1242.00 1241 .82 0 1241 .82

MPC 4862 (g) 36.60 0 36.85 36,85

Low mineral cheese WPC (g) 109.80 0 1 10.54 1 10.54

Cream (g) (g) 246.00 246.31 0 246.31

Potassium sorbate (g) 2.40 1.87 0 1 .87 Culture Christian Hansen YFL-702 (g) 2.40 na na 2.40

Water (g) 10363.20 7840.00 2522.62 10362.62

Total (g) 12002.40 9330.00 2670.00 12002.40

Final protein target (%) 4.50 4.50 4.50 4.50

Final fat target (%) 1 .00 1.22 0,28 1 .00

Final casein:whey ratio target 67 : 33 ^" 80 : 20 20 : 80 67 : 33

Final ash target (%) 0.86 1 .05 0.18 0.86 pH before heating (-) 6.74 6.73 6.80 *

Milk sample codes FCB24 FCB27 FCB29 FCB25

Table 15 - Physical properties of set and stirred yoghurts containing 1 % fat (control and sample obtained with inventive method variant R) using MPC 4862 and low mineral cheese WPC

The term "comprising" as used in this specification means "consisting at least in part of. When interpreting each statement in this specification that includes the term "comprising", features other than that or those prefaced by the term may also be present. Related terms such as "comprise" and "comprises" are to be interpreted in the same manner.

In this specification, where reference has been made to external sources of information, including patent specifications and other documents, tins is generally for the purpose of providing a context for discussing the features of the present invention. Unless stated otherwise, reference to such sources of information is not to be construed, in any jurisdiction, as an admission that such sources of information are prior art or form part of the common general knowledge in the art. ..

It is not the intention to limit the scope of the uivention to the abovementioned examples only. As would be appreciated by a skilled person in the art, many variations are possible without departing from the scope of the invention (as set out in the accompanying claims).

Claims

WHAT WE CLAIM IS

1. A process tor preparing a yoghurt comprising

a) combining (1) a heat treated daily composition comprising casein and (11) a heat-treated mineral-depleted whey protein stream, wherein the mineral depleted whey protein stream is supplemented with a calcium-depleted source of casein that is included in the heat ti'eatment step, b) acidifying the combined heat-treated streams to a pH of 3.8 - 5.5, c) optionally smoothing the acidified material; wherein the pH of the whey-rich stream is adjusted with acid or base to a pH in the range of 6.0-8.0, prior to heating with the calcium depleted casein source.

2. A process as claimed in claim 1 wherein the product is packaged after the acidifying step, or after the smoothing step when this is used.

3. A process as claimed in claim 1 or claim 2 wherein the acidifying step is a fermentation step.

4. A process as claimed in any one of claims 1 -3 wherein the dairy composition comprising casein is selected from fresh or pasteurised milk or sl m miik, reconstituted milks and retentates or milk protein concentrates (MPCs) dispersed or dissolved in water or milk.

5. A process as claimed in claim 4 wherein the casein content of the dairy composition is in the range 2-12% (w/v).

6. A process as claimed in any one of claims 1 -4 wherein the calcium depleted source of casein is sodium caseinate, potassium caseinate or ammonium caseinate.

7. A process as claimed in any one of claims 1 -5 wherein the calcium depleted source of casein is calcium depleted skim milk, calcium depleted milk powder, a calcium depleted retenate or a calcium depleted MPC.

8. A process as claimed in claim 7 wherein the calcium is replaced with sodium, potassium or ammonium.

9. A process as claimed in claim 8 wherein at least 50% of the calcium replaced with monovalent cations.

10. A process as claimed in claim 9 wherein more than 80%> of the calcium is replaced with monovalent cations.

1 . A process as claimed in any one ot claims 1 - 10 wherein the mineral depleted whey protein stream has an ash content of less than 2.5g per l OOg of solids.

12. A process as claimed in any one of claims 1 - 1 1 wherein the whey protein stream has been subjected to ultrafiltration, diafiltration or dialysis.

1 3. A process as claimed in an one of claims 1 - 12 wherein the calcium content of the low mineral whey protein stream is reduced in calcium ions to less than SOmg/ l OOg whey protein.

14. A process as claimed in any one of claims 1 - 1 3 wherein the whey protein stream, before heating, has a denaturation level less than 20%. 5. A process as claimed in any one of claims 1 - 1 3 wherein the proportions of low mineral whey protein and calcium depleted casein in the whey protein-rich stream are in the range 95: 5 to 45:55 by weight.

1 6. A process as claimed in any one of claims 1 - 1 5 wherein the heat treatments are in range from about 60°C for up to 30 minutes to about 140°C for at least 2 seconds.

1 7. A process as claimed in claim 16 wherein the heat treatments are at 80-90°C for 5 to 20 minutes.

1 8. A process as claimed in any one of claims 1 - 16 where the heat-treated casein rich stream and the heat-treated whey protein stream are mixed togedier to provide a yoghurt milk with casein to whey protein ratios by weight in the range 95: 5 to 30:70 by weight.

19. A process as claimed in claim 18 wherein the protein content of the yoghurt milk is between 1 .5% and 1 5% by weight.

20. A process as claimed in claim 19 wherein the overall protein content of the yoghurt milk is between 2% and 8% by weight.

21. A process as claimed in any one of claims 1 -20 wherein the acidifying is by the growth of food approved lactic acid producing bacteria.

22. A process as clauned in any one of claims 1 -21 wherein acidifying is by use of a slow release acidulant, for example glucono-delta-lactone (GDL).

23. A process as clauned in any one of clauns 1 -22 wherein the fat content of the yoghurt milk is 0.1 %- 15% by weight.