SOY-BASED BEVERAGE

The invention relates to a soy-based beverage precursor and beverage comprising a water- dispersible biopolymer with a Trouton ratio of at least 300, preferably 400, more preferably 450 measured as a 0.2 % w/w aqueous solution at 20 degrees Celsius.

Summary of the invention

The inventors have observed that the consumer perceived creaminess of a soy-based beverage is surprisingly significantly enhanced on addition of a selected water-dispersible biopolymer, in particular a water-dispersible biopolymer which has a Trouton ratio of at least 300, preferably 400, more preferably 450.

Thus in a first aspect of the invention, a soy-based beverage precursor is provided, the soy- based beverage precursor comprising:

(a) < 27, preferably < 25 % w/w soy protein; and

(b) an effective amount of a water-dispersible biopolymer,

wherein the water-dispersible biopolymer has a Trouton ratio of at least 300, preferably 400, more preferably 450 measured as a 0.2 % w/w aqueous solution at 20 degrees Celsius, and

wherein the soy-based beverage precursor excludes peanut.

The term "Trouton Ratio" means, for the purposes of the invention, the dimensionless number provided by the extensional viscosity divided by the apparent shear viscosity. The Trouton Ratio is as measured in a 0.2 % w/w aqueous solution of said ingredient at 20 degrees Celsius.

In a second aspect of the invention, a soy-based beverage is provided, the soy-based beverage in the form of an aqueous dispersion of the soy-based beverage precursor of the first aspect of the invention.

Detailed description of the invention

Preferably the water-dispersible biopolymer is obtained from the group consisting of okra, prickly pear (Opuntia ficus indica) and mekabu. The soy-based beverage precursor may additionally comprise a stabilizer. The stabilizer may be selected from the group consisting of high methoxy pectin, sodium carboxymethyl cellulose, propylene glycol alginate ester, water-soluble soybean polysaccharides, beet derived pectin, carrageenans, gellan gum and xanthan gum. The weight ratio of soy protein to stabilizer is preferably in the range 1 :1 to 30:1 , preferably 1 :1 to 25:1.

The soy-based beverage precursor is preferably in the form of a powder, granules or a liquid concentrate. The weight ratio of soy protein to water-dispersible biopolymer in the soy-based beverage precursor is preferably at least 3:1 , more preferably at least 5:1 , and most preferably at least 10:1 .

The weight ratio of soy protein to water-dispersible biopolymer in the soy-based beverage precursor biopolymer is preferably no more than 10000:1 , more preferably no more than 5000:1 , and most preferably no more than 1000:1

The soy-based beverage precursor preferably comprises 0.0005 to 0.5, more preferably 0.0025 to 0.5, and most preferably 0.025 to 0.25 % w/w water-dispersible biopolymer.

The soy-based beverage of the second aspect of the invention may be prepared from a one part soy beverage precursor of the first aspect of the invention by gently mixing water with the soy-based beverage precursor. Admixing conditions can influence the properties of the resulting soy-based beverage. Short mixing times and/or medium, preferably low, shear favour the formation of a soy-based beverage according to the invention. Mixing at high shear and/or for a prolonged time may lead to a soy-based beverage which does not show improved creaminess. Mixing at high temperature (≥ 90 degrees centigrade for a few minutes) will also inactivate the water-dispersible biopolymer. An alternative process for preparing the soy-based beverage of the second aspect of the invention is by starting with a soy-based beverage precursor of the first aspect of the invention wherein the soy protein and the water-dispersible biopolymer are separated into two separate components or parts. Water and the soy protein are homogenized and then the water-dispersible biopolymer is post-added under gentle shear. To ease dissolution, the water-dispersible biopolymer may be coated onto maltodextrin particles or granules.

However, a skilled person may easily determine the proper mixing conditions by measuring the Trouton ratio of the soy-based beverage. Thus, preferably the soy-based beverage has a Trouton ratio of at least 80, preferably at least 100, most preferably at least 120. The Trouton ratio of the soy-based beverage is the Trouton ratio as measured in a degassed sample of said soy-based beverage at 20 degrees Celsius. Preferably the Trouton ratio of the soy-based beverage of the second aspect of the invention is < 50 000, more preferably < 10 000, most preferably < 5 000.

Preferably the soy-based beverage comprises 0.01 to 10, preferably 0.01 to 5, most preferably 0.3 to 5 % w/w soy protein. Preferably the soy-based beverage comprises 0.0001 to 0.1 , preferably 0.0005 to 0.1 , most preferably 0.005 to 0.05 % w/w water-dispersible biopolymer.

As set out above, the soy-based beverage of the second aspect of the invention may be made by mixing the soy-based beverage precursor of the first aspect of the invention with water. As such, the weight ratio of soy protein to water-dispersible biopolymer is preferably the same in the soy-based beverage as it is in the soy-based beverage precursor.

The soy-based beverage optionally further comprises a fruit juice. Preferably the fruit juice is selected from the group consisting of orange, peach, mango, pineapple, apple, grape, lime, strawberry, pear, kiwi fruit, raspberry, mandarin, passion fruit, and tomato.

The soy-based beverage optionally comprises an animal-derived protein.

Preferably the soy-based beverage has a pH≥ 3.0, preferably≥ 3.5. For a soy-based beverage comprising a fruit juice, the pH is preferably in the range 3 to 5, more preferably 3.5 to 5. For soy-based beverage excluding a fruit juice, the pH is preferably in the range 5.5 to 8.5, more preferably 6 to 8.

Preferably the soy-based beverage has a Trouton ratio of at least 80, more preferably at least 100, most preferably at least 120. Example 1

The Trouton ratios of various compounds/extracts Materials

Okra fruit (Abelmoschus esculentus (L.) Moench)

Jew's mallow (Corchorus olitorius) leaves (Sonac Company, Alexandria, Egypt and The

United Company for Food Industry, Egypt)

Lime flowers (Just Ingredients Limited, UK)

Guar gum (Grindsted™ Guar 250, Danisco)

Locust bean gum (Grindsted™ LBG 246)

Tara gum (Solgum D21004/82, Foreign Domestic Chemicals Corporation)

Sodium carboxymethyl cellulose (FMC)

Xanthan gum (CP Kelco)

k-Carrageenan (Danisco A/S)

Flax seed gum (Shaanxi Fuheng (FH) Biotechnology Co. Ltd, China)

Sugar beet pectin (Pectin Betaspec RU 301 (Hernstreith & Fox KG)

Citrus pectin A (degree of esterification -40 % (GENU pectin LM-18 CG-Z (CP Kelco))

Citrus pectin B (degree of esterification -35 % (GENU pectin LM-12 CG-Z (CP Kelco))

Apple pectin powder (Solgar™ Vitamin and Herb, U.K)

OSA starch (octenyl succinic anhydride starch; National Starch).

SSPS (soluble soybean polysaccharides, obtained from SoyFIBE).

HM citrus pectin (high-methoxyl citrus pectin, JM-150, obtained from CP Kelco).

Gum Arabic (Super Gum EM10, San-Ei Gen FFI Incorporated)

Yellow mustard gum (extracted from yellow mustard bran obtained from G S Dunn) Prickly pear cactus (Opuntia ficus-indica) (opuntias.co.uk)

Mekabu (flowering sprout of Undaria pinnatifida) (Muso Limited)

Sodium alginate (Danisco A/S)

A degree of esterification below 50 % is considered low, and thus both citrus pectins A and B are considered to be low methoxy (LM) pectins.

Okra pectin was extracted from okra using the following method:

1. The okra was washed, the calyx removed and the remainder chopped roughly.

2. The chopped product was then blended with a double weight amount of 96 % w/w ethanol initially using a handheld blender and then a Silverson homogenizer. 3. The blend was then sieved through a 75 micron sieve and the filtrate discarded.

4. The solids were resuspended in a double weight amount of 96 % w/w ethanol and homogenised twice with a Silverson homogenizer.

5. The suspension was vacuum filtered through Miracloth (22-25 microns) and the filtrate discarded.

6. A suspension of 350 g of solid, 10 g NaCI and boiling water to a total volume of 5 litres was prepared.

7. The suspension was stirred with a paddle stirrer for at least 2 hours at 200 rpm.

8. The suspension was then centrifuged for 55 minutes at 4000 g and the supernatant decanted.

9. Ethanol was then added to the supernatant under hand stirring over 20 minutes to give a final concentration of 45 % w/w ethanol.

10. The mixture was left to stand and precipitate at room temperature for at least 1 hour.

1 1 . The suspension was filtered through 90 micron sieve and the filtrate discarded.

12. The precipitate (okra pectin) was rinsed using 96 % w/w ethanol and freeze dried.

Jew's mallow pectin and lime flower pectin was extracted from their respective leaves using the following method:

1. Blend the whole leaves in 2 times their weight of food-grade ethanol: first with a handheld blender (20-30 sees) then using a Silverson (large screen) for 5-10 minutes.

2. Remove the pulp from the ethanol by filtering through Miracloth using a vacuum pump.

3. Re-suspend the pulp using more food-grade ethanol and filter again.

4. Repeat washing in ethanol and filtering twice.

5. Take 350 g of the pulp, add 10 g of NaCI and 350 g boiling water and mix well using a spoon, Make up to 5 L with boiling water.

6. Mix continuously using a paddle stirrer for at least 2 hours.

7. Centrifuge the mixture for 55 minutes at 4000 g using a Sorvall RC-3C centrifuge.

8. Decant the supernatant into 2 x 5 litre beakers and precipitate the pectin by addition of food-grade 96 % ethanol up to approximately 47 % w/w.

9. Filter the mixture to remove the precipitate using a 70-90 μηη sieve.

10. Wash the precipitate using pure ethanol.

1 1 . Dry the precipitate under vacuum in a freeze drier for at least 24 hour. 12. Grind the precipitate to a fine powder using a grinder and store in a cool, dry place until required.

Yellow mustard gum was extracted from yellow mustard bran according to the following method:

1. Mix 1 kg of bran powder well with 2.5-3 times weight (food grade) ethanol and leave for 10-20 minutes.

2. Sieve (90 μηι).

3. Vacuum filter through Miracloth (single layer) discarding the ethanol and retaining the solid.

4. Wash twice with ethanol using vacuum filtration through Miracloth; mixing well between washes in order to remove pigments homogeneously.

5. Take 350 g of wet solid, add 10 g of NaCI, 350 g warm water and mix well.

6. Put under paddle stirrer and add boiling water to 4.8-5 litres total volume.

Ensure good mixing and stir gently for at least 2 hours.

7. Centrifuge for 55 minutes at 5000 g and decant supernatant into 2 x 5 litre beakers (discard solid).

8. Add supernatant in 400 ml aliquots to ethanol (90 % of the weight of the supernatant) and hand mix slowly using a gentle, folding action. Mix well and leave to precipitate at room temperature for at least 1 hour with occasional stirring.

9. Sieve (90-250 μηι).

10. Wash precipitate in ethanol.

1 1 . Dry precipitate in freeze drier.

12. Grind precipitate to a fine powder (<1 mm) using a grinder.

Methods

The extensional viscosity was determined using a commercially available instrument, which is a Capillary Break-up Extensional Rheometer (CaBER 1 from THERMO Electron Corporation) according to the following procedure. A liquid sample was placed at 20 °C between two 6-mm diameter parallel discs sitting 2 mm from each other. The upper disc was quickly pulled up and within 0.05 s it reaches 8 mm separation. A transient liquid bridge (i.e. a filament) was thus formed between the two plates, which ultimately breaks up upon separation of the discs. The instrument measured the diameter of the midpoint of the liquid filament formed between the two discs when fully separated and tracked its thinning until the break up point. The method has no means to control the rate at which the filament is thinning (i.e. the strain rate). This rate was determined by the balance of the capillary force, trying to shrink and break the liquid filament, and the opposing viscous force. The latter force was determined by the extensional viscosity which can vary as the contraction rate changes with time. The processing of the raw data and the calculation of the extensional viscosity was done using CaBER Analysis software (V 4.50 Built 29-Nov-04, designed by Cambridge Polymer Group, http://www.campoly.com). For the calculation of the Trouton ratio, the highest stable value of the extensional viscosity was used and the corresponding strain rate was recorded for later use to determine the corresponding shear viscosity value.

The CaBER Analysis software has a built-in function to select the usable range of data. It cuts off the data where the filament is too thick and its shrinkage is driven by the gravity and leaves the part where the shrinkage is due to the capillary force only. But in addition to this, the last data points were also removed where, after the break-up occurs, the retraction of the broken filament ends causes additional wavy features in the filament diameter curve.

Due to these instrumental limitations reliable values of the extensional viscosity were not obtained for all 0.2 wt. % solutions of ingredients in water, such as for some very thin and relatively inelastic solutions.

According to Jones et al (Rheologica Acta, 26, 20-30 (1987)), the Trouton ratio (TR) can be defined as the ratio between shear (η) and extensional viscosity (Γ)Ε) using the following equation, where έ is the strain rate

A high TR indicates a material with a high extensional viscosity or "stretchy" rheology. A material with a high extensional viscosity or "stretchy" rheology can also have a lower TR when the shear viscosity of the material is high. It is the maintenance of this "stretchy" rheology in the soy-based beverage of the second aspect of the invention which is believed important for sensory benefits.

The shear viscosity was measured using a Parallel-Plate geometry using either AR-2000 rheometer (from TA Instruments) or Physica MCR-501 (from Anton Paar). With the AR- 2000, a 40 mm diameter plate was used and with the MCR-501 a 50 mm diameter plate was used. The viscosity was measured at 20 °C for a range of shear rates between 1 and 1000 s-1.

0.2 % w/w aqueous solutions of the above-mentioned compounds/extracts were prepared and the Trouton ratio determined as described above.

Results

The extensional viscosities of 0.2 % w/w solutions of OSA starch, gum Arabic, SSPS, sodium carboxymethyl cellulose, xanthan gum and a range of commercially available pectins and other biopolymers could not be measured with the equipment available due to very short filament lifetime. In order to obtain a Trouton ratio for these compounds more concentrated solutions were prepared until a reliable measurement could be made. It is assumed that the Trouton ratio of 0.2 % w/w aqueous solutions of these compounds will be lower or at most equal to the Trouton ratio obtained at higher concentrations. The results are summarised in Table 1 .

Table 1 : Trouton ratio of various compounds/extracts as aqueous solutions at various % concentrations.

Concentration (wt. %) Trouton ratio

OSA starch 20 13.9

Gum Arabic 20 4.9

SSPS 20 8.1

Sugar beet pectin 4 3.6

Sodium alginate 2 3.9

Sodium carboxymethyl cellulose 2 35.0

HM citrus pectin 2 6.8

Citrus pectin A 2 3.3

Citrus pectin B 2 3.5

Apple pectin 2 3.2

Xanthan gum 1 12.7

Locust bean gum 1 29.5

Guar gum 1 13.3

k-Carrageenan 1 29.8

Tara gum 1 5.2

Okra pectin 0.2 572.1-950.1

Jews mallow pectin 0.2 250.9

Lime flower pectin 0.2 256.6 Yellow mustard gum 0.2 236

Flax seed gum 0.2 88

Prickly pear cactus (juice) ¹ 0.2 1569

Mekabu (aqueous extract) ² 0.2 660

1 Obtained by squeezing pulp through muslin.

2 Dried Mekabu was extracted by soaking in water (20 g in 100 ml) for 24 hours and squeezing through muslin. Conclusions

The Trouton ratios of 0.2 % w/w aqueous solutions of okra pectin, prickly pear cactus (aqueous extract) and Mekabu (aqueous extract) were found to be at least 300. The Trouton ratios of okra pectin, prickly pear cactus (aqueous extract) and Mekabu (aqueous extract) are clearly far above those of other commonly used plant extracts used in foods.

Example 2

Sensorial impact (creaminess) of high Trouton ratio ingredient on soy milk

Sensory evaluation of soy milk comprising a range of test ingredients was conducted by a panel of 15 trained sensory assessors. In particular the effect on creaminess was evaluated.

The soy milk chosen was an unflavoured soy milk (Ades). 40 ml of the soy milk was mixed with the test ingredient prepared as a 10 ml solution (or a 10 ml water blank for the control) and evaluated at room temperature.

An Absolute Scaling Method was used as it can measure objectively the intensity of sensory attributes in foods, perceived by a panel of trained sensory assessors. The reference used for the intensity scale of this method was a range of concentrations of citric acid. With this method, attributes values are absolute. For instance an attribute score of 5 is half as intense as an attribute score of 10. This method provides highly reproducible data over time and assessors. According to this reference scale, panellists have to rate the intensity of each attribute on a 16-point scale (0 - 15) (thus for citric acid concentrations of 0.2, 0.4, 0.6 and 0.8 g per litre, the values on the 16 point scale were 2, 5, 8 and 1 1 respectively). Definitions of attributes and anchored points for control were available for the panelists during the assessment.

The impact on sensorial character (creaminess) of the soy milk was evaluated with: • 0.02 % w/v Jews mallow pectin

• 0.02 % w/v lime flower pectin

• 0.02 % w/v guar gum

• 0.02 % w/v okra pectin

The corresponding Trouton ratios for each beverage were also determined using the method described in Example 1 with the exception that the Physica MCR-501 (from Anton Paar) was used with a 25 mm diameter plate and shear rates up to 10 000 s ^"1 were measured. As the low levels of test ingredient led to corresponding low shear viscosities for the soy milk, reliable measurement for Trouton ratios were only obtained for the soy milk comprising okra pectin. For the rest of the beverages, only upper limits are provided in Table 2.

Results

The results are summarized in Table 1 as a positive or negative percentage change from the control without a test ingredient. Statistical analysis was performed using Student's T- test (P < 0.05) and letters assigned wherein common letters between two values means there is no significant difference. Table 2: Change in perceived creaminess (attribute cream as mean of 30 values) of soy milk comprising 0.02 % w/v one of Jews mallow pectin, lime flower pectin, guar gum and okra pectin. Student's T-test (P < 0.05) with letters assigned wherein common letters between two values means there is no significant difference. Trouton ratios for each beverage are also included.

Unflavoured soy milk Unflavoured soy milk

Test ingredient Cream (%) Trouton ratio

Control 0 (a) < 60.9

Jews mallow pectin +2 (a) < 62.0

Lime flower pectin -2 (a) < 77.4

Guar gum -5 (a) < 79.3

Okra pectin +18 (b) 132.2 The results clearly show that not only does okra pectin surprisingly increase the creaminess of unflavoured soy milk, but is significantly better than any of Jew's mallow pectin, lime flower pectin, or guar gum. Conclusions

Okra pectin surprisingly increase perceived creaminess in soy milk and furthermore shows surprising efficacy at increasing creaminess in unflavoured soy milk compared to other test ingredients (Jew's mallow pectin, lime flower pectin, and guar gum).