Process for producing thermally modified starch blends

Description

[0001] The invention relates to a process for producing a mixture of at least two thermally modified starches, in which the starches are granular starches of distinct botanical origins, this process consisting of mixing under acidic conditions said starches of distinct botanical origins before carrying out the heat treatment itself.

[0002] Acidic conditions are understood to mean a treatment of the mixture of at least two starches of distinct botanical nature, in the milk phase, with a citrate buffer of pH between 4 and 6.

[0003] More particularly, the invention relates to a method for producing a mixture of thermally modified potato starch and waxy corn starch.

[0004] Such mixtures of at least two thermally modified starches make it possible to strengthen their viscosity properties while retaining the texturizing properties expressed by thermally modified starches prepared from a starch from a single botanical origin.

[0005] Such thermally modified starches then find use as thickening and texturizing agents in many food applications, primarily in soups and sauces, and in dairy products. Background to the invention

[0006] Synthesized biochemically, a source of carbohydrates, starch is one of the most widespread organic materials in the plant world, where it constitutes the nutritional reserve of organisms.

[0007] Starches have always been used in the food industry, not only as a nutritional ingredient but also for their technological properties, as a thickening, binding, stabilizing or gelling agent.

[0008] For example, native starches are used in preparations requiring cooking. Corn starch, in particular, is the basis of "custard powders."

[0009] As it is rich in amylose, it retrogrades and therefore gels strongly. It allows for firm flans to be obtained after cooking and cooling.

[0010] It is also suitable for pastry creams.

[0011] But these cannot be used in pastries intended for freezing because, when defrosting, the phenomenon of syneresis, which results in the expulsion of water, destroys the texture of the cream.

[0012] Thus, in its native state, starch is of limited application due to syneresis, but also due to: its low resistance to shear stresses and heat treatments, its high retrogradation, its low processability and its low solubility in common organic solvents.

[0013] Therefore, to meet today's demanding technological needs, the properties of starch must be optimized by various so-called "modification" methods.

[0014] These main modifications then aim to adapt the starch to the technological constraints resulting from cooking, but also from freezing / thawing, canning or sterilization, and to make it compatible with modern food (microwaves, instant preparations, "high temperatures"..).

[0015] Starch modification then aims to correct one or some of the defects mentioned above, thus improving its versatility and satisfying consumer demand.

[0016] Starch modification techniques have been broadly classified into four categories: physical, chemical, enzymatic and genetic, the aim being to produce various derivatives with optimized physicochemical properties.

[0017] Chemical and physical modifications are the most commonly implemented.

[0018] Chemical processing involves introducing functional groups into starch, which significantly alters its physicochemical properties. Such modifications of granular native starches profoundly alter gelatinization, sticking, and retrogradation behavior.

[0019] Typically, these modifications are achieved by chemical derivatization, such as esterification, etherification, crosslinking, or grafting.

[0020] However, chemical modifications are less sought after by consumers in food applications (also for environmental reasons), even though some modifications are considered safe.

[0021] Various physical modifications are therefore proposed, for example: Heat Moisture Treatment (HMT), which involves treating starch at controlled humidity levels (22-27%) and at high temperatures for 16 hours to alter the structure and physicochemical properties of the starch; Annealing, which involves treating starch with excess water at temperatures below the gelatinization temperature to approach the glass transition temperature; High Pressure Processing (HPP), which involves hydrating the amorphous regions of the starch granule, leading to distortion of the crystalline portions of the granule and promoting the accessibility of said crystalline regions to water; glow discharge plasma treatment, which generates high-energy electrons and other highly active species at room temperature.When applied to starch, these active species excite the chemical groups of the starch and cause significant crosslinking of the macromolecules; osmotic pressure treatment (OPT), carried out in the presence of solutions with a high salt content. The starch is suspended in sodium sulfate to produce a uniform suspension. The starch changes from type B to type A after treatment, thus acquiring a gelatinization temperature that increases significantly; treatment by "thermal inhibition". In general, thermal inhibition means the dehydration of a starch until it reaches an anhydrous or substantially anhydrous state (i.e. < 1% humidity), then a heat treatment at more than 100°C for a period of time sufficient to "inhibit" the starch, in this case to give it crosslinked starch properties. It is also necessary to place the starch in pH conditions at least neutral to preferably alkaline before proceeding with the advanced dehydration step. An alternative treatment by "thermal inhibition" has been proposed in the solvent phase, which consists of heating a non-pregelatinized granular starch in an alcoholic medium, in the presence of a base and salts, at a temperature of 120° to 200°C, for 5 minutes to 2 hours.

[0022] In any case, the thermal inhibition process then leads to obtaining a starch paste with properties of increased resistance to viscosity breakage, and a non-cohesive texture.

[0023] The technical field to which the invention relates is that of the treatment by thermal inhibition of starch, without hydro-alcoholic solvent.

[0024] In this particular technical field, we can cite more particularly US patent 6,221,420 which describes a thermally inhibited starch, obtained by dehydration then heat treatment.

[0025] The main steps are: the dehydration of the starch to a water content of less than 1% carried out at a temperature between 100 and 125°C, then the heat treatment of the dry starch thus obtained, at approximately 140°C, in a fluidized reaction bed, for a period of around 20 hours.

[0026] Preferably, before the starch dehydration step, it is recommended to carry out a starch alkalization step, making it possible to bring the pH of the starch suspension to a value between 7 and 10, preferably between 8 and 10.

[0027] At this stage, before the actual dehydration step which precedes the inhibition step, the water content of the starch (as exemplified) is then between 8 and 10%.

[0028] US patent application 2001 / 0017133 describes a similar process, in which the starch is also dehydrated below 125°C before the inhibition process is started (at a temperature of more than 100°C, preferably between 120 and 180°C, more preferably between 140 and 160°C) for a period of up to 20 hours, preferably between 3 and 4 hours 30 minutes.

[0029] Before the dehydration step, the conventional alkalization step results in a starch suspension having a pH value of between 7.5 and 11.2, preferably between 8 and 9.5%, and a water content of between 2 and 15%.

[0030] A variant has been proposed in patent application WO 2014 / 042537, which variant relates to heating an alkaline starch to temperatures between 140 and 190°C, ensuring that the inhibition process is initiated and carried out in the presence of sufficient water, i.e. more than 1% water.

[0031] In other words, this process recommends thermally inhibiting a previously alkalized starch without carrying out a dehydration step.

[0032] The starch preparation or starch is thus brought to a pH between 9.1 and 11.2, preferably to a value of around 10, and the humidity is adjusted between 2 and 22%, preferably between 5 and 10%.

[0033] Thermal inhibition is then carried out directly on this powder or starch, at a temperature between 140 and 190°C, preferably between 140 and 180°C, for a period of 30 minutes.

[0034] Another variant has been proposed, by lowering the pH of the treated starch before thermal inhibition. Indeed, a major drawback is that high pH levels tend to increase the browning of the starch during the heating step.

[0035] Thus, in patent application WO 2020 / 139997, a step of preparing starch is described which successively consists of: prepare a starch milk, bring the starch milk thus prepared to a pH between 5 and 6.5, by neutralization if necessary, add an acid buffer, and adjust the pH.

[0036] Only then does the actual heat treatment step take place, comprising a dehydration step and a thermal inhibition step.

[0037] In this patent application, it is essential to first bring the starch milk to a pH value between 5 and 6.5. Either the starch already has this pH value, considered to be its natural pH value, or a "neutralization" step (by adding acid or base) is carried out to reach this pH range. This step can take up to 24 hours.

[0038] Then, a citrate or citric acid pH buffer is added to bring the pH of the starch milk to a pH value between 4 and 6. This step can take up to 24 hours.

[0039] Finally, to produce a thermally inhibited starch, this patent application teaches the dehydration and thermal inhibition of the starch milk thus treated. This step can take up to 4 hours of reaction.

[0040] It has finally been established that this process makes it possible to obtain whiter thermally inhibited starches with an improved taste compared to thermally inhibited starches prepared by conventional means, i.e. after alkaline impregnation.

[0041] Nevertheless, this method of preparing thermally inhibited starches is perfectible. Indeed, it has the disadvantage, from the point of view of its implemented process, of having to carry out the impregnation of the starch milk in an acid medium only after having finely controlled the initial pH of the starch.

[0042] Then, a treatment of this type leads to the production of thermally inhibited starches with a high fixed citrate level, of the order of more than 0.2%, reflecting a significant degree of functionalization.

[0043] In the state of the art, other alternatives are presented to improve the thermal inhibition rate of starch, in order to improve its technological properties.

[0044] Thus, in patent EP 1,102,792 it is proposed to boost starch in the presence of oligosaccharides containing 1 to 20 sugar units.

[0045] Under certain implementation conditions, the mixture of starch and oligosaccharides then develops better stability during cold storage.

[0046] However, for some food applications, heavy purification steps must be added to remove oligosaccharides from starch after heat treatment.

[0047] In patent EP 2,251,358, it is preferred to use a powdery mixture of starch and water-soluble hemicellulose, in particular proportions.

[0048] Heat treatment consists of moist heat treatment at a temperature of 100 to 200°C.

[0049] It is also recommended to incorporate sodium carbonate (or similar alkaline compounds) with hemicellulose into tapioca starch, glutinous rice or waxy maize.

[0050] However, what is most sought is a method of modifying starch in such a way that swelling and / or disintegration (e.g., rupture) of starch granules are effectively suppressed without any chemical treatment.

[0051] According to this patent, the objective is not to improve the technological properties of the starch thus thermally modified, but to prevent the gelatinization of the starch for the production of pastry creams.

[0052] There therefore remains a need for a new process for preparing thermally modified starches which gives them improved viscosity properties, while retaining their excellent resistance properties, or even making it possible to improve the coloring of the products obtained.

[0053] And the Applicant company found that this need could be met by proposing a process for the heat treatment of a mixture of at least two granular starches of distinct botanical origins, under acidic conditions. Description of the invention

[0054] According to the invention, the process for preparing a mixture of at least two thermally modified starches, in which the starches are granular starches of distinct botanical origins, comprises the steps of: (i) Prepare a starch milk containing at least two starches of distinct botanical origins, having a total dry matter of between 30 and 40%, preferably between 35 and 37% by weight, (ii) Prepare a citrate buffer with a pH of between 4 and 6, preferably 6, and add said citrate buffer to the starch milk so as to obtain a milk pH of between 4 and 6, preferably 6, (iii) Ensure a contact time of between 0.5 and 5 hours, (iv) Dry to an equilibrium humidity of between 10 and 12%, (v) Heat to a temperature of more than 150°C, for a residence time of between 10 minutes and 6 hours, (vi) Resuspend, rectify the pH, optionally wash and dry again the thermally modified starches thus obtained.

[0055] By choosing to carry out the mixing of starches of distinct botanical origins under acidic conditions before the actual heat treatment, the Applicant company is going against the teachings of the state of the art.

[0056] It is indeed known to make mixtures of starches to optimize their functional properties, but this is achieved by mixing thermally modified starch varieties with native starches as described in international application WO 2020 / 018061.

[0057] Furthermore, thermal modification of a mixture of starches of different botanical origins ensures the same level of modification of the two (or more) components of the mixture.

[0058] To the knowledge of the Applicant company, no heat treatment processes have ever been described on a preliminary mixture of at least two starches of distinct botanical origins, under acidic conditions.

[0059] The starches to be used in the process of the invention may be of any origin, for example corn, waxy corn, amylocorn, wheat, waxy wheat, pea, field bean, potato, waxy potato, tapioca, waxy tapioca, rice, konjac, etc.

[0060] Preferably, we will choose to mix potato starch with corn starch, more particularly waxy corn starch (high in amylopectin).

[0061] According to a particular embodiment, the method according to the invention may relate to the preparation of a mixture of two thermally modified starches, in which the starches are granular starches of distinct botanical origin, present in the mixture in equal quantities, thus forming a 50 / 50 mixture by mass.

[0062] The process according to the invention requires the preparation of a milk of starches of at least two distinct botanical origins with a total dry matter of between 30 and 40%, preferably between 35 and 37% by weight. As will be exemplified below, the dry matter is set at 36.5% by weight.

[0063] The next step is then to bring the pH of the mixed starch milk to a pH between 4 and 6, preferably 6.

[0064] To do this, a citrate buffer must be prepared at the pH level and added to the starch milk so as to obtain a milk pH of between 4 and 6, preferably 6.

[0065] A contact time of between 0.5 and 5 hours is then ensured, preferably around 0.5 hours.

[0066] After filtration, the preparation is dried to an equilibrium starch humidity of between 10 and 12%.

[0067] Drying can be done at 60°C in a Retsch ® laboratory dryer, but also in a hood with natural ventilation at room temperature, or in a pilot / industrial dryer, at a temperature of over 100°C.

[0068] After drying at 60°C then grinding in Thermomix ®< in order to break up the starches and prevent the formation of aggregates, the heat treatment itself is carried out.

[0069] As will be exemplified below, the heat treatment is carried out at a temperature of 170°C, for a reaction time ranging from 0.5 to 2 hours. This kinetics makes it possible to vary the degree of functionalization of the thermally modified starches thus prepared (the more functionalized the starches are, the more resistant they are to restrictive conditions of use, i.e. acid pH, high shear, high temperature).

[0070] But these conditions are entirely transposable to implementation in a continuous turbodryer device or in a reaction fluidized bed.

[0071] In a first embodiment of the method according to the invention, the following heat treatment step itself can be carried out in heat treatment devices combining heat exchanges by conduction and by convection, a turbodryer type device, for example at least one continuous turbodryer of the VOMM type, which thus makes it possible, depending on the size of said VOMM, to achieve a very short reaction time, of the order of a few minutes, or less than 5 minutes per heat treatment stage.

[0072] The temperature settings are then set at values ​​of more than 190°C, preferably between 200 and 210°C, for a residence time of between 10 and 60 minutes, more preferably between 15 and 35 minutes.

[0073] Delta T, defined as the temperature difference between the set temperature and the product temperature at the reactor outlet, is between 17 and 27°C.

[0074] In a second embodiment of the method according to the invention, the heat treatment itself can be carried out in “reaction fluidized bed” type devices.

[0075] As is known to those skilled in the art, this device consists of a reactor which allows a divided solid to be suspended using a gas, in this case an air / nitrogen mixture. The speed of the gas is adjusted according to the raw material.

[0076] The heat treatment temperature (product temperature) is set at more than 130°C, preferably between 130 and 200°C with a reaction time varying between 30 min and 6 h, preferably between 2 and 4 hours.

[0077] According to a particular embodiment, the present invention relates to a mixture of at least two thermally modified starches of distinct botanical origin produced by the process as defined above, characterized in that it has a free citrate level of less than 0.05% and a fixed citrate level of between 0.05 and 0.15%.

[0078] According to a particular embodiment, the present invention relates to a mixture of at least two thermally modified starches as defined above, characterized in that it has a coloration expressed in L* value of between 80 and 90, and in Yie value of between 15 and 20.

[0079] The thermally modified starches according to the invention will be advantageously used, depending on their respective properties, as a thickening agent or texturizing agent in food applications, in particular in soups, sauces and in dairy products.

[0080] One of the main limitations of thermally modified starches made from a single botanical source is a slightly lower developed viscosity than commercially available chemically modified starches, requiring an overdose when replacing them in a 100% natural solution with "the least possible chemical transformation" (in English "clean label").

[0081] These mixtures therefore represent a simple and effective solution for industries manufacturing sauces, for example.

[0082] More specifically, these mixtures would be adapted to the technological requirements of pasteurization, medium shear and acid pH.

[0083] The invention will be better understood with the aid of the following examples, which are intended to be illustrative and not limiting. Examples Materials and methods

[0084] Below are the two products used for the pH 6 buffer. NAME CASE Brut Formula Mw g / mol Sodium citrate tribasic dihydrate 6132-04-3 C 6 H 5 Na 3 O 7- 2H 2 O 294,1 Citric acid, Anhydrous 77-92-9 C 6 H 8 O 7 192,12 Conductivity and pH measurement Conductivity

[0085] The method implemented here is adapted from the European Pharmacopoeia - official edition in force - Conductivity (§ 2.2.38). Materials:

[0086] KNICK 703 electronic conductivity meter also equipped with its measuring cell and checked according to the operating procedure described in the relevant instruction manual. Operating mode:

[0087] A solution is prepared containing 20 g of sample in powder form and 80 g of distilled water with a resistivity greater than 500,000 ohms.cm.

[0088] The measurement is carried out at 20°C using the conductivity meter, referring to the operating procedure indicated in the device's user manual.

[0089] Values ​​are expressed in microSiemens / cm (µS / cm) or milliSiemens / cm (mS / cm). pH

[0090] The method implemented here is adapted from the European Pharmacopoeia - official edition in force - pH (§ 2.2.3).

[0091] A 20% (W / W) suspension of the sample to be analyzed is prepared and the pH value is determined using a laboratory pH meter, referring to the procedure indicated in the device's user manual. The pH is expressed to the nearest 0.01 unit. Measurement of citrates by HPLC

[0092] The dosage of citrates is carried out: directly after heat treatment, on washed and dried products, (measurement of free citrates) then, after hydrolysis (measurement of fixed citrates) according to the following methods. Dosage method

[0093] After separation by ion exchange chromatography, the citrate ion is detected by conductimetry. Quantification is done by the internal calibration method.

[0094] The equipment used is a high-performance liquid chromatography set consisting of: A Thermofisher ICS 2100 system; A sampler to maintain samples at 10°C (TSP AS-AP type); An AERS 500-ultra suppressor; A Thermo AS11 HC 250*4 mm column with AG11-HC 4*50 mm pre-column. Filters for ion chromatography - IC 0.45 µm - Pall.

[0095] The following reagents are used: HPLC Grade Water Sigma Sodium Citrate Tribasic Dihydrate Sigma Trifluoroacetic Acid Internal Standard Solution: 400 mg / L Trifluoroacetic Acid Solution Merck 2N Hydrochloric Acid Thermo EGC-KOH Cartridge

[0096] The operating procedure is as follows: - Solvent A: HPLC quality water, - Elution program: Time (min) Flow rate (mL / min) KOH 0 1,3 1,5 15 1,3 1,5 28 1,3 1,2 48 1,3 28 60 1,3 40 60,1 1,3 60 67 1,3 60 67,1 1,3 1,5 77 1,3 1,5 - Injected volume: 25 µL - Column temperature: 36°C - Analysis time: 77 min - Sample temperature: 10°C - ASRS: 193 mA Calibration

[0097] Prepare 6 curve points. Weigh x mg of sodium citrate, i.e. x between 10 and 250 mg, and adjust to 500 mL of water. Shake. Take 0.5 mL, add 0.5 mL of internal standard, and make up to 20 mL with 1 mM sodium hydroxide. Filter and inject. Calculate the weight of the chloride standards (weight of sodium chloride (Mw of ion / Mw of salt)).

[0098] Plot the calibration curve: ratio of peak heights (= weight of chloride standard / weight of internal standard). Sample

[0099] Weigh 100 mg of sample + 0.5 mL of internal standard into 20 mL of water. Filter. Inject. Washing and drying

[0100] In a 250 ml beaker, approximately 20 g of the sample to be analyzed and 200 ml of demineralized water are introduced. Cover with a watch glass and stir for 20 minutes using a magnetic stirrer. Then filter through a Buchner funnel with a diameter of 150 mm equipped with a Durieux filter no. 111 white band, with a diameter of 150 mm or equivalent.

[0101] The filtrate is returned to a 250 ml beaker, dispersed in 200 ml of demineralized water and stirred for 20 minutes. Filter again and rinse with 200 ml of demineralized water.

[0102] The filtered product is dried in a laboratory oven overnight, then ground to avoid lumps. Hydrolysis

[0103] In the 250 ml flask with a flat bottom and a ground neck, a test portion "P" is introduced, exactly weighed from the sample to be analyzed. A quantity of distilled water equal to (100 - P), 100 ml of 2N hydrochloric acid and some boiling regulators (granular pumice stone) are added. Place in the electric heating mantle under reflux condenser and leave for 45 minutes from boiling. Cool then neutralize with the 40% sodium hydroxide solution to pH 7. Expression of results

[0104] The citrate ion content in % is determined by the following equation: Q / P × 100 Or : Q = quantity of citrate read on the curve (mg) P = weight of the sample in mg.

[0105] Measuring the viscosity of a starch suspension using the Rapid Viscosimeter Analyzer (RVA)

[0106] This measurement is carried out under specific concentration conditions and following an adapted temperature / time analysis profile.

[0107] Two buffer solutions are prepared: Stamp A

[0108] In a 1 liter beaker, containing 500 ml of demineralized water, add 91.0 g of citric acid monohydrate (purity > 99.5%) and homogenize, 33.0 g of sodium chloride (purity > 99.5%), and homogenize until completely dissolved, 300.0 g of 1N sodium hydroxide.

[0109] Transfer to a 1 L graduated flask and fill with demineralized water to 1 L. Buffer B

[0110] 100 g of Buffer A is mixed with 334.0 g of demineralized water.

[0111] The product to be analyzed is prepared as follows: A mass of 1.37 g of the dry product to be analyzed thus obtained is introduced directly into the bowl of the viscometer, and Buffer B solution is introduced until a mass equal to 28.00 ± 0.01 g is obtained. Homogenization is carried out using the stirring blade of the Rapid Visco Analyzer (RVA-NewPort Scientific).

[0112] The time / temperature and speed analysis profile in the RVA is then carried out as follows: [Table 1] Time hh:mm:ss Temperature °C Rotation speed Revolutions / min (RPM) 00:00:00 50 100 00:00:10 50 500 00:00:20 50 960 00:00:30 50 160 00:01:00 50 160 00:05:00 92 160 00:17:00 92 160 00:20:00 50 160 End of test: 00:20:05 (hh:mm:ss) Initial temperature: 50°C ± 0.5°C Data acquisition interval: 2 seconds Sensitivity: low

[0113] The measurement results are given in RVU (unit used to express the viscosity obtained on the RVA), knowing that 1 RVU unit = 12 cPoises (cP).

[0114] As a reminder, 1 cP = 1 mPa.s.

[0115] The results will therefore be presented in mPa.s.

[0116] Viscosity measurements will be carried out “at peak”, i.e. maximum viscosity value between 4 and 6 minutes, and “at drop”, i.e. the difference between the viscosity value at peak and that measured at 17 minutes. Sedimentation test

[0117] The product is dispersed in an aqueous medium and the decanted volume is measured. Solution A

[0118] Zinc chloride: 10 g Ammonium chloride: 26 g Distilled water: qsp 100 ml

[0119] In a 250 ml container, place a 1.0 g anhydrous sample of the product to be analyzed. Add 100 ml of solution A, stopper, homogenize, and place in a water bath for 10 minutes. Cool in a cold water bath, homogenize again, transfer to a 100 ml test tube, and measure the decanted volume after 24 hours.

[0120] The decanted volume, expressed in ml, is given by the following formula: volume dé canté d ' amidon / volume total × 100

[0121] It is then considered that a decantation volume: more than 70 ml means a product that is poorly functional / poorly resistant, between 60 and 70 ml: moderately functional / moderately resistant, between 40 and 60 ml: highly functional / resistant less than 40 ml: very highly functional or very resistant Measuring the color

[0122] Colorimetric measurement is based on the opposite color theory which specifies that the responses of cones (cells in the retina of the human eye responsible for color vision) to red, green and blue colors are recombined into opposite "black-white", "red-green" and "yellow-blue" signals when transmitted to the brain by the optic nerve.

[0123] This measurement is based on the color scales widely used in the food and polymer industries, called CIELAB L*, a*, b* scales.

[0124] The L*, a* and b* type scales are defined as follows: “L*” axis (lightness): 0 corresponds to black, 100 corresponds to white “a*” axis (red-green): positive values ​​are attributed to red; negative values ​​are attributed to green; 0 is neutrality “b*” axis (yellow-blue): positive values ​​are attributed to yellow; negative values ​​are attributed to blue; 0 is neutrality.

[0125] The "L*" index therefore has a value between 0 and 100, while the "b*" and "a*" indices have no numerical limitations. The measuring device is typically a Colorflex ®< EZ spectrocolorimeter, following the manufacturer's specifications (Version of the manual 1.2 of August 2013 for CFEZ firmware version 1.07 and higher - see pages 17 and 38).

[0126] The measurement is carried out in a 64 mm glass sample cup into which the sample is introduced so as to half fill the glass cup to have enough material to cover the surface in contact with the rays (for measurement homogeneity).

[0127] The Yie (Yellowness Index) is a number calculated from spectrophotometric data that describes the color change of a tested sample from “clear or white” to “yellow,” well known to those skilled in the art. Example 1 : Obtaining the mixture of thermally modified starches in a ventilated oven Preparation of the milk phase base:

[0128] The 50 / 50 mixture of potato starch + WAXILYS ® corn starch is suspended in demineralized water to a dry matter content of 36.5%, then the citrate buffer is added to obtain a milk pH of 6.

[0129] We wait for stabilization for at least 30 minutes and measure the pH and conductivity of the suspension.

[0130] We filter on sinter of porosity 3 then dry on a RETSCH ® type dryer at 60°C to around 10-12% humidity.

[0131] We grind on Thermomix ®< in order to break up the starch and prevent the formation of aggregates. Thermal treatment

[0132] The reaction is carried out at 170°C in an aluminum dish in a ventilated laboratory oven type MEMMERT ™< .

[0133] Weigh 35 to 40 g of prepared base per aluminum cup (type for METTLER TOLEDO ®< LJ16 Moisture measuring balance).

[0134] Place the cups in the MEMMERT ™ oven, previously set to 170°C.

[0135] Then samples are taken at different times for reaction kinetics analysis. Measurements: Humidity, pH, conductivity, color (L * and Yie)

[0136] Table 1. at 20% MS REF PE Humidity pH Driver. COLORING MEASUREMENTS on COLORFLEX HunterLab HOURS % µS L* Yi 043-- PRIMIMPREGNATED BASE 0 10,67 6,5 346 97 8 043 A 0,5 0,09 6,4 335 96 9 043 B 0,75 0,01 6,3 327 94 11 043 C 1 0,01 6,3 324 94 12 043 D 1,5 0,01 6,2 321 92 17 043 E 2 0,01 6,0 316 91 20

[0137] Resuspension of products.

[0138] After the reaction at 170°C, the sample is resuspended between 30 and 36% DM in demineralized water.

[0139] The pH is adjusted if necessary between 5.5 and 6.5 with NaOH.

[0140] We filter on a frit of porosity 2.

[0141] The cake is dried under a ventilated hood overnight at room temperature and ground (IKA ®< ) to break up the lumps and make the sample homogeneous.

[0142] Controls: Humidity, pH, conductivity, colorations, RVA, sedimentation test. Table 2 MEASUREMENTS AFTER pH RECTIFICATION, FILTRATION, DRYING AND GRINDING at 20% MS WASHING PRODUCT PE Humidity pH Driver. COLORING MEASUREMENTS on COLORFLEX HunterLab HOURS % µS L* Yi 043 AL 0,5 16,0 6,1 235 97 7 043 BL 0,75 14,7 6,1 207 95 10 043 CL 1 14,7 5,6 193 95 10 043 DL 1,5 15,0 5,7 173 94 14 043 EL 2 13,5 5,8 173 93 18

[0143] After washing, the products are obtained with pH values ​​between 5.5 and 6.5 and conductivity values ​​between 170 µS and 235 µS. The Yie (yellow) colors are between 7 and 18. By comparison, the native waxy corn base has a Yie of 8.11. Results

[0144] RVA viscosity measurements.

[0145] The results are presented in Table 3 and the Fig. 1 . Table 3 MCL 106H: 1.37 g anh. QSP 28g buffer B -- in cPs REF PE Hours V0 peak or max between 4 and 6 min ① Visco V 92°C +12' ② fall = ①-② MAX visco between 4 and 10' V 10' at 50°C Native 50 / 50 / / / 8,9 1145 187 958 1145 258 PdT / Waxy 043-- PRIMIMPREGNATED BASE 0 7,8 1066 203 863 1066 278 043 AL 0,5 12,0 648 430 218 651 587 043 BL 0,75 6,6 574 525 49 607 735 043 CL 1 7,4 562 576 -14 625 830 043 DL 1,5 7,1 394 563 -169 529 841 043 EL 2 3,2 183 357 -174 287 539

[0146] Each mixture of thermally modified starches according to the invention has improved stability during the use process compared to native starch: less viscosity gain and retrogradation phenomena are observed when using these starches. On this point, it is observed that the more the RVA drop tends towards 0 or becomes negative, the more the product will be functionalized, therefore the more resistant it will be and the less retrogradation it will express. Sedimentation test

[0147] Table 4 REF REACTION TIME HOURS DECANTATION VOLUME ml 043 - PRIMIMPREGNATED BASE 0 100 043 AL 0,5 97 043 BL 0,75 78 043 CL 1 73 043 DL 1,5 53 043 EL 2 43

[0148] After 2 hours of reaction, we therefore obtain a highly functional product. Example 2: obtaining a WAXILYS ® corn starch thermally modified in a ventilated oven

[0149] The aim of this experiment is to prepare a thermally modified starch according to the same process as Example 1, but from a single source of starch, in this case waxy corn starch, in order to then compare it with the product manufactured from the mixture of waxy corn starch and potato starch according to the invention. Preparation of the base in the milk phase

[0150] In a 1000 ml beaker, the waxy corn starch (WAXILYS ®< as marketed by the Applicant company) is introduced, then suspended in demineralized water for a total dry matter (DM) of 36.5% by weight. Sodium citrate and citric acid are added directly to the starch milk powder to obtain a pH of 4. Stabilization is allowed for at least 30 minutes and the pH and conductivity of the suspension are measured. Filter through a porosity 3 frit. Dry on a RETSCH ®< type dryer at 60°C to an equilibrium humidity of around 13%. Grind on a Thermomix ®< grinder to break up the starch and prevent the formation of aggregates. Thermal treatment

[0151] Weigh 40 g of sample and place it in an aluminum dish for METTLER TOLEDO ®< LJ16 balance (Humidity measuring balance). Place the dish in the MEMMERT ™< ventilated oven preheated to 170°C. Start the stopwatch as soon as the dishes have been introduced, and remove them according to the chosen reaction kinetics.

[0152] Washing and pH adjustment of the products after reaction by resuspension Resuspension of the sample at 36% DM in demineralized water, rectification of the pH between 4.5 and 6.5 by NaOH, filtration on a porosity 3 frit, drying of the cake under a ventilated hood overnight at room temperature. Grinding (IKA ®< ) in order to break up the sample and make it powder homogeneous. Measurements: Humidity, pH, conductivity, color (L* and Yie), RVA 4500 viscosity on washed products and sedimentation test. Results

[0153] Physicochemical measurements carried out on the final products after washing: humidity, pH, conductivity in µS, coloration. The results are presented in Table 5 below. Table 5 Tests Thermal treatment Humidity At 20% DM Colorflex ® color measurements< Hunter Lab pH Conductivity L* YIE Hours % µS Test 011 A 0.5 15 5.5 336 97.3 6.4 Test 011 B 1 14.7 5.9 176 96.9 6.9 Test 011 C 1.5 14.8 6.2 183 96.4 8.0 Test 011 D 2 14.1 4.6 203 96.1 8.5 Test 011 E 3 14.5 5.9 176 95 11,7

[0154] After washing, products are obtained with pH values ​​between 4.6 and 6.5 and conductivity between 176 µS and 336 µS.

[0155] The Yie (yellow) colorations are between 6.4 and 11.7. By comparison, the native Waxy base has a Yie of 8.11.

[0156] RVA viscosity measurements.

[0157] The results are presented in Table 6 and the Fig. 2 . Table 6 RVA 4500 at 92°C - MCL 106H: 1.37 g anh. QSP 28g buffer B -- in cPs REF PE Hours V0 peak or max between 4 and 6 min ① Visco V 92°C +12' ② fall = ①-② MAX visco between 4 and 10' V 10' at 50°C 011-- impregnated base 0 9,7 1015 107 908 1015 146 011 A 0,5 5,2 689 471 218 689 603 011 B 1 4,3 121 225 -104 170 346 011 C 1,5 3,9 71 124 -53 90 215 011 D 2 18,3 47 66 -19 53 135 011 E 3 11,1 36 50 -14 42 93

[0158] Each thermally modified starch presented above has improved stability during the use process compared to native starch: less viscosity gain and retrogradation phenomena are observed when using these starches. On this point, we note that the more the RVA drop tends towards 0 or becomes negative, the more the product will be functionalized, therefore the more resistant it will be and the less retrogradation it will express. Sedimentation test

[0159] The results are presented in the table below. Table 7 Tests Thermal treatment Decanting volume Hours mL Waxy corn base 0 100 Test 011 A 0,5 82 Test 011 B 1 37 Test 011 C 1,5 22 Test 011 D 2 30 Test 011 E 3 27

[0160] After 1 hour of reaction, very highly functional products are obtained, i.e. very resistant. Comparison of citrate-treated blend to citrate-treated waxy starch alone

[0161] The comparison is easy, by presenting on the same graph, the RVA curves of two products presenting the same level of functionalization (see Fig. 3 ), in this case, the mixture 043 LE of example 1 and the single product 011B.

[0162] We thus obtain, for a comparable level of functionalization (i.e. equivalent sedimentation tests), higher viscosity measurements for the citrate-treated mixture. Example 3. Obtaining the mixture of carbonated and thermally modified starches in a ventilated oven

[0163] The preparation of thermally modified starches is carried out conventionally, as we have said above, by implementing a preliminary step of alkaline impregnation of the starches, rather than an acid treatment as recommended by the Applicant company in the present invention.

[0164] The aim of this experiment is to compare the properties of the mixture of waxy starch and potato starch pretreated by these two different voices, with diametrically opposed pH. Preparation of the starch mixture and impregnation with sodium carbonate

[0165] The 50 / 50 potato starch + WAXILYS ® corn starch mixture is suspended in demineralized water to a dry matter content of 36.5%.

[0166] The pH and conductivity of the suspension are then measured.

[0167] Sodium carbonate is added to this milk under the following alternative conditions: If adding sodium carbonate in powder form: in sufficient quantity to obtain a final conductivity measured on a powder resuspended at 20% DM between 0.5 and 1 mS / cm. A contact time of 2 hours is allowed. If adding sodium carbonate in solution at 30% mass concentration: in sufficient quantity to obtain a conductivity, on the milk, between 2 and 4 mS / cm. A contact time of 30 minutes is sufficient, given that the sodium carbonate is already well dissolved in 30% solution.

[0168] Filter and dry to an equilibrium starch humidity of between 10 and 14%. Heat treatment in an oven

[0169] Weigh 40 g of sample and place it in an aluminum dish for METTLER TOLEDO ®< LJ16 balance (Humidity measuring balance). Place the dish in the MEMMERT ™< ventilated oven preheated to 170°C. Start the stopwatch as soon as the dishes have been introduced, and remove them according to the chosen reaction kinetics. Neutralization and washing of the products obtained

[0170] After the reaction at 170°C, the test is resuspended at 36% DM in demineralized water. The pH is adjusted to between 5.5 and 6 with HCl.

[0171] It is filtered and washed by percolation so as to obtain a conductivity of < 500 µS / cm in the final product resuspended at 20% MS.

[0172] The resulting “cake” is dried under a ventilated hood overnight at room temperature.

[0173] Coarsely ground on an IKA ® grinder then sieved through a 315 µm screen. Measurements: Humidity, pH, conductivity, color (L* and Yie), RVA 4500 viscosity on washed products and sedimentation test. Results

[0174] Physicochemical measurements: Table 8 MEASUREMENTS AFTER pH RECTIFICATION, FILTRATION, DRYING AND GRINDING at 20% MS WASHING PRODUCT PE HOURS Humidity pH Driver. CO LORATION MEASURES % µS L* Yi 038 AL 0,5 12,7 6,6 26 94 11 038 BL 1 13,0 6,3 31 89 23 038 CL 1,5 11,3 6,2 40 86 29 038 DL 2 12,2 6,2 48 85 32 RVA viscosity measurements

[0175] The results are presented in Table 9 and the Fig. 4 . Table 9 RVA 4500: 1.37g anh QSP 28g buffer B - cPs REF PE Reaction time Hours V0 peak or max between 4 and 6 min ① Visco after 12' at 92°C ② fall = ①-② MAX between 4 and 10' Visco FIN V 10' at 50°C 038 AL 0,5 9,2 756 428 328 756 539 038 BL 1 3,3 849 718 131 886 1017 038 CL 1,5 1,1 550 842 -292 861 1169 038 DL 2 -0,1 342 783 -441 678 1070

[0176] Each mixture of thermally modified starches described above exhibits improved stability during the use process compared to native starch: less viscosity gain and retrogradation phenomena are observed when using these starches. On this point, we note that the more the RVA drop tends towards 0 or becomes negative, the more functionalized the product will be, therefore the more resistant it will be and the less retrogradation it will express. Sedimentation test

[0177] The results are presented in the following Table 10. Table 10 REF TESTS PE time MCL 330N ml sedimentation test 038 AL 0,5 94 038 BL 1 67 038 CL 1,5 51 038 DL 2 46

[0178] We note that after 1 hour of reaction we already obtain a moderately functional product and after 1 hour 30 minutes of reaction and especially after 2 hours we obtain very highly functional products. Comparison of Citrate-treated Mixture to Carbonate-treated Mixture

[0179] The result of this comparison (sedimentation test, coloration and RVA viscosity) can be presented in the following table, based on two mixtures with the same level of functionalization, in this case the two products obtained after 2 hours of heat treatment.

[0180] The comparison is possible by presenting on the same graph the RVA curves of two products with the same level of functionalization (see Fig. 5 ). Table 11 RVA 4500 REF TESTS PE time MCL 330N ml sedimentation test Yi Coloring peak or max between 4 and 6 minutes fall Visco FIN V 10' at 50°C 038 DL 2 46 32 342 -441 1070 043 EL 2 43 18 183 -174 539

[0181] The advantage of the citrate treatment process according to the invention is to obtain a product with significant functionalities and therefore strong resistance to process conditions (temperature, acidity, shear) while obtaining a product with very little color compared to the carbonate treatment process.

[0182] We thus obtain, for a comparable level of functionalization (i.e. equivalent sedimentation tests), higher viscosity measurements for the citrate-treated mixture. Examples 4: comparison with commercial products of the same category

[0183] The commercial products shown in this table below are produced from a single botanical source, in this case waxy corn starch.

[0184] The results of the measurements are presented in the table below and represented in the Fig.6 . Table 12 COLORING SEDIMENTATION TEST RVA: MCL106H 1.37g anh QSP 28 g buffer B DOASGE CITRATES L* Yie MCL 330N ml peak or max between 4 and 6 minutes fall V 10' at 50°C % dry 038 DL 85 32 46 342 -441 1070 / / / 043EL 93 18 43 183 -174 539 0,03 NOVATION ENDURA 100 92 17 43 108 -110 354 0,10 NOVATION LUMINA 0100 95 9 44 136 -72 315 0,17

[0185] It appears that for the same level of resistance (same sedimentation test), the viscosity is higher with the products resulting from the invention than with commercial products.

[0186] By making a mixture of potato starch and waxy corn starch, the peak viscosities as well as the final viscosities are higher.

[0187] The 50 / 50 mixture gives the best compromise between its peak viscosity which represents the viscosity developed by the product and its falling viscosity which represents its level of resistance.

[0188] The 50 / 50 blend allows you to benefit from the advantages of both raw materials in terms of texture, strength and viscosity development.

Claims

1. A method for preparing a blend of at least two heat-modified starches, in which the starches are granular starches of different botanical origins, which comprises the steps consisting in: (i) preparing a starch milk containing at least two starches of different botanical origins, having total dry matter content comprised between 30 and 40%, and preferably between 35 and 37% by weight, (ii) preparing a citrate buffer with a pH comprised between 4 and 6, preferably 6, and adding said citrate buffer to the starch milk so as to obtain a milk pH comprised between 4 and 6, preferably 6, (iii) ensuring a contact time comprised between 0.5 and 5 hours, (iv) drying until reaching an equilibrium moisture content of 10 to 12%, (v) heating to a temperature higher than 150°C, for a residence time comprised between 10 minutes and 6 hours, (vi) resuspending, adjusting the pH, optionally washing and again drying the heat-modified starches thus obtained.

2. The method according to claim 1, characterized in that the botanical origin of the starches is selected from the group consisting of corn, waxy corn, amylomaize, wheat, waxy wheat, pea, faba bean, potato, waxy potato, tapioca, waxy tapioca, rice, konjac and is more preferentially potato starch and waxy corn starch.

3. The method according to claim 1, characterized in that the rise in temperature of the dry starch obtained in step (iv) is carried out in continuous turbo-dryer type devices, for which the setpoint temperature is set to more than 190 °C, preferably comprised between 200 and 210 °C, for a residence time comprised between 10 and 60 minutes, even more preferentially between 15 and 35 minutes and the delta T, defined as the difference in temperature between the setpoint temperature and the temperature of the product at the outlet of the reactor, is comprised between 17 and 27 °C.

4. The method according to claim 1, characterized in that the rise in temperature of the dry starch obtained in step (iv) is carried out in devices of the reaction fluidized bed type, for which the setpoint temperature is set to more than 130 °C, preferably comprised between 130 and 200 °C, for a residence time comprised between 30 minutes and 6 hours, even more preferentially between 2 hours and 4 hours.

5. A blend of at least two heat-modified starches of distinct botanical origin produced by the method of any one of the preceding claims, characterized in that it has a free citrate content of less than 0.05% and a bound citrate content comprised between 0.05 and 0.15%.

6. The blend of at least two heat-modified starches according to claim 5, characterized in that it has a coloring expressed with an L* value comprised between 80 and 90, and a Yie value comprised between 15 and 20.

7. Use of a heat-modified starch produced by the method according to any one of claims 1 to 4, as a thickening agent or texturizing agent in food applications, in particular in soups and sauces, and in dairy products.