CA2211372A1 - Paper containing thermally-inhibited starches - Google Patents

Abstract

Thermally-inhibited starches and flours, preferably cationic or amphoteric starches which are optionally chemically crosslinked, are added, primarily as wet end additives, to paper stock. The starch or flour is inhibited by dehydrating to anhydrous or substantailly anhydrous and then heat treating the dehydrated starch or flour for a time and at a temperature sufficient to inhibit the starch or flour and improve its viscosity stability when dispersed in water. The dehydration may be a termal or non-thermal dehydration (e.g., by alcohol extraction or freeze-drying). Preferably, the pH of the starch or flour is adjusted to neutral or above prior to dehydration.

Description

W O96/23104 PCTrUS9CJ~33 pAP~ CONTAINING THE~MATT~Y-INHIBITED STAR~

..
Technical Field This invention relates to paper and its manufacture using starches or flours.
Backqround Art ~ eat Treatment of Starches and Flours lo Heat/moisture treatment and annealing of starches and/or flours are taught in the literature and distinguished by the amount of water present.
~nne~ ling~ involves slurrying a granular starch with excess water at temperatures below the starch's or flour's gelatinization ~ ~~~ature. "Heat/moisture-treatment" involves a semi-dry treatment at temperatures below the starch's or flour's gelatinization temperature, with no added moisture and with the only moisture present being that normally present in a starch granule (which is typically 10% or more).
In the following discussion, a history of the various heat/moisture and annealing treatments of starch and/or flour is set out.
GB 263 897 (accepted Dec. 24, 1926) discloses an improvement in the heat treatment process of GB
228 829. The process of the '829 patent involves dry heating flour or wheat to a point at which substantially all of the gluten is rendered non-retainable in a washing test and then blending the treated flour or wheat with untreated flour or wheat to provide a blend having superior strength. The improvement of the '897 patent is continuing the dry heating, without, however, gelatinizing the starch, for a considerable time beyond that necessary to render all of the gluten non-retainable. "Dry-heating" excludes heating in a steam W O 96/23104 PCTrUS9 atmosphere or an atmosphere contA;n;ng considerable quantities of water vapor which would tend to gelatinize the starch. The wheat or flour may contain the usual amount of moisture, preferably not greater than 15%. The heat treatment may ~Yc~eA 7 hours at 77-93~C (170-200~F), e.g., 8 to 14 hours at 82~C (180~F) or 6 hours at 100~C
(212~F).
GB 530,226 (accepted Dec. 6, 1940) discloses a method for drying a starch cake containing about 40-50%
water with hot air or another gas at 149~C (300~F) or above without gelatinizing the starch. The starch cake is disintegrated by milling it to a finely divided state prior to drying.
GB-59S.552 (accepted December 9, 1947) discloses treatment of starch, more particularly a corn starch, which involves drying the starch to a relatively low moisture content of 1-2%, not exc-ee~;ng 3%, and subsequently dry heating the substantially moisture-free starch at 115-126~C for 1 to 3 hours. The treatment is intended to render the starch free from thermophilic bacteria. The starch should not be heated longer than necessary to effect the desired sterilization.
U.S. 3.490.917 (issued January 20, 1970 to C.A.F. Doe et al.) discloses a process for preparing a non-chlorinated cake flour suitable for use in cakes and sponges having a high sugar to flour ratio. The starch or a flour in which the gluten is substantially or completely detached from the starch granules is heated to a temperature of from 100-140~C and then cooled. The conditions are selected so that dextrinization does not occur, e.g., about 15 minutes at 100-115~C and no hold and rapid cooling at the higher temperatures. The heat treatment should be carried out under conditions which allow the water vapor to escape. The reduction in moisture content due to the heat treatment depends upon ,~. ~

3 l~EPLACEMENi PAGE
the temperature employed. At treatment te~peratures o~
100-105~C, the moisture content is reduced from 10-12% to 8-9~, by weight, while at medium and high temperatures the moisture content is typically reduced to 7~ or less.
Preferably, during cooling the moisture is allowed to reach molsture equilibrium with the atmosphere. The gelatinization temperature of the heat treated starch or ~1Our is approximately 0.5-1~C higher than that o~ a comparable chlorinated flour or starch. The heating can be carried out in many ways, including heating in a hot air fluidized bed.
U.S. 3,578,497 (issued May 11, 1971 to E. T.
Hjermstad) discloses a process for non-chemically improving the paste and gel properties of potato starch and imparting a swelling temperature as much as -7 to -1~C (20 to 30~F) higher. A concentrated suspension (20-40~ dry solids) at a neutral pH (5.5-8.0, preferably 6-7.5) is heated either for a long time at a relatively low temperature or for a short time at successively higher temperatures. The suspension is first heated at a temperature below the incipient swelling temperature of the particular batch of starch being treated preferably 49OC (120~F). Then the temperature is gradually raised until a temperature well above the original swelling temperature is attained. It is essential that swelling be avoided during the different heating periods so that gelatinization does not occur. After this steeping treatment the starch has a higher degree of granular stability. It resists rapid gelatinization and produces a rising or fairly flat viscosity curve on cooling. The pastes are very short textured, non-gumming, non-slimy, cloudy and non-cohesive. They form firm gels on cooling and aging.
U.S. 3,977,897 (issued August 31, 1976 to Wurzburg et al.) discloses a method for preparing non-~3129365.
AMEN~E3 S~~~

W0961~3104 PCT~S96JV~999 chemically inhibited amylose-containing starches. Both cereal and root starches can be inhibited, but the inhibition effects are more observable with root starches. Amylose-free starches, such as waxy corn starch, show no or very slight inhibition. The Brabender viscosity of cooked pastes derived from the treated starch was used to determine the inhibition level.
Inhibition was indicated by a delayed peak time in the case of the treated corn starch, by the lack of a peak and a higher final viscosity in the case of the treated achira starch, and by the loss of cohesiveness in the case of the treated tapioca starch. The granular starch is suspended in water in the presence of salts which raise the starch's gelatinization temperature so that the suspension may be heated to high temperatures without causing the starch granules to swell and rupture yielding a gelatinized product. The preferred salts are sodium, ammonium, magnesium or potassium sulfate; sodium, potassium or ammonium chloride; and sodium, potassium or ammonium phosphate. About 10-60 parts of salt are used per 100 parts by weight of starch. Preferably, about 110 to 220 parts of water are used per 100 parts by weight of starch. The suspension is heated at 50-100~C, preferably 60-90~C, for about 0.5 to 30 hours. The pH of the suspension is maintained at about 3-9, preferably 4-7.
Highly alkaline systems, i.e., pH levels above 9 retard inhibition.
U~S. 4 013.799 (issued March 22, 1977, to Smalligan et al.) discloses heating a tapioca starch above its gelatinization temperature with insufficient moisture (15 to 35% by total weight) to produce gelatinization. The starch is heated to 70-130~C for 1 to 72 hours. The starch is used as a thickener in wet, pre-cooked baby foods having a pH below about 4.5.

WO96123104 PCT~S96100999 U.S. 4.303.451 (issued December 1, 1981 to Seidel et al.) discloses a method for preparing a pregelatinized waxy maize starch having improved flavor characteristics reminiscent of a tapioca starch. ~he starch is heat treated at 120-200~C for 15 to 20 minutes.
The pregelatinized starch has gel strength and viscosity characteristics suitable for use in pudding mixes.
U.S. 4,303.452 (issued Dec. 1, 1981 to Ohira et al.) discloses smoking a waxy maize starch to improve gel strength and impart a smoky taste. In order to counteract the smoke's acidity and to obtain a final product with a pH of 4-7, the pH of the starch is raised to pH 9-11 before smoking. ~he preferred water content of the starch during smoking is 10-20~
The article "Differential ScAnn;ng Calorimetry of Heat-Moisture Treated Wheat and Potato Starches" by J.W. Donovan et al. in Cereal Chemistry, Vol. 60, No. 5, pp. 381-387 (1983) discloses that the gelatinization temperature of the starches increased as a result of the heat/moisture treatment or annealing. See also the article "A DSC Study Of The Effect Annealing On Gelatinization Behavior of Corn Starch" by B.R. Krueger et al. in Journal of Food Science, Vol. 52, No. 3, pp.
715-718 (1987).
U.S. 4,391,836 (issued July 5, 1983 to C.-W.
Chiu) discloses instant gelling tapioca and potato starches which are non-granular and which have a reduced viscosity. Unmodified potato and tapioca starches do not normally gel. The starches of the patent are rendered non-granular and cold-water-dispersible by forming an aqueous slurry of the native starch at a pH of about 5-12 and then drum-drying the slurry. The starches are ~ rendered gelling by heat treating the drum-dried starch for about 1.5 to 24 hours at 125-180~C to reduce the W O96/23104 PCTrUS~ 009~3 viscosity to within defined Brabender viscosity limitations.
U.S. 4,491,483 (issued January 1, 1985 to W.E.
Dudacek et al.) discloses subjecting a semi-moist blend of a granular starch with at least 0.25 wt. % of a fatty acid surfactant and sufficient water (about 10-40 wt. %) to a heat-moisture treatment at from about 50-120~C, followed by drying to about 5-15 wt. %, preferably 10 wt.
%, moisture. The heat-moisture treated starch-surfactant product is characterized by a hot water dispersibility of from about 60-100% and a higher pasting temperature than the granular starch from which it is derived.
Preferably, the treatment takes place in a closed container so that the moisture can be maintained at a constant level. The preferred conditions are 3 to 16 hours at 60-90~C. Degradation and dextrinization reactions are undesirable as they destroy the thickening ability of the starch. The use of conditions, such as, e.g., 35% moisture at 90~C for 16 hours results in reduced paste viscosity. It is believed the presence of the surfactant during the treatment permits formation of a complex within the partially swollen starch matrix with straight chain portions of the starch molecules. The limited moisture environment allows complex formation without gelatinization.
Japanese Patent Publication No. 61-254602, (published December 11, 1987) discloses a wet and dry method for heating waxy corn starch and derivatives thereof to impart emulsification properties. The wet or dry starch is heated at 100-200~C, preferably 130-150~C, for 0.5 to 6 hours. In the dry method, the water content is 10%, preferably 5%, or less. In the wet method, the water content is 5 to 50%, preferably 20-30%. The pH is 3.5-8, preferably 4-5.

WO 96~3104 PCTnUS9~ 3 The article "Hydrothermal Modification of Starches: The Difference between Annealing and Heat/Moisture-Treatment", by Rolf Stute, Starch/St~rke Vol. 44, No. 6, pp. 205-214 (1992) reports almost identical modifications in the properties of potato starch with annealing and heat/moisture treatments even through the alteration of the granular structure is different. The BrAhPn~r curves of the heat/moisture-treated and ~nne~led potato starches show the same typical changes, including a higher gelatinization temperature and a lower peak viscosity or no peak. The DSC curves also show a shift to higher gelatinization temperatures for both treatments. A combined treatment involving annealing a heat/moisture-treated potato starch leads to a further increase in gelatinization temperature without detectable changes in gelatinization enthalapy and with retention of the viscosity changes caused by the heat treatment. A combined treatment involving annealing a heat/moisture-treat potato starch does not lower the gelatinization temperature, when compared to the base starch, and increases the gelatinization temperature at higher heat/moisture treatment levels.
Chemiçal Crosslinking of Starches and Flours Starches and flours are chemically modified with difunctional reagents such as phosphorus oxychloride, sodium trimetaphosphate, adipic anhydride, acetic anhydride and epichlorohydrin to produce chemically crosslinked starches having excellent tolerance to processing variables such as heat, shear, and pH extremes. Such chemically crosslinked starches (also referred to as "inhibited starches") provide a desirable smooth texture and possess viscosity stability throughout processing operations and normal shelf life.
In contrast, when unmodified (i.e., non-crosslinked) starches, particularly waxy-based starches, W O96/23104 PCTrUS~'C~

are gelatinized, they reach a peak viscosity which soon begins to breakdown, loose thickening capacity and textural qualities, and behave unpredictably during storage as a result of the stresses encountered during processing. Heat, shear, and/or an extreme pH, especially an acidic pH, tend to fully disrupt the starch granules and disperse the starch.
PaPermaking Paperma~ing, as it is conventionally known, is a process of forming an aqueous slurry of pulp or wood cellulosic fibers, introducing the fibers onto a screen or wire to form an interlocking mat and to allow the water to drain through the screen or wire, squeezing the mat between rollers, drying it, and processing it into a dry roll or sheets.
Most paper is made on a Fourdrinier machine or a cylinder machine. In the Fourdrinier and multi-cylinder operations, and in other machine operations typical in papermaking, the aqueous slurry of the pulp or wood cellulosic fibers is formed at the feed or inlet to the machine, called the wet end. In the wet end, the pulp or wood cellulosic fibers are subjected to mec-h~n;cal beating or chemical refining to promote fiber-to-fiber bonding and ultimately strengthen the final paper product. However, a practical limit exists in the amount of refining that can be used for strength development. Excessive refining causes the sheet to lose other desirable characteristics, such as porosity, flexibility, brightness, and opacity. The addition of starch to the wet end obviates the need for excessive refining and improves the strength of the paper stock.
Historically, starch has been used in papermaking processes in a solubilized form so that the starch molecules are accessible for bonding to the cellulosic fibers. Starch retention on cellulosic fibers W O 96/23104 PCTnUS9~'C~

is increased by -k; ng the starch cationic or amphoteric.
Both starch and cellulosic fibers are anionic. Attracted to the negatively charged cellulosic fiber, and also to any of the common negatively charged f illers added to the wet end system, the cationic or amphoteric starch increases fiber-to-fiber and fiber-to-filler bonding, promoting a high degree of filler retention as well as strength.
Various cationic starches are known and used in lo the paper industry, with the tertiary amino and quaternary ammonium starch ethers being the most commercially significant derivatives. These and other cationic starches, as well as a method for their preparation, are described in "Cationic Starches" by D.B. Solarek, Modified Starches: Properties and Uses, Chapter 8, pp. 113-129 (1986).
While unmodified or modified chemically crosslinked starches have been used in a number of applications, they have not been used to a large extent in papermaking. The following publications disclose the use of chemically crosslinked starches in papermaking.
U.S. 3,417,078 (issued Decr h~r 7, 1968 to C. Patel) discloses the use of a cationic starch imidazoline derivative which is reacted with a crosslinking agent such as dichlorobutene. European Patent No. 097 371 (published January 4, 1984 to S. Frey) discloses the use of a nongelatinized starch which is cationized and partly crosslinked. U.S. 5.122,231 (issued June 16, 1992 to K.
Anderson) discloses the use of a cationic starch as a wet crosslinked end additive to produce increased starch loading capacity. JaPanese Patent No. 2-133695 (published May 22, 1990 to K. Maeda) discloses the use of ~ a cationic crosslinked starch having specified, but broad, degrees of crosslinking. U.S. 5 368,690 (issued November 29, 1994 to D. Solarek et al.) discloses the W O96/23104 PCTrUS~6/0~3 addition of jet-cooked chemically-crosslinked cationic or amphoteric starches having a viscosity breakdown of about 2-85% for improving calcium carbonate retention.
Disclosure of Invention This invention is directed to paper comprising, as an additive, a starch or flour that is thermally inhibited. The starches and flours may be non-pregelatinized granular starches and flours or pregelatinized granular or nongranular starches and flours. The starches or flours are thermally inhibited to impart the functionality previously provided by chemical crosslinking with a multifunctional crossl;nk;ng agent. The ther~ally-inhibited starches and flours can be used wherever starches are conventionally used in paper manufacture, for example, as wet end additives, coatings, and sizes.
Preferably, the starch is derivatized with cationic, anionic, non-ionic, or amphoteric substituents.
When anionic or non-ionic starches are used as wet end additives, they are typically used in combination with cationic or amphoteric starches.
The starches and flours are thermally inhibited, without the addition of chemical reagents, in a heat treatment process that results in the starch or flour becoming and remaining inhibited. The starches and flours are referred to as "inhibited" or "thermally-inhibited (abbreviated "T-I"). When these thermally-inhibited starches and flours are dispersed and/or cooked in water, they exhibit the textural and viscosity properties characteristic of a chemically-crosslinked starch. The starch granules are more resistant to viscosity breakdown. This resistance to breakdown results in what is subjectively considered a non-cohesive or "short" textured paste, meAn;ng that the gelatinized WO 96~23104 PCT~US96100999 starch or flour tends to be salve-like and heavy in viscosity rather than runny or gummy.
When the thermally-inhibited starches and flours are non-pregelatinized granular ~tarches or flours, the starches or flours exhibit an tlnchAnged or r~llc~A gelatinization temperature. In contrast, most annealed and heat/moisture treated starches show an increased gelatinization temperature. Chemically-crosslinked starches show an unchanged gelatinization temperature. It is believed the overall granular structure of the thermally-inhibited starches and flours has been altered.
The starches and flours that are substantially completely thermally inhibited will resist gelatinization. The starches and flours that are highly inhibited will gelatinize to a limited extent and show a continuing rise in viscosity but will not attain a peak viscosity. The starches and flours that are moderately inhibited will exhibit a lower peak viscosity and a lower percentage breakdown in viscosity compared to the same starch that is not inhibited. The starches and flours that are lightly inhibited will show a slight increase in peak viscosity and a lower percentage breakdown in viscosity compared to the same starch that is not inhibited.
The starches and flours are inhibited by a process which comprises the steps of dehydrating the starch or flour until it is anhydrous or substantially anhydrous and then heat treating the anhydrous or substantially anhydrous starch or flour at a temperature and for a period of time sufficient to inhibit the starch or flour. As used herein, "substantially anhydrous"
~ means containing less than 1% moisture by weight. The dehydration may be a thermal dehydration or a non-thermal dehydration such as alcohol extraction or freeze drying.

W 096/23104 PCT/us~D~

An optional, but preferred, step is adjusting the pH of the starch or flour to neutral or greater prior to the dehydration step.
The amount of thermal inhibition required will ~r~n~ on the reason the starch or flour is included in the paper, as well as the particular processing conditions used to prepare the paper. Paper pulps prepared with the thermally-inhibited starches and flours will poCcecs viscosity stability, and process tolerance such as resistance to heat, acid and shear. In addition, the viscosity of the jet-cooked thermally-inhibited starches is lower than the viscosity of jet-cooked chemically-crosslinked starches. This lower viscosity is a significant processing advantage.
Dep~n~;~g on the extent of the heat treatment, various levels of inhibition can be achieved. For example, lightly inhibited, higher viscosity products with little breakdown, as well as highly inhibited, low viscosity products with no breakdown, can be prepared by the thermal inhibition prnC~ce~ described herein.
For making paper having an alkaline pH, a thermally-inhibited, cationic, anionic, amphoteric, or non-ionic starch or flour is added to the wet end of the papermaking system. If a non-pregelatinized starch or flour is used, it is preferably cooked, e.g., by jet cooking, before addition. When used to improve the calcium carbonate retention, non-pregelatinized granular starch or flour will be thermally inhibited to a level such that, when the starch or flour is dispersed in water at 5% solids at 95~C, it will show a breakdown from peak viscosity in the range of 15-65%, preferably 25-45%.
Mode(s) For CarrYinq Out the Invention All starches and flours are suitable for use herein. The thermally-inhibited starches and flours are used in these compositions for their strength, retention, W096/23104 PCT~S9f'~9~9 thickening/bindings and surface modifying properties, which will depend on the starch or flour base selected as well as its modification.
The thermally-inhibited starches and flours can be derived from any native sourCe. A "native" starch or flour is one as it is found in nature in unmodified form.
Typical sources for the starches and flours are cereals, tubers, roots, legumes and fruits. The native source can be corn, pea, potato, sweet potato, h~n~A~ barley, wheat, rice, sago, amaranth, tapioca, sorghum, waxy maize, waxy tapioca, waxy rice, waxy barley, waxy potato, waxy sorghum, starches having an amylose content of 40%
or greater and the like. Preferred starches are waxy starches, potato, tapioca and corn.
The thermal inhibition process may be carried out prior to or after other starch or flour reactions are used to modify starch or flour. The starches may be modified by conversion (i.e., acid-, enzyme-, and/or heat-conversion), oxidation, phosphorylation, etherification (e.g., by reaction with propylene oxide), esterification (e.g., by reaction with acetic anhydride or octenylsuccinic anhydride), and/or chemical crossl;nki~q (e.g., by reaction with phosphorus oxychloride or sodium trimetaphosphate). The flours may be modified by bleaching or enzyme conversion.
Procedures for modifying starches are described in the Chapter "Starch and Its Modification" by M.W. Rutenberg, pages 22-26 to 22-47, HAn~hook of Water Soluble Gums and Resins, R.L. Davidson, Editor (McGraw-Hill, Inc., New York, NY 1980).
Native granular starches have a natural pH of about 5.0-6.5. When such starches are heated to temperatures above about 125 C in the presence of water, acid hydrolysis (i.e., degradation) of the starch occurs.
This degradation impedes or prevents inhibition.

W O96123104 PCTrUS9~0~

Therefore, the dehydration conditions need to be chosen so that degradation is avoided. Suitable conditions are dehydrating at low temperatures and the starch's natural pH or dehydrating at higher temperatures after increasing the pH of the starch to neutral or above. As used herein, "neutral" covers the range of pH values around pH
7 and is meant to include from about pH 6.5-7.5. A pH of at least 7 is preferred. More preferably, the pH is 7.5-10.5. The most preferred pH range is above 8 to below 10. At a pH above 12, gelatinization more easily occurs.
Therefore, pH adjustments below 12 are more effective.
It should be noted that the textural and viscosity benefits of the thermal inhibition process tend to be enhanced as the pH is increased, although higher pHs tend to increase browning of the starch or flour during the heat treating step.
To adjust the pH, the non-pregelatinized granular starch or flour is typically slurried in water or another aqueous medium, in a ratio of 1.5 to 2.0 parts of water to 1.0 part of starch or flour, and the pH is raised by the addition of any suitable base. Buffers, such as sodium phosphate, may be used to maintain the pH
if needed. Alternatively, a solution of a base may be sprayed onto the powdered starch or flour until the starch or flour attains the desired pH, or an alkaline gas such as ammonia can be infused into the starch or flour. After the pH adjustment, the slurry is then either dewatered and dried, or dried directly, typically to a 2-15% moisture content. These drying procedures are to be distinguished from the thermal inhibition process steps in which the starch or flour is dehydrated to anhydrous or substantially anhydrous and then heat treated.
The starches or flours can be pregelatinized prior to or after the thermal inhibition process using W O 96/23104 PC~US9~ g methods known in the art. The amount of pregelatinization, and conse~uently, whether the starch will display a high or a low initial viscosity when - dispersed in water, can be regulated by the pregelatinization procedure used, as is known in the art.
The resulting pregelatinized starches are useful in applications where cold-water-soluble or cold-water-dispersible ~tarches are used.
Pregelatinized granular starches and flours have retAi~e~ their granular structure but lost their polarization crosses. They are pregelatinized in such a way that a majority of the starch granules are swollen, but remain intact. Exemplary processes for preparing pregelatinized granular starches are disclosed in U.S.
4.280 851 (issued July 28, 1981 to E. Pitchon et al.), U.5. 4 465.702 (issued August 14, 1984 to J.E. Eastman et al.), U.S. 5.037.929 (issued August 6, 1991 to S.
Rajagopalan), and U.S. 5.149.799 (issued September 22, 1992 to Roger W. Rllh~n~), the disclosures of which are incorporated by reference.
Pregelatinized non-gr~nlllAr starches and flours have also lost their polarization crosses and have become so swollen that the starches have lost their granular structure and broken into fragments. They can be prepared according to any of the known physical, chemical or thermal pregelatinization processes that destroy the granule such as drum drying, extrusion, or jet-cooking.
See U.S. 1.516.512 (issued Nov. 25, 1924 to R.W. G.
Stutzke); U.S. 1.901.109, (issued March 14, 1933 to W.
Maier); US. 2 314.459 (issued March 23, 1943 to A.A.
Salzburg; US. 2 582 198 (issued January 8, 1957 to O.R.
Ethridge); US. 2 805 966 (issued September 10, 1957 to O.R. Ethridge); US. 2 919 21~ (issued December 29, 1959 to O.R. Ethridge); U.S. 2 940 876 (issued June 14, 1960 to N.E. Elsas); U.S. 3 086 890 (issued April 23, 1963 to , W O96/23104 PCTrUS9 A. Sarko et al.); U.S. 3.133 836 (issued May 19, 1964 to U.L. Winfrey); U.S. 3.137 592 (issued June 16, 1964 to T.F. Pratzman et al.); U.S. 3 234 046 (issued February 8, 1966 to G.R. Etchison); U.S. 3 607 394 (issued Sept~her 5 21, 1971 to F.J. Germino); U.S. 3 630 775 (issued December 18, 1971 to A.A. Winkler); and U.S. 5.131.953 (i~sued July 21, 1992 to J.J. Kasica et al.); the disclosures of which are incorporated by reference.
If the pregelatinization process is performed first and the pregelatinized starch or flour is granular, the pH is adjusted by slurrying the pregelatinized granular starch or flour in water in a ratio of 1.5-2.0 parts to 1.0 part starch, and optionally, the pH is adjusted to neutral or greater. In another embodiment, the slurry is simultaneously pregelatinized and dried and the dried, starch or flour is thermally inhibited. If the thermal inhibition process is performed first, the starch or flour is slurried in water, the pH of the starch or flour is adjusted to neutral or greater, and the starch or flour is dried to about 2-15% moisture.
The dried starch or flour is then dehydrated and heat treated. The inhibited starch or flour is reslurried in water, optionally pH adjusted, and simultaneously pregelatinized and dried.
For non-granular pregelatinized starches or flours prepared by drum drying, the pH is raised by slurrying the starch or flour in water at 30-40% solids and adding a sufficient amount of a solution of a base until the desired pH is reached.
For non-granular pregelatinized starches or flours prepared by the continuous coupled jet-cooking/spray-drying process of U.S. 5 131 953 or the dual atomization/spray-drying process of U.S. 4 280 851, the starch or flour is slurred at 6-10% solids in water and the pH is adjusted to the desired pH by adding a W O 96/23104 PCT~US9.~J00~5 sufficient amount of a solution of a base until the desired pH is reached.
Suitable bases for use in the pH adjustment step include, but are not limited to, sodium hydroxide, sodium carbonate, tetrasodium pyrophosphate, ammonium orthophosphate, ~;co~;um orthophosphate, trisodium phosphate, calcium carbonate, calcium hydroxide, potassium carbonate, and potassium hydroxide, and any other bases a~ ed for use under the applicable regulatory laws. The preferred base is sodium carbonate.
It may be possible to use bases not approved provided they can be washed from the starch or flour so that the final product conforms to good manufacturing practices for desired end use.
A thermal dehydration is carried out by heating the starch or flour in a heating device for a time and at a temperature sufficient to reduce the moisture content to less than 1%, preferably 0%. Preferably, the temperatures used are 125~C or less, more preferably 100-120~C. The dehydrating temperature can be lower than 100~C, but a temperature of at least 100~C will be more efficient for removing moisture.
Representative processes for carrying out a non-thermal dehydration include freeze drying or extracting the water from the starch or flour using a solvent, preferably a hydrophilic solvent, more preferably a hydrophilic solvent which forms an azeotropic mixture with water (e.g., ethanol).
For a laboratory scale dehydration with a solvent, the starch or flour (about 4-5% moisture) is placed in a Soxhlet thimble which is then placed in a Soxhlet apparatus. A suitable solvent is placed in the apparatus, heated to its reflux temperature, and refluxed for a time sufficient to dehydrate the starch or flour.
Since during the refluxing the solvent is condensed onto W O96/23104 PCTnUS96/00999 the starch or flour, the starch or flour is exposed to a lower temperature than the solvent's boiling point. For example, during ethanol extraction the temperature of the starch is only about 40-50~C even though ethanol's boiling point is about 78~C. When ethanol is used as the solvent, the refluxing is continued for about 17 hours.
The extracted starch or flour is removed from the thimble, spread out on a tray, and the excess solvent i8 allowed to flash off. The time required for ethanol to flash off is about 20-30 minutes. The dehydrated starch or flour is immediately placed in a suitable heating apparatus for the heat treatment. For a commercial scale dehydration any continuous extraction apparatus is suitable.
For dehydration by freeze drying, the starch or flour (4-5% moisture) is placed on a tray and put into a freeze dryer. A suitable bulk tray freeze dryer is available from FTS Systems of Stone Ridge, New York under the trademark Dura-Tap. The freeze dryer is run through a programmed cycle to remove the moisture. The temperature is held constant at about 20OC and a vacuum is drawn to about 50 milliTorr (mT). The starch or flour is removed from the freeze dryer and immediately placed into a suitable heating apparatus for the heat treatment.
After it is dehydrated, the starch or flour is heat treated for a time and at a temperature sufficient to inhibit the starch or flour. The preferred heating temperatures are greater than about 100~C. For practical purposes, the upper limit of the heat treating temperature is about 200~C. Typical temperatures are 120-180~C, preferably 140-160~C, most preferably 160~C.
The temperature selected will depend upon the amount of inhibition desired and the rate at which it is to be achieved.

CA 022ll372 l997-07-24 W O 96/23104 PCTAUS~ 0~5Y

The time at the final heating temperature will ~Pr~n~ upon the level of inhibition desired. When a conventional oven is used, the time ranges from 1 to 20 - hours, typically 2 to 5 hours, usually 3.5 to 4.5 hours.
When a fluidized bed is used, the times range from 0 - minutes to 20 hours, typically 0.5 to 3.0 hours. Longer times are required at lower temperatures to obtain more inhibited starches.
For most applications, the thermal dehydrating and heat treating steps will be continuous and accomplished by the application of heat to the starch or flour beginning from ambient temperature. The moisture will be driven off during the heating and the starch will become anhydrous or substantially anhydrous. Usually, at these initial levels of inhibition, the peak viscosities are higher than the peak viscosities of starches heated for longer times, although there will be greater breakdown in viscosity from the peak viscosity. With continued heat treating, the peak viscosities are lower, but the viscosity breakdowns are less.
Although starches with a wide range of inhibition can be obtained by the thermal inhibition process, the preferred starches for use in papermaking will have a level of inhibition which results in the starch, after dispersion in an aqueous medium, having a balance of both intact granules and solubilized starch molecules.
The process may be carried out as part of a continuous process involving the extraction of the starch from a plant material.
As will be seen in the following examples, the source of the starch or flour, the initial pH, the dehydrating conditions, the heating time and temperature, and the equipment used are all interrelated variables that affect the amount of inhibition.

W O96/23104 PCTrUS96/00999 ZO
The heating steps may be performed at normal pressures, under vacuum or under pressure, and may be accomplished by conventional means known in the art. The preferred method is by the application of dry heat in dry air or in an i~ert gaseous environment.
The heat treating step can be carried out in the same apparatus in which the thermal dehydration occurs. Most conveniently the process is continuous with the thermal dehydration and heat treating occurring in the same apparatus, as when a fluidized bed reactor is used.
The dehydrating and heat treating apparatus can be any industrial ovens, conventional ovens, microwave ovens, dextrinizers, dryers, mixers and blenders equipped with heating devices and other types of heaters, provided that the apparatus is fitted with a vent to the atmosphere so that moisture does not accumulate and precipitate onto the starch or flour. The preferred apparatus is a fluidized bed. Preferably, the apparatus is equipped with a means for removing water vapor, such as, a vacuum or a blower to sweep air or the fluidizing gas from the head-space of the fluidized bed. Suitable fluidizing gases are air and nitrogen. For safety reasons, it is preferable to use a gas containing less than 12% oxygen.
Superior inhibited starches having high viscosities with low percentage breakdown in viscosity are obtained in shorter times in the fluidized bed reactor than can be achieved using other conventional heating ovens or dryers.
The starches or flours may be inhibited individually or more than one may be inhibited at the same time. They may be inhibited in the presence of other materials or ingredients that would not interfere W O96123104 PCTlUS~C1~0533 with the thermal inhibition process or alter the properties of the starch or flour product.
~n~ustrial A~licability - The thermally-inhibited starches and flours can be used wherever starches and flours are conventionally used in papel ~k;~, e.g., as wet end additives, sizes, or coatings.
When used as wet end additives, they increase the dry strength and the retention of fillers or pigments and minimize or overcome strength losses due to the incorporation of fillers. They are usually added as an aqueous dispersion which is prepared by cooking a thermally-inhibited non-pregelatinized granular starch or flour or dispersing a thermally-inhibited pregelatinized starch or flour.
Converted starches, as well as starch ethers and starch esters, are useful as surface sizing agents, or coatings for providing barriers for water vapor, grease, and solvents.
Non-converted starches are useful as thickeners in surface sizings.
Cationic starches are useful as flocculants for suspensions or inorganic or organic matter having a negative charge. When used as surface sizes, they provide pick-up and penetration for inks and the like.
When used as coatings, they act as water vapor, grease, and solvents barriers.
Cationization of starch can be carried out by well known chemical reactions with reagents containing primary, secondary, tertiary or quaternary amines attached through an ether or ester linkage, as disclosed, for example, in Solarek, "Cationic Starches".
Typical cationic and cationogenic groups include the diethylaminoethyl ether groups introduced by reaction with 2-diethylaminoethylchloride hydrochloride W 096/23104 PCTrUS9GI~v3~9 or 3-(trimethyl ammonium chloride)-2-hydroxypropyl ether groups introduced by reaction with 3-chloro-2-hydroxypropyl trimethylammonium chloride. See, U.S.
2,813,093 (issued November 12, 1957 to Caldwell, et al.) for reactions with dialkylaminoalkyl halides to introduce tertiary amino ~L Ou~ which can then quaternized to ammonium groups; U.S. 2,876,217 (issued March 3, 1959 to PA~r-h~ll) for gelatinizable cationic starch ethers; U.S.
~,970,140 (issued January 31, 1961 to Hullenger et al.) for granular starches containing alkyl amine groups; U.S.
2,989,520 (issued June 20, 1961 to Rutenberg, et al.) for reactions with beta-halogeno akylsulfonium-, vinyl sulfonium-, or epoxyalkyl-sulfonium salts U.S. 3,077,469 (issued February 12, 1963 to A. Aszalos) for reactions with beta-halogenoalkyl phosphonium salts; U.S. 4,119,487 (issued October 10, 1978 to Tessler) for epihalohydrin-tertiary amino or tertiary amine salt reaction products;
U.S. 4,260,738 (issued April 7, 1981), U.S. 4,278,573 (issued July 14, 1981 to Tessler), and U.S. 4,387,221 (issued June 7, 1983 to Tessler) for alkyl- or alkenyl-sulfosuccinates; and U.S. 4,675,394 (issued January 23, 1987 to D. Solarek, et al.) for reactions with aminoalkyl anhydrides, amino epoxides, amino halides, or aryl amines. The disclosures of the above patents are incorporated herein by reference.
Converted starches containing cationic or cationogenic groups and sulfo-succinate groups, useful as pigment retention aids, are described in U.S. 4,029,544 (issued June 14, 1977 to Jarowenko et al.).
Amphoteric starches are also useful herein.
Dual treatments of starch with cationic and anionic modifying reagents have been used to prepare amphoteric derivatives for use in different applications, including as wet end additives. Cationic derivatization, particularly with tertiary amino or quaternary ammonium W O 96123~04 PCT~USg6,'~05 ether groups has been combined with further derivatization with anionic groups such as phosphate, phosphonate, sulfate, sulfonate or carboxyl groups. The - resulting amphoteric starches and methods for their preparation are disclosed in Solarek, "Cationic Starches", supra, pp. 120-121. See also, U.S. 3 459 632 ~ August 5, 1969 to Caldwell, et al.) for the preparation of starch derivatives cont~;~ing anionic phosphate groups and cationic groups; U.S. 3 562 103 (issued February 9, 1971 to Moser et al.); and U.S.
4.876,336 (issued October 24, 1987 to Solarek, et al.) for the preparation of starch derivatives containing anionic phosphate groups and cationic tertiary amino or quaternary ammonium groups. The disclosures of the above patents are incorporated herein by reference.
Anionic starches are prepared by substitution with anionic yLo~ such as those discussed above according to the proc~ res previously disclosed for preparing amphoteric starches but without the use of cationic reagents.
When anionic or non-ionic starches are used in the papermaking process, they are used in conjunction with cationic additives, such as a synthetic polymer containing the residues of cationic monomers or a cationic or amphoteric starch. Alternatively, the anionic or nonionic starch may be thermally inhibited in combination with a cationic or amphoteric starch, and this thermally inhibited starch blend used as an additive in the papermaking.
Aldehyde-containing starches, useful as strength aids starches, are described in U.S. 4 675.394 (issued June 23, 1987 to Jobe et al.), the disclosure of which is incorporated herein by reference.
Recovery of the starch derivatives may be readily accomplished, with the particular method employed -W O96/2310~ PCTrUS96/00999 being dependent on the form of the starch base. Thus, a granular starch is recovered by filtration, optionally washed with water to remove any residual salts, and dried. The granular starch products may also be drum-dried, jet-cooke~ and spray-dried, or gelatinized and isolated by alcohol precipitation or freeze drying to form non-granular products. If the starch derivative is non-granular, it may be purified by dialysis to remove residual salts and isolated by alcohol precipitation, lo freeze drying, or spray drying.
The amount of a substituent on the starch is defined as the degree of substitution (DS) which means the average number of substituent groups per anhydroglucose unit of the starch molecule. The degree of substitution can be varied, although generally a degree of substitution (DS) from about 0.005 to 0.2, and preferably from about 0.01 to 0.05, will be used. Higher degrees of substitution may be used, but this makes the starch more costly and difficult to make, and therefore, is not economically attractive.
When the starch is to be used in the wet end system of papermaking, it is derivatized, optionally pregelatinized, and then thermally inhibited. It may then be subjected again to controlled gelatinization.
Jet cooking provides a continuous process for gelatinizating the starch. It is well known in the art.
By varying the temperature, pressure, flow rate, solids content of the starch slurry, and equipment configuration (e.g., haffles and mixers) it is possible to provide gelatinization conditions where a proportion of the thermally-inhibited starch granules will hydrate and partially swell without fragmenting and a portion of the granules will disperse and solubilize. Gelatinization and dispersion of the starch occurs in a very short period of time, from a few seconds to a few minutes.

W O 96/23104 PCT~US9CJ'~g~g Suitable conditions are temperatures of about 90-165~C, preferably about 100-160~C, more preferably about 105-122~C, and an applied back pressure of at least 5 psi, preferably about 5-20 psi, although the pressure can be as high as loO p5i. The starch concentration in the c~Qking chamber should be at least 3~, preferably from 3-7% solids.
The derivatized and thermally-inhibited starches, optionally pregelatinized, may be effectively 0 A~ to pulp prepared from any type and combination of cellulosic fibers and/or synthetic fibers. Cellulosic materials include bleached and unbleached sulfate (kraft), bleached and unbleached sulfite, bleached and unbleached soda, neutral sulfite, semi-chemical, chemiground wood, and ground wood. Synthetic fibers include polyamides, polyesters, rayon and polyacrylic resins, as well as fibers from minerals, such as asbestos and glass. Viscose rayon or regenerated cellulose fibers can also be used for the pulp.
Inert mineral fillers, such as, clay, titanium dioxide, talc, calcium carbonate, calcium sulfate, and diatomaceous earths, rosin, and other additives commonly introduced into paper, such as, dyes, pigments, sizing additives, alum, and anionic retention aids, may be added to the pulp or furnish.
The amount of thermally-inhibited starch that is added to the wet end or paper pulp will be an effective amount, typically from about 0.05-5%, preferably from about 0.1-2~, by weight based on the dry weight of the pulp.
In a preferred embodiment, the thermally-inhibited starch is added to a simple alkaline papermaking system (i.e., one containing mainly pulp, alum and starch), or to a microparticle colloidal papermaking system. Suitable microparticles for W O96/23104 PCTrUS9G~ 3 inclusion in the alkaline systems include colloidal silica, bentonite and anionic alum, which are usually incorporated in amounts of at least 0.001%, more typically about 0.01-1% by weight based on the weight of dry pulp. Further description of such microparticle inorganic materials may be found in U.S. 4,388.150 (issued June 14, 1983 to Batelson et al.); U.S. 4.643.801 (issued February 17, 1987 to Johnson); U.S. 4.7S3.710 (issued June 28, 1988 to Holroyd et al.); and U.S.

4.913.775 (issued April 3, 1990 to Holroyd et al.).
It has been determined that significantly improved performance in papermaking is produced when the starch is thermally inhibited to a level sufficient to provide a percentage breakdown in viscosity of about 2-60% calculated according to the formula:
% Breakdown = Peak-(ViscositY at 95~C + 30') x 100, Peak where Peak is the peak viscosity in Brabender Units and is the viscosity after holding at 95~C for 30 minutes.
Sample Preparation Unless indicated otherwise, all the starches and flours used were granular and were provided by National Starch and Chemical Company of Bridgewater, New Jersey.
The controls for the test samples were from the same native sources as the test samples, were unmodified or modified in the same manner as the test samples, and were at the same pH, unless otherwise indicated.
All starches and flours, both test and control samples, were prepared and tested individually.
The pH of the samples was raised by slurrying the starch or flour in water at 30-40% solids and adding a sufficient amount of a 5% sodium carbonate solution until the desired pH was reached.
Measurements of pH, either on samples before or after the thermal inhibition steps, were made on samples W O96123104 PCTJUS~r,~9g consisting of one part starch or flour to four parts water.
After the pH adjustments, if any, all non-pregelatinized granular samples were spray-dried or flash-dried as conventional in the art (without gelatinization) to about 2-15% moisture.
After the pH adjustment, if any, slurries of the starches to be pregelatinized to granular pregelatinized starches were intro~t-~eA into a pilot spray dryer, Type 1-KA#4F, from APV Crepaco, Inc., Dryer Division, Attleboro Falls, Massachusetts, using a spray nozzle, Type 1/2J, from Spraying Systems Company of Wheaton, Ill. The spray nozzle had the following configurations: fluid cap 251376, air cap 4691312. The low initial cold viscosity samples were sprayed at a steam:starch ratio of 3.5-4.5:1, and the high initial cold viscosity samples were sprayed at a steam:starch ratio of 5.5-6.5:1. Moisture content of all pregelatinized samples after spray drying and before the dehydration step in the thermal inhibition process was 4-10%.
For the samples pregelatinized by drum drying the pH was raised by slurrying the starch or flour in water at 30-40% solids and adding a sufficient amount of a 5% sodium carbonate solution until the desired pH was reached. A single steam-heated steel drum at about 142-145~C was used for the drum drying.
For the samples pregelatinized by the continuous coupled jet-cooking/spray-drying process of Y.S. 5.131.953 or the dual atomization/spray-drying process of U.S. 4,280,851, the starch or flour was slurred at 6-10% solids in water and the pH was adjusted to the desired pH by adding a sufficient amount of 5%
sodium carbonate solution until the desired pH was reached.

CA 022ll372 l997-07-24 W O96/23104 PCTrUS9~'0~5 Except where a conventional oven or dextrinizer is specified, the test samples were dehydrated and heat treated in a fluidized bed reactor, model number FDR-100, manufactured by Procedyne Corporation of New Brunswick, New Jersey. The cross-sectional area of the fluidized bed reactor was 0.05 sq meter. The starting bed height was 0.3-0.8 meter, but usually 0.77 meter. The fluidizing gas was air except where otherwise indicated.
When granular non-pregelatinized starches were being heat treated, the gas was used at a velocity of 5-15 meter/min. When pregelatinized granular starches were being heat treated, the gas was used at a velocity of 15-21 meter/min. The side walls of the reactor were heated with hot oil, and the fluidizing gas was heated with an electric heater. The samples were loaded into the reactor and then the fluidizing gas was introduced, or the samples were loaded while the fluidizing gas was being introduced. No difference was noted in the samples in the order of loading. Unless otherwise specified, the samples were brought from ambient temperature up to no more than 125~C until the samples became anhydrous and were further heated to the specified heat treating temperatures. When the heating temperature was 160~C, the time to reach that temperature was less than three hours.
The moisture level of the samples at the final heating temperature was 0%, except where otherwise stated. Portions of the samples were removed and tested for inhibition at the temperatures and times indicated in the tables.
Unless specified otherwise, the samples were tested for inhibition using the following Brabender Procedures.

W O96/23104 PCTAUSg~lV~3~3 Brabender Procedure -~on-Pregelatinized Granular Starches Unless other stated, the following Br~h~n~e~

- 5 procedure was used. All samples, except for corn, tapioca and waxy rice flour, were slurried in a sufficient amount of distilled water to give a 5%
anhydrous solids starch slurry. Corn, tapioca, and waxy rice flour were slurried at 6.3% anhydrous solids. The pH was adjusted to pH 3.0 with a sodium citrate, citric acid buffer and the slurry was introduced into the sample cup of a BrAh~n~ VISCO/Amylo/GRAPH (manufactured by C.W. Brabender Instruments, Inc., Hackensack, NJ) fitted with a 350 cm/gram cartridge. The VISCO\Amylo\GRAPH
records the torque required to balance the viscosity that develops when a starch slurry is subjected to a programmed heating cycle. The record consists of a curve tracing the viscosity through the heating cycle in arbitrary units of measurement termed Br~h~nA~r Units (BU).
The starch slurry is heated rapidly to 92~C and held for 10 minutes. The peak viscosity and viscosity ten minutes after peak viscosity were recorded in Brabender Units (BU). The percentage breakdown in viscosity (+ 2%) was calculated according to the formula:
% Breakdown = Peak - (Peak + 10') x 100, Peak where "peak" is the peak viscosity in Brabender units, and "(peak + 10')" is the viscosity in Brabender Units at ten minutes after peak viscosity. If no peak viscosity was reached, i.e., the data indicate a rising (ris.) curve or a flat curve, the viscosity at 92~C and the viscosity at 30 minutes after att~; n; ng 92~C were recorded.
Using data from the Brabender curves, inhibition was determined to be present if, when dispersed at 5% or 6.3% solids in water at 92-95OC and pH

W O96/23104 PCTrUS~C

3, during the Brabender heating cycle, the Brabender data showed (i) no or almost no viscosity, indicating the starch was so inhibited it did not gelatinize or strongly resisted gelatinization; (ii) a continuous rising viscosity with no peak viscosity, indicating the starch was highly inhibited and gelatinized to a limited extent;
(iii) a lower peak viscosity and a lower percentage breakdown in viscosity from peak viscosity compared to a control, indicating a moderate level of inhibition; or (iv) a slight increase in peak viscosity and a lower percentage breakdown compared to a control, indicating a low level of inhibition.
Characterization Of Inhibition of Non-Pregelatinized Çranular Starches By Brabender Curves Characterization of a thermally-inhibited starch is made more conclusively by reference to a measurement of its Br~hen~ viscosity after it is dispersed in water and gelatinized.
For non-inhibited starches, the cycle passes through the initiation of viscosity, usually at about 60-70OC, the development of a peak viscosity in the range of 67-95~C, and any breakdown in viscosity when the starch is held at an elevated temperature, usually 92-95~C.
Inhibited starches will show a Brabender curve different from the curve of the same starch that has not been inhibited (hereinafter the control starch). At low levels of inhibition, an inhibited starch will attain a peak viscosity somewhat higher than the peak viscosity of the control, and there may be no decrease in percentage breakdown in viscosity compared to the control. As the amount of inhibition increases, the peak viscosity and the breakdown in viscosity decrease. At high levels of inhibition, the rate of gelatinization and swelling of the granules decreases, the peak viscosity disappears, and with prolonged cooking the Brabender trace b~co~c a W O 96123104 PCT~US96/00999 rising curve indicating a slow continuing increase in viscosity. At very high levels of inhibition, starch granules no longer gelatinize, and the Brabender curve - remains flat.
Br~h~n~er Procedure - Pregelatinized Granular and Non-Granular Starches The pregelatinized thermally-inhibited starch to be tested was slurried in a sufficient amount of distilled water to give a 4.6% anhydrous solids starch slurry at pH 3 as follows: 132.75 g sucrose, 26.55 g starch, 10.8 g acetic acid, and 405.9 g water were ;Y~
for three minutes in a st~n~rd home Mixmaster at setting #1. The slurry was then introduced to the sample cup of a Brabender VISC0/Amylo/GRAPH fitted with a 350 cm/gram cartridge and the viscosity measured as the slurry was heated to 30~C and held for 10 minutes. The viscosity at 300C and lo minutes after hold at 30OC were recorded.
The viscosity data at these temperatures are a measurement of the extent of pregelatinization. The higher the viscosity at 30~C, the grater the extent of granular swelling and hydration during the pregelatinization process.
Heating was continued to 95~C and held at that temperature for 10 minutes.
The peak viscosity and viscosity 10 minutes after 95~C were recorded in Brabender Units (BU). The percentage breakdown was calculated using the previous formula.
If no peak viscosity was reached, that is, the data indicated a rising curve or a flat curve, the viscosity at 950C and the viscosity at 10 minutes after attaining 95~C were recorded.

W O96/23104 PCTrUS~6/00999 Characterization of Inhibition of Pregelatinized Granular Starches bY Brabender Curves As discussed above, characterization of a thermally-inhibited starch is made more conclusively by reference to a measurement of its viscosity after it is dispersed in water and gelatinized using the instrument described above.
For pregelatinized granular starches, the level of viscosity when dispersed in cold water will be dependent on the extent to which the starch was initially cooked out during the pregelatinization process. If the granules were not fully swollen and hydrated during pregelatinization, gelatinization will continue when the starch is dispersed in water and heated. Inhibition was determined by a measurement of the starch viscosity when the starch was dispersed at 4.6% solids in water at pH 3 and heated to 95~C.
When the pregelatinized granular starch had a high initial cold viscosity, meaning it was highly cooked out in the pregelatinization process, the resulting Br~h~n~er traces will be as follows: for a highly inhibited that the starch, the trace will be a flat curve, indicating that the starch is already very swollen and is so inhibited starch it is resisting any further gelatinization or the trace will be a rising curve, indicating that further gelatinization is occurring at a slow rate and to a limited extent; for a less inhibited starch, the trace will he a dropping curve, indicating that some of the granules are fragmenting, but the overall breakdown in viscosity will be lower than that for a non-inhibited control or the trace will show a second peak but the breakdown in viscosity will be lower than that for a non-inhibited control.
When the pregelatinized starch had a low initial cold viscosity, meaning it was not highly cooked W O 96/23104 PCTrUS96/00999 out in the pregelatinization process and more cooking is needed to reach the initial peak viscosity, the resulting BrAh~n~er traces will be as follows: for a highly - inhibited starch, the trace will be a rising curve, indicating that further gelatinization is occurring at a slow rate and to a limited extent; for a less inhibited starch, the trace will show a peak viscosity as gelatinization occurs and then a drop in viscosity, but with a lower percentage breakdown in viscosity than for a non-inhibited control.
If no peak viscosity was reached, that is, the data indicated a rising curve or a flat curve, the viscosity at 95~C and the viscosity at lo minutes after attaining 95~C were recorded.
Characterization of Inhibition of Pregelatinized Non-Granular Starches b~ Brabender Curves The resulting Brabender traces will be as follows: for a highly inhibited starch the trace will be flat, indicating that the starch is so inhibited that it is resisting any further gelatinization or the trace will be a rising curve, indicating that further gelatinization is occurring at a slow rate and to a limited extent; for a less inhibited starch, the trace will show a dropping curve, but the overall breakdown in viscosity from the peak viscosity will be lower than that for a non-inhibited control.
Brabender Procedure - Crosslinked Starches The crosslinked, thermally-inhibited cationic and amphoteric starches (23.0 g) to be tested was combined with 30 ml of an aqueous solution of citric acid monohydrate (prepared by diluting 210.2 g of citric acid monohydrate to 1000 ml in a volumetric flask) and sufficient water was added to make the total charge weight 460.0 g. The slurry is added to the cooking chamber of the Brabender VISC0 amylo GRAPH fitted with a W 096/23104 PcTlus~ J~

700 cm/gram cartridge and rapidly heated from room temperature to 95~C. The peak viscosity (highest viscosity observed) and the viscosity after 30 minutes at 95~C were recorded. The percentage breakdown in viscosity (~2%) was calculated according to the formula % Breakdown = Peak - ~ViscositY after 30' at 95~C~ x 100 Peak ~hara~terization of Inhibition bY Cooks A dry blend of 7 g of starch or flour (anhydrous basis) and 14 g of sugar were added to 91 ml of water in a Waring blender cup at low speed, then transferred to a cook-up beaker, allowed to stand for 10 minutes, and then evaluated for viscosity, color, clarity and texture.
Some of the granular non-pregelatinized starch samples were tested for pasting temperature and/or gelatinization temperature using the following procedures.
Rapid Visco Analvzer (RVA) This test is used to determine the onset of gelatinization, i.e., the pasting temperature. The onset of gelatinization is indicated by an increase in the viscosity of the starch slurry as the starch granules begin to swell.
A 5 g starch sample (anhydrous basis) is placed in the analysis cup of a Model RVA-4 Analyzer and slurried in water at 20% solids. The total charge is 25 g. The cup is placed into the analyzer, rotated at 160 rpm, and heated from an initial temperature of 50~C up to a final temperature of 80~C at a rate of 3~C/minute. A
plot is generated showing time, temperature, and viscosity in centipoises (cP). The pasting temperature is the temperature at which the viscosity reaches 500 cP.
Both pasting temperature and pasting time are recorded.

W O 96/23104 PCTAUS9G,'~C5 Differential Sc~nnin~ Calorimetrv (DSC) This test provides a quantitative measurement of the enthalapy ( H) of the energy transformation that - occurs during the gelatinization of the starch granule.
The peak temperature and time required for gelatinization are recorded. A Perkin-Elmer DSC-4 differential ~C~nni ~g calorimeter with data station and large volume high pressure sample cells is used. The cells are prepared by weighing accurately lO mg of starch (dry basis) and the appropriate amount of distilled water to approximately equal 40 mg of total water weight (moisture of starch and distilled water). The cells are then sealed and allowed to equilibrate overnight at 4~C before being scanned at from 25-150~C at the rate of 10~C/minute. An empty cell is used as the blank.
Brookfield Viscometer Procedure Test samples are measured using a Model RVT
Brookfield Viscometer and the appropriate spindle (the spindle is selected h~e~ on the anticipated viscosity of the material). The test sample, usually a cooked starch paste, is placed in position and the spindle is lowered into the sample to the appropriate height. The viscometer is turned on and the spindle is rotated at a constant speed (e.g., 10 or 20 rpm) for at least 3 revolutions before a reading is taken. Using the appropriate conversion factors, the viscosity (in centipoises) of the sample is recorded.
Calcium Carbonate Retention The various samples were tested an alkaline Dynamic Retention Evaluation using a Britt Jar (modified TAPPI T26 pm 79 method) at 0.75% addition level. The retention was compared with that of a cationic waxy corn starch reacted with diethylaminoethyl chloride hydrochloride or an amphoteric waxy corn starch reacted with diethyl aminoethyl chloride hydrochloride and sodium W O 96/23104 PCTrUS~ C9 tripolyphosphate. The test was run while mixing and agitating using a Britt Jar with a screen having holes 76 microns in diameter.
To simulate a simple papermaking system, a sample of 500 ml of 0.5% by weight pulp stock was placed in the jar and agitated at about 800 rpm. Alum, at 0.5 wt. % of dry fiber in the pulp stock, was added and mixed at 400 rpm for one minute and then the mixing was increased to 1000 rpm. The starch, at 0.75 wt. % of dry fiber, was then added and mixing was continued for another minute.
To simulate a microparticle papermaking system, a sample of 500 ml of 0.5 wt. % by weight pulp stock was placed in the jar and agitated at about 800 rpm. Alum, at 0.25 wt. % of the of the dry fiber in the pulp stock, was added and mixed at 400 rpm for one minute and then the mixing was increased to 1000 rpm. The starch at 0.75 wt. % of the dry fiber, was then added and mixing was continued for another minute. Colloidal silica was added at 0.15 wt. % of the dry fiber and the sample was mixed for another minute.
After the addition/mixing se~uence, a 100 ml sample was collected by removing the clamp. The sample was acidified with 5 N hydrochloric acid to solubilize the calcium carbonate and then filtered onto tared filter paper to recover the fine solids. A standard water hardness titration was run by adding Eriochrome Black "T"
indicator and titrating with 0.1 N of the disodium salt (EDTA) ethylene diamine tetra-acetic acid, disodium salt) stAn~ard solution to a blue endpoint, using a calibrated burette. Identical titrations were made on an acidified portion of the starting pulp sample and a 25 ml sample of 100 ppm hardness water used and with this information the percent retention of calcium carbonate (CaC03) was determined using the following formula:

~ ~ 02211372 1997-07-24 _ ~37 ~EpLAcE~r~ PAG~
/ ~ CaCO3 Retention = (P-W) - (S-W) x 100, ( P -W) / where P is ml EDTA ~or pulp stock w is ml EDTA ~or raw ~ 5 water blank, and S is ml EDTA for sample.
¦ ~rainaae Resistance ¦ The drainage resistance test was performed on the furnish using a turbulent pulse sheet former ~TPSF), which is a modified Britt jar that incorporates air and vacuum to simulate the dynamics o~ an industrial~ er making machine. Furnish (200 ml) at 0.5~ consistency was diluted to one liter with water and added to the TPSF
to simulate a 0.118 kg/meter2 (80 lb/3330 sq ft) ~ine paper grade. The following were then added to the furnish in the order recited and mixed for 30 seconds at loOo rpm after each addition: alum (0.25 wt. ~ of dry fiber used in the pulp stock), starch (0.75 wt. ~~ dry ~i~er), colloidal silica (0.15 wt. ~ dry fiber). The conditions for the paper formation were: air pressure of 20 in. o~ vacuum pressure of 10 in. Hg; total pulse count for drawing air and vacuum, 3. The resistance of water drainage from the furnish and additives was measured for each sample and a comparison made using the cationic starch as the control. Comparative values are given in the following table measured against the cationic starch as a control with a value of 1~0.
Bond Strencth After the drainage resistance test, the sheets formed on the TPSF were pressed and dried tested for Scott Bond strength. The sheets were conditioned overnight at 21.5~C and 50~ relative humidity. TAPPI
test procedure T 541 om-89 were used. Comparative values are set out in the following tables measured against the cationic starch as a control with a value of 100.

Ns12s36s.l wos6r23104 PCT~US96/0-9~9 ~PLES
The following examples will more fully illustrate the embodiments of the invention. In the examples, all parts are given by weight and temperature are in degrees Celsius unless otherwise noted. The thermally-inhibited starches and controls in the following examples were prepared as described above and are defined by textural characteristics or in relation to data taken from 8r~h~n~e~ curves using the above described procedures. The thermally-inhibited starches and flours are referred to as "T-I" starches and flours and the conditions used for their preparation (i.e., pH
to which the starch is adjusted and heat treatment temperature and time at that temperature are included in parenthesis - (pH; temperature/hold time at that temperature). All pH adjustments are done with sodium carbonate unless specified otherwise. Unless otherwise specified, the thermally-inhibited starches and flours referred to as "grAntllAr" starches are non-pregelatinized granular starches and flours.
In the first three examples, the moisture indicated is the moisture of the starch before the dehydration and heat treating steps. As indicated above, as the starches were brought from ambient temperature up to the heating temperature, the starches became anhydrous or substantially anhydrous.
In the tables the abbreviations"sl.", "mod.", "v.", "ris." and "N.D." stand for slight or slightly, moderate or moderatly, very, rising, and not determined.

This example illustrates the preparation of the starches of this invention from a commercial granular waxy maize base starch by the heat treatment process of this invention.

L

3 9 REPI,ACEMEN7r PAGE
Processing conditions and their e~ects on viscosity and texture of waxy maize starch are set forth in the Tables below.
To obtain a heat-stable, non-cohesive thickener, samples of granular starch were slurried in 1.5 parts of water, the pH of the slurry was adjusted with the addition o~ a 5~ Na2CO3 solution and the slurry was agitated for 1 hour, then ~iltered, dried, and ground. The dry starch samples (150 g) were placed into o an aluminum foil pan a 18 cm. x 23 cm. x 7 cm. ~4 in. x 5 in. x 1.5 in.) and heated in a conventional oven under the conditions described in Ta~les I and II. Brabender viscosity measurements demonstrated that the most heat-sta~le starches were obtained by heating at 160 C and a pH of at least 8.0 for about 3.5 to 6.0 hours.
TABLE I - Process Variables Waxy Cold Evaluation of Maizea HeatinG - 160~Çelatinized SamPlesd~e Mois-~H ture Ti~e Viscositv Texture (~) (hrs.) 1 6.0 10.9 2 heavy to v. cohesive heavy 2 6.0 10.9 4 thin to mod. --3 8.2 ~10.6 3.5 heavy to v. cohesive, heavy less than unmodified control 4 8.2 10.6 4 heavy to v. sl. to heavy mod.
cohesive Nsl2s36s.1 AM~NDE~ S~EET

WO 96123104 PCT~US96~ 9!9 8.2 10.6 4.5 heavy non-cohesive 6 8.2 10.6 5.5 heavy, non-~h;nn~t cohesive 7 8.2 10.6 6 mod. heavy non-cohesive unmod- -- -- v. heavy cohesive ifiedb cross- -- -- v. heavy non-linkedC cohesive a. All samples were commercial samples of granular waxy maize starch obtained from National Starch and Chemical Company, Bridgewater, New Jersey.
b. The unmodified control was a commercial granular waxy maize starch obtained from National Starch and Chemical Company, Bridgewater, New Jersey.
c. The modified control was a commercial crosslinked lS (phosphorous oxychloride treated) granular waxy maize starch obtained from National Starch and Chemical Company, Bridgewater, New Jersey.
d. Samples were cooked by slurrying 7.0 g of starch (at 12% moisture) in 91 mls water at neutral pHs and heating the starch slurry for 20 minutes in a boiling water bath.
e. The cold evaluation was carried out at 25 C.

.

W O 96123104 PCTAUS~C/0~9 ~Rr~ Brabender Evaluation Waxy Br;~h~.n~lçr - Maizea Process Variables ViscositYb (BU) Viscosity Heating Peak at 95~C/
~emp. Time Viscositv 20 mins.
(~C) (hrs.) S 3 8.2 160 3.5 985 830 4 8.2 160 4.0 805 685 8.2 160 4.5 640 635 6 8.2 160 5.5 575 570 Unmodified -- none none 1640 630 control 1 6.0 160 2.0 1055 560 2 6.0 160 4.0 140 80 a. See Table I for a description of samples.
b. In the Brabender procedure, a sample cont~i n i ng 5.4%
anhydrous solids of starch dispersed in water was heated rapidly to 50~C, then the heat was increased by 1.5~C per minute to 95~C, and held for 20 minutes.
~MPLE 2 This example illustrates that a variety of granular starches may be processed by the method of this invention to provide a non-cohesive thickener with properties similar to chemically crosslinked starches.
Processing conditions and their effects on the viscosity and texture of waxy barley, tapioca, V.O.
hybrid and waxy rice starches are set forth in the tables below.

W O96123104 PCT~US96/00999 ~ABLE III - Process Variables Cold Evaluation of Samplea ~eatinq - 160~C Gelatinized Sampleb Mois-E~ ture Time Viscosity/Texture (%) (hrs.) W~yy BarleY Starch 1 8.7 8.5 1.5 heavy cohesive 2 8.7 8.5 2.5 heavy sl. mod.
cohesive 3 8.7 8.5 3.5 mod. heavy non-to heavy cohesive 4 5.2 10.8 1.5 thin --5.2 lO. 8 2. 5 thin/ --~h;nnect Waxy -- -- 0 heavy cohesive Barley Control Tapioca Starch 6 8.8 10. 3 2 heavy to v. cohesive heavy 7 8.8 10.3 3 heavy to v. cohesive/
heavy less than Sample 6 8 8.8 10. 3 4 heavy to v. sl.
heavy cohesive to sl.
lumpy W O 96/23104 PCTrUS96100999 9 8.8 10.3 5 heavy non-cohesive lumpy 5.5 10.9 3 mod. heavy --Tapioca -- -- 0 v. heavy cohesive Control 5 W~v Rice Starch 1 9.1 9.0 2 v. heavy cohesive 2 9.1 9.0 3 heavy sl.
cohesive 3 9.1 9.0 4 heavy sl.
cohesive 4 9.1 s.o 5 mod. heavy non-to heavy cohesive 10 Waxy -- -- 0 v. heavy cohesive Rice Control a. Tapioca starch samples were commercial granular starch obtained from National Starch and Chemical Company, Bridgewater, New Jersey. Waxy barley starch samples were commercial granular starch obtained from AlKo, Finland. Waxy rice starch samples were commercial granular starch obtained from Mitsubishi Corporation, Japan.~0 b. Samples were cooked by slurring 7.5 g of starch at 12% moisture in 100 mls of water and heating the starch slurry for 20 minutes in a boiling water bath.

W O96/23104 PCTrUS9~'~U~9 TABLE IV - Process Va~iables Cold Evaluation of $am~1eHeating - 160~C Gelatinized Sampleb Mois-E~ ture Time ViscositY/Texture (%) (hrs.) VØ Hybrid Starch 1 8.7 10.5 2.0 heavy cohesive v.
sl. less than control 2 8.7 10.5 3.0 heavy sl. mod.
cohesive 3 8.7 10.5 4.0 mod. heavy smooth, to heavy very sl.
cohesive 4 8.7 10.5 5.0 mod. heavy smooth, short, non-cohesive 8.7 10.5 6.0 moderate smooth, short, non-cohesive VØ 5.9 11.4 0 heavy cohesive Hybrid Control a. V.o. hybrid starch samples were granular starches obtained from National Starch and Chemical Company, Bridgewater, New Jersey.
b. Samples were cooked by slurrying 7.5 g of starch at 12% moisture in 100 mls of water and heating the starch slurry for 20 minutes in a boiling water bath.

W O96/23104 PCTAUS~6/~09 The viscosity and texture evaluation results show that a non-cohesive, heat-stable starch thi~-kener may be prepared from waxy barley, V.o. hybrid, tapioca - and waxY rice starches by the process of this invention.
The amount of inhibition (non-cohesive, thi~ke~;ng character in cooke~ aqueous dispersion) increased with increasing time of heat treatment.
~AMPLE 3 This example illustrates the effects of temperature, the pH, and starch moisture content on the viscosity and texture of the treated starch.
Part A
A waxy maize starch sample (100 g) contAin;ng 20.4% moisture was heated in an oven at lOO-C for 16 hours in a sealed glass jar. A second sample was heated for 4 hours and a third sample was heated for 7 hours under the same conditions. The product viscosity and texture were compared to a 12.1% moisture granular waxY
maize starch control using the cook evaluation method of Example 1, Table I. Results are shown in Table V, below.
T~RT~ V - Effect of Process Moisture Waxy Process Cold Evaluation of Maizea Variablesb Gelatinized StarchC
Heat Time viscositY Texture (hrs.) 1. Test 16 heavy, sl. cohesive (20.4% H20) thinner than control 2. Control 0 heavy cohesive (12.1% H20) 3. Test 4 heavY cohesive (20.4% H20) : CA 022ll372 l997-07-24 Waxy Process Cold Eva~uation of Maizea Variablesb Gelatinized StarchC
Heat Time Viscositv Texture (hrs.) 1. Test 16 heavy, sl. cohesive (20.4~ H2O) thinner than control 2. Control 0 heavy cohesive (12.1~ H2O) 3. Test 4 heavy cohesive (20.~% H2O) lo 4. Control 0 heavy cohesive (12.1~ H2O) 5. Test 7 heavy cohesive (20. 4% H2O) 6. Control 0 heavy cohesive (lZ.l~ HzO) a. Samples were obtained from National Starch and Chemical Company, Bridgewater, New Jersey.
b. Process was conducted at pH 5.2.
c. See Table III for cook conditions.
The results demonstrate that moisture added during the process yields a product which is as cohesive and undesirable as a control starch which had not been heated.
Part B
Samples (900 g) of a commercial granular waxy maize starch (obtained from National Starch and Chemical Company, Bridgewater, New Jersey) were placed in a 45 cm.
x 68 cm. x 3.4 cm. (10 in. x 15 in. x 0.75 in.) aluminum tray and heated in an oven at 180 C for 15, 30, 45 and 60 minutes. The pH of the starch was not adjusted and NB129365.1 _ AMEI~IC~D S~I~ET

R~PL~CEMENT PAGE

remained at about 5.2 during the heating process. Sample viscosity and texture were evaluated by the method o~
Exa~ple 1.
As shown in Table VI, below, the pH 5.2 samples were characterized by an undesirable, cohesive texture ~3129365.1 c~ S~E~J

W 096r23104 PCTnUS96/00999 similar to that of a waxy maize starch control which had not been heat treated.
~ARr~ VI - Effect of Acidic Process ~H
Process Cold Evaluation of Sam~le Variablesa Gelatinized Starchb Heating Time Viscositv Texture (minutes) 1 15 v. heavy cohesive 2 30 v. heavy cohesive 3 45 v. heavy cohesive 4 60 heavy to cohesive v. heavy control 0 v. heavy cohesive a. The pH was not adjusted from that of the native waxy maize starch (a pH = 5.2) and Samples 1-4 correspond to starch treated by the process of U.S. Pat. No. 4.303,451 (no pH adjustment).
b. See Table III for cook conditions.
Thus, a combination of selected factors, including the pH, moisture content and the type of native starch, determine whether a desirable, non-cohesive, heat-stable starch thickener is produced by the process of this invention.
~MPLE 4 This example shows carrying out the thermal inhibition in the fluidized bed previously described.
The effects of temperature and time at the indicated temperature on the level of inhibition of waxy maize granular starch at pH 9.5 are shown below.

W O ~6/23104 PCTrUS96100999 Viscosity tB.U.) He~tinq Temperature and Time Peak Break-~eak + 10' down (%) Control (none) 1135 730 64.3 110~C for 22 hrs. 1185 970 18.1 160~C for 0 hr. 1055 880 16.6 160~C for 2 hrs. 665 660 0.7 175~C for 0 hr. 850 755 11.2 180~C for 0 hr. 715 680 4.9 190~C for 0 hr. 555 550 o.g 200~C for 0 hr. ris. -- --200~C for 2 hrs. none -- --The data shows that inhibited anhydrous or substantially anhydrous samples can be obt~; neA at heat treating temperatures between 100-200~C, with more inhibition obtained at higher temperatures or at longer times at lower temperatures. The starch samples heated at 200~C were highly inhibited (rising curves) or completely inhibited (no gelatinization).
~MP~E 5 Samples of a high amylose starch (Hylon V - 50%
amylose) at its natural pH and pH 9.5 were evaluated for the effect of the high amylose content on inhibition.
The starches were thermally-inhibited at 160~C in the fluidized bed for the indicated time. Due to the high levels of amylose, it was necessary to use a pressurized Visco/amylo/Graph (C.W. Brabender, Hackensack, NJ) to obtain Brabender curves. Samples were slurried at 10%
starch solids, heated to 120~C, and held for 30 minutes.

CA 022ll372 l997-07-24 W 096~3104 PCTJU~ 9 The results are shown below:
Natural ~H pH 9.5 Viscosity ~BU)Viscosity (BU) Peak Break- Peak Break--Peak+ 10~down Peak+ 10'down (%) (%) Control 1180525 55.5 1180525 55.5 (0 min.) (120 min.) The data show that inhibition was obt~;ne~l only on the high pH sample.
E~MPLE 6 This example shows the preparation of pregelatinized granular, thermally-inhibited waxy maize starches. The pregelatinization step was carried out prior to the thermal inhibition. The fluidized bed described previously was used.
Starch slurries (30-40% solids), pH adjusted to 6, 8, and 10, were pregelatinized in a pilot size spray drier (Type-l-KA#4F, from APV Crepaco, Inc., Dryer Division, of Attle Boro Falls, Massachusetts) using a spray nozzle, Type 1/2 J, from Spraying Systems Company of Wheaton, Illinois. The spray nozzle had the following configuration: fluid cap, 251376, and air cap, 4691312.
The resulting high and low viscosity pregelatinized granular starches were dehydrated and heat treated at the temperature and time indicated. The thermally-inhibited starches were evaluated for inhibition using the Brabender procedure previously described.
The results are shown below:

W O96123104 PCT~US~/0~5 Heat Treatment CQnditions ViscositY (B.U.~
30~C 95~C Break-30~C+lO'~eak95~C+10' down (%) pH 6.0 - High Xnitial Viscosity Control 1280 960 960 170 90 91 160~C for 0 min. 700 980 700 610370 47 160~C for 30 min. 600 910 720 690370 49 160~C for 90 min. 450 780 915 740400 56 160~C for 150 min.360 590 925 800500 46 10 pH6.0-Low Initial Viscosity Control 230 250 750 340100 87 160~C for 30 min. 100 130 600 370210 65 160~C for 60 min. 100 140 730 500260 64 160~C for 120 min.100 130 630 430260 59 160~C for 180 min. 90 120 550 390240 56 pH8.0- High Initial Viscosity Control 1400 1020 1020 270100 90 160~C for 0 min. 700 1060 1050 760280 73 160~C for 60 min. 260 600 1340 1200780 42 160~C for 90 min. 240 440 1280 12401000 22 160~C for 120 min.280 420 1320 13201280 3 160~C for 150 min.120 200 860 860820 7 160~C for 180 min.180 260 980 980920 8 pH8.0-Low Initial Viscosity Control 250 250 820 340130 84 160~C for 0 min. 50 100 690 460270 61 160~C for 60 min. 40 50 840 590320 62 160~C for 120 min. 20 30 720 650450 38 160~C for 180 min. 20 30 590 570450 24 30 pH 10 - High Initial Viscosity Control 1010 740 1010 300160 84 140~C for 0 min. 550 850 1280 1080750 41 150~C for 0 min. 270 420 1680 16801540 8 W O 96/23104 PCT~U5gG, 160~C for 0 min. 170240 --1180 1440 ris.
160~C for 30 min. 80 85 -- 410 650 ris.
160~C for 60 min. 60 60 -- 150 300 ris.
- 160~C for 90 min. 50 50 -- 80 140 ris.
120~C for 120 min. 40 40 -- 80 130 ris.
150~C for 150 min. 40 40 -- 60 90 ris.
160~C for 160 min. 40 40 -- 45 70 ris.
pH 10 - Low Initial Viscosity Control 200190615 350 190 69 130~C for 0 min. 110 180 1500 880 530 65 150~C for 0 min. 50 80 1670 1540 1250 25 160~C for 0 min. 30 30 --1040 1320 ris.
160~C for 30 min. 30 30 -- 380 640 ris.
160~C for 60 min. 30 30 -- 150 310 ris.
160~C for 90 min. 10 10 -- 50 120 ris.
The results show some thermal inhibition was attained in all the dehydrated and heat treated pregelatinized granular starches and that increasing the initial pH and the heat treatment time increased the level of inhibition. For the samples at pH 6.0, at 0 and 30 minutes, the recorded peak was actually a second peak obtained after the initial high viscosity began to breakdown. For some of the samples at pH 10, no peak viscosity was reached, indicating a highly inhibited starch.
~MPLE 7 This example describes the preparation of thermally-inhibited pregelatinized granular starches from additional starch bases as well as a waxy maize starch.
The granular starches were adjusted to the indicated pH, pregelatinized using the procedure previously described, and heat treated in an oven at 140~C for the indicated time. The cook evaluation and Brabender results are shown below.

W O96/23104 PCTrUS9G~ 999 Cook Evaluation ~eat Treatment Viscosity ~ ~ours at 140~C of CookTexture of Cook 5 WAYY Maize 6 2 mod. sl. cohesive, smooth 6 4 mod. to thin sl. cohesive, smooth 6 6 mod. v. sl. cohesive, smooth 6 8 mod. v. sl. cohesive, smooth 8 2 mod. cohesive, smooth 8 4 mod. to heavy sl. cohesive, smooth 8 6 mod. v. sl. cohesive, smooth 8 8 mod. v. sl. cohesive, smooth 2 mod. sl. cohesive, smooth 4 mod. to heavy non-cohesive, short, smooth 6 mod. non-cohesive, short, smooth 8 mod. non-cohesive, short, smooth Ta~ioca 6 2 mod. to heavy v. cohesive, long 6 4 mod. to heavy cohesive W O 96/23104 PCTnU5n~'00~

6 6 mod. sl. cohesive, smooth 6 8 mod. non-cohesive, short, smooth 8 2 mod. to heavy v. cohesive 8 4 mod. to neavy cohesive 8 6 mod. to heavy non-cohesive, short, smooth 8 8 mod. to heavy non-cohesive, short, smooth 2 mod. to heavy cohesive, long lo 4 mod. to heavy v. sl. cohesive, smooth 6 mod. non-cohesive, short, smooth 8 mod. to heavy non-cohesive, short, smooth Potato 6 2 heavy to v. cohesive, long v. heavy 6 4 heavy cohesive 6 6 mod. to heavy sl. cohesive 6 8 mod. to heavy v. sl. cohesive 8 2 heavy to v. cohesive, long v. heavy 8 4 v. heavy sl. cohesive 8 6 heavy non-cohesive, sl. set, smooth Wos6/23104 PCT~S96/00999 8 8 mod. non-cohesive, v. sl.
set, smooth 2 heavy v. cohesive 4 heavy to mod. sl. cohesive, v. sl. set, smooth 6 heavy to mod. non-cohesive, short, mod. set, smooth 8 heavy to mod. non-cohesive, short, mod. set, smooth Heat Treatment Conditions ViscositY (BU) Breakdown 30~C 30~C+10' Peak 95~C95~C+10' (%) Waxy Maize at pH 8 and 140~C
2 hrs 400 1115 1115 515515 60 6 hrs 400 955 1120 11201023 38 TaPioca at ~H 8 and 140~C
2 hrs 1140 2685 2685 2685880 78 6 hrs 370 800 1110 1110890 46 The results show that thermally-inhibited pregelatinized granular starches can be prepared using other starch bases and that for non-cohesive starches longer times and/or higher pHs are required when an oveno rather than a fluidized bed is used for the dehydration and heat treatment.

This example shows the preparation of pregelatinized, non-granular starches which were pregelatinized by drum-drying and then thermally inhibited.

W O96/23104 PCTAUSg~'~O~gg Samples of waxy maize, tapioca and potato starches, at pH 6, 8, and lo, were pregelatinized by drum-drying. The samples were placed in a 140~C oven, dehydrated to anhydrous, and heat treated at 140~C for the indicated times.
The viscosity and textural characteristics of the thermally-inhibited starches are set out below.
im~ Cook Viscositv Cook Texture WAYY Maize - ~H 6 10 2 hrs heavy v. cohesive, pulpy 4 hrs heavy to v. heavy cohesive, pulpy 6 hrs heavy sl. cohesive, pulpy 8 hrs mod. to heavy v. sl. cohesive, pulpy W;~YY M;~; ze -- pH 8 15 2 hrs heavy v. cohesive, pulpy 4 hrs heavy sl. cohesive, pulpy 6 hrs mod. to heavy v. sl. cohesive, pulpy 8 hrs mod. to heavy v. sl. cohesive, pulpy Waxy Maize - pH lQ
20 2 hrs heavy cohesive, pulpy 4 hrs heavy to mod. v. sl. cohesive, pulpy 6 hrs mod. non-cohesive, short, pulpy 8 hrs mod. non-cohesive, short, pulpy Tapioca - PH 6 25 2 hrs v. heavy cohesive, pulpy 4 hrs heavy to v. heavy sl. cohesive, pulpy 6 hrs mod. heavy sl. cohesive, pulpy W O96/23104 PCTrUS96/00999 8 hrs heavy sl. cohesive, pulpy T~ioca - ~H 8 2 hrs heavy to v. heavy v. cohesive, pulpy 4 hrs heavy v. cohesive, pulpy 5 6 hrs N.D. N.D.
8 hrs heavy v. sl. cohesive, pulpy Ta~;oca 10 - ~H
2 hrs heavy cohesive, pulpy 4 hrs heavy to v. heavy sl. cohesive, pulpy 10 6 hrs heavy non-cohesive, short, pulpy 8 hrs mod. heavy non-cohesive, short, pulpy Potato - pH 6 2 hrs heavy to v. heavy cohesive, pulpy 4 hrs heavy cohesive, pulpy 15 6 hrs mod. to heavy cohesive, pulpy 8 hrs mod. to heavy cohesive, pulpy Potato - pH 8 2 hrs heavy to v. heavy v. cohesive, pulpy 4 hrs v. heavy cohesive, pulpy 20 6 hrs v. heavy cohesive, pulpy 8 hrs v. heavy cohesive, pulpy Potato - pH 10 2 hrs heavy to v. heavy v. cohesive, pulpy 4 hrs v. heavy slight set, sl. chunky 25 6 hrs heavy slight set, sl. chunky W O 96123104 PCTnUSg~ 9~g 8 hrs mod. heavy moderate set, sl.
chunky Brabenders were run on some of the above starches. The results are shown below.
Viscosity (B.U.) 30~C 95~C Break-30~C +lo~ Peak 95~C +10' down (%) Waxy Maize - pH 8 2 hrs. 665 3000 46201120 300 94 6 hrs. 700 1640 24452440 1900 22 Tapioca - pH 8 2 hrs. 1500 3170 3290680 600 82 6 hrs. 1180 1870 1873780 600 68 The results show that longer heating times and/or higher pHs are required to prepare non-cohesive starches at 140~C. It is expected that heating at 160~C, preferably in a fluidized bed, will provide non-cohesive starches.
ExAMpT~ g This example shows the preparation of another pregelatinized non-granular starch which was jet-cooked, spray-dried, and then thermally inhibited.
A granular high amylose starch (50% amylose) was jet-cooked and spray-dried using the continuous coupled jet-cooking/spray-drying process described in U.S. 5,131.953 and then thermally inhibited for 8 hours at 140~C. The jet-cooking/spray-drying conditions used were as follows: slurry - pH 8.5-9.0; cook solids - 10%;
moyno setting - about 1.5; cooking temperature - about 145~C; excess steam - 20%; boiler pressure - about 85 psi; back pressure - 65 psi; spray-dryer - Niro dryer;
inlet temperature - 245~C; outlet temperature - 115~C;

W O96123104 PCTrUS96/00999 atomizer - centrifugal wheel. The pregelatinized non-granular starch was adjusted to pH 8.7 and dehydrated and heat treated for 8 hours in an oven at 140~C. The characteristics of the resulting thermally-inhibited starches are set out below.
Viscositv (BU) 30~C 30~C~10' Peak 95~C 95~C+10' Breakdown (%) Control 200 195 245 245 130 47 High 350 240 420 410 335 20 Amylose The results show that even a high amylose starch can be inhibited. There was less breakdown for the thermally-inhibited starch and the overall viscosity was higher.
~MPLE 10 This example shows that thermally-inhibited waxy maize starches can be prepared by drum drying the starches prior to thermal inhibition. The resulting non-granular thermally-inhibited drum-dried starches are compared with the non-granular thermally-inhibited waxy maize starches prepared by the continuous coupled jet-cooking and spray-drying process used in Example 8 and with granular thermally-inhibited starches prepared by the dual atomization/spray drying process described in Y.S. 4.280.251 (which was used in Example 6). The conditions used for the oven dehydration and heat treatment were 8 hours at 140~C.
The characterization of the resulting thermally-inhibited pregelatinized starches is shown below.

W O 96/23104 PCTAUS96~!G~9~g ~um-Dried/Non-Granular T-I Waxy Maize (~H 8) Viscositv (BU) 30~C 30~C+10' Peak 95~C 95~C+10' Breakdown (96) Control 640 2770 3530 1690 1550 56 ~et-Cooked/S~raY-Dried/Non-Granular T-I Waxy Maize - PH 8 Control 60 90 100 41 30 70 Steam Atomized/S~ray-Dried/Granular T-I Waxy Maize - ~H 8 Control 100 1010 1080 340 170 84 lo T-I 360 950 970 860 650 33 The results show that after 8 hours heat treatment at 140~C all the pregelatinized thermally-inhibited starches showed much less breakdown. The results also show that a higher degree of inhibition along with a higher peak viscosity can be obtained if the starch granules are completely disrupted as by drum drying or jet cooking.
~MPLE 11 This example shows that a granular starch can be dehydrated by ethanol extraction and that a better tasting starch is obtained.
A granular waxy maize starch was slurried in 1.5 parts water based on the weight of the starch and adjusted to pH 7 and 9.5 with 5% sodium carbonate, held 2S for 30 minutes, filtered, and dried on a tray to a moisture content of about 5-6% moisture. The starch having the pH of 5.3 was a native starch which was not pH
adjusted.
For the dehydration, the dried pH 5.3, pH 7.0, and pH 9.5 starches were each separated into two samples.
One sample was dried on trays in a forced draft oven at W O96/23104 PCTrUS96/00999 80~C overnight to thermally dehydrate the starch to <1%
(o%) moisture. The other sample was placed in a Soxhlet extractor and allowed to reflux overnight (about 17 hours) with anhydrous ethanol (boiling point 78.32~C).
The ethanol-extracted sample was placed on paper so that the excess alcohol could flash off which took about 30 minutes. The ethanol-extracted starch was a free flowing powder which was dry to the touch.
For the heat treatment, the oven-dehydrated starches and ethanol-extracted starches were placed on trays in a forced draft oven and heated for 3, 5, and 7 hours at 160~C.
The thermally-inhibited (T-I) starches and the controls were evaluated using the BrAhen~e~ Procedure previously described was used. The results are shown below:
R~ABENDER RESUT'r'S
ViscositY (BU) Dehydra- Heat tion Treat- Peak Break-Method ment ~ + 10' down (160~C) WaxY Maize (pH 5.3) Control -- -- 1245 330 74 Dehydrated oven -- 1290 350 73 Dehydrated ethanol -- 1205 245 80 T-I oven 5 hrs. 95 45 53 T-I ethanol 5 hrs. 255 185 28 T-I oven 7 hrs. 60 35 42 T-I ethanol 7 hrs. 165 105 36 CA 022ll372 l997-07-24 W O 96/23104 PCTnUS9~ 9 Waxv Maize (~H 7.0) Dehydrated oven -- 1240 380 69 T-I oven 7 hrs. 298 240 20 T-I ethanol 7 hrs. 400 310 23 5 WAYV Maize (pH 9.5) Dehydrated oven -- 1250 400 68 Dehydrated ethanol --- 1070 350 67 T-I ethanol 3 hrs. 665 635 5 T-I oven 3 hrs. 680 655 4 T-I oven 5 hrs. 245 460 ris.
T-I ethanol 5 hrs. 160 375 ris.
T-I Oven 7 hrs. 110 295 ris.
T-I Ethanol 7 hrs. 110 299 ris.
The results show that the starches can be dehydrated by ethanol extraction. The results also show that dehydration without the subsequent heat treatment did not inhibit the starch. The viscosity breakdown was not significantly different from that of the native waxy maize starch. Both of the thermally-inhibited pH 7 starches were higher in viscosity than the pH 5. 3 (as is) thermally-inhibited starches. The starches which were thermally-inhibited at pH 9.5 were moderately highly inhibited or highly inhibited (rising curve).

Granular tapioca, corn, and waxy rice starches and waxy rice flour were adjusted to pH 9.5, dehydrated in an oven and by extraction with ethanol, and heat treated at 160~C for the indicated time. They were evaluated for Brabender viscosity using the procedure previously described.

W O96/23104 PCTrUS96/00999 The Brabender results are shown below.
Viscositv (BU) Heat Treat-Dehydration ment Peak Break-S~rch Method Time Peak +10' down f%) Tapioca (pH 9.5 and 160~C) Dehydrated oven -- 745 330 58 Dehydrated ethanol -- 720 330 54 T-I oven 5 hrs. 270 260 3 T-I ethanol 5 hrs. 260 258 T-I oven 7 hrs. 110 155 ris.
T-I ethanol 7 hrs. 100 145 ris.
Corn fpH 9.5 and 160~C) Dehydrated oven -- 330 280 15 Dehydrated ethanol -- 290 250 14 T-I oven 5 hrs. 10 80 ris.
T-I ethanol 5 hrs. 10 170 ris.
T-I oven 7 hrs. 10 65 ris.
T-I ethanol 7 hrs. 10 45 ris.
Waxy Rice (pH 9.5 and 160~C) Dehydrated oven -- 1200 590 50.8 Dehydrated ethanol -- 1155 450 61.0 T-I oven 5 hrs. 518 640 ris.
T-I oven 7 hrs. 265 458 ris.
T-I ethanol 7 hrs. 395 520 ris.

W O 96123104 PCTAUS9C~'~O~

WAXV Rice Flour (~H 9.5 and 160~C) Dehydrated oven -- 895 700 22 Dehydrated ethanol -- 870 410 53 T-I oven 5 hrs. 38 73 ris.
T-I ethanol 5 hrs. 140 260 ris.
T-I oven 7 hrs. 10 16 ris.
T-I ethanol 7 hrs. 40 100 ris.
The results show that pH 9.5-adjusted, ethanol-extracted, heat-treated tapioca and corn starches had viscosity profiles generally similar to those of the same thermally-inhibited starches which were oven-dehydrated.
The 7 hours heat-treated samples were more inhibited than the 5 hour heat-treated samples.
~XA~PLE 13 This example compares ethanol extracted granular waxy maize starches and oven-dehydrated granular waxy maize starches heat treated in an oven for 5 and 7 hours at 160~C at the same pH, i.e., pH 8.03.
The Brabender results are shown below.
viscositY (BU) Dehydration/ Break-Heat Treatment Peak Peak + 10' down (~) Oven/None 1160 360 69 EtOH/None 1120 370 67 Oven/5 hrs. 510 455 11 EtOH/5 hrs. 490 445 9 Oven/7 hrs. 430 395 8 EtOH/7 hrs. 360 330 8 W O96/23104 PCTrUS9G/00999 The thermally-inhibited starches were slurried at 6.6% solids (anhydrous basis), pH adjusted to 6.0-6.5, and then cooke~ out in a boiling water bath for 20 minutes. The resulting cooks were allowed to cool and then evaluated for viscosity, texture, and color.
Dehydration Time at Method 160~C Viscositv Texture Colox Oven None heavy to cohesive sl. off-v. heavy white Ethanol None heavy to cohesive sl. off-v. heavy white 10 Oven5 hours mod. heavy non- sl. tan, to heavy cohesive, darker*
smooth Ethanol 5 hours mod. heavy non- sl. tan to heavy cohesive, smooth Oven 7 hours mod. heavy non- mod. tan, to heavy cohesive, darker*
smooth Ethanol7 hours mod. heavy non- mod. tan to heavy cohesive, smooth * Slightly darker than ethanol-dehydrated samples.
These Brabender results show that highly inhibited starches can be obtained by both thermal and non-thermal dehydration. The cook evaluation results show that there is a benefit for the ethanol-dehydrated, thermally-inhibited starches in terms of reduced color.
As will be shown hereafter, there is also a flavor improvement with ethanol dehydration.

W O 96/23104 PCTnUS~ n~g ~xAMpLE 14 A granular waxy maize starch was pH adjusted to pH 9.5 as previously described. The starch was then placed in a freeze dryer and dried for 3 days until it was anhydrous (0% moisture). The freeze-dried (FD) starch was heat treated for 6 and 8 hours at 160~C in a forced draft oven.
BrAh~n~er evaluations were run. The results are shown below:
lo Viscosity (BU) Waxy Maize Time at Break-(pH 9.5) 160~C Peak Peak + 10' down (%) Control -- 1260 320 75 F.D. -- 1240 320 74 T-I 6 hrs. 340 465 ris.
T-I 8 hrs. 285 325 ris.
The results show that the starch can be dehydrated by freeze drying and that the subsequent heat treatment is n~c~ssAry to inhibit the starch. The starches are highly inhibited as shown by their rising viscosity.
~MPLE 15 This example shows that thermal inhibition reduced the gelatinization temperature of the granular waxy maize starches.
The gelatinization temperature of an untreated waxy maize, a thermally-inhibited (T-I) waxy maize (pH
adjusted and not pH adjusted), and chemically-crosslinked (X-linked) waxy maize starches (0.02%, 0.04%, and 0.06%
phosphorus oxychloride) were determined by Differential ScAnn;ng Calorimetry. The starches were thermally W O96/23104 PCTrUS9G~ 39 dehydrated and heat treated in an oven for the indicated time and temperature.
The peak gelatinization temperature and enthalpy ( H) are shown below.
Peak Gelatinization Waxy Maize Temperature (~C) ~nthalpv (cal/q) Unmodified 74 4.3 T--I 68 2.9 (pH 9.5j160~C
for 8.5 hrs.) T-I Waxy Maize 59 2. 8 (pH 6;160~C
for 8 hrs.) 15 X-linked 73 4.4 (0.02% POC13) X-linked 72 4.2 (0.04% POC13) X-linked 74 4.2 (0.06% POC13) The results show that there was a significant reduction in peak gelatinization temperature of the thermally inhibited (T-I) starches. The heat treatment reduced the enthalpy ( H) from 4.3 cal/g for the unmodified starch to 2.8 - 2.9 cal/g for the thermally-inhibited starch. The chemically crosslinked (X-linked) starches are essentially identical to the unmodified waxy starch in peak temperature (72-740C vs. 74~C) and enthalpy (4.2-4.4 vs 4.3 cal/g). The reduced gelatinization temperature and decrease in enthalpy suggest that the overall granular structure has been altered by the dehydration and heat treatment.

W O 96123104 PCTrUS9~ 3 ~A~DPLE 16 This example shows that the thermal inhibition may begin as early as 110~C (230~F), that it is substantially noticeable at 160~ (320~F), and that the gelatinization is lm~h~nged or reduced. Gr~nlllAr waxy maize starches were pH adjusted to 7.0 and 9.5 and dehydrated and heat treated using air having a Dew point below 9.4~C (15~F) in the fluidized bed previously described at the indicated temperature and time. The Brabender and DSC results are shown below.
WaxY Maize (pH 9.5) Dehydration/
Heat Treatment Conditions Brabender Viscosity fBU) Peak Peak + 10' Breakdown (%) Control (pH 9.5) 1240 300 75.8 93~C for 0 min.1200 300 75.0 104~C for o min.1205 320 73.4 110~C for 0 min.1260 400 68.3 121~C for 0 min.1230 430 65.0 127~C for 0 min.1255 420 66.5 138~C for 0 min.1245 465 62.7 149~C for 0 min.1300 490 62.3 160~C for 0 min.1120 910 18.8 160~C for 60 min. 750 730 2.7 160~C for 90 min. 690 680 1.4 W O 96/23104 PCTrUS~61~J~3 Dehydration/ Peak Heat Treatment Gelatinization C:Qnditions Temperature ~nthalpY (cal/q) Control (pH 9.5) 74.82 4.05 127~C for 0 min. 74.84 4.17 160~C for 0 min. 73.04 4.50 160~C for 60 min. 71.84 4.60 160~C for 90 min. 70.86 4.26 * Average of 2 readings.
The DSC results show that at the onset of inhibition there was a slight reduction in the peak gelatinization temperature and that as the inhibition temperature and time increased there was a reduction in peak gelatinization temperature. The enthalpy is 15 lln~h~nged or slightly higher, unlike the enthalpy of the more highly inhibited starches of the prior example.
~X~MPLE 17 This example shows the correlation between the RVA pasting temperature and time and DSC peak 20 gelatinization temperature and time and the reduction in Brabender viscosity breakdown for various granular starch bases and for granular waxy maize starches dehydrated by various methods including heating, ethanol extraction, and freeze drying. The base starches were unmodified.
25 The starches were all adjusted to pH 9.5 before dehydration. The ethanol-extracted and freçze-dried controls were pH adjusted and dehydrated but not heat treated. The dehydrated starches were all heat treated in an oven at 160~C for the indicated time except for the 30 starches chemically crosslinked with sodium trimetaphosphate (STMP) which were heat treated at 160~C
for the indicated time in the fluidized bed previously described.

CA 022ll372 l997-07-24 WO 96t23104 PCTIUS9~J'~03~3 The results are shown below.
Starch Pasting DSC Viscosity (B.U.) Peak Peak Break Te~. ~i~ Temp. ~imÇ~ Peak +10' down (~C) (min) (~C) (min) (%) Tapioca 68.20 3.7 70.61 6.6 1595 440 72.41 C~~

DehYdrated Thermally/Heat Treated at 160~C
T--I 66.65 3.4 68.31 6.3 1230 560 54.47 (2 hrs.) ~-~ 64.20 2.9 65.41 6.0 355 335 5.63 (6 hrs.) Potato 61.05 2.3 62.67 5.8 1825 1010 44.66 Control nehv~lrated Thermally/Heat Treated at 160~C
T--I 60.25 2.1 61.41 5.6 995 810 18.59 (3 hrs.) T--I 60.20 2.1 61.13 5.6 ris. ris. ris.
(6 hrs.) Waxy 70.95 4.3 73. 86 6 .9 1215 350 71.79 Maize Control CA 022ll372 l997-07-24 W O96/23104 PCT~U3gG/'~

Dehydrated Thermally/Heat Treated at 160~C
T--I 68.15 3. 7 70.71 6.6 760 720 S.26 (8 hrs-) Waxy 70.95 4.3 74.23 6.9 1250 400 68.00 Maize Control ~h~nol DehYdrated/Heat Treated at 160~C
T-I 65.00 3.1 71. 81 6.7 ris. ris. ris.
(2 hrs.) T-I 63. 85 2. 8 68.12 6.3 ris. ris. ris.
(7 hrs.) Waxy 71.30 4.4 74.16 6.9 1240 320 74.19 Maize ~ontrol Dehydrated bY Freeze Drying/Heat Treated at 160~C
T-I 69. 50 4.0 66.09 6.1 ris. ris. ris.
(6 hrs.) T-I 66.75 3.5 64.64 6.0 ris. ris. ris.
(8 hrs.) Cross- 71.70 N.D. 74. 33 6.9 ris. ris. ris.
linked Waxy Maize Control W O 96/23104 PCT~US~

Thermally Dehydrated Crosslinked Waxy Maize~
T-I 69.10 N.D. 71.66 6.7 ris. ris. ris.
(30 5 min.) T-I 66.00 N.D. 67.14 6.2 ris. ris. ris.
(150 min.) * Fluidized bed.
The results show that heat treatment of thermally and non-thermally dehydrated granular starches reduced the pasting and peak gelatinization temperatures while at the same time inhibiting the viscosity breakdown. Because the gelatinization temperature has been lowered by the heat treatment of the dehydrated starch, less time is required to reach the pasting and gelatinization temperatures. ~he more highly inhibited starches showed a lower pasting t~mr~ature and less breakdown in viscosity.
~PLE 18 A granular waxy maize starch which had been lightly crosslinked with 0.04% phosphorous oxychloride was thermally-inhibited. The granular starch was jet-cooked and spray-dried using the coupled continuous jet-cook;ng/spray-drying process and conditions described in Example 8. The spray-dried starch was oven dehydrated and heat treated for 8 hours at 140~C.
The Brabender results and viscosity and textural characteristics of the resulting thermally-inhibited starch are set out below.

W O96/23104 PCTrUS9~J'~J~9 Viscosity (BU) 30~C 95~C Break-30~C +10' Peak 95~C +10' down (%) Crosslinked 150 165215 120 70 67 Waxy Maize Control Crosslinked Waxy Maize Starch Cook Evaluation Viscosity 10of Cook Texture of Cook Crosslinked Waxy thin to moderate cohesive, pulpy Maize Control T-I Crosslinked very heavy non-cohesive, Waxy Maize very pulpy, short Starch The results show that after the dehydration and heat treatment steps the crosslinked starch was very highly inhibited.

This example shows the thermal inhibition of converted starches.
Samples of waxy maize and tapioca starch were slurried in 1.5 parts water. The slurries were placed in a 52 C water bath, with agitation, and allowed to equilibrate for one hour. Concentrated hydrochloric acid (HCl) was added at 0.8% on the weight of the samples.
The samples were allowed to convert at 52~C for one hour.
The pH was then adjusted to 5.5 with sodium carbonate, then to pH 8.5 with sodium hydroxide. The samples were CA 022ll372 l997-07-24 recovered by filtering and air drying (approximately 11%
moisture). The starches in 50g amounts were placed in an aluminum tray, covered and placed into a forced draft oven at 140~C for 5.5 hours. The starches were evaluated for inhibition.
The results set out in the following table.
Waxy Maize Tapioca ViscositY ~BU) ViscositY fBU) Peak Break- Peak Break-Peak + 10' down Peak ~ 10' down (%) (96) unmodified 1380 250 81.9 810 225 72.2 acid 640 110 82.3 432 115 73.4 lo converted T-~ acid 805 728 9. 6 495 350 29.3 converted The results show that converted starches can be thermally inhibited by this process.
~MPLE 20 Waxy maize samples reacted with 7% and at 3% by weight propylene oxide (P0), at the naturally occurring pH and at pH 9.5, were evaluated for inhibtion.
The results set out in the following tables.
ViscositY (BU) Temp Time Peak Peak + 92~C 92~C ~ Breakdown (~C) (min) 10' 30' (%) Waxy Maize (7~ PO and natural pH at 160~C~
Control -- 1420 395 -- -- 72 WO96/23104 PCTrUS~6/~0~39 160 60 685 430 -- __ 37 160 120 620 340 -- __ 45 WA~Y Maize (7% PO and ~H 9. 5 at 160~C) Control -- 1420 395 -- -- 72 160 90 910 890 -- ~ -- 2 15Waxy Maize r396 PO and natural pHat 160~C) Control -- 1155 280 -- -- 76 160 30 570 370 -- -_ 35 WaxY Maize r3~ P0 and ~H 9.5 at 160~C) 25Control -- 1155 280 -- -- 76 WO96/23104 PCT/u~3~/00999 160 o 1220 960 -- -- 21 160 90 -- -- 750 790 ris.
160 120 ~ 620 780 ris.
160 150 -- -- 510 750 ris.
160 180 -- -- 400 700 ris.
The data show that derivatized starches, in this case etherified starches, can be thermally inhibited by this process and that higher inhibition can be achieved at higher pH.
EXAMP~E 21 A converted hydroxypropylated waxy maize starch (25 WF starch reacted with 2% propylene oxide) was adjusted to pH 9.5 and thermally inhibited using the fluidized bed previously described. Samples were taken at 110~C, 125~C, and 140~C, all for o minutes.
The thermally-inhibited starch samples were cooked in tap water at 88-93 C (190-200 F) bath temperature for 30-60 minutes to yield solutions having a Brookfield viscosity of approximately 3000 cps. The viscosity stability at room temperature was evaluated.
The control was a hydroxy-propylated waxy maize starch which was not thermally-inhibited.
The results are tabulated below.
Solution Stability Control 110~C 125~C 140~C
Water Fluidity 25.0 25.5 20.6 21.8 Solids (%) 18 18 18 18 Initial 3160 2550 2820 2800 Viscosity (cps) W O96/23104 PCTrUS~J~

Viscosity after 3280 ~ 2640 24 hours (cps) Viscosity after 30202475 2730 2810 7 days (cps) - 5 Viscosity after 3000 1980 2140 2940 8 days (cps) Viscosity after 2850 1990 2230 2870 g days (cps) Appearance clear clear clear yellow ~MPLE 22 Waxy maize samples at the naturally occurring pH and at pH 8.5, were reacted with 1% by weight acetic anhydride (Ac20) and thermally-inhibited. The control was the non-thermally-inhibited waxy maize starch acetate.
The results are shown below.
Viscosity (BU) Break-Time Peak + 92 C + down (min) Peak 10' 92 C ~0' (%) WaxY Maize (1% Ac2O and natural pH at 160~C) Control -- 1480 490 -- -- 67 0 1030 570 -- __ 45 Waxy Maize (1~ Ac2O and natural PH at 160~C) Control -- 1480 490 -- -- 87 CA 022ll372 l997-07-24 W 096/23104 PCTJV~3C~ 3 The data show that derivatized starches, in this case esterified starches, can be inhibited to varying degrees and that higher inhibition can be obt~; ne~ at higher pH.
~AMPr~ 23 ~his example shows the preparation and use of a thermally-inhibited cationic starch in a simple papermaking system and a microparticle papermaking system.
A granular waxy corn starch (looo g) was slurried in 1500 CC water, 175 g of 4% sodium hydroxide were added, and the slurry was heated to 40~C. One hundred (100) g of a 50% aqueous solution of 3-chloro-2-hydroxypropyl trimethyl ammonium chloride was added while main~Ain;ng the pH at 11. 5 by adding 496 sodium hydroxide.
The mixture was allowed to react overnight at 40~C. The slurry was adjusted to pH 6. 5 with hydrochloric acid, filtered, washed and air dried to about 8-15% moisture.
The degree of substition was of 0.04.
A portion of the above cationic starch derivative was chemically crosslinked with 0.01 wt. % of epichlorohydrin at 40~C for 16 hours, neutralized to pH
6.0, filtered, water washed (2 parts water per part of starch), and air dried to about 8-15% moisture.
A portion of the above chemically-crosslinked, cationic starch was thermally inhibited by adjusting the pH to 9.5 with a 5% solution of sodium carbonate, spray-drying without gelatinization to between 3-15% moisture, 78 REPLACEMENi PAGE
and thermally dehydrating and heat treating the cationic, chemically-crosslinked granular starch in the fluidized bed previously described.
Prior to addition to the papermaking ~urnish, the starch samples were slurried at 4-6~ solids and cooked in a mini-jet cooker (scaled down jet cooker to simulate a commercial jet cooker) at a temperature o~
105-122~C and an applied back pressure of 34.5 to 137.8 kPa (5-20 psi) using controlled live steam. The mini-jet cooker had a cooking chamber capacity or volume of 5.0 ml. The starch was passed through the cooking --chamber at a flow rate o~ about 130 ml/min with a retention time of about 2.3 seconds.
A standard papermaking furnish was prepared using a pulp stock which comprised an aqueous slurry o~
bleached hardwood kra~t pulp (BHWK) and bleached so~twood kraft pulp (BSWK). The pulp stock (80 wt. ~ BHWK and 20 wt. ~ BSWK) was refined in an a~ueous solution to about 400 CSF (Canadian Standard Freeness) and a pH of 7.8-8.2.
The pulp stock contained precipitated calcium carbonate filler (30% by weight of fiber) with 8-10 wt. % fiber fines and 37-42 wt. ~ total fines.
Calcium carbonate retention, dry strength, and drainage resistance in both simple and microparticle papermaking systems were evaluated. The non-thermally-inhibited cationic starch was used as the control.
The results are set out below.
Simple Pa~ermakinq System CaCO3 Dry Bond Cationic StarchesRetentionStrencth Control 100 100 Chemically Crosslinked 148 106 Ns12936s.
A;U~.~.,~~ S~IE~

T-I (120~C
for 0 min.) 113 106 T-I (125~C
for 15 min.) 110 109 T-I (130~C
for 0 min.) 111 105 T-I (160~C
for 30 min.) 67 --T-I (160~C
for 120 min.) 60 --The results show that in the simple papermaking system the lightly inhibited starches (120~C for 0 min.
and 125~C for 15 min.) were better than the cationic ~Ull~' ol in both calcium carbonate retention and dry bond strength and were as good as the thermally-inhibited, chemically crosslinked cationic starch in dry bond strength but not in calcium carbonate retention.
Micro~article Pa~ermaking System CaCO3 Dry Bond Drainage Cationic StarchesRetention Strenqth Resistance Control 100 100 100 Chemically Crosslinked 119 118 108 T-I (120~C
for 0 min.) 101 119 106 T-I (125~C
for 0 min.) 120 139 102 T-I (130~C
for 0 min.) 123 134 100 T-I (160~C
for 0 min.) 63 52 101 W O96/23104 PCTrUS9~ J~9 T-I (160~C
for 30 min.) -- 71 95 T-I (160~C
for 120 min.) -- 75 --In the microparticle system, the very lightly inhibited starch (120~C for 0 min.) was better than the non-inhibited control in dry hand strength and drainage resistance and as good as the control in calcium carbonate retention. The lightly inhibited starches (125~C for 15 min. and 130~C for 0 min.) were better than the non-thermally inhibited control in calcium carbonate retention, dry bond strength, drainage resistance. The highly inhibited starches (160~C) were unsatisfactory in both calcium carbonate retention and dry bond strength.
The samples of starches were tested for Brookfield viscosity at 3% solids at 20 rpm with a No. 5 spindle.
Brookfield Viscosity ~cPs) Control 300 T-I Chemically Crosslinked 3650 T-I (120~C for o min.) 440 T-I (125~C for 15 min.) 790 T-I (130~C for 0 min.) 990 T-I (160~C for 0 min.) 1900 T-I (160~C for 30 min.) <60 T-I (160~C for 120 min.) <60 The results show that the thermally-inhibited starches are much lower in viscosity (<60 to 1900 cps) than the thermally-inhibited chemically crosslinked starch (3650 cps.). This is a significant advantage in papermaking since the pulp slurrier must be pumped.

,, . , ~ , . .. . .
.
- ~ ~ . . .. .. . . ... .
~ ~ . . . . .
81 ' '~UEPLACE~iE~rr PAGÉ
A Brabender analysis run on the starch which thermally inhibited at 160~C ~or 120 minutes. It showed a percentage breakdown o~ 2~.

s A cationic waxy maize starch was prepared using su~icient 3-chloro-2-hydroxypropyl trimethyl ammonium chloride to give about 0.30-0.36~ bound nltrogen a~
quaternary ammonium groups. The cationic starches were thermally inhibited in the fluidized bed previously _ descri~ed at the indicated temperature and time. The thermally-inhibited cat~onic starches were cooked at 4~
solids, a temperature of 104~C ~220~F), a pressure of 20 psi, and pump speed o~ 3.1 in the mini-jet cooker previously described.
The cooked starches were evaluated in two alkaline retention syste~s using a standard 80/20 hardwood:so~twood bleached Kra~t pulp (pH 7.8) with 30 calcium carbonate. The microparticle system contained .25% (5 lbs./ton) alum, .75% (15 lbs./ton) starch, and .15~ (3 lbs./ton) silica. The polymer system contained 0.5% (10 lbs./ton) alum, .75~ (15 lbs./ton) starch, and 0.05% (1 lb./ton) o~ a 33% emulsion o~ polyacrylamide (Nalco 625) as a retention aid. A non-thermally inhibited cationic starch was used as the control.
The results are shown below.
Microparticle SYstem Polvmer SYstem Cationic CaCO3 CaCO3 Starches Retention % Std. Retention ~ Std.
Control 31.1 100 49.1 100 T-I (130~C -- -- 59.1 120 for 0 min.) Ns12s36s.1 .\ CA 02211372 1997-07-24 3 . . C C ~
.
,, , ,, ,., __ _ T-I (130~C 43.7 150 54.8 112 for 15 min.) T-I (140~C 41.1 132 57 116 for 0 min.) T-I (140~C 24.1 77 42 86 for 15 min.) The results show that the more lightly inhibited starches (130~C for 0 and 15 min. and 140~C for 0 min.) had better calcium carbonate retention than the lo non-inhibited cationic starch. The more inhibited starch (140~C for 15 min.) had lower calcium carbonate retention than the non-inhibited cationic starch.

This example shows the use of thermally-inhibited cationic and amphoteric waxy maize starches inalkaline ~ine papers.
The cationic waxy maize starch was prepared as above.
The amphoteric waxy maize starch was prepared by reacting a granular anionic waxy maize starch containing about 0.08-0.12~ bound phosphate (provided by reaction with a sufficient amount of sodium tripolyphosphate) with a sufficient amount of 3-chloro-2-hydroxpropyl trimethyl ammonium chloride to give about 0.25-0.32% bound nitrogen.
The cationic and amphoteric starches were adjusted to pH ~.S and dehydrated and heat treated in the fluidized bed previously described ~or the indicated time at the indicated temperature.
The thermally-inhibited starches were jet cooked as previously described and evaluated in the microparticle paper making system containing .25~
(5 lbs./ton) alum, .75% (15 lbs./ton) starch, and .15%

Ns12s36s.1 ~ CA 02211372 1997-07-24 '83 ' REPLACEMENI PAGE
(3 lbs./ton) silica and in a polymer system containing r 0.5~ (10 lbs./ton) alum, .75~ (15 lbs./ton) starch, and 0.05~ (1 lb./ton) of an anionic polyacrylamide (Nalco 625) as a retention aid.
s The calcium carbonate retention data for the cationic starches are shown below. A non-thermally-inhibited cationic starch was used as the control.
Cationic Microparticle Starches Svstem Polymer Svs~em CaC03 CaC03 Retention ~ Std. Retention ~ Std.
J Control 31.1 100 49.1 100 T-I (130~C 43.7 140 59.1 120 for 15 min.) T-I (140~C 41.1 132 54.8 112 for 0 min.) T-I (140~C 24.1 77 58.5 119 for 15 min.) The results show that the more lightly inhibited cationic starches (130~C for 15 min. and 140~C
for 0 min.) were better than the control in the microparticle system and that all the lightly inhibited starches were better than the control in the polymer system.
The calcium carbonate retention data for the 2S amphoteric starches are shown below. A non-thermally-inhibited amphoteric starch was used as the control.
Microparticle Svstem ,PolYmer Svstem Amphoteric CaCO3 CaCO3 Starches Retention ~ Std. Retention ~ Std.
Control 31.1 100 49.1 100 NB129365.1 CA 022ll372 l997-07-24 w 096r23104 PCTAUS5Cf~0 T-I (110~C 34.6 111 42 86 for 15 min.) T-I (120~C 35.9 115 57.8 118 for 15 min.) T-I (130~C 34.6 111 59.4 121 for 15 min.) T-I (140~C 33.4 107 57.4 117 for 15 min.) The results show that the inhibited starches were better than the control.
The TPSF drainage data in the Microparticle and Polymer Systems are shown below.
Microparticle S~stem Starch Resis- % Std. Rate% Std.
tance (mls/
(mm) sec) Cationic Control 111100 167 100 T-I Cationic (130~C for 15 min.) 113 98 161 96 T-I Cationic (140~C for 15 min.) 108103 161 96 T-I Amphoteric (110~C for 15 min.) 108103 167 100 T-I Amphoteric (120~C for 15 min.) 107104 172 103 T-I Amphoteric (130~C for 15 min.) 106105 179 107 T-I Amphoteric (140~C for 15 min.) 114 97 156 94 ... . .. .

CA 022ll372 l997-07-24 W0 96/23104 PCT~US~C,S,~3 PolYmer System S~ch Resis- % Std. ~ % Std.
~' ~Ance (ml8/
(mm) sec) Cationic Control 90 100 165 100 Crosslinked Cationic 88 103 179 108 T-I Cationic (130~C for 15 min.) 91 99 179 108 T-I Cationic (140~C for 15 min.) 94 96 175 106 T-I Amphoteric (110~C for 15 min.) 90 100 165 100 T-I Amphoteric (120~C for 15 min.) 90 100 172 104 T-I Amphoteric (130~C for 15 min.) 91 99 185 112 T-I Amphoteric (140~C for 15 min.) 90 100 156 94 The results show that the drainage perfo. ~nce was satisfactory.
Now that the preferred ~ho~;ments of the present invention have been described in detail, various modifications and improvements thereto will become readily apparent to those skilled in the art.
Accordingly, the spirit and scope of the invention are to be limited only by the appended claims and foregoing specification.

Claims (20)

WE CLAIM:

1. A paper comprising, as a wet end additive, an effective amount of a thermally-inhibited starch or flour homogeneously dispersed therein, which thermally-inhibited starch as flour is prepared by (a) dehydrating a starch or flour to a moisture content of less than 1%
to render the starch or flour anhydrous or substantially anhydrous, and (b) heat treating the anhydrous or substantially anhydrous starch or flour for a time and at a temperature sufficient to inhibit the starch or flour, which temperature is 100°C or greater and which time is up to 20 hours, characterized in that the thermally-inhibited starch or flour, after dispersion in water, has improved viscosity stability in comparison to the non-thermally-inhibited base starch or flour.

2. The paper of Claim 1 wherein the starch or flour is a cereal starch or flour, a tuber starch or flour, a root starch or flour, a legume starch or flour, or a fruit starch or flour; wherein the starch or flour is adjusted to a pH of neutral or greater prior to the dehydrating; and wherein the thermally-inhibited starch or flour is a non-pregelatinized granular starch or flour which has an unchanged or reduced gelatinization temperature after the thermal inhibition or a pregelatinized granular or non-granular starch or flour.

3. The paper of Claim 2, wherein the dehydration is a thermal or a non-thermal dehydration;
wherein the pH is 7 to 10; wherein the heat-treating temperature is 110 to 160°C; and wherein the heating time is up to 2 hours.

4. The paper of Claim 3, wherein the pH is above 8 to below 10, the heating temperature is 120 to 140°C, and the heating time is up to 15 minutes.

5. The process of Claim 2, wherein the dehydrating and heat treating steps are carried out simultaneously in a in a fluidized bed.

6. The paper of Claim 4, wherein the thermally-inhibited starch or flour is selected from the group consisting of corn, pea, oat, potato, sweet potato, banana, barley, wheat, rice, sago, amaranth, tapioca, sorghum, waxy maize, waxy tapioca, waxy rice, waxy barley, waxy potato, waxy sorghum, and a starch or flour having an amylose content of 40% or greater.

7. The paper of Claim 6, wherein the starch is a derivatized starch selected from the group consisting of a cationic starch, an anionic starch, a non-ionic starch, and an amphoteric starch, which derivatized starch is optionally chemically crosslinked.

8. The paper of Claim 7, wherein the derivatized starch or derivatized, chemically-crosslinked starch, after dispersion in water, has a breakdown in Brabender viscosity of only from about 15-65%.

9. The paper of Claim 8, wherein the breakdown in Brabender viscosity is from about 25-45%.

10. The paper of Claim 8, wherein the cationic or amphoteric starches contain tertiary amino or quaternary ammonium groups.

11. The paper of Claim 10, wherein the cationic starch contains at least about 0.15% by weight of bound nitrogen and wherein the amphoteric starch contains at least about 0.15% by weight of bound nitrogen and at least about 0.04-1% bound phosphate groups.

12. The paper of Claim 11, wherein the cationic starch is a waxy maize starch containing diethylaminoethyl chloride hydrochloride groups and/or 2-hydroxypropyl trimethyl ammonium chloride groups in an amount sufficient to provide about 0.2-0.45% by weight of bound nitrogen; or an amphoteric waxy maize starch containing diethylaminoethyl chloride hydrochloride groups and/or 2-hydroxypropyl trimethyl ammonium chloride groups in an amount sufficient to provide about 0.2-0.45%
by weight of bound nitrogen and phosphate groups in an amount sufficient to provide about 0.1-0.3% by weight of bound phosphate.

13. The paper of Claim 7, wherein the starch is cooked at a temperature of from about 104-121°C
(220-250°F) and at a pressure of at least 15 psi.

14. The paper of Claim 1, wherein the wet end system further comprises an alkaline microparticle system containing an aluminum donor, an aluminum donor and colloidal silica or silicic acid, or an aluminum donor and bentonite and optionally a polyacrylamide.

15. In a method for making paper, the step which comprises adding to the stock, at any stage prior to passing the stock onto the wire, a dispersed thermally-inhibited cationic or amphoteric starch or flour which starch or flour, after dispersion in water, is characterized by its improved viscosity stability in comparison to the non-thermally-inhibited cationic or amphoteric base starch.

16. In the method of Claim 15, wherein the cationic or amphoteric starch is chemically-crosslinked prior to or after the thermal inhibition and wherein the thermally-inhibited and chemically-crosslinked starch is dispersed by cooking at a temperature of 104-121°C
(220-250°F) and a pressure of at least (15 psi).

17. A paper stock comprising (a) water, (b) cellulose fibers, (c) mineral fillers, and (d) a thermally-inhibited starch or flour homogeneously dispersed therein, which thermally-inhibited starch or flour is characterized by its improved viscosity stability in comparison to the non-thermally-inhibited base starch or flour.

18. The paper stock of Claim 17, further comprising (e) an aluminum donor, (f) colloidal silica or silicic acid or bentonite optionally together with a polyacrylamide as retention and drainage aids, and (g) optionally a sizing agent.

19. The paper stock of Claim 18, the aluminum donor is aluminum sulfate and/or polyaluminum chloride, wherein the mineral filler is calcium carbonate, and wherein the sizing agent is an alkenyl succinic anhydride and/or an alkyl ketene dimer.

20. A paper prepared from the stock of Claim 18.