CATION EXCHANGE RESIN BEADS WITH ROUGH OUTER SURFACE

FIELD

The invention relates the use of strong acid cation exchange resins to chromatographically separate sugars including monosaccharides such as fructose and glucose.

INTRODUCTION

The current state of the art for chromatographic separation of sugars (e.g. fructose and glucose) utilizes strong acid gel-type cation exchange resins in calcium form (Ca ²⁺ ). A

representative resin is DOWEX™ MONOSPHERE™ 99Ca/320 available from The Dow Chemical Company. See also US5176832. These types of chromatographic resins do not "exchange" ions in the traditional sense. Rather the bound Ca ²⁺ ions form ligand interactions with the hydroxyl (-OH) and carbonyl (C=0) groups of sugar molecules. Fructose has more "absorbing" interactions with the Ca ²⁺ ions and thus is more strongly retained by the resin as compared with glucose. Current techniques for chromatographically separating sugars utilize large quantities of water. New technologies are desired that require less water to accomplish the sugar separation.

SUMMARY

The subject cation exchange resins are characterized by possessing "rough" outer surfaces.

When used in the subject method, sugars may be chromatographically separated using less water and with lower operating costs than similar methods utilizing traditional cation exchange resins (i.e. those with "smooth" surfaces). In a preferred embodiment, the invention includes a method for chromatographically separating a first saccharide from a liquid (e.g. aqueous) eluent including the first saccharide and a second saccharide by passing the liquid eluent through a bed including a strong acid cation exchange resin. The resin is provided in bead form and is characterized by having a rough surface, i.e. SIOz of at least 1 and more preferably at least 5 or even 6 as measured by confocal laser scanning microscopy. In preferred embodiments, the resin is gel-type and provided in calcium form with a uniform particle size.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures la-b are micrographs of a conventional bead-form strong acid, gel-type cation exchange resin and Figures lc-d are micrographs of a similar resin provided with "rough" outer surfaces as per the subject invention.

Figure 2 includes four chromatographic elution curves (Brix vs. Bed Volume (BV)) corresponding to Example 2 with the first two elution curves correspond to glucose and fructose, respectively, using sample resin B and the second two elution curves correspond to glucose and fructose, respectively, using sample resin A. DETAILED DESCRIPTION

The invention includes a method for chromatographically separating a first saccharide (analyte) from a liquid eluent including multiple saccharides, (e.g. a first and second saccharide). While the liquid eluent may include a variety of constituents, e.g. monosaccharides, disaccharides, oligosaccharides, organic acids, amino acids, inorganic salts, etc., the first and second saccharides are preferably monosaccharides (e.g. glucose and fructose). For example, in the production of high fructose corn syrup, the liquid eluent typically includes an aqueous mixture of glucose (first saccharide) and fructose (second saccharide) along with various acids and salts. As with traditional chromatographic separations of saccharides, the liquid eluent (mobile phase) passes through a bed or stratum of resin (stationary phase). The set up and operation of the bed is not particularly limited, e.g. moving, simulated moving and stationary beds may be used. Given the nature of the interactions with the resin, the first and second saccharides pass through the resin bed at different rates, thus allowing their separation. For example, fructose (second saccharide) more strongly interacts with the resin as compared with glucose (first saccharide). As a consequence, glucose passes (elutes) through the bed more quickly followed by fructose as a separate product "cut". The individual product cuts can then be collected and used or further treated as is customary in the art.

The resin used in the present invention is a strong acid cation exchange resin, preferably provided in bead form. The cation exchange resin includes a crosslinked copolymer matrix that is functionalized (e.g. sulfonated) and which is preferably converted to a calcium form. While the crosslinked copolymer matrix may be macroporous or gel-type, gel-type copolymers are preferred. The terms "gel-type" and "macroporous" are well-known in the art and generally describe the nature of the copolymer bead porosity. The term "macroporous" as commonly used in the art means that the copolymer has both macropores and mesopores. The terms "microporous," "gellular," "gel" and "gel-type" are synonyms that describe copolymer beads having pore sizes less than about 20 Angstroms A , while macroporous copolymer beads have both mesopores of from about 20 A to about 500 A and macropores of greater than about 500 A . Gel-type and macroporous copolymer beads, as well as their preparation are further described in US 4256840 and US 5244926 - the entire contents of which are incorporated herein by reference. The crosslinked copolymer resin beads preferably have a median bead diameter from 100 to 2000 microns, and more preferably from 150 to 500 microns). The beads may have a Gaussian particle size distribution but preferably have a relatively uniform particle size distribution, i.e. "monodisperse" that is, at least 90 volume percent of the beads have a particle diameter from about 0.9 to about 1.1 times the volume average particle diameter. A representative example is AMBERLITE™ SR1L available from The Dow Chemical Company and which may be converted to a calcium ionic form by way of conventional techniques known in the art.

Preferred methods for making the subject resin are known in the art, including "seeded" polymerizations, sometimes also referred to as batch or multi-batch (as generally described in EP 62088 Al and EP 179133A1); and continuous or semi-continuous staged polymerizations (as generally described in US 4419245, US 4564644; and US 5244926). A seeded polymerization process typically adds monomers in two or more increments. Each increment is followed by complete or substantial polymerization of the monomers therein before adding a subsequent increment. A seeded polymerization is advantageously conducted as a suspension polymerization wherein monomers or mixtures of monomers and seed particles are dispersed and polymerized within a continuous suspending medium. In such a process, staged polymerization is readily accomplished by forming an initial suspension of monomers, wholly or partially polymerizing the monomers to form seed particles, and subsequently adding remaining monomers in one or more increments. Each increment may be added at once or continuously. Due to the insolubility of the monomers in the suspending medium and their solubility within the seed particles, the monomers are imbibed by the seed particles and polymerized therein. Multi-staged polymerization techniques can vary in the amount and type of monomers employed for each stage as well as the polymerizing conditions employed.

The seed particles employed may be prepared by known suspension polymerization techniques. In general the seed particles may be prepared by forming a suspension of a first monomer mixture in an agitated, continuous suspending medium as described by F. Helfferich in Ion Exchange, (McGraw-Hill 1962) at pp. 35-36. The first monomer mixture comprises: 1) a first monovinylidene monomer, 2) a first crosslinking monomer, and 3) an effective amount of a first free -radical initiator. The suspending medium may contain one or more suspending agents commonly employed in the art. Polymerization is initiated by heating the suspension to a temperature of generally from about 50-90°C. The suspension is maintained at such temperature or optionally increased temperatures of about 90-150° C until reaching a desired degree of conversion of monomer to copolymer. Other suitable polymerization methods are described in US 4444961, US 4623706, US 4666673 and US 5244926 - each of which is incorporated herein in its entirety.

The monovinylidene aromatic monomers employed herein are well-known and reference is made to Polymer Processes, edited by Calvin E. Schildknecht, published in 1956 by Interscience Publishers, Inc., New York, Chapter III, "Polymerization in Suspension" at pp. 69-109. Table II (pp. 78-81) of Schildknecht lists diverse types of monomers which are suitable in practicing the present invention. Of the monomers listed, styrene and substituted styrene are preferred. The term

"substituted styrene" includes substituents of either/or both the vinylidene group and phenyl group of styrene and include: vinyl naphthalene, alpha alkyl substituted styrene (e.g., alpha methyl styrene) alkylene-substituted styrenes (particularly monoaikyl-substituted styrenes such as vinyltoluene and ethylvinylbenzene) and halo-substituted styrenes, such as bromo or chlorostyrene and vinylbenzyl chloride. Additional monomers may be included along with the monovinylidene aromatic monomers, including monovinylidene non-styrenics such as: esters of α,β-ethylenically unsaturated carboxylic acids, particularly acrylic or methacrylic acid, methyl methacrylate, isobornyl- methacrylate, ethylacrylate, and butadiene, ethylene, propylene, acrylonitrile, and vinyl chloride; and mixtures of one or more of said monomers. Preferred monovinylidene monomers include styrene and substituted styrene such as ethylvinylbenzene. The term "monovinylidene monomer" is intended to include homogeneous monomer mixtures and mixtures of different types of monomers, e.g. styrene and isobornylmethacrylate. The seed polymer component preferably comprises a styrenic content greater than 50 molar percent, and more preferably greater than 75, and in some embodiments greater than 95 molar percent (based upon the total molar content). The term "styrenic content" refers to the quantity of monovinylidene monomer units of styrene and/or substituted styrene utilized to form the copolymer. "Substituted styrene" includes substituents of either/or both the vinylidene group and phenyl group of styrene as described above. In preferred embodiments, the first monomer mixture used to form the first polymer component (e.g. seed) comprises at least 75 molar percent, preferably at least 85 molar percent and in some embodiments at least 95 molar percent of styrene.

Examples of suitable crosslinking monomers (i.e., polyvinylidene compounds) include polyvinylidene aromatics such as divinylbenzene, divinyltoluene, divinylxylene, divinylnaphthalene, trivinylbenzene, divinyldiphenylsulfone, as well as diverse alkylene diacrylates and alkylene dimethacrylates. Preferred crosslinking monomers are divinylbenzene, trivinylbenzene, and ethylene glycol dimethacrylate. The terms "crosslinking agent," "crosslinker" and "crosslinking monomer" are used herein as synonyms and are intended to include both a single species of crosslinking agent along with combinations of different types of crosslinking agents. The proportion of crosslinking monomer in the copolymer seed particles is preferably sufficient to render the particles insoluble in subsequent polymerization steps (and also on conversion to an ion-exchange resin), yet still allow for adequate imbibition of an optional phase-separating diluent and monomers of the second monomer mixture. In some embodiments, no crosslinking monomer will be used. Generally, a suitable amount of crosslinking monomer in the seed particles is minor, i.e., desirably from about 0.01 to about 12 molar percent based on total moles of monomers in the first monomer mixture used to prepare the seed particles. In a preferred embodiment, the first polymer component (e.g. seed) is derived from polymerization of a first monomer mixture comprising at least 85 molar percent of styrene (or substituted styrene such as ethylvinylbenzene) and from 0.01 to about 10 molar percent of divinylbenzene.

Polymerization of the first monomer mixture may be conducted to a point short of substantially complete conversion of the monomers to copolymer or alternatively, to substantially complete conversion. If incomplete conversion is desired, the resulting partially polymerized seed particles advantageously contain a free -radical source therein capable of initiating further polymerization in subsequent polymerization stages. The term "free-radical source" refers to the presence of free -radicals, a residual amount of free -radical initiator or both, which is capable of inducing further polymerization of ethylenically unsaturated monomers. In such an embodiment of the invention, it is preferable that from about 20 to about 95 weight percent of the first monomer mixture, based on weight of the monomers therein, be converted to copolymer and more preferably from about 50 to about 90 weight percent. Due to the presence of the free radical source, the use of a free-radical initiator in a subsequent polymerization stage would be optional. For embodiments where conversion of the first monomer mixture is substantially complete, it may be necessary to use a free-radical initiator in subsequent polymerization stages.

The free-radical initiator may be any one or a combination of conventional initiators for generating free radicals in the polymerization of ethylenically unsaturated monomers. Representative initiators are UV radiation and chemical initiators, such as azo-compounds including

azobisisobutyronitrile; and peroxygen compounds such as benzoyl peroxide, t-butylperoctoate, t-butylperbenzoate and isopropylpercarbonate. Other suitable initiators are mentioned in US 4192921, US 4246386 and US 4283499 - each of which is incorporated in its entirety. The free-radical initiators are employed in amounts sufficient to induce polymerization of the monomers in a particular monomer mixture. The amount will vary as those skilled in the art can appreciate and will depend generally on the type of initiators employed, as well as the type and proportion of monomers being polymerized. Generally, an amount of from about 0.02 to about 2 weight percent is adequate, based on total weight of the monomer mixture.

The first monomer mixture used to prepare the seed particles is advantageously suspended within an agitated suspending medium comprising a liquid that is substantially immiscible with the monomers, (e.g. preferably water). Generally, the suspending medium is employed in an amount from about 30 to about 70 and preferably from about 35 to about 50 weight percent based on total weight of the monomer mixture and suspending medium. Various suspending agents are conventionally employed to assist with maintaining a relatively uniform suspension of monomer droplets within the suspending medium. Illustrative suspending agents are gelatin, polyvinyl alcohol, magnesium hydroxide, hydroxyethylcellulose, methylhydroxyethyl cellulose methylcellulose and carboxymethyl methylcellulose. Other suitable suspending agents are disclosed in US 4,419,245. The amount of suspending agent used can vary widely depending on the monomers and suspending agents employed. Latex inhibitors such as sodium dichromate may be used to minimize latex formation.

In the so-called "batch-seeded" process, seed particles comprising from about 10 to about 50 weight percent of the copolymer are preferably suspended within a continuous suspending medium. A second monomer mixture containing a free radical initiator is then added to the suspended seed particles, imbibed thereby, and then polymerized. Although less preferred, the seed particles can be imbibed with the second monomer mixture prior to being suspended in the continuous suspending medium. The second monomer mixture may be added in one amount or in stages. The second monomer mixture is preferably imbibed by the seed particles under conditions such that substantially no polymerization occurs until the mixture is substantially fully imbibed by the seed particles. The time required to substantially imbibe the monomers will vary depending on the copolymer seed composition and the monomers imbibed therein. However, the extent of imbibition can generally be determined by microscopic examination of the seed particles, or suspending media, seed particles and monomer droplets. The second monomer mixture desirably contains from about 0.5 to about 25 molar percent, preferably from about 2 to about 17 molar percent and more preferably 2.5 to about 8.5 molar percent of crosslinking monomer based on total weight of monomers in the second monomer mixture with the balance comprising a monovinylidene monomer; wherein the selection of crosslinking monomer and monovinylidene monomer are the same as those described above with reference to the preparation of the first monomer mixture, (i.e. seed preparation). As with the seed preparation, the preferred monovinylidene monomer includes styrene and/or a substituted styrene. In a preferred embodiment, the second polymer component (i.e. second monomer mixture, or "imbibed" polymer component) has a styrenic content greater than 50 molar percent, and more preferably at least 75 molar percent (based upon the total molar content of the second monomer mixture). In a preferred embodiment, the second polymer component is derived from polymerization of a second monomer mixture comprising at least 75 molar percent of styrene (and/or substituted styrene such as ethylvinylbenzene) and from about 1 to 20 molar percent divinylbenzene.

In an in-situ batch-seeded process, seed particles comprising from about 10 to about 80 weight percent of the copolymer product are initially formed by suspension polymerization of the first monomer mixture. The seed particles can have a free -radical source therein as previously described, which is capable of initiating further polymerization. Optionally, a polymerization initiator can be added with the second monomer mixture where the seed particles do not contain an adequate free radical source or where additional initiator is desired. In this embodiment, seed preparation and subsequent polymerization stages are conducted in-situ within a single reactor. A second monomer mixture is then added to the suspended seed particles, imbibed thereby, and polymerized. The second monomer mixture may be added under polymerizing conditions, but alternatively may be added to the suspending medium under conditions such that substantially no polymerization occurs until the mixture is substantially fully imbibed by the seed particles.

Once formed, the crosslinked copolymer resin beads are sulfonated with a sulfonating agent, e.g. concentrated sulfuric acid (e.g. at least 96 wt ), oleum (fuming sulfuric acid), chlorosulfonic acid, sulfur trioxide or combinations thereof. The sulfonation reaction is preferably conducted at elevated temperature, e.g. 110 to 150°C without a solvent. An applicable sulfonation technique is described in US 2005/0014853. See also: in US 4500652, US 6228896, US 6750259, US 6784213, US 2002/002267 and US 2004/0006145. Sulfonation of the resin in the absence of a solvent is believed to impart a wrinkled or dimpled surface to the resin bead. When used in the present method, this rough surface is believed to increase interaction of analytes with the bead, thus leading to faster elution rates.

By way of illustration, Figures la-d are micrographs of two comparable gel-type strong acid cation exchange resins. The resin shown in Figs.1 a) and b) was sulfonated using a traditional solvent (EDC) whereas the resin shown in Figs lc) and d) was sulfonated in the absence of a solvent. More specifically, Figures la-b are micrographs of AMBERLITE™ 120 Na brand cation exchange resin available from The Dow Chemical Company. This resin has a gel-type, crosslinked,

styrene-divinylbenzene copolymer matrix that has been sulfonated with H ₂ S0 ₄ using EDC as a swelling agent (solvent). Figures lc-d are micrographs of AMBERLITE™ SR1L brand cation exchange resin also available from The Dow Chemical Company. This resin is substantially similar to AMBERLITE 120 Na; however, this resin was sulfonated using H ₂ S0 ₄ without solvent.

In a preferred embodiment, the sulfonated resins are subsequently converted to a calcium form using standard techniques as used with respect to ion exchange resins. For example, the sulfonated resin may be combined, agitated and soaked within a 1M solution of CaCl ₂ . The resin may then be optionally soaked within a saturated solution of Ca(OH) ₂ followed by optionally pH adjustment, e.g. with a solution of H ₃ P0 ₄ . The treatment with CaCl ₂ may be repeated multiple times to ensure a high level of conversion.

As mentioned, the cation exchange resins of the subject invention possess a rough outer surface. As shown in Figures lc-d, the surface appears wrinkled or dimpled. The degree of roughness may be qualified by way of Confocal Laser Scanning Microscopy (CLSM), e.g. a Keyence VK-9700 microscope application viewer VK-Hl VIE with a 50x objective lens and superfine resolution) using a scanning violet laser (408 nm) light source for high resolution confocal surface profiling using a 283 μπι x 212 μπι sample surface area. Combined with an additional white light source, the system provides simultaneous color, laser intensity, and height information to generate high-resolution images. The CLSM is preferably operated with a 1 nm z resolution and 130 nm spatial resolution providing SEM-like images with a large 7mm through focus range. Using CLSM, the following parameters may be used to characterize the surface of a resin bead:

Sa = the arithmetic mean deviation of the actual surface relative to the center plane fit to the data. Sq = the geometric mean derivation of the all points on the surface from the mean value of the data

(RMS or root-mean-squared roughness).

SAD = The difference in the surface area between the imaged surface and a flat surface of the same lateral size, expressed as a percentage.

SIOz = the average difference between the 5 highest and 5 lowest points on the surface relative to the mean plane.

Smax = the difference between the highest and lowest point on the surface relative to the mean plan.

The resin beads of the subject invention possess at least one and preferably all of the following surface characteristics: Sa > 1 (e.g. 1 to 3); Sq > 1 (e.g. 1-3), preferably > 1.2; SAD > 2 (e.g. 2 to 4); SIOz > 1, preferably >5 and more preferably >6 (e.g. 2 to 8); and Smax > 1, preferably >5 and more preferably >6 (e.g. 2 to 8). EXAMPLES

Comparative testing was conducted on two cation exchange resins (samples A and B). Both resins were strong acid gel-type resins based upon a styrene-divinylbenzene crosslinked copolymer and provided in their calcium ionic form. Resin A was sulfonated at 140°C using sulfuric acid using EDC (as a swelling agent) and had a "smooth" outer surface; whereas resin B was sulfonated under the same conditions but without the use of a swelling agent and had a "rough" outer surface. Selected physical properties of the resins are summarized in Table 1.

Table 1

Confocal Laser Scanning Microscopy (CLSM) was used to determine general surface profiles for the smooth (A) and wrinkly (B) resin beads. In the method, lasers were directed at the bead to capture high-resolution optical images at different depths on the beads surface. The CLSM parameters (Sa, Sq, SAD, SIOz, and Smax) analyze the deviation from the average surface height compared to that of a perfect sphere. The results are summarized in Table 2.

Table 2

SAD or the surface area difference is the percent difference in the surface area of a resin bead versus that of a perfect sphere. The surface area of smooth resin (A) shows approximately 0.95% increase in surface area versus that of a perfect sphere. The wrinkly resin (B) shows approximately 2.10% increase in surface area. From the CLSM images and numerical data, one can conclude that the exaggerated peaks and valleys that result from sulfonation without EDC increase the surface area by 1.25% (2.10% - 0.95%). The resin samples were tested to determine their relative ability to chromatographically resolve sugars. The identity of the resins and their comparative resolutions ("R") are summarized in Table 4 provided below. Chromatographic pulse testing was conducted using the testing conditions summarized in Table 3.

Table 3: chromatographic pulse testing parameters

More specifically, a column was packed with the sample resin and heated to 60°C, and the individual sugar (glucose or fructose) to be pulsed were preheated to 60°C by placing a bottle containing the sugar in a heated water bath. Water was pumped through the column at the flowrate of the test (2.0 BV/hr) and the pressure in the column was maintained constant. To start the test, 0.05 BV (26.2 mL) of heated 30 Brix sugar was loaded into the injection valve using a syringe fitted with a blunt needle. The injection valve was then switched into the "inject" position and the sample loop containing the sugar was placed in-line to the column inlet stream. 10 minutes (for glucose) or 12 minutes (for fructose) after injection, the fraction collector was turned on and fraction collection began. Fractions were collected every 18 seconds (0.010 BV). The fractions were analyzed for total sugar content using a Reichert AR200 Digital Hand-Held Refractometer (Catalog # 13950000) - commonly referred to as a "Brixmeter" - to determine which fractions contain sugar and at what concentration. Each fraction was analyzed for sugar concentration using the Brixmeter, and a pulse test chromatogram for the sugar tested was constructed from the results by plotting bed volumes (calculated from fraction number) on the x-axis and sugar concentration in Brix on the y-axis. This is provided in Figure 2. This pulse test was repeated twice per resin - once using glucose and once using fructose. The resulting chromatograms were overlaid on top of one another and pulse test results such as resolution coefficient are calculated from the overlaid chromatogram. Results are illustrated in

Figure 2 and Table 4. In Table 4, μ _! is the first moment of the sugar elution peak, σ is the elution peak standard deviation, μ ₂ is the second moment or variance of the sugar elution peak, N is the number of theoretical separation plates, H is the theoretical plate height, and R is the glucose/fructose resolution value. Table 4: Glucose/Fructose Resolution values (R)