STARCH POLYESTER BLEND FROM REACTIVE EXTRUSION
Cross-Reference to Related Applications This application claims the benefit of U.S. Provisional Patent Application No. 60/567,646, filed May 4, 2004, the whole of which is incorporated herein by reference.
Government Support This invention was made with United States Government support from the U.S. Department of Agriculture (USDA) Grant No. 2001-35504-10681. The Government has certain rights in the invention.
Technical Field The invention is directed to starch or other hydroxyl containing polymer blend with polyester, of improved mechanical properties.
Background of the Invention The above-described blends require compatibilization steps to minimize mechanical property decrease. Previously compatibilization efforts have focused on grafting anhydride groups on polyester or starch or on grafting low molecular weight polyesters, e.g., low molecular weight polycaprolactones, on starch, or on polymerizing ester monomers, e.g., ε- caprolactone, directly on starch. These processes require either additional processing steps or high amounts of polyesters (e.g., 70 weight percent polycaprolactone) to achieve asserted mechanical properties.
Summary of the Invention It has been discovered herein that advantages over known methods can be obtained if starch or other hydroxyl containing polymer is oxidized during blend preparation. One embodiment herein, denoted the first embodiment, is directed at a hydroxyl containing polymer-polyester blend comprising plasticized oxidized hydroxyl containing polymer crosslinked with polyester, the blend having maximum tensile strength greater than
elongation at break greater than 50% of that of the polyester. Another embodiment herein, denoted the second embodiment, is directed at a method for preparing blend of hydroxyl containing polymer and polyester, of maximum tensile strength greater than 25%) of that of the polyester, Young's modulus greater than 25%> of that of the polyester and elongation at break greater than 50% of that of the polyester comprising the steps of (a) forming an admixture of hydroxyl containing polymer and polyester in a hydroxyl containing polymer-polyester weight ratio ranging from 1 :10 to 10:1, plasticizer in a hydroxyl containing polymer weight ratio ranging from 1 : 10 to 1:1 and H202 in an amount effective to oxidize the hydroxyl containing polymer, and (b) feeding the admixture into an extruder and extruding therein or therethrough to cause oxidation of the hydroxyl containing polymer to form oxidized hydroxyl containing polymer and cause crosslinking between oxidized hydroxyl containing polymer and polyester of the admixture. The maximum tensile strength data in MPa set forth herein, is determined by ASTM Test D638-96. The Young's modulus data in MPa set forth herein is determined by ASTM Test D638-96. The ultimate elongation data (%), i.e., elongation at break data (%) set forth herein, is determined by ASTM Test D638-96.
Brief Description of the Drawings FIG. 1A is a graph of wt % organoclay versus Youngs Modulus and shows results of Working Example I. FIG. 1 B is a graph of wt % organoclay versus elongation at break and shows results of Working Example I. FIG. 1C is a graph of wt % organoclay versus maximum tensile strength and shows results of Working Example I.
Detailed Description We turn now to the first embodiment which is directed at a hydroxyl containing polymer-polyester blend comprising plasticized oxidized hydroxyl containing polymer crosslinked with polyester, said blend having a maximum tensile strength greater than 25% of that of the polyester, Young's modulus greater than 25% of that of the polyester and elongation at break greater than 50% of that of the polyester. The hydroxyl-containing polymer is preferably starch (e.g., wheat starch, high amylose corn starch, waxy maize starch and their blends). The starch can also be obtained, for example, from rice, potato, peas and/or tapioca. Native wheat and corn starches, which are preferred for use herein, contain about 75% amylopectin fraction and 25% amylose fraction where the amylose fraction has a degree of polymerization (DP) ranging from lxlO2 to 4x105 and the amylopectin fraction has a DP ranging from lxlO4 to 4x107, with branches after every 19-25 linear units. Commercially available corn starches having amylose contents ranging from 0% (waxy maize) to about 70% (high amylose content) where amylose has DP range as above if present, and amylopectin has a DP range and branching set forth above, are also useful herein. The hydroxyl containing polymer can also be, for example, cellulosic polymer (Mn ranging from 160,000 to 500,000,000 g/mol as determined by gel permeation chromatography or intrinsic viscosity measurements) or polyvinyl alcohol (Mn ranging from 25,000 to 300,000 g/mol as determined by gel permeation chromatography). The oxidation converts the hydroxy of the hydroxyl containing polymer to carboxyl, aldehyde and ketone and fosters crosslinking by abstraction of hydrogen atoms adjacent to carbonyl on both oxidized hydroxyl containing polymer and polyester. The plasticizer is present to facilitate forming hydroxyl containing polymer melts during preparation processing without significant molecular breakdown. Preferably, the plasticizer is a trihydric alcohol, very preferably glycerol. Other suitable plasticizers include monohydric, dihydric and polyhydric alcohols. Some examples are ethylene glycol, propylene glycol, erythritol, pentaerythritol, sorbitol and higher molecular weight plasticizers such as polyglycerol. Plasticizers used herein may also act as substrates in the oxidation/crosslinking pathways.
The polyester is preferably a biodegradable polyester and when such polyester is used in a blend with oxidized starch or native starch (with or without clay), the blend is biodegradable. Biodegradable polyesters for use herein include, for example, polycaprolactone (e.g., poly (e-caprolactone), denoted PCL, polybutylene succinate, denoted PBS, polytetramethylene adipate-co-terephthalate (PAT), polyhydroxyl butyrate-valerate (PHBV), polylactic acid (PLA) and polyglycolic acid (PGA). The polyester can also be a non4?iodegradable polyester, e.g., polyethylene terephthalate. The polyesters typically have Mn ranging from 40,000 to 300,000 g/mol as determined by gel permeation chromatography using polystyrene standards. Preferably the blend comprises glycerol plasticized oxidized starch crosslinked with polyester. The blend can optionally contain nanoclay modified to contain organic cation. Suitable nanoclays include montmorillonite, hectorite and saponite modified to contain organic cation. A preferred clay ingredient is montmorillonite modified to contain quaternary ammonium octadecyl cations (C18H35NH ). The clay functions to allow reduction in temperature during preparation processing, for obtaining the stated mechanical properties. The blend is preferably formulated from (A) 9 to 91 weight percent hydroxyl containing polymer and (B) from 91 to 9 weight percent polyester, based on (A) plus (B) being 100%; very preferably is formulated from (A) 10 to 95 weight percent plasticized hydroxyl containing polymer and (B) 90 to 5 weight percent polyester, based on (A) and (B) being 100%. The hydroxyl-containing polymer: plasticizer weight ratio can range, for example, from 10:1 to 1 :1 ; 2:1 is used in Working Example I. When clay is present, the blend is additionally formulated to contain from 0.5 to 10%, preferably more than 1%, nanoclay modified to contain organic cation, based on the weight of the blend including the organic cation modified clay. We turn now to the second embodiment of the invention which is directed to preparing hydroxyl containing polymer-polyester blend of maximum tensile strength greater than 25%> of that of the polyester, Young's modulus greater than 25% of that of the polyester and elongation at break greater than 50% of that of the polyester, comprising the steps of (a) forming an admixture of hydroxyl containing polymer, plasticizer in a hydroxyl containing polymer plasticizer weight ratio ranging from 10:1 to 1 :1, polyester in a hydroxyl containing
polymer: polyester weight ratio ranging from 1 : 10 to 10:1 and H 0 in an amount effective to oxidize the hydroxyl containing polymer, and (b) feeding the admixture into an extruder and extruding the admixture therein or therethrough to cause oxidation of the hydroxyl containing polymer to form oxidized hydroxyl containing polymer and cause crosslinking between oxidized hydroxyl containing polymer and polyester of the admixture. The hydroxyl containing polymers, plasticizers and polyesters are those of the first embodiment. In data, the hydroxyl containing polymer is native wheat starch, the plasticizer is glycerol and the polyester is polycaprolactone (Mn of 80,000) or polytetramethylene adipate- co-terephthalate. (high molecular weight, Eastar Bio Ultra) The blends are useful as plastic substitutes. For the second embodiment, the hydroxy containing polymer: polyester weight ratio ranges from 1: 10 to 10:1, the hydroxyl-containing polymeπplasticizer weight ratio ranges from 10:1 to 1 :1 and the H 0 can be present in amount ranging from 0.01 ml/gm of hydroxyl containing polymer to 0.36 ml/gm of hydroxyl containing polymer on a 30% H 0 solution in water basis. The admixture for step (a) of the second embodiment can optionally include ferrous ions, e.g., in the form of ferrous sulfate, to catalyze the oxidation, e.g., in an amount of 0.0069-1.5 g/ml of peroxide (30%) solution in water basis); cupric ion, e.g., in the form of cupric sulfate is also optionally included to catalyze oxidation and decrease molecular weight of oxidized starch (by a pathway called the Ruff degradation pathway) but is omitted in applications requiring contact with food materials. When cupric sulfate is included, it is used in an amount of 0.0020-0.0125 g/gram starch. Other transition metal catalysts, like iron, can replace copper. The H 02 and the catalysts perform a dual role, generating free hydroxyl radicals (i) for hydroxyl containing polymer oxidation and (ii) for causing crosslinking of oxidized starch and polyester. Nanoclay, e.g., montmorillonite, hectorite or saponite, modified to contain an organic cation, e.g., montmorillonite modified to contain C|gH35NH4 +, is an optional component in the admixture of step (a) and allows reducing of processing temperature in step (b), for obtaining the stated mechanical properties. When included the nanoclay is ordinarily used in an amount by weight ranging from 0.5 to 10% of the admixture of step (a).
Step (b) can be carried out in a batch extruding step or in a single continuous reactive extrusion process where oxidation and crosslinking successively occur. The term "reactive extrusion" is used herein to mean extrusion of plasticized hydroxyl containing polymer and polyester blend in the presence of hydrogen peroxide and catalyst(s) with or without clay. The extrusion temperature generally ranges from 90°C to 150°C but clay is a necessary ingredient to obtain the stated mechanical properties if the extrusion temperature is less than 130°C. There is no significant effect on properties when higher temperatures (130° C or higher) are used with clay. The plasticizer facilitates forming hydroxyl containing polymer, e.g., starch, melts during extrusion without significant molecular breakdown. In addition to hydroxyl containing polymer and polyester, the plasticizer may also act as a substrate for oxidation/crosslinking pathways during extrusion. The final products obtained which are embraced by the first embodiment of the invention herein are, for example, mixtures of starch, oxidized starch, glycerol, polycaprolactone and crosslinked oxidized starch-polycaprolactone. The invention is further described in Kalambur, S.B. and Rizvi, S.S.H., Polymer International, 53 (10), 1413-1416 (published on-line 7/29/04), and Journal of Applied Polymer Science 96(4), 1072-1082 (published on4ine March 2005), the whole of both articles being incorporated herein by references. The invention is illustrated by the following working examples.
Working Example 1 Formulations were made up as set forth in Table 1 below.
Table 1
*H
20
2 is 30% solution in water, ferrous and cupric sulfate catalysts were used at 0.0025 and 0.002 g/g of starch on wet basis. **Two other levels of peroxide were also used, 0.27 (Level II) and 0.36 (Level III) ml/gram starch, indicated by STPCLPERII and STPCLPERII1 respectively.
The starch was native wheat starch (MIDSOL 50, Midwest Grain Products, Atchison, Kansas). The polycaprolactone (PCL) had Mn=80,000 g/mol. The organoclay was Nanocor I.30E (montmorillonite) type with sodium cations, modified to replace sodium cations with quaternary ammonium octadecyl cations (C18H35NH ). Ingredients were pre-mixed and fed to a co-rotating twin-screw microextruder (DACA Instruments, Goleta, California). The residence time was kept at 3 minutes. Melt was extruded out of the die of the microextruder in the form of cylindrical strands. The temperature throughout the extruder was maintained at 120°C. Screw speed was maintained at 125-130 rpm. The extruder barrel was blanketed with nitrogen during extrusion. Two batches of each formulation were extruded to ensure reproducibility. The extruded strands were injection molded in a micro-injector (DACA Instruments, Goleta, California) at 80-1 10 psi and 120°C in the form of dog-bone pieces of area of 1.5 x 4 mm2. The mold was maintained at ambient temperature. The conditions were such as to obtain peroxide oxidation to almost 100% conversion because of high temperature (120° C) during the reactive extrusion process. The exact percent conversion was not determined but since H20 was present in concentrated form in the presence of excess organic substrates (starch and PCL), it is expected that full conversion took place during the extrusion process. Tensile properties of the formulations of Table 1 were obtained as described above using an Instron Model 1 122 at crosshead speed of 50 mm/min. Young's modulus was measured up to 1 % strain. The mechanical property results are set forth in Table 2 below where the values in parentheses are standard deviations.
Table 2
Data was also obtained which is set forth in Table 3 below. Table 3
native wheat starch:PCL:glycerol wt. ratio 1 : 1 :0.5, ferrous and cupric sulfate catalyst at 0.0025 and 0.002 g/g starch, H
20
2 (30% solution in water) = 0.18ml/gram starch, 0% clay, extruded and molded at 140°C native wheat starch:PCL:glycerol wt. ratio 1 : 1 :0.5, ferrous and cupric sulfate catalyst at 0.0025 and 0.002 g/g starch, H
20 (30%) solution in water) = 0.18ml/gram starch, 3% clay, extruded and molded at 140°C c. 100%) PCL, extruded and molded at 120°C d. native wheat starch:PCL:glycerol wt. ratio 1 :1 :0.5, 0% clay, extruded and molded at 120 °C
The values for Maximum Tensile Strength and Elongation at Break differ for STPCLPERI-0 and STPCLPERI-3 between Tables 2 and 3 because of different extrusion temperatures (120° C for STPCLPERI-0 and STPCLPERI-3 for Table 2 and 140° C for Table 3). As shown in Tables 2 and 3, elongation at break was dramatically improved in blends subjected to reactive extrusion (i.e., extrusion with plasticizer and H
20
2 and optionally, ferrous, present), (denoted RB), compared to where blend was not subjected to reactive extrusion. Elongation at break reached a maximum at 3% organoclay and decreased with increasing amounts of organoclay. In the reactive extrusion provided blends at 6 and 9% organoclay, the maximum tensile strength and Young's Modulus approached and were each greater, respectively, over that of blend without reactive extrusion (STPCL). So far as the blends obtained by reaction extrusion are concerned, no large differences in maximum tensile strength were observed among the formulations and modulus increased as organoclay content increased between 3 and 9% organoclay. The STPCLPERI-0 sample in Table 2 (120° C extrusion) showed poor properties. Moreover, the blend with 1% nanoclay in Table 2 (120° C extrusion) did not have tensile properties comparable to 100% PCL. The effects of peroxide level at 0.18 (LEVEL I PEROXIDE), 0.27 (LEVEL II PEROXIDE) and 0.36 (LEVEL III PEROXIDE) ml peroxide/gram of starch on Young's modulus, elongation at break and maximum tensile strength are respectively shown in FIGS. 1 A, IB and 1C. As shown in FIGS. 1A-1C, the best properties at different peroxide levels were obtained at a level of 0.18 ml/gram of starch. SEM pictures of STPCL, STPCLPERI-3 and STPCLPERI-6, obtained from molded samples that were ultra-sonicated in water for 5 minutes at 40°C, showed better interfacial adhesion for the compositions of the invention with organoclay, that resulted in no phase separation of starch upon ultrasonication. The better interfacial adhesion manifests in the benefits of better elongational properties in reactive blends making compositions of the invention comparable to 100%> PCL. Diffractive spectra showed that organoclay was intercalated and flocculated and increased flocculation was noted as organoclay content increased from 1% to 3%. Though not constrained by any specific theory, intercalation and flocculation suggest a mechanism of
tensile property improvement not only in the traditional manner of interacting with starch and PCL, but more importantly, in catalyzing the reactive extrusion. Melting and crystallization properties determined on PCL and the formulations of Table 1 are set forth in Table 4 below.
Table 4
Number of replicates, n=6. Same letters indicate same mean and different letters indicate significant difference at α=0.05 by t-test for independent means.
Relative crystallinity that is the heading of the right hand column of Table 4 is a measure of change in PCL crystallinity of PCL in starch-PCL blend compared to 100% PCL (relative crystallinity of 1.0) As shown in Table 4, crystallization temperature in blends subjected to reactive extrusion increased as the amount of organoclay increased from 1 to 9% but were lower than that of blend without organoclay not subjected to reactive extrusion and also compared to 100% PCL. As shown in Table 4, percent crystallinity increased with increasing organoclay content for the compositions obtained by reactive extrusion. The differences, however, were small. Thus, significant improvement in tensile elongation and changes in mechanical properties were not due to changes in PCL crystalline morphology. Since increased crystallinity in RB composites is associated with decreased crosslinking density in the presence of nanoclay, biodegradation rate was not significantly changed compared to PCL as highly crosslinked PCL showed reduced biodegradation rates in other studies. Damping behavior results evidenced the occurrence of crosslinking. Starch oxidation levels at different organoclay levels were determined and the results are set forth in Table 5 below.
Table 5
Number of replicates, n=3-4. Same letters indicate same mean and different letters indicate significant difference at α=0.05 by t-test for independent means.
As shown in Table 5, increased level of organoclay resulted in decreased level of oxidation in the products. The decreased level of oxidation was accompanied by increased compatibilization between starch and polyester and starch and clay and improved mechanical properties without grafting polyester on starch or polymerizing ester monomer directly on starch or using anhydride modifications as in other studies. Immersion testing according to ASTM D 543-95 (2001) in which water was used solvent showed STPCLPERl-0 without any structural integrity left after seven days immersion. On the other hand, STPCLPERI-6 remained intact at the seven day mark. Significant swelling. was observed in the STPCLPERI-0 samples while there was no swelling observed in the STPCLPERI-6 samples. The significance of this in terms of benefit for STPCLPERI-6 over STPCLPERI-0 is that blends with nanoclay showed a higher relative water resistance than those without nanoclay. The data show that the reactive extrusion improved interfacial adhesion between starch and PCL and that elongation of products herein was comparable to that of 100%) polyester and that strength and modulus levels remained the same as in starch-PCL composites without crosslinking. Starch-PAT (high molecular weight copolyester, Eastar Bio Ultra) nanocomposites prepared by the same reactive extrusion procedure as set forth above, followed the same trend in tensile properties as shown above, with almost a 5-fold increase in elongation over starch- PAT composites prepared without reactive extrusion.
Working Example II Scale-up was carried out from the low output (0.2 kg/hr) conical twin-screw co- rotating micro-extruder (DACA Instruments, Goleta, CA) (length/diameter, L/D = 5) used in Working Example I, to a continuous high output (-3-6 kg/hr) parallel twin-screw co-rotating extruder (Wenger TX-52, Wenger Inc., Sabetha, K.S). Screw speed, configuration and feed rates were optimized to achieve desired PCL-like elongation of reactive blends. Similar specific mechanical energies (SME's) were obtained when the TX-52 was operated at screw speeds l/3rd of that in the micro-extruder (40 rpm compared to 120 rpm in the microextruder). Higher screw speeds resulted in polymer shear degradation and lower mechanical properties. Smooth extrudates without steam expansion were obtained in the TX-52 at low feed rates (3-6 kg/hr). At high feed rates/outputs, mechanical properties of extrudates were lower than those at low feed rates/outputs. Screw configurations in the two extruders that resulted in equal mechanical properties were different. The conical micro-extruder had a forward screw formation while similar configuration in the TX-52 gave poor extrudate tensile properties. A series of 6-12 forward kneading discs staggered at 30° angles and located near the extruder feed port in the TX-52 was found to result in mechanical properties equal to those from micro-extrusion.
Variations The foregoing description of the invention has been presented describing certain operable and preferred embodiments. It is not intended that the invention should be so limited since variations and modifications thereof will be obvious to those skilled in the art, all of which are within the spirit and scope of the invention.