US 5,258,488 AGrant
Continuous Process for Manufacture of Lactide Polymers With Controlled Optical Purity
Issue Date:1993-11-02
•14 Claims
•8 Drawing Sheets
Abstract
A process for the continuous production of polylactide polymers from lactic acid which incorporates removal of water or a solvent carrier to concentrate the lactic acid feed followed by polymerization to a low-molecular-weight prepolymer. This prepolymer is fed to a reactor in which a catalyst is added to facilitate generation of lactide, the depolymerization product of polylactic acid. The lactide generated is continuously fed to a distillation system as a liquid or vapor wherein water and other impurities are removed. The resultant purified liquid lactide is fed directly to a polymerization process.
Metadata
Assignee
- Cargill, Incorporated
Inventors
- Patrick R. Gruber
- Eric S. Hall
- Jeffrey J. Kolstad
- Matthew L. Iwen
- Richard D. Benson
- Ronald L. Borchardt
Application Information
Application Number:US 9269010
Filing Date:1992-08-07
Priority Date:1992-01-24
Art Unit:153
Classifications
IPC:
C08G 6308C07D31912
Field of Search:
528525549526354;35741527467;68
Patent Drawings (8 sheets)
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to processes for the continuous production of lactide polymers from crude lactic acid in the field of biodegradable polymers.
2. Description of the Prior Art
The continued depletion of landfill space and the problems associated with incineration of waste have led to the need for development of truly biodegradable polymers to be utilized as substitutes for non-biodegradable or partially biodegradable, petrochemical-based polymers. The use of lactic acid and lactide to manufacture a biodegradable polymer is well known in the medical industry. As disclosed by Nieuwenhuis et al. (U.S. Pat. No. 5,053,485), such polymers have been used for making biodegradable sutures, clamps, bone plates and biologically active controlled release devices. It will be appreciated that processes developed for the manufacture of polymers to be utilized in the medical industry have incorporated techniques which respond to the need for high purity and biocompatibility in the final polymer product. Furthermore, the processes were designed to produce small volumes of high dollar-value products, with less emphasis on manufacturing cost and yield. It is believed that prior to Applicants' development, viable, cost-competitive processes for the continuous manufacture of lactide polymers from lactic acid having physical properties suitable for replacing present petrochemical-based polymers in packaging, paper coating and other non-medical industry applications were unknown.
It is known that lactic acid undergoes a condensation reaction to form polylactic acid when water is removed by evaporation or other means. The overall polymerization reaction is represented by: ##STR1## While step n of said polymerization reaction is represented by: ##STR2##
As Dorough (U.S. Pat. No. 1,995,970) recognized and disclosed, the resulting polylactic acid is limited to a low molecular weight polymer of limited value, based on physical properties, due to a competing depolymerization reaction in which the cyclic dimer of lactic acid, lactide, is generated. As the polylactic acid chain lengthens, the polymerization reaction rate decelerates until it reaches the rate of the depolymerization reaction, which effectively limits the molecular weight of the resulting polymers. An example of this equilibrium reaction is represented below. ##STR3##
Given this understanding, Dorough was convinced that high molecular weight polymers could not be generated directly from lactic acid. He was, however, successful in generating high molecular weight polymers from lactide, through the lactic acid dimer generated from the low molecular weight polymers of lactic acid. Because these polymers are generated from lactide, they are known as polylactides.
It is well known that lactic acid exists in two forms which are optical enantiomers, designated as D-lactic acid and L-lactic acid. Either D-lactic acid, L-lactic acid or mixtures thereof may be polymerized to form an intermediate molecular weight polylactic acid which, upon further polymerization, generates lactide as earlier disclosed. The lactide, or the cyclic dimer of lactic acid, may have one of three types of optical activity depending on whether it consists of two L-lactic acid molecules, two D-lactic acid molecules or an L-lactic acid molecule and a D-lactic acid molecule combined to form the dimer. These three dimers are designated L-lactide, D-lactide and meso-lactide, respectively. In addition, a 50/50 mixture of L-lactide and D-lactide with a melting point of about 126.degree. C. is often referred to in the literature as D,L-lactide.
DeVries (U.S. Pat. No. 4,797,468) recently disclosed a process for the manufacture of lactide polymers utilizing a solvent extraction process to purify lactide prior to polymerization. With DeVries, disclosure, the inventor recognized that existing literature recommends purification of lactide by several recrystallization steps. It is believed that processes prior to DeVries solvent extraction method, have generally utilized a recrystallization step to purify the crude lactide in order to obtain a source of lactide suitable for polymerization. However, processes utilizing such recrystallization steps are known to have relatively poor yields due to significant losses of lactide during the recrystallization steps. It is believed that producers of medical-related biodegradable products have not been concerned with such low yields because of the high margin generally expected for sales of such products and the lack of competitive alternatives. It will be appreciated, however, that in developing a process for the large-scale, commercial manufacture of biodegradable polymers, such as polylactides, for use in nonmedical-products-oriented applications where such polymers will necessarily compete with low-cost polymers made from petrochemicals, it will be important to maximize yield and minimize other overall cost factors to produce a biodegradable polymer which is cost-competitive.
The biogradable polylactide polymers must also possess physical properties suitable for application in non-medical products presently utilizing petrochemical-based polymers such as packaging materials, paper coatings and any other disposable articles. Nieuwenhuis et al. disclose that lactide polymers derived from polymerization of mixtures of the three lactides result in polymers with a variety of useful physical properties, including improved biodegradability. However, no commercially viable process for the large-scale manufacture of such lactide polymers is believed to have been disclosed to date.
Lactic acid is commercially available and manufactured from several known processes. Representative examples of such processes are disclosed by Glassner et al. (European Patent Application, EP 393818, Oct. 24, 1990), G Machell, "Production and Applications of Lactic Acid", Industrial Chemist and Chemical Manufacturer, v. 35, pp. 283-90 (1959) and Kirk Othmer, Encyclopedia of Chemical Technology, "Lactic Acid", v. 12, pp. 177-78 (2nd ed. 1963).
The optical activity of either lactic acid or lactide is known to alter under certain conditions, with a tendency toward equilibrium at optical inactivity, where equal amounts of the D and L enantiomers are present. Relative concentrations of D and L in the starting materials, the presence of impurities or catalysts and time at varying temperatures and pressures are known to affect the rate of such racemization.
Muller (U.S. Pat. No. 5,053,522) discloses that the preparation of optically pure lactide from an optically pure lactic acid feed is possible when utilizing appropriate conditions and catalysts. However, there is no teaching of a process which controls the optical purity of the resulting lactide to desired degrees or minimizes overall costs and maximizes yield of the lactide product. Furthermore, there is no disclosure of a commercially-viable lactide purification system, which allows production of polymer grade lactide, from crude lactic acid, which may subsequently be polymerized to produce a variety of non-medical-related polylactide polymers suitable for replacing existing petrochemical-based polymers.
Accordingly, a need exists for a continuous manufacturing process which utilizes commercially-available lactic acid to produce polylactide polymers suitable as a cost-competitive replacement for petrochemical-based polymers. The present invention addresses this need as well as other problems associated with the production of lactide polymers. The present invention also offers further advantages over the prior art, and solves other problems associated therewith.
SUMMARY OF THE INVENTION
The present invention provides a continuous process for the production of lactide polymers from a crude lactic acid feed source. The crude lactic acid feed may be any available combination of the optical enantiomers D-lactic acid and L-lactic acid in solution with a hydroxylic medium such as water or other solvent such as methanol, ethanol, propanol, butanol, isopropanol, isobutanol, or the like, or mixtures thereof. The source of lactic acid could also be an ester of lactic acid, such as methyl lactate, ethyl lactate, propyl lactate, butyl lactate, isopropyl lactate, isobutyl lactate or the like, or mixtures thereof. It is, however, recognized that the composition of the lactic acid feed source and the design and operating conditions of the process disclosed herein will affect the optical purity of the final polylactide polymer product. The process disclosed herein provides for the control of racemization to advantageously produce a polymer grade lactide of selected optical purity and composition. Because racemization can be controlled, it is possible to project the optical purity and composition of the resulting product based on that of the starting material. When polymerized, the resulting polylactide can have desirable physical properties for a wide variety of non-medical related applications. Furthermore, impurities such as color bodies, carbohydrates, proteins, amino acids, salts, metal ions, and other carboxylic acids and organic acids may be present in the crude lactic acid feed. Applicants process disclosed herein overcomes problems associated with producing a polymer grade lactide when such contaminants are present.
Referring now briefly to FIG. 1, which provides a preferred flowchart of the overall process disclosed herein, the crude lactic acid is first fed to an evaporator, continuously. Within the evaporator a portion of the water or solvent or any condensation reaction by-product is removed from the crude lactic acid. The water or solvent or any condensation reaction by-product is removed as a vapor from the evaporator and discarded or recycled. The evaporator thus concentrates the lactic acid in the crude feed. It is believed there will be some condensation reaction occurring and the lactic acid may start to form oligomers and low molecular weight polymers during the evaporation step, producing a condensation reaction by-product. This concentrated lactic acid is next fed to a prepolymer reactor, which in reality is a further evaporator.
It is well known in the art that as water or solvent are removed from a solution of lactic acid, the remaining lactic acid will begin to polymerize. In the prepolymer reactor, sufficient water or solvent and condensation byproducts such as water, ethanol, methanol, propanol, butanol, isopropanol, isobutanol and the like are removed to cause the lactic acid to polymerize to form lactic acid polymers having an average molecular weight of about 100 to about 5000, preferably about 200 to about 3000, and more preferably about 400 to about 2500. The water or solvent removed is recycled or discarded. In preferred embodiments, the water or solvent is recycled back to the evaporation process, because it may be contaminated with lactic acid. In this preferred embodiment, loss of feed material is prevented and the overall yield is increased.
It is recognized by Applicants that the evaporation and prepolymerization stages may be combined into one step. However, Applicants have discovered the benefit of utilizing two steps that allow first removing uncontaminated water or solvent in the evaporation step which is readily discarded or reused without treatment. The vapor stream from the prepolymerization reactor is greatly reduced in volume, yet contains some lactic acid. Recycling back through the initial evaporation step allows recovery of any lactic acid carryover, thus preventing loss of any valuable feed material.
The prepolymer product from the prepolymer reactor, polylactic acid or PLA, is fed to a lactide reactor. A catalyst is simultaneously and continuously fed to the lactide reactor. Many suitable catalysts are known, such as metal oxides, metal halides, metal dusts and organic metal compounds derived from carboxylic acids or the like. It is believed, any such catalyst may be utilized in the process disclosed herein. Polymer properties will, however, vary. In a preferred embodiment, the prepolymer and catalyst are mixed together in a static mixer to facilitate an even distribution of catalyst within the prepolymer. The solution within the lactide reactor would quickly come to an equilibrium distribution of lactide and polylactic acid with the temperature and catalyst employed. Heat is added to vaporize the crude lactide which is continuously removed from the lactide reactor, thus driving the depolymerization reaction, resulting in the net production of lactide as the contents of the lactide reactor seek equilibrium. It is believed that concentrations of unreactive high-boiling polylactic acid and other non-volatile impurities will concentrate in the solution within the lactide reactor. It is believed this will require a purge stream to remove such impurities.
In a preferred embodiment of the present invention, a portion of the purge stream of unreactive high boiling polylactic acid or other non-volatile impurities in the solution within the lactide reactor may be recycled to a point prior to the lactide reactor system or fed to polymerization. Based on experimental data which will follow hereinbelow, it is believed that any long chain lactic acid polymers will undergo transesterification to form lower molecular weight polylactic acids which may be utilized as a feed source to the lactide reactor. This allows further maximization of yield due to reduced loss of valuable feed material.
The crude lactide vapor is composed of a mixture of all three possible lactides: L-lactide, D-lactide, and meso-lactide, in various combinations. Along with the lactide, there is residual water, lactic acid and condensation reaction byproducts. This crude lactide may be fed directly to a distillation system as a vapor for purification. In a preferred embodiment, this stream is fed to a partial condenser in which the lactide condenses and the majority of the water and other impurities remain as vapors and are recycled back to the lactide reactor or other upstream process equipment such as the evaporator or prepolymer reactor. Preferably, the condensed crude lactide is fed directly to a distillation system for purification. Within this distillation system residual water and lactic acid are preferably removed as a distillate product and recycled back to the lactide reactor or other upstream process equipment such as the evaporator or prepolymer reactor. In addition, provision may be made to remove low molecular weight oligomers which may be present in the crude lactide or formed during distillation. The purified lactide is preferably fed to a polymerization reactor of conventional design.
The preferred overall process disclosed herein allows for the continuous manufacture of lactide polymers from a crude lactic acid with little or no waste of raw material lactic acid feed. This is accomplished by maintaining the crude lactide which was generated in the lactide reactor as a liquid or vapor and avoiding the yield loss associated with the recrystallization step traditionally used to purify the lactide. The purified lactide leaving the distillation system is further maintained as a liquid and fed into a polymerization process. Other monomers may be added to this purified liquid lactide prior to polymerization to achieve production of co-polymers of polylactide. Representative co-polymers are disclosed by P. Dave, N. Ashar, R. Gross, S. McCarthy, "Survey of Polymer Blends Containing Poly (3-hydroxybutyrate-co-16% hydroxyvalerate), Polymer Preparation, American Chemical Society, v. 31 (1), pp. 442-3 (1990); B. Riedl and R. Prud'homme, "Thermodynamic Study of Poly(vinyl chloride)-Polyester Blends by Inverse Gas Phase Chromatography", J. Polymer Science, Part B, vol. 24(11), pp. 2565-82 (1986); H. Younes and D. Cohn, " Phase Separation in Poly(ethylene glycol)/Poly(lactic acid) Blends, European Polymer J., v. 24(8), pp. 765-73 (1988); Smith et al. (European Patent Application, EP 209371, Jan. 21, 1987); Pines et al. (European Patent Application EP 109197, May 23, 1984); J. Zhu, Y. Shao, W. Sui, S. Zhang, H. Xiao and X. Tao, "Homopolymers and Copolymers of Glycolide and Lactide", C-MRS Int. Symp. Proc. Meeting Date 1990, v. 3, pp. 387-90 (1990); Jarrett et al. (U.S. Pat. No. 4,788,979); and, T. Nakamura et al., "Surgical Application of Biodegradable Films Prepared from Lactide-Epsilon-Caprolactone Copolymers, Advanced Biomaterials, 7 (Biomater. Clin. Appl.) pp. 759-64 (1987).
Applicants believe any monomer capable of copolymerizing with lactide may be used with the process disclosed herein.
In particular, this system allows recovery of any meso-lactide which may be present or formed within the disclosed process and which is normally lost in a recrystallization process. Further, the problems associated with handling solid materials are eliminated. These problems are well-documented by D. D. Deane and E. G. Hammond in "Coagulation of Milk for Cheese-Making by Ester Hydrolysis", J. Dairy Science, v. 43, pp. 1421-1429 (1960) and Nieuwenhuis et al. (U.S. Pat. No. 5,053,485) which are incorporated herein by reference. The problems of storing such solids for any time period are also disclosed by Deprospero et al. (U.S. Pat. No. 3,597,449) and F. E. Kohn et al. in J. of Applied Polymer Science, Vol. 29, 4265-4277 (1984) which are incorporated herein by reference. These problems include contamination by water vapor which would lead to ring-opening side reactions causing the lactide to convert to lactic acid. The presence of lactic acid in the feed to the final polymerization step will result in polymers of limited molecular weight.
It is believed that the prior art does not teach use of distillation to purify crude lactide streams. Applicants believe that one would not turn to utilization of distillation due to the narrow differences between melting point and boiling point of lactide streams, which potentially could cause solid plugging problems within a distillation system. Furthermore, side reactions in which the lactide ring is opened and polymers of lactic acid are formed may occur during distillation. It is believed, the presence of such side reaction products would lead to undesirable molecular weight limitations in the final polymer product. Applicants have discovered that proper design and control of a distillation system coupled with direct feed of a crude lactide vapor stream or a liquid crude lactide stream after partial condensation to remove water and lactic acid vapor allows purification of crude lactide in a conventional distillation system. Previous to this disclosure, applicants believe, any polymer made from non-optically pure lactide relied on blending the various lactide components, each of which had been purified separately using recrystallization of a crude lactide produced by other techniques.
The present system also allows use of crude lactic acid streams which contain impurities. As designed, the present system allows for removal of both low boiling and high boiling impurities prior to distillation of the crude lactide stream which is subsequently polymerized. The prior art fails to disclose a process with such advantages. Further, Applicants have found that impurities may prevent catalyst activation. In a preferred embodiment, this problem is overcome by first activating the catalyst by heating a mixture of the catalyst and purified lactic acid or lactide, then feeding such activated catalyst with the crude lactic acid feed. It is believed the prior art contains no such teaching.
These and various other advantages and features of novelty which characterize the present invention are pointed out with particularity in the claims annexed hereto and forming a part hereof. However, for a better understanding of the invention, its advantages, and the objects attained by its use, reference should be made to the drawings which form a further part hereof, and to the accompanying descriptive matter, in which there are illustrated and described preferred embodiments of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
In the drawings, in which like referenced numerals indicate corresponding parts or elements of preferred embodiments of the present invention throughout the several views;
FIG. 1 is a flow diagram of the preferred overall process steps of the present invention;
FIG. 2 is a detailed schematic representation of a preferred polylactide polymer production system in accordance with the present invention;
FIG. 3 is a graph showing the effects of sodium on lactide production rate and optical purity;
FIG. 4 is a graph which represents the relationship between the optical purity of generated lactide in relation to the molecular weight of the feed to the lactide reactor;
FIG. 5 is a graph showing the relationship between catalyst concentration and optical purity of the resulting lactide;
FIG. 6 is a graph showing the effect of hydroxyl impurities on polymer molecular weight at different temperatures;
FIG. 7 is a graph showing the effect of hydroxyl impurities on polymer molecular weight at different catalyst concentrations; and,
FIG. 8 is a graph showing the equilibrium lactide concentration as a function of temperature.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
As required, detailed embodiments of the present invention are disclosed herein. However, it is to be understood that the disclosed embodiments are merely exemplary of the present invention which may be embodied in various systems. Therefore, specific details disclosed herein are not to be interpreted as limiting, but rather as a basis for the claims and as a representative basis for teaching one skilled in the art to variously practice the present invention.
Referring now to the figures, FIG. 2 represents an overall schematic flowchart encompassing the preferred process disclosed herein. A crude lactic acid feed (2) is provided. The crude lactic acid feed may be of various grades. This could include USP, food grade, or any other solution in a hydroxylic medium. A hydroxylic medium is a medium which contains molecules having a hydroxyl group, mediums such as water, methanol, ethanol, propanol, butanol, isopropanol, isobutanol and the like, preferably having a number of carbon atoms in a range from 0-4, more preferably in a range from 0-2. The crude lactic acid can include from about 1% to about 99% by weight lactic acid, preferably, from about 1% to about 85%, more preferably from about 5% to about 50%. In a preferred embodiment, the crude lactic acid feed is a solution of about 15% lactic acid and about 85% water which is commercially produced. Many manufacturing processes for producing crude lactic acid are known in the art, such as Glassner et al., (European Patent Application, EP 393818, Oct. 24, 1990); G. Machell, "Production and Applications of Lactic Acid", Industrial Chemist and Chemical Manufacturer, v. 35, pp. 283-90 (1959) and Kirk Othmer, Encyclopedia of Chemical Technology, "Lactic Acid", v. 12, pp. 177-78 (2nd ed. 1963), which are incorporated herein by reference. In an alternative embodiment, the source of crude lactic acid feed (2) could be in the form of the ester of lactic acid such as methyl lactate, ethyl lactate, and the like. These esters are known intermediate products of the lactic acid process disclosed above and incorporated herein by reference.
It is well known in the art that lactic acid includes two optical isomers, the L and D enantiomers. Either optical isomer or any combination thereof may be utilized as a crude lactic acid feed to the present reactor system. Furthermore, Applicants recognize that the crude lactic acid feed (2) may contain other impurities, such as color bodies, carbohydrates, proteins, amino acids, salts, metal ions, and other carboxylic acids or organic acids. As will be explained in greater detail below, and in Example 1, the overall system incorporated in the present invention includes outlets for such impurities so that their impact on final polymer products is limited. These outlets are designated as (10), (28), (99), (71) and (66) of FIG. 2. Each will be discussed in greater detail below. Thus, in preferred embodiments the need for a more costly purified crude lactic acid feed is eliminated.
A fluid transfer mechanism (4) is provided to transport the crude lactic acid feed (2) through an optional static in-line mixer (6) in a pipeline (8) to an evaporator system (22). The evaporator system (22) is utilized to concentrate the crude lactic acid feed (2) by removing water or any other solvent or hydroxylic medium which is used as a carrier for the lactic acid, such as methanol, ethanol or the like and any condensation reaction by-products. The evaporator system (22) may be of any conventional type known in the art, such as a multiple effect evaporator, a wiped film evaporator, a falling film evaporator, or any other conventional system. It is appreciated that such systems may be operated at pressures below atmospheric pressure, at atmospheric pressure or above atmospheric pressure with commensurate changes in heat load and operating temperatures. In a preferred embodiment, vacuum evaporation is utilized to reduce racemization. Water vapor or solvent vapor, in reference to the hydroxylic medium or condensation reaction by-product, is removed from the evaporator via a transfer line (18), and condensed in a condenser (16). The condensed liquid is transferred in a pipeline (14) to a fluid transfer mechanism (12), such as a pump or the like. The fluid transfer mechanism (12) transports the condensed water or solvent via pipeline (10) as a waste stream and is discarded.
The lactic acid may be concentrated to a weight percent lactic acid of about 50% to about 99%, preferably from about 75% to about 99% and more preferably from about 85% to about 99%. In a preferred embodiment, the evaporator system (22) is utilized to concentrate the crude lactic acid feed from about 15% lactic acid up to about 85% lactic acid.
The concentrated lactic acid is transferred via a fluid transfer mechanism (24) through a pipeline (26) to a prepolymer reactor (38). The prepolymer reactor (38) is essentially a second evaporator system of any conventional type which is utilized to further remove water or solvent from the crude lactic acid feed. A portion of the water or solvent vapor now also includes such water or solvent produced from the lactic acid polymerization reaction previously disclosed, the condensation reaction by-product. The water or solvent vapor leaves the prepolymer reactor (38) via line (32) and is condensed in a condenser (30). The condensed liquid is transferred via pipeline (36) to a transfer mechanism (34), with the transferred liquid comprising water or solvent with small amounts of lactic acid and other impurities present. This liquid may be discarded through line (28) or may be recycled through line (29) back to a static mixer or other mixing mechanism and fed once again through line (8) to the evaporator (22). The remaining liquid in the prepolymer reactor is continuously transferred via transfer mechanism (40) through line (42) to a hold tank (44).
As previously disclosed, it is well recognized in the art that lactic acid undergoes a condensation reaction to form polylactic acid, the polymer of lactic acid, as water is removed. In a preferred embodiment of the present system, the prepolymer reactor (38) is utilized to remove adequate water or solvent and condensation reaction by-product from the lactic acid to cause polymerization up to a molecular weight of about 100 to about 5000, preferably about 200 to about 3000, and more preferably about 400 to about 2500. As will be detailed in Example 2, which follows, in preferred embodiments the molecular weight of the polylactic acid leaving the prepolymer reactor impacts the chemical purity as well as the optical purity of the crude lactide. This in turn will affect the distillation and the properties of the final polymer product
Applicants recognize that the evaporator system (22) and the prepolymer reactor (38) could be combined into a single system which provided removal of water or solvent sufficient to concentrate the lactic acid feed and also to polymerize such lactic acid. In the preferred embodiment, as discussed above, the systems are separate to take advantage of recognized differences in the composition of the vapor leaving the evaporator (22) at line (18) and the vapor leaving the prepolymer reactor (38) in line (32). The first step of concentrating the crude lactic acid in the evaporator (22) from 15% lactic acid to 85% lactic acid results in substantially pure water or solvent leaving the evaporator in line (18), which may be readily discarded without treatment. The vapor in line (32) leaving the prepolymer reactor (38) will necessarily contain lactic acid and other impurities which are carried over in the evaporation process. These impurities will preferably need to be recycled or treated before discarded. Thus, in the preferred embodiment, Applicants take advantage of the economic benefits of removing nearly pure water (or solvent) in the evaporator (22) and reduced recycle or waste treatment of the vapor leaving the prepolymer reactor (38).
Applicants also recognize that the evaporator system (22) and prepolymer reactor (38) may be replaced by a series of batch evaporators that concentrate the lactic acid and produce prepolymer. The series of batch systems may be operated to provide a net continuous supply of prepolymer.
The hold tank (44) is maintained at a temperature sufficient to keep the polylactic acid in a flowable liquid state. The hold tank (44) is, however, only a feature of the preferred embodiment, recognizing the control problems which may result from direct feed to the lactide reactor (60). This liquid is transferred via transfer mechanism (46) through a pipeline (48) to a static in-line mixer or other mixing mechanism (50). Within the mixing mechanism (50) a catalyst is added to the polylactic acid. Applicants recognize that any means of adding the catalyst to the polylactic acid would be appropriate; however, the static mixer (50) utilized in the preferred embodiment allows more even distribution of the catalyst within the liquid. The catalyzed polylactic acid is transferred via transfer line (54) to the lactide reactor (60). It is well recognized in the art that polylactic acid maintains a dynamic equilibrium with its depolymerization product, lactide, as represented by the reaction below: ##STR4##
It is further recognized that this reaction is catalyzed by such compounds as tin dust, tin halide, tin oxide, and organic tin compounds derived from C.sub.1 -C.sub.20 carboxylic acids, as disclosed by Muller in U.S. Pat. No. 5,053,522, which is incorporated herein by reference. Other metals, such as zinc, and their halides, oxides and organic compounds thereof, have been recognized by the art as possible catalysts for the lactide reaction. It is believed any metals of Groups IV, V or VIII of the Periodic Table and compounds thereof, are possible catalysts for generating lactide. In a preferred embodiment tin oxide is utilized as catalyst. In a most preferred embodiment the catalyst is activated prior to feed.
As lactide is generated within the lactide reactor (60), it is removed as a vapor continuously through line (58). The removal of lactide further drives the depolymerization reaction. It is believed that some high-boiling or non-volatile contaminants present in the feed to the entire system will concentrate in the lactide reactor and necessitate inclusion of a purge stream (62). Example 1 below details the effect of cationic impurities and Example 11, the detrimental effects of concentrating metal ions in the lactide reactor (60). It is believed other impurities would have a similar effect and necessitate the purge stream (62). A portion of this purge stream (56) may be recycled back to the static mixer (50) and recatalyzed and fed to the lactide reactor (60). Alternatively, the purge stream may be fed to the polymerization reactor (110) via line (68), if such polymer is desired. A transfer mechanism (64) is provided to transport the purge stream optionally to a point prior to the evaporator (22) such as the static mixer (6) or to the feed line (26) to the prepolymer reactor (38) or to waste through line (66) or to a static mixer (104) for polymerization in a polymerization reactor (110).
The lactide vapor leaving the lactide reactor (60) in line (58) is partially condensed in a condenser (72). The uncondensed vapor consists of residual lactic acid and water or solvent, along with some lactide which remains uncondensed. This vapor stream leaves the system through line (90) and is condensed in a condenser (92), the liquid thus being transferred by transfer mechanism (96) through line (94). This liquid may optionally be discarded through line (99) or recycled through line (98) back to the crude lactic acid static mixer (6). In the preferred embodiment, this stream is recycled in order to recover and utilize any lactic acid or lactide which is not condensed in partial condenser (72).
The condensed crude lactide leaving condenser (72) via line (74) is transferred via fluid transfer mechanism (76) through line (78) to a distillation system (80) for purification of the lactide. Applicants recognize that partial condensation may not be necessary and the crude lactide vapor may be fed directly to the distillation system (80). The distillate leaving the distillation system (80) in line (82) is composed of water or solvent, some residual lactic acid, and some lactide carryover. This stream is condensed in condenser (84) and transferred via fluid transfer mechanism (88), in line (86), and may be discarded or recycled back to a point prior to the evaporator (22), such as the static mixer (6) or the feed line (26) to the prepolymer reactor (38) through line (71), or more preferably recycled in line (70) back to the static mixer (50) to be recatalyzed and re-fed to the lactide reactor (60). This preferred embodiment allows minimization of waste by preventing loss of lactic acid or converting lactic acid to lactide from the feedstock.
The refined lactide is removed from the distillation system (80) via transfer mechanism (100) in line (102) and fed to a polymerization reactor (110). Applicants recognize that the distillation system (80) may include more than one distillation column or a flash drum. The polymerization process may be of any conventional design known to the art, such as that disclosed by J. Leenslag and A. Pennings, "Synthesis of High Molecular Weight Poly (L-lactide) Initiated with Tin 2-Ethylhexanoate", Makromol. Chem., v. 188, pp. 1809-14 (1987) and F. Kohn et al., "The Ring-Opening Polymerization of D,L-Lactide in the Melt Initiated with Tetraphenyltin, J. Applied Polymer Science, v. 29, pp. 4265-77 (1984), which are incorporated herein by reference.
Applicants recognize that in a preferred embodiment one may choose to add a non-lactide monomer to the purified lactide leaving the distillation system (80). This co-monomer may be added via line (101). The co-monomers are fed to the polymerization reactor (110) and polymerized to form a co-polymer. Many co-polymers of polylactide are known to the art. These include P. Dave, N. Ashar, R. Gross, S. McCarthy, "Survey of Polymer Blends Containing Poly (3-hydroxybutyrate-co-16% hydroxyvalerate), Polymer Preparation, American Chemical Society, v. 31 (1), pp. 442-3 (1990); B. Riedl and R. Prud'homme, "Thermodynamic Study of Poly(vinyl chloride)-Polyester Blends by Inverse Gas Phase Chromatography", J. Polymer Science, Part B, vol. 24(11), pp. 2565-82 (1986); H. Younes and D. Cohn, "Phase Separation in Poly(ethylene glycol)/Poly(lactic acid) Blends, European Polymer J., v. 24(8), pp. 765-73 (1988); Smith et al. (European Patent Application, EP 209371, Jan. 21, 1987); Pines et al. (European Patent Application EP 109197, May 23, 1984); J. Zhu, Y. Shao, W. Sui, S. Zhang, H. Xiao and X. Tao, "Homopolymers and Copolymers of Glycolide and Lactide", C-MRS Int. Symp. Proc. Meeting Date 1990, v. 3, pp. 387-90 (1990); Jarrett et al. (U.S. Pat. No. 4,788,979); and, T. Nakamura et al., "Surgical Application of Biodegradable Films Prepared from Lactide-Epsilon-Caprolactone Copolymers, Advanced Biomaterials, 7 (Biomater. Clin. Appl.) pp. 759-64 (1987), which disclosures are incorporated herein by reference. Applicants believe any co-polymers of polylactide may be produced from the process disclosed herein.
Fluid transfer mechanisms disclosed throughout this detailed description would normally be a pump. However, Applicants recognize that through design choices other mechanisms for transfer, such as gravitational flow, may also be utilized.
Applicants further recognize that the preferred overall system described herein is a complex combination of many known chemical engineering unit operations. So that the benefit of the overall combination may be recognized, Applicants herein disclose in further detail the selection, operation, and benefits of selecting such unit operations, along with actual laboratory experimental results exemplifying the disclosed advantages.
As previously stated, the crude lactic acids fed to this process (2) may be made up of L-lactic acid or D-lactic acid, or combinations thereof. The composition of the feed, however, does not translate directly through the entire process to define the composition of the polymer product leaving the polymerization reactor (110) through line (108). Applicants recognize that racemization, or conversion of one optical enantiomer to the other, may occur. It is believed that such racemization is driven by such factors as temperature, pressure, time at a given temperature or pressure, the presence of catalysts or impurities, and relative concentrations of the two enantiomers at any given time. The degree of racemization is defined herein by the percent conversion of the optical enantiomer that is present in excess of 50%. As an equation, this calculation would be defined as: ##EQU1## Thus, an initial composition of 75% L and 25% D, which results after racemization to a 50% L, 50% D mixture, would equate to a degree of racemization of 100%. In all instances, no matter what initial composition, a 100% degree of racemization coincides with a composition of 50% each enantiomer, or optical inactivity. This recognizes the tendency toward equilibrium at a 50% of concentration of each enantiomer, corresponding to optical inactivity. In the most preferred embodiment of the system, each unit operation is controlled to a degree that allows production of a purified lactide mixture with selected chemical and optical composition. The optical composition of the lactide mixture is determined by the relative abundance of D- and L-lactic acid sub-units in the polylactic acid within the lactide reactor. As recognized by Nieuwenhuis et al. in U.S. Pat. No. 5,053,485, the disclosure of which is incorporated herein by reference, the blend of lactide isomers used to produce the polymer affects the physical properties of the polymer, including the biodegradability.
In a preferred embodiment, the evaporator (22) is operated to minimize residence time so that there is little or no effect on optical purity. The prepolymer reactor (38) is also operated to minimize racemization. This includes reducing the residence time within the reactor.
The crude lactide produced in the lactide reactor (60) will be a mixture of the three possible lactides which may be generated from L- and D-lactic acid. These include an L-lactide, a D-lactide, and meso-lactide.
As detailed in Example 3 hereinbelow, the concentration of catalyst added to static mixer (50) also affects the degree of racemization and composition of the crude lactide product. In a preferred embodiment, the catalyst concentration level is adjusted based on desired properties of the final polymer product.
Applicants have discovered, and detailed in Example 12, that the quality of the crude lactide charged to the distillation system has a significant effect on the operation of said system. In particular, acidic impurities such as lactic acid and low molecular weight oligomers, which are formed by ring opening reactions of lactic acid or water with lactide, can cause premature polymerization in the distillation system. In a preferred embodiment, applicants believe such side reactions may be controlled by partially condensing the crude vapor prior to feeding to distillation to remove impurities.
The distillation system (80) may also be operated to control racemization of the lactide and other side reactions. In a preferred embodiment, this system is designed to minimize racemization by utilizing a packed column distillation system which minimizes liquid holdup, along with a thermal-siphon reboiler which limits residence time of the bottom liquids, and utilizing a minimum reflux ratio to further reduce holdup time in the column. It is, however, recognized that other distillation systems may be utilized with varying impact on the optical purity of the purified lactide and resultant polymer product.
In a preferred embodiment, the distillation system (80) is utilized as a purification step for the lactide so that crystallization of the crude lactide is unnecessary in order to produce a lactide product of suitable purity for polymerization. The lactide reactor (60) is also designed in a preferred embodiment, maximizing surface area between liquid and vapor so that liquid lactide can more easily vaporize. This allows for rapid removal of the generated lactide, which in turn further drives the reaction. Furthermore, as recognized by DeVries in U.S. Pat. No. 4,797,468, which is incorporated herein by reference, a system which utilizes purification steps other than crystallization increases yield. The use of distillation as a purification step also prevents the need to handle solids with the problems with equipment and contamination inherent in such operations.
The following examples further detail advantages of the system disclosed herein:
EXAMPLE 1
Effect of Cationic Impurities on Optical Purity
Na.sub.2 S was added at levels of 20, 200, and 1000 ppm to purified lactic acid (Purac heat stable grade) of composition 85% L-lactic acid and 15% water. The lactic acid was then polymerized to form PLA with a number average molecular weight of about 650 g/mol. H.sub.2 S was removed under vacuum while the PLA was being formed, leaving Na.sup.+ ions in solution. The PLA was then used to generate lactide at 10 mm Hg with 0.05 wt-% SnO catalyst (Aldrich cat. no. 24,464-3, Tin (II) oxide, 99+%) and a constant heat input of 75%, allowing reactor temperature to float. The results are shown in the following table and in FIG. 3:
The table shows that although the addition of sodium had no effect on the lactide production rate or on the reaction temperature, it did have a pronounced effect on the amount of meso-lactide present in the crude product.
Applicants believe other cationic species will behave in a similar fashion.
A practical implication of this example is that it will be necessary in any continuous process, where management of optical composition is desired, to provide a mechanism for removing ionic impurities from the reactor. The ionic impurities will be present in all sources of lactic acid to some extent, and will concentrate in the liquid within the lactide reactor (reactor bottoms) over time. In a preferred embodiment of the present invention, a purge stream is provided to accomplish this objective. An alternative would be to shut the system down periodically and dump or recycle the reactor bottoms.
EXAMPLE 2
The Effect of PLA Molecular Weight on Optical Purity
Lactide was generated from several samples of PLA (also known as polylactic acid or prepolymer) each sample having a different average molecular weight. The conditions under which the lactide was generated in each experiment include: 10 mm Hg pressure, 0.05 wt-% SnO as catalyst (same catalyst as Example 1), run to approximately 73% conversion of PLA to lactide, constant heat input and stirring speed. The reaction temperature was allowed to float, reaching the value necessary to dissipate the heat input. This method of running tends to give a constant lactide production rate, within limits. Five samples of PLA were used, a reference case with molecular weight of 640 g/mol (determined by end group titration) and four additional cases obtained by further heating of the PLA under vacuum at 200.degree.-220.degree. C. The results of the experiments are tabulated below and shown in FIG. 4.
Increasing the feed molecular weight results in a very clear decrease in the concentration of lactic acid in the crude lactide. This is a processing benefit because it will be easier to achieve polymerization grade lactide from a cleaner starting material. However, as the table and figure show, the concentration of meso-lactide in the crude lactide increase significantly. The optimal operation will require careful balancing of these two factors based on the desired final polymer product. In particular, if lactide of high optical purity is desired then the process should be run with a low molecular weight feed.
EXAMPLE 3
The Effect of Catalyst Concentration on Optical Purity
Lactide was produced using SnO catalyst (catalyst of Example 1) from PLA with a molecular weight of 650 g/mol at several catalyst levels and two pressures. The power input was held constant (Varian setting of 75) and the reaction temperature was allowed to seek an equilibrium level. The percentage meso in the crude lactide for each experiment is shown in the table below and graphically in FIG. 5.
It is clear from this data that Catalyst levels above 0.1 wt-% lead to an increase in the content of meso-lactide, for both pressures studied. The increase occurs even though the overall lactide generation rate increases and the reaction temperature decreases. The content of meso-lactide also increases at very low concentrations of catalyst, resulting in a minimum for 0.02-0.10 wt-% SnO. In a preferred embodiment, catalyst concentrations may be varied depending on desired final polymer physical properties.
EXAMPLE 4
The Effect of Recycling Lactide Reactor Bottoms
Example 1 demonstrated that some form of purge of the liquid in the reactor (reactor bottoms) would probably be necessary for a continuous operation. This example demonstrates a surprising benefit if such purge stream is recycled back to the catalyst addition stage.
PLA was produced from the same lactic acid used in Example 1, utilizing the same method. This was used to generate lactide at 1 mm Hg with 0.05 wt-% SnO as catalyst (catalyst of Example 1). The reaction was run to 72% conversion, at which point the lactide production rate had begun to decline significantly. The heat was turned off and the flask was cooled to 150.degree. C. under a nitrogen atmosphere. 390 gms of 88% L-lactic acid was added to the 80 gms of residue. The flask was adapted to generate a new batch of PLA, simulating the recycle of reactor bottoms to the evaporator section. The new PLA was heated under vacuum until the molecular weight was about 650 g/mol (by end group titration). A GPC (gel permeation chromatography) analysis showed that the reactor bottoms had been completely broken down and been reabsorbed into the PLA, with no sign of high molecular weight reactor bottoms. The lactide generation was rerun using the same conditions as before, and it was surprisingly found that the lactide production rate increased: from 0.86 hr.sup.-1 for the first run (prior to recycle) to 1.03 hr.sup.-1 for the second run (recycle, no additional catalyst). The reaction temperature was 213.degree. C. for the first run and 215.degree. C. for the second run. The composition of the crude lactide and of the reactor bottoms were similar for the two cases. Thus, in a preferred embodiment the reactor bottoms is recycled to a point prior to the lactide reactor to increase overall yield from the lactic acid feed.
EXAMPLE 5
Polymerization Technique
The lactide is a mixture of 80% L- and 20% D,L-lactide, recrystallized to high purity. 40 gm of lactide is charged to a flask with magnetic stirring. A THF (Tetrahydrofuran, Burdick and Jackson, high purity, non-spectro) solution containing L-lactic acid, water, or both is added to the lactide. The flask is lowered into an oil bath at 140.degree.-160.degree. C. to melt and mix the monomer. This is held for five minutes after complete melting (about 15 minutes total). A starting sample is pulled for GC and/or water analysis. A catalyst solution of 10 wt. % Tin(II) 2-Ethylhexanoate (Johnson Mathey Electronics, Tech. Grade) in toluene is added and allowed to react for 1 minute. Five gram samples are then pipetted into silanized and nitrogen flushed 20 ml vials. These are quickly placed into a temperature controlled oil bath. Vials are pulled and frozen at various time intervals up to 4 hours.
The samples are prepared for analysis by breaking the polymer out of the vials and dissolving in THF at room temperature on an orbital shaker (about 1-6 hours for 5 grams of polymer in 125 mls THF). The mixture is then diluted to 1% in THF and analyzed utilizing GPC analysis to determine the molecular weight and percent conversion.
EXAMPLE 6
Polymer Molecular Weight is Controlled by Impurity Level and is Independent of Temperature
Experiments were conducted at three different temperatures with two levels of impurities, using the polymerization technique of Example 5. In each case, a projected molecular weight which the polymer would achieve at 100% conversion was determined by GPC analysis of the highest conversion sample and corrected for the unconverted monomer. This method has been shown to give reproducible values and accurately corrects for any effect of sampling at different conversion levels. The results of the experiments are tabulated below and shown graphically in FIG. 6.
A statistical analysis of variance showed that the molecular weight of the polymer was controlled solely by the level of impurities, with temperature having no effect. Thus, in a preferred embodiment hydroxyl impurities are controlled to desired levels to control the physical properties of the resulting polymer product.
EXAMPLE 7
Polymer Molecular Weight is Controlled by Impurity Level and is Nearly Independent of Catalyst Concentration
The polymers were prepared at 160.degree. C. using the polymerization technique of Example 5. Two levels of water (H=5.9-8.8 meq./mol., L=1.8-3.7 meq./mol.) and two levels of lactic acid (H=0.9-1.3 meq./mol., L=0.1-0.2 meq../mol.) were used in a duplicated factorial design experiment at each of two different levels of catalyst (0.0002 mol/mol; and 0.0004 mol/mol) (eight experiments total). Projected molecular weights were calculated as in Example 6. The results are shown in tabular form below and graphically in FIG. 7.
An analysis of variance revealed that the change in hydroxyl content accounted for 91% of the variance in the molecular weight, while the change in catalyst concentration accounted for only 4% of the variance. Both effects were found to be statistically significant.
These data show, in a preferred embodiment, the critical need to control the level of hydroxyl containing impurities in the lactide in order to control the molecular weight of the final polymer.
EXAMPLE 8
Equilibrium Concentration of Lactide in Polylactic-Acid
PLA of 650 MW was heated at atmospheric pressure with either 0.00, 0.05, or 0.15 wt % SnO as a catalyst (Catalyst of Example 1). The mixtures were held at three different desired temperature for 20 minutes, at which time 10 wt % of purified L-lactide was added to the mixture with stirring. The vessel was fitted with a condenser to prevent the loss of water or other volatile components. Samples were removed from the reaction vessel at times ranging from 5 minutes to 450 minutes and were analyzed using an Ultrastyragel.RTM.100A GPC column (Waters Chromatography, a division of Millipore Corp.) with THF as the mobile phase to determine the concentration of lactide. The concentration data were fit to a simple first order decay model using a non-linear regression software package (SAS Institute, Inc.) to determine the equilibrium values. The resulting projected values for the equilibrium concentrations of lactide are shown in the table below and plotted graphically in FIG. 8. The results show the beneficial effect of rapid removal of lactide from the lactide reactor in preferred embodiments to further drive the lactide generation reaction.
EXAMPLE 9
Relative Rates of Racemization
Samples of PLA (with and without SnO (Catalyst of Example 1) as catalyst) and lactide were heated and stirred for four hours at 200.degree. C. at atmospheric pressure in a round bottom flask fitted with a condenser to prevent loss of volatile components. The samples were then allowed to cool and the optical purity of the PLA was determined by saponification followed by a measurement of the optical rotation. The lactide sample was analyzed by GC to determine the meso-lactide content, which was then converted to a measurement of optical purity.
The results of this experiment demonstrate that racemization occurs fastest in PLA which is exposed to catalyst. Thus, in the most preferred embodiment racemization is controlled in the lactide generating reactor. It is however recognized that another area of racemization control will be the evaporators which are used to prepare PLA, because of the long residence times and the possible inclusion of catalyst and catalyzing impurities. In a preferred embodiment the residence time of the lactide in the distillation column will be kept low, minimizing the potential for racemization.
EXAMPLE 10
Effect of Mass Transfer Efficiency on Lactide Composition
Lactide was produced from PLA at several catalyst levels and at two pressures to determine the effect of mass transfer. The catalyst was SnO (Catalyst of Example 1) and the power setting of the Varian was 75%.
The table below shows the effect of changing mass transfer efficiency by adjusting the pressure (vapor phase lactide concentration). Note that the reaction temperatures were similar for each pair of cases.
The increased mass transfer efficiency at 1 mm Hg vs 10 mm Hg results in significantly higher net lactide production rates and a lower concentration of meso-lactide. In a preferred embodiment the lactide reactor is operated under vacuum to facilitate mass transfer.
EXAMPLE 11
The Effects of Metal Contaminants Concentrating In the Lactide Reactor
Lactic acid was concentrated and polymerized to form low molecular weight polylactic acid (MW range of about 600-2200) and fed to a continuous pilot scale reactor for the production of lactide. At the end of 1-week of operation a sample of the reactor liquid was taken and analyzed for metals. The results are shown below.
The metals profile clearly shows corrosion of the stainless steel reaction system, either in the formation of the prepolymer or in the lactide generating reactor.
The high metals content, which represents the build-up over a week with no purge on the reactor bottoms, is detrimental to the lactide formation process. The data below demonstrate this effect.
Three lactide runs were made following the usual laboratory process. A control using 650 MW PLA, the control with added iron and chromium (1000 ppm iron from FeCl.sub.3 *6H.sub.2 O, 1000 ppm iron from FeSO.sub.4 *7H.sub.2 O, and 1000 ppm chromium from CrCl.sub.3 *6H.sub.2 O), and the reactor bottoms sample (initial MW 2000). Fresh catalyst, 0.05 wt % SnO (Catalyst of Example 1), was added to each sample and lactide was generated at 10 mm Hg with a reactor temperature of 230-240.degree. C.
The runs with elevated metals content had lower yield and much higher weight average molecular weight at the end of the reaction, demonstrating the detrimental effects of a high metal content.
It is believed that in a preferred embodiment, a purge of the reactor bottoms will alleviate this problem.
EXAMPLE 12
The Effect of Acidic Impurities on Distillation
Lactide was produced in a continuous pilot plant at rates of 2-5 kg/hr. The starting materials were Purac lactic acid of about 85% concentration. A PLA prepolymer having a range of molecular weights from about 400-2000 MW was made batchwise by heating first at atmospheric pressure and then under vacuum. The prepolymer was used to supply the continuous feed to the lactide reactor. The reactor was run at a temperature of 220.degree.-240.degree. C. and pressure of about 35 mm Hg.
Two samples of lactide were distilled in a 2000 ml three-neck flask with mechanical stirring. The lactide was taken overhead through a 2 cm ID by 30 cm glass column with stainless steel packing. Reflux was not controlled, but the column was insulated. The rate of distillation ranged from about 150-370 gms/hr. After taking approximately 80% overhead, the bottoms were analyzed by GC to determine the concentration of oligomers and to calculate the amount of polymerization (based on feed) which had occurred. The table below shows the relationship between the concentration of acidic impurities in the crude lactide and the extent of polymerization during distillation. The data show the effect of acidic impurities on final polymer molecular due to the increased oligomer content in the purified lactide.
EXAMPLE 13
The Beneficial Effects of Catalyst Activation on Lactide Generation Rates
Three grades of PLA were evaluated for the production of lactide using various catalysts. The sources of the PLA were: A) Purac heat stable grade lactic acid, B) a test sample of lactic acid from Lactech, Inc. which was produced by fermentation, and A/B) a 50/50 mix of the PLA produced from each of the previous two sources.
Lactide was generated in a laboratory apparatus (three-neck flask with mechanical stirring, primary condenser operated at 85.degree.-100.degree. C., and received flasks and traps) at constant temperature of 230.degree. C. and constant pressure of 10 mm Hg. Initial PLA charge was 360 gms of material having an average molecular weight of about 650. The catalyst charge was 0.045 wt % as Sn for each type of catalyst.
Table 1 shows the results of several experiments for different combinations.
From the table above it can be readily seen that PLA B is not as reactive toward lactide production as is PLA A. The effect is particularly pronounced with SnO catalyst.
Analysis of the initial lactic acids did not reveal any significant impurities in B relative to A. It is believed that some unidentified contaminant in PLA B blocks catalyst activity.
Further experiments were performed using the liquid contents of the lactide reactor (reactor bottoms or bottoms) from previous runs with SnO catalyst as the catalyst. The overall Sn content was maintained at 0.045 wt %.
The table above clearly shows that lactide can be produced from PLA B as fast as from PLA A if the SnO catalyst is pretreated by first producing a batch of lactide from PLA A. The table also shows that if the catalyst is pretreated by producing a batch of lactide from the blend A/B PLA then it is rendered ineffective and does not promote lactide production at high rates even from PLA A. Thus, applicants have surprisingly found that apparent deficiencies in the lactic acid source can be overcome by proper conditioning of the catalyst.
Comparison of the rates for PLA A with SnO (0.70 hr.sup.-1 table 1) and SnO/A (1.20 hr table 2) shows that the conditioning also increases the overall rate of the reaction for the most productive lactic acid, providing an additional benefit.
Applicants believe that heating the catalyst at about 200.degree.-230.degree. C. for about 10-30 minutes in the presence of either 1) PLA made from a purified lactic acid, 2) PLA prepared by the partial hydrolysis of purified lactide, 3) purified lactic acid, or 4) lactide, would provide similar benefits as the method described above.
EXAMPLE 14
Distillation of Crude Lactide
The overhead fraction from example 12 was collected in three receivers, containing 14%, 36%, and 28% of the crude lactide charge, respectively. The first fraction contained over 5 wt % lactic acid and was discarded. Fractions 2 and 3 were combined and redistilled, yielding a lactide fraction with total acidic impurities of 4.4 milliequivalents/mol of lactide. This fraction was polymerized using the polymerization technique of example 5 (temperature 180.degree. C., catalyst/monomer ratio 1:5000), yielding a polymer with number average molecular weight of 42,100 at 100% conversion and weight average molecular weight of 76,300. Actual conversion was 84.6% at 2 hours.
Applicants believe this final example, coupled with previous examples demonstrates the overall feasibility and advantages of the disclosed process.
It will be understood, however, that even though these numerous characteristics and advantages of the invention have been set forth in the foregoing description, together with details of the structure and function of the invention, the disclosure is illustrative only, and changes may be made in detail, especially in matters of shape, size and arrangement of the parts or in the sequence or the timing of the steps, within the broad principle of the present invention to the full extent indicated by the broad general meaning of the terms in which the appended claims are expressed.
Claims
What is claimed is:
1. A process for the continuous conversion of a crude lactic acid feed in an aqueous medium to substantially purified lactide, said process comprising the steps of: a) providing a source of lactic acid in an aqueous medium; b) concentrating the lactic acid in the aqueous medium by evaporating a substantial portion of the aqueous medium to form a concentrated lactic acid solution; c) polymerizing lactic acid in the concentrated lactic acid solution of step (b) by further evaporation of the aqueous medium to form polylactic acid molecules having an average molecular weight of between about 100 and about 5000; d) forming a crude lactide in the presence of catalyst means for catalyzing the depolymerization of the polylactic acid molecules to form lactide molecules, and e) purifying the crude lactide formed in step (d) to form a substantially purified lactide by distilling the crude lactide.
2. The process of claim 1, wherein the source of lactic acid of step (a) is an aqueous solution of lactic acid and water in concentrations ranging from about 1% lactic acid to about 99% lactic acid.
3. The process of claim 1, wherein racemization of the lactic acid is controlled in a range of 0% to 100% by altering residence times in process equipment, catalyst levels, temperatures and pressures.
4. The process for of claim 1, wherein the crude lactide formed in step (d) is a vapor, said vapor being subsequently partially condensed to remove volatile contaminants to form a condensed partially purified liquid lactide, the condensed partially purified liquid lactide then being purified as in step (e).
5. The process of claim 1, wherein the source of lactic acid for step (a) contains impurities which include but are not limited to color bodies, carbohydrates, proteins, amino acids, salts, metal ions, or other carboxylic acids.
6. The process of claim 1, wherein the yield of lactide is maximized by recycling streams containing recoverable lactic acid, lactide or oligomers and polymers thereof.
7. The process of claim 11, wherein the crude lactide formed in step (d) is generated in a reactor under conditions of elevated temperature and reduced pressure.
8. A process for the continuous conversion of a crude lactic acid feed in a hydroxylic medium to a substantially purified lactide, said process comprising the steps of: a) providing a source of lactic acid in a hydroxylic medium; b) concentrating the lactic acid in the hydroxylic medium by evaporating a substantial portion of the hydroxylic medium to form a concentrated lactic acid solution; c) polymerizing lactic acid in the concentrated lactic acid solution of step (b) by further evaporation of the hydroxylic medium to form polylactic acid molecules having an average molecular weight of between about 100 and about 5000; d) forming a crude lactide in the presence of catalyst means for catalyzing the depolymerization of the polylactic acid molecules to form lactide molecules; and e) purifying the crude lactide formed in step (d) to form a substantially purified lactide by distilling the crude lactide.
9. The process of claim 8, wherein the source of lactic acid of step (a) is a hydroxylic solution of lactic acid and a hydroxylic medium in concentrations ranging from about 1% lactic acid to about 99% lactic acid.
10. The process of claim 8, wherein racemization of the lactic acid is controlled in a range of 0% to 100% by altering residence times in process equipment, catalyst levels, temperatures and pressures.
11. The process of claim 8, wherein the crude lactide formed in step (d) is a vapor being subsequently partially condensed to remove volatile contaminants to form a condensed partially purified liquid lactide, the condensed partially purified liquid lactide then being purified as in step (e).
12. The process of claim 8, wherein the source of lactic acid for step (a) contains impurities which include but are not limited to color bodies, carbohydrates, proteins, amino acids, salts, metal ions, or other carboxylic acids.
13. The process of claim 8, wherein the yield of lactide is maximized by recycling streams containing recoverable lactic acid, lactide or oligomers and polymers thereof.
14. The process of claim 8, wherein the crude lactide formed in step (d) is generated in a reactor under conditions of elevated temperature and reduced pressure.
Patent Citations (88)
| Patent | Date | Inventor | Cited By |
|---|---|---|---|
| US1095205 | 1914-05-01 | Gruter et al. | |
| US1849107 | 1932-03-01 | Moss | |
| US1995970 | 1933-04-01 | Dorough | |
| US2396994 | 1946-03-01 | Filachione et al. | |
| US2703316 | 1955-03-01 | Schneider | |
| US2758987 | 1956-08-01 | Salzberg | |
| US2951828 | 1960-09-01 | Zeile et al. | |
| US3268487 | 1966-08-01 | Klootwijk | |
| US3322791 | 1967-05-01 | Selman | |
| US3531561 | 1970-09-01 | Trehu | |
| US3636956 | 1972-01-01 | Schneider | |
| US3772420 | 1973-11-01 | Glick et al. | |
| US3773919 | 1973-11-01 | Boswell et al. | |
| US3839297 | 1974-10-01 | Wassermann et al. | |
| US3887699 | 1975-06-01 | Yolles | |
| US3912692 | 1975-10-01 | Casey et al. | |
| US4045418 | 1977-08-01 | Sinclair | |
| US4249531 | 1981-02-01 | Heller et al. | |
| US4273920 | 1981-06-01 | Nevin | |
| US4279249 | 1981-07-01 | Vert et al. | |
| US4595713 | 1986-06-01 | St. John | |
| US4643734 | 1987-02-01 | Lin | |
| US4677191 | 1987-06-01 | Tanaka et al. | |
| US4683288 | 1987-07-01 | Tanaka et al. | |
| US4719246 | 1988-01-01 | Murdoch et al. | |
| US4727163 | 1988-02-01 | Bellis | |
| US4728721 | 1988-03-01 | Yamamoto et al. | |
| US4766182 | 1988-08-01 | Murdoch et al. | |
| US4789726 | 1988-12-01 | Hutchinson | |
| US4797468 | 1989-01-01 | De Vries | |
| US4800219 | 1989-01-01 | Murdoch et al. | |
| US4835293 | 1989-05-01 | Bhatia | |
| US4902515 | 1990-02-01 | Loomis et al. | |
| US4950258 | 1990-08-01 | Kawai et al. | |
| US4960866 | 1990-10-01 | Bendix et al. | |
| US4981696 | 1991-01-01 | Loomis et al. | |
| US4983745 | 1991-01-01 | Muller et al. | |
| US4990222 | 1991-02-01 | Aigner et al. | |
| US5011946 | 1991-04-01 | Hess et al. | |
| US5023349 | 1991-06-01 | Bhatia | |
| US5023350 | 1991-06-01 | Bhatia | |
| US5028667 | 1991-07-01 | McLain et al. | |
| US5041529 | 1991-08-01 | Shinoda et al. | |
| US5043458 | 1991-08-01 | Bhatia | |
| US5053485 | 1991-10-01 | Nieuwenhuis et al. | |
| US5053522 | 1991-10-01 | Muller | |
| US5076983 | 1991-12-01 | Loomis et al. | |
| US5097005 | 1992-03-01 | Tietz | |
| US5108399 | 1992-04-01 | Eitenmuller et al. | |
| US5132397 | 1992-07-01 | DeGuia | |
| US5134171 | 1992-07-01 | Hammel et al. | |
| US5136017 | 1992-08-01 | Kharas et al. | |
| US5142023 | 1992-08-01 | Gruber et al. | |
| US5149833 | 1992-09-01 | Hess et al. | |
| US5180765 | 1992-01-01 | Sinclair | |
| CAX808731 | 1969-03-01 | ||
| CAX863673 | 1971-02-01 | ||
| CAX923245 | 1973-03-01 | ||
| EPX107591 | 1984-02-01 | ||
| EPX299730 | 1989-01-01 | ||
| EPX314245 | 1989-05-01 | ||
| EPX52510 | 1992-03-01 | ||
| EPX481732 | 1992-04-01 | ||
| EPX507554 | 1992-10-01 | ||
| EPX510998 | 1992-10-01 | ||
| EPX515203 | 1992-11-01 | ||
| DE2267826 | 1913-12-01 | ||
| DEX1083275 | 1960-12-01 | ||
| DEX1543958 | 1970-02-01 | ||
| DEX3632103 | 1988-03-01 | ||
| JPX4-283227 | 1992-07-01 | ||
| WOXWO90/01521 | 1990-02-01 | ||
| WOXWO91/02015 | 1991-02-01 | ||
| WOXWO91/06601 | 1991-05-01 | ||
| WOXWO92/00292 | 1992-01-01 | ||
| WOXWO92/00974 | 1992-01-01 | ||
| WOXWO92/04410 | 1992-03-01 | ||
| WOXWO92/04412 | 1992-03-01 | ||
| WOXWO92/04413 | 1992-03-01 | ||
| WOXWO92/05167 | 1992-04-01 | ||
| WOXWO92/05168 | 1992-04-01 | ||
| WOXWO92/05311 | 1992-04-01 | ||
| WOXWO92/15340 | 1992-09-01 | ||
| WOXWO91/17155 | 1992-11-01 | ||
| GBX1040168 | 1966-08-01 | ||
| GBX1108720 | 1968-04-01 | ||
| GBX1351409 | 1974-05-01 | ||
| GBX2145422 | 1985-03-01 |
Non-Patent Literature (58)
- T. M. Jackanicz, "Polylactic Acid as a Biodegradable Carrier for Contraceptive Steroids", Contraception, vol. 8, No. 3, 227-234 (1973).
- A. D. Schwope et al., "Lactic/Glycolic Acid Polymers as Narcotic Antagonist Delivery Systems", Life Sciences, vol. 17, 1877-1886 (1975).
- L. C. Anderson, "An Injectable Sustained Release Fertility Control System", Contraception, vol. 13, No. 3, 375-384 (1976).
- D. L. Wise et al., "Sustained Release of an Antimalarial Drug Using a Copolymer of Glycolic/Lactic Acid", Life Sciences, vol. 19, 867-874 (1976).
- R. A. Miller et al., "Degradation Rates of Resorbable Implants (Polylactates and Polyglycolates): Rate Modification with Changes in Pla/Pga Copolymer Rations", J. Biomed. Mater. Res., vol. 11, 711-719 (1977).
- D. K. Gilding et al., "Biodegradable Polymers for Use in Surgery--Polyglycolic/Polylactic Acid Homo and Copolymers: 1." Polymer, vol. 2, 1459-1464 (1979).
- D. K. Gilding, "Degradation of Polymers: Mechanisms and Implications for Biomedical Applications", Biocompatibility of Clinical Implant Materials, D. F. Williams, ed., vol. 1, 43-65 (1981).
- A. M. Reed and D. K. Gilding, "Biodegradable Polymers for Use in Surgery Polyglycolic/Polylactic Acid Homo and Copolymers: 2. In Vitro Degradation", Polymer, vol. 22, No. 4, 494-498 (1981).
- D. K. Gilding, "Biodegradable Polymers", Biocompatibility of Clinical Implant Materials, D. F. Williams, ed., vol. 2, 209-232 (1981).
- J. D. Strobel, "Biodegradable Polymers", paper presented at Medical Textiles and Biomedical Polymers and Materials Conference held at Clemson, S.C., U.S.A., Dec. 5-6, 1989, Stolle Research and Development Corp., PD 712-01, pp. 1-32 and Attachments A1-A21.
- "Biocompatible Composite Would Be Completely Absorbed in the Body", Advanced Materials, vol. 12, No. 15, Aug. 1990, p. 6.
- "Polylactides Exhibit Degradability", Tappi Journal, Sep. 1991, p. 42.
- P. V. Bonsignore et al., 1992, "Poly(lactic acid) Degradable Plastics, Coatings, and Binders", TAPPI Proceedings (Nonwovens Conference); pp. 129-140.
- W. Carothers, G. Dorough, and F. Van Natta ("Studies of Polymerization and Ring Formation. X. The Reversible Polymerization of Six-Membered Cyclic Esters", 1932, American Chemical Society Journal, v. 54, pp. 761-772).
- E. Filachione, E. Costello, T. Dietz and C. Fisher ("Lactic Acid Derivatives as Plasticizers Esters of Polymeric Lactic Acid", 1951, Bur. Agric. Ind. Chem., v. 11, pp. 1-11).
- D. Deane and E. Hammond ("Coagulation of Milk for Cheese-Making by Ester Hydrolysis", 1960, Journal of Dairy Science, v. 43, pp. 1421-1429).
- Kulkarni et al. ("Biodegradable Poly(lactic acid) Polymers", 1971, J. Biomed. Mater. Res., v. 5, pp. 169-181).
- A. Schindler, R. Jeffcoat, G. Kimmel, C. Pitt, M. Wall and R. Zweidinnger ("Biodegradable Polymers for Sustained Drug Delivery", 1977, Contemporary Topics in Polymer Science, v. 2, pp. 251-287).
- I. Luderwald ("Thermal Degradation of Polyesters in the Mass Spectrometer", 1979, Dev. Polymer Degradation, v. 2, pp. 77-98).
- M. Vert and F. Chabot ("Stereoregular Bioresorbable Polyesters for Orthopaedic Surgery", 1981, Makromol. Chem., Supp. 5, pp. 30-41).
- M. Gupta and V. Deshmukh ("Thermal Oxidative Degradation of Poly-lactic Acid; Part I: Activation Energy of Thermal Degradation in Air", 1982, Colloid & Polymer Science, v. 260, pp. 308-311).
- M. Gupta and V. Deshmukh ("Thermal Oxidative Degradation of Poly-lactic Acid; Part II: Molecular Weight and Electronic Spectra During Isothermal Heating", 1982, Colloid & Polymer Science, v. 260, pp. 514-517).
- G. Van Hummel and S. Harkema ("Structure of 3,6-Dimethyl-1,4-Dioxane-2,5-Dione [D-,D-{L-,L-}Lactide]", 1982, Acta. Crystallogr., v. B38, pp. 1679-1681).
- F. Chabot, M. Vert, S. Chapelle and P. Granger ("Configurational Structures of Lactic Acid Stereocopolymers as Determined by .sup.13 C(.sup.1 H) N.M.R.", 1983, Polymer, v. 24, pp. 53-59).
- F. Kohn, J. Van Don Berg, G. Van De Ridder and J. Feijen ("The Ring-Opening Polymerization of D,L-Lactide in the Melt Initiated with Tetraphenyltin", 1984, Journal of Applied Polymer Science, v. 29, pp. 4265-4277).
- H. Kricheldorf and A. Serra ("Polylactones 6. Influence of Various Metal Salts on the Optical Purity of Poly(L-lactide)", 1985, Polymer Bulletin, v. 14, pp. 497-502).
- A. Chawla and T. Chang ("In-Vivo Degradation of Poly(lactic acid) of Different Molecular Weights", 1985, Biomat., Med. Dev., Art. Org., v. 13, pp. 153-162).
- I. McNeill and H. Leiper ("Degradation Studies of Some Polyesters and Polycarbonates--1. Polylactide: General Features of the Degradation Under Programmed Heating Conditions", 1985, Polymer Degradation and Stability, v. 11, pp. 267-285).
- I. McNeill and H. Leiper ("Degradation Studies of Some Polyesters and Polycarbonates--2. Polylactide: Degradation Under Isothermal Conditions, Thermal Degradation Mechanism and Photolysis of the Polymer", 1985, Polymer Degradation and Stability, v. 11, pp. 309-326).
- Makino et al. ("Preparation and in Vitro Degradation Properties of Polylactide Microcapsules", 1985, Chem. Pharm. Bull., v. 33, pp. 1195-1201).
- D. Garozzo, M. Giuffrida and G. Montaudo ("Primary Thermal Decomposition Processes in Aliphatic Polyesters Investigated by Chemical Ionization Mass Spectrometry", 1986, Macromolecules, v. 19, pp. 1643-1649).
- "Irganox.RTM. 1076 Antioxidant and Thermal Stabilizer", (published on an unknown date in 1986 by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).
- J. Leenslag and A. Pennings ("Synthesis of high-molecular-weight poly(L-lactide) initiated with tin 2-ethylhexanoate", 1987, Makromol. Chem., v. 188, pp. 1809-1814).
- Nakamura et al. ("Surgical Application of Biodegradable Films Prepared from Lactide-.epsilon.-Caprolactone Copolymers", 1987, Bio. Materials and Clinical Applications, v. 7, pp. 759-764).
- H. Kricheldorf, M. Berl and N. Scharnagl ("Polymerization Mechanism of Metal Alkoxide Initiated Polymerizations of Lactide and Various Lactones", 1988, Makromol., v. 21, pp. 286-293).
- K. Jamshidi, S. Hyon and Y. Ikada ("Thermal Characterization of Polylactides", 1988, Polymer, v. 29, pp. 2229-2234).
- M. Vert ("Bioresorbable Polymers for Temporary Therapeutic Applications", 1989, Die Angwandte Makromolekulare Chemie, v. 166-167, pp. 155-168).
- "Hydrolytic Stability/Corrosivity of Phosphite Costabilizers", (Technical Bulletin 89-04, published on an unknown date in 1989, by Stars Laboratory, Additives Division, Ciba-Geigy Corporation, Ardsley, N.Y. 10502).
- "GE Specialty Chemicals Product Guide CA-4001E", (published on an unknown date in 1989, by General Electric Company, 5th and Avery Street, Parkersburg, W.V. 26102).
- "Tinuvin.RTM. 123 Hindered Aminoether Light Stabilizer for Coatings", (published on an unknown date in 1989, by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).
- "Irganox.RTM. B-Blends Antioxidants and Process Stabilizers for Polymers", (published on an unknown date in Mar. 1990, by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).
- "Naugard.RTM. 445, Specialty Chemicals", (a product brochure published on or before May 1, 1990, by Uniroyal Chemical Company, Inc., Middlebury, Conn. 06749).
- "Ethanox.RTM. 398 Antioxidant, The First Fluorophosphonite Antioxidant", (published on or before an unknown date in Oct. 1990, by Ethyl Corporation, 451 Florida Blvd., Baton Rouge, La. 70801).
- "The Resomer.RTM. Resorbable Polyesters" (published on or before an unknown date in Feb. 1991 by Boehringer Ingelheim KG, D-6507 Ingelheim, W. Germany).
- P. Klemchuk, ("Introduction to Polymer Degradation", lecture notes distributed at a seminar entitled: Principles of Polymer Degradation and Stabilization in Orlando, Fla., Oct. 28-30, 1991, sponsored by The Institute of Materials Science, State University of New York at New Paltz).
- R. Thomas, ("Degradation and Stabilization of Engineering Polymers", lecture notes distributed at a seminar entitled: Principles of Polymer Degradation and Stabilization in Orlando, Fla., Oct. 28-30, 1991, sponsored by The Institute of Materials Science, State University of New York at New Paltz).
- W. Enlow, ("Process Stabilization with Phosphite Antioxidants", lecture notes distributed at a seminar entitled: Principles of Polymer Degradation and Stabilization in Orlando, Fla., Oct. 28-30, 1991, sponsored by The Institute of Materials Science, State University of New York at New Paltz).
- "Naugard.RTM. XL-1 Specialty Chemicals", (product brochure published on an unknown date in Feb. 1992, by Uniroyal Chemical Co., Inc., Middlebury, Conn. 06749).
- Sir John Meurig Thomas, ("Solid Acid Catalysts", Apr. 1992, Scientific American, pp. 112-118).
- "Argus Product Data, Argus.RTM. Dimyristyl Thiodipropionate", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Argus Thiochemical Product Data, Argus.RTM. Thiodipropionate", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Argus Product Data, Argus.RTM. Distearyl Thiodipropionate", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Argus Product Data, Mark.RTM. 2140 Pentaerythrityl Octylthiopropionate", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Argus Thiochemical Product Data, Argus.RTM. Dilauryl Thiodipropionate", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Argus Product Data, Seenox.RTM. 412S Pentaerythritol Tetrakas (B-Laurylthiopropionate)", (published on or before an unknown date in Aug. 1992, by Argus Division, Witco Corporation, 633 Court Street, Brooklyn, N.Y. 11231-2193).
- "Irgonox.RTM. 1010", (a product brochure published on or before an unknown date in Aug. 1992, by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).
- "Irganox.RTM. MD 1024, Metal Deactivator/Antioxidant", (published on an unknown date prior to Aug. 1992, by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).
- "Tinuvin.RTM. 622LD Low Dust, Hindered Amine Light Stabilizer for Polymers FDA-Cleared for Polyolefins", (published on an unknown date before Aug. 1992, by Ciba-Geigy Corporation, Three Skyline Drive, Hawthorne, N.Y. 10532).