MXPA00003607A - Lactic acid processing;methods;arrangements;and, products - Google Patents

Abstract

The techniques for processing lactic acid/lactate salt mixtures are provided. Preferred mixtures for processing are obtained from fermentation broths, preferably from fermentation processes conducted at a pH of 4.8 or lower. The techniques generally concern the provision of separated lactic acid and lactate streams, from the mixtures. Preferred techniques of separation and processing of each of the streams are provided.

Description

PROCESSING OF LACTIC ACID; METHODS; ARRANGEMENTS; AND PRODUCTS FIELD OF THE NINETION The present invention relates to processing of lactic acid. Particularly concerns: methods for separating lactic acid streams and lactate salt streams from mixtures such as fermentation broths; isolate and process lactic acid; and isolating the lactate salt in preferred forms.

BACKGROUND OF THE INVENTION The potential of lactic acid as a convenience chemical is known, for example, for use in the production of various industrial polymers. This has been described, for example, in U.S. Patents: 5, 142, 023; 5,247,058; 5,258,488; 5,357,035; 5,338,822; 5,446, 1 23; 5, 539, 081; 5, 525, 706; 5,475, 080; 5, 359.026; 5,484, 881; 5, 585, 1 91; 5,536,807; 5,247, 059; 5,274, 073; 5.51, 0.526; and 5, 594, 095. (Full descriptions of these seventeen patents, which belong to the attorney of the present application, Cargill, Inc. of Minneapolis, Minnesota, are hereby incorporated by reference). There has been a general interest in developing improved techniques for the generation and isolation of lactic acid. In addition, due to its potential commercial value, there is great interest in the isolation of other valuable lactate-related products, such as lactide, lactate esters and amides, and oligomers; see, for example, the same 1 7 patents.

In general, large quantities of lactic acid can be easily generated by conducting large-scale industrial, bacterially driven fermentation processes, particularly using carbohydrates, such as dextrose as the raw material, together with mineral-based nutrients and suitable amino acids . Typically, such productions occur at broth temperatures of at least 45 ° C, usually around 48 ° C. Topics of interest with respect to the generation of lactic acid include, inter alia, appropriate pH control within the fermentation system to ensure an appropriate environment for bacterial action; separation and isolation of either or both of lactic acid and lactate salts from the fermentation processes; and downstream isolation and production involving the isolated lactic acid or lactic acid-derived product.

BRIEF DESCRIPTION OF THE INVENTION In accordance with the present disclosure, techniques are provided for processing mixtures of lactic acid and dissolved lactate salts. Preferred techniques are provided for processing fermentation broths, preferably fermentation broths produced with or adjusted to have a pH of less than about 4.8, normally and preferably less than about 4.5, more preferably less than 4.3 and most preferably within the range of about 3.0 to 4.2 inclusive.

The techniques include processing the mixtures in: (a) a lactic acid stream, component or phase; and (b) a lactate salt stream, component or phase. Preferred techniques are provided, so that the lactic acid stream, component or phase can be readily captured to produce desirable lactate products, such as lactate oligomers, lactate lactate esters, lactate amides and / or polylactate. The preferred process also provides the lactate salt in a form suitable for further use, such as, recycling to a fermentation broth; or as a fertilizer or food.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a process flow diagram of a process according to the present description; FIG. 2 is a process flow diagram alternate to one shown in FIG. 1; FIG. 3 is a process flow diagram alternate to those shown in FIGS. 1 and 2; FIG. 4 is a process flow diagram alternate to those shown in FIGS. 1-3; FIG. 5 is a process flow diagram alternate to those shown in FIGS. 1 -4; FIG. 6 is a process flow diagram alternate to those shown in FIGS. fifteen; FIG. 7 is a process flow diagram alternate to those shown in FIGS. 1-6; FIG. 8 is a diag ram of alternating process flow to those shown in the FI Gs. 1-7; FIG. 9 is a process flow diagram alternate to those shown in FIGS. 1-8; FIG. 1 0 is an alternate process flow chart to those shown in the FI Gs. 1-9; and FIG. 1 1 is a graph showing the percentage of lactic acid in free acid form, in a lactic acid mixture, as a function of pH.

DETAILED DISCLAIMER I. Selected issues of interest with respect to lactic acid processing, isolation and use A. Chirality Lactic acid has a chiral center and is found in both D and L forms. The chiral purity of lactic acid is important with respect to to meet the needs of industrial applications, see for example, US Patents: 5, 142, 023; 5, 338,822; 5,484, 881; and 5, 536, 807. There are bacteria, for example of the genus Lactobacillus, which can be either D-lactic acid or L-lactic acid. However, it is normal for any bacterial strain to produce a large majority of a single enantiomer. In fact, fermentation broths with high chiral purity (90% or greater) of lactic acid can be obtained easily. This chirality is obtained from the metabolism of dextrose or other carbohydrates by microorganism cells during fermentation. For example, Lactobacillus bulgaricas and Lactobacillus coryniformis normally produce almost exclusively enantiomer of D-lactic acid. It has been found that Lactobacillus casei produces, in its majority, L-lactic acid. For applications of polylactic acid, the chiral purity of the lactic acid has a strong influence on the properties of the polymer. The chiral purity of the polymer controls the ability of the polymer to crystallize. See for example, U.S. Patents 5,484, 881; 5, 585, 1 91; and 5,536, 807; and the commonly assigned US patent application 08/850, 319, filed May 2, 1997. (Each of these four references is incorporated herein by reference.) In some cases, polymers are desired. controlled amounts of crystallinity in order to obtain polymer properties which are advantageous in an industrial application, for example, to raise the heat distortion temperature of the polymer. Other advantages of controlled polymer crystallinity relate to the storage, transfer and processing of polylactic acid resins in fibers, non-woven fabrics, films and other end products. Lactic acid currently used in food applications has requirements of chiral purity greater than 95% of quarity, generally with a preference for the "L" form. The chiral purity of lactic acid is also important for terminal products, such as pharmaceuticals and other medical devices where lactic acid is a starting material. In the present, the term "95% chiral purity" means that 95% of the lactic acid / lactate content is one of the two possible enantiomers. (In this manner, the composition could alternatively be characterized as 1 0% racemic or 90% optically pure.) Herein, the terms "polylactic acid" or "polylactate" are intended to refer to any polymer comprising at least 50% in weight of polymer units of lactic acid residue or lactide residue. In this way, the two terms include polylactides within their scope. The terms "polylactic acid" and "polylactate" are not intended to specifically identify the polymerized monomer, for example, if the polymerized material was lactide (lactic acid dimer) or lactic acid by itself.

B. pH control during fermentation Most microorganisms have a pH range in which they are able to perform metabolism very efficiently. Consequently, the pH of the fermentation is a processing variable that strongly affects the overall productivity of the microorganism cells in the fermentation. The microorganisms of Lactobacillus produce lactic acid. Without a neutralizing agent, the pH of a normal, conventional fermentation broth quickly falls to a value at which most microorganisms die or cease useful production. Accordingly, the addition of a neutralizing agent has been normally required to satisfy the economical need for a fermentation with a high overall productivity. The pH value for many of the lactic acid fermentations with good productivity (ie, >0.5 g of lactic material (lactic acid and lactate salt) produced / liter / h) are in the range of 5.0 to 7.0; see, for example, US Patent 5, 51 0, 526. Much work has been done to look for organisms that retain high lactic acid productivities while operating in broths in pH ranges from 3.0 to 4.8. This is discussed below. Lactic acid (HLa or LaH) dissociates into a proton, H +, and a lactate anion, La "(sometimes referred to herein as a dissolved lactate salt when another source of cation is present, usually from the buffer salt) The amount of dissociation is related to the pH of the solution and the pKa of lactic acid.The pKa of the lactic acid at 25 ° C is 3.86 (at 50 ° C it is approximately 3.89) Equation 1 below describes how they are related the pH, pKa, and the degree of dissociation of lactic acid. pH = pKe + log í The ~ 1 Equation 1 [HLa] Equation 1 shows that the acid is dissociated when the pH equals the pKa of the acid. At higher pH values, most of the lactic acid is in the form of lactate anion. Figure 11 is a graph showing the percentage of lactic acid in the undissociated form (free acid) as the pH varies from 1 to 7. The graph shows that the percentage of undissociated lactic acid present in solution at the values of pH of 5 to 7 is relatively low.

If the fermentation broth has a pH value between 3.0 and 4.5, there will be a significant amount of lactic acid in the undissociated form, see Fig. eleven . Actually at a pH of 3.0, the molar ratio of free (non-dissociated) lactic acid to lactate ion at 25 ° C is approximately 7.0; and, at a pH of about 4.5, the ratio at 25 ° C, is about 0.23. A separation process that specifically separates the undissociated lactic acid or a lactate derivative and the lactate salt would be beneficial, because it would provide a stream of: (1) lactic acid products to be further purified (and / or converted to lactide or polymer); and, (2) a lactate salt, useful as a buffering agent for pH control in the fermenter. Equation 1 shows how the ratio of lactate anion, La-, to free lactic acid HLa is related to the pH of the solution. As free lactic acid is produced by the microorganism during fermentation, the addition of lactate salt can maintain the pH of the solution constant. If the added lactate salt, for pH control, is a recycled material, then it follows that the efficiency of conversion and recovery of raw material into lactic acid, in the overall fermentation processes, is improved. That is, a higher percentage of added raw material is converted into, and is retained and isolated as, lactic acid, as opposed to the lactate salt, due to the maintenance of the equilibrium suggested by Equation 1 with recycling of lactate salt. The preferred separation scheme to be employed will depend on the shape of the lactate material in solution. The isolation of lactate material from an aqueous solution may require substantial energy, especially if the aqueous solution is at a pH greater than 4.5 and the lactate material is mainly present as a lactate salt, as opposed to an aqueous solution to an aqueous solution. pH below 4.5, where a significant amount of lactate material is in the free acid form. When the pH of the fermentation broth is about 5 to 7, a normal step in the conventional separation processes has been to acidify the solution strongly by the addition of sulfuric acid. This process forms free lactic acid, but also forms a by-product salt (usually calcium sulfate). The formation of by-product salt represents the use of chemical energy to transform the lactate salt into lactic acid, and can create a waste disposal issue when lactic acid is manufactured on a large scale. In many alternative separation processes, in which direct acidification does not occur, the energy would be used to separate the lactate salt back into lactic acid and a base. Water separation electrodialysis is a good example of this type of separation process, in which electric energy is used to form an acid and a base from salt and water. It is noted that as the desired product (HLa) of the fermentation increases in concentration, the fermentation is often not only inhibited due to the pH, but also due to the concentration of HLa.

C. Processing downstream; DEFINITION OF MATERIALS Once the lactic acid is separated from the lactate salt, the lactic acid can be used to form high molecular weight polylactic acid (usually M.W average from 1,000 to about 300,000). Normally and preferably, processes such as those described in: 5,338,822 are used; 5,446, 1 23; 5,539,081; 5, 525,706; 5,475, 080; 5, 359.026; 5,484, 881; 5, 585, 1 91; 5, 536,807; 5,247, 059; 5,274,073; and 5, 594, 095. Such techniques generally include: (a) providing a mixture of lactide (optionally with other reagents, such as other monomers and / or epoxidized oils) with an appropriate catalyst and a presence of sufficiently low water; (b) polymerizing the lactide mixture, generally by application of heat; and, (c) devolatilization of the polylactide to remove unreacted monomer and residual water. Stabilizing agents, such as free radical scavengers, and catalyst deactivators, can be used to provide the final composition with preferred melt stability. Chemical intermediates formed from lactic acid, such as lactide, alkyl lactate esters, alkyl lactate amides and oligomers with an average molecular weight of less than about 5,000, are normally used to form polylactide polymers, some times when it is first reacted to form lactide when an intermediary other than lactide is involved by itself. In this way, it is of great interest the generation and / or isolation of these "training blocks" identified for polymers, from the LaH of a fermentation broth. The term "lactic acid products", as used herein, means that they include lactic acid, lactate salts, alkyl lactate esters, alkyl lactate amides, lactide, lactoyl lactate, lactic acid trimers and tetramers, and oligomers of lactic acid, usually with an average molecular weight of less than about 5,000. Of course, lactic acid is the smallest repeating unit (present as the acid residue of the condensation polymerization) in polylactic acid. It is the most basic starting material for polylactic acid and the other chemical intermediates, such as lactide and lactic acid oligomers are usually made of lactic acid (or lactate salts). Lactide is a cyclic ester comprising two molecules of lactic acid. That is, it is a lactic acid dimer. Due to the chiral nature of lactic acid, lactide can have one of three types of optical activity depending on whether it comprises two residues of D-lactic acid, two residues of L-lactic acid or one residue of L-lactic acid and one residue of D-lactic acid. These three dimers are designated D-lactide, L-lactide and meso-lactide, respectively. Lactide is completely dehydrated lactic acid, and is commonly used in the manufacture of polylactic acid (or polylactide) using an open ring reaction to grow the polymer at high molecular weights. Lactide can also be a key starting material in the production of other, indom- erately relevant chemicals. The alkyl lactate esters and alkyl lactate amides are compounds which can be used as raw material for oligomers of lactic acid, lactide or polylactide acid. To make lactic acid oligomers with an ester at the terminal end of the carboxylic acid, the alkyl lactate esters can be transesterified with the corresponding alcohol being obtained together with the oligomer. The simultaneous or sequential removal of the alcohol drives the reaction for oligomer formation. Lactide can be made from oligomers of esterified lactic acid. The alkyl lactate amides would have similar chemistry to the esters, but with an amine being obtained together with the oligomer. The lactide can be made from an oligomer of lactic acid with an amide group at the terminal carboxylic acid end. The formation of esters or amides from lactic acid can also assist in the separation of the lactic acid derivative from impurities. After a stream of alkyl lactate ester or amide of purified alkyl lacatate is obtained, the ester or amide can be hydrolyzed to obtain the lactic acid and the corresponding alcohol or amine. The lactic acid can be separated from this mixture, and the alcohol or amine is recycled back to the step of ester or amide formation. Of course, certain esters and lactic amides could be further purified if necessary. Useful alkyl lactate esters include: methyl lactate, ethyl lactate, butyl lactate, octyl lactate, docecyl lactate, 2-ethylhexyl lactate, and 1,4-butane diol lactate. In fact, alkyl lactates with 1-20 carbon atoms in the alcohol residue, whether saturated or unsaturated, or unsaturated, are potentially useful. With regard to lactate esters and their use, see, for example, US Pat. No. 5,247,059. Lactamide (the ammonia amide of lactic acid) is an industrially important lactic amide. It is used in hair care products.

Lactic acid oligomers having an average molecular weight of less than about 5,000 are useful for making lactide. Usable techniques are described in US Pat. No. 5,152,023, incorporated by reference. Certain preferred modifications described therein directly concern the formation of lactide even in the presence of residual extractor, such as residual trialkylamine. A catalyst can be used to increase the rate of lactide formation from oligomers of polylactic acid. Many suitable catalysts are known, such as metal oxides, metal powders and organic metal compounds, see, for example, U.S. Patents: 5, 142, 023; 5, 338, 822; and 5, 594, 095. To handle lactide formation, lactide is normally removed simultaneously or sequentially from the lactic acid oligomer stream. One method for this removal is the addition of heat to vaporize a stream of crude lactide from the oligomers. In addition to being used as precursors for lactides, the lactic acid oligomers are useful as antimicrobial agents and as controlled release acidulants for food and agricultural use. Of course, the oligomer can be terminated or functionalized, in some cases, such as the amide or ester.

II. Fermentation at lower pH The generation of lactic acid solutions, via bacteriological systems, having pHs in the order of 5.0 or less, preferably 4.8 or less and normally 3.5 to 4.5, leads to a higher percentage of production of the lactate material, in the form of lactic acid. This is described, for example, in the co-filed, commonly assigned US patent application (for Cargill, Inc. of Minnetonka, Minnesota), entitled LOW pH LACTIC ACI D FERM ENTATION (Fermentation of lactic acid at low pH), corresponding to the PCT application WO 99/1 9503 (the application of Carlson et al.). The US application of Carlson et al was filed on the same date of the present application (October 14, 1997) and is incorporated herein by reference. Again, the generation of relatively large amounts of product from the fermentation process in the form of lactic acid, rather than lactate salt, is advantageous, since it reduces the need for, in certain consecutive process steps of acidification and / or "salt separation". That is, if a greater amount of the material is generated as free lactic acid, a processing step for generating the lactic acid from lactate, and the expenses and consequences associated with it, are reduced or avoided. Even if some acidification is conducted, substantially less acid addition would be involved than would be the case with a high pH system. In general, it has been determined that with processes conducted to make fermentation broths at pHs in the order of about 4.8 or less (preferably 4.5 or less, most preferably 4.3 or less, typically 3.5 to 4.2), a global efficient process can be developed, wherein the lactic acid generated is used in the polymer production and the recovered lactate salt is recycled in the fermentation system as a buffering agent, or set differently for pH control. The process of Carlson et al. it allows the efficient production of lactate and, in particular, the efficient production of high concentrations of free lactic acid via incubation of an acid-tolerant homoloctic bacterium in a suitable nutrient medium. By "homolactic" it is meant that the bacterial strain produces substantially only lactic acid as the fermentation product. Acid-tolerant homoloctic bacteria are usually isolated from maize infusion water from a commercial corn milling facility. While different bacteria of this type can produce either racemic lactate or lactate predominantly either in isomeric form D or L, the Carlson et al process describes preferred fermentation using a homolactic bacterium that predominantly produces L-lactate, and most preferably in optically pure form. The Carlson et al process allows the efficient production of high concentrations of free acid form of an optical isomer of lactic acid. This efficiency can be expressed in a variety of ways. The concentration of lactic acid in the fermentation broth serves as a measure of the overall productivity of the process. The present process normally generates a solution that includes at least about 25 g / l, preferably at least about 30 g / l, and more preferably at least about 40 g / l of free lactic acid. In a very normal and preferable manner, the lactate produced by the fermentation process is predominantly of a chiral form, either D-lactate or L-lactate. For preferred downstream processing, an optical purity of the lactic acid from the fermentation process of at least 50%, more preferably at least 75% and most preferably at least 90% to optically pure is produced in the fermentation and is used. For example, one modality of the process described in Carlson et al. includes incubating an acid-tolerant, homolactic bacterium in a nutrient medium to produce L-lactate having an optical purity of at least about 50% (ie, has a chiral purity of at least about 75%). The Carlson et al process can even be applied to produce L-lactate in optically pure form (ie, in which essentially only the L-form of lactate is produced). As noted above, the amount of free lactic acid present in a solution is a function of both the pH of the solution and the overall concentration of lactate material (i.e., lactic acid plus dissolved lactate sai) in the mixture. In this way, specifying these two parameters for a given solution, (for example, a fermentation broth), effectively specifies the concentration of free lactic acid. The Carlson et al process normally generates a solution, containing at least about 50 g / 1, preferably at least about 80 g / l and more preferably at least about 1 00 g / l of lactate at a relatively low pH. The lower the pH of the solution, the higher the percentage of lactate that is present in its free acid form. Again, if the pH of the medium (solution or mixture) is equal to the pKa of lactic acid (which is about 3.8 to 25 ° C), 50% of the lactate material is present in the free acid form. The pH of the nutrient medium during the homolactic bacterial incubation step can be expressed in several different ways, for example, in terms of the average incubation pH or the final incubation pH. The fermentation process of Carlson et al is usually capable of producing high levels of lactate material at an average incubation pH of no more than about 4.3, preferably no more than about 4.2, and more preferably no more than about 4.0. Alternatively, the pH of the broth during incubation can be expressed in terms of the final incubation pH. The Carlson et al process typically allows the production of high lactate concentrations at a final incubation pH (or mix pH) of no more than about 4.2, preferably no more than about 4.0, and more preferably no more than about 3.9. Particularly effective modes of the present fermentation process are capable of producing at least about 80 g / L of lactate at an average incubation pH of no more than about 4.0 and / or a final incubation pH of no more than about 3.9. In the present, the terms "nutrient medium" and "fermentation broth" are used interchangeably. Both are mixtures of free lactic acid and lactate anion (salt). These terms may be used to refer to: (i) means in the manner originally provided, for example, for acid-tolerant bacteria and nutrient sources including carbohydrates; (ii) media produced after some or all of the originally provided nutrients have been consumed and fermentation products, including lactate, have been excreted in the media by the bacteria; and, (iii) clarified media after the removal of a fermantador and filtration. The process provided in Carlson et al for producing lactic acid includes incubating acid-tolerant bacteria, such as acid-tolerant homoloctic bacteria, in a nutrient medium at a pH that provides a substantial portion of the lactate material in the free acid form. In the present, when the term "acid tolerant" is used with reference to bacteria, it is intended to refer to bacteria that are capable of producing lactate material at a pH sufficient to provide a substantial portion of the lactate material in the form of free acid. Acid-tolerant bacteria described in Carlson et al. , they are usually capable of producing at least about 25 g / L of free lactic acid at an incubation temperature of at least about 40 ° C. such bacteria can generally also produce at least about 50 g / L of lactate material in nutrient medium at an average incubation pH of no more than about 4.2 and at a temperature above 40 ° C. In another embodiment of the invention, the homolactic bacterium is capable of generating a solution containing at least about 40 g / l, preferably at least about 75 g / l of lactate and preferably about 90 g / l of lactate at an average incubation pH of no more than about 4.3. Particularly effective strains are capable of producing these levels of L-lactate (or D-lactate) at an average incubation pH of no more than about 4.0 and / or at a final incubation pH of no more than about 3.9. If the fermentation is carried out to a point where the pH and / or concentration of lactic acid inhibits additional lactate production, the "average incubation pH" is determined based on an average of the pH values measured in ten (10). or more equal time intervals over the period necessary to produce 90% of the limiting lactate concentration. The present fermentation process can be run in a continuous manner. Under such fermentation conditions, steady-state conditions (in terms of pH, lactate concentration and concentrations of nutrients) are generally achieved and maintained after an initial start-up phase has been completed. When the fermentation is conducted in this manner, the average incubation pH is the average pH of the broth after the initial start phase has been completed. If the fermentation is not done to a point where the limiting lactate concentration is reached, the "average incubation pH" is determined based on an average of the pH values measured in ten (10) or more equal time intervals about the course of fermentation. As used herein, the "limiting lactate concentration" is the concentration of lactate (concentration of undissociated and dissociated lactic acid) under a given set of incubation conditions (nutrient medium, temperature, degree of aeration) to which the pH and / or concentration of lactic acid generated by the fermentation inhibits the production of additional lactate As used herein, the term "limiting incubation pH" means the pH of the fermentation broth for a given set of conditions of incubation in which the pH and / or concentration of lactic acid inhibits the production of additional lactate.It is considered that inhibition of lactate production occurs when the amount of lactate produced in a batch fermentation does not increase by more than about 3 hours. % on additional incubation for a period of up to approximately twelve (12) hours under the same conditions.This definition presumes that a sufficient nutrients are available for the production of lactate in the fermentation broth and is applicable for both continuous and batch operations. In the process of Carlson et al., the pH of the fermentation broth after incubation of the acid-tolerant bacteria to produce lactate, is usually not greater than about 4.2 ("final incubation pH"). As referred to herein, the "final incubation pH" is the pH of the fermentation broth to the extent that growth and / or production of lactate material by the acid-tolerant bacteria ceases. The cessation of growth and / or production of lactate material may be the result of a change in the reaction temperature, the depletion of one or more necessary nutrients in the fermentation broth, a deliberate change in pH, or separation of the fermentation broth of bacterial cells. In those cases where the fermentation is deliberately stopped by addition to the acid or base fermentation broth sufficient to stop lactate production, the final incubation pH is defined as the pH of the nutrient medium just before the addition.

Alternatively, the growth and / or production of the lactate material can stop due to the accumulation of one or more fermentation products and / or a change in the pH of the broth, which results from the production of fermentation products, ie , the fermentation reaction has reached a self-limiting point for the given set of incubation conditions. As noted before, it is quite common for bacterial fermentations, which produce an organic acid, such as lactic acid, to be subject to inhibition of the final product. The term "lactate material", as used in this application, refers to 2-hydroxypropionate either in its free acid or salt form. The terms "lactic acid" and "free lactic acid" are used interchangeably herein, to refer to the acid form, i.e., 2-hydroxypropionic acid. The salt or dissociated form of lactate is specifically referred to herein as a "lactate salt", for example, as the sodium (or calcium) salt of lactic acid or sodium lactate (or calcium lactate). It has been found that the nutrient media suitable for use in the present process, preferably include at least about 50 g / L of carbohydrate. More preferably, the nutrient medium includes at least about 70 g / l and, most preferably, at least about 90 g / l of the carbohydrate. The carbohydrate is usually made of glucose, fructose, galactose, melibiose, sucrose, raffinose, stachyose or a mixture thereof. Glucose, fructose and sucrose are particularly suitable for use as a source of carbon and energy in the nutrient medium. It is generally not useful to incorporate more than about 150 g / L of carbohydrate in the medium. It has been found that it may be advantageous to include a base, such as calcium carbonate (CaCO3), sodium hydroxide (NaOH), ammonium hydroxide (N H4OH) and / or sodium bicarbonate (NaHCO3). Normally, at least about 30 g / l of calcium carbonate (or an equivalent amount of another base) is added to the nutrient medium. In some embodiments of the process, for example, modalities that produce higher levels of lactate, it may be preferred to include up to about 40 g / L of calcium carbonate in the nutrient medium. Although higher levels of base may be employed, due to the limitations in the solubility of the calcium carbonate salts and the desire to maintain a relatively low broth pH, it is generally not useful to incorporate more than about 1000 g / l carbonate of calcium carbonate. calcium in the middle. Very often, the full amount of calcium carbonate present will not initially dissolve in the nutrient medium. As the fermentation proceeds, some of the calcium carbonate may react with the lactic acid being formed to generate calcium lactate. As this occurs, additional portions of the undissolved calcium carbonate may creep into the solution. The overall effect is to neutralize a portion of the lactic acid in formation and prevent the pH of the broth from falling below a desired level (eg, below about 3.8-3.9). It may not be necessary to add a base, such as calcium carbonate to achieve this effect. A solution containing a lactate salt (eg, calcium, sodium or ammonium lactate) can be added to help buffer the pH of the fermentation broth. An example of a process in which this could occur would involve the separation of a fraction of the fermentation broth from the incubating bacteria, and recycling the portion back to fermentation after the removal of some or all of the free lactic acid in the fraction . Alternatively, the calcium lactate could be isolated from the fermentation broth (eg, in solid form), and mixed together with the nutrient medium that is added to the fermentation. In general, the addition of lactate salt as a buffer salt can be advantageous because it minimizes the amount of neutralizing base added to the fermentation broth., thereby minimizing the amount of lactate produced that is converted to the salt form. Nutrient media including at least about 70 g / L of glucose and / or fructose and at least about 20 g / L of calcium carbonate are particularly suitable for use in the present process. Depending on the bacterial strain used in the process, the incorporation of maize infusate (eg, in an amount equivalent to at least about 25 g / l dry solids of maize infusion water) may also be preferred in this half nutrient It is particularly useful to add corn infusate containing only the same lactic form of lactate to be generated by the fermentation process. The strain of homolactic bacteria and the fermentation conditions are normally chosen so that the free lactic acid is produced at a global rate of at least about 0.5 g / l / h, preferably at least about 1.0 g / l / h, more preferably at least about 2.0 g / l / h, and most preferably at least about 4.0 g / l / h. As used herein, the overall production rate of either lactate or free lactic acid (or lactate) is calculated by dividing the total amount of free lactic acid (lactate) produced by the incubation time. For fermentations where a limiting lactate concentration occurs, the overall production rate of free lactic acid (lactate) is calculated over the time required to produce 90% of the limiting concentration of free lactic acid (lactate).

The productivity of the present process can also be expressed in terms of the overall production rate of lactate. The present fermentation process is generally carried out under lactate-producing conditions at a global rate of at least about 1.0 g / l / h, preferably at least about 2.0 g / l / h and more preferably, at least about 3.0 g / l / h. As indicated herein, lactate is preferably produced at these rates in a broth at an average incubation pH of no more than about 4.1, and more preferably, no more than about 4.0. lll. Separation of the fermentation broth - lactate vs. lactic acid A variety of issues arise about the development of a processing approach for lactate / lactic acid solutions that involve the generation of large amounts of lactic acid; for example, in solution at pHs no greater than about 4.8 (preferably not greater than about 4.2 or 4.3) of the fermentation broth; and with concomitant (and if desired, recycled) isolation of lactate salt (usually calcium lactate, potassium lactate, sodium lactate and / or ammonium lactate). The main concerns relate to the design of the system to achieve the two objectives of: 1. Isolation of lactic acid products for consecutive processing, for example, to generate polymer; and 2. Isolation of lactate salt, preferably in a desirable form for recycling to the fermentation broth. Three general approaches interest: 1. Separation of lactic acid from the solution leaving the lactate salt behind; and if it is desa, the direction of the residual solution having the lactate salt therein, after separation, towards a fermentor; 2. Isolation of the lactate salt from the solution; d direction of the lactate salt, if desired, towards a fermentor; and a consecutive isolation of the lactic acid product from the residual solution after separation of the lactate salt; and, 3. Simultaneous separation of lactic acid in one stream and lactate salt in another, leaving residual mixture. With the techniques described here, each is possible. However, the advantageous global processes will depend, in part, on the selection, among the approximations, of the one most widely facilitating an efficient and cost-effective overall processing scheme in a large-scale implementation.

The techniques herein can be practiced in a variety of solutions of lactate material (ie, solutions of lactic acid and dissolved lactate salt). These solutions may comprise fermentation broth or broth that has been removed from a fermentor and has been modified in some way, for example, by filtration or pH adjustment. Actually, the techniques can also be applied to solutions that are made in another way. However, the techniques and proposals described herein are developed particularly with a focus on efficient processing of fermentation broth solutions, especially relatively acid ones, in which pH modification is not required by the addition of acid and preferably It has not happened. Although the techniques described herein are particularly well suited for processing selected bacterial fermentation broths, they can be applied to mixtures of lactic acid, such as, those obtained from: fungal action or yeast; purge streams of lactide reactions; or, polylactic acid currents of polylactic acid processing. As described in the patent application of Carson et al. incorporated herein by reference before, alternative approaches based on the fermentations of other acid-tolerant microorganisms have also been reported. Yeasts, such as, Saccharomyces cereisiae, are capable of growth at much lower pH than lactate dehydrogenase gene from a bacterial (lactobacillus) or mammalian (bovine) source in Saccharomyces cerevisiae. Recombinant yeast strains are, it is said, capable of producing lactate at or below the pKa of lactic acid (approximately 3.8). However, ethanol is the main fermentation product generated by these recombinant yeast strains. This both decreases the efficiency of lactate production and introduces additional potential issues with respect to the separation and purification of free lactic acid. The production of lactic acid has also been reported by a pelleted form of the fungus, Rhizopus orgyzae. This fungal fermentation also normally produces glycerol and / or ethanol as main byproducts. The yield of free lactic acid was optimized in this case by continuous removal of the fermentation broth using a polyvinylpyridine column ("PVP"). It was not reported to have generated lactate concentrations greater than approximately 25 g / l using the Rhizopus / PVP method. The normal compositions in which techniques according to the present invention can be applied, with respect to pH, would be at least 0.86 and less than 6.0. That is, the normal compositions in which the techniques will be practiced will have a pH within this range. As indicated by Equation 1, and Figure 11, for such com positions, the molar ratio of free lactic acid to dissociated acid or dissolved lactate salt at 25 ° C is within a range of about 1,000: 1. up to 0.007: 1. The most preferred process will involve solutions with a pH in the order of approximately 1.98-5.00 (ratio of H LA: LA within the range of approximately 75: 1 to 0.070: 1); and, more preferred processing will involve solutions having a pH within the range of about 3.0-4.5 (ratio of HLA: LA within the range of about 7.0: 1 to 0.23: 1). As indicated above, solutions within the most preferred pH range described above are readily obtained via the present fermentation process with substantial concentrations of the lactate material therein. Alternatively, other fermentation broths may be used, for example, with pH adjustment by adding acid normally to the given more preferred pH range. Certain preferred methods of acidulation are described below. In the present, there will sometimes be reference to "preferential separation" of: lactic acid of a composition containing lactic acid and lactate salt; or, lactate salt of a composition containing lactic acid and lactate salt. The term "preferential separation" and variants thereof, in this context, refers to a separation technique that preferentially removes one of the two components (lactic acid or lactate salt) from the other. In a normal preferred processing according to the present invention, a mixture of lactic acid and lactate salt is divided into two "product streams". In a product stream, (ie, the stream rich in free lactic acid), preferably the molar ratio of free lactic acid to lactate salt obtained is at least 2/1 and preferably at least 3/1. With certain of the techniques described herein, proportions of at least 5/1 and actually in proportions of 1 0/1 or more are readily obtainable.

The other product stream is the lactate salt rich stream. In this stream, preferably the molar ratio of free lactic acid to lactate salt is not greater than about 0.5. With normal preferred processing as described herein, proportions of no more than 0.3, preferably no more than 0.2 and most preferably 0.1 or less are easily obtained. In the present, the term "current" as used in the context indicated by the previous two paragraphs, refers to an isolated phase or product segment, without considering whether that phase or segment of product is a solution, solid or a materials mix. In this way, a "lactic acid rich stream" is simply a phase or mixture rich in lactic acid (versus lactate salt) compared to the original processed mixture; and, a "lactate salt rich stream" is a stream rich in lactate salt (versus lactic acid) compared to the original processed mixture. When the product stream enriched in free lactic acid is obtained as a result of the separation of free lactic acid from the mixture, for example, from a fermentation broth, the aqueous mixture remaining after the lactic acid removal will sometimes be referred to as "suppressed" with respect to free lactic acid. Similarly, when the enriched stream of lactate salt results from the separation of the lactate salt from a mixture containing the lactic acid and the lactate salt, the remaining mixture will sometimes be referred to as "suppressed" with respect to a lactate salt. Preferably, when the processed solution is a fermentation broth, the product stream enriched in lactate salt is provided and formed such that the weight ratio of fermentor impurities, to lactate salt therein, is less than found in the fermentation broth, preferably by a factor of at least 5. This can be handled by techniques described herein which concern the control over the particular approach selected for the isolation of the lactate salt, as well as through use as various purification techniques, such as, retro washing or recrystallization. Preferably, the lactate product stream is optionally isolated as an aqueous solution or mixture of an aqueous phase and a solid phase, for convenient recycling in a fermentation system, in order to maintain water balance. If the concentration of an aqueous solution is used in order to facilitate water balance in the broth, preferably, relatively low cost concentration techniques are used, such as reverse osmosis and vapor recompression.

IV. Various options for separation of lactic acid / lactate salt; advantages and disadvantages A. Removal of lactic acid from the fermentation broth (or other mixture of lactic acid / lactate salt) One class of advantageous processing approaches involves the removal of lactic acid from the fermentation broth or other mixture, while leaving behind the soluble lactate salt in the fermentation broth. (Separation may occur, in some cases, inside the fermenter, or may be conducted in solution material removed from the fermenter). If, after such separation, the residual fermentation broth can then be recycled, one can also retain at least part of the various nutrient values in the broth, for use in the feed.

A number of approaches can be used to preferentially separate lactic acid from a fermentation broth (or other mixture) including materials such as lactate salt and other dissociated salts therein. Approaches include the following: 1. Extraction. It is possible to remove the lactic acid from a lactic acid / lactate mixture, such as a fermentation broth, by extraction. For example, extraction can be conducted with an insoluble amine to water, preferably amines having at least 18 carbon atoms, most preferably tertiary amines, see for example, U.S. Patents: 4,771, 001; 5, 1 32,456; and 5.51, 0.526; and Sh imizu et al, J. of Fermentation and Bioengineering (1 996), Vol. 81 pp. 240-246; Yabennauor and Wang, Biotech Bioeng. (1 991) Vol. 37, p. 1 095-1 1 00; and, Chen and Lee, Appl. Biochem. Biotech, (1 997), vol. 63-65, pp. 435-447. These six references are incorporated herein by appointment. The extraction of lactic acid is a favored approach when the lactic acid is split into two immiscible liquid phases. The scaling and performance of extraction processes is direct. The extraction processes are favorable due to the lack of solids handling, the wide variety of equipment available to contact two immiscible phases and the capacity to handle large flow velocities. The extraction processes may suffer when the phases tend to form stable emulsions or have a high viscosity. One also to worry about: (a) entrained components and soluble solvents that affect the productivity of the microbe; and (b) the extraction process that removes important nutrients from the recycled broth. The extraction process can be done in the fermenter, in an external contactor, or with the use of a membrane to maintain a phase of dispersing in the other. The use of a supported liquid membrane can be useful depending on the overall separation process. The choice of extraction solvent is important for the overall efficiency and economy of the separation process. A measurement of the extraction efficiency is the partition coefficient as calculated by the concentration (basis in weight) of the lactic acid in the organic phase (extractor) divided by the concentration of the lactic acid in the aqueous phase (phase in which it occurs the removal). It is desired to have a partition coefficient greater than 0.1, even more desirable is a partition coefficient greater than 1.0, and even better if the partition coefficient is greater than 3.0. The latter can be achieved by selecting the solvent or mixture of suitable solvents from the following preferred solvents. Of course, in the practice of the commercial scale, the extraction efficiency is the ability to achieve a combination of high performance, low extractor volume and concentrated product. This can be achieved with the techniques discussed here. Solvents that provide favorable partitioning include: oxygenated solvents, phosphate esters, phosphine oxides, amines and mixtures of these solvents. Suitable oxygenated solvents include alcohols, ketones, ethers, esters, acids or solvents having a multiple number of these functional groups. Solvents are preferred which include at least 60% by weight, more preferably at least 80% by weight, and most preferably at least 90% by weight, (usually 95% or more), component or components which are generally immiscible in water (solubility not greater than about 50 grams per liter in water at 25 ° C). The specific usable solvents are 1-butanol, 2-ethyl hexanol, 1-octanol, methyl isobutyl ketone, cyclohexanone, disobutilyl ketone, isopropyl ether, ethyl acetate, isobutyl acetate, ethyl lactate, butyl lactate, octyl lactate, N, N-dibutyl lactamide and hexanoic acid. Suitable phosphate compounds include tributyl phosphate, triphenyl phosphate, diethylhexylphosphoric acid, and trioctylphosphine oxide. Suitable amines include triethylamine, dioctylamine, trioctylamine, tridecylamine, methyl didodecylamine and industrial preparations, such as, Amberlite LA-1 (a mixture of dialkyl amine with twelve carbon atoms in each alkyl chain), Alamine 304 (tridodecylamine ina), Alam ina 308 (a mixture of branched chain trialkyl with a total of 8 carbon atoms in each chain), and Alamine 336 (a commercially available mixture of trioctyl "tridecyl", dioctyldecyl "and didecyloctyl amine). The extraction solvent may also preferably contain a hydrocarbon fraction, such as urea, usually (if used) at 1 to 40% by weight Such a hydrocarbon fraction, favorably modifies the viscosity, phase coalescence, and other properties of the system A usable and frequently preferred solvent system comprises, by weight, 0 to 15% ethanol; 65 to 85% of Alamina 336 and 1 to 35% of kerosene.

Depending on the lactic acid product of interest, the characteristics of the solvent will vary. If the lactic acid product of interest comprises oligomers of lactic acid, a solvent with a relatively low boiling point is advantageous in comparison with lactic acid / lactic acid oligomers (preferably lower than 200 ° C to 1.01 × 1 05 Pa). or 760 mm Hg), because the solvent can be vaporized and easily separated from the lactic acid oligomers. If the lactic acid product is an alkyl lactate ester, such as methyl lactate, a solvent with a relatively high boiling point in comparison with the ester (s) is preferable (preferably greater than 1 75 ° C to 760). mm Hg) to easily distill the methyl lactate away from the solvent. In addition, when a lactate ester is formed, it may be advantageous for the alcohol of that ester to be a component in the extraction solvent. If the product is a lactic acid amide, it may be useful to have the corresponding amine present. On the contrary, if the product is lactic acid, the presence of a non-tertiary alcohol or amine in the solvent may be unfavorable due to the possibility of loss of yield in the formation of esters or amides. 2. Adsorption. Another approach for the isolation of lactic acid from a fermentation broth including free lactic acid and dissolved lactate salt is through: adsorption of lactic acid on a solid adsorbent; consecutive physical separation of the solid adsorbent from the liquid phase; and, eventual generation of lactic acid from the solid adsorbent. (In the present, the term "adsorption" is intended to include absorption within its scope, ie, the specific mechanism of interaction is not referred to, unless otherwise specified.) The partition of free lactic acid in one phase Solid either by ion exchange or adsorption is another favorable method to separate the lactic acid from an aqueous solution. These methods show good efficiency when the solid phase has a high capacity for lactic acid, an efficient regeneration cycle, a long life in the process. Excessive pressure drop over a solid bed, bed swelling, possible dilution of product over regeneration, resin contamination and slow mass transfer rates can make solid phase processes difficult. The capacity of the resin is an important characteristic of the resin because it determines, together with the mass transfer rate, how much resin is required for a given amount of lactic acid. A resin with a capacity of 0.10 g of lactic acid per g of dry resin would be adequate, a capacity of 0.20 g of lactic acid per g of dry resin is better, and a capacity of 0.30 g of lactic acid per g of Dry resin is the best. The latter can be achieved, for example, with a Dowex MWA-1 resin in equilibrium with a lactic acid solution of 20 g / liter having a pH of no more than 4.5 at room temperature. The contact of a solid phase with the aqueous lactic acid solution can occur in the fermenter or in the external equipment of the fermenter. For contact within the fermentor, the microorganisms are immobilized and the solid phase adsorbent is separated from the microorganisms on the basis of differences in velocity of fall in a fluidized bioreactor. See, for example, Davidson and Scott, Biotechnology and Bioengineering, (1992), vol. 39, pg. 365-368, incorporated herein by reference. Ion exchange resins that would be suitable for the recovery of lactic acid are weak, moderate and strong basic anion exchangers. As the pH of the aqueous lactic acid stream increases, a stronger basic anion exchanger is required to recover the lactic acid. Consequently, the pH of the lactic acid stream will be a factor in the choice of the ion exchange resin. Commercially available tertiary amine ion exchange resins that would be suitable include Reillex 425 and Reillex H P (both poly-4-vinyl pyridine resins), Reilly I ndustries, I nc. from I ndianapolis, IN), Dowex MWA-1 and Dowex 66 (both tertiary amine resins of polystyrene-divinylbenzene, Dow Chemical Company of Midland, Ml), and Duolite A561 (a copolymer of tertiary amine acrylic-divindylbenzene) and Amberlite I RA-67 (a phenolic-formaldehyde-cross-linked tertiary ammonium resin) (Rohm and Haas Corp. of Philadelphia, PA). Both macroreticular and gel resins are suitable.

Another important factor in the choice of resin is the technique available to remove lactic acid from the resin. As the basicity of the resin increases, the regeneration method must be more "powerful" to remove the lactic acid from the resin. A suitable regeneration method would be to contact the resin with a polar liquid possibly at an elevated temperature. Suitable polar liquids include water, aqueous solutions, methanol, ethanol, triethylamine, methyl isobutyl ketone, dimethyl sulfoxide, N-methylpyrrolidinone, 1,4-dioxane, tributyl phosphate, trioctylphosphine oxide and various combinations thereof. The evaporation of the lactic acid product is also potentially a method. In the case of evaporation, thermally stable resins, such as Reillex, are very useful. King et al disclose the use of aqueous trimethyl amine solutions for regenerating adsorbents and distilling water and trimethylamine to isolate lactic acid in U.S. Patent 5,132,456, incorporated herein by reference. The selectivity of the resin is also important, since preferably the resin should be selective for lactic acid and none of the nutrients needed for microorganisms. The resin may also need to be washed with solvents, acids, and / or bases before use to minimize any leakage of monomers, olomers, or other compounds that may be toxic to the microbes. 3. Separation by vaporization. The distillation of lactic acid from the aqueous solution or mixture is an alternate method of separation. This method would not contaminate the recycle stream with residual extractor material that could be toxic to microorganisms and allows good control of the water balance. A disadvantage of this approach is that water needs to be distilled first. This is an energy consumer; and as the water is removed, the conditions are favorable for the condensation of lactic acid. Vacuum conditions for distillation (ie, less than 4.00 x 1 04 or 300 mm Hg) will be preferred to lower the distillation temperature because the condensation of lactic acid to dimer or oligomer is reduced. The recovery of lactic acid can be facilitated by using equipment, such as a thin film evaporator which minimizes the monomer residence time during distillation. One possibility is to add an alcohol, such as ethanol and form ethyl lactate, which has a higher volatility compared to lactic acid, and as a result, is easier to distill. 4. Separation via membrane. The lactic acid can pass through a membrane in a separate aqueous phase. This method will be favorable if the membrane chosen is one that is highly selective for lactic acid (versus lactate salt). A useful membrane is a dense hydrophilic membrane, such as, Celgard 3400 from Hoechst Celanese Co. of Somerville, New Jersey and anion exchange membranes that also allow the transfer of protons. An example of this type of process is having a solution of ammonia through the membrane from the lactic acid solution. The lactic acid is conducted through the membrane to neutralize the ammonia. The ammonium lactate solution can then be subjected to conditions that vaporize the ammonia to give a lactic acid solution. Other volatile bases, such as trimethylamine or triethylamine, could be used alternatively. The use of a strong base, such as sodium hydroxide, would normally undesirably neutralize lactic acid to sodium lactate. Thus, in general with such an approach, a weak base should be used on the opposite side of the membrane of the fermentation broth.

That is, a weak base, such as the amino acid bases mentioned above, will form an association that can be easily dissociated to regenerate the lactic acid. The term "weak base", as used herein, refers to a base with a pH of half the neutralization of less than 2.5.; a "moderate" base will be considered as a base with a pH of neutralization half of between 2.5 and 7.0; and a "strong" base will be considered as a base with a pH of neutralization half of more than 7.0. The term "pH of the neutralization moiety" is a measure of apparent basicity of a water-immiscible base, as defined in Grinstead, R. R. et al. , J. Phys. Chem. , vol. 72, # 5, p. 1630 (1968), incorporated herein by reference. Regardless of which method is used for the removal of lactic acid from the fermentation broth, it needs to be considered the consecutive destination of the lactate salt and residual fermentation broth. Of course, it would be preferable to use a technique which leaves the residual fermentation broth and the lactate salt in a desirable form for direct recycling, without further significant treatment. On the other hand, it may be desirable to isolate the lactate salt from the residual broth, so that the lactate salt can be recycled or otherwise used, with the fermentation broth either directly recycled, or disposed of or used in other ways. Several approaches for the removal of the lactate salt from the residual fermentation broth, after the removal of lactic acid, include: extraction; electrodialysis, exclusion of ions, adsorption with a solid adsorbent, with consecutive separation of the adsorbent; separation with membrane; and, crystallization. It is noted that in some cases, the techniques involved can also lead a material to the recirculation stream, having the lactate salt in solution. For example, extraction methods can affect the composition of this stream. When such approaches are chosen, it is important either to use materials that will have low toxicity to the microorganisms, or to develop consecutive treatments that modify the composition of the stream appropriately for recycling, for example, by flashing the volatile compounds from the residual broth or Bring the broth in contact with an immiscible liquid of low toxicity that extracts toxic components. A preferred method would be to use an immiscible liquid of low toxicity as the extraction solvent or as a component in the extraction solvent and as the immiscible liquid is used the toxic components are extracted. The techniques described in this section can be practiced in a continuous or batch manner. In fact, even the fermenter feed flow and fermenter operation can be practiced continuously or batchwise.

B. Removal of the lactate salt from the fermentation broth - leaving behind the lactic acid As indicated above, an alternate approach to the production of lactic acid involves the separation of lactate salt from the fermentation broth (or other mixture), leaving back the lactic acid in the residual mixture, with subsequent processing of the lactic acid from the residual mixture. The isolated lactate salt could be useful for directing (or recycling) the fermentation system for pH control, if desired.

A variety of approaches can be used for the isolation of lactate salt, a fermentation broth or another lactic acid / lactate salt mixture, leaving behind lactic acid. This could be developed, in general, around the same approximations as characterized in the previous section for the isolation of the lactate salt from the residual fermentation broth after the removal of lactic acid. As with the approaches of the other section, they can be practiced in a continuous or batch manner. Then, the approximations would generally be the following: 1. Extraction. A lactate salt can be extracted from an aqueous solution containing lactic acid with the use of a quaternary amine, such as methyl trioctyl ammonium chloride or a mixture of methyl trialkylamino chloride salts, such as ALI QUAT 336 ( corresponding ammonium methyl of Alamina 336, available from Henkel Corp. in Kankakee, IL). Typically, trialkyl ammonium (preferably chloride) halide salts of trialkyl amines of 1 8 carbons or more are used. In general, an anion exchange occurs, in which the lactate anion is exchanged for the chloride anion present in the amine phase. In this way, this approach can "load" the residual solution, containing the lactic acid therein, with chloride ion.

Another extraction approach is to completely extract the lactate salt using a coupled extractor consisting of both a liquid cation exchanger and liquid anion in a solvent. An example would be the use of quaternary amine as listed above, with diethylhexylphosphoric acid. The lactate salt is extracted with the formation of water. The quaternary amine may need to be pretreated to the free base form of the amine for this to work efficiently. 2. Solid adsorbent. The fermentation broth containing both the lactic acid and the lactate ion could come into contact with a solid adsorbent, for the removal of the lactate ion. The preferred solid adsorbents for this would be strong anion exchangers, such as quaternary ammonium fixed compounds. An example would be Amberlite I RA-400 and Amberlite I RE-900 available from Rohm and Haas Co., Philadelphia, PA. Such materials generally comprise quaternary ammonium functionality and copolymer styrene di vini I benzene. Another approach would be to use mixed bed ion exchange resins to separate the lactate salt from the aqueous solution. This is similar to the mixed liquid ion exchangers mentioned above.

Another approach to separating the mixture using a solid adsorbent is the ion exclusion technique. In an ion exclusion chromatography, an anion exchange resin is converted to the lactate form. The lactate anion in the feed solution exits with the empty volume while other ionic components are retained by the resin. 3. Separation with a membrane. The lactate ion can also be removed from an aqueous solution, such as a fermentation broth, leaving a lactic acid behind, by electrodialysis. More specifically, the fermentation broth, preferably pre-filtered, and a relatively pure stream of water are fed to an electrodialysis unit. The unit would include alternating cation and anion exchange membranes forming a plurality of compartments (or stacks) with a cathode and anode on opposite sides of the stack (to provide an electric field through the stack). The properties of the membranes would be such, that only the anions would pass substantially through the anion exchange membrane and only the cations would pass through the cation exchange membranes. An electrodialysis unit for water desalination would be suitable for separating and concentrating the lactic salt, and providing a lactic acid / suppressed stream in lactate salt. Companies such as, Aqualytics in Warren, NJ and Lonics, I nc. in Watertown MA, they provide appropriate desalination equipment for this use. 4. Crystallization. The lactate salts can be crystallized from aqueous solutions.

In this way, the lactate salt can be removed from the fermentation broth (or other mixture) via a crystallization process. This can be done through concentration (for example, by evaporation of water); by reduction in temperature and / or by addition of agents to facilitate crystallization (for example, water-soluble alcohols, such as C1 to C4 alcohols (methanol, ethanol, propanol and / or the various butanols). Physical separation of the crystallized product from the solution, the remaining lactic acid containing solution could then be further treated for lactic acid isolation.In some preferred processing, the added agent is preferably one that has a low solubility in water at about room temperature, but this solubility increases acutely on an increase in temperature Butanol provides a good example: the addition of butanol to a solution containing lactic acid and lactate salt at an elevated temperature, approximately 1000 ° C, and under appropriate pressure avoid vaporization, it would result in efficient crystallization of the lactate salt. Po agent has many disadvantages. First, it is relatively easy to separate from the remaining solution after crystallization, for example, by cooling. Second, on the cooling of the liquor after the crystallization of the lactate salt, the lactic acid will be distributed between the two phases (water and butanol) forming a very efficient combination of crystallization and extraction in one operation. That is, the lactate salt is crystallized and the lactic acid is extracted in butanol. Third, if the added agent is an appropriate alcohol, a lactate ester can be formed and separated in pure form, for example, by distillation. If the crystallization is the selected approach, often the calcium lactate (CaLa2) will be the salt selected because: (a) it has a relatively low solubility in water; and (b) its solubility in water is strongly dependent on temperature. The calcium lactate salt can be provided by the use of the appropriate soluble calcium salt, such as calcium carbonate, to the mixture. Of course, whenever the approach is used to isolate the lactate from the mixture, the overall processing scheme will require recovery of the lactic acid value, in some form, from the residual broth (or other mixture) after the removal of the salt. of lactate. Approximations analogous to those described above can be used, with respect to the removal of lactic acid from a fermentation broth or other mixture. More specifically: extraction; solid adsorption; vaporization; or membrane separation, as previously described, are feasible. For many of these options, a previous separation step of the lactate salt will be beneficial, because in some cases acid separation will be more efficient when conducted without the buffering effect of the lactate salt. Thus, the techniques described above for the recovery of lactic acid would be applied to a solution suppressed in lactate, to achieve a purification / isolation of lactic acid instead of the separation of lactate.

V. A preferred class of approaches - removal of lactic acid from the mixture by extraction In some cases, a preferred class of global processing approaches will involve the separation of lactic acid from the mixture via extraction. Among the reasons for this are the following: 1. With an extraction process, especially if it is practiced in a clarified fermentation broth or similar solution, it may be possible to leave the residual solution suppressed in lactic acid, with the lactate in it, in an appropriate form for recycling in the fermenter without treatment additional substantial, different perhaps to a diluent wash or similar treatment to remove any residual extractor that may be toxic to the microorganisms of the fermentor. 2. The extraction processes can be conducted, in many cases, efficiently and quickly on a large scale. 3. The extraction processes can be quite selective for lactic acid, versus other materials (such as amino acids and carbohydrates) in the fermentation mixture. Such high selectivity can be achieved by using basic extractors, such as trialkylamines, especially relatively insoluble trialkylamines having at least 1 8 carbon atoms, such as, Alam ina 336. A variety of approaches can be used for the recovery of lactic acid, say, the removal of the lactic acid, or lactic acid, values of the extract containing lactic acid. Approaches include the following: A. Phase separation. In general, when this technique is applied, the extractor containing the lactic acid is modified to generate lactide and / or oligomers of lactic acid. This would be done, for example, by handling the condensation reaction (lactic acid to dimer or oligomer) through evaporation of water during concentration. To facilitate such a sequential process, a hydrophobic extraction solvent is preferred for the original extraction, because most of the separation of the lactic acid from the water of the original aqueous phase (eg, fermentation broth) will have been done by the step of extraction and separation of phases. An example of a suitable hydrophobic extraction solvent is one with a high proportion of long chain alkylamines and at least 1-35% kerosene, by weight. Those extraction solvents usually coextract only one mole of water per mole of extracted lactic acid. The condensation reaction (to form lactide or oligomer) can be facilitated by the addition of catalyst. In general, during condensation / concentration, the resulting lactide or oligomers will form a phase separated from the remainder of the extractor, for example, the amine. The physical separation can then be used to achieve the recovery of the desired lactic product. The separated oligomer can then be taken directly to lactide, without the removal of residual extractor therein, if desired. It is noted that this approach, especially when an amine is used as the extractor, can lead to some degree of racemization of the lactic product. Racemization can be minimized through the use of low temperature, below 150 ° C and below 2.67 x 1 03 Pa or 20 mm Hg.

B. Extraction.

With this approach, one can again extract the lactic acid from the first extracted phase. This can often be done with an aqueous extraction or washing, due to the high solubility of lactic acid in water. Of course, other polar liquids may be used, such as dimethylsulfoxide (DMSO), N-methylpyrrolidinone, N, N-dimethylformamide (DMF), triethylamine and lactide. In some cases it may be desirable to use relatively hot extraction conditions again in comparison with the first extraction, for example, re-extracting with water at a temperature of at least 1 00 ° C, normally about 150 ° C or higher, for facilitate the process. (Assuming conditions of the first extraction of 1 5 ° -60 ° C at atmospheric pressure). Such back-extraction would normally be conducted under a pressure of at least 2.07 x 1 05 Pa.

C. Membrane separation. Membrane separation techniques can be used to facilitate the separation of lactic acid from the lactic acid-extraction phase. For example, one could use a hydrophilic barrier with the extractor phase on one side and a preferred phase for lactic acid on the other. This preferred phase could be, for example, a tertiary amine, as described above, for membrane separation for lactic acid from the fermentation broth. In some cases, aqueous systems can be used. D. Solvent distillation. If the solvent of the extractor phase is of relatively low molecular weight or high volatility, for example, butanol, it can be distilled or flashed from the extractor phase, leaving behind the lactic acid. This approach will be very useful when the extractor phase comprises solvent such as: butanol, methyl isobutyl ketone or triethylamine. It may be desirable to use conditions of relatively low pressure to facilitate distillation. For example, processing will be preferred at pressures in the order of about 6.67 x 1 04 Pa or 500 μm Hg or less. The use of a carrier and pervaporation gas is also preferred. In some cases, the concentration of lactic acid that occurs during distillation will lead to the formation of condensation products, such as lactate esters (if alcohols are present), lactide or oligomers of lactic acid. E. Distillation of lactic product. The distillation of lactic product, for example, lactic acid, from the extractor phase will be favored when the extractor phase comprises a material of relatively low volatility. For example, when a tertiary amine, especially tertiary amines of 1 8 carbon atoms or more, is used, in the extractor phase, the distillation can be easily used to recover the lactic acid. With respect to this, attention is directed to US Patent 5,51 0, 526 incorporated herein by reference. Of course, in some cases, the extraction phase may contain both high volatility and low volatility materials. When this is the case, multistage distillations may be preferred to obtain the isolation of the lactic acid or lactic acid product. Here again, a carrier gas and particularly pervaporation could be advantageous.

F. Crystallization of lactic product. When the lactic acid product is lactide, a favorable approximation for the separation of the extracted phase is crystallization. More specifically, lactide crystallizes easily from non-polar solvents, such as toluene. Normally it will be desired to generate lactide from the lactic acid recovered in the extraction. This can be done by removing water and condensation under controlled conditions. See, for example, 5, 142.023 incorporated herein by reference, with respect to lactide formation. G. Aqueous extraction, re-extraction of solvent. With this approach, the lactic acid is extracted from the extractor phase into an aqueous phase. Then it is removed from the aqueous phase to a preferred extractor phase for consecutive processing, such as, condensation to oligomer and eventual processing to lactide. A typical example would be a first extraction to a tertiary amine phase, preferably amines of 1 8 carbons or more, with a consecutive extraction of the tertiary amine phase in an aqueous phase. The lactic acid can be extracted in cyclohexanone or another polar solvent, not an amine, with condensation (a oligomer) occurring in the cyclohexanone (or other organic solvent, polar, not amine) during concentration / distillation. The temperature of the back-extraction in the aqueous phase may be higher than the temperature during extraction in the organic phases. This is preferred for condensation within the tertiary amine phase directly, if racemization is to be minimized or avoided.

Of course, the lactic acid could be isolated from the aqueous phase directly, for example, by distilling the water. However, in general, this may require more energy than condensation within a preferred polar organic phase of higher volatility.

SAW. Several approaches for the separation of lactic acid from an extracted phase - a closer look The following describes a normal approach to the problem of making lactic acid products from a fermentation of lactic acid or other aqueous solution of lactic acid, using the techniques described before. Assume a broth comprising approximately 50 to 1 10 g / l of total lactic material at a pH of 3.5 to 4.3. The broth is removed, continuously, from a fermentor. The broth is clarified to remove coarse impurities and other insoluble substances in the stream, for example, by passing through a filter (or through flocculation, centrifugation or a combination of these various techniques). This filter can be a dead end filter or a cross flow filter using micro- or ultrafiltration membranes. (In some cases, pretreatment with activated carbon can also be conducted to purify the mixture.) The undissociated lactic acid is then extracted from the remaining lactate salt solution or broth. The extraction solvent includes a tertiary alkylamine, an oxygenated solvent that increases the partition coefficient and a kerosene fraction that modifies the viscosity of the solvent mixture. Preferably, the extraction solvent contains 60 to 80% by weight of tertiary alkylamine, such as, Alamine 336, 5 to 20% by weight of methyl isobutyl ketone, and 1 to 30% kerosene (for example, IsoPar K). ). The aqueous solution of lactic acid and extraction solvent are contacted in a counter-current manner either in: a stirred column; a packed column; a column of perforated plate; a rain bucket contactor; a centrifugal contactor; or mixer / settler equipment. The temperature during this contact is between 0 ° C and 95 ° C, but more preferably between 1 5 ° C and 60 ° C. The streams that come out of the extraction process are an aqueous solution of lactate salt and an extract rich in lactic acid. The aqueous solution of lactate salt is recycled back to the fermenter. The extract rich in lactic acid, the lactic acid product in the extract should be separated from the solvent, in order to regenerate the solvent and isolate the lactic acid product. As indicated above, there are a number of methods for separating lactic acid from the solvent. For example, the lactic acid product can be: extracted in a second phase of low m iscibility with the extraction solvent; distilled either with the solvent or the lactic acid product going up; or passed through a membrane to another phase. In another method, all or some of the extraction solvent can be distilled while adding a second, less volatile solvent, in order to obtain the lactic acid in a different solvent composition. This method was described by Verser, et al. in U.S. Patent 5,420,304, incorporated herein by reference. A preferred separation scheme for obtaining lactic acid from the extraction solvent is to distill the lactic acid / extraction solvent stream to obtain a stream of crude lactic acid. There may be components in the extraction phase that are of greater or lesser boiling point than lactic acid. Efficient and economical distillation schemes of these components can be conducted with conventional distillation equipment. In a preferred method, high and high vacuum surface area equipment is used to efficiently isolate lactic acid with a minimum amount of condensation. A sliding film evaporator or a falling film evaporator would be suitable for this type of operation. A stream of vaporized lactic acid would be condensed to form a liquid lactic acid stream that could be further processed to oligomers of lactic acid and lactide in processes described by Gruber., et al. in 5, 142.023. A stream of vaporized lactic acid may come into contact with an appropriate catalyst to form lactide, as described by Bel I is and Bhatia in US Pat. No. 5,138,074, incorporated herein by reference. This stream of lactic acid could also be sold as a final product with purification as necessary. The lactic acid could also be reacted to form other valuable products, such as lactate esters, lactate amides and acrylic acid. Distilling the relatively small amount of lactic acid from the extraction solvent is a particularly attractive approach for the recovery of lactic acid, because the minor component of the solution is being captured at the top. Consequently, an extraction solvent with a lower volatility in relation to lactic acid would be preferred.

Alamina 336, a commercially available mixture of tertiary alkylamines with octyl and decyl alkyl groups has a lower volatility than lactic acid. It has been found that for a mixture of Alamina 336 and aqueous lactic acid at relatively low temperatures (approximately <65 ° C) and certain concentrations of aqueous free lactic acid, three liquid phases are in equilibrium; an aqueous free lactic acid concentration is about 2.2% by weight, while the concentrations of lactic acid are about 16% by weight and 1.4% by weight for the middle and upper organic phases, respectively. The total organic phase or the highly charged middle phase of lactic acid-Alamine 336 can be physically separated from the other phase (s). Lactic acid could be distilled from Alamina 336, or the lactic acid extract can be further processed using other methods described in this application to obtain lactic acid products. It should be noted that at room temperature, if the concentration of aqueous free lactic acid in equilibrium is significantly above or below 2.2% by weight, only a simple organic phase will be obtained. Example 2 reports an example of making a three-phase system, and how a simple organic phase with high concentration of lactic acid was obtained by bringing the two organic phases back into contact with a fresh aqueous lactic acid solution. As indicated above, another possible method for obtaining the lactic acid from the extractive solvent is to re-extract the lactic acid in a liquid phase, which is immiscible with the extraction solvent. The second immiscible phase can be water, polar organic compounds, or mixtures of these liquids. It has been found that some polar organic compounds are immiscible with the preferred extraction solvents described above. As the weight fraction of the trialkyl amine and kerosene increases, the probability that a polar organic compound is immiscible with the extraction solvent increases. The polar organic compounds of interest include: methanol; ethanol; lactide; oligomers of lactic acid; methyl sulfoxide; N. N-dimethyl foramide; N-methyl pyrrolidinone; and sulfolane. In general, the polar organic compounds of interest are those which have a solubility in water greater than 1 g per 1 0 g of water. The back-extraction preferably will be carried out at a higher temperature than the initial extraction of lactic acid in the extraction solvent (usually 30 ° C to 160 ° C or higher, usually at 90 ° C to 160 ° C). There are exceptions to this, and when an extraction solvent comprising a large proportion alcohol, such as hexanol or octanol, is used, it may be favorable to perform the back-extraction at a lower temperature than the initial extraction. The composition of the extraction solvent can be changed between the forward extraction of lactic acid and the retro-extraction. The proper equipment for forward extraction as listed above is also suitable for retro-extraction. The retro-extraction solvent may also have a basic compound to increase the distribution of lactic acid back to the second immiscible phase. It has been found that the ternary system of triethylamine-lactic acid-trioctylamine at room temperature has two phases. This is somewhat surprising because triethylamine and trioctylamine are miscible. The trioctylamine phase contains little lactic acid if the amount of triethylamine added is slightly more than stoichiometric. The triethylamine-rich phase is almost at a molar ratio of 1: 1 lactic acid to triethylamine, which gives a weight percent of lactic acid of 47%. In this way, this system is able to concentrate the lactic acid during the retro-extraction. Triethylamine is substantially more volatile than lactic acid and can be distilled to obtain a crude lactic acid product. Trimethylamine is expected, ammonia and other amines with a molecular weight of less than 200 would show similar behavior as triethylamine. Bailey et al. describe the use of tertiary triallets in an organic solvent with back-extraction in an aqueous phase (with a relatively strong base, such as ammonia) in U.S. Patent 4,771,001, incorporated herein by reference). Retro-extraction in a mixture of triethylamine in polar solvent with relatively low volatility is an efficient process, because the ratio of solvent to triethylamine can be carefully controlled. The presence of the solvent allows the viscosity to remain low during the distillation of lactic acid amine and would provide a means for further reactions of lactic acid to lactic acid products. This latter option of retro-extracting lactic acid has been described generally, as having a non-polar solvent with a basic extractor, such as a long chain alkylamine (1 8 carbon atoms or more), and using a solvent polar organic, as the retro-extraction phase. Of course, the opposite may be true. The initial extraction solvent may be relatively polar, but still immiscible in water, and the back-extraction liquid may be a non-polar solvent with a basic extractor. The fundamental concept is the ability to extract the lactic acid from an aqueous solution with an extraction solvent and retro-extract the lactic acid in a second liquid. In some cases, that liquid will be water, but it can also be an organic liquid that is appropriate for the efficient separation of lactic acid or for making and separating lactic acid products. When the lactic acid is retro-extracted in a second polar liquid phase, there will be a residual amount of the extraction solvent components in the lactic acid-rich retro-extraction phase. If desired, the residual extraction solvent may be reduced upon contacting the retro-extraction phase with a non-polar solvent, such as IsoPar K. This extra purification is shown in Example 3. Another option for separating the Lactic acid outside the extract is the use of a membrane process. In this case, the lactic acid passes through the membrane to a different phase in composition than the extraction solvent. A possible case is when the extraction solvent contains a long-chain alkylamine in a non-polar solvent. On the other side of the membrane is a volatile base, such as trimethylamine, in the same non-polar solvent. The membrane is an anion exchange membrane, which does not allow cations to pass, such as trimethylammonium. Lactic acid passes through the membrane to form a complex of Iactate: trimethylammonium. The volatile base is then removed by distillation and the lactic acid products can be made and separated in the non-polar mixture. In general, the use of the membrane allows the pipe of spearación of phases between two liquids miscibles of another way. As previously indicated, the formation of the alkyl lactate ester can be a condensation reaction between the lactic acid and a hydroxyl group in another molecule. This other molecule could be another lactic acid molecule or any other molecule that has a hydroxyl group. Possibilities include methanol, ethanol, butanol, octanol, dodecanol, 2-ethyl hexanol and 1,4-butane diol. The condensation reaction is driven to the production of the ester by removal of the ester and / or water from the reaction mixture. The condensation reaction can be carried out in the extraction solvent or the polar liquid that is used for the retro-extraction. The separation of the lactate ester can be carried out by evaporation or extraction. The alkyl lactate ester can also be formed by a transesterification reaction between the carboxylic acid group of lactic acid and an ester. The byproduct of a transesterification reaction is an acid, and the reaction is conducted forward to the production of the lactate ester by separation of the acid and / or lactate ester from the reaction mixture. The use of format esters, acetate esters, or other esters, which have a corresponding acid that is more volatile than the lactic acid and lactate ester formed are good choices, because the volatile acid can be evaporated out of the reaction mixture to drive the reaction to completion. The lactate ester can then be evaporated or retro-extracted in an immiscible phase of extraction solvent. An ester with a corresponding acid having a lower volatility than the lactate ester could be used. The reaction is conducted to the termination by separating the lactate ester from the reaction mixture. Suitable esters for this type of process would be a methyl ocanoate, dimethyl succinate and ethyl decanoate. The advantage of this system is that the lactate ester product is removed immediately from the reaction mixture. The disadvantage is that the by-product acid must be regenerated again to the desired ester efficiently. In the transesterification processes described above, the initial ester was chosen due to the relative volatility of the corresponding acid. This was due to the fact that evaporation was chosen as the method to remove the products of the transesterification reaction. If the back-extraction in a phase that is immiscible with the extraction solvent, is used to separate the products of the transesterification reaction, the initial ester would be chosen based on the selectivity of the immiscible phase of the corresponding acid or lactate ester. . For example, if an immiscible phase was found to select succinic acid on lactic acid, butyl lactate, butyl succinate and dibutyl succinate, the condensation reaction will be conducted to butyl lactate by extracting the succinic acid.

The initial alcohol or ester can be part of the extraction solvent or added to the lactic acid extract after it is separated from the aqueous phase. As mentioned, the removal of the products of the condensation reaction is important to handle the reaction towards the formation of the lactate esters. This removal can be simultaneous or sequential. Reactive distillation would be suitable for the simultaneous removal of one of the reaction products. A process of separation sequences may require a (recycled) direction of material back to the condensation reactor for efficient operation. The lactate ester can be further purified, if required, especially if the lactate ester is the final product of interest. Once an appropriate lactate ester stream is obtained, the polylactic acid can be obtained by the following method. Any free water in the system would be separated, and the current would be heated, possibly under subatmospheric pressure. The corresponding alcohol of the lactate ester would be evaporated from the reaction mixture to handle the transesterification reaction. For example, a stream of methyl lactate would give methanol and a lactoyl methyl lactate. As the methanol is evaporated, the molecular weight of the lactic acid oligomer capped with methyl increases. This stream would then be fed to a lactide forming reactor, where a catalyst is added. Lactide, the lactic acid cyclic ester, would then be formed and used to form polylactic acid. The processes for making lactide of alkyl lactate esters have been described by Gruber, et al. in U.S. Patent Nos. 5,247, 059 and 5,274, 073, each of which is incorporated herein by reference. The lactic acid oligomers are also lactic acid products of interest that could be used to make polylactic acid. Previously, a preferred process for capturing an extract rich in lactic acid and making oligomers of lactic acid was described. The oligomer formation is carried out in the presence of the extraction solvent, the reaction being conducted by the removal of water. The preferred process is to remove water by evaporation, usually under subatmospheric pressures. There are other methods for removing water that are suitable, such as, adsorption on molecular sieves and silica, reaction with an anhydrous salt to form a hydrated salt, and passing water preferentially through a membrane, such as pervaporation. The generation of lactic acid oligomers is discussed generally in 4, 142.023. If the extraction solvent is relatively volatile, the solvent could also be removed by evaporation, leaving a stream of concentrated lactide acid oligomer. A preferred process would have the volatile extraction solvent forming an azeotrope with water, thus aiding in the water removal step. If the extraction solvent can not be easily evaporated, other methods are used to separate the lactic acid oligomers from the extraction solvent. These methods are suitable when the extraction solvent includes a high molecular weight trialkylamine, such as a tridodecylamine, or high molecular weight oxygenated phosphorous compound, such as tributyl phosphine and trioctyl phosphine oxide. It has been found that the prepolymer can be substantially separated from the extraction solvent by causing the creation of two immiscible phases. It has been found that when Alamin 336 is used as the extraction solvent, a significant amount of the trialkylamine can be separated from lactic acid oligomers with little extra effort. The extract of Alamina 336 - lactic acid is subjected to conditions that cause the condensation of lactic acid to oligomers of lactic acid. It has been found that upon cooling these mixtures, two immiscible phases are obtained. One of the phases is the enriched Alamina 336 and the other is the enriched lactic acid oligomer. It was found that the enriched phase of lactic acid oligomer had a concentration of Alamina 336 approximately equal to a molar ratio of 1: 1 of amine to oligomer of lactic acid. In consecuenseas the average molecular weight of the oligomers increases, the average amount of residual Alamina 336 in the oligomer-rich phase will decrease. There are other ways to force the creation of a second liquid phase. The displacement of acid or base are suitable methods to be used when the extraction solvent contains a high molecular weight amine tertiary. The oligomer still has a carboxylic acid end which can strongly interact with an amna group. If another acid is added to the system and the amine selectively prefers this other acid, this acid will interact with the amine and the oligomer will be free to partition into a separate phase. The required amount of said other acid is equivalent to that of the oligomer. In this way, the higher the molecular weight of the oligomer, the lower the amount of the other acid will be required. Alternatively, another base could be added to the system and the carboxylic acid end of the oligomer will selectively prefer this other base. The amine could then form a second liquid phase. The displacing acid or displacement base needs to be separated from the amine phase or oligomer phase, respectively, possibly via distillation or ion exchange. The displacing acid or base may be added in a solution and the solvent associated with the displacing species may also need to be separated. If the lactic formation is conducted, and is carried out at temperatures between approximately 50 ° C and 250 ° C and pressures from approximately 2.67 x 1 O2 Pa to 1.33 x 1 04 Pa (2 mm Hg to 1 00 mm Hg) , the lactide produced will evaporate. This raw lactide stream may need additional purification to meet the purity requirements to make quality polylactic acid. If a stream of the lactic-forming reactor is cooled, the stream can produce a new solid phase. Lactide with a chiral purity of more than 95% has a melting point of about 96 ° C. Accordingly, the lactide can be crystallized from the extraction solvent provided that the lactide concentration exceeds the solubility of the lactide in that solvent. If the solubility of lactide is not exceeded, the addition of another liquid, an anti-solvent, to the lactic acid stream can decrease the lactide's solubility, so that the lactide crystallizes.

The filtration of the paste provides a raw lactide stream that can be used for polymer formation. Another method for the isolation of lactide, from the lactide-forming reactor, is to cause the lactide to be divided into the extraction solvent phase. This phase division could be caused by changing the temperature of a stream or adding liquid solutions to the stream leaving the lactide formation reactor. It has been found that lactide is immiscible with Alamina 336 and mixtures of Alamina 336 and IsoPar K, an aliphatic solvent of Exxon. Accordingly, the lactide could be concentrated and further processed to polylactic acid if Alam ina 336 is used as the extraction solvent. In some processes, it may be desirable to convert the lactate salt, from fermentation, into another salt, to facilitate separation, for example, the calcium salt could be converted to the sodium salt.

Vile. Some specific processing schemes The techniques described below, in connection with the Figures, are frequently presented with respect to the process system that involves the processing of a fermentation broth by removal, from the aqueous phase, of lactic acid. Sometimes processed downstream processing steps are described with respect to the lactic acid component. Of course, it is not necessary that the processing under the lactic acid / lactate mixture be conducted directly from a fermentation broth, that is, without any previous modification of the broth other than simply filtration. For example, pH adjustments to the broth can be conducted, for example at a pH of about 2.0. In addition, current processing below the lactic acid fraction can be conducted after a previous step of removing lactate from the broth or mixture, if desired. It is also noted that the techniques presented in the Figures can be applied both in continuous processes and batch processes, as desired. As a result, the techniques indicated in the figures are well adapted for commercial implementation.

A. Figure 1; Extraction of lactic acid, lactate salt and recycled nutrients, (optional) regenerated extract, concentrated lactic acid by distillation. Fig. 1 is a schematic representation of a preferred lactic acid recovery process, in which the lactic acid solution is formed by fermentation. Referring to Fig. 1, the fermenter is usually indicated as 1. Via line 2, the fermentation broth is removed through fermenter 1. The fermentation broth is passed through a filter unit 3, with removed solids (e.g., cellular material) shown taken from line 4 and the clarified or filtered liquid is transferred to an extraction process or extraction unit. 6. The filter unit 3 may comprise a simple physical filter, or may include adsorbent materials, such as activated carbon and / or physical ion exchange medium. Preferably, an approach is chosen so that sterilization of the material before recycling is not required. (Of course, the cells, in many cases, can be directed back to fermentor 1, if desired.). More specifically, the liquid in line 5, comprising an aqueous solution of lactic acid and lactate salt, is directed to an extraction unit 6, such as a stirred column, a perforated plate column or a series of mixers settlers. Two or more such units can be used in series, for a multi-step extraction process. The extractor is fed to the system via line 7, with the lactic acid carrying the extractor phase removed via line 8. The raffinate or residual aqueous phase (suppressed lactic acid), containing the lactate salt and any residual nutrient It is removed via line 10 and directed towards a pretreatment system 1 1, to return to the fermentation broth via line 1 2. The pretreatment system 1 1 can be, for example, a retro wash of solvent to remove any level of residual extract, which can be toxic to organisms in the fermenter, or to remove some other unwanted impurity to avoid the formation of impurities in the fermenter due to recycling. The extractor phase, containing the lactic acid, is directed in a distillation system 1 3. The lactic acid products are distilled via line 1 5. The resulting lactic acid product comprises lactic acid and condensation products (oligomers) of lactic acid (depending on the degree of concentration in the distillation system 1 3). It can be used to form lactide and polymer. The extraction solvent is removed and recycled in the extraction.

Then, the scan of Fig. 1 is particularly well suited for use in systems, wherein the lactic acid is removed from the fermentation broth by extraction, and the recovery of the lactic acid results from distillation of the lactic acid away from the extractor. The scheme of Fig. 1 is suitable, for example, for the application wherein the extractor phase comprises a mixture of tertiary amines and alkanes, such as, Alamina 336 and kerosene. The specific preferred conditions for the extraction would involve contacting the aqueous lactic acid solution and extraction solvent at a temperature between 30 ° C and 50 ° C. The proportion of aqueous to organic phase is preferably between 0.1 and 10, and more preferably between 0.2 and 5. Of course, a process similar to Fig. 1 could be practiced in a variety of solutions other than simply fermentation broths. The material in line 5 could be modified broth (acidified, for example) or could be from a source other than fermentation. Alternatively, if the extraction solvent is more volatile than lactic acid, the same flow diagram as in Fig. 1 . However, the extraction solvent would be distilled from the lactic acid products.

B. Figure 2; Crystallization of lactate from broth, recycling of lactate salt to fermentation; recovery of acid from the broth suppressed from salt.

Now we turn our attention to the scheme of Fig. 2. In this alternative approach, a lactate salt of relatively low solubility, eg, calcium lactate, is precipitated from the mixture, and the lactic acid is recovered from the mother liquid. Referring to Fig. 2, the fermenter is indicated generally as 31. The fermentation broth is shown removed via line 32 to be directed through the filter or clarifier 33. The clarifier solids are removed via line 34. The clarified broth is then directed, via line 35, to the unit of evaporation 36. (Of course, the material on line 35 could be modified broth or a mixture of some other source.) During evaporation, the concentration and crystallization of the lactate salt contained within the broth will occur. Evaporation water is entrained via line 37. The physical separation of the mother liquor from the crystallized material is shown by directing the result of evaporation through the filter 40. The solids recovered from the filter 40, comprising crystallized lactate, are directed via the line 45 to a purification unit and eventually (if desired) are recycled to the fermentation via line 46 (with optional purification at station 47 if desired). Of course, they can be alternatively directed to another process (line 48). A combination of the two may be preferred in some circumstances. The mother liquor of the filtration is directed via line 49 to the recovery of lactic acid. The step of recovery of lactic acid can be any of the variety of steps characterized above. In general, the scheme of Fig. 2 will be particularly useful when the lactate salt is calcium lactate due to low solubility of calcium lactate in aqueous solutions. It is noted that the recovered calcium lactate will make an excellent buffer or pH adjuster for the fermentation broth.

C. Fig. 3; Extraction of lactic acid, formation of oligomer and lactide in extraction solvent Now we turn our attention to the scheme of Fig. 3. This approach will be particularly useful when the attempt is to produce directly, as a "lactic product", lactide and / or oligomer of lactic acid, with previous isolation of lactic acid and, preferably, without a back-extraction step. Referring to Fig. 3, the fermenter is generally indicated at 60. The fermentation broth is removed to fermenter 60 via line 61, for the direction towards the clarifier filtration unit 62. The solids of the filtration unit are removed via line 64. The extraction unit is indicated, generally at 65, and may comprise more than one extraction stage. Generally, these steps can be as described above in connection with Fig. 1. Alternatively, the salt is separated first as in Figure 2 and the mother liquor is extracted. The extractor phase is removed via line 66 and directed towards unit 67 for oligomer formation. The unit 67 may comprise, for example, a multi-stage evaporation unit, from which water and other volatiles are conducted away via line 68 as the extractor is concentrated and the lactic acid is condensed to form an oligomer.

The resulting lactic product phase (oligomer) is removed from the multi-stage evaporator via line 69 and is directed to reactor stage 70. In reactor stage 70, catalysts can be added, for example, via line 75, to facilitate the formation of lactide. The lactide formation step, indicated generally at 76, results in the generation of crude lactide, entrained via line 77, and the bottoms of the reactor are removed via line 78, for catalyst recovery or other treatment 79. The catalyst, of course, can be recycled, if desired, via line 75. The extractor solvent is removed from the lactide formation phase via line 80, to be recycled towards extraction. Still referring to Fig. 3, in the extraction step, the aqueous phase is removed via line 81 for optional recycling, as desired, back to the fermentation broth. The purification of this, if needed, is indicated in the equipment 82. Such purification may be desired, for example, if the extractor is one that is toxic to the microorganisms of the fermentation broth. Approaches for the removal of such materials were discussed earlier. From a review of the above, and Fig. 3, it should be evident that with the techniques described, the formation of "direct" oligomers can be carried out without a back-extraction step or another that separates the lactic acid from the extraction; and, that the direct beating formation, from the oligomer, can be conducted even if residual extractor is present in the oligomer. In this way, with the techniques presented, highly efficient processes can be developed.

D. Fig. 4; Extraction of lactic acid; distillation of solvents; Formation of lactide by condensation oligomer. Now turn attention to Fig. 4. In Fig. 4, the aqueous lactate and lactic acid feed of a fermentor is shown, usually following the clarification on line 91. The feed (or other mixture) is directed towards an extraction system 92. As with the previous arrangements described, the extraction system 92 of FIG. 4 may comprise multiple steps, each of which may comprise extraction equipment as previously characterized. The aqueous raffinate (suppressed lactic acid) is removed from the system via line 93. The raffinate will include the lactate salt and can be treated for recycling to the fermentation broth via the techniques previously described. The extractor phase is directed to an evaporator 95, via line 96. In the evaporator, the extraction solvent is removed under distillation conditions, usually at low pressure. For example, hexanol and other alkanols with 4 to 7 carbons and methyl isobutyl ketone or other ketones with 5 to 9 carbons can be used to extract lactic acid from an aqueous solution. The alkanols (and / or ketones) can be distilled from the lactic acid at temperatures below about 120 ° C. It is important to keep the temperature low to reduce the condensation reaction with lactic acid and alkanol. The non-volatiles of the evaporator are directed via line 97 to a system 98 for oligomer formation. Within system 98, a condensation reaction of lactic acid is generally facilitated by concentration and removal of water. The oligomer is removed from reactor 98 via line 99. Catalyst is added via line 100 and lactide formation is generated, as indicated in system 1 01. The crude lactide stream is then removed from line reactor 1 05, with the catalyst purge removed via line 1 06. The lactic formation step can be conducted as generally described in US Pat. , 142, 023; 5,247, 058; 5,258,488; and 5,357, 035, incorporated herein by reference.

E. Fig. 5; Extraction of lactic acid; recycling of raffinate; retro-extraction of lactic acid in the second polar phase; formation of oligomer and lactide in the subsequent polar phase Attention is now directed to Fig. 5. In Fig. 5, the fermenter is usually indicated in 1 20. The fermentation broth is removed from the fermenter 120 via line 1 21, to be directed through the clarifying filter 1 22. The solids are removed from the filter via line 123. The phase The aqueous phase, containing lactic acid and lactate salt, is directed to an extract system 1 via line 1 24. (Of course, this phase could also be modified broth or some other mixture.) The aqueous phase (raffinate) is removed via line 125 to be directed through a purifier 1 27, if desired, and eventually recycled to fermenter 120.

The extractor phase is removed from the extractor system 1 25 via line 1 30. The extractor phase is directed to a second extractor step or system 131. (Of course, the same physical extraction equipment can be used for both extractions.) A second polar fluid is directed to the extractor 1 31 via line 1 32. The lactic acid will be extracted preferentially in the second polar fluid, being Removed the original extraction solvent from extractor 125 via line 133, for recycling. The second polar liquid, containing the lactic acid therein, is removed from the extractor system via line 1 34, to be directed towards a system 135 for oligomer formation. This would be conducted as previously described, with condensation occurring as a result of the water being conducted off via line 36. The oligomer is then shown directly to a reactor system 37, to be mixed with a catalyst directed via line 1. 38. The formation of lactide is indicated generally at 139, with crude lactide removed via line 140, and the reactor purge, which contains a catalyst, being directed via line 141, and the recovery of catalyst 142. This liquid The recovered polar can be recycled back to the system as shown by line 1 32.

F. Fig. 6; Extraction of lactic acid; recycling of raffinate; retro-extraction to the second polar phase; lactic acid purified by distillation; Formation of lactide and consecutive oligomer (optional) Attention is now directed to Fig. 6. A fermentor is usually indicated at 1 60. The fermentation broth containing lactic acid and lactate salt is removed via line 161 and directed to filter 1 62. The solids are removed via line 1 63. The mixture of lactate salt and lactic acid is directed to an extractor system via line 166. (Of course, this mixture could also be modified broth or a different mixture to the fermentation broth.) The extractor is fed via the line 1 67, containing the lactate salt, removed via line 168. The residual lactate phase is then directed for purification of the broth, if desired, to the purifier 170 and eventually for recycling to the fermenter 160. The second l Polar fluid is directed to the retro-extractor system 1 75 via line 1 76, with the original extraction solvent (from the extraction at 165) drawn via line 178 for recycling to the first extraction system. n1 65; and, with the second polar liquid, containing lactic acid, removed via line 1 80 and directed in distillation system 181. Within the distillation system 181, either the lactic acid will be distilled from the second polar liquid, or the second polar liquid will be distilled from the lactic acid, depending on the relative volatilities. The separated lactic acid is directed via line 182 to the oligomer formation downstream in 1 83, with the addition of a catalyst in 1 84 and lactic formation in 1 85. The crude lactide is removed via line 1 86, with the purge of lactide formation shown at 1 87. In line 1 90, water is expelled during oligomer formation. The second polar fluid is removed from distillation step 1 81 via line 1 91. The second polar fluid, of course, can be recycled to the second extraction system 1 75.

G. Fig. 7; adsorption of lactic acid; liquid levigation; Lactide formation and optional oligomer Attention is now directed to Fig. 7. A feed of fermentation broth (or other mixture of lactic acid / lactate salt) is shown on line 200. This feed is directed to a system 201 containing solid adsorbent lactic acid. In this system, the aqueous feed comes into contact with the solid adsorbent, with the suppressed aqueous phase removed via line 202. For this system, the solid adsorbent would be an adsorbent preferentially adsorbing lactic acid versus lactate. Weak anion exchangers would be preferred for this, as was characterized above. The solid adsorbent is removed from the contact passage or system 201, via line 203. The solid adsorbent is treated with an egvigating liquid introduced via line 205. The levigating liquid would remove the lactic acid from the solid adsorbent. The levigating liquid is removed via line 206. Of course, the levigating liquid can be directed to downstream oligomer formation steps and / or lactide formation steps (as indicated), or other processing for lactic acid isolation recovered, as desired. After the levigation step, the solid adsorbent can be prepared in an appropriate manner to be used (recycled route) still in additional adsorption steps.

H. Fig. 8; Extraction of lactic acid; oligomer and lactide formation in extraction solvents; purification of phase division lactide. Attention is now directed to Fig. 8. Feeding fermentation broth (or other mixture), containing lactic acid and lactate, is directed to an extraction system 221 via line 220. The aqueous phase (raffinate) containing the The lactate salt is removed via line 222. The extractor phase, containing the extracted lactic acid, is removed via line 223. Then it is directed to processing for oligomer formation in 225 with addition of 225a catalyst, and eventually formation of lactide in 226, using techniques previously described. The lactide is removed from the lactide formation step via line 227 and is directed towards a phase division of lactide / solvent. This could be, for example, a system in which the reaction mixture is cooled to a temperature between about 70 ° C and 150 ° C, causing the lactide to phase-divide spontaneously with the extraction solvent. At this point, the extraction solvent is removed from the lactide and recycled via line 230. During the formation of the oligomer, the water resulting from the condensation is removed via line 231. The oligomer can then be directed to lactide formation. In the present, such process will sometimes be referred to as "direct" formation of lactide from the non-aqueous extractor phase, because no retro-extraction step for the lactic acid was involved from the extractor phase. Instead, the lactic acid was condensed and then reacted to lactide. Such "direct" training can be practiced with a variety of the approaches described herein.

I. Fig. 9; Extraction of lactic acid; formation of oligomer and extraction solvents; purification of oligomer by phase separation; Lactide formation from oligomer.

Attention is now directed to the scheme of Fig. 9. The feed of lactic acid solution / lactate solution from a fermentor or other source is shown on line 250 which is directed to an extractor unit 251. The aqueous raffinate, containing the lactate salt, is removed from the extractor unit 251 via line 252. The extraction solvent containing lactic acid therein is removed via line 253 and is directed to an oligomer formation step as it was previously described at 254. Water is expelled from the oligomer formation via line 255, with the oligomer directed to the next processing via line 256. In the phase division / extraction unit 270, the oligomer mixture of Lactic acid and extraction solvent is cooled to approximately 0 ° C to 60 ° C. With relatively non-polar extraction solvents, the lactic acid oligomers will phase-divide spontaneously and nothing is needed to be added via line 282. The relatively polar extraction solvent may need to have a phase division compound, such as those described. in Example 1 8, added to generate two phases. The phase rich in oligomer-lactic acid is removed via 271 and treated by the addition of catalyst in 275 for the formation of lactide in reactor 276. The crude lactide stream is removed via line 277, with the pure reactor, containing the catalyst, removed in line 278. In line 280, the phase division compound, if any, is removed from the extraction solvent via distillation or ion exchange in 281. The regenerated extraction solvent is recycled back to the extractor 251.

J. Figure 10; Extraction of lactic acid; formation of alkyl lactate ester; purification of lactate ester by distillation Now turn attention to Fig. 1 0. In Fig. 1 0 a fermentor is shown in 300. The fermentation broth is removed from the fermenter 300 at 301, and directed through the filter unit 302. The solids are removed via line 303. The aqueous solution of the filtration unit, or another source (modified or not) containing lactic acid and lactate, is directed to an extractor 305. The extractor is fed via line 306, with the resulting aqueous raffinate, containing the lactate salt therein, directed via the line 31 0 to a purification unit 31 1, if desired, and then recycled as necessary, to the fermenter 300. The extractor, containing the lactic acid, is directed to a condensation reactor 31 via the line 6. From the condensation reactor 31 5, it is drained via line 31 7. The product is then directed to distillation 31 8 for a distillation resulting in solvent separation of the residual lactate product. If the selected extractive solvent contains an appropriate alcohol, the product within the distillation unit 31 8 will comprise an alkyl lactate ester. For example, the extraction solvent could contain ethanol, which easily forms esters with lactic acid. The alkyl lactate ester is purified in distillation unit 31 and is removed via line 31 9. The extraction solvent leaves the distillation system via line 320 to possibly recycle back to extraction unit 305. The alcohol that reacted in the condensation reactor 31 can be replaced via line 321 addition.

IX. Some schemes of usable processes; conditions This section provides some descriptions of hypothetical processes, to indicate how the techniques described above can be applied. A. Direct lactic acid distillation from the extractor A bacterial strain would be used to ferment dextrose to lactic acid at 45 ° C in a batch mode. The fermentation medium would include dextrose, maize infusion water and other salts for the efficient productivity of lactic acid. At the end of the batch, the final pH would be 3.9 with a concentration of lactate material (lactide acid + dissociated salt) of 80 g per I of broth. This provides approximately 38 g / l of non-dissociated Jctido acid in the broth. Preferably, a bacterium is used that produces L-lactate in a chiral purity of at least 90%. The broth would be filtered to remove cell mass and other insolubles and it would be put in contact with an extraction solvent in a series of mixing settlers at 20 ° C up to 30 ° C. The extraction solvent would be 50% by weight of Alamina 336, 40% of dodecanol, and 1.0% of IsoPar K. IsoParK is a mixture of alkanes. The weight ratio of aqueous phase to extract would be 1.2. The aqueous extract and raffinate would settle and they would separate carefully to avoid entrainment. The raffinate would be sent to a holding tank to be used for pH control in the next batch. The extract would be sent to a falling film evaporator at pressure of 1.33 x 103 Pa (10 mm Hg) and 175 ° C. The lactic acid would be evaporated and then condensed to obtain a concentrated lactic acid solution with a small amount of residual solvent. The liquid lactic acid stream would be sent to a distillation column with a forced circulation re-kettle with a bottom at 1 50 ° C and 2.67 x 1 04 Pa (200 mm Hg) pressure to remove water. The average molecular weight of the oligomer leaving this distillation column would be about 500 g per mole. The catalyst FASCAT 91 02 would be added to the oligomer stream and the mixture would be recirculated through a falling film evaporator at 190 ° C and 1.33 x 1 03 Pa pressure (10 mm Hg). A raw lactide stream would be obtained from the vapor phase of the falling film evaporator. Approximately 5% of the material would be purged from the lactide reactor as oligomers of lactic acid with an average molecular weight greater than 1.500 g per mole.

B. Retro-aqueous extract, removal of water to make prepolymer, and then lactide. A bacterial strain would be used to ferment dextrose to lactic acid at 45 ° C in a batch mode. The fermentation medium would include dextrose, maize infusion water and other salts for the efficient productivity of lactic acid. At the end of the batch, the final pH is 3.9 with a concentration of lactate material of 80 g per liter of broth. This provides approximately 38 g / l of undissociated lactic acid in the broth. Preferably, a bacterium is used that produces L-lactic acid with a chiral purity of at least 90%. The broth would be filtered to remove the cell mass and other insolubles and to come in contact with an extraction solvent in a rotary disk contactor at 20 ° C up to 40 ° C. Tributyl phosphate would be the extraction solvent, and the aqueous ratio to extract would be 1: 3. The extract is retro-extracted in water in a packed bed extraction column at 80 ° C up to 1 00 ° C. The column would be pressurized to 1.03 x 1 05 Pa by nitrogen. The aqueous ratio to extract in the retro-extraction would be 1: 2. The resulting aqueous phase would be about 1.9% by weight of lactic acid. This aqueous stream would be sent to a triple effect evaporation system, which removes water so that the concentration of lactic acid increases to more than 88% by weight, more preferably to more than 92% by weight and most preferably to more than 95% by weight. At this point, the lactic acid oligomers will have been formed and the lactide can be made from this stream as described above.

C. Crystallization of calcium lactate with recovery of lactic acid A bacterial strain would be used to ferment dextrose to lactic acid at 48 ° C in a continuous two-step fermentation, providing 1000 kg per hour of fermentation broth. The fermentation medium would have dextrose, maize infusion water and other salts for an efficient productivity of lactic acid. The fermentation broth would have a pH of 3.86 and a total lactate anion concentration of lactate material of 90 grams per kilogram of broth. This would give approximately 45 g / kg of free lactic acid in the broth and 55 g / kg of calcium lactate. Preferably, a system is used that provides L-lactic acid at a chiral purity of at least 90%. The broth would be filtered to remove cell mass and other insolubles. The clarified broth would be sent to an evaporator which runs at atmospheric pressure. Approximately 700 kg / h of water is distilled from the broth. The broth would be cooled in a forced circulation cooling crystallizer that crystallizes the calcium lactate out of solution at 25 ° C. Ethanol would be added to the crystallizer at a rate of 85.1 kg / h to decrease the solubility of calcium lactate to 3.0% by weight of calcium lactate. Solid calcium lactate would be recovered via filtration at a rate of 44.8 kg / h, with 10.2 kg / h of calcium lactate remaining in the mother liquors. The solid salt would be recycled back to the fermenter for pH control by adding calcium carbonate as needed. The suppressed salt broth would be contacted with 350 kg / h of extractor solvent consisting of 20% by weight of urea and 80% of trioctylamine in a series of centrifugal contactors. The lactic acid and ethanol will be distributed between the two liquid phases, so that the organic layer contains about 10% by weight of lactic acid and about 5% by weight of ethanol. The aqueous stream would have ethanol, residual lactic acid, calcium lactate and other broth components. The ethanol in this stream would be recovered via distillation or other technology for recycling, while the remaining aqueous stream would go to animal feed or another system. This aqueous stream with ethanol could be suitable feed for a larger ethanol plant.

The extract with lactic acid and ethanol could be processed in a variety of ways to make necessary lactic acid products. The stream could be subjected to conditions to make ethyl lactate, which would then be distilled away from the remaining extractant solvent. The lactic acid could be back-extracted in ethanol at an elevated temperature, and the ethyl lactate could be made in the retro-extraction phase. Ethyl lactate made by these methods could be sold or used to make lactide. Of course, the ethanol could be separated from the lactic acid / extraction solvent phase and the lactic acid could be processed by any of the methods described in this application to make lactic acid products that could be sold or used in the manufacturing of PLA.

D. Direct condensation in the extractor phase. A bacterial strain is used to ferment dextrose to lactic acid at 40 ° C in a batch mode using calcium carbonate as a neutralizing agent. At the end of the batch, the final pH value is approximately 5.7 with a total lactate material concentration of 120 g / l. The broth is filtered to remove cell mass and other insolubles. The concentration of free lactic acid is less than 2 g / l so that a strong acid, either sulfuric acid or phosphoric acid, was added to decrease the pH value to about 2.0. The concentration of free lactic acid is now approximately 1 1 8 g / l. Calcium sulfate or calcium phosphate will form and crystallize out of solution. The solution will be filtered to remove the calcium salt.

The acidified and clarified fermentation broth is placed in countercurrent contact with Alamina 336 in a series of centrifugal contactors at about 40 ° C. Alam ina 336 can be pre-distilled under similar conditions to lactide formation conditions if necessary, to remove any impurity that may be volatile. The ratio of aqueous to organic phase is 3: 1 and the concentration of lactic acid in the extractor phase is 21% by weight and is a simple phase. The water refining can be recycled to the extraction if necessary. The extractor phase is then taken to an evaporator at atmospheric pressure and 1 30 ° C, where the water evaporates. A second evaporator at 6.67 x 1 04 Pa (50 mm Hg) and 1 60 ° C also evaporates water and conducts the condensation of lactic acid to oligomers of lactic acid. The average molecular weight of the oligomers at this point is from about 600 to 800. After this step, the reaction mixture is cooled to about 60 ° C, where the mixture spontaneously is divided into two phases.; an almost pure Alamina 336 phase and an oligomer phase of lactic acid-Alamine 336. These phases are physically separated using a normal settler and the almost pure Alamina 336 phase is recycled back to the extractor. The tin octane (I I) is added to the oligomer phase of lactic acid-Alam ina 336 at about 0.1 to 0.5 wt% tin. The mixture is recirculated through a sliding film evaporator at approximately 1 80 ° C and 5 mmHg pressure. A stream of crude lactide is obtained in the vapor phase.

A purge is taken from the evaporated recirculation circuit and processed to separate tin, possibly via ion exchange. The lactic acid-Alamine 336 oligomer is recycled or can be separated into a rich stream of lactic acid oligomer and a current rich in Alamina 336 via acid or base displacement as shown in example 1 8 and 19. These two streams can then recycled back into the process.

X. Experimental Example 1 600 ml of Alamine 336 subjected to caustic scrubbing, 800 ml of 1% wt.% Aqueous lactic acid solution and 1 00 ml of 50% by weight aqueous lactic acid solution were added to a funnel of separation and mixed at room temperature. The phases were allowed to settle during the night; the phases were divided and the upper organic phase was centrifuged to remove the entrained aqueous phase. It was determined that the concentration of lactic acid in the organic phase was 9.75% by weight by titration with a sodium hydroxide solution with phenolphthalein as an indicator. 304.6 g of the Alamina 336 and lactic acid solution were added to a four-necked round bottom flask with a stirring shaft, thermocouple, condenser, heating blanket and nitrogen purge. The solution was heated to 200 ° C and atmospheric pressure over 45 minutes. Then it was allowed to cool to approximately 64 ° C. Then it was heated to 200 ° C at a pressure of 8.00 x 1 03 Pa (60mm Hg) for 30 minutes. The flask was held at 200 ° C and 9.33 x 103 Pa (70 mmHg) for 45 minutes. The flask was cooled and the bottom was divided into two phases on cooling. It was determined that the upper phase was virtually all Alamina 336 by gas chromatography. The bottom phase was viscous and consisted of oligomers of lactic acid and a small one of Alamine 336. Aliquid of 85.9 grams of Alamina 336 and lactic acid oligomer solution (approximately 54.8% by weight at average MW of 476) were added to a flask round bottom, 4-neck, 500 ml, with a stirring arrow, high vacuum system, nitrogen purge, condenser, thermocouple and heating blanket. With the solution at 125 ° C, 900 μl of FASCAT 91 02, a butyltin tris-2-ethylhexanoate catalyst from Atochem, was added. The solution was heated at 200 ° C for four hours, and the mixture was held at 200 ° C for 60 minutes. The pressure was kept constant at approximately 1.33 x 1 02 Pa (1 mm Hg) throughout the heating time. The temperature of the condenser medium was maintained at 1110 ° C. The upper material crystallized on cooling. The bottoms of the flask after heating were determined by gas chromatography as visually all of Alamina 336. 1 39 g of material went to the top, virtually all of the oligomer was converted to lactide and distilled on top. Some Alamina 336 was also distilled above due to the high temperature and low pressure. The presence of significant quantities of lactide in the distillate was confirmed by gas chromatography. The lactide obtained had a chiral purity of less than 80%. The chiral purity can be improved by using lower temperatures and using high surface area equipment for the lactic reactor to allow a good transfer of lactide mass out of the reactor. This example shows how the extraction solvent can be used for the formation of lactide and lactic acid oligomer.

Example 2 300 ml of Alamine 336 and 200 ml of a 22% by weight aqueous lactic acid solution were added to a separating funnel. The mixture was stirred and allowed to settle. Three liquid phases were obtained, which is normal for extractions of pure Alamina 336 under these conditions. The lower aqueous phase was discarded. The two higher organic phases were contacted with 100 ml of a 22% aqueous lactic acid solution. The mixture was stirred and allowed to settle overnight. Only two phases were obtained, and the lower aqueous phase was discarded. The upper organic phase was centrifuged to remove any entrained water. The concentration of lactic acid in the organic phase was 1 9.4% by weight as determined by titration. The water content in solution was 4.6% by weight as determined by titration using an automatic Karl Fischer titrant. 143.0 g of this Alamina 336 and lactic acid solution were added to a 500 ml round bottom, 3-neck flask with thermocouple, vacuum, nitrogen purge, condenser and stirring shaft. The pressure was adjusted to 2.67 x 1 03 Pa (20 mm Hg) and the solution was heated from room temperature to 21 0 ° C. Fraction 1 was taken from the vapor phase with the kettle temperature between room temperature and 1 03 ° C. Fraction 2 was taken at the pot temperature between 1 03 ° C and 1 50 ° C. Fraction 3 was taken at the kettle temperature between 1 50 ° C and 169 ° C. Fraction 4 was taken at the kettle temperature between 1 69 ° C and 21 0 ° C. It was determined that the concentration of acid in fractions 1, 2, 3 and 4 per titration was 0.23% by weight, 16.1% by weight, 73.2% by weight and 60.8% by weight, respectively. The bottoms of the kettle weighed 109.1 g and were two liquid phases at room temperature showing that some condensation occurred during the distillation. Fraction 4 was found to be approximately 2% lactide showing additional evidence of condensation. The addition of 23.8 g of octanol caused the two lower phases to become miscible. Single-phase funds were titled to find only 2.2% lactic acid when corrected for octanol. Approximately 60% of the lactic acid was recovered above. This example shows that the distillation of lactic acid from a less volatile extraction solvent is a viable process option.

EXAMPLE 3 200 ml of dimethyl sulfoxide (DMSO) and 200 ml of a previously made Alamin 335 and lactic acid solution with 8.4% by weight lactic acid were added to a 500 ml round bottom flask of 3 ml. necks with a stirring arrow, temperature control, condenser and heating mantilla. The mixture was stirred and heated to 140 ° C and held at 140 ° C for 15 minutes. The phases settled quickly, separated, and were allowed to cool to room temperature. Samples of the bottom DMSO phase showed 1.1% by weight of lactic acid per titration and 0.58% by weight of Alamina 336 by gas chromatography. 40 ml of Iso Par K from Exxon were added to the DMSO phase in a separating funnel. The funnel was stirred at room temperature and the phases were allowed to settle and separate. Samples of the bottom DMSO phase showed 1.1% by weight of lactic acid, 2.7% by weight of water by Karl Fischer titration, and 0.05% by weight of Alamine 336. Then 230.0 g of this solution was placed. of DMSO and lactic acid in a 500 ml, round bottom, 4-neck flask with a stirring arrow, vacuum, condenser, thermocouple and heating mantle. The material was heated to atmospheric pressure at 1 80 ° C, collecting 42.0 g above. The material was allowed to cool. It was determined that the acid concentration in the lower phase was 12.4% by weight, showing some concentration by loss of acidity, assuming no lactic acid was evaporated. The material was then heated from room temperature to 1 1 7 ° C with approximately a pressure of 8.00 x 1 03 Pa (60 mm Hg) for 60 minutes. Another 34.7 g of material was distilled over. This completed the step of forming the lactic acid oligomer. 146.7 g of DMSO solution and lactic acid oligomer remained for the lactide forming portion. 1.53 g of FASCAT 91 02, a butyltin tris-2-ethylhexanoate catalyst, were added. A cold trap of dry ice and nitrogen purge was added and the condenser was changed to ethylene glycol medium at 110 ° C.

The mixture was heated at room temperature to 145 ° C under a pressure of 100 mm Hg for 80 minutes. Only 7.8 g of material remained in the bottom of the flask. The receiver contained 1 1 6.2 g of material. The boiling point of DMSO is close enough to the lactide that a significant amount of DMSO is expected to distill. The presence of lactide in the upper part was confirmed by gas chromatography. This example shows the feasibility of retro-extracting the lactic acid in a polar liquid from the extraction solvent and using the polar fluid as a solvent to make lactic acid and lactide oligomer.

Example 4 Two solutions of lactic acid and Alamine 336 were made by contacting Alamina 336 with various amounts and concentrations of aqueous lactic acid solutions. Mixtures of Alamina 336 with 4.35%) by weight and 1.89% by weight of lactic acid were obtained. 2 μl of solutions of Alamine 336 and lactic acid were contacted separately with the following solvents - dimethyl sulfoxide (DMSO); N. N-dimethyl foramide (DMF); 1,4-dioxane; N-methyl pyrrolidinone (NM P); and 1,3-dioxane. The samples were held at the specified temperature in an oil bath for about 45 to 60 minutes with regular mixing. Mixtures of 1,4-dioxane and 1,3-dioxalan formed a simple liquid phase at temperatures between 20 ° C and about 80 ° C. A similar procedure was used to contact solutions of Alam ina 336 and lactic acid with lactide and tetramethylene sulfone (TMSF). The phases were allowed to settle to a specified temperature and then quickly separated when the bottom phase was removed by pipe. Samples were taken for titration with sodium hydroxide solution with phenophthalein as an indicator, to determine the concentration of lactic acid, and gas chromatography to determine the concentrations of Alamina 336. In all cases, the Alamina 336 phase was the less dense phase or the higher phase. Table 1 reports the concentrations of lactic acid and Alamine 336 in the upper and lower phases. The partition coefficient is calculated by dividing the concentration of lactic acid in the upper phase of Alamina 336 between the concentration of lactic acid in the polar liquid phase of the bottom. The results show that significant amounts of lactic acid are distributed in the polar liquid phase under these conditions. In some of the solvents, a significant amount of Alam ina 336 was co-extracted in the polar liquid phase. Dimethyl sulfoxide seems a favorable solvent for this type of processes due to the good selectivity for lactic acid on Alamine 336. This example shows that lactic acid can be retro-extracted in a polar liquid from the extraction solvent initial with good efficiency. This example supports the feasibility of a process that uses a retro-extraction of lactic acid in a second polar liquid.

Table 1 . Results of retro-extraction of lactic acid in second polar phase Solvent Sample Temp ° C% by weight of acid% by weight of Lactic Coefficient Alamina 336 of partition Top DMSO 140 0.18 72.4 0.19 Bottom 0.93 0.0 Top 140 0.39 66.03 0.06 Bottom 11.78 0.98 Bottom DMF 110 0.16 61.09 0.14 Bottom 1.13 1.17 Top 110 1.49 70.17 0.13 Bottom 11.36 19.23 Superior SPM 90 0.33 65.26 0.29 Bottom 1.13 1.72 Superior 110 2.73 67.15 0.22 Bottom 12.2 19.71 Top TMSF 140 0.88 71.69 0.21 Bottom 4.15 5.28 Top 140 1.25 73.06 0.17 Bottom 7.52 10.09 Top 140 1.50 78.10 0.17 Bottom 8.66 13.56 Top Laps 140 2.44 nd 0.82 Lower 2.982 2 = calculated from a mass balance obal nd = not determined Example 5 Alamina 336 and a solution of aqueous lactic acid were contacted to obtain a lactic acid 26.74% by weight in the phase of Alamina 336 Ten grams of the lactic acid loaded in the Alamina 336 phase was contacted with 5 grams of triethylamine in a separating funnel of 1 25 ml. The flask was stirred for one minute at 24 ° C and the phases allowed to settle. The upper phase contained excess triethylamine of Alamina 336 and virtually no lactic acid, while the lower phase contained 43% lactic acid and triethylamine. The acid concentrations were determined by titration with sodium hydroxide. The retro-extraction was scaled to allow the distillation experiment. Thirty grams of a 43% by weight lactic acid in a triethylamine mixture was added to a 500 ml round bottom, 3-necked flask equipped with a dry ice trap, pressure gauge, condenser, thermocouple and blanket. heating. The triethyl sheet was initially evaporated at 23 ° C and 1.33 x 1 03 Pa (1 0 mm Hg). The temperature increased to 1 20 ° C and the mixture was held at that temperature for 90 minutes. Approximately 69% of triethylamine was evaporated. The chiral purity of the lactic acid did not change significantly after heating. The removal of triethylamine can be dramatically increased in the presence of a solvent. A solution of lactic acid at 21.5% by weight in a mixture of triethylamine and N-methyl-2-pyrrolidinone at 55 ° C and 1.33 x 1 03 Pa (1.0 mm Hg) of pressure was heated for two hours and 48% of the triethylamine was evaporated from the solution. The remaining mixture was heated to 1110 ° C, where it was kept for 80 minutes. At this point, 96% of the triethylamine was evaporated. The chiral purity of the material did not change significantly. This example shows the retro-extraction of the lactic acid from the extraction solvent and then the ability to evaporate the back-extraction solvent to obtain a concentrated lactic acid product.

EXAMPLE 6 An excess of calcium lactate pentahydrate crystals was mixed for 2 hours at 30 ° C with a solution containing 9% lactic acid and no ethanol. The resulting aqueous solution was analyzed for calcium ions to determine the concentration of dissolved calcium lactate. It was found that there was 7.49% calcium lactate. An excess of calcium lactate pentahydrate crystals was mixed for 2 hours at 30 ° C with a solution containing 1.1% of lactic acid and 1.0% of ethanol. The aqueous solution was analyzed by calcium ions to determine the concentration of dissolved calcium lactate. It was found to be 5.1 3% calcium lactate. An excess of calcium pentahydrate crystals was mixed for 2 hours at 30 ° C with a solution containing 1.89% of lactic acid and 24.8% of ethanol. The aqueous solution was analyzed by calcium ions to determine the concentration of dissolved calcium lactate. It was found to be 2.99% calcium lactate.

These solubility measurements show the decrease in calcium lactate concentration as the amount of ethanol in solution increases. In one process, the addition of ethanol to the broth provides an additional driving force for the crystallization of calcium lactate from the broth.

Example 7 An aqueous feed solution containing 25% of sodium lactate and 2.9 mol / kg of lactic acid with hexanol at 80 ° C was drawn upstream. The ratio of aqueous phase to organic phase was 1: 2.3 w / w and the number of steps was 5. The concentrations of lactic acid in the extract and in the raffinate were 1.0 mol / kg and 0.2 mol / kg respectively. The extract was retro-extracted countercurrently with water at 30 ° C. The ratio of aqueous phase to organic was 1: 1.6 p / p and the number of stages was 6. The concentration of lactic acid in the regenerated extractor was less than 0.1 mol / kg and was in the solution of the resulting aqueous product it was approximately 1.6 milligrams / kg. This example shows the efficient recovery of lactic acid from a stream of lactic acid and lactate salt using extraction and back extraction in water with an alcohol solvent.

Example 8 An aqueous feed solution containing 25% of sodium lactate and 3.0 mol / kg of lactic acid with TBP at 30 ° C was drawn upstream. The proportion of aqueous to organic phase was 1: 2.3 w / w and the number of stages was 5. The concentrations of lactic acid in the extract and in the raffinate were 1 .3 mol / kg and 0.2 mol / kg, respectively. The extract was retro-extracted countercurrently with water at 85 ° C. The ratio of aqueous phase to organic was 1: 1 .7 p / p and the number of stages was 6. The concentration of lactic acid in the regenerated extractor was approximately 0.03 mol / kg and that was in the solution of the resulting aqueous product was approximately 2.1 mol / kg. This example shows the efficient recovery of lactic acid from a stream of lactic acid and lactate salt using extraction and retro-extraction in water with an oxygenated phosphorous compound.

Example 9 An aqueous feed solution containing 0.5 mol / kg of lactic acid and 0.5 mol / kg of sodium lactate with Alamina 336 at 25 ° C was drawn upstream. The proportion of aqueous phase to organic was 5.6: 1 w / w and the number of stages was 4. The concentrations of lactic acid in the extract and in the raffinate were 2.3 mol / kg and 0.1 mol / kg, respectively. The extract was back-extracted countercurrently with water at 1 60 ° C. The ratio of aqueous to organic phase was 1: 1 .2 w / w and the number of steps was 4. The concentration of lactic acid in the regenerated extractor was approximately 0.1 mol / kg and that in the resulting aqueous product solution was approximately 2.7. mol / kg. This example shows the efficient recovery of lactic acid from a stream of lactic acid and lactate salt using extraction and back extraction in water with a trialkylamine. Compared with the initial solution, the aqueous back-extraction product has a higher concentration of lactic acid.

Example 10 An aqueous feed solution containing 4 mol / kg of lactic acid was extracted countercurrently with hexanol at 80 ° C. The proportion of aqueous phase to organic was 1: 2.3 w / w and the number of stages was 6. The concentrations of lactic acid in the extract and in the raffinate were 1.8 mol / kg and 0.2 mol / kg respectively. The extract was back-extracted countercurrently with water at 30 ° C. The proportion of aqueous phase to organic was 1: 1.5 p / p and the number of stages was 7. The concentration of lactic acid in the regenerated extractor was less than 0. 1 mol / kg and it was in the aqueous product solution resulting was approximately 2.7 mol / kg. This example shows the recovery of lactic acid from an aqueous solution with an alcohol solvent.

Example 1 1 An aqueous feed solution containing 4.5 mol / kg lactic acid was extracted countercurrently with tri-butyl phosphate (TBP) at 25 ° C. The proportion of aqueous phase to organic was 1: 2.3 w / w and the number of stages was 6. The concentrations of lactic acid in the extract and in the raffinate were 2.0 mol / kg and 0.2 mol / kg, respectively. The extract was back-extracted counter-current with water at 85 ° C. The ratio of aqueous phase to organic was 1: 1 .7 p / p and the number of stages was 8. The concentration of lactic acid in the regenerated extractor was approximately 0.03 mol / kg and that in the resulting aqueous product solution was approximately 3.5 mol / kg. This example shows the recovery of lactic acid from an aqueous solution with an oxygenated phosphorous compound.

Example 12 An aqueous feed solution containing 0.5 mol / kg of lactic acid was extracted countercurrent with Alamina 336 at 25 ° C. The proportion of aqueous phase to organic was 5.6: 1 w / w and the number of stages was 4. The concentrations of lactic acid in the extract and in the raffinate were 2.3 mol / kg and 0. 1 mol / kg, respectively. The extract was back-extracted countercurrently with water at 160 ° C. The ratio of aqueous phase to organic was 1: 1 .2 w / w and the number of stages was 4. The concentration of lactic acid in the regenerated extractor was approximately 0.1 mol / kg and that in the resulting aqueous product solution was approximately 2.7. mol / kg. This example shows the efficient recovery of lactic acid from an aqueous solution using a trialkylamine. Compared with the initial solution, the aqueous back-extraction product has a higher lactic acid concentration.

Example 13 An aqueous feed solution containing 2 mol / kg of lactic acid was extracted by Alam ina 336 at 25 ° C in a simple step.

The ratio of aqueous phase to organic was 1: 1 w / w. Three phases were formed: an aqueous bottom phase and two organic phases. The concentrations of lactic acid in the combined organic extract and in the raffinate were 2.3 mol / kg and 0.4 mol / kg, respectively. The combined organic extract was retro-extracted countercurrently with water at 1 60 ° C. The ratio of aqueous phase to organic was 1: 1 .2 w / w and the number of stages was 4. The concentration of lactic acid in the regenerated extractor was approximately 0. 1 mol / kg and that in the solution of the resulting aqueous product was approximately 2.7 mol / kg. This example shows that a significant amount of lactic acid can be extracted in a simple step using a trialkylamine solvent. This system has the advantage of the three-phase system that is formed when an extraction solvent has a high amount of Alamina 336 and other non-polar compounds, such as kerosene, and a minimum amount of oxygenated solvents, such as hexanol or methyl isobutyl ketone.

Example 14 An organic phase containing 3.1 3 mol of lactic acid / kg in Alamina 336 was added to a beaker. The beaker flask was heated on a hot plate at 50-160 ° C and atmospheric pressure and kept under these conditions for 7 hours. A sample of the contents was titrated with 0.1 N sodium hydroxide and found to contain 0.639 mol of acid / kg. The drop in acid concentration in the organic phase is a result of converting lactic acid molecules to lactic acid oligomers. The conversion efficiency of lactic acid to the oligomer form was 79%. This example shows the ability to make lactic acid oligomers at atmospheric pressure in a trialkylamine solvent.

Example 15 1 7.7 g of a solution containing 1.92 mol / kg of Alamina 336, 1.98 mol / kg of lactic acid and one drop of an antioxidant, were heated in a beaker, placed in an oil bath, at about 1 35-1 50 ° C and kept at that temperature for 42 hours. Nitrogen was bubbled through the solution during the heating period. The beaker was connected to a distillation column, which was connected to a trap filled with water. At the end of the experiment, while it was still at an elevated temperature there was only one phase in the beaker. After cooling, two organic phases were observed; 1.59 g of a viscous bottom phase and 1 1 .4 g of an upper phase. The concentrations of amine and protons in the bottom phase were determined by titration with a solution of 0. 1 N hydrochloric acid and 0.1 N sodium hydroxide solution, respectively. It contained 1 mol / kg of amine and 0.957 mol / kg of protons. As the heavy organic phase contains only the amine and a lactic acid oligomer, these figures make it possible to calculate the molecular weight of the oligomer. It was found to be approximately 635, equivalent to that of an oligomer consisting of 8 lactic acid monomers. The I R spectra support the conclusions based on this calculation. The concentration of the amine and protons was determined in the upper phase as 2.54 mol / kg and 0.02 mol / kg, respectively. This example shows that the extracted lactic acid can be converted into an oligomer of lactic acid while in the extractor. In addition, it shows the ability of the Alamina 336 system and lactic acid oligomer to spontaneously divide into phases on the cooling of the reaction mixture. Analysis of these phases shows that the significantly higher upper phase is virtually all Alamina 336 and that the lower lower phase is the oligomer product and Alamina 336.

Example 16 An organic phase was prepared, containing 1.63 mole of lactic acid in Alamina 336. It was heated on a hot plate, in an open glass vessel, at 140-150 ° C and kept at that temperature for 6 hours . Then, a drop of tin 2-ethylhexanoate was added, the solution was heated to 1 80 ° C and kept at that temperature for 3.5 hours. Part of the distilled vapors during the heating was condensed in a cold glass (held above the heated container). The condensate was washed from the glass with chloroform and the NMR spectrum was taken from the chloroform solution. The NMR spectrum confirmed that chloroform contained lactide as the main lactic acid product in chloroform. A drop of tin 2-ethylhexanoate catalyst was added to a solution containing 1.02 mol / kg of lactic acid oligomer of DP4-5 in Alamina 336. This mixture was heated on a hot plate while it was in a glass of precipitates connected to a trap, at 1 70-1 90 ° C and kept at that temperature for 5 hours. The condensate was flushed from the trap with chloroform and the I R spectrum was taken from the chloroform solution. Based on these spectra, u can not conclude that the condensate collected in the trap contains a significant amount of lactide. These two examples show that lactide can be made from lactic acid in the presence of a trialkylamine. In this case, the production of lactide was at atmospheric pressure.

Example 17 1 6.2 g of octanol and 0.222 g of phosphoric acid were mixed for 1 5 minutes at 25 ° C with 0.936 g of solution containing 1. 1 9 mmol of Alamine 336 and 0.74 mmol of lactic acid oligomer (DP8-9). After settling for 4 hours in a refrigerator, two phases were observed. The acid-base titers show that 85% of the total amine was present in the light phase and a significant amount of the oligomer was present in the heavy phase. 1 6 g of isopropanol and 0.468 of phosphoric acid were mixed for 1 5 minutes at 25 ° C with 1. 1 56 g of solution containing 1.66 mmol of Alam ina 336 and 1.5 mmol of lactic acid oligomer (DP8-9). After settling for 4 hours in a refrigerator, two phases were observed. The acid-base titers show that 73% of the total amine was present in the light phase and a significant amount of the oligomer was present in the heavy phase. 3.06 g of isopropanol and 0.324 g of acetic acid were mixed for 1 5 min at 25 ° C with 0.733 g of solution containing 1.04 mmol of Alamine 336 and 0.57 mmol of lactic acid oligomer (DP8-9). After settling for 4 hours in a refrigerator, two phases were observed. The acid-base titers show that 82% of the total amine was present in the light phase and a significant amount of the oligomer was present in the heavy phase. These three examples show that the addition of an alcohol solvent and either a relatively strong acid (phosphoric acid) or a weak acid (acetic acid) can separate trialkylamines from lactic acid oligomers via extraction or division of phases.

Example 18 4.66 g of a solution containing 1.69 mol / kg of lactic acid oligomer (DP-5) and 1 was mixed. 1 6 mol / kg of Alam ina 336 were mixed with 2618 g of hexane. Then, 0.585 g of a concentrated ammonia solution (1 2.4 mmol of ammonia) was added. After mixing and settling, two phases were observed. The liquid phase was titrated by HCl and by NaOH to determine the concentrations of amine + ammonia and protons (as lactic acid oligomer), respectively. 0.19 mmol of prepolymer and 3.27 mmol of amine + ammonia were found in said light phase. Based on the fact that the solubility of ammonia in hexane is negligible, the concentration of base in the light phase represents a separation of about 60% of the oligomer of the amine in a single step.

This example shows that the addition of a base such as ammonia can force the formation of a second phase and lactic acid oligomer of trialkylamine extract.

Example 19 1.79 g of a phosphoric acid solution of 8.7 mol / kg was added to a mixture of 5.48 g of calcium lactate pentahydrate and 20.25 g of water. The solution was mixed for 2.5 hours at 85 ° C. A solid phase and an aqueous liquid phase were found. The solid phase was filtered, washed with water and sampled. The solid was found to be virtually free of lactic material and contained 80.2% and 77.0% of total calcium and phosphate, respectively. The remaining aqueous solution contains 77% of the lactate material in free acid form and 23% of the lactate material as a calcium lactate salt. 2.84 g of phosphoric acid solution of 8.7 mol / kg were added and 16. 46 g of butanol to a mixture of 7.83 g of calcium lactate pentahydrate and 20.6 g of water. The mixture was amalgamated for 30 m inutes at 20 ° C. Three phases were found, one solid and two lines. It was found that the solid was virtually free of lactate material and contained 68.7% and 72.3% of the total calcium and phosphate, respectively. The bottom aqueous phase contained 66.9% of the lactate material, 31% of the total phosphate and 26.7% of the total calcium. The organic phase contained 33% of the total lactate material. In this manner, an amount of lactate salt was acidified simultaneously and extracted into the organic phase.

These examples show the ability to acidify with phosphoric acid and form calcium phosphate salt. The acidulated lactic acid can then be extracted with an appropriate solvent or isolated from the aqueous phase by other means.

Example 20 A fermentation broth with a final pH value of 3.87 and a total lactate material concentration of 79 g / 1 was obtained, as determined by high performance liquid chromatography. A variety of different solvents were contacted with the broth to determine the recovery of lactic acid in a simple step. The amount of free lactic acid in each phase was determined by titration with an aqueous sodium hydroxide solution. For the 1 00% Alamina 336 solvent, two organic phases were isolated and titrated, and both values are reported. The partition coefficient is not reported for this system.

A336 = Alamine 336 OctOH = 1 octanol IPK = IsoPar K DodecOH = 1 dodecanol This example shows that a variety of different extraction solvents provide partition coefficient values that are suitable for industrial processes.

Example 21 A fermentation broth with a final pH of 3.87 and a total lactate material concentration of 79 g / l was obtained as determined by high performance liquid chromatography. The broth was contacted three times with fresh extraction solvent consisting of 89% or by weight of Alamina 336, 9% by weight of dodecanol, and 2% by weight of IsoPar K with an aqueous to organic phase ratio of 3.0. The concentration of free lactic acid was determined in each phase by titration with an aqueous sodium hydroxide solution.

This example shows how the partition coefficient decreases, which is a measure of the efficiency of the extraction, as more lactic acid is extracted from the broth, that is, as the pH of the remaining broth increases.

Claims (23)

1. REVIVAL DICTION EN 1 . A process for the production of lactic acid products from a mixture containing free lactic acid and dissolved lactate salt; said method including the steps of: (a) incubating a culture of acid-tolerant lactate-producing microorganisms in nutrient medium at an average incubation pH of no more than 4.8 to produce an aqueous mixture, which includes at least 40 g / l of lactate material comprising lactic acid, lactate salt or a mixture thereof; and (b) separating the aqueous mixture from acid-tolerant lactate-producing microorganisms; and (c) preferentially separating lactic acid from the mixture to form a stream containing lactic acid.
2. 2. A process according to claim 1, wherein the step of preferentially separating lactic acid from the aqueous mixture is selected from: (a) extracting the lactic acid from the mixture in a first non-aqueous phase; (b) adsorbing the lactic acid on an adsorbent solid; (c) distilling the lactic acid from the aqueous mixture; and (d) passing the lactic acid through a membrane.
3. 3. A process according to claim 2, wherein the step of extracting in a first non-aqueous phase includes extracting lactic acid from the mixture in a first non-aqueous phase comprising a tertiary amine.
4. 4. A process according to claim 3, further comprising a step for separating the lactic acid from the non-aqueous phase, enriched with lactic acid, wherein the step of separating the lactic acid from the non-aqueous phase, enriched with lactic acid, is selected from: (a) back-extracting the lactic acid from the non-aqueous phase, enriched with lactic acid in a second non-aqueous phase, which comprises a polar organic solvent and separating the lactic acid from the polar organic solvent; (b) retro-extracting the lactic acid from the non-aqueous phase, enriched with lactic acid, in a second aqueous phase; (c) separating the lactic acid from the non-aqueous phase, enriched with lactic acid, by distillation.
5. 5. A process according to any of claims 1-4, further comprising a step of condensing lactic acid to form lactic acid oligomer.
6. 6. A process according to any of claims 1-5, wherein: (a) the step of preferentially separating lactic acid includes a simueous step of preferentially separating lactate salt in a stream containing lactate salt.
7. 7. A process for the production of lactic acid products from a mixture containing free lactic acid and dissolved lactate salt; said method including the steps of: (a) incubating a culture of acid-tolerant lactate-producing microorganisms in nutrient medium at an average incubation pH of no more than 4.
8. 8 to produce an aqueous mixture, which includes at least 40 g / liter of lactate material comprising lactic acid, lactate salt or a mixture thereof; and (b) separating the aqueous mixture from acid-tolerant lactate-producing microorganisms; and (c) preferentially separating lactate salt from the mixture to form a lactate salt-containing stream. A process according to claim 7, wherein the step of preferentially separating lactate salt is selected from: (a) extracting lactate salt from the mixture in a first non-aqueous phase; (b) adsorbing the lactate anion salt in a solid adsorbent; (c) electrodialysis; and (d) crystallizing lactate salt from the mixture.
9. 9. A process according to claim 8, wherein: (a) the step of crystallizing lactate salt comprises a step of crystallizing calcium lactate. 1.
10. A process according to any of claims 7-9, wherein: (a) the step of preferably separating lactate salt includes a simueous step of preferentially separating lactic acid in a stream containing lactic acid. eleven .
11. A process according to any of claims 1-10, wherein the lactate material includes 40 g / l of L-lactate or 40 g / l of D-lactate.
12. 12. A process according to any of claims 1-11, wherein the lactate material includes 75 g / L of L-lactate or 75 g / L of D-lactate.
13. 13. A process according to any of claims 1 -12, wherein the lactate material has an optical purity of at least 50%.
14. 14. A process according to any of claims 1 -1 3, wherein the lactate material has an optical purity of at least 75%. 5.
15. A process according to any of claims 1-14, further comprising the step of: (a) ~ modifying the aqueous mixture by adding acid thereto prior to the step of separating lactic acid or lactate salt from the mixture . 1 6.
16. A process according to any of claims 1-15, further comprising a step of: (a) adding phosphoric acid to the aqueous mixture to obtain at least one calcium salt of phosphoric acid before the step of separating lactic acid or lactate salt of the mixture. 7.
17. A process according to any of claims 1-16, wherein the step of incubating includes incubating a culture of acid-tolerant, lactate-producing microorganisms selected from lactate-producing bacteria, lactate-producing yeasts and fungi. lactate producers.
18. 18. A process according to any of claims 1-17, wherein the step of incubating includes incubation of a culture of acid-tolerant lactate producing microorganisms, which include recombinant microorganisms.
19. 1 9. A process according to claim 1 7 or 1 8, wherein: (a) the step of incubating comprises incubating a compound, which includes acid-tolerant lactate-producing yeasts.
20. 20. A process according to claims 1 7 or 1 8, wherein: (a) the step of incubating comprises incubating a culture, which includes acid-tolerant, lactate-producing bacteria. twenty-one .
21. A process according to any of claims 1-20, wherein: (a) the step of incubating comprises incubating an acid-tolerant lactate-producing microorganism culture at an average incubation pH of 3.0 to 4.8.
22. 22. A process according to any of claims 1-21, further comprising a step for adding lactate salt to the nutrient medium during the incubation step.
23. 23. A process according to any of claims 1 -22, wherein the step of separating the aqueous mixture from the acid-tolerant lactate-producing microorganism includes a step of filtering the aqueous mixture. SUMMARY The techniques for processing lactic acid / lactate salt mixtures are provided. Preferred mixes for processing are obtained from fermentation broths, preferably from fermentation processes conducted at a pH of 4.8 or less. The techniques generally concern the provision of separate lactic acid and lactate streams from the mixtures. Preferred separation and processing techniques are provided for each of the streams.