RELATED APPLICATIONS

[0001] This application claims the benefit of US Application No. 13/234,856 filed September 16, 2011 (US Publication No. 2011/0272269 published November 10, 2012), which claims the benefit of US Application No. 13/051,443 filed March 18, 2011, which claims the benefit of US Provisional Application Nos. 61/320,063, filed April 1 , 2010 and 61/327,789, filed April 26, 2010, the subject matter of which is hereby incorporated by reference.

TECHNICAL FIELD

[0002] This disclosure relates to processes for the direct production of carboxylic acids, such as succinic acid (SA) or adipic acid (AA), from fermentation broths containing ammonium salts thereof, such as diammonium succinate (DAS), monoammonium succinate (MAS), and/or SA, or monoammonium adipate (MAA), diammonium adipate (DAA), and/or AA.

BACKGROUND

[0003] Certain carbonaceous products of sugar fermentation are seen as replacements for petroleum-derived materials for use as feedstocks for the manufacture of carbon-containing chemicals. Two such products are SA and AA.

[0004] Succinic acid can be produced by microorganisms using fermentable carbon sources such as sugars as starting materials. However, most commercially viable, succinate producing microorganisms described in the literature neutralize the fermentation broth to maintain an appropriate pH for maximum growth, conversion and productivity. Typically, the pH of the fermentation broth is maintained at or near a pH of 7 by introduction of ammonium hydroxide into the broth, thereby converting the succinic acid to DAS.

[0005] Kushiki (Japanese Published Patent Application, Publication No. 2005-

139156) discloses a method of obtaining MAS from an aqueous solution of DAS that could be obtained from a fermentation broth to which an ammonium salt is added as a counter ion. Specifically, MAS is crystallized from an aqueous solution of DAS by adding acetic acid to the solution to adjust the pH of the solution to a value between 4.6 and 6.3, causing impure MAS to crystallize from the solution.

[0006] Masuda (Japanese Unexamined Application Publication P2007-254354, Oct.

4, 2007) describes partial deammoniation of dilute aqueous solutions of "ammonium succinate" of the formula H ₄ NOOCCH CH ₂ COONH ₄ . From the molecular formula disclosed, it can be seen that "ammonium succinate" is diammonium succinate. Masuda removes water and ammonia by heating solutions of the ammonium succinate to yield a solid succinic acid-based composition containing, in addition to ammonium succinate, at least one of monoammonium succinate, succinic acid, monoamide succinate, succinimide, succinamide or ester succinate. Thus, it can be inferred that like Kushiki, Masuda discloses a process that results in production of impure MAS. The processes of both Kushiki and Masuda lead to materials that need to be subjected to various purification regimes to produce high purity MAS.

[0007] It would be desirable to have a process for the direct production of substantially pure S A from a D AS-containing fermentation broth.

SUMMARY

[0008] We provide a process for making SA from a clarified DAS-containing fermentation broth including distilling the broth under super atmospheric pressure at a temperature of >100°C to about 300°C to form an overhead that comprises water and ammonia, and a liquid bottoms that comprises SA and at least about 20 wt% water, cooling and/or evaporating the bottoms to attain a temperature and composition sufficient to cause the bottoms to separate into a liquid portion and a solid portion that is substantially pure SA, and separating the solid portion from the liquid portion.

[0009] We also provide a process for making SA from a clarified DAS-containing fermentation broth including adding an ammonia separating solvent and/or a water azeotroping solvent to the broth, distilling the broth at a temperature and pressure sufficient to form an overhead that comprises water and ammonia, and a liquid bottoms that comprises SA and at least about 20 wt% water, cooling and/or evaporating the bottoms to attain a temperature and composition sufficient to cause the bottoms to separate into a liquid portion and a solid portion that is substantially pure SA, and separating the solid portion from the liquid portion.

[0010] We further provide a process for making SA from a clarified MAS- containing fermentation broth including distilling the broth under super atmospheric pressure at a temperature of >100°C to about 300°C to form an overhead that comprises water and ammonia, and a liquid bottoms that comprises SA and at least about 20 wt% water, cooling and/or evaporating the bottoms to attain a temperature and composition sufficient to cause the bottoms to separate into a liquid portion and a solid portion that is substantially pure SA, and separating the solid portion from the liquid portion. [0011] We further provide a process for making SA from a clarified MAS- containing fermentation broth including adding an ammonia separating solvent and/or a water azeotroping solvent to the broth, distilling the broth at a temperature and pressure sufficient to form an overhead that includes water and ammonia, and a liquid bottoms that comprises SA, and at least about 20 wt% water, cooling and/or evaporating the bottoms to attain a temperature and composition sufficient to cause the bottoms to separate into a liquid portion and a solid portion that is substantially pure SA, and separating the solid portion from the liquid portion.

[0012] We further provide an integrated process for making carboxylic acids, such as SA, from a fermentation broth containing mono or diammonium succinate salts, including reactive distillation whereby ammonium salts and/or amides and/or imides are converted to the coiTesponding carboxylic acids and ammonia, low temperature crystallization, separation of essentially pure carboxylic acids, and recycle of resulting mother liquor containing unreacted ammonium salts and process byproducts. This integrated process allows conversion of byproduct carboxylic acid amides and imides and ammonium salts to the free carboxylic acid. Near total recycle of crystallization mother liquor to the process is provided, which yields near quantitative conversion of the biologically produced ammonium carboxylates to the corresponding mono-, di- and poly carboxylic acids.

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Fig. 1 is a block diagram of the process for making SA from a DAS containing broth.

[0014] Fig. 2 is a graph showing the solubility of SA as a function of temperature in both water and a 20 wt% aqueous MAS solution.

[0015] Fig. 3 is a block diagram of an integrated process for making carboxylic acids from fermentation broth containing ammonium salts thereof.

[0016] Fig. 4 is a block diagram of the process for making SA from a DAS containing broth using a one stage reactive distillation without a lower pressure concentration step.

[0017] Fig. 5 is a block diagram of the process for making SA from a DAS containing broth using a two stage reactive distillation without a lower pressure concentration step.

[0018] Fig. 6 is a block diagram of the process for making SA from a DAS containing broth using a one stage reactive distillation with a lower pressure concentration step. 6

[0019] Fig. 7 is a block diagram of the process for making SA from a DAS containing broth using a two stage reactive distillation with a lower pressure concentration step. DETAILED DESCRIPTION

[0020] It will be appreciated that at least a portion of the following description is intended to refer to representative examples of processes selected for illustration in the drawings and is not intended to define or limit the disclosure, other than in the appended claims.

[0021] Our processes may be appreciated by reference to Fig. 1 , which shows in block diagram form one representative example, 10, of our methods.

[0022] A growth vessel 12, typically an in-place steam sterilizable fermentor, may be used to grow a microbial culture (not shown) that is subsequently utilized for the production of the DAS, MAS, and/or SA -containing fermentation broth. Such growth vessels are known in the art and are not further discussed.

[0023] The microbial culture may comprise microorganisms capable of producing

SA from fermentable carbon sources such as carbohydrate sugars. Representative examples of microorganisms include Escherichia coli (E. coli), Aspergillus niger, Corynebacterium glutamicum (also called Brevibacterium flavum), Enterococcus faecalis, Veillonella parvula, Actinobacilhis succinogenes, Mannheimia succiniciproducens, Anaerobio spirillum succiniciproducens, Paecilomyces Varioti, Saccharomyces cerevisiae, Bacteroides fragilis, Bacteroides ruminicola, Bacteroides amylophilus, Alcaligenes eutrophus, Brevibacterium ammoniagenes, Brevibacterium lactofermentum, Candida brumptii, Candida catenulate, Candida mycoderma, Candida zeylanoides, Candida paludigena, Candida sonorensis, Candida utilis, Candida zeylanoides, Debaryomyces hansenii, Fusarium oxysporum, Humicola lanuginosa, Kloeckera apiculata, Kluyveromyces lactis, Kluyveromyces wickerhamii, Penicillium simplicissimum, Pichia anomala, Pichia besseyi, Pichia media, Pichia guilliermondii, Pichia inositovora, Pichia stipidis, Saccharomyces bayanus, Schizosaccharomyces pombe, Torulopsis Candida, Yarrowia lipolytica, mixtures thereof and the like.

[0024] A preferred microorganism is an E. coli strain deposited at the ATCC under accession number PTA-5132. More preferred is this strain with its three antibiotic resistance genes {cat, amphl, tetA) removed. Removal of the antibiotic resistance genes cat (coding for the resistance to chloramphenicol), and amphl (coding for the resistance to kanamycin) can U 2012/051326 be performed by the so-called "Lambda-red (A-red)" procedure as described in Datsenko KA and Wanner BL., Proc. Natl. Acad. Sci. U S A 2000 Jun 6; 97(12) 6640-5, the subject matter of which is incorporated herein by reference. The tetracycline resistant gene tetA can be removed using the procedure originally described by Bochner et al., J Bacteriol. 1980 August; 143(2): 926-933, the subject matter of which is incorporated herein by reference. Glucose is a preferred fermentable carbon source for this microorganism.

[0025] A fermentable carbon source (e.g., carbohydrates and sugars), optionally a source of nitrogen and complex nutrients (e.g., corn ^' steep liquor), additional media components such as vitamins, salts and other materials that can improve cellular growth and/or product formation, and water may be fed to the growth vessel 12 for growth and sustenance of the microbial culture. Typically, the microbial culture is grown under aerobic conditions provided by sparging an oxygen-rich gas (e.g., air or the like). Typically, an acid (e.g., sulphuric acid or the like) and ammonium hydroxide are provided for pH control during the growth of the microbial culture.

[0026] In one example (not shown), the aerobic conditions in growth vessel

(provided by sparging an oxygen-rich gas) are switched to anaerobic conditions by changing the oxygen-rich gas to an oxygen-deficient gas (e.g., C0 ₂ or the like). The anaerobic environment triggers bioconversion of the fermentable carbon source to succinic acid in situ in growth vessel 12. Ammonium hydroxide is provided for pH control during bioconversion of the fermentable carbon source to SA. The SA that is produced is at least partially if not totally neutralized to DAS due to the presence of the ammonium hydroxide, leading to the production of a broth comprising DAS. The C0 ₂ provides an additional source of carbon for the production of SA.

[0027] In another example, the contents of growth vessel 12 may be transferred via stream 14 to a separate bioconversion vessel 16 for bioconversion of a carbohydrate source to SA. An oxygen-deficient gas (e.g., C0 ₂ or the like) is sparged in bioconversion vessel 16 to provide anaerobic conditions that trigger production of SA. Ammonium hydroxide is provided for pH control during bioconversion of the carbohydrate source to SA. Due to the presence of the ammonium hydroxide, the SA produced is at least partially neutralized to DAS, leading to production of a broth that comprises DAS. The C0 ₂ provides an additional source of carbon for production of SA.

[0028] In yet another example, the bioconversion may be conducted at relatively low pH (e.g., 3 - 6). A base (ammonium hydroxide or ammonia) may be provided for pH control during bioconversion of the carbohydrate source to SA. Depending on the desired pH, due to the presence or lack of the ammonium hydroxide, either SA is produced or the SA produced is at least partially neutralized to MAS, DAS or a mixture comprising SA, MAS and/or DAS. Thus, the SA produced during bioconversion can be subsequently neutralized, optionally in an additional step, by providing either ammonia or ammonium hydroxide leading to a broth comprising DAS. As a consequence, a "DAS-containing fermentation broth" generally means that the fermentation broth comprises DAS and possibly any number of other components such as MAS and/or SA, whether added and/or produced by bioconversion or otherwise. Similarly, a "MAS-containing fermentation broth" generally means that the fermentation broth comprises MAS and possibly any number of other components such as DAS and/or SA, whether added and/or produced by bioconversion or otherwise.

[0029] The broth resulting from the bioconversion of the fermentable carbon source

(in either vessel 12 or vessel 16, depending on where the bioconversion takes place), typically contains insoluble solids such as cellular biomass and other suspended material, which are transferred via stream 18 to clarification apparatus 20 before distillation. Removal of insoluble solids clarifies the broth. This reduces or prevents fouling of subsequent distillation equipment. The insoluble solids can be removed by any one of several solid-liquid separation techniques, alone or in combination, including but not limited to centrifugation and filtration (including, but not limited to ultra-filtration, micro-filtration or depth filtration). The choice of filtration can be made using techniques known in the art. Soluble inorganic compounds can be removed by any number of known methods such as, but not limited to, ion-exchange, physical adsorption and the like.

[0030] An example of centrifugation is a continuous disc stack centrifuge. It can be useful to add a polishing filtration step following centrifugation such as dead-end or cross- flow filtration that may include the use of a filter aide such as diatomaceous earth or the like or, more preferably, ultra-filtration or micro-filtration. The ultra-filtration or micro-filtration membrane can be ceramic or polymeric, for example. One example of a polymeric membrane is SelRO MPS-U20P (pH stable ultra-filtration membrane) manufactured by Koch Membrane Systems (850 Main Street, Wilmington, MA, USA). This is a commercially available polyethersulfone membrane with a 25,000 Dalton molecular weight cut-off which typically operates at pressures of 0.35 to 1.38 MPa (maximum pressure of 1.55 MPa) and at temperatures up to 50°C. As an alternative to using centrifugation and a polishing filtration in combination, cross-flow filtration may be employed alone using ultra- or micro-filtration membranes. [0031] The resulting clarified DAS-containing broth or MAS-containing broth, substantially free of the microbial culture and other solids, is transferred via stream 22 to distillation apparatus 24.

[0032] The clarified broth should contain DAS and/or MAS in an amount that is at least a majority of, preferably at least about 70 wt%, more preferably 80 wt% and most preferably at least about 90 wt% of all the ammonium dicarboxylate salts in the broth. The concentration of DAS and/or MAS as a weight percent (wt%) of the total dicarboxylic acid salts in the fermentation broth can be easily determined by high pressure liquid chromatography (HPLC) or other known means.

[0033] Water and ammonia are removed from distillation apparatus 24 as an overhead, and at least a portion is optionally recycled via stream 26 to bioconversion vessel 16 (or growth vessel 12 operated in the anaerobic mode).

[0034] Distillation temperature and pressure are not critical as long as the distillation is carried out in a way that ensures that the distillation overhead contains water and ammonia, and the distillation bottoms preferably comprises at least some MAS and at least about 20 wt% water. A more preferred amount of water is at least about 30 wt% and an even more preferred amount is at least about 40 wt%. The rate of ammonia removal from the distillation step increases with increasing temperature and also can be increased by injecting steam (not shown) during distillation. The rate of ammonia removal during distillation may also be increased by conducting distillation under a vacuum, under pressure or by sparging the distillation apparatus with a non-reactive gas such as air, nitrogen or the like.

[0035] Removal of water during the distillation step can be enhanced by the use of an organic azeotroping agent such as toluene, xylene, methylcyclohexane, methyl isobutyl ketone, cyclohexane, heptane or the like, provided that the bottoms contains at least about 20 wt% water. If the distillation is carried out in the presence of an organic agent capable of forming an azeotrope consisting of the water and the agent, distillation produces a biphasic bottoms that comprises an aqueous phase and an organic phase, in which case the aqueous phase can be separated from the organic phase, and the aqueous phase used as the distillation bottoms. Byproducts such as succinamide and succinimide are substantially avoided provided the water level in the bottoms is maintained at a level of at least about 30 wt%.

[0036] A preferred temperature for the distillation step is in the range of about 50 to about 300° C, depending on the pressure. A more preferred temperature range is about 150 to about 240° C, depending on the pressure. A distillation temperature of about 170 to about 230° C is preferred. "Distillation temperature" refers to the temperature of the bottoms (for batch distillations this may be the temperature at the time when the last desired amount of overhead is taken).

[0037] Adding a water miscible organic solvent or an ammonia separating solvent facilitates deammoniation over a variety of distillation temperatures and pressures as discussed above. Such solvents include aprotic, bipolar, oxygen-containing solvents that may be able to form passive hydrogen bonds. Examples include, but are not limited to, diglyme, triglyme, tetraglyme, sulfoxides such as dimethylsulfoxide (DMSO), amides such as dimethylformamide (DMF) and dimethylacetamide, sulfones such as dimethylsulfone, gamma-butyrolactone (GBL), sulfolane, polyethyleneglycol (PEG), butoxytriglycol, N- methylpyrolidone (NMP), ethers such as dioxane, methyl ethyl ketone (MEK) and the like. Such solvents aid in the removal of ammonia from the DAS or MAS in the clarified broth. Regardless of the distillation technique, it is preferable that the distillation be carried out in a way that ensures that at least some MAS and at least about 20 wt% water remain in the bottoms and even more advantageously at least about 30 wt%. The distillation can be performed at atmospheric, sub-atmospheric or super-atmospheric pressures.

[0038] Under other conditions such as when the distillation is conducted in the absence of an azeotropic agent or ammonia separating solvent, the distillation is conducted at super atmospheric pressure at a temperature of > 100° C to about 300° C to form an overhead that comprises water and ammonia and a liquid bottoms that comprises SA and at least about 20 wt% water. Super atmospheric pressure typically falls within a range of >ambient atmosphere up to and including about 25 atmospheres. Advantageously the amount of water is at least about 30 wt%.

[0039] The distillation can be a one-stage flash, a multistage distillation (i.e., a multistage column distillation), multiple columns or the like. The one-stage flash can be conducted in a fiasher (e.g., a wiped film evaporator, thin film evaporator, thermosiphon flasher, forced circulation flasher and the like). The multistages of the distillation column can be achieved by using trays, packing or the like. The packing can be random packing (e.g., Raschig rings, Pall rings, Berl saddles and the like) or structured packing (e.g., Koch-Sulzer packing, Intalox pack-ing, Mellapak and the like). The trays can be of any design (e.g., sieve trays, valve trays, bubble-cap trays and the like). The distillation can be performed with any number of theoretical stages.

[0040] If the distillation apparatus is a column, the configuration is not particularly critical, and the column can be designed using well known criteria. The column can be operated in either stripping mode, rectifying mode or fractionation mode. Distillation can be conducted in either batch, semi-continuous or continuous mode. In the continuous mode, the broth is fed continuously into the distillation apparatus, and the overhead and bottoms are continuously removed from the apparatus as they are formed in a purge stream. The distillate from distillation is an ammonia/water solution, and the distillation bottoms is a liquid, aqueous solution of MAS and SA, which may also contain other fermentation by-product salts (i.e., ammonium acetate, ammonium formate, ammonium lactate and the like) and color bodies.

[0041] The distillation bottoms can be transferred via stream 28 to cooling apparatus

30 and cooled by conventional techniques. Cooling technique is not critical. A heat exchanger (with heat recovery) can be used. A flash vaporization cooler can be used to cool the bottoms down to about 15° C. Cooling to 0° C typically employs a refrigerated coolant such as, for example, glycol solution or, less preferably, brine. A concentration step can be included prior to cooling to help increase product yield. Further, both concentration and cooling can be combined using known methods such as vacuum evaporation and heat removal using integrated cooling jackets and/or external heat exchangers.

[0042] We found that the presence of some MAS in the liquid bottoms facilitates cooling-induced separation of the bottoms into a liquid portion in contact with a solid portion that at least "consists essentially" of SA (meaning that the solid portion is at least substantially pure crystalline SA) by reducing the solubility of SA in the liquid, aqueous, MAS-containing bottoms. Fig. 2 illustrates the reduced solubility of SA in an aqueous 20 wt% MAS solution at various temperatures ranging from 5 to 45°C. We discovered, therefore, that SA can be more completely crystallized out of an aqueous solution if some MAS is also present in that solution. A preferred concentration of MAS in such a solution is about 20 wt%. This phenomenon allows crystallization of SA (i.e., formation of the solid portion of the distillation bottoms) at temperatures higher than those that would be required in the absence of MAS.

[0043] The distillation bottoms, after cooling, is fed via stream 32 to separator 34 for separation of the solid portion from the liquid portion. Separation can be accomplished via pressure filtration (e.g., using Nutsche or Rosenmond type pressure filters), centrifugation and the like. The resulting solid product can be recovered as product 36 and dried, if desired, by standard methods.

[0044] After separation, it may be desirable to treat the solid portion to ensure that no liquid portion remains on the surface(s) of the solid portion. One way to minimize the amount of liquid portion that remains on the surface of the solid portion is to wash the 12 051326 separated solid portion with water and dry the resulting washed solid portion (not shown). A convenient way to wash the solid portion is to use a so-called "basket centrifuge" (not shown). Suitable basket centrifuges are available from The Western States Machine Company (Hamilton, OH, USA).

[0045] The liquid portion of the separator 34 (i.e., the mother liquor) may contain remaining dissolved SA, any unconverted MAS, any fermentation byproducts such as ammonium acetate, lactate, or formate, and other minor impurities. This liquid portion can be fed via stream 38 to a downstream apparatus 40. In one instance, apparatus 40 may be a means for making a de-icer by treating in the mixture with an appropriate amount of potassium hydroxide, for example, to convert the ammonium salts to potassium salts. Ammonia generated in this reaction can be recovered for reuse in the bioconversion vessel 16 (or growth vessel 12 operating in the anaerobic mode). The resulting mixture of potassium salts is valuable as a de-icer and anti-icer.

[0046] The mother liquor from the solids separation step, 34, can be recycled (or partially recycled) to distillation apparatus 24 via stream 42 to further enhance recovery of SA, as well as further convert MAS to SA.

[0047] The solid portion of the cooling-induced crystallization is substantially pure

S A and is, therefore, useful for the known utilities of SA.

[0048] HPLC can be used to detect the presence of nitrogen-containing impurities such as succinamide and succinimide. The purity of SA can be determined by elemental carbon and nitrogen analysis. An ammonia electrode can be used to determine a crude approximation of SA purity.

[0049] Depending on the circumstances and various operating inputs, there are instances when the fermentation broth may be a clarified MAS-containing fermentation broth or a clarified SA-containing fermentation broth. In those circumstances, it can be advantageous to add MAS, DAS and/or SA and, optionally, ammonia, and/or ammonium hydroxide to those fermentation broths to facilitate the production of substantially pure SA. For example, the operating pH of the fermentation broth may be oriented such that the broth is a MAS-containing broth or a SA-containing broth. MAS, DAS, SA, ammonia, and/or ammonium hydroxide may be optionally added to those broths to attain a broth pH preferably less than about 6 to facilitate production of the above-mentioned substantially pure SA. In one particular form, it is especially advantageous to recycle SA, MAS and water from the liquid bottoms resulting from the distillation step 24, and/or the liquid portion from the separator 34, into the fermentation broth and/or clarified fermentation broth. In referring to the MAS-containing broth, such broth generally means that the fermentation broth comprises MAS and possibly any number of other components such as DAS and/or SA, whether added and/or produced by bioconversion or otherwise.

[0050] In one embodiment, the processes described above may be an integrated process for making carboxylic acids, such as SA, from a fermentation broth containing mono or diammonium succinate salts, including reactive distillation whereby ammonium salts and/or amides and/or imides are converted to the corresponding carboxylic acids and ammonia, low temperature crystallization, separation of essentially pure carboxylic acids, and recycle of resulting mother liquor containing unreacted ammonium salts and process byproducts. It has been surprisingly discovered that the process allows conversion of byproduct carboxylic acid amides and imides and ammonium salts to the free carboxylic acid with equal efficiency, without the need for catalysis. Near total recycle of crystallization mother liquor to the process is provided, which yields near quantitative conversion of the biologically produced ammonium carboxylates to the corresponding mono-, di- and poly carboxylic acids.

[0051] Fig. 3 shows a block diagram of an embodiment of an integrated process with recycling of mother liquor. In the integrated process, fermentation broth containing ammonium salts or desired carboxylic acids may be introduced (3) along with optionally water or steam (4) to a reactive distillation system operating at elevated pressure (>50psig) and elevated temperature (>140°C) where free carboxylic acids and ammonia are eporduced. Ammonia, water and acetic acid (if present) are collected as overhead. A small amount of caustic may optionally be added whether separately (20) or with the broth to partially neutralize any strong by-product acids present (e.g. formic acid). The tails stream (6) from the reactive distillation containing the desired free carboxylic acids together with some unconverted feed are concentrated at atmospheric or lower pressure, if desired; this step is optional and depends on the tails stream concentration exiting the distillation. Water and trace amounts of remaining low boiling acids are removed overhead. The reactive distillation tails (6) or the concentrated tails (8) are cooled to less than 25°C, preferably less than 5°C and most preferably less than 0°C, where desired solid, pure carboxylic acid is selectively separated (9) and the crystallization mother liquor recycled (10, 1 1 & 2). The pure carboxylic acid may be subject to further purification (e.g., distillation, recrystallization) as may be desired, with distillation residues and/or mother liquors recycled to the process. Depending on the broth feed byproducts a small mother liquor purge (12) may be required. The purge stream 12 may be concentrated and chilled to obtain further unconverted feed ammonium U 2012/051326 salts for recycle (16). Any mother liquors from recrystallization (19) can be used to dissolve recovered ammonium salts (16 & 17) and combined for recycle to the process (2). Recycle of the mother liquor may include recycle of all the crystallization mother liquor or a lesser portion thereof.

[0052] In an embodiment of an integrated process, a DAS containing broth together with recycled downstream crystallization mother liquor is fed to a pressure distillation column operating at a head pressure of -200 psig and with a column base temperature of ~200°C. Water and /or steam may be fed to the base of the column at a rate of 0.1 to 10 times the weight of the succinate being fed to the top of the column. The water taken overhead in the column preferably exceed the rate of water and/or steam fed to the bottom of the column. Approximately 75% of the ammonia contained in the feed stream is recovered overhead together with water and byproduct such as acetic acid. The feed, tails and overhead feed stream rates may be adjusted, such that the residence time in the base of the reactive distillation column is between 1 and 1000 minutes.

[0053] The resulting conversion to succinic acid is approximately 50% of the succinate contained in the feed. The concentration of the succinic acid in the tails stream is between 1% and 50%o, most preferably between 5% and 35% and typically -20%. No further concentration is required. The tails stream may be cooled to a temperature as low as 0°C in the crystallization unit, where succinic acid (SA) is separated for further purification and the mother liquor recycled. The mother liquor recycle stream (1 1) to purge stream (12) ratio is at least 5: 1 , preferably 10:1 and typically 20:1. The purge stream is concentrated to >50% dissolved solids and cooled to <15°C. Solid monoamnionium succinate and alkali metal succinate, if present are recycled to the process. The process gives a net overall recovery of SA exceeding 95% of the theoretical in a purity >98%. The final purge stream (15) may be used as fertilizer or as a deicer additive.

[0054] It should be noted that recycle of the mother liquor can be fed into the process also into the broth or during the removal of water.

[0055] Figs. 4 to 7 show integrated processes using one and two stage reactive distillation either with or without a lower pressure concentration step. In particular, Figs. 4 and 5 show processes without low pressure concentration, whereas Figs. 6 and 7 show processes featuring the optional low pressure concentration step. In the low pressure concentration step, water is removed as overhead to concentrate the distillate prior to crystallization. In some instances, the low pressure concentration step is performed at pressure of less than about 30psig and most preferably at a pressure lower than about 20psig. [0056] It is believed that the low pressure resulting in lower temperature concentration of the product from the reactive distillation step(s) improves SA production. The low temperature concentration results in lower by-product content in the product going to crystallization. It is believed that this impacts both isolated SA purity and yield.

[0057] Additionally, Figs. 5 and 7 show a process using a two step reactive distillation. In some instances, the first stage of distillation may be performed at temperature of about 190°C and under pressure of about 160 psig. The second stage of reactive distillation may be performed at temperature of about 210°C and under pressure of about 250 psig.

[0058] Operating individual reactive distillation stages, if used at different pressures from each other and from the concentration step, allows for much more effective utilization of steam (i.e energy input). Effectively, at least triple use of steam. It is preferable that the reactive distillation stage pressures differ by at least about 15psi and preferably differ by more than about 30psi. In particular, the overhead steam from a reactive distillation step can effectively be used either directly or indirectly to provide steam to prior pressure reactive distillation steps. In addition, the overhead vapor (steam/NH ₃ ) from the lowest pressure reactive distillation step can be conveniently used to supply energy for the low pressure concentration step.

[0059] A preferred process scheme as shown in Fig. 7 offers numerous opportunities for cross exchange of tails and/or overhead streams with feed streams for heat integration. For example, the process shown in Fig. 7 provides cross exchange opportunities for chilled streams. In such an integrated process, heat and energy produced by or residual from one step can be applied to other steps of the process.

[0060] It should be noted, that although portions of this disclosure provides detailed description of processes for making SA, that the processes can be modified to make other carboxylic acids. For example, processes from making AA from fermentation broth containing DAA, MAA, AA, and or other biologically produced compounds or impurities. One skilled in the art would appreciate how to modify the distillation, crystallization, or other conditions of the present disclosure as necessary to make and isolate other carboxylic acids.

EXAMPLES

[0061] The processes are illustrated by the following non-limiting representative examples. In the first two examples, a synthetic, aqueous DAS solution was used in place of an actual clarified DAS-containing fermentation broth. The other examples employed an actual clarified DAS-containing fermentation broth.

[0062] For the first two examples, the use of a synthetic DAS solution is believed to be a good model for the behavior of an actual broth in our processes because of the solubility of the typical fermentation by-products found in actual broth. The major by-products produced during fermentation are ammonium acetate, ammonium lactate and ammonium formate. Ammonium acetate, ammonium lactate and ammonium formate are significantly more soluble in water than S A, and each is typically present in the broth at less than 10% of the DAS concentration. In addition, even when the acids (acetic, formic and lactic acids) were formed during the distillation step, they are miscible with water and will not crystallize from water. This means that the SA reaches saturation and crystallizes from solution (i.e., forming the solid portion), leaving the acid impurities dissolved in the mother liquor (i.e., the liquid portion).

EXAMPLE 1

[0063] This experiment shows the conversion of DAS to S A in an aqueous media.

[0064] An experiment was conducted in a 300 ml Hastelloy C stirred Parr reactor using a 15% (1.0 M) synthetic DAS solution. The reactor was charged with 200 g of solution and pressurized to 200 psig. The contents were then heated to begin distillation, bringing the temperature to approximately 200°C. Ammonia and water vapor were condensed overhead with cooling water and collected in a reservoir. Fresh water was pumped back to the system at a rate equal to the make rate (approximately 2 g/min) to maintain a constant succinate concentration and volume of material. The run continued for 7 hours. At the end of the run, analysis of the mother liquor showed 59% conversion to SA, 2.4% to succinamic acid, and 2.9% to succinimide. Cooling the mother liquor would result in a liquid portion and a solid portion that would be substantially pure SA.

EXAMPLE 2

[0065] This example demonstrates the effect of solvents on ammonia evolution from aqueous DAS. Run 10 is the control experiment where no solvent is present.

[0066] The outer necks of a three neck 1 -L round bottom flask were fitted with a thermometer and a stopper. The center neck was fitted with a five tray 1" Oldershaw section. The Oldershaw section was topped with a distillation head. An ice cooled 500 mL round bottom flask was used as the receiver for the distillation head. The 1 -L round bottom flask was charged with distilled water, the solvent being tested, SA and concentrated ammonium hydroxide solution. The contents were stirred with a magnetic stirrer to dissolve all the solids. After the solids dissolved, the contents were heated with the heating mantle to distill 350g of distillate. The distillate was collected in the ice cooled 500 mL round bottom flask. The pot temperature was recorded as the last drop of distillate was collected. The pot contents were allowed to cool to room temperature and the weight of the residue and weight of the distillate were recorded. The ammonia content of the distillate was then determined via titration. The results were recorded in Tables 1 and 2.

Table 1

Run # 1 2 3 4 5

Name of Acid charged Succinic Succinic Succinic Succinic Succinic

Wt Acid Charged (g) 11.8 11.81 1 1.83 1 1.8 1 1.78

Moles Acid Charged 0.1 0.1 0.1 0.1 0.1

Wt 28% NH3 Solution Charged (g) 12.76 12.78 12.01 12.98 13.1

Moles NH3 Charged 0.21 0.21 0.2 0.215 0.217

Name of Solvent DMSO DMF NMP sulfolane triglyme

Wt Solvent Charged (g) 400.9 400 400 400 400

Wt Water Charged (g) 400 400 400 400 401

Wt Distillate (g) 350.1 365.9 351.3 352.1 351.2

Wt Residue (g) 467.8 455 460.5 457.1 473

%Mass Accountability 99.1 99.6 98.5 98.1 99.8

Wt% NH3 in distillate (titration) 0.91 0.81 0.78 0.71 0.91

Moles NH3 in distillate 0.187 0.174 0.161 0.147 0.188

%NH3 removed in Distillate 89 83 81 66 86

%First NH3 removed in Distillate 100 100 100 100 100

%Second NH3 removed in Distillate 78 66 62 32 72

Final Pot Temp (°C) 138 1 14 126 1 13 103

Final DAS/MAS/SA ratio 0/22/78 0/34/66 0/38/62 0/68/32 0/28/72

[0068]

Table 2

Run # 6 7 8 9 10

Name of Acid charged Succinic Succinic Succinic Succinic Succinic

Wt Acid Charged (g) 1 1.84 11.81 1 1.8 11.81 11.8

Moles Acid Charged 0.1 0.1 0.1 0.1 0.1

Wt 28% NH3 Solution

Charged (g) 12.1 1 12.1 1 12.1 12.15 12.1

Moles NH3 Charged 0.2 0.2 0.2 0.2 0.2

Dowanol Tetra HeavyMe

Name of Solvent TPM Tetraglyme PentaEG GlyEther none

Wt Solvent Charged (g) 400.1 400 400 400.1 0

Wt Water Charged (g) 400 400 400 400.1 800

Wt Distillate (g) 350 345 350 349 351

Wt Residue (g) 468.4 473.8 465 470.4 466

%Mass Accountability 99.3 99.4 98.9 99.4 99.2

Wt% NH3 in distillate

(titration) 0.58 0.62 0.55 0.6 0.13

Moles NH3 in distillate 0.1 19 0.126 0.1 13 0.123 0.027

%NH3 removed in

Distillate 60 63 57 62 13.4

%First NH3 removed in

Distillate 100 100 100 100 27

%Second NH3 removed in

Distillate 20 26 14 24 0

Final Pot Temp (°C) 104 1 10 115 113 100

Final DAS/MAS/SA ratio 0/80/20 0/74/26 0/86/14 0/76/24 83/27/0

EXAMPLE 3

[0069] This example used a DAS-containing, clarified fermentation broth derived from a fermentation broth containing E. coli strain ATCC PTA-5132.

[0070] The initial fermentation broth was clarified, thereby resulting in a clarified fermentation broth containing -4.5% diammonium succinate (DAS). That clarified broth was used to produce crystalline SA as follows. The broth was first concentrated to approximately 9% using an RO membrane and then subjected to distillation at atmospheric pressure to further concentrate the broth to around 40%.

[0071] The concentrated broth was used as the starting material for conversion of

DAS to SA, carried out batchwise in a 300 ml Parr reactor. A 200 g portion of the solution was reacted at 200°C/200 psig for 11 hours. As the reaction proceeded, water vapor and ammonia liberated from the DAS were condensed and collected overhead. Condensate was T/US2012/051326 collected at about 2 g/min, and makeup water was fed back to the system at approximately the same rate.

[0072] Multiple samples were taken throughout the experiment. Samples taken early in the reaction indicated the presence of succinamide, succinamic acid, and succinimide. However, these nitrogen-containing byproducts ^' decreased throughout the experiment. Conversion to SA was observed to be 55% in the final bottoms sample. The final solution was concentrated by evaporation and cooled to 4°C. The resulting crystalline solids were isolated via vacuum filtration, washed with ice water and dried under vacuum. The product (7 g) was essentially pure SA as determined by HPLC.

EXAMPLE 4

[0073] A 500 mL round bottom flask was charged with 80g of an aqueous 36% DAS solution and 80g of triglyme. The flask was fitted with a 5 tray 1" glass Oldershaw column section which was topped with a distillation head. An addition funnel containing 3300g of water was also connected to the flask. The flask was stirred with a magnetic stirrer and heated with a heating mantel. The distillate was collected in an ice cooled receiver. When the distillate started coming over the water in the addition funnel was added to the flask at the same rate as the distillate was being taken. A total of 3313g of distillate was taken. The distillate contained 4.4g of ammonia, as determined by titration. This means -37% of the DAS was converted to SA with the rest being converted to MAS. The residue in the flask was then placed in an Erlenmeyer flask and cooled to -4°C while stirring. After stirring for 30 minutes the slurry was filtered while cold yielding 7.1g of solids. The solids were dissolved in 7.1 g of hot water and then cooled in an ice bath while stirring. The cold slurry was filtered and the solids dried in a vacuum oven at 100°C for 2 hrs yielding 3.9g of SA. HPLC analysis indicated that the solids were SA with 0.099%) succinamic acid present.

EXAMPLE 5

[0074] A pressure distillation column was made using an 8 ft long 1.5" 316 SS

Schedule 40 pipe that was packed with 316 SS Propak packing. The base of the column was equipped with an immersion heater to serve as a reboiler. Nitrogen was injected into the reboiler via a needle valve to pressure. The overhead of the column had a total take-off line which went to a 316 SS shell and tube condenser with a receiver. The receiver was equipped with a pressure gauge and a back pressure regulator. Material was removed from the overhead receiver via blowcasing through a needle valve. Preheated feed was injected into the column at the top of the packing via a pump along with a dilute 0.4% sodium hydroxide solution. Preheated water was also injected into the reboiler via a pump. This column was first operated at 50 psig pressure which gave a column temperature of 150° C. The top of the column was fed a 4.7% DAS containing broth at a rate of 8 mL/min along with 0.15 mL/min of 0.4% sodium hydroxide solution. Water was fed to the reboiler at a rate of 4 mL/min. The overhead distillate rate was taken at 8 mL/min and the residue rate was taken at 4 mL/min. A total of 2565g of broth was fed to the column along with 53g of 0.4% sodium hydroxide solution. A total of 2750g of distillate was taken and 1269g of residue taken during the run. Titration of the distillate indicated that -71% of the total ammonia contained in the DAS was removed (i.e. the residue was a 42/58 mixture of SA/MAS). The composited residue was then fed back to the same column the next day under the following conditions; pressure 100 psig and temperature 173 ⁰ C. The composited residue was fed to the top of the column at 4 mL/min along with 0.15 mL/min of 0.4% sodium hydroxide solution. The reboiler was fed water at 9.2 mL/min. A total of 1240g of residue from the previous day was fed to the column along with 58g of sodium hydroxide solution and 2890g of water. A total of 3183g of distillate was taken along with 1132g of residue during the run. Titration of the distillate revealed an additional -14% of the ammonia was removed yielding a 70/30 mixture of SA/MAS in the residue.

EXAMPLE 6

[0075] A pressure distillation column was made using an 8 ft long 1.5" 316 SS

Schedule 40 pipe, of which about 6 ft were packed with 316 SS Propak packing. The base of the column was equipped with an immersion heater to serve as the reboiler. Nitrogen was injected into the reboiler via a needle valve to pressure the system. The overhead of the column had a total take-off line which went to a 316 SS shell and tube condenser with a receiver. The receiver was equipped with a pressure gauge and a back pressure regulator. Material was removed from the overhead receiver via blow casing through a needle valve. Preheated aqueous ammonium carboxylate feed was injected into the column typically at the top l/3 ^rd of the packing via a pump unless otherwise specified. Preheated water was also injected into the reboiler via a pump. This column was first operated in the 15 to 200 psig pressure range, which gave a column operating temperature in the 110°C to 205°C range. Optionally, when desired a dilute <1% sodium hydroxide solution was co-fed to the top of the column packing. Both overhead and tails streams were continuously taken. A bottoms kettle volume of -180 ml was typically maintained. The overhead water stream was typically analyzed for N¾ and acetate if expected to be present. Feeds and tails streams were controlled by volume, and the total feed and make weights were measured. The tails stream U 2012/051326 was typically analyzed by LC (liquid chromatography), acidity and ammonium ion content. On recycle runs the crystallized acid removed was made up by mixing fresh feed with the recycle mother liquor.

[0076] Multiple stage runs were conducted by using the unaltered tails of a run as the feed for a subsequent run, which may have been conducted at various reaction conditions and pressures. Data for several representative runs using various aqueous monoammonium succinate (MAS) and diammonium succinate (DAS) feeds are included in Table 3. Unless otherwise stated, runs were conducted for about 6 hours, which was long enough to approximate steady state operation. The percentage yields were estimated by determining the percentage of the available ammonia in the feed that was recovered in the aqueous overhead (OH) make. The experiments in Table 3 utilized either aqueous diammonium succinate (DAS) or monoammmonium succinate (MAS) feeds. The respective tails streams were analyzed to determine the relative amounts of succinamic acid (SAM) and succinimide (SIM) produced as well as wt% succinate [SA + MAS + DAS]. The last column in Table 3 reports the weight ratio of water taken overhead to aqueous feed, which is a measure of the energy efficiency of the process. Less water taken overhead is preferred because it allows for a lower energy cost. The following observations are made based on these experiments:

1. Experiments 18, 19 & 20 with DAS show that below 40 psig reaction

pressure and 140°C only the first ammonia is removed and little free

SA is produced.

2. Experiments 37-50, 55-56 & 60-64 show that ammonia removal and

conversion to SA increase with both the weight ratio of overhead water to MAS feed and with increasing reaction pressure/temperature.

3. Experiments 51-52 and 57-59 show that feed of the MAS to the

reaction kettle rather than to the column results in lower removal of N¾ and lower conversion to SA.

4. The experiments also show byproduct SAM + SIM tend to increase

with both the tails concentration of succinate and with increased temperature.

[0077] Data for several continuous staged reactive distillation experiments are shown in Table 4. These experiments were conducted by using the tails from the respective prior stage as feed to the column. The experiments further demonstrate the impacts of additional stripping water and increased reaction pressure/temperature on conversion. The following observations are made based on the Table 4 staged reactions.

5. Five stage experiment 70a-e further demonstrates the impact of

increased stripping water and reaction pressure/temperature on MAS to SA conversion. Note that an overall conversion of 74% was achieved, although the weight ratio of water overhead was 120.

[0078] Table 3. Continuous Pressure Distillation Experiments Feeding Aqueous

Diammonium Succinate (DAS) or Monoammonium Succinate (MAS)

Two stage experiments 72 & 73 demonstrate that over 60% conversion can be achieved in two stages and with less stripping water by raising the initial reaction stage pressure.

Experiment 74 was run with the same conditions as experiment 72 except 20% of the MAS in the feed was replaced with SIM. Experiment 74 conversion (68%) was equivalent to experiment 72 conversion (67%) to SA. This result surprisingly demonstrates that byproduct amides and imides may be converted equally as well as ammonium salts to the respective acids.

Experiments 83 and 84 were conducted with three stages all at 162 psig. It is observed that little additional conversion is being achieved with additional stages considering the disadvantage of excessive stripping water.

Two stage experiments 77 demonstrate that the reaction may be effectively run without caustic feed at the top of the column.

Experiment 80 was conducted in two stages using monoammonium lactate (MAL) feed. Over 50% conversion to lactic acid was achieved, which demonstrates the versatility of the process.

0079] Table 4. Staged Continuous Reactive Distillation Experiments

[0080] Data for several continuous staged succinate broth reactive distillation experiments are shown in Table 5. Biologically derived broth identified as PUF 20-33 was used for the experiments in Table 5. It contained about 3.9 wt% succinic acid and 0.71 wt% acetic acid, both present as their mixed ammonium/sodium salts. The molar ammonium to sodium ratio was estimated to be approximately 10: 1. Undetermined small amounts of other carboxylic acids, such as formic, pyruvic and fumarate, together with miscellaneous fermentation residues were also present. The following observations are made based on the Table 5 succinate broth staged reactions.

11. Six stage succinate broth experiment 9a-f confirmed that reaction

pressures and temperature exceeding 50 psig and 145°C, respectively, were required for good conversion to free S A.

12. Experiments 75a-c and 79a-c verified that good conversion of broth to

free acid could be achieved in three stages.

13. These experiments also showed the small marginal benefit of

additional reaction stages relative to the energy cost of distilling additional stripping water.

14. Surprisingly the overhead water cuts contained increasing quantities of

acetic acid as the reaction pressure was increased. In fact the majority of the contained feed acetic acid could be accounted for in the overhead product.

2012/051326

Table 5. Staged Succinate Broth Continuous Reactive Distillation

EXAMPLE 7

[0082] The consolidated tails (3199g) from CPD-79c was concentrated with vacuum and at a maximum temperature of 88°C to 533g and allowed to cool to room temperature. About 70g of tan crystals, 79C1 , were filtered and air dried. The liquid chromatography (LC) analysis indicated 94.3% succinate and 0.38% SAM. No acetic acid or SIM was present. The acid to ammonium ratio was measured to be 414 indicating less than 0.5% of the acid was present as its ammonium salt (MAS). Assuming the balance of the solid was water, the SA purity was >99%. It was estimated a total of 286g of DAS (222g as SA) was fed in run CPD- 79. The 79C1 yield was -32% of the theoretical and -58% of the SA available. [Note that 77%o of the theoretical ammonia was removed in the reaction, indicating all of the DAS was converted to MAS and 54%> of the resulting MAS was converted to SA]. The mother liquor from 79C1 was further concentrated and allowed to crystallize at ambient temperature. An additional 35g, 79C2, was isolated. LC analysis showed 86.3% succinate and 0.2% SAM. The acid to ammonium ratio was 2, which indicated that more than 66% of the succinate was present as its ammonium salt (MAS).

[0083] Table 5. Staged Succinate Broth Continuous Reactive Distillation

Experiments

[0084] Summary run data for several single stage succinate run experiments are detailed in Table 6. The runs were conducted for approximately 6 hours per day, in some cases up to 3 days. The column feeds were 25% MAS, synthetic MAS Broth I (25% MAS + 3% AmAc), Syn MAS Broth II (25% MAS + 3% AmAc + 0.2% Amformate + 0.5% mMonoAmFumarate) or Syn DAS Broth III (28% DAS + 3% AmAc + 0.2% Amformate + 0.5% DiamFumarate). (AmAc = ammonium acetate, Amformate = Monoammonium formate, DiamFumarate = diammonium fumarate.) The product tails from most runs (86, 86R, 88, 88R, 88R2, 90, 90R, & 90R2) were concentrated and chilled to 5°C in order to crystallize and isolate the product SA. The crystallization mother liquors were replenished with makeup respective feed to replace the crystallized SA, and were then subjected to recycle reactive distillation to simulate continuous recycle. SA product recoveries typically averaged -37% of that fed. The crystallization details including product purity are summarized in Table 7. The following observations are made based on the Table 6 single stage reactions.

[0085] Table 6. Single Stage Synthetic Broth and Mother Liquor Recycle Reactive

Distillation Experiments

15. Over 40% conversion of ammonium succinates (DAS and /or MAS) to

S A can be achieved at 190°C with an overhead to feed ratio as low as 13 parts stripping water to succinate feed. U 2012/051326 Crystallization mother liquor recycle together with makeup feed surprisingly gives similar SA conversion as the previous run (86R, 88R, 88R2, 90R & 90R2)..

The acetic acid in the overhead runs 88, 89 & 90 accounted for over 90% of that fed in the respective synthetic feeds.

Observation 16 together with 17 suggest that essentially total recycle of the crystallization mother liquors is possible.

Run 95 with only SIM and SAM feed (no MAS or DAS) gave equivalent conversion (N¾ removal) as equivalent run 88 and 89, which validates this finding.

In run 93 the column feed was moved to the top of the packed column and in run 94 the column feed was moved to a point ~ 1/3 up the packed section of column. The runs were conducted essentially the same as run 86, giving identical conversion to SA. This result suggests that only a few distillation stages are required.

Note that after two recycles, more SA has been isolated than was fed initially as diammonium salt in run series 88 and 90. Thus we have demonstrated efficient, cost effective quantitative conversion of diammonium succinate to SA in the described process.

U 2012/051326

[0086] Table 7. Succinic Acid Crystallization/Recovery Summary

[0087] The tails product was concentrated (from 25 wt% to 45 wt %) based on the total contained succinate, calculated as SA. It was unexpectedly found that reducing the crystallization temperature was far preferable to concentration by water evaporation as a means to increase yield and product purity [Additional examples will be provided to demonstrate this more clearly]. Note that lowering the temperature to enhance recovery is far less costly than evaporating water to concentrate the tails stream. In the preferred process we will likely eliminate the concentration unit shown in Figure 1. The preferred crystallization temperature is <25°C and most preferred is less than 5°C. The entire SA in the table was isolated at ~5°C with the exception of 93 which was crystallized at ambient room temperature. The following observations are made based on the crystallization/recovery data in Table 7. Note that the available recovery is calculated from the conversion and the total ammonium succinates fed. 12 051326

22. The results in the table demonstrate recovery of >90% of the available

contained SA in >98% purity, even in the recycle runs (86R, 88R, 88R2, 90R & 90R2), using the preferred crystallization temperature.

23. Although run 93 was concentrated the most (45%), the available

recovery is -30% less than those chilled to 5°C. It is expected its purity won't be as good as well. Note as mentioned, evaporating water is far more costly than cooling it.

[0088] Although our processes have been described in connection with specific steps and forms thereof, it will be appreciated that a wide variety of equivalents may be substituted for the specified elements and steps described herein without departing from the spirit and scope of this disclosure as described in the appended claims.