EP2451761A2

EP2451761A2 - Method for separating alcohol-ketone mixtures

Info

Publication number: EP2451761A2
Application number: EP10731478A
Authority: EP
Inventors: Kirstin Suck; Ulrich Sohling; Friedrich Ruf; Andreas Liese; Katja Goldberg; Paul Bubenheim
Original assignee: Sued Chemie AG
Current assignee: Clariant Produkte Deutschland GmbH
Priority date: 2009-07-07
Filing date: 2010-07-07
Publication date: 2012-05-16
Also published as: DE102009032080A1; WO2011003607A3; WO2011003607A2

Abstract

The invention relates to a method for separating a mixture having a first compound dissolved in a solvent, which first compound comprises at least one keto group, and having a second compound, which comprises a chiral carbon atom with a hydroxyl group, which was yielded from the stereoselective reduction of the keto group from the first compound. The mixture is brought into contact with an inorganic adsorber material and at least part of the second compound is adsorbed on the inorganic adsorber material, and a mixture depleted in the second compound is separated from the inorganic adsorber material.

Description

Process for the separation of alcohol-ketone mixtures

The invention relates to a process for the separation of a mixture with a first compound dissolved in a solvent, which comprises at least one keto group, and a second compound which comprises a chiral carbon atom having a hydroxy group which has arisen by reduction of the keto group from the first compound.

Chiral hydroxy compounds are valuable building blocks for the preparation of pharmaceutically active compounds. Chiral compounds are usually difficult to prepare enantiomerically pure by classical chemical methods. If, in a synthesis step, both enantiomers are formed, they must subsequently be subjected to complex separation processes

Use of optically active auxiliaries are separated. Therefore, attempts are made to prepare chiral compounds

to apply biotechnological procedures, whereby the

Stereoselectivity of enzymatic reactions is used.

In this case, both complete microorganisms can be used as well as completely or partially purified

isolated enzymes. For the enzymatic synthesis of chiral alcohols can

will label different ways. Examples of suitable enzymes are oxidoreductases, hydrolases and lyases. K.

Goldberg, K. Schroer, S. Lutz and A. Liese, Appl. Microbiol. Biotechnol. (2007) 76: 237-248 gives an overview of processes for preparing chiral alcohols using isolated enzymes.

Stereoselective syntheses can also be used

complete microorganisms are performed, such as

Bacteria or fungi. In this reaction, the metabolism of the cells can be used for an internal regeneration of the cofactor. An overview of the use of complete cells for the reduction of ketones to chiral alcohols is given by K. Goldberg, K. Schroer, S. Lutz and A. Liese in Appl. Microbiol. Biotechnol. (2007) 76: 249-255.

Catalyze enzymes from the class of oxireductases

Redox reactions, that is reactions in which electrons are transferred to or from the substrate. To the group of

Oxireduktases include, among others, the alcohol dehydrogenases (ADH), which are also referred to as carbonyl reductases (CR). Alcohol dehydrogenases, for example, catalyze the reduction of prochiral ketones to chiral alcohols.

Most alcohol dehydrogenases are dependent on the redox cofactors ß-1, 4-nicotinamide adenine dinucleotide (NADH) or ß- 1, 4-nicotinamide adenine dinucleotide phosphate (NADPH), which are stoichiometrically consumed in the enzymatic reduction. A stoichiometric use of these cofactors is uneconomical because of the high prices. The enzymatic reaction is therefore usually carried out so that the cofactor is regenerated in a further enzyme-catalyzed reaction by oxidation of a cosubstrate. A co-substrate is a compound which is oxidized enzymatically as a reducing agent, the released electrons being transferred to NAD or NADP, which are thereby regenerated to NADH or NADPH. In order to be able to use the enzymes used more efficiently in industrial processes and thus reduce costs, the enzymes can also be recovered so that they can be used in a further reaction. In EP 1 568 780 Bl, a method for the

Enantioselective reduction of keto compounds by enzymes described. The keto compound is first in an aqueous reaction medium, which in addition to water

Reducing agent, alcohol dehydrogenase and coenzyme, reduced to a secondary alcohol. The reaction is carried out under reduced pressure, so that volatile

Components can be removed from the reaction system. The aqueous phase is then extracted with a water-immiscible organic solvent such that the secondary alcohol formed transitions from the aqueous phase to the organic phase. The organic phase becomes

separated and reused the aqueous phase and added again for the next reaction cycle with keto compound and optionally reducing agent, enzyme and coenzyme. Enzymatic reactions are equilibrium reactions, that is, they occur during the course of the reaction

thermodynamic equilibrium. If whole cells are used for the reaction, the product may also be toxic to the cell. Thus, if the concentration of the product in the batch exceeds a certain value, the cells die. In both cases, therefore, the reaction comes to an end after an initial phase. In order to avoid deactivation or reduction of the synthesis activity, it is possible to remove the product in situ from the reaction mixture, so that the equilibrium is shifted to the side of the product. This can be done for example by a gaseous

Product is continuously removed from the reaction mixture or a product is continuously precipitated. JT Vincenzi, MJ Zmijewski, MR Reinhard, BE Landen, WL Muth and PG Marler, Enzyme and Microbiol. Technology 20: 494-499, 1997 describe a stereoselective enzymatic reduction of a ketone, wherein the resulting alcohol is bound in situ to a polymeric resin and thus from the

Reaction equilibrium is removed. The authors describe the reduction of 3, 4-methylenedioxyphenylacetone to S-3,4-methylenedioxyphenylisopropanol by Zygosaccharomyces rouxii. Complete cells were used. The yield was more than 95% with an enantiomeric excess (ee) of more than 99.9%. Since both the substrate and the product are toxic to the cells employed, hydrophobic polymer resins are used which have been loaded with the substrate. The substrate bound to the solid polymer phase is in equilibrium with the aqueous phase in which the cells are provided. The small amount of dissolved in the water phase substrate is from the cells to

reduced corresponding optically active alcohol, which is then bound back to the solid resin phase. In this way, a low constant concentration of the substrate or of the product in the water phase can be set, with overall very high yields of the desired product being obtained. A disadvantage of this method is that a relatively expensive resin is used for the separation of the product. This is particularly disadvantageous when the chiral

Alcohol is used for the synthesis of a product which is under great cost pressure, for example products for the agricultural industry or intermediates in the chemical industry. Furthermore, the resin used on the

respective reaction are tuned, so that a highly preferred adsorption of the alcohol and thus a high

Concentration of the remaining ketone in the

Reaction mixture is achieved. Another way to reduce the cost of an enzymatically catalyzed reaction is to bind the enzyme to an adsorbent, which can be separated from the reaction mixture after the end of a reaction and used in a further approach. In DE 10 2006 010 994 Al a method for enzymatic

Preparation of chiral alcohols described, wherein the reaction mixture contains an adsorbent, which with a

Oxireduktase is associated and which is separated from the reaction mixture after the end of the reaction. In the

Examples is used as an adsorbent for the enzyme diatomaceous earth (Celite ^®). After filtration of the reaction mixture, the filter cake can be placed in a further batch and used there again for the catalysis of a reduction of a ketone to an optically active alcohol.

The prior art methods described above use relatively expensive components for the prior art

Synthesis. Such processes are therefore only justified if the optically active alcohols which are obtained in the reaction, for the synthesis of very expensive substances

be used. Therefore, if chiral alcohols for the

Production of products to be used, which are under a high cost pressure, the described methods are not suitable per se. The invention therefore an object of the invention to provide a method for separating a mixture of a ketone and an obtained from the ketone by reduction optically active alcohol available, which can be carried out easily and inexpensively. This object is achieved by a method having the features of patent claim 1. Advantageous embodiments of the method are the subject of the dependent claims. The invention was based on the observation that ketones are less strongly bound by inorganic adsorbents than alcohols. Inorganic adsorbents are generally inexpensive and available in large quantities. The process can therefore also be adapted very easily to industrial standards, without high cost pressure from the

Provision of the inorganic adsorbent is triggered. The method is therefore particularly suitable for the provision of such optically active alcohols, which u. a. used as building blocks for the production of fine chemicals, for example for the chemical industry, where profit margins are limited by relatively high competitive pressure.

According to the invention, a method for separating a

Mixture with a first compound dissolved in a solvent, which comprises at least one keto group, and a second compound which comprises a chiral carbon atom having a hydroxy group, which is prepared by reduction of the

Keto group originated from the first compound, to

Provided, wherein the mixture is brought into contact with an inorganic adsorbent material and at least a part of the second compound to the inorganic

Adsorber material is adsorbed, and a depleted in the second compound mixture of the inorganic

Adsorber material is separated.

The inventive method is therefore suitable for the

Separation of mixtures typically obtained by reducing a prochiral keto compound to an alcohol. The reduction is carried out in solution, which is why both the starting material and the product are present in solution. The reduction is preferably carried out in such a way that essentially only one of the enantiomers is formed. It is therefore not necessary to use the inorganic Adsorber material perform a separation of enantiomers. The information about the stereochemistry, which occurs during the reduction at the carbon atom of the keto group, is therefore already introduced during the reduction of the keto group in the second compound. It is initially not significant how the stereoselective reduction takes place. So it's also possible the reduction

for example, to carry out with a stereoselective catalyst in the course of a conventional hydrogenation, wherein on the

Catalyst the information on the stereochemistry at

Carbon atom of the keto group is introduced, for example by using asymmetric catalysts.

In the reduction, the carbon skeleton of the first compound is substantially retained, so that the first compound and the second compound preferably differ in their structure only at the carbon atom of the keto group of the first compound which is converted into an optically active alcohol. In the reduction, the highest possible stereoselectivity is sought. The mixture preferably has an enantiomeric excess in favor of one of the two enantiomeric alcohols of more than 98%, preferably more than 99%, particularly preferably more than 99.5%.

The mixture containing both the first compound, ie the ketone, and the second compound, ie the optically active alcohol, is then brought into contact with an inorganic adsorbent material. The inorganic one

Adsorber material is selected so that it preferably has the highest possible adsorptive power in favor of the alcohol, so the second compound. exemplary

inorganic adsorbent materials are aluminum oxides,

Alumina hydrates, bentonites, silica gels, silicates such as aluminosilicates, magnesium silicates, calcium silicates and

Hydrotalcites. The mixture is so long with the inorganic Adsorber material brought into contact that at least one

Proportion of the second compound on the inorganic

Adsorber material is adsorbed and thus the mixture is depleted in the second compound. The second compound is thus removed from the reaction, so that in the case of an equilibrium reaction the equilibrium in favor of

Product is shifted or, if the product is inhibitory or toxic to the catalyst, is removed from the reaction mixture. The method is particularly advantageous

when a solvent is used in the reaction mixture which does not involve adsorptive sites on the inorganic adsorber material with the second compound

competes. Since water is a very polar solvent and therefore very strong with the resulting alcohol in the reduction of the keto group to binding sites on the inorganic

Adsorber material can compete, the inventive method according to an embodiment is preferably carried out in such a way that as an organic solvent

Solvent is used. Particular preference is given to using organic solvents which have no protonatable or deprotonatable groups.

Preferably, the organic solvent is selected so that it has a lower polarity than the alcohol resulting from the reduction. Particularly preferred

Used solvents which have a relatively low polarity. Such solvents are characterized by a relatively low boiling point at atmospheric pressure. Preference is given to using organic solvents which are added at

Normal pressure have a boiling point of less than 9O ⁰ C, preferably less than 80 ⁰ C, particularly preferably less than 70 ⁰ C. Suitable solvents are, for example, esters, such as ethyl acetate, ethers, such as methyl t-butyl ether, Alkanes, such as hexane or petroleum ether, or aromatic hydrocarbons, such as toluene.

According to a further preferred embodiment, the solvents used are those organic solvents which are substantially immiscible with water. As organic solvents which are substantially immiscible with water, such organic ones are preferred

Used solvent which is less than 5 wt .-% at 20 ⁰ C and atmospheric pressure, less than 2 wt .-%,

more preferably less than 1 wt .-%, in particular less than 0.5 wt .-% water. Unless with this

Embodiment the mixture contains larger amounts of water, therefore, two phases are formed, wherein the reduction of the keto group to the chiral hydroxy group takes place substantially in the organic phase. In this way, the inventive method can also be the implementation of polar

Include compounds as reactants, which dissolve only in the water phase. Without wishing to be bound by this theory, the inventors start from the idea that in this case the reaction takes place at interfaces between the organic phase and the water phase, the product in the

Essentially dissolved in the organic phase and is adsorbed on the inorganic adsorbent material from there.

As already explained above, in the inventive

Method of reducing the keto group in the way

that essentially only one of the enantiomers is formed. These can be, for example, stereoselective

Catalysts are used. According to a preferred

Embodiment, the reduction of the keto group to the hydroxy group bonded to a chiral carbon atom is carried out in such a manner that the reduction occurs under catalysis by an enzyme. Enzymes have a high regio- as well

Stereoselectivity, so that the reduction with a high Enantiomeric excess in favor of one enantiomer

can be carried out. The enzymes are preferably selected from the class of the oxidoreductases. The specific one

Oxireduktase becomes dependent on the substrate

selected. The enzymes can be used both in isolated and possibly purified form as well as in cells, so that the reduction of the keto group is carried out in the form of a fermentation. As such, the method according to the invention is not subject to any restrictions here. If the enzymes are used in substance, they can both in shape

open-minded cells are used as well

purified form. In the latter case, it is both possible to use the enzymes in dissolved form as well as in one

bonded form, that is, in a form in which the enzyme is bound to a solid matrix. The skilled person can refer here to known techniques.

The enzymatic reduction by oxidoreductases generally requires the presence of a cofactor which is consumed in the reduction in stoichiometric amount. According to one embodiment it is therefore envisaged that the reduction of the keto group to the hydroxy group bound to a chiral carbon is effected by the enzyme in the presence of a cofactor. Cofactors used are conventional cofactors. Exemplary cofactors are NADH and NADPH. However, it is also possible to use other known cofactors.

According to a preferred embodiment, an alcohol dehydrogenase is used as the enzyme for the stereoselective reduction of the ketone. The alcohol dehydrogenase is selected depending on the substrate. Suitable alcohol dehydrogenases are, for example, yeast alcoholdehydrogenase (YADH),

alcoholdehydrogenase A (ADH-A), Lactobacillus brevis

alcoholdehydrogenase (LbADH), 6-hydroxyhexanoic acid dehydrogenase (HCADH) or the reductase carbonyl reductase from Rhodococcus erythropolis (RECR) or also carbonyl reductase (CPCR) from Candida parapsilosis.

The alcohol dehydrogenases are recovered by conventional methods. Preferably, such alcohol dehydrogenases are used, which also in organic solvents, the reduction of the keto group to a chiral

Catalyze bonded hydroxy group.

As already explained above, the inventive

Method even in the presence of a small amount of a

Water phase can be performed. This may be required, for example, to solve a very polar cofactor, such as NAD ⁺ or NADP ⁺ . After the oxidation, the cofactor, for example NADH or NADPH, may optionally pass into the organic phase and be used there in the enzymatically catalyzed reduction of the keto group. Since large amounts of water cause water molecules to compete for adsorption sites on the inorganic adsorber, it is provided according to a preferred embodiment of the method according to the invention that the cofactor is dissolved in a water phase and the water phase in addition to the organic solvent in a proportion of less than 1 wt .-%, based on the

organic solvent contained in the mixture. In one embodiment, the water phase is in a proportion of at least 0.1 wt .-%, based on the organic

Solvent, contained in the mixture. The share of

Water phase is chosen according to an embodiment at least so large that the cofactor can be completely dissolved in the water phase. As already explained above, the cofactor can be dissolved, for example, only in the reduced form in the water phase and, after its oxidation, back into the organic phase pass. But it may also be that the cofactor in both the reduced and the oxidized form in the

Water phase is included and the enzyme-catalyzed reduction of the ketone takes place at a boundary phase between the organic phase and the water phase. In this embodiment of the

inventive method is preferred that the

Water phase is dispersed in finely divided form in the organic phase.

The enzymes which are used in the enzymatic reduction of the first compound, ie the reduction of the keto group, usually require the presence of a cofactor, which is oxidized during the enzymatic reaction. These

Cofactors are consumed in stoichiometric amounts as discussed above. Since they are very expensive, it is provided according to a preferred embodiment that the cofactor is regenerated during the reduction of the keto group, ie the cofactor is cycled between its oxidized and its reduced state.

The regeneration of the cofactor can be carried out by conventional methods. Thus, the regeneration can be done electrochemically. However, it is also possible to carry out the regeneration chemically by adding a further reducing agent to the mixture, for example

Sodium dithionite. Preference is given to the regeneration of

Cofactors, however, also carried out enzyme-catalyzed.

In this case, conventional systems can be used.

For example, formic acid or formates may be through

Formate dehydrogenase are oxidized to carbon dioxide. Next can be used as a reducing agent, for example, glucose or glucose-6-phosphate, which by

Glucose dehydrogenase or glucose-6-phosphate dehydrogenase are oxidized to the corresponding sugar acids. In one embodiment, the regeneration of the cofactor is also by an alcohol dehydrogenase

wherein the alcohol dehydrogenase can be used both for the reduction of the ketone to the chiral alcohol and for the oxidation of a so-called sacrificial alcohol to the corresponding ketone. In this embodiment of the method according to the invention is therefore intended that the mixture as a further constituent of a sacrificial alcohol for

Regeneration of the cofactor contains. As the sacrificial alcohol, lower secondary alcohols are preferably used, which preferably comprise 3 to 10 carbon atoms, more preferably isopropanol or isobutanol. The alcohol used can be used in excess. According to one

Embodiment, the proportion of the sacrificial alcohol in the mixture is less than 25 wt .-%, preferably less than 10 wt. ^■ %, particularly preferably less than 5 wt .-%, based on the organic solvent, which does not participate in the enzymatic reduction. Preferably, the sacrificial alcohol is used at least in an amount in the mixture which corresponds to twice the stoichiometric amount, particularly preferably at least three times the stoichiometric amount required for the regeneration of the cofactor.

According to one embodiment it is provided that the

Sacrificial alcohol has a lower molecular weight than the second compound. Preferably, the molecular weight of the sacrificial alcohol is less than 75%, more preferably less than 55%, of the molecular weight of the second compound.

As already explained above, the information about the stereochemistry at the chiral carbon atom to which the

Hydroxy group is bonded, preferably introduced during the reduction of the keto group. It is then not necessary to carry out a separation of enantiomers in the resulting alcohol. According to one embodiment, it is therefore provided that the inorganic adsorbent is an achiral adsorbent. Under an achiral adsorbent is a

Adsorbent understood in which both enantiomers are adsorbed substantially the same, so that by the

Adsorption no separation of the two possible enantiomers takes place. Very favorable inorganic adsorbents can therefore be used in the process according to the invention, which is advantageous in particular for industrial application.

The process according to the invention is preferably carried out in such a way that the alcohol formed during the reduction of the keto group is continuously removed from the reaction mixture. For this purpose, the enzyme can be separated, for example, from the solvent phase. The solvent which is the mixture of first and second compound

contains, with the inorganic adsorbent in

Contact, so that the second compound is at least partially removed by the inorganic adsorbent from the solvent. For this purpose it can be provided that the reaction mixture in which the enzymatic reaction is carried out, for example via a membrane

is separated so that the enzyme from a part of

Solvent is separated. The solvent is then contacted with the inorganic adsorbent by passing the solvent through, for example, a column packed from the inorganic adsorbent so that the solvent phase on the second compound is depleted. The depleted in the second compound solvent phase can then be returned to the reaction mixture. According to one embodiment of the process according to the invention, it is provided that the reduction of the keto group to a hydroxy group bonded to a chiral carbon atom in Presence of the inorganic adsorbent is carried out.

In this embodiment, therefore, the inorganic

Adsorbent added to the reaction mixture. The amount of inorganic adsorbent is chosen to be sufficiently high, so that a sufficiently large amount of the second

Compound can be adsorbed on the inorganic adsorbent to the reduction of the keto group of the first

Set the connection in favor of the highest possible yield in favor of the second compound.

After the reaction, the reaction mixture can be separated very easily by allowing the inorganic adsorbent to sediment, for example, and the

Solvent phase, which contains the enzyme and optionally also portions of the first compound is decanted off.

It is also possible to separate the inorganic adsorbent after the reduction of the keto group by filtration from the solvent phase.

The inorganic adsorbent is preferably so

selected that the second compound compared to the first compound to an increased degree of the inorganic

Adsorbent is adsorbed.

A suitable inorganic adsorbent is

For example, silica, which is preferably used in finely dispersed form.

In the process according to the invention, preference is given to using inorganic adsorbents which have an aluminum content, calculated as Al ₂ O _3, of more than 40% by weight. The aluminum may be contained in the inorganic adsorbent in the form of alumina. But it is also possible the inorganic adsorbent contains the aluminum in the form of, for example, a mixed oxide. Next to the

Aluminum may contain other metals in the inorganic adsorbent, such as silicon. The alumina-containing inorganic adsorbent may be degraded from a natural source, so a

be natural mineral. Preferably, however, synthetic compounds are used. These can be under

in controlled conditions and therefore lead to reproducible properties and results

inventive method. The inorganic one

Adsorbent is preferably predominantly of an alumina-containing inorganic adsorbent. It is possible that the inorganic adsorbent in addition to the actual adsorbent, for example, contains a binder, which itself may also have no adsorption properties. Preferably, the inorganic comprises

Adsorbent more than 60 wt .-%, preferably more than

80 wt .-%, more preferably more than 90 wt .-% Al ₂ O ₃ . The difference to 100% could be formed, for example, each by a proportion of a binder, with which the inorganic adsorbent, for example, to a

Particle is bound. According to a particularly preferred embodiment, the inorganic adsorbent consists only of alumina.

The inorganic adsorbent used in the process according to the invention preferably has a very high

specific surface on. Preferably, the inorganic adsorbent has a specific surface area of more than 100 m ² / g. According to a preferred embodiment, the inorganic adsorbent has a specific surface area in the range of 100 to 750 m ² / g, particularly preferably 120 to 700 m ² / g, particularly preferably 140 to 650 m ² / g. The specific surface area is determined by the BET method.

According to a further preferred embodiment, the inorganic used in the process according to the invention

Adsorbent on a high pore volume. According to a preferred embodiment, the inorganic

Adsorbent a pore volume of more than 0.1 ml / g, more preferably a pore volume of more than 0.2 ml / g. The pore volume is determined to be cumulative pore volume according to BJH (I.P. Barret, L.J. Joiner, P.P. Haienda, J. Am. Chem. Soc., 73, 1991, 373) for pores having a diameter of 1.7 to 300 mm. In one embodiment, the inorganic adsorbent has a pore volume of less than 1.4 ml / g. According to another embodiment of the method, the pore volume of the inorganic adsorbent is less than 1.3 ml / g and in another embodiment less than 1.2 ml / g.

As the alumina-containing component are preferred

Aluminum oxides, aluminum hydroxides, boehmite or aluminosilicates used. These compounds can be used in pure form or else in the form of mixtures of the compounds mentioned. As alumina, both α-, γ- and S-Al ₂ O ₃ can be used. According to one embodiment, aluminum oxides are used as the inorganic adsorbent, with γ-Al ₂ O _{3 being} more preferred. The aluminas can be used in pure form or as a mixture of two or three of said phases. It is possible to use aluminum hydroxides of different composition which also have a different degree of dehydration or

May have polymerization. Boehmite is AlOOH. For example, aluminas and their hydrates can be prepared by neutralizing basic aluminate solutions by the addition of acid. The precipitation of

Aluminum hydroxides or their hydrates can be supported by the addition of nuclei. It is also possible hydrated aluminum hydroxides from acidic solutions of aluminum salts by the addition of bases or by

Combination of basic solutions of aluminates with acidic solutions of aluminum salts. The of the

aqueous phase separated hydrated aluminas can then be converted to aluminas by annealing.

A suitable method for producing hydrated

Aluminum oxides are described, for example, in WO 01/02297

described. The aluminas and their hydrates can also have a

Sol-gel process can be produced. For example, aluminum alkoxides can be hydrolyzed for this purpose. The hydrolysis can generally in a temperature range of 30 to 150 ⁰ C.

be performed. The solid alumina hydrate is separated from the aqueous alcohol phase. The obtained

For example, crystals can be aged under hydrothermal conditions.

A suitable method is described for example in WO 00/09445. According to another preferred embodiment, the alumina-containing component is a synthetic

Aluminosilicate.

The aluminosilicates preferably have a silicon content, calculated as SiO ₂ , of less than 60% by weight, preferably less than 55% by weight, more preferably less than 50% by weight. on. According to one embodiment, the silicon content of the aluminosilicate, calculated as SiO ₂ , is more than 0.5% by weight, according to a further embodiment more than 0.75% by weight.

Aluminosilicates can be prepared, for example, by hydrolyzing organic aluminum compounds under acidic conditions and then together with silica or

Silica compounds are aged under hydrothermal conditions. Suitable aluminum compounds are

for example, aluminum alcoholates,

Aluminum hydroxyalcoholates, aluminum acetylacetonates,

Aluminum alkyl chlorides or aluminum carboxylates. A suitable method is described for example in US 6,245,310 Bl.

In order to obtain particularly pure aluminosilicates, hydrolyzable ones can also be used instead of silica

Organosilicon compounds are used, wherein the

Hydrolysis of organosilicon compounds and the

hydrolyzable aluminum compounds is carried out together. Such a method is described for example in EP 0 931 017 Bl.

Particular preference is given to using aluminosilicates which contain only SiO ₂ and Al ₂ O ₃ as constituents. The proportion of other metals, calculated as the most stable oxide, is

preferably chosen less than 1 wt .-%. The inorganic adsorbent may be provided, for example, in the form of a powder. The particle size of the powder is generally adjusted so that the

Inorganic adsorbents without difficulty with a suitable method, such as filtration, can be separated from the reaction mixture within a suitable period of time. In particular for an application of the inorganic

However, adsorbents in the form of a column pack are also suitable for larger particle sizes. For this purpose, the inorganic adsorbent is preferably in the form of granules

used. In particular for the production of

Saulenpackungen is preferably used a granule having a grain size of more than 0.1 mm. Preferably, the granules have a grain size in the range of 0.2 to 5 mm, particularly preferably 0.3 to 2 mm. The grain size can be adjusted, for example, by sieving.

The granules can be prepared by conventional methods, such as the finely ground inorganic

Adsorbent with a granulating agent, for example, water, applied and then in a conventional

Granulating device in a mechanically generated

Fluidized bed is granulated. However, other methods can be used to prepare the granules. Thus, the powdery inorganic adsorbent

be formed for example by compaction to a granulate.

The inorganic adsorbent can also be provided as a shaped article, which can be obtained, for example, by extrusion of a plastic mass. For this purpose, from the alumina-containing component and optionally other components, such as a binder by

Addition of a liquid, preferably water, a paste

produced. This paste is then extruded and the extrudate comminuted, for example by cutting the extruded strand into short cylindrical pieces, and then drying the resulting shaped articles. In addition to massive cylinders, for example, hollow cylinders can be produced in this way. After shaping, the granules or the shaped bodies can still be heat-treated and sintered, for example, by heating. As a result, the stability of the granules or the shaped bodies can be increased. For the heat treatment, the moldings or the granules are preferably at a temperature of more than 300 ⁰ C, according to another

Embodiment heated to a temperature of more than 400 ⁰ C, and according to yet another embodiment to a temperature of more than 500 ⁰ C. According to one embodiment, the temperature is less than 1200 ⁰ C, chosen according to another temperature lower than 1000 ⁰ C.

In order to obtain stable moldings, the heat treatment is preferred for a duration of at least 30 minutes, according to a further embodiment for a duration of at least 60

Minutes dialed. According to one embodiment, the

Treatment duration less than 5 hours chosen.

As already explained, the inorganic adsorbent can be added directly to the reaction mixture, wherein according to one embodiment it is agitated, for example shaken. According to a further embodiment, the inorganic

Adsorbent provided in the form of a column packing. The reaction mixture can then be passed through the column packing. The depleted in the second compound

Solvent may then be recycled back to the reaction batch in one embodiment.

The column packing may be provided, for example, in the form of a cartridge. In practice, the reaction mixture or the solvent from which the enzyme was previously separated, then be passed through the cartridge until the adsorption capacity of the in the

Cartridge contained inorganic adsorbent is exhausted. The cartridge can then be easily replaced with a new cartridge. The in the cartridge at

inorganic adsorbent adsorbed second compound can then be eluted with a suitable eluent. If the inorganic adsorbent in the form of a

Column packing provided, the inorganic

Larger grain adsorbents are provided to prevent excessive pressure drop across the column packing. In this embodiment, therefore, the inorganic

Adsorbent preferably used in the form of granules having a particle diameter of more than 0.1 mm,

particularly preferably has a particle diameter in the range of 0.2 to 5 mm. It may also be advantageous to use the inorganic

To heat adsorbent to a higher temperature to remove physically bound water. For this purpose, according to one embodiment, the inorganic adsorbent can be heated to a temperature of more than 600 ⁰ C. According to one embodiment, the temperature is chosen lower than 1000 ⁰ C. By means of such a treatment, for example, boehmite can first be used to remove the superficially bound water and to convert the hydrate into an oxide phase.

The invention will be explained in more detail by way of examples and with reference to the accompanying figures. The figures show in detail:

Fig. 1: a schematic representation of an apparatus for

Implementation of the method according to the invention; Fig. 2 is a graph showing the concentration of 2,5-hexanedione and 2,5-hexanediol bound from a solution in ethyl acetate to adsorbent 3; Fig. 3 is a graph showing the concentration of 2,5-hexanedione and 2,5-hexanediol bound from a solution in ethyl acetate to adsorbent 5;

Fig. 4: a diagram in which the concentration of 2,5-hexanedione and 2, 5-hexanediol is indicated, which consists of a solution in heptane / 10% 2-propanol

Adsorbent 3 is bound;

5 shows a chromatogram for a separation of 2,5-hexanedione and 2,5-hexanediol on adsorbent 5; Is a diagram in which the concentration of 2,5 hexanediol in the supernatant in the adsorption of 2,5-hexanediol from a mixture of toluene and 10% vv ^"1 2-propanol for various adsorbents shown, Figure 7: Fig. 6. Figure 3 is a graph showing the adsorption of NADH from an aqueous potassium phosphate buffer;

Fig. 8 is a graph showing adsorption of NADH from a mixture of a potassium phosphate aqueous buffer and isopropanol; 9 is a diagram in which the change of

Concentration of 2, 5-hexanedione and 2, 5-hexanediol during an enzymatically catalyzed reduction is shown, wherein the reaction without presence an inorganic adsorbent is carried out;

10 is a diagram in which the change of

Concentration of 2, 5-hexanedione and 2, 5-hexanediol during an enzymatically catalyzed reduction is shown, wherein the reaction is carried out in the presence of an inorganic adsorbent.

Fig. 1 shows a structure in which the method according to the invention can be carried out. In a reactor 1, there is contained a reaction mixture 2 containing an organic solvent in which a first compound comprising a prochiral keto group is dissolved, and a second compound comprising a hydroxy group bonded to a chiral carbon atom and those of the first Compound has emerged. Next, the mixture contains 2 a

Alcohol dehydrogenase and a sacrificial alcohol, for example isopropanol. Furthermore, the mixture 2 comprises a water phase in which a cofactor, for example NADP ⁺ , is dissolved. The mixture is agitated by a stirrer 3. Via a drain 4, the mixture 2 is removed from the reactor 1 and in a separator 5 by means of a membrane 6 in a proportion

split, which contains the enzyme and the one over

Return line 7 is returned to the reactor 1, and a proportion containing the organic solvent, but no enzyme, and via a feed line 8 a

Adsorberkartusche 9 is supplied. The adsorber cartridge 9 is packed with an inorganic adsorbent material,

For example, alumina, so that the second compound is partially adsorbed on the inorganic adsorbent material and thus the organic solvent is depleted in the second compound. The organic solvent, which still contains a portion of the first compound is then returned via return line 10 back to the reactor 1.

In order to be able to elute this on the product absorbed in the absorber cartridge 9, an eluent feed 11 and a drain 12 are provided, via which the feed with the

Product loaded eluants can be removed. For this purpose, the leading via the reactor 1 cycle of

Solvent stopped and rinsed the absorber cartridge 9 with eluent, which is fed via inlet 11 and discharged via outlet 12.

Determination of physical parameters

The physical properties of the inorganic

Adsorbent were prepared by the following methods

determined: BET surface area / pore volume after BJH and BET:

The surface and the pore volume were combined with a

fully automatic nitrogen porosimeter of the company

Micromeritics type ASAP 2010 determined.

The sample is cooled in a high vacuum to the temperature of liquid nitrogen. Subsequently, it becomes continuous

Nitrogen dosed into the sample chambers. By detecting the adsorbed amount of gas as a function of pressure, an adsorption isotherm is determined at a constant temperature. In a pressure equalization, the analysis gas is gradually

removed and recorded a desorption isotherm.

To determine the specific surface area and the porosity according to the BET theory, the data according to DIN 66131

evaluated. The pore volume is also determined from the measurement data using the BJH method (IP Barret, LG Joiner, PP

Haienda, J. Am. Chem. Soc. 73, 1991, 373). This procedure also takes into account effects of capillary condensation. Pore volumes of certain volume size ranges are determined by summing up incremental pore volumes obtained from the evaluation of the BJH adsorption isotherm. The total pore volume by BJH method refers to pores with a diameter of 1.7 to 300 nm.

The water content of the products at 105 ⁰ C is determined using the method DIN / ISO-787/2.

Loss on ignition:

In a calcined, weighed porcelain crucible with lid, approx. 1 g of dried sample is weighed to the nearest 0.1 mg and annealed for 2 hours at 1000 ^° C. in a muffle furnace. Thereafter, the crucible is cooled in a desiccator and weighed.

Determination of bulk density

A graduated cylinder cut off at the 1000 ml mark is weighed. Then, the sample to be examined is filled by means of a Pulvertrichters so in a train in the measuring cylinder that above the conclusion of the

Measuring cylinder forms a Schüttkegel. The pouring cone is made with the help of a ruler, which over the opening of the

Measuring cylinder is guided, stripped and the filled

Weigh the measuring cylinder again. The difference corresponds to the bulk density. Used inorganic adsorbent materials:

In the following examples, a number of synthetic alumina compounds were used. As a first group of commercially available aluminosilicates were tested under the names ^® Siral 5, Siral 30 ^®, ^® Siral 40

are available. Furthermore, a boehmite was tested, which is sold under the name Pural ^® SCC 150. As a further adsorbent aluminas were used, which are offered under the name Puralox ^® . These adsorbents were manufactured by Sasol Germany GmbH,

Ankelmannsplatz 1, 20537 Hamburg, DE, related. Finally, aluminas have been used as commonly used for chromatographic separation processes

(Manufacturer Macherey-Nagel, Düren, DE). As a comparative material, a commercially available silica gel for column chromatography was tested, which was obtained from Sigma-Aldrich. Silica gel Merck type 9385, 230-400 mesh, 60 A pore diameter,

EC: 231-545-4.

The following designations are used below for the inorganic adsorber materials investigated:

Table 1: Adsorber materials used

Adsorbent 1 Siral 5 ^® (Sasol, Hamburg, DE)

Adsorbent 2 Siral ^® 30 (Sasol, Hamburg, DE)

Adsorbent 3 Siral ^® 40 (Sasol, Hamburg, DE)

Adsorbent 4 Pural ^® SB (Sasol, Hamburg, DE)

Adsorbent 5 Puralox ^® KR-160 (Sasol, Hamburg, DE)

Adsorbent 6 Pural SCC ^® 150 (Sasol, Hamburg, DE)

Puralox ^® SCCa adsorbent 7-150/230 (Sasol, Hamburg, DE)

Adsorbent 8 aluminum oxide neutral (Macherey-Nagel, Düren,

DE)

Adsorbent 9 aluminum oxide basic (Macherey-Nagel, Düren,

DE)

Adsorbent 10 silica gel (Sigma-Aldrich)

The properties of the adsorbents are summarized in Table 2.

Stolmär Scheele & Partner 7th of June 2010

34282-SOUTH P-WO / DrJun

Table 2: Physical properties of the adsorbents (according to the manufacturer)

K)

Stolmär Scheele & Partner 7th of June 2010

34282-SUD-P-WO / DrJun

Table 2 (continued): Physical properties of the adsorbents

O

Example 1 Adsorption of 2,5-hexanedione and 2,5-hexanediol on inorganic adsorbents using different solvents

It was the adsorption of a mixture of 2, 5-hexanedione and 2, 5-hexanediol, which had been solved in each case in different solvents, to the adsorbent 3 and to the adsorbent 5 examined.

The following solvents were tested: a) distilled water

b) mixture of 18% vv ^"1 2-propanol and KPi buffer 100 mM,

pH 7.0

c) ethyl acetate

General procedure:

In each case, a solution was prepared from the solvents to be investigated, which had a specific, indicated in the examples concentration of 2, 5-hexanedione or 2, 5-hexanediol.

In each case 15 ml of the solution were added at room temperature (20 ° C) with a defined amount of the adsorbent. The addition of the adsorbent was either once or in several

Servings at a certain time interval. After addition of the adsorbent, the suspension was at 150 rpm on a

Rotationsschuttier incubated at room temperature. After a certain time, a sample was taken, the

Adsorbent separated by centrifugation and the supernatant analytically by gas chromatography or im

Measure spectrophotometer at 340 nm (on examination of the cofactor NADH). When measured in a gas chromatograph, the amount of sample was 70-100 μL. The experimental conditions used and the results are given in the following tables. a) Adsorption from water

The investigation was carried out according to the general rule given above. A solution containing 50 mmol of 2,5-hexanediol but no 2,5-hexanedione was used. At the beginning of the experiment, 3.0 g of adsorbent were used once

added. After sampling, the reaction mixture was further incubated at room temperature. The measured

Concentrations are summarized in Table 3.

Table 3: Adsorption of 2,5-hexanediol from water;

Concentration in the supernatant of the sample after

certain adsorption times; one-time addition of the adsorbent

b) Adsorption from KPi buffer / isopropanol

The investigation was made after the above

general rule. A solution containing 50 mM 2, 5-hexanediol but no 2,4-hexanedione was used. At the beginning of the experiment, 3.0 g of adsorbent was added. After 20 h, an additional 0.5 g of adsorbent was added. The results are summarized in Table 4. Table 4: Adsorption of 2,5-hexanediol from KP ₁ -Puffer /

isopropanol; Concentration in the supernatant of the sample after certain adsorption times; portionwise addition of the adsorbent

Also from a Kpi buffer no significant adsorption of the 2, 5-hexanediol to the investigated adsorbents. c) Adsorption from ethyl acetate

The test was carried out in accordance with the general instructions given above. In a first

A series of experiments was carried out with a stock solution of 51.45 mM 2,5-hexanedione and 48.65 mM 2, 5-hexanediol, in a second

Series of experiments using 50.68 mM 2, 5-hexanedione and 53.85 mM 2,5-hexanediol. At the beginning of the experiment, 0.5 g of adsorbent was added to the solution. At intervals of 30 minutes, in each case a further 0.5 g of the adsorbent was added.

The concentrations determined are shown in Tables 5a and 5b. Table 5a: Adsorption of 2,5-hexanediol / 2,5-hexanedione

ethyl acetate; 51.45 mM 2, 5-hexanedione and 48.65 mM 2, 5-hexanediol; Concentration in the supernatant of the sample after certain adsorption times; portionwise addition of the adsorbent

Table 5a (continued)

Table 5b: Adsorption of 2,5-hexanediol / 2,5-hexanedione

ethyl acetate; 50.68 mM 2, 5-hexanedione and 53.85 mM 2, 5-hexanediol; Concentration in the supernatant of the sample after certain adsorption times; portionwise addition of the adsorbent

Table 5b (continued)

Table 5c: adsorption of 2,5-hexanediol / 2,5-hexanedione

ethyl acetate; 51.23 mM 2, 5-hexanedione and 48.67 mM 2, 5-hexanediol; Concentration in the supernatant of the sample after certain adsorption times; in portions of the adsorbent at intervals of 30 minutes

In contrast to the aqueous or aqueous / organic medium, when using ethyl acetate as the solvent, the diol was preferably adsorbed. This can be for both boehmite,

Boehmit / SiO ₂ mixed phases, γ-Al ₂ O ₃ , as well as for the amorphous Al oxides are shown.

By way of example, the adsorption capacities for two of the carriers under the experimental conditions given are shown graphically in FIGS. 2 and 3.

Example 2 Adsorption of 2,5-hexanedione and 2,5-hexanediol on inorganic adsorbents in the presence of another polar solvent

The experiment was carried out analogously to the general procedure of Example 1. As a solvent for adsorption of the mixture of hexanediol / hexanedione heptane with 10% 2-propanol used. 2-Propanol is used as a sacrificial alcohol to reduce NADH or NADP in enzymatic reactions.

A stamen solution was prepared containing 70 mM 2,5-hexanedione and 50 mM 2, 5-hexanediol in heptane / 10% 2-propanol. From this stock solution was added by adding

Heptane / 10% 2-propanol prepared a dilution series.

In glass vials 200 mg of adsorbent 3 were weighed in each case and mixed with 1 ml of the respective mixture from the dilution series. The samples were then incubated for 60 minutes on the rotary shaker (150 rpm) at room temperature. Subsequently, a sample was taken from each sample vessel, the adsorbent by centrifugation at 13,000 rpm / 1 min.

separated and transferred 100 .mu.l of the supernatant into a GC vial with Inlet. The examination of the samples was carried out by means of

Gas chromatography.

The gas chromatographically determined concentrations and the capacity of the carrier derived therefrom are summarized in Table 6.

Table 6: Adsorption of 2,5-hexanediol / 2,5-hexanedione

Heptane with 10% 2-propanol on adsorbent 3

Also in the presence of 10% of the further polar solvent 2-propanol is a selective adsorption of the target alcohol hexanediol. By way of example, the adsorption capacity is shown in FIG. 4.

Example 3: Use of the Adsorbent 5 in a

chromatography column

2, 5-hexanedione and 2, 5-hexanediol in a concentration of 50 mM were dissolved in ethyl acetate and to a 4.1 g

Adsorbent 5 packed and equilibrated with ethyl acetate column (length 24 cm, diameter: 0.7 cm) was added. The mixture was then eluted with ethyl acetate as eluent. The 2,5-hexanedione was first eluted from the column and thereby completely separated from the 2,5-hexanediol. The chromatogram with the elution peak for 2,5-hexanedione is shown in FIG. 5

shown. The eluent for the 2,5-hexanediol is methanol or acetonitrile.

Example 4: Adsorption tests from toluene

There was a solution of 50 mM 2, 5-hexanediol in toluene

and to 15 ml of this solution 0.5 g of the zu

given examining adsorbent. Subsequently, the

Suspension at 150 rpm on the rotary shaker at

Room temperature incubated. After 30 minutes, the adsorbent was separated by centrifugation at 4 ° C for 5 minutes. From the clear supernatant, 100 μL was placed in a sample jar

transferred. The remainder of the supernatant was added to 0.5 g of fresh adsorbent and incubated at the above conditions for an additional 30 minutes. The adsorbent was again separated by centrifugation and the supernatant

measured by gas chromatography. In the further course of the

Experiments were added every 30 minutes another 0.5 g of the adsorbent and the suspension for a further 30 minutes at

Room temperature incubated. This process was repeated two more times, in each case before the addition of the adsorbent, the concentration of 2, 5-hexanediol was determined in the supernatant in the manner described above. The experimental data and the determined concentrations are shown in Table 7 played. The decrease in the concentration of 2,5-hexanediol in the supernatant is also shown graphically in FIG.

With several of the adsorbents, the concentration of 2, 5-hexanediol in the supernatant of 50.00 mM could be reduced to as much as 10.21 mM when using 2.0 g.

Stolmär Scheele S. Partner 7th of June 2010

34282-SUD-P-WO / DrJun

Table 7: Adsorption of 2, 5-hexanediol from toluene to adsorbent 3, adsorbent 5, adsorbent 8 and

Adsorbent 9

Example 5: Adsorption of the Cofactor NADH

To determine the suitability of the adsorbents for an enzymatic

To test the conversion of ketones into chiral alcohols using the enzyme ADH- ¹ A 'from Rhodococcus ruber DSM 44541, the adsorption of the cofactor NADH used in this system was investigated.

In order to investigate the influence of the adsorbents on the pH of the solutions, initially 3 g of adsorbent 3 or 1.5 g of adsorbent 5 were added to 15 ml each of potassium phosphate buffer (100 mM) or distilled water. The suspensions were incubated at room temperature at 150 rpm for 60 minutes

Reaction detractor moves. The adsorbent was through

Centrifugation separated and the pH of the clear

Supernatant determined with a glass electrode. Subsequently, the adsorbent was in fresh suspending agent

slurried and incubated again at the above conditions for 60 minutes. This process has been repeated

repeated. The values obtained in the treatment with water are given in Table 8 and in the case of treatment with

Potassium phosphate buffer values given in Table 9.

Table 8: Change in the pH value by stirring the adsorbents with distilled water

Table 9: Change of the pH by coils of the adsorbents with enzyme buffer (KP ₁ -PUffer)

For the measurement of the adsorption of NADH to the adsorbent 3 or the adsorbent 5, 20 ml of the reaction mixture indicated in Table 10 were added to 4 g of adsorbent and the suspension was incubated at 150 rpm on a rotary vaporizer at room temperature. Samples were taken at regular intervals, the carriers were separated by centrifugation and the supernatant was measured photometrically at 340 nm. The determined concentrations are shown graphically in FIGS. 7 (approach 1) and 8 (approach 2).

Table 10: Composition of reaction mixtures for

Investigation of the adsorption of NADH on different adsorbents

As shown in Figures 7 and 8, the cofactor is not adsorbed by the adsorbents in significant amounts. Example 6: Carrying out an enzymatically catalyzed

Reduction of 2, 5-hexanedione in the presence of adsorbents

To study the influence of adsorbents on the enzymatic reduction of 2, 5-hexanedione a supported alcohol was ADH 'A' used, as carrier for the enzyme porous glass particles TRISOPERL ^® (VitraBio, Steinach, D) were used. The immobilization of the ADH- ¹ A 'on the glass particles is described in Goldberg, K.; Krueger, A .; Meinhardt, T .; Kroutil, W .; Mautner, B. & Liese, A. (2008), Novel immobilization routes for the covalent binding of alcohol dehydrogenase from Rhodococcus ruber DSM 44541 ', Tetrahedron: Asymmetry 19 (10), 1171-1173.

The cofactor was used as the stock solution, but 14.2 mg NADH were dissolved in 200 μL KPi buffer. The reaction medium (toluene) was saturated with KPi buffer, for 30 mL toluene were mixed with 10 mL KPi buffer (100 mM, pH 7.0), the toluene phase was removed and used in the experiment.

First, the supported enzyme was diluted with 100 μL NADH

Mixed stock solution. Next, 8.8 mL of toluene

(saturated with buffer) and 1 mL 2-propanol (10% vv ^"1 )

added. The reaction was started by adding 97 μL hexanedione (80 mM). The sample was incubated at 30 ^° C. and 150 rpm on the rotary shaker.

For the determination of the concentration of 2, 5-hexanedione and 2,5-hexanediol, the reaction mixture in certain

time intervals each 50 μL sample taken. The sample was diluted with 50 μL of toluene and measured by gas chromatography.

In a first series of experiments, the conversion of the reaction was investigated without addition of adsorbent. In these conditions the enzyme was intact and the concentration of 2,5-hexanedione decreased with increasing reaction time while 2,5-hexanediol formed. The concentration of 2, 5-hexanedione or 2, 5-hexanediol is shown graphically in Figure 9 as a function of time.

In a second series of reactions, the experiment was repeated, but in addition 3 g of adsorbent 5 were added. The decrease in the diketone concentration in the supernatant over time also indicates a functioning enzymatic conversion, the concentration of 2, 5-hexanediol in the supernatant was lower. It can be assumed that that

enzymatic product is bound to the adsorbent 5. The concentration of 2, 5-hexanedione or 2, 5-hexanediol is graphically represented in Figure 10 as a function of time. Example 7: Performance of an enzymatically catalyzed

Reduction of 2, 5-hexanedione with subsequent work-up of the product alcohol

It was in a thermostatable reaction flask, a 2-phase system consisting of an aqueous phase ADH-A

Crude extract in potassium phosphate buffer (0.1 M, pH 7.0) and 1 mM NADH and an organic phase toluene + 10% v / v 2-propanol (for the regeneration of the cofactor NADH in a

Substrate-coupled approach) submitted. To start the

Enzyme reaction was initially introduced as starting material 80 mM 2, 5-hexanedione and converted to the product 2, 5-hexanediol. The

Reaction temperature was 30 ⁰ C, stirred only the upper (organic) toluene phase with a Clipfish stirrer at 100 rpm.

As a control on product formation, the organic phase was measured in the GC. After the detection of a high

Product concentration, the reaction was stopped and the organic phase decreased. As reference values (O-samples) final concentrations of starting material / product were measured.

Subsequently, 3 ml of the organic phase were added to 0.5 g of initially introduced carriers and incubated for 30 minutes at 150 rpm at RT on a rotary shaker.

Table 11: Treatment of the organic phase with the adsorbent

3 and 5, initial concentrations after enzymatic reaction (blank samples) and concentrations after treatment with 0.5 g of adsorbent

LIST OF REFERENCE NUMBERS

1 reactor

2 reaction mixture

3 stirrer

4 outflow

5 separator

6 membrane

7 return

8 supply line

9 Adsorber cartridge

10 return

11 feed eluent

12 drain eluent

Claims

claims

1. A process for separating a mixture with one in one

Solvent-solubilized first compound comprising at least one keto group, and a second compound comprising a chiral carbon atom having a hydroxy group formed by stereoselective reduction of the keto group from the first compound, wherein the

Mixture with an inorganic adsorber material in

Contact is brought and at least part of the second

Compound is adsorbed on the inorganic adsorbent material, and a mixture depleted in the second compound is separated from the inorganic adsorbent material.

2. The method of claim 1, wherein the solvent is a

is organic solvent.

3. The method according to claim 2, wherein the organic

Solvent with water is substantially immiscible.

4. The method according to any one of claims 1 to 3, wherein the

Reduction of the keto group to that of a chiral one

Carbon atom bonded hydroxy group is carried out under catalysis by an enzyme.

5. The method of claim 4, wherein the reduction of

Keto group to that at a chiral carbon atom

bonded hydroxy group is carried by the enzyme in the presence of a cofactor.

6. The method according to claim 4 or 5, wherein the enzyme is an alcohol dehydrogenase.

7. The method of claim 6, wherein the cofactor in a

Water phase is dissolved and the water phase in addition to the organic solvent in an amount of less than 1 wt .-%, based on the organic solvent in the mixture is included.

8. The method of claim 6 or 7, wherein the mixture contains a sacrificial alcohol for the regeneration of the cofactor.

9. The method of claim 8, wherein the sacrificial alcohol

has lower molecular weight than the second

Connection.

10. The method according to any one of the preceding claims, wherein the inorganic adsorbent is an achiral

Adsorbent is.

A process according to any one of the preceding claims, wherein the reduction of the keto group to a hydroxy group bonded to a chiral carbon atom is carried out in the presence of the inorganic adsorbent.

12. The method according to any one of the preceding claims, wherein the inorganic adsorbent has an aluminum content, calculated as Al ₂ O ₃ of more than 40 wt .-%.

13. The method according to any one of the preceding claims, wherein the inorganic adsorbent is a specific

Surface of more than 100 m ² / g.

14. The method according to any one of the preceding claims, wherein the inorganic adsorber material has a pore volume of more as 0.2 ml / g, determined as cumulative pore volume according to BJH for pores with a diameter of 1.7 to 300 nm.

15. The method according to any one of the preceding claims, wherein the inorganic adsorbent material is an aluminum silicate

16. The method according to any one of the preceding claims, wherein after the separation of the second compound

depleted mixture of the inorganic

Adsorber material adsorbed second compound is desorbed from the inorganic adsorber.