Method for the continuous production of ketones from epoxides in a fixed bed

Abstract

A method is useful for the continuous production of ketones from a compound with at least one epoxide group in at least one fixed bed reactor. A catalyst composition is used with at least one noble metal and at least one metal oxide. To reduce the proportions of high-boilers which form in the reaction, an inert gas is introduced so that a carbon monoxide partial pressure of 50 mbar or less is set in the reactor.

Claims

1. A method, comprising: continuously producing a ketone from a compound comprising at least one epoxide group, in a device comprising at least one fixed bed reactor, wherein the at least one fixed bed reactor contains a catalyst composition comprising at least one noble metal and at least one metal oxide, wherein at least one non-reactive gas is introduced in the reactor and wherein a. the pressure in the reactor upstream of the catalyst is at least 1.2 bar, b. the gas phase of the reactor has a maximun hydrogen partial pressure of 0.2 bar, and c. a carbon monoxide partial pressure of 50 mbar or less is set in a gas phase of the reactor downstream of the catalyst by introducing at least one inert gas.

2. The method according to claim 1, wherein a carbon monoxide partial pressure of 30 mbar or less is set in the gas phase of the fixed bed reactor.

3. The method according to claim 1, wherein the metal oxide of the catalyst system comprises titanium dioxide, zirconium dioxide, or mixtures thereof, or consists of titanium dioxide, zirconium dioxide, or mixtures thereof.

4. The method according to claim 1, wherein the total pressure in the reactor is 4 bar or less.

5. The method according to claim 1, wherein a ratio of the amount of inert gas and the amount of the compound comprising epoxide group is at least 0.5.

6. The method according to claim 1, wherein at least one further fixed bed reactor is downstream of the fixed bed reactor.

7. The method according to claim 6, wherein offgas is introduced in countercurrent between the fixed bed reactors.

8. The method according to claim 1, wherein the compound comprising at least one epoxide group is a cycloaliphatic compound having 4 to 20 carbon atoms.

9. The method according to claim 7, wherein the compound comprising at least one epoxide group comprises monoepoxycyclododecane.

10. A method for synthesizing a lactam, the method comprising: rearranging at least one compound comprising an epoxide group to a ketone, b. oximating the ketone to an oxime, and c. rearranging, by Beckmann rearrangement, the oxime to the lactam, wherein the rearrangement a. is conducted according to the method of claim 1.

11. The method according to claim 1, wherein the gas phase of the reactor in b. has no hydrogen.

12.The method according to claim 1, wherein the total pressure in the reactor is 2 bar or less.

13. The method according to claim 1, wherein a proportion of high-boiling by-products is less than 5% by weight, based on the epoxide converted.

Description

[0052] FIG. 1 shows a device (6) having a circulation reactor, i.e. a fixed bed reactor (3) which is operated in a closed loop. The feed (1) is introduced into the system at this point. Inert so gas (2) is metered in prior to the fixed bed reactor(s) (3) comprising the catalyst system. Offgas (4) and the liquid product mixture (5) are separated in a separator (7). The offgas is removed from the system, The product mixture is in part fed back to the fixed bed reactor(s) and in part to the next method step. The pressure in the device (6) is increased by a pump (8). The pressure between the pump and catalyst, which is located in the fixed bed reactor (3), is generally elevated compared to other regions of the device. This corresponds to the aforementioned pressure in the reactor upstream of the catalyst which is at least 1.2 bar.

[0053] A device (6) having a pump (8) and a tubular reactor as the fixed bed reactor (3) is depicted in FIG. 2 in which the feed (1) is fed and inert gas (2) is added prior to the fixed bed reactor (3). Offgas (4) and product (5) are separated in the separator (7).

[0054] The process according to the invention can be carried out in organic solvents, it being preferred to work without solvents and thus to use no organic solvents. Suitable solvents are, for example, alkanes such as n-hexane, n-heptane, n-tetradecane and cyclohexane; ethers such as tetrahydrofuran and dioxane; alkanols such as methanol, ethanol and t-butanol; esters such as ethyl acetate and butyl acetate. The solvents can be used on their own or in mixtures. The solvent is preferably used in an amount which is 20 times or less than, preferably 10 times or less than, the weight of compound E.

[0055] In a preferred embodiment of the invention, monoepoxycyclododecane is converted continuously to cyclododecanone with a fixed bed catalyst without solvent at temperatures of 170 to 250° C., wherein the partial pressure of carbon monoxide in the gas phase of the reactor is controlled downstream of the catalyst below 50 mbar. The partial pressure of carbon monoxide is determined, for example, by measuring the total pressure using a manometer and by measuring the proportion of carbon monoxide by means of FT-IR spectrometry.

[0056] The invention further provides a process for the synthesis of lactams (lactam process according to the invention), in which the aforementioned process according to the invention for producing ketones is used; firstly, rearrangement of a compound (compound E) comprising at least one epoxide group to the ketone takes place. Then, oximation of the ketone to the oxime is carried out. Subsequently, Beckmann rearrangement of the oxime to the lactam takes place. The compound E is preferably selected from aliphatic monoepoxycycloalkanes, aliphatic rnonoepoxycycloalkanedienes and aliphatic monoepoxycycloalkenes, with monoepoxycycloalkanes being preferred.

[0057] If the ketone is present in a mixture with the corresponding alcohol derivative, a dehydrogenation of the alcohol to the ketone can take place. The Beckmann rearrangement may be carried out using sulphuric acid or cyanuric chloride. The lactams may be subjected to further processing by polycondensation to give polyamides.

[0058] The dehydrogenation, the oximation, the Beckmann rearrangement and the condensation reaction are known to the person skilled in the art.

[0059] In a preferred embodiment of the lactam process according to the invention, laurolactam is prepared from monoepoxycyclododecane (or cyclododecane epoxide or 1,2-cyclododecane epoxide).

[0060] In the context of the preferred lactam method, monoepoxycyclododecane is obtainable by the following reaction steps: 1,3-butadiene is reacted to give cyclododecatriene by cyclotrimerization. This is followed by a hydrogenation to give the cyclododecene. The cyclododecane epoxide is obtained by subsequent epoxidation. The person skilled in the art in the field of the synthesis of organic compounds can prepare other aliphatic and cycloaliphatic compounds E analogously to the synthesis of monoepoxycyclododecane.

[0061] The present invention is more particularly elucidated hereinbelow with reference to examples. Alternative embodiments of the present invention are obtainable analogously.

LIST OF REFERENCE NUMERALS

[0062] 1 Feed [0063] 2 Inert gas [0064] 3 Fixed bed reactor [0065] 4 Offgas [0066] 5 Product [0067] 6 Device [0068] 7Separator [0069] 8 Pump

EXAMPLES

[0070] The percentages in the case of catalysts give the weight fraction of the noble metal, based on the total weight of the catalyst comprising noble metal and support. The abbreviation “calc.” stands for “calcined”. The abbreviations for the substances are: CDAN: Cyclododecene; CDEN: Cyclododecene; ECD: Epoxycyclododecane; CDON: Cyclododecanone; CDENON: Cyclododecenone (isomer mixture); CDOL: Cyclododecanol; CDENOL: Cyclododecenol (isomer mixture).

[0071] The catalyst system used consisted of a ZrO.sub.2-SiO.sub.2 mixed oxide (95% ZrO.sub.2, 5% SiO.sub.2) and a 0.5% Pd/SiO.sub.2 catalyst. Both catalysts were produced in accordance with EP3006107 (mixed oxide corresponding to Example B and Pd/SiO.sub.2 corresponding to Example D).

[0072] Gas chromatography (GC): Gas chromatographic investigations were carried out using a GC-2010 (Shirnadzu) chromatograph, fitted with autosampler, flame ionization detector (FID), and GC capillary column Supelcowax® (60 m×0.32 mm×0.25 μm, Supelco). Measurements were carried out in the split mode (Split rate 1:66) with helium as carrier is gas (flow rate 0.89 ml/min, linear carrier gas rate 17.9 cm/s). Temperature programme for GC oven: Start temperature 150° C.; heat to 180° C. at 5° C./min, hold for 10 min; heat to 200° C. at 5° C./min, hold for 10 min. Detector and injector temperatures were 340° C. and 220° C.

[0073] By adding an external standard (tetradecane) to each sample and applying the factor method, the composition of the reaction mixture was calculated in % by weight. Using the molar mass of each substance, the composition of the mixture could then be calculated in mol %. The conversion of the epoxide could then be calculated. The selectivity of each product was calculated on the basis of the difference in concentration of this product in the reaction mixture and in the reactant, based on the reacted epoxide. For the high boilers, the selectivity was calculated on the basis of the molar mass of the epoxide, which provides a statement about the amount of epoxide which had been converted to high boilers (loss of selectivity during the reaction).

[0074] The proportion of carbon monoxide was determined by IR spectrometry by introducing the offgas from the reactor into an IR spectrometer. The spectroscopic measurements were carried out using a Gasmet DX4000 Fourier transform (FT) mid-IR spectrometer from Ansyco, which records the absorption in the spectral range of 600-4200 cm.sup.−1. The gaseous material stream from the reactor was fed via appropriate lines to the measurement cell of the spectrometer and secured by means of a Swagelok fitting. To avoid condensation in the material stream, the lines were heated to 110° C. by means of electrical trace heating. In the mid-IR spectrometer used, a ceramic Si-C material (Globar) served as mid-IR source. The measurement cell had a volume of 0.45 L and an optical path length of 500 cm, which was accomplished via multiple reflections. The measurement cell was also heated to a temperature of 110° C. A thermoelectrically cooled MCT detector served as detector. The CO determination was calibrated by reference spectra starting from test gases (range: 0.9 to 7 vol% carbon monoxide in nitrogen). For this purpose, the spectral range of the CO signal from 1850-2060 cm.sup.−1 was evaluated, since no interferences with other components were observed here. The calibration and the measurements were carried out using the Calcmet software from Ansyco. The measurements of the gas phase of the reactor were carried out against pure nitrogen as background spectrum. The measurement durations used were 60 s with a measurement interval of 1800 s.

[0075] The figures for the mass flow of the nitrogen are specified for differentiation to the volume flow rate in NL/h. DIN 1343 is used as standard (back pressure 1013.25 mbar, gas temperature 0° C.).

Example 1

Non-Inventive

[0076] The reaction was carried out in a laboratory scale plant. The system consisted of two fixed-bed reactors in series (ca. 200 ml per reactor) and a storage container (1 L). The lower fixed bed reactor was filled with 45 g of ZrO.sub.2-SiO.sub.2 mixed oxide (95% ZrO.sub.2, 5% SiO.sub.2) and the upper fixed bed reactor with 90 g of 0.5% Pd/SiO.sub.2, The container was filled with 1000 g of cyclododecanone. The liquid was pumped in a cycle from the storage container through the catalyst bed back to the storage container using a circulating pump (10 I/h). The reactors were heated to an internal temperature of 185° C. in the reaction mixture using a thermostat.

[0077] Then, 65 g/h of feed comprising 85.4% by weight epoxycyclododecane, 9.3% by weight CDAN and 5.3% CDEN were metered continuously into the circuit. This corresponds to a metered addition of 0.31 mol/h of epoxycyclododecane. Product was discharged continuously from the system via an overflow tube such that the fill level in the storage container remained constant. In addition, 2 NL/h of nitrogen were metered continuously into the system, which corresponds to an amount of 0.08 mol/h of N.sub.2. Thus, the ratio of the amount of nitrogen to the amount of epoxycyclododecane was 0.26. Using a supply pressure regulator, a total pressure in the system of 3.2 bar was set and the offgas was discharged continuously from the system.

[0078] After a run time of 48 h, the reaction mixture was in steady state. A conversion of the epoxide of 77% was achieved. The selectivity for CDON was only 6.8 mol % and 17 mol % of high boilers were formed.

[0079] The proportion of carbon monoxide in the offgas was determined by IR spectrometry and was 2.6 mol %. At a total pressure of 3.2 bar, a partial pressure of carbon monoxide of 83 mbar was attained.

TABLE-US-00001 TABLE 1 Conversion of epoxide (mol %, GC with external standard) and selectivity for various products (mol %, GC with external standard) Conversion Selectivity (mol %) of epoxide Cycloundecane + High (mol %) cycloundecene CDAN CDEN CDON CDENON CDOL CDENOL boilers 77 0.2 0.1 6.5 6.8 8.3 1.0 60.1 17 CDAN, CDEN, CDON, CDENON, CDOL and CDENOL are substances which can all be converted to CDON by known methods. They are therefore all utilizable products. On this basis, the selectivity for utilizable products is 82.8 mol %.

Example 2

Non-Inventive

[0080] The reaction was carried out in a laboratory scale plant in accordance with Example 1. 65 g/h of feed comprising 85.4% by weight epoxycyclododecane. 9.3% by weight CDAN and 5.3% CDEN were metered continuously into the circuit. This corresponds to a metered addition of 0.31 mol/h of epoxycyclododecane. Product was discharged continuously from the system via an overflow tube such that the fill level in the storage container remained constant. In addition, 5 NL/h of nitrogen were metered continuously into the system, which corresponds to an amount of 0.20 mol/h of N.sub.2, Thus, the ratio of the amount of nitrogen to the amount of epoxycyclododecane was 0.67. Using a supply pressure regulator, a total pressure in the system of 3.2 bar was set and the offgas was discharged continuously from the system.

[0081] After a run time of 48 h, the reaction mixture was in steady state. A conversion of the epoxide of 78% was achieved. The selectivity for CDON was 37 mol % and 16.5 mol % of high boilers were formed.

[0082] The proportion of carbon monoxide in the offgas was determined by IR spectrometry and was 1.83 mol %. At a total pressure of 3.2 bar, a partial pressure of carbon monoxide of 59 mbar was attained.

TABLE-US-00002 TABLE 2 Conversion of epoxide (mol %, GC with external standard) and selectivity for various products (mol %, GC with external standard) Conversion Selectivity (mol %) of epoxide Cycloundecane + High (mol %) cycloundecene CDAN CDEN CDON CDENON CDOL CDENOL boilers 78 1.0 0.1 2.0 37.4 11.3 3.4 28.3 16.5

[0083] The selectivity for utilizable products was 82.5 mol %.

Example 3

Inventive

[0084] The reaction was carried out in a laboratory scale plant in accordance with Example 1. 65 g/h of feed comprising 85.4% by weight epoxycyclododecane, 9.3% by weight CDAN and 5.3% CDEN were metered continuously into the circuit. This corresponds to a metered addition of 0.31 mall of epoxycyclododecane. Product was discharged continuously from the system via an overflow tube such that the fill level in the storage container remained constant. In addition, 20 NL/h of nitrogen were metered continuously into the system, which corresponds to an amount of 0.81 mon of N.sub.2. Thus, the ratio of the amount of nitrogen to the amount of epoxycyclododecane was 2.6. Using a supply pressure regulator. a total pressure in the system of 3.2 bar was set and the offgas was discharged continuously from the system.

[0085] After a run time of 48 h, the reaction mixture was in steady state. A conversion of the epoxide of 79% was achieved. The selectivity for CDON was 90 mol % and only 2 mol % of high boilers were formed.

[0086] The proportion of carbon monoxide in the offgas was determined by IR spectrometry and was 0.85 mol %. At a total pressure of 3.2 bar, a partial pressure of carbon monoxide of 27 mbar was attained.

TABLE-US-00003 TABLE 3 Conversion of epoxide (mol %, GC with external standard) and selectivity for various products (mol %, GC with external standard) Conversion Selectivity (mol %) of epoxide Cycloundecane + High (mol %) cycloundecene CDAN CDEN CDON CDENON CDOL CDENOL boilers 79 1.4 0.5 0.1 89.9 3.6 1.4 1.1 2.0

[0087] The selectivity for utilizable products was 97.6 mol %.

Example 4

Inventive

[0088] The reaction was carried out in a laboratory scale plant in accordance with Example 1. 65 g/h of feed comprising 85.4% by weight epoxycyclododecane, 9.3% by weight CDAN and 5.3% CDEN were metered continuously into the circuit. This corresponds to a metered addition of 0.31 mol/h of epoxycyclododecane. Product was discharged continuously from the system via an overflow tube such that the fill level in the storage container remained constant. In addition, 5 NL/h of nitrogen were metered continuously into the system, which corresponds to an amount of 0.20 mol/h of N.sub.2. Thus, the ratio of the amount of nitrogen to the amount of epoxycyclododecane is 0.67. Using a supply pressure regulator. a total pressure in the system of 1.2 bar was set and the offgas was discharged continuously from the system.

[0089] After a run time of 48 h, the reaction mixture was in steady state. A conversion of the epoxide of 78% was achieved. The selectivity for CDON was 89 mol % and only 2.7 mol % of high boilers were formed.

[0090] The proportion of carbon monoxide in the offgas was determined by IR spectrometry and was 1.71 mol %. At a total pressure of 1.2 bar, a partial pressure of carbon monoxide of 21 mbar was attained.

TABLE-US-00004 TABLE 4 Conversion of epoxide (mol %, GC with external standard) and selectivity for various products (mol %, GC with external standard) Conversion Selectivity (mol %) of epoxide Cycloundecane + High (mol %) cycloundecene CDAN CDEN CDON CDENON CDOL CDENOL boilers 77 1.5 0.8 0.1 88.5 5.2 0.1 1.0 2.9

[0091] The selectivity for utilizable products was 95.6 mol %.

Example 5

Inventive

[0092] The reaction was carried out in a laboratory scale plant in accordance with Example 1. 65 g/h of feed comprising 85.4% by weight epoxycyclododecane, 9.3% by weight CDAN and 5.3% CDEN were metered continuously into the circuit. This corresponds to a metered addition of 0.31 mol/h of epoxycyclododecane. Product was discharged continuously from the system via an overflow tube such that the fill level in the storage container remained constant. In addition, 20 NL/h of nitrogen were metered continuously into the system, which corresponds to an amount of 0.81 mol/h of N.sub.2. Thus, the ratio of the amount of nitrogen to the amount of epoxycyclododecane is 2.6. Using a supply pressure regulator, a total pressure in the system of 1.2 bar was set and the offgas was discharged continuously from the system.

[0093] After a run time of 48 h, the reaction mixture was in steady state. A conversion of the epoxide of 76% was achieved. The selectivity for CDON was 90 mol % and only 2.3 mol % of high boilers were formed.

[0094] The proportion of carbon monoxide in the offgas was determined by IR spectrometry and was 0.70 mol %. At a total pressure of 1.2 bar, a partial pressure of carbon monoxide of 8 mbar was attained.

TABLE-US-00005 TABLE 5 Conversion of epoxide (mol %, GC with external standard) and selectivity for various products (mol %, GC with external standard) Conversion Selectivity (mol %) of epoxide Cycloundecane + High (mol %) cycloundecene CDAN CDEN CDON CDENON CDOL CDENOL boilers 76 1.6 0.5 0.1 90.2 3.6 1.0 0.7 2.3

[0095] The selectivity for utilizable products was 96.1 mol %.

Result

[0096] It could be demonstrated by Examples 1 to 5 that reducing the CO partial pressure to below 50 mbar, by means of an inert gas mixture, substantially reduces the proportion of high boilers as by-product of the conversion of the epoxide to the ketone.

TABLE-US-00006 TABLE 6 Overview of the partial pressures and the proportions of reaction products resulting therefrom. Total Partial pressure in Selectivity in Selectivity pressure the reactor mol-% mol % Example (CO) in mbar in bar CDON High boilers 1 83 3.2 6.8 17 2 59 3.2 37.4 16.5 3* 27 3.2 89.9 2 4* 21 1.2 88.5 2.9 5* 8 1.2 90.2 2.3 *inventive