Beckmann rearrangement of cyclohexanone oxime to ε-caprolactam catalyzed by sulfonic acid resin in DMSO

Beckmann rearrangement of cyclohexanone oxime to ε-caprolactam catalyzed by sulfonic acid resin in DMSO

Available online at www.sciencedirect.com Catalysis Communications 9 (2008) 1521–1526 www.elsevier.com/locate/catcom Beckmann rearrangement of cyclo...

131KB Sizes 0 Downloads 72 Views

Available online at www.sciencedirect.com

Catalysis Communications 9 (2008) 1521–1526 www.elsevier.com/locate/catcom

Beckmann rearrangement of cyclohexanone oxime to e-caprolactam catalyzed by sulfonic acid resin in DMSO Kuiyi You a, Liqiu Mao b, Dulin Yin b,*, Pingle Liu a, He’an Luo a,* b

a College of Chemical Engineering, Xiangtan University, Xiangtan 411105, PR China Institute of Fine Catalysis and Synthesis, Hunan Normal University, Changsha 410081, PR China

Received 9 October 2007; received in revised form 9 January 2008; accepted 9 January 2008 Available online 15 January 2008

Abstract The catalytic liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam over solid sulfonic acid resin in dimethyl sulfoxide (DMSO) has been successfully performed. The influences of solvent, catalyst, reaction temperature and reaction time on the rearrangement of cyclohexanone oxime in the liquid-phase were examined. The results indicate that conversion of oxime can reach 100% with 97.9% of selectivity to e-caprolactam under optimal reaction conditions. This catalytic system involved is environmentally harmless, easily separate, and the catalyst can be conveniently recovered for recycled use. Ó 2008 Elsevier B.V. All rights reserved. Keywords: Cyclohexanone oxime; e-Caprolactam; Liquid-phase Beckmann rearrangement; Solid sulfonic acid resin; DMSO

1. Introduction e-Caprolactam is an important monomer for the production of nylon-6 synthetic fibers and resins. Most of the current caprolactam production methods involve the conversion of cyclohexanone with hydroxylamine (in its sulfate or phosphate form) into cyclohexanone oxime, followed by Beckmann rearrangement by the action of fuming sulfuric acid, and then treatment with ammonia giving ecaprolactam [1,2]. However, this process suffers from many problems such as the formation of large quantities of lowvalue ammonium sulfate, the corrosion of equipment and environmental pollution caused by the use of fuming sulfuric acid. To overcome these problems, many research groups have attempted to carry out vapor-phase Beckmann rearrangement using solid acid catalysts instead of fuming sul* Corresponding authors. Tel./fax: +86 732 8293545 (H. Luo); Tel./fax: +86 731 8872576 (D. Yin). E-mail addresses: [email protected] (D. Yin), [email protected] (H. Luo).

1566-7367/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2008.01.011

furic acid [3–12]. It has been suggested that various hydroxyl groups and silanol nests on the surface of the zeolite catalysts enhance the formation of e-caprolactam [13– 16]. The vapor-phase reaction, however, suffers from its intrinsic features such as the requirement of high temperature and the need to use a fluidized bed due to coke formation on the catalyst. As for the liquid-phase Beckmann rearrangement process, it usually proceeded under milder conditions and could afford various amides, and good results have been obtained by using cyanuric chloride/DMF [17], sulfamic acid [18], anhydrous oxalic acid [19], solid metaboric acid [20], boron trifluoride etherate [21], tetrabutylammonium perrhenate [22] and chlorosulfonic acid–dimethyl formamide reagent [23] etc., as catalysts. Recently, in development of such liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam process, the use of room temperature ionic liquids as environmentally friendly media for catalytic processes is widely recognized and accepted [24–27]. The Beckmann rearrangement was found to be efficiently progressed using ionic liquids in the presence of PCl5, PCl3 or P2O5 as catalyst [28,29]. However,

K. You et al. / Catalysis Communications 9 (2008) 1521–1526

2. Experimental

3. Results and discussion 3.1. Comparison of catalytic performances among solid acid catalysts The catalytic activity of sulfonic acid resin for rearrangement of cyclohexanone oxime in dimethyl sulfoxide solvent was compared with the activities of several other solid acid catalysts. As seen in Fig. 1, sulfonic acid resin, sulphamic acid and phosphomolybdic acid show similar high conversion of the substrate, however, the selectivity to e-caprolactam over sulfonic acid resin catalyst is much higher than those over sulphamic acid or phosphomolybdic acid. The sulfonic acid resin gave a 100% conversion of cyclohexanone oxime and 97.9% of selectivity to e-caprolactam, however, the sulphamic acid and phosphomolybdic acid gave only 73.7% and 19.0% of selectivity to e-caprolactam, respectively. These phenomena may be caused by two factors. On the one hand, as an electron-withdrawing substituent, the phenyl ring adjacent to the sulfonic group can disperse the negative charge and stabilize the anion, and thus increase the acid strength of the sulfonic group. On the other hand, the phenyl group is more hydrophobic than other group and thus the formation of cyclohexanone as by-product through hydrolysis of cyclohexanone oxime is reduced. In addition, HTS-1 zeolite and sulfonic carbon gave comparatively poor conversion of cyclohexanone oxime and low selectivity to e-caprolactam though they posses large amounts of acid sites and high acid strengths, respectively. The catalytic reaction probably proceeded mainly on the exterior surface due to the diffusion limitation of the liquid in the micropores. Similar results on the micropore zeolites, however, were also observed in

2.1. Reagents and instrument

The Beckmann rearrangement of cyclohexanone oxime was carried out in liquid-phase at a certain set temperature in a 50 ml three-necked, round-bottomed flask reactor equipped with a reflux condenser and a magnetic stirrer. The reactor was immersed in an oil bath. A typical reaction run was as follows: 0.5 g catalyst was added in a solution of 20 mmol of cyclohexanone oxime in 10 ml solvent that was allowed to equilibrate at the set temperature. The obtained mixture was heated to the reaction temperature and stirred for 12 h. The resulting mixture was filtered and analyzed by using an Agilent 6890 gas chromatography.

Conversion and selectivity (%)

2.2. Typical reaction procedure

100 80 60 40 20 0 Su lfo ni ca pci To d re lu sin en es ul fo ni ca ci d Su lp ha m Ph ic os ac ph id om ol yb di ca ci d Su lfo ni cc ar bo n

All reagents and solvents used were analytical grade and were obtained commercially. The sulfonic acid resin was purchased from chemical reagent factory of Shanghai. Gas chromatography was performed on Agilent 6890 equipped with a DB-5 (30 m  0.25 mm  0.25 lm) column. Mass spectra were run on an America Varian Saturn 2100T GC–MS.

conversion of cyclohexanone oxime selectivity of caprolactam

-1

this rearrangement route suffers some problems such as the difficulty in separation products from ionic liquids and environmental pollution caused by the use of the phosphorated compounds. In contrast, the solid–liquid phase catalytic system appears to be more promising, where the reaction can proceed at a moderate temperature and the catalyst deactivation is minimized due to the presence of solvent. In this regard, the possibility of using a solid–liquid phase process at a moderate temperature is worth to be reexamined. These arguments suggest that the development of a green, simple, and cheap catalyst system for liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam over solid acid catalyst is highly desirable. In this paper, we report a simple and efficient process for liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam over solid acid catalyst, and the excellent conversion and selectivity were obtained. In addition, concerning the solvent effect, it has been reported that the solvent strongly affected the activity, selectivity and stability of zeolite catalysts in the vapor-phase Beckmann rearrangement of cyclohenanone oxime [30–32]. It is suggested that a polar protic solvent is effective in accelerating the reaction rate by assisting desorption of e-caprolactam. However, the results of this study show that both the dielectric constant and the basicity of a solvent may be related to the catalytic activity in the liquid-phase Beckmann rearrangement of cyclohexanone oxime. We hope that this work may serve as a guideline to the selection of an appropriate solvent in the liquid-phase Beckmann rearrangement of oxime.

H TS

1522

Fig. 1. Activities of oxime rearrangement reaction and lactam selectivities in DMSO solvent over different solid acid catalysts. Reaction condition: 0.5 g catalyst; 20 mmol cyclohexanone oxime; 10 ml DMSO; 130 °C; 12 h.

K. You et al. / Catalysis Communications 9 (2008) 1521–1526

the heterogeneous vapor-phase Beckmann rearrangement reaction [33]. Therefore, the catalytic activity strongly depends on the amounts of acid sites accessible by the reactant and the strength of the acid sites. The strong acid strength would enhance the selectivity to e-caprolactam. 3.2. Effect of solvent on liquid-phase rearrangement of cyclohexanone oxime The effects of different solvents on the reaction of Beckmann rearrangement of cyclohexanone oxime catalyzed by sulfonic acid resin were investigated and the experimental results were summarized in Table 1. The solvent was found to have great influence on the oxime conversion and lactam selectivity. The Beckmann rearrangement of cyclohexanone oxime over sulfonic acid resin afforded a 100% conversion of cyclohexanone oxime and 97.9% of selectivity to e-caprolactam at 130 °C in dimethyl sulfoxide solvent. The main by-product is cyclohexanone, which derived from the hydrolysis of cyclohexanone oxime under acidic condition. From Table 1, it is clearly seen that the conversion of cyclohexanone oxime and the selectivity to e-caprolactam were very poor in apolar cyclohexane. In contrast to cyclohexane, improved conversion and selectivity are the features obtained in polar organic solvents. Generally speaking, oxime rearrangement is initiated by the attack of the proton from a Brønsted acid site to the nitrogen atom of oxime, followed by the proton transfer from nitrogen to oxygen atom. Then, the alkyl migration occurs simultaneously with the elimination of the protonated hydroxyl group. The proton transfer from nitrogen to oxygen is a 1,2-H-shift reaction with high activation energy. This process of forming O-protonated oxime is considered the rate-determining step in Beckmann rearrangement [34,35]. We now speculate that both the strength of adsorption and the dielectric constant of a solvent possibly play central roles in determining the performance of

Table 1 Effect of solvent on rearrangement reactiona Solvent

Ethyl acetate Alcohol Cyclohexane Chlorobenzene Toluene DMFd,c DMSOe

Dielectric constantb

Conversion (%)

Selectivity (%) ecaprolactam

cyclohexanone

6.0 24.6 2.0 5.6 2.4 36.7 46.7

2.1 2.5 1.0 33.7 18.1 12.6 100

1.7 2.8 0 90.1 92.2 20.2 97.9

98.3 96.9 100 9.9 7.2 24.7 2.1

1523

the reaction. From the viewpoint of the accessibility of the reactant to the active sites, a non-polar solvent such as chlorobenzene or toluene is favorable. However, the relatively lower dielectric constants of non-polar solvents counterbalance the advantage (see Table 1). This means that the 1,2-H-shift of the N-protonated oxime may be more difficult in the non-polar solvents presumably due to the lower ability to stabilize the transition structure or to reduce the energy barrier of the transition state in comparison with other solvents which have large proton affinities. Moreover, these non-polar solvents are also less efficient in assisting the migration of OHþ 2 group from nitrogen to carbon atom, and hence, the promotion of the rearrangement rate by a solvent is smaller. In this process, maybe DMSO can play an important role in stabilizing N-protonated oxime cation and migrating hydrogen, and the possible reaction path is shown in Scheme 1. There exists a quite strong five-membered ring formed via a direct interaction between DMSO and N-protonated oxime. Such a stabilizing interaction between the solvent molecule and the migrating hydrogen in the transition structure would lower the energy barrier of 1,2-H-shift. This transition state plays a crucial role in which the reactants are oriented in an optimal manner creating the most favorable spatial conditions for proton transfer in a subsequent rapid step. In contrast, N,N-dimethylfomamide (DMF) may be regarded as better solvent to accelerate not only the 1,2-H-shift but also the subsequent rearrangement. However, DMF possesses itself weak basicity, probably poisons the active acid sites on the catalyst and therefore hinders the catalytic reaction. The GC–MS analysis results show that other by-products mainly were nitriles (5-cyanopent-1-ene and cyanopentane) formed by ring-opening dehydration, and small amount of nitro-cyclohenxane and octahydrophenadine formed by the oxidation and dimerization of cyclohexanone oxime in DMF. On the basis of these results, it is clear that the facility of protonation of oxime through the adsorption of substrate on the active site depends on the competitive adsorption between substrate and solvent. On the other hand, a solvent having a higher dielectric constant or a more polar nature is preferred in the two subsequent energy-demanding steps in which a solvent may accelerate the reaction by stabilizing transition structure and also by promoting the migration of OHþ 2 group. Therefore, the selection of a suitable solvent balancing the two aspects of adsorption and the dielectric constant is the most important in the liquid-phase Beckmann rearrangement of oxime over solid acid catalyst. The results indicate that DMSO is proved to be the best media for Beckmann rearrangement of cyclohexanone oxime over sulfonic acid resin among present solvents.

a

Reaction condition: cyclohexanone oxime : 20 mmol; catalyst mass : 0.5 g, solvent : 10 ml; reaction time : 12 h. b Value at 20 °C. c Other by-products mainly are nitriles (5-cyanopent-1-ene and cyanopentane) and small amount of nitro-cyclohenxane and octahydrophenadine in DMF. d,e Reaction temperature: 130 °C.

3.3. Effect of reaction temperature on liquid-phase rearrangement of cyclohexanone oxime The effect of the reaction temperature on cyclohexanone oxime conversion and selectivity to e-caprolactam was

1524

K. You et al. / Catalysis Communications 9 (2008) 1521–1526

N

H

OH

+

N

CH3

H O

CH3 S O H

+

N

H+

H O

CH3

CH3 S O

H+ N

DMSO

O H

N-protonated oxime

CH3

CH3 S O OH2+

H N

O N

O H -H+

N

+

OH2 H2O

N

H+

N +

-H2O

O H N

-DMSO

O-protonated oxime Scheme 1. The possible reaction path of liquid-phase rearrangement of cyclohexanone oxime in DMSO.

examined in dimethyl sulfoxide solvent over sulfonic acid resin. The results are illustrated in Fig. 2. It is clearly found that the oxime conversion increases rapidly with the reaction temperature from 8.6% at 80 °C to 100% at 130 °C. The selectivity to product e-caprolactam is maintained around 93–98% and doesnot significantly vary with temperature, while the selectivity to cyclohexanone slightly decreases with the reaction temperature. The conversion of cyclohexanone oxime reaches the highest (100%) when the temperature rises to 130 °C, while the selectivity to ecaprolactam slightly decreased to 97.9%. These results indicate that elevated temperature might favor the reaction for liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam.

The catalytic performance of sulfonic acid resin catalyst as a function of reaction period was examined in DMSO at 130 °C. The change in catalytic activity of sulfonic acid resin with time on stream was shown in Fig. 3. In the first 4 h, the cyclohexanone oxime conversion and selectivity to e-caprolactam increase dramatically to 87.7% and 98.5%, respectively. Then, the reaction rate slows down. The cyclohexanone oxime conversion gradually increases to 100% as the reaction proceeds up to 24 h, while the selectivity to e-caprolactam only decreases slightly. These results also show that the catalytic active sites of sulfonic acid resin for Beckmann rearrangement do not deactivate during the reaction period. 3.5. Effect of the amount of catalyst on liquid-phase rearrangement of cyclohexanone oxime

100

Conversion or selectivity (%)

3.4. Effect of reaction time on liquid-phase rearrangement of cyclohexanone oxime

80 60 40 20 0 60

80

100

120

140

160

0

Temperature/ C Fig. 2. Activities of oxime rearrangement reaction and lactam selectivities in DMSO solvent over sulfonic acid resin catalyst at different temperature: (j) conversion of cyclohexanone oxime; () selectivity of e-caprolactam; (N) selectivity of cyclohexanone. Reaction condition: 0.5 g catalyst; 20 mmol cyclohexanone oxime; 10 ml DMSO; 12 h.

The experimental results of cyclohexanone oxime rearrangement catalyzed by the sulfonic acid resin with different amount of catalyst in dimethyl sulfoxide are listed in Table 2. Obviously, the conversion of cyclohexanone oxime and selectivity to e-caprolactam increased gradually, whereas the selectivity to cyclohexanone decreased gradually with the elevated amount of catalyst. The conversion of oxime reached 100% with 97.9% of selectivity to e-caprolactam when reaction was added 0.5 g of sulfonic acid resin. The conversion of cyclohexanone oxime and the selectivity to e-caprolactam or cyclohexanone were almost unchanged when the amount of catalyst exceeded 0.5 g. Therefore, the suitable amount of catalyst adequately and efficiently catalyzed the Beckmann rearrangement of cyclohexanone oxime to the main product, enhancing the yield of e-caprolactam. These results illuminated that the acid capacity of catalyst has important influence on the

K. You et al. / Catalysis Communications 9 (2008) 1521–1526

Table 3 The results of reuse of catalyst in rearrangement reaction of cyclohexanone oximea

Conversion or Selectivity (%)

100 98

Run

Conversion (%)

96 1 2 3 4 5

94 92 90

100 100 99.8 95.5 90.6

Selectivity (%) e-caprolactam

cyclohexanone

97.9 97.8 97.8 97.7 97.5

2.1 2.2 2.2 2.3 2.5

a

Reaction condition: cyclohexanone oxime: 20 mmol; DMSO: 10 ml; reaction temperature: 130 °C; reaction time: 12 h.

88 86

1525

0

5

10

15

20

25

4. Conclusions

Reaction time (h) Fig. 3. Activities of oxime rearrangement reaction and lactam selectivities in DMSO solvent over sulfonic acid resin catalyst in different reaction time: (N) conversion of cyclohexanone oxime; () selectivity of ecaprolactam. Reaction condition: 0.5 g catalyst; 20 mmol cyclohexanone oxime; 10 ml DMSO; 130 °C.

Table 2 Effect of the amount of catalyst on rearrangement reactiona Catalyst (g)

0.05 0.1 0.2 0.5 1.0

Conversion (%)

21.1 45.5 96.1 100 100

Selectivity (%) e-caprolactam

cyclohexanone

86.8 91.3 95.6 97.9 97.8

13.2 8.1 4.4 2.1 2.0

In conclusion, an environmentally friendly catalytic liquid-phase Beckmann rearrangement of cyclohexanone oxime to e-caprolactam can be successfully carried out over solid sulfonic acid resin under mild condition, and excellent conversion and selectivity were obtained. This catalytic system involved is environmentally harmless, easily separate, and the catalyst can be reused. Moreover, the influences of solvent, catalyst, reaction temperature and reaction time on rearrangement of cyclohexanone oxime had been investigated. The results show that DMSO give the highest conversion and selectivity to e-caprolactam, and is proved to be the best media for Beckmann rearrangement of cyclohexanone oxime over sulfonic acid resin among present solvents. Maybe, this work can serve as a direction to the selection of an appropriate solvent in the liquid-phase Beckmann rearrangement of oxime over solid acid catalyst.

a

Reaction condition: cyclohexanone oxime: 20 mmol; DMSO: 10 ml; reaction temperature: 130 °C; reaction time: 12 h.

Acknowledgement

liquid-phase Beckmann rearrangement of cyclohexanone oxime.

The authors thank the financial support for this work by the National Natural Sciences Foundation of China (No. 20233040, 20572021).

3.6. Recycling of solid sulfonic acid resin in Beckmann rearrangement of cyclohexanone oxime

References

Based on itself performance of solid sulfonic acid resin, it can be easily recovered by filtrating and then was washed for three times by ethanol solvent. The results of the recovered sulfonic acid resin after five recycling are listed in Table 3. Obviously, reuse of the sulfonic acid resin for three times decreased slightly in conversion of cyclohexanone oxime, however, the selectivity to e-caprolactam was hardly reduced. As can be seen from Table 3, the evident decrease in conversion was observed in fourth and fifth runs. We suspected the possible reason is that the mass leaching of sulfonic acid resin resulted in decrease of conversion. Moreover, in this experiment, it is clearly observed that the globular grains were ground to farinose grains in presence of the magnetic stirrer; maybe the structure of sulfonic acid resin was damaged, and therefore resulted in decrease of the activity for the fourth and fifth runs.

[1] W.J. Wassen, R. Geottsch, US Patent 4081442. [2] U. Brand, R. Schmitz, E. Deuker and H. Fuchs, US Patent 4806638. [3] K. Chaudhari, B. Rajaram, A.J. Chandwadkar, S. Sivasanker, J. Mol. Catal. A 177 (2002) 247. [4] M. Ishida, T. Suzuki, H. Ichihashi, A. Shiga, Catal. Today 87 (2003) 187. [5] Y.J. Zhang, Y.Q. Wang, Y.F. Bu, Z.T. Mi, W. Wu, E.Z. Min, S. Han, S.B. Fu, Catal. Commun. 6 (2005) 53. [6] A. Aucejo, M.C. Burguet, A. Corma, V. Forne´s, Appl. Catal. 22 (1986) 187. [7] H. Sato, K. Hirose, M. Kitamura, Y. Nakamura, Stud. Surf. Sci. Catal. 49 (1989) 1213. [8] T. Curtin, J.B. McMonagle, B.K. Hodnett, Appl. Catal. A: Gen. 93 (1992) 75. [9] L.X. Dai, Y. Iwaki, K. Koyama, T. Tatsumi, Appl. Surf. Sci. 122 (1997) 335. [10] G. Dahlhoff, U. Barsnick, W.F. Ho¨lderich, Appl. Catal. A 210 (2001) 83. [11] D. Shouro, Y. Moriya, T. Nakajima, S. Mishima, Appl. Catal. A 198 (2000) 275.

1526

K. You et al. / Catalysis Communications 9 (2008) 1521–1526

[12] F.M. Bautista, J.M. Campelo, A. Garcı´a, J.M. Marinas, M.S. Moreno, Stud. Surf. Sci. Catal. 78 (1993) 615. [13] A. Thangaraj, S. Sivasanker, P. Ratnasamy, J. Catal. 137 (1992) 252. [14] L.X. Dai, R. Hayasaka, Y. Iwaki, K.A. Koyano, T. Tatsumi, Chem. Commun. (1996) 1071. [15] W.F. Ho¨lderich, G. Heitmann, Catal. Today. 38 (1997) 227. [16] W.F. Ho¨lderich, G. Dahlhoff, Chem. Innov. 31 (2001) 29. [17] L. De. Luca, G. Giacomelli, A. Porcheddu, J. Org. Chem. 67 (2002) 6272. [18] B. Wang, Y. Gu, C. Luo, T. Yang, L. Yang, J. Suo, Tetrahedron Lett. 45 (2004) 3369. [19] S. Chandrasekhar, K. Gopalaiah, Tetrahedron Lett. 44 (2003) 7437. [20] S. Chandrasekhar, K. Gopalaiah, Tetrahedron Lett. 43 (2002) 2455. [21] R. Anilkumar, S. Chandrasekhar, Tetrahedron Lett. 41 (2000) 5427. [22] K. Narasaka, H. Kusama, Y. Yamashita, H. Sato, Chem. Lett. (1993) 489. [23] M.A. Kira, Y.M. Shaker, Egypt. J. Chem. 6 (1973) 551.

[24] J.H. Davis, P.A. Fox, Chem. Commun. (2003) 1209. [25] R. Sheldon, Chem. Commun. (2001) 2399. [26] J. Dupont, R.F. de Souza, P.A.Z. Suarez, Chem. Rev. 102 (2002) 3667. [27] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 39 (2000) 3772. [28] R.X. Ren, L. Zueva, W. Ou, Tetrahedron Lett. 42 (2001) 8441. [29] J. Peng, Y. Deng, Tetrahedron Lett. 42 (2001) 402. [30] J. Roseler, G. Heitmann, W.F. Holderich, Appl. Catal. 144 (1996) 319. [31] T. Yashima, N. Oka, T. Komatsu, Catal. Today 38 (1997) 249. [32] T. Komatsu, T. Maeda, T. Yashima, Micropor. Mesopor. Mater. 36 (2000) 173. [33] G.P. Heitmann, G. Dahlhoff, W.F. Holderich, Appl. Catal. A 185 (1999) 99. [34] M.T. Nguyen, L.G. Vanquickenborne, J. Chem. Soc., Perkin Trans. 2 (1993) 1969. [35] M.T. Nguyen, G.T. Raspoet, L.G. Vanquickenborne, J. Am. Chem. Soc. 19 (1997) 2552.