A novel route for the synthesis of piperazine from N-(2,3-dihydroxy- propyl)ethylenediamine over composite photocatalysts K. V. Subba Rao,* a V. Kandavelu, a B. Srinivas, b M. Subrahmanyam b and K. Ravindranathan Thampi* a a Laboratory of Photonics and Interfaces, Swiss Federal Institute of Technology (EPFL), CH-1015, Lausanne, Switzerland; Fax: +41 21 693 4111; Tel: +41 21 693 6127 b Catalysis and Physical Chemistry Division, Indian Institute of Chemical Technology, Hyderabad 500 007, India. E-mail: ravindranathan.thampi@epfl.ch; rao.kambala@epfl.ch Received (in Cambridge, UK) 23rd July 2003, Accepted 11th September 2003 First published as an Advance Article on the web 30th September 2003 Semiconductor loaded zeolite composite catalysts (5 wt% TiO 2 /Hb) have been used to photocatalytically synthesize piperazine from N-(2,3-dihydroxypropyl)ethylenediamine with yields up to 59.0 mol%. The synthesis of piperazine and its derivatives has received growing interest because of their potent applications in the drugs, perfumery and pharmaceutical industries. Although several methods 1–3 are reported for their synthesis, the im- portance of these molecules has provided the impetus to develop more convenient and environmentally friendly proce- dures. Selective production of these compounds comprises cyclization, dehydronitrogenation and dehydration reactions. A great variety of chemical transformations in the presence of CdS or TiO 2 have been reported in earlier reviews. 4,5 Despite several semiconductor mediated reactions reported, the examples of inter and intramolecular cyclizations are rather limited. 6–8 Semiconductor loaded zeolites have recently drawn increased attention as potential composite photocatalysts due to their unique pore structure and adsorption properties. Reports on semiconductor loaded zeolites for organic sythesis are scarce. 9 The present work demonstrates a new photocatalytic route with the combined advantages of semiconductor and zeolite catalysts for the synthesis of piperazine at room temperature. The photocatalysts were prepared as described elsewhere 10 and characterized by BET, XRD, NH 3 TPD and elemental analysis (EDX and AAS). The results of AAS and EDX analysis showed good agreement for the known and observed amounts (wt%) of semiconductor loading on zeolites. 11 N-(2,3-Dihy- droxypropyl)ethylenediamine (NPEDA) was prepared as re- ported by Surrey et al. 12 The respective catalysts (100 mg) were suspended with 0.268 g of NPEDA (2 mmol) in 20 ml of acetonitrile solvent. The suspension was magnetically stirred and irradiated under a constant stream of molecular O 2 (20 ml h 21 ) at room temp., using a 250 W high pressure mercury lamp in a cylindrical round bottomed quartz photoreactor of 200 ml capacity (Ø = 2 cm, L = 20 cm) and equipped with a refluxing condensor at the top. The reaction mixture was centrifuged after irradiation (12 h) to separate the catalyst. Thin layer chroma- torgraphy (TLC) was performed in a CH 3 OH : CHCl 3 (10 : 90) solvent system to detect the product spots. Further purification was done by column chromatography and the product was characterized by 1 H NMR, melting point, electron impact mass spectroscopy (EI-MS), GC-MS and C,H,N analysis. The progress of the reaction was monitored by TLC and yields reported were estimated on the basis of isolated yields. The reaction conditions were optimized by several trials to get the maximum yield. Parallel experiments were also carried out using bare TiO 2 , CdS, ZnO and different semiconductor loaded zeolites to compare the photocatalytic activity. The results are summarized in Table 1. Table 1 shows the representative results of the photocatalytic reaction of non-aqueous NPEDA solution by various types of semiconductor loaded zeolite catalysts. On the other hand, bare semiconductor powder (TiO 2 , CdS, ZnO) photocatalyzed reaction did not yield piperazine. The reaction, in this case, predominantly progressed to total oxidation of NPEDA. It has been experimentally proved (Table 2) that the reaction is photocatalytic since this reaction proceeds only in the co- presence of oxygen, irradiation and a photocatalyst. As clearly seen in Table 1, the formation of piperazine depended on the nature of the zeolite support. In this study, one of the best photcatalysts for the production of piperazine was found to be 5 wt% TiO 2 /Hb. A change in the physical properties of the semiconductor loaded zeolite catalysts was not observed and XRD analysis (results not shown) confirms that no loss of crystallinity occurs after the modification and subsequent Table 1 Photocatalytic intramolecular cyclization of NPEDA achieved by semiconductor/zeolite composite catalysts Catalyst Surface area a / m 2 g 21 Acidity b / mmol g 21 Isolated product III c (mol%) TiO 2 (2%)/HZSM5 (30) d 350 0.37 35.2 TiO 2 (5%)/HZSM5 (30) 320 0.32 38.6 TiO 2 (2%)/HY (4.4) 400 0.25 39.4 TiO 2 (5%)/HY (4.4) 341 0.20 40.5 TiO 2 (2%)/Hb (30) 495 0.43 46.1 TiO 2 (5%)/Hb (30) 485 0.47 59.0 ZnO(2%)/HZSM5 (30) 326 0.13 17.9 ZnO(5%)/HZSM5 (30) 320 0.15 22.1 ZnO(2%)/HY (4.4) 336 0.18 25.3 ZnO(5%)/HY (4.4) 300 0.21 27.4 ZnO(2%)/Hb (30) 490 0.41 29.6 ZnO(5%)/Hb (30) 483 0.40 30.4 CdS(2%)/HZSM5 (30) 356 0.45 31.9 CdS(5%)/HZSM5 (30) 349 0.49 33.0 CdS(2%)/HY (4.4) 409 0.31 37.1 CdS(5%)/HY (4.4) 389 0.36 38.2 CdS(2%)/Hb (30) 468 0.33 41.8 CdS(5%)/Hb (30) 459 0.37 52.7 TiO 2 50 — — ZnO 30 — — CdS — — — a BET surface area measured with liquid N 2 at 77 K. b TPD of NH 3 measured on Autochem 2910 (Micromeritics, USA). c Irradiation (12 h; 250 W HP mercury lamp). d SiO 2 /Al 2 O 3 ratios of zeolites are shown in parentheses (). Table 2 Control experiments for the photocatalytic synthesis of piper- azine Conditions Irradiation time/h Isolated product III (mol%) Semiconductor/zeolite, light, O 2 12 59.0 Semiconductor/zeolite, dark, O 2 12 0 Light, O 2 12 0 Zeolite, light, O 2 12 0 Semiconductor, light, O 2 12 0 Semiconductor, light 12 0 This journal is © The Royal Society of Chemistry 2003 2706 CHEM. COMMUN. , 2003 2706–2707 DOI: 10.1039/b308520a