Introduction

The transformation of alcohols into aldehydes and ketones is of great importance in organic chemistry, both in laboratory-scale experiments and in manufacturing processes [1]. Numerous oxidizing reagents (for example, CrO3, KMnO4, MnO2, etc.) in stoichiometric amounts have been traditionally used to accomplish this transformation, but with many drawbacks, for example use of expensive reagents, volatile organic solvents, and discharge of environmentally pernicious wastes. Environmentally and economically, catalytic oxidation processes are thus valuable, and those using molecular oxygen as primary oxidant are particularly attractive. Many highly efficient systems have been developed for catalytic aerobic oxidation of alcohols by use of transition metals alone (for example palladium [2–4], gold [5–7], ruthenium [8, 9], copper [10, 11], vanadium [12, 13]) or in combination with the nitroxy radical 2,2,6,6-tetramethylpiperidyl-1-oxy (TEMPO). Of particular interest are catalytic systems consisting of an inexpensive transition metal compound and TEMPO for mild and selective aerobic oxidation of alcohols. However, separation of the products from the TEMPO can require lengthy workup of these systems, especially for reactions on a large industrial scale.

In this context, use of a supported TEMPO catalyst seemed desirable, because it enables both catalyst recycling and simplified workup of the reaction mixture. TEMPO has been immobilized on inorganic and organic supports, for example SiO2 [14, 15], molecular sieve [16], polymers [17–19], affording solid catalysts which are readily separated from the reaction mixtures. Nevertheless, these solid catalysts often have disadvantages of their own, and when oxygen, especially, is used as the terminal oxidant, supported TEMPO has less catalytic activity than under comparable homogeneous conditions [20, 21].

Ionic liquids (ILs) are, in contrast, eco-friendly reaction media or catalysts which have attracted increasing attention because of their particular properties—undetectable vapor pressure, high thermal stability, excellent solubility, and ease of recovery and reuse. However, ionic liquids are still expensive, despite being commercially available. Furthermore, the high viscosity of ILs may result in mass transfer limitations if the chemical reaction is fast, causing the reaction to proceed in the diffusion layer of the catalyst rather than in the bulk solvent.

Recently, by analogy with well-established supported aqueous-phase catalysis [22], ILs have been introduced to the immobilization of homogeneous catalysts. The “solid catalyst with ILs (SCILs)” technique, in which the ionic liquid phase is immobilized as a film on a conventional supported catalyst, is a new method for circumventing the limitations described above [23, 24]. With SCILs, the activity and selectivity of the original solid catalyst may be changed substantially by coating its internal surfaces with the IL. Thus, processes have been designed for immobilization of ILs on solid catalysts [25–30].

Our recent work has focused on oxidation of organic compounds with molecular oxygen as the oxidant [31–34]. We have previously reported an efficient and recyclable catalyst system comprising CuCl2 and TEMPO task-specific ionic liquid (TEMPO-IL), supported by the sol–gel technique [10] or by physical adsorption [35], for aerobic oxidation of alcohols. Herein, as part of our studies on the oxidation of organic compounds, we report a facile and efficient procedure for aerobic oxidation of alcohols to the corresponding carbonyl compounds by use of an effective and reusable catalyst prepared by coating the surface of ready-made polystyrene–TEMPO resin with the ionic liquid [bmim]PF6 containing a copper salt (Fig. 1).

Fig. 1
figure 1

Supported ionic-liquid catalytic system used for selective oxidation of alcohols

Full size image

Results and discussion

Initial research was conducted with 4-methoxybenzyl alcohol as substrate to investigate the effects on the selective oxidation of the IL layer supported on the polystyrene–TEMPO (PS-TEMPO) resin; the results are summarized in Table 1. First, the catalytic activity of original PS-TEMPO resin was evaluated in different solvents. Among the solvents tested, catalytic activity of PS-TEMPO/CuCl2 was much greater in n-octane than in other solvents, with 58 % conversion and 99 % selectivity for benzaldehyde (entry 1). For the solvents toluene, [bmim]PF6, and [bmim]BF4, the rate of oxidation and the selectivity were substantially lower (entries 2–4). In DMF, the reaction barely proceeded (entry 5).

Table 1 Effects of IL layer supported on PS-TEMPO resin on oxidation of 4-methoxybenzyl alcohol
Full size table

The activity and selectivity of the PS-TEMPO resin were increased substantially by the coating its surface with the IL. When n-octane was used as solvent, 85 % conversion was achieved with over 99 % selectivity, and similar results were observed in toluene. Immobilizing an IL layer on the solid support can have a marked effect on the catalytic activity of this system, improving the “chemical” properties of the catalyst (i.e. the cocatalytic effect), and providing higher effective concentrations of substrates or intermediates on the surface of the solid catalyst because of the ionic properties of the IL phase (i.e. a physical solvent effect) [23, 26]. In DMF or [bmim]PF6 conversion and selectivity were still very low (entries 10, 11). Consequently, for this catalytic system, it is critical to choose an appropriate solvent to form a biphasic system with ILs, otherwise the IL film would be washed off the catalyst, and the catalytic system would be deactivated.

After determining the optimum reaction conditions, we studied oxidation of a variety of alcohols catalyzed by PS-TEMPO@Cu-IL in n-octane; the results are listed in Table 2. All benzylic alcohols can be converted into the corresponding aldehydes with high conversion (entries 1–6). Similar reactivity was observed in the oxidation of substrates with electron-withdrawing or donating groups in the aromatic ring. The rates of these benzylic alcohol oxidations were not significantly affected by the electronic properties of the substituents on the benzene ring. Moreover, allylic and heterocyclic primary alcohols could be smoothly oxidized to the corresponding aldehydes, with high conversion, by use of this catalyst system (entries 7, 8). The catalytic system had high selectivity, and only traces of carboxylic acid (below 1 %), which was easily removed by washing the organic phase with sodium bicarbonate solution, were observed. The double bond of the allylic alcohol was unaffected. Despite the good results obtained for benzylic, allylic, and heterocyclic alcohols, conversion of aliphatic alcohols under the same conditions was quite low (entries 9, 10).

Table 2 Oxidation of alcohols with molecular oxygen catalyzed by PS-TEMPO@Cu-IL
Full size table

The reusability of the catalyst was investigated by using 4-methoxybenzyl alcohol as model substrate. The catalyst was easily recovered by filtration and washing with n-octane after the reaction. After being dried, it was subjected to another reaction with identical substrates. The procedure was repeated, and the results indicated that the catalyst could be cycled four times, with gradual loss of catalytic activity, with selectivity for aldehyde formation always remaining very high (Fig. 2). The structure of the catalyst recovered after five runs had not obviously changed, as was apparent from comparison of its FT–IR spectrum with that of the fresh catalyst. This result indicated that the TEMPO groups remained tightly bonded to the PS resin after five runs. In the supernatant fraction, Cu was not detected by ICP analysis. We suggested that the decrease in activity may be because of loss of [bmim]PF6 adsorbed on the PS-TEMPO. To validate this postulate, the ionic-liquid layer was reloaded, and it was found that catalytic activity increased to a level close to that of the fresh catalyst.

Fig. 2
figure 2

Recycling of the catalyst used for selective oxidation of 4-methoxybenzyl alcohol

Full size image

In conclusion, we have developed a facile, efficient procedure for selective oxidation of activated alcohols by use of a novel ionic-liquid ([bmim]PF6) catalyst containing CuCl2 supported on PS-TEMPO resin. The supported ionic-liquid layer makes an important contribution to the high catalytic activity. Notable advantages of this method are high catalytic activity and selectivity, simple work-up, reusable catalyst, and mild reaction conditions.

Experimental

All chemicals were obtained from commercial sources; they were >99 % pure and used without prior purification. Gas chromatographic (GC) analysis was performed on a Shimadzu 17A equipped with a 30 m × 0.32 mm × 0.5 μm DB-5 capillary column and a flame ionization detector. Elemental analysis was performed by ICP–AES (Thermo icap 6300) and use of a CHNS/O analyzer (PerkinElmer 2400 Series II). Infrared spectra were recorded as KBr disks on a Tensor 27 FT–IR spectrometer.

Catalyst preparation

Polystyrene–TEMPO resin was prepared in accordance with a literature method [19]; loading of the TEMPO was determined to be 0.90 mmol g−1 by elemental analysis (CHNS/O analyzer). To a stirred solution of 0.6 g [bmim]PF6 in 25 cm3 methanol, 0.25 g CuCl2·2H2O (1.45 mmol) was added, then 1.0 g polystyrene–TEMPO resin. The mixture was stirred for 4 h at room temperature, then the solvent was evaporated under reduced pressure for 3 h to give a light red bead denoted PS-TEMPO@Cu-IL. Before use, the product was dried overnight at 50 °C under vacuum.

Typical experimental procedure for oxidation of alcohols

Alcohol (1.0 mmol) and 100 mg PS-TEMPO@Cu-IL were mixed with 5 cm3 n-octane and stirred at 65 °C under O2 (1 atm) for the given time. After completion (monitored by TLC or GC), the catalyst can be easily collected by filtration. The recovered catalyst was used directly for the next run after simply washing with n-octane. The combined solvent phase was concentrated under vacuum. The crude product was purified by column chromatography (10:1 petroleum ether–ethyl acetate) to provide the analytically pure aldehyde. All products were known compounds and were characterized by 1H NMR and IR spectroscopy; results were identical with data reported in the literature [36, 37].