α‐Zirconium phosphate (ZP) is one of the most important compounds in inorganic chemistry, and the layered structure of this material has led to its use in a variety of different fields [1, 2, 3]. ZP behaves as a unique ion exchanger because of its exceptionally poor aqueous solubility, high thermal stability, resistance to radiation and abrasive properties [4, 5]. The H+ of the P-OH moiety in ZP can be exchanged for various ions, resulting in an enlargement of the interlayer distance [6, 7, 8, 9]. Several studies pertaining to the successful exchange of this proton with various divalent and trivalent cations have been presented in the literature [10, 11, 12, 13, 14]. It has also been reported that ZP possesses excellent selectivity towards Pb2+, Zn2+, and Fe3+ as an ion exchanger [15, 16, 17]. Furthermore, ZP has been reported to exhibit antibacterial activity when loaded with Cu2+, Zn2+, or Ce3+ [5, 6, 13, 14]. There have also been several reports concerning the catalytic activity of ion-exchanged materials of this type, including the use of zinc zirconium phosphate (ZPZn) and copper zirconium phosphate (ZPCu) as catalysts in the acetylation of alcohols and phenols and the use of potassium iron zirconium phosphate as a catalyst in Friedel-Crafts benzoylation reactions [18, 19, 20, 21, 22, 23, 24].
The selective oxidation of alcohols into the corresponding carbonyl compounds is of importance in research because the corresponding aldehydes, ketones and carboxylic derivatives serve as important and versatile intermediates for the synthesis of various chemicals, vitamins, drugs and fragrances [25, 26, 27]. For example, benzaldehyde (BzH) is a typical product of alcohol oxidation and a starting material for the preparation of intermediates for dyestuffs, agrochemicals, perfumery and pharmaceuticals [27, 28, 29]. From economic and environmental viewpoints, there have been many recent publications emphasizing environmentally benign methods for the oxidation of alcohols, using molecular O2 or aqueous H2O2 as the oxidant, in the presence and/or absence of solvents such as CuSO4 [25], CuBr2 [26], AMPA [28], MPA/V2O5-Al2O3 [29], TEMPO-IL/CuCl [30], Ni3[Fe(CN)6]2 [31], (Pd/Fe@C) [32], silica-gel-TEMPO-NOx [33], CoTM4PyP-MT [34], Zn4(P2W15O56)16- [35], (TEAH)H2PW12O40 [36], Zn-Co-LDH [37], RuCl3∙3H2O [38], H4SiW12O40/SiO2 [39], AuRu/AC [4 0 ], oxidovanadium(V) complexes [41], Ag/SBA-15 [42], DHPDMDO [43], H2WO4/[C8mim][NTf2] [44], SF-3-APTS-Fe(TClPP) [45], NaBrO3/[bmim]Br [46], KBr/Oxone [47], WO4@PMO-IL [48], Cu-NHC-TEMPO complexes [49], Cu/AlO(OH)x [50], Na4H3 [SiW9Al3∙(H2O)3 O37] [51], copper/imidazolium/TEMPO [52], Ca(ClO)2/Al2O3 [53], PVPTB [54], Au/UiO-66 [55], PSFC [56], Au/Al2O3 [57], PMo11M (M = Co, Mn, Ni) [58], Mn(salen)OAc [59], BPFC [60], KMnO4-aluminum silicate [61], MnTPPS-silica [62] and PVP-H2O2 [63]. Homogeneous catalysts have drawbacks in terms of their corrosive nature, pollution of the product with catalyst, tedious catalyst separation and post-synthesis disposals, and recovery from the effluents. However, the design of a new catalyst, which gives excellent conversion with maximum selectivity for organic transformation, is one of the challenges in the field of catalysis. With growing environmental concerns, one of the most promising ways to achieve these goals seems to be the use of green and insoluble catalysts or of ecofriendly solvent-free conditions. H2O2 is a green and very clean oxidant for liquid phase oxidations because it provides a high content of active oxygen species where water is the only byproduct. This oxidant is much cheaper and safer than most other organic and inorganic oxidants and is also readily available. When an insoluble catalyst is used, it can be easily recovered from the reaction mixture by simple filtration and recycled and can be reused several times, making the process more economically and environmentally viable. Furthermore, reported examples have demonstrated that heterogeneous catalysts typically require easier work-up procedures. With this in mind, and as part of ongoing work towards the development of efficient green catalysts for organic transformations [64,65 ] with particular emphasis on the oxidation of alcohols [23], we report herein the use of nickel zirconium phosphate (ZPNi) as an efficient catalyst for the mild and convenient selective oxidation of alcohols as characterized by ICP-OES, XRD, BET, NH3-TPD, Py-FTIR, SEM and TEM.
All the reagents and solvents used in the current study were purchased from Merck Chemical Company and used without further purification. The catalyst was prepared according to previously published procedures, with minor modifications [2, 8, 9, 10]. As an initial step, ZP was prepared according to the following procedure. ZrOCl2·8H2O (5 g) was heated at reflux in a solution of H3PO4 (50 mL, 12 mol/L) for 24 h. The resulting mixture was cooled to ambient temperature to give a suspension, which was filtered and then washed with a solution of H3PO4 (0.1 mol/L) until the filtrate was free of chloride ions. The filter cake was then washed several times with distilled water until the pH of the filtrate was neutral. The solid was collected and dried in an oven at 110 °C for 24 h.
ZPNi was prepared through an ion exchange reaction [8, 9, 10]. Briefly, ZP (3 g) was dispersed in deionized water (50 mL) at 50 °C and the resulting suspension was treated with a solution of Ni(OAc)2 (100 mL, 0.1 mol/L) in water (excess amount of Ni2+). This mixture was then heated at reflux for 4 d. It is noteworthy that the acetate ion performed effectively as a base to keep the hydrogen ion concentration in solution sufficiently low to achieve high loadings of the catalyst [24]. A complete exchange between the cations and the hydrogen ions of the P-OH groups could not be achieved in less than 3 d or at temperatures below 80 °C [13]. The resulting slurry was filtered hot to give a light green solid, which was washed with distilled water until no Ni2+ ions could be detected in the filtrate (i.e., until the filtrate was colorless). The solid product was then dried at 100 °C for 24 h before being calcined at 600 °C for 4 h to give the final product, ZPNi, as a pale green solid (Scheme 1).
The chemical composition of the ZPNi catalyst was evaluated at different stages of the reaction (i.e., before and after the catalytic reaction) using an Optima 7300 V ICP-OES spectrometer (PerkinElmer). The samples were ground into a fine powder and analyzed by XRD on a Philips X’pert X-ray diffractometer. The specific surface areas of the samples were determined from their N2 adsorption-desorption isotherms using the Brunauer-Emmett-Teller (BET) method on a Quantachrome ChemBET 3000 instrument. Each sample was degassed at 400 °C for 2 h before being analyzed to remove any adsorbed species from their surfaces. The BET surface areas of the materials were estimated from their N2 adsorption-desorption isotherms. The surface morphologies of the ZP and ZPNi materials were studied by SEM on a Philips XL scanning electron microscope (Philips). TEM images of ZPNi were obtained on a CENTRA 100 TEM system (Zeiss).
ZPNi (0.5 mol%) was added to 5-mmol substrate in a 25-mL two-necked flask. The mixture was heated in an oil bath to 50 °C and 30% H2O2 (0.015 mol) was added slowly with continuous stirring for the specified time. The reaction progress was monitored by GC. At the end of the reaction, the mixture was cooled to room temperature and the catalyst was removed from the reaction mixture by centrifugation. Afterwards, the organic layer was separated from the aqueous phase by extraction with n-hexane and drying over anhydrous CaCl2. The identities of reaction products were confirmed by FT-IR, GC-MS and 1H NMR.
To examine the recyclability of the catalyst, the used ZPNi was recovered from the reaction media and reused. After the first use, the catalyst was separated from the reaction mixture by centrifugation, washed sequentially with ethanol and water before being dried at 110 °C for 2 h, and then activated at 450 °C for 2 h.
The ICP-OES analyses of ZP and ZPNi are shown in Table 1. The results obtained in the current study for ZPNi were compared with those reported previously in the literature [8, 9, 10]. Our results revealed that there was a negligible leach of nickel ions into the reaction media after the reaction (i.e., following the first use of the catalyst).
Figure 1 shows the powder XRD patterns of the ZP and ZPNi materials. The results show some characteristic reflections in the 2θ range of 5°-40°. The diffraction peak in ZP at 2θ ≈ 12° was assigned to a d002 basal spacing of 0.75 nm between the planes, which was consistent with the patterns previously reported for ZP and its derivatives with a hexagonal crystal system [2]. It shows that the d-spacing of the (002) plane of ZPNi had increased, which indicated that the Ni2+ ions had intercalated into the interlayer of ZP and increased the d002 basal interlamellar spacing of ZP from 0.75 to 0.98 nm. It is well known that the ion radii of Ni2+ (0.069 nm) and hydrated Ni2+ (0.404 nm) are smaller than the basal spacing of ZP (0.75 nm) [66, 67].
The XRD results indicated that Ni2+ ions had inserted into the interlayer of ZP and increased the basal spacing of the modified ZP after the exchange [8, 9, 10]. These data indicated that ZPNi had been formed successfully. The XRD pattern of the ZPNi catalyst after the 8th run showed that the basal spacing of ZP was about 1.03 nm, which was only a little larger than that of the fresh ZPNi catalyst. This increase may have occurred because of a decrease in Ni2+ on the surface of ZP, and an increase in the number of water molecules between the layers following the seventh run (i.e., Ni2+ ions may have been washed off during the regeneration of the catalyst, section 2.3 and Table 1).
Figure 2 shows the N2 adsorption-desorption isotherm of ZPNi as a representative example in the relative pressure range (p/p0) of 0.1-1.0. The surface area of ZPNi was determined to be 121.5 m2/g. The isotherm for ZPNi shows three adsorption stages. The first of these stages was observed at p/p0 < 0.37, whereas the second stage was observed in the range of 0.37 < p/p0 < 0.93, and the third stage was observed at higher relative pressures (p/p0 > 0.93). The N2 adsorption-desorption isotherm of ZPNi exhibited a typical “type IV” isotherm shape with a distinct hysteresis loop that is characteristic of a mesoporous material [68]. The hysteresis loop (type H3) is associated with the occurrence of capillary condensation in the mesopores and indicates the presence of a mesoporous structure in the ZPNi catalyst. The observed increase in adsorption at the higher p/p0 value indicated the presence of larger mesopores in the sample [9]. The surface area of ZPNi after the 8th run was found to be 72.3 m2/g.
Pyridine adsorption was used to determine the acidic sites using FTIR. Prior to the measurements, 20 mg of a catalyst was pressed into a self-supporting disc and activated in the IR cell attached to a vacuum line at 350 °C for 4 h. The adsorption of pyridine was performed at 150 °C for 30 min. The excess of probe molecules was further evacuated at 150 °C for 30 min. The adsorption-evacuation was repeated several times until no changes in the spectra were observed (Fig. 3(1)).
The main bands observed over the samples are assigned according to the literature data [69, 70]. The pyridine-desorbed FTIR spectra of ZPNi shows strong bands at 1632 and 1541 cm-1, indicating the presence of the pyridinium ion. The band at 1488 cm-1 is a combination band between those at 1541 and 1444 cm-1 and correspond to Brӧnsted and Lewis acid sites, respectively [18].
Figure 3(2) shows the Py-FTIR spectrum of the catalyst after the 8th run. It clearly indicates that the replacement of Ni2+ on the surface of the catalyst with H+ during the catalyst regeneration process has reduced the number of Lewis acid sites (band at 1444 cm-1 was reduced). Conversely, the amount of Bronsted acid sites were increased (band at 1632 cm-1 was reduced).
The total acidity of samples were determined by temperature-programmed desorption of ammonia (NH3-TPD) with a Quantachrome ChemBET 3000. Before the adsorption of ammonia, the samples were pre-treated in He at 250 °C for 30 min followed by 1 h at 350 °C and then cooled to 100 °C. Ammonia was then adsorbed onto the samples for 1 h. The NH3-TPD was carried out between 150 and 550 °C, at 10 °C/min, and analyzed by a thermal conductivity detector (TCD) for continuous monitoring of the desorbed ammonia. NH3-TPD provides a quantitative estimation of the total number of acid sites and the distribution of acid strengths. Because of the strong basicity of NH3 gas, it was expected that all acid sites on the catalysts would interact with NH3. The total amount of NH3 desorbed after saturation permits the quantification of the number of acid sites on the surface, while the position of the peak and the desorption temperature indicates the strength of the catalyst, i.e., the higher temperature of desorption, the stronger the acid strength [64]. The NH3-TPD curves of ZPNi are shown in Fig. 4.
ZPNi desorbed ammonia at a wide range of temperatures from 212 to 538 °C, which mostly corresponds to the medium and strong acidic sites. The NH3 desorption peak at temperatures below 250 °C belongs to the physisorption/chemisorption of NH3 molecules on weak acidic sites. The peak at about 250-450 °C shows the existence of intermediate strength acidic sites while the peak at 450-538 °C demonstrates the presence of strong acidic sites on the surface of ZPNi. Figure 4 shows that the desorption of ammonia starts at almost 212 °C and peaks at 317 °C. The NH3-TPD curves decreased with further increases in temperature and were almost complete at 538 °C.
Figure 4 indicates that ZPNi contains a considerable number of acidic sites owing to the presence of Ni2+ groups on the surface of ZP layers making it suitable as a solid acid catalyst. The extent of desorption is found to be ca. 1.8 mmol/g for the catalyst. A TPD experiment was carried out after the 8th cycle on recovered catalyst to compare with fresh catalyst (Table 1).
The SEM image of ZP (Figure 5(a)) revealed the presence of hexagonal plates with well-defined shapes and very smooth surfaces. Figures 5(b) and (c) show the SEM images of ZPNi. These images revealed that the structure of ZPNi was much less ordered than that of ZP, and that the ZPNi particles had aggregated to form both sheets and spheres of different shapes and sizes [4, 9].
Figure 6 shows the TEM images of ZPNi. It shows that ZPNi catalyst retained the original morphology of ZP (layered structure) and that the particles were approximately 150 nm in size. These images also showed nanoparticles of different size on the smooth surface of the ZP. The presence of metallic crystal nanoparticles on the surface of ZP indicated that the nickel deposited on the surface of the ZP had agglomerated. Similar observations have also been reported for zinc, copper and cerium with ZP [6, 14, 24].
Figures 5(d) and 6(c) show SEM and TEM images, respectively, of the catalyst following its 8th run. Both of these images showed that the sheets and particles had conglomerated to a much greater extent following the 8th run because of the process used to regenerate the catalyst.
Initially, the benzyl alcohol (BzOH) was selected as the model substrate to determine the optimal conditions. The results are summarized in Table 2. Although the major product was BzH, benzoic acid was also formed as a side product (Table 2, entries 4, 5, 8, 11, 14, 15). It was found that by changing reaction parameters, such as time (Table 2, entries 1-5), mole ratio (Table 2, entries 6-8), temperature (Table 2, entries 9-11) and catalyst amount (Table 2, entries 12-15), the selectivity and/or yield were changed.
As can be seen from Table 2, increase in the reaction time up to 15 min increase the yield of BzH where it is also the only product. However, at longer reaction time, benzoic acid was produced as a byproduct and the BzH yield was reduced to 75%. When the mole ratio of BzOH:H2O2 reached 1:3, the highest yield was obtained. Further increases in the BzOH:H2O2 mole ratio slightly increased the yield but sharply decreased selectivity towards BzH (Table 2, entries 2, 6-8). The optimum reaction temperature for the oxidation of BzOH was determined by performing the reaction at different temperatures. The yield of BzH was very low at room temperature even after 90 min (Table 2, entry 9). Table 2 shows that increases in temperature up to a maximum temperature of 50 °C can accelerate the reaction. However, further temperature increases result in a decrease in the yield, which might be owing to the over-oxidation of BzH to benzoic acid or the self-decomposition of H2O2 at higher temperatures leading to insufficient oxidation of BzOH (Table 2, entries 2, 9-11). Hence, the best results were achieved when the reaction was performed with a BzOH:H2O2 mole ratio of 1:3 at 50 °C for 15 min (Table 2, entry 2). To evaluate the role of our catalyst, the oxidation reaction was performed with ZP used as the catalyst for comparison (Table 2, entry 17). The oxidation reaction was also performed in the absence of catalysts (Table 2, entry 16). No significant amount of BzH was detected indicating that H2O2 alone is unable to oxidize BzOH to BzH.
As shown in Table 3, a wide range of alcohols bearing either electron-donating or electron-drawing groups were successfully oxidized into their corresponding carbonyl compounds within short reaction time. Notably, the efficient transformations of primary alcohols into the desired aldehydes were observed without any over-oxidation to the corresponding carboxylic acid (Table 3, entries 1-14). Both electron-donating and electron-drawing groups accelerated the oxidation reaction, but ortho-substituted substrates (Table 3, entries 2, 5, 7, 9) gave relatively lower yields compared with the corresponding para-isomer because of steric hindrance [26, 50, 52]. Compared with benzylic alcohols, aliphatic alcohols showed relatively low reactivity toward oxidation (Table 3, entries 22-25). It should be noted that under these reaction conditions, various secondary alcohols were oxidized to their corresponding ketones in fair yields (Table 3, entries 15-21, 23, 25). To investigate the selectivity of this method, a mixture of 1-hexanol and BzOH was subjected to oxidation. BzH was produced with a yield of 85% and only 8% of 1-hexanal was detected. Such selectivity is a useful practical achievement in the oxidation of alcohols.
Based on the observations, the typical mechanism proposed for ZPNi catalyzed oxidation of alcohols with H2O2 is given in Scheme 2. At first, the Ni Lewis acid site (on the surface of ZPNi) interacts with H2O2 to form (I) and then reacts with the alcohol to give an intermediate (II) that subsequently gives the corresponding carbonyl compound and regenerated active sites by two steps of dehydration. Hence, any steric hindrance around the hydroxyl group of alcohols, such as the presence of a substituent at the ortho position (Table 3, entries 2, 5, 7, 9) or more hindered secondary alcohols (Table 3, entries 15-21), causes longer reaction time and lower yields. Almost similar behavior was previously reported [26, 50, 52].
The reusability of the ZPNi catalyst was investigated under the optimum reaction conditions for the oxidation of alcohols, and the results are shown in Table 4. The elemental composition of the catalyst remained largely unchanged following its 8th run, although the amount of nickel in the catalyst was reduced by almost 50% compared with the first run (Table 1). The recycled ZPNi catalyst gave a similar product yield to the freshly prepared catalyst up until the sixth cycle.
All products were identified by comparing GC-MS (GC-Mass, Agilent 5975C) and 1H NMR spectra (Bruker-Avance AQS 400 MHz spectrometer; in CDCl3) with authentic samples [23, 28, 47, 49, 51]. 1H NMR spectral data of some compounds from Table 3.
PhCHO (Table 3, entry 1): 1H NMR (400 MHz, CDCl3) δ = 10.01 (s, 1H), 7.85-7.45 (m, 5H).
4-MeO-C6H4CHO (Table 3, entry 3): 1H NMR (400 MHz, CDCl3) δ = 9.90 (1H, s), 7.85 (2H, d, J = 8.7 Hz), 7.0 (2H, d, J = 8.7 Hz), 3.93 (3H, s).
4-Me-C6H4CHO (Table 3, entry 4): 1H NMR (400 MHz, CDCl3) δ = 9.92 (s, 1H), 7.75 (d, J = 8.4 Hz, 2H), 7.32 (d, J = 8.4 Hz, 2H), 2.43 (s, 3H).
2-Cl-C6H4CHO (Table 3, entry 5): 1H NMR (400 MHz, CDCl3) δ = 10.40 (s, 1H), 7.88-7.86 (q, 1H), 7.54-7.31 (m, 3H).
4-NC-C6H4CHO (Table 3, entry 11) 1H NMR (400 MHz, CDCl3) δ = 10.13 (s, 1H), 8.11 (d, J = 8.2 Hz, 2H), 7.90 (d, J = 8.2 Hz, 2H).
Ph2CO (Table 3, entry 15): 1H NMR (400 MHz, CDCl3) δ = 7.81 (d, J = 7.5 Hz, 4H), 7.59 (t, J = 7.3 Hz, 2H), 7.49 (t, J = 8.0 Hz, 4H).
C6H5COCH3 (Table 3, entry 16): 1H NMR (400 MHz, CDCl3) δ = 7.97 (d, J = 7.6 Hz, 2H), 7.57 (t, J = 7.3 Hz, 1H), 7.47 (t, J = 7.7 Hz, 2H), 2.62 (s, 3H).
4-MeO-C6H4COCH3(Table 3, entry 17): 1H NMR (400 MHz, CDCl3) δ = 7.90 (d, J = 9.0 Hz, 2H), 6.9 (d, J = 9.0 Hz, 2H), 3.85 (s, 3H), 2.58 (s, 3H).
Cyclohexanone (Table 3, entry 25): 1H NMR (400 MHz, CDCl3) δ = 2.52-2.38 (m, 4H), 2.2-1.64 (m, 6H).
A comparison of the catalytic efficiency of ZPNi with selected known catalysts is collected in Table 5. BzOH was oxidized to BzH in less than 15 min at 50 °C with 90% isolated yield using the present protocol (Table 5, entry 18). Although the reaction conditions are different, the catalyst used in this study (ZPNi) may be regarded as one of the better catalysts with regards to lower reaction time and temperatures, better BzH yields and selectivity toward BzH as the major product. The catalyst is also easily regenerated for further application.
In summary, we have reported the catalytic performance of water-insoluble ZPNi in the oxidation of alcohols using H2O2. The catalyst was characterized by various methods and the results showed good agreement with literature. ZPNi showed outstanding catalytic performance with excellent conversion of BzOH and selectivity to BzH at 50 °C at 15 min. The selectivity remained unchanged for all reactions, although the yields were found to be affected by steric hindrance. Alcohols having bulky groups required longer reaction time. The results clearly reveal that this method can be applied for chemoselective oxidation of BzOHs in the presence of aliphatic alcohols. This procedure is environmentally benign, applicable to a wide range of alcohols, efficient, high yielding, safe and operationally simple.
Acknowledgments
We gratefully acknowledge the funding support received for this project from the Isfahan University of Technology, IR Iran.
2015, Vol. 36 Issue (7): 1109-1116
PDF (649 KB)
2015, Vol. 36 Issue (7): 1109-1116
PDF (649 KB)