Sandeep S.
Kahandal
,
Sandip R.
Kale
,
Shamrao T.
Disale
and
Radha V.
Jayaram
*
Department of Chemistry, Institute of Chemical Technology, Matunga, Mumbai 400 019, India. E-mail: rv.jayaram@ictmumbai.edu.in; Fax: +91-22 3361 1020; Tel: +91-22 3361 2607
First published on 11th April 2012
It has been reported that sulphated yttria–zirconia efficiently catalysed the epoxide ring opening with a variety of alcohols under solvent free conditions to give corresponding β-alkoxy alcohols. In most of the cases reaction of epoxides with 1°, 2° and 3° alcohols in the presence of catalytic amounts of SO42−/YxZr1−xO2 gives exclusively one regioisomer with excellent yields. The catalysts were characterized by XRD, SEM/EDAX, total acidity and TGA. The protocol is highly efficient, regioselective, atom economical, greener and applicable for a wide range of aliphatic and aromatic epoxides and alcohols with excellent yields of the corresponding β-alkoxy alcohols. The catalysts show reusability for up to four consecutive catalytic cycles without any significant loss of catalytic activity.
The epoxide ring opening reaction with alcohols for the synthesis of corresponding β-alkoxy alcohols is a widely studied reaction. However due to the poor nucleophilicity of the alcohols, strong acidic or basic catalysts are required for the reaction. Recently, various approaches have been tried to bring about this transformation using, e.g., copper(II)tetrafluoroborate,4 SnCl4,5 InCl3,6 Mg(HSO4)2,7 tin(IV) porphyrinato trifluoromethane sulfonate,8 and heterogeneous catalysts such as FeCl3 supported on silica,9 polymer supported ferric chloride,10 K5[CoW12O40]3H2O,11 mesoporous alumina-silicate,12 zirconium doped silica13 and saponite clays.14 Also for this transformation various metal triflates and perchlorates such as Yb(OTf)3,15 TiCl3(OTf),16 Al(OTf)17 and Fe(ClO4)18 have been used.
Recently CuO/SiO219 and metal–organic frameworks20 have also been explored for the synthesis of β-alkoxy alcohols. Although a number of methods are available and despite the potential utility of the protocols, they suffer from disadvantages such as requirement of harsh reaction conditions, need of organic solvents, inert atmosphere, non-reusability of the catalysts, tedious workup, lower yields and, importantly, poor regioselectivity of the corresponding β-alkoxy alcohols.
Environmental and economic considerations have recently raised strong attention in redesigning commercially important processes so that the use of harmful substances and the generation of toxic waste can be avoided. In this respect, heterogeneous catalytic processes are more favourable over their homogeneous counterparts in the production of fine chemicals, owing to their ease of handling, simple workup procedures and, most importantly, their reusability.
Sulphated metal and mixed metal oxides gained much more recognition than metal and mixed metal oxides. Sulphated yttria–zirconia as a Lewis acid has been successfully used for Diels–Alder reaction,21 Mannich type of reaction,22 acylation of alcohols, thiols and amines with carboxylic acids,23tert-butoxycarbonylation of amines,24 and carbamate synthesis.25 We report herein our results of highly regioselective epoxide ring opening with alcohols using sulphated yttria–zirconia under solvent free conditions.
In continuation of our efforts to develop an environmentally benign protocol for various organic transformations,26 we describe in the present work that a catalytic amount of sulphated yttria–zirconia can bring about the efficient, smooth, highly regioselective ring opening of epoxides with 1°, 2° and 3° alcohols and amines.
We have used surface modified metal oxides and mixed oxides as catalysts for ring opening of epoxides with alcohols. Various factors were found to decide the reactivity and regioselectivity of the products (Scheme 1). The regioselectivity observed makes this synthetic route highly useful for the preparation of β-alkoxy alcohols. Reusability of the catalysts, solvent free conditions and high regioselectivity make the protocol more attractive and environmentally benign.
![]() | ||
Scheme 1 Sulphated yttria–zirconia catalysed alcoholysis of epoxides. |
![]() | ||
Fig. 2 FT-IR spectrum of SO42−/Y0.16Zr0.84O2. |
The band at 1625 cm−1 is assigned to the deformation vibration mode of the adsorbed water.
Entry | Catalystb | Time/h | Acidityc/mmol g−1 | E i d/mV | Initial rate (mmol h −1 g−1) × 104 | Yielde (%) |
---|---|---|---|---|---|---|
a Reaction conditions: styrene oxide (2 mmol), ethanol (2 mmol), solvent free, rt (32 °C). b 10 wt% of catalyst. c Acidity measured by potentiometric method. d Initial electrode potential. e GC yield. | ||||||
1 | — | 5 | — | — | — | — |
2 | ZrO2 | 5 | 0.81 | 55 | — | — |
3 | SO42−/ZrO2 | 0.5 | 2.07 | 168 | 2.21 | 59 |
4 | Y0.16–Zr0.84O2 | 0.5 | 0.90 | 91 | 1.18 | 35 |
5 | SO42−/Y0.04Zr0.96O2 | 0.5 | 1.23 | 148 | 1.37 | 42 |
6 | SO42−/Y0.08Zr0.92O2 | 0.5 | 1.64 | 195 | 2.06 | 64 |
7 | SO42−/Y0.12Zr0.88O2 | 0.5 | 1.93 | 310 | 4.57 | 83 |
8 | SO42−/Y0.16Zr0.84O2 | 0.5 | 4.19 | 530 | 7.0 | 96 |
9 | SO42−/Y0.20Zr0.80O2 | 0.5 | 2.91 | 330 | 4.62 | 79 |
10 | SO42−/Y0.24Zr0.76O2 | 0.5 | 2.65 | 248 | 1.20 | 52 |
11 | SO42−/Ce0.02Zr0.98O2 | 0.5 | 3.17 | 440 | 2.22 | 65 |
12 | SO42−/Ce0.07Zr0.93O2 | 0.5 | 4.23 | 560 | 4.98 | 87 |
13 | SO42−/Ce0.10Zr0.90O2 | 0.5 | 3.52 | 450 | 3.57 | 74 |
14 | SO42−/Ce0.15Zr0.85O2 | 0.5 | 3.50 | 460 | 2.82 | 69 |
![]() | ||
Fig. 3 Potentiometric titration curves for (a) SO42−/Y0.16Zr0.84O2, (b) yttria–zirconia and (c) sulphated zirconia. |
![]() | ||
Fig. 4 TGA profile of SO42−/Y0.16Zr0.84O2. |
![]() | ||
Fig. 6 Time–conversion plot for the ring opening of styrene oxide with ethanol catalysed by (a) ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Ring opening of styrene oxide with ethanol was selected as the model reaction and the results obtained with various catalysts are given in Table 1 (entries 1–14). In the absence of a catalyst the reaction did not take place even up to 5 h (Table 1, entry 1). While performing the reaction with ZrO2, no product formation was observed, which clearly indicates that ZrO2 itself was not highly active (Table 1, entry 2). SO42−/ZrO2 and Y0.16–Zr0.84O2 exhibit better catalytic activity as compared to the zirconia (Table 1, entries 3 and 4). It was found that sulphated yttria–zirconia showed good catalytic activity under solvent free conditions with very high regioselectivity. The experiments with different compositions of sulphated yttria–zirconia (Table 1, entries 5–10) show that SO42−/Y0.16Zr0.84O2 gives 96% yield with 100% regioselectivity under experimental conditions (Table 1, entry 8). Encouraged by these results, we extended our studies to sulphated mixed oxides of ZrO2 and CeO2. We have observed that with the different compositions of SO42−/CexZr1−xO2 the reaction gives good yields of the β-alkoxy alcohol (Table 1, entries 11–14). SO42−/Ce0.07Zr0.93O2 exhibits highest total acidity in the series of catalysts studied but it fails to give 100% regioselectivity (Table 1, entry 12). The highest activity of the SO42−/Y0.16Zr0.84O2 catalyst in terms of initial rate and yield of the product could be attributed to the total acidic sites (4.19 mmol g−1) and the acid strength, as indicated by the initial electrode potential (530 mV) (Table 1, entry 8). The results showed that sulphated yttria–zirconia is a highly active and selective heterogeneous acid catalyst system for the ring opening of styrene epoxide with ethanol.
We started with our aim to carry out the reaction under solvent free conditions so as to make the protocol environmentally benign. We observed that under solvent free conditions the reaction smoothly gives the desired β-alkoxy alcohols with very high regioselectivity.
In the case of surface modified mixed metal oxide catalysed reactions, the amount of catalyst employed is one of the important factors and hence efforts were made to determine the optimum concentration of catalyst required. Reactions of styrene oxide (2.0 mmol) and ethanol (2 mmol) were performed in the presence SO42−/Y0.16Zr0.84O2 at various catalyst loadings ranging from 5 to 20 wt% (Table 2, entries 1–4). Initially, the reaction was carried out using 5 wt% of the catalyst, which provided the desired product β-alkoxy alcohols with 55% yield (Table 2, entry 1). The increase in the catalyst concentration up to 10 wt% shows increase in product yield (Table 1, entry 2). Further increase in catalyst amount did not show any considerable increase in product yield (Table 2, entries 2–4). Thus 10 wt% of catalyst loading was selected as the optimum concentration for further studies.
After optimising the reaction conditions for ethanol as a nucleophile with styrene oxide as a model reaction, we scrutinized the generality of the protocol, and also applicability of the catalyst to other epoxides, alcohols and amines. The results are shown in Tables 3 and 4. The reaction of styrene oxide with ethanol gives 96% yield of 2-ethoxy-2-phenylethanol (Table 2, entry 2). When the reaction was carried out with styrene epoxide and methanol, 97% yield of 2-methoxy-2-phenylethanol was obtained (Table 3, entry 1). The reaction of 2-propanol gave slightly lower yield as compared to 1-propanol as a nucleophilic source. This may be due to the steric factor of two methyl groups, which hinders the reaction site resulting in the lowering of styrene oxide conversion into the desired product (Table 3, entries 3 and 4). Similar behaviour was also observed when the reaction was carried out with 1-butanol, 2-methyl-1-propanol and 2-methyl-2-propanol as nucleophiles (Table 3, entries 5–7). This observation can be explained by the fact that 2-methyl-2-propanol is even more crowded, having three methyl groups increases the steric hindrance at the reaction site which in general inhibits the nucleophilic substitution reaction. Importantly, in all cases a single regioisomer was observed in the presence of sulphated yttria–zirconia as catalyst. The reaction of styrene oxide with allyl alcohol and 2-methoxy ethanol gives 95% and 93% yield of the corresponding product without affecting the olefinic bond of allyl alcohol and the ether functionality in 2-methoxy ethanol (Table 3, entries 8 and 9).
Entry | Nucleophile | β-Alkoxy alcoholsb | Time/h | Yieldb (%) |
---|---|---|---|---|
a Reaction conditions: epoxide (2 mmol), ethanol (2 mmol), solvent free, 10 wt% of catalyst with respect to styrene oxide, rt (32 °C). b GC yield. c Reactions carried out at 45 °C. | ||||
1 | MeOH |
![]() |
0.5 | 97 |
2 | EtOH |
![]() |
0.5 | 96 |
3 |
![]() |
![]() |
0.5 | 87 |
4 |
![]() |
![]() |
0.5 | 82 |
5 |
![]() |
![]() |
0.5 | 85 |
6 |
![]() |
![]() |
0.5 | 81 |
7 |
![]() |
![]() |
0.5 | 72 |
8 |
![]() |
![]() |
0.5 | 95 |
9 |
![]() |
![]() |
0.5 | 93 |
10 |
![]() |
![]() |
2.0 | 67 |
11 |
![]() |
![]() |
2.0 | 31 |
12 |
![]() |
![]() |
0.5 | 89 |
13c |
![]() |
![]() |
2.0 | 85 |
14c |
![]() |
![]() |
2.0 | 86 |
Entry | Nucleophile | Epoxide | β-Alkoxy alcohols | Time/h | Yieldb (%) |
---|---|---|---|---|---|
a Reaction conditions: epoxide (2 mmol), alcohol (2 mmol), solvent free, 10 wt% of catalyst w.r.t. styrene oxide, 45 °C. b GC Yield. c Epoxide (2 mmol), aniline (2 mmol), solvent free, 10 wt% of catalyst with respect to epoxide, rt (32 °C). | |||||
1 | MeOH |
![]() |
![]() |
1.5 | 84 |
2 | EtOH |
![]() |
![]() |
1.5 | 95 |
3 | MeOH |
![]() |
![]() |
2.0 | 81 |
4 | EtOH |
![]() |
![]() |
2.0 | 83 |
5 |
![]() |
![]() |
![]() |
2.0 | 78 |
6 | MeOH |
![]() |
![]() |
4.0 | 93 |
7 | EtOH |
![]() |
![]() |
4.0 | 95 |
8c |
![]() |
![]() |
![]() |
0.25 | 96 |
Interestingly, we observed that the reaction of higher homologues of alcohol such as 1-hexanol and 1-octanol with styrene oxide for 2 h gives moderate yields of the corresponding products with 5% and 8% of phenyl acetaldehyde, respectively, as a rearrangement product (Table 3, entries 10 and 11). The reaction between cyclohexanol and styrene oxide furnishes the desired product in 89% yield (Table 3, entry 12). When phenol and benzyl alcohol reacted with styrene oxide at 45 °C, 85% and 86% yield of the products was obtained, respectively (Table 3, entries 13 and 14). It shows that the styrene oxide ring opening occurred at the more hindered carbon atom of the epoxide ring, as the benzylic cation is more stabilized than the primary carbonation. As the bulkiness of the incoming nucleophile increases, the rate of the reaction decreases.
We also studied the generality of the protocol for other epoxides and nucleophiles. Reaction of sterically crowded cyclohexene oxide with ethanol at room temperature gives only 15% yield of the desired β-alkoxy alcohol, but with an increase in temperature up to 45 °C, it gives 95% yield, and in the case of methanol it gives 84% yield of the desired β-alkoxy alcohol (Table 4, entries 1 and 2). This experiment again suggests that the size of the molecule has strong influence on the reactivity and conversion of epoxide with alcohols when the reaction is performed with sulphated yttria–zirconia as catalyst. Reaction of epichlorohydrin with ethanol gives slightly good yield as compared to methanol and isopropanol at 45 °C reaction temperature with attack of the nucleophile at the less hindered carbon centre of the epoxide (Table 4, entries 3–5).
A similar trend was observed in the case of 3-phenoxy propylene oxide when reacted with methanol and ethanol at 45 °C for 4 h. The attack of nucleophile at the less hindered carbon centre of the epoxide provides good yields of β-alkoxy alcohols (Table 4, entries 6 and 7).
Encouraged by the results, epoxide ring opening with alcohols was explored to extend the scope of the present methodology (Tables 3 and 4). It was observed that when aniline was reacted with styrene oxide, 2-(phenylamino)-2-phenylethanol was obtained in 96% yield (Table 4, entry 8).
To determine the activity of the sulphated yttria–zirconia for the conversion of styrene oxide to 2-ethoxy-2-phenylethanol, the reaction was carried out with eight-fold excess of ethanol and styrene oxide with respect to the sulphated yttria–zirconia catalyst. After 8 h of reaction time, 68% of styrene oxide converted to 2-ethoxy-2-phenylethanol with 99% regioselectivity. As expected, an eightfold excess of styrene oxide concentration decreases the rate of the reaction, but we observed that after 24 h, 95% styrene oxide converted to 2-ethoxy-2-phenylethanol without much decrease in the product regioselectivity, indicating that sulphated yttria–zirconia was not deactivated.
The reusability of the sulphated yttria–zirconia was tested for ring opening of styrene epoxide with ethanol. At the end of the reaction, the heterogeneous catalysts were separated by simple filtration and the recovered catalysts were again used for the next cycle after washing with acetone and drying in an oven. Catalysts showed reusability for four consecutive cycles without much decrease in catalytic activity (Table 5).
Run | Isolated yield |
---|---|
a Reaction conditions: reactions were performed with styrene oxide (3 mmol), ethanol (3 mmol), solvent free, 10 wt% of catalyst, 0.5 h, rt. | |
1 | 96 |
2 | 96 |
3 | 95 |
4 | 94 |
The plausible reaction mechanism for the synthesis of β-alkoxy alcohols involves the interaction between the oxygen atom of epoxide and acid sites of the catalyst. The interaction resulted in an increase in partial electropositive nature of the carbon attached to the phenyl ring or one which is less hindered, attacked by the corresponding nucleophile to furnish the product. The high regioselectivity is possibly due to the polarization of the C–O bond of the epoxide and predominantly attack of the nucleophile.
Sulphated yttria–zirconia was characterized by XRD, FT-IR, TGA, total acidity determined by a n-butylamine potentiometric titration method and SEM/EDAX. The percentage conversion, purity and relative yields of the final products were determined by using a Thermofisher GC-1000 equipped with a capillary column (30 m × 0.32 mm ID – 0.25 μm BP-10) with an FID detector and nitrogen as the carrier gas. The products were identified by GC-MS by using a Shimadzu (GCMS-QP 2010) EI mode with high purity helium as carrier gas. Powder XRD patterns of catalysts were recorded in a Bruker AXS diffractometer with D8Cu-Kα radiation (λ = 1.540562) over a 2θ range of 0–80°. SEM-EDAX spectra were recorded using a Tungsten source on a JEOL model JSM-6390 instrument. Thermograms were recorded on a SDT Q 600v 8.2 Build 100 model of TA instruments. Potentiometric titrations were carried out by using Equip-tronic model (EQ-614A) instruments with double junction electrodes. The regioselectivity of all products was determined by mass spectrometry, which shows different fragmentation pattern changes depending on the position of the nucleophilic groups. All the products obtained and discussed in this work have been previously reported and representative products were characterized by suitable techniques such as 1H NMR (Varian 300 MHz) and GC-MS (Shimadzu QP 2010) analysis.
White solid, 1H NMR (300 MHz, CDCl3) δ (ppm) = 7.23–7.36 (m,5 H), 4.62 (dd, J = 4.4 Hz, 8.2 Hz, 1H), 3.43–3.57 (m, 2H), 2.27 (dd, J = 3.8 Hz, 9.5 Hz, 1H), 1.17 (s, 9H); GC-MS (EI, 70 eV): m/z = 194 (M+), 163, 121, 107, 91, 79, 57, 51.
Footnote |
† Electronic supplementary information (ESI) available: Experimental details and characterization data of selected products. See DOI: 10.1039/c2cy20116j |
This journal is © The Royal Society of Chemistry 2012 |