- Email: [email protected]
Lewis acid metal ion-exchanged MAPO-5 molecular sieves for solvent free synthesis of coumarin derivative
Catalysis Communications 10 (2008) 23–28
Contents lists available at ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Lewis acid metal ion-exchanged MAPO-5 molecular sieves for solvent free synthesis of coumarin derivative S. Gopalakrishnan, K.R. Viswanathan, S. Vishnu Priya, J. Herbert Mabel, M. Palanichamy, V. Murugesan * Department of Chemistry, Anna University, Chennai 600 025, Tamil Nadu, India
a r t i c l e
i n f o
Article history: Received 7 March 2008 Received in revised form 26 July 2008 Accepted 28 July 2008 Available online 5 August 2008 Keywords: Magnesium aluminophosphate Ion-exchange TPD (ammonia) Coumarin
a b s t r a c t Magnesium aluminophosphate (MAPO-5) molecular sieve was synthesised hydrothermally and ionexchanged with Lewis acid metal ions such as Ga3+, In3+, La3+ and Ce3+ by wet method. The materials were characterised using XRD, FT-IR, SEM, TPD (ammonia) and TGA. The XRD patterns of ion-exchanged molecular sieves are similar to the parent MAPO-5 thus showed absence of structural degradation and non-framework metal oxide phase during ion-exchange. The thermograms of metal ion-exchanged MAPO-5 showed a weight loss between 550 and 600 °C which is assigned to the decomposition of + MðOHÞþ 2 into MO species. The total acidity of materials decreased after ion-exchange as revealed from TPD (ammonia) studies. Their catalytic activity was tested in the solvent free synthesis of coumarin. Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction Synthesis of aluminophosphate molecular sieves has attracted great attention because of their potential application in catalysis, adsorption and separation [1–4]. The major disadvantage of parent AlPO molecular sieve is the absence of acid sites. Hence incorporation of Bronsted acid sites is necessary. The incorporation of divalent cations such as Mg2+, Zn2+, Co2+, Mn2+, etc into AlPO and SAPO results MAPO and MSAPO, respectively [5–7]. Such isomorphous substitution introduces charge imbalance in the framework which is balanced by protons, thus generating Bronsted acidity and offering catalytic activity and ion-exchange capability [8]. Mg2+ substituted aluminophosphate molecular sieve exhibited strong Bronsted acidity and it was reported to be active catalyst in acid catalysed chemical reactions [9–11]. Among the aluminophosphate molecular sieves AlPO-5 has attracted much attention due to its large pore size with 0.7 nm pore opening. It is composed of 12-membered rings of regular alternation of Al and P in the tetrahedral framework [12]. The catalytic application of both transition and non-transition divalent metal ions substituted AlPO-5 molecular sieves has been reported already [13,14]. Although there are reports on the catalytic activity of metal ions substituted AlPO-5 molecular sieves with Bronsted acid sites as catalytically active sites, Lewis acid ion-exchanged MAPO-5 catalysts yet to receive considerable importance. However, Lewis acid ion-exchanged zeolites have been reported to be catalytically active for many reactions of industrial importance [15,16]. Tynjala and Pakkanen [17] * Corresponding author. Tel.: +91 44 22203144; fax: +91 44 22200660/22350397. E-mail address: [email protected] (V. Murugesan). 1566-7367/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2008.07.032
compared the acidic properties of Ba2+, Al3+ and La3+ ion-exchanged ZSM-5 zeolite. This report revealed clearly that during ion-exchange with trivalent metal ions, the splitting of water molecules results in the formation of new Bronsted acid sites and acidic MðOHÞþ 2 based on Plank–Hirschler mechanism. This is more probable in La3+ and Al3+ ion-exchanged zeolite than Ba2+. Rane et al. [18] reported wet ion-exchange of Ga3+ in ZSM-5 zeolite which is more Lewis acidic and exhibit high catalytic active in the cracking of n-heptane. Mavrodinova et al. [19–22] reported the introduction of InO+ and AlO+ into HY and Hb zeolites. These reports revealed the formation of MO+ species during calcination. Considering in the above lines, there is no report on Ga3+, In3+ and rare-earth metal-ions-exchanged MAPO-5 molecular sieves. In the present study we report the preparation of Ga3, In3+, La3+ and Ce3+ ion-exchanged MAPO-5, their characterisation and catalytic activity in the liquid phase condensation of salicylaldehyde and diethyl malonate. 2. Experimental 2.1. Preparation of catalysts Hydrothermal synthesis of MAPO-5 was carried out using a gel composition of 0.1MgO:1TEA:1Al2O3:1P2O5:40H2O adopting the procedure reported already [23]. Aluminium isopropoxide (Merck), phosphoric acid (Merck) (88%) and magnesium acetate (Merck) were used as the sources for Al, P and Mg, respectively. Aluminium isopropoxide (28.37 g) was soaked in 30 ml distilled water for 24 h in a stainless steel autoclave (316 type) and stirred vigorously for 1 h. While stirring, 1.48 g magnesium acetate dissolved in 7.8 ml
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
phosphoric acid and 20 ml distilled water were added drop by drop and the stirring continued for additional 1 h. Triethylamine (9.7 ml) was added drop wise to it and the pH of the gel was found to be 4.4. The autoclave was tightly closed and kept at 175 °C under autogeneous pressure for 24 h. Then it was cooled to room temperature to obtain the solid product. The product was washed several times with distilled water and dried in an air oven at 110 °C for 12 h. It was then calcined at 550 °C for 8 h to remove the template. The final composition of MAPO-5 was found to be 0.07MgO:0.93Al2O3:1P2O5. The ion-exchanged materials were prepared by stirring 2 g of the calcined MAPO-5 with corresponding metal nitrate solution (30 ml; 0.1 M) for 12 h at 80 °C. The solid was filtered, washed thoroughly with distilled water and dried at ambient temperature. The same procedure was repeated thrice. These samples were used for characterisation and catalytic studies.
a
Intensity (a.u)
24
b
c
d
e 5
10
15
20
25
30
35
40
2 Theta (degree)
X-ray diffraction patterns were recorded on a PANalytical X’pert PRO diffractometer using nickel filtered Cu Ka (k = 0.154 nm) radiation and a liquid nitrogen cooled germanium solid state detector. The diffractograms were recorded in the 2h range from 5 to 40° in steps of 1.2° min1 with a count time of 10 s at each point. Fourier transform infrared spectra (FT-IR) of the materials were recorded on a FT-IR spectrometer (Nicolet Avatar 360) using KBr pellet technique. About 15 mg of the sample was pressed (under a pressure of 2 tons cm2) into a self-supported wafer of 13 mm diameter. This pellet was used to record the infrared spectra in the range 400– 4000 cm1. Thermogravimetric analysis (TGA) of the materials was performed on a high resolution thermogravimetric analyser (Perkin–Elmer diamond series) under nitrogen atmosphere in the temperature from 50 to 800 °C at a heating rate of 10 °C min1. The size and morphology of the samples were recorded using a scanning electron microscope (SEM) (JEOL 640). The density and strength of acid sites were determined by temperature programmed desorption (TPD) of ammonia on a Micromeritics chemisorb 2750 pulse chemisorption system. About 0.08 g of the sample was pretreated in an oxygen stream at 450 °C for 3 h. Ammonia was adsorbed at 100 °C and the physisorbed ammonia was removed by applying vacuum. The temperature was then increased up to 600 °C in a helium stream (20 ml min1) at a heating rate of 10 °C min1 and the desorption was monitored by a thermal conductivity detector (TCD) [24]. The liquid products obtained in the reaction were analysed by a gas chromatograph (Shimadzu GC-17A; crosslinked 5% phenyl methyl siloxane capillary column, FID detector). Further, the products were also confirmed using a gas chromatograph coupled with Turbo mass spectrometer (GC–MS Perkin–Elmer auto system XL (EI, 70 eV) using helium as the carrier gas at a flow rate of 1 ml min1.
3. Results and discussion 3.1. Characterisation The powder X-ray diffraction patterns of MAPO-5 and metal ion-exchanged MAPO-5 catalysts are shown in Fig. 1. The patterns of MAPO-5 are similar to that of AlPO-5 [25,3]. All the ion-exchanged XRD patterns are almost similar to the patterns of MAPO-5. However, the patterns corresponding to non-framework oxide of exchanged metal ions were not seen and hence such oxides were absent or the dimension of such oxides, if present, may not be in sufficient dimension to be detected by XRD. The patterns also revealed absence of structural degradation during ion-ex-
Fig. 1. XRD patterns of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
change with metal nitrates. The FT-IR spectra of calcined MAPO5 catalysts are shown in Fig. 2a. The broad envelope between 2750 and 3700 cm1 in all the spectra corresponds to O–H stretching modes of surface hydroxyls (P–OH and Bronsted acid sites). The presence of water is confirmed by its OH2 bending vibration close to 1630 cm1. The water content appears to be different in these catalysts. The broad absorption band around 1000 cm1 is due to tetrahedral asymmetric vibrations of P–O–Al and P–O–M groups. The corresponding symmetric vibration gives a peak close to 750 cm1. The peak around 500 cm1 corresponds to the bending mode of T–O–T. These are the characteristic vibrations of MAPO5 molecular sieves [26]. The spectra of LaMAPO-5, CeMAPO-5, GaMAPO-5 and InMAPO-5 are shown in Fig. 2b–e. The spectral characteristics of ion-exchanged MAPO-5 are almost similar to MAPO-5. However, the increase in the peak height corresponding to OH stretching vibration for ion-exchanged MAPO-5 is due to the OH stretching modes of surface hydroxyls interacting with water and also water adsorbed on Lewis sites. Thermogravimetric analysis of metal ion-exchanged MAPO-5 was carried out between 30 and 800 °C. However, the thermograms shown in Fig. 3a–d are drawn between 400 and 650 °C for clarity. There is a characteristic weight loss between 550 and 600 °C. This weight loss is assigned to decomposition of MðOHÞþ 2 into MO+ species where M is the trivalent ions used for ion-ex-
a
Transmittance (%)
2.2. Physicochemical characterization
b c d e
3500
3000
2500
2000
1500
1000
500
Wavenumber (cm-1) Fig. 2. FT-IR spectra of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
25
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
3.65
Weight loss (%)
3.60 a
3.55
b 3.50 c 3.45 d
3.40
3.35 400
450
500
550
600
650
Temperature (°C ) Fig. 3. TGA curves of (a) LaMAPO-5, (b) CeMAPO-5, (c) InMAPO-5 and (d) GaMAPO5.
TCDSignal (a.u)
change. This is the most supportive evidence for the formation of MO+ species. The formation of MðOHÞþ 2 is based on Plank–Hirschler mechanism as discussed by Tynjala and Pakkanen [17]. The SEM pictures of MAPO-5 and ion-exchanged MAPO-5 catalysts are shown in Fig. 4a–e. The SEM picture of MAPO-5 shows a clear hexagon. The edges and sides are clearly distinct in addition to well grown hexagon. There are particles with irregular morphology. The size of these particles appears to be less than the size of hexagonal crystal. The hexagon also appears to be agglomerated
but along the same plane. The surface of hexagon does not appear to possess any smaller grains on its terrace. The SEM picture of GaMAPO-5 is shown in Fig. 4e. This hexagon also shows well developed edges and phases. The tiny grains without definite morphology are also present. The SEM image of CeMAPO-5 is shown in Fig. 4d. The fully developed hexagon and its precursor with different sizes and morphologies are evident. Both La and InMAPO-5 also possess hexagonal morphology but the surface appears different. Agglomeration of plate hexagon is clearly evident from the Fig. 4b and c. The acidity of the catalysts was determined by temperature programmed desorption (TPD) of ammonia technique. TPD (ammonia) was carried out between 100 and 650 °C for calcined MAPO-5 and ion-exchanged MAPO-5 catalysts. The results are depicted in Fig. 5. This profile matches with already reported for boron-beta zeolite, beta zeolite and H-ZSM-5 modified with indium [27–29]. The amount of strong and weak acid sites present in these catalysts was calculated from the volume of ammonia desorbed and the values are presented in Table 1. All the metal ion-exchanged catalysts showed higher amount of weak acid sites than MAPO-5. The weakly acidic nature of La3+ ions has been reported already in La3+ ion-exchanged H-ZSM-5 [30]. In addition La3+ ions were not shown to reduce the effective pore volume of the catalyst or increase the steric hindrance for the passage of linear molecules through the pores of the catalyst [30]. According to Plank–Hirschler mechanism, the highly charged metal ions like trivalent cations undergo hydrolysis in aqueous solution producing MðOHÞþ 2 species. Hence the actual species involved in the ion-exchange may be þ MðOHÞþ 2 . The formation of MðOHÞ2 takes place most probably in the case of smaller cations with high polarizing power. However, these ions are dehydrated to produce MO+ species during calcination [17]. This is also verified by TGA analysis as discussed above. During the ion-exchange of M3+ ions there will be the loss of protons in the framework. However, there may be splitting of coordinated water molecules around M3+ ions to form two protons and MðOHÞþ 2 species. These two newly formed protons will be taken up by two oxygen bridges of the framework. These two protons may not be as much acidic as that of actually exchanged strongly acidic protons. This confirms that weak acid sites are generated as a result of ion-exchange. These weak acid sites include weak Bronsted acid sites, Lewis acid sites and defective sites [30]. Hence
a b
c d e 100
150
200
250
300
350
400
450 500
550
600
650
Temperature (°C ) Fig. 4. SEM pictures of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
Fig. 5. Temperature programmed desorption (TPD) of ammonia profiles of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
26
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Table 1 TPD (ammonia) results of MAPO-5 and ion-exchanged MAPO-5 Catalyst
Temperature maximum (°C)
Volume (mL/g at STP)
Weak and strong acid sites (mmol/g)
MAPO-5
168.5 580.2
1.1717 0.1955
0.0523 0.0087
LaMAPO-5
148.6 574
1.9667 0.1116
0.0878 0.005
CeMAPO-5
172.7 506.7
1.745 0.0697
0.0779 0.0031
InMAPO-5
205.3 585.9
1.9504 0.06
0.0871 0.0027
GaMAPO-5
164 522.2
2.2106 0.0581
0.0987 0.0026
as a result of metal-ion-exchange some of the strong acid sites may be converted to weak acid sites. The data also indicate feasibility of ion-exchange of strong acid sites. The order of ion-exchange of strong acid sites is found to be GaMAPO-5 > InMAPO-5 > CeMAPO-5 > LaMAPO-5. Hence Ga3+, In3+, La3+ and Ce3+ may not exist as M3+ during ion-exchange but as MðOHÞþ 2 as predicted by Plank–Hirschler mechanism. The same may be converted to MO+ during calcination as shown below:
Table 2 Conversion of salicylaldehyde, yield and selectivity of 3-ethyl coumarin carboxylate over different catalysts Catalyst
Temperature (°C)
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
MAPO-5
100 120 140 160
21.5 23 25.7 28.8
89.3 90.4 91 92
19.2 20.8 23.4 26.5
LaMAPO-5
100 120 140 160
31 36.6 44.7 53.4
90.9 91.2 91.4 92.6
28.2 33.4 40.9 49.5
CeMAPO-5
100 120 140 160
34.8 42.3 47.2 55.8
86.4 89.3 92.5 93.1
30.1 37.8 43.7 52
InMAPO-5
100 120 140 160
40.2 47.5 51.4 61.3
89 91.1 91.6 94.6
35.8 43.3 47.1 58
GaMAPO-5
100 120 140 160
46.9 56.3 64.4 71
91.2 92.5 92.7 95.9
42.8 52.1 59.7 68.1
Calcination
þ þ MðH2 OÞ3þ 6 ! MO þ 5H2 O þ 2H
Reaction conditions: catalyst amount 0.1 g, time 12 h, feed ratio 1:3 (salicylaldehyde: DEM).
3.2. Catalytic performance Intermolecular cyclisation of salicylaldehyde and diethyl malonate was carried over MAPO-5 and ion-exchanged MAPO-5 at 100, 120, 140 and 160 °C in the liquid phase. The feed ratio was kept at 1:3 (salicylaldehyde: diethyl malonate) and the reaction was carried for 12 h in all the cases. The major product was found to be 3-ethyl coumarin carboxylate. The results of salicylaldehyde conversion, selectivity and yield of products over all the catalysts are presented in Table 2. The salicylaldehyde conversion and selectivity to 3-ethyl coumarin carboxylate increased with increase in temperature. The reaction required initial condensation between carbonyl group of salicylaldehyde and active methylene group of diethyl malonate on the active sites like Bronsted acid sites. There may be enolisation of diethyl malonate and the enolised species reacted with salicylaldehyde in a sequential manner giving the precursor of 3-ethyl coumarin carboxylate as shown in the reaction Scheme 1. In addition, the reaction may also involve protonation at the ester carbonyl followed by nucleophilic attack of the salicylaldehyde –OH group as shown in the reaction Scheme 2. The latter appeared to be more probable over catalyst possessing Bronsted acid sites. The subsequent intermolecular cyclisation of the precursor also requires acid sites as protonation of carbonyl oxygen of the aldehyde group can facilitate easy condensation with the active methylene group at the ester portion. Since the precursor is not observed as product, its conversion appears to be rapid. Salicylaldehyde conversion increased with increase in temperature and so the reaction appeared to be activation energy demanding. The yield of 3-ethyl coumarin carboxylate increased with increase in temperature and the selectivity of product is above 86% at all temperatures. Comparison of the results of salicylaldehyde conversion and product yield shows that MAPO-5 is the least active among the catalysts. This revealed that the reaction is largely controlled by Lewis acid sites. LaMAPO-5 showed higher conversion of salicylaldehyde and yield of 3-ethyl coumarin carboxylate than MAPO-5. This observation established that metal ion dependent route is an additional one to the Bronsted acid site catalysed route. The metal ion
COOCH2CH3 H2C _ 2 Mg
+ H O
OH
+C + P
COOCH2CH3 + H2C
_ 2 Mg
OCH 2CH3
O + P
COOCH2CH3 OH CHO
O
O O C CH2 CHO C O OCH 2CH3 O C + CH CH C O
O
O C
+ H MAPO
CH + CH
C O
HO H O
OCH 2CH3 O COOCH2CH3
OCH 2CH3 Scheme 1. Intermolecular condensation of salicylaldehyde and diethyl malonate over MAPO-5.
may play a template role in bringing together salicylaldehyde and diethyl malonate in its co-ordination sphere and simultaneously allowing them to react further to yield the product. Co-ordination of both the reactants to LaO+ may be quite possible as it has high co-ordination capability. The important observation is that ion-exchange of La3+ mainly occurs on strong acid sites leaving weak acid sites unaffected as discussed above. Hence, the metal ion dependent route is more pronounced than Bronsted acid sites. The reaction over CeMAPO-5 exhibited nearly similar trend in terms of conversion and product yield as that of LaMAPO-5. However, there is further increase in conversion of salicylaldehyde, product yield and selectivity of 3-ethyl coumarin carboxylate over CeMAPO-5. Therefore Ce3+ may play a similar role as that La3+. The conversion and yield are higher over CeMAPO-5 than LaMAPO-5 since Ce3+ sites are more Lewis acidic than La3+ sites. The higher Le-
27
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Table 4 Effect of catalyst amount on salicylaldehyde conversion, yield and product selectivity over GaMAPO-5
OCH2CH3 OH +
CHOO=C
H
CH2CH3
O O=C CH2
O=C
O
+ MO
CH2 -CH CH OH 3 2
MO+
O=C
C=O
OCH 2CH3
OCH 2CH3
H O
O
C=O MO+
C=O MO+
CH 2
C=O O=C H
C=O
OCH 2CH3
C H
C
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
0.05 0.10 0.15
55.8 71.0 73.5
86.5 95.9 95.2
48.3 68.1 70.0
Reaction condition: temperature 160 °C, time 12 h, feed ratio 1:3 (salicylaldehyde: DEM).
H H O C OCH 2 CH3
O -H2O
C=O
-MO+
CH
Catalyst amount (g)
O
O C
COOCH 2 CH3
COOC2 H5
OH H
Scheme 2. Intermolecular condensation of salicylaldehyde and diethyl malonate over MO+.
wis acidity could be attributed to smaller size of Ce3+ than La3+ due to lanthanide contraction. The results of conversion and product yield obtained over InMAPO-5 also illustrate the same trend as that of La3+ and Ce3+ exchanged catalysts. But both conversion and yield over InMAPO-5 are higher than La3+ and Ce3+ exchanged catalysts. This may be due to its high Lewis acid strength as a result of the smallest size. The results of GaMAPO-5 illustrate similar results as that of InMAPO-5 but the conversion and yield are higher than InMAPO5. GaMAPO-5 is more active than all other catalysts as Ga3+ is the smallest ion. Since Ga3+ is a small size ion, it can form more stable complex than others. This complex also exhibits sufficient lability to catalyze the condensation. The yield of the product is also higher over GaMAPO-5 than other catalysts. Bigi et al. [31] reported the same condensation reaction over montmorillonite KSF and K10. They reported a maximum yield of 46% at 160 °C without any solvent and with a feed ratio 1:2. The results obtained with GaMAPO5 therefore demonstrate that it is better than the catalysts already reported. The effect of feed ratio (salicylaldehyde: diethyl malonate) viz, 1:1, 1:3 and 1:5 on salicylaldehyde conversion, yield and product selectivity was studied over GaMAPO-5 and the results are shown in Table 3. The feed ratio 1:3 showed higher conversion, yield and selectivity of 3-ethyl coumarin carboxylate than other feed ratios. The low conversion at 1:1 feed ratio may be due to less amount of diethyl malonate which could suppress co-adsorption of diethyl malonate and salicylaldehyde. The conversion and product yield with feed ratio 1:5 are lower than 1:3. This may be attributed to dilution of salicylaldehyde by diethyl malonate. Similar trend is also reflected in the yield and selectivity of 3-ethyl coumarin carboxylate and hence the optimum feed ratio is 1:3.
Table 5 Effect of reaction time on salicylaldehyde conversion, product selectivity and yield over GaMAPO-5 Time (h)
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
3 6 9 12 18
48.0 57.0 64.8 71.0 70.2
88.9 90.0 90.2 95.9 95.8
42.7 51.3 58.5 68.1 67.3
Reaction condition: temperature 160 °C, catalyst amount 0.1 g, feed ratio 1:3 (salicylaldehyde: DEM).
The effect of catalyst loading on salicylaldehyde conversion and product selectivity was studied using 0.05, 0.1 and 0.15 g catalyst with a constant feed ratio of 1:3 at 160 °C and the results are presented in Table 4. The conversion increased with increase in the catalyst amount. Therefore once the product formed it should be immediately desorbed from the catalyst surface without allowing it to chemisorb again. The selectivity of 3-ethyl coumarin carboxylate also increased with increase in catalyst amount from 0.05 to 0.15 g. GaMAPO-5 with more active sites is better than other catalysts. This study revealed that the reaction still depends on the number of active sites. So the optimum amount of catalyst is found to be 0.1 g. The effect of reaction time on conversion and product yield was carried out and the results are presented in Table 5. The conversion increased with increase in reaction time up to 18 h beyond which the conversion remained steady illustrating the attainment of equilibrium. Based on the results of conversion it could be visualised that the reactants may be reacting in a sequential manner on the metal ion to form 3-ethyl coumarin carboxylate. Ga3+ and InMAPO-5 showed better results than La3+ and CeMAPO-5. It is also concluded that templating role of metal ions during condensation is better at 1:3 than at 1:1. This revealed that salicylaldehyde may be better co-ordinating than diethyl malonate. 4. Conclusion
Table 3 Effect of feed ratio on salicylaldehyde conversion, yield and product selectivity over GaMAPO-5 Feed ratio
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
1:1 1:3 1:5
64.0 71.0 67.7
86.4 95.9 90.5
55.3 68.1 61.3
Reaction condition: temperature 160 °C, catalyst amount 0.1 g, time 12 h.
The study established the feasibility of ion-exchange in MAPO-5 with Lewis acid metal ions such as La3+, Ce3+, Ga3+ and In3+. The exchanged metal ions are suggested to remain as MO+ as charge compensating cation based on Plank–Hirschler mechanism. It is also concluded that strong acid sites are exchanged with Lewis acid metal ions during ion-exchange The catalytic activity of Lewis acid metal ion-exchanged MAPO-5 materials is better than Bronsted acid sites for the condensation of salicylaldehyde and diethyl malonate to form coumarin derivative. The absence of side reaction is also one of the important features of the catalysts.
28
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Acknowledgements The authors gratefully acknowledge the financial support from the Defence Research and Development Organisation (DRDO), Government of India, New Delhi, for this research work. One of the authors (Mr. S. Gopalakrishnan) is grateful to DRDO for the award of Senior Research Fellowship (SRF). References [1] S.T. Wilson, B.M.C. Lok, A. Messina, T.R. Cannan, E.M. Flanigen, J. Am. Chem. Soc. 104 (1982) 1146. [2] C. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite Structure Types, Elsevier, London, 2001. [3] J.M. Bennett, W.J. Dytrych, J.J. Pluth, J.W. Richardson Jr., J.V. Smith, Zeolites 6 (1986) 349. [4] J.M. Thomas, R. Raja, G. Sankar, R.G. Bell, Acc. Chem. Res. 34 (2001) 191. [5] E.M. Flanigen, B.M. Lok, R.L. Patton, S.T. Wilson, Stud. Surf.Sci. Catal. 46 (1989) 191. [6] Z. Levi, A.M. Raitsimring, D. Goldfarb, J. Phy. Chem. 95 (1991) 191. [7] C. Montes, M.E. Davis, B. Murray, M. Narayanan, J. Phy. Chem. 94 (1990) 6425. [8] E.M. Flanigen, R.L. Patton, S.T. Wilson, in: P.J. Gorbet, W.J. Mortier, E.F. Vansant, G. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science, Elsevier, Amsterdam, 1988, p. 13. [9] R.J. Gorte, G.T. Kokotailo, A.I. Biaglow, D. Parrillo, C. Pereira, Stud. Surf. Sci. Catal. 69 (1991) 181. [10] D.B. Akolekar, J. Catal. 144 (1993) 148. [11] L.T. Yuen, S.I. Zones, T.V. Harris, E.J. Gallegos, A. Auroux, Micropor. Mater. 2 (1994) 105. [12] J.M. Bennett, W.J. Dytrch, J.J. Pluth, J.W. Richardson, J.V. Smith, Zoelites 6 (1986) 349.
[13] M.S. Rigutto, H. Van Bekkum, J. Mol. Catal. 81 (1993) 77. [14] M.J. Happen, A.M. Elemans-Mehring, J.H.C. Van Hoff, Appl. Catal. A: Gen. 152 (1997) 203. [15] V.V. Balasubramanian, A. Pandurangan, M. Palanichamy, V. Murugesan, Ind. J. Chem. Tech. 7 (2000) 149. [16] V.V. Balasubramanian, V. Umamaheswari, I. Sudarsankumar, M. Palanichamy, V. Murugesan, Proc. Ind. Acad. Sci. (Chem. Sci.) 110 (1998) 453. [17] P. Tynjala, T.T. Pakkanen, J. Mol. Catal. A: Chem. 110 (1996) 153. [18] N. Rane, M. Kersbulck, R.A. Van Santen, E.J.M. Hensen, Micropor. Mesopor. Mater. 110 (2008) 279. [19] V. Mavrodinova, M. Popova, Y. Neinska, Ch. Minchev, Appl. Catal. A: Gen. 210 (2001) 397. [20] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, Appl. Catal. A: Gen. 248 (2003) 197. [21] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, Appl. Catal. A: Gen. 262 (2004) 75. [22] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, J. Mol. Catal. A: Chem. 220 (2004) 239. [23] K. Joseph Antony Raj, V.R. Vijayaraghavan, J. Chem. Sci. (Ind. Acad. Sci.) 116 (2004) 107. [24] P. Concepcion, J.M. Lopez Nieto, J. Perez-Pariente, A. Misfud, Zeolites 16 (1995) 107. [25] S.T. Wilson, B.M. Lok, E.M. Flanigen, US Patent, 4.310.440, 1982. [26] V.R. Vijayaraghavan, K. Joseph Antony Raj, J. Mol. Catal. A: Chem. 207 (2004) 41. [27] R. Dimitrova, Y. Neinska, M.R. Mihalyi, T. Tsoncheva, M. Spassova, React. Kinet. Catal. Lett. 74 (2001) 353. [28] O.A. Anunziata, A.R. Beltramone, F.G. Requejo, J. Mol. Catal. A: Chem. 267 (2007) 194. [29] F.G. Requejo, J.M. Ramallo-Lopez, E.J. Ledo, E.E. Miro, L.B. Pierella, O.A. Anunziata, Catal. Today 54 (1999) 553. [30] R.W. Hartford, M. Kojima, C.T. O’Connor, Ind. Eng. Chem. Res. 28 (1989) 1748. [31] F. Bigi, L. Chesini, R. Maggi, G. Sartori, J. Org. Chem. 64 (1999) 1033.
Contents lists available at ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Lewis acid metal ion-exchanged MAPO-5 molecular sieves for solvent free synthesis of coumarin derivative S. Gopalakrishnan, K.R. Viswanathan, S. Vishnu Priya, J. Herbert Mabel, M. Palanichamy, V. Murugesan * Department of Chemistry, Anna University, Chennai 600 025, Tamil Nadu, India
a r t i c l e
i n f o
Article history: Received 7 March 2008 Received in revised form 26 July 2008 Accepted 28 July 2008 Available online 5 August 2008 Keywords: Magnesium aluminophosphate Ion-exchange TPD (ammonia) Coumarin
a b s t r a c t Magnesium aluminophosphate (MAPO-5) molecular sieve was synthesised hydrothermally and ionexchanged with Lewis acid metal ions such as Ga3+, In3+, La3+ and Ce3+ by wet method. The materials were characterised using XRD, FT-IR, SEM, TPD (ammonia) and TGA. The XRD patterns of ion-exchanged molecular sieves are similar to the parent MAPO-5 thus showed absence of structural degradation and non-framework metal oxide phase during ion-exchange. The thermograms of metal ion-exchanged MAPO-5 showed a weight loss between 550 and 600 °C which is assigned to the decomposition of + MðOHÞþ 2 into MO species. The total acidity of materials decreased after ion-exchange as revealed from TPD (ammonia) studies. Their catalytic activity was tested in the solvent free synthesis of coumarin. Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction Synthesis of aluminophosphate molecular sieves has attracted great attention because of their potential application in catalysis, adsorption and separation [1–4]. The major disadvantage of parent AlPO molecular sieve is the absence of acid sites. Hence incorporation of Bronsted acid sites is necessary. The incorporation of divalent cations such as Mg2+, Zn2+, Co2+, Mn2+, etc into AlPO and SAPO results MAPO and MSAPO, respectively [5–7]. Such isomorphous substitution introduces charge imbalance in the framework which is balanced by protons, thus generating Bronsted acidity and offering catalytic activity and ion-exchange capability [8]. Mg2+ substituted aluminophosphate molecular sieve exhibited strong Bronsted acidity and it was reported to be active catalyst in acid catalysed chemical reactions [9–11]. Among the aluminophosphate molecular sieves AlPO-5 has attracted much attention due to its large pore size with 0.7 nm pore opening. It is composed of 12-membered rings of regular alternation of Al and P in the tetrahedral framework [12]. The catalytic application of both transition and non-transition divalent metal ions substituted AlPO-5 molecular sieves has been reported already [13,14]. Although there are reports on the catalytic activity of metal ions substituted AlPO-5 molecular sieves with Bronsted acid sites as catalytically active sites, Lewis acid ion-exchanged MAPO-5 catalysts yet to receive considerable importance. However, Lewis acid ion-exchanged zeolites have been reported to be catalytically active for many reactions of industrial importance [15,16]. Tynjala and Pakkanen [17] * Corresponding author. Tel.: +91 44 22203144; fax: +91 44 22200660/22350397. E-mail address: [email protected] (V. Murugesan). 1566-7367/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2008.07.032
compared the acidic properties of Ba2+, Al3+ and La3+ ion-exchanged ZSM-5 zeolite. This report revealed clearly that during ion-exchange with trivalent metal ions, the splitting of water molecules results in the formation of new Bronsted acid sites and acidic MðOHÞþ 2 based on Plank–Hirschler mechanism. This is more probable in La3+ and Al3+ ion-exchanged zeolite than Ba2+. Rane et al. [18] reported wet ion-exchange of Ga3+ in ZSM-5 zeolite which is more Lewis acidic and exhibit high catalytic active in the cracking of n-heptane. Mavrodinova et al. [19–22] reported the introduction of InO+ and AlO+ into HY and Hb zeolites. These reports revealed the formation of MO+ species during calcination. Considering in the above lines, there is no report on Ga3+, In3+ and rare-earth metal-ions-exchanged MAPO-5 molecular sieves. In the present study we report the preparation of Ga3, In3+, La3+ and Ce3+ ion-exchanged MAPO-5, their characterisation and catalytic activity in the liquid phase condensation of salicylaldehyde and diethyl malonate. 2. Experimental 2.1. Preparation of catalysts Hydrothermal synthesis of MAPO-5 was carried out using a gel composition of 0.1MgO:1TEA:1Al2O3:1P2O5:40H2O adopting the procedure reported already [23]. Aluminium isopropoxide (Merck), phosphoric acid (Merck) (88%) and magnesium acetate (Merck) were used as the sources for Al, P and Mg, respectively. Aluminium isopropoxide (28.37 g) was soaked in 30 ml distilled water for 24 h in a stainless steel autoclave (316 type) and stirred vigorously for 1 h. While stirring, 1.48 g magnesium acetate dissolved in 7.8 ml
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
phosphoric acid and 20 ml distilled water were added drop by drop and the stirring continued for additional 1 h. Triethylamine (9.7 ml) was added drop wise to it and the pH of the gel was found to be 4.4. The autoclave was tightly closed and kept at 175 °C under autogeneous pressure for 24 h. Then it was cooled to room temperature to obtain the solid product. The product was washed several times with distilled water and dried in an air oven at 110 °C for 12 h. It was then calcined at 550 °C for 8 h to remove the template. The final composition of MAPO-5 was found to be 0.07MgO:0.93Al2O3:1P2O5. The ion-exchanged materials were prepared by stirring 2 g of the calcined MAPO-5 with corresponding metal nitrate solution (30 ml; 0.1 M) for 12 h at 80 °C. The solid was filtered, washed thoroughly with distilled water and dried at ambient temperature. The same procedure was repeated thrice. These samples were used for characterisation and catalytic studies.
a
Intensity (a.u)
24
b
c
d
e 5
10
15
20
25
30
35
40
2 Theta (degree)
X-ray diffraction patterns were recorded on a PANalytical X’pert PRO diffractometer using nickel filtered Cu Ka (k = 0.154 nm) radiation and a liquid nitrogen cooled germanium solid state detector. The diffractograms were recorded in the 2h range from 5 to 40° in steps of 1.2° min1 with a count time of 10 s at each point. Fourier transform infrared spectra (FT-IR) of the materials were recorded on a FT-IR spectrometer (Nicolet Avatar 360) using KBr pellet technique. About 15 mg of the sample was pressed (under a pressure of 2 tons cm2) into a self-supported wafer of 13 mm diameter. This pellet was used to record the infrared spectra in the range 400– 4000 cm1. Thermogravimetric analysis (TGA) of the materials was performed on a high resolution thermogravimetric analyser (Perkin–Elmer diamond series) under nitrogen atmosphere in the temperature from 50 to 800 °C at a heating rate of 10 °C min1. The size and morphology of the samples were recorded using a scanning electron microscope (SEM) (JEOL 640). The density and strength of acid sites were determined by temperature programmed desorption (TPD) of ammonia on a Micromeritics chemisorb 2750 pulse chemisorption system. About 0.08 g of the sample was pretreated in an oxygen stream at 450 °C for 3 h. Ammonia was adsorbed at 100 °C and the physisorbed ammonia was removed by applying vacuum. The temperature was then increased up to 600 °C in a helium stream (20 ml min1) at a heating rate of 10 °C min1 and the desorption was monitored by a thermal conductivity detector (TCD) [24]. The liquid products obtained in the reaction were analysed by a gas chromatograph (Shimadzu GC-17A; crosslinked 5% phenyl methyl siloxane capillary column, FID detector). Further, the products were also confirmed using a gas chromatograph coupled with Turbo mass spectrometer (GC–MS Perkin–Elmer auto system XL (EI, 70 eV) using helium as the carrier gas at a flow rate of 1 ml min1.
3. Results and discussion 3.1. Characterisation The powder X-ray diffraction patterns of MAPO-5 and metal ion-exchanged MAPO-5 catalysts are shown in Fig. 1. The patterns of MAPO-5 are similar to that of AlPO-5 [25,3]. All the ion-exchanged XRD patterns are almost similar to the patterns of MAPO-5. However, the patterns corresponding to non-framework oxide of exchanged metal ions were not seen and hence such oxides were absent or the dimension of such oxides, if present, may not be in sufficient dimension to be detected by XRD. The patterns also revealed absence of structural degradation during ion-ex-
Fig. 1. XRD patterns of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
change with metal nitrates. The FT-IR spectra of calcined MAPO5 catalysts are shown in Fig. 2a. The broad envelope between 2750 and 3700 cm1 in all the spectra corresponds to O–H stretching modes of surface hydroxyls (P–OH and Bronsted acid sites). The presence of water is confirmed by its OH2 bending vibration close to 1630 cm1. The water content appears to be different in these catalysts. The broad absorption band around 1000 cm1 is due to tetrahedral asymmetric vibrations of P–O–Al and P–O–M groups. The corresponding symmetric vibration gives a peak close to 750 cm1. The peak around 500 cm1 corresponds to the bending mode of T–O–T. These are the characteristic vibrations of MAPO5 molecular sieves [26]. The spectra of LaMAPO-5, CeMAPO-5, GaMAPO-5 and InMAPO-5 are shown in Fig. 2b–e. The spectral characteristics of ion-exchanged MAPO-5 are almost similar to MAPO-5. However, the increase in the peak height corresponding to OH stretching vibration for ion-exchanged MAPO-5 is due to the OH stretching modes of surface hydroxyls interacting with water and also water adsorbed on Lewis sites. Thermogravimetric analysis of metal ion-exchanged MAPO-5 was carried out between 30 and 800 °C. However, the thermograms shown in Fig. 3a–d are drawn between 400 and 650 °C for clarity. There is a characteristic weight loss between 550 and 600 °C. This weight loss is assigned to decomposition of MðOHÞþ 2 into MO+ species where M is the trivalent ions used for ion-ex-
a
Transmittance (%)
2.2. Physicochemical characterization
b c d e
3500
3000
2500
2000
1500
1000
500
Wavenumber (cm-1) Fig. 2. FT-IR spectra of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
25
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
3.65
Weight loss (%)
3.60 a
3.55
b 3.50 c 3.45 d
3.40
3.35 400
450
500
550
600
650
Temperature (°C ) Fig. 3. TGA curves of (a) LaMAPO-5, (b) CeMAPO-5, (c) InMAPO-5 and (d) GaMAPO5.
TCDSignal (a.u)
change. This is the most supportive evidence for the formation of MO+ species. The formation of MðOHÞþ 2 is based on Plank–Hirschler mechanism as discussed by Tynjala and Pakkanen [17]. The SEM pictures of MAPO-5 and ion-exchanged MAPO-5 catalysts are shown in Fig. 4a–e. The SEM picture of MAPO-5 shows a clear hexagon. The edges and sides are clearly distinct in addition to well grown hexagon. There are particles with irregular morphology. The size of these particles appears to be less than the size of hexagonal crystal. The hexagon also appears to be agglomerated
but along the same plane. The surface of hexagon does not appear to possess any smaller grains on its terrace. The SEM picture of GaMAPO-5 is shown in Fig. 4e. This hexagon also shows well developed edges and phases. The tiny grains without definite morphology are also present. The SEM image of CeMAPO-5 is shown in Fig. 4d. The fully developed hexagon and its precursor with different sizes and morphologies are evident. Both La and InMAPO-5 also possess hexagonal morphology but the surface appears different. Agglomeration of plate hexagon is clearly evident from the Fig. 4b and c. The acidity of the catalysts was determined by temperature programmed desorption (TPD) of ammonia technique. TPD (ammonia) was carried out between 100 and 650 °C for calcined MAPO-5 and ion-exchanged MAPO-5 catalysts. The results are depicted in Fig. 5. This profile matches with already reported for boron-beta zeolite, beta zeolite and H-ZSM-5 modified with indium [27–29]. The amount of strong and weak acid sites present in these catalysts was calculated from the volume of ammonia desorbed and the values are presented in Table 1. All the metal ion-exchanged catalysts showed higher amount of weak acid sites than MAPO-5. The weakly acidic nature of La3+ ions has been reported already in La3+ ion-exchanged H-ZSM-5 [30]. In addition La3+ ions were not shown to reduce the effective pore volume of the catalyst or increase the steric hindrance for the passage of linear molecules through the pores of the catalyst [30]. According to Plank–Hirschler mechanism, the highly charged metal ions like trivalent cations undergo hydrolysis in aqueous solution producing MðOHÞþ 2 species. Hence the actual species involved in the ion-exchange may be þ MðOHÞþ 2 . The formation of MðOHÞ2 takes place most probably in the case of smaller cations with high polarizing power. However, these ions are dehydrated to produce MO+ species during calcination [17]. This is also verified by TGA analysis as discussed above. During the ion-exchange of M3+ ions there will be the loss of protons in the framework. However, there may be splitting of coordinated water molecules around M3+ ions to form two protons and MðOHÞþ 2 species. These two newly formed protons will be taken up by two oxygen bridges of the framework. These two protons may not be as much acidic as that of actually exchanged strongly acidic protons. This confirms that weak acid sites are generated as a result of ion-exchange. These weak acid sites include weak Bronsted acid sites, Lewis acid sites and defective sites [30]. Hence
a b
c d e 100
150
200
250
300
350
400
450 500
550
600
650
Temperature (°C ) Fig. 4. SEM pictures of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
Fig. 5. Temperature programmed desorption (TPD) of ammonia profiles of (a) MAPO-5, (b) LaMAPO-5, (c) CeMAPO-5, (d) InMAPO-5 and (e) GaMAPO-5.
26
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Table 1 TPD (ammonia) results of MAPO-5 and ion-exchanged MAPO-5 Catalyst
Temperature maximum (°C)
Volume (mL/g at STP)
Weak and strong acid sites (mmol/g)
MAPO-5
168.5 580.2
1.1717 0.1955
0.0523 0.0087
LaMAPO-5
148.6 574
1.9667 0.1116
0.0878 0.005
CeMAPO-5
172.7 506.7
1.745 0.0697
0.0779 0.0031
InMAPO-5
205.3 585.9
1.9504 0.06
0.0871 0.0027
GaMAPO-5
164 522.2
2.2106 0.0581
0.0987 0.0026
as a result of metal-ion-exchange some of the strong acid sites may be converted to weak acid sites. The data also indicate feasibility of ion-exchange of strong acid sites. The order of ion-exchange of strong acid sites is found to be GaMAPO-5 > InMAPO-5 > CeMAPO-5 > LaMAPO-5. Hence Ga3+, In3+, La3+ and Ce3+ may not exist as M3+ during ion-exchange but as MðOHÞþ 2 as predicted by Plank–Hirschler mechanism. The same may be converted to MO+ during calcination as shown below:
Table 2 Conversion of salicylaldehyde, yield and selectivity of 3-ethyl coumarin carboxylate over different catalysts Catalyst
Temperature (°C)
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
MAPO-5
100 120 140 160
21.5 23 25.7 28.8
89.3 90.4 91 92
19.2 20.8 23.4 26.5
LaMAPO-5
100 120 140 160
31 36.6 44.7 53.4
90.9 91.2 91.4 92.6
28.2 33.4 40.9 49.5
CeMAPO-5
100 120 140 160
34.8 42.3 47.2 55.8
86.4 89.3 92.5 93.1
30.1 37.8 43.7 52
InMAPO-5
100 120 140 160
40.2 47.5 51.4 61.3
89 91.1 91.6 94.6
35.8 43.3 47.1 58
GaMAPO-5
100 120 140 160
46.9 56.3 64.4 71
91.2 92.5 92.7 95.9
42.8 52.1 59.7 68.1
Calcination
þ þ MðH2 OÞ3þ 6 ! MO þ 5H2 O þ 2H
Reaction conditions: catalyst amount 0.1 g, time 12 h, feed ratio 1:3 (salicylaldehyde: DEM).
3.2. Catalytic performance Intermolecular cyclisation of salicylaldehyde and diethyl malonate was carried over MAPO-5 and ion-exchanged MAPO-5 at 100, 120, 140 and 160 °C in the liquid phase. The feed ratio was kept at 1:3 (salicylaldehyde: diethyl malonate) and the reaction was carried for 12 h in all the cases. The major product was found to be 3-ethyl coumarin carboxylate. The results of salicylaldehyde conversion, selectivity and yield of products over all the catalysts are presented in Table 2. The salicylaldehyde conversion and selectivity to 3-ethyl coumarin carboxylate increased with increase in temperature. The reaction required initial condensation between carbonyl group of salicylaldehyde and active methylene group of diethyl malonate on the active sites like Bronsted acid sites. There may be enolisation of diethyl malonate and the enolised species reacted with salicylaldehyde in a sequential manner giving the precursor of 3-ethyl coumarin carboxylate as shown in the reaction Scheme 1. In addition, the reaction may also involve protonation at the ester carbonyl followed by nucleophilic attack of the salicylaldehyde –OH group as shown in the reaction Scheme 2. The latter appeared to be more probable over catalyst possessing Bronsted acid sites. The subsequent intermolecular cyclisation of the precursor also requires acid sites as protonation of carbonyl oxygen of the aldehyde group can facilitate easy condensation with the active methylene group at the ester portion. Since the precursor is not observed as product, its conversion appears to be rapid. Salicylaldehyde conversion increased with increase in temperature and so the reaction appeared to be activation energy demanding. The yield of 3-ethyl coumarin carboxylate increased with increase in temperature and the selectivity of product is above 86% at all temperatures. Comparison of the results of salicylaldehyde conversion and product yield shows that MAPO-5 is the least active among the catalysts. This revealed that the reaction is largely controlled by Lewis acid sites. LaMAPO-5 showed higher conversion of salicylaldehyde and yield of 3-ethyl coumarin carboxylate than MAPO-5. This observation established that metal ion dependent route is an additional one to the Bronsted acid site catalysed route. The metal ion
COOCH2CH3 H2C _ 2 Mg
+ H O
OH
+C + P
COOCH2CH3 + H2C
_ 2 Mg
OCH 2CH3
O + P
COOCH2CH3 OH CHO
O
O O C CH2 CHO C O OCH 2CH3 O C + CH CH C O
O
O C
+ H MAPO
CH + CH
C O
HO H O
OCH 2CH3 O COOCH2CH3
OCH 2CH3 Scheme 1. Intermolecular condensation of salicylaldehyde and diethyl malonate over MAPO-5.
may play a template role in bringing together salicylaldehyde and diethyl malonate in its co-ordination sphere and simultaneously allowing them to react further to yield the product. Co-ordination of both the reactants to LaO+ may be quite possible as it has high co-ordination capability. The important observation is that ion-exchange of La3+ mainly occurs on strong acid sites leaving weak acid sites unaffected as discussed above. Hence, the metal ion dependent route is more pronounced than Bronsted acid sites. The reaction over CeMAPO-5 exhibited nearly similar trend in terms of conversion and product yield as that of LaMAPO-5. However, there is further increase in conversion of salicylaldehyde, product yield and selectivity of 3-ethyl coumarin carboxylate over CeMAPO-5. Therefore Ce3+ may play a similar role as that La3+. The conversion and yield are higher over CeMAPO-5 than LaMAPO-5 since Ce3+ sites are more Lewis acidic than La3+ sites. The higher Le-
27
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Table 4 Effect of catalyst amount on salicylaldehyde conversion, yield and product selectivity over GaMAPO-5
OCH2CH3 OH +
CHOO=C
H
CH2CH3
O O=C CH2
O=C
O
+ MO
CH2 -CH CH OH 3 2
MO+
O=C
C=O
OCH 2CH3
OCH 2CH3
H O
O
C=O MO+
C=O MO+
CH 2
C=O O=C H
C=O
OCH 2CH3
C H
C
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
0.05 0.10 0.15
55.8 71.0 73.5
86.5 95.9 95.2
48.3 68.1 70.0
Reaction condition: temperature 160 °C, time 12 h, feed ratio 1:3 (salicylaldehyde: DEM).
H H O C OCH 2 CH3
O -H2O
C=O
-MO+
CH
Catalyst amount (g)
O
O C
COOCH 2 CH3
COOC2 H5
OH H
Scheme 2. Intermolecular condensation of salicylaldehyde and diethyl malonate over MO+.
wis acidity could be attributed to smaller size of Ce3+ than La3+ due to lanthanide contraction. The results of conversion and product yield obtained over InMAPO-5 also illustrate the same trend as that of La3+ and Ce3+ exchanged catalysts. But both conversion and yield over InMAPO-5 are higher than La3+ and Ce3+ exchanged catalysts. This may be due to its high Lewis acid strength as a result of the smallest size. The results of GaMAPO-5 illustrate similar results as that of InMAPO-5 but the conversion and yield are higher than InMAPO5. GaMAPO-5 is more active than all other catalysts as Ga3+ is the smallest ion. Since Ga3+ is a small size ion, it can form more stable complex than others. This complex also exhibits sufficient lability to catalyze the condensation. The yield of the product is also higher over GaMAPO-5 than other catalysts. Bigi et al. [31] reported the same condensation reaction over montmorillonite KSF and K10. They reported a maximum yield of 46% at 160 °C without any solvent and with a feed ratio 1:2. The results obtained with GaMAPO5 therefore demonstrate that it is better than the catalysts already reported. The effect of feed ratio (salicylaldehyde: diethyl malonate) viz, 1:1, 1:3 and 1:5 on salicylaldehyde conversion, yield and product selectivity was studied over GaMAPO-5 and the results are shown in Table 3. The feed ratio 1:3 showed higher conversion, yield and selectivity of 3-ethyl coumarin carboxylate than other feed ratios. The low conversion at 1:1 feed ratio may be due to less amount of diethyl malonate which could suppress co-adsorption of diethyl malonate and salicylaldehyde. The conversion and product yield with feed ratio 1:5 are lower than 1:3. This may be attributed to dilution of salicylaldehyde by diethyl malonate. Similar trend is also reflected in the yield and selectivity of 3-ethyl coumarin carboxylate and hence the optimum feed ratio is 1:3.
Table 5 Effect of reaction time on salicylaldehyde conversion, product selectivity and yield over GaMAPO-5 Time (h)
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
3 6 9 12 18
48.0 57.0 64.8 71.0 70.2
88.9 90.0 90.2 95.9 95.8
42.7 51.3 58.5 68.1 67.3
Reaction condition: temperature 160 °C, catalyst amount 0.1 g, feed ratio 1:3 (salicylaldehyde: DEM).
The effect of catalyst loading on salicylaldehyde conversion and product selectivity was studied using 0.05, 0.1 and 0.15 g catalyst with a constant feed ratio of 1:3 at 160 °C and the results are presented in Table 4. The conversion increased with increase in the catalyst amount. Therefore once the product formed it should be immediately desorbed from the catalyst surface without allowing it to chemisorb again. The selectivity of 3-ethyl coumarin carboxylate also increased with increase in catalyst amount from 0.05 to 0.15 g. GaMAPO-5 with more active sites is better than other catalysts. This study revealed that the reaction still depends on the number of active sites. So the optimum amount of catalyst is found to be 0.1 g. The effect of reaction time on conversion and product yield was carried out and the results are presented in Table 5. The conversion increased with increase in reaction time up to 18 h beyond which the conversion remained steady illustrating the attainment of equilibrium. Based on the results of conversion it could be visualised that the reactants may be reacting in a sequential manner on the metal ion to form 3-ethyl coumarin carboxylate. Ga3+ and InMAPO-5 showed better results than La3+ and CeMAPO-5. It is also concluded that templating role of metal ions during condensation is better at 1:3 than at 1:1. This revealed that salicylaldehyde may be better co-ordinating than diethyl malonate. 4. Conclusion
Table 3 Effect of feed ratio on salicylaldehyde conversion, yield and product selectivity over GaMAPO-5 Feed ratio
Salicylaldehyde conversion (%)
Selectivity of 3-ethyl coumarin carboxylate (%)
Yield of 3-ethyl coumarin carboxylate (%)
1:1 1:3 1:5
64.0 71.0 67.7
86.4 95.9 90.5
55.3 68.1 61.3
Reaction condition: temperature 160 °C, catalyst amount 0.1 g, time 12 h.
The study established the feasibility of ion-exchange in MAPO-5 with Lewis acid metal ions such as La3+, Ce3+, Ga3+ and In3+. The exchanged metal ions are suggested to remain as MO+ as charge compensating cation based on Plank–Hirschler mechanism. It is also concluded that strong acid sites are exchanged with Lewis acid metal ions during ion-exchange The catalytic activity of Lewis acid metal ion-exchanged MAPO-5 materials is better than Bronsted acid sites for the condensation of salicylaldehyde and diethyl malonate to form coumarin derivative. The absence of side reaction is also one of the important features of the catalysts.
28
S. Gopalakrishnan et al. / Catalysis Communications 10 (2008) 23–28
Acknowledgements The authors gratefully acknowledge the financial support from the Defence Research and Development Organisation (DRDO), Government of India, New Delhi, for this research work. One of the authors (Mr. S. Gopalakrishnan) is grateful to DRDO for the award of Senior Research Fellowship (SRF). References [1] S.T. Wilson, B.M.C. Lok, A. Messina, T.R. Cannan, E.M. Flanigen, J. Am. Chem. Soc. 104 (1982) 1146. [2] C. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite Structure Types, Elsevier, London, 2001. [3] J.M. Bennett, W.J. Dytrych, J.J. Pluth, J.W. Richardson Jr., J.V. Smith, Zeolites 6 (1986) 349. [4] J.M. Thomas, R. Raja, G. Sankar, R.G. Bell, Acc. Chem. Res. 34 (2001) 191. [5] E.M. Flanigen, B.M. Lok, R.L. Patton, S.T. Wilson, Stud. Surf.Sci. Catal. 46 (1989) 191. [6] Z. Levi, A.M. Raitsimring, D. Goldfarb, J. Phy. Chem. 95 (1991) 191. [7] C. Montes, M.E. Davis, B. Murray, M. Narayanan, J. Phy. Chem. 94 (1990) 6425. [8] E.M. Flanigen, R.L. Patton, S.T. Wilson, in: P.J. Gorbet, W.J. Mortier, E.F. Vansant, G. Schulz-Ekloff (Eds.), Innovation in Zeolite Materials Science, Elsevier, Amsterdam, 1988, p. 13. [9] R.J. Gorte, G.T. Kokotailo, A.I. Biaglow, D. Parrillo, C. Pereira, Stud. Surf. Sci. Catal. 69 (1991) 181. [10] D.B. Akolekar, J. Catal. 144 (1993) 148. [11] L.T. Yuen, S.I. Zones, T.V. Harris, E.J. Gallegos, A. Auroux, Micropor. Mater. 2 (1994) 105. [12] J.M. Bennett, W.J. Dytrch, J.J. Pluth, J.W. Richardson, J.V. Smith, Zoelites 6 (1986) 349.
[13] M.S. Rigutto, H. Van Bekkum, J. Mol. Catal. 81 (1993) 77. [14] M.J. Happen, A.M. Elemans-Mehring, J.H.C. Van Hoff, Appl. Catal. A: Gen. 152 (1997) 203. [15] V.V. Balasubramanian, A. Pandurangan, M. Palanichamy, V. Murugesan, Ind. J. Chem. Tech. 7 (2000) 149. [16] V.V. Balasubramanian, V. Umamaheswari, I. Sudarsankumar, M. Palanichamy, V. Murugesan, Proc. Ind. Acad. Sci. (Chem. Sci.) 110 (1998) 453. [17] P. Tynjala, T.T. Pakkanen, J. Mol. Catal. A: Chem. 110 (1996) 153. [18] N. Rane, M. Kersbulck, R.A. Van Santen, E.J.M. Hensen, Micropor. Mesopor. Mater. 110 (2008) 279. [19] V. Mavrodinova, M. Popova, Y. Neinska, Ch. Minchev, Appl. Catal. A: Gen. 210 (2001) 397. [20] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, Appl. Catal. A: Gen. 248 (2003) 197. [21] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, Appl. Catal. A: Gen. 262 (2004) 75. [22] V. Mavrodinova, M. Popova, R.M. Mihalyi, G. Pal-Borbely, Ch. Minchev, J. Mol. Catal. A: Chem. 220 (2004) 239. [23] K. Joseph Antony Raj, V.R. Vijayaraghavan, J. Chem. Sci. (Ind. Acad. Sci.) 116 (2004) 107. [24] P. Concepcion, J.M. Lopez Nieto, J. Perez-Pariente, A. Misfud, Zeolites 16 (1995) 107. [25] S.T. Wilson, B.M. Lok, E.M. Flanigen, US Patent, 4.310.440, 1982. [26] V.R. Vijayaraghavan, K. Joseph Antony Raj, J. Mol. Catal. A: Chem. 207 (2004) 41. [27] R. Dimitrova, Y. Neinska, M.R. Mihalyi, T. Tsoncheva, M. Spassova, React. Kinet. Catal. Lett. 74 (2001) 353. [28] O.A. Anunziata, A.R. Beltramone, F.G. Requejo, J. Mol. Catal. A: Chem. 267 (2007) 194. [29] F.G. Requejo, J.M. Ramallo-Lopez, E.J. Ledo, E.E. Miro, L.B. Pierella, O.A. Anunziata, Catal. Today 54 (1999) 553. [30] R.W. Hartford, M. Kojima, C.T. O’Connor, Ind. Eng. Chem. Res. 28 (1989) 1748. [31] F. Bigi, L. Chesini, R. Maggi, G. Sartori, J. Org. Chem. 64 (1999) 1033.