Iron aluminophosphates (FeAlP) with different percentage of iron were synthesized and characterized for their surface and bulk properties. The catalytic activity was determined in the transesterification of diethyl malonate with benzyl alcohol. Benzyl ethylmalonate and dibenzyl malonate were obtained as the only products. FeAlP with 0.025 mole % of iron was found to be distinctly different in its textural and catalytic properties. Formation of diester was found to be favored by the acid sites of intermediate strength. The presence of hydrated alumina and the polycondensed phosphates in the materials reduced the catalytic activity of iron aluminophosphates in transesterification reaction.
Iron compounds catalyse several organic reactions
amorphous transition metal aluminophosphates have also been found effective catalysts in several organic transformations.
Recently we have observed iron aluminophosphate exhibited distinctly different catalytic activity compared to other transition metal phosphates. The catalytic activity, selectivity and yield of the products were found to depend on the method of preparation of the catalysts as well as on the nature of the transition metal. These observations prompted us to further work on iron aluminophosphates to find out the role of iron and phosphate species and also understand the scientific basis for the catalytic activity of these materials in the transesterification reaction between diethyl malonate and benzyl alcohol. To the best of our knowledge such studies have not been reported in the literature.
- Preparation of Catalysts
Aluminophosphate (AlP) and iron phosphate (FeP):
Pure Aluminophosphate (AlP) samples with three different mole ratios (Al and P = 0.95:1, 1:1and 1:0.95) were prepared by coprecipitation method. Aluminum nitrate (Al(NO
O) and 85 % of H
were mixed in the desired molar ratio in 500 mL of deionised water and heated to 60 ~ 70℃ to get a homogeneous solution. To this hot solution 28% liquor ammonia was added drop wise from a micro burette. Aluminium phosphate gel thus obtained was separated by filtration, washed and dried at 120 ℃ in a hot air oven for 12 hours. The solids thus prepared were powdered and further calcined at 550 ℃ for 5 hours in a muffle furnace. A sample of iron phosphate (FeP) was also prepared by a similar method by taking ferric nitrate (Fe(NO
O) in the place of aluminum nitrate.
Iron aluminophosphates (FeAlP):
Iron aluminophosphates with 0.01, 0.025, 0.05 and 0.1 molar concentration of iron were also prepared by co precipitation technique as described above using a mixture of Aluminum nitrate (Al(NO
O), ferric nitrate [Fe(NO
O] and 85% of H
with Al : P ratio of 0.95 : 1.
- Characterization of catalysts
BET specific surface area of the materials was determined using NOVA 1000 Quantachrome high speed gas sorption analyser using liquid nitrogen as the adsorbent. The samples prior to their BET analysis were degassed at 120 ℃ for 2 hours then the usual adsorption-desorption procedure was followed to obtain the data on specific surface area , pore diameter and pore volume. The PXRD diffraction patterns were obtained with Xpert Pro Philips diffractrometer equipped with a Ni filtered Cu-Kα radiation with λ = 1.5418 Å using a graphite crystal monochromator in 2θ range 5 ~ 80° FTIR spectra were recorded using Nicolet IR 200 instrument by KBr pellet technique. The acid strength of the catalysts was evaluated by Temperature Programmed Desorption of NH
gas using Puls Chemisorb 2705 from Micrometrics. SEM studies were carried out in a JEOL JED-2300 analysis station apparatus.
- Catalytic activity studies
Transesterification of diethyl malonate with benzyl alcohol was carried out in liquid phase under refluxing conditions. After a predetermined period the reaction mixture was cooled and the solid catalyst was separated by centrifugation. The components in the filtrate were analyzed by gas chromatography using 10% carbowax column and further by GC-MS.
RESULTS AND DISCUSSION
The composition and surface properties of the catalysts are given in
In the case of iron aluminophosphates (FeAlP) the numbers given in the brackets correspond to the mole % of iron in the respective sample.
Composition & surface properties of catalysts
amg of NH3 per gm of catalyst.
- Surface area
The BET surface area of AlP, iron phosphate (FeP) and FeAlP calcined to 550 ℃ was in the following increasing order, FeP < AlP < FeAlP. An increase in the mole% of Phosphorus in AlP and Fe in FeAlP decreased the surface area of the respective samples. At the outset these changes in surface area qualitatively indicate the effect of phosphorous and iron on the textural properties of AlPs and FeAlPs.
The decrease in surface area with increase in phosphorous content in AlP may be attributed to the formation of poly condensed phosphates,
whereas the decrease in the surface area of FeAlP with increase in iron loading due to the iron pyrophosphate.
It is worth noting that FeAlP samples containing 0.025 mole% of iron exhibited highest surface area. This may be attributed to the reduced possibility of poly condensed phosphates formation due to the utilisation of all the phosphate ions in the formation of FeP and AlP in the presence of phosphate as the limiting reagent.
- Total surface acidity (NH3-TPD)
Surface acidity of the calcined AlP, FeP and FeAlP in terms of mg of NH
/g of catalyst are presented in
The values are calculated from the following standard method. The area of the peak obtained from the graph of temperature Vs desorption of NH
gives the amount of ammonia desorbed from each sample. The TPD-NH
profile for FeAlP is given in
The AlP samples showed lower acidity than FeAlP samples. Further among the different AlP samples the one which had lowest mole% of P showed highest surface acidity. In the case of FeAlP samples total acidity increased with an increase in iron content from 0.012 to 0.025 mole%. Loading iron beyond 0.025 mole%, the acid strength decreased, this may be attributed to the formation of poly phosphate. It is supported by the fact that this effect was not observed when the iron content is < 0.025 mole%.
NH3-TPD profile for FeAlP samples.
- Adsorption-desorption isotherms
These studies were useful in identifying the nature of the pore structure of the catalysts. The isotherm of AlP-1 (
) was a straight line. This indicated an unclear nature of pores in the material. The adsorption-desorption isotherms of FeAlP samples calcined at 550 ℃ for 5 h are given in
The shapes of these isotherms are of the type-IV and hysteresis loops H1 type, characteristic of mesoporous materials.
It is to be noted that the starting point of the hysteresis loop is displaced towards higher relative pressures on increasing iron loading indicating a decrease in pore volume. FeAlP is thus found to have mesoporous structure. Campello and co-workers have also made similar kind of observations with metal aluminophosphates and attributed this behaviour to continuous restructuring of the porous texture resulting in a variation of the mesopore diameter and pore volume due to the different iron content in the material.
Adsorption- desorption isotherm of AlP-1.
Adsorption desorption isotherms of FAlP samples. (a) Fe(0.012) AlP (b) Fe(0.025)AlP (c) Fe(0.050)AlP (d) Fe(0.100)AlP.
- Powder X ray diffraction
The PXRD patterns of both calcined and uncalcined AlP, FeP (not given) and FeAlP samples except the one containing 0.025% of iron exhibited only a broad peak in the 2θ range of 20° ~ 30° (
). This indicated an amorphous nature and small particle size of the materials. However, FeAlP sample containing 0.025 mole % of iron exhibited sharp diffraction peaks at 2θ = 23.3, 26.99, 38.5 corresponding to FePO
PXRD patterns of FAlP samples with different iron loading, (a) Fe(0.012)AlP (b) Fe(0.025)AlP (c) Fe(0.050)AlP (d) Fe(0.100)AlP samples calcined at 550 ℃.
SEM observation showed a very widely varied distribution in morphology, texture and particle sizes for all the samples. SEM micrographs of AlP and FeAlP samples are shown in
(a) SEM picture of AlP-1 (b) SEM picture of Fe(0.025)AlP.
The FTIR absorption spectra of AlP, FeP & FeAlP samples are given in the
respectively. It is to be noted that all the samples exhibited characteristic absorption peaks due to the asymmetric vibrations of phosphate in the range of 1100 ~ 1128 cm
. The IR spectra also exhibited shoulder peaks at 725 and 507 cm
which are respectively assigned to the symmetric stretching mode of P-O-P and bending mode of O-P-O bonds in pyrophosphate. The absorption bands in the range 3430 ~ 3448 cm
are due to the surface hydroxyl groups.
Thus FTIR analysis indicated the formation of pyrophosphate.
FTIR spectra of AlP-1 sample.
FTIR spectra of FeP sample.
FTIR spectra of FAlP samples with different iron loading (a) Fe(0.012)AlP (b) Fe(0.025)AlP (c) Fe(0.050)AlP (d) Fe(0.100)AlP samples.
Hence, analysis of the materials for their acidity, crystallinity and adsorption-desorption properties clearly indicated that FeAlP containing 0.025 mole % of Fe had distinctly different textural properties.
Al-NMR investigations on FeAlP containing 0.025 mole% of iron and calcined at 550 ℃ for 5 h exhibited two signals, a broad peak at 38.1 and short one at -13.5 ppm assigned to four and six coordinated Al species respectively (
a). This defines the sharing of oxygen atoms with four tetrahedra of phosphorus [Al(OP)
]. It was also proposed that six coordination of Al species resulted from the coordination with not only PO
units but also water molecules.
P NMR spectra exhibited one broad signal at -26.6 ppm which support the presence of P(-OAl)
units with, the P-O-Al bonds in a tetrahedral coordination environment (
b). A. Sayari
and Z. Luan
have assigned the value in the region -19.0 to -30.0 ppm are due to full condensation of phosphorous species. These facts clearly support limited poly condensation of PO
H NMR showed a single signal at -5 ppm indicating the presence of highly shielded environment of proton.
(a) 27Al-MAS NMR of Fe(0.025)AlP sample. (b) 31P-MAS NMR of Fe(0.025)AlP sample.
- Catalytic activity
Transesterification of diethyl malonate (DEM) with benzyl alcohol (BA) was found to be effective in the presence of all the metal phosphates used. The major products were found to be only benzylethyl malonate (BEM) and dibenzyl malonates (DBM). However the yield and selectivity of the products were significantly influenced by the composition and the surface acidity of the catalyst.
Catalytic activity of the AlP, FeP & FeAlP in transesterification reaction of DEM with BA
Reaction conditions: amount of catalyst = 0.1g, DEM : BA = 1 : 3, time = 4 h
Aluminium phosphates (with different Al/P ratios) and Iron phosphate exhibited 100% selectivity for BEM (
entries 1-4). The total surface acidity of these samples was also found to be in the same order. NH
-TPD studies on these samples revealed that these materials had only weak acid sites. From this it may be inferred that the selectivity towards the formation of BEM is associated with the weak acid sites.
The lower catalytic activity of the sample with higher concentration of Al (Al/P > 1) may be attributed to the basic Al(OH)
that was precipitated along with the AlPO
. Low catalytic activity of iron phosphate indicated the presence of such acid sites in very low concentration.
The catalytic activity of FeAlP samples is given in the
in terms of the selectivity and percentage yield towards transester products. It is interesting to note that the presence of Fe in aluminophosphate resulted in the formation of the diester, dibenzyl malonate (DBM). The optimum amount of Fe for the formation of diester was found to be 0.025 mole % of Fe. Further the catalytic activity was also found to depend on the iron loading. The catalytic activity of various FeAlP’s is in the order Fe(0.025)AlP > Fe(0.012)AlP > Fe(0.05)AlP > Fe(0.1)AP.
It is reported earlier that acidity of catalysts was one of the surface properties that favour the formation of a diester.
This aspect is further confirmed by our observation based on the NH
-TPD studies that iron containing catalysts possessed higher acidity compared to AlP and FeP. The catalyst, Fe(0.025)AlP which exhibited stronger acidity among the other FeAlP’s also exhibited greater diester formation activity. The stronger acidity of this sample may be due to the generation of Lewis acid sites by the Fe species present in association with aluminium species. It is also noteworthy that this sample had acid sites of intermediate strength as evidenced by TPD-NH
adsorption studies. This observation further makes it clear that the intermediate strength acid sites are catalytically active in the formation of dibenzyl esters where as the weaker acid sites are active for the mono benzylester. Thus it may be inferred from these studies that iron loading in iron aluminophosphate influence the surface acidity by altering its textural properties. This aspect may thus be taken as a guiding principle to alter the surface properties of iron aluminophosphate to direct its catalytic activity to increase the selectivity either towards the monoester or diester in the transesterification between DEM and BA.
The textural properties of the FeAlP catalyst are influenced by the amount of iron loading. The presence of iron reduces the crystallinty and the particle size of the pure aluminophosphate. The acid strength of the pure aluminophosphate increases with incorporation of iron. The catalytic activity depends on the amount of iron present in the catalyst. The mole ratio of Al:Fe:P is 0.95:0.025:1 in FeAlP is most suitable for the synthesis of mono and dibenzyl substituted malonates. The high surface area, surface acidity and mesoporous nature of Fe-(0.025)AlP were responsible for its better catalytic activity. The presence of acid sites of intermediate strength favored the formation of the dibenzyl substituted esters. The formation of hydrated alumina and polycondensed phosphates in the catalysts reduced the catalytic activity of the material. Thus the amount of iron present in catalyst can be used as a controlling measure for determining the surface properties and catalytic properties of the amorphous iron aluminophosphates.
We thank the MRG at SJC Research Centre, Bangalore for PXRD and BIT, Bangalore for BET analysis, we also thank NMR centre I.I.Sc. Bangalore for NMR analysis and UGC, New Delhi for financial assistance.
Bautista F. M.
Campelo J. M.
Raquel M. L.
Marinas J. M.
Romero A. A.
Jose A. N.
J. Mater. Chem.
DOI : 10.1039/a807538g
Reddy J. S.
Ratcliffe C. I.
Ripmeester J. A.
Preston K. F.
DOI : 10.1021/cm960029v