1-s2.0-s0926860x05002188-main.pdf

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

1/9

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

2/9

product, thereby requiring azeotropic distillation to remove

the water formed during the reaction, or else leading to the

deactivation of the catalyst[10]. Due to this, the equilibrium

for the stoichiometric mixture reached in the liquid phase

was about 6668% [11] of conversion for straight-chain

saturated alcohol; complete conversion can only be achieved

by elimination of the water formed. But the same reaction isthermodynamically favourable when performed in the

vapour phase due to the higher values of equilibrium

constants in comparison with those of the liquid phase

reaction[12]. Hence, in order to increase the conversion, the

reaction was performed in the vapour phase with the use of

mesoporous catalyst, Al-MCM-41, which was discovered by

Mobils central research laboratory in 1992 [13,14]. Such

catalysts can easily be separated from the product and

reactants byfiltration and can also be regenerated with ease

[15]. The activity values presented here are far better than

reported by Michael Verhoef et al.[7],who found a very low

activity for MCM-41 (Si/Al = 16) in terms of reactant

conversion and product selectivity. In the present study we

have found excellent catalytic activity of H-MCM-41 for the

esterification of acetic acid with NBA, IBA and TBA under

autogeneous pressure in the batch process.

2. Experimental

2.1. Materials

The syntheses of Al-MCM-41 materials were carried out

by a hydrothermal method using sodium metasilicate

(Na2SiO35H2O), aluminum sulfate (Al2SO418H2O), cetyl-trimethylammonium bromide (C16H33(CH3)2N

+Br), and

sulfuric acid (H2SO4). The AR grade chemicals used were

purchased from Aldrich & Co., USA.

2.2. Commercial catalytic materials

HM (Si/Al = 12, PQ), Hb (Si/Al = 8, PQ), HY (Si/Al = 4,

PQ), H-ZSM-5 (Si/Al = 15, PQ), H3PW12O40nH2O,

H3PMo12O40nH2O, H4SiW12O40nH2O were obtained from

commercial sources.

2.3. Synthesis of Al-MCM-41

The Al-MCM-41 with various Si/Al ratios: 25, 50, 75

and 100, were synthesised according to the previous report

[13,16]using a hydrothermal method with the gel composi-

tion of SiO2:xAl2O3:0.2CTAB:0.89H2SO4:120H2O, sodium

meta silicate was used as the silicon source, cetyltrimethy-

lammonium bromide as the structure directing agent and

aluminium sulphate as the aluminium source. Sodium meta

silicate (21.21 g) was dissolved in 80 ml of water and the

mixture was stirred for half an hour. Then the required

quantity of aluminiumsulphate, whichwas dissolved in 15 ml

of water,was added and this was stirred for 1 h. Then40 ml of

4N sulphuric acid was added drop by drop until the gel

formed. The stirring was continued for 2 h. Exactly 7.28 g of

cetyltrimethylammonium bromide (CTAB), dissolved in

25 ml of water, was added and stirring was continued for a

further 2 h. After that, the gel was transferred to an autoclave

that was kept in a hot air oven at 145 8C for 36 h. Then the

product obtained was filtered, washed several times withdoubledistilled water, anddriedat 80 8C inan air ovenfor2 h.

Then thesample was calcinedin a mufflefurnaceat550 8Cfor

6 h to remove the template. The sample calcined by this

procedure was ion exchanged repeatedly with one molar

solution ammonium nitrate and thenfiltered, dried, calcinedat

550 8C for 12 h.

2.4. Catalytic runs

Esterification reactions were carried out under batch

reaction conditions using an autoclave in the temperature

range of 100200 8C, in 15 ml stainless steel batch reactors

under autogeneous pressure conditions. A typical reaction

mixture in the reactor contained acetic acid (0.1 mol),

alcohol (0.1 mol) and a freshly activated catalyst (0.1 g).

Activation of the catalyst was done by calcinations at 500 8C

in air for 5 h. The autoclave temperature was then slowly

raised to 100, 125, 150, 175 and 200 8C as required and

maintained at the desired temperature during the specified

reaction periods. The effect of the reaction period, the molar

ratios of the reactants, and the amounts of catalyst required

on various alcohol conversions and product selectivity were

studied.

2.5. Analysis

The reaction mixture was collected from the autoclave

after it had been cooled to room temperature. This solution

was removed from the catalyst by filtration. The reaction

mixture was analysed by an Schimadzu gas chromatograph

GC-17A using a DB-5 capillary column with an FID

detector. Comparing the retention values of the known

standards with those of the products confirmed the latter.

The product analysed by GC revealed the formation ofn-

butyl acetate, isobutyl acetate and tertiary butyl acetate with

100% selectivity. Some small peaks corresponding to the

dehydrated products of alcohols were also observed. The

percentage conversion and selectivity calculation are based

on the GC analysis. The selectivity to a product is expressed

as the percentage weight of the product alkyl acetate divided

by sum of the weight percentage of the entire product.

%Conversion initial wt:% final wt:%

initial wt:% 100

%Product selectivity

wt:% of product

sum of the wt:% values of all products 100

B.R. Jermy, A. Pandurangan / Applied Catalysis A: General 288 (2005) 253326

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

3/9

2.6. Characterisation

Mesoporous materials in general are characterized by a

variety of techniques including XRD (Rigaku, D-Max/111-

VC model) using nickel filtered Cu K a radiation

l= 1.5406 A. Surface area, pore volume and pore size

distribution were measured by nitrogen adsorption at 77 Kusing an ASAP-2010 porosimeter from Micromeritics

Corporation, GA. The samples were degassed at 623 K

and 105 Torr overnight prior to the adsorption experiments.

The mesopore volume was estimated from the amount of

nitrogen adsorbed at a relative pressure of 0.5 by assuming

that all the mesopores werefilled with condensed nitrogen in

the normal liquid state. Pore size distribution was estimated

using the Barrett, Joyner and Halenda (BJH) algorithm

(ASAP-2010) available as built-in software from Micro-

meritics. Mid-infrared spectra of MCM-41 molecular sieves

were recorded on a Nicolet (Avatar 360) instrument using a

KBr pellet technique. About 4 mg of the sample was ground

with 200 mg of spectral grade KBr to form a mixture, which

was then made into a pellet using a hydraulic press. This

pellet was used for recording the infrared spectra in the

range 4000400 cm1. Thermal analysis was carried out in

Mettler TA 3001 analyser. Zeolites used in this study were of

commercial origin.

2.7. Acidity measurements

The acidity of Al-MCM-41 was analysed by pyridine

adsorption followed by FT-IR spectroscopy. Finely ground

catalyst sample (1015) was pressed for 2 min at

10 Torr cm2

pressure under vacuum) into a self-supportingwafer. The wafers were calcined under vacuum

(133.322 103 N m2) at 500 8C for 2 h, followed by

exposure to pyridine vapour at ambient temperature for 1 h

to allow the pyridine to permeate the samples. Each thin

wafer was placed in the FT-IR cell and the spectrum was

recorded in absorbance mode on a Nicolet 800 (AVATAR)

FT-IR spectrometer, fully controlled by the OMNIC

software, within an all-glass high-vacuum system. The

difference between the spectra of pyridine adsorbed on the

samples and that of the reference was obtained by

subtraction.

3. Results and discussion

3.1. Characterization of Al-MCM-41

3.1.1. XRD

The diffraction patterns are shown inFig. 1and the data

are presented in Table 1. The patterns illustrate the

characteristics of a typical mesoporous MCM-41 structure.

As can be seen from the diffraction patterns, the d100reflections of calcined Al-MCM-41 have been shifted to

higher values compared to its as-synthesised analogue.

This is in agreement with Borade and Clearfield [17],

suggesting the framework substitution of alumina in MCM-

41 structure.During calcinations at 550 8C, thedvalues are generally

shifted towards the lower values or higher 2uvalues, though

to a smaller extent, implying shrinkage in the unit cell as a

result of the removal of the surfactant molecules used as

templates[18].

3.1.2. Nitrogen adsorption isotherms

BET surface area, pore size and pore volume of calcined

materials are presented in Table 2. Adsorption and

desorption isotherms and pore size distribution for calcined

materials (BJH method) are shown inFigs. 2 and 3; the data

coincide with the reported values[19,20].It can be seen that

B.R. Jermy, A. Pandurangan / Applied Catalysis A: General 288 (2005) 2533 27

Fig. 1. X-ray diffraction patterns of calcined Al-MCM-41: (a) Al-MCM-41

(25); (b) Al-MCM-41 (50); (c) Al-MCM-41 (75); and (d) Al-MCM-41

(100).

Table 1

Textural properties of the calcined catalysts (Si/Al = 25, 50, 75 and 100)

Catalysts Si/Al Calcined

d100(A) Unit cell a0 (nm)

Al-MCM-41 (25) 25 35.38 4.09

Al-MCM-41 (50) 50 40.60 4.69

Al-MCM-41 (75) 75 42.50 4.90

Al-MCM-41 (100) 100 42.50 4.90

Table 2

Surface area, pore size and pore volume of the catalysts

Catalysts Surface area

(m2 g1)

Pore size:

BJHAds (nm)

Pore volume:

BJHAds(cc g1)

Al-MCM-41 (100) 1023 2.644 0.9575

Al-MCM-41 (75) 1018 2.631 0.9540

Al-MCM-41 (50) 976.6 2.538 0.9407

Al-MCM-41 (25) 950.8 2.501 0.9457

HM (12) 431

Hb (8) 694

HY (4) 821

HZSM-5 (15) 393

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

4/9

MCM-41 presents the highest surface area and pore volume,

with all pores being in the mesopore range. The pore size

distribution of calcined MCM-41 shows a unique peak

centered at about 25 A diameter (Fig. 3) as given in the

literature [21]. All samples exhibited type IV isotherms, with

capillary condensation steps occurring at a partial pressure

corresponding to HorvathKawazoe pore size distributions

centered around 25 A. The BET surface areas were

calculated by fitting the straight part of the p/x(p po)

versusp/pocurve (wherepis the pressure of nitrogen, andx

is the number of grams of adsorbed nitrogen per gram ofsolid). The resulting surface area ranged from 950.8 to

1023 m2 g1.

3.1.3. FT-IR Spectroscopy

The FT-IR spectra of the as-synthesised and calcined

samples are given in Figs. 4 and 5, respectively. The

presence of absorption bands around 2921 and 2851 cm1

for the as-synthesised materials corresponds to asymmetric

and symmetric CH2 vibrations of the surfactant molecules.

In the FT-IR spectrum of calcined samples, comparison of

the broad envelope due toOH stretch of water in the higher

energy region and the corresponding OH2 bending mode

around 1637 cm1 very well correlate with the water

adsorption property (hydrophilic property) of the catalysts.

The intensity of the bands due to water at the catalyst

decreases in the order MCM-41 (25) > MCM-41

(50) > MCM-41 (75) > MCM-41 (100), which is also the

order of the hydrophilic property of the catalysts.

3.1.4. FT-IR spectroscopy of pyridine-adsorbed samplesThe FT-IR spectra for Al-MCM-41 (25), Al-MCM-41

(50), Al-MCM-41 (75) and Al-MCM-41 (100), containing

adsorbed pyridine are presented inFig. 6.It is observed that

all the catalysts have both Bronsted and Lewis acid sites. A

typical sharp peak appearing at 1545 cm1 is the indication

of pyridine adsorbed on Bronsted acid sites. A small peak at


Fig. 2. Adsorption isotherms of nitrogen on MCM-41 at 77 K: (a) Al-

MCM-41 (100); (b) Al-MCM-41 (75); (c) Al-MCM-41 (50); and (d) Al-

MCM-41 (25).

Fig. 3. Pore size distributions in Al-MCM-41 (adsorption isotherms): (a)

Al-MCM-41 (100); (b) Al-MCM-41 (75); (c) Al-MCM-41 (50); and (d) Al-

MCM-41 (25).

Fig. 4. FT-IR spectra of as-synthesised mesoporous materials (using KBr

method): (a) Al-MCM-41 (25); (b) Al-MCM-41 (50); (c) Al-MCM-41 (75);

and (d) Al-MCM-41 (100).

Fig. 5. FT-IR spectra of calcined mesoporous materials (using KBr

method): (a) Al-MCM-41 (25); (b) Al-MCM-41 (50); (c) Al-MCM-41

(75); and (d) Al-MCM-41 (100).

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

5/9

1455 cm1 and a high-intensity peak around 1620 cm1

indicate the pyridine adsorbed on Lewis acid sites.

3.1.5. Thermal analysis

The thermal properties of the samples were investigated

by TGA. Their data are presented in Table 3 and the

thermograms are presented inFig. 7. The initial weight loss

up to 120 8C is due to desorption of physically adsorbed

water. Weight loss from 120 to 350 8C is due to organic

template. The oxidative desorption of the organic template

takes place at 180 8C; the minute quantity of weight loss

above 350 to 550 8C is related to water loss from the

condensation of adjacent SiOH groups to form siloxane

bonds [22]. It is seen that, as the Al content in MCM-41

framework increases, the amount of water desorbed

increases and the organic species decrease, which confirms

the decrease in the hydrophobic character of the catalyst

with increasing Al content.

3.2. Application of Al-MCM-41 (25), Al-MCM-41 (50),

Al-MCM-41 (75) and Al-MCM-41 (100) catalysts to the

esterification of acetic acid

3.2.1. Esterification

The esterification of acetic acid with various alcohols

is an electrophilic substitution. The reaction is relatively

slow and needs activation either by high temperature or

by a catalyst to achieve higher conversion to a reasonable

amount. The effects of various parameters on the

esterification reaction are discussed later. This study was


Fig. 6. FT-IR spectra of: (a) Al-MCM-41 (25); (b) Al-MCM-41 (50);

(c) Al-MCM-41 (75); and (d) Al-MCM-41 (100) catalysts containing

adsorbed pyridine (by self-supporting disk).

Fig. 7. TGA and DTA spectra of uncalcined mesoporous Al-MCM-41: (a)

Al-MCM-41 (100); (b) Al-MCM-41 (75); (c) Al-MCM-41 (50); and (d) Al-

MCM-41 (25).

Table 3

TGA spectral data of as-synthesised mesoporous Al-MCM-41 (Si/Al = 25,

50, 75 and 100) molecular sieves (in the presence of air)

Catalyst Weight loss (wt.%)

Total 43120 8C 120350 8C 350680 8C

Al-MCM-41 (25) 54.29 3.89 32.32 18.08

Al-MCM-41 (50) 51.17 5.09 29.71 16.37

Al-MCM-41 (75) 54.26 4.77 33.58 15.91

Al-MCM-41 (100) 50.44 5.26 31.75 13.44

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

6/9

also extended to heteropolyacids such as H3PW12O40nH2O,

H3PMo12O40nH2O, and H4SiW12O40nH2O. The zeolites

that are tested are H-Mordenite, Hb, H-ZSM-5 and HY in

order to have a comparative understanding of the catalytic

activity and selectivity of the products for the reaction.

3.2.2. Variation with reaction temperature

The reaction was carried out at various reactiontemperatures, ranging from 100 to 200 8C at a given alcohol

to acetic acid ratio of 1:2 for 8 h over Al-MCM-41 with

various Si/Al ratios: 25, 50, 75 and 100. The results are given

in Table 4. A common trend in the conversion over all

catalysts is an increase in alcohol conversion with increase

in temperature. Hence, the reaction requires activation

energy. The activation energy may be required to reduce

intermolecular associations of n-butanol for dispersed

adsorption and to avoid clustering of alcohols around the

Bronsted acid sites by hydrogen bonding. For each catalyst,

the conversion at the particular temperature decreases in the

order NBA > IBA > TBA, since each alcohol after chemi-

sorption on the Bronsted acid sites is to give a carbonium ion

for nucleophilic reaction with acetic acid. The degree of

positive charge of carbonium ion may be important to

account the difference in their conversion. Since NBA can

give a carbonium ion of high degree of positive charge

compared to that of IBA and TBA, the conversion for

esterification with NBA becomes higher than other alcohols.

Isobutyl cation and tertiary butyl cation are to have less

degree of positive charge thenn-butyl cation due to hyper-

conjugation. Hence the conversion becomes less than that

withn-butanol. Further, it may also be inferred that it is not

the formation of carbonium ion that is important in this

esterification, but it is the reactivity of the carbonium that is

important. Since the temperature employed in this reaction

is sufficiently high, the rate of formation of carbonium ion

may not be the slow step in the esterification. The activities

for NBA and TBA were compared at 125 8C, while that for

IBA was compared at 150 8C. The kinetics data under

pseudo-first-order conditions are also derived; the results are

presented inFig. 8. The rate constants derived by the first-order plot for all the three alcohols are 0.0495, 0.0286 and

0.026, respectively. The rate constants also follow the order

K1B > K2I > K3T, which also matches the activity of the

catalyst.


Table 4

Catalytic activity of Al-MCM-41 (25, 50, 75 and 100) for esteri fication of acetic acid with various alcohols

Catalyst Temperature (8C) %Conversion of NBA %Conversion of IBA %Conversion of TBA

Al-MCM-41 (100) 100 58.5 56.5 19.5

125 78.2 50.0 35.7

150 79.0 71.6 36.2

175 81.0 70.6 36.8

200 83.2 71.6 36.9

Al-MCM-41 (75) 100 60.9 66.9 18.0

125 70.3 61.9 33.6

150 80.0 71.2 33.9

175 85.2 71.3 34.9

200 87.1 72.2 35.0

Al-MCM-41 (50) 100 58.2 56.9 15.0

125 79.0 54.9 33.9

150 80.0 69.8 33.8

175 87.2 74.0 34.4

200 88.2 75.1 34.5

Al-MCM-41 (25) 100 61.6 57.2 11.2

125 87.3 41.0 33.2

150 88.5 55.8 33.5175 88.7 66.5 34.2

200 90.1 69.7 34.4

Time = 8 h; feed ratio = 1:2 (alcohol:acid).

Fig. 8. Relation betweenln (1 conversion) and time (h) with catalyst

loading of 0.05 g ml1 for NBA and 0.15 g ml1 for IBA and TBA,

respectively. Reaction conditions: temperature = 125 8C for NBA and

TBA; 150 8C for IBA; feed ratio = 1:4 (alcohol:acid).

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

7/9

3.2.3. Influence of reaction time

The effect of reaction period on the esterification was

studied on Al-MCM-41 (25) at the optimised temperature of

125 8C with the feed ratio of 1:2 for NBA and on Al-MCM-

41 (100) for IBA and TBA at 150 and 125 8C, respectively.

Results are presented inFig. 9. Conversion with respect to

NBA increases from 60 wt.% at 2 h to a maximum of

88.30 wt.% at 10 h. Similarly, the conversion with respect to

IBAwas 66.41 wt.% at 2 h to a maximum of 80.68 wt.%. For

TBA, the conversion at 2 h was 28 wt.%; after 10 h, it is

35.84 wt.%. As seen from the table, in the case on NBA, 6 h

of reaction time completes 80.0 wt.% of the reaction,

whereas at the end of 10 h, 88.0 wt.% of the reaction is

complete. Similarly, in the cases of IBA and TBA, 9.0 and5.0 wt.% increase in the conversion was found to occur after

6 and 4 h. A gradual rise in the conversion were seen with

increase in the duration of the reaction period. This

observation is a normal feature for reaction procedures of

this kind.

3.2.4. Influence of mole ratio of the reactants

The effects of different feed ratio on NBA, IBA and TBA

conversion were studied. Different feed ratios were used

over Al-MCM-41 (25) for NBA at 125 8C for 6 h and Al-

MCM-41 (100) at 150 8C, and 125 8C for IBA andTBA for 6

and 4 h, respectively. The results are presented in Table 5.

When the feed increased from 1:1 to 1:5, a non-linear trend

in alcohol conversion was observed. The conversion of NBA

increased from 1:1 to 1:3, followed by a decrease thereafter.

Since acetic acid probably gets chemisorbed on the Bronstedacid sites, increase in acetic acid content increased the

conversion. The decrease above 1:3 might be due to dilution

of alcohol by excess acetic acid, thus preventing its

accessibility for nucleophilic reaction with chemisorbed

acetic acid. A similar trend in alcohol conversion was also

observed for IBA and for TBA. The optimum feed ratio for

NBA was found to be 1:3, for IBA 1:2 and for TBA 1:5,

respectively. Although the optimum feed reaction for TBA is

1:5, 1:3 might be sufficient, because the increase in

conversion for the change of feed ratio from 1:3 to 1:5

only about 2%. The reaction was also studied by taking more

alcohol content in the feed. The conversion was not much

altered for a change of feed ratio from 2:1 to 5:1 in the case

of NBA. Although the increase in alcohol content might be

expected to increase conversion, there might be dilution of

acetic acid, hence the expected increase in conversion must

be balanced by decrease in the conversion due to increase in

dilution of acetic acid. The conversion of IBA for similar

variation in the feed ratio decreased. The same trend was

also observed for the conversion of TBA. Hence in both

these cases there might be dilution of acetic acid by

increased alcohol contents, thereby preventing acetic acid

adsorption on the Bronsted acid sites. In addition, both

isobutyl and TBA can yield their corresponding carboca-

tions, which are more hydrophobic and stearically hinderedthan those from NBA; as a result, the reaction with acetic

acid for the reverse nucleophilic reaction might be more

hindered than with NBA. So far, this study can conclude that

an increase in the acetic acid content in the feed is better for

esterification than a decrease in the acetic acid content in the

feed.


Fig. 9. The effect of reaction time on the esterification of acetic acid over

Al-MCM-41 (25) for NBA and Al-MCM-41 (100) for IBA and TBA.

Reaction conditions: temperature = 125 8C for NBA and TBA; 150 8C for

IBA; feed ratio = 1:2 (acid:alcohol).

Table 5

Effect of feed ratio on esterification of acetic acid over Al-MCM-41 (100)Mole ratio %Conversion

of NBA

%Conversion

of IBA

%Conversion

of TBA

1:1 50.4 47.1 24.6

1:2 80.5 71.6 35.8

1:3 91.8 71.7 38.3

1:4 87.7 60.1 40.0

1:5 85.0 60.4 41.6

2:1 50.4 63.3 16.2

3:1 42.8 33.3 13.1

4:1 56.6 26.9 11.3

5:1 57.1 20.0 9.3

Temperature = 125 8C for NBA and TBA; 150 8C for IBA. Time = 6 h for

NBA and IBA and 4 h for TBA.

Fig. 10. The effect of amount of catalyst on the esterification of acetic acid

over Al-MCM-41 (25) for NBA and Al-MCM-41 (100) for IBA and TBA.

Reaction conditions: catalyst = Al-MCM-41 (25); temperature = 125 8C for

NBA and TBA; 150 8C for IBA; feed ratio = 1:2 for IBA and 1:3 for NBA

and TBA; time = 6 h for NBA and IBA and 4 h for TBA.

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

8/9

3.2.5. Influence of the catalyst loading

The effect of catalyst loading on alcohol conversion was

studied at various optimised conditions over Al-MCM-41

(25) for NBA and Al-MCM-41 (100) for IBA and TBA. The

results are presented in Fig. 10. With increase in catalyst

loading, conversion increased. Increase was not much for

NBA. The optimum loading can taken as 0.05 g. SimilarlyIBA conversion increased with increase in catalyst loading,

but the optimum loading may be taken as 0.15 g. Similarly,

for TBA conversion the optimum loading is taken to be

0.15 g. The requirement of higher catalyst loading for IBA

and TBA as compared to that for NBA might be due to the

effect of hydrophobic property of the alcohols, which could

be reduced by increase in the catalyst loading. There might

be some prevention of adsorption of acetic acid on the

catalyst surface at low loading.

3.2.6. Comparative study on various zeolites

The esterification of acetic acid with various alcohols was

also tested with different catalysts at optimised temperature

of 125 8C with the feed ratio of 1:3 for NBA and TBA for 6 h

and 150 8C with the feed ratio of 1:2 for IBA for 4 h. The

results are presented inTable 6. Except zeolites, all catalysts

were found to be more active. The lower activity for zeolites

might be due to their micropores, with which the diffusional

constraints for the reaction and the product might be more.

The zeolites offer resistance to diffusion for the reactant in to

the pores, as well as for the product out of the pores. So from

this study it can be concluded that Al-MCM-41 material can

also be exploited as a catalyst for this reaction in addition to

zeolites, sulphated zirconia and HPA catalyst. The HPA

catalyst showed nearly the same activity as that of Al-MCM-41 molecular sieves for all the three-esterification reactions.

This observation clearly supported molecular concession

free esterification inside the pores of MCM-41. The reaction

was also performed with as-prepared Al-MCM-41 (25) and

Al-MCM-41 (100) under the same conditions at the

temperature of 125 8C for NBA and TBA and 150 8C for

IBA with the feed ratio of 1:2. The alcohol conversion found

to be very less for all the three alcohols. This clearly proves

that the reaction occurs within the pores of Al-MCM-41

molecular sieves. In addition, Al-MCM-41 (100) was tested

for its recyclability by running the reaction three times. Nochange in conversion was observed, illustrating the stability

of the catalyst. In order to verify the active influence of the

catalysts, we also studied the reaction in the absence of the

catalyst under the optimum conditions: feed ratio of 1:2 at

125 8C for NBA, TBA for 6 h and 150 8C for IBA for 4 h.

After completion of the reaction, the reaction mixture was

analysed. The acetic acid conversions for NBA, IBA and

TBA were found to be 13, 10 and 14 wt.%. These are 74, 61

and 20% less than the results in the presence of catalyst.

Further in order to know whether the reaction occurs mainly

inside the pores, outside the pores or both, the study was

carried out with as-synthesised catalyst, which shows the

conversions 28, 12 and 13 wt.% for NBA, IBA and TBA,

respectively. Whereas, the calcined materials conversion

was 87.3, 71.6 and 35.7 wt.%, respectively, which was 58,

59 and 22 wt.% less conversion than calcined sample. So

the reaction is more prone to occur within the pores of the

catalyst rather than on the outer surface.

4. Conclusions

From the studies on the esterification of acetic acid over

various protonated Al-MCM-41 with different Si/Al ratios,

the following conclusions can be drawn. Al-MCM-41molecular sieves can be conveniently exploited from the

esterification of acetic acid with NBA, IBA and TBA. For

NBA, Al-MCM-41 (25) was found to be more active where as

Al-MCM-41 (100) found to be more active for IBA and TBA.

The hydrophobicity of the catalyst surface and the hydro-

phobicity of the alcohol are also found to be deciding factors.

In the esterification, the reaction was found to follow Eley

Rideal type with chemisorption of acetic acid and nucleo-

philic attack of alcohol. The activity of Al-MCM-41 was also

comparable to HPA and sulphated zirconia catalyst. Zeolite

was found to be less active than either MCM-41 molecular

sieves or HPA-supported catalysts with respect to NBA

conversion. The reaction over as-prepared catalyst showed

68% less conversion than that of calcined sample. Hence, the

reaction is proposed to occur mainly within the pores of

the catalyst. This observation indirectly proves planting of

Bronsted acid sites insides the pores of the catalyst.

Acknowledgement

The authors would like to thank All India Council for

Technical Education (8020/RID/R&D-94/2001-02) for

providingfinancial support.


Table 6

Effect of various catalysts on esterification of acetic acid over Al-MCM-41

(100)

Catalyst %Conversion

of NBA

%Conversion

of IBA

%Conversion

of TBA

Al-MCM-41 (25) 87.3 55.8 33.2

Al-MCM-41 (50) 75.5 69.8 33.9

Al-MCM-41 (75) 70.3 71.2 33.6Al-MCM-41 (100) 80.5 71.6 35.7

H3PW12O40nH2O 79.0 72.7 36.6

H3PMo12O40nH2O 84.8 72.4 35.1

H4SiW12O40nH2O 86.9 69.8 34.6

HM (12) 40.4 60.4 38.2

Hb (8) 48.0 54.0 22.3

HY (4) 53.5 57.1 25.7

HZSM-5 (15) 42.7 60.2 25.4

Without catalyst 13.2 10.1 14.1

As-synthesised

Al-MCM-41 (100)

28.5 12.4 13.3

Temperature = 125 8C for NBA and TBA; 150 8C for IBA. Time = 6 h for

NBA and IBA and 4 h for TBA. Feed ratio = 1:2 for IBA and 1:3 for NBA

and TBA. Catalyst loading for NBA = 0.05 g, 0.15 g for IBA and TBA.

8/11/2019 1-s2.0-S0926860X05002188-main.pdf

9/9

Reference

[1] R. Koster, B. van der Linden, E. Poels, A. Bliek, J. Catal. 204 (2001)

333.

[2] J. Gimenez, J. Costa, S. Cervera, Ind. Eng. Chem. 26 (1987) 198.

[3] H.B. Zhang, B.Z. Zhang, H.X. Li, J. Nat. Gas Chem. 1 (1992) 49.

[4] A. Corma, H. Garcia, S. Iborra, J. Primo, J. Catal. 120 (1989) 78.

[5] Z.H. Chen, T. Lizuka, K. Tanabe, Chem. Lett. (1984) 1085.[6] M. Hino, K. Arata, Chem. Lett. (1981) 1671.

[7] J. Michael Verhoef, J.P. Kooyman, A. Joop Peters, H. van Bekkum,

Microporous Mesoporous Mater. 27 (1999) 365.

[8] W. Chu, X. Yang, X.K. Ye, Y. Wu, Appl. Catal. A: Gen. 145 (1996)

125.

[9] Y.O. Li, Petrochem. Technol. 54 (1981) 309 (in Chinese).

[10] S.E. Sen, Tetrahedron 55 (1998) 12657.

[11] F.A. Lowenheim, M.K. Moran, Industrial Chemicals, Wiley/Inter-

science, New York, 1975.

[12] P.W. Hawes, R.L. Kabel, AIChEJ 14 (1968) 606.

[13] J.S. Beck, J.C. Vartuli, W.I. Roth, M.E. Leonowicz, C.T. Kresge, K.D.

Schmidt, C.T.W. Chu, D.H. Olson, F.W. Sheppard, S.B. McCullen,

J.B. Higgins, J.I. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834.

[14] C.T. Kresge, M.E. Leonowicz, M.E. Roth, J.C. Vartuli, J.S. Beck,

Nature 114 (1992) 10834.

[15] P. Selvam, Ind. Eng. Chem. Res. 40 (2001) 15.

[16] M. Selvaraj, A. Pandurangan, K.S. Seshadri, P.K. Sinha, V. Krishna-

samy, K.B. Lal, J. Mol. Catal. 186 (2002) 173.

[17] R.B. Borade, A. Clearfield, Catal. Lett. 31 (1995) 267.

[18] R. Maheswari, K. Shanthi, T. Sivakumar, S. Narayanan, Appl. Catal.

A: Gen. 245 (2003) 221.

[19] S.J. Greggand, K.S.W. Sing, Adsorption, Surface Area and Porosity,

second ed., Academic Press, New York, 1982.

[20] T.R. Pauly, Y. Liu, T.J. Pinnavaia, S.J.L. Billinge, T.P. Rieler, J. Am.

Chem. Soc. 121 (1992) 8835.

[21] A. Corma, A. Martinez, V. Martinez-Soria, J.B. Morton, J. Catal. 153

(1995) 25.

[22] M.L. Occelli, S. Biz, A. Auroux, G.J. Ray, Microporous Mesoporous

Mater. 26 (1998) 193.


1-s2.0-s0926860x05002188-main.pdf

Documents