Al-MCM-41

Al-MCM-41

Catalysis Communications 10 (2009) 1486–1492 Contents lists available at ScienceDirect Catalysis Communications journal homepage: www.elsevier.com/l...

369KB Sizes 6 Downloads 13 Views

Catalysis Communications 10 (2009) 1486–1492

Contents lists available at ScienceDirect

Catalysis Communications journal homepage: www.elsevier.com/locate/catcom

Synthesis of paracetamol by liquid phase Beckmann rearrangement of 4-hydroxyacetophenone oxime over H3PO4/Al-MCM-41 M. Ghiaci a,*, H. Aghaei a, M. Oroojeni a, B. Aghabarari a, V. Rives b, M.A. Vicente b, I. Sobrados c, J. Sanz c a

Department of Chemistry, Isfahan University of Technology, Isfahan 8415683111, Iran GIR-QUESCAT, Departamento de Química Inorgánica, Universidad de Salamanca, 37008 Salamanca, Spain c Instituto de Ciencia de Materiales, C.S.I.C., C/Sor Juana Inés de la Cruz, 3, 28049 Cantoblanco-Madrid, Spain b

a r t i c l e

i n f o

Article history: Received 22 January 2009 Received in revised form 18 March 2009 Accepted 24 March 2009 Available online 31 March 2009 Keywords: Paracetamol 4-hydroxyacetophenone oxime Beckmann rearrangement H3PO4/MCM-41 catalysts

a b s t r a c t Paracetamol was synthesized via the environmentally benign liquid phase Beckmann rearrangement of 4-hydroxyacetophenone oxime over Al-MCM-41 modified with H3PO4. Al-MCM-41 materials were synthesized using different aluminium sources. Aprotic solvents with a high dielectric constant or high polar nature are preferred for this transformation, acetone being the optimal solvent. The conversion of 4hydroxyacetophenone oxime to paracetamol increased with the H3PO4 content in the catalysts up to a maximum value for a H3PO4 loading of 30 wt% and decreased for higher loadings. Recovery of the catalysts was also investigated. Ó 2009 Elsevier B.V. All rights reserved.

1. Introduction The rearrangement of ketoximes to amides or lactams in the presence of acid catalysts is a process commonly used for the preparation of caprolactam, the monomer used in the production of nylon 6. Paracetamol (N-acetyl-p-aminophenol or acetaminophen) is a valuable non-steroidal anti-inflammatory drug in widespread use for the management of pain and fever in a variety of patients, including children, pregnant women, the elderly; and those with osteoarthritis, simple headaches, and non-inflammatory musculoskeletal diseases [1,2]. Since paracetamol was first synthesized by Morse in 1878, it has been one of the most widely used analgesics and an important commodity. Conventionally, large-scale preparation of paracetamol mainly employs the acetylation of p-aminophenol with acetic anhydride. As the main use for paracetamol is as a pharmaceutical, the presence of impurities must be kept at very low values. In the mid 1980s, Davenport and Hilton [3] reported an innovative technology for the preparation of paracetamol that involves a two-step process. The first step involves reacting 4-hydroxyacetophenone with hydroxylamine hydrochloride to obtain the ketoxime (4-hydroxyacetophenone oxime) followed by the Beckmann rearrangement in the presence of an acid catalyst, such as fuming sulfuric, hydrochloric, trifluoroacetic, methanesulfonic or p-toluenesulfonic acids, amberlyst, nafion, or thionyl chloride in liquid sulfur dioxide. The use of homogeneous acid catalysts * Corresponding author. Tel.: +98 311 3912351; fax: +98 311 3912350. E-mail address: [email protected] (M. Ghiaci). 1566-7367/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2009.03.025

requires tedious workup procedures and the necessary neutralization of the strong media, producing undesired wastes. Thus, when sulfuric acid is used as a catalyst, paracetamol is recovered from the reaction mixture by neutralization of the oleum with aqueous ammonia. In this case a large amount of ammonium sulfate is formed. Fritch et al. [4] have described an alternative to the above Beckmann rearrangement process that uses an alkyl alkanoate ester as a solvent and thionyl chloride or phosphorous oxytrichloride as the acid catalyst to give high conversion of paracetamol. The use of insoluble acid catalysts will allow an easy separation workup (no neutralization step required) and catalyst recycling, avoiding equipment corrosion and contaminant wastes. This has been largely attempted for the Beckmann rearrangement of cyclohexanone oxime to caprolactam in vapor or liquid phase, with the use of a large number of heterogeneous catalysts such as silica [5,6], alumina [7–9], mesoporous molecular sieves [10–13], mixed oxides [14], zeolites [15–20] and nanoparticles [21,22]. The main goal of the present work is the synthesis of paracetamol using new catalysts obtained by impregnation of silica with phosphoric acid. We have recently reported the preparation of the catalysts, their characterization by usual solid state techniques (XR-diffraction, thermal analysis, FTIR spectroscopy, nitrogen adsorption and SEM microscopy), and their use in the vapor-phase alkylation of toluene by benzyl alcohol [23]. The characterization of these solids is now extended by a detailed MAS NMR spectroscopy study (29Si, 27Al, 31P, 23Na and 1H), and their application to the synthesis of paracetamol is reported.

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492

1487

2. Experimental

2.4. Reaction procedure

2.1. Materials

The Beckmann rearrangement of 4-hydroxyacetophenone oxime (Scheme 1) was carried out in the liquid phase in a 50-mL two-necked, round-bottomed flask immersed in a thermo-stated bath and equipped with a reflux condenser and a magnetic stirrer. For the cases when acetone or methanol was used, the reaction temperature was maintained at the reflux temperature. A typical reaction run was as follows: 0.1 g catalyst, pre-activated in air at 400 °C, was suspended in a solution of 0.15 g (1 mmol) of 4hydroxyacetophenone oxime in 20 mL of solvent, which was allowed to equilibrate to the set temperature. The reaction mixture was heated to reaction temperature and stirred for 1 h. When the reaction was completed, the reaction mixture was filtered and diluted with methanol and the reaction products were analyzed using a Shimadzu, Model C-R4AX HPLC instrument, equipped with a Symmetry C18 5 lm  6 mm  150 mm column with a UV detector operating at 254 nm. As paracetamol is not the only product of the reaction, results are given as conversion (amount of 4-hydroxyacetophenone that reacts) and selectivity (content of paracetamol in the products).

HPLC grade methanol, phosphoric acid, 4-hydroxyacetophenone, Na2HPO4 and solvents were obtained from Merck with the maximum purity degree, and used as received. In the case of 4-hydroxyacetophenone, the purity of the commercial reagent was >98%, but after preparing the 4-hydroxyacetophenone oxime, it was purified by crystallization and its purity was over 99.8%. The synthesis of 4-hydroxyacetophenone oxime was carried out by the general method reported in the literature [24]. Its identity was confirmed by FT-IR and proton NMR spectroscopies. 2.2. Preparation of the catalysts Al-MCM-41 molecular sieves were synthesized by a previously described method [23]. After some preliminary catalytic tests, the sample with a SiO2/Al2O3 molar ratio of 70 was chosen for preparing a series of H3PO4/Al-MCM-41 catalysts, varying the H3PO4 content. These catalysts were prepared by impregnation of the calcined Al-MCM-41 solid with different amounts of H3PO4 dissolved in distilled water. The impregnated catalysts were dried at 110 °C for 12 h and then calcined in air at 550 °C for 4 h before using them for catalytic reactions.

3. Results and discussion 3.1. Characterization of the catalysts

2.3. Characterization techniques The acidity of the samples was evaluated from the Fourier transform IR spectra of adsorbed pyridine, measured using a JASCO FT/IR (680 plus) spectrometer following a previously described method [25]. Textural analyzes were determined by nitrogen adsorption– desorption isotherms at 196 °C using a Micromeritics Gemini apparatus, after outgassing the solids at 110 °C for 2 h. Specific surface area was obtained by application of the BET method. X-ray diffraction (XRD) patterns were recorded over non-oriented powder samples between 1 and 70° (2h) at a scanning speed of 2°/min by using a Siemens D-500 diffractometer, at 40 kV and 30 mA, and employing filtered Cu Ka radiation. The equipment was connected to a DACO-MP microprocessor and used DiffractAT software. High resolution MAS-NMR experiments were performed at room temperature in a Bruker AVANCE-400 spectrometer, operating at 79.49 MHz (29Si signal), 104.26 MHz (27Al signal), 161.98 MHz (31P signal), 105.84 MHz (23Na signal) and 400.13 MHz (1H signal). 29Si, 31 P, 1H (I = 1/2) MAS-NMR spectra were recorded after p/2 pulse irradiation (4 ls), using a 500 kHz filter. To record satellite transitions in 23Na (I = 3/2) and 27Al (I = 5/2) MAS-NMR signals, spectra were recorded after p/8 pulse irradiation (1.5 ls), using a 1 MHz filter. In MAS experiments, powder samples were spun at 5 and 12 kHz. The number of scans was 400 for silicon and 50 for other signals. In all cases, time between accumulations was chosen between 1–30 s to minimize saturation effects in different signals. The experimental error in peak position values was estimated to be ±0.2 ppm. The quantitative analysis of MAS-NMR spectra was carried out with the Winfit software package (Bruker). This program allows the position, linewidth, and intensity of components to be determined with a nonlinear iterative least-squares method. However, quadrupolar CQ and g constants must be determined from 23Na and 27Al MAS-NMR spectra with a trial and error procedure. In this case, experimental profiles were simulated, considering second order quadrupolar interactions, by using the DM2006 software package. Chemical shift values of components were deduced after correction of second order quadrupole effects, following the procedure described by Massiot et al. [26].

3.1.1. X-ray diffraction measurements XRD data indicate that MCM-41 samples exhibit the ordered hexagonal structure characterized by an intense reflection peak at d spacing of 3.8 nm [23]. The intensity of this low angle reflection peak did not change as the H3PO4 loading increased. Although H3PO4 loading induces a drastic decrease in the specific surface area, there are not any structural distortions of the MCM-41. Consequently, it seems reasonable to propose that loading of H3PO4 on MCM-41 does not change the regular arrangement of the uniform channels of the support [27]. 3.1.2. N2 adsorption measurements The textural properties show a relatively important variation between the different solids, as observed in Table 1. The specific surface area of the supports varies between 668–1038 m2/g, increasing when the SiO2/Al2O3 ratio increases, that is, when the amount of Al incorporated to the silica decreases. In other words, if pure silica is taken as a reference, incorporation of Al clearly decreases the specific surface area of the resulting solids. All the supports have very similar external surface areas, between 83–90 m2/g; thus, the impregnation with H3PO4 seems to block the channels of the solids, although such a blocking may not be expected considering the inherently large pores of the solid and the size of the phosphate entity. Incorporation of H3PO4 produces a drastic decrease in the surface of the solids, to values between 46–88 m2/g, clearly suggesting that H3PO4 blocks the access to the MCM channels. 3.1.3. Acidity measurements Acidity was measured by adsorption–desorption of pyridine by IR spectroscopy. Results for acidity measurements are summarized in Table 1. All catalysts show both Brønsted and Lewis acid sites. Brønsted acidity increases upon impregnation with H3PO4 while Lewis acidity shows the opposite trend. Both types of acidity clearly decrease when heating the solids from 200 to 300 °C. 3.1.4. Multinuclear NMR study The multinuclear NMR study has been carried out on the 30% H3PO4/Al-MCM-41 sample that displays the highest reactivity. This study has been divided in two parts. In the first part, spectra of

1488

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492

NOH HO

HO

Catalyst

NHCOOCH3

N-acetyl-p-aminophenol

4-hydroxyacetophenone oxime Scheme 1.

Table 1 BET surface area and acidity of the catalysts. Sample

BET surface area (m2/g)

Acidity* (l mol g1) Brønsted

Al-MCM-41(2 0) Al-MCM-41(7 0) Al-MCM-41(1 1 0) Al-MCM-41(1 5 0) Al-MCM-41(2 0 0) 30% H3PO4/Al-MCM-41(20) 30% H3PO4/Al-MCM-41(7 0) 30% H3PO4/Al-MCM-41(1 1 0) 30% H3PO4/Al-MCM-41(1 5 0) 30% H3PO4/Al-MCM-41(2 0 0) *

668 810 870 950 1038 88 66 54 58 46

Lewis

Total

200 °C

300 °C

200 °C

300 °C

200 °C

300 °C

41 29 17 11 19 108 135 78 99 121

18 13 8 8 8 44 51 26 58 67

90 75 31 28 28 34 48 16 22 25

71 62 22 25 25 23 22 6 14 17

131 104 48 39 47 142 183 94 121 146

89 75 30 33 33 67 73 32 72 84

Acidity (l mol pyridine adsorbed per gram of catalyst) calculated using the extinction coefficients from Refs. [36,37].

structure-forming cations (Si, Al, Na) will be described, then spectra of surface deposited species (P, H) will be analyzed. The 29Si MAS-NMR spectra of samples show the presence of two main components at 112 and 102 ppm (Fig. 1) that correspond to Si atoms surrounded by 4Si and 3Si1OH environments. In catalysts with lower SiO2/Al2O3 ratios, two additional components are detected at 90 and 80 ppm. These two new components correspond to Si atoms in 2Si2OH and 1Si3OH environments. In general, the higher the Si content of samples, the higher is the intensity of NMR components associated to more condensed – Q4(4Si) and Q4(3Si1OH) – species. The 27Al MAS-NMR spectra of analyzed catalysts display one broad and symmetric component at 10 ppm, which was ascribed to octahedral aluminium (Fig. 2). The 31P MAS-NMR spectra of catalysts are formed by two components at 1 and 10 ppm that correspond to P atoms in Q0 and Q1 environments (Fig. 3). The 1H MAS-NMR spectra are formed by two components at 0 and 7.5 ppm that have been ascribed to OH groups of Al–OH and P–OH associations (Fig. 4). 23Na MAS-NMR spectra show a single broad component at 3 ppm (Fig. 5). As indicated, 29Si MAS-NMR of 30% H3PO4/Al-MCM-41 catalysts are mainly formed by two signals at 112 and 102 ppm that correspond to Si atoms surrounded by 4Si and 3Si1OH environments, respectively. Both signals can be ascribed to Si atoms located inside and at the surface of MCM compounds. Relative intensities of the two components change with SiO2/Al2O3 ratios, indicating that the relative amount of Si atoms located at the surface increase when the amount of Al increases. In the analyzed samples, the polymerization degree decreases progressively from 3.8 to 3.4 when SiO2/ Al2O3 ratios decrease from 200 to 20. This observation could be explained if the thickness of MCM walls (Si atoms in Q4(4Si)) increases with the relative amount of Si (SiO2/Al2O3 ratios). In MCM compounds, Si atoms should only occupy Q4 and Q3 environments. However, the presence of Q2 and Q1 environments indicates the presence of structural defects, probably located at the surface of MCM compounds that further reduces the surface tetrahedral polymerization when the SiO2/Al2O3 ratio decreases below 70. The octahedral arrangement of Al suggests that Al is preferentially located at the MCM surface, what would explain the absence of tetrahedral Al–O–Si associations and the lack of Q4(mAl) environments in 29Si MAS-NMR spectra.

29

-112

Si

M200

-102

M110

-112 -102

M20

-80 -90

-50

-100

ppm

-150

Fig. 1. 29Si MAS-NMR spectra of the samples 30% H3PO4/Al-MCM-41(20), 30% H3PO4/Al-MCM-41(1 1 0) and 30% H3PO4/Al-MCM-41(2 0 0), abbreviated as M20, M110 and M200, respectively.

On the other hand, the degree of condensation of PO3 4 species is very low, what favours the stabilization of Q1 and Q0 species on the surface of MCM catalysts. This fact favours the interaction of P tetrahedra with other polyhedra. Taking into account that formation of tetrahedral P–O–Si associations is not chemically

1489

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492

27

-10

Al

1

H

7.5

M200

M200 -0.5

M110

M110

M20

100

50

0

ppm

-50

M20

-100

Fig. 2. 27Al MAS-NMR spectra of the samples indicated (for notation of samples, see Fig. 1).

31

0

P

30

20

10

ppm

0

-10

-20

Fig. 4. 1H MAS-NMR spectra of the samples indicated (for notation of samples, see Fig. 1).

23

M200

-3

Na

M200

-10

M110 M110

M20 M20

80

40

0

ppm

-40

-80

Fig. 3. 31P MAS-NMR spectra of the samples indicated (for notation of samples, see Fig. 1).

100

50

0

ppm

-50

-100

Fig. 5. 23Na MAS-NMR spectra of the samples indicated (for notation of samples, see Fig. 1).

1490

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492

favourable; the interaction of tetrahedral P with Al must be enhanced. In the 31P MAS-NMR spectra of modified MCM catalysts, the peak at 10 ppm has been ascribed to pyrophosphate groups, and that at 0 ppm to monomeric PO3 4 -species. In general, the amount of pyrophosphate groups increases with the aluminium content of the sample. However, the dispersion of P and Al on the MCM surface precludes the formation of AlPO4 and Al2O3 phases. From the analysis of Si, Al and P NMR signals, the formation of Si–O–Al–O–P bridges seems to be favoured at the surface of MCM-41 catalysts. Formation of Si–O–Al–O–P bridges should decrease the amount of surface Si–OH groups in MCM compounds. On the other hand, the interaction of acid PO3 4 species with Al-MCM-41 seems to favour the octahedral arrangement of Al. A similar arrangement of Si, Al and P is produced in SAPO compounds [28]; however, structural requirements impose the tetrahedral coordination of Al. This arrangement underlines the importance of surface Al in immobilization of impregnated PO3 4 species. The presence of octahedral Al explains the small effect of Al in 29Si and 31P MAS-NMR spectra, but the bigger influence of these strong polarizing cations on the octahedral Al signal. The up-field shift observed in the octahedral Al signal should result from this interaction. On the other hand, impregnation of MCM-41 surface with acid PO3 4 species justifies the presence of acidic OH groups in Q0 and Q1 phosphorous environments. Taking into account that polarization of OH groups (acidity) favours the shift of 1H NMR components towards more positive positions, the signals detected at 7.5 and 0.5 ppm have been ascribed to protons in P–OH and Al–OH associations. This result suggests that most of OH groups are preferentially retained by P but not by tetrahedral Si. The position of 1H MAS-NMR components must also be affected by the interaction of OH groups with adsorbed water. The interaction of water with acidic P–OH groups (band near 10 ppm) could favour the formation of hydronium cations, H3 Oþ , which would shift the OH band towards more negative values. Taking into account that the chemical shift of water is 4.8 ppm; the formation of H3 Oþ groups would produce the detection of one 1H MAS-NMR component at 7 ppm, that has been experimentally observed.

HO

The small polarizing strength of Na cations explains that the positions of 29Si and 31P MAS-NMR components were not considerably affected when protons are substituted by Na cations. This fact makes difficult the study of sodium location by NMR spectroscopy. On the other hand, taking into account the sensitivity of the 23Na signal to hydration processes, the chemical shift values could be used to estimate the coordination and mobility of sodium cations. Chemical shifts near 3 ppm could be assigned to hydrated Na cations; however, important broadenings detected in Na signals suggest that Na compensating cations are not directly coordinated to water molecules. According to the widely accepted reaction mechanism for the Beckmann rearrangement, the first step in the formation of paracetamol is the protonation of the oxime oxygen (O-protonated oxime), followed by the migration of the 4-hydroxyphenyl group and the removal of a water molecule, giving a nitrilium cation. The nitrilium cation reacts with one molecule of water to form the amide tautomer and finally the corresponding amide [29]. However, more recently it was proposed that the most energetically favourable path in the Beckmann rearrangement involves the protonation of the oxime, giving a N-protonated oxime. This is followed by a 1,2-H shift, giving the O-protonated oxime, which evolves toward the nitrilium cation and then gives the amide [30]. According to the data presented in Table 1, by loading the MCM-41 with 30% H3PO4, the support loses most of its internal surface area, indicating that reaction occurs on the external surface of catalysts (Scheme 2). 3.2. Catalytic studies 3.2.1. Solvent effect The nature of the solvent strongly influences the performance of the reaction (Table 2). In polar protic media, the conversion is low, but in the presence of polar aprotic solvent, the reaction progressed to a considerable conversion, up to 100% in acetone (which was selected as a solvent for further studies). It may be proposed that when using a polar protic solvent, 4-hydroxyacetophenone has difficulties to come in contact with the catalyst, therefore the

HO

HO

C H3C

H

N

C OH

O

N

H 3C

H

O

O

P

-H 2 O

O

CH3 O

Al

OH O O

O

H

H3 C

Si

O H

Si

C Si

H 3C

N O

H3 C

Al

CH3

Si

O

O P

O

Al O

HO HO O O

Si

Al

Si

OH O

H 2O

C H3C

C

N

OH C

N

H3C

Scheme 2.

OH

H 3C

NH

OH

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492 Table 2 Effect of solvent on conversion and selectivity over 30% H3PO4/Al-MCM-41(7 0) catalyst. Solvent

Dielectric constant*

Conversion

Selectivity

Acetone Acetonitrile Methanol Water

21.01 36.64 33.0 80.1

100 81.6 42.3 15.2

100 100 100 100

Reaction conditions: 1 mmol oxime, 0.1 g of catalyst, 20 mL solvent, 343 K, 1 h reaction time. * Values at 293.2 K.

Table 3 Effect of Al-MCM-41 with different SiO2/Al2O3 ratio on conversion and selectivity. Catalyst

Conversion

Selectivity

Al-MCM-41(2 0) Al-MCM-41(7 0) Al-MCM-41(1 1 0) Al-MCM-41(1 5 0) Al-MCM-41(2 0 0)

84.3 80.2 83.3 65.8 60.7

75.3 89.0 83.2 64.1 65.2

Reaction conditions: 1 mmol oxime, 0.1 g of catalyst, 20 mL acetone, 329 K, 1 h reaction time.

Table 4 Effect of H3PO4 loading on Al-MCM-41(7 0) on conversion and selectivity. Amount of H3PO4

Conversion

Selectivity

0% H3PO4 10% H3PO4 20% H3PO4 30% H3PO4 35% H3PO4

80.2 80.4 91.5 100 100

89.0 86.7 95.4 100 91.6

Reaction conditions: 1 mmol oxime, 0.1 g of catalyst, 20 mL acetone, 329 K, 1 h reaction time.

conversion is low, in agreement with the results reported by Chung and Rhee on the effect of the solvent on this reaction [31]. Aprotic media allow obtaining much better results. However, it is remarkable that we got at least 15% and about 40% conversion when using water and methanol as solvent, respectively. We think that highlighting the effect of solvent without taking into account the role of the support is not consistent. 3.2.2. Effect of the SiO2/Al2O3 ratio of the Al-MCM-41 catalysts As indicated in Table 3, the samples with high hydrophilic character (low SiO2/Al2O3 ratio) are the most active catalysts due to their improved adsorption properties for polar molecules [32]. The amount of Brønsted acid sites with moderate strength (Table 1) may explain the higher activity of Al-MCM-41 with SiO2/Al2O3 from 20 to 110 when compared to Al-MCM-41 with SiO2/Al2O3 = 150–200. Concerning the selectivity it should be noted that due to the characteristics of this type of materials (regular array of pores with a large diameter), steric constraints imposed by the structure are not expected. The differences observed should be mainly due to the different distribution and strength of the acid sites. 3.2.3. Effect of modification with H3PO4 Improvement of the H3PO4/Al-MCM-41 catalyst is made by optimizing the H3PO4 loading. Variation of H3PO4 loading between 0 and 35 wt% has different effects on the activity and selectivity for Beckmann rearrangement; the evolution of the activity of the catalysts with different H3PO4 loadings is compared in Table 4. The first important result is that H3PO4-containing catalysts are more active than the bare support, without H3PO4. The results obtained

1491

also show that the maximum 4-hydroxyacetophenone oxime conversion and best selectivity is recorded for the catalyst with 30 wt% H3PO4. According to the literature, H3PO4 supported on silica (solid phosphoric acid-SPA) were industrially developed in the 1940s. It is assumed [33–35] that by treating the aluminosilicate molecular sieves with phosphoric acid, the bridged OH groups between Si and Al atoms are replaced by H2PO4 groups. The substitution of the Brønsted acidic hydroxyl groups by the H2PO4 groups implies different consequences on the strength and density of the Brønsted acid sites. Strong Brønsted acid sites are weakened, and the resulting terminal hydroxyl groups have a lower acid strength than the bridged hydroxyl groups. 3.2.4. Catalyst recovery In order to study the catalyst deactivation occurring during the synthesis of paracetamol, we carried out the Beckmann rearrangement several times with modified Al-MCM-41 in acetone which was selected as the best solvent for the reaction. After the reaction was complete, the catalyst was filtered, washed with acetone, and used again in a new experiment. The yields of five consecutive reactions leading to paracetamol were 100, 100, 96, 89 and 90%, respectively, thus no considerable decrease in the yield was observed. No leaching of H3PO4 was observed during the reaction, its percentage in the catalyst remains constant. The slight decrease in the activity may be explained by the fact that acetone can polymerize through aldol condensation in acidic conditions, contaminating the surface of the catalyst. Although it may be mostly removed during calcination, small amounts of coke may be retained on the catalyst, being responsible for the decrease in conversion. Thus, Al-MCM-41 modified with H3PO4 can be reused as a catalyst in Beckmann rearrangement of 4-hydroxyacetophenone oxime. 4. Conclusions Paracetamol was synthesized by an environmentally benign synthetic method over Al-MCM-41 modified with H3PO4. Solid acid catalysts present clear advantages over the conventional homogeneous acid catalysts used commercially for the preparation of paracetamol. Al-MCM-41 modified with H3PO4 is very active and selective for the preparation of paracetamol by Beckmann rearrangement, achieving good conversion of oximes in relatively short reaction times. It was found that for 4-hydroxyacetophenone oxime, the nature of the solvent has a strong effect on the activity of the catalysts. Acknowledgment Thanks are due to the Research Council of Isfahan University of Technology and Center of Excellency in the Chemistry Department of Isfahan University of Technology for supporting of this work. References [1] W.T. Beaver, D. McMillian, Br. J. Clin. Pharmacol. 10 (Suppl. 2) (1980) 215S. [2] D.R. Mehlisch, J. Am. Dent. Assoc. 133 (2002) 861. [3] K.G. Davenport, C.B. Hilton, US Patent 4,524,217 (1985) to Celanese Corporation. [4] J.R. Fritch, D.A. Aguilla, T. Horlenko, O.S. Fruchey, EP 469,742 (1992) to Celanese Corporation. [5] T. Ushikubo, K. Wada, J. Catal. 148 (1994) 138. [6] S. Sato, K. Urabe, Y. Izumi, J. Catal. 102 (1986) 99. [7] T. Curtin, J.B. McMonagle, B.K. Hodnett, Appl. Catal. A 93 (1992) 91. [8] T. Curtin, J.B. McMonagle, B.K. Hodnett, Catal. Lett. 17 (1993) 145. [9] S. Sato, S. Hasebe, H. Sakurai, K. Urabe, Y. Izumi, Appl. Catal. 29 (1987) 107. [10] C. Ngamcharussrivichai, P. Wu, T. Tatsumi, J. Catal. 227 (2004) 448. [11] L.X. Dai, K. Koyama, T. Tatsumi, Catal. Lett. 53 (1998) 211. [12] C.C. Tsai, C.Y. Zhong, I. Wang, S.B. Liu, W.H. Chen, T.C. Tsai, Appl. Catal. A 267 (2004) 87.

1492

M. Ghiaci et al. / Catalysis Communications 10 (2009) 1486–1492

[13] C. Ngamcharussrivichai, P. Wu, T. Tatsumi, Chem. Lett. 33 (2004) 1288. [14] M. Ghiaci, A. Abbaspur, R.J. Kalbasi, Appl. Catal. A 287 (2005) 83. [15] B. Bonelli, L. Forni, A. Aloise, J.B. Nagy, G. Fornasari, E. Garrone, A. Gedeon, G. Giordano, F. Trifirò, Micropor. Mesopor. Mater. 101 (2007) 153. [16] C. Ngamcharussrivichai, P. Wu, T. Tatsumi, Appl. Catal. A 288 (2005) 158. [17] Y. Zhang, Y. Wang, Y. Bu, Z. Mi, W. Wu, E. Min, S. Han, S. Fu, Catal. Comm. 6 (2005) 53. [18] G.P. Heitmann, G. Dahlhoff, J.P.M. Niederer, W.F. Hölderich, J. Catal. 194 (2000) 122. [19] L.X. Dai, K. Koyama, M. Miyamoto, T. Tatsumi, Appl. Catal. A 189 (1999) 237. [20] G.P. Heitmann, G. Dahlhoff, W.F. Hölderich, Appl. Catal. A 185 (1999) 99. [21] M. Ghiaci, H. Aghaei, A. Abbaspur, Mater. Res. Bull. 43 (2008) 1255. [22] P. Botella, A. Corma, S. Iborra, R. Montón, I. Rodríguez, V. Costa, J. Catal. 250 (2007) 161. [23] M. Ghiaci, A. Abbaspur, R. Kia, C. Belver, R. Trujillano, V. Rives, M.A. Vicente, Catal. Commun. 8 (2007) 49. [24] L. Vogel, A Text Book of Practical Organic Chemistry, ELBS/Longmann, London, 1971.

[25] M. Ghiaci, R.J. Kalbasi, M. Mollahasani, H. Aghaei, Appl. Catal. A 320 (2007) 35. [26] D. Massiot, C. Bessada, J.P. Coutures, F. Taulelle, J. Magn. Reson. 90 (1990) 231. [27] R.A.A. Melo, M.V. Giotto, J. Rocha, E.A. Urquieta-Gonzalez, Mater. Res. 2 (1999) 173. [28] J. Sanz, J.M. Campelo, J.M. Marinas, J. Catal. 130 (1991) 642. [29] M.B. Smith, J. March, March’s Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, sixth ed., John Wiley & Sons, Inc., Hoboken, New Jersey, 2007. [30] J.C. Jochims, S. Hehl, S. Herzberger, Synthesis (1990) 1128. [31] Y.-M. Chung, H.-K. Rhee, J. Mol. Catal. A 159 (2000) 389. [32] V.N. Ipatieff, R.E. Schaad, US Patent 2,382,318 (1945). [33] W.W. Kaeding, C. Chu, L.B. Young, B. Weinstein, S.A. Butter, J. Catal. 67 (1981) 159. [34] J.C. Védrine, A. Auroux, P. Dejaifve, V. Ducarme, H. Hoser, S. Zhou, J. Catal. 73 (1982) 147. [35] J.A. Lercher, G. Rumplmayr, H. Noller, Acta Phys. Chem. 31 (1985) 71. [36] T.R. Hughes, H.M. White, J. Phys. Chem. 71 (1967) 2192. [37] J. Datka, A.M. Turek, J.M. Jehng, I.E. Wachs, J. Catal. 135 (1992) 186.