It is the cache of ${baseHref}. It is a snapshot of the page. The current page could have changed in the meantime.
Tip: To quickly find your search term on this page, press Ctrl+F or ⌘-F (Mac) and use the find bar.

Journal of the Chilean Chemical Society - CATALYTIC HYDROGENATION OF THYMOL OVER Pd/MgO PREPARED BY SMAD METHOD

SciELO - Scientific Electronic Library Online

 
vol.51 número4CATALYTIC OZONATION OF OXALIC ACID WITH MnO2/TiO2 AND Rh/TiO2DITERPENOIDS FROM Calceolaria filicaulis SSP LUXURIANS índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Journal of the Chilean Chemical Society

versión On-line ISSN 0717-9707

J. Chil. Chem. Soc. v.51 n.4 Concepción dic. 2006

http://dx.doi.org/10.4067/S0717-97072006000400013 

J. Chil. Chi. Soc., 51, N°.4 (2006), p.1053-1056

 

CATALYTIC HYDROGENATION OF THYMOL OVER Pd/MgO PREPARED BY SMAD METHOD

 

GALO CÁRDENAS T*.1, RICARDO OLIVA C. AND PATRICIO REYES N.2

1Departamento de Polímeros, Laboratorio Materiales Avanzados
2 Departamento de Físico-Química Facultad de Ciencias Químicas, Universidad de Concepción, Casilla 160-C, Concepción (Chile)


SUMMARY

In this work are summarized the results obtained in the thymol hydrogenation using Pd/MgO catalyst. The catalyst was prepared from a colloidal dispersion of Pd in 2-propanol using the SMAD ( Solvent Metal Atom Dispersion) methodology.

The catalyst was characterized by means of Transmission Electron Microscopy (TEM) and by chemisorption of hydrogen (metallic area and dispersion). The particles show a bimodal distribution in the histogram of particle distribution with a range of particle size between 11 and 16.6 nm. The catalyst metallic area was 26m2/g with a 5.8 % dispersion.

The products obtained from thymol hydrogenation were menthone and isomenthone. The catalyst was not able to hydrogenate the carbonyl group and did not produce the menthol and /or isomers. The different temperatures (80, 100, 125 and 150ºC) only produce an effect in the catalyst activity, but not in the selectivity. The same results were obtained with increase in pressure (3, 5, 7 and 10 bars) and the charge catalyst (3, 7, 10 and 20%) was used. The studies were carried out with different thymol amounts over the same catalyst, and demonstrated no change in the activity. This reveals that the catalyst is poisoned most probably by unreacted thymol. The addition of a promotor (Fe+3) to activate the carbonyl group, allows the obtention of a mixture of menthol isomers.

The Pd/MgO catalyst is more active than Pd/C and Pd/Al2O3. This demonstrates that the support plays an important role, however, the Pd is not able to hydrogenate the carbonyl group in menthone or isomenthone.


INTRODUCTION

The thymol hydrogenation in liquid phase to produce menthol is frequently used.

This product is widely used in pharmaceutical industry, cosmetics, perfume and elsewhere (1-4).

The thymol hydrogenation produces a final complex mixture of 4 diasteroisomers: (±) menthone, (±) isomenthone, (±) neomenthone and (±) neoisomenthone (1) which widely depend on the composition on the catalyst and on the reaction parameters.

An important aspect of this reaction is that always thymol reduction is carried out in a first stage to the menthone and / or isomenthone, depending on the reaction conditions, especially pH. Only when the menthone or isomenthone amount reaches an important concentration the menthol isomers will be produced. The selectivity of the catalyst beyond the menthol formation depends exclusively of the formation of menthone (±) isomenthone. The first carries out the (±) menthone and (±) neomenthol formation and the isomenthone to (±) neoisomenthol and (±) neomenthol (5-7).

In general, it is more easy to produce isomenthone than menthone, but this product can exchange practically in 100% to menthone in basic media (5) . For this reason, the hydrogenation is carried out at pH > 10 (1). In neutral or acidic media the isomenthone formation is predominant and as a consequence a mixture of isomenthol and neoisomenthol is obtained (1,8). In most of the literature reports carbon is used as support and water as solvent due to the basicity of the solutions used.

The kinetic of thymol hydrogenation over Ni-Cr catalyst with formation of four menthol, diasteroisomers have been investigated in a batch reactor at constant hydrogen pressure (0.4 -4.0 Mpa) at temperatures between 373 and 433K in n-hexane and cyclohexane solutions. Initially It was observed that more thermodinamically unstable isomers (neomenthol and neoisomenthol) were produced. At relatively high conversions trans isomers (menthol and isomenthol) were formed. However, the stereo selectivity values at a particular conversion were seen to be independent of hydrogen pressure and reaction temperature (9). In other study, the stereo selective hydrogenation of thymol was studied in liquid phase over several Ni catalysts, modified by impregnation of inorganic compounds containing chlorine. The total activity of all catalysts studied was decreased on the same order of magnitude, indicating that chlorine even after reduction remains on the surface and is responsible for the poisoning effects. Selectivity and stereo selectivity to menthols and menthons changed significantly. The modifiers could influence the rate of keto-enol transformation, which is probably the control selectivity and stereoselectivity in the main step (10).

In other studies, the catalytic properties of Pt, Rh and Ir catalyst in the reduction of thymol have been compared, with respect to their activities and stereo selectivities in the formation of menthones and menthols. The reduction over Pt and Rh proceeds essentially via the ketone intermediates, whereas the direct way is predominant on Ir. The formation of the cis isomer is always highly favored (11). Experimental kinetics and product distribution in thymol hydrogenation over Rh/C and (Pd + Rh)/C were consistent with a kinetic scheme based on the mechanism of hydrogenation of aromatic compounds. For these catalysts, ratios of isomeric menthones and menthols varied little with temperatures; the ratio of isomeric menthols was nearly the same on the two catalysts; the menthone: isomenthone ratio on (Pd +Rh) was greater than Rh. With a Ir/C catalyst, hydrogenolysis processes were observed (12).

In this work, the presence of a basic support like MgO is studied in the activity and selectivity in the thymol hydrogenation over a catalyst from 3 to 20% Pd/MgO prepared by SMAD (13 ).

The hydrogenation was performed in aprotic solvents such as n-hexane in order to minimize the solvent effect. The reaction was carried out under mild conditions compared with what has been reported in literature. The temperature effect (80-150º C), pressure (3-10 bar), the amount of catalyst and the catalyst recycling are reported.

EXPERIMENTAL

Catalyst Preparation

The catalyst formation was carried out in a bimetal reactor atom (14,15,16). Two Tungsten crucibles (W-Al2O3 from Sylvania Emissive) were charged with around 0.1500 g of Pd metal in lumps. Distilled and dried solvents ( 2-propanol, 100ml) were placed in a ligand inlet tube and freeze-pump-thaw degassed for five cycles. The reactor is kept under vacuum until reaching 5-10 microns of Hg, previously 3.0 g of MgO (BET area 90m2/g) with a magnetic stir bar have been introduced.

A liquid nitrogen filled dewar was placed around the vessel and Pd and 100ml 2-propanol were codeposited over a 1.5h period. The matrix was of a black color at the end of the code position. The matrix formed was allowed to warm slowly for 1.0 h at room temperature under vacuum by removal of the liquid nitrogen dewar. Upon meltdown, the black dispersion was allowed to warm for another 0.5 h at room temperature under N2(g) flow. Finally, the metal dispersed in solvent is stirred with MgO for 24h in the reactor at room temperature under N2(g). atmosphere.

Transmission Electron Microscopy (TEM)

Electron micrographs were obtained on a JEOL JEM 1200 EXII. The supported catalyst is grounded in an agate mortar and dispersed in 2-propanol or acetone. A drop of each dispersion was placed on a 150 mesh copper grid coated with carbon. Several magnifications were used. Four to five electron micrographs in different places of the copper grid were taken. Then 80 to 100 particles in each micrograph were measured. Finally, the 6.0 Origin program (Microcal Software Inc.) was used to plot the frequency histogram to determine the mean particle size.

Selection Area Electron Diffraction (SAD)

The electron diffraction of the catalyst was obtained in a JEOL microscope internally calibrated with gold standard (Merck, 99.99% of pure) (120 KV, K= 3.848 cm. Å). The diffraction patterns were obtained using an aperture of the terminal field of 20 m. In this way the diffractions coming from the grid are avoided and the observed area is minimized.

Chemisorptions.

Hydrogen chemisorptions at 343 K was carried out by a pulse method in a TPD/TPR 2900 Micromeritics system provided with a thermal conductivity detector. Before the experiments, the samples were reduced in situ under hydrogen flow (50 cc/min) at 673 K for 1 h. Then, the gas was shifted to Ar and kept at this temperature for 2h and cooled down to 343 K. Once the baseline was restored, different pulses of H2 were sent to the sample holder, up to complete saturation of the metallic surface. By evaluating the amount of H2 uptake at 343K, the H/Pd ratio was obtained.

Catalyst impregnation with Fe+3

50 ml were prepared from a solution of Fe(NO3)3.9H2O in 2-propanol. Sufficient amount of catalyst ( 5%Pd/MgO) were prepared in atomic ratio of Pd/Fe+3 =1. The mixture was stirred overnight. The solvent was evaporated under vacuum in order to use the catalyst, this is heated under N2 flow at 130ºC to decompose the nitrate.

Catalytic hydrogenation of thymol

The thymol hydrogenation in liquid phase was carried out in a stainless steel reactor of 200ml, with a manometer to control the pressure and connected to a hydrogen tank (99,99% ). The catalyst powdered was added in the reactor, and a H2(g) flow is constant to reduce the surface oxide and then heated at 130ºC for 20 min, to clean the surface from the remaining solvent .

The reactor is cooled at room temperature and the thymol solution in hexane (70 ml) was added.

The mixture was homogeneized with a magnetic stirrer. A N2(g) atmosphere was used.

Hydrogen gas was flow to purge the system and the partial adequate pressure was set up. The reactor was immersed in an oven with digital temperature controlled at heating rate of 5ºC/min. The reaction was stirred until all the thymol is reacted, samples were taken periodically.

The samples were analyzed, in a GC gas HNU Systems GC-321 with thermal conductivity detector and a packed column ALLTECH 254, 80/100 for alcohols with a flow of He of 30 cc/min was used. Only two peaks were detected and the mixture of menthone and isomenthone.

These isomers are different from menthol presenting the same retention time that menthone or isomenthone, for that reason they were detected by FTIR.

RESULTS AND DISCUSSION

The following scheme shows the hydrogenation of thymol.

The more thermodinamically unstable isomers are neomenthol and neoisomenthol. On the other hand, the ketones should be more favored in the presence of a basic support that improves the catalytic hydrogenation of thymol.

Catalyst Characterization

The catalyst prepared shows a metallic area of 26 m2/g and a 5.8% dispersion. The size obtained by TEM demonstrate a bimodal size distribution (fig. 1 and 2), with an average size of 11nm and 16.6nm, however, it is probable that a second distribution belongs to the growing of primary particles. The particles distribution over the support is enough homogeneous to indicate that the sizes exhibited are representative of the catalysts. Another important characteristic of the particles is revealed by the electron diffraction pattern (SAD). This reveals the presence of metallic particles well formed and random distributed over the support surface (the diffraction pattern shows only rings).



Temperature effect .

In the described conditions the product was a racemic mixture of menthone and isomenthone. The effect of temperature between 80-150 ºC in the initial rate of thymol consumption is observed in figure 3 .


The increase in temperature from 80 to 150ºC accelerates 3.6 times the reaction rate, however, no menthol isomers were detected only a racemic mixture of menthone and isomenthone without optical activity. After the thymol consumption, the mixture menthone and isomenthone does not undergo to the menthol formation or some isomer. The reason of this is attributed to the Pd presence where it is not very efficient to hydrogenate the carbonyl group (17 ). On the other hand, metals such as Pt and Ni in neutral aprotic solvents like cyclohexane and n-hexane are more effective. Then the corresponding isomers of menthol, specially neomenthol were obtained (3,8). The support does not play an important role in the reaction and the results obtained were similar to the thymol hydrogenation over Pd/Selcat Q (the supports is a modified carbon) in gas phase (2), where menthone and isomenthone were obtained.

Pressure effect.

In figure 4, the results obtained for pressures at 3, 5, 7 and 10 bars are presented. In this case, the hydrogen concentration, produces a more dramatic change in the speed of thymol conversion.


The rate of reaction at 7 bars is 30 times bigger than at 3 bars pressure. The results obtained are similar to those described previously, the increase in the pressure and hence the increase in the hydrogen solubility does not favor the menthol or isomers production. The reaction is stopped in the production of the mixture menthone and isomenthone. Even though hydrogenation in liquid phase at high pressures (over 40 bars) can end up with menthol formation. The pH effect and the presence of a protic solvent, especially water is fundamental in the thymol hydrogenation over Pd. Several tests were carried out under the same conditions with catalysts Pd/C and Pd/Al2O3 SMAD type, exhibiting a low activity in the hydrogenation, which is indicative that the presence of a basic support like MgO plays an important role in the Pd activity. However, it is not able to hydrogenate menthone or isomenthone. The experiment of menthone hydrogenation at 150ºC, 7 bar and 10% de Pd/MgO in n-hexane, during 24 h does not exhibit activity. This is a result that confirms that Pd is not able to hydrogenate the carbonyl group under these conditions.

The stereoselectivity of hydrogenation of p-tert-butylphenol and thymol to cis-4-tert-butyl cyclohexanol and neoisomenthol on Rh or Rh black catalyst depends on the nature of solvent and catalyst support and on the reaction temperature and pH rather than hydrogen pressure. The reaction is studied varying hydrogen pressure in the 1-10Mpa range, pH 1.9-12.5, at 20-120?C, using isopropanol, hexane and water as solvents and TiO2' ", -Al2O3 and MgO as catalyst supports. The highest yields reported of cis- and neo- are 67 and 92 %, respectively (12).

A kinetic study of thymol hydrogenation on a well characterized supported catalyst in cyclohexane at 313-373 K under 3 mPa hydrogen was reported. The relative rate constants of the different reaction pathways either hydrogenation via menthone or isomenthone and direct hydrogenation to the 4 menthol diastereoisomers were determined from the changes in composition of the reaction medium during the reaction process. Hydrogenation via the menthone intermediates is the main route, the formation of isomenthone being favored. The configuration of the menthols, produced from direct hydrogenation or from the ketone intermediates, is controlled by the geometry of adsorption of the precursors on the metal surface, so that neoisomenthol with all substituents in the cis position is by far the most abundant stereoisomers produced (18).

Preliminary work was carried out with a Pd/MgO and doped with Fe+3 (nitrate salt) in atomic ratio Pd/Fe=1 as a promotor to activate the carbonyl group and to improve hydrogenation.

The thymol reduction over the catalyst under the above conditions with respect to the thymol used, produces a racemic mixture of menthol isomers with a mixture of the isomers neomenthol, neoisomenthol and isomenthol.

Effect of Catalyst Loading.

The effect of catalyst with respect to the amount of thymol added can be observed in figure 5. This effect is not so dramatic as it occurs with the pressure. The initial speed of conversion for a 20% catalyst is 2.8 bigger than for a 3% catalyst. Again, the selectivity conducts to a mixture of menthone and isomenthone, however in this case, the total thymol conversion decreases with the amount of catalyst. For a 3% catalyst the reaction after 90% conversion is much slower and does not reach the 100% conversion (reaches 95%). On the other hand, for a 10 and 20% catalyst the conversion is complete. This can be explained by the poisoning of the active site of Pd by the unreacted thymol (5 ) and the possible formation of a surface oxide layer over Pd, due to the humidity presence in the reactive and solvent, which affects the selectivity and activity of Pd (2). However, there are some reports of the hydrogenation of the solid 3-methyl-6-isopropylphenol at room temperature and under 1 bar hydrogen pressure using Pd/C, Pd/Al2O3, Rh/C, Rh/Al2O3, Pt/C, Pt/Al2O3, PtO2 or Raney Ni catalyst gave mainly the thermodynamic unstable products isomenthone and neosiomenthol. Pt and Rh are the most active catalysts. Ru/C gives iso products mainly (19).


Thymol has been hydrogenated over Raney Ni, stabilized Ni, or alloy Ni-Al-Fe at various temperatures an the products mixture containing (±) menthone, (±) isomenthone, (±) menthol and (±) isomenthol isomerized in different solvents in the presence of Raney Ni or Na. In the isomerization process, menthone or decalin as solvent gave yields of 80% of (±) menthol (20).

Effect of Thymol Loadings

The observations from the number of loadings are exhibited in figure 6. The rate of reaction with the number of thymol feeds over the same catalyst decreases slowly and the effect is bigger for the first catalyst portions. This demonstrates a decrease in the activity most probably by the catalyst poisoning.


Another studies on the effects of the nature of the catalyst, the temperature and the pressure on the yield of dl-menthol versus other menthol stereoisomers formed in the catalytic hydrogenation of thymol were studied. The optimum conditions of hydrogenation on Ni sponge (1 % w/w of the thymol amount) were a temperature of 195-205° C and a pressure of 20-40 atm., at which the content of dl-menthol in the hydrogenated products was 44% after 2 hrs of contact with the catalyst. Keeping the hydrogenated mixture in the presence of hydrogen at 20-40 atm and 210-220° C in contact with the catalyst for 5 hrs. produced rearrangement which raised the dl-menthol content in the mixture to 54-58%. The amount of dl-menthol was brought to the equilibrium point (57%) for a mixture of the 4 stereoisomeric menthols. Hydrogenation experiments with other catalysts (Cu-Cr, Cu-Ni, Ni prepared by decomposition of Ni formate, Ni deposited on kieselguhr) indicated that Cu-Cr was the best of the catalysts tested. With this catalyst, the amount of dl-menthol in the hydrogenated mixture was 54-58% after a conversion at 200-205° C, a pressure of 40-80 atm., and a time of contact with the catalyst of 4.0-4.5 hrs.


CONCLUSIONS

1. The presence of a basic support (MgO) increases the activity of Pd in the hydrogenation of thymol compared with the catalyst Pd/C y Pd/Al2O3. The hydrogenation of thymol over Pd/MgO produces only a racemic mixture of (±) menthone and (±) isomenthone.

2. The Pd is not able to hydrogenate the menthone and isomenthone. The addition of a promotor such as Fe+3, activates the carbonyl group and allows the formation of a mixture of (±) menthol isomers.

3. The temperature, pressure and amount of catalyst does not change the selectivity of the reaction in the conditions reported.

ACKNOWLEDGEMENTS

The authors would like to acknowledge the financial support from an operating grant Scientific Millennium Initiative (ICM 99-092) and R. Oliva the postdoctoral fellowship. We also thank Dirección de Investigación from Universidad de Concepción and the laboratories of Facultad de Ciencias Químicas.

 

REFERENCES

1. A. Tungler, T. Mathe, Z. Bende and J. Petró, Appl. Catal. 19, 365 (1985)         [ Links ]

2. A. Tungler, T. Mathe, J. Petró and T. Tarnai, Appl. Catal. A: General 79, 161 (1991)         [ Links ]

3. M. Besson, L. Bullivant, N. Nicolaus and P. Gallezot, J. of Catal. 140, 30 (1993)         [ Links ]

4. P. J. Porcaro and V. Johnston, Analytical. Chem., 33, 1748 (1961)         [ Links ]

5. A. L. Barney and M. B. Hass, Ind. and Eng. Chem., 36, 85 (1994)         [ Links ]

6. P. Kukula and L. Cerveny, Appl. Catal. A: General, 193, 285 (2000)         [ Links ]

7. J. Solodar, J. Org. Chem., 41, 3461 (1976)         [ Links ]

8. A. I. Allakhverdiev, N. V. Kulkova and D. Yu Murzin, Ind. Eng. Chem. Res., 34, 1539 (1995)         [ Links ]

9. S.R. Konuspaev, Kh. N. Zhanbekov, Zh. A. Bizhanov; T.S. Imankulov; R.K. Nurbaeva Kinetika I Kataliz, 35 (6), 893 (1994).         [ Links ]

10.A. I. Allakhverdiev, N. V. Kulkova and D. Yu Murzin, Cat. Lett. 29, 57 (1994)         [ Links ]

11. M. Besson, L-Bullivant; N. Nicolaus-Dechamp, P. Gallezot, Studies in Surf. Sci. and Catal. 78, 115 (1993).         [ Links ]

12. S.R. Konuspaev, N.V. Kulkova, D. Yu, Murzin, Kinetika I Kataliz, 34 (6), 1038(1994).         [ Links ]

13.Yong-Xi Li and K. J. Klabunde, J. of Catal., 126, 173 (1990)         [ Links ]

14.G. Cárdenas, J. Chil. Chem. Soc. 50,3 (2005)         [ Links ]

15.R. Segura, J. Reyes-Gasga, G.Cárdenas, Colloid Polym. Sci. 283,455 (2005)         [ Links ]

16.G. Cárdenas, S.Salinas, J.Chil. Chem. Soc. 49, 267 (2204).         [ Links ]

17.V. Ponec, Appl. Catal. A: General 149, 27 (1997)         [ Links ]

18. M. Besson, L. Bullivant, N. Nicolaus, P. Gallezot, J. of Catal. 140, 30 (1993).         [ Links ]

19. R. Lamartine; R. Perrin; G. Bertholon, Comptes Rendus des Seances de l'Academie desSciences, Serie C: Sciences Chimiques 219, 291 (1980).         [ Links ]

20.Teh-Chou, Chang; Hsueh-Ling, Su. Zhongyuan Xuebao (Eng) 6, 67 (1977).        [ Links ]

 

*corresponding author : gcardena@udec.cl