CATALISIS 2 (2013) 1-13.

 

Transformation of methanol into aromatic compounds over MFI type zeolites

 

Transformación de metanol en compuestos aromáticos sobre zeolitas del tipo MFI

 

Albania Villarroel1, Djamal Djaouadi1, FÉlix AvendaÑo1, Luis GarcÍa2

 

1Petróleos de Venezuela (PDVSA)-Intevep, Gerencia General de Refinación e Industrialización, Gerencia de Valorización de Corrientes, Apartado 76343, Caracas 1070-A, República Bolivariana de Venezuela

E-mail: villarroelax@pdvsa.com

2Escuela de Ingeniería Química, Facultad de Ingeniería, Universidad Central de Venezuela. A.P. 47102, Caracas 1020-A, Venezuela .

 

 

ABSTRACT

 

               A large amount of petcoke is produced in Venezuela due to the crude oil processing in the refining industry. Gasification of this pet-coke could be an alternative to its valorization. Thus the syngas produced can be transformed in methanol and then in more added-value products such as aromatics compounds (MOGD Process). This study has been done to explore the transformation of methanol to aromatics compounds over medium pore size zeolites (MFI type), taking into account that olefins compounds will be also generate as intermediaries. For all used zeolites, at any operational condition, high selectivity was found towards aromatics compounds - mainly xylenes - , regarding with the olefins, ethylene and propylene were the most favored. The overall conversion value was close to 90%, based on methanol transformation, for all evaluated catalyst.

 

Keywords: Aromatics, methanol transformation, MFI type zeolites, olefins.

 

 

RESUMEN

 

En Venezuela se produce una gran cantidad de coque de petróleo (pet-coke) debido al procesamiento de crudo en la industria de refinación. La gasificación de este pet-coke podría ser una ruta alterna a su valorización. De esta forma, el gas de síntesis producido se podría transformar en metanol y luego en productos de mayor valor agregado, tales como los compuestos aromáticos (proceso MOGD). Este estudio ha sido realizado para explorar la transformación de metanol en compuestos aromáticos sobre zeolitas de tamaño de poro medio (tipo MFI), tomando en cuenta que se generarán compuestos olefínicos como intermediarios. Para todas las zeolitas empleadas, en cualquier condición operacional, se encontró alta selectividad hacia compuestos aromáticos - principalmente xilenos -, mientas que en cuanto a las olefinas, etileno y propileno fueron las más favorecidas. El valor de conversión global estuvo cerca de 90%, basado en la transformación del metanol, para todos los catalizadores evaluados.

 

Palabras clave: Compuestos aromáticos, olefinas, transformación de metanol, zeolitas tipo MFI.

 

© Sociedad Venezolana de Catálisis. Todos los derechos reservados. Para permiso, envíe un correo electrónico a: revista.catalisis@gmail.com

 

 

INTRODUCTION

 

            The syngas obtained through gasification of pet-coke can be used to produce methanol, which can be transformed in more added-value products as aromatics compounds. Those products are one of the basic raw materials in the refinery and petrochemical industry; benzene, toluene and xylenes (BTX) are the most commonly used.

 

            The commercial processes for aromatics production are almost based on processes such as reforming and cracking using crude oil. However, it is important to explore cleaner technologies to produce petrochemicals compounds, especially BTX.

 

            Methanol conversion to olefins over H-ZSM-5 zeolites (MTO Process) has been the subject of many researches [1, 2, 3] and is considered the first stage of MTA process [4], as can be seen in the following scheme:

 

 

Figure 1. General scheme of MTA Process.

 

            Chang and Silvestri [5] reported that at 644K and atmospheric pressure, the aromatics fraction in the products was 41% by weight; according to the accepted mechanism, the aromatic hydrocarbons are formed from olefins. The major secondary products that result in MTA process are ethylene, propylene, butenes and pentenes; the heavier compounds as gasoline fraction C5-C10 were found in lesser amount.

 

            The general aim of this research is to study the methanol conversion to aromatic compounds over medium pore size zeolites (MFI type). It was examined the influence of the following factors on selectivity and overall conversion: a) modification of the Si/Al relation for two zeolites in order to vary the number and strength of acid sites, b) modification of process conditions such as reaction temperature and space velocity.

 

 

EXPERIMENTAL

 

Catalyst preparation. For evaluating the effect of reaction temperature and space velocity, ZSM-5 commercial zeolite catalyst was used.

 

            The study of the influence of the Si/Al ratio was carried out with MFI types zeolites synthesized by Intevep and named INT-MFI [6]. Both catalysts (ZSM-5 and INT-MFI) were subjected to ion exchange treatment with a solution of (NH4)2SO4, followed by calcination at 500 ° C, to obtain the protonated zeolites.

 

            For catalysts characterization it were used five techniques: X-ray diffraction (XRD), pyridine adsorption by infrared spectrophotometer, inductively coupled plasma optical emission spectrometry (ICP-OES), BET analysis by nitrogen adsorption and temperature programmed oxidation (TPO).

 

            The crystallinity degree of the catalysts was obtained by X-ray diffraction on a difractometer PANalytical X’Pert PRO model.

 

            Brönsted and Lewis acidity was calculated by pyridine adsorption with an infrared spectroscopy Perkin-Elmer 1750 model. With the infrared spectra of samples containing basic molecules such as pyridine, it can get information about the nature of acid centers.

 

            The ICP-OES was used to obtain the elemental composition of a solid and thus the unit cell formulas of the zeolites, which are shown in Table 1.

 

Table 1. Chemical composition of zeolites.

Catalyst

Ratio Si/Al (% molar)

Unit cell formula

ZSM-5 (11)

10,85

H+8,10 [Al8,10 Si87,90 O192]·16H2O

INT-MFI (12)

11,83

H+7,48 [Al7,48 Si88,52 O192]·16H2O

INT-MFI (35)

35,45

H+2,63 [Al2,63 Si93,36 O192]·16H2O

INT-MFI (100)

100,1

H+0,95 [Al0,95 Si95,05 O192]·16H2O

 

            To determine textural properties, it was used BET analysis by nitrogen adsorption with a Micromeritics equipment Tristar 3000 model.

 

            TPO technique was used to obtain the temperature profile of coke deposited on the zeolites surface, through a thermal oxidation. The equipment used was a Micromeritics Autochem 2 model.

 

            The catalysts used were ZSM-5 zeolite ratio Si/Al of 11 – ZSM-5 (11) – and INT-MFI zeolites ratios Si/Al of 12, 35 and 100 – INT-MFI (12), INT-MFI (35), INT-MFI (100).

 

Experimental setup. For the catalytic tests, methanol was used as a reactive (99% Aldrich) and nitrogen was used as a diluent (99%). It was used a Pyrex glass reactor of 610mm long and 18mm wide, with a bed catalyst of particle size between 355 and 425 µm. Reaction heat was supplied by an infrared furnace Research Inc. Box 24064 model connected to a Micristrar controller. Nitrogen flow was regulated by an electrically valve and methanol injection was done through a positive displacement pump ISCO. Experimental setup used in this research is shown in Fig. 2.

 

            The temperature effect was studied between 350 and 500 °C, using a nitrogen flow of 10cc/min and 10h-1 of WHSV. To the study of WHSV effect the following values were selected: 10h-1, 24h-1 and 50h-1, conditions of temperature and nitrogen flow were set at 450 °C and 10 cc/min respectively.

 

            The study of relation Si/Al effect was carried out under optimal operating conditions found in the catalytic tests. The reaction time used for comparative purposes was about 200min.

 

 

Figure 2. Experimental setup used for MTA process.

 

Product analysis. Products at reactor outlet were flowing through a condenser connected to a cryostat, which operated at 4 °C and allowing the separation of gases and liquids. Reaction gas products were analyzed online by gas chromatography using a Hewlett-Packard 5890 Series II, with a DB1 column of 60m in length and 32mm in diameter, and an ionization detector (FID).

 

            Two phases were obtained into liquid stream: an aqueous phase (mainly water and remaining methanol) and an organic phase composed by paraffins, olefins and aromatics. This phases were separated and the organic compounds were analyzed by gas chromatography using a Agilent Technologies 6890N model with PIONA application and a WASSON×ECE×NP×3013KC 148 column of 100m in length.

 

Experimental data processing. It was obtained the yield both in the organic phase as in the gas phase as through chromatographic analysis. In addition, from a global balance and considering that it was possible to measure the amounts of methanol fed and the resulting liquid product for a given operating time, the amount of gaseous products were determined.

 

      With the global mass and the yields, it was determined the mass of each component present in the mixture reaction, using the following equations:

 

                   (1)

                        (2)

 

      Once known the masses, it were calculated the corresponding moles and therefore the conversion were determined. Assuming that the stoichiometric ratio is met, it can establish a relationship between the methanol moles that have to react to form the product i and the carbon numbers in that product. For benzene:

 

 

           (3)

 

      This procedure it was repeated for all the reaction products and the sum represents the moles of methanol that reacted:

 

              (4)

 

 

Finally, methanol conversion was calculated using the following expression:

 

             (5)

 

 

RESULTS AND DISCUSSION

 

      In this section the results obtained for studying the effect of the variation of temperature, space velocity and Si / Al ratio on product distribution and methanol conversion are shown.

 

1. Temperature effect over product distribution and conversion.

      Space velocity and nitrogen flow were kept constants    at    10h-1    and    10 cc/min respectively.   Methanol conversion is about 13% for all temperatures studied, as shown in Table 2.

 

Table 2. Methanol conversion over ZSM-5 (11)

(200 min, 10 h-1, 10 cc/min N2)

Temperature (°C)

350

400

450

500

Conversion (% mol)

13,6

13,1

13,5

13,5

 

      Selectivity to paraffins, olefins and aromatics for different temperatures is shown in Fig. 3. It can be seen a drop in the olefins production and a growth in aromatics formation, when the temperature rises.

 

      It is clearly observed the behavior of the olefins as intermediates for the production of aromatic compounds.

 

      As for the paraffinic compounds, it was observed a slight decrease with temperature. These results indicate that oligomerization and cyclization reactions were favored with temperature increases. The appearance of paraffins implies the existence of hydrogen transfer reactions. Hydrocarbons distribution for all temperatures studied can be seen in Table 3.

 

 

Figure 3. Selectivity to paraffins, olefins and aromatics at different temperature

 

 

Table 3. Temperature effect over hydrocarbon distribution using ZSM-5 (11) zeolites

(200min, 10h-1, 10cc/min N2)

Temperature (°C)

350

400

450

500

                                      molar %

 

 

Methane

1,44

3,52

7,71

10,88

Ehtylene

17,02

15,16

11,51

9,51

Ethane

0,32

0,53

0,66

0,41

Propylene

8,22

9,40

9,09

6,15

Propane

6,08

10,39

4,41

2,74

Isobutane

9,97

15,38

7,29

5,23

n-Butane

0,00

0,00

0,00

0,00

t-2-Butene

2,15

1,86

1,61

1,51

1-Butene

0,18

0,00

0,00

0,28

Isobutene

1,62

2,74

1,36

0,80

c-2-Butene

0,00

0,00

0,00

0,00

Isopentane

0,01

0,03

0,03

0,02

n-Pentane

0,00

0,00

0,00

0,00

t-2-Pentene

5,41

6,61

3,25

2,94

1-Pentene

0,00

0,00

0,00

0,00

c-2-Pentene

0,03

0,04

0,05

0,04

Benzene

0,73

0,20

0,11

0,05

Toluene

3,32

3,53

3,72

4,32

Ethylbenzene

1,14

0,75

1,14

1,42

Xylenes

29,13

20,79

32,70

39,35

1,2,4-TMB

16,28

7,58

14,94

17,79

 

            According to mechanisms proposed to explain the transformation of mixture DME / Methanol into light olefins and aromatics compounds, the first product formed is ethylene [7, 8, 9]. The results observed in Table 3 are consistent with these theories.

 

            It found a higher selectivity to aromatics compounds, principally to xylenes. Aromatics compounds can be formed by ethylene and propylene oligomerization reactions, followed by ciclization and aromatization processes. TMB formation is affected by the shape selectivity of zeolites. It was observed the formation of 1,2,4-TMB  and  not   the   other   isomers,   this  is because the Van Der Waals diameter (7,4 Å) is similar to the o-xylene molecule.

 

            As to the olefins, the ethylene selectivity is higher than for the propylene for all temperatures tested, however, the difference between both molar percentages decreased with the increasing of reaction temperature. This is an indicative that the alkylation of ethylene with methanol, that could be one way for the propylene formation, is favored with increasing temperature.

 

            At temperatures studied, propylene has higher selectivity than pentenes, especially at high temperatures (450 – 500 °C). These results are an indicative that propylene can also be formed due to cracking reactions of pentene or heavier olefins. Butenes could be formed by ethylene dimerization or propylene alkylation with methanol [9, 10, 11]. The molar percentages of butenes is lower than ethylene and propylene, these compounds could be forming heavier compounds such as xylenes.

 

2. Space velocity effect over product distribution and conversion. Temperature (450 °C) and nitrogen flow (10 cc/min) were kept constants. When space velocity was 50h-1, global conversion reached values close to 13%. Selectivity (% molar) to paraffins, olefins and aromatics for different space velocities is shown in Table 4.

 

Table 4. WHSV effect over product distribution using ZSM-5 (11) zeolite.

 

WHSV (h-1)

10

24

50

 

molar %

Paraffins C1-C5

22,96

35,77

12,06

Olefins C2-C5

27,51

43,43

32,81

Aromatics (BTX)

48,38

20,80

55,13

 

            The higher selectivity to aromatics was obtained operating to 50h-1. It was found a maximum in selectivity to paraffins and olefins when the space velocity was between 10h-1 and 24h-1.

 

3. Ratio Si/Al effect over product distribution and conversion. Temperature (450 °C), nitrogen flow (10 cc/min) and space velocity (10 h-1) were kept constants. For all the ratios Si/Al global conversion achieved values close to 13%.

 

            Selectivity to paraffins, olefins and aromatic is reported in Table 5. It was observed that molar percentage of olefins and aromatics increases with the ratio Si/Al, while the molar percentage of paraffins decreases.

 

(200min, 450°C, 10h-1 10 cc/min N2)

Ratio Si/Al

12

35

100

 

molar %

Paraffins C1-C5

45,61

14,68

8,13

Olefins C2-C5

20,13

41,69

44,59

Aromatics (BTX)

31,86

43,78

47,24

Table 5. Effect of ratio Si/Al over product distribution using the INT-MFI zeolite

 

            Paraffins formation involves hydrogen transfer reactions (bimolecular reactions); the increase of the ratio Si/Al implies an active sites separation, so the bimolecular reactions are unfavorable. These effects explain the trend observed in Table 5. 

 

            The tendency resulting in olefins and aromatics formation can be explained by the increases in acid sites strength that occur product of the dealumination process, coupled with the fact that cracking reactions for olefins formation and cyclization do not require two adjacent active sites to take place.

 

            Aromatics compound formation is favored at high Si/Al ratio, however, these conditions also favor the catalyst deactivation.

 

4. Brönsted and Lewis acidity by pyridine adsorption. It was determined Brönsted and Lewis acidity for   ZSM-5 (11) and INT-MFI (12) by pyridine adsorption; the results are shown in Tables 6. INT-MFI (12) zeolites have largest number of strong Brönsted acid sites than ZSM-5 (11).

 

Table 6.  Brönsted and Lewis acidity for ZSM-5 (11) and INT-MFI (12) zeolites

Sites Type

Brönsted

N° Lewis

(μmol/gr)

(μmol/gr)

ZSM-5 (11)

Weak

91

118

Media

77

79

Strong

127

84

 

INT-MFI (12)

Weak

23

84

Media

215

131

Strong

249

74

 

            Acidity analysis was done in order to compare zeolites with a similar Si/Al relation, which is why the results for INT-MFI (35) and INT-MFI (100) were not presented.

5 Characterization of zeolites by Temperature Programmed Oxidation (TPO). Using this technique it could be observed the signals of coke deposited on the catalyst surface after had been used. Fig. 4 shows the spectra of the coke oxidation as a function of temperature for ZSM-5 (11) and INT-MFI (12) zeolites.

 

 

Figure 4. Thermograms of coke oxidation for spent zeolites ZSM-5 (11) and INT-MFI (12).

 

            The thermogram of ZSM-5 (11) zeolites has a maximum signal at 600°C. There are two signals in the thermogram of INT-MFI (12) associated with the oxidation of two types of coke, with a maximum of 450°C and 600°C. Importantly, the signal to a maximum temperature of 600°C for INT-MFI (12) zeolites is similar to that of ZSM-5 (11). The results observed in Fig. 4 indicate that the coke presents in ZSM-5 (11) zeolites differs with the coke that is located in INT-MFI (12) zeolites.

 

6. Surface area determination by BET analysis with nitrogen adsorption. Nitrogen adsorption isotherms for ZSM-5 (11) and INT-MFI (12) zeolites fresh and spent are shown in Fig.5.

 

            In the isotherm of zeolites fresh ZSM-5 (11) a hysteresis is observed at a relative pressure range of 0,45 to 0,9, this phenomenon is  characteristic  of  mesoporosity materials. The isotherm of INT-MFI (12) without using reflects microporosity.

 

            In zeolites with coke impregnated on the surface, the isotherm of ZSM-5 (11) shows hysteresis due to the mesoporosity, which is smaller than observed to the same zeolites before use. The isotherms of INT-MFI (12) zeolites are similar in both cases.

 

 

Figure 5. Nitrogen adsorption isotherms of ZSM-5 (11) and INT-MFI (12) zeolites before and after use.

 

            Surface area values of zeolites used in this research are show in Table 7. Both zeolites - ZSM-5 (11) and INT-MFI (12) - without using have a similar surface area. Values of fresh INT-MFI zeolites are bounded in a range of approximately 306 to 426 m2/gr.

 

            Comparing the values of fresh and spent zeolites, it was observed that surface area of used INT-MFI (12) zeolite was 92% lower than the surface area of the same fresh zeolites. In the case of ZSM-5 the reduction was 69%.

 

Table 7. Surface area of used zeolites.

 

Catalyst

BET Surface area

 (m2/gr)

Fresh ZSM-5 (11)

418

Fresh INT-MFI (12)

407

Fresh INT-MFI (35)

306

Fresh INT-MFI (100)

426

Spent ZSM-5 (11)

130

Spent INT-MFI (12)

33

 

            The difference in percentages may be due that    INT-MFI (12) zeolite presents higher microporosity, so coke formation on the surface is more favored than in ZSM-5 (11) zeolite. The oxygen treatment (TPO) for used zeolites allowed to remove certain amount of the coke deposited on the solids. The superficial area of treated catalyst was also calculate, for   ZSM-5 (11) the area was 366 m2/gr while for the INT-MFI (12) was 403 m2/gr.

 

            Unlike what was found for ZSM-5 (11) zeolites, the superficial area of INT-MFI (12) treated is slightly lower than the area of the same zeolites before use; this difference indicates that this solid has a higher regenerability.

 

7. Mechanism of methanol transformation into aromatics compounds. As shown in Figure 1, to convert methanol into aromatic compounds must pass through an intermediate stage of production of olefins. First light olefins are obtained and then, by alkylation or oligomerization reactions are obtained heavier olefins which are the precursors of aromatic compounds.

 

            Kaeding and Butter [8] proposed a carbocations mechanism to explain the formation of light olefins from methanol.

 

 

Figure 6. Light olefins formation from carbocations as intermediates.

 

            Olefin chain growth was explained by Cormerais [9] through oligomerization reactions:

 

 

Figure 7. Oligomerization reactions of light olefins.

 

            Finally, the olefins are converted into aromatic compounds by a mechanism that includes, besides the cyclization, a series of reactions of hydrogen transfers:

 

 

Figure 8. Aromatization reactions of light olefins.

 

CONCLUSIONS

 

            INT-MFI zeolite is a catalyst with high potential for used in the transformation of methanol into olefins and aromatics compounds.

 

            The optimum conditions found in this research to the transformation of methanol into olefins and aromatics compounds over medium pore zeolites are: 450 °C, 10 h-1 and 10 cc/in of nitrogen.

 

            Si/Al ratio is a variable that determine the catalyst deactivation, and high temperatures decrease the life of the solid because it promotes the coke formation. On the other, the increase of Si/Al ratio leads to more stable catalysts.

 

            TPO analysis showed that INT-MFI (12) zeolites recover most of the initial textural properties.

 

GLOSSARY

 

: Total mass of methanol fed at a given time t

: Total mass of liquid product obtained at a given time t

: Total mass of gas product obtained at a given time t

: Mass of component i at stream j

: Total mass of stream j

: Mass fraction of component i at stream j

 

 

REFERENCES

 

1.      C. Chu, C. Chang. J. Catal. 86 (1984) 297.

2.      G. J. Hutchings, R. Hunter. Catal. Today 6 (1990) 279.

3.      F. Patcas. J. Catal. 231 (2005) 194.

4.      S. L. Meisel. Chemtech, 1 (1988) 32.

5.      C. Chang, A. Silvestri. J. Catal. 47 (1977) 249.

6.      N. Martinez and coll. US Patent 5,254,327; Oct. 19, 1993. (7)

7.      W. Zatorski, S. Krzyzanowski. Acta Phys. Chem. 24 (1978) 347.

8.      W. Kaeding, S. Butter. J. Catal. 61 (1980) 155.

9.      F. Cormerais. Transformation de méthanol et le diméthyléther en oléfines légères”. Doctoral Thesis. Poitiers University. 1981.

10.   M. Bjørgen, S. Svelle, F. Joensen, J. Nerlov, S. Kolboe, F. Bonino, L. Palumbo, S. Bordiga, U. Olsbye. J. Catal. 249 (2007) 195.

11.   J. R. Anderson, T. Mole, V. Christov. J. Catal. 61 (1980) 477.

12.   M. R. F. Siggel, T. D. Thomas. J. Electron Spectrosc. Relat. Phenom. 48 (1989) 101.

13.   M. A. Fahim, T. A. Alsahlaf, A. Elkilani. “Fundamentals of Petroelum Refining.” Elsevier, Amsterdam (2010). 95-152.

14.   Y. M. Ni, A. M. Sun, X. L. Wu, J. L. Hu, T. Li, G. X. Li. Chin. J. Chem. Eng. 19 (2011) 439.