(281a) Hydrocarbon Conversion on Zeolite and Catalyst Deactivation

Summary

Coke formation, zeolitic acid sites deactivation and product distribution were studied during catalytic reaction of 1-pentene over USHY zeolite. Two novel techniques for coke classification and acid sites characterisation were developed to further understanding of coking. Cracking and hydride transfer were the predominant reaction in the first few minutes of time-on-stream which deactivated rapidly allowing isomerisations to become the main reactions afterwards. Coke formation was very strong initially which is in good correlation with the initial rapid deactivation. The hydrogens freed from high C/H ratio coke components contributed to gas phase formation of cracking and hydride transfer products. The amount of free acid sites agrees well with the total amount of coke. Coke preferentially deactivated the strongest acid sites which contributed to cracking reactions. 1. Introduction

The reaction of alkenes over zeolite-based catalysts is a long-established and important industrial process. The reaction mechanisms and deactivation have received considerable research attention due to their theoretical and commercial importance [1, 2]. Most postulated deactivation mechanisms fall into two categories, coke deposition directly on the acid sites, or coke blocking access to acid sites [3]. The variables of coke formation have a profound effect on product distribution and catalyst deactivation especially at initial reaction time. Albeit the advance in knowledge of coking and the introduction of advanced techniques for identifying these carbonaceous deposits, characterisation of coke remains the bottleneck in the elucidation of mechanisms of coke formation and hence development of new catalysts more resistant to coking and less sensitive to deactivation [1]. In this work we studied two aspects of catalytic reaction of 1-pentene on USY zeolite. Firstly, the coke components and acidity characterisation of coked zeolite were studied by two novel techniques: coke classification into hard coke and coke precursors by TGA (Thermal Gravimetric Analysis) [4, 5] and the discrimination of acid sites according to their strength by the novel TPD (Temperature Programmed Desorption) method using mild temperature pre-treatment and combining of TPD without and with NH₃ [6, 7]. Secondly, product distribution was discussed [5, 6, 7]. 2. Experimental Section

2.1 Experimental procedure

Catalytic reactions of 1-pentene over USHY zeolite were carried out in the temperature range of 523-623 K and atmospheric pressure in a fixed-bed reactor. The amount of calcined catalyst used in each experiment was 0.65g (1-cm-long catalyst bed). Steel wool was used in the reactor to ensure isothermal conditions. In these experiments, the reactant partial pressure was P(1-pentene) = 0.8 bar, (P(N2)= 0.2 bar). Products were collected in a ten-way sampling valve at various time-on-stream and analysed by GC. 2.2 Thermogravimetric analysis procedure

About 150 mg coked sample was first heated to 473 K (10 K/min) and maintained there for 60 min under flowing N₂ (60 mL/min) to remove adsorbed water and reaction-mixture components. Secondly the temperature was raised to 873 K (10 K/min) and kept for 30 min under N₂ flow (60 mL/min). During this period coke precursors were removed resulting into a sample weight decrease. By switching from N₂ to air at the final temperature (873K), the hard coke was burnt off and its weight was measured. The amount of coke precursors in the catalyst was calculated as the difference between the sample mass after drying at 473 K and switching from N₂ to air at 873 K. The amount of hard coke was estimated by the mass difference of the catalyst sample between before and after switching from N₂ to air, when the hard coke was completely burnt off. The whole procedure of calculating coke precursors/hard coke is illustrated in Fig. 1. 2.3 Temperature-programmed desorption procedure

The common method of acidity estimation with TPD needs sample pre-treatment at high temperatures which causes coke precursors to volatilise or decompose into smaller volatile fragments resulting into a falsification of the TPD signal. To avoid this falsification due to the chemically active character of coke precursors, we adopt indirect TPD methods with mild temperature sample pre-treatment to study the acid sites of coked zeolites. Around 50 mg coked catalyst was preheated for 1 h at 473 K (10 K/min) in He flow to remove water and reaction mixture components adsorbed on the catalyst. After saturated with NH₃ at 353 K it was heated to 383 K to desorb physisorbed NH₃. The linear temperature program (10 K/min) was then started from 383 K to 873 K and remained at 873 K for 30 min. The desorbed NH₃ and coke precursors were monitored continuously with a thermal conductivity detector. This is called First TPD. The fresh zeolite was also analysed by this method to measure the total acid sites. The same TPD was carried out initially without preceding NH₃ adsorbed. This was called TPD without NH₃. Second TPD was carried out with 873 K preheating instead of 473 K. During this period of preheating, coke precursors as well as water adsorbed and reaction mixture components were removed. The First TPD contains both NH₃ and coke precursors adsorbed on the zeolite. However, the TPD without NH₃ only contains coke precursors. By subtracting the signal of TPD without NH₃ from the signal of First TPD, the free acid sites of coked sample inhabited by both coke precursors and hard coke can be calculated. First TPD of fresh zeolite indicated the total acid sites of fresh zeolite. Since coke precursors have been removed after the pre-treatment at 873 K in He stream during Second TPD, only hard coke deposited on the catalyst. Therefore, the signal of Second TPD shows the free acid sites of coked zeolite inhabited only by hard coke. All these TCD signals were normalized by the zeolite weight. These calculation processes can be clearly illustrated in Fig. 2. With these methods, not only the amount of acid sites but also the corresponding acid strength distribution can be achieved. 3. Results and Discussion

With the coke classification by TGA (Figs.3-4) we can clearly interpret TOS dependence of coke precursors and hard coke as well as to elucidate steps of overall coking mechanism. Coke formation was an extremely rapid process at the beginning of catalyst exposure to the reaction mixture. The formation rate of total coke content was particularly high during the first minute of TOS while it became much slower afterwards. More hard coke was formed than coke precursors during the deactivation. The amount of coke precursors decreased with increasing reaction temperature due to higher desorption of coke precursors into gas phase while hard coke amount increased with temperature as expected from an activated process. From Figs. 5-6, it can be observed that USHY zeolite suffered a strong acid sites deactivation especially at initial stage while much slow afterwards during 1-pentene reaction. The amount of free acid sites is correlated well with the total amount of coke. Coke preferentially deactivates the strongest acid sites. In Fig. 7, we can see that cracking and hydride transfer were predominant reactions only in the first couple of minutes experiencing a rapid deactivation, giving rise afterwards to isomerisation reactions, especially double bond isomerisation. The main products after three minutes of TOS were trans- and cis-2-pentene. The initial hydride transfer products were formed due to the release of hydrogen during the transformation of hydrogen rich gas phase reaction components to hydrogen poor coke components on the catalyst surface.

REFERENCES

[1] Shuo Chen and George Manos, Catal. Lett. 96 (2004) 195-200.

[2] Aristidis A. Brillis and George Manos, Catal. Lett. 91 (2003) 185-191.

[3] P. D Hopkins and A. T. Miller, Appl. Catal. A: Gen. 136 (1996) 29-48.

[4] B. Wang, G. Manos, A Novel Thermogravimetric Method for Coke Precursors Characterisation (submitted to J. Catal.).

[5] B. Wang, G. Manos, Deactivation Studies during Catalytic Cracking of 1-Pentene over USHY Zeolite (submitted to Appl. Catal. A: General)

[6] B. Wang, G. Manos, A Novel TPD Method for Study of Acid Sites in USHY Zeolite during Hydrocarbon Conversion (in preparing)

[7] B. Wang, J. Yu, G. Manos, Role of Zeolitic Strong Acid Sites on Hydrocarbon Reactions (in preparing)

Fig. 1. Coke precursors and hard coke of a coked sample during thermogravimetric anaylsis.

Fig. 2. Procedure of indirect TPD method with mild temperature pre-treatment.

Fig. 3. Coke content during 1-pentene reactions over USHY zeolite at 623 K.

Fig. 4. Coke content after 20 min TOS during 1-pentene reactions over USHY zeolite at different reaction temperatures.

Fig. 5. Free acid sites of deactivated USHY zeolite coked from 1-pentene reaction at different TOS (T = 623 K).

Fig. 6. TPD of fresh catalyst, Free acid sites and Second TPD (sample from deactivated USHY zeolite coked from 1-pentene reaction at 623 K and 20 min-on stream).

Fig. 7. Product distribution of 1-pentene reaction over USHY zeolite at 623 K.