Effect of combining metallic and acid functions in CZA/HZSM-5 desilicated zeolite catalysts on the DME steam reforming in a fluidized bed

https://doi.org/10.1016/j.ijhydene.2013.05.134Get rights and content

Highlights

  • Bifunctional catalysts have been used in the dimethyl ether steam reforming.

  • Balance should be struck between acidity and number and dispersion of metal sites.

  • The method for contacting the phases and their mass ratio are crucial aspects.

  • A high H2 yield is obtained with the catalyst prepared by wet physical mixing.

  • A mass ratio of 1:1 between metallic and acid functions is suitable for the process.

Abstract

A study has been carried out on the effect of combining the metallic (CuO–ZnO–Al2O3) and acid (HZSM-5 zeolite treated with NaOH) functions, and of their mass ratio, on the kinetic performance in the steam reforming of dimethyl ether (SRD). The runs have been conducted in a fluidized bed reactor in the 225–325 °C range, and the reaction indices (dimethyl ether and methanol conversions, and H2 and CO yields) have been explained based on the physico-chemical properties of the catalyst.

A suitable synergism between dimethyl ether hydrolysis and methanol reforming steps requires striking a balance between acidity and the number and dispersion of metal sites, as happens in the catalyst prepared by wet physical mixing with a mass ratio of 1:1 between the metallic function and the acid one. This catalyst allows obtaining high values of dimethyl ether conversion and H2 yield in the 275–300 °C range by minimizing the CO formation rate and deactivation by coke.

Introduction

Amongst the compounds available for use as H2 vectors in proton exchange membrane fuel cells (PEMFC), dimethyl ether (DME) has advantages (even over methanol) thanks to its high hydrogen content (13 wt% vs. 12.5 wt% of methanol), low toxicity (harmless), gas-like properties, liquid-storage density, and the infrastructure available for its handling (similar to that for LPG) [1], [2]. Furthermore, the steam reforming of DME (SRD) can also proceed at low temperatures, similarly to the steam reforming of methanol (SRM) [3], [4], [5], [6], [7], [8], [9], [10].

The DME synthesis by co-feeding CO2 with syngas in a single reaction step (by integrating methanol synthesis and its subsequent dehydration to DME) is therefore considered a key process for both CO2 sequestration [11], [12], [13] and the viability of lignocellulosic biomass gasification [14]. The large-scale reforming of DME obtained from biomass is an interesting short-term solution for meeting the growing H2 requirements of fuel hydroreforming processes in refineries [15], and a medium-term solution for H2 requirements as an automotive fuel [16].

The SRD process consists of two steps in series:DME hydrolysis (on the acid function in the catalyst): (CH3)2O + H2 2CH3OHMeOH steam reforming (on the metallic function): CH3OH + H2 3H2 + CO2Overall reaction: (CH3)2O + 3H2 6H2 + 2CO2

Besides, the reverse water-gas shift reaction (r-WGS) takes place over the metallic function:CO2 + H2  H2O + CO

Methane can also be generated via DME decomposition, when a strong acid catalyst or high reforming temperatures are used [17], [18]:(CH3)2O → CH4 + H2 + CO

Moreover, the conversion of methanol and DME into hydrocarbons can take place at temperatures above 300 °C on strong acid sites [19], and significant coking may occur on the catalyst surface due to the dehydrogenation of the hydrocarbons formed [20]:Oxygenates → hydrocarbons → coke

A suitable catalyst should be used to attain high yields of H2, given that: (i) steps (1) and (2) need to be sufficiently high and well matched and (ii) steps (4)–(6) should be minimized and therefore the production of coke (it deactivates the SRD catalyst) and CO (it poisons the fuel cell anode) should also be minimized.

This paper studies the kinetic performance in the SRD process of catalysts prepared by different methods for combining the metallic and acid functions and using different mass ratios between the functions. The suitable composition of each one of these functions has been reported in previous papers [21], [22]. Thus, an HZSM-5 zeolite desilicated by treating with NaOH (the Si/Al ratio decreases from 15.3 for the parent zeolite to 13.0 for the treated zeolite) minimizes the formation of hydrocarbons and coke (Eq. (6)) due to the decrease in zeolite acidity [21]. Furthermore, this alkaline treatment generates mesopores by micropore opening, and also new micropores and mesopores. The zeolite treated has sufficient activity for DME hydrolysis below 300 °C (Eq. (1), SRD rate-limiting step), which is the ceiling temperature for avoiding Cu sintering in the metallic function [23].

A previous paper [22] deals with the effect the composition of CZA metallic function (CuO–ZnO–Al2O3) has in the methanol reforming step and in DME reforming. Thus, the CZA metallic function with Cu/Zn/Al atomic ratio = 4.5:4.5:1.0 and calcined at 325 °C is suitable for preparing the bifunctional catalyst for SRD, given that it is highly active, stable below 300 °C (resistant to deactivation by metallic site blockage by coke deposition) and gives way to low CO formation rate.

This paper approaches the study of SRD process in a fluidized bed reactor to ensure the isothermicity of the catalytic bed.

Section snippets

Catalyst preparation

The acid function is a ZSM-5 zeolite obtained by treating a commercial ZSM-5 zeolite (SiO2/Al2O3 = 30, Zeolyst International) supplied in ammonium form, with a 0.2 M solution of NaOH (Panreac, 99%), by stirring at 80 °C for 300 min, using a reflux condenser to prevent evaporation [21]. The concentration of 0.2 M for the NaOH solution is advisable for avoiding the dealuminization of the zeolite [24], [25]. The sodium ZSM-5 zeolite treated has been ion-exchanged with an ammonium nitrate solution,

Effect on the catalyst properties

Fig. 2 shows the X-ray diffractogram patterns for the catalysts synthesized following the methods described in Section 2.1 (wet impregnation, coprecipitation and wet physical mixing), with mass ratio between the metallic and acid functions (M/A) being 1:1. The peaks corresponding to CuO and ZnO metallic phases are observed, as are those corresponding to the crystalline structure of the zeolite, but not those corresponding to Al2O3 (which is probably due to a high dispersion of this phase or to

Conclusions

The method for contacting the metallic and acid functions in the preparation of the bifunctional catalyst does not affect the resulting crystalline structure of the functions nor their acidity, but significantly affects the incorporation of metal charge into the catalyst and its metal surface area. The methods of wet physical mixing and coprecipitation are suitable for obtaining catalysts for SRD in a reproducible way with nominal metal charge. The metal surface area of the catalysts obtained

Acknowledgements

This work has been carried out with the financial support of the Ministry of Science and Technology of the Spanish Government (Projects CTQ2006-12006 and CTQ2009-13428), University of the Basque Country, UPV/EHU (UFI 11/39) and of the Basque Government (Project GIC07/24-IT-220-07).

References (41)

Cited by (35)

  • Inside the reaction mechanism of direct CO<inf>2</inf> conversion to DME over zeolite-based hybrid catalysts

    2021, Applied Catalysis B: Environmental
    Citation Excerpt :

    In this respect, several approaches have been reported by different research groups, like the preparation of core–shell hybrid catalysts (with a core of Cu–based catalyst and a shell of a zeolite) [46], the mixing of a zeolite with Cu–ZnO active sites confined in a mesoporous silica [47], the combination of an acid matrix with a Cu-based catalyst supported on carbon nanotubes [48], the immobilization of nanoparticles of CuO and ZnO onto acid supports [49], the mixing of colloidal nanoparticles of Cu/ZnO with γ-Al2O3 or HZSM-5 [50,51]. Other studies have shown how the generation both of metal-oxides and acid sites in a single catalyst grain is able to improve the conversion of CO2 compared to conventional mechanical mixing of a methanol synthesis catalyst and a zeolite, also allowing a higher rate of MeOH formation/dehydration on neighbouring surface sites [52,53]. Nevertheless, there are different opinions on the efficiency of bifunctional catalysts in comparison to admixed systems.

View all citing articles on Scopus
View full text