Numerical and experimental study on hydrogen production via dimethyl ether steam reforming

https://doi.org/10.1016/j.ijhydene.2020.02.091Get rights and content

Highlights

  • 3D CFD model for DME/steam reforming model is developed.

  • Test rig for DME/steam reforming with catalyst is developed.

  • Comparisons between CFD and experimental results are made for DME conversion and hydrogen yield.

Abstract

This work presents the characteristics of catalytic dimethyl ether (DME)/steam reforming based on a Cu–Zn/γ-Al2O3 catalyst for hydrogen production. A kinetic model for a reformer that operates at low temperature (200 °C–500 °C) is simulated using COMSOL 5.2 software. Experimental verification is performed to examine the critical parameters for the reforming process. During the experiment, superior Cu–Zn/γ-Al2O3catalysts are manufactured using the sol-gel method, and ceramic honeycombs coated with this catalyst (1.77 g on each honeycomb, five honeycombs in the reactor) are utilized as catalyst bed in the reformer to enhance performance. The steam, DME mass ratio is stabilized at 3:1 using a mass flow controller (MFC) and a generator. The hydrogen production rate can be significantly affected depending on the reactant's mass flow rate and temperature. And the maximum hydrogen yield can reach 90% at 400 °C. Maximum 8% error for the hydrogen yield is achieved between modeling and experimental results. These experiments can be further explored for directly feeding hydrogen to proton exchange membrane fuel cell (PEMFC) under the load variations.

Introduction

In consideration of the fossil fuel crisis, proton exchange membrane fuel cells (PEMFCs) are considered potential alternative energy sources [1]. Hydrogen is a fuel source for PEMFCs, in which chemical energy is converted into electrical energy. PEMFCs are regarded as promising energy sources for meeting the growing energy demand because of their high energy conversion efficiency, low noise, and environmental friendliness. However, hydrogen supply and storage are two major barriers to the commercial application of PEMFCs due to the lack of hydrogen supply infrastructure. Therefore, supplying hydrogen to PEMFCs is an important factor for the widespread usage [2]. Several researchers have focused on producing hydrogen through a steam reforming (SR), which can flexibly supply hydrogen to PEMFCs [[3], [4], [5], [6]].

Among the different SR processes that use hydrocarbon-based fuels, the simplest ether, namely, dimethyl ether (DME) has emerged as a clean substitute for other fuels due to its high cetane number (55–56), high hydrogen yield ratio, and low working temperature [7,8]. Therefore, DME can provide an excellent solution to hydrogen storage and carrier [9]. Several technologies such as SR, autothermal reforming [10], and partial oxidation (PO) [11] are available for producing hydrogen from DME. Among these techniques, the SR of DME can be performed at considerably lower temperatures than those of other methods [12]. SR also has other advantages, such as achieving an effective hydrogen yield in stream and low CO yield; therefore, it has been recently recognized as one of the most important routes in producing hydrogen [13]. DME is easily mass-produced from coal, natural gas, and biomass; thus, it can be considered as a potential fuel for hydrogen production in fuel cells. A conventional DME production route for methanol dehydration catalyzed by HZSM-5 or γ-Al2O3 catalyst was proposed, providing a choice for the continuous supply of DME [14,15]. Meanwhile, numerous researchers have developed catalysts for DME/SR reactor. Gonzalez-Gil et al. [16] proposed a DME/SR using a VeNi/Al2O3 catalyst and demonstrated that DME conversion is possible in DME/SR at 500 °C with a hydrogen yielding ratio of 60%–70%. Chu et al. [17] designed a compact plate reactor for DME/SR with an innovative Cu/Ni/Fe/γ-Al2O3/Al mesh-type catalyst prepared using an anodic oxidation method. They proved that 66% of DME conversion was possible at 400 °C. In general, Cu/ZnO/γ-Al2O3 exhibits excellent catalytic performance in DME/SR [[18], [19], [20]]. Takeishi et al. [9] studied a synthesized Cu/ZnO catalyst with the γ-Al2O3 as carrier and found that Cu/ZnO worked well in DME/SR. With regard to reformer structure, Hayer et al. [21] reported that the microchannel reactors have become a new technology for methanol SR. In their study, a microchannel reactor coated with a Cu/CeO2/γ-Al2O3 catalysts was prepared by filling the etched channels of the stainless-steel platelets with aluminum suspension. A honeycomb structure was also used to develop the catalyst bed. Numerous research results have indicated that ceramic honeycomb can be utilized as the skeletal structure for supporting catalysts [[22], [23], [24], [25], [26]].

Hydrogen production rate through DME/SR can be varied depending on the DME:steam ratio, operating temperature, and reactor location. Therefore, a simulation model for hydrogen production is necessary to determine reactor size, catalyst, and operating conditions before producing hydrogen via DME/SR. Chiu et al. [27] developed a simulation model for hydrogen production via methanol SR and found that DME can be a possible candidate for hydrogen production due to its high energy density and superior transportability. Castedo et al. [28] also developed an accurate simulation model for DME/SR to explore reaction kinetics. However, there are no research results proving their numerical results via real experiment.

In the present study, a 3D simulation model for DME/SR is first constructed using commercial software (i.e., COMSOL 5.2). The effect of the geometric dimensionality of the computational domain on the hydrogen production through a DME/SR reactor is then determined depending on various operating conditions, such as reaction temperature (200 °C–500 °C) and reaction gas flow rate (DME: steam = 1:3). After the simulation, an actual experiment with an appropriate scale is performed to confirm the reaction conditions obtained from the simulation. A Cu–Zn/γ-Al2O3 catalyst is prepared using the sol-gel method, and a ceramic honeycomb is used as catalyst support to optimize catalyst performance. Lastly, the simulation and experiment results for the hydrogen production are compared.

This research aims to explore the characteristics of DME/SR for hydrogen production through computational fluid dynamics (CFD) simulation and actual experiments. As a first approach, the established DME/SR is evaluated, and the performances of the catalyst is investigated under different conditions. The present development of a reformer/PEMFC can be integrated into the system in the future. Good agreement between the simulation results and the experimental data is achieved. A series of integrated experimental devices is assembled to build a practical hydrogen production unit. Hydrogen yield efficiency can reach 90%, which is considerable improvement compared with literature.

Section snippets

Model boundary conditions and meshing scheme

In the 3D model simulation, the interface of the diluted porous medium is used to describe the catalyst bed inside the reactor; thus, the catalyst is defined as a porous solid phase (porosity is set to 0.3). Then, the steam carbon ratio (SCR) is set to steam: DME = 3:1 assuming that the gas phase has no slippage condition. Other assumptions for the CFD modeling are as follows.

  • 1.

    Various gases are considered ideal gases, indicating that gas properties can be calculated using ideal gas laws. The gas

Catalyst preparation

Copper alumina catalysts can be prepared using different methods. For example, Kim et al. [31] presented a DME/SR catalyst formed into a powder via an impregnation method. Vicente et al. [38] studied a catalyst for DME/SR with an atomic ratio of Cu:Zn:Al = 4.5:4.5:1. To obtain the optimal catalytic effect, Pashchenko et al. [39] introduced steam methane reforming to Ni-based catalysts with the shape of filled sphere. Reaction was simulated in 1D, 2D and 3D models. Porous ceramic honeycombs were

Characterization of Cu–Zn/γ-Al2O3 catalyst

The performance of Cu–Zn/γ-Al2O3 catalyst can be correlated with the activity surface area. To optimize the catalyst performance, the Cu–Zn/γ-Al2O3 catalyst is prepared using the sol-gel method as mentioned previously. Through this method, a uniform distribution of different metals is achieved and the excellent activity of the catalyst is predicted. In the experiment, ceramic honeycombs are used as support to improve the surface area. Through the sol-gel method and by coating on the ceramic

Conclusion

A 3D isothermal CFD model is developed in this study to investigate the reaction characteristics at different reaction temperatures. An appropriate scale experiment is conducted to validate the accuracy of the CFD model. Good agreement is achieved between the modelling and experiment data by comparing reaction properties at different temperatures. The sol-gel method is adopted to manufacture a high-efficiency catalyst, and ceramic honeycombs are used as catalyst carrier by coating the surface

Acknowledgement

This work was supported in part by the National Research Foundation of Korea (18R1D1A1B07044005 and 15R1A4A1041746).

References (40)

Cited by (17)

  • An autonomous fuel cell: Methanol and dimethyl ether steam reforming direct fed to fuel cell

    2023, International Journal of Hydrogen Energy
    Citation Excerpt :

    Xing et al. concluded that the best temperature region, S/C ratio, and pressure for Hydrogen-rich gas production would be at 200–300 °C, 1.6 to 2.0, and 1 atm, respectively [34]. DME conversion and hydrogen yield can also be related to the coupling effect of the catalyst preparation method, and the reaction temperature effect is positive within an appropriate range (200–400 °C) according to Zhang et al. [50]. The best-operating conditions for steam feed and temperature for the steam reforming reactor were defined based on the literature for methanol and DME steam reforming.

  • Kinetic modeling and reactor design of the direct synthesis of dimethyl ether for CO<inf>2</inf> valorization. A review

    2022, Fuel
    Citation Excerpt :

    DME is also adequate as H2 vector, through steam reforming using bifunctional catalysts [80]. Due to its characteristics, it can be used in proton exchange membrane fuel cells (PEMFC) [81] and solid oxide fuel cells (SOFC) [82], as well as for energy storage, through a cycle comprising DME synthesis from CO2 (exothermic) and DME reforming to H2 (endothermic), using energy intermittently generated from renewable sources [83]. The production of DME (10 Million tons/year, mostly in Asian countries) has been carried out using syngas feeds in a two-step process, that is, the synthesis of methanol and its dehydration towards DME [84].

  • Hydrogen generation characteristics of reduced DME steam reforming catalysts according to heating methods

    2022, International Journal of Hydrogen Energy
    Citation Excerpt :

    As the demand for carbon reduction increases, the demand for renewable energy and hydrogen energy increases, thereby increasing the importance of hydrogen energy. When combined with fuels, hydrogen is a highly promising energy carrier that can diversify energy supply and effectively store energy produced via renewable power generation, eliminating the spatial constraints of renewable energy [2,3]. In terms of carbon neutrality, hydrogen energy should be based on renewable energy; however, electrolysis-produced hydrogen is too expensive and scarce to satisfy future hydrogen demand.

  • Texture and acidity of amorphous silica-alumina regulated by the complex-decomposition method for steam reforming of dimethyl ether

    2022, Catalysis Today
    Citation Excerpt :

    Based on these reasons, the SRD reaction has been a subject of intense research in recent years [3]. At the beginning of the 21st century, the pioneering works of Sobyanin in the aspects of the thermodynamics analyses [4] and kinetics feasibility [5] are reported, and the subsequent reports on the thermodynamics [6–8] and the kinetics behaviors of different catalytic systems [9–12] lay the foundation of SRD. At the same time, several studies are reported on the development of catalysts [13–23], the reaction engineering including the types of reactors [24–27], and the technical feasibility and economy [3] for SRD.

View all citing articles on Scopus
View full text