Liquid foam assisted sol–gel synthesis of iron oxides for hydrogen storage via chemical looping

https://doi.org/10.1016/j.ijhydene.2016.07.019Get rights and content

Highlights

  • Oxygen carrier is prepared by liquid foam assisted sol–gel method.

  • The pore configuration is tuned by adding surfactant SDS in the precursor sols.

  • Material with 2D pores can better resist the particle sintering up to 900 °C.

Abstract

Chemical looping of iron oxides is capable of storing hydrogen with high capacity and low material costs. In this paper, we prepared iron oxides materials from the liquid foam templated sol–gel precursors and demonstrated its application in the chemical looping hydrogen storage process. By tuning the precursor chemistry, we obtained a series of materials with different pore configuration, and found this approach a convenient access to define the material configuration. With assistance of SDS, materials showed promoted porosity and better reactivity compared with the sample without SDS. In particular, samples with isolated pore structure were better able to limit the sintering. By characterizing the materials before and after cycles, we found the confinement effects of pore wall was capable of restraining the growth of the crystalline size, and thus mitigated the material deactivation in the high temperature redox cycles.

Introduction

The stochastic nature of the wind energy input makes the large-scale energy storage technology an inevitable requirement. Using H2 as a storage medium has been proposed as a promising method to address this problem, in which the electricity is fed in an electrolyzer, and then stored and carried as hydrogen. Currently, hydrogen can be stored physically in liquid or high-pressure state or chemically absorbed by a series of materials including hydrocarbons, metal alloys, metal hydrides as well as the currently active-developing carbon nanotubes and the metal organic frameworks (MOFs) [1], [2], [3], [4], [5], [6], [7], [8]. Unfortunately, although significant progress has made on improving these materials, producing a high performance hydrogen storage material that combines good safety, high capacity with low manufacturing cost is still challenging.

An alternative technology capable of storing hydrogen is the chemical looping of iron oxides [9], [10], [11], [12], [13], [14]. In the first step, the pseudo-storage of hydrogen is realized by the reduction of the iron oxides oxygen carrier (OC). To maximize storage capacity (theoretically 4.8 wt.%), metallic iron is generally used as the nominal storage material.

Hydrogen charge:Fe3O4 + 4H2 ↔ 3Fe + 4H2O

Hydrogen discharge:3Fe + 4H2O ↔ Fe3O4 + 4H2

By recycling the water produced in the first step, the OC is then regenerated to its original state and simultaneously releases the stored hydrogen. In this process, one could note that the feedstock is only H2O, Fe3O4, which is readily available, environmentally friendly and very importantly, with low costs. Therefore, this technology may be promising in the application of large-scale hydrogen storage system, such as solar or wind plant where the hydrogen can be inexpensively produced from water [1], [15], [16], [17], [18].

However, there remain many challenges to successfully use this method to store hydrogen. One of the key challenges is that when using pure iron oxide, the process suffers from severe sintering of particles, caused by the high temperature cyclic operation of reduction and oxidation steps. This implies that the practical hydrogen storage capacity will decay along with time. In the material science, sintering is susceptible to a number of influencing factors, including operating temperature, reaction atmosphere, promoters, material porosity, support species, etc [19], [20], [21]. Since this method was first proposed by Otsuka et al. [11], extensive studies have been performed on adjusting these parameters to limit the sintering. The mixing of some promoters, such as Cr, Ce, Rh, has been demonstrated efficient in lowering the reaction temperature, and thereby mitigates the sintering, mainly by Otsuka’ group and Kim's group [9], [11], [12], [13]. In some extreme cases, the operating temperature can be less than 300 °C, under which the sintering was negligibly slow but the kinetics of chemical looping is significantly limited. Dispersing the active component on refractory supports is an alternative way to limit the sintering with the material reduced and oxidized in high temperature. Currently, a number of supported OC has been prepared using various supports, including Al2O3, CeO2, ZrO2, TiO2, yttria-stabilized zirconia (YSZ), MgAl2O4, etc, with different preparation methods, such as mechanical mixing, impregnation, co-precipitation, sol–gel, etc; the effects of support species, interaction of the support and iron oxides, dispersion of the cyclic phase, support loadings on the chemical looping performance are extensively studied in these works [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]. However, it is noticeable that changing the preparation conditions, e.g. support species, loadings, will inevitably lead to different pore configuration of the material, which was also reported as influential to the particle sintering. To be best of our knowledge, no such work concerning the effects of material porosity on chemical looping performance has been found to date.

Conventionally, the material porosity could be characterized as 2D or 3D structure configuration depending on the interconnectivity of the pores. Within the 2D matrix, the pores are one-dimensional and confined to the support surface. In such porosity, we can imagine that the active component within a given pore can only interact with other particles in the same cavity, and has no opportunity to incorporate those in the neighboring pores. In contrast, the 3D pores possess the connecting channels between different pores, so the particles can circumvent the pore walls to interact with the components in the neighboring pores. At present, the effects of material porosity on the sintering have been carried out on the supported Au catalysts with a result showing that the sintering of Au particles is significantly affected by material pore size, pore wall thickness, pore connectivity, and pore strength. Datye and coworkers [33] found gold particles inside 3D pores with interconnecting channels sintered into larger ones more readily than did those inside 2D pores. At the same time, they observed smaller pores produced a smaller final particle size of the sintered nanoparticles than larger pores.

Apart from the impact on the sintering, it is well-known that the material porosity is important in the kinetics of the gas–solid reactions. For instance, the larger surface density of active sites can be generally achieved on the high specific surface of the micropores or mesopores, whilst macropores dramatically reduce the transport limitations and facilitate the mass flow. In addition, for a higher performance, the material should combine the macropores with mesopores, i.e. hierarchically organized and not just segregated pores [34], [35].

The aim of this work is to investigate the effects of the material porosity on the OC performance in chemical looping hydrogen storage cycles. To obtain OC particles with different pore structure, we modified the conventional sol precursor with the assistance of liquid foam template. To figure out the precursor-structure-property relationship, we synthesized a series of OC particles by varying the surfactant SDS (sodium dodecyl sulfonate) amount and characterized the metal oxides with various physical and chemical techniques, and then we carried out the hydrogen charge and multicycles experiments to study the kinetics and the sintering of the prepared materials.

Section snippets

OC preparation

The basic Fe–Al aqueous sol was prepared by dissolving Fe(NO3)3.9H2O, Al(NO3)3.9H2O, citric acid (CA), polyethylene glycol (PEG, Mw = 400) in the deionized water, followed by being heated in the oil bath at 60 °C with magnetic stir. The molar ratio of total metal irons, CA and PEG was set at 1:1:4 for the completely chelating and polymerizing the precursor during the sol–gel process. The loading of Fe2O3 in the final OC was designed at 75% in mass basis, according to the results obtained by

Effect of SDS amount on the porous characteristics of OC

Due to the fact that SDS has the ability to define the bubble size distribution (Fig. 1), the microstructure of the final material can be efficiently controlled, that is, varying the micelles concentration in the precursor sols to realize the desired large and small size pore configuration. The feasibility of this approach is demonstrated by the results shown in Table 1. Clearly, when SDS added, the porosity was dramatically promoted with a maximum value of 87.3%, in which there was a 30%

Mechanism of sintering

In the case of our proposed process, two sintering mechanism, crystallite migration and atomic migration, were possibly involved. Crystallite migration involves the migration, collision and coalescence of entire crystallites within the support network. On the other hand, atomic migration, referred to as Ostwald ripening, involves detachment of metal atoms or metal clusters from crystallites, and then transportation from small particles to larger particles. Thermodynamically, Ostwald ripening

Conclusions

We reported the synthesis of iron oxides using liquid foam assisted sol–gel process and demonstrated its application in the chemical looping hydrogen storage process. By investigating the precursor–structure relationship, we found this method a convenient access to define the material pore configuration, e.g. the 2D or 3D pores can be obtained by just tuning the SDS concentration in the precursor sols. Among the five prepared samples, SDS-10, 20 showed better activities due to the promotions in

Acknowledgements

The authors gratefully acknowledge the National Natural Science Foundation of China (Grant No. 51476035), National Science Foundation for Distinguished Young Scholars of China (Grant No. 51525601) and the Scientific Research Foundation of Graduate School of Southeast University (Grant No. YBJJ1609) for financial support of this project.

References (46)

  • M. Korpaas et al.

    Operation and sizing of energy storage for wind power plants in a market system

    Int J Electr Power Energy Syst

    (2003)
  • M. Thaler et al.

    Investigations of cycle behaviour of the contact mass in the RESC process for hydrogen production

    Int J Hydrogen Energy

    (2006)
  • R. Siriwardane et al.

    Production of pure hydrogen and synthesis gas with Cu–Fe oxygen carriers using combined processes of chemical looping combustion and methane decomposition/reforming

    Int J Hydrogen Energy

    (2015)
  • X. Zhu et al.

    Ce–Fe oxygen carriers for chemical-looping steam methane reforming

    Int J Hydrogen Energy

    (2013)
  • B. Wang et al.

    Characterization and evaluation of Fe2O3/Al2O3 oxygen carrier prepared by sol–gel combustion synthesis

    J Anal Appl Pyrolysis

    (2011)
  • K. Li et al.

    Modification of CeO2 on the redox property of Fe2O3

    Mater Lett

    (2013)
  • J. Adanez et al.

    Progress in chemical-looping combustion and reforming technologies

    Prog Energy Combust Sci

    (2012)
  • F. Kosaka et al.

    Iron oxide redox reaction with oxide ion conducting supports for hydrogen production and storage systems

    Chem Eng Sci

    (2015)
  • Z. Xu et al.

    Self-assembly template combustion synthesis of a core–shell CuO@TiO2–Al2O3 hierarchical structure as an oxygen carrier for the chemical-looping processes

    Combust Flame

    (2015)
  • D.-W. Zeng et al.

    Bio-oil heavy fraction for hydrogen production by iron-based oxygen carrier redox cycle

    Fuel Process Technol

    (2015)
  • R. Xiao et al.

    Use of heavy fraction of bio-oil as fuel for hydrogen production in iron-based chemical looping process

    Int J Hydrogen Energy

    (2014)
  • L. Martins et al.

    Preparation of hierarchically structured porous aluminas by a dual soft template method

    Microporous Mesoporous Mater

    (2010)
  • R. Pellegrini et al.

    0.5wt.% Pd/C catalyst for purification of terephthalic acid: irreversible deactivation in industrial plants

    J Catal

    (2011)
  • Cited by (0)

    View full text