High performance stainless-steel supported Pd membranes with a finger-like and gap structure and its application in NH3 decomposition membrane reactor

https://doi.org/10.1016/j.cej.2020.124245Get rights and content

Highlights

  • Supported Pd membranes with enhanced stability by a gap and finger-like structure.

  • Pd membranes show strong robustness under fast heating up/cooling down cycles.

  • Nearly complete ammonia conversion at a low temperature of 673 K and 3 bar (ΔP)

  • Excellent long-term stability test of NH3 decomposition Pd membrane reactor at 673 K.

Abstract

Pd composite membranes show great potential in several important fields, such as H2 production for fuel cell or pre-combustion applications. This study presents a facile and effective approach to develop high performance stainless-steel supported Pd composite membranes with a finger-like and gap structure by coating the finger-like porous stainless-steel support (PSS) with MnCO3 particles, which forms a small gap of ca. 1 µm during subsequent thermal treatment. The stability of this novel configuration was demonstrated under 20–50 fast heating up/cooling down cycles at a maximum ramp rate of 10 K/min, under near practical fuel cell application conditions. Such a structure imitates semi-free-standing bulk Pd membranes, which not only avoids the direct contact between Pd layer and PSS and ensures a high H2 permeance, but also avoids the shear stresses between the metal membrane and module. The membrane with a finger-like and gap structure was applied in NH3 decomposition membrane reactor, which achieved nearly complete NH3 conversion at a relatively low temperature of 673 K and remained stable for 200 h, exhibiting potential in future applications.

Introduction

Due to unique permeability to hydrogen and its isotopes, Pd-based membranes play an important role in ultrapure hydrogen generation in semiconductor industry as well as pure hydrogen production for fuel-cell application in both vehicular and stationary scenarios [1]. Moreover, simultaneous reaction and separation principle can potentially lower the temperature and improve the conversion of chemical reactions such as natural gas steam reforming and NH3 decomposition reaction [2], [3], [4].

The thermal/chemical stability remains the most critical challenge towards the commercial application of Pd-based composite membranes. Under fuel cell application conditions, the fast response requirements during startup/shutdown process further imposes a high demand on the stability of Pd composite membranes. In the literature studies, great efforts have been dedicated to the improvement of thermal stability in the current studies [5] including the formation of Pd alloys, such as Pd/Ag, Pd/Cu, Pd/Fe or Pd/Y alloy membranes [6], [7], [8], internal coating of Pd layer onto the porous support [9], anchorage of Pd layer to the support [10].

Thermal stresses may exist between Pd layer and membrane module as well as between Pd layer and porous substrate underneath, which need to be taken into account:

  • 1)

    The presence of constrains between the membrane tube and the module leads to significant mechanical (cyclic) stresses.

Demange et al. [11] observed significant elongation in a finger-type Pd membrane measured as 0.6% relative to the length of the membrane during nominal operation at 400 °C and 1 bar H2, which is ascribed to thermal expansion and H2 absorption. Such deformations and change in length are irreversible and remain after the experiment to some extent.

Tosti et al. [12] designed a flexible coupling between the Pd–Ag and the steel tube in order to compensate the differential thermal expansion between the membrane and the Pyrex shell module. In a later study [13], they proposed a finger-like configuration that imposes no constrains between the membrane tube and the module and permits free expansion and contraction during hydrogenation and thermal cycles.

  • 2)

    Shear stresses occur due to the differential elongation between the metal and the ceramic under thermal cycling and hydrogen loading, at the interface of Pd-based layer and the ceramic substrate [14].

Tong et al. [15] proposed a novel structure with a small gap between the Pd layer and the substrate layer by a combined organic-inorganic process, which makes the acting length almost equal to the membrane tube length, in other words, shifting the whole Pd membrane while enduring thermal cycles because force is put on the whole Pd layer rather than on the small Pd anchor in the pores [16]. Therefore, the shear stress for such a novel Pd-based supported membrane configuration is about two orders smaller than that for traditional Pd-based supported composite membrane.

Such a gap structure on porous ceramic substrate was derived by a combined organic-inorganic process which requires high temperature treatment in air at 873 K, thus not suitable for stainless-steel supported membranes. In this study, we proposed a facile and effective approach to derive a small interstice by modifying the support with MnCO3 particles that eventually decomposed to form MnOx [17] and thus leave a small gap between top metal layer and support during subsequent annealing treatment at 773 K. Here the finger-like configuration and gap design imitate the semi-free standing structure of bulk Pd tubes [13], which can also effectively avoid the direct contact between Pd layer and PSS. An excellent stability and high permeation performance was proved during fast and repeated heating up/cooling down cycles simulating practical fuel cell vehicle scenarios.

Liquid fuels, such as ammonia, methanol and ethanol, are regarded as ideal hydrogen carriers due to high hydrogen storage capacity (17.6 wt%, 12.5 wt% and 13.0 wt%, respectively), ease of hydrogen storage and transportation as well as safety, which is regarded as liquid sunshine in the prospect of large-scale deployment of solar energy at gigaton levels [18]. In particular, ammonia possesses a considerable hydrogen storage capacity (17.6 wt%) among these liquid carriers. Besides, the Pd membrane technology exhibits significant advantages in terms of compactness, low energy consumption and high recovery rate (>90%) compared to PSA or cryogenic separation (CS). This is very attractive as to ammonia decomposition due to applicability of Pd membrane separation in case of H2/N2 mixtures. Here, the stainless steel supported membranes were integrated within NH3 decomposition membrane reactor, achieving nearly complete conversion at a low temperature of 673 K that remained stable during the following 200 h long-term stability test, exhibiting a great potential in future applications.

Section snippets

Preparation of Pd composite membranes

The preparation procedure of Pd and Pd-Ag alloy composite membranes with a finger-like configuration and gap structure(Pd/MnOx/PSS, Pd-Ag/MnOx/PSS) consisted of four steps: (i) ceramic suspension was prepared containing MnCO3 powders (0.5 μm), polyvinyl alcohol (PVA), and polyethylene glycol (PEG), where PVA and PEG acted as dispersant and binder, respectively; (ii) the above suspension was then ultrasonically dispersed and stirred to form a mixed suspension and the mixed suspension was coated

Performance of Pd composite membranes

Due to the difficulty to obtain a sample of Pd/MnOx/PSS for SEM, the sample of Pd/MnOx/Al2O3 composite membranes was an alternative. The SEM image of Pd/MnOx/Al2O3 composite membrane is depicted in Fig. 2b, which exhibits a small interstice of ca. 1 µm between Pd layer and Al2O3 substrate from SEM analysis. Meanwhile, it is the photo taken by metalloscope to show an interstice between Pd layer and PSS (Fig. 2c). The semi-free-standing design with a porous and closed end is illustrated in Fig. 3

Conclusions

Thermal stability remains as one of the critical challenges towards the commercial application of Pd-based composite membranes, and there exist thermal stresses between Pd layer and membrane module as well as between Pd layer and porous substrate underneath. This study presents a facile and effective approach to develop stainless-steel supported Pd and Pd90Ag10 composite membranes with a finger-like configuration and gap structure which minimizes the thermal stresses mentioned above and the

Acknowledgements

We are grateful for the financial support from the 100-Talent Project of CAS, National Natural Science Foundation of China (Grant No. 21676265; 51501177; 21306183), The Ministry of Science and Technology (MOST) of the People's Republic of China (Grant No. 2016YFE0118300), and the K. C. Wong Education Foundation (GJTD-2018-06).

References (22)

  • S. Tosti et al.

    Long-term tests of Pd–Ag thin wall permeator tube

    J. Membr. Sci.

    (2006)
  • Cited by (43)

    • Carbon-free green hydrogen production process with induction heating-based ammonia decomposition reactor

      2023, Chemical Engineering Journal
      Citation Excerpt :

      In addition to the storage and transportation of hydrogen in the form of ammonia, the decomposition process must also be considered to release the desired hydrogen. Accordingly, significant research has focused on catalyst and process development for the decomposition of ammonia to yield hydrogen [20–23]. More specifically, numerous metals, alloys, and noble metal compounds have been studied [1,24–28], wherein Ru catalysts have received particular attention among the metal-based catalysts such as Fe, Ni, Pt, Ru, Ir, Pd, and Rh.

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text