Elsevier

Ceramics International

Volume 45, Issue 2, Part A, 1 February 2019, Pages 2219-2225
Ceramics International

Room- and high-temperature torsional shear strength of solid oxide fuel/electrolysis cell sealing material

https://doi.org/10.1016/j.ceramint.2018.10.134Get rights and content

Abstract

The structural integrity of the sealant material is critical for the reliability of solid oxide fuel/electrolysis stacks. In the current study, a torsion test is implemented to evaluate and compare its shear strength with a partially crystallized glass sealant at room- and operation relevant high-temperatures. Hourglass-shaped specimens with different configurations of hollow- and full-halves are utilized for testing. The fracture surfaces are visualized via optical microscopy and complementary scanning electron microscopy. In addition, cyclic loading is used to investigate potential subcritical crack growth effects in the sealants. Both, the specimens with a hollow-half as well as the ones with two full-halve steel plates yield almost the same nominal shear strengths. The shear fracture stresses decrease with rising temperature, while the fracture mode changes from brittle at room temperature and 600 °C to ductile at 800 °C. The cyclic loading condition indicates subcritical crack growth in the sealant at 600 °C and creep associated damage at 800 °C.

Introduction

Solid oxide fuel/electrolysis cells (SOFCs/SOECs) are electrochemical devices that convert chemical energy of fuels into electricity and vice versa by promoting a redox reaction across a solid electrolyte [1]. In order to achieve higher production capacity, individual cells are connected together in series to form a stack. Among the different designs of SOFC/SOEC stacks, the planar types with metallic interconnect have received increasing attention because of their higher power density, lower cost, and relative ease of fabrication [2]. The intermediate-temperature SOFCs/SOECs, which are currently undergoing commercialization, operate in a temperature range of 650–850 °C [3]. The stack components should be stable and capable of a service life of more than 40,000 h and hundreds of thermal cycles for stationary systems, or thousands of thermal cycles for mobile systems [4].

The dissimilar component materials joined in the stack, depending on their coefficients of thermal expansion (CTE), show different strains with temperature changes [5]. The thermochemical environmental conditions may also have different effects on bulk dimension of stack materials [1]. As a consequence of these thermal and chemical strains, normal as well as shear stresses arise even for a single component in case of a temperature gradient or for the joined components during thermal cycling or steady state operation [1], [5]. These internally generated stresses, as well as the stack clamping loads [6], can impair components of even stationary stacks, which are generally exposed to less demanding thermal boundary conditions than in the case of mobile applications [7]. Therefore, mechanical issues of any stack component may have a serious impact on performance and degradation rate.

Sealing has been identified as a critical issue for commercializing the entire planar SOFCs/SOECs technology [8]. Sealants are needed to join components and hermetically separate fuel and oxidant [9]. Glass-ceramics are most widely used as sealant materials due to the possibility of tailoring their composition for stable performance in the vicinity of operation associated gases, contacting ceramic cells and metallic interconnects/frames [10]. One of the major challenges is to evaluate the candidate sealing materials regarding its mechanical robustness. It has to be considered that various stress conditions existing in the stack [11], hence, a comprehensive mechanical assessment of seal is necessary. Although extensive work has been dedicated to evaluate the sealants under tensile- and/or bending-dominant loading condition [12], [13], [14], [15], [16], [17], [18], only a limited number of studies exist on shear strength evaluation [18], [19], [20], [21].

In the current study, the shear behavior of a rapid crystalizing glass (sealant G) is investigated and compared with a slowly crystalizing glass (sealant H-F) at room- and high-temperatures. The implemented method on the hourglass-shaped specimens is the torsional shear test, which has recently received growing attention [22], [23], [24], [25] because of its pure shear loading, relatively low stress concentration, and easy alignment [26]. The fracture surfaces after room- and high-temperature tests are examined to improve the understanding of the fracture behavior under shear stress loading. Addition cyclic loading tests are used to investigate potential subcritical crack growth and its effect on final failure.

Section snippets

Experimental

The used glasses termed “H” and “G” are based on the BaO-CaO-SiO2 ternary system modified with the minor additions, as shown in Table 1. The raw materials were obtained from Merck KGaA Darmstadt with a grade of purity higher than 99%. Each batch was prepared by mixing an appropriate mole fraction of oxide ingredients and melting at 1480 °C in a platinum crucible in an induction furnace [27]. For a better homogenization of the glass, the melting procedure was carried out twice. The prepared

Results and discussion

Fig. 2 shows the average shear fracture stress of specimens with different configurations obtained on the basis of Eq. (1) and previous finite simulation results [24], which are termed in the following nominal and simulated shear strengths, respectively. As it can be seen, the full-full and hollow-full specimens revealed almost the same nominal shear strengths, being slightly higher than that of hollow-hollow specimens. The simulation results revealed that perforation of either one half or both

Conclusions

The shear strengths of two sealants, one partially and one fully crystallized, were evaluated by a torsion test applied on hourglass-shaped specimens at room- and high-temperatures. The specimens with one hollow half as well as the ones with two full halves revealed almost the same nominal shear fracture stresses but higher values than that of hollow-hollow specimens. The fracture surfaces confirmed brittle fracture of specimens at room temperature and 600 °C, which turned into ductile mode at

Acknowledgments

The authors wish to thank Mr. D. Federmann and Ms. T. Osipova for the support in specimens’ preparations and testing, Dr. E. Wessel and Dr. D. Grüner for SEM investigations, and Prof. L. Singheiser and Prof. M. Krüger for hosting at Forschungszentrum Jülich. M. Fakouri Hasanabadi, Prof. A.H. Kokabi and Dr. M.A. Faghihi-Sani express their gratitude to Ministry of Science, Research and Technology of Iran and also the research board of Sharif University of Technology for financial support and

References (46)

  • C.K. Lin et al.

    Creep rupture of the joint between a glass-ceramic sealant and lanthanum strontium manganite-coated ferritic stainless steel interconnect for solid oxide fuel cells

    J. Eur. Ceram. Soc.

    (2018)
  • E.V. Stephens et al.

    Experimental characterization of glass – ceramic seal properties and their constitutive implementation in solid oxide fuel cell stack models

    J. Power Sources

    (2009)
  • T. Osipova et al.

    Room and elevated temperature shear strength of sealants for solid oxide fuel cells

    Ceram. Int.

    (2016)
  • M. Fakouri Hasanabadi et al.

    The analysis of torsional shear strength test of sealants for solid oxide fuel cells

    Ceram. Int.

    (2017)
  • L. Blum et al.

    Investigation of solid oxide fuel cell sealing behavior under stack relevant conditions at Forschungszentrum Jülich

    J. Power Sources

    (2011)
  • N.H. Menzler et al.

    Post-test characterization of a solid oxide fuel cell stack operated for more than 30,000 hours: the cell

    J. Power Sources

    (2018)
  • M. Fakouri Hasanabadi et al.

    Interactions near the triple-phase boundaries metal/glass/air in planar solid oxide fuel cells

    Int. J. Hydrog. Energy

    (2017)
  • J.W. Hutchinson et al.

    Mixed Mode Cracking in Layered Materials

    Adv. Appl. Mech.

    (1991)
  • C.-K. Lin et al.

    Thermal stress analysis of a planar SOFC stack

    J. Power Sources

    (2007)
  • N.P. Brandon et al.

    Solid Oxide Fuel Cell Lifetime and Reliability

    (2017)
  • L. Blum et al.

    Anode-supported solid oxide fuel cell achieves 70 000 hours of continuous operation

    Energy Technol.

    (2016)
  • Z. Gao et al.

    A perspective on low-temperature solid oxide fuel cells

    Energy Environ. Sci.

    (2016)
  • Q. Zhu et al.

    Stable glass seals for intermediate temperature (IT) SOFC applications

  • Cited by (12)

    • Effect of YSZ sol-gel coating on interaction of Crofer22 APU with sealing glass for solid oxide fuel/electrolysis cell

      2020, Journal of Alloys and Compounds
      Citation Excerpt :

      Merits including low environmental pollution, flexible fuels and high efficiency make SOFC a promising power generation technology for the future [1,2]. However, due to relatively high operating temperatures (650–850 °C), the reliability of SOFC stack components is a serious challenge [3–5]. Sealing materials have been received significant attention in recent research and development activities [6,7].

    • Torsional shear strength behavior of advanced glass-ceramic sealants for SOFC/SOEC applications

      2020, Journal of the European Ceramic Society
      Citation Excerpt :

      The crosshead speed was 0.5 mm min −1 with an estimated rotation speed of 0.010 rad min−1 [25]. In the torsion tests at FZJ the specimens were twisted with a speed of ∼ 4° min−1 until fracture occurred [27]. A round-robin test with the two torsion machines has been done using epoxy adhesive bonded steel hourglasses (THG-5) prior to this work, to test the comparability of the obtained results.

    • Finite element optimization of sample geometry for measuring the torsional shear strength of glass/metal joints

      2020, Ceramics International
      Citation Excerpt :

      With increasing torque one of the cracks reaches the critical stress intensity for propagation along the interfaces and leads then to catastrophic failure (Fig. 4b). During torsion tests, the brittle materials tend to break along planes perpendicular to the direction in which tension possesses its maximum value, i.e., along fracture surfaces at an angle of ~45° to the z axis [20]. It is worth noting that this angle can be affected by the elastic moduli mismatch of sealant and steel [44].

    View all citing articles on Scopus
    View full text