Characterization and analysis methods for the examination of the heterogeneous solid oxide fuel cell electrode microstructure: Part 2. Quantitative measurement of the microstructure and contributions to transport losses

Advanced characterization and analysis of multifunctional materials, such as the materials found in heterogeneous solid oxide fuel cell (SOFC) electrode architectures, can help to provide a qualitative and quantitative understanding of how these structures respond to different manufacturing and operating practices. Dense, opaque materials, which have large X-ray mass absorption coefficients and features on sub-micrometer length scales, can make characterization difficult. Advances in tomographic X-ray imaging can permit this level of detailed characterization, and complement stereographic scanning electron microscope measurements that have also been reported. In this second part of a two-part study, details regarding quantitative characterization methods that have been used to examine the SOFC anode microstructure are reported. The detailed formulation and validation of a phase size distributions for the three constitutive phases, as well as resistive loss microstructure-induced resistive loss distributions in the nickel (Ni) and yttria-stabilized zirconia (YSZ) phases are provided in this section.     

Stress analysis of solid oxide fuel cell anode microstructure reconstructed from focused ion beam tomography

The degradation and ultimately lifetime of solid oxide fuel cells (SOFCs) is determined in part by the stresses generated within the different layers of the device. For fully dense materials such as the electrolyte, when modelling these stresses on a macro-scale the material properties can be considered to be homogeneous (evenly distributed) allowing the prediction of volume average stresses due to differential thermal expansion in the layer. However, detailed stress analysis of real, multiphase porous layers such as those found in SOFC electrodes, on the micron and sub-micron scale has not been possible to date as detailed geometry and convenient methods to generate a finite element model have not been available.In this paper we present work that combines microstructural characterisation of a porous solid oxide fuel cell anode with three dimensional stress analysis to inspect the stresses within the individual phases of the anode, and at phase boundaries. The electrode microstructure has been characterised using focused ion beam (FIB) tomography and the resulting microstructure used to generate a solid mesh of three dimensional tetrahedral elements. A temperature field was applied to simulate the heating of the sample from room temperature (298 K) to operating temperature (1073 K). The maximum principal stress in the nickel phase was found to exceed the yield strength, while the minimum principal stress in the yttria-stabilized zirconia (YSZ) phase was found to exceed the characteristic strength of that volume of YSZ, indicating that the probability of failure of the YSZ matrix is significant.     

Quantitative analysis of micro structural and conductivity evolution of Ni-YSZ anodes during thermal cycling based on nano-computed tomography

Understanding the mechanism of degradation in solid oxide fuel cells (SOFCs) using nickel/yttria-stabilized zirconia (Ni-YSZ) as the anode material is very important for the optimization of cell performance. In this work, the effects of thermal cycling on the microstructure of the Ni-YSZ anode are explored using the three-dimensional X-ray nano computed tomography (nano-CT) imaging technique. It is found that the average Ni particle size increased with thermal cycling, which is associated with the decreased connectivity of the Ni phase and the three-phase-boundary (TPB) length. Moreover, the conductivities of the anode samples are also reduced with the increase in thermal cycle times. The implication of these observations is discussed in terms of the relationship between the conductivity and connectivity of the Ni phase.     

Effect of nickel-phosphorus interactions on structural integrity of anode-supported solid oxide fuel cells

An integrated experimental/modeling approach was utilized to assess the structural integrity of Ni-yttria-stabilized zirconia (YSZ) porous anode supports during the solid oxide fuel cell (SOFC) operation on coal gas containing trace amounts of phosphorus impurities. Phosphorus was chosen as a typical impurity exhibiting strong interactions with the nickel followed by second phase formation. Tests were performed using Ni-YSZ anode-supported button cells exposed to 0.5-10 ppm of phosphine in synthetic coal gas at 700-800 °C. The extent of Ni-P interactions was determined by a post-test scanning electron microscopy (SEM) analysis. Severe damage to the anode support due to nickel phosphide phase formation and extensive crystal coalescence was revealed, resulting in electric percolation loss. The subsequent finite element stress analyses were conducted using the actual anode support microstructures to assist in degradation mechanism explanation. Volume expansion induced by the Ni phase alteration was found to produce high stress levels such that local failure of the Ni-YSZ anode became possible under the operating conditions.     

Simulation of coarsening in three-phase solid oxide fuel cell anodes

Coarsening of the nickel phase is known to occur in solid oxide fuel cell (SOFC) anodes consisting of Ni and yttria-stabilized zirconia (YSZ). However, the exact nature of the coarsening process is not known, nor how it affects three-phase boundaries (TPBs) and the resulting electrochemical performance. We apply a phase-field approach to simulate the microstructural evolution of Ni-YSZ anode functional layers. An experimentally obtained three-dimensional reconstruction of a functional layer from an anode-supported SOFC is used as the initial microstructure. The evolution of the microstructure is characterized quantitatively by examining the TPB density, interfacial area per unit volume, and tortuosity versus time. The assumed TPB contact angles are found to have a strong effect on the microstructural evolution; in particular, reducing the contact angle of nickel on YSZ yields less TPB reduction.     

Functionally-graded composite cathodes for durable and high performance solid oxide fuel cells

A double-layer dual-composite cathode is fabricated and has an ideal cathode microstructure with large electrochemical active sites and enhanced the durability in solid oxide fuel cells (SOFCs). The insertion of a yttria-stabilized zirconia (YSZ)-rich functional layer between the electrolyte and the electrode allows for a graded transition of the YSZ phase, which enhances ionic percolation and minimizes the thermal expansion coefficient mismatch. Electrochemical measurements reveal that the double-layer composite cathode exhibits improved cathodic performance and long-term stability compared with a single-layer composite cathode. A cell with a well-controlled cathode maintains nearly constant interfacial polarization resistance during an 80 h accelerated lifetime test.     

Thermoelastic analysis of friction pairs in hydro-viscous drive combined thermal and mechanical loads under soft startup condition

Thermo-mechanical characteristics significantly impact the operating performance of the friction pairs in hydro-viscous drive. A three-dimensional transient thermo-mechanical coupling model was established for investigation of the distributions and coupling relationships of the temperature, stress, and displacement generated by the combination of the thermal and mechanical loads under soft startup condition. Then, the time-dependent contact pressure and velocity were extensively examined, as well as their effects on the thermo-mechanical coupling characteristics. Finally, the effects of the material and structural parameters were analyzed by the one-factor-at-a-time method. The observations confirmed plastic deformation of the steel disc will occur when the steel disc and friction disc are, respectively, restrained at the outer and inner diameters. However, there is a low likelihood of plastic deformation of the friction lining occurring. The thermal load and constraint locations were also found to crucially affect the stress and displacement distributions. Moreover, the circumferential stress was determined to be the most important stress component, and can thus be used as a basis for judging whether plastic deformation will occur. The results provide theoretical reference for predicting the thermal characteristics of the friction pairs, as well as a foundation for selecting the material and structural parameters.     

Numerical simulation and analysis of thermal stress distributions for a planar solid oxide fuel cell stack with external manifold structure

A three-dimensional numerical model based on the finite element method (FEM) is constructed to calculate the thermal stress distribution in a planar solid oxide fuel cell (SOFC) stack with external manifold structure. The stack is composed of 5 units which include cell, metallic interconnect, seal and anode/cathode current collectors. The temperature profile is described according to measured temperature points in the stack. It can be clearly seen that the maximum stress concentration area appears at the corner of the components when the stack is heated from room temperature (RT) to 780 °C. The effects of stack components on maximum stress concentration have been investigated under the operation temperature, as well as the thermal stress simulation results. It is obvious that the coefficient of thermal expansion (CTE) mismatch between the interconnect and the seal plays an important role in determining the thermal stress distribution in the stack. However, different compressive loads have almost no effect on stress distribution, and the influence of glass-based seal depends on the elastic modulus. The simulation results can be applied for optimizing the structural design of the stack and minimizing the high stress concentration in components.     

Durability of Ni anodes during reoxidation cycles

Anodes manufactured from NiO- and yttria-stabilized zirconia (Y₂O₃ doped ZrO₂, YSZ) powders are today's state of the art for solid oxide fuel cells (SOFCs) because they are easy to manufacture and have high performance in both anode-supported and electrolyte-supported cells. However, such cells can show significant degradation or fail completely if nickel is reoxidized during high-temperature operation even though it can be reduced again. Tests with stacks and systems have shown that system shutdown procedures, accidental air entry due to component failure or controlled air feed to the anode side as a result of operational necessities may occur and result in the reoxidation of the metallic nickel. This reoxidation is not only associated with a volume expansion, but also with significant structural changes in the anode microstructure, generating stresses in the anode itself, as well as in the electrolyte. These stresses can exceed the stability of the components, potentially promoting crack growth, which leads to degradation of the SOFC or complete failure.This problem has been addressed by a number of contributions in the literature over the last decade, but interest is increasing, particularly because SOFC systems are being discussed for transport and mobile applications requiring new system specifications. The most critical problem to be overcome is the tolerance of a large number of intentional redox cycles due to system requirements during operating lifetime.This article gives an overview of the various approaches to the redox problem by summarizing many of the contributions, starting with a basic understanding of the underlying physicochemical processes of Ni reduction and oxidation and ending at stack-level results, leading finally to their combination with recent findings. It aims to elaborate reliable results and open questions on this topic considering the mechanical and electrochemical aspects of the problem.     

Structural and electrical properties of Ni-YSZ cermet materials

Ceramic-metal composites (cermets) containing yttria-stabilized zirconia, YSZ, and Ni particles are commonly used as anode materials in solid oxide fuel cells. The long-term performance of fuel cells is strictly related to both the structural and electrical properties of anode materials. In order to achieve high mixed electrical conductivity and high activity of electrochemical reactions and hydrocarbon fuel reforming, it is necessary to select an appropriate chemical composition and a suitable method of preparation. Materials containing 8 mol% yttria-stabilized zirconia and Ni were prepared by means of two methods: co-precipitation and impregnation. The structure of the materials was characterized by means of X-ray diffraction (XRD), scanning electron microscopy (SEM) and porosity studies. The thermal expansion coefficient (TEC) was determined using the dilathometric method. Electrochemical impedance spectroscopy (EIS) and the Wagner polarization method were used to determine electrical conductivity and the electron transference numbers, respectively.     

Analysis of the three-dimensional microstructure of a solid-oxide fuel cell anode using nano X-ray tomography

The visualization of three-dimensional (3D) microstructures of solid oxide fuel cells helps to understand the efficiency of the electrochemical conversion process, study the device's reliability, and improve manufacturing processes. Here, we used X-ray nanotomography to investigate a porous nickel-yttria-stabilized zirconia (Ni-YSZ) composite anode. These results were used to characterize and quantify the key structural parameters, such as the volume ratio of the three phases (Ni, YSZ, and pore), connected porosity, surface area of each phase, interface of Ni/YSZ, volume-specific three-phase boundary length (TPB where the Ni, YSZ and fuel gas phases come together), and electrical conductivity of the anode.     

Thermal stresses in an operating micro-tubular solid oxide fuel cell

A multi-physics model is developed to investigate the thermal stresses in a micro-tubular SOFC, based on a previously developed thermal-fluids model predicting cell operation. Mechanical properties of the anode and cathode are determined theoretically through composite structure approximation. Residual stresses arisen during the fabrication of the cell due to the mismatch in thermal expansion coefficients are calculated by accounting for each fabrication process separately. The interactions between the cell, the sealant and the alumina tube are accounted for a better representation of the actual fuel cell test setup. The effect of sealant and alumina tube on the stress distribution in the cell is investigated and it is found out that near the fuel cell-sealant interface stress distribution changes significantly. The effect of spatial temperature gradient on the stress distribution is also analyzed and found to have a minimal impact for a typical fuel cell operation at mid-range current densities. The effects of oxygen vacancies caused by the reduction of the GDC electrolyte on the overall stress distribution are also shown. Oxygen vacancies of the electrolyte result in relaxation of the stresses due to the alleviation of mismatch in Young's modulus between different layers of the cell.     

A novel design and performance of cone-shaped tubular anode-supported segmented-in-series solid oxide fuel cell stack

A novel design of cone-shaped tubular segmented-in-series solid oxide fuel cell (SOFC) stack is presented in this paper. The cone-shaped tubular anode substrates are fabricated by slip casting technique and the yttria-stabilized zirconia (YSZ) electrolyte films are deposited onto the anode tubes by dip coating method. After sintering at 1400 °C for 4 h, a dense and crack-free YSZ film with a thickness of about 7 μm is successfully obtained. The single cell, NiO-YSZ/YSZ (7 μm)/LSM-YSZ, provides a maximum power density of 1.78 W cm⁻² at 800 °C, using moist hydrogen (75 ml min⁻¹) as fuel and ambient air as oxidant.A two-cell-stack based on the above-mentioned cone-shaped tubular anode-supported SOFC is fabricated. Its typical operating characteristics are investigated, particularly with respect to the thermal cycling test. The results show that the two-cell-stack has good thermo-mechanical properties and that the developed segmented-in-series SOFC stack is highly promising for portable applications.     

RedOx study of anode-supported solid oxide fuel cell

The common technology for solid oxide fuel cells (SOFC) is based on a cermet (ceramic-metal composite) anode of nickel with yttria stabilized zirconia (YSZ), often used as the supporting structure. One of the main limitations of this technology is the tolerance of the anode towards reduction and oxidation (“RedOx”) cycles.In this study, two techniques are used to quantify the anode expansion after a RedOx cycle of the nickel at different temperatures. The first method considers the anode expansion above the electrolyte fracture limit by measuring the crack width in the electrolyte layer. In the second method, the anode porosity is measured using scanning electron microscopy (SEM) image quantification. The same measurement techniques are used to quantify anode expansion after consecutive RedOx cycles at constant temperature.The quantification technique is then applied to cells tested in real stack conditions. The cell corners can undergo several RedOx cycles depending on stack design and fuel utilization. The study of such zones allows estimating the number of cycles that the anode experienced locally.     

3D thermo-electro-chemo-mechanical coupled modeling of solid oxide fuel cell with double-sided cathodes

A solid oxide fuel cell based on double-sided cathodes is developed in our group, showing special properties and many advantages under some harsh conditions. To optimize the cell further, a thermo-electro-chemo-mechanical coupled 3D model is developed to simulate the distributions of temperature, current density, fuel gas and thermal stress under different voltages. The numerical results indicate that the temperature distribution, current, fuel gases and thermal stress is non-uniform in the cell at different voltages. The distribution of thermal stress in the electrolyte is also non-uniform because of the un-even electrochemical reaction and convective heat transfer. Furthermore, the result shows that about 47%~54% of maximum 1st principal stress in SOFC is caused by the mismatch of coefficients of thermal expansion (CTEs) among materials, while the other part of the maximum 1st principal stress is mainly caused by temperature gradient.     

Pore-scale investigation of mass transport and electrochemistry in a solid oxide fuel cell anode

The development and validation of a model for the study of pore-scale transport phenomena and electrochemistry in a Solid Oxide Fuel Cell (SOFC) anode are presented in this work. This model couples mass transport processes with a detailed reaction mechanism, which is used to model the electrochemical oxidation kinetics. Detailed electrochemical oxidation reaction kinetics, which is known to occur in the vicinity of the three-phase boundary (TPB) interfaces, is discretely considered in this work. The TPB regions connect percolating regions of electronic and ionic conducting phases of the anode, nickel (Ni) and yttria-stabilized zirconia (YSZ), respectively; with porous regions supporting mass transport of the fuel and product. A two-dimensional (2D), multi-species lattice Boltzmann method (LBM) is used to describe the diffusion process in complex pore structures that are representative of the SOFC anode. This diffusion model is discretely coupled to a kinetic electrochemical oxidation mechanism using localized flux boundary conditions. The details of the oxidation kinetics are prescribed as a function of applied activation overpotential and the localized hydrogen and water mole fractions. This development effort is aimed at understanding the effects of the anode microstructure within TPB regions. This work describes the methods used so that future studies can consider the details of SOFC anode microstructure.     

Thermophysical Properties of Zirconia Toughened Alumina Ceramics with Boron Nitride Nanotubes Addition

Abstract

The demand for high thermal conductivity substrates with electrically insulating materials are increasing with the emerging markets in power electronics and mobile telecommunication device packages. Effective heat transfer in those packages is important to provide high performance and reliability of the product. This paper mainly presents the thermophysical properties of zirconia toughened alumina ceramics with the addition of small amount of boron nitride nanotubes (BNNTs). The effects of the boron nanotubes addition on the sintering behavior, the microstructure and the thermal properties of the yttria-stabilized zirconia toughened alumina (YZTA), nanocomposite ceramics are investigated. The addition of 0.3?wt% boron nitride nanotubes into the YZTA matrix enhanced the thermal diffusivity as well as a mechanical strength. Above all, the addition of boron nitride nanotubes greatly decreased the coefficient of thermal expansion (CTE) of the composites in which the CTE of pure alumina increases with increasing temperatures. Moreover, the BNNTs added YZTA composites revealed a drastic decrease in CTE at high temperature range, 400–800?°C. This enhanced thermal stability of YZTA–BNNT composites may have a potential application to the high temperature structural ceramics and high power semiconductor packaging substrate.     

Effect of alumina on the curvature, Young's modulus, thermal expansion coefficient and residual stress of planar solid oxide fuel cells

Planar solid oxide fuel cells (SOFCs) are composites consisting of porous and dense functional layers as electrodes and electrolytes, respectively. Because of the thermo-elastic mismatch between the individual layers, residual stresses develop during manufacturing and cause unconstrained cells to warp. The addition of alumina decreases the thermal expansion coefficient (TEC) of the NiO and yttria-stabilized zirconia (YSZ) anode-support material. Correspondingly, the lower TECs have flattened the half cells during fabrication. In addition, the residual stress at room temperature (RT) for samples with more than 4 wt% alumina is only 20% of the residual stress of the samples without alumina, at approximately 100 MPa. The effects of Al₂O₃ on the curvature, Young's modulus, TEC and residual stress of the SOFC with (NiO-YSZ)_1−x(Al₂O₃)_x (x = 1-5 wt%) anode support are discussed in this work.     

An SOFC anode model using TPB-based kinetics

Fundamental studies focusing on the electrode kinetics are essential in understanding the fuel cell operation and optimizing the electrode designs. In this study, we determined the triple-phase boundary (TPB)-based kinetics of hydrogen electrochemical oxidation using nickel patterned electrode experimental data and the Butler-Volmer formalism of the oxidation process. The same kinetics are then incorporated in a cermet electrode electrochemical model to estimate the effective TPB density of the nickel/yittrium-stabilized zirconia cermet anode. The kinetics are found to be of the same order of magnitude as previously determined by the microstructure reconstruction of cermet anode. Simulation results further revealed that the effective TPB density is several orders of magnitude lower than the typically reported physical densities of the cermet anode that possibly suggests that only a minor fraction of the physical TPB is actually required or available to produce the cell current at given cell voltage. The effect of various operating conditions on the anode activation overpotential is also investigated and discussed in this study.     

Influence of composition and Cu impregnation method on the performance of Cu/CeO2/YSZ SOFC anodes

In this study, we report on how different impregnation procedures affect the distribution and morphology of the Cu component in Cu/CeO₂/YSZ (YSZ, yttria-stabilized zirconia) composite anodes and how this affects anode performance. Two different methods for Cu addition to the porous YSZ anode were investigated: impregnation using aqueous solutions of Cu(NO₃)₂ and impregnation using aqueous solutions of Cu(NO₃)₂ plus urea. The latter method produced a homogeneous distribution of Cu throughout the anode while the former resulted in a higher concentration of Cu near the exposed surface relative to that in the bulk. Studies of the thermal stability of the deposited copper layers and the influence of the Cu distribution on cell performance when operating with humidified H₂ as the fuel are also presented.