Elsevier

Journal of Power Sources

Volume 407, 15 December 2018, Pages 63-69
Journal of Power Sources

Transient simulation of a tubular micro-solid oxide fuel cell

https://doi.org/10.1016/j.jpowsour.2018.10.062Get rights and content

Abstract

A 2D-axisymmetric model is developed for a tubular micro-solid oxide fuel cell (TμSOFC), detailing heat, mass, momentum, and charge transfer as well as electrochemical reactions. Thermal transients are simulated. No parameter fitting was done, and only kinetics parameters estimated in our previous work for isothermal, steady state conditions are used. Transient experiments were conducted for this study and the data are presented. The purely predictive simulation results are found to be consistent with experimental results. The model is used to simulate the transient behavior of a step in cell current and air flow. The sensitivity of transients to several parameters are simulated and discussed. It is found that in general, transients of the system were controlled by the thermal transients indicating that other transients were much faster. Therefore, a fast electric response results from a small thermal mass of the system. Two modes could be identified in the thermal transients: fast and slow; the former is found to be associated with the cell itself, while the latter was found to be due to the furnace wall surrounding the cell. A step in current resulted in a sharp drop in voltage and its gradual recovery, which is a result of the cell slowly heating up.

Introduction

The transient behavior of a solid oxide fuel cell (SOFC) has a significant practical interest, more so as the commercialization of this technology becomes more viable. The operation of an SOFC is by nature transient; the current and voltage can change significantly during its operation as a result of changing load requirements. Critical components of a SOFC power generation system, such as the control and power conditioning sub-units, are typically designed to handle transients. As the load changes in the SOFC, concomitant changes in operating parameters are required to keep it running both effectively and within the acceptable range. For example, a load increase requires a higher flow rate of reactants to maintain the same utilization and efficiency. Furthermore, to avoid temperatures that exceed permissible limits, the extra thermal energy generated at higher load levels must be removed, for example by increasing the air circulation rate. Managing and controlling such transient is vital to proper operation of the SOFC system [1]. During start-up and shut down cycles thermal transients are usually significant, and it is important to manage the thermal stresses so as to ensure mechanical stability and safe operation. Optimum control of such transients will maximize the efficiency of the SOFC system [2].

Computer simulation is a powerful tool that can be used to improve understanding of SOFC transients, how to predict them, and to control them effectively. There are many SOFC models reported in the literature, and these show variations which vary in their details, designs, and purposes. Bhattacharyya et al. [3] provide a detailed review of a wide range of these models. One of the main purposes of transient modelling is effective SOFC control, and many dynamic models have been used to study and optimize control strategies [2,[4], [5], [6], [7], [8]]. A simplified model that correlates the SOFC input parameters to parameters that need to be controlled, for example, power, voltage, current and temperature, might be sufficient for this purpose. For this reason, control based models are often simplified zero [[9], [10], [11], [12], [13]] or one-dimensional [[14], [15], [16], [17], [18]]. Transient models have been used to study the systems aspect of the SOFC operation, i.e. the balance of plant, which focuses on the interaction of the components [19,20], the control and optimization of the overall system [[21], [22], [23], [24], [25], [26]], to analyze system performance in the grid [27], to understand the effect of load transients [28], and to develop load-following strategies [[29], [30], [31]]. In addition, dynamic models based on physics have been developed that model SOFC in two dimensions [32,33], quasi-three dimension [34,35], and three dimensions [[36], [37], [38]]. Some transient models are used to gain a better understanding of various SOFC operational details such as startup conditions [39,40], carbon deposition [41], effect of flow configuration [42], and the induced mechanical stress effects [42]. Models simulating transient behavior of SOFCs go beyond the time-domain transient simulations in cases where in frequency domain the impedance response of SOFC cells and electrodes are modeled [33,[43], [44], [45], [46], [47], [48], [49], [50], [51]].

The significant phenomena observed in a fuel cell have widely differing transient time-scales. The electrochemical reaction steps are very fast and have time constants on the order of a fraction of a second [52]. The mass transport phenomena at high temperatures are quite fast and sub-second as well [52]. The slowest are the thermal transients, which are highly dependant on factors such as the design and architecture of the cell, and typically greater than 100 s [52]. Transient time constants, especially the ones for heat and mass transport, depend on the design and operating conditions of the cell. For example, cells with thicker electrodes or lower gas velocities have slower mass transport transients. The thermal transients are often the controlling ones, and therefore they are most importance from a practical point of view. This observation has induced some modellers to include only the thermal transients in their models. One of the earliest dynamic SOFC only modeled temperature distribution and electrochemical reactions [53,54]. Furthermore, there are even transient models that assume isothermal conditions [55,56].

TμSOFCs offer several advantage compared to other geometries. Because of the micro deign, they have small thermal mass, and also the thermal gradients across the cell are minimized. Both these advantages makes them superior for rapid star-up, shut-down, and fast load-following applications, which are all inherently transient phenomenon.

In this paper, the TμSOFC steady-state model reported previously [57] is extended to predict transient behavior. The resulting transient model comprehensively captures all of the relevant phenomena in detail. The transient transport phenomena within the gas channels and the electrodes are included. A unique aspect of the presented transient model is its experimental validation. Experimental validation is a significant aspect of any simulation work, to ensure the relevance and applicability of the results. However, little work has been done on experimental validation of transient models [56], which makes a simulation work presented along with experimental validation, such as this work noteworthy. Hence, this work is significant because it presents such details in the transient model along with the presentation of a comprehensive experimental validation; it is one of the few if not the only paper in that regard.

Section snippets

Experimental

The tubular micro-solid oxide fuel cell that is the subject of this work is shown in Fig. 1. The cell was developed at the Alberta Research Council using the electrophoretic deposition technique (EPD). EPD allows for fabricating very thin layers of anode and electrolyte as shown in Fig. 1(b). The cell is anode supported, with a Ni/YSZ anode, YSZ electrolyte, and LSM/YSZ cathode and has one end closed. A diffuser tube was used to inject hydrogen to the closed end which flowed back along the

Results and discussion

The effect of various quantities on the thermal and voltage transients to a step change in the current load from 0 to 1 A was simulated. Because the behavior of the TμSOFC itself is the focus, the results presented are for the case without a thermocouple effect.

As a general rule, the time constant for a transient thermal system is proportional to its thermal mass, mcp, and to the inverse of the heat transfer coefficient [61]:ΔT(t)=ΔT(0)etτ;τ=mcpU

The heat transfer coefficient increases when

Conclusion

A transient model for a TuSOFC was developed and solved. As no parameter fitting was done, the simulation results were purely predictive. Simulations were validated against experimental results and were found to make reliable predictions. When the effect of thermocouples attached to the cell was included, the model predictions were closer to the experimental values.

The thermal behavior of the system was distinguished by two transients. The faster transient corresponded to the cell, and a slow

Acknowledgement

This work was financially supported by the Alberta Ingenuity Fund in Nanotechnology, and the Natural Science and Engineering Council of Canada (NSERC) strategic grant.

References (69)

  • M. Fardadi et al.

    Appl. Energy

    (2016)
  • L. Zhang et al.

    Electrochim. Acta

    (2015)
  • H. Cheng et al.

    Int. J. Hydrogen Energy

    (2016)
  • S.A. Taher et al.

    Int. J. Electr. Power Energy Syst.

    (2014)
  • B.J. Spivey et al.

    J. Process Contr.

    (2012)
  • P.H. Lin et al.

    J. Power Sources

    (2009)
  • R. Gaynor et al.

    J. Power Sources

    (2008)
  • C. Haynes

    J. Power Sources

    (2002)
  • R.S. Gemmen et al.

    J. Power Sources

    (2005)
  • P. Aguiar et al.

    J. Power Sources

    (2005)
  • Y.G. Huangfu et al.

    Energy Convers. Manag.

    (2013)
  • K.J. Albrecht et al.

    J. Power Sources

    (2016)
  • N.F. Harun et al.

    Appl. Energy

    (2016)
  • F. Mueller et al.

    J. Power Sources

    (2009)
  • M.L. Ferrari et al.

    J. Power Sources

    (2005)
  • T. Kaneko et al.

    J. Power Sources

    (2006)
  • L. Barelli et al.

    Appl. Energy

    (2013)
  • L. Barelli et al.

    Energy

    (2016)
  • Y. Luo et al.

    Energy

    (2015)
  • M. Fadaei et al.

    Energy Convers. Manag.

    (2014)
  • M. Hering et al.

    J. Power Sources

    (2017)
  • M. Hering et al.

    J. Power Sources

    (2016)
  • T.X. Ho

    Int. J. Hydrogen Energy

    (2014)
  • M. Peksen

    Int. J. Hydrogen Energy

    (2014)
  • M. Peksen et al.

    Int. J. Hydrogen Energy

    (2013)
  • M.H. Chen et al.

    J. Power Sources

    (2012)
  • T. Jacobsen et al.

    Electrochim. Acta

    (2008)
  • A. Kromp et al.

    Electrochim. Acta

    (2013)
  • Y.X. Shi et al.

    J. Power Sources

    (2012)
  • M. Fadaei et al.

    Energy Convers. Manag.

    (2015)
  • E. Achenbach

    Elsevier Science Sa Lausanne

    (1994)
  • E. Achenbach

    J. Power Sources

    (1995)
  • A. Murshed et al.

    J. Power Sources

    (2007)
  • D. Bhattacharyya et al.

    Chem. Eng. Sci.

    (2009)
  • Cited by (18)

    • Significance of radiative heat transfer in modeling a tubular micro-solid oxide fuel cell

      2020, International Journal of Thermal Sciences
      Citation Excerpt :

      The electrochemical performance of the cell was experimentally validated as reported in Ref. [35]. In addition, the simulated temperature distribution was validated against steady-state experiments [36], and also the cell's thermal transients was validated against experiments [37]. Special attention was paid to the validity of the simplifications used in the development of the model, as well as the numerical consistency of the model.

    • Mathematical modeling and simulation of hydrogen-fueled solid oxide fuel cell system for micro-grid applications - Effect of failure and degradation on transient performance

      2020, Energy
      Citation Excerpt :

      Mass transport phenomena at high temperatures occur on time scales between those of electrochemistry and heat transport. Transient time constants especially those for heat and mass transport strongly depend on operating conditions [49–51]. Degradation usually occurs over time periods of several hundred hours.

    • Influence of structure size on voltage uniformity of flat tubular segmented-in-series solid oxide fuel cell

      2020, Journal of Power Sources
      Citation Excerpt :

      Solid oxide fuel cell (SOFC) becomes one of the most promising power generation methods because of its high efficiency, clean power generation and silent operation [1–7].

    • Influence of fuel flow rate on the performance of micro tubular solid oxide fuel cell

      2020, International Journal of Hydrogen Energy
      Citation Excerpt :

      Solid oxide fuel cell (SOFC) has become one of the most promising power generation methods because of its high efficiency, clean power generation and silent operation [1–8].

    • High performance of anode supported BaZr<inf>0.1</inf>Ce<inf>0.7</inf>Y<inf>0.1</inf>Yb<inf>0.1</inf>O<inf>3-δ</inf> proton-conducting electrolyte micro-tubular cells with asymmetric structure for IT-SOFCs

      2019, Journal of Electroanalytical Chemistry
      Citation Excerpt :

      The porous anode membrane near the electrolyte (~4 μm) is expected as electrode functional layer where electrochemical reactions occur [40]. This asymmetric pore structure of “sponge-like micro-pores electrode|homogeneous porous functional layer” looks forward to being beneficial of gas diffusion and surface reaction for cells during operation [41]. As shown illustrated in Fig. 2(c), such interface is especially suitable for expanding the TPBs length, which can lead to decreasing electrodes polarization.

    View all citing articles on Scopus
    View full text