Thermodynamic analysis of 100% system fuel utilization solid oxide fuel cell (SOFC) system fueled with ammonia

https://doi.org/10.1016/j.enconman.2021.114839Get rights and content

Highlights

  • Feasibility of 100% system fuel utilization SOFC system fed with NH3 is studied.

  • 100% fuel utilization is achieved by operating SOFC in dead-end anode (DEA) loop.

  • System efficiency as high as 75% LHV can be achieved with NH3 fed DEA loop SOFC.

  • The energy and exergy balance studies are conducted and compared with standard SOFC.

  • Parametric analyses of critical operating parameters of the DEA system are studied.

Abstract

The system efficiency of solid oxide fuel cells (SOFCs) fueled with ammonia or pure hydrogen always underperforms compared to the SOFCs fueled with carbon-based fuels. Hence, improving the efficiency of SOFCs fueled with pure hydrogen or hydrogen carriers such as ammonia is required to promote hydrogen economy. This study deals with the thermodynamic analyses of SOFC with 100% system fuel utilization configuration achieved by a dead-end anode (DEA) loop. The results are compared with the conventional SOFC configuration with the afterburner setup. It is shown that the energy efficiency and exergy efficiency of ammonia-fueled recirculation system are 12.17% points and 11.39% points higher than the conventional system. Energy balance comparison shows that the reduction in heat loss through exhaust gas is one of the significant factors for improving the system efficiency. The improvement in the exergy efficiency of the recirculation system is attributed to the reduction in exergy destructions in air preheaters and the elimination of exergy destruction at the afterburner. In addition, a parametric analysis and artificial neural network– genetic algorithm (ANN-GA) based optimization are performed to understand the effects of stack fuel utilization, current density, stack operating temperature, radiation loss, the pressure ratio across the membrane, and species removal rate in anode off-gas on system efficiency. The system efficiency as high as 75% lower heating value (LHV) can be achieved by the DEA loop configuration with operating the system at lower stack fuel utilizations.

Introduction

Solid oxide fuel cells (SOFCs) are ideal candidates for stationary power generation, regarding their fuel flexibility and high efficiency. Their high operating temperature makes it attractive to be combined with bottoming thermal cycles to improve the total energy efficiency. Standalone SOFCs application can reach system efficiency over 50%, and their integration with bottoming cycles enables reaching the efficiency over 70% [1], [2]. The distinctive feature of fuel flexibility has attracted considerable attention in recent years. The most commonly investigated fuels are carbon-containing fuels such as natural gas, methane, ethanol, and biomass [3], [4], [5], [6]. Synthesized fuels such as hydrogen and ammonia for use in SOFCs are still in the research and development phase [7], [8], [9].

Hydrogen is an ideal fuel because of its highest gravimetric energy density (LHV = 120 MJ/kg), and the combustion of hydrogen produces only H2O, and no anthropogenic CO2 is released. However, the storage of hydrogen is a major concern due to its low density (0.0813 g/L at 25 °C and 1 bar), i.e., hydrogen is 3.2 and 2700 times less energy-dense than natural gas and gasoline. Hydrogen storage requires either gaseous form storage by compression to around 700 bar or liquid form storage by cooling down below −253 °C. Compression requires energy equivalent to 13–18% of its LHV value [10], and around 30–40% of its LHV [11] to liquefy. Besides compression or liquefaction storage, another option is chemically combining hydrogen storage with other metals or metal hydrides, but thermal management limits the feasibility of these storage techniques.

Other widely used hydrogen carriers are methanol and ammonia because of their higher hydrogen content of 18.8 wt% and, 16.6 wt%, respectively. Nevertheless, the attractive feature of ammonia over methanol is that cracking of ammonia yields only hydrogen and nitrogen, whereas reforming of methanol yields carbon monoxide and carbon dioxide in addition to hydrogen. Other advantages that make ammonia promising are: (i) ease of synthesis via highly productive Haber-Bosch process [12], (ii) ease of storage than hydrogen. Ammonia is easily liquefiable at around 10 bar pressure at ambient temperature or −33 °C at ambient pressure [8], [12], [13], and ammonia is less flammable than hydrogen and can be easily detected in the event of leakage.

The potential of using ammonia for fuel cells was first investigated by Cairens et al. [14] for alkaline fuel cells. Later, Vayenas and Farr [15] showed the feasibility of using ammonia in an SOFC. Metkemeijer and Achard [16] presented an indirect ammonia SOFC using an ammonia cracker. Wojcik et al. [17] experimentally demonstrated that ammonia could be directly used as a fuel in the SOFC system. Furthermore, the authors also concluded that the nickel-based anode SOFC system is ideal because nickel also acts as a catalyst for ammonia cracking. Later, many researchers studied the performance of SOFCs based on oxygen ion-conducting electrolyte (SOFC-O) and proton conducting electrolyte (SOFC-H). It was shown that the SOFC-H performs better than SOFC-O due to the higher partial pressure of H2 in the anode of SOFC-H [18], [19]. Cinti et al. [20] experimentally showed that diluted ammonia performs on par with the pure ammonia-fed SOFC system. This result is convincing to use dilute ammonia which is non-toxic and non-flammable for mobile applications. In another experimental study, Cinti et al. [21] investigated a commercial planar SOFC with nickel-based anode using three fuels: pure hydrogen, ammonia, and their mixture to represent cracked ammonia. It was shown that the SOFC with pure ammonia performed better than the others due to the decrease in operating temperature because of the cracking reaction that absorbed heat produced in the cell. The catalytic activity and long-term stability of NH3 decomposition reactor and NH3 autothermal cracker were investigated by Okanishi et al. [22].

However, the efficiency of the SOFC system fueled with ammonia was not high compared with SOFCs fueled with carbon-based fuels. The main factors that govern the improvement in electrical efficiency of SOFC system are: (i) high fuel utilization leading to reduced fuel consumption, (ii) fuel reforming technology, (iii) the materials choices for the cell to reduce overpotentials, (iv) reduced system heat losses specifically in small-sized SOFC systems, (v) efficient BoP components to reduce excessive parasitic losses [23]. Operating the SOFC at high fuel utilization is an attractive option; however, at the expense of a risk of Ni oxidization [24]. It is possible to improve the system fuel utilization without making the stack fuel utilization high. Higher system fuel utilization can be achieved in two ways: (i) multi-staging stack and (ii) dead-end anode (DEA) loop by 100% recirculating the anode gas with or without purging arrangement. Elangovan et al. [25] first noticed the improvement in efficiency due to multi-staging, which enhanced the system fuel utilization and suppressed temperature gradients in the cell. Matsuzaki et al. [26] demonstrated that net AC efficiency of approximately 76% (LHV) can be achieved by using proton-conducting electrolyte in a multistage stack.

A detailed study on the influence of fuel supply ratio and fuel utilization factor on each stack for a two-staged stack design was performed by Tachikawa et al. [27]. Nakamura et al. [28] and Dohkoh et al. [29] emphasized the importance of regeneration of anode gas by separating water and carbon dioxide before feeding the off-gas from the upstream stack to the downstream stack in a two-stage SOFC stack configuration. Furthermore, Dohkoh et al. [29] introduced a membrane separation concept to remove steam and CO2 from the anode gas of the first stack. They reported that a DC efficiency of about 74.8% LHV can be achieved using a two-stage SOFC.

A concept of full recirculation of anode off-gas, referred to as dead-end anode (DEA) loop, is prevalent in polymer electrolyte membrane fuel cells (PEFCs) [30], [31]. Dead-end anode loop for hydrogen-fed SOFC was first investigated by Peters et al. [32]. They compared three hydrogen-fueled-SOFC system configurations: (i) conventional configuration without anode recycle, (ii) configuration with anode recycling and condensing moisture in anode off-gas, (iii) configuration with 100% system anode gas recycling. The corresponding electrical efficiencies were 48%, 54.5%, and 60% for the three cases, respectively. Furthermore, exergy analysis by Selvam et al. [33] revealed that the full anode off-gas recirculation in the SOFC also exhibited higher exergetic efficiency in comparison to the conventional configurations.

In order to realize the DEA loop system for ammonia-fueled SOFC system, it is critical to remove nitrogen and water from the anode off-gas, unlike only water condensation is required in the case of a hydrogen-fueled SOFC system [34]. The most common techniques used for nitrogen separation are solvent adsorption, pressure swing adsorption, and cryogenic distillation. Nevertheless, these separation techniques are energy-intensive, i.e., a significant amount of energy is required for solvent regeneration in the case of the adsorption method and for refrigeration in the case of cryogenic distillation [35]. Another promising method that has gained attention over the years is membrane-based hydrogen separation. This technique is further motivating with the fact that the energy required for the separation of gases is lower than the other techniques. In addition, high selectivity and purity can be obtained [36], [37]. The most widely used membrane is dense metallic membranes, specifically Palladium (Pd) based membranes due to their high exceptional H2 permeability and selectivity [38]. However, the major disadvantages of pure Pd membranes are that it is brittle in a pure hydrogen environment at low temperatures and expensive [39]. Hence, Pd is always alloyed with other metals to suppress hydrogen embrittlement and cost. In addition, alloying with a silver (Ag) further increases the membrane's permeability [40].

While several approaches have been conducted to enhance the efficiency of ammonia-fueled SOFC systems using better electrode and electrolyte materials [18], [19], and better fuel processing techniques [22], there were no studies to the best of the authors' knowledge to improve the system efficiency of ammonia-fueled SOFC by maximizing system fuel utilization. The objective of this study is to propose an innovative concept for 100% fuel utilization SOFC system fueled with ammonia. This concept is achieved by full recirculation of anode off-gas with moisture removal and membrane separation of nitrogen. The proposed configuration is compared with the conventional configuration in terms of system performance, energy efficiency, and exergy efficiency. A parametric study is performed to understand the impacts of design parameters, and an ANN-GA-based optimization is also conducted to investigate the optimum input parameters in terms of highest energy and exergy efficiencies. The novelty of this study is to propose and assess the feasibility of 100% system fuel utilization ammonia-fueled SOFC system to achieve the highest energy and exergy efficiencies.

Section snippets

Conventional SOFC system

A conventional SOFC system fueled with ammonia is schematically shown in Fig. 1(a). Ammonia from a high-pressure tank is preheated using the anode off-gas and enters the stack at 670 °C. The cathode air is slightly pressurized above the atmospheric pressure to overcome the pressure loss within the stack. The pressurized air is preheated to 620 °C before entering the stack by means of the flue gas from the afterburner. The flue gas after air preheating exits the system. On the other side, the

System modeling

The system modeling is performed under the following assumptions:

  • i.

    SOFC model is a 0D model.

  • ii.

    SOFC operates adiabatically, i.e. no heat loss to the surroundings. The heat loss is treated exceptionally in the parametric study.

  • iii.

    NH3 decomposition reaction occurs at equilibrium conditions inside the stack.

  • iv.

    Mass transport through the membrane is steady, and heat loss across the membrane is negligible.

  • v.

    Cathode air inlet composition (mole %) is 77.29 N2, 20.75 O2, 1.01 H2O, 0.92 Ar, 0.03 CO2.

  • vi.

    Ambient pressure

Electrochemical and system performance

The electrochemical performance and the system performance of the SOFCs in the conventional and DEA loop configurations are shown in Table 5. The results show that the cell voltage of the conventional SOFC system is slightly higher than the DEA configuration. This is attributed to the fact that the anode gas entering the SOFC contains pure ammonia in the case of conventional configuration, whereas it is a mixture of NH3, H2, and H2O in the DEA configuration. This causes a slight decrease in

Conclusions

Thermodynamic analysis of a novel SOFC system fueled with ammonia in full recirculation of anode off-gas mode is designed, and the performance is compared with the conventional configuration. Full recirculation of the anode off-gas is achieved by condensing the moisture and membrane separation of N2 at the fuel regeneration unit. The residual hydrogen separated from the anode off-gas is mixed with the incoming fresh fuel and is preheated and sent to the SOFC stack with 100% system fuel

CRediT authorship contribution statement

Kalimuthu Selvam: Conceptualization, Methodology, Investigation, Software, Visualization, Validation, Writing – original draft. Yosuke Komatsu: Methodology, Visualization, Writing - review & editing. Anna Sciazko: Methodology, Visualization, Writing - review & editing. Shozo Kaneko: Conceptualization, Supervision, Writing - review & editing. Naoki Shikazono: Conceptualization, Resources, Supervision, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (60)

  • A. Afif et al.

    Ammonia-fed fuel cells: a comprehensive review

    Renew Sustain Energy Rev

    (2016)
  • R. Metkemeijer et al.

    Ammonia as a feedstock for a hydrogen fuel cell; reformer and fuel cell behaviour

    J Power Sources

    (1994)
  • A. Wojcik et al.

    Ammonia as a fuel in solid oxide fuel cells

    J Power Sources

    (2003)
  • F. Ishak et al.

    Thermodynamic analysis of ammonia-fed solid oxide fuel cells

    J Power Sources

    (2012)
  • G. Cinti et al.

    SOFC operating with ammonia: stack test and system analysis

    Int J Hydrogen Energy

    (2016)
  • Q. Fang et al.

    SOFC stack performance under high fuel utilization

    Int J Hydrogen Energy

    (2015)
  • R.K. Ahluwalia et al.

    Fuel cell systems for transportation: status and trends

    J Power Sources

    (2008)
  • S. Hikita et al.

    Power-generation characteristics of hydrogen fuel cell with dead-end system

    JSAE Rev

    (2002)
  • R. Peters et al.

    Efficiency analysis of a hydrogen-fueled solid oxide fuel cell system with anode off-gas recirculation

    J Power Sources

    (2016)
  • W.H. Chen et al.

    Influence of vacuum degree on hydrogen permeation through a Pd membrane in different H2/N2 gas mixtures

    Renew Energy

    (2020)
  • M. Nordio et al.

    Effect of sweep gas on hydrogen permeation of supported Pd membranes: experimental and modeling

    Int J Hydrogen Energy

    (2019)
  • M. Ni et al.

    Thermodynamic analysis of ammonia fed solid oxide fuel cells: comparison between proton-conducting electrolyte and oxygen ion-conducting electrolyte

    J Power Sources

    (2008)
  • J.R. Ferguson et al.

    Three-dimensional numerical simulation for various geometries of solid oxide fuel cells

    J Power Sources

    (1996)
  • S.H. Chan et al.

    Complete polarization model of a solid oxide fuel cell and its sensitivity to the change of cell component thickness

    J Power Sources

    (2001)
  • S. Kalimuthu et al.

    3-E analysis of a pressurized pulverized combined cycle (PPCC) power plant using high ash Indian coal

    Energy

    (2017)
  • J. Catalano et al.

    Influence of water vapor on hydrogen permeation through 2.5 μm Pd-Ag membranes

    Int J Hydrogen Energy

    (2011)
  • J. Arriagada et al.

    Artificial neural network simulator for SOFC performance prediction

    J Power Sources

    (2002)
  • J. Milewski et al.

    Modelling the SOFC behaviours by artificial neural network

    Int J Hydrogen Energy

    (2009)
  • L. Van Biert et al.

    A thermodynamic comparison of solid oxide fuel cell-combined cycles

    J Power Sources

    (2018)
  • F. Ishak et al.

    Energy and exergy analyses of direct ammonia solid oxide fuel cell integrated with gas turbine power cycle

    J Power Sources

    (2012)
  • Cited by (0)

    View full text