Recent advances and challenges in degradation issues of direct ammonia solid oxide fuel cells: comprehensive review

Abstract

Direct ammonia solid oxide fuel cells (DA-SOFCs) are promising energy-conversion devices that serve as alternates to hydrogen (H₂)-fueled SOFCs, given their potential to overcome the current limitations of green hydrogen, e.g., the high costs of production and transportation as well as the low volumetric energy density and high storage levels. Unlike H₂-fueled SOFCs, DA-SOFCs can offer high efficiency and approachability based on the already well-established value chain of ammonia. However, despite these advantages, recent studies on DA-SOFCs have reported significant degradation issues in many aspects, all of which should be carefully considered for broader commercialization. In this review, we introduce the recent progress and challenges related to DA-SOFCs, focusing more on degradation issues and methods capable of suppressing them in three directions: (1) materials, (2) cells, and (3) systems. Therefore, the study provides motivation for additional research directions for the development of DA-SOFCs.

---

Hyunho Lee

Hyunho Lee is a M.S. student in the Department of Future Energy Convergence at Seoul National University of Science & Technology (SeoulTech) under the supervision of Prof. Mingi Choi. His research focuses on degradation mechanisms and surface engineering of electrodes in electrochemical devices.

Jaewan Baek

Jaewan Baek is a M.S. student in the Department of Future Energy Convergence at SeoulTech under the supervision of Prof. Mingi Choi. His research focuses on applying machine-learning and deep-learning techniques to electrochemical devices.

Mingi Choi

Mingi Choi is an Assistant Professor in the Department of Future Energy Convergence at SeoulTech. He received his PhD in Mechanical Engineering from Sungkyunkwan University (SKKU). After obtaining his PhD, he worked as a postdoctoral researcher at SKKU from 2021 to 2022. His current research interests focus on ammonia-fueled fuel cells, electrolysis cells, artificial intelligence, and computational fluid dynamics simulations.

---

1. Introduction

Given the increase in the global demand for carbon neutrality, hydrogen (H₂), one of the most promising green energy sources, has been attracting significant attention.^1–3 Specifically, in fuel cells, H₂ is utilized as a type of fuel for electrochemical reactions, reacting with oxygen as an oxidant and producing water, with high environmental friendliness. However, the water electrolysis technology to produce green hydrogen is not yet cost-competitive compared to the technologies related to gray and blue types of hydrogen, which are made from methane-steam reforming.^4–6 In addition, hydrogen requires a temperature of −253 °C at 1 atm to liquefy, thus requiring a considerable amount of energy for storage and transportation. Furthermore, hydrogen has a very low volumetric energy density (8.49 MJ L⁻¹ for H₂), which necessitates large sites for storage, making it difficult to apply to portable devices.^7,8 Thus, various hydrogen carriers have been proposed to overcome these limitations. To date, the most representative hydrogen carriers have been hydrocarbon fuels, most notably methane (CH₄), which is a natural gas. However, there is also the critical problem of generating CO₂ after combustion and utilization in fuel cells.^9,10 Regarding these concerns, ammonia (NH₃), a carbon-free fuel that may serve as an alternative to hydrogen and hydrocarbon fuels, is of great interest among researchers who study renewable energy.^11–14 NH₃ liquefies at a relatively high temperature compared to H₂ (25 °C at 10 atm, −33 °C at 1 atm), and has approximately a ∼1.5-fold higher volumetric energy density (12.92 MJ L⁻¹ for NH₃), making it easy to store and transport, and thus imparting great cost-competitiveness.^15,16 In addition, the infrastructure for ammonia is already in place, and globally it is the second most commonly produced chemical with producing about 183 million metric tons in 2020, finding active uses in various fields, such as agriculture and chemical synthesis.¹⁷ Therefore, it can be readily produced and utilized in a robust value chain, providing comparable cost-effectiveness ($2.95 per kg_H2 for ammonia) compared to that of natural gas ($1.37 per kg_H2), blue hydrogen ($2 per kg_H2), and green hydrogen ($ 3.6–4 per kg_H2).^18–22 Notably, the market size of green ammonia is growing rapidly with a compound annual growth rate (CAGR) of 75% from 2023 to 2031 for eco-friendly production.²³

Common power generation technologies that use ammonia are combustion and fuel cells. Combustion is the technology of using ammonia to produce gas that can be used to operate a turbine to generate electricity. Combustion enables large-scale power generation by utilizing existing power plants, but its disadvantages include NO_x emissions and low efficiency.²⁴ Also, when combusting ammonia alone, incomplete combustion may occur, so it needs to be mixed with hydrogen for combustion.²⁵ Fuel cells, in contrast, are powered by electrochemical reactions, with a relatively high efficiency of 40–60%.²⁶ And it is very eco-friendly as the only emissions are water and nitrogen, and it can be applied directly to buildings, vehicles, ships, etc. because it can generate power on a small scale. Specifically, solid oxide fuel cells (SOFCs) have excellent compatibility with ammonia as a fuel.^27–29 Thermodynamically, 99.99% of ammonia can be decomposed at 400 °C.³⁰ Because SOFCs commonly operate above 700 °C, the temperature is sufficient for full ammonia decomposition. A large amount of nickel (Ni) is used as a catalyst in the fuel electrode (anode), with Ni being the second most reactive catalyst for ammonia cracking after Ru.^31,32 Therefore, it is possible to construct a direct ammonia solid oxide fuel cell (DA-SOFC) system that directly utilizes ammonia as a fuel without external reforming (Fig. 1(a)).^34,35 This is accordingly a strategic approach that saves both the space and capital required to build and operate an external reformer and simplifies the system by mitigating additional considerations, such as thermal management and the recirculation of unreacted gas that must be fed into the external reformer. More interestingly, given that the hydrogen oxidation reaction (HOR) inside the fuel cell (2H₂ + O₂ → 2H₂O, ΔH = −286 kJ mol_H2⁻¹) is exothermic and the ammonia cracking reaction (2NH₃ → N₂ + 3H₂, ΔH = +36.3 kJ mol_H2⁻¹) is endothermic, there are significant advantages in terms of energy utilization.³⁶ As shown in Fig. 1(b), the efficiency exceeds 86% in the range of 600–900 °C, and if the operating temperature can be reduced to 400 °C or lower, an efficiency rate of even 100% becomes theoretically possible, as the heat released by the HOR under an exothermic reaction can be used to decompose ammonia, which is an endothermic reaction.^33,36 This process also allows a heat balance to be established, eliminating the need for additional thermal management. In addition to the thermodynamic efficiency, various papers have demonstrated electrochemical performance outcomes of DA-SOFCs compared to those of hydrogen-fueled SOFCs (Fig. 1(c) and Table 1). Both the sufficient operating temperature range and the catalysts used in DA-SOFCs allow high electrochemical performance capabilities in the given structure compared to H₂-fueled SOFCs. Despite these advantages, the most critical problem to overcome is stability, as shown in Fig. 1(d) and Table 2. DA-SOFCs show significant degradation rates during long-term operation compared to common SOFCs.^34,66 Notably, stabilities in DA-SOFCs present a particular trend showing lower stability at the lower operating temperature, which is the opposite of that in H₂-fueled SOFCs, showing higher stability at the lower operating temperature. Complicate degradation factors and their influence on DA-SOFCs bring this opposite trend. A number of the degradation factors affect SOFCs, such as Sr segregation, Cr and sulfur poisoning, interfacial chemical interdiffusion, etc. In addition, in the fuel electrode of DA-SOFCs, the ammonia cracking reaction, as an additional chemical process that occurs before the HOR, can induce the undesired chemical reactions in the cell itself and the balance of plant (BOP) of the systems.^67–69 Various papers have reported different and complicated degradation behaviors under direct ammonia operating conditions. Degradation behaviors and their effects on stability can differ according to the materials used, including the catalyst and the catalyst support, and also according to the different operating temperatures and the electrochemical reactions depending on the different charge carriers and support layers. These factors make it difficult to verify unified mechanisms and degradation processes that affect DA-SOFCs, thus, it provided the motivation to summarize and review overall the degradation behaviors that affect DA-SOFCs from multiple perspectives.

Fig. 1 (a) Schematic of the reaction in a DA-SOFC, (b) theoretical efficiency of H₂-SOFCs and DA-SOFCs, reprinted with permission.³³ Copyright 2022, Elsevier. (c) Electrochemical performance of DA-SOFCs (color: magenta) and H₂-SOFCs (color: blue), and (d) degradation rates of DA-SOFCs during long-term operation.

Table 1 Summary of anodes, electrolytes, and cathodes used for DA-SOFCs and their power density

Anode	Electrolyte	Cathode	Power density in H2 (mW cm^−2)	Power density in NH3 (mW cm^−2)	Temperature (°C)	Year	Ref.
Pd	BaZr0.1Ce0.7Y0.2O3−δ	LSCF	810	580	600	2017	37
			490	340	550
			240	210	500
Ni–BZCYYb	BaCe0.7Zr0.1Y0.1Yb0.1Ni0.04O3−δ	BaCo0.4Fe0.4Zr0.1Y0.1O3−δ	944	877	650	2021	38
			680	638	600
			442	445	550
			267	255	500
Ni–GDC	YSZ	LSCF/LSC	1961	1330	650	2022	39
			1587	557	600
			1220	342	550
			833	204	500
Ni–YSZ	YSZ	Pr2Zr2O7–YSZ	1220	1220	800	2022	40
			780	760	700
					600
Pr0.6Sr0.4Co0.2Fe0.75Ru0.05O3−δ	Sm0.2Ce0.8O1.9	BaCo0.4Fe0.4Zr0.1Y0.1O3−δ	483	374	800	2022	41
			381	267	750
			280	179	700
Fe/Ni–BZCYYb	BZCYYb	PBSCF	1507	1078	700	2022	42
			1157	685	650
			776	327	600
Fe–Ni–BZCYYb	BZCYYb	PBSCF	2062	1609	700	2022	43
				1263	650
				723	600
				366	550
Pd–Ni–BZCYYb	BZCYYb	PBSCF	1059	850	600	2023	44
			826	610	550
			550	340	500
Ni–BZCYYb	BZCYYb	Ba0.62Sr0.38CoO3−δ–Pr1.44Ba0.11Sr0.45Co1.32Fe0.68O6−δ	1651	1372	600	2023	45
			1214	976	550
			843	658	500
			543	293	450
Sr(Ti, Fe, Ru)O3−δ	La0.8Sr0.2Ga0.8Mg0.2O3−δ	La0.4Ce0.6O2−δ/SrTi0.3Fe0.6Co0.1O3−δ	1430	1250	800	2024	46
			830	720	700
Ni–YSZ	YSZ	GDC/LSCF	704	709	750	2024	47
			604	593	700
			494	404	650
			350	247	600
La0.6Sr0.4Fe0.7Ni0.2Mo0.1O3−δ–SDC	SDC	La0.6Sr0.4Fe0.7Ni0.2Mo0.1O3−δ–SDC	618	487	800	2024	48
			494	360	750
			388	225	700
			293	125	650
Ni–YSZ	YSZ	Pr2Zr1.95Sc0.05O7+δ–60YSZ	1460	1450	800	2024	49
			920	900	700
			470	440	600
Ni–BCZYYb	BCZYYb	PBSCF	1350	1140	600	2024	50
			925	758	550
			590	440	500
			353	208	450
Ru–Ni–BZCYYb	BZCYYb	PrBaCo2O5+δ	576	457	650	2024	51
			365	296	600
			236	194	550
			143	110	500
20Co–80BaZr0.8Y0.2O3−δ/Ni–BZCYYb	BZCYYb	PBSCF	2147	1788	700	2024	52
				1387	650
				974	600
				626	550
Fe/Ni–BZCYYb	BZCYYb	PBSCF–BZCYYb	724	608.7	700	2024	53
			561	450.7	650
			384	300	600
			242	167.5	550

---

Table 2 Summary of anodes, electrolytes, and cathodes used for DA-SOFCs and their degradation rates

Anode	Electrolyte	Cathode	Temperature (°C)	Operating condition	Degradation rate (% per 100 h)	Year	Ref.
Ni–YSZ	YSZ	La1−xSrxMnO3–YSZ	850	1 A cm^−2	0.47	2007	54
Ni–GDC	GDC	Ba0.9Co0.7Fe0.2Nb0.1O3−δ	600	0.6 V	0.27	2013	55
Ni–YSZ	YSZ	LSCF	700	0.2 A cm^−2	20	2017	56
Ni–YSZ	YSZ	LSCF	600	0.2 A cm^−2	2.3
Ni–YSZ	YSZ	CeO2/LSCF	700	0.3 A cm^−2	8.7	2019	57
Ni–SDC	SDC	Ba0.5Sr0.5Co0.8Fe0.2O3−δ	700	0.1 A cm^−2	25.2	2020	58
Ni–3YSZ/Ni–8YSZ	8YSZ	GDC/LSCF	750	0.2 A cm^−2	8.8	2020	59
Ni–Ba(Zr0.1Ce0.7Y0.1Yb0.1)0.95Pd0.05O3−δ	BZCYYb	BaCo0.4Fe0.4Zr0.1Y0.1O3−δ	550	0.2 A cm^−2	6.8	2021	60
Pr0.6Sr0.4Co0.2Fe0.75Ru0.05O3−δ	SDC	BaCo0.4Fe0.4Zr0.1Y0.1O3−δ	700	0.1 A cm^−2	6.5	2022	41
Ni–BZCYYb	BZCYYb	PBSCF	650	0.5 A cm^−2	3.9	2022	43
NiFe–BZCYYb	BZCYYb	PBSCF	650	0.5 A cm^−2	10.7
Ni–YSZ	YSZ	GDC/PrBa0.8Ca0.2Co2O6−δ	700	0.5 A cm^−2	29.4	2023	61
CeO2–Ni–YSZ	YSZ	GDC/PrBa0.8Ca0.2Co2O6−δ	700	0.5 A cm^−2	12.7
Sr2Fe1.35Mo0.45Cu0.2O6−δ + Ni–BZCYYb	BZCYYb	PBSCF	650	0.5 A cm^−2	4.67	2023	62
Ni–BZCYYb	BZCYYb	PBSCF	650	0.5 A cm^−2	12.5
Ni–BZCYYb	BZCYYb	LSCF	650	0.32 A cm^−2	1.4	2023	63
Fe–Ni–BZCYYb	BZCYYb	LSCF	650	0.36 A cm^−2	2.47
Fe/Ni–BZCYYb	BZCYYb	PBSCF–BZCYYb	650	0.75 V	12	2023	53
Ni–BZCYYb	BZCYYb	PBSCF	500	0.3 A cm^−2	17	2023	44
Pd–Ni–BZCYYb	BZCYYb	PBSCF	500	0.3 A cm^−2	2
Sr1.9Fe0.4Ni0.1Mo0.5O6−δ–GDC	GDC	Sr1.9Fe0.4Ni0.1Mo0.5O6−δ–GDC	800	0.1 A cm^−2	0.48	2024	64
Fe–Sr(Ti, Fe, Ru)O3−δ	La0.8Sr0.2Ga0.8Mg0.2O2−δ	SrTi0.3Fe0.6Co0.1O3−δ	800	0.75 V	0.6	2024	46
Fe–Ni–GDC	La0.8Sr0.2Ga0.8Mg0.2O3−δ	SrTi0.3Fe0.6Co0.1O3−δ	800	0.75 V	28.3
Ni–YSZ	YSZ	GDC/LSCF	600	0.7 V	17.2	2024	47
20Co–80BaZr0.8Y0.2O3−δ/Ni–BZCYYb	BZCYYb	PBSCF	550	0.2 A cm^−2	4.7	2024	52
Ni–BZCYYb	BZCYYb	PBSCF	550	0.2 A cm^−2	15.6
Ni–BZCYYb	BZCYYb	PBSCF	550	0.2 A cm^−2	9.8	2024	65
Ru0.95Cu0.05Nix–Ni–BZCYYb	BZCYYb	PBSCF	550	0.2 A cm^−2	1.36

---

In this review, we comprehensively discuss the degradation factors and corresponding effects on DA-SOFCs. In particular, the degradation factors in DA-SOFCs can be organized as mechanical and chemical degradations of materials and/or cells. Mechanical degradation includes mechanical cracks, nickel coarsening, and delamination and chemical degradation includes nickel nitridation, and nickel oxidation. To consider these degradation factors from various perspectives, we categorize these factors in terms of materials, cells, and systems, as shown in Fig. 2. First, material aspects such as the catalyst, catalyst support (electrolyte), and BOP are considered. During the NH₃ decomposition process, adsorbed nitrogen or other byproducts can facilitate outcomes such as metal nitriding and oxidation, resulting in chemical, electrochemical, and mechanical degradation. Second, with regard to cell aspects, there are two representative types of SOFCs, anode-supported cells (ASCs) and electrolyte-supported cells (ESCs), as well as two representative charge carriers in SOFC, proton (H⁺) and oxygen ion (O²⁻). According to these cell types and charge carriers, the byproducts associated with the electrochemical reactions and operating temperature ranges differ, significantly affecting the degradation behaviors of the DA-SOFCs. Finally, with regard to system aspects, computational approaches such as computational fluid dynamics (CFD) and/or machine learning (ML) techniques have been applied to estimate degradation in large-cells and stack systems. Discussing recent advances in this regard can provide motivation and research guidelines by which to advance toward high performance capabilities and sustainability of the DA-SOFCs.

Fig. 2 Schematic diagram of degradation factors in direct ammonia solid oxide fuel cells.

2. Material aspects

2.1. Catalysts

In the fuel electrodes of SOFCs, Ni has been widely used as a catalyst for the HOR. In addition to common SOFCs, DA-SOFCs are also compatible with Ni because it is the second most active catalyst for ammonia decomposition after Ru.^30,32,70 The high decomposition rate of Ni, exceeding 90% above 600 °C, and the high composition in a fuel electrode (>50%) enables successful DA-SOFC operation.^71,72 Therefore, DA-SOFCs have significant advantages, which can be utilized in a given structure without the addition of other noble catalysts to improve ammonia decomposition. However, in addition to the HOR process, the ammonia decomposition process concurrently occurs at the fuel electrode, involving ammonia adsorption, dehydrogenation and the desorption of hydrogen and nitrogen on the catalyst or the catalyst support, as expressed by eqn (1)–(7). While oxygen can conduct through the electrolyte to the fuel electrode, direct oxidation of ammonia can occur with reacting ammonia and oxygen ions, resulting in NO_x production, as shown in eqn (8) and (9).³⁴ These complicated processes, occurring in the fuel electrode, can lead to unexpected chemical reactions.⁷³

Overall reaction

2NH₃ ↔ N₂ + 3H₂ (1)

Ammonia decomposition mechanism

NH* + * ↔ N* + H* (5)

2H* ↔ H₂ + 2* (6)

2N* ↔ N₂ + 2* (7)

Direct ammonia oxidation reaction

2NH₃ + 5O²⁻ → 2NO + 3H₂O + 10e⁻ (8)

N* + xO²⁻ → NO_x + 2xe⁻ (9)

2.1.1. Nickel nitridation. A typical degradation factor that has been reported in relation to DA-SOFCs is the nitridation of nickel, described here using eqn (10) and (11).^42,74 As shown in Fig. 3(a), nickel nitridation can be induced during the ammonia decomposition process in an NH₃ atmosphere. Adsorbed ammonia on a nickel catalyst starts to decompose, forming NH_x and H atoms due to dehydrogenation reactions, with N* finally remaining on the catalyst surface. At high temperature, if nitrogen desorption is not facilitated, N* on the nickel surface can diffuse into the metal, and nickel nitride (Ni₃N) formation can occur. According to density functional theory (DFT) calculations, nitrogen desorption is the rate-determining step with an activation energy of 1.86 eV during the ammonia decomposition reaction on the Ni (111) surface, weakening Ni against nickel nitride formation.⁷⁵ Nickel nitride formation reduces the electrical conductivity through the electrode, with approximately a 10⁴-fold higher resistivity of nearly 7.5 × 10⁻⁴ Ω m compared to that of pure nickel at 7.5 × 10⁻⁸ Ω m at a temperature of 300 K.^76,77 In addition, the lower ammonia decomposition rate of Ni₃N compared to that of nickel can result in a drop of the open-circuit voltage (OCV) and a large concentration loss due to the insufficient hydrogen supply at the fuel electrode.^71–73

3Ni + N* → Ni₃N (10)

6Ni + 2NH₃ → 2Ni₃N + 3H₂ (11)

2Ni₃N → 6Ni + N₂ (12)

2Ni₃N + 3H₂ → 6Ni + 2NH₃ (13)

Fig. 3 (a) Schematic of the nitriding reaction in Ni of a fuel electrode. (b) Gibbs free energy changes during the formation (ΔG_for (eV)) and decomposition (ΔG_dec (eV)) of Ni₃N from 600 °C to 700 °C, reproduced with permission.⁴² Copyright 2022, Elsevier. (c) XRD pattern of a Ni film on a YSZ disk after exposure to NH₃ (100 sccm) at 600 °C for 1 h, reprinted with permission.⁷⁴ Copyright 2015, American Chemical Society. (d) Backscattered scanning electron microscopy (SEM) image of the anode surface in the fuel inlet region after the operation with ammonia-based fuels at 700–800 °C, reprinted with permission.⁵⁷ Copyright 2019, Elsevier.

However, nickel nitridation becomes energetically unfavorable at high temperatures, as shown in Fig. 3(b).⁴² At a high temperature, nitride becomes unstable and reduces back to Ni, as expressed by eqn (12) and (13).⁷⁴ Therefore, high operating temperatures of SOFCs, commonly above 750 °C, may allow the degradation caused by the nickel nitridation to be avoided. Despite the fact that SOFCs with H₂ fuel are generally more stable at lower temperatures with less Sr segregation and chemical interdiffusion, DA-SOFCs are less stable at lower temperatures, as shown in Fig. 1(d).^78–81 This may be caused by the facile nickel nitridation at lower temperatures. Yang et al. reported XRD analysis of nickel nitridation by exposing a Ni film deposited on a YSZ disk to ammonia at 600 °C for 1 h, as shown in Fig. 3(c).⁷⁴ Through this analysis, they confirmed that Ni₃N is the stable phase of nickel nitride in pure NH₃ at 600 °C. However, even if nickel nitride formed by nitridation can be reduced back to nickel, the repetition of nickel nitridation and the reduction process at a high temperature can also cause morphological changes, mechanical cracks and deterioration due to the nickel redox cycle stemming from volume differences and crystal structures of Ni₃N (76.81 ų, hcp) and pure metallic nickel (41.97 ų, fcc).^82,83 Stoeckl reported that nickel can be agglomerated and enlarged by ammonia because of nickel nitriding, as shown in Fig. 3(d).⁵⁷ These changes in microstructures are more significant in the gas inlet region than at the outlet, which is more exposed to high concentrations of ammonia. And Yang et al. reported that mechanical cracking can occur in DA-SOFCs due to nickel nitridation after thermal cycling in the range of 600–700 °C every 3–4 h for 20 h under an ammonia atmosphere, resulting in an OCV drop.⁷⁴ Furthermore, although nickel is more stable with regard to nickel nitridation at high temperatures, other degradation factors, such as Sr segregation, Cr poisoning and chemical interdiffusion, can be facilitated.^34,84,85 Therefore, understanding the nickel nitridation process and finding a means of suppressing it at lower temperatures can concurrently reduce all degradation behaviors.^60,86,87 Because the factors inducing nickel nitridation are sluggish ammonia decomposition kinetics and nitrogen remaining on the catalyst surface, designing a new catalyst that can expedite the surface kinetics and change the rate-determining step (RDS) can be a strategic approach by which to realize sustainable DA-SOFC operation in a wide range of operating temperatures.

To improve the ammonia decomposition kinetics, the nitrogen adsorption energy should be considered. As shown in Fig. 4(a), Boisen et al. reported ammonia conversion calculated at 773 K for various transition-metal catalysts.⁸⁸ One of the important factors to determine a better catalyst for ammonia decomposition is considering N₂ adsorption energy. Ru has been reported as the best metal catalyst for ammonia decomposition. In addition, non-noble metals such as Co, Ni, Fe, and Co₃Mo₃N also show comparable catalytic performance on the ammonia decomposition. Noble metals such as Rh, Ir, Pd, and Pt also have good catalytic performance.^90–92 Rathore et al. reported a Pd-infiltrated Ag-LSCF fuel electrode created by mixing LSCF perovskite and silver powder utilized as the fuel electrode and adding a Pd precursor solution to enhance ammonia dissociation by allowing faster dissolution of hydrogen.⁹³ Before Pd infiltration, it showed 15% lower electrochemical performance in ammonia fuel than hydrogen at 800 °C, but after infiltration, it showed electrochemical performance in NH₃ fuel nearly identical to that of H₂. Zhu et al. and Li et al. fabricated Ru-based catalysts to improve the ammonia decomposition rate of DA-SOFCs, showing improved catalytic and electrochemical performance and durability outcomes of DA-SOFCs.^38,94 And Yang et al. reported the ammonia conversion of the Ru-infiltrated Ni–YSZ anode for DA-SOFCs.³⁶ Adding Ru as a catalyst to NiO–YSZ improves the ammonia conversion rate, even with 0.2 wt% of Ru (∼50% at 600 °C) compared to bare Ni–YSZ (∼20% at 600 °C).

Fig. 4 (a) Calculated turnover frequencies of ammonia synthesis/decomposition at 773 K as a function of the dissociative N₂ adsorption energy, reprinted with permission.⁸⁸ Copyright 2005, Elsevier. (b) Potential energy diagrams of ammonia decomposition on Pt, Ni/Pt and Ni–Pt at 0 K calculated by DFT, reprinted with permission.⁸⁹ Copyright 2015, Springer Nature.

However, because noble metals have very low cost-competitiveness, a bimetallic catalyst composed of noble and/or transition metals can provide synergetic effects regarding catalytic performance capabilities more economically.^95–97 Vlachos et al. reported a Pt–Ni bimetallic catalyst for ammonia decomposition that outperformed pure Ni and Pt types according to multiscale steady-state simulations.⁸⁹ In their simulations, Ni/Pt served as a bifunctional catalyst, where the Ni terrace sites catalyze the N–H bond scission and the (110) edges of Ni/Pt patches catalyze the N₂ association. Especially according to DFT calculations, the Ni/Pt bimetallic material is more active with regard to dehydrogenation and has a lower N–N association barrier (1.83 eV) compared to pure Ni (2.10 eV) and Pt (1.96 eV), as shown in Fig. 4(b). He et al. reported a Ni_xCo_10−x/CeO₂ bimetallic catalyst for ammonia decomposition as a function of the ratio of nickel to cobalt, outperforming pure Ni and Co.⁹⁸ At 650 °C, the NC-0/CeO₂ (Ni : Co = 10 : 0) sample shows an NH₃ conversion rate of only 65.8%, whereas the NC-2.5/CeO₂ (Ni : Co = 7.5 : 2.5) sample shows a corresponding rate of 96.96%. DFT calculations show that the CoNi (111) bimetallic material has a lower activation energy for the N–H cleavage process and the N association reaction, with an activation energy level of 1.73 eV, compared to pure Ni (111), for which the activation energy level is 1.90 eV for N association. Guo et al. calculated the turnover frequencies (TOFs) of ammonia by first-principles calculations and ab initio molecular dynamics.⁹⁹ They found that the metal–nitrogen (M–N) binding energy of 134 kcal mol⁻¹ exhibited the best ammonia decomposition catalytic performance. In their study, Ru single-metal samples exhibited the highest TOF, followed by Fe–Pt and Fe–Ir bimetallic samples, which showed an M–N binding energy close to 134 kcal mol⁻¹. Given that the constituent metals in the bimetallic catalyst change the RDS and process during ammonia decomposition, the ammonia decomposition kinetics can be enhanced.

High-entropy alloy (HEA) catalysts, composed of more than four or five elements, can also significantly maximize the synergetic effects of each component.^100–103 Xie et al. reported HEA catalysts for high-efficiency ammonia decomposition with earth-abundant elements.¹⁰⁴ They fabricated HEA CoMoFeNiCu nanoparticles with control of the Co/Mo atomic ratio. DFT calculations showed that increasing the proportion of cobalt decreased the desorption energy barrier of N and increased the kinetic barrier of dehydrogenation in NH₃, which became the RDS. In contrast, as the Mo ratio was increased, the kinetic barrier of dehydrogenation increased, and the energy barrier of desorption of N decreased. The optimized novel catalyst Co₂₅Mo₄₅Fe₁₀Ni₁₀Cu₁₀ achieved an NH₃ conversion rate of 100% at ∼525 °C, which is even higher than that of metallic Ru (which showed a 73% conversion rate at 600 °C).

In addition, a pre-surface treatment of the metal catalyst with nitrogen to form a metal nitride can suppress severe nickel nitridation.^105,106 Hashinokuchi et al. fabricated Ni–Mo anodes by impregnating a Mo precursor into Ni–SDC anodes and then exposing them to ammonia in the DA-SOFC operating environment to synthesize metal nitrides for Ni_xMo_1−xN.¹⁰⁷ The Ni_xMo_1−xN/SDC cell exhibited approximately a 1.6-fold increase in its electrochemical performance under an ammonia atmosphere, with an outcome of approximately 250 mW cm⁻² for a bare cell and 400 mW cm⁻² for the treated cell at 900 °C. Sorcar et al. reported that a LaTiO₂N fuel electrode with 5% Cu added in a DA-SOFC showed electrochemical performance capabilities of approximately 228 mW cm⁻² at 750 °C and high stability under thermal cycling between 600 and 700 °C compared to the corresponding outcomes of the Ni–yttria-stabilized zirconia (Ni–YSZ) electrode.¹⁰⁶

2.1.2. Nickel oxidation. In addition to nickel nitridation, nickel oxidation can also cause the degradation of fuel electrodes in DA-SOFCs.^109–111 Because oxygen ions are conducted from the air electrode (cathode) to the fuel electrode, the conducted oxygen ions can facilitate nickel oxidation, as expressed by eqn (14).^112,113 In particular, under an ammonia environment, the possible reaction processes of oxidation are still veiled, such as NO_x production, mentioned in the Materials section, and other oxidation reactions. In addition, inevitable water evolution through the HOR in the fuel electrode can complicate the process. Ya et al. reported that nickel in the fuel electrode was oxidized at a low temperature (550 °C) by oxygen ions conducting through the electrolyte from the air electrode in an ammonia atmosphere, as shown in Fig. 5(a).¹⁰⁸ At a high temperature of 700 °C, the decomposition rate of ammonia at the fuel electrode was increased by oxygen ions from the air electrode, which promotes H₂ dissociation from the active sites. Thereby, it enhances the decomposition rate of ammonia at high temperatures due to an increase in oxygen partial pressure in the cathode. In contrast, at a low temperature of 550 °C, the excess oxygen conducted from the air electrode facilitated the oxidation of the nickel, resulting in a decrease in the ammonia decomposition rate. And Lee et al. demonstrated the effect of water generated in the fuel electrode on nickel oxidation in the operating environment of DA-SOFCs.⁴⁷ During a 50 h stability test at a temperature of 600 °C, the degradation of electrochemical performance was accelerated in an ammonia atmosphere compared to hydrogen, especially in a wet environment compared to a dry environment. As shown in Fig. 5(b), according to the XRD analysis, a Ni–YSZ fuel electrode under wet NH₃ shows a high intensity of the NiO peak. With Rietveld refinement, a dry H₂/N₂ sample shows that only 4.5% of nickel was oxidized, whereas a wet NH₃ sample shows 53%. And DFT calculations also verified that adsorbed N on the surface of Ni expedites H₂O decomposition, resulting in a high concentration of OH groups on the Ni surface. Thus, an OH group on the Ni surface can cause Ni oxidation due to the inevitable water evolution in the fuel electrode of DA-SOFCs.¹¹⁴ Given that nickel oxide undergoes significantly less ammonia decomposition and electrical conductivity compared to pure metallic nickel, it causes the performance to degrade over long-term operation.^59,115,116 In addition, because ammonia is also a highly reducing substance, there is the possibility of repeated re-oxidation and reduction, which can impart mechanical stress on the cell, resulting in delamination or cracking.⁷⁴ Therefore, changing the operating conditions and environment, optimizing the internal microstructure of the fuel electrode, inserting catalysts into the fuel electrode, or synthesizing new materials are required to reduce the re-oxidation of Ni. Nickel reoxidation can cause mechanical stress, resulting in cracks, due to the different molar volumes of pure metallic nickel and nickel oxide.^111,117,118 Pihlatie et al. reported that re-designing the fuel electrode with high porosity can reduce mechanical stress.¹¹¹ In addition, Welander et al. reported that aluminate spinel (NiAl₂O₄) and zirconium titanate (Zr₅Ti₇O₂₄) can suppress Ni coarsening by the nickel reoxidation.¹¹⁸ Furthermore, in the redox cycling test, the NiAl₂O₄ decorated fuel electrode shows lower degradation rates of less than 0.3% h⁻¹ after 12 hours, whereas, the undecorated samples show degradation rates of more than 0.6% h⁻¹.

Fig. 5 (a) Schematics of chemical changes in Ni in the Ni–YSZ fuel electrode depending on the DA-SOFC environment and operating conditions, reprinted with permission.¹⁰⁸ Copyright 2024, Elsevier. (b) XRD patterns of Ni–YSZ fuel electrode surfaces in the initial state and after a 50 h accelerated stability test to verify the changes in the chemical phase of nickel under dry H₂/N₂, wet H₂/N₂, dry NH₃, and wet NH₃ fuel conditions, respectively, reprinted with permission.⁴⁷ Copyright 2024, Elsevier.

2.2. Catalyst supports

Given that catalyst supports also have a considerable influence on the increase in the ammonia decomposition reactivity, they can also affect the degradation behaviors in DA-SOFCs.¹¹⁹ Depending on the catalyst support, the adsorption energy of the gas on the catalyst surface can be controlled, which leads to a change in the reaction pathway or the rate-determining step (RDS). For example, Miyazaki et al. demonstrated ammonia decomposition according to the catalyst support in DA-SOFCs.¹²⁰ As shown in Fig. 6(a), Ni/BaCe_0.4Zr_0.4Y_0.2O_3−δ (BCZY) showed the highest ammonia decomposition rate among Ni/gadolinium-doped ceria (GDC) and Ni/YSZ. Ni/BCZY has a decomposition rate close to 100% at 600 °C, while Ni/GDC and Ni/YSZ have a decomposition rate of less than 40%. This difference in the ammonia decomposition rate is due to the effect of the basicity of the catalyst support. A catalyst support with high basicity can actively interact with ammonia, expediting the decomposition of ammonia. The strong basicity of Ba in catalyst supports such as BCZY can promote charge transfer reactions that provide electrons to the adsorbed gas, resulting in active ammonia decomposition. In good agreement with the findings of Miyazaki et al., when the ammonia decomposition rate was compared according to the basicity of doped cations in ZrO₂ (Fig. 6(b)), Ba, which has the lowest Smith acidity, i.e., the strongest basicity, presented the highest ammonia decomposition rate among Sr- and Ca-doped ZrO₂.^121,122 Therefore, modification of the fuel electrode with basic materials can activate ammonia decomposition and improve the durability of DA-SOFCs. Eguchi et al. reported that Ba–Ni/YSZ fuel electrodes could be fabricated by mixing barium and Ni, with high basicity relative to those of existing conventional Ni–YSZ-based SOFC fuel electrode materials, to increase the ammonia decomposition rate.⁵⁹ While the NH₃ decomposition rate of bare Ni–YSZ at about 600 °C was as low as 40%, Ba–Ni/YSZ showed a decomposition rate of 100%. In addition, when the bare Ni/YSZ sample was operated at 750 °C and at 200 mA cm⁻² for 50 h, the electrochemical performance decrease was 1.32%, whereas the performance of the Ba–Ni/YSZ sample decreased by only 0.47%.

Fig. 6 (a) Ammonia conversion rate of Ni–BCZY, GDC, and YSZ cermet. Reaction condition: 100% NH₃, reprinted with permission.¹²⁰ Copyright 2020, Royal Society of Chemistry. (b) Ammonia conversion as a function of the reaction temperature of 40 wt% Ni/ZrO₂, Ni/CaZrO₃, Ni/SrZrO₃, and Ni/BaZrO₃. Reaction condition: 100 vol% NH₃, reprinted with permission.¹²¹ Copyright 2018, Royal Society of Chemistry.

2.3. Physical issues (delamination and agglomeration)

An ammonia environment changes not only the chemical composition but also the mechanical structure by the delamination and agglomeration of both the catalyst and the catalyst support. In SOFCs, the cathode, electrolyte, and anode should be well bonded to reduce the contact resistance.^87,123–125 However, given that nickel, nickel nitride, and nickel oxide have different molar volumes of 41.97 ų, 76.81 ų and 73.37 ų, respectively, chemical changes can also induce delamination, thermal and mechanical stress, or the agglomeration of a porous fuel electrode, resulting in cracks and/or undesired changes in the microstructure of the fuel electrode.¹²⁶ Furthermore, ammonia is a strong reducing agent of metal and support materials, resulting in abrupt changes in microstructures.¹²⁷ Zhu et al. reported that the fuel electrode (Ni/BaCe_0.7Zr_0.1Y_0.1Yb_0.1O_3−δ (BZCYYb)) of DA-SOFCs in their study was significantly reduced in an NH₃ atmosphere.³⁸ Through strong reduction under an NH₃ atmosphere at 600 °C, the grains of the BZCYYb support were detached, as shown in Fig. 7(a) and (b). This gap can lead to the separation of the ionic conductor, which increases the ohmic resistance of the cell and degrades its electrochemical performance. In addition, Ni also became more porous under harsh reduction conditions, making it difficult to secure the microstructure. Stoeckl et al. also reported the degradation of Ni–YSZ fuel electrodes and Ni current collectors, as shown in Fig. 7(d).⁵⁷ These reports confirm that the ammonia decomposition process clearly affects the microstructure of nickel, such as the particle size and porosity, leading to a decrease in the number of active sites.^128–130 Yang et al. reported Ni–YSZ anode-supported DA-SOFCs subjected to thermal cycling tests between 600 and 700 °C.⁷⁴ Under hydrogen fuel, the OCV was well maintained over six thermal cycles in 20 h. However, the OCV tended to decrease gradually in an ammonia environment. A post-mortem analysis of the samples showed that the OCV decreased due to gas leakage caused by cracks in the fuel electrode of the DA-SOFC, as shown in Fig. 7(e).

Fig. 7 SEM images of (a and b) BCZYYb (1 wt% NiO) and (c) NiO after non-cracked NH₃ reduction, reprinted with permission.³⁸ Copyright 2021, Springer Nature. (d) Backscattered SEM image of nickel contact mesh at the fuel inlet after operation with ammonia-based fuels at 700–800 °C, reprinted with permission.⁵⁷ Copyright 2018, Elsevier. (e) Illustration of an anode supported cell after a temperature cycling test in NH₃ in the range of 600–700 °C for 20 h, reprinted with permission.⁷⁴ Copyright 2015, American Chemical Society.

2.4. Interconnect materials

NH₃ affects not only the fuel electrode but also the interconnector and current collector, where the flow channel is designed to receive fuel. Because interconnectors are composed of SUS or Inconel, which are alloys containing Cr, Fe, and Ni, they are susceptible to nitriding.^131–133 Under an ammonia atmosphere, it has been reported that interconnectors degrade significantly in the form of peeling and mechanical cracking due to volume expansion/contraction, composition reconstruction, and phase collapse.¹³⁴ In particular, given that metal interconnectors are also ammonia-decomposable, nitriding can be accelerated by residual ammonia, especially due to the nitrogen adsorbed on the surface. Yang et al. compared the surface and chemical phase changes in SUS430 and Inconel 600 under hydrogen and ammonia atmospheres, respectively, as shown in Fig. 8(a)–(d).¹³⁴ The surface areas of SUS430 and Inconel 600 increased rapidly after exposure to an ammonia atmosphere due to metal nitridation. Through an XRD analysis, it was found that Fe₃N and CrN were formed by ammonia. More importantly, the changes in the structural and chemical compositions of these interconnectors resulted in a decrease in the ammonia decomposition rate. Kishimoto et al. also reported that after the operation of a DA-SOFC stack at 750 °C for 1000 h, surface changes of the SUS430 interconnect were observed, as shown in Fig. 8(e)–(h).³⁵ SEM images and results from an electron probe microanalyzer showed that Fe was phase-separated toward the surface and that nitrogen became incorporated into the metal after exposure to the ammonia atmosphere for 1000 h, indicating that nitridation induces structural and chemical changes concurrently on the surface and in the bulk. In addition, Stoeckl et al. reported that a bipolar plate composed of chromium iron yttrium in DA-SOFCs demonstrated the formation of substantial amounts of chromium nitride after operation for 1000 h.¹³⁵ Because these structural and chemical changes in the interconnector due to ammonia detrimentally affect the stability of DA-SOFCs, it is necessary to design new interconnectors by selecting materials that are not susceptible to nitridation or to improve the durability by using coating technologies such as sputtering, electrodeposition, or atomic layer deposition (ALD) on existing interconnect materials with low nitridation reactivity levels.^136–139

Fig. 8 SEM images of (a) SUS430 powder after 750 °C H₂ reduction (H₂-SUS430), (b) SUS430 powder following an ammonia decomposition test (NH₃-SUS430), (c) IN600 powder after 750 °C H₂ reduction (H₂-IN600), and (d) IN600 powder following the ammonia decomposition test (NH₃-IN600), reprinted with permission.¹³⁴ Copyright 2024, Elsevier. Cross-sectional images of the separator on the anode side after a 1000 h durability test with direct ammonia fuel at 750 °C, (e) SEM image, and the distribution of (f) Cr, (g) Fe, and (h) N, reprinted with permission.³⁵ Copyright 2020, John Wiley and Sons.

In addition to metallic interconnectors, the sealants to ensure gas tightness between air electrodes and fuel electrodes in DA-SOFCs can deteriorate under an ammonia atmosphere. The composition of the glass sealants used, Al₂O₃, means that it can react with undesirable chemical species, such as Sr from an air electrode, Cr from a metallic interconnector, and sulfur from the air or fuel.^60,140,141 However, it is unclear whether the N-source from the ammonia affects the degradation of glass sealants.

3. Cell aspects

For the development of high-performance and sustainable DA-SOFCs, not only the materials but also the cell types should be considered. In particular, SOFCs are categorized into O-SOFCs (SOFCs), which conduct oxygen ions (O²⁻), and H-SOFCs (protonic ceramic fuel cells, PCFCs), which conduct protons (H⁺), depending on the defects in the electrolyte.^71,142,143 According to the charge carrier, the byproducts at the fuel electrode differ. For example, SOFCs produce water as a byproduct during the HOR at the fuel electrode. In contrast, PCFCs produce water not at the fuel electrode but at the air electrode, as shown in Fig. 9. As discussed in Section 2.1.2, given that oxygen ion and water can induce nickel oxidation, the degradation behavior depending on the charge carrier can differ. In addition, depending on the type of support, anode-supported cells (ASCs) and electrolyte-supported cells (ESCs) can be distinguished. As the fabrication processes are different between ASCs and ESCs, the thickness, porosity, materials, composition and mechanical strength of the fuel electrode layer are also different, thus affecting the ammonia decomposition rate.^144–147 Furthermore, according to the types of charge carriers and supporting layers, the target operating temperature varies, significantly affecting the catalytic and electrochemical performance and stability factors. In this regard, performance and degradation behavior of DA-SOFCs as a function of the type of charge carriers and the type of supporting layers are discussed below, after which we introduce examples of cell modifications caused by the insertion of catalysts into cells or by the changing of cell structures in various ways to improve the performance and durability.

Fig. 9 Schematic of the operating reactions in the (a) DA-SOFC and (b) DA-PCFC.

3.1. Effects of charge carriers

The major difference between SOFCs and PCFCs is the charge carrier that conducts through the electrolyte. The mechanism of operation and the direction of ion transport, depending on the type of charge carrier in SOFCs and PCFCs, are shown in Fig. 9. O²⁻ ions are conducted from the air electrode in a SOFC, and H₂O is generated through the HOR at the fuel electrode, while in a PCFC, protons are conducted through the electrolyte, and H₂O is generated at the air electrode.⁷¹ Therefore, a fuel dilution effect at the fuel electrode may occur in SOFCs compared to PCFCs, and the performance and degradation may be affected by the conducted O²⁻ and H₂O.^71,148–150 As we addressed in the previous section, the presence of water or oxygen ions can facilitate nickel oxidation. In contrast, given that PCFCs do not conduct O²⁻ toward the fuel electrode, nickel oxidation is avoidable.

There are material differences in the electrolytes used in SOFCs and PCFCs. Materials that conduct oxygen ions are typically fluorite-based materials such as YSZ, GDC, and scandium-stabilized zirconia. On the other hand, materials that conduct protons are perovskite-based materials containing mainly Ba, such as yttrium-doped barium cerate (BCY), yttrium-doped barium zirconate (BZY), and BZCYYb.^151–153 Because protons are approximately 1.6 × 10⁵ times smaller (0.00087 pm) than oxygen ions (140 pm), proton conductors commonly have higher ionic conductivity. For example, the ionic conductivity of the proton conductor BZCYYb is higher than that of YSZ (9.0 × 10⁻³ S cm⁻¹ for YSZ at 700 °C and 1.9 × 10⁻² S cm⁻¹ for BZCYYb at 500 °C).^154,155 Therefore, both higher ionic conductivity and higher basicity from the Ba enable H-SOFCs to operate at lower temperatures below 500 °C. Jeong et al. and Yun et al. demonstrated that a high-performance PCFC with direct ammonia fuel could operate below 500 °C.^44,50 Without catalyst modification, Jeong et al. and Yun et al. achieved power densities of 0.34 and 0.44 W cm⁻² at 500 °C, respectively. However, because nickel nitridation is facilitated at lower temperatures, the stability issue remains. Jeong et al. demonstrated rapidly decreased electrochemical performance with a degradation rate of 0.17% h⁻¹ at 0.3 A cm⁻² and 500 °C.⁴⁴ In addition, Zhu et al. demonstrated a significant durability issue with a PCFC with direct ammonia fuel compared to injection of cracked ammonia from an external reformer.³⁸ When reformed ammonia was injected via an external reformer at 400 mA cm⁻² and 600 °C, the PCFC ran smoothly almost without degradation for 300 h, while the electrochemical performance of a DA-PCFC at 500 mA cm⁻² and 650 °C dropped significantly over 15 h.

In SOFCs, O²⁻ and H₂O generated by electrochemical reactions in the SOFC fuel electrodes have the potential to influence nickel nitriding, oxidation, and other undesired chemical reactions. First, NO_x can be generated as a byproduct during operation of the cell by direct oxidation of NH₃, as mentioned in eqn (8) and (9).¹⁵⁶ Yang et al. reported that SOFCs operated at 600 °C under fuel conditions of NH₃: 42.9%, H₂O: 1.4% and N₂: 55.7% at current densities of 0.07, 0.11, and 0.14 A cm⁻² generated the NO_x below 20 ppm in all cases.¹⁵⁷ And Yang et al. reported a trend in the amount of NO_x formed for each fuel utilization rate and temperature.¹³⁴ According to this study, NO_x generation in DA-SOFCs is relatively small in the range of 1 × 10⁻³⁵ to 1 × 10⁻⁵ ppm compared to that in combustion.¹⁵⁸ In addition, according to Singh et al., under DA-SOFC operation at 550 °C, nitriding was not detected in the XRD peak according to a post-mortem analysis.⁵⁶ In contrast, Yang et al., reported that when a Ni film on the YSZ disk was exposed to ammonia at 100 sccm for 1 h at 600 °C without electrochemical operation, as mentioned in Fig. 3(c), nickel nitriding was detected in the XRD peaks.⁷⁴ Despite the fact that the detailed mechanisms affected by the conducted O²⁻ have not been unveiled, it can be speculated that adsorbed nitrogen (N*) can be desorbed by the evolution of the NO_x through the conducted O²⁻ according to eqn (8) and (9). Thus, nickel nitridation could be suppressed under DA-SOFC operating conditions, while only an ammonia-exposed condition could facilitate nickel nitridation. However, since O²⁻ is not present in the fuel electrodes of PCFCs, suppressing nickel nitridation through O²⁻ as in a SOFC is challenging.

3.2. Effects of different support layer types

As shown in Fig. 10(a) and (b), the cell type can differ depending on the support layer used for fabrication. ASCs are fabricated based on a thick anode support layer (∼500 μm), while ESCs are fabricated based on a thick electrolyte layer (∼50 μm). Because ASCs have a thin electrolyte layer (∼10 μm), they show low ohmic resistance and high electrochemical performance, resulting in a relatively low operating temperature (∼750 °C).^144,145,159 Moreover, the high Ni content (>50%) in the porous and thick fuel electrode provides better compatibility with DA-SOFC operation compared to that of ESCs by providing a sufficient reforming area. However, the drawbacks of ASCs are their mechanical strength and durability due to the high Ni content of the fuel electrode, which is susceptible to redox stability or Ni agglomeration.^160–162 In contrast, because ESCs have a thick electrolyte layer, they show high ohmic resistance and low electrochemical performance, thus generally operating at high temperatures exceeding 800 °C. Despite the fact that the high operating temperatures of ESCs can facilitate ammonia decomposition, the thin fuel electrode in ESCs does not provide a sufficient amount of catalyst for ammonia decomposition.

Fig. 10 Schematic of (a) an anode-supported cell (ASC) and (b) an electrolyte-supported cell (ESC), and the degradation rate of ASCs and ESCs fueled with (c) 60% H₂ – 5% H₂O – 35% N₂ and (d) 66.67% NH₃ – 8.33% H₂O – 25% N₂ at 0.2 A cm⁻². Temperature: 700 °C; cathode gas: 100% O₂.⁵⁶

Depending on their advantages and disadvantages, ASCs and ESCs show different degradation behaviors, as shown in Fig. 10(c) and (d). Singh et al. reported different stabilities of DA-SOFCs between ASCs and ESCs.⁵⁶ ASC and ESC samples showed stable operation with degradation rates of 0.18% h⁻¹ and 0.22% h⁻¹, respectively, in a H₂ fuel environment at 0.2 A cm⁻² for 10 h. In contrast, although the ASC still exhibited reasonable stability with a degradation rate of 0.2% h⁻¹ in the ammonia fuel environment, the ESC showed a significant voltage drop with fluctuation and degradation rate of 5.4% h⁻¹. It could be ascribed to the fact that the ASC provided sufficient hydrogen through self-ammonia decomposition in a thick fuel electrode, resulting in better stability. However, in the ESC, the hydrogen supply was not sufficient due to the lack of ammonia decomposition in the thin fuel electrode, resulting in an abrupt decrease in performance. Therefore, ASCs would be more suitable for DA-SOFCs than ESCs because a sufficient ammonia reforming layer is required for both catalytic and electrochemical performance and stability.

3.3. Cell modifications

Because all cell types have individual characteristics when operating under direct ammonia, modifications of the fuel electrode to improve compatibility have been extensively investigated. As well as alternating the Ni catalyst and enhancing the catalytic activity of the fuel electrode, surface coating techniques, including the ex-solution, infiltration, physical vacuum deposition (PVD), and ALD techniques, have been applied.^163–169

The ex-solution technique causes the doped cation to evolve toward the surface under a reducing atmosphere.^170–172 Ex-solved catalysts can provide an enlarged reaction area and anchored structure on the surface, possibly improving catalytic and electrochemical performance and stability of the ammonia decomposition in the fuel electrode. Zhong et al. fabricated Sr_1−xTi_1−yNi_yO_3−δ by doping nickel into the B-sites of the SrTiO₃ perovskite material and then ex-solved nickel particles through an exsolution process, as shown in Fig. 11(a).¹⁶⁷ The as-prepared Sr_0.9Ti_0.8Ni_0.2O_3−δ (STN0.2) exhibits a lower flat-band potential as well as a smaller conduction band and band gap and high conductivity compared to the NiO–YSZ anode, resulting in a feasible bound electron transition to a free electron. Therefore, the peak power density (PPD) correspondingly reaches 287.1 and 262.7 mW cm⁻² in H₂ and NH₃ fuels at 800 °C, respectively, showing an improvement of 1.5 times compared to a bare Ni–YSZ sample in an ammonia atmosphere. Nickel doping at the B-site promoted the reduction of Ti⁴⁺ to Ti³⁺ to form oxygen vacancies and facilitated the electron transfer process during the ammonia reaction, facilitating ammonia decomposition, as shown in Fig. 11(b).¹⁶⁷ Similarly, ex-solved FeNi₃ from a Ni-doped SFM double-perovskite (Sr_1.9Fe_0.4Ni_0.1Mo_0.5O_6−ε) fuel electrode demonstrated an ammonia decomposition rate of 97.9% in the range of 500–800 °C, also showing the good stability with a degradation rate of 0.0048% h⁻¹.⁶⁴ Shao et al. reported a novel anode material, Ni–Ba(Zr_0.1Ce_0.7Y_0.1Yb_0.1)_0.94Ru_0.03Fe_0.03O_3−δ (BZCYYbRF), developed by co-doping with Ru and Fe via a secondary redox treatment.¹⁷⁴ In the co-doped sample, Ru and Fe were ex-solved onto the surface under reducing conditions, with a secondary process further enhancing the redistribution of the nanoparticles. This secondary redox sample shows increased catalytic activity and improves both ammonia decomposition and hydrogen oxidation. The PPDs of the secondary redoxed BZCYYbRF-based cell at 650, 600, 550, and 500 °C were 807, 654, 447, and 268 mW cm⁻², respectively, which were 15% higher than those of an untreated BZCYYbRF-based cell at 650 °C, exhibiting a voltage decay rate of 0.003 V h⁻¹ during a stability test with ammonia fuel when discharging 200 mA cm⁻² at 600 °C for a continuous duration of 100 h.

Fig. 11 (a) TEM image of STN0.2 and the distribution of the ex-solved Ni particle sizes; (b) proposed reaction mechanism of a STN0.2–40YSZ anode in a NH₃-SOFC, reprinted with permission.¹⁶⁷ Copyright 2023, Elsevier. (c and d) SEM images of an Fe-infiltrated Ni–BZCYYb anode after modification and electrochemical performance of a (e) bare Ni–BZCYYb anode-based PCFC and a (f) Fe-infiltrated Fe–Ni bimetallic anode-based PCFC using H₂ or NH₃ as the fuel at 700–800 °C, reprinted with permission.⁴³ Copyright 2022, Royal Society of Chemistry. (g and h) SEM images of the catalytic CeO_x layer and ASL, reprinted with permission.¹⁷³ Copyright 2022, Elsevier.

Infiltration, which has been extensively investigated, is the simplest and easiest way to modify the catalyst surface.^{58,169,175,176} Xu et al. infiltrated CeO_2−δ onto a Ni-based cermet anode.⁶¹ A DA-SOFC with CeO_2−δ demonstrated high peak power densities of 0.941, 1.351, and 1.893 W cm⁻², while the bare sample shows 0.673, 0.997, and 1.375 W cm⁻² at 700, 750, and 800 °C, respectively, with high durability also found compared to the bare anode with retaining of the catalyst.⁶¹ Zhang et al. infiltrated iron onto a Ni/BZCYYb fuel electrode in an effort to improve improving electrochemical performance of DA-PCFCs.⁴³ Although iron demonstrates a worse ammonia decomposition rate than nickel, because it has a higher nitrogen desorption energy (2.86 eV) compared to that of nickel (1.86 eV), it can enhance nitrogen desorption during the ammonia decomposition process by forming a Ni–Fe bimetallic alloy, as shown in Fig. 11(c) and (d).^43,75 Thus, the peak power densities in the ammonia environment increased by approximately 20% compared to the bare sample, as shown in Fig. 11(e) and (f) (1.609 mW cm⁻² in the Fe-modified case and 1.398 mW cm⁻² for the bare sample at 700 °C).⁴³ Infiltrated Fe also exhibited improved stability by approximately 3.64-fold with a degradation rate of 0.0022 V h⁻¹, compared to that of the bare sample (0.008 V h⁻¹ at 650 °C and 0.5 A cm⁻²). DFT calculations confirmed that the associative desorption barriers of the adsorbed N* were reduced from 1.92 eV (Ni) to 1.80 eV (FeNi), resulting in improved electrochemical performance and durability.

In addition, Shim et al. modified a Ni/BZCYYb fuel electrode with Pd using the ALD technique.⁴⁴ Because ALD can deposit a coating layer on the sub-nanometer scale, the trade-off between performance and stability, depending on the coating layer, can be precisely controlled. The bare Ni/BZCYYb cell demonstrated PPD outcomes of 0.65, 0.38, and 0.16 W cm⁻² at 600, 550, and 500 °C, respectively. In contrast, the Pd-treated cell with 150 ALD cycles exhibited substantially improved electrochemical performance rates of 0.85, 0.61, and 0.34 W cm⁻² at 600, 550 and 500 °C, respectively. During long-term operation at 0.3 A cm⁻² and 500 °C for 100 h, the Pd-treated cell showed higher stability compared to that of the bare cell. Interestingly, in contrast to the nickel nitridation in the bare cell after the operation, the Pd-treated cell did not show nickel nitridation.

To support ammonia decomposition in a fuel electrode, installing an additional reforming layer onto the fuel electrode is also a strategic approach.^177–179 As shown in Fig. 11(g) and (h), Chen et al. applied an additional Fe–CeO_x reforming layer onto Ni–BZCYYb fuel electrodes.¹⁷³ The Fe–CeO_x layer improved the catalytic performance, showing peak power density (1.06 W cm⁻² at 700 °C) compared to that of an additional Fe-layered cell (0.78 W cm⁻² at 700 °C). It also showed better long-term stability by inhibiting nitriding by CeO_x. Li et al. fabricated a cell with an additional CeO₂-supported Ru/Ni reforming layer.⁹⁴ By applying the additional reforming layer, the cell showed greater ammonia decomposition, with an improved peak power density (60.4 mW cm⁻² at 700 °C) compared to that of a bare cell (49.5 mW cm⁻² at 700 °C).

4. Stacks and systems

To commercialize DA-SOFCs, the performance and degradation behaviors of large cells and the stack should be addressed. In the large cells and stacks, the current densities, temperature, pressure and gas composition vary along the lateral and vertical direction through the flow channel, as shown in Fig. 12(a).¹⁸⁰ Therefore, verifying and characterizing the changes in the structure and chemical characteristics at a certain position can become complicated. Therefore, studies of issues in DA-SOFC large cells and stacks necessarily accompany computational simulations. Conventionally, numerical modelling and computational fluid dynamics (CFD) have been utilized extensively, and machine-learning and/or deep neural network techniques have recently been applied to analyze the complicated multiphysics processes in DA-SOFC systems and optimize the conditions for the high efficiency and performance.^183–188

Fig. 12 Schematic of the concept of the multiscale multiphysics modeling of an ammonia fueled SOFC: (a) cell level, and (b) stack level, reprinted with permission.¹⁸⁰ Copyright 2024, Elsevier. (c-1) Distribution of the current density and (c-2) temperature in a co-flow designed planar DA-SOFC, reprinted with permission.¹⁸¹ Copyright 2023, Elsevier. SEM micrographs of the fuel support layer of the cell obtained after 1000 h of testing at (d-1) the fuel inlet and (d-2) the fuel outlet under high flow rate. Green arrows indicate instances of Ni nitriding. The distributions of the ratio of K_n/K_n,cr under (d-3) high flow rate and (d-4) low flow rate are also shown, and for better visualization, the z-axis scale is doubled, reprinted with permission.¹⁸² Copyright 2024, Elsevier.

In large scale cell and stack systems, the differences in the physicochemical characteristics in the lateral and/or vertical direction become significantly greater compared to those in a small lab-scale cell, as shown in Fig. 12(a) and (b).^{180,189–192} These factors are affected more by the fluid mechanics, with gas diffusion along the channel becoming important and not negligible. Fig. 12(a) and (b) show a schematic of the reaction and gas migration when ammonia fuel is injected into a large area cell. When ammonia is directly injected, an endothermic reaction occurs due to NH₃ decomposition, and an exothermic reaction occurs due to the oxidation of the hydrogen generated in a different area, possibly resulting in temperature and electrochemical performance distributions.^33,193 During the reaction process, ammonia decomposition mainly occurs at the inlet part, reducing the temperature through an endothermic reaction. Thereafter, the HOR occurs mainly at the middle and outlet part, increasing the temperature through an exothermic reaction. Therefore, there are temperature, current density, and gas composition differences in the lateral direction, as shown in Fig. 12(c). For example, according to Omer et al., there is a 226 K temperature difference between the inlet part and the outlet part during 0.3 A cm⁻² operation of a DA-SOFC ten-cell stack.¹⁸¹ As mentioned above, the degree of degradation could differ in the lateral direction through the temperature and gas composition gradient, resulting in significant mechanical cracks induced by the volume changes among the Ni, Ni₃N and NiO.

Stoeckl et al. reported a comparison of degradation phenomena depending on the location of the fuel electrode and Ni current collector in a planar DA-SOFC.¹³⁵ A Ni mesh current collector at the fuel inlet was nitrided due to the relatively low temperature caused by the endothermic reaction of ammonia decomposition and the high partial pressure of ammonia. In contrast, in the outlet region, no nitriding occurred due to the relatively high temperature from the exothermic reaction of the HOR and the low partial pressure of ammonia. Therefore, it is necessary to optimize the fuel injection conditions and stabilize the temperature gradient. The two main control methods are the operating temperature and the flow rate. Electrochemical performance outcomes of DA-SOFCs tend to increase as the flow rate of ammonia increases. However, a low fuel utilization rate can reduce economic efficiency and cause an excessive ammonia decomposition reaction compared to the HOR. This excessive endothermic reaction at the inlet increases the temperature gradient between the inlet and outlet, causing mechanical stress.

To overcome the temperature and gas composition gradient, redesigning the flow channel in the metallic interconnect has been investigated.^194–197 The fuel can be injected into a channel along the co-flow, cross-flow, and counter-flow. To reduce the gradient, finding a compatible injecting strategy should be accomplished. Bae et al. reported the redesign of a flow channel based on CFD simulations to suppress the temperature gradient.¹⁹⁸ As a result of calculating the DA-SOFC stack through CFD simulation, the temperature difference was found to be 128–163 °C according to various operating conditions in the conventionally designed in-flow configuration. Due to the significant temperature difference at the inlet and outlet sites, high mechanical stress can arise, resulting in a high probability of thermal degradation. However, the temperature deviation was reduced to 77–97 °C under various operating conditions when the new alternative design was applied, indicating values nearly 38% lower than those with the conventional in-flow shape. The temperature gradient was located at the center, greatly reducing the thermal stress. Dong et al. also reported the effects of various parameters on the performance of a planar DA-SOFC.¹⁹⁴ To address the gap in the thermal stress analysis when ammonia is fed directly into planar SOFCs, three different flows (co-flow, counter-flow, and cross-flow) were manufactured, and CFD simulations were used. The temperature differences on the cell surface in a lateral direction, depending on the flow channel type, were 81, 237, and 284 K for co-, counter-, and cross-flow configurations, respectively.

In large-scale stack experiments, a large number of SOFCs is required, making it challenging to conduct an experiment and characterize the degradation phenomena by means of a post-mortem analysis without considering economic and time-related perspectives.^199,200 Therefore, researchers have attempted to predict the performance or degradation of a DA-SOFC through 0 to 3D modeling and a simulation.^{188,201–203} Ni et al. reported the 2D thermo-electrochemical modelling of DA-SOFCs to predict the electrochemical performance according to various parameters.²⁰⁴ In this research, through 2D CFD model simulations, they found that the inclusion of NH₃ thermal decomposition is an important factor to increase the electric output and the temperature field in a DA-SOFC. Oh et al. utilized 2D multiphysics modeling to obtain a detailed perspective on the limited performance capabilities of thin-film DA-SOFCs, especially at low temperatures.³⁹ The performances were well validated with experimental results from large-scale DA-SOFCs, with 2D multiphysics modeling showing that an insufficient supply of H₂ from the reduced ammonia decomposition reaction at a low temperature and poor mass transport resulted in a considerable performance drop. They concluded that an improved ammonia decomposition reaction and an appropriate cell design that facilitates the diffusion process may further increase the performance of low-temperature DA-SOFCs. Frandsen et al. reported that a 3D multiphysics model of a DA-SOFC was devised at the cell level to compare the nitriding process of nickel using transport equations, heat transfer, electrochemical reactions, and the NH₃ cracking reaction in parallel with simulations and experiments.¹⁸² The nickel nitriding potential (K_n, x_NH3/x_H2^1.5) was calculated from the simulation based on a Lehrer diagram for Ni. In order for nitriding not to occur, K_n must be less than the critical nitriding potential (K_n,cr) (K_n < K_n,cr). Comparing this with the experimental data, no nitriding was observed in the specific region of the fuel electrode where K_n < K_n,cr was calculated, but in the region where K_n > K_n,cr, nickel nitriding was observed, an outcome consistent with the calculation, as shown in Fig. 12(d). Rizvandi et al. reported that the multiscale modeling of H₂-SOFCs can predict stack-wide performance degradation.²⁰⁵ The simulation predicted 38 000 h of stack operation in approximately one hour and 15 minutes, also predicting that Ni particle coarsening, chromium poisoning of the cathode, and oxidation of the interconnect were the main causes of electrochemical performance degradation. Because DA-SOFCs have not yet been extensively tested, there are no results pertaining to such long-term operating simulations. However, many of the studies on H₂-SOFCs can be built upon to predict the lifetimes of DA-SOFCs. These simulations of the interactions among these degradation mechanisms and operating conditions can provide insight into how to improve long-term stack performance capabilities.^{183,206–208}

At the system level, it is important to build a more efficient DA-SOFC system for economical operation. Even if the DA-SOFC system runs at a high fuel utilization rate, 20–30% of the fuel is wasted. Therefore, recirculating the anode off-gas or using engines, gas turbines, or other fuel cells such as PEMFCs can improve the efficiency.^209–213 Cinti reported a simulation analysis of anode off gas recirculation (AOGR) in a DA-SOFC system, as shown in Fig. 13(a).²⁰⁹ By using various strategies such as power rating, fuel recirculation, and stack cascading through various system designs, the net efficiency was improved from 52.1% to 66% compared to the reference design, with the heat surface area reduced by 67% and the heat exchanger surface area also minimized. Frandsen et al. also reported an efficiency improvement through the AOGR by using 3D multiphysics modeling.¹⁸² In their model, the efficiency was measured at different AOGR ratios under high flow (low fuel utilization) and low flow (high fuel utilization) conditions. At a low flow, ammonia was not detected in the off-gas due to the long retention of ammonia in the cell and its decomposition before reaching the outlet, while at a high flow, an approximate mole fraction of ammonia gas of 0.15 was detected. When the fuel was injected through 70–90% AOGR, electrochemical performance was reduced slightly by 0.34–0.88% compared to that without recirculation. This result achieved high economic efficiency with a fuel saving of 21–27%. They also suggested that nickel nitriding can be avoided in the operating range with a fuel utilization factor of 70% and temperature range of 750–850 °C. Ashar et al. also reported an attempt to utilize off-gas as fuel for combustion in combination with a gas turbine rather than recirculation.²¹⁵ When the results of a ceria-based intermediate temperature SOFC stack simulation were analyzed through process simulation software, it was found to be possible to achieve lower heating value (LHV) efficiency rates close to 60% relative to the optimal efficiency considering the fuel utilization rate. These DA-SOFC systems combined with recirculation or other power generation methods will facilitate higher efficiency and can bring commercialization closer.

Fig. 13 (a) System design with the power rate and anode recirculation, reprinted with permission.²⁰⁹ Copyright 2023, Elsevier. (b) 3D scatter plots of experimental points using a machine learning technique to find the optimized conditions to realize the best electrochemical performance in a total of ten experimental trials. The black arrows indicate the path explored by the Bayesian optimization algorithm based on PI-M3/2, reprinted with permission.²¹⁴ Copyright 2024, Elsevier.

These simulation results convey the importance of finding the appropriate fuel-injecting conditions for achieving high efficiency and stability from an economic perspective. However, utilizing only human actions introduces limitations when seeking to find the optimal point by considering the various multiphysics processes of heat transfer, fluid dynamics, electrochemistry, thermodynamics, and thermocatalysis simultaneously. Therefore, recent studies have attempted to optimize the operating conditions through artificial intelligence, machine learning (ML) and theoretical analysis methods.^216–219 Wang et al. reported machine learning tools for optimizing SOFC-based system design.²²⁰ They leveraged multiphysics models and machine learning tools to automate the design process and improve the reliability of their solutions. The physics-informed deep neural network technique reduces the prediction error by approximately 5% compared to traditional empirical designs or simple reduced-order model techniques and is highly reliable, even when the training data are scarce. After optimizing the fuel temperature, fuel utilization, internal reforming percentage, and oxidant recirculation fraction using the interior point optimizer, a maximum efficiency of 61.48% was achieved, which is higher than the reference efficiency of 60.1%. This combined approach of physics models and machine learning significantly reduced the computational cost of building and optimizing the ROM, speeding up design time. And Baek et al. suggested a method that rapidly optimizes the operating conditions of DA-SOFCs.²¹⁴ Through Bayesian optimization, a ML technique, it was possible to find the optimal performance with only a small number of iterations, as shown in Fig. 13(b). When running fuel cells in real-world stacks and systems, there are numerous factors involved, such that it is difficult to find the optimal conditions for each one. Accordingly, these AI tools can help to achieve optimal conditions more economically. We introduced various examples of how to use computational simulations and various tools to analyze the causes of degradation. Given that it is difficult to identify degradation phenomena on the stack in certain situations during experiments, these computational studies can overcome the challenges associated with the commercialization of DA-SOFCs.

5. Conclusion and future outlook

SOFCs can offer high compatibility with ammonia, which is a promising alternative fuel that may replace hydrogen given its high volumetric energy density, high liquefying temperature, and ease of transportation and storage. Moreover, the material and operating temperature of SOFCs enable them to operate without an external ammonia reformer. However, DA-SOFCs are associated with significant degradation. Therefore, this review discusses the three aspects of materials, cells, and stacks/systems to address these challenges.

(1) Material aspects: the materials utilized in DA-SOFCs can be categorized into the catalyst, catalyst support and the metallic interconnects. Because the ammonia decomposition reaction concurrently occurs at the fuel electrode and during the HOR, adsorbed nitrogen induces undesirable chemical reactions, mostly nitridation. Nitridation significantly reduces the electrochemical and thermocatalytic performance capabilities of DA-SOFCs, also affecting the stability. In addition, nickel oxidation can arise in DA-SOFCs due to the presence of water and/or conducted oxygen ions and the synergetic effects of ammonia during their operation. Thus, these processes can result in mechanical degradation in the form of cracks or delamination due to volume differences. To suppress these types of deterioration, improving the ammonia decomposition kinetics and surface coating to mitigate the influence of ammonia on the bare material can be considered.

(2) Cell aspects: some SOFCs and PCFCs depend on the charge carriers of O²⁻ and H⁺, and ASCs and ESCs depend on the type of supporting layers. Because SOFCs and PCFCs create different byproducts in the fuel electrode and have different operating temperatures due to the different electrochemical reactions according to the charge carrier used, the degradation behaviors can also differ. In particular, the Ni in a DA-SOFC can be oxidized and can generate NO_x through the oxygen ions conducted from the electrolyte and evolved water in the fuel electrode. In contrast, PCFCs can avoid this issue due to the absence of conducting oxygen ions and water in the fuel electrode, but Ni can be nitrided because of lower operating temperature. In addition, between ASCs and ESCs, ASCs have a more compatible cell structure for DA-SOFCs due to their thick reforming layer as the fuel electrode. To compensate for the different characteristics depending on the cell types, cell modifications with a catalyst, a catalyst support and an additional reforming layer could improve catalytic and electrochemical performance and sustainability of DA-SOFCs.

(3) Stacks/systems: in large-scale cells and/or stacks, more complicated multiphysics processes, including heat transfers, fluid dynamics, thermodynamics, electrochemistry, and thermocatalysis, have been simultaneously considered. For this reason, it is difficult to acquire reliable experimental results and characterize the degradation behaviors. Moreover, different characteristics of the temperature, gas composition, and electrochemical performance in the lateral and vertical directions significantly induce undesirable stability issues. To clear these hurdles, computational approaches such as 0 to 3D modelling, CFD, and ML techniques have been applied to provide unexpected insight via experiments. A recent model successfully predicted nickel nitridation in a large-scale cell, providing a thoroughly validated performance estimation.

(4) Outlooks: DA-SOFCs have shown a strong possibility to replace the H₂-fueled SOFCs due to their high comparability of cost-effectiveness with largely established infrastructures and chemical characteristics of high volumetric energy density and easy liquefaction. DA-SOFCs show comparable performance and efficiency compared to the H₂-fueled SOFCs. However, the challenge is still veiled degradation behavior under various conditions, specifically voltage, flow rate, partial pressure, temperature, and other undesired secondary effects with external factors of sulfur, Cr, water, etc. Therefore, first, revealing these degradation factors and their effects on the electrochemical stabilities under various conditions should be further conducted. Then, based on these researches, we could consider the appropriate operating strategies for sustainable operation and propose novel approaches to suppress degradation or preserve stability. Delving into this from various perspectives, including materials, cells, and systems, can enlarge the approaches to solve it, such as using new material synthesis, wet-chemical process, vacuum deposition techniques, machine learning, deep learning, computational fluid dynamics, and so on. We believe that this review paper can brighten the future research directions for the DA-SOFCs.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Author contributions

Hyunho Lee: writing – original draft, review & editing, figure curation, conceptualization. Jaewan Baek: writing – original draft, figure curation. Mingi Choi: writing – review & editing, supervision, conceptualization.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

This study was financially supported by Seoul National University of Science and Technology and National research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No. 2021R1C1C2006657), and the result of a research project conducted with the funds of the Open R&D program of Korea Electric Power Corporation (No. R23XO03).

References

1. P. P. Edwards, V. L. Kuznetsov and W. I. David, Philos. Trans. R. Soc., A, 2007, 365, 1043–1056.
2. M. A. Rosen and S. Koohi-Fayegh, Energy, Ecol. Environ., 2016, 1, 10–29.
3. G. Cipriani, V. Di Dio, F. Genduso, D. La Cascia, R. Liga, R. Miceli and G. R. Galluzzo, Int. J. Hydrogen Energy, 2014, 39, 8482–8494.
4. R. Dell and D. A. J. Rand, Clean Energy, Royal Society of Chemistry, 2004.
5. J. O. Abe, A. Popoola, E. Ajenifuja and O. M. Popoola, Int. J. Hydrogen Energy, 2019, 44, 15072–15086.
6. O. Massarweh, M. Al-khuzaei, M. Al-Shafi, Y. Bicer and A. S. Abushaikha, J. CO2 Util., 2023, 70, 102438.
7. S. K. Dwivedi and M. Vishwakarma, Int. J. Hydrogen Energy, 2018, 43, 21603–21616.
8. X. Li, X. Ma, J. Zhang, E. Akiyama, Y. Wang and X. Song, Acta Metall. Sin. (Engl. Lett.), 2020, 33, 759–773.
9. H. Taghvaei, A. Jahanmiri, M. R. Rahimpour, M. M. Shirazi and N. Hooshmand, Chem. Eng. J., 2013, 226, 384–392.
10. M. Otto, K. L. Chagoya, R. G. Blair, S. M. Hick and J. S. Kapat, J. Energy Storage, 2022, 55, 105714.
11. C. Zamfirescu and I. Dincer, J. Power Sources, 2008, 185, 459–465.
12. D. Erdemir and I. Dincer, Int. J. Energy Res., 2021, 45, 4827–4834.
13. O. Herbinet, P. Bartocci and A. G. Dana, Fuel Commun., 2022, 11, 100064.
14. A. Valera-Medina, F. Amer-Hatem, A. K. Azad, I. Dedoussi, M. De Joannon, R. Fernandes, P. Glarborg, H. Hashemi, X. He and S. Mashruk, Energy Fuels, 2021, 35, 6964–7029.
15. M. Aziz, A. T. Wijayanta and A. B. D. Nandiyanto, Energies, 2020, 13, 3062.
16. S. Giddey, S. P. Badwal, C. Munnings and M. Dolan, ACS Sustain. Chem. Eng., 2017, 5, 10231–10239.
17. D. Saygin, H. Blanco, F. Boshell, J. Cordonnier, K. Rouwenhorst, P. Lathwal and D. Gielen, Sustainability, 2023, 15, 1623.
18. L. Ye, R. Nayak-Luke, R. Banares-Alcantara and E. Tsang, Chem, 2017, 3, 712–714.
19. V. Pattabathula and J. Richardson, Chem. Eng. Prog., 2016, 112, 69–75.
20. C. Makhloufi and N. Kezibri, Int. J. Hydrogen Energy, 2021, 46, 34777–34787.
21. T. N. Do, H. Kwon, M. Park, C. Kim, Y. T. Kim and J. Kim, Energy Convers. Manage., 2023, 279, 116758.
22. A. Pal, S. Kakran, A. Kumar, A. B. Youssef, U. P. Singh and A. Sidhu, Int. J. Hydrogen Energy, 2024, 49, 16–41.
23. Y. B. Shapovalov, O. Salavor and I. Yakymenko, IOP Conf. Ser.: Earth Environ. Sci, 2023, 1, 012025.
24. T. Zhu, X. Yan, Z. Gao, Y. Qiu, L. Zhu and Z. Huang, Int. J. Hydrogen Energy, 2024, 62, 579–590.
25. R. Sitar, J. Shah, J. D. Way and C. A. Wolden, Energy Fuels, 2022, 36, 9357–9364.
26. N. A. Qasem and G. A. Abdulrahman, Int. J. Energy Res., 2024, 2024, 7271748.
27. H. Lee and M.-J. Lee, Energies, 2021, 14, 5604.
28. N. Kim, M. Lee, J. Park, J. Park and T. Lee, Energies, 2022, 15, 9617.
29. A. Afif, N. Radenahmad, Q. Cheok, S. Shams, J. H. Kim and A. K. Azad, Renewable Sustainable Energy Rev., 2016, 60, 822–835.
30. I. Lucentini, X. Garcia, X. Vendrell and J. Llorca, Ind. Eng. Chem. Res., 2021, 60, 18560–18611.
31. T. A. Le, Q. C. Do, Y. Kim, T.-W. Kim and H.-J. Chae, Korean J. Chem. Eng., 2021, 38, 1087–1103.
32. J. C. Ganley, F. Thomas, E. Seebauer and R. I. Masel, Catal. Lett., 2004, 96, 117–122.
33. Y. Luo, S. Liao, S. Chen, H. Fang, F. Zhong, L. Lin, C. Zhou, C. Chen, G. Cai and C.-T. Au, Appl. Energy, 2022, 307, 118158.
34. S. S. Rathore, S. Biswas, D. Fini, A. P. Kulkarni and S. Giddey, Int. J. Hydrogen Energy, 2021, 46, 35365–35384.
35. M. Kishimoto, H. Muroyama, S. Suzuki, M. Saito, T. Koide, Y. Takahashi, T. Horiuchi, H. Yamasaki, S. Matsumoto and H. Kubo, Fuel Cells, 2020, 20, 80–88.
36. S. Yang, H. Kim, S. Oh, A. A. Gokbayrak, J. Lee, M. J. Oh, H.-i. Ji, K. J. Yoon, J.-W. Son and J.-H. Lee, ECS Trans., 2023, 111, 2111.
37. Y. Aoki, T. Yamaguchi, S. Kobayashi, D. Kowalski, C. Zhu and H. Habazaki, Global Challenges, 2018, 2, 1700088.
38. L. Zhu, C. Cadigan, C. Duan, J. Huang, L. Bian, L. Le, C. H. Hernandez, V. Avance, R. O'Hayre and N. P. Sullivan, Commun. Chem., 2021, 4, 121.
39. S. Oh, M. J. Oh, J. Hong, K. J. Yoon, H.-I. Ji, J.-H. Lee, H. Kang, J.-W. Son and S. Yang, iScience, 2022, 25(9), 105009.
40. F. Zhong, S. Yang, C. Chen, H. Fang, K. Chen, C. Zhou, L. Lin, Y. Luo, C. Au and L. Jiang, J. Power Sources, 2022, 520, 230847.
41. X. Xiong, J. Yu, X. Huang, D. Zou, Y. Song, M. Xu, R. Ran, W. Wang, W. Zhou and Z. Shao, J. Mater. Sci. Technol., 2022, 125, 51–58.
42. Y. Pan, H. Zhang, K. Xu, Y. Zhou, B. Zhao, W. Yuan, K. Sasaki, Y. Choi, Y. Chen and M. Liu, Appl. Catal., B, 2022, 306, 121071.
43. H. Zhang, Y. Zhou, K. Pei, Y. Pan, K. Xu, Y. Ding, B. Zhao, K. Sasaki, Y. Choi and Y. Chen, Energy Environ. Sci., 2022, 15, 287–295.
44. H. J. Jeong, W. Chang, B. G. Seo, Y. S. Choi, K. H. Kim, D. H. Kim and J. H. Shim, Small, 2023, 19, 2208149.
45. F. Liu, H. Deng, D. Diercks, P. Kumar, M. H. A. Jabbar, C. Gumeci, Y. Furuya, N. Dale, T. Oku and M. Usuda, Nat. Energy, 2023, 8, 1145–1157.
46. H.-T. Huang, D.-C. Lu, C.-X. Li and S.-L. Zhang, Int. J. Hydrogen Energy, 2024, 87, 89–99.
47. H. Lee, J. Kim, J. Baek, H. Kwon and M. Choi, Chem. Eng. J., 2024, 497, 154764.
48. H. Jiang, Z. Liang, H. Qiu, Y. Yi, S. Jiang, J. Xu, W. Wang, C. Su and T. Yang, Nanomaterials, 2024, 14, 673.
49. H. Fang, S. Yang, W. Ye, F. Zhong, Y. Luo, S. Wang, C. Chen and L. Jiang, Chem. Eng. Sci., 2024, 290, 119778.
50. J. Yun, G. Xiong, S. Kim, D. Bardgett, S. Choi and S. M. Haile, ACS Energy Lett., 2024, 9, 5520–5528.
51. K. Kitabayashi, J. Hyodo, N. Shi and Y. Yamazaki, Solid State Ionics, 2024, 417, 116701.
52. L. Chen, H. Zhang, K. Xu, Y. Xu, X. Zhang, F. Zhu, F. He and Y. Chen, Mater. Today Catal., 2024, 7, 100072.
53. Z. Huang, T. Chen, X. Zhang, K. Liu, T. Li, S. Duo, H. Zhang, Y. Ling and S. Wang, Ceram. Int., 2024, 50, 10551–10559.
54. A. Hagen, Risø International Energy Conference 2007: Energy Solutions for Sustainable Development, 2007, pp. 347–356.
55. M.-F. Han, J. Xiong, T. Zhu and S. C. Singhal, ECS Trans., 2013, 50, 163.
56. V. Singh, H. Muroyama, T. Matsui and K. Eguchi, ECS Trans., 2017, 78, 2527.
57. B. Stoeckl, V. Subotić, M. Preininger, M. Schwaiger, N. Evic, H. Schroettner and C. Hochenauer, Electrochim. Acta, 2019, 298, 874–883.
58. Y. Song, H. Li, M. Xu, G. Yang, W. Wang, R. Ran, W. Zhou and Z. Shao, Small, 2020, 16, 2001859.
59. Y. Wang, J. Yang, J. Wang, W. Guan, B. Chi, L. Jia, J. Chen, H. Muroyama, T. Matsui and K. Eguchi, J. Electrochem. Soc., 2020, 167, 064501.
60. F. He, Q. Gao, Z. Liu, M. Yang, R. Ran, G. Yang, W. Wang, W. Zhou and Z. Shao, Adv. Energy Mater., 2021, 11, 2003916.
61. K. Xu, F. Zhu, M. Hou, C. Li, H. Zhang and Y. Chen, Nano Res., 2023, 16, 2454–2462.
62. F. He, M. Hou, Z. Du, F. Zhu, X. Cao, Y. Ding, Y. Zhou, M. Liu and Y. Chen, Adv. Mater., 2023, 35, 2304957.
63. H. Lan, J. Chu, X. Chen, Q. Zhou, W. Jin, Y. Zhang and J. Zhou, J. Power Sources, 2024, 593, 233987.
64. O. Rahumi, M. K. Rath, L. Meshi, I. Rozenblium and K. Borodianskiy, ACS Appl. Mater. Interfaces, 2024, 16, 53652–53664.
65. H. Zhang, K. Xu, Y. Xu, F. He, F. Zhu, K. Sasaki, Y. Choi and Y. Chen, Energy Environ. Sci., 2024, 17, 3433–3442.
66. S. Z. Golkhatmi, M. I. Asghar and P. D. Lund, Renewable Sustainable Energy Rev., 2022, 161, 112339.
67. F. Wang, H. Kishimoto, T. Ishiyama, K. Develos-Bagarinao, K. Yamaji, T. Horita and H. Yokokawa, J. Power Sources, 2020, 478, 228763.
68. P. Steiger, D. Burnat, H. Madi, A. Mai, L. Holzer, J. Van Herle, O. Krocher, A. Heel and D. Ferri, Chem. Mater., 2019, 31, 748–758.
69. S. Bang, J. Lee, J. G. Kim, J. Kim, M. Choi, Y. Chen and W. Lee, J. Mater. Chem. A, 2024, 12, 24103–24113.
70. A. Talukdar, A. Chakrovorty, P. Sarmah, P. Paramasivam, V. Kumar, S. K. Yadav and S. Manickkam, Int. J. Energy Res., 2024, 2024, 6443247.
71. A. C. Chien, W. Y. Chen and M. S. Zheng, J. Electrochem. Soc., 2023, 170, 044505.
72. M. Chen, B. H. Kim, Q. Xu, B. G. Ahn and D. P. Huang, Solid State Ionics, 2010, 181, 1119–1124.
73. F. García-García, A. Guerrero-Ruiz, I. Rodríguez-Ramos, A. Goguet, S. Shekhtman and C. Hardacre, Phys. Chem. Chem. Phys., 2011, 13, 12892–12899.
74. J. Yang, A. F. S. Molouk, T. Okanishi, H. Muroyama, T. Matsui and K. Eguchi, ACS Appl. Mater. Interfaces, 2015, 7, 28701–28707.
75. X. Duan, J. Ji, G. Qian, C. Fan, Y. Zhu, X. Zhou, D. Chen and W. Yuan, J. Mol. Catal. A: Chem., 2012, 357, 81–86.
76. L. Abadlia, F. Gasser, K. Khalouk, M. Mayoufi and J.-G. Gasser, Rev. Sci. Instrum., 2014, 85, 095121.
77. J. Liang, B. Zhang, H. Shen, Y. Yin, L. Liu, Y. Ma, X. Wang, C. Xiao, J. Kong and S. Ding, Appl. Surf. Sci., 2020, 503, 144143.
78. S. Badwal, Solid State Ionics, 2001, 143, 39–46.
79. Z. Gao, L. V. Mogni, E. C. Miller, J. G. Railsback and S. A. Barnett, Energy Environ. Sci., 2016, 9, 1602–1644.
80. N. Shaigan, W. Qu, D. G. Ivey and W. Chen, J. Power Sources, 2010, 195, 1529–1542.
81. M. Shah, P. W. Voorhees and S. A. Barnett, Solid State Ionics, 2011, 187, 64–67.
82. H. Wriedt, Bull. Alloy Phase Diagrams, 1985, 6, 558–563.
83. A. Leineweber, H. Jacobs and S. Hull, Inorg. Chem., 2001, 40, 5818–5822.
84. W. Jung and H. L. Tuller, Energy Environ. Sci., 2012, 5, 5370–5378.
85. O. Elmutasim, S. Giddey, D. S. Dhawale and S. Bhattacharya, Int. J. Hydrogen Energy, 2024, 96, 192–209.
86. J.-S. Shin, M. Saqib, M. Jo, K. Park, K. M. Park, J. S. Ahn, H.-T. Lim and J.-Y. Park, ACS Appl. Mater. Interfaces, 2021, 13, 49868–49878.
87. N. Shah, X. Xu, J. Love, H. Wang, Z. Zhu and L. Ge, J. Power Sources, 2024, 599, 234211.
88. A. Boisen, S. Dahl, J. K. Nørskov and C. H. Christensen, J. Catal., 2005, 230, 309–312.
89. W. Guo and D. G. Vlachos, Nat. Commun., 2015, 6, 8619.
90. D. A. Hansgen, D. G. Vlachos and J. G. Chen, Nat. Chem., 2010, 2, 484–489.
91. J. W. Makepeace, T. He, C. Weidenthaler, T. R. Jensen, F. Chang, T. Vegge, P. Ngene, Y. Kojima, P. E. de Jongh and P. Chen, Int. J. Hydrogen Energy, 2019, 44, 7746–7767.
92. A. J. Medford, A. Vojvodic, J. S. Hummelshøj, J. Voss, F. Abild-Pedersen, F. Studt, T. Bligaard, A. Nilsson and J. K. Nørskov, J. Catal., 2015, 328, 36–42.
93. S. S. Rathore, A. P. Kulkarni, D. Fini, S. Giddey and A. Seeber, Solids, 2021, 2, 177–191.
94. X. Li, J. Chen, Y. Huang, H. Fang, C. Chen, F. Zhong, L. Lin, Y. Luo, Y. Wang and L. Jiang, Front. Energy, 2024, 18, 875–884.
95. A. Hatta, A. A. Jalil, N. Hassan, M. Hamid, A. Rahman, L. Teh and D. Prasetyoko, Int. J. Hydrogen Energy, 2022, 47, 30981–31002.
96. K. Hong, M. Choi, Y. Bae, J. Min, J. Lee, D. Kim, S. Bang, H.-K. Lee, W. Lee and J. Hong, Nat. Commun., 2023, 14, 7485.
97. R. K. Rai, W. Al Maksoud, N. Morlanés, M. Harb, R. Ahmad, A. Genovese, M. N. Hedhili, L. Cavallo and J.-M. Basset, ACS Catal., 2021, 12, 587–599.
98. H. He, H. Jiang, F. Yang, J. Liu, W. Zhang, M. Jin and Z. Li, Int. J. Hydrogen Energy, 2023, 48, 5030–5041.
99. H. Wu, C. Liu and W. Guo, J. Phys. Chem. C, 2021, 126, 192–202.
100. K. X. Lee, B. Hu, P. K. Dubey, M. Anisur, S. Belko, A. N. Aphale and P. Singh, Int. J. Hydrogen Energy, 2022, 47, 38372–38385.
101. Q. Wang, J. Xie, Y. Qin, Y. Kong, S. Zhou, Q. Li, Q. Sun, B. Chen, P. Xie and Z. Wei, Adv. Mater. Interfaces, 2024, 11, 2301020.
102. W. A. Saidi, W. Shadid and G. t. Veser, J. Phys. Chem. Lett., 2021, 12, 5185–5192.
103. Y. Zhang, Z. Wang, L. Wang and L. Zong, Small, 2024, 20, 2400892.
104. P. Xie, Y. Yao, Z. Huang, Z. Liu, J. Zhang, T. Li, G. Wang, R. Shahbazian-Yassar, L. Hu and C. Wang, Nat. Commun., 2019, 10, 4011.
105. L. I. Holz, F. J. Loureiro, V. C. Graça, S. M. Mikhalev, D. Mendes, A. Mendes and D. P. Fagg, Renewable Energy, 2022, 201, 124–130.
106. S. Sorcar, H. Zinowits, E. P. Komarala, N. Moshe, I. Agranovich and B. A. Rosen, J. Mater. Chem. A, 2022, 10, 24115–24126.
107. M. Hashinokuchi, R. Yokochi, W. Akimoto, T. Doi, M. Inaba and J. Kugai, Solid State Ionics, 2016, 285, 222–226.
108. Y. Ya, Y. Xu, Y. Liu, B. Sun, J. Liu and X. Cheng, Int. J. Hydrogen Energy, 2024, 91, 659–672.
109. K. Lou, F. H. Wang, Y. J. Lu and X. Zhao, Int. J. Mod. Phys. B, 2016, 30, 1650200.
110. J. Laurencin, G. Delette, B. Morel, F. Lefebvre-Joud and M. Dupeux, J. Power Sources, 2009, 192, 344–352.
111. M. Pihlatie, T. Ramos and A. Kaiser, J. Power Sources, 2009, 193, 322–330.
112. F. El Gabaly, K. F. McCarty, H. Bluhm and A. H. McDaniel, Phys. Chem. Chem. Phys., 2013, 15, 8334–8341.
113. A. Faes, A. Hessler-Wyser, A. Zryd and J. Van Herle, Membranes, 2012, 2, 585–664.
114. T.-L. Cheng, Y. Lei, Y. Chen, Y. Fan, H. Abernathy, X. Song and Y.-H. Wen, J. Power Sources, 2023, 569, 232991.
115. H. Yan, Y.-J. Xu, Y.-Q. Gu, H. Li, X. Wang, Z. Jin, S. Shi, R. Si, C.-J. Jia and C.-H. Yan, J. Phys. Chem. C, 2016, 120, 7685–7696.
116. D. S. Dhawale, S. Biswas, G. Kaur and S. Giddey, Inorg. Chem. Front., 2023, 10, 6176–6192.
117. A. Faes, A. Nakajo, A. Hessler-Wyser, D. Dubois, A. Brisse and S. Modena, J. Power Sources, 2009, 193, 55–64.
118. M. Welander, M. Zachariasen, C. Hunt, S. Sofie and R. Walker, J. Electrochem. Soc., 2018, 165, F152.
119. J. Yang, T. Akagi, T. Okanishi, H. Muroyama, T. Matsui and K. Eguchi, Fuel Cells, 2015, 15, 390–397.
120. K. Miyazaki, H. Muroyama, T. Matsui and K. Eguchi, Sustainable Energy Fuels, 2020, 4, 5238–5246.
121. K. Okura, K. Miyazaki, H. Muroyama, T. Matsui and K. Eguchi, RSC Adv., 2018, 8, 32102–32110.
122. C. Nicollet, C. Toparli, G. F. Harrington, T. Defferriere, B. Yildiz and H. L. Tuller, Nat. Catal., 2020, 3, 913–920.
123. F. Tietz, Ionics, 1999, 5, 129–139.
124. M. Mori, T. Yamamoto, H. Itoh, H. Inaba and H. Tagawa, J. Electrochem. Soc., 1998, 145, 1374.
125. Y. Sun, D. Jin, X. Zhang, Q. Shao, C. Guan, R. Li, F. Cheng, X. Lin, G. Xiao and J. Wang, Materials, 2024, 17, 1216.
126. A. Jain, S. P. Ong, G. Hautier, W. Chen, W. D. Richards, S. Dacek, S. Cholia, D. Gunter, D. Skinner and G. Ceder, APL Mater., 2013, 1, 011002.
127. S. Dan, G. Maalhagh, H. Bagheri and R. Ranjbar-Karimi, in Progresses in Ammonia: Science, Technology and Membranes, Elsevier, 2024, pp. 157–177.
128. M. A. Laguna-Bercero, A. R. Hanifi, L. Menand, N. K. Sandhu, N. E. Anderson, T. H. Etsell and P. Sarkar, Electrochim. Acta, 2018, 268, 195–201.
129. A. R. Hanifi, M. A. Laguna-Bercero, N. K. Sandhu, T. H. Etsell and P. Sarkar, Sci. Rep., 2016, 6, 27359.
130. K. Machaj, J. Kupecki, A. Niemczyk, Z. Malecha, J. Brouwer and D. Porwisiak, Int. J. Hydrogen Energy, 2024, 52, 936–951.
131. P. Ettmayer and W. Lengauer, Ullmann's Encyclopedia of Industrial Chemistry, 2000.
132. P. Feng, M. Lee, D. Wang and Y. Suzuki, Int. J. Hydrogen Energy, 2023, 48, 29209–29219.
133. A.-C. Kimmel, T. F. Malkowski, S. Griffiths, B. Hertweck, T. G. Steigerwald, L. P. Freund, S. Neumeier, M. Göken, J. S. Speck and E. Schluecker, J. Cryst. Growth, 2018, 498, 289–300.
134. H. Kim, J. Lee, A. A. Gokbayrak, Y. Seo, S. Oh, M. J. Oh, Y. Jun, J.-W. Son and S. Yang, Int. J. Hydrogen Energy, 2024, 68, 1312–1321.
135. B. Stoeckl, M. Preininger, V. Subotić, S. Megel, C. Folgner and C. Hochenauer, J. Power Sources, 2020, 450, 227608.
136. M. Tomas, A. Visibile, J.-E. Svensson and J. Froitzheim, Int. J. Hydrogen Energy, 2023, 48, 18405–18419.
137. H. Abdoli, S. Molin and H. Farnoush, Mater. Lett., 2020, 259, 126898.
138. J. Zhu, Y. Zhang, A. Basu, Z. Lu, M. Paranthaman, D. Lee and E. Payzant, Surf. Coat. Technol., 2004, 177, 65–72.
139. J. Zhu, M. Lewis, S. Du and Y. Li, Thin Solid Films, 2015, 596, 179–184.
140. M. Garai and B. Karmakar, Mater. Today Energy, 2020, 18, 100535.
141. D. Krainova, N. Saetova, A. Farlenkov, A. Khodimchuk, I. Polyakova and A. Kuzmin, Ceram. Int., 2021, 47, 8973–8979.
142. J.-M. Klein and J. Deseure, in Computer Aided Chemical Engineering, Elsevier, 2008, vol. 25, pp. 781–785.
143. M. Choi, J. Paik, D. Kim, D. Woo, J. Lee, S. J. Kim, J. Lee and W. Lee, Energy Environ. Sci., 2021, 14, 6476–6483.
144. C. Mendonça, A. Ferreira and D. M. Santos, Fuels, 2021, 2, 393–419.
145. S. Y. Shin, D.-K. Lim, T. Lee and S.-Y. Jeon, J. Electrochem. Sci. Technol., 2023, 14, 145–151.
146. L. Mathur, Y. Namgung, H. Kim and S.-J. Song, J. Korean Ceram. Soc., 2023, 60, 614–636.
147. A. M. Hussain, Y.-L. Huang, K.-J. Pan, I. A. Robinson, X. Wang and E. D. Wachsman, ACS Appl. Mater. Interfaces, 2020, 12, 18526–18532.
148. K. Kendall, C. Finnerty, G. Saunders and J. Chung, J. Power Sources, 2002, 106, 323–327.
149. S. Hussain and L. Yangping, Energy Transitions, 2020, 4, 113–126.
150. M. Choi, D. Kim, C. B. Lee, J. Baek, S. Bang, Y. Jung, K. Hong, J. Hong, D. Chen and K. Kim, J. Mater. Chem. A, 2024, 12, 6955–6967.
151. M. Tang, Y. Niu, W. Muhammad, S. Muhammad, Z. Zhong, S. Muhammad, Y. Pang, Z. Wan, N. Chen and L. Qiao, Int. J. Hydrogen Energy, 2024, 50, 618–632.
152. J. W. Fergus, J. Power Sources, 2006, 162, 30–40.
153. A. V. Kasyanova, I. A. Zvonareva, N. A. Tarasova, L. Bi, D. A. Medvedev and Z. Shao, Mater. Rep.: Energy, 2022, 2, 100158.
154. K. H. Stern and E. S. Amis, Chem. Rev., 1959, 59, 1–64.
155. M. Choi, D. Kim, T. K. Lee, J. Lee, H. S. Yoo and W. Lee, Adv. Energy Mater., 2025, 15, 2400124.
156. A. Wojcik, H. Middleton and I. Damopoulos, J. Power Sources, 2003, 118, 342–348.
157. J. Yang, A. F. S. Molouk, T. Okanishi, H. Muroyama, T. Matsui and K. Eguchi, ACS Appl. Mater. Interfaces, 2015, 7, 7406–7412.
158. G. B. Ariemma, G. Sorrentino, R. Ragucci, M. de Joannon and P. Sabia, Combust. Flame, 2022, 241, 112071.
159. A. Faes, H. L. Frandsen, A. Kaiser and M. Pihlatie, Fuel Cells, 2011, 11, 682–689.
160. M. Hubert, J. Laurencin, P. Cloetens, B. Morel, D. Montinaro and F. Lefebvre-Joud, J. Power Sources, 2018, 397, 240–251.
161. Q. Fu, C. Tian, L. Hun, X. Wang, Z. Li, Z. Liu and W. Wei, Energy, 2024, 289, 129997.
162. A. Zekri, K. Herbrig, M. Knipper, J. Parisi and T. Plaggenborg, Fuel Cells, 2017, 17, 359–366.
163. D. Y. Jang, J. Koo, H. R. Choi, J. W. Kim, H. J. Jeong, F. B. Prinz and J. H. Shim, ACS Sustain. Chem. Eng., 2020, 8, 10529–10535.
164. X. Li, Y. Zhu, N. Zhang, Y. Gong, R. Wang, H. Wang, J. Jin, L. Zhao and B. He, Int. J. Hydrogen Energy, 2024, 55, 208–216.
165. M. A. Umer, C.-Y. Cheng, K.-C. Tsai, C.-J. Tseng, S.-y. Chen and S.-W. Lee, J. Power Sources, 2024, 613, 234872.
166. C.-A. Thieu, S. Yang, H.-I. Ji, H. Kim, K. J. Yoon, J.-H. Lee and J.-W. Son, J. Mater. Chem. A, 2022, 10, 2460–2473.
167. F. Zhong, L. Wang, H. Fang, Y. Luo, C. Chen, L. Lin, K. Chen and L. Jiang, Chem. Eng. J., 2023, 471, 144650.
168. Y. Yi, J. Chen, M. Xu, G. Yang, R. Ran, W. Zhou, W. Wang and Z. Shao, Catalysts, 2023, 13, 996.
169. L. Liu, K. Sun, X. Wu, X. Li, M. Zhang, N. Zhang and X. Zhou, Int. J. Hydrogen Energy, 2012, 37, 10857–10865.
170. T. Ruh, D. Berkovec, F. Schrenk and C. Rameshan, Chem. Commun., 2023, 59, 3948–3956.
171. J. Mei, T. Liao and Z. Sun, Mater. Today Energy, 2023, 31, 101216.
172. Y. H. Kim, H. Jeong, B.-R. Won, H. Jeon, C.-h. Park, D. Park, Y. Kim, S. Lee and J.-h. Myung, Nano-Micro Lett., 2024, 16, 33.
173. M. Hou, Y. Pan and Y. Chen, Sep. Purif. Technol., 2022, 297, 121483.
174. Z. Liu, H. Di, D. Liu, G. Yang, Y. Zhu, Z. Luo, R. Ran, W. Wang, W. Zhou and Z. Shao, Adv. Funct. Mater., 2024, 2420214.
175. K. J. Yoon, M. Biswas, H.-J. Kim, M. Park, J. Hong, H. Kim, J.-W. Son, J.-H. Lee, B.-K. Kim and H.-W. Lee, Nano Energy, 2017, 36, 9–20.
176. Y. Namgung, J. Hong, A. Kumar, D.-K. Lim and S.-J. Song, Appl. Catal., B, 2020, 267, 118374.
177. P. Qiu, S. Sun, X. Yang, F. Chen, C. Xiong, L. Jia and J. Li, Int. J. Hydrogen Energy, 2021, 46, 25208–25224.
178. Z. Lyu, Y. Wang, Y. Zhang and M. Han, Chem. Eng. J., 2020, 393, 124755.
179. W. Wang, C. Su, Y. Wu, R. Ran and Z. Shao, J. Power Sources, 2010, 195, 402–411.
180. A. Nemati, O. B. Rizvandi, R. N. Nakashima, J. Beyrami and H. L. Frandsen, Appl. Energy, 2024, 373, 123913.
181. A. Omer, I. Rahimipetroudi, K. Rashid, K. S. Ullah, W. W. Kazmi, I. Kariim, J. E. Hong and S. K. Dong, Renewable Energy, 2025, 238, 121758.
182. A. Nemati, O. B. Rizvandi, F. Mondi and H. L. Frandsen, Energy Convers. Manage., 2024, 308, 118396.
183. M. M. Rahman, A. M. Abdalla, L. A. Omeiza, V. Raj, S. Afroze, M. S. Reza, M. R. Somalu and A. K. Azad, Processes, 2023, 11, 2728.
184. M. Duarte, L. I. Holz, C. Fernandes, P. Ribeirinha, D. P. Fagg and A. Mendes, Appl. Energy, 2024, 368, 123465.
185. F. Mütter, C. Berger, B. Königshofer, M. Höber, C. Hochenauer and V. Subotić, Energy Convers. Manage., 2023, 291, 117263.
186. M. Asmare, M. Ilbas, F. M. Cimen, C. Timurkutluk and S. Onbilgin, Int. J. Hydrogen Energy, 2022, 47, 15865–15874.
187. M. Asmare, M. Ilbas and S. Yalcin, Int. J. Hydrogen Energy, 2021, 46, 36878–36889.
188. M. Ilbas, M. A. Alemu and F. M. Cimen, Int. J. Hydrogen Energy, 2022, 47, 3416–3428.
189. N. Alhazmi, G. Almutairi, F. Alenazey and B. AlOtaibi, Electrochim. Acta, 2021, 390, 138838.
190. H. Zhu, A. Ashar, R. J. Kee, R. J. Braun and G. S. Jackson, J. Electrochem. Soc., 2023, 170, 104506.
191. L. Mahmud, A. Muchtar and M. Somalu, Renewable Sustainable Energy Rev., 2017, 72, 105–116.
192. R.-T. Wang, H.-Y. Chang and J.-C. Wang, Polymers, 2021, 13, 2774.
193. Y. Lai, Z. Wang, D. Cui, F. Han, Y. Ji and W. Cai, Int. J. Hydrogen Energy, 2023, 48, 2351–2367.
194. A. Omer, I. Rahimipetroudi, K. Rashid, J. B. Yang, J. E. Hong and S. K. Dong, J. Power Sources, 2023, 573, 233135.
195. Y. Liu, Y. Xu, J. Liu, M. Xiang, B. Sun, Y. Ya, Z. Guo and X. Cheng, Int. J. Hydrogen Energy, 2024, 50, 765–785.
196. I. Khazaee and A. Rava, Energy, 2017, 119, 235–244.
197. Y. Liu, Y. Xu, J. Liu, Y. Ya, B. Sun, M. Xiang and X. Cheng, Int. J. Hydrogen Energy, 2024, 94, 1329–1343.
198. T.-Q. Quach, Y. S. Kim, D. K. Lee, K. Y. Ahn, S. Lee and Y. Bae, Energy Convers. Manage., 2023, 291, 117248.
199. R. Scataglini, M. Wei, A. Mayyas, S. H. Chan, T. Lipman and M. Santarelli, Fuel Cells, 2017, 17, 825–842.
200. J. Wang, H. Wang and Y. Fan, Engineering, 2018, 4, 352–360.
201. M. Ni, D. Y. Leung and M. K. Leung, Int. J. Hydrogen Energy, 2008, 33, 5765–5772.
202. M. Ilbas, B. Kumuk, M. A. Alemu and B. Arslan, Int. J. Hydrogen Energy, 2020, 45, 35108–35117.
203. V.-T. Giap, T.-Q. Quach, K. Y. Ahn, Y. Bae, S. Lee and Y. S. Kim, J. Hydrogen New Energy, 2022, 33, 723–732.
204. M. Ni, Int. J. Hydrogen Energy, 2011, 36, 3153–3166.
205. O. B. Rizvandi, X.-Y. Miao and H. L. Frandsen, Int. J. Hydrogen Energy, 2021, 46, 27709–27730.
206. T. Yang, Y. Fan, J. Liu, H. Finklea, S. Lee, B. Guan, H. W. Abernathy, T. L. Kalapos and G. A. Hackett, Int. J. Hydrogen Energy, 2022, 47, 41124–41137.
207. S. Yu, D. Schäfer, S. Zhang, R. Peters, F. Kunz and R.-A. Eichel, Energies, 2023, 16, 7841.
208. B. Bosio and F. R. Bianchi, Sustainable Energy Fuels, 2023, 7, 280–293.
209. G. Cinti, V. Liso and S. S. Araya, Int. J. Hydrogen Energy, 2023, 48, 15269–15279.
210. B. Sun, Y. Xu, Y. Liu, Y. Ya, X. Cheng, Y. Quan, H. Ding and J. Xie, Energy Convers. Manage., 2023, 298, 117804.
211. K. Al-Hamed and I. Dincer, eTransportation, 2019, 2, 100027.
212. F. Ishak, I. Dincer and C. Zamfirescu, J. Power Sources, 2012, 212, 73–85.
213. T. Meng, D. Cui, Y. Shi, Y. Ji, M. Cheng, B. Tu and Z. Lan, Int. J. Hydrogen Energy, 2023, 48, 30887–30898.
214. J. Baek, J. Kim, H. Lee, M. Lee and M. Choi, J. Power Sources, 2024, 619, 235194.
215. A. Ashar, L. Wehrle, O. Deutschmann and R. J. Braun, ECS Trans., 2023, 111, 2103.
216. H. Qin, B. Zhang, R. Zhou, Y. Sun, H. Tian, Z. Wang, S. Wu, X. Li and J. Jiang, Appl. Energy, 2024, 372, 123722.
217. M. Kishimoto, Y. Matsui and H. Iwai, J. Power Sources, 2023, 580, 233411.
218. H. Xu, J. Ma, P. Tan, B. Chen, Z. Wu, Y. Zhang, H. Wang, J. Xuan and M. Ni, Energy and AI, 2020, 1, 100003.
219. V. Subotić, M. Eibl and C. Hochenauer, Energy Convers. Manage., 2021, 230, 113764.
220. D. Wang, J. Bao, Z. Xu, B. Koeppel, O. A. Marina, A. Noring, M. Zamarripa-Perez, A. Iyengar, E. Eggleton and D. T. Schwartz, ECS Trans., 2021, 103, 2283.

---