A Review of Proton Exchange Membrane Degradation Pathways, Mechanisms, and Mitigation Strategies in a Fuel Cell

Abstract

Proton exchange membrane fuel cells (PEMFCs) have the potential to tackle major challenges associated with fossil fuel-sourced energy consumption. Nafion, a perfluorosulfonic acid (PFSA) membrane that has high proton conductivity and good chemical stability, is a standard proton exchange membrane (PEM) used in PEMFCs. However, PEM degradation is one of the significant issues in the long-term operation of PEMFCs. Membrane degradation can lead to a decrease in the performance and the lifespan of PEMFCs. The membrane can degrade through chemical, mechanical, and thermal pathways. This paper reviews the different causes of all three routes of PFSA degradation, underlying mechanisms, their effects, and mitigation strategies. A better understanding of different degradation pathways and mechanisms is valuable in producing robust fuel cell membranes. Hence, the progress in membrane fabrication for PEMFC application is also explored and summarized.

1. Introduction

Rapid population growth and industrialization requiring enormous amounts of fossil fuels-based energy have led to excessive emissions of greenhouse gases, mainly CO_2, causing severe environmental issues. Recent environmental awareness and stringent governmental policies have forced the world to transition from quickly depleting and pollution-causing fossil fuel-based energy to green and alternative renewable sources of energy. Hence, researchers are exploring greener alternative sources of energy [1,2] and, at the same time, possible ways to utilize the CO₂ emission streams [3,4,5]. There are several sources of energy, such as wind, solar, geothermal, and hydropower, that are greener and renewable. Nevertheless, the intermittent nature and unpredictability of these sources present significant challenges to ensuring their stable operation as reliable energy sources for end users. Fuel cells, which are electrochemical devices that can convert the chemical energy stored in a fuel into electrical energy without actual fuel combustion, could be a potential solution. Depending on the nature of the electrolyte used, fuel cells can be categorized as polymer electrolyte (proton exchange) membrane fuel cells (PEMFCs), phosphoric acid fuel cells (PAFCs), molten carbonate fuel cells (MCFCs), direct methanol fuel cells (DMFCs), alkaline fuel cells (AFCs), and solid oxide fuel cells (SOFCs) [6,7]. Proton exchange membrane fuel cells (PEMFCs), which convert the chemical energy of hydrogen that can be produced using renewable energy sources such as solar, wind, etc., into electrical energy, have the potential to tackle the issues originating from the intermittent and unpredictable nature of renewable sources [8]. Features such as high compactness, high power density, noiseless operation, and efficient energy production without releasing pollutants or CO₂ make PEMFCs a promising cleaner energy source [9].

PEMFCs mainly consist of membrane–electrode assembly (MEA), bipolar plates, and sealing gaskets in a sandwich structure (Figure 1a). A fuel cell stack is formed by several MEAs, and each MEA is sandwiched between two bipolar plates, providing a foundation and mechanical support by acting as a shield to the fuel cell. Gaskets are added around the edges of the MEA to prevent any gas leaks. The electrochemical reaction takes place at the MEA, which is composed of an anode and a cathode, each containing a gas diffusion layer and an active catalyst layer separated by a proton exchange membrane (PEM). Hydrogen is injected as fuel on the anode side and diffuses through the gas diffusion layer to reach the anode catalyst layer. The catalyst facilitates hydrogen oxidization to produce protons (H⁺) and electrons (e⁻) (1). The proton exchange membrane allows protons to pass through to reach the cathode catalyst layer while electrons are transferred to the cathode via an external circuit. Air is injected on the cathode side and reaches the cathode catalyst layer, flowing through the gas diffusion layer. The cathodic catalyst enables oxygen reduction by reacting with the protons that have crossed the membrane and electrons available from the external circuit (2), producing water and heat using H₂ and O₂ in the overall process (3) (Figure 1b).

$$\mathrm{Anode} \mathrm{reaction}: {\mathrm{H}}_2\to 2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}$$

(1)

$$\mathrm{Cathode} \mathrm{reaction}: {\mathrm{O}}_2+4{\mathrm{H}}^{+}+4{\mathrm{e}}^{+}\to 2{\mathrm{H}}_2\mathrm{O}$$

(2)

$$\mathrm{Overall} \mathrm{reaction}: 2{\mathrm{H}}_2+{\mathrm{O}}_2\to 2{\mathrm{H}}_2\mathrm{O}$$

(3)

The proton exchange membrane is the key element of the PEMFC, as it must permit only the protons to pass while preventing direct electron conduction and gas permeation between the anode and cathode. Such a requirement makes it essential for PEM to have excellent proton conductivity and extremely low gas and electron permeability. Furthermore, PEMFCs are operated in harsh working conditions integrating water, air, hydrogen, and heat, necessitating the PEM to have sufficient chemical, mechanical, thermal, and dimensional stability [10]. Several polymers have been considered as PEMs for their application in PEMFCs under diverse operating conditions [11,12,13,14]. They can be mainly categorized as non-fluorinated and fluorinated polymeric membranes. The latter has been preferred because most non-fluorinated membranes, although having attractive prices, have poor resistance to oxidation and undergo thermal degradation. Fluorinated proton exchange membranes, especially fully fluorinated (perfluorinated) ones, have the advantageous combination of chemical, thermal, mechanical, and dimensional stability.

Nafion, a perfluorosulfonic acid membrane, was developed in the mid-1960s by DuPont and is the most used PEM in PEMFCs. It consists of an aliphatic perfluorocarbon hydrophobic chain and hydrophilic sulfonic acid terminated perfluoro vinyl ether side groups that exhibit excellent proton conductivity and good chemical, oxidative, and dimensional stability [15]. However, it has been found that PEM degradation is one of the significant issues in the long-term operation of PEMFCs. The membrane degradation can lead to a decrease in fuel cell performance and a shorter lifespan. While Nafion has been a dominant proton exchange membrane, several polymers with similar core structures, such as Aquivion^® and Hyflon^® (earlier DOW) by Solvay (Brussels, Belgium), Flemion^® by Asahi Glass, (Chiyoda City, Tokyo, Japan) and Dyneon^TM by 3M (Saint Paul, MN, USA), etc. (Figure 2), have been studied as PEMs in fuel cells [16,17]. Given their similar core structures, the degradation issue is analogous to that of Nafion; therefore, the discussions in this review could be applied to all similar polymers.

The different pathways for PFSA membrane degradation in PEMFCs can be classified into chemical, thermal, and mechanical degradation. We review various causes leading to each route of degradation, the underlying mechanisms, and mitigation strategies. Furthermore, we analyze the progress in membrane fabrication processes for PEMFC applications.

2. Chemical Degradation

The chemical degradation of a proton exchange membrane can be triggered by metal poisoning, impurities in the fuel and the air streams, and radical attack on the polymer structure. Each of these causes, along with the degradation mechanisms, are discussed next. Finally, at the end of the section, overall mitigation strategies for chemical degradation are explored.

2.1. Metal Poisoning

There are two main sources of undesired metal contaminants in the fuel cell system: (i) impurities in fuel streams or humidifier sources and (ii) corrosion of the metal bipolar plates [18,19]. Metal ions can affect the performance of membranes in two ways. First, several metal cations have a stronger affinity to the sulfonic acid groups of membranes than H⁺. Consequently, a significant number of active sites are occupied by undesired cations during fuel cell operation, negatively affecting the membrane ionic conductivity, hydration, and H⁺ transfer rate [20,21]. When hydrogen is produced from water, it may contain cations such as Ca²⁺ and Na⁺ in the fuel stream [22,23]. The same applies to the air stream; it may contain several impurities depending on the site of operation. For example, the air in a sea region or industrial area may contain extremely high concentrations of Na⁺ or airborne particulate matter, respectively [24,25]. The second effect of metal ion contaminants is the catalysis of the radical formation reaction, discussed in Section 2.3.

Some common metal ion impurities that can originate from water in the humidifier, fuel streams, metal alloy components, or alloy catalysts of the fuel cell system include Fe³⁺, Fe²⁺, Ca²⁺, Cu²⁺, Na⁺, Li⁺, K⁺, Mg²⁺, Ni²⁺, Co²⁺, Cr³⁺, and Al³⁺ (

Table 1. Metal ion contaminant sources.

Typical Pollutant	Sources
Ions of Fe, Cr, Ni, Mo, Mn, Cu	Metal bipolar plates
Ions of Ca, Na	Membrane, gaskets, fuel
Ions of Si, Ca, Mg, K	Gaskets
Ions of Si, Na, Ca, Mg, Al, S, K, Fe, Cu, V, Cr	Coolants, DI water

) [26,27].

Each metal ion affects the fuel cell performance and membrane degradation differently. The presence of 5 ppm Fe³⁺ in the air stream resulted in the fuel cell performance decline of 174 mV within 191 h, whereas the presence of 5 ppm Al³⁺ led to a degradation of only 65 mV over a span of 282 h. The stronger effect of Fe³⁺ was attributed to its catalytic effect in the production of radicals, causing the formation of pinholes in the membrane, evidenced by the sudden death of the fuel cell during tests and based on scanning electron microscopy (SEM) analysis [28]. However, another study reported the ranking of four metal cations in terms of the greatest reduction in fuel cell performance as Al³⁺ ≫ Fe²⁺ > Ni²⁺, Cr³⁺ when membranes were contaminated by immersion in metal sulfide solutions before use in the fuel cells [29]. This suggests that the procedure of contamination introduction in the fuel cell itself has a significant impact. Ca²⁺ is present in water in considerably high concentrations, making it an inevitable impurity in air humidifier streams. The injection of 5 ppm and higher Ca²⁺ in the air stream led to large cell performance degradation at 1 A cm⁻² with severe membrane degradation, whereas at 0.6 A cm⁻², only a slight performance loss was observed with the same level of impurity. Furthermore, precipitation of CaSO₄ significant enough to induce mass transport resistance was observed between the gas diffusion layers and bipolar plates of the cathode at all test conditions, even at lower Ca²⁺ concentrations of 2 ppm [30,31]. Moreover, Ca²⁺ causes a reduction in the O₂ permeability by being absorbed in the perfluorinated ionomer in the gas diffusion layer [32]. Mg²⁺, which is also a common metal cation found in water streams and could leach from bipolar plates and elastomeric gaskets degradation, decreases the voltage and maximum power density of the fuel cell with an increase in Mg²⁺ concentration and contamination time [33,34]. Qi et al. investigated the effects of K⁺, Ba²⁺, Ca²⁺, and Al³⁺ by in situ injection of cationic salt solutions with ClO₄⁻ anions in the air stream [35]. The degradation in cell performance in the presence of metal ion contaminants was attributed to the increase in mass transport resistance and the decrease in the electrochemical surface area caused by the replacement of protons on the sulfonate groups. Another study revealed that higher valence cations exhibit stronger interactions and a higher affinity for sulfonic sites compared to lower valence cations [36]. The replacement of H⁺ with metal cations causes physical deterioration of the membrane, such as shrinkage and density increment [37]. Furthermore, the presence of cations decreases the water content, ionic conductivity, H⁺ transference, and water permeability proportionally to the number of cations. Metal cations affect not only the membrane but also the catalyst/support and ionomer. Lee et al. [38] found that Co²⁺ contamination has a negative impact on ionomer conductivity and oxygen transport through the catalyst layer. They discovered that when the membrane area is large compared to the active area being used in MEA, Co²⁺ migrates to the inactive membrane area, acting as a Co²⁺ sink and resulting in a negligible effect of Co²⁺ on performance. However, when the inactive area was decreased to the proximity of the active area, the performance remained unaffected only up to ∼44% of Co²⁺ exchange for a 25 μm thick membrane with an active area of 5 cm² in the total membrane area of 8.6 cm² [38]. The performance reduction was attributed to increased O₂ and proton transport resistance due to reduced water uptake and availability of sulfonic acid groups (Figure 3).

Molecular dynamics (MD) simulations have provided new insights into the mechanisms of metal cation-induced membrane or fuel cell performance degradation. MD simulations have confirmed that Co²⁺ contamination reduces water content and increases the modulus of the ionomer film covering the catalyst surface while reducing the distance between the sulfonic acid groups and increasing the aggregation size of the ionomer [39]. In addition to the effect on membranes, studies suggest that some cationic contaminants prefer to be more concentrated on the cathode side of the PEM when current is drawn through the PEM. Kienitz et al. [40] modeled the effect of Na⁺ and Ca²⁺ by solving the Nernst–Planck equation and found them to be more concentrated on the cathode side due to the tradeoff between diffusion and migration of contaminants in the membrane. By contrast, Zhu et al. [41] found Cu²⁺ to be accumulated at the cathode due to the electric field through the thickness direction of the MEA. As indicated in several experimental studies, MD simulations revealed that the diffusion coefficients of H₃O⁺ and H₂O are significantly reduced by the presence of Na⁺ and Ca²⁺, with Ca²⁺ having more adverse effects. Moreover, cations affected intramembrane transport properties by occupying sulfonic groups and disrupting the connectivity of water clusters [42]. Conforming observations were made in our previous work where sodium ions replaced hydrogen ions in degraded Nafion fragments, and the rate of substitution increased with increasing temperature [24]. The synergetic effect of metal ions and air pollutants is even more devastating, as it was found to accelerate the degradation of stack performance [43].

2.2. Impurities in the Fuel and Air Streams

The air stream and hydrogen fuel stream expose the fuel cell components to several pollutants that can vary greatly in concentration depending on the sources and location of operation. The primary method for hydrogen production involves reforming hydrocarbons or oxygenated hydrocarbons, such as methane from natural gases and methanol from biomass. The reforming process, which includes steam reforming, autothermal reforming, and partial oxidation, dominates over alternative methods, like fermentation or partial oxidation of biomass and water electrolysis [22]. However, the reformation of hydrogen from hydrocarbon fuels introduces inevitable impurities such as carbon oxides (CO₂ and CO) and sulfur compounds (H₂S and sulfur organics). Typically, the reforming process generates a hydrogen-rich gas known as “reformate”, containing approximately 40 to 70% H₂, 15 to 25% CO₂, 1 to 2% CO, along with small amounts of inert gases (nitrogen and water vapor) and sulfur impurities [26]. Additionally, using ammonia as a tracer gas in natural gas distribution systems can result in the presence of NH₃ and HCN in the reformate. Furthermore, hydrogen production from different fuels leads to the presence of several hydrocarbons, oxygenated hydrocarbons, and odorants. In alternative hydrogen productions, the biomass-based process involves gasification, pyrolysis, conversion to liquid fuels by supercritical extraction, liquefaction, and hydrolysis of biomass, followed in some cases by reformation and biological hydrogen production. The main impurities in biomass-sourced hydrogen are aldehydes, alcohols, acetic acid, formic acid, NH₃, H₂S, and HCN. Hydrogen production from water is a mature and traditional hydrogen production method with commercial uses dating back to the 1890s [22]. There are three main ways to generate hydrogen from water: electrolysis, thermolysis, and photoelectrolysis, and the impurities are the ions from water, such as Ca²⁺, Na⁺, and Cl⁻.

Table 2. Sources of different contaminants in the fuel stream.

Typical Pollutants in Hydrogen	Hydrogen Sources
CO, NH3, H2S, HCN	Crude oil, natural gas
Hydrocarbons, aldehydes	Gasolines
Mercaptans	Diesels
CO, odorants, alcohols	Methanol/dimethyl ether
Cations, aldehydes, alcohols, formic acid, NH3, H2S, HCN	Biomass
CO, CO2, SO2, NH3, H2S, benzene	Industrial by-products
Ions of Ca, Na, Cl	Water

summarizes the different contaminants and their sources in the hydrogen stream. For optimal fuel cell performance and longevity, a purer form of hydrogen is required as the fuel feed. To achieve the necessary purity levels, separation processes are essential for removing undesired impurities from the reformate gas. Despite their effectiveness, these processes are associated with high costs. Filtration, especially for trace amounts of CO and sulfur compounds, poses an even greater challenge and comes with a higher price tag.

CO, H₂S, NH₃, and hydrocarbons are considered the main impurities entering the fuel cell system through the hydrogen stream. These contaminants can enter the MEA structure, poisoning the catalyst sites and altering membrane properties, such as hydrophobicity and hydrophilicity and proton transportation routes, and disturbing water management, thus causing performance decay [26]. Although these contaminants mostly cause catalyst poisoning, we discuss the associated effect on the membrane. For example, CO and H₂S interact strongly and are adsorbed on the metal anode catalyst surface, drastically reducing the available active catalyst sites for H₂ electro-oxidation, an essential step for fuel cell operation [44,45,46]. Increasing the operating temperature is one of several catalyst poisoning mitigation strategies [45]. However, Nafion and similar membranes need to be hydrated for proton transport, limiting the operating temperature to below 90 °C. Increasing operating temperatures to tackle the catalyst poisoning may lead to dehydrated and mechanically damaged membranes, causing performance loss. NH₃, as a fuel contaminant, has a more prominent and direct effect on the membrane. Gomez et al. [47] studied the effect of NH₃ on each fuel cell components and overall performance. They found that NH₃ severely degrades the overall fuel cell performance by affecting the membrane conductivity, catalyst ionomer of both electrodes, the hydrogen oxidation reaction, and the oxygen reduction reaction. Zhang et al. [48] found that the effect of NH₃ on the membrane is more intense than that on the electrodes. The electrochemical impedance spectroscopy (EIS) results at a relative humidity (RH) of 49% for both the anode gas and the cathode gas streams revealed that the high-frequency response increases gradually, indicating an increase in membrane resistance due to contamination (Figure 4a). The recovery test indicated that membrane resistance can only recover about 60% (Figure 4b), which appears to be more permanent damage compared to electrodes. It has been speculated by several researchers that the membrane and ionomer conductivity reduction could be due to NH₄⁺ formation (NH₃ + H⁺ → NH₄⁺), which can occupy the charged sites of the PEM [47,48,49]. Furthermore, like CO and H₂S, NH₃ adsorption onto metal catalysts (mostly Pt) may block H₂ oxidation at the anode and, similarly, the oxygen reduction reaction at the cathode catalyst by crossing over to the cathode side [23].

Hydrocarbon impurities in the fuel stream do not generally have a drastic effect on fuel cell performance or membrane stability. Therefore, the studies have also been limited. Common hydrocarbon methane is believed to have no poisoning effect, while formaldehyde (HCHO) and formic acid (HCOOH) show a very small and reversible catalyst poisoning effect [50]. Methanol in the hydrogen stream, depending on its concentration and exposure time, results in reversible degradation in the performance of the PEMFC [51]. The poisoning effect of methanol at low concentrations, up to 200 ppm, is reversible, and beyond that, it takes a very long time to recover. Another study determined that 5 ppm HCHO, 2 ppm HCOOH, 19 ppm chloromethane, 30 ppm acetaldehyde, 5% ethylene, 20 ppm toluene, and 10 ppm benzene do not lead to significant degradation of the fuel cell when present individually in the fuel stream [52]. Furthermore, up to a certain concentration, a mixture of CO, H₂S, HCOOH, benzene, and ammonia does not significantly affect fuel cell performance. Overall, hydrocarbon impurities, in some cases, affect the fuel cell performance; however, the effect on membrane stability is not clear yet. The effects of cations are analogous to what has been discussed in Section 2.1.

The air stream is the most convenient way to feed oxygen to fuel cells. However, ambient air, depending on the location, contains several impurities that are detrimental to the fuel cell membrane electrode assembly and overall performance [53]. Nitrogen oxides (NO and NO₂), carbon oxides (CO and CO₂), sulfur oxides (SO₂ and SO₃), O₃, hydrocarbons, soot, and particulates are common pollutants originating from fossil fuel combustion. Considerable amounts of NH₃, ions of salts, and dust are present in ambient air in farming areas, seaside, and industrial zones, respectively. De-icers and humidifiers introduce several metal cations into the fuel cell system. Other air impurities that can be encountered are chemical warfare agents, including SO₂, NO₂, CO, propane, and benzene present on a battlefield.

Table 3. Sources of different contaminants in air stream.

Typical Pollutants in Air	Sources
SOx, NOx, COx, O3, hydrocarbons, soot, particulates	Fuel combustion, pollution
NH3	Ambient air, farming
Cations and anions from ocean salts, dust	Natural sources
Ions of Na, Ca, Mg, Al, K, Fe, Cu, V, Cr, Cl	De-icers, humidifiers
SO2, NO2, CO, propane, benzene	Battlefield pollutants

lists the different air contaminants and their sources.

The contaminants in the air stream that are similar in nature to those in the fuel stream have similar effects, i.e., the poisoning of the catalyst on the cathode side and influencing the oxygen reduction reaction, and if they cross the membrane, they affect H₂ oxidation on the anode side. Other contaminants that are not present in the fuel stream also have an analogous effect. For instance, exposure to NO and NO₂ leads to a current drop even at concentrations of 1 ppm [54]. However, the effects are reversible and even faster at elevated temperatures. Another study reported the reversibility of the NO₂ impact until at least 4 ppm, which is much higher than the NO₂ concentration that can be encountered in real PEMFC operations [55]. Misz et al. [56] conducted a systemic review of the effect of different air contaminants on fuel cells and experimentally investigated the effects of NO, NO₂, SO₂, NH₃, toluene, and ethane. Contamination caused a significant degradation in performance, mostly due to the effect on the cathode with the possibility of regeneration. PEM is mainly affected by two types of species that can originate from air impurities, namely NH₄⁺ and metal cations [43,57]. The effect of metal cations on membranes has been discussed in Section 2.1. NH₄⁺ can originate in fuel cell systems through different paths. One route has been discussed above, where NH₃ enters the anode side as a fuel impurity and can react with H⁺ to form NH₄⁺, occupying the charged sites. A similar reaction can occur when NH₃ enters the fuel cell as an air impurity, causing the same issue of reduced membrane conductivity. The other route of NH₄⁺ formation is through the reduction of NO_x impurities according to Reactions (4) and (5).

NO₂ + 8H⁺ + 7e⁻ → NH₄⁺ + 2H₂O

(4)

NO + 6H⁺ + 5e⁻ → NH₄⁺ + 2H₂O

(5)

NH₄⁺ has several detrimental effects on the membrane. Hongsirikarn et al. [58] compared the conductivities of Nafion membranes with different NH₄⁺ concentrations in deionized water at room temperature and in a gas phase at 80 °C at various relative humidities. The conductivity in the water phase decreased linearly with an increase in the NH₄⁺ concentration in the membrane, with a 75% decrease in conductivity for the fully exchanged NH₄⁺ form compared to the membrane, with SO₃⁻ groups fully occupied by H⁺. By contrast, depending on humidity, the gas phase conductivity of the NH₄⁺ form exhibited a reduction ranging from 66% to 98% compared to the H⁺ form. The conductivity of a cation generally decreases with an increase in its radius [59]. The Pauling cation radius of NH₄⁺ (∼1.52 Å) is approximately four times larger than that of H⁺ (∼0.4 Å), which explains the drastic decrease in membrane conductivity once H⁺ is exchanged with NH₄⁺. Furthermore, membrane thinning due to NH₄⁺ uptake was observed, which is possibly caused by the formation of hydrogen bonds by a lone pair of electrons of nitrogen and of hydrogen in the N–H bond with adjacent water molecules (H–N⋯H–O–H, N–H⋯H–O–H) and thus a tighter, more compact structure formation [58]. Furthermore, SO₂ and NO₂ in the air stream are dissolved into water and generate H₊, SO₄²⁻, and NO₃⁻, reducing the pH of the water on the surface of the alloy bipolar plates [43]. Reduced pH leads to accelerated corrosion of the air outlet at the end plate, which may result in metal ion contaminants known to accelerate membrane degradation. Recently, there has been tremendous interest in using fuel cells for military applications, which raises concerns due to battlefield pollutants [60]. Moore et al. [61] investigated the impact of chemical warfare agents, sarin, sulfur mustard, cyanogen chloride, and hydrogen cyanide on PEMFC performance. All chemical warfare agents tested seriously compromised the performance of the fuel cells irreversibly.

2.3. Radical Attack

The degradation of PEM due to radical attack is the most disastrous for fuel performance and longevity, as it dismantles the fluorocarbon backbone in membranes, affecting the proton conductivity and mechanical strength [24,62]. Radical attack on the membrane is considered the most prominent reason for membrane chemical degradation. Oxygen molecules from the air stream permeate the membrane from the cathode side and are reduced at the anode metal catalyst to form hydrogen peroxide via reactions (6)–(8).

H₂ + Pt → Pt-H (at anode)

(6)

Pt-H + O₂ (permeated through PEM) → •OOH

(7)

•OOH + Pt-H → H₂O₂

(8)

H₂O₂ is a very reactive species; however, H₂O₂ itself is not a capable oxidant to substantially damage the polymer structure [63]. In fact, perfluorosulfonic acid membranes are stable against 30% H₂O₂ up to 80 °C in the absence of metal ions or other sources generating radicals, such as UV light [64]. Contamination by trace metal ions, mainly Fe²⁺, Cu²⁺, and other multivalent cations like Cr²⁺ or Ti³⁺ originating from different sources (Table 1), catalyze the decomposition of H₂O₂ to produce hydroxyl (•OH) (9) and hydroperoxyl (•OOH) (10) radicals. •OH can further react with H₂ to generate •H (11) [65].

H₂O₂ + M^x+ → M^(x+1)+ + •OH + OH⁻

(9)

•OH + H₂O₂ → H₂O + •OOH

(10)

•OH + H₂ → H₂O + •H

(11)

The produced radicals attack and degrade the membrane in several ways. In situ membrane degradation is tested by open circuit voltage hold testing [66], while the Fenton reagent, consisting of hydrogen peroxide and ferrous ions, is employed to study the membrane degradation ex situ [24]. The ex situ investigations give better insight into the degradation behavior of PEM under different conditions that can be mimicked precisely. There are four vulnerable sites in perfluorosulfonic acid membranes where these radicals can attack [67]. The first is the carboxylic acid end groups, which are inevitably generated during the polymer manufacturing process or through the transformation of other non-perfluorinated end groups via radical attack [68]. The highly unstable radicals formed through the reactions described above (9) and (10) attack the carboxylic groups to produce CO₂, HF, and a perfluorocarbon radical (12) (Figure 5) [69]. The produced perfluorocarbon radical reacts with another •OH to form an acid fluoride and HF (13) and (14). Finally, the acid fluoride undergoes hydrolysis to generate carboxylic acid end groups in the polymer chain with one carbon removed (15). The carboxylic acid end group generation makes this mode of degradation continuous, with each cycle removing one carbon from the chain, resembling the unzipping process. When all carbons before the side chain (m) become unzipped through this process, the degradation product CO₂H–CF(CF₃)–O–CF₂–CF₂–SO₃H is produced, which may further decompose to HO–CF₂–CF₂–SO₃H and trifluoroacetic acid due to •OH attack on the ether bond [70]. Furthermore, the degradation product itself can undergo a further unzipping process starting from the carboxylic acid group and produce HOOC–CF₂–SO₃H and trifluoroacetic acid [71]. A recent study has found evidence of chemical degradation via the unzipping process due to radical attack, observing weight loss, fluoride, and CO₂H–CF(CF₃)–O–CF₂–CF₂–SO₃H emissions [72].

R–(CF₂)_mCOOH + •OH → R−CF₂• + CO₂ + H₂O

(12)

R−(CF₂)_m−1CF₂• + •OH → R−(CF₂)_m−1CF₂OH

(13)

R−(CF₂)_m−1CF₂OH → R−(CF₂)_m−1COF + HF

(14)

R−(CF₂)_m−1COF + H₂O → R−(CF₂)_m−1COOH + HF

(15)

Mittal et al. [73] found that the presence of sulfonic acid groups occupied by H⁺ in the polymer side chain significantly affects the fluoride emission rate and hence membrane degradation. If the sulfonic acid group is occupied by metal cations instead of H⁺, the fluoride emission rate decreases by two orders of magnitude, suggesting the significant role of sulfonic acid groups in membrane degradation. Consequently, the second vulnerable site in the polymer chain is the C–S bond at the end of the side chain. The •OH radical attacks the C–S bond of the side chain, which is also the weakest bond within this segment [74]. In the first step, SO₃, H₂O, and perfluorocarbon radicals are produced (16), and from there, polymer degradation proceeds similarly to the unzipping mechanism (17)–(19), progressively shortening the side chain with HF and CO₂ formation (Figure 6). The detection of –O–CF₂–CF₂• radicals after a UV-induced Fenton test of Nafion and Dow perfluorinated membranes aligns with this degradation mechanism [75].

R–O–CF₂–CF₂–SO₃H + •OH → R–O–CF₂–CF₂• + SO₃ + H₂O

(16)

R–O–CF₂–CF₂• + •OH → R–O–CF₂–CF₂OH

(17)

R–O–CF₂–CF₂OH → R–O–CF₂–COF + HF

(18)

R–O–CF₂–COF + H₂O → R–O–CF₂–COOH + HF

(19)

Tertiary C-F bonds are the most susceptible to radical attack among the fluorocarbons because they have the lowest bond strength [76]. The side chain degradation or main chain scission may also start at the tertiary carbon in the main chain or the tertiary carbon between two ether groups within the side chain. These reactions can lead to complete or partial unzipping of the side chains, releasing HF [74]. The •H removes the fluoride from the tertiary carbon to produce HF and tertiary carbon radical (Figure 7) [76]. The tertiary carbon radical reacts with 2 •OH to release H₂O and a fragment of the main chain while producing a carbonyl at the tertiary carbon. Carbonyl at the tertiary carbon converts the ether linkage between the main and side chain to the ester group, which undergoes hydrolysis to produce two fragments, the main chain with a carboxylic acid end group and the side chain with an alcohol group. The side chain fragment further reacts with water to release 2 HF while generating the carboxylic acid end group on the side chain as well. The degradation through this mechanism releases a relatively low amount of fluoride; nevertheless, an increase in COOH groups is observed, which can lead to degradation by unzipping.

The ether group in the side chain has been proposed as another site for radical attack [77,78]. There are several possibilities through which the ether bond closer to the C-S bond, which is the preferred site for radical attack, can be cleaved (Figure 8). However, in all cases, the fragmented part ends up with a carboxylic acid end group, which would follow the unzipping mechanism, as explained above.

Conventionally, hydroxyl (•OH) or hydrogen atoms (•H) are considered the main radicals responsible for membrane degradation. Recent research has indicated the presence of hydronium radical (H₃O•), which can be produced by the reaction between •H and H₂O in PEMFCs [79]. The ab initio modeling suggested that H₃O• can be stabilized by the sulfonic anion on the polymer side chain. The polymer side chain can undergo a conformational change with the help of water, leading to a greatly reduced barrier for F extraction from the tertiary carbon (Figure 7), supporting the •H attack on the tertiary carbon mechanism.

All these radical attacks on the membrane cause a release of fluoride; different acids, including HF; and polymer fragments in the fuel cell system. The membrane resistance increases due to the loss of conducting functional groups. Weight, density, and water uptake properties are changed due to changes in the polymer backbone structure. Local or uniform membrane thinning causes severely reduced mechanical properties. Mostly, the formation of pinholes and cracks takes place, causing a high gas crossover rate, which is responsible for severe performance decay [80].

2.4. Mitigation Strategies

The main mitigation strategies to avoid chemical degradation are to avoid foreign impurities that can trigger radical formation and to improve the stability of membranes against radical attacks. To avoid the impurities that trigger unstable radical formation, especially the Fe²⁺ and Cu²⁺, the materials of the bipolar plate can be modified. Composite bipolar plates could be one way to avoid such metal impurities coming from bipolar plate corrosion [81]. For the contamination coming through the air streams, several researchers have suggested using a more efficient air filter in the fuel cell system [43,82]. Filtering out NOx, sulfide, or other acidic substances from the air reduces the impact of gas impurities on fuel cell stack corrosion [83,84]. Several mitigation strategies, including on-site pre-treatment of reformation during hydrogen production, onboard CO removal, in operando CO mitigation, and the development of CO-tolerant catalysts, have been proposed to tackle the CO poisoning of the fuel cell [44,85]. Finally, to mitigate the radical attack on membranes, enhanced radical-tolerant membranes should be developed and commercialized [86]. Furthermore, some tuning of operating conditions can also be helpful in reducing membrane degradation. A moderate operating temperature, a lower hydrogen pressure, and a stable and uniform humidity distribution within the cell have been suggested to slow down membrane degradation rates [87].

3. Mechanical Degradation

The mechanical degradation of the membrane is caused by several factors, which are discussed next. Fuel cell testing over an extended period shows time-dependent deformation, such as micro-crack fractures appearing in the membrane. Such mechanical degradation results from cyclic stress applied to the membrane and strain from changes in the operating conditions. Relative humidity cycling, transients at open circuit voltage, temperature cycling, and potential cycling are among the most important factors causing mechanical defects in the membrane [88,89]. Some of these factors affect the mechanical robustness of polymers more significantly than others; for instance, temperature cycling has been reported to have less influence on membrane breakdown than RH cycling.

3.1. Mechanical Stress

Localized mechanical stress caused by foreign particles, the gas diffusion layer’s roughness, non-homogeneous conditions within the cell, and uneven clamping stress promote cracks and their expansion [90,91]. Moreover, the cyclic swelling and shrinking of the membrane reduce its strength, which finally leads to permanent plastic deformation, poor membrane–electrode interfacial contact, and micro defects [88]. Additionally, the active and non-active areas of the membrane expand to a different extent, causing mechanical stress at the corners [92]. Apart from the improper PEMFC manufacturing process, leaving particles, non-uniform clamping stress, and chemical degradation also make the membrane susceptible to mechanical degradation. Once the membrane is chemically degraded, its mechanical properties are reduced, and it becomes more vulnerable to the effects of changes in humidity, temperature, and pressure [93,94]. Moreover, the molecular weight of the membrane material can be significantly reduced by chemical degradation, causing the membrane to be brittle and accelerating the fracture and break due to uneven mechanical stress on the membrane [87].

3.2. Hygrothermal Cycling

The humidity of the membrane significantly affects its mechanical properties, such as strength and modulus. Hygrothermal cycling refers to the cyclic increase and decrease in temperature and humidity, especially during start-up procedures, that can influence the strength, modulus, and elongation at the break of the membrane. Kundu et al. [95] studied the mechanical properties of Nafion membranes under dry and hydrated conditions and found that membranes in a hydrated state exhibit reduced strength and modulus. Tang et al. [96] found that RH cycling of Nafion membranes from water-soaked state to 25% RH state caused cyclic stress up to 2.23 MPa with a dimensional change of ∼11%, while the safety limit of the cyclic stress for the particular Nafion membrane was only ∼1.5 MPa (Figure 9a).

3.3. Freeze–Thaw Cycling

The freeze–thaw cycle of the water confined in the MEA can lead to fuel cell performance loss and membrane degradation. The increased volume of ice in the enclosed MEA leads to the development of stress, causing microcrack growth and deformations [97]. McDonald et al. [98] studied the effects of freeze–thaw cycling on the chemical and physical properties of Nafion by subjecting it to 385 freeze–thaw cycles between +80 °C and −40 °C over a period of three months. The freeze–thaw cycling of membranes resulted in decreased strength and elongation at break.

3.4. Mitigation Strategies

To avoid mechanical degradation of membranes due to foreign particles, the gas diffusion layer’s roughness and processing and manufacturing conditions should be improved. For example, Taylor et al. [99] suggested a calendering technique to flatten stray fibers within the gas diffusion media to mitigate pinhole formation in hot-pressed MEAs. Bipolar plate and cell design should be optimized to ensure uniform contact pressure. Excessive pressure cycling, overpressure, and pressure gradients should be avoided. The reinforcement of the membrane with, e.g., a porous polyethylene, PTFE, or oxides, enhances the dimensional stability and reduces the shrinkage stress in the membrane during drying [90,100]. Gas purging is among the thermal management strategies employed to control the temperature of water within the membrane during shutdown. Other approaches, such as heat preservation to prevent freezing and passive/active rapid startups from sub-zero environments, have also been explored [101]. Additionally, exposing the cell to dry gas (Figure 9b) or filling its components with an antifreeze solution have been proposed as effective ways of averting ice accumulation in the MEA [102,103]. Furthermore, optimizing membrane humidity management involving avoiding air supply with supersaturated steam, minimizing humidity cycling, particularly at elevated humidification levels, and ensuring operating conditions that prevent membrane dry-out could limit membrane degradation [87].

Figure 9. (a) Shrinkage stress of Nafion membrane at 25 °C at different humidities. (b) Effects of thermal cycle from 80 to −10 °C on the open-circuit voltage and current density after gas purging process. Adapted with permission (a) from [96] and (b) from [102].

Figure 9. (a) Shrinkage stress of Nafion membrane at 25 °C at different humidities. (b) Effects of thermal cycle from 80 to −10 °C on the open-circuit voltage and current density after gas purging process. Adapted with permission (a) from [96] and (b) from [102].

4. Thermal Degradation

High-temperature operating conditions can increase the ion transport capacity of the membrane and promote the electrochemical reaction. However, if the temperature is excessively high (thermal stress), membrane failure can occur due to reduced water content (drying) and material degradation.

4.1. Thermal Stress and Cycling

The most favorable working temperature of a PEM fuel cell is usually from 60 to 80 °C to maintain the adequate hydration of the membrane. The conventional perfluorosulfonic acid membranes undergo serious breakdown at high temperatures because their glass transition temperature is around 80 °C [18]. Frensch et al. [104] found that the fluoride emission rate (Figure 10a) from Nafion and membrane thinning (Figure 10b) significantly increases at temperatures above 80 °C. Furthermore, hot spots or excess heat can severely accelerate membrane degradation. Current research mainly focuses on mechanical degradation and chemical degradation since PEMFCs are operated at lower temperatures. However, high-temperature operation can improve fuel cell performance by increasing the ion transport capacity of the membrane, enhancing the electrochemical reaction rate, and improving the water management system. Consequently, significant efforts have been made to develop PEM fuel cells that operate at a temperature of >100 °C. Nonetheless, the proton conductivity of PFSA membranes drops drastically when the fuel cell is operated at high temperatures due to reduced water content [105]. The mechanism of thermal degradation has not been sufficiently studied. The most widely accepted one is that high temperature leads to the decomposition of the side sulfonate acid group, which is similar to chemical degradation [106,107]; however, for such degradation, a considerably high temperature is required. In addition to high temperature, the subfreezing temperature (freeze–thaw cycling) causes membrane degradation, which has been discussed in Section 3.3.

4.2. Drying

The drying of the membrane is related to high-temperature operation, which reduces the water content of the membrane. The proton conductivity of the membrane degrades significantly in dehydrated conditions at a high temperature and low humidity [105]. Furthermore, dry membranes are susceptible to crack formation, which critically affects their mechanical stability and promotes the formation of pinholes and gas crossover [108]. Apart from high-temperature operation, contaminant ions contribute to membrane drying and conductivity reduction [109,110]. Furthermore, long-term application may cause irreversible drying due to changes in the chemical nature of the membrane. Collette et al. [111] found that membrane degradation causes the alteration of sulfonic acid end groups when used over a period of 400 days. They discovered that the condensation of sulfonic acids produces sulfonic anhydride, leading to a considerable decrease in water uptake and thus the hydrophilicity of Nafion due to the substitution of ionic end groups with less hydrophilic anhydrides.

4.3. Mitigation Strategies

Thermal material degradation of membranes can be avoided by using moderate temperatures for operation, especially for PFSA membranes. The drying can be avoided by operating under proper humid conditions or using a humidifier. However, to operate the fuel cell at a higher temperature, membranes with better thermal stability and adequate conductivity under low humidity conditions should be developed. The thermal stability of membranes can be improved by using reinforcement or through modifications of the chemistry of the membrane materials. For example, a phosphoric acid-doped polybenzimidazole membrane has been developed to achieve higher thermal stability [112]. Furthermore, nanocomposite membranes with several fillers, such as SiO₂, mordenite, zirconia, TiO₂, and graphene oxide, have been explored to improve mechanical and thermal stability [113]. Degradation due to hot spots or excess heat can be avoided by proper heat management and water management and their combined optimization [114,115,116,117].

5. Progress in PEM Development for Fuel Cells

The advances in PEM fuel cell membrane development target improving thermal stability for high-temperature applications, along with enhanced conductivity and mechanical properties [118,119,120]. The approaches can be divided into (i) modifications in the chemistry of membrane materials and (ii) developing composite membranes.

5.1. Chemically Modified PEMs

Several attempts have been made to modify the structure of Nafion to achieve advantageous properties. The research has led to several commercial membranes, such as Aquivion^®, Hyflon^®, Flemion^®, and Dyneon_TM, based on shorter side sulfonic chains compared with Nafion [119]. Furthermore, the whole range of new materials has been tested as proton exchange membranes for high-temperature fuel cell applications. Several sulfonated polymers, such as polyether ketones, polyphenylsulfone, and poly(arylene ether), have been widely studied as PEMs due to their good stability at high temperatures and low cost [121,122,123,124,125,126]. However, sulfonated polyether ketone membranes, like the PFSA membranes, undergo degradation due to dehydration under high operating temperatures. Polybenzimidazole (PBI) is an aromatic linear heterocyclic rigid semi-crystalline polymer containing repeated benzimidazole units with excellent chemical resistance, high mechanical strength, good moisture regain, and thermo-oxidative stability above 80 °C [127]. PBI and acid-doped PBI-based membranes have been considered for high-temperature applications because of their excellent chemical and thermal properties at elevated temperatures without the need for humidification and their low cost [11,113]. Different acids, such as HNO₃, H₂SO₄, HCl, and HClO₄, can be doped into PBI membranes to act as proton donors. However, phosphoric acid (H₃PO₄) has been a favored dopant because compared with others, it induces higher conductivity and excellent thermal stability in PBI membranes without the need for high RH [128,129]. Jin et al. [130] developed poly(arylene pyridine) using one-step Friedel–Crafts polymerization of 4-acetylpyridine and para-terphenyl/biphenyl; the resulting membrane from the polymer exhibited excellent phosphoric acid uptake (up to 220%) with a conductivity of 0.102 S cm⁻¹ at 180 °C. Hu et al. [131] fabricated a phosphoric acid-doped triazole-functionalized poly(arylene perfluorophenyl)-based membrane with similar conductivity at 180 °C (0.109 S cm⁻¹), whereas Li et al. [132] synthesized poly(terphenyl pyridine) polymers using 4-acetylpyridine, p-terphenyl, and comonomer 2,2,2-trifluoroacetophenone via the acid-catalyzed polycondensation reaction. The prepared membranes using the synthesized polymer doped with phosphoric acid exhibited a proton conductivity of 0.142 S cm⁻¹ at 160 °C under anhydrous conditions and a peak power density of 395 mW cm⁻² at 180 °C in a single fuel cell fueled by non-humidified H₂ and O₂. The major issues with the H₃PO₄-doped PBI membranes are catalyst poisoning and other types of fuel cell component degradation due to acid leaching and undesirable electrochemical reactions. Another chemically different membrane is an acid–base membrane in which a non-aqueous solvent, like acid or protic ionic liquid, replaces water and acts as a proton carrier while a solid material provides the medium for proton conduction. The proton-conducting medium is mostly based on polymers with basic moieties, such as amine, amide, imide, alcohol, ether, or aprotic ionic liquid, which can react with acid to form a complex acid–base polymer membrane with proton conductivity (such as acid-doped PBI). For example, Huang et al. [133] incorporated hindered amine and imidazole in the PFSA membranes, forming an acid–base crosslinking and interpenetrating network structure for continuous proton transport and providing membranes with excellent dimensional stability and radical scavenging properties. An increase in the proton carrier acid content of the membrane can improve the proton conductivity but leads to poor mechanical stability, especially at temperatures greater than 100 °C [113].

In addition to improving the mechanical and thermal stability, the chemical modification of the membrane to incorporate radical scavenging properties has also been attempted. Wang et al. [134] grafted organic free radical scavengers, such as 2-mercapto-1-methylimidazole, 3-mercapto-1,2,4-triazole, and 2-mercaptobenzimidazole into the polyarylethersulfone backbone. Around 200 wt% phosphoric acid-doped 2-mercapto-1-methylimidazole-grafted polyarylethersulfone membranes exhibited an anhydrous proton conductivity of 78.3 mS cm⁻¹ at 180 °C, a tensile strength of 7.8 MPa at room temperature, and a peak power density of 423 mW cm⁻² at 160 °C with a fuel cell fueled by non-humidified H₂ and O₂. Agarwal et al. [135] incorporated phosphonic acid into Nafion membranes as a radical scavenger and compared the performance with membranes incorporated with widely the reported radical scavenger cerium. The fluoride emission rates of phosphonic acid-Nafion membranes were much lower than the baseline (non-modified Nafion) and cerium-incorporated Nafion (Ce), especially for the phosphonic acid with the shortest perfluoro chains (C6) (Figure 11a). The ratio of fluorescence intensity (I) after the addition of Fenton’s reagent with and without scavengers to the fluorescence intensity of the undegraded dye (I₀) exhibited higher retention of the intensity of the fluorescent dye (I) for phosphonic acid–Nafion compared to cerium–Nafion, indicating that phosphonic acid has a higher radical scavenging capability (Figure 11b). Even the conductivity (although not surprising given that phosphonic acids are proton conductors) and water uptake of phosphonic acid–Nafion increased (Figure 11c,d). However, the strength of the phosphonic acid–Nafion decreased from baseline, while the strength of cerium–Nafion increased from baseline, although this made it more brittle, with a higher modulus and a lower elongation at break (Figure 11e). Moreover, the Nafion membrane itself has been chemically modified to improve the radical tolerance and proton conductivity. Teixeira et al. [136] doped Nafion membranes with bisphosphonic acids and found that the doped membranes have better chemical stability after oxidative degradation under Fenton’s test conditions at 80 °C.

5.2. Composite PEMs

The sulfonic acid groups of PFSA membranes can undergo dynamic cross-linking with hygroscopic inorganic materials, like Al₂O₃, TiO₂, SiO₂, ZrO₂, mordenite, graphene oxide, ionic liquids, and layered double hydroxide nanoparticles [113]. The dispersion of these particles in the polymer matrix enhances water retention and porosity while reducing gas crossover in the membrane [15,137]. However, incorporating hydrophilic inorganic additives into conducting membranes generally decreases proton conductivity due to the insulation effects of fillers. Numerous strategies have been employed to improve conductivity, including (i) functionalizing inorganic additives, (ii) integrating inorganic binary component materials, and (iii) introducing proton-conductive fillers [138]. Each type of inorganic particle provides specific benefits to membranes; for example, silica particles and ZrO₂ reduce membrane swelling, increase water uptake, and improve proton conductivity [139,140,141]. TiO₂ reduces methanol crossover, improves mechanical and thermal properties, and increases water absorption [142,143]. However, the low compatibility between polymers and fillers may impact the physical and mechanical properties of the membranes [144]. Carbon-based nanomaterials, such as carbon nanotubes [145], graphene [146], fullerenes [147], and nanodiamond [148] are promising reinforcements for improving the performance of PEM. Metal–organic frameworks (MOFs) are crystalline porous materials with three-dimensional network structures formed by organic–inorganic self-assembly [15,149]. MOFs have been extensively investigated as fillers in PEMs due to their ability to be tuned for target applications. MOFs can improve the water retention ability and proton conductivity of PFSA membranes [150]. Ionic liquids (ILs) are molten salts in the liquid state at room temperature, typically defined as liquid electrolytes composed entirely of ions without the use of solvents. Some ILs have excellent ionic conductivities even under anhydrous conditions and are, therefore, widely studied as reinforcement in membranes for PEMFC applications [151,152]. Polymeric reinforcement is another technique that has been investigated for developing composite PEMs [143]. Several polymers, such as polytetrafluoroethylene, polyvinylidene fluoride, and polybenzimidazole, have been incorporated into PSFA membranes to improve their performance [138,143].

Several nanomaterials with radical scavenging properties have been incorporated into membranes to improve radical tolerance, with cerium-based materials being the most popular [153,154,155]. Tinh et al. [156] used cerium oxide particles (5–7 nm) produced by the sol-gel method to fabricate durable layered membranes containing cerium oxide spheres on both surface layers. The smaller particles did not affect the proton conductivity of the membrane while providing radical scavenging effects. Cerium is widely used and is an effective radical scavenger; however, the migration of cerium ions to the catalyst layer during fuel cell operation causes a loss in durability and performance due to adverse effects on various components of the fuel cell. Xu et al. [157] co-doped organic antioxidant alizarin and cerium ions to fabricate PFSA composite membranes with improved oxidative stability and the reduced migration of cerium ions from the membrane. Agarwal et al. [158] used 15-Crown-5 to immobilize cerium within PFSA membranes by forming an organometallic complex of cerium with 15-Crown-5. Another way of encapsulating free radical scavengers could be fabricating a thin conductive layer on the CeO_x-containing membranes [159]. Zhiyan et al. [160] changed the morphology of ceria from conventional nanoparticles to nanorods, significantly reducing the migration of cerium. Moreover, ferrocyanide, ferricyanide, and MnO_x are among other radical scavenging materials being used to fabricate composite PEMs [161,162,163].

6. Conclusions and Outlook

Proton exchange membrane fuel cells have the potential to tackle major challenges associated with fossil fuel-sourced energy consumption and accompanying environmental complications. However, PEM degradation is one of the significant issues in the operation of PEMFCs in the long term or at subfreezing and elevated temperatures. There are several routes of membrane degradation, which can be divided into chemical, mechanical, and thermal degradation. We have reviewed different pathways of PFSA degradation, the underlying mechanisms, and the mitigation strategies. The chemical degradation of the membrane caused by radical attack, which is accelerated by the presence of metal impurities such as Fe²⁺ and Cu²⁺, is the most severe threat, which can be avoided by using composite bipolar plates. There are several advantages of composite bipolar plates, such as their light weight, high corrosion resistance, and stability under harsh conditions; however, it has been difficult for composite materials to satisfy the electrical conductivity requirements due to the poor intrinsic electrical properties of polymers [81]. Therefore, extensive research is needed to develop high-performance, advantageous composite bipolar plates. The mechanical degradation of membranes is mostly caused by non-uniform conditions, cyclic humidity change, and cyclic thermal stress, and these triggers can be avoided by membrane modification, better manufacturing practices, and optimized operating conditions. The research should be focused on fabricating robust membranes and optimizing the thermal, water, and gas management in the fuel cell system. Furthermore, the use of conventional PFSA membranes such as PEM shows limitations in terms of low proton conductivity in anhydrous conditions and degradation in mechanical and physical properties at elevated temperatures. Consequently, several advances in the manufacturing of PEMs have been made to develop membranes that could be used under high-temperature and low-humidity conditions and that are tolerant to impurities and radical attacks. However, these membranes still have some limitations, such as the leaching of acid or metal ions from fillers. Recently, many advancements have been made, especially in membranes incorporated with radical scavengers. Nevertheless, most of these membranes have yet to reach the level of commercial applications that PFSA membranes have reached. Therefore, producing robust fuel cell membranes tolerant to diverse operating conditions, impurities, and radical attacks at the commercial level is still a fruitful direction of research and will boost the widespread application of fuel cells in diverse conditions.