Highly sulfonated carbon nano-onions as an excellent nanofiller for the fabrication of composite proton exchange membranes with enhanced water retention and durability

Highlights

• Highly sulfonated carbon nanon-onions (SP–CNOs) were synthesized.

• SP-CNOs enhanced water retention and proton conductivity of the composite PEMs.

• The composite PEMs showed prominent mechanical and antioxidative stabilities.

• 0D SP-CNOs was a promising candidate for the constructing long-range H⁺-channels.

• The composite PEMs displayed high fuel cell performance and durability.

Abstract

Highly sulfonated carbon nano-onions (SP–CNOs) with large specific surface area are used as a new type of nanofiller in sulfonated poly(arylene ether sulfone) (SPAES) to construct and adjust the proton transfer channels efficiently. SP-CNOs are synthesized from nanometer diamonds via thermal annealing, phenylation and sulfonation, and characterized by XRD, XPS, BET, Raman and HRTEM. The obtained mesoporous SP-CNOs possess a large specific surface of 350 m²/g and an average pore size of 9.7 nm, which endow the nanoparticles with good dispersivity and hydrophilicity. A series of composite membranes based on SP-CNOs/SPAES are prepared through the solution casting approach and evaluated by FESEM, XRD, TGA, water uptake, proton conductivity, chemical stability and fuel cell performance. The results indicate that the composite membranes all show excellent mechanical toughness and greatly enhanced water-retention capacity, thermal, dimensional and oxidative stability due to the good interfacial compatibility and the formation of hydrogen-bond interaction between SP-CNOs and SPAES. The SPAES/SP-CNOs-1.5 membrane achieves the highest proton conductivity of 181.2 mS/cm at 90 °C, which is 45% higher than that of SPAES; H₂/O₂ fuel cell performance records a power density of 735 mW/cm² at 80 °C, which is slightly better than that of Nafion® 112. In addition, the SPAES/SP-CNOs-1.5 membrane undergoes the CV decay of 0.38 mV/h after 168 h at 80 °C, which is comparable to Nafion® 112 (0.43 mV/h) but lower than that for the pristine SPAES membrane (0.54 mV/h). After the accelerated stress test, the SPAES/SP-CNOs-1.5 membrane exhibits superior cell performance and lower hydrogen crossover than the pristine SPAES membrane.

Introduction

Proton exchange membrane fuel cell (PEMFC) has drawn extensive interest for its advantages of high efficiency, low pollutant emission, and rapid start-stop. It is considered as a promising and environmentally friendly power source for both portable and automobile applications [[1], [2]]. The major challenges for commercializing PEMFC are to improve the performance and durability of the elements that constitute the fuel cell at a bare minimum cost. The key element of a PEMFC is the membrane electrode assembly (MEA), which comprises three main components of gas diffusion electrodes (anode and cathode) and proton exchange membrane (PEM) [3]. Among them, PEM acts as a barrier for hindering fuel/oxidant mixing between electrodes and a proton carrier, which has great influence on PEMFC performance and stability [4]. During fuel cell operation, PEM experiences mechanical and chemical deterioration, which lowers electrochemical efficiency, exacerbates fuel crossover, and reduces PEMFC lifetime [5].

Commercialized Nafion® series are currently considered the benchmark PEMs as they offer superior proton conductivity, excellent chemical and mechanical stability. However, their applications in PEMFC have been limited because of the high cost, strong humidity dependence for proton conduction, and severe fuel crossover [6]. To address those shortcomings, many researchers are dedicated to the performance improvement or the development of new membranes in PEMFC. For example, sulfonated hydrocarbon polymers including sulfonated poly(arylene ether sulfone) (SPAES) and sulfonated poly(ether ether ketone) (SPEEK) are recognized as good PEM candidates due to the low cost and fuel permeation, good chemical stability, and excellent mechanical property [7,8]. However, PEMs with sulfonic acid groups linked directly to the polymer main chains usually exhibit undesirable hydrophilic and hydrophobic phase-separation morphology because of their rigid backbones, leading to the reduction of proton conductivity [9]. Increasing the sulfonation degree is an effective way to enhance conductivity, but it usually sacrifices the mechanical strength, dimensional and chemical stability. Consequently, sulfonated hydrocarbon membranes are vulnerable to mechanical damage such as cracks and perforations during MEA fabrication and PEMFC assembling; they are also prone to be degraded by HOO∙ and HO∙ radicals during PEMFC application, which decreases their service performance and lifetime [10,11].

Mechanical and particularly chemical stability of hydrocarbon membrane have demonstrated significant effect on PEM development as the fuel cell failure during long-term operation is closely affiliated with its chemical degradation. In PEMFC operation, the attack of highly reactive free radicals (HOO∙ and HO∙) are believed to be the main reason for chemical degradation of the membrane [5,12,13]. Two pathways commonly exist for free radical formation. First, H₂O₂ produced by incomplete electrochemical reaction decomposes into free radicals. Second, H₂O₂ or free radicals generated on Pt surface by the direct reaction of O₂ and H₂, which occurs upon H₂ or/and O₂ permeation [14].

Many approaches including blending, grafting, crosslinking, and inorganic-organic compositing have been employed to balance the stability and conductivity issues for PEMs. Incorporating surface-functionalized inorganic nanomaterials has been generally considered as an effective strategy to improve the chemical durability because they can catalyze the decomposition of H₂O₂ and free radicals during PEMFC operation [3]. The choice of inorganic fillers includes nano-spheres, nano-tubes, nano-sheets, and metal-organic frameworks [[15], [16], [17], [18]]. For example, Kim et al. [16] reported a series of functionalized carbon nanotubes (ACNTs)/SPEEK composite membranes. The resulting composite membranes exhibited enhanced proton conductivity and durable performance. Wu et al. [18] obtained a high peak power density and improved long-term stability for cross-linking branched polybenzimidazole by the incorporation of metal-organic framework (MOF) UiO-66. However, the size of these fillers is usually larger than 100 nm, which is 10–100 times larger than the separated nanophase or nano-scale polymer chain. Incorporation of these fillers will disrupt the long-range proton transfer channels, or even cut them into several blocks [19]. Gong et al. [15] introduced functionalized CNTs into the SPEEK polymer, which led to the embrittlement of membrane and decrease in elongation at break by 85%. Therefore, it is still a challenging issue to control the size and aggregation of nanomaterials to match the proton transfer channels.

Zero-dimensional (0D) nanoparticles, such as ZrO₂ [20], SiO₂ [21], TiO₂ [22], SnO₂ [23], polyhedral oligomeric silsesquioxane (POSS) [24] and quantum dots (QDs) [25], have smaller sizes, larger specific surface areas and more active sites. Their intrinsic structural characteristics make 0D nanoparticles easier to be modified, also facilitate better interfacial compatibility between polymer and inorganic fillers. Specifically, molecular-scale hybridization provides the hope to solve the issue of enhancing the stability and proton conductivity of PEMs concurrently [19,[25], [26], [27], [28]]. It can improve the compatibility of polymer and nanofillers for their uniform distribution in the membrane. Besides, fillers sizes of 2–10 nm match well with the diameter of proton transport channels, which will help in constructing continuous proton transfer channels. Wu et al. [27] reported increased proton conductivity and good dimensional stability when functionalized POSS (1–3 nm) was incorporated into SPAES membranes. Wang et al. [19] prepared Nafion-based composite membranes doped with 2–5 nm QDs. The composite membranes displayed good water affinity and improved proton conductivity. However, owing to the particle aggregation and chemical stability problems, there have been only a few studies so far on the chemical durability of composite membranes with molecular level fillers.

Carbon nano-onions (CNOs) are typical 0D nanomaterials and have attracted research interest recently. They have been investigated in many fields including electronics, catalysis, sensors, energy storage and conversion thanks to its unique structural properties, such as a large amount of pores and defects in the graphitic layers of the onion-like structure, ultra-small size, large specific surface, and good chemical stability [[29], [30], [31], [32]]. Surface-functionalized CNOs have good dispersivity and negligible agglomeration because of the high specific surface areas and intermolecular forces [30]. Moreover, functionalized CNOs have good chemical and mechanical stability, high conductivity, and adjustable surface structure, which make them applicable as electrode materials in supercapacitors and adsorbents in environmental remediation [33]. However, the researches on functionalized CNOs in membrane development are still limited, especially the influences on proton conductivity and durability of PEMs have not been reported.

In this work, CNOs prepared by nanometer diamonds (NDs) via thermal annealing and functionalized CNOs with sulfonated phenyl groups have been prepared. The obtained sulfonated carbon nano-onions (SP–CNOs) exhibit a uniform particle size (∼5 nm), high specific surface area as well as mesoporous structure, which profoundly increase the dispersivity and hydrophilicity. SPAES with a moderate IEC (∼1.50 mmol/g) was used to prepare the SP-CNOs composite membranes. High specific surface area and abundant hydrophilic –SO₃H groups of SP-CNOs possess excellent water-retention property. In the composite membrane, the absorbed water of SP-CNOs can form hydration layer to facilitate the diffusion of H₃O⁺. SP-CNOs can also provide excess proton hopping sites, which could construct low-energy-barrier proton transfer channels and enhance the interaction between polymers and inorganic particles. Furthermore, the good mechanical and chemical stability of SP-CNOs make it an appealing nanofiller to improve physicochemical performance and reduce the fuel crossover, which is helpful to the durability and long-term stability of the composite membrane. Proton conductivity, water retention, dimensional stability, and oxidative stability of the prepared composite membranes were investigated in detail. H₂/O₂ single fuel cell performance and chemical durability test were also conducted to assess the membrane feasibility in practical applications.