Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Jul 27;16(38):42941–42954. doi: 10.1039/d6ra03454c

Intrinsic electrocatalytic activity combined with oxygen preconcentration at non-metalated hypercrosslinked polymers for selective oxygen reduction reaction

Shumaila Razzaque a,✉, Katarzyna Wielondek a, Alla Dyachenko a, Kostyantyn Nikiforow a, Adam Kubas a, Krzysztof Noworyta a, Michal Bielejewski b, Bien Tan c, Frank Marken d, Marcin Opallo a,✉
PMCID: PMC13404368  PMID: 42516281

Abstract

The oxygen reduction reaction (ORR) remains a pivotal process in sustainable energy conversion and electrochemical synthesis. However, the development of efficient and cost-effective metal-free catalysts remains a challenge. In the last few years, it has become clear that polymers are a promising option. However, relatively easy-to-synthesize porous materials, such as non-metalated, non-pyrolyzed hyper-crosslinked polymers (HCPs), remain unexplored for the ORR. Here, we have focused on the ORR performance of HCPs synthesized by solvent knitting of triphenylamine (TPA) and triphenylbenzene (TPB) using different catalysts: FeCl3 (Fe-TPBA) and AlCl3 (Al-TPBA). Both polymers exhibit a hierarchical pore structure, with the latter showing a predominant contribution from micropores. HCP particles were immobilized in a Nafion® film deposited on the electrode surface, and their electrocatalytic activity towards the ORR was evaluated in both acidic and alkaline electrolytes. Both HCPs exhibit a catalytic effect and promote the 2-electron ORR. It was found that oxygen is preconcentrated in the catalyst film, contributing to the increase in its flux as compared to the bulk electrolyte, thereby increasing the ORR current. Three-phase junctions (electrode|nonconductive polymeric catalyst|ionomer filled with aqueous electrolyte) are suggested as the reaction site. DFT quantum chemical calculations indicate a weak interaction between HCPs' molecular motifs and O2, making it a suitable O2 flux promoter. Moreover, the calculations indicate that the presence of nitrogen-containing TPA units enhances the oxygen reduction rate as compared to TPB. This study paves the way for exploration of other non-metalated HCPs prepared from a wide array of precursors for ORR electrocatalysis.

Oxygen reduction electrocatalysis is demonstrated at an electrode modified with nonmetalated hypercrosslinked polymer particles. This modification enhances oxygen flux towards the electrode.graphic file with name d6ra03454c-ga.webp

1. Introduction

The oxygen reduction reaction (ORR) is one of the most extensively studied fundamental electrochemical processes. It plays a crucial role in energy conversion and storage systems, such as fuel cells or metal–air batteries.1–4 Recently, the electrochemical ORR has been considered as a sustainable alternative approach for H2O2 synthesis.5–9 One of the major challenges associated with ORR application is its inherently sluggish kinetics. This necessitates the development of highly active electrocatalysts to accelerate the reaction and improve the overall efficiency of energy conversion or H2O2 synthesis. Pt, Pd, and their alloys are widely used as catalysts, but their high cost and/or limited availability hamper commercial application. Conversely, most nonmetallic ORR catalysts, like porous carbons10–14 or polymers, ranging from highly conductive materials15–17 to insulating and highly porous structures,18–22 offer promising alternatives. Unfortunately, synthesis of porous carbon electrocatalysts requires high-temperature (700–1100 °C) pyrolysis and produces poorly defined catalytic sites. This poses a huge challenge to understanding their structure–property relationships, severely hampering the insightful understanding of the reaction mechanism. Notably, polymers' mild synthesis conditions and durable active sites favor them as ORR catalysts. Apart from conducting polymers, the majority of research in this area is focused on metalated or pyrolyzed polymers. Therefore, studying the ORR at non-metalated and non-pyrolyzed porous polymer catalysts is indispensable for a better understanding, and this work aims to address that gap.

Conducting polymers were first reported as ORR catalysts 20 years ago,23 and some progress has been made in studies of their electrocatalytic properties.16,17,24 More recently, porous materials, such as metal organic frameworks (MOFs),21,25,26 or POPs,27,28 particularly covalent organic frameworks (COFs),20,21,29,30 have attracted attention due to their large specific surface area, tunable pore size, tailorable frameworks, and high chemical stability in electrochemical environments.31 Despite these advantages and the synthetic tunability of porous frameworks, several limitations of MOFs and COFs hinder their practical application. For instance, their preparation relies on a limited range of precursors and often requires expensive or rare starting materials.32,33 Furthermore, their synthesis requires rigorous, time-consuming, and costly procedures. Additionally, scaling up the production of these frameworks remains challenging due to issues such as reproducibility, stability, and cost-effectiveness.34 Although porous frameworks offer a versatile platform for electrocatalyst development, their widespread application for electrochemical processes demands further advancements in structural robustness, structural tunability, stability, and scalable and cost-effective synthesis.18 Importantly, POPs' ability for gas storage and/or separation35–38 is a clear advantage for the ORR, because of the limited solubility of oxygen in water. Consequently, POPs may provide a more efficient O2 transport path towards the electrode surface, as recently demonstrated for a film of polymer of intrinsic microporosity (PIM).39,40 Last but not least, understanding structure–reactivity relationships22,41 and the mechanism of electrochemical reactions catalyzed by POPs creates a foundation for their application in electrocatalysis.

In this context, HCPs have emerged as a promising class of POPs offering a highly tunable framework with permanent microporosity and a variety of dense functional sites that can be tailored for ORR catalysis.42–46 Despite their widespread use in gas storage and separation,36,45,47–49 or heterogeneous catalysis,33,50 a few studies have focused exclusively on the application of the porous carbons derived from HCPs42,51–54 or metal-coordinated macrocyclic unit-based HCPs55,56 to achieve enhanced ORR activity. These strategies involve high-temperature pyrolysis or multi-step post-synthetic modification, which increases processing complexity and produces unstable catalysts. The comparison of electrocatalytic ORR efficiency at metalated and non-metalated COFs having the same structure indicates the contribution of the catalytic effect of the polymer itself.57–59 Therefore, we have focused on exploring non-metalated and non-pyrolysed HCPs as ORR catalysts. Our preliminary study demonstrated that immobilization of such polymer particles on a noncatalytic electrode decreases the ORR overpotential.60 Here, we aim to identify the effect of polymer structures generated by different synthetic conditions on ORR electrocatalysis, understand HCPs' intrinsic electrocatalytic properties, examine their ability to preconcentrate oxygen, and evaluate the ORR mechanism. This may bring an assessment of whether non-metalated and non-pyrolysed HCPs are promising ORR catalysts.

Several HCP features are beneficial for ORR electrocatalysis. High surface area and permanent microporosity may facilitate rapid mass transport of oxygen to active sites, which is favorable for faster ORR kinetics.40,61 HCPs' chemical tunability enables the incorporation of heteroatoms (e.g., N, S, P) or coordination of transition metals, allowing for the design of well-defined, accessible catalytically active sites.44,62–65 It was suggested that the electrocatalytic activity of POPs is strongly influenced by their heteroatom functionalities and pore size.66 Integration of heteroatoms in the porous structure can alter the local electronic structure and charge distribution. Interestingly, our preliminary study revealed that the presence of N heteroatoms in HCPs enhances the ORR, not affecting overpotential,60 which may indicate acceleration of O2 transport within the porous structure. Indeed, oxygen preconcentration within POPs pores was reported.67,68 Micro- and meso-pores may improve product selectivity and interactions between active sites and the substrate, whereas macropores may facilitate mass transport, shorten the diffusion path, and reduce the residence time.69 Importantly, HCPs' hierarchical pore distribution can be tuned by the selection of a synthetic path.70,71 Thus, investigation of the combination of well-defined porosity and structural flexibility of non-pyrolysed HCPs may uncover their inherent electrocatalytic potential and offer a more sustainable, low-cost pathway for the development of metal-free ORR catalysts.

Here, three-dimensional HCPs were prepared from the same substrates by two different synthetic routes to obtain different porosity, specific surface area and pore size distribution. Triphenylamine (TPA) was employed as a key monomer and co-polymerized with 1,3,5-triphenylbenzene (TPB) in the presence of the different Lewis acid catalysts. Scholl coupling with AlCl3 and Friedel–Crafts reaction with FeCl3 were expected to generate HCPs with distinct structural and catalytic properties:70–72 TPA increases the electron density of adjacent phenyl rings through the lone pair on nitrogen, which should enhance interactions with molecular oxygen, highly desirable for ORR catalysis.73,74 The specific surface area and porosity of the HCPs were modulated by the introduction of a rigid, non-planar, and sterically twisted TPB.60 Here, we will demonstrate the effect of the synthetic route on the pore distribution and morphology of the HCP particles embedded in the ionomer film deposited on the solid electrode surface. Furthermore, we will show that the type of porous structure generated by a certain polymer preparation route affects the ORR overpotential decrease. We will also demonstrate that HCP particles, immobilized on the electrode surface in the ionomer film, increase oxygen flux towards the electrode surface. We will prove that the ORR follows a two-electron path, and this mechanism does not depend much on the type of HCP. Complementary density functional theory (DFT) calculations will point out the role of specific polymer functional groups and structural motifs in facilitating O2 adsorption and reduction, providing atomic-level insights into the reaction energetics and kinetics.

2. Experimental

2.1. Materials

All chemicals and reagents were of analytical grade and were used without further purification. TPA, anhydrous FeCl3, anhydrous AlCl3, KCl, K3Fe(CN)6, ferrocenemethanol (FcOH) and anhydrous dichloromethane (DCM) were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Ethanol, methanol, and acetone were obtained from Linegal Chemicals (Warsaw, Poland). TPB was purchased from Tokyo Chemical Industry Co., Ltd (TCI, Japan). KOH and HClO4 were obtained from Merck KGaA (Darmstadt, Germany). Nafion® solution (5 wt%) was sourced from DuPont China Holding Co., Ltd (Shanghai, China). Vulcan XC-72 carbon nanoparticles (CNPs) were supplied by Aldrich. Deionized water with a resistivity of 18.2 MΩ cm, obtained from an Arium® Comfort Lab Water System (Sartorius), was used in all experiments.

2.2. Synthesis of TPBA

The synthetic route (Scheme 1) depends on the type of catalyst and follows either the Friedel–Crafts or Scholl coupling mechanism (see SI for details).

Scheme 1. Scheme of synthesis of Fe-TPBA and Al-TPBA with their proposed structures.

Scheme 1

Open in a new tab

2.3. Equipment for polymer characterization

The synthesized HCPs were characterized with a range of techniques. Fourier-transform infrared spectroscopy (FTIR) was conducted on a Vertex 80 V Fourier spectrometer (Bruker Inc., USA) to confirm the chemical structure. Solid-state UV-vis absorption spectra were recorded using a Shimadzu UV-2600 spectrophotometer to investigate the degree of HCP π-conjugation. Nitrogen adsorption–desorption isotherms were measured at 77 K using a Micromeritics ASAP 2460 analyzer to evaluate the surface area and porosity of the materials. X-ray photoelectron spectroscopy (XPS) was performed on an XPS/AES spectrometer (VG Microtech) to investigate the surface elemental composition and chemical states of the elements. Solid-state 13C NMR spectra were recorded on a Bruker Avance III spectrometer equipped with an 11.4 T superconducting magnet operating at 125.76 MHz (Larmor frequency of 13C) to identify the linkages (methylene) formed during polymerization. The morphology of the obtained HCPs was examined by an FEI Nova NanoSEM 450 microscope. The electrical conductivity of the studied polymers was obtained from electrochemical impedance spectroscopy (EIS) data performed in a four-electrode setup using a Zurich Instruments frequency analyzer.

2.4. Electrochemical equipment and procedures

Electrochemical studies of the synthesized HCPs were carried out to determine their electrocatalytic activity, ORR mechanism, O2 diffusion coefficient and concentration in the polymer film, and intrinsic redox activity. For this purpose, a noncatalytic glassy carbon (GC) or a “catalytic” Pt electrode was modified with Fe-TPBA or Al-TPBA powder embedded in a Nafion® film (ionomer), as recommended.75 HCP ink was prepared by dispersing HCP powder in an ionomer solution. Typically, 5 mg of the HCP powder was dispersed in 100 µL of ethanol and sonicated for 30 minutes to obtain a homogeneous dispersion. After addition of 30 µL of 5 wt% Nafion® solution, it was sonicated for an additional 30 minutes. Finally, 5 µL of HCP dispersion was dropped on the electrode surface and evaporated for 30 minutes under ambient conditions. Similarly, polymer ink with CNPs was prepared by the same procedure. In some experiments, before each modification, the electrode was polished on a micro-cloth with the alumina suspensions (Buehler) and rinsed.

For cyclic voltammetry (CV), linear sweep voltammetry (LSV) and chronoamperometry (CA) in quiescent conditions, PSTrace 5 (PalmSens) electrochemical workstations with 3 mm diameter modified or bare GC or Pt disk (Mineral, Poland) as the working electrode were employed. For hydrodynamic voltammetry, a Metrohm-Autolab potentiostat with an Autolab rotating GC disk electrode (RDE) or a Pt ring and GC disk electrode (RRDE) (both from Metrohm, Switzerland) was used. The collection coefficient of the Pt ring was estimated as 0.2 (see SI). All electrochemical measurements were carried out in a three-electrode electrochemical cell filled with 0.1 M HClO4 or 0.1 M KOH aqueous solution using Pt wire and Ag/AgCl (3 M KCl) or Hg/HgO (6 M KOH) as the counter and reference electrodes, respectively. All measured potentials were converted to the RHE scale.

2.5. Quantum chemical calculations

All calculations were performed using the ORCA 6.0.1 suite of programs76 for structural motifs (Scheme S1) related to proposed structures. In principle, we followed the protocol to describe oxygen reactivity proposed by some of us.77 Briefly, the BP86 functional,78,79 supplemented with the D3 dispersion correction,80 was used for geometry optimization. Solvent (water) was represented using a Conductor-like Polarizable Continuum Model (CPCM)81 with a dielectric constant of 80.1. For the ORR, two explicit water molecules were added to the model to more accurately describe the hydrogen-bond network. All structures obtained were confirmed to be energy minima by calculating Hessian matrices and inspecting their eigenvalues; no imaginary modes were found for the structures reported. Single-point calculations were performed using the B3LYP functional,82,83 which was augmented with the D3 correction. All calculations utilized the def2-TZVP basis set.84–86 Final energies reported in the manuscript were obtained at the B3LYP+D3/def2-TZVP/CPCM (water) level and are inclusive of zero-point energies obtained with the BP86+D3/def2-TZVP/CPCM (water) setup used for structure optimizations. The same structures were used for vertical electronic spectra simulations using the time-dependent DFT approach (TD-DFT). Here, the range-separated CAM-B3LYP functional87 was employed, along with the def2-TZVP basis set. 20, 40, and 60 excited states were considered for monomers, dimers, and trimers, respectively (see Scheme S1), and spectral lines were plotted using a broadening parameter of 1500 cm−1.

Half-reactions were referenced against the theoretical standard hydrogen electrode. Its absolute potential of 4.45 V was obtained at the same level of theory as used for systems under study (vide supra) by considering the ½H2 = Haq+ + e− half-reaction. The proton solvation energy was taken from the literature (−11.38 eV = −262.4 kcal mol−1).88

3. Results and discussion

3.1. Physicochemical characterization

The formation of HCPs synthesized using FeCl3 (Fe-TPBA) or AlCl3 (Al-TPBA) was confirmed by FTIR spectra (Fig. 1a). The core structure of the HCPs appears similar due to the similar crosslinking mechanism, except for the broader stretching C Created by potrace 1.16, written by Peter Selinger 2001-2019 C bands in the aromatic region of the Al-TPBA spectrum, which may be an indication of extended aromatic structure formation (Scheme 1). Absorption in the range of 2800–3100 cm−1 attributed to the –CH stretching vibration is associated with the presence of benzene rings and methylene (–CH2–) bridges. They come from DCM, which acts not only as the reaction medium but also as an external crosslinking agent, participating directly in the formation of the HCP network. Strong bands between 1400 and 1600 cm−1, corresponding to the aromatic C Created by potrace 1.16, written by Peter Selinger 2001-2019 C stretching vibrations, are also present in the monomer spectra (Fig. S1), indicating that bonds connecting aromatic rings in monomers are retained after polymerization. Moreover, the copolymerization is further supported by the C–N bending vibration band observed at around 1311 cm−1.89

Fig. 1. (a) FTIR spectra, (b) XPS survey spectra, (c) UV-vis spectra, (d and e) SEM images and (f) N2 adsorption–desorption isotherms of Fe-TPBA and Al-TPBA.

Fig. 1

Open in a new tab

Next, HCPs were characterized by XPS to gain information on the surface elemental composition and chemical states of functional groups that may affect the electrocatalytic process. The survey spectra of both polymers (Fig. 1b) reveal the presence of main constituent elements C, N, and O, and allow us to calculate their atomic content (%; Table S1). Upon fitting the C 1s XPS spectrum, the distinct chemical environment of the carbon in the HCP backbone is identified (Fig. S2a and c). The dominant peak at 284.5 eV is attributed to C Created by potrace 1.16, written by Peter Selinger 2001-2019 C and C–H bonds,90 whereas the peaks at higher binding energy (285.5 eV) correspond to C–N bonds, signifying the successful copolymerization of TPA with TPB. The N 1s high-resolution XPS spectrum shows a single peak at 399.8 eV and 399.6 eV (Fig. S2b and d), ascribed to the sp2-hybridized N, again confirming the copolymerization. The oxygen detected in the HCP samples is most likely adsorbed from the air or from oxidation of unreacted chloromethyl bridges during washing (HCl aq., H2O).91,92 More importantly, the XPS analysis revealed no significant Fe content retained in the HCPs, which could otherwise interfere with the electrocatalytic activity. The 13C cross-polarization/magic angle spinning nuclear magnetic resonance (CP/MAS NMR) spectra (Fig. S3) of the polymers show broad resonance peaks at around 110 and 140 ppm due to the substituted and unsubstituted aromatic carbons, respectively, while the resonance peaks below 50 ppm can be ascribed to the carbon of methylene linkers, stemming from DCM working as solvent and cross-linker. The presence of an aliphatic signal in both polymers is a clear indication of Friedel–Crafts hypercrosslinking.

The solid-state UV-vis absorption spectrum (Fig. 1c) depends on the type of catalyst used for synthesis, indicating different optical properties for Fe-TPBA and Al-TPBA. The Al-TPBA sample spectrum exhibits a noticeable red shift, indicating a lower optical bandgap and more extended π-conjugation due to efficient intra- and inter-molecular Scholl coupling, resulting in a more tightly packed and conjugated polymer.72,93 In contrast, Fe-TPBA shows a blue-shifted absorption maximum (370–380 nm) and a more steep absorbance decrease beyond 650 nm, indicating a higher bandgap and less conjugated structure (Scheme 1).

SEM images of the HCP particles reveal a noticeable difference in the surface morphology. The aggregated, irregular particulate structure of Fe-TPBA (Fig. 1d) is typical of HCPs formed via extensive crosslinking reaction.94 Conversely, Al-TPBA shows a smooth sheet-like structure (Fig. 1e), attributed to the more uniform crosslinking of the monomers.

The N2 adsorption–desorption isotherms (Fig. 1f) illustrate the distinct porosity features of the HCPs prepared via two different Lewis acid catalysts. The Fe-TPBA isotherm reveals the dominance of meso- and macro-pores with a minimal contribution of micropores.95 Furthermore, the hysteresis loop at medium pressure provides additional evidence for the dominance of mesoporous structures. On the contrary, the Al-TPBA isotherm demonstrates significant nitrogen uptake at low relative pressure (P/P0 ≤ 0.1), indicative of a dominant microporous structure. Furthermore, the presence of a hysteresis loop suggests the existence of mesopores within the material. However, only a limited contribution from macropores is evidenced by the absence of a steep increase in adsorption at high relative pressure (P/P0 ≤ 1). Importantly, Al-TPBA exhibits a significantly larger BET surface area than Fe-TPBA (1137.25 m2 g−1), as well as a larger total pore volume (0.642 cm3 g−1) (Table S2). Application of the nonlocal density functional theory (NLDFT) provides information about the hierarchical pore distribution in both polymers (Fig. S4). Importantly, the predominant contribution of micropores in Al-TPBA can be attributed to the higher Lewis acidity and catalytic reactivity of AlCl3 compared with FeCl3, which facilitate rapid electrophilic activation of aromatic monomers, resulting in higher crosslinking density.72 For the example of porous carbon derived from HCP, such hierarchical porosity was suggested to be favorable for electrocatalysis limited by mass transport.96

The specific conductivity of both dry polymers is low (<0.3 µS m−1; Table S3). Importantly, Al-TPBA's conductivity is two orders of magnitude higher than that of Fe-TPBA. This may result from more extensive delocalization of π-electrons arising from the higher degree of conjugation in Al-TPBA.

3.2. Electrochemical characterization

CV curves obtained with GC|Fe-TPBA (Fig. 2a) and GC|Al-TPBA (Fig. 2b) electrodes in air-saturated acidic solution exhibit features corresponding to at least four electrode processes. Importantly, the current density of all signals is smaller for the GC|Al-TPBA electrode (Fig. 2b; please note the different scale of the y-axis as compared to Fig. 2a). A sharp increase of the cathodic current at potentials below −0.4 V and a cathodic wave with onset potential ca. 0.0 V are assigned to the hydrogen evolution reaction (HER) and ORR, respectively. Importantly, the cathodic wave is ca. two times smaller for the GC|Al-TPBA electrode, and it is not seen in the absence of oxygen at both electrodes (Fig. S5).

Fig. 2. (a) Cyclic voltammograms obtained at the (a) GC|Fe-TPBA and (b) GC|Al-TPBA electrodes in 0.1 M HClO4 with a scan rate of 0.02 V s−1. Linear sweep voltammograms obtained with GC, GC|Fe-TPBA and GC|Al-TPBA electrodes in (c) 0.1 M HClO4 and (d) 0.1 M KOH at a scan rate of 0.02 V s−1. Chronoamperograms obtained with Pt|Nafion and Pt|Fe-TPBA electrodes (modified with one, two or three drops of polymer ink) plotted on (e) linear and (f) semilogarithmic scale.

Fig. 2

Open in a new tab

The anodic and cathodic peaks with similar peak current and mid-peak potential of 0.71 V are a signature of the intrinsic polymer redox process arising from the reversible electrochemical oxidation of the TPA unit to the corresponding stable radical cation:97,98

TPA − e− ↔ TPA+˙ 1

However, the observed pH dependence of the mid-peak potential (Emidpeak) (Fig. S6) indicates that protons are involved in this reaction, despite the extremely low basicity of TPA. This may indicate that the electron-donating properties of the TPB unit and the methylene bridges increase the basicity of TPA. The slope of the linear dependence of Emidpeakvs. pH equals 0.042 V, indicating a 2 : 3 H+ : e− ratio in the elementary reaction step. This may suggest the distribution of the TPA units' basicity due to different local environments. Such a feature of an intrinsic redox reaction is not seen in alkaline solution, because of the instability of the triphenylamine cation in this medium. Clearly, reaction (1) does not influence the ORR, because at potentials below 0.0 V TPA units are in a reduced state and the redox catalysis cannot occur. However, the voltammetric signal related to reaction (1) provides information useful for understanding the ORR at the HCP-modified electrode, because the carbon atom next to the N atom66 of the TPA unit can be identified as a catalytic site.65 Importantly, the peak currents corresponding to the oxidation–reduction of the TPA unit (Fig. 2a, b and S7) are ca. one order of magnitude larger for the GC|Fe-TPBA electrode as compared to GC|Al-TPBA. This may indicate that a much larger number of TPA units of Fe-TPBA, presumably catalytic sites, are in contact with the GC surface. At the positive side of the potential window (Fig. 2a and b), an anodic current increase is seen at 1.2 V, which can be attributed to the oxygen evolution reaction (OER).

The cathodic peak or wave on the LSV curves obtained under static conditions in O2-saturated acidic or alkaline solution with the GC|Fe-TPBA or GC|Al-TPBA electrodes is shifted towards higher potential as compared with bare GC (Fig. 2c and d). This feature is not seen in Ar-saturated electrolyte solution (Fig. S5 and S8). Both observations confirm ORR electrocatalysis at the electrodes modified with synthesized HCPs. The wave shape of the ORR current in acidic solution may result from the overlap of this signal with the HER current increase at slightly lower potentials (Fig. S5). This is more pronounced for the GC|Al-TPBA electrode, indicating the strong electrocatalytic effect of this polymer towards the HER.

Although the use of the onset potential as a measure of electrocatalytic activity has been criticized,99,100 it remains a reasonable option for comparing structurally similar polymeric catalysts. In acidic solution, the difference between onset potential values (defined as the potential where the LSV current density reaches −0.01 mA cm−2) at the bare and modified GC electrode is equal to 0.06 V and 0.16 V for GC|Al-TPBA and GC|Fe-TPBA, respectively (Fig. 2c and d). Clearly, the overpotential decrease depends on the synthetic route and thereby the polymer structure, and this effect is smaller in alkaline solution (0.08 V vs. 0.13 V for GC|Al-TPBA and GC|Fe-TPBA, respectively). Although the Fe-TPBA surface area determined from N2 adsorption measurements is almost three times smaller than that of Al-TPBA, it seems that a larger fraction of mesopores promotes ORR catalysis. The larger shift of the onset potential may indicate that a larger number of catalytic sites of Fe-TPBA are in direct contact with the electrode surface.99 Indeed, such a conclusion can be drawn from the magnitude of the CV signal related to the intrinsic redox reaction of the polymer (Fig. S7). The larger ORR current seen for the GC|Fe-TPBA electrode may result from faster transport of O2 molecules through large Fe-TPBA pores than O2 molecules squeezing through Al-TPBA micro- and meso-pores.

To understand the role of oxygen transport towards an electrode substrate modified with a porous polymer particulate film, one must consider its structure. Here, particles of the porous insulator (HCP) with catalytic sites dispersed in the ionomer – Nafion® film form O2 percolation paths (Scheme 2). Apart from that, O2 diffusion in aqueous electrolyte solution filled with ionomer represents another route for O2 transport to the electrode (Scheme 2). Importantly, HCP particles block the GC surface as demonstrated by the CV experiment in the redox probe (FcOH) solution. The electrode modification with Fe-TPBA or Al-TPBA particles decreases CV peak currents by 4 and 8 times, respectively (Fig. S9). The stronger blocking effect of Al-TPBA can be attributed to the hydrophobic nature of the polymer, which causes its poor wettability by the aqueous electrolyte.101

Scheme 2. Scheme showing a cross-section of TPBA particle-modified electrode. The Nafion|electrolyte interface is not marked.

Scheme 2

Open in a new tab

Contrary to commonly employed conductive ORR catalysts like Pt/C nanoparticles embedded in ionomer, only HCP catalytic sites next to the GC surface may participate in the electrode reaction, because of a lack of electronic communication. The catalytic sites also have to be in contact with the electrolyte to allow for access of ions to the reaction site. Taking into account that the electrolyte does not penetrate the hydrophobic TPBA particles, only the catalytic sites located at the GC|TPBA|Nafion three-phase junctions may participate in the electrocatalytic reaction (Scheme 2), and the onset potential shift should depend on their number.99,102 Some penetration of HCP by aqueous solution, noted earlier for electrodes modified with particles of polymer of intrinsic porosity,103,104 may increase the number of effective catalytic sites. A similar effect may be envisaged if some conductive polymer domains are formed during the polymerization, and they are in direct contact with the electrode surface. Significantly, the CV peak current resulting from the intrinsic redox process (reaction (1)) is ten times larger for the GC|Fe-TPBA electrode as compared to the GC|Al-TPBA electrode (Fig. S7). The electric charge passed during reaction (1) calculated from these voltammograms corresponds to ca. 50 and 5 layers of TPA units of Fe-TPBA and Al-TPBA, respectively. This indicates that redox-conductive zones may expand a few nanometers from the electrode, and this effect is more pronounced for Fe-TPBA. The observed different peak current vs. scan rate dependences point out that the Fe-TPBA redox reaction is limited by diffusion of counterions (Fig. S10), whereas an entirely surface redox process is seen for Al-TPBA (Fig. S11). This indicates that counterion motion keeping local charge balance during the redox reaction (1) is severely restricted for the Al-TPBA electrode, which may result from its ultra-microporosity. Moreover, electrode modification with HCP increases the capacitive current measured in a potential range where faradaic processes are absent by approximately five times (Fig. S12a). Such an effect, which can be related to the double-layer capacity increase (Cdl) (Fig. S12b), is similar for both HCPs and may indicate the presence of conductive HCP domains, perhaps due to local electron delocalization.

Importantly, Nafion® film consists of macaroni-like bundles or lamellae separated by aqueous electrolyte,105–107 and it does not impede O2 diffusion.108,109 In turn, HCPs are known to be excellent materials for the storage of gases like O2, N2, or CO2 (ref. 45 and 110) and even for the separation of O2 from N2.111,112 Therefore, the flux of gaseous O2 across the TPBA particle percolation network (Scheme 2) may be significantly larger than the flux of dissolved O2 through the Nafion® electrolyte, due to limited concentration of O2 in the latter.

To confirm this hypothesis, chronoamperometric (CA) experiments were performed with Pt|Fe-TPBA and Pt|Al-TPBA (Fig. 2e and S13a) electrodes, as the current is expected to be proportional to the O2 transport rate and its concentration gradient. Here, the noncatalytic GC was replaced by “catalytic” Pt to eliminate the polymer contribution to ORR electrocatalysis and keep the ORR controlled by the O2 transport process.40 An alkaline electrolyte was selected for these experiments to avoid overlap of the ORR with the HER (Fig. 2c and S5b). The current density (J) recorded as a function of time (t) is entirely dependent on the oxygen diffusion coefficient (D) and concentration (c):

J(t) = nFD1/2cπ−1/2t−1/2 2

where n is the number of electrons, and F is the Faraday constant.113 As expected, J decreases with the passage of time (Fig. 2e and S13a). Additionally, the J value is roughly proportional to TPBA particle loading into the Nafion film, which was earlier seen for polymers of intrinsic porosity.40 However, here some saturation effect can be seen. Importantly, the J value at longer times is 2–3 times higher than that on a bare Pt electrode or a Pt electrode covered by Nafion film. One may conclude that the larger current results from higher O2 flux (proportional to D1/2 × c product) across polymer particles. Perhaps larger pores of Fe-TPBA promote larger current at the Pt|Fe-TPBA electrode. Therefore, one may conclude that the O2 for the ORR is mainly supplied by a network of TPBA particles, and the reported use of porous organic polymers as oxygen carriers67,68,114–116 supports this conclusion.

Further analysis based on the procedure earlier proposed40 aimed at the estimation of D and c in HCP@Nafion polymer film. A clear drop in J is seen at 1–10 s in the plot of J vs. log t dependence (Fig. 2f and S13b). This indicates that O2 flux is much higher next to the electrode surface than in the bulk electrolyte. SEM images (Fig. S14) of the cross-section of the modified electrodes show that the polymer deposit on Pt|Fe-TPBA is thinner (ca. 30 µm) than that on Pt|Al-TPBA (ca. 100 µm). This difference is reflected in the J vs. log t plots (compare Fig. 2f and S13b), where the current decreases at longer times. We used these values to calculate both the O2 diffusion coefficient and concentration (Fig. 2f, S13 and Table 1). The product D1/2 × c is proportional to O2 flux and is approximately four times larger for both polymer films as compared to water. This explains the larger ORR current at the HCP-modified Pt electrodes than on bare or Nafion® modified Pt.

Table 1. The parameters obtained from chronoamperometric experiments performed at a Pt electrode modified with HCP particles suspended in a Nafion film immersed in O2-saturated 0.1 M KOH aqueous solution.

Medium Transition time/s Film thickness/µm O2 diffusion coefficient/m2 s−1 O2 concentration/mM D 1/2 × c/mM s−1/2
Fe-TPBA 7 100 7.1 × 10−10 8 2.1 × 10−4
Al-TPBA 12 30 3.8 × 10−11 44 2.7 × 10−4
Aqueous electrolyte 2.1 × 10−9 1.2 5.5 × 10−5
Open in a new tab

We also carried out 24 hour stability test CA experiments with GC|Fe-TPBA and GC|Al-TPBA electrodes in alkaline electrolyte (Fig. S15). Especially, the latter electrode exhibits a stable j vs. t response. More importantly, current density is ca. two times higher for the GC|Al-TPBA electrode, which may result from the higher O2 concentration in the Al-TPBA@Nafion film and therefore a larger concentration gradient.

Next, we performed experiments with a RRDE to determine the ORR mechanism. Clearly, a larger fraction of H2O2 is produced at the GC|Fe-TPBA electrode in alkaline solution as compared to acidic solution. The number of electrons calculated from RRDE data (see SI) is 2.0–2.5 and 2.0–3.0 (depending on the electrode potentials) for acidic and alkaline electrolytes, respectively (Fig. 3a and b). Analysis based on the Koutecky–Levich equation yields a lower number of electrons. However, one must keep in mind that this equation was derived for conductive and flat electrodes, which is not the case here. The same number of electrons (2–3) was also obtained from Koutecky–Levich analysis of the RRDE experiments for the GC|Al-TPBA electrode in alkaline solution (Fig. 3c). The data obtained with the same electrode in acidic solution could not be analysed because of the strong overlap of the ORR and HER currents. The obtained results demonstrate significant conversion of oxygen to hydrogen peroxide.

Fig. 3. Results of RRDE experiments (GC|HCP disk and Pt ring). LSV curves obtained with a GC|Fe-TPBA electrode in (a) 0.1 M HClO4 and (b) 0.1 M KOH solution and (c) with a GC|Al-TPBA electrode in 0.1 M KOH solution with different rotation rates. Scan rate: 0.002 V s−1. The number of electrons (n) as a function of electrode potential obtained with GC|Fe-TPBA in (d) 0.1 M HClO4 and (e) 0.1 M KOH and with GC|Al-TPBA in (f) 0.1 M KOH. These values were obtained from LSV at different rotation rates and from the Koutecky–Levich plot (see SI for details).

Fig. 3

Open in a new tab

The electrocatalytic activity was further investigated in a dynamic setup using the rotating disk electrode (RDE) at a rotation of 1600 rpm (Fig. 4a). The onset potential difference between dynamic and quiescent conditions is not larger than 0.01 V. Obviously, the increase in the cathodic current density is due to enhanced mass transport. As almost all studies on ORR electrocatalysis at POPs are done with electrodes modified with polymer particles and carbon nanomaterials encapsulated in Nafion® film, here we employed Vulcan XC-72 carbon CNPs. From the comparison of RDE voltammograms (Fig. 4b), one can see that the presence of CNPs in the film increases the plateau current density (approximately 3–4 times) and further decreases the overpotential by 0.08–0.09 V (Fig. 4b). These effects result from the formation of CNP percolation paths, which increase the number of catalytic sites having electronic communication with the electrode surface. As the current increase is rather obvious, the onset potential was recently postulated for an electrocatalysis electrode modified with conductive nanoparticles.99

Fig. 4. (a) Dynamic LSV obtained with GC, GC|Fe-TPBA and GC|Al-TPBA electrodes, rotating at 1600 rpm in 0.1 M KOH aq. (b) Comparison of the LSV curves obtained with GC|Fe-TPBA (red) and GC|Al-TPBA (blue) RDE without (solid line) and with (dashed line) Vulcan XC-72 added to the film. Scan rate 0.02 V s−1.

Fig. 4

Open in a new tab

3.3. DFT calculations

To gain further insights into the structure and activity of the Fe-TPBA and Al-TPBA HCPs, we performed a series of DFT calculations. A series of simple to more extended structures (Scheme S1) was selected to represent the structural motifs present in the studied HCPs. These cover the monomers used (TPA and TPB), two heterodimeric structures connected either via a CH2 bridge (dimer 1) or linked directly (dimer 2), and five trimers with different TPA : TPB ratios and connectivity (motifs 1–5). These structures were subject to TD-DFT electronic spectra calculations (Fig. S16). Notably, the spectra of the TPB and TPA monomers have very distinct absorption maxima of ca. 230 nm and 290 nm, respectively. These bands are rather preserved in the case of dimer 1, which features a CH2 bridge between units and matches very well with the broad 220–290 nm band present in the Al-TPBA spectrum (Fig. 1c). This indicates the dominance of methylene-bridged motifs in the final material. Interestingly, the maxima of the UV-vis spectra shift to lower energies with a higher number of direct phenyl–phenyl bonds. This is generally consistent with the difference between the proposed HCP structures: Fe-TPBA with methylene-bridge-dominated motifs and Al-TPBA with some directly connected phenyl groups.

In the next step, we computed the molecular oxygen interaction energies with selected monomers and dimers (Scheme S1). According to the interaction energies (Table 2), they are rather low (less than 2 kcal mol−1). A significant difference is observed only between TPA and TPB monomers, with oxygen preferring interaction with TPB by ca. 0.8 kcal mol−1. This is mainly due to dispersion interactions being stronger in the case of TPB (larger molecule, more extended π-system). Interestingly, the O–O bond is longer for the TPA monomer, even if the interaction energy is smaller. This indicates that in this case, the electronic factor is dominant, with some charge-transfer to the O2 moiety (total Mulliken charge of −0.02). For the two dimers tested, the oxygen molecule prefers to interact with the TPB subunit, consistent with interaction energies derived for monomers. The distance to the central phenyl ring of the TPB unit is about 3 Å, and the O–O bond does not seem to be significantly elongated with respect to free O2 (1.217 to 1.219 Å vs. 1.216 Å for free oxygen). All these facts point us to the conclusion that the examined HCPs do not bind oxygen strongly, which may be advantageous from the point of view of the transport of oxygen to the electrode.

Table 2. Computed interaction energies (dE) of molecular oxygen with monomers and dimers shown in Scheme S1 along with the O–O bond distance (rO–O) in the complex. The reference value of rO–O for a free O2 molecule obtained at the same level of theory is 1.216 Å.

Monomer/dimer dE (kcal mol−1) r O–O [Å]
TPA −0.8 1.222
TPB −1.6 1.219
Dimer 1 −1.5 1.217
Dimer 2 −1.6 1.219
Open in a new tab

Almost all coupled electron/proton transfer (ET/PT) reaction steps for O2 complexes with TPB and TPA are downhill in energy (Table 3). One distinct difference is the mildly endothermic (+0.15 eV) first ET/PT step for O2 at TPB, which reflects the relatively stronger interaction of molecular oxygen with TPB as compared to TPA, as well as the very weak interaction of the OOH radical with both monomers. The energy changes during the two- and four-electron ORR for O2 at both monomers were found to be similar, with values of ca. −1.4 eV and −5.2 eV, respectively. This parameter includes contributions from the interactions of H2O2 and H2O with the TPB or TPA monomers. Interaction energies of H2O2 with both monomers seem to be virtually identical, so once this state is formed, only the morphology of the material and the interface with the electrode will dictate the probability of the next ET/PT steps. However, if a TPB-rich polymer is present in the vicinity of the electrode, then the first ET/PT step (one-electron oxygen reduction) shall be slower as compared to TPA sites. This is consistent with experimental findings showing that Fe-TPBA provides higher ORR currents and contains significant TPA unit concentration near the operating electrode surface (Fig. 2a).

Table 3. Theoretical redox potentials of TPB and TPA monomers as well as step-by-step energy changes for the 4e−/4H+ molecular oxygen reduction reaction for oxygen at the respective monomer.

Property/process TPB TPA
Redox potentials [V]
E ox (monomer) 1.42 0.51
E red (monomer) 2.61 3.13
 
Energy changes [eV]
O2 + 1e− + 1H+ → ˙OOH 0.15 −0.06
˙OOH + 1e− + 1H+ → H2O2 −1.51 −1.35
H2O2 + 1e− + 1H+ → H2O + ˙OH −1.82 −1.58
˙OH + 1e− + 1H+ → H2O −1.98 −2.19
Open in a new tab

4. Conclusions

Here, we have demonstrated the potential of non-metalated and non-pyrolyzed HCPs synthesized by solvent knitting for ORR electrocatalysis both in alkaline and acidic conditions. Their use in alkaline electrolytes is more advantageous because of the good separation of ORR and HER currents. Nevertheless, the studied HCPs appear to be promising non-metalated HER catalysts in acidic electrolytes, and this will be the focus of our future work. The onset potentials are within the range of reported data for extensively studied electrodes modified with other porous polymer particles, such as COFs embedded in Nafion film (Table S4).117 However, the electrocatalytic effect is not impressive when compared with porous polymers deposited on carbon nanomaterials or pyrolyzed HCPs.42,52,96 The present study should help to understand the intrinsic electrocatalytic activity of HCPs in the absence of conductive nanoparticulate additives or high-temperature treatment.

Although the heteroatom density of the studied amorphous HCPs is much smaller than that of crystalline COFs, their straightforward and scalable one-step synthesis from commercial monomers is noteworthy. Importantly, the pore size distribution and surface area depend on the catalyst employed in the polymerization process. The electrode modified with HCP particles with bigger (micro- and meso-) pores exhibits larger ORR voltammetric current. The differences between the conductivities of the two polymers reflect the differences in electron delocalization, supporting structural differences arising from the polymerization pathways (Scheme 1). Moreover, a negligible difference in onset potential is observed between the GC|Fe-TPBA and GC|Al-TPBA electrodes. Importantly, the two-electron ORR is predominant at both polymers, which makes these HCP-modified electrodes a prospective tool for sustainable H2O2 generation. This mechanism may result from the weak interactions of the reactants with the studied HCPs, as demonstrated by the results of DFT calculations, and does not indicate a preference to accept another two electrons.

Another important result is the existence of an O2 gradient in the HCP particle film encapsulated in Nafion®. This is due to O2 absorption in HCP particles, which overcomes its slow diffusion, producing a larger oxygen flux than in the bulk electrolyte. It can be inferred that these O2 molecules diffuse along percolation paths formed by insulating HCP particles embedded in the ionomer film to the three-phase junction (electrode|polymer particle|electrolyte in ionomer; Scheme 2), where they are reduced. This aspect of the ORR reaction was not previously reported for particulate porous polymer films embedded in an ionomer and may be generalisable for other porous polymer particles. The recently advocated concept of “triphasic gas storage” seems to fit the TPBA particulate film embedded in Nafion®. Such an effect was similar to intrinsically microporous polymer particulate film prepared without the help of ionomer40,118 studied earlier, as they are typically immobilized in Nafion®, as in our study. The increase in O2 flux may contribute to the onset potential shift and to the difference between the onset potentials of the electrodes modified with different HCPs. The separation of intrinsic catalytic effects from substrate concentration effects requires modeling, which is challenging for such a complex film geometry containing two O2 diffusion pathways: through hydrophobic pores between touching HCP particles and through electrolyte domains in Nafion®. Anyway, HCPs' intrinsic catalytic activity has to be taken into account in the analysis of ORR catalysis at metalated HCPs.55 However, for practical application of such insulating polymer particles, the use of conductive nanomaterials as a film component is highly recommended. This is demonstrated here by the effect of the addition of CNPs to the ink employed for electrode modification. This increases the current density and apparently decreases the ORR overpotential. The latter observation shows that this recently suggested effect may be observed for more complex systems than electrodes modified with catalytic conductive nanoparticles.99

Building upon this foundation, future efforts should focus on the rational design of HCPs incorporating different combinations, configurations, and densities of heteroatoms to systematically evaluate their electronic and structural effects on the ORR activity and selectivity. Overall, this study opens a new direction for tuning the catalytic behavior of HCPs through heteroatom engineering, advancing the development of sustainable, metal-free electrocatalysts for energy conversion technologies.

Author contributions

S. R.: conceptualization, methodology, investigation, validation, formal analysis, writing – original draft, writing – review and editing. K. W.: investigation, formal analysis. A. D.: investigation, formal analysis. K. N.: investigation, formal analysis A. K. investigation, formal analysis, writing – original draft, writing – review and editing. M. B.: investigation. B. T.: investigation, writing – review and editing. F. M.: writing – review and editing. M. O. conceptualization, methodology, validation, writing – original draft, writing – review and editing, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Supplementary Material

RA-016-D6RA03454C-s001

Acknowledgments

This work was financially supported by the National Science Centre (NCN, Poland) through OPUS 27 research grant No. 2024/53/B/ST4/01997 “Knitting polymers for oxygen electrocatalysis”.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: detailed description of synthesis, description of analysis of RDE, RRDE and CA data, IR spectra, XPS spectra, 13C CP/MAS NMR spectra, pore size distribution, CV and CA curves, SEM images, images of structural motifs for DFT calculations, simulated electronic excitation spectra, elemental composition data, summary of porosity measurements, reported values of onset potentials, and XYZ coordinates of structural motifs for DFT calculations. See DOI: https://doi.org/10.1039/d6ra03454c.

References

  1. Liang Z. Lei H. Zheng H. Wang H.-Y. Zhang W. Cao R. Chem. Soc. Rev. 2025;54:5248–5291. doi: 10.1039/D4CS01199F. [DOI] [PubMed] [Google Scholar]
  2. Cui X. Gao L. Ma R. Wei Z. Lu C.-H. Li Z. Yang Y. J. Mater. Chem. A. 2021;9(37):20985–21004. doi: 10.1039/D1TA02795F. [DOI] [Google Scholar]
  3. Zhao Y. Adiyeri Saseendran D. P. Huang C. Triana C. A. Marks W. R. Chen H. Zhao H. Patzke G. R. Chem. Rev. 2023;123(9):6257–6358. doi: 10.1021/acs.chemrev.2c00515. [DOI] [PubMed] [Google Scholar]
  4. Yang H. Han X. Douka A. I. Huang L. Gong L. Xia C. Park H. S. Xia B. Y. Adv. Funct. Mater. 2021;31(12):2007602. doi: 10.1002/adfm.202007602. [DOI] [Google Scholar]
  5. Wu Z. Ding P. Gueskine V. Boyd R. Głowacki E. D. Odén M. Crispin X. Berggren M. Björk E. M. Vagin M. Energy Environ. Mater. 2024;7(2):e12551. doi: 10.1002/eem2.12551. [DOI] [Google Scholar]
  6. Yuan K. Li H. Gu X. Zheng Y. Wu X. Zhao Y. Zhou J. Cui S. ChemSusChem. 2025;18(6):e202401952. doi: 10.1002/cssc.202401952. [DOI] [PubMed] [Google Scholar]
  7. Wen Y. Zhang T. Wang J. Pan Z. Wang T. Yamashita H. Qian X. Zhao Y. Angew. Chem. 2022;134(35):e202205972. doi: 10.1002/ange.202205972. [DOI] [PubMed] [Google Scholar]
  8. Zhao L. Yan R. Mao B. Paul R. Duan W. Dai L. Hu C. Small. 2024;20(43):2403029. doi: 10.1002/smll.202403029. [DOI] [PubMed] [Google Scholar]
  9. Chen H. Chen R. Liu S. Zhou Y. Chen X. Cai J. Lan X. Jiang H. Lin L. Sun Z. ChemPlusChem. 2024;89(11):e202400422. doi: 10.1002/cplu.202400422. [DOI] [PubMed] [Google Scholar]
  10. Li C. Yang W. Hu C. Sun Y.-M. Hu X. Chem. Eng. J. 2025;520:165922. doi: 10.1016/j.cej.2025.165922. https://dx.doi.org/10.1016/j.cej.2025.16592211 [DOI] [Google Scholar]
  11. Liu D. Tao L. Yan D. Zou Y. Wang S. ChemElectroChem. 2018;5(14):1775–1785. doi: 10.1002/celc.201800086. [DOI] [Google Scholar]
  12. Li M. Song P. Hou M. Feng W. Zhao D. Zhang J. Chem. Eng. J. 2024;479:147608. doi: 10.1016/j.cej.2023.147608. [DOI] [Google Scholar]
  13. He H. Liu S. Liu Y. Zhou L. Wen H. Shen R. Zhang H. Guo X. Jiang J. Li B. Green Chem. 2023;25(23):9501–9542. doi: 10.1039/D3GC02856A. [DOI] [Google Scholar]
  14. Zhao G. Chen T. Tang A. Yang H. Chem.–Eur. J. 2024;30(29):e202304065. doi: 10.1002/chem.202304065. [DOI] [PubMed] [Google Scholar]
  15. Gueskine V. Singh A. Vagin M. Crispin X. Zozoulenko I. J. Phys. Chem. C. 2020;124(24):13263–13272. doi: 10.1021/acs.jpcc.0c03508. [DOI] [Google Scholar]
  16. Vagin M. Gueskine V. Mitraka E. Wang S. Singh A. Zozoulenko I. Berggren M. Fabiano S. Crispin X. Adv. Energy Mater. 2021;11(3):2002664. doi: 10.1002/aenm.202002664. [DOI] [Google Scholar]
  17. Sepat N. Vagin M. Carli S. Marchini E. Caramori S. Zhang Q. Braun S. Wu Z. Ding P. Wijeratne K. Small. 2025;21(5):2409471. doi: 10.1002/smll.202409471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Yang D.-H. Tao Y. Ding X. Han B.-H. Chem. Soc. Rev. 2022;51(2):761–791. doi: 10.1039/D1CS00887K. [DOI] [PubMed] [Google Scholar]
  19. Cui X. Wu M. Liu X. He B. Zhu Y. Jiang Y. Yang Y. Chem. Soc. Rev. 2024;53(3):1447–1494. doi: 10.1039/D3CS00727H. [DOI] [PubMed] [Google Scholar]
  20. Li X. Chen T. Liu D. Mu Z. Yang B. Xiang Z. Acc. Chem. Res. 2024;57(4):590–601. doi: 10.1021/acs.accounts.3c00730. [DOI] [PubMed] [Google Scholar]
  21. Xu Y. Sun Z. Fan S. Han X. Li L. Gao Z. Wang C. J. Mater. Chem. A. 2024;12(40):27180–27205. doi: 10.1039/D4TA05404K. [DOI] [Google Scholar]
  22. Cheng R. He X. Li K. Ran B. Zhang X. Qin Y. He G. Li H. Fu C. Adv. Mater. 2024;36(25):2402184. doi: 10.1002/adma.202402184. [DOI] [PubMed] [Google Scholar]
  23. Khomenko V. Barsukov V. Katashinskii A. Electrochim. Acta. 2005;50(7–8):1675–1683. doi: 10.1016/j.electacta.2004.10.024. [DOI] [Google Scholar]
  24. Warczak M. Gryszel M. Jakešová M. Đerek V. Głowacki E. D. Chem. Commun. 2018;54(16):1960–1963. doi: 10.1039/C7CC08471D. [DOI] [PubMed] [Google Scholar]
  25. Tao S. Wang J. Zhang J. ACS Nano. 2025;19(10):9484–9512. doi: 10.1021/acsnano.4c14989. [DOI] [PubMed] [Google Scholar]
  26. Sun N. Shah S. S. A. Lin Z. Zheng Y.-Z. Jiao L. Jiang H.-L. Chem. Rev. 2025;125(5):2703–2792. doi: 10.1021/acs.chemrev.4c00664. [DOI] [PubMed] [Google Scholar]
  27. Yang D.-H. Tao Y. Ding X. Han B.-H. Chem. Soc. Rev. 2022;51:761–791. doi: 10.1039/D1CS00887K. [DOI] [PubMed] [Google Scholar]
  28. Li H. Pan F. Qin C. Wang T. Chen K. J. Adv. Energy Mater. 2023;13(28):2301378. doi: 10.1002/aenm.202301378. [DOI] [Google Scholar]
  29. He Y. Huang S. Li X. Huang Y. Cao D. Wang S. Coord. Chem. Rev. 2025;544:216972. doi: 10.1016/j.ccr.2025.216972. [DOI] [Google Scholar]
  30. Zhao X. Pachfule P. Thomas A. Chem. Soc. Rev. 2021;50(12):6871–6913. doi: 10.1039/D0CS01569E. [DOI] [PubMed] [Google Scholar]
  31. Huang B. Li L. Tang X. Zhai W. Hong Y. Hu T. Yuan K. Chen Y. Energy Environ. Sci. 2021;14(5):2789–2808. doi: 10.1039/D1EE00306B. [DOI] [Google Scholar]
  32. Song K. S. Fritz P. W. Coskun A. Chem. Soc. Rev. 2022;51(23):9831–9852. doi: 10.1039/D2CS00727D. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Gu Y. Son S. U. Li T. Tan B. Adv. Funct. Mater. 2021;31(12):2008265. doi: 10.1002/adfm.202008265. [DOI] [Google Scholar]
  34. Liu Z. Yang T. Song Y. Zhao N. Wu S. Ma Z. Gong X. Tian X. Liu Z. RSC Appl. Polym. 2025;3:746–766. doi: 10.1039/D4LP00379A. [DOI] [Google Scholar]
  35. Liu W. Yang Y. Guo L. Di J. Lau C. H. Bermeshev M. V. Shao L. Chem. Eng. J. 2024;498:155569. doi: 10.1016/j.cej.2024.155569. [DOI] [Google Scholar]
  36. Goudar S. H. Ingle D. S. Sahu R. Kotha S. Reddy S. K. Babu D. J. Kotagiri V. R. ACS Appl. Polym. Mater. 2023;5(3):2097–2104. doi: 10.1021/acsapm.2c02102. [DOI] [Google Scholar]
  37. Chanchaona N. Ding L. Lin S. Sarwar S. Dimartino S. Fletcher A. J. Dawson D. M. Konstas K. Hill M. R. Lau C. H. J. Mater. Chem. A. 2023;11(18):9859–9867. doi: 10.1039/D2TA09253K. [DOI] [Google Scholar]
  38. Li Z. Scott C. T. Suzuki-Osborne T. Lowe J. P. Taylor D. Carta M. McKeown N. B. Burrows A. D. Mascaro L. H. Marken F. Adv. Sci. 2025:e14182. doi: 10.1002/advs.202514182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Burba M. Pargoletti E. Marken F. McKeown N. B. Carta M. Vertova A. Minguzzi A. ChemElectroChem. 2024;11(2):e202300481. doi: 10.1002/celc.202300481. [DOI] [Google Scholar]
  40. Azevedo Beluomini M. Ramos Stradiotto N. Boldrin Zanoni M. V. Carta M. McKeown N. B. Fletcher P. J. Sain S. Li Z. Marken F. ACS Appl. Mater. Interfaces. 2024;16(29):37865–37873. doi: 10.1021/acsami.4c04459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Han J. Sun J. Chen S. Zhang S. Qi L. Husile A. Guan J. Adv. Mater. 2024;36(47):2408139. doi: 10.1002/adma.202408139. [DOI] [PubMed] [Google Scholar]
  42. Zhu Z. Cui J. Cao X. Yang L. Sun H. Liang W. Li J. Li A. Energy Technol. 2022;10(5):2101097. doi: 10.1002/ente.202101097. [DOI] [Google Scholar]
  43. Zhan Z. Yu J. Li S. Yi X. Wang J. Wang S. Tan B. Small. 2023;19(16):2207646. doi: 10.1002/smll.202207646. [DOI] [PubMed] [Google Scholar]
  44. Gu Y. Son S. U. Li T. Tan B. Adv. Funct. Mater. 2021;31(12):2008265. doi: 10.1002/adfm.202008265. [DOI] [Google Scholar]
  45. Tan L. Tan B. Chem. Soc. Rev. 2017;46(11):3322–3356. doi: 10.1039/C6CS00851H. https://dx.doi.org/10.1039/C6CS00851H6 [DOI] [PubMed] [Google Scholar]
  46. Huang J. Turner S. R. Polym. Rev. 2018;58(1):1–41. doi: 10.1080/15583724.2017.1344703. [DOI] [Google Scholar]
  47. Maleki F. Mashhadimoslem H. Ghaemi A. Sadeghi G. M. M. Mater. Chem. Phys. 2022;292:126779. doi: 10.1016/j.matchemphys.2022.126779. [DOI] [Google Scholar]
  48. Song W. Tang Y. Moon B.-R. Liao Q. Xu H. Hou Q. Zhang H. Yu D.-G. Liao Y. Kim I. Green Chem. 2024;26:2476–2504. doi: 10.1039/D3GC04222G. [DOI] [Google Scholar]
  49. Shu G. Zhang C. Li Y. Jiang J. X. Wang X. Li H. Wang F. J. Appl. Polym. Sci. 2018;135(8):45907. doi: 10.1002/app.45907. [DOI] [Google Scholar]
  50. Wang K. Jia Z. Yang X. Wang L. Gu Y. Tan B. J. Catal. 2017;348:168–176. doi: 10.1016/j.jcat.2017.02.024. [DOI] [Google Scholar]
  51. Yang Y. He Z. Liu Y. Wang S. Wang H.-g. Zhu G. Appl. Surf. Sci. 2021;565:150579. doi: 10.1016/j.apsusc.2021.150579. [DOI] [Google Scholar]
  52. Yang Z. Han J. Jiao R. Sun H. Zhu Z. Liang W. Li A. J. Colloid Interface Sci. 2019;557:664–672. doi: 10.1016/j.jcis.2019.09.069. [DOI] [PubMed] [Google Scholar]
  53. Barman J. Deka N. Maji P. K. Dutta G. K. ChemElectroChem. 2022;9(15):e202200677. doi: 10.1002/celc.202200677. [DOI] [Google Scholar]
  54. Chen X. Yang J. Zhou K. Lei Z. Zhao X. Zhang J. Luo H. Electrochim. Acta. 2024;491:144338. doi: 10.1016/j.electacta.2024.144338. [DOI] [Google Scholar]
  55. Dong X. Luo Y. Tao S. Liu J. Tan X. Lu Z. Wang G. Chen J. Wang R. Zhang J. J. Mater. Chem. A. 2024;12(23):13800–13809. doi: 10.1039/D4TA01084A. [DOI] [Google Scholar]
  56. Valverde-González A. Guan L. Z. Ferrer M. L. Iglesias M. Maya E. M. ACS Appl. Mater. Interfaces. 2020;12:32681–32688. doi: 10.1021/acsami.0c07412. [DOI] [PubMed] [Google Scholar]
  57. Motori A. Sandrolini F. Davolio G. J. Power Sources. 1994;48(3):361–370. doi: 10.1016/0378-7753(94)80032-4. [DOI] [Google Scholar]
  58. Yue J.-Y. Wang Y.-T. Wu X. Yang P. Ma Y. Liu X.-H. Tang B. Chem. Commun. 2021;57(94):12619–12622. doi: 10.1039/D1CC04928C. https://dx.doi.org/10.1039/d1cc04928c [DOI] [PubMed] [Google Scholar]
  59. Huang S. Zhang B. Sun H. Hu H. Wang J. Duan F. Zhu H. Du M. Lu S. Chem. Commun. 2023;59(69):10424–10427. doi: 10.1039/D3CC02948D. [DOI] [PubMed] [Google Scholar]
  60. Razzaque S. Dyachenko A. Waqar A. Dusilo K. Opallo M. Electrochem. Commun. 2024;167:107800. doi: 10.1016/j.elecom.2024.107800. [DOI] [Google Scholar]
  61. Sharp C. H. Bukowski B. C. Li H. Johnson E. M. Ilic S. Morris A. J. Gersappe D. Snurr R. Q. Morris J. R. Chem. Soc. Rev. 2021;50(20):11530–11558. doi: 10.1039/D1CS00558H. [DOI] [PubMed] [Google Scholar]
  62. Zhang H. Yin R. Wang C. Environ. Technol. Rev. 2025;14(1):427–457. doi: 10.1080/21622515.2025.2485364. [DOI] [Google Scholar]
  63. Kotp M. G. Kuo S.-W. Electrochim. Acta. 2025:146440. doi: 10.1016/j.electacta.2025.146440. [DOI] [Google Scholar]
  64. Ansari M. Helal A. Khan A. Abdulazeez I. Khan M. Y. ACS Appl. Polym. Mater. 2025;7(8):4838–4848. doi: 10.1021/acsapm.4c04148. [DOI] [Google Scholar]
  65. Li W. Wang C. Lu X. Coord. Chem. Rev. 2022;464:214555. doi: 10.1016/j.ccr.2022.214555. [DOI] [Google Scholar]
  66. Lei H. Zhang Q. Liang Z. Guo H. Wang Y. Lv H. Li X. Zhang W. Apfel U. P. Cao R. Angew. Chem. 2022;134(24):e202201104. doi: 10.1002/ange.202201104. [DOI] [PubMed] [Google Scholar]
  67. Sheng D. Li X. Zhao S. Sun C. Ma Q. Feng X. Wang B. J. Am. Chem. Soc. 2025;147(28):24838–24846. doi: 10.1021/jacs.5c06822. [DOI] [PubMed] [Google Scholar]
  68. Yu S. Guo Y. Bie M. Yang G. Huang J. Pan F. Jiang Z. Sep. Purif. Technol. 2025;356:129857. doi: 10.1016/j.seppur.2024.129857. [DOI] [Google Scholar]
  69. Liu X. Wang C. Chen C. Pan Z. Gao C. Lai W. Zhao J. Tian T. Xiao W. Coord. Chem. Rev. 2025;544:216927. doi: 10.1016/j.ccr.2025.216927. [DOI] [Google Scholar]
  70. Hou S. Razzaque S. Tan B. Polym. Chem. 2019;10(11):1299–1311. doi: 10.1039/C8PY01730A. [DOI] [Google Scholar]
  71. Shao L. Sang Y. Liu N. Wei Q. Wang F. Zhan P. Luo W. Huang J. Chen J. Sep. Purif. Technol. 2021;262:118352. doi: 10.1016/j.seppur.2021.118352. [DOI] [Google Scholar]
  72. Urban J. Škeříková V. J. Sep. Sci. 2014;37(21):3082–3089. doi: 10.1002/jssc.201400730. [DOI] [PubMed] [Google Scholar]
  73. Zhang D. Tao L. Wang Q. Wang T. Polymer. 2016;82:114–120. doi: 10.1016/j.polymer.2015.11.041. [DOI] [Google Scholar]
  74. Zhao L. Xu Q. Shao Z. Chen Y. Xue Z. Li H. Zhang J. Interfaces. ACS Appl. Mater. Interfaces. 2020;12(41):45976–45986. doi: 10.1021/acsami.0c11742. [DOI] [PubMed] [Google Scholar]
  75. Wei C. Rao R. R. Peng J. Huang B. Stephens I. E. Risch M. Xu Z. J. Shao-Horn Y. Adv. Mater. 2019;31(31):1806296. doi: 10.1002/adma.201806296. [DOI] [PubMed] [Google Scholar]
  76. Neese F. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2022;12(5):e1606. doi: 10.1002/wcms.81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Kubas A. Orain C. De Sancho D. Saujet L. Sensi M. Gauquelin C. Meynial-Salles I. Soucaille P. Bottin H. Baffert C. Nat. Chem. 2017;9(1):88–95. doi: 10.1038/nchem.2592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Becke A. D. Phys. Rev. A. 1988;38(6):3098. doi: 10.1103/PhysRevA.38.3098. [DOI] [PubMed] [Google Scholar]
  79. Perdew J. P. Phys. Rev. B: Condens. Matter Mater. Phys. 1986;33(12):8822. doi: 10.1103/PhysRevB.33.8822. [DOI] [PubMed] [Google Scholar]
  80. Grimme S. Antony J. Ehrlich S. Krieg H. J. Chem. Phys. 2010;132:154104. doi: 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]
  81. Barone V. Cossi M. J. Phys. Chem. A. 1998;102(11):1995–2001. doi: 10.1021/jp9716997. [DOI] [Google Scholar]
  82. Lee C. Y. Parr W. Phys. Rev. B: Condens. Matter Mater. Phys. 1988;37:785. doi: 10.1103/PhysRevB.37.785. [DOI] [PubMed] [Google Scholar]
  83. Becke A. D. J. Chem. Phys. 1993;98(7):5648–5652. doi: 10.1063/1.464913. [DOI] [Google Scholar]
  84. Hellweg A. Hättig C. Höfener S. Klopper W. Theor. Chem. Acc. 2007;117(4):587–597. doi: 10.1007/s00214-007-0250-5. [DOI] [Google Scholar]
  85. Weigend F. Ahlrichs R. Phys. Chem. Chem. Phys. 2005;7(18):3297–3305. doi: 10.1039/B508541A. [DOI] [PubMed] [Google Scholar]
  86. Weigend F. Häser M. Patzelt H. Ahlrichs R. Chem. Phys. Lett. 1998;294(1–3):143–152. doi: 10.1016/S0009-2614(98)00862-8. [DOI] [Google Scholar]
  87. Yanai T. Tew D. P. Handy N. C. Chem. Phys. Lett. 2004;393(1–3):51–57. doi: 10.1016/j.cplett.2004.06.011. [DOI] [Google Scholar]
  88. Sotoudeh M. Laasonen K. Busch M. Electrochim. Acta. 2023;443:141785. doi: 10.1016/j.electacta.2022.141785. [DOI] [Google Scholar]
  89. Ejaz M. Mohamed M. G. Kotp M. G. Elewa A. M. Kuo S.-W. Colloids Surf., A. 2025:137239. doi: 10.1016/j.colsurfa.2025.137239. [DOI] [Google Scholar]
  90. Shao L. Sang Y. Liu N. Wei Q. Wang F. Zhan P. Luo W. Huang J. Chen J. Sep. Purif. Technol. 2021;262:118352. doi: 10.1016/j.seppur.2021.118352. [DOI] [Google Scholar]
  91. Qian Q. Hu H. Huang S. Li Y. Lin L. Duan F. Zhu H. Du M. Lu S. Carbon. 2023;202:81–89. doi: 10.1016/j.carbon.2022.11.014. [DOI] [Google Scholar]
  92. Wang H. Li Z. Meng Z. Guo X. Du Y. Yang H. New J. Chem. 2021;45:7060–7064. doi: 10.1039/D1NJ00089F. [DOI] [Google Scholar]
  93. Wang K. Wu Z. Tang L. Ji N. Zhang C. Jia Z. Liu M. J. Polym. Sci. 2024;62(8):1714–1721. doi: 10.1002/pol.20230759. [DOI] [Google Scholar]
  94. Yang S. Wang X. Tan B. Macromol. 2023;56(3):1213–1222. doi: 10.1021/acs.macromol.2c01949. [DOI] [Google Scholar]
  95. Dawson R. Cooper A. I. Adams D. J. Prog. Polym. Sci. 2012;37(4):530–563. doi: 10.1016/j.progpolymsci.2011.09.002. [DOI] [Google Scholar]
  96. Xu J. Shi L. Wang J. Lu S. Wang Y. Gao G. Ding S. Carbon. 2018;138:348–356. doi: 10.1016/j.carbon.2018.07.013. [DOI] [Google Scholar]
  97. Iwan A. Sek D. Prog. Polym. Sci. 2011;36(10):1277–1325. doi: 10.1016/j.progpolymsci.2011.05.001. [DOI] [Google Scholar]
  98. Rybakiewicz R. Zagorska M. Pron A. Chem. Pap. 2017;71(2):243–268. doi: 10.1007/s11696-016-0097-0. [DOI] [Google Scholar]
  99. Batchelor-McAuley C. Curr. Opin. Electrochem. 2023;37:101176. doi: 10.1016/j.coelec.2022.101176. [DOI] [Google Scholar]
  100. Martínez-Fernández M. Segura J. L. ChemSusChem. 2024;17(18):e202400558. doi: 10.1002/cssc.202400558. [DOI] [PubMed] [Google Scholar]
  101. Hu X. M. Mendoza D. Madsen M. R. Joulié D. Lassalle-Kaiser B. Robert M. Pedersen S. U. Skrydstrup T. Daasbjerg K. ChemSusChem. 2020;13(23):6360–6369. doi: 10.1002/cssc.202001311. [DOI] [PubMed] [Google Scholar]
  102. Masa J. Batchelor-McAuley C. Schuhmann W. Compton R. G. Nano Res. 2014;7(1):71–78. doi: 10.1007/s12274-013-0372-0. [DOI] [Google Scholar]
  103. Hašková A. Bashta B. Titlová Š. Brus J. Vagenknechtová A. Vyskočilová E. Sedláček J. Macromol. Rapid Commun. 2021;42(15):2100209. doi: 10.1002/marc.202100209. [DOI] [PubMed] [Google Scholar]
  104. Wang G. Dou B. Wang J. Wang W. Hao Z. RSC Adv. 2013;3(43):20523–20531. doi: 10.1039/C3RA41450G. [DOI] [Google Scholar]
  105. Bass M. Berman A. Singh A. Konovalov O. Freger V. J. Phys. Chem. B. 2010;114(11):3784–3790. doi: 10.1021/jp9113128. [DOI] [PubMed] [Google Scholar]
  106. DeCaluwe S. C. Baker A. M. Bhargava P. Fischer J. E. Dura J. A. Nano Energy. 2018;46:91–100. doi: 10.1016/j.nanoen.2018.01.008. [DOI] [Google Scholar]
  107. Wood III D. L. Chlistunoff J. Majewski J. Borup R. L. J. Am. Chem. Soc. 2009;131(50):18096–18104. doi: 10.1021/ja9033928. [DOI] [PubMed] [Google Scholar]
  108. Chlistunoff J. J. Power Sources. 2014;245:203–207. doi: 10.1016/j.jpowsour.2013.06.128. [DOI] [Google Scholar]
  109. Ayad A. Naimi Y. Bouet J. Fauvarque J.-F. J. Power Sources. 2004;130(1–2):50–55. doi: 10.1016/j.jpowsour.2003.11.064. [DOI] [Google Scholar]
  110. Ramezanipour Penchah H. Ghanadzadeh Gilani H. Ghaemi A. J. Chem. Eng. Data. 2020;65(10):4905–4913. doi: 10.1021/acs.jced.0c00541. [DOI] [Google Scholar]
  111. Cucu E. Dalkılıç E. Altundas R. Sadak A. E. Microporous Mesoporous Mater. 2022;330:111567. doi: 10.1016/j.micromeso.2021.111567. [DOI] [Google Scholar]
  112. Sadak A. E. Microporous Mesoporous Mater. 2021;311:110727. doi: 10.1016/j.micromeso.2020.110727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  113. Bard A. J., Faulkner L. R. and White H. S., Electrochemical Methods: Fundamentals and Applications, John Wiley & Sons, 2022 [Google Scholar]
  114. Dutta D. Wang J. Li X. Zhou Q. Ge Z. Small. 2022;18(37):2202369. doi: 10.1002/smll.202202369. [DOI] [PubMed] [Google Scholar]
  115. Tao D. Feng L. Chao Y. Liang C. Song X. Wang H. Yang K. Liu Z. Adv. Funct. Mater. 2018;28(43):1804901. doi: 10.1002/adfm.201804901. [DOI] [Google Scholar]
  116. Li L. Xu L. Hu Z. Yu J. C. Adv. Funct. Mater. 2021;31(52):2106120. doi: 10.1002/adfm.202106120. [DOI] [Google Scholar]
  117. Zhang H. Zhu M. Schmidt O. G. Chen S. Zhang K. Adv. Energy Sustainability Res. 2021;2(4):2000090. doi: 10.1002/aesr.202000090. [DOI] [Google Scholar]
  118. Marken F. Carta M. McKeown N. B. Anal. Chem. 2021;93(3):1213–1220. doi: 10.1021/acs.analchem.0c04554. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

RA-016-D6RA03454C-s001
RA-016-D6RA03454C-s001.pdf (2MB, pdf)

Data Availability Statement

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: detailed description of synthesis, description of analysis of RDE, RRDE and CA data, IR spectra, XPS spectra, 13C CP/MAS NMR spectra, pore size distribution, CV and CA curves, SEM images, images of structural motifs for DFT calculations, simulated electronic excitation spectra, elemental composition data, summary of porosity measurements, reported values of onset potentials, and XYZ coordinates of structural motifs for DFT calculations. See DOI: https://doi.org/10.1039/d6ra03454c.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES