View PDF Version
 
DOI: 10.1039/D4TA02721C (Paper) J. Mater. Chem. A, 2024, Advance Article

Influence of surface terminal groups on the efficiency of two-electron oxygen reduction reaction catalyzed by iron single atoms on Ti3C2Tx (T = Cl, Br, NH) MXene†

Dawid Daniël Krugera, Juan José Delgadob, F. Javier Recio*c, Sara Goberna-Ferróna, Ana Primo*a and Hermenegildo García*a
aInstituto Universitario de Tecnología Química CSIC-UPV, Universitat Politècnica de València, Av. De los Naranjos s/n, València, 46022, Spain. E-mail: aprimoar@itq.upv.es; hgarcia@itq.upv.es
bMaterial Science and Metallurgy Engineering and Inorganic Chemistry, University of Cádiz, 11510 Puerto Real, Cádiz, Spain
cDepartamento de Química Física Aplicada, Facultad de Ciencias, Universidad Autónoma de Madrid, C/Francisco Tomás y Valiente, 7, Cantoblanco, 28049 Madrid, Spain. E-mail: javier.recioc@uam.es

Received 20th April 2024 , Accepted 4th August 2024

First published on 13th August 2024

Abstract

Two Ti3C2 MXenes having single atom Fe were prepared from a Ti3AlC2 precursor by Al etching using an FeX2 (X = Cl, Br) Lewis acid as an etching reagent in NaCl/KCl molten salt. After removal of the excess Fe metal using concentrated aqueous HX solution, the resulting MXenes having Cl or Br surface terminations were concomitantly doped with single Fe atoms. A third sample was prepared from Fe(SA)-Ti3C2Brx by exchanging Br with –NH– using LiNH2 in molten salt. The presence of single atom Fe was determined by aberration corrected high-angle annular dark-field scanning transmission electron microscopy (AC-HAADF-STEM) with atomic resolution where isolated Fe atoms were observed. These Fe(SA)-Ti3C2Tx materials exhibit electrocatalytic activity for the oxygen reduction reaction (ORR), with the selectivity to H2O2 depending on the surface termination and the electrolyte pH. Involvement of Fe SA was inferred from the onset potential of the ORR and the Fe(III)/Fe(II) reduction potential and the change of the Tafel slope in this region. Under the best conditions, at pH 13, Fe(SA)-Ti3C2(NH)x exhibits for the ORR an average electron transfer number (n) of 2.5, which corresponds to an H2O2 generation efficiency of 75%; these data compare well with those of the best Fe single-atom electrocatalysts.

Introduction

Due to their high electrical conductivity and large metal content, MXenes have been used since their discovery in 2011 as electrocatalysts for reactions of high importance for renewable electricity use and storage, including the hydrogen evolution reaction (HER), CO2 reduction reaction (CO2RR), and oxygen evolution reaction (OER), among others.1–5 MXenes offer several possibilities to optimize their performance as electrocatalysts, including selection of the composition, structure and morphology of the material as well as the nature of the surface terminal groups.3,6 It is known that for a given MXene, the surface termination can tune the conducting/semiconducting properties of the material and this can play a significant role in the efficiency of the MXene as an electrocatalyst.7–9

Besides the intrinsic electrocatalytic properties of MXenes or their use as additives in electrode preparation, MXenes are suitable supports of single metal species as active catalytic sites.1,10–12 Due to the harsh etching conditions required in MXene preparation from MAX phase precursors, generation of atom vacancies and defects frequently occurs as a side effect of the etching of the “A” element with aqueous fluoride-based etching methods.2,13,14 These metal vacancies have been found to be especially appropriate to nest single atoms of other metals that can exhibit high activity and stability in electrocatalytic processes.10,11,15–17 More recently, it has been found that vacancy-confined transition-metal single atoms on MXene could also be achieved during MXene synthesis by the Lewis acid molten salt method, where the species of the metal single-atom is derived from the Lewis acid molten salt etchant itself (usually a late transition metal halide).12,18

Electrochemical production of H2O2 by the 2e− oxygen reduction reaction is an attractive technology as an alternative to the ubiquitous anthraquinone process, which is an energy and material intensive process that depends on large-scale, centralized production to be economically competitive. In contrast, selective 2e− ORR offers a means to produce H2O2 on-site in a scalable way, avoiding many of the issues associated with the anthraquinone process.19,20 However, electrochemical H2O2 production is still limited by the lack of efficient electrocatalysts. This is due to the electrochemical instability of H2O2 that tends to undergo further reduction to H2O, thus resulting in a low 2e− ORR selectivity. Noble metals supported on MXene, including single-atom catalysts, have shown promise as ORR electrocatalysts.21,22 Electrocatalysts based on non-precious materials are highly desirable, in terms of both material costs and sustainability. It well known that Fe single-atom systems exhibit exceptional performance for the ORR, although, for many of the commonly studied Fe-based materials, the 4e− pathway is favored.23 In one example, it has been reported that Fe phthalocyanine (FePc) as a source of Fe–N4 moieties in combination with Ti3C2Tx MXene as a support exhibits high ORR efficiency, better than that of the unsupported electrocatalyst.24 Importantly, the selectivity between the 2e− and 4e− ORR pathways is highly dependent on the local coordination environment.25,26 Recent progress in experimental methods has enabled post-synthetic surface modifications of molten-salt derived MXene, with substitution of halide (Cl, Br) surface groups with O, S, and NH, for example.7 In this context, more efforts are still necessary to exploit the opportunities that MXenes offer as electrocatalysts for 2e− ORR and specifically to combine surface functionalization with the use of MXene as a single-atom support.

In the present work we report that a molten-salt preparation procedure of Ti3C2Tx affords in a straightforward manner Fe single atoms supported on Ti3C2Tx MXene with various surface terminations, achieving simultaneously: (i) Al etching of the MAX phase, (ii) adequate surface functionalization, and (iii) healing of structural Ti vacancies by installing Fe single atoms (Fe(SA)-Ti3C2Tx). The resulting Fe(SA)-Ti3C2Tx materials (T = Cl, Br, NH) all exhibit electrocatalytic activity for the ORR, there being at basic pH values a clear influence of the nature of the surface terminal group on the electrocatalytic activity and the selectivity, with Fe(SA)-Ti3C2(NH)x being the best performing material for the 2e− pathway.

Experimental

Materials and methods

Materials. Ti3AlC2 used as a MXene precursor was a commercial sample from Nanochemazone. FeCl2 and FeBr2 used as Lewis acids, KCl, NaCl, LiBr, KBr and CsBr to form the molten salt, LiNH2 and (TiO)SO4 used for colorimetric H2O2 titration and PTFE in 60 wt% emulsion for electrode preparation were supplied by Sigma-Aldrich. Concentrated HCl and HBr aqueous solutions were procured from Scharlau.
Synthesis of Fe single atoms on Ti3C2Clx and Ti3C2Brx MXene. The Lewis acid molten salt (LAMS) method was used to prepare Ti3C2Clx MXene,27 containing Fe single atoms, from the parent Ti3AlC2 MAX phase. Anhydrous FeCl2 was used as the A-site etchant in combination with dry NaCl/KCl in the 1[thin space (1/6-em)]:[thin space (1/6-em)]1 eutectic proportion as a diluting molten salt solvent. A mass of 1.00 g of Ti3AlC2 powder was introduced into a N2-filled glovebox and ground together with FeCl2, NaCl, and KCl using a mortar and pestle, with Ti3AlC2[thin space (1/6-em)]:[thin space (1/6-em)]FeCl2[thin space (1/6-em)]:[thin space (1/6-em)]NaCl[thin space (1/6-em)]:[thin space (1/6-em)]KCl in the molar ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]3[thin space (1/6-em)]:[thin space (1/6-em)]2[thin space (1/6-em)]:[thin space (1/6-em)]2. The solid mixture was transferred to an alumina crucible. The crucible was then placed into a tube furnace, continuously flushed with Ar (150 ml min−1) and heated at a rate of 5 °C min−1 to 700 °C, holding this temperature for 8 h. After this time, the system was allowed to cool naturally, all under Ar flow. To remove excess FeCl2, NaCl and KCl, the solid was dispersed into Milli-Q water (200 ml), filtered, and then washed with an additional 200 ml of Milli-Q water. The filtered product containing the Fe/Ti3C2Clx was then dried at 60 °C under vacuum. The Fe/Ti3C2Clx powder was then stirred in 4 M HCl (300 mg powder in 50 ml) for 12 h to dissolve the bulk Fe metal particles. The liquid was filtered, and the solid was washed with 200 ml of Milli-Q water and finally dried at 60 °C under vacuum for 3 h. This sample is referred to as Fe(SA)-Ti3C2Clx.

For the preparation of Ti3C2Brx MXene, containing Fe single atoms, the same method as for Fe(SA)-Ti3C2Clx was used, except that the etchant was anhydrous FeBr2, and an LiBr/KBr mixture, in the eutectic proportion (60[thin space (1/6-em)]:[thin space (1/6-em)]40 molar ratio), was used as the diluting molten salt solvent.28 The starting proportions of Ti3AlC2[thin space (1/6-em)]:[thin space (1/6-em)]FeBr2[thin space (1/6-em)]:[thin space (1/6-em)]LiBr[thin space (1/6-em)]:[thin space (1/6-em)]KBr were thus 1[thin space (1/6-em)]:[thin space (1/6-em)]3[thin space (1/6-em)]:[thin space (1/6-em)]3.6[thin space (1/6-em)]:[thin space (1/6-em)]2.4 in molar ratio. Furthermore, instead of HCl, concentrated HBr (38 wt%) was used to dissolve the Fe excess from the Fe/Ti3C2Brx powder (300 mg powder in 10.9 ml). This sample is referred to as Fe(SA)-Ti3C2Brx.

Synthesis of Fe single atoms on NH-modified Ti3C2 MXene. A method of surface modification reported by Kamysbayev et al. was adapted for the synthesis of Fe single atoms on NH-functionalized Ti3C2 MXene from Fe(SA)-Ti3C2Brx.7 Briefly, in a N2-filled glovebox, 70 mg of Fe(SA)-Ti3C2Brx was mixed with 1.777 g of finely ground and dried CsBr/KBr/LiBr (25[thin space (1/6-em)]:[thin space (1/6-em)]18.9[thin space (1/6-em)]:[thin space (1/6-em)]56.1 molar ratio) and 20 mg of LiNH2 into a round-bottom flask. The mixture was heated to 280 °C under a gentle flow of N2 under magnetic stirring and held at that temperature for 3 h. After cooling naturally, 30 ml of pure formamide was injected into the flask. The flask was then sonicated in a bath sonicator for 1 h to disperse the product and dissolve the salts. The solid was filtered and washed with additional acetonitrile and methanol and allowed to dry under ambient conditions. This sample is denoted as Fe(SA)-Ti3C2(NH)x.

Physicochemical characterization

Inductively Coupled Plasma-Optical Emission Spectroscopy (ICP-OES) elemental analyses were performed in a PerkinElmer Avio 560 Max instrument after completely dissolving the samples with aqua regia using standard samples for calibration. Attenuated total reflectance Fourier transform IR spectroscopy (ATR-FTIR) was performed in a Bruker Tensor 27 spectrophotometer equipped with a diamond ATR accessory in the range from 4000 to 400 cm−1 with a resolution of 4 cm−1 and averaging four spectra. X-ray photoelectron spectroscopy (XPS) analyses were performed in a Kratos AXIS Ultra DLD spectrometer equipped with a 165 mm hemispherical MCD-9 detector under ultrahigh vacuum (1.3 × 10−13 atm) using Al Kα monochromatic X-ray radiation (hν = 1486.74 eV) for sample excitation. The binding energy values were calibrated against graphitic C 1s as a reference. XPS data were treated using the Casa XPS processing software using the Shirley function for background correction. Powder XRD was carried out using a Philips XPert diffractometer with Ni filtered Cu Kα radiation.

FESEM images were acquired using a ZEISS GeminiSEM 500 electron microscope having an EDX detector. The samples were deposited directly on the support and subsequently investigated. Transmission electron microscopy (TEM) images of the Fe(SA)-Ti3C2Tx MXene samples were taken in a JEOL JEM 2100F electron microscope under 200 kV as accelerating voltage. Fe(SA)-Ti3C2Tx samples were suspended on methanol by ultrasound depositing one microdrop of this suspension of the material immediately after sonication onto a carbon-coated holey Cu TEM sample holder. The solvent was spontaneously evaporated at room temperature, before introducing the sample in the microscope chamber.

For aberration-corrected transmission electron microscopy (AC-TEM) a double Cs aberration-corrected FEI Titan3 Themis 60–300 microscope was used. This advanced equipment, operating at 200 kV, is set up with a monochromator, an X-FEG gun, and a high efficiency XEDS ChemiSTEM with 4 windowless SDD detectors. High-resolution scanning transmission electron microscopy (HR-STEM) imaging was conducted with a high-angle annular dark-field (HAADF) detector, allowing for the accurate detection of single atoms of heavier elements than the support. XEDS mappings were carried out with a beam current of 200 pA and a dwell time per pixel of 128 μs. The elemental maps obtained were enhanced using a Gaussian blur filter of 0.8 through Velox software to improve visual quality. The XEDS included in this work were obtained by adding all the individual spectra of the mapping. This was done to reduce the sample damage as much as possible.

The Fe K-edge X-ray absorption fine structure spectra were collected at the BL14W beamline in the Shanghai Synchrotron Radiation Facility (SSRF). The storage rings of SSRF were operated at 3.5 GeV with a stable current of 200 mA. Data collection was carried out in fluorescence mode with a Lytle detector using a Si(111) double-crystal monochromator. All spectra were collected under ambient conditions.

Electrochemical characterization

Electrocatalyst inks were prepared in each case by dispersing 5 mg of the MXene powder and 5 mg of Vulcan XC-72 carbon black in a vial containing 1 ml of isopropanol and 4 ml of Milli-Q water, and stirring the mixture for 12 h. A volume of 35 μl of Nafion (10 wt% in alcohol, Sigma-Aldrich) was added to this dispersion, and sonicated with a microtip sonicator for 20 pulses (50% duty cycle, 40% amplitude).

Electrochemical measurements were performed using a rotating ring disk electrode (RRDE, E6R2 Series, Pine Research Instrumentation) with a glassy carbon (GC) disk (5.5 mm diameter) with a Pt ring, with a collection efficiency (Nc) of 38%. The RRDE was modified with the electrocatalyst by spin coating, dropping 20 μl of catalyst ink onto the GC disk while rotating at 300 rpm, and then drying under a gentle N2 flow. An Ag|AgCl (sat. KCl) electrode was used as the reference electrode, and a graphite rod as the counter electrode. Potentials are reported against the reversible hydrogen electrode (RHE), using the expression: ERHE = EAg|AgCl + 0.0592 pH + 0.199 V. Electrochemical experiments were performed in a standard three-electrode RRDE cell. Supporting electrolytes were 0.1 M KOH as the alkaline electrolyte, 0.5 M Na2SO4 as the neutral electrolyte, and 0.5 M H2SO4 as the acidic electrolyte. Unless stated otherwise, cyclic voltammetry (CV) measurements were performed with N2 saturated electrolytes at a scan rate of 5 mV s−1.

The electrochemical ORR performance was tested by linear sweep voltammetry (LSV) in O2 saturated electrolyte, with a rotation rate of 1600 rpm, unless otherwise indicated, and with the Pt ring biased at 1.2 V vs. RHE. An Autolab PGSTAT302N bipotentiostat/galvanostat (Metrohm) was used to conduct CV and LSV electrochemical tests. The mean number of electrons transferred was determined using eqn (1), where n is the electron transfer number, idisk and iring are the currents measured for the GC ring and the Pt ring, and Nc is the collection efficiency of the RRDE Pt ring. The kinetic current density, jKinetic, was determined using eqn (2), considering the total measured current density, j, and the diffusion limited current density, jDiff, which were estimated from the current plateau of the catalyst modified glassy carbon disk at a high overpotential.

 
image file: d4ta02721c-t1.tif(1)
 
image file: d4ta02721c-t2.tif(2)

Measurements of the electrochemically active surface area (ECSA) were performed by the capacitance method, using standard glassy carbon electrodes (3 mm diameter), modified with 7.07 μl of catalyst ink (0.1 mgcat cm−2) and dried under a N2 flow. For these measurements, the cyclic voltammograms were recorded at scan rates of 5, 10, 20, 40, 80, and 160 mV s−1 in N2 saturated electrolytes using a potential window of 0.2 V, centred on the open-circuit potential (OCP). The double-layer capacitance was then determined using eqn (3), where ian and icat are the anodic and cathodic currents at the OCP, Cdl is the double-layer capacitance, and ν is the scan rate. The ECSA was calculated using eqn (4), considering a specific capacitance, Cs, of 0.040 mF cm−2, which is a typical value for atomically smooth planar surfaces.29

 
image file: d4ta02721c-t3.tif(3)
 
image file: d4ta02721c-t4.tif(4)

H2O2 production tests

For hydrogen peroxide generation and quantification tests, circular Toray carbon papers (TGPH-120) with a radius of 1.5 cm were used as carbon supports to deposit the active electrocatalyst. The carbon substrates were modified with different catalysts (Fe(SA)-Ti3C2Tx, with T = Cl, Br, NH). Inks of the electrocatalysts were prepared by suspending 10 mg ml−1 of the solid in ethanol and 3.5 mg ml−1 of PTFE. The suspensions were sonicated for 30 min and then added dropwise to the Toray carbon support up to a final density of 2 mg cm−2. The prepared electrodes were then dried in an oven at 60 °C for 30 min.

Experiments for H2O2 production were performed in a cell filled with 50 ml of 0.05 M Na2SO4 at pH 7 adjusted with a diluted solution of KOH. The modified carbon paper electrodes were assembled in a 3D printed air-cathode prototype using a mesh of stainless steel as contact. A constant air flow of 1.2 L min−1 was supplied during the experiments to the electrode to ensure a constant oxygen concentration. A graphite rod was used in the experiments as a counter-electrode and a constant current density of −10 mA cm−2 was applied for 90 min. Aliquots of 900 μl were drawn from the solution at 5, 10, 20, 30, 45, 60 and 90 min and were mixed with 100 μl of the commercial solution of titanium(IV) oxysulfate forming a yellow complex. The H2O2 concentration was analyzed by measuring the complex absorbance at λmax = 410 nm in the UV-Vis spectrum with a PerkinElmer UV-Vis Lambda 365 spectrophotometer. The stability of the best performing material was tested by repeating the chronopotentiometric H2O2 generation 5 times.

Results and discussion

Two of the samples under study were readily obtained starting from the Ti3AlC2 precursor by molten salt Lewis acid etching using either FeCl2 or FeBr2 as an etchant, followed by acid HCl or HBr treatment to remove the metallic Fe formed in the process. Scheme 1 illustrates the preparation procedure, while Table 1 lists the samples under study and their main analytical and textural data. A third material studied was Fe single-atom anchored on Ti3C2(NH)x. This sample was prepared from Fe(SA)-Ti3C2Br2 by exchanging surface Br atoms with NH groups using LiNH2 as a reagent under flux conditions.
image file: d4ta02721c-s1.tif
Scheme 1 Preparation of Fe single atoms on Ti3C2Tx (T = Cl, Br) MXene. The Ti3AlC2 MAX phase is etched with a molten Fe(II) halide salt in an alkali metal halide eutectic melt, forming an Fe/Ti3C2Tx composite and evolving AlX3 (X = Cl, Br) as a gaseous by-product. Bulk metallic Fe is removed by washing with concentrated acid (HCl or HBr), after which single atoms of Fe on MXene remain.
Table 1 Elemental composition of Fe(SA)-Ti3C2Tx samples determined by combined techniques of EDX under STEM (Ti, Fe, Al, Cl, Br), and combustion elemental analysis (C, N, H)
Sample Lattice parametersa Atomic ratiob
c, nm a, nm Ti Fe Al Cl Br C N H
a Lattice parameters c and a (as P63/mmc space group) were calculated from XRD (002) and (110) peaks respectively; values in brackets were estimated from TEM images.b Atomic ratios have been calculated assuming a Ti[thin space (1/6-em)]:[thin space (1/6-em)]C ratio of 3[thin space (1/6-em)]:[thin space (1/6-em)]2.
Fe(SA)-Ti3C2(NH)x 2.50 (2.4) 0.32 (0.32) 3 0.13 0.18 — — 2 0.34 2.4
Fe(SA)-Ti3C2(Brx) 2.34 (2.4) 0.32 (0.31) 3 0.07 0.21 — 0.63 2 0.02 2.5
Fe(SA)-Ti3C2(Clx) 2.21 (2.3) 0.32 (0.31) 3 0.04 0.11 1.9 — 2 0.03 1.2

Physicochemical characterization

After their synthesis, the samples were characterized by analytical tools to determine the residual Fe content as well as the percentage of Ti and the nature of surface terminal groups. The results are presented in Table 1. It should be noted at this point that even though the Fe content measured by EDX is similar for the three samples, variations occur at the nano-scale, and the experimental procedure does not allow a more precise control of the exact Fe percentage. Nevertheless, the bulk measurements by ICP-OES indicate iron contents of 1.1 wt%, 0.87 wt%, and 1.04 wt% for the Fe(SA)-Ti3C2Tx samples with T = Cl, Br, and NH, respectively. The variability can be attributed to the etching treatment employing different Fe halide salts having different Lewis acid strength and hardness–softness nature that should generate different densities of atom vacancies depending on the Fe salt, either FeCl2 or FeBr2. It is also worth noting that, probably due to the stronger Ti–Cl bond, the stoichiometric Cl content was experimentally found to be closer to the theoretical value of 2, while in the case of Br surface termination, the weaker Ti–Br bond makes it easier to be replaced by oxygen group terminations during the aqueous work-up. For this reason, the Br content for the sample under study was lower than the expected stoichiometric value of 2, indicating that in the process of removing the excess metallic Fe, some Br− surface atoms are replaced by OH or O groups on the surface.

The success of Al etching from the MAX phase precursor in the molten salt treatment was firmly supported by XRD, whereby the characteristic (002) peak and the corresponding harmonics were observed. Furthermore, the bulkiness of the surface terminal groups and their interaction with the H-bridge are reflected by the position of the (002) peak as well as the related (00l) diffraction bands in the XRD pattern. Fig. 1 presents the XRD patterns for the accordion-like samples, showing the shift in the most intense (002) peak position after Al etching of the Ti3AlC2 as a function of the surface termination. From these 2θ values, the corresponding interlayer distances can be determined by applying Bragg's law. The resulting c lattice parameters are included in Table 1. These data are in agreement with previous reports for –Cl and –Br terminated Ti3C2Tx.7,30,31


image file: d4ta02721c-f1.tif
Fig. 1 PXRD patterns of (a) the Ti3AlC2 MAX phase precursor and its derived MXene samples, (b) Fe(SA)-Ti3C2Clx, (c) Ti3C2Brx, and (d) Ti3C2(NH)x with * symbols showing the presence of a minor phase of α-Al2O3. The inset shows the shift in the position of the (002) peak indicating differences in the interlayer spacing depending on the surface termination.

The morphology of the MXene samples was analysed by SEM that also allowed elemental analysis using EDX. Fig. 2a–c show a selected set of SEM images of the accordion-like Fe(SA)-Ti3C2Tx MXene forms, while the ESI† provides additional images with EDX elemental mapping of the samples under study (Fig. S1–S3†). As expected, SEM images show the typical morphology for the expanded multilayer MXene phase resulting after the molten salt etching in which the particles show that they are constituted by the loose stacking of MXene sheets. EDX analysis of the samples shows, in addition to the expected Ti and C elements for all samples, the presence of Cl or Br homogeneously distributed throughout the particles of the samples etched with FeCl2 or FeBr2, respectively. These halogens are associated with the surface terminations of these Ti3C2Tx MXene.27,31,32 In the case of the LiNH2 treated MXene, unambiguous detection of N using EDX was not possible due to the overlap of the Ti LIIIMI (0.3953 keV) and LIIMI (0.4013 keV) signals with the N KL transition (0.3924 keV).33 The presence of N could be confirmed by combustion elemental analysis, with a C[thin space (1/6-em)]:[thin space (1/6-em)]N ratio of 2[thin space (1/6-em)]:[thin space (1/6-em)]0.34. Besides the elements of the corresponding Ti3C2Tx formula, the presence of residual Al, in an atomic percentage below 3%, was also monitored. This residual Al has been frequently reported and has been attributed to the formation of some Al oxides remaining on the sample.32 Importantly, no Fe nanoparticles that should have been formed in the Al etching were observed in SEM, indicating that the HX acid treatment has been efficient in removing them. However, the presence of Fe as a minor constituent was still observed with a very weak EDX signal, but clearly above the background noise (Fig. S1–S3†).


image file: d4ta02721c-f2.tif
Fig. 2 SEM images of (a) Fe(SA)-Ti3C2Clx, (b) Fe(SA)-Ti3C2Brx and (c) Fe(SA)-Ti3C2(NH)x. Scale bars are 1 μm. Cross section (d) and frontal (e) TEM images of Fe(SA)-Ti3C2(NH)x.

In agreement with the SEM study, high-resolution cross-section TEM images show the multilayer structure of the accordion-like form of the Fe(SA)-Ti3C2Tx samples (Fig. 2e and S7a and c†). From these images the interlayer spacings and c lattice parameter of the samples could be measured. Frontal views, along the <001> axis, reveal the high crystallinity of the Fe(SA)-Ti3C2Tx samples under study, and allow for the measurement of the a lattice parameters from the corresponding Fast Fourier Transforms (FFT). These values are summarized in Table 1 and they are in accordance with the values obtained by XRD for these samples, showing minor variations due to the various surface terminations and local deviations under TEM conditions. The shortest interlayer distance was that of the Cl-terminated Ti3C2Cl2, increasing in the Br-terminated Ti3C2Brx, most likely due to the relatively larger atomic size of Br. The largest interlayer distance corresponds to the –NH termination, which could reflect the presence of strongly co-adsorbed intercalated H2O within the layers due to the formation of hydrogen-bridge bonds, and consequently a more expanded structure. As commented earlier, the presence of residual Fe was detected in the Fe(SA)-Ti3C2Tx samples by EDX elemental mapping showing a high Fe dispersion throughout the material, without the observation of Fe nanoparticles (see Table 1 and Fig. S4–S6†). From the electrocatalytic point of view it is of interest to determine the structure and distribution of these sites on the samples.

To address how Fe was present on the Fe(SA)-Ti3C2Tx samples, atomic resolution AC-HAADF-STEM images of the Fe(SA)-Ti3C2Clx sample were taken. Fig. 3 shows representative AC-HAADF-STEM images of the Fe(SA)-Ti3C2Clx sample. In structural terms, the FFT of the image shows a hexagonal pattern with features corresponding to an atomic interplanar distance of 2.7 Å. This is consistent with a simulated electron diffraction pattern along the <001> zone axis of the P63/mmc space group of the Ti3C2 MXene structure with a lattice parameter of 3.2 Å. These AC-HAADF-STEM images clearly show the presence of isolated atoms with higher Z-contrast compared to the surrounding matrix. The fact that these dots correspond to Fe was confirmed by EDX analyses that conclusively detected a weak Fe Kα peak accompanying the intense Ti K signals. In related precedents, it has also been shown that the molten salt etching of MAX precursors can lead to the installation of single metal atoms of Cu and Co in the MXene structure,12,18 and this occurrence is also apparent in the present case with Fe(SA)-Ti3C2Tx samples.


image file: d4ta02721c-f3.tif
Fig. 3 Representative high resolution STEM images (a and b) of sample Fe(SA)-Ti3C2Cl2. The experimental digital diffraction pattern and simulated diffraction pattern of the selected area are included in (a).

The harsh conditions of the molten salt etching cause the leaching of some Ti atoms in the structure that are suitable nests for coordination with other transition metals present in the medium. The HX treatment to remove the Fe0 nanoparticles is unable to completely remove those Fe species that are strongly bound to the MXene structure.

Further information on the nature of the surface terminal groups was obtained by IR spectroscopy. The ATR-FTIR spectra of the samples are shown in Fig. S8.† They show the presence of weak O–H stretching vibration bands at about 3400 cm−1, accompanied by a bending vibration at 1630 cm−1, both attributable to co-adsorbed H2O present on the samples. A band at around 450 cm−1 with shifts depending on the surface can be attributed to the Ti–C skeletal vibration of the Ti3C2Tx MXene (see the inset of Fig. S8†).34–36 Overall these IR spectra conclusively rule out that the Fe(SA)-Ti3C2Tx samples contain oxygenated groups on the surface, this being in agreement with precedents in the molten salt synthetic method as a procedure that selectively introduces halogens as surface groups in MXenes.7,12,37

Further information on the distribution of the elements present on the Fe(SA)-Ti3C2Tx samples among different oxidation states and coordination spheres was obtained by XPS. Fig. 4 and S9† show the XPS analysis of the samples under study The XPS Ti 2p core level can be deconvoluted into six components appearing at 464.8 (459.1), 461.8 (456.5) and 459.4 (455.5) eV that agree well with the values reported for TiO2-xNH, C–Ti2+ and C–Ti3+, respectively,7,38 with the numbers in brackets corresponding to the Ti 2p3/2 splitting of the Ti 2p1/2 components indicated outside the brackets. The Ti 2p3/2 vs. Ti 2p1/2 relative intensity was 2[thin space (1/6-em)]:[thin space (1/6-em)]1.25. The XPS C 1s spectrum was adequately fitted to four components at 288.8, 285.9, 284.8 and 282.5 eV that can be ascribed to C atoms in COO, CO, C–C and Ti–C, respectively. For Ti3C2Cl2 and Ti3C2Brx, a single component due to Cl or Br at 198.4/200.0 eV (Cl 2p1/2/Cl 2p3/2, relative intensity ratio 1) and 69.7 (75.4) eV (Br 3d5/2/Br 3d3/2, relative intensity ratio 4.25) was recorded. In the case of Ti3C2(NH)x (Fig. 4) two N 1s peaks appear with the peak at 395.7 eV matching well with the expected value for Ti–N, while the peak at 399.6 eV matched with that of Ti–N–O (or Ti–O–N or N–Ti–O),38,39 as well as an expected contribution from Fe–N.40 Overall, the XPS data here recorded agree well with prior XPS data in the literature for Ti3C2 samples obtained by the Lewis acid molten salt method,7 in which the most salient features are the small component of the Ti–C in the C 1s peak deconvolution and the shape of the complex Ti 2p band with poorly separated Ti 2p3/2 and Ti 2p1/2 components.


image file: d4ta02721c-f4.tif
Fig. 4 Experimental high resolution XPS of (a) the Ti 2p, (b) C 1s and (c) N 1s levels of the Fe(SA)-Ti3C2(NH)x sample, with individual components deconvoluted.

The Fe K-edge of XANES spectra recorded for all the FeSA-Ti3C2Tx samples (se Fig. 5a) exhibits absorption edges positioned between those of FeO and Fe2O3, with a tendency towards the FeO position. This suggests that the average oxidation state of Fe in these samples is approximately +2. The Fourier transforms of the k2 weighted Fe K-edge EXAFS oscillations (see Fig. 5b) display a prominent peak around 1.5 Å in R space, which can be attributed to Fe–O coordination, as it is close to the first shell peak observed in Fe oxide reference samples. This first shell peak is more intense for Fe(SA)-Ti3C2Clx and Fe(SA)-Ti3C2(NH)x compared to Fe(SA)-Ti3C2Brx, suggesting a higher degree of local structural ordering or a higher coordination number. Peaks around 2.2 and 2.55 Å, corresponding to the Fe–Fe pair in Fe(0) foil and Fe oxides respectively, are negligible for all samples, indicating the absence of metal or metal oxide nanoparticles and the formation of atomically dispersed Fe single atoms.41,42 These results are consistent with atomically resolved STEM images, and support the successful immobilization of Fe-SA on Ti3C2Tx.


image file: d4ta02721c-f5.tif
Fig. 5 (a) XANES spectra at the Fe K-edge of Fe(SA)-Ti3C2Tx samples, and (b) Fourier transforms of the k2-weighted χ(k) of the Fe K-edge EXAFS.

Electrochemical characterization

Initial electrochemical measurements were aimed at determining the electrochemical surface areas (ECSAs) of the three electrocatalysts by cyclic voltammetry at different scan rates in the potential range corresponding to the electric double layer in alkaline (pH 13) and neutral (pH 7) media. Fig. 6a and b show the cyclic voltammetry registered for Ti3C2(NH)x MXene in alkaline and neutral media, respectively, while Figures S10a–d† show the CV for the Ti3C2Clx and Ti3C2Brx MXenes. In the range of potentials of these measurements, the materials present non-faradaic behaviour at high scan rates at both pH values, required for ECSA analysis. The capacitive current was determined at a constant potential of 0.57 V vs. RHE and at 0.71 V vs. RHE, in alkaline and neutral media, respectively.
image file: d4ta02721c-f6.tif
Fig. 6 Cyclic voltammetry plots in the 0.47–0.67 V ((a) alkaline medium) or 0.67–0.82 V ((b) neutral medium) vs. RHE window for Ti3C2(NH)x recorded at a 5 mV s−1 scan rate. Capacitance in an alkaline (c) or neutral medium (d) for the three Ti3C2Tx MXenes under study.

Fig. 6c and d show the determination of the double-layer capacitance from the capacitive currents for the three electrocatalysts. It was observed that at each pH value, Ti3C2Tx MXene with any of the two halogens, either Cl or Br, exhibits similar capacitance. However, surface functionalization with N-groups on the Ti3C2Tx MXene structure considerably increases the electrochemical surface area of the MXene. The values of the double layer capacitance (Cdl) and ECSA (Cs = 0.040 mF cm−2) calculated from eqn (3) and (4) are summarized in Table 2. For comparison, results from the determination of the ECSA in acid media are also presented in Fig. S12a–d.† Additionally, the ECSAs normalized by the catalyst loading (mF mg−1) are compared in Table 3 to other recently reported MXene-based ORR electrocatalysts. The ECSA of the halogen-terminated Ti3C2Tx is comparable to previous reports for solitary MXene samples.43,46 The remarkable increase in the ECSA observed for Fe(SA)-Ti3C2(NH)x is comparable to nanocomposites of MXene, such as MOF-derived Co–N–C/Ti3C2Tx (ref. 44) and Fe-PQD/Ti3C2Ox.46

Table 2 Electrochemical characterization data of the Ti3C2Tx (T = Cl, Br, NH) MXenes under study at two different pH values
Catalyst Eonset (V vs. RHE) e− transfer number, na Ered (V vs. RHE) Cdl (mF) ECSA (m2 g−1)
a Number of electrons measured at 0.55 or 0.3 V vs. RHE for pH 13 and 7, respectively.
pH 13
Fe(SA)-Ti3C2Cl2 0.734 3.34 0.639 0.058 20.5
Fe(SA)-Ti3C2Brx 0.740 2.88 0.625 0.068 24.1
Fe(SA)-Ti3C2(NH)x 0.751 2.51 0.631 0.150 53.1
[thin space (1/6-em)]
pH 7
Fe(SA)-Ti3C2Cl2 0.430 2.79 0.225 0.030 10.6
Fe(SA)-Ti3C2Brx 0.409 2.93 0.050 0.045 15.9
Fe(SA)-Ti3C2(NH)x 0.446 2.85 0.315 0.082 29.0

Table 3 Comparison of the 2 electron ORR in alkaline media performance of Fe(SA)-Ti3C2Tx with data reported in the literature for related MXene samples
Electrocatalyst Eonseta (V vs. RHE) e− transfer number, n H2O2 selectivity (%) ECSA, Cdl (mF mg−1) Electrolyte Ref.
a Number of electrons measured at 0.55 or 0.3 V vs. RHE for alkaline and neutral electrolytes, respectively, except for ref. 49.b Number of electrons measured at 0.15 V vs. RHE.
MXene-based electrocatalysts in alkaline electrolyte
Fe(SA)-Ti3C2Clx 0.73 3.34 33% 8.2 0.1 M KOH This work
Fe(SA)-Ti3C2Brx 0.74 2.88 56% 9.6 0.1 M KOH This work
Fe(SA)-Ti3C2(NH)x 0.75 2.51 75% 21.2 0.1 M KOH This work
Ti3C2Tx (HF etched) 0.73 2.61 70% 4.1 0.1 M KOH 43
Nb2CTx (HF etched) 0.74 2.44 78% 5.1 0.1 M KOH 43
V2CTx (HF etched) 0.72 2.18 91% 3.5 0.1 M KOH 43
Co–N–C/Ti3C2Tx 0.85 2.61 70% 15.3 0.1 M KOH 44
FePc/Ti3C2Tx 0.97 3.99 0.6% — 0.1 M KOH 24
Mo2CTx 0.68 2.25 87% — 0.1 M KOH 45
Mo2CTx:Fe 0.74 2.12 94% — 0.1 M KOH 45
Fe/PQD/Ti3C2Ox (“o-MQFe”) 0.98 3.97 2% 23.4 0.1 M KOH 46
Ti3C2Ox 0.75 3.34 33% 7.5 0.1 M KOH 46
[thin space (1/6-em)]
MXene-based electrocatalysts in neutral electrolyte
O–Mo2TiC2 0.36 2.36 82% — 0.1 M Na2SO4 47
Co–N–C/Ti3C2Tx 0.74 3.3 35% — 0.1 M PBS 44
Co–NC/Nb2CTx 0.73 2.74 63% — 0.1 M PBS 48
Ag–MnO2–Ti3C2Tx 0.34 3.7b 15% — 0.05 M PBS 49
[thin space (1/6-em)]
Other Fe single-atom electrocatalysts
3DOM Fe–N–C (900 °C) 0.96 3.98 1% ∼137 0.1 M KOH 50
Fe-CNT (O coordinated Fe) 0.822 2.19 91% — 0.1 M KOH 25
Fe1/NSOC 1.08 ∼4 0% — 0.1 M KOH 51
FeN2O2/HNC 0.79 2.26 87% 522 0.1 M KOH 52

The electrochemical characterization of the redox process of the atomically dispersed Fe-SA present on the Ti3C2Tx MXenes was performed by cyclic voltammetry at 5 mV s−1 in alkaline, neutral, and acidic media. Fig. S11† shows the corresponding voltammograms obtained for the different electrocatalysts in alkaline and neutral media. In alkaline media, the catalysts show an irreversible faradaic process detected in the first cycle that attenuates with successive cycles. The cathodic and anodic faradaic processes occur in the range of 0.6–0.7 V vs. RHE (Ered) and 0.8–1.2 V vs. RHE, respectively, and are associated with the electrochemical redox of Fe(III)/Fe(II) species present in the MXene structure as iron single-atom moieties. In neutral media, the reduction process of Fe(III)/Fe(II) takes place at lower values of 0–0.35 V vs. RHE (Fig. S11d–f†), and the clear faradaic process is also attenuated with successive scans. The difference between the potential values in alkaline and neutral media is close to 0.059 V per pH unit, indicating that the redox process is pH-dependent. The decrease of the intensity of the electrochemical response of the Fe(III)/Fe(II) redox would indicate an incomplete reversibility of the process. These electrochemical data are summarized in Table 2. In acidic media, these Fe redox processes cannot be distinguished clearly except in the case of Fe(SA)-Ti3C2Br and may be obscured by other electrochemical processes at low potentials, near 0 V vs. RHE (Fig. S12e–g†).

Electrocatalytic ORR as a function of the surface termination

The electrocatalytic activity for the ORR has also been evaluated at both alkaline and neutral pH with an RRDE. The polarization curves (Fig. 7a and d) were registered at an RRDE rotation rate of 1600 rpm and a scan rate of 5 mV s−1. To quantify the H2O2 generation during the polarization and determine the number of electrons transferred during the tests, a constant potential of 1.1 V vs. RHE was applied to the Pt ring. As can be observed in Fig. 7a and d, all the materials present catalytic activity for the ORR. It was found that the highest catalytic activity is found in alkaline media than in other media, with an onset potential about 0.3 V lower for neutral pH, and 0.7 V in acid media (Fig. S13 and Table S1†), which indicates that in acid media the activity is the lowest. For this reason, the following electrochemical tests were carried out in neutral and alkaline media. In alkaline media, the MXene samples present remarkable ORR activities with onset potentials between about 0.73 and 0.75 V vs. RHE (Table 2). The selectivity for the ORR can be estimated from the total number of electrons transferred and the H2O2 electrogenerated during the polarization tests.
image file: d4ta02721c-f7.tif
Fig. 7 (a) RRDE linear sweep voltammetry polarization curves recorded at 0.005 V s−1 for Fe(SA)-Ti3C2Tx (T = Cl, Br, NH) in O2 saturated 0.1 M KOH electrolyte, with (b) corresponding electron transfer numbers, n, and H2O2 selectivity, and (c) Tafel slopes at high and low overpotentials, and (d–f) RRDE polarization curves, electron transfer numbers, and Tafel slopes in O2 saturated 0.5 M Na2SO4 electrolyte.

As can be observed in Fig. 7b and e, the ORR selectivity depends on the media pH. Importantly, in alkaline media the selectivity was also greatly dependent on the different surface groups (Tx) of the Fe(SA)-Ti3C2Tx MXene electrocatalyst. In this medium the selectivity changes from 2.5e− with a selectivity for H2O2 production of 75% for Fe(SA)-Ti3C2(NH)x, to 3.3e− and a selectivity for H2O2 production of 33% for the Fe(SA)-Ti3C2Clx. In comparison, at neutral pH and acid pH the FeSA-Ti3C2Tx MXenes present very similar selectivity following a mixed mechanism of 3e− and 50% selectivity for H2O2 production. To elucidate the active sites in the Fe(SA)-Ti3C2Tx electrocatalysis, Fig. S11† presents the kinetic currents of the polarization curves and the CV curves of the different materials.

As can be observed in these figures, for both media the onset potential of the polarization curves matches with the reduction processes of the Fe(III)/Fe(II), strongly suggesting that the ORR process follows an inner sphere mechanism in which the oxygen reduction takes place with the electrogeneration of Fe(II) actives sites. This mechanism is the common pathway for the ORR with single-atom catalysts in a wide pH range.

Taking into consideration this relationship between the ORR onset potential and the Fe(III)/Fe(II) reduction potential, the mechanism for the ORR was studied for calculating the Tafel slopes from the corrected kinetic currents at high and low overpotentials.

Fig. 6c and f show the Tafel plots in alkaline and neutral media. A transition between the two Tafel slopes is clearly observable, from close to 0.060 V dec−1 to 0.12 V dec−1 from low to high overpotentials. This change in the Tafel slope values takes place close to the Fe(III)/Fe(II) reduction potential previously determined by CV.

This change in the Tafel slope implies a shift in the mechanism depending on the total surface concentration of Fe(II) in the catalysts (θFe(II)). Considering that the θFe(II) is potential dependent, two different mechanisms, depending on whether or not Fe(III) is reduced to Fe(II) prior to the rate determining step, can be proposed for FeSA-Ti3C2Tx, following the previously reported mechanism for other Fe-SA electrocatalysts in which Fe was coordinated to N atoms in a N-doped carbon.40,53

At low overpotentials (0.06 V dec−1)

Fe(III) + e− → Fe(II)

Fe(II) + O2 → [Fe(II)–O–O] (rate determining step)

At high overpotentials (0.120 V dec−1)

Fe(II) + O2 + e− → [Fe(II)–O–O]− (rate determining step)

In neutral media, this change occurs but is less pronounced, and the Tafel slope values indicate that there is a mixed mechanism for the ORR at high and low overpotentials.

To put the performance of Fe(SA)-Ti3C2Tx into a broader context, Table 3 provides a comparison of the electrocatalytic activity for the 2 electron ORR of the Fe(SA)-Ti3C2Tx samples under study with data reported in the literature for related MXene electrodes. While this table shows the extraordinary performance of Mo or V MXenes in comparison to Ti MXenes, the Fe(SA)-Ti3C2(NH)x reported here is the highest among the Ti MXenes reported so far. Co, Mo and Ag can also promote 2 e− ORR under neutral conditions with similar maximum selectivity values to those found here for Fe(SA)-Ti3C2(NH)x. Table 3 also shows precedents of Fe(SA) on other supports as electrocatalysts, showing that the performance of Fe(SA)-Ti3C2(NH)x is among those of the best materials, with less H2O2 selectivity than Fe(SA) on carbon nanotubes but requiring a lower onset potential.

For practical applications, electrochemically generated H2O2 was quantified during galvanostatic tests in a prototype cell, performed at 10 mA cm−2 for 90 min in neutral media (shown in Fig. S14†). For Fe(SA)-Ti3C2(NH)x an initial rate of 372 mM h−1 g−1 of electrogenerated H2O2 was determined, reaching a stationary concentration at 40 min close to 225 mM gcat−1. In these measurements, the performance of Fe(SA)-Ti3C2Brx and Fe(SA)-Ti3C2Clx was similar with a significantly lower initial rate of 262 and 182 mM h−1 g−1, respectively. All the catalysts present higher accumulation than the carbon paper used as a substrate, which indicates the excellent performance of the catalysts for the H2O2 generation and accumulation. Among the catalysts, Fe(SA)-Ti3C2(NH)x presents the best performance despite the similar selectivity towards the H2O2 production obtained in the electrocatalytic studies in neutral pH. This increase in H2O2 generation could be associated with the higher electrochemical surface area of the NH catalysts determined by CV tests. The stability of Fe(SA)-Ti3C2(NH)x was investigated by repeated galvanostatic tests lasting 90 minutes each with the same electrode. During the 5th repeat measurement, still 98.4% of the initial H2O2 generation was maintained, demonstrating the excellent stability of the Fe(SA)-Ti3C2(NH)x air cathode (Fig. S15†).

Conclusions

Preparation of Ti3C2 MXene samples having Fe single atoms and defined surface terminations has allowed their 2 electron ORR electrocatalytic pathway to be determined, showing a notable influence of the surface terminal groups on the electrocatalytic activity. The best catalytic performance was found with Fe(SA)-Ti3C2(NH)x in alkaline media, which exhibits 2.5e− ORR selectivity, and a H2O2 production of 75%, which is among the highest reported so far for Ti MXenes. As can be observed from the catalytic studies the ORR follows the inner sphere mechanism, where the onset potential of the reaction is connected to the redox potential of the material and presents a pH dependance in the studied pH range. Overall, these results show the role that surface terminations play in electrocatalysis by MXenes, offering in this way the potential for performance optimization by appropriate selection of these surface groups.

Data availability

Data will be made available upon reasonable request to the corresponding authors.

Author contributions

D. D. K. prepared and characterized the materials and performed most of the electrochemical measurements, J. J. D. made the aberration corrected high resolution STEM images, S. G. F. performed the analysis and interpretation of the X-ray absorption spectra, F. J. R. was in charge of the electrochemical part and supervised D. D. K., and F. J. R., A. P. and H. G. conceived the study. The draft was written with the contribution of all the authors. H. G. revised the final version.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support was provided by the Spanish Ministry of Science and Innovation (CEX-2021-001230-S, PID2021-126071OB-C21\ and PID2021-123431OB-I00 funded by MCIN/AEI/10.13039/501100011033) and Generalitat Valenciana (Prometeo 2021/038 and Advanced Materials programme Graphica MFA/2022/023 with funding from European Union NextGenerationEU PRTR-C17.I1). The European Commission (ERC Adv., Grant Discovery) is also acknowledged. We thank the Advanced Microscopy Center of the University of Cadiz for granting instrument time.

References

  1. X. Bai and J. Guan, Small Struct., 2023, 4, 2200354 CrossRef CAS .
  2. Y. Cui, Z. Cao, Y. Zhang, H. Chen, J. Gu, Z. Du, Y. Shi, B. Li and S. Yang, Small Sci., 2021, 1, 2100017 CrossRef CAS .
  3. X. Wu, Y. Wang and Z. S. Wu, iScience, 2024, 27, 108906 CrossRef CAS PubMed .
  4. A. Liu, X. Liang, X. Ren, W. Guan, M. Gao, Y. Yang, Q. Yang, L. Gao, Y. Li and T. Ma, Adv. Funct. Mater., 2020, 30, 2003437 CrossRef CAS .
  5. X. Li, Z. Huang, C. E. Shuck, G. Liang, Y. Gogotsi and C. Zhi, Nat. Rev. Chem, 2022, 6, 389–404 CrossRef PubMed .
  6. H. Wang and J.-M. Lee, J. Mater. Chem. A, 2020, 8, 10604–10624 RSC .
  7. V. Kamysbayev, A. S. Filatov, H. Hu, X. Rui, F. Lagunas, D. Wang, R. F. Klie and D. V. Talapin, Science, 2020, 369, 979–983 CrossRef CAS PubMed .
  8. R. Khanal and S. Irle, J. Chem. Phys., 2023, 158, 194701 CrossRef CAS PubMed .
  9. C. Qiao, H. Wu, X. Xu, Z. Guan and W. Ou-Yang, Adv. Mater. Interfaces, 2021, 8, 2100903 CrossRef CAS .
  10. J. Zhang, Y. Zhao, X. Guo, C. Chen, C.-L. Dong, R.-S. Liu, C.-P. Han, Y. Li, Y. Gogotsi and G. Wang, Nat. Catal., 2018, 1, 985–992 CrossRef CAS .
  11. D. Zhao, Z. Chen, W. Yang, S. Liu, X. Zhang, Y. Yu, W. C. Cheong, L. Zheng, F. Ren, G. Ying, X. Cao, D. Wang, Q. Peng, G. Wang and C. Chen, J. Am. Chem. Soc., 2019, 141, 4086–4093 CrossRef CAS PubMed .
  12. P. Yang, Y. Long, W. Huang and D. Liu, Appl. Catal., B, 2023, 324, 122245 CrossRef CAS .
  13. X. Sang, Y. Xie, M. W. Lin, M. Alhabeb, K. L. Van Aken, Y. Gogotsi, P. R. C. Kent, K. Xiao and R. R. Unocic, ACS Nano, 2016, 10, 9193–9200 CrossRef CAS PubMed .
  14. R. Ibragimova, P. Rinke and H.-P. Komsa, Chem. Mater., 2022, 34, 2896–2906 CrossRef CAS .
  15. D. Zhang, S. Wang, R. Hu, J. Gu, Y. Cui, B. Li, W. Chen, C. Liu, J. Shang and S. Yang, Adv. Funct. Mater., 2020, 30, 2002471 CrossRef CAS .
  16. H. Liu, Z. Hu, Q. Liu, P. Sun, Y. Wang, S. Chou, Z. Hu and Z. Zhang, J. Mater. Chem. A, 2020, 8, 24710–24717 RSC .
  17. Q. Zhao, C. Zhang, R. Hu, Z. Du, J. Gu, Y. Cui, X. Chen, W. Xu, Z. Cheng, S. Li, B. Li, Y. Liu, W. Chen, C. Liu, J. Shang, L. Song and S. Yang, ACS Nano, 2021, 15, 4927–4936 CrossRef CAS PubMed .
  18. H. Song, R. Du, Y. Wang, D. Zu, R. Zhou, Y. Cai, F. Wang, Z. Li, Y. Shen and C. Li, Appl. Catal., B, 2021, 286, 119898 CrossRef CAS .
  19. S. Yang, A. Verdaguer-Casadevall, L. Arnarson, L. Silvioli, V. Čolić, R. Frydendal, J. Rossmeisl, I. Chorkendorff and I. E. L. Stephens, ACS Catal., 2018, 8, 4064–4081 CrossRef CAS .
  20. Y. Jiang, P. Ni, C. Chen, Y. Lu, P. Yang, B. Kong, A. Fisher and X. Wang, Adv. Energy Mater., 2018, 8, 1801909 CrossRef .
  21. S. G. Peera, C. Liu, A. K. Sahu, M. Selvaraj, M. C. Rao, T. G. Lee, R. Koutavarapu, J. Shim and L. Singh, Adv. Mater. Interfaces, 2021, 8, 2100975 CrossRef CAS .
  22. X. Peng, S. Zhao, Y. Mi, L. Han, X. Liu, D. Qi, J. Sun, Y. Liu, H. Bao, L. Zhuo, H. L. Xin, J. Luo and X. Sun, Small, 2020, 16, e2002888 CrossRef PubMed .
  23. A. A. Gewirth, J. A. Varnell and A. M. DiAscro, Chem. Rev., 2018, 118, 2313–2339 CrossRef CAS PubMed .
  24. Z. Li, Z. Zhuang, F. Lv, H. Zhu, L. Zhou, M. Luo, J. Zhu, Z. Lang, S. Feng, W. Chen, L. Mai and S. Guo, Adv. Mater., 2018, 30, e1803220 CrossRef PubMed .
  25. K. Jiang, S. Back, A. J. Akey, C. Xia, Y. Hu, W. Liang, D. Schaak, E. Stavitski, J. K. Norskov, S. Siahrostami and H. Wang, Nat. Commun., 2019, 10, 3997 CrossRef PubMed .
  26. P. Zhu, P. Song, W. Feng, D. Zhao, T. Liu, J. Zhang and C. Chen, J. Mater. Chem. A, 2022, 10, 17948–17967 RSC .
  27. Y. Li, H. Shao, Z. Lin, J. Lu, L. Liu, B. Duployer, P. O. A. Persson, P. Eklund, L. Hultman, M. Li, K. Chen, X. H. Zha, S. Du, P. Rozier, Z. Chai, E. Raymundo-Pinero, P. L. Taberna, P. Simon and Q. Huang, Nat. Mater., 2020, 19, 894–899 CrossRef CAS PubMed .
  28. J. Sangster and A. D. Pelton, J. Phys. Chem. Ref. Data, 1987, 16, 509–561 CrossRef CAS .
  29. N. C. S. Selvam, J. Lee, G. H. Choi, M. J. Oh, S. Xu, B. Lim and P. J. Yoo, J. Mater. Chem. A, 2019, 7, 27383–27393 RSC .
  30. M. Li, X. Li, G. Qin, K. Luo, J. Lu, Y. Li, G. Liang, Z. Huang, J. Zhou, L. Hultman, P. Eklund, P. O. A. Persson, S. Du, Z. Chai, C. Zhi and Q. Huang, ACS Nano, 2021, 15, 1077–1085 CrossRef CAS PubMed .
  31. J. Lu, I. Persson, H. Lind, J. Palisaitis, M. Li, Y. Li, K. Chen, J. Zhou, S. Du, Z. Chai, Z. Huang, L. Hultman, P. Eklund, J. Rosen, Q. Huang and P. O. A. Persson, Nanoscale Adv., 2019, 1, 3680–3685 RSC .
  32. M. Li, J. Lu, K. Luo, Y. Li, K. Chang, K. Chen, J. Zhou, J. Rosen, L. Hultman and P. Eklund, J. Am. Chem. Soc., 2019, 141, 4730–4737 CrossRef CAS PubMed .
  33. J. A. Bearden, Rev. Mod. Phys., 1967, 39, 78–124 CrossRef CAS .
  34. Q. Peng, J. Guo, Q. Zhang, J. Xiang, B. Liu, A. Zhou, R. Liu and Y. Tian, J. Am. Chem. Soc., 2014, 136, 4113–4116 CrossRef CAS PubMed .
  35. T. Hu, J. Wang, H. Zhang, Z. Li, M. Hu and X. Wang, Phys. Chem. Chem. Phys., 2015, 17, 9997–10003 RSC .
  36. K. Nakamoto, in Part B: Applications in Coordination, Organometallic, and Bioinorganic Chemistry, John Wiley & Sons, Inc., 6th edn, 2008, pp. 1–331 Search PubMed .
  37. L. Liu, H. Zschiesche, M. Antonietti, B. Daffos, N. V. Tarakina, M. Gibilaro, P. Chamelot, L. Massot, B. Duployer, P. L. Taberna and P. Simon, Adv. Energy Mater., 2022, 13, 2202709 CrossRef .
  38. Y. Yoon, A. P. Tiwari, M. Lee, M. Choi, W. Song, J. Im, T. Zyung, H.-K. Jung, S. S. Lee, S. Jeon and K.-S. An, J. Mater. Chem. A, 2018, 6, 20869–20877 RSC .
  39. Y. Wen, T. E. Rufford, X. Chen, N. Li, M. Lyu, L. Dai and L. Wang, Nano Energy, 2017, 38, 368–376 CrossRef CAS .
  40. C. Zúñiga, C. Candia-Onfray, R. Venegas, K. Muñoz, J. Urra, M. Sánchez-Arenillas, J. F. Marco, J. H. Zagal and F. J. Recio, Electrochem. Commun., 2019, 102, 78–82 CrossRef .
  41. Y. Ren, F. Tian, L. Jin, Y. Wang, J. Yang, S. You and Y. Liu, Environ. Sci. Technol., 2023, 57, 10458–10466 CrossRef CAS PubMed .
  42. L. Jin, S. You, N. Ren, B. Ding and Y. Liu, Environ. Sci. Technol., 2022, 56, 11750–11759 CrossRef CAS PubMed .
  43. X. Huang, M. Song, J. Zhang, J. Zhang, W. Liu, C. Zhang, W. Zhang and D. Wang, Nano Res., 2022, 15, 3927–3932 CrossRef CAS .
  44. X. Huang, W. Liu, J. Zhang, M. Song, C. Zhang, J. Li, J. Zhang and D. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 11350–11358 CrossRef CAS PubMed .
  45. D. A. Kuznetsov, Z. Chen, P. M. Abdala, O. V. Safonova, A. Fedorov and C. R. Muller, J. Am. Chem. Soc., 2021, 143, 5771–5778 CrossRef CAS PubMed .
  46. Y. Liu, X. Liu, Z. Lv, R. Liu, L. Li, J. Wang, W. Yang, X. Jiang, X. Feng and B. Wang, Angew Chem. Int. Ed. Engl., 2022, 61, e202117617 CrossRef CAS PubMed .
  47. G. Li, B. Zhou, P. Wang, M. He, Z. Fang, X. Yuan, W. Wang, X. Sun and Z. Li, Catalysts, 2022, 12, 850 CrossRef CAS .
  48. X. Huang, W. Zhang, W. Liu, J. Zhang, M. Song, C. Zhang, J. Zhang and D. Wang, Appl. Catal., B, 2022, 317, 121737 CrossRef CAS .
  49. X. Wang, X. Gong, L. Chen, S. Li, J. Xie and Y. Liu, Catal. Sci. Technol., 2021, 11, 4823–4830 RSC .
  50. X. Zhang, X. Han, Z. Jiang, J. Xu, L. Chen, Y. Xue, A. Nie, Z. Xie, Q. Kuang and L. Zheng, Nano Energy, 2020, 71, 104547 CrossRef CAS .
  51. Y. Zhao, Z. Shen, J. Huo, X. Cao, P. Ou, J. Qu, X. Nie, J. Zhang, M. Wu, G. Wang and H. Liu, Angew Chem. Int. Ed. Engl., 2023, 62, e202308349 CrossRef CAS PubMed .
  52. Y. Wu, Y. Ding, X. Han, B. Li, Y. Wang, S. Dong, Q. Li, S. Dou, J. Sun and J. Sun, Appl. Catal., B, 2022, 315, 121578 CrossRef CAS .
  53. R. Venegas, K. Muñoz-Becerra, C. Candia-Onfray, J. F. Marco, J. H. Zagal and F. J. Recio, Electrochim. Acta, 2020, 332, 135340 CrossRef CAS .

Footnote

† Electronic supplementary information (ESI) available: Additional electron microscopy images and associated EDX mapping and spectra, XPS, ATR-FTIR, and electrochemical characterisations and ORR performance data of the related samples. See DOI: https://doi.org/10.1039/d4ta02721c
This journal is © The Royal Society of Chemistry 2024