Hybrid PEDOT/MnO_x nanostructured electrocatalysts for oxygen reduction

Abstract

A series of hybrid poly(3,4-ethylenedioxythiophene)/manganese oxide (PEDOT/MnO_x) thin films have been prepared via a stepwise approach: electrodeposition of PEDOT, followed by formation of MnO_x particles by a spontaneous redox reaction between PEDOT and KMnO₄. Electrocatalytic characterization of the PEDOT/MnO_x thin films demonstrates high activity toward the oxygen reduction reaction (ORR), with a shift in intrinsic ORR onset and half-wave potentials by ca. 0.2 V to lower overpotential relative to the PEDOT thin film. The most active PEDOT/MnO_x thin film electrocatalyst, P-MnO_x-20, demonstrates superior activity relative to the commercial 20% Pt/C catalyst in the half-wave region of the ORR potential window at equal mass loading, with a half-wave potential of 0.83 V (20% Pt/C, 0.81 V) and charge transfer resistance of 479 Ω (20% Pt/C, 862 Ω). The P-MnO_x-20 film also demonstrates preference to a pseudo-four electron ORR pathway (n = 3.8) and high specific ORR activity, when considered on both a total mass (−96 mA mg_total⁻¹; 20% Pt/C: −108 mA mg_total⁻¹) and metal (or metal oxide) mass basis (−296 mA mg_MnOx⁻¹; 20% Pt/C: −540 mA mg_Pt⁻¹). The P-MnO_x-20 film has been identified as the most active PEDOT/ceramic composite electrocatalyst reported to date, which is rationalized by the high surface concentration of Mn^(III), strong electronic coupling between PEDOT and MnO_x, as well as a high active site density and efficiency achieved by the stepwise electrodeposition-redox approach.

---

Introduction

Electrochemical energy conversion and storage devices with high specific energy and efficiency are needed to stabilize an increasingly renewable electricity grid and enable electromobility (e.g., electric vehicles).¹ Among these, fuel cells and metal–air batteries are promising candidates for the future of power generation and energy storage, respectively. The performance of both of these classes of devices, however, is limited by the sluggish kinetics of the oxygen reduction reaction (ORR) at the air cathode. Furthermore, the overpotential to drive the four electron/four proton transfer to surface oxygen species is in excess of 0.2 V, a significant source of efficiency loss in these devices.^2,3 Platinum is one of the most active surfaces for facilitating the ORR at the solid–liquid–gas interface, and supported Pt catalysts are the leading choice for electrode catalyst layers. Besides its prohibitive cost, Pt can suffer from deactivation by particle coalescence, CO poisoning and competing alcohol oxidation reactions upon fuel crossover.^4,5 In order to effectively replace Pt in these energy conversion and storage devices, new electrocatalytic materials are being pursued with a focus on raw material abundance, overall performance/stability and nanostructuring for improved active site density.^6–10

The rate determining step (RDS) of the ORR in both acidic and alkaline environments is acknowledged as the electron transfer to molecular oxygen, or equivalently, the displacement of the surface hydroxide.^6,11 At low pH, the electron is transferred to chemisorbed O₂ in the Inner Helmholtz Plane (IHP). The competing Outer Helmholtz Plane (OHP) electron transfer to solvated O₂ tends to dominate at high pH, due to (1) stable surface OH coverage, and (2) reduction in overpotential for superoxide radical formation in the OHP.^11–13 Although this phenomenon translates to lower alkaline ORR overpotential and relative independence to the catalyst surface thermodynamics (“nonspecificity”^11–13), the OHP electron transfer favors the undesirable 2 e⁻ ORR pathway to form HO₂⁻. Active sites that take advantage of this alkaline RDS “nonspecificity” while maintaining 4 e⁻ pathway selectivity are required to realize the stability benefits of operating in alkaline conditions. Catalyst materials from abundant sources with competitive activity against Pt have been discovered, returning interest to the alkaline fuel cell.¹³ Briefly, transition metal–nitrogen–carbon (M–N–C) materials (e.g., FeN₄/C¹² and NPC-Co45¹⁴), metal oxides (e.g., LaNiO₃ and LaMnO₃,¹⁵ Ni_xCo_3−xO₄,^8,16 MnO₂ and Mn₂O₃^6,9,17–19), heteroatom-doped carbon (e.g., BNC-2²⁰ and S-graphene-1050²¹), and conductive polymers (e.g., PEDOT and PANI^22–24) have all demonstrated high ORR activity in alkaline environments.¹³

The high conductivity, stability, and electroactivity of intrinsically conductive polymers (ICPs) makes them compelling candidates for functional electrochemical and photo(electro)chemical materials. The conjugated backbone of the ICP not only provides rapid electron transfer as a purely conductive support, but results in a lower bandgap relative to conventional polymers, suggesting the ability to tune electronics for electron transfer (electrocatalysis) and light absorption (photocatalysis/photosensing).^23,24 The ORR activity of poly(3,4-ethylenedioxythiophene) (PEDOT) was considered first by Khomenko et al. on a chemically polymerized PEDOT electrode (with PTFE binder).²³ The authors rationalized the lack of observed ORR activity for PEDOT (relative to polythiophene and polymethylthiophene) by evidence that PEDOT is fully un-doped at the highly reducing ORR potentials. In contrast, the discovery of Pt-comparable ORR activity from a vapor polymerized PEDOT air electrode by Winther-Jensen et al. prompted renewed interest in PEDOT as an electrocatalytic material.²² The vapor polymerized PEDOT film reached a stable oxidation state only in the presence of air and remained more conductive by orders of magnitude relative to the film in the absence of air. It was later determined that the vapor polymerization method provides an inherent advantage for the ORR pathway on the PEDOT surface, observing a transition from 2 e⁻ to 4 e⁻ pathway at −0.45 V vs. Ag/AgCl²⁵ (ca. 0.54 V vs. RHE), while electropolymerized PEDOT operated by a 2 e⁻ pathway across the full ORR potential range.^25,26

Jiang and co-workers have reported a series of PEDOT based catalysts: self-assembled PEDOT structures with ORR activity from pH 1–13,²⁷ PEDOT doped with hemin for the incorporation of the active FeN₄ center,²⁸ and related S-, N-, Fe-doped C foams via pyrolysis of PEDOT:PSS for alkaline ORR.²⁹ PEDOT:PSS anchored on rGO has also demonstrated enhanced ORR activity, proposed to originate from a conformational change of PEDOT and electronic interactions with rGO.³⁰ Most relevant to the study reported here, PEDOT/ceramic composites have been reported with ORR activity: CoMn₂O₄–PEDOT³¹ and PEDOT/MnO₂.³² PEDOT/MnO₂/10AA carbon paper air electrodes were prepared by hydrothermal synthesis of α- and β-MnO₂, followed by the direct polymerization of EDOT on the MnO₂via chemical vapor deposition (CVD).³² The nanostructure of the hybrid electrode was not characterized, but hybridization resulted in an increase in ORR current (at −0.7 V vs. SCE) of 15.53 mA cm⁻² for the more active phase α-MnO₂/10AA to 30.91 mA cm⁻² for PEDOT/α-MnO₂/10AA.³² A composite CoMn₂O₄–PEDOT material was prepared by conversion of α-MnO₂ nanorods to spinel CoMn₂O₄ nanoparticles, followed by chemical polymerization of EDOT in the presence of colloidal CoMn₂O₄ nanoparticles.³¹ The CoMn₂O₄–PEDOT electrocatalyst exhibits a preference for the 4 e⁻ ORR pathway and half-wave potential only 50 mV more negative than the commercial Pt/C catalyst.³¹

Despite these developments in PEDOT-based electrocatalysts, none of these materials/composites have approached Pt-like ORR onset potential or specific mass activity;⁷ however, we recently reported on the co-electrodeposition of a MnO_x/PEDOT composite thin film with ORR activity rivaling that of commercial 20% Pt/C.¹⁰ The unprecedented synergism between MnO_x and PEDOT resulted in a >0.2 V shift in intrinsic ORR potential to lower overpotential while electrochemical impedance spectroscopy (EIS) indicated a low ORR charge transfer resistance (R_CT). Thus, the co-electrodeposition approach was proposed to afford intimate contact and increased interfacial area between PEDOT and MnO_x, providing facile electron transfer and enhancing the electronic coupling effect responsible for the lower ORR overpotential.

Here, we report on a series of hybrid PEDOT/MnO_x thin film electrocatalysts derived from a stepwise process: PEDOT is electrodeposited from an organic solvent, and then decorated with MnO_x particles via spontaneous redox with KMnO₄ on the film surface. We provide chemical and electrochemical characterization of the hybrid PEDOT/MnO_x films using scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), cyclic voltammetry (CV) and a quartz crystal microbalance (QCM). The electrocatalytic properties of the films for ORR in alkaline electrolyte have also been determined using a rotating disk electrode (RDE) with CV, linear scanning voltammetry (LSV), and EIS methods, and compared to commercial benchmark 20% Pt/C. Finally, the present work is compared to results on PEDOT–ceramic hybrid electrocatalysts in order to analyze property-activity relations to serve as a guide for future catalyst design in composite polymer–ceramic systems.

Experimental

Materials

Acetonitrile (99.8%, anhydrous), LiClO₄ (>95%, ACS grade), KMnO₄ (99+%, ACS grade), Nafion solution (5 wt% in alcohol) and 3,4-ethylenedioxythiophene (EDOT) were obtained from Sigma-Aldrich. Commercial benchmark electrocatalyst powder 20% Pt/C was obtained from E-Tek. All chemicals and materials were used as-received without purification or modification.

Preparation of PEDOT and PEDOT/MnO_x thin films

An EDOT acetonitrile solution containing 0.05 M EDOT monomer and 0.1 M LiClO₄ was galvanostatically electropolymerized at a current density of 1 mA cm⁻². The three-electrode cell consisted of a Pt foil counter electrode, Ag/Ag⁺ reference electrode and glassy carbon (GC) working electrode. The PEDOT film was deposited on the GC surface of two types of working electrodes: planar, silica-supported GC electrodes (A = 0.1963 cm²) were utilized for characterization, while RDE GC working electrodes (A = 0.0707 cm², Bioanalytical Systems, Inc.) were utilized for electrochemical studies. Films were deposited for 20 s at 1 mA cm⁻². The working electrode was then removed from the acetonitrile solution and allowed to soak in DI H₂O to remove excess solvent, monomer and electrolyte from the film. The resulting blue PEDOT film was either used as-prepared (denoted PEDOT), or proceeded to soak in a 0.01 M KMnO₄ solution to achieve the formation of MnO_x on the film surface: a static soak in the KMnO₄ solution for 10, 20 or 30 min resulted in hybrid PEDOT/MnO_x films, denoted P-MnO_x-10, P-MnO_x-20, and P-MnO_x-30, respectively. The hybrid films were then allowed to soak in DI H₂O for 30 min to remove excess KMnO₄ and dry under ambient conditions. No further preparation (e.g., binder application) was necessary before electrochemical investigation of the thin catalyst films.

Results and discussion

Hybrid PEDOT/MnO_x thin film electrocatalysts: electrodeposited PEDOT and spontaneous formation of MnO_x

The preparation of hybrid PEDOT/MnO_x thin film electrocatalysts is described fully in the Experimental Section, and depicted in (Scheme 1). Briefly, EDOT was electropolymerized from acetonitrile solvent containing 0.1 M LiClO₄, at a constant current density of 1 mA cm⁻² for 20 s (Scheme 1, 1).³³ The PEDOT film was grown directly on a GC working electrode designed for the RDE cell. This PEDOT film was immersed in a bath of 0.01 M KMnO₄ for varying times (10, 20, 30 min) to induce MnO_x particle formation via spontaneous reduction of Mn^(VII) (Scheme 1, 2).^34–36 A SEM image and the corresponding energy-dispersive spectroscopy (EDS) element maps of the as-prepared P-MnO_x-20 film are shown in Fig. 1a. C, O, and S appear to be equally abundant across the surface. Mn is present across the whole surface (consistent with nanoparticle growth),³⁶ but also apparently concentrated on the PEDOT outgrowths of the film. Fig. 1a and b, as well as cross-sectional analysis in Fig. S1 (ESI†), show the irregular nature of the film growth. The bulk of the film ranges in thickness from 15–70 nm, with the larger extending features reaching as much as ca. 370 nm from the surface of the GC electrode. The irregular film is rationalized by the mechanism proposed by Aaron et al., where PEDOT chains are expected to polymerize and grow parallel to the electrode surface in organic solvent, and are thus less likely to form the close-packed, uniform thickness films that are characteristic of the micellar electropolymerization.^10,33

Scheme 1 (1) Electropolymerization of EDOT (acetonitrile, 0.1 M LiClO₄, 0.05 M EDOT, 1 mA cm⁻²); (2) spontaneous formation of MnO_x particles on the PEDOT surface (0.01 M KMnO₄, 10–30 min).

Fig. 1 (a) Low-magnification SEM image and corresponding EDS C, O, S and Mn elemental maps of a P-MnO_x-20 film; (b) high-magnification SEM image of a P-MnO_x-20 film; (c) FTIR spectra of PEDOT/MnO_x hybrid thin films after PEDOT background subtraction (arrow denotes 1044 cm⁻¹ S(O) stretching mode of interest).

KMnO₄ is a common reagent known to react with alkenes, and has been successful in preparing MnO_x particles in situ on conductive and high surface area carbons, such as acetylene black and carbon nanotubes.^34,35 Similarly, Liu et al. recently adapted this methodology for conductive polymers, reporting redox-induced formation of MnO₂ on PEDOT nanowire templates for supercapacitor arrays using KMnO₄.³⁶ Following MnO₄⁻ exposure, the authors showed that there is no measureable oxidative damage to the carbon backbone of PEDOT, resolved a small chemical shift in the S 2p binding energy via XPS, and showed the appearance of a S(O) stretching mode in the FTIR spectra. Thus, the authors’ suggested mechanism for the spontaneous growth of MnO_x particles on a PEDOT surface is the oxidation of the thiophene sulfur atom to sulfoxide and/or sulfone, as Mn^(VII) is reduced to Mn^(IV).³⁶ This hypothesis was further supported by DFT studies showing significantly favored orientation and charge transfer between the EDOT sulfur atom and the Mn atom in two phases of crystalline MnO₂.³² FTIR spectra of the PEDOT/MnO_x films prepared here are shown in Fig. 1c, and the presence of surface sulfoxide species via increased absorbance at the characteristic frequency of 1044 cm⁻¹ (denoted by an arrow) after exposure to MnO₄⁻ (and PEDOT background subtraction)^36,37 is consistent with the previously suggested growth mechanism.³⁶

QCM measurements were used to understand the dynamics of the electrodeposition and to quantify the mass of PEDOT and subsequent MnO_x on the PEDOT surface. Fig. 2a shows the mass change (calculated by the Sauerbrey equation) on a Ti/Pt crystal working electrode during the electropolymerization of PEDOT. The average mass loading of the PEDOT film after 20 s of electropolymerization is 20.8 ± 1.8 μg cm⁻². Next, a PEDOT/MnO_x film was prepared on the crystal identically to the method on a GC electrode: a 20 s PEDOT film was electropolymerized on the crystal, rinsed in DI H₂O and transferred to a 0.01 M KMnO₄ solution to monitor the uptake of MnO_x by formation directly on the PEDOT film (Fig. 2b). 10, 20, and 30 min were chosen as the KMnO₄ immersion times for the PEDOT films to assess the effects of changing MnO_x formation time on film properties. Average MnO_x mass loading of the hybrid films are 7.62 ± 1.28 μg cm⁻², 9.94 ± 1.50 μg cm⁻² and 11.51 ± 1.70 μg cm⁻² for P-MnO_x-10, P-MnO_x-20 and P-MnO_x-30, respectively. Thus, the average total mass loading for P-MnO_x-10, P-MnO_x-20 and P-MnO_x-30 are 28.28 ± 3.13 μg cm⁻², 30.6 ± 3.35 μg cm⁻² and 32.17 ± 3.55 μg cm⁻², respectively.

Fig. 2 (a) Electrode mass change during a typical galvanostatic electropolymerization of EDOT on a Ti/Pt QCM crystal; (b) electrode mass change during MnO_x uptake on a 20 s (20.8 ± 1.8 μg cm⁻²) PEDOT film on the QCM crystal, when immersed in aqueous 0.01 M KMnO₄.

XPS survey spectra of the PEDOT, P-MnO_x-10, P-MnO_x-20, and P-MnO_x-30 films are presented in Fig. 3a. The spectra indicate the presence of Mn (49 eV, 3p; 82–90 eV, 3s; 640–655 eV, 2p), S (163 eV, 2p; 228 eV, 2s), C (285 eV, 1s) and O (530 eV, 1s) for all PEDOT/MnO_x films. The observation of Cl (207 eV, 2p) in the PEDOT only film suggests that the PEDOT is originally doped with ClO₄⁻ ions, which are removed (to the limit of detection) from the film upon reaction with aqueous MnO₄⁻ for as little as 10 min. Representative high resolution XPS spectra in the Mn 3s binding energy region are shown in Fig. 3b. Six measurements were taken in different locations on each film to analyze the characteristic ⁷S–⁵S multiplet peak splitting, which is strongly affected by Mn valence. This binding energy difference, ΔE(Mn 3s), can be used to determine relative concentrations of Mn^(III) and Mn^(IV) on the surface available for catalysis, according to known ΔE(Mn 3s) values for Mn^(III)₂O₃ (5.4 eV) and Mn^(IV)O₂ (4.5 eV).³⁸ Average experimental ΔE(Mn 3s) values as a function of MnO_x formation time are shown in Fig. 3c, and the value is very close among the three films: all approximately within the standard error, the ΔE(Mn 3s) values are 4.95 ± 0.03 eV (P-MnO_x-20) < 5.01 ± 0.06 eV (P-MnO_x-30) < 5.05 ± 0.02 eV (P-MnO_x-10). The values indicate a large concentration of Mn^(III) on the surface for all three films.³⁸ The electron transfer to O₂ in the RDS of ORR occurs through the Mn^(III)/Mn^(IV) couple; thus, the Mn^(III)/Mn^(IV) ratio has been identified as a key activity descriptor for manganese oxide ORR electrocatalysts by our group and others.^6,9,39 This result also shows that the spontaneous reduction of Mn^(VII) on a PEDOT film results in a significantly reduced, mixed-valent manganese oxide (Mn^(III)_xMn^(IV)_yO_z), as opposed to MnO₂ (ΔE(Mn 3s) = 4.5 eV) as suggested by Liu et al.³⁶

Fig. 3 (a) XPS survey spectra of the PEDOT and PEDOT/MnO_x thin films on glassy carbon; (b) representative high resolution XPS spectra of the Mn 3s binding energy region; (c) Mn 3s multiplet splitting energy, ΔE(Mn 3s), as a function of the MnO_x formation time for the PEDOT/MnO_x thin films.

Electrochemical properties of the PEDOT and PEDOT/MnO_x films were determined in a three-electrode cell with a rotating glassy carbon working electrode, Ag/AgCl reference electrode (reported vs. RHE), and Pt counter electrode in 0.1 M KOH electrolyte. CV scans at a rate of 50 mV s⁻¹ in the non-Faradaic potential window of 0.95–1.35 V are shown in Fig. 4a. The capacity of the electrode does not increase significantly upon formation of the PEDOT/MnO_x hybrid (as compared to the 4-fold increase in capacitive current observed upon conversion of PEDOT nanowires to MnO₂-NP/PEDOT nanowires by Liu et al.³⁶), with integral charge of 244 μC (PEDOT), 354 μC (P-MnO_x-10), 250 μC (P-MnO_x-20), and 289 μC (P-MnO_x-30) (Fig. 1a). Rigorous determination of the electrochemically active surface area (ECSA) or real surface area (RSA) requires experimentally determined double layer capacitance (C_DL), and precise values of specific or reference capacitance, respectively. Because the specific and reference capacitance values are not known for the hybrid nanostructured film of PEDOT/MnO_x, we have chosen to report the double-layer capacitance (C_DL) as an approximation of the surface area accessible to the electrolyte and thus available for catalysis. C_DL is proposed to provide a satisfactory comparison between PEDOT/MnO_x films, as the PEDOT mass is the same for all films and the MnO_x mass varies approximately within a standard deviation between the three samples. Fig. 4b shows the results of the repeated CV scans in a 0.1 V non-Faradaic potential window (all curves provided in Fig. S2, ESI†) at increasing scan rates. C_DL values calculated from the slope of the correlation between scan rate and current near the switching potential demonstrate similar surface area for all films and reinforce the trends in charge capacity (Fig. 4a): 3.9 mF cm⁻² (PEDOT) < 4.3 mF cm⁻² (P-MnO_x-20) < 4.9 mF cm⁻² (P-MnO_x-30) < 6.2 mF cm⁻² (P-MnO_x-10).

Fig. 4 (a) CV scans at 50 mV s⁻¹ in Ar-saturated 0.1 M KOH electrolyte; (b) current-scan rate plots to determine double-layer capacitance, where each point represents the near-switching potential (1.05 V vs. RHE) current of a CV curve at a discrete scan rate in Ar-saturated 0.1 M KOH electrolyte (all CV scans provided in Fig. S2, ESI†); (c) CV scans in the ORR potential window, in O₂-free electrolyte (dotted curves, Ar-saturated 0.1 M KOH) and O₂-saturated electrolyte (solid curves, O₂-saturated 0.1 M KOH).

CV scans in the ORR potential window in Ar-saturated and O₂-saturated electrolyte are provided in Fig. 4c. The scans in O₂-free electrolyte (dashed lines) show no redox activity from PEDOT and a sharp oxidation wave and weak reduction wave centered at ca. 0.77 V for the PEDOT/MnO_x films, which can be attributed to the Mn^(IV)/Mn^(III) redox transitions on the surface.³⁹ In O₂-saturated electrolyte, a strong reduction wave is evident for the PEDOT film, signaling its ORR activity, however at a high overpotential (0.57 V). The O₂ reduction wave for the PEDOT/MnO_x films is shifted by more than 0.2 V to lower overpotential relative to PEDOT. The ORR peak potentials increase (improve) on the order of: PEDOT (0.57 V) < P-MnO_x-10 (0.805 V) ∼ P-MnO_x-30 (0.805 V) < P-MnO_x-20 (0.81 V). The significant reduction in overpotential demonstrates the synergistic activity between MnO_x and PEDOT, similar to that observed for the previously reported co-electrodeposited PEDOT/MnO_x hybrid film.¹⁰ The stepwise synthetic method of MnO_x formation directly on a PEDOT film precludes the evaluation of a MnO_x control electrocatalyst on a GC electrode; however, the ORR peak potential for other amorphous MnO_x catalysts are negative relative to the PEDOT/MnO_x films,^40,41 indicating there is beneficial electronic interaction between MnO_x and PEDOT that facilitates the ORR at lower overpotential.¹⁰

Hydrodynamic LSV methods were employed to further investigate the ORR activity and determine the ORR electron transfer number (n) and kinetic ORR rate constant (k) values by Koutecky–Levich (K–L) analysis. Fig. 5a shows single LSV curves at 1600 RPM for the PEDOT, P-MnO_x-10, P-MnO_x-20 and P-MnO_x-30 films, as well as a commercial benchmark, 20% Pt supported on Vulcan XC-72 (20% Pt/C). The ORR onset potential exhibits a similar positive shift upon the hybridization of MnO_x with PEDOT, as seen using CV (Fig. 4c), from 0.68 V (PEDOT) to ca. 0.87 V (P-MnO_x-10, P-MnO_x-20, P-MnO_x-30). The onset potential of 20% Pt/C is 0.89 V. Half-wave potentials for the five catalysts increase in the order of 0.57 V (PEDOT) < 0.81 V (20% Pt/C) < 0.82 V (P-MnO_x-30) < 0.83 V (P-MnO_x-10, P-MnO_x-20). Diffusion-limiting ORR current is maximized with the P-MnO_x-20 film, achieving −2.93 mA cm⁻², vs. −1.07 mA cm⁻², −2.74 mA cm⁻² and −2.85 mA cm⁻² for PEDOT, P-MnO_x-10 and P-MnO_x-30, respectively. It is worth noting that P-MnO_x-20 demonstrates the highest diffusion-limiting current despite having the smallest charge capacity and C_DL among the PEDOT/MnO_x films (Fig. 4a and b). 20% Pt/C achieves diffusion-limiting current of −3.27 mA cm⁻². Fig. 5b shows the LSV curves of the P-MnO_x-20 and commercial 20% Pt/C catalysts normalized to specific catalyst mass, rather than the geometric area (Fig. 5a), in order to analyze the activity on an approximate ‘per-active-site’ basis. The P-MnO_x-20 film outperforms 20% Pt/C at low overpotential (>0.7 V) when considered on a total catalyst mass basis, while the 20% Pt/C is clearly superior when normalized to metal (or metal oxide) mass basis. The diffusion-limiting specific mass activity (j_s) of 20% Pt/C is −108 mA mg_total⁻¹ (−540 mA mg_Pt⁻¹), compared to −96 mA mg_total⁻¹ (−296 mA mg_MnOx⁻¹) for P-MnO_x-20.

Fig. 5 (a) Background-subtracted ORR geometric current density LSV curves at 1600 RPM in 0.1 M KOH electrolyte; (b) background-subtracted ORR specific mass activity LSV curves at 1600 RPM in 0.1 M KOH electrolyte; (c) K–L analysis, calculated n (dotted lines) and k (solid lines) metric values; (d) potentiostatic EIS Nyquist plots at the ORR half-wave potential in 0.1 M KOH electrolyte.

Because of the similar activity in the onset and half-wave regions of the LSV curves for the PEDOT/MnO_x films, K–L analysis and EIS were employed to determine the ORR pathway, kinetic constant, and charge transfer characteristics (All LSV curves are included in Fig. S3, ESI†). Fig. 5c shows the n and k values calculated from the slope and intercept, respectively, of the K–L plots, as a function of potential. Due to known issues in assumptions employed in the K–L analysis (particularly in alkaline electrolyte),⁴²n was sometimes found to exceed the theoretical limit of 4 e⁻ and is thus only considered in the range of 0.35–0.55 V. The average n values indicate a preference for the pseudo-four electron pathway for all PEDOT/MnO_x catalysts, increasing in the order of: n = 3.7 (P-MnO_x-10, P-MnO_x-30) < 3.8 (P-MnO_x-20). The pseudo-four electron pathway refers to the ORR pathway where O₂ is first reduced electrochemically by two electrons to form a HO₂⁻ intermediate, followed by either (1) two-electron electrochemical reduction of HO₂⁻, or (2) chemical disproportionation of HO₂⁻, leading to regeneration and subsequent two-electron electrochemical reduction of O₂. The competing electrochemical reduction and disproportionation of HO₂⁻ are evidenced by the high activity of Mn^(III)/Mn^(IV) oxides toward both processes.⁴³k values show similarity in the intrinsic rate for the three PEDOT/MnO_x catalysts, which could be expected from the identical slope in the kinetic-limited region of the LSV curves (Fig. 5a). Average k values show the following trend: 0.0175 cm s⁻¹ (P-MnO_x-10) > 0.0173 cm s⁻¹ (P-MnO_x-20) > 0.0162 cm s⁻¹ (P-MnO_x-30). Based on the ΔE(Mn 3s) analysis via XPS (Fig. 3c), P-MnO_x-10 could be expected to exhibit the highest ORR rate, with (possibly) the highest concentration of surface Mn^(III).^6,9 However, it is likely that the Mn^(III) surface concentration is similar between the three films, as ΔE(Mn 3s) values of both P-MnO_x-10 and P-MnO_x-20 are approximately within experimental standard deviation of P-MnO_x-30. Optimal ORR charge transfer characteristics were determined by modelling EIS spectra to the equivalent Randles circuit. Fig. 5d shows the Nyquist plot spectra of the potentiostatic EIS experiments at 0.82 V with 0.01 V RMS perturbations. The real-axis intercepts demonstrate that P-MnO_x-20 achieves the lowest ORR charge transfer resistance (R_CT) in the half-wave region: 479 Ω (P-MnO_x-20) < 545 Ω (P-MnO_x-30) < 608 Ω (P-MnO_x-10) < 862 Ω (20% Pt/C). Due to mechanical separation of the thin 30 μg cm⁻² P-MnO_x films from the electrode surface during long-term stability testing, the loading was increased to 80 μg cm⁻² to assess the stability and MeOH tolerance of P-MnO_x-20. The chronopotentiometric response is shown in Fig. S4 (ESI†) and demonstrates the stability of P-MnO_x-20 at −1 mA cm⁻² for three hours, including one hour in the presence of MeOH. The potential decreases only 15 mV over the three-hour period. The high stability rules out any near-term detrimental, irreversible MnO_x reduction or competing MeOH oxidation.

Property-activity relations of PEDOT/MnO_x electrocatalysts

The limited reports on PEDOT/ceramic composite electrocatalysts can be considered in two categories, PEDOT-modified metal oxide nanostructures^31,32 and strongly coupled PEDOT-amorphous manganese oxide hybrid nanostructures¹⁰ (including this work). The high ORR overpotential and low mass activity of the CoMn₂O₄–PEDOT³¹ and PEDOT/α-MnO₂/10AA³² composites (relative to the PEDOT/MnO_x electrocatalysts reported here and in previous work¹⁰) is likely a synthetic limitation of the ceramic–polymer interfacial area, as the hybridization does not appear to significantly reduce intrinsic ORR overpotentials (onset, half-wave) of the ceramic. In contrast, the synergistic activity of hybrids derived from co-electrodeposition¹⁰ and redox-induced formation approaches are proposed to result in a nanostructure with strong electronic coupling and improved active site density by maximizing interfacial area between PEDOT and ceramic materials.

We first reported a MnO_x/PEDOT electrocatalyst prepared by an anodic, potentiostatic co-electrodeposition of MnO_x and PEDOT from an aqueous micellar solution.¹⁰ The co-electrodeposition resulted in a uniformly distributed thin film of polymer and ceramic: uniform ca. 110 nm thickness (Fig. S5, ESI†), >50% MnO_x content suggested by QCM studies, and nanoporosity on the order of ca. 10–50 nm.¹⁰ The MnO_x/PEDOT exhibited a pseudo-four electron ORR pathway, half-wave potential 40 mV positive relative to 20% Pt/C, diffusion-limited mass activity of 40 mA mg⁻¹, and Tafel slope (39 mV dec⁻¹) lower than 20% Pt/C (Table 1).¹⁰ Here, the reported PEDOT/MnO_x electrocatalysts were prepared via anodic, galvanostatic electrodeposition of PEDOT from organic solvent, followed by formation of MnO_x particles via spontaneous reduction of aqueous MnO₄⁻ (Scheme 1). The route presented here results in a significantly different nanostructure to the co-electrodeposited films, with irregular thickness from ca. 20–370 nm, larger apparent porosity on the scale of ca. 50–200 nm, and <50% MnO_x content (Fig. S5, ESI†). The P-MnO_x-20 film showed similar high activity toward the ORR, with an n value of 3.8, half-wave potential 10 mV positive of 20% Pt/C, and R_CT and Tafel slope (41 mV dec⁻¹) lower than 20% Pt/C (Table 1 and Fig. S6, ESI†). The mass-transport corrected Tafel plots are also provided in Fig. S6 (ESI†) and confirm the low slope for P-MnO_x-20. The similar n and Tafel slope values between co-electrodeposited MnO_x/PEDOT and P-MnO_x-20 suggest the same pathway/mechanism of ORR is taking place on the two hybrid surfaces. Furthermore, the Mn^(III)/Mn^(IV) ratio is a crucial indicator of ORR activity on manganese oxide surfaces,^6,9 and the characteristic XPS Mn 3s multiplet splitting, ΔE(Mn 3s), suggest a very similar Mn^(III)/Mn^(IV) ratio on the surface of the co-electrodeposited MnO_x/PEDOT film (4.98 ± 0.06 eV) as the P-MnO_x-20 film (4.95 ± 0.03 eV) to facilitate the ORR (Fig. S5, ESI†). The diffusion-limited mass activity at 500 RPM for P-MnO_x-20 is provided in Table 1 in order to compare to the electrodeposited PEDOT/MnO_x¹⁰ catalyst that was reported at 400 RPM. The mass activity is higher for the P-MnO_x-20 than the co-electrodeposited MnO_x/PEDOT despite lower relative MnO_x content and lower mass loading. For an ORR activity comparison of P-MnO_x-20 to other PEDOT-based ORR electrocatalysts, the reader is referred to our previous articles^7,10 for activity comparison tables.

Table 1 Comparison of ORR activity metrics: PEDOT/MnO_x prepared via stepwise electrodeposited/spontaneous redox process vs. co-electrodeposition

Catalyst	Ref.	Mass loading (μg cm^−2)	Electrolyte/rotation rate	n (e^−)	E half-wave (V vs. RHE)	Diffusion-limited mass activity (mA mg^−1)	Tafel slope (mV dec^−1)	EIS RCT (Ω)
P-MnOx-20	This work	30	0.1 M KOH/500 RPM	3.8	0.83	−65.3	41	479
MnOx/PEDOT	10	40	1 M KOH/400 RPM	3.9	0.83	−40.2	39	361

---

Thus, not only is the stepwise electropolymerization/spontaneous MnO_x formation approach reported here potentially more versatile in nature, but it also results in a film with higher mass activity and catalyst (MnO_x) efficiency than the co-electrodeposited PEDOT/MnO_x. This result is proposed to originate from the MnO_x residing exclusively on the surface of the PEDOT film in the P-MnO_x-20 catalyst. In the co-electrodeposited approach, it is likely that potential MnO_x electrocatalytic sites are obscured by PEDOT or internal to the larger MnO_x platelets and not accessible. Thus, the stepwise approach of MnO_x formation on a PEDOT film by spontaneous redox has proven an effective method for improved ORR activity via increased active site density and efficiency and represents an advance in PEDOT-based alkaline electrocatalysis for the ORR.

Conclusions

Hybrid PEDOT/MnO_x thin film electrocatalysts were prepared by galvanostatic electrodeposition of PEDOT in organic solvent, followed by formation of MnO_x particles on the surface via spontaneous reduction of aqueous MnO₄⁻. High-resolution XPS studies in the Mn 3s binding energy region indicate a significant presence of Mn^(III), in addition to the expected Mn^(IV), on the surface of the electrode, with average ΔE(Mn 3s) multiplet splitting energy values of 5.05 eV (P-MnO_x-10), 4.95 eV (P-MnO_x-20), and 5.01 eV (P-MnO_x-30). The electrocatalytic activity of the PEDOT/MnO_x hybrid thin films towards the ORR was determined by CV and RDE LSV methods. The results indicate a strong electronic coupling between PEDOT and MnO_x to facilitate the ORR at low overpotential, with a positive shift of >0.2 V in CV peak potential and LSV half-wave potential from PEDOT to P-MnO_x-10, P-MnO_x-20, and P-MnO_x-30. Although all composite catalysts films show high activity at low overpotential, the P-MnO_x-20 film exhibited the highest overall ORR activity, with onset and half-wave potentials of 0.87 and 0.83 V, respectively, n value of 3.8, and half-wave R_CT of 479 Ω. Comparatively, commercial 20% Pt/C exhibited onset and half-wave ORR potentials of 0.89 and 0.81 V, respectively, n value of 3.9, and half-wave R_CT of 862 Ω. Besides a more positive half-wave potential and lower R_CT than 20% Pt/C, the P-MnO_x-20 film also demonstrated high mass activity when considered on both a total mass (−96 mA mg_total⁻¹; 20% Pt/C: −108 mA mg_total⁻¹) and metal (or metal oxide) mass basis (−296 mA mg_MnOx⁻¹; 20% Pt/C: −540 mA mg_Pt⁻¹). The reported P-MnO_x-20 thin film was demonstrated to be the most active PEDOT–ceramic hybrid electrocatalyst reported so far for alkaline ORR. The origin of activity is attributed to strong electronic coupling between PEDOT and MnO_x, high concentration of surface Mn^(III), and the increase in active site density and efficiency realized by the spontaneous redox formation approach.

Acknowledgements

Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy*s National Nuclear Security Administration under contract DE-NA0003525. Drs Michael Brumbach (XPS) and Erik Spoerke (IR) are thanked for their technical assistance and access to instrumentation.

Notes and references

1. O. Gröger, H. A. Gasteiger and J.-P. Suchsland, J. Electrochem. Soc., 2015, 162, A2605–A2622.
2. J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. R. Kitchin, T. Bligaard and H. Jónsson, J. Phys. Chem. B, 2004, 108, 17886–17892.
3. J. Larminie and A. Dicks, in Fuel Cell Systems Explained, John Wiley & Sons, Ltd, 2013, pp. 25–43 DOI:10.1002/9781118878330.ch2.
4. R. Borup, J. Meyers, B. Pivovar, Y. S. Kim, R. Mukundan, N. Garland, D. Myers, M. Wilson, F. Garzon, D. Wood, P. Zelenay, K. More, K. Stroh, T. Zawodzinski, J. Boncella, J. E. McGrath, M. Inaba, K. Miyatake, M. Hori, K. Ota, Z. Ogumi, S. Miyata, A. Nishikata, Z. Siroma, Y. Uchimoto, K. Yasuda, K.-I. Kimijima and N. Iwashita, Chem. Rev., 2007, 107, 3904–3951.
5. R. J. Stanis, T. N. Lambert and M. A. Yaklin, Energy Fuels, 2010, 24, 3125–3129.
6. D. J. Davis, T. N. Lambert, J. A. Vigil, M. A. Rodriguez, M. T. Brumbach, E. N. Coker and S. J. Limmer, J. Phys. Chem. C, 2014, 118, 17342–17350.
7. T. N. Lambert and J. A. Vigil, ECS Trans., 2016, 75, 965–970.
8. T. N. Lambert, J. A. Vigil, S. E. White, D. J. Davis, S. J. Limmer, P. D. Burton, E. N. Coker, T. E. Beechem and M. T. Brumbach, Chem. Commun., 2015, 51, 9511–9514.
9. T. N. Lambert, J. A. Vigil, S. E. White, C. J. Delker, D. J. Davis, M. Kelly, M. T. Brumbach, M. A. Rodriguez and B. S. Swartzentruber, J. Phys. Chem. C, 2017, 121, 2789–2797.
10. J. A. Vigil, T. N. Lambert and K. Eldred, ACS Appl. Mater. Interfaces, 2015, 7, 22745–22750.
11. N. Ramaswamy and S. Mukerjee, Adv. Phys. Chem., 2012, 2012, 491604.
12. N. Ramaswamy, U. Tylus, Q. Jia and S. Mukerjee, J. Am. Chem. Soc., 2013, 135, 15443–15449.
13. Q. He and E. J. Cairns, J. Electrochem. Soc., 2015, 162, F1504–F1539.
14. Z. Liu, G. Zhang, Z. Lu, X. Jin, Z. Chang and X. Sun, Nano Res., 2013, 6, 293–301.
15. J. Suntivich, H. A. Gasteiger, N. Yabuuchi, H. Nakanishi, J. B. Goodenough and Y. Shao-Horn, Nat. Chem., 2011, 3, 546–550.
16. C. Jin, F. Lu, X. Cao, Z. Yang and R. Yang, J. Mater. Chem. A, 2013, 1, 12170–12177.
17. Y. Gorlin and T. F. Jaramillo, J. Am. Chem. Soc., 2010, 132, 13612–13614.
18. T. N. Lambert, D. J. Davis, W. Lu, S. J. Limmer, P. G. Kotula, A. Thuli, M. Hungate, G. Ruan, Z. Jin and J. M. Tour, Chem. Commun., 2012, 48, 7931–7933.
19. C. Jiang, Z. Guo, Y. Zhu, H. Liu, M. Wan and L. Jiang, ChemSusChem, 2015, 8, 158–163.
20. Y. Gong, H. Fei, X. Zou, W. Zhou, S. Yang, G. Ye, Z. Liu, Z. Peng, J. Lou, R. Vajtai, B. I. Yakobson, J. M. Tour and P. M. Ajayan, Chem. Mater., 2015, 27, 1181–1186.
21. Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, X. A. Chen and S. Huang, ACS Nano, 2012, 6, 205–211.
22. B. Winther-Jensen, O. Winther-Jensen, M. Forsyth and D. R. MacFarlane, Science, 2008, 321, 671–674.
23. V. G. Khomenko, V. Z. Barsukov and A. S. Katashinskii, Electrochim. Acta, 2005, 50, 1675–1683.
24. Q. Zhou and G. Shi, J. Am. Chem. Soc., 2016, 138, 2868–2876.
25. R. Kerr, C. Pozo-Gonzalo, M. Forsyth and B. Winther-Jensen, ECS Electrochem. Lett., 2013, 2, F29–F31.
26. J. Ma, T. Xue, Z. Zou, B. Wang and M. Luo, Int. J. Electrochem. Sci., 2015, 10, 4562–4570.
27. Z. Guo, Y. Qiao, H. Liu, C. Ding, Y. Zhu, M. Wan and L. Jiang, J. Mater. Chem., 2012, 22, 17153–17158.
28. Z. Guo, H. Liu, C. Jiang, Y. Zhu, M. Wan, L. Dai and L. Jiang, Small, 2014, 10, 2087–2095.
29. Z. Guo, C. Jiang, C. Teng, G. Ren, Y. Zhu and L. Jiang, ACS Appl. Mater. Interfaces, 2014, 6, 21454–21460.
30. M. Zhang, W. Yuan, B. Yao, C. Li and G. Shi, ACS Appl. Mater. Interfaces, 2014, 6, 3587–3593.
31. A. D. Chowdhury, N. Agnihotri, P. Sen and A. De, Electrochim. Acta, 2014, 118, 81–87.
32. Y.-L. Kuo, C.-C. Wu, W.-S. Chang, C.-R. Yang and H.-L. Chou, Electrochim. Acta, 2015, 176, 1324–1331.
33. N. Sakmeche, S. Aeiyach, J.-J. Aaron, M. Jouini, J. C. Lacroix and P.-C. Lacaze, Langmuir, 1999, 15, 2566–2574.
34. S.-B. Ma, K.-Y. Ahn, E.-S. Lee, K.-H. Oh and K.-B. Kim, Carbon, 2007, 45, 375–382.
35. S.-B. Ma, Y.-H. Lee, K.-Y. Ahn, C.-M. Kim, K.-H. Oh and K.-B. Kim, J. Electrochem. Soc., 2006, 153, C27–C32.
36. R. Liu, J. Duay and S. B. Lee, ACS Nano, 2010, 4, 4299–4307.
37. W. R. Fawcett and A. A. Kloss, J. Phys. Chem., 1996, 100, 2019–2024.
38. A. J. Nelson, J. G. Reynolds and J. W. Roos, J. Vac. Sci. Technol., A, 2000, 18, 1072–1076.
39. A. S. Ryabova, F. S. Napolskiy, T. Poux, S. Y. Istomin, A. Bonnefont, D. M. Antipin, A. Y. Baranchikov, E. E. Levin, A. M. Abakumov, G. Kéranguéven, E. V. Antipov, G. A. Tsirlina and E. R. Savinova, Electrochim. Acta, 2016, 187, 161–172.
40. J. Yang and J. J. Xu, Electrochem. Commun., 2003, 5, 306–311.
41. Y. Meng, W. Song, H. Huang, Z. Ren, S.-Y. Chen and S. L. Suib, J. Am. Chem. Soc., 2014, 136, 11452–11464.
42. R. Zhou, Y. Zheng, M. Jaroniec and S.-Z. Qiao, ACS Catal., 2016, 6, 4720–4728.
43. A. S. Ryabova, A. Bonnefont, P. Zagrebin, T. Poux, R. Paria Sena, J. Hadermann, A. M. Abakumov, G. Kéranguéven, S. Y. Istomin, E. V. Antipov, G. A. Tsirlina and E. R. Savinova, ChemElectroChem, 2016, 3, 1667–1677.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7qm00147a

---