Conducting Polymer‐Based e‐Refinery for Sustainable Hydrogen Peroxide Production

Electrocatalysis enables the industrial transition to sustainable production of chemicals using abundant precursors and electricity from renewable sources. De‐centralized production of hydrogen peroxide (H2O2) from water and oxygen of air is highly desirable for daily life and industry. We report an effective electrochemical refinery (e‐refinery) for H2O2 by means of electrocatalysis‐controlled comproportionation reaction ( 2 H 2 O + O 2 → 2 H 2 O 2 ), feeding pure water and oxygen only. Mesoporous nickel (II) oxide (NiO) was used as electrocatalyst for oxygen evolution reaction (OER), producing oxygen at the anode. Conducting polymer poly(3,4‐ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) drove the oxygen reduction reaction (ORR), forming H2O2 on the cathode. The reactions were evaluated in both half‐cell and device configurations. The performance of the H2O2 e‐refinery, assembled on anion‐exchange solid electrolyte and fed with pure water, was limited by the unbalanced ionic transport. Optimization of the operation conditions allowed a conversion efficiency of 80%.


Introduction
The recent UN warning report on the global climate changes [1] is uncompromising in ascertainment of human activities with direct negative impact on climate change.The electrochemical technologies based on direct transformation of green electrical energy accessible from renewable sources to chemical energy are prioritized since it bypasses greenhouse gas emission.
Air and water are two vital and ample resources of Earth.The electrochemical interconversion of them enables the green technologies related to energy storage such as rechargeable metal-air batteries, fuel cells and water electrolysers for hydrogen production.The sluggish kinetics of oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) is causing loss in electrical energy conversion during the molecular transformations.This can be significantly reduced by scaling up electrocatalytic interfaces at industrial level.To enable this, the cost of efficient electrocatalysts must be decreased, for example, by bypassing platinum group metals (PGM).This is a major economical driving force to the mass implementation of sustainable technologies.The half-reaction of H 2 O 2 reduction: is characterized by its high equilibrium potential ( E ) of +1.763 V (vs reversible hydrogen electrode (RHE) at pH 0).The equilibrium potential of Reaction ( 1) is higher than the characteristic potentials of oxidation of a wide variety of organics implying the general oxidative aptitude of H 2 O 2 .The combination of this property with the liquid form of oxidant and the fact, that water is the only product of oxidation, defines the advancement of H 2 O 2 use as a powerful and green liquid oxidant.As an example, H 2 O 2 is widely used in the pulp and paper industry for bleaching and recycling.This application justifies 40% of the global production; other applications include food packaging disinfection, biomedical applications, organic synthesis, [2] COVID-19 virus inactivation, [3] water treatment, [4,5] and it can potentially be used as an energy carrier. [6]The annual production of H 2 O 2 exceeds 3 million tons and is based on the non-sustainable anthraquinone process, which relies on use of organic solvents, hydrogen gas, significant energy input as well as PGM-based catalysts.Such demanding production, developed by BASF (1935-1945)  and commercialized by Dupont (1950s), requires a centralized infrastructure, which implies economical loses on distribution and chemical safety due to the high concentration of the produced H 2 O 2 solution (70 wt.%).
[9] The rapid decrease in price for electricity from renewable sources in the last decade makes e-refinery economically viable.E-refinery for on-site H 2 O 2 production using only abundant oxygen, water and the electricity relies on the two-electron ORR to H 2 O 2 (ORR-to-H 2 O 2 ) (E eq = +0.695V (RHE)), which can be written for acidic media: and basic media: The achievement of the first In contrast to Reactions (2) and (3) that are characterized by an electron to proton ratio of This implies that the equilibrium potential (V (RHE)) of Reaction ( 4) is given by the following equation: a1 .The instability of H 2 O 2 in the conditions of reductive e-refinery is due to three vents of different origins.Vent A (thermodynamic) is due to the downhill character of the equilibrium potentials for H 2 O 2 production (Reactions 2-4) and possible reactive losses (via Reaction 1 and via the reaction O þ 4 H þ 4 e ! 2 H O (E eq = +1.223V (RHE))) at any pH.This implies that the electrochemical production via ORR-to-H 2 O 2 is feasible only at the conditions of kinetic control.Vent B (thermodynamic) is due to the two-stage mechanism of ORR-to-H 2 O 2 (Reactions 2-4) based on two independent transfers of energetically unequal electrons in the absence of heterogeneous catalysts.Here, the initial step of ORR is the first electron transfer: with the equilibrium potential of reaction ( 5) with the equilibrium potential of reaction ( 6) The difference in the equilibrium potentials for ORR-to-H 2 O 2 (Reactions 2-4) and for the initial step of ORR (Reactions 5 and 6), the so-called thermodynamic overvoltage (Δ E ), defines the thermodynamic downhill of the driving force inherent for sequential ORR-to-H 2 O 2 .The chemisorption of oxygen on PGM-based electrocatalysts enables the mitigation of vent B in acidic media with possible aggravation of vent A. Alternatively, Δ E decreases as pH increases at pH ≥ pK H2O2 a1 : This motivates the use of less expensive general electrocatalytic interfaces in combination with basic media for the mitigation of the vent B. [10] Vent C (chemical) is exergonic (electrochemistry-free) disproportionation of H 2 O 2 to oxygen and water: which happens fast on metal ion contaminants.Therefore, the realistic strategy towards efficient H 2 O 2 e-refinery is the combination of an alkaline media, which can be reconstructed by anion (hydroxide) exchanging membrane, with a metal-free electrocatalytic interface assuring ORR-to-H 2 O 2 .Decisively, the alkaline media enables the use of inexpensive anodes sustaining oxygen evolution reaction (OER) [11][12][13] as an auxiliary electrode system essential for realistic H 2 O 2 e-refinery.[16] Introducing nanoporosity enlarged the available surface area of these electrocatalysts and led to the decrease in overpotential.However, the pore size must be optimized to allow efficient diffusion of reagents. [16]RR-to-H 2 O 2 electrocatalysts are most often based on precious and scarce PGM nanoparticles, which prevent them from large-scale deployment. [17,18]Mercury-based ORR-to-H 2 O 2 electrocatalysts are limited by the potential release of toxic species into the water-based environment when used in harsh operational conditions. [19,20]][23] It is today not fully understood how this class of electrocatalysts operates as electrocatalytic data is not easily reproducible.This stems likely from their sensitivity to the complex surface states of carbons.Consequently, the structure-performance relationships of carbon-based electrocatalysts are not fully elucidated and thus difficult to further optimize. [24]onducting polymers constitute another class of electrocatalysts.They are organic molecular systems with high electronic conductivity and broad metal-free electrocatalytic capabilities including ORR-to-H 2 O 2.
[ [25][26][27][28][29][30] The key difference between the molecular systems, such as conducting polymers, and atomic crystals, [31][32][33][34] such as graphitic materials and metals, is the absence or the presence of the unsaturated valences on the surface respectively.This results in a ground-level distinction of electrocatalytic reactions on conducting polymers manifested by the absence of chemosorbed species on surface.The absence of surface adsorbates during the reduction of triplet di-radical molecule, namely molecular oxygen, on conducting polymers, [26] implies the non-Norskovian type of the electrode reaction in the bulk, which can be theorized in a frame of Lavironian formalism of proton-coupled electron transfers.Poly(3,4 ethylenedioxythiophene) composited with poly(styrenesulfonate) (PEDOT:PSS) is one of the most successful conducting polymers and it stands out owing to the near-unity efficiency for affordable H 2 O 2 generation. [27]In contrast to transition metals and carbon-based catalysts for which the reactivity with dioxygen depends on spin-selectivity rules and electronic structure matching to define an intermediate adsorption, while the conducting polymers drive the electrocatalysis via outer-sphere electron transfer, [26] which positions them out from conventional catalysts.
Here, we report on the development of cost-effective de-centralized H 2 O 2 e-refinery based on pure water and oxygen only.We propose the unique combination of both PEDOT:PSS and nickel (II) oxide as electrocatalysts for ORR-to-H 2 O 2 and OER, respectively, in alkaline environment.Both catalysts showed, in half-cell configuration, mass normalized activities comparable with PGM-based electrocatalysts.The Energy Environ.Mater.2024, 7, e12551 operation of a membrane electrolyser (Scheme 1) in the H 2 O 2 erefinery assembly with ORR cathode and OER anode, fed only with oxygen and pure water, was limited by the ionic transport.The device performance can be significantly improved by using an external feed of electrolyte yielding an alkaline solution of H 2 O 2 , an important product for pulp and paper industry with an efficiency up to 80%.

ORR
The advantage of operating in alkali media is the use of efficient and low-cost catalysts based on transition-metal oxides/hydroxides.The potential challenge is the loss of H 2 O 2 via disproportionation (Reaction 8).This loss is known to be significant in presence of metal ions, [35,36] and it needs to be evaluated for PEDOT:PSS catalytic electrodes in alkaline media (Figure 1).Films of PEDOT:PSS were deposited by dropcasting of a water-based suspension of PEDOT:PSS onto glassy carbon electrode (GCE).A secondary dopant (DMSO) was added to the suspension contained to increase the conductivity. [37]The voltammetry on PEDOT:PSS-modified GCE in argon-saturated electrolyte solution showed a box-shaped curve of capacitive currents (Figure 1a) originating from the relaxation of the electrical double layer established within the porous film of the conducting polymer. [38]The presence of oxygen in the electrolyte showed the appearance of negative currents manifesting ORR.Linear sweep voltammetry of the hydrodynamic electrode (rotating disc electrode (RDE)) in the oxygen-saturated electrolyte (Figure 1b) was used to remove the limitation caused by slow reagent and product diffusion and to obtain the ORR kinetic currents.An increase in the rotation speed increased the recorded raw currents due to the diffusion enhancement of dissolved oxygen.Sets of voltammograms recorded at different rotation speeds were analysed by the Koutecky-Levich equation for sluggish electrode reaction: where I E is the total current recorded on RDE at the certain potential E, I K is the kinetic current free from any diffusion limitations, while I L is the current contribution due to diffusion limitation; ω is the angular rotation rate of electrode (rad s −1 ) and B is a constant.The plots of the reciprocal total currents 1 IE against the reciprocal root of the angular rotation rate ω À1=2 À Á (Koutecky-Levich plot, Figure 1c) gives a straight line with a finite Y-axis intercept equal to the reciprocal kinetic current 1

IK
at the potential E and the slope equal to 1

B
À Á (Table S1, Supporting Information).The constant B in Equation ( 9) is defined by the equation: where n is the number of transferred electrons per oxygen molecule, .This is close to the values reported for electrochemically polymerized PEDOT and PEDOT:PSS [27] in neutral media.
The hydrodynamic voltammetry of the rotating ring disk electrode (RRDE), PEDOT:PSS-modified RDE fenced with a platinum ring electrode at the independently applied potential of H 2 O 2 oxidation, showed in situ the appearance of H 2 O 2 in the course of ORR (Figure 1b, Note S1, Supporting Information).The in situ estimated values of the number of transferred electrons (ca.2.6, Figure 1d) and peroxide yield (ca. 65-70%, Figure 1d) deviate from the values expected for ORR-to-H 2 O 2 (2 and 100% respectively).This is explained by the presence of a competitive four-electron ORR pathway with water as the terminal product which lowers the peroxide yield.This could be caused by the 'peroxide escape' effect [39] due to the fast H 2 O 2 disproportionation to oxygen and water (Reaction 8) on the traces of transition-metal ions, always present in the aqueous alkaline electrolytes prepared from solid hydroxides. [35,36]he dependence of the calculated diffusion-free (rotationindependent) kinetic current of ORR on the applied potential is available for direct Tafel slope analysis (Figure S1, Supporting Information).The low current densities of ORR on PEDOT:PSS in alkaline media are characterized by a Tafel slope of 30 mV dec −1 implying the EEC mechanism, [40] where the first electron transfer (E) is followed by the second electron transfer (E) and, finally, the slow and rate- Energy Environ.Mater.2024, 7, e12551 determining chemical step (CÞ .This means that the first electron transfer during ORR, computed to occur at the outer-sphere (free from reactant adsorption) on PEDOT, [26] is faster than the final chemical step.The Tafel slope is interpreted as losses caused by the slow reaction rate.Its estimated value of 30 mV dec −1 is smaller than those reported for ORR on PGM-based catalysts in alkaline media.An increase in the current densities of ORR on PEDOT:PSS leads to an increase of the Tafel slope up to 60 mV dec −1 , which implies a change in the reaction mechanism to ECE: [40] Normalizing the diffusion-free kinetic current of ORR-to-H 2 O 2 on PEDOT:PSS in alkaline media by the mass of the electrocatalyst loaded on the RDE enables direct comparison with the mass activities with the performance of other ORR-to-H 2 O 2 electrocatalysts (Figure 2).The common doping level of PEDOT in a composite with PSS is 30%, [41] which implies the ratio between monomer units and dopant units as 3:1. [37]However, only 30% of the sulfonate groups of PSS are involved in the compensation of the positive charge carried by PEDOT. [37]These assumptions correspond to the ratio between PEDOT and PSS units as 1:1.Considering the corresponding molecular weights, one can estimate of the mass of PEDOT to be 43% of the total mass of PEDOT:PSS.This mass correction of the electronic conductor driving the actual electron transfer in ORR gives an increase in the mass activity.The estimated mass activities of PEDOT:PSS and the pure electronic conductor, PEDOT, are comparable with the performance of reported state-of-the-art ORR-to-H 2 O 2 electrocatalysts such as nitrogen-doped carbon. [42]However, the financial costs associated with the degree of complexity of the synthesis and purification steps of the electrocatalytic composites ultimately define the cost-efficiency of a H 2 O 2 e-refinery.From this perspective, the scalable [43] drop-casting route used for the PEDOT:PSS composite is a costeffective approach in comparison to the routes used for the other the state-of-the-art catalysts.

OER
Mesoporous NiO (mesoporous-NiO) can be used as an electrocatalytic auxiliary anode for H 2 O 2 erefinery and accordingly drive the OER in alkaline media: Driving Reaction ( 11) from left to right can be described as a synchronized loss of equal number of hydroxide anions, OH À and electrons, where the latter ones are extracted with an external electric circuit powered by a DC source.The loss of a hydroxide anion in the aqueous media is accompanied by the appearance of a virtual proton.Consequently, the OER on the anode is a protoncoupled electron acceptor process, which can be utilized as the auxiliary anode for a protoncoupled electron donor process, as the ORR-to-H 2 O 2 on the cathode.
The structure-directing agent sodium dodecyl sulphate (SDS) was used as a template in a hydrothermal synthesis to form a mesoporous NiO electrocatalyst. [16]-ray diffractometry of the as synthesized mesoporous-NiO (Figure 3a) showed only peaks consistent with cubic NiO (JCPDS card number 014-0117).Nitrogen sorption analysis of NiO shows a type IV isotherm with a H1 hysteresis loop (Figure 3b) [44] which is consistent with mesoporosity.Brunauer-Emmett-Teller (BET) analysis of the adsorption isotherm yield a specific surface area of ca.110 m 2 g −1 with a total pore volume of 0.24 cm 3 g −1 (Table S1).The maximum of the Barret-Joyner-Haleneda (BJH) pore size distribution is located at ca. 3.3 nm (Figure 3c).Transmission electron microscopy (Figure 3d) confirms the presence of mesopores within highly crystalline NiO nanosheet.Selected area electron diffractometry (Figure 3d insert) suggests the nanosheet to have a preferred crystal orientation with [111]  normal to the sheet surface.
The half-cell evaluation of the OER anode was performed with mesoporous-NiO-modified carbon fibre paper (CFP) as the working electrode under different loads.The voltammetry (Figure 3e) showed the appearance of oxidation currents at positive potentials higher than ca.1.55 V (RHE), which manifests OER at the anode.The slight influence of the current collector is illustrated by the blank CFP as a minor increase in the oxidation currents caused by the sluggish kinetics of OER on CFP.The presence of a Ni 2+/3+ redox peak at ca.Energy Environ.Mater.2024, 7, e12551 loaded with 1.0 mg cm −2 mesoporous NiO was confirmed by the estimation of the charge transfer resistance using electrochemical impedance spectroscopy (Note S2, Supporting Information), which is in agreement with published reports. [45,46]This optimized anode was utilized in the assembly of H 2 O 2 e-refinery.

Electrolyser
A PGM-free H 2 O 2 e-refinery (Scheme 1) was assembled using an anion -exchange membrane (AEM), which assures the transport of hydroxide-ions.The transport of hydroxide anions, the products of ORR and the reagents of OER was maintained on the AEM, which reconstructs of the alkaline environment in the electrolyser cell.The AEM sandwiched between the ORR-to-H 2 O 2 cathode (PEDOT:PSS) and the OER anode (mesoporous-NiO) served as a solid electrolyte in between the liquid feds to the cathode and anode compartments.An oxygen-saturated cathode feed was used in all measurements.Ramping up the constant currents supplied to the electrolyser led to a rise of the cell voltage (Figure 4a) and the amount of H 2 O 2 in a liquid in the cathode compartment (Figure 4c).This illustrates the launch of electrolysis-driven comproportionation: A supply of 0.02 mA cm −2 to the electrolyser fed with pure water resulted in a cell voltage of 1.25 V and trace concentrations of H 2 O 2 (up to 0.2 mM after 5 h of operation (Figure S4A, Supporting Information)).To increase the production rate, the current was increased to 0.4 mA cm −2 (20 times), which more than doubled the cell voltage to 3.3 V (Figure 4a).This also resulted in a linear increase of H 2 O 2 concentration up to 3.5 mM during 4 h of electrolysis.The Faradaic efficiency of the e-refinery in pure water was rather low (Figure S4B, Supporting Information, up to 40% at 0.4 mA cm −2 ) and showed a minor dependence on the applied current density (Figure S4B, Supporting Information).Ideally, the electrode reactions take place at their equilibrium potential at any current.The increase of cell voltage with increased current illustrates an energy loss in the electrolyser, which contributes to the operational cost of the H 2 O 2 erefinery.
Persuasively, the change of the fed media from pure water to alkaline electrolyte resulted in a significant mitigation of the voltage loss at all values of applied current densities (Figure 4a), which was accompanied by the higher rate capabilities (Figure 4b) and the higher efficiencies of H 2 O 2 production (Figure 4c,d).Indeed, 12-and 40-times higher currents (0.25 mA cm −2 and 0.8 mA cm −2 respectively) need to be applied on the electrolyser fed with 0.01 M and 0.1 M KOH, respectively, to create cell voltages close to 1 V, in comparison with the currents observed in operation in pure water (0.02 mA cm −2 ).Coherently, the presence of alkaline electrolyte in the liquid media fed to the electrolyser led to higher H 2 O 2 production rates (Figure 4b) and improved efficiency (Figure 4d, Figure S4, Supporting Information).During first 4 h with a current of 0.4 mA cm −2 , the H 2 O 2 generation was approximately linear with time with rates of 1.08 and 1.35 μmol min −1 in 0.01 M and 0.1 M KOH respectively.At longer operation times, the performance of the e-refinery decreased in all investigated conditions (Figure 4d, Figure S4, Supporting Information), which might be due to the degradation of active components of e-refinery, such as graphite moieties on OER anode.This issue is beyond the scope of this report.The maximum peroxide concentration obtained (in 0.1 M KOH after 5 h of electrolysis) was 10 mM.
The oxygen concentrations in water saturated by pure oxygen and air (at 25 °C and pressure of 1 bar) are 1.25 mM and 0.25 mM respectively.The upper limit of the H 2 O 2 production rate was estimated considering an electrolyser flow rate of 7.8 mL min −1 and 100% conversion of oxygen (i.e.ideal diffusional transport, fast ORR-to-H 2 O 2 kinetics and full selectivity) to 9.75 and 1.95 μmol min −1 for the electrolyser fed with oxygen-and airsaturated water respectively.The production limit defined by air saturation (Figure 4b) delimits the operational conditions when air can be used as an oxygen source.
The electrolyser fed with pure water and diluted electrolyte (0.01 M KOH) demonstrated a H 2 O 2 production rate independent of current density (Figure 4b).In these conditions, an increase in the driving force of the Faradaic reactions, ORR and OER, did not increase the production rate.This implies that the limiting process of H 2 O 2 production in these conditions is diffusional transport of reagents and products, which is affected by the flow rate of the liquid media.Oppositely, the operation in concentrated electrolyte (0.1 M KOH) showed a close-tolinear dependence of production rate with respect to current density (Figure 4b), which implies a contribution by the rate of the Faradaic reaction.In these conditions, the production has a mixed control by both slow Faradaic reactions and slow diffusional mass transfer.This is illustrated by the higher Faradaic efficiency obtained at a higher flow rate (Figure 4d).Indeed, a doubled flow rate of the alkaline electrolyte increased the Faradaic efficiency of the electrolyser to 80% at short operation times (1 h).
Loss A is insensitive to the presence of the electrolyte in the feed.This is illustrated by the measurements of the high-frequency impedance of the electrolyser fed with either pure water or alkaline electrolyte (Figure S5, Supporting Information).The extrapolation of impedance data on the real axis yields shows that the estimated value of AEM ohmic resistance (<1 Ohm) is insensitive to the presence of electrolyte.In other words, the smallest electrolyte-insensitive ohmic resistance of the sandwiched membrane at the equilibrium conditions is not a main contributor to the overall voltage losses in the electrolyser.Loss C is not the cause either to the main origin of the overall losses as the rate of the overall Faradaic reaction (2 H O þ O ! 2 HO ) is independent on the concentration of hydroxide anion, which is at the same time the reagent and the product of cathode and anode half-reactions (ORR Reactions 2-4 and OER Reaction 11, respectively).The origin of loss D should be found in porosity and flow rate.Indeed, the transport of reagents and products within the pores of electrocatalysts is limited at high current densities. [16]he main cause of voltage loss is the resistance due to the slow ionic transport through the porous electrodes and AEM at the electrolysis conditions, loss B. This is clearly shown by the significant improvement of the electrolyser performance upon the addition of the electrolyte (Figure 4a-d).
Importantly, in ionically nonconductive media (water in this case), electrochemical transformations such as ORR-to-H 2 O 2 and OER can only occur at the points where the three transport modalities ionic, electronic and reagent/product fluxes are in physical contact, so-called triple points.Importantly, both PSS in PEDOT:PSS-based ORR cathode and Nafion in NiO-based OER anode maintain the selective transport of cations.On contrary, AEM sustains selective anion transport.Hence, there is a conflict between the cation transport within the electrodes and the anion transport in the membrane.This might lead to the isolation of the electrode bulks from the AEM-maintained transport of hydroxide anions in pure water.And indeed, by adding external electrolyte (KOH) in high concentration, one levels out the contradiction in ionic transport by creating excess of available ions (so-called screening of immobile cations in AEM).The low Faradaic yield of H 2 O 2 electrosynthesis observed for the device operation in pure water (Figure 4d, Figure S4A,B, Supporting Information) is a result of limited ionic transport to the bulk of the porous electrodes.Indeed, in pure water, the electrolysis can take place only at a small number of triple points created on a thin 2D region of the direct contact of AEM and the porous electrode accessible for the liquid.This implies significantly higher values of the actual current density than the geometrical current densities in idealistic half-cell measurements in electrolyte solution (Figures 1a, 2 and 3e).This can lead to higher overpotentials at these points and hence provoke background electrochemical reactions such as further reduction of H 2 O 2 to water, water splitting and degradation of the electrocatalysts lowering the Faradaic yield of H 2 O 2 .The addition of electrolyte increases in the volumetric 3D density of triple points in the porous electrode, which leads to optimization of the actual current density spent on ORR-to-H 2 O 2 .
The mitigation of the loss upon slow diffusional transport of reagents/products (loss D) is visible upon the increase in the flow rate as a further improvement of efficiency up to 80% (Figure 4d) at short operation times (1 h) of the electrolyser with optimized ionic transport (0.1 M KOH feed).In the presence of an electrolyte the capacitive behaviour of electrodes is introduced, leading to non-faradaic leakage currents due to the self-discharge at the conditions of constant current electrolysis, which implies the possibility of the electrode design optimization to increase the efficiency.
As illustrated, the kinetic control of the H 2 O 2 e-refinery implies strict control of the operational conditions.This means that the increase of the current density to increase the production rates (Figure 4b) will end up in the loss of faradaic efficiency of production due to the launching of background electrochemical reactions.The H 2 O 2 production rate can be boosted by both stacking in series using bi-polar plates and the use of the gas diffusion electrodes, which enables the direct uptake of oxygen from the air mitigating the oxygen solubility limit.

Conclusions
To conclude, we demonstrated e-refinery of H 2 O 2 using ORR and OER electrocatalysis on cost-effective and scalable organic (PEDOT:PSS) and inorganic (mesoporous NiO) interfaces.The evaluation of ORR on PEDOT:PSS in alkaline media using a halfcell configuration showed the specific production of H 2 O 2 with mass activity comparable with state-ofthe-art electrocatalysts.The operation of H 2 O 2 erefinery assembled using anion-exchange membrane in weakly supported media showed a limitation by the ionic transport.Optimized operation in high flow rate and external electrolyte of high concentration showed 80% production efficiency.

Methods
Reagents: Sodium dodecyl sulphate (SDS, ≥99.0%,Sigma-Aldrich), nickel (II) chloride hexahydrate (≥ 98%, puriss; Sigma-Aldrich), ammonium hydroxide solution (28.0-30.0%,ACS reagent; Sigma-Aldrich), Nafion ® perfluorinated resin solution (5 wt%, Sigma-Aldrich), potassium hydroxide (Reagent grade; Sigma-Aldrich), ethanol (99.5%, analytical grade, Solveco), horseradish peroxidase (HRP), 3,3 0 ,5,5 0 -tetramethylbenzidine (TMB) and dimethyl sulfoxide (DMSO) were purchased from Sigma (Sweden) and used as received.Experiments were carried out with Milli-Q water from a Millipore Milli-Q system.Procedures: Mesoporous NiO-Hydrothermal synthesis [16] was adopted for the preparation of mesoporous NiO.SDS and nickel (II) chloride were dissolved in distilled water in a ratio of 2:1 yielding a transparent solution after stirring at room temperature for 1 h with SDS at a concentration of 1 wt.%.The pH of the solution was adjusted to 10 with dropwise addition of 28-30% ammonia.The mixture was continuously stirred for 3 h and then transferred into a PTFE bottle.After the hydrothermal treatment (100 °C for 48 h) the precursor was obtained by filtration followed by washing with ethanol and water for several times and drying at 80 °C overnight.The final mesoporous NiO powder was achieved after the calcination of the precursor in a muffle furnace (400 °C for 4 h using a temperature ramp of 10 °C min −1 ).
The synthesized mesoporous NiO powder was dispersed in ethanol/Nafion solution (49:1, v:v) by ultrasonication for 30 min yielding a homogeneous ink with a OER catalyst concentration of 5 mg mL −1 .
The electrochemical measurements were carried out on a SP-300 potentiostat (Bio Logic Science Instruments) using three electrodes: GCE (5.61 mm diameter), Energy Environ.Mater.2024, 7, e12551 platinum wire and Hg/HgO as working, counter and reference electrodes respectively.The potential scale conversion to RHE was performed as: , where E vsRHE and E vsHg=HgO are potentials measured with respect to RHE and Hg/HgO reference electrodes, respectively (V), ΔE Hg=HgOvsSHE is the difference between SHE and Hg/HgO reference electrodes (0.098 V) and pH =13 and pH = 14 for 0.1 M and 1 M KOH respectively.Ohmic drop compensation was performed as: E iRÀcorrected ¼ E measured Ài Â R, where E iRÀcorrected and E measured are corrected and measured (raw) potentials (V), i is the current (A) and R is the ohmic resistance (Ohms) determined from EIS measurements as interception at high frequencies.
Prior to use, GCE was successively polished with 0.05 μm Al 2 O 3 powders and sonicated in Milli-Q water.The rotating disk ring electrode (RRDE, 5.61 mm OD GCE, 320 μm gap, platinum ring 6.25 mm ID, 7.92 mm OD; Pine Research Instrumentation Inc.) has been utilized with the controlled rotation speed.
The CFP modified by OER catalyst (mesoporous-NiO/CFP) was activated first by cyclic voltammetry (CV) in oxygen-saturated 1 M KOH in a potential range of +0.2 -+0.9 V (vs Hg/HgO) at scan rate of 100 mV s −1 for 5 cycles (Figure S2, Supporting Information).
PEDOT:PSS modified RRDE was prepared by drop-casting of the PEDOT:PSS ink (2.47 μL) on the surface of blank RRDE followed by drying (60 °C for 30 min).
X-ray diffraction (XRD) was performed on Panalytical X'Pert Pro X-ray diffractometer with Cu Kα radiation (λ = 0.15406 nm).N 2 physisorption measurements were carried out on an ASAP 2020 (Micromeritics) at −196 °C.The samples were degassed at 300 °C for 6 h under vacuum before analysis.The Brunaure-Emmert-Teller (BET) method was used to calculate the specific surface area (SSA, relative pressure ca.0.99).The Barret-Joyner-Halenda (BJH) method was used for calculating the pore size distributions from the desorption branch of isotherms.Transmission electron microscopy (TEM) and selected area electron diffraction (SAED) were performed using a FEI Tecnai G2 microscope operated at 200 and 380 kV voltage respectively.
H 2 O 2 electrolyser: Anode-Carbon fibre paper (CFP, Toray carbon paper 060; Fuelcellstore, College Station, TX, USA) was calcined at 600 °C for 30 min to remove PTFE from the surface followed by washing with ethanol, water and drying at 60 °C overnight.The OER catalyst ink was deposited on CFP by dropcasting with a loading in a range of 0.5-4.0mg cm −2 followed by drying (100 °C).
Cathode-CFP was modified via drop-casting by PEDOT:PSS ink with a loading of 1 mg cm −2 followed by drying at drying (60 °C).
The H 2 O 2 electrolyser was assembled using flow membrane cell (C-flow 5 × 5 (active area of 25 cm 2 ), C-Tech Innovation Ltd. (UK) and anion-exchange membrane (AEM, Fumasep FAA3PKBO; FuelCellStore)) sandwiched between CFP-based anode and cathode without additional integration.The graphite felt (AvCarb G200; FuelCellStore) used as diffusion layer of the electrolyser was hydrophilized by the immersion in a concentrated H 2 SO 4 for about 5 s followed by washing with an excess of water.The anode and cathode compartments were fed independently with aqueous media using peristaltic pump (flow rate 7.8 mL min −1 ).
Off-line UV-vis analysis was utilized for the quantification of produced H 2 O 2 .An aliquot (10 μL) of aqueous media from cathode compartment was collected every 20 min.Then 290 μL of stirred freshly prepared solution of HRP (0.75 ng mL −1 ) and TMB (30 μg mL −1 ) in 0.1 M phosphate-citrate buffer solution pH 6 were added to each of the aliquot.Then the absorbance of the mixture was measured (at 653 nm) by UV-vis plate reader (BioTek Synergy H1 Hybrid Multi-Mode Reader).Finally, the concentration of H 2 O 2 was determined from the calibration line with known H 2 O 2 concentrations (0, 10, 20, 30 and 40 mkM) by standard additions method.The Faradaic efficiency (FE) over time was calculated as: where F is Faraday constant (96485.3A mol s −1 ), C is the concentration of H 2 O 2 (mol L −1 ), V is the volume of the cathode compartment (L), I is the current (A) and t is the time of the electrolysis (s).
is the kinematic viscosity of the electrolyte (1 × 10 −2 cm 2 s −1 ) and C O is the bulk concentration of O 2 (1.2 × 10 −6 mol cm −3 ).The number of transferred electrons per oxygen molecule calculated from the slopes of the Koutecky-Levich plots is close to 2 (Figure 1d), which demonstrates the two-electron pathway for ORR on PEDOT:PSS in alkaline media with H 2 O 2 as terminal product (ORR-to-H 2 O 2 )
~1.40 V (RHE) illustrates the formation of NiOOH on the electrocatalyst as a surface intermediate of OER.A mesoporous-NiO loading of 1.0 mg cm −2 resulted in the highest oxidation currents and the lowest kinetic potential loss of ca.1.67 V (RHE) to reach a current density of 100 mA cm −2 among the tested loadings.The fastest OER rate on CFP

Figure 1 .
Figure 1.ORR-to-H 2 O 2 electrocatalysts on PEDOT:PSS in alkaline media.a) cyclic voltammograms on stagnant PEDOT:PSS-modified GCE in argon-and oxygen-saturated electrolyte (black and red curves, respectively, 0.1 M KOH); b) linear sweep voltammograms measured on PEDOT:PSSmodified RDE at different rotation speeds and the current on ring electrode (at 1.27 V (RHE), 400 rpm) in oxygen-saturated 0.1 M KOH; c) Koutechý−Levich plot of RDE voltammetry data; d) potential dependencies of the number of transferred electrons per oxygen molecule (grey curvefrom K-L analysis; empty symbols-from RRDE analysis) and the in situ H 2 O 2 yield (at 400 rpm).

Figure 4 .
Figure 4. Performance of H 2 O 2 e-refinery.The dependences of a) electrolyser cell voltage and b) H 2 O 2 production rate on the current density; time dependencies of c) H 2 O 2 concentration (inset: electrolyser scheme) and d) Faradaic yield of the electrolyser (at 0.4 mA cm −2 ) fed with pure water, 0.01 M and 0.1 M KOH (black, red and blue curves respectively); e) scheme of the losses in electrolyser.