Assessing the Intrinsic Activity of Pt‐Group Electrocatalysts for Carbon Monoxide Oxidation: Best Practices and Benchmarking Parameters

Pt‐group elements are the state‐of‐the‐art electrocatalysts for various fuel cells. However, their CO‐poising is a critical hitch for large‐scale applications, so the researchers are exerting huge efforts to solve this issue. The exponentially increasing attention in this field pressures the researchers to publish their findings quickly, leading to unavoidable flawed evaluation parameters for the intrinsic activity of electrocatalysts. The CO oxidation (COOxid) is highly sensitive to various factors. Thus, it is urgent to afford a deeper understanding of the inherent COOxid activity of state‐of‐the‐art electrocatalysts and adopt accurate guidelines for researchers to test, optimize, and compare their electrocatalysts. This review provides exactitude in the evaluation and precise assessment of the key descriptors related to electrocatalysts (i. e., effect of both size, shape, and support) and CO oxidation (i. e., effect of electrolyte, working electrode, and CO surface diffusion). This is besides the fundamental aspects (i. e., COOxid Process, mechanism, measurements, calculations, thermodynamics, and kinetics). Various experimental results from our group and others besides in‐situ analysis were provided to support our deep discussion. Finally, we provide a brief synopsis of the relevant milestones of the up‐to‐date challenges and perspectives.


Introduction
The inevitable utilization of non-renewable, excessive fossil fuels led to emissions of hazardous gasses, [1] so various solutions were adapted to solve this problem in gas conversion reactions [2] and the exploitation of green and renewable energy sources. [3]Anodic and cathodic proton electrolyte membrane fuel cells (PEMFCs) are the most promising energy sources due to their low-operation conditions, ease of handling, and earthabundance of their fuels (i.e., methanol, ethanol, and glycerol). [4]t-group elements are the main anodes or cathodes for such fuel cells, but their high-cost, earth rarity, and poisoning by CO are crucial barriers in the commercialization process. [5]Mainly, Pt-group catalysts are easily poisoned and loose catalytic active sites by adsorption of small traces of adsorbed CO (i.e., 10-100 ppm), so the CO should be promptly removed from the cells via the electrochemical oxidation to allow the long-term operation. [6]Thus, optimization and understanding the fundamental factors controlling the electrocatalytic CO Oxid on Ptbased electrocatalysts is a crucial step toward large-scale utilization of PEMFCs as green and efficient energy resources. [7]sically, the inherent CO Oxid mechanism, high enthalpy of COadsorption on Pt (i.e., up to 140 kJ/mol at low coverages), and slow oxidation kinetics (i.e., starts over ~0.6 V vs. RHE) are necessary fences.These blockades could be solved by tailoring the morphology, size, [8] and composition of the Pt-group catalyst besides using support (i. e., metal oxide, carbon). [9]ainly, mixing Pt-group elements with other inexpensive and earth-abundant metals (i.e., Co, Cu, Bi, Ni, and Fe) [10] modulates their d-band level (i.e., upshifting or downshifting), which can optimize the adsorption/activation of reactants and tolerate the adsorption energies of intermediates and products. [11]Altering the catalyst's size and morphology to provide multiple surface facets, enhances the CO surface diffusion, and maximizes the utilization of metal active sites. [12]Using catalyst supports (i.e., carbon, metal oxide, and graphene) endows high dispersion of Pt-catalysts, reduces Pt-loading, enhances electrolyte interaction, increases surface area, and provides stability against the dissolution and aggregation in aqueous electrolytes under severe oxidation conditions [13] Also, the strong electronic interaction with the supports eases the generation of OH and its adsorption on Pt-based catalysts at low potential, which accelerates the CO Oxid kinetics and reduces CO-poisoning. [14]lobal researchers are exerting huge efforts to understand the effect of such parameters and to ensure accurate and suitable utilization of evaluation factors in intrinsic electrocatalytic CO Oxid . [15]However, the exponentially increasing number of publications in this field pressurizes researchers worldwide to publish their findings quickly, which somewhat leads to unavoidable flawed evaluation parameters reflecting the intrinsic activity of electrocatalysts, attributed to the absence of obvious insights on the activity parameters.Thus, enormous efforts were devoted to tolerating the CO-tolerance on Pt-group electrocatalysts for the oxidation of organic molecules (i.e., ethanol, methanol, and formic acid), which led to over ~104000 published articles according to; however, their CO Oxid activity was highlighted in ~230 articles (Figure 1a).Thus, this research arena was comprehensively reviewed in various excellent reviews on Pt-group catalysts for electrocatalytic CO Oxid and others, but not on assessment Factors. [6,16]Thereby, it is urgent to better understand the inherent CO Oxid performance of Ptgroup electrocatalysts, besides precise assessments of parameters for analyzing and optimizing the catalytic performance. [17]his review provides qualitative and quantitative analysis on the effect of state-of-the-art catalysts (i.e., particle size, morphology, and support) and CO oxidation factors (i.e., effect of electrolyte, working electrode, and CO surface diffusion) (Figure 1b).This, besides elucidation of the fundamental issues (i.e., CO Oxid process, mechanism, calculations, reaction thermodynamics, and kinetics), is supported with the needed experimental results and in-situ analysis to support our discussion.Meanwhile, we have added various experimental data to ensure best practices to optimize the electrocatalytic CO Oxid performance of Pt-based and Pd-based catalysts as a function of alloying, support, and shape.This review ends with a conclusion on the current barriers and perspectives to tailor the electrocatalytic merits of CO Oxid activity on Pt-group elements.

Function and Advantages of Commercial Pt/C
Commercial Pt/C catalysts are important in various industrial, environmental, and fundamental applications because they couple unique catalytic meets of Pt and inimitable physiochemical properties of activated carbon nanosheets.Mainly, Pt/C comprises Pt nanoparticles (3-5 nm) distributed over twodimensional carbon nanosheets with an average loading of (5-30 wt.%). [18]So Pt nanoparticles with small sizes are highly active, have a quantum effect, and are accessible during catalytic reactions, whereas carbon has a great surface area, outstanding electrical conductivity, and ease of dispersion over various substrates for commercial usage (i.e., Nafion and metalfoams).Also, Pt/C reduces the cost because it has low mass loading of expensive and scarce Pt and could be easily prepared in high yield (several kilograms).These merits enhance the catalytic activity and durability, accelerating charge mobility and electron transfer during electrocatalytic reactions.Thereby, Pt/C electrocatalyst is versatile for various applications like fuel cells (i.e., alcohol, acid, and oxygen), electrolyzers (i.e., water splitting), and batteries (i.e., metalÀ air batteries), and environmental remediation (i. This is owing to the ability of Pt/C to accelerate redox reactions at lower applied potential besides high tolerance for intermediates.Regarding the CO Oxid , Pt is the active site for the adsorption of reactants (i. However, some technological and environmental challenges should be solved, like the cost and scarcity of Pt, besides its leaching and aggregation in electrolytes alongside carbon corrosion under certain conditions.19b,20] Using porous carbon support with a high surface area could also enhance the performance of Pt. [21]

CO Oxidation Process
The electrochemical CO Oxid on the surface of Pt-group catalysts in or out of alcohol fuel cells is undeniably a complex process, so it is essential to decipher its pathway and mechanisms to improve the catalytic performance of such fuel cells and reduce the CO-poisonous. [22]There are various proposed processes for the CO Oxid process on Pt-group catalysts, but mainly four mechanisms are the most acceptable, including the ligand effect, electronic effect, bi-functional effect, and hydrogen spillover effect, [6] which all start with CO adsorption, followed by diffusion, oxidation at Pt-group active sites, and finally, CO 2desorption (Figure 2a).In the ligand effect, the metal interacts with CO molecules, decreasing the binding energy between CO and the Pt surface and thus easing the CO Oxid .In the electronic effect, the metal alters the electronic structure by modulating the d-band center of Pt.Meanwhile, in the bifunctional effect, the delocalized d-orbital electrons on the metal surface induce the electron transfer to the Pt-group. [6,23]In the hydrogen spillover, a hydrogen atom is dissociated from the Pt-group surface and transferred to the metal sites, creating an interaction of COÀ Mo bond, whereas the metalÀ H interaction decreases the bond between CO and metal, resulting in facilitating the CO oxidation on the Pt surface.Various studies reported that Pt-based electrocatalysts with a uniform dispersion of Pt tend to show a stronger spill effect. [6,24]These four processes are driven mainly by two factors: the first is driven by the oxophilic effect in the case of mixing Pt with other transition metals that induce the generation of OH species on the oxophilic sites of transition metals via the activation and dissociation of water in electrolytes (H 2 O + ** * OH ads + H + + e À ) then OH is transferred to the Pt-group to oxidize CO adsorbed on Pt (PtÀ CO). [6]The second pathway is endowed by the electronic effect, resulting from the modulation of the d-band center of Pt-group elements by second transition metals, which alters the chemisorption properties of the Pt-group towards hydrogen and CO, subsequently tolerates the coverage of CO, and increases the number of unoccupied Pt-group active sites, allowing the oxidation of H 2 . [6]ll these processes mainly follow the Langmuir-Hinshelwood mechanism, [25] which entails the initial CO adsorption on the Pt-surface (PtÀ CO ads ) simultaneously with the activation/ dissociation of H 2 O in the electrolyte to produce active (OH ads ) at the second metal surface (MÀ OH). [6]Meanwhile, both PtÀ CO ads and MÀ OH are oxidizing on their respective sites via the bifunctional mechanism to induce the reaction of PtÀ CO ads with MÀ OH to yield intermediate formate species (COOH) at the reactive sites. [6,26]Finally, the intermediate is dissociated to produce CO 2 , which is easily desorbed from the catalyst site Eq.1-4.
It is noteworthy that, despite the great achievements made in the electrocatalytic CO Oxid .The mechanism and pathways need a lot of in-situ analysis that is still required (i.e., DRFIT, FTIR, and Raman) to understand it and to explore the role of interaction with the support rather than the role of metals mixed with Pt-group elements.

CO Oxidation Measurements
The electrochemical CO Oxid measurements are usually carried out in a three-electrode cell using a counter electrode (i.e., platinum wire), reference electrode (i.e., silver/silver chloride (Ag/AgCl), and saturated calomel electrode (Hg/HgCl), and the working electrode (i.e., glassy carbon, carbon sheet, and metal  foam) covered with Pt-group electrocatalysts in the presence of aqueous solutions of acidic (i.e., HClO 4 and H 2 SO 4 ), alkaline (i.e., KOH and NaOH), and neutral (i.e., PBS and NaHCO 3 ) electrolytes saturated with CO gas (i.e., 2 % and 5 % balanced with N 2 or Ar).Another measurement is called CO-striping, which is measured after.
To estimate the CO Oxid performance, various electrochemical tests should be measured, like the cyclic voltammogram (CV), linear sweep voltammogram (LSV), electrochemical impedance spectroscopy (EIS), and chronoamperometry test.Notably, prior to the electrochemical CO Oxid or CO-stripping, the working electrodes should be cleaned via carrying out CV cycles (i.e., 40-100 cycles) at high scan rates (i.e., 150 or 200 mV/s) in electrolytes under an inert atmosphere (i.e., N 2 or Ar) till we get stable voltammogram current.Then, an additional three CV cycles should be measured at a lower scan rate (i.e., 50 mV/s) to get the typical voltammogram features of the Pt-group (i.e., Hydrogen ads/des and formation of metal-redox) (Figure 3a) [27] and also to calculate the electrochemical active surface area.Following that, for the electrochemical CO Oxid , all the tests are directly measured in the electrolyte solutions under continuous CO-pursing to get the CO-anodic oxidation features (Figure 2b).Meanwhile, for the CO-tripping test, the monolayer of CO-gas is initially adsorbed on the Pt-group catalysts surface via changing N 2 gas to CO gas under constant potential (i.e., 0.09-0.2V, where CO is not oxidized) for a known time (i.e., 120-3600 sec) and then the N 2 gas should be switched back on for (i.e., 60-300 sec) to remove non-adsorbed CO gas.Then, the CV test is measured for 3 cycles in an anodic direction to get the faradaic current in the first cycle and the disappearance of hydrogen ads/ des area, followed by the disappearance of the CO-anodic current and recovery of the hydrogen ads/des in the 2 nd or 3 rd cycle.Notably, the reaction parameters of both CO Oxid and COstripping are not yet optimized, and several factors should be studied, like gas flow rate and time, besides the type of working electrode and interfering with other gases like H 2 .Also, the effect of electrolytes (i.e., type, concertation, and composition) is still ambiguous and not emphasized enough.

CO Oxidation Calculations
16b] The CO Oxid current density (I Co ) is obtained via normalizing the anodic current (I Anod ) to the geometric area of the working electrode (A WE ) using Equation 1 (Eq. 1) The electrochemical active surface area (ECSA) represents the portion of the catalyst surface that actively participates in the electrochemical reaction, so a higher ECSA is preferred for promoting the catalytic activity.ECSA is calculated by dividing the charge obtained from the integration of the CO-adsorption area in the forward direction (Q CO ) by the mass loading of catalysts on the working electrode (m Cat ) and the charge attributed to the adsorption of monolayer CO on the catalyst surface (420 μC/cm 2 ) Eq. 2 (2) The onset potential (E Onset ) is the initial potential point that allows the initiation of the CO Oxid to get I Anod , meanwhile the CO Oxid (E Oxid ) is the potential needed to completely oxidize CO to CO 2 and generates the maximum I Anod value (Figure 2b).Therefore, a higher I CO at a lower E Oxid, besides an earlier E Onset are benchmarks of a quick CO Oxid kinetic.
The mass activity (I mass ) is obtained by dividing the I CO over the m Cat on the working electrode (Eq.3).
The specific activity (I s ) is obtained via dividing the I CO over the m Cat and ECSA (Eq.4).
To determine the CO Oxid kinetics, the CV should be measured at different scan rates (υ) (i.e., 25-300 mV/s), which allows increasing I CO with increasing the scan rate υ.Then, the obtained I CO is plotted against υ 1 = 2 following the Randles-Sevcik equation to get the slope.The linear relationship and higher slope indicates the diffusion coefficient of CO Oxid processes and quicker reaction kinetics.
Turnover frequency (TOF) is a crucial factor in describing the intrinsic catalytic activity of a catalyst, which can be calculated based on normalizing the I CO to the Faraday's constant (F) (96485 C/mol), the number of electrons transferred in the reaction (i.e., 2 electrons), and ECSA (Eq.5).A higher TOF indicates better catalytic activity and an abundance of active sites.However, it is rarely reported in the electrocatalytic CO Oxid .TOF ¼ 16b] The apparent electron transfer rate constant (k app ) is calculated by normalizing the universal gas constant (R = 8.3145 J/mol/K) and absolute temperature (T) to the number of electron transfers, Faraday constant, geometric area of the working electrode, concentration of the electrolyte (C), and R ct (Eq.6) The Impedance of the constant phase elements (Z CPE ) is calculated based on the numerical value (Q0) of the admittance at ω = 1 rad/s, ideality factor (n), phase angle of the Impedance (0 < α < 1), angular frequency (ω), and imaginary component [28] of (À 1 1/2 ) (Eq. 7) The value of n indicates the type and behavior of electrodes (Z CPE is a pure resistor, capacitor, inductor, and W d at n = 0, 1, À 1, and 0.5, respectively. [29]Higher CPE, besides lower R ct and R s are cogent indicators for superior charge mobility and better electrolyte-electrode interaction.However, few studies reported the EIS data and their fitting, [16b] meanwhile, the shape of Nyquist plots depends on catalyst shape, composition, and support besides electrolyte (i.e., type and pH), during the CO Oxid .
The catalyst's stability is one of the main factors for largescale applications, so it is usually estimated through chronoamperometry or accelerated durability test (ADT) under COpursing under constant applied potential (i.e., E Oxid or E Onset ) for several hours.Following that, the CV and LSV should be measured again to see the change in the I CO , E Oxid , E Onset , I m , I s , and ECSA.Then, the catalyst's shape, composition, and interaction with the support could be investigated using TEM and XPS analysis after durability tests.The electrocatalysts that can maintain their active sites for the long term are highly required.Also, the stability could be performed via applying the CV test for several cycles (i.e., 1k-10k) at a high sweeping rate (i.e., 100-200 mV/s) to check the change in the voltammogram features and ECSA. [30]However, the durability has only been studied for a few minutes or hours, but the practical applications need long-term durability for several days or weeks.

Fabrication of the Working Electrode
There are various working electrodes like glassy carbon, carbon paper, and carbon cloth sheet, [31] but the first one is the most common one, and it should be polished conductively by 1 μm and 0.3 μm aluminum powder and then rinsed in double deionized water/ethanol (3/1 v/v) for 3-10 sec under sonication treatment at 25 °C before being covered with the catalyst ink.The catalyst ink of Pt-group catalysts is prepared by dispersion of 2-4 mg catalyst in a 1-2 mL of inorganic, organic, or hybrid solvent (i.e., water, ethanol, toluene, and isopropanol) contains a suitable binder (i.16a] The structure-directing agents that are usually used during the preparation of Pt-group catalysts are the main factor in selecting the solvents for the catalyst ink.Pt-group elements capped with ionic/non-ionic copolymers (i.e., PVP, F127, and CTAB) [32] are highly soluble in water, while other capping agents (i.e., Triton-X100 and oleylamine) are only soluble in organic solvents (i.e., hexane, DMF, and toluene).
After preparation the catalyst inks, they should be directly deposited on the working electrodes using different methods like drop casting, impregnation, spray coating, and electrochemical deposition, [33] but drop casting is the most common and facile one.Then, prior to the electrochemical tests, the electrodes should be left to dry under vacuum at 80 °C to allow the polymerization of the binder with the catalysts or at a lower temperature if the catalysts are not stable at high temperatures.The film uniformity on the working electrode is essential to show the typical voltammogram features of the catalysts and ensure higher current density, ECSA, and maximize Pt utilization (Figure 3). [27]Early study optimized the film uniformity of commercial Pt/C, and found that, the ECSA and ORR activity depended on the film uniformly (Figure 3). [27]

Effect of Catalyst Capping Agent
The capping agents usually affect the electrocatalytic activity, so they should be stripped from Pt-group catalysts via heating under an oxidation atmosphere (i.e., PVP decomposed at (473-623 °K) in a 20 % O 2 /He), [34] exposure to UV light illumination under an ozone atmosphere, [35] acid/based etching (i.e., acetic acid removes oleylamine, [36] and H 2 O 2 /H 2 SO 4 removes PVP [37] ), displacement by hydride generated from an aqueous solution of NaBH 4 with/without tert-butylamine (TBA) (i.e., eliminates organothiols, thiophene, adenine, rhodamine, halide ions, and PVP), [38] electrochemical potential CV between 0 and 1.0 V RHE in aqueous solution of 0.5 M NaOH at a sweeping arte 0.5 V/s for 100 cycles (i.e., removes PVP and oleylamine/oleic acid), [39] electrochemical displacement.However these approaches affect shape (i.e., cause aggregation), cause catalyst posing (i.e., cook formed from heating of polymer), and change composition (i.e., leaching).Also, the CO stripping promotes the removal of surfactants (i.e., PVP and others) at a high potential (1.0 V RHE ) from the surface of Pd and Pt. [40]Recently, Xia group reported the efficient removal of capping agents and Br ions on the Pd surface via the direct heating of the sample in pure water, besides maintaining the morphology and composition of Pd and enhancing its catalytic activity. [41]The effect of different cleaning methods on the catalytic activity of metal nanoparticles was discussed in this review, [42] but the effect of removing surfactant from the surface of Pt-group catalysts on the CO Oxid is rarely reported.For instance, Pt/C nanoparticles prepared using the water-in-oil microemulsion method with different ways for surfactant removal, including thermal method (i.e., two heating at 200 °C under H 2 (10 %)/Ar), chemical treatment in acetone, THF, or no treatment, but the effect of cleaning method on the CO-strpping was insignificant (Figure 4a-b). [43]However, ctalayst 1 obtained without chemical treatment showed earlier E Onset than others, possibly because of its higher ECSA.This study lacks surface and bulk analysis to confirm the removal of surfactants and to understand the effect of the cleaning method.So, it is imperative to study the electrocatalytic behavior of the Pt-group towards the CO Oxid using various cleaning methods for surface capping agents, besides using characterization tools to investigate maintained shape, composition, and catalyst-support interaction.

Effect of Catalyst Binder and Type of Working Electrode
Notably, the binder is essential to protect the catalysts from detaching, decrease the diffusion resistance of H 2 , reduce an ohmic (IR) drop, and enhance the electrolyte-electrode interaction, especially at proper concentration (i.e., Nafion 10 % v ratio of catalyst ink). [44]The effect of binder type and concentration on the electrocatalytic CO Oxid of Pt-group catalysts has not yet been studied.Few studies report the effect of Nafion concentration (i.e., 0-80 wt.%) on the HER activity of commercial Pt/C catalysts in alkaline electrolytes, and they found a significant difference, and the activity changed significantly (i.e., the HER activity decreased with increasing Nafion loading and with an optimal amount of 25 wt.%). [45]he effect of the working electrode on the electrocatalytic CO Oxid is also not studied, and previous articles mainly reported the glassy carbon electrode as the working electrode without using other electrodes such as metal foam/sheet (i.e., Ni, Co, Ti, Pt), [46] carbon (i.e., carbon cloth, carbon foam carbon paper, and carbon fleece). [47]These electrodes have a greater electrical conductivity and surface area than glassy carbon and their interconnected porosity can endow the dispersion of catalysts ink and enhance support interaction during the CO Oxid .Meanwhile, their low cost and ease of fabrication are important merits for sustainability and practical applications.Another crucial issue is that the majority of present methods (i.10a,48] However, hydrothermal, CVD, and electrochemical deposition allowed the in situ synthesizing of Pt-group catalysts over various supports like Ni foam and carbon cloth sheet. [49]For instance, Pt single-atom captured by NiCo-layered double hydroxides on Ni foam was synthesized via the electrochemical-deposition method, which enhanced the HER and glycerol electrooxidation more than Pt/C. [50]Thereby, the rational design of Pt-group catalysts over solid substrates like metal foam/sheet (i.e., Ni, Co, Ti, Pt), [46,51] carbon (i.e., carbon cloth, carbon paper, and carbon fleece) [47] can pave the way for practical applications and also eliminating the need for binders.

Thermodynamics and Kinetics of CO Oxid
In the alcohol fuel cells, CO is highly poisoning for Pt-group active sites even at small concentrations (i.e., 10-100 ppm), and CO Oxid is a thermodynamic uphill reaction (ΔG > 0), originating from the durability of adsorbed CO molecules and the inherent mechanism of CO Oxid and the relatively great enthalpy of COadsorption on the catalysts surface, (i.16c] Meanwhile, sufficient positive potential is also required for promoting the CO Oxid , which merely begins at a substantial rate at potentials (> 0.6VRHE).Thereby, in the experimental CO Oxid , Langmuir-Hinshelwood kinetics (E a = 75.4kJ mol À 1 ) is highly favored to describe the process than a power law expression (E a = 90.6 kJ mol À 1 ). [53]The high potential is needed to not only induce the activation and dissociation of CO but also to generate abundant active oxygenated species (i.e., OH), which react with adsorbed CO molecules to form CO 2 , H + , electrons, and two free Pt sites through two Faraday/mol of reaction.So, the availability of active sites, CO adsorption, and ease of formation of OH species are critical factors in the thermodynamics and kinetics of CO Oxid .Notably, CO, with its linear geometry, can coordinate vertically via a carbon atom to three sites on the Pt surface (i.e., top, bridge, and hollow sites) (Scheme 1), and their electronic densities could be investigated to determine the parameters affecting the kinetics and energetics of CO Oxid .The CO diffusion or coordination between Pt-group elements and CO molecules usually results in peak multiplicity (i.e., different peaks at different potentials) found in CO Oxid or CO-stripping.[43] Copyright (2007), with permission from Elsevier).(c) CVs of CO-stripping on Pt UME in 0.5 M H 2 SO 4 solution at 0.1, 0.5, 1, 2, and 3 Vs À 1 .(d) Separated current 1 st peak (i) and 2 nd peak (ii).Linear relationship between (e) 1 st peak j and ν 1/2 , and (f) 2 nd peak j and ν 1/2 .(Reproduced from ref. [60] Copyright (2016), with permission from American Chemical Society).

Assessment of CO Oxid Parameters CO Surface Diffusion on Catalysts
17a] An early study on the CO diffusion on an ordered Pt (111) electrode in 0.1 M HClO 4 proposed that the CO Oxid arises from the nucleation and growth at the island edges, as evidenced in the island formation, which survives low CO coverages. [54]Also, CO ads is a motionless molecule and its oxidation occurs when domains of adsorbed oxygen-donors grow from the vacancies of a CO adlayer. [54]Meanwhile, more coverage-dependent CO patches are gained in the presence of coadsorbed H 2 , which eases the electrooxidation more than the CO islands formed at intermediate coverages by earlier stripping. [54]The simulations of CO stripping voltammogram studies showed the significant effect of CO-diffusion over supported Pt (111) and Pt (100) on the CO Oxid (i.e., current, potential, and peaks position/number). [55]Also, the CO Oxid mechanism was proposed, where CO diffuses from ( 111) to (100) facets.The current profile of CO Oxid appeared with two separated peaks at a sluggish interface diffusion, whereas the two peaks were overlapped, and the (100) facet was more active than the (111) facet. [55]In this case, the CO Oxid rate is comparable to the CO interface diffusion.In contrast, the shoulder observed occurs due to the reaction on (111) sites.A single narrow peak in the CO stripping voltammogram was observed at quick CO diffusion, and the stripping of both facets would happen simultaneously.The same results in terms of the shape of the peaks were also observed on unsupported Pt aggregates, [56] and the simulated studies were in agreement with the experimental data obtained using Pt nanoparticles [57] and Pt nanoparticles with a clean surface and different ratios of ( 100) and ( 111) facets. [58]A positive shift of the CO Oxid peak was noticed when the size of Pt nanoparticles decreased especially between 1 and 4 nm. [59]he CO Oxid voltammogram with two distinct peaks was observed on the Pt ultra-microelectrode even at the high sweeping rate of 3 V/s (Figure 4c), and the I CO of the 1 st and 2 nd peaks was proportional to the υ and υ 1/2 , respectively (Figure 4d). [60]The 1 st peak originated from the oxidative desorption of the CO ads at the Pt/electrolyte interface, and the 2 nd peak was due to the surface mobility of either O 2ad s or CO ads at the adjacent Pt surface.However, based on the linear relationship of I Co vs. υ 1/2 (Figure 4e-f), the calculated surface diffusion coefficient of 6.8×10 À 10 cm 2 /s, was close to the mobility of O 2ads and greater than that of CO ads (3.6×10 À 13 cm 2 /s) that was estimated previously using EC-NMR, [61] which prove the CO Oxid is a bimolecular reaction comprising the surface mobility of O 2ads on Pt surface.However, this clarification is uncommon because CO Oxid occurs on Pt-group catalysts between CO ads and OH ads at defect sites, but the higher diffusion coefficient for O 2ads in the form of OH ad is interesting and can raise a lot of questions about the reaction mechanism and CO diffusion.
Although various studies were conducted to study the CO diffusion and kinetics of the Pt-based catalysts, [62] further in-situ analysis is required to explain and understand this behavior, and also further studies of supported Pt-group catalysts with different shapes are still needed.This review summarized the utilization of various electrochemical scanning probe microscopies (i.e., EC-STM, EC-AFM, and SECPM) in electrochemical applications and CO oxidation. [63]Notably, 13 C nuclear magnetic resonance (EC-NMR) spectroscopy was used for the first time to evaluate the CO diffusion on commercial Pt-black nanoparticles at the solid/liquid interface at 253-293 K, which showed the CO diffusion factors matched Arrhenius behavior. [61]Also, the CO diffusion mechanism was attributed to the interchange between various CO coverages originating from the chemical potential gradient, and the liquid electrolyte decreased the CO diffusion relative to that observed on surfaces of bulk metals under UHV conditions. [61]Later, the 13 C EC-NMR was used to investigate the CO diffusion on Pt and Pt/Ru nanoparticles with different diameters under temperatures of 253-293 K, in liquid electrolytes.The diffusion and chemisorption energies were affected significantly by the sizes. [64]Mainly, the CO surface diffusion was very quickly considered as the rate-limiting parameter in CH 3 OH reactivity on Pt nanoparticles, and the CO diffusion rates were enhanced on PtRu and with a correlation between the Fermi level local density of the 2π* molecular orbital of adsorbed CO and the activation energy for surface diffusion. [64]he way in which CO occupies and detaches from exact Ptbased catalyst surface sites is important to get insights into the dynamics and mobility of CO ads and their oxidation. [28]This was studied by observing the potentiostatic growth of dissimilar coverage CO adlayers at different doses θ CO occurred for several CO dosing times (θ CO ), with time (i.e., 150 sec) on cubic and octahedral Pt nanoparticles in 0.5 M H 2 SO 4 electrolyte.
The results showed the slight shift of CO ads molecules towards only low coordination sites on octahedral Pt nanoparticles, while on cubic Pt nanoparticles, adsorbed CO acts as an immobile species, low coordinated sites, and (100) terraces are filled consistently and concurrently (Figure 5a-b). [65]However, during CO Oxid , the CO ads perform as an immobile species on octahedral and cubic Pt nanoparticles after completing the adlayer, regardless of whether the CO Oxid occurs in a single step or in a sequence of various potential steps (Figure 5c-h). [65]The in-situ FTIRS demonstrated that the CO ads stretching frequencies depend on θCO (achieved by partial stripping from a saturated CO adlayer or by direct dosing) for various 12CO/13CO mixtures on Pt (111). [54]The early study utilized in-situ scanning tunneling microscopy (STM) coupled with infrared reflection-absorption spectroscopy (IRAS) to unravel the 3D structure of CO adlayers on Pt (111) in an acidic electrolyte, besides providing a detailed structural image of this electrochemical interface. [66]Likewise, Scheme 1.The CO adsorption on Pt electrocatalysts.the structures of a CO adlayer on the Pt(100) electrode surface in 0.1 M HClO 4 electrolyte saturated with CO were observed with a clear sharp peak (Figure 6a). [67]That was further studied by the in-situ STM, which displayed the well-ordered structures of the CO adlayer with CO coverages of (n-1)/n, and it was dynamically diverse with the potential, CO concertation, and the CO partial pressure in solution. [67]The STM image demonstrated the ordered structure of the CO adlayer on Pt(100), as confirmed by the obvious bright spots, and it was allied along the close-packed atomic row [01 %1], and brightness changed within each CO molecular (Figure 6b-c).The pressure led to compressed structures of the CO adlayer formed on Pt(100), and unique structures were observed in the vicinity of monoatomic steps in N 2 À and 1 % CO/He-saturated HClO 4 , as evidenced in the STM images of the Pt(100) obtained under different potential and CO %. [67] These results were confirmed later on h-BN/Pt (111) under pressure from UHV to 300 mbar CO using dynamic STM imaging, which showed the adsorption of CO adlayers on Pt(111) but confined under h-BN over layers at nanoscale and in near ambient pressure CO atmospheres in line with the DFT simulation. [28]This is important for the fundamental understanding of CO adsorption chemistry confined to nanospace and for providing precise guidelines for the rational design of Pt-based catalysts confined 2D supports.Both STM, subtractive normalized interfacial Fourier transform infrared spectroscopy (SNIFTIRS), and DFT revealed that the finite size effects induce local compressive strain, which adjusts the CO chemisorption properties of Pt sML /Pd(100)in the lack of coherent strain in Pt/Pd. [68]These results are important because the size effect was widely ignored in other reports.Notably, some theoretical and simulation studies investigated the size-selective kinetic characteristics of CO Oxid on multiscale nanostructured Pt/ GC involving nanodisks of 120 nm and nanoparticles of 6 nm, which showed their dissimilar CO stripping voltammogram features in line with the experiments result. [69]This originates from the altered CO surface diffusion induced by the size and initial configuration of CO.Therefore the particle size, crystal facets, and morphology of Pt-based catalysts are the main factors determining the CO diffusion and CO Oxid even features and activity.

Effect of Catalyst Size
The size of the electrocatalysts also affects the CO diffusion and subsequent CO Oxid kinetics, as well as the shape of voltammogram properties, but is rarely studied.Usually, smaller sizes possess higher surface area, accessible active sites, and lower mass, which are important merits for promoting the CO Oxid , but decreasing the size of the electrocatalysts did not essentially enhance the CO Oxid and electrocatalysts with larger sizes were more active than those of smaller sizes, and other factors like support, shape, and reaction conditions may affect. [70]An early study showed that the CO Oxid in H 2 SO 4 electrolyte on Pt nanoparticles with different loadings (10 to 78 wt %) and sizes (1-6.5 nm) supported on a glassy carbon electrode revealed dissimilar voltammogram features (i.e., number of peaks) and differing kinetics (i.e., E Oxid and E Onset ) (6e-m). [71]Mainly single CO Oxid peak was observed using 10, 20, and 30wt % Pt, and two peaks for 30-78 wt % of Pt besides lower E Oxid /E Onset and lower I CO (Figure 6d-l); thus, higher loading with larger particle size accelerates the CO Oxid kinetics but reduces the I CO . [71]Also, unsupported Pt black catalysts displayed the lowest E Onset , with a neglected shoulder beside the main CO Oxid peak.This implies the occurrence of CO Oxid at the lower potential on terrace sites is the main at larger nanoparticles; meanwhile, at smaller particle sizes, the reaction CO Oxid takes place on or near edge and corner sites because the number of edges and corner sites increases with increasing particle size.Contrarily, a positive shift of the CO Oxid peak was noticed when the size of Pt nanoparticles decreased, especially between 1 and 4 nm, due to the limited CO ads mobility on small nanoparticles (up to 2 nm), [59] due to the co-relation between the CO ad diffusion coefficient and particle size; Pt nanoparticles with particle size over 2 nm, CO ad diffusion increased.Notably, the studied particle sizes are close to each other and were relatively low; meanwhile, the electronic effects and particle size distribution should be considered because both are particle size dependent.A series of studies conducted the CO Oixd using various sizes of Pt nanoparticles supported on carbon (Pt/C), [72] which showed that the CO Oxid voltammogram features could be used as a probing technique for the particle diameter of Pt/C nanoparticles distribution.The multiple CO Oid peaks are attributed to the difference in the particle size of Pt/ C, which was not in line with the previous study that linked the difference with the dissimilarity in the terraces and edges of Pt/ C. [71] Moreover, the CO Oxid ensues at the bigger and aggregated Pt/C nanoparticles, yielding a lower overpotential peak than that at distributed nanoparticles.Using FTIR spectroscopy and double-potential step measurements, the CO Oxid on Pt nanoparticles (0.5 to 5 nm) supported on low surface area (1 m 2 g À 1 ) carbon are controlled by the different sizes rather than dissimilar adsorption sites. [73]Also, the CV voltammogram had two peaks, due to the stepwise CO Oxid from "large" (~3.6 nm)  (d,g) successive E steps for a partial Co oxid (sequence 0.10 V!0.65 V!0.10 V!0.65 V). (e,h) multiple steps of CA after j subtraction at 0.1 V from (d,g).(Reproduced from ref. [65] Copyright (2017), with permission from American Chemical Society).
and "small" (~1.7 nm) Pt nanoparticles, which allowed the isolation between the contribution of distinct sizes as observed in their infrared and CV.This implies the significant role of insitu analysis in understanding these phenomena; however, the difference in the diameter of such studied Pt nanoparticles was insignificant.
The DEMS coupled with CO-stripping was used to study the effect of size on the CO Oxid in H 2 SO 4 electrolyte using Pt nanodisks and/or supported on glassy carbon. [74]This is driven by the colloidal lithography method, which allowed the formation of Pt nanodisks (94, 118, 140,143,147 nm in diameter) mixed with or without Pt nanoparticles (3.9, 4.3, 6.5, 7.3 nm) and the voltammogram features were altered significantly with the particle size.The CV voltammogram of the catalysts that contain nanodiscs (140, 143, 147 nm) with nanoparticles (3.9, 4.3, and 6.5 nm) showed two peaks with a positive shift in the E Oxid , while the catalysts that had individual Pt nanodiscs (94 and 118 nm) or nanoparticles (7.3 nm) revealed only one peak with a negative shift in the E Oxid , which agree with the mass signal observed in DEMS. [74]This implies the significant effect of smaller particle sizes, despite their lower surface area, on the enhancement of the CO Oxid ; meanwhile, the 2 nd peak at high potential could be used to detect small-size Pt nanoparticles.Therefore, Pt-group electrocatalysts with smaller sizes were more active than larger ones; however, previous studies on the effect of particle size ignored the surface facets, shape, and electronic structure, which makes the effect of size somehow arbitrary because diverse particle sizes usually have different surface facets, shapes, defects, etc. which also endow the CO Oxid activity.100) electrode (Reproduced from ref. [67]  Copyright (2013), with permission from Royal Society of Chemistry).CO-stripping for a series of Pt/C deposits (d) 10, (e) 20, (f) 30, (g) 40, (h) 50, (i) 60, (j) 70, (k) 78 wt.% Pt and of (l) Pt black at 10 m/Vs in 0.5 M H 2 SO 4 .(Reproduced from ref. [71] Copyright (2004), with permission from American Chemical Society).

Effect of Catalyst Shape
The morphology of the Pt-based group is among the most critical factors for the enhancement of the CO Oxid , due to their multiple surface facets and surface defects, which can provide diverse sites for CO ads . [75]Pt nanostructures with different surfaces (i.e. the ratio of 100/111 facet) were prepared by various methods (i.e., water-in-oil microemulsion and polyacrylate method) to understand the effect of surface facets on the CO Oxid [57] in a 0.5 M H 2 SO 4 electrolyte after being saturated with CO at 0.1 V. Spherical Pt nanoparticles (3 nm), nanocubic (12 nm), nanooctahedral (10 nm), and nanohexagonal (9 nm) showed dissimilar voltammogram features in the absence of CO, with two mean peaks in all catalysts attributed to the adsorption/desorption of H on edges mimicking the (110) facet and atomically ordered sides having a (100) orientation (Figure 7a-d). [57]However, Pt nanocubic, nanooctahedral, and nanohexagonal nanostructures revealed additional broad peak attributed to anion adsorption/desorption on (111) facet; the difference fraction of the (110/111) facet could be seen in the dissimilar intensity, and broadening. [57]The CV of CO Oxid showed only one sharp peak on Pt nanoshere and two peaks on other shapes, but the I CO was in the order of Pt nanoshere > nanocubic > nanohexagonal > nanooctahedral, due to their variation in the ratio of surface facets in line with previous results in. [76]The 1 st peak at lower potentials originated from the oxidation of CO chem adsorbed on the (111) facet.
Meanwhile, the 2 nd peak at higher potentials is attributed to the oxidation of CO chem adsorbed on low-coordination sites, but the intensity of the 2nd peak increases with an increasing fraction of (100) facet.The @I/@E vs. E profiles did not show an inflection point in the case of nanospheres and nanocubes, indicating their peak 1 is a single voltammetric feature, while in the case of nanohexagonal and nanooctahedral, indicating the presence of two overlapped and closely located peaks in peak 1 (Figure 7e). [57]Considering these results, the CO Oxid , entails four features of oxidation peaks due to CO Oxid on the (111) facet (peak 1 st ), on low-coordinated surface sites (2 nd ), and on the (100) facet (3 rd ) (Figure 7e).However, the size of the surface domains should be considered in the CO Oxid , which was investigated later on as unsupported Pt nanospheres, nanocubes, nanooctahedrons/tetrahedrons, and nanocuboctahedrons with various sizes of surface facets in H 2 SO 4 electrolyte. [77]he nanostructures revealed different oxidation voltammogram features that depended on the number of "reconstruction CV cycles" that altered the nature and relative amounts of ( 100) and ( 111) surface facets, owing to the incessant formation/ reduction of Pt oxides/hydroxides.Self-standing PtPd hollow nanocubes formed via (PtPd HNCsÀ E) formed via galvanic replacement and co-reduction enhanced the electrocatalytic CO Oxid significantly with an earlier E Onset of 0.77 V relative to PtPd nanocubes (0.79 V) and commercial Pt/C (0.84 V) in 0.1 M KOH, due to porous morphology and alloy effect. [78]Likewise, Fractal PtPdCu hollow sponges synthesized via galvanic replacement and chemical etching enhanced the COOxid with a lower E Onset (0.76 V) than those of PtCu hollow sponges (0.841 V) and PtPdCu networks (0.846 V), and Pt/C (0.851 V) in 0.1 M HClO4 electrolyte, due to porous shape, higher ECSA, and accessible sites, which ease directed mass-and electron-transfer. [79]This study is of great importance because, the electrocatalytic CO Oxid of porous ternary Pt-based catalysts is not studied enough.However, the effect of shape remains ambiguous due to the lack of systematic investigation of the EIS data that are important to understand charge mobility and electrolyte interaction, and the studied PtPdCu was used without support.

Effect of Electrolytes
There are various aqueous and non-aqueous electrolytes for multidisciplinary electrocatalytic applications.The ability of the electrolyte to act as a reservoir for dissolving and storing the reactants (i.e., CO and OH) and promoting their adsorption and activation/dissociation on the catalyst surface is of great importance in the enhancement of the CO Oxid .Meanwhile, electrolytes with abundant protons (H + ) (i. e., acidic electrolytes) and ease of splitting to generate active OH species from the electrolyte on the catalyst under low potential are also crucial factors for endowing the CO Oxid .There are wide ranges of aqueous electrolytes entailing acid (i.e., H 2 SO 4 , HClO 4 ), alkaline (i.e., NaOH and KOH), neutral (i.e., PBS), and carbonate-based (i.e., KHCO 3 , H 2 CO 3 , and Na 2 CO 3 ). [80]80a] Carbonate-based electrolytes have a higher ability to store/ dissolve CO 2 and high proton donation; however, some reported greater Faradic efficiency and production rate in K 2 HPO 4 solution than in K 2 SO 4 , KClO 4 , and KCl. [80]The acidic electrolytes have abundant protons and easily dissociate to generate OH or get H + to form hydronium under low potential, which facilitates the CO Oxid kinetics and eases the desorption of intermediates and products.The alkaline electrolytes with their low H + , and high pH can ease the leaching of catalysts.Meanwhile, carbonate-based electrolytes are imminent with their outstanding ability to adsorb, dissolve, and store CO besides their great proton donation.Non-aqueous electrolytes such as ionic liquids and organic solvents (i.80b] The effect of electrolytes on the electrocatalytic CO Oxid activity was rarely reported and not studied enough, and previous studies focused merely on a few electrolytes (i.e., H 2 SO 4 , KOH, and KHCO 3 ).We have studied the effect of electrolytes on the electrocatalytic CO Oxid over self-standing Pdbased nanostructures, but the activity was in the order of H 2 SO 4 > KOH > NaHCO 3 , as shown in the higher I Co and lower E Oxid . [81]he ability of electrolytes to adsorb anions in the supporting electrolytes also led to different voltammogram features and activity.16c] This is due to the high ability of H 2 SO 4 to strongly adsorb anions (i.e., bi-sulfate), while HClO 4 is a non-adsorbing anion.However, the effect of adsorbed anions on the voltammogram features and the CO Oxid mechanism is not studied enough, and their effect in accordance with CO diffusion, catalyst active sites, surface facets, and shape is still to be demonstrated.A review published long ago summarized the utilization of FTIR, STM/AFM, and SXS to demonstrate the effect of the adsorption of intermediates on the reaction pathway and kinetics on Pt catalysts, which concluded the garter adsorption strength of (bi)sulfate on Pt (111) than on ( 100) and ( 110) facets. [82]This was also proved later for Pt nanoparticles supported on carbon (with a low surface area) using in-situ FTIR spectroscopy and also showed the effect of Pt size on the E Oxid and E Onest (i.e., a positive shift at a larger size). [73]However, exploring the effect of electrolytes pH, type, and anion/cations (i.e., Br, Na + , K +) on the CO Oxid activity is still important.Also, acidic electrolytes were always better than alkaline and neutral ones.

Best Practice in Benchmarking the CO Oxid
To ensure the best practice for testing and optimizing the CO Oxid on Pt-group catalysts, we have carried out several studies, including tailoring shapes, compositions, support, and electrolytes pH.

Formation of Porous Catalyst
We synthesized porous spatial Pt nanodendrites (11 nm in diameter) by the ultrasonic irradiation method and 1D Pt nanochains (6 nm in diameter) by the chemical reduction method (Figure 8a). [83]The electrocatalytic CO Oxid of unsupported Pt nanodendrites and nanochains were compared with 0D commercial Pt/C (3-5 nm in size) in HClO 4 , KOH, and NaHCO 3 electrolytes, but Pt nanodendrites were the most active (i.e., higher I CO and lower E Oxid /E Onset ), and durable than nanochains and Pt/C in all electrolytes (Figure 8d-e, g-h,j-k). [83]This is due to the 3D porous structure with multiple arms that provide higher ECSA and abundant sites for adsorption and activation of the reactants.In terms of the electrolytes, the performance of Pt nanodendrites was in the order of HClO 4 > NaHCO 3 > KOH under the same reaction conditions.The EIS data and their fitting revealed lower R s and R ct besides a higher CPE of Pd nanodendrites in all electrolytes than nanochains and Pt/C, which indicates the superior charge mobility and better electrolyte-nano dendrites interaction during CO Oxid (Figure 8f, i, l).The CO Oxid activity of our developed Pt nanodendrites [83] was superior to previously reported PtNi multicubes, [84] PtPd Nanodendrites, [85] Pt dendrimer-encapsulated nanoparticles, [86] Pt(110)À Ru [87] Pt(FAM), [56] and ordered Pt(111) [88] tested in various electrolytes.This study indicated that porous 3D Pt nanostructures are preferred over 1D and supported 0D nanostructures, which may pave the way for further utilization of such shapes at low mass in fuel cells and evading COpoisoning. [83]ecently, we have synthesized and tailored the morphology of unsupported Pd nanocrystals (i.e., nanocubes and nanosponge) with a clean surface and high surface area using the chemical reduction method with NaBH 4 at 0 °C and ascorbic acid at 25 °C, respectively (Figure 9a-c). [81]The electrocatalytic CO Oxid activity and stability of Pd nanocube nanosponge were systematically investigated relative to commercial Pd/C in different electrolytes (i.e., H 2 SO 4 , NaHCO 3 , and KOH), and the activities depended on the shape of Pd and electrolyte pH. [81]nterestingly, self-standing Pd nanocubes and nanosponges were more active and durable than commercial Pd/C in all electrolytes in terms of higher I CO and lower E Oxid and E Onset (Figure 9d-e, g-h, j-k), originated from the higher ECSA of Pd nanocubes and superior CO ads charge.Pd nanocubes and nanosponge were close, but nanocubes were the best in H 2 SO 4, and nanosponge was more active in other electrolytes.All nanostructures showed only one CO Oxid peak in all solutions, but only nanocubes showed two CO Oxid peaks in KOH (Figure 9h).The CO Oxid activity of Pd nanosponge was at least three times higher (i.e., I CO ) than that previously reported for polycrystalline Pd in H 2 SO 4 , [89] PtPd(50 %) nanodendrites in H 2 SO 4 , [90] and PdÀ Pd(4 : 1)/C in KOH. [91]The detailed EIS analysis demonstrated semicircle lines but with a lesser diameter for Pd nanocube and Pd nanosponge than Pd/C in all solutions, which implies better electrolyte-interaction and higher charge mobility (Figure 9f, i, l), as seen in the EIS data fitting, which clearly warranted the lower R s and R ct besides the greater CPE impedance of Pd nanosponge and nanocubes relative to Pd/ C. [81] These results display the superiority of unsupported Pd and Pt porous anisotropic nanostructures than that of supported spherical-like Pd/C and Pt/C in different electrolytes, but acidic electrolyte was better.Also, 3D porous nanosponge was the most preferred structure, which may open new gates for understanding the electrocatalytic activities of Pt-group catalysts and preventing their CO poisoning in fuel cells, as shown in the higher ethanol oxidation activity of anisotropic Pd nanostructures than Pd/C in our recent report. [92]owever, other Pt-group elements (i.e., Au, Ag, Ir, and Rh) should be studied due to their importance in multidisciplinary electrocatalytic reactions.

Alloying Pt with Other Metals
To check the effect of composition, we fabricated bimetallic porous 3D porous PdCu (3.21 nm), PdMn (7.33 nm), and AuPd (7.96 nm) nanosponge with clean surfaces (i.e., no capping agent or surfactant) and average pore diameter (2-10 nm) by the aqueous-phase ice-reduction method in presence of NaBH 4 method. [93]The electrocatalytic CO oxid performance of Pd alloys was superior to commercial Pd/C catalysts in H 2 SO 4 , KOH, and NaHCO 3 , but PdCu had the utmost CO Oxid activity in all electrolytes.The I CO of porous PdCu nanosponge was more than 1.74, 1.14, and 2.63 times higher than those of PdMn, AuPd, and Pd/ C nanosponges in H 2 SO 4 , KOH, and NaHCO 3 , respectively, besides of great stability for 1000 cycles. [93]Also, the EIS data demonstrated the lower charge transfer resistance and better electrolyte interaction of PdCu, PdMn, AuPd, and Pd/C than Pd/ C, as shown in the lower R s , R ct , and higher CPE.Notably, PdCu nanosponge [83] outperformed previously reported PtPd nanodendrites in KOH, [85] PtPd nanodendrites in H 2 SO 4 , [90] and AuPd/ C in H 2 SO 4. [94] This is due to the greater strain effect and oxophilicity of, which enhance H 2 O activation/dissociation at a lower potential needed for accelerating the CO Oxid .
To get more insights into the composition effect and whether strain or synergetic effect is preferred for the CO Oxid , we have synthesized unsupported foam-like PdFe, PdCo, and PdNi nanocrystals via the coalescence and growth mechanism mediated by NaBH 4 for the CO Oxid .PdFe, PdCo, and PdNi had porous foam-like nanostructures with 3D interconnected pores (Figure 10a-c). [95]he composition of this obtained alloy was found to be PdFe (Pd/Fe are 83.46/16.54at.%),PdCo (Pd/Co are 54.09/45.91 at.%), and PdNi (Pd/Ni are 63.05/36.95at.%), respectively.The XPS analysis revealed the modulation of the d-band level of Pd by Ni, Co, and Fe, which is important for tailoring their catalytic merits, as seen in the higher CO Oxid performance of Pdbased alloys than supported Pd/C catalysts in KOH, HClO 4 , and NaHCO 3 electrolytes.Notably, all the catalysts had the same CV voltammogram features (i.e., reversible oxidation/red peaks with sharp ones in the forward direction) in KOH and NaHCO 3 , but PdNi and PdCo showed additional small peaks in the forward direction in HClO 4 (Figure 10d-e, g-h, j-k). [95]Meanwhile, the I CO of PdFe was more than 2.18, 4.35, and 1.56 times higher than that of PdCo, PdNi, and Pd/C in KOH and HClO 4 media, while PdCo was the highest in NaHCO 3 solution along-side greater durability for 1000 cycles.This is due to the higher strain and alloying effect, which downshifted the d-band level of Pd and subsequently accelerated the generation of active OH species at low potential and tolerated the intermediates. [95]herefore, PdFe foam-like [95] was more active than those PtRu (1 : 1) in HClO 4 , [96] PdÀ Pd(4 : 1)/C in KOH in KOH, [97] and others.The linear relationship between I CO and υ 1/2 indicates the diffusion-controlled CO Oxid process on porous Pd alloy nanostructure.The systematic EIS Nyquist plots obtained in all electrolytes showed semicircle lines but with a smaller diameter for PdFe than those of PdCo, PdNi, and commercial Pd/C catalysts (Figure 10f, i, l), indicating better charge transfer and electrolyte-electrode interaction as seen in the fitting of EIS data that revealed the lower R s and R ct along grater CPE on PdFe than its counterparts.Notably, when we used carbon paper as a working electrode instead of the traditional glassy carbon electrode, the CO Oxid of the I CO of PdFe nanosponge enhanced by nearly 1.5 times (7.6 mA/cm 2 ) besides a significant decrease in the E CO and E Onset (Figure 10m-o), due to the higher ECSA and electrical conductivity and better electrolyte-electrode interaction of carbon paper.Thus, carbon paper should be used as a working electrode for Pt-group catalysts in CO Oxid and other electrochemical oxidation reactions.This study unraveled the significant alloying effect and porous structure on promoting the CO Oxid in different electrolytes besides the effect of working electrode type and it may allow their utilization in real fuel cell applications, as shown in superior electrocatalytic ethanol oxidation on such alloys than commercial Pd/C catalysts in our recent studies. [98]e have prepared porous PtPdCu nanodendrites (35 nm in size) and atomic content of Pt/Pd/Cu (52/29/19 at.%) by the ultrasonic irradiation method, which enhanced the CO Oxid significantly compared to PtPd nanodendrites, PtCu nanodendrites, and commercial Pt/C in H 2 SO 4 . [99]The I CO of PtPdCu nanodendrites (10 mA/cm 2 ) was higher than those of PtPd nanodendrites (~7.8 mA/cm 2 ), PtCu nanodendrites (~5 mA/cm 2 ), and commercial Pt/C (~3 mA/cm 2 ) in H 2 SO 4 electrolyte, owing to the ternary metal composition, porosity, higher surface area, and modulated d-band of Pt with Pd/Cu.Notably, the electrocatalytic CO Oxidation activity of porous multimetallic Pt-based catalysts is rarely reported as a function of support and electrolyte pH, which could affect the catalytic merits substantially.Therefore, self-standing porous ternary Pt-based catalysts, especially with transition metals (i.e., Cu, Co, and Fe), were more active and durable than those of binary and mono metallic systems and supported commercial Pt/C.Also, using transition metals with high synergism is more favored for CO Oxid .

Using Carbon-Based Support
To study the effect of support, we have synthesized 1D carbon nitride nanorods doped with Pt and Pd (PtPd/CNs) by mixing metal salts and melamine ethylene glycol under stirring, followed by the addition of NaNO 3 with HCl to induce the protonation and subsequent deamination step, followed by the annealing under N 2 (Figure 11a). [100]PtPd/CNs had well-defined  nanorods (~94 nm in longitude and ~11 nm in width), with a great surface area (155.2 m 2 g À 1 ) and codoped with Pt/Pd (1.5 wt%) (Figure 11b).The electrocatalytic CO Oxid activity of PtPd/CNs (I CO = 14.75 mA cm À 2 ) was higher than that of commercial Pt/C and metal-free CN by 2.01 and 23.41 times, respectively, in terms of the I CO besides a lower E CO and E Onset in KOH (Figure 11c). [100]The I CO on PtPd/CNs (14.75 mA cm À 2 ) was superior to previously reported porous PtPdRu nanodendrites (12 mA cm À 2 ) and PtPd alloy nanoparticles (11.2 mA cm À 2 ), and Pt nanowires (0.12 mA cm À 2 ) besides higher durability. [101]The linear relationship between I CO and υ 1/2 designates the diffusioncontrolled CO Oxid process on PtPd/CNs.This study is critical not only because it endowed the utilization of CN as a support for Pt and Pd in CO Oxid but also showed the higher activity of metal dopants relative to Pt nanoparticles, which may pave the road for using Pt-group in the form of dopants with lower mass loading in electrocatalytic applications.PtPd nanoparticles with heavy d-π overlap and high strength of PtÀ C bonding in sp 2 bonded graphitic carbons nanofibers (PtPd/aÀ CNF) and a low level of defects synthesized via the interfacial method showed higher ECSA and better electrocatalytic activity than that of PtPd nanoparticles over acidized CNFs and commercial Vulcan-XC 72R carbon. [102]The heavy d-π overlap sheds could enhance the CO Oxid activity of Pt-group catalysts especially on 1D carbon nanostructure, due to their strong bond and short pathway for electron transfer.

Replacement of Carbon Support with Others
Later, we used 2D exfoliated Ti 3 C 2 T x MXene (T X = OH, O, and F) nanosheets as a support for endowing the in-situ formation of Pd nanoparticles with a size of 10 nm and mass loading (2.5 Wt. %) ( Pd/Ti 3 C 2 T x ) (Figure 11d), [103] which enhanced the electrocatalytic CO Oxid (I CO = 0.318 mA/cm 2 and E Oxid = 0.9 V) than Pd-free Ti 3 C 2 T x beside a lower R s and R ct (Figure 11e-f).This study not only unveiled the utilization of Ti 3 C 2 T x as a support and as a reducing agent for the formation of Pd nanoparticles without a reducing agent, heating, and organic solvent, but also showed the possibility of using Ti 3 C 2 T x as a support for the CO Oxid . [103]However, Ti 3 C 2 T x should be used with other Pt-group elements to verify the feasibility of this concept and also to get more insights on the CO Oxid performance in the presence of Ti 3 C 2 T x; meanwhile, other MXene phases should be tried.
To check the effect of other supports on the catalytic merits of Pd nanoparticles, we have prepared a Co-metal-organic framework (MOF)-derived porous carbon nanostructure decorated with Pd nanoparticles (Pd/ZIFÀ 67/C) via coupling the microwave-irradiation method with annealing and chemical etching structure (Figure 12a). [104]This formed a 3D porous flower-like structure (Figure 12b) with interconnected pores composed of small nanosheets supported by Pd (6.8 nm) nanoparticles (Figure 12c) with atomic contents of 54.12, 8.02, 15.08, and 22.15 %, for C, N, Ni, and Pd, respectively. [104]The CO Oxid activity and stability of Pd/ZIFÀ 67/C were significantly greater than that of commercial Pd/C and Pt/C in different electrolytes, including higher I CO , lower E Oxid , and E Onset (Figure 12 e, g, i).The I CO of Pd/ZIFÀ 67/C was (4.2 and 4.4), (4.0 and 2.7), and (3.59 and 2.7) folds greater than that of Pd/C and Pt/C in HClO 4 , KOH, and NaHCO 3 respectively, due to the well-dispersed Pd nanoparticles and abundant NiÀ N x active sites, porous morphology, and higher electrical conductivity of ZIFÀ 67/C. [104]he lower diameter of the semi-circle lines of Nyquist plots of Pd/ZIFÀ 67/C, in all electrolytes than Pd/C and Pt/C, implying its lower charge transfer resistance and better interaction with the electrolytes, as also shown in the lower R s and R ct and higher CPE of Pd/ZIFÀ 67/C (Figure 12f, h, j). [104]Also, Bode plots displayed the minimal logarithm of total impedance of Pd/ ZIFÀ 67/C than Pt/C and Pd/C besides a phase angle of 67-80°, which designates that occurrence of CO diffusion with the adsorption process during CO Oxid .The relationship of I CO vs. v 1/2 was linear on all electrocatalysts, implying the diffusioncontrolled CO Oxid process, but the slope of Pd/ZIFÀ 67/C was larger, which implies its quicker kinetics.The I CO of Pd/ZIFÀ 67/C was superior to previously reported Pd or Pt-based electrocatalysts supported over carbon supports like PdAg/C in KOH, [105] 60 wt.%Pt/C in H 2 SO 4 , [106] and PtRu@hÀ BN/C in H 2 SO 4 . [107]ikewise, NiÀ MOF-derived porous carbon nanosheets (NiÀ MOF/PC) enriched with NiÀ N x sites supported by Pd nanocrystals (Pd/NiÀ MOF/PC) enhanced the CO Oxid performance more than those of Pd/NiÀ MOF/C (obtained without etching) and commercial Pt/C in HClO 4 , KOH, and NaHCO 3 electrolytes, respectively, [108] due to the higher porosity and NiÀ N x sites of Pd/NiÀ MOF/PC.Also, the voltammogram features (i.e., number of oxidation peaks, E Oxid , E Onset , I CO ) were electrolyte pH-dependent.Pd/ NiÀ MOF/PC showed the highest I CO , and the lowest E Oxid , E Onset , and impedance parameters (R ct and R s ) in the order of HClO 4 > KOH > NaHCO 3 . [108]These results clearly indicated that porous carbon support derived from MOF is preferred for enhancing the catalytic activity and stability of Pd nanoparticles than traditional activated carbon.
To check whether coupling two supports is more preferred than one support, we have prepared PdNiO nanoparticles (8 nm in size) supported on porous sponge-like CeO 2 /onion-like carbon nanostructures (PdNiOÀ CeO 2 /OLC) via the sol-gel and impregnation methods. [109]PdNiOÀ CeO 2 /OLC composed of PdNiO nanocrystals (2.54 nm) distributed over CeO 2 flower-like (20 nm) and OLC (30 nm) with overall porosity (0.30 cm 3 /g), and outstanding surface area (155.66 m 2 /g).These merits endowed the CO Oxid activity and stability of PdNiOÀ CeO 2 /OLC (i.e., higher I CO , lower E Oxid /E Onset ) than that of PdNiO/OLC, PdNiOÀ CeO 2 , and commercial Pd/C by more than 1.66, 1.88, and 2.12 times, respectively, in different electrolytes but the activity was in the order of HClO 4 > KOH > NaHCO 3 . [109]This is besides higher durability for 1000 cycles and better charge transfer and interaction with the electrolyte, as shown in the lower R s and R ct and higher CPE. [109]This is due to the electronic interaction of PdNiO with CeO 2 /OLC co-supports, which facilitates the CO adsorption/activation, besides activation/dissociation of H 2 O to form OH, resulting in higher kinetics. [109]This study implies that coupling two support CeO 2 /OLC is preferred for enhancing the CO Oxdia activity of PdNiO nanocrystals than one support OLC or CeO 2 or C. [109] Likewise, we prepared PdNiOÀ CeO 2 /CB with pore volume (0.29 cm 3 /g) and surface area (151.86 m 2 /g) by the sol-gel of carbon black (CB) with cerium salt to form (CeO 2 /CB) flake-like nanostructure, then impregnation with Pd/Ni precursors to induce their reduction to form PdNiO nanocrystals (4.9 nm) without any reducing agent, surfactant, or organic solvents (Figure 13a-c). [110]The I CO of PdNiOÀ CeO 2 /CB was significantly higher than those of PdNiO/CB, PdNiOÀ CeO 2 , and Pd/C by more than 1.52, 1.58, and 1.67 times in all electrolytes, besides a greater durability for 1000 cycles in different electrolytes, but with better activity in HClO 4 than in KOH, and NaHCO 3 (Figure 13d,f,h).This originated from coupling the unique physicochemical merits of CB and CeO 2 beside their interaction with PdNiO, which accelerates the generation of active OH needed for promoting the CO at low potential and maximizing utilization of Pd active sites.
The I CO vs. υ 1/2 showed the diffusion-controlled CO Oxid process on all catalysts but with a higher slope on PdNiOÀ CeO 2 / CB, implying its better CO-diffusion and quicker kinetics, as also shown in the lower semi-circle lines of PdNiOÀ CeO 2 /CB in Nyquist plots and its lower Rs and Rct beside higher CPE than its counterparts (Figure 13e, g, i).The I CO of PdNiOÀ CeO 2 /OLC and PdNiOÀ CeO 2 /CB were superior to previously reorted Pt/ SnO xin HClO 4 , [111] PtRu@hÀ BN/Cin H 2 SO 4 , [107] and Pd/CMK-3-R8-1500-10 in H 2 SO 4 , [112] and our Pd/ZIFÀ 67/C. [104]Pd/NiÀ MOF/ PC. [108] The study indicates the highest activity of PdNiO with two supporters CeO 2 with CB than one supports CeO 2 or CB.
Lastly, other supports like MOF, MXenes, CB, gCN, and CeO 2 were more favored for enhancing CO Oxid activity of Pt and Pdbased catalysts, and also the two supports were the best.However, using in-situ FTIRS, STM, and IRAS is urgent to decipher the effect of supports on the CO Oxid mechanism.

Measurements Consideration
To ensure best practice for optimization of the CO Oxid on Ptgroup catalysts, the experimental design (i.e., pre-treatment, cell design, Ohmic drop, and catalyst loading).The Ohmic drop should be considered because it can affect the E Oxid and position originating from the uncompensated solution resistance. [113]The typical loading of Pt-group catalysts ranged from 14 to 80 μg/ cm 2 , which can affect the ECSA, Ohmic losses, and I CO , so the loading should be optimized, mewanwhile, the EIS analysis and its data fitting should be done.The cell design can lead to different resistances (i.e., > 10 Ω), so the E Oxid peak could be displaced by � 10mV at nearly 1mA, so the area of the working electrode relative to cell volume should be considered.This also affects the potential window (i.e., higher potential may lead to one oxidation peak, while low potential leads to multiple peaks). [56]The initial CV cleaning cycles eliminate the impurities from the electrode surface, and activate the catalysts(i.e., alter the microstructure of the electrode), and subsequently affect the CO Oxid activity and features, as proved long ago. [114]The pretreatment of electrodes via annealing or oxide-annealing (COannealing) method could eliminate the surface capping agents (i.e., surfactants) and remove defects (i.e., high index hkl planes surfaces) [115] via rearrangements of atoms from the 1 st layer can result in dissimilar voltammogram features and higher CO Oxid performance.This could be explained according to the great CO ads energy that led to the expansion of the surface layer (i.e., 4 % at 0.05V), followed by the consequent shrinkage after COremoval by oxidation. [82]Contrarily, CO-annealing of Pt (100) single crystal generated less defects than those obtained by other methods. [116]Therefore, from the perspective of largescale applications, CO-annealing could be used as an effective way to remove the defects before conducting the electrochemical oxidation reactions and enhance the catalytic activity.However, this effect was carried out before oxygen reduction reactions [117] and it was not studied enough and needed further investigations.The detailed comparison between the electrocatalytic CO Oxid activities of Pt-group electrocatalysts using different factors (i.e., preeration methods, morphology, electrolytes) are listed in (Table S1).These parameters tailored the activity significantly (i.e., I CO and E Oxid ).

Conclusions
This review provides a deep analysis of the effect of Pt-group (i.e., nanosize, morphology, and support) and CO Oxid parameter (i.e., the effect of electrolyte, working electrode, and CO surface diffusion) besides the fundamental aspects (i.e., CO Oxid Process, mechanism, measurements, calculations, thermodynamics, and kinetics).Pt-group electrocatalysts were prepared in various shapes (i.e., nanodendrites, nanocubes, nanospheres, and octahedrons) sizes (1.5-50 nm), and compositions by different methods (i.e., chemical reduction, solvothermal, and template), which varied in their CO Oxid activity.Various metals (i.e., Ru, Fe, Mo, W, Sn, Co, Cu, Mo) were alloyed with Pt-group for enhanced CO Oxid , [118] but they showed lower activities than mono metallic nanoparticles (Table S1), so much effect needs more investigations.Various supports (i.e., CB, activated carbon, graphene, WO 2.75 , ZrO 2 , and CeO 2 ) [13] were used for enhanced CO Oxid , but carbon supports were better, and using two supports with Ptgroup catalysts outperformed one support.Mesoporous carbon (CMK-3-R8-1500-10) showed the highest activity (I CO = 145 mA/ cm 2 at 0.9 RHE ) [112] among all reported Pt-group electrocatalysts, owing to high surface area, mesoporoisty, and maximized Pd utilization.Our recent results warranted that coupling two supports (i.e., CeO 2 /OLC, CeO 2 /CB) is preferred over one support for promoting the CO Oxid activity and durability of PdNiO nanoparticles in different electrolytes. [110]Porous carbon derived from MOF is more effective than traditional activated carbon for enhancing the CO Oxid activity of Pd nanoparticles, due to porosity and additional MÀ NÀ C species.Modulation the d-band center of Pd with Co, Cu, Ni, Fe, Au, and Mn without support enhanced the performance significantly than Pd/C catalysts in different electrolytes, but it was better in KOH. [93,95]PdFe with a higher strain effect was superior to all other alloys. [95]Porous Pt nanodendrites and Pd nanocubes are the most promising in different electrolytes,due to higher ECSA and maximized atom utilization.Using carbon paper as a working electrode for porous PdFe alloy was better than glassy carbon electrodes.Carbon nitride and Ti 3 C 2 T X Mxenes are promising supports for Pt-group catalysts (i.e., Pt, Pd, and Au), due to their unique physiochemical merits and enhanced metal dispersion.
Although the observed progress in Pt-based electrocatalysts for CO Oxid , several barriers should be solved, like intolerable cost, earth-rarity, and high mass loading on supports (10-30 wt.%).This is besides the infeasibility of their fabrication approaches (i.16a] Besides, segregating metal atoms over support rather than their mixing atomically devalues their mass utilization. [84,119]To defeat such fences, several perspectives could be applied, as follows; The expansion of aqueous solution-based techniques to facilitate the rational design of in-situ supported noble-metal nanoparticles.This is through judiciously selecting proper reducing agents (i.e., boron hydride, ascorbic, and hydrazine) and capping agents (i.e., polyvinylpyrrolidone, cetrimonium chloride, and poloxamer-407) under ambient temperature (i.e., 25-40 °C) the presence of various supports characterized by outstanding ECSA (i.e., graphene, activated carbon, carbon nitride, carbon dots, and MXenes). [120]32a,48b,85] Direct synthesis of porous Pt-group over solid supports (i.e., carbon paper, metal oxide/hydroxide sheet, and carbon foam) is proposed bypass additional steps required for forming working electrodes.This in addition to maximizes the utilization of active sites and fosters electrolyte-electrode interaction.
119a] Another way that could enhance the CO Oxid , is tailoring the synthesis of Pt-based alloys by interfacial method, which produces alloys with heavy d-π overlap and robust PtÀ C bonding in sp 2 bonded graphitic carbon and with minimal defects, thereby enhances the CO Oxid activity and stabilizes Ptagainst aggeration and leaching in different electrolytes. [102]he deployment of porous Pt-based catalysts like foams, sponges, and aerogels achieved by crosslinking of Pt-group metals with CNTS, MOF, and graphene can maximize the utilization of metal active sites during the CO Oxid . [121]inally, the prospectus underscores the significance of coupling in-situ (i.e., operando, NMR, FTIR, and DRIFT) and exsitu electrocatalytic CO Oxid with computational simulation to investigate the effect of different factors on the CO Oxid process, thereby providing a holistic understanding and enabling targeted improvements.
Dr. Eid obtained a Bachelor's Degree in Chemistry from Al Azhar University, an MSc.from Helwan University in Egypt, and a Ph.D. in analytical chemistry from the Chinese Academy of Sciences in China through a scholarship from TWAS-UCAS in 2016.He published over 100 Q1 research papers, edited one book with the RSC (London), filled out 15 granted patents at the USA Patent and Trademark Office, and participated in 32 international conferences.His research interests include gas conversion reactions, heterogenous catalysts, electrocatalysts, and water splitting.

Figure 1 .
Figure 1.(a) Number of published papers in the last decade from the Web-of-Science with keywords "Electrocatalysts Fuel cells'' and ''Electrocatalysts electrochemical CO oxidation/stripping" and (b) the main focus of this review article.

Figure 2 .
Figure 2. (a) The proposed electrochemical CO oxidation mechanism on Pt-based alloys drawn based on the data on Ref. [6] (b) Cyclic voltammetry of Pt with CO ads .

Figure 3 .
Figure 3. (a) Comparison of CVs obtained with different films uniformity at 20 mV s À 1 in N 2 -saturated 0.1 M HClO 4 and (b) Optical images of electrodes with different films uniformity, (Reproduced from ref. [27] Copyright (2010), with permission from American Chemical Society).

Figure 5 .
Figure 5. CVs at 0.2 V s À 1 after the CO ads at 0.1 V: (a) cubic and (b) octahedral Pt NPs.CA curves on (c-e) Pt cubic, and (f-h) Pt octahedral NPs of CO oxid at 0.1 V. (c,f) single E step (0.1-0.65 V).(d,g) successive E steps for a partial Co oxid (sequence 0.10 V!0.65 V!0.10 V!0.65 V). (e,h) multiple steps of CA after j subtraction at 0.1 V from (d,g).(Reproduced from ref.[65]Copyright (2017), with permission from American Chemical Society).

Figure 7 .
Figure 7. CVs of various shapes for Pt NPs (a) spherical, (b) cubic, (c) octahedral, and (d) cuboctahedral.(Top) CV for CO-stripping (red lines) and subsequent CVs (black lines) with scan rate s = 20 mV s À 1 .(Bottom) First derivative, @I/@E, vs. E curve for the CO-stripping transients.(e) Schematic represents CV features (CO-stripping) and its assigns to surface domains and sites (Reproduced from ref. [57] Copyright (2012), with permission from American Chemical Society).CVs of CO-stripping and the subsequent CVs at 20mVs À 1 obtained for a commercial unsupported Pt black recorded in (f) 0.5M H 2 SO 4 , and (g) 0.1M HClO 4 .(Reproduced from ref. [16c] Copyright (2018), with permission from Elsevier).