Electrochemistry Communications 60 (2015) 30–33 Contents lists available at ScienceDirect Electrochemistry Communications j ourna l homepage: www.e lsev ie r .com/ locate /e lecom Surface spectators and their role in relationships between activity and selectivity of the oxygen reduction reaction in acid environments Eduardo G. Ciapina a,b,c, Pietro P. Lopes a, Ram Subbaraman a, Edson A. Ticianelli c, Vojislav Stamenkovic a, Dusan Strmcnik a,⁎, Nenad M. Markovic a a Materials Science Division, Argonne National Laboratory, Argonne, IL, USA b Faculdade de Engenharia, UNESP—Univ. Estadual Paulista, Guaratinguetá, SP, Brazil c Instituto de Quimica de Sao Carlos, Universidade de Sao Paulo, Sao Carlos, SP, Brazil ⁎ Corresponding author. E-mail address: strmcnik@anl.gov (D. Strmcnik). http://dx.doi.org/10.1016/j.elecom.2015.07.020 1388-2481/Published by Elsevier B.V. a b s t r a c t a r t i c l e i n f o Article history: Received 7 July 2015 Received in revised form 20 July 2015 Accepted 22 July 2015 Available online 29 July 2015 Keywords: Oxygen reduction reaction Catalyst selectivity Electrocatalysis Electrochemical interfaces Electronic effects Ensemble effects Weuse the rotating ring disk (RRDE)method to study activity–selectivity relationships for the oxygen reduction reaction (ORR) on Pt(111) modified by various surface coverages of adsorbed CNad (ΘCNad). The results demon- strate that small variations inΘCNad have dramatic effect on theORR activity and peroxide production, resulting in “volcano-like” dependence with an optimal surface coverage of ΘCNad= 0.3 ML. These relationships can be sim- ply explained by balancing electronic and ensemble effects of co-adsorbed CNad and adsorbed spectator species from the supporting electrolytes, without the need for intermediate adsorption energy arguments. Although this study has focused on the Pt(111)–CNad/H2SO4 interface, the results and insight gained here are invaluable for controlling another dimension in the properties of electrochemical interfaces. Published by Elsevier B.V. 1. Introduction The oxygen reduction reaction (ORR), the cathodic half-cell reaction in fuel cells [1,2], is one class of electrocatalytic reaction exhibiting strong relationships between interfacial properties and reactivity, due its multi-electron reaction nature that includes a number of elementary steps involving different reaction intermediates (e.g., O2*, H2O2*, and OH*) [3–5]. From studying the ORR on well-characterized metal single crystal surfaces it was found that the reaction kinetics varies with the crystal face differently according to the electrolyte used [6], strongly suggesting that structure sensitivity arises mainly from the geometry dependent adsorption of spectator species [7–11]. As a consequence, during the ORR metal surfaces are always covered with different kind of spectators, for which the corresponding surface coverage (Θs) is strongly dependent on a synergy between the substrate–adsorbate en- ergetics, the applied electrode potential and the symmetry match be- tween the geometry of surface atoms and adsorbates. Recently, it has been shown that a Pt(111) electrode modified by an optimal coverage of irreversibly adsorbed CN species (ΘCNad) can effectively suppress ad- sorption of undesired tetrahedral oxyanions, while providing sufficient number of free metal sites to effectively chemisorb O2 molecule and to break the O\\O bond [12]. It was proposed that adsorption of phosphoric and sulfuric oxyanions on Pt(111)–CNad are suppressed largely via a steric, site-blocking (ensemble) mechanism [12]. In con- trast, the effect of CNad on adsorption of Hupd and OHad is markedly dif- ferent, affecting the Pt–Hupd and Pt–OHad energetics via dipole–dipole CNad–Hupd/OHad interactions (electronic effects) [13]. Combination of these two mechanisms may serve as a basis in finding true functional links between the activity and selectivity of Pt(111)–CNad electrochem- ical interface in environments containing undesirable bisulfate/phos- phate anions. Given this opportunity, we decided to extend our recent rotating disk exploration of the ORR activity on the Pt(111)–CNad electrode to encompass utilization of the rotating ring disk method for monitoring CNad surface coverage dependent selective production of peroxide in- termediate. Our findings address some key points in selectivity for the ORR at the same time as discuss the importance of adsorbate lateral interactions. 2. Experimental Crystal preparation procedures and electrochemical measurements have been described in previous publications [12,14,15]. Electrode po- tentials are given versus the reversible hydrogen electrode (RHE), cali- brated using H2 oxidation in a separate experiment. Cyanide-modified Pt(111) electrodes were prepared by immersion of an annealed Pt(111) single crystal in 0.2 mol L−1 KCN solution for 30 min at open http://crossmark.crossref.org/dialog/?doi=10.1016/j.elecom.2015.07.020&domain=pdf http://dx.doi.org/10.1016/j.elecom.2015.07.020 mailto:strmcnik@anl.gov http://dx.doi.org/10.1016/j.elecom.2015.07.020 http://www.sciencedirect.com/science/journal/13882481 www.elsevier.com/locate/elecom 31E.G. Ciapina et al. / Electrochemistry Communications 60 (2015) 30–33 circuit, thus forming an irreversibly adsorbed CN adlayer. After exten- sive rinsing, the electrode was embedded into the rotating ring-disk electrode assembly (RRDE) and transferred into a standard three- compartment electrochemical cell containing 0.05 mol L−1 H2SO4 (EMD) solution saturated with O2, under potential control at 0.27 V. Electrodes with distinct ΘCNad were obtained by cycling the CN- modified Pt(111) electrode between0.07 and 0.9 V in aO2-saturated so- lution at 1600 rpm. Base cyclic voltammograms characteristic to a given ΘCNad was recorded by purging out the dissolved oxygen by argon for at least 30 min under at 0.27 V. The sweep rate for all measurements was 50 mV s−1. Peroxide oxidation signal was measured at the ring held at 1.2 V. Collection efficiency found for the RRDE setupwas 0.24±0.05 de- termined from separate experiments, as described elsewhere [16]. All gases were 5 N5 quality acquired from Airgas. 3. Results and discussion Cyclic voltammetrywas used to show how the surface coverage by a “static” cyanide adlayer on Pt(111) affects the potential-dependent (“dynamic”) adsorption of Hupd, OHad and HSO4ad in sulfuric acid solu- tions (Fig. 1). The voltammetric profile of Pt(111) in 0.05 M H2SO4 (Fig. 1a) is divided in three potential regions: (i) the Hupd potential region between 0.05 V and 0.35 V; (ii) adsorption of (bi)sulfate ions between 0.35 V and 0.6 V, with a distinctive sharp adsorption peak (so-called butterfly peak) related to disorder–order transition of HSO4ad [17,18]; and (iii) 0.6 to 0.9 V where a small hump corresponds to OHad adsorption. For comparison, the cyclic voltammetric (CV) pro- file of Pt(111) in 0.1 M HClO4 is displayed in Fig. 1a. Consistent with the previous literature [12], while there is no difference in the Hupd re- gion between these two electrolytes, in 0.1 M HClO4 a characteristic re- versible peak centered at 0.8 V (also called the butterfly feature) corresponds to adsorption of OH−. Nevertheless, Fig. 1b–f show that in sulfuric acid solution the potential-dependent adsorption of spectator species is altered on the Pt(111)–CNad electrode. Three characteristic voltammetric features are noteworthy. First, theHupd region is extended up to ca. 0.6 V, suggesting the existence of electrostatic interactions (dipole–dipole forces) between partially positively charged Hupd and a partially negatively charged CNad. Notice also that while between 0.05 and 0.4 V the adsorption profile (peak shape) on Pt–CN–Hupd is consis- tent with a Frumkin-like adsorption, at more positive potential the Fig. 1. Cyclic voltammograms of Pt(111)–CNad electrode in several distinct CNad coverages in 0 eredwith θCNad = 0.04ML, (c) 0.16ML, (d) 0.31 ML, (e) 0.36 ML and (f) 0.45 ML. Potential regi and OHad, respectively. All experiments were conducted at 25 °C with 50 mV s−1 sweep rates. adsorption of Hupd is also controlled by some attractive forces between CNad and Hupd. Second, the formation of the butterfly feature in region (ii) is rapidly suppressed even at ΘCNad = 0.04ML, confirming previous observation that an ordered adsorption of bisulfate anions requires large ensemble of Pt surface sites. A close inspection of Fig. 1 reveals that increase by CNad to 0.16 ML leads to a further decrease in ΘHSO4ad so that at ca. ΘCNad = 0.31 ML the adsorption of HSO4 is observed only above 0.6 V (a small increase in the “double layer”pseudocapacitive fea- ture). Complete suppression of bisulfate anions appears to occur on the surface which is covered by ΘCNad = 0.36 ML. Third, the surface cover- age and the potential window of adsorption of OHad are also dependent on ΘCNad. Due to decrease in the surface coverage by HSO4ad an initial enhanced OH adsorption is observed already at ΘCNad = 0.04 ML (a peak at 0.9 V). Further increase in ΘCNad, however, is first mirrored by the concomitant increase in ΘOHad, that later becomes suppressed for ΘCNad N 0.31 ML. It is reasonable to suggest that the observed rela- tionship betweenΘCNad andΘOHad is controlled by a delicate balance be- tween the surface coverage by CNad and bisulfate anions and thus the availability of Pt sites for OHad formation that requires a single Pt site. Unlike Hupd, the peakpotential for OHad is shifted towardsmore positive potentials relative to one observed in perchloric acid solution because of the existence of lateral repulsion between partially negative charges of CNad and OHad. All of these changes observed for the “dynamic” specta- tors can be related to the nature of the Pt–cyanide bonding and its sur- face structure, as they can interact through both electrostatic forces and site ensemble selectivity. Taken together, we conclude that CNad “static” layer can affect adsorption of “dynamic” adsorbates by eithermodifying or eliminating surface site-ensembles and by electrostatic dipole–dipole forces. In our previous work we used rotating disk electrode (RDE) method to study the ORR on Pt(111) modified by an optimal coverage of CNad [12]. Of concern here is tomonitor how the peroxide formation depends on various surface coverages of both the “static” and “dynamic” specta- tors. For monitoring simultaneously the ORR and concomitant peroxide formation we employed the RRDEmethod. Selected polarization curves for theORRon Pt(111)–CNad disk electrode and correspondingperoxide oxidation currents on the ring electrode are summarized in Fig. 2a and b, respectively, also including the results on Pt(111) in 0.1 M HClO4, for comparison. Qualitatively, peroxide formation occurs below 0.5 V, e.g., in the “double layer” and Hupd potential regions. For our purposes .05 mol L−1 H2SO4 electrolytes. a) bare Pt(111) in both H2SO4 and HClO4. (b) Pt(111) cov- on i(dashed), ii(solid) and iii(dotted) highlights the presence of HUPD, adsorbed (bi)sulfate Fig. 2.RRDE results for ORR in Pt(111)–CNad electrode in several distinct CNad coverages in 0.05mol L−1 H2SO4. The disk and ring currents are shown in a) and b), respectively. Polarization curves for Pt(111) in pure HClO4 are included for comparison. “Volcano”-like trendwith varying ΘCNad wasmeasured at 0.85 V for activity (c) and peroxide production at 0.60 V (d). ORR selectivity determined at 0.60 V is shown in (e), using Eq. (1). Note that H2O2 is only produced below the reversible potential for O2+2H++2e- –N H2O2. 32 E.G. Ciapina et al. / Electrochemistry Communications 60 (2015) 30–33 here we will first focus on the role of CNad on the reaction pathway and then we use the RRDE polarization curves recorded in Fig. 2a and b to explore the manner and extent of Pt–OHad and Pt–Hupd interactions may affect the ORR on Pt(111)–CNad. As expected, activity–selectivity relationships are strongly depen- dent on the ΘCNad. A quantitative analysis of the results depicted in Figs. 1a–f, 2a and b, summarized in the form of ΘCNad vs. ORR activity at 0.85 V (Fig. 2c) and ΘCNad vs. ring currents at 0.6 V (Fig. 2d) provides insight into the role of the surface coverage of adsorbed species (Θs) on the rate of the ORR and peroxide production. While 0.85 V was chosen because in this region the ORR is under kinetic control, 0.6 V is an opti- mal potential where the role of CNad can be studiedwithout an interfer- ence with the Hupd potential region. Notice that by changing ΘCNad the surface coverage by bisulfate will affect the activity of ORR (blocking O2 adsorption) but not the peroxide production after the ORR starts. This is because desorption of large bi-sulfate anions leave behind an en- semble of at least four Pt sites that are required for further reduction of H2O2* to water (see the result for the ORR on Pt(111) in 0.05 MH2SO4). Thus, one can anticipate that at 0.6 V it is indeed possible to explore the role of CNad on the functional links between activity and selectivity. Fig. 2e shows the fraction of H2O2* (xH2O2) produced at 0.6 V as well as the production of H2O (a 4e− process). The fraction of peroxide is cal- culated using the equation: [16,19]: H2O2 %ð Þ ¼ 2Iring=N Idisk þ Iring=N � � ð1Þ where Iring and Idisk are ring and disc currents at 0.6 V andN is the collec- tion efficiency (determined as 0.24). As expected, for high CNad cover- ages (ΘCNad N 0.36 ML) very small currents for ORR are observed, signaling that the Pt sites for adsorption of O2 are highly blocked by CNad. This observation is in harmony with CVs depicted in Fig. 1e to f, and with the fact that at high ΘCNad on Pt(111) adopts an ordered close packed structure [20–22]. Even more striking is the effect of CNad on the selectivity, as for ΘCNad N 0.31 ML the ORR almost entirely goes through the 2e− pathway generating H2O2 (see Fig. 2e). As pro- posed above, under this experimental conditions the selectivity is pre- dominantly governed by the lack of large Pt ensembles (at least 4 Pt atoms) needed to further reduce H2O2* to water. As expected, Fig. 2c and d show a ΘCNad-controlled “volcano” like behavior; e.g., by decreas- ing theΘCNad the ORR first increases, reaching its maximum forΘCNad= 0.29 ML, and then decreases so that for ΘCNad b 0.1 ML the activity ap- proaches same as on bare Pt(111) in sulfuric acid solution (Fig. 2a). Also worth noting briefly is that in certain potential range the activity–selectivity relationships are controlled by the combined effects of ΘCNad and Θs of dynamic spectators. Two potential regions are of par- ticular interest; namely, the potential window where on Pt(111)–CNad surface either OHad or Hupd are adsorbed on CNad-free Pt sites (see Fig. 1, regions i and iii). In contrast to adsorption of bisulfate ions, these two dynamic spectators need only one Pt site for adsorption. It is reasonable to anticipate that reactivity and selectivity will depend strongly on the intermolecular interactions between adsorbed species. For example, the onset potential for the ORR is strongly dependent on ΘOHad, the effect arising from the need of removal (reduction) of OHad that is controlling adsorption of O2. Interestingly, irrespectively of ΘCNad the ORR below 0.8 V (region iii, Fig. 1) proceedswithout peroxide formation. In contrast, for the same CNad coverages in theHupd potential region the ORR is always accompanied by peroxide production. A possible origin for this difference could be due both distinct nature of Hupd–Hupd vs. OHad–OHad interactions and/or the role Hupd and OHad may have in the overall reaction pathway. For instance, while Hupd tends to form a randomly distributed disordered structure (the Frumkin-type isotherm with repulsive lateral interactions), it appears that OHad adlayer has a tendency to cluster on the Pt(111) surface, at least at lower OHad coverages. As a consequence, a small ΘHupd can alter significantly the availability of Pt ensembles required for breaking the O\\O bond in H2O2; thus acting as a third-body effect. On the other hand, if OHad has a tendency to cluster on the Pt(111) surface then there are always free sites (ensemble of four atoms) for further reduction of the peroxide intermediates. Clearly, the interactions observed between 33E.G. Ciapina et al. / Electrochemistry Communications 60 (2015) 30–33 static and dynamic spectators can help us to infer not only that certain ensemble of Pt sites is required for the 4e− reduction but, in addition, provide more substantial evidence on how the switching between the “peroxide pathway” and the “water pathway” might take place. 4. Conclusion By combining the use of a “static” spectator, (CNad, with a known surface coverage) with the common “dynamic” spectators observed in sulfuric acid solutions (Hupd, HSO4ad and OHad) wewere able to propose the functional links between activity and selectivity for the ORR. These relationships can be simply explained by balancing electronic and en- semble effects of adsorbed spectators without invoking intermediate adsorption energy arguments. Although this study has focused on the Pt(111)–CNad/H2SO4 interface, the results and insight gained here are valuable for controlling another dimension in the properties of electro- chemical interfaces. Acknowledgments This work was supported by the Office of Science, Office of Basic Energy Sciences, Division of Materials Sciences, U.S. Department of Energy, under contract DE-AC02-06CH11357 (BES-DMSE). E. G. Ciapina acknowledges support from FAPESP (grant numbers 2010/02905-2 and 2013/16930-7). References [1] H.A. Gasteiger, N.M. Markovic, Chemistry. Just a dream—or future reality? Science 324 (2009) 48–49. [2] M.K. Debe, Electrocatalyst approaches and challenges for automotive fuel cells, Nature 486 (2012) 43–51. [3] M.H. Shao, R.R. Adzic, Spectroscopic identification of the reaction intermediates in oxygen reduction on gold in alkaline solutions, J. Phys. Chem. B 109 (2005) 16563–16566. [4] N. Ohta, K. Nomura, I. Yagi, Adsorption and electroreduction of oxygen on gold in acidic media: in situ spectroscopic identification of adsorbed molecular oxygen and hydrogen superoxide, J. Phys. Chem. C 116 (2012) 14390–14400. [5] J.K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J.R. Kitchin, T. Bligaard, et al., Origin of the overpotential for oxygen reduction at a fuel-cell cathode, J. Phys. Chem. B 108 (2004) 17886–17892. [6] N.M. Markovic, P.R. Jr, Surface science studies of model fuel cell electrocatalysts, Surf. Sci. Rep. 45 (2002) 117–229. [7] F.C. Nart, T. Iwasita, M. Weber, Vibrational spectroscopy of adsorbed sulfate on Pt(111), Electrochim. Acta 39 (1994) 961–968. [8] F.C. Nart, T. Iwasita, An FTIR study of adsorbed sulfate species on polycrystalline platinum with application of group theory to the problem of band assignment for adsorbed species, J. Electroanal. Chem. 322 (1992) 289–300. [9] V.R. Stamenkovic, N.M. Markovic, P.N. Ross, Structure-relationships in electrocatalysis: oxygen reduction and hydrogen oxidation reactions on Pt (111) and Pt (100) in solutions containing chloride ions, J. Electroanal. Chem. 500 (2001) 44–51. [10] N.M. Markovic, H.A. Gasteiger, B.N. Grgur, P.N. Ross, Oxygen reduction reaction on Pt(111): effects of bromide, J. Electroanal. Chem. 467 (1999) 157–163. [11] H.A. Gasteiger, N.M. Markovic, P.N. Ross, Bromide adsorption on Pt (111): adsorp- tion isotherm and electrosorption valency deduced from RRD Pt (111) E measure- ments, Langmuir 12 (1996) 1414–1418. [12] D. Strmcnik, M. Escudero-Escribano, K. Kodama, V.R. Stamenkovic, A. Cuesta, N.M. Marković, Enhanced electrocatalysis of the oxygen reduction reaction based on pat- terning of platinum surfaces with cyanide, Nat. Chem. 2 (2010) 880–885. [13] M. Escudero-Escribano, G.J. Soldano, P. Quaino, M.E. Zoloff Michoff, E.P.M.M. Leiva, W. Schmickler, et al., Cyanide-modified Pt(111): structure, stability and hydrogen adsorption, Electrochim. Acta 82 (2012) 524–533. [14] D.S. Strmcnik, P. Rebec, M. Gaberscek, D. Tripkovic, V. Stamenkovic, C. Lucas, et al., Relationship between the surface coverage of spectator species and the rate of elec- trocatalytic reactions, J. Phys. Chem. C 111 (2007) 18672–18678. [15] B. Genorio, D. Strmcnik, R. Subbaraman, D. Tripkovic, G. Karapetrov, V.R. Stamenkovic, et al., Selective catalysts for the hydrogen oxidation and oxygen re- duction reactions by patterning of platinum with calix[4]arene molecules, Nat. Mater. 9 (2010) 998–1003. [16] U.A. Paulus, T.J. Schmidt, Oxygen reduction on a high-surface area Pt/Vulcan carbon catalyst: a thin-film rotating ring-disk electrode study, J. Electroanal. Chem. 495 (2001) 134–145. [17] A.M. Funtikov, U. Stimming, R. Vogel, Anion adsorption from sulfuric acid solutions on Pt(111) single crystal electrodes, J. Electroanal. Chem. 428 (1997) 147–153. [18] N. García, V. Climent, J.M. Orts, J.M. Feliu, A. Aldaz, Effect of pH and alkaline metal cations on the voltammetry of Pt(111) single crystal electrodes in sulfuric acid solu- tion, ChemPhysChem 5 (2004) 1221–1227. [19] J.S. Jirkovský, M. Halasa, D.J. Schiffrin, Kinetics of electrocatalytic reduction of oxygen and hydrogen peroxide on dispersed gold nanoparticles, Phys. Chem. Chem. Phys. 12 (2010) 8042–8052. [20] A. Cuesta, At least three contiguous atoms are necessary for CO formation during methanol electrooxidation on platinum, J. Am. Chem. Soc. 128 (2006) 13332–13333. [21] C. Stuhlmann, I. Villegas, M.J. Weaver, Scanning tunneling microscopy and infrared spectroscopy as combined in situ probes of electrochemical adlayer structure. Cyanide on Pt(111), Chem. Phys. Lett. 219 (1994) 319–324. [22] C. Stuhlmann, Characterization of an electrode adlayer by in-situ infrared spectroscopy: cyanide on Pt(111), Surf. Sci. 335 (1995) 221–226. http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0005 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0005 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0010 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0010 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0015 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0015 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0015 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0020 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0020 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0020 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0025 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0025 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0025 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0030 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0030 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0035 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0035 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0040 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0040 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0040 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0045 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0045 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0045 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0045 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0050 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0050 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0055 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0055 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0055 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0060 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0060 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0060 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0065 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0065 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0065 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0070 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0070 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0070 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0075 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0075 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0075 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0075 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0080 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0080 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0080 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0085 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0085 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0090 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0090 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0090 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0095 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0095 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0095 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0100 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0100 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0100 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0105 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0105 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0105 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0110 http://refhub.elsevier.com/S1388-2481(15)00200-3/rf0110 Surface spectators and their role in relationships between activity and selectivity of the oxygen reduction reaction in aci... 1. Introduction 2. Experimental 3. Results and discussion 4. Conclusion Acknowledgments References