Revealing the Dominant Chemistry for Oxygen Reduction Reaction on

Dec 11, 2017 - The bulk chemistry has been successfully used as a descriptor for oxygen reduction reaction (ORR) activities of various metal oxides. H...
0 downloads 17 Views 6MB Size
Subscriber access provided by RMIT University Library

Article

Revealing the Dominant Chemistry for Oxygen Reduction Reaction on Small Oxide Nanoparticles Ye Zhou, Shibo Xi, Jingxian Wang, Shengnan Sun, Chao Wei, Zhenxing Feng, Yonghua Du, and Zhichuan J. Xu ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03864 • Publication Date (Web): 11 Dec 2017 Downloaded from http://pubs.acs.org on December 11, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Revealing the Dominant Chemistry for Oxygen Reduction Reaction on Small Oxide Nanoparticles Ye Zhou†,‡, Shibo Xi§, Jingxian Wang†, Shengnan Sun†,‡, Chao Wei†,‡, Zhenxing Feng#, and Yonghua Du§,*, Zhichuan J. Xu†,‡,ǁ,* †

School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang

Avenue, 639798, Singapore §

Institute of Chemical and Engineering Sciences A*STAR, 1 Pesek Road, 627833, Singapore

#

School of Chemical, Biological, and Environmental Engineering, Oregon State University, Cor-

vallis, OR, 97331, United States ‡

Solar Fuels Laboratory and Energy Research Institute, Nanyang Technological University, 50

Nanyang Avenue, 639798, Singapore ǁ

Energy Research Institute @ Nanyang Technological University, 50 Nanyang Avenue, 639798,

Singapore Corresponding Author * ([email protected]). * ([email protected]).

1 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: The bulk chemistry has been successfully used as descriptor for oxygen reduction reaction (ORR) activities of various metal oxides. However, as the size of oxides become small, the bulk chemistry may not be sufficient for describing the activities. Here, we report a systematic study on Mn substituted ferrite MnxFe3-xO4 (x=0.5~2.5) nanoparticles and the roles of surface Mn in determining their ORR activities. Gradual Mn substitution induced changes in Mn valence and crystal structure. However, there is no remarkable correlation that can be found between their bulk chemistry and ORR activities. Instead, the surface Mn density and valency were found to play dominant roles in determining the ORR. This work shows that at a small particle size, the bulk chemistry of oxides may not be the descriptor for their electrochemical properties. Due to the significantly high surface/bulk ratio, the surface chemistry has to be carefully characterized to interpret the activities of oxide nanoparticles.

KEYWORDS: Surface density; Mn valence state; Manganese ferrite; Oxygen reduction reaction; Nanoparticles

2 ACS Paragon Plus Environment

Page 2 of 17

Page 3 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

INTRODUCTION The growing level of aspiration for higher standards of living is continuously calling for clean renewable energy and high energy efficiency. Technologies like fuel cell and metal-air batteries have attracted great attention for their important roles in renewable energy infrastructures. For large-scale application, seeking cheap and highly efficient catalysts, for example, earth-abundant transition metal oxides, becomes extremely important. To develop highly active oxide catalyst, efforts to reveal the underlying mechanisms and descriptors for their catalytic behaviors are essential. Earlier attempts have been successful in revealing the relation of oxygen electrocatalytic activity with the eg electron of transition cation1, and O p-band center2. Most of these correlations and theories are developed for perovskites which are of large particle size. More recently, growing attention has been moved onto spinel-type materials because spinels have octahedral sites (which were widely accepted as the catalytic active sites3) as that of perovskite oxides. Additionally, nanoscale spinel oxides can be readily developed to promote the mass efficiency4. For spinel oxides catalysts, growing efforts have been given to identifying the active site5, understanding the composition-dependent activities4, 6, deliberately substituting cations or dopants to promote the activity of host materials7-8. In particular, the spinel oxides containing Mn cations have been found highly active for oxygen reduction reaction (ORR)4, 9. For example, Mn was found to be the most effective cation in promoting ORR activity of ferrite nanoparticles over other substitution cations like Co and Ni9. It has been disclosed that the ORR activity through Mn substitution are dependent on the nominal Mn content. It is because the enhancement were found with increasing Mn substitution in MnxFe3-xO4 and MnxCo3-xO4 (0 ≤ x ≤ 1).

9-10

All indicate

that Mn species are the active sites. However, it was also reported that the ORR activities of MnxCu3-xO4 ሺ1.6 ≤ x ≤ 2) were adversely affected by a high nominal Mn content, suggesting 3 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

that the ORR activity does not solely depend on the Mn content11. Typically, progressively substituting Mn in spinel structure results in a transition from cubic to tetragonal spinels due to the Jahn-Teller distortion of Mn3+. Findings in MxMn3-xO4 (M = divalent metals) has shown that Mnrich tetragonal spinel is less active than that of cubic counterparts owing to its weaker oxygen binding energy4. The optimal Mn-substitution level in promoting oxygen reduction of MnxCo3xO4

(0 ≤ x ≤ 3) was determined experimentally by Lee6, while the dominating factor determin-

ing the ORR activity remains elusive. In fact, the attention has been given more on the bulk chemistry of these Mn-based spinel oxides. The effort to directly correlate the ORR activity with the oxide surface property remains limited. However, it has to be noticed that the surface is the place where electrochemistry occurs. In this work, we substituted Mn into spinel ferrite and investigated MnxFe3-xO4 (x=0.5~2.5) nanoparticles (NPs) for ORR in KOH. The bulk chemistry of the MnxFe3-xO4 NPs was carefully examined by bulk-sensitive tools including XRD, XANES, and EXAFS. And classical surfacesensitive tool, XPS, was employed to probe the surface chemistry of the MnxFe3-xO4 NPs. Interestingly, XPS gives similar information as those characterized by the bulk tools because the oxide particle sizes (< 10 nm) are within the average depth (5~10 nm) of XPS signal collection. Correlating the ORR activities of MnxFe3-xO4 (x=0.5~2.5) to NPs’ bulk properties produces contradictory observations as compared to previous reports. In contrast, information obtained from an electrochemical approach reveals the dominating role of NPs’ surface chemistry in ORR. For the first time, the interplay between surface Mn density and Mn valency in governing the electrocatalysis of MnxFe3-xO4 was established by this study. Through taking the various manganese ferrite NPs as model materials, our work helps solve the puzzle of oxide catalysts and will encourage ongoing efforts in exploring the surface of oxide nanoparticles. 4 ACS Paragon Plus Environment

Page 4 of 17

Page 5 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

RESULTS AND DISCUSSION Manganese ferrite nanoparticles (MnxFe3-xO4) were prepared by a wet-chemical method,12 and their crystal structures ware examined by XRD technique. As displayed in Figure 1a, Fe-rich manganese ferrites (x=0.5, 1.0, 1.5) possess a single cubic spinel phase as all the diffraction peaks match well with that of the standard Fe3O4. In contrast, their Mn-rich counterparts (x=2.0, 2.5) are subjected to certain distortion and possess a tetragonal spinel phase. To be specific, the extra peak appeared at ~33° can be assigned to the standard tetragonal spinel phase (Mn3O4). It is typical for high Mn-concentration spinel oxides to experience a cubic-to-tetragonal phase transition due to the Jahn-Teller distortion of MnO6 octahedrons13. The cubic-tetragonal transition in MnxFe3-xO4 was reported to occur when Mn fraction (x) is beyond 1.74~1.84,13 and this is consistent with the observation in the as-prepared Mn ferrites. Figure 1b shows a representative TEM image of the as-synthesized nanoparticles loaded on Vulcan carbon. High-resolution transmission electron microscopy (HRTEM) image (Figure 1c) show the crystal (400) planes of the ferrite particles (x=1.0) which belong to the cubic phase. The phase purity of the as-prepared ferrite (x=1.0, 2.5 as representative, Figure 1d, 1e) was further confirmed by the selected area electron diffraction (SAED) where all the crystal planes identified coincide well with the XRD results. The average particle sizes of these MnxFe3-xO4 nanoparticles are within 6~13 nm by counting at least 200 nanoparticles for each (Figure S1). The specific surface areas of the ferrite particles given in Figure 1f were calculated according to the particle sizes and the densities displayed in Table S1. The cation valence states were inspected by XPS. In Figure 1g, the Mn 2p3/2 peak is split into two peaks subordinating to Mn3+ (641.89 eV) and Mn2+ (640.78 eV), respectively.14 The ratio between the two integrated peak areas (which represent the Mn3+/Mn2+ ratio) were used to quantify the Mn valence states.14 The results displayed in Table S2 show that the 5 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

average Mn oxidation state in MnFe2O4 (x=1.0) is of the lowest (~+2.5) and that in other ferrites varies between +2.8 to +2.9. In contrast, Fe cations remains stable at +3 in all of the compositions as indicated by the consistent 2p3/2 peak positions at ~711.3 eV15. Thus, the chemical formula can be approximated as MnxFe3-xO4+δ where δ signifies oxygen non-stoichiometry and ranges from 0.27 to 0.47.16-17

Figure 1. Crystal structure, particle size and metal valance state characterization. a) XRD patterns of MnxFe3-xO4; b) A typical TEM image of as-synthesized ferrite nanoparticles (MnFe2O4, x=1) loaded on Vulcan carbon; c) The STEM dark-field image of a ferrite nanoparticle (x=1.0); The SAED patterns of the d) cubic spinel ferrite (x=1.0) particles and e) tetragonal spinel ferrite (x=2.5); f) The average particle sizes and specific surface areas, and g) Mn 2p spectra (Fitted) and Fe 2p spectra of these ferrite MnxFe3-xO4 nanoparticles.

6 ACS Paragon Plus Environment

Page 6 of 17

Page 7 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

The ORR performance of spinel MnxFe3-xO4 catalysts in O2-saturated KOH were examined using a rotating ring-disk electrode (RRDE) in order to determine kinetic parameters for O2 reduction and clarify the peroxide intermediate formation (Figure S2). The representative ORR curves comprising ring and disk currents were shown in Figure 2a. To make a fair comparison, the specific disk current densities at 0.75 V vs. RHE were mapped out (Figure 2b). The specific current densities were obtained by normalizing the ORR currents to the oxide surface area after background, mass transport, and iR corrections18. As shown, the ferrite with a Mn fraction of x=1.0 gives the highest current density, which is twice of the density value as given by x=0.5. Being aware that Fe species are inactive comparing with Mn species in catalyzing oxygen reduction in oxide forms from earlier studies9. We noted, increasing Mn concentration beyond x=1.0 failed to boost the activity further. Instead, the ORR specific activity displays “N”-shape trend with respect to the Mn nominal concentration.

7 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. ORR electrocatalytic properties and EXAFS analysis. a) Representative ORR polarization curves of MnxFe3-xO4 carried out with RRDE at a rotating speed of 1600 rpm; b) Specific ORR activity based on surface area of oxides (current density at 0.75V vs. RHE); c) Potentialdependent electron transfer numbers calculated within at low overpotential region (0.6-0.82 V vs RHE); d) Fourier transform of Mn K-edge EXAFS spectra (colored curves) and corresponding fitting curves (grey curves) of MnxFe3-xO4. To understand the composition-dependent ORR trend, we first pay attention to the ring currents as shown in Figure 2a. A notable exception in terms of ring currents is catalyst x=1.0, where its ring currents started rising at a relatively more positive potential and attained a much higher current density than the other ferrites. Since the ring currents indicating the formation of peroxide directly reflect the reduction pathway in ORR, the potential-dependent electron transfer numbers (n) were calculated accordingly (Figure 2c). A combined 2e- and 4e- oxygen reduction path in MnxFe3-xO4 was indicated from the calculated n (ranges between 3~4). Since a high electron transfer signifies high efficiency in catalyzing ORR, we may able to estimate the intrinsic activity of the active species from the n obtained. As shown in Figure 2c, the n in the cubic spinels increases in a sequence: x=1.0, x=1.5, and x=0.5, demonstrating a high catalyzing efficiency at x=0.5 and a lower efficiency especially at x=1.0. For the tetragonal spinels (x=2.0, 2.5), their n are quite similar and both are higher than cubic spinel x=1.0. These observations indicate that there exist substantial differences in the intrinsic activity of active species in MnxFe3-xO4. If such a difference can entirely account for their distinctive ORR performance, we would have expected a linear correlation between n and ORR activity. However, intriguing discrepancies can be seen, especially for cubic spinel ferrites with substitution numbers x=0.5 and x=1.0 (Figure S3). It should be noticed that with attaining the lowest n, ferrite with x=1.0 inconsistently yields a high8 ACS Paragon Plus Environment

Page 8 of 17

Page 9 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

est specific current and, on the contrary, its counterpart x=0.5 gives a much lower current with proceeding through a higher electron transfer pathway. Since n reflects the intrinsic activity of the catalytic-active cations that are independent of their concentration, changes in Mn content of MnxFe3-xO4 appears to be one potential factor that was neglected. In contrast, Fe cations are not considered as catalytically active because it is stable at +3 and has an eg occupancy of 2.1 Previous studies have also claimed that the catalytic performance of oxides is site dependent, where the coordination of transition metal plays important role16, 19. As spinel oxide consists of two geometric sites, thus, extended X-ray absorption fine structure (EXAFS) was conducted to inspect the site occupancy of transition metals in spinel MnxFe3-xO4. Figure 2d shows the Fourier transforms (FT) of Mn K-edge EXAFS spectra of MnxFe3-xO4. As revealed from EXAFS analysis20, the amount of Mn in octahedral sites increase with increasing x in MnxFe3-xO4 (Figure S4, Table S3). In an effort to eliminate the effects from Mn concentration, we corrected the ORR current density (µA cm-2ox) to bulk Mn composition (x) in MnxFe3-xO4 or alternatively to Mn occupancy in tetrahedral/octahedral site (Figure S5a) in consideration of the site-dependence16, 19. The activity displays a clear difference before and after those corrections (Figure 2b vs. Figure S5a). By observing how the corrected activity changes as a function of x in MnxFe3-xO4, it is convinced that in addition to the Mn content, there exists another activity-determining factor. Enlightened by previous findings that Mn valence state could be an influencing factor associated with the substantial difference in ORR activity or reduction path3, attempts were then made to investigate the relationship between Mn valence and Mn composition (x)-corrected or Mn site occupancycorrected ORR activity (Figure S5b-d). Unfortunately, none of these correlations seems to give a clear clue on assisting the interpretation of the ORR performance by MnxFe3-xO4.

9 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Note that the well-defined redox peaks associated with transition metal species intrigued by electrochemical methods can afford abundant information to uncover the dynamics of active species at the oxide surface21-22. We scanned the CV of spinel MnxFe3-xO4 in KOH within a potential window of 0.2 to 1.4 V (vs. RHE). As shown in Figure 3a-e, capacitive characteristics featured by predominant redox peaks present in all of the ferrites. Specifically, the second reduction wave is fading with increasing x in MnxFe3-xO4 which is accompanied by the broadening and negative shift of the first reduction wave. The reduction waves can be described as the gradual evolution of Mn species to its reduced valence state, prior to the conversion into Mn2+ 22-23. Interestingly, the redox peak current fluctuates with the composition (x) which signifies different availability of surface redox active sites. Considering that the redox reaction of Fe species is more negative than the investigated potential window, the contribution of Fe cations to the redox electron density is negligible24. Thus, we adopted a integration method to calculate the surface Mn density that is actively engaged in the electrochemical process22. To be specific, the cathodic response reflecting the Mn species from its original valence state (determined by XPS) to Mn2+ was integrated to extrapolate the amount of active Mn density. A linear background subtraction was first employed in order to integrate the cathodic redox faradaic charge based on the principles adopted in Celorrio’s work22. As shown in Figure 3a-e, the shaded region was integrated as the cathodic charge (Q) for spinel MnxFe3-xO4. More details regarding the calculations were covered in Supporting information (Experimental and Table S4).

10 ACS Paragon Plus Environment

Page 10 of 17

Page 11 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 3. Determination of surface Mn density. Cyclic voltammetry of MnxFe3-xO4 (a. x=0.5, b. x=1.0, c. x=1.5, d. x=2.0 and e. x=2.5) at a scan rate of 10 mV s-1 in 0.1 M KOH and f) surface Mn density. The surface Mn density provided in Figure 3f strongly suggests that the amount of active Mn significantly differs from the nominal Mn composition. Noting that nanoparticle surface can be another version of its bulk regarding atom arrangements, coordination environment, the mismatch between bulk and surface composition is not unusual25. It was proposed that segregation in the near-surface region is one of the attributes of metal oxide nanoparticles.26-27 Elastic/electrostatic interaction of the substitutes with their surrounding lattice, surface defects interaction and cation size mismatch were proposed to be responsible in affecting the cation distribution on the oxide surface27-28. The atomic ratio of Mn/Fe probed by XPS (Figure S6) suggests an 11 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

underestimated and conversely, an overestimated Mn content in composition (x=0.5, 1.0) and composition (x=2.0, 2.5), respectively. This verifies the difference in bulk and surface composition regarding MnxFe3-xO4 nanoparticles. However, the significant differences in surface composition characterized by the CVs does not well coincides with the XPS results. We believe this is mainly due to the limitations of XPS technique in probing the outmost surface information of ultra-small nanoparticles. Here, the nanoparticle sizes are within the average depth (5~10 nm) of XPS signal collection. In addition, in view of the consistency in cation valency characterized by XANES (Figure S7) and XPS, it is convinced that the Mn touched by XPS are nearly from the whole oxide particles. Intriguingly, the fluctuations of surface Mn density as a function of the nominal Mn composition (x) resembles the activity trend in Figure 2b. This is a strong indication of the tight connections between surface Mn density and ORR activity of spinel MnxFe3-xO4 oxides. To see if the surface Mn density can totally account for the activity difference, we correlated the catalytic activity (Figure 2b) to the surface Mn density. Figure 4a shows that the surface Mn density-rationalized ORR activity decreases in the sequence: x=0.5, x=1.5, x=2.0, x=2.5 and x=1.0, signifying the existence of other activity-determining factor. Of notable interest is that the rationalized ORR activity are in good agreement with the n characterized by RRDE, where a higher rationalized activity matches with a higher n, and vice versa. This consistency, to a large extent, has verified the reliability of surface Mn density obtained using electrochemical CVs scans. To further inspect the role of Mn valency, we plotted the rationalized ORR activity as a function of the Mn oxidation state characterized by XPS (Figure 4b). It is clear that increasing Mn oxidation state promotes their intrinsic activity (as indicated by the dash line), which is in agreement with those reported earlier in perovskites, further confirming that Mn3+ is more active than Mn2+ because 12 ACS Paragon Plus Environment

Page 12 of 17

Page 13 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

the high spin Mn3+ fulfills a moderate eg filling1. Note for the tetragonal spinels (x=2.0, 2.5), a slight downshift of ORR activity deviated from the correlated dash line can be observed. This can be ascribed to its weaker oxygen adsorption ability compared with cubic counterparts4 (Figure S8). Thus, our interpretation of the changes in MnxFe3-xO4’s ORR activity suggested in Figure 2b can be explained from the interplay between surface Mn density and its oxidation state.

Figure 4. Interplay between surface Mn density and Mn valence state. a) Rationalized ORR specific activity of MnxFe3-xO4 (Current was normalized to surface Mn density); b) Correlation between Mn valence state and rationalized specific activity.

CONCLUSION In summary, MnxFe3-xO4 (x=0.5~2.5) nanoparticles were investigated as ORR catalysts and thorough understanding on their distinctive specific activity were unfolded from the perspectives of both surface Mn density and Mn valency. Due to the mismatch between surface and bulk Mn concentration in MnxFe3-xO4 nanoparticles, we integrated the electrochemical redox responses of Mn cation triggered in CVs and revealed the near-surface Mn density. Particularly, the surface 13 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Mn density was found to be one influential factor in governing the ORR activity brought by gradual Mn substituting in the investigated ferrites. In addition, the integration method adopted could afford as an alternative to inspect the surface information for ultra-small oxide nanoparticles, for which XPS might not be that surface-sensitive. This work brings the surface density of active cations to attention in understanding the electro-catalytic properties of transition metal oxides. ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publication website at http://pubs.acs.org and includes detailed experimental procedures and characterization data. AUTHOR INFORMATION Corresponding Author * Email: [email protected], [email protected]. Notes The authors declare no competing financial interests. ACKNOWLEDGEMENT This work was supported by the Singapore Ministry of Education Tier 1 Grant (RG131/14), Tier 2 Grant (MOE2015-T2-1-020) and the Singapore National Research Foundation under its Campus for Research Excellence And Technological Enterprise (CREATE) programme. Authors thank the Facility for Analysis, Characterization, Testing, and Simulation (FACTS) in Nanyang

14 ACS Paragon Plus Environment

Page 14 of 17

Page 15 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Technological University and appreciate the XAFCA beamline29 of the Singapore Synchrotron Light Source for XAFS characterization. REFERENCES

(1) Suntivich, J.; Gasteiger, H. A.; Yabuuchi, N.; Nakanishi, H.; Goodenough, J. B.; Shao-Horn, Y., Nat. Chem. 2011, 3, 546-550. (2) Lee, Y.-L.; Kleis, J.; Rossmeisl, J.; Shao-Horn, Y.; Morgan, D., Energy Environ. Sci. 2011, 4, 3966-3970. (3) Stoerzinger, K. A.; Risch, M.; Han, B.; Shao-Horn, Y., ACS Catal. 2015, 5, 6021-6031. (4) Cheng, F.; Shen, J.; Peng, B.; Pan, Y.; Tao, Z.; Chen, J., Nat. Chem. 2011, 3, 79-84. (5) Friebel, D.; Louie, M. W.; Bajdich, M.; Sanwald, K. E.; Cai, Y.; Wise, A. M.; Cheng, M.-J.; Sokaras, D.; Nilsson, A.; Bell, A. T., J. Am. Chem. Soc. 2015, 137, 1305-1313. (6) Lee, E.; Jang, J.-H.; Kwon, Y.-U., J. Power Sources 2015, 273, 735-741. (7) Guo, Y.; Tong, Y.; Chen, P.; Xu, K.; Zhao, J.; Lin, Y.; Chu, W.; Peng, Z.; Wu, C.; Xie, Y., Adv. Mater. 2015, 27, 5989-5994. (8) Xu, X.; Su, C.; Zhou, W.; Zhu, Y.; Chen, Y.; Shao, Z., Advanced Science 2016, 3, 1500187. (9) Zhu, H.; Zhang, S.; Huang, Y.-X.; Wu, L.; Sun, S., Nano Lett. 2013, 13, 2947-2951. (10) Restovic, A.; Rıos, E.; Barbato, S.; Ortiz, J.; Gautier, J., J. Electroanal. Chem. 2002, 522, 141-151. (11) Ríos, E.; Abarca, S.; Daccarett, P.; Cong, H. N.; Martel, D.; Marco, J.; Gancedo, J.; Gautier, J., Int. J. Hydrogen Energy 2008, 33, 4945-4954. (12) Xu, Z.; Shen, C.; Hou, Y.; Gao, H.; Sun, S., Chem. Mater. 2009, 21, 1778-1780. (13) Brabers, V., J. Phys. Chem. Solids 1971, 32, 2181-2191. (14) Tang, Q.; Jiang, L.; Liu, J.; Wang, S.; Sun, G., ACS Catal. 2014, 4, 457-463. (15) Huang, G.; He, E.; Wang, Z.; Fan, H.; Shangguan, J.; Croiset, E.; Chen, Z., Ind. Eng. Chem. Res. 2015, 54, 8469-8478. (16) Chao Wei; Feng, Z.; Scherer, G. G.; Barber, J.; Shao-Horn, Y.; Xu, Z. J., Adv. Mater. 2017, 29, 1606800. (17) Naghash, A.; Lee, J. Y., J. Power Sources 2001, 102, 68-73. (18) Wei, C.; Yu, L.; Cui, C.; Lin, J.; Wei, C.; Mathews, N.; Huo, F.; Sritharan, T.; Xu, Z., Chem. Commun. 2014, 50, 7885-7888. (19) Grimaud, A.; Carlton, C. E.; Risch, M.; Hong, W. T.; May, K. J.; Shao-Horn, Y., J. Phys. Chem. C 2013, 117, 25926-25932. (20) Du, Y.; Wang, J.-o.; Jiang, L.; Borgna, L. S.; Wang, Y.; Zheng, Y.; Hu, T., J. Synchrotron Radiat. 2014, 21, 756-761. (21) Chen, J. Y. C.; Miller, J. T.; Gerken, J. B.; Stahl, S. S., Energy Environ. Sci. 2014, 7, 1382. (22) Celorrio, V.; Calvillo, L.; Dann, E.; Granozzi, G.; Aguadero, A.; Kramer, D.; Russell, A. E.; Fermín, D. J., Catal. Sci. Technol. 2016, 6, 7231-7238. (23) Celorrio, V.; Dann, E.; Calvillo, L.; Morgan, D. J.; Hall, S. R.; Fermin, D. J., ChemElectroChem 2016, 3, 283291. (24) Castro, P. A.; Vago, E. R.; Calvo, E. J., J. Chem. Soc., Faraday Trans. 1996, 92, 3371-3379. (25) Feng, Z.; Hong, W. T.; Fong, D. D.; Lee, Y.-L.; Yacoby, Y.; Morgan, D.; Shao-Horn, Y., Acc. Chem. Res. 2016, 49, 966-973. (26) Jacobs, J.-P.; Maltha, A.; Reintjes, J. G.; Drimal, J.; Ponec, V.; Brongersma, H. H., J. Catal. 1994, 147, 294300. (27) Lee, W.; Han, J. W.; Chen, Y.; Cai, Z.; Yildiz, B., J. Am. Chem. Soc. 2013, 135, 7909-7925. (28) Lee, H. B.; Prinz, F. B.; Cai, W., Acta Mater. 2010, 58, 2197-2206. (29) Du, Y.; Zhu, Y.; Xi, S.; Yang, P.; Moser, H. O.; Breese, M. B.; Borgna, A., J. Synchrotron Radiat. 2015, 22, 839-843.

15 ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

16 ACS Paragon Plus Environment

Page 16 of 17

Page 17 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

TABLE OF CONTENTS

17 ACS Paragon Plus Environment