Role of Local Carbon Structure Surrounding FeN4 Sites in Boosting

May 5, 2017 - Development of effective nonprecious metal and nitrogen codoped carbon catalysts for the oxygen reduction reaction (ORR) requires a fund...
0 downloads 12 Views 2MB Size
Download PDF
Article pubs.acs.org/JPCC

Role of Local Carbon Structure Surrounding FeN4 Sites in Boosting the Catalytic Activity for Oxygen Reduction Kexi Liu,† Gang Wu,‡ and Guofeng Wang*,† †

Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, United States ‡ Department of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, New York 14260, United States S Supporting Information *

ABSTRACT: Development of effective nonprecious metal and nitrogen codoped carbon catalysts for the oxygen reduction reaction (ORR) requires a fundamental understanding of the mechanisms underlying their catalytic activity. In this study, we employed the first-principles density functional theory calculations to predict some key parameters (such as activation energy for O−O bond breaking and free-energy evolution as a function of electrode potential) of ORR on three FeN4-type active sites with different local carbon structures. We find that the FeN4 site surrounded by eight carbon atoms and at the edge of micropores has the lowest activation energy (about 0.20 eV) for O−O bond breaking among the three FeN4-type active sites for promoting a direct four-electron ORR. Consequently, our computational results suggest that introduction of micropores in the nonprecious metal catalysts could enhance their catalytic activity for ORR through facilitating the formation of FeN4−C8 active sites with high specific activity.

1. INTRODUCTION A class of pyrolyzed transition metal and nitrogen-derived carbon materials (denoted as TM−N−C) has shown promise as economical yet effective electrocatalysts for the oxygen reduction reaction (ORR) in acid media.1−6 For example, a Fe−N−C catalyst derived from polyaniline and iron was found to exhibit an ORR activity with a half-wave potential (E1/2) by only 59 mV more negative relative to that of a commercial Pt/ C and with a H2O2 yield below 1%.4 Recently, the half-wave potential for ORR in the Fe−N−C catalyst has been improved by another 20−30 mV.7,8 Moreover, a bimetallic (Fe,Mn)−N− C catalyst has been demonstrated to possess improved durability with only 4% activity loss after 9000 potential cycles in acid.8 Despite much progress, there still exists a considerable performance gap between these TM−N−C catalysts and the state-of-the-art platinum group metal (PGM) based catalysts.8 Consequently, it is imperative to acquire fundamental understanding on the mechanisms underlying various synthesis procedures that improve the catalytic performance of the TM− N−C catalysts.9−13 It has been recognized that increasing the microporosity in the TM−N−C catalysts by using microporous carbon black as supports,14 ammonia or oxygen etching,15,16 or sacrificial support method17,18 could lead to pronounced improvement in the ORR activity of the catalysts. A simple explanation to this enhancement in ORR activity might be that the introduced pores enlarge the total area of exposed surface and thus increase © XXXX American Chemical Society

the number of accessible active sites in the TM−N−C catalysts.19 However, quantitative characterization results reveal clearly that the catalytic activity of the Fe−N−C catalysts for ORR is actually correlated with the specific area of those micropores with a size of 5−15 Å rather than the total area of the exposed surface.14,20−22 Hence, this experimental finding strongly suggests that the micropores in the TM−N−C catalysts could host highly active sites for catalyzing ORR. Inspired by this insightful experimental result, we carried out a first-principles density functional theory (DFT) study on how the ORR activity can be enhanced on the FeN4 moieties formed on the edge of micropores as compared to those formed within a graphene layer.

2. COMPUTATIONAL METHOD In this study, we performed spin-polarized DFT23,24 calculations using the Vienna ab initio simulation package (VASP) code.25,26 A projector augmented wave (PAW) pseudopotential27,28 was used to describe the core electrons, and a plane wave basis set with a kinetic energy cutoff of 400 eV was used to expand the wave functions. Electronic exchange and correlation was described within the framework of the generalized gradient approximation (GGA) of Perdew, Burke, Received: January 28, 2017 Revised: April 6, 2017 Published: May 5, 2017 A

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Figure 1. Atomistic structures of the ORR active site (a) FeN4−C10, (b) FeN4−C12, and (c) FeN4−C8 and O2 adsorption with (d) an end-on configuration on FeN4−C10, (e) an end-on configuration on FeN4−C12, and (f) a side-on configuration on FeN4−C8. In the figure, the gray, blue, orange, red, and white balls represent C, N, Fe, O, and H atoms, respectively.

and Ernzernhof (PBE) functionals.29 The Brillouin zone was sampled using a Monkhorst−pack30 4 × 4 × 1 k-point grid for active site FeN4−C10, 3 × 3 × 1 for FeN4−C12, and 4 × 3 × 1 for FeN4−C8. The atomic positions were optimized until the forces were below 0.02 eV/Å during structural optimization. The transition state of the chemical reactions was located using the climbing image nudged elastic band (Cl-NEB) method31 with six intermediate images and with a convergence of 0.05 eV/Å for the force components both along and perpendicular to the tangent of the reaction path. Zero-point energy (ZPE) corrections were included in all energies reported in this work. 1 ZPE corrections were calculated as ZPE = ∑i 2 hνi , where h is Planck’s constant and vi is the frequency of the ith vibrational mode of binding molecules.

sites are stable only if proper pore structures have already been present in the graphene layer. In acid media, a four-electron pathway of ORR on an active site starts from adsorption of the O2 molecule, undergoes a series of elementary reactions involving OOH, O, and OH, and ends with desorption of H2O. Hence, we calculated the adsorption energies of the ORR species (O2, OOH, O, OH, and H2O) on the three FeN4-type active sites and presented our results in Table 1. It should be noted that, for each ORR Table 1. Predicted Adsorption Energies of Various ORR Species on the Three FeN4 Active Sitesa

3. RESULTS AND DISCUSSION Several models involving FeN4 moieties embedded in carbon have been proposed to explain the activity of the Fe−N−C catalysts for ORR in acid.12,32 In a previous study, we predicted that the four-electron OOH dissociation pathway is both thermodynamically and kinetically favorable on the FeN4 site embedded in an otherwise intact graphitic layer and surrounded by 10 carbon atoms (denoted as FeN4−C10, Figures 1a and S1a).12 In the present study, we used this FeN4−C10 site to study the ORR activity of the Fe−N−C catalysts in the region far away from micropores. Regarding the active sites hosted by micropores, we constructed a FeN4−C12 site (Figures 1b and S1b) near mircopores and with a FeN4 moiety bridging two adjacent zigzag graphitic edges with a porphyrin-like architecture and a FeN4−C8 site (Figures 1c and S1c) on the edge of micropores and with a FeN4 moiety bridging two adjacent armchair-like graphitic edges. It is noted that FeN4− C12 was identified as the primary active site in the Fe−N−C catalysts by Jaouen et al.33 from their analysis of X-ray absorption near-edge structure (XANES) results, whereas FeN4−C8 was proposed to be the ORR active site by Dodelet et al.3,14 based on X-ray photoelectron and diffraction characterization results. Recent XANES studies indicated that the catalytic activity of the Fe−N−C catalysts varied with the amount of the three different types of FeN4 moieties in the sample.33−35 The formation energies of these three FeN4 active sites have been calculated in this study and are presented in Figure S2. Our calculation results suggest that it is indeed energetically favorable to form a FeN4−C10 site from a perfect graphene layer, whereas both FeN4−C12 and FeN4−C8 active

Ead (eV)

FeN4−C10

FeN4−C12

FeN4−C8

O2 OOH O OH H2O

−0.95 −1.71 −4.12 −2.69 −0.13

−0.63 −1.41 −3.89 −2.38 −0.27

−1.67 −1.97 −4.37 −2.83 −0.21

a

Negative value of adsorption energy indicates attractive interaction between the ORR species and the ORR active sites.

species, there are several different adsorption configurations on the FeN4-type active site. In this work, we examined various possible adsorption configurations with the central Fe as the adsorption site (Figure S3) and located the most stable one (Figure S4) for each adsorption system in our restricted search. Correspondingly, the predicted lowest adsorption energy is reported in Table 1. Figure 1d, 1e, and 1f shows the optimized atomistic structures of the lowest energy configuration of the adsorption of O 2 on FeN 4 −C 10 , FeN 4 −C 12, and FeN 4−C 8 sites, respectively. It can be seen that O2 adopts an end-on adsorption configuration with bent Fe−O−O geometry on FeN4−C10 and FeN4−C12 whereas a side-on adsorption configuration with O−O lying along the N−Fe−N bonding direction on FeN4−C8. Our DFT calculations predict that the adsorption energy for O2 is −0.63 eV on FeN4−C12, −0.95 eV on FeN4−C10, and −1.67 eV on FeN4−C8, suggesting a trend that the binding strength of O2 to the FeN4-type active sites becomes stronger with less local carbon atoms around the FeN4 moiety. Indeed, we found that the Fe−O bonding orbital in the O2−FeN4−C8 adsorption system was much deeper below the Fermi energy than those of O2 adsorption on the other two FeN4 sites surrounded by more carbon atoms (Figure S6). These predicted O2 adsorption energies are also found to be B

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

easier on FeN4−C8 than on FeN4−C10 and FeN4−C12 sites, apparently relating to the relatively stronger O2 adsorption on FeN4−C8 than the other sites. Besides the potential-independent O−O bond breaking reaction, the four-electron pathways of ORR include several protonation reactions (such as O2 protonation, OH protonation, and O protonation) whose activation energies vary with external electrode potential. Examining this effect of electrode potential on ORR, we predicted the free energy evolution following the OOH dissociation ORR pathway on the three FeN4-type active sites. These free energy landscapes were calculated using the adsorption energies given in Table 1 and a computational approach (see “Free energy calculation” in Supporting Information) proposed by Nørskov et al.38,39 The step of OOH dissociation into O and OH is included in this diagram. The free energy change and activation energy of this nonelectrochemical step are independent of the electrode potential. Hence, Figure 3 depicts the same reaction pathway

comparable to those predicted for O2 on Pt catalysts (−0.69 eV on Pt(111)36 and −1.10 eV on Pt(100)).37 Moreover, our DFT results in Table 1 indicate that the binding of H2O on the FeN4-type active sites is weaker than the solvation stabilization energy of bulk water by about 0.40 eV. Consequently, we predict here that FeN4−C10, FeN4−C12, and FeN4−C8 active sites are able to attract reactant O2 to initiate the ORR as well as release product H2O to complete the ORR. Following four-electron pathways of ORR, the O−O bond in the adsorbed O2 must be broken through either direct dissociation of O2 or dissociation of OOH, which is after O2 is protonated. It is noted that this O−O bond breaking process is not influenced by the external electrode potential. In Table 2 Table 2. Calculated Activation Energies for O2 Dissociation and OOH Dissociation on the FeN4 Active Sites Ea (eV)

FeN4−C10

FeN4−C12

FeN4−C8

O2 dissociation OOH dissociation

1.19 0.56

1.39 0.72

0.94 0.20

we report our calculated activation energies for the two types of O−O bond breaking reactions (O2 vs OOH) on the three FeN4-type active sites. Our results show that the activation energy for the direct O2 dissociation reaction is always higher than that for the OOH dissociation reaction on the FeN4 sites. In addition, the activation energy for the O2 hydrogenation reaction to form OOH was found to be quite low (for example, 0.04 eV on FeN4−C8). It is thus inferred from our DFT calculation that the OOH dissociation pathway would be kinetically more feasible than the O2 dissociation pathways for ORR on the FeN4-type active sites. Our calculation results in Table 2 also indicate that the OOH dissociation reaction on FeN4−C8 requires an activation energy of only 0.20 eV, which is appreciably lower by at least 0.36 eV than those on the other two FeN4-type active sites. In addition, we find that the central Fe of FeN4−C8 is fully capable of breaking the OOH molecule into O and OH on itself (Figure 2c, Figure S5, Table S2). In contrast, the OOH dissociation reaction on both FeN4−C10 and FeN4−C12 is predicted to require transportation of product OH to the top of the adjacent carbon (Figure 2a and 2b). Consequently, our DFT calculations predict the O−O bond breaking of ORR is fairly

Figure 3. Calculated free energy evolution diagrams for the O2 reduction through an OOH dissociation pathway on the active sites FeN4−C10, FeN4−C12, and FeN4−C8 at a temperature of 300 K and under an electrode potential of 0.69 V in acid medium.

adopted in other studies.40 Our results indicate that all of the elementary ORR reactions will be exergonic and thus thermodynamically favorable below a limiting electrode potential, whereas some reactions involving charge transfer will become endergonic above this limiting electrode potential. In Figure 3 we plot the free energy evolution for ORR under an

Figure 2. Atomistic structures of the initial state, transition state, and final state for the OOH dissociation reaction on (a) FeN4−C10, (b) FeN4−C12, and (c) FeN4−C8 active sites. In the figure, the gray, blue, yellow, red, and white balls represent C, N, Fe, O, and H atoms, respectively. C

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

Figure 4. Calculated partial density of states (pDOS) for Fe, N, and O in the relaxed FeN4−C10, FeN4−C12, and FeN4−C8 sites (upper panels), and the adsorption of the O2 molecule on these three FeN4-type active sites (lower panels). In the lower panels, O1 refers to the O atom close to the Fe atom whereas O2 refers to the other O atom away from the Fe atom in an end-on O2 adsorption configuration; the two O atoms are equivalent in a side-on O2 adsorption configuration.

6% shorter than the Fe−N bond length of 2.02 Å in FeN4−C12. It is noted that the Fe−N bond length in the FeN4−C12 site is very close to 1.99 Å as previously predicted for an isolated Fe− porphyrin molecule.43 In such a square planar ligand field of FeN4 moieties,44 the 3dxy, 4s, 4px, and 4py orbitals of the central Fe would participate in the formation of four Fe−N bonding orbitals with the hybrid 2p orbitals of the surrounding N, whereas the other four 3d orbitals (dx2−y2, dyz, dyz, and dz2) of Fe would transform into four nonbonding d-type orbitals and interact with the electronic orbitals of adsorbates. Illustrating this point, we plot in Figure 4 the partial density of states (pDOS) for the elements in the three FeN4 active sites. It can be seen in the upper panels of Figure 4 that N 2p orbitals mainly overlap the Fe 3dxy orbital to form the Fe−N bonds in the clean FeN4 active sites. More interestingly, we observe in the lower panels of Figure 4 that the 2p orbitals of O2 mainly overlap with the 3dz2 orbital of the central Fe when O2 is adsorbed with an end-on adsorption configuration on FeN4−C10 and FeN4−C12 but overlap with the 3dxz and/or 3dyz orbitals of the central Fe when O2 is adsorbed with a side-on adsorption configuration on FeN4−C8. Our results of the decomposed pDOS (Figure S7) show that the Fe 3dz2 orbital is close to the Fermi energy in FeN4−C10 and FeN4−C12, whereas Fe 3dxz and/or 3dyz are near the Fermi energy in FeN4−C8. Hence, it appears that the orientation of the Fe nonbonding d-type orbitals near their Fermi energy affects the binding strength and adsorption configuration of O2 (or other ORR species) on the FeN4 active sites. Furthermore, the binding strength and optimized configurations of OOH adsorption affect the activation energy for O−O bond breaking on the FeN4 active sites. Our DFT calculations predict that the adsorbed OOH binds to FeN4−C8 the most strongly among the three FeN4 sites (see results in Table 1) and adopts a configuration with the two O atoms nearly parallel to the graphene plane (the angle O−O−Fe is about 100° in Figure 2c). As a result, we find that the O−O bond in the adsorbed OOH (with initial bond length of 1.48 Å)

electrode potential of 0.69 V. Moreover, it can be seen that under 0.69 V the last OH protonation reaction has a zero change in free energy, but all other ORR steps are still exergonic on FeN4−C10. Consequently, we predict that the thermodynamic limiting potential for ORR on FeN4−C10 is 0.69 V. Figure 3 also shows that under 0.69 V the O protonation reaction requires an increase in free energy by 0.04 eV on FeN4−C12 and the OH protonation reaction has an increase in free energy by 0.15 eV on FeN4−C8. Thus, we predict that the thermodynamic limiting potential for ORR is 0.65 and 0.54 V on FeN4−C12 and FeN4−C8 sites, respectively. It should be pointed out that our predictions on the ORR limiting potential for the three FeN4-type active sites are close to a previous theoretical prediction of 0.64 V for FeN4 cluster.41 Therefore, our DFT calculations predict that although all three FeN4-type active sites are able to catalyze the fourelectron ORR under their respective thermodynamic limiting electrode potentials, the OOH dissociation pathway requires the lowest activation energy (about 0.20 eV) on FeN4−C8 which lies on the edge of micropores in the graphitic layer whereas the highest activation energy (about 0.72 eV) on FeN4−C12 which has a porphyrin-like structure. These theoretical predictions agree well with a previous experimental finding that the Fe−N−C catalysts containing the FeN4−C10 and/or FeN4−C8 moieties exhibited enhanced catalytic activity for ORR than the sample containing mainly the porphyrin-like FeN4−C12 moieties.34 Consequently, it could be a practical route to enhance the activity of Fe−N−C catalysts for ORR through controlled fabrication of these highly active FeN4−C8 sites using designed metal−organic frameworks.42 In order to understand the predicted local carbon structuredependent ORR activity, we further performed analysis on the geometric and electronic structures of the three FeN4-type active sites. All three FeN4 moieties are found to have a planar structure embedded in a graphene layer and with a nearly square ligand field (as shown in Figure 1a−c). In their fully optimized structures, the lengths of the Fe−N bond in FeN4− C10 and FeN4−C8 are calculated to be 1.90 Å, which is about D

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C Notes

can be broken fairly easily (with an activation energy of 0.20 eV and final O−O separation of 2.53 Å) on the central Fe atom of FeN4−C8, which is thus predicted to possess enhanced ORR activity than other possible active sites. This result confirms our previous proposal that the side-on adsorption configuration of O2 and OOH would favor easy O−O bond scission on FeN4 macrocyclic complexes.45,46 Comparing the calculated binding strength of various ORR species on the three FeN4-type active sites (Figure S8) we established a linear scaling relation between the adsorption energy of ORR species and the number of local carbon atoms immediately adjacent to the FeN4 moiety. Our finding here is in support of the previous proposition that the position of the nonbonding Fe 3d orbitals could be tuned by the degree of πelectron delocalization on the carbon plane with sp2-hybridized graphitic carbon atoms and is related to the adsorption energy of ORR species.10,43 In this study we find that a lower number of local carbon atoms around a FeN4 moiety (i.e., the FeN4−C8 on the edge of micropores) results in a Fe 3dxz and/or 3dyz orbital near the Fermi energy, strong binding of ORR species, easy O−O bond breaking, and thus enhanced ORR activity.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS G.F.W. acknowledges the research grant from the National Science Foundation (Grant No. DMR-1410597). G.W. is thankful for the financial support from the National Science Foundation (Grant No. CBET-1604392). The authors gratefully acknowledge the computational resources provided by the computer facility at the Center for Simulation and Modeling of the University of Pittsburgh and at the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by the National Science Foundation grant number ACI-1053575.

■

4. CONCLUSIONS In conclusion, we predicted the adsorption energies of ORR species, activation energy for O−O bond breaking, and freeenergy evolution of ORR on three FeN4-type active sites with different local carbon structures using first-principles DFT calculations. Our DFT calculations reveal that the FeN4−C8 sites hosted by micropores could favorably promote the fourelectron ORR below a limiting electrode potential of 0.54 V and break the O−O bond in the adsorbed OOH with a pretty low activation energy of 0.20 eV. Consequently, our theoretical study suggests that introduction of micropores in the nonprecious Fe−N−C catalysts could enhance their intrinsic catalytic activity for ORR through generating the FeN4−C8 active sites with high specific activity. Therefore, this computational study provides useful guidance to the rational design and controlled synthesis of nonprecious metal and nitrogen-derived carbon electrocatalysts.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.7b00913. Extended simulation cells; formation energy of the FeN4 active site; locating the most stable adsorption configuration; atomistic structures of the optimized configurations of various ORR species adsorbed on the three FeN4-type active sites; free energy calculation; determination of the final state for the O2 and OOH dissociation reaction; predicted activation energies of the two different pathways for O2 and OOH dissociation on FeN4−C10, FeN4−C12, and FeN4−C8; electronic structures for O2−FeN4 systems and FeN4 sites; pDOS for decomposed d orbitals of Fe in the FeN4-type active sites; scaling relation for adsorption energies (PDF)

■

REFERENCES

(1) Tang, C.; Zhang, Q. Can Metal-Nitrogen-Carbon Catalysts Satisfy Oxygen Electrochemistry? J. Mater. Chem. A 2016, 4, 4998− 5001. (2) Bashyam, R.; Zelenay, P. A Class of Non-Precious Metal Composite Catalysts for Fuel Cells. Nature 2006, 443, 63−66. (3) Lefèvre, M.; Proietti, E.; Jaouen, F.; Dodelet, J. P. Iron-Based Catalysts with Improved Oxygen Reduction Activity in Polymer Electrolyte Fuel Cells. Science 2009, 324, 71−74. (4) Wu, G.; More, K. L.; Johnston, C. M.; Zelenay, P. HighPerformance Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and Cobalt. Science 2011, 332, 443−447. (5) Wang, M. Q.; Yang, W. H.; Wang, H. H.; Chen, C.; Zhou, Z. Y.; Sun, S. G. Pyrolyzed Fe-N-C Composite as an Efficient Non-Precious Metal Catalyst for Oxygen Reduction Reaction in Acidic Medium. ACS Catal. 2014, 4, 3928−3936. (6) Shao, M. H.; Chang, Q. W.; Dodelet, J. P.; Chenitz, R. Recent Advances in Electrocatalysts for Oxygen Reduction Reaction. Chem. Rev. 2016, 116, 3594−3657. (7) Wang, X. J.; Zhang, H. G.; Lin, H. H.; Gupta, S.; Wang, C.; Tao, Z. X.; Fu, H.; Wang, T.; Zheng, J.; Wu, G.; et al. Directly Converting Fe-Doped Metal Organic Frameworks into Highly Active and Stable Fe-N-C Catalysts for Oxygen Reduction in Acid. Nano Energy 2016, 25, 110−119. (8) Sahraie, N. R.; Kramm, U. I.; Steinberg, J.; Zhang, Y. J.; Thomas, A.; Reier, T.; Paraknowitsch, J. P.; Strasser, P. Quantifying the Density and Utilization of Active Sites in Non-Precious Metal Oxygen Electroreduction Catalysts. Nat. Commun. 2015, 6, 8618. (9) Ferrandon, M.; Kropf, A. J.; Myers, D. J.; Artyushkova, K.; Kramm, U.; Bogdanoff, P.; Wu, G.; Johnston, C. M.; Zelenay, P. Multitechnique Characterization of a Polyaniline-Iron-Carbon Oxygen Reduction Catalyst. J. Phys. Chem. C 2012, 116, 16001−16013. (10) Ramaswamy, N.; Tylus, U.; Jia, Q. Y.; Mukerjee, S. Activity Descriptor Identification for Oxygen Reduction on Nonprecious Electrocatalysts: Linking Surface Science to Coordination Chemistry. J. Am. Chem. Soc. 2013, 135, 15443−15449. (11) Kattel, S.; Atanassov, P.; Kiefer, B. Catalytic Activity of Co-Nx/C Electrocatalysts for Oxygen Reduction Reaction: A Density Functional Theory Study. Phys. Chem. Chem. Phys. 2013, 15, 148−153. (12) Kattel, S.; Wang, G. Reaction Pathway for Oxygen Reduction on FeN4 Embedded Graphene. J. Phys. Chem. Lett. 2014, 5, 452−456. (13) Holby, E. F.; Zelenay, P. Linking Structure to Function: The Search for Active Sites in Non-Platinum Group Metal Oxygen Reduction Reaction Catalysts. Nano Energy 2016, 29, 54−64. (14) Charreteur, F.; Jaouen, F.; Ruggeri, S.; Dodelet, J. P. Fe/N/C Non-Precious Catalysts for PEM Fuel Cells: Influence of the Structural Parameters of Pristine Commercial Carbon Blacks on Their Activity for Oxygen Reduction. Electrochim. Acta 2008, 53, 2925−2938. (15) Kramm, U. I.; Herrmann-Geppert, I.; Bogdanoff, P.; Fiechter, S. Effect of an Ammonia Treatment on Structure, Composition, and Oxygen Reduction Reaction Activity of Fe-N-C Catalysts. J. Phys. Chem. C 2011, 115, 23417−23427.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Guofeng Wang: 0000-0001-8249-4101 E

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C (16) Yan, X.; Gan, L.; Lin, Y.-C.; Bai, L.; Wang, T.; Wang, X.; Luo, J.; Zhu, J. Controllable Synthesis and Enhanced Electrocatalysis of IronBased Catalysts Derived from Electrospun Nanofibers. Small 2014, 10, 4072−4079. (17) Serov, A.; Artyushkova, K.; Andersen, N. I.; Stariha, S.; Atanassov, P. Original Mechanochemical Synthesis of Non-Platinum Group Metals Oxygen Reduction Reaction Catalysts Assisted by Sacrificial Support Method. Electrochim. Acta 2015, 179, 154−160. (18) Serov, A.; Artyushkova, K.; Niangar, E.; Wang, C. M.; Dale, N.; Jaouen, F.; Sougrati, M. T.; Jia, Q. Y.; Mukerjee, S.; Atanassov, P. Nano-Structured Non-Platinum Catalysts for Automotive Fuel Cell Application. Nano Energy 2015, 16, 293−300. (19) Liang, H. W.; Wei, W.; Wu, Z. S.; Feng, X. L.; Mullen, K. Mesoporous Metal-Nitrogen-Doped Carbon Electrocatalysts for Highly Efficient Oxygen Reduction Reaction. J. Am. Chem. Soc. 2013, 135, 16002−16005. (20) Jaouen, F.; Lefèvre, M.; Dodelet, J. P.; Cai, M. Heat-Treated Fe/ N/C Catalysts for O2 Electroreduction: Are Active Sites Hosted in Micropores? J. Phys. Chem. B 2006, 110, 5553−5558. (21) Jaouen, F.; Herranz, J.; Lefèvre, M.; Dodelet, J. P.; Kramm, U. I.; Herrmann, I.; Bogdanoff, P.; Maruyama, J.; Nagaoka, T.; Garsuch, A.; et al. Cross-Laboratory Experimental Study of Non-Noble-Metal Electrocatalysts for the Oxygen Reduction Reaction. ACS Appl. Mater. Interfaces 2009, 1, 1623−1639. (22) Wu, G.; Johnston, C. M.; Mack, N. H.; Artyushkova, K.; Ferrandon, M.; Nelson, M.; Lezama-Pacheco, J. S.; Conradson, S. D.; More, K. L.; Myers, D. J.; et al. Synthesis-Structure-Performance Correlation for Polyaniline-Me-C Non-Precious Metal Cathode Catalysts for Oxygen Reduction in Fuel Cells. J. Mater. Chem. 2011, 21, 11392−11405. (23) Kohn, W.; Sham, L. J. Self-Consistent Equations Including Exchange and Correlation Effects. Phys. Rev. 1965, 140, A1133− A1138. (24) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Phys. Rev. 1964, 136, B864−B871. (25) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 47, 558−561. (26) Kresse, G.; Furthmuller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6, 15−50. (27) Blochl, P. E. Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50, 17953−17979. (28) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 1758−1775. (29) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (30) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B 1976, 13, 5188−5192. (31) Henkelman, G.; Uberuaga, B. P.; Jonsson, H. A Climbing Image Nudged Elastic Band Method for Finding Saddle Points and Minimum Energy Paths. J. Chem. Phys. 2000, 113, 9901−9904. (32) Kattel, S.; Wang, G. A Density Functional Theory Study of Oxygen Reduction Reaction on Me-N4 (Me = Fe, Co, or Ni) Clusters between Graphitic Pores. J. Mater. Chem. A 2013, 1, 10790−10797. (33) Zitolo, A.; Goellner, V.; Armel, V.; Sougrati, M. T.; Mineva, T.; Stievano, L.; Fonda, E.; Jaouen, F. Identification of Catalytic Sites for Oxygen Reduction in Iron- and Nitrogen-Doped Graphene Materials. Nat. Mater. 2015, 14, 937−942. (34) Jia, Q.; Ramaswamy, N.; Hafiz, H.; Tylus, U.; Strickland, K.; Wu, G.; Barbiellini, B.; Bansil, A.; Holby, E. F.; Zelenay, P.; et al. Experimental Observation of Redox-Induced Fe-N Switching Behavior as a Determinant Role for Oxygen Reduction Activity. ACS Nano 2015, 9, 12496−12505. (35) Kramm, U. I.; Lefèvre, M.; Larouche, N.; Schmeisser, D.; Dodelet, J. P. Correlations between Mass Activity and Physicochemical Properties of Fe/N/C Catalysts for the ORR in PEM Fuel Cell via Fe57 Mössbauer Spectroscopy and Other Techniques. J. Am. Chem. Soc. 2014, 136, 978−985.

(36) Duan, Z.; Wang, G. A First Principles Study of Oxygen Reduction Reaction on a Pt(111) Surface Modified by a Subsurface Transition Metal M (M = Ni, Co, or Fe). Phys. Chem. Chem. Phys. 2011, 13, 20178−20187. (37) Duan, Z.; Wang, G. Comparison of Reaction Energetics for Oxygen Reduction Reactions on Pt(100), Pt(111), Pt/Ni(100), and Pt/Ni(111) Surfaces: A First-Principles Study. J. Phys. Chem. C 2013, 117, 6284−6292. (38) Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.; Jonsson, H. Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode. J. Phys. Chem. B 2004, 108, 17886−17892. (39) Peterson, A. A.; Abild-Pedersen, F.; Studt, F.; Rossmeisl, J.; Norskov, J. K. How Copper Catalyzes the Electroreduction of Carbon Dioxide into Hydrocarbon Fuels. Energy Environ. Sci. 2010, 3, 1311− 1315. (40) Studt, F. The Oxygen Reduction Reaction on Nitrogen-Doped Graphene. Catal. Lett. 2013, 143, 58−60. (41) Anderson, A. B.; Sidik, R. A. Oxygen Electroreduction on Feii and Feiii Coordinated to N4 Chelates. Reversible Potentials for the Intermediate Steps from Quantum Theory. J. Phys. Chem. B 2004, 108, 5031−5035. (42) Zhang, H. G.; Osgood, H.; Xie, X. H.; Shao, Y. Y.; Wu, G. Engineering Nanostructures of PGM-Free Oxygen-Reduction Catalysts Using Metal-Organic Frameworks. Nano Energy 2017, 31, 331− 350. (43) Liu, K.; Lei, Y.; Wang, G. Correlation between Oxygen Adsorption Energy and Electronic Structure of Transition Metal Macrocyclic Complexes. J. Chem. Phys. 2013, 139, 204306. (44) Jean, Y.; Marsden, C. Molecular Orbitals of Transition Metal Complexes; Oxford University Press: New York, 2005; p 270. (45) Wang, G. F.; Ramesh, N.; Hsu, A.; Chu, D.; Chen, R. R. Density Functional Theory Study of the Adsorption of Oxygen Molecule on Iron Phthalocyanine and Cobalt Phthalocyanine. Mol. Simul. 2008, 34, 1051−1056. (46) He, H.; Lei, Y.; Xiao, C.; Chu, D.; Chen, R.; Wang, G. Molecular and Electronic Structures of Transition-Metal Macrocyclic Complexes as Related to Catalyzing Oxygen Reduction Reactions: A Density Functional Theory Study. J. Phys. Chem. C 2012, 116, 16038−16046.

F

DOI: 10.1021/acs.jpcc.7b00913 J. Phys. Chem. C XXXX, XXX, XXX−XXX