1. Overview of Our Research

     The coupled challenges of a doubling in the world’s energy needs by the year 2050 and the ever-increasing demands for “clean” energy sources have resulted in increased attention worldwide to the possibility of a “hydrogen economy” as a long-term solution for securing energy future. While the hydrogen economy offers a compelling vision of an energy future for the world that is abundant, clean, flexible, and secure, significant scientific and technical challenges should be addressed to achieve its implementation.
     The key components for hydrogen-based energy cycles are integrated electrochemical energy devices such as fuel cells, water electrolyzers, and solar fuel systems. The performance of these energy conversion devices depends critically on the efficiency and durability/stability of catalysts for electrochemical reactions at the electrodes of these devices. The reactions include the electrocatalytic hydrogen oxidation reaction (HOR) and oxygen reduction reaction (ORR), which take place on the anode and cathode of a hydrogen fuel cell, respectively; and the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) at the cathode and anode of a water electrolyzer, respectively. These reactions involve multi-electron transfers and are kinetically demanding. Hence, precious metal-based materials such as Pt, Ru, or Ir with high reaction kinetics have been prevalent choice of catalysts. However, precious metal-based catalysts commonly show declining activity during long-term operation and are susceptible to poisoning; furthermore, their prohibitively high cost and scarcity have also been bottlenecks that impede the widespread use of fuel cells and water electrolyzers. Hence, the development of economic electrocatalysts with high activity and durability/stability has been of utmost importance in this area of research.


     Since the inception of our research group in 2010, we have endeavored to build a research program under the unified theme of “Materials chemistry and electrocatalysis of renewable energy conversion reactions”. Our efforts have been focused on (i) developing highly active, stable, and cost-effective electrocatalysts for renewable energy conversion reactions, (ii) identifying the activity descriptor and active sites of catalysts by exploiting in situ spectroscopic methods in combination with theoretical calculations, and (iii) translating the newly developed catalyst for device-level implementation.
     Our group has been involved in the following projects: (i) ORR electrocatalysts for fuel cells, (ii) bifunctional OER-ORR electrocatalysts for regenerative fuel cells, and (iii) HER electrocatalysts for water electrolyzers.

2. ORR Electrocatalysts for Fuel Cells

     Polymer electrolyte fuel cells (PEFCs) with hydrogen fuel have been considered a promising power source of alternative energy because of their high efficiency, environmental benignity, the reusability of exhaust heat, and their flexibility for mobile, transportation, and stationary applications. Of two electrode reactions, the ORR taking place at the cathode requires the transfer of four electrons coupled with protons and the cleavage of strong double bond, rendering this reaction very sluggish. Hence, electrocatalysts for the ORR play a pivotal role in dictating the overall performance of PEFCs. Pt-based electrocatalysts have hitherto been predominantly used for the ORR; however, their low durability, very high cost, and scarcity has hampered the widespread deployment of PEFC systems. In this context, tremendous efforts have been geared towards the development of highly active and stable, yet low-cost ORR electrocatalysts that are based on Pt-free or low-Pt compositions. In this project, we have been focused on three classes of electrocatalysts: (i) metal-free carbon-based catalysts, (ii) non-precious metal M–N/C (M=Fe and/or Co) catalysts, and (iii) Pt-based catalysts.

2.1. Metal-free, carbon-based ORR catalysts
     Over the last few years, various carbon nanostructures, including carbon nanotubes, graphene, and nanoporous carbons that are doped with various heteroatoms have been exploited as electrocatalysts for the ORR in alkaline media. Despite the rapid progress in doped nanocarbon-based catalysts, there remains a multitude of challenges, which include the lack of fundamental understanding on the working principles that underpin the ORR activity in the doped nanocarbons, relatively lower ORR activity compared to Pt/C catalysts, and only sporadic demonstration of these catalysts in alkaline fuel cell systems.


     In this area, we discovered that the nanoscale work function of doped nanocarbons is strongly correlated with the activity and reaction kinetics of doped nanocarbon catalysts for the ORR [1]. This work represents the first experimental evidence revealing the critical role of work function as a governing factor of ORR activity.
     Based on the above work, we developed a CNT-based core-sheath nanocomposite that comprise CNT core and multiple heteroatom-doped carbon (HDC) sheath layers [2]. The CNT/HDC nanostructures showed excellent electrocatalytic activity for the ORR, which is one of the best performances among heteroatom-doped nanocarbon catalysts in alkaline media. Importantly, the CNT/HDC nanostructures showed very high current and power densities when employed as the cathode catalyst in an alkaline anion exchange membrane fuel cell (AEMFC).

[1] Intrinsic Relationship between Enhanced Oxygen Reduction Reaction Activity and Nanoscale Work Function of Doped Carbons.J. Am. Chem. Soc. 136, 8875 (2014).
[2] Carbon Nanotubes/Heteroatom-Doped Carbon Core-Sheath Nanostructures as Highly Active, Metal-Free Oxygen Reduction Electrocatalysts for Alkaline Fuel Cells.Angew. Chem. Int. Ed. 53, 4102 (2014).

2.2. Non-precious metal M-N/C (M=Fe and/or Co) ORR catalysts
     Among various classes of non-precious metal catalysts, M–N/C (M=Fe and/or Co) catalysts have been considered as the most attractive candidates that can replace Pt-based catalysts in acidic media, owing their high ORR activity. Although synthetic optimization in recent years has led to improved activities and durability of M–N/C catalysts, their ORR activities had been still lower than Pt-based catalysts in acidic electrolytes. Furthermore, high-performance M–N/C catalysts commonly requires complex preparatory steps and the use of toxic reactive gas such as ammonia. Finally, the identification of active site structure in M–N/C catalysts remains elusive due to the high-temperature annealing step during the preparation of these catalysts. We have made multi-directional efforts to address the above-mentioned issues [3].
     We developed a simple, scalable, and highly reproducible synthesis route to highly active M–N/C catalysts, self-supported, transition metal-doped ordered mesoporous porphyrinic carbons (M-OMPCs; M=Fe and/or Co), which exhibit Pt-like catalytic activity for the ORR in acidic media [4]. The FeCo-OMPC showed an extremely high electrocatalytic activity for ORR in acidic media, due to its large surface area, hierarchical micro-mesoporosity, and the formation of a high density of active sites; its activity is one of the best among M–N/C catalysts. Density functional theory (DFT) calculations suggested a weakening of the interaction between oxygen atom and FeCo-OMPC compared to Pt/C, thereby enhancing the ORR activity of FeCo-OMPC.
     The preparation of M–N/C catalysts commonly involve a high-temperature annealing to endow high conductivity as well as structural integrity, thereby giving rise to high activity and stability. However, this step inevitably generates a heterogeneity of active sites that include an atomically dispersed M–Nx site and a metal encapsulated within carbon shell (M@C) site. There has been a continued controversy over a genuine active site for the ORR.


     We prepared a series of Fe–N/C model catalysts that selectively comprised Fe–Nx and Fe@C sites [5]. It was revealed that Fe−Nx sites dominantly catalyze ORR via 4-electron (4 e) pathway, exerting a major role for high ORR activity, whereas Fe@C sites mainly promote 2 e reduction of oxygen followed by 2 e peroxide reduction, playing an auxiliary role. Based on this study, we developed a general “silica-protective-layer-assisted” synthesis that could preferentially yield catalytically active Fe–Nx sites while suppressing the formation of large Fe-based particles [6,7]. The catalyst synthesis includes a silica coating step before high-temperature pyrolysis step, which was found to preserve the Fe−Nx sites. As a result, Fe–N/C catalysts prepared with the silica coating step contained a higher density of active Fe−Nx sites compared to catalysts without silica coating. The resulting catalysts showed very high ORR activity and excellent stability in alkaline media, and when employed as cathode catalyst they demonstrated excellent performances for an alkaline AEMFC as well as acidic proton exchange membrane fuel cell (PEMFC).
     Toward uncovering a new active site for M–N/C catalysts, we constructed model hybrid catalysts by the reaction of an organometallic complex, [CoII(acac)2] (acac=acetylacetonate), with N-doped graphene-based materials at room temperature [8]. It was revealed that the cobalt-containing species is coordinated to heterocyclic groups in N-doped graphene as well as to its parental acac ligands. The hybrid material showed high electrocatalytic activity for the ORR in alkaline media, and superior durability and methanol tolerance to a Pt/C catalyst. Based on the chemical structures and ORR experiments, we could identify a new active species for the ORR: “Co–O4–N” structure.

[3] Strategies for Enhancing the Electrocatalytic Activity of M‒N/C Catalysts for the Oxygen Reduction Reaction.Top. Catal. 61, 1077-1100 (2018).
[4] Ordered Mesoporous Porphyrinic Carbons with Very High Electrocatalytic Activity for the Oxygen Reduction Reaction.Scientific Reports 3, 2715 (2013).
[5] Roles of Fe–Nx and Fe–Fe3C@C Species in Fe–N/C Electrocatalysts for Oxygen Reduction Reaction.ACS Appl. Mater. Interfaces 9, 9567-9575 (2017).
[6] A General Approach to Preferential Formation of Active Fe–Nx Sites in Fe-N/C Electrocatalysts for Efficient Oxygen Reduction Reaction.J. Am. Chem. Soc. 138, 15046-15056 (2016).
[7] Promoting Oxygen Reduction Activity of Fe-N/C Electrocatalysts by Silica Coating-Mediated Synthesis for Anion Exchange Membrane Fuel Cells.Chem. Mater. 30, 6684-6701 (2018).
[8] Coordination Chemistry of [Co(acac)2] with N-Doped Graphene: Implications for Oxygen Reduction Reaction Reactivity of Organometallic Co-O4-N Species.Angew. Chem. Int. Ed. 54, 12622-12626 (2015).

2-3. Pt-based ORR catalysts
     Pt-based catalysts have been the best-performing ORR catalysts, and most of commercial ORR catalysts comprise Pt-based compositions. Our efforts in this direction have been (i) the design of multi-metallic electrocatalysts based on skeletal nanostructures that can substantially reduce the amount of Pt while enhancing maximize the ORR activity and (ii) the development of Pt-based intermetallic nanostructures toward highly active and durable electrocatalysts for the ORR.


     In the former approach, we designed and prepared octahedral Pt-Ni skeletal nanostructures by exploiting a phase segregation phenomenon in a Pt@Ni core-shell octahedron in combination with the selective etching of Ni-rich region [9]. The resulting hollow Pt-Ni skeletal nanostructures showed very high ORR activity in acidic media, ranking one of the best ORR activities among the Pt-based catalysts. While this skeletal nanostructure could provide very high activity, it is inherently subject to degradation during a prolonged electrochemical operation due to abundance of low-coordination sites such as vertices. In this context, a vertex-reinforced skeletal Pt-Cu-Co ternary structure designed, which showed enhanced ORR activity and durability [10].
     For practical implementation of newly developed ORR catalysts in fuel cell devices, their durable operation and long-term stability are as important as their intrinsic activity. Although many of recent nanocatalysts showed very high ORR activity, far exceeding the targets of US Department of Energy (DOE), their durability is comparatively underdeveloped. Intermetallic structures wherein Pt and a transition metal are arranged in an atomically ordered manner have shown great promise as active and durable ORR catalysts, compared to Pt-based alloys with random atomic arrangement. However, the origin of enhanced activity and durability of intermetallic structure is not yet elucidated fully. Further, their implementation into single cells for PEMFCs have rarely been reported. Our group tackled this issue by preparing carbon-free, self-supported Pt-based catalysts with intermetallic structures [11].

[9] Skeletal Octahedral Nanoframe with Cartesian Coordinates via Geometrically Precise Nanoscale Phase Segregation in a Pt@Ni Core-Shell Nanocrystal.ACS Nano 9, 2856 (2015).
[10] Vertex-Reinforced PtCuCo Ternary Nanoframes as Efficient and Stable Electrocatalysts for the Oxygen Reduction Reaction and the Methanol Oxidation Reaction.Adv. Funct. Mater. 28, 1706440 (2018).
[11] Self-Supported Mesostructured Pt-Based Bimetallic Nanospheres Containing an Intermetallic Phase as Ultrastable Oxygen Reduction Electrocatalysts.Small 12, 5347-5353 (2016).

3. Bifunctional OER-ORR Electrocatalysts

     Bifunctional oxygen electrocatalysis involving both the OER and ORR are ubiquitous and play a pivotal role in energy conversion and storage devices, such as unitized regenerative fuel cells (URFCs) and metal-air batteries. Similar to ORR catalysts, precious metals such as IrO2 and RuO2 have thus far been the prevalent choice of the materials for the OER. Hence, non-precious metal-based bifunctional electrocatalysts, including nanostructured carbons doped with transition metals and heteroatoms and metal oxides, have been actively pursued [12]. However, the realization of high catalytic activity for both reactions using these non-precious metal catalysts remains a challenge.
     We designed and prepared highly integrated, high-performance, bifunctional oxygen electrocatalysts composed of highly graphitic nanoshells embedded in mesoporous carbon (GNS/MC) [13]. The GNS/MC exhibited very high oxygen electrode activity, which is one of the best performances among non-precious metal bifunctional oxygen electrocatalysts, and substantially outperformed Ir- and Pt-based catalysts for the OER and ORR, respectively. Moreover, the GNS/MC showed excellent durability for both the OER and ORR. The presence of Ni and Fe species coordinated to nitrogen was found to be critical in enhanced activity and durability. In aqueous Na-air battery tests, the GNS/MC air cathode-based cell exhibited superior performance to Ir/C- and Pt/C-based batteries. Significantly, the GNS/MC-based cell demonstrated the first example of rechargeable aqueous Na-air battery.


     Metal oxides represent another prominent class of non-precious metal OER-ORR catalysts. Using the most widely explored cobalt oxides as model catalysts, we tried to identify active sites during the OER and ORR using in situ X-ray absorption spectroscopy [14]. We prepared four different sized cobalt oxide nanoparticles (NPs) from 3 to 10 nm with varying ratios of CoO and Co3O4. Under OER conditions, all cobalt oxide nanocatalysts underwent a structural transformation into cobalt oxyhydroxide phase, regardless of their size, suggesting cobalt oxyhydroxide as a common active species during the OER. The OER activity increased with decreased NP size, which correlated to the enhanced oxidation state and larger surface area in smaller NPs, whereas the ORR activity was independent of NP size.
     Among metal oxides, perovskites constitute a highly intriguing class of materials as they can host a myriad of compositions and structures. We developed a high-performance oxygen electrocatalyst based on a triple perovskite, which showed superior activity and durability for oxygen electrode reactions to single and double perovskites [15]. When hybridized with nitrogen-doped reduced graphene oxide (N-rGO), the resulting NBCFM/N-rGO catalyst showed further boosted bifunctional oxygen electrode activity, which surpasses that of Pt/C and Ir/C catalysts and which, among the perovskite-based electrocatalysts, is the best activity reported to date. The superior catalytic performances of the triple perovskite was correlated to its oxygen defect–rich structure, lower charge transfer resistance, and smaller hybridization strength between O 2p and Co 3d orbitals.


     Although carbon-based and metal oxide-based materials promising catalytic activity and sufficient stability in alkaline electrolytes, their intrinsic instability hampers their use in acidic electrolytes. Hence, the precious-metal-economic design of Ir or Ru-based catalysts is of prime importance. By exploiting phase-segregation and selective etching phenomena, we prepared skeletal structures of IrO2 and RuO2, which demonstrated prominent OER activity in acidic conditions [16,17].

[12]  Water Electrolysis: A Magnetic Boost.
        Nature Energy 3, 451-452 (2018).
[13] Graphitic Nanoshell/Mesoporous Carbon Nanohybrids as Highly Efficient and Stable Bifunctional Oxygen Electrocatalysts for Rechargeable Aqueous Na-Air Batteries. Adv. Energy Mater. 6, 1501794 (2016).
[14] Size-Dependent Activity Trends Combined with In Situ X-Ray Absorption Spectroscopy Reveal Insights into Cobalt Oxide/Carbon Nanotube Catalyzed Bifunctional Oxygen Electrocatalysis.ACS Catal. 6, 4347-4355 (2016).
[15] Oxygen-Deficient Triple Perovskites as Highly Active and Durable Bifunctional Electrocatalysts for Oxygen Electrode Reactions.Science Advances 4, eaap9360 (2018).
[16] Iridium-Based Multimetallic Nanoframe@Nanoframe Structure: An Efficient and Robust Electrocatalyst toward Oxygen Evolution Reaction.ACS Nano 11, 5500-5509 (2017).
[17] Topotactic Transformations in an Icosahedral Nanocrystal to Form Efficient Water-Splitting Catalysts.Adv. Mater. Published on the Web (2018).

4. HER Electrocatalysts for Water Electrolyzers

     Hydrogen has been of pivotal importance as a clean energy carrier for the realization of a hydrogen economy. Current hydrogen production depends predominantly on the steam methane reforming. Water splitting, when coupled with renewable energy sources such as solar energy, is considered an ideal method for hydrogen production, owing to its unparalleled capacity and carbon-neutral nature. For water electrolyzers, expensive Pt is the most effective catalyst for the HER. Hence, intensive efforts have been devoted to replacing Pt-based catalysts with non-precious metal catalysts, among which metal sulfides have emerged as one of the most promising classes of catalysts.
     In this endeavor, we investigated the nanoscale size effect in MoS2-based HER catalysts [18]. We developed a “confined nanospace” approach to generate MoS2 nanoplates with monolayer precision from one to four layers with the nearly constant basal plane size of 5 nm. HER activity trends over this series of model catalysts revealed that the turnover frequency (TOF) increases with decreased layer numbers, which could be attributed to the higher degree of oxidation, higher Mo–S coordination number, formation of the 1T phase, and lower activation energy required to overcome transition state.


     We also investigated the growth behavior of tungsten sulfide nanostructure within confined nanospace [19]. It was found that initially generated ultra-small WSx nanoclusters preferentially grow toward horizontal direction to yield monolayer WS2 nanoplate. Interestingly, WSx nanoclusters showed higher HER activity compared to WS2 nanoplates, which was attributed to the larger density of active bridging S22- sites.

[18] Monolayer-Precision Synthesis of Molybdenum Sulfide Nanoparticles and Their Nanoscale Size Effects in the Hydrogen Evolution Reaction.ACS Nano 9, 3728-3739 (2015).
[19] Preferential Horizontal Growth of Tungsten Sulfide on Carbon and Insight into Active Sulfur Site for the Hydrogen Evolution Reaction.Nanoscale 10, 3838-3848 (2018).
[20] Facet-Controlled Hollow Rh2S3 Hexagonal Nanoprisms as Highly Active and Structurally Robust Catalysts toward Hydrogen Evolution Reaction.Energy Environ. Sci. 9, 850-856 (2016).