Monodisperse Ultrasmall Manganese-Doped Multimetallic Oxysulfide

Apr 3, 2018 - Herein, we report a straightforward top-down synthesis of monodisperse ultrasmall manganese-doped multimetallic (ZnGe) oxysulfide ...
0 downloads 0 Views 11MB Size
Subscriber access provided by Kaohsiung Medical University

Energy, Environmental, and Catalysis Applications

Monodisperse Ultrasmall Manganese-Doped Multimetallic Oxysulfide Nanoparticles as Highly Efficient Oxygen Reduction Electrocatalyst Yingying Zhang, Xiang Wang, Dandan Hu, Chaozhuang Xue, Wei Wang, Huajun Yang, Dongsheng Li, and Tao Wu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 03 Apr 2018 Downloaded from http://pubs.acs.org on April 3, 2018

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 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 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.

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 42 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 Applied Materials & Interfaces

Monodisperse Ultrasmall Manganese-Doped Multimetallic Oxysulfide Nanoparticles as Highly Efficient Oxygen Reduction Electrocatalyst Yingying Zhang,† Xiang Wang,*,† Dandan Hu,† Chaozhuang Xue,† Wei Wang,† Huajun Yang,† Dongsheng Li,‡ and Tao Wu*,† †

College of Chemistry, Chemical Engineering and Materials Science, Soochow University,

Suzhou, Jiangsu 215123, China. ‡

College of Materials and Chemical Engineering, Hubei Provincial Collaborative Innovation

Center for New Energy Microgrid, Key Laboratory of Inorganic Nonmetallic Crystalline and Energy Conversion Materials, China Three Gorges University, Yichang, Hubei 443002, China. KEYWORDS: manganese-doped oxysulfide,

chalcogenidometalates, oxygen

reduction

reaction, ultrasonication, non-Pt electrocatalysts, ultrasmall nanoparticles

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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 highly efficient and cheap non-Pt-based electrocatalysts such as transitionbased catalysts prepared via facile methods for oxygen reduction reaction (ORR) is desirable for large-scale practical industry applications in energy conversion and storage systems. Herein, we report a straightforward top-down synthesis of monodisperse ultrasmall manganese-doped multimetallic (ZnGe) oxysulfide nanoparticles (NPs) as efficient ORR electrocatalyst by simple ultrasonic treatment of the Mn-doped Zn-Ge-S chalcogenidometalates crystal precursors in H2O/EtOH for only one hour at room temperature. Thus obtained ultrasmall monodisperse Mndoped oxysulfide NPs with ultralow Mn loading level (3.92 wt%) not only exhibit comparable onset and half-wave potential (0.92 V and 0.86 V vs. RHE, respectively) to the commercial 20 wt% Pt/C, but also exceptionally high metal mass activity (189 mA/mg at 0.8 V) and good methanol tolerance. A combination of transmission electron microscope (TEM), scanning electron microscope (SEM), X-ray photoelectron spectroscopy (XPS) and electrochemical analysis demonstrated that the homogenous distribution of a large amount of Mn(III) on the surface of NPs mainly accounts for the high ORR activity. We believe that this simple synthesis of Mn-doped multimetallic (ZnGe) oxysulfide NPs derived from chalcogenidometalates will open a new route to explore the utilization of discrete-cluster-based chalcogenidometalates as novel non-Pt electrocatalysts for energy applications and provide a facile way to realize the effective reduction of the amount of catalyst while keeping desired catalytic performances.

1. INTRODUCTION The oxygen reduction reaction (ORR) is generally considered as the major bottleneck in many renewable energy conversion and storage processes, such as fuel cells and metal-air batteries, due to its sluggish kinetics.1, 2 To overcome the problem, Pt and Pt-based alloys have been used as the state-of-the-art electrocatalysts due to their high activity. However, the high cost, scarcity 2 ACS Paragon Plus Environment

Page 2 of 42

Page 3 of 42 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 Applied Materials & Interfaces

and relatively poor long-term stability in the operating environment greatly impede their largescaled application.1,

2

Consequently, the ongoing search for alternative cost-effective non-

precious catalysts with earth-abundant elements would facilitate the global scalability of cleanenergy

technologies.2

Toward

this

target,

transition-metal

oxides,3-7

nitrides,1,8-10

oxynitrides,1,11carbonitrides,1 chalcogenides,1, 12 carbon-based noble-metal-free materials13-16 and metal-organic frameworks (MOFs)17-22 have attracted much attention and been extensively studied as effective candidates for the ORR over the last few decades. Among these non-precious metal catalysts, manganese-based compounds (mainly oxides) are widely used as ORR catalysts for alkaline battery due to their rich electrochemical properties, high abundance, low cost, less toxic and rather high activity toward the ORR, which is desirable for commercialization.1,

2, 23

Over the past several decades, manganese oxide catalysts with

improved activity and stability for ORR have been extensively investigated.4, 7, 23-29 The factors affecting catalytic performance of Mn-based oxides include several physical and chemical parameters,23,

30

such as crystallographic structure, phases (crystalline or amorphous), size,

surface area, composition, Mn oxidation state, oxygen defects, and electrical conductivity, etc..2326, 29, 31-34

Among those parameters, the valence state of Mn is a key factor to the ORR

performance as Mn(III) ion in MnOx typically exhibits higher activity than Mn(II) or Mn(IV).23, 27-29

Therefore, attempts have been made to modify the nominal oxidation state of Mn by

introducing different metal ions or anions (such as S and Se) into manganese oxide catalysts to influence the catalytic activity of manganese oxides.26, 33-36 Besides, decreasing the size of NPs has been demonstrated to be an effective strategy for enhancing the catalytic performance as a larger fraction of active metal sites are exposed to the surface of smaller particles, often exhibiting highly enhanced catalytic activities.37 However, manganese-based electrocatalysts

3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 4 of 42

synthesized by various solution-assisted bottom-up methods are often of relative large size and distribution and usually need additional post calcination to obtain high activity.4, 38, 39 Thus, it is a challenge to develop a simple and energy-saving synthetic approach to synthesis monodisperse small manganese-based electrocatalysts with high activity, which avoids multiple postpreparative purification. Chalcogenidometalates, which consider to be porous metal chalcogenides that typically linked through sulfur sites at the corners of each cluster, varying from discrete nanoclusters (NCs) to extended three-dimensional (3D) networks with large pores and cavities,40, 41 are among the most advanced solid-state materials for applications in optoelectronics,40, 42, 43 photocatalysis,44 thinfilm solar cells,45 or ion exchanging for environment governance due to the synergetic effect of both inorganic and organic components.46,

47

Moreover, chalcogenidometalates, especially

composed of isolated NCs, may be useful in electrochemistry, as the possible synergetic effect in electrochemical reactions caused by electrochemically active multimetallic (ZnGe) ions coexisted in isolated chalcogenides NCs. In addition, the NCs could be dispersed in some solvent uniformly if possible,48,

49

which is beneficial for solution processing to complex with other

functional materials such as carbon black, graphene and CNTs etc.45, 48 Nevertheless, few works have been reported about the exploitation of chalcogenidometalates as electrocatalysts for ORR. The reason for it may be their poor dispersibility in common solvents due to the strong ionic force in crystals, which impedes the application of chalcogenidometalates in electrochemistry. Thus, we conjecture that some new nanomaterials can be obtained with the assistance of ultrasonication just like other nanomaterials (graphene,50,

51

quantum dots,52 perovskite

nanocrystals53-55 etc.) synthesized by liquid exfoliation under ultrasonication, and be effectively utilized to electrochemical applications.

4 ACS Paragon Plus Environment

Page 5 of 42 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 Applied Materials & Interfaces

Herein, we developed a simple top-down strategy to obtain monodisperse ultrasmall manganese doped multimetallic (ZnGe) oxysulfide NPs as a new efficient electrocatalyst for ORR in 1 M KOH by ultrasonication of Mn-doped P1-type chalcogenidometalate crystals as precursor in H2O/EtOH (4:1, V/V) for 1h at room temperature. The deliberately chosen P1-type chalcogenidometalate is composed of discrete P1-type supertetrahedral NCs (Pn, P means pentasupertetrahedra and n is the layers of tetrahedral chalcogenide cluster in the supertetrahedral cluster)56, 57 and doped with trace amount of manganese.58 The monodisperse NPs of ca. 3 nm obtained from bulk P1-type chalcogenidometalate crystal precursors via ultrasonic treatment show comparable ORR activity to commercial Pt/C catalyst via 4e- transfer reduction pathway, as well as long-term stability and good methanol tolerance in alkaline media. In addition, the Mn-doped multimetallic (ZnGe) oxysulfide NPs exhibit high metal mass activity with ultralow Mn content (3.92 wt%) compared to other Mn-based compounds.7,

59, 60

This

superior catalytic activity of thus obtained NPs is probably related to the ultrasmall size, homogenous distribution, existence of large amount Mn(III) on the surface and the dopant effect of multimetallic (ZnGe) cations, which derived from the composition of chalcogenidometalates. To the best of our knowledge, this is the first report about utilization of chalcogenidometalates bulk crystals as precursor to prepare ORR electrocatalyst with highly efficient activity. Compared with traditional bottom-up chemical synthesis of MnOx, this top-down ultrasonic approach is convenient due to the short reaction time and moderate condition, allowing us to skip multistep post-preparative purification. This study not only offers a new route to develop new kind of non-Pt ORR electrocatalysts with ultralow amount of Mn species while maintaining high activities but also paves the potential way for discrete-cluster-based chalcogenidometalates as novel electrocatalysts to be utilized in more energy conversion and storage processes and

5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 6 of 42

technologies, such as hydrogen evolution reaction (HER), oxygen evolution reaction (OER), CO2 reduction and lithium-sulfur batteries with improved active metal utilization efficiency by simplified handling procedures. 2. EXPERIMENTAL SECTION 2.1. Chemicals. K2Sx (AR, powder, ACROS), GeO2 (AR, powder, HWRK), S (AR, powder, Sinopharm), Zn(Ac)2·2H2O (AR, powder, MACKLIN), Mn(Ac)2·4H2O (AR, powder, MACKLIN), Mn2O3 (98%, powder, MACKLIN), KOH (GR, Sinopharm Group Co., Ltd), 1,5diazabicyclo[4.3.0]non-5-ene (DBN, C7H12N2, 98 %, liquid, MACKLIN), Pt/C (20 wt%, Johnson Matthey), KB (Ketjen black, powder, Innochem) were all used as supplied without any further purification. Milli-Q water (Millipore, 18.2 MΩ/cm) was used for all experiments. 2.2. Synthesis of K6[Ge4Zn4S13(OH)4](C7H12N2)0.5(H2O)X (denoted as P1). A mixture of S powder (180.0 mg, 5.625 mmol), K2Sx powder (90.0 mg), Zn(Ac)2·2H2O (60.0 mg, 0.27 mmol), GeO2 (50.0 mg, 0.48 mmol), DBN (2.0 mL) and H2O (1.0 mL) were placed in a sealed autoclave and stirred for 30 min. Then the vessel was heated to 200 °C for 8 days. After cooling to room temperature, a large amount of light pink cubic crystals were obtained. The chemical formula was determined by single-crystal X-ray diffraction analysis, elemental analysis (EA) and SEM/EDS (see supporting information). Yield of P1 is 68.5mg, 43.0% based on Ge. Elemental analysis, calcd. (wt %): C, 2.98; N,1.01; H, 1.01; found: C, 2.97; N, 1.07; H, 1.14; 2.3. Synthesis of K6[Ge4Mn1.2Zn2.8S13(OH)4](C7H12N2)0.5(H2O)X (denoted as P1-Mn). A mixture of S powder (180.0 mg, 5.63 mmol), K2Sx powder (150.0 mg), GeO2 (50.0 mg, 0.48 mmol), Zn(Ac)2·2H2O (35.0 mg, 0.16 mmol), Mn(Ac)2·4H2O (27.0 mg, 0.11 mmol), DBN (2.0 mL) and H2O (1.0 mL) were placed in a sealed autoclave and stirred for 30 min. The vessel was heated to 200 °C for 8 days. A large amount of laurel-green cubic crystals were obtained. The 6 ACS Paragon Plus Environment

Page 7 of 42 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 Applied Materials & Interfaces

chemical formula was determined by single-crystal X-ray diffraction analysis, elemental analysis (EA) and SEM/EDS (see supporting information). The percentage of Mn was determined by inductively coupled plasma mass spectrometer (ICP-MS) (Table S1). Yield of P1-Mn is 22.8 mg, 14.5 % based on Ge. Elemental analysis, calcd. (wt %): C, 3.01; N, 1.00; H, 1.00; found: C, 2.95; N, 1.24; H, 1.13. 2.4. Synthesis of manganese-doped multimetallic (ZnGe) oxysulfide NPs (denoted P1-Mn NPs). A mixture of P1-Mn crystals (2 mg), H2O (400 µL) and ethanol (100 µL) was added into a glass bottle and then ultrasonicated strongly for 1 h. The frequency of the ultrasonication is 40KHz, and the power is 240W. We carried out the ultrasonication at room temperature (293 K). The percentage of Mn and other metals were determined by inductively coupled plasma mass spectrometer (ICP-MS) (Table S1 and Table S2). 2.5. Physical characterization. The physical characterization methods of as-synthesized crystalline samples, including powder X-ray diffraction patterns (PXRD), elemental analysis (EA) of C, H, and N, scanning electron microscope (SEM), energy dispersive spectroscopy (EDS) analysis, solid-state UV-Vis diffusion reflectance and X-ray photoelectron spectroscopy (XPS), were the same as referred in previously reported reference.61 The UV-Vis absorption spectra were calculated by using Kubelka-Munk function: F(R) = α/S = (1-R)2/2R, where R, α, and S are the reflection, the absorption and the scattering coefficient, respectively. The XPS spectra were all calibrated by C1s line at 284.6 eV. An iCAPQ (Thermofisher Scientific) inductively coupled plasma mass spectrometer (ICP-MS) was used to determine the percentage of Mn of P1-Mn, P1Mn NPs and other metals of P1-Mn NPs. Fourier transform-infrared spectral analysis was performed on a Thermo Nicolet Avatar 6700 FT-IR spectrometer with cesium iodide optics allowing the instrument to observe from 400 to 4000 cm-1. TEM, high-resolution TEM

7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 8 of 42

(HRTEM), angle annular dark field (HAADF)/scanning transmission electron microscopy (STEM) (HAADF/STEM) and elemental mapping images were measured on a FEI Tecnai F20 TEM at an acceleration voltage of 200 kV. The samples were prepared by water/ethanol (4:1, V/V) dispersion of samples onto ultrathin carbon-coated copper TEM grids using micropipettes and dried under ambient condition. Electron paramagnetic resonance (EPR) measurement was carried out at -150 °C on a JEOL JES-X320 spectrometer with the settings of center field (325 mT), microwave frequency (9157.288 MHz), MOD frequency (100.00kHz), width (0.0400mT) and power (0.998 mW). 2.6. Single Crystal Characterization. The single crystal characterization and the processing of single crystal data of P1 crystal are the same as referred in previous report.61 The crystal data and refinement details of P1 are summarized in Table S1. 2.7. Electrochemical measurements. Electrochemical experiments were performed in a three-electrode electrochemical cell and conducted on a Rotating Ring Disk Electrode (RRDE) apparatus (RRDE-3A, BAS Inc.) equipped with computer-controlled CHI-760E. A rotating disk electrode (RDE) with a glassy carbon (GC) disk (φ = 4 mm) was used as the working electrode. A KCl saturated no-leak Ag/AgCl electrode and a graphite rod were used as the reference electrode and counter electrode, respectively. The procedure of pretreatment and modification of the working electrode is as follows: it was firstly polished mechanically with 0.05 mm alumina slurry to obtain a mirror-like surface and then rinsed thoroughly with water, ethanol, and water and then allowed to dry. The homogeneous inks were prepared by dispersing 2 mg catalyst and 0.5 mg Ketjen black in mixture of 400 µL water, 100 µL ethanol and 10 µL 5% Nafion (5 wt% in propanol, Alfa Aesar) which were denoted as P1/KB and P1-Mn/KB, respectively. The mixture was then sonicated strongly for 1 h to form a uniform catalyst ink. 14 µL of as-prepared catalyst

8 ACS Paragon Plus Environment

Page 9 of 42 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 Applied Materials & Interfaces

ink with catalyst loading of 0.437 mg/cm2 was coated onto the pre-cleaned working electrode and dried at room temperature for the following electrochemical measurements. For comparison, the commercial 20 wt% Pt/C was used with Pt loading amount of 43.7 µg/cm2 by the same modification but with 7 µL of as-prepared catalyst ink. All the potentials used in this study were reported relative to the reversible hydrogen electrode (RHE). Cyclic voltammetry (CV) was performed at a scanning rate of 50 mV/s after purging high pure O2 or N2 gas for 30 min. Linear sweep voltammetry (LSV) was performed with the RDE in the O2-saturated 1.0 M KOH solution at rotation speeds varying from 600 to 2500 rpm with a scan rate of 2 mV/s. During the measurements, a gentle flow of O2 or N2 was maintained above the electrolyte. The background currents were collected in N2-saturated electrolyte and corrected for the LSVs under O2. Note that all electrochemical tests were not iR compensated and performed at room temperature. 2.8. Calculation method. The electron transfer number (n) during the ORR was calculated at various electrode potential according to the Koutecky-Levich (K-L) equation:62 1 1 1 1 1 = + = + 1/ 2 j jL jK Bω jK

(1)

B = 0.62nFC0 D03/ 2ν −1/6

(2)

J K = nFkC0

(3)

Where J represents the measured current density, JK and JL are the kinetic- and diffusion-limiting current densities, ω is the angular velocity (ω = 2πN, N is the electrode rotation rate expressed in rpm), n is the number of electrons transferred, F is the Faraday constant (96485 C/mol), C0 is the bulk concentration of O2 (8.3×10-7 mol/cm3 in 1.0 M KOH), D0 is the diffusion coefficient of O2 (1.65×10-5 cm2/s), ν is the kinematic viscosity of the electrolyte (0.00947 cm2/s), and k is the electron-transfer rate constant.62 To detect HO2- yield, a rotating ring disk electrode (RRDE)

9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 10 of 42

(ALS, Japan) with a GC disk (φ = 4 mm) and a Pt ring was used as working electrode. The working electrode was scanned at a rate of 2 mV/s while keeping the ring electrode at 1.3 V (vs. RHE) and rotated at 1600 rpm in 1.0 M KOH. The HO2- yield and electron transfer number (n) were calculated by following equations:

n=

4 I disk I ring

N

(4)

+ I disk

200 HO −2 % =

I ring

N

I ring N

(5)

+ I disk

where Idisk is the disk electrode current, Iring is the ring electrode current and N is collection efficiency of the ring electrode (Pt). N was determined to be 0.424 according to the literature.63

3. RESULTS AND DISCUSSION As shown in Figure 1a, pure P1 and P1-Mn bulk crystals were successfully synthesized via one-pot hydrothermal process without other co-products, which was much more convenient than traditional approaches that usually required high temperature (generally above 500 oC) or multiple procedures under strict waterless anaerobic operation using standard glove box and Schlenk line techniques.56, 58, 64-67 The morphology, composition and optical band gap of the assynthesized crystals were characterized by microscope photographs, SEM/EDS, UV-Vis absorption and infrared spectral measurements, respectively (Figure S1 and S2). The content of Mn dopants in P1-Mn crystals was determined by ICP-MS (Table S1), accounting 5.34 % of the total mass. The crystal data and refinement details of P1 were summarized in Table S3. According to the crystal structure, the basic structure units of P1 and P1-Mn clusters are illustrated in Figure 1b. Compared with formerly reported P1-type chalcogenidometalates that 10 ACS Paragon Plus Environment

Page 11 of 42 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 Applied Materials & Interfaces

are very sensitive to oxygen and humid air,56, 58, 64-67 our synthesized compounds are stable under air condition for at least one month as demonstrated in Figure S3. Such enhanced stability is probably attributed to the replacement of S2- by OH- at the four corners of the P1 cluster according to the crystal structure refinement data (see Figure S4), resulting in less charges of clusters and correspondingly less potassium cations as counter ions for balancing the charge of P1 cluster (Figure S4). In addition, some protonated organic amine templates disorderly locating in the crystal lattice, as shown in the infrared spectroscopy of Figure S2b, also reinforce the stability of as-synthesized compounds in air to some extent. This is in favor of exploring the applications of chalcogenidometalates. As P1-Mn crystals are composed of isolated P1-type NCs, we tried to obtain small isolated NCs from the P1-Mn bulk crystals by ultrasonication (Scheme 1) to study their electrochemically properties since Mn-containing compounds are usually considered as electrochemical active in both oxygen reduction and evolution reactions. Fortunately, a stable P1-Mn dispersion solution of H2O/EtOH was obtained as shown in Figure 2b.

Figure 1. (a) The XRD patterns of simulated P1 (black), as synthesized P1 (red) and P1-Mn (blue), respectively; (b) Basic structural unit of P1 and P1-Mn cluster. 11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 12 of 42

Scheme 1. Schematic illustration of the preparation of Mn-doped multimetallic (ZnGe) oxysulfide NPs from P1-Mn crystals by ultrasonication.

As can be seen in Figure 2a, the structure of original P1-Mn crystals changed after ultrasonication, indicating the transformation of P1- Mn crystals to some new kind of compound or structure. To figure out the morphology and composition of the finally obtained species, TEM and HRTEM was then used to investigate. The low-magnification TEM image of Figure 2b clearly shows that monodisperse NPs (denoted as P1-Mn NPs) distributed on carbon-coated copper TEM grid were uniform with a mean size of ca. 3.5 nm (inset of Figure 2b), even though little aggregation was observed. Furthermore, the HRTEM image of Figure 2c reveals that the well-resolved lattice fringes with a d-spacing of about 0.306 nm, consistent with the corresponding XRD peak of P1-Mn NPs at 2θ = 29.2º in Figure 2a (from Bragg’s equation), revealing the high crystallinity of the obtained NPs. Figure 2d shows HAADF/STEM and elemental mapping images of the P1-Mn NPs, demonstrating the elements of P1-Mn were all retained and distributed uniformly in the monodisperse ultrafine NPs. Furthermore, we compared the obtained the XRD pattern of P1-Mn NPs with possible compounds (Figure S5), which revealed that the P1-Mn NPs does not match all possible compounds according to the elements of P1-Mn crystals. Combined with TEM and elemental mapping results, P1-Mn NPs were of pure phase and Mn was doped in the NPs as Zn is the main metal in NPs as determined by ICPMS (Table S2). What is more, as seen in Figure S5a, the different (marked by red line) and 12 ACS Paragon Plus Environment

Page 13 of 42 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 Applied Materials & Interfaces

unchanged peaks (marked by blue line) between P1-Mn NPs and P1-Mn crystal indicated P1-Mn crystals endured an incomplete structural reformation to some extent during ultrasonication. Moreover, P1 crystals could also be dispersed into small NPs as shown in Figure S6. Different from that of P1-Mn crystal, the structure of obtained NPs from ultrasonication of P1 crystals did not change much except the ratio of some peaks’ intensity according to the XRD patterns shown in Figure S6a. The reason for the different results will be discussed below. Though the exact structure of P1-Mn NPs is unclear at present, it does not affect exploring its application in electrochemistry.

Figure 2. (a) The XRD patterns of P1-Mn before (black) and after (red) ultrasonication; (b) TEM image of the obtained NPs after ultrasonication of P1-Mn crystals (Inset, particle-size distribution histogram and photograph of a stable P1-Mn dispersion in H2O/EtOH, demonstrating the Tyndall effect with a laser pointer); (c) High resolution TEM image and the corresponding 13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 14 of 42

Fast Fourier Transform (FFT) pattern of the obtained NPs; (d) HAADF/STEM image and elemental mapping images of the NPs in the selected area, scale bar: 20 nm. In order to ascertain whether P1-Mn NPs could serve as a highly efficient ORR electrocatalyst just as usually used Mn-based ORR electrocatalysts, the sample was mixed with KB to form a catalyst ink and then coated onto RDE electrodes for comprehensive electrochemical measurements. After loading on KB, the P1-Mn NPs/KB composite was characterized by XRD, TEM and SEM/EDS (Figure S7-S9). And the P1-Mn NPs were distributed homogeneously on the KB obviously (Figure S8). Corresponding EDS spots results (Figure S9) also showed that all elements of original P1-Mn crystal with changed stoichiometry were retained in single P1-Mn NP, which is consistent of the results of elemental mapping and ICP-MS and confirmed that Mn doping in NPs as the P1-Mn NPs are of pure phase (Figure S5).

14 ACS Paragon Plus Environment

Page 15 of 42 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 Applied Materials & Interfaces

Figure 3. (a) Cyclic voltammograms (CVs) of P1-Mn NPs/KB in N2- (black) and O2- (red) saturated 1 M KOH solution. Scan rate, 50 mV/s; (b) LSV curves of KB, P1 NPs/KB, Mn2O3/KB, P1-Mn NPs/KB and 20% Pt/C obtained on RDE electrodes in O2-saturated 1 M KOH solution at the rotation speed of 1600 rpm. Scan rate, 2 mV/s; (c) Tafel slope values at low overpotential regions derived by the mass-transport correction of corresponding RDE data in (b) for Mn2O3/KB, P1-Mn NPs/KB and 20% Pt/C; (d) The metal mass activities (MMA) of Mn2O3/KB, P1-Mn NPs/KB and 20% Pt/C at potentials of 0.90 V, 0.85 V, and 0.80 V versus RHE. As shown in Figure 3a, the CV curve of P1-Mn NPs/KB composite displays a remarkably enhanced cathodic peak at ca. 0.83 V in O2-saturated 1 M KOH, while a much smaller reduction peak was obtained at the same potential when the electrolyte was saturated with N2, which suggested P1-Mn NPs/KB composites show the pronounced electrocatalytic activity for the ORR. Moreover, the CV is similar to the commercial Mn2O3 compound deposited on glassy carbon electrode in alkaline solution (Figure S10) and the voltammetric peaks are associated with the redox processes of Mn(II)/Mn(III) and Mn(III)/Mn(IV) with overlapped features.60,

68, 69

The

displayed high peak current for Mn3+ oxidation at ca. 0.85 V indicated that a higher density of Mn(III) species on the surfaces of the NPs, which suggested a large number of ORR active sites according to the reported ORR mechanism of MnOx in the literatures.60, 69 For comparison, the P1 NPs sample was also prepared by the same method (Figure S6b). As displayed in Figure 3b, the LSV polarization curve of P1-Mn NPs/KB catalyst exhibited a remarkable ORR activity with onset potential (Eonset) and half-wavepotential (E1/2) of ca. 0.92 and 0.86 V (vs. RHE), respectively, which are better than those of commercial Mn2O3/KB (Eonset, 0.91 V and E1/2, 0.82 V), while only 50~60mV lower than those of commercial 20 % Pt/C 15 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Page 16 of 42

catalyst (Eonset, 0.97 V and E1/2, 0.87 V). In view of the quite low activity of both KB and P1 NPs, we realized that the P1-Mn NPs may play a crucial role in ORR activity. Although the obtained diffusion-limited current density of the P1-Mn NPs/KB is ca. 3.43 mA/cm2, which is a bit smaller than that of 20% Pt/C (3.6 mA/cm2), it is better than that of Mn2O3/KB (3.0 mA/cm2). The electrochemical characteristics of measured ORR are summarized in Table 1. It is important to note that the current of ORR is smaller in 1 M KOH than in 0.1 M KOH due to the less oxygen solubility.62 To the best of our knowledge, the catalytic activity of the P1-Mn NPs/KB catalyst is among the best performance reported to date for Mn-based ORR electrocatalysts in 1 M alkaline electrolyte and could be compared to other non-precious ORR electrocatalysts reported recently, as summerized in Table S4 in the Supporting Information.

Table 1. Summary of ORR parameters and activities of 20% Pt/C, P1-Mn NPs and Mn2O3.a ORR parameters

a

20% Pt/C P1-Mn NPs/KB Mn2O3/KB

Onset potential (V)b

0.97

0.92

0.91

Half-wave potential (V)

0.87

0.86

0.82

Electron transfer number

3.94

3.91

3.87

HO2− yield (%)